Venous thromboembolism and prophylaxis therapy after elective spine surgery: a population-based study

Currently, there is no generalized consensus regarding perioperative prophylaxis of venous thromboembolism (VTE) in patients undergoing spine surgery. In the absence of large-scale studies, we aimed to use national data to study the association between anticoagulant prophylaxis and VTE in spine surgical patients. Our secondary outcomes were hematoma and blood transfusion. We included anterior cervical discectomy and fusion (ACDF) and posterior lumbar fusion (PLF) cases from 2006 to 2016 recorded in the Premier Healthcare database. Anticoagulant prophylaxis was categorized into aspirin, regular heparin, and low molecular weight heparin given on the day of surgery. Mixed-effects models measured the association between anticoagulation categories and outcomes. Cohorts were adjusted to reduce the risk of "confounding by indication" and to distinguish between prophylactic and therapeutic use of anticoagulants. We report odds ratios (OR) and Bonferroni-corrected confidence intervals (CI). Among 83,839 individuals undergoing ACDF and PLF, 0.45% (n = 374) had a hematoma, 8.1% (n = 6,769) received a blood transfusion, and 0.13% (n = 113) experienced VTE. After adjustment for relevant covariates, prophylactic aspirin (OR, 1.48; CI, 1.17 to 1.86) and regular heparin (OR, 2.01; CI, 1.81 to 2.24) were associated with increased odds of blood transfusion. No detectable differences in the odds of hematoma or VTE were observed for any anticoagulant. Although low molecular weight heparin was used much less frequently than regular heparin, it was associated with a lower incidence of transfusion compared with aspirin and regular heparin. All three anticoagulants were associated with similar incidence of VTE and hematoma. Varying subgroup-specific VTE risks may further inform future studies to identify patients expected to benefit the most from chemical thromboprophylaxis.