Association of Prolonged-Duration Chemoprophylaxis With Venous Thromboembolism in High-risk Patients With Head and Neck Cancer

Abstract

Importance: Venous thromboembolism (VTE) is associated with substantial morbidity and is the most common factor associated with preventable death among hospitalized patients. Data from otolaryngologic studies suggest that the risk of VTE may be underestimated among high-risk patients, particularly among those undergoing oncologic procedures. The incorporation of prolonged-duration chemoprophylaxis (PDC) into preventive therapy has been associated with substantial decreases in VTE incidence among patients undergoing oncologic surgery. However, bleeding remains a major concern among otolaryngologists, and substantial variation exists in the use of thromboprophylaxis. Objective: To assess the association between PDC and VTE in high-risk patients with head and neck cancer undergoing oncologic procedures. Design, Setting, and Participants: This retrospective cohort study identified 750 patients with biopsy-confirmed head and neck cancer and a Caprini risk score of 8 or higher who underwent inpatient oncologic surgery at a tertiary care referral center between January 1, 2014, and February 1, 2020. After exclusions, 247 patients were included in the study; patients were divided into 2 cohorts, traditional and PDC, based on the duration of prophylaxis. Univariate and multivariate analyses were performed to examine the development of VTE and bleeding-associated complications during the 30-day postoperative period. Data were analyzed from April 1 to April 30, 2020. Exposures: PDC, defined as 7 or more postoperative days of chemoprophylaxis. Main Outcomes and Measures: VTE and bleeding events during the 30-day postoperative period. Results: Among 247 patients (mean [SD] age, 63.1 [11.1] years; 180 men [72.9%]) included in the study, 106 patients (42.9%) received traditional prophylaxis, and 141 patients (57.1%) received PDC. The incidence of VTE was 5 of 106 patients (4.7%) in the traditional cohort and 1 of 141 patients (0.7%) in the PDC cohort (odds ratio [OR], 0.15; 95% CI, 0.003-1.33). In the multivariate logistic regression analysis, PDC was independently associated with reductions in the risk of VTE (OR, 0.04; 95% CI, 0.001-0.46). The incidence of bleeding events was 1 of 106 patients (0.9%) in the traditional cohort and 6 of 141 patients (4.3%) in the PDC cohort (OR, 4.64; 95% CI, 0.55-217.00). Conclusions and Relevance: The use of chemoprophylaxis for high-risk patients with head and neck cancer remains a high-priority topic. The results of this study suggest that PDC may be associated with reductions in VTE among this patient population. However, the associated increase in nonfatal bleeding events warrants careful consideration and further highlights the need to determine an optimal duration for chemoprophylaxis among this distinct cohort.