INTRODUCTIONShoulder instability is a common problem with annual incidence between 0.084%-1.7%.1 There are 46%-100% incidence of recurrent dislocation for patients aged less than 40 years without surgical intervention, which decreases to 7%-15% after surgical stablization.1 To improve the pre-operative planning and post-operative result of surgery, exact diagnosis of pathology and anatomy of labroligamentous structure is essential.Conventional computerized tomographic arthrography (CTA) has limited soft tissue contrast and spatial resolution which has led to it being replaced by conventional and contrast enhanced magnetic resonance imaging (MRI) of shoulder.2-6 At present, magnetic resonance arthrography (MRA) of shoulder is well proven and useful technique for diagnosis of intraarticular lesions.5,11 However it is still expensive and metal in the vicinity interferes with the true signal.Since multidetector CT (MDCT) was introduced in 1988, cross-sectional imaging has been revolutionized by the ABSTRACT Background: Purpose of this study was to compare diagnostic effectiveness of MDCT arthrography (MDCTA) in shoulder instability and pain in throwing and its comparison to MR arthrography (MRA) and arthroscopy taking arthroscopy as gold standard. Methods: 20 patients with history of recurrent shoulder dislocation in activity were included in this study. After detailed clinical examination, each patient underwent MDCT-MR arthrography in one sitting followed by diagnostic arthroscopy within 6 weeks. Results were compared with the help of statistician. Results: At arthroscopy, 10 Bankart's lesions, 7 Hill Sachs lesion, 6 SLAP lesion, 1 ALPSA, 1 capsular laxity, 1 partial subscapularis tear and 1 supraspinatus fraying were visualized in 20 shoulders. For Bankart's lesion MDCT has sensitivity 80%, specificity 100%, positive predictive value (PPV) 100% and negative predictive value (NPV) 83.3%. MRA has sensitivity of 90%, specificity 100%, PPV 100% and NPV 90.9%. For SLAP lesions sensitivity, specificity, PPV and NPV for MDCTA and MRA are 88.3%, 100%, 100%, 93.3%. For Hill-Sachs lesion; sensitivity, specificity, PPV and NPV for MDCTA are all 100% and for MRA they are 85.7%, 100%, 100%, 92.8% respectively. For ALPSA; sensitivity is 100%, specificity is 95%, PPV is 50% and NPV is 100% both for MDCTA and MRA. К value for MRA is 0.60 and for CTA is 0.55 suggesting moderate agreement. Conclusions: Considering availability, cost, time consumption, superior detection of bony lesions and comparable detection of soft tissue lesions; MDCTA can be used as single investigation of choice in shoulder instability pain.