Diagnosis of Sacroiliac Joint Pain: Predictive Value of Three Diagnostic Clinical Tests

To date, there have been no acceptable and accurate diagnostic criteria or standards of care for the management of sacroiliac joint (SIJ) pain. Several studies have yielded different contributions of clinical presentation, history, and physical examination in the diagnosis of SIJ pain. Our goal in this study was to assess the sensitivity and specificity of the diagnostic clinical tests and their predictive value in accurately diagnosing SIJ pain.

Upon enrolling 200 eligible patients with SIJ pain as their primary diagnosis, they were re-evaluated and their verbal rating scale (VRS) pain scores and demographic data were obtained. Thereafter, three SIJ diagnostic tests were performed: the thigh thrust test, the Patrick test, and a modified version of the Gaenslen test that is referred to as the Mekhail test. Subsequently, the patients were taken to the procedure room to undergo SIJ injection, for which a confirmative result was ≥50% pain relief. The physicians performing the procedure were blinded of the results of the 3 tests performed. Results from the 3 tests were incorporated with the procedure results, from which we drew statistical and medical conclusions determining their predictive value and degree of aid to physicians in diagnosing SIJ pain.

We found that the cumulative effect of adding simultaneous tests increased the sensitivity of the testing but decreased the specificity, which generates a powerful screening tool. The combination of the Patrick and Mekhail tests demonstrated the best accuracy, with 94% sensitivity, 17% specificity, 81% positive predictive value, and 44% negative predictive value. The Patrick test was better than other tests for differentiating patients with SIJ pain from those with non-SIJ pain. No combination yielded both significant sensitivity and specificity. Generally, the overall predictive value of any of the tests on their own or their combination did not vary significantly from the predictive value of baseline demographics, including pre-injection pain score and pain referral diagram.

In conclusion, our study results were similar to those of previous investigators who found that physical examination plays a limited role in diagnosing SIJ pain. Specifically, we found that the clinical tests and/or their combinations added no significant predictive capacity compared to patients' baseline characteristics in predicting the response to diagnostic SIJ injection, albeit the combination of the Mekhail and Patrick tests yielded high sensitivity (94%), making them viable for consecutive screening, possibly reducing the unnecessary costs of diagnostic SIJ injection procedures.