Long Head of Biceps Tenotomy Is Not Inferior to Suprapectoral Tenodesis in Arthroscopic Repair of Nontraumatic Rotator Cuff Tears: A Multicenter, Non-inferiority, Randomized, Controlled Clinical Trial

Arthroscopy. 2021 Jun;37(6):1767-1776.e1. doi: 10.1016/j.arthro.2021.01.036. Epub 2021 Feb 6.


Purpose: To determine if long head of the biceps (LHB) tenotomy is not inferior to suprapectoral LHB tenodesis when performed in conjunction with arthroscopic repair of small- to medium-sized nontraumatic rotator cuff tears.

Methods: This multicenter, randomized, non-inferiority trial recruited 100 participants older than 50 years who had a supraspinatus and/or infraspinatus tear sagittally smaller than 3 cm and arthroscopically confirmed LHB pathology. During arthroscopic rotator cuff repair, we randomized 48 patients to undergo suprapectoral LHB tenodesis and 52 patients to undergo LHB tenotomy. Data were collected preoperatively and at 6 weeks, 3 months, and 1 year postoperatively. The primary outcome was non-inferiority of the Constant-Murley score (CMS) at 1-year follow-up. Secondary outcomes included the Dutch Oxford Shoulder Score; Disabilities of the Arm, Shoulder and Hand questionnaire; Popeye deformity; elbow flexion strength index; arm cramping pain; and quality of life (EQ-5D score). The integrity of the rotator cuff repair was assessed with magnetic resonance imaging. Differences between intervention groups were analyzed by mixed modeling.

Results: The mean CMS in the LHB tenotomy group improved from 44 (95% confidence interval [CI], 39-48) to 73 (95% CI, 68-79). In patients with LHB tenodesis, the mean CMS improved from 42 (95% CI, 37-48) to 78 (95% CI, 74-82). The difference between groups at 1-year follow-up was 4.8 (97.5% CI, -∞ to 11.4), with a P value for non-inferiority of .06. The secondary outcomes also improved over time, with no remarkable differences between groups. A Popeye deformity occurred in 33% of tenodesis patients and 47% of tenotomy patients (P = .17). Tenotomy was performed with a shorter operative time (73 minutes vs 82 minutes, P = .03). Magnetic resonance imaging showed a recurrent rotator cuff tear in 20% of all cases.

Conclusions: Although statistically "inconclusive" regarding non-inferiority of the CMS at 1-year follow-up, any observed differences between patients with LHB tenotomy and those with LHB tenodesis in all outcome scores were small.

Level of evidence: Level I, randomized controlled trial and treatment study.

Publication types

  • Multicenter Study
  • Randomized Controlled Trial

MeSH terms

  • Arm
  • Arthroscopy
  • Humans
  • Prospective Studies
  • Quality of Life
  • Rotator Cuff Injuries* / surgery
  • Tenodesis*
  • Tenotomy