Radiographic Cortical Thickness of the Humerus in Detecting Post-Stroke Regional Bone Loss

Post-stroke immobilization and reduced weight-bearing frequently lead to significant regional bone mineral density loss and asymmetry, particularly in the paretic upper extremity, which increases fracture risks. While Dual-Energy X-ray Absorptiometry (DXA) is the gold standard for evaluating systemic bone loss, it primarily focuses on axial or lower extremity sites and lacks universal accessibility. Since routine shoulder radiographs offer an opportunistic screening tool to evaluate regional bone quality without additional radiation , this study aims to compare proximal humerus cortical bone thickness between the paretic and non-paretic sides in stroke patients and assess its correlation with systemic DXA values to determine its clinical utility.

This cross-sectional, observational study involves a retrospective data analysis of patients aged 18 and older who experienced a single unilateral stroke at least 6 months prior and present with clinical hemiparesis. Eligible participants must have concurrent standard anteroposterior shoulder radiographs and DXA measurements available from their routine clinical follow-ups. Patient demographic data, stroke characteristics, Brunnstrom stages, and systemic DXA measurements (femoral neck and lumbar spine T-scores and bone mineral density values) are systematically recorded for analysis.

Cortical bone thickness measurements are performed using ImageJ software on standard radiographs at points 10 cm and 12 cm distal to the highest point of the humerus. To ensure reliability, measurements for both the paretic and non-paretic sides are conducted independently by two researchers who are completely blinded to the DXA results. Statistical analyses, including paired t-tests or Wilcoxon tests and Pearson or Spearman correlations, will be used to compare the sides and evaluate the relationship between radiographic cortical thickness and systemic bone density.