Meta-analysis favors anticoagulation for patients with cirrhosis and portal vein thrombosis


Key clinical point: Anticoagulation produced favorable outcomes with no increase in bleeding risk in patients with cirrhosis and portal vein thrombosis.

Major finding: Rates of any recanalization were 71% in treated patients and 42% in untreated patients (P less than .0001); rates of complete recanalization were 53% and 33%, respectively (P = .002), rates of spontaneous variceal bleeding were 2% and 12% (P = .04), and bleeding affected 11% of patients in each group.

Data source: A systematic review and meta-analysis of eight studies of 353 patients with cirrhosis and portal vein thrombosis.

Disclosures: The reviewers reported having no funding sources or conflicts of interest.




Patients with cirrhosis and portal vein thrombosis (PVT) who received anticoagulation therapy had nearly fivefold greater odds of recanalization compared with untreated patients, and were no more likely to experience major or minor bleeding, in a pooled analysis of eight studies published in the August issue of Gastroenterology (doi: 10.1053/j.gastro.2017.04.042).

cirrhosis Copyright Sebastian Kaulitzki/Thinkstock
Bleeding from portal hypertension is a major complication in cirrhosis, but PVT affects about 20% of patients and predicts poor outcomes, they noted. Anticoagulation in this setting can be difficult because patients often have concurrent coagulopathies that are hard to assess with standard techniques, such as PT-INR (international normalized ratio). Although some studies support anticoagulating these patients, data are limited. Therefore, the reviewers searched PubMed, the ISI Web of Science, SCOPUS, and the Cochrane database through Feb. 14, 2017, for trials comparing anticoagulation with no treatment in patients with cirrhosis and PVT.

This search yielded eight trials of 353 patients who received low-molecular-weight heparin, warfarin, or no treatment for about 6 months, with a typical follow-up period of 2 years. The reviewers found no evidence of publication bias or significant heterogeneity among the trials. Six studies evaluated complete recanalization, another set of six studies tracked progression of PVT, a third set of six studies evaluated major or minor bleeding events, and four studies evaluated spontaneous variceal bleeding. Compared with no treatment, anticoagulation was tied to a significantly greater likelihood of complete recanalization (pooled odds ratio, 3.4; 95% confidence interval, 1.5-7.4; P = .002), a significantly lower chance of PVT progressing (9% vs. 33%; pooled odds ratio, 0.14; 95% CI, 0.06-0.31; P less than .0001), no difference in bleeding rates (11% in each pooled group), and a significantly lower risk of spontaneous variceal bleeding (OR, 0.23; 95% CI, 0.06-0.94; P = .04).

“Metaregression analysis showed that duration of anticoagulation did not influence outcomes,” the reviewers wrote. “Low-molecular-weight heparin, but not warfarin, was significantly associated with a complete PVT resolution as compared to untreated patients, while both low-molecular-weight heparin and warfarin were effective in reducing PVT progression.” That finding merits careful interpretation, however, because most studies on warfarin were retrospective and lacked data on the quality of anticoagulation, they added.

“It is a challenge to treat patients with cirrhosis using anticoagulants because of the perception that the coexistent coagulopathy could promote bleeding,” the researchers wrote. Nonetheless, their analysis suggests that anticoagulation has significant benefits and does not increase bleeding risk, regardless of the severity of liver failure, they concluded.

The reviewers reported having no funding sources or conflicts of interest.

Next Article: