Antithrombin activity levels for predicting long-term outcomes in the early phase of isolated traumatic brain injury

Abstract

Coagulopathy management is an important strategy for preventing secondary brain damage in patients with traumatic brain injury (TBI). Antithrombin (AT) is a natural anticoagulant that controls coagulation and inflammation pathways. However, the significance of AT activity levels for outcomes in patients with trauma remains unclear. This study aimed to investigate the relationship between AT activity levels and long-term outcomes in patients with TBI; this was a sub-analysis of a prior study that collected blood samples of trauma patients prospectively in a tertiary care center in Kawaguchi City, Japan. We included patients with isolated TBI (iTBI) aged ≥16 years admitted directly to our hospital within 1 h after injury between April 2018 and March 2021. General coagulofibrinolytic and specific molecular biomarkers, including AT, were measured at 1, 3, 6, 12, and 24 h after injury. We analyzed changes in the AT activity levels during the study period and the impact of the AT activity levels on long-term outcomes, the Glasgow Outcome Scale-Extended (GOSE), 6 months after injury. 49 patients were included in this study; 24 had good neurological outcomes (GOSE 6-8), and 25 had poor neurological outcomes (GOSE 1-5). Low AT activity levels were shown within 1 h after injury in patients in the poor GOSE group; this was associated with poor outcomes. Furthermore, AT activity levels 1 h after injury had a strong predictive value for long-term outcomes (area under the receiver operating characteristic curve of 0.871; 95% CI: 0.747-0.994). Multivariate logistic regression analysis with various biomarkers showed that AT was an independent factor of long-term outcome (adjusted odds ratio: 0.873; 95% CI: 0.765-0.996; p=0.043). Another multivariate analysis with severity scores showed that low AT activity levels were associated with poor outcomes (adjusted odds ratio: 0.909; 95% CI: 0.822-1.010; p=0.063). We demonstrated that the AT activity level soon after injury could be a predictor of long-term neurological prognosis in patients with iTBI.

Keywords: antithrombin; disseminated intravascular coagulation; long-term outcome; trauma-induced coagulopathy; traumatic brain injury.

Figure 1

ROC curve predicting long-term neurological outcome. Red line: multivariate logistic regression model with age, GCS, ISS, and antithrombin 1 h after injury Blue line: univariate of antithrombin 1h after injury. ROC, receiver operating characteristic; AUC, area under the ROC curve; GCS, Glasgow Coma Scale; ISS, Injury Severity Score.

Figure 2

Changes in AT activity levels of each GOSE group after injury. Significant differences between good (white box) and poor (gray box) GOSE groups at all the time points are shown in AT activity levels. Neither group showed a significant change in AT activity levels after injury (good GOSE group, Friedman test, χ² = 3.59, df = 4, p = 0.464; poor GOSE group, Friedman test, χ² = 2.83, df = 4, p = 0.587). Horizontal bars, boxes, whiskers, and dots represent the median, interquartile range, a distance of 1.5 times the interquartile range, and outliers, respectively. AT, antithrombin; GOSE, Glasgow Outcome Scale-Extended.