Physical Therapy of the Low Bck

Second Edition
Otto D. Payton, Ph.D., Chairman
Louis R Amundsen, Ph.D.
Suzann K. Campbell, Ph.D.
John L. Echternach, Ed.D.
Already Published
Hand Rehabilitation
Christine A. Moran, M.S., R.P.T., guest editor
Spors Physical Therapy
Donna Bernhardt, M.s., R.P.T., A.T.e., guest editor
John L. Echternach, Ed.D., guest editor
Therapeutic Considerations for the Elderly
Osa Littrup Jackson, Ph.D., guest editor
Physical Therapy Management of Arhritis
Barbara Banwell, M.A., P.T., and Victoria Gall, M.Ed.,
P.T., guest editors
Physical Therapy of the Cerical and
Thoracic Spine
Ruth Grant,, Grad.Dip.Adv.Man.Ther.,
guest editor
TM) Disorders: Management of the
Cranlomandlbular Complex
Steven L. Kraus, P.T., guest editor
Physical Therapy of the Geriatric Patient,
2nd Ed.
Osa L. jackson, Ph.D., R.P.T., guest editor
Physical Therapy for the Cancer Patient
Charles L. McGarey III, M.s., P.T., guest editor
Gt In Rehabilitation
Gar L. Smidt, Ph.D., guest editor
Physical Therapy of the Hlp
john L. Echternach, Ed.D., guest editor
Physical Therapy of the Shoulder, 2nd Ed.
Robert Donatelli, M.A., P.T., guest editor
Pediatric Neurologc Physical Therapy, 2nd Ed.
Suzann K. Campbell, Ph.D., P.T., F.A.P.T.A., guest editor
Physical Therapy Management of Parkinson's Disease
George I. Turnbull, M.A., P.T., guest editor
Pulmonar Management In Physical Therapy
Cynthia Cofn Zadai, M.S., P.T., guest editor
Physical Therapy Assessment In Eary Infancy
Irma j. Wilhelm, M.s., P.T., guest editor
Forthcoming Volumes in the Series
Physical Therapy for Closed Head Injur
jacqueline Montgomer, P.T., guest editor
Physical Therapy of the Knee, 2nd Ed.
Robert E. Mangine, M.Ed., P.T., A.T.e., guest editor
Physical Therapy of the Foot and Ankle, 2nd Ed.
Gar C. Hunt, M.A., P.T., a.e.s., and
Thomas McPoil, Ph.D., P.T., A.T.e., guest editors
Physical Therapy
of the Low Back
Second Edition
Edited by
Lance T. Twomey, Ph.D.
Deputy Vice-Chancellor
Professor of Physiotherapy and Clinical Anatomy
Curtin University ofTechnolog
Perth, Western Australia
James R. Taylor, M.D., Ph.D.
Associate Professor
Department of Anatomy and Human Biolog
University of Western Australia
Medical Faculty
Nedlands, Western Australia
Research Fellow
Department of Neuropatholog
Clinical Assistant
Sir George Bedbrook Spinal Unit
Royal Perth Hospital
Perth, Western Australia
Spinal Physician
Perth Pain Management Centre
Applecross, Western Australia
New York, Edinburgh, London, Madrid, Melbourne, Toko
To our wives,
Meg and Maie
Nikolai Bogduk, M.D., Ph.D., B.Se.(Med.), F.A.C.R.M.(Hon.)
Professor, Department of Anatomy, University of Newcastle Faculty of Medicine;
Director, Cervical Spine Research Unit, Mater Misericordiae Hospital, Newcastle,
New South Wales, Australia
Margret I. Bullock, Ph.D., F.T.S.
Professor, Department of Physiotherapy, The University of Queensland, Brisbane,
Queensland, Australia
Joanne E. Bullock-Ston, Ph.D.
Lecturer, Department of Physiotherapy, The University of Queensland, Brisbane,
Queensland, Australia
Caroline Dre, M.S., P.T.
Clinical Instructor, Kaiser Hayward Physical Therapy Residency Program i n
Advanced Orthopaedic Physical Therapy, Hayward, Califoria; Private Practitioner,
Redwood Orthopaedic Physical Therapy, Castro Valley, Califoria
Brian C. Edwards, B.Se., B.App.Se., Grad.Dlp.Man.Ther.
Specialist Manipulative Physiotherapist and Honorary Fellow, Curtin University of
Technology; Principal, Brian C. Edwards and Associates, Perth, Wester Australia,
Joe Farrell, M.S., P.T.
Senior Clinical Instructor, Kaiser Hayward Physical Therapy Residency Program in
Advanced Orthopaedic Physical Therapy, Hayward, Califoria; Private Practitioner,
Redwood Orthopaedic Physical Therapy, Castro Valley, Califoria
Ruth Grant, M.App.Se., Grad.Dlp.Adv.Man.Ther.
Professor, Department of Physiotherapy, and Dean, Faculty of Health and Biomedical
Sciences, University of South Australia, Adelaide, South Australia, Australia
Gwendolen A. Jull, M.Pht., Grad.Dlp.Man.Ther., F.A.C.P.
Senior Lecturer, Department of Physiotherapy, The University of Queensland;
Specialist Manipulative Physiotherapist, Private Practice, Brisbane, Queensland,
Mike Kour, M.S., P.T.
Clinical Instructor, Kaiser Hayward Physical Therapy Residency Program in
Advanced Orthopaedic Physical Therapy, Hayward, Califoria; Private Practitioner,
Redwood Orthopaedic Physical Therapy, Castro Valley, Califoria
x Contributors
Colleen B. Liston, A.U.A., <rad.Dlp.Hlth.Sc., M.App.Sc., M.A.P.A., M.C.S.P.
Senior Lecturer, School of Physiotherapy, Curtin University of Technology;
Consultant Physiotherapist, Cerebral Palsy Association, and Consultant
Physiotherapist, Community Health Services, Western Australia Department of
Health, Perth, Western Australia, Australia
Geoffrey D. Maitland, M.B.E., A.U.A., F.C.S.P., F.A.C.P.(Monog.), F.A.C.P.,
Visiting Specialist Lecturer for the Graduate Diploma in Advanced Manipulative
Therapy, School of Physiotherapy, University of South Australia, Adelaide, South
Australia, Australia
Robin A. McKenzie, O.B.E., F.C.S.P., F.N.Z.S.P.(Hon.), Dlp.M.T.
Consultant and Director, Spinal Therapy and Rehabilitation Centre, The McKenzie
Institute International, Pauatahanui, New Zealand; Director, Clinical Services, The
McKenzie Institute Interational Clinics, Wellington, New Zealand
Carolyn A. Richardson, Ph.D., B.Pht.(Hon.)
Senior Lecturer, Department of Physiotherapy, The University of Queensland,
Brisbane, Queensland, Australia
Nils Schonstrom, Ph.D.
Head, Department of Orthopaedic Surgery, Ryhov Hospital, Jonkoping, Sweden
M. Scot Sullivan, M.S., P.T.
Assistant Professor, Department of Physical Therapy, Virginia Commonwealth
University Medical College of Virginia, Richmond, Virginia
James R. Taylor, M.D., Ph.D.
Associate Professor, Department of Anatomy and Human Biology, University of
Wester Australia Medical Faculty, Nedlands, Wester Australia, Australia; Research
Fellow, Department of Neuropathology, and Clinical Assistant, Sir George Bedbrook
Spinal Unit, Royal Perth Hospital, Perth, Western Australia, Australia; Spinal
Physician, Perth Pain Management Centre, Applecross, Western Australia, Australia
Patricia H. Trot, M.Sc., <rad.Dlp.Adv.Man.Ther., F.A.C.P.
Associate Professor and Head, School of Physiotherapy, University of South
Australia; Specialist Manipulative Physiotherapist, Adelaide, South Australia,
Lance T. Twomey, Ph.D.
Deputy Vice-Chancellor and Professor of Physiotherapy and Clinical Anatomy,
Curtin University of Technology, Perth, Wester Australia, Australia
. �:
In the seven years that have passed since the publication of the first edition of
Physical Therapy of the Low Back, the general demographic statistics on low back
pain have remained virtually unaltered. Thus, 80 percent of us will still suffer at least
one disabling episode of low back pain during our life, while within any large group
30 to 35 percent will have some degree of backache at any point in time. However it
is measured, low back pain remains the most costly musculoskeletal disorder in
Western society, and it is depressing to have to report that recent evidence has shown
that when back clinics are opened in developing agricultural societies, large numbers
of people seek treatment. I
There have been very considerable advances in the knowledge of the structure
and function of the vertebral column in recent years, and a considerable body of
further evidence has been gathered to help develop a better understanding of the
pathogenesis of low back pain. This additional information is reflected in changes to
the first four chapters of this book, where morphology, pathology, and biomechanics
are considered in detail. There has also been a considerable advance in our
understanding of the effects of the physical therapies on low back pain. In particular,
we are now much more aware of the importance of mobility and exercise in the
treatment of back pain, while the use of analgesics and rest as treatment of choice has
been shown to have little effect.2•3 The use of intensive exercise programs for the
treatment of chronic back pain allied with better measurement tools remains an
avenue for further progression in the future.4
This second edition has expanded the book from 12 to 16 chapters-indeed,
there are six new chapters in this edition. They include chapters on exercise (Ch. 15)
and intensive rehabilitation (Ch. 10), on ergonomics (Ch. 12) and lifting (Ch. 13), on
back pain in children and adolescents (Ch. 14), and on surgery for intractable low
back pain (Ch. 11). While all chapters show progression and change from the first
edition, Chapters 6, 8, and 9 contain entirely new material. Similarly, the final
chapter-which attempts to gather together the information and provide a rationale
for physical therapy for low back pain-has been entirely rewritten based on
contemporary knowledge.
Low back pain remains an almost universal condition. There have been
improvements in public health education in a number of countries in recent years, and
certainly there is a better understanding of how this disabling condition can be
effectively managed and treated. Back education, changes in lifestyle that emphasize
mobility and exercise, and a more intelligent approach to working and leisure
xii Preface
activities are still important measures in the prevention and management of back pain.
The physical therapist remains a most important element in this equation.
Lance T Twomey, Ph.D.
James R. Taylor, M.D., Ph.D.
I. Frymoyer JW, Cats-Baril WL: An overview of the incidences and costs of low back pain.
Orthop Clin North Am 22:263, 1991
2. Waddell G: A new clinical model for the treatment of low back pain. Spine 12:632, 1987
3. Deyo RH, Diehl AK, Rosenthal M: How many days of bed rest for acute low back pain?
N Engl J Med 315:1064,1986
4. Twomey LT: A rationale for the treatment of back pain and joint pain by manual therapy.
Phys Ther 72:885, 1992
The Lumbar Spine fom Infancy to Old Age
James R Taylor and Lance T. Twomey
2 Lumbar Posture, Movement, and Mechanics 57
Lance T. Twomey and James R Taylor
3 Inneration, Pain Paterns, and Mechanisms of
Pain Production 93
Nikolai Bogduk
4 Anatomy and Function of the Lumbar Back Muscles
and Their Fascia 111
Nikolai Bogduk
5 The Maitland Concept: Assessment, Examination,
and Treatment by Passive Movement 149
Geoffrey D. Maitland
6 Mechanical Diagosis and Therapy for Disorders
of the Low Back 171
Robin A. McKenzie
7 Clinical Assessment: The Use of Combined
Movements In Assessment and Treatment 197
Brian C. Edwards
8 Manipulative Physical Therapy In the Management
of Selected Low Lumbar Syndromes 221
Patricia H. Trott. Ruth Grant. and Geoffey D. Maitland
9 Rehabilitation of Active Stabilization of the
Lumbar Spine 251
Gwendolen A. Jull and Carolyn A. Richardson
10 Intensive Physical Rehabilitaion for Back Pain 275
Lance T. Twomey and James R Taylor
11 Lumbar Spinal Stenosis 285
Nils Schonstrom
xiv Contents
12 Low Back Pain I n the Workplace: An Ergonomic
Approach to Control
Margaret I. Bullock and Joanne E. Bullock-Saxton
13 lifing and Back Pain
M. Scott Sullivan
14 Low Back Pain: Physical Treament In Children
and Adolescents
Colleen B. Liston
15 Therapeutic Exercise for Back Pain
Joe Farrell, Caroline Dre, and Mike Kour
16 Back and Joint Pain: A Rationale for Treatment
by Manual Therapy
Lance T. Twomey and James R Taylor
1 The Lumbar Spine
from Infancy to Old
James R. Taylor
Lance T. Twomey
The human lumbar spine, balanced on the pel vi s by i t s muscl es and l iga­
ment s, supports the whole length of the spi ne above it in the erect posture.
The human lumbar spi ne is uni que in i ts ful l y erect posture. In four-footed
animals where the tasks of weightbeari ng and locomoti on are shared by four
l i mbs, t he thoracol umbar spi ne forms an upwardl y convex bridge between the
forel i mbs and the hi ndl i mbs. In humans, the erect posture adopted i n i nfancy
develops a secondary curve or lordosi s i n the l umbar spi ne, which is general l y
maintained throughout l ife.
In thi s erect col umn, the lowest parts bear the highest loads. Thi s funct ional
requirement i s retected i n the large si ze of the l umbar vertebrae and i n the
thickness and high proteogl ycan content of lumbar i ntervertebral di scs . I n addi ­
tion, the l umbar spi ne has wide ranges oftlexi on, extensi on, and lateral bendi ng.
These dual functi ons of thi s strong dynami c organ shoul d be mai ntai ned over
a life span of 70 to 80 years or more. The nature of this dynami c structure
varies according to fami l ial characteri sti cs, growth and maturi t y, gender, and
long-term responses to wear and tear. Unfortunatel y the fl exi bi l i ty and bounce
of youth are usual l y replaced by the stiffness and slow movement of ol d age,
but vertebral osteophytes and degenerate di scs, wi th backache and stiffness ,
are not i nevitable accompani ments of agi ng. Thi s chapter describes how the
complex structure of the l umbar spi ne vari es throughout l i fe and how i t i s
designed to cope wi t h i ts di verse functions. Thi s chapter begi ns wi th a descrip-
2 Physical Therapy of the LOlli Back
tion of the normal adult lumbar spi ne and then describes the development and
growth of the spi ne before deal i ng with age-related changes i n spinal st ructure
and function. Fi nal l y, it exami nes how structure and function i nfuence the
common t ypes of injury to the lumbar spi ne.
The spi ne is the central skeletal axi s of the body. The musculoskeletal and
vi sceral structures of t he head, neck, upper l i mbs, and torso are supported by
i t or hung from i t . Thi s axi s is strong but not static ; i t provi des wide ranges of
bendi ng and twi st i ng in different di recti ons. The contrary requirements of
strength and mobi l i ty are met by combi ni ng strong i ndi vidual intervertebral
joi nts al l owi ng l i mi ted movement , with a l arge number of motions segments,
whi ch col l ecti vel y gi ve l arge ranges of movement .
Consi deri ng the lumbosacral spi ne in particular, i ts strength, resi l ience,
and mobi l i ty are essenti al to normal human posture and act i vi t y. lts posi tion
affects t he posture of the whol e spi ne and head ( e. g. , pelvic obliquity results
in lumbar scol i osi s and a whole variety of compensatory changes at higher
level s. I. 2 Stiffness in the lumbosacral spi ne restri cts mobi l i ty i n locomot ion,
bendi ng, and l ifti ng, and many other normal , everyday acti viti es . Lumbar spi nal
osteoporosi s and shorteni ng wi th aging al low the rib cage to abut on the pel vi s,
further restri cti ng movement and respi rati on.
A thorough knowledge of the normal adult anatomy of the lumbar spi ne
i s necessary for a good understandi ng of the effects of age-related changes.
The anatomy of t he lumbar spi ne i s descri bed below i n t he fol lowi ng se­
quence: the functional anatomy of the vertebrae and spi nal l igaments; the spi nal
and i ntervertebral canal s and thei r contents ; the bi omechani cs of i njuries; and
the functi onal anatomy of i ntervertebral di scs and zygapophyseal joi nts. These
descri pti ons are based on our own studi es of l arge series of lumbar spines3-7
and on other authori ti es as referenced.
The vertebral bodies and i ntervertebral di scs bear over 80 percent of the
static compressi ve l oad in erect posture. 8 The progressi ve i ncrease i n the sur­
face area of the vertebral end-plates from C2 to t he lower lumbar spine refl ects
the progressi ve i ncrease i n l oadi ng from above down. According to Davi s,9
vertebral body end surface area i s maximal at L4. The end surface of L5 can
be smal l er because its forward tilt transfers a l arger part of i ts load to the
vertebral arches t hrough t he very thick pedi cl es and transverse processes that
characterize L5.
Except for the bony ri m a few mi l l i meters wide, each vertebral end-plate
The Lumbar Spine from Infancy to Old Age 3
is covered by a plate of hyal i ne cart i l age about I mm t hi ck. The outer fi brous
annulus i s attached to t he bony ri m and the inner fibrocarti l agi nous annul us is
attached to the peripheral parts of t he cart i l age pl ate, 3 formi ng an el l i pt ical
envelope for the i ncompressi ble "fl ui d" nucleus pulposus ( Fi gs. 1 - \ and \ -2) .
The average height of adul t mal e l umbar vertebral bodies i s about 25 mm and the
di scs are from \ 0 to \ 2 mm thi ck at thei r centers. Thei r average anteroposterior
diameter i s from 30 to 35 mm. Thei r t ransverse di mensi ons i ncrease from about
40 mm at LI to about 50 mm at L5. Female vertebral dimensi ons are on average
about 1 5 percent smal l er. 3
4 The girth of t he col umn is dependent in part on
the effect of muscular forces duri ng growt h, and the female vertebral col umn
i s more slender than the mal e col umn.
Each vertebral body i s ki dney-shaped and covered by a t hi n shel l of com­
pact bone. Its i nternal body archi tecture refl ects i ts weight-bearing functi on
( Fig. \ -2) . A lateral x-ray shows the predomi nantl y vert i cal ori entati on of i ts
trabecul ae, wi th cross-ti es, part i cularl y numerous near the upper and lower
end-plates. The vertical bony t rabecul ae are quite rigid but they can bend
sl ight l y when loaded. Unsupported verti cal col umns might bend beyond thei r
elastic capaci ty and fracture, if t hey di d not have the support of the t ransverse
cross-t ies , whi ch i ncrease their rigidi ty. Red hemopoeti c marrow fi l l s the honey­
combed i nterstices of this t rabecular scaffol di ng. Thi s composite structure pro­
vides both strength and resi l ience with a smal l amount of "gi ve" i n response
to loading, but the mai n shock absorbers are the i ntervertebral di scs , whose
structure and function will be described i n detai l l ater.
"--( mve )
( poteO )
(wegtbear )
Fig. 1-1. Relationship of the structures of vertebral parts to thei r functi ons.
4 Physical Therapy of the Low Back
Fig. 1-2. A median sagittal section shows the internal structure or "architecture" of
a vertebral body and an intervertebral disc. The vertebral body has vertical l y oriented
weightbeari ng, trabeculae, and cross-ties formed by transverse trabeculae. The interver­
tebral disc has a lamel l ar outer annulus fibrosus (AF). The i nner lamellae of the annulus
are conti nuous with the l amel l ar struct ure of t he carti lage plates (ep) as revealed by
polari zed l ight . The annul us and carti lage plates form an envelope enclosing the nucleus
pulposus ( NP) . Fusion of the ring apophysis with the centrum forms the bony ri m. P,
pedi cl e; L, l ami na; the interlaminar l igaments are the ligamenta tava.
Protection of Neural Structures
The vertebral arches protect the neural structures in the spi nal canal and
i ntervertebral forami na. In contrast to the di rect conti nui ty of the anterior ele­
ments , speci al i zed for l oadbeari ng, the protecti ve arches are only in contact
through thei r articular processes, the gaps between them bei ng bridged by the
el asti c l igamenta fava ( Fi g. 1 -2) . Each arch i s formed by two pedi cl es, which
project backwards from the upper outer margi ns of a vertebral body and by
two l ami nae, whi ch meet i n t he mi dl i ne behind the spinal canal . The pedi cles
are oval i n secti on wi th a thi ck cortex, whi ch i s why surgeons use t hem as
points of fixation for metal pl ates, i nsert i ng t ranspedi cular screws through t he
plates, above and bel ow an unstable segment in a spi nal fusi on.
The lumbar l ami nae are flat plates of bone that cover the posterior aspect
of the spi nal canal , contai ni ng the lower part of the spinal cord and the cauda
equina wi thi n the dural sac . In spi nal stenosi s, parts of the lami nae are removed
The Lumbar Spine from infancy to Old Age 5
"Bllanar" zygapophyseal joit
wih fibrous capsul posteriorly
a imm fJvl anteedially
Subarachnoid space,
�cauda equina"
containing fat and
venous plexus
Fig. 1-3. The appearance of the zygapophyseal (facet) joints i s shown in horizontal
section. The cancel l ous bone of the vertebral body can be seen wi thi n the compact bony
ring. The contents of the vertebral canal i n the l ower l umbar region include the cauda
equina within the dural sac . The extradural veins form a val vel ess venous plexus that
connects regional vei ns outside the vertebral column with basivertebral veins at the
center of each vertebral body.
by surgeons to "decompress" the cauda equi na, i n the procedure known as
laminectomy. The canal can be entered by a needle i nserted i n the gap between
the adjacent lami nae of lower lumbar vertebrae, below the l evel of the spi nal
cord. The epidural space i s entered first , by passi ng the needle through the
l i gamentum favum i nto the vascular epidural fat , wi thout pi erci ng the dura.
This space surrounds t he dura and is t raversed by the spi nal nerves on thei r
way to the i ntervertebral forami na. When the needl e i s passed through the
fi brous dura and i ts membranous i nner l i ni ng of arachnoi d, i t enters t he sub­
arachnoid space contai ni ng the cerebrospi nal fl uid contai ni ng the cauda equi na.
Control of Movements: Levers and Joints
Each vertebral arch has a number of processes t hat act as l evers for move­
ment and as guides and restrai nts of different types of movement s. For each
arch there are two t ransverse processes and one spi nous process to whi ch
muscles and l igaments are attached; they act l i ke l evers . Two superior art i cular
processes project upward from the arch and two i nferior articular processes
project downward. The articular facets on these processes form zygapophyseal
6 Physical Therapy of the Low Back
joi nts ( Fi g. 1 -3) that act l i ke guide rai l s for movement i n flexi on, extensi on, and
lateral bendi ng; they al so resi st axial rotary movements.
The transverse processes are mostl y l ong flat and "spatul ate" ; those of
L3 are longest and those of L I are shortest, but those of L5 are rounded,
stout, and strong. Occasional l y the t ransverse processes of LI are enl arged
wi th joi nts, formi ng l umbar ri bs . Thei r anterior surfaces form parts of the attach­
ments of psoas major and quadrat us l umborum. The ti ps of the L5 transverse
processes are attached to the i l i ac crests by i l i ol umbar l igaments , whi ch contrib­
ute to the stabi l i ty of the l umbosacral junction.
Lumbar t ransverse processes are costal el ement s; on the back of the base
of each process is an accessory tubercl e from whi ch a smal l l igament bridges
over the medi al branch of a dorsal ramus to the mami l lary process of the supe­
rior arti cul ar process. Thi s forms a smal l tunnel for the nerve as i t descends
to suppl y the inferior part of a zygapophyseal joint and the superior recess of
the next zygapophyseal joi nt.
The l umbar spi nous processes are l arge and hatchet-shaped, projecting
straight backward from the j unction of the right and l eft l ami nae at a level I
cm or more bel ow the correspondi ng vertebral bodi es . The wide gutter-shaped
hollow on each si de of the spi nous processes, between the spi nous and trans­
verse processes, is fi l led by the post spi nal muscl es , the longitudinal erector
spi nae and the obl i que mul tifi dus. These are enclosed i n a strong envelope
formed by the posterior and mi ddl e layers of the l umbodorsal fascia, which
take attachment from t he tips of the spi nous and t ransverse processes ( Fig. 1 -4).
These muscl es, wi thi n thei r fascial envel ope, cover the backwardl y projecting
l umbar zygapophyseal joi nts, the l ami nae, and the l igamenta fava. Al l these
muscl es are extensors , but the deepest fasci cl es of the l umbar mul tifdus mus­
cl es also play an i mportant role as stabi l i zers of the zygapophyseal joints . De­
tai l ed descri pti ons of all the muscl es appear i n Chapter 4.
Spinal Ligaments
The anterior and posterior l ongitudinal l igaments form long ribbonl i ke liga­
ments that l i ne the anteri or and posterior surfaces of the vertebral bodies and
i ntervertebral di scs of the whole vertebral col umn. They form contrasting
shapes: the broad ribbon of the anterior l ongitudinal l igament has paral l el mar­
gi ns and the posterior l ongi tudi nal l igament is dentate in outl i ne, widest at the
discs and narrowest at the middle of each vertebral body. The anterior l igament
is qui te firml y attached to both the anterior vertebral periosteum and to the
anterior annul us of each di sc, but the posterior l igament is only attached to the
di scs. Its narrow part bridges over a sl ight concavi ty behi nd each vertebral
body, whi ch contai ns the anterior epi dural vei ns where they connect to the
basi vertebral vei ns wi thi n each vertebral body. These longi tudi nal l igaments
are conti nuous , in contrast to the l igaments of the vertebral arches. They are
i mportant structures because thei r fi bers may bridge over several segments.
3 layers of lumbar
The Lumbar Spine from Infancy to Old Age 7
Posterir layer of Unbar fasci
and eector spiae
fascia fusing :-
Fig. 1 -4. The three layers of the l umbar fascia encl osing the quadratus l umborum in
a lateral compart ment and the dorsal spinal muscl es in a posterior compartment ( showing
erector spinae and multifidus as one muscl e mass). The psoas i s enclosed by i ts own
psoas fascia. The divi sion of the spinal nerve into dorsal and ventral rami and of the
dorsal ramus into medial and lateral branches i s shown. ( Modi fied from Last ,79 wi th
permission. )
Whi l e they are thi nner than the anterior and posterior annul us of the di sc, i n
the cervical spine t hey may form the last l i ne of defense agai nst i nstabi l it y, i n
severe flexi on and extensi on i njuries causi ng avul si on of a di sc from an adjacent
vertebral body.
The spaces between t he lami nae are bridged by l igamenta flava, whi ch al so
form the anteri or capsules of the zygapophyseal joi nt s . The spi nous processes
are connected by obl i que i nterspi nous and supraspi nous l igaments. So-cal l ed
intertransverse l igaments are parts of the mi ddl e layer of the l umbar fasci a.
Ligamenta Flava
Ligamenta fl ava are yel l ow, el asti c l igaments that are thi ckest i n the l umbar
region, where they are 2 to 3 mm t hi ck. Medi al to the zygapophyseal joi nts, thei r
el asti c fi bers pass verti cal l y between adjacent l ami nae and wi th the l ami nae they
form the posterior boundary of the spi nal canal . There may be a narrow mi dl i ne
cleft between the right and left hal ves of each ligamentum favum. Over the
anterior aspect of each zygapophyseal joi nt, the lateral fi bers of the l igamentum
8 Physical Therapy of the Low Back
fl avum are di rected obl i quel y upward and lateral l y to form the anterior capsule
of the joint and the posterior boundary of the i ntervertebral foramen. The elas­
ti ci ty of these l igaments mai ntai ns the smooth regular contour of the spi nal
canal in all postures of t he spi ne and they mai ntai n congrui ty in the anterior
parts of the zygapophyseal joi nts by hol di ng the articular surfaces together.
They are stretched by fexion and contract and thi cken during extensi on, with­
out buckl i ng, in the healthy spi ne. If an i ntervertebral di sc loses i ts normal
thi ckness or a vertebral body l oses i t s normal height wi th agi ng, the l igamentum
flavum wi l l thi cken as i t shortens and may lose its el asti ci ty with aging. lo•11
The more fibrous l igamentum tlavum of an el derl y person may buckl e forward
on extension, narrowing the spi nal canal . Apparent thi ckeni ng of t he l igamen­
tum fl avum is someti mes due to underl yi ng bony hypertrophy i n the superior
arti cul ar process of the zygapophyseal joi nts. These l igaments help to resist
excessi ve flexi on of the spi ne, but they are less i mportant i n restrai ni ng exces­
sive fl exion t han the i ntegri t y of congruous zygapophyseal arti cul ar facets. 1
Supraspinous and Interspinous Ligaments
The supraspi nous l igaments and most of the i nterspinous l igaments are
fibrous or col l agenous structures, rei nforced by the most medial fibers of the
erector spi nae and by the i nterspi nal es. They extend down to L5 but not below
L5, except as fbrous muscl e i nsertions. The fi bers of each i nterspi nous l igament
run upward and backward from one spi nous process to the next , as a double
layer with a narrow i nterval between them. 1 3 Anteriorl y they are conti nuous
with the l igamentum favum on each si de. These l igaments appear designed to
l i mi t fl exi on but posterior rel ease experi ments 1
suggest that their role in thi s
respect i s l ess than that of other posterior el ement structures.
The Spinal Canal and the Intervertebral Foramen
The l umbar spi ne doubl es as a mobile supporti ng structure and as a conduit
for the lower end of the spi nal cord and the cauda equi na. The lower part of
the spi nal cord is a vital part of the central nervous system, contai ni ng control
centers for refl ex acti vi ty in the lower l i mbs, bladder, bowel , and reproducti ve
organs. It is al so a pathway for motor and sensory communication between the
brain and the lower parts of the body. The lumbar enl argement of the cord l ies
i n the thoracol umbar region of the spi nal canal and it tapers to a point (the
conus) , whi ch generall y termi nates at the level of the L I -L2 di sc. Occasional l y,
i t ends sl i ght l y higher or some di stance l ower i n the range between TI2 and
L3; i n spi na bifi da, an abnormal cord may termi nate lower than normal if i t i s
tethered to t he meni nges. The cauda equi na i s an i mportant part of the periph­
eral nervous system, col l ecti ng the motor and sensory roots i nto the l umbar
and sacral spi nal nerves to supply the lower l i mbs, bl adder, and bowel , as wel l
The Lumbar Spine from Infancy to Old Age 9
as other pelvic and perineal structures. The cord and cauda equi na are contained
wi thi n the dural sac , whi ch i s l i ned on i ts inner aspect by the arachnoid mem­
brane containing the cerebrospi nal fluid i n the subarachnoi d space. The dural
sac contai ni ng the subarachnoi d space terminates at the S2 l evel , marked by
the posterior superior i l iac spi nes ( Fi g. 1 -5). Above thi s level , there i s an angled
hollow at the l umbosacral junction and the lumbar spi nous processes can be
readi l y palpated.
The nerve rootl ets that form the spinal nerves arise in paral l el conti nuous
l i nes along the front and along t he back of the spi nal cord. Sets of these anterior
or posterior rootl ets converge to form si ngle anterior and posterior roots, whi ch
i n tur uni te to form a spi nal nerve. The posterior roots contai n about three
times as many axons as the anterior roots . The anterior roots t ransmi t the
somatic motor fi bers from the anterior horn cel l s ( lower motor neurons) and
sympathetic preganglionic fi bers , from the lateral horn of gray matter i n the
spinal cord, to the sympathetic gangl i a. The dorsal roots t ransmi t a variety of
sensory fi bers. Those parts of the nerves wi thi n the dural sac do not have the
strong connect i ve ti ssue coverings, whi ch they wi l l recei ve as they pi erce the
dura to enter the i ntervertebral forami na. Therefore, the rootl ets are not as
strong as the spi nal nerves and are more vul nerable to i njury. The cerebrospi nal
$Çín$l ÇOfO
[$ÇI$l DØfV=
Fig. 1-5. A dorsal view of the l umbosacral spine from which the vertebral arches have
been removed shows the spinal canal contai ni ng the dural sac , which terminates opposite
the posterior superior i l iac spines. The dura fuses with the outer fibrous covering ( epi neu­
ri um) of each spinal nerve as it exi ts the dural sac.
1 0 Physical Therapy of the Low Back
flui d t hat bathes the roots suppl i es some nutri t ion to the nervous t i ssue of the
cord and provi des some protection from mechanical forces. When the nerve
roots pierce the arachnoi d and the dura they are surrounded by a short funnel­
shaped extensi on of the arachnoi d i nto the medial end of the foramen. This
contains cerebrospinal fl ui d and can be outl i ned i n a myelogram. The sleeves
of arachnoi d end i n the i ntervertebral forami na, but the dura continues as a
connecti ve t i ssue sl eeve along each spi nal nerve, formi ng i ts epi neuri um. Pe­
ripheral nerves are tough structures because a high proporti on of each nerve
i s connecti ve ti ssue. They al so contain many small vessel s and can become red,
swol l en, and edematous when "i rri tated" by noxi ous mechanical or chemical
st i mul i . Withi n the foramen, or in the lateral recess of the spinal canal , each
dorsal nerve root forms a fusiform swel l i ng ( the dorsal root gangl ion). Thi s
contai ns the cel l bodies of al l sensory neurons, both somatic and visceral ,
whether they be from ski n, muscl e, bone, vi scera, blood vessel s, or the dura
i tself.
The di mensi ons and shape of the spinal canal vary according to spinal level
and age.
4 At LI, the canal is usual l y oval in transverse section; at thi s level ,
the average sagittal di ameter of the bony canal i s 20 mm and the average coronal
( i nterpedi cular) di ameter is 25 mm. ls Below Ll the canal changes to a triangular
outl i ne. It becomes trefoil in shape in older peopl e, especi al l y at lower l umbar
l evel s,
due to facet hypertrophy and osteophytosi s bulging the lateral part
of the l igamentum flavum forward. At the same ti me as these age changes in the
facets , there is l ikel y to be some age-related shortening of the osteol igamentous
col umn. These common changes, together with posterior di sc bUlging or hernia­
t ion, and lordotic posture with anterior buckl i ng of the l igamentum flavum,
col l ecti vel y or i ndi vi dual l y reduce the space i n the nerve root canal s and may
produce spinal stenosi s , wi th compressi on of the nerves. In young adul ts, nu­
cl eus pul posus ti ssue is soft enough to be extruded from a ruptured di sc, when
it may i mpi nge on and deform the subarachnoid space. Thi s would be vi si ble
on a myelogram as a "fi l l i ng defect . "
The dural sac , contai ni ng the cauda equi na, i s ci rcul ar in cross section and
onl y occupies the central part of t he triangular spinal canal . The epi dural or
extradural space contai ns fat , the spinal nerves descendi ng obl i quel y to their
forami na, and a ri ch pl exus of val vel ess vei ns. Thi s i nternal vertebral venous
plexus has wi despread connections. It recei ves at least two veins through each
i ntervertebral foramen and connects to the central marrow of each vertebral
body by a basi vertebral vei n. Wi thi n the anterior epidural space the venous
plexus forms a ladder pattern of vei ns. Two longit udinal channels (one on each
si de) have lateral communi cations through each i ntervertebral foramen and a
cross communication to each basi vertebral vei n. At the upper end of the spinal
canal the l ongi tudi nal veins are conti nuous with i ntracranial venous si nuses;
The Lumbar Spine from Infancy to Old Age I I
below, they are connected by sacral vei ns to pel vi c venous pl exuses. At each
intervertebral foramen the plexus i s connected to the regional segmental vei ns:
the l umbar, posterior i ntercostal , and vertebral vei ns . Blood can flow i n any
direction i n the pl exus, accordi ng to regional di fferences i n pressure, affected
by respirati on, coughi ng, and strai ni ng. If a vena cava i s blocked, i t can form
a bypass route for the venous return to t he heart . Cancer cel l s may spread by
it from a primary tumor (e . g. , i n the breast or prostate) to the vertebral spongi­
osa, gi vi ng vertebral metastases , with the possi bi l i t y of pathologic fracture.
Small arteries also enter the spi nal canal from each i ntervertebral foramen
to supply the vertebrae, meni nges, and cauda equi na. These are mostl y very
smal l , butt hey di vide i nto three branches, 1 8 one to supply the vertebral body,
one t o fol low t he nerve roots ( radi cular branch) , and one to suppl y t he posterior
elements. Like the veins t hey form a ladder pattern of anastomosi s with thei r
neighboring vessel s i n the anterior epi dural space. The radi cul ar branches sup­
pl y the nerve roots and fol low them through t he dura and arachnoi d i nto the
subarachnoid space. One of the radi cular arteries, i n the thoracol umbar region,
i s very i mportant because i t suppl i es t he lower thoracic and l umbosacral parts
of the spi nal cord. This "great spi nal artery" of Adamki ewi cz usual l y arises
from a lower thoracic posterior i ntercostal artery ( most often TI O on the left ) ,
and i t anastomoses wi th t he anterior and posterior spi nal arteries of t he spi nal
cord. 1
20 Because i ts origin is variabl e, it may be at risk in operati ons on aortic
aneurysms. If i t i s i nadvertentl y damaged, paraplegia i s t he l ikel y resul t.
The i ntervertebral forami na are regularly spaced lateral openi ngs from the
spi nal canal . Each l ies between the pedi cl es above and bel ow, wi th a vertebral
body and i ntervertebral disc i n front, and a zygapophyseal joi nt , covered by
the l igamentum flavum, behi nd. The average adult foramen i s oval and measures
IS mm i n height and 8. S mm i n i ts wi dest anteroposterior extent , except at
LS-S I, whi ch i s more rounded and i s 1 2 mm from front to back and 13 mm i n
height. The space i s widest above where i t contai ns t he nerve; t he l ower part
i s rel ati vel y narrow and occupi ed by vei ns, whi ch may be separated from the
upper part by a smal l transforami nal l i gament . In addi ti on to a spi nal nerve ,
each l umbar foramen transmi ts a smal l branch of segmental artery, two or more
quite large vei ns, and a smal l recurrent branch of the spi nal nerve cal l ed a
si nuvertebral nerve. The si nuvertebral nerves suppl y the dura and posterior
longitudi nal ligament, and parti ci pate i n the supply of the outer lamel l ae of
the posterior annul us fibrosus. In the sacrum there are anterior and posterior
forami na that separately t ransmi t t he ventral and dorsal rami of the sacral
nerves. The anterior sacral forami na also transmi t smal l l ateral sacral arteries
and vei ns.
The lumbar forami na are short canal s, whi ch range i n l ength from 9 mm
at LJ-L2 t o 2 cm at LS-S 1,
4 i n proportion to t he t hi ckness of t he pedi cl es
above and below. Each nerve-root canal recei ves the spi nal nerve roots i n thei r
dural sheat h, i n t he l ateral recess of the spi nal canal , and passes obl i quel y
downward and lateral l y bel ow the pedi cl e, i nto the upper part of t he i nterverte­
bral foramen. I n the medial part of the foramen, the anterior root and the poste­
rior root (with i ts gangl i on) unite to form a mixed spi nal nerve and the sl eeve
1 2 Physical Therapy of the Low Back
of dura becomes conti nuous with the epi neuri um of the spinal nerve.
1 Thi s
nerve, contai ni ng motor, sensory, and sympathetic fi bers, passes out through
the wide upper part of the foramen , behind the lower part of the vertebral body
and above the l evel of the i ntervertebral di sc . It i mmediatel y di vi des into ventral
and dorsal rami . The l umbar ventral rami form a plexus i n the psoas muscle
and parti ci pate wi th sacral ventral rami i n the l umbosacral plexus, forming
femoral , obturator, gluteal , and sci ati c nerves to suppl y lower l i mb structures .
The ventral rami of S I-L3 groove t he anterior aspect of the sacrum as they
pass lateral l y from their anterior forami na to joi n the l umbosacral trunk
(L4-LS), which descends over the al a of the sacrum. These nerves are stretched
fai rl y taut over the al a and anterior sUlface of the sacrum as they form the
sciatic nerve, whi ch passes out of the pelvi s through the greater sciatic notch,
then down behi nd the hi p joi nt . When L4, LS, or S I i s entrapped, combined
hip fexion and knee extensi on stretches the sciatic nerve and pulls on the
entrapped nerve and i ts dural sheath, el i ci ti ng pain and refl ex muscle spasm.
The sl ump test i s a more general test for dural i rritati on, depending on t he same
biomechanical pri nci pl es .
Each dorsal ramus di vides i nto medial and lateral branches: the lateral
branch suppl i es spinal muscl es and become cutaneous , suppl ying the ski n of
t he low back and gl uteal region; t he medial branch wi nds around the articular
pi l lar of the superior arti cul ar process and suppl i es two zygapophyseal joi nts.
The i ntervertebral foramen may be reduced i n si ze by the same age-related
and pathologic processes that affect the spi nal canal . These i ncl ude motion
segment i nstabi l i t y, wi th retrol i sthesi s of the upper vertebra,S di sc thi nni ng or
vertebral end-plate col l apse, lordotic posture, and osteophytosi s of the zygapo­
physeal joi nt . The combination of Z joi nt osteophytes and reduction i n height
of the i ntervertebral foramen are t he changes most l ikel y to put the spi nal nerve
at risk of entrapment ( Fi g. 1 -6) . A herniated di sc is likely to affect the nerve
Fig. 1 -6. Lower lumbar intervertebral
foramen, with the narrowing that results
from shortening of the column ( whether
from osteoporostic bowing of the end­
plates or thi nning of the disc) and retro­
l i sthesis of the upper vertebra associated
with instabi l i ty.
The Lumbar Spine from Infancy to Old Age 13
descendi ng to the next i ntervertebral foramen, i n the l ateral recess of the spi nal
canal .
Common Injuries to the Lumbosacral Spine
From our experience of secti oni ng over 200 l umbar spi nes and over 1 00
cervical spines from autopsi es, a substantial proporti on of them from acute
i njuri es, we have observed a wide variety of bone and sof-ti ssue i njuries. It i s
interesti ng t o contrast t he i njuries i n the t wo regi ons, because thei r different
biomechanical behaviors result i n di fferent regional patterns of i njury.
The 45° orientation of cervical facets contrasts wi th the orientat ion of l um­
bar facets almost paral l el to the l ong axi s of the spi ne. Cervi cal movements,
around a center of motion i n the vertebral body bel ow, are i nevi tabl y accompa­
nied by translation between adjacent vertebrae, but the verti cal l umbar facets
severel y l i mi t t ransl ati on, protect i ng the di sc from sheari ng. Lumbar fl exi on
and extensi on are mai nl y rocking movements of one vertebral body on another,
accompanied by upward and downward gl i di ng between the facets. The t hick­
ness of the disc can accommodate the very smal l amount of transl atory move­
ment . The relati vel y large forward sl i de of one cervi cal vertebra on the next
vertebra, which accompanies cervi cal flexi on, produces sheari ng forces wi thi n
t he di sc. These forces of everyday l i fe produce transverse fi ssures i n the di scs ,
fi rst at the uncovertebral joi nts, formed during adolescence, and then extendi ng
transversely through the whol e posterior half of t he di sc i n young adults . Fi ssur­
ing in l umbar di scs is seldom present in young adul t s. When i t appears in later
life i t i s l ess uni versal , and usual l y l ess obvious, than in cervical di scs .
The resul ts of trauma to the l umbar or cervi cal spi nes, i n high speed accel ­
eration or deceleration i njuri es, are i nfl uenced by t he different regional bi o­
mechani cs descri bed. I n very rapid flexi on of t he neck, the facets sl i de freel y
to end range, leavi ng onl y a smal l part of thei r arti cul ar surfaces in contact ,
and most of t he bendi ng and transl atory force i n t he motion segment i s absorbed
by the di scs. At the same t i me, the posterior muscl es and the posterior l igamen­
tous complex are strai ned. I n severe flexion i njuries, the posterior muscl es tear,
the facets may be di slocated, and the di sc may be avul sed from the adjacent
vertebra, wi th a high ri sk of cord damage. Quite frequentl y there i s no bony
i njury. In l ess severe flexi on i njuri es, an incomplete transverse tear appears at
the di scovertebral juncti on, wi thout damage to the l ongi tudi nal l igaments , and
l igamentous and muscular strai ns , short of tearing, occur i n zygapophyseal
capsul es, adjacent l igaments, and postvertebral muscl es. The neck is less well
protected by muscl es i n cervical extension, but the l ikel i hood of fract ure is
even l ess than with fl exi on, except at end range, where the tips of art icul ar
processes may be broken. The mai n i njuries are to t he di scs, whi ch tear, and to
the vascul ar synovial fol ds wi thi n the facet joi nts, whi ch are brui sed. Extensi on
i njuries are often mul ti l evel i njuries.
By contrast , thoracic or l umbar i njuries more frequentl y resul t i n fractures.
In deceleration-fl exi on, t he verti cal l umbar facets absorb t he fi rst force. Smal l
facet fractures, wi th avul si on of the mami l lary process or i nfracti on of the
14 Physical Therapy of the Low Back
art i cular surface of the superior arti cular process, are common. In more severe
i njuries a facet may fracture across i ts base, or through the pars i nterarticulari s,
as i n spondylol ysi s. I f t he facets are not severely fractured i n t he flexion i njury,
the motion segment "hi nges" on the facets and the anterior el ements are com­
pressed. Thi s often resul ts in a wedge compression fracture, most frequently
of T 1 2 or L I; alternati vel y, there is a compression fracture of a vertebral end­
plate, with i ntradi scal bleedi ng or a compression i njury to the i ntervertebral
di sc i tself, wi th di sc contusi on or traumatic herniati on, but the di sc is protected
from severe di srupti on as long as the facets remai n i ntact . Disc tears are less
frequent than i n the cervi cal region because there i s l ess translati on. Extension
i njuries to the l umbar spine are uncommon i n motor vehicle acci dent s, but they
may occur as a result of occupati onal or sporting i njuries. The i nferior facets
are driven down agai nst the lamina below and may cause spondylol ysi s through
the i sthmus of the pars i nterarti cul ari s . Thi s i njury may occur ei ther as an acute
i njury ( e. g. , in Austral ian footbal l ) or as a resul t of a repeated sporting acti vi ty
that consi stentl y stresses the l ow back, such as fast bowl i ng in cricket, or
gymnasti cs . 23
Muscles and Fasciae
The l umbar postvertebral muscl e masses ( erector spi nae or multifidus) are
l arger than the space between the spi nous and transverse processes. Thus the
ti ps of the spinous processes are palpated i n a longitudi nal mi dl i ne groove be­
tween t he two muscl e masses, and t he ti ps of the t ransverse processes lie deep
to the l ateral parts of the muscl es . The mi ddl e l ayer of the l umbar fascia passes
lateral l y from the tips of the transverse processes, separating quadratus l um­
borum from erector spi nae, and the posterior layer of the l umbar fascia extends
lateral l y from the tips of the spinous processes, the two layers enclosing the
erector spi nae and mul ti fi dus muscles ( see Fi g. 1 -4) . The term erector spinae
is reserved for the superfi ci al , longitudi nal l y runni ng fi bers t hat span many
segments . The muscl e erector spi nae forms a compact mass in the l umbar region
and spl i ts i nto three col umns (the spi nal i s, longi ssi mus, and i l iocostalis), as it
ascends i nto the thoracic regi on. Multifi dus l i es deep and medial to erector
spi nae. The fasci cl es of mul tifidus run in an obl i que di rection and only span
two or three segments as they ascend or descend. The deeply placed zygapo­
physeal joi nts are covered by the fasci cl es of multi fidus, which are in turn
covered by the erector spi nae and the posterior, aponeurotic l umbar fasci a.
The deeper parts of mul ti fi dus may pl ay an i mportant role i n stabi l i zi ng the
zygapophyseal joi nt s. 24
25 Detailed descri pti ons of spi nal musculature appear
i n Chapter 4.
Sacrum and Sacroiliac Joints
The central part of the bony sacrum i s formed by the fusion of five centra
but the adult sacrum sti l l contai ns remnants of i ntervertebral di scs wi thi n the
bone. The transverse elements fuse to form the alae and lateral masses . The
adult sacrum i s triangular with its broad base di rected upward and forward and
The Lumbar Spine from Infancy to Old Age 1 5
i ts apex di rected downward andjoi ned t o the coccyx below. I t s anterior suIace
i s smooth, with transverse ridges where the di scs would have been, with four
pairs of anterior sacral forami na between the fused bodi es and the l ateral
masses, opposite the transverse ridges ( Fi g. 1 -7) . The rough posterior suIiace
shows a mi dl i ne spi nous crest , and two lateral arti cular crests wi th the posterior
sacral foramina bet ween the central and lateral crest s. I nferi orl y, above the
joint wi th the coccyx, the posterior wall of the spi nal canal i s defi ci ent i n an
inverted V shape called the sacral hi at us. This gi ves access to the l ower end
of the spi nal canal , whi ch is fi l l ed by fat and vei ns, the dural sac ending at the
S2 level .
On each lateral sacral suIiace, two areas can be di sti ngui shed: a smooth
arti cul ar area i n front and a rough l igamentous area behi nd. The rough posteri or
area i s for the attachment of t he enormousl y strong sacroil i ac l igament s, whi ch
suspend the sacrum between the two i l i a. The smooth arti cul ar suIiace of t he
sacroi l i ac joi nt is auricular in shape. I t may be descri bed as havi ng two "l i mbs ":
an upper shorter l i mb di rected upward and backward and a longer lower l i mb
Fig. 1 -7. Anterior and superior surfaces of a femal e sacrum. The superior articular
process (SAP) of S I and the anterior sacral foramina (AF), which transmit the ventral
rami of sacral spinal nerves, are designated. The ridges mark the positions of fusion
between the five centra forming the central part of the sacrum. The ala, or upper end
of the sacral lateral mass, is crossed by the lumbosacral trunk as it descends to form
the sciatic nerve wi th S I, S2, and S3.
1 6 Physical Therapy of the Low Back
di rected downward and backward. The superior surface of the S I vertebral
body of the sacrum bears the weight of the head, trunk, and upper l i mbs, trans­
mi tted t hrough the l umbar vertebral col umn . Thi s axial force would tend to
rotate the sacrum forward, but t hi s tendency is resi sted by the sacrotuberous
and sacrospi nous accessory l igaments ( Fi g. 1 -8) , which bind the lower parts of
the sacrum and coccyx down to the lower parts of the hip bones. Equally
i mportant i n preventi ng or reduci ng sacroi l i ac movement are t he reci procal
irregularities of the adul t arti cular surfaces, together wi th the great strengt h of
the posterior and i nterosseous sacroi l i ac l igaments ( Fi g. 1 -9) . The carti lage­
covered arti cular surfaces are smooth and fl at in the chi l d, where rotary move­
ment in the sagittal plane i s possi bl e . They become irregular in the mature adult
due to the growth of a ridge or ridges on the i l iac surfaces. Thi s makes move­
ment almost i mpossi bl e i n most adul t s. 26. 27 In young femal es, 8° of movement
is said to be possi bl e, 28 and duri ng pregnancy there is a degree of ligamentous
laxi t y, probably due to the effect of the hormone relaxin on fi brous ti ssues.
The sacroi l iac joi nts shoul d not be consi dered in isolation because they
are part of a three-joi nt compl ex i n the pel vi s. A sacroi l iac joint cannot rotate
wi thout some correspondi ng movement at the symphysi s pUbi s. Bilateral differ-
Body Weiht
Axis of
Fig. 1-8. Body weight , acti ng through the l umbar vertebral column, would tend to
rotate the sacrum forward if its lower end were not "anchored" by the sacrospi nous
and sacrotuberous l igaments. The l igaments resist the anterior rotational effect of body
weight .
The Lumbar Spine from Infancy to Old Age 17
Fig. 1·9. The sacrum i s "suspended" between the two i l ia by very strong posterior
sacroiliac ligaments. Body weight tends to compress the i l ia agai nst the sacrum at the
sacroiliac joints.
ences i n sacroiliac joi nt posture are someti mes seen on radiographs of adoles­
cents who have a leg l ength di screpancy. Pelvic torsion occurs when one sacroil­
iac joi nt i s rotated more than the other and a "step" is apparent between t he
right and lef pubic bones on the anteroposterior radiograph. 29.30 The sacroi l i ac
joi nt is deeply si t uated, medial to t he posterior i l iac spi nes, whi ch are readi l y
palpabl e, but t he joi nt itself i s di ffi cul t to palpate because i t i s covered by
muscl es.
Joint: Mobility with Stability
The mul ti segmental constructi on of the spi ne combi nes strength wi t h stabi l ­
i ty. The l umbar vertebral col umn i s requi red to provi de stabi l i t y i n loadbearing
and a wide range of mobi l i t y. Mobi l i t y and stabi l i ty are usual l y in i nverse pro­
portion to each other, but t hese t wo apparentl y contradictory requi rements
are achi eved i n the l umbar vertebral col umn by vi rtue of i ts mul t i segmental
construction. Each mobile segment , consi st i ng of one i ntervertebral di sc and
two zygapophyseal joi nt s, has only a l i mi ted range of movement and therefore
remains stable . However, the fi ve l umbar mobi l e segments together provide
large ranges of sagittal and coronal pl ane movement . There is a l i mi ted range
of axial rotation, whi ch is part l y dependent on coupl i ng wi t h coronal and sagittal
plane movement .
Articular Triad
The i ntervertebral di sc and i ts t wo associ ated synovi al joi nts combi ne i n
a uni que way to give strengt h and stabi l i ty to t he mobi l e segment , but at t he
same ti me provide i t wi t h adequate mobi l i t y.
1 8 Physical Therapy of the Low Back
The i ntervertebral di sc provides the main strength and sti ffness of the mo­
t ion segment , but i t s sl ight compl iance and i t s consi derable thi ckness ensure a
useful movement range. It i s a structure of uni que si mpl i ci ty in concept , but
wi th a complexity of fine structure i n its part s. Though general l y described as
formed by the annul us fibrosus and the nucl eus pul posus, i t shoul d be regarded
as i ncl udi ng the cart i l age pl ates, whi ch bi nd and uni te it to the vertebral bodies
above and bel ow Y The carti lage plates and the i nner annul us fi brosus form a
conti nuous envelope encl osi ng the nucl eus pul posus ( see Fi g. 1 -2) .
Annulus Fibrosus. The annul us fibrosus consi sts of 1 2 t o 1 6 concentric
lamel lae, with two di sti ngui shable part s . The outer fi brous annulus contai ns
rel ati vel y l i ttl e proteogl ycan and attaches to the vertebral ri m. The i nner fibro­
cart i lagi nous annul us is ri ch i n proteoglycan and is conti nuous wi th the cart i l age
plates . The annul ar l amel l ae have an outwardl y convex arrangement and are
arranged in spi ral l i ng sheets around the circumference of t he nucleus. The paral­
lel fi bers of each successi ve sheet of col lagen bundl es cross the fi bers of the
next sheet at an i nterstriation angl e of about 57°. 32 The arrangement i s not
unl i ke t hat of the l ayeri ng of an oni on and i s si mi lar to the archi tecture of
col lagen i n the osteons of compact bone. I n the i ntervertebral disc the arrange­
ment gi ves the annul us great strengt h. The outer fi brous lamel l ae of the annul us
are fi rml y embedded i n the bony vertebral ri m. The i nner fibrocarti l agi nous
l amel l ae of the annul us are shown by polarized light sti dues3 1 to be di rect l y
conti nuous wi th the horizontal l amel l ae of the "hyal i ne" carti l age plates above
and below the nucl eus. The i nextensi bl e but deformable envel ope formed by
the annul us and t he cart i l age plates encl oses the el l i pti cal sphere, which i s the
nucl eus pul posus.
The cart i l age pl ates not onl y form an essenti al part of the envelope contain­
ing the nucl eus, they are al so fi rml y bound to the end sUce of the vertebral
body; they were devel opmental l y parts of the carti lage model of the vertebral
body, and are someti mes described as its unossifed epi physes. In the growing
i ndi vi dual , growth pl ates at the junction of the bony vertebral body and the
cart i l age plate ensure growth in vertebral height . The cart i l age plates are best
regarded as t hose parts where the vertebra and di sc i nterl ock. Observations of
i nj uries suggest that they are more fi rml y attached to the di sc than to the verte­
bra, part icul arl y in chi l dren where i njury may resul t in a cleft at the carti l age
plate vertebral end-plate juncti onY
Nucleus Pulposus. The i nfant nucl eus pul posus i s a vi scous-fl uid struct ure
wi th a cl ear, watery matri x. 3 I t s appearance and consi stency are qui te different
in the adul t , where the nucl eus contai ns many col lagen bundl es , has a reduced
water content , and i s di ffi cul t to di ssect cl ear of its envelope. However, the
heal t hy young adult nucl eus behaves hydrostati cal l y as a vi scous fl uid, which
i s i ncompressi bl e and changes shape freel y Y By changing shape the di sc act s
as a joi nt . The nucl eus recei ves axial loads and redi stri butes them centri petall y
to t he surroundi ng envel ope, di ssi pati ng verti cal forces i n horizontal di recti ons
and act i ng as a shock absorber. The el l i pti cal envelope around the nucl eus.
formed by the i nner annul us and carti lage pl ates wi th a high proteoglycan con-
The Lumbar Spine from Infancy to Old Age 19
tent, deforms under axi al loadi ng. The proteogl ycan mol ecul es resi st t hi s defor­
mat ion, rei nforcing the shock-absorbi ng functi on of the nucleus.
The outer si x layers of t he l umbar annul us are i nnervated, but nerves do
not usual l y penetrate beyond i t s outer thi rd; no nerves have been demonstrated
i n t he nucleus or the cart i l age plates. 34-36 The outer annul us and cart i l age plates
are vascular i n t he fet us and i nfant, but thei r vascul ari ty is progressi vel y re­
duced with maturation. 3. 37 I n t he adult a few vessel s penetrate the calcified
carti l age layer bi ndi ng the carti l age plate to the bony centrum and a few small
blood vessel s persist i n the surface layers of the annul us. 38-4o The avascular
nucl eus contains a sparse cel l populat ion in a watery matrix rich in proteogl y­
cans ( PGs) . The sparse cel l population of the adul t nucl eus recei ves i ts nutri ti on
by diffusion from the few vessel s i n the outer annul us and from the few vascul ar
buds that penetrate the carti l age plates for a short di stance from t he vertebral
marrow spaces . 4
The nucl eus is held under tensi on wi thi n the envel ope formed by the annu­
l us and carti lage plates. Thi s tensi on ( or turgor) i s dependent on the i nextensi bi l ­
ity of the envelope, and i s produced by the chemical force resul ti ng from the
water-attract i ng capaci ty of the PG macromol ecul es. These macromol ecul es
make space for, or "i mbi be, " water. In recumbent post ure, at night , the di sc
tends to swel l , and duri ng the course of each day when the di sc i s compressed
by axi al loading it "creeps" by squeezi ng out water, to become sl ightl y thi nner.
Al l i ndi vi dual s l ose about 1 7 mm of height due to axi al weightbearing duri ng
the day, and regain it when recumbent at night . 3 At autopsy, di scs that are cut
open swell quite soon by absorbing water from the atmosphere. At di scography,
heal thy di scs act i vel y resi st the i njecti on of contrast , whi l e degenerate di scs
allow the contrast to enter wi th relati vel y l i ttl e resi stance.
Zygapophyseal (Facet) Joints
Zygapophyseal joi nts are the principal gui di ng and restrai ni ng mechani sm
of t he mobile segment . Al t hough the di sc i s the strongest part of the motion
segment , the zygapophyseal joi nts are essenti al to protect the l umbar discs from
the rotational and translati onal strai ns that woul d damage the di sc. Wi thout the
zygapophyseal joi nts t he mobi l e segment woul d become unstabl e. The two
zygapophyseal joi nts permit movement to occur i n the sagittal and coronal
planes, but t hey restrain axial rotati on, and bri ng fexion to a halt at the end
of the physi ologic range . 1
4 1 .
They al so wi den the axial load-bearing base.
In normal erect posture they bear 1 5 to 20 percent of axial loadi ng. 33 Thi s load­
bearing function i s l arger i n t he flexed spi ne, part i cul arl y when l ift i ng l oads.
The zygapophyseal joi nts can al so be sources of back pain when thei r fi brous
capsules or synovial fol ds are i rritated.
Joint Anatomy. The zygapophyseal joi nts6.43 are formed between the medi­
al l y faci ng superior articular processes and the l ateral l y faci ng i nferior arti cular
processes of the vertebra above. They are described as havi ng pl ane or fl at
articular surfaces, but flat facets are relati vel y uncommon. The superior articu-
20 Physical Therapy of the Low Back
lar facets are usual l y concave in the horizontal pl ane, encl osi ng the smal ler
convex i nferior art icul ar facets, and resembl e segments of t he surface of a
cyl i nder. Both arti cular surfaces are approxi mately vertical l y oriented, paral lel
to the long axi s of the spi ne , al though they may show a sl ight forward slope.
From the funct ional poi nt of vi ew, it is best to consi der the facets as bipla­
nar ( see Fig. 1 -3) . The anterior thi rd of each superi or arti cul ar process i s ori­
ented cl ose to t he coronal plane and may be cal led the coronal component .
Thi s part prevents or severely l i mi t s forward translation of the upper vertebra
in flexi on and hel ps to control fl exi on. The posterior thi rd is oriented cl ose to
t he sagittal plane and may be cal l ed the sagittal component ; i t restrains or
prevents axial rotation ( Fi g. 1 - 1 0) . The arti cul ar surfaces of the superior articu­
lar process may be curved in a regular concavi ty, but not infrequentl y the
coronal and sagittal components are di sti nct and meet at an angle greater than
90°; in t ransverse secti ons, the joi nts resemble a boomerang. The l umbar zyga­
pophyseal joi nts show a gradual change in orientation from L I down to LS:
the sagittal component is l argest i n L1 -L2 and the coronal component i s largest
in LS-S 1 . The upper l umbar and mi dl umbar joi nts are more consi stentl y bipla­
nar, whi l e the lower joi nts are often truly planar ( Fi g. I - I I ) .
The joi nt i s encl osed by a fibrous capsul e posteri orl y, and the elastic l iga­
mentum flavum forms i ts anteromedial capsule ( see Fi g. 1 -3) . The posterior
" Z
" \ I
Fig. 1 - 1 0. This tracing from a CT scan shows soft-tissue outl i nes of psoas, lateral and
medial parts of erector spi nae, and the outl ines of the L3-L4 zygapophyseal joints. The
"joint space" (articular cartilage) and the compact bone of the subchondral bone plate
with its coronal and sagittal components are distinguished. LF, l igamentum tlavum.
The Lllmbar Spine from Infancy to Old Age 2 1

Fig. 1 - 1 1 . The segmental variation i n the orienta-
tion of l umbar zygapophyseal j oint planes i s
shown based on tracings from CT scans. It should ..
be noted that there is a good deal of variation i n
structure between individual s. ( From Tayl or and
Twomey,6 with permission. )
( `
capsule i s frequent l y di rectl y conti nuous wi th the posterior margin of t he articu­
lar carti lage l i ni ng the superior arti cul ar facet. Thi s capsul ar attachment gives
the appearance of extendi ng the concave socket for the convex i nferior facet .
The capsule i s not as tight posteriorly as it is anteriorl y, and depends on mul tifi­
dus for ' ' tensi oni ng" ( Fi g. 1 - 1 2) ; i t is qui te loose above and below at the superior
and i nferior joint recesses. 6
43.44 A neurovascul ar bundl e lies cl ose to, and may
be seen entering, each joint recess, where a l arge, vascul ar, fat-fi l led synovi al
fol d extends from the i nsi de of the fibrous capsul e, projecti ng for a short di s­
tance between the arti cular sUlfaces. In young heal thy joi nt s, these vascul ar
fat-pads adapt easi l y to the changing shape of t he joi nt cavi ty in movement,
but with aging, or after i njury, they become fi brous, especial l y at the ti ps where
they have been repeatedly compressed between the arti cul ar surfaces .45 A vari­
ety of synovial , fibrofatt y, or fi brous fri nges extend around each joint as space
fi l l ers, from a base on t he arti cular capsule to an apex projecti ng between the
articular surfaces. The vascular fat-pads in the arti cul ar recesses of the lower
l umbar and l umbosacral joi nts are parti cul arl y l arge,44 and contrary to popul ar
bel ief,46 they are i nnervated by smal l nerves, whi ch contai n vasomotor fi bers
and also fi bers that are separate from bl ood vessel s, may contai n substance P,
and are probably both sensory and nocicepti ve. 47
Small vertical ridges or tubercl es, close to the posterior art i cul ar margin
of each superior arti cular process , are termed the mamillary tubercles. These
are formed by the attachments of t he l umbar fasci cl es of mul ti fi dus. A l umbar
fascicle of mul ti fdus descends obl i quel y from the base of a spi nous process
two segments above, to the l ateral margin of each joi nt (see Fi g. 1 - 1 2) . I t uses
22 Physical Therapy of the Low Back
Fig. ) - ) 2. Based on dissection studies, a single
fascicl e of the l umbar mul tifidus i s shown descend­
ing obl iquel y from the spinous process of L2 to the
mami l lary tubercl e on the superior articular pro­
cess of L4, with some capsular insertion at the su­
perior part of the L3-L4 joint Y This is one unit
of a l arger, more compl ex muscl e. This obl iquity
of this deep fasci cl e would al l ow it both to extend
the joints and to exert a small rotary force at the
two mobile segments it crosses, but its principal
function i s probabl y as a fi xator or stabilizer at the
L3-L4 zygapophyseal joi nt . Thi s view contrasts
with most textbook descripti ons. ( From Taylor and
Twomey, 6 with permission. )
thi s attachment t o control joi nt posture and mai ntai n congrui ty i n the posterior
sagittal component of t he zygapophyseal joi nt. 6 This action i s somewhat analo­
gous to the rotator cuff functi on at the shoul der joi nt. Li ke the rotator cuff
muscl es, mul ti fi dus i s partl y i nserted i nto t he joi nt capsule. Observations on
postmortem l umbar spi nes suggest that it hel ps to maintain joint congrui ty,
because t he posterior capsul e i s sl ack and the posterior joi nt sUlfaces separate
sl ightl y when the muscl e is removed. 6,4 1
The sagittal component of the joi nts appears well designed to block axial
rotation around the usual axis of rotation near the posterior sUlce of the
i ntervertebral di sc, but some axial twi sti ng is possi bl e, as measurement studi es
show. 48 Thi s movement i s severely l i mited when i ndi vidual mobi l e segments
are tested in rigid experi mental condi ti ons, but when torque i s applied to the
whole l umbar spi ne, t he resul tant movement is a combi nation of true axial
rotation wi th some coupl ed sagittal and coronal plane movements.
The functi on of the coronal components of the joi nts i n l i mi ti ng flexion i s
l ess wel l known. Posterior-release studi es show that loadi ng of these joint s, by
the forward translational force that accompanies flexi on, i s t he si ngle most
i mportant restrai nt to flexi on at end range, and i s more i mportant than tensi on
The Lumbar Spine from Infancy to Old Age 23
i n the posterior l igaments i n bringing fl exi on to a hal t . 4 The chondromalacia
that select i vel y occurs i n the coronal components of the joints confirms that
these are subject to greater compressi ve loads than the sagittal components. 6
Lumbosacral Joints
The anterior sUlface of the sacrum is normall y i ncl i ned at an angle of about
60° to the vertical plane, gi ving a sharp change in direction from the approxi ­
mately verti cal l y oriented l umbar spi ne. Thi s angulation i nvol ves a di vi si on of
the vertical weight-bearing force through the l umbar spine into two vectors : one
obliquely downward and backward t hrough the sacrum, and the other obl i quel y
downward and forward paral l el to the upper surface of S I . The anteriorly di ­
rected vector of the weight-bearing force exerts a shearing force on the l umbosa­
cral joint s.
Thi s shearing force i s resi sted by ( I ) the two l umbosacral zygapo­
physeal joint s; (2) t he l umbosacral i ntervertebral di sc ; and ( 3) the i l i ol umbar
l igaments . The l umbosacral facets are the most ri gi d structures of the moti on
segment and bear most of the sheari ng stress.
The loading from each l umbosacral zygapophyseal joi nt i s transmitted up­
ward through the i nferior arti cular process of L5 and through the pars i nterarti c­
ul ari s of L5 , a narrow i sthmus of bone between i ts superior and i nferior arti cul ar
facet s. Thi s narrow bridge of bone i s the weakest l i nk i n the chai n, and i t i s
the part most l i kel y to fail under loadi ng. 49 A fract ure of the pars i nterarti cul ari s
i s quite common and i s called spondyl ol ysi s. Depri ved of thi s bony support ,
the lumbar col umn is now supported by l igamentous structures onl y, pri nci pal l y
the lumbosacral di sc. The L5 vertebral body may separate from i t s arch and
sl i p forward ( ol i sthesi s) due to creep, stretch, and fi ssuri ng i n the ti ssues of the
disc. This anterior di splacement i s termed spondylolisthesis. Spondyl ol ysi s i s
seen frequentl y i n sports i nvol vi ng repeti ti ve and sudden loadi ng of t he l umbo­
sacral joints i n extension. I t i s found i n L5 most often, from chi l dhood and
adolescence onward, but i t i s not found i n fetuses and i nfant s. It may be ei ther
painful or asymptomati c. There may be a devel opmental component i n i t s eti ol ­
ogy ( e. g. , a developmental l y narrow i sthmus i n the pars i nterarti culari s) , but
i t i s fundamental l y a stress fracture and i s painful i n those cases where i t occurs
as a result of an i dentifiable sporting i njury; in t hese i ndi vidual s it remai ns
painful when stressed.
Spondyl ol i sthesi s is more l i kel y to fol l ow spondyl ol ysi s in young people
( under 25 years of age) wi th compl i ant discs than i n ol der people wi th stif
di scs. Spondyl ol ysi s i s not t he onl y possi ble resul t of the part i cular stresses at
the lumbosacral joi nts and i n the lower l umbar spi ne. Degenerati ve changes i n
the lumbosacral art icular triad and i n the L4-L5 arti cul ar t riad are al so com­
mon. Intervertebral di sc rupture wi th nucl ear extrusi on i s most common i n the
lowest two intervertebral di scs ,50 and zygapophyseal joi nt art hri ti s i s al so most
common i n the l umbosacral joi nts . 5 1
Retrol i sthesi s of L4 on L5 i s al so qui te common i n a l umbar spi ne wi tl
degenerat ive changes of the mobile segment .
24 Physical Therapy of the Low Back
General Principles
The processes of development and growth i nfl uence structure, function,
and pathology of the l umbar vertebral col umn in a number of i mportant areas.
Mal formati ons resul t from abnormal genetic, chemi cal , or mechanical influ­
ences on growt h. By growth we mean measurable i ncrease in size. This i nvolves
i ncrease i n cell numbers and cell si ze, i ncreased producti on of matri x and fi bers,
or any combi nati on of these. Prenatal growth is characterized by cel l mul ti pl ica­
tion and postnatal growth pri nci pal l y, but not ent i rel y by i ncreases i n cell size
and cell product s. Development also i nvol ves differenti ati on of cel l s ( i . e. , cel l s
become more special i zed and l ess versati l e i n thei r function as they mul ti pl y) .
Connecti ve t i ssue cel l s have a self-differentiati ng capaci ty dependi ng on thei r
geneti c program, position i n t he devel opi ng embryo, and contact and i nteraction
with other cel l s or ti ssues and on local and systemi c hormonal i nfuences. Some
cel l s produce diffusi bl e chemical substances, whi ch i nfluence the development
of neighboring ti ssues. Other ti ssues i nfl uence the development of neighboring
ti ssues by mechanical pressure, but the control mechani sms for many aspects
of growth are sti l l not compl etel y understood. 52
Summary of Early Development
Before there i s any vertebral col umn, in the thi rd week of embryonic l ife
t he axi s of the flat embryonic di sc i s determi ned by t he appearance and growth
of the notochord between the ectoderm and the endoderm. The appearance
and growth of the notochord defi nes t he axi s of the flat embryonic di sc, and it
can now be described as havi ng head and tail ends. At about the same ti me,
mesoderm, t he thi rd pri mary l ayer of the embryo, develops on each si de of the
notochord, between the ectoderm and the endoderm.
The notochord i s a l ong, t hi n rod of pri mi ti ve cel l s, whi ch have the potential
to i nfl uence the devel opment of other cel l s around them. The notochord i nduces
thi ckeni ng of t he adjacent dorsal ectoderm to form the neural plate, which
grows rapidl y, fol di ng to form the neural tube. The notochord and neural tube
extend from the head to the tai l of the embryo and together they attract the
migration and condensation of mesodermal cells around them i n order to form
the original blastemal vertebral col umn. The paraxial mesoderm alongsi de the
notochord and neural t ube is origi nal l y i n the form of two conti nuous col umns
but before the blastemal vertebral col umn i s formed the paraxial mesoderm
segments i nto a large number of somites or bl ocks of mesoderm. The medial
parts of these bl ocks of mesoderm ( somi tes) on either side the notochord and
neural tube are used as bui l di ng material to form t he blastemal col umn.
The notochord, the neural tube, and t he col umns of paraxial mesoderm on
each si de of them are t he essent ial el ements for the format ion of the vertebral
The LlImbar Spine from Infancy to Old Age 25
not ochord
eural tube
Fig. 1 - 1 3. A 7-mm crown-rump length ( CRL) human embryo shows somites on the
external surface. The center secti on shows median plane axi al structures at the bl astemal
stage of vertebral devel opment . The i ntersegmental branches of the aorta pass around
the l ight bands, which are the promordia of the vertebral bodi es. Neural processes grow
around the neural tube to form the vertebral arches. The cyl i ndrical notochord ( bl ack)
passes through the centers of the l ight bands ( pri mi ti ve vertebrae) and dark bands ( pri mi ­
ti ve intervertebral di scs). ( From Taylor and Twomey,54 wi th permission. )
col umn. As the embryo grows , i t fol ds and bends ; the transverse fol di ng of the
flat disc encl oses the endoderm to form a pri mi ti ve gut tube and the l ongi tudi nal
bending forms a ventral concavity with the pri mi ti ve axial structures curved in
a bow around the gut tube. Anterior to these structures, the pri mi ti ve aorta
runs down i n the mi dl i ne with blood al ready ci rcul at i ng to t he area where the
blastemal vertebral col umn wi l l be formed ( Fi g. 1 - 1 3) .
Developmental Stages
The vertebral col umn wi l l pass through three devel opmental stages: blaste­
mal , carti l aginous , and osseous.
The blastemal or mesenchymal col umn is formed around t he notochord by
the mesoderm from the ventromedial porti ons of the somites. s3-s6 Although
formed from segmented mesoderm, this original mesodermal condensati on
around the notochord i s conti nuous and unsegmented. It resegments i nto al ter-
26 Physical Therapy of the Low Back
nate l ight and dark bands all the way along its l engt h. Neural processes grow
around the neural tube from each l ight band. The aorta, whi ch l i es i mmediately
i n front of the bl astemal col umn, sends i ntersegmental branches around the
middle of each l ight band. The l ight bands grow i n height four t i mes more
rapi dl y t han the adjacent dark bands.
Each l ight band wi th i t s associ ated neural processes di fferentiates i nto a
cartilagi nous vertebra at the begi nni ng of the fetal stage of development (2
months' gestation) . Thi s di fferenti at ion takes pl ace throughout al l the ti ssue of
the l ight band at about the same t i me. 55 There i s no evidence that there are
two centers of chondrifi cati on as descri bed by Schmorl and J unghanns. 57 The
rapid di fferent iation and growth of the fetal carti l age model s of vertebral bodies
i s accompanied by notochordal segmentation ( Fi g. 1 - 1 4) . Each notochordal
Fig. 1 - 1 4. A median sagittal section of the thoracic spine of a 7S-mm CRL ( I I th week)
human fetus shows the cartilaginous stage of vertebral col umn devel opment. At the
center of each carti l age model of a vertebral body ( v. h. ) , cal ci fied zones with hypertro­
phied chondrocytes herald the formation of centers of ossification (centr). The dark
anterior rim around the cal cified zone represents the first bone formation wi thi n the
vertebral body. Vascul ar buds are seen wi thi n the posterior parts of the cartilaginous
vertebrae. The notochord has segmented into notochordal aggregations ( n. c. ) , which
will each form a nucl eus pul posus. All that remai ns of the notochordal track through
the vertebra is the mucoid streak ( m. s. ) .
The Lumbar Spine from Infancy to Old Age 27
segment will form a nucl eus pul posus at the center of a dark, band, 3 At t he
periphery of thi s pri mi ti ve i ntervertebral di sc, fi brobl asts and col l agen bundl es
appear in lamel l ar form. The cart i laginous stage of vertebral devel opment i s a
short one. Soon, blood vessel s grow i nto t he cartilagi nous vertebra, as centers
of ossification appear.
Three primary centers of ossi fication are formed in each vertebra. Bi lateral
centers for the vertebral arch appear fi rst , one for each half arch. The earliest
vertebral-arch centers are i n the cervi cothoraci c region, but the process rapi dl y
extends up and down the col umn. The appearance of arch centers i s general l y
sequential , with the most caudal appearing l ast , except that the appearance of
midthoracic centers is delayed unti l al l l umbar centers have appeared. 58 A si ngle
primary center for each vertebral body forms the centrum. There is no evi dence
that doubl e centers appear i n normal development of the centra, but a bi lobed
appearance i s common i n vertical sections through the plane of the notochord,
due to a temporary i nhi bi tion of ossification i n the i mmedi ate vi ci ni ty of the
notochord. 3.59 The centra appear earliest near the thoracolumbar juncti on and
then appear i n sequence up and down the col umn.
The process of ossification extends through the cart i l age model of each
vertebra except for t he growth plates, which persi st to ensure conti nui ng
growt h, and the carti lage pl ates on t he upper and l ower vertebral surfaces,
which remain carti lagi nous throughout l i fe .
Dorsal-Midline Growth Plate. A si ngle growth plate' i n the mi dl i ne of the
vertebral arch, dorsal l y ( Fi g. 1 - 1 5 ) , which persists till about I year postnatal l y.
Neurocentral Growth Plates. Neurocentral growth plates persi st on each
side, between the arch and the centrum, unti l 3 to 7 years , ensuri ng growth of
the spinal canal to accommodate growth of the spi nal cord and cauda equi na.
Cartilage-Plate Growth Plates. Growth plates at the upper and l ower verte­
bral end-plates ensure growth i n height of the vertebra. These growth plates
are parts of the cart i l age plates cappi ng t he cephal i c and caudal surfaces of the
vertebral body ( Fi g. 1 - 1 6) . A ri m of bone appears in the peri phery of each
carti l age plate between 9 and 1 2 years . This "ri ng apophysi s" fuses wi th the
vertebral body at 1 8 to 20 years i n males,6
and 2 years earl i er i n femal es ( Fi g.
1 - 1 7) . An apophysi s i s comparable to an epi physi s, except that i t does not i tself
contribute to growt h. 53
Growth in Length of the Vertebral Column as a Whole
Growth is most rapid prenatal l y, and the rate of growth decreases progres­
sively throughout i nfancy and chi l dhood with a final i ncrease in growth rate
during the adolescent growth spurt . Measurements of si tti ng height at different
ages can be used to chart postnatal growth i n l ength of the spi ne . 4 1 The spi ne
28 Physical Therapy of the Low Back
Fig. I - IS. Thi s tracing of a horizontal section of a human fetal l umbar vertebra ( 28. 5-
cm CRL, 32nd week) shows the centrum, the vertebral arches, the neurocentral growth
plates (ncgp) , and the vertebral arch growth pl ate (gp) . There is onl y one center of
ossification for each side of the vertebral arch, but the plane of the section has missed
the bone of the mi ddl e part of each hal f arch, which appears as carti l age.
contributes 60 percent of si tti ng height . Si tti ng-height growth rate decl i nes from
5 cm per year in the second year of l ife, to 2. 5 cm per year at 4 years. I t remains
steady at this rate until 7 years , then decl i nes furt her to 1 . 5 cm per year just
before adolescence . The adol escent growth spurt for the spi ne begi ns at 9 years
i n femal es, l asti ng unti l 1 4 years, and peaki ng at 1 2 years with a si tti ng-height
growth vel oci ty of 4 cm per year. In males the growth spurt lasts from 1 2 to
1 7 years wi th a peak growth vel oci ty of 4 cm per year at 14 years.
The t horacol umbar spi ne grows and matures earl i er than the cervical or
sacral regions. The l umbar spi ne grows more rapi dl y t han the thoracic spi ne
before pubert y, but the thoracic spi ne grows more rapi dl y afer pubert y. Growth
in length of the thoracol umbar spine i s 60 percent more rapid i n the femal e than
i n the male between the ages of 9 and [ 3 years ( Fig. [ - ( 8) . After [ 3 years the
The Lumbar Spine from Infancy to Old Age 29
Fig. ) · )6. ( A) The intervertebral di sc
as it would appear in the median sec­
tion of a ful l -term fetus when viewed
by polarized l i ght. There i s direct con­
ti nui ty of the i nner two-thirds of the
annul us with the l amel l ar structure of
the carti lage plates. (8) The lower di sc
as seen by normal transmitted light
shows the outli nes of the blood vessel s
suppl yi ng the di sc. The angul ar i nden­
tations of the carti lage plates from the
nucl eus pul posus i ndicate where the
notochordal track origi nal l y passed
through the col umn. The growth plates
of the vertebral end-pl ates are seen.
The black area i n t he posterocentral
vertebral body is the basivertebral
vein. ( From Taylor and Twomey,54
with permission. )
male spi ne grows more rapi dl y. Sitti ng height and thoracol umbar spi ne length
both reach 99 percent of their maxi mum by 1 5 years in girls and 1 7 years in
boys. Ri sser' s sign , i nvol vi ng the l ateral appearance, medial excursion, and
fusion of the i l iac crest apophyses, is somewhat arbitrari l y used tojudge i ndi vid­
ual completion of spi nal growth, which takes about 2 years from frst appear­
ance to fusion of the apophysi s. 4
1 At compl eti on of growth, the spi ne usual l y
forms a greater proporti on of the total stature i n femal es than i n males .
Control Mechanisms in Normal and Abnormal
Notochordal, Neural, Vascular, and Mechanical Infuences
A summary of normal and abnormal aspects of devel opment fol l ows.
I . The notochord and neural tube i nduce formation of the bl astemal verte­
bral column around them from the mesenchyme of the adjacent somi tes. 3
2. Segmentation of t he bl astemal col umn i s probabl y determi ned by the
regul ar arrangement of i ntersegmental arteri es wi t hi n i t . 61
3 . The notochord forms the original nucl eus pul posus; after rapid growth
30 Physical Therapy of the Low Back
Fig. 1 - 1 7. ( A&B) The supe­
rior and anterior surfaces of a
juveni l e thoracic vertebrae
show the radi al grooves due to
the vascul ar canal s entering the
carti lage pl ates ( Fig. B). (C)
The lines of neurocentral fu­
si on can al so be seen. A ful l y
grown vertebra shows t he ring
apophysi s fusi ng with the cen­
trum. Small vascular foramina
pass between the apophysi s
and the centrum. ( From Taylor
and Twomey, ´ with per­
mi ssion. )
. 350
: 300
The Lumbar Spine from Infancy to Old Age 3 1
• Males
o Females
5 6 7 8 Ü 10 1 1 12 13 14 15 16 17 18 Adult
Age (years)
Fig. 1 - 1 8. Cross-sectional data for mal e and female growth i n thoracolumbar spi ne
l ength based on measurements i n 1 500 subjects.
of a notochordal nucl eus pul posus i n fetuses and i nfants, notochordal ti ssue
atrophies and di sappears duri ng chi l dhood. 3
4. Persistence of l i ve notochord cel l s i n vertebrae may lead to t he forma­
tion of chordomas in adul t s. 3 These rare, mal ignant t umors are usual l y seen i n
a high retropharyngeal or sacrococcygeal si tuati on.
S. Anomalous vertebral devel opment , such as bl ock vertebrae, butterfl y
vertebra, or hemi vertebra, may resul t from abnormal devel opment of the noto­
chord or the segmental blood vessel s . 41
6. When the notochordal track and vascul ar canals di sappear from the
carti lage pl ates of devel opi ng vertebrae, they leave weak areas , whi ch are t he
sites of potential prolapse of di sc material 54 i nto vertebral bodi es ( Schmorl ' s
nodes) .
7. Growth of the spi nal cord and cauda equi na i nfuences growth of the
vertebral arches and canal , just as brain growth i nfl uences skul l-vaul t growth.
An enl arged spi nal cord resul ts i n an enl arged canal . 62 Spi na bifida i s a devel op­
mental anomal y, whi ch varies from si mpl e spl i tti ng of the skeletal el ements of
the vertebral arch ( spi na bifida occul ta) , whi ch is common and i nnocuous, to
complete spl i tt i ng of ski n, vertebral arch, and underl yi ng neural tube ( ra­
chi schi si s) wi th associated neurologic defici t s. In spinal bifida, abnormal devel ­
opment of the neural tube i s probably the primary event and the skel etal defects
are secondary.
32 Physical Therapy of the Low Back
8. Asymmetri c growth of right and left halves of the vertebral arches is
very common in normal i ndi vidual s . It produces sl ight rotation of t he anterior
el ements to the left side in i nfancy and to the right side in adolescence. This
probably determi nes the lateral i ty (directi on of curvature) of scol i osi s in both
its physiologic and progressi ve forms . 58
9. Lumbar zygapophyseal joi nt facets grow backward from t hei r lateral
margi ns, and change from pl anar coronal l y oriented facets i n i nfant s, to bi planar
facets wi th coronal and sagittal components in older chi l dren and adult s. The
coronal and sagittal components of the joi nt facets have di fferent functi ons i n
control l i ng movements . 6.4 1
1 0. When infants assume erect posture, there i s an i ncrease in the l umbosa­
cral angl e, i ncreased lordosi s, changes in the shape of i ntervertebral di scs and
the posi ti on of the nucl eus pul posus. The vertebral bodies i ncrease thei r antero­
posterior growth rate, and the vertebral end-plates change thei r shape. These
vertebral end-plates are convex i n i nfants , but they become concave in chi ldren
and adults . The end-plate concavi ty i s opposite t he maxi mum bulge of the
nucl eus . I
63 Leg-length di screpancy and pel vi c obl i qui ty are associated with
asymmetric concavi ti es in the vertebral end-plates.
I I . Sexual di morphi sm i n vertebral-body shape and spi nal posture devel­
ops i n chi ldhood and adolescence i n association with diferent hormonal i nfu­
ences and differences i n muscl e development . 64 These di fferences contribute
to the greater prevalence of progressi ve scol i osi s i n females than i n males.
1 2. Scheuermann' s di sease, wi t h irregularity of the vertebral end-plates i n
adol escence, i s associ ated wi th t he development of mUl ti ple Schmorl ' s
nodes. 4 1 . 65
Segmentation, Segmental, and Other Vertebral Anomalies
Normal Segmentation. A condensation of mesenchyme is formed around
the notochord by the medial mi grati on of cel l s from the scl erotomes, or ventro­
medial porti ons of the somi tes . Al t hough the perichordal blastemal column is
formed from segmented bl ocks of mesoderm, i t is i tsel f conti nuous and unseg­
mented. It resegments i nto alternate l ight and dark bands in such a way that
t he l ight bands , formi ng carti l agi nous vertebrae, are at the level of the interseg­
mental branches of the aorta ( see Fi g. 1 - 1 3) . The muscl es, derived from the
myotomes of the somi tes, bridge over the dark bands t hat form the di scs, and
are attached to upper and lower vertebral borders . The alteration of muscle
and bone i s essenti al to the proper function of t he locomotor system. 55
The i ntersegmental branches of t he dorsal aorta have an i mportant i nfu­
ence i n vertebral -col umn resegmentation by vi rtue of their placement around
the centers of the light bands, where they provi de nutri ti on for the more rapid
growth of the pri mi ti ve vertebrae. 4
They are the onl y constant and regularly
recurring structures in the bl astemal vertebral col umn, and vascular anomal ies
may resul t i n anomal i es of segmentati on. 6 1
Segmental Anomalies. A uni lateral hemi vertebra resul ts if one side of the
vertebral body fai l s to devel op. Normal l y there i s onl y one primary center of
The Lllmbar Spine from Infancy to Old Age 33
ossification for the centrum, and i t is l i kel y t hat the anomaly origi nates duri ng
a preosseous stage of development . Absence of an i ntersegmental vessel on
one si de may gi ve ri se to a uni l ateral hemi vertebra, whi ch wi l l be associ ated
with a sharpl y angled congeni tal scol i osi s after assumpti on of an erect post ure .
Absence of the anterior part of a vertebral body i s al so termed hemi vertebra.
The origin of this anomaly i s di ffi cul t to explai n. I t gi ves rise to a congeni tal
kyphosis ( Fi g. 1 - \ 9) .
As the l ight bands grow rapidl y, they appear t o expel notochordal ti ssue
into the more sl owl y growing i ntervertebral di scs . Each notochordal segment
forms a nucl eus pul posus. Notochordal t i ssue grows rapidl y by cell mul ti pl ica­
tion and product ion of mucoid matri x at the center of the fetal di sc . Absence
of a notochordal segment may cause the centra to fuse, formi ng congeni tal
block vertebrae, because no nucl eus pul posus forms to separate t hem.
Later Notochordal Development, Chordoma, and Butterfy Vertebra. The
i nvasive nature of notochordal cel l s assi sts the rapid growth of the notochordal
nucleus pulposus i n the fetus and infant . 3 The notochordal cel l s produce sub­
stances that loosen and digest the i nner margi ns of the surroundi ng envel ope,
incorporati ng these ti ssues i nto the expandi ng nucleus . The t i ssues of the enve­
lope (annulus and carti lage pl ates) must grow rapi dl y just to keep pace with the
"erosi on" of thei r i nner margins by the rapi dl y expandi ng nucl eus. Notochordal
Butterfly Verebra
Fig. 1 - 1 9. ( A) A tracing of vertebral outl i nes from an anteroposterior radiograph shows
the appearance of a uni l ateral hemi vertebra, wi th consequent angled scol i osi s. (B) A
tracing from an anteroposterior radiograph shows a butterfly vertebra. Soft ti ssues are
continuolls through the vertebral body from one i ntervertebral di sc to the next.
34 Physical Therapy of the Low Back
cel l s do not normal l y survi ve beyond earl y chi l dhood-except perhaps deeply
buried in the devel opi ng sacrum or possi bl y in Schmorl ' s nodes. Their fortunate
demi se during chi l dhood i s associated wi th the progressi ve decrease i n vascular­
ity of the surroundi ng ti ssues. If notochordal cel l s do survi ve , they may be
"rel eased" by t rauma to the contai ni ng t i ssues, and begi n to mul ti pl y agai n.
The cel l s of mal ignant chordomas strongl y resemble embryonic notochordal
cel l s. 3
The mucoi d streak persi sts for a whi l e i n t he cartilage models of the verte­
brae as an acel l ul ar notochordal track, but when ossificat i on of the centrum
commences, i t is usual l y obl i terated. The mucoid streak has a temporary i nhi bit­
ing effect on ossification of the centrum, which may have a bi lobed appearance
in vertical secti ons t hrough the plane of the notochord. Persistence of parts of
the notochordal track t hrough the centrum is quite common in i nfancy, 3 but i s
probabl y rare after i nfancy. If a compl ete notochordal track persi sts, a butterfl y
vertebra is the result ( see Fi g. 1 - 1 9) . Persistence of the notochordal track i n
infancy produces a mi sshapen vertebra wi th a ni pple-shaped deformity ( Fi g. 1 -
20) of the vertebral end-plates . Thi s may be responsi bl e for the "Cupi d' s bow"
appearance frequent l y seen in lower l umbar vertebral bodi es i n adul t s. 5
Other Vascular and Notochordal Infuences
Disappearance of Notochordal Cells from the Nucleus Pulposus. The carti­
lage pl ates of i nfants have an excel l ent blood suppl y from the adjacent vertebral
periosteum. Thi s brings nutri ti on to the rapi dl y growing i ntervertebral di sc. 3 1
The capi l l ary pl exuses of the vascul ar arcades approach cl ose t o the growing
notochordal nucl eus pul posus ( see Fig. l - 1 6B) . When t he disc mass grows and
its vessel s di sappear duri ng chi l dhood, the notochordal cel l s die off and are
replaced by chondrocytes , cel l s better adapted to an avascul ar envi ronment .
Schmorl's Nodes. The vascul ar canals are pl ugged by loose connective
ti ssue, l eavi ng channels of reduced resi stance from near the nucl eus to the
peripheral vertebral spongiosa. 54 These channel s, archi ng around the advancing
ossification front of the centrum, i nhi bi t ossi fi cati on local l y, causi ng a toothed
or grooved surface on the vertebral end surfces of adolescents ( see Fig. 1 - 1 7
A & B) .
The cart i l age plates al so have a consi stentl y sit uated funnel -shaped defect
on thei r nuclear aspect , where the notochordal track formerl y penetrated the
col umn, just behind the center of each vertebral end-plate ( see Fig. 1 - 1 6) . These
notochordal weak poi nts are the si tes of central Schmorl ' s nodes and the vascu­
lar channel s are probably the si tes of peripheral Schmorl ' s nodes, either i n the
anterior spongiosa or formi ng a l i mbus vertebra ( Fi g. 1 -2 1 ) . They occur almost
as frequentl y in adolescents and young adults as i n ol der adul ts ,57 i ndi cati ng
thei r association wi t h devel opmental factors.
Schmorl ' s nodes occur i n thoraci c and l umbar vertebrae of 38 percent of
adult spi nes. The central nodes are an i nci dental fi ndi ng, and may not cause
The Lumbar Spine from Infancy to Old Age 35
Fig. 1 -20. A montage of a median sagittal section of three thoracic vertebrae and two
discs i n a I - month-old i nfant demonstrates conti nui ty of the notochord track through
the thoracic vertebral bodi es. The notochordal track normal l y disappears from the verte­
brae at about 20 weeks' gestation. The track may temporarily i nhi bit ossifi cati on, but
i n this case the deformities of the vertebral end-pl ates appear to resul t from attempts
by the bone to "grow around" it. The nucl eus pul posus i s bi l ocul ar because of the bony
deformi ty. ( From Taylor,59 with permission. )
36 Physical Therapy of the Low Back
l i mbus
Fig. 1 -21 . Varieties of Schmorl ' s nodes ( i ntraspongious di sc prolapse) . Central nodes
are very common and occur through weaknesses i n the cartilage plates l eft by the noto­
chordal track ( see Fig. 1 - 1 6) . Peripheral nodes are less frequent and occur along t he
l i nes of vascul ar canal s. Bl ood vessel s atrophy duri ng maturation and the vascular canals
are pl ugged by loose connective ti ssue ( see Figs. 1 - 1 6 and 1 - 1 7) . Disc prolapse between
the centrum and the ring apophysi s i s termed a limbus vertebra. Vascular canal s are
more pl enti ful anteriorl y, and peripheral pro l apses are usual l y anterior. Not e: Central
nodes are sl i ghtl y posterior of center. ( Modified from Tayl or and Twomey,54 wi th per­
mi ssi on. )
any symptoms, but the anterior nodes may be associated wi th a traumatic inci­
dent, and the pati ent may compl ai n of localized somatic back pai n. 41
Scheuermann's Disease. Mul ti pl e Schmorl ' s nodes, both large and smal l ,
are seen i n Scheuermann' s juveni l e kyphosi s, wi t h a radiologic appearance of
irregul ari ty of the vertebral end-pl ates. Large anterior nodes are frequentl y
seen, and are someti mes associ ated wi th anterior vertebral body col l apse and
wedging. The cause of the mul ti pl e Schmorl ' s nodes i s uncertai n, but one sug­
gesti on is that there may be an associ ated osteoporosi s predi sposing to vertebral
end-plate weakness. 67 Al ternati vel y, an "abnormal l y" vascul ar end-plate
would have a correspondingly l arge number of "weak poi nts" on attrition of
the vessel s , predi sposi ng to a larger number of Schmorl ' s nodes than usual .
We have noted an i ncreased vascul ari ty of vertebral end-plates contain
Schmorl ' s nodes. 65
Scoliosis and Growth
When ossifcation centers appear i n the fetal thoraci c vertebral arches,
those on the ri ght commonl y appear before the correspondi ng left centers. Tay­
lor68 advanced the hypothesi s that in the fetal ci rcul at i on, the better oxygenated
The Lumbar Spine fom Infancy to Old Age 37
blood from the left ventricle suppl ied the right mid thoraci c vertebral arches,
and the l ess wel l oxygenated blood from the right ventri cl e, t hrough the ductus
arteriosus, suppl i ed the l eft vertebral arches. Thi s woul d occur t hrough sl ow
mi xi ng of the two streams of blood from the proxi mal aorta and the ductus
This coul d expl ai n why right arch centers often appear before correspond­
i ng left arch centers i n mi dthoracic vertebrae. 58
68 Thi s asynchrony in appear­
ance of paired ossification centers correlates with measured asymmetry in i n­
fant vertebral arches ( right greater t han left ) , and wi th t he observation t hat
i nfant i l e thoracic scoliosi s tends to be convex to the l eft ( Fi g. 1 -22) .
There i s evi dence of asynchronous maturation and growth of vertebral
arches ( right greater than left) i n fet uses and i nfants persi sti ng unti l cl osure of
the neurocentral growth plates at 6 to 7 years , and of a reversal of thi s asyn­
chrony and asymmetry i n thoraci c vertebral arches in ol der chi l dren (l eft greater
than right) after cl osure of t he neurocentral growth pl ates.
I n older chi l dren and adolescents t he left arches are larger than the corre­
sponding right arches . 51 The consequent reversal of the twi st of the mi dthoracic
anterior elements from 7 years onward changes the di recti on of any spi nal
curvature so that it becomes convex to the right in t he mi dthoraci c spi ne. Thi s
change i s related to asynchrony of cl osure of the neurocentral growth pl ates.
At the same ti me, left-sided fl atteni ng of the anterior surfaces of mi dthoraci c
vertebrae t,egi ns to appear i n older chi l dren, where the vertebral bodi es are i n
contact with the aorta. The aortic pressure t hat flattens the anteri or l eft surfaces
of thoracic vertebral bodies also suppl i es the force that twi sts t hese vertebral
bodies to the right . 4
1 .58. 68 The di fferent position of the l umbar aorta on the
anterior sUces of L2 to L4 and the di fferent shape of the l umbar vertebral
bodies may be related to the observation that sl ight twi st ing of vertebral bodi es
0-5 years 6-7 years 7- 12 years
A B c
Fig. 1 -22. Diagram representing the normal asynchrony in growth of thoracic verte­
brae. ( A) I n later fetal and infant l ife, the right neural arch general l y matures earl i er
than the corresponding l ef arch and therefore i s sl ightl y l arger. ( 8&C) At neurocentra I
closure thi s symmetry is "reversed" by earl i er cl osure on the right ( see text ) . ( From
Taylor and Twomey, 41 with permission. )
38 Physical Therapy of the Low Back
to the l eft is more common here. 68.69 The asymmetri es described are mi nor i n
nat ure. They accompany normal growth and probably determine t he side of
scol i osi s, whether it be physi ol ogic or progressi ve, but they are not by them­
sel ves the cause of progressi ve scol i osi s . Other causes al so operate in the mul ti ­
factorial etiology of structural scoli oses.
Growth of Lumbar Zygapophyseal Joints
Lumbar zygapophyseal joi nt s, l i ke thoraci c facet s, are oriented in a coronal
pl ane in fetuses and i nfants, but from late i nfancy onward, the l umbar articular
facets grow in a posterior di rection from thei r l ateral margi ns, changing their
shape and ori entati on. The original coronal l y oriented part remai ns, but a sagit­
tal component is added to the joi nt by posterior growth from i t s lateral margi ns
( Fi g. 1 -23) . In the adul t joi nt s, from L I to L2 to L3 to L4, the posterior two­
t hi rds of the joi nt is approxi matel y sagittal and the anterior thi rd i s approxi­
matel y coronal ( see Fig. 1 - 1 1 ) . Remodel i ng at the junction of t he two compo­
nents usual l y makes the joi nts surfaces curved. In transverse sections or com­
puted tomography (CT) scans the concave superi or arti cular facet i s seen to
"embrace" the smal l er convex i nferior facet ( see Fig. 1 - 1 0) . Both curved articu­
l ar surfaces are i ncl i ned approximatel y paral l el to the long axi s of the l umbar
spi ne so that they resemble segments of cyl i nders . The lower l umbar joi nts are
more nearl y coronal in thei r overal l orientation than the upper joints and tend
to be flatter or less curved than the upper joi nts. 6
43. 7
The coronal and sagittal components of the arti cul ar sUlfaces relate to two
different functi ons: restraint of fexi on and restraint of axial rotation, respec­
t i vel y. The subchondral bone pl ate (SCP) of the superior arti cular facet becomes
much thi cker in i t s anterior coronal part than i n t he sagittal part , developing
a wedge shape in transverse secti on, whi ch refl ects the greater physi ologic
compressi ve loading in the coronal part of the joi nt . 6 The SCP and arti cular
cart i l age reach thei r maxi mum thi ckness i n young adult l ife. The adult hyal i ne
carti l age i s about I mm thi ck on each facet. It has a very smooth surface and
the matrix and cel l s stai n l i ghtl y and evenl y i n the heal thy young adult joint .
It i s joi ned to the underl yi ng SCP by a thi n, regular calcified layer. In the
growing joint the subchondral bone is very vascul ar, but wi th maturation thi s
vascul ari t y decl i nes.
The arti cul ar cart i l age i s avascul ar and has no nerves , so that it i s i nsensi­
tive, except at i ts peri phery, where i t may be i n cont i nui ty wi th the wel l -i nner­
vated joi nt capsul e. It recei ves its nutrition from t he synovial fl uid that bathes
it, the ci rcul ati on of the fl ui d bei ng aided by movements of the joi nt. From the
fourt h decade onward, there are changes i n the stai ni ng characteri sti cs of the
arti cular carti l age of the coronal component of the joi nt , wi th hypertrophy of
chondrocytes and i ncreased i ntensi ty of stai ning of the matri x, especial l y i n
the mid-zone, suggesti ng a response to high compressi ve loadi ng i n these parts
of the joi nt s, probabl y in fexi on. 6 The cel l ul ar changes eventual l y change the
Caneelu. bone
Compact bone
Artiulr eartlge
The Lumbar Spine from Infancy 10 Old Age 39
Fig. 1 -23. Tracings from horizontal sections of three L4-L5 zygopophyseal joi nts at
I , 8, and 1 5 years show the "rotation" of the joint plane from the coronal pl ane, toward
the sagittal plane during growt h. The concave anterior facet is on the superior arti cul ar
process of the lower vertebra. The convex facet is on the inferior arti cul ar process of
the upper vertebra. Note that the anteromedial capsul e is formed by the l igamentum
tlavum (LF), and that mul tifidus (M) partly insets i nto t he posterior fi brous capsul e
(Caps). Articul ar cartilage i s seen (stippled) , and deep to it the subchondral bone pl ate
(Ivhite) becomes thi cker with growth and i s general l y wedge-shaped and thus thi cker
anteriorly in the concave facet .
40 Physical Therapy of the Low Back
nature of the matrix so that spl i tti ng of the carti l age occurs, along l i nes parallel
to the col l agen framework. Thi s is characteri sti c of chondromalaci a.
Growth of Vertebral Bodies and Intervertebral Discs
Shape Changes on Assumption of Erect Posture
There are marked changes in the shape of l umbar vertebral bodies after
i nfancy. These are due to the changed mechanical forces on the spine with
erect posture and are closel y related to a change i n the position of the growing
nucl eus pul posus as lordosi s is establ i shed. 7
Fetal and infant l umbar vertebral
bodies have convex upper and lower surfaces, so that the i ntervertebral di scs
are bi concave. In the fet us and smal l i nfant, the pri mary spi nal curve i s main­
tained and each l umbar nucl eus pulposus appears wedge shaped i n sagittal
secti on, wi th its main mass si tuated posteriorl y. As the i nfant si ts up, stands ,
and learns t o wal k, the secondary l umbar curvature appears , the di sc changes
i ts shape, and the nucl eus pul posus moves forward to a central posi ti on. The
weight-bearing vertebral end-plates gradual l y change thei r shape from convex
to concave during chi l dhood, the concavi ti es appearing opposite the maxi mum
bulge of the nucl eus pulposus. 3 . 63
Other i mportant changes in l umbar spi nal shape fol l owi ng assumption of
the erect posture are the relati ve i ncreases i n the anteroposterior and transverse
di mensi ons of vertebral bodi es and di scs. These changes i n the predominant
di rection of growth give the l umbar spine more stabi l i ty in the sagittal and
coronal planes . We have studi ed the i ncreased anteroposterior growth of the
l umbar vertebral bodies and i ntervertebral di scs. The changes i n vertebral­
body shape fol l owi ng weightbearing in the erect posture reflect the plast icity
of vertebral bone and i ts ready response to mechanical forces at this age. I n
nonambul atory chi l dren, these normal growth changes of early chi l dhood do
not appear. Such chi l dren have very square vertebrae on l ateral radiographs,
without t he normal concavi ti es i n t he endplates and wi t h relati vel y short antero­
posterior di mensi ons ( Fi g. 1 -24) . Thi s abnormal shape is due to decreased hori-
_¸-¬ :. ¬¬
:: \
\ I
, I
non-ambulant ( 1 4 year female)
ambulant ( 1 5 year female)
Fig. 1 -24. Tracings from lateral radio­
graphs show the effect of weightbearing
on growth. In the absence of normal
weightbearing in erect posture, no end­
plate concavity appears and anteropos­
terior vertebral body growth is severely
retarded. ( From Taylor,63 with per­
mission. )
The Lllmbar Spine from Infancy to Old Age 4 1
zontal growth rather than to any i ncrease i n vertical growth. 72 A si mi l ar square
shape, but from a different cause, is seen when the posterior surfaces of verte­
bral bodi es are scal loped by resorption i n the presence of a t umor i n t he verte­
bral canal .
Sexual Dimorphism in Vertebral Body Shape
From the age of 8 or 9 years onward, further i nteresti ng differences i n
shape appear i n normal l umbar vertebrae i n mal es and females. 64 Female verte­
brae grow i n height more rapidl y t han male vertebrae, gi vi ng a more sl ender
vertebral col umn. Mal e vertebrae grow more i n both t ransverse and anteropos­
terior di mensi ons than female vertebrae throughout the whole adolescent
growth period, and appear on radiographs wi der and more squat than female
vertebrae ( Fi g. 1 -25) . 64 A measurement survey of a growing popul ation showed
that the thoracol umbar spine grows i n height 1 . 7 t i mes more rapi dl y i n femal es
than mal es bet ween the ages of 9 and 13 years . 3
After the age of 14 years, the
male spi ne grows i n l ength more rapi dl y than the female spi ne, but the greater
transverse growth of male vertebrae mai ntai ns the shape di fference descri bed.
Accompanying these di fferences in vertebral growth patterns in mal es and fe­
males are differences in muscular support , si nce the effect of testosterone on
muscle is to i ncrease both i ts bul k and i ts strength-per-uni t cross-sectional area.
Thus the broader or t hicker mal e vertebral col umn al so has better muscul ar
support than the average female col umn. When axi al l y loaded, an average sl en­
der female col umn wi l l buckl e more easi l y than a wi de mal e col umn, part icul arl y
if i t already has a sl ight physi ologic scol i osi s. Thi s may expl ai n the greater
tendency to progressi on of scoliosi s i n females than in mal es . 64
The i ncreased anteroposterior vertebral growth we have observed in both
sexes i n earl y chi l dhood, and the greater i ncrease i n transverse vertebral growth
i n males than i n females around pubert y, relate to mechani cal i nfl uences on
bone growt h. In the fi rst case, the horizontal growth i s t he resul t of the new
Fig. 1 -25. The appearance of sex differ­
ences in vertebral-body shape is shown in
tracings from anteroposterior radiographs of
normal vertebrae. Mal e vertebral bodies
grow more in width than femal e vertebrae,
and female vertebral bodies tend to be rela­
tively tal l er than male vertebrae. ( From
Taylor and Twomey,54 with permission. )
Male 1 2 years
Female 12 years
42 Physical Therapy of the Low Back
-Young adult outline
- --- Elderly adult outline
Fig. 1 -26. Di agram based on tracings of vertebral
bodies and on measurements of vertebral bodies and
i ntervertebral di scs shows the changes i n their shapes
with age. " Bal l ooni ng" of the central part of the disc
i s related to col l apse of the vertebral end-plates, and
there i s some i ncrease in anteroposterior di mension
wi th aging. ( From Twomey and Taylor,SO with per­
mi ssion. )
forces of wei ghtbearing i n erect posture, and i n the second case i t i s probably
related to di fferences i n muscl e acti vi ty on growing vertebrae. There may also
be hormonal i nfl uences independent of muscl e act i vi ty . It i s i nteresti ng to com­
pare these growth changes in shape to the aging changes in shape i n el derly
adul t s, where there is "thi ckeni ng of the wai st " of vertebral bodi es, and loss
of the sex di ference i n shape. 4. 73 Thi s is associated wi th reduced mechanical
l oading (and reduced response of osteoblasts to mechanical loads) , and reduced
estrogen l evel s in postmenopausal women ( Fi g. 1 -26) .
Anterior Elements
Vertebral Bodies
Measurement studi es on large series of l umbar vertebral col umns of all
ages4. 5.73 .74 demonstrated that the length of the column decreases wi th aging,
as expected, but the reason for this i s not usual l y loss i n di sc height , but gradual
col lapse of vertebral end-plates , associated with reduced bone densit y. The
primary osseous change is l oss of horizontal trabeculae i n the vertebral bodi es.
These horizontal trabeculae form cross-ti es, bi ndi ng together and stiffeni ng the
verti cal trabeculae, whi ch act like weight-bearing "beams. "
The l oss of the cross-ti es leads to buckl ing and fracturi ng of the unsup­
ported verti cal trabeculae or beams of bone, which hold up the vertebral end­
pl at es. Measurements demonstrate a gradual i ncrease i n vertebral end-plate
concavi ty wi th agi ng. Thi s appears earl i er i n females than i n mal es. In hemi­
sec ted spi nes, the i ncreased bowing of the vertebral end-plates i s accompanied
The Lumbar Spine from Infancy to Old Age 43
by "ballooning" of t he center of each di sc i nto t he adjacent vertebral bodi es.
The tradi tional assumption t hat di scs general l y get thi nner wi th aging i s i ncor­
rect . Measurements of average di sc thi ckness and mass i n a large series of
l umbar spi nes show i ncreases in central di sc thi ckness and di sc mass in the
majori ty of old l umbar spi nes. 74
There are also age changes in the horizontal di mensi ons of the vertebral
bodi es, characterized as a "thickeni ng of the wai st" ( see Fig. 1 -26) . These late
changes i n vertebral shape gradual l y el i mi nate the sex differences i n vertebral­
body shape that are present from adolescence to mi ddl e age. Female vertebrae
are relati vel y taller and more sl ender duri ng thi s chi l d-bearing period. The ef­
fects of testosterone on muscl e may provi de a mechani sm for the development
of broader vertebrae i n adult males, as the pul l of muscl e attachments affects
bone shape. Other hormonal i nfl uences , such as the fai rl y dramati c est rogen
loss at the female menopause, lead to the earl ier l oss in bone densi ty in aging
females . However, i t i s unwi se to regard bone loss with aging as i nevi tabl e.
Some of i t may be di ffi cul t t o avoid, but as noted i n Chapter 2, regular exerci se,
adequate cal ci um intake, and possi bl y hormone replacement therapy, can re­
duce or prevent some of the bone l oss associ ated wi th agi ng. 75
Intervertebral Discs
Contrary to popular bel ief, aging l umbar di scs do not , in the majori ty of
peopl e, become thi nner and bulge l i ke underi nflated car ti res. A mi nority of
the populati on shows di sc thi nni ng and degeneration in one or more l umbar
di scs with aging, but the majori ty of di scs, in a majority of i ndi vi dual s, mai ntai n
or even i ncrease thei r average thi ckness, due to thei r central expansi on, i n an
osteoporotic spi ne. When t hi nni ng and degeneration are found, they are seen
most often in the L4-L5 and L5-S I di scs. In subjects over 60 years of age, 30
percent of the L4-L5 and L5-S I di scs are cl assified as Rolander grade 3 . 4. 74
Aging of intervertebral di scs i s general l y associated with a reduct ion in water
content, parti cul arl y i n the nucl eus pul posus, but t he greater part of t hi s loss
occurs during maturation rather than i n ol d age. The nucl eus becomes l ess
wel l differentiated from the annul us bot h i n i ts water content and i t s hi stologic
struct ure. 3 There are i ncreases i n the absolute amounts of col lagen i n the nu­
cl eus pulposus and changes in t he types of col l agen present . There i s al so an
increase i n the ratio of kerati n sulfate ( KS) to chondroi ti n sulfate ( CS) i n the
di sc duri ng chi l dhood growth and maturation, i n parallel wi th the decreased
water-bi ndi ng capaci ty of the di sc. The progressi ve substi tuti on of KS for CS
takes place i n chi l dhood when growth of the disc i s accompanied by a di sappear­
ance of its blood vessel s, and not in ol d age. In condi ti ons of oxygen l ack, KS
acts as a functi onal substi tute for CS in mai ntai ni ng the turgor and water content
of the di sc. 37 The degenerative changes observed i n l umbar di scs wi th agi ng
incl ude the appearance of fi ssures in the annul us, fi rst ci rcumferential and later
radial fi ssures. The nucl eus has usual l y, by this stage , lost some of its fl ui di ty,
so that i t does not "flow" i nto the fi ssures. However, i f fi ssures or tears occur
44 Physical Therapy of the Low Back
in relati vel y young di scs , wi th a high turgor, t he nucl eus may be forced out
through the fi ssure, most often i n a posterolateral di rection. Di sc rupture with
extrusi on of its nucl eus is l ess frequent than was once supposed. Di sc aging
is general l y accompanied by i ncreased di sc sti ffness and decreased ranges of
movement , whi ch is ful l y di scussed in Chapter 2.
Kissing Spines
Wi th the col lapse of vertebral end-plates al l owi ng the di scs to "si nk into"
shortened vertebr
l bodi es, t here is shorteni ng of the col umn, and spi nous pro­
cesses, whi ch were formerly wel l spaced, may come i nto contact with each
other. Hi stologic studi es on four l umbar spi nes from el derl y cadavers showing
radiologic evidence of "ki ssi ng spi nes" revealed adventi ti ous joi nts with a fi­
brocart i l agi nous coveri ng on the bone and a bursal i ke cavi ty , surrounded by
fat and l i ned by synovi al membrane, between the "ki ssi ng spi nes" (Taylor
1 984, unpubl i shed data) . In the view of Sartori s et al ,7
"ki ssi ng spi nes" are
attributable to i ncreased lordosi s wi th agi ng.
Zygapophyseal Joints
Articular Surfaces and Subchondral Bone Plate
The geometry of these bi planar joi nts at different l umbar l evel s has already
been described. Age changes in young and mi ddl e-aged adults are described
fi rst . 6.43
Coronal Component (Anteromedial Third of Joint ) . The loadi ng stress im­
posed on the coronal components of the joi nts, by the tendency to forward
translational during flexi on, i s refl ected by changes in the arti cular carti lage
and i ts support i ng SCPo In normal adolescents and young adults a thi cker SCP
devel ops i n the coronal component of the concave superior arti cul ar facet .
Later, in mi ddl e l ife, thi s subchondral bone in the coronal component of the
concave facet shows further thi ckeni ng and an i ntense hematoxyl i n stai ni ng
suggesti ve of scl erosi s. In young adul ts between 30 and 40 years old, the arti cu­
l ar cart i l age l i ni ng thi s thi cker part of the SCP i n the anteromedial thi rd of t he
joi nt generall y shows cel l hypertrophy and i ncreased stai ni ng of chondrocytes
and thei r peri cel l ul ar matri x, whi ch begi n in the mid-zone of the arti cular carti­
lage of the concave facet . These arti cular carti l age changes occur i n t he concave
facet fi rst , anteriorly and at t he deepest part of its concavi ty. They occur in
the coronal component of the convex facet soon afterward. General l y, they do
not affect the sagittal component of either facet . They appear i n the fourth
The Lumbar Spine from Infancy to Old Age 45
decade of life and often progress to swel l i ng and vert i cal spl i tti ng (chondroma­
lacia) of the carti l age ( Fi g. 1 -27) .
These changes i n bone and cart i l age appear to be reacti ons to compressi ve
loading of the coronal components of the joi nt . Thi s woul d occur i n flexi on, as
the convex i nferior arti cul ar process ( lAP) gl i des upward i n the concavi ty of
the superi or arti cular process ( SAP) . The resi stance offered by the SAP to
forward translation by the lAP, whi ch hooks down behi nd i t from the vertebra
above, would create this compressi ve loading. I n many respects the changes
described are analogous to those of patel lofemoral chondromalaci a, where the
stress al so i nvol ves gl i di ng movement accompanied by compressi on of t he con­
vex patella agai nst the concave trochl ea of the femur.
Sagittal Component (Posterior Two-Thirds of Joint). The age changes i n
the posterior, sagi ttal l y oriented two-thirds of t he zygapophyseal joi nts are qui te
different i n character, and tend to occur at a l ater stage than t he changes de­
scribed i n the coronal component of t he art icul ar facets. The sagi ttal l y oriented
parts of the SCPs are relati vel y t hi n, suggesti ng l ess compressi ve loadi ng. How­
ever, i n middle-aged or elderly adul t s, the sagittal components of a number of
joints show spl i tti ng of the art i cular carti lage near the posterior joint margi n,
parallel to the art i cul ar carti l age-SCP i nterface. The directi on of the tear i s
related to the force produci ng i t . Thi s i s condi ti oned by the pecul i ar anatomy
at the posterior joi nt margi n, where we have frequentl y observed a di rect attach­
ment of the posterior fi brous capsul e to t he posterior margin of t he art i cular
carti lage. Thi s capsule-carti lage conti nui ty i s seen most often at the posterior
margin of the superior arti cul ar facet . The potenti al for bi omechani cal mi schief
i s accentuated by the part ial i nsertion of mul ti fi dus through the posterior cap­
sule ( see Fig. 1 -27) . Tension transmitted to the carti lage from the capsul e, and
frequentl y repeated, may shear the arti cul ar carti l age from t he subchondral
bone. Even movement wi thout muscle acti vi ty could exert this tensi on. The
acti vi ty of axial rotation around an axi s i n the posterior disc would exert tensi on
i n the posterior capsul e on one si de and compressi on of t he art icul ar carti lage
on the other si de, part icul arl y i n a motion segment that had lost its ti ghtness
and i ts original mechanical effi ci ency. Some sl ight forward or backward sl i di ng
of one facet on the other may accompany axi al rotation in loose joi nts . Thi s
would produce compl ex forces on the arti cul ar cart i l age, i ncl udi ng tensi on at
the capsular margin and compressi on and shearing deeper in the sagittal compo­
nent of the joint .
Meniscoid Inclusions, Back Pain, and "Locked Back"
In a previous section we described the different varieties of synovi al -l i ned,
vascul ar, fibrofatty pads, whi ch project up to 3 mm between the arti cul ar sur­
faces from the polar joint recesses of a normal joi nt . Smal l er fol ds from the
anterior and posterior capsul es fi l l the smal l t riangular gaps between the
rounded joint margi ns. A survey of 80 adult joi nts found that there are larger
fat-pads i n olderjoi nts, whi ch show the wear and tear changes of osteoarthrosi s,
46 Physical Therapy of the Low Back
Fig. 1 -27. Drawings of mi d-j oi nt horizontal sections of L3-L4 zygapophyseal joi nts,
whi ch were stained by hematoxyl i n and light green. ( A) A normal joi nt from a 20-year­
old woman shows smooth arti cul ar carti l age wi th regularly di stributed chondrocytes
and even stai ni ng of the matri x. The subchondral bone plate ( white) i s thi cker anteriorly
than posteriorly i n the concave facet . The el ast ic l igamentum favum (LF) forms the
anterior capsul e, whi ch mai ntai ns close apposition of the anterior arti cul ar surfaces,
but , i n the absence of mul tifidus, the posterior fi brous capsul e (Caps) is lax and the
posterior part of the joi nt i s open. ( B) A joi nt of a 37-year-old man shows chrondrocyte
hypertrophy in the arti cul ar cartilages of the coronal component . This is most evident
i n the concave facet of the superior arti cul ar process of L4, where two smal l splits i n
the arti cul ar carti lage are seen. There i s thi ckeni ng of the SCP, parti cul arly i n the coronal
component , compared to the joint in ( A) . I ncreased stai ni ng of the thi ckened SCP sug­
gests hypercal ci fication or "scl erosis. " The posterior fibrous capsule is covered by
some fibers of mul tifidus. Note the conti nui ty of the capsul e wi th the posterior margin
of the arti cular cart i l age of the concave facet. Spl i tti ng of thi s carti lage away from
the SCP is apparent . LF, ligamentum flavum; Caps , posterior fbrous capsul e. ( Fig lire
The Lumbar Spine from Infancy 10 Old Age 47
Fig. 1 -27. (Continlled). (C) Ajoi nt from a 62-year-old man where there i s some atrophy
of chondrocytes and loss of staining i n the concave facet , and some cel l hypertrophy
i n the convex facet. Fibri l l ation of both arti cul ar cartilages, most marked i n the coronal
parts of the joi nt, may be seen. The SCP has lost thi ckness compared to the SCP in
Fig. B. There are osteophytes at the posterior joint margi n. (D) A joint from a 6 1 -year­
old man showing advanced arthritic changes with wi despread fibril lation of arti cul ar
carti l age (AC). There are i nfracti ons and irregul arity of the SCPo The coronal component
of the SCP appears to have col lapsed into the spongy bone of the concave facet . The SCP
at thi s point shows local thi ckeni ng and i ncreased stai ni ng suggesti ve of "sclerosi s. "
48 Physical Therapy of the Low Back
than in younger adult joi nts. They appear to act as cushi ons i n lateral or medial
joint recesses adjacent to osteophytes, and to extend further between arti cul ar
surfaces where cart i l age l oss has occurred (Tayl or and Connel l , unpubli shed
dat a) . These vascul ar structures are i nnervated, and also di rect l y connected to
the joi nt capsul e, whi ch has many pain-sensi ti ve nerve fi bers . They are parti cu­
larl y l arge i n the i nferior joi nt recess of the l owest two l umbar mobile seg­
ments. 44.45.47 They are capable of causi ng back pain if entrapped between articu­
l ar surfaces, either due to di rect sti mul ati on of the i nnervated i ncl usions or due
to tracti on on the i nnervated capsul e. Their potent ial i mportance i n this latter
regard has been questi oned by Engel and Bogduk77 on the grounds that they
are too friable to exerci se tension on the capsul e if caught between the art icular
I n our study,6 we have found a number of i nstances of fibrocart i l aginous
i ncl usi ons projecti ng i nto the posterior aspect of zygapophyseal joi nt s, which
Fig. 1 -28. (A) Horizontal section from the l ower thi rd of a l ef L3-L4 joint i n a 37-
year-old man that shows ajoi nt i ncl usi on, which may be a torn-off portion of the damaged
arti cul ar cartil age of the concave facet, attached at its base to the posterior fibrous
capsul e. There is conti nui ty of the capsul e with the posterior margin of the cartilage on
the concave facet, and al so metapl astic formation of arti cular cartilage around the poste­
rior margin of the convex facet under the capsul e i n the form of a "wrap-around bumper"
( x 4). (Figllre continlles. )
The Lumbar Spine from Infancy to Old Age 49
Fig. 1 -28. (Continlled). (8) A higher-power vi ew of the j oi nt i ncl usion i n Fi g. A shows
the templatel i ke fit of the i ncl usion on the underl yi ng damaged cartilage, whi ch shows
signs of repair ( x 1 0) . ( Fig. 8 from Tayl or and Twomey, 81 wi th permission. )
appeared t o be tor-off port i ons of arti cular carti lage that remai ned attached
to the fi brous capsul e. Thei r origi n i n t he arti cul ar carti l age is attested by thei r
templatel i ke fit on the underl yi ng damaged carti lage, whi ch i s repai ri ng ( Fi g.
\ -28) . They appear to be analogous to torn meni sci i n the knee joi nt. Li ke torn
knee joint menisci , they could be di splaced i n certain ci rcumstances, and bei ng
firmer and more strongly attached to the capsul e than other fibrofatty i ncl u­
si ons, thei r entrapment coul d cause a reduction i n the normal range of spi nal
50 Physical Therapy of the Low Back
movement or even "l ocked back. " It is highl y l i kel y that thi s would be pai nful
because of traction on the joi nt capsul e. In thi s regard, the action of mul tifidus
in control l ing accurate apposi ti on of the joi nt surfaces would be i mportant. If
joi nt congrui ty i s not maintained by tone i n mul ti fdus there would be an oppor­
tuni ty for the torn porti ons of cart i l age to be di splaced, parti cul arly if the joint
surfaces separate in sudden rotary movement s. Mani pul ati ve techniques woul d
be appropriate in "freei ng" such entrapped torn pieces of arti cular carti l age.
On the other hand, mani pul ati ve techni ques that "gap" the joi nts too forceful l y
coul d al so damage joi nts wi th capsule-cart i l age cont i nui ty by "shearing off'
arti cular cart i l age.
In comparati ve anatomic and radiologic studi es , Taylor and McCormick4
showed that loose joi nts may contain enlarged fat-pads that extend from a polar
recess down to mi d-joi nt l evel s. These may represent an attempt to fi l l the
potential space in unstable joi nts wi th a loose capsul e.
Age Changes in Middle-Aged and Elderly Zygapophyseal Joints
Age changes in mi ddl e-aged and el derl y zygapophyseal joi nts i ncl ude those
already descri bed: the chondromalacia and subchondral scl erosi s i n the coronal
components of the facet s, whi ch probabl y resul t from repeti tive compressive
l oading, and also the loosening or "shearing off" of articular carti l age flaps at
the posterior joint margi ns from wear and tear associated with capslli e-carti lage
conti nui ty. They al so i ncl ude the more general age changes of osteoarthrosis
( e. g. , surface fi bri l l ati on and i rregul arit y) , with thi nni ng of arti cul ar carti l age,
osteophytic l ippi ng, facet enlargement , and i rregulari ty at the arti cul ar margi ns,
and scl eroti c thi ckeni ng of the subchondral bone. Subchondral bone cysts are
also found, but l ess often t han in the hi p and knee joi nts. Constant rubbing of
enl arged posterior joi nt margi ns against an overl yi ng thi ckened capsule l eads
to the extensi on of metapl asti c carti lage around the posterior joint margi ns
under the posteri or fibrous capsul e. These rounded "wrap-around bumpers"
can be recognized on CT scans of the joi nt s.
Measurements of arti cul ar cart i l age thi ckness suggest that an increase i n
central thi ckness may occur i n mi ddl e age. Hi stologic exami nation shows chon­
drocyte hypert rophy wi th swel l i ng of the matri x. Thi s swel l i ng may di srupt the
matri x, spl i ts occurring normal to t he subchondral bone, paral lel to the col lagen
fi bers i n the mid-zone. The underl yi ng SCP may show thi ckeni ng, or i nfractions
and col lapse of i ts anterocentral porti on. Measurements of concavi ty indexes
of the zygapophyseal joints at L l -L2, L3-L4, and L4-L5 , from 70 l umbar
spi nes of all ages, show i ncreased concavi ty of the superior art i cul ar facets in
old age. Thi s i ndex measures the degree of concavity at the bone-carti l age
i nterface i n the superior art i cul ar facet . The concurrent i ncrease in central carti­
l age thi ckness may si mpl y represent an attempt to maintain joint congrui ty.
Some osteoporotic joi nts wi th very thi n SCPs show a marked i ncrease i n con­
cavi ty. Other joi nts show i nfracti ons or local col lapse of central parts of the
The Lumbar Spine from Infancy to Old Age 5 1
SCPo This phenomenon i s si mi l ar to the osteoporoti c col l apse of the vertebral
body end-plate, which has reduced bony trabecul ar support .
Changes at Joint Margins: Tension Efects
Articular Cartilage. Whi l e the i ni t ial and more dramat i c age changes i n
articular carti l age take place i n t he coronal and central parts of t he joi nts as
described, i n t he long term, t he more destructi ve effects are at t he joi nt margi ns,
parti cularly at the polar regions and at the posterior margi ns of the joi nt s. Actual
loss of carti lage i s most common i n these regions, probabl y due to the "shearing
off' of carti lage by tension from capsular attachments.
This is often accompa­
nied by the i ntrusion of enl arged fat-pads i nto t he defect .
Bone Changes. I n mi ddl e-aged joi nts, osteophytes appear at the joi nt mar­
gins i n addition to the subchondral bony scl erosi s previousl y descri bed in the
coronal components of the superior arti cular facets . They are another mani fes­
tation of response to tensi on, both by extensi on of ossificati on from the l ami nae
into the l igamenta flava, and at the posterior arti cul ar margi ns wi th associated
enl argement of the mamil lary processes .
Radiologic Diagnosis. The assumpti on that subchondral bone thi ckeni ng
and sclerosi s are rel iable i ndi cators of wear and tear changes may be val i d i n
the joints of mi ddl e-aged active pati ents, but i n elderl y osteoporoti c joi nts there
may be arthri ti c changes wi thout subchondral bony sclerosi s . Our studi es show
that many elderly subjects wi th advanced fibri ll ati on and cart i l age loss have
extremel y thi n SCP due to concomitant osteoporosi s.
Relation of Age Changes in Discs and Facet Joints
It is commonl y stated78 that di sc degeneration (as a primary change) leads
to secondary zygapophyseal joint degenerati on. The coexi stence of disc and
facet changes i n advanced degeneration of the motion segment i s often ob­
served, but i t cannot be assumed that facet changes are always secondary to
di sc degenerat ions. The facets are vital l y i mportant in protecti ng the di scs from
damage, and in fulfi l l i ng thi s functi on they endure repeti ti ve loadbeari ng. I n
spondylol ysi s, when t he protecti ve infl uence of the zygapophyseal joi nt i s re­
moved, damage to the i ntervertebral di sc wi l l ensue. The anterior, coronal l y
oriented parts of l umbar zygapophyseal joi nts protect i ntervertebral di scs from
translational shearing forces and bear high loads when they bri ng flexi on to a
halt at the end of the physiologic range or when the spi ne i s loaded in a flexed
posit ion . The facets also l i mi t the range of extensi on when the ti ps of the inferior
articular processes abut on the subjacent l ami nae. The observati on of a bui ld­
up of scl erotic compact bone i n the l ami nae at the i nferior joi nt recesses i n
adul ts of al l ages i s wi tness to t hi s effect . The i mportant protecti ve effects of
the zygapophyseal joi nts on the i ntervertebral di scs are largely substanti ated
by the evidence we have presented.
D. ¹ ¯
¬ ¹ ·¬^
The i ncreased ranges of movement
52 Physical Therapy of the Low Back
amounti ng to i nstabi l i ty that resul t from experi mental sectioni ng of pedi cl es
would place additional stress on the i ntervertebral di scs. It i s apparent from
our studi es of di scs and zygapophyseal joi nts from the same i ndi viduals that
in some cases the zygapophyseal joi nts show more advanced degeneration than
i s apparent in the i ntervertebral di scs. However, the mobile segments i nteract
i n a mutual l y dependent way and i t is l i kel y that any defect i n one joint of the
arti cular tri ad woul d adversely i nfl uence the other joints of the triad. This is
recognized by spi nal surgeons who fuse the l umbar spi ne to avoid i nstabi l i ty,
when a wi de l ami nectomy i s required for spinal stenosi s ( see Ch. I I ) .
I . The strong constructi on of l umbar vertebrae and motion segments re­
flects the need for regional stabi l i ty in weightbearing and movement .
2. The thi ck young di scs, ri ch i n proteoglycans, are designed t o allow
useful ranges of movements and bear high compressi ve loads.
3. The biplanar facets restrain torsion and t ranslation, protecting t he annu­
lar fi bers of the discs from the stresses to which they are most vul nerable.
4. I n resi sti ng translation in fl exi on, the coronal components of the facets
may devel op chondromalacia and thi ckeni ng of the SCP i n young adul t s.
5. I n protecti ng the di scs i n deceleration i njuries, t he facets show i nfrac­
tions of their arti cular surfaces. I n fexion compression i njuries, the motion
segment hi nges on t he facets and the vertebral bodies are compressed with end­
plate fractures or vertebral wedging.
6. Spondyl ol ysi s i n young athletes i mpl i es fai l ure of the facet function,
wi th consequent exposure of t he di sc to shearing forces, whi ch may lead to
spondylol i sthesi s.
7. Degenerati ve spondyl ol i sthesi s i n t he el derl y i mpl i es slow fai l ure of the
facets' protective functi on wi th segmental i nstabi l it y.
8. I n t he "dysfuncti onal " phase of early looseni ng of t he motion segment,
wi th di sc fi ssuri ng, tension at the posterior facet margi ns, where the capsule and
arti cular carti l age are conti nuous, may "shear off' a flap of arti cul ar carti l age
formi ng a l oose carti l agi nous i ncl usi on i n thejoi nt, wi th t he possi bi l i ty of locki ng
of the joi nt.
9. I n el derl y spi nes, osteoporosi s l eads to vertebral end-plate col lapse with
"bal l ooning" of discs i nto the concave vertebral end-plates and shorteni ng of
the spi ne; thi s reduces the vertical di mensi ons of the nerve-root canal s.
1 0. Looseni ng of t he motion segment may al l ow vertebral retrol i sthesi s
wi th reducti on i n the horizontal di mensi ons of the nerve-root canal s.
I . Gi l es LGF, Taylor J R: Lumbar spine structural changes associated wi th leg length
i nequal i t y. Spi ne 7 : 1 59. 1 982
2. Gi l es LGF, Taylor J R: The effect of postural scol iosis on l umbar apophyseal joi nts.
Scand J Rheumatol 1 3: 209, 1 984
The Lumbar Spine from Infancy to Old Age 53
3. Taylor JR: Growth and development of the human intervertebral di sc. Ph. D. Thesi s,
Uni versity of Edi nburgh, 1 973
4. Twomey LT: Age changes i n the human l umbar vertebral col umn. Ph. D. Thesi s ,
Uni versity of Western Austral i a, 1 98 1
5 . Twomey L T, Taylor J R: Age-related changes i n the l umbar spi ne and spi nal rehabi l i ­
tati on. CRC cri ti cal revi ew. Phys Rehab Med 2: 1 53 , 1 99 1
6 . Taylor J R, Twomey L T: Age changes i n l umbar zygapophyseal joi nts: observations
on structure and function. Spi ne 1 1 : 739, 1 986
7. Taylor J R, Twomey LT, Corker M: Bone and sof ti ssue i njuri es in postmortem
l umbar spines. Paraplegia, 28: 1 1 9, 1 990
8. Hutton W, Adams MA: The forces acting on the neural arch and thei r rel evance to
low bac,k pai n. p. 49. In: Engineering Aspects of the Spi ne. Mechanical Engineeri ng,
London, 1 980
9. Davis PR: Human l ower l umbar vertebrae: some mechanical and osteological con­
siderations. J Anat 95 : 337, 1 961
1 0. Ramsey RH: Anatomy of l igamentum flava. Cl i n Orthop 44: 1 29, 1 966
I I . Yong-Hi ng MD, Rei l l y J , Ki rkal dy-Wi l l i s WH: The l igamentum flavum. Spi ne I :
226, 1 976
1 2 . Twomey L T, Taylor J R: Sagittal movements of the human l umbar vertebral col umn:
a quantitati ve study of t he role of t he posterior vertebral el ements. Arch Phys Med
Rehab 64: 322, 1 983
1 3 . Heyl i ngs 0: Supraspinous and interspinous ligaments of the human l umbar spi ne.
J Anat 1 25 : 1 27, 1 978
1 4. Twomey L, Taylor J R: Age changes i n the l umbar spine and i ntervertebral canal s.
Paraplegia 26: 238, 1 988
1 5 . Knutsson F: Growth and di fferentiation of postnatal vertebrae. Acta Radiol 55: 401 ,
1 96 1
1 6. Eisenstein S: The morphometry and pathol ogical anatomy of the l umbar spi ne i n
South Afri can negroes and caucasoids wi th specific reference to spinal stenosi s. J
Bone Joi nt Surg 59B: 1 73 , 1 977
1 7. Parkin IG, Harrison GR: The topographical anatomy of the l umbar epidural space.
J Anat 1 4 1 : 2 1 1 , 1 985
1 8. Crock HV, Yoshizawa H: Blood suppl y of l umbar vertebral col umn. Cl i n Orthop
1 55: 6, 1 976
1 9. Romanes GJR: The arterial blood suppl y of the spinal cord. Paraplegia 2: 1 99, 1 965
20. Dommisse GF: Bl ood suppl y of spi nal cord. J Bone Joi nt Surg 56B: 225, 1 974
2 1 . Sunderland S: Meni ngeal-neural rel ati onshi ps i n the i ntervertebral space. J Neuro­
surg 40: 756, 1 974
22. Kakul as BA, Taylor JR: Pathology of i nj uries of the vertebral col umn and spinal
cord. p. 2 1 . I n Frankel H L ( ed. ) : Handbook of Cl i nical Neurol ogy. El sevi er, New
York, 1 992
23. Twomey LT, Taylor JR, Ol iver M: Sustained flexi on loadi ng, rapid extension l oad­
ing of the l umbar spine and the physiotherapy of related i njuri es. Physiother Pract
4: 1 29, 1 988
24. Taylor JR: Age changes i n l umbar zygapophyseal joi nts. p. 1 2 1 . I n: Archi ves of
Bi ol ogy Vllth Symposi um I nternationale des Sciences Morphologi ques. Louvai n­
en-Wol ue, 1 986
25. Taylor JR, Twomey LT: Lumbar mul ti fi di : rotator-cuff muscl es of zygapophyseal
joints. J Anat 1 49: 266, 1 986
26. Bowen V, Cassidy J D: Macroscopic and mi croscopic anatomy of the sacro-i l iac
joint from embryonic l i fe unti l the 8th decade. Spi ne 6: 620, 1 98 1
54 Physical Therapy ofthe Low Back
27. Palfrey AJ : The shape of sacroiliac joi nt surfaces . J Anat 1 32: 457, 1 98 1
28. Cl ayson SJ: Eval uation of mobi l i ty of hi p and l umbar vertebrae of normal young
women. Arch Phys Med 43: I , 1 962
29. Bourdi l l on JF: Spi nal Mani pul ation. 2nd Ed. Hei nemann, London, 1 973
30. Taylor JR, Sl i nger BA: Scol i osi s screening and growth i n Western Australian stu­
dents. Med J Aust 1 : 475 , 1 980
3 1 . Taylor JR, Twomey LT: Devel opment of the human i ntervertebral disc . p. 39. [ n
Ghosh P ( ed) : Biology of the I ntervertebral Di sc . CRC Press, Boca Raton, FL, 1 988
32. Horton WG: Further observations on the el astic mechani sm of the intervertebral
di sc. J Bone Joi nt Surg 40B: 552, 1 958
33. Nachemson A, El fstrom G: I ntravital dynami c pressure measurements i n lumbar
di scs. Scand J Rehabi l Med 1 : I , 1 970
34. Tayl or JR, Twomey L T: I nnervation of l umbar i ntervertebral di scs. Med J Aust 2:
701 , 1 979
35. Taylor JR, Twomey L T: I nnervation of l umbar i ntervertebral di scs. N Z J Physiol
8: 36, 1 980
36. Bogduk N: The i nnervation of the l umbar spi ne. Spi ne 8: 286, 1 983
37. Taylor JR, Scott J E, Cri bb AM, Bosworth TR: Human intervertebral disc acid
gl ycosaminogl ycans. J Anat 1 80: 1 37, 1 992
38. Hirsch C, Schazowicz F: Studi es on structural changes in the l umbar annul us fibro­
sus. Acta Orthop Scand 22: 1 84, 1 953
39. Wal msl ey R: The human i ntervertebral di sc: devel opment and growt h. Edi n Med
J 60: 341 , 1 953
40. Maroudas A, Nachemson A, Stockwel l RA: Factors i nvol ved i n the nutrition of the
human l umbar intervertebral di sc : cel l ul arity and di ffusi on of glucose i n vi tro. J
Anat [ 20: 1 1 3 , 1 975
4 1 . Taylor JR, Twomey, LT: Vertebral col umn devel opment and i ts relation to adul t
pathology. Aust J Physiother 3 1 : 83, 1 985
42. Twomey L, Tayl or J : Age changes i n the l umbar arti cul ar triad. Aust J Physiother
3 1 : 1 06, 1 985
43 . Tayl or JR, Twomey LT: Structure and function of l umbar zygapophyseal ( facet )
joi nts. Ch. 8. [ n Boyl i ng JD, Palastanga N (eds) : Grieve' s Modern Manual Therapy
of the Vertebral Col umn. Churchi l l Li vi ngstone, Edi nburgh, 1 993
44. Tayl or JR, McCormi ck CC: Lumbar facet joi nt fat pads. Neuroradiology 33: 38, 1 99 1
45. Gi l es LGF, Taylor J : Intra-arti cul ar synovial protrusions i n the lower l umbar apoph­
yseal joi nts. Bul l Hosp Joi nt Dis 42: 248, 1 982
46. Wyke B: The neurol ogy of joi nts: a revi ew of general pri nci pl es. Clin Rheum Dis
7: 223, 1 98 1
47. Gi l es LG, Taylor JR, Cockson A: Human zygapophyseal joint synovial fol ds. Acta
Anat 1 26: 1 1 0, 1 986
48. Taylor JR, Twomey L: Sagittal and horizontal plane movement of the l umbar verte­
bral col umn in cadavers and in the l ivi ng. Rheum Rehabil 1 9: 223, 1 980
49. McCormi ck CC, Taylor JR, Twomey LT: Facet joi nt arthrography in l umbar spon­
dyl ol ysi s: anatomic basis for spread of contrast . Radiology 1 7 1 : 1 93 , 1 989
50. Spangfort EV: The l umbar di sc herni ati on. Acta Orthop Scand suppl 1 42: 1 972
5 1 . McRae DL: Radiology of the l umbar spinal canal . p. 92. In Wei nstei n PR, Ehni
G, Wil son CB (eds): Lumbar Spondyl osi s: Diagnosi s, Management and Surgical
Treatment. Year Book Medi cal Publ i shers, Chicago, 1 977
52. Wolpert L: The cel l ul ar basis of skel etal growth duri ng development. Br Med Bul l
37: 1 52, 1 98 1
The Lumbar Spine from Infancy to Old Age 55
53. Taylor JR, Twomey LT: Factors infl uenci ng growth of the vertebral col umn. p. 30.
I n Grieve GP (ed): Modern Manual Therapy. Churchi l l Li vi ngstone, Edinburgh,
1 986
54. Taylor JR, Twomey LT: The role of the notochord and blood vessel s in devel opment
of the vertebral col umn and in the aetiol ogy of Schmorl ' s nodes. p. 2 1 . In Gri eve,
GP ( ed) : Modern Manual Therapy of the Vertebral Col umn. Churchi l l Li vi ngstone,
Edinburgh, 1 986
55. Verbout AJ : The devel opment of the vertebral col umn. p. 1 . I n : Beck F, Hi l d W,
Ortmann R ( eds) : The Devel opment of the Vertebral Col umn. Advances i n Anat­
omy, Embryology and Cell Biol ogy. Vol . 90. Springer Verlag, Berl i n, 1 985
56. O' Rahi l l y R, Meyer DB: The ti mi ng and sequence of events i n the development of
the human vertebral col umn during the embryonic period proper. Anat Embryol
1 57: 1 67, 1 979
57. Schmorl G, Junghanns H: p. 68. The Human Spine in Heal th and Disease. 2nd Ed.
(Translated by Besemann EF. ) Grune & Stratton, New York, 197 1
58. Taylor JR: Scoliosis and growth: patterns of asymmetry in normal growth. Acta
Orthop Scand 54: 596, 1 983
59. Tayl or JR: Persistence of the notochordal canal i n vertebrae. J Anat I I I : 2 1 1 , 1 972
60. Dal e Stewart TD, Kerl ey ER: Essenti al s of Forensic Anthropology. p. 1 36. Charles
C. Thomas, Spri ngfiel d, I L, 1 979
6 1 . Tanaka T, Uht hoff HK: The pathogenesis of congenital vertebral mal formations.
Acta Orthop Scand 52: 4 1 3 , 1 98 1
62. Watterson RL, Fowler I , Fowler BJ : The role of the neural tube and notochord i n
development of the axial skel eton of the chi ck. Am J Anat 95: 337, 1 954
63 . Taylor JR: Growth of human intervertebral discs and vertebral bodi es. J Anat 1 20:
49, 1 975
64. Taylor JR, Twomey L T: Sexual di morphi sm i n human vertebral growt h. J Anat
1 38: 28 1 , 1 984
65 . McFadden KD, Taylor J R: End-plate l esions of the l umbar spi ne. Spine 1 4: 867,
1 989
66. Dietz GW, Christensen EE: Normal ""cupi d' s bow' ' ' ' contour of lower l umbar
vertebrae. Radiology 1 2 1 : 577, 1 976
67. Bradford DS, Moe J H: Scheuermann' s juveni le kyphosi s: a histol ogical study. Cl i n
Orthop 1 1 0: 45 , 1 975
68. Taylor JR: Vascul ar causes of vertebral asymmetry and the laterality of scoliosi s .
Med J Aust 1 44: 533, 1 986
69. Dickson R, Lawton JO, Butt WP: Pathogenesis of idiopathic scoliosi s . p. 1 6. I n
Dickson R, Bradford D ( eds) : Management of Spinal Deformities. Butterwort hs,
London, 1 984
70. Van Shaik J PJ , Verbiest H, Van Shaik FDJ : The orientation of laminae and facet
joints i n the lower l umbar spine. Spine 1 0: 59, 1 985
7 1 . Taylor JR, Corker M: Age rel ated responses to stress i n the vertebral col umn: a
review. I n: The Growing Scope of Human Biology: Proceedings of the Australasian
Society for Human Biology 2: 2 1 3. Centre for Human Bi ology U W A, Uni versity
of Wester Austral i a, 1 989
72. Taylor J R: Growth of l umbar vertebral bodies and I VDs i n relation to weightbearing
in the erect posture. J Anat 1 1 9: 4 1 3 , 1 975
73. Twomey L T, Taylor J R, Furniss B: Age changes i n the bone density and structure
of the l umbar vertebral col umn. J Anat 1 36: 1 5 , 1 983
74. Twomey L, Taylor J: Age changes in l umbar vertebrae and intervertebral di scs.
Cl in Orthop 224: 97, 1 987
56 Physical Therapy of the Low Back
75. Twomey L T, Taylor J R: Old age and physical capaci ty: use it or lose i t . Aust J
Physiother 30: 1 1 5, 1 984
76. Sartoris DJ , Resni ck D, Tyson R, Haghighi P: Age-related alterations i n the vertebral
spi nous processes and i nterveni ng soft ti ssues: radiologic-pathol ogic correl ation.
Am J Roentgenol [45: 1 025, [ 985
77. Engel R, Bogduk N: The meni sci of the l umbar zygapophyseal joints. J Anat 1 35:
795, 1 982
78. Vernon-Roberts B: The pathology and interrelation of i ntervertebral di scs l esions,
osteoarthrosis of the apophyseal joi nts, l umbar spondyl osi s and low back pai n. p.
83. I n Jayson MIV (ed): The Lumbar Spi ne and Back Pai n. 2nd Ed. Pitman, London,
1 985
79. Last S: Anatomy: Regional and Appl i ed. 2nd Ed. Churchi l l Li vi ngstone, Edinburgh,
1 959.
80. Twomey LT, Tayl or JR: Age changes i n l umbar i ntervertebral di scs. Acta Orthop
Scand 56: 496, 1 986
8 1 . Taylor J R, Twomey L T: Structure, functi on, and age changes i n the l umbar zyga­
phophyseal joi nts. Spi ne I I : 739, 1 986.
2 Lumbar Posture,
Movement, and
Lance T. Twomey
James R. Taylor
The adult vertebral column is a segmented, jointed, flexible rod that sup­
ports the loads of weightbearing in the erect posture, protects the spinal cord
and emerging spinal nerves, allows a considerable range of movements in all
directions, and serves as the axial support for the limbs. The vertebral column's
capacity to fully subserve these functions alters through the different phases
of the life cycle. The extremely malleable C-shaped column of the neonate
remains almost as flexible and mobile in childhood as it grows and develops
its finely balanced curves. Further growth and maturation are associated with
progressive increases in the strength and "dynamic" stability of the adolescent
and young-adult columns, and with a continuing small decline in its mobility.
The middle years demonstrate an increasing incidence of minor traumatic and
degenerative pathology, with a further decline in range of movement. In old age,
with its osteoporotic decrease in bone strength, there is progressive increase in
joint stiffness with a considerable decline in movement ranges, and a flattening
of the lumbar spine. The lumbar spine is markedly lordotic in children, with a
small decline in lordosis in adolescents and young adults, and a pronounced
flattening of the region in middle life and old age.
Posture is a term that indicates the relative position of the body segments
during rest or activity, while stature indicates the height of a subject. In most
individuals their resting supine length exceeds their standing height (or stature)
by about 2 cm.
58 Physical Therapy of the Low Back
Posture refers to a composite of the positions of all of the joints of the
body at any given moment. I A minimum of muscle work is required for the
maintenance of good posture in any human static or dynamic situation. In good
standing posture, the head is held tall and level, while the spine is nicely bal­
anced so that its sagittal curves allow free movement of the chest and abdomen,
and prevent the shoulders from sagging forward. The lower limbs serve as
balanced support. In a side view of most individuals, a plumb line would inter­
sect the mastoid process, the acromion process, and the greater trochanter. It
would pass just anterior to the center of the knee joint and through the ankle
The usual static posture for the lumbar spine is that of lordosis. Normal
spinal posture is expressed as a balanced series of curves when viewed from
the side (Fig. 2-\). The adult spine is supported on a symmetrical level pelvis
by two equal-length lower limbs. In normal sitting posture, the level pelvis is
supported with body weight distributed equally through both ischial tuberosi­
ties. There is no discernible lateral curvature or rotation of the spine when
viewed from the front or behind.
line of gravity
center of
-- gravity
Fig. 2-1. The relati onshi p between the vertebral col­
umn and the l i ne of gravi ty.
Lumbar Posture, Movement, and Mechanics 59
The cervical lordosis begins to appear at birth and develops as a permanent
curve at about 3 months of age, while the permanent lumbar lordosis appears
with the extension of the legs and weightbearing in the erect posture, usually
between 12 and 18 months of age. These curves continue to change until the
completion of spinal growth, usually between the ages of 13 and 18 years.2-4
Sagittal Pelvic Tilt and Muscle Action
Pelvic tilt in the sagittal plane and lumbar lordosis are inextricably linked
together, because the lumbar spine and the sacrum are united at the strong,
relatively immobile sacroiliac joints. Thus, when the pelvis is tilted further
forward, it brings about an increase in lordosis, and when tilted backward, the
lumbar spine flattens. The muscles responsible for pelvic posture include erec­
tor spinae (sacrospinalis), abdominals (rectus abdominus and oblique muscles),
psoas major and iliacus, gluteus maximus, and hamstrings (Fig: 2-2).
It is the interaction between these muscles that is the major factor determin­
ing pelvic tilt and lumbar lordosis at any point in time. Thus, while the back
extensor muscles primarily increase lumbar lordosis, the abdominal muscles,
Fig. 2-2. The muscles responsible for the maintenance
of pelvic ti l t. Abel. abdominal muscles; ES, erector spinae;
GM. gluteus maximus; H. hamstri ngs; PM. psoas major.
60 Physical Therapy of the Low Back
gluteals, and hamstrings act together to flatten lumbar lordosis by their action
about the lever of the pelvis. The psoas muscles, which attach to the lateral
margins of the lumbar vertebrae, can also increase lumbar lordosis when the
lower limbs are extended by pulling the lumbar vertebrae forward around the
"pulley" of the hip joint (see Fig. 2-2). In this way they pull the pelvis and
lumbar vertebrae ventrally. In the course of everyday activity we are constantly
adjusting our posture to allow for comfort, ergonomic advantage, and in re­
sponse to our environment. Muscle tightness affecting any of the muscle groups
listed can change both habitual resting posture and the total range of dynamic
postures available. Tightness of psoas major and hamstring muscles is associ­
ated with increased lordosis in some individuals. Shortened hamstrings should
pull down on the pelvis and flatten the lumbar spine. For this reason, it is
possible that tight hamstrings are the result of lordotic posture rather than its
Analysis of the Lumbar Curve
The lumbosacral lordosis is a compound curve (Fig. 2-3) with the degree
of curvature greatest at the LS-S I level, and least at the L I-L2Ievel. In general,
at all ages the intervertebral discs contribute to a greater proportion of the
L-1 L 1-81 Angle
( degrees)
¯ Ý
3 4 5 6
Age category
Fig. 2-3. Changes that occur through l ife to the l umbosacral angl e.
Lumbar Posture, Movement, and Mechanics 61
lordosis in both sexes than do the vertebrae. However, at the lumbosacral
junction, the L5 vertebral body makes a significant contribution. The L5-S I
disc is also more wedge-shaped than any of the higher discs.
Sexual Dimorphism
During the childbearing years (i .e:, between adolescence and middle age)
the LJ-SI angle of lumbar lordosis is greater in women than in men.5 This
sexual dimorphism is not apparent in childhood and disappears again in old
age. Although the reasons for this diference remain obscure, its place in the
life cycle suggests that it has a hormonal basis. One of the hormones that may
be involved consists of three closely related polypeptides collectively called
relaxin. It is secreted by the ovary and has been shown to relax the symphysis
pubis, sacroiliac joints, and spinal ligaments.6 These hormones, which are se­
creted in relatively large amounts by the corpus luteum of pregnancy, are also
found in small amounts in the circulating blood of nonpregnant women of child­
bearing age. It is suggested that the effects of relaxin in "loosening" pelvic
and lumbar ligaments may coincidentally allow an increase in the lumbar curve
during that period in the female life cycle (adolescence and early adult life)
when the hormones are present in relatively large amounts.
Another suggestion advanced to explain the gender differences in lordosis
relates to Treanor's demonstration
that the wearing of high-heel shoes tips the
body's center of gravity forward and brings about an associated increase in
pelvic tilt and an increase in lumbar lordosis, because lumbar lordosis is depen­
dent on pelvic posture (Fig. 2-4). The habitual wearing of high heels (predomi­
nantly a female fashion) in the "developed" countries may eventually contrib­
ute to a difference in lordosis between men and women in Western societies.
This is less likely to be a feature in societies where outdoor activity in bare
feet or soft-heel shoes is possible for most of the year.s
Recent studiess.9 show that an increase in pelvic tilt due to high heels does
not always bring about an increase in lumbar lordosis in nonpregnant young
women. An equal number of women demonstrated a flattening of their lumbar
curve as showed an increase. However, pregnant women showed a significant
increase in lordosis when wearing 2-inch-high heels. The response of the lumbar
spine to changes in pelvic posture would appear to relate to the location of the
center of gravity. In pregnancy, the center of gravity is displaced ventrally and
is balanced by an increase in lumbar lordosis.
An analysis of the components of the L I-S I angle in a major cadaveric
study5 showed that the principal difference between the lumbar posture of fe­
males and males occurs at the lumbosacral junction, because the composite
LJ-LS angles are similar throughout life in both sexes (Fig. 2-5). Thus, it is
the increased sacral and pelvic tilt of females that is primarily responsible for
the difference in lordosis during adolescence and early adult life. In this regard,
it is interesting to note that anatomists such as Romanes 10 consider that in
normal erect posture, the anterosuperior iliac spines and the symphysis pubis
62 Physical Therapy of the Low Back
Fig. 2-4. The i ncrease in l umbar lordosis due to wearing high-heel shoes.
lie in the same plane. This was confirmed in a measurement study by Taylor
and Alexanderll of 39 nonpregnant females. Obstetric experience suggests that
the symphysis pubis lies in a more anterior plane than do the iliac spines.12
The greater LS-S I angle in females may well be related to a greater degree of
pelvic tilt during the childbearing years.
Variations in Lumbar Lordosis
Lumbar lordosis is maintained by intrinsic features such as the shape of
the vertebrae, discs, and the sacrum, and by extrinsic factors such as position
of center of gravity, body weight and its distribution, muscle strength, and
Anl ( degrees )
Lumbar Posture, Movement, and Mechanics 63
L 1-L5 ang

. ¯¯

.¯¯ ¯

� L5-S1 an

�¯¯�¯ �¯¯¯�¯��
0-1-5 1_5-12 13-19M 20-35M 36-59 8
13-17F 18-35F
Ag ( yeas)
Fig. 2-5. An analysis of the change in L I to LS and LS-S I angles wi th increasing age.
sociocultural preferences_ In regard to sociocultural preferences, it is commonly
observed that an individual's lumbar lordosis and thus habitual posture is based
on factors such as fashion (e.g., wearing high heels), repetitive daily activity
(e.g., a schoolchild carrying a heavy schoolbag), life-style (e.g., the "typical"
military posture), afect and attitude (e.g., depression or elation), and aesthetics
or training (e.g., model's swayback posture).
Prolonged Maintenance of the Static Erect Posture
Standing in the upright position for long periods of time tends to produce
an increase in lumbar lordosis as musc\e� begin to fatigue and as the slow creep
of the soft tissues often emphasizes the natural tendency toward extension of
the region. This occurs because the center of gravity is usually ventral to the
sacral promontory in most adults.13 The effect of gravity acting through this
center pulls the lumbar spine into a more lordotic posture in those individuals.
This postural change accompanies the diurnal decline in height (see Diurnal
Variation in Stature, below). Vertebral column posture is ideally dynamic rather
than static, as the tissues adapt to prolonged static loading by further creep of
the column.
The Effects of Age on Lumbar Lordosis
The lumbar spinal lordosis flattens considerably in old age in both sexes
(see Fig. 2_5),2.5 although a few individuals do show small increases in lordosis.
Increases are usually associated with increased abdominal girth and weight and
64 Physical Therapy of the Low Back
declines in abdominal-muscle strength, so that their physique approximates that
of a pregnant female. Thus, the explanation for an increase in lordosis is the
same as that for pregnant women (see above) . However, Twomey's large study5
of a typical Australian society clearly shows a significant decrease in lumbar
lordosis with increasing age after adolescence of 32 percent in females and 20
percent in males.
Lumbar Lordosis in Association with Back Pain
Clinicians often report a flattening of the lumbar lordosis during episodes
of back pain. However, a study of 600 men between the ages of 23 and 60 years
has shown that the distribution and range of lordosis (as viewed on radiographs)
do not vary in acute or chronic low-back pain more than they do in men without
back pain. 1 4
Leg-Length Inequality and Pelvic Obliquity
Whereas pelvic tilt in the sagittal plane is inextricably linked with lumbar
lordosis, coronal- plane obliquity of the pelvis is associated with lumbar sco­
liosis. The most common cause of this functional situation is leg-length inequal­
Functional scoliosis must be distinguished from idiopathic structural sco­
liosis. Structural scoliosis progresses during the growth period, is seen most
frequently in girls, 1
and is usually convex to the left in the lumbar spine. Leg­
length inequality and postural scoliosis have been associated with low back
pain, degenerative changes in the intervertebral discs and zygapophysealj oints,
and with a higher incidence of osteoarthritic changes in hip and knee j oints.
I nterestingly, it is most often the left leg that tends to be longest, particularly
in men, 17 since most right-handed people put more weight on their left foot. 18
Unequal leg length may be associated with pathologic conditions (e. g. ,
Perthes disease, previous fracture) , but in the vast maj ority of situations, it
accompanies normal growth. I n most of this latter group, the degree of asym­
metric growth of the lower limbs is very common. I t is associated with "out
of phase" growth, where one of a pair of bones is more advanced than the other
at maturity. 17 A leg-length difference of I cm or greater is twice as common at
the peak of the adolescent growth spurt (13 percent) than at maturity (7 per­
cent) . 19 I ndeed, it is very rare to find exactly equal leg lengths in normal commu­
Giles and Taylor
1 showed that unequal leg length and the associated
"postural" scoliosis are linked with minor structural changes in lumbar discs
and vertebral end-plates, and with asymmetric changes to the articular cartilage
and subchondral bone of the lumbar zygapophyseal j oints. The j oints on the
convex side of the scoliotic curve show thicker subchondral bone plates and
thinner articular cartilage than those from the concave side, suggesting greater
loading on the convex side of the curve. This may be related to the greater
Lumbar Posture, Movement, and Mechanics 65
postural muscl e forces necessary on the convex side t o prevent buckl ing of
the scoliotic column under axial l oading. A number of surveys have shown a
stat istical association between leg-l ength inequal ity and low back pain. 1 5
Leg-l engt h inequal ity is twice as common in low back pain patients (13 to 22
percent) than in control populat ions (4 to 8 percent) . Giles and Tayl orl 5 also
suggested that t he response to manipulat ive therapy in l ow back pain associated
with l eg-l engt h inequal ity is much improved when a foot-raise shoe insert is
provided as part of the treatment .
The t opics of stat ure and post ure are cl osely related. Stature is affected
by posture in a number of ways.
Postural Fatigue
Laxity in posture causes " creep" of sof tissues (see Prolonged Mainte­
nance of the Static Erect Posture, above) .
Diurnal Variation in Stature
I n 1 777, Buffon noted t hat a young man was considerably shorter after
spending t he night at a ball but he regained his previous height after a rest in
bed. Merkel (1881) measured his own daily l oss in height (2 cm standing, 1.6
cm sitting) by measuring t he height of his "visual plane" from the foor. He
found that half his loss in statute occurred in the first hour after rising, and t hat
a greater l oss occurred afer vigorous exercise.
De Puky
4 measured diurnal variation in stature and found the daily loss
in height to average 1 5. 7 mm. Diurnal variation as a percentage of total body
height decreases steadily wit h increasing age. Blackman25 showed a decrease
in st ature of 0. 76 cm 1 hour afer rising and 1.77 cm 4 hours after rising; this
order of decrease was confrmed by Stone and Tayl or, 26 who showed that l oss
in sitting height was equivalent to 80 percent of t he loss in standing height.
I n an interesting study, Tyrrell et al27 showed that average daily variation
in stature was about 1 percent of normal stature, and that the greatest l oss
occurred in the first hour after rising in the morning. Approximately 70 percent
of this l ost stature was gained during t he first hal f of t he night. The carrying
of heavy loads increased t he rate of shrinkage loss (i.e., by creep) . I nt erestingly,
rest with the l umbar spine in ful l flexion produced more rapid gains in stature
than in other positions. This also suggests that the diurnal loss invol ves creep
into extension (see Prol onged Maintenance of the Static Erect Posture, above).
Adams and Hutton
have recently demonstrated that the fexed position in­
duces t he t ransport of metabol ites and fuids into the intervertebral discs. I f
66 Physical Therapy of the Low Back
most of t he diurnal loss in st at ure is a l oss of t runk l ength due to smal l diurnal
reductions in disc height ,
6 t hen t he us
of flexion as a t ool t o maintain disc
height and t o preserve normal erect post ure wit hout excessive l ordosis becomes
of cl inical interest for physical t herapists.
The mechanism invol ved in diurnal variation in st at ure is discussed furt her
under l ater sections on creep of vertebral st ruct ures.
The Infuence of Changes in Posture on Stature
When parent s measure t heir chil d' s stat ure as a record of t heir growt h rate,
t he usual instructions given are t o "st and t al l , l ike a sol dier. " This impl ies a
general underst anding t hat st at ure is dependent in part on a person's posture.
When "st anding t all , " the chil d flattens t he spine as much as possible, tucks
t he chin in, and att empts t o push t he t op of t he head as far upward as possibl e.
Simil arl y, aft er surgical correction of moderate t o severe scol iosis, when t he
spine is surgicall y " st raightened, " chil dren can gain up t o 8 cm in height .
The t horacic spine makes t he l argest cont ribution to spine l ength. The
l umbar spine constit utes one-third, t he cervical spine one- fift h, and t he t horacic
spine t he remainder of t he t ot al length of t he presacral spine in t he adul t .
The Effect of Growth and Aging on Stature
The rates of growth and decl ine in st at ure are described in Chapter I. The
spinal component in t he decline in st at ure t hat occurs in old age is much more
a result of a decrease in vertebral height t han it is a decrease in intervertebral
disc height . 30•3
An increasing t horacic kyphosis (part icul arly in women) also
contributes significant ly to t he loss of stat ure in ol d age.
At each l evel in t he vertebral column t here are t hree interacting joint s
allowing and control l ing movement. This unique combinat ion is known as t he
articular triad or as t he mobile segment (see Ch. I). Each art icul ar t riad allows
onl y a few degrees of movement . However, l umbar movement usual l y involves
a complex interaction of mobile segment s at mul t iple l evel s. The t hickness of
each int ervertebral disc, t he compl iance of it s fibrocart ilage, and t he dimensions
and shape of it s adj acent vertebral end-plates are of primary importance in
governing t he ext ent of movement possible. The shape and orientation of t he
vertebral-arch articul ar facet s, wit h t he ligaments and muscl es of t he arch and
its processes, guide t he t ypes of movement possible and provide restraint s
against excessive movement .
Lllmbar Postllre, Movement, and Mechanics 67
Ranges of Disc Movement
The ant erior el ement s (vert ebrae and discs) of t he articular t riads are capa­
ble of certain ranges in movement depending on disc dimensions (thickness and
horizontal dimensions) and disc st iffness.
Disc Dimensions
A large range of movement woul d occur when disc height was relatively
great and vert ebral end-pl ate horizontal dimensions rel ativel y short (Fig. 2-6).
Adol escent and young adult femal es have shorter vertebral end-plates (a) than
mal es, whereas disc height (bl, b2, and b3) and disc st iffness are substant ial l y
t he same. Thus, femal es possess t he necessary combinat ion of dimensions for
a larger range of movement s than is possibl e in mal es.5.3 2 I n ol d age, when
mal e and femal e vert ebrae and disc shapes become very simil ar and hormonal
differences are reduced, t he range of movement of men and women becomes
al most ident ical (Fig. 2-7).
Disc Stifness
The general reduct ion in movement ranges in both sexes is att ribut abl e t o
increased disc st iffness. This has been demonst rat ed by t he posterior release
experiment of Twomey and Tayl or,33 which demonst rat ed a 40 percent increase
in disc st iffness in t he elderly.
Fig. 2-6. The anterior vertebral el ements (mobile segment): a, vertebral end-plate, and
b, disc thickness.
68 Physical Therapy of the Low Back
flexion extension
20 10 V
10 20 30 40


2 ¸ 1·5 • 1
20 10
lateral Flexion
o 10
. l
| .
left right
10 1
10 20
O. 1-5
1·5. 12
36. 59

. .
Laterol flexion

IS' 35



Fig. 2-7. Age changes in the range of l umbar movements in both sexes. (From Taylor
and Twomey,32 with permission. )
Planes of Movement
The movements possible at each l umbar- mot ion segment are t raditional l y
described as being i n the sagit t al (flexion-extension), coronal (lateral flexion) ,
and horizontal (axial rotat ion) pl anes. Each movement occurs along one of
t hree coordinate axes x, y, and z (Fig. 2-8). Thus, all mobile segment s of the
lumbar spine possess 6° of freedom and each movement consist s of an angul ar
or rotary displ acement t oget her with t ransl ation of a vertebra on it s subj acent
vertebra. I t is rare for movement t o occur exclusivel y in a single pl ane. Move-
Lumbar Posture, Movement, and Mechanics 69
Fig. 2-8. Planes and axes of movement. Sagittal plane movements occur along the x
axis; rotational plane movements occur along the y axis; and coronal plane movement
occurs along the z axi s.
ments are generally "coupled" in habitual movement, 34 and occur across the
standard descriptive planes of motion.
Ranges of Lumbar Movement
Despite the availability of simple, reliable methods for measuring spinal
mobility, these have not been applied to studies of normal lumbar spine move­
ment until recently. The literature records considerable variation in the values
given for the ranges of movements of the lumbar spine. This variation stems
largely from the different measurement methods used and the differences in
age, sex, race, and numbers of subjects studied. The clinical measurements
most frequently used include indirect estimates of spinal mobility from measure­
ment of (Ì )the distance from the fingertips to the floor when the patient bends
forward, and (2) the use of a tape measure to measure the increase in distance
between two skin landmarks, often the SI and Ll spinous processes. These
methods are most inaccurate, and give no direct measure of the range (angular
deflection) of spinal movement. The former is dependent on hamstring muscle
length and the latter fails to show a reasonable level of consistency between
repeated measures. Published studies of lumbar spinal movement have mostly
concentrated on sagittal and coronal plane movements, and include direct mea-
70 Physical Therapy of the Low Back
surement in living subjects, using a wide variety of equipment, 3
35 -43 radio­
graphic studies,44-5 1 cadaveric studies that have mostly involved a single mobile
segment in a small number of specimens, 3.49
-64 photographic techniques,6
and theoretical studies based on mathematical models.
Estimates of the range of sagittal motion of the lumbar region vary widely
from 121 ° in a young male acrobat,44 to 21.8° in elderly women.4
Begg and Falconer
3 considered 70° to be the "normal" average total range of
lumbar fexion-extension. Few studies have attempted to measure axial rota­
tion in the lumbar spine, largely because of methodologic problems. I t has
proven diffcult to measure lumbar rotation in the living either directly or radio­
graphically with any degree of accuracy, and cadaveric studies have mostly
been confined to motion segments rather than to the whole lumbar column.
Some authorities maintain that rotary movement does not exist as a sep­
arate entity in the lumbar region,
4-76 or that if rotation does occur, it is in
spite of the fact that the facets are designed to prevent it.
Other sources have
assessed the total range of rotation as between 5° and 36° of move­
ment. 3
Clinical Measurement
I n an effort to provide instrumentation that would be relatively easily ap­
plied in the clinical situation and provide reasonably accurate objective data,
two instruments have been devised to measure lumbar sagittal anc horizontal
plane movement, and have been tested in clinical trials.3
The lumbar spon­
dylometer is noninvasive, has good interperson and intertest reliability, and
measures lumbar sagittal motion (Fig. 2-9). Since its base rests on the sacrum,
the measurement is not confounded and invalidated by the inclusion of hip
motion. Tests of its accuracy made by comparing living subjects with fresh,
cadaveric specimens suggest that it underestimates the range of movement by
an average of J 0. 32 I nter- and intraoperator repeatability trials show high correla­
The lumbar spondylometer is comparable in accuracy and in some
respects in principle to an inclinometer, but with a more complex geometry. :5
I t is also easier to use in a clinical situation where separate readings from two
inclinometers would be required. I ts use requires a thorough knowledge of the
surface anatomy of the lumbar region, with consistently accurate placement of
the cushions, and the precise location of the L I spinous process.
The cushions of the lumbar spondylometer rest on the dorsal surface of
the sacrum, with the top cushion at the level of S I. The distal end of the instru­
ment rests on the spinous process of LI (Fig. 2-9). The physical therapist reads
of the initial starting position in degrees, asks the subject to fully extend, and
reads off the new position in degrees. The subject returns to the starting position
(checked by the operator) , and then moves into full-range fexion, with the
operator recording the result. Thus, flexion, extension, and full-range sagittal
motion are recorded. This entire process takes an experienced physical thera­
pist less than 2 minutes to administer and record, and is a useful objective
Lumbar Posture, Movement, and Mechanics 7 1
Fig. 2-9, The lumbar spondyl ometer.
cl inical measurement in t he assessment of t he progress of t reat ment for back
condit ions. !!
The ext eral measurement of rotation in t he cl inical situation has been
made possibl e by t he development of a l umbar rot amet er.32
The apparat us
consist s of a l arge prot ractor strapped at right angl es t o t he subject ' s sacrum,
and a belt wit h a pointer st rapped around t he t runk at L I. The t ip of t he point er
rest s just above t he prot ractor (Fig. 2-(0). The subj ect is asked t o rot at e ful l y
t o t he right and the l eft , and angular defl ections of t he point er are read off on
the prot ract or. lnt ert rial and interoperat or rel iabil it y t ests show a maximum
variat ion of 5° in a range of 56°, and t hese measurement s correl ate well wit h
cadaveric motion. 3
2 The rot amet er is rel at ivel y cumbersome, and it t akes about
3 minut es for an experienced physical t herapist t o use it in a clinical sett ing.
For t hese reasons, it has proven l ess useful as a clinical t ool . I t has t he additional
disadvant age t hat it s reading may be influenced t o a minor degree by l ower rib
cage movement s.
Ranges of lumbar movement s, for bot h sexes in six age-group categories
using t he spondyl ometer and the rotameter, and t he gravit y incl inomet er3
side fl exion, are l isted in Tabl e 2- 1 .
Age Changes in Ranges of Movements
Tabl e 2- 1 clearly demonst rat es a decline in t he ranges of all l umbar move­
ment s in t he l iving wit h increasing age. This decline paral l el s t he reductions
observed in cadaveric studies by other authors. 5
. 4
. 4
72 Physical Therapy of the Low Back
Fig. 2-10. The lumbar rotameter.
Table 2·1. The Mean and Standard Deviation for the Total Ranges of Sagittal, Horizontal, and
Coronal Plane Movements in Living Subjects (Population 960 Persons)
Ranges of Movements
Horizontal Range
Sagittal Range (Rotation to Both Coronal Range (Side
(Flexion-Extension) Sides) Flexion)
Age (yrJ Male Female Male Female Male Female
5-12 58° ± 9° 58° ± 9° 34° ± 6° 34° ±
° 47° ±
° 47° ±
13-19 45°
± 10° 57" ± 8° 30° ± 4° 34° ± 4° 38° ± 5° 37° ± 4°
19-35 42° ±
° 42° ± 7° 33° ±
° 33° ±
° 40° ± 5° 40° ± 5°
35-59 38° ± 7° 38° ± 7° 26° ±
° 27° ±
° 32" ± 4° 30° ± 3°
0+ 30° ± 7" 28° ±
° 22° ± 5° 20° ± 4° 28° ± 4° 30° ± 5°
Lumbar Posture, Movement, and Mechanics 73
In old age the ranges of lumbar movement in men and women become
almost ident ical. It would appear t hat when hormonal differences are reduced,
sexual different iations in vertebral shape, posture, and spinal-movement ranges
disappear. The general reduction in ranges in both sexes occurs as a result of
increased "stiffening" of t he int ervert ebral disc in association with disc-shape
changes involving increases in t he anteroposterior lengt h and concavit y of t he
vert ebral end-plat e. 33 A reason often provided for t he decline in average ranges
of movement s in aging populat ions (i.e., a general t endency t o t hinning of int er­
vert ebral discs in old age) has recently been shown t o be false. 3o.3
I n old age
most discs increase in volume and become t hicker and more convex at t he
disc-vert ebral interface. Only about 30 percent of discs become t hinner. The
principal r
ason for decreased range of movement is increased disc st iffness.33
The 40 percent increase in disc stiffness wit h age is associated wit h well
documented hist ologic and biochemical changes. These include an increase in
the total number of collagen fibers and in t he ratio of t ype I t o t ype I I collagen,
a decrease in wat er content and a change in t he proteoglycan rat ios where t he
proport ion of kerat an sulfat e; chondroitin sulfat e increases.
4 There is also an
associat ed increase in "fatigue failure" of collagen in older cartilage. It is uncer­
t ain whet her it is collagen fibers t hat undergo "fatigue" or split t ing or whether
it is t he bonds between adjacent collagen fibers t hat separate. Collectively,
t hese changes and t he associat ed decrease in compliance render t he disc fibro­
cart ilage less capable of acting efficient ly as a shock absorber or joint , and of
t ransmitt ing loads along t he vert ebral column.33
Lumbar Intersegmental Motion
It is an essent ial part of a physical t herapist ' s examinat ion t o det ermine
ranges of movement of t he whole lumbar spine. I n addit ion t o measurement of
these physiologic ranges of movement s, t he manipulat ive physical t herapist
always conduct s a manual assessment of lumbar int ersegment al motion. This
involves t he displacement of a lumbar mot ion segment by t he applicat ion of
an external, manual force applied direct ly t hrough t he spinous processes, or
indirectly via t he ligament s and joint s of t he adjacent vert ebrae. Small rotations
and t ranslat ions about and along t he axes of movement can be achieved in t his
way (see Fig. 2-8). No t echniques are current ly available for object ive clinical
local anomalies (increases or decreases' in movement) to be confident ly diag­
nosed. While these t echniques involve subject ive evaluat ion of vertebral mo­
tion, it is interest ing to not e it s current excellent correlat ion wit h other diagnos­
t ic studies. 3
Many clinical reports have associat ed eit her hypo- or hypermobilit y with
a variet y of lumbar disorders and low back pain.34 At t his point in t ime, t here
is very little real evidence linking ranges of mot ion and back pain,
anecdot al and clinical stories abound. However, Farrell and Twomey,
1 .82
a study of acute low back pain and manipulative t herapy, did show an improve­
ment in lumbar sagittal motion (measured manually) associat ed with improve-
74 Physical Therapy of the Low Back
ment in back pain symptoms. Giles and Taylorl 5 also showed increased lumbar
range of movement following recovery from episodes of low back pain, but
this increase was only in patients under the age of 50 years. Similarly, J ull88
has shown an increase in intersegmental motion associated with remission of
symptoms in patients with low back pain. I t is likely that current research will
throw further light on this question over the next few years.
The orientation of the lumbar articular processes facilitates sagittal move­
ment and allows for a considerable range of motion in this plane. From the
"normal" erect standing posture, flexion usually comprises about 80 percent
and extension 20 percent of the total range of sagittal movement. Flexion ceases
due to apposition of lumbar zygapophyseal joint surfaces and tightening of
posterior ligaments and muscles, whereas extension is blocked by bone contact
when the inferior joint facets come into contact with the laminae of the vertebra
below or the spinous processes meet. 33
Control of Flexion
Muscular Control
The lumbar back muscles exert considerable control over active ranges of
lumbar movement (see Ch. 4). Erector spinae and multifidis are principally
responsible for all movements72 by exerting an eccentric control (i. e. , by paying
out) on movements that are gravity assisted. Thus trunk flexion in standing or
sitting is controlled by an eccentric contraction of these muscle groups. I n
exerting this control, the muscles tend to restrict the total range of movements
possible, particularly in the sagittal plane. 3
79 This helps explain why cada­
varic studies show a slightly greater range of lumbar sagittal movement than
is usually recorded in the living. 5
I t has been shown that after suitable warm-up exercises, ranges of lumbar
flexion increase by a few degrees, 32 and that a change in posture from the
upright to the side-lying position brings about an additional increase, which
equates with the ranges observed in the cadaveric studies. I t would appear
that warm-up exercises achieve their effect by relaxation or stretching of the
sacrospinalis muscle group and it is not unreasonable to assume that the slightly
larger increase obtained in side lying is due to the elimination of antigravity
activity in the long back muscles. Research has shown64 an "electrically silent"
phase in the back muscles at the limit of lumbar flexion. Although they conclude
that the spine is supported passively by tension in postvertebral connective
tissue structures at this point, it may also be due in part to passive elastic tension
of the posterior muscles themselves. I ndeed, the apposed zygapophyseal facets
play the greater restraining roleY
Lllmbar Postllre, Movement, and Mechanics 75
Each l umbar multifidus muscl e attaches strongl y to a mamil l ary process
on a superior articul ar process and also into the capsul e of zygapophyseal joint.
I t acts as a rotator cuf muscl e and maintains the approximation and congruity
of the zygapophyseal facets on the posterol ateral aspect of the j oint (the l iga­
mentum fl avum maintains the articul ar surfaces in close apposition on the ante­
romedial side of the joint) . The cl ose relation of this muscl e to the joint capsul e
and its simil ar innervation woul d readil y explain how with other postvertebral
muscl es it woul d severel y l imit flexion and rotation in any painful condition of
the joints.
Other Factors Controlling Lumbar Flexion
In addition to the postvertebral muscles, the posterior elements of the
l umbar spine consist of a compl ex l igamentous system and the articul ating bony
arches. Over the years there have been a number of conflicting views on the
relative roles pl ayed by these posterior elements in l imiting and control ling the
range of lumbar fexion. I n general , it has been considered that the interspinous
and supraspinous l igaments and the strong el astic l igamentum flavum served
principal l y to act as a "brake" to fexion.
9 I n this regard, the elasticity of the
l igamentum favum was seen as important because tension increased as the
movement continued, whil e the dense, strong, and inelastic inter- and supraspi­
nous l igaments acted as a physical barrier to the movement.
Adams et al , 90 in a sequential posterior release experiment, quantitated
the rel ative parts pl ayed by the supraspinous-infraspinous l igaments, the l iga­
mentum favum, the zygapophyseal joint fibrous capsul e, and the intervertebral
disc in resisting fl exion of individual motion segments. They showed that the
joint fibrous capsule and intervertebral disc play the more important roles, with
the l igamentum flavum and spinous l igaments making l esser contributions. They
found it most surprising that the rel ativel y unimpressive fibrous capsul e shoul d
exert such l arge restraining forces, and noted that technical problems in section­
ing all capsul ar fibers made it difficul t to distinguish the rol e of capsul ar forces
from articul ar facet forces exerted through the articul ar processes.
In our study33 of the role of the posterior elements, each of the posterior
l igaments was sectioned in turn (supraspinous and interspinous, l igamentum
favum and capsul e) , as was the bony arch to assess the influence of each on
the range of l umbar fl exion. The range of flexion was measured before and
afer sectioning of each of the elements and the results are l isted in Table 2-2.
Table 2-2. Average Increases (Degrees) in Sagittal Range Following Section
Flexion Extension Sagittal Range
Section Increase Increase Increase
Supraspinous and interspinous ligaments 2.0 1.5 3.5
Ligamenta flava 2.5 1.0 3.5
Joint capsules 3.0 2.0 5.0
Pedicles 14.0 3.0 17.0
76 Physical Therapy of the Low Back
Analysis demonstrated small regular increases in sagittal range on each succes­
sive ligamentous release and a large abrupt increase in range following section
of the vertebral arches. Young and middle-aged subjects showed almost a 100
percent increase in lumbar sagittal range after removal of all posterior elements.
whereas elderly subjects showed a 60 percent increase (Fig. 2-11).
This study confirms that all the ligamentous elements offer some resistance
to lumbar fexion, with the joint capsules having the greatest influence, as sug­
gested by Adams, et al.9
However, by far the greatest restraining influence
on flexion is the pressure between the apposed articular facets of the zygapo­
physeal joints. Radiographic analysis of flexion in a young cadaveric lumbar
spine showed that the movement includes both forward rotation of a vertebra
on the vertebra beneath along an axis in the posterior part of the intervertebral
disc, and an associated forward translation or slide of the superior vertebrae
on the inferior vertebra (Fig. 2-12). The zygapophyseal joints guide the plane
of rotation and resist the forward slide. The coronally oriented anterior compo­
nent of each articular facet bears the resultant stress. When the pedicles are
cut, a greater degree of forward slide permits further rotation.
Joint loading in axial weightbearing has been described by Nachemson91
and by Shah et al.9
1t seems clear from our studies that flexion involves progres­
sive joint loading to the point where the horizontal moment prevents further
rotation from occurring. The lumbar vertebral arches through the zygapophy­
seal joints thus provide an essential restraint limiting or preventing the transmis­
sion of shearing forces to the intervertebral discs. This could lead to instability
(as in spondylolisthesis) with danger to the cauda equina and to the nerve roots
in the intervertebral canals.
-- --
.-- ----
-- -
o �--


Ín!Çc! ÍÍg�men cO5ue peOÍ�Íe
5pectmen teÍeO5e reÍeO5e 5ec!IOn
P 1Oung(14 - 22 YRS ) 5 cO5e5
A- -A NÍOOÍe ÞgeO( 35 - 58 YRS
6 cO5e5
+- -+ LÍOetÍj( 61 - 78 YRS) 5 cO5e5
Fig. 2-11. The effects of the release of the posterior vertebral elements on the range
of l umbar fexion. (From Twomey and Taylor,33 with permission. )
Lumbar Posture, Movement, and Mechanics 77
Fig. 2-12. Tracings of superimposed radiographi c plates showing the range of fexion
produced and indicating the types of movement i nvolved. ( From Twomey and Taylor,33
with permission. )
Age differences in response to the posterior sectioning procedure cited
above throw light on the effect of increased stiffness with aging in the interverte­
bral discs. Foll owing ligamentous rel ease and also more dramaticall y foll owing
pedicle section, the available increase in movement range is much reduced in
elderl y subjects compared to young subjects. This demonstrates that the in­
creased stiffness or t he reduction in disc compliance with aging is t he principal
reason for the observed decrease in lumbar range of movement. The important
conclusions drawn from the studies described above are summarized bel ow:
I. Lumbar sagittal movement involves both forward rotation and slide of
one vertebra on t he vertebra beneath around a coronal axis l ocated in the poste­
rior annul us of t he intervertebral disc.
2. Sagittal movement of the vertebral column is restricted more by progres­
sive increase in loading of apposed facetal joint surfaces in flexion than by
tension or compression in posterior ligaments.
3. The decline in sagittal range of movement in old age is principally due
to increasing stiffness in t he intervertebral disc.
Creep in the Lumbar Spine
We have argued (above) t hat the stiffness in intervertebral discs and the
progressive loading of the zygapophyseal joints are t he factors bringing t he
normal range of fexion to a halt. However, prolonged l oading in fexion (10
78 Physical Therapy of the Low Back
minut es or more) does produce further fexion of the spine. This movement is
due t o creep, which is t he progressive deformation of a st ructure under const ant
l oad when the forces are not l arge enough t o cause permanent damage to the
vertebral structures.
Axial Creep
I n t he normal erect posture, approximatel y 16 to 20 percent of axial com­
pressive l oad on t he l umbar spine is borne by zygapophyseal joint s, whil e the
rest is carried by t he int ervert ebral discs, 9
. 93 which are wel l suited t o this
purpose. When axial l oads to intervertebral discs are maintained, the discs
progressivel y lose height until t he chemical forces devel oped wit hin t hem equal
t hose mechanical forces appl ied exteral l y. 94 Provided t hat t he forces used are
bel ow the l evels that woul d cause permanent damage, the greater t he external
force then t he great er the l oss of height t hat occurs. 94, 95
During the day, a person' s body weight acts as an axial compression force
through the vertebral col umn, and the subsequent creep brings about a reduc­
t ion in stature. When body weight is rel ieved (e. g. , at night in bed) and axial
l oads reduced, t he int ervertebral discs and other sof t issues are abl e to rehy­
drate and stature increases. 9
It has recentl y been demonst rated that a period
of rest in full flexion brings about a more rapid increase in st ature than does
rest in t he ful l y extended position.
7 This presumabl y occurs because flexion
act s as a dis�racting force on t he l umbar region, causing the discs to "suck in"
water at a greater rate.
Creep in Flexion
When ful l -range l umbar fexion is maint ained under l oad for a period of
t ime, t he articul ar t riad is distorted so that t he anterior disc region is
"squeezed" while t he posterior region is stretched; the zygapophyseal joint
surfaces are compressed tight l y toget her as t he coronal part of the articul ar
surfaces bear most of t he load; the soft tissues adjust by creep. 91 , 94, 95 , 97
Creep in fexion is observed as progressive ventral movement , nto further
fl exion, so t hat the endpoint of fl exion is increased (i. e. , range increases) . The
amount of creep in t he el derl y is great er t han in t he young and both t he creep
and the recovery from creep t ake pl ace over a l onger period of time. During
the process fluid is extruded from the soft tissues and they become rel ativel y
deprived of their nutrition. 9
Repetitive l oading causes cartil age degeneration
and bone hypertrophy in the various el ements of t he art icular t riad. 99
I f the amount of creep invol ved after prol onged l oadbearing in fl exion is
considerabl e, then recovery back t o the original starting posture (hysteresis) is
extremel y slow. It takes considerabl e time for the soft tissues to imbibe fuid
after it has been expressed during prol onged fl exion l oading. Many occupat ional
groups (e. g. , stonemasons, brickl ayers, roofing carpent ers) regul arl y submit
Lllmbar Posture, Movement, and Mechanics 79
thei r l umbar spi nes to thi s category of i nsul t. They work wi th thei r lumbar
col umn ful l y flexed and under l oad for consi derable periods of t i me. There i s
often l i ttle movement away from the ful l y flexed posi t i on once i t has been
reached and l i ttl e opportuni t y for recovery between episodes of work i n this
posi ti on. It is therefore not at all surpri si ng to fi nd so many bri cklayers, for
i nstance, wi th chroni c back pai n and wi t h occasi onal epi sodes of acute pai n.
These occupat i onal groups need consi derable ergonomi c advi ce and requi re
alterati ons to thei r working condi t i ons i f t hi s si tuati on is to be rectified.95
Control of Extension
The Role of Muscles
The cont rol of lumbar extensi on has not been i nvestigated and anal yzed
to t he same extent as flexi on. In the erect standi ng or si tti ng postures, t he
movement is i ni ti ated by contracti on of t he l ong back-extensor muscl es, and
then control l ed by t he eccent ri c contraction of t he abdomi nal muscle group
once t he movement has begun. The range of ext ensi on from t he neutral erect
standing posi ti on i s much less t han the range of flexi on, but t he muscular control
mechani sms are very si mi lar.
Other Contributing Factors
It seems probable t hat the range of extensi on is not control l ed by l igamen­
t ous tensi on, but that i t ceases when the t wo i nferi or arti cular processes at any
l evel are forced agai nst the laminae of t he vertebrae bel ow or perhaps when
the spi nous processes "ki ss . " Thi s i s wi tnessed by the build-up i n compact
bone, whi ch is evi dent in the lamina beneath the i nferi or process. Extensi on
occurs al ong an axis in the posterior part of the i ntervertebral di sc at that l evel.
Thi s posi tion probabl y does not place the s oft ti ssues under the same constant
strain as flexion does except at t he l i mi t s of l ordosi s after prolonged standi ng,
when zygapophyseal joi nts probabl y take a larger amount of t he l oad of body
weight. When. hyperextensi on occurs, it is l i kel y that t he axis of movement
shifts even further posteri orl y and i s l ocated where the tips of the i nferi or facets
arti culate wi th the l ami nae. Thi s woul d cause stretchi ng of t he anteri or s oft
ti ssues, notabl y t he anteri or l ongi tudi nal l igament and the anteri or annul us fi ­
brosus, whi ch are ext remel y strong and capable of wi thstanding such forces.
It i s apparent from our i nvestigations that consi derable osteoarthri ti c change
takes place in t he articul ar carti lage and subchondral bone. Thi s change occurs
at t he pol ar region of the i nferi or recesses of t he zygapophyseal j oi nts corre­
sponding t o t he areas of extensi on i mpact and compressi on. 99
80 Physical Therapy of the Low Back
Creep in Extension
Prolonged maintenance of an extended posture or of l oading in extension
are rare, and few if any occupational groups have such a working situation.
The position of the line of gravity is described as passing anterior to the thoracic
vertebrae and through the l umbar vertebrae. 1 00 Theoretical ly, prol onged stand­
ing (i.e., axial l oading with body weight) woul d tend to increase thoracic kypho­
sis (by creep) , but not alter l umbar l ordosis. However, present evidence indi­
cates that in prol onged standing there is a tendency for the axial l oad of body
weight to increase l umbar l ordosis. I n this way, the zygapophyseal joints will
take an increasing proportion of the l oad of body weight. Al though- l ong, contin­
ued l ordotic posture and l oading are rare, a number of sports activities involve
full extension movements of an explosive nature. These may be repetitive
movements and may involve high peaks of loading in ful l extension. Thus, fast
bowl ers in cricket, gymnasts, and high j umpers are three groups that pl ace
tremendous impact forces through this posterior arch compl ex. At heel strike
during these sports, the chisel- l ike inferior articular processes are forced down
suddenl y into the laminae of the vertebrae below. The forces involved are very
considerable, as the l oad borne by the facets increases dramaticall y with the
amount of extension of the region.I OI Repetition over l ong periods of time re­
sults in soft-tissue infammation and bone sclerosis that may become obvious
on radiographic examination, but which may later resul t in fracture and perhaps
displ acement. The bone area that absorbs this force is the isthmus between the
zygapophyseal facets (i.e. , the pars interarticularis). As is wel l known, this is
the site at which spondylol ysis occurs. I t is even more l ikel y that the repetitive
combination of al ternative expl osive ful l extension fol l owed suddenl y by full
fexion places enormous strain on the pars interarticularis region. This exten­
sionlexion repetition moment at the pars may cause fatigue fracture in a similar
way to that of fatigue in metal caused by successive opposite movements.
Considerabl e research on the mechanics of this region has taken place in
recent years.93
The nucleus of the intervertebral disc is c ontained under pressure within
its protective fbrous and cartil aginous envelope. I ntradiscal pressure is a useful
index of disc function and has been shown to vary according to posture, move­
ment, external l oading, and age. Nachemson' s comprehensive study91 on l um­
bar intradiscal pressure in 1 28 discs from 38 cadavers of both sexes from 6 to
82 years of age concl uded the fol l owing:
I . The l oaded disc behaves hydrostatical ly in that the nucl eus acts as a
fluid, distributing external pressures equal ly in al l outward directions to the
annul us.
Lumbar Posture, Movement, and Mechanics 81
2. Axial loading produces l ower pressure readings in children under the
age of 1 6 years than in adults.
3. The level of the lumbar spine does not influence the pressures recorded
in "loaded" or resting discs (the L5-S I disc was not included in the s tudy).
4. The posterior vertebral s tructures (pedicles and articular processes) ab­
sorb 1 6 to 20 percent of the axial loading forces.
5. " Moderately degenerated" discs (as suggested by disc "thinning")
show simil ar pressure behavior to " intact" discs, and the mechanical behavior
of a lumbar disc does not change appreciably if ' ' degeneration" is not advanced.
Since Nachemson's original study, 9
it has been s hown in living subjects
that intradiscal pressures are higher in the sitting than the s tanding pos­
. 1
they are less in the physiologic lordotic posture than in the s traight
or kyphotic posturel03 ; pressures are increased with passive lumbar fl exion of
20° 1 04; they are further increased during active trunk fl exion exercises 1 0
; and
the largest increases accompany heavy lifting, particularly when the Val s alva
maneuver is performed
1 06 Nachemson et al30 and Merriam et all 0
7 showed that
abnormal degenerated discs did not behave in a consistent way, as they showed
patterns of pressure changes in different postures that were often dissimilar
from that shown by normal discs. Similarly, other studies have s hown that the
ability of the disc to withstand compressive forces depends on both the integrity
of the disc envelope and the turgor of the contained nucl eus pulpOSUS.
1 09
This contrasts with the claim of Belytschko et ai l i o that i n a theoretical model
annular tears would reduce intradiscal pressures more than degenerative nu­
clear lesions.
Clinical Considerations
The effects of different postures, exercises, and loading conditions on in­
tradiscal pressure are of interest to the physical therapist as an indicator of
how the disc responds to these variables and of the possible abnormal response
in disc degeneration. However, intradiscal pressure alone does not appear to
be able to indicate which activities are likely to be either s afe or dangerous for
a patient' s back. Although it is probably important to take the environmental
condition of fhe disc into account when prescribing exercise, as yet there are
no clear indications that any particular exercises are contraindicated on the
grounds that a rise in intradiscal pressure would be prejudicial to the disc or
risk injury to the vertebral end-plates.
It is quite cl ear that movements such as lumbar flexion and lateral bending
and tasks such as the lifting of heavy weights increase intradiscal pressure. For
this reason, many clinicians approach these activities with considerable care
and watch their patients very carefully when they begin such activities. Simi­
larly, patients need to be aware of the effect of these activities on their lumbar
spine and approach tasks that involve rises in intradiscal pressure with reasona­
ble care. I n the same way, because it is known that intradiscal pressure is
82 Physical Therapy of the Low Back
lowest in supine and prone lying, is lower in standing than in sitting, and remains
quite low in activities involving lumbar extension and rotation, then some clini­
cians may wish to use this information in the exercise programs t hey prescribe.
While t his may be useful, no direct link has yet been established between a rise
in intradiscal pressure and the production of a disc lesion. Similarly, ergonomic
advice as to lifting t echnique, seat design (including car seat design) , working
posture, and activities of daily living that emphasizes the maintenance of a
lordosis is often given on the basis of t he effects of these t asks on intradiscal
pressure. I t still remains uncertain just what role raised intradiscal pressure
may play in the production of disc lesions and low back pain or whether reduced
pressure may prevent disc lesions. Therefore, it remains an interesting observa­
tion t hat may ultimat ely prove to be of pract ical use, but at present its relevance
is uncertain.
Until quite recent times, intrat horacic and intraabdominal pressure have
been considered to be import ant in relieving t he spine of a large part of the
axial compression and shear loads by convert ing the t runk into a more solid
cylinder and t ransmitting part of the load over t he wider area. I t was considered
highly likely that intraabdominal pressure exert ed a major force in this regard
by t he simultaneous contraction of abdominal muscles, t he diaphragm, and t he
muscles of t he pelvic diaphragm. These are most ly t ransverse and obliquely
oriented muscles, all capable of exerting considerable torque and compressive
force on the cavity t hey enclose.
The t horacic spine bears less weight than the lumbar spine and is supported
ventrally and laterally by t he physical presence of t he ribs and to a lesser extent,
when the glot t is is closed, by a rise in intrat horacic pressure. The role of intraab­
dominal pressure in reducing forces acting on the spine and protecting vulnera­
ble vertebral bodies and int ervertebral discs from excessive loading was initially
investigated by Bartelink,
and since then has excit ed the imagination of other
1 06. 1 1 2-1 1 6
Contraction of the muscles of t he t runk cavity to raise intraabdominal pres­
sure probably functions as a protect ive reflex mechanism, both to protect ab­
dominal viscera from damage by a blow and to assist in protecting the vertebral
column from excessive loading. Thus, when loads are placed on the vertebral
column, t he muscles are involuntarily called into action to fix the rib cage and
to restrain and compress t he content s of the abdominal cavit y so t hat it becomes
like a "balloon." 1 1 5 The positions of the trunk and t he load influence the extent
of any rise in intraabdominal pressure. The greater t he spine is flexed and the
furt her the load is away from t he body, the greater is t he increase in intraabdom­
inal pressure required to balance t he load and help distribut e the compression
The mechanism may be compared to an inflat ed balloon, which act s on an
Lumbar Posture, Movement, and Mechanics 83
anterior moment arm t wo or t hree t imes the lengt h of t he posterior moment
arm of t he back ext ensor muscles. Pressures generat ed in t he abdominal cavity
will produce a net positive moment , and t end to restore t he lordotic curvature
of the lumbar spine. This will counteract t he flexion moments produced by t he
upper body and t he anterior load carried. Eiel 1 4 described t he relieving force
of intraabdominal' pressure as reducing by about 40 percent t he required com­
pressive effect of the cont raction force of t he erector spinae muscles. However,
recent studies have clearly shown t hat t he effect is nowhere near as substan­
0 They have shown a much smaller relieving net effect due to intratrunkal
pressure and have also clearly demonstrat ed t hat there is not a linear relation­
ship between t he increase in intrat runkal pressure and t he strength of contrac­
t ion of the abdominal musculature.3
1 1
I ntraabdominal pressure thus plays a small role in stabilizing t he spine and
pelvis at t he onset of the lift by resisting t runk flexion, although it probably
does little to reduce intervertebral compression forces. This helps allow t he
pelvis to be rot at ed backward and t he lumbar curve t o be flattened by t he
powerful gluteal and hamst ring muscles, which have a longer moment arm and
a great er cross-sectional area (and t hus power) t han do t he spinal extensors.
Thus they are the most suitable and capable muscles t o be recruited in initiating
the t ask of heavy lift ing,
l i S
and help t o reduce t he moment of the load, allowing
the erect or spinae muscles to t ake over and ext end t he spine on a stable
1 9
While the suggest ed ability of int rat runkal pressure t o relieve loads acting
on the lumbar spine received significant support in t he 30 years since Bartel­
t here are current ly many quest ions raised about it s validity. These have
arisen since studies of lift ing have not demonstrated suffciently large rises in
intraabdominal pressure t hat correlate wit h either t he size of t he load being
lift ed, or the stresses measured on t he vertebral column. 3 0.
7. 1
2 0 Similarly,
increasing intraabdominal pressure by using t he Valsalva maneuver may act u­
ally increase t he load on the lumbar spine, whereas strengthening abdominal
muscles in normal people or in those wit h back pain does not appear t o increase
the capacit y to raise intraabdominal pressure as measured during lifting. 3o.
1 1
20 It should also be t aken int o account t hat it has been clearly shown t hat
to raise int raabdominal pressure high enough t o generat e a sufficient antiflexor
moment , the pressure would be so high as t o obst ruct blood flow in t he vena
cava and abdominal aort a,
a point t hat Bart elink
not ed in his early study.
The lowest intraabdominal pressures have been recorded in t he smallest
people, whereas largest pressures are evident in t aller, heavier subj ect s.
St rong at helet es are able to produce enormous rises in intraabdominal pres­
Gait shows phasic changes in this pressure, with increases as t he speed
of the act ivit y increases. J umping in place or from a height raises t he pressure,
as do pushing and pulling activit ies. I t is uncertain whet her or not t he Val salva
maneuver or t he use of a lumbosacral corset does any more t han produce mar­
ginal increases in intraabdominal pressure.
84 PhysicaL Therapy of the Low Back
Clinical Considerations
I n the past, intraabdominal pressure was considered to be a potent influ­
ence for reducing the loads applied to the spine. This rationale was used to
explain the need for the development of strong abdominal musculature sur­
rounding the abdominopelvic cavity. However, more recent research has dem­
onstrated that intraabdominal pressure does not dramatically reduce loads on
the spine. The previous estimates of the loads generated at the LS-S 1 junction
during maximal lifting were an overestimate and the back musculature has been
shown to be stronger and able to generate a considerably greater power than was
first estimated. 1
7 These observations, together with a greater understanding of
the role of the thoracolumbar fascia in assisting in distributing a small percent­
age of the load and in "tying" the long back muscles down (see Ch. 4) has
meant that the inherent strength of the back mechanism as a whole is better
recognized. Similarly, Waddelll
3 has demonstrated that those with back pain
respond very favorably to programs of intensive exercise, including a strong
trunk- strengthening program. This adds weight to the argument that musculo­
skeletal fitness is a major factor in the management of back pain and lifting
disorders. The role of intraabdominal pressure in this equation is now under
serious debate and woul d benefit from further research. I t may be that improve­
ments in abdominal muscle strength achieve an effect through the better control
of pelvic and spinal posture rather than by a greater capacity to raise intraab­
dominal pressure.
I n 1 983, physical therapists Giovanelli, Thompson, and Elveyl
4 conducted
a pilot trial investigating lumbar zygapophyseal intracapsular joint pressures in
living subjects. They placed two needles within the joint under radiographic
control, one needle to inject saline and the other to record pressure changes.
They showed that there is no intracapsular pressure at rest. Once fluid was
injected into the joint, most active and passive movements caused a drop in
the pressure produced by the injection, the pressure rising again on return to
the starting position. The greatest drops in pressure occurred when passive
techniques were directed specifically to the joint concerned. This highlighted
a possible mechanism of pain relief by the use of localized manipUlative and
mobilizing techniques because raised intracapsular pressure with outpouring of
fluid may result from some forms of joint pathology. This pilot trial needs to
be extended considerably before much regard can be placed on its conclusions,
but it does provide interesting information useful to the manipulative physical
therapist on the ways in which manipulative techniques directly infuence the
joints moved. The anatomic studies of the fat-pads of the lumbar facet polar
recesses show that in movement, the fat moves in and out of the joint in re­
sponse to any changes in pressure.
Lumbar Posture, Movement, and Mechanics 85
As descri bed in Chapter I, t he i nternal architecture of lumbar vertebrae
consists of verti cal bony trabeculae (beams or strut s of bone) supported by
hori zontal trabeculae, which are aligned parallel to the li nes of stress. Thus t he
vertical trabeculae absorb the axial loads of weight bearing, and transmit t he
load downward and outward to t he vertebral shell via the transverse trabeculae,
which resist buckling of the vertical weight-bearing beams. I t i s li kely t hat the
horizontal trabeculae are also important in absorbing and transmitti ng the lateral
forces appli ed through the body as a consequence of muscular acti vity. Old
age is associ ated wi t h a signi fi cant selecti ve decli ne in the numbers of hori zontal
trabeculae. The compressi ve load of body weight, which is usually mai ntai ned
in old age, brings about fractures of the now less-well-supported vert ical trabec­
ulae, and collapse of the vertebral end-plate. Lumbar vertebrae become shorter
and wi der in old age, and more concave at the disc-vertebral juncti on.
Thi s pattern of selecti ve bone loss and associ ated changes in vertebral
body shape is part of the general pi cture of osteopeni a and osteoporosi s seen
i n t he elderly. I n Western society, at age 65 radiographic compari son with a
"standard" suggest s that 66 percent of women and 22 percent of men have
osteoporosis. I n women the i nci dence i ncreases by about 8 percent for each
addi ti onal decade, whereas a large i ncrease does not occur unti l afer the age
of 76 in men.
26 The principal si tes of fracture and pai n due to osteoporosis are
the vertebral column, the distal radius, and t he neck of t he femur. I t causes
over 200, 000 hi p fractures annually in the Uni ted States. Pain and shortened
stature accompani ed by "dowager' s hump" or hunchback i n elderly women
are maj or symptoms of advanced osteoporosis, which often leads to vertebral
collapse and funct i onal disability.
The prevention of osteoporosi s at present focuses on the need for relatively
high levels of dietary calcium (1000 to 1 200 mg/day), particularly in women, and
for estrogen replacement therapy i n some women.
9 Recently, considerable
attenti on has also been pai d t o the i mportant role of exercise i n preventi on.
I mportant st udies by Aloia et al1 3
and Smith et al
have shown bone gain to
foll ow exercise even in very elderly subjects. Physical therapists deali ng with
t he prevent ion and treatment of back pai n and disabi li t y need to stress these
factors with thei r mi ddle-aged and elderly pat ients. There i s no doubt that bone
loss occurs in the absence of physi cal activit y, and t hat bone hypertrophy fol­
lows i ncreased exerci se activi ty.
I t i s likely t hat t he incidence of osteoporoti c
bone fractures in the elderly could be reduced i f regular exercise was generally
mai ntained i nto old age. Thi s reduced ri sk of fracture may relate as much to
the mai ntenance of muscle strength and neuromuscular coordi nation as to the
associated maintenance of bone mass.
I . Kendall FP, McCreary EK: Muscl es, Testing and Functi on. 3rd Ed. Wi l l i ams &
Wi l ki ns, Baltimore, 1 983
2. Schmorl G, Junghanns H: The Human Spine i n Health and Disease. 2nd Ed. Grune
& Stratton, New York, 1 97 1
86 Physical Therapy of the Low Back
3. Horak FB: Cl i nical measurement of postural control in adul ts. Phys Ther 67: 1 88 1 ,
1 987
4. Taylor JR: Growth of human IN discs and vertebral bodi es. J Anat 1 20: 49, 1 975
5 . Twomey LT: Age changes i n the human l umbar spi ne. PhD Thesis, University of
Western Austral ia, 1 98 1
6 . Landau BR: Essential Human Anatomy and Physiology. 2nd Ed. Scott Foresman,
Gl enview, I L, 1 980
7. Treanor WJ: Motions of the hand and foot . [n Licht SH ( ed): Therapeutic Exer­
ci se. Waverly Press, Balti more, 1 965
8. Opil a KA, Wagner SS, Schiowit Z, Chen J : Postural alignment in barefoot and
high heel stance. Spine 1 3: 542, 1 988
9. Youngman A, El l iott M: The effect of high heel shoes on l umbar lordosi s. BAppSc
project , Western Australian I nsti tute of Technology, 1 985
1 0. Romanes CJ : Cunningham' s Textbook of Anatomy. 1 1 th Ed. Oxford Uni versity
Press, Oxford, 1 972
I I . Taylor JR, Al exander R: BSc project, Uni versity of Western Australia, 1 983
1 2. Ostgaard HC, Andersson GBJ , Karlsson K: Prevalence of back pain in pregnancy.
Spi ne 1 6: 549, 1 991
1 3 . Day J W, Smi dt GL, Lehmann T: Effect of pelvic ti l t on standing posture. Phys
Ther 64: 5 1 0, 1 984
1 4. Hansson T, Sandstrom J , Roos B et al : The bone mineral content of the lumbar
spine i n pati ents with chronic low back pai n. Spine 1 0: 1 58, 1 985
1 5 . Gi l es LGF, Taylor JR: Low back pain associated with leg length i nequal i ty. Spine
6: 5 1 0, 1 98 1
1 6. Wi l l ner S : A study of height, weight and menarche i n girls with idiopathic structural
scoliosi s. Acta Orthop Scand 46: 7 1 , 1 975
1 7. Taylor J R, Halliday M: Limb l ength asymmetry and growt h. J Anat 1 26: 634, 1 978
1 8. I ngelmark BE, Li ndstrom J : Asymmetries of the lower extremi ti es and pelvis and
thei r rel ati onship to lumbar scoli osi s. Acute Morphol Neerl Scand 5/6: 227, 1 963
1 9. Taylor J R, Slinger BS: Scoliosis screening and growth in Western Australian stu­
dents. Med J Aust 1 7: 475, 1 980
20. Gi l es LGF, Taylor JR: I ntra-articul ar synovial protrusi ons. Bull Hosp J Dis Orthop
Inst 42: 248, 1 982
2 1 . Gi l es LGF, Taylor JR: The effect of postural scoliosis on lumbar apophyseal joints.
Scand J Rheumatol 1 3 : 209, 1 984
22. Terjesen T, Benum P, Rossvoll l et al: Leg length discrepancy measured by ultraso­
nography. Acta Orthop Scand 62: 1 2 1 , 1 991
23 . Sicuranza BJ, Richards J , Tisdall LH: The short leg syndrome i n obstetrics and
gynaecology. Am J Obstet Gynecol 1 07: 2 1 7, 1 970
24. De Puky MD: Diurnal variation in stature. Acta Orthop Scand 6: 338, 1 935
25 . Bl ackman J : Experimental error i nherent i n measuring the growing human being.
p. 389. In Boyd E ( ed) : Am J Phys Anthropol 1 3, 1 924
26. Stone M, Taylor J R: Factors influencing stature. BSc anatomy dissertation, Uni­
versity of Western Australia, 1 977
27. Tyrrell AP, Rei l l y T, Troup JDG: Ci rcadian variation i n stature and the effects of
spi nal loading. Spi ne 1 0: 1 61 , 1 985
28. Adams MA, Hutton WC: The effect of posture on the lumbar spine. Bone Joint
Surg 67B: 625, 1 985
29. Sul l i van WE, Mi l es M: The l umbar segment of the vertebral col umn. Anat Rec
1 33: 61 9, 1 959
Lumbar Posture, Movement, and Mechanics 87
30. Nachemson AL, Schultz AB, Berkson MH: Mechanical properties of human l um­
bar spine motion segments. Spine 4: I , 1 979
3 1 . Twomey L T, Taylor JR: Age changes in the l umbar i ntervertebral di scs. Acta
Orthop Scand 56: 496, 1 985
32. Taylor J R, Twomey L T: Sagittal and horizontal plane movement of the human
l umbar vertebral col umn i n cadavers and i n the l i vi ng. Rheumatol Rehabi l 1 9: 223,
1 980
33. Twomey L T, Taylor JR: Sagittal movements of the human l umbar vertebral col­
umn: a quantitative study of the role of the posterior vertebral elements. Arch
Phys Med Rehabil 64: 322, 1 983
34. Whi te AA, Panjabi MM: The Cl i nical Biomechanics of the Spi ne. JB Li ppincot t ,
Philadel phia, 1 978
35. Dunham WF: Ankylosing spondyl i ti s: measurement of hip and spi ne movements.
Br J Phys Med 1 2: 1 26, 1 949
36. Lindahl 0: Determination of the sagittal mobi l i ty of the l umbar spi ne: a cl i nical
method. Acta Orthop Scand 37: 24 1 , 1 966
37. Leighton JR: The Leighton fexometer and flexi bi l i ty test. J Assoc Phys Ment
Rehabil 20: 86, 1 966
38. Gregersen G, Lucas DB: An in vi vo study of the axial rotation of the human
thoracolumbar spine. J Bone Joint Surg 49A: 247, 1 967
39. Loebl WY: Measurement of spinal posture and range of spinal movement. Ann
Phys Med 9: 1 03 , 1 967
40. Loebl WY: Regional rotation of the spi ne. Rheumatol Rehabil 1 2: 223, 1 973
4 1 . Macrae I F, Wright V: Measurement of back movement . Ann Rheum Di s 28: 584,
1 969
42. Moll J MH, Li yanage SP, Wright V: An objecti ve cli ni cal method to measure spinal
extension. Rheum Phys Med 1 1 : 293 , 1 972
43 . Moll J, Wright V: Measurement of spinal movement. p. 93 . In Jayson M ( ed) : The
Lumbar Spine and Back Pai n, Pitman Medical , Kent, England, 1 976
44. Wil es P: Movements of the l umbar vertebra duri ng fexi on and extension. Proc R
Soc Med 28: 647, 1 935
45 . Gianturco C: A roentgen anal ysis of the motion of the lower l umbar vertebrae.
Am J Roentgenol 52: 26 1 , 1 944
46. Hasner E, Schali ntzek M, Snorrason E: Roentgenological exami nati on of the func­
tion of the lumbar spine. Acta Radiol 37: 1 4 1 , 1 952
47. Tanz SS: Motion of the l umbar spi ne. A roentgenologic study. Am J Roentgenol
69:399, 1 953
48. All brook D: Movements of the l umbar spinal col umn. J Bone Joint Surg 39B: 339,
1 957
49. Rolander SD: Motion of the l umbar spine with special reference to the stabilising
effect of posterior fusi on. Acta Orthop Scand, suppl . 90, 1 966
50. Troup J DG, Hood CA, Chapman AE: Measurements of the sagittal mobi l i ty of
the lumbar spine and hi ps. Ann Phys Med 9: 308, 1 967
5 1 . Froning EC, Frohman B: Motion of the l umbosacral spine after l aminectomy and
spine fusion. J Bone Joint Surg 50A: 897, 1 968
52. Virgin WJ : Experimental i nvestigations into the physical properties of the interver­
tebral disc. J Bone Joint Surg 33B:607, 1 95 1
53. Hi rsch K: The reaction of the intervertebral di scs to compression forces. J Bone
Joint Surg 37A: 1 1 88, 1 955
54. Hi rsch K, Nachemson A: A new observation on the mechanical behavior of l umbar
di scs. Acta Orthop Scand 23: 254, 1 954
88 Physical Therapy of the Low Back
55. Brown T, Hansen RJ , Yorra AJ : Some mechanical tests on the lumbosacral spine
with parti cul ar reference to the i ntervertebral di scs. J Bone Joi nt Surg 39A: 1 1 35,
1 957
56. Evans FG, Li ssner HR: Biomechanical studies on t he lumbar spine and pel vi s. J
Bone Joi nt Surg 4 I A: 278, 1 959
57. Roaf R: Vertebral growth and its mechanical control . J Bone Joint Surg 42B: 40,
1 960
58. Galante JO: Tensile properties of the human lumbar annul us fibrosi s. Acta Orthop
Scand, suppl . 1 00: I , 1 967
59. Whi te AA: Anal ysi s of the mechanics of the thoracic spine i n man. Acta Orthop
Scand, supp! . 1 27, 1 969
60. Farfan HF, Cossette JW, Robertson GH et al : The effects of torsion on the lumbar
intervertebral joi nts: the role of torsion i n the production of di sc degenerati on. J
Bone Joi nt Surg 52A: 468, 1 970
6 1 . King AL, Vul can AP: Elastic deformation characteristics of the spine. J Biomech
4: 4 1 3 , J 97 1
62. Kazarian L: Dynamic response characteristics of the human l umbar vertebral col­
umn. Acta Orthop Scand, supp! . 1 46: I , 1 972
63 . Panjabi MM: Experi mental determi nation of spinal moti on segment behavior. Or­
thop Cl i n North Am 8: 1 69, 1 977
64. Twomey LT, Taylor JR: Physical Therapy of the Low Back. Churchill Li vi ngstone,
New York, 1 987
65 . Keegan 11: Alterati ons of the l umbar curve related to posture and seating. J Bone
Joint Surg 35A: 589, 1 953
66. Davis PR, Troup JDG, Burnard J H: Movements of t he thoracic and lumbar spine
when l ifting: a chrono-cyclophotographic study. J Anat 99: 1 3 , 1 965
67. Schultz AB, Bel ytschko TP, Andriacchi TP et al : Analog studies of forces i n the
human spine: mechanical properti es and motion segment behaviour. J Biomech 6:
373 , 1 973
68. Panjabi MM: Three-di mensi onal mathematical model of the human spi ne structure.
J Biomech 6: 67 1 , 1 973
69. Bel ytschko TB, Andriacchi TP, Schultz AB et al: Analog studies of forces i n the
human spi ne: computational techni ques. J Biomech 6: 36 1 , 1 973
70. Panjabi MM, Brand RA, Whi te AA: Mechanical properties of the human thoracic
spi ne. J Bone Joi nt Surg 58A: 642, 1 976
7 1 . Panjabi MM, Krag MH, White AA: Effects of preload on load di splacement curves
of the l umbar spi ne. Orthop Cl i n North Am 8: 1 8 1 , 1 977
72. Bogduk N, Macintosh J E, Pearcy MJ: A uni versal model of the lumbar back mus­
cles i n the upright posi ti on. Spine 1 7: 897, 1 992
73. Begg AG, Falconer MA: Pl ain radiographs in i ntraspinal protrusion of lumbar inter­
vertebral di scs: a correlation with operative fi ndings. Br J Surg 36: 225, 1 949
74. Lovett RW: A contribution to the study of the mechanics of the spine. Am J Anat
2: 457, 1 902
75 . Tondury G: Functional anatomy of the smal l joi nts of the spine. Ann Med Phys
1 5 : 2, 1 97 J
76. Kapandji I A: The Physiology of the Joi nts. 2nd Ed. Vol . 3 . Trunk and Vertebral
Col umn. Churchi l l Li vi ngstone, London, 1 974
77. Lewi n T, Moffett B, Vi i di k A: The morphology of l umbar synovial intervertebral
joi nts. Acta Morphol Neerl Scand 4: 229, 1 961
LlImbar Posture, Movement, and Mechanics 89
78. Rissanen PM: The surgical anatomy and pathology of the supraspi nous and inter­
spinous ligaments of the l umbar spine with special reference to l igament ruptures.
Acta Orthop Scand, suppJ . 46, 1 960
79. Bogduk N, Twomey LT: Cl i ni cal Anatomy of the Lumbar Spi ne. 2nd Ed. Churchi l l
Li vingstone, Mel bourne, Austral i a, 1 99 1
80. Lumsden RM, Morris J M: An i n vi vo study of axial rotation and i mmobil isation
at the l umbosacral joint. J Bone Joi nt Surg 50A: 1 591 , 1 968
8 1 . Farrell JP, Twomey LT: Acute low back pai n: comparison of two conservat ive
treatment approaches. Med J Aust I : 1 60, 1 982
82. Farrell J: A comparison of two conservative treatment approaches to acute l ow
back pai n. MAppSc Thesi s, Wester Austral ian I nstitute of Technology, 1 982
83 . Nachemson A: Lumbar spine i nstabi l i ty: a critical update and symposi um sum­
mary. Spine 1 0: 290, 1 985
84. Adams P, Eyre DR, Mui r H: Biomechanical aspects of devel opment and ageing
of human l umbar i ntervertebral di scs. Rheum Rehabi l 1 6: 22, 1 977
85 . Bushel l GR, Ghosh P, Tayl or TFK et al : Proteoglycan chemi stry of the i nterverte­
bral disc. Clin Orthop 1 29: 1 1 5 , 1 977
86. Nachemson AL: The l umbar spine: an orthopaedic chal lenge. Spi ne 1 : 59, 1 976
87. Hi lton RC, Ball J , Benn RT: I n-vi tro mobi l ity of the l umbar spine. Ann Rheum
Dis 38: 378, 1 979
88. Jul l G: The changes with age in l umbar segmental motion as assessed by manual
examination. Master' s Thesi s , Uni versity of Queensl and, 1 985
89. Ramsey RH: The anatomy of the l igamenta flava. Clin Orthop 44: 1 29, 1 966
90. Adams MA, Hutton WC, Stott MA: The resistance of flexi on of the l umbar i nter­
vertebral joi nt. Spine 5: 245, 1 980
9 1 . Nachemson A: Lumbar intradiscal pressure. Acta Orthop Scand, suppJ . 43, 1 960
92. Shah JS, Hampson WGJ , Jayson MI V: The distribution of surface strain in the
cadaveric l umbar spine. J Bone Joi nt Surg 60B: 246, 1 978
93 . Mi l l er JAA, Haderspeck KA, Schul tz AB: Posterior el ement l oads i n l umbar mo­
ti on segments. Spine 8: 33 1 , 1 983
94. Kazarian LE: Creep characteristics of the human spi nal col umn. Orthop Cl i n North
Am 6: 3 , 1 975
95 . Twomey LT, Taylor J R, Oi l ver MJ : Sustained fl exi on l oading, rapid extension
l oadi ng of the l umbar spine and the physical therapy of rel ated i njuries. Physiother
Pract 4: 1 29, 1 988
96. Shirazi-Adl A: Finite el ement si mulation of changes i n the fui d content of interver­
tebral di scs. Spine 1 7: 206, 1 992
97. Twomey L T, Taylor JR: Fl exi on creep deformation and hysteresi s in the l umbar
vertebral col umn. Spine 7: 1 1 6, 1 982
98. Adams MA, Hutton WC: The effect of posture on the flui d content of l umbar
i ntervertebral di scs. Spine 8: 665 , 1 983
99. Twomey LT, Taylor JR: Age changes i n the l umbar articular triad. Aust J Physi ­
other 3 1 : 1 06, 1 985
1 00. Pearsal l OJ , Reid JG: Li ne of gravity relative to upright vertebral posture. Cl i n
Biomech 7: 80, 1 992
1 01 . Yang KH, King AI : Mechanism of facet load transmi ssi on as a hypothesi s for l ow
back pai n. Spine 9: 557, 1 984
1 02. Nachemson A, Morris J M: I n vi vo measurements of intradiscal pressure. J Bone
Joi nt Surg 46A: 1 077, 1 964
90 Physical Therapy of the Low Back
1 03 . Andersson BJG, Ortengren R, Nachemson A et al : Lumbar di sc pressure and
myoelectric back muscl e acti vi ty during si tti ng. Scand J Rehabil Med 6: 1 04, 1 974
1 04. Nachemson A: The effect of forward bearing on l umbar i ntradiscal pressure. Acta
Orthop Scand 35: 3 1 4, 1 965
1 05. Nachemson A, Elfstrom G: Intravital dynami c pressure measurements i n l umbar
di scs. Scand J Rehabi l Med, supp\. I : I , 1 970
1 06. Andersson B1G, Ortengren R, Nachemson A: Quantitative studies of back loads
in l ifti ng. Spi ne I : 1 78, 1 976
1 07. Merriam WF, Qui nnel l RC, Stockdale HR et al : The effect of postural changes on
the i nferred pressures wi thi n the nucl eus pul posus during l umbar di scography.
Spine 9: 406, 1 984
1 08. Virgin WJ : Anatomical and pathological aspects of the i ntervertebral di sc. I ndian
J Surg 20: 1 1 3, 1 958
1 09. Panjabi MM, Krag MH, Chung TQ: Effects of di sc injury on mechanical behavior
of the human spi ne. Spine 9: 707, 1 984
1 1 0. Bel ytschko T, Kul ak RF, Schul tz AB et al: Fi ni te element stress anal ysis of an
intervertebral di sc. J Biomech 4: 277, 1 974
I I I . Bartel i nk DL: The role of abdominal pressure i n rel i evi ng the pressure on the
lumbar i ntervertebral di scs. 1 Bone 10int Surg 39B: 7 1 8, 1 957
1 1 2 . Morris 1 M, Lucas DB, Bresl er B: Rol e of the trunk i n stabi l i ty of the spi ne. 1 Bone
10i nt Surg 43A: 328, 1 961
1 1 3 . Davi s PR, Troup lDG: Pressures i n the trunk cavities when pul l i ng, pushing and
l i fti ng. Ergonomics 7: 465 , 1 964
1 1 4. Ei e N: Load capacity of t he l ow back. J Osl o City Hosp 1 6: 73, 1 966
1 1 5 . Kumar S, Davis PR: Lumbar vertebral i nnervation and i ntra-abdominal pressure.
1 Anat 1 1 4: 47, 1 973
1 1 6. Grillner S, Nil sson J, Thorstensson A: I ntra-abdominal pressure changes during
natural movements i n man. Acta Physi ol Scand 1 03: 275, 1 978
1 1 7. McGi l l SM, Norman RW: Portioning of the L4-L5 dynamic movement into di sc,
ligamentous and muscul ar components duri ng l i fti ng. Spine I I : 666, 1 986
1 1 8. Farfan HF: Muscul ar mechani sm of the l umbar spine and the position of power
and effici ency. Orthop Cl i n North Am 6: 1 35, 1 975
1 1 9. Ortengren R, Andersson GB1: Electromyographic studies of trunk muscles, with
special reference to the functi onal anatomy of the l umbar spine. Spine 2: 44, 1 977
1 20. Ortengren R, Andersson GB1, Nachemson AL: Studies of relationships between
l umbar disc pressure, myoelectric back muscl e acti vi ty and i ntra-abdominal pres­
sure. Spi ne 6: 98, 1 98 1
1 2 1 . Gracovetsky S , Farfan HF: The opti mum spine. Spi ne 1 1 : 543 , 1 986
1 22. Davis PR, Stubbs DA, Ridd J E: Radio pi l l s: thei r use in monitoring back stress.
1 Med Eng Technol 1 : 209, 1 977
1 23 . Waddel l G: A new cl inical model for the treatment of low back pai n. Spine 1 2:
632, 1 987
1 24. Giovanel l i B, Thompson E, El vey R: Measurement of variations i n l umbar zygapo­
physeal joi nt intracapsular pressure: a pil ot study. Aust J Physiother 3 1 : 1 1 5, 1 985
1 25 . Twomey LT, Taylor lR, Furni ss B: Age changes in the bone density and structure
of the l umbar vertebral col umn. J Anat 1 36: 1 5 , 1 983
1 26. Ei sman J A: Osteoporosis: preventi on, preventi on and preventi on. Cur Ther 33:
25, 1 992
1 27. Pardini A: Exerci se, vi tal i ty and aging. Age Ageing 344: 1 9, 1 984
1 28. Di xon AStJ: Non hormonal treatment of osteoporosi s. Br Med 1 286:999, 1 983
Lumbar Posture, Movement, and Mechanics 9 1
1 29. Spencer H, Kramer L, Lesniak M et al : Calcium requirements i n humans. Cl i n
Orthop 1 84: 270, 1 984
1 30. Aloia JF, Cohn SH, Ostuni JA et al : Prevention of i nvolutional bone l oss by exer­
cise. J Clin Endocrinol Metab 43 :992, 1 978
1 3 1 . Smith EL, Reddan W, Smi th PE: Physical activity and cal ci um modalities for bone
mineral increase in aged women. Med Sci Sports Exerc 1 3 : 60, 1 98 1
1 32. Menard D, Stanish WD: The aging athl ete. Am J Sports Med 1 7: 1 87, 1 991
3 Innervation, Pain
Patterns, and
Mechanisms of Pain
Nikolai Bogduk
Fundamental to the interpretation of lumbar pain syndromes is a knowledge
of the mechanisms by which local and referred pain can be produced. If the
mechanisms involved in a particular patient are properly understood, then treat­
ment can be prescribed logically and appropriately. However, when interpreta­
tions are based on misconception or restricted knowledge, there is a risk of
treatment being inappropriate and unsuccessful.
What has compromised the evolution of thought on lumbar pain syndromes
has been a tendency simply to infer or deem that a particular mechanism or
cause is responsible for a particular syndrome, without actually proving it to
be so. Indeed, if reiterated often and strongly enough, such inferences, even
if incorrect, seem to gain the respectability of a "rule" or "dogma," and some­
times become so "sacred" as to be exempt from challenge. Yet in some in­
stances, new facts or correct logic expose the limitations or errors in concepts
evolved in this way.
As far as possible, concepts should be based on scientific fact. This chapter
presents those experimental observations that are relevant to the comprehen­
sion of the mechanisms of lumbar pain syndromes, to provide a rational basis
for interpretation and challenging or dispelling certain common misconceptions
that are no longer tenable.
94 Physical Therapy of the Low Back
For any structure to be a source of pain it must be connected to the nervous
system; conversely, any structure that has a nerve supply is a potential source
of pain. For this reason, the innervation of the lumbar spine has been the subject
of several studies
-4 and reviews5
6; only a summary is provided here.
The posterior elements of the lumbar vertebral column are those parts that
lie dorsal to the intervertebral foramina, and they are all innervated by branches
of the dorsal rami of the lumbar spinal nerves. The dorsal rami themselves are
very short nerves directed backward and caudally through the intertransverse
spaces (Fig. 3-1). As each dorsal ramus approaches the subjacent transverse
process, it divides into two or three branches.3 Lateral branches are distributed
to the lateral column of the lumbar erector spinae-the iliocostalis muscle. The
lateral branches of the Lito L3 dorsi rami emerge from this muscle and cross
the iliac crest to become cutaneous over the buttock.3.7 Intermediate branches
of the lumbar dorsal rami arise independently from each dorsal ramus or from
l1 vn
ioø ¡¸¸
Ls vn
Fig. 3-1. Left posterolateral view of
the lumbar dorsal rami showing the dis­
position of their branches. VR, ventral
ramus; DR, dorsal ramus; mb, medial
branch; ib. intermediate branch; Ib, lat­
eral branch; ibp, intermediate branch
plexus; is, interspinous branch; a, artic­
ular branch; ZJ, zygapophyseal joint.
(From Bogduk and Twomey,52 with per­
mission. )
Innervation, Pain Patters, and Mechanisms of Pain Production 95
its lateral branch3 and are distributed to the medial column of the lumbar erector
spinae-the longissimus thoracic muscle. 8
The medial branches of the lumbar dorsal rami have been considered the
most relevant clinically; in addition to innervating the multifidus muscle and
the interspinous muscles and ligaments they supply the lumbar zygapophyseal
The anterior elements of the lumbar vertebral column are the vertebral
bodies and the intervertebral discs and their related ligaments, but include also
the anterior aspect of the lumbar dural sac, which not only lies anterior to the
spinal nerves and their roots, but is innervated by branches of the same nerves
that innervate the other anterior elements. No nerve endings occur in the poste­
rior dura4.13; therefore, for present purposes, no controversy need be raised as
to whether the posterior dura should be classified as an anterior or posterior
The anterior elements of the lumbar vertebral column are innervated by
dense, microscopic plexuses of nerves stemming from the lumbar sympathetic
trunk, the gray rami communicantes, and the ventral rami of the lumbar, spinal
nerves. An anterior plexus accompanies the anterior longitudinal ligament along
the front of the vertebral column and a posterior plexus accompanies the poste­
rior longitudinal ligament4 (Figs. 3-2 and 3-3). Within the posterior plexus cer­
tain larger nerves can be identified on microdissection, and in the past have been
referred to as the sinuvertebral nerves2.8.9.IO
4 (see Fig. 3-3). They represent the
main sources of the posterior plexus and stem from the gray rami communican­
tes and the lumbar ventral rami. They reach the plexus by passing through the
intervertebral foramina.
From both the anterior and posterior plexuses many fine branches enter
the vertebral bodies and intervertebral discs from their posterior, anterior, and
lateral aspects (Fig. 3-4). Osseous branches accompany vessels into the verte­
bral bodies and penetrate deeply into the spongiosa. Branches to the discs
penetrate only the outer third or so of the annulus fbrosus and do not extend
further than the outer half of the annulus. Ligament branches innervate the
posterior and anterior longitudinal ligaments. From the posterior plexus,
branches innervate the anterior aspect of the dural sac and the dural sleeves
of the lumbar nerve rootS. 4
Histologic studies have demonstrated nerve fbers and nerve endings not
only in the superficial laminae of the annulus fibrosus,
-18 but also as deeply
as the outer third or outer half of the annulus.
1 .2.
.1 9 More comprehensive re­
views of this issue are available elsewhere. 5.6
This summary of neurologic anatomy establishes that the possible sources
of lumbar pain are the lumbar zygapophyseal joints, the various back muscles,
the interspinous ligaments (all innervated by dorsal rami), the vertebral bodies,
intervertebral discs, longitudinal ligaments, and the dura mater (all innervated
by ventral rami). Anatomy alone, however, does not prove that a structure can
be a source of pain. It shows only that the necessary nerve supply is present.
Additional physiologic evidence is needed to show that stimulation of a struc­
ture can actually cause pain.
96 Physical Therapy of the Low Back
Fig. 3-2. The nerve plexus accompanying
the anterior longitudinal ligament at the lev­
els of the L3 and lower vertebrae, as seen in
whole mounts of human fetuses. ST, lumbar
sympathetic trunk. (Adapted from Groen et
al,4 as appeared in Bogduk and Twomey,52
with permission.)
At one time or another, each of the structures listed above has been incrimi­
nated as source of low back pain. However, it has taken many years to collect
the experimental evidence that proves these assertions. Thus, in 1938 and 1939,
Kellgren20.21 demonstrated that low back pain could be induced by noxious
stimulation of the lumbar back muscles and interspinous ligaments, while, recip­
rocally, Steindler and Luck22 showed that certain forms of low back pain syn­
dromes could be relieved, at least temporarily, by anesthetizing these same
structures. Although the term facet syndrome was introduced in 1933,23 it was
not shown until 1976 that experimental stimulation of lumbar zygapophyseal
joints could cause low back pain in normal volunteers,24.25 and that back pain
stemming from these joints could be relieved by radiologically controlled blocks
of the joints themselves26-
33 or their nerve supply. 34
Two lines of evidence revealed that the intervertebral discs could be a
source of back pain. Operating on patients under local anesthesia, Wiberg3H
showed that pressing on the posterior annulus fibrosus could evoke low back
pain, as did Falconer et al.39 Later, after the introduction of discography as a
diagnostic procedure, it was recognized that back pain could be reproduced by
Innervation, Pain Pal/ers, and Mechanisms of Pain Production 97
Fig. 3-3. The nerve plexus accompanying
the posterior longitudinal ligament at the
levels of the L3 and lower vertebrae, as
seen in whole mounts of human fetuses.
The large fibers (arroll's) represent the si­
nuvertebral nerves. (Adapted from Groen
et al,4 as appeared in Bogduk and Two­
mey,52 with permission.)
injections of contrast medium into lumbar intervertebral discs. 40 Subsequent
experience with provocation discography has confirmed that the injection of
contrast medium or even normal saline into intervertebral discs can evoke back
pain, even if the disc is structurally intact and appears normal on myelog­
The dura mater has been shown to be capable of causing back pain in two
types of clinical experiments. First, it was shown that back pain could be evoked
by traction on the dural sleeves of lumbar nerve roots by pulling on sutures
threaded through the dura at operation for laminectomy.46 More recently it has
been shown that chemical irritation of the dura in the form of injections of
hypertonic saline can evoke back pain. 47 Pain stemming from the dura mater
of nerve root sleeves has been relieved by microsurgical transection of the
nerves supplying them.48.49
There is, therefore, a wealth of clinical experimental data confirming that
the ligaments, muscles, joints, discs, and dura mater of the lumbar spine are
all capable of being a source of back pain. Of the innervated structures of the
lumbar spine, only the epidural blood vessels and the vertebral bodies have
not been subjected to experimental study to determine whether they too can
be a source of pain. Circumstantial evidence is conducive to the notion that
98 Physical Therapy of the Low Back
Fig. 3-4. Innervation of the lumbar spine. A cross-sectional view incorporating the
level of the vertebral body ( VB) and its periosteum (p) on the right and the intervertebral
disc (/VD) on the left. PM, psoas major; QL, quadratus lumborum; IL, iliocostalis
lumborum; LT, longissimus thoracis; M, multifidus; alllf , anterior layer of thoracolum­
bar fascia; pill posterior layer of thoracolumbar fascia; esa, erector spinae aponeurosis;
ds, dural sac; zj, zygapophyseal joint; pll, posterior longitudinal ligament; all, anterior
longitudinal ligament; VI', ventral ramus; dr, dorsal ramus; m, medial branch; i, interme­
diate branch; I, lateral branch; svn, sinuvertebral nerve; grc, gray ramus communicans;
sl, sympathetic trunk. (From Bogduk and Twomey,52 with permission.)
distension of epidural veins can cause pain50; however, although it is presumed
that the pain of spinal osteoporosis arises from vertebral bodies, there have
been no formal experimental studies of low back pain stemming from bone.
Given that various structures in the lumbar spine have been shown to be
capable of producing low back pain, it is important to realize that in each case,
the mechanism involved is the stimulation of nerve endings in the affected
structure. Lumbar nerve-root compression is in no way involved. To relate this
to pathology, the possible causes of low back pain would be any pathologic
Innervation, Pain Patters, and Mechanisms of Pain Production 99
process that stimulates the nociceptive nerve endings in one or other of the
pain-sensitive structures of the lumbar spine. In this respect, there are only
two known mechanisms by which nerve endings can be stimulated: chemical
or mechanical irritation.
Chemical irritation occurs in inflammatory diseases or follows tissue dam­
age. Although it is very difficult to validate experimentally, the mechanism
seems to involve the direct stimulation of nerve endings by chemicals. These
may include hydrogen and potassium ions or enzymes that are liberated from
inflammatory cells or damaged tissue cells. Mechanical irritation, on the other
hand, involves the stretching of connective tissue without the involvement of
any chemical mediators. Exactly how mechanical irritation causes pain remains
unclear, but a plausible explanation is that when an array of collagen fibers in
a ligament, joint capsule, or periosteum is placed under tension, it deforms and
closes the available space between individual collagen fibers. Nerve endings
or perhaps nerve fibers within the array would then be stimulated by being
squeezed between the encroaching collagen fibers.
It is beyond the scope of this chapter to expand on the pathology of the
lumbar spine, but other publications have explored the pathomechanics of pain
arising from discs, 5.6.5
dural pain, 52 and pain mediated by the lumbar dorsal
rami.53.54 However, it is worth elaborating on the issue of primary disc pain.
Primary Disc Pain
In contradistinction to pain caused by the compression of spinal nerves by
herniated intervertebral discs, primary disc pain is pain that stems directly from
the disc itself. It is caused by the stimulation of the nerve endings within the
annulus fibrosus. Pathologic processes theoretically responsible for this stimu­
lation include excessive mechanical strain of the annulus, chemical irritation
as a result of infammation following trauma to the annulus, and involvement
of the annulus in the chemical degrading processes that occur in internal disc
Logical deduction reveals that not every pathologic process affecting a
disc will necessarily be painful, as only the peripheral annulus is innervated.
Disc pain will occur only if a pain-producing process affects the innervated
periphery of the disc. Processes such as disc degradation or degeneration that
are restricted to the nucleus and central portions of the annulus do not have
access to the nerve supply of the disc, and therefore cannot directly cause pain.
Thus even severely degenerated discs may not be painful. However, should
centrally located processes extend to the innervated periphery, as in a radial
fissure5.6.51 or if the peripheral annulus itself is primarily damaged, as in a
torsional strain, 5.6.51 then nerve endings may be affected, and if enough are
stimulated, then pain may ensue. On the other hand, an otherwise healthy and
innervated portion of an annulus could become painful if it is called on to bear
a greater mechanical load as a result of disease in other portions of the disc,
and is thereby secondarily subjected to excessive strain. 5.6.51 Variations and
100 Physical Therapy of the LolV Back
permutations like these explain why discs apparently affected by similar disease
processes may be inconsistently painful or painless.
Referred Pain
Referred pain is pain perceived in a region topographically displaced from
the area that is the source of the pain. In lumbar pain syndromes, referred pain
is generated by lesions in the lumbar spine but is perceived in the buttocks and
lower limbs or sometimes in the groin or abdominal wall.
There are two common types of referred pain from the lumbar spine. The
mechanisms and their diagnostic and therapeutic implications are so different
in each that the two must be recognized and distinguished. The first is somatic­
referred pain and the other is radicular pain.
Somatic-Referred Pain
The term somatic-referred pain is used to emphasize the skeletal or "so­
matic" origin for this form of pain, and to distinguish it from radicular pain
and referred pain caused by visceral or vascular disease.
Virtually any source of local lumbar or lumbosacral pain is also capable
of producing somatic-referred pain. The mechanism appears to be that afferent
impulses from the lumbar spine activate neurons in the central nervous system,
which also happen to receive afferents from the lower limbs, buttocks, or groin.
Stimulation of such central nervous system neurons by impulses from the lum­
bar spine results in the perception of pain arising from all the tissues subtended
by these neurons. Thus the patient complains of pain in the lower limbs as well
as the back even though there is no signal actually emanating from the limbs.
The evidence for such a mechanism stems from several experimental and
clinical studies. Kellgren20.21 showed that low ba
k pain experimentally induced
by stimulating interspinous ligaments and back muscles could be accompanied
by referred pain in the lower limbs. These observations were later corroborated
by other investigators. 55.56
Experimental noxious stimulation of lumbar zygapophyseal joints can
cause referred pain in various regions of the lower limbs, buttocks, and
24.25 Traction on the dura mater has been shown to produce buttock and
thigh pain, 46 and in some patients, disc stimulation can reproduce not only
their back pain but their referred pain as well.
7.42-44 Complementing these
experimental studies are the reports that anesthetizing intervertebral discs, 44
and anesthetizing24- or denervating3
.37.53 zygapophyseal joints in cer­
tain patients relieves not only their local pain but also their lower limb pain.
The critical feature of these various studies is that the stimuli used to evoke
referred pain or the anesthetics used to relieve it were delivered directly to
somatic elements of the lumbar spine. Nerve roots were not stimulated or anes-
Innervation, Pain Patterns, and Mechanisms of Pain Production 101
thetized. The mechanisms for the referred pain, therefore, must lie beyond the
nerve roots, and the only possible site is in the central nervous system.
An overemphasized aspect of somatic-referred pain is its apparent segmen­
tal distribution. Early investigators sought to establish charts of the segmental
pattern of pain referral in the anticipation that the axial origin of referred pain
could be diagnosed on the basis of its peripheral distribution, just as dermatomes
are used to diagnose the segmental level of a root compression or spinal cord
injury.21.55.56 However, it is now evident that the fields of referred pain from
individual segments overlap greatly within a given individual and the patterns
exhibited by different individuals vary significantly. 25.53.54 These irregularities
preclude the use of charts of so-called sclerotomes for any legitimate diagnostic
purpose. Such charts serve only to illustrate that lumbar pain may be referred
into the lower limbs, but do not pinpoint constant locations.
In this context, it is sometimes maintained that somatic-referred pain does
not extend beyond the knee and that pain distal to the knee must be radicular
in origin. However, while it is true that somatic-referred pain most commonly
is distributed in the region of the buttock,
24.39.56 it nevertheless can extend as
far as the foot. 2
1 .24.55 Indeed, there is even some evidence that the distance of
referral into the lower limb is proportional to the intensity of the stimulus to
the spine.24
An important (although overlooked) legacy of the experimental studies on
somatic-referred pain relates to its quality. All the studies showed that the
referred pain was deep and aching in quality, and was hard to localize. This
contrasts with the sharper, lancinating nature of radicular pain and putatively
may be used to distinguish somatic-referred pain from radicular pain.
Radicular Pain
The concept of sciatica stems from the coincidental similarity between the
distribution of some forms of referred pain and the course of the sciatic nerve.
Consequently, sciatica was originally ascribed to intrinsic disease of the sciatic
nerve, then later to muscular compression, and eventually to compression of
the lumbosacral nerve roots by disorders of the vertebral column (hence the
term radiclliar pain).
These notions on the causation of sciatica, however, were based only on
inference or circumstantial evidence. Thus, because arthritic changes could be
demonstrated radiologically in patients with sciatic pain, the cause was deemed
to be compression of the L5 spinal nerve by lumbosacral "arthritis. "57.58 Later,
this notion was superceded by the revelation that herniated intervertebral discs
could compress lumbosacral nerve roots. 59 The compressive causes of sciatica,
however, were introduced without it being demonstrated that root compression
could, in fact, cause pain.
Early investigators were probably drawn to their conclusions by the obser­
vations that most of their patients had weakness or numbness in association
with their sciatic pain. Because weakness and numbness are features of nerve
102 Physical Therapy of the Low Back
compression, it was understandably attractive to ascribe the pain to the same
cause and mechanism. Moreover, these conclusions were made and the nerve
root compression theory established before the earliest experiments on somatic­
referred pain. 20.2
It is surprising that nerve-root compression was sustained as the mecha­
nism for referred pain, because it is known that compression of nerves else­
where in the body does not cause pain. 60 Indeed, this paradox led to criticisms
of the nerve-root compression theory. 6
However, subsequent clinical and labo­
ratory experiments have helped resolve this paradox, albeit at the expense of
raising new questions.
MacNab62 reported that experimental compression of normal nerve roots,
using catheters inserted into intervertebral foramina, evoked paresthesia and
numbness, but did not cause pain. On the other hand, Smyth and Wright46
demonstrated that pulling on nerve roots previously affected by disc herniation
did evoke sciatic pain. Thus clinically damaged, but not normal, nerve roots
are capable of generating pain.
These clinical observations have been corroborated by animal experi­
ments, which showed that activity in nociceptive afferent fibers could be elicited
by mechanical stimulation of previously damaged nerve roots, but not by stimu­
lation of normal roots. 63 The questions raised by these experiments are (I) how
do normal and damaged roots differ and (2) how soon after a compressive
lesion is a normal root sufficiently damaged to become painful? These questions
remain unanswered.
Another observation from these same animal experiments63 is t hat nocicep­
tive activity could be elicited by stimulation of dorsal root ganglia irrespective
of whether they were normal or damaged. Thus dorsal root ganglia are appar­
ently more susceptible to mechanical stimulation than axons, and this difference
may explain why root compression is capable of producing pain when compres­
sion of nerve trunks is not. Compression of the cell bodies in the dorsal root
ganglia seems to be the key difference.
Other issues aside, there is no doubt that under the appropriate circum­
stances, compression of nerve roots can cause pain, but an unfortunate legacy
of the concept of sciatica is the tendency in some circles to interpret all forms
of pain in the lower limb as due to nerve-root compression. This is not justified.
The experiments of Smyth and Wright46 showed that traction on nerve
roots produced only a particular form of pain. It was lancinating or shooting
in quality, and was felt along a relatively narrow band "no more than one-and­
a-half inches wide. "46 This neuralgic type of pain is the only type that has been
shown to be produced by root compression. Therefore, only this form of pain
can legitimately be called sciatica and ascribed to root compression. In contrast,
somatic-referred pain is static, aching in quality, hard to localize, and should
be recognized as a different entity.
There are two further irregularities concerning the concepts of sciatica
and nerve-root compression. First, there is no known mechanism whereby a
compressive lesion can selectively affect only nociceptive axons (i. e. , without
also affecting large-diameter afferent fibers that convey touch and other sensa-
Innervation, Pain Patterns, and Mechanisms of Pain Production 103
tions). Therefore, there is no mechanism whereby root compression can cause
pain without causing other neurologic abnormalities as well. Thus, for root
compression to be deemed the cause, radicular pain must be accompanied by
other features of nerve compression: numbness, weakness, or paresthesia. In
the absence of such accompanying features, it is very dificult to maintain that
root compression is the cause of any pain. Pain in the lower limb in the absence
of objective neurologic signs is most likely to be somatic-referred pain.
The second irregularity relates to back pain. All the experimental studies
on radicular pain emphasize that root compression causes pain in the lower
limb. Thus, although radicular pain may feel as if it starts in the back and
radiates into the lower limb, there is no evidence that root compression can or
should cause isolated low back pain. It is implausible that a compressive lesion
could stimulate only those afferents in a root that come from the lumbar spine
but spare those from the lower limb. Isolated low back pain suggests a somatic
lesion, which should be sought, and the pain should not be dismissed as due
to nerve-root compression when no evidence of compression can be found.
One reservation, however, must be raised: apparently local back pain may in
fact be referred pain from pelvic or abdominal visceral or vascular disease.
For this reason, abdominal and pelvic examinations are essential parts of the
assessment of any patient presenting with low back pain.
Combined States
Whereas it is evident that back pain and referred pain may be caused by
a variety of disorders and mechanisms, it is critical to realize that a patient's
complaints may not be due to a single disorder or a single mechanism. Several
disorders may coexist and different mechanisms may be coactive. The simplest
examples are the coexistence of zygapophyseal disorders and disc disorders at
the same or different segmental levels, with each disorder contributing sepa­
rately to the patient's overall complaint.
A more complex example relates to nerve-root compression syndromes.
The cardinal features of nerve-root compression are the objective neurologic
signs of weakness or numbness. In the presence of such signs accompanied by
the lancinating pain that is characteristic of radicular pain, the syndrome may
legitimately be ascribed to nerve-root compression. However, nerve-root
compression may only be part of a patient's complaint. Local somatic and
somatic-referred pain may occur in addition to the symptoms of nerve-root
compression. In such cases, the most likely source or sources of the somatic
pain are the structures immediately adjacent to the compressed root.
The closest relation of a nerve root is its dural sleeve, and it is obvious
that any lesion that might compress a root must first affect its dural sleeve.
Given that the dura is pain-sensitive,46,47 it becomes a potent possible source
of low back pain and even referred pain,46 which can occur alone or be superim­
posed on any radicular pain. However, the mechanism involved is distinctly
104 Physical Therapy of the Low Back
different from that of any radicular pain, because dural pain is caused by the
stimulation of nerve endings in the dural sleeve, not by nerve compression.
Since the dura is mechanosensitive,46 traction of the dura over a space­
occupying lesion, like a heriated disc, could be the possible cause of dural
pain. The dura is also chemosensitive,47 and therefore an additional or alterna­
tive process could be chemical irritation of the dura. With regard to the latter,
it has been demonstrated that disc material contains potent inflammatory chemi­
66 and when disc material ruptures into the epidural space it seems to
elicit an autoimmune inflammatory reaction that can affect not only the roots
but the dura as we1 1.6
The other two possible sources of pain concurrent with root compression
are the adjacent disc and zygapophyseal joint. Regardless of any herniation
that compresses a root, a disc itself may be an intrinsic source of pain, the pain
being mechanical in origin caused by strain of the annulus fibrosus of the dis­
eased disc. In such cases, treating the nerve-root compression may relieve the
objective neurologic signs and any radicular pain, but the discogenic pain may
continue unless it is treated as well.
A zygapophyseal joint may compress or traumatize the underlying roots
by developing osteophytes,7
but a degenerative zygapophyseal joint may also
be independently painful, causing both local and somatic-referred pain. Thus,
while resecting the osteophytes may decompress the roots, it may not relieve
the intrinsic low back pain and referred pain stemming from the diseased joint.
This concept has particular ramifications in the interpretation of spinal stenosis
where not all the symptoms are necessarily due to the overt nerve-root
It might be expected that different causes of lumbar pain should be distin­
guishable from one another on the basis of differences in the distribution and
behavior of symptoms. Frustratingly, however, this is not so. Because different
structures in the lumbar spine share a similar segmental nerve supply, and
because different disorders share similar mechanisms, no single disorder has a
characteristic distribution of local or referred pain.
Even the classic syndrome of herniated nucleus pulposus is fraught with
diagnostic pitfalls. The sensitivity of straight leg raising as a test for this condi­
tion has been measured to be between 72 and 97 percent,72 meaning that be­
tween 3 and 28 percent of cases remain undetected. Other, classical features
such as weakness of extensor hallucis, weakness of ankle dorsiflexion, and
absent ankle reflex have sensitivities for less than this.72 The specificity of
straight leg raise, however, is only II to 45 percent,72 meaning that too often
conditions other than herniated nucleus pulposus are the cause of the abnor­
With respect to zygapophyseal joint disorders, experimental studies have
shown that local and referred pain patterns from joints at different levels vary
Innervation, Pain Patters, and Mechanisms of Pain Production 105
considerably in different individuals and that even in a given individual they
overlap greatly. 24
.25 Furthermore, the incidence of other clinical features in
zygapophyseal syndromes, including various aggravating factors, is insvffi­
ciently different from their incidence in other syndromes. Fairbank et aJ29 per­
formed diagnostic joint blocks on patients presenting with back pain and re­
ferred pain, and analyzed the differences between those who responded and
those who did not. Although certain features did occur more commonly in
responders, they also occurred so frequently in nonresponders that no clinical
feature could be identified that could be held to be indicative or pathognomonic
of zygapophyseal joint pain.
Other forms of pain, such as disc pain and muscular-pain syndromes, have
not been studied in this same rigorous way. Consequently, there is no scientific
evidence that permits any claim that certain pain patterns are characteristic of
these syndromes. Only the diagnosis of lumbar-disc herniation has withstood
scrutiny, while the diagnosis of zygapophyseal syndromes on the basis of con­
ventional clinical signs has been shown to be impossible. Therefore, the diagno­
sis of lumbar-pain syndromes, other than those caused by disc herniation, relies
on investigations outside the realm of symptomatic and conventional physical
In this regard, plain radiography has little value as a diagnostic tool in low
back pain,44 while electromyography, myelography, and computed tomography
are of relevance only in nerve-root compression syndromes. For conditions in
which pain alone is the complaint, and there are no objective neurologic signs
indicative of nerve-root compression, other investigations are required.
The mainstay for the diagnosis of lumbar pain in the absence of neurologic
signs are diagnostic blocks and provocation radiology. These techniques are
based on the principles that if a structure is the cause of pain, then stressing
that structure should reproduce the pain and anesthetizing the structure should
relieve it. Thus zygapophyseal joints suspected of being the source of pain can
be infiltrated with local anesthetic,33 and relief of pain implicates the inected
joint as the source. Similarly, intervertebral discs can be injected with saline
or contrast medium to reproduce pain or with local anesthetic to relieve pain. 44
Radicular pain and dural pain can be diagnosed by infiltrating the root thought
to be responsible with local anesthetic. 73
75 In all of these procedures, failure
to provoke or relieve the pain excludes the investigated structure as the source
of pain, whereupon other structures or other segmental levels in the lumbar
spine may be investigated until the responsible site is identified. Although sub­
ject to certain technical limitations,44 these techniques are the only available
means of objectively confirming particular causes of pain suspected on the basis
of clinical examination.
In conflict with this conclusion are the claims of manipulative therapists
who maintain that they are able to diagnose the source of pain in the vertebral
column by manual examination. Elaborate methods of assessment for this pur­
pose have been described in manipulative therapy texts. 76
78 However, no re­
search has been undertaken to validate the purported accuracy of these meth­
ods. Thus, although perhaps attractive in principle, manual examination as a
106 Physical Therapy of the Low Back
diagnostic effort is still open to skepticism from those unconvinced of a manipu­
lative therapist's ability to detect changes in discrete components of the spine,
and that these changes are at all diagnostic of any cause of pain.
I. Yoshizawa H, O'Brien JP, Smith WT, Trumper M: The neuropathology of interver­
tebral discs removed for low-back pain. J Pathol 132:95, 1980
2. Bogduk N, Tynan W, Wilson AS: The nerve supply to the human lumbar interverte­
bral discs. J Anat 132:39, 1981
3. Bogduk N, Wilson AS, Tynan W: The human lumbar dorsal rami. J Anat 134:383,
4. Groen G, Baljet B, Drukker J: The nerves and nerve plexuses of the human vertebral
column. Am J Anat 188:282, 1990
5. Bogduk N: The innervation of the intervertebral discs. p. 135.1n Ghosh P (ed): The
Biology of the Intervertebral Disc. CRC Press, Boca Raton, FL, 1988
6. Bogduk N: The innervation of the discs. In Boyling JD, Palastanga N (eds): Grieve's
Modern Manual Therapy of the Vertebral Column, 2nd Ed. Churchill Livingstone,
Edingurgh (in press)
7. Johnston HM: The cutaneous branches of the posterior primary divisions of the
spinal nerves and their distribution in the skin. J Anat 43:80, 1908
8. Bogduk N: The innervation of the lumbar spine. Spine 8:285, 1983
9. Paris SV: Anatomy was related to function and pain. Orthop Clin North Am 14:
475, 1982
10. Pedersen HE, Blunck CFJ, Gardner E: The anatomy of lumbosacral posterior rami
and meningeal branches of spinal nerves (sinu-vertebral nerves). J Bone Joint Surg
38A:377, 1956
II. Lazorthes G, Gaubert J: L'innervation des articulations interapophysaires verte­
brales. CR Assoc Anat 43e Reunion:488, 1956
12. Lazorthes G, Juskiewenski S: Etude comparative des branches posterieures des
nerfs dorsaux et lombaires et de leurs rapports avec les articulations interapophy­
saires vertebrales. Bull Assoc Anat 4ge Reunion: 1025, 1964
13. Edgar MA, Nundy S: Innervation of the spinal dura mater. J Neurol Neurosurg
Psychiatry 29:530, 1966
14. Lazorthes G, Pouhles J, Espagne J: Etude sur les nerfs sinu-vertebraux lombaires.
Le nerf de Roofe existe-t-il? CR Assoc Anat 34e Reunion:317. 1947
15. Groen G, Balet B, Drukker J: The innervation of the spinal dura mater: anatomy
and clinical implications. Acta Neurochir 92:39, 1988
16. Jung A, Brunschwig A: Recherches histologiques des articulations des corps verte­
braux. Presse Med 40:316, 1932
17. Hirsch C, Ingelmark B, Miller M: The anatomical basis for low back pain. Acta
Orthop Scand 33: I, 1963
18. Jackson HC, Winkelmann RK, Bickel WH: Nerve endings in the human lumbar
spinal column and related structures. J Bone Joint Surg 48A: 1272, 1966
19. Malinsky J: The ontogenetic development of nerve terminations in the intervertebral
discs of man. Acta Anat 38:96, 1959
20. Kellgren JH: Observations on referred pain arising from muscle. Clin Sci 3: 175.
Innervation, Pain Patterns, and Mechanisms of Pain Production 107
21. Kellgren JH: On the distribution of pain arising from deep somatic structures with
charts of segmental pain areas. Clin Sci 4:35, 1939
22. Steindler A, Luck JV: Differential diagnosis of pain low in the back: allocation of
the source of pain by procain hydrochloride method. JAMA 110: 106, 1938
23. Ghormley RK: Low back pain with special reference to the articular facets with
presentation of an operative procedure. J AMA 10: 1773, 1933
24. Mooney V, Robertson J: The facet syndrome. Clin Orthop 115: 149, 1976
25. McCall IW, Park WM, O'Brien JP: Induced pain referral from posterior lumbar
elements in normal subjects. Spine 4:441, 1979
26. Mehta M, Sluijter ME: The treatment of chronic back pain. Anaesthesia 34:768,
27. Carrera. GF: Lumbar facet joint injection in low back pain and sciatica. Radiology
137:665, 1980
28. Dory MA: Arthrography of the lumbar facet joints. Radiology 140:23, 1981
29. Fairbank JCT, Park WM, McCall IW, O'Brien JP: Apophyseal injection of local
anaesthetic as a diagnostic aid in primary low-back pain syndromes. Spine 6:598,
30. Destouet JM, Gilula LA, Murphy WA, Monsees B: Lumbar facet joint injection:
indication, technique, clinical correlation, and preliminary results. Radiology 145:
32 1, 1982
31. Rashbaum RF: Radiofrequency facet denervation. Orthop Clin North Am 14:569,
32. Lippit AB: The facet joint and its role in spine pain. Spine 9:746, 1984
33. Carrera GF, Williams AL: Current concepts in evaluation of the lumbar facet joints.
CRC Crit Rev Diagn Imaging 21 :85, 1984
34. Pawl RP: Results in the treatment of low back syndrome from sensory neurolysis
of the lumbar facets (facet rhizotomy) by thermal coagulation. Proc Inst Med Chi­
cago 30: 150, 1974
35. Lora J, Long DM: So-called facet denervation in the management of intractable
back pain. Spine 1:121, 1976
36. Ogsbury JS, Simons H, Lehman RAW: Facet "denervation" in the treatment of
low back syndrome. Pain 3:257, 1977
37. Sluijter ME, Mehta M: Treatment of chronic back and neck pain by percutaneous
thermal lesions. p. 141. In Lipton S, Miles J (eds): Persistent Pain: Modern Methods
of Treatment. Vol. 3. Academic Press, London, 1981
38. Wiberg G: Back pain in relation to the nerve supply of the intervertebral disc. Acta
Orthop Scand 19:211, 1947
39. Falconer MA, McGeorge M, Begg AC: Observations on the cause and mechanism
of symptom production in sciatica and low back pain. J Neurol Neurosurg Psychiatry
11:13, 1948
40. Lindblom K: Technique and results in myelography and disc puncture. Acta Radiol
34: 321, 1950
41. Collis JS, Gardner WJ: Lumbar discography-an analysis of 1,000 cases. J Neuro­
surg 19:452, 1962
42. Wiley 11, MacNab 1, Wortzman G: Lumbar discography and its clinical applications.
Can J Surg 11:280, 1968
43. Simmons EH, Segil CM: An evaluation of discography in the localization of symp­
tomatic levels in discogenic disease of the spione. Clin Orthop 108:57, 1975
44. Park WM: The place of radiology in the investigation of low back pain. Clin Rheum
Dis 6:93, 1980
108 Physical Therapy of the Low Back
45. White AH: Injection techniques for the diagnosis and treatment of low back pain.
Orthop Clin North Am 14:553, 1983
46. Smyth MJ, Wright V: Sciatica and the intervertebral disc: an experimental study.
J Bone Joint Surg 40A: 1401, 1958
47. EI Mahdi MA, Latif FY A, Janko M: The spinal nerve root innervation and a new
concept of the clinicopathological interrelations in back pain and sciatica. Neuro­
chirurgia 24: 137, 1981
48. Cuatico W, Parker JC, Pappert E, Pilsl S: An anatomical and clinical investigation
of spinal meningeal nerves. Acta Neurochir 90: 139, 1988
49. Cuatico W, Parker JC: Further observations on spinal meningeal nerves and their
role in pain production. Acta Neurochir 101: 126, 1989
50. Boas RA: Post-surgical low back pain. p. 188. In Peck C, Wallace M (eds): Problems
in Pain. Pergamon, Sydney, 1980
51. Bogduk N: The lumbar disc and low back pain. In Loeser J (ed): Neurosurg Clin
North Am 2:79 1, 199 1
52. Bogduk N, Twomey L T: Nerves of the lumbar spine. p. 107. Clinical Anatomy of
the Lumbar Spine. 2nd Ed. Churchill Livingstone, Melboure, 1991
53. Bogduk N: Lumbar dorsal ramus syndrome. Med J Aust 2:537, 1980
54. Bogduk N: Lumbar dorsal ramus syndromes. In Boyling JD, Palastanga N (eds):
Grieve's Modern Manual Therapy of the Vertebral Column, 2nd Ed. Churchill Liv­
ingstone, Edinburgh (in press)
55. Feinstein B, Langton JNK, Jameson RM, Schiller F: Experiments on pain referred
from deep structures. J Bone Joint Surg 36A:981, 1954
56. Hockaday JM, Whitty CWM: Patterns of referred pain in the normal subject. Brain
90:481, 1967
57. Danforth MS, Wilson PD: The anatomy of the lumbosacral region in relation to
sciatic pain. J Bone Joint Surgery 7: 109, 1925
58. Williams PC: Reduced lumbosacral joint space: its relation to sciatic irritation.
JAMA 99: 1677, 1962
59. Mixter WJ, Barr JS: Rupture of the intervertebral disc with involvement of the
spinal canal. N Engl J Med 211 :210, 1934
60. Fisher CM: Pain states: a neurological commentary. Clin Neurosurg 31 :32, 1984
61. Kelly M: Is pain due to pressure on nerves? Neurology 6:32, 1956
62. MacNab I: The mechanism of spondylogenic pain. p. 89. In Hirsch C, Zotterman
Y (eds): Cervical Pain. Pergamon, Oxford, 1972
63. Howe JF, Loeser JD, Calvin WH: Mechano-sensitivity of dorsal root ganglia and
chronically injured axons: a physiological basis for the radicular pain of nerve root
compression. Pain 3:25, 1977
64. Marshall LL, Trethewie ER, Curtain CC: Chemical radiculitis: a clinical, physiologi­
cal and immunological study. Clin Orthop 129:61, 1977
65. Saal JS, Franson RC, Dobrow R et al: High levels of inflammatory phospholipase
A2 activity in lumbar disc herniation. Spine 15:674, 1990
66. Franson RC, Saal JS, Saal JF: Human disc phospholipase A2 is intlammatory. Spine
17:S 129, 1992
67. Gertzbein SD, Tile M, Gross A, Falk R: Autoimmunity and degenerative disease
of the lumbar spine. Orthop Clin North Am 6:67, 1975
68. Gertzbein SD: Degenerative disk disease of the lumbar spine. Clin Orthop 129:68,
69. Gertzbein SD, Tai JH, Devlin SR: The stimulation of lymphocytes by nucleus pUlpo­
sus in patients with degenerative disc disease of the lumbar spine. C1in Orthop 123:
149, 1977
Innervation, Pain Patters, and Mechanisms of Pain Production 1 09
70. Murphy RW: Nerve roots and spinal nerves in degenerative disk disease. Clin Or­
thop 129:46, 1977
71. Epstein JA, Epstein BS, Lavine LS et al : Lumbar nerve root compression at the
intervertebral foramina caused by arthritis of the posterior facets. J Neurosurg 39:
362, 1973
72. Andersson GBJ: Sensitivity, specificity and predictive value. p. 277. In Frymoyer
JW (ed): The Adult Spine, Principles and Practice. Vol. I. Raven Press, New York,
73. Stanley 0, McLaren MI, Euinton HA, Getty CJM: A prospective study of nerve
root infiltration in the di agnosis of sciatica. A comparison with radiculography,
computed tomography and operative findings. Spine 15:540, 1990
74. Tajima T, Furukawa K, Kuramochi E: Selective lumbosacral radiculography and
block. Spine 5:68, 1980
75. Dooley JF, McBroom RJ, Taguchi T, McNab I: Nerve root infiltration in t he diagno­
sis of radicular pain. Spine 13:79, 1988
76. Maitland GO: Vertebral Manipul ation. 4th Ed. Butterworths, London, 1977
77. Grieve GP: Common Vertebral Joint Problems. Churchill Livingstone, Edinburgh,
78. Bourdillon JF: Spinal Manipul ation. 3rd Ed. Heinemann, London, 1982
4 Anatomy and Function
of the Lumbar Back
Muscles and Their
Nikolai Bogduk
The l umbar spine is surrounded by muscl es t hat , for descri pti ve purposes
and on functional grounds, can be di vided i nto t hree groups. These are ( I) psoas
major and psoas mi nor, whi ch cover the anterol ateral aspects of the l umbar
spine ; (2) lateral intertransverse muscl es and quadratus l umborum, whi ch con­
nect and cover t he front of t he transverse processes ; and (3) the l umbar back
muscl es, which l i e behi nd and cover the posterior elements of the l umbar spine .
The psoas major i s a long muscl e that ari ses from the anterolateral aspect
of the l umbar spi ne and descends over the bri m of the pel vi s to i nsert i nto the
l esser trochanter of the femur. It i s essential l y a muscle of the t high whose
princi pal action is fexion of t he hi p. Some i nterpretati ons had maintained t hat
• This chapter is from Bogduk N, Twomey L T: The lumbar muscles and their fascia. pp. 83-105.
In: Clinical Anatomy of the Lumbar Spine. 2nd Ed. Churchill Livingstone, Melbourne, 1991, with
1 1 2 Physical Therapy of the LolV Back
because it arises from the l umbar spi ne, the psoas major might have an action
on the l umbar spi ne, ei ther to flex the l umbar spi ne when the thigh i s fi xed, as
in the exerci se of "si t-ups, " or at least to "stabi l ize" the l umbar spi ne. I
However, t hese views are not consonant wi th ei ther i ts anatomy or i ts biomech­
ani cs .
The psoas major has di verse, but systemat i c, attachments to the l umbar
spi ne (Fi g. 4- 1 ) . At each segmental l evel , i t i s attached to the medial half or so
of the anterior surface of the transverse process, to the i ntervertebral di sc, and
to the margi ns of the vertebral bodies adjacent to the di sc.
The muscl e fi bers
from the LS-S I i ntervertebral di sc, the LS vertebral body, and the LS trans­
verse process form the deepest and l owest bundl e of fi bers wi thi n the muscl e.
These fi bers are systemat i cal l y overl apped by fi bers from the di sc, vertebral
margi ns, and transverse process at L4, and i n tur these fi bers are overlapped
by t hose from L3, and so on. As a resul t, the muscl e can be seen i n cross
section to be concentrical l y layered, with fi bers from higher l evel s forming t he
outer surface of the muscl e and those from l ower l evel s being buried sequen­
ti al l y, deeper wi thi n i ts substance.
Fig. 4- 1. Psoas major (PM) and quadratus l umborum (QL) . At each segmental l evel
psoas major attaches to the transverse process, the intervertebral disc, and adjacent
vertebral margins, and to the tendinous arch covering the vertebral body. The attach­
ments of quadratus l umborum are to t he i l i ac crest (A), t he i lio-Iumbar l igament (B),
the transverse processes (C), and the 1 2th rib (D). (From Bogduk and Twomey,92 with
permission. )
The Lumbar Back Muscles and Their Fascia 1 1 3
Al l the muscular fi bers converge to a common tendon t hat crosses the
i l iopubic emi nence, but wi thi n t he muscl e al l muscul ar fibers are t he same
lengt h.
As a resul t the fi bers from L I become tendi nous before those from
L2, and so on. Thi s arrangement consti tutes morphologic evi dence t hat the
psoas major cannot act on the l umbar spi ne. It i ndi cates t hat al l fi bers are
designed to contract in uni son to the same extent . Thi s is appropriate for a
muscl e acti ng to flex the t high but i nappropriate to move the l umbar spi ne.
During flexion of t he l umbar spi ne (whether i n the sagittal or coronal plane) ,
each l umbar vertebra undergoes a different l i near excursi on t han t he others;
the upper ones move much further than the l ower ones. If the psoas was respon­
sible for such movements, the l ength of i ts fbers should be proportional t o the
excursion undergone by their vertebra of origin, which i s not the case. How­
ever, more compel l i ng evi dence arises from the bi omechani cs of the psoas
Plotting the l i nes of acti on of each fasci cl e of psoas reveal s t hat they have
a varied and variable rel ati onshi p to the axes of sagi ttal rotati on of the l umbar
vertebrae; some fasci cl es pass behi nd the upper axes but cross the mi ddl e axes
and l ie i n front of the l ower axes; others cross some axes but l i e i n front of the
The net effect i s that the psoas major tends to extend the upper l umbar
spi ne but flex the lower l umbar spine . However, i ts capaci ty to do so i s l i mi ted
by the proxi mi ty of the l i nes of acti on of i ts fasci cl es to t he axes of rotati on.
The fasci cles pass so close to the axes that the moment s t hat they exert are
next to trivial . Even when the moments of �I fasci cl es are summed, the total
i s smal l . I n essence, t he psoas major, acti ng at maxi mum force-capaci t y, can
generate a fl exion moment that i s barel y enough to raise one-quarter of an
average body weight, in t he exerci se of "si t-ups . "
The emphasi s pl aced on psoas major as a muscl e of t he l umbar spi ne i s
not justifed. Morphological l y and mechanical l y i t i s si mpl y a fl exor of t he thigh
that has assumed an adventi ti ous origin on the l umbar spi ne.
The psoas mi nor i s an i nconstant smal l muscl e bel l y t hat ari ses from the
T 1 2-L I i ntervertebral di sc and forms a very l ong narrow tendon that i nserts i nto
the region of the i l iopubic emi nence. Its bi omechani cal signifi cance i s unknown.
The lateral i ntertransverse muscl es consi st of two part s: the ventral lateral
intertransverse and the dorsal lateral i ntertransverse. The ventral i ntertrans­
verse muscles connect the margi ns of consecuti ve transverse processes , whi l e
the dorsal intertransverse muscl es each connect an accessory process to the
transverse process bel ow (Fi g. 4-2) . Both the ventral and dorsal i ntertransverse
muscl es are innervated by the ventral rami of the l umbar spi nal nerves,s and
consequent l y cannot be classified among the back muscl es, whi ch are al l i nner­
vated by the dorsal rami . On the basis of thei r attachments and thei r nerve
supply, the ventral and dorsal i ntertransverse muscles are consi dered to be
1 1 4 Physical Therapy of the Low Back
Fig. 4-2. The short, intersegmental
muscl es. ITL V, ventral intertransverse
laterals; ITLD, dorsal intertransverse
laterals; ITM, medial intertransverse
muscl es; IS, interspinals; AP, acces­
sory process; MP, mamil lary process;
MAL, mami l l o-accessory l igament .
(From Bogduk and Twomey,92 with
permission. )
homologous t o the i ntercostal and l evator costae muscl es of the thoracic
regi on. 5
The function of the lateral ietertransverse muscl es has never been deter­
mi ned experimental l y. One mi ght presume t hat on the basi s of thei r attachments
they act synergi sti cal l y with the quadratus l umborum i n l ateral flexion of the
l umbar spi ne. However, t hei r small size lends doubt to this notion. I nstead, i t
may be that they pl ay a propriocepti ve rol e l i ke the other intersegmental mus­
cl es of the lumbar spi ne ( see below) .
The quadratus l umborum is a wide, more or l ess rectangular muscl e t hat
covers the l ateral t wo-t hi rds or so of the anteri or sUces of the L I to L4
transverse processes and extends lateral l y a few centi meters beyond the t i ps
of the t ransverse processes. I n detail , the muscl e is a complex aggregation
of various obl i que and l ongi tudi nal l y runni ng fi bers t hat connect the l umbar
t ransverse processes, the i l i um, and the 1 2t h rib6 ( see Fi g. 4- 1 ) .
Caudal l y, t he muscl e ari ses from the L5 t ransverse process, the trough
formed by the superior and anterior i l i ol umbar l igament s, and from the i l iac
crest lateral to the poi nt of attachment of the i l i ol umbar l igament . From t hi s
seri es of attachments the most lateral fibers pass di rect l y toward t he lower,
anterior surface of the 1 2th rib. More medial fibers pass obl i quel y upward and
medi al l y to the anterior surfaces of each of the l umbar t ransverse processes
above L5 . These obl i que fibers i ntermi ngle wi th other obl i que fibers that run
The Lumbar Back Muscles and Their Fascia 1 1 5
upward and lateral l y from each of the l umbar transverse processes to the 1 2th
The majori ty of the fibers of the quadrat us l umborum are connected to the
1 2th rib and one of the functi ons of thi s muscle is said to be to fix the 12th rib
duri ng respi rat ion. 7 The remai ni ng fibers of quadratus l umborum connect the
i l ium to the upper four lumbar transverse processes, and these are the onl y
fi bers of quadrat us lumborum that are suitabl y di sposed to execute l ateral flex­
ion of the l umbar spi ne.
The lumbar back muscl es are those muscl es that l i e behi nd the pl ane of
the lumbar transverse processes and that exert an acti on on the lumbar spi ne.
They i ncl ude muscl es that attach to the l umbar vertebrae and thereby act di ­
rectl y on the lumbar spi ne, and certain other muscl es that although not attaching
to the lumbar vertebrae, neverthel ess exert an action on the l umbar spi ne.
For descri pt i ve purposes and on morphologic grounds, the l umbar back
muscl es can be di vided i nto three groups:
I . The short i ntersegmental muscl es-the i nterspi nal s and the medi al i n­
tertransverse muscl es.
2. The pol ysegmental muscl es that attach to t he lumbar vertebrae-the
mul tifi dus and the lumbar components of longi ssi mus and i l i ocostal i s .
3. The long polysegmental muscl es, represented by the thoracic compo­
nents of longi ssimus and i l iocostal i s l umborum, which, i n general , do not attach
to the l umbar vertebrae but cross the l umbar region from thoracic l evel s to fi nd
attachments on the i l i um and sacrum.
The descriptions of the back muscl es offered in thi s chapter, notabl y those
of the mul t ifi dus and erector spi nae, di ffer substanti al l y from those gi ven i n
standard textbooks . Tradi t ional l y, these muscl es have been regarded as stem­
ming from a common origin on the sacrum and i l i um and passi ng upward to
assume di verse attachments to the l umbar and thoracic vertebrae and ribs.
However, i n the face of several studi es of these muscl es,8
J2 i t i s consi dered
more appropriate to view these muscl es in the reverse di recti on: from above
downward. Not onl y i s this more consi stent wi th the pattern of their nerve
suppl y, 12. J
but it cl arifies the i denti ty of certain muscl es and the i denti ty of
the erector spi nae aponeurosi s, and reveal s the segmental biomechanical di spo­
sition of the muscles.
The l umbar i nterspinal s are short paired muscl es that l i e on ei ther si de of
the interspinous l igament and connect the spi nous processes of adjacent l umbar
vertebrae (see Fig. 4-2) . There are four pairs i n the l umbar region. Al though
116 Physical Therapy of the Low Back
di sposed to act synergi sti cal l y with the mul t ifi dus to produce posterior sagittal
rotation of the vertebra above, the i nterspi nal s are quite small and would not
contribute appreci abl y to the force requi red to move a vertebra. Thi s paradox
is si mil ar to that whi ch appl i es for the medi al i ntertransverse muscl es and is
di scussed further in t hat context .
Medial Intertransverse Muscles
The medi al i ntertransverse muscl es can be consi dered to be true back mus­
cl es, because, unl i ke the l ateral i ntert ransverse muscl es, they are i nnervated
by the l umbar dorsal rami .
The medi al i ntertransverse muscl es arise from an
accessory process, the adjoini ng mami l lary process and the mami l lo-accessory
l igament that connects these two processes . 1
They insert into the superior
aspect of the mami l l ary process of the vertebra below ( see Fig. 4-2) .
The medi al i ntertransverse muscl es l i e l ateral t o t he axi s of lateral flexion
and behi nd the axi s of sagittal rotati on. However, they lie very close to these
axes and are very smal l muscl es. Therefore, it is quest ionable whether they
coul d contri bute any appreciabl e force i n ei ther lateral fl exi on or posterior sagit­
tal rotati on. It might be argued that perhaps l arger muscl es might provide the
bul k of the power to move the vertebrae, and t he i ntert ransverse muscl es could
act to "fine t une" the movement. However, t hi s suggestion i s highly specula­
ti ve, i f not fanciful, and does not take i nto account thei r small size and consider­
able mechani cal di sadvantage.
A tantal izi ng al ternati ve suggesti on i s that the i ntert ransverse muscl es act
as l arge, propriocepti ve transducers ; thei r value l i es not i n t he force they can
exert , but in the muscl e spi ndl es t hey contai n. Placed cl ose to the l umbar verte­
bral col umn, the i ntertransverse muscl es could monitor the movements of the
column and provi de feedback t hat infl uences the act ion of the surrounding
muscles . Such a role has been suggested for the cervical i ntertransverse mus­
cl es, whi ch have been found to contai n a high densi ty of muscl e spi ndles. 1
l ndeed, al l uni segmental muscl es of the vertebral col umn have between two
and six ti mes the densi ty of muscl es spi ndl es found i n the longer, pol ysegmental
muscl es, and there i s growi ng specul ati on t hat t hi s underscores the propriocep­
ti ve functi on of all short , smal l muscl es of the body.
Mul t ifi dus is the largest and most medial of the l umbar back muscl es. I t
consi sts of a repeati ng series offasci cl es that st em from t he l ami nae and spinous
processes of the l umbar vertebrae and exhibit a constant patter of attachments
caudal l y.
1 2
The shortest fasci cl es of t he mul tifidus are t he "lami nar fibers," whi ch
ari se from the caudal end of the dorsal surface of each vertebral l ami na and
The LlImbar Back Mllscles and Their Fascia 1 1 7
Fig. 4-3. The component fascicl es of multifidus. (A) The l ami nar fibers of mul tifi dus.
(B-F) The fascicles from the L l to L5 spinous processes, respectivel y. ( From Bogduk
and Twomey,n with permission. )
i nsert into the mamil lary process of the vertebra two l evel s caudad ( Fi g. 4-3A) .
The L5 l aminar fi bers have no mami l lary process i nto whi ch t hey can i nsert ,
and i nsert i nstead i nto an area on the sacrum just above the fi rst dorsal sacral
foramen. Because of their attachments, t he l ami nar fi bers may be considered
homologous to the thoracic rotatores.
The bul k of t he l umbar mul tifidus consi sts of much larger fasci cles that
radiate from t he l umbar spi nous processes. These fasci cl es are arranged i n fi ve
overlapping groups such that each l umbar vertebra gives rise to one of t hese
groups . At each segmental level , a fasci cle arises from the base and caudol ateral
dge of the spinous process , and several fasci cles arise, by way of a common
tendon, from the caudal ti p of t he spi nous process. Thi s tendon i s referred to
1 1 8 Physical Therapy of the Low Back
hereafter as the "common tendon. " Al t hough confuent wi th one anot her at
t hei r origi n, the fasci cl es in each group di verge caudal l y to assume separate
attachments to mami l l ary processes, the i l iac crest, and the sacrum.
The fasci cl e from the base of the LI spi nous process i nserts into the L4
mami l lary process, whereas t hose from the common tendon i nsert i nto the
mami l lary processes of LS, S I, and the posterior superior i l iac spi ne ( Fig.
4-3B) .
The fasci cl e from the base of the spi nous process of L2 i nserts i nto the
mami l l ary process of LS, whereas t hose from the common tendon i nsert into
the S 1 mami l lary process, the posterior superior i l iac spine, and an area on the
i l iac crest j ust caudoventral to the posteri or superior i l iac spi ne ( Fi g. 4-3C) .
The fasci cl e from t he base of the L3 spi nous process i nserts i nto the mami l ­
lary process of the sacrum, whereas those fasci cl es from the common tendon
i nsert into a narrow area extendi ng caudal l y from the caudal extent of the
posterior superior i l iac spine to the l ateral edge of the thi rd sacral segment ( Fi g.
4-3D) . The L4 fasci cl es i nsert onto the sacrum i n an area medi al to the L3 area
of i nserti on, but lateral to the dorsal sacral forami na ( Fi g. 4-3E) , whereas those
from the LS vertebra i nsert onto an area medial to the dorsal sacral forami na
(Fi g. 4-3F) .
I t i s noteworthy t hat whi l e many of the fasci cl es of mul t ifidus attach to
mami l l ary processes, some of the deeper fibers of t hese fasci cl es attach to t he
capsul es of the zygapophyseal joi nts next to the mami l lary processesY Thi s
attachment al l ows the mul ti fi dus t o protect the joi nt capsul e from bei ng caught
i nsi de the joi nt duri ng the movements executed by the mul ti fi dus.
The key feature of t he morphology of the l umbar mul tifi dus i s that i ts
fasci cl es are arranged segmental l y. Each l umbar vertebrae i s endowed wi th a
group of fasci cl es t hat radiate from i ts spi nous process, anchori ng it below to
mami l l ary processes, the i l iac crest , and the sacrum. Thi s disposition suggests
t hat the fibers of mul ti fi dus are arranged i n such a way t hat their princi pal
acti on i s focused on i ndi vi dual l umbar spi nous processes.
They are designed
to act in concert on a si ngle spi nous process. Thi s contention i s supported by the
pattern of i nnervati on of the muscl e. All the fasci cl es arisi ng from the spinous
processes of a gi ven vertebra are i nnervated by the medial branch of the dorsal
ramus t hat i ssues from below t hat vertebra.
Thus the muscles that di rectl y
act on a parti cul ar vertebral segment are i nnervated by the nerve of t hat
segment .
I n a posterior vi ew, the fasci cl es of mul ti fi dus are seen to have an obl i que,
caudolateral ori entati on. Thei r l i ne of acti on, therefore, can be resolved into
two vectors: a large verti cal vector, and a consi derabl y smal l er horizontal vec­
tor9 ( Fi g. 4-4A) .
The smal l horizontal vector suggests t hat the mul tifdus coul d pul l the spi­
nous processes sideways, and t herefore produce horizontal rotati on. However,
horizontal rotati on of l umbar vertebrae is impeded by the impaction of the
contralateral zygapophyseal joi nts . Hori zontal rotation occurs after impaction
of the joi nts onl y i f an appropriate shear force i s appl i ed to the i ntervertebral
di scs, but the horizontal vector of mul tifidus is so small that it is unl i kel y that
The Lumbar Back Muscles and Their Fascia 1 1 9
Fig. 4-4. The force vectors of multifidus. (A) In an anteroposterior vi ew, the obl ique
l i ne of action of the multifi dus at each l evel (bold arrows) can be resol ved i nto a major
vertical vector ( V) and a smal ler horizontal vector (H). (B) I n a lateral vi ew, the vertical
vectors of the mul t ifidus are seen to be aligned at right angles to the spinous processes.
(From Bogduk and Twomey,n with permission. )
mul tifidus would be capable of exert i ng such a shear force on the di sc by acti ng
on the spinous process. I ndeed, el ectromyographic studi es reveal that mul tifi­
dus is i nconsi stent l y acti ve in derotation and that, paradoxical l y, it is acti ve
in both i psi lateral and contral ateral rotati on.
Rotati on, therefore, cannot be
i nferred to be a primary action of mul ti fi dus. In this context , mul ti fi dus has
been sai d to act onl y as a stabi l izer i n rotat ion,21.
but the aberrant movements
that i t i s supposed to stabi l ize have not been defined (al though see bel ow) .
The pri nci pal action of mul t ifidus i s expressed by i t s vertical vector, and
further i nsight i s gained when this vector i s vi ewed i n a l ateral projecti on ( Fi g.
4-48) . Each fasci cl e of mul t ifidus, at every l evel , act s vi rtual l y at right angles
to i ts spi nous process of origi n. 9 Thus, usi ng the spinous process as a l ever,
every fasci cl e i s ideal l y di sposed to produce posterior sagi t tal rotati on of i ts
vertebra. The right-angle orientati on, however, precl udes any acti on as a poste­
rior horizontal transl ator. Therefore , the mul ti fi dus can onl y exert the "rock­
i ng" component of extensi on of the l umbar spi ne or control thi s component
during fl exion.
Havi ng establ i shed that mul ti fi dus i s pri mari l y a posterior sagittal rotator
of the l umbar spine, it i s possible to resolve the paradox about its acti vi ty duri ng
1 20 Physical Therapy of the Low Back
horizontal rotati on of t he t runk. 9 In the fi rst i nstance, i t shoul d be realized that
rotation of the l umbar spine is an i ndi rect acti on. Active rotation of the l umbar
spine occurs onl y if the t horax is fi rst rotat ed, and is therefore secondary to
t horacic rotation. Secondl y, it must be realized t hat a muscle wi th two vectors
of action cannot use these vectors i ndependentl y. If the muscle contracts, then
both vectors are exerted. Thus, mul tifidus cannot exert axial rotation wi thout
simultaneousl y exerting a much l arger posterior sagittal rotat ion .
The princi pal muscl es that produce rotation of the t horax are the obl i que
abdomi nal muscl es. The horizontal component of thei r orientation is able to
t urn the t horacic cage in the horizontal pl ane and thereby impart axial rotation
to t he l umbar spi ne. However, the obl i que abdomi nal muscl es also have a
vertical component to thei r ori entati on. Therefore, if they contract to produce
rotati on t hey wi l l also si mul taneousl y cause fexi on of the trunk, and therefore,
of the l umbar spi ne. To counteract t hi s fl exi on, and maintain pure axial rotation,
extensors of the l umbar spi ne must be recrui ted, and t hi s i s how mul tifidus
becomes i nvol ved i n rotati on.
The rol e of mul ti fi dus i n rotati on i s not to produce rotati on, but to oppose
the flexi on effect of the abdomi nal muscl es as they produce rotation. The aber­
rant moti on stabil ized by mul t ifi dus during rotation i s, therefore, the unwanted
fexion unavoidably produced by the abdomi nal muscl es. 9
Apart from i ts acti on on i ndi vidual l umbar vertebrae, the mul t ifi dus, be­
cause of i ts pol ysegmental nature, can al so exert i ndi rect effects on any i nter­
posed vertebrae. Si nce the l i ne of acti on of any long fasci cl e of mul tifidus
l i es behi nd the l ordotic curve of the l umbar spine, such fasci cl es can act l i ke
bowstri ngs on t hose segments of the curve that i ntervene between the attach­
ments of the fasci cl e. The bowstri ng effect would tend to accentuate the l umbar
lordosi s, resul t i ng in compressi on of i ntervertebral di scs posteriorly and strai n
of the di scs and longit udi nal l igament anteriorl y. Thus a secondary effect of
the acti on of mul tifi dus is to i ncrease the lumbar lordosi s and the compressi ve
and tensi l e loads on any vertebrae and i ntervertebral di scs i nterposed between
its attachments .
Lumbar Erector Spinae
The lumbar erector spinae l i es lateral to the mul tifidus and forms the promi­
nent dorsolateral contour of t he back muscl es in the l umbar region. It consi sts
of two muscl es: the l ongi ssimus thoraci s and the i l iocostal i s l umborum. Further­
more, each of these muscl es has two components: a l umbar part, consi st i ng of
fasci cl es ari si ng from l umbar vertebrae, and a thoracic part, consi sti ng of fasci­
cles ari sing from t horacic vertebrae or ri bs. �·
10. I I
These four parts may be re­
felTed to, respecti vel y, as longi ssimus t horaci s pars l umborum, i l iocostal i s
l umborum pars l umborum, l ongissimus t horaci s pars l umborum, and longi ssi­
mus thoraci s pars thoraci s.
I n the l umbar region, the l ongi ssimus and i l iocostal i s are separated from
each other by the l umbar i ntermuscul ar aponeurosi s, an anteroposterior cont in-
The LlImbar Back Mllscles and Their Fascia 1 2 1
Fig. 4-5. The l umbar fibers of longissi­
mus ( longi ssimus thoracis pars l umb­
orum) . On the left. the five fascicl es of
the intact muscle are drawn. The forma­
tion of the lumbar intermuscul ar aponeu­
rosis (L1A) by the l umbar fascicl es of lon­
gissimus is depicted. On the right, the
lines indicate the attachments and span
of the fascicl es. ( From Bogduk and Two­
mey,n with permission. )
o 0 0
uation of the erector spi nae aponeurosi s.
S. I O
I t appears as a flat sheet of col lagen
fi bers that extend rostral l y from the medial aspect of the posterior superior i l iac
spine for 6 to 8 cm. It i s formed mai nl y by the caudal tendons of the rostral
four fasci cl es of the l umbar component of l ongi ssi mus ( Fi g. 4-5) .
Longissimus Thoracis Pars Lumborum
The longi ssi mus thoraci s pars l umborum is composed of five fasci cl es,
each ari si ng from t he accessory process and the adjacent medial end of the
dorsal surface of the transverse process of a l umbar vertebra (see Fi g. 4-5) .
The fasci cl e from the L5 vertebra i s the deepest and shortest. I t s fi bers
i nsert di rectl y i nto the medial aspect of the posterior superi or i l iac spi ne. The
fasci cl e from L4 al so l i es deepl y, but lateral to that from L5. Succeedi ng fasci ­
cl es l i e progressi vel y more dorsal l y so t hat the L3 fasci cl e covers t hose from
L4 and L5, but is i tsel f covered by the L2 fasci cl e, whereas the LI fasci cl e l i es
most superfi ci al l y.
The Lito L4 fasci cl es al l form tendons at thei r caudal ends, whi ch con­
verge to form the l umbar i ntermuscular aponeurosi s , which eventual l y attaches
to a narrow area on the i l i um i mmediatel y l ateral to the i nserti on of the L5
fasci cl e. The l umbar i ntermuscul ar aponeurosi s thus represents a common ten­
don of i nsertion, or the aponeurosi s, of the bul k of the l umbar fi bers of l ongi s­
si mus.
Each fasci cl e of the l umbar longi ssi mus has both a dorsoventral and a
122 Physical Therapy of the Low Back
rostrocaudal orientation. 10 Therefore, t he action of each fasci cl e can be re­
sol ved into a vertical vector and a hori zontal vtctor, the relative si zes of which
differ from Ll to L5 ( Fi g. 4-6A) . Consequent l y, the rel ative actions of longi ssi ­
mus di ffer at each segmental l evel. Furthermore, the acti on of longi ssi mus , as
a whol e, wi l l differ according to whether the muscl e contracts uni lateral l y or
bi lateral l y.
The l arge vertical vector of each fasci cl e l i es lateral t o the axi s of lateral
fexion and behind the axi s of sagittal rotation of each vertebra. Thus contract­
ing uni l ateral l y t he longi ssi mus can lateral l y fl ex the vertebral col umn, but act­
ing bi l ateral l y the various fasci cl es can act , l i ke mul tifidus, to produce posterior
sagittal rotation of thei r vertebra of origi n. However, thei r attachments to the
accessory and transverse processes lie close to the axes of sagittal rotat ion,
and therefore thei r capaci ty to produce posterior sagittal rotation i s less effi ci ent
than t hat of mul tifidus, whi ch acts through the long l evers of the spinous pro­
cesses. IO
The horizontal vectors of the longi ssi mus are di rected backward. There­
fore, when contracting bi l ateral l y the longi ssi mus i s capable of drawing the
( 0 0
, A
Fig. 4-6. The force vectors of the longissi mus thoracis pars l umborum. (A) I n a lateral
vi ew, the obl i que l i ne of acti on of each fasci cl e of longi ssi mus can be resolved i nto a
vertical ( V) and a horizontal (H) vector. The horizontal vectors of l umbar fasci cl es are
larger. (B) In an anteroposterior vi ew, the l i ne of action of the fasci cl es can be resolved
i nto a major vertical vector and a much smaller horizontal vector. (From Bogduk and
Twomey,92 wi th permi ssi on. )
The Lumbar Back Muscles and Their Fascia 123
l umbar vertebrae backward. This action of posterior translation can restore the
anterior t ranslat ion of the l umbar vertebrae t hat occurs duri ng flexi on of the
l umbar col umn. The capaci ty for posterior translation i s greatest at l ower l um­
bar l evel s where the fasci cl es of longissimus assume a greater dorsoventral
orientation ( Fig. 4-6B) .
Revi ewi ng the horizontal and verti cal acti ons of longissi mus together, i t
can be seen that longissimus expresses a conti nuum of combi ned acti ons along
the l ength of the l umbar vertebral col umn. From below upward, i ts capaci ty
as a posterior sagittal rotator i ncreases, whi l e reci procal l y, from above down­
ward, the fasci cl es are better designed to resist or restore anteri or translati on.
I t is emphasized that the longissimus cannot exert i t s horizontal and verti cal
vectors i ndependent l y. Thus whatever horizontal translation i t exerts must
occur simul taneousl y with posterior sagittal rotati on. The resol uti on i nto vec­
tors simply reveals the rel ati ve amounts of si mul taneous translation and sagittal
rotat ion exerted at different segmental l evel s.
It might be deduced that because of the horizontal vector of longissi mus,
this muscl e, act i ng uni lateral l y, coul d draw the accessory and transverse pro­
cesses backward and therefore produce axi al rotati on. However, in thi s regard,
the fasci cl es of longissimus are oriented al most di rectl y toward the axis of
axial rotation and so are at a marked mechani cal disadvantage to produce axi al
Iliocostalis Lumborum Pars Lumborum
The l umbar component of i l iocostalis l umborum consists of four overl yi ng
fascicles arising from the L I to L4 vertebrae.
Rostral l y, each fasci cl e attaches
to the tip of the t ransverse process and to an area extendi ng 2 to 3 cm l ateral l y
onto the mi ddl e layer of the thoracol umbar fascia ( Fi g. 4-7) .
The fasci cl e from L4 i s the deepest, and caudal l y i t i s attached di rectl y t o
the i l iac crest just lateral to t he posterior superior i l iac spi ne. Thi s fasci cl e i s
covered by t he fasci cl e from L3 t hat has a si mi l ar but more dorsolateral l y
located attachment on the i l iac crest . I n sequence, L2 covers L3 and L I covers
L2 with insertions on the i l i ac crest becoming successi vel y more dorsal and
lateral . The most lateral fasci cl es attach to the i l iac crest just medial to the
attachment of the "l ateral raphe" of the thoracol umbar fasci a ( see bel ow) .
The most medial fi bers of i l iocostalis contri bute t o t he l umbar i ntermuscul ar
aponeurosis, but onl y to a minor extent .
Alt hough an L5 fasci cl e of i l iocostal i s l umborum is not described in t he
l i terature, i t i s represented i n the i l i ol umbar "l igament . " I n neonates and chi l ­
dren t hi s "l igament" is completel y muscul ar i n structure. By the t hi rd decade
of l i fe the muscl e fi bers are enti rel y replaced by col lagen, gi vi ng ri se to the
fami l iar i l iol umbar l igament. 23 On the basis of sites of attachment and rel ati ve
orientat ion, the posterior band of the i l iol umbar l igament woul d appear to be
deri ved from the L5 fasci cl e of i l iocostal i s, whereas the anterior band of the
ligament is a deri vati ve of the quadratus l umborum.
1 24 Physical Therapy of the Low Back
C 1 )
Fig. 4-7. The l umbar fibers of i l iocostalis (il iocostal i s lumborum pars l umborum) . On
the left , the four l umbar fasci cl es of i l i ocostal i s are shown. On the right, their span and
attachments are i ndicated by t he l i nes. ( From Bogduk and Twomey,92 with permission. )
The di sposi ti on of t he l umbar fasci cl es of i l iocostalis i s simi l ar t o that of
the l umbar longissi mus, except t hat the fasci cl es are si tuated more lateral l y.
Like that of t he l umbar longissi mus, thei r action can be resol ved into horizontal
and vert ical vectors ( Fig. 4-8A) .
The verti cal vector i s st i l l predomi nant , and therefore the l umbar fasci cl es
of i l iocostal i s contract i ng bilateral l y can act as posterior sagittal rotators (Fi g. 4-
8B) , but because of t he horizontal vector, a posterior transl ation will be exerted
simul taneousl y, principal l y at lower l umbar l evel s where the fascicl es of iliocos­
tal i s have a greater forward ori entati on. Contracti ng unilateral l y, the l umbar
fasci cl es of i l iocostal i s can act as lateral fl exors of the l umbar vertebrae, for
whi ch action t he t ransverse processes provi de very substantial l evers .
Contracting uni l ateral l y, the fi bers of i liocostalis are better suited to exert
axial rotati on t han the fasci cl es of l umbar longi ssimus, because thei r attachment
to the tips of the transverse processes displ aces them from t he axis of horizontal
rotation and provi des them with substantial l evers for this acti on. Because of
t his l everage, the lower fascicl es of iliocostalis are the only intrinsi c muscl es
of the l umbar spi ne reasonabl y di sposed to produce horizontal rotat ion . Their
The Lumbar Back Muscles and Their Fascia 1 25
G 0 l
Fig. 4-8. TI-e force vectors of the i l iocostal i s l umborum pars l umborum. (A) I n a lateral
vi ew, the l i ne of action of the fascicles can be resol ved into vertical ( V) and horizontal
(H) vectors. The horizontal vectors are larger at l ower l umbar l evel s. (B) I n an anteropos­
terior vi ew, the l i ne of action is resolved i nto a vertical vector and a very smal l horizontal
vector. ( From Bogduk and Twomey,92 wi th permi ssion. )
effecti veness as rotators, however, i s dwarfed by t he obl i que abdomi nal mus­
cles that act on the ribs and produce l umbar rotati on i ndirect l y by rotati ng the
thoracic cage. However, because i liocostalis cannot exert axi al rotation wi thout
simul taneously exerti ng posterior sagittal rotation, the muscl e i s wel l sui ted t o
cooperate with mul t ifidus to oppose the flexi on effect of the abdomi nal muscl es
when they act to rotate the t runk.
Longissimus Thoracis Pars Thoracis
The thoracic fibers of longissimus thoraci s t ypi cal l y consist of I I or 12
pairs of small fasci cl es arising from the ribs and t ransverse processes of Tl or
T2 down to T 1 2 ( Fi g. 4- 9) . At each l evel , two t endons can usual l y be recognized:
a medial one from the tip of the transverse process, and a lateral one from t he
rib. although in t he upper 3 or 4 l evel s, the l atter may merge medi all y wi th t he
fascicle from t he transverse process. Each rostral tendon extends 3 to 4 cm
before forming a smal l muscle bel l y measuring 7 to 8 cm i n l engt h. The muscl e
bel l ies from the hi gher l evel s overl ap those from l ower l evel s. Each muscl e
1 26 Physical Therapy of the Low Back
Fig. 4-9. The thoracic fibers of longi ssi mus ( longi ssi mus thoracis pars thoraci s) . The
intact fasci cl es are shown on the l ef. The darkened areas represent the short muscle
bel l i es of each fasci cl e. Note the short rostral tendons of each fascicl e, and the long
caudal tendons, whi ch col l ecti vel y constitute most of the erector spinae aponeurosis
(ESA). The span of the i ndi vi dual fasci cl es i s i ndi cated on the right. ( From Bogduk and
TwomeY,n with permi ssi on. )
bel l y eventual l y forms a caudal tendon that extends i nto the l umbar region.
The tendons run in paral l el , wi th those from higher l evel s being most medi al .
The fasci cles from t he T2 l evel attach to the L3 spinous process, whereas the
fasci cl es from t he remai ni ng l evel s i nsert into spinous processes at progres­
sivel y l ower l evel s. For exampl e, t hose from T5 attach to L5 and those from
T7 attach to S2 or S3. Those from T8 to TI 2 di verge from the mi dline to find
attachment to the sacrum al ong a l i ne extending from the S3 spinous process
to the caudal extent of the posterior superior i l iac spi ne.
The lateral edge of the
caudal tendon of TI 2 lies alongside the dorsal edge of the l umbar intermuscular
aponeurosis formed by the caudal tendon of the L I l ongissimus bundl e.
The side-to-side aggregati on of the caudal tendons of longissimus thoracis
pars thoracis forms much of what i s termed the erector spinae aponeurosis,
which covers the l umbar fibers of l ongissimus and iliocostal i s, but affords no
attachment to them.
The Lumbar Back Muscles and Their Fascia 1 27
The longi ssi mus t horaci s pars t horaci s i s designed to act on t horacic verte­
brae and ri bs. Nonethel ess, when contracti ng bi l ateral l y i t acts i ndi rect l y on the
l umbar vertebral col umn, and uses the erector spi nae aponeurosi s to produce an
i ncrease i n the l umbar lordosi s . However, not al l of the fasci cl es of longi ssi mus
thoraci s span the ent ire l umbar vertebral col umn. Those from t he second ri b
and T2 reach onl y as far as L3, and onl y t hose fasci cl es ari si ng between t he
T6 or T7 and the TI 2 l evel s actual l y span t he ent i re l umbar regi on . Conse-
, quent l y, onl y a porti on of the whol e t horacic l ongissi mus acts on al l the l umbar
The obl i que orientation of t he l ongi ssi mus thoraci s pars t horaci s al so per­
mits i t to lateral l y flex the t horacic vertebral col umn and t hereby i ndirect l y fl ex
the l umbar vertebral col umn l ateral l y.
Iliocostalis Lumborum Pars Thoracis
The i l iocostal i s l umborum pars thoraci s consi st s of fasci cl es from t he l ower
seven or eight ribs that attach caudal l y to t he i l i um and sacrumlO (Fi g. 4- 1 0) .
These fasci cl es represent t he thoraci c component of i l iocostal i s l umborum, and
should not be confused with the i l iocostali s t horaci s, whi ch is restricted to t he
t horacic region between t he upper si x and l ower si x ri bs.
Each fasci cl e of t he i l i ocostal i s l umborum pars t horaci s ari ses from t he
angle of t he rib vi a a ribbonl i ke tendon measuring some 9 to 10 cm i n l engt h.
I t then forms a muscl e bel l y of 8 to 1 0 cm i n l engt h. Thereafer, each fasci cl e
conti nues as a tendon, contri buti ng to the erector spi nae aponeurosi s, and ul t i­
mat el y attachi ng to the posterior superior i l iac spi ne. The most medi al tendons,
from t he more rostral fasci cl es , often attach more medi al l y t o t he dorsal surface
of the sacrum, caudal to the i nserti on of mul tifidus.
The thoracic fasci cl es of i l iocostal i s l umborum have no attachment t o l um­
bar vertebrae. They attach to t he i l i ac crest and t hereby span t he l umbar regi on.
Consequentl y, by acti ng bi l ateral l y, it i s possi bl e for t hem to exert an i ndirect
"bowstring" effect on the vertebral col umn, ca)Jsi ng an i ncrease in the lordosi s
of the l umbar spi ne. Acti ng uni l ateral l y, t he i l iocostal i s l umborum pars t horaci s
can use t he l everage afforded by t he ri bs to lateral l y flex t he t horacic cage
and t hereby lateral l y flex the l umbar vertebral col umn i ndi rect l y . The di stance
between the ri bs and i l i um does not shorten greatl y duri ng rotation of the trunk,
and therefore the i l iocostal i s l umborum pars t horaci s can have l i t t l e act i on as
an axi al rotator. However, contralateral rotat i on greatl y i ncreases thi s di stance,
and the i l iocostal i s l umborum pars t horaci s can serve to derotate t he thoraci c
cage and, therefore, t he l umbar spi ne.
One of t he cardi nal revel ati ons of recent studi es of t he l umbar erector
spinaes.,o i s that t hi s muscle consi sts of both l umbar and thoracic fi bers. Modern
textbook descriptions largel y do not recognize t he l umbar fibers , especial l y
1 28 Physical Therapy of the Low Back
Fig. 4-10. The thoracic fibers of i l iocostal i s lumborum (i l iocostalis lumborum pars thor­
aci s) . The intact fasci cl es are shown on the left, and thei r span is shown on the right .
The caudal tendons of the fasci cl es col l ecti vel y form the lateral part s of the erector
spi nae aponeurosis (ESA). (From Bogduk and Twomey,92 wi th permi ssion. )
those of i l iocostal i s
; moreover, they do not note that the l umbar fi bers (of both
longi ssi mus and i l iocostal i s) have attachments quite separate from t hose of the
t horacic fi bers. The l umbar fi bers of the l ongi ssi mus and i l iocostal i s pass be­
tween the l umbar vertebrae and the i l i um. Thus, t hrough these muscl es, the
l umbar vertebrae are anchored di rectl y to the i l i um. They -do not gain any
attachment to the erector spi nae aponeurosi s, whi ch i s the impl ication of al l
modern textbook descri pti ons t hat deal wi th the erector spi nae.
The erector spi nae aponeurosi s i s descri bed as a broad sheet of tendi nous
fi bers t hat i s attached to the i l i um, the sacrum, and the l umbar and sacral
spi nous processes, and whi ch forms a common origin for the lower part of
erector spi nae. However, as described above, the erector spi nae aponeurosis
i s formed vi rtual l y excl usi vel y by the tendons of the longi ssi mus t horaci s pars
thoraci s and i l iocostal i s pars t horaci s.
8. 1
The medial half or so of the aponeuro­
sis i s formed by the tendons of longi ssi mus thoraci s, and the lateral half is
formed by the i l iocostal i s l umborum (Fig. 4- 1 1 ). The onl y addi ti onal contribu­
ti on comes from the most superfi ci al fi bers of mul t ifi dus from upper l umbar
The Lllmbar Back Mllscles and Their Fascia 1 2 9
Fig. 4-11. The erector spi nae aponeurosis (ESA). Thi s broad sheet i s formed by the
caudal tendons of the thoracic fibers of longi ssi mus thoracis (LT) and iliocostal i s l umb­
orum (lL). ( From Bogduk and Twomey.n wi th permi ssion. )
l evel s, whi ch contribute a smal l number of fi bers to the aponeurosi s 12 ( see Fi gs.
4- 9 and 4-(0) . Nonethel ess, the erector spi nae aponeurosi s i s essenti al l y formed
only by the caudal attachments of muscl es acti ng from t horacic l evel s.
The l umbar fi bers of erector spi nae do not attach to t he erector spi nae
aponeurosi s. Indeed, the aponeurosi s is free to move over the surface of the
underl yi ng l umbar fi bers, and t hi s suggests that the l umbar fi bers , whi ch form
the bul k of the l umbar back musculat ure, can act i ndependentl y from the rest
of the erector spinae.
The thoracolumbar fasci a consi sts of t hree l ayers of fasci a t hat envelop
the muscl es of the l umbar spine, effecti vel y separating them i nto t hree compart­
ments. The anterior l ayer of t horacol umbar fasci a is quite t hi n, and i s deri ved
1 30 Physical Therapy of the Low Back
from t he fasci a of quadratus l umborum. It covers t he anterior sUce of quadra­
tus l umborum, and i s attached medi al l y to t he anterior surfaces of the l umbar
transverse processes. In the i ntertransverse spaces i t bl ends with the intertrans­
verse l igaments, and may be viewed as one of the lateral extensi ons of the
i ntertransverse ligament s. Lateral to t he quadratus l umborum, the anterior layer
bl ends wi th the other l ayers of the t horacol umbar fasci a.
The mi ddl e l ayer of t horacol umbar fasci a l i es behi nd the quadratus l um­
borum. Mediall y, it is attached to the ti ps of t he l umbar transverse processes,
and i s di rectl y conti nuous wi th the i ntertransverse l igament s. Lateral l y, it gives
ri se to the aponeurosi s of the transversus abdomi ni s. Its actual identity i s debat­
able. It may represent a lateral conti nuati on of the i ntert ransverse l igaments,
a medi al conti nuation of the transversus aponeurosi s, a thi ckeni ng of the poste­
rior fasci a of the quadratus, or a combi nati on of any or all of these.
The posterior l ayer of t horacol umbar fasci a covers the back muscles. It
ari ses from the l umbar spinous processes i n the mi dl i ne posteriorl y, and wraps
around the back muscl es to bl end wi th the other layers of the thoracol umbar
fasci a along the lateral border of the i l i ocostal i s l umborum. The union of the
fasci ae is quite dense at t hi s si te, and the mi ddl e and posterior layers in part i cu­
lar form a dense raphe that, for purposes of reference, has been cal led t he
lateral raphe.
Tradi ti onal l y, the thoracol umbar fascia has been ascri bed no other function
t han to i nvest the back muscl es and to provide an attachment for t he t ransversus
abdomi ni s and t he i nternal obl i que muscl es. 7 However, i n recent years there
has been consi derable i nterest i n i ts bi omechani cal role in the stabi l i ty of the
l umbar spi ne, part icul arl y i n t he flexed posture and i n l ifti ng. Thi s has resulted
in anatomic and bi omechani cal studi es of the anatomy and functi on of t he thora­
col umbar fasci a, notabl y i t s posterior l ayer.
The posterior l ayer of thoracol umbar fasci a covers t he back muscl es from
the l umbosacral region t hrough to the t horacic region as far rost ral l y as the
spl eni us muscl e. I n t he l umbar region, it is attached to the ti ps of the spinous
processes in the mi dl i ne. Lateral to the erector spi nae, between the 1 2th rib
and the i l i ac crest , it unites with the middle layer of thoracolumbar fasci a in
the lateral raphe. At sacral l evel s, the posterior layer extends from the mi dl i ne
to the posterior superior i liac spi ne and t he posterior segment of the i l i ac crest.
On close i nspection, the posterior l ayer exhi bi ts a cross-hatched appear­
ance, mani fest because it consi sts of two l ami nae: a superfici al l ami na wi th
fibers oriented caudomedi al l y and a deep l ami na wi th fibers oriented caudolat­
eral l y.
The superfi ci al l ami na i s formed by the aponeurosi s of lati ssi mus dorsi ,
but the di sposi ti on and attachments of i ts consti tuent fi bers differ according to
t he porti on of l at i ssi mus dorsi from whi ch they are derived (Fi g. 4- 1 2) . Those
fibers derived from the most lateral 2 to 3 cm of the muscl e are short and i nsert
directl y i nto t he i liac crest wi thout contri buti ng to t he thoracol umbar fasci a.
Fi bers from the next most lateral 2 cm of t he muscl e approach the i l iac crest near
t he lateral margin of the erector spi nae, but then defl ect medial l y, bypassi ng the
crest t o attach to the L5 and sacral spinous processes. These fibers form the
The Lllmbar Back Muscles and Their Fascia 1 3 1
Fig. 4-12. The superficial lamina of the posterior layer of thoracol umbar fasci a. J ,
Aponeurotic fibers of the most lateral fasci cl es of l ati ssimus dorsi i nsert di rectl y i nto
the i l iac crest . 2, Aponeuroti c fibers of the next most lateral part of l ati ssi mus dorsi
glance past the i l iac crest and reach the mi dl i ne at sacral l evel s. 3, Aponeuroti c fibers
from thi s portion of the muscl e attach to the underl yi ng l ateral raphe (LR), and t hen
deflect medial l y to reach the mi dl i ne at the L3 to L5 l evel s. 4, Aponeurotic fibers from
the upper portions of l ati ssi mus dorsi pass di rectl y to the mi dl i ne at thoracol umbar
l evel s. (From Bogduk and Twomey,92 wi th permi ssi on. )
sacral portion of the superficial l ami na. A thi rd series of fi bers become aponeu­
rotic just lateral to the l umbar erector spi nae. At the lateral border of the erector
spinae they blend wi th t he other layers of t horacol umbar fasci a i n the lateral
raphe, but t hey they defl ect medi al l y, conti nui ng over the back muscl es to reach
the mi dl i ne at the l evel s of t he L3, L4, and L5 spi nous processes. These fi bers
form the l umbar portion of the superfi cial l ami na of the posterior layer of t hora­
col umbar fasci a.
The rostral porti ons of the l ati ssi mus dorsi cross the back muscl es and do
not become aponeurotic unti l some 5 cm lateral to the mi dl i ne at the L3 and
higher level s. These aponeuroti c fibers form the thoracol umbar and thoracic
porti ons of the thoracol umbar fasci a.
Beneath the superficial l ami na, t he deep l ami na of the posterior l ayer con­
si sts of bands of col l agen fibers emanati ng from the mi dl i ne, pri nci pal l y from the
l umbar spinous processes (Fi g. 4- 1 3) . The bands from L4, L5, and S I spi nous
processes pass caudolateral l y to t he posterior superi or i l iac spi ne. Those from
the L3 spinous process and L3-L4 i nterspinous l igament wrap around the lat­
eral margin of the erector spi nae to fuse with the middl e layer of t horacol umbar
fascia i n the lateral raphe. Above L3 the deep l ami na progressi vel y becomes
thinner, consi sti ng of sparse bands of col lagen t hat di ssi pate lateral l y over the
erector spi nae. A deep l ami na i s not formed at t horaci c l evel s.
Col l ecti vel y, t he superfi ci al and deep l aminae of t he posterior l ayer of
thoracolumbar fasci a form a reti nacul um over the back muscl es. Attached to
1 32 Physical Therapy of the Lol l ' Back
Fig. 4-13. The deep lamina of the posterior layer of thoracolumbar fascia. Bands of
collagen fibers pass from the mi dl ine to the posterior superior i l iac spine and to the
lateral raphe (LR). Those bands from the L4 and L5 spinoll s processes form alarl i ke
l igaments that anchor these processes to the i l i um. Attaching to the lateral raphe lateral l y
are the aponeurosis of transversus abdomi ni s (Ia). and a variable number of the most
posterior fibers of i nternal obl ique (io) . ES. erector spi nae. ( From Bogduk and Two­
mey,n with permission. )
the mi dl i ne medi al l y and the posterior superior i l iac spi ne and lateral raphe
lateral l y, the fascia covers or ensheaths the back muscl es preventi ng thei r di s­
placement dorsal l y. Addi ti onal l y, the deep l ami na alone forms a series of di s­
ti nct l i gament s. When vi ewed bi l ateral l y, the bands of fi bers from the L4 and
L5 spi nous processes appear l i ke al ar l igaments anchoring these spinous pro­
cesses to the i l i a. The band from the L3 spi nous process anchors thi s process
i ndi rect l y to the i l i um via the lateral raphe. Thi rdl y, the lateral raphe forms a
site where the two l ami nae of the posterior l ayer fuse not onl y wi th the middle
layer of t horacol umbar fasci a, but al so with the t ransversus abdomi ni s, whose
mi ddl e fibers arise from the lateral raphe (see Fig. 4- 1 3 ) . The posterior layer
of t horacol umbar fasci a thereby provi des an i ndi rect attachment for the
transversus abdomi ni s to the l umbar spinous processes. The mechanical signifi­
cance of t hese t hree morphologic features is expl ored below.
Each of the l umbar back muscl es is capabl e of several possible actions.
No acti on i s unique to a muscl e and no muscl e has a si ngle action. l nstead,
the back muscl es provide a pool of possi bl e actions that may be recruited to
sui t the needs of the vertebral col umn . Therefore, the functions of the back
The Lumbar Back Muscles and Their Fascia 1 33
muscl es need to be consi dered in terms of the observed movements of the
vertebral col umn. I n t hi s regard, t hree t ypes of movements can be addressed:
(I ) mi nor acti ve movements of the vertebral col umn, ( 2) postural movement s,
and ( 3) major movements i n forward bendi ng and l ifti ng. I n t hi s context postural
movements refers to movement s, usual l y subconsci ous , that occur to adj ust
and mai ntai n a desi red post ure when thi s i s di sturbed, usual l y by the i nfl uence
of gravi ty.
Minor Active Movements
In the upright posit ion, the l umbar back muscl es pl ay a mi nor, or no acti ve
role i n executi ng movement , because gravi ty provi des the necessary force.
During extension, the back muscl es contri bute to t he i ni ti al t i l t , drawi ng the
l i ne of gravi ty backward,
� but are unnecessary for further extensi on. Muscl e
acti vi ty i s recruited when the movement i s forced or resi sted,
but i s restricted
to muscl es act i ng on the t horax. The l umbar mul tifidus, for exampl e, shows
l i ttl e or no i nvolvement . 3o
The lateral flexors can bend the l umbar spi ne sideways, but once the center
of gravi ty of the trunk i s di spl aced, lateral fl exion can conti nue under the i nflu­
ence of gravi t y. However, the i psi l ateral lateral flexors are used to di rect the
movement , and the contralateral muscl es are required to balance the action of
gravi ty and control the rate and extent of movement. Consequent l y, lateral
fl exion i s accompanied by bi lateral acti vi ty of the l umbar back muscl es, but
the contralateral muscles are relati vel y more acti ve, because t hey must balance
the l oad of the lateral l y flexing spi neY·
If a weight is hel d i n the hand
on the side to whi ch the spine i s lateral l y flexed, a greater load i s appl i ed to
t he spine, and the contralateral back muscl es show greater act i vi t y to balance
this load. 3 1
Maintenance of Posture
The upright vertebral col umn is wel l stabi l ized by i ts joi nts and l igaments,
but it i s sti l l l i able to di splacement by gravi ty or when subject to asymmet rical
weightbearing. The back muscl es serve to correct such di spl acement s, and
depending on the di recti on of any di splacement , the appropriate back muscl es
wi l l be recruited.
During standi ng at ease, the back muscl es may show sl i ght cont i nuous
acti vi ty ,
. �
i ntermi ttent acti vi ty,
or no act i vi t y,
4 1 -43
and the amount of acti vi ty can be i nfl uenced by changing the posi ti on of the
head or al l owi ng the trunk to sway .
The explanation for t hese differences probabl y l i es in the l ocation
of the l i ne of gravi ty i n rel ati on to the l umbar spi ne i n different i ndi vid­
ual s.
7 In about 75 percent of i ndi vidual s the l i ne of gravi ty passes
in front of t he center of the L4 vertebra, and therefore essenti al l y i n front of
1 34 Physical Therapy of the Low Back
the l umbar spi ne. 38. 43 Consequentl y, gravi ty wi l l exert a constant tendency to
pul l the t horax and l umbar spi ne i nto fexi on. To preserve an upright posture,
a constant l evel of acti vi ty in the posterior sagittal rotators of the l umbar spine
wi l l be needed to oppose the tendency to fexion. Conversel y, when the l i ne
of gravi ty passes behi nd t he l umbar spi ne, gravi ty tends to extend it, and back
muscl e acti vi t y is not requi red. I nstead, abdominal muscl e act i vi ty is recruited
to prevent the spi ne from extendi ng under gravi t y. 38.43
Acti vi ti es t hat di spl ace the center of gravi ty of the trunk sideways will tend
to cause lateral fexi on. To prevent undesi red lateral fexion, the contralateral
lateral flexors wi l l contract . Thi s occurs when wei ghts are carried i n one
handY. 41 Carrying equal wei ghts in both hands does not di splace the line of
gravi t y, and back muscl e acti vi ty is not i ncreased substanti al l y on ei ther side
of the bodyY.4
Duri ng si tti ng, the acti vi ty of the back muscl es is si mi l ar to that during
standi ng,3
. 37.48.49 but i n supported si tti ng, as with the el bows resti ng on t he
knees, t here i s no act i vi ty in the l umbar back muscles ,27.34 and wi th arms resting
on a desk, back muscl e acti vi ty i s substanti al l y decreased. 3
.37.48 In reclined
si tti ng, the back rest supports the weight of the thorax, l esseni ng t he need for
muscular support . Consequent l y, i ncreasi ng the recl i ne of the back rest of a
seat decreases l umbar back muscl e acti vi t y. 3
6. 37
Major Active Movements
Forward flexi on and extensi on of the spi ne from the fexed position are
movements duri ng whi ch the back muscl es have thei r most i mportant function.
As the spi ne bends forward, there is an i ncrease in the acti vi ty of the back
muscl es,22
. 5
1 -
54 and thi s i ncrease is proport ional to the angle
of fl exi on and the si ze of any load carri ed. 3
. 33. 55. 5
The movement of forward
flexi on i s produced by gravi ty, but the extent and the rate at whi ch it proceeds
is control led by the eccentri c contraction of the back muscl es . Movement of
the thorax on the l umbar spi ne is control l ed by the long thoracic fi bers of
longi ssi mus and i l iocostal i s. The long tendons of i nsertion al l ow these muscl es
to act around the convexi ty of the i ncreasi ng t horacic kyphosi s and anchor the
thorax to the i l i um and sacrum. I n the l umbar region, the mul tifidus and the
l umbar fasci cl es of longi ssi mus and i l iocostal i s act to control the anterior sagittal
rotation of the l umbar vertebrae. At the same ti me the l umbar fasci cl es of
longi ssi mus and i l i ocostal i s also act to control the associated anterior translation
of the l umbar vertebrae.
At a certai n point duri ng forward flexi on, the acti vi ty in the back muscl es
ceases, and the vertebral col umn i s braced by the l ocki ng of the zygapophyseal
joi nts and tensi on in its posterior l igaments (Ch. 7) . Thi s phenomenon is known
as " cri tical point " . 44
45. 47. 57 However, cri t ical point does not occur in al l i ndi vid­
ual s, or i n al l muscl es . 22. 27. 34. 45.5
When it does occur, i t does so when the spi ne
has reached about 90 percent maxi mum flexion, even though at t hi s stage, the
hip fl exi on that occurs i n forward bendi ng is sti l l only 60 percent complete.47. 57
The Lllmbar Back Muscles and Their Fascia 1 35
Carrying weights during flexi on causes the cri ti cal poi nt to occur later in t he
range of vertebral flexion.
The physiologic basi s for cri ti cal point is st i l l obscure. It may be due to
refl ex i nhi bition i ni ti ated by proprioceptors i n the l umbar joi nts and l igaments ,
or in muscl e stretch and l ength receptors.
7 Whatever the mechani sm, the signif­
icance of critical point i s t hat it marks the transi ti on of spi nal loadbearing from
muscl es to the l igamentous system.
Extension of t he trunk from t he fexed posi ti on i s characterized by hi gh
l evel s of back muscl e acti vit y.
I n the thoracic region, the i l i ocostal i s
and longi ssi mus, acti ng around the t horacic kyphosi s, l i ft the t horax by rotating
it backward. The l umbar vertebrae are rotated backward, pri nci pal l y by the
l umbar mul tifi dus, causi ng thei r superior surfaces to be progressi vel y t i l ted
upward to support the ri si ng t horax.
Compressive Loads of the Back Muscles
Because of the downward di recti on of thei r acti on, as the back muscl es
contract they exert a l ongi tudi nal compressi on of t he l umbar vertebral col umn,
and thi s compression rai ses the pressure i n t he l umbar i ntervertebral di scs.
Any acti vi ty that i nvol ves the back muscl es, therefore, i s associ ated wi th a ri se
in nuclear pressure. As measured i n the L3-L4 i ntervertebral di sc, the nuclear
pressure correlates with the degree of myoelectric acti vi ty i n the back mus­
cl es. 3 1
8 As muscl e acti vi ty i ncreases, di sc pressure ri ses.
Di sc pressures and myoel ectri c acti vi ty of t he back muscl es have been
used extensi vel y to quantify the stresses appl i ed to the l umbar spi ne i n various
postures and by various acti vi ti es. 36
From the standi ng posi ti on, for­
ward bendi ng causes t he greatest i ncrease in di sc pressure. Lifti ng a weight i n
t hi s position raises di sc pressure even further, and the pressure i s great l y i n­
creased if a l oad i s l i fted wi th the l umbar spi ne both flexed and rotated. Through­
out these various maneuvers, back muscl e acti vi ty i ncreases i n proporti on to
t he di sc pressure.
One of the prime revel ati ons of combi ned di scometric and el ect rom yo­
graphic studies of the l umbar spine duri ng l ifti ng relates to the comparati ve
stresses applied to the l umbar spi ne by di fferent l i ft i ng t acti cs. I n essence, i t
has been shown that , on the basi s of changes i n di sc pressure and back muscl e
acti vi ty, there are no di ferences between usi ng a "stoop" l i ft or a "leg" l i ft
( i . e . , l ift ing a weight wi th a bent back versus l ifti ng wi th a straight back) . 3
The critical factor is the di stance of the load from the body. The further the
load i s from t he chest the greater the stresses on the l umbar spine, and the
greater the di sc pressure and back muscl e act i vi t y.
Strength of the Back Muscles
The strength of the back muscl es has been determined in experi ments on
normal vol unteers.
8 Two measures of strength are avail abl e: the absol ute maxi­
mum force of contraction i n the upright posture and the moment generated on
1 36 Physical Therapy of the Lol l ' Back
the l umbar spi ne. The absol ute maxi mum strength of the back muscl es as a
whole is about 4000 N. Acti ng on the short moment arms provided by the
spinous processes and pedi cl es of the l umbar vertebrae, this force converts to
an extensor moment of 200 Nm. These figures appl y to average males under
the age of 30 years ; young femal es exhibit about 60 percent of this strengt h,
whereas i ndi vidual s over t he age of 30 years are about 1 0 t o 30 percent weaker,
respecti vel y. 2
By comparing these val ues wi th the detai l ed morphol ogy of the back mus­
cl es, model i ng studi es have revealed the extent to whi ch i ndi vi dual muscl es
contri bute to the overal l strength of the back muscl es.
It emerges that with
respect to extensi on moments , about half the total , maxi mum moment exerted
on the LS-S I joi nt i s generated by the t horacic fi bers of longi ssi mus t horaci s and
i l i ocostal i s l umborum. The remainder i s generated by those fibers of mul ti fi dus,
longi ssi mus, and i l iocostal i s that arise from the l umbar vertebrae, wi th mult ifi­
dus contri buti ng about half of t hi s remai nder.
In biomechanical terms, the act of l ifting consti tutes a problem in bal ancing
moment s. When an i ndi vidual bends forward to execute a l i ft , fexi on occurs
at the hip joi nt and in the l umbar spi ne. I ndeed most of the forward movement
seen duri ng t runk fl exi on occurs at t he hip joi nt .
7 The fexion forces are gener­
ated by gravi ty acti ng on the mass of the object to be l i fted and on the mass
of the trunk above the l evel of the hip joi nt and l umbar spine (Fig. 4- 1 4) . These
forces exert flexi on moments on both the hip joint and the l umbar spine . In
each case the moment wi l l be the product of the force and its perpendi cul ar
di stance from the joi nt i n quest i on. The total fl exi on moment acti ng on each
joi nt wi l l be the sum of the moments exerted by the mass to be l i fted and the
mass of the t runk. For a l i ft to be executed these fl exi on moments have to be
overcome by a moment acti ng in the opposite di rection . This could be exerted
by l ongi tudi nal forces act i ng downward behi nd the hi p joint and vertebral col­
umn or by forces acti ng upward i n front of the joi nts pushi ng the trunk upward.
There are no doubts as to the capaci ty of t he hi p extensors to generate
large moments and overcome the fl exi on moments exerted on the hip joi nt even
by the heaviest of loads that might be l i fted.
.68 However, the hi p extensors
are onl y able to rotate the pel vi s backward on the femurs ; they do not act on
the l umbar spi ne. Thus, regardl ess of what happens at the hi p joi nt, the l umbar
spine sti l l remai ns subject to a fl exi on moment that must be overcome i n some
other way. Wi thout an appropriate mechani sm the l umbar spine would stay
flexed as the hi ps extended; i ndeed, as the pel vi s rotated backward, flexion of
the l umbar spi ne woul d be accent uated as its bottom end was pul led backward
wi th the pel vi s whi l e i ts top end remained stat i onary under the load of the
flexion moment . A mechani sm is requi red to al l ow the l umbar spi ne to resist
t hi s deformati on or to cause i t to extend in unison with the hip joi nt .
Despi te much i nvestigati on and debat e, the exact nature of t hi s mechanism
The Lllmbar Back Muscles and Their Fascia 1 37
Fig. 4-14. The fl exi on moments exerted on a flexed trunk. Forces generated by the
weight of the trunk and the load to be l i fted act vertical l y in front of the lumbar spi ne
and hi p joi nt. The moments they exert on each joi nt are proportional to the di stance
between the l i ne of action of each force and the joint in questi on. The mass of the trunk
(ml ) exerts a force ( WI ) that acts a measurable di stance i n front of the lumbar spi ne ( dl )
and the hi p joint (d3). The mass t o be l ifted (m2) exerts a force ( W2) that acts a measurable
distance from the lumbar spi ne (d2) and hip joi nt (d4) . The respecti ve moments acting
on the lumbar spi ne will be Wl dl and Wl d3; those on the hip joint wi l l be W2d2 and
W2d4. ( From Bogduk and Twomey,92 wi th permi ssi on. )
remains unresol ved. I n various ways the back muscl es, i nt raabdomi nal pres­
sure, the thoracol umbar fascia, and the posterior l igamentous system have been
believed to parti ci pate.
For l ight l ifts the fl exi on moments generated are relati vel y smal l . I n the
case of a 70-kg man l ift i ng a 1 0-kg mass i n a ful l y stooped posi ti on, the upper
trunk weighs about 40 kg and acts about 30 cm i n front of the l umbar spi ne
whi l e the arms hol di ng the mass to be l ifed l i e about 45 cm i n front of the
l umbar spi ne. The respecti ve flexi on moments are therefore 40 x 9. 8 x 0. 30
= 1 1 7. 6 Nm and 1 0 x 9. 8 x 0. 45 = 44. 1 Nm, for a total of 1 6 1 . 7 Nm. Thi s
load i s wel l wi thi n the capaci ty of the back muscl es (200 Nm, see Strength of
the Back Muscles) . Thus, as the hi ps extend, the l umbar back muscl es are
capable of resi sti ng furt her flexion of the l umbar spi ne, and i ndeed, could even
acti vel y extend i t, and the wei ght woul d be l i fted.
1 38 Physical Therapy of the Low Back
I ncreasing the l oad to be l i fted to over 30 kg i ncreases the flexion moment
to 1 32. 2 Nm, whi ch when added to the flexi on moment of t he upper trunk
exceeds t he capaci ty of t he back muscl es. To remai n wi thi n the capaci ty of the
back muscl es such loads must be carried closer to the l umbar spi ne ( i . e . , they
must be borne with a much shorter moment arm) . Even so, decreasing the
moment arm to about 1 5 cm l i mi t s the load to be carried to about 90 kg. The
back muscl es are si mpl y not strong enough to raise greater loads. Such realiza­
ti ons have generated concepts of several additional mechani sms that serve to
ai d the back muscl es in overcomi ng large fexi on moments.
In 1 957 Bartel i nk
9 raised the proposi ti on t hat i ntraabdominal pressure
coul d aid t he l umbar spi ne in resi sti ng fl exi on by act i ng upward on the dia­
phragm-the so-called i ntraabdominal bal l oon mechani sm. Bartel i nk hi mself
was circumspect and reserved i n raising t hi s conjecture, but the concept was
rapidl y popul arized, parti cul arl y among physi ot herapi st s. Even though i t was
never val idated, the concept seemed to be treated as proven fact . It received
early endorsement in orthopedi c ci rcl es ,30 and i ntraabdominal pressure was
adopted by ergonomi sts and others as a measure of spi nal stress and safe l ifi ng
standards. 70
I n more contemporary studi es, i ntraabdominal pressure has
been moni tored duri ng various spi nal movements and l ift ing tasks. 3 1 . 65
Reservati ons about the val i di ty of the abdomi nal bal l oon mechani sm have
ari sen from several quarters . Studi es of l ifti ng tasks reveal that unl i ke myoelec­
tric act i vit y, i nt raabdominal pressure does not correlate well with the size of
the load bei ng l ifed or the appl i ed stress on the vertebral column as measured
by i ntradi scal pressure.
77. 78 I ndeed, del i beratel y i ncreasi ng i ntraabdominal
pressure by a Val salva maneuver does not rel ieve the load on the l umbar spine
but actual l y i ncreases i t . 79 Cl i nical studi es have shown that al though abdominal
muscl es are weaker than normal in pati ents with back pai n, i ntraabdominal
pressure is not different. 80 Furt hermore, strengtheni ng the abdomi nal muscl es
both i n normal i ndi vi dual s8
and i n patients wi th back pain8
does not infl uence
i ntraabdomi nal pressure duri ng l ifting.
The most strident cri ti ci sm of the i ntraabdomi nal bal l oon theory comes
from bioengineers and others who maintain that (I ) to generate any significant
anti fexi on moment the pressure required would exceed the maximum hoop
tensi on of the abdomi nal muscl es83
; (2) such a pressure would be so high as
to obstruct the abdomi nal aorta83 (a reservati on raised by Bartel i nk hi msel f
and ( 3) because the abdomi nal muscl es l i e i n front of the l umbar spi ne and
connect the t horax to the pel vi s, whenever t hey contract to generate pressure
t hey must also exert a fl exi on moment on the trunk, whi ch would negate any
antifl exi on val ue of the i ntraabdomi nal pressure.
These reservati ons i nspi red an al ternat i ve expl anation of the role of the
abdomi nal muscl es duri ng l i fti ng. Farfan, Gracovet sky, and col leagues2
7. 8
noted the cri sscross arrangement of the fi bers in t he posterior layer of thoraco­
l umbar fasci a and surmi sed that i f lateral tensi on was applied to t hi s fascia it
woul d resul t in an extension moment bei ng exerted on the l umbar spinous
processes. Such tensi on coul d be exerted by the abdomi nal muscl es that arise
from the t horacol umbar fasci a, and the trigonometry of the fi bers i n the thoraco-
The Lllmbar Back Muscles and Their Fascia 1 39
l umbar fascia was such that they coul d convert l ateral tensi on i nto an apprecia­
ble extension moment-the so-cal led gain of the t horacol umbar fasci a. t
role of the abdomi nal muscl es duri ng lifting was t hus to brace, i f not actual l y
extend, the l umbar spi ne by pul l i ng on t he t horacol umbar fasci a. Any ri ses i n
i ntraabdominal pressure were thereby onl y coi nci dental , occurring because of
the contraction of the abdomi nal muscl es acti ng on the t horacol umbar fasci a.
Subsequent anatomic studi es revealed several l i abi l i ti es of t hi s mode. 89
First , the posterior layer of t horacol umbar fasci a is wel l devel oped onl y i n the
lower l umbar regi on, but nevert hel ess i t s fi bers are appropriatel y oriented to
enable lateral tension exerted on the fasci a to produce extensi on moments at
least on t he L2 to LS spinous processes ( Fi g. 4- I S) . However, di ssecti on reveal s
that of the abdomi nal muscl es i nternal obl i que offers onl y a few fi bers t hat
irregularly attach to the thoracol umbar fasci a; transversus abdomi ni s i s t he onl y
muscle that consi stentl y attaches to the t horacol umbar fasci a, but onl y i ts very
middle fi bers do so. The si ze of these fi bers is such t hat even on maxi mum
contraction the force t hey exert i s very smal l . Cal cul ati ons revealed t hat the
extensor moment they coul d exert on the l umbar spi ne amounted to l ess t han
6 N m. 89 Thus the contribution that abdomi nal muscl es mi ght make to antitlexi on
moments i s t ri vial , a concl usi on al so borne out by subsequent , i ndependent
model i ng studi es. 9
Fig. 4- 15. The mechani cs of the t horacol umbar fasci a. From any poi nt in the lateral
raphe (LR), lateral tensi on i n the posterior l ayer of thoracolumbar fasci a i s transmi tted
upward through the deep l ami na of the posterior l ayer, and downward through the
superfi ci al l ayer. Because of the obl i qui ty of these l i nes of tensi on, a smal l downward
vector i s generated at the mi dl i ne attachment of t he deep l ami na, and a smal l upward
vector i s generated at the mi dl i ne attachment of the superfici al l ami na. These mutual l y
opposite vectors tend to approximate or oppose the separation of L2 and L4 and of L3
and L5 spinous processes. Lateral tensi on on the fasci a can be exerted by the transversus
abdomi ni s ( TA) . and to a l esser extent by the few fibers of i nternal obl i que when they
attach to the lateral raphe. ( From Bogduk and Twomey,92 wi th permi ssion. )
1 40 Physical Therapy ofthe Low Back
A total l y di fferent model of l ifti ng was elaborated by Falfan and Gracovet­
sky. 26
7. 8
Noti ng the weakness of the back muscl es, t hese authors proposed
that extensi on of the l umbar spi ne was not requi red to l i ft heavy loads or loads
wi th l ong moment arms. They proposed that the l umbar spine should remain
ful l y flexed in order to engage (i . e. , maxi mal l y stretch) , what they referred to as
the "posterior l igamentous system, " namel y the capsul es of the zygapophyseal
joi nt s, the i nterspi nous and supraspinous l igament s, and the posterior layer of
t horacol umbar fasci a, the l atter act i ng passi vel y to t ransmi t tensi on between
the l umbar spi nous processes and i l i um.
Under such condi t ions the acti ve energy for a l i f was provided by the
powerful hip extensor muscl es. These rotated the pel vi s backward. Meanwhi l e,
the external l oad act i ng on the upper trunk kept the l umbar spi ne fl exed. Ten­
sion woul d devel op in the posterior l igamentous system, whi ch bridged the
thorax and pel vi s. With the posterior ligamentous system so engaged, as the
pelvis rotated backward the l umbar spine woul d be passi vel y raised while re­
mai ni ng in a ful l y fl exed posi ti on. In essence, the posterior sagittal rotation of
the pel vi s woul d be transmi tted t hrough the posterior l igaments fi rst to t he L5
vertebra, t hen to L4, and so on, up through the l umbar spine into t he thorax.
Al l t hat was requi red was that the posterior ligamentous system be sufci entl y
strong to wi thstand t he passi ve tensi on generated i n i t by the movement of the
pel vi s at one end and the weight of the trunk and exteral load at the other.
The l umbar spi ne woul d thereby be rai sed l i ke a long, ri gi d arm rotat i ng on the
pel vi s and rai si ng the external load with i t .
Contracti on of the back muscl es was not requi red i f the l igaments could
take the load. I ndeed, muscl e contraction was di sti nctl y undesi rabl e, because
any acti ve extensi on of the l umbar spine woul d di sengage the posterior l iga­
ments and precl ude them from t ransmi tti ng tensi on. The back muscl es could
be recrui ted only once the trunk had been raised suffi ci entl y to shorten the
moment arm of the external load, reduci ng i ts flexion moment to within the
capaci ty of the back muscl es.
The attracti on of t hi s model was that i t overcame the problem of the relat ive
weakness of the back muscl es by di spensi ng with their need to act , which i n
turn was consi stent wi th the myoel ectric si l ence of the back muscl es at ful l
flexi on of t he trunk and t he recrui tment of muscl e acti vi ty onl y once t he t runk
had been el evated and the flexion moment arm had been reduced. Support for
t he model al so came from surgical studi es that reported t hat if the midl i ne
l igaments and t horacol umbar fasci a were consci enti ousl y reconstructed after
mul ti l evel l ami nectomi es, the postoperati ve recovery and rehabi l i tation of pa­
ti ents were enhanced. 91
However, al though attracti ve i n a qual i tati ve sense, the mechanism of the
posterior l igamentous system was not val idated quanti tati vel y. The model re­
qui res t hat the l igaments be strong enough to sustain the loads appl ied. I n thi s
regard, data on the strength of the posterior l igaments are scant and i rregular,
but suffi ci ent data are available to permit an i ni ti al appraisal of the feasibi l i ty
of the posterior l igament model .
The strength of spi nal l igaments varies consi derabl y, but average val ues
The Lumbar Back Muscles and Their Fascia 1 4 1
Table 4- 1 . Strength of the Posterior Ligamentous System
Average Force
at Failure" Moment Armb Maximum Moment
Reference ( N) ( m) ( Nm)
94 90 0.02 1. 8
94 244 0.03 7.3
94 680 0.04 27.2
95 672
94 1 07 0.05 5.4
94 500 0.06 30.0
5 1 . 7
Abbrel'ialiolls: ISL, interspinous ligament; LF, ligment um tlavum; PLL, posterior longitudinal
ligament ; TLF, the posterior layer of thoracolumbar fascia and the erector spinae aponeurosis that
forms the so-called supraspinous ligament; ZJC, zygapophyseal joint capsules (bilaterally).
" Average force at failure has been calculated using raw data provided in the references cited.
b Moment arms are estimates based on inspection of a representative vertebra measuring the
perpendicular distance between the location of the instantaneous axis of rotation and the sites of
attachment of the various ligaments.
( From Bogduk and Twomey,n with permission. )
can be calculated. Tabl e 4- 1 summarizes some of the avai l abl e data. The strong­
est posterior "l igaments" of the l umbar spi ne are the zygapophyseal joi nt cap­
sules and the thoracol umbar fasci a forming the mi dl i ne "supraspi nous l iga­
ment . " However, when the relati vel y short moment arms over whi ch t hese
ligaments act are consi dered, it t ranspi res that the maxi mum moment they can
sustai n i s rel ati vel y smal l . Even the sum total of all their moments is consi der­
ably less than that requi red for heavy l ifti ng and i s some four ti mes l ess t han
the maxi mum strength of the back muscl es. Of course, i t is possi bl e t hat the
data quoted may not be representative of the true mean val ues of t he strength
of these ligaments, but i t does not seem l i kel y t hat the l i terature quoted underes­
ti mated thei r strength by a factor of four or more. Under t hese condi ti ons, i t
i s evident that posterior l igamentous system alone i s not strong enough to per­
form the role requi red of i t i n heavy l i fti ng. The posterior l igamentous system
i s not strong enough to repl ace the back muscl es as a mechani sm to prevent
flexion of the l umbar spi ne duri ng l ifti ng. Some other mechani sms must operate.
One such mechani sm i s that of t he hydraul i c ampl ifi er effect.
8 Origi nal l y
set forth by Gracovetsky et al ,88 i t proposed that because the t horacol umbar
fascia surrounded the back muscl es as a reti nacul um i t could serve to brace
these muscl es and enhance thei r power. The engi neering basi s for t hi s effect
i s compl i cated, and the concept remained unexpl ored until very recentl y. A
mathematical proof has been publ i shed that suggests that by i nvesti ng the back
muscles the thoracol umbar fasci a enhances the strength of the back muscl es
by some 30 percent . 9
Thi s i s an appreci abl e i ncrease and an attracti ve mecha­
nism for enhanci ng the antifl exi on capaci ty of the back muscl es . However, the
val idity of thi s proof i s sti l l bei ng questioned on the grounds that t he pri nci pl es
used, al though appl icabl e to the behavior of sol i ds, may not be appl i cabl e to
muscl es; the concept of the hydraul i c ampl ifier mechani sm sti l l remai ns under
scrut i ny.
Quite a contrasti ng model has been proposed to expl ai n the mechani cs of
1 42 Physical Therapy of the Low Back
t he l umbar spi ne in l i ft i ng. It i s based on arch theory and mai ntai ns that the
behavi or, stabi l i t y, and strength of the l umbar spi ne duri ng l ift ing can be ex­
pl ai ned by vi ewi ng t he l umbar spi ne as an arch braced by i ntraabdomi nal pres­
sure. 9
. 97 Thi s i ntrigui ng concept , however, has not met wi th any degree of
acceptance, and i ndeed, has been chal lenged from some quarters . 98
Despi te much effort over recent years the exact mechani sm of heavy lifting
st i l l remai ns unexpl ai ned. The back muscl es are too weak to extend the l umbar
spi ne against l arge flexi on moment s; t he i ntraabdomi nal bal l oon has been re­
futed; the abdomi nal mechanj sm and thoracol umbar fasci a has been refuted;
and the posterior l igamentous system appears too weak to replace t he back
muscl es. Engineering model s of the hydraul i c ampl ifier effect and the arch
model are sti l l subject to debate.
What remai ns to be explai ned i s what provi des the mi ssi ng force t o sustain
heavy loads, and why i ntraabdomi nal pressure is so consi stentl y generated
duri ng l ifts i f it is nei ther to brace the thoracol umbar fasci a nor to provide an
i ntraabdomi nal bal l oon. At present t hese questi ons can only be addressed by
conjecture, but certain concepts appear wort hy of consi deration.
Wi th regard to i ntraabdomi nal pressure, one concept that has been over­
looked i n studi es of l ifti ng is the rol e of the abdomi nal muscl es i n control l i ng
axi al rotation of the t runk. Investi gators have focused thei r attention on move­
ments in the sagittal plane duri ng l ifti ng and have ignored the fact that when
bent forward to address an object to be l i fted the trunk is l i able to axi al rotation.
Unl ess the external load i s perfect l y balanced and l i es exactl y in the mi dl i ne,
i t wi l l cause the t runk to twi st to the l eft or t he right . Thus, to keep the weight
in the mi dl i ne and in the sagittal pl ane, the l ifter must control any twi sti ng
efect . The obl i que abdomi nal muscl es are the pri nci pal rotators of the trunk
and would be responsi bl e for t hi s braci ng. In contracti ng to control axi al rota­
ti on, t he abdominal muscl es woul d secondari l y raise i ntraabdominal pressure.
Thi s pressure rise i s t herefore an epi phenomenon and woul d reflect not the size
of any external load but its tendency to twi st the flexed trunk.
Wi th regard to loads i n t he sagittal pl ane, the passive strength of the back
muscl es has been neglected in discussions of l ift i ng. From the behavior of isolate
muscl e fi bers i t is known that as a muscl e el ongates its maxi mum contractile
force di mi ni shes, but its passi ve el asti c tensi on ri ses; t hi s is to such an extent
t hat i n an elongated muscl e the total passi ve and acti ve tension generated is at
l east equal to the maxi mum contracti l e capaci ty of the muscl e at resting length.
Thus, al though they become el ectrical l y si l ent at ful l flexi on, t he back muscl es
are sti l l capabl e of provi di ng passi ve tensi on equal to thei r maxi mum contractile
strengt h. This woul d allow the si l ent muscl es to suppl ement the engaged poste­
rior l igamentous system. With t he back muscl es provi di ng some 200 Nm and
t he l igaments some 50 Nm or more, t he total anti fexion capaci ty of the l umbar
spi ne rises to about 250 Nm, whi ch would al l ow some 30 kg to be safel y l i fted
The Lumbar Back Muscles and Their Fascia 1 43
at 90° trunk fl exi on. Larger loads coul d be sustai ned by proportional l y shorten­
ing t he moment arm. Consequent l y, the mechani sm of l i fi ng may well be essen­
ti al l y as proposed by Farfan and Gracovetsky,
. 6
7.88 except t hat the passi ve
tensi on i n the back muscl es consti tutes the major component of the "posteri or
ligamentous system. "
I . Basmajian J V: Electromyography of iliopsoas. Anat Rec 1 32 : 1 27, 1 958
2. Nachemson A: El ectromyographic studies on the vertebral portion of the psoas
muscle. Acta Orthop Scand 37: 1 77, 1 966
3. Nachemson A: The possible i mportance of the psoas muscle for stabil ization of the
l umbar spine. Acta Orthop Scand 39: 47, 1 968
4. Bogduk N, Pearcy M, Hadfield G: Anatomy and biomechanics of psoas maj or. Cl i n
Biomech 7: 1 09, 1 992
5. Cave AJ E: The i nnervation and morphology of the cervical intertransverse muscl es.
J Anat 7 1 :497, 1 937
6. Poirier P: Myologi e. p. 1 39. I n Poirier P, Charpy A ( eds) : Traite d' Anatomie Hu­
mai ne. 3rd Ed. Vol . 2, Fasc I . Masson, Pari s, 1 9 1 2
7. Wi l l iams PL, Warwick R: Gray' s Anatomy. 37th Ed. Churchi l l Li vingstone, Lon­
don, 1 98<
8. Bogduk N: A reappraisal of the anatomy of the human l umbar erector spinae. J
Anat 1 3 1 : 525 , 1 980
9. Macintosh J E, Bogduk N: The biomechanics of the l umbar multifi dus. Cl i n Biomech
1 : 205 , 1 986
1 0. Macintosh JE, Bogduk N: The morphology of the l umbar erector spinae. Spine 1 2:
658, 1 986
I I . Macintosh J E, Bogduk N: The attachments of the l umbar erector spinae. Spi ne 1 6:
783, 1 991
1 2 . Macintosh J E, Val encia F, Bogduk N, Munro RR: The morphol ogy of the l umbar
mul tifidus muscl es. Cl in Biomech I : 1 96, 1 986
1 3 . Bogduk N, Wi l son AS, Tynan W: The human l umbar dorsal rami . J Anat 1 34: 383,
1 982
1 4. Bogduk N: The l umbar mami l l o-accessory ligament. I t s anatomical and neuro­
surgical significance. Spine 6: 1 62, 1 98 1
1 5 . Abrahams VC: The physiology of neck muscl es; their role i n head movement and
maintenance of posture. Can J Physiol Pharmacol 55: 332, 1 977
1 6. Abrahams VC: Sensory and motor special ization i n some muscl es of the neck.
Trends Neuosci 4: 24, 1 98 1
1 7. Cooper S, Danial PM: Muscl es spindl es i n man, their morphol ogy i n the l umbricals
and the deep muscl es of the neck. Brain 86: 563 , 1 963
1 8. Bastide G, Zadeh J, Lefebvre D: Are the "l ittl e muscl es" what we think they are?
Surg Radiol Anat I I : 255, 1 989
1 9. Nitz AJ, Peck D: Comparison of muscl e spi ndl e concentrations i n large and smal l
human epaxi al muscl es acti ng in paral l el combinations. Am Surg 52: 273, 1 986
20. Peck D, Buxton DF, Nitz A: A comparison of spi ndl e concentrations i n large and
smal l muscl es acting i n paral l el combinations. J Morphol 1 80: 243, 1 984
1 44 Physical Therapy of the Low Back
2 1 . Lewi n T, Moffet B, Vi i di k A: The morphology of t he l umbar synovial i ntervertebral
joi nts. Acta Morphol Neere-Scand 4: 299, 1 962
22. Doni sch EW, Basmaj ian J V: El ectromyography of deep back muscl es i n man. Am
J Anat 1 33: 25, 1 972
23. Luk KDK, Ho HC, Leong JCY: The i l i ol umbar l igament . A study of i ts anatomy,
devel opment and cl i nical si gni ficance. J Bone Joint Surg 68B: 1 97, 1 986
24. Bogduk N , Macintosh J : The appl i ed anatomy of the thoracolumbar fasci a. Spine
9: 1 64, 1 984
25 . Fairbank JCT, O' Bri en J P: The abdomi nal cavi ty and thoracol umbar fascia as stabi­
l i sers of the l umbar spine in patients with low back pai n. Engineering Aspects of
the Spine 2: 83, 1 980
26. Gracovetsky S, Farfan HF, Lamy C: The mechani sm of the l umbar spi ne. Spi ne
6: 249, 1 98 1
27. Floyd WF, Si l ver PHS: The functi on of the erectores spi nae muscl es in certain
movements and postures in man. J Physiol 1 29: 1 84, 1 955
28. McNei l l T, Warwick D, Andersson G, Schultz A: Trunk strengths i n attempted
fexi on, extensi on and lateral bendi ng i n heal thy subjects and pati ents with low back
di sorders. Spine 5 : 529, 1 980
29. Ortengren R, Andersson GBJ: El ectromyographi c studi es of trunk muscl es wi th
speci al reference to t he functional anatomy of the l umbar spi ne. Spi ne 2: 44, 1 977
30. Morris J M, Lucas DB, Bresl er B: Role of t he trunk in stabi l i ty of the spi ne. J Bone
Joi nt Surg 43A: 327, 1 961
3 1 . Andersson G BJ , Ortengren R, Nachemson A: I ntradiscal pressure, i ntra-abdominal
pressure and myoel ectric back muscl e acti vi ty related to posture and loadi ng. Cl i n
Orthop 1 29: 1 56, 1 977
32. Carlsoo S: The static muscl e l oad i n di fferent work positi ons: an electromyographic
study. Ergonomi cs 4: 1 93 , 1 96 1
33. Ortengren R, Andersson G, Nachemson A: Lumbar loads i n fixed working postures
duri ng fexion and rotation. p. 1 59. In Asmussen E, Jorgensen K ( eds) : Biomecha­
nics VI - B, I nternati onal Series on Bi omechani cs , Vo! . 2B. Uni versity Park Press,
Bal ti more, 1 978
34. Portnoy H, Mori n F: El ectromyographic study of t he postural muscl es i n various
posi ti ons and movements. Am J Physiol 1 86: 1 22, 1 956
35. Al l en CEL: Muscl e action potenti al s used in the study of dynamic anatomy. Br J
Phys Med I I : 66, 1 948
36. Andersson BJG, Ortengren R: Myoelectric acti vi ty during si tti ng. Scand J Rehabil
Med, supp! . 3 : 73, 1 974
37. Andersson BJG, Jonsson B, Ortengren R: Myoel ectric acti vi ty i n i ndi vi dual l umbar
erector spi nae muscl es i n si tti ng: a study with surface and wi re el ectrodes. Scand
J Rehabi l Med, supp! . 3: 9 1 , 1 974
38. Asmussen E, Kl ausen K: Form and function of the erect human spi ne. Cl i n Orthop
25 : 55, 1 962
39. Carlsoo S: I nfluence of frontal and dorsal loads on muscl e acti vi ty and on the weight
di stribution in the feet . Acta Orthop Scand 34: 299, 1 964
40. De Vri es HA: Muscl e tonus i n postural muscl es. Am J Phys Med 44: 275 , 1 965
4 1 . Jonsson B: The functions of the i ndi vidual muscl es i n the l umbar part of the spinae
muscl e. El ectromyography 1 0: 5, 1 970
42. Joseph J , McCol l I : El ectromyography of muscl es of posture: posterior vertebral
muscl es in mal es. J Physi ol 1 57: 33, 1 961
43. Kl ausen K: The form and function of the loaded human spi ne. Acta Physiol Scand
65: 1 76, 1 965
The Lumbar Back Muscles and Their Fascia 1 45
44. Morris J M, Benner G, Lucas DB: An electro myographi c study of the i ntri nsi c mus­
cl es of the back i n man. J Anat 96: 509, 1 962
45 . Valencia FP, Munro RR: An el ectromyographic study of the l umbar mul t ifidus i n
man. El ectromyogr Cl i n Neurophysi ol 25: 205 , 1 985
46. Floyd WF, Si l ver PHS: Functi on of erectores spi nae in fl exion of the trunk. Lancet
I : 1 33 , 1 95 1
47. Kippers V, Parker A W: El ectromyographic studi es of erectores spi nae: symmetri cal
postures and sagittal trunk moti on. Aust J Physi other 3 1 : 95, 1 985
48. Andersson BJG, Ortengren R: Lumbar di sc pressure and myoel ectric back muscl e
acti vi ty during si tti ng. I I . Studi es of an office chai r. Scandi nav J Rehab Med 6: 1 1 5,
1 974
49. Andersson BJG, Ortengren R, Nachemson AL et al : The si tti ng posture: an el ectro­
myographic and di scometric study. Orthop Cl i n North Am 6: 1 05, 1 975
50. Nachemson AL: The l umbar spi ne. An orthopaedic chal l enge. Spi ne 1 : 59, 1 976
5 1 . Okada M: El ectromyographi c assessment of t he muscul ar load in forward bendi ng
postures. J Fac Sci Uni v Tokyo 8: 3 1 1 , 1 970
52. Golding JSR: El ectromyography of the erector spi nae i n l ow back pai n. Postgrad
Med J 28:40 I , 1 952
53. Koreska J, Robertson D, Mi l l s RH: Biomechani cs of t he l umbar spi ne and i ts cl i nical
significance. Orthop Cl i n North Am 8: 1 2 1 , 1 977
54. Pauly J E: An el ectromyographi c anal ysi s of certai n movements and exerci ses. I .
Some deep muscl es of t he back. Anat Rec 1 55: 223, 1 966
55. Andersson G BJ . Ortengren R, Herberts P: Quanti tat i ve el ectromyographic studi es
of back muscl e acti vi ty related to posture and loadi ng. Orthop Cl i n North Am 8:
85, 1 977
56. Schul z A, Andersson GBJ, Ortengren R et al : Anal ysi s and quanti tat i ve myoelectric
measurements of loads on the l umbar spi ne when hol di ng weights i n standi ng post­
rues. Spi ne 7: 390, 1 982
57. Ki ppers V. Parker A W: Posture related to myoelectric si l ence of erectores spi nae
during trunk flexi on. Spine 7: 740, 1 984
57. Andersson GBJ : Loads on the l umbar spi ne: in vi vo measurements and bi omechani ­
cal anal yses. p. 32. I n Wi nter DA, Norman RW, Wel l s RP et al ( eds) : Biomechani cs
I X- B, I nternational Seri es on Biomechanics . Human Ki neti cs, Champai gn, I L, 1 983
58. Ortengren R, Andersson GBJ, Nachemson AL: Studi es of rel ati onshi ps between
lumbar di sc pressure. myoelectric back muscle act i vi t y, and i ntra-abdomi nal ( i ntra­
gastric) pressure. Spi ne 6: 98, 1 98 1
59. Andersson GBJ , Ortengren R, Nachemson A: Quanti tati ve studi es of the back i n
di fferent working postures. Scand J Rehabi l Med, suppl . 6: 1 73, 1 978
60. Andersson BJG, Ortengren R, Nachemson A, El fstron G: Lumbar disc pressure
and myoelectric acti vi ty duri ng si tti ng. I . Studi es on an experi mental chair. Scand
J Rehabil Med 6: 1 04, 1 974
6 1 . Andersson BJG. Ortengren R, Nachemson A, El fstrom G: Lumbar di sc pressure
and myoelectric back muscl e acti vi ty duri ng si tti ng. I V. Studi es on a car dri ver' s
seat. Scand J Rehabi l Med 6: 1 28, 1 974
62. Nachemson A: The load on l umbar di sks in di fferent posi ti ons of the body. Cl i n
Orthop 45: 1 07, 1 966
63 . Nachemson AL. El fstron G: I ntravi tal dynami c pressure measurements i n l umbar
di scs. A study of common movements, maneuvers and exerci ses. Scand J Rehabi l
Med suppl . I : I , 1 970
64. Nachemson A, Morris J M: In vi vo measurements of i ntradi scal pressure. J Bone
Joi nt Surg 46: 1 077, 1 964
1 46 Physical Therapy of the Low Back
65 . Andersson GBJ, Ortengren R, Nachemson A: Quantitative studies of back loads in
lifting. Spi ne 1 : 1 78, 1 976
66. Bogduk N, Macintosh J E, Pearcy MJ : A uni versal model of the l umbar back muscl es
in the upright posture. Spine 1 7: 897, 1 992
67. Farfan HF: Muscul ar mechanism of the l umbar spine and the position of power and
effi ci ency. Orthop C1 i n North Am 6: 1 35, 1 975
68. Farfan HF: A reorientation i n the surgical approach to degenerative l umbar interver­
tebral joint disease. Orthop Cl i n North Am 8: 9, 1 977
69. Bartel i nk DL: The role of abdomi nal pressure i n rel i eving the pressure on the l umbar
intervertebral di scs. J Bone Joint Surg 39B: 7 1 8, 1 957
70. Davis PR: Posture of the trunk during the l ifing of weights. Br Med J 1 : 87, 1 959
7 1 . Davis PR: The use of i ntra-abdominal pressure in eval uating stresses on the l umbar
spi ne. Spine 6: 90, 1 98 1
72. Davis PR, Stubbs DA: Safe l evel s of manual forces for young mal es. Appl Ergon
8: 1 4 1 , 1 977
73. Davis PR, Troup J DG: Pressures i n the trunk cavities when pul l i ng, pushi ng and
lifting. Ergonomics 7: 465, 1 964
74. Troup JDG: Dynamic factors in the anal ysis of stoop and crouch lifting methods:
a methodol ogical approach to the devel opment of safe materials handl i ng standards.
Orthop Clin North Am 8: 20 1 , 1 977
75 . Troup JDG: Bi omechanics of the vertebral col umn. Physiotherapy 65 : 238, 1 979
76. Kumar S, Davis PR: Lumbar vertebral i nnervation and intra-abdominal pressure.
J Anat 1 1 4: 47, 1 973
77. Granhed H, Johnson R, Hansson T: The loads on the l umbar spine during extreme
weight l ift ing. Spine 1 2 : 1 46, 1 987
78. Leski nen TPJ, Stal hammar HR, Kuorinka I AA, Troup J DG: Hi p torque, l umbosa­
cral compression, and intraabdominal pressure in l ifting and lowering tasks. p. 55.
I n Wi nter DA, Norman RW, Wel l s RP et al ( eds) : Biomechanics I XB, I nternational
Series on Biomechanics. Human Ki netics, Champaign, IL 1 983
79. Nachemson AL, Andersson GBJ , Schul tz AB: Val sal va maneuver biomechanics.
Effects on trunk load of el evated i ntraabdominal pressure. Spine 1 1 :476, 1 986
80. Hemborg B, Mortiz U: I nt ra-abdominal pressure and trunk muscl e activity during
lifting I I : chronic low-back pati ents. Scand J Rehabil Med 1 7: 5, 1 985
8 1 . Hemborg B, Moritz U, Hamberg J et al : I ntra-abdominal pressure and trunk muscl e
activity during l ifti ng-effect of abdominal muscle training in heal thy subjects.
Scand J Rehabil Med 1 5 : 1 83, 1 983
82. Hemborg B, Moritz U, Hamberg J et al : I ntra-abdomi nal pressure and trunk muscl e
acti vity during lifting I l l : effects of abdominal muscl e training in chronic low-back
pati ents. Scand J Rehabil Med 1 7: 1 5, 1 985
83 . Farfan HF, Gracovetsky S: The abdominal mechani sm. Paper presented at the Inter­
national Society for the Study of the Lumbar Spine Meeting, Paris, May 1 6-20,
1 98 1
84. Farfan HF, Gracovetsky S , Hel l eur C: The role of mathematical models in the
assessment of task i n t he workpl ace. p. 38. I n Winter DA, Norman RW, Wel l s
RP et al ( eds) : Bi omechanics I XB, I nternational Series on Biomechanics. Human
Kinetics, Champaign, I L, 1 983
85. Gracovetsky S, Farfan HF, Hel l eur C: The abdominal mechanism. Spine 1 0: 31 7,
1 985
86. Bearn JG: The signifcance of the activity of the abdominal muscl es in weight lifting.
Acta Anat 45 : 83, 1 961
The Lumbar Back Muscles and Their Fascia 1 47
87. Farfan HF: The biomechanical advantage of l ordosis and hi p extension for upright
act i vi ty. Man as compared with other anthropoi ds. Spine 3 : 336, 1 978
88. Gracovetsky S, Farfan HF, Lamy C: A mathematical model of the l umbar spine
using an optimal system to control muscl es and l igaments . Orthop Cl in North Am
8: 1 35 , 1 977
89. Macintosh I E, Bogduk N, Gracovetsky S: The biomechanics of t he thoracol umbar
fascia. Cl in Biomech 2: 78, 1 987
90. McGil l SM, Norman RW: Potential of l umbodorsal fascia forces to generate back
extension moments during squat l i ft s. J Biomed Eng 1 0: 3 1 2, 1 988
9 1 . Crock HV, Crock MC: A technique for decompression of the l umbar spinal canal .
Neuro-orthopaedics 5: 96, 1 988
92. Bogduk N, Twomey LT: The l umbar muscl es and thei r fasci a. pp. 83-1 05. I n:
Cl inical Anatomy of t he Lumbar Spine. 2nd Ed. Churchi l l Livingstone, Mel bourne,
1 99 1
93. Mykelbust I B, Pintar F, Yoganandan N e t al : Tensile strength of spinal l igaments.
Spine 1 3 : 526, 1 988
94. Cyron BM, Hutton WC: The tensi l e strength of the capsular ligaments of the apophy­
seal joints. J Anat 1 32: 1 45, 1 98 1
95 . Huki ns OWL, Aspden RM, Hickey DS: Thoracol umbar fascia can i ncrease t he
effciency of the erector spinae muscl es. Cl i n Bi omech 5 : 30, 1 990
96. Aspden RM: I ntra-abdominal pressure and its role in spinal mechanics. Cl i n Bio­
mech 2: 1 68, 1 987
97. Aspden RM: The spine as an arch. A new mathematical model . Spine 1 4: 266, 1 989
98. Adams M: Letter to the editor. Spine 1 4: 1 272, 1 989
5 The Maitland Concept:
Examination, and
Treatment by Passive
Geoffrey D. Maitland
It would be difficult for me, as one who has been involved in the practice
of manipulative physical therapy in Australia for the past four decades, to objec­
tively assess my particular contribution to the discipline. I therefore begin this
chapter, by way of explanation and justification, with a relevant and pertinent
quotation from Lance Twomey, who asked me to write it.
In my view, the Maitland approach to treatment differs from others, not in
the mechanics of the technique, but rather in its approach to the patient and
his particular problem. Your attention to detail in examination, treatment,
and response is unique in physical therapy, and I believe is worth spelling
out in some detail:
The development of your concepts of assessment and treatment
Your insistence on sound foundations of basic biologic knowledge
The necessity for high levels of skill
The evolution of the concepts: it did not "come" to you fully developed,
but is a living thing, developing and extending
The necessity for detailed examination and for the examination/treatment/
re-examination approach
150 Physical Therapy of the LolV Back
(This last area is well worth very considerable attention because, to me,
it is the essence of "Maitland". )
Although the text of this chapter deals with "passive movement," it must
be very clearly understood that I do not believe that passive movement is the
only form of treatment that will alleviate neuromusculoskeletal disorders. What
the chapter does set out to do is to provide a conceptual framework for treat­
ment, which is considered by many to be unique. Thus, for want of a better
expression, the particular approach to assessment, examination, and treatment
outlined in this paper is described as "the Maitland concept," and referred to
hereafter as "the concept."
To portray all aspects of the concept by the written word alone is difficult
because so much of it depends on a particular clinical pattern of reasoning.
The approach is methodical and involved, and therefore difficult to describe
adequately without clinical demonstration. The Maitland concept reqllires
open-mindedness, mental agility, and mental discipline linked with a logical
and methodical process of assessing cause and effect. The central theme de­
mands a positive personal commitment (empathy) to IInderstand what the per­
son (patient) is enduring. The key issues of the concept that require explanation
are personal commitment, mode of thinking, techniques, examination, and as­
All clinicians would claim that they have a high level of personal commit­
ment to every patient. True as that may be, many areas of physical therapy
require that a deeper commitment to certain therapeutic concepts be developed
than is usual. Thus the therapist must have a personal commitment to care,
reassure, communicate, listen and believe, and inspire confidence.
All therapists must make a conscious effort (particularly during the first
consultation), to gain the patient's confidence, trust, and relaxed comfort in
what may be at first an anxious experience. The achievement of this trusting
relationship requires many skills, but it is essential if proper care is to be pro­
Within the frst few minutes, we clinicians must make the patient believe
that we want to know what the patient feels; not what the doctor or anyone
else feels, but what the patient feels is the main issue. This approach immedi­
ately puts patients at ease by showing that we are concered about their symp­
toms and the effect they are having.
We must use the patient's terminology in our discussions; we must adapt
our language (and jargon) to that of the patient. We must make our concer
for the patient's symptoms show in a way that matches the patient's feelings
about the symptoms. In other words, we should adapt our approach to match
the patient's mode of expression, not make or expect the patient to adapt to
our personality and our knowledge. The patient also needs to be reassllred of
the belief and understanding of the therapist.
The Maitland Concept 151
Commllnication is another skill that clinicians must learn to use effectively
and appropriately. As far as personal commitment is concerned, this involves
understanding the nonverbal as well as the verbal aspects of communication,
so that it can be used to further enhance the relationship between patient and
clinician. Some people find that this is a very difficult skill to acquire, but
however much effort is required to learn it, it must be learned and used.
Listening to the patient must be done in an open-minded and nonjudgmental
manner. It is most important to accept the story the patient weaves, while at
the same time being prepared to question it closely. Accepting and listening
are very demanding skills, requiring a high level of objectivity.
It is a very sad thing to hear patients say that their doctor or physical
therapist does not listen to them carefully enough or with enough sympathy,
sensitivity, or attention to detail. The following quotation from The Age, J an
Australian daily newspaper, sets out the demands of "listening" very clearly:
Listening is itself, of course, an art: that is where it differs from merely
hearing. Hearing is passive; listening is active. Hearing is involuntary; listen­
ing demands attention. Hearing is natural; listening is an acquired discipline.
Acceptance of the patient and his or her story is essential if trust between
patient and clinician is to be established. We must accept and note the subtleties
of the patient's comments about a disorder even if they may sound peculiar.
Expressed in another way, the patient and his or her symptoms are "innocent
until proven guilty" (i . e. , the patient's report is true and reliable until found
to be unreliable, biased or false). In this context, the patient needs to be guided
to understand that the body can tell things about a disorder and its behavior
that we (the clinicians) cannot know unless the patient expresses them. This
relationship should inspire confidence and build up trust between both parties.
This central core of the concept of total commitment must begin at the
outset of the first consultation and carry through to the end of the total treatment
Other important aspects of communication are discussed under Examina­
tion and Assessment.
As qualified physical therapists, we have absorbed much scientific informa­
tion and gained a great deal of clinical experience, both of which are essential
for providing effective treatment. The "science" of our discipline enables us
to make diagnoses and apply the appropriate "art" of our physical skill. How­
ever, the accepted theoretical basis of our profession is continually developing
and changing. The gospel of yesterday becomes the heresy of tomorrow. It is
essential that we remain open to new knowledge and open-minded in areas of
uncertainty, so that inflexibility and tunnel vision do not result in a misapplica-
[52 Physical Therapy of the Low Back
tion of our "art. " Even with properly attested science applied in its right con­
text, with precise information concerning the patient's history of disorder,
symptoms, and signs, a correct diagnosis is often difficult. Matching of the
clinical findings to particular theories of anatomic, biomechanical, and patho­
logic knowledge, so as to attach a particular "label" to the patient's condition,
may not always be appropriate. Therapists must remain open-minded so that
as treatment progresses, the patient is reassessed in relation to the evolution
of the condition and the responses to treatment.
In summary, the scientific basis underlying the current range of diagnoses
of disorders of the spine is incompletely understood. It is also changing rapidly
with advances in knowledge and will continue to do so. In this context, the
therapist may be sure of the clinical evidence from the patient's history and
clinical signs, but should beware of the temptation to "ft the diagnosis" to
the infexible and incomplete list of options currently available. The physical
therapist must remain open-minded, not only at the initial consultation, but
also as noting the changing responses of the patient during assessment and
treatment. When the therapist is working in a relatively "uncharted area" like
human spinal disorders, one should not be influenced too much by the unreliable
mass of inadequately understood biomechanics, symptomatology, and pa­
This section, in essence, is the most important aspect of the concept that
makes it different from that which others teach. It leaves one's mind totally
open and not blocked by the theoretical compartment, the medical diagnostic
titles, and the rigid philosophies of some manipulators (both medical and lay).
Consequently, I have drawn up a list of practical steps to follow. In the
early era of its evolution, the Maitland concept had as its basis the following
stages within a treatment:
I. Having assessed the effect of a patient's disorder, perform a single treat­
ment technique
2. Take careful note of what happens during the performance of the tech­
3. Having completed the technique, assess the effect of the technique on
the patient's symptoms including movements
4. Having completed steps 2 and 3, and taken into account the available
theoretical knowledge, plan the next treatment approach and repeat the cycle
from step 1 again
Table Sol. One Diagnosis with Many Presentations
Diagnosis: disc herniation
H, (history); Sy, (symptoms); S, (signs)
H3; SY3: S3
The Maitland Concept 153
Table 5-2. Different Dia
noses for One Set of Symptoms and Signs
Theory Clinical
Diagnosis I �======�������_
Diagnosis 2 _
Diagnosis 3
nosis 4
H; Sy; S
Ahhrel'ialiolls: H. history; Sy, symptoms: S. signs.
It becomes obvious that this sequence can only be useful and informative
ifboth the clinical history taking and physical examinations have been accurate.
The actual pattern of the concept requires us to keep our thoughts in two
separate but interdependent compartments: the theoretical framework and the
clinical assessment. An example may help to clarify these concepts. We know
that a lumbar intervertebral disc can herniate and cause pain, which can be
referred into the leg. However, there are many presentations that can result
from such a herniation (Table 5-1).
The reverse is also true: a patient may have one set of symptoms for which
more than one diagnostic title can be applied2 (Table 5-2).
Because of the circumstances shown in Tables 5-1 and 5-2, it is obvious
that it is not always possible to have a precise diagnosis for every patient
treated. The more accurate and complete our theoretical framework, the more
appropriate will be our treatment. If the theoretical framework is faulty or
deficient (as most are admitted to be), a full and accurate understanding of
the patient's disorder may be impossible. The therapist's humility and open­
mindedness are therefore essential, and inappropriate diagnostic labels must
not be attached to a patient prematurely. The theoretical and clinical compo­
nents must, however, influence one another. With this in mind, I have devel­
oped an approach separating theoretical knowledge from clinical information
by what I have called the symbolic, permeable brick wall (Table 5-3). This
serves to separate theory and practice, and to allow each to occupy (although
not exclusively) its own compartment. That is, information from one side is able
to filter through to the other side. In this way, theoretical concepts influence
Table 5-3. Symbolic, Permeable Brick Wall
Theory Clinical
H: Sy; S
Abbrevialiolls: H. history; Sy, symptoms; S, signs.
154 Physical Therapy of the Low Back
examination and treatment, while examination and treatment lead one back to
a reconsideration of theoretical premises.
Using this mode of thinking, the brick-wall concept frees the clinician's
mind from prejudice, allowing the therapist to ponder the possible reasons for
a patient's disorder; to speculate, consider a hypothesis, and discuss with others
the possibilities regarding other diagnoses without anyone really knowing all
the answers, yet all having clear understanding of the patient's symptoms and
related signs (Fig. 5-1).
This mode of thinking requires the use of accurate language, whereas inac­
curate use of words betrays faulty logic. The way in which an individual makes
a statement provides the listener with an idea both of the way that person is
thinking and of the frame of reference for the statement.
A simple example may help to make this point clear. Imagine a clinician
presenting a patient at a clinical seminar, and on request the patient demon­
strates his area of pain. During the ensuing discussion, the clinician may refer
to the patient's pain as "sacroiliac pain." This is a wrong choice of words. To be
true to concept we have outlined, of keeping clinical information and theoretical
interpretations separate, one should describe the pain simply as a "pain in the
sacroiliac area." It would be an unjustified assumption to suggest that pathology
in the sacroiliac joint was the source of pain, but the former description above
could be interpreted in this way. On the other hand, describing the pain as "in
the sacroiliac area" indicates that we are considering other possible sites of
origin for the pain besides the sacroiliac joints, thereby keeping our diagnostic
options open until we have more evidence. This is an essential element to
the concept. Some readers may believe that attention to this kind of detail is
unnecessary and pedantic. Quite the opposite is true. The correct and careful
choice of words indicates a discipline of mind and an absence of prejudice,
which influences all our diagnostic procedures including the whole process of
examination, treatment, and interpretation of the patient's response.
; Ph
Biomechanics; Patholog
Facts and "Impressions�
Fig.5-1. Flow chart demonstrating relationships and contexts for theoretical and clini­
cal knowledge with related hypotheses. H, history; Sy, symptoms; S, signs.
The Maitland Concept 155
A clinician's written record of a patient's examination and treatment find­
ings also shows clearly whether the therapist's thinking processes are right or
wrong. A genuine scientific approach involves logical thinking, vertical and
lateral thinking, and inductive and deductive reasoning. It requires a mind that
is uncluttered by confused and unproven theory, which is at the same time able
to use proven facts, and has the critical ability to distinguish between well­
attested facts and unsubstantiated opinions. It requires a mind that is honest,
methodical, and self-critical. It also requires a mind that has the widest possible
scope in the areas of improvisation and innovation.
Many physical therapy clinicians are continually seeking new techniques
of joint mobilization. When they hear a new name or when a new author has
written a book on manipulation, they attempt to acquire the "new" technical
skills, and immediately apply them. In reality, the techniques are of secondary
importance. Of course, if they are poorly performed or misapplied, treatment
may fail and the therapist may lose confidence in the techniques. However,
there are many acceptable techniques each of which can be modified to suit a
patient's disorder and the clinician's style and physique. Accordingly, there is
no absolute set of techniques that can belong or be attributed to any one person.
There should be no limit to the selection of technique: the biomechanically
based techniques of Kaltenbor; the "shift" techniques of McKenzie; the com­
bined-movements technique of Edwards; the neural techniques of Elvey and
Butler; the osteopathic and chiropractic technique; the Cyriax techniques; the
Stoddard technique; the bonesetters' techniques; the Maigne techniques; and
the Mennell techniques. All of these techniques are of the present era. Every
experienced practitioner must feel totally free to make use of any of them. The
most important consideration is that the technique chosen be appropriate to
the particular patient or situation, and that its effect should be carefully and
continually assessed.
Within the broad concept of this chapter, there are certain techniques of
management that are continually used, but are not described by other authors.
These techniques are as follows.
When treating patients having very painful disorders, passive-treatment
movements can be used in an oscillatory fashion ("sUce stirring" as de­
scribed by Maitland
), but with two important provisos: (I) the oscillatory
movement is performed without the patient experiencing any pain whatsoever,
or even any discomfort; and (2) the movement is performed only in that part
156 Physical Therapy of the Low Back
of the range of movement where there is no resistance (i .e., where there is no
stiffness or muscle spasm restricting the oscillations).
One may question how a pain free oscillatory movement, which avoids
every attempt to stretch structures, can produce any improvement in a patient's
symptoms. A scientific answer to this question has been suggested
but there
is a far more important clinical conclusion. It has been repeatedly shown clini­
cally that such a technique does consistently produce a measurable improve­
ment in range of movement with reduction in pain and disability and no demon­
strable harmful effects. This demonstrates that the treatment is clinically, and
therefore ' ' scientifcally, " correct even though an adequate theoretical explana­
tion for its effectiveness may not yet be available. Reliable and repeated demon­
stration of effectiveness must validate a treatment method. To know how the
method achieves the result is a theoretical problem for science to solve. The
"scientific" examination must match the primary clinical observation, the latter
being the aspect of which we can be sure.
This example demonstrates once more how this mode of thinking, so essen­
tial to the concept, is so necessary for the further development of treatment
methods. Without this mode of thinking we would never have found that pas­
sive-movement treatment procedures can successfully promote union in non­
uniting fractures.
Oscillatory movements as an important component of passive movement
are referred to above in relation to the treatment of pain. There is another
treatment procedure that requires oscillatory movement to be effective. This
is related to the intermittent stretching of ligamentous and capsular structures.
There are clearly defined areas of application for this treatment, which are
described elsewhere.
There are occasions when a passive treatment movement needs to be per­
formed with the opposing joint surfaces compressed together. 6 Without the
compression component, the technique would fail to produce any improvement
in the patient's symptoms.
Using the movements and positions by which a patient is able to reproduce
his symptoms as an initial mandatory test is essential to the concept. This
tactic, like the formalized examination of combined movements (the original
contribution in cooperation with Edwards
) is very special to the concept.
Although it is frequently recognized that straight-leg raising can be used
as a treatment technique for low lumbar disorders, it is not widely appreciated
that the technique can be made more effective by using straight-leg raising in
the "slump test" position.
In the same slumped position, the neck flexion
component of the position can be effectively used when such movement repro­
duces a patient's low back pain.
Sometimes a patient is able to guide the therapist as to what to do because
his body tells him what it wants (and what it does not want). Figure 5-2 is a
perfect example of such circumstances. The disorder had been very difficult
to treat because progress gained at a treatment would not be retained well
enough. The disorder was at the level of T6-T7 and had been responding to
extremely gentle traction. Then one day the patient came in saying that he
needed the traction but he also needed to have the vertebra pushed backward
The Maitland Concept 157
Fi�. 5-2. Thoracic mobilization via the sternum.
and toward the left while having the levels above being twisted to the right.
Figure 5-2 shows how the position was obtained while the mobilizing was pro­
duced through his sternum. He claimed he was 60 percent better after the
first such treatment, and 80 percent better after the second. At his suggestion
treatment was discontinued and on review 12 months later he showed no signs
of slipping back.
"Accessory" movements produced by applying alternating pressure on
palpable parts of the vertebrae are also very important in terms of techniques
and the Maitland concept. Any treatment concept that does not include such
techniques is missing a critical link essential to a full understanding of the effects
of manipulation on patients with low lumbar disorders.
It is important to remember that there is no dogma or clear set of rules
that can be applied to the selection and use of passive-movement techniques;
the choice is open-ended. A technique is the brainchild of ingenuity. "The
achievements are limited to the extent of one's lateral and logical thinking"
(K. Hunkin, unpublished data, 1985).
History Taking
The care, precision, and scope of examination required by those using this
concept are greater and more demanding than other clinical methods I have
observed. The concept's demands differ from those of other methods in many
158 Physical Therapy of the Low Back
The history taking and examination demand a total commitment to under­
standing what the patient is suffering and the effects of the pain and disability
on the patient. Naturally, one is also continually attempting to understand the
cause of the disorder (the theoretical compartment of the concept).
Examination must include a sensitive elucidation of the person's symptoms
in relation to (I) precise area(s) indicated on the sUlface of the body; (2) the
depth at which symptoms are experienced; (3) whether there is more than one
site of pain, or whether multiple sites overlap or are separate; and (4) changes
in the symptoms in response to movements or differences in joint positions in
different regions of the body.
The next important and unique part of the examination is for the patient
to re-enact the movement that best reveals the disorder or, if applicable, to re­
enact the movement that produced the injury. The function or movement is
then analyzed by breaking it into components in order to make clinical sense
of particular joint-movement pain responses, which are applicable to the com­
The routine examination of physiologic movements is performed with a
degree of precision rarely used by other practitioners. If the person's disorder
is an "end-of-range" type of problem, the details of the movement examination
required are as follows:
I. At which point in the range are the symptoms first experienced; how
do they vary with continuation of the movement; and in what manner do the
symptoms behave during the symptomatic range?
2. In the same way and with the same degree of precision, how does muscle
spasm or resistance vary during the symptomatic range?
3. Finally, what is the relationship of the symptom to the resistance or
spasm and during that same movement? There may be no relationship whatso­
ever, in which case the stiffness is relatively unimportant. However, if the
behavior of the symptoms matches the behavior of the stiffness, both should
improve in parallel during treatment.
An effective method of recording the findings of all components of a move­
ment disorder is to depict them in a "movement diagram." These also are
an innovative part of the concept. The use of movement diagrams facilitates
demonstration of changes in the patient's condition in a more precise and objec­
tive manner. They are discussed at length in the fifth edition of Vertebral Manip­
If the patient's disorder is a "pain through range" type of problem, the
details of the movement examination required are as follows:
I. At what point in the range does discomfort or pain first increase?
2. How do the symptoms behave if movement is taken a short distance
beyond the onset of discomfort? Does intensity markedly increase or is the
area of referred pain extended?
3. Is the movement a normal physiologic movement in the available range
The Maitland Concept 159
or is it protected by muscle spasm or stiffness? Opposing the abnormal move­
ment and noting any change in the symptomatic response compared with entry
2 is performed to assess its relevance to the course of treatment.
Palpatory Techniques
The accessory movements are tested by palpation techniques and seek the
same amount and type of information as described above. They are tested in
a variety of different joint positions. The three main positions are: (I) the neutral
mid-range position for each available movement (i.e. , midway between flexion/
extension, rotation left and right, lateral flexion left and right, and distraction/
compression; (2) the joint in a "loose-packed position,"
at the particular posi­
tion where the person's symptoms begin, or begin to increase; and (3) position
at the limits of the available range.
These palpatory techniques of examination and treatment have been pecul­
iar to this concept from its beginnings. As well as seeking symptomatic re­
sponses to the movement as described above, the palpation is also used to
assess positional anomalies and soft-tissue abnormalities, which are at least as
critical to the concept as the movement tests.
The testing of physiologic and accessory movement can be combined in a
variety of ways in an endeavor to find the comparable movement sign most
closely related to the person's disorder. Edwards
originally described a
formal method of investigating symptomatic responses and treating appropriate
patients using "combined movement" techniques. In addition, joint surfaces
may be compressed, both as a prolonged, sustained firm pressure, and as an
adjunct to physiologic and accessory movement. These are two further exam­
ples of examination developed as part of the Maitland concept.
Differentiation tests are perfect examples of physical examination proce­
dures that demonstrate the mode of thinking so basic to the Maitland concept.
When any group of movements reproduces symptoms, the concept requires a
logical and thoughtful analysis to establish which movement of which joint is
affected. The simplest example of this is passive supination of the hand and
forearm, which when held in a stretched position, reproduces the patient's
symptoms. The stages of this test are as follows:
I. Hold the fully supinated hand/forearm in the position that is known to
reproduce the pain.
2. Hold the hand stationary and pronate the distal radioulnar joint 2°
or 3°.
3. If the pain arises from the wrist, the pain will increase because in pronat­
ing the distal radioulnar joint, added supination stress is applied at the radiocar­
pal and midcarpal joints.
4. While in the position listed in stage I, again hold the hand stationary,
but this time increase supination of the distal radioulnar joint. This decreases
the supination st,
etch at the wrist joints and will reduce any pain arising from
1 60 Physical Therapy of the Low Back
the wrist. However, if the distal radioulnar joint is the source of pain, the
increased supination stretch will cause the pain to increase.
All types of differentiation tests require the same logically ordered proce­
dure. These objective tests follow the same logic as the subjective modes of
assessment described at the beginning of this chapter, and provide additional
evidence leading to accurate diagnosis.
In the last few years it would appear that physical therapists have discov­
ered a new' 'skill," with the lofty title of' 'problem solving." This is, and always
should be, the key part of all physical therapy treatment. Being able to solve
the diagnostic and therapeutic problems and thus relieve the patient of the
complaint is just what physical therapists are trained to do. For many years,
manipUlative physical therapy has been rightly classified as empirical treatment.
However, since manipUlative physical therapists began to be more strongly
involved in problem-solving skills, treatment has become less empirical and
more logical. On the basis that the pathology remains unknown in the majority
of cases and the effects of the treatment on the tissues (as opposed to symptoms)
are unknown, the treatment remains empirical in form. This is true with almost
all of the medical science. Nevertheless, the approach to the patient and to
physical treatment has become more logical and scientific within the Maitland
Minds existed before computers were developed, and manipulative thera­
pists are trained to sort out and access "input" so that appropriate and logical
"output" can be produced. Appropriate problem-solving logic will relate clini­
cal findings to pathology and mechanical disorders. We have called this process
of "sorting out" assessment. Assessment is the key to successful, appropriate,
manipulative treatment, which, because of the reliability of its careful and logi­
cal approach, should lead to better treatment for our patients.
Assessment is used in six different situations: (I) analytical assessment at
a first consultation; (2) pretreatment assessment; (3) reassessment during every
treatment session proving the efficacy of a technique at a particular stage of
treatment; (4) progressive assessment; (5) retrospective assessment; (6) final
analytical assessment.
Analytical Assessment
A first consultation requires skills in many areas, but the goals require
decisions and judgments from the following five areas: (I) the diagnosis; (2)
the phase of the disorder; (3) the degree of stability and irritability of the disorder
at the time of treatment; (4) the presenting symptoms and signs; (5) the charac­
teristics of the person.
The Maitland Concept 161
The answers to the different assessment procedures (\ to 5) cannot be
reliably determined without communication and an atmosphere of trust. By
using one's own frame of reference, and endeavoring to understand the patient's
frame of reference, the characteristics of the patient can be judged. By making
use of nonverbal skills, picking out key words or phrases, knowing what type
of information to listen for, and recognizing and using "immediate-automatic
response" questions (all described later), accurate information can be gained at
this first consultation. The physical examination is discussed in the Examination
Pretreatment Assessment
At the beginning of each treatment session a specific kind of assessment
is made of the effect of the previous session on the patient's disorder, its symp­
toms, and changes in movement. Since the first consultation includes both
examination of movements and treatment of movements, the assessment at the
second treatment session will not be as useful for therapy as it will be at the
following treatment sessions.
When the patient attends subsequent treatment sessions, it is necessary
to make both sUbjective and physical assessments (i.e., sUbjective in terms of
how they feel; physical in terms of what changes can be found in quality and
range of movement, and in related pain response). When dealing with the sub­
jective side of assessment, it is important to seek spontaneous comment. It is
wrong to ask, "How did it feel this morning when you got out of bed, compared
with how it used to feel?'" The start should be "How have you been?" or some
such general question, allowing the patient to provide some information that
seems most important to him. This information may be more valuable because
of its spontaneous nature.
Another important aspect of the subjective assessment is that statements
of fact made by a patient must always be converted to comparisons to previous
statements. Having made the subjective assessment, the comparative statement
should be the first item recorded on the patient's case notes. And it must be
recorded as a comparison-quotation of his opinion of the effect of treatment.
(The second record in the case notes is the comparative changes determined
by the physical movement tests). To attain this subjective assessment, commu­
nication skills are of paramount importance. There are many components that
make up the skill, but two are of particular importance:
I. Key words or key phrases: Having been asked the question, "How has
it been?" a patient may respond in a very general and uninformative way.
However, during his statements he may include, for example, the word Mon­
day. Latch on to Monday, because Monday meant something to him. Find out
what it was and use it. "What is it that happened on Monday? Why did you
say Monday?"
2. The immediate-automatic response: A patient frequently says things
162 Physical Therapy of the Low Back
that demand an immediate-automatic response question. As a response to the
opening question given above, the patient may respond by saying, "J am feeling
better." The immediate-automatic response to that statement, even before he
has had a chance to take a breath and say anything else, is "Better than what?"
or "Better than when?" It may be that after treatment he was worse and that
he is better than he was then, but that he is not better than he was before the
One aspect of the previous treatment is that it (often intentionally) provokes
a degree of discomfort. This will produce soreness, but if the patient says he
has more pain, the clinician needs to determine if it is treatment soreness or
disorder soreness. For example, a patient may have pain radiating across his
lower back and treatment involves pushing on his lumbar spine. He is asked
to stand up and is asked, "How do you feel now compared with before I was
pushing on your back?" He may say, "it feels pretty sore." He is then asked,
"Where does it feel sore?" If he answers, "It's sore in the center," the clinician
may consider that it is likely to be treatment soreness. But if he answers, "It's
sore across my back," then the clinician may conclude that it is disorder sore­
ness. If it were treatment soreness it would only be felt where the pressure had
been applied. If the soreness spreads across his back, the treatment technique
must have disturbed the disorder.
In making subjective assessments, a process is included of educating the
patient in how to refect. If a patient is a very good witness, the answers to
questions are very clear, but if the patient is not a good witness, then subjective
assessment becomes difficult. Patients should learn to understand what the
clinician needs to know. At the end of the first consultation, patients need to
be instructed in how important it is for them to take notice of any changes in
their symptoms. They should report all changes, even ones they believe are
trivial. The clinician should explain, "Nothing is too trivial. You can't tell me
too much; if you leave out observations that you believe to be unimportant,
this may cause me to make wrong treatment judgments." People need to be
reassured that they are not complaining, they are informing. Under circum­
stances when a patient will not be seen for some days or if full and apparently
trivial detail is needed, they should be asked to write down the details. The
criticism that is made of asking patients to write things down is that they become
hypochondriacs. This is a wrong assessment in my experience, because the
exercise provides information that might otherwise never be obtained.
There are four specific times when changes in the patient's symptoms can
indicate the effect of treatment. They are as follows:
I. Immediately after treatment: The question can be asked, "How did you
feel when you walked out of here last time compared with when you walked in?"
A patient can feel much improved immediately after treatment yet experience
exacerbation of symptoms 1 or 2 hours later. Any improvement that does not
last longer than I hour indicates that the effect of the treatment was only pallia-
The Maitland Concept 163
tive. Improvement that lasts more than 4 hours indicates a change related to
2. Four hours after treatment: The time interval of 4 hours is an arbitrary
time and could be any time from 3 to 6 hours. It is a "threshold" time interval
beyond which any improvement or examination can be taken to indicate the
success or failure of the treatment. Similarly, if a patient's syndrome is exacer­
bated by treatment, the patient will be aware of it at about this time.
3. The evening of the treatment: The evening of the day of treatment pro­
vides information in regard to how well any improvement from treatment has
been sustained. Similarly, an exacerbation immediately following treatment
may have further increased by evening. This is unfavorable. Conversely, if the
exacerbation has decreased, it is then necessary to know whether it decreased
to its pretreatment level or decreased to a level that was better than before that
day's treatment. This would be a very favorable response, clearly showing that
the treatment had alleviated the original disorder.
4. On rising the next morning: This is probably the most informative time
of all for signaling a general improvement. A patient may have no noticeable
change in his symptoms on the day or night of the treatment session, but may
notice that on getting out of bed the next morning that the usual lower back
stifness and pain are less, or that they may pass more quickly than usual. Even
at this time span, any changes can be attributed to treatment. However, changes
that are noticed during the day after treatment, or on getting out of bed the
second morning after treatment, are far less likely to be a result of treatment.
Nevertheless, the patient should be questioned in depth to ascertain what rea­
sons exist, other than treatment, to which the changes might be attributed.
Because accurate assessment is so vitally and closely related to treatment
response, each treatment session must be organized in such a way that the
assessments are not confused by changes in treatment. For example, if a patient
has a disorder that is proving very difcult to treat, and at the eighth treatment
session he reports that he feels there may have been some slight favorable
change from the last treatment, the clinician has no alternative in planning the
eighth treatment session. In the eighth treatment, that which was done at the
seventh must be repeated in exactly the same manner in every respect. To do
otherwise could render the assessment at the ninth treatment confusing. If the
seventh treatment is repeated at the eighth session, there is nothing that the
patient can say or demonstrate that can confuse the effect attributable to that
treatment. If there was an improvement between the seventh and eighth treat­
ment (and the eighth treatment was an identical repetition of the seventh treat­
ment), yet no improvement between the eighth and the ninth treatment, the
improvement between treatments seven and eight could not have been due to
There is another instance when the clinician must recognize that there can
be no choice as to what the eighth treatment must be. If there had been no
improvement with the first six treatments, and at the seventh treatment session
a totally new technique was used, the patient may report at the eighth session
\64 Physical Therapy of the Low Back
that there had been a surprisingly marked improvement in symptoms. It may
be that this unexpected improvement was due to treatment or it may have been
due to some other unknown reason. There is only one way that the answer can
be found-the treatment session should consist of no treatment techniques at
all. Physical assessment may be made but no treatment techniques should be
performed. At the ninth session, if the patient's symptoms have worsened con­
siderably, the treatment cannot be implicated in the cause because none had
been administered. The clinician can then repeat the seventh treatment and see
if the dramatic improvement is achieved again. If it is, then the improvement
is highly likely to have been due to that treatment.
Whatever is done at one treatment session is done in such a way that when
the patient comes back the next time, the assessment cannot be confusing.
Another example of a different kind is that a patient may say at each treat­
ment session that he is "the same," yet assessment of his movement signs
indicates that they are improving in a satisfactory manner, and therefore that
one would expect an improvement in his symptoms. To clarify this discrepancy,
specific questions must be asked. It may be that he considers he is "the same"
because his back is still just as stiff and painful on first getting out of bed in
the morning as it was at the outset of treatment. The specific questioning may
divulge that he now has no problems with sitting, and that he can now walk
up and down the stairs at work without pain. Although his sitting, climbing,
and descending stairs have improved, his symptoms on getting out of bed are
the same, and this explains his statement of being "the same." The physical
movement tests will have improved in parallel with his sitting and stair-climbing
Assessment During Every Treatment Session
Proving the value or failure of a technique applied through a treatment
session is imperative. Assessment (problem solving) should be part of all as­
pects of physical therapy. In this chapter it is related to passive movement.
There are four kinds of assessment, and probably the one that most people
think of first is the one in which the clinician is trying to prove the value of a
technique that is being performed on a patient.
Proving the Value of a Technique
Before even choosing the technique to be used it is necessary to know
what symptoms the patient has and how his movements are affected in terms of
both range and the pain response during the movement. Selection of a treatment
technique depends partly on knowing what that technique should achieve while
it is being performed. In other words, is it the aim to provoke discomfort and,
if so, how much "hurt" is permissible? It is also necessary to have an expecta­
tion of what the technique should achieve afer it has been performed.
The Maitland Concept 1 65
With these considerations in mind, it is necessary to keep modifying the
treatment technique until it achieves the expected goal during its performance.
Assuming that this is achieved and that the technique has been performed for
the necessary length of time, the patient is then asked to stand, during which
time he is watched to see if there are any nuances that may provide a clue as
to how his back is feeling. The first thing to ask him is, "How do you feel now
compared with when you were standing there before the technique?" It is then
necessary to clarify any doubts concerning the interpretation of what he says
he is feeling. It is important to understand what the patient means to say if the
sUbjective effect of the technique is to be determined usefully.
Having subjectively assessed the effect of the technique, it is then neces­
sary to re-examine the major movements that were faulty, to compare them
with their state before the technique. An important aspect of checking and
rechecking the movements is that there may be more than one component to
the patient's problems. For example, a patient may have back pain, hip pain,
and vertebral-canal pain. Each of these may contribute to the symptoms in his
lower leg. On reassessing him after a technique, it is necessary to assess at
least one separate movement for each of the components, so it can be deter­
mined what the technique has achieved for each component. It is still necessary
to check all of the components even if it is expected that a change will only be
effected in one of the components. Having completed all of these comparison
assessments, the effect of that technique at that particular stage of the disorder
is now recorded in detail.
Progressive Assessment
At each treatment session the symptoms and signs are assessed for changes
for their relation to the previous treatment session and to "extracurricular"
activities. At about each fourth treatment session a subjective assessment is
made, comparing how the patient feels today with how he felt four treatments
previously. The purpose of this progressive assessment is to clarify and confrm
the treatment by assessment of the treatment response. One is often surprised
by the patient's reply to a question, " How do you feel now compared with 1 0
days (i.e., four treatments) ago?" The goal is to keep the treatment-by-treat­
ments assessment in the right perspective in relation to the patient's original
Retrospective Assessment
The first kind of retrospective assessment is that made routinely at each
group of three or four treatment sessions when the patient's symptoms and
signs are compared with before treatment began, as described above.
A second kind of retrospective assessment is made toward the end of treat­
ment when the considerations relate to a final assessment. This means that the
1 66 Physical Therapy of the Low Back
clinician is determining ( I) whether treatment should be continued; (2) whether
spontaneous recovery is occurring; (3) whether other medical treatments or
investigations are required; (4) whether medical components of the disorder
are preventing full recovery; (5) what the patient's future in terms of prognosis
is likely to be.
A third kind of retrospective assessment is made when the patient's disor­
der has not continued to improve over the last few treatment sessions. Under
these circumstances, it is the subjective assessment that requires the greatest
skill, and its findings are far more important than the assessment of the physical­
movement tests. The clinician needs to know what specific information to look
for. This is not a facetious remark, since it is the most common area where
mistakes are made, thereby ruining any value in the assessment. The kinds of
question the clinician should ask are as follows:
"During the whole time of treatment, is there anything I have done that has
made you worse?"
"Of the things I have done to you, is there any one particular thing (or more)
that you feel has helped you?"
"Does your body tell you anything about what it would like to have done to
it to make it start improving?"
"My record of your treatment indicates that after the lumbar-traction treat­
ment last Friday, you had a bad day Saturday, but that by Monday the
pain had subsided and you thought you might have been better than you
were before the traction. Looking back to that weekend now, do you feel
that the traction did help you? Do you feel that if the traction had been
gentler and of shorter duration that you might not have had the recurrence
of symptoms, and that you might then have been more sure of the treat­
ment' s effect?"
"Do your symptoms tell you that it might be a good plan to stop treatment
for say, 2 weeks, after which a further assessment and decision could be
And so the probing interrogation continues until two or three positive an­
swers emerge, which will guide the further measures that should be taken. The
questions are the kind that involve the patient in making decisions, and that
guide the clinician in making a final decision regarding treatment.
There is a fourth kind of retrospective assessment. If treatment is still
producing improvement but its rate is less than anticipated, a good plan is to
stop treatment for 2 weeks and to then reassess the situation. If the patient
has improved over the 2-week period, it is necessary to know whether the
improvement has been a day-to-day affair, thus indicating a degree of sponta­
neous improvement. If the improvement only occurred for the first 2 days after
the last treatment, then it would seem that the last treatment session was of
value and that three or four additional treatments should be given followed by
another 2-week break and reassessment.
The Maitland Concept 167
Final Analytical Assessment
When treatment has achieved all it can, the clinician needs to make an
assessment in relation to the possibility of recurrence, the effectiveness of any
prophylactic measures, the suggestion of any medical measures t hat can be
carried out, and finally an assessment of the percentage of remaining disability.
The answers to these matters are to be found analyzing all the information
derived from ( I ) the initial examination; (2) the behavior of the disorder through­
out treatment; (3) the details derived from retrospective assessments; and (4)
the state of affairs at the end of treatment, taking into account t he subjective
and physical changes.
This final analytical assessment is made easier as a clinician gains experi­
ence. It is necessary for this experience to be based on a self-critical approach
and on analysis of the result s, with t he reasons for these results.
The question has often been asked, "How did this method of treatment
evolve?" The attributes necessary to succeed in t his treatment method are an
analytical, self-critical mind and a talent for improvisation.
With t his as a basis, t he next step is to learn to understand how a patient's
disorder affects him. Coupled with this is the need to have sound reasons for
trying a particular technique and then the patience to assess its efect. In the
Maitland concept, this has developed over the years into a complex interrelated
series of assessments:
Why are painless techniques used to relieve pain?
A: Experience with patients who have had manipulative treatment else­
where allows us to inquire as to which kind of technique was used and
to observe its effect. When patients emphasize the extreme gentleness
of some successful clinicians, one is forced to the conclusion that there
must be ways of moving a joint extremely gently and thus improving
patients' symptoms. Having accepted this fact (and t hat is not always
easy), the obvious next step is to reproduce these techniques. For exam­
ple, a technique one patient may describe can t hen be used on other
patient s who fit into the same kind of category. The clinician can learn
what its possibilities are via the assessment process.
Why, conversely, are some of the techniques quite vigorous and painful?
A: When treatment reaches a stage when nothing seems to help, a useful
axiom is "Find the thing that hurts them, and hurt them." This should
not be interpreted as being cruel to a patient, or that one is "out to hurt
them," come what may. The hurting is a controlled progressive process
with a strong emphasis on assessment. From using this kind of treatment
on appropriate patient s, it has become obvious that some disorders need
to be pushed to the point of eliciting pain in order to aid recovery. This
168 Physical Therapy of the Low Back
approach may be seriously questioned by some practitioners, but it can
be a most useful technique in appropriate circumstances.
How did treating joints using strong compression of the joint surfaces
come about?
A: If, for example, a patient has shoulder symptoms, felt only when lying
on it, and if normal examination methods reveal very little, then the
thought processes go something like this:
"I believe him when he says he has a shoulder problem. "
"There is nothing to indicate any serious or sinister disorder."
"He has not responded to other treatments. "
"So i t must be possible to find something on examination that relates
to his problem. "
"How can I find that something? What lead is there?"
"He says, ' I cannot lie on it."
"So I will ask him to lie on it and then move it around and see what
happens. "
By experiment with techniques (improvisation) until the patient's
pain can be reproduced, and having found the thing that hurts him,
treatment should then aim to hurt him in this controlled manner:
"As the patient doesn't move his shoulder around when he's asleep
and lying on it, why is my examination using compression only,
without movement, not painful?"
One would expect it to be painful !
"However, he has to lie on it for half an hour before pain forces him
to change his position, so try compression again but make it stronger
and sustain it longer." After half a minute or so of sustained maxi­
mum compression without movement his pain will certainly appear.
How did the slump test and treatment evolve?
A: Some patients who have low back pain complain about difficulty getting
into a car. By re-enacting the action and analyzing it, it is found that it
was not the fexing of the lumbar spine that made getting into the car
difcult; it was the head/neck flexion that provoked the low back symp­
toms. Examination using standard movement tests for structures be­
tween the head and the sacrum do not reveal anything; therefore, re­
enact the particular movements and remember that the only structure
connecting both areas must be in the vertebral column, most likely
within the vertebral canal. To put these structures on stretch was the
only method that reproduced the complaint. The maximum stretch posi­
tion is the position now referred to as the slump position.
We now read of using mobilizing techniques to make a nonuniting frac­
ture unite. How did this come about?
A: I n the past, traditional methods used t o stimulate union have been (I)
remove all support for the fracture site and allow the patient to take
weight through the fracture, and (2) surgically explore the area and make
both ends of the fracture site bleed, and then splint them in apposition
again. If such things can promote union, then why not try passively
The Maitland Concept 169
moving the fracture site? Based on this reasoning and linking it with
our axiom "find the thing that hurts and hurt them," it was found that
it was possible to cause "fracture-site pain." This characteristic pain
was found to have two other characteristics: (I) the pain stopped imme­
diately when the treating movement was stopped, and (2) no side effects
were provoked. This then meant that the treatment could be repeated,
and in fact pain became harder to provoke (union took place).
I . The Age. p. 5. 2 1 August 1 982
2. McNab I: Negative disc exploration: an analysis of the causes of nerve root involve­
ment in 68 patients. J Bone Joint Surg 53A: 89 1 , 1 97 1
3. Maitland GD: Passive movement techniques for intra-articular and periarticular dis­
orders. Aust J Physiother 3 1 : 3, 1 985
4. McNair JFS: Non-uniting fractures management by manual passive mobilization.
Proceedings Manipulative Therapists' Association of Australia, p. 88, Brisbane,
1 985
5. McNair J FS, Maitland GD: The role of passive mobilization in the treatment of a
nonuniting fracture site-a case study. International Conference on Manipulative
Therapy, Perth, 1 983
6. Maitlanr GD: The hypothesis of adding compression when examining and treating
synovial joints. Orthop Sports Phys Ther 2: 7, 1 980
7. Edwards BC: Combined movements of the lumbar spine: examination and clinical
significance. Aust J Physiother 25: 1 47, 1 979
8. Maitland GD: Negative disc exploration: positive canal signs. Aust J Physiother 25:
6, 1 979
9. Maitland GD: Vertebral manipulation. 5th Ed. Butterworths, London, 1 986
1 0. MacConaill MA, Basmajian SV: Muscles and Movements. Waverly Press, Balti­
more, 1 969
I I . Edwards BC: Movement patterns. International Conference on Manipulative Ther­
apy, Manipulative Therapists' Association of Australia, Perth, 1 983
1 2. Edwards BC: Manual of Combined Movements. Churchill Livingstone, London,
1 992
6 Mechanical Diagnosis
and Therapy for
Disorders of the
Low Back
Robin A. McKenzie
In the first edition of thi s book, I I wrote, "Over the next few years, physical
therapy will have the opportunity to become the key profession wi thi n medicine
responsible for the del i very of conservative care for mechanical di sorders of
the spine. If appropriate steps are not taken now we may never again have
another such opportuni ty. "
Since 1987 the practice of physical therapy has changed l i ttle. Assessment
methods and therapies are lamentably outdated. Ti me is running out . Chi roprac­
tors i n the United States al ready see more patients with mechanical back pain
than the combined total of patients seen by al l other heal th providers. Scientific
support for chi ropractic is accumul ati ng at a very credi table rate. Conversel y,
t he scientific l i terature contains little t o support t he control l ed use by physical
t herapists of such modal ities as heat , col d, 2 laser therapy, 3 transcutane. ous elec­
trical stimulation, 4 infrared, short-wave diathermy, interferential t herapy, or
magnetic therapy. 5 Al though Nwuga6 found a beneft from the use of ultrasound
in treating herniated di scs, the experi ment has not been successful l y repeated
by others .
On the other hand, Gnatz7 documented case studies of two patients wi th
disc herniation where paraspinal ul trasound fol lowi ng hot pack application
caused severe increase in low back and radi cular pain duri ng appl ication.
In the United States , legislative and scientific advances by chi ropractors
17 1
1 72 Physical Therapy of the Low Back
suggest that physical therapists are fal l i ng further behind in the race for the
patient's spi ne. Medical tolerance of chi ropractors, if not actual support, is
growing worldwide. 8 Si nce the study by Meade et al , 9 chi ropractic is taking on
a new and confi dent profl e. Education of chi ropractors is movi ng toward the
traditional medical traini ng model , and more medical professionals are involved
in their education.
Physical therapists i n general are not special l y trained, nor are they ski l led
in the del i very of treatment for spinal disorders. Only those physical therapists
with special traini ng i n mechanical procedures and manual techniques can ex­
pect to attai n the ski l l s necessary to compete successful l y wi th chi ropractors.
Some i nstitutions now provi de postgraduate trai ni ng in manual and manipula­
ti ve techni ques for physical therapi sts. Even so, doubts sometimes remain as
to the level of expertise di spensed.
Al though a myriad of techniques may be taught for each and every skeletal
arti culation, the i ndications for the use of such techniques are frequently based
on findi ngs derived from palpation of i ntersegmental motion or from radiologic
screeni ng. Neither palpation nor radiographs5 are hel pful i n determining diagno­
sis i n nonspecific mechanical l ow back pai n. Fai l ure to obtain consistent results
with manual or manipulati ve procedures frequentl y drives manual therapists
and chi ropractors down the l ong road in the never endi ng search for the ul timate
technique that surel y must somewhere exi st, and that must surel y succeed when
al l el se has fail ed.
A major obstacle to the attainment of excel lence in the del i very of spinal
therapy i s the requirement that physical therapists become expert in the treat­
ment of all muscul oskel etal di sorders. Such expertise is difi cul t if not impossi­
ble to acquire when spread so wi del y. Chi ropractors, on the other hand, devote
the greater part of their traini ng sol el y to the study of knowledge related to
spinal di sorders and their treatment.
Perhaps the ti me has come for the pol i tical forces wi thi n the profession of
physical therapy to make a stand and insist that onl y those with specialist ski l l s
and qualifications shoul d be permitted to treat activi ty-related spinal di sorders.
It is now common for orthopedic surgeons to specialize in specific surgical
techni ques for specific anatomic regions. Thus there are surgeons specializing
i n hip replacements or knee replacements, as wel l as those devoting their entire
ski l l s to spinal surgery. If physical therapists are to compete in the current
envi ronments they must l ikewise establish soundl y based specialties. Such spe­
ci al ists must have the abil ity to critical l y analyze the sci entific l i terature. The
evi dence is there for all to see. Adopt scientifical l y valid methods of assessment
and treatment or perish. Society wi l l no longer permit the dispensation of use­
less treatment. Our treatment methods must be directed at i mproving the patho­
logic condi ti on rather than modul ating pai n.
Increasingl y, for the treatment of mechanical back di sorders, the scientific
l i terature is recommendi ng that active care replace passi ve; that modalities are
out; that exercise, acti vity, and movement are the means by which we are most
l i kel y to i mprove the health of i nj ured spinal structures.
Disorders of the Low Back 1 73
Modern medicine has advanced with such rapi di ty that it can be fai rl y
stated that in the past 20 years we have learned more of the precise nature of
the causes of spinal pain than in al l of previous recorded hi story, and the search
has narrowed to two structures l i kel y to be i nvol ved i n the producti on of most
mechanical back pai n. 5
The i ntervertebral di sc , with i ts strong annul us fibrosus retai ni ng t he gel­
l i ke nucl eus, probabl y attract s the most attenti on.
-17 I ndeed, Mooney 18 has
recent l y stated that, ''In summary, what is the answer to the questi on of where
i s the pai n coming from in the chronic l ow back pain pati ent? I believe i ts
source ul timately, is i n the di sc. "
The journal Spine has publ ished t he resul ts of many studi es aimed at the
examination of intervertebral di sc structure, functi on, pathology, and treat­
ment. The other mobile structures to capture the attention of those i nvestigating
back pain are the zygapophyseal joi nts. These are also probable sources of
pai n, but the precise pathology causi ng the pai n is unknown. The frequency of
"facet " joint pai n, however, has been chal lenged i n two recent studi es.
Disorders of the sacroiliac joint occur wi thout doubt , but most are i nflam­
matory in origin. That true mechanical l esi ons occur is also recogni zed. They
are, however, uncommon and usual l y onl y occur fol lowing pregnancy Y Physi­
cal therapists, especial l y in the United States, 22 or wherever t herapi sts are
recei vi ng i nstruction from osteopat hs, are "di scovering" sacroi l iac pathology
in many of their pat i ents. It i s l i kel y that ei ther the proponents are wrong or
the literature is i n error.
The hi storical obsession of physical therapists wi th the musculature as the
main source of backache has al ready been exposed.
3-25 Al though orthopedic
opinion does not support the proposal , 26 muscle imbal ance as wel l as muscl e
strain are stil l considered by some physical t herapists to be common causes of
persistent back pai n.
It appears now t o be general l y accepted wi t hi n medici ne t hat many l ow­
back problems are mechanical i n ori gi n, probably ari si ng in the i ntervertebral
di sc early in life and i n t he apophyseal joi nts much later in l ifey-3
the treatment for these particular probl ems shoul d mai nl y be mechanical . Thi s
fact has been recognized through 2500 years of recorded history, 3
and most
treatments today given for the al l eviation of back pain contain mechanical com­
Physical therapists have wi del y adopted (as occurred with the chiroprac­
tors and osteopaths) a system of "pathology" quite separate and contrary to
that which exists wi thi n medi ci ne as a whol e, and orthopedi cs in particular.
Medicine has always been able to exert control over and put pressure on i ts
wayward practitioners i n order to protect the publ i c from outrageous cl ai ms
and methods of treatment .
ls it not ti me to adopt those control s wi thi n physical therapy, especial l y
when we see such fri nge concepts as craniosacral techni que and myofascial
release techniques being taught wi thout the sl ightest sci entific evi dence to sup-
174 Physical Therapy of the Low Back
port thei r consideration as tool s for the treatment of spi nal di sorders? If we
fai l to curb the development of unscientific cul ti sm, we deserve to lose what
shoul d be our rightful place wi thi n the medical team.
By adopti ng acti ve mechanical procedures for treating mechanical low back
pai n, physical therapists have much to gai n. In the 1950s physiotherapists,
especial l y i n Bri ti sh Commonwealth countries, began i ncorporati ng manipula­
tive procedures i n thei r treatments. Prior to this, mechanical therapy within
physiotherapy consi sted of techniques of massage and exercise in which the
proponents were parti cul arly adept . Then along came spi nal manipulative ther­
apy (SMT), but there was no amalgamation between the exercise of physiother­
apy and the manipulation of the osteopath and chi ropractor. The conceptual
model for the di spensi ng of exercise was completely different from the concep­
tual model for the use of manipul ative therapy. There was no marriage between
patient-generated and therapi st-generated force. The two partners were i ncom­
pat ible.
Presentl y al l over the world enthusiastic physiotherapists, frustrated and
stifled by years of control by medicine or di si l l usioned by the use of ineffectual
methods of physiotherapy, are "di scovering" mobil ization and manipulation
and are del i veri ng SMT as the treatment of choice for most patients with spinal
pai n. This i ni tial enthusiasm, al though understandable, must be tempered and
brought into perspective. Those of us well experienced i n the use of manipula­
tive therapy still derive that unique sati sfaction whenever a spectacular im­
provement i s obtai ned. There is no doubt that many patients benefit from SMT.
Several studies demonstrate that t here i s a short-term benefit obtained from
SMT. 32-37
A recent study9 has found that a long-term benefit may be obtained from
SMT admini stered by a chi ropractor. The study has attracted much attention
by chi ropractors who obviousl y fi nd much that i s pleasing in the concl usions
drawn. The study has many flaws however, which are el egantl y described by
Assendelft et al . 38 SMT neverthel ess has a particular and i mportant part to
pl ay in the treatment of spi nal pai n. Those in the profession who, due to long
experience with SMT, are ful l y aware of the l i mitations as well as the benefits
of thi s form of t reatment, must uti l ize thi s experience to moderate the excesses
t hat emerge from the overuse of mani pul ati ve techniques .
Unti l we have learned to di sti nguish between improvement that occurs
directl y as a resul t of treatment, and improvement that resul ts from spontaneous
heal ing or the nat ural hi story, our credi bi l i ty i s at risk. When our patients im­
prove over a period of 3 to 4 months, can we seri ousl y attribute their recovery
to our manipulative or mechanical prowess applied over thi s period? Failure
to recognize and understand the natural hi story and the self-l i mi ti ng characteris­
tics of the nonspecific spi nal di sorder wi l l perpetuate the i nabi l i ty of so many
to critical l y review thei r treatment methods and resul ts.
Disorders of the Low Back 1 75
Fi nal l y, we must be increasingl y concered that al l currentl y prescribed
mechanical treatments for mechanical spinal di sorders create dependence on
such therapies. Whenever we do somethi ng to or for the patient, the patient, and
very often the therapist, attributes any improvement to the treatment di spensed.
Ignorance or di sregard of the natural hi story of mechanical back di sorders al­
lows all health providers to cl aim a success rate of over 90 percent!
In the future, because of the ever-rising cost of health care, society wi l l
demand that treatments for musculoskeletal di sorders contain "self-treatment
concepts" that will allow the patient future independence from therapy and
therapists . If there is the sl ightest chance that patients can be educated in a
method of treatment that enables them to reduce thei r own pai n and di sabi l i ty
using their own understanding and resources, they shoul d receive that educa­
tion. Every patient is entitled to this information, and every therapist shoul d
be obliged to provide it. It is appropriate at this poi nt to quote di rectl y from
In the therapeutic field today, it i s virtually impossible to i nt roduce a new
drug without cl i nical and laboratory tests to prove its effecti veness and we
are i ncreasingly alert to and critical of different types of pharmacological
side effects. The same approach should be used for the different forms of
treatment of low back pain and we should criticall y re-assess our present
The fi ndings and recommendations of the Quebec Task Force5 (QTF) are
important to all health providers i nvol ved with the diagnosis and treatment of
mechanical spi nal di sorders.
Publi shed in Spine, the Report of the Quebec Task Force on Acti vi ty Re­
lated Spinal Disorders was commi ssioned and funded by the I nstitute for Work­
ers Health and Safety of Quebec, whi ch was concerned about the i ncreasing
cost of treating spinal di sorders, especial l y the cost of physical therapy. The
problems of diagnosis are highl ighted i n some of the fol lowi ng excerpts.
Pain i s the primordial , and often the onl y, symptom of the vast majority of
spinal disorders.
Of the numerous pathologic conditions of the spi ne, nonspecifc ail ­
ments of back pain in the l umbar, dorsal , and cervical regions, wi th or
without radiation of pai n, comprise the vast majority of problems found
among workers (and the i ncidence in general populations can only be
It is estimated that 85 percent of back pai n epi sodes are nonspecific. 5 Thus
only 1 5 percent of patients can be specifical l y di agnosed wi th our present tech­
nology and understanding.
176 Physical Therapy of the Low Back
Although there are considerably more cl i nical studies on patients suffer­
ing from problems of the l umbar area than there are on patients with prob­
l ems in the cervical region, pain develops because of the irritation of struc­
tures sensitive to pai n, and these are the same for all segments of the spi ne.
These structures are bones, di scs, joi nts, nerves, muscl es and soft tissues.
It i s difficul t to identify precisel y the origin of the pai n, because even
if its characteristics may sometimes point to a given structure, the pain often
remains unspecific. In addition, it is generally impossible to corroborate
clinical observations through histologic studies, because on one hand the
usual benignity of spinal di sorders does not justify that ti ssue be removed
and on the other, there i s ofen no modification of t i ssue identifiable through
current methods.
Nachemson39 confirms our di l emma: "Wi th one exception, we are not
yet in a posi tion to posi ti vel y i denti fy preci sel y t he structures involved in the
producti on of common l ow back probl ems. The great majority of these problems
are mechanical i n origin but defy identification as to the structure invol ved or
the nature of the faul t . "
I n spite of the technological advances that have provided computed tomog­
raphy (CT) scanni ng and magnetic resonance imagi ng (MRI ) , positive findi ngs
from these imaging techni ques are found i n large numbers of normal asymptom­
atic subjects.
I n order to promote a better system of identification of spi nal di sorders,
the QTF made several recommendations. It was fel t that it would be hel pful
if all professions i nvol ved in the care of spinal di sorders were able to better
communicate wi t h one another. The fol lowi ng recommendations were made
with regard to t he devel opment of the ideal system of classification:
Reliability: A gi ven case of a vertebral di sorder shall be cl assified i n the same
manner by two or several practitioners.
Clinical usefulness: It wi l l faci l i tate the maki ng of cl i nical decisions as well
as the eval uation of care.
Simplicity: I t s use wi l l be si mpl e and wi l l nei ther call for complete paracl i nical
examinations nor encourage superfuous i nvesti gati ons.
Using these criteria as a guide, the QTF has recommended the fol lowing
categories be adopted by al l heal th-related professi ons i nvol ved in the treatment
of back pain:
1 . Pai n i n t he l umbar, dorsal, or cervical areas, without radiation below
the gluteal fol d or beyond the shoul der, respecti vel y, and i n the absence of
neurologic signs
2. Pain in the l umbar dorsal, or cervi cal areas, wi th radiation proximal l y
(i . e. , to an upper or lower l i mb but not beyond t he el bow or t he knee, respec­
t i vel y) and not accompanied by neurologic si gns
Disorders of the Low Back 177
3. Pain in the lumbar, dorsal , or cervical areas , wi th radiation distal l y ( i . e. ,
beyond the knee or t he el bow, respecti vel y) but without neurologic signs
4. Pain in the l umbar, dorsal, or cervical areas, with radiation to a l i mb
and wi th the presence of neurologic si gns (e.g. , focal muscul ar weakness, asym­
metry of refl exes, sensory loss i n a dermatome, or specific loss of i ntest inal ,
bladder, or sexual function)
"Thi s category i ncl udes the radicular syndromes, which are well described
in classic textbooks. These radicular syndromes may be due to various afec­
tions, the most frequent one being the discal hernia. However, other mechanical
di stortions of the spine may trigger an i rritation or a radicular defici t. "
The conditions described in t he fi rst four categories of the QTF represent
90 percent of activi ty-related low-back di sorders. They also represent progres­
sively complex pathologies . As the symptoms radiate further from t he midl i ne,
a pathology evolves t hat is more likel y to be resi stant to si mpl e treatment meth­
ods . Thus, category I is the least compl ex disorder and category 4, representi ng
compression or interference of t he conducti vi ty of spinal nerves, i s t he most
complex and potential l y di sabl i ng entity.
I n recommending the use of pain patterns to cl assify nonspecific mechani­
cal disorders, the QTF attempts to avoid the perpetuation of meaningless diag­
noses that are more l i kel y to be wrong than right.
We may not know the precise nature of the injury or the exact location of
the structure affected, but i n simple terms, mechanical pai n ari ses onl y from
trauma ( which wi l l be of l i mited durati on), internal derangement of the i nterver­
tebral di sc ( which may often appear as a recurring and episodic i njury) , or
contracture or adaptive shorteni ng, causi ng persi stent pai n l ong after repair is
compl ete. A further cause of mechanical pai n, and probably the most frequentl y
encountered, is that ari si ng from prolonged stati c loadi ng i n t he absence of
If we confine oursel ves to the mechanical disorders of the back, there are
relatively few atypical pathologies occurring in our dai l y pati ent caseload. Most
problems can be related to ei ther internal or external disruption and/or di splace­
ment of the interverbetral disc, or to damage and consequent repair of ligamen­
tous or capsular periarticular soft tissues. Each di sorder causes pai n to behave
in a typical l y characteristic fashion. Pain l ocation and i ntensi ty can change,
and the manner in which thi s occurs al l ows us to separate probl ems caused by
internal disc di splacement from problems ari si ng from contractures and fi brosis .
This is necessary because the former require di fferent t reatment than the l atter.
The use of pain patterns to identify varying categories in the nonspecific
spectrum of mechanical back pain, as contained i n the QTF report, is al so
a key feature of t he McKenzie cl assification of mechanical spi nal di sorders
178 Physical Therapy of the Low Back
developed in the earl y 1960s. 4
Indeed, the McKenzie system of treatment
originall y evol ved from an abil i ty to systematicall y alter pain location by apply­
ing repeated movement s, whi ch caused pain to move from a di stal to a central
locati on. "Centralization of pain" is i ndicative of good outcome and is a prog­
nostic sign in al l patients wi t h radiating or referred symptoms. 42
By repeating movements that cause pain to move from a distal to a more
proximal locati on, a QTF cl assification can be changed from a more to a less
complex pathology. Expressed in another way, rapid nonsurgical treatment
of referred and radiating symptoms is frequentl y possible usi ng specifically
customized end-range repeated movements that centralize pai n. The McKenzie
system provides a logical and compatible t reatment protocol that fi ts wel l wi th
the QTF recommendati ons. The McKenzie protocol s are described el se­
Mechanical Diagnosis
After 40 years of experience in treating mechanical di sorders of the spine,
I stil l consider i t preferable that the patient's general practi tioner or a speciali st
establ i sh a differential diagnosi s . Thus serious pathol ogies shoul d be excl uded
from the patient popul ation. Radiologic i nformation el i mi nates most serious
di sorders and exposes archi tectural faul ts unsuitable for the mechanical ap­
There are many differing vi ews regarding the best method of establishing
a mechanical diagnosis. It is agreed almost universal l y t hat it is necessary to
obtain detailed information from the pat ient by way of hi story, but t here is
often di sagreement about the i mportance and relevance of much of this i nforma­
tion. Some therapi sts attach much importance to one particular response,
whereas others will consider the same response to be irrelevant.
Cl inicians have a simpl e choice. If a large range of detailed information is
to be gathered, much of it will be irrelevant or unrel iable. 53 If cl i nicians are
prepared to l i mi t the information, then they wi l l i ncrease its rel iabi l i ty and
rel evance.
There are those, mai nl y osteopathical l y ori ented, who decide the nature
of the mechanical problem principal l y by pal patory means; some even claim
to be able to determi ne by pal pation al one the levels of exi st ing pathology.
However, i ntertherapist rel iabil it y usi ng palpation alone has yet to be demon­
strated. 54
Chiropractors rel y on a combination of diagnostic criteri a, but mai nl y on
information obtained from radiologic and palpatory findings . These have always
been their main tools of diagnosi s and remain so today. Their treatments, how­
ever, have expanded significant l y as they have embraced more of orthodox
medicine in recent years.
In the l ong run, the most appropriate mechanical procedure for any given
condi tion wil l be identified from the pati ent's descripti on of pai n responses
to the applied mechanical forces. If the i mpl i cations of these responses are
considered within the conceptual model s described here, successful appl ication
of the McKenzie methods of t reatment shoul d fol l ow.
Disorders of the Low Back 1 79
My concl usions on completion of the evaluation process are based on the
effects on pain resul ti ng from the appl i cation of static and dynamic loadi ng
forces. Alteration i n pai n i ntensi ty and location fol lowing static l oadi ng i n flexed
or extended postures are correlated with pain responses obtained from repeti­
tive end-range sagittal and frontal pl ane movements performed i n l oaded and
unloaded positions.
Fol l owi ng t he performance of these movements, a subdi vi sion of pati ents
wi thi n the nonspecific spectrum of back pai n is possi bl e.
Subgroups in Nonspecific Low Back Pain
Three categories can be identified: postural, dysfuncti on, and derangement
syndrome. Patients i n the derangement group are by far the most frequentl y
encountered. The precise means of identi fication and the concepts and methods
of treatment of these syndromes are described in detail el sewhere. 4
Postural Syndrome
Pat ients wi th the postural syndrome are usual l y under 30 years of age,
have sedentary occupations, and frequentl y are underexerci sed. As a resul t of
prolonged end-range static loading, such as occurs in si tti ng and bendi ng, they
develop pai n local l y, adjacent to the mi dl i ne of the spinal col umn. Such pati ents
frequently complain of pai n fel t separatel y or si mul taneousl y i n the cervical ,
thoracic, and l umbar areas .
Pain from the post ural syndrome is never i nduced by movement, rarel y
referred, and never constant . There is no pathology, no l oss of movement, and
symptoms are entirely subjecti ve. 40
Dysfunction Syndrome
Pain in the dysfunction syndrome usual l y devel ops i nsi di ousl y, appearing
local l y, adjacent to the mi dl i ne of t he spi nal col umn, and is provoked on at­
tempting full movement that deforms adapti vel y shortened or contracted soft
tissues. Pain is always fel t at end range and does not occur duri ng the move­
ment. With the exception of a patient with nerve-root adherence, pai n from
dysfunction is never referred.
Loss of movement i n t he dysfunction syndrome has t wo common causes.
The most common cause of reduced spi nal mobi l i ty i s poor postural habi ts
and l ack of exercise. Poor postural habi ts al l ow adapti ve shorteni ng of certain
structures. The resul t is a gradual reduction of mobi l i ty with agi ng. The move­
ments reduced are usually those extension movements essential for the mai nte­
nance of the very erect posture.
The second cause of reduced spinal mobi l i ty is contracture offibrous repair
fol l owing trauma. Thus an i nextensible scar forms wi thi n or adjacent to other-
180 Physical Therapy of the Low Back
wi se healthy structures, and wi l l cause reduced mobi l i t y. The pain resul ti ng
from stretchi ng of thi s i nextensibl e scar appears onl y on attempti ng ful l end­
range movement s. The pain does not occur duri ng the movement or before the
scar is stretched. Surroundi ng heal thy structures capable offurther extensibil ity
are restri cted by the scar ti ssue i tself. Thus t he persi sti ng pain resul ts from the
repair process i tself.
It is not possible to i dentify the structure causi ng the pain of dysfunction,
but any of the soft ti ssues adjacent to the vertebral col umn may adapti vel y
shorten or may be damaged. Thus the pai n may resul t from i njury to any of
the l igamentous structures in the segment from the i ntervertebral di sc, the zyga­
pophyseal joi nts, or the superfi ci al or deep muscl es or thei r attachment s. The
pain may also resul t from adherence of t he spi nal nerve root or dura fol l owi ng
i ntervertebral di sc prolapse or herniati on. Described simpl y, the pai n of dys­
functi on i s produced i mmediatel y when stretchi ng of shortened ti ssue occurs.
Conversel y, the pai n ceases i mmediatel y when stretchi ng ceases. 40
Derangement Syndrome
By far the most common cause of back pain, the derangement syndrome
affects those between 20 and 55 years of age. Those affected devel op pain,
usual l y of sudden onset; t hat i s, i n a matter of a few hours or over a day or
two they change from compl etel y normal to significantl y di sabled beings. Very
ofen this syndrome appears for no apparent reason. The symptoms may be
fel t l ocal l y, i n the mi dl i ne or adjacent to the spi nal column, and may radiate
or be referred di stall y in the form of pai n, parasthesia, or numbness . The symp­
toms are produced or abol ished, i ncreased or reduced, may central i ze or periph­
eral ize, or remai n better or worse fol lowi ng the repetition of certain movements
or t he mai ntenance of certain posi ti ons.
Pai n from t he derangement syndrome may al ter and change both i n regard
to the location of the pai n, or the extent of the area afected, which may i ncrease
or decrease. Pai n from the derangement syndrome may cross the mi dl i ne, for
exampl e, and move from the right of the l ow back to the left .
Discogeni c pathology shoul d always be suspected when t he patient de­
scribes that symptoms change location with repeated movements or prolonged
posi tioning.
Change i n the di stribution or location offuid nucleus or sequestrum wi thi n
the i ntact annul us may i n t urn alter the locati on of pain accordi ng to the si te
of deformati on. 56 Posterocentral bUlging wi l l cause central or bilateral pain,
whereas posterolateral bul gi ng wi l l cause unilateral pai n. 57•58
Pain from the derangement syndrome wi l l be constant as long as t i ssue
wi thi n the i ntact annul us remains di splaced. There may be no position i n which
the patient can fnd relief. The pain may be present whether movement i s per­
formed or not and thi s pai n is usual l y described as an ache. That ache is then
made worse by movement i n certai n di recti ons and reduced by movement in
other directions .
Disorders of the Low Back 181
In the derangement syndrome, especial l y i n severe cases, gross l oss of
movement may occur. Displaced ti ssue obstructs movement in the di recti on
of the di splacement . I n severe cases t he degree of di spl acement i s such that
postural deformities, such as kyphosi s and scol iosi s, are forced. Sudden l oss
of spinal mobi l i ty and the sudden appearance of postural deformi ty i n acute
l ow back and neck pain may be l i kened to t he sudden l ocki ng that may occur
in the knee joi nt where internal derangement of the meni scus i s common.
The mechanism of i nternal derangement of the i ntervertebral di sc i s not
ful l y understood. That the nucl eus pul posus can be di spl aced toward and escape
through a damaged annul ar wall i s i narguabl e. 59,6o It i s hi ghl y l i kel y that thi s
wil l follow as a consequence of sudden or viol ent movement, or wi th sustained
postures in younger pati ents. Ol der patients have a stifer, l ess flui d nucl eus,6
which is less l ikel y to be di splaced from wi thi n i ts annul ar envelope. 59.6o
62 It
i s al so hypothesized that prior to a frank annul ar l esi on wi th nuclear herniation,
there exi sts i ncomplete tears i nto whi ch nuclear material may be di splaced.
Thi s alters the joi nt biomechanics and may be responsible for the postural
deformities ( e. g. , localized scol i osi s) observed.
Creep of the fluid nucl eus/annul us compl ex wi l l di sturb the normal align­
ment of adjacent vertebrae63 and change the resti ng shape of the di sc. 62 Thi s
change of shape wi l l al so affect the abi l i ty of the joint surface to move i n i ts
normal pathway63 and movement deviation to the right or left of t he sagittal
plane will resul t on attempti ng fexi on or extension.
Described simply, the pain of derangement occurs as a consequence of a
change in di sc shape wi th related mi sal i gnment of the mobil e segment and i ts
associated abnormal stresses . 40
Identification of the different syndromes is based on the effects that re­
peated movements have on the i ni t iation of the pain: the poi nt in the movement
pathway where pain i s first percei ved; the si te of the pain and subsequent
change of location of the pain; the i ncreasing or decreasing intensity of t he
pain; and fi nal l y abol i tion of the pai n. Mechanical pain can arise from a l imited
number of events or combination of events causi ng force to be appl ied to i nner­
vated soft tissues . Those soft ti ssues may be in a normal state, a contracted
state, or in an anatomical l y altered state with a change in the shape of the di sc.
Any of these events can be identified by the response of the patient's pai n to
the del i berate appl icati on of certai n mechanical stresses.
Pati ents wi th i nflammatory disorders, with spondyl ol i st hesi s or other unde­
tected minor fracture, and wi th pathologies unsuited to mechanical therapies
will behave atypical l y and be qui ckl y exposed and recognized when tested i n
the manner I have described.
Postural Syndrome
Normal tissues can become pai nful i n everyday l ife by the application of
prolonged stresses commonly appearing during static postural loadi ng condi­
tions, such as prolonged sitting, standing, or bending.
182 Physical Therapy ofthe Lol Back
Correction of faul ty postural habits removes inappropriate causative
stresses. No other treatment is required. I n order to remove the cause of pain,
the therapist must educate the patient.
Dysfunction Syndrome
Shortened structures cause limited movement and simultaneousl y cause
pain when the shortened structure is stretched.
Treatment woul d include remodel ing short structures by t he regular appli­
cation of stretching exercises. Dysfunction is not rapidl y reversible; weeks are
required to remodel and lengt hen. Structures t hat have adaptivel y shortened
over weeks and months cannot suddenl y lengthen by the application of high
velocity thrusts without incurring damage. Regular end-range exercise must be
considered as the most l ikel y therapy to infl uence shortened structures. It re­
mains to be seen whether exercise ( REPEX, repeated end-range passive exer­
cise) therapy or manual mobilization is superior in regards to the recovery of
function by remodeling.
Derangement Syndrome
An example of derangement syndrome is frank tears of the annul us fibrosus
with nucl ear displacement or annul ar bulging. The patient experiences aching
without movement, increased pain with movement in certain directions as dis­
placement increases, and reduced pain in other directions as displacement de­
creases .
This syndrome is subject to mostl y rapid reversal . The use of patient self­
treatment methods using repeated end-range movement has been successful in
the reduction of derangement within the lower l umbar segments. The rate of
reduction of derangement can be accel erated significantl y in a large number
of cases with the use of REPEX associated with appropriate self-treatment
protocols; maintaining reduction by correcting posture and avoiding wrong po­
sitions; restoring function before adaptive changes are established; and teaching
prevention of recurrence and self-treatment . Mobilization and SMT now be­
come important and may be required if self-treatment provides insufficient re­
ductive pressures.
Mechanical forces used to treat mechanical disorders of t he low back
shoul d be applied in a graduated form, first using patient self-treatment repeated
movements, progressing through mobilization, and final l y the application of
manipulative procedures.
The time t o appl y our special techniques of mobilization and manipulation
arrives when the patient, having exhausted all possibilities of self-treatment ,
requires an increase in the degree of pressure in the appropriate direction; that
direction, having already been determined during exercise, allows mobil izing
procedures to commence. Failing improvement with the use of mobilization,
Disorders of the Low Back 1 83
then and onl y then is mani pulation i ndi cated. Thus the gradual devel opment
of increasing force to bri ng about change is a logical and safe method of appl yi ng
mechanical therapies, assumi ng that vertebral and vascul ar pathology are ex­
cl uded. The ul ti mate weapon we have is t he mani pul ative t hrust techni que.
Why use that weapon on day I when it may wel l be t hat t he pati ent wi thout
being approached by the therapist i s capable of causi ng the change hi msel f (and
learning an important self-management lesson)?
I have, therefore, proposed si nce t hat ti me that we shoul d postpone or
avoi d mobi l i zation and mani pul ation unti l we have determined that resol uti on
of the problem is i mpossible usi ng the pat ient's own positi ons and movements.
This concept offers up to 70 percent of people referred to physi cal therapi sts
with mechanical low back pain the opportuni ty to t reat and manage thei r own
problem and thus become i ndependent of therapi st s. 46
The remaining 30 percent , i ncl udi ng patients categorized by myself as hav­
ing derangement 4 or 6 and therefore unable to apply self-treatment procedures
with lasting benefi t , wi l l always requi re the special ski l l s of the mani pul ati ve
therapist for correction of any l umbar or sci ati c l i st . 40-65 Others wi l l require
techniques in the form of mobil ization and some of those wi l l addi tional l y re­
quire manipulative thrust procedures.
This is not t o suggest that mobi l i zation and manipUl ation are no l onger
requi red in our armamentari um. Spinal mani pulative t herapy has a parti cul ar
and important part to play i n the treatment of mechanical spinal pain but i ts
di spensation is great l y mi sused. I am proposing that the t i me has come when
we must rat ionalize the use of such methods. We are now abl e to determi ne
within at least 48 hours from commencement of assessment whether manipula­
ti ve therapy will be necessary at al l . It shoul d no longer be ethical to apply the
technique i n order to find out retrospecti vel y if the procedure was i ndi cated.
Spinal mobilization and manipulation shoul d not be di spensed to the enti re
population wi th back pai n in order to ensure that the very few who real l y need
it actual l y receive it.
When all mechanical therapies fai l to i mprove the patient's condition, and
modulation of pai n i s consi dered necessary, physi cal therapi sts must ask
whether modulating pai n wi th physical therapy "gadgetry" i s better for the
patient and society as a whole than di spensation of rather i nexpensi ve medi­
As reported by the Quebec Task Force , the l iterature is sadl y l acki ng i n
wel l -control led, randomized studies evaluating t he various treatments for low
back and neck pai n. Studies of the McKenzie approach were absent at the t ime
of that report but are now begi nni ng to appear. The McKenzi e system i s in use
worldwide and is the subject of increasing scrut i ny. 42-45
Many cl i nical studi es have been compl eted, and many more are requi red.
The system l ends i tsel f to research because the conceptual model s of pathology
1 84 Physical Therapy of the Low Back
establ i shed to i dentify subgroups i n the nonspecific spectrum of back pain relate
to the common pathologic processes wel l described in t he l i terature and encoun­
tered dai l y in orthopedic cl i ni cs.
It has now been demonstrated that several different patient populations
can be taught to manage their own back problems usi ng self-applied movement
as described here and el sewhere. Studies by McKenzi e, 64 Ponte et al , 68 Nwuga
and Nwuga, 69 Kopp et al , 7
Di Maggio and Mooney, 43 Wil liams et al/' Donelson
et al , 42 Al exander et al , 7
Stankovic/3 Spratt et al/4 Roberts/5 and McKin­
ney /6 have al l demonstrated the benefts of this approach to the assessment
and treatment of mechanical spi nal di sorders.
Three studies (Ponte et al,68 Nwuga and Nwuga,69 and El naggar et al77)
have compared the use of Wi l liams' fl exi on exercises with McKenzie treatment
in small pati ent populati ons. Al l patients were acute, but patients in one study
(Nwuga and Nwuga69) al l had confrmed disc l esions by myel ogram and with
refl ex changes. Both studies demonstrated a significant difference in favor of
the McKenzie approach in terms of the rate of decreasing the pai n, restoration
of ful l range of moti on, increasing straight-leg-raising comfort , and increasing
si tti ng comfort . The average number of treatment sessions was also significantly
lower with the McKenzie approach than with the approach of Wi l l iams . Of
course, Wi l l iams' approach t ypical l y has l i ttl e or no assessment portion but
treats all pati ents essential l y the same. The McKenzie system, however, defines
the treatment on the basis of the patient's i ndi vi dual i zed mechanical needs
identifed by monitoring pain behavior during the assessment .
One of these studies (Nwuga and Nwuga69) fol lowed pat ients for 2 months
and noted a cl ear reduction i n the number of recurrences after t he McKenzie
method. This would seem to refl ect the abil i ty of these pat ients to prevent
further probl ems, thus defeating the high i nci dence of recurrences.
El naggar77 compared flexi on wi th extension exercises and not surprisingl y
found no difference i n the outcomes between t he two groups. By i nference,
the authors attempted to relate their resul ts to a lack of effecti veness of the
McKenzie protocol s. The authors were general l y unaware of the requirements
of t he McKenzie treatment methods and incorrectl y related the extension exer­
cises used in thei r study to those described in the Mc Kenzie protocol .
I n a randomized, bl i nd study comparing the use of McKenzie wi th two
other treat ments (back school and 90/90 tracti on), 97 percent of the patients
treated with McKenzie had i mproved after 1 week, whereas less than 50 percent
improved wi th ei ther of the other two treatment s. 71 The resul ts of both of the
latter treatments were simi lar to, or lower than, what might be expected with
no treatment at al l . The study attrition rate after the first week of treatment
i nval idated further concl usi ons, but t he first week's resul ts were fel t to be valid.
In a Swedi sh study73 the McKenzie system was compared to a mi ni -back
school program in treating acute low back pain in a working population. The
McKenzie-treated workers had a median sick leave of 10 days versus 17.5 days
in the comparison group and a mean sick-l i sting for recurrences t he frst year of
27 days versus 40 days with mi ni-back school . Onl y 45 percent of the McKenzie­
treated workers had first year relapses versus 80 percent of the comparison
Disorders of the Low Back 185
group. The average number of McKenzie treatments during the acute epi sode
was only 5.5. It would appear that t he McKenzie pati ents resolved thei r acute
episode and di sabil ity faster, were better able to prevent recurrences, and were
able to mi ni mi ze di sabi l i ty when symptoms di d recur.
Donelson et al42 have found that central ization of pain i s a reliable predictor
of outcome i n treating low back and referred pai n. I n applying repeated sagittal
movement testing according to the McKenzie protocol s, they found that pa­
tients consi stentl y demonstrated a di rectional preference, the majority central­
izing or reduci ng pai n with extensi on movements. Conversely fexi on was found
to increase pai n. 4
In a further study on centralization of pain, Donel son et aJ79 found that a
simi lar directional preference was shown in a majority of pati ents tested i n the
frontal pl ane, most experi enci ng centralization or reducti on of symptoms when
laterally flexi ng toward t he painful si de.
found that the adoption of extended l umbar si tti ng postures
reduced or centralized pain i n patients with nonspecifi c l ow back and referred
pai n. The effects were most marked in t hose wi th leg symptoms. Flexed pos­
tures, on t he other hand, showed no si mi lar reducti on i n pain i ntensi ty or loca­
tion change.
Roberts75 compared McKenzie treatment wi th nonsteroidal anti inflamma­
tory drug therapy. His patients were recruited prospecti vel y and were t reated
within 3 weeks of onset of an attack of low back pain. Both groups of patients
were encouraged to mobil i ze acti vel y. Rest , after the first 2 days, was di s­
The major measure of outcome was a wi del y used di sabil i ty questi onnaire.
At 7 weeks after onset of the attack, the McKenzie-treated pati ents were l ess
disabled compared wi th the drug treatment pati ents. Thi s difference became
significant when those McKenzie patients who could not be di agnosed accord­
ing to McKenzie's classification at the time offi .rst assessment by the physi o­
therapist were excl uded. This suggests that if a defi ni te diagnosis cannot be
made on initial assessment the result wi l l be less certai n, and that more experi­
enced practitioners will have greater success when treating groups of pati ents ,
because correct assessment seems to be a key part of treatment. 75
The patients in the McKenzie group who had not recovered after si x treat­
ments were found to represent a very di fficult group to treat-at least by the
physiotherapists i nvol ved i n the study. 75 Thi s l i mi t of si x treatments matches
the experience of Rath et al so i n America and Stankovic and 10hneIF3 in Scandi­
Patients who received McKenzie therapy were, however, away from work
for longer than the drug patients and it i s suggested that a di recti ve to return
to work from the physiotherapist is an i mportant element to the regi men.75
Careful psychological assessment was performed on al l patients i n connec­
tion with their personal responsi bi l i ty for pain control. Pati ents undergoing
McKenzie therapy were significantly more responsi ble for personal pain control
than the drug treatment pati ents 7 weeks after the onset of t he low back pain.
186 Physical Therapy ofthe Low Back
Thi s responsi bi l i ty alteration was sti l l significantly different when measured 6
months later. McKenzie t herapy alters the way patients thi nk about pain.75
The mechanism of centralization was examined i n detai l , and again those
pati ents who could be diagnosed on t hei r fi rst attendance with the physiothera­
pist showed better responses than the pati ents whose syndrome was unclear.
Scores of pain i ntensi ty matched a score for peripheralization wi th a very signifi­
cant degree of correlati on. 75
I n 1 986 Kopp et a170 reported that of 67 patients wi th herniated nucl eus
pul posus (HNP) t reated wi th the McKenzie extension protocol, 35 patients
were able to achieve normal l umbar extension wi thi n 3 days of admission to
the hospital . The remai ni ng 32 patients all requi red surgery and of these onl y
two were able to achi eve extension before surgery. Al l 67 had failed 6 weeks
of conservati ve treatment prior to being treated by the McKenzie protocol .
Kopp and co-workers concl uded that the abi l i ty of patients wi t h HNP and
radi culopathy to achieve ful l passive l umbar extension is a useful predictor
to sel ect patients who can be expected to respond favorably to conservative
management. They further reported t hat the i nabi l i ty to achieve extension ( posi­
tive extensi on sign) i s an early predictor of the need for surgical intervention,
and recommended extensi on exercises as a therapeutic modal i ty.
In 1990 Alexander et al72 reported on a fol l owup of the pati ents i n t he Kopp
study. It was found that after an average of almost 6 years from onset, 33 of
the 35 patients who did not require surgery were sati sfied with the resul t and
82 percent had been able to resume thei r old jobs. At long-term fol lowup, Alex­
ander and co-workers found that a negati ve extensi on sign was confirmed as
a predictor of a favorable response to nonoperati ve treatment of HNP i n 91
percent of the nonsurgical group i n Kopp' s study.
Onl y one study has assessed the long-term value of thi s treatment approach
for mechani cal probl ems in the cervical spi ne.76 I n a si ngle-blind randomized
prospecti ve study, McKi nney 76 found that advice to exerci se and correct pos­
ture in the early phase after i njury was superior to outpati ent physiotherapy
consisting of hot and cold applications, pul sed short-wave diathermy, hydro­
therapy, traction, and active and passive repeti ti ve movements. At 2-year fol ­
l owup fewer pati ents i n the exerci se group had persi sti ng symptoms. McKinney
suggests that the reason for the superior resul ts i n the exercise group is that
pati ents gi ven responsi bi l i ty for thei r own treatment may become self-suffi cient
at managing epi sodes of a mi nor nature, and there may be psychological advan­
tages in making patients responsible for thei r own treatment rather than victims
of their own symptoms.
Ri ddl e8 ) studi ed a much abbreviated version of the McKenzie system of
assessment for l ow back pai n and found the system to have low i ntertherapist
reliabil i ty. The therapists used in the Riddle study, unfortunatel y, did not re­
ceive traini ng i n the McKenzie protocol , even to a mi ni mum level of compe­
The procedures described i n the McKenzie protocol are relatively uncom­
pl icated and safe when applied by persons with the appropriate training. Re­
centl y, however, cauda equi na l esi ons have been reported fol lowi ng inappro-
Disorders of the Low Back 187
priate application of the McKenzie protocols by untrai ned persons. 83 Thi s
reinforces the need for adequate trai ni ng and qualifications for those therapists
wishi ng to claim competence i n the use of the McKenzi e system.
Specific customized exerci ses for i ndi vi dual pati ent needs can be i dentified
and successful l y del i vered for sel f-treatment of l ow back pai n. There are, how­
ever, l i mitations to the system i n some cases, usual l y i nvol vi ng pati ent fatigue
and lack of compl iance.
Ideal l y, to obtain the best resul ts, pati ent self-treatment exercises shoul d
be appl ied 10 to 15 repetitions every 2 hours. A total number of 80 to 100 end­
range movements per day can be achieved wi th thi s formul a. Most pati ents
wi th acute and subacute low back pain experience reducti on or central ization
of pain fol l owi ng the practice of customized sel f-treatment exerci ses. They
remain better and rapidl y and progressi vel y i mprove with practice. Even so,
there remain some whose pai n reduces or central i zes for a short period onl y,
and who do not progressi vel y i mprove dai l y.
Several possibil ities account for lack of response to repeated movement s.
It can be that t he pati ents condi ti on is irreversi bl e usi ng mechanical therapy.
This is not always cl ear at the first evaluati on. It could be the wrong mechani cal
diagnosi s. For i nstance, the patient may have dysfuncti on but has been cl assi ­
fied as havi ng derangement syndrome. Thi s can be confirmed by reassessi ng
and appl yi ng al ternative testing procedures.
Lack of response to self-treatment exercises coul d also i ndicate that al­
though the chosen di rection of movement i s correct, the frequency, force, or
degree of motion appl ied i s inadequate. Would the application of large numbers
of cycles to end range accelerate t he recovery i n such resi stant cases?
To enable this question to be answered, a conti nuous passive moti on device
has been devel oped that enables unl i mi ted cycl es of progressi ve end-range exer­
cise to be applied to the l umbar spi ne. REPEX ( Howard Wright Ltd, New
Pl ymouth, New Zealand) has now been well trialed cl i ni cal l y at four McKenzie
I nstitute Cl i nics in New Zealand. Wi th the advent of REPEX, a new concept
and dimension i n the treatment of spi nal therapy has emerged. Fi ndi ngs thus
far i ndi cate that REPEX i s as efective as the McKenzi e exercise protocols i n
resol vi ng certain mechanical di sorders of t he lower back, and is superior to
exercise i n certain other categories ( Fig. 6- 1).
REPEX enables fi ne incremental progressi ons to be made i n the applied
range of fexion or extension. The equipment del i vers 10 cycl es/mi n of end­
range motion, which in the case of patients wi th the derangement syndrome is
progressivel y i ncreased as reduction of the derangement i s achi eved. Di fferen­
tiation between dysfunction and derangement i s i mperative and such di fferentia­
tion must be made prior to the application of REPEX. The pati ent should be
assessed and the nature of the derangement determi ned using the McKenzie
assessment protocol .
188 Physical Therapy of the Low Back
Fig. 6- 1 . REPEX: repeated end-range passive movements.
REPEX in Acute and Subacute Low Back Pain
Cl i nical experience i ndicates that in patients wi th acute and subacute low
back pai n, use of REPEX accelerates t he rate of recovery signifcantl y. Central­
izati on, reducti on, or abol i ti on of symptoms is achieved i n one treatment ses­
sion more frequentl y and at a faster rate when using REPEX than i s possible
usi ng pati ent sel f-treatment methods al one.
Wi th the use of REPEX i t is possible to achieve centralization, reduction,
or abol i ti on of pain i n cases where patient-generated exercises carried out over
several days have fai led to produce change. We bel ieve i t is al so significant
that the frequency of appl ication of spinal manipulative thrust techniques by
staff at the McKenzie Institute cl i ni cs has further reduced si nce the advent of
REPEX has the abi l i ty to rapi dl y reduce derangement even when move­
ment is significantl y obstructed, such as occurs in patients fi xed in flexion with
acute l umbago. Prior to the advent of REPEX, curve reversal from kyphosis
to lordosis in this acute di sorder was achi eved sl owly over a period of time
usi ng careful positioning and mi l d-range movement as behavior of pain al lowed.
With the correct use of REPEX, i t i s now possible i n most cases to restore ful l
movement wi thi n 20 to 30 mi nutes .
A most signifcant fi ndi ng from the use of REPEX is the total absence of
beneficial effect on pai n or range of motion when patient motion is l i mited to
repetition i n midrange. Onl y when applied progressi vel y to end range does pain
reduce and motion i mprove.
Disorders of the Low Back 189
REPEX and Remodeling
The aim of treatment for pati ents wi th the dysfuncti on syndrome is to
remodel contracted or adherent ti ssue to a functi onal l ength, thus restoring
mobi l i ty to t he greatest extent possible consideri ng the nature of the in­
Cl i nical experience to date suggests that t he remodel i ng process
may be significantl y accelerated with REPEX.
The fi ne control of REPEX and the unl i mited number of del i very cycles
provide a remodel i ng process previ ousl y unobtainable by pati ent sel f-generated
exerci ses.
It is i mportant that spi nal segments restricted by contracture or fibrosi s
are not overstretched. REPEX has the capaci ty to overstretch, and treatment
of dysfunction shoul d be progressed slowly i n comparison to the rapi d progres­
sions applied to pati ents with the derangement syndrome.
Never before has a t herapy been available that appl ies unl i mi ted cycles of
end-range movement to the l umbar spine. REPEX is a potent tooL for the aLLevia­
tion of pain and the recover of function in mechanicaL spinal disorders. As
such it also has the potential to cause harm if used improperly.
The McKenzie I nstitute I nternational establ ished a resi denti al spi nal ther­
apy and rehabilitation center i n 1 990 for the treatment of chronic and recurrent
mechanical spinal di sorders. In l ight of our experience to date, we believe that
the defi ni tion of chroni c l ow back pai n shoul d be amended to specifi cal l y ex­
cl ude those patients who experience recurrent epi sodi c l ow back pai n. Large
numbers of pati ents in thi s category are inappropriately cl assified as havi ng
chronic low back pai n.
Low back pain chroni ci ty i mpl i es that those affected suffer more or l ess
continuous pain, and have few if any, pai n-free periods. The vi ew i s al so wi del y
held that these patients must learn to l i ve wi th thei r probl em and t he chances
for recovery are remote. The di agnosis of chronicity i mpl i es a hopel essness
that creates and perpetuates both physical and mental di stress.
We have found that a majority of pati ents referred to the rehabil itation
center, perhaps as many as 60 percent, do not have continuous l ow back pai n as
such. I nstead, they sufer recurring epi sodes so frequentl y that the i mpression i s
given that pain i s conti nuous. I n real i ty there may be some days or even weeks
when no pai n i s experienced. Because thei r symptoms are so frequent, patients
state that thei r symptoms are "always present . " Thus, i nstead of recei vi ng
appropriate treatment for recurring i nternal derangement (whi ch requires edu­
cation in sel f-reduction procedures), the pati ent i s referred to a chroni c pai n
Management strategies for patients with recurrent episodi c low back pain
must take i nto account the rapid reversi bi l i ty of derangement pathol ogies, and
must i ncl ude education i n procedures of self-reduction and self-treatment . Once
190 Physical Therapy of the Low Back
identified and treated appropriatel y, many of these patients are able to resume
normal act i vi ti es and manage thei r own probl em. Onl y those pat ients with truly
long-standing and unremi tting symptoms shoul d be placed in the chronic cat­
It does not seem appropriate to i nsi st that pati ents with recurrent derange­
ment shoul d be forced to undergo work-hardeni ng programs and "work through
t he pain. " No health provider woul d force such acti vi ty on a patient with an
internal l y deranged carti l age wi thi n the knee!
Some centers treating chronic pai n patients advise reduction of activity
l evel s by ei ther cutti ng out acti vi ty enti rel y, or restricting motion by attempting
to stabi l i ze segments of the spi ne during acti vi ty. The consequences of such
advice create i n the pati ent' s mind t he idea that acti vi ti es and movement are
harmful . Unfortunatel y, thi s advi ce is al l too frequentl y adhered to; chronic
back pain pati ents afraid of movi ng lose mobi l i ty, and soft tissues contract ,
joi nts stiffen, and the muscl es weaken. Thus, wi th progressive deterioration of
the spi nal muscul oskeletal system, segments above the affected area become
stiff and painful wi th neck and thoracic symptoms devel oping long after the
onset of l ow back pai n.
For pat ients wi th chronic low back symptoms who experience trul y con­
stant pain and who have contractures or fi brosis causing loss of function and
movement , we have devel oped a treatment and rehabilitation program in whi ch
REPEX is a pri me factor i n the remodeling process, the recovery of mobi l ity,
and reducti on of i nternal derangement . Rehabi l i tation fol l owing REPEX con­
sists of self-treatment procedures to mai ntain mobi l i ty and i nsti l l prophylactic
pri nci ples, and gymnasi um and general exercise programs to provide functional
and physical reacti vation.
Prel i mi nary fi ndings from an ongoing study by Wi l l iamsl2 show significant
benefits for pati ents wi th chronic low back pain.
Wi l l iams studied pati ents wi th chronic (> 3 months) mechanical spinal pain
who were off work and had been on earnings-related compensation for 3 mont hs
or longer and who were referred to the center by general practiti oners through­
out New Zealand. Pati ents i ni ti al l y completed a battery of questionnaires col­
l ecti ng data relat i ng to demographi cs, cl i nical history, pai n, functional status
( Dal l as Pain Questionnai re) , depression (Zung), somatic perceptions ( Modified
Somatic Percepti ons Questionnai re) , fear avoidance beliefs ( Fear Avoidance
Bel iefs Questionnaire), and pai n l ocus of control ( modified Mul tidimensional
Health Locus of Control ) . Physical examination fol lowed assessment of spinal
mobi l i t y, straight-leg rai se, neurologic deficit , and signs and symptoms of non­
organic i l l ness. Pati ents were then assessed accordi ng to the standard McKen­
zie protocol i ncl udi ng repeated end-range spi nal movements and static loading
tests to determi ne the directi on of therapeutic spi nal movement and appropriate
resti ng posi ti ons. Pati ents were prescribed i ndi vi dual ized exercise programs
i nvol vi ng pati ent-generated repeated end-range passive exercise and repeated
end-range passive exercise on REPEX equi pment. Patients also undertook ac­
ti ve strengtheni ng exercises and reacti vati on. Indi vidual i zed home exercise and
gymnasi um programs were prescribed for self-management fol l owing di s-
Disorders of the Low Back 1 9 1
charge. Patients were reassessed at exi t and fol lowups undertaken at 3 months
and I year.
Of 79 patients referred to the center, 15 were excl uded, 64 entered the
program, and 3 dropped out . Patients entering had a mean age of 37 years , a
mean current episode duration of 3 . 1 years , and had been off work an average
of 1 . 9 years . At di scharge, 62 percent reported a "marked i mprovement" or
better in their pain level , and overal l the group yi el ded a median value of 70
percent relief on a pai n relief analog scal e. At entry, the numbers of pati ents
report ing most di stal pai n location as the leg, the t high, or the l ow back were
35 , 16, and 10, respect i vel y. At discharge the correspondi ng val ues were 21 ,
1 5, and 25 (P < 0. 00 1). Pre- and posttreatment mean values for pai n i mpact on
the Dallas Pain Questionnaire were 57 and 33 percent, respecti vel y ( P < 0.00 1)
wi th each of the four subscales yi el di ng significant changes ( P < 0. 00 I ) . Signifi­
cant improvements in spi nal mobi l i ty ( 21 percent, P < 0.001) and straight-leg
raise (P < 0. 001 ) were also obtai ned. At entry, 7 pati ents were cl assified as
"normal , " 33 as "risk, " and 21 as "di stressed" according to the Di stress and
Risk Assessment Method. Corresponding figures at di scharge were 28, 22, and
I I . Patients exhibited significant mean reductions i n fears regarding t he effect
of acti vi ty on thei r pain ( P < 0. 00 I ) , a decreased chance l ocus of pai n control
(P < 0. 01 ) , and an increased i nternal l ocus of pain control (P < 0.00 1).
Wi l l i ams concluded that although the uncontrol led nature of thi s study
precluded conclusions as to the cause(s) of these changes, the l ength of ti me
these patients had had their symptoms and been off work suggested their condi­
tions were l i kel y to have reached rel ati vel y stabl e states. Al though other causes
could not be ruled out , the marked changes i n pain location and i ntensi t y,
function, and psychological stat us over t he 2-week period were consistent wi t h
McKenzie' s cl aim that repeated end-range spi nal movements were able to rap­
idly reverse underlying pathology. Follow-up assessments were being under­
taken to determine the stabi l i ty of these fi ndi ngs.
The McKenzie approach to the diagnosis of spi nal pain is often mi sunder­
stood. It does not treat all patients with extension as though it were the opposite
of Williams' fexion exercises. It is also a much more complex di sci pl i ne to
carry out than thi s brief description can convey.
The McKenzie system is based primaril y on pai n and its behavior, as wel l
as the presence of acute spi nal deformity , as the onl y di rect l i nks we have wi th
the unknown underl yi ng pathology. Meanwhi l e, t he large majority of acute and
chronic patients can be efecti vel y t reated and recurrences can be prevented
or qui ckl y resolved si mpl y by monitoring the pai n and its behavior duri ng a
comprehensive mechanical assessment that identifies and uses sel f-treatment
pri nciples to the ful lest extent for a wide range of pati ents.
To impact the low-back population and its hi gh cost to our soci ety, we
cannot remain content wi th l etting patients spontaneousl y recover as we di s-
192 Physical Therapy ofthe Low Back
pense unproven, passive modal i ti es whi l e at the same ti me cl ai ming credi t.
We al so cannot justify performing i nvasive i njection procedures, prescribing
medications and bed rest, or appl yi ng manual therapy to a population of patients
who can effecti vel y treat themsel ves.
The success of the McKenzie system is dependent on the training and
expertise of the physician and therapi st especial l y. Typical l y, 2 years of cl i nical
experience with assessment and treatment of mechanical spinal di sorders , cou­
pled with appropriate education and certificati on, is a mi ni mum requirement
for effecti ve education in thi s method.
Progressi ve repeated end-range passive exercise offers an exci ti ng tool for
the treatment of l ow back pain i n the future. Much research is required to
val i date i ts val ue and i dentify i ts most useful rol e. Earl y fi ndings, however,
and my own experience allow me to specul ate that this "devi ce" wi l l alter the
course of conservative therapy of the lower back. Only ti me will tel l .
I . McKenzie RA: Mechanical diagnosis and therapy for low back pai n: toward a better
understanding. p. 1 57. In Twomey LT, Taylor JR ( eds) : Cl i nics in Physical Therapy.
Physical Therapy of the Low Back. Churchi l l Livi ngstone, New York, 1 987
2. Landen BR: Heat and cold for the relief of low back pain? Phys Ther 47: 1 1 26, 1 967
3 . Kl ei n RG, Eek BC: Low energy laser and exercise for chronic low back pain:
double-blind controlled trial. Arch Phys Med Rehabil 7 1 : 34, 1 990
4. Deyo RA, Wal sh N, Martin D et al : A controlled trial of transcutaneous electronic
nerve sti mul ation (TENS) and exercise for chronic low back pai n. N Engl J Med
322: 1 627, 1 990
5. Spitzer WO, LeBlanc FE, Dupuis M et al : Scientific approach to the assessment
and management of activity-related spinal disorders. A monograph for clinicians.
Report of the Quebec Task Force on spi nal disorders. Spi ne 1 2: 75, 1 987
6. Nwuga YCB: Ult rasound in treatment of back pain resulting from prolapsed inter­
vertebral disc. Arch Phys Med Rehabil 64: 88, 1 989
7. Gnatz SM: I ncreased radicular pain due to therapeutic ul trasound applied to the
back. Arch Phys Med Rehabil 70:493, 1 989
8. Nachemson A: Cl i n Orthop Relat Res 279: 8, 1 992
9. Meade TW, Dyer S, Browne W et al : Low back pain of mechanical origin: random­
ised comparison of chi ropractic and hospital outpatient treat ment . Br Med J 300:
1 43 1 , 1 990
1 0. Yanharanta H, Sachs BL, Spivey MA et al : The relationship of pain provocation
to l umbar disc deterioration as seen by CT/discography study. Spine 1 2: 295, 1 987
1 1 . Yanharanta H, Guyer RD, Ohnmeiss DD et al: Disc deterioration in low back syn­
dromes: a prospecti ve, mul ticenter CT/discography study. Spine 1 3 : 1 345, 1 988
1 2 . Zucherman J , Derby R, Hsu K et al : Normal magnetic resonance imaging with
abnormal discography. Spi ne 1 3 : 1 355, 1 988
1 3 . Kornberg M: Discography and magnetic resonance imaging i n the diagnosis of lum­
bar di sc disruption. Spine 1 4: 1 368, 1 989
1 4. Butl er D, Trafimow J H, Anderson GBJ et al : Discs degenerate before facets. Spine
1 5: I I I , 1 990
Disorders of the Low Back 193
1 5 . Kusl ich SD, Ul strom RN, Mi chael CJ : The tissue origin of low back pain and scia­
tica: a report of pain response to tissue sti mul ation during operations on the l umbar
spine using local anesthesia. Orthop Clin North Am 22: 1 8 1 , 1 99 1
1 6. Smyth MJ , Wright V: Sciatica and the intervertebral di sc. An experimental study.
J Bone Joint Surg 40A: 1 40 1 , 1 958
1 7. Hi rsch C: The anatomical basis for low back pain. Acta Orthop Scand 33: 1 , 1 963
1 8. Mooney V: Uni versity of Arizona Symposium on Back Pai n: The present and the
future, Tucson, Arizona, March 1 1 - 1 4, 1 992
1 9. Jackson RP, Jacobs RR, Montesano PX: 1 988 Vol vo Award in Clinical Sciences.
Facet joint injection i n l ow-back pai n. A prospective statistical study. Spine 1 3: 966,
1 988
20. Lui l i us G, Laasoner EM, Myl lynen P et al : Lumbar facet joint syndrome. A random­
ized cl inical trial . J Bone Joint Surg 77B: 1 1 08, 1 989
2 1 . Dixon A: Diagnosis of low back pai n. In Jayson M ( ed) : The Lumbar Spine and
Back Pai n. Pitman Medical , Tunbridge Wel l s, England, 1 980
22. Potter NA, Rothstein J M: Intertester rel iabil ity for selected clinical tests of the
sacroiliac joint . Phys Ther 65: 1 67 1 , 1 985
23. Cyriax J: Textbook of Orthopaedic Medici ne. 5th Ed. Vol. I . Bai l l i ere Tindal l ,
London, 1 980
24. Bogduk N: The causes of low back pai n. Med J Aust 1 56: 1 5 1 , 1 992
25. Bogduk N: The sources of low back pai n. In Jayson MI V ( ed) : The Lumbar Spine
and Back Pai n. 4th Ed. Churchi l l Livingstone, Edinburgh, 1 992
26. Suzuki N, Seiichi E: A quantitative study of trunk muscle strength and fatigabil ity
in the low back pain syndrome. Spine 8: I , 1 983
27. Twomey L, Taylor J : Age changes i n the l umbar intervertebral discs. Acta Orthop
Scand 56: 496, 1 985
28. Coventry MB, Ghormley RK, Kernohan JW: The intervertebral disc: part I I .
Changes i n the intervertebral disc concomitant with age. J Bone Joint Surg 27: 233,
1 945
29. Coventry MB, Ghormley RK, Kernohan JW: The intervertebral di sc: part I I I . Path­
ological changes i n the intervertebral disc. J Bone Joint Surg 27: 460, 1 945
30. Vernon-Roberts B: The pathology and interrelation of intervertebral disc l esions.
In Jayson M ( ed) : The Lumbar Spine and Back Pain. 2nd Ed. Pitman Medical ,
Tunbridge Wel l s, England, 1 980
3 1 . Schiotz EH, Cyriax J : Manipulation past and present. Hei nemann, London, 1 975
32. Brunarski DJ : Cl i nical trials of spinal manipul ation. J Manip Phys Ther 7: 4, 1 984
33. Doran DML, Newel l DJ : Manipulation i n treatment of low back pai n: a mul ticenter
study. Br Med J 2: 1 6 1 , 1 975
34. Farrell J B, Twomey L T: Acute low back pai n. Comparison of two conservative
treatment approaches. p. 1 62. Proceedings of Manipulative Therapists Association
of Australia, Perth, Western Austral i a, 1 983
35 . Hadler NM, Curtis P, Gi l l ings DB, Stinnett S: A benefit of spinal manipulation as
adjunctive therapy for acute low back pain: a stratified, controlled trial . Spi ne 1 2:
7, 1 987
36. Hoehler FK, Tobis J S, Buerger AA: Spinal manipulation for low back pai n. J AMA
245: 1 835, 1 98 1
37. Sims-Wi l l iams H, Jayson MI V, Young SMS: Control l ed trial of mobilisation and
manipulation for patients with low back pain in general practice. Br Med J 2: 1 338,
1 978
38. AssendeLft WJJ , Bouter LM, Kessels AGH: Effectiveness of chi ropractic and phys-
1 94 Physical Therapy of the Low Back
iotherapy i n the treatment of low back pai n: a cri tical discussion of the British
randomized cl inical trial . J Manip Phys Ther 1 4: 28 1 , 1 99 1
39. Nachemson A: Conservative treatment of low back pai n. I n Jayson M (ed) : The
Lumbar Spine and Back Pai n. 2nd Ed. Pitman Medical , Tunbridge Wel l s, England,
1 980
40. McKenzie RA: The Lumbar Spine. Mechanical Diagnosis and Therapy. Spinal Pub­
lications Limited, Waikanae, New Zealand, 1 98 1
4 1 . McKenzie RA: The Cervical and Thoracic Spine. Mechanical Diagnosis and Ther­
apy. 1 st Ed. Spinal Publications ( N. Z. ) Limited, Waikanae, New Zealand, 1 990
42. Donelson R, Murphy K, Sil va G: Central ization phenomenon: its usefulness in eval­
uating and treating referred pai n. Spine 1 5 : 3, 1 990
43 . Di Maggio A, Mooney V: Conservative care for low back pai n; what works? J Muscu­
loskel Med 4: 27, 1 987
44. Di Maggio A, Mooney V: The McKenzie program: exercise effective against back
pai n. J Muscul oskel Med 4: 63, 1 987
45. Van Wijmen PM: Lumbar pain syndromes. p. 442. In Grieve GP ( ed) : Modern
manual therapy of the vertebral col umn. Churchil l Livingstone, New York, 1 986
46. McKenzie RA: Treat Your Own Back. Spinal Publications, Lower Hutt, New
Zealand, 1 98 1
47. McKenzie RA: Treat Your Own Neck. Spinal Publications, Lower HUl l , New
Zealand, 1 983
48. McKenzie RA: A physical therapy perspective on acute spinal disorders. p. 2 1 1 .
I n Mayer TG, Mooney V , and Gatchel RJ ( eds) : Contemporary Conservative Care
for Painful Spinal Disorders: Concept s, Diagnosis and Treatment. Lea & Febiger,
Mal vern, PA, 1 99 1
49. Stevens BJ, McKenzie RA: Mechanical diagnosis and sel f treatment of the cervical
spine. In Grant R ( ed) : Physical Therapy of the Cervical and Thoracic Spine. Church­
ill Li vi ngstone, New York, 1 988
50. Donelson R: The McKenzie approach to eval uating and treating low back pai n.
Orthop Rev XI X:68 1 , 1 990
5 1 . Donelson RG, McKenzie R: Mechanical assessment and treatment of spinal pain.
p. 1 627. In Frymoyer JW (edt): The Adult Spine: Principles and Practice. Vol. 2.
Raven Press, New York, 1 99 1
52. Donelson RG: Identifyi ng appropriate exercises for your low back pain patient . J
Muscul oskel Med 8: 1 4, 1 99 1
53. Nelson MA, Al l en P , Cl amp S et al : Rel iabil ity and reproducibility of cl inical fi ndings
in low-back pai n. Spine 4: 2, 1 979
54. Matyas TA, Bach TM: The rel iabil ity of selected techniques in cl inical arthrome­
trics. Aust J Physiother 3 1 : 1 75, 1 985
55. Gonnel l a C, Paris S, Kutner M: Reliability in evaluating passive intervertebral mo­
tion. Phys Ther 62: 437, 1 982
56. Shepperd JAN: Patterns of internal disc dynamics, cadaver motion studies. Video
presented at the International Society for the Study of the Lumbar Spine Meeting,
Boston, June 1 3-1 4, 1 990
57. Cloward RB: Cervical di skography: a contribution to the etiology and mechanism
of neck, shoulder and arm pain. Ann Surg 1 50: 1 052, 1 959
58. Murphey F: Sources and patterns of pain i n disc disease. Cl i n Neurosurg 1 5: 343 ,
1 968
59. Adams MA, Hutton WC: Prolapsed intervertebral di sc: a hyperflexion injury. Spine
7: 3, 1 982
Disorders of the Low Back 195
60. Adams MA, Hutton WC: Gradual disc prolapse. Spine 1 0: 6, 1 985
6 1 . Taylor JR, Twomey L T: The lumbar spine from infancy to old age. p. I . I n Twomey
L T and Taylor JR (eds) : Cl inics in Physical Therapy. Physical Therapy of the Low
Back. Churchi l l Livingstone, New York, 1 987
62. Farfan HF: Mechanical Disorders of the Low Back. Lea & Febiger, Phi l adel phi a,
1 973
63 . Panjabi M, Krag MH, Chung TQ: Effects of disc injury on mechanical behaviour
of the human spine. Spine 9: 7, 1 984
6. McKenzie RA: Prophyl axis i n recurrent low back pai n. NZ Med J 89: 627, 1 979
65 . McKenzie RA: Manual correction of sciatic scoliosi s. NZ Med J 76:484, 1 972
66. Cherkin D: Pilot randomized trials of physical and educational interventions for
low back pain. p. 2. BOAT Newsl etter. Vol . 2. University of Washington, Seattl e,
Washington, June 1 992
67. The Backletter 7: 7, 1 992, Skol Publ ishi ng, Madison, New Hampshire
68. Ponte DJ , Jensen GJ , Kent BE: A preliminary report on the use of the McKenzie
protocol versus Wil l iams protocol in the treatment of l ow back pai n. J Orthop Sports
Phys Ther 6: 1 30, 1 984
69. Nwuga G, Nwuga V: Relative therapeutic efficacy of the Wi l l iams and McKenzie
protocols in back pain management . Physiother Pract 1 : 99, 1 985
70. Kopp JR, Alexander AH, Turocy RH et al : The use of l umbar extension i n the
evaluation and treatment of patients with acute herniated nucl eus pul posus. A pre­
limi nary report. Cl i n Orthop 202: 2 1 1 , 1 986
7 1 . Wil l iams MM, Hawley JA, McKenzie RA, Van Wij men PM: A comparison of the
effects of two sitting postures on back and referred pai n. Spine 1 6: 1 1 85 , 1 99 1
72. Alexander AH, Jones AM, Rosenbaum J r DH: Nonoperative management of her­
niated nucleus pulposus: patient selection by the extension sign-long term fol l ow­
up. Presented to North American Spine Society Annual Meeting, Monterey, Cal ifor­
nia, August 8- 1 1 , 1 990
73 . Stankovic R, Johnel l 0: Conservative treatment of acute low-back pai n. A prospec­
tive randomized trial : McKenzie method of treatment versus patient education in
"Mini Back School . " Spine 1 5 : 2, 1 990
74. Spratt KF, Lehmann TR, Weinstein IN, Sayre HA: A new approach to the low­
back physical examination. Behavioral assessment of mechanical signs . Spine 1 5 :
2, 1 990
75 . Roberts AP: The conservative treatment of low back pai n. Thesi s, Notti ngham,
1 990
76. McKinney LA: Early mobi l i sation and outcome in acute sprains of the neck. Br
Med J 299: 1 006, 1 989
77. Elnaggar 1 M, Nordin M, Sheikhzadeh A et al: Effects of spinal fexion and extension
exercises on low-back pain and spinal mobility i n chronic mechanical low-back pain
patients. Spine 1 6: 967, 1 99 1
78. Vanharanta H, Videman T, Mooney V: Comparison of McKenzie exercises, back
trac and back school in lumbar syndrome; preliminary resul ts. Annual Meeting of
International Society for the Study of the Lumbar Spine, Dal l as, Texas, May
28-June 9, 1 986
79. Donelson RG, Grant WD, Kamps C, Medcalf R: Low back and referred pain re­
sponse to mechanical l umbar movements i n the frontal pl ane. Presented at I nterna­
tional Society for the Study of the Lumbar Spine Meeting, Heidel berg, May 1 2- 1 6,
1 99 1
80. Rath WW, Rath J ND, Duffy CG: A comparison of pain location and duration with
1 96 Physical Therapy of the Low Back
treatment outcome and frequency. Presented at First I nternational McKenzie Con­
ference, Newport Beach, Cali fornia, Jul y 20-2 1 , 1 989
8 1 . Riddl e DL, Rothstein JM: I ntertester rel iabil ity of McKenzi e' s cl assifications of the
syndrome types present i n patients with low back pai n. Spine 1 8: 1 333, 1 993
82. Wil liams M, Grant R: Effects of a McKenzie spinal therapy and rehabilitation pro­
gramme: prel i minary findings ( abstract) . Presented at The Society for Back Pain
Research, Annual Scientific Meeting, London, October 30, 1 992
83. Anderson J , Boumphrey F, Bel l G: Early decompression in cauda equina syndrome.
Presented at the I nternational Society for the Study of the Lumbar Spine Meeting,
Chicago, May 20-24, 1 992
84. McI lwain WA: A technique for improving results of soft tissue repair in low back
patients util i sing the LPM 1 00. 1 99 1 ( i n press)
7 Clinical Assessment:
The Use of Combined
Movements in
Assessment and
Brian C. Edwards
The physical examination of the lumbar spine requires a high degree of
skill combined with a thorough knowledge of structure and function, particu­
larly when the treatment to be prescribed is manipulative therapy. This is be­
cause very close attention needs to be paid to the way in which the patient's
signs and symptoms react to relatively small changes in posture and movement.
If movement is the preferred method of treatment, the more precise the informa-'
tion on the reaction of signs and symptoms to movement the more specific and
effective the passive movement (manipulative procedure) will be.
However, it is important to expand the usual orthopedic examination by
including movements that will highlight signs and symptoms that might normally
be masked. This involves combining movements, which will increase or de­
crease the stretch and compressive effects on structures associated with the
joint (e.g. , the capsule, ligaments, and muscle attachments). This enables the
therapist to establish movement patterns that assist not only in the choice of
the movement techniques to be used but also in the progression of treatment
and prognosis.
198 Physical Therapy of the Low Back
A standard orthopedic examination of the lumbar spine initially involves
history taking. At the outset an accurate account of the symptoms and their
distribution must be obtained. This can be achieved in a number of ways. The
patient can be given a simple body chart (Fig. 7-1) and asked to draw in the areas
of pain or discomfort. However, such a method usually requires the therapist to
repeat the examination in more detail, which can be very time consuming. A
useful compromise is to ask the patient to map out on the body chart using one
finger the areas of paresthesia and anesthesia, including a description of the
type and depth of pain. In the case of the lumbar spine particularly, specific
questions need to be asked related to perineal anesthesia or paresthesia alter­
ation in micturition or bowel habit (particularly if associated with severe pain)
and the responses require careful assessment. Care must be taken at this point
to explain to the patient that it is important and necessary to describe the
symptoms experienced at the time of assessment. If the distribution of symp­
toms has changed, the change can be superimposed on the same diagram or a
separate diagram can be used. This is an important aspect of the examination
since a clear understanding of those symptoms that are currently experienced
as compared to those originally experienced can have an important bearing on
both the diagnosis of the condition and selection of the treatment techniques.
Once descriptions of the symptoms and their distribution have been estab­
lished, the patient should be questioned about how symptoms in different areas
Deep ache
( caostant)
Only when
Fig. 7-1. A body chart used to show distribution of symptoms in a patient.
Combined Movements in Assessment and Treatment 199
are related to each other (e.g., has there been an increase in pain in one area with
a corresponding change in another area? In what sequence did the symptoms
originally appear?). Such questioning is often omitted in history taking. How­
ever, the answers provided by the patient highlight and implicate the anatomic
structures likely to be affected, which may be the cause of the patient's symp­
toms. They can also help in identifying the level of involvement of referred
symptoms when it comes to the objective examination.
The constancy of a patient's symptoms and the variability in the intensity
of pain are also important aspects of the history taking and must be established.
Following this, activities that cause any change in symptoms should be noted
(i. e., the ease with which symptoms are aggravated; the activities that cause
this aggravation; the relation between the type of activity and the duration and
intensity of the symptoms produced, sometimes called irritability). These are
often a useful guide to the amount of physical examination or treatment that
may be carried out on the first day.
Activities that aggravate, relieve, or do not affect the symptoms need to be
carefully described and analyzed in relation to the anatomy and biomechanics of
the vertebral column, and to the distribution of the patient's symptoms. The
simple activity of digging in the garden may be performed quite differently by
two patients.
Once aggravating and easing factors have been identified, it is useful to
evaluate which particular combined movements are being performed during the
activity. This is especially important in the case of a patient with a severe or
irritable condition, when it is important to identify a pain-easing position. If
sitting, standing, or lying positions aggravate or relieve the symptoms, this
should be carefully noted. Particular attention should be paid to the positions
adopted by the patient at the time of examination and treatment.
24-Hour Variation
The history of present and past attacks of 10' back pain in terms of the
type of activity responsible (if any) and the mode of onset of the symptoms
needs to be clearly described by the patient and noted by the physical therapist.
The onset of the symptoms is frequently related to and may have resulted from
a particular incident or activity. However, it is not unusual for patients to have
difficulty in remembering the particular incident, as they may regard it as trivial
or it may have occurred some time before the onset of the symptoms. I
The need for accurate questioning and skill in the interpretation of answers
cannot be overemphasized. It is essential to have this information as a sound
base before proceeding to the next step of the objective examination. The inter­
pretation of the subjective examination also requires considerable patience and
skill on the part of the therapist.
200 Physical Therapy of the Low Back
The term objective examination is something of a misnomer. Objectivity
in its pure form is diffcult to achieve when the physical therapist includes in
such examination not only movements but also the patient's description of the
symptoms reproduced by the movements. The objective examination therefore
contains some elements that are subjective in the sense that patient response
requires interpretation by the therapist. It is important that reference is con­
stantly made to the specific areas of pain and the symptoms for which the
patient has come seeking treatment. It must be emphasized that attention to
small details of a patient's answers and to individual movements is essential if
the objective examination is to help identify particular structures as likely
sources of the patient's symptoms.
The principal aim of the objective examination is to establish the effect
of movement of the lumbar spine on those symptoms that have already been
described by the patient. In doing this the identification of the muscles, joints,
and ligaments involved in the patient's disorder is of primary importance. Care­
ful observation of the way the vertebral column moves areas of hyper- and
hypomobility, and areas of relative muscle hypertrophy or atrophy are as­
The first part of the examination consists of observation. Three important
aspects of observation are general movement, posture and shape of joints, and
General Movement
Observation of the care with which a patient moves while adopting the
sitting position or moves out of such a position, and how the patient moves
while disrobing, with any changes in facial expression, assists the interpretation
of the patient's symptoms. A pertinent question to be asked when a particular
posture produces pain is "Is it 'the' pain or is it different from the pain for which
you are seeking treatment?" Such observations may suggest to the therapist the
movements that are likely to reproduce the symptoms.
Posture and Shape of Joints
Alteration in posture and joint outline may be of recent or long-standing
duration and include many so-called postural deformities that may well be per­
fectly normal for a particular individual. It should be emphasized that some
Combined Movements in Assessment and Treatment 201
fairly obvious deformities (e.g., marked kyphosis, lordosis, or scoliosis) may
be of no significance in the patient's current problem.
Obvious gait alterations can be observed initially. Changes such as altered
weight distribution and lack of mobility in hips, knees, or ankles, or a positive
Trendelenburg sign may be noted. With the exception of the latter, lack of
mobility may be inhibition of movement due to pain originating in the lumbar
spine, or due to a previous unrelated peripheral joint involvement.
At this time, it is important to compare the symptoms and signs produced
on movement to the answers given to the related subjective questions. Thus
links may be established between the symptoms described by the patient and
those elicited by the movement. The patient needs to be undressed sufficiently
to observe the whole of the spine as well as the lower limbs.
Observation from Behind
When observing from behind the following may be observed and variation
I. Altered leg length
2. Altered shoulder height
3. Position of head on neck and neck on shoulders
4. Kyphosis or lordosis (exaggerated or diminished)
5. Position of scapulae
6. Valgus or varus deformity of knees and feet
7. Scoliosis (postural or structural)
8. Position of sacrum and iliac crests
9. Prominence or depression of vertebral spinous processes
10. Skin contour and color
Observation from the Front
When observing from the front, a clinician should take note of the following
I. Height or level of iliac crests
2. Position or level of the knees
202 Physical Therapy of the Low Back
3. Shape of trunk
4. Relative position of shoulders, head, and feet
5. Skin contour
Observation from the Side
When observing from the side, a clinician should be aware of the following:
I. Position of head
2. Shape of cervical, thoracic, and lumbar spinal curves (any increased or
decreased kyphosis or lordosis)
3. Skin contour
The lumbar spine is most easily examined from behind.
When suitably undressed the patient is asked first to describe exactly where
the symptoms are at present. The patient is then asked to benrl forward to
where there is any increase in any part of the symptom complex. This flexion
range is usually recorded by measuring the distance from the outstretched fin­
gertips to the floor or in relation to the position of the fingertips on the legs
(e.g., patella, midthigh). The patient is then asked to move in a controlled
manner further into the painful range (Fig. 7-2). This range and any alteration
in the symptoms are noted.
In addition to noting the full range of movement, considerable attention is
paid to the way in which the individual vertebrae move during fexion. Areas
of hyper- and hypomobility, as well as any deviation from the median sagittal
plane, are recorded. Surface contour should be carefully considered, particu­
larly noting areas of prominence or depression. At this time it is important to
compare those symptoms and signs produced on flexion to those answers given
to the related subjective questions. Thus any links may be established between
the symptoms described by the patient and those elicited by the movement.
The distribution of the symptoms and the range of movement needs to be very
carefully recorded. In one patient buttock pain may be produced during the
first 10° of fexion, but the patient may also be able to continue to full range
without any alteration in the distribution of the pain. In another patient, buttock
pain may be produced in the first 10° of flexion but on continuing the movement
toward full range, the pain may progress to the calf. Both patients have the
same range of forward flexion but they produce quite difrent symptoms, which
need to be treated quite differently.
Combined Movements in Assessment and Treatment 203
Fig. 7-2. Part I of the general
examination: fexion.
The effect of a controlled amount of overpressure (i.e., gentle passive forc­
ing of the movement from the patient's end range further into range) is also
necessary under certain circumstances, not only to observe the way the symp­
toms react but also to test the end feel of the physiologic movement. The end
feel of a movement is the relationship between the pain experienced and the
resistance to movement. Such resistance may be due to intrinsic muscle spasm
or tightness of the ligaments and capsule of the joint.
The end feel of the physiologic movement, mentioned above, may be differ­
ent from the end feel with localized passive movement procedures (described
later). However, quite distinct solid, springy, soft, or hard end feelings may be
distinguished. The end feel needs to be noted because if there is a difference
between what is found with localized procedures compared to the more general
movement procedures, then an attempt needs to be made to define those differ­
ences and the possible reasons for them.
On occasion it is important to hold the full range position of flexion for a
period of time. This becomes a necessary part of the examination, especially
if during the subjective examination and activity involving sustained flexion is
204 Physical Therapy of the Low Back
reported by the patient as a position where symptoms are eased. Such a proce­
dure is valuable because if the symptoms are not eased, more detailed question­
ing and examination is necessary.
The return from the flexed to the upright position is also an important
movement to monitor, both in terms of the way the vertebral column moves
and the production of symptoms. A postural scoliosis or tilt may become on
adoption of the erect position, which is not evident on bending forward. Another
important aspect of assessment of flexion and the return to the upright position
is the reproduction of a painful arc; that is, pain that is produced through a
part of the range and then is eased as the movement continues. This can happen
either during the fexion movement or during the return to the upright position.
The range within which such symptoms are produced as well as the distribution
of the symptoms should be carefully recorded and related if possible to the
subjective findings. Often those patients with painful arcs are slower to respond
to treatment, particularly if the painful arc is variable in its position in the range.
On occasion, symptoms may be produced some time after the movement
has been completed (i.e., latent pain). This latent pain possibly may have a large
chemical/infammatory component in its production or etiology. Occasionally
repeated flexion movements or varying the speed of the movement may be
necessary to reproduce this symptom.
In addition to the general observation of changes in signs and symptoms
on full range flexion, particular consideration needs to be given to the way in
which motion segments are moving on fexion. There is a cephalad movement
of the inferior zygapophyseal facets at one level in relation to the superior facets
of the level below. This is accompanied by a stretching of the sof tissues of
the posterior elements of the motion segment, including the posterior parts of
the disc and the canal structures, as well as the posterior ligaments, capsules,
and muscles. There is an accompanying compression of the anterior parts of the
disc (nucleus and anterior annulus including the anterior longitudinal ligament).
Lateral Flexion
The patient stands in the same position as for fexion and is asked to slide
the hand down the lateral aspect of the leg. Measurement is usually taken of
the distance from the fingertips to the head of the fibula. Areas of hypo- and
hypermobility can be observed at segmental levels by closely matching the
movement behavior of the vertebrae, comparing the relative movement of the
motion segment(s) with those above and below, and their sideways movement
behavior compared one to the other. As with flexion, the use of overpressure
and repeated and sustained movements may be necessary, in addition to the
observance of deformity and presence or absence of a painful arc.
The effect of lateral fexion on that part of the motion segment on the
contralateral extended side (away from that which the movement is performed
on) is similar to that observed with sagittal plane flexion. There is a cephalad
movement of the inferior zygapophyseal facet of the superior vertebra on the
Combined Movements in Assessment and Treatment 205
Fig. 7-3. Part I of the general ex­
amination: right lateral flexion.
superior zygapophyseal facet of the vertebra below. This stretches the soft­
tissue structures (e.g., discs, joint capsules,) on the side opposite the direction
of lateral flexion. On the side toward which the lateral flexion is performed
these structures are compressed (Fig. 7-3).
The therapist stands behind the patient, who is asked to bend backward.
Measurement can be made of the distance the fingertips pass down the posterior
aspect of the thigh. Areas of hypo- and hypermobility are observed as well as
the distribution of symptoms at the end of range and through range. Overpres­
sure and repeated and sustained movements are used as necessary.
The effect of extension on the motion segment is such that there is a caudal
movement of the inferior zygapophyseal facets of a vertebra on the facets of
206 Physical Therapy of the Low Back
Fig. 7-4. Part 3 of the general
examination: thoracolumbar ex­
the vertebra below. There is also a compression of the posterior parts and a
stretching of the anterior parts of the intervertebral disc (Fig. 7-4).
Axial Rotation
Rotation as a testing procedure in the lumbar spine is not a movement that
often produces significant alteration in signs and symptoms. Strangely enough,
it is a movement often preferred as a passive movement treatment technique
by therapists. One method of testing rotation is for the therapist to stand on
the lef side of the patient and take hold of the patient's right ileum with the
right hand and the patient's right shoulder with the left. The patient's pelvis is
then rotated to the right while applying counterresistance to the shoulder (Fig.
Comhined Movements in Assessment and Treatment 207
Fig. 7-5. Part 4 of the general
examination: lumbar rotation to
the left.
General Assessment of Standard Active Physiologic Movements
In addition to recording the ranges of movements that are available and
the way in which the vertebral segments move, detailed attention must also be
given to the distribution of the symptoms and the type of symptoms involved
with each movement. The importance of the patient's descriptions of these
symptoms cannot be overemphasized. Comparison needs to be made between
the way the patient describes the symptoms produced with various activities
as elicited during the sUbjective questioning to those produced and observed
in the objective testing of lumbar movements mentioned above. Similar descrip­
tions of type of symptoms as well as distribution are important (i.e., the pain
may be described as diffuse, lancinating, or referred to a limb, etc.). At this
stage, the therapist should look for similarities between the movements of gen­
eral daily activities that bring on the symptoms and the active movements that
208 Physical Therapy (�l the Low Back
elicit the pain. For example, a patient may report that bending activity in the
garden for about I hour brings on back pain, while on subsequent examination
one repetition of forward fexion is shown to produce the same pain. Careful
questioning is required to define the type and distribution of the symptoms,
because sustained flexion (if that is the movement adopted while gardening) is
unlikely to produce the same quality and quantity of pain as would be produced
by one movement of flexion. Such careful questioning can help both in diagnosis
as well as the selection of a treatment technique.
Combined Movements
Habitually, movements of the vertebral column occur in combination
across planes rather than as pure movements in one plane only. Many aspects
of this phenomenon have been investigated.2-6 Gregerson and Lucas7 found
that axial rotation of the lumbar spine to the left accompanied lateral flexion
to the lef and rotation to the right accompanied lateral fexion to the right.
However, in one case they found the reverse combination occurred. Stoddard5
stated that the direction of this conjoined rotation varied, depending on whether
the lateral flexion was performed with the lumbar spine in flexion or extension.
He suggested that the conjoined rotation is to the same side when the movement
of lateral flexion is performed in fexion and to the opposite side when the
movement of lateral flexion is performed in extension. Kapandji8 also stated
that lateral rotation occurs in conjunction with lateral flexion. Personal labora­
tory observation on fresh human cadaveric specimens seems to indicate that
the direction of axial rotation is in the opposite direction to that to which the
lumbar spine is laterally flexed, regardless of whether the spine is in fexion or
extension. The presence of degenerative processes within the disc or zygapo­
physeal joints affects the amount of rotation, and occasionally there is an unex­
plained apparent reversal of the axial rotation.
Because of this, the usual objective examination of the lumbar spine should
be expanded to incorporate combined movements. This is because symptoms
and signs produced by lateral flexion, flexion, extension, and rotation as pure
movements may alter when these movements are performed in a combined
Lateral Flexion in Flexion
The therapist stands on the right-hand side of the patient so that the thera­
pist's right anterosuperior iliac spine is in contact with the lateral aspect of the
patient's right hip. The therapist's right hand is placed over the posterior aspect
of the patient's left shoulder. The therapist's lef hand grips the patient's left
ilium. The patient is asked to bend forward: the range at which the symptoms
Combined Movements in Assessment and Treatment 209
Fig. 7-6. Part I of the combined
movements examination: lateral
fexion in flexion.
are reproduced is noted and while this position is maintained, lateral fexion
to the right is included as part of the total pattern (Fig. 7-6).
Lateral Flexion in Extension
The therapist stands on the right-hand side of the patient, with the right
arm placed around the patient's chest so that the therapist's right hand grips
the patient's left shoulder. The thumbs and index finger of the therapist's lef
hand are placed over the transverse process of the vertebral level to be exam­
ined. The patient is then bent backward and laterally flexed to the right (Fig.
210 Physical Therapy of the Low Back
Flexion and Rotation
Fig. 7-7. Part 2 of the combined
movements examination: lateral
flexion in extension.
The therapist stands on the right-hand side of the patient and places the
hands posteriorly on the patient's shoulder. The patient then bends forward
and rotates to the right (Fig. 7-8).
Extension and Rotation
The same hand positions are adopted as with extension and lateral fexion.
The patient's lumbar spine is extended and rotated to the right as well (Fig. 7-
Passive Accessory Movements in Combined Positions
The usual accessory movements of transverse, central, and unilateral pres­
sure9•IO may also be carried out in combined positions. The lumbar spine is
placed in the combined positions described above and the appropriate accessory
movements are performed.
Combined Movements in Assessment and Treatment 211
Fig. 7-8. Part 3 of the combined
movements examination: flexion
and rotation.
Central Vertebral Pressure in Extension and Right Lateral
The patient lies prone in a position of extension and right lateral fexion.
Central vertebral pressure is applied over the spinous process. Then the pres­
sure is directed caudad on the spinous process of L4; the compressive effect
particularly on the right-hand side will be increased on the right between L4-L5
but decreased between L3-L4 (Fig. 7-10).
Transverse Pressure to the Left in Flexion and Lef Lateral
With the patient prone and in a position of flexion and left lateral flexion,
transverse pressure over the right side of the spinous process is applied. This
will tend to increase the stretching effect on the right (Fig. 7-11).
212 Physical Therapy of the Lol Back
Fig. 7-9. Part 4 of the combined
movements examination: exten­
sion and rotation.
Unilateral Pressure on the Right in Flexion and Lef Lateral
With the patient positioned in flexion and right lateral fexion, unilateral
pressure on the right over the transverse process of the vertebrae is applied.
If this pressure is directed cephalad on the transverse process of L4 there will be
an increase in the stretching effect on the right between L4-L5 and a decrease
between L3-L4 (Fig. 7-12).
Passive Testing of Physiologic Movements
Standard passive physiologic tests of the movements of flexion, extension,
lateral fexion, and axial rotation can also be carried out and are a useful adjunct
to the examination procedures.
Combined Movements in Assessment and Treatment 213
Fig. 7-]0. Passive accessory movements in part I of the combined positions examina­
tion: central vertebral pressure in extension and right lateral fiexion.
Fig. 7-11. Passive accessory movements in part 2 of the combined positions examina­
tion: transverse pressure to the left in fiexion and left lateral fiexion.
214 Physical Therapy of the LOlli Back
Fig. 7-12. Passive accessory movements in part 3 of the combined positions examina­
tion: unilateral pressure on the right in fexion and left lateral flexion.
Assessment of Combined Movements
The movements of flexion, extension, lateral flexion, and rotation per­
formed in the neutral position are termed primary movements. On the initial
examination it is usual for one of these primary movements to reproduce part
or all of the symptoms of which the patient is complaining. It is to this primary
movement that the other movements are added. It is essential when performing
the primary movement that the movement is taken to the point where the symp­
toms begin, and then taken in a controlled manner further into the painful range
of the movement. At this stage the movement of lateral flexion or rotation or
both is added.
Care must be taken when adding other movements that the starting position
into the range of the primary movement is not altered. The basic principle is
to combine movements that have similar mechanical effects on the motion seg­
ment and to observe if symptoms are increased or decreased by such maneu­
vers. On flexion there is a cephalad movement of the inferior zygapophyseal
facet, for instance of L4 on the superior zygapophyseal facet of L5. The poste­
rior elements are stretched: the posterior part of the intervertebral disc, the
posterior longitudinal ligament, the ligamentum flavum, and capsules of the
zygapophyseal joints. The anterior structures are compressed.
With right lateral flexion, for example, the left interior zygapophyseal facet
of L4 moves upward on the left superior zygapophyseal facet of L5. This move­
ment produces a stretching of the elements on the left side of the motion segment
with completion of the right side of the motion segment. When flexion and right
lateral flexion are combined, the stretching effects on the left are increased and
slight stretching effects are produced on the right.
Combined Movements in Assessment and Treatment 215
With extension there is a downward movement of the inferior zygapophy­
seal facet (e.g., of the L4 vertebrae on the superior zygapophyseal facet of L5).
This is accompanied by compression of the posterior elements of the motion
segment. When the movement of right lateral flexion is combined with exten­
sion, there is an increase on the compressive effects on the right and a decrease
on the left.
Regular and Irregular Patterns
The combination of these movements can, under many circumstances,
produce recognizable symptom patterns, which may be described as regular
or irregular. I Regular patterns are those in which similar movements at the
intervertebral joint produce the same symptoms whenever the movements are
performed. The symptoms, however, may differ in quality or severity. The
regular pattern may be divided further: ( I) regular patterns: stretch and com­
pressive; (2) irregular patterns: those that show no recognizable pattern.
Regular Compressive Pattern. Right lateral flexion increases right buttock
pain. This pain is made worse when right lateral flexion is performed in exten­
sion, and eased when right lateral flexion is performed in flexion. Therefore,
if the patient's symptoms are produced on the side to which the movement is
directed, then the pattern is a compressing pattern.
Regular Stretch Pattern. Right lateral flexion increases left buttock pain.
This pain is made worse when right lateral flexion is performed in flexion and
eased when right lateral flexion is performed in extension. Therefore, if the
symptoms are present on the opposite side from that to which the movement
is directed, then the pattern can be considered a stretching pattern.
Irregular Pattern. All of those patients who do not fit into the regular
category are classified as irregular. There are many of these (e.g., when right
lateral fexion reproduces right buttock pain, which is made worse when right
lateral fexion is combined with fexion on the right. In this movement, the right
compression of right lateral flexion is counteracted by the stretching of flexion).
Another example is left lateral flexion producing right buttock pain (a stretching
movement). This pain is made worse when the same movement is performed
in extension (a compressing movement) and eased when the movement of left
lateral flexion is performed in flexion (another stretching movement).
The irregular pattern may indicate that there is more than one component
to the joint disorder (e.g., zygapophyseal joint and intervertebral disc, canal
and foranimal structures). Generally traumatic injuries (e.g., whiplash and early
disc lesions) have irregular patters, whereas chronic disc lesions or zygapo­
physeal joint lesions with no history of trauma leave regular patterns.
Contained within one type of pattern there may be elements of other pat­
terns. These should be recognized, and time provided so as to decide whether
any irregular patters have some recognizable, regular components.
The recognition of diferent patterns can assist in choosing the direction
of technique, allowing a combined movement end-of-range (CME), procedure
to be used, and assessing the manner in which the signs and symptoms may
216 Physical Therapy of the Low Back
When choosing the direction of the technique in the case of regular patterns
(either stretch or compressive), the first choice of movement is in the direction
that is away from the movement that reproduces the symptom. For example,
if right lateral flexion reproduces the right buttock pain and this pain is made
worse when the right lateral flexion is performed in extension (that is, a regular
compressive patter), the first choice would be to carry out the opposite move­
ment (i.e., left lateral flexion in flexion). As the symptoms improve, the tech­
nique of lef lateral flexion in flexion is changed to right lateral flexion in flexion
and progresses to right lateral flexion in extension as symptoms improve
If the patient presents with a regular stretch pattern (e.g., right lateral
fexion reproduces left buttock pain and this pain is made worse when right
lateral fexion is performed in flexion and eased when right lateral fexion is
performed in extension), the first choice technique would be the opposite move­
ment (i. e. , left lateral flexion in extension). This technique can then progress
to performing right lateral flexion in flexion as the symptoms improve.
The use of the box diagram is a simple way to show this:
R Buttock Pain
Progression: (I) left lateral flexion in flexion, (2) right lateral flexion In
fexion, (3) right lateral flexion in extension.
Or, for example 2:
L Buttock Pain
Combined Movements in Assessment and Treatment 217
Progression: ( I) left lateral flexion in extension, (2) left lateral flexion in
fexion, (3) right lateral flexion in flexion.
These procedures are described in greater detail elsewhere. I
CME procedures should be used. There is evidence that end-of-range pro­
cedures (i.e., passive movements pelformed at the limit of a range of movement)
are more effective in reducing pain perception and improving the range of move­
ment.IO-13 By combining movements, it is possible to use end-of-range move­
ments in a relatively painless way. The most painful movement for the patient
can be used as a treatment technique in its least painful combined position.
The movement technique can be carried out in a manner that causes the least
pain but is pelformed at the end of the available range. This position is pro­
gressed to what was the most painful position for the patient as the symptoms
Assessing the improvement in a patient's condition means recognizing the
patters, regular or irregular, that can help in predicting the way in which the
symptoms will improve. In the case of a regular compressive pattern where,
for example, right lateral flexion produces right buttock pain and this pain is
worse when the movement is pelformed in extension, then right lateral fexion
in neutral will improve before right lateral flexion in extension. With a regular
stretch patter, for example, of right lateral flexion causing left buttock pain,
this pain being worse when right lateral flexion is performed in flexion, then
right lateral flexion in neutral will improve before right lateral fexion in flexion.
The response in the case of irregular patters is not as predictable and the
improvement in the signs and symptoms may appear in an apparently random
Right Lateral Flexion in Neutral
The treatment table is adjusted so as to have the thoracic spine at the
desired position of lateral flexion. This position will correspond to that found
on examination. The lumbar spine is placed in a neutral position. The therapist's
right hand is placed at the level at which the lateral flexion is to be centered,
with the therapist's left hand placed over the patient's right greater trochanter.
The moment of lateral flexion is pelfrmed by moving the left hand in a cephalad
direction so as to laterally flex the pelvis to the right (Fig. 7-13).
Right Lateral Flexion in Flexion
The treatment table is adjusted and the therapist's hands are placed in a
manner as described above. The patient's hips and knees are fexed so as to
flex the lumbar spine to the required position. The patient's feet are placed
over the therapist's flexed left thigh. This maintains the angle of lumbar flexion.
218 Physical Therapy of the Low Back
Fig. 7-13. Part I of the technique examination: right lateral flexion in neutral.
Fig. 7-14. Part 2 of the technique examination: right lateral flexion in fexion.
Combined Movements in Assessment and Treatment 219
Fig. 7·15. Part 3 of the technique examination: right lateral flexion in extension.
The movem:nt of lateral flexion is carried out by moving the left hand cephalad
while maintaining the flexion of the lumbar spine (Fig. 7-14).
Right Lateral Flexion in Extension
The table and hand position are placed in a manner as described above.
The lumbar spine is placed in a position of extension. Lateral flexion is carried
out by moving the left hand in a cephalad direction while maintaining the posi­
tion of extension (Fig. 7-15).
The examination of the lumbar spine requires considerable patience and
skill. The principle of combining movements provides an addition to the stan­
dard examination and a means by which changes in signs and symptoms can
be readily assessed and specific symptom patterns observed. These symptom
patterns are of assistance in the selection of passive movement treatment tech­
nique, because the ability to predict the likely result enables the therapist to
use the CME technique, even in acutely painful conditions.
220 Physical Therapy of the Low Back
I wish to acknowledge the considerable assistance of David Watkins in the
preparation of the photographs used in this chapter.
I. Edwards BC: Manual of Combined Movement. Churchill Livingstone, London,
2. Farfan HF: Muscular mechanics of the lumbar spine and the position of power and
efficiency. Orthop Clin N Am 6: 135, 1975
3. Loehl WY: Regional rotation of the spine. Rheumatol Rehabil 12:223, 1973
4. Rolander SD: Motion of the lumbar spine with special reference to the stabilising
effect of posterior fusion-an experimental study of autopsy specimens. Acta Or­
thop Scand, suppl 90, 1966
5. Stoddard AC: Manual of Osteopathic Technique. Hutchison, London, 1959
6. Troup JDG, Hood CA, Chapman AE: Measurements of the sagittal mobility of the
lumbar spine and hips. Am Phys Med 9:308,1968
7. Gregerson GC, Lucas DB: An in vivo study of the axial rotation of the human
thoracolumbar spine. J Bone Joint Surg 49A:247. 1967
8. Kapandji IA: Physiology of Joints. Vol. 3. Churchill Livingstone. London, 1974
9. Maitland GD: Vertebral Manipulation. Butterworths, London, 1985
10. Nade S, Newbold PJ: Factors determining the level and changes in intra-articular
pressure in the knee joint of the dog. J Physiol 338:21, 1983
I I . Baxendale RH, Ferrell WR: The effect of knee joint afferent discharge on transmis­
sion in flexion reflex pathways in decerebrate cats. J Physiol 315:231, 1981
12. Grigg P, Greenspan BJ: Response of primate joint afferent neurons to mechanical
stimulation of the knee joint. J Neurophys 40: I, 1977
13. Lundberg A, Malmgren K, Schomburg ED: Role of joint afferents in motor control
exemplified by effects on reflex pathways from ib afferents. J Physiol 284:327, 1978
8 Manipulative Physical
Therapy in the
Management of
Selected Low Lumbar
Patricia H. Trott
Ruth Grant
Geoffrey D. Maitland
Patients referred for manipulative physical therapy can be divi ded i nto two
groups according to the hi story. Patients in the first group have a history of
injury, such as a fall or di rect blow, or are referred followi ng surgery. The
ti ssues that are i njured depend on the direction and force of the i njury, and
thus the therapi st cannot predict the pattern of the symptoms and signs or the
response to treatment. The second group i ncludes patients who have a history
of symptoms occurring spontaneously or following some trivial i nci dent such
as sneezing or bendi ng to pick up a light object. Pati ents in this second group
have symptoms, signs, and histories that are easily recogni zed, and these condi ­
tions respond in a predi ctable way to manipulative physical therapy.
This chapter is predicated on a recogni tion and understandi ng of the pathol­
ogy and patters of pain presentation given in previous chapters, and is there­
fore specifically di rected to technique selecti on and appli cation in the manage-
222 Physical Therapy of the Low Back
ment of some common syndromes (clinical presentations) of the low lumbar
spine ( L4-S I ) seen in physical therapy practice. Although discussion is re­
stricted to the use of passive movement techniques, the need for a detailed
assessment of the soft-tissue components and the muscular control of the spine
and pelvis is stressed, and these aspects together with ergonomic advice are
included in the overall management of low lumbar problems (these aspects are
covered in other chapters of this book).
Before describing such lumbar conditions, discussion of the factors that
govern the selection of passive movement techniques is necessary.
In the clinical setting, reaching a definitive diagnosis is not always possible.
This is particularly the case for low lumbar disorders, and relates to the follow­
ing factors.
I . In many cases the etiology is multifactorial and includes both an inflam­
matory and a mechanical cause.
2. Pain arising from certain tissues does not follow a specific anatomic
3. A particular pathologic process can give differing patterns of symptoms
and signs.
For example, patients with a diagnosis of disc herniation with nerve-root in·ita­
tion can exhibit differing clinical presentations; that is, the symptoms may be
acute and severe, or chronic. The distribution of pain may vary, being worse
either proximally or distally, with or without neurologic changes. The pattern
of limitation of movements may vary from gross restriction of fexion and
straight-leg raising ( SLR) due to pain, to full range and pain-free flexion but
with marked restriction of extension.
Clearly, a diagnostic label on its own is of limited assistance when choosing
physical treatment modalities, and in particular when selecting passive move­
ment techniques. Rather, treatment selection is based on the way the condition
presents in terms of the patient' s symptoms, abnormalities of movement, and
the history of the disorder. Knowing which structures can cause pain and the
different patterns of pain response that can occur during test movements is
fundamental to the selection of passive movement techniques.
Physical therapists, and in particular manipulative physical therapists, are
skilled in the diagnosis of mechanical disorders of the neuromusculoskeletal
system. They are also trained to recognize when symptoms do not have a
mechanical basis and when to suspect an inflammatory component. In many
cases radiologic and hematologic tests are required to exclude other pathol­
Manipulative Physical Therapy 223
Pain-Sensitive Structures and Their Pain Patterns
In the lumbar spine the common structures that cause symptoms are the
joints 1
2 and their supportive tissues, 2 and the pain-sensitive structures in the
vertebral and foraminal canals.3-7
Intervertebral Joints
Intervertebral Disc. Pain from disorders of the intervertebral disc is com­
monly deep and ill defined, presenting as a wide area across the low back or
as a vague buttock pain. This pain may spread to the upper posterior thigh or
lower abdomen, but pain originating in the disc itself is not referred into the
lower leg. s The pain may be central, to one side, or bilateral (symmetric or
A damaged disc may impinge against the posterior longitudinal ligament
or the dura or, as it heriates, disc material can impinge on or irritate the nerve­
root sleeve or the nerve root causing referred pain. 2
Discogenic pain behaves differently from pain arising from other struc­
tures, in that ( I ) following a sustained posture, rapid reversal of that posture
is both painful and stiff (e. g. , standing up quickly after prolonged sitting), and
(2) speed of movement will vary the position in range at which pain is experi­
enced (i. e. , with increased speed, pain is experienced earlier in the range).
Discogenic pain may be aggravated by either compressive or stretching move­
ments. IO
Zygapophyseal Joint. Like other synovial joints, the zygapophyseal joint
may present with an intraarticular disorder (which is made worse, that is, more
painful, when the articular surfaces are compressed) or a periarticular disorder
(which is worsened by movements that place stress on the capsule).
Commonly, zygapophyseal joint pain is felt locally as a unilateral back
pain, which when severe can spread down the entire limb.
.11 The site of pain
is not exclusive to one zygapophyseal joint; therefore, the source of pain must
be confirmed by clinical examination. In its chronic form, there may be no
local pain over the affected joint, but a distal localized patch of pain. This is a
common phenomenon in the thoracolumbar region; similar clinical findings for
the lower lumbar area have not been substantiated by research.
Ligamentous and Capsular Structures. Referred pain from specifc spinal
ligamentous structures follows no known neurologic pattern.
12 Based on clini­
cal experience, Maitland 10 reported that ligamentous and capsular pain is felt
maximally over the ligament and that the pain may spread into the lower limb.
Movements that stretch the ligament/capsule may produce sharp local pain or
a stretched sensation at the symptomatic site.
Structures in the Vertebral and Foraminal Canals
The pain-sensitive structures in the vertebral and foraminal canals are the
dura anteriorly ,5.6 the nerve-root sleeves, the ventral nerve roots, and the blood
vessels of the epidural space. 7
224 Physical Therapy of the Low Back
The structures compri sing the vertebral canal that are pain sensi tive are
the posterior longitudinal ligament, posteri or portions of the annuli fibrosi , and
the anterior aspect of the laminae. I The pai n-sensitive components of the fora­
minal canals are the posterolateral aspect of the i ntervertebral discs and the
zygapophyseal joints. I
Passi ve-movement tests will i mplicate a loss of mobi lity and/or i ncrease
in tension of the neuromeningeal ti ssues.
Dura and Nerve-Root Sleeve. Dural pain does not have a segmental pattern
of reference. 3 However, stimulation of the nerve-root sleeve gives rise to symp­
toms of simi lar distribution to those ari sing from stimulation of that nerve root.
The only reported difference i s that the nerve root frequently gives rise to
symptoms that are more severe di stally and the pain is often associated with
paresthesia. These phenomena are not seen with i rri tation of the nerve-root
sleeve. '4
Radicular Pain. Mechanical or chemical i rritation of the sensory nerve
root causes pain and/or paresthesi a to be experi enced in the di stal part of a
dermatome, or i f felt throughout a dermatome, these symptoms are often worse
di stally. 15 Movements that narrow the i ntervertebral canal and foramen (exten­
sion, rotation, and lateral fexion to the affected si de) are likely to reproduce or
aggravate the nerve-root pai n/paresthesia. Cli ni cally, pain can be conclusively
attributed to the nerve root only if there are neurologi c changes i ndicating a
loss of conduction along that nerve root.
Passive movement tests that specifically test the neuromeningeal struc­
tures are SLR, prone knee flexi on (PKF), passive neck flexi on (PNF), and the
slump test. These tests are described in major textbooks on manipulati ve ther­
apy. 9
IO.16.17 They are used not only as examination techniques but also i n
Range/Pain Response to Movement
Test movements of the low lumbar i ntervertebral joints and the neuromen­
i ngeal ti ssues produce common patterns.
Stretching or Compressing Pain
Unilateral back pain may be reproduced by either stretching (e. g. , lateral
flexi on away from the pai nful si de)or by compressing the faulty tissues (e. g. ,
lateral flexion toward the pai nful si de).
End-of-Range or Through Range Pain
Pain may be reproduced at the limit of a parti cular movement (i.e. , when
the sof-ti ssue restraints are put on stretch) or during the performance of a
movement, i ncreasing near the limit of the movement. Through range pain i s
common i n joints i n which there i s a constant ache.
Local and Referred Pain
Manipulative Physical Therapy 225
In pat i ents who have referred pain, t he pain response t o t est movements
influences t he select i on of passive movement t echniques. For example, t est
movement s even when firmly applied may eli cit only local back pain. In t hese
pati ent s the movement may be applied frmly without risk of exacerbati on of
sympt oms
In cases where t he t est movement has t o be sustained at end of
range in order to reproduce t he referred pain, a t reatment t echni que t hat is
sust ained will be requi red. In contrast to t his, t est movements t hat i mmediat ely
cause distal leg sympt oms requi re very gentle treatment i n a manner t hat does
not reproduce the dist al symptoms. Test movements t hat cause latent referred
pain or t hat cause t he referred pain t o linger also i ndicate caution i n treatment .
Any history t aken should include t he onset and progression of t he di sorder.
Condi tions t hat have a spontaneous ( nontraumat i c) onset have a charact eristi c
progressive history; t hat i s, t here i s a patt ern t hat i s t ypical of a degenerati ng
disc or of postural ligamentous pai n. Knowing t he history t hat is t ypical for
t hese condi ti ons helps t he cli nici an t o recogni ze t he present stage of t he disorder
and to match this wi t h t he symptoms and signs t o form a syndrome. Typical
histories are presented in t he case studies at the end of t hi s chapter.
A det ailed history gives i nformation as to the stabili ty or progressi ve nature
of t he disorder. Thi s will guide t he extent and strength of t echniques used and
may contraindi cat e cert ain t echniques. This is particularly i mportant in cases
of a progressive disc di sorder when i njUdici ous treatment may convert a pot en­
ti al disc prot rusion i nt o a heri ated disc wit h neurologic changes.
The progression of t he di sorder allows predi ction of t he out come of t reat ­
ment, number of t reatment sessions needed, and long-term prognosis.
The followi ng case history illust rates these aspect s of history t aking.
A 25-year-old gardener presented with a IO-year history of low back
pain, which started one school vacation when he worked as a buil der's
laborer. He then remained symptom free until he began work as a gardener
7 years ago. Prolonged digging caused low back pai n, which initial l y would
be gone by the next morning, but this slowly worsened to the extent that
the pain spread to his l eft buttock and posterior thigh and took longer to
settle. I n the last 6 months he has required treatment; two or three treatments
of heat and extension exercises have completely relieved his symptoms.
Two weeks ago, he tripped over a stone and experienced sharp pain in his
left calf and paresthesia of his lef fifth toe. This has not responded to heat
and exercises, but he has been able to continue his gardening.
Thi s history i s t ypical of a progressive and worseni ng disc disorder and
t he pati ent is now at a stage where he has nerve-root i rrit at i on. Although a
t rivial i nci dent provoked t hi s epi sode, t he disorder i s relatively stable in t hat
he can continue gardening without worsening hi s sympt oms.
226 Physical Therapy of the Low Back
More specifc treatment will be required and can be perfrmed firmly with­
out risk of exacerbation of his symptoms. The expectati on of treatment is to
make him symptom free while antici pating that there may be further epi sodes
due to the progressive nature of his di sorder.
The area i n whi ch a patient feels the symptoms and the manner in whi ch
they vary i n relation to posture and movement assist i n the recognition of syn­
dromes, and i f they match the response to physical exami nation, can assist in
the selection of passive movement techniques. A movement or combination of
movements that simulate a position or movement described by the patient as
one that causes the pain, can be used as the treatment technique. The following
case history i llustrates this.
A right-handed t ennis player complains of chronic right-side low back
pain as he commences serving. I n this position his low l umbar spine is
extended, laterally fexed, and rotated to the right . Examination confirms
that this combined position reproduces his pain and testing of i ntervertebral
movement reveals hypomobi l i ty at the LS-S I joi nt .
An effecti ve treatment would consi s� of placing hi s low lumbar spine i nto
this combined position and then passively stretching one of these movements,
carefully localizing the movement to the L5-S I joint.
Two other i mportant aspects of the pati ent' s symptoms are the severity of
the pain and the irritability of the disorder. Severity relates to the examiner' s
i nterpretation of the severity of the pain based on the pati ent' s descri ption and
functional limitations due to the pain. Irri tability (or touchiness) of the disorder
is based on three things: (1) how much activity the patient can pelform before
being stopped by pai n; (2) the degree and distribution of pai n provoked by that
activity; and (3) how long the pain takes to subsi de to i ts original level. (Thi s
is the most informati ve part and serves as a guide to the probable response of
the symptoms to exami nation and treatment. )
In the previous example of the tennis player, a nonirri table disorder would
be one in which he experiences momentary pai n each time he serves (in thi s
case the treatment described previously would be applicable). In contrast to
this, an i rri table di sorder would be one i n which his back pain lasts for several
minutes after servi ng a ball, and this pain increases to the extent that after
serving one game hi s back is so painful that he cannot continue to play and
has to rest for I hour to ease his pain. In this example, a technique that repro­
duced his symptoms would not be the i ni tial choice of treatment, but rather
his lumbar spine would be positioned in the most comfortable positi on and a
techni que performed that was pain free.
Manipulative Physical Therapy 227
Signs refer to physical examination findi ngs. Physical exami nation tests
are used to i ncrimi nate or exclude certai n structures as the source of a pati ent' s
symptoms. In particular, the tests determine the i nvolvement of the i nterverte­
bral joints and neuromeningeal ti ssues and whether conduction of the spinal
cord and cauda equina is altered. They help indi cate the degree of irritability of
the di sorder and demonstrate whether symptoms have a stretch or compression
The physical examination of movements i ncludes three sections. These
are exami nation of
I . The gross physi ologi c movements of the lumbar spine (flexion, exten­
sion, lateral flexion; and rotation). It may be necessary to examine these move­
ments in di fferent combinations and i n varied sequences, to sustain these posi­
tions or to perform them with di straction or compressi on
2. Passi ve physi ologi c and accessory movements at each i ntervertebral
3. The neuromeningeal ti ssues i n the vertebral and forami nal canals ( using
SLR, PNF, PKF, and the slump test)
Reachi ng a di agnosi s of a mechanical di sorder of the neuromusculoskeletal
tissues is i mportant, to i solate the structures at fault by knowing the symptom
distribution and the response to physical tests. Knowledge of the movements
that i ncrease or decrease the symptom response are the main determinants of
how to apply passive movement in treatment (not the di agnosti c title per se).
Selection of Technique Based on Efect
Passive movement as a treatment techni que can be broadly divided i nto
its use as mobilization (passive osci llatory movements) or manipulation ( small
amplitude thrust/stretch performed at speed at the limit of a range of
Mobili zation i s the method of choice for most lumbar di sorders because i t
can be used as a treatment for pai n or for restoring movement i n a hypomobi le
joint. It can be adapted to suit the severity of the pain, the i rritability of the
di sorder, and the stages and stability of the pathology.
Manipulation is the treatment of choice when an i ntervertebral joint i s
locked. To regain mobi lity i n cases of an i rritable joint condi tion, a single local­
ized manipulation may be less aggravating than repeated stretching by mobili ­
228 Physical Therapy of the Low Back
Position of the Intervertebral Joint and Direction of the
Movement Technique
Treatment by passive movement i nvolves careful positioni ng of the parti cu­
lar i ntervertebral segment and the selecti on of the most effective direction of
movement. These are based on a knowledge of spinal bi omechanics and the
desired symptom response.
Manipulation. Manipulation i s applied in the direction of limi tation in order
to stretch the tissues in that parti cular direction. For example, using bi omechan­
i cal pri nciples, the lumbar spine can be positi oned (in lateral flexion and contra­
lateral rotation) to i solate movement to the desi red i ntervertebral segment and
a rotary thrust applied i n the appropriate di recti on.
Passive Mobilization. When using passive mobilization both the position
of the i ntervertebral joi nt and the di rection of movement are varied according
to the desired effect of the technique. Some examples are:
I. To avoid any discomfort or pain: In cases where the pai n is severe or
the di sorder i s irri table, the symptoms should not be provoked or aggravated.
The lumbar spine would be positioned so that the painful i ntervertebral segment
was pain free and the movement techni que performed must also be pain free.
2. To cause or to avoid reproduction of referred pain: Provocati on of re­
ferred pain is safe when the pain i s chroni c and nonirri table, and when it i s not
an acute radi cular pain. In these cases i t may be necessary to cause some leg
pain to gain i mprovement; thus the spine is positioned to ei ther provoke some
symptoms or to enable the treatment techni que to provoke the referred symp­
toms. Findings i ndi cating radicular pain (i . e. , pain worse di stally and the pres­
ence of neurologi c changes), when the examination of movements reproduces
the di stal pain, should warn the clini cian against using a technique that provoked
the referred pain.
3. To open one side of the intervertebral joint (i . e. , to stretch the di sc,
distract the zygapophyseal j oint, and wi den the foraminal canal on one si de):
This would be the choice i n cases of nerve-root irri tati on/compression or i n
cases of a progressive unilateral disc di sorder. For example, to widen the right
side of the L4 to L5 intervertebral space, the spine would be posi tioned in the
combined posi tion of flexion, lateral flexi on, and rotation to the left. Which of
these movements would be emphasized as the treatment technique would de­
pend on the pain response.
4. To stretch tissues that are contracted: Joints that are both painful and
hypomobile can respond di fferently to passive mobilization dependi ng to a large
extent on the i rritability. The pain response during the pelformance of a tech­
nique and i ts effect over a 24-hour peri od will guide the clini cian in the choice
of whi ch direction to move the joint and how firmly to stretch the contracted
tissues. A favorable response to gentle oscillatory stretches is that the pain
experi enced duri ng the techni que decreases, thus allowing the movement to be
performed more strongly. A worseni ng of the pai n response indi cates that thi s
direction of movement is aggravating the conditi on.
Manipulative Physical Therapy 229
5. To move the intervertebral joints or the canal structures: During the
physical examination, if movements of the intervertebraijoints and of the neuro­
meningeal structures in the canal both reproduce the patient' s leg symptoms,
a technique directed at altering the intervertebral joint movements should be
the first choice of treatment. The effect on the intervertebral joint signs and
the canal signs is noted and if the latter are not improving, movement of the
neuromeningeal tissues is added or substituted. In cases of only back and/or
buttock pain and where the canal tests more effectively reproduce this pain,
then the first choice would be to use movement of the canal structures (e. g. ,
PNF, SLR, PKF, or the slump test).
Manner of Movement Technique Performance
Selection of a treatment technique does not merely relate to the direction
of movement but also to the manner in which it is applied. The amplitude can
be varied from a barely perceptible movement to one that makes use of the
total available range. The rhythm can be varied from a smooth, evenly applied
movement to one that is staccato. Similarly, the speed and the position in range
in which the movement is performed can be altered.
Passive movement techniques must be modified according to the intention
of the technique, and this is based on the symptoms experienced by the patient
during the technique, the quality of the movement, the presence of spasm, and
the end feel. I t is not possible to discuss these details in any depth in this
chapter, but only to present the two ends of the symptom spectrum that ranges
from a constant ache with pain experienced through range, to stiffness with
mild discomfort felt only at the end of range of certain movements as presented
below. For a full description, see Maitland.10
Constant Aching with Pain Through Range. The lumbar spine must be
placed in a position of maximal comfort (usually one of slight flexion and mid­
position for the other movements). The treatment technique will be of small
amplitude, performed slowly and smoothly (so that there is no discomfort pro­
duced or where discomfort is constant, with no increase in the level of the
aching). The movement technique may be a physiologic or an accessory move­
ment, and its performance should result in an immediate lessening of the level
of aching. I n some patients there may be an immediate effect, but in others the
effect should be noted over a 24-hour period.
Stiffness with Mild Discomfort Felt Only at the End of Range of Certain
Movements. The lumbar spine is carefully positioned at or near the limit of the
stiff directions of movement (i . e. , in the position that best reproduces the stiff­
ness and discomfort). The treatment technique will be one that places maximal
stretch on the hypomobile intervertebral segment. The technique should be
firmly applied, of small amplitude, and either sustained or staccato in rhythm.
If the level of discomfort increases with the firm stretching, large amplitude
movements can be interspersed every 40 to 60 seconds.
230 Physical Therapy of the Low Back
In this section some of the common syndromes with a history of sponta­
neous onset of symptoms are presented. Management is restricted to treatment
by passive movement. The syndromes are presented as case histories; however,
the reader is encouraged to read Grant et al .
Acute Back Pain (Disco genic)
A 35-year-old man experienced a mild central low backache after push­
ing a car one morning. This ache intensified during a 40-minute drive to
work, and he was unable to get out of the car unaided due to severe l ow
back pain. He had no previous history of backache and radiographs of his
lumbosacral region were reported as being normal. After 2 days in bed
without any improvement in his symptoms, physical therapy was requested
by his doctor.
There was a constant dull backache centered over the L4 to L5 region.
The patient was unabl e to move in bed due to sharp jabs of pain. His most
comfortable position was supine with his hips and knees flexed over two
pillows in crook lying or side l ying with his legs (and l umbar spine) flexed.
The patient was examined supine. SLR was almost full range as was
PNF, but both tests slightly increased his back pain. Spreading and
compression of the ilia were pain free and there was no abnormality in
lower limb reflexes or sensation (testing muscl e strength was not under­
taken due to back pain). I n crook lying, lumbar rotation to each side was
reduced to half range due to pain.
The history suggests discogenic pain-pushing a heavy car would raise
intradiscal pressure-with pain worsened by sitting followed by inability
to extend the spine. The physical examination was too restricted by pain
to be helpful in confirming the source of his symptoms.
Manipulative Physical Therapy 231
Fig. 8-1. Longitudinal caudad mobilization of the lower l umbar spi ne, produced by
manual traction to the lower li mbs.
Day 1 (Treatment 1 ). Because SLR was of good range, both legs were
comfortably fexed at the hips to 50° and gentle manual traction was applied
by pulling on his legs (Fig. 8-1). During traction his backache decreased.
A series of four gentle but sharp tugs were applied to his legs. With each
tug, a jab of back pain was experienced but there was no increase in his
constant backache. On reassessment, PNF was full and painless but SLR
remained unchanged. The treatment was repeated resulting in a slight im­
provement in his SLR on both sides and in his range of lumbar rotations
performed in supine.
Day 2 (Treatment 2). The patient reported greater freedom of move­
ments in bed and had been out of bed twice for a hot shower. Examination
of his l umbar mobility in standing showed marked l imitation of fexion by
pain centered over L5 and there was obvious spasm in his erector spinae.
Extension was half range and he had a full range of lateral flexion to each
side. Bilateral SLR was full and painless. Manual traction efected no
change in his mobility so he was placed in a prone position with two pillows
under his abdomen. From this position of comfort, he was asked to gently
passively extend his spine using a modified push-up technique. 19 This tech­
nique was repeated 10 times, sustaining the position for 5 seconds. The
patient was encouraged to extend his lumbar spine to his comfortabl e limit.
However, particular care was taken to avoid development of a backache.
On reassessment in standing, fl exion had improved so that he could reach
fingertips to his patellae and extension was three-quarters of usual range.
232 Physical Therapy of the Low Back
The technique was repeated but this time lying over only one pillow.
Following two more applications the patient could lie comfortably prone
without a pillow, and in weightbearing there was a further slight increase
in his ranges of flexion and extension.
The patient was asked to repeat this technique hourl y (three sets of
10 push-ups, at the end of which he allowed his back to sag into sustained
extension for I minute, providing that there was no reproduction of back­
ache). He was allowed out of bed for short periods but was to avoid sitting.
Day 3 (Treatment 3). The patient was pain free when moving in bed
and could be ambulant for more than I hour before his ache retured. His
range of fl exion was such that he was able to reach his fingertips to midshin
level (normally he could reach his ankles) and to fully extend his lumbar
spine with only a mild ache.
The extension push-ups were repeated with the manipulative therapist
stabilizing the patients' pelvis flat on the fl oor. As this was pain free, the
patient was asked to sustain the position all owing his lumbar spine to sag
fully into extension. In this extended position, gentle posteroanterior pres­
sures were applied to L5, taking care to cause only mild discomfort. On
resuming the flat prone position, sharp deep pain was felt over L5 but this
quickly subsided with repeated gentle extension.
The above regimen was repeated prior to reassessing his mobility in
standing. Flexion and extension were now full range. Gentle overpressure
to extension reproduced the same deep, sharp pain.
Days 4, 6, and 1 0 (Treatments 4, 5, and 6). Subsequent treatments
were conducted at the manipulative therapist's clinic and consisted of res­
toration of full pain-free range of lumbar extension using posteroanterior
pressures on L5 with the patient's lumbar spine in extension. By day 6 the
patient was symptom free, but experienced a deep ache with firm sustained
posteroanterior pressure on L5. When seen 4 days later, no pain could be
elicited by sustained or staccato posteroanterior oscillatory movement.
The patient was discharged with advice regarding lifting and care of
his back during sustained postures (especially fl exion).
Severe Nerve-Root (Radicular) Pain
A 35-year-old man had suffered from recurrent attacks of low back
pain over the last 5 years. These were associated with lifting strains. This
present episode commenced 4 days ago when he bent to move the garden
hose. He experienced only mild aching in his low back, but over the next
few hours his back pain disappeared and he felt strong pain in his left
buttock and calf. His calf pain had worsened in the last 24 hours and spread
into his left foot.
Manipulative Physical Therapy 233
The patient had constant severe pain in the lateral aspect of his left
calf and foot and numbness of the l ateral aspect of his left foot. Less severe
aching was experienced in his left buttock. Weight bearing and sitting ag­
gravated his back and calf pain, and coughing aggravated his back pain.
He could gain some relief of symptoms by l ying on his right side with his
legs (and lumbar spine) fexed.
The following movements aggravated his buttock and calf/foot pain:
fexion to touch his patellae, extension and left lateral flexion half range,
and left SLR limited to 25°. Neurologic examination revealed a reduced
lef ankl e jerk, reduced sensation over the l ateral border of the foot, and
weak toe fl exors (calf power could not be tested due to pain on
There was evidence of S I nerve-root compression. The history of a
trivial incident causing this episode and the presence of worsening symp­
toms indicated that the disorder is both unstabl e and progressive, requiring
care with treatment so as not to worsen the condition.
Day 1 (Treatment I), In position of ease (i. e. , lying on right side with
lumbar spine comfortably flexed), the pelvis was gently rotated to the right
taking care not to aggravate calf/foot or buttock pain. The technique was
performed as far as possible into range without aggravating symptoms;
SLR was used as reassessment. After two applications of rotation, fl exion
was also reassessed. The patient reported easing of his calf symptoms and
both flexion and left SLR had minimally improved; however, extension
range remained unal tered.
The patient was advised to rest in bed as much as possibl e and to
avoid sitting.
Day 2 (Treatment 2), The patient reported that his symptoms were
unchanged, and his physical signs and neurologic status were found to be
unaltered. Rotary mobil ization was repeated (as on day 1) with a similar
Day 3 (Treatment 3), No alteration in symptoms or signs. As his symp­
toms were aggravated by weightbearing, lumbar traction was given (15 Ib
234 Physical Therapy of the Low Back
for 10 minutes) . During traction his calf/foot pain was eased and aferward
his SLR improved by 10°.
Day 4 (Treatment 4). Definite reduction in calf/foot pain and improve­
ment in all physical signs. Traction was repeated (15 Ib for 20 minutes).
At subsequent treatments both the time and the strength of the traction
were increased.
Day 10 ( Treatment 8). There were no leg symptoms, but buttock pain
was experienced with prolonged sitting. Physical examination revealed full
recovery of neurologic function. The extreme range of fl exion, with the
addition of neck fl exion, reproduced buttock pain; the other spinal move­
ments were full range and pain free. On passive overpressure, left SLR
lacked 20° and also reproduced left buttock pain.
At this stage, spinal mobility was full and painless but movement of
the neuromeningeal tissues was restricted and reproduced the patient's
only remaining symptom. A gentle technique to stretch the neuromeningeal
tissues should be used, but if reproduction of nerve-root symptoms oc­
curred it would contraindicate its use at this stage.
Treatment 8 (Continued). A gentle stretch was applied to l eft SLR,
causing only buttock pain. Following this, flexion plus neck fl exion were
pain free.
Day 12 (Treatment 9). The patient was now symptom free, but left
SLR still caused buttock pain at 75°.
Despite an excellent response to treatment, in view of the progressive
disorder, a decision not to stretch his SLR more firmly was taken. It was
decided to review his progress in 2 weeks.
When seen 2 weeks later he had remained symptom free but his left
SLR had not improved. Now that his disorder had stabilized, his SLR was
strongly stretched, restoring full range with no retur of symptoms.
Chronic Nerve-Root (Radicular) Aching
This may present as either ( I ) residual symptoms from an acute episode
of nerve-root pain; or (2) chronic aching (not pain) with signs of nerve-root
In both cases, the disorder is nonirritable and does not restrict the patients'
activities; however, because most low lumbar nerve-root problems are of disco­
genic origin, sitting causes an increase in leg symptoms. The disorder is stable
Manipulative Physical Therapy 235
and permits stronger techniques to be applied safely. The following case history
illustrates the second type.
A 40-year-old housewife presented with a past history of recurrent
low back pain for 7 years. One year ago she noticed a dull ache in her left
leg. At that time, two or three treatments of passive mobilization com­
pletely relieved her symptoms. The current episode began 3 weeks ago
following paving of the garden path with bricks. While stooping to lay the
bricks she was conscious of aching in her buttock and down the posterior
aspect of her left leg. Rotary mobilization had not helped.
A constant dull ache spread from her left sacroiliac area, down the
posterior aspect of her buttock to the heel, together with paresthesia of
the lateral aspect of her foot. The ache in her calf and the paresthesia were
worsened if she sustained a flexed posture (e. g., vacuuming carpets) for
more than 30 minutes or sat for more than 60 minutes.
There was a full range of pain-free spinal movements, even when these
were sustained. Poor intervertebral movement was noted below L3 on
extension and on lef lateral flexion. By adding left lateral flexion to the
fully extended position, buttock aching was reproduced. The addition of
left and right rotation made no change in the symptoms. Testing of interver­
tebral movement confrmed hypomobility at both L4-L5 and L5-S I mo­
tion segments and posteroanterior pressure over L5 (performed with her
spine in extension/left lateral fexion) caused buttock pain.
Tests for the neuromeningeal tissues revealed full SLR but the left leg
had a tighter end feel; the slump test was positive (i . e. , left knee extension
lacked 30° and caused calf pain, which was eased by releasing cervical
flexion). She had slight weakness of her left calf, but otherwise showed
no neurologic deficit.
The history implicated a disc disorder that was slowly progressing to
interfere with nerve-root function. The disorder was stable in that the pa­
tient could continue with her daily activities as a housewife. Treatment to
236 Physical Therapy of the Low Back
change both her intervertebral joint signs and neuromeningeal signs was
necessary, using techniques that would temporarily aggravate her leg ache.
Treating the intervertebral joint hypomobility first was safer, while observ­
ing its effect on both the joint and neuromeningeal signs.
Day 1 (Treatment 1 ). With the patient's low lumbar spine positioned
in extension/left lateral flexion, firm posteroanterior pressures were applied
to L4 and L5 spinous processes for 60 to 90 seconds, causing local pain
and a mild increase in left buttock aching (Fig. 8-2). On reassessment,
lumbar extension/left lateral flexion no longer reproduced an increase in
buttock ache and low lumbar mobility had improved. The slump test had
improved (left knee extension improved by 10°). Treatment was repeated
with no further gain in mobility; neurologic function was unchanged.
Day 3 (Treatment 2). The patient reported no ill effects from treat­
ment, and a lessening of her left leg aching. Physical examination showed
that she had maintained the improvement gained on day I . Neurologic
function was unchanged. The above treatment was repeated even more
strongly and sustained to stretch the tight tissues. The reproduction of only
local pain (no referred buttock pain) supported the safety of using a strong
stretch. The result of this stretch was that extension/left lateral flexion was
painful only when sustained, and in the slump position, knee extension
Fig. 8-2. Posteroanterior pressures on L5 with the l umbar spine positioned in exten­
sion/lef lateral fexi on.
Manipulative Physical Therapy 237
improved by another 10°. Repeating the technique twice gained no further
change to the slump test.
Day 5 (Treatment 3). The patient was delighted with her progress. She
no longer had a constant ache down her left leg. The ache and paresthesia
retured only if she sat for more than 1 hour. On examination, her left calf
had regained full strength. Left SLR was no longer tighter than on the
right; however, in the slump position, left knee extension still lacked 10°
and caused buttock pain.
Treatment was changed to restore mobility in the neuromeningeal tis­
sues. In the slump position, her left knee was stretched into full extension,
causing sharp buttock pain (Fig. 8-3) . This technique was not repeated
until its efect on nerve-root conduction was known. This can only be
assessed over 24 to 48 hours.
Day 7 (Treatment 4). There was no return of symptoms following the
last treatment. Her calf strength and ankle reflex were normal. In the slump
position, knee extension still reproduced sharp buttock pain. The stretch
to the neuromeningeal tissues was increased by stretching the knee into
Fig. 8·3. Passi ve extension of
the l eft knee performed whi l e i n
the sl ump posi tion.
238 Physical Therapy of the Low Back
full extension and stretching the ankle into full dorsiflexion in the slump
position. This again caused sharp buttock pain. Reassessment of knee ex­
tension in the slump position showed it to be full range with minimal but­
tock pain. The technique was repeated once more. The patient was asked
to experiment with activities such as sustained fexion and sitting for long
periods during the next week.
Day 14 (Treatment 5). The patient reported that she experienced no
leg symptoms but that her back ached after activities involving sustained
flexion for more than 45 minutes and after sitting more than 90 minutes.
She considered that this was better than she had been for several years.
Examination of the slump test revealed full mobility, but caused slight
buttock pain. The patient was discharged with a home exercise program
to maintain the mobility of both her lower lumbar spine and her neuro­
meningeal tissues.
Mechanical Locking
A 20-year-old man complained of a sudden onset of unilateral back
pain, which prevented him from standing upright. He had bent forward
quickly to catch a ball near his left foot and was unable to straighten be­
cause of sharp back pain. He had no past history of back pain and no spinal
radiographs had been taken.
There was no pain when his back was held in slight flexion, but on
standing upright pain was experienced to the right of the L5 spinous
The patient was prevented by pain from extending,laterally flexing,
or rotating his low lumbar spine to the right. The other movements were
full and painless. Passive testing of intersegmental movement revealed an
inability to produce the above painful movements at L4-L5 with marked
spasm on attempting to do so. Unilateral posteroanterior pressures over
the right L4-L5 zygapophyseal joint produced marked pain and spasm.
Manipulative Physical Therapy 239
A rapid unguarded movement in flexion/lef lateral fexion gapped the
right lumbar zygapophyseal joints, following which there was mechanical
blocking of the movements that normally appose the articular surfaces
(extension, lateral flexion, and rotation of the trunk to the right). The mech­
anism of mechanical locking remains a contentious issue.
A manipulation, localized to the affected intervertebral level, to gap
in this case the right L4-L5 zygapophyseal joint, will restore normal joint
A rotary manipulation was performed to gap the right L4-L5 zygapo­
physeal joint. The patient was positioned on his left side with his low
lumbar spine fexed and laterally flexed to the right until movement could
be palpated at the L4-L5 intervertebral level (Fig. 8-4). In this position,
with thumb pressure against the right side of the spinous process of L4 (to
stabilize L4), a quick left rotary thrust was applied through the pelvis and
to L5 by finger pressure against the left side of L5 spinous process (to pull
L5 into left rotation).
Immediately afterward the patient could fully extend, laterally flex,
Fig. 8-4. Rotary manipulation is used to open the right L4-LS zygapophyseal joint.
240 Physical Therapy of the Low Back
and rotate hi s trunk to the ri ght with only soreness experienced at the
extreme of these movements. Thi s soreness was lessened by gentle large
amplitude posteroanteri or pressures performed unilaterally over the right
L4-L5 zygapophysealjoint. The next day the patient reported by telephone
that he was symptom free.
Zygapophyseal Joint Arthropathy (Causing Only
Referred Symptoms)
A 50-year-old man descri bed a gradual onset, over 3 days, of achi ng
i n the right trochanteric area. Thi s had been present for I month. He could
not recall any i njury to hi s back, hip, or leg, and he had not experienced
pain i n any other area. He had no past hi story of back or lower extremity
Radi ographs of the lumbar spi ne and hi p were reported to be normal.
He was di agnosed by the referri ng doctor as suffering from trochanteri c
bursiti s.
The patient experienced a constant deep ache over his right greater
trochanter, whi ch was unaltered by posture or acti vity.
Lumbar movements were full and pain free with overpressure. Tests
for the neuromeningeal tissues, hip, and trochanteric bursi tis were neg­
ati ve.
Deep palpation (through the erector spinae) over the right L4-L5 zyga­
pophyseal joint revealed stiffness, l ocal spasm, and tenderness, and there
was an area of thickening at the right side of the interspinous space between
L4 and L5. These signs were absent on the left side. Reproduction of
referred pain was not possible.
Anatomical ly, pathology of the L4-L5 zygapophyseal joint could give
rise to referred pain at the trochanteri c area. In the absence of other physi­
cal signs, it would be appropri ate to mobilize the hypomobile L4-L5 zyga­
pophyseal joint and note any effect on hi s trochanteri c aching. An associ a-
Manipulative Physical Therapy 241
tion between the hypomobile L4-L5 zygapophyseal joint and the
trochanteric pain can be made only in retrospect.
The hypomobility was localized to the L4-L5 right zygapophyseal
joint; therefore, passive stretching should be localized to this joint.
Day 1 (Treatment 1). Posteroanterior oscillatory pressures were ap­
plied firmly for 60 seconds to the spinous processes of L4 and L5, and
unilaterally over the painful joint. On being questioned, the patient reported
no change in his constant trochanteric ache.
Day 2 (Treatment 2). The patient reported that his right trochanteric
pain was now intermittent ( and continued unrelated to movement or to
changes of posture of his trunk). The treatment was repeated giving three
applications of posteroanterior pressures lasting 60 seconds each. These
were interspersed with gentle large amplitude oscillations to ease the local
Days 4, 6, and 8 (Treatments 3, 4, and 5). The patient reported contin­
ued improvement of his symptoms and the mobilization of L4-L5 was
progressed in strength and sustained for longer periods to achieve a better
stretch on the tight soft tissues (capsule and ligaments). At the last visit,
it was necessary to place his lower lumbar spine into full extension and
direct the posteroanterior pressures more caudally in order to detect any
residual hypomobility. By this stage, he experienced only occasional tran­
sient aching in his thigh, so treatment was stopped, with a review in 2
Day 22 (Treatment 6). When reassessed 2 weeks later the patient re­
ported that he was symptom free. Passive mobility tests showed no hypo­
mobility or thickening of soft tissues on the right of the L4-L5 joint.
Zygapophyseal Joint Arthropathy (Intra-articular
A 75-year-old woman complained of a sharp pain to the right of L5
following stepping awkwardly with her right foot into a shallow depression
in the pavement I week previously. She had become aware of aching in
her back and this worsened with each step until it became constant. Be­
cause her back was both painful and stif the following morning, she con­
sulted her doctor. He ordered radiographs to be taken, which revealed
narrowing of her lumbosacral disc space and osteoarthritic changes in her
242 Physical Therapy of the Low Back
l umbosacral zygapophyseal j oints. She was given anti-inflammatory medi­
cation and advised to rest as much as possible. Five days later her pain
was no l onger constant but certain spinal movements still caused considera­
ble pain and she was referred for physical therapy. She had a past history
of low backache for many years if she stood for long periods.
Sharp pain j ust lateral to the spinous process of L5 on the right was
provoked by turning in bed from a supine position onto her right side, and
on bending to the right when standing.
All movements of her low lumbar spine were hypomobile. Lumbar
extension and lateral fexion to the right reproduced her pain. Pain was
experienced at half range, increasing at the limit of these movements.
Palpation of passive accessory movements revealed marked hypomo­
bility of L4-L5 and L5-S I segments and posteroanterior pressure and
transverse pressure to the left reproduced her back pain. Tests for neural
mobility and conduction were negative.
This was an elderly woman with a degenerative, stiff lower lumbar
spine. A trivial inj ury (i. e. , ajar up through her right leg) caused the hypo­
mobile joint to become painful.
This was thought to be an intraarticul ar problem because pain was
experienced with movements that cl osed the right side of the intervertebral
joint and because pain was felt early and throughout these movements.
Such articular problems respond well to large ampl itude passive mobiliza­
tion, performed carefully to avoid compression of the articul ar sUlfaces.
Later, when pain is minimal, one may progress to mobilization with the
surfaces compressed.
Day 1 (Treatment I), With the patient lying comfortably on her right
side in slight flexion, gentle large amplitude left lateral flexion oscillations
were produced by moving the pelvis (Fig. 8-5). Care was taken not to cause
any discomfort. Following this, extension and right lateral flexion were
reassessed. A favorabl e response was noted in that pain started later in
Manipulative Physical Therapy 243
Fig. 8-5. Passive lef lateral flexion mobilization of t he lower l umbar spi ne, produced
by moving the lower l i mbs and pelvi s.
the range of both of these movements. The t echnique was repeated but no
further improvement was noted.
Day 3 (Treatment 2) . The patient reported no change in her symptoms
but the improvement in the signs gained with the first t reatment had been
maintained. The day I treatment was repeated t wice, following which the
pain response to right lateral flexion improved, but extension was un­
changed. The t echnique was changed to accessory posteroanterior central
pressures to L4 and L5, again employing large amplitude movements. This
achieved an immediate improvement in the range of extension and a reduc­
tion in the pain response.
Day 5 (Treatment 3). The patient was del ighted with her progress in
that she had no pain turning in bed and her daily movements were painless.
Mil d pain was experienced at the l imit of both extension and right lateral
flexion. By adding extension to right lateral flexion, sharp pain was pro­
duced. For treatment, her lumbar spine was placed in t he position of slight
right lateral fexion combined with extension. In this position, posteroan­
terior pressures were performed for 30 seconds as a large amplit ude move­
ment, causing slight pain at first . During performance of the technique,
the pain disappeared, so the spine was placed further into extension and
right lateral fexion. Slight pain was again experienced and again this disap­
peared with another application of the mobilization. On reassessment, com­
bined right lateral fexion with extension was pain free. Slight pain was
experienced only on overpressure.
244 Physical Therapy of the Low Back
Day 7 (Treatment 4). The patient was still symptom free and ri ght
lateral fexion with extension was no longer painful when performed with
overpressure. However, when right lateral flexion was added to extension,
slight pain was experienced.
The spine was placed in this combined position in the same order (full
extension and then full right lateral flexion) and posteroanteri or central
pressures performed as a strong stretching technique. Thi s caused marked
pai n and required gentle large amplitude posteroanteri or pressures to ease
the soreness. Following this the patient complained of aching across her
lower lumbar area. Pulsed short-wave diathermy for 1 5 minutes (on a low
frequency and low dosage) eased her ache.
Days 10 and 24 (Treatments 5 and 6). The patient remained symptom
free but extension plus right lateral fexion still reproduced slight pai n to
the ri ght of L5. By combining these movements on the left si de, a similar
pain was produced. This was consi dered li kely to be her normal response
and no further treatment was given. Thi s was veri fied by finding the same
signs 2 weeks later, during which time she had remained symptom free.
Postural Pain
A 28-year-old mother of three chi ldren presented with a 6-month hi s­
tory of gradual onset of low back pain. She could not recall an i ncident
that had caused her symptoms. During her thi rd pregnancy 2 years ago,
she had experi enced the same pain but this had settled after the birth.
There was no hi story of trauma and her radi ographs were reported to be
The patient was asymptomatic in the morning but by mi dafternoon
her low back began to ache. Thi s ache worsened as the day progressed,
especi ally during activities requi ring her lumbar spine to be held in sus­
tained flexion (bathing the chi ldren, making beds, sweeping, vacuuming)
and when lifting the children. Sitting and lying eased the pain.
The patient stood with an increased lumbar lordosis; she had a full
range of pain-free movements. Overpressure into full extension was painful
and by combining this with lateral flexion to ei ther side the pain was made
worse on that side. Testing of i ntervertebral movements revealed excellent
Manipulative Physical Therapy 245
mobility with the exception of posteroanterior gliding of L5 on the sacrum,
which was slightly hypomobile. Tests for the neuromeningeal tissues were
normal. Her lower abdominal and gluteal muscles muscles were slack and
This was a patient with a young, mobile spine that became painful
when the tissues restraining flexion were stressed (posterior ligamentous
structures and zygapophyseal joints). There was poor support by the ab­
dominal muscles, which co-contract with the erector spinae and gluteals
to stabilize the spine during flexion. Pain was relieved by rest (when stress
was taken off the painful tissues).
Day 1 (Treatment 1) . An explanation of the cause of the symptoms
was given, the need to strengthen the lower abdominals, gluteals, and erec­
tor spinae was stressed. Large amplitude accessory posteroanterior mobili­
zation of her L5-S 1 joint to restore her normal mobility at this segment
will help this joint to become pain free, but the primary objective is to
strengthen abdominal, gluteal, and erector spinae muscle groups. This was
complemented by postural correction (pelvic tilting) and by giving advice
on correct lifting techniques and how to restore the lumbar lordosis afer
periods of sustained flexion (discussed in more detail in other chapters).
Day 5 (Treatment 2). The patient was seen again to check that she
was performing her exercises correctly and regularly. As her strength im­
proved the exercises were progressively increased in difficulty. Her lumbo­
sacral joint was again mobilized. She reported no change in her symptoms.
She was urged to continue her exercises regularly.
Day 21 (Treatment 3). The patient reported that she was virtually
symptom free, experiencing slight aching if she was excessively busy and
tired. On examination, lumbar extension combined with lateral flexion to
each side was full and painless, as was posteroanterior accessory gliding
of L5. The patient was discharged with the advice to maintain good muscle
support of her spine by regular exercise.
Coccygeal Pain
A 22-year-old female bank clerk complained of a gradual onset over
3 days of localized coccygeal pain. There was no history of lumbar or
pelvic symptoms, trauma, or childbirth.
246 Physical Therapy of the Low Back
The patient' s coccygeal pain was present only in sitting and worsened
if she slouched. Standing eased her pain almost immediately. Her pain was
unaltered by sitting on a hard surface, defecation, or squatting.
The patient' s sitting posture was poor and passive overpressure to the
slouched sitting posture reproduced her coccygeal pain. The addition of
cervical flexion and knee extension did not alter her pain. Also the addition
of posteroanterior pressure to the coccyx (flexion of the sacrococcygeal
joint) did not alter her pain, although the coccyx was very tender.
Lumbar movements were full and pain free to overpressure; however,
there was a loss of intersegmental mobility below L3 on both flexion and
extension. Testing of intervertebral mobility revealed hypomobility, pain,
and spasm on central posteroanterior pressure over LS, but the coccygeal
pain was not reproduced. Pain associated with pressure on LS was unal­
tered by the addition of posteroanterior pressure on L4. There was no
thickening on careful palpation of the sacrococcygeal joint and the liga­
ments attaching to the coccyx.
Coccygeal pain of musculoskeletal origin can arise from the sacrococ­
cygeal joint or ligamentous attachments to the coccyx, secondary to direct
trauma such as sitting heavily on the buttocks, or childbirth, or be referred
from the low lumbar spine. In this case, the lack of coccygeal trauma, pain
on passive movement of the coccyx, and thickening of the sacrococcygeal
joint or coccygeal ligamentous attachments negated a local source of pain.
Local j oint signs at LS-S I incriminated this joint as the source of symp­
toms. This could be verified by treating the LS-S I joint and reassessing
symptom production by sitting.
Day 1 (Treatment 1 ). The explanation was given that the likely source
of symptoms was the lumbar spine and that treatment would entail correc­
tion of sitting posture and restoration of low lumbar joint mobility. Large
amplitude accessory posteroanterior mobilization was applied for 60 sec­
onds to LS to restore normal pain-free mobility. This effected an immediate
lessening of the degree of pain on retesting of overpressure to slouched
sitting. The technique was repeated with further improvement. The session
Manipulative Physical Therapy 247
ended with postural correction of the patient' s sitting posture and she was
asked to set up a protocol by which she could monitor her sitting posture
at work.
Day 4 (Treatment 2). The patient reported increased pain associated
with sitting, for 2 days following her examination and first treatment; since
then her symptoms had greatly improved. On questioning, she reported
that she monitored her posture at the completion of each customer transac­
tion. On reassessment, slouched sitting with overpressure was symptom
free but her LS-S I joint remained hypomobile and painful to posteroanter­
ior pressure testing. Day I treatment was repeated.
DlY 7 (Treatment 3). The patient reported that she was symptom free.
Since her second treatment she had experienced coccygeal aching only
twice, associated with periods of prolonged sitting. Posture correction im­
mediately abolished the aching. The LS-S I j oint was firmly mobilized and
home exercises checked and progressed.
Spinal Stenosis
A 60-year-old ex-nurse complained of a 20-year history of low back
pain for which she had never sought treatment. In the last 6 months she
had experienced vague aching in both lower limbs, which had become
progressively worse to the stage where it now severely interfered with
walking. She found that she could still ride her bicycle without pain. Having
been a nurse she suspected peripheral vascular disease, possibly secondary
to diabetes mellitus. Tests for these were negative. Plain radiographs of
her lumbar spine showed narrowing of the L4-LS disc space and marked
bilateral osteoarthrosis of the zygapophyseal joints at L4-LS and LS-S I .
No spondylolisthesis was present.
Symptoms included deep aching and pain in both calves and the dorsal
and lateral aspects of both feet, worse on the left side. Symptoms were
provoked by standing for more than I S minutes or walking for more than
3 to 4 minutes. They were eased only by sitting.
Signs included excessive lumbar lordotic curve with adaptive shorten­
ing of erector spinae muscles. Extension of the lumbar spine reproduced
low back pain and if sustained with passive overpressure for 20 seconds,
248 Physical Therapy of the Low Back
bilateral calf aching developed. Other lumbar movements were hypomobile
below L3, but pain free. SLR was equal on both sides, but there was an
i ncreased feeling of posterior thigh ti ghtness on the left. There was no
neurologi c defi cit, and peri pheral pulses were normal. Deep palpation
( through the erector spinae) revealed bilateral thickening, hypomobi lity,
local spasm, and tenderness over the L4-LS and LS-S I zygapophyseal
joints. Firm pressure did not reproduce referred leg symptoms.
Gradual onset of bilateral extrasegmental lower extremity pain associ ­
ated with walking suggested vascular or neurogeni c ( cauda equina) etiol­
ogy. A neurogeni c source was supported by the easing of symptoms with
sitting, which wi dens the spi nal canal, and by reproduction of both back
and bi lateral leg pains with lumbar extension. A vascular source was un­
likely because leg symptoms were not eased by standi ng still and her pe­
ripheral pulses were normal.
In consultati on wi th her medical practiti oner, a computed tomography
(CT) scan of her low lumbar spine was taken. Thi s confi rmed the presence
of spi nal canal stenosi s at LS-S I with i ndentati on of the thecal sack.
Day 1 ( Treatment 1). Management consisted of an explanation to the
patient of the likely eti ology of her symptoms and for the need to widen
her spi nal canal by flattening her lumbar lordosi s. Passive mobili zation
of her lower lumbar zygapophyseal joints would help to ease her back
symptoms, possibly by lesseni ng joint effusion.
In crook lying the patient was unable to actively flatten her lumbar
curve; therefore, the tight erector spinae were lengthened using a combi na­
tion of reciprocal relaxation and passi ve stretching techniques performed
at the li mi t of lumbar flexion in the sitting posture. Thi s resulted in the
pati ent being able to achieve some flattening of her lumbar spine and she
was i nstructed to repeat this 10 times at home, four times daily.
Day 3 (Treatment 2) . Symptoms were unchanged. Review of active
posteri or pelvi c tilting revealed an i mproved range of lumbar flexi on but
poor movement below L3. Lengthening techniques for erector spinae were
repeated, but this effected no palpable i mprovement i n low lumbar flexion.
In crook lying with the abdomi nal wall relaxed, anteroposterior mobili za­
tion was applied by slowly depressing the abdomen with the thumb pads
unti l the anterior aspect of the vertebral bodies were palpated. Following
this, there was i mprovement in the passive range of low lumbar fexion.
This was followed by refining the active posterior tilting exerci se to encour­
age recrui tment of the low abdomi nals. The patient was then asked to
Manipulative Physical Therapy 249
gently (submaximal effort) flatt en her lumbar spine by hollowing her abdo­
men. 23 The same regimen of home exercises was prescribed
Day S (Treatment 3) . Symptoms were unchanged. Treatment 2 was
repeated wit h emphasis on re-education of low abdominal control of low
lumbar/pelvic flexion. This included being able to hold t he fatt ened lumbar
position while flexing alternat e hips.
Day 8 (Treatment 4). Symptoms were unchanged and lumbar exten­
sion was the same as on day I . Treatment 2 was repeated with the addition
of unilateral post eroant erior mobilization, performed bilaterally L3 to L5.
Lumbar extension caused less back pain and, after a second application,
ext ension had to be sustained for 5 seconds before producing back pain.
Abdominal hollowing was progressed to standing and maintenance during
Day 1 4 (Treatment S). The patient was now able to st and for 20 min­
utes before leg symptoms developed, and walking was unaltered. The im­
provement gained in lumbar extension from unilateral posteroanterior mo­
bilization had been maintained, so t his was repeated. Following this only
a vague ache was experienced in t he calves after sustaining full extension
for 40 seconds. The avoidance of lumbar lordosis when walking was empha­
Day 21 (Treatment 6). The patient was now able t o walk for 6 minutes
before needing to sit to relieve her leg symptoms. Treatment 5 was repeated
with most time spent on maintenance of a fatt ened lumbar spine during
walking and climbing stairs.
Day 28 (Treatment 7). The patient was able to walk 10 to 15 minut es
before needing to sit. Treatment 6 was repeated.
Day S6 (Treatment 8). When reviewed 1 month lat er patient reported
that she could walk for 20 minut es and t hat st anding to do household duties
was symptom free as long as she remembered to maintain her slightly
fexed lumbar posture. Her lumbar extension and passive intersegmental
mobility L3 to L5 were reassessed and found to have been maint ained.
The need for ongoing abdominal exercises and lumbar posture correction
was emphasized.
Most lumbar dysfunctions have a mechanical component t hat responds
well to carefully applied manipulative physical t herapy. Manipulative physical
t herapy is safe, effective, and an import ant part in t he overall management of
patients with these conditions. However, t he decision to apply manipulative
therapy must be based on a t horough examination, sound judgment of which
techniques to select, and repeated reassessment of t he effects of t hese t ech­
niques if the optimal results are to be achieved.
250 Physical Therapy of the Low Back
I . Bogduk N, Tynan W, Wilson AS: The nerve suppl y to the human l umbar interverte-
bral di scs. J Anat 1 32: 39, 1 98 1
2 . Bogduk N: The i nnervation of the l umbar spi ne. Spi ne 8: 286, 1 983
3. Cyriax J: Dural pai n. Lancet 1 : 9 1 9, 1 978
4. Wyke B: Neurological aspects of low back pai n. p. 1 89. In Jayson MI V ( ed) : The
Lumbar Spi ne and Back Pain. Sector, New York, 1 976
5. Ki mmel D: Innervation of the spinal dural mater and dura mater of the posterior
cranial fossa. Neurology 1 0: 800, 1 961
6. EI Mahdi MA, Latif FY A, Janko M: The spinal nerve root i nnervation and a new
concept of the cl i nicopathol ogical i nterrelat ions i n back pain and sciati ca. Neuro­
chirurgia 24: 1 37, 1 98 1
7 . Pedersen HE, Bl unck CFJ, Gardner E: The anatomy of l umbo-sacral posterior rami
and meni ngeal branches of spi nal nerves ( sinu-vertebral nerves) . J Bone Joint Surg
38A: 377, 1 956
8. Si mmons FH, Segil CM: An eval uation of di scography i n the l ocal i zation of symp­
tomatic l evel s in discogenic disease of the spi ne. Clin Orthop 1 08: 57, 1 975
9. Grieve GP: Common Vertebral Joint Problems. 2nd Ed. p. 1 44. Churchi l l Living-
stone, London, 1 98 1
1 0. Mai tl and GD: Vertebral Manipul ation. 5th Ed. Butterwort hs, London, 1 986
I I . Mooney V, Robertson J: The facet syndrome. Cl i n Orthop 1 1 5: 1 49, 1 976
1 2. McCall I W, Park WM, O' Brien JP: I nduced pain referral from posterior l umbar
el ements in normal subjects. Spi ne 4: 44 1 , 1 979
1 3 . Edgar MA, Nundy S: I nnervation of the spinal dura mater. J Neurol Neurosurg
Psychiatry 29: 530, 1 966
1 4. Edgar MA, Park WM: Induced pain patterns on passive straight-leg-raising in lower
l umbar disc protrusion. J Bone Joint Surg 56B: 658, 1 974
1 5 . Austen R: The distribution and characteri sti cs of l umbar-lower limb symptoms in
subjects wi th and without a neurological defi cit . p. 252. In Proceedings 7th Biennial
Conference, Manipul at i ve Physiotherapists Association of Austral ia, North Fitzroy,
Austral ia, 1 99 1
1 6. Cyriax J : Textbook of Orthopaedic Medici ne. 7th Ed. Vol I
Bai l l iere Tindall , Lon­
don, 1 978
1 7. Butler DS: Mobi l isation of t he Nervous System. Churchi l l Livingstone, Melbourne,
1 99 1
1 8 . Grant R, Jones M, Maitland GD: Cl i nical decision making in upper quadrant dys­
function. p. 5 1 . In Grant R ( ed) : Physical Therapy of the Cervical and Thoracic
Spi ne. Churchi l l Li vi ngstone, New York, 1 988
1 9. McKenzie RA: The Lumbar Spi ne. Mechani cal Diagnosis and Therapy. Spinal Pub­
l i cati ons, Waikanae, New Zeal and, 1 98 1
20. Kos J , Wolf J : Les menisques intervertebraux et l eur role possible dans l es blocages
vertebraux. Ann Med Phys 1 5: 203 , 1 972
2 1 . Bogduk N, Jul l G: The theoretical pathology of acute locked back: a basis for manip­
ul ative therapy. Man Med 1 : 78, 1 985
22. Bogduk N, Twomey LT: Clinical anatomy of t he l umbar spine. 2nd Ed. Churchi l l
Livingstone, Mel bourne, 1 99 1
23. Richardson C, Jull G, Toppenburg R, Comerford M: Technique for active l umbar
stabi l i sation for spinal protect i on: a pi l ot study. Aust J Physiother 38: 1 05, 1 992
9 Rehabilitation of
Active Stabilization of
the Lumbar Spine
Gwendolen A. Jull
Carolyn A. Richardson
Rehabi l i tation of the trunk muscle system is one of the most important
aspects of treatment undertaken by physical therapists to help pati ents regain
function and to prevent recurring episodes of back pai n. A vital functi on of the
muscle system is to support and control the back in posture, movement , and
its load-bearing acti vi ties. Not surprisingl y, current exercise programs empha­
size the aim of enhancing active trunk muscle stabi l izati on. 1-3
The provision of support and control for the l umbar spi ne involves a com­
plex i nteraction between many muscles of the trunk and gi rdl es. Whi l e some
muscles perform and control the primary action, other muscles must work auto­
matically i n synergy to balance any asymmetrical forces, control unwanted
movement , and offer support to articul ar structures. The posi ti onal and di rec­
tional requirements of the task dictate the combi nation of muscles used and
the nature of their work. 4
7 The rehabi l i tation of active trunk stabi l ization is
concerned not only with torque-producing capabi l i ti es of muscles in each plane
of moti on, but more perti nentl y wi th a person' s abil i t y to automati cal l y coordi ­
nate an optimal pattern of muscle acti vi ty to control postures and functi ons
safel y and effecti vel y.
252 Physical Therapy of the Low Back
Supporting Role of Muscles of the Lumbopelvic Area
Al l muscl es of the trunk and pel vi s contribute to trunk control . However,
cl i ni ci ans and researchers recognize that trunk muscles such as the obl i que
abdomi nal s and t he deep transversus abdomi ni s have key roles in spi nal support
and control . 2
14 Thei r anatomic design and location provide a dynamic cor­
setl ike structure suitable for fixation and support . The transversus abdomi ni s,
and to a variable extent the i nternal obl i ques, furt her contribute to l umbar
noncontracti l e supporti ve mechani sms through thei r attachments to the thora­
col umbar fasci a.
12.15 Here they augment sagittal and coronal pl ane control .
The i nternal and external obl i que abdomi nals act in synergy wi th the mul tifidus
to provi de rotatory control of the trunk. 16 However, wi thi n this synergy, i t is
primari l y the obl i ques that have the torque-produci ng capaci ty to provide t he
torsional stabi l i ty in sagittal plane functi on, 17 whereas rectus abdomi ni s and
the erector spi nae produce and control the primary movement. As a group,
these lateral abdomi nal muscl es have the anatomic design to provide and aug­
ment control i n all t hree planes of moti on. Opti mal and safe functi on, especially
i n the sagittal plane, i s commensurate wi th thei r automatic recruitment to a
l evel suffi ci ent to support the t runk when it is required to carry load and to
perform common dai l y acti vi ti es. These muscl es must be also capable of offer­
i ng thi s supporti ng acti vi ty over ti me.
The short one-joi nt muscl es i n the pel vi c region also have the anatomic
design and al i gnment to provi de major support to the l umbar region. For exam­
pl e, of the hip extensors, gl uteus maxi mus would have the prime supporting
rol e. Of the many, often long fusiform muscl es that can perform hip flexion,
the deep i l iopsoas woul d be more suited to the supporti ng rol e.
I n summary, al l muscl es associated wi th the l umbar region must contribute
in some way to i ts stabi l i zati on. It i s argued that the function of some specific
muscl es is more aligned to a supporti ng rol e.
Dysfunction in Muscles in Their Supporting Role
Studi es comparing back pain pati ents and normal control s, although
plagued with variabl es, have shown that the back muscl es and the abdominals
become weaker with chronic back pai n, and more concl usi vel y, that they lose
endurance capaci ty (see Bei mborn and Morri sseyls for revi ew). Al though these
studi es provi de data on gross muscle function, more specific i nformat ion is
required on the pattern and degree to whi ch i ndi vidual muscles contribute to
the dysfuncti on. This is necessary because cl i nical evidence suggests that dys­
functi on may be di sproportionate between muscl es. 19
Al though al l abdomi nal muscl es can become weak, the lateral abdominal
muscl es ( i . e . , obl i que abdomi nal s and transversus abdomi ni s) appear to be par­
ti cularl y vulnerable to l oss of thei r trunk-supporting rol e. 22.23 There is also some
suggestion that lack of control and fatigabi l i ty of these parti cular muscles may
predispose to back strai n. Parnianpour et al24 requi red subjects to repeat [5
Rehabilitation of Active Stabilization of the Lumbar Spine 253
cycles of ful l trunk fl exi on and extensi on as qui ckl y and as accurately as they
could at 70 percent of thei r maxi mum extensi on torque. They found a loss of
preci sion and control , measured parti cul arly as an i ncrease in movement in the
transverse and coronal planes wi th the onset of fatigue. These authors consi d­
ered that this loss of active control may deny protection to the l umbar structures
and could be a mechani sm for i ndustrial and recreational i nj ur y.
Our own i nitial i nvestigat i on of a method t o measure the stabi l i t y capaci ty
of these lateral abdominal muscl es found that 8 of 20 asymptomatic subjects
demonstrated poor automat i c rotatory control of thei r trunks when requi red
to take the weight of one l eg. This was i mproved when subjects consci ousl y
contracted thei r obli ques and transversus abdomi ni s pri or to holdi ng thei r leg
weight .25 Such fi ndi ngs are i n accordance with the cl i nical practi ce of focusi ng
on acti vating these muscl es.
Other muscles i nvol ved in the control of the l umbopelvi c area al so appear
to be vul nerable to loss of thei r supporti ng functi on. Gluteus maxi mus (com­
pared to other hip extensors), gluteus medi us (compared to other hi p abduc­
tors) , and i l iopsoas ( compared to other hi p fexors) are recognized as muscl es
that often show signs of dysfuncti on. Dysfuncti on i n these muscl es together
with the abdominals could contri bute signifi cantl y to the onset of back
pai n.
The conclusion that dysfuncti on occurs specifi cal l y in the supporti ng mus­
cles has been deduced through cl i nical observati on and l aboratory research.
There is a need for physi cal therapists and thei r pati ents to have an objecti ve,
practical assessment of the specific muscle dysfuncti on that can be used easi l y
i n the cl i nical situati on. Such a measurement woul d also al l ow pati ents to moni­
tor their own i mprovement wi th home exerci ses.
Clinical Assessment of Specific Supporting Muscle
Recognition of specific defi ci ts in these muscl es has led to various i ndirect
methods of cl i ni cal l y assessing the muscl es' supporti ng dysfuncti on. These i n­
cl ude postural anal ysi s and assessment of muscl e l engths. Such assessments
are based on the premi se that a muscl e that loses i ts supporti ng role usual l y
lengthens due to the effect of gravi ty. Thi s wi l l i ndi rectl y affect various body
postures. 9
6.27 Movement patterns are also used to detect dysfuncti on through
a change in muscle recrui tment patterns. Muscl es such as gl uteus maximus
may either not be recruited at al l or di splay delayed recrui tment i n movements
such as hip extensi on.
These assessment procedures, when performed in the cl i nic, rel y on the
observational ski l l s of the physi cal therapist and lack object i vi t y. It was there­
fore considered i mportant to devel op additional specific tests for these muscl es
that coul d i l l ustrate a wel l-defined defici t i n muscl e functi on and that coul d be
easi l y and objecti vel y measured in the cl i ni c.
The test advocated i s one that exami nes the muscl e' s abil i ty to acti vate i n
254 Physical Therapy of the Low Back
i solation under low-load condi ti ons and further tests i ts ability to hold an isomet­
ric contracti on. The basis for this test evolved from studyi ng the possible mech­
ani sms that coul d have caused the loss of support and control in these muscl es.
I nsight i nto the physi ol ogi c changes that occur i n antigravity muscles when
subjected to di suse coul d expl ai n cl i ni cal fi ndi ngs and rat ional i ze this objective
test .
proposed that patterns of normal use as wel l as changes that
occur due to pai n and refl ex i nhi bi ti on lead to a progressi ve lack of use in such
muscl es as the obl i que abdomi nal s and gl uteals. Such supporti ng, antigravity
muscl es when subject to di suse are l ikel y to undergo some physiologic changes.
Reduced neural i nput (neural traffi c) seems to affect the slow twi tch fi bers
wi thi n the muscl e to a greater extent than the fast twitch fi bers. 30 Many research
studi es both on ani mal s and humans suggest that the slow-twitch fi bers of a
di sused antigravi ty muscl e take on more of the characteristics and funct ional
role of fast-twitch fibersY-4o
Because it is the sl ow twi tch fi bers within a muscl e that are primarily
concerned with post ural support ,
thei r changing functi on would affect the
support and control offered by the whole muscl e. Therefore, when an antigrav­
i ty muscl e is not used over a period of ti me, i t woul d be reasonable to suggest
that i t would have parti cular diffi cul ty sustai ni ng and control l i ng a low-load
isometri c contracti on (i . e. , 30 to 40 percent maxi mum) without phasi c, errat ic
contracti ons occurri ng.
The test advocated requi res that the target muscle i s assessed i n relative
isolation so that its functi onal status i s not masked by substi tuti on by other
neighboring muscl es. Body posi tioni ng and l i mb load i s used to hel p i solate the
muscl e but no added weight is appl ied. The pati ent is asked to hold an i sometric
contracti on wi th good control . The l ength of ti me this contraction is held while
mai ntai ni ng the limb in a steady position provides an objecti ve measure of any
dysfuncti on of the slow twitch fbers within a muscl e.
This cl i nical test has proved very useful for depi cti ng a functi onal defi cit
in the support i ng muscles. The pel vic-hip muscl es such as the gluteal s can be
tested in some degree of i solation wi th a careful l y appl ied grade 3 classical
muscl e test.
9 More sensi ti vi ty seems to be obtained if the physical therapist
passi vel y l ifs the l i mb i nto the i nner range, then requi ri ng the pati ent to hold
the posi ti on. The pati ent ' s abi l i ty to activate the muscl e and mai ntain the con­
tracti on wi thout l osi ng control of l i mb position can then be assessed.
In relati on to the obl i ques/transversus group, a special low-load test had
to be devi sed that woul d test thei r capaci ty sel ecti vel y wi thout contribution
from rectus abdomi ni s. Thi s was a chal l enging task, because the classical tests
for the obl i que abdomi nal s i nvol ve a si gnificant contribution by often more
acti ve muscl es. For exampl e, the external obl i que abdominals are usual l y tested
wi th a trunk curl -up wi th rotati on acti on. 4
This would necessari l y i nvol ve rec­
tus abdomi ni s contracti on. The i nternal obl i ques are tested by i psi lateral pelvic
t i l t i ng and rotation toward the contralateral si de.4
The obl i que abdominal acti­
vation woul d be di ffi cul t to di fferentiate from the contraction of rectus abdom­
i ni s and quadratus l umborum in such a test . It was the muscl e test for transver-
Rehabilitation of Active Stabilization of the Lumbar Spine 255
sus abdomi ni s that provided the possible answer. Lacote et al4
used the act ion
of sucking i n the stomach and depressi ng the abdomi nal wal l to test transversus
abdomi ni s funct ion. This i s a si mi lar acti on to drawi ng i n the stomach suggested
by Kendall and McCreary
9 as a method of act i vat i ng an obl i que abdomi nal
contraction. It was proposed that such a test may act i vate both the obl i que
abdominals and transversus and separate thei r acti on from rectus abdomi ni s.
Studies were undertaken to test t hi s assumpti on. Mul ti channel el ectro­
myography (EMG) was used to i nvestigate this and other methods of act i vat i ng
the supporting muscl es of the trunk. 25.42 The resul ts revealed that setti ng the
abdominals by drawing in the stomach and t ighteni ng the waist9 act i vated the
obliquesltransversus abdomi ni s muscl es and most i mportant l y di ssociated thei r
acti vi ty from that of rectus abdomi ni s. A si mi lar pattern of acti vi ty was demon­
strated with an abdominal braci ng acti on.
These techni ques fulfi l l ed the re­
quirements of the test in that they separated the lateral abdomi nal s from rectus
abdomi ni s, but objecti ve methods of quantification were lacki ng as wel l as a
method of assessi ng the hol di ng capaci ty or fati gabi l i ty of these muscl es. A
new assessment tool had to be devi sed to al l ow these two parameters to be
easi l y quantified in the cl i nical si tuati on.
This need l ed to the development of a si mpl e pressure sensor ( Stabil izer,
Chattanooga, Austral ia) . It consi sts of a t ri sectional , si ngle-cell uni t made of a
nonelastic material . The sensor is i nserted between the l ow back and exercise
sUiface and i s i nfated to fi l l the irregularl y shaped space. It operates on the
pri nciple that body movement or change of position i n any plane causes vol ume
changes i n the cel l , whi ch are measured as pressure changes.
Out studi es4
i n the supi ne crook l yi ng and recli ned si tti ng posi ti ons demon­
strated that the pressure sensor can be used as a cl i nical measure to i ndi cate
the successful , or not , act i vat ion of the lateral abdomi nal musculature or to
detect if incorrect substi tuti on strategies are used (Fi g. 9-1). From a basel i ne
pressure of 40 mmHg ( i . e . , the pressure in the cel l that fil l s the space behi nd
the back giving the patient an awareness only of i ts presence), the correct
abdominal sett i ng or braci ng act ions cause a sl ight flatteni ng of the l umbar
spine, which registers as a pressure i ncrease of approxi matel y 10 mmHg. I nabi l ­
ity to acti vate the muscl es registers as a ni l i ncrease i n pressure. Inappropriate
recrui tment of rectus abdomi ni s, to substi tute for the correct muscl e contrac­
ti on, causes posterior pel vi c ti l t and l umbar flexi on and this resul ts in a more
marked i ncrease i n pressure ( e. g. , up to 20 mmHg) . 4
2 The pressure sensor can
be used in conjunction wi th mul ti channel EMG to check the correct act i vation
of the abdominal muscles.
The hol di ng capacity or fatigability of these muscl es i s j udged by the t i me
that the correct acti vati on i s held. There wi l l be a sl ight fuctuati on (± 2 mmHg)
registered by the pressure sensor wi t h the movement associated wi th breathi ng,
but fat igue i n the muscl es can be i nstantl y detected by ei ther a gradual or often
quite rapid l oss i n pressure. A gradual switch to rectus abdomi ni s acti vi ty i s
di scernible by a subtle , gradual l y i ncreasing pressure over ti me. In thi s way, the
pressure sensor provides both a measure and feedback system for the pati ent.
This test of abdominal sett i ng by drawi ng i n the stomach and tighteni ng
256 Physical Therapy of the Low Back
Fig.9-1. The pressure sensor ( i nserted) used in the test of the abdominal setting action.
the waist or by a braci ng action, as a test of the acti vation of the lateral abdomi­
nals and their hol di ng capaci ty, appears to be very relevant to these muscles'
functional supporting and post ural rol e. I t i s proving useful as a cl i nical test.
However, further refi nement of the techni que is needed for its validation as a
research tool .
When the dysfuncti ons i n the abi l i ty to consci ousl y activate and statically
hold and control a muscle contraction are detected i n muscles such as the lateral
abdominals and key girdle muscl es, it provides a specific focus for the initial
stages of the stabi l i zation program. We bel i eve that these muscles require spe­
cific trai ni ng in isolation in the first i nstance, to enhance their activation and
to train their supporti ve role. This i nitial step is necessary to ensure their appro­
priate parti cipation in higher l evel s offuncti on where they are required to coacti­
vate with all trunk and girdle muscl e to support and protect the spi ne.
The aim of thi s stabi l i zation program is to re-acti vate the stabi l izing mus­
cl es, retrain their hol di ng capaci t y, and retrain their ability to automatical l y
contract appropriatel y wi th other synergists to support and protect the spine
under various functional loads and ski l l s. Based on our research and cl inical
observations, a four-stage progressive approach to stabi l ization training has
Rehabilitation of Active Stabilization of the Lumbar Spine 257
been developed. Rehabi l i tati on is commenced at the l evel appropriate to the
patient and progressed sequential l y through higher degrees of loaded control
and ski l l . Progression i s guided by cont i nuous reassessment.
I n focusi ng on a parti cul ar aspect of rehabi l itati on, i t i s easy to detract
attention from the often compl ex nature of the physi cal dysfuncti on that may
be present in the back pai n pati ent. The art icular, neural, and muscl e systems
and their central nervous control are i nterdependent i n function and dys­
In the first i nstance, pai n, swel l i ng, and pathology have powerful infl uences
on the muscl e system and muscl es qui ckl y react by either i nhi bi ti on or spasm. 43
Lumbar osseoligamentous stabi l ity is aided by an opti mal return of movement
to the l umbar joi nts so that the joints can share the load created in normal
functional acti vi ti es. The most obvi ous i l l ustration of thi s necessi ty i s demon­
strated in the i nstabi l i ty that can occur i n the segment adjacent to fusi on sites
i n the l umbar spi ne. 44 Likewi se, the range of movement of the joi nts and ti ght­
ness in the muscl es of the pel vic gi rdl e, hi ps, and lower l i mbs can exert substan­
tial i nfl uences on l umbar spine posture and movement . Probl ems i n these struc­
tures are often present i n the back pai n pati ent and are part of the total
Movement and muscle control of these regions are cl osel y
related. Therefore, any rehabi l itation program must eval uate and address the
total probl em, and emphasis must be placed on the component of rehabi l itation
of active trunk stabil izati on.
Stage 1: Isolation and Facilitation of Target Muscles
In the i ni tial assessment of the obl i que abdomi nal s and transversus abdom­
i ni s, i t i s found that many back pain pati ents have an i nabi l i t y to i solate, acti vat e,
hol d, and control a sett i ng contracti on. This consti tutes a probl em at a very
basic level and these dysfuncti ons must be addressed and abi l iti es restored
before the patient can progress further into the program. The patient trai ns to
gai n consci ous control of these muscl es at submaximal l evel s i n this stage.
It is not uncommon to fi nd that pati ents often experience i ni ti al diffi cul ty
i n re-acti vating the muscl e pattern of the obl i ques and transversus abdomi ni s
voluntari l y. There does not seem to be a si mi l ar probl em i n activati ng rectus
abdomi ni s,3 and pati ents wi l l readi l y substi tute wi th thi s muscl e acti on. I t i s
therefore i mportant from the outset that the patient has a consci ous awareness
and perception of the correct activation i n the setti ng contracti on and can them­
selves detect when their performance i s i ncorrect or i neffect i ve. The emphasi s
i n thi s stage is on accuracy. There i s not an emphasi s on effort , because thi s
wi l l encourage recrui tment of inappropriate muscl es near and far.
The first essential step i n retrai ni ng i s good demonstrati on and verbal i n­
structi on, because different patients respond to different verbal cues. Varyi ng
i nstructions such as drawing in and hol lowi ng the lower abdomen , drawi ng the
navel up and i n toward the spi ne, II or feel i ng the muscl es ti ghten at the wai st
may vari ousl y cue the pati ent . When the pati ent uses substi tuti on strategies
258 Physical Therapy of the Low Back
for the correct muscle acti on, these must be ident ified and explai ned to the
pati ent . As previ ously mentioned, the most common is the often subt l e substi tu­
tion wi th rectus abdomi ni s that is observed by a posterior pel vi c ti l t acti on, a
depression in the anterior rib cage, and a tendency for a roundi ng of anterior
abdominal wall rather than a hol l owing. Al ternati vel y, patients may effecti vel y
mi mi c the setti ng acti on by i nhal i ng and merely el evati ng thei r rib cage, which
makes thei r abdomen look fat . From the begi nni ng, the patient must be taught
to dissociate breathing from the setti ng action and learn to breathe normal l y
whi l e activat i ng and hol di ng the abdominal set .
As the patient trains to i mprove the holding capaci ty of these muscles, it
is not al ways easy to know whether the abdomi nal setti ng action is being per­
formed correctl y or when muscle fat igue is occurri ng. The vi sual feedback
provided by the pressure sensor i s provi ng to be a very potent monitoring and
feedback system. I t is used in both the teaching and practice of the setting
action, provi di ng feedback to the patient, through the pressure readi ngs, of
thei r successful or unsuccessful performance. An i nabi l i ty or poor abi l i ty to
act i vate the muscles registers as a nil or mi ni mal i ncrease i n pressure, whereas
substi tuti on wi th rectus abdomi ni s wi l l cause pressure changes greater than
those associ ated wi th the isolated contracti on of the obl i ques and transversus
abdomi ni s. In this way, the pressure sensor assists the pat ient i n the motor
learning process for acqui ri ng this abdominal setting ski l l and for trai ni ng the
hol di ng capacity of these muscl es. This quantification of performance thus helps
in the teaching process and also enhances compl iance, especi al l y when pract ic­
i ng at home.
Motor learning abi l i ties and the l evel of dysfuncti on i n the lateral abdominal
muscl es wi l l vary between patients, and physical therapists wi l l use the various
faci l itation techni ques at thei r di sposal to encourage conscious activation of
the muscles. When pati ents are havi ng di ffi cul ties i nitial l y, the easiest positions
to teach conscious muscl e activation is in four-point kneel or prone l yi ng (Fig.
9-2). These positions use the forward drift of the abdomi nal contents as a stretch
faci l i tation to the sett i ng acti on. In the prone position, visual feedback can
augment the faci l itati on. The pressure sensor i s placed under the abdomen and
i nfated to a basel i ne pressure of approximately 70 mmHg. Instead of tryi ng to
i ncrease pressure, as required when the sensor i s behi nd the back, the patient
sets the lateral abdominals and draws the stomach off the pad, ai mi ng for a
decrease i n pressure of at l east 10 mmHg.
Manual guidance subtl y di rected through mul tifidus in the prone, standi ng,
and si tti ng posi ti ons can al so assist activation of the muscle pattern and can
be appl i ed by ei ther the therapist or patient . This action di rectl y guides the
very subtl e l umbar fexi on that occurs with the abdomi nal setting action and
also directl y faci l i tates the mul tifidus, which cocontracts with the obl i ques to
support the spi ne.
The abdomi nal setti ng acti on is also taught and practiced i n the supported
standi ng and sitt i ng posi ti ons as wel l as in supi ne crook l yi ng. This i s a necessary
component for postural retrai ni ng and for future exercise. Faci l itation tech­
ni ques such as sweep tappi ng in conj uncti on with vi sual feedback can be helpful
Rehabilitation of Active Stabilization of the Lumbar Spine 259
Fig. 9-2. Abdominal setting in four-point kneel i ng position.
here. Al though a IO-mmHg i ncrease in the pressure sensor is sought with cor­
rect activation in these posi ti ons , pati ents may achieve lesser i ncreases i n the
prel i mi nary stages. These readi ngs wi l l gradual l y i mprove as pat ients enhance
their ski l l s. While the abi l i ty to act i vate the lateral abdomi nal muscl es is re­
qui red in all the posi ti ons mentioned, pati ents must start trai ni ng in those where
they can more easi l y achieve the i solated contracti on. The emphasis is on attain­
able goals, because unachievable tasks will encourage substi tuti on strategi es.
If patients cannot consci ousl y acti vate the lateral abdomi nal muscl es and
when marked weakness i s considered to be a major probl em, i t may be neces­
sary to begin retrai ni ng by usi ng di rect resi stance to the trunk to encourage
activation . It has been shown that a rotatory resi stance applied to the trunk
via the pel vi s or shoulders will di rectl y activate the supporti ng muscl es with
l i ttl e contribution from rectus abdomi ni s. 14 It i s also very appropriate to use a
technique that emphasizes faci l itation of the mul tifidus wi thi n the cocontracti on
pattern because localized dysfuncti on i n this muscl e has been identi fi ed. Usi ng
diagnostic ul trasound and i magi ng, a study of pati ents with thei r fi rst epi sode
of acute uni lateral back pai n revealed that there i s uni lateral i nhi bi ti on wi thi n
thi s multifasci cl e muscl e, primari l y at the l umbar l evel of dysfuncti on. 50 Provi d­
ing a reason for this selective segmental inhi bi ti on i s chal l engi ng and i t may
represent pain inhibi ti on mediated via a long loop refl ex pathway. 51 Neverthe­
less, these fi ndi ngs do provide a basi s for i ncl udi ng exerci se techni ques that
aim to faci l i tate mul tifi dus at the dysfuncti onal segmental l evel (Fig. 9-3).
The resistance applied in any of these exercises shoul d be l ow load and
faci l itatory, with the patient encouraged to sustain the submaxi mal contracti on .
Rhythmic stabi l i zations and alternating isometrics are al so sui table techni ques
to empl oy to activate parti cular supporti ng muscl es. 52 Whi l e exercises usi ng
direct resistance proceed, the pati ent cont i nues to work to achieve a consci ous
isolated activation of these muscl es.
260 Physical Therapy of the Low Back
Fig. 9-3. The mul tifidus is faci l i tated via a gentle rotatory resistance applied to the
dysfunctional segment .
The second phase of this initial stage of rehabilitation is directed toward
improving t he holding capacity of these muscl es. Once the patient can con­
sciousl y activate the abdominal set , they practice to increase t he holding time of
this contraction. Preference is given to practicing repeated holding contractions
short of muscl e fatigue. This is to ensure accuracy of training to the lateral
abdominal muscles and to prevent the recruitment of unwanted synergists that
fatigue may provoke. For example, the patient may practice to hold for 5 sec­
onds, rest 5 seconds, repeating the sequence 10 times and building up the se­
quence to 10-second interval s. Whenever possible they monitor their perfor­
mance with the pressure sensor and cease the instant loss of control or
substitution is detected.
The hol ding capacity of these muscles can al so be trained by monitoring
the control of the abdominal setting action under very low loads provided by
unilateral leg movement . A leg movement such as abduction/external rotation
( Fig. 9-4) provides a low force that tends to rotate the pelvis and trunk. This
addition of a low facilitatory eccentric load is a good method for measurement
of holding abil it y. A loss of control during a continuous abdominal setting con­
traction while performing repetitions of this leg movement resul ts in rotatory
movement of the pel vis and l umbar spine. This movement can be easily detected
either t hrough observation or use of feedback from the pressure sensor.
Repetition is essential to this retraining process . Formal practice of these
types of exercises should be undertaken at least twice a day to monitor and
reinforce effective performance. This is augmented by incorporating regular
repetitions of the holding contraction into dail y activities. This helps the regi­
men to not be too invasive on a patient's time . The repeated activation and
endurance training of the lateral abdominal supporting muscl es is also intimately
related to the re-education of static and dynamic postural control and form.
A similar dysfunction to that found in the lateral abdominal muscl es is
often detected in pel vic girdle muscl es such as the gluteals and iliopsoas . It is
treated with the same approach of facilitating the muscl e in isolation at low
Rehabilitation of Active Stabilization of the Lumbar Spine 261
Fig. 9-4. Monitoring the ability to sustain an abdominal set wi th l ow levels of leg load.
Leg rotation in the prone position is also used.
loads and training its holding capacit y. Stability of the whol e l umbopelvic area
is dependent on the support of these muscl es, and their rehabilitation is in
tandem with that of t he lateral abdominal s.
This first stage of conscious activation and control of key muscl es of the
trunk and girdle at low l evel s of effort must be achieved before the trunk stabili­
zation program can be progressed. The stage can be frustrating for t he patient
but if these muscl es are not capable of functioning at the low l evel s, their
appropriate contribution to more complex motor behaviors and loads is un­
likel y. The muscl e activation is commenced immediatel y in the rehabilitation
program even in the treatment of many acute back pain patients. The minimal
loads involved in the exercise techniques do not overload injured struct ures.
The added benefit is that control of pain i s often commensurate with the re­
activation of the muscle support system.
Stage 2: Training of Trunk Stabilization Under Static
Conditions of Increasing Load
Following their activation in isolation, the lateral abdominal muscl es are
now trained in their stability role under more functional conditions. This in­
volves retraining the total motor program used for trunk stabilization, which
requires a balanced interaction between the lateral abdominal s, rectus abdom­
inis, and the back muscl es.
262 Physical Therapy of the Low Back
The aim of this stage is to enhance muscl e control for trunk stabilization
by gradually adding load both directl y to the trunk and indirectl y with the use
of limb l oad. This second stage of stabilization training requires the trunk to
be held and controlled in a mid position and continues to train the static capacity
of the trunk muscl es.
The static mode of training as the initial functional level has several advan­
tages. From a pathoanatomic viewpoint , the l umbar spine is maintained in a
more mid- or neutral position. This lessens the risks of adverse stress on the
inj ured or pathol ogically compromised l umbar articul ar structures that may be
provoked by through range or end-of-range exercises. Training can be pro­
gressed to t his l evel even in the early stages of a treatment program.
The l evel and effectiveness of a patient's muscle control of trunk position
for stabilization can be monitored objectively with the pressure sensor in this
static method oftraining. I t is placed behind the back to ensure that stabilization
of the lumbopel vic region is maintained during the progressively loaded exer­
cise. The principle employed for the use of such a device is that if the trunk
muscle control is at an appropriate level and balanced between muscles for the
load applied, no movement in the lumbopelvic region will occur and hence there
will be no pressure variation during the exercise. In dynamic trunk exercise, it
is very difficul t if not impossible to quantitatively evaluate if this balance of
activity is occurring between the muscl es producing and controlling the primary
movement and those having a supportive function in that plane.
Another important consideration when training an interaction of muscle
activity under conditions of increasing loads is that too much resistance on a
weakened muscl e can be inhibitory. 19 As already discussed, it is often found
cl inically that the oblique abdominals and transversus abdominis are often
either not automaticall y recruited appropriatel y or demonstrate weakness in
the back pain patient. With too much load, the patient can easily lose control
of trunk stability or substitute with inappropriate muscles and lose t he correct
synergistic interaction of the trunk muscles. Having t he accuracy provided by
the pressure sensor in the clinic, it has clearly emphasized that back pain pa­
tients often have to start training at very low loads, because it is onl y at these
l evel s t hat they can maintain trunk control initial ly. The objective monitoring
system also hel ps the physical therapist and the patient resist the temptation
to add too much load too quickl y, but rather concentrate on precision and
control . By accuratel y monitoring the l evel of control of the coac
ivation of
t he trunk muscl es , the pressure sensor is used to direct progression of the load
safely and appropriatel y during t his stage of stabilization training.
A great deal of emphasis is given to training active trunk stabilization with
leg load resistance, because this allows monitoring of control , the feedback
hel ping the patient to focus on the concept of muscl e coordination and control
rather than the more familiar strengthening types of exercise.
Static Training with Leg Loading
In the low-load leg exercises, patients continue to consciousl y set their
abdominals prior to and during the exercise to ensure activation of the obliques
and transversus abdominis in the total trunk motor program. They monitor their
Rehabilitation of Active Stabilization of the Lumbar Spine 263
A __ ii l
B . _____ _
Fig. 9-5. Crook l ying, low-l evel uni lateral leg loads usi ng straight or diagonal lifts.
Arms should not provide trunk support. (A) Si ngl e leg l if to 90° hi p fexion and lowering.
(B) Single leg extension and fexion movement .
performance with the pressure sensor and cease the exercise i f control is lost
or with the onset of fatigue. I t should be noted that preference is given to
exercises invol vi ng unilateral leg-loading activities. The leg movement can be
in the sagittal plane or the leg placed in some degree of abduction to allow
diagonal movement. Such uni lateral exercises highlight the demand for rotatory
control by the lateral abdominal muscl es to compl ement the acti vity of rectus
abdominis and the back extensors ( Fig. 9-5).
Progression of leg loading can be given in a variety of starting positions
( Fig. 9-6). Attention is si mul taneousl y given to correct activation and control
of the girdle muscles during these exercises. For exampl e, both trunk control
and appropriate gluteus maximus acti vi ty should be monitored in prone, leg
extension exercises. Load is gradual l y and careful l y increased to both the trunk
and girdle muscles by methods such as increasing limb lever l ength or by adding
resistance to the limbs with el astic straps or light weight s.
Once t he patient has progressed through t he program to the higher l eg load
exercises ( Fig. 9-7), automatic activation of the lateral abdominal muscl es in
balance with rectus abdominis shoul d now be occurring. To reinforce thi s bal­
ance, the patient aims to maintain a steady reading on the pressure sensor as
well as maintai ni ng a fattish abdominal wall during t he exercise.
Additional types of exercises are given in thi s stage in conjunction with
264 Physical Therapy of the LolV Back
Fig. 9-6. Exampl es of leg loading activities in the sagittal pl ane: (A) standing unil ateral
hip flexi on, (B) si tti ng unilateral hip flexi on, (C) prone l yi ng uni l ateral hip extension.
l eg-loading activities. This is necessary so that the patient learns muscle control
in a variet y of situations.
Static Training with Direct Resistance to the Trunk
Isometric trunk exercises using such techniques as rhythmic stabilizations
in a variety of starting positions are useful exercises for the clinic. 52 Control
of the lumbar spine in a neutral or mid position continues to be emphasized.
Rehabilitation of Active Stabilization of the Lumbar Spine 265
Fig. 9-7. An exampl e of higher l evel of unilateral leg loadi ng.
This type of exercise can be adapted for self-application at home ( Fig. 9-8).
The manual resistance applied by either the therapist or t he patient at home
should be well controlled, the magnitude of resistance aimed at working the
trunk muscles at approximatel y 30 percent of t heir maximum in line with their
holding and controlling rol e. At later stages , the speed of change of application
of resistance can be increased to facilitate automatic postural responses.
Static Training Using Diferent Body Positions
Trunk muscle control in the neutral position of t he l umbar spine is trained
in a variety of body positions, with the four-point kneel ing and bridging posi­
tions being especial l y useful . Exercises are progressed through increasing de­
grees of difficul ty, introducing more unstable bases and making more strength
and endurance demands on the trunk stabilizers ( Fig. 9-9).
Fig. 9-8. Self-applied rotatory resistance in the bridge position while the patient con­
sciously set s the abdomen.
266 Physical Therapy of the Low Back
Fig.9-9(A&B). Static training in different positions. The patient must control the trunk
i n a neutral position.
Static Trunk Control in Conjunction with Other Treatment
Trunk control is emphasized and monitored in other components of the
total management program of the back pain patient. This is particul arl y relevant
to muscle-lengthening techniques where the shortened muscle has attachments
to either the l umbar spine or pel vic girdl e. Such muscl es include the hamstrings,
tensor fascia lata and iliotibial band, rectus femoris , the iliopsoas (when tight)
and upper l imb muscl es such as latissimus dorsi (Fig. 9-10). Substitute move­
ment or loss of control of t he neutral position of t he l umbar spine can mimic
apparent muscl e l ength gains . This not onl y reduces the effectiveness of muscle
l engthening techniques, but more importantl y can make the l umbar joints vul ­
nerable to strain. Likewise, trunk control is emphasized in exercise used to
strengthen t he l ower or upper limb muscul ature.
Throughout al l exercises in this second stage of trunk stabilization training,
Rehabilitation of Active Stabilization of the Lumbar Spine 267
Fig. 9·10. Lengthening of hamstri ngs in sitting combined with training for active trunk
control .
t he patient works t o control a mid· or neutral position of the l umbopelvic com­
plex and attention is simul taneously directed toward the correct activation of
the girdle muscl es. Substitution strategies t hat the patient may attempt to adopt
to control t he trunk position need to be careful l y monitored, explained to the
patient , and corrected.
Stage 3: Development of Trunk Stabilization During Slow
Controlled Movement of the Lumbar Spine
Once stability has been trained through static stabilization procedures, the
program is progressed to introduce control l ed trunk movement (i. e. , movement
of the trunk with appropriate activation of the supporting muscles) . If progres­
sive training has been followed, automatic protective muscl e stabilization
should now begin to occur during movement . If necessary, the patient can still
voluntarily activate the supporting muscl es, because movement is slow enough
to gain conscious feedback for activation.
Trunk movement i s resisted in each plane of motion. Resistance or load
268 Physical Therapy of the Low Back
Fig. 9-11. Slow control led curl-up with emphasis on the correct abdominal interaction.
can be applied to the trunk muscl es using body weight and exercise apparatus,
and through functional retraining. Exercises such as curl-ups and curl-ups with
rotation are appropriate at this stage, provided that t hey are undertaken at a
slow speed and the l ateral abdominal muscl es are active with rectus abdominis
( Fig. 9-11). A tlattish abdominal wall during the curl -up is indicative of a more
balanced activity. I I Trunk control during back extension exercises can be en­
hanced by incorporating activation of the lower scapul ar stabil izers and lateral
abdominal muscl es .
I sokinetic apparatus can be used in this stage to provide accommodating
resistance to through range trunk
movement s. Exercise can be conducted in
each of the sagittal , coronal , and importantl y, transverse planes. Programs can
be instituted for both concentric and eccentric work.
This stage al so introduces t he more advanced functional retraining or work
hardening. Emphasis is on qual it y of movement and muscl e control . Lifing,
carrying, pushing, or other activities as required by t he patient's occupation
are gradual l y progressed to heavier l oads.
Physical t herapists are familiar with the types of exercise and functional
retraining that are performed in this third stage of stabilization training. The
important issue is that with movement , there are difficul ties in objectively moni­
toring if the appropriate l evel s of activation are occurring in the trunk muscles
in their stability role. For t his reason it is prudent not to progress to this stage
until the patient can demonstrate trunk muscl e control under the measurable
static positions of stage 2.
Stage 4: Lumbar Stabilization During High-Speed and
Skilled Movement
When the patient's sport or occupation involves high-speed or skilled
movement s, it is necessary to ensure that there is adequate stabilization and
muscl e control for these activities. These high-speed activities should not be
Rehabilitation of Active Stabilization of the Lumbar Spine 269
introduced too earl y i nto a treatment program, and i ndeed more ball i sti c-type
exercises should not form part of a program to enhance stabi l i zati on. High­
speed phasic act i vi ty is the anti thesi s of the tonic conti nuous acti vi ty requi red
for joint support .
This delay in i ntroduci ng high-speed movement is i n response to several
factors . Our approach to the retrai ning of stabi l i zation has been essential l y
based on the abi l i ty to monitor if t he supporti ng muscl es are capable of control­
l i ng the lumbopel vi c area, and secondly on t he patient ' s ability to consciousl y
correct and retrain the muscl es i n order to gai n automatic act i vation in the
supporting muscl es. Both these el ements become very difficul t to achi eve when
the patient i s performing fast , hi ghl y ski l led acti vi ti es and exerci ses. For thi s
reason the first three stages of the program are more suited to the training of
dysfunction in the supporting muscl es .
In addition, the acti on of the muscl es aligned wi th stabi l i ty may be ad­
versel y affected by a person performi ng a predominance offast repeti ti ve move­
ments. This may even be a factor in t he development of poor postural control
and supporting abi l i t y. It i s perti nent to briefl y revi ew some studies on t he
effects of fast movements on the muscl e system, because t hei r resul ts expl ai n
why an emphasis on thi s stage of rehabil i tation i s unwi se and why those engag­
i ng i n various trai ni ng and fitness programs i nvol vi ng a predomi nance of such
acti vi ties need to be closel y monitored.
Research has been di rected to muscl es of the knee, ankl e, hi p, and more
recentl y the trunk. Richardson and Bul lock5
studi ed the effect of i ncreasi ng
speeds of bal l i stic knee flexi on-extensi on movements on the quadriceps muscu­
lature. They found that wi thi n the knee extensor muscl e group, i ncreasi ng
speeds of movement were associated wi th i ncreasingl y hi gher l evel s of rectus
femoris acti vi ty. The vast us medial i s and vastus lateral i s, i n contrast , demon­
strated no i ncrease i n acti vi ty levels with i ncreasi ng speeds of knee movement.
These resul ts highl ight a possi bl e relationshi p between rapid knee joi nt move­
ment and reduced acti vi ty in t he surroundi ng one-joi nt support i ng muscl es.
Similar fi ndings have been described for the ankl e plantarfexors wi th the
two-joint gastrocnemi us more activated i n fast repeti ti ve pl antar flexi on than
the antigravi t y, stabi l i zi ng muscl e, soleus. Smi th et aJ54 found that the gastroc­
nemius of the cat performed the very rapid movements of paw shaki ng whi l e
soleus was i nhi bited. A study on humans confirmed t hi s pattern of muscl e
acti vation. 55 Subjects were trained over a 4-week period i n rapid pl antar fl exi on,
heel-raising movements i n standing. Significant i ncreases i njumpi ng height abi l ­
ities were found but these were accompanied by a si gnificant l oss of stati c
function of the sol eus muscl e. Repeati ng rapi d pl antar fexi on movements coul d
reduce the stabi l i ty role of the soleus .
It appears that gl uteus maxi mus may react i n a si mi lar way when exposed
to repeti ti ve, fast hip extensi on acti vi ti es. A study of el i te track cycl i st s deter­
mi ned that these subjects had a significantl y reduced abi l i ty to hold an i nner
range gluteus maxi mus contraction when compared to a noncycl i ng control
population. 5
I n fact , most cycl i st subjects coul d not hol d the posi tion at al l .
There is also evidence that repeti ti on of fast trunk movements may be
270 Physical Therapy of the Low Back
related to reduced trunk stabi l izati on. Thortensson et al57 reported that i ncreas­
ing the speed of trunk flexi on movement resulted in i ncreased rectus abdomi ni s
act i vi t y, whi ch was not accompanied by si milar i ncreases i n the obl i que abdomi­
nal muscl e acti vi t y. A recent study i nvestigated the effect of speed of perfor­
mance of curl -up exercises on subjects' abi l i ty to acti vel y stabil ize thei r l umbar
spi ne. 5
Thi s was measured by the abi l i ty to hold the l umbar spi ne steady duri ng
leg extensi on movements in supi ne whi l e load was i ncreased by i ncrements of
leg weight . I t was shown that subjects who regularly performed curl-ups at high
speeds demonstrated lower l evel s of act i ve stabi l i zati on, whereas higher levels
were demonstrated by t hose subjects who regularly performed curl-ups in a
sl ow and control l ed fashi on.
The resul ts of these studi es of the effects of high-speed movement on the
nature of muscle acti vi ty highl i ght the i mportance i n stabil ization trai ni ng of
progressi ng the pat i ent slowly and methodi cal l y through the earl ier stages of
t he program. Thi s i s to ensure that automatic and appropriate l evel s of stabi l i za­
ti on can be achieved before thi s more advanced fi nal stage. Thi s fourth stage
of trai ni ng is pri nci pal l y requi red for those pat i ents whose l ife-styl e demands
i ncl ude fast and ski l l ed movements ( e. g. , sports persons or those i nvolved in
hi gh l evel s of physical act i vi t y) . It is not routi nel y given to al l pat ients . When
requi red, retrai ni ng i s approached by fi rst progressi vel y i ncreasing the speed
at whi ch various formal exercises for the trunk or l i mbs are performed. This
can then be transferred to i ncreasing the rate at which funct ional acti viti es are
attempted. I n thi s way the rehabi l i tation of trunk stabil izati on can be progressed
to thi s highest l evel of speed and ski l l .
Physi cal therapi sts worldwide recognize the i mportance of retrai ni ng acti ve
trunk stabil ization wi thi n the total management program of the back pain pa­
ti ent, as wel l as i n preventati ve programs. Thi s is reflected both in the l iterature
and in current i nstructional programs. Thi s type of trai ni ng, whi ch i ni ti al l y
emphasi zes control i n a mi dtrunk posi ti on, has the advantage of al l owi ng exer­
ci se to be i ntroduced earl y in the treatment program.
The key factors that have i nfuenced the devel opment of the program pre­
sented here are evol vi ng from our research. Thi s has been concerned with
methods of i dentifyi ng and measuring the nature of dysfunction of muscl es, i n
thei r supporti ng rol e, t hrough an i sol ated test of thei r acti vation and hol di ng
capaci t y. Addi ti onal l y a pressure sensor has been developed as a simple cl i ni cal
method to measure the compl ex i nteracti on of all trunk muscl es to provide
support for t he trunk by moni toring a steady trunk position when di stal load
is applied.
Provi di ng methods by whi ch both the physical therapist and patient can
appreciate the muscle dysfuncti on, in objecti ve terms, has given a posi ti ve step
forward in enhanci ng the effecti veness and effi ci ency of stabil izat ion exerci ses.
Physi cal therapi sts are able to quantify the dysfuncti on, and therefore plan a
Rehabilitation of Active Stabilization of the Lumbar Spine 271
suitable level of exerci se for the pati ent . In addi ti on, patients are most compliant
when they can understand t he dysfuncti on i n their muscles and can monitor
their own progress duri ng rehabi l i tati on. This i s vi tal because the rel earni ng
and retrai ni ng of t he abi l i ty of the t runk muscles to opti mal l y support the spine
may take some pati ents several weeks and may al so requi re long-term exerci se
trai ni ng to prevent the recurrence of thei r pai n.
There are mUl ti pl e causes of muscl e dysfuncti on and t he l oss of optimal
control for trunk stabi l i zati on. Thi s program i s presented as a step forward i n
the gradual understandi ng of a compl ex probl em. I t i s one that bl ends cl i nical
effectiveness with a rationale based on current research. It i s hoped that i t
stimulates further research i ni ti ati ves i nto the effecti vefess of physi cal therapy
exerci se programs for t he rehabil itation and preventi on of back pai n.
I . Saal J A, Saal JS: Nonoperative treatment of herniated l umbar i ntervertebral disc
with radicul opathy. An outcome study. Spi ne 1 4: 43 1 , 1 989
2. Liemohn W: Exercise and the back. Rheum Dis Cl i n North Am 1 6:945 , 1 990
3. Robison R: The new back school prescription: stabilization training part 1 . Occup
Med 7: 1 7, 1 992
4. Farfan H: Muscular mechanism of t he l umbar spine and t he position of power and
efficiency. Orthop Cl i n Nort h Am 6: 1 35 , 1 975
5. Kumar S: Physiological responses to weight lifting i n different planes. Ergonomics
23: 987, 1 980
6. Pope MH, Anderson GBJ, Broman et al : El ectromyographic studies of l umbar trunk
musculature during the development of axial torques. J Orthop Res 4: 288, 1 986
7. Oddson L, Thorstensson A: Task specificity in the control of intrinsic trunk muscl es
in man. Acta Physiol Scand 1 39: 1 23 , 1 990
8. Kennedy B: An Australian program for management of back probl ems. Physiother­
apy 66: 1 08, 1 980
9. Kendall FP, McCreary EK: Muscl es. Testing and Function. 3rd Ed. Wi l l iams &
Wi l ki ns, Balti more, 1 983
10. Gracovetsky S, Farfan H, Hel leur C: The abdominal mechanism. Spine 1 0: 3 1 7 , 1 985
I I . Mil ler MI , Medeiros J M: Recruitment of internal oblique and transversus abdomi ni s
muscles during the eccentric phase of the curl -up exercise. Phys Ther 67: 1 2 1 3 , 1 987
1 2. Tesh KM, Dunn JS, Evans J H: The abdominal muscles and vertebral stabi l i ty.
Spi ne 1 2: 50 1 , 1 987
1 3 . Zetterberg C, Anderson GBJ, Schultz AB: The activity of i ndi vi dual trunk muscles
during heavy physical loading. Spi ne 1 2: 1 035, 1 987
1 4. Richardson C, Toppenberg R, Jull G: An initial eval uation of eight abdominal exer­
cises for t hei r abi l i ty t o provide stabili sation for the l umbar spine. Aust J Physiother
36: 6, 1 990
1 5 . Macintosh JE, Bogduk N, Gracovetsky S: The biomechanics of the t horacolumbar
fascia. Clin Biomech 2: 78, 1 987
16. Macintosh J E, Bogduk N: The detailed biomechanics of t he l umbar multifi dus. Cl i n
Biomech I : 1 96, 1 986
1 7. Bogduk N, Twomey LT: Cl inical Anatomy of the Lumbar Spi ne. 2nd Ed. Churchi l l
Livingstone, Mel bourne, 1 99 1
272 Physical Therapy of the Low Back
1 8. Bei mborn OS, Morrissey MC: A review of the l iterature related to trunk muscle
performance. Spine 1 3: 655, 1 988
1 9. Janda V: Muscl es, motor regulation and back probl ems. p. 27. In Korr 1 M ( ed) :
The Neurologic Mechani sms i n Mani pul ati ve Therapy. Plenum, New York, 1 978
20. Janda V, Schmi d HJA: Muscl es as a pathogenic factor i n back pai n. p. I . I n Proceed­
ings 4th I nternational Conference of the I nterational Federation of Orthopaedic
Manipul ati ve Therapists. Chri stchurch, New Zealand , 1 980
2 1 . Sahrmann SA: Muscl e imbalances i n the orthopaedic and neurological pati ent. p.
836. I n Proceedi ngs 1 0t h I nternational Congress of the World Confederation for
Physical Therapy. Sydney, Austral i a, 1 987
22. Caix M, Outrequin G, Descottes B et al: The muscl es of the abdominal wal l : a new
functional approach with anatomical deducti ons. Anat Cl i n 6: 1 0 1 , 1 984
23. Nouwen A, Van Akkerveeken PF, Versloot J M: Patterns of muscular acti vity during
movement i n pati ents with chronic low back pai n. Spine 1 2 : 777, 1 987
24. Parnianpour M, Nordi n M, Kahanovitz N, Frankel V: The triaxial coupl i ng of torque
generation of trunk muscl es during isometric exerti ons and the effect of fatiguing
isoinertial movements on the motor output and movement patterns. Spine 1 3 :982,
1 988
25 . Jul l G, Richardson C, Toppenberg R et al: Towards a measurement of active muscle
control for l umbar stabi l i sati on. Aust J Physiother 39: 1 87, 1 993
26. Jul l GA, Janda V: Muscl es and motor control in low back pai n: assessment and
management . p. 253 . In Twomey LT, Taylor JR (eds) : Physical Therapy of the Low
Back. Churchi l l Li vi ngstone, New York, 1 987
27. Sahrmann S: Postural appl i cati ons i n the chi l d and adul t . Neurodevelopmental as­
pects. Adul t posturi ng. p. 295. In Kraus S ( ed) : TMJ Disorders. Management of
t he Craniomandibular Compl ex. Churchi l l Li vingstone, New York, 1 988
28. Bul l ock-Saxton J : Changes in muscl e function at hip and low back fol l owi ng chronic
ankle sprai n. p. 1 470. ]n Proceedings 1 1 th I nternational Congress of the World
Confederation for Physical Therapy. London, 1 99 1
29. Richardson CA: The need for more precise methods of testi ng and exercising stabil­
i ty muscle functi on. ( Submitted for publ ication)
30. Kuno M: An hypothesis for neural control of the speed of muscle contraction i n
the mammal . Adv Biophys 1 7: 69, 1 984
3 1 . Fischbach GO, Robbi ns N: Changes i n contracti l e properties of disused soleus mus­
cl e. J PhysioI 201 : 305 , 1 969
32. Edstrom L: Sel ecti ve atrophy of red muscl e fibres in the quadriceps in long standing
knee joint dysfuncti on: i njuries to the anterior cruciate ligament. J Neurol Sci I I :
55 1 , 1 970
33. I anuzzo CD: The cel l ular composi tion of human skel etal muscl e. p. 3 1 . In Knuttgren
HG ( ed) : Neuromuscular Mechani sms for Therapeutic and Conditioning Exercise.
Uni versity Park Press, Bal t imore, 1 976
34. Oganov V, Skuratova S, Potapov N, Shi rvi ndkaye M: Physiological mechani sms
of adaptation of skel etal muscl es of mammals to the weightl ess state. p. 1 7 . In Cuba
F, Maredal G, Takacs 0 ( eds) : Advances in Physi ological Science. Pergamon Press,
Hungary, 1 980
35. Zetterberg C, Aniansson A, Grimby G: Morphology of t he paravertebral muscles
i n adolescent idiopathic scol i osi s. Spi ne 8: 457, 1 983
36. Whi te MJ, Davies CTM: The effects of i mmobi l i sation, after lower leg fractures,
on t he contracti l e properti es of human triceps surae. Cl i n Sci 66: 277, 1 984
37. Fi tts RH, Bri mmer CJ : Recovery in skel etal muscl e contracti l e function after pro­
longed hindl i mb i mmobil ization. J Appl Physiol 59: 9 1 6, 1 985
Rehabilitation of Active Stabilization of the Lumbar Spine 273
38. Richardson CA: Atrophy of vastus medial is in patel l ofemoral pain syndrome. p.
400. I n Proceedings 1 0t h I nternational Congress of the World Confederation for
Physical Therapy. Sydney, Australia, 1 987
39. Templeton GH, Sweeney HL, Hi mson BF et al: Changes in fibre composition of
soleus muscle during hi nd limb suspension. J Appl Physiol 65 : 1 1 9 1 , 1 988
40. Appell HJ: Muscul ar atrophy fol l owing i mmobil isation: a revi ew. Sports Med 1 0:
42, 1 990
4 1 . Lacote M, Cheval ier AM, Mi randa A et al : Cl i nical Eval uation of Muscl e Function.
Churchi l l Livingstone, London, 1 987
42. Richardson CA, lul l GA, Toppenberg R, Comerford M: Techniques for active l um­
bar stabilisation for spi nal protection: a pil ot study. Aust 1 Physiother 38: 1 05 , 1 992
43. Ekholm 1 , Ekl und G, Skoglund S: On the retlex effects from the knee joint of the
cat . Acta Physiol Scand 50: 1 67, 1 960
44. Froning EC, Frohman B: Motion of t he l umbosacral spine after l ami nectomy and
spine fusion. Correlation of motion with the resul t . 1 Bone Joi nt Surg 50A: 897, 1 968
45 . Pope MH, Bevins T, Wil der DG, Frymoyer W: The rel ationship between ant hropo­
metric, postural , muscul ar and mobi l i ty characteri stics of males aged 1 8-55. Spine
1 0:644, 1 985
46. Mel l i n G: Correlations of hip mobil i ty with degree of back pain and l umbar spine
mobi l ity in chronic l ow back pain patient s. Spine 1 3 : 668, 1 988
47. Mierau D, Cassidy JD, Yong- Hing K: Low-back pain and straight leg raising i n
chil dren and adolescents. Spi ne 1 4: 526, 1 989
48. Sward L, Erikssen B, Peterson L: Ant hropometric characteri sti cs, passi ve hip tlex­
ion and spinal mobil ity in relation to back pain i n athl etes. Spi ne 1 5 : 376, 1 990
49. Waddel l G, Somervi l l e D, Henderson I , Newton M: Objective cl i nical eval uation
of physical impairment i n chronic low back pai n. Spine 1 7: 61 7, 1 992
50. Hi des JA, Stokes MJ, Sa ide M et al: Asymmetry of l umbar mul tifidus muscl e size
measured by real-time ultrasound imaging i n patients with acute back pai n. Spine
(i n press)
5 1 . Stokes MJ, Hides JA, Jul l GA, Cooper DH: Mechanism of human paraspinal muscl e
wasting with acute l ow back pain, abstracted. J Physiol 452: 280p, 1 992
52. Sul l ivan PE, Markos PD: Cl i nical Procedures i n Therapeutic Exerci se. Reston Pub­
lishing, Reston, VA, 1 987
53. Richardson CA, Bul l ock MI : Changes in muscle activity during fast, al ternating
texion-extension movements of the knee. Scand J Rehabil Med 1 8: 5 1 , 1 986
54. Smith 10, Betts B, Ederton VR, Zernicke RF: Rapid ankle extension during paw
shakes: sel ective recruitment of fast ankl e extensors. 1 Neurophysiol 43: 61 2 , 1 980
55. Ng G, Richardson CA: The effects of training triceps surae using progressive speed
loading. Physiother Theory Pract 6: 77, 1 990
56. Richardson CA, Sims K: An i nner range hol ding contracti on. An objective measure
of stabilising function of an antigravity muscl e. p. 829. I n Proceedings I I th I nterna­
tional Congress of the World Confederation for Physical Therapy, London, 1 99 1
57. Thortensson A, Oddsson L, Carlson H : Motor control of vol untary trunk move­
ments in standing. Acta Physiol Scand 1 25 : 309, 1 985
58. Wohlfahrt DA, lull GA, Richardson CA: The rel ationship between the dynamic and
static function of the abdominal muscles. Aust J Physiother 39: 9, 1 993
10 Intensive Physical
Rehabilitation for
Back Pain
Lance T. Twomey
James R. Taylor
The current epidemic of low back pain and spinal dysfunction in Western
society, with its consequent huge costs to the individual and to the communities
concerned, has demanded increasing attention in recent years. It is suggested
that contemporary health science is currently unable to contain the ever-in­
creasing costs of treatment, I and is failing to provide definitive answers to the
multifaceted problem. However, closer analysis and investigation reveals that
in some areas there have been significant recent advances, particularly in our
ability to rehabilitate those with chronic back pain.2
Few people escape back problems and associated pain during their lives
and all vertebral columns show changes with age that make them potentially
less able to cope with the variety of physical stresses of daily life. It is now
generally considered that 80 percent of adults in Western industrialized societies
suffer low back pain during their lifetime. In terms of work loss and treatment
costs, back pain is the single most expensive musculoskeletal ailment in West­
ern society. 104
At any time, about 40 percent of any large gathering will indicate
that they currently have low back pain, while as many as 60 percent of the
population will have experienced some degree of low back pain in the past
For many years it has been assumed that low back pain is an epidemic
only in industrialized societies, but is not a significant problem in less "devel­
oped" countries. However, recent research indicates that when back pain clin­
ics are opened in less industrialized societies, people flock there for treatment.
276 Physical Therapy of the Low Back
It is unlikely that the prevalence of back pain has significantly altered in recent
years, but the cost to the community is now much better understood.
In the absence of a complete knowledge of the pathogenesis of back pain,
and because of the inadequacy of some diagnostic procedures, many of the
diagnostic labels attached to patients are uncertain and treatment is often empir­
ical.2.4 However, recent biologic and epidemiologic studies continue to fill in
the gaps in our knowledge of normal spinal structure and function, and of the
patterns of age changes and related pathology in the spine.
Low back pain and
low-back dysfunction or disability need to be carefully distinguished. 1 While
both may reflect pathology and structural change, soft-tissue or biochemical
pathology are not readily diagnosed by current investigative methods and the
assessment of pain and dysfunction still rests primarily on the individual's sub­
jective history. This is infuenced by the patient's beliefs and attitudes,8 and
the subjective report of the severity of back pain may reflect a host of other
influences. These influences include concepts as diverse as a person's learned
response to pain, their interpretation of the meaning of their back pain, and
avoidance behavior associated with unpleasant aspects of the patient's life­
style or occupation.
This chapter does not consider the treatment of acute low back pain, since
this is covered elsewhere in this volume and because there are so many treat­
ment methodologies used by orthodox and alternative medicine that appear to
be effective. The reader is referred to Grieve9 and to Corrigan and Maitland
for excellent descriptions of many of the methods used. This chapter does
consider the treatment of chronic low back pain, since it is this disorder that
is most prevalent in all Western societies and has been the focus of considerable
research effort with some important progress made during the last decade.
In recent years, it has become abundantly clear that physical activity is
beneficial and necessary to patients with back pain; active rehabilitation not
only restores function, but is also strongly associated with a reduction in
pain. 1
1 1
13 There is no evidence that prolonged rest, or the avoidance of
exercise/activity brings about a reduction in chronic back pain; indeed the evi­
dence supports the view that except for a short period after injury, rest has no
efect on the natural history of back pain. A number of studies suggest that
prolonged inactivity accentuates the problem and may increase the severity of
the pain. 13. 1
In spite of this evidence, bed rest, analgesics, the prescription of
corsets, and the avoidance of physical activity are still the most commonly
prescribed forms of medical treatment.I. 1
There is no doubt that prolonged bed rest and inactivity brings about pro­
found deleterious changes to the musculoskeletal system.1
. 1
There is a decline
in physical fitness, a marked reduction in muscle strength, a reduction in joint
range and flexibility, and a decline in bone mass in those individuals whose
activity levels are severely curtailed. 1
6. 18
19 All parts of the musculoskeletal
Intensive Physical Rehabilitation for Back Pain 277
system demand constant use throughout life and even into extreme old age to
maintain their strength and efficiency. In the spine, the health of joints is largely
dependent on repeated low stress movements. The intervertebral discs and the
articular cartilage of the facet joints are dependent on the "stirring effect" of
movement for the maintenance of adequate fluid transfer and nutrition in their
avascular cartilage.
A habitual reduction in activity levels is inevitably asso­
ciated with a decline in a person's ability to react appropriately to changes in
the environment. A return to physical work, after a period of bed rest for
chronic back pain, exposes an individual to risk of further back injury, since
the individual's state of musculoskeletal fitness is much lower than it was when
the back pain initially caused the person to stop work. Such individuals are
weaker, less mobile, and fatigue more rapidly than before they stopped work
and it is little wonder they often suffer an early recurrence of their back pain
problem on return to work.3. 13 Much of the lost productivity in industry, result­
ing from work absence due to back pain, relates to the physical changes directly
resulting from bed rest and inactivity. 1
In rece�t years, an improved understanding of the relationship between
back pain and activity and of the deleterious effects of prolonged disuse, has
resulted in a new approach to treatment in many centers.I•3. 13 This approach
concentrates initially on the need to gain a comprehensive "baseline" measure
of back function and fitness and pain levels. Patients are then ftted into a
program of intensive physical therapy appropriate to their physical status. This
will often include a behavior modification program, aimed at rapid functional
restoration and a return to work as soon as possible. The success of such
programs can be measured by marked improvements in physical capacity and
function and particularly by the ability to return to the workplace. Improvement
during the course of the program should not be judged solely by the patients'
sUbjective self-reports on their pain levels, since this is often modified substan­
tially by legal, psychological, financial, and social factors.21 It is important to
note that at the conclusion of such programs of intensive physical reconditioning
a significant reduction in subjective pain measures are consistently re­
ported. 1
21-23 As a general rule, people who are physically fit recover faster
from any musculoskeletal ailment.22.24.2
Thus McQuade et aJ25 have shown
that the stronger the individual with chronic back pain, the less that person
appears to be limited by the condition, and also that the higher the aerobic
work capacity, the more active and flexible is that individual. They showed
that reports of back pain intensity were not increased by exercise; indeed many
subjects reported feeling better following vigorous activity. The recent ad­
vances in the intensive physical rehabilitation of patients of all ages with chronic
back pain have developed principally from the treatment of young athletes with
sports injuries.3.24 Jt has been known for some years that athletes with severe
278 Physical Therapy of the Low Back
musculoskeletal damage (including back injury) respond very well to intensive
programs of physical treatment. 2
Measurements of spinal performance have proved difficult to devise and
slow to gain acceptance because of the complexity of the vertebral column,
which consists of a multitude of small joints, covered by large fleshy muscles
with overlapping attachments, allowing complex multiplanar movementsY
However, it is now possible to measure many aspects of the movement behavior
and mechanical capacity of the vertebral column by noninvasive technologies
that have been verified in laboratory and clinical trials.3
.. �.28
Measures of Physical Function
The following techniques are used to measure and assess physical function.
1. Ranges of lumbar movements, using instruments such as the lumbar
spondylometer and rotameter29
2. Isokinetic trunk strength, using isokinetic dynamometers such as
Cybex, Kin Com, and Isostation B20023
3. Measures of muscle endurance for trunk extensor and flexor
3 1
4. Measures of cardiovascular fitness and efficiency using standardized
fitness tests32
5. Static and dynamic lifting using fixed loads or dynamometry3
6. Measures of functional capacity: these are usually especially con­
structed for each subject and reflect the working conditions to which it is ex­
pected they will return2
7. Dynamic obstacle course: a timed test stimulating activities of daily
living, and requiring the person to complete the series of tasks in many different
Measures of Behavioral Function
The behavioral self-report measures available are many and different clin­
ics use many combinations of them. They include such tests as the Oswestry
low-back pain/disability questionnaire; visual pain analog scale; quantitative
pain drawing; Minnesota Multiphasic Personality Inventory; and Middlesex
Hospital Questionnaire.
Patients are measured prior to and at the conclusion of the intensive physi­
cal treatment programs and at regular intervals during the program. In addition,
every aspect of their physical work activity is carefully monitored to assess
Intensive Physical Rehabilitation for Back Pain 279
ongoing performance levels. Regular follow-up assessment is done at 6 weeks,
and 3-, 9-, and 12-month intervals after the conclusion of the programs.
After the initial measurement of full medical diagnosis and assessment,
patients begin a program of intensive physical therapy and work conditioning
for 8 to 10 hours a day over 3 or 4 weeks.3.'3 Each program is carefully tailored
to meet the particular requirements of the entering individuals, and is closely
monitored by appropriately trained staff. The patient's working day consists of
a well-constructed mix of weight training, aerobic and fitness work, functional
activity, relaxation and stress training, work hardening, educational instruction,
and where appropriate, behavior modification. All programs emphasize active
participation, encourage the individuals to continue to persist at their tasks
irrespective of their back pain levels, and demand a progressive increase in
work output as the program proceeds. Cardiovascular fitness is always included
as a central element, since it plays a most important role in back injury and in
There is always an accompanying back education component that teaches
the patients about the structure, function, and pathology of the vertebral col­
umn. This segment needs to allow considerable opportunity for dialogue be­
tween educator and patients and seeks to ensure that patients develop a proper
understanding of their particular back problem and its management. Thus the
educational component should include both group and individual counseling.
This is usually done by a clinical psychologist who may be using behavior
modification, relaxation, and pain management techniques.
The whole thrust of these programs is directed toward preparing each indi­
vidual for retur to work. Thus specific programs of work hardening need to
be constructed and implemented. The aim is to return the individual to the
previous occupation wherever possible. At times, either the workplace requires
modification or the patient requires retraining for a modified or different occupa­
tion. In these circumstances, the physical and occupational therapists should
work in close harmony with the rehabilitation counselor and representatives
of the patient's workplace.
Patients from occupations requiring heavy and/or repetitive lifting proce­
dures are not precluded from such programs. All programs include substantial
lift-training components. Recent research clearly indicates that most individuals
can be trained to manage heavy, repetitive lifting tasks, using well-proven and
long-known weight training techniques.34 Thus the initial assessment procedure
determines the type and duration of lifting activity that any individual will need
to perform at work and trains that person specifically for that task. If, for
example, the job requires the lifing of 10-kg legs of ham from a freezer and
placing them on a shelf, then the individual is trained to a level so that they
may adequately pelfrm that particular task. The physical therapist ensures
that the training initially uses low loads and lower repetitions. Progressively, the
280 Physical Therapy of the Low Back
size of the loads and the numbers of repetitions are increased until a satisfactory
standard is reached. These methods will generally ensure that the patient's
strength and endurance capacities are progressively and rapidly increased.
Other work activities can be measured and appropriate training is provided
in the same way as for lifting. The whole process of fitness for a specific occupa­
tion forms a most important part of the total rehabilitation process and is the goal
to which the physical program is directed. This requires a close investigation of
the worksite by the rehabilitation team, a proper assessment of the physicaV
mental requirements of the task, and the subsequent development of a training
program aimed at developing those skills in the injured worker. It must involve
dialogue with management, and often leads to changes in the workplace so as
to provide an ergonomically safer and more efficient worksite. Thus the heights
of benches, tables, and chairs may need modification and often the ergonomist/
therapist is able to provide information on alternative techniques or facilities
that might be used to help the worker perform the tasks more appropriately
and safely. Where repetition of physical activity is a central part of the occupa­
tion, the employees and employers are educated as to the length and nature of
pauses that are necessary throughout the working day to avoid fatigue and
prevent musculoskeletal problems. It is important that pause activities and exer­
cises be taught to the employee as an essential part of a job. The close coopera­
tion of the employer is central to the success of this approach. Just as much
as the employees need to be fit for the tasks that the job requires, employers
need to understand the physical difficulties of particular tasks and appreciate
the need for some alternative activity at appropriate times throughout the work­
ing day.
Similarly, a strong association has been shown between poor physical work
skills, low levels of strength and endurance, and the incidence of occupational
back pain and dysfunction. 13.35 By increasing the participant's strength, endur­
ance, and cardiovascular fitness, together with improving the specific handling
skills necessary for a particular occupation, there is usually a concomitant re­
duction in the level of back pain reported. 13.32 This provides a cogent argument
for ensuring that physical rehabilitation programs are especially tailored to suit
the work skills and requirements of particular individuals. While there are many
physical exercises and activities that can be done in common within a larger
group, there is no doubt that specific work hardening is an essential part of
any program. 13
In Mayer's classic study3.3
of objective assessment and intensive physical
treatment at the PRIDE establishment in Dallas, Texas, 84 percent of his study
group (n = 62) returned to work and continued at work for 12 months after
the conclusion of the program. This compared very favorably with a return to
work of 55 percent for his control group, and only 20 percent for those who
dropped out of his initial cohort. Mayer's control group consisted of 38 patients,
Intensive Physical Rehabilitation for Back Pain 28 1
each with a similar history of chronic back pain, who were eligible for entry
into the program, but where there was prolonged delay or refusal by their
insurance companies to authorize participation. Continuing data for the PRIDE
clinic show that the initial population is continuing in employment for some
years after the completion of their initial treatment/education program.
However, the figures indicating an 84 percent return to work are higher
than those obtained at most other clinics. In West Australia, data from a clinic23
with a similar philosophy demonstrate a 60 percent return to work, with that
group still retained in employment at the end of 12 months.
A recent retrospective cohort study by Saal and SaaP3 considered the
functional outcome of an aggressive physical rehabilitation program in 64 pa­
tients with herniated lumbar discs. The underlying premise of the treatment
was that patients should be involved in active rather than passive therapy. The
study showed that 90 percent of the patients had a good or an excellent outcome
based on measurement criteria and there was 92 percent return to work. Four
of the six patients who required surgery were found to have spinal stenosis.
The study demonstrated that patients with herniated lumbar discs can be suc­
cessfully treated by intensive physical therapy and that surgery should be re­
served for those patients in whom function has not been improved by aggressive
physical rehabilitationY
A review of the recent current literature reveals that patients with chronic
low back pain react better to active rather than to passive treatment regimens.
Indeed, the existing information demonstrates that rest, which is currently the
most popularly prescribed treatment, often exacerbates rather than improves
the condition of those with low back pain. Deyo et aJl5 have shown that 2 days
of bed rest after an episode of acute back pain is sufficient, and that further
bed rest causes a significant decline in a patient's functional capacity. They
believe that if a policy of no more than 2 days of bed rest was universally
applied that there would be significant benefits for a community. These would
be realized in terms of a reduction in the indirect costs associated with low back
pain for patients and employees and that absenteeism would be substantially
I. Waddell G: Clinical assessment of lumbar impairment. Clin Orthop Relat Res 221:
110, 1987
2. Nachemson A: Work for all: for those with low back pain as well. Clin Orthop Relat
Res 179:77, 1983
3. Mayer TG, Gatchel RJ, Kishino N et al: Objective assessment of spine function
following industrial injury. Spine 10:482, 1985
282 Physical Therapy of the Low Back
4. Nachemson AL, La Rocca H: Editorial: Spine. Spine 12:427, 1987
5. Twomey L T, Taylor JR: Physical Therapy of the Low Back. Churchill Livingstone,
New York, 1987
6. Andersson GJ: The biomechanics of the posterior elements of the lumbar spine.
Spine 8:326, 1983
7. Frymoyer JW, Cats-Baril WC: An overview of the incidences and cost of low back
pain. Orthop Clin North Am 22:263, 1991
8. Waddell G, Main CJ, Morris EW et al: Chronic low back pain. psychological distress
and illness behavior. Spine 9: 209, 1984
9. Grieve G: Modern Manual Therapy. Churchill Livingstone, Edinburgh, 1986
10. Corrigan B, Maitland GD: Practical Orthopaedic Medicine. Butterworths, London,
II. Smith S, Mayer TG, Gatchel RJ, Becker TJ: Quantification of lumbar function. Part
I: isometric and mUltispeed isokinetic trunk strength measures in sagittal and axial
planes in normal subjects. Spine 10:757, 1985
12. Mayer TG, Barnes D, Kishino ND et al: Progressive isoinertial lifting evaluation
I. A standardised protocol and normative database. Spine 13: 993, 1988
13. Ganora A: Rehabilitation of work related back injury. Aust Family Phys 15: 430,
14. Dcyo RA, Tsui-Wu YR: Descriptive epidemiology of low back pain and its related
medical care in the United States. Spine 12:264, 1987
15. Deyo RH, Diehl AK, Rosenthal M: How many days of bed rest for acute low back
pain? New Engl J Med 315: 1064, 1986
16. Twomey L T: Physical activity and ageing bones. Patient Management 27:34, 1989
17. Lowther D: The effect of compression and tension on the behavior of connective
tissues. In Glasgow EF, Twomey L T, Scull ER et al (eds): Aspects of Manipulative
Therapy. 2nd Ed. Churchill Livingstone, Melbourne, 1985
18. Pardini A: Exercise, vitality and ageing. Ageing 344: 19. 1984
19. Shephard RJ: Management of exercise in the elderly. Appl Sports Sci 9: 109, 1984
20. Frank C, Akeson WH, Woo SL-Y et al: Physiology and therapeutic value of passive
joint motion. Clin Orthop Relat Res 185:113, 1984
21. Gatchel RJ, Mayer TJ, Capra P et al: Quantification of lumbar function, part 6: the
use of psychological measures in guiding physical function restoration. Spine II:
36, 1986
22. Deyo RA, Bass JE: Lifestyle and low back pain: the infuence of smoking and
obesity. Spine 14:501, 1989
23. Edwards BC, Zusman M, Hardcastle D et al: A physical approach to the rehabilita­
tion of patients disabled by chronic low back pain. Med J Aust 156: 167, 1992
24. Cinque C: Back pain prescription: out of bed and into the gym. Phys Sports Med
17: 185, 1989
25. McQuade KJ, Turner JA, Buchner DM: Physical fitness and chronic low back pain.
Clin Orthop Relat Res 233: 198, 1988
26. Saal JA: Rehabilitation of football players with lumbar spine injury. Phys Sports
Med 16:61, 16:117, 1988
27. Bogduk N, Twomey L T: Clinical Anatomy of the Lumbar Spine. Churchill Living­
stone, Melbourne, 1987
28. Nordin M, Kahanovitz N, Verderane R et al: Normal muscle strength and endurance
in women and the effect of exercises and electrical stimulation, Part I: normal
endurance and trunk muscle strength in 101 women. Spine 12: 105, 1987
29. Taylor JR, Twomey L T: Age related change in the range of movement of the lumbar
spine. J Anat 133: 473, 1981
Intensive Physical Rehabilitation for Back Pain 283
30. Lindgren S, Twomey L T: Spinal mobility and trunk muscle strength in elite hockey
players. Aust J Phys 34: 123, 1988
31. Cady LD, Thomas PC, Korwasky RJ et al: Programs for increasing health and
physical fitness in firefighters. J Occup Med 2: Ill, 1985
32. Cairns D, Mooney V, Crane P et al: Spinal pain rehabilitation: inpatient and outpa­
tient treatment results and development of predictors for outcome. Spine 9:91, 1984
33. Saal JA, Saal JS: Non operative treatment of herniated lumbar intervertebral disc
with radiculopathy: an outcome study. Spine 14:431, 1989
34. Sullivan S: Back support mechanisms during manual lifting. Phys Ther 69:38, 1989
35. Vide man T, Malmivaara A, Mooney V et al: The value of the axial view in assessing
discograms: an experimental study with cadavers. Spine 12: 299, 1987
36. Mayer TG, Gatchel RJ, Kishino N et al: A prospective short term study of chronic
low back pain patients utilising novel objective functional measurement. Pain 25:
53, 1986
11 Lumbar Spinal
Nils Schonstrom
The concept of spinal stenosis was first introduced almost four decades
ago by Verbiest,'·2 who described a radiculopathy caused by a narrowing of
the lumbar spinal canal. There had been sporadic reports earlier in the literature
of a peculiar disease with radiculopathy that was cured by laminectomy and
where no obvious pathology was found during surgery.3-6 With Verbiest's re­
ports, however, a possible explanation for these early results was given, an
appreciation of the importance of the narrow lumbar spinal canal slowly devel­
oped, and the concept of developmental narrowing of the spinal canal was
accepted. A true congenital form, with a stenosis present at birth, had been
described earlier by Sarpyener in 1945.7 The term developmental, chosen by
Verbiest, was intended to indicate that symptoms did not develop until the
patient reached maturity. This form of stenosis was characterized by a short
anteroposterior diameter between the posterior part of the vertebral body and
the vertebral arch.
Later Epstein and co-workers8
9 described narrowing
around the spinal nerve as it was leaving the dural sac and on its way down
the root canal under the pedicle and out through the intervertebral foramen.
This form was called lateral stenosis and the original form of stenosis around
the dural sac and the cauda equina was called central stenosis.
The work of Kirkaldy-Willis and co-workers 10 has shown that another com­
mon type of stenosis occurs as a consequence of degenerative processes in the
three-joint complex (Fig. 11-1), that is, in the disc and the two zygapophyseal
joints holding two adjacent vertebrae together. This type of stenosis has a differ­
ent etiology from the type initially described by Verbiest, and it is now common
286 Physical Therapy of the Low Back
Fig. 11-1. The three-joint complex includes the disc, which is the joint between the
vertebral bodies, and the two zygapophyseal joints between the posterior structures.
to distinguish between developmental and degenerative stenosis. Two reports
described a prevalence ratio of about 10 to I, comparing degenerative and
developmental stenosis.
1 .1
2 A broad definition of stenosis was given in 1976
by a group of authors where the stenosis was defined as "any type of narrowing
of the spinal canal, nerve root canals or intervertebral foramina." 13
From an etiologic point of view, in current practice we can generally distin­
guish at least seven different forms of lumbar spinal stenosis:
I. Congenital, which is present at birth7
2. Developmental, with a genetic disposition presenting itself with symp­
toms in adult life,
1 .
14 the most pronounced being a consequence of achondro­
3. Degenerative, as a result of degenerative processes in the disc and facet
4. Metabolic, as a consequence of pathologic changes in the bone sub­
stance (e.g. , Paget's disease of the bone,
9 fuorosis, and diffuse idiopathic
skeletal hyperostosis?O
Iatrogenic, usually postlaminectomy
6. Post-traumatic, with a distorted anatomy of the spinal canal as a conse­
quence of trauma
7. Miscellaneous, such as epidural lipomatosis2
and cysts of the ligamen­
tum favum
From a morphologic point of view, lumbar spinal stenosis can be divided
into central stenosis engaging the nerve roots of the cauda equina, and lateral
stenosis disturbing the spinal nerve in the lateral recess, the nerve-root canal,
or intervertebral foramina. Andersson and McNeiH23 clarified the "lateral anat­
omy," by dividing this region into three different zones related to the pedicle,
C N ,.
c c c
Lumbar Spinal Stenosis 287
Fig. 11-2. The root canal containing the spinal nerve can be divided into three different
zones in relation to the pedicle. (Adapted from Andersson and McNeill,23 with per­
mission. )
which is an important surgical landmark (Fig. 11-2). Zone I is the lateral recess;
this is the area under the superior articular process medial to the pedicle. Zone
2 is below the pedicle and zone 3 is lateral to the pedicle. It is open to question
if a mechanical disturbance of the spinal nerve in zone 3 should be categorized
as stenosis.
The developmental form of stenosis is characterized by a short pedicle and
thick lamina, usually at mUltiple levels. Its most pronounced form is seen in
achondroplastic dwarfs, where it is a common finding. The stenosis in achon­
droplasia is also different from other forms in that there is also a short interpe­
dicular distance. 15 In the developmental form of stenosis, the emphasis is on
an encroachment of the nervous structures by the skeletal structures. The steno­
sis is of a central type affecting the roots of the cauda equina and possibly
the spinal nerve in the lateral recess (zone I). According to Verbiest,
4 an
anteroposterior diameter below 12 mm but over 10 mm is a relative stenosis
where other problems, such as a herniated disc, will lead to symptoms. If the
diameter is less than 10 mm there is an absolute stenosis capable of causing
288 Physical Therapy of the Low Back
symptoms without the addition of any other factors. Measurement studies of
skeletal collections show that these small dimensions are rare, indicating that
developmental stenosis might be a rare disease.
5 A diameter of 13 mm or
less was present in only 6.3 percent of Eisenstein's 433 skeletons.
In the degenerative type of stenosis, the skeletal dimensions of each verte­
bra might well be within normal limits, but changes in the disc, ligamentum
flavum, and zygapophyseal joints lead to narrowing of the spinal canal.
a morphologic point of view, these changes can lead to both a central and a
lateral stenosis as defined previously. These two types of stenosis will be dis­
cussed separately, but it is very common for central and lateral stenosis to
occur together in the same individual.
The central stenosis in this group has three different components, ofen
resulting in a typical trefoil shape of the spinal canal that can be present in the
skeletal structure as well as being the result of a deformation of the dural sac
by the surrounding soft tissues. A bulging disc, or sometimes a herniated disc,
can cause an impression in the dural sac on its anterior aspect. From the pos­
terolateral side, on both sides of the midline, enlarged zygapophyseal joints
covered by a thick ligamentum flavum cause concave impressions (Fig. 11-3).
Another common factor contributing to stenosis is an anterior or posterior slip­
ping of one vertebra on the next. The pars interarticularis is often intact and
the phenomenon is referred to as pseudolisthesis or degenerative olisthesisl
8 The slip can, in itself, be sufficient to cause stenosis, or it can
signifcantly contribute to the deformation of the dural sac described previously
as a result of sof-tissue encroachment.
Fig. 11-3. The typical deformation of the dural sac at disc height is in the form of a
Lumbar Spinal Stenosis 289
Fig. IJ-4. Arthritic changes in one zygapophyseal joint can lead to impingement on
two different spinal nerves: the upper nerve in zone 2 and the lower nerve in zone I.
In the lateral type of stenosis, degenerative enlargement of a zygapophyseal
joint can lead to encroachment on two consecutive spinal nerves.26 It can either
impinge on the nerve about it as it leaves the dural sac and passes out through
the intervertebral foramen (in zone 2 in relation to the pedicle above), or on
the next spinal nerve as it lies in the lateral recess (zone I of the next pedicle)
(Fig. 11-4).
The metabolic type of stenosis is seen as a result of postmature growth of
bone resulting in a central stenosis of much the same type as developmental
The iatrogenic type of stenosis following surgery involves three different
mechanisms: ( I) new bone formation from raw bone surfaces after laminectomy
or posterior fusion; (2) postlaminectomy membrane proliferation or cyst forma­
tion; and (3) instability caused by laminectomy with rapid degeneration and
accentuated slipping between adjacent vertebrae (pseudolisthesis).
Dynamic Concept in Degenerative Stenosis
The short description in the previous section gave a static picture of the
basic changes leading to a degenerative stenosis. However, to understand the
pathophysiologic description that follows, it is helpful to adopt a more dynamic
concept of degenerative stenosis.
The three-joint complex formed by the disc and zygapophyseal joints is
290 Physical Therapy of the Low Back
designed to permit movements between the vertebrae. Normally this is done
in such a way that the space in the spinal canal and the nerve-root canals
remains sufficiently large to accommodate the nervous structures inside it. We
have shown in in vitro experiments that when a lumbar spine specimen was
moved from full flexion to full extension, the transverse sectional area of the
spinal canal was diminished by an average of 40 mm2Y The same magnitude
of change in the transverse area of the spinal canal was found after a shift in
axial loading from 200 N of axial distraction to 200 N of axial compression.
During the axial loading the spine was held in a neutral position. Since this is
approximately equivalent to an increase in axial load of about 40 kg, the weight
of a full-grown torso, it might represent the decrease in size of the spinal canal
when a subject moves from lying down to standing up. These changes are well
tolerated with a normal canal but it can have significant implications in a spine
with a narrow canal.
It has been claimed that true hypertrophy of the ligamentum flavum is
extremely rare.28 This has led to the assumption that this ligament does not
contribute to a central stenosis. However, several authors29
3o have claimed
that it has been their impression during surgery that it does play a role in the
compression of the dural sac. In a measurement study, we have shown that
the fully relaxed thickness of the ligamentum flavum is on average 2 mm greater
than the ligament when distracted under 8 kg of 10adY This is explained by
the elastic behavior of the ligamentum flavum, which can contain as much as
80 percent of elastic fbers. 32 Thus, when the normal ligament is fully relaxed
in extension, the resulting thickening can contribute significantly to a further
constriction of an already narrow canal, even without hypertrophy of the lig­
The spinal nerve in the lateral recess (zone I) may or may not be affected
by changes of posture and load. An increased backward bulging of the disc can
increase pressure on the spinal nerve in the proximal part of zone I , which is
at disc height, but a few millimeters further distally, the vertebral body forms
the anterior limit of this space and here the dimensions are constant during
shifts in load or posture. Under the pedicle, in zone 2, however, great changes
occur when the spine is moved from fexion to extension. In extension the
superior articular facet moves upward, toward the pedicle where the spinal
nerve is situated in the uppermost of the foramen. Normally, the foramen,
which contains fat and vascular structures together with the nerve, is large
enough to accommodate these changes. When there is degenerative enlarge­
ment of the superior articular process, the difference between extension and
flexion might mean the difference between encroachment and no encroachment
on that spinal nerve. In addition, when an axial load is applied to the spine
over a period of time, creep in the disc results in a reduced disc height. As a
consequence, there is a migration upward of the superior facet toward the
pedicle above, with a risk of nerve impingement by an enlarged zygapophyseal
This dynamic view of the size of the spinal canal and nerve-root canals is
the key to understanding how changes in posture and load, which are problem
Lumbar Spinal Stenosis 291
free in a normal canal, can lead to symptoms when the canal is narrow. It also
provides one explanation for the fluctuating nature of the symptoms that are
so characteristic of lumbar spinal stenosis.
Size of the Canal and Cauda Equina
Various measurements have been used to describe the available space in
the lumbar spinal canal. The most frequently used measurement of the canal
has been its anteroposterior diameter. The interpedicular distance has been
proposed, but not generally accepted as useful, with the exception of stenosis
in achondroplasia. The transverse area of the spinal canal, as outlined by the
skeletal structures, has been used,33 and this area will be reduced in cases
with developmental stenosis. It is much less valuable in degenerative stenosis
because the skeletal measurements could be normal despite severe stenosis,
as described earlier. Based on a morphologic study in patients with central
stenosis confirmed during surgery, we have found the transverse area of the
dural sac to be the best measurement for confirming a central stenosis on trans­
verse sections of the lumbar spine (Fig. 11
What about the size of the neural elements to be accommodated in the
canal? In two in vitro experiments, we recorded this size expressed as the
transverse area of the dural sac and its contents.34,35 A carefully calibrated
circular clamp was placed around the dural sac at the L3 level. A thin pressure­
recording catheter was placed among the roots of the cauda equina inside the
clamp. The clamp was tightened until the first sign of a pressure increase among
the roots. The transverse area where this first pressure increase was noted was
called the critical size of the dural sac. The critical size was surprisingly con­
stant, among different individuals and using different measurement and experi-
D Experimental study, critical size
[ Stenotic patientsl min size
El Normals

��� �
� 0

i •
Transverse area of dural sac ...
Fig. 11-5. The transverse area of the dural sac in one clinical and two experimental
studies. (From Schonstrom,41 with permission.)
292 Physical Therapy of the Low Back
mental procedures, at an average of 75 mm2, with a standard deviation of 15
mm2 (see Fig. 11-5). To reach a further pressure increase of about 50 mmHg
among the roots, the clamp had to be tightened to reduce the area by another
19 percent below the critical size; to produce a pressure increase of 100 mmHg
the area had to be reduced by an average of 26 percent below the critical size.
We concluded that the size of the spinal canal varies considerably between
different individuals, but variation in the size of its neural content is small. We
therefore defined central stenosis in terms of the size of the dural sac on com­
puted tomography (CT) scans, based on our experimental data. No similar
experiments have been done for lateral stenosis, and one has to rely on other
methods, such as nerve-root blocks, to confirm the diagnosis.36
Efect of Pressure on Nerve Roots
The pathophysiologic response to acute mechanical pressure on the nerve
roots of the cauda equina has been investigated by Olmarker37 in porcine experi­
ments. A translucent pressure chamber was attached to the spine of the pig
after laminectomy under general anesthesia. An acute pressure was applied to
the nerve roots, using a balloon inside the chamber. The effects of a stepwise
increase in pressure on the microcirculation of the roots was studied through
the translucent chamber. The average occlusion pressure for the arterioles was
slightly below the systolic blood pressure. It was also concluded that the flow
in the capillary network was affected by venous stasis, which began to occur
at low pressures of only 5 to 10 mmHg. A retrograde stasis could be responsible
for disturbances of nerve function due to venous congestion. Observations of
gradual decompression after initial acute compression revealed that the pres­
sure levels had to go down to zero to obtain a full restitution of the blood flow.
The possibility of edema and blocking of axonal transport are other mechanisms
that may add to the disturbance of nerve function.
In an anatomic study on cadaver spines, Hoyland and co-workers38 found
a correlation between venous stasis in the intervertebral foramen and interstitial
fbrosis in the corresponding spinal nerve. It seems possible that a chronic stasis
can lead to edema as seen in the experimental study by Olmarker,37 and over
time this edema could be transformed into fibrosis. Thus the fibrosis seen by
Hoyland et aJ38 was interpreted as the end result of chronic nerve damage
associated with venous congestion.
Encroachment on Nerve Roots of the Cauda Equina
If we put the experimental information together, it is possible to create a
hypothetical model for the pathogenesis of central stenosis of the degenerative
type based on a dynamic concept.
The normal spinal canal has a reserve capacity over and above the space
required for the contents of the canal, a concept introduced by Weisz and Lee.39
Lumbar Spinal Stenosis 293
Fig. 11-6. At disc height the spinal canal is made up of two adjacent vertebrae.
The smaller the canal the closer one approaches the critical size beyond which
the nerve roots would be compressed. Canal narrowing occurs in most instances
at the disc level, where the posterior boundary of the canal is formed on each
side by a zygapophyseal joint (Fig. 11-6). The dural sac is usually deformed
into a trefoil shape (see Fig. 11-3). Considering frst the anterior aspect of the
canal, a bulging disc can easily make an impression of 2 mm as a result of
increased axial load (Reuber et aI. 40). The posterolateral impressions are made
by enlarged zygapophyseal joints, with a possible further encroachment of2 mm
by the thic'ening of the ligamentum f1avum that accompanies their relaxation in
extension. Figure 11-7 shows how it can be calculated that an impression from
the anterior aspect of 2 mm (disc) with a further 2 mm from the posterolateral
aspects (ligamentum f1avum) will reduce the area of the dural sac by about 40
percent, if the undeformed area is circular with lO-mm diameter. 4
Thus when
the canal has reached the small dimensions of the "critical size," the normal
changes in disc bulge and the thickness of the ligamentum f1avum can reduce
the remaining space considerably as a result of axial loading or extension of the
spine. This was confirmed in our experiment with spine specimens in various
postures and axial loads, described earlierY
Fig. 11-7. The trefoil deformation of the initially circular dural sac can be graded as a
percentage of the initial diameter, whereby the magnitude of the deformation can be
estimated. (From Schonstrom,41 with permission.)
294 Physical Therapy of the Low Back
If the dural sac has already reached the critical size in fexion, when the
canal size has its maximum, a further reduction of the available space by 40
percent as a result of extension could easily result in an acute pressure increase
of over 100 mmHg among the nerve roots of the cauda equina.35 The work of
Olmarker et al37 showed that this could lead to pronounced acute disturbances
of nerve function. The chronic effects on the nerve roots are more unpredictable
and a certain adaptation of the nervous structures to a narrow surrounding is
possible. In our experiments with acute constriction of the nerve roots, we saw
a creep downward with time of the initial pressure increase if the constriction
was left constant. 35 One interpretation of this phenomenon could be that it was
a result of a deformation over time of the nervous structures as a consequence
of the sustained compression. However, with the dynamic concept described
above, it is also justifable to discuss acute changes in size superimposed on
chronic deformation.
Clinical History
Spinal stenosis can be seen in all age groups, but its prevalence is greatest
in patients over 60 years of age. Many of the patients with lumbar spinal stenosis
have a long history of low back pain of a "mechanical" type, with sciatica and
evidence of disc disease during their thirties and forties (Kirkaldy-Willis
Symptoms are usually less pronounced in the fifties and they gradually become
typical of established spinal stenosis in the sixties.
Although back pain is a common symptom, the clinical picture is dominated
by various disturbances in the lower extremities. The principal complaint may
be pain, but is more often described in terms of numbness, paresthesia, and
weakness. Often the patient has difficulty in clearly expressing the quality of
the symptoms. The sometimes bizarre nature of the symptoms can result in
the patient being accused of malingering.
The well-known symptoms of neurogenic claudication are the unique fea­
ture of the disease and need special attention when taking the patient's history.
These symptoms are usually brought on by walking and relieved by rest. The
characteristic feature of neurogenic claudication is the influence of posture on
physical activity. In this respect, neurogenic claudication differs from periph­
eral vascular claudication. Extension aggravates the symptoms whereas flexion
facilitates the physical activity and diminishes the symptoms. Thus walking
bent forward increases the maximum walking distance. 4
Riding a bike is often
possible for considerable distances,43 and teaning forward or squatting while
resting relieves the discomfort. In one specific variant of this condition, the
symptoms arise, not so much after walking or physical exercise, but simply by
persistent extension of the back, even when standing still. This led Wilson44
to define two types of clinical picture in stenosis, one postural type and one
claudication type. He also stated that motor symptoms would ofen precede
Lumbar Spinal Stenosis 295
sensory changes, leading to "drop attacks," where during walking the patient
gets a sudden weakness leading to a fall. A more infrequent symptom IS a
chronic cauda equina syndrome with genital pain and disturbances.
Physical Examination
The physical examination follows the same principles as for all other prob­
lems of the lumbar spine with disturbances in the legs. Special emphasis is
placed on a few points.
I. The patient's ability to walk and peliorm other forms of physical activity
and the influence of posture on these abilities: The objective findings during a
physical examination may be few at rest but could become more pronounced
immediately after physical exercise or following a period of prolonged lumbar
2. The presence or absence of arterial disease: Palpation of the abdominal
aorta and auscultation for bruits in the iliac or femoral arteries are essential.
Palpation of peripheral pulses and when necessary, recording of ankle blood
pressure with the Doppler technique are also helpful.
3. Hip disease as degenerative arthritis is a common differential diagnosis
in this age group and examination of these joints is recommended.45
Since the typical patient with lumbar spinal stenosis usually has few patho­
logic signs in the physical examination, the great importance of the examination
is to exclude other serious diseases such as spinal tumors, neurologic problems
with demyelinating or peripheral nervous disease, hip disease, and vascular
problems. One should also bear in mind that in this age category it is quite
possible to have various combinations of spinal stenosis with hip disease or
vascular disease.
Morphologic Examination of the Spinal Canal
A morphologic examination of the spinal canal is essential to confirm the
diagnosis. On a plain x-ray of the lumbar spine the morphologic prerequisites
for the disease can be identified. It is also important to exclude unpleasant
surprises such as vertebral metastases. If any form of surgical intervention is
contemplated, a more detailed analysis of the spinal canal and nerve-root canals
is required.
Myelography has, ever since the syndrome was described by Verbiest,
been the method of choice to analyze the dimensions of the canal and visualize
any encroachment on the nerve roots. Despite the introduction of modern imag-
296 Physical Therapy of the Low Back
ing techniques such as CT scanning and magnetic resonance tomography
(MRT), myelography still has an advantage in certain respects. It gives a very
good overview of the canal, which is essential to determine how many levels
are affected and how surgery is to be planned. It also gives unique possibilities
for dynamic investigations, with pictures in neutral, flexion, and extension as
described by Sortland et al.46lt is also possible to obtain pictures with various
amounts of axial loading as proposed by Schumacher.
47 This not only helps us
to understand the mechanics of the root compression, but can also assist in
sharpening the diagnostic ability of the myelography by exaggerating the patho­
logic changes in borderline cases. The drawback of myelography is that it is
an invasive procedure with attendant risks and discomfort for the patient, and
it is also usually done as an inpatient procedure with the increased costs of
Computed Tomography
CT scans provide transverse cuts, perpendicular to the long axis of the
spine, with excellent opportunities for a detailed analysis of the size and shape
of the spinal canal and nerve-root canals. With an appropriate window setting
of the machine, it is usually possible to visualize the encroachment on the
nervous structures, also by the soft tissues such as the disc and ligamentum
flavum. It is easy to appreciate the importance of the soft tissues in this respect
by looking at Figure 11-8, which is a tracing from the CT scans of a patient
Fig. 11-8. Tracings from preoperative CT scans from a patient with central stenosis.
The tracings are made with and without the soft tissues to emphasize the importance
of the soft tissues in the deformation of the dural sac.
Lumbar Spinal Stenosis 297
with a central stenosis of the degenerative type. Under certain circumstances
an analysis of the influence of the soft tissues can be very diffcult on CT scans
(e. g. , with very obese patients or after previous surgery in the area). Some of
these diffculties can be overcome if the CT investigation is combined with
myelography, so that a small amount of contrast remains in the thecal sac when
the scans are done. Modern machines may also have the ability to provide
sagittal views and three-dimensional reconstructions, which give significant
contributions to the diagnostic information. In many centers today, it is custom­
ary to begin with the noninvasive CT scanning and proceed to myelography
only if necessary. In terms of central stenosis, the measurement of the critical
size of the dural sac, as outlined above, can be helpful in establishing the diagno­
sis. However, CT is still not suitable for dynamic investigations and one has
to remember that with the patient lying down, the axial load is removed from
the spine. Furthermore, to help the patients lie still and avoid distortion by
movement, they are told to assume the most comfortable position. This means
that they try not to provoke the nerve roots and in borderline cases the diagnosis
can be missed. These deductions can be drawn from the concept of dynamic
stenosis discussed above.
Magnetic Resonance Tomography
MRT has the advantage of giving both an overview and a detailed analysis,
both along the axis of the spine and perpendicular to it. It also gives unique
information about the hydration of the discs and their state of degeneration.
Furthermore it is noninvasive, cost effective, and without the dangers of ioniz­
ing radiation. The general impression is that MRT is better at depicting the soft
tissues and CT scans are best for demonstrating skeletal structures. Surpris­
ingly, one of the major contraindications for MRT is claustrophobia. Other
contraindications are metal implants in the patient and especially cardiac pace­
makers. MRT is not suitable for dynamic investigations in different postures
or with axial loads.
It is a general belief that the morphologic analysis of lateral stenosis is
more difficult than that of central stenosis, mainly because of the difficulty of
clearly visualizing the interaction between the spinal nerve and the root canal.
Van Akkerveken36 has shown convincingly that nerve-root blocks can help in
determining which spinal nerve is affected. This is of great importance in plan­
ning surgery so that the surgical trauma can be minimized.
Diagnostic Synthesis
Diagnostic analysis is based primarily on the clinical history. The clinical
examination reinforces the diagnosis, mainly by excluding other reasons for
the symptoms, but also by helping to decide the level of lateral stenosis with
rhizopathy. The morphologic evaluation of the spinal canal confirms the diagno-
298 Physical Therapy of the Low Back
sis and forms the basis for the planning of surgical intervention, if that is neces­
sary. One cannot warn too much against overreliance on the morphologic analy­
sis alone. A narrow lumbar canal without an appropriate clinical history does
not justify the diagnosis of spinal stenosis! Even with an appropriate history
and a narrow canal, it is absolutely essential to exclude other reasons for the
symptoms, such as neurologic disturbances of other kinds, hip disease, and
vascular problems. If surgery is contemplated, the risks of major lumbar surgery
are only justified by a sound clinical diagnosis following a complete investiga­
tion and not simply by a morphologic confirmation of a narrow canal.
As with most forms of low back pain, with or without sciatica, good treat­
ment is based on a correct clinical diagnosis and starts with conservative mea­
sures. A mild degree of numbness or weakness, combined with a gait where the
patient prefers to walk somewhat bent forward, is by many people considered to
be a natural consequence of old age. However, nowadays quality of life is an
important issue and even a mild form of lumbar spinal stenosis is often a reason
for a medical consultation. Usually, it is a natural anxiety about the possibility
of serious disease, with a concern about a rapid decline in physical ability, that
is the reason for the visit. Therefore it is important, even in the milder forms,
to secure the diagnosis with a reasonable degree of care and to exclude a more
serious disease. If this is done, sometimes the information about the spinal
stenosis and some advice on how to live with it is all that is needed at this
point. However, if the symptoms progress, a more active strategy is required.
Considering that the symptoms by definition are brought on by encroachment
on the cauda equina or the spinal nerves by the walls of the spinal canal or
nerve-root canals, it could seem natural to contemplate surgical intervention
as soon as possible when the symptoms have reached a certain level. However,
surgery for this disease is not without risks, some patients do not obtain relief
from an operation, and there is a risk of recurrence of symptoms. It is therefore
worthwhile to discuss the conservative, nonoperative alteratives to surgery.
It has been shown by 10hnsson et al48 that even with clinical and radiologic
signs of lumbar spinal stenosis, the prognosis was not bad even when surgery
was not performed and the patients were followed up for at least 4 years.
The reasons for this are still obscure, because one has to assume that
the degeneration responsible for the stenosis is likely to progress with time.
However, emphasis should be placed on the importance of studying the natural
history of any disease where major surgery is contemplated. I have no doubt
that surgery has a well-defended position in the treatment of spinal stenosis
and when symptoms are severe and progressive it is often the best solution.
However, there is almost always time to undertake a period of conservative
treatment and to evaluate its effectiveness. A conservative strategy could be
based on the following modalities.
Lumbar Spinal Stenosis 299
Bed Rest and Reduced Physical Activity
Bed rest and reduced physical activity should be contemplated only in
acute situations with severe back pain and sciatica. It is important to remember
that in patients at a relatively advanced age, inactivity brings special risks. Bed
rest for a maximum of 2 to 3 days can be an effective way of alleviating the
pain of sciatica. This may be achieved by a reduced disc pressure, an increased
size of the spinal canal due to reduced axial loading, or a reduction of an irrita­
tion of the nerve roots with increased blood flow after a reduction of pressure
on the roots. However, with this kind of treatment, reduced physical strength
comes rapidly and the accompanying mental inactivity does not help the pa­
tient's outlook. It is a potentially dangerous treatment in these respects, but
can be justified and effective if the pain is severe.
Corsets and Braces
The motive for the use of corsets and braces is to limit the motion of the
different segments of the lumbar spine, thereby achieving a symptomatic relief.
It may be argued whether a reduction in motion really is achieved by these
devices, but in some instances they give relief. In a study from 1985, Willner
et al49 showed a positive effect of a rigid plastic brace on the symptoms of
spinal stenosis and spondylolisthesis. A rationale could be that it helps to avoid
extension, and may thereby avoid part of the insult on the nervous structures.
It should be noted that they used a special jig to custom-fit the brace to the
patient so that the best possible posture for symptomatic relief was obtained.
They also excluded patients not suitable for treatment with a brace. This might
be an alternative to the short bed rest described above, avoiding the hazards
of bed rest.
Physical Exercise
Different forms of physical exercise programs have for many years been
part of the standard repertoire in the treatment of low back pain. They have been
advocated for two main reasons: first, to strengthen the muscles controlling the
movements of the lower back, achieving a better control of the motion segment,
and second, to increase the sense of comfort and physical security that a general
exercise program can give. This has been claimed to be the result of increased
levels of endorphins reducing the pain level.
Since this is mainly a geriatric clientele, exercise programs should be indi­
vidualized and monitored to avoid an increase in pain level as a result of the
program. As has been mentioned several times the symptoms are usually aggra­
vated by extension and, in my opinion, extension exercises should be avoided
in spinal stenosis. Riding a properly adjusted bicycle is usually a good way of
getting proper exercise and also gives a method of alternative transportation
300 Physical Therapy of the Low Back
when walking is difficult. It is important to restore and strengthen the patient's
physical function, and at the same time to educate patients on how to use their
body so as to avoid irritation of the nerve roots.
Drug Therapy
In an acute exacerbation of back pain with sciatica, drug therapy is often
a good alternative if the nerve roots or spinal nerves have been irritated by
prolonged intermittent compression producing "inflammation." According to
Rydevik and co-workers,50 edema in the nervous structures is part of the dis­
turbance caused by acute compression. A period of 3 to 6 weeks of anti-inflam­
matory treatment, usually with a nonsteroidal anti-inflammatory drug (NSAID),
may reduce the symptoms. If a reduction of edema is achieved, the space
available for the nerve roots is greater. This is especially important in the case
of a narrow canal where dynamic factors play an important role. One must
make sure that there are no contraindications to that form of treatment, such
as a tendency to gastric bleeding.
It is usually a good policy to use simple analgesics like paracetamol, instead
of more potent drugs that carry a risk of making the patient dependent on the
drug. Spinal stenosis is a more or less chronic condition and long-standing
medication with analgesics is always a problem. It should only be instituted
together with other conservative treatment modalities and should be carefully
monitored with very precise instructions on dose and length of medication from
the treating physician.
Back Schools and Other Education
Knowledge about the disease and the mechanisms behind the symptoms
helps the patient to deal with the problems of the disease. Whether this informa­
tion is delivered by the treating physician, the physiotherapist, or in an organ­
ized form such as a back school, is not the main issue. In my experience, the
physiotherapist is the best suited to give this kind of information, which can
be organized in a very practical format, together with an individualized physical
exercise program. Sometimes a geriatric clientele responds better to individual
instruction. The goal is to make the patients aware of how to deal with existing
symptoms and how to avoid acute exacerbations by adjusting their style of
living, if possible, without reducing their activity level or quality of life.
The conservative treatment strategy is based on whether the patient pre­
sents during an acute exacerbation or in the chronic phase. If acute problems
are present, a few days of bed rest combined with suitable drug treatment would
be a good start. Alternatively, a period with a rigid brace, individually fitted
according to the principles of Willner et al,49 could be of help. If the patient
presents without acute symptoms, or once these have settled, a change in life­
style should be instituted, avoiding harmful postures and with an individualized
Lumbar Spinal Stenosis 301
and carefully monitored exercise program. A good conservative program to­
gether with some optimism often leads to a significant improvement.
Surgical Alternatives
If conservative treatment fails to achieve a tolerable level of discomfort,
or if progressive neurologic deficit threatens the patient's activities of daily life,
surgical treatment should be considered.
The aim of surgery is to relieve the discrepancy in size between the spinal
canal and its neural content. This is achieved by removing parts of the wall of
the spinal canal or nerve-root canals. The main procedures in central stenosis
are laminectomy or laminotomy, where the whole lamina or part of the lamina
is removed. After that the ligamentum flavum is removed and in a few instances
the anterior canal is decompressed by a discectomy. There are variants in these
procedures, where the laminae are merely opened and hinged outward on a
lateral attachment and left in that position. It has also been suggested that the
lateral portion of the ligamentum flavum should be saved and a decompression
be confined to removal of the skeletal structures posterior to it, with the aim
of providing a soft-tissue barrier between the raw bone surfaces and the nerve
roots. To achieve a good decompression it is often necessary to sacrifice the
inner third of the zygapophyseal joint.
In lat�ral stenosis, a decompression by a foraminotomy has been the
method of choice. Various forms of undercuts, taking only the most anterior
part of the superior articular zygapophyseal joint, has been developed in order
to save as much as possible of the zygapophyseal joint and avoid secondary
It is common to perform a combination of both types of procedures. In
both cases there is a delicate balance between achieving a good decompression
and causing a significant instability. Secondary instability is a common postop­
erative finding, and therefore various types of fusion of the decompressed seg­
ment should be considered. The development of transpedicular screw fixation
has provided the surgeon with methods of providing mechanical stability until
a solid bony fusion of the stabilized segment is achieved.
Despite the combination of decompression and fusion, there is always a
risk of a recurrence of the symptoms after surgery. Two common factors are
new bone growth from the raw surfaces afer partial resection and a progression
of the degenerative changes responsible for the disease in the first place. A
postsurgical instability can significantly add to these problems. However, with
a successful outcome of the surgery there is a good relief of the symptoms with
a dramatic improvement in the quality of life for the patient.
I. Verbiest H: A radicular syndrome from developmental narrowing of the lumbar
vertebral canal. J Bone Joint Surg 36B:230, 1954
2. Verbiest H: Further experiences on the pathological infuence of developmental
narrowness of the bony lumbar vertebral canal. J Bone Joint Surg 37B:576, 1955
302 Physical Therapy of the Low Back
3. Bailey P, Casamajor L: Osteoarthritis of the spine as a cause of compression of the
spinal cord and its roots; with report of 5 cases. J Nerv Ment Dis 38:588, 1911
4. Elsberg CA: Experiences in spinal surgery. Observations upon 60 laminectomies
for spinal disease. Surg Gynecol Obstet 16: 117, 1913
5. Kennedy F, Elsberg CA, Lambert CI: A peculiar and undescribed disease of the
nerves of the cauda equina. Am J Med Sci 147:645, 1914
6. Sachs B, Fraenkel J: Progressive ankylotic rigidity of the spine. J Nerv Ment Dis
27: 1,1900
7. Sarpyener MA: Congenital stricture of the spinal canal. J Bone Joint Surg 27:70,
8. Epstein JA, Epstein BS, Lavine LS et al: Sciatica caused by nerve root entrapment
in the lateral recess: the superior facet syndrome. J Neurosrug 36:584, 1972
9. Epstein JA, Epstein BS, Lavine LS et al: Lumbar nerve root compression at the
intervertebral foramina caused by arthritis of the posterior facets. J Neurosurg 39:
362, 1973
10. Kirkaldy-Willis WH, Wedge JH, Yong-Hing K et al: pathology and pathogenesis
of lumbar spondylosis and stenosis. Spine 3:319, 1978
I I. Getty CJM: Lumbar spinal stenosis. The clinical spectrum and results of operation.
J Bone Joint Surg 62B:481, 1980
12. Schonstrom NSR, Bolender NFr, Spengler DM: The pathomorphology of spinal
stenosis as seen on CT-scans of the lumbar spine. Spine 10:806, 1985
13. Arnoldi CC, Brodsky AE, Cauchoix J et al: Lumbar spinal stenosis and nerve root
entrapment syndromes, definition and classification. Clin Orthop Relat Res 115:4,
14. Verbiest H: Neurogenic intermittent claudication in cases with absolute and relative
stenosis of the lumbar vertebral canal (ASLC and RSLC), in cases with narrow
lumbar intervertebral foramina, and in cases with both entities. Clin Neurosurg 20:
204, 1973
15. Gelman MI: Cauda equina compression in acromegaly. Radiology 112:357, 1974
16. Lutter LD, Lonstein JE, Winter RB et al: Anatomy of the achondroplastic lumbar
canal. Clin Orthop Relat Res 126: 139, 1977
17. Lutter LD, Langer LO: Neurologic symptoms in achondroplastic dwarfs-surgical
treatment. J Bone Joint Surg 59A:87, 1977
18. MacNab 1: Spondylolisthesis with an intact neural arch: the so-called pseudospondy­
lolisthesis. J Bone Joint Surg 32B:325, 1950
19. Weisz GM: Lumbar spinal canal stenosis in Paget's disease. Spine 8: 192, 1983
20. Johnsson KE, Petersson H, Wollheim FA et al: Diffuse idiopathic skeletal hyperos­
tosis (DISH) causing spinal stenosis and sudden paraplegia. J Rheumatol ' 10:784,
21. Lipson SJ, Haheedy MH, Kaplan MM et al: Spinal stenosis caused by lipomatosis
in Cushing's syndrome. New Engl J Med 302:36, 1980
22. Abdullah AF, Chambers RW, Daut DP: Lumbar nerve root compression by synovial
cysts of the ligamentum favum. Report of four cases. J Neurosurg 60:617, 1984
23. Andersson GBJ, McNeill TW: Lumbar Spine Syndromes. p. II. Springer-Verlag,
Vienna, 1989
24. Eisenstein S: The morphometry and pathological anatomy of the lumbar spine in
South African Negroes and Caucasoids with specific reference to spinal stenosis.
J Bone Joint Surg 59B: 173, 1977
25. Postachini F, Ripani M, Carpano S: Morphometry of the lumbar vertebrae. Clin
Orthop Relat Res 172:296, 1983
Lumbar Spinal Stenosis 303
26. Kirkaldy-Willis WH: Managing Low Back Pain. Churchill Livingstone, New York,
1 983
27. Schonstrom NSR, Lindahl S, Willen J et al : Dynamic changes in the dimensions
of the lumbar spinal canal. J Orthop Res 7: 155, 1989
28. Yong-Hing K, Reilly J, Kirkaldy-Willis WH: The ligamentum favum. Spine 1: 226,
29. Towne EB, Reichert FL: Compression of the lumbosacral roots of the spinal cord
by thickened ligamenta fava. Ann Surg 94: 327, 1931
30. Yamada H, Ohya M, Okada T et al : Intermittent cauda equina compression due to
narrow spinal canal. J Neurosurg 37: 83, 1972
31. Schonstrom NSR, Hansson TH: Thickness of the human ligamentum flavum as a
function of load. An in vitro experimental study. Clin Biomech 6: 1 9, 1991
32. Nachemson AL, Evans JH: Some mechanical properties of the third human l umbar
interiaminar ligament. J Biomech 1: 211, 1968
33. Ul lrich CG, Binet EF, Sanecki MG et al: Quantitative assessment of the l umbar
spinal canal by computed tomography. Radiology 134: 1 37, 1 980
34. Schonstrom NSR, Bolender NFr, Spengler DM et al : Pressure changes within the
cauda equina following constriction of the dural sac. An in vitro experimental study.
Spine 9: 604, 1984
35. Schonstrom NSR, Hansson TH: Pressure changes following constriction of the
cauda equina. An experimental study in situ. Spine 13: 385, 1988
36. Van Akkerveken PF: Lateral stenosis of the lumbar spine. A new diagnostic test
and its influence on management of patients with pain only. Thesis, University of
Utrecht, 1989
37. Olmarker K: Spinal nerve root compression. Nutrition and function of the porcine
cauda equina compressed in vivo. Acta Orthop Scand, supp!. 242: 1, 1991
38. Hoyland JA, Freemont AJ, Jayson MI V: Intervertebral foramen venous obstruc­
tion. A cause of periradicular fibrosis. Spine 14: 558, 1989
39. Weisz GM, Lee P: Spinal canal stenosis. Concept of spinal reserve capacity: radio­
logic measurements and clinical applications. Clin Orthop Relat Res 179: 1 34, 1983
40. Reuber M, Schultz A, Denis F et al: Bulging of lumbar intervertebral disks. J Bio­
mech Eng 104:187, 1982
41. Schonstrom NSR: The narrow l umbar canal and the size of the cauda equina in
man. Thesis, University of Goteborg, 1988
42. Dyck P: The stoop-test in lumbar entrapment radiculopathy. Spine 4: 89, 1979
43. Dyck P, Doyle JB: "Bicycle test" of Van Gelderen in diagnosis of intermittent
cauda equina compression syndrome. Case report. J Neurosurg 46:667, 1977
44. Wilson CB, Ehni G, Grollmus J: Neurogenic intermittent claudication. Clin Neuro­
surg 18:62, 197 1
45. Bohl WR, Steffe AD: Lumbar spinal stenosis. A cause of continued pain and disabil­
ity in patients after total hip arthroplasty. Spine 4: 168, 1979
46. Sortland 0, Magnaes B, Hauge T: Functional myelography with metrizamide in the
diagnosis of lumbar spinal stenosis. Acta Radio, supp!. 355: 42, 1977
47. Schumacher M: Die Belastungsmyelographie. Fortschr Rontgensrt 145: 642, 1986
48. Johnsson KE, Rosen I, Uden A: The natural course of lumbar spinal stenosis. Clin
Orthop Relat Res 279:82, 1992
49. Willner S: Effect of a rigid brace on back pain. Acta Orthop Scand 56: 40, 1985
50. Rydevik BL, Pedowitz RA, Hargens AR et al: Effects of acute graded compression
on spinal nerve root function and structure. An experimental study on the pig cauda
equina. Spine 16: 487, 1991
12 Low Back Pain in the
Workplace: An
Ergonomic Approach
to Control
Margaret I. Bullock
Joanne E. Bullock-Saxton
Low back pain ( LBP) is one of the most common di sorders seen today
and many studies of its incidence have revealed that i t may occur in as many
as 80 percent of the popul ation. 1 . 2 Biering-Sorensen3 has al so pointed out the
hi gh prevalence of LBP associated wi th work. Al though the i nformation on the
importance of various workplace factors to the occurrence and etiology of LBP
i s i ncomplete, workers i n many occupations complain of tension i n the shoul ­
ders and in the upper and lower back, which i nterferes wi th their workpl ace
Accordi ng to Rowe, 4 85 percent of LBP sufferers have i ntermi ttent attacks
of di sabl ing pain every 3 months to 3 years and, as work absence i s often
consequential on i ndustrial injury, thi s high inci dence is costl y to industry. Yu
et al5 have drawn attention to the economi c consequences of LBP and indeed,
according to some estimates , in a pl ant wi th 1 000 employees , an absenteei sm
rate of 5 percent woul d cost about $1 mi l l ion per year .
Back pain can arise from many causes, incl uding those related to pathologic
problems el sewhere i n the body, or to a traumatic i ncidence in which some
component of the vertebral system i s i njured. However , i n many i nstances,
back pain develops unexpectedl y wi thout di sr uption of the normal daily pattern
306 Physical Therapy of the Low Back
of acti vi ty. I n such cases, it might be assumed that l ocal degenerative changes
or conditioni ng from previ ous postural or vi bratory stress had increased the
suscepti bi l i ty to a sudden onset of symptoms.
Some authors bel i eve the back t o be parti cul arl y susceptible to earl y degen­
erative phenomena.6-8 If degeneration is advanced, insignificant amounts of
trauma may precipitate the onset of pain in a person wi th no previous history
of back di sorders. This is confirmed if the figures for i ndustrial accident reports
are consi dered. I n a study of 148 workers from a Briti sh company presenting
to the physiotherapy department with back pain, Strachan9 found that 35 per­
cent of cases coul d name no known reason for the onset of pai n. In a further
14 percent of cases, an unexpected onset of pain occurred duri ng the perfor­
mance of an everyday task that had previousl y presented no problem.
It is diffcul t to diagnose and cl assify back pain and i njury. Many variables
may be i nvol ved i n the cause of LBP, i ncl udi ng physi cal characteri stics; the
experience of the worker; the demands of the task; organizational, social , and
cultural i nfl uences; the characteri sti cs of the work environment; and the out­
comes of the i nj ury. Furthermore, causes of LBP are not necessari l y physical .
Back pai n experienced by the worker may be based on a mi xture of organic
and psychological causes, as wel l as social di sturbances that may be i nfl uenced
by the cul t ural background of t he person.
Such a multifactori al probl em demands resol ution through a variety of
measures and often by a group of peopl e wi th differi ng but complementary
ski l l s. A physiotherapi st i s one of such a team. The physiotherapist ' s role in­
cl udes both therapeuti c acti viti es and rehabil itat ion and prevention of i njury,
and concern with t he l atter process i mpl i es an understanding of the objectives
and pri ncipl es of ergonomi cs.
Ergonomi cs i s concerned wi th ensuri ng that the work place i s so designed
that work-i nduced i njuri es, di sease, or di scomfort are prevented and safety is
ensured, whi l e effci ency and producti vi ty are maintai ned or i ncreased. Work
stress probl ems at home, school, or at the place of employment need attention
and often require the cooperati ve acti vi ty of representatives of a number of
di sci pl i nes, such as engineeri ng, psychology, medici ne, and physiot herapy, to
provi de sui tabl e sol ut ions. Physiotherapists fit i nto thi s study of man-mach­
ine-task relationshi ps because of their special abi l i ty to analyze body move­
ments in detail and to eval uate postural abuse during dynamic situations. 10 The
physiotherapist can hel p to el i mi nate mi suse of the body and assist in the design
of equi pment and work areas so that the situations so arranged are better suited
to the physical wel l-bei ng of the person usi ng them.
Bul l ock
has described the many facets of the preventive role of the
physiotherapist, whi ch i ncl ude such responsi bi l ities as job anal ysi s; work pos­
ture monitoring; task design; personnel sel ection and placement; education;
supervi sion of work methods; i nfl uenci ng of moti vation and attitudes; and pro­
vi sion of appropriate activi ty breaks, exerci se, and physical fitness programs.
Al t hough onl y physiotherapists wi th further education in the practi ce of ergo­
nomics are l i kel y to act as consul tants to industry, al l physiotherapists have a
part to pl ay not onl y i n health prevention, but al so in t he prevention of injury
Low Back Pain in the Workplace: Ergonomics 307
or re-injur y in cl i ents under their care. The hi gh i ncidence of LBP in the commu­
ni ty and i n cl i ents presenti ng for physiotherapy suggests that methods of con­
trol ling i t , whether associated with work, home, school , or l ei sure, shoul d be
addressed by the physiotherapi st .
Prevention of LBP or injury rel i es on an under standi ng of the factors that
may contribute to that pai n, the i mplication of those factors for the production
of LBP, and an. appreciation for the rational e for use and the rel ati ve effecti ve­
ness of different approaches that may be taken to control the ri sk factors . These
aspects are considered in thi s chapter .
It appears that a number of vocational factors are associated with mechani­
cal i njuri es , al though indivi dual factors and abnormal iti es al so play a part . An­
dersson 1 2 has proposed that these factors i ncl ude physi cal l y heavy work, static
work postures, frequent bending and twi sti ng, l ifti ng and forceful movement s,
repetitive work, and vi brations.
Manning and Shannon 1 3 suggest three possi bi l i ti es to expl ain how these
factors coul d produce pai n in the low back: abnormal strai n on a normal back;
normal stress on an abnormal back; and normal stress on an unprepared normal
Postures adopted during activity and devel oped over a period of ti me ap­
pear to have a major association wi th the presence of LBP. For exampl e, An­
dersson 1 2 suggests that prolonged si tti ng, dri vi ng of vehi cles, and bent-over
work postures seem to carry an increased ri sk of LBP. It i s general l y accepted
that certain occupational and postural stresses on an already inflicted back wi l l
produce further episodes of pain.2• 1 4
The stresses of vertical compression, hori zontal shear , rotary torque, or
a combination of these are determi ned by the i niti al , final , and i ntermediate
postures of the person, the velocity of movement s, and the load carried by the
subject during the task performed. IS Nachemson et al 1 6 consi der that mechani­
cal stress has at least some role to play i n the etiology of LBP syndrome.
The Infuence of Position
It is interesting to note that attempts have been made to cal cul ate the load
on the l umbar discs in different positions on the body. Nachemson 17 found an
increase in load from l yi ng, to standi ng to si tti ng. He also found that forward
leaning i ncreased the load. Nachemson 17 considered t hat these hi gh stresses
produced wi thin the l umbar disc possibl y pl ay a rol e i n the occurrence of poste­
rior annul us ruptures and that dynami c forces woul d increase the magnitude
of the stresses on the annul us. Such fi ndings coul d be reflected i n the results of
Kel sey's experi ments, 1
where it was demonstrated that sedentary occupat ions
308 Physical Therapy of the Low Back
were conduci ve to the development of herniated l umbar di scs, most especial l y
among t hose aged 35 years and older, and more specifi cal l y among those who
sat for half of the time or more at thei r jobs.
Physiotherapi sts need to be aware of the implications of postural load for
LBP and the importance of provi di ng relevant advice that could help to reduce
postural stress. In t hei r consi deration of the features of work that infl uence
postural l oad, special attention needs to be given to demands for sustained
positi ons, bent-over work postures, and the influence on posture and activity
of the seated positi on.
Sustained Positions
Occupati ons i nvol ving light work with sustained postures have been re­
ported to have a high i nci dence of back ache . Partridge et al19 have clai med
that sustained bent-over worki ng postures are l ikel y to precipitate back pai n,
whereas Magora's study20 i ndicated that the i ncidence of back ai l ments was
extremel y hi gh in occupations i nvol vi ng prolonged sitting of longer than 4 hours
or sustained standi ng i n one pl ace.
Static work i s characterized by slow contracti ons wi th heavy l oads or by
long-lasting hol di ng postures. I n a strong static contraction, the blood supply
i s i mpai red and waste products accumulate i n the muscl es. According to
Grandjean and Hunti ng, 21 thi s i s the reason for acute pain i n the static pre loaded
muscl e. If thi s i s repeated frequently and for long periods of ti me, chronic pains
may resul t. These are due to pathol ogi c changes not onl y i n the muscl es, but
al so in the connecti ve ti ssue of tendons, joi nt capsules, and joint ligaments.
Thi s may play a part i n the postural backache suffered by workers who are
not abl e to carry out normal movement during thei r worki ng day. Corl ett and
Manencia22 have argued that because many muscl e groups are i nvol ved i n hold­
ing a posture, it is possi bl e that thei r relative contribution to the total supporti ng
force required is changed during the period for whi ch the posture i s held. The
importance of avoi di ng or reduci ng the durati on of stati c muscl e stress can
therefore be appreciated.
Bent-Over Work Postures
Investigati ons i ndi cate that LBP i s more frequent in people with predomi­
nantl y bent-over work postures, where the load on the back i s increased. It
shoul d be noted t hat when the l umbar spine is flattened as in the stooped posi­
tion, the zygapophyseal joints are less abl e to resist compressi ve forces and
the majority of the i ntervertebral compressi ve force must be resisted by the
di sc. Ki sner and Col by23 attri bute postural pai n to the mechanical stresses on
various structures when a faulty posture i s maintai ned for a lengthy period.
Twomey et al 24 have explained how the process of creep in fl exion occurs when
the spi ne i s loaded i n ful l fl exi on for a sustained period. These authors have
Low Back Pain in the Workplace: Ergonomics 309
pointed out that if such loads are prolonged beyond I hour, wi th mi ni mal acti vi ty
i nto another posit ion, as may occur in some occupati ons, there may be a consi d­
erable degree of extrusion of fl uid from the i ntervertebral di scs, the arti cular
cart ilage of t he zygapophyseal joint s , and t he spinal l i gament s. Twomey and
Taylor25 suggest that it may take many hours of rest for fl ui d to be reabsorbed
into the sof ti ssues, and thei r shape re-establ i shed. These considerati ons rein­
force the need to maintain the natural forward curve in the l umbar region duri ng
worki ng activities and expl ai ns why it is i nadvi sabl e for those wi th LBP to si t
leaning over a desk. Furthermore, opportuni ties for a change of position shoul d
be provided.
Seated Postures and Activities
The cause of the high i ncidence of back pai n i n sedentary workers is not
entirel y clear since those wi th back pai n may have el ected employment in a
sedentary occupation.26 On t he other hand, it may be the sedentary l ife-styl e
itself that contributes to the onset of symptoms.
Griec027 has noted that because of t he automati on and i ntegrati on of previ­
ously physi cal tasks, much heavy manual work has been replaced by f i xed
postures, particularl y the si tti ng position. However, the attenti on given to con­
trol l ing risks in heavy manual work has not been appl ied to other aspects of
work postures. The prolonged sitting posture has brought wi th it a new problem,
not only for the workplace, but also for l ife-styl e in general . Janda28 has attrib­
uted thi s sedentary l ife-styl e and associated decrease i n movement to the devel ­
opment of muscl e i mbal ances, whi ch he bel i eves often predispose workers to
Physiologic and epidemiol ogic studies have demonstrated that prolonged
sitting work can cause lower back pai n. 29 I n studyi ng the al terati ons of the
l umbar curve related to posture and seati ng i n various posi ti ons, Keegan30 noted
that the reduction of the l umbar curve in some posi ti ons tends to force the
central portion of t he l ower l umbar di scs posteriorly by hydraul i c pressure f rom
anterior wedgi ng. Such flatteni ng of the l umbar curve could be caused by the
tightening of the posterior thigh and gl uteal muscl es duri ng hip fl exi on, due to
their attachment to the i schi um, the sacrum, and the i l i um. Keegan' s resul t s30
emphasize the fact that sitting with a 90° angl e at the hip joi nts causes consi dera­
ble strai n at the l umbosacral junction, whi ch woul d be i ncreased by f urther hip
fl exi on. Bodguk and Twomey3) have explai ned how in si tti ng, due to the de­
crease in l umbar lordosi s, the i ntervertebral discs are compressed.
Flattening of the l umbar spi ne al so occurs in the si tting position when the
legs are stretched out i n front of the body. Thi s can occur when a seat i s too
deep. If the person si ts wel l back to make use of the leg rest , the knees do not
reach the edge and the l egs must be extended further than normal . Thi s pul l s
on t he muscl es behind the t highs, ti l ti ng t he pel vi s backward and flatteni ng the
lumbar spi ne. Si tti ng wi th the l egs stretched out because the seat i s too l ow
has a similar effect . Sel ecti ng a chai r that i s the correct height and depth and
310 Physical Therapy of the Low Back
that has a l ow back support to encourage forward movement of the pel vis and
maintai n the l umbar curve is t herefore critical i n ensuri ng comfort and safety.
Dri vi ng a vehi cl e, an occupation i nvol vi ng prolonged sitti ng, al so presents
a risk of LBP. Kel seyl8 observed that dri vi ng for long periods of ti me often
requires prolonged sitting in a seat with i nsufficient support for the low back
and with legs extended, i n a positi on subject to the vi brati on from the road and
mechanical stress from starti ng and stoppi ng. The absence of power steering
in such ci rcumstances can create a substantial ri sk to the operator, where the
appl ication of pushi ng and pul l i ng forces on the steering wheel could induce
major spinal stresses.
Andersson et al32 consider that for the seated operator, maneuvers such
as the gear shift and preparation for cl utch depression can al so increase load
on· the l umbar spi ne and can i ncrease the myoelectric activity of the erector
spinae muscl es. These researchers found that l ifting the lower l imb to place
the foot on the pedal placed stress on the l umbar spine due to the contraction
of the psoas major muscle, whereas intradi scal pressure of the thi rd lumbar
disc was found to i ncrease markedl y duri ng the depression of the cl utch.
Nachemson33 had previ ousl y found that myoelectric activity i n the psoas mus­
cle corresponded with an i ncrease in the i ntradi scal pressure in the l umbar
spi ne. This suggests that a cl utch pedal , whi ch requires greater acti vity of the
psoas muscl e (due to a high location above the floor), may produce more stress
in the l umbar spine than a lower-placed pedal .
Such fi ndi ngs are parti cularl y rel evant to drivers of heavy vehicl es, such
as trucks and tractors, i n whom the inci dence of LBP i s high . They support
the fi ndi ngs of Bul lock,34.35 who determi ned the opti mal relati onship of a pedal
to the operator in terms of mi ni mal spinal movements. Among her recommenda­
tions was the use of a pedal that requi red mi ni mal hip fl exion and abduction
to reach it . Advice to drivers with LBP should i ncl ude recommendations either
for selecti on of trucks or tractors with suitabl y designed pedal locations or for
incorporation of appropriate modifications to the seat/pedal relationship to
avoid excessive l umbar movement during pedal use.
Vibrational Conditions
Workers such as bus and truck drivers and heavy equipment operators are
subjected to conti nuous vi brati on. The effect of vi bration on height , i mpl yi ng
spinal load, has been demonstrated by a number of researchers,36.37 and the
high frequency of l ow-back complaints among truck drivers and drivers of work
machi nes has focused i nterest on a possi bl e relati onship between LBP and
vi bration. Troup2 has advised that epidemiologic data suggest that the longer
the time spent dri vi ng, the greater the risk of back trouble. Kakosky38 has
asserted t hat "vi brati on transferred from a machi ne to the human body may
cause di scomfort , a reduction i n performance or even injury, " whereas expo­
sure to whole body vibrati on i s l i sted as one of the factors causi ng LBP at
Low Back Pain in the Workplace: Ergonomics 31 1
work.39 Control of vi brational conditions is therefore an important aspect of
prevention in i ndustry.
The greatest dynamic l oad on the trunk and probabl y on t he spi ne occurs
when it is vi brated at i ts natural frequency. Stress is therefore l ikel y to be
greatest when the seated body is vi brated verti cal l y i n t he range of 4 to 8 Hz. 2
I n a truck, vertical vi bration i s t he domi nant vi bratory mode and occurs i n the
range of 2 to IS Hz. To prevent vi bration i n the range of 4 to 8 Hz, firm
cushions should be suppl ied and the seat shoul d be suspended to gi ve it a natural
frequency of less than 1.5 Hz.2
Effect of Other Musculoskeletal Injuries in the Workplace
The LBP syndrome may sometimes present a wide variety of neuro­
orthopedic signs .20 Some of t hese manifestations are the resul t of the underl yi ng
process causi ng LBP, whi l e other cl i nical l y objecti ve si gns may be related ei ther
directl y or i ndi rect l y to the actual cause of the LBP. For exampl e, sacrol umbar
postural di sorders , affections of the hip and kneejoi nts , foot deformi ty, unequal
length of the lower l i mbs, muscle contractures, or poor muscl e coordi nation
and function have al l been found to be accompanied by LBP.40-42 The interrela­
tionships between the l umbar vertebrae, sacrum, pel vi s, and femur are prime
factors i n standing posture as well as duri ng acti vi ty and it is not surpri si ng
that strains in one area can lead to stresses i n another.43
Magora20 found that 1 1.4 percent of LBP patients had at least a mi l d degree
of l i mitation of movement in the hip, whereas onl y about 2 percent of the
controls had mild uni l ateral or bi lateral l i mitations. Thi s di screpancy poi nts to
a possibl e connection between hip l esi ons and LBP. Thi s coul d be expl ai ned
by a secondary postural di sorder and possi bl y by the compensatory degree of
additional mobi l i ty and use demanded of the l umbosacral spi ne.
The infuence of injuries i n t he l ower l i mbs on the development of LBP
must be considered by those concerned wi th preventi on. Sl ippi ng and tripping
are common industrial accidents and occur also i n the domestic situation. 13
The i ncidence of ankle and knee sprain that is often associated wi th these
accidents should alert the attendi ng physiotherapi st to the possi bi l i ty of future
LBP, so that appropriate preventi ve measures can be t aken. Bul lock-Saxton44
has demonstrated that the function of muscl es around the hip and l ow back
changes significant l y after ankle sprain and has recommended attenti on to nor­
malizing muscl e i mbalances and then the appl ication of sensory-motor pro­
grams, such as those advocated by Janda and Vavrova.45 Such an approach
would ensure maintenance of proprioception as wel l as muscl e functi on. Thi s
recognizes that t he maintenance of the correct upright posture depends on the
coordinated acti vi ty and endurance of many muscl es and that muscl e control i s
largely automati c. But, as Robert s46 poi nted out , after i njury or duri ng di sease,
altered neural act i vi ty reachi ng the central nervous system is l ikel y to i nfuence
the automatic mechanisms for muscl e control . I n time, the person may develop
new patters of muscl e act i vi ty and thi s in turn may lead to muscl e i mbalance,
31 2 Physical Therapy of the Low Back
unusual postures, or interference with the capacity to carry out certain move­
ments. Much can often be learned about the nature and location of an inj ury
from the changes in posture and movement patterns t hat fol l ow it . Unfortu­
natel y, such new patterns may persist long after the original i njuries have
heal ed, l eavi ng abnormalities of gait and posture attributabl e to habit . 46
Active programs for prevention of LBP should i ncl ude proper considera­
ti on of the l ong-term i mpl icati ons that coul d occur in areas remote from the
site of i njury, and shoul d i ncl ude comprehensi ve assessment and management
of the muscul oskeletal system fol lowi ng lower l i mb i njuries.
Muscl e weakness and muscl e i mbal ance may al so devel op as a result of
i nj ury, through poor postural habits or through overactivity or underacti vity
of certai n groups of muscl es i nvol ved i n work or play. Once an imbalance has
occurred, Janda and Schmi d47 bel i eve that changes i n muscl e function play an
important role in the devel opment of many subsequent painful conditions of
the motor system, such as LBP syndromes . They are an integral part in the
postural defects that can be associated with work. These authors argue that
the acti vi ty of i ndi vi dual muscl es is not of such great importance as thei r coordi­
nated acti vity withi n different movement patterns. As a result of their cl i nical
observati ons, they suggest t hat certai n muscl es appear to respond to pain by
t ightness and shorteni ng, whi l e others react by i nhi bition, atrophy, and
I t has been observed t hat i mbalance between muscl es often starts t o de­
velop in the pel vi c-hip compl ex. 47 . 48 Here, an i mbalance develops between
shortened and tight hip fexors and trunk erectors in the l umbar region on the
one hand, and weakened gl uteal and abdomi nal muscl es on the other. Muscles
that become tight or hypertonic are readi l y activated in most movement patterns
and thi s i s reflected by thei r earl i er recruitment during movement . The changes
in muscl e l ength and recruitment patterns are considered to lead to anteversion
of the pel vi s, together with l umbar hyperlordosi s and sl ight fl exi on in the hip.
Thi s in turn may cause unfavorabl e changes of pressure di stribution on the
di scs, joi nt s, and l igaments in the l umbar region and hip joi nts.
These views tend to confirm Michel i's i mpression49 that many young ath­
letes with LBP have relati vel y tight musculotendi nous and l igamentous struc­
tures about the low back, hips, and knees. He has observed that many appear
to have relati ve weakness of the anterior muscl es i ncl udi ng the abdominals,
and a tightness of the hi p flexors and anterior el ements of the hip. It shoul d be
remembered, however, that on the other hand, many young gymnasts with
back pai n of either a spondylogenic or mechanical nature have a great flexibility
although they can have relati vel y weak abdominal muscl es. Mi chel i49 suggests
that thi s weakness may represent a relative muscul otendi nous imbalance. The
i nfl uence of a muscl e imbalance on the recruitment of muscl es about the pel vis
may be the most sal i ent point , where a tendency to decreased pel vic stabi l ity
and control i s noted. The possi bi l ity of t hese muscl e i mbalances needs to be
recognized by the physiotherapi st i nvol ved in desi gni ng acti vity exerci se pro­
grams for i ndi vi dual cl ients or for groups in i ndustry or sport .
Low Back Pain in the Workplace: Ergonomics 31 3
Manual Handling and Lifing
Although mechani zation has been i ntroduced i nto many areas of work,
prolonged heavy manual work i s sti l l necessary withi n some workplaces, partic­
ularly i n relation to loadi ng and unl oadi ng or to transferring goods. Further,
l ifting and carrying i n the home or i n l ei sure acti vi ti es i s not an infrequent
acti vi ty. The relationship between heavy manual work and. the frequency of
LBP is a much di scussed subject. For exampl e, Troup7 found a combi nati on
of flexion and rotation when lifting to be the most common cause of back pai n
at work. Stubbs50 has al so noted that LBP ari si ng from manual handl ing i s a
constant hazard in the i ndustrial envi ronment . Workers wi th back pai n l ose
more days from work when t hei r jobs i nvol ve heavy l oads51 and, because pre­
exi sti ng l umbar spine condi ti ons can be aggravated by heavy loads , i t fol l ows
that l oads on the l umbar spi ne should be kept as light as possibl e.
Occupati ons i nvol vi ng onl y occasional l if ti ng have al so shown a high i nci ­
dence of back pai n. 1 4 Magora20 cl ai ms that infrequent physi cal demands and
sudden unexpected movements such as a rapi d stretch, fl exi on, or rotational
movement gives" . . . more low back pain ofen than conti nuous heavy work. "
That thi s may be related to poor executi on of t he task i s supported by the
evidence that a moderate amount of dai l y lifti ng was not found to infl uence t he
rate of back ache. 20
The load on the spi ne duri ng a l ift i s related to a number offactors, i ncl udi ng
the weight of the object to be l ifed, the horizontal di stance from the body from
which or to whi ch it is l ifted, the height or vertical di stance of the l ift , the
frequency of l ifti ng, the durati on or peri od of l ifti ng, the speed of the l i f, t he
body posture of the worker, and the bulk of the object l ifted.
It has been argued that l umbosacral strain when l i fti ng can occur through
lifti ng a load too heavy for t he person's physi cal strength or l ifting a load withi n
the person' s strength capaci t y, but i n an i ncorrect manner. Goldberg et aJ52 have
suggested that l umbosacral strai n can al so occur through improperl y arching the
spine into a lordotic position once the load i s l i fted. The proposal that the
method of lifting may be an important factor i n the production of LBP has come
from a number of authors . For example, Chaffn and Park53 have stated that
the stresses induced at the lower back during weight l ifting are due to a combi na­
ti on of the weight lifted and the person's method of l ifti ng. The latter may be
due to the person's approach or to work si tuati ons t hat impose restri ctions on
the way a l ift may be performed.
Physiotherapi sts need to gi ve some consi derati on to l ifti ng methods i n thei r
prevention program, and they need to be aware of the potential hazards of the
various approaches to l ifti ng. Advice about manual handl i ng depends on many
factors, but i n parti cular, must be approached i n terms of the i ndi vi dual person's
capaciti es and movement abi l i ti es.
Gravi tational forces acti ng on the load hel d i n the hand and the person's
body mass create rotational moments or torques at the vari ous arti culations of
the body. The skeletal muscl es are positioned to exert forces in such a manner
t hat they counteract these torques. The amount of torque at any joi nt i s depen-
31 4 Physical Therapy of the Low Back
! < -
-- a
Fig. 12-1. Extensor muscle force i n
a stooped posi tion:
E(= -
C) =
W x a
E. extensor muscle force; W. weight
of upper body and load; a, moment
arm of body weight; h. moment arm
of the extensor muscl e force; and C.
compressive force acting l ongitudi­
nal l y through the vertebral bodies
and discs.
dent on the product of the force tendi ng to rotate the segments and the moment
arm of that force (i . e. , the di stance from the joint to the force vector measured
normal to t he force vector) , as shown in Fi g. 12-1 . The magnitude of the moment
arm varies wi th the person' s posture. If the load l ifted is held cl ose to the body,
the moment arms are smal l and the resulti ng torques at the joints are smal l . If
the load i s hel d away from the body, the large moment arms wi l l cause large
torques. Thi s emphasizes the importance of posi ti oni ng the body cl ose to the
load, which al so has i mpl icati ons for both access and the bulk of the object.
Studi es54 have shown that during weight l ifting, t he bending moment at
the l umbosacral juncti on can become qui te large. To counteract thi s torque,
the muscl es of the low-back region, pri mari l y the erector spinae group, must
exert correspondi ngl y hi gh forces, si nce they operate on smal l moment arms .
The high forces generated by the l ow-back muscl es are the primary source of
compressi on forces on the l umbosacral di sc.
I n l ifti ng, the greatest stresses occur at the begi nni ng when the inertia of
the weight has to be overcome and the load accelerated. This is why it i s
important for the person to be i n a posi ti on to move freel y into t he upright
posture wi th the l oad. It is not wi se to start a l ift in an awkward position, which
will prevent a smooth l ift . Storage of i tems and arrangement of the workplace
are i mportant factors i n t hi s respect .
The trunk extensor muscl es act strongl y around the vertical but they do
not work when the back is ful l y flexed and onl y commence thei r acti vi ty afer
30° of extensi on. Electromyographic ( EMG) studies55-57 have demonstrated the
el ectrical si l ence of the erector spinae i n the ful l y flexed position, the activity
of hip extensors as the trunk begi ns to rise, and the later vigorous activity of
the erector spinae as the load nears the vertical . That i s, return to t he upright
Low Back Pain in the Workplace: Ergonomics 31 5
position from a fexed posture i s begun by the hip extensor muscl es rather than
the l umbar spine extensors so that the posterior part of t he di sc, intervertebral
disc , and the posterior vertebral l igament are not protected by acti vel y contract­
ing trunk muscl es during the early stages of lifting from a fexed positi on. The
l ikel ihood of damage i s therefore greater if load is placed on the back in thi s
position. Thi s emphasi zes the val ue of i nitiati ng a l ift through movement of the
legs rather than the trunk. Further, where a stoopi ng positi on i s prol onged,
muscles fatigue and thei r supporti ng function i s decreased. Agai n, thi s pl aces
a greater demand on the posterior aspect of the i ntervertebral disc and the
posterior ligaments. It has been argued that conti nued work from a flexed pos­
ture coul d ulti matel y produce strai n and could predi spose to LBP.
Disc damage has been l i nked with sudden hi gh l oadi ng of t he spi ne, espe­
cial l y in flexed postures (e. g. , when a worker catches a heavy l oad) . This can
also occur when one of two partners loses grip on an object being l ifted, so
that the remai ning partner unexpectedl y recei ves an i ncreased l oad. For thi s
reason, group l ifti ng needs careful coordi nati on, i n terms of number of person­
nel used, size of lifers, understandi ng of the proposed movements, appreciation
of desti nati on, prel i mi nary position and grip to be used, and action commands
for lifti ng, carryi ng, and l owering.
Speed of movement i n l ifting and handl i ng may also be important , for if
the movement i s not adequatel y under muscular control , additional stress may
be suddenly imposed on the spinal ligaments and a sprai n may occur. Work
demands that requi re rapid transfers of weights shoul d be avoided.
For those who lift from a knee fl exed positi on, the principal muscl e work
invol ved in rai si ng the load relates to the extensi on of the knees and hips.
Because the trunk i s held more cl osel y to t he vertical , t he vertebral col umn i s
not placed in a position of vul nerabi l ity. If some movement does occur i n t he
spi ne, it is through a range where the di stance of the combined load and weight
of the trunk from the center of gravity is smal l and the effort required by t he
extensor muscl es is l ess than that requi red if the trunk were horizontal .
Probabl y the major probl em with the knee action l ift i s i n whether the
person has sufficient strength in the extensors of knees and hips to rai se the
load. These muscl es work most efficientl y and effecti vel y when the joi nts are
at 90°. If the person squats before l ifti ng, the extensor muscl es are placed in a
position of mechanical di sadvantage and it may be very difficult for the person
to raise up to standi ng. Thi s highl ights the need to provi de hi p and knee extensor
exercises in a prevention program.
Noti ng the chal l enges to the val idity of the recommended lifting method
with straight back and bent knees53 and to the practical uti l ity of such a
method,58 Parnianpour et al59 applied a lifting stress cal culator to i ndi vidual i ze
and optimize the l ifting techni que and to take i nto account cl i ni cal compl ai nts.
Parnianpour et al59 reported that thi s model showed variabi l it y of knee joi nt
and back joint angles for different loading conditi ons, but they emphasi zed that
although the idea of di stributing l oads between the knee and the back accordi ng
to a patient's symptoms may be appeal i ng, there is a l imited number of postures
that can be assumed to carry out a l ift i n view of the physical characteristi cs
316 Physical Therapy of the Low Back
of that load. These experimenters coul d find no si ngle "safe" method of l ifti ng,
supporti ng previ ous assertions of the inadequacy of trai ni ng in proper l ifing
techni ques i n reduci ng LBP and the possi bi l i ty for greater success i n injury
preventi on t hrough the redesign of the task or workplace.
When the physiotherapi st ' s approach to LBP is sol el y that of "treatment , "
t he focus i s on helpi ng t he i ndi vidual t o regai n function or compensate for its
loss. De Jong60 has cri ti ci zed this l imi ted approach i n whi ch the physiotherapist
aims to mi nimi ze the effect of sensory or physi cal i mpairment on movement s,
bel i evi ng that whi le it may resolve i mmediate problems, it ignores t he infuences
of attitudes and environments on the experience of di sabi l i t y. A broader ap­
proach, that of di sabi l i ty management ,6 1 is concerned not onl y with developing
the abi l i ties of those who are di sabled, but also with changing attitudes of the
professional s and employers to disabi l i t y. Thus, i n di sabi l i ty management, ob­
stacl es occurring i n t he envi ronment are removed i n order to restore the individ­
ual's worki ng capaci ty. Such programs incorporate mechani sms commonl y
used i n ergonomics to prevent the occurrence of di sabi l i t y. Because LBP i s
a mul tifactorial probl em, a team approach i s necessary to provide adequate
management . Modes of treatment need to be varied according to the individual
ci rcumstances.
The emphasi s on prevention i n ergonomics i s on risk control . Thi s rel i es
on t he i dentification of risk factors associated wi th i njury, evaluation of those
risks, and i mpl ementat ion of control s that take t hose risks into account . Fol l ow­
ing a job anal ysi s , correcti ve and preventive measures can incl ude such features
as advi ce, education, and trai ni ng i n ergonomic principl es; a program of instruc­
tion for newcomers to the workplace; redesign of equipment and the work
envi ronment; changes i n organi zati on and work methods; relaxation and exer­
ci ses; stress rel ief; and appropriate assessment and treatment should symptoms
Job Analysis
Through risk anal ysi s , those aspects of the work si tuation that represent
a risk to the worker can be identified and assessed, so t hat areas of risk can
be placed in order of priori t y. Three basic approaches are usual l y adopted in
risk i dentifcation: anal yses of stati stics, consul tation with relevant personnel ,
and surveys and observati ons in the workplace. Stati sti cs such as incidence of
acci dents and injuries can reveal where and in what jobs specific injuries have
occurred. The frequency and severity of i njuries relative to the number of em­
pl oyees, the hours worked, or the areas of work can be exami ned, and i njury
Low Back Pain in the Workplace: Ergonomics 317
i ncidence relati ve to location, occupation, or task can reveal areas of greater
priority for risk assessment and risk control .
Di scussion with the personnel i nvol ved often helps to reveal probl em areas.
Because of their i nti mate knowl edge of t he work si te and t he parti cular task
to be performed, the person at risk for muscul oskeletal i njury has i nformati on
on potential hazards that may not be apparent to management unti l after an
accident occurs .
Surveys and observations at the workplace can be carried out using si mpl e
checkli st approaches or by i mpl ementi ng more compl ex procedures . Checkl i sts
provide a focus for exami ni ng specifi c probl em areas and may deal wi th general
i ssues, or more specifical l y with the type of handl i ng task or the methods of
carrying out a task by the operators . Checkl i st s may provide a basi s for apprai s­
ing design of equipment, work space, or the envi ronment, and they may al so
be used to review the approach to supervi si on and work organi zati on.
Luopajarvi62 has pointed out that when workers study thei r work and work­
place acti vel y through the use of worker checkl i sts, they are abl e to learn the
guidel ines and, because of thei r personal i nvol vement, become more motivated.
They reflect the form of teachi ng that emphasizes acti ve participati on and l earn­
i ng by doing.
The approach to work anal ysi s has been wel l described by Luopajarvi .63
From the point of view of physi cal ri sk factors that may lead to LBP, anal yses
of work load that relate to observati on of work postures and movements are
most relevant. Direct vi sual observati ons, i n whi ch postures adopted by the
worker are recorded, or more advanced methods of i ncorporat i ng computerized
data recording and possi bl y i nvol vi ng photography and videotapi ng may be
used. U si ng work sampl ing methods, the postures may be observed at predeter­
mined i nterval s, so that a profi l e of the work demands may be gai ned. Anal yses
can reveal the frequency of stressful postures or movements that coul d l ead to
Special attention shoul d be paid to the availabi l i ty of adequate space for
the worker's operation, the avai labi l i ty of access for the worker to equipment
and storage areas, and the hei ght relationships between the worker and the
work surface that coul d influence the degree of stati c work posture.
Physiotherapi sts shoul d appl y thei r understandi ng of body mechanics to
anal yze t he dynamic posture of the person duri ng performance of thei r acti vi t y.
The stabi l i ty, symmetry, and qual i ty of al ignment of the posture assumed and
the way in whi ch movements undertaken fol low the principl es of good posture
could be appraised. Detai l s of whether the sequence of movements i n vari ous
body parts was appropriate; whether sati sfactory posture was mai ntained
throughout the acti vi ty; and the duration, range, di rection of, and resi stance
to i ndi vi dual specifc movements coul d be determined. These appraisal s, with
or without the quantitative measures gained duri ng work posture monitori ng,
would help to clarify the operati onal demands on the posture and on the muscu­
loskeletal system, l eadi ng to the development of sui tabl e modificati ons for the
preservation of postural l y safe worki ng condi ti ons.
31 8 Physical Therapy of the Low Back
Education and Training
Perhaps more subtl e than the effects of physical factors on the worker i s
the potential risk associated wi th l ack of knowledge in relation to work de­
mands. Here, the adequacy of communi cati on, education, and teachi ng i s im­
portant . Wi thi n the overall program of risk control, employee education has
been shown to reduce lost work ti me. 64 Trai ni ng workers to recognize and
avoid hazards during manual work, to improve coordination and handl i ng ski l l,
and to devel op an awareness of thei r capaci ty for handl i ng or tolerance for
postural stress wi thout LBP is of considerabl e val ue.65 However, poor tech­
ni ques establ i shed after years of habit are very diffi cul t to break.
Bul l ock66 has outl i ned the objecti ves and possibl e content of educational
programs that may be used for the various categories of worker in industry,
and has hi ghl ighted the need to use diferent approaches in educati ng work
supervi sors and managers, so that al l personnel understand their role in preven­
tion. Supervi sors have an added responsi bi l i ty for ri sk control and need more
guidance not onl y in methods of reduci ng work demands, but al so in gai ni ng
the cooperati on of workers .
Objecti ves for education depend on the program recipients and their type
of work, but in most cases, educational programs wi l l need to address the
concept of ergonomics, the mechani sm of injury relevant to the workplace, and
approaches to i njury prevention appl icabl e to i t . Participants would al so be
encouraged to questi on the sui tabi l i ty of thei r own environment . 66 For example,
when guidance about l ifti ng and handl ing was given, the person would be en­
couraged to ask whether the job i tsel f needed to be done, whether it might be
done more effecti vel y by a machine, whether the frequency of handl ing material
or the distance moved coul d be reduced, and whether the nature of the load
needed to be considered. Participants would al so be advi sed to query the num­
ber, capabi l i ti es, and l ifti ng habi ts of people i nvol ved i n handl i ng the task. It
i s important that workers are encouraged to examine the si tuation i tself before
deci di ng on how to cope wi th a part icular l ift or method of handl ing.
I t i s recommended that workers be shown those positions and movements
t hat could be potential l y dangerous in their own working envi ronment, and
that they be given the opportuni ty of devel opi ng sensi bl e and safe methods of
handling and l ifti ng under guidance. Most i mportantl y, those who are invol ved
in manual handl i ng and l ifting shoul d be taught how to use their own body
power in an appropriate way. Thi s i mpl i es t hat t here i s sufficient space in which
to move and room i n which to place the feet and to coordinate the total move­
ment of feet, body, arms, and load during a transfer. The need for suitable
exerci se and physical ftness must al so be emphasized i n education programs.
Special attenti on needs to be given to the education of new employees in
an i ndustry or of workers transferred to new manual tasks. Earl y trai ni ng in
appropriate approaches to movement and to safe working techniques could
help to control back i njury.
Low Back Pain in the Workplace: Ergonomics 319
It has been clai med that the most important way in whi ch ergonomic pri nci­
ples can be appl ied for prevention of i njury i s through design, i n whi ch al l factors
that affect the performance of a task are consi dered, to avoid its presenti ng a
risk to the health and safety of the worker. An important aspect in thi s regard
is the participation of workers in deci si on maki ng about thei r own worki ng
place. The physiotherapi st must be sure to i nvol ve the worker i n di scussi ons
about design needs and improvements that woul d contribute to safety as wel l
as effecti veness.
Developing a more effecti ve desi gn for a work operati on or work method
requires an appreciation for the purpose and requi rements of the task, as wel l
as the rules that govern desi gn of worki ng techni ques matched to i ndi vi dual
needs. I n particul ar, these principl es relate to the fol l owi ng poi nts: (I) the avoid­
ance of extreme positi ons of the trunk or shoul ders; ( 2) the use of fowi ng
dynami c movements rather t han rigi dl y hel d static positi ons; (3) the i mportance
of relaxing muscl es not requi red for a particul ar task; and (4) the need to di strib­
ute heavy work l oads over as many joints and muscl es as possi bl e.
The characteristi cs of the operators and the task are of prime consi deration
in planning a workplace. One of the most important aspects of the ergonomic
design approach i s the concern for careful specification of the worker-task
relationship, and thi s demands an appreci ati on for the variati ons in body si ze.
Basic requirements of si ze and shape shoul d be establ i shed so that the appropri­
ate height of equipment and workplace relative to the operator and the task
demands can be determi ned, so that posture wi l l be adequate and excessi ve
static muscl e work for the trunk muscl es avoided. Reference shoul d be made
to anthropometric studies that define the range of di mensi ons rel evant to the
particular worker population and to the demands of the task. Where a workplace
is to be used by many peopl e, adjustability over a range of di mensi ons relevant
to the users needs to be incorporated into the design.
The design of the work process is important al so. It shoul d ensure that
prolonged fi xed postures are avoided, that the worker is not required repeatedl y
to manipulate heavy objects, that movements demanded by the process do not
require rapid acceleration or decel eration, and that there is varied acti vi ty by
the one worker, combi ni ng gross and fi ne movements.
To avoid some of the muscul oskeletal problems that may be associ ated
with poor positi oni ng in sitti ng, attention has been gi ven to seat and tabl e design.
However, although many new ergonomical l y designed chai rs and work tabl es
have been provided for those i nvol ved i n office work, many of whi ch provide
sufficient adjustability to accommodate the large variety of sizes found i n work­
ers, thi s adjustabi l ity i s not always the case for many i ndustrial acti viti es .67 I n
circumstances where the worker i s i nvol ved i n prolonged acti vity requiring
some precision or dexterous work, a l arge amount of torso and upper extremity
mobility i s needed, while support for the lower torso and l ower limbs i s needed
to avoid muscl e fatigue and circul atory probl ems. This dual requi rement for
320 Physical Therapy of the Low Back
good postural support together wi th a capaci ty for adequate reach and vi sion,
makes the chair and workplace design more complex.67
The seated posture shoul d preferabl y be such that the person can work
wi th the trunk erect; wi th the lower back supported by the back rest; wi th the
el bows at the si de of the body; where practi cabl e, wi th forearms horizontal or
i ncl i ned sl ightl y downwards; wi thout excessive bending of the neck; and with­
out pressure being exerted on the underside of the thigh from t he front of the
seat , when the feet are flat on t he foor and knees at right angl es.
The seat desi gn shoul d al l ow for comfortabl e postural adj ustment and
shoul d also al l ow adequate space for the buttocks below the back rest , for the
legs if stretched out and for the knees and t highs under the bench or desk. The
l umbar support shoul d be firm and shoul d not restrict movement of the vertebral
col umn or the arms. Preferabl y, the back rest shoul d be clear of the sacrum
below and the shoul der blades above and shoul d conform to the forward curva­
t ure of t he l umbar regi on. Appropriate pelvic stabi l i zati on can al so be provided
by sl i ghtl y i ncl i ni ng the seat surface and ensuring adequate support for both
feet on the floor. Wi ll iams et ai ' s study68 has shown that with a l umbar support
to i ncrease lordosis in si tti ng, back pain can be significantl y reduced.
The relati onshi p of the seat to t he work bench or the desk is al so important
in terms of preventi ng back i njuries, as a too high bench requires the work to
be done wi th arms rai sed, whi ch i s l ikel y to lead to increased lumbar lordosi s,
whi l e a too l ow bench encourages stooped worki ng posi tions. I n terms of func­
tion, the recommended tabl e height i s related to the position of the el bow,
whi ch i n turn depends on the acti vi ty. Where forces need to be exerted through
the arms, the person shoul d be i n a standi ng posi ti on. The greater the force to
be appl i ed, the more elbow extensi on i s necessary. A worki ng surface should
be at the correct height for the task and, where the position i s likely to be
mai ntai ned for long periods , t he work surface should be sl oped toward the
worker to reduce the need for stooping.
Care needs to be taken about the design of acti vi ties carried out i n t he sitting
positi on. For exampl e, pedal use i n sitting, whether i n i ndustry, agri cul ture, or
general transport , can be a risk factor for LBP. Where the line of action of the
pedal is verti cal , the hip extensor muscl e acti vi ty needed for pedal depression
may demand excessi ve stabi l i zati on forces i n the trunk. As a result of extensive
experimentati on, Bul lock35 recommended t hat for the seated operator a pedal
path shoul d be at 45° to the horizontal , conti nuous with the "hip to foot on the
pedal " l i ne. Troup2 also recommended t hat the l i ne of action for pedal depres­
sion shoul d pass from the foot through the hip joint and t hat the back rest
shoul d fi rml y resi st t he tendency of the pel vi s to rotate.
Ergonomic guidel i nes have been provided for various aspects of work.
For exampl e, McCormick and Sanders69 and Chaffi n and Andersson7o have
provided gui del i nes for the maxi mum hol ding t i me of a static forceful exert ion,
whi l e Eastman Kodak Company ,71 McCormick and Sanders,69 and Grandjean72
have provided ergonomic gui del i nes for l ifting. The maxi mum al lowable l ifti ng
for parti cul ar situations has al so been recommended in a number of ergonomic
texts. 69.72 Appreciation for the relati onshi p between heavy manual work and
Low Back Pain in the Workplace: Ergonomics 321
LBP led to gui del i nes bei ng devel oped by the National I nst i t ute for Occupa­
tional Safety and Heal th ( NIOSH).73
Dul and Hi ldebrandt14 have revi ewed the gui del i nes outl i ned by a number
of aut hors concerned wi th ergonomi cs and the protecti on of the l ow back. Their
concl usi ons suggest that l i mitati ons exist i n the currentl y avai labl e gui del i nes
and recommend t he need for thei r val i dati on. These authors hi ghl ight the need
to advance knowledge about t he rel ationship between back load variabl es such
as EMG signal s of back muscl e acti vi ty, bi omechani cal torques or forces, i ntra­
abdominal pressure, i ntradi scal pressure, and the long-term i nci dence of LBP.
Dul and Hi l debrandt14 emphasize t he need for t he development of gui del i nes
that wi l l focus specifi cal l y on the preventi on of LBP.
Work should be designed t o be safe and al so to sui t t he populati on i nvol ved
i n it. Through i ntroducti on of ergonomic principl es , adverse handl i ng stresses
should be abl e to be mi ni mized, the cause of postural backache shoul d be
removed, and vi bratory stresses reduced. 65 The i mpli cati ons of certai n design
features for LBP and a knowledge of factors t hat can rel ieve back pai n are
necessary i n approachi ng the desi gn of techni ques for specified physi cal ac­
ti vi ti es.
Work Organization
Although these areas are vi tal l y i mportant to the physi cal wel l -bei ng of the
worker, other risks associated wi th the work acti vi ty must be appraised. These
i ncl ude the effects of work organi zation, such as the way work i s schedul ed,
the part icular demands on the worker of peaks of acti vi t y, and the t i me al lowed
for carrying out an acti vi ty. Si mi l arl y, attitudes to occupational heal th and
safety held throughout t he organi zati on, which may have an i nfl uence on the
provi si on of assi stance for heavy work tasks (i n terms of addi tional personnel
or appropriate equipment) and the arrangements made when workers are feel i ng
fatigued as a resul t of the heavy work demands, need to be apprai sed.
To be effecti ve, a preventi on strategy shoul d focus on the necessary organi ­
zational change that wi l l ensure probl em i dentificati on, documentati on, and
sol ution. A work environment must be created that wi l l ensure the participati on
of the workers , ensure the proper coordi nati on of al l preventi ve measures,
encourage both t he devel opment of ski l l s and the observance of pri nci pl es relat­
ing to safe manual handl i ng, and that wi l l i ntegrate i nj ury preventi on with work
procedures and work organi zati on. Indeed, some preventi on programs have
fai l ed in the past because they have had too excl usi ve an emphasi s on specifi c
l ifting technique i nstructions, too l i ttl e coordi nati on between functi ons affecti ng
occupational heal th and safety, and too l i ttl e i nvol vement by admi ni strators
and management in the preventi on process.
It has been acknowledged that there are mi ni mal as wel l as maxi mum
acceptabl e work loads i n keepi ng wi th safety of worker performance. Al though
emphasi s i s placed on loads t hat are beyond the capaci ty of the worker, i t must
322 Physical Therapy of the Low Back
al so be appreciated that too l i ttl e acti vi ty may be an eventual cause of LBP or
i njury. Magora20 found the smal l est i ncidence of back pain in a group of police­
men who underwent i ntensi ve physical trai ni ng before acceptance into t he po­
l i ce force and had to mai ntai n thi s fit ness t hroughout their career.
Preventi on programs shoul d incorporate instruction and guidance i n exer­
ci ses for a number of purposes. For example, in addition to providing education
in movement , programs can i ncl ude advice on methods of general and specific
relaxati on, exercises to provide acti ve work for muscl es maintained in a state
of stati c contracti on duri ng the work period, specific exercises to strengthen
muscl es concerned with l ifti ng and handl i ng, as well as exerci ses for general
physical fitness. Parti cular attenti on shoul d be paid to the correction of muscle
i mbalances and to the strengtheni ng of postural muscl es.
Although appropriate design of a job mi ni mi zes many of the factors leading
to LBP, much of the responsi bi l i ty for protection of t he back must rest with
the person. Back pai n and back i njuries do not onl y occur in the working envi­
ronment , but may al so occur out of worki ng hours . It is therefore important
t hat all peopl e recei ve some education not only in ergonomic design, but al so
in correct approaches to movement , so that they can appl y them to their leisure
acti vi ties , whether it be sport, gardeni ng, or si tti ng as a spectator.
Acti vi ty programs shoul d also emphasize the importance of changing both
posi ti on and work acti vi ty duri ng the worki ng hours. Opportunities for job
rotat ion t hat al l ow a change in muscl e acti vi ty should be provided by man­
agement .
Prevention of Reinjury
The effecti veness of a man-machi ne system depends on the way in which
t he mechani cal design of the equipment and workplace matches the capacities
of t he operator. Physi otherapists concerned wi th the care of the LBP patient
shoul d gai n an appreciati on of the various demands placed on the body by
work, home, or l ei sure acti vi ti es . Through assessment of relevant attributes
such as joint flexi bi l i t y, muscle strength and endurance, and functional capac­
i t y, they shoul d provide advice about modifications of the work envi ronment
or the work method that would acknowledge any l i mitations of the operator.
Work Site Visits
In consi deri ng how, when, and where t he worker with LBP may resume
work acti vi ti es , detai l s of the work environment and work activities, as pro­
vided by t he job anal ysi s , gi ve an understandi ng of the demands placed on t he
Thi s understandi ng of work demands provi des the basi s for development
of safe and effective methods for the operator' s performance of work, which
should control risk factors ari si ng from inadequate design of t he work process.
Low Back Pain in the Workplace: Ergonomics 323
Where vi si ts can be made by the physiotherapi st to the work si te for supervi si on
of work methods, t he opportuni ty for correction of operati ng diffi cul ti es or
potential l y damaging postures provides rei nforcement of good work methods
and encourages consul tation by the worker who may be uncertai n of the correct
approaches to prevention. Consultations with a person' s foreman may resul t
in the adaptation of the job for the period of rehabi l itati on.
Assessment of Functional Capacity
To protect the person agai nst exacerbati on of i njury, knowledge of thei r
functional capacity-possibl y changed by t he i ncidence of the LBP-needs to
be gained. I serhagen 75 has emphasized t hat for t he physiotherapi st to provide
correct and comprehensi ve information about a person' s functional capaci ty,
the appropriate design of functional capaci ty assessments i s essential . She
points out that the best functional capaci ty evaluators test the total person so
that al l aspects of movement and movement patterns can be observed. As
proposed by Janda76 in hi s rule of horizontal and vertical general i zation, I sern­
hagen77 out l i nes how a dysfunction in one area (e. g. , the knee) may affect
normal body mechanics and lead to i njuries i n another area (e. g. , LBP) .
The importance of eval uati ng body movements i n the context of work
and life si tuations is al so highlighted by I sernhagen. 77 Thus tests of strengt h,
endurance, range of movement , balance, and coordi nati on shoul d be related
to relevant work acti vi ty. This positive approach, in which the person's level
of capacity i s assessed, l eads to specificati ons of acti vities that may stil l be
achieved by the worker.
Functional capaci ty assessment can provide i nformati on that woul d be
helpful i n relation to advi si ng the LBP pati ent when return to work acti vi ty i s
safe and how much can be done. I sernhagen asserts that such an assessment
can act as a guide to pati ents to reach thei r ful l maxi mum potential . She explai ns
that later assessments provide the background to graduated return to work t hat
i s at a higher but safe functional l evel .
Watson et al78 suggest that the ergonomi st ' s approach to assessi ng human
abi l i ty in relation t o work demands i s appeal i ng i n t hat i t seeks to measure
interactions between the employee and the work place t hrough the structured,
systematic exami nati on of functional items relating to bot h. These authors de­
veloped the Acti vi ty Matchi ng Abi l ity System ( AMAS), whereby criteria are
identified agai nst which measures coul d be set . They highl ight the need for
instruments or assessment techni ques that measure differences i n functional
abi l i ty. Watson et aF8 believe t hat the abi l i ty assessment offers cl i ni cians a
potential instrument for identifying functional defi ci ts and for monitoring func­
tional outcome and change, because i t i mpl i es that treatments may be focused
on restoring or compensating for functions vi tal to work performance.
324 Physical Therapy of the Low Back
The Back School
The "back school " was establ i shed at Danderyd Hospital , Stockhol m,
in 1 970 as an efficient organization for LBP treatment . 79 The approach of the
back school was based on the knowledge that LBP i s often exacerbated by in­
creased mechanical strain and on sci entific research concered with etiology
of LBP, i ntradiscal pressure measurements, and epidemiology. Zachrisson­
ForsseW9 used ergonomics and education as the main el ements of this Swedi sh
program. The back school aims to enable pati ents to pl ay an acti ve part i n
improvi ng thei r working envi ronment i n order to reduce their back probl em,
to provide increased knowl edge, so gi vi ng better understanding and reducing
the risk of inappropriate therapy and reduci ng the demand for social , medical ,
and economic resources resul ting from avoidable back pai n. Through their con­
trol l ed study, Berqguist-Ul l man and Larsson 1 4 demonstrated a reduction in
days lost from work through use of the Swedi sh Back School , which emphasized
the application of ergonomic principl es.
Back school s have al so been developed in other countries, i ncl udi ng Can­
ada80 and the United States. 81 Hal l80 placed a greater emphasis on encouraging
a change of attitude in the person. Hi s study revealed signi fi cant subjective
improvement in 69 percent of those i nvol ved in the Canadian Back School .
The variabi l i ty in success of back school programs may reflect the great
di versi ty in physical , psychol ogical , and social factors infl uenci ng people with
LBP as well as the nature, i ntensi ty, and duration of pai n, and the l evel of
education of the participant s. Responsi veness of cl ients to particular ap­
proaches may depend on whether the components of the program match indi vid­
ual needs, can elicit a reaction, or can motivate the cl i ent. Most back schools
i ncl ude di scussi ons of anatomy and biomechanics, pain and i ts control , relaxa­
tion and stress management , and the i mportance of the person' s own attitude
and motivation to taking positive steps for risk avoidance. Many programs also
rel y on operant conditioning and major contributi ons from a psychologist . Back
school s usual l y offer a program of 4 to 10 cl asses,. at weekl y interval s. Rein­
forcement afer a period of appl yi ng pri ncipl es learned i n daily l ife i s essential .
It must be acknowl edged that the back school has not been seen as an
effecti ve approach to the management of LBP by al l who have used i t . For
exampl e, Lankhorst et ai ' s study82 of t he effect of t he Swedi sh Back School
in chroni c idiopathi c LBP subjects and Donchi n et aI ' s comparison8) of the
effecti veness of the back school with a second intervention program produced
resul ts t hat question the val ue of the program in reduci ng LBP episodes.
Ergonomi sts, recognizing the l i mitati ons of educational programs such as
these, have emphasized the i mportance of design changes in the workplace and
of modifications to work organizati on.
The application of t he principles of ergonomics to t he prevention of LBP
has consi derable val ue to the person at ri sk for i njury. Because of t he cost
benefits of control l i ng back pain and days lost from work, it also has great
value to producti vi ty in the industrial situation.
LolV Back Pain in the Workplace: Ergonomics 325
The physiotherapist can contribute to t hi s i mportant area i n a signifcant
way. Encouragement of the worker to be aware of hazards i n the workplace
or i n l ei sure acti vi ties and invol vement i n identification of risks can help to
produce the motivation for t he cl i ents' participation in control of thei r own
environment . Education in aspects of preventi on, in methods of movi ng the
body effecti vel y and safel y, and in mi ni mi zi ng or el i mi nati ng postural stress
through careful design, are essential components of a program. Al so i mportant
i s the need to modify attitudes to occupational heal t h and safet y, so t hat an
environment i s created i n whi ch protection of the person' s wel l -bei ng is seen
as vital by al l . It i s hoped t hat this positive attitude to preventi on will be taken
increasingly by those who are consulted by sufferers of LBP.
\ . Pedersen OF, Petersen R, Statfeldt ES: Back pain and isometric back muscl e
strength of workers i n a Dani sh factory. Scand J Rehabi l Med 7: 1 25 , 1 975
2. Troup J OG: Driver' s back pain and i ts preventi on: a revi ew of t he postural , vi bratory
and muscular factors, together with the problem of transmi tted road-shock. Appl
Ergon 9: 207, 1 978
3. Biering-Sorensen F: A prospecti ve study of low back pain i n a general population.
\ . Occurrence, recurrence and aetiology. Scand J Rehabil Med 1 5 : 7 1 , 1 983
4. Rowe ML: Low back pai n i n i ndustry. A position paper. J Occup Med I I : 1 6 1 , 1 969
5. Yu T, Roht LH, Wi se RA et al : Low back pain in i ndustry an old problem revi sited.
J Occup Med 26: 5 1 7, 1 984
6. Hershenson A: Cumul ati ve i njury: a national probl em. J Occup Med 2 1 : 674, 1 979
7. Troup J OG: Relation of lumbar spine disorders to heavy manual work and lifting.
Lancet I : 857, 1 965
8. Wickstrom G: Effect of work on degenerati ve back di sease. Scand J Work Environ
Heal th, Supp\ . I : I , 1 978
9. Strachan A: Back care i n industry. Physiotherapy 65: 249, 1 979
10. Bul lock MI : Ergonomi cs-a broad chal l enge for the physiotherapi st. p. 3. I n Bul lock
MI (ed): Ergonomi cs: The Physiotherapist i n t he Workpl ace. Churchi l l Li vi ngstone,
Edinburgh, 1 990
I I . Bul lock MI : The physiotherapi st ' s role in the control of i ndustrial i njuri es. Control
1 : 69, 1 974
1 2. Andersson GBJ : Epidemi ologic aspects on low back pai n in i ndustry. Spi ne 6: 53,
1 98 1
1 3 . Manning DP, Shannon HS : Sl i pping acci dents causi ng low back pai n i n a gearbox
factory. Spine 6: 70, 1 98 1
1 4. Berqguist -Ul l man M, Larsson V: Acute low back pai n i n i ndustry. Acta Orthop
Scand, supp\ . 1 70: I , 1 977
1 5 . Kumar S: Lifting and ergonomi cs. p. 1 83 . I n Bul lock MI ( ed) : Ergonomi cs: The
Physiotherapist i n the Workplace. Churchi l l Li vingstone, Edi nburgh, 1 990
1 6. Nachemson AL, Schul tz AB, Berkson MH: Mechanical propert i es of human l umbar
spine motion segments. I nfluences of age, sex, disc level and degenerati on. Spi ne
4: 2, 1 979
1 7. Nachemson A: The effect of forward learning on l umbar i ntra-discal pressure. Acta
Orthop Scand 35: 3 1 4, 1 965
326 Physical Therapy of the Low Back
1 8. Kel sey J L: An epi demi ological study of the rel ati onshi p between occupations and
acute herniated l umbar i ntervertebral di scs. Int J Epidemiol 4: 1 97, 1 975
1 9. Partridge RE, Anderson JAD, McCarthy JA et al : Rheumatic compl aints among
workers in Iran foundries . Ann Rheumatol Dis 27: 44 1 , 1 968
20. Magora A: I nvestigation of the relation between low back pain and occupation.
Scand J Rehabi l Med 7: 1 46, 1 975
2 1 . Grandjean E, Hunti ng W: Ergonomics of posture-review of various probl ems of
standing and si tti ng posture. Appl Ergon 8: 1 35, 1 977
22. Corlett EN, Manenci a I: The effect and measurement of working postures. Appl
Ergon 1 1 : 7, 1 980
23. Ki sner C, Colby LA: Therapeutic Exerci ses: Foundations and Techni ques. 2nd Ed.
FA Davi s, Phi l adel phi a, 1 990
24. Twomey LT, Taylor J R, Ol i ver MJ : Sustained fl exi on loading, rapid extension load­
ing of the l umbar spi ne, and the physical therapy of related i njuri es. Physiother
Pract 4: 1 29, 1 988
25. Twomey LT, Taylor J R: Fl exi on creep deformation and hysteresis i n the l umbar
vertebral col umn. Spi ne 7: 1 1 6, 1 982
26. Jayson NEV: The l umbar spine and back pai n. Pitman Medical Company, Edin­
burgh, 1 980
27. Grieco A: Si tti ng posture: an old problem and a new one. Ergonomics 29: 345, 1 986
28. Janda V: Movement patterns in the pel vi c and the hi p region with special reference
to pathogenesi s of vertebrogenic di sturbances. Habil itation Thesi s, Charles Univer­
sit y, Pragve, Czechosl ovaki a, 1 964
29. Pope MH, Frymoyer JW, Andersson GBJ : Occupational Low Back Pai n. Praeger,
New York, 1 984
30. Keegan J J : Al terations to the l umbar curve related to posture and seating. J Bone
Joi nt Surg 35A: 589, 1 953
3 1 . Bogduk N, Twomey L T: Cl i nical Anatomy of the Lumbar Spi ne. 2nd Ed. Longman,
Si ngapore, 1 991
3 2. Andersson BJG, Ortengrem R, Nachemson A et al : The si tti ng posture-an el ectro­
myographic and discometric study. Orthop Cl i n North Am 6: 1 05 , 1 975
33. Nachemson A: El ectromyographic studies on the vertebral portion of the psoas
muscl e. Acta Orthop Scand 37: 1 77, 1 966
34. Bul l ock MI : The determi nation of an opti mal pedal -operator rel ationshi p by the use
of stereo photo-gammetry. Biostereometrics: 290. Proceedings of the Symposi um
of Commissi on V I nternational Society for Photogrammetry. Washington, 1 974
35. Bul l ock MI : Muscul oskeletal di sorders i n the workplace. p. 93 I n Kumashiro M
and Megaw ED ( eds) : Towards Human Work. Taylor and Franci s, London, 1 991
36. Sul l i van A, McGi l l SM: Changes i n spine length during and after seated whole body
vi brati on. Spi ne 1 5: 1 257, 1 990
37. Magnussen M, Al mqvi st M, Broman H et al : Measurement of height loss during
whole body vi brati ons. J Spi nal Disorders 5: 1 98, 1 992
38. Kakosky T: Vibration di sease. Bai l l i ere' s Cl i n Rheumat 3: 25, 1 989
39. Troup JDG: Causes, prediction and prevention of back pain at work. Scand J Work
Envi ron Heal th 1 0: 41 9, 1 984
40. Stevens J: Low back pai n. Med Cl i n North Am 52: 55, 1 968
4 1 . Sypher F: Pain in t he back. A general theory. J I nt Col i Surg 33 : 7 1 8, 1 960
42. Janda V: Muscl es, cent ral nervous regul ati on and back probl ems. p. 27. In Kerr
1M ( ed) : The Neurobiologic Mechani sms i n Mani pul ati ve Therapy. 1 st Ed. Plenum
Press, New York and London, 1 978
Loll' Back Pain in the Workplace: Ergonomics 327
43 . Clayson SJ , Newman 1 M, Debevec OF et al : Eval uati on of mobi l i t y of hi p and
l umbar vertebrae i n normal young women. Arch Phys Med Rehabil 43: I , 1 962
44. Bul lock-Saxton J E: Changes i n muscl e function at hi p and low back fol l owing
chroni c ankl e sprai n. p. 57. Proceedi ngs of the 1 1 t h I nternational Congress World
Confederation for Physical Therapy. Vol . I . Chartered Society of Physiotherapy,
London, 1 991
45. Janda V, Vavrova M: Sensory Motor Sti mul ati on: A vi deo, presented by J E Bul l ock­
Saxton. Body Control Syst ems, Brisbane, 1 990
46. Roberts TOM: The mechanics of the upright posture. Physi otherapy 398: 404, 1 969
47. Janda V, Schmi d HJA: Muscl es as a pathogenic factor in back pai n. p. I . Proceed­
ings of the I FOMT Conference, Auckl and, New Zeal and, 1 980
48. Jul l GA, Janda V: Muscl es and motor control in low back pai n: assessment and
management. p. 1 97. I n Twomey LJ and Taylor J R ( eds) : Physi cal Therapy of t he
Low Back. Churchi l l Li vi ngstone, New York, 1 987
49. Mi chel i LJ : Low back pain i n the adol escent: di fferenti al diagnosi s. Am J Sports
Med 7: 362, 1 979
50. Stubbs OA: Trunk stressors i n constructi on and other i ndust rial workers. Spine 6:
83 , 1 981
5 1 . Schul tz AB, Andersson GBJ : Anal ysi s of loads on t he l umbar spine. Spi ne 6: 76,
1 981
52. Goldberg HM, Kohn HS, Dehn T et al : Diagnosis and management of low back
pai n. Occup Health Safet y, June 1 4, 1 980
53. Chaffin DB, Park KS: A longi tudi nal study of low back pain associated wi th occupa­
tional weight l i fi ng factors. Am I nd Hyg Assoc J 34: 5 1 3 , 1 973
54. Tichauer ER: The biomechanics of the arm-back aggregate under i ndustrial working
conditions. ASME 6: 28, 1 966
55. Schul tz AB, Haderspeck-Grip K, Si nkora G et al : Quanti tati ve studi es of t he fexi on­
relaxation phenomenon i n t he back muscl es. J Orthop Res 3: 1 89, 1 985
56. Kumar S: The study of spinal moti on duri ng wei ght l ifting. I r J Med Sci 1 43: 86,
1 974
57. Kumar S, Davi s PR: Spi nal loading i n static and dynamic postures: EMG an i ntraab­
dominal pressure study. Ergonomi cs 26: 9 1 3 , 1 983
58. Graveling RA, Si mpson GC, Si ms MT: Lift with your legs, not with your back: A
realistic di recti ve? In Brown 1 0, Gol dsmi th R, Coombes K et al (eds) : Ergonomics
International 85 , Proceedings of the 9th Congress of t he I nternational Ergonomics
Associ ation, Bournemout h, Engl and, Tayl or and Franci s, London, 1 985
59. Parnianpour M, Beiiani FJ , Pavl i di s L: Worker trai ni ng: t he fal l acy of a singl e,
correct l ift ing techni que. Ergonomics 30: 33 1 , 1 987
60. De Jong G: I ndependent l i ving as an anal ytic paradigm. Aust Rehabil Rev 6: 45 ,
1 982
61 . Gal vi n DE: Empl oyer-based di sabi l i ty management and rehabil itation programs.
Ann Rev Rehabi l 5: 23, 1 985
62. Luopajarvi T: Workers' education. Ergonomi cs 30: 305, 1 987
63 . Luopajarvi T: Ergonomic anal ysi s of workplace and postural load. p. 5 1 . In Bul lock
MI ( Ed) : Ergonomics: The Physiotherapist i n the Workpl ace. Churchi l l Li vingstone,
Edinburgh, 1 990
6. Tabor M: Reconstructing the scene: back i njury. Occup Heal th Safet y, February
1 6, 1 982
65 . Troup J OG: Bone mechanics of the vertebral col umn, i t s appl ication to preventi on
of back pain in the popul ation and to assessment of working capaci ty in pat i ents
with l umbar spi nal disabi l i t y. Physiotherapy 65 : 238, 1 979
328 Physical Therapy of the Low Back
66. Bul l ock MI : Heal th education in the workplace. p. 9. In I sernhagen S ( ed) : Work
I njury, Management and Preventi on. Aspen Publ i shers, Rockvi l l e, MD, 1 988
67. Chaffin D: Occupational biomechanics-a basis for workplace design to prevent
muscul oskel etal i njuri es. Ergonomics 30: 32 1 , 1 987
68. Wi l l iams MM, Hawley JA, McKenzie RA et al: A comparison of the effects of two
si tti ng postures on back and referred pai n. Spine 1 6: 1 1 85, 1 99 1
69. McCormick EJ , Sanders MS: Human Factors i n Engineering and Design. McGraw­
Hi l l I nternational , Auckl and, 1 984
70. Chaffi n DB, Andersson GBJ : Occupational Biomechani cs. Wi l ey, New York, 1 984
7 1 . Eastman Kodak Company: Ergonomics Design for People at Work. Vol . I . Lifeti me
Learning Publ i cati ons, Bel mont , 1 983
72. Grandjean E: Fi tti ng the Task to the Man. Taylor and Franci s, London, 1 980
73. National I nstitute of Occupational Safety and Health: Work Practices Guide for
Manual Lifting. Publication 8 1 - 1 22, Government Pri nti ng Offi ce, Washington, D. C. ,
1 98 1
74. Dul J , Hildebrandt VH: Ergonomi c gui del i nes for the prevention of low back pain
at the work pl ace. Ergonomics 30: 41 9, 1 987
75. Isernhagen J : Work i njury, management and preventi on. Aspen Publishers, Rock­
vi l l e, MD, 1 988
76. Janda V: I nt roduction to functional pathology of the motor system. p. 39. I n Bul lock
MI ( ed) : Proceedings of the VI I Commonwealth Games International Conference.
Vol . 3 . Uni versity of Queensl and Press, Brisbane, 1 982
77. Isernhagen J : The role of functional capacities assessment after rehabi l i tation.
p. 259. I n Bull ock MI ( ed) : Ergonomi cs: The Physiotherapist in the Workplace.
Churchi l l Li vingstone, Edi nburgh, 1 990
78. Watson H, Whal l ey S, McCl el land l : Matchi ng work demands to functional abi l i ty.
p. 23 1 . I n Bul l ock MI ( ed) : Ergonomi cs: The Physiotherapist i n the Workplace.
Churchi l l Li vi ngstone, Edinburgh, 1 990
79. Zachrisson-Forssell M: The Swedi sh back school . Physiotherapy 66: 1 1 2, 1 980
80. Hal l H: The Canadian back education uni ts. Physiotherapy 66: 1 1 5 , 1 980
8 1 . Mattmi l l er AW: The Cal i fornia back school . Physiotherapy 66: 1 1 8, 1 980
82. Lankhorst GJ , Van de Statd RJ , Vogelaar TW et al : The effect of the Swedish Back
School i n chronic idiopathic low back pain. Scand J Rehabil Med 1 5 : 1 4 1 , 1 983
83 . Donchi n M, Woolf 0, Kaplan L et al: Secondary prevention of low back pai n, a
cl i nical trial . Spi ne 1 5 : 1 3 1 7, 1 990
13 Lifting and Back Pain
M. Scott Sullivan
Epidemiologic studies have identified heavy manual labor as a major risk
factor for the development of low back pain. I According to Kelsey, lone major
risk factor for developing acute intervertebral disc prolapse is frequent lifting
of objects weighing 25 Ib or more. The risk is greater when objects are held
away from the body while lifting, and when twisting of the trunk occurs during
lifting.2 Further evidence supporting lifting as a cause of low back pain stems
from reports that sudden onset of low back pain is frequently preceded by an
event involving manual lifing.3
Physical therapists frequently administer treatment to individuals with
back injuries who return to jobs involving manual lifting. In addition to perform­
ing physical therapy assessments, physical therapists frequently evaluate pa­
tients' abilities to return to work through the use of functional capacities evalua­
tions. These evaluations usually involve assessments of patients' abilities to lift
manually. Physical therapy treatments are frequently complimented by patient
education5 and work-conditioning programs,6 which have been developed to
prepare the patient to safely tolerate the physical demands of the job. These
programs frequently spend a great deal of time teaching proper lifting technique.
In recent years physical therapists have also become more involved in
back-pain prevention programs involving pre-employment screening and back
schools for workers.7 The pre-employment screening programs often involve
analysis of the critical demands of the jobs and measurement of the worker 's
physical capabilities to perform these tasks. Pre-employment screening is fre­
quently used in industries with jobs involving manual lifing. Back school educa­
tion programs instruct participants in basic spinal structure and function, basic
mechanical principles, and proper body mechanics during lifting in an effort to
prevent back injuries. Because physical therapists evaluate lifting ability and
instruct patients and workers in lifting techniques, it is imperative for the physi­
cal therapist to have a working knowledge of the relationship between (1) lifting
330 Physical Therapy of the Low Back
and low back pain, ( 2) biomechanics of lifting, and ( 3) current theories of back
support mechanisms employed during lifting.
Physical therapists also need to analyze critically the reliability and validity
of lifting assessments performed in clinical and industrial settings, but explora­
tion of the topic is beyond the scope of this chapter.
The purposes of this chapter are to consider the current literature on the
biomechanics of lifting and the back support mechanisms, to discuss mecha­
nisms of back injuries while lifting, and to offer rationale for instructing patients
and industrial workers in lifting technique.
The forces acting on the spine during lifting may be estimated by the use
of free body diagrams and static equilibrium equations. The free body diagram
uses known quantities of certain variables to predict unknown quantities of
other variables. For example, the biomechanist can predict joint reaction forces
and muscle force requirements needed for a person to lift a given weight, by
using some of the following information: the body positions during the lift,
weight of the person and of the load lifted, accel