Venous thromboembolism in critically ill cirrhotic patients: practices of prophylaxis and incidence.

Objectives. We compared venous thromboembolism (VTE) prophylaxis practices and incidence in critically ill cirrhotic versus noncirrhotic patients and evaluated cirrhosis as a VTE risk factor. Methods. A cohort of 798 critically ill patients followed for the development of clinically detected VTE were categorized according to the diagnosis of cirrhosis. VTE prophylaxis practices and incidence were compared. Results. Seventy-five (9.4%) patients had cirrhosis with significantly higher INR (2.2 ± 0.9 versus 1.3 ± 0.6, P < 0.0001), lower platelet counts (115,000 ± 90,000 versus 258,000 ± 155,000/ μ L, P < 0.0001), and higher creatinine compared to noncirrhotic patients. Among cirrhotics, 31 patients received only mechanical prophylaxis, 24 received pharmacologic prophylaxis, and 20 did not have any prophylaxis. Cirrhotic patients were less likely to receive pharmacologic prophylaxis (odds ratio, 0.08; 95% confidence interval (CI), 0.04-0.14). VTE occurred in only two (2.7%) cirrhotic patients compared to 7.6% in noncirrhotic patients (P = 0.11). The incidence rate was 2.2 events per 1000 patient-ICU days for cirrhotic patients and 3.6 events per 1000 patient-ICU days for noncirrhotics (incidence rate ratio, 0.61; 95% CI, 0.15-2.52). On multivariate Cox regression analysis, cirrhosis was not associated with VTE risk (hazard ratio, 0.40; 95% CI, 0.10-1.67). Conclusions. In critically ill cirrhotic patients, VTE incidence did not statistically differ from that in noncirrhotic patients.