The aim of this study was to compare the efficacy of indirect MR arthrography images obtained following intravenous contrast injection and conventional MR imaging in the diagnosis of rotator cuff tears. Twenty-four patients with clinically suspected rotator cuff disease were examined. Conventional MR images and post-contrast indirect MR arthrography images were obtained. All images were evaluated in a blinded fashion by two musculoskeletal radiologist. Results were than analyzed depending on surgical output. The correlation coefficient (Spearman rank correlation test) and the kappa values for agreement between surgery and imaging techniques were calculated. The correlation coefficients between indirect MR arthrography and surgery for reader 1 and reader 2 were 0.9137 and 0.9773, respectively. Whereas the agreement between conventional MR imaging and surgery was moderate (kappa = 0.383-0.571), the agreement between indirect MR arthrography and surgery was excellent (kappa = 0.873-0.936). We suggest the use of indirect MR arthrography technique when conventional MR images are equivocal in diagnosis of rotator cuff disease.