Cancer and surgery are major independent risk factors for venous thromboembolism (VTE). While postoperative VTE prophylaxis is considered standard-of-care, the optimal duration remains controversial. Clinical trials have demonstrated a lower rate of VTE with extended thromboprophylaxis (ExTP) for 3-4 weeks compared to standard thromboprophylaxis (STP) for 1-2 weeks, but data from real world settings is lacking. We sought to investigate differences in outcomes between ExTP and STP following major abdominal surgery by conducting a retrospective analysis of the SEER Medicare database, a large national dataset enriched for older adults with cancer.
Eligible patients were entitled to Medicare based on age, enrolled in Medicare Parts A and B for ≥12 months and Medicare Part D for ≥3 months prior to index surgery, were diagnosed with gastrointestinal or genitourinary cancers based on first primary site, and underwent major abdominal/pelvic surgery within one month before or any time after index cancer diagnosis. For patients with multiple admissions, only the first admission was included. We restricted to patients with a prescription for anticoagulation <7 days preoperatively or 30 days postoperatively. Patients with a prescription for vitamin K antagonists (i.e., patients undergoing a heparin to warfarin bridge for long-term anticoagulation) were excluded. We defined the exposure based on the length of the anticoagulation prescription, with STP defined as 1-14 days and ExTP defined as ≥21 days. We defined VTE as the presence of a diagnosis code for VTE in the primary or secondary position of an inpatient admission, and we also stratified VTE into pulmonary embolism (PE) and deep vein thrombosis (DVT).We used Poisson regression with robust variance estimators to calculate the risk ratio (RR) and 95% confidence interval (CI) for VTE at 30-day and 90-day following surgery and for 30-day and 90-day readmission while controlling for age, primary cancer site, and Elixhauser Comorbidity Score (ECS).
A total of 1,178 eligible patients were identified, with 559 (47.5%) receiving STP, 584 (49.6%) receiving ExTP, and 35 (3.0%) receiving other durations of thromboprophylaxis. Of the 1,143 individuals with either STP or ExTP, the median duration of VTE prophylaxis was 7 days (IQR: 5-11) in the STP group and 30 days (IQR: 28-30) in the ExTP group. Median ages were 74.2 (IQR: 69.8-79.5) and 74.2 (IQR: 69.8-78.9) in the STP and ExTP groups, respectively. Females comprised 63.0% of the STP group and 72.1% of the ExTP group, and the majority patients in both groups identified as white (85.8% and 86.8%, respectively). The most common primary cancer site overall was colorectal (56.1%), which was more common in the STP group (64.2%) than the ExTP group (47.6%). Ovarian cancer was the second most common in both groups (22.4% and 40.3%, respectively), followed by pancreas cancer (7.2% and 8.2%, respectively) and gastric cancer (6.2% and 3.9%, respectively). Both groups had a median ECS of 4 (IQR: 3-5). Overall, the most common anticoagulant used was enoxaparin (83.4%) followed by rivaroxaban (11.6%), apixaban (4.3%) and fondaparinux (0.6%).
The 30-day VTE rate was 2% in the ExTP group compared to 2.7% in the STP group (adjusted RR: 0.77, 95% CI: 0.35-1.67). Results were similar for VTE at 90 days (2.1% vs. 3.3%, respectively; aRR: 0.68, 95% CI: 0.33-1.42). The incidence of PE at 30-day and 90-day was 1.4%, and 1.4% in the ExTP group, respectively, and 1.5% and 1.7% in the STP group, respectively. The incidence of DVT at 30-day and 90-day was 0.5% and 0.7% in the ExTP group respectively, and 0.7% and 1.5% in the STP group, respectively. Readmission by 30 days was lower in the ExTP group (14.9%) than in the STP group (19.5%) (aRR: 0.78, 95% CI: 0.58-1.04) and 90-days (24.5% vs. 30.0%, respectively; aRR: 0.82, 95% CI: 0.65-1.03).
In this real-world data set postoperative extended thromboprophylaxis after major abdominopelvic cancer surgery was associated with numerically lower postoperative VTE and 30-day readmission rates, but the absolute benefit may be marginal. Although current practice guidelines suggest pharmacologic thromboprophylaxis for up to 28 days in this high-risk subpopulation, our findings support individualized decision making given the benefit may be relatively low and needs to be balanced with postoperative bleeding risk.
Zwicker:UpToDate: Patents & Royalties; Med Learning Group: Consultancy; Regeneron: Consultancy; BMS: Consultancy; Calyx: Consultancy; Parexel: Consultancy; Quercegen: Research Funding; Incyte Corporation: Research Funding. Patell:Merck Research: Consultancy, Other: Personal fees.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal