Examining the Link Between Cirrhosis and Venous Thromboembolism

A meta-analysis published in Thrombosis and Haemostasis, by Pasquale Ambrosino, MD, from the Department of Clinical Medicine and Surgery at the Federico II University in Naples, Italy, and authors confirmed the association between cirrhosis and venous thromboembolism (VTE), finding that patients with cirrhosis have a 1.2-percent higher absolute risk of  VTE than patients without cirrhosis (3.7% vs. 1.8%; odds ratio [OR] = 1.703; 95% CI 1.333-2.175; p<0.0001), which amounted to an attributable risk, defined as the risk of VTE among control patients divided by the risk for VTE among cirrhotic patients, of 51.3 percent.1

Previous research has suggested a connection between cirrhosis and increased VTE risk, but the results from this study support the role of screening for VTE in patients with this chronic liver disease, according to Dr. Ambrosino and authors. “An increased thromboembolic risk in cirrhotic patients compared [with] the general population may suggest the need for strict monitoring of risk factors for VTE to identify high-risk patients who could benefit from prophylaxis,” the authors wrote.

The literature review included 11 retrospective studies with 15 datasets, including data on 695,012 cirrhotic patients and 1,494,660 non-cirrhotic controls. The mean patient age ranged from 49.5 years to 58.8 years, and the percentage of male patients ranged from 49.3 percent to 88 percent. All studies evaluated the risk of VTE (including deep vein thrombosis [DVT] and/or pulmonary embolism [PE]) and reported on the prevalence of VTE and on adjusted risk estimates, hazard ratio (HR), relative risk (RR), or incidence rate ratio (IRR) for VTE. Case reports, reviews, animal studies, and studies without a control group were excluded.

Data on sample size; major clinical and demographic variables; and ORs, HRs, RR, and IRRs for DVT and/or PE were extracted, with data including only information on DVT and PE in the lower limbs.

The risk for VTE was significantly increased among patients with cirrhosis, compared with controls, and the adjusted risk estimates (including OR, HR, RR, and IRR) ranged from 0.40 to 2.76 (TABLE).

In a subgroup analysis, looking specifically at DVT risk, seven studies showed a significantly increased risk of DVT in 19,237 cirrhotic patients compared with 847,082 controls (OR=2.038; 95% CI 1.817-2.285; p<0.001). In an analysis looking specifically at PE risk, five studies showed a significantly increased risk of PE in 18,223 cirrhotic patients compared with 834,626 controls (OR=1.655; 95% CI 1.042-2.630; p=0.033).

A sensitivity analysis of nine studies “consistently confirmed” the increased risk of VTE in cirrhotic patients compared with the control group (OR=1.493; 95% CI 1.266-1.762; p<0.0001), the authors reported. The sensitivity analysis was conducted “given the confounding factors that might affect the difference in the risk of VTE between cirrhotic patients and non-cirrhotic controls.”

A regression model also determined that men with cirrhosis had a greater absolute difference in risk of VTE, compared with female patients and controls (p<0.001).

“Making physicians more aware of the potential association between cirrhosis and VTE could help reduce the incidence of this potentially avoidable and costly disease,” the authors concluded. “The higher risk of VTE emerging from our meta-analysis consistently supports the need for large long-term interventional trials with specific endpoints to confirm our results and to investigate whether appropriate strategies may modify the thromboembolic risk in cirrhotics.”

Several factors may mediate this association, Dr. Ambrosino and authors noted, including that most coagulation factors are synthesized in the liver, and, therefore, patients with chronic liver diseases are “naturally” anticoagulated because of low levels of coagulation factors. Patients with cirrhosis also may have more hypercoagulable features than controls, such as enhanced thrombin generation.

Limitations of this study include its heterogeneous patient population (including clinical and demographic characteristics), retrospective design, and study samples (including differences in inclusion and exclusion criterial, as well as methodology to screen for DVT and PE).

“Cirrhosis is a risk factor of venous thrombosis, and this reinforces the notion that routine thromboprophylaxis should not be withheld despite prolongations in the international normalized ratio,” Ton Lisman, PhD, from the Department of Surgery at the University of Groningen in the Netherlands, and Francesco Violi, MD, from the Department of Internal Medicine and Medical Specialties at the Sapienza University of Rome in Italy, wrote in an accompanying editorial published in Thrombosis and Haemostasis.2Clinical studies on optimal thromboprophylactic strategies in these patients that frequently have profound hemostatic alterations are urgently required.”


  1. Ambrosino P, Tarantino L, Di Minno G, et al. The risk of venous thromboembolism in patients with cirrhosis: A systematic review and meta-analysis. Thromb Haemost. 2016 October 20. [Epub ahead of print]
  2. Lisman T, Violi F. Cirrhosis as a risk factor for venous thrombosis. Thromb Haemost. 2016 October 27. [Epub ahead of print]

TABLE. Adjusted Estimates of the Thrombotic Risk Associated With Cirrhosis in Included Studies
Author Cirrhosis Controls Measure of Association Adjusted Risk Estimates
Venous thromboembolism
Ahmed 2012 23,673 50,291 IRR 1.41 (1.05–1.91)
Al-Dorzi 2013 75 723 HR 0.40 (0.10–1.67)
Ng 2015 2,223 22,230 OR 1.71 (1.05–2.78)
Ramagopalan 2011 – ORLS1 N/A N/A IRR 1.29 (0.52–2.67)
Ramagopalan 2011 – ORLS2 N/A N/A IRR 2.76 (1.37–4.95)
Ramagopalan 2011 – England N/A N/A IRR 1.49 (1.26–1.74)
Søgaard 2009 1,602 59,4714 RR 1.74 (1.54–1.95)
Wu 2010 – Compensated LC 40,8253


57,5057 OR 1.01 (0.96–1.06)
Wu 2010 – Decompensated LC 24,1626 57,5057 OR 1.10 (1.03–1.18)
Yang 2015 1,296 19,3532 OR 1.50 (1.20–2.00)
Deep venous thrombosis (DVT)
Enger 2014 15,158 45,473 IRR 1.95 (1.31–2.90)
Gulley 2008 963 12,405 OR 1.78 (1.10–2.20)
Pulmonary embolism (PE)
Enger 2014 15,158 45,473 IRR 2.51 (1.54–4.07)
OR=odds ratio; IRR=incidence rate ratio; HR=hazard ratio; N/A=not assessed; compensated LC=compensated liver cirrhosis; decompensated LC=decompensated liver cirrhosis