Surgery Open Science (Jul 2022)

Venous thromboembolism in cancer surgery: A report from the nationwide readmissions database

  • Chelsea S. Pan, BA,
  • Yas Sanaiha, MD,
  • Joseph Hadaya, MD,
  • Cory Lee, DO,
  • Zachary Tran, MD,
  • Peyman Benharash, MD

Journal volume & issue
Vol. 9
pp. 58 – 63

Abstract

Read online

Background: The present study characterized the incidence of venous thromboembolism in a contemporary cohort of surgical oncology patients and its association with index hospitalization and postdischarge outcomes. Methods: Adults undergoing 7 major thoracic and abdominal cancer resections were identified in the 2016–2019 Nationwide Readmissions Database. Multivariable models stratified by operative subtype were developed to evaluate the association of venous thromboembolism with outcomes of interest. Results: Of an estimated 436,368 patients, venous thromboembolism was identified in 9,811 (2.2%) patients during index hospitalization. Esophageal (4.1%) and gastric (4.1%) resections exhibited the highest rates of venous thromboembolism, whereas pulmonary resection (1.0%) the lowest. Following adjustment, cancer resection type demonstrated the strongest association with venous thromboembolism development among all factors analyzed (adjusted odds ratio: 3.13, 95% confidence interval: 2.60–3.78). Diagnosis of venous thromboembolism was associated with increased mortality (10.2%, 95% confidence interval: 9.4–11.1 vs 1.7, 95% confidence interval: 1.6–1.7) and prolonged index hospital stay (19.5 days, 95% confidence interval: 19.1–20.0 vs 7.5, 95% confidence interval: 7.4–7.5). Of patients who survived index hospitalization, venous thromboembolism occurrence was associated with increased risk of nonhome discharge (56.4%, 95% confidence interval: 54.7–58.0 vs 14.4, 95% confidence interval: 14.2–14.7) and readmission (30.0%, 95% confidence interval: 28.5–31.1 vs 16.9, 95% confidence interval: 16.7–17.1). Additionally, venous thromboembolism substantially increased index hospitalization ($40,000, 95% confidence interval: $38,000–$42,000) and readmission costs ($3,200, 95% confidence interval: $1,700–$4,700). Conclusion: Rates of venous thromboembolism remain high in surgical oncology patients, with cancer resection type as a major predictor of venous thromboembolism incidence. Venous thromboembolism was associated with inferior clinical and financial outcomes that extended beyond discharge. These findings underscore the importance of continued vigilance and procedure-specific prophylaxis measures.