重症急性胰腺炎早期炎性介质清除率与可溶性血栓调节蛋白对临床转归的预测价值OA

Objective To investigate the predictive value of early inflammatory clearance rate and soluble thrombo-modulin(sTM)levels for clinical outcomes in patients with severe acute pancreatitis(SAP),construct a combined predictive model,and evaluate its efficacy.Methods We retrospectively analyzed clinical data from 110 SAP patients admitted to our hospital between June 2022 and June 2025.Patients were categorized into a favorable prognosis group(n=72)and an unfavor-able prognosis group(n=38)based on clinical outcomes within 28 days of onset.Clinical baseline characteristics,laboratory parameters,early inflammatory clearance rate,and sTM levels were compared between groups.Multivariate logistic regression identified independent prognostic risk factors.Receiver operating characteristic(ROC)curves assessed the predictive value of individual markers and the combined model,leading to the development of a nomogram-based risk prediction model.Results The unfavorable prognosis group exhibited significantly higher sTM levels and significantly lower early inflammatory clear-ance rate compared with the favorable prognosis group(t/P=5.709/<0.001;6.981/<0.001).The unfavorable prognosis group also showed significantly higher APACHE II scores,SOFA scores,and PCT levels than the favorable prognosis group(t/P=7.299/<0.001;10.433/<0.001;7.445/<0.001),as well as a significantly higher incidence of MODS(χ2/P=32.894/<0.001).Ele-vated PCT levels,high APACHE II scores,high SOFA scores,high MODS incidence,and elevated sTM were all independent risk factors for poor 28-day outcomes in SAP patients,while a high early inflammatory clearance rate was an independent pro-tective factor(all P<0.001).ROC curve analysis revealed that the predictive model combining these six indicators(AUC=0.999;95%CI:0.996-1.000)demonstrated optimal predictive performance,with a sensitivity of 100.00%and a specificity of 97.22%,surpassing both individual indicator models and the model combining sTM with early inflammatory clearance rate(AUC=0.888).Bootstrap validation of the nomogram confirmed good consistency between predicted and observed probabili-ties,and the DCA curve indicated significant clinical applicability.Conclusion The predictive model combining early inflam-matory clearance rate,sTM levels,APACHE II score,SOFA score,PCT levels,and MODS occurrence accurately predicts 28-day adverse outcomes in SAP patients.The nomogram provides reliable guidance for early clinical risk stratification and indi-vidualized treatment.