The Spine Journal 3 (2003) 460–465

Correlation of clinical examination characteristics with three

sources of chronic low back pain
Sharon Young, PT, Cert. MDTa,*, Charles Aprill, MDb,
Mark Laslett, PT, Dip. MT, Dip. MDTc
a
Mobile Spine and Rehabilitation Center, 6051 Airport Blvd. Suite A-1, Mobile, AL 36608, USA
b
Magnolia Diagnostics, New Orleans, LA, USA
c
Linköpings Universitet, Linköpings, Sweden
Received 6 November 2002; accepted 27 May 2003

Abstract BACKGROUND CONTEXT: Research has demonstrated some progress in using a clinical exami-
nation to predict discogenic or sacroiliac (SI) joint sources of pain. No clear predictors of symptomatic
lumbar zygapophysial joints have yet been demonstrated.
PURPOSE: To identify significant components of a clinical examination that are associated with
symptomatic lumbar discs, zygapophysial joints and SI joints.
STUDY DESIGN: A prospective, criterion-related concurrent validity study performed at a private
radiology practice specializing in spinal diagnostics.
PATIENT SAMPLE: The sample consisted of 81 patients with chronic lumbopelvic pain referred
for diagnostic injections.
OUTCOME MEASURES: Contingency tables were constructed for nine features of the clinical
evaluation compared with the results of diagnostic injections. Statistical analysis included chi-
squared test for independence, phi and odds ratios with confidence intervals.
METHOD: Patients received blinded clinical examinations by physical therapists, and diagnostic
injections were used as the criterion standard.
RESULTS: Significant relationships were found between discogenic pain and centralization of pain
during repeated movement testing, and pain when rising from sitting. Lumbar zygapophysial joint
pain was associated with absence of pain when rising from sitting. Sacroiliac joint pain was related
to three or more positive pain provocation tests, pain when rising from sitting, unilateral pain
and absence of lumbar pain.
CONCLUSIONS: Significant correlations exist between clinical examination findings and symp-
tomatic lumbar discs, zygapophysial and SI joints. The strongest relationships were seen between
SI joint pain and three or more positive pain provocation tests, centralization of pain for symptomatic
discs and absence of pain when rising from sitting for symptomatic lumbar zygapophysial joints. 쑖
2003 Elsevier Inc. All rights reserved.
Keywords: Lumbar zygapophysial joint; Sacroiliac joint; Intervertebral disk; Physical examination; Low back pain; Intra-
articular injections; Spinal injections; Diagnosis

Introduction examination [3–11], clinicians are often at a loss to provide

Patients with chronic low back pain desire answers about
why they have pain [1,2]. Given the limitations of clinical
FDA device/drug status: not applicable.
Nothing of value received from a commercial entity related to this research.
* Corresponding author. Mobile Spine and Rehabilitation Center, 6051
Airport Boulevard Suite A-1, Mobile, AL 36608, USA. Tel.: (251) 460-
0201; fax: (251) 460-2848.
E-mail address: sbyoung@bellsouth.net (S. Young)
1529-9430/03/$ – see front matter 쑖 2003 Elsevier Inc. All rights reserved.
doi:10.1016/S1529-9430(03)00151-7
answers with any degree of certainty. Imaging studies demon-
strating neural compromise are likely to be clinically signifi-
cant [12,13], but plain radiographs, computed tomography
and magnetic resonance imaging are of limited value in identi-
fying painful lumbar structures because they are often abnor-
mal in asymptomatic subjects and may be unremarkable in
patients with significant symptoms [12–17].
Progress has been made with clinical identification of
subgroups within low back pain patient populations. Specific
features of the clinical examination have been found to have
 S. Young et al. / The Spine Journal 3 (2003) 460–465
high sensitivity and specificity for nerve root pain associated
with disc extrusion [18–20], and for spinal stenosis [21].
The phenomenon of pain moving centrally toward the lumbar
spine or peripherally from the lumbar spine in response to
repeated lumbar movements has been associated with dis-
cogenic pain [22,23]. This centralization phenomenon has
high sensitivity (94%) but low specificity (52%) in patients
with chronic low back pain [24]. Young et al. [25] used specific
pain provocation tests and found them to be useful in pre-
dicting symptomatic sacroiliac (SI) joints (sensitivity 83%,
specificity 75%). Identification of patients with symptomatic
lumbar zygapophysial joints by clinical examination remains
problematic [3,26–29]. The composite of signs and symp-
toms reported to be associated with zygapophysial joint pain
[30] was not confirmed by a recent study [8].
Disc, zygapophysial and SI joint pain can be established
by specific spinal injection techniques. The criteria for a
positive discogram include presence of abnormal morphol-
ogy and provocation of similar or exact pain during injection.
At least one adjacent disc must be negative to injection as
a control [31]. Substantial symptom relief after delivery
of an anesthetic into lumbar zygapophysial or SI joints is
the criterion for identifying symptomatic joints [32,33].
The purpose of this study was to assess the significance
of components of the history and clinical examination of
patients with chronic pain associated with symptomatic
lumbar discs, lumbar zygapophysial joints or SI joints identi-
fied by diagnostic injection. The null hypothesis was that
there is no correlation between features of the clinical exami-
nation and the results of diagnostic injection.
Patients and methods
A total of 102 patients with chronic lumbar or lumbopel-
vic pain were referred to a private radiology practice spe-
cializing in spinal diagnostics. They were evaluated by
visiting physical therapists. Each patient had previously un-
dergone imaging studies and had prior unsuccessful thera-
peutic interventions. Fifty-five patients were part of a study
examining the diagnostic power of clinical evaluation in
predicting the presence or absence of symptomatic SI joints.
Fifty-seven additional patients were evaluated to explore
possible means of differentiating between painful SI and
zygapophysial joints, and intervertebral discs. The same clin-
ical evaluation was used throughout.
The clinical evaluation was carried out by physical thera-
pists with credentials in the McKenzie method of spinal me-
chanical diagnosis and therapy. Two training sessions were
held before the study to ensure consistency of methods and
standardization of test procedures. Patients were scheduled for
the clinical evaluation in an opportunistic fashion on days
when one of the physical therapists would be in the clinic. The
therapists were blinded to previous radiology investigations,
and the radiologist was blinded to results of the physical thera-
pist clinical evaluations. Informed consent was obtained, clin-
ical evaluations were conducted and conclusions recorded
461
before injections. Patients were not considered for the clinical
evaluation if they were unwilling to participate, had clear
signs of nerve root compression, were deemed too frail or were
unable to tolerate the full clinical evaluation.
The evaluation included patient history, determination of
symptom location, aggravating and relieving factors and
performance of repeated end-range movements and positions
to assess changes in symptom behavior [34]. Centralization,
peripheralization or other change in pain status was specifi-
cally noted, as was any obvious change in spinal mobility with
repeated movement testing. The term “centralization” refers
to a retreat of referred symptoms from the periphery toward the
midline of the spine. “Peripheralization” refers to progression
of symptoms moving from the spine toward the periphery.
The SI joint was examined using pain provocation tests,
including distraction, compression, sacral thrust, thigh thrust
and Gaenslen’s test (pelvic torsion), as described by Laslett
and Williams [35].
Injection procedures
All patients were referred for evaluation employing one
or more injection procedures. Lumbar discography, lumbar
zygapophysial joint arthrography/injection or SI joint
arthrography/injection were performed when specifically re-
quested by the referring physician or if deemed indicated
by the radiologist. Considerations included patient consent
and ability to tolerate the procedures. Discography was per-
formed employing a standard technique [31]. Using an extra-
pedicular approach, small-gauge needles were directed into
the nucleus of the discs. Contrast material was then slowly
injected in a controlled fashion. Spot films in two planes
were obtained to demonstrate the enhanced nucleus and any
annular or end plate abnormalities. Discometric (pressure/
volume) data were also collected during the injection, as
was the patient’s response to disc stimulation. Local anesthe-
tic was instilled into painful discs, and anesthetic response
was recorded before discharge. The injection was deemed
positive when it provoked similar or exact pain, when im-
aging demonstrated abnormal morphology and at least one
adjacent disc was negative upon injection.
SI and lumbar zygapophysial joint injections were per-
formed employing standard techniques [4,33]. Small-gauge
spinal needles were directed into the joint space with fluoro-
scopic guidance. Contrast material was slowly instilled
to confirm needle placement in the joint and to assess capsu-
lar integrity. The SI and/or lumbar zygapophysial joint injec-
tion was considered positive if the slow injection of solution
provoked familiar pain of moderate intensity and instilla-
tion of a small volume of anesthetic (less than 1.5 cc) resulted
in 80% or greater relief of the primary pain based on 10-point
verbal analogue scale. Pain ratings were recorded before
and 30 to 60 minutes after the injection. Injections were
considered indeterminate when there was a concordant pain
response but insufficient pain relief. Such indeterminate re-
sponses were considered negative for analysis. Where cir-
cumstances dictated, the study of several components was
462 S. Young et al. / The Spine Journal 3 (2003) 460–465
accomplished on the same day. SI and zygapophysial joint
injections were completed first. Discography was performed
in a second session after evaluating response to the initial pro-
cedures. The radiologist documented the procedures per-
formed, radiographic findings and pain and analgesic
response to injection.
Statistical analysis
Post hoc review of the clinical evaluation of raw data
suggested that pain stemming from the three sources differed
in regard to pain location and response to mechanical
stresses. Nine specific characteristics of the clinical assess-
ment were selected for statistical analysis. These were mid-
line lumbar pain, asymmetrical pain, unilateral pain, bilateral
pain; change in lumbar range of motion, centralization or
peripheralization with repeated movement testing; pain
when rising from sitting and presence of three or more
positive SI joint pain provocation tests.
Data were analyzed using SPSS statistical software [36].
Two-by two contingency tables were constructed, and the
chi-squared test of independence was used as a screening
test to identify significant associations among the characteris-
tics as compared with the results of diagnostic injections
(α ⫽ .05). The phi coefficient, which is an index of associa-
tion analogous to Pearson’s correlation coefficient (r), was
also obtained [37]. The phi coefficient is used to demonstrate a
correlation between two dichotomous variables, and ranges
from values of ⫺1 to +1; ⫺1 indicates a perfect inverse
relationship, 0 relates to no association, and +1 indicates a
perfect positive correlation between the two variables. Phi
is calculated by dividing the square root of chi-squared by
the number of subjects in the sample. Phi-squared indicates the
strength of the effect size. The effect size is an estimate used
to quantify the degree to which the results deviate from
null expectations.
For those characteristics showing a significant correla-
tion, odds ratios with 95% confidence intervals were ob-
tained to assess the strength of the correlation between the
variables. An odds ratio of one indicates no relationship
between the diagnostic injection and clinical examination
characteristic; two indicates a weak relationship; over four
demonstrates a strong relationship [38]. Sensitivity, specific-
ity and confidence intervals for the relationship of centraliza-
tion to each of the three pain generators were calculated
using the Wilson method [39].
Results
Of the 102 cases referred for diagnostic evaluation, 21
were excluded for a variety of reasons: 9 patients had epi-
dural injections only, 6 patients were unable to tolerate the
clinical evaluation, there was a technical failure to inject a
joint in 2 cases, 1 patient had no injection and 3 patients
were pain free at the time of the examination so that no
conclusion could be reached by clinical evaluation.
The study group comprised 81 individuals (49 women
and 32 men) ranging in age from 18 to 79 years (mean, 40.8
years; SD, 12.1 years). Time from onset of symptoms to
evaluation ranged from 2 to 156 months (median, 18 months;
mean, 26 months; SD, 28 months). Forty-three percent were
off work because of their symptoms. Ninety-eight percent of
the study group included buttock pain on their preprocedure
pain drawing. Table 1 shows the distribution of the dominant
pain location for the 81 patients. Fifty-four percent reported
buttock pain as the dominant pain.
A total of 104 injection procedures were performed on the
81 subjects. Thirty patients had only negative responses to
one or more injection procedures. A pain generator was posi-
tively identified in 51 patients. The number of patients and
results of the diagnostic injections are shown in Table 2.
There was a significant association between a positive SI
joint injection and four of the nine characteristics (Table
3). A negative relationship was found between presence of
midline lumbar pain and positive SI joint injection. Patients
having SI joint pain rarely had pain at or above the level
of the L5 spinous process. In contrast, 80% of those with
discogenic pain reported midline lumbar pain. Positive rela-
tionships were noted for unilateral pain (p⫽.05), pain pro-
duced or increased when rising from sitting (p⫽.02), and
presence of three or more positive SI joint pain provocation
tests (p⬍.001). As indicated by the odds ratios, there was a
fairly strong relationship between positive SI joint injection
and unilateral pain and a very strong relationship with three or
more pain provocation tests. Presence of three or more positive
SI joint pain provocation tests was found to have a phi coeffi-
cient of .6 when compared with the results of SI joint diagnos-
tic injection. Squaring this value gives an effect size of 36%.
All patients with positive SI joint injections had pain when
rising from sitting, precluding the calculation of an odds ratio,
because there were no patients with a positive SI joint injec-
tion who did not have pain when rising from sitting.
Both centralization of pain and pain when rising from
sitting were significantly associated with a positive disco-
gram (Table 4). All patients with positive discograms re-
ported pain when rising from sitting, precluding calculation
of an odds ratio. Seven of 15 patients with positive disco-
grams (47%) reported centralization of pain during the
repeated movement portion of the clinical examination.
The odds of centralization occurring for patients having
Table 1
Dominant pain distribution
Dominant pain Percentage of patients
Lumbar 27
Sacrum 13
Buttock 54
Groin 4
Lower extremity 2
 S. Young et al. / The Spine Journal 3 (2003) 460–465 463

Table 2 Table 4
Diagnostic injection data Significant clinical examination findings for lumbar discogenic pain
Discogram Lumbar facet Sacroiliac Chi squared
Positive injection 15 14 22 (1, N⫽24) p Value Phi
Negative injection 9 9 35 Centralization 5.9 .025 0.5
Totals 24 23 57 Pain rising from sitting 5.7 .017 0.5

positive discograms were 2.13 (95% CI 1.28, 3.52). This
represents a weak but positive relationship between the two
variables. Patients with negative discograms did not report
centralization of pain during repeated movement testing.
To further assess the value of centralization, the 2×2 con-
tingency tables for each of the three pain generators (Tables
5, 6 and 7) were evaluated to establish the sensitivity and
specificity of this clinical sign. When compared with the
results of discography, centralization was found to have a
low sensitivity of 0.47 (95% CI 0.25, 0.70), but a high
specificity of 1.00 (95% CI 0.7, 1.00). All patients whose
pain centralized had positive discography. Comparison of
centralization to the results of diagnostic zygapophysial joint
injection yielded a sensitivity of 0.0 (95% CI 0.0, 0.22) and
specificity of 0.89 (95% CI 0.57, 0.98). None of the patients
with positive zygapophysial joint injections reported central-
ization of pain. Comparison of centralization with SI joint
injections gave a sensitivity of 0.09 (95% CI 0.03, 0.28) and
specificity of 0.79 (95% CI 0.63, 0.90).
Given that all patients with a positive injection into either
the SI joint or a lumbar disc reported provocation of pain
when rising from sitting, it is interesting to note that the
lack of pain provocation when rising from sitting was
the one significant characteristic associated with patients with
positive zygapophysial injections (Table 8). There was a
strong correlation between not having pain provoked when
rising from sitting for patients with positive zygapophysial
injections. This characteristic had an effect size of 36%.
Discussion
The results of this study allow rejection of the null hypoth-
esis that there would be no significant associations between
the clinical evaluation and diagnostic injections. Significant
Table 3
Significant clinical examination findings for sacroiliac joint pain
Chi squared Odds ratio
(1, N⫽56) p Value Phi (95% CI)
Midline pain 3.9 .05 ⫺0.3 0.3 (.01, 1.0)
Pain rising 5.3 .02 0.3
from sitting
ⱖ3 SIJ PPT/s 22.2 ⬍.001 0.6 27.9 (5.4, 143.4)
Unilateral pain 3.7 .05 0.3 3.4 (0.9, 12.1)
CI⫽confidence interval; PPTs⫽pain prorocation tests; SIJ⫽sacroiliac
joint.
associations were seen for each of the three pain generators
investigated by diagnostic injection.
With regards to pain location, 80% of those with painful
discs had midline lumbar pain. The presence of midline
lumbar pain tends to exclude the SI joint as a potential
pain generator. While this study found no correlation between
midline lumbar pain and zygapophysial joints, a previous
study found that patients with confirmed zygapophysial joint
pain did not have midline pain [3]. The small number of
symptomatic zygapophysial joints in the sample provides
insufficient statistical power to support or reject the pre-
viously cited negative relationship between midline pain and
these joints.
The one characteristic that was significantly associated
with lumbar zygapophysial joint pain was lack of provoca-
tion of pain when rising from sitting. This contrasts to sig-
nificant correlations between pain provoked on rising from
sitting for patients who had positive disc or SI joint injec-
tions. Patients reporting midline lumbar pain and pain when
rising from sitting are more likely to have discogenic pain.
While only a weak correlation was observed between
centralization and discogenic pain, it is of note that patients
with zygapophysial joint pain did not report the centraliza-
tion phenomenon. When centralization is observed, a painful
facet joint is less likely to be the pain generator.
This study supports the findings of Donelson et al. [23]
regarding the value of centralization in identifying symptom-
atic discs. The sensitivity and specificity of centralization can
be calculated from their data to be 0.94 (95% CI 0.82, 0.99)
and 0.52 (95% CI 0.34, 0.69), respectively. Nearly half of
the centralizers had negative discograms. In contrast to the
Donelson et al. study, all patients in the present study receiving
discography and reporting centralization had positive disco-
grams. The two patients reporting centralization who did not
undergo discography did have positive SI joint injections. Be-
cause these two patients did not receive discography, it is un-
known if the discs of these patients contributed to their
complaints of pain.
In the present study, patients with unilateral pain below
the level of the L5 spinous process, no lumbar pain and pain
Table 5
Contingency table comparing centralization of pain to the results
of discography.
Positive discogram Negative discogram
Centralizers 7 0
Noncentralizers 8 9
464 S. Young et al. / The Spine Journal 3 (2003) 460–465
Table 6
Contingency table comparing centralization of pain with the
results of diagnostic zygapophysial joint (z-joint) injections
Positive z-joint Negative z-joint
Noncentralizers 14 8 Pain rising
Centralizers 0 1
when rising from sitting were more likely to have a painful
SI joint. The use of SI joint pain provocation tests aids
significantly in clarifying the clinical picture. Presence of a
SI joint source of pain as identified by diagnostic injection
is 28 times more likely when there are three or more positive
SI joint pain provocation tests.
A limitation of this study is that the comparisons are
based on single diagnostic injections. This is not an issue for
those receiving discography, because internal controls are
used to minimize false-positive responses. However, there is
a 32% false-positive rate associated with single lumbar zyga-
pophysial injections [40]. A false-positive rate for SI joint
injections has not been reported. A false-positive rate of
7.7% may be calculated from one study [4] or 47% in another
[6]. Replication of this study using double (confirmatory)
diagnostic blocks is warranted to better clarify the degree
of significance of these clinical characteristics.
Given the limitations of a routine orthopedic examination
and radiographic findings in identifying sources of back
pain, this paper provides preliminary evidence that pain
location, historical factors and examination findings facili-
tate better clinical classification of patients with low back
pain. Individuals with symptomatic lumbar discs, zygapo-
physial joints or SI joins constitute at least 68% of all patients
with chronic low back pain [41]. The use of readily available
clinical findings to identify possible SI joint, facet or disco-
genic sources of pain may enable more precise use of diag-
nostic procedures and therapeutic interventions.
Conclusion
Significant associations were noted between SI joint pain
provocation tests, pain when rising from sitting, unilateral
pain and lack of midline lumbar pain and SI joint pain.
Centralization of pain and pain when rising from sitting
were correlated with discogenic pain. No provocation of
pain when rising from sitting was associated with zygapo-
physial joint pain. Consideration of specific pain location,
Table 7
Contingency table comparing centralization of pain with the
results of diagnostic sacroiliac joint injections
Positive SI joint Negative SI joint
Noncentralizer 20 27
Centralizer 2 7
SI⫽sacroiliac.
Table 8
Significant clinical examination finding for lumbar facet pain
Chi squared Odds ratio
(1, N⫽23) p Value Phi (95% CI)
7.1 .008 ⫺0.6 0.08 (0.01, 0.59)
from sitting
CI⫽confidence interval.
historical factors and examination findings can allow for
better clinical classification of patients with low back pain.
References
[1] McPhillips-Tangum CA, Cherkin DC, Rhodes LA, Markham C. Rea-
sons for repeated medical visits among patients with chronic back
pain. J Gen Intern Med 1998;13:289–95.
[2] Cherkin DC. Primary care research on low back pain. The state of
the science. Spine 1998;23:1997–2002.
[3] Schwarzer AC, Aprill C, Derby R, Fortin JD, Kine G, Bogduk N.
Clinical features of patients with pain stemming from the lumbar
zygapophysial joints. Is the lumbar facet syndrome a clinical entity?
Spine 1994;15:1132–7.
[4] Schwarzer AC, Aprill C, Bogduk N. The sacroiliac joint in chronic
low back pain. Spine 1995;20:31–7.
[5] Dreyfuss PH, Michaelsen M, Pauza K, McLarty J, Bogduk N. The
value of history and physical examination in diagnosing sacroiliac
joint pain. Spine 1996;2594–602.
[6] Maigne J-Y, Aivaliklis A, Pfefer F. Results of sacroiliac joint double
block and value of sacroiliac pain provocation tests in 54 patients
with low back pain. Spine 1996;21:1889–92.
[7] Slipman CW, Sterenfeld EB, Chou LH, et al. The predictive value
of provocative sacroiliac joint stress maneuvers in the diagnosis of
sacroiliac joint syndrome. Arch Phys Med Rehabil 1998;79:288–92.
[8] Manchikanti L, Pampati V, Fellows B, Baha AG. The inability of the
clinical picture to characterize pain from the facet joints. Pain Physi-
cian 2000;3:158–66.
[9] Spitzer WO. Scientific approach to the assessment and management
of activity-related spinal disorders. A monograph for clinicians.
Report of the Quebec Task Force on Spinal Disorders. Spine 1987;
12(suppl 7):S16–9.
[10] Waddell G. The back pain revolution. Edinburgh: Churchill Living-
stone, 1998.
[11] Nachemson A, Vingard E. Assessment of patients with neck and back
pain: a best evidence synthesis. In: Nachemson AL, Jonsson E, editors.
Neck and back pain—the scientific evidence of causes, diagnosis, and
treatment, 3rd ed. Philadelphia: Lippincott Williams and Wilkins,
2000:189–235.
[12] Boos N, Reider R. The diagnostic accuracy of magnetic resonance
imaging, work perception and psychological factors in identifying
symptomatic disc herniations. Spine 1995;20:2613–25.
[13] Jarvik JG, Hollingworth W, Heagerty P. The longitudinal assessment
of imaging and disability of the back (LAIDBack) study. Spine
2001;26:1158–66.
[14] Twomey L, Taylor JR, Taylor MM. Unsuspected damage to lumbar
zygapophyseal (facet) joints after motor vehicle accidents. Med J
Aust 1989;151:210–7.
[15] Fardon DF, Milette PC. Nomenclature and classification of lumbar
disc pathology. Spine 2001;26:E93–113.
[16] Schwarzer AC, Wang SC, O’Driscoll D. The ability of computerized
tomography to identify a painful zygapophysial joint in patients with
chronic low back pain. Spine 1995;20:907–12.
 S. Young et al. / The Spine Journal 3 (2003) 460–465
[17] Elgafy H, Semaan HB, Ebraheim NA, Coombs RJ. Computed tomog-
raphy findings in patient with sacroiliac pain. Clin Orthop 2001;382:
112–118.
[18] Hakelius A. A prognosis in sciatica. Acta Orthop Scand 1970;129:
1–76.
[19] Kortelainen P, Puranen J, Koivisto E, Lahde S. Symptoms and signs
of sciatica and their relation to the localization of the lumbar disc
herniation. Spine 1985;10:88–92.
[20] Kosteljanetz M, Espersen JO, Halaburth H, Miletic T. Predictive value
of clinical and surgical findings in patients with lumbago-sciatica. A
prospective study. Part 1. Acta Neurochir 1984;73:67–76.
[21] Fritz JM, Delitto A, Welch WC, Erhard RE. Lumbar spinal stenosis:
a review of current concepts in evaluation, management, and outcome
measurements. Arch Phys Med Rehabil 1998;79:700–8.
[22] Donelson R, Murphy K, Silva G. The centralization phenomenon: its
usefulness in evaluating and treating sciatica. Spine 1990;15:211–3.
[23] Donelson R, Aprill C, Medcalf R, Grant W. A prospective study of
centralization of lumbar and referred pain: a predictor of symptomatic
discs and annular competence. Spine 1997;22:1115–22.
[24] Bogduk N. Commentary on a prospective study of centralization and
lumbar and referred pain: a predictor of symptomatic discs and anular
competence. Pain Medicine Journal Club Journal 1997;3:246–8.
[25] Young SB, Laslett M, Aprill CN, Kelly C. The sacroiliac joint: compar-
ing physical examination and diagnostic block arthrography. Journal
of Orthopedic and Sports Physical Therapy 2000;30:A34.
[26] Jackson RP, Jacobs RR, Montesano PX. 1988 Volvo award in clinical
sciences. Facet joint injection in low-back pain. A prospective statisti-
cal study. Spine 1988;13:966–71.
[27] Helbig T, Lee CK. The lumbar facet syndrome. Spine 1988;13:61–4.
[28] Revel M, Poiraudeau S, Auleley GR, et al. Capacity of the clinical
picture to characterize low back pain relieved by facet joint anesthesia.
Proposed criteria to identify patients with painful facet joints. Spine
1998;23:1972–7.
465
[29] McCall IW, Park WM, O’Brien JP. Induced pain referral from posterior
lumbar elements in normal subjects. Spine 1979;4:441–6.
[30] Revel M, Poiraudeau S, Auieley GR, et al. Capacity of the clinical
picture to characterize low back pain relieved by facet joint anesthesia.
Spine 1998;23:1972–7.
[31] Aprill CN. Diagnostic disc injection. In: Frymoyer JW, editor. The
adult spine: principles and practice. Philadelphia: Lippincott-Raven
Publishers, 1997:539–62.
[32] Merskey H, Bogbuk N. Classification of chronic pain: descriptions
of chronic pain syndromes and definitions of pain terms, 2nd ed.
Seattle: IASP Press, 1994;190.
[33] Bogduk N, Aprill CN, Derby R. Diagnostic blocks of spinal synovial
joints. In: White AH, Schofferman JA, editors. Spine care. St Louis,
MI: Mosby-Year Book, 1995:298–321.
[34] McKenzie RA. The lumbar spine: mechanical diagnosis and therapy.
Waikanae, New Zealand: Spinal Publications Ltd., 1981.
[35] Laslett M, Williams M. The reliability of selected pain provocation
tests for sacroiliac joint pain. Spine 1994;19:1243–9.
[36] SPSS Graduate Pack 10.0 for Windows [computer program]. Chi-
cago: SPSS, 1999.
[37] Huck SW. Reading statistics and research, 3rd ed. New York: Addition.
Wesly Longman, 2000;628–9.
[38] Garb JL. Understanding medical research. Boston: Little, Brown and
Company, 1996:104.
[39] Altman DG. Diagnostic tests. In: Altman DG, Machin D, Bryant TN,
Gardner MJ, editors. Statistics with confidence, 2nd ed. Bristol: British
Medical Journal, 2000:105–19.
[40] Schwarzer AC, Aprill CN, Derby R, et al. The false-positive rate of
uncontrolled diagnostic blocks of the lumbar zygapophysial joints.
Pain 1994;58:195–200.
[41] Bogduk N. The anatomical basis for spinal pain syndromes. J Manipul
Physiol Ther 1995;18:603–5.