Postoperative thromboprophylaxis with oral rivaroxaban versus subcutaneous low-molecular-weight heparin: a retrospective comparison in women with a gynecologic malignancy

Abstract

<h3>Objectives:</h3> Perioperative venous thromboembolism (VTE) is the most common preventable cause of in-hospital death in the US. Current guidelines recommend perioperative pharmacologic prophylaxis for patients undergoing abdominal or pelvic surgery for a malignancy. However, there is limited evidence comparing methods of thromboprophylaxis, more specifically the administration of a subcutaneous low molecular weight heparin (LMWH) versus a direct oral anticoagulant (DOAC) in women with gynecologic malignancies. We aim to determine the incidence of post-operative VTE among gynecologic cancer patients stratified by method of pharmacologic VTE prophylaxis. <h3>Methods:</h3> Women who underwent a laparotomy as part of their treatment for a gynecologic malignancy and subsequently received extended post-operative thromboprophylaxis with either subcutaneous enoxaparin or oral rivaroxaban were included. All patients received care at Carilion Roanoke Memorial Hospital between January 2010 - December 2017. A retrospective record review of demographics and clinical characteristics, as well as method and duration of VTE prophylaxis after hospital discharge were collected. The primary outcome of incidence of VTE was defined as deep venous thrombosis and/or pulmonary embolism within 30 days and 90 days of surgery. The secondary outcome of incidence of major bleeding events was defined as a re-admission or repeat surgery for bleeding/acute blood loss or transfusion for acute blood loss anemia. Wilcoxon Two Sample Test and Fisher's Exact Test were performed to determine the rate of VTE, stratified by method of thromboprophylaxis. <h3>Results:</h3> A total of 315 women were included who received extended thromboprophylaxis, of which 233 (74%) received enoxaparin and 82 (26%) received rivaroxaban. Baseline demographics and surgical-pathologic variables were overall well matched, with exception of significantly longer median surgical duration in the rivaroxaban group (128 vs 113 min) and higher rate of lymph node dissection in the enoxaparin group. The incidence of VTE at 30 days in the enoxaparin group was 1.7% (4/233) compared to 1.2% (1/82) in the rivaroxaban group (p=1.00). The cumulative incidence of VTE at 90 days in the enoxaparin group was 4.3% (10/233) compared to 2.4% (2/82) in the rivaroxaban group (p=0.74). Major bleeding events were 0.4% (1/233) in patients receiving enoxaparin compared to 3.7% (3/82) in those receiving rivaroxaban (p=0.06). <h3>Conclusions:</h3> No significant difference was seen in the incidence of VTE or rate of major bleeding events following laparotomy in patients with gynecologic malignancies who received subcutaneous enoxaparin versus rivaroxaban for extended thromboprophylaxis. Rivaroxaban may be an effective and safe alternative for extended thromboprophylaxis. Given the small sample size, these findings should be taken with caution but provide support for future multi-institutional and prospective studies.