The Validity of the Lag Signs in Diagnosing Full-Thickness Tears of the Rotator Cuff: A Preliminary Investigation

Abstract

Miller CA, Forrester GA, Lewis JS. The validity of the lag signs in diagnosing full-thickness tears of the rotator cuff: a preliminary investigation. Objective To investigate whether the lag signs were valid tools in diagnosing full-thickness tears of the rotator cuff. Design A same-subject, correlation, double-blinded design was used. The results of the external rotation lag sign, drop sign, and internal rotation lag sign were compared with the criterion standard of diagnostic ultrasound to establish their accuracy. Setting A regional orthopedic hospital. Participants Consecutive subjects (N=37), 21 women and 16 men, with shoulder pain referred to a consultant orthopedic surgeon specializing in shoulder conditions were recruited for this investigation. Interventions Not applicable. Main Outcome Measures Sensitivity, specificity, and positive and negative likelihood ratios of the lag signs when using ultrasound as the reference test. Results The specificities of the drop sign and internal rotation lag sign were 77% and 84%, respectively, which, together with low positive likelihood ratios 3.2 (95% confidence interval [CI], 1.5–6.7) and 6.2 (95% CI, 1.9–12.0), indicate that a positive result was poor at recognizing the presence of full-thickness tears. The drop sign had a sensitivity of 73% with a negative likelihood ratio of .34 (95% CI, 0.2–0.8), suggesting that a negative test was fair at ruling out the presence of full-thickness tears. The sensitivity of the internal rotation lag sign (100%) supported by the negative likelihood ratio of 0 (95% CI, 0.0–2.5) suggests that a negative test will effectively rule out the presence of full-thickness tears of the subscapularis. A positive external rotation lag sign is the clinical test most likely to indicate that full-thickness tears of the supraspinatus and infraspinatus are present (specificity, 94%). However, the external rotation lag sign did demonstrate a low sensitivity score of 46% and negative likelihood ratio of .57 (95% CI, 0.4–0.9), which means that a negative test will not rule out the presence of full-thickness tears. Conclusions The findings of this investigation suggest that a clinical diagnosis of a full-thickness tear of the rotator cuff cannot be conclusively reached using one or more of the lag signs.