Summary

Bilateral CT imaging and advanced imaging analysis of the glenoid in heavy weightlifting athletes quantified the distribution of subchondral bone mineralization and showed the highest Hounsfield concentration in the central and superior glenoid regions.

Abstract

Background

Excessive glenohumeral loads can induce stress and may lead to injury and progressive degeneration. Although standard radiographic imaging including X-rays and CT provides insight into advanced stages, evaluating early changes at the joint surface and subchondral bone is limited. Distribution of mineralization of the subchondral bone plate (DMSB) can measure stress modifications across the glenoid. The purpose of this study is to determine which region of interest is most impacted in weightlifting athletes.

Methods

This prospective study included athletes between 18 and 50 with normal glenohumeral radiographs and at least two years of bodybuilding, powerlifting, or heavy weightlifting. Training type, powerlifting total, and Wilks classification were collected. Glenoid retroversion and Walch stage were assessed following bilateral CT imaging. Image acquisition was standardized and imported into an imaging analysis software (Amira, Thermofisher Scientific Waltham, MA) for 3D reconstruction and glenoid segmentation. DMSB was quantified using Hounsfield units (HU) and assessed at the subchondral bone plate (L1), subchondral bone trabeculate (L2), and trabecular bone (L3). 75 HU measurements were taken per glenoid (L1-L3). Regions of interest (ROI) were defined as 6 surface areas (full glenoid, superior, inferior, anterior, posterior, and central) and 6 rims (superior, inferior, anterior, posterior, antero-inferior, and postero-inferior) and assessed using L1. The mean (± standard deviation) and median (interquartile range, IQR) were used for quantitative analysis based on data distribution. Chi-squared and Kruskal-Wallis analyses were used to determine significance.

Results

14 athletes (10 males, 4 females) with a mean age of 31.9 ± 6.4 and an average BMI of 26.9 ± 2.0 were enrolled (4 bodybuilders, 2 powerlifters, and 8 heavy weightlifters). The mean powerlifting total was 1011 lbs ± 217.0. Wilks classification categorized 50% of athletes as advanced and 50% intermediate. The glenoid Walch classification included 67% A1, 29% B1, and 4% B2. Mean retroversion was 6.9 ± 3.8, with 46% of shoulders below and 54% above the average. The mean HU of the full glenoid were highest at L1, 473.6 [391.9, 558.3], and significantly declined to 261.8 [208.3, 325.0] and 160.4 [122.6, 198.1] at L2 and L3 respectively (p < 0.001). This observation was consistent for all ROIs. HU were highest in the central glenoid, 490.3 [411.9,597.4] followed by the superior, 485.5 [385.3,573.2], anterior 476.4 [403.8,568.4], posterior 474.9 [379.4, 555.9], and inferior 452.4 [378.5, 535.2]. In the rims, HU were highest in the superior rim, followed by the anterior, posterior, postero-inferior, antero-inferior, and inferior. Individuals aged 32 and above had significantly higher HU in the full glenoid (p < 0.001) and 6 ROI (5 surface areas, 1 rim): superior, inferior, central, anterior, posterior, and posterior rim (p < 0.03). L1 HU measurements were not impacted by retroversion, side of dominance, and Wilks classification in any ROI.

Conclusion

Athletes with significant exposure to mechanical stress in the glenohumeral joint showed consistently high distribution of mineralization of the subchondral bone plate in the central and superior glenoid. Future comparison in a larger dataset will provide insight into the development of shoulder arthritis due to heavy weightlifting.