Popeye Sign After Tenotomy Versus Tenodesis: Letter to the Editor:

Dear Editor: With great interest we have read the article of Cho and colleagues. The authors concluded that both tenotomy and tenodesis are equally effective for the treatment of long head biceps lesions associated with arthroscopic rotator cuff repair. They found no significant effect on clinical outcome as measured by the Constant score, no significant effect on the University of California, Los Angeles (UCLA) score, and no significant difference in the overall incidence of cosmetic deformity, as measured by the popeye deformity. The presented study is an evidence level 3 trial based on 83 patients and contributes to the need for high-level evidence on this topic. Unfortunately, no sample size calculation was performed prior to the initiation of this study. Because the highest level of evidence is a meta-analysis, we conducted a literature search in Pubmed/Medline with the search terms biceps tendon, tenotomy, tenodesis, rotator cuff, and shoulder. Based on this search, we found 4 similar studies that provided data on Constant score or popeye signs, which enabled us to conduct a meta-analysis on the effectiveness of tenotomy versus tenodesis of the long head of the biceps when performed with rotator cuff repair, and we found that 372 patients could be added to the 83 patients in Cho et al study, which is the most recent trial published in this journal. The results of our meta-analysis showed a higher incidence of the popeye sign in the patients with a tenotomy compared with patients with a tenodesis of the long head of the biceps (Figure 1). In previous studies there seemed to be no difference in clinical outcome assessed by the Constant score. Our meta-analysis revealed a strong trend to a difference in functional outcome (Figure 2) in favor of the tenodesis patients. However, a significant difference was not established (95% CI, 0.09-5.68). The Constant score was also assessed with the fixed-effects model, and this did not show different results. The Constant score is not specifically designed or validated to measure functional performance of the shoulder after long head biceps treatment. It is difficult to isolate the influence of the biceps treatment on the functional outcome when performed in combination with a cuff repair. The lack of a statistically significant difference in clinical outcome may therefore not be a correct reflection of the patients’ perceived functional results of biceps treatment. Pooling the data from different trials can help in this respect. Updating meta-analyses on a regular basis provides the physician with up-to-date best evidence and helps them in decision making in clinical practice. Currently, 3 new trials are recruiting patients, and an update of our meta-analysis should be performed after these highquality trials, reporting on patient-related outcome measures such as complication rate and validated patient-reported outcomes, are finished.