Abstract
BACKGROUND: Outcome reporting for extensor tendon surgery is fragmented. Legacy systems-Miller for repairs and Bincaz for transfers-use different success definitions, limiting comparison. Building on the composite framework proposed by Jerome et al. (wrist arc, forearm rotation, MCP extension/flexion, first-web opening, grip/pinch; all normalized to the contralateral side), we adopted this construct as the EXTensor Repair & Transfer Outcome Score (EXTROS) and validated its measurement properties across repairs and transfers. METHODS: EXTROS summarizes domain percentages into an overall grade: Good (>75%), Fair (50-75%), Poor (<50%). We evaluated 80 patients at ≥4 months (mean 6): 40 acute repairs (zones III-VII) and 40 tendon transfers (radial palsy or rheumatoid attrition). Two blinded raters scored EXTROS. We calculated Cohen's κ and ICC for reliability, Cronbach's α for internal consistency, correlations with DASH, ROM%, grip%, and pulp-to-palm for construct validity, and agreement with Miller (repairs) and Bincaz (transfers) for criterion validity. RESULTS: Mean age 36.8 ± 11.5 years; 70% male. EXTROS graded 85% as Good. Inter-rater agreement 90%; κ = 0.85 (95% CI 0.78-0.92); continuous score ICC = 0.95. Internal consistency α = 0.88. Construct validity was strong: EXTROS correlated with DASH (r = -0.72), ROM% (r = +0.80), grip% (r = +0.77), and pulp-to-palm (ρ = -0.65) (all p < 0.001). Criterion validity was substantial: Bincaz vs EXTROS 95% agreement (κ = 0.78) and Miller vs EXTROS 90% (κ = 0.82). CONCLUSIONS: EXTROS is reliable, valid, and straightforward. As a unified, percentage-based Good/Fair/Poor grade, it enables consistent reporting across repairs and transfers, supports clinical decisions, and provides a pragmatic standard for future studies of extensor tendon injuries. LEVEL OF EVIDENCE: Level IV (observational validation/case series).