Effect of platelet dynamic changes on disseminated intravascular coagulation and prognosis in severe heatstroke patients.

Objective: This study aimed to investigate the dynamic changes in the platelets of patients with severe heatstroke and the impact of these changes on the occurrence of disseminated intravascular coagulation (DIC) and prognosis in them.

Methods: This retrospective cohort study conducted at two tertiary hospitals recruited 264 patients with severe heatstroke. Logistic regression was used to analyze the association between platelet counts and DIC. The receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of platelets count for DIC occurrence. We used mediation effect to analysis the role of DIC as a mediating variable to mediate the relationship between platelet count decrease after 24 hours and death.

Results: There were 214 patients with lower platelet counts compared to admission (107 × 10⁹/L[69,168] vs.171 × 10⁹/L[126,215], p < 0.001). The DIC patients had lower platelet counts than the non-DIC patients when measured in the emergency department and after 24 hours. The platelet count decrease after 24 hours was a risk factor for DIC (odds ratio [OR] = 2.710, 95% confidence interval [CI] = 1.069-6.869). The results of the ROC curve revealed that the predictive performance of the platelet count after 24 hours (area under the curve [AUC] = 0.8685, 95% CI = 0.8173-0.9197) was significantly better than that of the platelet count measured in the emergency department (AUC = 0.7080, 95% CI = 0.6345-0.7815). Mediation analyses showed that PLT decrease after 24 hours did not directly lead to death, but can indirectly cause death by inducing the development of DIC.

Conclusions: Decreased platelet count is an independent risk factor for DIC in patients with severe heatstroke. Although the platelet counts measured in the emergency department and after 24 hours show a good predictive performance for DIC occurrence, the prediction performance of the latter is better.