Combined SII, LAR, and D-dimer outperform sPESI in predicting adverse outcomes in acute pulmonary embolism
The study reports a single-center retrospective analysis of 151 patients with computed tomography-confirmed acute pulmonary embolism to evaluate the prognostic value of the systemic immune-inflammation index (SII), lactate-to-albumin ratio (LAR), and D-dimer for in-hospital adverse outcomes. A three-biomarker model achieved an AUC of 0.891 for predicting a composite endpoint of ICU transfer, shock, or death, substantially outperforming the standard sPESI score (AUC 0.637). Internal bootstrap validation confirmed the model's calibration and decision-curve net benefit, though adding sPESI to the biomarker panel did not significantly improve discrimination. These findings suggest that routine admission SII, LAR, and D-dimer measurements could provide a cost-effective prognostic adjunct to current risk stratification, pending multicenter external validation.
The original study
Association of SII, LAR, and D-Dimer with In-Hospital Adverse Outcomes in Acute Pulmonary Embolism.
- Authors
- Chen Z, Huang S, Song Y, Gong W, Zhang L, Ding W, et al.
- Journal
- Clinical and applied thrombosis/hemostasis : official journal of the International Academy of Clinical and Applied Thrombosis/Hemostasis
- Type
- Journal Article
- PMID
- 42758097
Original abstract
ObjectivesTo investigate the prognostic value of the systemic immune-inflammation index (SII), lactate-to-albumin ratio (LAR), and D-dimer for in-hospital adverse outcomes in patients with acute pulmonary embolism (APE).MethodsThis single-center retrospective cohort study screened 248 patients with computed tomography pulmonary angiography-confirmed APE between June 2022 and February 2026; 151 patients were included after 97 exclusions. Predictors were defined using the first measurements obtained within 24 hours after admission and before anticoagulation, thrombolysis, vasopressor therapy, or any related adverse event. The primary outcome was an in-hospital composite of intensive care unit transfer, shock or vasopressor use, cardiopulmonary resuscitation, catheter-based therapy or thrombolysis, and fatal outcome (in-hospital death or withdrawal of life-sustaining treatment after irreversible deterioration). Multiple imputation and logistic regression were used to construct and compare an sPESI model, a biomarker model, and a combined model. Performance was assessed using receiver operating characteristic curves, calibration analysis, bootstrap internal validation, and decision curve analysis.ResultsAmong 151 patients, 37 (24.5%) experienced the composite outcome and 114 did not. SII, LAR, and D-dimer remained associated with the outcome after adjustment for sPESI. Model 2 had an AUC of 0.891, compared with 0.637 for model 1 and 0.661 for ESC risk classification. Adding sPESI or ESC classification to model 2 increased the AUC to 0.912 and 0.900, respectively, without a significant further improvement in discrimination (P=.221 and P=.409). The bootstrap-corrected AUC for model 2 was 0.885, the Brier score was 0.091, and internal decision-curve analysis showed potential net benefit.ConclusionsAdmission SII, LAR, and D-dimer were associated with in-hospital adverse outcomes and remained associated after adjustment for sPESI. The exploratory three-biomarker model may complement sPESI and ESC risk classification, but multicenter external validation is required before clinical application.