Frontiers in neurology · 2026
Cardiac troponin T and 28-day mortality in traumatic brain injury: a retrospective cohort study and machine learning prediction model.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Zhang Liwen, Hu Wenwen, He Wenjie, Huang Feiyu, Zhang Jing
Аннотация
BACKGROUND: Traumatic brain injury (TBI) frequently induces cardiac dysfunction via the brain-heart axis, yet the prognostic value of cardiac troponin T (cTnT) in TBI remains incompletely characterized. This study investigated thye association between cTnT levels and 28-day all-cause mortality in TBI patients and developed a machine learning-based prediction model.
METHODS: This retrospective cohort study included 837 adult TBI patients from the MIMIC-III/IV databases (development cohort) and 121 patients from a single Chinese center (external validation cohort). The first cTnT measurement within 24 h of ICU admission was analyzed. Restricted cubic spline (RCS) regression was used to explore nonlinear relationships. cTnT was categorized into quartiles (Q1 ≤ 0.03, Q2 0.03-0.06, Q3 0.06-0.17, Q4 > 0.17 ng/mL). A two-step Boruta plus LASSO feature selection identified 13 robust predictors. Five machine learning algorithms (Random Forest, BART, GBM, CTREE, SVM) were trained. The primary outcome was 28-day all-cause mortality.
RESULTS: Among 837 patients, 217 (25.9%) died within 28 days. RCS revealed a significant nonlinear association (P for nonlinearity < 0.001). In multivariable Cox regression adjusted for the 12 LASSO-selected covariates (excluding cTnT quartile itself), patients in Q3 and Q4 had nearly twofold higher mortality risks compared with Q1 (HR = 2.01, 95% CI: 1.37-2.95, p < 0.001; and HR = 1.96, 95% CI: 1.31-2.93, p = 0.001, respectively). Random Forest achieved the best discriminative performance, with an internal AUC of 0.783 and an external validation AUC of 0.706 (95% CI: 0.584-0.828). The model also demonstrated good calibration (Brier score = 0.155) and favorable clinical utility across clinically relevant threshold probabilities as shown by decision curve analysis.
CONCLUSION: Elevated cTnT is independently associated with 28-day mortality in TBI patients, with a potential risk threshold of ≥0.06 ng/mL. The Random Forest model integrating cTnT with clinical variables provides moderate mortality prediction. cTnT may serve as an integrative marker of systemic stress and multi-organ dysfunction in TBI.
METHODS: This retrospective cohort study included 837 adult TBI patients from the MIMIC-III/IV databases (development cohort) and 121 patients from a single Chinese center (external validation cohort). The first cTnT measurement within 24 h of ICU admission was analyzed. Restricted cubic spline (RCS) regression was used to explore nonlinear relationships. cTnT was categorized into quartiles (Q1 ≤ 0.03, Q2 0.03-0.06, Q3 0.06-0.17, Q4 > 0.17 ng/mL). A two-step Boruta plus LASSO feature selection identified 13 robust predictors. Five machine learning algorithms (Random Forest, BART, GBM, CTREE, SVM) were trained. The primary outcome was 28-day all-cause mortality.
RESULTS: Among 837 patients, 217 (25.9%) died within 28 days. RCS revealed a significant nonlinear association (P for nonlinearity < 0.001). In multivariable Cox regression adjusted for the 12 LASSO-selected covariates (excluding cTnT quartile itself), patients in Q3 and Q4 had nearly twofold higher mortality risks compared with Q1 (HR = 2.01, 95% CI: 1.37-2.95, p < 0.001; and HR = 1.96, 95% CI: 1.31-2.93, p = 0.001, respectively). Random Forest achieved the best discriminative performance, with an internal AUC of 0.783 and an external validation AUC of 0.706 (95% CI: 0.584-0.828). The model also demonstrated good calibration (Brier score = 0.155) and favorable clinical utility across clinically relevant threshold probabilities as shown by decision curve analysis.
CONCLUSION: Elevated cTnT is independently associated with 28-day mortality in TBI patients, with a potential risk threshold of ≥0.06 ng/mL. The Random Forest model integrating cTnT with clinical variables provides moderate mortality prediction. cTnT may serve as an integrative marker of systemic stress and multi-organ dysfunction in TBI.