Different predictive value for short-term all-cause mortality with commonly used biomarkers regarding the cause of pulmonary embolism

Abstract

Background/Aim. The evaluation of blood levels of cardiac troponin I (cTnI), D-dimer, B-type natriuretic peptide (BNP), and C-reactive protein (CRP) on admission and during the treatment of pulmonary embolism (PE) are the part of routine diagnostic process and estimation of mortality risk. The aim of this study was to evaluate the predictive value of these biomarkers on admission for all-cause 30-day mortality in consecutive PE patients regarding whether they classified as spontaneous, transiently provoked, or permanently provoked PE. Methods. This retrospective analysis was gained from the data of 590 PE patients from the Serbian University Multicenter Pulmonary Embolism Registry (SUPER). Patients had at least one of these biomarkers (BNP, CRP, cTnI, and D-dimer) measured during the first 24 hours upon admission. Results. Receiver operating characteristic (ROC) curve analyses demonstrated that BNP had the highest prognostic accuracy for 30-day mortality in patients (n = 219) who had data for all examined biomarkers. BNP provided an AUC of 0.785 (p < 0.001). Separately, BNP had the highest c-statistic for all three groups of patients. CRP had a modest predictive value for the 30-day all-cause mortality in the group with transient provoked PE. Troponin I had a very modest predictive value for the 30-day all-cause mortality only in patients with spontaneous PE, and D-dimer was a very weak predictor of this end-point only in patients with persistent provoked PE. Conclusion. Patients with spontaneous, transient provoked, and persistent provoked PE have a significantly different pro-file of blood biomarkers level with different prognostic significance for early all-cause mortality.