Review Article

Challenges in Treating
Acromioclavicular Separations:
Current Concepts

Abstract
Jay B. Cook, MD Injuries to the acromioclavicular joint constitute approximately 3.2% of
Kevin P. Krul, MD shoulder injuries. Although the overall goal of treatment continues to
be return to activity with a pain-free shoulder, the treatment of
acromioclavicular joint separations has been fraught with conflict
since the earliest reports in both ancient and modern literature.
Accurate diagnosis and classification are important to determine the
optimal treatment. Nonsurgical therapy remains the mainstay for
treatment of low- and most mid-grade injuries, although recent
Downloaded from http://journals.lww.com/jaaos by BhDMf5ePHKbH4TTImqenVHsCMLYwnrnSg1cI4Hjh89rS5cyzKw9EeGZj1yWJCLDX on 09/27/2018

biomechanical and biokinetic data might suggest that patients are

more affected than traditionally thought. High-grade injuries often
necessitate surgical intervention, although little consensus exists on
the timing or technique. New surgical techniques continue to evolve as
more biomechanical data emerge and kinematic understanding
improves. Challenges associated with management of this injury
abound from diagnosis to reconstruction.

A cromioclavicular joint (ACJ) sep-

From the Department of
Orthopaedics, the Winn Army
Community Hospital, Fort Stewart,
Georgia (Dr. Cook), and the
Department of Orthopaedics, the
Tripler Army Medical Center,
Honolulu, Hawaii (Dr. Krul).
Neither of the following authors nor
any immediate family member has
received anything of value from or has
stock or stock options held in a
commercial company or institution
related directly or indirectly to the
subject of this article: Dr. Cook and
Dr. Krul.
J Am Acad Orthop Surg 2018;26:
669-677
DOI: 10.5435/JAAOS-D-16-00776
Copyright 2018 by the American
Academy of Orthopaedic Surgeons.
arations are common injuries and
compose a sizeable portion of shoulder
injuries.1 Rates of ACJ injuries are
higher in men and contact athletes,
specifically those who participate in
rugby, wrestling, and hockey.2 Tradi-
tional management has largely been
nonsurgical for low-grade injuries and
surgical for high-grade injuries.1 Con-
troversy has persisted on the optimal
treatment of the mid-grade injury type.
Changes in surgical techniques and
biomechanical analysis have created
new areas of interest with regard to
treatment of this injury and introduced
questions on how, when, and why the
injury should be optimally treated.
Anatomy
The ACJ is a true synovial joint with
the articular cartilage, a capsule, and
several stabilizing ligaments. It is an
important stabilizer to the shoulder
girdle, providing support to the entire
suspensory complex. When a com-
plete ACJ separation occurs, this
suspensory complex is no longer able
to provide support to the shoulder
and the acromion, scapula, and upper
extremity sag creating the appear-
ance of the “shoulder separation.”3,4
The stability of the ACJ is twofold:
the stability of the acromion is bony
with contributions from the cor-
acoacromial ligament, whereas the
stability of the distal clavicle is largely
ligamentous. The acromioclavicular
(AC) ligaments provide stability
directly at the ACJ; they are the pri-
mary restraint to the anterior and
posterior motion of the distal clavicle
and consist of the anterior, posterior,
superior, and inferior ligaments.5
The coracoclavicular (CC) ligaments
connect the coracoid to the clavicle

October 1, 2018, Vol 26, No 19 669

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Challenges in Treating Acromioclavicular Separations
Figure 1
Anterior view of 3D CT
reconstruction of a right shoulder
showing the conoid ligament (red
trapezoid) has a much more broad
insertion than the trapezoid ligament
(blue trapezoid). The center of each
clavicular insertion is located at a
ratio of 0.24 and 0.17 (distance from
the lateral edge divided by the
clavicular length) for the conoid and
trapezoid, respectively.
and provide the main restraint to
superior translation of the clavicle
through the laterally based trapezoid
and the medial conoid ligaments.5
The locations of the conoid and
trapezoid ligament insertion sites
have been studied rather extensively.
The coracoid insertions sites are
located near the base of the coracoid,
with the conoid more medial and
posterior than the trapezoid.4 The
clavicular insertions are described in
terms of distances from the lateral
border of the clavicle and ratios of
this distance with respect to clavic-
ular length.6 The center of the trap-
ezoid attachment is roughly 26 mm
from the ACJ, with an average ratio
of approximately 0.17; the medial
border of the conoid insertion is
approximately 46 mm from the ACJ,
with an average ratio of 0.31 (note
measurements to the center of the
conoid insertion average 35 mm or a
ratio of 0.24)6 (Figure 1).
Physical Examination
ACJ separations are characterized by
pain, tenderness, and swelling at the
ACJ. Because of the mechanism of
injury, a complete evaluation of the
appendicular and axial skeleton is
warranted. Most patients will have
670
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
difficulty complying with a complete
shoulder examination. A 1 to 2 mL
injection of local analgesic may assist
with the examination. If there is gross
deformity, the reducibility of the ACJ
has been used to define the lesion.7 For
displaced injuries, because the injury
represents depression of the scapula
and not true elevation of the clavicle,
the reduction maneuver must include
a superior directed force on the
scapula, shoulder, or elbow and an
inferiorly directed force on the clavi-
cle. An irreducible ACJ represents
interposition of fascia or cartilage and
has been described as an indication
for surgery.7
Radiographic Evaluation
Standard radiographs with or with-
out stress are usually sufficient to make
the radiographic diagnosis. Typically,
the radiographic examination begins
with a shoulder series (ie, AP, scapular
Y, and axillary views) but may also
include weighted, cross-arm AP, or
Zanca views. The cranial tilt of a Zanca
view allows better visualization of the
ACJ. An axillary view can demonstrate
posterior displacement of the distal
clavicle. Weighted views can differen-
tiate grades of injury, and a cross-arm
(adducted) view may demonstrate
dynamic instability if the clavicle rides
over the acromion. Imaging of bilateral
shoulders is required for accurate clas-
sification purposes (discussed later).
MRI can provide further informa-
tion, but it is not routinely necessary
for making the diagnosis. It can assist
with identifying associated injuries
such as labral tears or rotator cuff
injuries requiring treatment, which
have been cited to be present in up to
30% of injuries.8 Typically, the
authors order MRI to assist with
surgical planning if the ACJ injury
meets indications for surgery or if
there is suspicion for concomitant
injuries based on the examination in
lower-grade ACJ injuries.
Journal of the American Academy of Orthopaedic Surgeons
Classification
The classification of ACJ separations
is based on the type or amount of
displacement associated with the
injury. Rockwood adapted the origi-
nal work of Tossy into the current
classification that is used by most
surgeons.3,9
This classification uses AP radio-
graphs of the shoulder. The amount
of displacement is traditionally mea-
sured using the CC distance defined
as the distance from the superior
aspect of the coracoid vertically to the
clavicle. Once the measurement on
the injured side is complete, it is
compared with the contralateral side
to determine displacement. Dis-
placement has also been measured
from the superior-medial border of
the acromion to the superior-lateral
aspect of the clavicle to determine the
percentage of displacement (Figure 2).
The anatomic correlations to the
radiographic findings are described
later and summarized in Table 1.
Type I injury: characterized by
injury to the AC ligaments without
complete tear. The CC ligaments are
uninjured. No deformity exists.
Type II injury: represents an injury
to the AC ligaments with complete
disruption; the CC ligaments are
partially injured without a complete
tear. Complete disruption of the AC
ligaments leads to horizontal insta-
bility and frequently some slight
radiographic asymmetry.
Type III injury: complete tears of
both the CC and AC ligaments exist.
Because of the horizontal and vertical
instability, there will be gross defor-
mity at the joint. The deltoid origin and
trapezius are completely detached.
Alternatively, the force of injury is
transmitted to the coracoid and results
in a fracture at the insertion of the CC
ligaments.
Type IV injury: posterior separa-
tion of the ACJ exists. The force at the
lateral edge of the acromion results in

the anterior displacement of the ac-
romion and the posterior displace-
ment of the clavicle. The result is
complete tears of the CC and AC
ligaments and notable deformity of
the posteriorly displaced clavicle.
Type V injury: a more severe type
III injury. The CC and AC ligaments
are completely detached, as are the
fibers of the trapezius and anterior
deltoid attachments, with increased
displacement of the joint versus a
type III.
Type VI injury: an inferior separa-
tion of the clavicle. The clavicle is
displaced either below the acromion
or below the coracoid. The AC liga-
ments are disrupted. The CC liga-
ments, trapezial fascia, and anterior
deltoid are all likely to be injured.
Although there is some controversy
with regard to clinical validation, the
Rockwood classification was intro-
duced to provide an anatomic de-
scription that guides treatment. The
randomized controlled trials that
have defined modern nonsurgical
treatment used the older classifica-
tions of either Tossy or Allman.10,11
In contrast to the older classifica-
tions, the Rockwood classification
further delineates the mid- to high-
grade injuries and recommends sur-
gical intervention in grades IV, V,
and VI. In the previous trials, a Tossy
or Allman grade III would include
Rockwood types III and V, making
the applicability of these classifica-
tion systems questionable in high-
grade injuries.9 Furthermore, type III
injuries can include both stable and
unstable injuries. Although these
injuries are traditionally treated
nonsurgically, some patients do
poorly with nonsurgical manage-
ment and require surgical interven-
tion. A statement by the International
Society of Arthroscopy, Knee Surgery
and Orthopaedic Sports Medicine
suggests the need for a IIIa and IIIb
modification; the former refers to an
inherently stable injury likely to be
successfully treated with nonsurgical
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Jay B. Cook, MD and Kevin P. Krul, MD
Figure 2
Bilateral standing AC joint non-weighted view. The Rockwood classification is
traditionally based on the distance between the coracoid and the clavicle and
compared with the uninjured side. However, frequently the distance from the
acromion to the clavicle is measured and compared with the contralateral shoulder.
Table 1
Rockwood Classification
Type AC Ligaments CC Ligaments Displacement on Radiograph
I Partially torn Intact None
II Disrupted Intact Minimal
III Disrupted Disrupted ,100% increase in the CC distance
IV Disrupted Disrupted Posterior through the trapezius
V Disrupted Disrupted .100% increase in the CC distance
VI Disrupted Disrupted Subacromial or subcoracoid
AC = acromioclavicular, CC = coracoclavicular
management, and the latter refers to and went on to surgery. Mouhsine
an unstable injury likely to result in et al13 also examined the outcome of
continued pain, instability, and nonsurgical treatment for low-grade
scapular dyskinesia if treated non- injuries and reported that half
surgically.12 Currently, the most became asymptomatic with time.
accepted classification is the Rock-
wood classification. However, the
authors agree that stable and unstable
Type III Injuries
type III injuries need to be The optimal treatment of type III
differentiated. injuries has long been the subject of
debate. The review by Johansen
et al11 thoroughly discusses the sev-
Nonsurgical Treatment eral prospective randomized articles,
performed nearly 20 to 40 years ago,
Type I and II Injuries that have shown no benefit to surgical
Nonsurgical treatment has been the treatment because the nonsurgical
mainstay of treatment for type I and cohorts had fewer complications and
type II injuries.7 Evidence to support faster return to work. Systematic
this treatment is limited, but several reviews, the most recent by Beitzel
studies have demonstrated its effi- et al14 in 2013, have not shown sig-
cacy. Pallis et al2 reported on 145 nificant functional outcome benefit
college athletes with low-grade (type with surgical treatment compared
I and type II) injuries, and only 6 with nonsurgical treatment. Despite
patients failed nonsurgical treatment more anatomic outcomes, surgical
671
Challenges in Treating Acromioclavicular Separations
treatment results in higher complica-
tion rates, slower return to work, and
equivalent range of motion.14 Con-
sequently, type III injuries have
largely been treated nonsurgically the
past few decades. However, more
recent studies have described altered
shoulder mechanics and scapular
dyskinesia with AC separations.15,16
Type V Injuries
Type V injuries are treated surgically,
although little evidence exists to
support this treatment. The only level
I or level II published data on non–
surgically treated severe Tossy type
III separations come from Bannis-
ter’s10 randomized controlled trial.
These patients had 2 cm of dis-
placement, and, in the authors’
opinion, these would be consistent
with a Rockwood type V. In the
nonsurgical group, four of five
patients had fair or poor outcomes.10
A more recent study examined non-
surgical management of type V
injuries demonstrating that most
patients will return to work; how-
ever, those with .2 cm of displace-
ment of the clavicle above the
acromion were more likely to fail
nonsurgical therapy.17 One other
review examined type V injuries and
noted that 77% of patients were able
to return to work with nonsurgical
management, half being manual
laborers, despite modest ASES
(American Shoulder and Elbow
Society) and DASH (Disabilities of
the Arm, Shoulder, and Hand)
scores.18
Author’s Preferred
Nonsurgical Treatment
Protocol
Multiple methods of casting and sling
wear attempting to externally hold an
AC reduction have been used.7
Notably, patient compliance is low,
and no method has been proven to
be more effective than a simple sling
672
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
or shoulder immobilizer and
activity modification.7,14 Patients
undergoing nonsurgical treatment
are removed from sport until
symptoms resolve.
Currently, the authors do not
attempt a reduction when undergo-
ing nonsurgical treatment. Patients
are treated with a sling for 2 to 3
weeks until much of the acute pain
resolved, followed by therapy and
early range of motion. For type I and
II injuries, surgery is considered if
patients remain symptomatic or
unable to return to sport after 3 to
6 months of therapy and rehabilita-
tion. Patients with a type III injury or
a type V injury with ,2 cm dis-
placement and without medial-
lateral instability with the clavicle
not overriding acromion are brought
back after 3 to 4 weeks from injury
for repeat evaluation. Those who
report significant improvement in
pain and motion, as well as minimal
scapular dyskinesia, are counseled to
continue nonsurgical management. If
the patients are noted to have marked
scapular dysfunction and minimal
improvement in pain, stability of the
joint is evaluated clinically and
radiographically, and surgical inter-
vention may be offered at this point.
Other considerations with regard to
work demands and the ability to
comply with postoperative restric-
tions are also taken into account in
the treatment algorithm.19
Surgical Treatment
More than 150 techniques for surgi-
cal treatment of AC injuries have
been described.14 These techniques
have generally fallen into several
categories: AC fixation, CC fixation,
or ligament reconstruction. On
principle, acute injuries with the
capacity to heal can do well with
techniques that hold the reduction
and allow for healing. Typically,
chronic injuries require some form of
Journal of the American Academy of Orthopaedic Surgeons
biologic augmentation, most com-
monly with the use of a tendon graft.
In the low-grade, symptomatic type I
and II injuries that have failed non-
surgical treatment, an open or
arthroscopic distal clavicle excision
may be appropriate to provide pain
relief.
AC fixation has historically involved
plates, screws, or wires across the ACJ
in the acute setting.7 Most commonly
today, this consists of hook plates,
which are secured to the clavicle with
screws and span the ACJ maintaining
reduction by hooking under the ac-
romion. The plates are frequently
removed approximately 3 months
postoperatively. The hook plate has
previously been demonstrated to result
in higher outcome scores compared
with nonsurgical treatment, however
lower scores compared with a modi-
fied Weaver-Dunn technique.20 A
more recent randomized control trial
demonstrated equally good functional
outcome scores between hook plate
fixation and nonsurgical treatment.21
CC fixation was traditionally per-
formed with a screw from the clavicle
to the coracoid; however, modern
techniques have used suspensory
fixation for CC fixation in place of
the rigid screw, particularly with
acute injuries.22 One study com-
pared the Bosworth screw with a
suture button for treatment of acute
injuries and noted no difference
in maintenance of reduction, but
increased patient satisfaction with
the suture button.23 Fixation with
one or two suture buttons as an
acute repair technique has been
shown to have high biomechanical
stability.24 This technique is optimal
for repair of acutely torn ligaments,
providing stabilization to allow the
native ligaments to heal.25
Modern techniques have also
moved toward anatomic reconstruc-
tions using tendon grafts with or
without suspensory devices used in
conjunction with the graft, particularly
for more chronic injuries.19,26-29

Multiple biomechanical studies have
shown anatomic CC ligament recon-
struction to have biomechanical
properties similar to the native joint,
significantly better than older tech-
niques.30,31 The anatomic recon-
struction was first described by
Mazzocca et al32 over a decade ago.
Since this description, the technique
has been modified with regard to
graft type, graft configuration, graft
placement, fixation method, aug-
mentation, and incorporation of
graft limbs into the AC liga-
ments.1,14,19,28,31-34 As of yet, there
has been no establishment of a single
modification of the anatomic tech-
nique that is superior to the rest,
although anatomic reconstruction
seems to be preferable to non-
anatomic reconstruction, given
available data. 14
Other Controversies in
Surgical Treatments
Timing
Early versus delayed treatment con-
tinues to be a subject of debate. Few
articles have compared early surgery
versus delayed treatment, but the
existing data trend toward improved
patient satisfaction and radiographic
outcomes with early treatment.35-37
Nevertheless, there is some difficulty
drawing generalized conclusions
because the available data include
low-level studies with different defi-
nitions on “acute.”14 Conceptually,
acute surgical treatment affords the
ability to stabilize the ACJ and
allows the native ligaments to heal,
ideal for techniques not including
biologics.25 However, the recom-
mendation cannot be made to treat
all potentially surgical ACJ separa-
tions with an acute repair, given the
available data on surgical versus
nonsurgical treatments. Such an
algorithm would result in a potentially
high number of patients undergoing
surgery who would otherwise do well
nonsurgically.
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
The counterargument is that de-
layed reconstruction better selects
those patients who require surgery as
they have failed nonsurgical man-
agement. Despite the aforementioned
literature suggesting that acute
repairs have better outcomes, delayed
reconstructions improve outcomes
from preoperative levels without
subjecting patients to potentially
unnecessary surgery (see Outcomes
section). Delaying the reconstruction
maximizes the number of patients
who can be successfully treated
nonsurgically.
If patients could be identified who
would fail nonsurgical management,
rehabilitation would be expedited if
early surgery were performed. Sev-
eral factors have been identified that
might suggest which patients will
likely fail nonsurgical management.
The first is highly unstable injuries,
types IV and VI specifically.7 The
second is type V injuries, but spe-
cifically, if there is .2 cm displace-
ment at the ACJ as some type V
injuries can do well with nonsurgical
treatment.17,18 Finally, dynamically
unstable type III injuries will often fail
nonsurgical treatment and merit
consideration for surgery as recom-
mended by International Society of
Arthroscopy, Knee Surgery and
Orthopaedic Sports; these can be
identified by a cross-arm AP radio-
graph noting the clavicle overriding
the acromion or axillary radiographs
with the arm abducted and adducted
to evaluate for posterior displace-
ment.12,19 However, these factors
have not yet been prospectively
validated.
Distal Clavicle Excision
Reduction of the ACJ can usually be
achieved without much difficulty in
the more acute setting. However, a
distal clavicle excision is sometimes
required to obtain reduction of chron-
ically dislocated ACJs. Although it has
been argued to retain the distal clavicle
Jay B. Cook, MD and Kevin P. Krul, MD
for bony stability, two recent bio-
mechanical studies have shown that
there does not appear to be a significant
increase of strain on the graft with a
small excision of the distal clavicle.
Beitzel et al38 showed that with an
intact posterior-superior AC capsule, a
5-mm resection added minimal ante-
rior or posterior translation, but this
increased with a 10-mm resection.
The authors recommended main-
taining the posterior-superior cap-
sule or considering AC capsule
reconstruction if needed in the set-
ting of an AC separation.38 Beaver
et al39 demonstrated no increased
anterior to posterior or superior to
inferior translation with bio-
mechanical testing of a CC ligament
reconstruction with and without
7 mm of distal clavicle resection.
Two studies have examined the
presence or absence of distal clavicle
excision as it related to early radio-
graphic failure, but did not show any
statistically significant difference
between the two groups with regard to
loss of reduction postoperatively.36,40
No other study has directly compared
reconstructions with and without a
distal clavicle excision with regard to
functional outcome scores or revision
rates. Currently, the authors’ pre-
ferred approach is to attempt a closed
reduction visualized with fluoroscopy
after the patient is placed under
anesthesia prior to prepping the
shoulder. If the ACJ is unable to be
reduced fully, the incision will be
extended to allow for removal of the
meniscal homologue or any inter-
posed tissue and then proceed with a
5-mm distal clavicle resection with an
imbrication of the capsule and pos-
sible augmentation with a limb from
the graft.
Graft Type
Most techniques describe the use of
allograft as a tendon source, but auto-
graft has also been described.19,26-29,36
One study looked at allograft versus
673
Challenges in Treating Acromioclavicular Separations
Figure 3
Superior view of 3D CT
reconstruction of a right shoulder
showing an example of the planned
tunnel location of the conoid (red)
and trapezoid (blue) limbs based on
anatomic ratios.
autograft as a risk factor for early
radiographic failure, noting a higher
failure rate in the allograft group
(37.5% versus 16.7%) but was
underpowered to show significance
with regard to radiographic fail-
ure.36 There are no studies designed
to directly comparing allograft and
autograft for use in biologic aug-
mentation in CC ligament recon-
struction, and it is unknown at this
time what role graft type plays in
failure or loss of reduction in CC
ligament reconstruction.
Author’s Preferred Surgical
Technique
As stated previously, most patients
are given a trial of nonsurgical treat-
ment for Rockwood types I, II, III,
and V with ,2 cm of ACJ displace-
ment. Early surgical indications
include Rockwood types IV, VI, and
V with .2 cm of displacement or
with medial-lateral instability re-
sulting in the clavicle overriding the
acromion in a high demand patient.
Other surgical indications include
open injuries, or injuries with
neurologic deficits, or low-grade
injuries (types I and II) that have
failed nonsurgical therapy for 3 to
6 months. Patients with types III or
674
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Figure 4
Preoperative and postoperative radiographs of a coracoclavicular ligament
reconstruction performed for a chronic ACJ separation. Placement of both
buttons on the clavicle allows vertical compressive force without placing a hole in
the coracoid. Both the graft and the suture are slung under the coracoid together.
Table 2
Complications
Complication Incidence (Source)
Loss of reduction (early or late) Up to 53%36,40
Clavicle fracture Up to 18%34
Coracoid fracture Up to 20%34,a
Infection Up to 6%28,34
Total complication rates Up to 53%33,34,36,40,41
a
In patients with the coracoid tunnel.
V injuries who have failed non- same technique is used for both sub-
surgical treatment as mentioned acute and chronic injuries.
above are also indicated for surgi- The clavicle is preoperatively tem-
cal intervention. plated to place the conoid tunnel at
Patients with low-grade injuries 20% to 25% of the clavicular length
and persistent pain can often be from the distal clavicle, and the trape-
treated with a simple distal clavicle zoid tunnel is placed 1.5 to 2 cm lateral
excision. Those with higher-grade to this (near the anatomic insertion at
injuries, symptomatic instability, or 17% of clavicular length) (Figure 3).
significant deformity who have failed Both the tendon graft and the Dog
nonsurgical management undergo Bone (Arthrex) are shuttled from the
arthroscopically assisted, anatomic medial tunnel, under the coracoid, up
reconstruction using biologic and the lateral tunnel. The ACJ is reduced
synthetic fixation. As surgical inter- and the suture button is secured, then
vention is rarely undertaken within the graft is secured on top of the
the first 2 to 3 weeks from injury, the clavicle. The remaining limbs may be
Journal of the American Academy of Orthopaedic Surgeons
 Jay B. Cook, MD and Kevin P. Krul, MD

Table 3
Outcomes of Anatomic Reconstructions
Study Patients Follow-up Subjective Satisfaction Objective Scores

Tauber et al31 12 37 mo (average) Average 9.6 of 10 ASES 96

— — — Constant 93
Carofino and Mazzocca28 17 21 mo (average) SANE 94.4 ASES 92
— — — SST 11.8
— — — Constant 94.7
Martetschlager et al33 43a 2.4 yr (average) Average 9 of 10 ASES 91
— — SANE 89 QuickDASH 7
Millet et al26 20a 3.5 yr (average) Average 9 of 10 ASES 94
— — SANE 89 QuickDASH 5.6

ASES = American Shoulder and Elbow Society, DASH = Disabilities of the Arm, Shoulder, and Hand, SANE = Single Assessment Numerical
Evaluation, SST = Simple Shoulder Test, VAS = Visual Analog Scale
a
Only reported on patients without a complication.

brought over laterally to augment the Table 4

AC ligament if desired (Figure 4).
Outcomes of Acute Repairs Using a Suture Button Construct
Postoperative rehabilitation in-
cludes strict sling wear for 6 weeks. Study Patients Follow-up Outcome Scores
Motion is gradually increased after Glanzmann et al45 19 24 mo (min) Constant 90.2
cessation of sling wear with a goal of — — SST 11.5
full motion at 3 months. Strengthen-
Venjakob et al22 23 58 mo (average) Constant 91.5
ing begins at this point, and patients
— — VAS 0.3
are allowed back to contact sports or
Torkaman et al46 28 16 mo (average) Constant 89.4
combat missions in the military at 6
— — DASH 1.4
months.
Loriaut et al47 39 42 mo (average) Constant 94.7
— — QuickDASH 1.7
— — VAS 0.5
Complications Rosslenbroich et al48 83 39 mo (average) Constant 94.7
Given the vast number of techniques DASH = Disabilities of the Arm, Shoulder, and Hand, SST = Simple Shoulder Test, VAS = Visual
described for ACJ separations, the Analog Scale
array of complications accompany-
ing them has also been quite large.
This discussion will focus on those plating tunnel position based on the the graft around the clavicle in lieu
that are specific to recent techniques clavicular length, placing the tunnels of placing bone tunnels. Coracoid
and are summarized in Table 2. at the center of the native insertions. fractures can also occur, particu-
Other, more infrequent complica- However, despite rates of radio- larly in techniques involving a cor-
tions can include adhesive capsulitis, graphic “failure” reaching up to 53% acoid tunnel.33 The graft or suture
neuropathy, distal clavicle hyper- in these military cohorts, over 82% of may be placed under the coracoid in
trophy, and ACJ pain. patients were functionally able to re- a sling fashion to avoid creating a
A frequently reported complication turn to their military duties.36,40 coracoid tunnel. The size of the
is partial or complete loss of reduc- Clavicle fractures have been a re- tunnel and orientation can affect
tion.41 Two studies have examined the ported postoperative complica- the biomechanical properties of the
causes of failure and noted tunnel tion.26,33 Biomechanical studies coracoid and risk of fracture; cora-
position, particularly medialization of have shown that the holes drilled for coid tunnels should be as small as
the conoid tunnel, to be a statistically bone tunnels in the clavicle render possible and placed center-center
significant risk factor.36,40 Both stud- the bone more susceptible to frac- from superior to inferior to mini-
ies recommend preoperatively tem- ture.42 One alternative is to wrap mize the risk of fracture.43,44

October 1, 2018, Vol 26, No 19 675

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Challenges in Treating Acromioclavicular Separations
The overall rate of fracture, either
coracoid or clavicle, has been cal-
culated at 5.3% in a recent sys-
tematic review.41
Outcomes
Patient outcomes of anatomic recon-
structions are frequently reported as
highly successful (Table 3). Many of
these reconstructions are chronic
injuries. However, even in a military
population, return to full duty in
service has been reported in over
82%.36,40 Acute repairs have also
reported high rates of maintenance
of reduction and good patient out-
comes (Table 4).41
As mentioned earlier, many stud-
ies have examined acute recon-
struction or repair for type III
injuries versus nonsurgical treat-
ment and have not found significant
differences in outcome mea-
sures.11,14 A more recent study,
including types III, IV, and V,
demonstrated again no significant
difference in outcome scores at 2
years between surgical intervention
with a hook plate and nonsurgical
management of acute injuries.21
This study, though well performed,
did not describe activity levels of
the patients. All patients did return
to work; nonsurgical patients ten-
ded to have improved outcomes
earlier, decreased chance of com-
plications, yet had an increased
chance of being dissatisfied with
cosmesis.21
Nonsurgical management has been
shown to result in altered shoulder
mechanics and scapular dyskinesia,
decreased bench press strength, and
decreased intensity in sport.15,16,49
Still, there remains a paucity of data
evaluating the biomechanics of shoul-
der motion and strength in a com-
parison study between a surgical and
nonsurgical shoulder. Yet, although
scapular dyskinesia is thought to
contribute to failure of nonsurgical
676
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
management, it can persist even after
surgical reconstruction.50 This fact
again raises the question of whether
early treatment with modern recon-
struction techniques can lead to
better outcomes than nonsurgical
treatment.
Summary
ACJ separations are common in the
athletic population, yet continue to
be challenging to treat. Most ACJ
injuries are low or middle grade and
should be treated nonsurgically.
Surgical intervention is traditionally
reserved only for those who fail
nonsurgical treatment or those who
have the most severe of injuries.
However, evidence of good or better
results with acute surgical interven-
tion may lead to a change in this
treatment algorithm. Currently, data
have failed to show improved long-
term outcomes with acute repair
versus initial nonsurgical treatment.
Finally, at present, there is no single
superior surgical technique with
quality long-term follow-up.
References
References printed in bold type are
those published within the past 5
years.
1. Mazzocca AD, Arciero RA, Bicos J:
Evaluation and treatment of
acromioclavicular joint injuries. Am J
Sports Med 2007;35:316-329.
2. Pallis M, Cameron KL, Svoboda SJ, Owens
BD: Epidemiology of acromioclavicular
joint injury in young athletes. Am J Sports
Med 2012;40:2072-2077.
3. Tossy JD, Mead NC, Sigmond HM:
Acromioclavicular separations: Useful and
practical classification for treatment. Clin
Orthop Relat Res 1963;28:111-119.
4. Salzmann GM, Paul J, Sandmann GH,
Imhoff AB, Schottle PB: The coracoidal
insertion of the coracoclavicular ligaments:
An anatomic study. Am J Sports Med 2008;
36:2392-2397.
5. Fukuda K, Craig EV, An KN, Cofield RH,
Chao EY: Biomechanical study of the
ligamentous system of the
Journal of the American Academy of Orthopaedic Surgeons
acromioclavicular joint. J Bone Joint Surg
Am 1986;68:434-440.
6. Rios CG, Arciero RA, Mazzocca AD:
Anatomy of the clavicle and coracoid
process for reconstruction of the
coracoclavicular ligaments. Am J Sports
Med 2007;35:811-817.
7. Rockwood CA, Green DP, Bucholz RW:
Rockwood and Green’s Fractures in Adults,
ed 7. Philadelphia, PA, Wolters Kluwer
Health/Lippincott Williams & Wilkins,
2010.
8. Arrigoni P, Brady PC, Zottarelli L, et al:
Associated lesions requiring additional
surgical treatment in grade 3
acromioclavicular joint dislocations.
Arthroscopy 2014;30:6-10.
9. Williams G, Nguyen V, Rockwood C:
Classification and radiographic analysis of
acromioclavicular dislocations. Appl
Radiol 1989;18:29-34.
10. Bannister GC, Wallace WA, Stableforth
PG, Hutson MA: The management of acute
acromioclavicular dislocation: A
randomised prospective controlled trial. J
Bone Joint Surg Br 1989;71:848-850.
11. Johansen JA, Grutter PW, McFarland EG,
Petersen SA: Acromioclavicular joint
injuries: Indications for treatment and
treatment options. J Shoulder Elbow Surg
2011;20(suppl 2):S70-S82.
12. Beitzel K, Mazzocca AD, Bak K, et al:
ISAKOS upper extremity committee
consensus statement on the need for
diversification of the Rockwood
classification for acromioclavicular
joint injuries. Arthroscopy 2014;30:
271-278.
13. Mouhsine E, Garofalo R, Crevoisier X,
Farron A: Grade I and II acromioclavicular
dislocations: Results of conservative
treatment. J Shoulder Elbow Surg 2003;12:
599-602.
14. Beitzel K, Cote MP, Apostolakos J, et al:
Current concepts in the treatment of
acromioclavicular joint dislocations.
Arthroscopy 2013;29:387-397.
15. Oki S, Matsumura N, Iwamoto W, et al:
Acromioclavicular joint ligamentous
system contributing to clavicular strut
function: A cadaveric study. J Shoulder
Elbow Surg 2013;22:1433-1439.
16. Gumina S, Carbone S, Postacchini F:
Scapular dyskinesis and SICK scapula
syndrome in patients with chronic type III
acromioclavicular dislocation. Arthroscopy
2009;25:40-45.
17. Krul KPC, Cook JB, Cage JM, Rowles DJ,
Bottoni CR, Tokish JM: The displacement
of the clavicle is a better predictor of
surgical intervention in the non-operatively
treated acromioclavicular dislocation than
the increase in coracoclavicular distance.
Orthop J Sports Med 2015;3(7 suppl 2):
2325967115S00077.

18. Dunphy TR, Damodar D, Heckmann ND,
Sivasundaram L, Omid R, Hatch GF III:
Functional outcomes of type V
acromioclavicular injuries with nonsurgical
treatment. J Am Acad Orthop Surg 2016;
24:728-734.
19. Cook JB, Tokish JM: Surgical management
of acromioclavicular dislocations. Clin
Sports Med 2014;33:721-737.
20. Gstettner C, Tauber M, Hitzl W, Resch H:
Rockwood type III acromioclavicular
dislocation: Surgical versus conservative
treatment. J Shoulder Elbow Surg 2008;17:
220-225.
21. Canadian Orthopaedic Trauma Society:
Multicenter randomized clinical trial of
nonoperative versus operative treatment of
acute acromio-clavicular joint dislocation.
J Orthop Trauma 2015;29:479-487.
22. Venjakob AJ, Salzmann GM, Gabel F, et al:
Arthroscopically assisted 2-bundle anatomic
reduction of acute acromioclavicular joint
separations: 58-month findings. Am J Sports
Med 2013;41:615-621.
23. Darabos N, Vlahovic I, Gusic N, Darabos
A, Bakota B, Miklic D: Is AC TightRope
fixation better than Bosworth screw
fixation for minimally invasive operative
treatment of Rockwood III AC joint injury?
Injury 2015;46(suppl 6):S113-S118.
24. Beitzel K, Obopilwe E, Chowaniec DM,
et al: Biomechanical comparison of
arthroscopic repairs for acromioclavicular
joint instability: Suture button systems
without biological augmentation: Am J
Sports Med 2011;39:2218-2225.
25. Di Francesco A, Zoccali C, Colafarina O,
Pizzoferrato R, Flamini S: The use of hook
plate in type III and V acromio-clavicular
Rockwood dislocations: Clinical and
radiological midterm results and MRI
evaluation in 42 patients. Injury 2012;43:
147-152.
26. Millett PJ, Horan MP, Warth RJ: Two-year
outcomes after primary anatomic
coracoclavicular ligament reconstruction.
Arthroscopy 2015;31:1962-1973.
27. Frank RM, Trenhaile SW: Arthroscopic-
assisted acromioclavicular joint
reconstruction using the TightRope device
with allograft augmentation: Surgical
technique. Arthrosc Tech 2015;4:
e293-e297.
28. Carofino BC, Mazzocca AD: The anatomic
coracoclavicular ligament reconstruction:
Surgical technique and indications.
J Shoulder Elbow Surg 2010;19:37-46.
29. Spencer HT, Hsu L, Sodl J, Arianjam A,
Yian EH: Radiographic failure and rates of
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
re-operation after acromioclavicular joint
reconstruction: A comparison of surgical
techniques. Bone Joint J 2016;98-B:
512-518.
30. Thomas K, Litsky A, Jones G, Bishop JY:
Biomechanical comparison of
coracoclavicular reconstructive techniques.
Am J Sports Med 2011;39:804-810.
31. Tauber M, Gordon K, Koller H, Fox M,
Resch H: Semitendinosus tendon graft
versus a modified weaver-dunn procedure
for acromioclavicular joint reconstruction
in chronic cases: A prospective comparative
study. Am J Sports Med 2009;37:181-190.
32. Mazzocca AD, Conway JE, Johnson S, et al:
The anatomic coracoclavicular ligament
reconstruction. Oper Tech Sports Med
2004;12:56-61.
33. Martetschlager F, Horan MP, Warth RJ,
Millett PJ: Complications after anatomic
fixation and reconstruction of the
coracoclavicular ligaments. Am J Sports
Med 2013;41:2896-2903.
34. Milewski MD, Tompkins M, Giugale JM,
Carson EW, Miller MD, Diduch DR:
Complications related to anatomic
reconstruction of the coracoclavicular
ligaments. Am J Sports Med 2012;40:
1628-1634.
35. Rolf O, Hann von Weyhern A, Ewers A,
Boehm TD, Gohlke F: Acromioclavicular
dislocation Rockwood III-V: Results of
early versus delayed surgical treatment.
Arch Orthop Trauma Surg 2008;128:
1153-1157.
36. Cook JB, Shaha JS, Rowles DJ, Bottoni CR,
Shaha SH, Tokish JM: Clavicular bone
tunnel malposition leads to early failures in
coracoclavicular ligament reconstructions.
Am J Sports Med 2013;41:142-148.
37. Weinstein DM, McCann PD, McIlveen SJ,
Flatow EL, Bigliani LU: Surgical treatment
of complete acromioclavicular dislocations.
Am J Sports Med 1995;23:324-331.
38. Beitzel K, Sablan N, Chowaniec DM, et al:
Sequential resection of the distal clavicle
and its effects on horizontal
acromioclavicular joint translation. Am J
Sports Med 2012;40:681-685.
39. Beaver AB, Parks BG, Hinton RY:
Biomechanical analysis of distal clavicle
excision with acromioclavicular joint
reconstruction. Am J Sports Med 2013;41:
1684-1688.
40. Eisenstein ED, Lanzi JT, Waterman BR,
Bader JM, Pallis MP: Medialized clavicular
bone tunnel position predicts failure after
anatomic coracoclavicular ligament
reconstruction in young, active male
Jay B. Cook, MD and Kevin P. Krul, MD
patients. Am J Sports Med 2016;44:
2682-2689.
41. Woodmass JM, Esposito JG, Ono Y, et al:
Complications following arthroscopic
fixation of acromioclavicular separations:
A systematic review of the literature. Open
Access J Sports Med 2015;6:97-107.
42. Spiegl UJ, Smith SD, Euler SA, Dornan GJ,
Millett PJ, Wijdicks CA: Biomechanical
consequences of coracoclavicular
reconstruction techniques on clavicle
strength. Am J Sports Med 2014;42:
1724-1730.
43. Martetschlager F, Saier T, Weigert A, et al:
Effect of coracoid drilling for
acromioclavicular joint reconstruction
techniques on coracoid fracture risk: A
biomechanical study. Arthroscopy 2016;
32:982-987.
44. Campbell ST, Heckmann ND, Shin SJ, et al:
Biomechanical evaluation of coracoid
tunnel size and location for
coracoclavicular ligament reconstruction.
Arthroscopy 2015;31:825-830.
45. Glanzmann MC, Buchmann S, Audige L,
Kolling C, Flury M: Clinical and
radiographical results after double flip
button stabilization of acute grade III and
IV acromioclavicular joint separations.
Arch Orthop Trauma Surg 2013;133:
1699-1707.
46. Torkaman A, Bagherifard A, Mokhatri T,
et al: Double-button fixation system for
management of acute acromioclavicular
joint dislocation. Arch Bone Joint Surg
2016;4:41-46.
47. Loriaut P, Casabianca L, Alkhaili J, et al:
Arthroscopic treatment of acute
acromioclavicular dislocations using a
double button device: Clinical and MRI
results. Orthop Traumatol Surg Res 2015;
101:895-901.
48. Rosslenbroich SB, Schliemann B, Schneider
KN, et al: Minimally invasive
coracoclavicular ligament reconstruction
with a flip-button technique (MINAR):
Clinical and radiological midterm results.
Am J Sports Med 2015;43:1751-1757.
49. Schlegel TF, Burks RT, Marcus RL, Dunn
HK: A prospective evaluation of untreated
acute grade III acromioclavicular separations.
Am J Sports Med 2001;29:699-703.
50. Murena L, Canton G, Vulcano E,
Cherubino P: Scapular dyskinesis and SICK
scapula syndrome following surgical
treatment of type III acute
acromioclavicular dislocations. Knee Surg
Sports Traumatol Arthrosc 2013;21:
1146-1150.
677
 Review Article

Comprehensive Review of the

Elbow Physical Examination

Abstract
Matthew V. Smith, MD, MSc Physical examination of the elbow is a critical component in
Joseph D. Lamplot, MD formulating an accurate diagnosis. Various special physical
examinations have been described to improve the clinician’s ability to
Rick W. Wright, MD
establish an accurate diagnosis. A comprehensive approach to the
Robert H. Brophy, MD physical examination of the elbow, including special tests, may
facilitate improved diagnosis of elbow pathology.

From the Department of Orthopedics,

Washington University Saint Louis, St.
Louis, MO.
Downloaded from http://journals.lww.com/jaaos by BhDMf5ePHKbH4TTImqenVBhf5K8zlU/3BW1Xi8ssKXpWgeQFDWB+87sRjl7ZuVul on 09/27/2018
P hysical examination of the elbow
is a critical component in formu-
tures comprising the LUCL complex,
is the most important for elbow sta-

Dr. Smith or an immediate family
member is a member of a speakers’
bureau or has made paid
presentations on behalf of Arthrex.
Dr. Wright or an immediate family
member has received research or
institutional support from the National
Institutes of Health (NIAMS and
NICHD) and serves as a board
member, owner, officer, or committee
member of the American Board of
Orthopaedic Surgery, the American
Orthopaedic Association, and the
American Orthopaedic Society for
Sports Medicine. Dr. Brophy or an
immediate family member is a
member of a speakers’ bureau or has
made paid presentations on behalf of
Arthrex; has received research or
institutional support from Orteq Sports
Medicine; and serves as a board
member, owner, officer, or committee
member of the American Academy of
Orthopaedic Surgeons, the American
Orthopaedic Association, the
American Orthopaedic Society for
Sports Medicine, and the Orthopaedic
Research Society. Neither
Dr. Lamplot nor any immediate family
member has received anything of
value from or has stock or stock
options held in a commercial company
or institution related directly or
indirectly to the subject of this article.
J Am Acad Orthop Surg 2018;26:
678-687
DOI: 10.5435/JAAOS-D-16-00622
Copyright 2018 by the American
Academy of Orthopaedic Surgeons.
lating an accurate diagnosis. Various
special physical examinations that
improve the clinician’s ability to
establish an accurate diagnosis have
been described. In this article, we
present a comprehensive approach
to the physical examination of the
elbow, with a focus on special tests
as described by their original authors.
We have included interpretations of
each test and statistical information
regarding the accuracy of each test, if
available.
Anatomy and
Biomechanics
The elbow is a complex hinge joint
that comprises three bony articu-
lations that provide stability to the
joint.1,2 The medial ulnar collateral
ligament (MUCL) and the lateral ul-
nar collateral ligament (LUCL) com-
plex provide additional static
constraints to the elbow.3 The MUCL
resists valgus stress to the elbow,
particularly within the functional
range of throwing from 20° to 120°,2
and consists of anterior, posterior,
and transverse bundles.4 The LUCL
complex comprises the LUCL, radial
collateral ligament, annular ligament,
and accessory LUCL. Of the struc-
bility.5 The examiner should also
know the course of the ulnar, median,
and radial nerves because they cross
the elbow.
Physical Examination
The examination begins with inspec-
tion of the affected elbow with com-
parison with the contralateral side.
The examiner should observe the
resting position of the elbow. In pa-
tients with an effusion, the elbow is
often held in 70° to 80° of flexion, a
position accommodating the greatest
capsular volume.6 The examiner
should also assess the carrying angle
of the elbow. In full extension, a
normal valgus carrying angle is
approximately 11° in men and 13° in
women.7 The carrying angle changes
in a linear fashion during flexion,
moving from valgus to varus as the
elbow moves from extension into
flexion, making accurate assessment
of the resting carrying angle difficult
in the setting of a flexion contrac-
ture.8 An increased carrying angle
has been observed in professional
throwing athletes and may indicate
an adaptation to repetitive valgus
stress.9,10

678 Journal of the American Academy of Orthopaedic Surgeons

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.

Palpation
Tenderness along the epicondyles
frequently indicates epicondylitis in
skeletally mature patients and may
indicate an epiphyseal plate injury in
skeletally immature patients.11 In
patients with medial epicondylitis,
maximal tenderness occurs 5 to
10 mm distal and anterior to the
medial epicondyle along the pro-
nator teres and flexor carpi radia-
lis.12 With the elbow in 50° to 70° of
flexion to move the flexor pronator
mass anterior to the ulnar collateral
ligament (UCL), tenderness in the
sulcus just distal to the medial epi-
condyle of the humerus or along the
ulnar sublime tubercle may indicate
injury to the UCL11 (Figure 1). Ten-
derness posterior to the medial epi-
condyle may indicate ulnar neuritis or
olecranon stress fracture. Tenderness
along the lateral border of the olec-
ranon may also suggest olecranon
stress fracture.11 In patients with lat-
eral epicondylitis, tenderness is most
commonly experienced on the exten-
sor carpi radialis brevis (ECRB) ten-
don origin just anterior to the border
of the lateral epicondyle.13 The radial
head can be palpated while passively
rotating the forearm, and tenderness
may indicate fracture, dislocation, or
annular ligament injury.11 The lateral
soft spot can be palpated between the
olecranon tip, radial head, and lateral
epicondyle. Pain in this soft spot may
indicate synovitis, a posterolateral
plica, or osteochondritis dissecans of
the capitellum. Fullness in the soft
spot is indicative of an effusion or
hemarthrosis.
Range of Motion
Normal elbow range of motion is
from full extension to approximately
145° of flexion,14 with approxi-
mately 75° and 85° of pronation and
supination, respectively.8,14 Loss of
terminal extension is often the first
motion lost after injury and may
suggest intra-articular pathology.1
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
However, a notable loss of terminal
extension of ,16° has been noted
in ,50% of professional throwing
athletes and does not necessarily
indicate injury in this population.9,10
While assessing motion, the exam-
iner should check for any crepitus,
pain, or mechanical symptoms, which
may indicate cartilage irregularities or
loose bodies.11 In an elbow that has
lost motion in flexion or extension, a
soft end point suggests an effusion,
soft-tissue swelling, or capsular
tightness,1 whereas a firm end point
may indicate an osteophyte or loose
body causing a mechanical block to
motion. Pain at the end point of
flexion and extension localized to the
medial olecranon may reflect degen-
erative changes associated with
chronic valgus extension overload and
posteromedial impingement, particu-
larly in throwing athletes.15
Strength
Isometric strength should be com-
pared with the contralateral side.
Inability to actively extend the elbow
indicates a complete triceps tendon
rupture. Triceps strength should
be assessed with the shoulder at 90°
of forward elevation and internal
rotation, thereby avoiding the influ-
ence of gravity on the examination.
In certain cases, such as suspected
lateral or medial epicondylitis, grip
strength should be evaluated and
compared with the unaffected side.12,16
Weakness around the elbow can also
be associated with cervical nerve root
injury or brachial plexus injury.
Special Tests for the Elbow
Ulnar Collateral Ligament
Assessment
Milking Maneuver
The integrity of the anterior bundle of
the MUCL can be tested using the
milking maneuver, which attempts to
simulate the position of stress in a
Matthew V. Smith, MD, MSc, et al
Figure 1
Clinical photograph demonstrating
the ulnar collateral ligament (UCL;
arrow). The ligament should be
palpated just distal to the medial
epicondyle (arrow) with the elbow
flexed 50° to 70° to move the flexor
pronator mass anterior to the UCL.
throwing athlete.4,17 With the affected
elbow flexed to .90°, the examiner
pulls on the patient’s thumb with the
patient’s forearm supinated, the
shoulder forward-flexed, and elbow
flexed past 90° (Figure 2). The test is
positive if the patient experiences a
subjective feeling of apprehension
and instability along with medial
elbow pain. The maneuver’s sensitiv-
ity has been reported as 87.5%, with a
negative predictive value of 100%.15
Moving Valgus Stress Test
Initially, the manual valgus stress test
was used to assess the integrity of the
anterior bundle of the MUCL.18,19
However, during this test, stabilizing
the humerus at flexion angles .30°
is difficult2,5 and subtle laxity may be
too small to detect on examination in
the setting of partial injuries. The
sensitivity and specificity of the man-
ual valgus stress test for laxity has been
described as 19% (95% confidence
interval [CI], 4% to 46%) and 100%
(95% CI, 40% to 100%), respec-
tively.15 In an effort to produce a more
sensitive test for detecting UCL tears,
O’Driscoll et al15 described the mov-
ing valgus stress test. With the patient
seated or standing, the shoulder is
679
Comprehensive Review of the Elbow Physical Examination

Figure 2 by the patient with aggravating activ-

ities, and the patient should experience
maximal pain between approximately
120° (late cocking) and 70° (early
acceleration) as the elbow is extended
from a flexed position. The sensitivity
of the moving valgus stress test has
been described as 100% (95% CI,
81% to 100%) and specificity as 75%
(95% CI, 19% to 99%).15 By com-
parison, in the same study, the sensi-
tivity and specificity of MRI was 40%
(95% CI, 16% to 68%) and 100%
(95% CI, 29% to 100%), respectively.

Epicondylitis
Maudsley Test
Resisted middle finger extension selec-
tively recruits the ECRB tendon and
can reproduce symptoms of lateral
A and B, Clinical photographs demonstrating the milking maneuver. A, The
affected elbow is flexed .90°. B, The examiner pulls on the patient’s thumb with epicondylitis but may also be painful in
the patient’s forearm supinated, the shoulder forward-flexed and elbow flexed the setting of radial tunnel syndrome.20
.90°. The entire course of the ulnar collateral ligament should be palpated Resisted wrist extension with the
during this maneuver. A subjective feeling of apprehension and instability along
elbow in full extension and pronation
with medial elbow pain indicates a positive test result.
stretches the common extensor origin
and can also recreate symptoms of
Figure 3 lateral epicondylitis.1 In patients with
lateral epicondylitis, grip strength de-
creases as the elbow moves from a
position of flexion to extension, with a
29% decrease in grip strength while in
extension compared with flexion.21

Chair Test
The patient is asked to lift a chair with
the shoulder forward-flexed, elbow
extended, and forearm pronated. If
this maneuver provokes lateral elbow
pain, it indicates lateral epicondylitis.22

Medial Epicondylitis Test

The patient’s elbow is fully extended
A and B, Clinical photographs showing the moving valgus stress test, which may
and the examiner resists the patient’s
be performed with the patient seated or standing and with the shoulder abducted wrist flexion and forearm pronation.
to 90° and the elbow maximally flexed. A, Modest valgus stress is applied to the Medial epicondylitis is likely if the
elbow while the shoulder is externally rotated to its limit. B, While maintaining patient experiences pain over the
constant valgus stress, the elbow is quickly extended to 30°.
medial epicondyle with these maneu-
vers.23 Although grip strength is
abducted to 90°, and the elbow While maintaining constant valgus decreased compared with baseline and
maximally flexed as modest valgus stress, the elbow is quickly extended to to the contralateral side, the magni-
stress is applied to the elbow with the 30° (Figure 3). A positive test must tude of impairment is often less than
shoulder externally rotated to its limit. generate the pain that is experienced that seen in lateral epicondylitis.24

680 Journal of the American Academy of Orthopaedic Surgeons

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
 Matthew V. Smith, MD, MSc, et al

Triceps Rupture Figure 4

Localized tenderness and swelling
with a palpable defect in the triceps
tendon can confirm the diagnosis
of rupture.25 However, swelling
often prevents the examiner from
palpating a defect in the tendon.
Viegas26 described a modification of
the Thompson test for Achilles ten-
don rupture. The patient’s upper arm
is placed on a table with the patient
prone and the forearm hanging
freely over the table edge. The
examiner manually squeezes the tri-
ceps muscle belly while the patient’s
arm is relaxed. If the triceps tendon is
intact, the elbow should slightly
extend.25,26

Valgus Extension Overload

The valgus extension overload test,
or extension impingement test, is
performed with the patient seated
and the shoulder slightly forward-
flexed. The examiner repeatedly
brings the slightly flexed elbow into
full extension while applying valgus
stress, attempting to reproduce pain
with impingement of the postero-
medial tip of the olecranon on the
medial wall of the olecranon fossa.27
The test is considered positive if the
posterior or posteromedial pain re-
produced during testing is similar to
the pain experienced by the patient
during throwing. Valgus loading
during the maneuver generally in-
creases pain, whereas varus loading
decreases pain.28
Posterolateral Rotatory
Instability
Lateral Pivot Shift Test
The lateral pivot shift test (ie,
posterolateral rotatory apprehension
test) may be performed with the
patient either awake or anesthetized
(Figure 4).29 The patient is posi-
tioned supine with the affected
extremity raised over the patient’s
head, the shoulder fully externally
A through D, Clinical photographs showing the lateral pivot shift test, which may
be performed with the patient awake or anesthetized. A, The patient is positioned
supine with the affected extremity raised over the patient’s head, the shoulder
fully externally rotated, and the forearm fully supinated. B, The examiner grasps
the patient’s wrist or forearm and, starting from a position of full extension, slowly
flexes the elbow while applying valgus and supination moments along with axial
compression. C, At approximately 40° of flexion, rotatory displacement is
maximized, and the patient may demonstrate apprehension and pain. With
further flexion, a clunk may occur when the joint is reduced. D, Posterolateral
dislocation of the radiocapitellar joint results in radial head prominence and a
skin dimple just proximal to the dislocated radial head. (Panel D reproduced from
Mehta JA, Bain GI: Posterolateral rotatory instability of the elbow. J Am Acad
Orthop Surg 2004;12[6]:405-415.)
rotated, and the forearm fully supi- the dislocated radiohumeral joint.
nated. The examiner grasps the pa- With further flexion, a clunk will
tient’s wrist or forearm and, starting occur when the joint is reduced.
from a position of full extension,
slowly flexes the elbow while ap- Posterolateral Rotatory Drawer
plying valgus and supination mo- Test
ments along with axial compression. The elbow is placed in 40° of flexion,
At approximately 40° of flexion, and an anterior-to-posterior force
rotatory displacement is maximized, on the lateral aspect of the proximal
and the patient may demonstrate radius and ulna is applied in an
apprehension and pain. A positive attempt to translate the forearm away
result in an anesthetized patient will from the humerus on the lateral side,
demonstrate a palpable and visible pivoting about the intact medial lig-
clunk along with posterior promi- aments.3 A positive test is indicated by
nence and associated dimple proxi- apprehension or the presence of a skin
mal to the radial head representing dimple.30

October 1, 2018, Vol 26, No 19 681

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Comprehensive Review of the Elbow Physical Examination

Figure 5

A through C, Clinical photographs showing the table-top relocation test. A, The patient is asked to stand in front of a table
with the hand placed around the outer edge of the table. The patient performs a press-up maneuver with the elbow pointing
laterally, maintaining the forearm in supination. B, The patient is asked to push down through the hand onto the edge of the
table, allowing the elbow to flex, and bringing the chest toward the table. C, The maneuver is repeated with the examiner
placing his or her thumb over the radial head while the patient performs the same maneuver.

Prone Push-up Test
The patient is positioned prone on the
floor with the elbows flexed at 90°,
the forearms supinated, and the arms
abducted to greater than the shoul-
der width. The patient is asked to
perform an active push-up. A test is
considered positive if apprehension
and guarding occur as the affected
elbow is terminally extended from a
flexed position.31
Chair Push-up Test
The patient begins seated in a chair
with the elbows flexed at 90°, fore-
arms supinated, and arms abducted
to greater than the shoulder width.
The patient is asked to push up from
the chair using exclusively upper
extremity forces. A test is considered
positive if the patient demonstrates
apprehension while extending the
elbow as he or she raises his or her
body from the chair.31
Regan and Lapner31 performed a
prospective evaluation of the prone
push-up test, the chair push-up test,
and the lateral pivot shift test; the
prone and chair push-up tests
demonstrated a higher diagnostic
sensitivity compared with the lateral
pivot shift test. For the prone and
chair push-up tests, they reported
87.5% sensitivity for each test, with
a combined sensitivity of 100%,
compared with 37.5% for the lateral
pivot shift test.
Table-top Relocation Test
The table-top relocation test com-
prises three parts32 (Figure 5). First,
the patient is asked to stand in
front of a table with the hand of
the symptomatic extremity placed
around the outer edge of the table.
The patient is asked to perform a
press-up maneuver with the elbow
pointing laterally, maintaining the
forearm in supination. The patient is
then asked to push down through the
hand onto the edge of the table, al-
lowing the elbow to flex while
bringing the chest toward the table.
The test is considered positive if pain
and apprehension occur at approxi-
mately 40° of flexion. Next, the ma-
neuver is repeated with the examiner
placing his or her thumb over the
radial head while the patient per-
forms the same maneuver. Symptoms
of pain and instability should be
relieved as the examiner’s thumb
prevents posterior subluxation of the
radial head. Finally, the examiner
removes his or her supportive thumb
during mid-elbow flexion of the same

682 Journal of the American Academy of Orthopaedic Surgeons

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.

maneuver performed a third time.
Pain and apprehension during the
first and third portions of this test,
with relief during the second portion,
reinforce the diagnosis of postero-
lateral rotatory instability and help
to exclude intra-articular injury,
in which pain may be experi-
enced during all three parts of this
examination.
Radiocapitellar Joint
Pathology
Physical examination may demon-
strate an effusion, crepitus, and
tenderness over the radiocapitellar
articulation. Swelling may cause
loss of terminal extension. For the
radiocapitellar compression test, the
examiner brings the patient’s fore-
arm through a passive range of
forearm pronation and supination
at the midrange of elbow flexion
while applying an axial load to the
radiocapitellar joint. Crepitus or
pain with axial loading represents a
positive test and is indicative of
osteochondritis dissecans or radio-
capitellar joint chondrosis.33
Biceps Rupture
Hook Test
The patient is asked to actively flex
the elbow to 90° and fully supinate
the forearm. The examiner attempts
to “hook” the biceps tendon with the
finger from the lateral side of the
antecubital fossa beneath the lateral
edge of the biceps tendon34 (Figure
6). With an intact biceps tendon,
the finger can be inserted beneath
the tendon approximately 1 cm, or
roughly to the distal interphalangeal
joint. The lacertus fibrosis can be
palpated with the opposite index
finger, palpating from the medial
side in a similar manner. The space
beneath the lacertus fibrosis is
smaller and tighter, and the finger
cannot be hooked beneath the edge
as far as with the biceps tendon.
Furthermore, the medial edge of the
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
lacertus is thin and sheet-like. In a
positive hook test, no cord-like ten-
don can be hooked. In the setting of a
lacertus avulsion, no medial sheet-
like edge can be palpated. A biceps
avulsion with an intact lacertus is
commonly encountered. If the biceps
tendon can be hooked but elicits a
painful response, the test suggests a
partial tear or injury of the tendon
sheath. For the hook test, the authors
report a sensitivity, specificity, posi-
tive predictive value, and negative
predictive value of 100%, whereas
they report the sensitivity and spec-
ificity of MRI to be 92% and 85%,
respectively.
Biceps Squeeze Test
Analogous to the Thompson test for
Achilles tendon rupture, the biceps
squeeze test is performed with the
patient seated and the forearm of the
affected extremity resting in the pa-
tient’s lap with the elbow flexed at
approximately 70° to relax the bra-
chialis.35 The forearm is slightly
pronated to place tension on the
biceps tendon. The examiner squee-
zes the biceps with both hands,
placing one on the distal my-
otendinous junction and the other
around the biceps muscle belly. As the
intact biceps is squeezed, the muscle
belly is drawn away from the
humerus, eliciting an anterior bow of
the muscle belly and forearm supina-
tion. A lack of forearm supination is
considered a positive test and indicates
rupture of the biceps tendon or mus-
cle. The reported sensitivity of the test
is 96%, with an unreliable specificity.
Biceps Crease Interval
The biceps crease interval (BCI) is
used to assess for biceps rupture by
measuring the distance between pal-
pable anatomic landmarks.36 First,
the patient’s elbow is brought from
flexion into full extension while
supinating the forearm (Figure 7).
The main flexion crease in the an-
Matthew V. Smith, MD, MSc, et al
Figure 6
Clinical photograph showing the
hook test. The patient fully
supinates and actively flexes the
elbow to 90°. The examiner places
an index finger from the lateral side
of the antecubital fossa beneath the
lateral edge of the biceps tendon in
an attempt to hook the thick, cord-
like tendon. With an intact biceps
tendon, the finger can be inserted
beneath the tendon approximately
1 cm, or roughly to the distal
interphalangeal joint, allowing the
examiner to pull the tendon
anteriorly.
tecubital fossa is marked. The con-
tour of the distal biceps is lightly
palpated back and forth along a line
parallel to its long axis to identify
the point at which the curve of the
distal biceps begins to turn most
sharply toward the antecubital fossa
(cusp). The cusp is marked with a
transverse line. The distance be-
tween the crease and the cusp is
then measured in centimeters to one
decimal place, and this distance is
recorded as the BCI. The contra-
lateral, unaffected arm is also
measured to calculate a biceps
crease ratio. Using a threshold of
6.0 cm for the BCI or 1.2 for the
biceps crease ratio, the authors
report a sensitivity, specificity, and
overall accuracy of 92%, 100%,
and 93%, respectively.
Passive Forearm Pronation Test
The passive forearm pronation test is
based on the observation that the
biceps muscle belly moves proxi-
mally with forearm supination and
distally with forearm pronation,
both actively and passively.37 The
683
Comprehensive Review of the Elbow Physical Examination

Figure 7

A through C, Clinical photographs showing the biceps crease interval (BCI). A and B, The main flexion crease (arrow) is
marked. The contour of the distal biceps is palpated along a line parallel to its long axis to identify the point at which the
curve of the distal biceps begins to turn most sharply toward the antecubital fossa (cusp). C, The cusp (arrow) is
marked. The distance between the crease (arrow) and the cusp (arrow) is measured and recorded as the BCI. The
contralateral, unaffected arm is also measured to calculate a biceps crease ratio (BCR). Using a threshold of 6.0 cm for
the BCI or 1.2 for the BCR, sensitivity, specificity, and overall accuracy of 92%, 100%, and 93%, respectively, have
been reported.

examiner moves the patient’s fore-
arm through a passive range of
supination and pronation, using
both palpation and visualization to
assess for appropriate movement of
the biceps muscle belly.
Application of the hook test, passive
forearm pronation test, and BCI test in
sequence result in 100% sensitivity
and specificity for complete biceps
ruptures when the outcomes of each
test are in agreement.38
Compressive Neuropathies
About the Elbow
Ulnar Neuritis/Cubital Tunnel
Syndrome
Ulnar neuritis/cubital tunnel syn-
drome can be seen in isolation or
concurrently with other medial
elbow conditions, including medial
epicondylitis and UCL injuries.4,12
Patients often present with medial
elbow pain; paresthesias of the
small finger, ulnar half of the ring
finger, and ulnar aspect of the hand;
hand weakness secondary to intrinsic
muscle weakness; and diminished
grip strength, pinch strength, and
fatigue.39 Symptoms may be exacer-
bated with prolonged elbow flexion.
Careful examination may demon-
strate interosseus atrophy in the set-
ting of prolonged disease. With the
elbow in full extension, gentle palpa-
tion and percussion (ie, Tinel sign) of
the ulnar nerve along its full course
starting at the axilla and moving dis-
tally behind the medial epicondyle
through the cubital tunnel and into the
FCU muscle may elicit pain or radi-
ating symptoms into the forearm
and/or ring and little fingers.39 A two-
point discrimination test may be
abnormal in the setting of both
cubital tunnel syndrome and most
distal ulnar nerve entrapment. How-
ever, an abnormal sensory exami-
nation over the ulnar aspect of the
dorsal hand in the distribution of the
dorsal sensory branch of ulnar nerve
can differentiate cubital tunnel from
more distal entrapment because this
nerve branches approximately 5 cm
proximal to the wrist.39 Furthermore,
compared with more distal nerve
entrapment, cubital tunnel syndrome
generally demonstrates a motor
deficit to ulnar innervated extrinsic
muscles and less clawing. Additional
examination maneuvers include di-
rect compression of the nerve and
prolonged maximal elbow flexion
(ie, elbow flexion test), both of which
may provoke or exacerbate symp-
toms in the setting of cubital tunnel
syndrome. Because of their high false-
positive rates, no tests for cubital tun-
nel syndrome are highly specific.40
Ulnar nerve hypermobility occurs in
greater than one third of the adult
population and does not seem to be
associated with an increased inci-
dence of symptomatology.41 Snap-
ping of the medial elbow, often
associated with ulnar nerve symp-
toms, is not necessarily caused by
dislocation or subluxation of the
ulnar nerve.42 Rather, recognizing
the possibility of a snapping medial
head of the triceps over the medial
epicondyle is important. A snapping
medial triceps can be elicited with
resisted extension of a fully flexed
elbow.

684 Journal of the American Academy of Orthopaedic Surgeons

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.

Proximal Median Nerve
Entrapment
Proximal median nerve entrapment
(PMNE), also called pronator syn-
drome, has been classically described
as resulting from compression of the
median nerve because it courses
between the two heads of the pro-
nator teres muscle or at the proximal
arch of the flexor digitorum super-
ficialis muscle.43 However, more
proximal median nerve compression
may occur at the lacertus fibrosis,
ligament of Struthers, or by an
accessory head of the flexor pollicis
longus.44-46 Determination of the
specific location of PMNE is impor-
tant because entrapment at specific
sites such as the lacertus (lacertus
tunnel syndrome) may not require an
extensive release with surgery.47 Pa-
tients with PMNE often report loss of
pinch strength and fine motor skills,
clumsiness, and occasional par-
esthesias in the radial three and one-
half digits or the palmar cutaneous
branch of median nerve distribu-
tion.47 PMNE can be distinguished
from carpal tunnel syndrome by
numbness and/or paresthesias in the
distribution of the palmar cutaneous
branch of the median nerve (ie, thenar
eminence) and an absence of positive
provocative testing at the wrist.44
The pronator compression test,
the most common finding in PMNE,
is performed by applying pressure to
the proximal volar forearm just
proximal and lateral to the pronator
teres muscle belly. Reproduction of
pain or paresthesias within 30 sec-
onds of compression indicates a
positive test.48 A positive Tinel sign
in this location may also indicate
PMNE.44,48,49 Median nerve com-
pression by the pronator teres or
lacertus fibrosis can be reproduced
with resisted pronation and supi-
nation.44,50 Resisted flexion of the
middle finger PIP joint may cause
median nerve compression by the
heads of the flexor digitorum su-
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
perficialis but may also be seen in
carpal tunnel syndrome.51 Hagert47
reported a symptomatic triad in
patients with PMNE at the level
of the lacertus fibrosis consisting of
weakness with manual testing of
median innervated muscles distal to
the lacertus (flexor pollicis longus,
flexor digitorum profundus to the
index finger [FDP II], and flexor
carpi radialis), external pressure on
the lacertus eliciting pain or a pos-
itive Tinel sign, and a positive
scratch-collapse test.47 Electromy-
ography (EMG) and nerve conduc-
tion studies are often normal or
inconclusive.47
Radial Tunnel
Syndrome/Supinator
Syndrome
Compression of the radial nerve at
various points may result in radial
tunnel syndrome, which can present
as lateral elbow pain, vague
dorsal/radial wrist pain, and weak-
ness of the wrist and/or finger ex-
tensors.52 Because there are more
potential sites of compression and a
lower incidence of severe compres-
sion causing nerve dysfunction, the
diagnosis of radial tunnel syndrome
may be more difficult than other
compressive neuropathies.52 In the
setting of lateral elbow pain, radial
tunnel syndrome may be mis-
diagnosed as lateral epicondylitis,
and both conditions may demon-
strate pain with resisted middle fin-
ger extension and may coexist. The
nerve may be compressed at the in-
termuscular septum between the tri-
ceps and brachialis, the proximal
edge of the ECRB, the radiocapitellar
joint, the arcade of Frohse, between
the two heads of the supinator, or
beneath the brachioradialis.52-54
Compression above the elbow may
cause weakness of both wrist and
finger extensors, whereas compres-
sion below the elbow does not affect
wrist strength. Decreased sensation
Matthew V. Smith, MD, MSc, et al
Figure 8
Clinical photograph demonstrating
the rule-of-nine test. Pressure is
applied to nine areas denoted by
circles. In the setting of median or
radial nerve compression, pressure
on the three medial circles (1 to 3)
should not reproduce symptoms.
Pressure on the biceps tendon, the
most proximal of the middle circles
(4), should be painful only in the
setting of biceps tendonitis. Of the
middle circles, pressure on the two
most distal circles (5 and 6) may
reproduce pain in the setting of
pronator syndrome. Of the lateral
circles, the two more proximal circles
(1 and 2) overlie the radial nerve.
Because of the course of the
posterior interosseous nerves,
pressure on the most distal lateral
circle does not compress the radial
nerve.
in the distribution of the superficial
radial nerve indicates compression
proximal to the arcade of Frohse.
Clinical examination for radial
tunnel syndrome distal to the elbow
(ie, supinator syndrome) can be per-
formed using the rule-of-nine test52,55
(Figure 8). Manual pressure on the
middle distal circles presses on the
course of the median nerve and can
cause pain in the setting of entrap-
ment. The proximal lateral circles
overlay the radial nerve, and manual
pressure on these circles may repro-
duce symptoms in the setting of
radial tunnel syndrome. Addition-
ally, pressure applied over the
proximal supinator with the hand in
685
Comprehensive Review of the Elbow Physical Examination
full supination can be very painful
and may be relieved with pronation
of the forearm.
Summary
Clinicians must have a thorough
understanding of the physical exam-
ination of the elbow and the special
tests that have been described to
diagnose specific pathologies. The
maneuvers commonly used to assess
for certain pathologies, including
biceps tendon rupture and UCL tears,
have been reported to be more sensi-
tive than advanced imaging studies.
Diagnoses involving instability, in-
cluding valgus extension overload,
rely on the physical examination to
establish a diagnosis. Clinicians may
use this review as a reference con-
taining the description of special
diagnostic examination maneuvers
for various elbow pathologies.
References
References printed in bold type are
those published within the past 5
years.
1. Hsu SH, Moen TC, Levine WN, Ahmad
CS: Physical examination of the athlete’s
elbow. Am J Sports Med 2012;40:
699-708.
2. Morrey BF, Tanaka S, An KN: Valgus
stability of the elbow: A definition of
primary and secondary constraints. Clin
Orthop Relat Res 1991:187-195.
3. O’Driscoll SW, Jupiter JB, King GJ,
Hotchkiss RN, Morrey BF: The unstable
elbow. Instr Course Lect 2001;50:89-102.
4. Chen FS, Rokito AS, Jobe FW: Medial
elbow problems in the overhead-throwing
athlete. J Am Acad Orthop Surg 2001;9:
99-113.
5. Morrey BF, An KN: Functional anatomy of
the ligaments of the elbow. Clin Orthop
Relat Res 1985:84-90.
6. Morrey BF: Applied anatomy and
biomechanics of the elbow joint. Instr
Course Lect 1986;35:59-68.
7. Beals RK: The normal carrying angle of the
elbow: A radiographic study of 422
patients. Clin Orthop Relat Res 1976:
194-196.
686
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
8. Morrey BF, Chao EY: Passive motion of the
elbow joint. J Bone Joint Surg Am 1976;58:
501-508.
9. King GJ, Morrey BF, An KN: Stabilizers of
the elbow. J Shoulder Elbow Surg 1993;2:
165-174.
10. Wright RW, Steger-May K, Wasserlauf BL,
O’Neal ME, Weinberg BW, Paletta GA:
Elbow range of motion in professional
baseball pitchers. Am J Sports Med 2006;
34:190-193.
11. Cain EL Jr, Dugas JR, Wolf RS, Andrews
JR: Elbow injuries in throwing athletes: A
current concepts review. Am J Sports Med
2003;31:621-635.
12. Amin NH, Kumar NS, Schickendantz MS:
Medial epicondylitis: Evaluation and
management. J Am Acad Orthop Surg
2015;23:348-355.
13. Rineer CA, Ruch DS: Elbow tendinopathy
and tendon ruptures: Epicondylitis, biceps
and triceps ruptures. J Hand Surg Am 2009;
34:566-576.
14. Boone DC, Azen SP: Normal range of
motion of joints in male subjects. J Bone
Joint Surg Am 1979;61:756-759.
15. O’Driscoll SW, Lawton RL, Smith AM: The
“moving valgus stress test” for medial
collateral ligament tears of the elbow. Am J
Sports Med 2005;33:231-239.
16. Ahmad Z, Siddiqui N, Malik SS, Abdus-
Samee M, Tytherleigh-Strong G, Rushton
N: Lateral epicondylitis: A review of
pathology and management. Bone Joint J
2013;95-B:1158-1164.
17. Veltri DM, Field LD, Deutsch A, Altchek A,
Potter H: The milking maneuver - a new test
to evaluate the MCL of the elbow in the
throwing athlete. Presented at the 10th
Open Meeting of the American Shoulder
and Elbow Surgeons Specialty Day, New
Orleans, LA, 1994.
18. Bruce JR, Andrews JR: Ulnar collateral
ligament injuries in the throwing athlete.
J Am Acad Orthop Surg 2014;22:315-325.
19. Conway JE, Jobe FW, Glousman RE, Pink
M: Medial instability of the elbow in
throwing athletes: Treatment by repair or
reconstruction of the ulnar collateral
ligament. J Bone Joint Surg Am 1992;74:
67-83.
20. McCallum SD, Paoloni JA, Murrell GA:
Five-year prospective comparison study of
topical glyceryl trinitrate treatment of
chronic lateral epicondylosis at the elbow.
Br J Sports Med 2011;45:416-420.
21. Dorf ER, Chhabra AB, Golish SR, McGinty
JL, Pannunzio ME: Effect of elbow position
on grip strength in the evaluation of lateral
epicondylitis. J Hand Surg Am 2007;32:
882-886.
22. Gardner RC: Tennis elbow: Diagnosis,
pathology and treatment. Nine severe cases
Journal of the American Academy of Orthopaedic Surgeons
treated by a new reconstructive operation.
Clin Orthop Relat Res 1970;72:248-253.
23. Verhaar J, Walenkamp G, Kester A, van
Mameren H, van der Linden T: Lateral
extensor release for tennis elbow: A
prospective long-term follow-up study.
J Bone Joint Surg Am 1993;75:1034-1043.
24. Pienimaki TT, Siira PT, Vanharanta H:
Chronic medial and lateral epicondylitis:
A comparison of pain, disability, and
function. Arch Phys Med Rehabil 2002;83:
317-321.
25. Keener JD, Sethi PM: Distal triceps tendon
injuries. Hand Clin 2015;31:641-650.
26. Viegas SF: Avulsion of the triceps tendon.
Orthop Rev 1990;19:533-536.
27. Wilson FD, Andrews JR, Blackburn TA,
McCluskey G: Valgus extension overload
in the pitching elbow. Am J Sports Med
1983;11:83-88.
28. Ahmad CS, Conway JE: Elbow
arthroscopy: Valgus extension overload.
Instr Course Lect 2011;60:191-197.
29. O’Driscoll SW, Bell DF, Morrey BF:
Posterolateral rotatory instability of the
elbow. J Bone Joint Surg Am 1991;73:
440-446.
30. Anakwenze OA, Kancherla VK, Iyengar J,
Ahmad CS, Levine WN: Posterolateral
rotatory instability of the elbow. Am J
Sports Med 2014;42:485-491.
31. Regan W, Lapner PC: Prospective evaluation
of two diagnostic apprehension signs for
posterolateral instability of the elbow. J
Shoulder Elbow Surg 2006;15:344-346.
32. Arvind CH, Hargreaves DG: Tabletop
relocation test: A new clinical test for
posterolateral rotatory instability of the
elbow. J Shoulder Elbow Surg 2006;15:
707-708.
33. Baumgarten T: Osteochondritis dissecans
of the capitellum. Sports Med Arthrosc Rev
1995;3:219-223.
34. O’Driscoll SW, Goncalves LB, Dietz P: The
hook test for distal biceps tendon avulsion.
Am J Sports Med 2007;35:1865-1869.
35. Ruland RT, Dunbar RP, Bowen JD: The
biceps squeeze test for diagnosis of distal
biceps tendon ruptures. Clin Orthop Relat
Res 2005:128-131.
36. ElMaraghy A, Devereaux M, Tsoi K: The
biceps crease interval for diagnosing
complete distal biceps tendon ruptures. Clin
Orthop Relat Res 2008;466:2255-2262.
37. Harding WG III: A new clinical test for
avulsion of the insertion of the biceps
tendon. Orthopedics 2005;28:27-29.
38. Devereaux MW, ElMaraghy AW: Improving
the rapid and reliable diagnosis of complete
distal biceps tendon rupture: A nuanced
approach to the clinical examination. Am J
Sports Med 2013;41:1998-2004.

39. Elhassan B, Steinmann SP: Entrapment
neuropathy of the ulnar nerve. J Am Acad
Orthop Surg 2007;15:672-681.
40. Gelberman RH: Ulnar tunnel syndrome,
in Gelberman RH, ed: Operative Nerve
Repair and Reconstruction.
Philadelphia, PA, JB Lippincott, 1991,
pp 1131-1143.
41. Calfee RP, Manske PR, Gelberman RH,
Van Steyn MO, Steffen J, Goldfarb CA:
Clinical assessment of the ulnar nerve at the
elbow: Reliability of instability testing and
the association of hypermobility with
clinical symptoms. J Bone Joint Surg Am
2010;92:2801-2808.
42. Spinner RJ, Goldner RD: Snapping of the
medial head of the triceps and recurrent
dislocation of the ulnar nerve: Anatomical
and dynamic factors. J Bone Joint Surg Am
1998;80:239-247.
43. Seyffarth H: Primary myoses in the M.
pronator teres as cause of lesion of the N.
medianus (the pronator syndrome). Acta
Psychiatr Neurol Scand Suppl 1951;74:
251-254.
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
44. Rodner CM, Tinsley BA, O’Malley MP:
Pronator syndrome and anterior
interosseous nerve syndrome. J Am Acad
Orthop Surg 2013;21:268-275.
45. Johnson RK, Spinner M, Shrewsbury MM:
Median nerve entrapment syndrome in the
proximal forearm. J Hand Surg Am 1979;4:
48-51.
46. Seitz WH Jr, Matsuoka H, McAdoo J,
Sherman G, Stickney DP: Acute
compression of the median nerve at the
elbow by the lacertus fibrosus. J Shoulder
Elbow Surg 2007;16:91-94.
47. Hagert E: Clinical diagnosis and wide-awake
surgical treatment of proximal median nerve
entrapment at the elbow: A prospective
study. Hand (N Y) 2013;8:41-46.
48. Gainor BJ: The pronator compression test
revisited: A forgotten physical sign. Orthop
Rev 1990;19:888-892.
49. Hartz CR, Linscheid RL, Gramse RR, Daube
JR: The pronator teres syndrome:
Compressive neuropathy of the median nerve.
J Bone Joint Surg Am 1981;63:885-890.
Matthew V. Smith, MD, MSc, et al
50. Olehnik WK, Manske PR, Szerzinski J:
Median nerve compression in the proximal
forearm. J Hand Surg Am 1994;19:
121-126.
51. Cobb TK, An KN, Cooney WP, Berger RA:
Lumbrical muscle incursion into the carpal
tunnel during finger flexion. J Hand Surg Br
1994;19:434-438.
52. Stanley J: Radial tunnel syndrome: A
surgeon’s perspective. J Hand Ther 2006;
19:180-184.
53. Erak S, Day R, Wang A: The role of
supinator in the pathogenesis of chronic
lateral elbow pain: A biomechanical study.
J Hand Surg Br 2004;29:461-464.
54. Konjengbam M, Elangbam J: Radial nerve
in the radial tunnel: Anatomic sites of
entrapment neuropathy. Clin Anat 2004;
17:21-25.
55. Loh YC, Lam WL, Stanley JK, Soames RW:
A new clinical test for radial tunnel
syndrome: The rule-of-nine test: A
cadaveric study. J Orthop Surg (Hong
Kong) 2004;12:83-86.
687
 Review Article

Management of Bunionette
Deformity

Abstract
Glenn Guangyu Shi, MD Bunionette deformity, historically known as tailor’s bunion, is a
Ammar Humayun, MD forefoot protuberance laterally, dorsolaterally, or plantarlaterally
along the fifth metatarsal head. Although bunionette deformity has
Joseph L. Whalen, MD, PhD
been compared to hallux valgus deformity, it is likely due to a
Harold B. Kitaoka, MD multifactorial, anatomic interplay between fifth metatarsal bony
morphology and forefoot soft-tissue imbalance. Friction generated
between the bony prominence, soft tissue, and associated
Downloaded from http://journals.lww.com/jaaos by BhDMf5ePHKbH4TTImqenVIdHfOa5cT8d0MW97pzEYFI3JSWm4It5g6XurCeS+P2g on 09/27/2018

constrictive footwear can result in keratosis, inflammation, pain, and

ulceration. Symptomatic bunionettes are usually responsive to
nonsurgical management. Surgical options are available based on
the underlying bony deformity when nonsurgical treatment fails.

B unionette deformity involves

From the Department of Orthopedic
Surgery, Mayo Clinic Florida,
Jacksonville, FL (Dr. Shi,
Dr. Humayun, and Dr. Whalen), and
Department of Orthopedic Surgery,
Mayo Clinic Minnesota, Rochester,
MN (Dr. Kitaoka).
None of the following authors or any
immediate family member has
received anything of value from or has
stock or stock options held in a
commercial company or institution
related directly or indirectly to the
subject of this article: Dr. Shi,
Dr. Humayun, Dr. Whalen, and
Dr. Kitaoka.
J Am Acad Orthop Surg 2018;26:
e396-e404
DOI: 10.5435/JAAOS-D-17-00345
Copyright 2018 by the American
Academy of Orthopaedic Surgeons.
lateral bony prominence along
the fifth metatarsal head that can be
painful, especially with constrictive
footwear. The condition was origi-
nally described in tailors and was
thus referred to as tailor’s bunion; it
was theorized that it occurred in
tailors sitting with their legs crossed
and driving pressure and friction
along the lateral aspect of the foot.
Anatomic predispositions and mod-
ifiable triggers have been identified.
Although nonsurgical management
is effective in most patients, indi-
vidualized surgical intervention is
required in others.
Anatomy and
Pathophysiology
Anatomic variations of the lateral
forefoot predispose patients to pain-
ful bunionette deformities (Figure 1).
Previously described fifth metatarsal
bony anomalies that contribute to
symptomatic bunionettes include a
prominent metatarsal head, lateral
bending of the metatarsal shaft,
increased four to five intermetatarsal
angle (IMA), and congenital plan-
tarflexed or dorsiflexed metatarsal.1,2
Associated soft-tissue keratosis, bur-
sitis, and contractures can also trigger
pain. Coexisting plantar callosity
secondary to a plantarflexed fifth
metatarsal shaft causes metatarsalgia
in one-third of patients with bun-
ionette deformities.3 Ankle and
hindfoot deformities, such as flatfoot
deformity, have been theorized to
accentuate the laterally driven pres-
sure, particularly against a constric-
tive shoe.3
Clinical Evaluation
The most common presentation of
bunionette deformity is pain associ-
ated with lateral pressure with foot-
wear. The patient history should rule
out other causes of pain such as gout,
septic joint, inflammatory arthrop-
athy, or trauma.
On standing examination, hindfoot
alignment and the presence of pes
planus deformity may be noted.
Hallux valgus deformity can coexist

e396 Journal of the American Academy of Orthopaedic Surgeons

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
 Glenn Guangyu Shi, MD, et al

with bunionette deformity in patients Figure 1 Figure 2

with wide splayfoot deformities be-
cause hallux valgus deformity shares
common risk factors with bunionette
deformity. This increase in width
of the forefoot can cause once
asymptomatic bunionette deformity
to become painful. During seated
examination, the clinician must
evaluate for associated lateral emi-
nence swelling, erythema, keratosis,
and ulceration. The location of the
callosity, plantar or lateral, and com- Clinical photograph of the right foot
bined variations have been described demonstrating a painful bunionette
as important guides for surgical deformity in a 45-year-old woman.
treatment.1,4 In patients with painful
plantar callosity, osteotomy that is
caused by increased lateral bowing
only translational without the ability
of the fifth metatarsal shaft with a
for elevation of the metatarsal head
normal four to five IMA. Type III
would be incomplete. Furthermore,
deformity is caused by divergent four
an isolated plantar callosity may be
to five metatarsal shafts evidenced by
the result of pathologies other than
an increased IMA, which is the most
bunionette deformity, such as gas-
common deformity and is commonly
trocnemius contracture or cavovarus
associated with pain.1,4,9
foot deformity.
Patients who do not fall into any
Radiographic evaluation includes
of the traditional categories can be
anteroposterior, oblique, and lateral
placed into the lesser known type
views of the weight-bearing foot.
IV. Type IV deformity is character-
Common measurements obtained
ized by a combination of deformities
using the AP view include metatarsal
with a metatarsal width .13 mm
head width, IMA, and fifth meta-
with associated lateral deviation
tarsophalangeal angle (MTPA).
or increase in four to five IMA 8
Average normal IMA has historically
(Table 1).
been considered to be 6.5°, with
angles .8° defined as abnormal.5 AP weight-bearing radiograph of the
Fifth MTPA is measured by the foot demonstrating measurement of
Nonsurgical Management the intermetatarsal angle (IMA) and fifth
degree of divergence of the fifth toe metatarsophalangeal angle (MTPA).
from the long axis of the metatar- Strong evidence for the use of nonsur-
sal shaft. The MTPA measures gical treatment options for symptom-
on average 10.2° to 14° in normal atic bunionette deformity is lacking. In keratosis also can temporarily relieve
feet but 16° in feet with our experience, symptomatic bunion- pressure. The use of localized injections
symptomatic bunionette deform- ette deformities resolve without an to address metatarsophalangeal joint
ities (Figure 2). One study defined invasive procedure in ,90% of (MTPJ) synovitis has been studied,
the normal fifth metatarsal head patients. Nonsurgical management of with moderate, short-term, and
width to be ,13 mm. 6 bunionette deformities often begins sustained improvement in less than
Bunionette deformities have been with patient education regarding one-third of the patients at 2-year
classified into three types based on proper footwear because most symp- follow-up.10
findings observed on AP weight- toms are caused by constrictive
bearing radiographs7 (Figure 3). footwear. Wide shoes, custom
Type I deformity is caused by a mass accommodative orthotics, nonsteroi- Surgical Management
effect secondary to prominent lateral dal anti-inflammatory medications,
condyle or dumbbell-shaped fifth and barrier pads can alleviate the When nonsurgical management fails
metatarsal head. Type II deformity is pain. Skincare with callous shaving of to relieve the pain associated with

October 1, 2018, Vol 26, No 19 e397

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Management of Bunionette Deformity
Figure 3
Classification of bunionette deformities using AP weight-bearing radiographs of a foot to identify an enlarged metatarsal
head (type I deformity; A), lateral bowing of fifth metatarsal shaft (type II deformity; B), and large 4 to 5 intermetatarsal angle
(type III deformity; C).
Table 1
Classification of Bunionette
Deformity
Type Description
I Enlarged fifth metatarsal head
II Lateral bowing of fifth
metatarsal shaft
III Increased IMA
IV Combination of an enlarged
head size and bow or IMA
IMA = intermetatarsal angle
bunionette deformity, surgical options
range from simple lateral eminence
resection to various distal or proximal
fifth metatarsal shaft osteotomies.
Surgical management is mainly
dependent on the severity and the
location of pain and correlates with
clinical and radiographic findings.
Lateral eminence resection is
reserved for patients with a focal
e398
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
prominence and no angular defor-
mity of the metatarsal. Metatarsal
osteotomies can be used to reduce the
width of the forefoot and subsequent
pain while preserving the fifth meta-
tarsal length and joint function.
Similar to hallux valgus correction,
proximal osteotomies allow for
larger correction of the IMA com-
pared with distal osteotomies and
carry a risk of potential injury to the
already tenuous proximal fifth
metatarsal blood supply.11-15 Suffi-
ciently powered, prospective, ran-
domized, controlled trials related to
surgical bunionette correction are
not available. Most of the evidence
related to surgical outcomes is
derived from level IV case series
that use a wide variety of surgical
techniques, inclusion criteria, out-
come measures, concomitant sur-
geries, postoperative protocols, and
lengths of follow-up. The validity of
many conclusions is questioned
Journal of the American Academy of Orthopaedic Surgeons
because of shortcomings of the
study design.
Lateral Eminence Resection
Resection of the lateral condyle is
typically reserved for patients who
have an isolated prominent lateral
condyle, but it can also be considered
for patients who cannot tolerate the
limitations or comply with postoper-
ative care (Figure 4). Some surgeons
consider the use of simple resection
in patients who have symptomatic
bunionette deformities and in those
who wish to avoid or are not can-
didates for osteotomy. Resection
does not correct malalignment but
rather reduces the mass effect.
Recovery is often fast with surgical
options that preserve joint mobility
and metatarsal length without oste-
otomy, avoiding the osteotomy-
associated complications of nonunion
and malunion. A case series of

16 patients (21 feet) treated with
lateral condyle resection reported a
71% patient satisfaction rate, with
no notable correction of angular
deformity and no relationship be-
tween the amount of resection and
patient satisfaction.16 Relative contra-
indications of lateral condylar resec-
tion include pes planus deformity and
forefoot pronation because the later-
ally driven force will continue to pro-
duce pain despite resection because of
the position of the foot during weight
bearing.
Distal Fifth Metatarsal Head
Osteotomy
Distal osteotomies have evolved over
time and have more corrective power
than simple lateral condyle resection
alone.17-23 Distal osteotomies are
indicated for painful type II bun-
ionette deformity. Although trans-
verse and oblique distal osteotomies
have been criticized for the risk of
instability, malunion, and the high
potential of recurrence, Cooper and
Coughlin24 reported that, in a study
of 14 patients with type I deformity
treated with subcapital oblique
osteotomy, the rate of good or
excellent clinical results was 88%
at a mean follow-up of 2.9 years.
Alternatively, chevron osteotomy
of distal fifth metatarsal allows for
correction with an inherently stable
osteotomy design. In a case series by
Kitaoka et al23 with a mean follow-
up of 7.1 years, the authors reported
improvement in pain control, IMA,
forefoot width, and fifth MTPA after
distal chevron osteotomy. Satisfac-
tion was 89.5% in 17 of 19 patients.
The osteotomy is stable by design;
however, fixation is recommended
to prevent transfer metatarsalgia,
malunion, nonunion, and recur-
rence.4 Boyer and Deorio20 re-
ported a 90% satisfaction rate in
their series on distal chevron oste-
otomy with single bioabsorbable
pin fixation for bunionette defor-
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
mity correction. All patients ach-
ieved radiographic union at 12
weeks; no nonunion or malunion
was reported.
Techniques for minimally invasive
osteotomy with or without fixation
continue to emerge, with good to
excellent results.25-30 In a study of in
31 feet, Waizy et al31 described
a minimally invasive approach to
distal fifth metatarsal osteotomy
with Kirschner wire (K-wire) fixa-
tion after oblique medial displace-
ment osteotomy. The mean four to
five IMA was reduced from 12° to
7.5°, and one patient sustained a pin
tract infection.31 Magnan et al32
reported the case of 30 consecutive
percutaneous distal osteotomies of
the fifth metatarsal for bunionette
deformity correction and reported
improvement of the average Ameri-
can Orthopaedic Foot and Ankle
Society lesser toe score from 51.9
preoperatively to 98.4 postopera-
tively at a mean follow-up of
96 months. Concerns of inferior
outcomes after minimally invasive
correction of type II and III de-
formities are possibly the result of an
inability to address the underlying
pathology that caused bunionette
deformity, bowing, and increased
IMA. Translation of .50% of the
metatarsal neck width can result in
instability, malunion, and unstable
surgical fixation.33
Diaphyseal Fifth Metatarsal
Osteotomy
In patients with an increased four to
five IMA or substantial lateral bow-
ing of the metatarsal shaft, diaphy-
seal fifth metatarsal osteotomy is
indicated.34 Historically, adverse
events, such as malrotation, non-
union, and malunion, have been
associated with transverse osteoto-
mies. Surgeons favor stable oste-
otomy constructs adapted from
those used to manage hallux valgus
deformity. Coughlin4 demonstrated
Glenn Guangyu Shi, MD, et al
Figure 4
AP radiograph of the right foot after
lateral eminence resection in an 85-
year-old woman. She had no
subluxation or recurrence.
the success of long oblique Ludloff
osteotomies combined with lateral
condylectomy and soft-tissue bal-
ancing in 93% of their study pop-
ulation. Several authors modified the
oblique osteotomy to be more bio-
mechanically stable with reverse
Ludloff and scarf osteotomies with a
success rate similar to that of the
standard Ludloff osteotomy.33,35,36
In a recent case series, 16 patients
underwent a reverse Ludloff-type
osteotomy for symptomatic type II
and III bunionette deformities.37
After a mean follow-up of 41.9
months, 15 of 16 patients reported
satisfactory outcomes, with an
average final American Orthopaedic
Foot and Ankle Society lesser toe
score of 86.6 points. MTPAs and
IMAs were considerably reduced
from 24.9° preoperatively to 4.3°
postoperatively and 13.2° preoper-
atively to 5.2° postoperatively,
respectively. One patient reported
dissatisfaction because of a persistent
fifth toe contracture.37
e399
Management of Bunionette Deformity
Figure 5
AP radiograph of the left foot after
fifth metatarsal head resection for a
failed symptomatic bunionette
correction in a 64-year-old woman.
Proximal Fifth Metatarsal
Osteotomy
Proximal or base fifth metatarsal os-
teotomies were designed to address
increased four to five IMAs. Re-
cently, Okuda et al15 reported on a
series of 10 patients who underwent
proximal third osteotomy of the fifth
metatarsal for bunionette correction
with sustained correction of a large
IMA (12.2° preoperatively to 4.8°
postoperatively). Osteotomies in this
region lost favor because of concerns
with nonunion, given the tenuous
blood supply at 2 cm proximal to the
fifth metatarsal base.
Metatarsal Head Resection
Metatarsal head resection is often
considered a salvage option to
e400
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Figure 6
AP radiograph of a right foot
demonstrating metatarsophalangeal
joint subluxation, which is a
complication of lateral eminence
resection. Other complications
include early arthritis and instability.
address painful bunionette deform-
ities (Figure 5). Concerns with this
procedure are related to MTPJ
instability, flail toe deformity, and
transfer metatarsalgia. Kitaoka and
Holiday38 reported on a series of 11
feet in seven patients who underwent
fifth metatarsal head resection with
an average follow-up of 9.1 years.
Poor results were seen in seven of 11
feet, with a 64% complication rate
reported. Common complications
included transfer metatarsalgia and
persistent painful fifth toe. Fifth
metatarsal head resection still has a
role in rheumatoid forefoot re-
constructions and in the management
of failed osteotomies, infections, and
neuropathy in patients with impend-
ing ulceration who cannot adhere to
Journal of the American Academy of Orthopaedic Surgeons
the immobilization required after
osteotomy.39
Complications
The overall reported complication
rate after surgical bunionette defor-
mity correction is low. Complications
including delayed wound healing,
malunion, nonunion, transfer metatar-
salgia, and recurrence vary depending
on the specific procedure and the im-
plants used. Recurrence can occur
after bunionette deformity correction
regardless of the technique used. When
the procedure chosen does not address
the underlying deformity, the patient is
at a higher risk for recurrence. Sublux-
ation is unique to undercorrection and
poor capsular imbrication, and dislo-
cation can occur with overcorrection or
undercorrection of bunionette defor-
mity (Figure 6). In addition, over-
resection of the lateral condyle can
produce enough instability to cause
immediate or delayed dislocation.
Valgus fifth toe is a poorly tolerated
condition in shoe-wearing societies,
when the fifth metatarsal phalangeal
joint becomes unstable after osteotomy
and medial soft-tissue release. Transfer
metatarsalgia can occur when the fifth
metatarsal is shortened, elevated, or
resected. Implant-related complica-
tions are most commonly reported
with the use of K-wire, which can
cause soft-tissue irritation and pin tract
infections.21,25,30,40
Authors’ Preferred Surgical
Approach
After diagnosis of bunionette defor-
mity, it is important to determine the
exact location of the pain: lateral,
plantarlateral, or plantar. Radio-
graphs can help the surgeon identify
and classify the deformity.
We reserve lateral condyle resec-
tion for a type I bunionette deformity
with painful lateral callosity. A 3-cm
longitudinal incision is made lateral

to the fifth MTPJ. Dissection isolates
and retracts the dorsolateral cutane-
ous nerve. Lateral capsulotomy is
made parallel to the skin incision to
expose lateral eminence. A micro-
sagittal saw is used to remove the
lateral eminence in line with the fifth
metatarsal shaft. Depending on the
preexisting medial capsular contrac-
ture, intra-articular pie crusting of the
medial capsule can be done after joint
distraction. Capsular repair is per-
formed, followed by skin closure,
after which a forefoot compression
dressing is applied. Sutures are
removed 2 weeks postoperatively.
Patients can bear weight in a postop-
erative shoe for a total of 4 to 6 weeks
before transitioning to regular shoes.
We routinely use distal chevron
osteotomy for mild type II bunionette
deformity. A lateral incision is made
similar to the approach used for lat-
eral eminence resection. Lateral emi-
nence resection is performed in line
with the metatarsal shaft. A 0.045
K-wire is placed in the center of the
metatarsal head. This pin can be left
in place and checked under fluoros-
copy to evaluate the direction of
translation. Then, 60° distal chevron
osteotomy is performed. Medial
translation can be performed on no
more than 50% of the metatarsal
head width before instability risks
arise. Fixation is completed with a
2-mm snap-off screw (Figure 7).
Fluoroscopic examination is per-
formed to ensure that the degree of
correction is adequate before remo-
val of the lateral prominence after
the displacement osteotomy. Cap-
sular repair is performed, followed
by skin closure. For distal osteoto-
mies, patients tolerate bearing
weight through the heel in a post-
operative shoe for 4 to 6 weeks.
Patients who present with more
pronounced symptomatic type II and
all type III deformities are candidates
for correction of the deformity
using a longitudinal diaphyseal oste-
otomy of the fifth metatarsal. Differ-
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Glenn Guangyu Shi, MD, et al
Figure 7
A, Intraoperative photograph of a foot demonstrating a distal chevron osteotomy
to correct a bunionette deformity. B, Postoperative AP radiograph of a foot
demonstrating healing after a successful distal chevron osteotomy that was
transfixed with a 2.0-mm snap-off screw. (Panel A courtesy of Mark E. Easley,
MD, Durham, NC.)
ent methods of performing this metatarsal shaft (Figure 8, B). The
osteotomy have been described.4,34 diaphyseal osteotomy begins dor-
We prefer the reverse Ludloff distal- sally just proximal to the metatarsal
dorsal to proximal-plantar osteotomy head and continues in a straight
with some modifications.4,34,35,37 We lateral-to-medial direction obliquely
often select this osteotomy for type II to the plantar proximal aspect of the
and III deformities for its variable fifth metatarsal diaphysis (Figure 8,
alignment corrective power and the C and D). The angle of the oste-
ability of the surgeon to control length, otomy must be judged carefully at
rotation, elevation, and depression this point to ensure that there is no
while preserving the blood supply. excessive elevation or depression of
An incision is made along the the metatarsal head (Figure 8, E).
straight lateral aspect of the fifth Some elevation may be favorable in
MTPJ, extending along the subcuta- patients who also have plantar cal-
neous border to within 1 cm of the losity and pain. Just before com-
proximal metaphyseal-diaphyseal pleting osteotomy, a 2-mm cortical
junction (Figure 8, A). Care is screw or K-wire is placed approxi-
taken to prevent injury to the mately midway along the shaft but is
dorsolateral cutaneous branch of not tightened completely to allow a
the sural nerve. The abductor rotation point and to prevent short-
digiti minimi muscle may need to be ening. If osteotomy is completed
elevated to visualize the lateral cor- without this provisional fixation,
tex. The fourth dorsal interosseous then the osteotomy can be difficult to
muscle medial to the fifth metatarsal later control with a bone clamp and
shaft is preserved. The capsule of the fix with screws because of the nar-
fifth MTPJ is incised in line with row diaphysis in some patients. After
the skin incision and reflected on the the osteotomy is completed, rotation
plantar aspect, but the capsule is left is achieved by pushing the distal
attached dorsally. The lateral emi- fragment medially and stabilized
nence is shaved at the edge of with a K-wire to prevent rotation
the articular cartilage flush with the (Figure 8, F and G). The screw is then
e401
Management of Bunionette Deformity

Figure 8

A, Preoperative clinical photograph of a foot showing the marking for a lateral incision along the fifth metatarsophalangeal
joint for longitudinal diaphyseal osteotomy to correct a type II deformity. B, Intraoperative photograph showing a
microsagittal saw being used to shave the lateral eminence flush with the fifth metatarsal shaft. C, Intraoperative
photograph showing the diaphyseal osteotomy line, which is made along the fifth metatarsal shaft. D, AP fluoroscopic
image of a foot showing guidewires that are inserted to identify the length of osteotomy. E, Intraoperative photograph
showing the use of a microsagittal saw in plane with the guidewires to avoid elevation or plantar flexion of the metatarsal
head. F, After completion of the proximal half of the metatarsal shaft osteotomy, a temporary Kirschner wire is placed
rather than a 2.0-mm screw to secure the proximal half of the osteotomy before completion of the distal half of the
osteotomy, preventing the surgeon from losing control of bony stability during manipulation. Subsequently, the metatarsal
head is shifted and pinned in place. G, AP fluoroscopic image of the foot demonstrating adequate correction.
Intraoperative photographs of a foot showing insertion of two 2.0-mm screws into the metatarsal shaft (H) and the use of a
microsagittal saw to remove the overhanging edges of the osteotomy (I). Postoperative AP weight-bearing (J) and lateral
(K) radiographs of a foot showing correction and fixation at 12 weeks.

tightened and the correction assessed
with fluoroscopy. Overcorrection or
undercorrection can be changed by
simply loosening the screw and re-
tightening it after the desired align-
ment is achieved.
A second screw is placed either
proximal or distal to the first screw
and is angled to provide another bi-
cortical fixation point, replacing the
provisional K-wire (Figure 8, H).
Careful shaving or smoothing of the
sharp edges of the osteotomy may be
necessary (Figures 8, I).
Care should also be taken to avoid
overpenetration of the plantar cortex
with the screws because it could
potentially cause weight-bearing
pain. The capsule of the MTPJ is
then repaired using absorbable su-
tures with the toe held at neutral,
advancing the plantar capsule with
the abductor digiti minimi slightly
dorsally. Final AP, lateral, and obli-
que intraoperative fluoroscopic im-
ages are obtained. Routine skin
closure is then completed with nylon
sutures.
Unlike simple lateral eminence
resection or distal osteotomy, feet are
splinted for 2 weeks after diaphyseal
osteotomies until the skin sutures are
removed, and then, a short leg cast is

e402 Journal of the American Academy of Orthopaedic Surgeons

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.

worn for 2 to 4 more weeks. Patients
are allowed to bear weight in a post-
operative shoe at 6 weeks after
surgery. The osteotomy may be
visualized radiographically for 3 to
4 months, but the patient may return
to normal activities in supportive
shoes in 8 to 10 weeks. Impact exer-
cise is usually tolerated by 12 weeks
(Figure 8, J and K).
Summary
Most painful bunionette deformities
respond to nonsurgical management.
When initial nonsurgical methods
have failed to relieve the symptoms,
surgical options allow for deformity
correction, with low reported com-
plication rates. Clinical and radio-
graphic evaluation guide surgical
intervention. Newer minimally inva-
sive techniques with expanded cor-
rective power have been reported,
with promising results.27-29 Type I
deformity is still managed with lat-
eral eminence resection with or
without distal osteotomy. Types II
and III can be managed with either
distal or diaphyseal osteotomy.
References
References printed in bold type are
those published within the past 5
years.
1. Coughlin MJ: Etiology and treatment of the
bunionette deformity. Instr Course Lect
1990;39:37-48.
2. Steinke MS, Boll KL: Hohmann-Thomasen
metatarsal osteotomy for Tailor’s bunion
(bunionette). J Bone Joint Surg Am 1989;
71:423-426.
3. Diebold PF: Basal osteotomy of the fifth
metatarsal for the bunionette. Foot Ankle
1991;12:74-79.
4. Coughlin MJ: Treatment of bunionette
deformity with longitudinal diaphyseal
osteotomy with distal soft tissue repair.
Foot Ankle 1991;11:
195-203.
5. Koti M, Maffulli N: Bunionette. J
Bone Joint Surg Am 2001;83-A:
1076-1082.
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
6. Steel MW III, Johnson KA, DeWitz MA,
Ilstrup DM: Radiographic measurements of
the normal adult foot. Foot Ankle 1980;1:
151-158.
7. Shimobayashi M, Tanaka Y, Taniguchi A,
Kurokawa H, Tomiwa K, Higashiyama I:
Radiographic morphologic characteristics
of bunionette deformity. Foot Ankle Int
2016;37:320-326.
8. Fallat LM: Pathology of the fifth
ray, including the Tailor’s bunion
deformity. Clin Podiatr Med Surg 1990;
7:689-715.
9. Nestor BJ, Kitaoka HB, Ilstrup DM,
Berquist TH, Bergmann AD: Radiologic
anatomy of the painful bunionette. Foot
Ankle 1990;11:6-11.
10. Grice J, Marsland D, Smith G, Calder J:
Efficacy of foot and Ankle
corticosteroid injections. Foot Ankle Int
2017;38:8-13.
11. Shrum DG, Sprandel DC, Marshall H:
Triplanar closing base wedge osteotomy for
Tailor’s bunion. J Am Podiatr Med Assoc
1989;79:124-127.
12. Weitzel S, Trnka HJ, Petroutsas J:
Transverse medial slide osteotomy for
bunionette deformity: Long-term results.
Foot Ankle Int 2007;28:794-798.
13. Moran MM, Claridge RJ: Chevron
osteotomy for bunionette. Foot Ankle Int
1994;15:684-688.
14. Shereff MJ, Yang QM, Kummer FJ, Frey
CC, Greenidge N: Vascular anatomy of the
fifth metatarsal. Foot Ankle 1991;11:
350-353.
15. Okuda R, Kinoshita M, Morikawa J,
Jotoku T, Abe M: Proximal dome-
shaped osteotomy for symptomatic
bunionette. Clin Orthop Relat Res 2002:
173-178.
16. Kitaoka HB, Holiday AD Jr: Lateral
condylar resection for bunionette. Clin
Orthop Relat Res 1992:183-192.
17. Sponsel KH: Bunionette correction by
metatarsal osteotomy: Preliminary
report. Orthop Clin North Am 1976;7:
809-819.
18. Leach RE, Igou R: Metatarsal osteotomy
for bunionette deformity. Clin Orthop
Relat Res 1974:171-175.
19. Haber JH, Kraft J: Crescentic osteotomy for
fifth metatarsal head lesions. J Foot Surg
1980;19:66-67.
20. Boyer ML, Deorio JK: Bunionette
deformity correction with distal chevron
osteotomy and single absorbable pin
fixation. Foot Ankle Int 2003;24:
834-837.
21. Frankel JP, Turf RM, King BA: Tailor’s
bunion: Clinical evaluation and correction
by distal metaphyseal osteotomy with
Glenn Guangyu Shi, MD, et al
cortical screw fixation. J Foot Surg 1989;
28:237-243.
22. Kitaoka HB, Leventen EO: Medial
displacement metatarsal osteotomy
for treatment of painful bunionette.
Clin Orthop Relat Res 1989:
172-179.
23. Kitaoka HB, Holiday AD Jr, Campbell
DC II: Distal Chevron metatarsal
osteotomy for bunionette. Foot Ankle
1991;12:80-85.
24. Cooper MT, Coughlin MJ: Subcapital
oblique osteotomy for correction of
bunionette deformity: Medium-term
results. Foot Ankle Int 2013;34:
1376-1380.
25. Legenstein R, Bonomo J, Huber W,
Boesch P: Correction of Tailor’s bunion
with the Boesch technique: A
retrospective study. Foot Ankle Int 2007;
28:799-803.
26. White DL: Minimal incision approach
to osteotomies of the lesser metatarsals:
For treatment of intractable
keratosis, metatarsalgia, and Tailor’s
bunion. Clin Podiatr Med Surg 1991;8:
25-39.
27. Giannini S, Faldini C, Vannini F, Digennaro
V, Bevoni R, Luciani D: The minimally
invasive osteotomy “S.E.R.I.” (simple,
effective, rapid, inexpensive) for correction
of bunionette deformity. Foot Ankle Int
2008;29:282-286.
28. Michels F, Van Der Bauwhede J, Guillo S,
Oosterlinck D, de Lavigne C: Percutaneous
bunionette correction. Foot Ankle Surg
2013;19:9-14.
29. Laffenetre O, Millet-Barbe B, Darcel V,
Lucas YHJ, Chauveaux D: Percutaneous
bunionette correction: Results of a
49-case retrospective study at a mean
34 months’ follow-up. Orthop
Traumatol Surg Res 2015;101:
179-184.
30. Radl R, Leithner A, Koehler W, Scheipl S,
Windhager R: The modified distal
horizontal metatarsal osteotomy for
correction of bunionette deformity. Foot
Ankle Int 2005;26:454-457.
31. Waizy H, Olender G, Mansouri F,
Floerkemeier T, Stukenborg-Colsman C:
Minimally invasive osteotomy for
symptomatic bunionette deformity is
not advisable for severe deformities: A
critical retrospective analysis of
the results. Foot Ankle Spec 2012;5:
91-96.
32. Magnan B, Samaila E, Merlini M, Bondi
M, Mezzari S, Bartolozzi P:
Percutaneous distal osteotomy of the
fifth metatarsal for correction of
bunionette. J Bone Joint Surg Am 2011;
93:2116-2122.
33. Maher AJ, Kilmartin TE: Scarf osteotomy
for correction of Tailor’s bunion: Mid- to
e403
Management of Bunionette Deformity
long-term followup. Foot Ankle Int 2010;
31:676-682.
34. London BP, Stern SF, Quist MA, Lee RK,
Picklesimer EK: Long oblique distal
osteotomy of the fifth metatarsal for
correction of Tailor’s bunion: A
retrospective review. J Foot Ankle Surg
2003;42:36-42.
35. Guha AR, Mukhopadhyay S, Thomas RH:
“Reverse” scarf osteotomy for bunionette
correction: Initial results of a new surgical
e404
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
technique. Foot Ankle Surg 2012;18:
50-54.
36. Glover JP, Weil L Jr, Weil LS Sr: Scarfette
osteotomy for surgical treatment of
bunionette deformity. Foot Ankle Spec
2009;2:73-78.
37. Waizy H, Jastifer JR, Stukenborg-
Colsman C, Claassen L: The reverse
Ludloff osteotomy for bunionette
deformity. Foot Ankle Spec 2016;9:
324-329.
Journal of the American Academy of Orthopaedic Surgeons
38. Kitaoka HB, Holiday AD Jr: Metatarsal
head resection for bunionette: Long-term
follow-up. Foot Ankle 1991;11:345-349.
39. Reize P, Leichtle CI, Leichtle UG, Schanbacher
J: Long-term results after metatarsal head
resection in the treatment of rheumatoid
arthritis. Foot Ankle Int 2006;27:586-590.
40. Friend G, Grace K, Stone HA: L-
osteotomy with absorbable fixation
for correction of Tailor’s bunion.
J Foot Ankle Surg 1993;32:14-19.
 Review Article

Surgical Management of
Patellofemoral Instability in the
Skeletally Immature Patient

Lauren H. Redler, MD
Margaret L. Wright, MD
Downloaded from http://journals.lww.com/jaaos by BhDMf5ePHKbH4TTImqenVL8/J7OdHJl5KE3NkejjbT4OSOzvI6kP1JNPXvI1w16Ut5XXzWOLXt8= on 10/01/2018
Abstract
Compared with skeletally mature patients, skeletally immature
patients are at a higher risk of acute traumatic patellar dislocation.
Surgical treatment is the standard of care for patients with recurrent
instability and requires important and technically challenging physeal

considerations. Physeal-sparing medial patellofemoral ligament

reconstruction is the treatment of choice for these patients, replacing
older nonanatomic extensor mechanism realignment techniques.
Implant-mediated guided growth is an important adjunct to correct
genu valgum angular deformities that contribute to patellar instability.
Patient-specific surgical techniques and proper surgical indications
are crucial for successful outcomes.

and recurvatum deformity. Trochlear

JAAOS Plus Webinar
Join Dr. Redler and Dr. M. Wright for
the interactive JAAOS Plus Webinar
discussing “Surgical Management of
Patellofemoral Instability in the
Skeletally Immature Patient,” on
Tuesday, October 16, 2018, at 8 PM
Eastern Time. The moderator will be
Rick W. Wright, MD, the Journal’s
Deputy Editor for Sports Medicine
topics. Sign up now at www.aaos.org/
coursecalendar/.
From the Department of Orthopedic
Surgery, Columbia University Medical
Center, New York, NY.
Neither of the following authors nor
any immediate family member has
received anything of value from or has
stock or stock options held in a
commercial company or institution
related directly or indirectly to the
subject of this article: Dr. Redler and
Dr. Wright
J Am Acad Orthop Surg 2018;26:
e405-e415
DOI: 10.5435/JAAOS-D-17-00255
Copyright 2018 by the American
Academy of Orthopaedic Surgeons.
Natural History and Risk
Factors
Patellofemoral dislocation is com-
mon in the pediatric population
and occurs at a rate of 29 per
100,000 patients between the ages of
10 and 17 years.1-3 Nearly 70% of
dislocations happen during sports
activities or dancing.3 Previous patel-
lar dislocation, ligamentous laxity,
open physes, and trochlear dysplasia
all represent risk factors for recur-
rence, with rates as high as 69%.4
The pathoanatomy of patello-
femoral instability has many factors
that are challenging to modify in the
skeletally immature patient. An ele-
vated tibial tubercle to trochlear
groove (TT-TG) distance creates a
lateral vector that contributes to
patellar instability. Because of the
open apophysis, this cannot be sur-
gically addressed with an osteotomy
in the skeletally immature patient.
Similarly, patella alta cannot be
treated with a distalizing osteotomy
because of the risk of growth arrest
dysplasia is the strongest individual
radiographic risk factor for instabil-
ity, which can be evaluated using
MRI or a lateral radiograph of the
knee.5,6 Although they are gaining
popularity, trochleoplasties are con-
traindicated in skeletally immature
patients because of the risk of injury to
the open distal femoral physis. In
contrast, genu valgum is an important
modifiable risk factor in skeletally
immature patients; it may be treated
with implant-mediated guided growth
before skeletal maturity and can
reduce the risk of future dislocations.
Classification
Although there is no universal
classification of patellofemoral in-
stability in pediatric patients, a
proposed classification that may be
useful for surgical planning includes
four groups: syndromic, obligatory,
fixed, and traumatic dislocations.2
Surgical indications differ in these
patients.

October 1, 2018, Vol 26, No 19 e405

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Patellofemoral Instability in Skeletally Immature Patient
Figure 1
AP radiograph of a skeletally immature
knee showing the undulating course of
the distal femoral physis.
Syndromic dislocation occurs in pa-
tients with syndromes associated with
ligamentous laxity or osseous defor-
mity (eg, Marfan syndrome, Ehlers-
Danlos, Down syndrome). Allografts
should be used for medial patello-
femoral ligament (MPFL) reconstruc-
tion in this group.7 Obligatory
dislocation results from tight lateral
structures and occurs every time the
knee is flexed, but reduces in knee
extension. Fixed dislocation cannot be
reduced even with knee extension.
Obligatory and fixed dislocations may
require additional soft-tissue proce-
dures for stabilization beyond the lat-
eral release and MPFL reconstruction
techniques described in this review.2
Surgical Indications
Skeletally immature patients who
sustain an initial, acute patellar dis-
location without loose osteochondral
e406
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
fragments have not been shown to
benefit from surgical treatment.8,9
Current standard treatment includes
activity restriction, patellar taping
or bracing, and physical therapy
focused on stretching the iliotibial
band and strengthening the vastus
medialis oblique (VMO), gluteal
muscles, and core. Osteochondral
injuries occur in as many as 75% of
pediatric patients with acute patellar
dislocation.10 The osteochondral
lesion is typically on the medial facet
of the patella or lateral femoral con-
dyle. Surgery is indicated in patients
with loose osteochondral fractures to
avoid a mechanical block to motion
and development of early chondral
wear. Lesions in non–weight-bearing
portions of the knee or irreparable
osteochondral fragments may be
excised, but large fragments from
weight-bearing surfaces have high
healing capacity and should be re-
paired. There are also reports of large,
chondral only fragments healing after
fixation in adolescent patients, and
these may benefit from fixation.11
Patellar stabilization is indicated in
patients with recurrent instability;
49% of patients with recurrent
instability who are treated non-
surgically will have further instability,
as opposed to a 4% of those treated
with MPFL reconstruction.3,12 Eighty
percent of patients return to their
preinjury activity level after MPFL
reconstruction compared to only
52% of patients treated nonsurgically
or with MPFL repair.8,13 Patients
with recurrent instability have lower
short- and long-term outcome scores
than those treated with MPFL
reconstruction, as well as increased
progression of patellofemoral carti-
lage erosion compared to those
without recurrence. 14-16 MPFL
reconstruction alone can be per-
formed successfully in pediatric
patients who have recurrent insta-
bility associated with trochlear
dysplasia and no other structural
deformities.17 It is important to
Journal of the American Academy of Orthopaedic Surgeons
recognize the features that affect the
risk of failure of surgical intervention
and recurrent instability, including
patella alta, elevated TT-TG, and
trochlear dysplasia. Although these
factors guide patient counseling, they
rarely affect surgical planning in this
age group. Surgeons should ensure that
children have reached skeletal maturity
and physeal closure before considering
tibial tubercle (TT) osteotomy.
Surgical Techniques
Surgical Considerations
In the pediatric population, a thor-
ough understanding of the distal
femoral and proximal tibial physes,
as well as the TT apophysis, is critical
when considering the multitude of
available surgical techniques. The
distal femoral physis has a charac-
teristic undulating structure with
relatively proximal medial and lateral
borders (Figure 1). It is the largest
and fastest growing physis in the
body and contributes 70% of the
femoral length and 37% of overall
lower limb growth, which amounts
to approximately 1 cm per year
during skeletal immaturity. This
growth plate fuses between the ages
of 14 and 16 years in females and 16
and 18 years in males. The proximal
tibial physis contributes approxi-
mately 55% of the length of the tibia
and 25% of the length of the entire
limb. On average, the proximal tibial
physis contributes 0.65 cm of growth
per year. This physis fuses between
the ages of 13 and 15 years in fe-
males and 15 and 19 years in males.
The TT apophysis fuses between the
ages of 13 and 15 years in females
and 15 and 19 years in males. The
patella begins to ossify at age 3 years
in females and 4 to 5 years in males.
Growth disturbance as a result of
injury of the patellar physis after
MPFL reconstruction has not been
reported. However, reconstruction
technique and patellar fixation
 Lauren H. Redler, MD and Margaret L. Wright, MD

Figure 2

A, AP intraoperative fluoroscopic image showing drilling of the femoral socket away from the physis. B, Lateral intraoperative
fluoroscopic image showing the drill located at the Schottle point. C, Postoperative coronal magnetic resonance image
showing the location of the interference screw on the distal medial femur distal to the physis. (Courtesy of Beth E. Shubin
Stein, MD, New York, NY.)

should be carefully considered in
young patients and is discussed in
more detail later.
The femoral insertion of the MPFL
is variable in pediatric patients,
although radiographic and cadaver
studies have found that in most pa-
tients, the midpoint of the femoral
attachment is just distal to the
physis.18,19 Grafts placed proximal
to the physis have the unique com-
plication of proximal migration of
the insertion after reconstruction and
high tension across the physis.
Fluoroscopic guidance is mandatory
to avoid physeal violation and to
determine appropriate tunnel place-
ment. Care should be taken to con-
firm that the trajectory of the guide
pin is completely distal to the physis
on the AP view, given the concave
shape of the distal femoral physis20
(Figure 2). Although growth of the
distal femoral epiphysis arises from
the secondary subchondral growth
plate, and is theoretically at risk of
injury during creation of tunnels in
the epiphysis, the concern for teth-
ering of the distal femoral physis
outweighs this issue, and thus, epi-
physeal tunnels remain the location
of choice. It is also important to
consider the perichondrium of the
physis because it is sensitive to in-
jury. Caution must be exercised to
avoid violation of the perichondrium
during both surgical dissection and
tunnel placement.
Distal Realignment
Procedures
Distal realignment procedures include
the Modified Roux-Goldthwait, Ga-
leazzi, and Nietosvaara techniques.
These procedures are often combined
with proximal realignments, includ-
ing medial imbrication and lateral
retinacular release.
The modified Roux-Goldthwait
procedure is performed by detach-
ing the lateral half of the patellar
tendon from the TT and passing it
medially under the patellar tendon.
The tendon is sutured to the medial
tissues and periosteum at the junction
of the medial TT and pes anserine
insertion. A lateral release, sometimes
with medial imbrication, is also per-
formed. In an early study of 30 knees
that underwent the procedure, 26
knees had excellent results based on
the Insall criteria21 (Figure 3).
The Galeazzi semitendinosus
tenodesis is performed by harvesting
the semitendinosus tendon, leaving it
attached at its distal insertion, and
sewing it in an oblique manner to the
inferomedial patella. A 2012 study of
34 knees treated with the procedure
found that 35% required a second
surgery and 82% had recurrent sub-
luxation or dislocation, so it has
largely been abandoned for more
effective and anatomic reconstruc-
tion procedures22 (Figure 4).
Nietosvaara et al23 described a
more anatomic modification of
the Galeazzi distal reconstruction,
in which the semitendinosus and
gracilis are left attached at the pes
insertion, passed through a medial
longitudinal patellar bone tunnel
from inferior to superior, and then
fixed to the femoral insertion of the
MPFL with an interference screw.
Although this technique seems
promising in the original case series,
long-term outcome studies have not
yet been performed23 (Figure 5).

October 1, 2018, Vol 26, No 19 e407

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Patellofemoral Instability in Skeletally Immature Patient

Figure 3

Schematic drawing of the modified Roux-Goldthwait distal realignment technique. (Reproduced with permission from Weeks
KD III, Fabricant PD, Ladenhauf HN, Green DW: Surgical options for patellar stabilization in the skeletally immature patient.
Sports Med Arthrosc Rev 2012;20[3]:194-202.)

Soft-tissue Procedures
MPFL Repair
The MPFL is the primary restraint
against lateral patellar translation
and is stretched or torn in dislocation
events, most commonly at the patel-
lar origin.19 Direct MPFL repair does
not require patellar bone tunnels
and poses minimal risk of femoral
physeal injury. However, biomechan-
ical tests of MPFL repair show weak-
ness compared to the native ligament
and MPFL reconstruction.24 Pa-
tients who undergo MPFL repair
are more likely than those who
undergo reconstruction to have
recurrent dislocation, and they are
nearly as likely as patients who are
treated nonsurgically to have recur-
rent dislocation.23,25,26 MPFL repair
is therefore only considered in pa-
tients with an initial dislocation
who are undergoing a procedure for
additional injuries (ie, osteochondral
repair).
Hamstring MPFL Reconstruction
MPFL reconstruction is favored in
patients with recurrent instability
because the chronically injured
medial retinacular structures are
insufficient to prevent recurrent dis-
location. Anatomic MPFL recon-
struction procedures for skeletally
immature patients have been de-
scribed, with multiple fixation tech-
niques available in the patella and
femur to avoid injury to the physis. A
recent meta-analysis found that anat-
omic grafts have the lowest recur-
rence rate when a double-limb graft
configuration is used.7 Similar out-
come scores and rates of recurrent
instability have been reported in
both autograft and allograft ham-
string reconstruction in skeletally
immature patients.7,20,27
Patellar fixation using bone tun-
nels, docking technique, interference
screw, or suture anchor fixation has
been described.20,28 Creation of a
bony sulcus and suture anchor fix-
ation are gaining favor because they
avoid the risk of patella fracture
(Figure 6). In addition, if cartilage
restoration procedures (eg, osteo-
chondral fracture fixation, OATS
[osteochondral allograft transfer sys-
tem], minced chondral allograft) are
concurrently performed on the patella,
creation of bone tunnels in the patella
increases the possibility of the tunnels
communicating and compromising the

e408 Journal of the American Academy of Orthopaedic Surgeons

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
 Lauren H. Redler, MD and Margaret L. Wright, MD

repair. The graft is placed between the Figure 4

capsule and medial retinaculum and
fixed to the Schottle point at 30°
flexion using an interference screw,
suture button, or suture anchor tech-
nique. A schematic demonstrating
hamstring MPFL reconstruction is
shown in Figure 7, F.

Quadriceps and Patellar Tendon

Reconstruction
Additional options beyond free ten-
don grafting include patellar and
quadriceps tendon grafts, which
eliminate patellar fixation, thus
obviating the risk of patella fracture
from bone tunnels that place the
proportionally smaller patellar at
higher risk. A 10- to 12-mm slip of the
middle third of the superficial quad
tendon is dissected and left attached
at its insertion on the proximal
patella. The tendon is subperiosteally
dissected more distally on the lateral
side so that when rotated, the attach-
ment is primarily at the superomedial
Schematic drawing of the Galeazzi semitendinosus tenodesis distal realignment
patella. A stay suture is placed to technique. (Reproduced with permission from Weeks KD III, Fabricant PD,
maintain an anatomic patellar origin, Ladenhauf HN, Green DW: Surgical options for patellar stabilization in the
and then it is passed through the sub- skeletally immature patient. Sports Med Arthrosc Rev 2012;20[3]:194-202.)
vastus space and fixed to the medial
femur29 (Figure 7, A). A small series
has shown good outcomes with min- technique in young, skeletally imma- the MPFL and medial patellotibial
imal recurrence of dislocation in ture patients have not been reported ligament (MPTL).33 The superficial
skeletally mature patients.30 No long- and therefore may not apply to quadriceps tendon is harvested, and
term outcome studies have been per- pediatric patients. the MPFL reconstruction is per-
formed in pediatric patients. The use of a combined quadriceps formed as described earlier. The
Some authors have described the use and patellar tendon graft in an effort medial patellar tendon is also har-
of a pedicled medial patellar tendon to create adequate tension through- vested, leaving the proximal end
graft to reconstruct the MPFL31,32 out knee flexion has been described. attached to the inferior patella, and
(Figure 7, B). The medial third of the Setting the graft length to appropri- it is attached to the tibia more
patellar tendon is detached from the ately establish graft tension is a medially to reconstruct the medial
TT. The tendon is subperiosteally challenge in any MPFL reconstruc- patellotibial ligament. With two
dissected to the proximal third of the tion technique, because placing the points of fixation, there may be a
patella to maintain an anatomic femoral insertion of the graft too secondary restraint to graft loosen-
MPFL origin. The tendon is rotated, proximally or distally can increase ing. Hinckel et al33 have specifically
and a stay suture is placed to prevent tension in flexion and extension, advocated for the use of a combined
further detachment from the patella. respectively, causing stress on the quadriceps and patellar tendon
The graft is passed in the subvastus medial trochlea and patellar carti- graft in patients with flexion insta-
space and fixed to the femoral inser- lage. The combined quadriceps and bility, in children with unmodifiable
tion.32 Similar to the quadriceps patellar tendon graft uses both the anatomic risk factors, and for knee
tendon technique, outcomes of the superficial quadriceps tendon and hyperextension associated with gen-
pedicled medial patellar tendon the patellar tendon to reconstruct eralized ligamentous laxity. With this

October 1, 2018, Vol 26, No 19 e409

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Patellofemoral Instability in Skeletally Immature Patient

Figure 5 tendon is also an ideal graft for MPFL

reconstruction (Figure 7, C). The
medial two thirds of the adductor
magnus tendon are harvested, leav-
ing it attached to the adductor
tubercle, and the tendon is passed
between the capsule and medial
retinaculum and secured to the
superomedial patella. A variation of
this technique is to create a small
bone tunnel in the distal femur from
the MPFL insertion to the adductor
tubercle, through which the graft can
be passed to provide a more anat-
omic femoral insertion. Compared
with patellar and quadriceps tendon
MPFL reconstruction techniques, the
adductor tendon graft was found to
have similar outcomes and no re-
currences of patellar instability.34
An important technical point to
consider when performing this tech-
nique is that the tendon harvester
should be pointed medially when
harvesting the adductor tendon graft,
which minimizes the risk of injury to
Schematic drawing of the Nietosvaara distal realignment technique. the saphenous nerve and vessels
(Reproduced with permission from Weeks KD III, Fabricant PD, Ladenhauf HN, passing through the adductor canal.
Green DW: Surgical options for patellar stabilization in the skeletally immature In a cadaver study, Jacobi et al35
patient. Sports Med Arthrosc Rev 2012;20[3]:194-202.) explored the potential anatomic
dangers of this technique. They
Figure 6 found that the neurovascular bundle
of the adductor hiatus, the saphe-
nous nerve, and the saphenous
branch of the descending genicular
artery are typically approximately
10 cm away from the instruments
during the graft harvest.

Sling Procedures
In an effort to maximize the ability
to create an anatomic graft while
minimizing the risk of femoral phys-
Intraoperative photographs demonstrating the placement of suture anchors (A) eal injury, the use of sling procedures
for fixation of the hamstring graft (B) to the medial border of the patella (asterisk) for MPFL reconstruction in pediatric
during medial patellofemoral ligament reconstruction. patients has been described. Monllau
et al36 described a technique using
gracilis autograft, in which the ad-
technique, it is important to consider Adductor Tendon Reconstruction ductor tendon insertion is used for
both the tibial and femoral physes Because of the proximity of the the femoral MPFL attachment point.
when determining the location of the adductor tubercle to the femoral Two tunnels are drilled through the
MPFL and MTFL fixation sites. insertion of the MPFL, the adductor patella in a V shape (Figure 7, E).

e410 Journal of the American Academy of Orthopaedic Surgeons

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
 Lauren H. Redler, MD and Margaret L. Wright, MD

Figure 7

Schematic drawing of several physeal-sparing medial patellofemoral ligament reconstruction techniques: hemiquadriceps
tendon transfer (A), hemipatellar tendon transfer (B), adductor tendon pedicle graft (C), hamstring graft using MCL as a
pulley (D), hamstring graft using adductor tendon as a pulley (E), and double-limb hamstring allograft using patellar and
femoral sockets (F). (Reproduced with permission from Gausden EB, Green DW: Medial patellofemoral ligament
reconstruction: Hamstring technique, in Cordasco FA, Green DW, eds: Pediatric and Adolescent Knee Surgery.
Philadelphia, PA, Wolters Kluwer, 2015, pp 140-147.)

The gracilis tendon is passed through
one bone tunnel, under the VMO,
around the adductor magnus tendon
near its femoral insertion, back
across the medial knee, and through
the second patellar tunnel, and the
ends are sutured together.
A similar technique has been
described using the femoral MCL
(medial collateral ligament) insertion.
A free hamstring graft is fixed to the
patella and then passed medially and
under the posterior third of the femoral
insertion of the MCL. The graft is then
looped back over the MCL and
sutured to the anterior patellar reti-
naculum.37 A variation of this tech-
nique is to leave the semitendinosus
attached to its pes insertion site. The
free end of the graft can then be
passed behind the femoral attachment
of the MCL and sutured to the ante-
rior patellar retinaculum. When per-
forming the MCL sling techniques,
care should be taken to avoid iatro-
genic injury to the MCL, although
no MCL injuries have been reported
in the literature to date (Figure 7, D).
Although these techniques are
appealing because they avoid tunnels

October 1, 2018, Vol 26, No 19 e411

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Patellofemoral Instability in Skeletally Immature Patient
Figure 8
Intraoperative photograph of the
knee demonstrating a combined
MPFL and MQTFL reconstruction.
MPFL = medial patellofemoral
ligament, MQTFL = medial
quadriceps tendon femoral ligament.
(Courtesy of Mininder S. Kocher,
MD, MPH, Boston, MA.)
on the medial femur and minimize the
risk of growth disturbance, results
have been mixed. One study found
a higher risk of patellar instability
after the adductor sling procedure
compared to anatomic techniques.38
A study of the adductor tendon
reconstruction compared to the
adductor sling procedure using
hamstring autograft found no dif-
ference in postoperative instability or
complications, but found that pa-
tients who had undergone the sling
procedure were more likely than
those who underwent reconstruction
using the adductor magnus tendon
to return to sports and had subjec-
tively better outcome scores.39 Longer
follow-up of the same patients found
that 11 of 15 had an excellent outcome
with the sling procedure and only one
required further surgery.39
Medial Quadriceps Tendon
Femoral Ligament
Reconstruction
Anatomic studies have shown vari-
ability in the patellar origin of the
MPFL, with some patients having
greater attachment to the VMO and
quadriceps tendon as opposed to
the patella directly.40 A separate
structure just superior to the
e412
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
MPFL, the medial quadriceps ten-
don femoral ligament (MQTFL),
has been identified. It originates
from the deep aspect of the quad-
riceps tendon and inserts superior
to the MPFL.41 MQTFL recon-
struction is an alternative to MPFL
reconstruction that avoids the use
of patellar tunnels.
MQTFL reconstruction is per-
formed with either semitendinosus
autograft or allograft. The graft is
fixed just anterior to the adductor
tendon insertion with an interference
screw or suture anchor. A second
parallel incision is made over the
medial patella and quadriceps ten-
don, and the graft is passed ex-
tracapsularly. The graft is sutured to
the distal aspect of the medial quad-
riceps tendon, and patellar tracking is
observed throughout range of motion
to evaluate graft tension. In a series of
17 skeletally mature patients who
underwent an MQTFL reconstruc-
tion for instability, none had recur-
rence at 1 year postoperatively.41 A
combination MPFL-MQTFL recon-
struction technique is gaining popu-
larity, sparking interest in renaming
it the medial patellofemoral complex
(Figure 8).
Lateral Release
Lateral release has not been found to
be effective when performed in iso-
lation, likely because it does not
improve patellar tracking. However,
in some cases, it may be an effective
addition to medial soft-tissue proce-
dures. Indications for a lateral release
in combination with medial proce-
dures include abnormal patellar tilt,
defined as the inability to evert the
patella to neutral, and TT-TG
distance .20 mm. It is important
to distinguish abnormal patellar tilt,
which represents true lateral tightness,
from lateral patellar tilt seen on
imaging in all cases of patellar insta-
bility as a result of injured incompe-
tent medial tissues. It is vital to avoid
Journal of the American Academy of Orthopaedic Surgeons
excessive lateral release because
medial instability after lateral release
has also been described. 42 Lateral
release may also be indicated in
patients with obligatory or fixed
patellar dislocation because these
patients have tight lateral restraints
that limit the ability to maintain
patellar reduction.2
Complications
Although MPFL reconstruction
techniques are currently the standard
of care for pediatric patients with
patellofemoral instability, complica-
tions remain a considerable chal-
lenge. One of the largest series of
young patients (aged ,21 years)
undergoing MPFL reconstruction
reported an overall complication
rate of 16.2%.12 The most common
complications were loss of complete
knee flexion and recurrent patello-
femoral instability. In the patients
with loss of flexion, all improved
with manipulation under anesthesia
and nearly all had anterior place-
ment of the femoral insertion point
identified on MRI.12 One study
found that the primary radiographic
risk factor for recurrent dislocation
was the severity of trochlear
dysplasia.43
Although specific methods of fixation
and graft type have not been associated
with complications after anatomic
reconstruction, patella fracture is a
known complication of the bone tun-
nels used for patellar fixation. Six cases
of patella fracture (4.3%) were identi-
fied in one series, and five of the six
required surgical fixation.44 Although
uncommon, this is an important
consideration when choosing the
technique of MPFL reconstruction,
especially in younger patients with a
less ossified patella and small margin
for error. This may sway the surgeon
to use techniques that do not require
patellar bone tunnels (quadriceps or
patellar tendon autograft or suture
anchor fixation).

Implant-mediated Guided
Growth
Indications
Guided growth techniques have been
used for many years to correct genu
valgum in children with open physes.
The goals of these procedures are to
maximize growth potential and to
avoid osteotomies after growth is
complete. Genu valgum is a known
risk factor for patellar instability, and
correction of genu valgum in isola-
tion or in addition to medial soft-
tissue procedures decreases the risk of
recurrent instability.45 Radiographs
should be obtained to confirm bone
age; a three-joint standing radiograph
of the lower extremities should
be obtained to assess remaining
growth and the degree of deformity.
Guided growth techniques are typi-
cally indicated in patients with genu
valgum .10° (defined by the lateral
distal femoral angle ,79°) that is
associated with patellar instability,
who have at least 6 months to 1 year
of remaining growth. Temporary
hemiepiphysiodesis through the
application of an extraphyseal ten-
sion band plate is a safe, effective,
and minimally invasive technique.
However, it should not be used in
children aged ,8 years because
spontaneous correction of the
deformity is likely in children this
young.46
Technique
Temporary hemiepiphysiodesis, or
tension band plating, is most often
done at the distal medial femur but
can also be applied to the proximal
medial tibia in cases in which both the
femur and tibia contribute to the
deformity. The surgical site is identi-
fied with fluoroscopy, and the pro-
cedure is performed through a 2-cm
skin incision or through the incision
over the femoral insertion of the
MPFL if the procedure is done con-
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Lauren H. Redler, MD and Margaret L. Wright, MD
Figure 9
A, Preoperative standing AP three-joint radiograph of the lower extremities
showing genu valgum of the left knee. B, Postoperative standing AP three-joint
radiograph of the lower extremities showing correction of genu valgum with distal
medial femoral and proximal medial tibial hemiepiphysiodesis.
currently with a medial soft-tissue until skeletal maturity with standing
procedure. Using blunt dissection, radiographs to confirm that no
the periosteum over the physis is overcorrection or other deformity
identified but not violated. The develops.
figure-of-8 plate is placed and held
provisionally with a hypodermic
needle or K-wire fixation. Fluoros- Technical Considerations
copy is used to confirm the position of
the plate over the midsagittal line of Physeal stapling and other forms of
the distal femoral physis on the lateral hemiepiphysiodesis have been used to
view, with one hole of the plate on correct genu valgum; however, they
each side of the physis. When posi- are not reversible and therefore must
tion is confirmed, K-wires are placed be used in older children who will
in each hole, and cannulated screws reach completion of growth before
are placed for definitive fixation. overcorrection. Figure-of-8 plates act
Weight bearing as tolerated may be as a tension band across the physis
permitted or weight bearing within and slow growth relative to the lateral
the limitations of any other proce- physis without creating permanent
dures that were performed concur- physeal bars or tethers.46 They can
rently.46 Patients are followed at therefore be used for growth modu-
3- to 4-month intervals with stand- lation techniques in younger chil-
ing three-joint radiographs to mea- dren, allowing for faster correction.
sure correction of the deformity Correction of approximately 0.7°
(Figure 9). The plates are removed per month in the femur and 0.5°
when correction is complete, and the per month in the tibia can be ex-
patient should continue to be fol- pected, for a combined average of
lowed at 4- to 6-month intervals 1.2° per month if used together.
e413
Patellofemoral Instability in Skeletally Immature Patient
Valgus alignment can improve an
average 8° per year with a femoral
hemiepiphysiodesis and an addi-
tional 4° per year with tibial hemi-
epiphysiodesis.46 Faster and greater
correction can be achieved when the
technique is used with two plates
instead of one and when it is per-
formed on younger children.46 Most
studies advocate for overcorrection
of 5° to account for the rebound
phenomenon. The plates should be
removed when correction is com-
plete to avoid overcorrection, al-
though plates may be placed on the
opposite side of the physis to reverse
this complication if it occurs.1
There are additional considerations
for when MPFL reconstruction and
guided growth procedures are per-
formed together because of the prox-
imity of the medial femoral physis to
the femoral insertion of the MPFL. A
femoral plate can be placed through
the same incision used for the femoral
fixation of the MPFL reconstruction,
but should be placed before the fixa-
tion of the MPFL so that it sits directly
on the periosteum. The plate is placed
on the midsagittal line, with concur-
rent planning of the femoral MPFL
fixation so that the ideal tunnel loca-
tion is not blocked by the plate. Dur-
ing the second surgery for plate
removal, the dissection should be
performed carefully so that the re-
constructed ligament is not disrupted.
Summary
Patellar dislocation is the most
common acute knee disorder in
children and adolescents.3 Predis-
posing factors are multifactorial.
Nonsurgical treatment is typically
used to treat an initial acute patellar
dislocation in a skeletally immature
athlete. In patients with recurrent
instability, surgical reconstruction is
indicated. MPFL reconstruction is
the standard of care for these pa-
tients; however, graft choice and
e414
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
fixation technique should be tai-
lored to the patient’s anatomy and
remaining physeal growth. Lower
limb alignment is an important
factor and offers the unique adjunct
of implant-mediated guided growth
to treatment of the skeletally
immature patient.
References
References printed in bold type are
those published within the past 5
years.
1. Khormaee S, Kramer DE, Yen YM,
Heyworth BE: Evaluation and management
of patellar instability in pediatric and
adolescent athletes. Sports Health 2015;7:
115-123.
2. Weeks KD III, Fabricant P, Ladenhauf HG,
Green DW: Surgical options for patellar
stabilization in the skeletally immature
patient. Sports Med Arthrosc Rev 2012;20:
194-202.
3. Fithian DC, Paxton EW, Stone ML, Silva P,
Davis DK, Elias DA, White LM:
Epidemiology and natural history of acute
patellar dislocation. Am J Sports Med 2004;
32:1114-1121.
4. Lewallen LW, McIntosh AL, Dahm DL:
Predictors of recurrent instability after
acute patellofemoral dislocation in
pediatric and adolescent patients. Am J
Sports Med 2013;41:575-581.
5. Askenberger M, Janarv PM, Finnbogason
T, Arendt EA: Morphology and anatomic
patellar instability risk factors in first-
time traumatic lateral patellar
dislocations. Am J Sports Med 2017;45:
50-58.
6. Lippacher S, Reichel H, Nelitz M:
Radiological criteria for trochlear dysplasia
in children and adolescents. J Pediatr
Orthop B 2011;20:341-344.
7. Weinberger J, Fabricant PD, Taylor SA,
Mei JY, Jones KJ: Influence of graft
source and configuration on revision rate
and patient-reported outcomes after
MPFL reconstruction: A systematic
review and meta-analysis. Knee Surg
Sports Traumatol Arthrosc 2017;25:
2511–2519.
8. Palmu S, Kallio PE, Donell ST, Helenius I,
Nietosvaara Y: Acute patellar dislocation in
children and adolescents: A randomized
clinical trial. J Bone Joint Surg Am 2008;
90:463-470.
9. Apostolovic M, Vukomanovic B, Slavkovic
N, Vuckovic V, Vukcevic M, Djuricic G,
Kocev N: Acute patellar dislocation in
Journal of the American Academy of Orthopaedic Surgeons
adolescents: Operative versus nonoperative
treatment. Int Orthop 2011;35:1483-1487.
10. Kramer DE, Pace JL: Acute traumatic and
sports-related osteochondral injury of the
pediatric knee. Orthop Clin North Am
2012;43:227-236.
11. Fabricant PD, Yen YM, Kramer DE,
Kocher MS, Micheli LJ, Heyworth BE:
Fixation of chondral-only shear fractures of
the knee in pediatric and adolescent
athletes. J Pediatr Orthop 2017;37:156.
12. Parikh SN, Nathan ST, Wall EJ, Eismann
EA: Complications of medial
patellofemoral ligament reconstruction in
young patients. Am J Sports Med 2013;41:
1030-1038.
13. Lippacher S, Dreyhaupt J, Williams SR,
Reichel H, Nelitz M: Reconstruction of the
medial patellofemoral ligament: Clinical
outcomes and return to sports. Am J Sports
Med 2014;42:1661-1668.
14. Sillanpaa PJ, Mattila VM, Mäenpää H,
Kiuru M, Visuri T, Pihlajamäki H:
Treatment with and without initial
stabilizing surgery for primary traumatic
patellar dislocation: A prospective
randomized study. J Bone Joint Surg Am
2009;91:263-273.
15. Bitar AC, D’Elia CO, Demange MK, Viegas
AC, Camanho GL: Randomized
prospective study on traumatic patellar
dislocation: Conservative treatment versus
reconstruction of the medial patellofemoral
ligament using the patellar tendon, with
a minimum of two years of follow-up. Rev
Bras Ortop 2011;46:675-683.
16. Nomura E, Inoue M: Second-look
arthroscopy of cartilage changes of the
patellofemoral joint, especially the patella,
following acute and recurrent patellar
dislocation. Osteoarthritis Cartilage 2005;
13:1029-1036.
17. Bollier M, Fulkerson JP: The role of
trochlear dysplasia in patellofemoral
instability. J Am Acad Orthop Surg 2011;
19:8-16.
18. Nelitz M, Dornacher D, Dreyhaupt J,
Reichel H, Lippacher S: The relation of the
distal femoral physis and the medial
patellofemoral ligament. Knee Surg Sports
Traumatol Arthrosc 2011;19:2067-2071.
19. Kepler CK, Bogner EA, Hammoud S,
Malcolmson G, Potter HG, Green DW:
Zone of injury of the medial patellofemoral
ligament after acute patellar dislocation in
children and adolescents. Am J Sports Med
2011;39:1444-1449.
20. Nelitz M, Dreyhaupt J, Reichel H, Woelfle
J, Lippacher S: Anatomic reconstruction of
the medial patellofemoral ligament in
children and adolescents with open growth
plates: Surgical technique and clinical
outcome. Am J Sports Med 2013;41:58-63.
21. Marsh JS, Daigneault JP, Sethi P, Polzhofer
GK: Treatment of recurrent patellar

instability with a modification of the Roux-
Goldthwait technique. J Pediatr Orthop
2006;26:461-465.
22. Grannatt K, Heyworth BE, Ogunwole O,
Micheli LJ, Kocher MS: Galeazzi
semitendinosus tenodesis for patellofemoral
instability in skeletally immature patients. J
Pediatr Orthop 2012;32:621-625.
23. Nietosvaara Y, Paukku R, Palmu S, Donell
ST: Acute patellar dislocation in children
and adolescents: Surgical technique. J Bone
Joint Surg Am 2009;91:139-145.
24. Mountney J, Senavongse W, Amis AA,
Thomas NP: Tensile strength of the medial
patellofemoral ligament before and after
repair or reconstruction. J Bone Joint Surg
Br 2005;87:36-40.
25. Matic GT, Magnussen RA, Kolovich GP,
Flanigan DC: Return to activity after
medial patellofemoral ligament repair or
reconstruction. Arthroscopy 2014;30:
1018-1025.
26. Camp CL, Krych AJ, Dahm DL, Levy BA,
Stuart MJ: Medial patellofemoral ligament
repair for recurrent patellar dislocation.
Am J Sports Med 2010;38:2248-2254.
27. Hohn E, Pandya NK: Does the utilization of
allograft tissue in medial patellofemoral
ligament reconstruction in pediatric and
adolescent patients restore patellar
stability? Clin Orthop Relat Res 2017;475:
1563-1569.
28. Wylie JD, Burks RT: Medial patellofemoral
ligament reconstruction with
semitendinosus autograft. Arthrosc Tech
2013;2:e417-e421.
29. Vavalle G, Capozzi M: Isolated
reconstruction of the medial patellofemoral
ligament with autologous quadriceps
tendon. J Orthop Traumatol 2016;17:
155-162.
30. Fink C, Veselko M, Herbort M, Hoser C:
MPFL reconstruction using a quadriceps
tendon graft: Part 2: Operative technique
and short term clinical results. Knee 2014;
21:1175-1179.
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Lauren H. Redler, MD and Margaret L. Wright, MD
31. Witonski D, Kęska R, Synder M, Sibi nski
M: An isolated medial patellofemoral
ligament reconstruction with patellar
tendon autograft. Biomed Res Int 2013;
2013:637678.
32. Bitar AC, Demange MK, D’Elia CO,
Camanho GL: Traumatic patellar
dislocation: Nonoperative treatment
compared with MPFL reconstruction using
patellar tendon. Am J Sports Med 2012;40:
114-122.
33. Hinckel BB, Gobbi RG, Demange MK,
Bonadio MB, Pécora JR, Camanho GL:
Combined reconstruction of the medial
patellofemoral ligament with quadricipital
tendon and the medial patellotibial
ligament with patellar tendon. Arthrosc
Tech 2016;5:e79-e84.
34. Steiner TM, Torga-Spak R, Teitge RA:
Medial patellofemoral ligament
reconstruction in patients with lateral
patellar instability and trochlear dysplasia.
Am J Sports Med 2006;34:1254-1261.
35. Jacobi M, Reischl N, Bergmann M,
Bouaicha S, Djonov V, Magnussen RA:
Reconstruction of the medial
patellofemoral ligament using the adductor
magnus tendon: An anatomic study.
Arthroscopy 2012;28:105-109.
36. Monllau JC, Masferrer-Pino À, Ginovart
G, Pérez-Prieto D, Gelber PE, Sanchis-
Alfonso V: Clinical and radiological
outcomes after a quasi-anatomical
reconstruction of medial patellofemoral
ligament with gracilis tendon autograft.
Knee Surg Sports Traumatol Arthrosc
2017;25:2453-2459.
37. Kumahashi N, Kuwata S, Tadenuma T,
Kadowaki M, Uchio Y: A “sandwich”
method of reconstruction of the medial
patellofemoral ligament using a titanium
interference screw for patellar instability in
skeletally immature patients. Arch Orthop
Trauma Surg 2012;132:1077-1083.
38. Lind M, Enderlein D, Nielsen T,
Christiansen SE, Faunø P: Clinical outcome
after reconstruction of the medial
patellofemoral ligament in paediatric
patients with recurrent patella instability.
Knee Surg Sports Traumatol Arthrosc
2016;24:666-671.
39. Gomes JE: Comparison between a static
and a dynamic technique for medial
patellofemoral ligament reconstruction.
Arthroscopy 2008;24:430-435.
40. Tanaka MJ: Variability in the patellar
attachment of the medial patellofemoral
ligament. Arthroscopy 2016;32:
1667-1670.
41. Fulkerson JP, Edgar C: Medial quadriceps
tendon-femoral ligament: Surgical anatomy
and reconstruction technique to prevent
patella instability. Arthrosc Tech 2013;2:
e125-e128.
42. Fithian DC, Paxton EW, Post WR, Panni
AS; International Patellofemoral Study
Group: Lateral retinacular release: A survey
of the International Patellofemoral Study
Group. Arthroscopy 2004;20:463-468.
43. Nelitz M, Theile M, Dornacher D, Wölfle
J, Reichel H, Lippacher S: Analysis of
failed surgery for patellar instability in
children with open growth plates. Knee
Surg Sports Traumatol Arthrosc 2012;20:
822-828.
44. Stupay KL, Swart E, Shubin Stein BE:
Widespread implementation of medial
patellofemoral ligament reconstruction for
recurrent patellar instability maintains
functional outcomes at midterm to long-
term follow-up while decreasing
complication rates: A systematic review.
Arthroscopy 2015;31:1372-1380.
45. Kearney SP, Mosca VS: Selective
hemiepiphyseodesis for patellar instability
with associated genu valgum. J Orthop
2015;12:17-22.
46. Ballal MS, Bruce CE, Nayagum S:
Correcting genu varum and genu valgum in
children by guided growth: Temporary
hemiepiphysiodesis using tension band
plates. J Bone Joint Surg Br 2010;92:
723-726.
e415
 Research Article

Construct Validation of a Novel Hip

Fracture Fixation Surgical
Simulator

Abstract
Matthew William Christian, MD Introduction: A surgical simulation platform has been developed to
Cullen Griffith, MD simulate fluoroscopically guided surgical procedures by coupling
computer modeling with a force-feedback device as a training tool for
Carrie Schoonover, BS
orthopaedic resident education in an effort to enhance motor skills and
Tim Zerhusen, Jr, BS potentially minimize radiation exposure. The objective of this study
Max Coale, BA was to determine whether the simulation platform can distinguish
Nathan O’Hara, MHA between novice and experienced practitioners of percutaneous
pinning of hip fractures.
Ralph Frank Henn III, MD
Downloaded from http://journals.lww.com/jaaos by BhDMf5ePHKbH4TTImqenVGyrqcySNDVCPViA6S+jqIBWiieGfYCXyCPfBTM1cR0/ on 10/02/2018

Methods: Medical students, orthopaedic residents, orthopaedic

Robert V. O’Toole, MD, MSME
Marcus Sciadini, MD
trauma fellows, and attending surgeons completed in situ hip-pinning
simulation that recorded performance measures related to surgical
accuracy, time, and use of fluoroscopy. Linear regression models
were used to compare the association between performance and
practitioner experience.
Results: Notable associations were shown between performance
and practitioner experience in 10 of the 15 overall measures (P ,
0.05) and 9 of 11 surgical accuracy parameters (P , 0.05).
Conclusion: This novel simulation platform can distinguish between
novice and experienced practitioners and defines a performance
curve for completion of simulated in situ hip pinning. This important
first step lays the groundwork for subsequent validation studies, which
will seek to demonstrate the efficacy of this simulator in improving
clinical performance by trainees completing a sequence of skills-
training modules.

S imulation training has long been

From the Department of
Orthopaedics, University of Maryland
School of Medicine, Baltimore, MD.
Correspondence to Dr. Sciadini:
msciadini@umoa.umm.edu
J Am Acad Orthop Surg 2018;26:
689-697
DOI: 10.5435/JAAOS-D-16-00724
Copyright 2018 by the American
Academy of Orthopaedic Surgeons.
recognized as an effective way to
enhance performance and expertise
and to improve safety in fields such as
aviation.1-3 More recently, this par-
adigm has been applied to surgical
training, most successfully in relation
to laparoscopic general surgical
procedures.4-11 Although orthopae-
dic surgery has been slow to adopt
surgical simulation as a routine
component of the education and
training process, great strides have
been made in recent years in devel-
oping effective and validated surgical
simulation platforms for arthro-
scopic procedures.4,12-15 In addition
to recognition of the potential ben-
efits of surgical simulation training
for resident education, concern
regarding the potential risks posed to
surgeons, operating room personnel,
and patients by radiation exposure
during fluoroscopically guided pro-
cedures has increased.
In response to these issues, the
American Academy of Orthopaedic
Surgeons (AAOS), in partnership with

October 1, 2018, Vol 26, No 19 689

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Novel Surgical Simulator for Flouroscopic Procedures

Figure 1 Figure 2

Photograph showing the foot pedal

used during simulation.

survey was administered to partic-

Photograph showing a participant’s visual perspective during simulation.
the Orthopaedic Trauma Association
(OTA), convened a task force charged
with developing a surgical simulator
for fluoroscopically assisted surgical
procedures in an effort to expand the
role of simulation in orthopaedic resi-
dent education. This effort has led to
the development of a computer-based
fluoroscopy simulation platform
(Augmented Reality Systems; Marxent
Labs LLC), which combines three-
dimensional computer-generated
graphics with a robotic force-feedback
arm to accurately mimic common sur-
gical procedures without subjecting
patients to the potential risks posed by
less skilled residents developing basic
motor skills in the operating room.
A preliminary training module with
multiple exercises of increasing com-
plexity was developed for percutaneous
fixation of a valgus-impacted femoral
neck fracture. We attempted to provide
construct validation by seeking to
determine the ability of the surgical
simulator to accurately distinguish
between experienced and novice prac-
titioners. Our hypothesis was that
experienced surgeons perform better
than novices on a simulated percutane-
ous procedure as determined by per-
formance metrics measured by the
AAOS/OTA hip fracture simulator.
Methods
Institutional review board approval
was obtained, and volunteer partic-
ipants were recruited for three study
groups: (1) novice (medical student
members of an orthopaedic interest
group, n = 15), (2) orthopaedic resi-
dents (postgraduate years 2 through
5, n = 17), and (3) experienced
(attending orthopaedic surgeons and
orthopaedic trauma fellows, n = 18). A
ipants to determine the number of
years of training, previous simulation
experience, video game experience,
and satisfaction with the appearance of
the simulator. All participants were
provided with a brief introduction to
the simulator and a 5-minute practice
period to familiarize themselves with
its use. Each participant was then al-
lowed two attempts to perform the
task of placing three guidewires in an
inverted triangle configuration into a
computer-generated model of a valgus-
impacted femoral neck fracture.
The simulated fluoroscopically
guided procedure was designed to
mimic as closely as possible the
clinical task. One computer moni-
tor displayed a split-screen depic-
tion of the fluoroscopic image, and
another displayed a computer-
generated image of the “surgical
field” (Figure 1). Activation of the
fluoroscopy unit and changing
between AP and lateral (LAT) views
were controlled by the participant
using a foot pedal (Figure 2). Only

Dr. Christian or an immediate family member serves as a board member, owner, officer, or committee member of the American Orthopaedic
Foot and Ankle Society and the Maryland Orthopaedic Association. Dr. O’Toole or an immediate family member has received royalties from
Coorstek; serves as a paid consultant to Coorstek, Imagen, and Smith & Nephew; has stock or stock options held in Imagen; has received
research or institutional support from DePuy Synthes and Stryker; and serves as a board member, owner, officer, or committee member of
the Orthopaedic Trauma Association. Dr. Sciadini or an immediate family member serves as a paid consultant to and has stock or stock
options held in Stryker. None of the following authors or any immediate family member has received anything of value from or has stock or
stock options held in a commercial company or institution related directly or indirectly to the subject of this article: Dr. Griffith, Ms.
Schoonover, Mr. Zerhusen, Mr. Coale, Mr. O’Hara, and Dr. Henn.

690 Journal of the American Academy of Orthopaedic Surgeons

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.

Table 1
Characteristics of Study Participants (n = 50)
Characteristic
Years of orthopaedic training, mean (SD)
No. of hip fractures treated, mean (SD)
Previous simulator experience, n (%)
Previous hip simulator experience, n (%)
Experience with computer games, n (%)
Is simulator appearance sufficient, n (%)
Is appearance important, n (%)
one of the two views was available
“live” at any given time, mimicking
the surgical equivalent of transfer-
ring fluoroscopic images from the
left to the right screen during pro-
cedures and requiring the clinical
motor skill of making biplanar
changes in pin position and orienta-
tion with uniplanar imaging.
Predetermined performance met-
rics were anonymously and auto-
matically recorded by the simulator
for each attempt by each participant,
including time to task completion,
total simulated fluoroscopy time, and
distance from predetermined “ideal”
pin positions. All participants were
shown an example of the inverted
triangle construct with ideally posi-
tioned pins demonstrated by simu-
lated AP and LAT fluoroscopic
views before task completion and
were instructed to replicate the con-
struct. However, they were not told
how the test was scored, nor were
they provided with any clinical
rationale behind the “ideal” con-
struct. Care was taken to ensure that
no study participants had previous
exposure to or experience with the
simulator. Although a series of 12
training exercises of increasing com-
plexity has been developed for
this training module, culminating
in the final task of placement of
the three pins, participants in
this study were not permitted to
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Medical Students Residents
(n = 15) (n = 17)
0 (0) 3.5 (1.2)
0 (0) 2 (35)
0 (0) 15 (88.24)
0 (0) 1 (5.88)
10 (66.67) 12 (70.59)
14 (93.33) 16 (94.12)
4 (26.67) 3 (17.65)
view or complete any of those
exercises.
Participants were categorized into
three training levels: medical students
(n = 15), orthopaedic residents (n =
17), and fellows/attending ortho-
paedic surgeons (n = 18). Responses
to the survey were summarized using
counts and proportions for categor-
ical responses and means with SDs
for continuous variables. Data from
the survey were compared by the
training level using analysis of vari-
ance for continuous variables and
the chi-square test for categorical
variables. Linear regression was used
to compare the effect of the partic-
ipants’ training levels on the simula-
tor parameters. The medical student
category served as the reference level
in all of the regression modes. A
power analysis demonstrated that a
sample size of 12 in each study group
would be required to provide .80%
power to detect statistically signifi-
cant differences with a one-sided
alpha level of 0.025. All analyses
were conducted using JMP Version
12 (SAS Institute).
Results
Table 1 presents a summary of the
survey responses stratified by the
training level. As expected, orthopae-
dic residents, fellows, and attendings
had markedly more experience than
medical students in years of training,
Matthew William Christian, MD, et al
Fellows and Attendings
(n = 18) P value
10.6 (6.2) ,0.0001
212 (34) ,0.0001
9 (50) ,0.0001
3 (16.67) 0.20
12 (66.67) 0.96
17 (94.44) 0.99
5 (27.78) 0.75
number of hip fractures treated, and
previous surgical simulator experience.
No notable difference was shown in
experience with computer games or
satisfaction with simulator appearance.
As shown in Table 2, orthopaedic
fellows and attendings significantly
outperformed medical students in
distance to the anterior cortex (P =
0.02), distance above the bottom of
the lesser trochanter (P = 0.04),
inferior guidewire tip distance to the
center (P = 0.04), and angle between
posterior guidewire and anterior
guidewire (P = 0.04).
Table 3 shows the parameters
in which residents, fellows, and
attendings significantly outperformed
medical students. The residents and
the fellows and attendings groups
used significantly fewer radiographs
(P = 0.03 and P = 0.04, respectively),
placed guidewires significantly closer
to the joint surface anteriorly,
posteriorly, and inferiorly (P , 0.001
in all cases), and were significantly
closer to the ideal starting point on
the LAT cortex for the inferior pin
(defined as the mid-sagittal position
on the LAT fluoroscopic view) (P ,
0.001). The more experienced groups
also placed the anterior and inferior
guidewires significantly more parallel
(P = 0.04 and P = 0.07, respectively).
Figure 3 demonstrates the relation-
ship between years of training of the
participant and surgical precision
parameters, such as the distance from
691
Novel Surgical Simulator for Flouroscopic Procedures

Table 2
Simulator Parameters in Which Fellows and Attending Orthopaedic Surgeons Outperform Medical Students
Training Parameter
Parameter Level Estimate SE P value

Distance to the anterior cortex (mm)

Medical student, 5.31 6 0.87a
Resident 2.24 1.2 0.07
Fellow/attending 22.8 1.18 0.02
Distance above bottom of the lesser trochanter (mm)
Medical student, 15.15 6 8.19a
Resident 23.11 2.67 0.25
Fellow/attending 25.65 2.63 0.04
Inferior guidewire tip distance to the center (mm)
Medical student, 8.24 6 1.55a
Resident 23.84 2.13 0.08
Fellow/attending 24.5 2.1 0.04
Angle between posterior guidewire and anterior
guidewire (°)
Medical student, 3.59 6 0.47a
Resident 21.25 0.65 0.06
Fellow/attending 21.35 0.64 0.04

SE = standard error
a
Medical student training level served as the reference level; values shown are mean 6 SD.

the posterior guidewire to the joint
surface.
As noted in Table 4, a number of
parameters were not significantly dif-
ferent despite differing levels of training
and experience. The amount of time
on the simulator required to place
the three pins and number of wire
retries were not markedly different.
We noted a trend toward decreased
amount of fluoroscopy time for both
the residents and the fellows and at-
tendings groups; however, it was not
significantly different (P = 0.28 and
P = 0.13, respectively). In addition,
the number of guidewires penetrating
the joint surface, either transiently or
left in this position, and parallel
placement of the posterior guidewire
relative to the inferior guidewire were
not markedly different.
Discussion
Traditionally, surgical training has
been undertaken using an appren-
ticeship model in which trainees
participate in surgical procedures on
patients under the supervision of
experienced surgeons. This approach
has remained largely unchanged
since Halsted16 proposed the tech-
nique in the 1900s. Increasing levels
of involvement and autonomy are
conferred based on the trainee’s
demonstration of mastery of con-
cepts and techniques and chrono-
logical progression through surgical
residency training. Much of the
assessment of a trainee’s progress is
highly subjective and dependent on
the teaching physician’s observations
in the operating room. Some surgical
residents are naturally gifted, and
others struggle with certain tasks in
the operating room. In either case,
the potential exists for adverse ef-
fects on patients, ranging from
increased surgical times to iatrogenic
injury. Additionally, many ortho-
paedic trauma procedures involve
the use of intraoperative fluoroscopy.
Resulting radiation exposure may
place surgeons, patients, trainees,
and operating room personnel at risk,
particularly if radiation is increased
because of the training component
on actual patients. The amount of
radiation exposure has been shown
to be higher in cases performed
by less-experienced surgeons. This
phenomenon is thought to be caused
by lower levels of confidence, less
experience with surgical procedures,
and higher level of reliance on fluo-
roscopic imaging.17,18
Percutaneous, fluoroscopically guided
procedures are common in orthopaedic
surgery. Performing them safely
and effectively depends on a thor-
ough understanding of anatomy, the
pathophysiology of the injury or
condition being treated, and the
mechanical and biologic implications
of the applied treatment. Most of
these concepts can be taught through
traditional didactic means. However,
equally important to the success of

692 Journal of the American Academy of Orthopaedic Surgeons

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
 Matthew William Christian, MD, et al

Table 3
Simulator Parameters in Which Residents, Fellows, and Attending Orthopaedic Surgeons Outperform Medical
Students
Parameter
Parameter Training Level Estimate SE P value

Radiographs (n)
Medical student, 41.73 6 32.25a
Resident 23.91 10.91 0.03
Fellow/attending 22.66 10.76 0.04
Inferior guidewire distance to the joint surface (mm)
Medical student, 16.21 6 1.41a
Resident 212.5 1.94 ,0.001
Fellow/attending 211.8 1.91 ,0.001
Posterior guidewire distance to the joint surface (mm)
Medical student, 13.5 6 0.74a
Resident 29.26 1.02 ,0.001
Fellow/attending 29.61 1.01 ,0.001
Anterior guidewire distance to the joint surface (mm)
Medical student, 14.54 6 1.09a
Resident 210.76 1.5 ,0.001
Fellow/attending 211.15 1.48 ,0.001
Distance to the center of the lateral cortex (mm)
Medical student, 5.35 6 0.6a
Resident 23.12 0.83 ,0.001
Fellow/attending 23.19 0.82 ,0.001
Angle between inferior guidewire and anterior
guidewire (°)
Medical student, 4.69 6 0.67a
Resident 21.9 0.92 0.04
Fellow/attending 21.69 0.9 0.07

SE = standard error
a
Medical student training level served as the reference level; values shown are mean 6 SD.

these procedures and the safety of
patients undergoing them is the
acquisition of specialized motor
skills, an understanding of three-
dimensional structures as evaluated
by two-dimensional images, and the
simultaneous processing of didactic
knowledge and visual and tactile
feedback received by the surgeon
while performing the procedure.
These latter skills are not learned in a
classroom. Hands-on practice and
repetitive exposure are required to
achieve proficiency. Although this
has traditionally occurred in the
operating room during surgical pro-
cedures on live patients, the inherent
risk of patient harm with that
approach is obvious, particularly
during a trainee’s learning phase of
technically demanding skills. In the
case of percutaneous pinning of
femoral neck fractures, penetration
of the articular surface, damage to
the femoral head blood supply,
neurologic injury, and prolonged
surgical time leading to increased
infection and anesthesia-related com-
plications are all potential hazards
encountered when training new
surgeons. Developing a tool to allow
beginning practitioners to progress
through a substantial portion of the
learning phase in an environment
that is risk-free for patients and
trainees is the ultimate goal of sim-
ulator development.
Surgical simulation has been shown
to be an effective adjunct to tradi-
tional clinical training in other dis-
ciplines, with positive effects on
procedural time and intraoperative
errors.7-10 General surgeons first
routinely used surgical simulation
training with the development of
laparoscopic surgical procedures in
the 1990s.6 Because of the inherently
different skill set required to perform
laparoscopic surgical procedures
compared with traditional open
techniques, including triangulation

October 1, 2018, Vol 26, No 19 693

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Novel Surgical Simulator for Flouroscopic Procedures
Figure 3
Graph showing the distance from the posterior guidewire to the joint surface in a
simulated procedure compared with the years of training of the participant and fit
with a smoothing spline (lambda = 600, R-squared = 0.61).
of instruments and performing the
procedure based on images displayed
on a video screen intraoperatively,
surgeons were required to complete
surgical skills training before being
credentialed to perform the clinical
procedure. This training included
surgical simulation, cadaveric human,
and live animal practice surgeries.
Building on this foundation, the fun-
damentals of laparoscopic surgery
have mandated surgical simulation
as a requisite component of general
surgical residency training since
2008. 11
Arthroscopic surgical procedures
have been a large part of the practice of
orthopaedic surgery for .50 years.15
The motor skills required of surgeons
performing surgical arthroscopy are
similar to those used in laparoscopic
surgery. Because the routine use of
arthroscopy by orthopaedic surgeons
predated the advent of laparoscopy in
general surgery, arthroscopic training
694
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
for orthopaedic surgeons was well
incorporated into the traditional ap-
prenticeship model. Recognizing the
benefits afforded by simulation train-
ing in laparoscopic surgery to general
surgery training programs, the
Arthroscopic Association of North
America developed their own surgical
simulation initiative resulting in sev-
eral well-established and validated
arthroscopy simulation platforms.14
Arthroscopy simulators currently in
use range from simple inexpensive box
models to highly sophisticated ex-
pensive anatomic models and video
workstations. Because the fluoroscopy
simulation platform is computer
based, it does not require specific
anatomic models for expansion to
other procedures or anatomic loca-
tions but can be easily expanded
through simple software upgrades. If it
is shown to be an effective training
tool in subsequent validation studies,
the adaptability to a range of fluoro-
Journal of the American Academy of Orthopaedic Surgeons
scopically guided procedures may
prove advantageous. Although the
cost associated with this simulator is
higher than that of the simplest box
models available for arthroscopy
training, it is still considerably less
than that of the more sophisticated
anatomic model arthroscopic simu-
lators and, as noted, includes the
potential for expanded applications
through software upgrades alone.
The AAOS/OTA fluoroscopy sim-
ulation project was undertaken in
recognition of the potential educa-
tional benefits for orthopaedic sur-
gery residents and for the potential to
improve patient safety afforded by
an effective fluoroscopy simulation
platform. The resulting simulation
platform consists of a personal
computer-based program that gen-
erates simulated fluoroscopic images
coupled with a haptic feedback arm
to mimic tactile feedback experienced
intraoperatively. For an introductory
training module, the relatively simple
procedure of percutaneous pinning
of a valgus-impacted femoral neck
fracture was selected. Skill develop-
ment exercises were intended to
“deconstruct” the surgical task into
its most basic components (eg,
developing a sense of “feel” for
the guidewire against bone, dis-
tinguishing between unicortical and
bicortical pin placement, adjusting
directionality of pin placement).
Each subsequent skill built on the
previous one, ultimately culminating
in the accurate placement of three
pins in an inverted triangle configu-
ration. Validation studies to dem-
onstrate whether these training
exercise programs improve perfor-
mance not only on the simulator but
in the operating room are ongoing.
Construct validation is defined as
the extent to which an instrument
measures the characteristic being
investigated.19 Our study sought to
accomplish this important first step
in construct validation, that is, to
demonstrate that the final exercise of
 Matthew William Christian, MD, et al

Table 4
Simulator Parameters in Which There Was No Effect of Training on Performance
Training Parameter
Parameter Level Estimate SE P value

Time in simulator (s)

Medical student, 250.2 6 36.08a
Resident 3.6 49.5 0.94
Fellow/attending 22 48.85 0.97
Time under fluoroscopy (s)
Medical student, 28.14 6 5.31a
Resident 27.95 7.29 0.28
Fellow/attending 211.13 7.2 0.13
No. of retries
Medical student, 3.33 6 0.92a
Resident 21.63 1.26 0.20
Fellow/attending 20.72 1.24 0.56
Angle between inferior guidewire and posterior
guidewire (°)
Medical student, 2.64 6 0.44a
Resident 0.45 0.61 0.47
Fellow/attending 20.16 0.6 0.79
Guidewires through the joint surface (n)
Medical student, 0.27 6 0.14a
Resident 20.21 0.2 0.29
Fellow/attending 0.18 0.19 0.36

SE = standard error
a
Medical student training level served as the reference level; values shown are mean 6 SD.

accurately placing three guidewires
in an inverted triangle configura-
tion without first completing the
sequence of training exercises would
distinguish between experienced and
inexperienced users.
Our results suggest that this simu-
lator is capable of distinguishing
between groups with differing clini-
cal experience: medical students with
an interest in orthopaedic surgery but
no previous experience with percu-
taneous pin placement, orthopaedic
residents with some experience, and
orthopaedic trauma fellows and at-
tendings with the most experience.
One of the parameters for which
performance metrics were similar
between the groups was the total time
to completion of the exercise. This
finding may reflect a lack of under-
standing on the part of the medical
students regarding the implications
of incorrect pin placement, causing
them to take less time and care in
positioning the pins. Although the
number of fluoroscopic images ob-
tained showed a statistically signifi-
cant difference between novice and
experienced users, total fluoroscopic
imaging time did not, perhaps re-
flecting an increased propensity of
the experienced practitioners to use
longer runs of “live fluoroscopy” but
fewer spot films. The performance
metrics assessed by the simulation
program were based on certain as-
sumptions and parameters defined
by the investigators and incorpo-
rated into the computer program.
The ideal starting point and trajec-
tory for the inferior pin, for example,
were defined as being no lower than
the inferior border of the lesser tro-
chanter, traversing the femoral neck
at the mid-sagittal level, and passing
no closer than 2 mm and no further
than 3 mm from the inferior cortex of
the femoral neck. Similar “ideals”
were defined for the superior anterior
and superior posterior pins. Pin pen-
etration of the articular surface was
also critically assessed. As further
development and validation of the
simulator progresses, adjustments to
these parameters may be necessary to
more accurately assess clinically rel-
evant performance. However, this
initial investigation seems to confirm
the ability of the simulator to assess
and distinguish between performance
by users with differing amounts of
surgical training and experience.
Subjective feedback from the more
experienced practitioners (residents,
fellows, and attendings) regarding

October 1, 2018, Vol 26, No 19 695

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Novel Surgical Simulator for Flouroscopic Procedures
how accurately the simulator repro-
duced the tactile, visual, and three-
dimensional elements of percutaneous
hip pinning was not specifically solic-
ited as part of the study. However, the
anecdotal feedback received was uni-
versally positive. Most of them com-
mented that performing the task on
the simulator actually seemed more
difficult than it is in real life. This
phenomenon may reflect the fact that
when performing the task on the
simulator, the surgeon is required to
activate the appropriate foot pedal
himself or herself to obtain the desired
fluoroscopic image rather than simply
asking the radiology technician for the
view. This perceived increased level of
difficulty on the simulator might
prove to be a desirable feature for
improving ultimate clinical task per-
formance. Criticism of the simulator
was centered on the force-feedback
(haptic) arm, with most stating that
although the qualitative feel of
passing a terminally threaded pin
through bone was reasonably con-
veyed, the resistance encountered on
the simulator was quantitatively much
less than in clinical practice.
This study should be interpreted
within the context of its limitations.
Although we demonstrated proper
pin placement to all participants, we
did not confirm whether they under-
stood it before testing their perfor-
mance on the simulator, thereby
introducing the risk that differences
in their performances could have
reflected a lack of understanding of
the task rather than a lack of technical
ability. The study could have been
strengthened by testing the partic-
ipants before the task was begun to
ensure that they actually understood
the proper pin placement. The simu-
lator’s most profound differences
were found between surgeons with
any experience and novices, which
makes sense if it is measuring a very
basic skill, such as pin placement,
that surgeons pick up relatively
quickly during residency. We at-
696
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
tempted to closely match our control
group to the resident population by
choosing students from the ortho-
paedic interest group only, but per-
haps there is something intrinsically
different about the groups, such as
better hand–eye coordination in those
who make it into the group compared
with those who are just interested. This
would be an alternate explanation for
our results; however, other factors
favor the medical students such as
increased experience with video games,
which is likely to produce improved
performance on the simulator. Our
data are from only one institution, and
other institutions might not yield
the same results. Furthermore, our
results are based in part on the
“ideal” performance metrics for
each parameter. These metrics
were predetermined before testing;
however, different ideal parameters
would have changed our results,
favoring either the novices or the
more experienced participants.
Further studies are necessary and
ongoing to determine the following: (1)
the effectiveness of the skill develop-
ment exercises in improving perfor-
mance by users at all levels; (2) the
period of time that positive effects from
completion of training modules persist;
and (3) the effectiveness of skills train-
ing on the simulator in improving
performance in the operating room.
Before these studies can be conducted,
the critical step of construct validity
must be undertaken, which is what we
believe we have done here.
Conclusions
Our study demonstrates preliminary
construct validation of the AAOS/OTA
fluoroscopy simulator—specifically, its
ability to distinguish between novice
and experienced practitioners in the
performance of a simple hip-pinning
procedure. Further validation is nec-
essary and ongoing to determine
whether this platform will prove
Journal of the American Academy of Orthopaedic Surgeons
capable of enhancing resident educa-
tion and improving clinical perfor-
mance, once expanded to include
appropriate training modules and
incorporated into residency training.
References
Evidence-based Medicine: Levels of
evidence are described in the table of
contents. In this article, references
8, 9, 18 are level I studies. References
3-5, 7, 11, and 17 are level II studies.
References 14 is a level III study.
References 1, 2, 6, 10, 12, and 13 are
level V reports or expert opinions.
References printed in bold type are
those published within the past 5
years.
1. Bell HH, Waag WL: Evaluating the
effectiveness of flight simulators for training
combat skills: A review. Int J Aviat Psychol
1998;8:223-242.
2. Carretta TR, Dunlap RD: Transfer of
Training Effectiveness in Flight Simulation:
1986 to 1997. Mesa, AZ, Air Force
Materiel Command, 1998.
3. Hays RT, Jacobs JW, Prince C, Salas E:
Flight simulator training effectiveness: A
meta-analysis. Mil Psychol 1992;4:63-74.
4. Gomoll AH, O’Toole RV, Czarnecki J,
Warner JJ: Surgical experience correlates
with performance on a virtual reality
simulator for shoulder arthroscopy. Am J
Sports Med 2007;35:883-888.
5. O’Toole RV, Playter RR, Krummel TM,
et al: Measuring and training surgical skill
with a virtual reality surgical simulator. J
Am Coll Surg 1999;189:114-127.
6. Wignall GR, Denstedt JD, Preminger GM,
et al: Surgical simulation: A urological
perspective. J Urol 2008;179:1690-1699.
7. Grantcharov TP, Kristiansen VB, Bendix J,
Bardram L, Rosenberg J, Funch-Jensen P:
Randomized clinical trial of virtual reality
simulation for laparoscopic skills training.
Br J Surg 2004;91:146-150.
8. Seymour NE, Gallagher AG, Roman SA,
et al: Virtual reality training improves
operating room performance: Results of a
randomized, double-blinded study. Ann
Surg 2002;236:458-463.
9. Ahlberg G, Enochsson L, Gallagher AG, et al:
Proficiency-based virtual reality training
significantly reduces the error rate for residents
during their first 10 laparoscopic
cholecystectomies. Am J Surg 2007;193:
797-804.

10. Seymour NE, Gallagher AG, Roman SA,
O’Brien MK, Andersen DK, Satava RM:
Analysis of errors in laparoscopic surgical
procedures. Surg Endosc 2004;18:592-595.
11. Loukas C, Lahanas V, Kanakis M,
Georgiou E: The effect of mixed-task basic
training in the acquisition of advanced
laparoscopic skills. Surg Innov 2014;22:
418-425.
12. Mabrey JD, Reinig KD, Cannon WD: Virtual
reality in orthopaedics: Is it a reality? Clin
Orthop Relat Res 2010;468:2586-2591.
13. Atesok K, Mabrey JD, Jazrawi LM, Egol
KA: Surgical simulation in orthopaedic
October 1, 2018, Vol 26, No 19
Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
skills training. J Am Acad Orthop Surg
2012;20:410-422.
14. Modi CS, Morris G, Mukherjee R:
Computer-simulation training for knee and
shoulder arthroscopic surgery. Arthroscopy
2010;26:832-840.
15. DeMaio M: Giants of orthopaedic surgery:
Masaki Watanabe MD. Clin Orthop Relat
Res 2013;471:2443-2448.
16. Halsted W: The training of surgeons.
Bull Johns Hopkins Hosp 1904;15:
267-276.
17. Bar-On E, Weigl DM, Becker T, Katz K,
Konen O: Intraoperative C-arm radiation
Matthew William Christian, MD, et al
affecting factors and reduction by an
intervention program. J Pediatr Orthop
2010;30:320-323.
18. Blattert TR, Fill UA, Kunz E, Panzer W,
Weckbach A, Regulla DF: Skill
dependence of radiation exposure for the
orthopaedic surgeon during interlocking
nailing of long-bone shaft fractures: A
clinical study. Arch Orthop Trauma Surg
2004;124:659-664.
19. Miller-Keane, O’Toole MT: Construct
validity. in Miller-Keane, O’Toole MT, eds:
Miller-Keane Encyclopedia & Dictionary of
Medicine, Nursing, & Allied Health, ed 7.
Philadelphia, PA, Saunders, 2003.
697

Harpreet Bawa, MD
Jack W. Weick, MD
Douglas R. Dirschl, MD
Hue H. Luu, MD
From the Department of Orthopaedic
Surgery and Rehabilitation Medicine,
University of Chicago Medicine and
Biological Sciences, University of
Chicago Medical Center, Chicago, IL.
Correspondence to Dr. Bawa:
harpreetbawa@gmail.com
Dr. Dirschl or an immediate family
member serves as a paid consultant
to Bone Support and Stryker and
serves as a board member, owner,
officer, or committee member of the
American Orthopaedic Association,
the Foundation for Orthopaedic
Trauma, and the Orthopaedic Trauma
Association. Dr. Luu or an immediate
family member serves as a paid
consultant to DePuy Synthes and
Stryker and serves as a board
member, owner, officer, or committee
member of the American Orthopaedic
Association and the Musculoskeletal
Tumor Society. Neither of the
following authors nor any immediate
family member has received anything
of value from or has stock or stock
options held in a commercial company
or institution related directly or
indirectly to the subject of this article:
Dr. Bawa and Dr. Weick.
J Am Acad Orthop Surg 2018;00:1-8
DOI: 10.5435/JAAOS-D-17-00235
Copyright 2018 by the American
Academy of Orthopaedic Surgeons.
Month 2018, Vol 00, No 00
Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Research Article
Trends in Deep Vein Thrombosis
Prophylaxis and Deep Vein
Thrombosis Rates After Total Hip
and Knee Arthroplasty
Abstract
Introduction: Patients undergoing total hip arthroplasty (THA) and
total knee arthroplasty (TKA) are at high risk of deep vein thrombosis
(DVT) postoperatively, necessitating the use of prophylaxis
medications. This investigation used a large claims database to
evaluate trends in postoperative DVT prophylaxis and rates of DVT
within 6 months after THA or TKA.
Methods: Truven Health MarketScan Commercial Claims and
Encounters and Medicare Supplemental and Coordination of Benefits
databases were reviewed from 2004 to 2013 for patients who
underwent THA or TKA. Data were collected on patient age, sex,
Charlson Comorbidity Index, and hypercoagulability diagnoses.
Postoperative medication claims were reviewed for prescribed aspirin,
warfarin, enoxaparin, fondaparinux, rivaroxaban, and dabigatran.
Results: A total of 369,483 patients were included in the analysis, of
which 239,949 patients had prescription medication claims. Warfarin
was the most commonly prescribed anticoagulant. Patients with a
hypercoagulable diagnosis had markedly more DVTs within 6 months
after THA or TKA. More patients with a hypercoagulable diagnosis
were treated with warfarin or lovenox than other types of
anticoagulants. A multivariate regression analysis was performed,
showing that patients prescribed aspirin, fondaparinux, and
rivaroxaban were markedly less likely than those prescribed warfarin
or enoxaparin to have a DVT within 6 months after THA or TKA.
Conclusion: After THA and TKA, warfarin is the most commonly
prescribed prophylaxis. Patients with hypercoagulability diagnoses
are at a higher risk of postoperative DVT. The likelihood of DVT within
6 months of THA and TKA was markedly higher in patients treated with
warfarin and lovenox and markedly lower in those treated with aspirin,
fondaparinux, and rivaroxaban.
Level of Evidence: Level III
M ore than 1 million total hip
arthroplasty (THA) and total
knee arthroplasty (TKA) procedures
of thromboembolic complications.2–4
The most common complication is
deep vein thrombosis (DVT) because
are performed in the United States patients are placed at a higher risk
each year.1 Patients undergoing these as a result of venous stasis with
procedures are at an increased risk leg positioning, increased risk of
1
Trends in Deep Vein Thrombosis Prophylaxis
Figure 1
Flowchart showing the cohort selection.
endothelial injury, and aberrant acti-
vation of the clotting cascade.4,5
Prophylaxis after THA and TKA is
recommended by both the American
Academy of Orthopaedic Surgeons
and American College of Chest Physi-
cians; however, no consensus exists on
the optimal prophylactic regimen.6
Anticoagulation after THA or TKA
can pose unique challenges because
anticoagulation medications must bal-
ance the reduction in blood clot for-
mation, with the risk of postoperative
bleeding, hematoma formation, revi-
sion surgery, and infection.7-10 The
vitamin K antagonist warfarin has been
shown to be effective in reducing the
rate of proximal DVTs and pulmonary
embolisms.11,12 The major advantage
of warfarin is that it can be reversed
if bleeding complications arise or
if patients require urgent surgical
intervention.13,14 Warfarin’s most
notable disadvantage is that it is only
effective within a narrow therapeutic
window, necessitating frequent lab-
oratory monitoring and dose ad-
justments. These drawbacks have led
to the use of alternative methods of
chemoprophylaxis.
Recent studies have demonstrated the
effectiveness of aspirin prophylaxis
2 Journal of the American Academy of Orthopaedic Surgeons
Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
after TKA or THA. In a trial of 13,356
patients undergoing THA, low-dose
aspirin reduced the rate of DVT by
29%, PE by 43%, and fatal pulmo-
nary embolism by 58% compared
with placebo.15 Low-molecular-weight
heparin agents, such as enoxaparin,
have also been shown to be effective;
however, they must be administered
through subcutaneous injection.16
Other therapies such as factor Xa
and direct thrombin inhibitors are
appealing because they can be deliv-
ered orally, do not require monitor-
ing, and have constant dosing for
most patients. Unfortunately, this
group of medications is costly and
requires fresh frozen plasma for
reversal.5,17,18
Although DVT prophylaxis after
THA and TKA is assumed to be the
standard of care, given the high like-
lihood of thromboembolic events
without prophylaxis, a paucity of
data exists on surgeon practice pat-
terns and changes over time. In
addition, few studies have evaluated
postoperative DVT rates by anti-
coagulants at a large-scale pop-
ulation level. The purpose of this
investigation was to use a large claims
database to evaluate trends over time
in national practice patterns of post-
operative DVT prophylaxis and rates
of DVT within 6 months after THA
or TKA procedures.
Methods
Data Source
A retrospective review was conducted
using the Truven Health MarketScan
Commercial Claims and Encounters
(commercial insurance) and Medicare
Supplemental and Coordination of
Benefits (Medicare with commercial
supplement) databases (Truven Health
Analytics). The databases contain
deidentified, integrated, person-specific
claims data for approximately 17 to 51
million individuals per year. The com-
mercial insurance database includes
healthcare claims for individuals with
insurance through a commercial pro-
vider or a self-insuring employer under
fee-for-service, fully capitated, or par-
tially capitated health plans. The
Medicare with commercial supplement
database includes claims information
for individuals who have both Medi-
care and commercial employer-
sponsored coverage. All claims
from the Medicare with commercial
supplement database reflect the coor-
dination of benefits between the com-
mercial insurer and Medicare such that
all payments made by either entity are
captured within the database. The
age distribution in the Medicare with
commercial supplement database is
representative of the overall Medicare
population. These databases, when
combined, constitute approximately
20% of the overall insurance market.
International Classification of Disease
(ICD-9) diagnoses codes and Current
Procedure Terminology (CPT) codes
can be identified in individual claims.
The data include claims made from
both inpatient and outpatient clinical
encounters and prescription medi-
cations. National Drug Codes (NDCs)
are used to organize prescription
medication claims. The NDCs specify
 Harpreet Bawa, MD, et al

Table 1
Demographics of Patient Deep Vein Thrombosis Prophylaxis After Total Hip Arthroplasty or Total Knee
Arthroplasty Procedures
No
Prescription Anticoagulation
Factor Aspirin Warfarin Enoxaparin Fondaparinux Rivaroxaban Dabigatran Claim

Age 61.9 64.9 62.5 61.8 61.3 70.2 66.0

% Female 56.6 57.7 56.0 55.9 56.7 41.9 61.4
% Hypercoagulability 1.8 3.0 2.9 1.8 1.7 2.5 0.9
diagnosis
CCI 0.043 0.081 0.085 0.081 0.043 0.067 0.089
Percent of total 0.44 38.0 20.8 3.4 5.1 0.3 35.1
population (%)

CCI = Charlson Comorbidity Index

Table 2
Postoperative Deep Vein Thrombosis Prophylaxis Trends by Drug Type
Postoperative Medication 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Prescription aspirin 0.03% 0.02% 0.03% 0.00% 0.03% 0.08% 0.20% 1.04% 1.59% 2.91%
Warfarin 75.02% 72.29% 68.04% 63.18% 61.65% 60.36% 58.92% 54.77% 45.51% 41.17%
Enoxaparin 25.26% 28.46% 30.91% 33.81% 35.65% 37.21% 37.94% 36.88% 26.19% 24.63%
Fondaparinux 2.72% 3.33% 5.21% 7.27% 7.09% 7.14% 7.18% 5.85% 2.98% 2.15%
Rivaroxaban 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 5.30% 27.89% 33.58%
Dabigatran 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.25% 1.39% 1.25% 1.05%

both the type and dosage of the med-
ication prescribed.
Study Sample
The databases were reviewed from
2004 to 2013 for subjects with a CPT
code for THA (CPT = 27,130) or TKA
(CPT = 27,447). Patients were
required to be enrolled in the database
continuously for 6 months before and
6 months immediately after the THA
or TKA procedure. Data were col-
lected on patient age and sex. We
identified a group of patients who
were hypercoagulable, defined as
subjects with a previous DVT (ie,
ICD-9 code 453.40) or who had an
ICD-9 code for primary hypercoagu-
lable state (ie, ICD-9 289.81, 289.82,
286.53, and 795.79), which includes
diagnoses of factor V leiden, anti-
phospholipid antibody, lupus antico-
agulant, and protein C/S deficiency.
Prescription medication claim infor-
mation was collected based on NDCs
on prescribed aspirin, warfarin, enox-
aparin, fondaparinux, rivaroxaban,
and dabigatran. Patients were ex-
cluded if they were on any prescrip-
tion anticoagulation medications
.2 months before the THA or TKA
procedure. In addition, patients with
no claims for a prescription antico-
agulant are presented as a separate
group. Data were gathered on all
anticoagulation prescription claims in
the 6-month period after THA and
TKA procedures. Postoperative
DVTs were identified in the 6 months
after THA and TKA procedures using
the ICD-9 code for DVT (453.40).
We excluded any anticoagulation
prescription claims that occurred
after a DVT. Charlson Comorbidity
Index (CCI) scores were calculated
for each subject using ICD-9 codes
for comorbidities, as described in
previous studies.19
Statistical Analysis
Mean age, percent female, percent of
population with the hypercoagulable
group, and CCI were calculated for
each type of anticoagulant studied.
Trends of utilization by year for each
anticoagulant were compared. Rates
of DVT in the 6-month period after
THA and TKA were calculated by
each type of prescription anticoagu-
lant using the ICD-9 code for DVT
(ie, 453.40). DVT rates by year were
also calculated over the course of the
study period to analyze trends in
DVT rates over time. A best fit line
was used to calculate the R2 value
and statistical significance of the
yearly trend.

Month 2018, Vol 00, No 00 3

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Trends in Deep Vein Thrombosis Prophylaxis

Figure 2

Graph showing yearly trends in deep vein thrombosis prophylaxis after total hip arthroplasty and total knee arthroplasty procedures.

Table 3 239,949 patients had prescription

Deep Vein Thrombosis Rates Within 6 Months of Total Hip Arthroplasty or
Total Knee Arthroplasty Procedures by Drug Type
Postoperative Anticoagulation
Prescription aspirin
Warfarin
Enoxaparin
Fondaparinux
Rivaroxaban
Dabigatran
No anticoagulation claim
DVT = deep vein thrombosis
DVT rates were compared in pa-
tients in the hypercoagulable group
versus patients not in the hypercoag-
ulable group using chi-squared tests.
All statistical analysis were performed
with SAS software, version 9.3 (SAS).
Comparison of Deep Vein
Thrombosis Rates by
Anticoagulants
Further analysis was conducted to
attempt to compare the rate of DVT
observed within 6 months after THA or
TKA procedures by the anticoagulant
anticoagulant information. A flow-
chart of patient selection is presented
in Figure 1. The average age of the
Percent w/DVT
cohort was 66.7 years, 58.7% were
2.20 female, 1.8% were in the hyperco-
4.74 agulable group, 28.3% underwent
3.73 THA, and 71.7% underwent TKA.
2.69 Demographics by anticoagulant
1.86 type and for patients without anti-
3.83 coagulation claims are presented
2.16
in Table 1. Warfarin was the most
commonly prescribed anticoagulant
within the cohort at a rate of 58.44%.
At the beginning of the study period
in 2004, 0.03% of patients had claims
type after controlling for confound- for prescribed aspirin, 75.02% warfa-
ing variables. Multivariate logistic rin, 25.26% enoxaparin, 2.72% fon-
regression analysis was performed daparinux, zero rivaroxaban, and zero
controlling for the effects of age, sex, dabigatran. Throughout the study
hypercoagulability, and CCI. Odds period, the most notable changes
ratios .1.0 signified the increased included a decrease in warfarin claims
risk of DVT in relation to the other and increase in rivaroxaban claims.
forms of prescription anticoagulation In 2013, 2.19% of patients had claims
for prescribed aspirin, 41.17% warfa-
studied.
rin, 24.63% enoxaparin, 2.15% fon-
daparinux, 33.58% rivaroxaban, and
Results 1.05% dabigatran. The prescription
medication claims per year are pre-
A total of 369,483 patients were sented in Table 2. Figure 2 demon-
included in the analysis, of which strates the change in postoperative

4 Journal of the American Academy of Orthopaedic Surgeons

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
 Harpreet Bawa, MD, et al

Table 4
Deep Vein Thrombosis Rate Within 6 Months After Total Hip Arthroplasty or Total Knee Arthroplasty Procedures by Year
Year 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Annual DVT rate 3.33% 3.40% 3.39% 3.28% 3.41% 3.21% 3.12% 3.26% 3.27% 3.21%

DVT = deep vein thrombosis

DVT prophylaxis drug prescriptions Figure 3

over the study period time.
DVT rates in patients within
6 months after THA or TKA proce-
dures were calculated and are pre-
sented in Table 3. DVTs were recorded
in 2.20% of patients with claims for
prescribed aspirin, 4.74% for warfa-
rin, 3.73% for enoxaparin, 2.69% for
fondaparinux, 1.86% for rivarox-
aban, 3.83% for dabigatran, and
2.17% for those patients with no
postoperative anticoagulation claims.
DVT rates within 6 months after THA Graph showing the annual rate of deep vein thrombosis within 6 months after
or TKA for each year during the study total hip arthroplasty and total knee arthroplasty procedures.
period were calculated (Table 4).
Figure 3 is a graph demonstrating that
Table 5
no significant change in DVT rates
was observed over the study period Deep Vein Thrombosis Rates Within 6 Months of Total Hip Arthroplasty or
(P = 0.07; r2 = 0.41). Total Knee Arthroplasty Procedures Based on Hypercoagulable Diagnosis
A subanalysis was performed Postoperative Hypercoagulable No Hypercoagulable P
comparing postoperative DVT rates Anticoagulation Diagnosis* Diagnosis value
within 6 months after THA or TKA Prescription aspirin 0.45% 0.44% 0.98
procedures in the hypercoagulable Warfarin 64.26% 37.48% ,0.001
group versus the remainder of the Enoxaparin 34.75% 20.55% ,0.001
cohort. Patients in the hypercoagu-
Fondaparinux 3.41% 3.35% 0.76
lable group had significantly more
Rivaroxaban 4.81% 5.10% 0.29
DVTs within 6 months after THA or
Dabigatran 0.47% 0.32% 0.05
TKA procedures compared with the
No anticoagulation claim 18.54% 35.35% ,0.001
remainder of the cohort (24.30%
Overall DVT rate 24.30% 2.50% ,0.001
versus 2.5%; P , 0.001) (Table 5). In
addition, a significantly greater DVT = deep vein thrombosis
proportion of patients in the hyper- *Hypercoagulable Dx: history of factor V Leiden, antiphospholipid antibody, lupus anticoagulant,
hypercoagulable state, or previous DVT in database.
coagulable group were treated with
warfarin (64.3% in the hypercoag-
ulable group versus 37.5% in the
remainder of the cohort, P , 0.001) others (Table 6). The adjusted analysis less likely to have a DVT within
or lovenox (34.8% in the hyperco- showed patients on prescribed aspirin 6 months after THA or TKA proce-
agulable group versus 20.6% in the (odds ratio: 0.69, 95% confidence dures when compared individually to
remainder of the cohort, P , 0.001). interval, 0.49–0.96), fondaparinux patients who were treated with all
A multivariate regression analysis (odds ratio: 0.85, 95% confidence other types of DVT prophylaxis. Pa-
was performed, adjusting for age, sex, interval, 0.76–0.95), and rivaroxaban tients on warfarin (odds ratio: 3.60,
hypercoagulability, and CCI, compar- (odds ratio: 0.57, 95% confidence 95% confidence interval, 3.38–3.84)
ing each type of anticoagulation to the interval, 0.51–0.63) were markedly and enoxaparin (odds ratio: 1.14,

Month 2018, Vol 00, No 00 5

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
Trends in Deep Vein Thrombosis Prophylaxis
Table 6
Odds Ratio of Deep Vein Thrombosis Within 6 Months of Total Hip
Arthroplasty or Total Knee Arthroplasty Procedure by Anticoagulant Type
Adjusted for Age, Sex, Hypercoagulability Diagnosis, and Charlson
Comorbidity Index
Postoperative Medication Odds Ratio
Prescription aspirin
Warfarin
Enoxaparin
Fondaparinux
Rivaroxaban
Dabigatran
95% confidence interval, 1.09–1.20)
were markedly more likely to have a
DVT when compared individually to
all other forms of anticoagulation
during the 6 months after THA or
TKA procedures (Table 6).
Conclusion
Patients are in a hypercoagulable state
after THA and TKA procedures.2-4,20
DVT after an elective procedure such
as a THA or TKA can have serious
ramifications including pulmonary
embolism and, in extreme situations,
death. Variation exists in the American
Academy of Orthopaedic Surgeons
and American College of Chest
Physicians recommendations on op-
timal postoperative anticoagulation
management of these patients.6 Con-
sequently, physician preference and
clinical decision making are the
primary determinants of a patient’s
postoperative anticoagulant selection.
This investigation demonstrated the
changes between year 2004 and year
2013 in surgeon preference for post-
operative DVT prophylaxis in pa-
tients undergoing THA and TKA
procedures. Limiting the study to this
period allowed this investigation to
obtain a large sample size and observe
the changes in practice patterns seen
with the introduction of novel post-
operative DVT prophylaxis agents.
6 Journal of the American Academy of Orthopaedic Surgeons
Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
95% Confidence
Interval (Lower, Upper)
0.69 0.49, 0.96
3.60 3.38, 3.84
1.14 1.09, 1.20
0.85 0.76, 0.95
0.57 0.51, 0.63
1.09 0.80, 1.47
During the study period, a notable
decrease was observed in patients pre-
scribed warfarin and an increase seen
in patients prescribed rivaroxaban.
During the same period, no statistically
significant change was noted in the
overall DVT rates in the 6 months after
THA or TKA. Patients with a hyper-
coagulable diagnosis had markedly
higher rates of DVT within 6 months
of a THA or TKA procedure compared
with those patients without hyperco-
agulable diagnosis. No other clinically
significant risk factors were identified
to distinguish patients at a high risk of
DVT. Furthermore, a greater portion
of patients with hypercoagulable diag-
nosis were treated with warfarin and
lovenox (Table 5).
Vitamin K antagonists such as war-
farin pose a variety of challenges,
which have likely contributed to its
decreasing popularity. Patients must
be closely monitored to ensure that
their international normalized ratio
(INR) is within the therapeutic range.
If the INR is above or below the
therapeutic range, patients will have
to adjust their doses accordingly. In
addition, dietary changes can influ-
ence the metabolism of the drug
subsequently affecting the overall
effectiveness and influence on the
INR.14,21 Warfarin has been classi-
cally considered to be relatively low
cost; however, after taking into
account the cost of laboratory mon-
itoring, the overall cost of warfarin
prophylaxis has been shown to actu-
ally be greater than prophylaxis
with low-molecular-weight hepa-
rin.22 However, warfarin provides
the greatest advantage over other
types of anticoagulant because it can
be readily reversed if bleeding or
anticoagulation-related complica-
tions arise. Newer medications, such
as factor Xa inhibitors and direct
thrombin inhibitors are attractive
solutions because they can be ad-
ministered orally, do not require
laboratory monitoring, and have
predictable pharmacokinetics irre-
spective of body mass or diet.23,24
Downsides of these medications
include cost and the only option to
reverse the drug with fresh frozen
plasma if bleeding complications
arise; however, studies have shown
no increase in wound complication
and thromboembolic rates in factor
Xa inhibitors compared with other
forms of anticoagulation.25
Aspirin is appealing because of its
low cost, oral administration, pre-
dictable pharmacokinetics, and easy
access over the counter. A recent meta-
analysis showed that patients under-
going THA or TKA did not have any
difference in proximal DVT risk when
comparing aspirin versus other anti-
coagulants.26 However, this analysis
included studies, which performed
screening for DVT on all patients; this
practice is neither practical nor rec-
ommended.5,6,11 In this investigation,
information was collected on pre-
scribed aspirin claims. The number of
patients taking aspirin as the primary
prophylaxis after TKA and THA
may be higher than reported in this
investigation because this drug is also
available without a prescription. In
a subanalysis, we included data on
129,505 patients who did not have
any prescription anticoagulant. This
cohort did not have markedly higher
DVT rates within 6 months of
THA or TKA procedures compared
with prescription anticoagulants. It

is possible that these patients
were taking over-the-counter anti-
coagulants; however, this cannot be
confirmed with the available data.
The study found markedly increased
DVT rates in the 6 months after THA
and TKA procedures in patients treated
with warfarin (odds ratio: 3.60, 95%
confidence interval, 3.38–3.84) and
enoxaparin (odds ratio: 1.14, 95%
confidence interval, 1.09–1.20). The
exact reason for the increased DVT
rate in this subgroup cannot be
determined from this study and
beyond the scope of this investigation.
This study focused on claims in-
formation and did not have any
information on INR levels or patient
compliance. Studies have demon-
strated that patients are within the
therapeutic range only 45.9% of the
time postoperatively.27 Enoxaparin
is administered subcutaneously and
provides challenges inherent to peo-
ple hesitant to administer the drug
on themselves; however, this has
not been shown to prevent compli-
ance.28 It is unclear whether the in-
creased rate of DVT observed is due
to failure of patient compliance or
another confounding variable unac-
counted for in the regression model.
Importantly, the results in this
study represent the actual observed
population-level rates over a decade
of time with the use of various pro-
phylactic regimens. Consequently,
these findings are informative to
both surgeons and patients as to the
expected population rates of DVT.
This study has limitations inherent
to any retrospective review. The
database is likely to represent a rep-
resentative sample of the overall pop-
ulation but does not include patients
without insurance or on Medicaid.
The information in the database is
limited to the specified time; however,
this provides the benefit of a large
sample size. Patients who did not have
medication claims data were also
excluded from this analysis. Medi-
cations that were obtained over the
Month 2018, Vol 00, No 00
Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
counter were also not captured by the
database. Patient compliance with
medications could not be determined
from our data because the usual
course of anticoagulants prescribed
after a THA or TKA is too brief
to calculate medication possession
ratios.
Overall, this investigation demon-
strates, at a population level, recent
trends in DVT prophylaxis after
THA and TKA procedures. Although
warfarin is still the most commonly
prescribed prophylaxis after THA
and TKA procedures, use of this
agent is decreasing. Patients with a
history of DVT or hypercoagulability
are at a higher risk of postoperative
DVT, regardless of the type of
prophylaxis prescribed. After ad-
justing for age, sex, and hypercoag-
ulability diagnosis, the likelihood
of DVT within 6 months of THA
and TKA procedures was markedly
increased in patients treated with
warfarin and lovenox and markedly
decreased in those treated with aspi-
rin, fondaparinux, and rivaroxaban.
Given the sample size of the data
over a prolonged period in the actual
observed clinical setting, this infor-
mation is useful for surgeons when
counseling their patients on the dif-
ferent types of anticoagulants avail-
able before undergoing THA or TKA
procedures.
References
References printed in bold type are
those published within the past 5 years.
1. Maradit Kremers H, Larson DR, Crowson CS,
et al: Prevalence of total hip and knee
replacement in the United States. J Bone Joint
Surg Am 2015;97:1386-1397.
2. Visuri T, Pulkkinen P, Paavolainen P,
Koskenvuo M, Turula KB: Causes of death
after hip arthroplasty in primary arthrosis.
J Arthroplasty 1997;12:397-402.
3. Harris WH, Sledge CB: Total hip and total
knee replacement. N Engl J Med 1990;323:
801-807.
4. Sculco T, Colwell C, Pellegrini V, Westrich G,
Bottner F: Prophylaxis against venous
Harpreet Bawa, MD, et al
thromboembolic disease in patients having a
total hip or knee arthroplasty. J Bone Joint
Surg 2002;84:466-477.
5. Falck-Ytter Y, Francis CW, Johanson NA,
et al: Prevention of VTE in orthopedic
surgery patients: Antithrombotic Therapy
and Prevention of Thrombosis, 9th ed:
American College of Chest Physicians
Evidence-Based Clinical Practice
Guidelines. Chest 2012;141:e278S-e325S.
6. Eikelboom JW, Karthikeyan G, Fagel N,
Hirsh J: American Association of
Orthopedic Surgeons and American
College of Chest Physicians guidelines for
venous thromboembolism prevention in hip
and knee arthroplasty differ: What are the
implications for clinicians and patients?
Chest 2009;135:513-520.
7. Barrack RL: Current guidelines for total
joint VTE prophylaxis: Dawn of a new day.
J Bone Joint Surg Br 2012;94:3-7.
8. McDougall CJ, Gray HS, Simpson PM,
Whitehouse SL, Crawford RW, Donnelly WJ:
Complications related to therapeutic
anticoagulation in total hip arthroplasty.
J Arthroplasty 2013;28:187-192.
9. Parvizi J, Ghanem E, Joshi A, Sharkey PF,
Hozack WJ, Rothman RH: Does
“excessive” anticoagulation predispose to
periprosthetic infection? J Arthroplasty
2007;22:24-28.
10. Pulido L, Ghanem E, Joshi A, Purtill JJ,
Parvizi J: Periprosthetic joint infection: The
incidence, timing, and predisposing factors.
Clin Orthop Relat Res 2008;466:
1710-1715.
11. Johanson NA, Lachiewicz PF, Lieberman JR,
et al: American Academy of Orthopaedic
Surgeons Clinical Practice guideline on:
Prevention of symptomatic pulmonary
embolism in patients undergoing total hip or
knee arthroplasty. J Bone Joint Surg Am
2009;91:1756-1757.
12. Francis CW, Pellegrini VD Jr, Marder VJ,
et al: Comparison of warfarin and external
pneumatic compression in prevention
of venous thrombosis after total hip
replacement. JAMA 1992;267:
2911-2915.
13. Nam D, Sadhu A, Hirsh J, Keeney JA,
Nunley RM, Barrack RL: The use of
warfarin for DVT prophylaxis following
hip and knee arthroplasty: How often are
patients within their target INR range? J
Arthroplast 2015;30:315-319.
14. Aynardi M, Brown PB, Post Z, Orozco F,
Ong A: Warfarin for thromboprophylaxis
following total joint arthroplasty: Are
patients safely anti-coagulated? J
Arthroplasty 2013;28:1251-1253.
15. Prevention of pulmonary embolism
and deep vein thrombosis with low
dose aspirin: Pulmonary embolism
prevention (PEP) trial. Lancet 2015;
355:1295-1302.
7
Trends in Deep Vein Thrombosis Prophylaxis
16. Colwell CW, Hardwick ME: Rationale for
low-molecular-weight heparin prophylaxis
after total knee arthroplasty. Clin Orthop
Relat Res 2006;452:181-185.
17. Stewart DW, Freshour JE: Aspirin for the
prophylaxis of venous thromboembolic
events in orthopedic surgery patients: A
comparison of the AAOS and ACCP
guidelines with review of the evidence. Ann
Pharmacother 2013;47:63-74.
18. Colwell CW: Managing thromboembolic
risk in hip and knee arthroplasty: State of
the art. Orthopedics 2003;26:S231-S236.
19. Quan H, Sundararajan V, Halfon P,
et al: Coding algorithms for defining
comorbidities in ICD-9-CM and ICD-10
administrative data. Med Care 2005;43:
1130-1139.
20. Warwick D, Williams MH, Bannister GC:
Death and thromboembolic disease after
total hip replacement: A series of 1162 cases
with no routine chemical prophylaxis. J
Bone Joint Surg Br 1995;77:6-10.
8 Journal of the American Academy of Orthopaedic Surgeons
Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.
21. Asnis PD. Gardner MJ, Ranawat A,
Leitzes AH, Peterson MGE, Bass AR: The
effectiveness of warfarin dosing from a
nomogram compared with house staff
dosing. J Arthroplasty 2007;22:213-218.
22. Caprini JA, Botteman MF, Stephens JM,
et al: Economic burden of long-term
complications of deep vein thrombosis after
total hip replacement surgery in the United
States. Value Health 2003;6:59-74.
23. Sobieraj DM, Coleman CI, Tongbram V,
et al: Comparative effectiveness of low-
molecular-weight heparins versus other
anticoagulants in major orthopedic surgery:
A systematic review and meta-analysis.
Bone 2012;51:376-382.
24. Kinov P, Tanchev PP, Ellis M, Volpin G:
Antithrombotic prophylaxis in major
orthopaedic surgery: An historical overview
and update of current recommendations.
Int Orthop 2014;38:169-175.
25. Kim SM, Moon YW, Lim SJ, Kim DW,
Park YS: Effect of oral factor Xa inhibitor
and low-molecular-weight heparin on
surgical complications following total hip
arthroplasty. Thromb Haemost 2016;115:
600-607.
26. Drescher FS, Sirovich BE, Lee A,
Morrison DH, Chiang WH, Larson RJ:
Aspirin versus anticoagulation for
prevention of venous thromboembolism
major lower extremity orthopedic
surgery: A systematic review and
meta-analysis. J Hosp Med 2014;9:
579-585.
27. Nam D, Nunley RM, Johnson SR, Keeney JA,
Clohisy JC, Barrack RL: Thromboembolism
prophylaxis in hip arthroplasty: Routine and
high risk patients. J Arthroplasty 2015;30:
2299-2303.
28. Bergqvist D, Arcelus JI, Felicissimo P;
ETHOS Investigators: Evaluation of
the duration of thromboembolic
prophylaxis after high-risk orthopaedic
surgery: The ETHOS observational
study. Thromb Haemost 2012;107:
270-279.

Robert L. Buly, MD, MS
Branden R. Sosa, HS
Lazaros A. Poultsides, MD, MSc,
PhD
Elaine Caldwell, BS, RN
S. Robert Rozbruch, MD
From the Hospital for Special Surgery,
New York, NY.
Correspondence to Dr. Buly:
bulyr@hss.edu
J Am Acad Orthop Surg 2018;00:1-10
DOI: 10.5435/JAAOS-D-17-00623
Copyright © 2018 The Author(s).
Published by Wolters Kluwer Health,
Inc. on behalf of the American
Academy of Orthopaedic
Surgeons.This is an open access
article distributed under the terms of
the Creative Commons Attribution-
NonCommercial-NoDerivatives
License 4.0 (CC BY-NC-ND), which
permits downloading and sharing the
work provided it is properly cited. The
work cannot be changed in any way or
used commercially without permission
from the journal.
Month 2018, Vol 00, No 00
Research Article
Femoral Derotation Osteotomy in
Adults for Version Abnormalities
Abstract
Background: Version abnormalities of the femur can cause pain and
hip joint damage due to impingement or instability. A retrospective
clinical review was conducted on patients undergoing a
subtrochanteric derotation osteotomy for either excessive
anteversion or retroversion of the femur.
Methods: A total of 55 derotation osteotomies were performed in 43
patients: 36 females and 7 males. The average age was 29 years
(range, 14 to 59 years). The osteotomies were performed closed with
an intramedullary saw. Fixation was performed with a variety of
intramedullary nails. Twenty-nine percent of patients had a
retroversion deformity (average, 29° of retroversion; range, 12°
to 223°) and 71% had excessive anteversion of the femur
(average, 137° of anteversion; range, 122° to 153°). The etiology
was posttraumatic in 5 patients (12%), diplegic cerebral palsy in 2
patients (5%), Prader-Willi syndrome in 1 patient (2%), and idiopathic
in 35 patients (81%). Forty-nine percent underwent concomitant
surgery with the index femoral derotation osteotomy, including hip
arthroscopy in 40%, tibial derotation osteotomy in 13%, and a
periacetabular osteotomy in 5%. Tibial osteotomies were performed
to correct a compensatory excessive external tibial torsion that would
be exacerbated in the correction of excessive femoral anteversion.
Results: No patient was lost to follow-up. Failures occurred in three
hips in three patients (5%): two hip arthroplasties and one nonunion
that healed after rerodding. There was one late infection treated
successfully with implant removal and antibiotics with an excellent
final clinical outcome. At an average follow-up of 6.5 years (range, 2 to
19.7 years), the modified Harris Hip Score improved by 29 points in the
remaining 52 cases (P , 0.001, Wilcoxon signed-rank test). The
results were rated as excellent in 75%, good in 23%, and fair in 2%.
Subsequent surgery was required in 78% of hips, 91% of which were
implant removals.
Conclusions: A closed, subtrochanteric derotation osteotomy of the
femur is a safe and effective procedure to treat either femoral
retroversion or excessive anteversion. Excellent or good results were
obtained in 93%, despite the need for subsequent implant removal in
more than two-thirds of the patients.
I n the surgical treatment of hip
disorders, a major cause of failure
is either insufficient correction or a
failure to fully recognize the under-
lying deformities causing pain and
joint damage.1-3 One type of femoral
1
Femoral Derotation Osteotomy in Adults

deformity that is still frequently
overlooked are rotational defor-
mities of the femur, that is, excessive
anteversion or femoral retroversion.
These rotational deformities may
occur alone or may coexist with
acetabular dysplasia4-6 or various
types of hip impingement.2,5,7-9
Rotational deformities may also be
associated with cerebral palsy10 and
labral tears11 and are not unusual
after the fixation of femoral shaft
fractures.12-14
Excessive femoral anteversion can
cause instability, damage of the artic-
ular cartilage and acetabular labrum,
and eventually osteoarthritis.15-17
Furthermore, it can cause a decrease
in the length of the abductor lever
arm, 18 posterior extra-articular
impingement,9 and ischiofemoral
impingement.19 Finally, excessive fem-
oral anteversion may cause increased
hip and knee adduction moments, an
intoeing gait and patellofemoral mal-
tracking, with resultant knee pain and
arthritis.20-22
Femoral retroversion, on the other
hand, causes damage due to impinge-
ment between the femoral neck and
acetabulum, which may result in dam-
age to the labrum and articular carti-
lage, ultimately resulting in
osteoarthritis of the hip.16,23 Other
potential retroversion problems in-
clude an increased risk of slipped
capital femoral epiphysis24 and sus-
ceptibility to a traumatic posterior hip
dislocation.25,26 Residual, untreated
femoral retroversion may be a reason
why hip preserving surgeries may fail,
especially after the arthroscopic treat-
ment of hip impingement.1-3
Despite the important role of fem-
oral rotational deformity in the
pathogenesis of hip disease, there is
little written about the outcomes of
treatment. This article describes the
technique and outcomes of a closed,
derotation osteotomy of the femur
to correct either excessive femoral
anteversion or retroversion as part
of a hip preservation effort. The
question to be answered is whether
hip pain as a result of version abnor-
malities of the femur can be alleviated
by this type of surgery.
Methods
Patients selected for the index proce-
dure had hip pain secondary to
increased femoral anteversion or
femoral retroversion. Surgery was
offered if the patient had failed all
nonsurgical treatment measures and
if the range of motion aberration
correlated with the version abnor-
mality (ie, excessive hip internal rota-
tion with excessive anteversion or a
lack of internal rotation associated
with femoral retroversion). Patients
with coxa vara (a neck-shaft angle
of ,125°) or coxa valga (a neck-shaft
angle of .140°) were excluded, with
the understanding that a varus or
valgus derotation intertrochanteric
osteotomy would be more appropri-
ate to address the concomitant neck-
shaft angulation.27
In addition to plain radiographs, all
patients underwent preoperative
MRI to assess the condition of the
articular cartilage and labrum and
three-dimensional CT scans to accu-
rately define the anatomic deform-
ities. The measurement of femoral
version was performed by the CT
technique described by Murphy
et al.28 All readings and measure-
ments were performed by board-
certified musculoskeletal radiologists.
The osteotomies were performed
with the patient in the supine position
under regional, hypotensive anes-
thesia. The operated leg was draped
free, and traction was not used. An
intramedullary hand saw was used
that did not require exposure of the
osteotomy site.14 A small, longitu-
dinal skin incision was made just
proximal to the greater trochanter.
The isthmus of the femur was over-
reamed by 0.5 mm in accordance
with the nail to be used. The
subtrochanteric region was then
reamed 0.5 mm larger than the
diameter of the proposed intra-
medullary saw. Rotational control
was achieved by placing 1/8-inch
smooth Steinmann pins into the
femur proximal and distal to the
osteotomy in the desired amount
of rotational correction (Figure 1).
The location and progress of the
osteotomy were controlled by
fluoroscopy (Figure 2). The angular
correction was controlled visually by
using flat, triangular guides from a
blade plate instrument set (Figure 3).
The osteotomy was performed
in the subtrochanteric region by
inserting the hand saw, which was
then rotated in a stepwise fashion with
progressive protrusion of the blade
from the cam. The distal fragment was
then rotated to align the two pins
parallel, thus effecting the rotational
correction. The goal was to achieve
approximately 15° of femoral ante-
version. Fixation was then achieved
using a variety of trochanteric entry
intramedullary nails that were locked
proximal and distal to the osteotomy.

Dr. Buly or an immediate family member has stock or stock options held in Blue Belt Technology and serves as a board member, owner,
officer, or committee member of the Maurice Mueller Foundation of North America and the International Society for Hip Arthroscopy.
Dr. Rozburch or an immediate family member has received royalties from Stryker; is a member of a speakers’ bureau or has made paid
presentations on behalf of NuVasive, Smith & Nephew, and Stryker; serves as a paid consultant to NuVasive, Smith & Nephew, and Stryker;
and serves as a board member, owner, officer, or committee member of the Limb Lengthening Reconstruction Society. None of the following
authors or any immediate family member has received anything of value from or has stock or stock options held in a commercial company or
institution related directly or indirectly to the subject of this article: Mr. Sosa, Dr. Poultsides, and Ms. Caldwell.

2 Journal of the American Academy of Orthopaedic Surgeons


Figure 1
Schematic illustration demonstrating the osteotomy technique.
The intramedullary devices used were
42 TriGen Trochanteric Antegrade
Nails (Smith & Nephew), five Tro-
chanteric Fixation Nails and three
Intramedullary Femoral Nails (DePuy
Synthes), four Gamma Nails (Stryker),
and one piriformis fossa entry
Phoenix Femoral Nail (Zimmer
Biomet).
Concomitant hip arthroscopy was
performed just prior to the osteotomy
if the magnetic resonance image re-
vealed labral and/or articular carti-
lage lesions or the presence of a cam
lesion of the femoral neck that would
impinge if a retroverting derotation
femoral osteotomy was to be per-
formed for excessive anteversion.
A concomitant periacetabular
osteotomy was performed at the
same setting if there was coexisting,
severe dysplasia that required cor-
rection along with the femoral ver-
Month 2018, Vol 00, No 00
sion. The periacetabular osteotomy
was performed first with the same
preparation and drape setup used for
both procedures.
A concomitant tibial/fibular oste-
otomy was performed if the patient
had a compensatory external tibial
torsion coexisting with excessive
femoral anteversion, as described by
Tönnis and Heinecke.15 This proce-
dure was done to prevent an exag-
gerated external foot progression
angle that would result from der-
otating the excessively anteverted
femur in patients with this rotational
deformity. The tibia was either
internally rotated with gradual cor-
rection using an external hexapod
frame in the supramalleolar or
proximal tibial regions or corrected
acutely over an intramedullary nail,
depending on the morphology of the
tibial deformity.
Robert L. Buly, MD, MS, et al
Postoperatively, epidural patient-
controlled anesthesia was used if a
tibial osteotomy was not performed.
Intravenous patient-controlled anes-
thesia was used instead with tibial
osteotomies to allow monitoring of
the lower leg and vigilance regarding
a possible compartment syndrome.
No braces or casts were used after
surgery. There were no range of
motion restrictions. Weight bearing
as tolerated was permitted with
crutches unless a concomitant peri-
acetabular or tibial osteotomy was
performed, in which case the weight
bearing was restricted to 20% for
6 weeks. Follow-up examinations
with AP and lateral radiographs were
performed at 6 weeks, 3 months,
6 months, and 1 year after surgery.
The modified Harris Hip Score
(mHHS) was used, and scores were
documented before surgery and at
3
Femoral Derotation Osteotomy in Adults

Figure 2

A, Photograph showing the Winquist intramedullary saw. B, Photograph showing the saw blade extended from the cam
mechanism. C–E, Intraoperative fluoroscopic images showing gradual transection of the lateral and medial cortices,
followed by osteotomy completion.

the latest follow-up. The minimum
follow-up time was 24 months.
Results
Starting in 1997, a total of 81 femoral
osteotomies have been performed in
67 patients. Forty-three patients
having undergone 55 derotation
osteotomies had a minimum follow-
up of 2 years. All bilateral cases were
done staged. There were 36 females
and 7 males; the average age was 29.0
years (range, 14 to 59 years).
The deformity etiology was post-
traumatic in 5 patients (12%), diple-
gic cerebral palsy in 2 patients (5%),
Prader-Willi syndrome in 1 patient
(2%), and idiopathic in 35 patients
(81%) (Table 1). All hips had a
Tönnis arthritis grade of zero (no
evidence of arthritis).15 Twenty-nine
percent of the hips (16 hips in 14
patients) had a retroversion defor-
mity (average, 29° of retroversion;
range, 12° anteversion to 224°
retroversion). Seventy-one percent
(39 hips in 29 patients) had excessive
anteversion of the femur (average, 1
37° of anteversion; range, 122°
to 153° anteversion). The average
rotational correction was 24° for the
retroverted hips (range, 18° to 35°)
and 23° (range, 15° to 40°) for an-
teverted hips (Table 2). All 16 ret-
roverted hips were considered to
have “severely diminished ante-
version” by the criteria of Tönnis,

4 Journal of the American Academy of Orthopaedic Surgeons


Figure 3
A, Triangles used to set the degree of rotation correction. B, A 20° triangle was used to set the correction between the proximal
and distal pins in a case of excessive anteversion. C, The femoral nail is inserted while maintaining rotational correction.
whereas 37 of 38 excessively ante-
verted hips (97%) were considered
to be “severely increased” (.25° of
anteversion), with 1 (3%) being
“moderately increased” (21° to 25°
of anteversion).15
The clinical hip range of motion
assessment for all patients included
measuring internal and external ro-
tations, with the hip flexed to 90°.
Retroverted hips typically lacked or
had diminished internal rotation.
Conversely, hips with excessive
anteversion had more internal ro-
tation than external rotation. For
the cases with excessive anteversion,
the preoperative average of internal
rotation at 90° of hip flexion
was 173° (range, 145° to 110°) and
external rotation 122° (range, 25°
to 160°). After osteotomy, the
internal rotation diminished to 126°
(range, 15° to 45°), whereas the
external rotation improved to 148°
(range, 115° to 70°). This was
significant at P , 0.01 (Wilcoxon
signed-rank test). For the cases of
retroversion, the preoperative average
of internal rotation at 90° of hip
flexion was 21° (range, 220° to
110°) and external rotation 181°
(range, 150° to 190°). After oste-
otomy, the internal rotation improved
to 123° (range, 110° to 135°),
whereas the external rotation dimin-
ished to 142° (range, 130° to 50°).
This was significant at P , 0.001
(Wilcoxon signed-rank test).
Month 2018, Vol 00, No 00
Previous surgery had been per-
formed in 26 hips (47%) (Table 1).
Twenty-seven hips (49%) underwent
concomitant surgery with the index
femoral derotation osteotomy, in-
cluding hip arthroscopy with labral
débridement and chondroplasty in
16 (29%), 6 hip arthroscopies with
an additional femoral osteochon-
droplasty (11%), a tibial derotation
osteotomy in 7 (13%), and an ipsi-
lateral periacetabular osteotomy with
3 of the femoral osteotomies (5%)
(Table 1). Three of the ipsilateral
tibial osteotomies were performed in
the supramalleolar region and two in
the proximal tibia, with external frame
fixation. Two of the tibial osteotomies
were performed at midshaft with
immediate rotational correction and
fixation with an intramedullary nail.
No patient was lost to follow-up.
One patient died of cancer 12.7 years
after surgery with a hip score of 85. The
average time to femoral osteotomy
union was 3.3 months (range, 2 to
16 months). All the tibial and pelvic
osteotomies healed uneventfully.
At an average follow-up of 6.5 years
(range, 2 to 19.7 years), the mHHS
improved by 27 points (P , 0.001,
Wilcoxon signed-rank test) from 66
to 93 points. When taken separately,
there was a statistically significant
improvement in both the groups with
either retroversion or excessive ante-
version (Table 2). The results were
rated as excellent in 70%, good in
Robert L. Buly, MD, MS, et al
23%, and fair in 7%, including the
revision femoral osteotomy and two
total hip replacements.
Failure was defined as conversion to
total hip arthroplasty, refixation of the
index osteotomy, or an mHHS of ,70.
Failures occurred in three hips in three
patients (5%): two hip arthroplasties
and one rerodding for a femoral
nonunion. A total hip arthroplasty was
performed 46 months after osteotomy
in a 46-year-old woman with Ehlers-
Danlos syndrome with only minimal
osteoarthritic change seen on MRI
and a normal joint space on plain
radiographs (Tönnis stage zero). The
patient continues to do well with the
contralateral osteotomy and has an
mHHS of 74 points. Another hip
replacement was performed in an 18-
year-old man with Prader-Willi syn-
drome 15 months after the index
procedure because of the failure of
the concomitant periacetabular oste-
otomy. The third failure was in a 26-
year-old woman with Ehlers-Danlos
syndrome with a nonunion that
was rerodded successfully. All three
presented initially with excessive
anteversion.
Subsequent surgery was required in
78% of hips, 39 of 43 (91%) were
implant removals. The implant was
removed in patients with radiographic
evidence of bone union and only if
there was notable pain refractory to
nonsurgical treatment, usually irrita-
tion from the screw heads or a thigh
5
Femoral Derotation Osteotomy in Adults

Table 1 after the revision femoral osteotomy,

Summary of Results
Total patients
Average age
Total osteotomies
Excessively anteverted
Retroverted
Etiology (patients)
Idiopathic
Posttraumatic
Cerebral palsy
Prader-Willi syndrome
Previous surgery (hips)
Total
Hip arthroscopy 1 femoral
osteochondroplasty
Open reduction 1 internal fixation
Previous femoral osteotomy
Hip arthroscopy 1 labral débridement
Femoral lengthening
Slipped capital femoral epiphysis pinning
Concomitant surgery (hips)
Total: all ipsilateral
Hip arthroscopy 1 labrum 1
chondroplasty
Tibial derotation osteotomy
Hip arthroscopy 1 osteochondroplasty
Periacetabular osteotomy
ache that occurred with loading or
unloading the femur (Figure 4). A
revision femoral osteotomy was per-
formed in one hip (2%) by another
surgeon to add additional rotational
the mHHS was 85 points.
A late infection occurred in one
43 (36 females and 7 males) femur of a 14-year-old girl 8 months
29.0 y (range, 14 to 59 y) after surgery on the second femur.
55 The organism was a minimally re-
39 (71%) sistant Staphylococcus aureus and
16 (29%) was treated successfully with im-
plant removal and antibiotics. At
35 (81%) follow-up, the patient’s mHHS was
5 (12%) 100 points.
2 (5%)
1 (2%)
Discussion
26 (47%)
13 (24%) It has been reported that osteo-
arthritis may occur with either
5 (9%) femoral retroversion or increased
3 (5%) anteversion.15-17,23,29,30 Retrover-
3 (5%) sion of the femur, either alone or in
1 (2%) combination with other defor-
1 (2%) mities, can cause hip damage sec-
ondary to impingement.16,23,29,30
27 (49%) The damage with excessive ante-
16 (29%) version occurs at the periphery of the
acetabulum secondary to high com-
7 (13%)
pressive and shear forces on the
6 (11%) articular cartilage and labrum, caus-
3 (5%) ing hip pain and arthritis.15-17
The onset of pain in patients with
dysplasia occurs earlier if there
stability while performing a surgical
is coexistent excessive combined
hip dislocation in a 39-year-old anteversion.4 Increased femoral an-
woman with anterior capsular defi- teversion has also been associated
ciency and instability after four pre- with posterior greater trochanteric
vious surgeries. Currently, 9 months impingement,9,31 decreased abductor

Table 2
Comparison of the Anteverted Versus Retroverted Cases
Periacetabular Tibial
Hips Osteotomy Osteotomy Average Average Preoperative Postoperative
Condition (%) (%) (%) Deformity Correction mHHS mHHS

Excessive 39 (71) 3 (8) 7 (18) 137° (122°653°) 23° (15°–40°) 64 94 (P , 0.01)a

femoral
anteversion
Femoral 16 (29) 0 (0) 0 (0) 29° (12° to 224°) 24° (18°–35°) 70 96 (P , 0.05)a
retroversion
Total 55 (43) 3 (5) 7 (13) — — 65 94 (P , 0.001)a
(patients)

mHHS = modified Harris Hip Score

a
Wilcoxon signed-rank test.

6 Journal of the American Academy of Orthopaedic Surgeons

 Robert L. Buly, MD, MS, et al

power by ,28% due to diminished Figure 4

femoral offset,18 and hip instability.32
Psoas irritation may be due to ante-
rior instability; the tendon may act
as a dynamic stabilizer and releasing
it may exacerbate the problem.2,33
Patients with symptomatic ischiofe-
moral impingement with diminished
clearance between these two structures
are more likely to have excessive
femoral anteversion compared to
asymptomatic patients.34 Other prob-
lems associated with excessive femoral
anteversion include increased hip and
knee adduction moments, an intoeing
gait and patellofemoral maltracking,
pain, and arthritis.20-22
The arthroscopic treatment of hip
impingement may fail if bony
débridement is inadequate. In addi-
tion, these procedures may also fail if
coexisting femoral retroversion is
not detected or treated.2,35 Six of the
patients in this study with retrover-
sion of the femur had previously
undergone an arthroscopic débride-
ment with initial symptomatic relief,
but eventually had a relapse of
painful impingement and a lack of
internal rotation.
There is no uniform agreement as to
the value of normal femoral and ace-
tabular version. The reported values
for the acetabulum range from 13° to
20° of anteversion in three studies,
averaging 17°.30,36,37 For the femur,
the range of anteversion is 10° to 20°
in three studies, averaging 15°.15,37,38
The McKibbin index is the sum of
acetabular version and femoral ver-
sion; “normal” is approximately 30°, AP (A) and lateral (B) radiographs of a 17 year-old female who presented with
excessive anteversion, showing a healed femoral osteotomy at 12 months
and values .60° are considered to be
postoperatively. The intramedullary nail was subsequently removed.
highly unstable.15 The goal of surgi-
cal correction was to approach 15° of
femoral version. Three-dimensional abnormal neck-shaft angle (ie, coxa also no shortening of the abductor
CT scans were used to measure ver- vara or coxa valga) and rotational muscle fibers as occurs with a varus
sion as precisely as possible and have deformities by rotating the distal producing intertrochanteric oste-
long been considered the benchmark fragment by the desired amount of otomy. Because of the much lower
technique to measure version.39 correction before applying the blade profile compared to a blade plate,
Version abnormalities of the femur plate.27 This technique may not be there can be considerably less peri-
can also be treated with an inter- necessary if the neck-shaft angle is trochanteric bursitis and pain. A
trochanteric osteotomy. These tech- normal, and the only femoral defor- subtrochanteric derotation femoral
niques allow the correction of an mity is purely rotational. There is osteotomy may also be performed by

Month 2018, Vol 00, No 00 7

Femoral Derotation Osteotomy in Adults
plating, but it requires a much more
invasive approach.40 The advantage
of the described technique is that it
allows for a much less invasive
approach, lessening surgical mor-
bidity and theoretically a lower
chance of infection. In addition, the
vastus lateralis is not dissected from
the femur, maintaining more of the
periosteal blood supply to enhance
bone union. In all the cases, it was
not necessary to expose the oste-
otomy site because the transection
was performed with an intra-
medullary bone saw. Other advan-
tages include the ability to allow
weight bearing as tolerated immedi-
ately because the fixation is provided
with a locked intramedullary nail.
Placing the distal interlocking screw
in the dynamic mode allows com-
pression at the osteotomy site with
weight bearing. In contrast, patients
treated with an intertrochanteric
osteotomy and plating are main-
tained at 20% weight bearing for at
least 6 weeks after surgery. In
addition, a pure derotation oste-
otomy performed in the subtro-
chanteric region does not deform the
proximal femur. Should a total hip
arthroplasty be required in the
future, it does not hamper stem
insertion as can occur after a previ-
ous intertrochanteric osteotomy.
The disadvantage of the described
technique is that bone healing is
slower, averaging 3 to 4 months, and
in some cases even longer, which may
be due to the diminished healing
potential of cortical bone versus
cancellous bone. In addition, there is
much less surface area at the site of
the transverse subtrochanteric oste-
otomy than with an intertrochanteric
or supracondylar type. Another dis-
advantage is potential damage to the
hip abductors because of the reaming
necessary to insert a nail. Care was
taken to enter the greater trochanter
through the posterosuperior “bare
area” if possible to leave the
abductors minimally disrupted. At
8 Journal of the American Academy of Orthopaedic Surgeons
follow-up, no patient had notable
hip abductor weakness or Trende-
lenburg limp or sign.
It was necessary to perform a con-
comitant tibial osteotomy in seven cases
(13%) with excessive femoral ante-
version and a compensatory external
tibial torsion instead of the usual
intoeing gait associated with excessive
anteversion, dubbed as “miserable
malalignment syndrome.”21 Surgical
correction of increased femoral ante-
version requires externally rotating the
distal fragment. In these patients, there
would have been a greatly exaggerated
external foot progression angle.
A concomitant periacetabular oste-
otomy was performed in patients with
severe acetabular dysplasia and coex-
isting femoral malrotation where it was
felt that correction of only one or the
other would leave a notable deformity
that is often an indication for surgery
when occurring alone. This procedure
was performed in three patients (5%).
Concomitant hip arthroscopy was
performed for two reasons: to address
intra-articular pathology (ie, torn
acetabular labrum and articular car-
tilage damage) that would ordinarily
not be accessed during the osteotomy
and to remove a sizable cam lesion
that would impinge after a femoral
retroverting osteotomy.
The failures all occurred in patients
with excessive anteversion and con-
nective tissue disorders: Ehlers-
Danlos or Prader-Willi syndrome.
Interestingly, failure did not occur in
the contralateral osteotomy of the
two Ehlers-Danlos patients. Collagen
abnormalities associated with these
conditions may have contributed to
the problems of instability and poor
bone healing.
Although subsequent surgery was
required in 78% of hips, 93% of these
were implant removals. Overall, 70%
of patients underwent removal of the
implant. Although generally better
tolerated than a blade plate after
intertrochanteric osteotomy, most pa-
tients had either irritation from the in-
terlocking screws or a thigh ache that
resolved in most cases after implant
removal. Other than the hip arthro-
plasties or osteotomy revisions, the
remainder of cases were hip arthro-
scopic débridements in two patients.
Winquist14 reported the ability to
perform a closed osteotomy and in-
tramedullary nailing to correct sim-
ple rotational deformities.
Chapman et al13 reported closed
osteotomy nailing performed in 31
patients for leg-length inequality and
6 with rotational deformities. Preop-
erative rotational deformities aver-
aged 58° and all were corrected to
within 5° of normal.
Stahl et al12 treated 14 patients with
posttraumatic rotational deformities
of the femur that ranged from 26° to
63° with a closed technique over an
intramedullary nail. Postoperative CT
scans revealed excellent correction of
the deformity within 4° in all cases.
Kamath et al40 reported 28 rota-
tional femoral osteotomies in 26
patients, 93% for excessive femoral
anteversion. Clinical outcomes were
not reported. After two initial fail-
ures for nonunion, all subsequently
went on to union with refixation.
Pailhe et al41 reported nine der-
otation osteotomies in six adolescents
(average age, 13.6 years) for excessive
anteversion. The technique was done
with a distal supracondylar oste-
otomy and fixation with an antegrade
intramedullary nail. The average
correction was 19°. Patient-reported
outcome scores were not recorded.
All patients were satisfied or very
satisfied and had better foot pro-
gression angles and less internal
rotation on range of motion testing.
Putz et al10 performed 96 der-
otation femoral osteotomies (proxi-
mal or distal) in 63 adult cerebral
palsy patients with excessive ante-
version. Although patient-reported
outcome scores were not recorded,
the group experienced statistical
improvement in foot progression
angle and passive and stance range of

motion. Tibial rotation osteotomy
was required in 16.7% of cases to
compensate for excessive external
tibial torsion.10 In the present study, a
similar need for concomitant tibial
derotation osteotomy was noted.
The limitation of this study is that it
is a retrospective case series without a
control group. However, it is a single-
surgeon series with a consistent
technique over a 20-year period. The
present study seems to be the only
series in which a patient-recorded out-
come score was used. In addition, no
patients were lost to follow-up.
Conclusion
In conclusion, hip pain and deterio-
ration can be caused by a variety of
deformities, acting either alone or in
combination. There can be consider-
able overlap with acetabular dyspla-
sia, hip impingement, and neck-shaft
abnormalities, while femoral version
may be diminished, normal, or exces-
sive. It is important to identify all the
deformities present to ensure the best
chance of success after hip preserva-
tion surgery. Version abnormalities of
the hip, often overlooked, must be
assessed because of the damage
caused by these deformities. A closed,
subtrochanteric derotation osteotomy
of the femur is a safe and effective
procedure to treat either femoral ret-
roversion or excessive anteversion.
Excellent or good results were obtained
in 93%, with a statistically significant
improvement in the mHHS, despite the
need for subsequent implant removal in
more than two-thirds of the patients.
Failures may occur in patients with
genetic defects associated with abnor-
mal collagen or bone density.
References
References printed in bold type are
those published within the past 5
years.
Month 2018, Vol 00, No 00
1. Ross JR, Larson CM, Adeoye O, Kelly BT,
Bedi A: Residual deformity is the most
common reason for revision hip
arthroscopy: A three-dimensional CT
study. Clin Orthop Relat Res 2015;473:
1388-1395.
2. Fabricant PD, Fields KG, Taylor SA,
Magennis E, Bedi A, Kelly BT: The effect of
femoral and acetabular version on clinical
outcomes after arthroscopic
femoroacetabular impingement surgery. J
Bone Joint Surg Am 2015;97:537-543.
3. Clohisy JC, Nepple JJ, Larson CM, Zaltz I,
Millis M: Persistent structural disease is the
most common cause of repeat hip
preservation surgery. Clin Orthop Relat
Res 2013;471:3788-3794.
4. Kohno Y, Nakashima Y, Akiyama M, Fujii
M, Iwamoto Y: Does native combined
anteversion influence pain onset in patients
with dysplastic hips? Clin Orthop Relat Res
2015;473:3716-3722.
5. Tibor LM, Liebert G, Sutter R,
Impellizzeri FM, Leunig M: Two or more
impingement and/or instability deformities
are often present in patients with hip pain.
Clin Orthop Relat Res 2013;471:
3762-3773.
6. Thawrani DP, Feldman DS, Sala DA: Not
all hip dysplasias are the same: Preoperative
CT version study and the need for reverse
bernese periacetabular osteotomy. J Pediatr
Orthop 2017;37:47-52.
7. Bedi A, Dolan M, Magennis E, Lipman J,
Buly R, Kelly BT: Computer-assisted
modeling of osseous impingement and
resection in femoroacetabular
impingement. Arthroscopy 2012;28:
204-210.
8. Fabricant PD, Bedi A, De La Torre K, Kelly
BT: Clinical outcomes after arthroscopic
psoas lengthening: The effect of femoral
version. Arthroscopy 2012;28:965-971.
9. Siebenrock KA, Steppacher SD, Haefeli PC,
Schwab JM, Tannast M: Valgus hip with
high antetorsion causes pain through
posterior extraarticular FAI. Clin Orthop
Relat Res 2013;471:3774-3780.
10. Putz C, Wolf SI, Geisbusch A, Niklasch M,
Doderlein L, Dreher T: Femoral derotation
osteotomy in adults with cerebral palsy.
Gait Posture 2016;49:290-296.
11. Dolan MM, Heyworth BE, Bedi A, Duke G,
Kelly BT: CT reveals a high incidence of
osseous abnormalities in hips with labral
tears. Clin Orthop Relat Res 2011;469:
831-838.
12. Stahl JP, Alt V, Kraus R, Hoerbelt R,
Itoman M, Schnettler R: Derotation of post-
traumatic femoral deformities by closed
intramedullary sawing. Injury 2006;37:
145-151.
13. Chapman ME, Duwelius PJ, Bray TJ,
Gordon JE: Closed intramedullary femoral
osteotomy: Shortening and derotation
Robert L. Buly, MD, MS, et al
procedures. Clin Orthop Relat Res 1993:
245-251.
14. Winquist RA: Closed intramedullary
osteotomies of the femur. Clin Orthop
Relat Res.1986:155-164.
15. Tonnis D, Heinecke A: Acetabular and
femoral anteversion: Relationship with
osteoarthritis of the hip. J Bone Joint Surg
Am 1999;81:1747-1770.
16. Eckhoff DG: Effect of limb malrotation on
malalignment and osteoarthritis. Orthop
Clin North Am 1994;25:405-414.
17. Terjesen T, Benum P, Anda S, Svenningsen
S: Increased femoral anteversion and
osteoarthritis of the hip joint. Acta Orthop
Scand 1982;53:571-575.
18. Scheys L, Spaepen A, Suetens P, Jonkers I:
Calculated moment-arm and muscle-
tendon lengths during gait differ
substantially using MR based versus
rescaled generic lower-limb musculoskeletal
models. Gait Posture 2008;28:640-648.
19. Gomez-Hoyos J, Schroder R, Reddy M,
Palmer IJ, Martin HD: Femoral neck
anteversion and lesser trochanteric
retroversion in patients with ischiofemoral
impingement: A case-control magnetic
resonance imaging study. Arthroscopy
2016;32:13-18.
20. Eckhoff DG, Montgomery WK, Kilcoyne
RF, Stamm ER: Femoral morphometry and
anterior knee pain. Clin Orthop Relat Res
1994:64-68.
21. Bruce WD, Stevens PM: Surgical correction
of miserable malalignment syndrome. J
Pediatr Orthop 2004;24:392-396.
22. MacWilliams BA, McMulkin ML, Davis
RB, Westberry DE, Baird GO, Stevens PM:
Biomechanical changes associated with
femoral derotational osteotomy. Gait
Posture 2016;49:202-206.
23. Moya LE, Buly RL, Henn RF, Kelly BT, Ma
Y, Molisani D: Femoral retroversion in
patients with femoroacetabular
impingement: A cofactor in the
development of hip osteoarthritis. J Bone
Joint Surg Br 2010;92(suppl IV):526.
24. Gelberman RH, Cohen MS, Shaw BA,
Kasser JR, Griffin PP, Wilkinson RH: The
association of femoral retroversion with
slipped capital femoral epiphysis. J Bone
Joint Surg Am 1986;68:1000-1007.
25. Steppacher SD, Albers CE, Siebenrock KA,
Tannast M, Ganz R: Femoroacetabular
impingement predisposes to traumatic
posterior hip dislocation. Clin Orthop
Relat Res 2013;471:1937-1943.
26. Canham CD, Yen YM, Giordano BD: Does
femoroacetabular impingement cause hip
Instability? A systematic review.
Arthroscopy 2016;32:203-208.
27. Buly R: Femoral deformities and hip
osteotomies, in Anil Ranawat M, Bryan
Kelly M, eds: Musculoskeletal
9
Femoral Derotation Osteotomy in Adults
Examination of the Hip and Knee.
Thorofare, NJ, SLACK Incorporated,
2010, pp 252–273.
28. Murphy SB, Simon SR, Kijewski PK,
Wilkinson RH, Griscom NT: Femoral
anteversion. J Bone Joint Surg Am 1987;69:
1169-1176.
29. Ganz R, Parvizi J, Beck M, Leunig M,
Notzli H, Siebenrock KA:
Femoroacetabular impingement: A cause
for osteoarthritis of the hip. Clin Orthop
Relat Res 2003:112-120.
30. Jamali AA, Mladenov K, Meyer DC, et al:
Anteroposterior pelvic radiographs to
assess acetabular retroversion: High
validity of the “cross-over-sign”. J Orthop
Res 2007;25:758-765.
31. Steppacher SD, Zurmuhle CA, Puls M,
et al: Periacetabular osteotomy restores
the typically excessive range of motion in
dysplastic hips with a spherical head.
Clin Orthop Relat Res 2015;473:
1404-1416.
32. Kraeutler MJ, Garabekyan T, Pascual-
Garrido C, Mei-Dan O: Hip instability: A
10
review of hip dysplasia and other
contributing factors. Muscles Ligaments
Tendons J 2016;6:343-353.
33. Ejnisman L, Philippon MJ, Lertwanich P,
et al: Relationship between femoral
anteversion and findings in hips with
femoroacetabular impingement.
Orthopedics 2013;36:e293-e300.
34. Gomez-Hoyos J, Schroder R, Reddy M,
Palmer IJ, Khoury A, Martin HD: Is
there a relationship between psoas
impingement and increased trochanteric
retroversion? J Hip Preserv Surg 2015;2:
164-169.
35. Kain MS, Novais EN, Vallim C, Millis MB,
Kim YJ: Periacetabular osteotomy after
failed hip arthroscopy for labral tears in
patients with acetabular dysplasia. J Bone
Joint Surg Am 2011;93(suppl 2):57-61.
36. Reynolds D, Lucas J, Klaue K: Retroversion
of the acetabulum: A cause of hip pain. J
Bone Joint Surg Br 1999;81:281-288.
37. Maruyama M, Feinberg JR, Capello WN,
D’Antonio JA: The Frank Stinchfield
Award: Morphologic features of the
Journal of the American Academy of Orthopaedic Surgeons
acetabulum and femur: Anteversion angle
and implant positioning. Clin Orthop Relat
Res 2001;52-65.
38. Hartel MJ, Petersik A, Schmidt A, et al:
Determination of femoral neck angle and
torsion angle utilizing a novel three-
dimensional modeling and analytical
Technology based on CT datasets. PLoS
One 2016;11:e0149480.
39. Sugano N, Noble PC, Kamaric E: A
comparison of alternative methods of
measuring femoral anteversion. J Comput
Assist Tomogr 1998;22:610-614.
40. Kamath AF, Ganz R, Zhang H, Grappiolo
G, Leunig M: Subtrochanteric osteotomy
for femoral mal-torsion through a surgical
dislocation approach. J Hip Preserv Surg
2015;2:65-79.
41. Pailhe R, Bedes L, Sales de Gauzy J, Tran R,
Cavaignac E, Accadbled F: Derotational
femoral osteotomy technique with locking
nail fixation for adolescent femoral
antetorsion: Surgical technique and
preliminary study. J Pediatr Orthop B
2014;23:523-528.
 ORTHOPAEDIC VIDEO THEATER ABSTRACTS

Stay Out of Trouble in Reverse Shoulder Arthroplasty - Award Winner

Enrico Bellato, MD; Lorenzo Mattei, MD; Michel J. Calò, MD; Davide Blonna, MD; Andrea Arpaia, MD;
Giovanni Ferrero, MD; Filippo Castoldi, MD
DOI: 10.5435/JAAOS-D-18-00492
Abstract: This video discusses the key steps a surgeon should follow to perform safe and successful
primary reverse shoulder arthroplasty. The first step is to select the appropriate candidate for reverse
shoulder arthroplasty and to ensure adequate preoperative preparation. CT is necessary to evaluate
glenoid version, identify possible glenoid bone defects, and determine the eventual need for patient-
specific instrumentation. Proper patient positioning is essential to avoid intraoperative patient shifting
and neck stretching and to allow free range of shoulder motion. The deltopectoral or anterolateral approach can be used,
and the surgeon should understand the advantages and disadvantages of both approaches. Retractors should be placed to
obtain optimal joint visualization with caution exercised to avoid neurovascular structures. During preparation of the
humeral head, the surgeon must be mindful of possible proximal migration of the humerus, the subjective variability of
humeral version, and the patient’s expectations for range of motion. Glenoid exposure is critical and must be performed
meticulously. The use of appropriate instruments, release of the long head of the triceps, release of the posterior capsule
and eventual revision of the humeral osteotomy allow adequate visualization of the glenoid surface and neck. Drawing the
Downloaded from http://journals.lww.com/jaaos by BhDMf5ePHKbH4TTImqenVCDV3BItTVWTdpjRZIgqbDP7ddUWBH19AZznP8Nb7E6d2wqdwyywKPo= on 10/02/2018

anatomic axes of the glenoid aids in placement of the guidewire in the glenoid, which should be inserted in the center of
the neck slightly inferiorly and with some inferior tilt. To achieve optimal baseplate fixation, the screws should be directed
toward the three columns of the scapula. The use of baseplates with variable-angle locking screws should be considered
to freely orientate the screws. The suprascapular nerve must be avoided during posterior screw insertion. Finally, implant
stability is assessed by evaluating shoulder range of motion and tension of the deltoid and conjoint tendons. Watch the
video trailer: http://links.lww.com/JAAOS/A158.

Anterior Cervical Diskectomy and Fusion for Myeloradiculopathy - Honorable

Mention
Vincenzo Denaro, MD; Gianluca Vadala, MD; Alberto C. Di Martino, MD, PhD;
Lorenzo Alirio Diaz Balzani, MD; Luca Denaro, MD
DOI: 10.5435/JAAOS-D-18-00493
Abstract: Cervical disk herniation can result in focal compression of the spinal cord, giving rise to an
ischemic lesion leading to myelopathy. Moreover, it can compress a nerve root, causing brachialgia
and deficits of the upper limbs. Cervical radiculopathy and myelopathy are common in adults and their
consequences may be as serious as spastic tetraplegia or nerve root paralysis. Anterior cervical
diskectomy and fusion is one of the most commonly performed procedures, and it usually is
associated with a good clinical outcome. It is mainly used if radiculopathy and/or myelopathy are the result of anterior
compression and if one or two disk levels are involved. This video shows anterior cervical diskectomy and fusion in a
45-year-old man with C5-C6 disk herniation and myeloradiculopathy who had brachialgia, pectoral fasciculation, and
weakness of the lower limbs associated with impaired ambulation. The surgical steps in this video include the anterolateral
approach to the midcervical spine, diskectomy and decompression as described by Smith and Robinson, interbody fusion
with the use of autologous bone graft harvested from the iliac crest, and stabilization with a plate and screws. Graphic
illustrations are included in the video to better explain the technique. Anterior cervical diskectomy and fusion is a reliable
procedure, leading to notable clinical and electrophysiologic improvement. This video may help surgeons at any stage of
their career understand how to appropriately approach and manage cervical myeloradiculopathy. Potential pitfalls
associated with this technique are reviewed and discussed. Watch the video trailer: http://links.lww.com/JAAOS/A159.

© 2018 The Author(s). Published by Wolters Kluwer Health, Inc., on behalf of the American Academy of Orthopaedic Surgeons.
Video trailers that accompany these abstracts are available on www.jaaos.org.

706 October 1, 2018, Vol 26, No 19

Copyright ª the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.