|
|
 |
* Please Note: The following article is provided by Body Mechanics for the sole purpose of
educating and informing our current and future patients.
The Value of Medical History and Physical Examination in Diagnosing Sacroiliac Joint Pain
Paul Dreyfuss , MD *-; Mark Michaelsen, DC ±; Kevin Pauza, MD *; Jerry McLarty, PhD Nikolai Bogduk, MD, PhD II
From *Spine Specialists, Tyler, Texas, -Department of Rehabilitation Medicine, University of Texas Health Science Center, San Antonio, Texas, ±Michaelsen Chiropractic Center, Tyler, Texas, Department of Epidemiology/Biomathematics, University of Texas Health Center, Tyler, Texas, and iFaculty of Medicine and Health Sciences, University of Newcastle, Newcastle, Australia.
SPINE 1996;21:2594-2602
Study Design
This prospective study evaluated the diagnostic utility of historically accepted sacroiliac joint tests. A multidisciplinary
expert panel recommended 12 of the "best" sacroiliac joint tests to be evaluated against a criterion standard of unequivocal pain relief
after an intra-articular injection of local anesthetic into the sacroiliac joint.
Objectives
To identify a single sacroiliac joint test or ensemble of tests that are sufficiently useful in diagnosing sacroiliac
joint disorders to be clinically valuable.
Summary of Background Data
No previous research has been done to evaluate any physical test of sacroiliac joint pain against an
accepted criterion standard.
Methods
Historical data was obtained, and the 12 tests were performed by two examiners on 85 patients who subsequently underwent
sacroiliac joint blocks. Ninety percent or more relief was considered a positive response, and less then 90% relief was
considered a negative response.
Results
There were 45 positive and 40 negative responses. No historical feature, none of the 12 sacroiliac joint tests, and
no ensemble of these 12 tests demonstrated worthwhile diagnostic value.
Conclusion
Sacroiliac joint pain is resistant to identification by the historical and physical examination data from
tests evaluated in this study.
It is quite clear that the sacroiliac joint can be a source of pain. Studies of healthy volunteers have shown that
intra-articular injections of contrast medium produce path over the region of the joint, which occasionally radiates into
the buttock and thigh. Even broader areas of pain in symptomatic patients have been relieved by anesthetizing the joint
with intra-articular injections of local anesthetic.r Known causes of sacroiliac joint pain include spondyloarthropathy,
crystal and pyogenic arthropathy, fracture of the sacrum and pelvis, and diastasis resulting from pregnancy or
childbirth.
References
1. April CN. The role of anatomically specific injections into the sacroiliac joint. In: Vleeming A, Mooney V, Snijders C, Dorman T, eds. Proceedings: First Interdisciplinary- World Congress on Low Back Pain and its Relation to the Sacroiliac Joint. San Diego, CA: 1992:373-80.
2. Barnsley L, Lord S, Wallis S, Bogduk N. False positive rates of cervical zygapophysial joint blocks. Clin J Pain 1993;9:124-30.
Beal MC. The sacroiliac problem: Review of anatomy, mechanics, and diagnosis. J Am Osteopath Assoc 1982;81:667-79.
4. Bellamy N, Park W, Rooney PJ. What do we know about the sacroiliac joint? Semin Arthritis Rheum 1983;12:282313.
5. Bernard TNJr, Cassidy SD. The sacroiliac joint syndrome: Pathophysiology, diagnosis, and management. In: Frymoyer JW, ed. The Adult Spine: Principles and Practice. New York: Raven Press, 1991:2107-30.
6. Bernard TN Jr, Kirkaldy-Willis WH. Recognizing specific characteristics of non-specific low back pain. Clin Orthop 1987;217:266-80.
7. Bogduk N, April' C, Derby R. Diagnostic blocks of spinal synovial joints. hi: White A, ed. Spine Care: Diagnosis and Conservative Treatment. St Louis: Mosby, 1995:298-321.
8. Borenstein D, Wiesel SW. Low back pain: Medical Diagnosis and Comprehensive Management. Philadelphia: W.B. Saunders, 1989:59-8 I.
9. Carmichael JP. Inter- and intra-tester reliability of palpation for sacroiliac joint dysfunction. J Manipulative Physiol Ther 1987;10:164-71.
10. Cassidy DJ, Mierau DR. Pathophysiology of the sacroiliac joint. In: Haldeman Scott, ed. Principles and Practice of Chiropractic. 2nd ed. San Mateo, CA: Appleton & Lange, 1992:211-24.
11. Cilbulka MT, Delitto A, Koldehoff RM. Changes in innominate tilt after manipulation of the sacroiliac joint in
patients with low back pain. An experimental study. Phys Ther 1988;68:1359-63.
12. Cibulka MT, Rose SJ, Delitto A, et al. Hamstring muscle strain treated by mobilizing the sacroiliac joint. Phys Ther I 986;66:1220-3.
13. Coy JT, Wolf, CR, Brower JD, Winter WG. Pyogenic arthritis of the sacroiliac joint. J Bone Joint Surg [Am] 1976;58:845-9.
14, Dreyfuss P, Cole A, Pauza A. Sacroiliac joint injections. In: Weinstein S, ed. Injection Techniques: Principles and
Practice. Clinics of North America in Physical Medicine and Rehabilitation, Philadelphia: WB Saunders, 1995.
15. Dreyfuss P, Dreyer S, Griffin J, Hoffman J, Walsh N. Positive sacroiliac screening tests in asymptomatic adults. 1994;19:1138-43.
16. Dvorak J, Dvorak V. Manual Medicine: Diagnostics. New York: Thieme Medical Publishers Inc., 1990:109-25.
17. Fortin JD. Sacroiliac joint dysfunction: a new perspective. J Back Musculoskel Rehabil 1993;3:31-43.
18. Fortin JD. Sacroiliac joint injection and arthrography with imaging correlation. In: Lennard T, ed. Physiatric Procedures in Clinical Practice. Philadelphia: Hanley & Belfus, Inc. 1995:242-53.
19. Fortin JD, Aprill CN, Ponthieux B, Pier J. Sacroiliac joint: pain referral maps upon applying a new injection/arthrography technique. Part II: Clinical evaluation. Spine 1994;19:1483-9.
20. Fortin JD, Dwyer AI', West S, Pier J. Sacroiliac Joint: pain referral maps upon applying a new injectionlarthrography technique. Part I: Asymptomatic volunteers. Spine 1994;19:1475-82.
21. Greenman PE. Principles of diagnosis and treatment of pelvic girdle dysfunction. In: Principles of Manual Medicine. Baltimore: Williams and Wilkins, 1989:225-70.
22. Haldeman KO, Soto-Hall R. The diagnosis and treatment of sacroiliac conditions by the injection of procaine (novocain). J Bone Joint Surg [Am] 1938;20:675-85.
23. Herzog W, Read LJ, Conway PJW, Shaw LD, McEwen MC. Reliability of motion palpation procedures to detect sacroiliac joint fixations. J Manipulative Physiol Ther 1989;12:86-92.
24. Hesch I, Aisenbrey JA, Guarino J. Manual therapy evaluation of the pelvic joints using palpatory and articular spring tests. In: Vleeming A, Mooney V, Snijders C, Dorman T, eds. Proceedings: First Interdisciplinary World Congress on Low Back Pain and its Relation to the Sacroiliac Joint. San Diego, CA: 1992;435-59.
25. Stanley Hoppenfeld. Physical Examination of the Spine and Extremities. Norwalk, CT: Appleton and Lange, 1976:261-2.
26. Laslett M, Williams M. The reliability of selected provocation tests for sacroiliac joint pathology. Spine 1994;19:1243-9.
27. Last John M, ed. A Dictionary of Epidemiology. 2nd ed. Oxford: Oxford University Press, 1988.
28. David J Magee. Orthopedic Physical Assessment. Philadelphia: W.B. Saunders, 1987:219-238, 253.
29. McCombe PF, Fairbank JCT, Cockersole BC, Pynsent PB. Reproducibility of physical signs in low back pain. Spine 1989;14:908-8.
30. Merskey H, Bogduk N, eds. Classification of Chronic Pain: Descriptions of Chronic Pain Syndromes and Definitions of Pain Terms. 2nd ed. Seattle: IASP Press, 1993:190-1.
31. Mitchell FL Jr, Moran PS, Pruzzo NA. An Evaluation and Treatment Manual of Osteopathic Muscle Energy Techniques. Valley Park, MO: Mitchell, Moran, and Pruzzo Associates, 1979:49-62, 109-55.
32. Norman GF. Sacroiliac disease and its relationship to lower abdominal pain. Am J Surg 1968;1 16:54-6.
33. Porterfield James A, Derosa Carl. Functional assessment of the lumbopelvic region. In: Mechanical Low Back Pain: Perspectives in Functional Anatomy. Philadelphia: W.B. Saunders, 1991:123-63.
34. Potter NA, Rothstein JM. Intertester reliability for selected clinical tests of the sacroiliac joint. Phys Ther 1985;65:1671-5.
35. John Prior, Jack Silberstein, John Stang. Physical Diagnosis: The History and Physical Examination of the Patient. St. Louis: C.V. Mosby Co., 1981:462-3.
36. Russell AS, Maksymowych, LeClercq S. Clinical examination of the sacroiliac joints: A prospective study. Arthritis Rheum 1981;24:1575-7.
37. Schwarzer AC, Aprill CN, Bogduk N. The sacroiliac joint in chronic low back pain. Spine 1995;20:31-7.
38. Schwarzer AC, Aprill CN, Derby R, Fortin J, Kine G, Bogduk N. The false positive rate of uncontrolled diagnostic blocks of the lumbar zygapophysial joints. Pain 1994;58:195-200_
39. Schwarzer AC, Wang S, Bogduk N, MeNaught PJ, Laurent R. Prevalence and clinical features of lumbar zygapophysial joint pain: A study in an Australian population with chronic low back pain. Ann Rheum Dis 1995;54:100-6.
40. Schuchmann JA, Cannon CL. Sacroiliac strain syndrome: Diagnosis and treatment. Tex Med 1986;82:33-6.
41. Van Duersen LLJM, Patijn Ockhuysen AL, Vortman BJ. The value of some clinical tests of the sacro-iliac joint. Manual Med 1990;5:96-9.
42. Walker JM. The sacroiliac joint: A critical review. Phys Ther 1992;72:903-16.
43. Woerman AL. Evaluation and treatment of dysfunction in the lumbar-pelvic-hip complex. In: Donatelli and Wooten, eds. Orthopedic Physical Therapy. New York: Churchill Livingstone, 1989:403-84.
In Gaenslen's test the patient lies supine as the contralateral hip is flexed, and the ipsilateral hip is extended:L*-= The midline sacral thrust test requires the examiner to apply a posteroanterior force to the sacrum as the patient lies prone.— The results of these four tests are considered positive if the maneuver aggravates the patient's typical pain.
Sacral sulcus tenderness refers to tenderness immediately medial to the posterior superior iliac spine.- Joint "play" is assessed over the superior pole of the joint by palpating the spring of the joint at end range and comparing the asymptomatic and symptomatic sides. For this test, crossed hands are placed over the sacrum; the overlying hand delivers a postero-anterior thrust, while the pisiform region of the underlying hand serves to detect joint play. This test is performed in a manner similar to the technique used for a sacroiliac shear test described by Bernard and Cassidy:-
Patients for the present study were selected from among those referred to one of the authors for sacroiliac joint blocks. These patients were referred by local physiatrists, chiropractors, family physicians, orthopedic surgeons, and neurosurgeons. Patients were selected for study if their pain was principally, although not exclusively, below L5 in a pattern consistent with patterns reported to be characteristic of sacroiliac joint pain in previous studies of healthy volunteers and symptomatic patients.
The study was approved by the Institutional Review Board for Human Research Rights at East Texas Medical Center Hospital in Tyler, Texas. After informed consent was obtained, patients answered a two-page questionnaire about their medical history and completed a drawing to indicate the location of their pain. Questions asked were: "are you on Worker's Compensation"; "does an attorney represent you"; "how long have you had your pain"; and "what was your average level of pain during the last week and on the day of the injection on a 0-10 analog scale (0 = no pain, 10 = worst imaginable pain)."
Patients also were asked if any of the following treatments were used and what effect this treatment had (increased, decreased, or did not affect their pain). The treatments listed were: anti-inflammatory medications, "muscle relaxers," physical therapy, home exercise, application of local heat and cold, and manipulations of the sacroiliac joint.
Patients were asked if a specific, traumatic event caused their pain and if this event was caused by twisting, heavy lifting, a fall on the buttocks, a motor vehicle accident, a fall in a hole, a sports-related injury, a repetitive work injury, or another type of injury.
Patients were asked if the following activities increased, decreased, or did not affect their pain: walking, sitting, lying down, standing in place, wearing high heels or boots, straining with a bowel movement, coughing or sneezing, and usual job activities.
After completing the questionnaire and drawing, all patients were examined independently and sequentially by a physician and a chiropractor using all 12 physical examination tests described previously. Next, all patients were injected intra-articularly with an anesthetic and corticosteroid into the sacroiliac joint on the same side of their body that their pain was located. Specific details of the
technique used have been described in a previous report.li In brief, the patient lies in a prone position, and his or her back is prepped with povoiodine, chlorhexadine, and isopropyl alcohol. A 22-gauge, 3.5-inch, spinal needle is placed into the inferior quarter of the joint; the patient is not sedated. The needle is placed into a hyperlucent zone (if visible) in the medial joint silhouette (posterior joint plane) when the medial joint silhouette is defined maximally by rotating the C-arm toward the side of injection to separate the images of the anterior and posterior joint margins (Figure 1). If this zone is not visible, the needle is inserted as well as possible into the lower end of the sacroiliac joint cavity.
After the needle is placed into the joint, 0.2-0.5 cc of contrast medium is injected until intra-articular spread is obtained and confirmed. If necessary, the needle is repositioned to obtain confined intra-articular spread. After the contrast medium is injected, 1.5 cc of 2% lignocaine and 0.5 cc of celestone soluspan is injected unless a firm end-point is reached before this volume. Intermittent imaging is used to determine if extra-articular
spread occurs, in which case the injection is ceased.
During the injection, pain provocation was assessed. The patient was asked whether the provoked pain was in the exact area of their "main pain."
After the injection, the patients moved about freely, ambulated, and went through a general range of trunk motion. At 20 minutes postinjection, the patients were assessed for a change in their pain status. The patients graded the reduction in their "main pain" for which they sought medical attention in 5% increments. A grade of 0% was used to indicate no pain reduction, and a grade of 100% indicated complete reduction of their pain. The patient had to experience a 90-100% reduction in pain after the block to obtain a diagnosis of sacroiliac joint pain. Any reduction of pain less than this was considered a negative block. The patient's pain reduction was reported to an independent technician or nurse.
Once all blocks had been completed, 2 x 2 contingency tables were constructed to determine the association between response to block and results of the questionnaire and physical examinations. Separate tables were constructed for the physical tests performed by each of the two examiners, and additional tables were constructed based on those cases in which the two examiners agreed on their findings.
Based on the definitions provided in "A Dictionary of Epidemiology,"= the number of true positive and true negative observations were recorded for each table; the sensitivity, specificity, and likelihood ratio of each test and a Pearson chisquared P value were calculated using response to block as the criterion standard. Separately. agreement between examiners was determined for each test executed, and a kappa statistic was calculated.
Results
Three patients who initially were invited to participate in the study subsequently refused because of time concerns. Otherwise, a total of 85 consecutive, eligible patients were assessed. There were 61 women Ind 24 men aged 18-87 years (median age, 44.5 years). The mean pain rating on the day of injection was 5.9 (standard deviation of 1.9); there were eight patients who had had pain for less than 3 months, 19 whose pain had lasted for 3-6 months, 17 whose pain had lasted for 7-11 months, 13 whose pain had asted for 12-24 months, and 28 whose pain had lasted for more than 24 months. Thirty-eight patients were under Worker's pensation, 21 were represented by an attorney, and 26 were private payers or under nedicare/medicaid.
411 patients were examined. Sixty-eight patients underwent unilateral blocks; 17 underwent bilateral )locks. In only one instance could the joint not be successfully infiltrated. There were 45 patients who .esponded positively to the diagnostic block, and 40 patients who responded negatively. Of the 40 )atients who responded negatively, seven reported between 51% and 89% reduction in pain, and 33 reported less than 50% reduction in pain.
Patients with positive responses to diagnostic blocks experienced various and broad patterns of referred pain that encompassed the entire lower limb.
Similar patterns were experienced, however, by patients who responded negatively to diagnostic blocks.
The only relative difference between the composite pain maps of patients with positive and negative responses was the presence of pain above L5 in those patients with no sacroiliac joint pain. Only two patients with confirmed sacroiliac joint pain drew even a trace of pain above L5.
When comparing results of physical examination with response to diagnostic blocks, it was found that the results of left and right procedures in patients who underwent bilateral blocks were not statistically independent. Consequently, for these 17 patients only one result was entered, that of the right-hand side, which resulted in a total of 85 patients/blocks for analysis.
With respect to medical history, no aggravating or relieving factor was of value for diagnosing the presence of sacroiliac joint pain as established by intra-articular blocks. All of these features in the patients medical history had poor sensitivity, poor specificity, or both.
Relief of pain by standing had a high specificity and likelihood ratio, but this may not be a significant finding because very few patients were relieved by standing. Provocation of pain using diagnostic blocks was not indicative of subsequent relief.
Pain relief as a result of previous therapy generally was not indicative of sacroiliac joint pain. Significant associations did emerge, however, between positive response to previous physical therapy and manipulation of the sacroiliac joint. Those associations, however, were for the most part a result of the few false-positive responses to these interventions, regardless of the number of false-negative responses to these interventions.
None of the 12 physical examination tests proved to be diagnostically sound whether they were administered by the physician or by the chiropractor. Although the tests varied with respect to sensitivity and specificity, none had a likelihood ratio of greater than J.3.
Agreement between the findings of the physician and chiropractor was appreciably better than it might have been by chance; many-kappa scores were greater than, 0.6. The worst kappa scores were seen with the Gillet test (k = 0.22) and the spring test (k = I/15), which are motion demand tests. There was no improvement in the diagnostic value of any test, however, among those cases in which the
physician and the chiropractor agreed completely in their physical findings.
Four tests demonstrated better sensitivity than the other eight. Sacral sulcus tenderness was the most sensitive test, followed by pain over the sacroiliac joint, buttock pain, and patient pointing to the PSIS as the main pain source. Using combinations of sacral sulcus tenderness with the most specific, reliable test (groin pain) and the test with the best positive predictive value (PSIS pointing) did not improve the diagnostic utility.
Increasing the number of tests that yielded positive results did not improve the diagnostic power. The sensitivity, specificity, and likelihood ratios remained poor even if 6, 7, 8, 9, 10, or 11 of 12 tests yielded positive results as determined by the physician, chiropractor, or both.
Discussion
The diagnostic value of examination tests for sacroiliac joint pain has been questioned previously by others in regard to pain resulting from inflammatory arthropathy, pyogenic infection, and sacroiliac joint dysfunction: The results of the present study vindicate these reservations and offer little
support to proponents of the use of physical examination for diagnosis.
Tests such as Patrick's test and Gaenslen's test have been promulgated in the literature as reliable for the diagnosis of sacroiliac joint pain, but their accepted validity stems not from scientific study but from repetitive citations in standard orthopedic, medical, manual medicine, osteopathic, and chiropractic.
The results of the present study, at least with respect to the 12 tests evaluated, refute the validity of these purported diagnostic tests.
Some aficionados of manual examination may feel disaffected because their particular favorite test was not assessed. It was not practical, however, to investigate every test devised to detect sacroiliac joint pain. In the interests of efficiency, the study focused on those signs most widely endorsed by the experts consulted. If proponents of other tests believe that their tests are superior, they have the responsibility and the means to validate those tests by challenging them with diagnostic, intra-articular sacroiliac joint blocks as described in this study.
A stringent criterion was used in this study for the diagnosis of sacroiliac joint pain; patients had to report 90% or more relief of their pain after a diagnostic block was done. This ensured that the physical tests were challenged against patients most likely to experience pain purely of intra-articular, sacroiliac origin. Under this condition, the sensitivity and specificity of the physical test optimally would be revealed. Even so, the tests failed to show any diagnostic value.
Control blocks previously have been advocated for evaluating synovial joint-mediated pain in the
cervical and lumbar zygapophysial joints.=:-=='22 In the present study, control blocks were not used for ethical and logistic reasons. Patients who are treated by physicians in private practice are generally reluctant to undergo multiple procedures. Control blocks are necessary; however, they are critical only if the conclusions of a study are positive and assertive. The lacleof controls does not compromise a study with negative results. Given the poor sensitivity and specificity already evident without controls, implementing controls would have served only to increase the negativity of the results.
Much attention has been paid to evaluating the intertester and intratester reliability of various sacroiliac
physical tests.2:=1-'2-1'='2:-='= Such investigations were rendered immaterial by the results of the present study, however, because if the tests have no external validity there is no point in measuring their reliability.
Nevertheless, the Gillet test and evaluation of "joint restriction" in the superior pole of the sacroiliac joint were found to have poor intertester reliabilities. This finding agrees with those of other investigators.
The results reinforce those of Schwarzer et al- with respect to clinical features. Patients with sacroiliac joint pain exhibit no characteristic feature such as aggravation or relief of their pain by sitting, walking, standing, flexion. or extension. Nor does provocation of pain by injection predict relief of pain by intra-articular local anesthetic.— Although provocation has been suggested as a diagnostic tool, it is obvious
this should not be used in the diagnosis of sacroiliac joint pain.
Unlike Schwarzer et the authors of the present study did not find that groin pain was a discriminating feature of sacroiliac joint pain. No explanation is offered for this difference.
Injury as the usual cause of sacroiliac joint pain was not substantiated. This is contrary to the belief of other clinicians, but in the light of the present results, superior data is necessary to either refute or perpetuate this belief encompass even the calf and foot in some patients, a feature previously reported by Schwarzer et In this respect, however, the patterns do not differ from those experienced by patients with no sacroiliac
joint pain. This is similar to the finding of Schwarzer et al-L that the presence of referred pain in the buttock, thigh, calf, or foot does not discriminate for or against sacroiliac joint pain. It is noteworthy, however, that in the present study only two patients with confirmed sacroiliac joint pain drew even a trace of pain above L5; yet it was not uncommon to have pain above L5 in patients with no sacroiliac joint pain. This finding may well be a discriminating feature worthy of formal analysis in the future. Presently, it remains a putative discriminating feature of unknown power.
While refuting the diagnostic value of physical examination tests for sacroiliac joint pain, this study offers little as a positive conclusion. It is evident from previous studies that the sacroiliac joint can be and often is a source of back pain and referred pain in the lower limb, but this pain cannot be diagnosed from history or physical examination.-- No single physical test or combination of tests evaluated carried a likelihood ratio substantially greater than 1.0, which means that these physical examination tests do not alter appreciably the likelihood of the diagnosis beyond its pretest probability, Le., its natural prevalence. If the diagnosis is to be made, controlled diagnostic blocks are, at present, the only means to make that diagnosis.
Acknowledgment
The authors thank Gary Eaton, DC, DO, for his endorsement of this study by a generous referral of patients.
|
