We conducted a focused review of the current guidelines related to medical management and risks of anticoagulation therapy. We then performed a comprehensive review of the literature related to thoracic surgery and anticoagulation therapy. Literature searches were conducted in the PubMed database using the key words: anticoagulation, deep venous thrombosis, pulmonary embolism, atrial fibrillation, drug-eluting stent, heparin, warfarin, antiplatelet therapy, clopidogrel, pulmonary resection, lobectomy, thoracic surgery, noncardiac thoracic surgery. Searches were limited to the English language, human subjects, and literature published in the last 5 years. Our search returned 400 articles; we critically reviewed 45 articles related to thoracic surgery as well as national guidelines from the American College of Chest Physicians (ACCP) and the American Heart Association (AHA). Emphasis was made on current national guidelines and recommendations.

Patients undergoing thoracic surgery are considered at least at moderate risk for VTE. In studies of patients undergoing thoracotomy for pulmonary resection, VTE occurred in approximately 1.7 % of patients and pulmonary embolism occurred in 1.2 % of patients despite perioperative prophylaxis [1]. Patients undergoing extended pulmonary resections or pneumonectomy are at even higher VTE risk, with reported rates as high as 7.4 % [2, 3].

Data evaluating the use of VTE prophylaxis in patients undergoing thoracic surgery are limited. Two small trials evaluating the use of perioperative VTE prophylaxis have were published over 20 years ago. The first paper compared the used of differing doses (5,000 units twice daily versus 7,500 units twice daily) of unfractionated heparin (UFH) while the second paper compared the use of UFH with the use of nadroparin [4, 5]. The remaining data on the topic were retrospective reviews, mostly evaluating the use of VTE prophylaxis in patients undergoing extrapleural pneumonectomy, who are considered high risk for VTE (Table 9.1). Mason et al. reported the results of their review of 336 patients undergoing pneumonectomy for malignancy who were all treated with twice daily dosing of UFH with intermittent compression devices [2]. Five percent of patients were diagnosed with deep venous thrombosis (DVT) while 2.3 % were diagnosed with a pulmonary embolism. Gomez-Hernandez et al. published their retrospective review of over 6,000 patients undergoing elective thoracic surgery [10]. In their study, they followed the ACCP guidelines for VTE risk stratification, prophylaxis, and diagnosis. Patients in whom VTE prophylaxis was recommended received the first dose of low molecular weight heparin (LMWH) prior to surgery with continuation post-operatively without delay. They reported a DVT rate of 0.18 % and pulmonary embolism rate of 0.11 %. Interestingly, three of seven patients who were diagnosed with a pulmonary embolism had a right pneumonectomy. The remaining studies compared LMWH with warfarin, LMWH with fondapurinox, or UFH with LMWH and found no significant differences in treatment [6–9]. The risk of post-operative bleeding was mentioned in only one of these studies and also demonstrated no significant difference in bleeding between treatment groups. Therefore, due to the limited data, VTE prophylaxis guidelines in the setting of thoracic surgery have been inferred from data in patients undergoing general or abdomino-pelvic surgery.

Patients present to the thoracic surgeon while taking VKAs for several known and frequently unknown indications. It is crucial to ascertain the true indication for treatment with VKAs in order to appropriately manage perioperative anticoagulation. The most common indications for VKA therapy are chronic atrial fibrillation, pulmonary embolism, and presence of a mechanical heart valve. We address each of these clinical scenarios individually. In managing the use of VKAs perioperatively, one must consider each individual patient’s risk for thrombosis (particularly in the prothrombotic state induced by surgery) with the patient’s risk for bleeding. Most of the data addressing this topic are from studies in patients in a non-operative setting. Therefore, guidelines on the perioperative management of VKAs are inferred from data in non-operative patients and based on risks of thrombus formation.

Approximately 600,000 patients undergo coronary artery interventions with either bare metal or drug eluting stents in the United States each year [14]. These patients are maintained on an appropriate antiplatelet regimen, which typically includes aspirin and clopidogrel (Plavix), for a defined period of time. The 2011 American College of Cardiology Foundation (ACCF)/American Heart Association (AHA) guidelines for percutaneous coronary intervention recommend aspirin use indefinitely, and P2Y12 inhibition with clopidogrel, or in certain circumstances, prasugrel or ticagrelor, for a minimum of 1 month. Patients should continue antiplatelet therapy with a P2Y12 inhibitor up to 12 months following implantation of a bare metal stent (BMS). Patients undergoing drug eluting stent (DES) implantation require treatment with an aspirin and P2Y12 inhibition for a minimum of 12 months.

Up to 5 % of patients undergoing percutaneous coronary intervention undergo surgery within the first year after coronary stenting [15]. Moreover, the perioperative morbidity and mortality in patients undergoing surgical procedures following percutaneous coronary intervention (PCI) with stenting are as high as 40 and 20 %, respectively [16, 17]. The risk of early discontinuation of antiplatelet therapy is severe. Discontinuation of clopidogrel therapy within 1 month of DES placement carried a 25 % stent thrombosis rate in one series [18]. In a large observational cohort study of more than 2,000 patients, thrombosis of DES occurred in 29 % of patients in whom antiplatelet therapy was discontinued prematurely (<3 months after sirolimus-eluting stents and <6 months after paclitaxel-eluting stents), with a mortality rate of 45 % [19].

Data on patients with DES <12 months old and BMS <1 month, or high risk patients with either, undergoing thoracic surgery is limited to small case series (Table 9.2). The increased thrombotic risk following surgery is well described. Two small series of 32 and 33 patients demonstrated major adverse cardiac events at rates of 9.3 % and 36 % [20, 21], respectively, when clopidogrel was held 5–7 days prior to thoracic surgery. A recent review of the SEER (Surveillance Epidemiology and End Results) Medicare data by Fernandez et al. demonstrated a similar rate of 9.3 % major adverse cardiac events versus 4.9 % in controls in over 500 patients who underwent major lung resection within 1 year following the placement of a BMS [24].