Purpose: The aim of the study was to investigate red cell distribution width (RDW) in predicting 30-day mortality in patients with pulmonary embolism (PE). Methods: A single-center, retrospective study design was used between January 1, 2014, and February 1, 2016. The primary end point was 30-day mortality after admission. The RDW predicting value was assessed by receiver operating characteristic curves and area under the curve. Results: A total of 309 patients with PE were included. The 30-day mortality was 14.9% (46/309). The mean RDW level was 13.9% +/- 0.6% (range, 10.7%-21.9%) at admission. The 30-day mortality was higher in the high-RDW-level group compared with the normal-RDW-level group (12.5% vs 23.5%, chi(2)=5.140, P=.023), with an odds ratio of 2.164 (95% confidence interval [CI], 1.019-4.450). Logistic regression showed that presence of shock, RDW level, and simplified pulmonary embolism severity index (sPESI) were independent risk factors for 30-day mortality in patients with PE. After adjustment by these risk factors, the adjusted odds ratio was 1.439 (95% CI, 1.024-2.116). The area under the curve for RDW predicting the 30-day mortality was 0.6646 (95% CI, 0.5585-0.7518). The cutoff was 16%. The Youden index for RDW and sPESI was 0.400 and 0.453, respectively. When adding RDW into sPESI, the modified sPESI showed highest prediction accuracy, with Youden index 0.499. Conclusions: Our results suggested that the RDW is a simple and useful indicator in predicting 30-day mortality in patients with PE. However, this conclusion showed be confirmed by prospective study with large sample. (C) 2016 Elsevier Inc. All rights reserved.