When not deferrable surgery is indicated in a patient on antithrombotic therapy, the overall management is complex. Ideally, antithrombotic therapy should be on the one hand interrupted to prevent perioperative bleeding complications and, on the other hand, continued to prevent perioperative ischemic complications, which may be even more harmful [2, 3]. Therefore, stratification of both the risk of bleeding associated with surgery and the risk of thrombosis and/or thromboembolism associated with the interruption and/or modification of antithrombotic therapy should be carried out [2, 3]. Further complexity to the perioperative management of antithrombotic therapy is added when it comprises both OAC and antiplatelet agents. For practical purposes, the inherent risks associated with the management of the two classes of antithrombotic agents should be evaluated (and managed) separately.

There is a high mortality rate (up to 8 %) associated with postoperative bleeding in patients undergoing surgery while on OAC with warfarin (as a result of blood transfusion, possible wound infection, occasional need for reoperation, and delayed resumption of antithrombotic therapy) [4]. Therefore, only when bleeding risk is low, it is advisable to perform surgery while being on effective OAC (i.e., INR ≥ 2.0) [3, 5] (Table 9.3). In patients at intermediate or high bleeding risk, timely interruption and/or reversal of OAC (depending on whether it is an elective of emergent procedure) seems reasonable [3, 5] (Table 9.4). In the elective setting, consideration should also be given to whether bridging anticoagulation with low-molecular-weight heparin (LMWH) may be required during interruption of OAC [2, 3]. Because of the increase in major bleeding complications, with no associated benefit on the incidence of thromboembolic events, perioperative bridging therapy with LMWH in patients with AF undergoing invasive procedures [2, 3] should generally not be performed. Possible exceptions may include a very high risk of stroke (such as a CHA₂DS₂-VASc score ≥ 6) and/or a history of previous stroke [3], given that interruption of OAC with warfarin has indeed been proven not to be without risks [6]. An algorithm of warfarin interruption for elective surgery is provided in Fig. 9.3 [2]. In the urgent/emergency setting, there is no time for warfarin interruption, and therefore reversal of OAC with the specific antidote vitamin K and/or nonspecific reversal agents [6] (Table 9.4) should be considered when the estimated risk of bleeding of surgery is high (or even intermediate) [2].

The perioperative management of antiplatelet therapy in patients with coronary artery disease is also complex and depends on the indication for and intensity of antiplatelet therapy. Of note, a 1.5 increase in the risk of bleeding has been reported for surgery on ongoing aspirin therapy compared to no aspirin [7], whereas that of dual antiplatelet therapy (DAPT) with aspirin and clopidogrel may be 3.4 times higher than with aspirin alone [8]. A 4 % and 21 % absolute rate of severe bleeding with ongoing single and DAPT, respectively, have been reported within 30 days of noncardiac surgery [9]. In addition to that, consideration to perform surgery on DAPT precludes the use of locoregional anesthesiological techniques (namely, neuroaxial), which are currently preferred due to their greater ability to lower sympathetic stimulation and give better control of perioperative pain [10]. Of note, perioperative withdrawal of aspirin in patients at risk of, or with proven, coronary artery disease increases the risk of adverse cardiac events by a factor of three [3]. General recommendations for the perioperative management of antiplatelet therapy in patients with coronary artery disease are given in Table 9.5 [11].

Given that every year > 1 million PCI (with stent implantation in approximately 85 % of cases) are performed in Europe and the USA and that 4–8 % of these patients undergo surgery within 1 year and 25 % within 5 years [12, 13], the management of antiplatelet therapy in this setting is an issue of great relevance. Early interruption of antiplatelet therapy is in fact the most potent predictor of stent thrombosis [14], which is in turn, associated with a rate of myocardial infarction and death of 50–70 % and 20–40 %, respectively [14]. In addition, surgery leads to an inflammatory, hypercoagulable, and hypoxic condition, which may result in plaque instability and perioperative arterial thrombosis [13]. These factors contribute to increased rates of adverse cardiac events, including death myocardial infarction and need for revascularization, when surgery is performed early after stent implantation, especially within the first month [15, 16].