Anticoagulation in non-malignant portal vein thrombosis is safe and improves hepatic function

Abstract

SummaryBackgroundNon-malignant portal vein thrombosis (PVT) is common in patients with advanced liver disease. Anticoagulation (AC) increases the chances of recanalization and may improve liver function in patients with cirrhosis.AimWe retrospectively assessed the course of non-malignant PVT in patients receiving AC.MethodsParameters related to hepatic injury (aspartate aminotransferase [AST]/alanine aminotransferase [ALT]), severity of disease (ascites) and synthesis function (albumin) as well as AC, rates of PVT regression/progression and AC-associated complications were documented.ResultsAmong 122 patients with PVT, 51 patients with non-malignant PVT (27 incomplete, 24 complete) were included, 12 patients (25%) received long-term AC therapy (≥9 months) as compared to 36 patients without long-term AC. We observed a trend towards higher regression rates with long-term AC of 58% (vs. 28% without AC; p = 0.08) and lower progression rates of 25% (vs. 42% without AC; p = 0.15). In the subgroup of patients with decompensation prior to PVT diagnosis (n = 39), long-term AC (n = 10, 25.6%) resulted in a significantly higher rate of PVT regression/resolution (70% vs. 24%, p = 0.031). Interestingly, AST/ALT tended to decrease (−19%/−16%) and the proportion of patients with ascites became lower (−33%) with long-term AC (without AC: ±0%). Furthermore, there was a significant improvement in albumin levels (+9%/+3.6 g/dl) when compared to patients without long-term AC (−2%/−0.8 g/dl; p = 0.04). Additionally, 10 patients were treated with direct oral anticoagulants (DOACs) for splanchnic vein thrombosis. Importantly, there were no AC-associated bleeding events in patients with conventional AC and one bleeding event in patients with DOAC treatment (10%).ConclusionOur findings support anticoagulation in patients with non-malignant PVT, since AC seems safe and associated with superior PVT regression rates and might also decrease hepatic injury and improve liver synthesis.

Recent developments in the field of vascular liver diseases

“…Apparently, this does not agree with an increasing body of data on the impact of recanalization during anticoagulation therapy. Recanalization during anticoagulant therapy is related to improved event‐free survival after adjustment for the severity of liver disease and portal hypertension, improved liver function tests as well as decreased mortality, particularly in patients with the most severe liver disease . It is frustrating to note that no predictors for the resolution of portal vein thrombus have been identified .…”

“…Results on analyses to identify predictors for recanalization on anticoagulation therapy have been varied. The delay in initiating anticoagulation is the only consistent factor …”

“…Not surprisingly multivariate analyses in these retrospective observational studies showed significant differences between patients who received anticoagulants and those who did not. In particular, a lower proportion of past variceal bleeding or cavernoma, lower blood urea levels, a smaller proportion of alcoholics, severe liver disease or cavernoma were found as well as an increased proportion of women or non‐specific beta‐blockers and a higher proportion of involvement of the superior mesenteric vein or multiple venous sites . These factors were not taken into account in the 2017 meta‐analysis …”

“…Although the use of direct oral anticoagulants for Budd‐Chiari syndrome and portal vein thrombosis is rapidly increasing, studies remain limited if not anecdotal. Indeed, the small numbers of patients in studies evaluating Budd‐Chiari syndrome, recent portal vein thrombosis and portal cavernoma in the absence of cirrhosis, and portal vein thrombosis in patients with cirrhosis, prevents drawing any conclusions and making any recommendations. Experience is increasing in patients with cirrhosis without splanchnic vein thrombosis but with other indications (mostly atrial fibrillation) .…”

“…The delay in initiating anticoagulation is the only consistent factor. [30][31][32][34][35][36] Two low-powered studies evaluated recurrence after spontaneous resolution, which was found to occur in 9 out of 20 patients 12 and 19 out of 89 patients 13…”