Diagnostic Performance of Ultrasonographic Median Nerve Cross-Sectional Area for Carpal Tunnel Syndrome in Wheelchair Users With Spinal Cord Injury

Abstract

OBJECTIVE: To evaluate the diagnostic performance of ultrasonographic median nerve cross-sectional area (CSA-M) for carpal tunnel syndrome (CTS) in wheelchair users with spinal cord injury (SCI), and to identify clinically relevant diagnostic cutoff values. DESIGN: Prospective, cross-sectional diagnostic accuracy study using registry data from Thai wheelchair users with SCI receiving outpatient rehabilitation care at a tertiary university hospital. Ultrasonographic CSA-M (index test) and electrodiagnostic studies (reference standard) were performed by blinded examiners. RESULTS: Sixty-seven participants were included, of whom 25(37.3%) had CTS. Among 126 wrists analyzed, 42(33.3%) had electrodiagnostically confirmed CTS. Mean CSA-M was significantly larger in CTS compared with non-CTS wrists (14.4±4.4 vs. 10.7±3.3 mm²; P<0.001). The AuROC was 0.76(95% CI, 0.67-0.84). At a 10 mm² cutoff sensitivity was 90.5%, and specificity was 45.2%. At 15 mm², sensitivity was 33.3%, and specificity was 91.7%. Age, sex, and body mass index did not influence diagnostic performance. The wrist-to-forearm ratio showed comparable, but not superior, diagnostic performance to CSA-M. CONCLUSIONS: Ultrasonographic CSA-M demonstrates moderate diagnostic accuracy for CTS in wheelchair users with SCI. Lower cutoffs favor screening sensitivity, whereas higher cutoffs improve diagnostic specificity.