Abstract
Objectives:
Advanced fatty infiltration is correlated with poor outcomes after rotator cuff repair, and high-grade fatty infiltration is considered a contraindication for repair. The influence of lower levels of fatty infiltration on outcomes after rotator cuff repair remains unclear. Quantitative magnetic resonance (MR) imaging sequences, specifically IDEAL imaging (iterative decomposition of water and fat with echo asymmetry and least-squares estimation), has been recently applied to measuring fatty infiltration of the rotator cuff muscles. Our purpose was to evaluate the relationship between rotator cuff intramuscular fat and patient-reported outcome measures after rotator cuff repair. We hypothesized that higher pre-operative fat content would be negatively correlated with post-operative outcomes.
Methods:
We retrospectively identified patients who underwent arthroscopic rotator cuff repair with pre-operative MRI scan with sagittal-oblique IDEAL imaging. All procedures were approved by our Institutional Review Board. Pre-operative tear size, tendon involvement, and tendon retraction were measured by a musculoskeletal radiologist. Image segmentation was performed manually on four consecutive slices with perimuscular fat excluded. Patients completed the Patient-Reported Outcomes Measurement Information System (PROMIS) Upper Extremity computer adapted survey at a minimum of two years after repair. Correlations between intramuscular fat measurements and PROMIS scores were determined with Spearman’s rank correlation coefficient. Patients were grouped by PROMIS scores above and below 50, as 50 represents population mean. Mann-Whitney U tests were used to compare fat fractions between patients with high PROMIS scores (at or above 50) or low PROMIS scores (less than 50). Multivariate linear regression was performed with PROMIS score as the dependent variable, and individual muscle fat fractions, age, BMI, sex, and total tear size as independent predictors. Significance was defined as p<0.05.
Results:
A total of 80 patients were included (Table 1). Mean follow-up was 42.5 ±10.7 months. Post-operative PROMIS scores were significantly correlated with the infraspinatus fat fraction (rho = -0.25, p = 0.02) and subscapularis fat fraction (rho = -0.29, p = 0.009). The infraspinatus fat fraction for patients with a low PROMIS score (N=31) was significantly higher relative to those with a PROMIS score above 50 (N=49) (7.2±4.9% vs. 5.2 ±3.0%; p=0.046) (Figure 1). The subscapularis fat fraction was significantly higher for patients with a low PROMIS score relative to those with a PROMIS score above 50 (10.4 ±5.1% vs. 8.2 ±5.0%; p=0.001). In controlling for age, BMI, sex, and total tear size, multivariate regression modeling identified infraspinatus fat fraction (beta = -0.68, p = 0.029) as the only significant independent predictor of post-operative PROMIS score.
Conclusion:
We observed significant relationships between infraspinatus and subscapularis muscle quality and post-operative patient-reported outcomes after rotator cuff repair. Infraspinatus fat fraction was the only significant predictor when accounting for demographics and rotator cuff tear size. Importantly, these patients were selected for rotator cuff repair and therefore excluded patients with advanced fatty infiltration. Even in patients with lower degrees of muscle degeneration, small differences in muscle quality may impact outcomes after tendon repair.
Demographics
| Mean | Standard Deviation | |
|---|---|---|
| Age (years) | 62.3 | 7.5 |
| BMI (kg/m2) | 26.7 | 4.9 |
| Tear Size (mm) | 16.3 | 14.1 |
| Retraction (mm) | 15.9 | 14.5 |
| Supraspinatus Fat Fraction (%) | 8.53 | 5.34 |
| Infraspinatus Fat Fraction (%) | 5.99 | 3.96 |
| Subscapularis Fat Fraction (%) | 9.04 | 5.14 |
| Teres Minor Fat Fraction (%) | 6.55 | 6.03 |
| Post-Operative PROMIS-UE | 51.29 | 7.33 |
Box plots for the pre-operative fat fractions for suprapinsatus, infraspinatus, subscapularis, and teres minor are shown for patients with PROMIS-UE scores at two years post-operatively less than 50 (N = 31) and PROMIS-UE at two years post-operatively greater than or equal to 50 (N = 49). * indicates p = 0.046, # indicates p = 0.001 through Mann-Whitney U tests.