Assessments of Fatty Infiltration and Muscle Atrophy From a Single Magnetic Resonance Image Slice Are Not Predictive of 3-Dimensional Measurements

Abstract

Purpose: To (1) determine whether standard clinical muscle fatty infiltration and atrophy assessment techniques using a single image slice for patients with a rotator cuff tear (RCT) are correlated with 3-dimensional measures in older individuals (60+ years) and (2) to determine whether age-associated changes to muscle morphology and strength are compounded by an RCT.

Methods: Twenty older individuals were studied: 10 with an RCT of the supraspinatus (5 men and 5 women) and 10 matched controls. Clinical imaging assessments (Goutallier and Fuchs scores and cross-sectional area ratio) were performed for participants with RCTs. Three-dimensional measurements of rotator cuff muscle and fat tissues were obtained for all participants using magnetic resonance imaging (MRI). Isometric joint moment was measured at the shoulder.

Results: There were no significant associations between single-image assessments and 3-dimensional measurements of fatty infiltration for the supraspinatus and infraspinatus muscles. Compared with controls, participants with RCTs had significantly increased percentages of fatty infiltration for each rotator cuff muscle (all P ≤ .023); reduced whole muscle volume for the supraspinatus, infraspinatus, and subscapularis muscles (all P ≤ .038); and reduced fat-free muscle volume for the supraspinatus, infraspinatus, and subscapularis muscles (all P ≤ .027). Only the teres minor (P = .017) fatty infiltration volume was significantly greater for participants with RCTs. Adduction, flexion, and external rotation strength (all P ≤ .021) were significantly reduced for participants with RCTs, and muscle volume was a significant predictor of strength for all comparisons.

Conclusions: Clinical scores using a single image slice do not represent 3-dimensional muscle measurements. Efficient methods are needed to more effectively capture 3-dimensional information for clinical applications. Participants with RCTs had increased fatty infiltration percentages that were likely driven by muscle atrophy rather than increased fat volume. The significant association of muscle volume with strength production suggests that treatments to preserve muscle volume should be pursued for older patients with RCTs.

Level of evidence: Level II, diagnostic study, with development of diagnostic criteria on the basis of consecutive patients with universally applied reference gold standard.

Trial registration: ClinicalTrials.gov NCT01459536.

Conflict of interest statement

Declaration: We declare that abstracts describing small portions of this manuscript have been presented at the Gait and Clinical Movement Analysis Society Conference, the International Society of Biomechanics Conference, the American Society of Biomechanics Conference, the World Congress of Biomechanics, the Wake Forest Graduate Student and Post-doc Research Symposium, and the Virginia Tech-Wake Forest School of Biomedical Engineering and Sciences Symposium. An abstract describing a portion of this work was awarded 3rd Place (certificate and $80 award to MEV) for Outstanding Podium Presentation by a PhD Student at the 13th Annual Virginia Tech-Wake Forest School of Biomedical Engineering and Sciences Symposium. We declare that author MTF serves as a consultant for Smith and Nephew, it does not represent a conflict of interest and no financial remuneration was received related to the work described in this manuscript. Author CJT has an ownership interest in a medical device measuring tension in rotator cuff tendon repairs with research applications; participation with device development and testing is outside the scope of this manuscript and does not represent a conflict of interest.

Copyright © 2016 Arthroscopy Association of North America. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1

Study participant recruitment flow chart. The left side of the chart shows the recruitment criteria and process for participants with a rotator cuff tear, while the right side of the chart shows the recruitment criteria and process for participants in the control group.

Figure 2

Mean Goutallier score (A) and mean Fuchs score (B) versus percentage of fatty infiltration for rotator cuff muscles. The injured arm (3 right /6 left) was assessed for participants with a rotator cuff tear and the dominant (9 right/1 left) was assessed for controls. * denotes a significant linear relationship, whereby significance indicates that the Goutallier score or Fuchs score captures three-dimensional measures of fatty infiltration. The relationship between mean Goutallier score (A) and percentage of fatty infiltration was significant for subscapularis and teres minor (supraspinatus, p=0.116; infraspinatus, p=0.094; subscapularis, p=0.008; teres minor, p=0.009). The relationship between Fuchs score (B) and percentage of fatty infiltration was significant for subscapularis and teres minor (supraspinatus, p=0.065; infraspinatus, p=0.086; subscapularis, p=0.027; teres minor, p=0.021).

Figure 3

Three-dimensional muscle volume measurements versus cross-sectional area ratio (muscle cross-sectional area/supraspinatus fossa area) for supraspinatus (p=0.315), infraspinatus (p=0.240), subscapularis (p=0.637), and teres minor (p=0.056) muscles. The injured arm (3 right /6 left) was assessed for participants with a rotator cuff tear and the dominant arm was assessed for controls. The presence of a significant correlation would have indicated that the cross-sectional area ratio effectively estimates three-dimensional muscle volume.

Figure 4

Mean±SD for quantitative measurements of muscle and fatty infiltration volume for rotator cuff muscles for older adults with a rotator cuff tear (gray bars) and healthy controls (white bars). For participants with a rotator cuff tear, the injured arm (3 right /7 left) was assessed and for controls, the dominant arm (9 right/1 left) was assessed. (A) There was a significantly greater percentage of fatty infiltration for all muscles (supraspinatus, p=0.002; infraspinatus, p=0.023; teres minor, p=0.022; subscapularis, p=0.017) for the rotator cuff tear group. (B) Only teres minor (p=0.017) had a significantly greater volume of fatty infiltration for the rotator cuff tear subjects. (C) The rotator cuff tear group had significantly reduced whole volume measurements for supraspinatus (p=0.003), infraspinatus (p=0.038), and subscapularis (p=0.015) muscles. (D) Fat-free muscle volume was significantly reduced for the rotator cuff tear group for supraspinatus (p=0.002), infraspinatus (p=0.027), and subscapularis (p=0.011) muscles.

Figure 5

Mean±SD isometric joint moment for the rotator cuff tear group (gray bars) and healthy controls (white bars). The injured arm (3 right/7 left) was evaluated for participants with a rotator cuff tear and the dominant arm (9 right/1 left) was assessed for controls. One participant could not perform abduction and three participants could not perform flexion due to pain; these trials were not included in the statistical analyses. Isometric joint moment was significantly reduced for adduction (p=0.021), flexion (p=0.019), and external rotation (p=0.003).

Figure 6

Isometric joint moment versus whole muscle volume (A) and fat-free muscle volume (B) for subjects with a rotator cuff tear (hollow markers) and healthy controls (filled markers). The injured arm (3 right/7 left) was assessed for participants with a rotator cuff tear and the dominant arm (9 right/1 left) was evaluated for controls. * denotes a significant linear relationship, which indicates that strength production is dependent on three-dimensional muscle volume measures. There are significant relationships for (A) whole muscle volume and the joint moment of each muscle’s primary action (supraspinatus, p=0.005; infraspinatus, p