brand logo

Am Fam Physician. 2001;64(7):1262

Tears of the rotator cuff muscles account for about one half of significant injuries involving the shoulder but can be difficult to diagnose. A variety of clinical tests have been recommended in addition to several radiologic modalities and direct arthroscopic examination of the joint. Murrell and Walton studied the ability of common clinical tests to diagnose rotator cuff tears in patients with shoulder injuries sufficiently severe enough to warrant arthroscopy.

Number of positive diagnostic features*Age group (years)Number of patients with rotator cuff tear (%)Number of patients with no tear (%)Post-Test probability (95% CI)
All 3Any48 (24)1 (0.5)0.98 (0.89 to 1.00)
Any 2< 6026 (13)14 (7)0.64 (0.47 to 0.79)
Any 2≥ 6048 (24)1 (0.5)0.98 (0.89 to 1.00)
Any 1< 403 (1.5)22 (11)0.12 (0.25 to 0.31)
Any 140-6952 (26)60 (30)0.45 (0.36 to 0.55)
Any 1≥ 7022 (11)6 (3.0)0.76 (0.56 to 0.90)
NoneAny1 (0.5)96 (48)0.05 (0.02 to 0.11)

The authors studied 400 patients who were referred to an Australian orthopedic surgeon because of significant shoulder injuries. The patients were systematically assessed using 23 common clinical tests of muscle wasting, tenderness and impaired function. Tests included the “drop-arm” sign, impingement, O'Brien's sign, active and passive range of motion, and strength tests. All patients subsequently underwent arthroscopy. The first 100 patients with partial or full-thickness tears of the rotator cuff but no other major shoulder pathology on arthroscopy were compared with the first 100 patients who were found to have other shoulder pathology but no damage to the rotator cuff.

Most clinical tests were unable to distinguish between the two groups. However, the tests for three clinical features were significantly predictive for rotator cuff tears. Supraspinatus weakness, weakness in external rotation, and impingement in external or internal rotation or both were more frequently positive in patients with rotator cuff tears. The drop-arm sign was highly specific (98 percent) but had a sensitivity of only 10 percent. At least 15 percent of the patients had both rotator cuff tears and additional shoulder pathology.

The authors calculate that patients who present with severe shoulder pain, supraspinatus weakness, impingement and weakness in external rotation have a 98 percent chance of rotator cuff tear on arthroscopy (see the accompanying table). In patients 60 years of age or older, any two positive tests are associated with a 98 percent chance of the diagnosis. A similar statistical probability exists for any patient with a positive drop-arm sign. Patients with none of the three clinical findings and a negative drop-arm sign have less than a 5 percent chance of having a rotator cuff tear. If only one of the three clinical tests is positive, imaging is indicated.

editor's note: This study may help physicians to accurately diagnose rotator cuff tears more quickly and with fewer confirmatory tests. However, the high levels of confidence reported may not be applicable to family practice because the study patients were selected based on severity of symptoms. In practice, it is also important to be alert for additional pathology when making the diagnosis of rotator cuff tear. In primary care, the rate may not be as high as the 15 percent reported in this study, but missing an additional condition in any patient could have serious consequences.—a.d.w.

Continue Reading

More in AFP

Copyright © 2001 by the American Academy of Family Physicians.

This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP.  See permissions for copyright questions and/or permission requests.