Predicting mortality and intensive care needs in geriatric trauma patients: A ROC analysis of frailty and trauma scoring systems

Abstract

Background: The global increase in the geriatric population presents significant challenges for healthcare systems. Trauma, mainly resulting from falls, constitutes a primary cause of morbidity and mortality among older people. It is imperative to evaluate mortality risk and the need for intensive care unit (ICU) admission among geriatric trauma patients to optimize clinical outcomes. This study aims to assess the efficacy of frailty and trauma scoring systems in predicting these critical outcomes. Methods: A prospective cross-sectional study was conducted at a tertiary care hospital in Türkiye from September 2023 to October 2024. The study population consisted of geriatric patients (≥ 65 years) presenting blunt trauma. Comprehensive demographic, clinical, and laboratory data were collected. Mortality risk and ICU admission requirements were assessed utilizing various scoring systems: the Frailty Index (FI), Clinical Frailty Score (CFS), Trauma-Specific Frailty Index (TSFI), Shock Index (SI), Modified Shock Index (MSI), Age Shock Index (ASI), Revised Trauma Score (RTS), Abbreviated Injury Scale (AIS), Injury Severity Score (ISS), Geriatric Trauma Outcome Score (GTOS), and Glasgow Coma Scale (GCS). A Receiver Operating Characteristic (ROC) analysis was performed to determine the predictive accuracy of these scores. Results: The cohort comprised 350 patients with a median age of 74 years. The mortality rate observed was 11%, with non-survivors had significantly higher frailty and trauma scores compared to survivors (p < 0.001). Admission to the intensive care unit (ICU) was necessary in 20% of the cases. A Receiver Operating Characteristic (ROC) analysis revealed the GTOS and RTS exhibited the highest and identical Area Under the Curve (AUC) values, both recorded at 0.846. Furthermore, the 95% confidence intervals (CI) for GTOS and RTS were 0.790–0.901 and 0.762–0.930, respectively, while the AUC for ICU admission requirements reached 0.830 (95% CI: 0.764–0.864) for RTS and 0.815 (95% CI: 0.755–0.876) for GTOS. Among the frailty scores, TSFI proved to be superior in predicting mortality, achieving an AUC of 0.805 (95% CI: 0.725–0.885), whereas the Frailty Index (FI) demonstrated the lowest predictive capacity for ICU needs, with an AUC of 0.666 (95% CI: 0.592–0.740). Conclusion: Frailty and trauma scoring systems serve as critical tools in predicting outcomes for geriatric trauma patients. The Revised Trauma Score and the Geriatric Trauma Outcome Score have demonstrated the highest reliability in forecasting both mortality and the need for ICU admissions. While the Trauma-Specific Frailty Index has exceeded other frailty indices in terms of mortality prediction, the Frailty Index exhibits limited utility for ICU evaluations. Integrating these scoring systems into clinical practice can significantly enhance the identification of high-risk patients and improve care strategies.