Background: Despite routine prophylaxis, thromboembolic events (TEs) in surgical patients remain a substantial problem. Furthermore, the timing and incidence of hypercoagulability, which predisposes to these events is unknown, with institutional screening programs serving primarily to establish a diagnosis after an event has occurred. Emerging evidence suggests that point of care (POC) rapid thrombelastography (r-TEG) provides a real-time analysis of comprehensive thrombostatic function, which represents an analysis of both enzymatic and platelet components of thrombus formation. We hypothesized that r-TEG can be used as a screening tool to identify hypercoagulable states in surgical patients and would predict subsequent thromboembolic events.
Methods: Rapid thrombelastography r-TEG analyses were performed on 152 critically ill patients in the surgical intensive care unit (ICU) during 7 months. Hypercoagulability was defined as clot strength (G)>12.4 dynes/cm(2). Variables of interest for identifying hypercoagulability and thromboembolic events included sex, age, operating hospital service, specific injury patterns, injury severity score (ISS), transfusion within first 24 h, ICU duration of stay, ventilator days, hospital admission days, and thromboprophylaxis. Comparisons between the hypercoagulable and normal groups or between the groups with and without thromboembolic events were performed using Chi-square tests or the Fisher exact test for categorical variables and independent sample t tests or Wilcoxon rank sum tests for continuous variables. Multivariate logistic regression analysis (LR) was performed to identify independent predictors of thromboembolic events. A receiver operating characteristic curve was used to measure the performance of G for predicting the occurrence of a TE event. All tests were 2-sided with significance of P < .05.
Results: In all, 86 patients (67%) were hypercoagulable by r-TEG. More than 85% of patients in the hypercoagulable group and 79% in the normal group received thromboprophylaxis during the study period. The differences between hypercoagulable and normal groups by bivariate analysis included high-risk injuries (52% vs 35%; P = .03), spinal cord injury (27% vs 12%; P = .03), median ICU duration of stay (13 vs 7 days; P < .001), median ventilator days (6 vs 2; P < .001), and median hospital duration of stay (20 vs 13 days; P < .001). A total of 16 patients (19%) of the hypercoagulable group suffered a thromboembolic event, and 10 hypercoagulable patients (12%) had thromboembolic events predicted by prior r-TEG hypercoagulability. No patients with normal coagulability by r-TEG had an event (P < .001). LR analysis showed that the strongest predictor of TE after controlling for the presence of thromboprophylaxis was elevated G value (odds ratio: 1.25, 95% confidence interval [CI]: 1.12-1.39). For every 1 dyne/cm(2) increase in G, the odds of a TE increased by 25%.
Conclusion: These results indicate that the presence of hypercoagulability identified by r-TEG is predictive of thromboembolic events in surgical patients. Subsequent study is necessary to define optimal prophylactic treatment strategies for patients with r-TEG proven hypercoagulability.