Abstract
Data to guide management of isolated bland cancer-associated splanchnic vein thrombosis (CA-SpVT) are limited. We aimed to assess the role of anticoagulation (AC) and bleeding and thrombosis in patients with CA-SpVT. We conducted a dual-center retrospective cohort study of adults with incident, isolated, bland CA-SpVT from 2011 to 2020. The primary outcome was major bleeding (MB); other outcomes included usual-site venous thromboembolism (VTE) recurrence and progression/recanalization of CA-SpVT. Time-to-event outcomes were analyzed with weighted Cox models adjusting for cancer type, stage, SpVT location, and whether symptomatic. For SpVT recanalization/progression, differences were estimated using weighted average treatment effects (ATEs). After excluding tumor thrombus, we included 437 patients with notable characteristics of median age 60 years, portal vein thrombosis (81.2%), and underlying hepatocellular cancer (35.9%). Of these, 29.5% received therapeutic AC. At 6 months, there were 11.9% MB and 6.4% incident usual-site VTE events. Among 308 patients with follow-up imaging, the 1-year thrombus progression rate was 19.8% and thrombus recanalization was 28.6%. In the adjusted analysis, there were numerically higher rates of MB with AC (adjusted hazard ratio [aHR] 1.93, 95% confidence interval [CI] 0.97-3.87) and no significant difference in the incidence of VTE (aHR 1.41, 95% CI 0.56-3.51). AC was associated with significantly higher likelihood of venous recanalization (ATE +24% 95% CI 13%-35%) and significantly lower likelihood of thrombus progression (ATE -14% 95% CI -23% to -5%). In patients with isolated bland CA-SpVT, AC was associated with thrombus recanalization and limited thrombus progression; effects were offset by a potentially higher risk of MB.