The Bear-Hug Test: A New and Sensitive Test for Diagnosing a Subscapularis Tear Johannes R. H. Barth, M.D., Stephen S. Burkhart, M.D., and Joe F. De Beer, M.D. Purpose: It was our intent to devise a new clinical test that would more accurately diagnose subscapularis tears than the current clinical tests. This new test is called the bear-hug test. The purpose of this study was to assess the bear-hug test and compare it with the current tests of subscapularis function (lift-off, belly-press, and Napoleon tests). Methods: Between January 2004 and March 2004, 68 consecutive patients scheduled for an arthroscopic procedure were evaluated preoperatively; the preoperative clinical examination findings were then correlated with arthroscopic findings. Lift-off, belly-press, Napoleon, and bear-hug tests were included in the examination. Furthermore, for the belly-press and bear-hug tests, the strength was precisely quantified by means of an electronic digital tensiometer (Kern HBC). Diagnostic arthroscopy was the reference that determined the actual pathologic lesions. Results: Subscapularis tears occurred with a prevalence rate of 29.4%. Of the subscapularis tears, 40% were not predicted by preoperative assessment by use of all of the tests. The bear-hug test was found to be the most sensitive test (60%) of all of those studied (belly-press test, 40%; Napoleon test, 25%; and lift-off test, 17.6%). In contrast, all 4 tests had a high specificity (lift-off test, 100%; Napoleon test, 97.9%; belly-press test, 97.9%; and bear-hug test, 91.7%). No statistically significant difference was found between the area under the receiver operating characteristic curve of the bear-hug test and that of the belly-press test in diagnosing a torn subscapularis. However, the areas under the receiver operating characteristic curve for both the bear-hug test and the belly-press test were significantly greater than those for the lift-off and Napoleon tests (P  .05). Positive bear-hug and belly-press tests suggest a tear of at least 30% of the subscapularis, whereas a positive Napoleon test indicates that greater than 50% of the subscapularis is torn. Furthermore, a positive lift-off test is not found until at least 75% of the subscapularis is torn. Conclusions: The bear-hug test optimizes the chance of detecting a tear of the upper part of the subscapularis tendon. Moreover, because the bear-hug test represents the most sensitive test, it can be considered to be the most likely clinical test to alert the surgeon to a possible subscapularis tear. Performing all of the subscapularis tests is useful in predicting the size of the tear. Level of Evidence: Level I, diagnostic study: testing of previously developed criteria in a series of consecutive patients with arthroscopy used as the criterion standard. Key Words: Bear-hug test—Subscapularis—Rotator cuff tear—Shoulder—Rotator cuff tendon. D espite the importance of the subscapularis as a major muscle of the rotator cuff, there are few clinical tests that can detect tears or dysfunction of this muscle-tendon unit. Gerber and colleagues described the lift-off test in 1991 1 and the belly-press test in 1996. 2 The Napoleon test, first described by Schwam- born and Imhoff 3 and further refined by Burkhart and Tehrany, 4 is a graded test related to the percentage of the tendon that is torn. Tokish et al., 5 in an electromyographic study, showed that the belly-press test activated the upper subscapularis muscle significantly more than the lift-off test whereas the lift-off test caused greater activation of the lower From The San Antonio Orthopaedic Group (J.R.H.B., S.S.B.), San Antonio, Texas, U.S.A.; Cape Shoulder Institute (J.F.B.), Cape Town, South Africa; and Department of Anatomy and Histology, Faculty of Health Sciences, University of Stellenbosch (J.F.B.), Cape Town, South Africa. Address correspondence and reprint requests to Stephen S. Burkhart, M.D., 540 Madison Oak Dr, Suite 620, San Antonio, TX 78258, U.S.A. E-mail: ssburkhart@msn.com © 2006 by the Arthroscopy Association of North America 0749-8063/06/2210-4534$32.00/0 doi:10.1016/j.arthro.2006.05.005 1076 Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 22, No 10 (October), 2006: pp 1076-1084