Prothrombin time predicting time-dependent and risk-stratified mortality in polytrauma patients.

Abstract

Background: Polytrauma is associated with a high mortality rate and often accompanied by coagulopathy. Prothrombin time (PT) is a prognostic factor for mortality in polytrauma patients. The aim was to analyze the time- and severity-dependent role of PT in polytrauma patients related to mortality.

Methods: Patients (≥ 16 years) with an Injury Severity Score ≥ 16 were retrospectively included, yielding 2890 cases after exclusion criteria. PT was measured at admission and 1, 2, 3, 4, 6, 8, 12, 24, and 48 h thereafter, reported as percentage activity of the reference reagence [%]. According to survival status, two groups were formed and compared. Binary logistic regression was used to test PT as an independent predictor for mortality. A closest top-left threshold method served for calculating threshold values between the survivor and non-survivor group. Patients were divided into subgroups according to PT levels and mortality was assessed for each subgroup at each time point.

Results: PT values in the non-survivor group were lower throughout the measuring period (p < 0.05). PT threshold values declined from admission until 2 h afterwards, reaching less than 50%. Already a slightly compromised PT (≤ 70%) represented a significant factor (p < 0.05) for mortality at early and late time points, associated with a rate of more than 20%. In extremis, PT values of ≤ 25% were related to a mortality rate of more than 50% up to four hours after admission.

Conclusion: There are early and significant differences in mortality according to PT values in polytrauma patients (despite resuscitation measures), urging for a fast correction of PT. Time-dependent and stratified referencing may help clinicians estimate the mortality risk and decide upon the extent of surgical care.