Perioperative antithrombotic management of patients who receive direct oral anticoagulants during gastroenterological surgery

Abstract Aim We investigated the effect of perioperative management of direct oral anticoagulants (DOACs) on bleeding and thromboembolic complications during gastroenterological (GE) surgery. Methods A total of 334 patients receiving anticoagulants and undergoing elective GE surgery between 2012 and 2018 were enrolled. The patients were divided into three groups: patients receiving warfarin (WF, n = 231), patients receiving DOACs with heparin bridging (DOAC‐HB, n = 34), and patients receiving DOAC without heparin bridging (DOAC‐NHB, n = 69). Outcome variables were compared between the groups and the risk factors of postoperative bleeding were assessed using logistic multivariate analysis. Results No significant differences were observed in background characteristics between the groups. There were similarities between the groups in surgical blood loss (P = .772) and rate of intraoperative transfusion (P = .952). Thromboembolic complications only occurred in two patients in the WF group (0.9%), and no thromboembolism occurred in the DOAC groups. The incidence of major postoperative bleeding was significantly higher in DOAC‐HB group than in the other groups (14.7% vs 4.8% vs 1.4%, P = .011). Multivariate analysis showed DOAC with heparin bridging to be the most significant risk factor of major postoperative bleeding (odds ratio = 11.60, P = .028). Conclusions Elective GE surgery can be safely performed in patients receiving DOACs without heparin bridging. Perioperative heparin bridging during DOAC interruption is not recommended even for patients undergoing major GE surgery due to increased postoperative bleeding.


| INTRODUC TI ON
The number of patients who receive anticoagulation therapy (ACT) for the prevention of ischemic stroke or venous thromboembolism is increasing with the aging of the population. Since clinical trials showed non-inferiority or superiority of direct oral anticoagulant (DOAC) therapy over vitamin K antagonist (warfarin) therapy in terms of safety and efficacy for patients with atrial fibrillation, 1-5 the number of patients who receive DOAC therapy has been increasing. Recent reports also indicate that approximately 10%-15% of patients treated with DOACs have to interrupt their anticoagulant treatment before an invasive procedure every year. 6,7 Direct oral anticoagulants, also known as non-vitamin K antagonist oral anticoagulants (NOACs), include direct thrombin inhibitors, such as dabigatran, and factor Xa inhibitors, such as rivaroxaban, apixaban, and edoxaban. The potential advantages of DOACs over warfarin include rapid onset and offset of action, reduced effect of dietary vitamin K intake or drug interaction on their activity, and predictable anticoagulant effects with no need for routine monitoring. 5,8 Although updated guidelines on digestive endoscopic procedures indicate the optimal perioperative management for patients who receive DOAC treatment, 9,10 perioperative management during gastroenterological (GE) surgery, which is considered to carry a high risk of bleeding, is yet to be established as it remains a challenge. In this study, we reviewed 334 consecutive patients who received ACT and underwent elective GE surgery, and we investigated the effect of perioperative management with DOACs on bleeding and thromboembolic complications.

| Patients
Our institutional review board approved this study (#19061903).
We searched the prospectively collected surgery database of a single institution for relevant cases, and we included 334 consecutive patients who underwent GE surgery in this study and excluded patients who underwent emergency surgery ( In case of DOAC-received patients with bridging heparin, DOAC was interrupted 1-3 days before the operation, heparin was started and continued until the day of surgery, DOACs and heparin bridging were resumed on POD 1-2, and heparin was discontinued on POD 4-5. If DOAC therapy was managed without bridging heparin, DOAC was stopped 0-1 days before the surgery and resumed 1-2 days after the surgery when hemostasis was secured. If a patient received antiplatelet therapy such as aspirin or clopidogrel, the perioperative management was performed according to the previously described perioperative management protocol (the Kokura Protocol). 11,12 Generally, antiplatelet agents were discontinued one week before surgery, but in patients with high thromboembolic risks, preoperative aspirin monotherapy was continued until the day before surgery. Postoperatively, early reinstitution of antiplatelets was performed unless there were signs of bleeding. For prevention of venous thromboembolism, mechanical prophylaxis (intermittent pneumatic compression and/or graduated compression stockings) and enforcement of early postoperative walking were generally performed, although routine use of medical prophylaxis with heparin was not adopted, except in the case of high venous thromboembolic risk patients with previous venous thrombosis or immobilization.
To assess the predicted thromboembolic risk of patients in each group, we used the revised CHADS 2 scoring system, 13-15 which is widely used for the prediction of ischemic stroke or transient ischemic attack (TIA) in patients with atrial fibrillation or atrial flutter.
This scoring system is an assessment tool that evaluates congestive heart failure, hypertension, age, diabetes, and ischemic stroke, and it categorizes patients with a score of 2 or higher into the high-risk group. It has been reported that the revised CHADS 2 scoring system F I G U R E 1 Consort diagram of the study. Abbreviations: ACT, anticoagulation therapy; DOAC, direct oral anticoagulant; GE, gastroenterological; HB, heparin bridging; NHB, non-heparin bridging; WF, warfarin also predicts ischemic stroke and death in patients without a history of atrial fibrillation or atrial flutter. 16,17 To assess the predicted bleeding risk of patients in each group, we used the HAS-BLED score, 18 which is widely used for the predicting bleeding risk in anticoagulated patients with atrial fibrillation.
The severity of patient symptoms and level of patient functioning in terms of ambulation were reported according to the Eastern Cooperative Oncology Group scale of performance status. 19 Postoperative complications were assessed and categorized according to the Clavien-Dindo classification (CDC) and CDC class 2 or higher was considered significant. 20 Postoperative thromboembolic complication was defined as per earlier reports. 11,12 In brief, thromboembolism included cerebral infarction, myocardial infarction, mesenteric infarction, pulmonary thromboembolism, and acute arterial embolism. Postoperative bleeding complication was catego-

| Statistical analysis
Continuous values were expressed as mean (standard deviation) or median (interquartile range), while categorical variables were presented as absolute numbers and percentages. For univariate comparisons, Fisher's exact probability test was used to evaluate categorical variables, and continuous variables were analyzed using one-way analysis of variance and Kruskal-Wallis test for normally distributed data and non-normally distributed data, respectively. Multivariate logistic regression analysis was performed to determine the risk factors of thromboembolic complications. All P-values were two-sided, and P-values less than .05 were considered statistically significant.
All statistical analyses were performed using EZR (Saitama Medical Centre, Jichi Medical University), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria, version 2.13.0). 22

| Patient and operative characteristics
A total of 334 patients were enrolled in this study. Figure

| Risk factors affecting thromboembolic and bleeding complications
Univariate and multivariate analyses of postoperative bleeding complications in the cohort were conducted and are shown in Table 4.
Univariate analysis revealed that only DOAC treatment with heparin bridging is associated with postoperative bleeding. Multivariate analysis also showed that DOAC treatment with heparin bridging is independently and significantly associated with postoperative bleeding (odds ratio = 11.60, P = .028).

| D ISCUSS I ON
In this study, which reviewed 334 patients who received ACT and underwent elective GE surgery, it was found that the incidences of overall postoperative complication, major bleeding complication, and thromboembolism were 28.4%, 5.1%, and 0.6%, respectively.
Thirty-four patients (33.0%) treated with DOACs received perioperative heparin bridging. Surgical blood loss and rate of RBC transfusion were identical between the groups, and thromboembolism was observed only in the WF group (0.9% of patients in the WF group).
The incidence of major postoperative bleeding was significantly higher in the DOAC-HB group than in the other groups, and multivariate analysis showed that DOAC therapy with heparin bridging is the most significant risk factor of major postoperative bleeding (odds ratio = 11.60, P = .028). Therefore, perioperative bridging with heparin is not recommended during DOAC therapy interruption even for patients who undergo major GE surgery.
Currently, DOACs are increasingly prescribed for the preven-  indications. They are "easy-to-use" drugs with a wide range of safety. Compared to warfarin, DOACs have superior pharmacological properties such as better dose-response, less difference in anticoagulant activity between individuals, no effect of vitamin K intake on anticoagulant activity, and few drug interactions. 5,23 Clinically, DOAC therapy also has many advantages including a 19% reduction in mortality and a 52% reduction in the incidence of intracranial hemorrhage compared to warfarin therapy. 5,23 There is strong evidence from large-scale randomized controlled trials supporting the use of all four DOAC agents, and the results of a meta-analysis on these four agents have been published in The Lancet. 5 The efficacy of DOACs (in preventing thromboembolism) is significantly higher than that of warfarin, and the safety of DOACs (in preventing bleeding events) is similar to that of warfarin.
DOACs are fast-acting agents that reach peak blood concentra- during the surgery and procedure but did not recommend heparin bridging during DOAC cessation even for procedures associated with a high risk of bleeding. 9,29,30 Several reviews and large-scale cohort studies 31 This study has some limitations. First, it is a retrospective review from a single center, and this reduces the strength of the conclusion.
This limitation will be mitigated in a later follow-up study. Second, our institution is a high-volume tertiary referral hospital for surgical patients who receive antithrombotic therapy; therefore, our findings may not be generalizable to lower-volume centers. This limitation can be minimized by conducting prospective multi-institutional studies.

| CON CLUS IONS
This study, in reviewing 334 patients who received ACT and underwent elective GE surgery, showed that DOAC therapy with heparin bridging is the most significant risk factor for postoperative bleeding complication. Perioperative bridging with heparin is not recommended during DOAC therapy interruption even for patients who undergo major GE surgery. TA B L E 4 Univariate and multivariate analyses of postoperative bleeding complication in the whole cohort (n = 334)

ACK N OWLED G EM ENTS
We would like to thank Editage (www.edita ge.com) for English language editing.

D I SCLOS U R E S
Conflicts of Interest: All authors declare no conflicts of interest.