Use of limited MR protocol (coronal STIR) in the evaluation of patients with hip pain

To assess the role of a limited MR protocol (coronal STIR) as the initial part of the MR examination in patients with hip pain. Eighty-five patients presenting with hip pain, and normal radiographs of the pelvis, and who underwent our full MR protocol for hips were included retrospectively in the study. The full protocol consists of coronal T1-weighted and short tau inversion-recovery (STIR), and axial T2-weighted sequences. Ninety-three MR examinations were performed. Two radiologists interpreted the STIR (limited) examinations and the full studies separately, masked to each other's findings and to the final diagnosis. Comparison between the two protocols was then undertaken. For both readers, all normal MR examinations on the coronal STIR limited protocol were normal on the full protocol, with an interobserver reliability of 0.96. The STIR protocol was able to detect the presence or absence of an abnormality in 100% of cases (sensitivity). The STIR-only protocol provided a specific diagnosis in only 65% of cases (specificity). A normal coronal STIR study of the hips in patients with hip pain and normal radiographs precludes the need for further pelvic MR sequences. Any abnormality detected on this limited protocol should be further assessed by additional MR sequences. (orig.)