Prognostic Value of Inflammatory Biomarkers for Mortality in Intensive Care Patients with Acute Pulmonary Embolism: A Retrospective Observational Study.

ObjectiveTo develop a new inflammatory biomarker-based and simple-to-use nomogram for the early identification of acute pulmonary embolism (PE) patients at a high-risk mortality in intensive care unit (ICU).MethodsWe randomly allocated 1083 acute PE patients from the Medical Information Mart for Intensive Care IV database to derivation and internal validation cohort. We used logistic regression analysis to determine independent risk factors and to construct a predictive nomogram. We subsequently evaluated the calibration, discrimination and clinical usefulness of the nomogram.ResultsAge>66, neutrophil-to-lymphocyte ratio (NLR) > 10.1, lymphocyte-to-monocyte ratio (LMR) < 1.5, red cell distribution width (RDW) > 14.35, respiratory rate (RR) > 26bpm, oxygen saturation (SPO2), vasopressor use and malignant cancer were detected as important determinants of 28-day mortality and included in our nomogram. The calibration plot revealed an adequate fit of the nomogram for predicting the risk of 28-day mortality. Regarding discriminative power, receiver operating characteristic curve analysis showed that the nomogram had an area under the curve of 0.772 (95% CI:0.732, 0.811, P < .001) in the primary cohort, outperforming other scores.ConclusionsThis proposed simple-to-use nomogram based on age, NLR, LMR, RDW, vasopressor use, RR, SPO2 and malignant cancer provides accurate death prediction for acute PE patients in ICU.