Postpancreatectomy hemorrhage (PPH) is the most severe type of complication after pancreatic surgery, although the effect of antithrombotic therapy (ATT) on PPH is largely unknown. The safety and efficacy of chemical thromboprophylaxis for venous thromboembolism (VTE) remains controversial.

To elucidate the effect of ATT on PPH.

Published articles between 2013 and 2020 were searched from PubMed and Google Scholar, and after careful reviewing of all studies, studies concerning ATT and pancreatic surgery were included. Data such as study design, type of surgical procedures, type of antithrombotic drugs, and surgical outcome were extracted from the studies.

Nineteen published articles with a total of 37863 patients who underwent pancreatic surgery were included in the systematic review. Fourteen were cohort studies, with only three being prospective in nature. Two studies demonstrated that in patients receiving chronic ATT, which were mostly managed by heparin bridging, the risk of PPH was higher compared with those without ATT, and one study showed that patients with direct-acting oral anticoagulants managed by heparin bridging had significantly higher postoperative bleeding rates than others. The remaining six studies reported that pancreatic surgery can be safely performed in patients receiving chronic ATT, even under preoperative aspirin continuation. Concerning chemical thromboprophylaxis for VTE, most studies have shown a potentially high risk of PPH in patients undergoing chemical thromboprophylaxis; however, its effectiveness against VTE has not been statistically demonstrated, particularly among Asian patients.

Pancreatic surgery in chronically ATT-received patients can be safely performed without an increase in the occurrence of PPH, although the safety and efficacy of chemical thromboprophylaxis for VTE during pancreatic surgery is still controversial. Further investigation using reliable studies with good design is required to establish definite protocols or guidelines.

Heart disease, cerebrovascular disease, and cancer are the three leading causes of death in the world. With the aging of society in recent years, more patients with cardiovascular and/or cerebrovascular disease can have opportunities to require non-cardiac surgery. Most patients receive antithrombotic therapy (ATT) to prevent thromboembolism. The perioperative period of these patients is at high risk of both bleeding and thromboembolism, which is very troublesome for surgeons[1,2].

ATT is classified into two types of drugs: antiplatelet drugs and anticoagulants. Antiplatelet drugs are frequently used for prevention of cerebrovascular or cardiovascular diseases, and can prevent thromboembolism by reduction of platelet aggregation. Antiplatelet agents include thienopyridine (e.g., clopidogrel, prasugrel, or ticlopidine), aspirin, and type III phosphodiesterase inhibitor (e.g., cilostazol)[3,4]. Anticoagulants, on the other hand, prevent coagulation of blood by suppressing the coagulation cascade. They are usually used for deep vein thrombosis, atrial fibrillation, and acute coronary syndrome, and cardiac endoprostheses. Anticoagulants are also used for perioperative thromboprophylaxis of venous thromboembolism (VTE). Oral anticoagulants include warfarin, factor Xa inhibitors (e.g., rivaroxaban, apixaban, edoxaban), and direct thrombin inhibitors (e.g., dabigatran)[4,5]. The latter two types are called direct-acting oral anticoagulants (DOACs) or non-vitamin K antagonist oral anticoagulants (NOACs), and now increasingly used. Table 1 summarizes the type and the duration of action of each antithrombotic agent.

Pancreatic surgery for malignancy is one of the most invasive abdominal surgical procedures and may expose patients to high risks of severe postoperative complications. Although post-pancreatectomy mortality has dropped significantly to less than 5% owing to advances in perioperative management and surgical procedures[6,7], postoperative complication rates remain high at 18%-50%[6-9]. Common types of post-pancreatectomy complications are post-pancreatectomy hemorrhage (PPH), deep surgical site infection (SSI) and postoperative pancreatic fistula (POPF). PPH is related to high mortality rates of up to 60%[6-14], with an incidence of 3%-16% after overall pancreatic resection[6-9] and 8%-29% after pancreaticoduodenectomy[11-14]. When performing pancreatic surgery in ATT-received patients, both rigorous hemostasis procedures and strict perioperative antithrombotic management are required to prevent both thromboembolic and bleeding complications. To date, there is no consensus on the safety of pancreatic resection and proper perioperative antithrombotic management in patients receiving ATT, and the optimal thromboprophylaxis for venous thromboembolism (VTE) also remains unclear.

The aim of the present review study is to assess the effect of ATT and chemical thromboprophylaxis on thromboembolic complications and PPH in pancreatic surgery.

Articles written in English and published between 2013 and 2020 were collected from PubMed and Google Scholar (Figure 1). The following key words were used for the search: Antithrombotic therapy, aspirin, clopidogrel, antiplatelet therapy, anticoagulation, warfarin, DOAC, NOAC, bleeding, postpancreatectomy hemorrhage, pancreatic surgery, and pancreatectomy. The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Protocols (PRISMA-P) 2015 checklist[15]. Articles were included when published in the peer review journal. The eligible studies included randomized clinical trials, prospective or retrospective cohort studies, or case-control studies; guidelines, review articles, or case series/reports were not included.

Open in New Tab   Full Size Figure   Download Figure

Figure 1  PRISMA flow diagram demonstrating articles selection process.

After removing duplicates, articles were excluded systematically by careful review of each study. The quality of each study was assessed depending on the study design, and eligible articles were determined. Complete data were extracted from each study, including the design of the study, year of publication, sample size, type of surgical procedures, type of antithrombotic drugs, and surgical outcome (both bleeding and thromboembolic complications). Bleeding events included two categories: Increased surgical blood loss (SBL) and PPH. PPH was defined according to the International Study Group of Pancreatic Surgery (ISGPS)[16].

To the best of our knowledge, this is the first systematic review study to assess the effect of ATT and medical thromboprophylaxis on thromboembolic complications and PPH during and after pancreatic surgery. The current review summarized 19 published articles with a total of 37863 patients receiving pancreatic surgery with special reference to ATT. Regarding the effect of ATT on PPH in patients undergoing pancreatectomy, some studies demonstrated that chronically ATT-received patients managed by heparin bridging potentially had an elevated risk of PPH, although the majority of included studies reported that pancreatic surgery can be safely performed in patients receiving chronic ATT, even under preoperative aspirin continuation. Concerning chemical thromboprophylaxis for VTE, most studies have shown a potentially high risk of PPH in patients undergoing chemical thromboprophylaxis, although their effectiveness against VTE has not been statistically demonstrated.

For the management of patients receiving chronic ATT, guidelines regarding antithrombotic management during non-cardiac surgery were recently updated and demonstrated that the prevention of thromboembolic events is more crucial than bleeding complications, since it might cause severe sequelae or death[4,35-38]. However, major gastroenterological surgery including pancreatic surgery showed scarce evidence concerning the definite protocol or guidelines.

Our hospital is a high-volume center for referrals to patients with gastrointestinal cancer who are taking antithrombotic drugs. For this reason, we currently use a centralized protocol for antithrombotic drug management in ATT-received patients receiving elective gastroenterological surgery including pancreatic surgery, which was established and has been updated with reference to several guidelines and recently reported studies concerning perioperative antithrombotic management for endoscopic procedures or non-cardiac surgeries[4,5,35-37]. In general, the management includes three ways according to the type of ATT: antiplatelets, warfarin, and DOACs. In patients with thromboembolic risks, continued aspirin monotherapy is performed in patients receiving antiplatelet therapy, and warfarin is substituted by DOAC bridging (preferred) or heparin bridging. In the case of DOAC therapy, short-period discontinuation of DOACs (usually 1-2 d) without heparin bridging is recommended. However, if the thromboembolic risk is very high, heparin bridging might be considered. Postoperatively, every antithrombotic agent is re-instituted as soon as possible (POD1-2).

Concerning the management of patients with antiplatelet drugs, some studies such as POISE-2 study have suggested that a slight increase in bleeding risk was observed in patients with continued antiplatelets during non-cardiac surgery[39,40], but most of other studies demonstrated that the bleeding events were not significantly increased[24,41]. Moreover, one recently published large-scale retrospective cohort studies showed that the continuation of preoperative aspirin therapy significantly reduced the occurrence of thromboembolic complication but was not related to bleeding complications or PPH[18]. In the current systematic review, two recently reported studies demonstrated that in patients receiving chronic antiplatelets or anticoagulants, which were mostly managed by heparin bridging, the risk of PPH was higher compared with those without ATT[10,17]. However, three other studies reported that preoperative continuation of aspirin monotherapy is safe and feasible during hepatobiliary-pancreatic or gastrointestinal surgery[18,23,24], and the remaining studies showed no difference in the occurrence of bleeding between ATT-received group and the group without ATT. Although the proper management of patients undergoing antiplatelet therapy during pancreatic surgery remains controversial, preferred perioperative antithrombotic management such as continuation of preoperative aspirin monotherapy should be considered.

In the clinical setting, some institutions may be instructed to use heparin bridging during the perioperative cessation of antiplatelet agents when cardiologists judge the risk of thromboembolism as high in APT-recipient patients, which may be because most surgeons and cardiologists do not recognize the preferred option of continued aspirin monotherapy for perioperative antiplatelet management. Since bridging therapy with heparin is currently considered a strong risk factor for postoperative bleeding and should be avoided as much as possible[42], and because its mechanism of action is different from that of antiplatelet drugs, heparin bridging during cessation of antiplatelet therapy should not be used.

Regarding DOACs, only two reports were included in the current review[20]. These studies suggested that perioperative DOAC management without heparin bridging was safe and feasible even for patients who underwent hepatobiliary-pancreatic or gastrointestinal surgery; however, patients who were managed by heparin bridging during DOAC cessation are at high risk of postoperative bleeding. Currently, DOACs are increasingly prescribed for the prevention of arterial or venous thromboembolism. DOACs are fast-acting agents and their anticoagulant effects also quickly fade within 48 h after their withdrawal[35]. One large-scale multicenter prospective cohort study (the PAUSE study) was recently published, which examined outcomes in 3,007 adult patients with atrial fibrillation who underwent DOAC therapy and received an elective non-cardiac procedure or surgery[43]. DOAC therapy was interrupted 1–2 days prior and reinstituted 1–2 days after the procedure or surgery. The occurrence of major bleeding 30 days after the procedure or surgery was 0.90%–1.85%, and arterial thromboembolic complication was occurred at the rate of 0.16%–0.60%. The study recommended that a centralized perioperative management of DOACs without heparin bridging can be performed safely for patients with atrial fibrillation. Although the PAUSE study included a relatively small number of patients undergoing major gastroenterological surgery, the included studies in the current review also suggested that optimal DOAC management without heparin bridging is recommended even for patients who undergo major digestive surgeries such as hepatobiliary-pancreatic or gastrointestinal cancer surgery[20,22].

Regarding the thromboprophylaxis for VTE during pancreatic surgery, most studies in the current review have shown a potential risk of PPH in patients undergoing chemical thromboprophylaxis, but their effectiveness against VTE has not been statistically demonstrated, especially in Asian patients undergoing surgery. VTE is a fatal complication during the perioperative period, and prevention is of utmost importance. Although some Western guidelines for VTE treatment recommend chemical thromboprophylaxis during non-cardiac surgery[44-46], the racial differences in the incidence of VTE between Western people and Asians were observed[47]. Additionally, in one systematic review concerning chemical prophylaxis for VTE in Asian surgical patients[48], the risk of perioperative VTE in Asian patients is low even in the context of risk factors typically regarded as high risk. The most recently published two large-scale cohort studies (with more than 1000 patients) conducted in Japan also showed that the incidence of clinically relevant VTE during or after pancreatic surgery was 0%-0.3%[17,18]. Currently, the safety and efficacy of chemical thromboprophylaxis with anticoagulants for VTE during pancreatic surgery is still controversial, particularly in Asian populations. It is important to build evidence to classify risks individually according to race.

Pancreatic surgery in chronically ATT-burdened patients can be safely performed without an increase in the occurrence of PPH and thromboembolic complications. The safety and efficacy of chemical thromboprophylaxis with anticoagulants for VTE during pancreatic surgery is still controversial. Further investigation using reliable studies with good design is required to establish definite protocols or guidelines.