Predictors of Deep Venous Thrombosis in Hospitalized Patients With Liver Cirrhosis in the US

Abstract

Background Patients with liver cirrhosis were previously considered as anticoagulated; thus, their risk of developing venous thromboembolism (VTE) is lower. Recently, several studies showed contradicting results regarding deep venous thrombosis (DVT) occurrence in cirrhotic patients. The aim of this study is to evaluate the prevalence and risk associated with developing DVT in hospitalized cirrhotic patients in a large US population. Methods We queried the commercial database Explorys (IMB Inc., Armonk, New York), an aggregate of electronic health record data from 26 US healthcare systems. After excluding patients under 20 years old, a cohort of patients with a Systematized Nomenclature of Medicine - Clinical Terms of "cirrhosis of the liver" and "inpatient care" between 2015-2019 were identified, and prevalence of DVT was calculated in the exposure and the control groups. Statistical analysis for a multivariable model was performed. Factors adjusted for include gender, race, obesity, hypoalbuminemia, diabetes mellitus, viral hepatitis, and liver malignancy. Results Among 9,990,290 patients who were hospitalized between 2015 and 2019, 157,400 patients had a diagnosis of liver cirrhosis. The prevalence of DVT in hospitalized patients with liver cirrhosis was 3.29% compared to 3.18% in non-cirrhotic patients. Using the multivariate analysis model, DVT was inversely associated with cirrhosis in hospitalized patients [OR: 0.921; p<0.0001] compared to patients without liver cirrhosis. Predictors of developing DVT among patients with cirrhosis were non-Caucasian race, obesity (BMI>30), liver malignancy, hypoalbuminemia, and diabetes mellitus. Cirrhotic patients due to viral hepatitis were less likely to develop DVT [OR: 0.775; p<0.001] compared to non-cirrhotic patients. Conclusion In this database, although the prevalence of DVT in cirrhotic hospitalized patients was slightly higher than in non-cirrhotic patients (3.29% vs. 3.18%, respectively), cirrhosis as an independent factor was associated with less risk of DVT during hospitalization. This poses a question regarding DVT prophylaxis necessity in this group of patients. Further studies are needed to clarify the benefit and risks of DVT prophylaxis in cirrhotic patients.