Prevalence of venous thromboembolism in admissions and readmissions with and without syncope: a nationwide cohort study.

Abstract

The Pulmonary Embolism in Syncope Italian Trial reported 17.3% prevalence of pulmonary embolism (PE) in patients admitted with syncope. We investigated the prevalence of venous thromboembolism [VTE, including PE and deep vein thrombosis (DVT)] in syncope vs. non-syncope admissions and readmissions, and if syncope is an independent predictor of VTE. We conducted an observational study of index admissions of the 2013-14 Nationwide Readmission Database. We excluded patients <18 years, December discharges, died during hospitalization, hospital transfers, and missing length of stay. Encounters were stratified by the presence or absence of DVT/PE and syncope diagnoses. Multivariable logistic regression analysis was used to evaluate the association between syncope and VTE. There were 38 655 570 admissions, of whom 285 511 had syncope. In the overall cohort, syncope occurred in 1.6% of VTE and 1.8% in non-VTE admissions. In a multivariable model, syncope was associated with a lower prevalence of VTE [odds ratio (OR) 0.76, 95% confidence interval (CI) 0.75-0.78; P < 0.001]. In index syncope vs. non-syncope admissions, the prevalence of DVT, PE, and VTE were 0.4 ± 0.06% vs. 1.3 ± 0.12%, 0.2 ± 0.04% vs. 1.2 ± 0.11%, and 0.5 ± 0.07% vs. 2.1 ± 0.14% (all P < 0.001), respectively. At 30 days, the prevalence of DVT, PE, and VTE in syncope vs. non-syncope were 2.2 ± 0.14% vs. 2.1 ± 0.14% (P = 0.38), 1.4 ± 0.12% vs. 1.2 ± 0.11% (P = 0.01), and 2.6 ± 0.17% vs. 3.0 ± 0.17% (P = 0.99), respectively. Syncope admissions were associated with a lower prevalence of VTE as compared to non-syncope admissions. Syncope should not trigger an automatic PE workup, rather, should be put into context of patient presentation.