RESUMO
Objective:To investigate the predictive value of mechanism Glasgow age blood pressure score (MGAPS), revised trauma score (RTS) and modified rapid emergency medicine score (mREMS) in predicting the mortality risk of patients with acute traumatic brain injury (TBI) within 24 hours.Methods:A case control study was performed for clinical data of 1 156 patients with acute TBI admitted to Affiliated Hospital of Nantong Hospital from January to December of 2020, including 745 males and 411 females; aged 18-100 years [(59.9±15.1)years]. Glasgow coma score (GCS) was 3-15 points [15(9, 15)points]. The patients were divided into death group ( n=87) and survival group ( n=1 069) according to death or not within 24 hours. Vital signs, general data, MGAPS, RTS and mREMS were documented at emergency visit. Differences in the specific scores and severity levels of the patients using the three scoring systems were compared between the two groups. Receiver operating characteristic (ROC) curve was plotted for the three scoring systems based on the specific scores and severity levels of the patients. While the area under the curve (AUC), sensitivity, specificity, optimal threshold and Youden index were determined to estimate the value of the three scoring systems in predicting death risk in patients with acute TBI within 24 hours. Results:Death group showed significantly decreased scores in MGAPS [17(12, 19)points] and RTS [5.0(4.1, 6.0)points] and significantly increased score in mREMS [9(7, 12)points] when compared with survival group (all P<0.01). The proportion of moderate- and high-risk patients for MGAPS and proportion of high-risk patients for RTS and mREMS in death group were significantly higher than those in survival group (all P<0.01). As indicated by the ROC curve plotted based on the specific scores, mREMS had the maximum AUC (0.88), followed by MGAPS (0.86) and RTS (0.86); the sensitivity of mREMS, MGAPS and RTS was similar (80.5%, 86.2% and 82.8%, respectively), while mREMS showed the highest specificity (83.4%) compared to MGAPS (78.0%) and RTS (82.3%); the optimum threshold of mREMS, MGAPS and RTS, was 6 points, 6.08 points and 20 points; the Youden index of MGAPS, RTS and mREMS was 0.64, 0.64 and 0.65. As indicated by the ROC curve plotted based on the injury severity, MGAPS had the highest AUC (0.84), followed by RTS (0.70) and mREMS (0.59); MGAPS also had the highest sensitivity (92.0%), higher than RTS (47.1%) and RTS (18.4%); when mREMS showed the highest specificity(98.8%) compared to RTS (93.7%) and MGAPS (68.8%); the optimal threshold of MGAPS, RTS and mREMS was 22 points, 4 points and 13 points; the Youden index of MGAPS, RTS and mREMS was 0.61, 0.41 and 0.17. Conclusions:MGAPS, RTS and mREMS can be predictive in assessing the mortality risk of patients with acute TBI within 24 hours. mREMS has the highest prediction value, with an optimal threshold of 6 points when the risk assessment is made in accordance with specific scores of the patients. MGAPS has the highest prediction value when the risk assessment is assessed by the injury severity.