Postinjury platelet aggregation and venous thromboembolism

Abstract

Background: Posttraumatic venous thromboembolism (VTE) remains prevalent in severely injured patients despite chemoprophylaxis. Importantly, although platelets are central to thrombosis, they are not routinely targeted in prevention of posttraumatic VTE. Furthermore, platelets from injured patients show ex vivo evidence of increased activation yet impaired aggregation, consistent with functional exhaustion. However, the relationship of this platelet functional phenotype with development of posttraumatic VTE is unknown. We hypothesized that, following injury, impaired ex vivo platelet aggregation (PA) is associated with the development of posttraumatic VTE.

Methods: We performed a secondary analysis of 133 severely injured patients from a prospective observational study investigating coagulation and inflammation (2011-2019). Platelet aggregation in response to stimulation with adenosine diphosphate (ADP), collagen, and thrombin was measured at presentation (preresuscitation) and 24 hours (postresuscitation). Viscoelastic clot strength and lysis were measured in parallel by thromboelastography. Multivariable regression examined relationships between PA at presentation, 24 hours, and the change (δ) in PA between presentation and 24 hours with development of VTE.

Results: The 133 patients were severely injured (median Injury Severity Score, 25), and 14% developed VTE (all >48 hours after admission). At presentation, platelet count and PA were not significantly different between those with and without incident VTE. However, at 24 hours, those who subsequently developed VTE had significantly lower platelet counts (126 × 10 9 /L vs. 164 × 10 9 /L, p = 0.01) and lower PA in response to ADP ( p < 0.05), collagen ( p < 0.05), and thrombin ( p = 0.06). Importantly, the magnitude of decrease in PA (δ) from presentation to 24 hours was independently associated with development of VTE (adjusted odds ratios per 10 aggregation unit decrease: δ-ADP, 1.31 [ p = 0.03]; δ-collagen, 1.36 [ p = 0.01]; δ-thrombin, 1.41 [ p < 0.01]).

Conclusion: Severely injured patients with decreasing ex vivo measures of PA despite resuscitation have an increased risk of developing VTE. This may have implications for predicting development of VTE and for studying platelet targeted chemoprophylaxis regimens.

Level of evidence: Prognostic/Epidemiological; Level III.

Figure 1.

Coefficient Plot of Association of Platelet Aggregation on Presentation, at 24 hours, and Change from 0 to 24 hours with Development of Posttraumatic Venous Thromboembolism (VTE). X axis is the odds ratio per 10 aggregation units, adjusted for length of stay, initiation of VTE chemoprophylaxis within 72 hours, platelet count, injury severity score, shock index, and whether patient was transfused in initial 24 hours.

Figure 2.

Coefficient Plot of Association of Viscoelastic Clot Strength and Lysis at presentation (0 hours) and at 24 hours with Development of Posttraumatic VTE. X axis is the odds ratio per 10 aggregation units, adjusted for length of stay, initiation of chemoprophylaxis with in 72 hours, injury severity score, shock index, and whether patients were transfused in initial 24 hours.