<- Back to digest

Arthroscopy - 2026-09-20 - Journal Article

Postoperative Supraspinatus Muscle Volume Restoration Is Associated With Tendon Healing and Strength Recovery After Rotator Cuff Repair: Deep-Learning-Based Volumetric Analysis of the Entire Muscle With Clinical Correlation.

Kim YT, Yoo HJ, Oh SY, Kim SH

retrospective cohortLOE IIIn = 102Mean 38.9 months (range 24–77 months)

Topics

sportsshoulder elbow
PMID: 42763871DOI: 10.1002/arj.70531View on PubMed ->

Key Takeaway

Each 10% increase in postoperative supraspinatus muscle volume independently reduces retear odds by 51% (OR 0.49, 95% CI 0.30–0.80) and correlates with superior Constant scores and abduction strength at mean 38.9 months after rotator cuff repair.

Summary Depth

Choose how much analysis to show on this article page.

Summary

This retrospective study used deep-learning-based MRI segmentation to quantify supraspinatus volume (SPV) change from preoperative to 1-year postoperative in 102 rotator cuff repair patients and correlated volume change with healing and functional outcomes at minimum 2-year follow-up. Patients with SPV increase (n=70) had a significantly lower retear rate than those with SPV decrease (14.3% vs 37.5%, P=.008) and greater postoperative Constant scores, abduction strength, and IR strength. On multivariate analysis, every 10% SPV increase independently predicted higher Constant score, greater abduction and IR strength, and lower retear odds (OR 0.49, P=.004).

Key Limitation

The requirement for full MRI coverage of the supraspinatus to its medial origin excluded an unspecified number of patients, potentially selecting for less atrophic muscles and limiting generalizability to the highest-risk retear population.

Original Abstract

PURPOSE

To quantify supraspinatus muscle volume (SPV) change after rotator cuff repair using deep-learning-based segmentation of full-coverage magnetic resonance imaging and evaluate its associations with clinical, radiologic parameters, and retears.

METHODS

This retrospective study analyzed rotator cuff repair patients with a minimum 2-year follow-up, with preoperative and 1-year postoperative magnetic resonance imaging both achieving full coverage of the supraspinatus to its medial origin. A deep-learning model was used for automatic supraspinatus segmentation and SPV calculation. Intergroup analyses "increase" versus "decrease," "healed" versus "retear," and "male" versus "female" were performed depending on SPV change, tendon healing, and sex, respectively. SPV change was further correlated with radiologic and clinical parameters.

RESULTS

A total of 102 patients were analyzed. The mean follow-up was 38.9 ± 13.4 months (range, 24-77). In the SPV increased group (n = 70), a lower retear rate (14.3% vs 37.5%, P = .008), a better postoperative Constant score (P = .021), greater scapular plane abduction (P = .011), and internal rotation (IR) strength (P = .034) were shown compared with the SPV decreased (n = 32). Percentage increase in SPV was significantly correlated with the postoperative increase in the Constant score (r = 0.272, P = .008), scapular plane abduction strength (r = 0.262, P = .014), and IR strength (r = 0.304, P = .004). Mean SPV increase was shown in the healed group (n = 80), contrary to the retear group (n = 22) (107.6 ± 14.7% vs 94.1 ± 19.8%, P = .006), with significantly better clinical scores and strength, except in IR. Male patients showed significantly greater pre- and postoperative strength and SPV, while SPV change was similar between sexes. A greater postoperative SPV increase (per 10%) was independently associated with a higher postoperative Constant score (P = .049), scapular plane abduction strength (P = .018), IR strength (P = .008), and a lower odds of retear (odds ratio, 0.49; 95% confidence interval, 0.30-0.80; P = .004).

CONCLUSIONS

Postoperative volume increase in the supraspinatus muscle after rotator cuff repair was strongly associated with tendon healing and showed significant correlation with the increase in postoperative strength and the Constant score.

LEVEL OF EVIDENCE

Level III, retrospective comparative case series.