Superiority of a combined antithrombin and activated protein C model for predicting coagulopathy and mortality in pneumonia-related sepsis: A prospective cohort study.

BACKGROUND

Pneumonia-related sepsis is frequently complicated by coagulation disorders, which significantly increase mortality risk. The current ISTH criteria demonstrate limited sensitivity in diagnosing pneumonia-specific coagulopathy, and the mechanisms underlying clinical phenotypic heterogeneity remain poorly understood. There is a growing need to identify novel biomarkers that can enhance the early detection and stratification of disseminated intravascular coagulation (DIC), particularly in its non-overt stages.

AIMS: The study aimed to evaluate the prognostic value of antithrombin (AT) and activated protein C (aPC) testing for predicting clinical deterioration and early DIC progression in adults with severe pneumonia-related sepsis.

METHODS

We conducted a prospective observational cohort study of 86 adults with severe pneumonia-related sepsis admitted to the Intensive Care Unit (ICU). All patients underwent testing of AT and aPC upon enrollment. The predictive value of this biomarker combination was evaluated for the development of overt DIC within 72 h and 28-day all-cause mortality using receiver operating characteristic (ROC) analysis.

RESULTS

Among 86 adults with severe pneumonia-related sepsis (median age 65 years; 56.9 % male), 18 (20.9 %) progressed to overt DIC within 72 h, and the 28-day mortality rate was 26.7 % (23/86). Compared with survivors, non-survivors exhibited significantly lower antithrombin activity (median 65 % vs. 78 %; P = 0.005) and higher activated protein C levels (median 127 pg/mL vs. 93.9 pg/mL; P < 0.001). Notably, the presence of overt DIC was the strongest predictor of mortality (OR = 144, 95 % CI: 18.66-1111.07). Similarly, patients with overt DIC exhibited markedly elevated aPC (P < 0.001). Among non-survivors, aPC was positively correlated with the 24-hour vasoactive-inotropic score (VIS; r = 0.567, P = 0.005). ROC analysis revealed that for predicting mortality, both aPC (AUC = 0.804, 95 % CI: 0.709-0.900) and the combined AT/aPC model (AUC = 0.814, 95 % CI: 0.714-0.913) significantly outperformed AT alone (AUC = 0.736, 95 % CI: 0.595-0.878).For predicting overt DIC, aPC (AUC = 0.859, 95 % CI: 0.778-0.939) and the combined model (AUC = 0.850, 95 % CI: 0.764-0.935) both demonstrated high accuracy. Notably, only the combined model effectively identified non-overt DIC (AUC = 0.850, 95 % CI: 0.768-0.932), whereas AT (AUC = 0.723) and aPC (AUC = 0.742) alone showed suboptimal performance.

CONCLUSION

In conclusion, this preliminary study identifies a combined AT/aPC model as a promising early predictor of DIC progression and mortality in pneumonia-sepsis. If validated in larger, multicentre cohorts, this simple two-biomarker approach provides a potential biomarker foundation for future studies investigating personalised anticoagulation strategies by identifying high-risk patients in the non-overt DIC stage.