Performance of five trauma scoring systems in predicting in-hospital outcomes in geriatric trauma: a retrospective cohort study

Résumé fourni par la source

BACKGROUND: Predicting unfavorable outcomes in geriatric trauma patients (GTPs) remains a significant challenge, as trauma remains a leading cause of mortality in this population. This study aimed to compare the performance of five trauma scoring systems, including three age-specific models, in predicting in-hospital mortality and the need for mechanical ventilation (MV) among Iranian GTPs. METHODS: This retrospective cohort study included trauma patients ≥ 65 years from July 29, 2016 to September 19, 2024. The GERtality score, Geriatric Trauma Outcome Score (GTOS), GTOS II, Trauma Injury Severity Score (TRISS), and adjusted TRISS (aTRISS) were calculated for each patient. We evaluated the statistical performance of these scores using the concordance statistic (C-index) and assessed calibration by comparing observed versus predicted mortality risks through calibration plots. Clinical usefulness was evaluated with decision curve analysis (DCA) and net benefit (NB). RESULTS: Of 1,081 GTPs, a total of 86 (7.9%) required MV, and 93 individuals (8.6%) were deceased during hospitalization. The GERtality score (C-index = 0.89, 95% CI: 0.85 to 0.93) and GTOS (C-index = 0.86, 95% CI: 0.84 to 0.93) demonstrated the highest predictive value for in-hospital mortality. For MV, GERtality and GTOS exhibited the highest predictive performance (C-index = 0.82). The calibration slopes for all scoring systems were close to 1.0 (range: 0.97 to 0.99), with confidence intervals including 1.0, indicating good agreement between predicted and observed mortality risks. Based on decision curve analysis, the GERtality score demonstrated the highest NB across a broad range of threshold probabilities (approximately 0.05 to 0.75), supporting its superior clinical utility in predicting both mortality and MV. CONCLUSIONS: Although all scoring systems demonstrated acceptable performance, the GERtality score exhibited the highest predictive value among the five systems for both in-hospital mortality and the need for MV in GTPs. TRIAL REGISTRATION NUMBER: Not applicable.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.