Retrospective Cohort Analysis of Aspirin Use and Venous Thromboembolism in Patients with Pancreatic Cancer and an Indwelling Central Venous Catheter.

Patients with pancreatic cancer are at high risk of developing venous thromboembolism (VTE). It is unknown if aspirin reduces the risk of VTE in this setting.  We sought to determine whether there is an association between aspirin use and VTE risk in patients with pancreatic cancer receiving chemotherapy with a central venous catheter (CVC).  We conducted a single-center, retrospective cohort study of adult patients diagnosed with pancreatic cancer and treated with chemotherapy using a CVC. Subjects were excluded if they were on anticoagulation at the time of CVC placement. The probability of VTE was analyzed using a time-to-event analysis framework for the development of VTE using the product-limit method of Kaplan and Meier (univariate) and adjusting for important confounding covariates using Cox proportional hazards regression (cause-specific hazard) and again using Fine and Gray regression (subdistributional hazard) with death prior to VTE considered a competing event.  The final analysis included 314 cases (125 with any aspirin use and 189 without). Patients with any aspirin use had fewer VTE events (34.4%) compared with those without aspirin use (42.3%;  = 0.021) by log-rank test and after adjustment for multiple covariates using a Cox proportional hazards model (hazard ratio [HR] = 0.60; 95% confidence interval [CI]: 0.40-0.92;  = 0.019). Using Fine and Gray regression to account for death as a competing event, the effect of aspirin remained in the direction of benefit, but was not statistically significant (HR = 0.70; 95% CI: 0.47-1.05,  = 0.083). Higher body mass index, active smoking, and metastatic stage of cancer were associated with VTE events in the Cox proportional hazards model. Rates of major bleeding or clinically relevant minor bleeding were similar between treatment groups.  Aspirin may reduce the risk of VTE in patients with pancreatic cancer with a CVC. We did not observe a significant increase in the rates of major bleeding or clinically relevant nonmajor bleeding.