RESUMEN
BACKGROUND: Arthroscopic examination of the tendon has been described as the "gold standard" for diagnosis of tendinitis of the long head of the biceps (LHB). An arthroscopic finding of an inflamed and hyperemic LHB within the bicipital groove has been described as the "lipstick sign." Studies evaluating direct visualization in diagnosis of LHB tendinitis are lacking. METHODS: During a 1-year period, 363 arthroscopic shoulder procedures were performed, with 16 and 39 patients prospectively selected as positive cases and negative controls, respectively. All positive controls had groove tenderness, positive Speed maneuver, and diagnostic ultrasound-guided bicipital injection. Negative controls had none of these findings. Six surgeons reviewed randomized deidentified arthroscopic pictures of enrolled patients The surgeons were asked whether the images demonstrated LHB tendinitis and if the lipstick sign was present. RESULTS: Overall sensitivity and specificity were 49% and 66%, respectively, for detecting LHB tendinitis and 64% and 31%, respectively, for erythema. The nonweighted κ score for interobserver reliability ranged from 0.042 to 0.419 (mean, 0.215 ± 0.116) for tendinitis and from 0.486 to 0.835 (mean, 0.680 ± 0.102) for erythema. The nonweighted κ score for intraobserver reliability ranged from 0.264 to 0.854 (mean, 0.615) for tendinitis and from 0.641 to 0.951 (mean, 0.783) for erythema. CONCLUSIONS: The presence of the lipstick sign performed only moderately well in a rigorously designed level III study to evaluate its sensitivity and specificity. There is only fair agreement among participating surgeons in diagnosing LHB tendinitis arthroscopically. Consequently, LHB tendinitis requiring tenodesis remains a clinical diagnosis that should be made before arthroscopic examination.