THE BEAR-HUG TEST FOR THE DIAGNOSIS OF SMALL INFRASCAPULAR TEARS: 68 CLINICAL TESTS COMPARED WITH ARTHROSCOPIC DATA

Abstract

Purpose of the study: The objective of this study was to investigate a new test (the bear-hug test) in search of a more sensitive way of diagnosing small infrascapular tears. The bear-hug test was compared with other tests (lift-off, belly-press, Napoleon).

Material and methods: From January to March 2004, 68 patients were scheduled for arthroscopic shoulder surgery. We searched for correlations between the preoperative clinical results and the anatomic observations during the diagnostic phase of the arthroscopy.

Results: The prevalence of infrascapularis lesions was 29.4%. Forty percent of the infrascapularis lesions had not been identified by any of the clinical diagnostic tests. The bear-hug test was the most sensitive (60%) compared with the belly-press test (40%), the Napoleon test (25%), and the lift-off test (17.6%). The lift-off test was the most specific (specificity 100%) versus 97.9% for the Napoleon test, 97.9% for the belly-press test and 91.7% for the bear-hug test. The bear-hug test was more sensitive than the Napoleon test and the lift-off test with statistically significant difference, irrespective of the size of the tear (p< 0.05), but this difference was not found for the belly-press test (p> 0.06). Conversely, for small tears (50%), there was a statistically significant difference in favor of the bear-hug test (Se bear-hug 50%; Se belly-press 21.4%; p< 0.035). The bear-hug test was thus moe sensitive for diagnosing small tears of the upper third of the infrascapularis. A positive bear-hug or belly-press tes would suggest at least a 30% infra-scapularis tear, while a positive Napoleon test suggests at least a 50% tear. The lift-off test is only positive when 75% of the infraspinatus is injured.

Conclusion: The sensitivity of the bear-hug test optimizes chances of detecting a tear of the infrascapularis at the physical examination. Combining all of these tests is useful for predicting the size of the tear.