Quantitative analysis of level-specific fusion outcomes and scapular positioning in scapulothoracic arthrodesis

Facioscapulohumeral Muscular Dystrophy (FSHD) is among the most common hereditary muscular dystrophies, often affecting periscapular muscles and causing scapular winging, which severely impairs activities of daily living. Scapulothoracic arthrodesis (STA) aims to stabilize the scapula, improving upper extremity range of motion. Despite its long-standing use, the ideal scapular fixation position remains undetermined, and level-specific fusion outcomes have not been evaluated. This retrospective cohort study evaluates fusion rates based on early and late computed tomography (CT) images, introduces a novel fusion grading system, and quantifies three-dimensional fixation positions in bilateral STA cases.

We hypothesized that STA would achieve a high rate of solid fusion detectable on CT within 12 months. Furthermore, we expected that bilateral procedures would yield symmetric scapular orientation within individual patients, and that the proposed fusion grading system would demonstrate substantial inter-observer reliability.

CT imaging of 36 patients (56 shoulders) was reviewed using 3D Slicer software. A novel five-point Qualitative Fusion Score (QFS) was developed, categorizing fusion at each rib level as: A (clear union), B (probable union), C (probable non-union), D (clear non-union), and F (scapula fracture). For bilateral STA cases (14 patients, 28 shoulders), scapular positions were assessed three-dimensionally using anatomical landmarks: glenoid center, scapular spine midpoint, and inferior angle. Angles between the scapular plane and coronal, sagittal, and transverse planes were analyzed using paired samples t-tests. Subgroup analysis compared early (3-6 months) and late (≥12 months) CT fusion scores in 17 shoulders.

Overall fusion success (Grade A + B) was achieved in 82.3% of sites at final follow-up (mean 38.4 months), with significant variation by rib level. Grade A (solid union) rates ranged from 37.78% (7th rib) to 58.82% (2nd rib). The 7th rib demonstrated the poorest outcomes with the highest rates of definite non-union (13.33%) and fracture (8.89%), while most fractures occurred exclusively at this level. The prevalence of cable fixation varied between 78.43 and 100% across fixation levels, with high-strength polyethylene tape sutures utilized selectively in 0-21.57%. In bilateral cases, scapular positioning showed excellent symmetry with no significant differences in angles relative to coronal (p = 0.271), sagittal (p = 0.297), or transverse planes (p = 0.053), and mean vertical position difference of 12.15 mm. Early CT evaluation demonstrated strong predictive value: all Grade B sites progressed to Grade A (100%), while only 35% of Grade C sites showed improvement. The QFS showed significant improvement from early to late imaging (20.8 ± 3.3 vs 24.8 ± 4.6, p < 0.001) with moderate-to-good inter-observer reliability (ICC = 0.725).

STA achieved high overall fusion success with significant level-specific variation, demonstrating superior outcomes at upper rib levels and increased complications at the 7th rib level. Bilateral procedures yielded highly symmetric scapular positioning. Early CT evaluation at 3 months effectively predicts final fusion outcomes, particularly identifying patients at risk for poor healing. The novel QFS provides a standardized assessment tool for postoperative monitoring and clinical decision-making.

III; Retrospective Cohort Study.