An assessment of the inter examiner reliability of clinical tests for subacromial impingement and rotator cuff integrity

Abstract

Over 30 separate clinical signs for the shoulder have been described, most with little evidence to support their accuracy and reliability. The aim of our study was to evaluate the accuracy and reliability of some of the commonly used tests for rotator cuff disease. Two clinicians, a consultant with an established shoulder practice and a registrar with an interest in shoulder surgery, examined 63 patients with history suggestive of rotator cuff disease. A set of pre-determined clinical tests for impingement syndrome and rotator cuff tear were standardised and agreed upon before the study was commenced. The examination included eliciting a painful arc, the drop arm test, the Neer’s and the Hawkin’s sign for impingement syndrome. Integrity of the individual components of the rotator cuff was assessed by the strength of abduction initiation and at 90° abduction for supraspinatus tear (Jobe’s test), Speed’s and Yergason’s test for biceps pathology, strength of shoulder external rotation for infraspinatus tears and the Gerber lift-off test for subscapularis tears. Accuracy assessments were determined by comparing clinical findings against findings at arthroscopy in 50 of the 63 patients. The two observers did not differ significantly in their assessments. Un-weighted kappa values defining agreement between the two observers and the positive predictive values were evaluated. Our findings indicate that clinical signs can be relied upon for diagnosis of impingement but not for rotator cuff deficiency. Compared with other previously published studies, our study demonstrates that the inter observer agreement is better when the assessors have a special interest and understanding of shoulder disorders.