Bismil Ali Ali1, Rolf Lefering2, Mariano Fortún Moral3, Tomás Belzunegui Otano4. 1. Servicio de Urgencias del Complejo Hospitalario de Navarra, Servicio Navarro de Salud-Osasunbidea, Pamplona, España. 2. Institute for Research in Operative Medicine (IFOM), Universidad de Witten/Herdecke, Alemania. 3. Médico de Servicio de Urgencias/ UVI-Móvil. Hospital de Tudela. Servicio de Navarro de Salud- Osasunbidea, Tudela, España. 4. Departamento de Salud, Universidad Pública de Navarra, Pamplona, España.
Abstract
OBJECTIVES: To validate the Mortality Prediction Model of Navarre (MPMN) to predict death after severe trauma and compare it to the Revised Injury Severity Classification Score II (RISCII). MATERIAL AND METHODS: Retrospective analysis of a cohort of severe trauma patients (New Injury Severity Score >15) who were attended by emergency services in the Spanish autonomous community of Navarre between 2013 and 2015. The outcome variable was 30-day all-cause mortality. Risk was calculated with the MPMN and the RISCII. The performance of each model was assessed with the area under the receiver operating characteristic (ROC) curve and precision with respect to observed mortality. Calibration was assessed with the Hosmer-Lemeshow test. RESULTS: We included 516 patients. The mean (SD) age was 56 (23) years, and 363 (70%) were males. Ninety patients (17.4%) died within 30 days. The 30-day mortality rates predicted by the MPMN and RISCII were 16.4% and 15.4%, respectively. The areas under the ROC curves were 0.925 (95% CI, 0.902-0.952) for the MPMN and 0.941 (95% CI, 0.921-0.962) for the RISCII (P=0.269, DeLong test). Calibration statistics were 13.6 (P=.09) for the MPMN and 8.9 (P=.35) for the RISCII. CONCLUSION: Both the MPMN and the RISCII show good ability to discriminate risk and predict 30-day all-cause mortality in severe trauma patients.
OBJECTIVES: To validate the Mortality Prediction Model of Navarre (MPMN) to predict death after severe trauma and compare it to the Revised Injury Severity Classification Score II (RISCII). MATERIAL AND METHODS: Retrospective analysis of a cohort of severe traumapatients (New Injury Severity Score >15) who were attended by emergency services in the Spanish autonomous community of Navarre between 2013 and 2015. The outcome variable was 30-day all-cause mortality. Risk was calculated with the MPMN and the RISCII. The performance of each model was assessed with the area under the receiver operating characteristic (ROC) curve and precision with respect to observed mortality. Calibration was assessed with the Hosmer-Lemeshow test. RESULTS: We included 516 patients. The mean (SD) age was 56 (23) years, and 363 (70%) were males. Ninety patients (17.4%) died within 30 days. The 30-day mortality rates predicted by the MPMN and RISCII were 16.4% and 15.4%, respectively. The areas under the ROC curves were 0.925 (95% CI, 0.902-0.952) for the MPMN and 0.941 (95% CI, 0.921-0.962) for the RISCII (P=0.269, DeLong test). Calibration statistics were 13.6 (P=.09) for the MPMN and 8.9 (P=.35) for the RISCII. CONCLUSION: Both the MPMN and the RISCII show good ability to discriminate risk and predict 30-day all-cause mortality in severe traumapatients.
Entities:
Keywords:
Calidad asistencial al trauma; Modelos predictivos; Mortalidad; Mortality; Quality of trauma care; Risk models; Trauma
Authors: Sascha Halvachizadeh; P J Störmann; Orkun Özkurtul; Till Berk; Michel Teuben; Kai Sprengel; Hans-Christoph Pape; Rolf Lefering; Kai Oliver Jensen Journal: BMJ Open Date: 2022-04-13 Impact factor: 3.006