Prognostic markers in pulmonary embolism: from risk scores to clinical outcomes

Abstract Background Pulmonary embolism (PE) is a potentially fatal cardiopulmonary emergency with unpredictable outcomes. Accurate risk stratification is critical for guiding intensive care unit (ICU) admission, optimizing resources, and improving prognosis. Clinical scores such as Pulmonary Embolism Severity Index (PESI), simplified PESI (sPESI), and Wells are widely used, but their limited sensitivity and specificity highlight the need for additional biomarkers. Laboratory parameters, including troponin, N-terminal pro-brain natriuretic peptide (NT-proBNP), lactate, D-dimer, and novel inflammatory indices such as platelet-to-lymphocyte ratio (PLR), may refine prognostic assessment and provide practical, low-cost tools for clinical decision-making. Methods This retrospective cohort study included 255 consecutive patients hospitalized with acute PE at a tertiary care center between January 2022 and June 2025. Diagnosis was confirmed by computed tomography pulmonary angiography. Demographics, comorbidities, clinical scores (PESI, sPESI, Wells), and laboratory biomarkers were analyzed. The primary outcome was ICU admission, and the secondary outcome was in-hospital mortality. Group comparisons were conducted using appropriate statistical tests. Receiver operating characteristic (ROC) analyses were performed to determine predictive accuracy and optimal cut-off values. Results The median age was 66 years, and 50.2% were female. ICU admission was required in 27.8% of patients ( n = 71), while in-hospital mortality occurred in 3.5% ( n = 9). Troponin demonstrated the highest predictive value for ICU admission (Area Under the Curve (AUC) = 0.783, cutoff = 22 ng/L), followed by D-dimer, sPESI, and PESI. Mortality was significantly associated with elevated lactate, troponin, urea, creatinine, and reduced albumin. PESI demonstrated the highest discriminatory performance for in-hospital mortality (AUC = 0.782), followed by sPESI (AUC = 0.742) and albumin (AUC = 0.720). PLR was associated with ICU admission. Malignancy was also identified as a major determinant of ICU admission and mortality. Conclusions The integration of clinical risk scores with readily available laboratory biomarkers may improve prognostic assessment in PE. Troponin and NT-proBNP reflected right ventricular dysfunction, lactate captured hemodynamic compromise, and inflammatory indices such as PLR provided additional information. Incorporating these tools into clinical algorithms may enhance individualized management, guide ICU triage, and optimize resource allocation, particularly in settings with limited access to advanced imaging.