Single Institution Validation Study of the Mangled Digit Severity Score for Predicting Salvageability of Severe Digital Injuries.

PURPOSE: A single institution external validation study was conducted to evaluate the recently created Mangled Digit Severity Score (MDSS). We hypothesized that the MDSS would demonstrate good discriminatory ability in predicting amputation versus successful salvage in an independent external cohort of patients with mangled digital injuries.

METHODS: A retrospective review was conducted of all patients diagnosed with a mangled digit who presented to a level 1 trauma center and subsequently underwent revision amputation or digit salvage procedures from 2019 to 2025. MDSS scores were calculated as published. Performance statistics for predicting amputation versus salvage were first evaluated using the original published MDSS cutoffs. Institution specific cutoff optimization was then performed using receiver operating characteristic analysis to determine the refined lower cutoff (favoring salvage) and upper cutoff (favoring revision amputation).

RESULTS: Forty patients with mangled digital injuries were identified, involving a total of 85 affected digits. Using the original published cutoffs, the MDSS achieved an overall accuracy of 76.2% (95% CI, 65.6% to 85.9%), sensitivity of 74.4% (95% CI: 55.2% to 90.6%), specificity of 77.2% (95% CI, 61.8% to 90.0%), and area under the curve of 0.76 (95% CI, 0.70-0.89). Following institution specific receiver operating characteristic-based optimization of cutoffs (score <23 indicates salvage, score ≥23 and <39 indicates indeterminant, score ≥39 indicates amputate), performance improved, with accuracy increasing to 82.5% (95% CI, 71.6% to 92.4%), sensitivity to 87.5% (95% CI, 75.0% to 100%), specificity to 78.9% (95% CI, 61.8% to 93.8%), and area under the curve of 0.84 (95% CI, 0.73-0.93).

CONCLUSIONS: This external validation supports the generalizability of the MDSS in our independent cohort, where it demonstrated reliable performance using the original published cutoffs. Institution specific optimization resulted in modest changes to threshold values while maintaining overall performance characteristics. These findings support the generalizability of the MDSS as a decision-support tool for predicting digit survival following severe digital trauma.

TYPE OF STUDY/LEVEL OF EVIDENCE: Prognostic IIb.