The Diagnostic and Prognostic Value of sTM, t-PA, and NLRP3 in ICU Patients with Sepsis

Abstract Background This study aimed to evaluate the diagnostic and prognostic values of soluble thrombomodulin (sTM), tissue plasminogen activator (t-PA), and NLRP3 in sepsis patients admitted to the intensive care unit (ICU). Methods This retrospective study enrolled 206 patients with sepsis or septic shock. Baseline characteristics and inflammatory markers were compared among sepsis, septic shock, and ICU control groups, as well as between survivors and non-survivors. Serum levels of sTM, t-PA, and NLRP3 were measured. ROC curves were used to evaluate diagnostic performance, while univariate and multivariate analyses identified significant prognostic predictors. Results Significant differences were observed in BMI, heart disease, lung disease, APACHE II scores, mechanical ventilation, vasopressor use, CRRT, and 28-day mortality across the three groups (all p < 0.05). Compared to ICU controls, the sepsis group showed significant differences in WBC and sTM levels, while the septic shock group exhibited variations in ten biomarkers, including WBC, neutrophil count, AST, BUN, Cr, CRP, IL-6, Lac, D-dimer, and sTM. Distinct biomarker profiles differentiated sepsis from septic shock (NLR, AST, CRP, IL-6, Lac). PCT, t-PA, and NLRP3 levels differed significantly across all group comparisons (all p < 0.05). Age, APACHE II score, neutrophil count, PLT, NLR, AST, BUN, Na+, PCT, P/F ratio, Lac, sTM, and t-PA were significantly different between survivors and non-survivors (all p < 0.05). Among individual indicators, NLRP3 showed the highest diagnostic accuracy (AUC = 0.992). The combination of sTM, t-PA, and NLRP3 demonstrated the best overall diagnostic performance (AUC = 0.822), followed by sTM and NLRP3 combined (AUC = 0.820). Prognostic stratification using APACHE II, IL-6, Lac, sTM, t-PA, and NLRP3 revealed significant differences in survival outcomes. APACHE II score, PCT, sTM, and t-PA were identified as independent prognostic factors. Conclusions sTM, t-PA, and NLRP3 effectively distinguish sepsis from ICU controls, while sTM and t-PA show strong discriminatory power for identifying sepsis survivors and non-survivors. The combination of all three biomarkers provides the best predictive performance for mortality, with sTM and t-PA serving as independent prognostic indicators.