Coronary Stents and Noncardiac Surgery: Current Clinical Challenges and Conundrums


Address for correspondence: Jagtar S. Heir, DO, Assistant Professor, Department of Anesthesiology and Perioperative Medicine, The University of Texas M. D. Anderson Cancer Center, Houston, TX


This article discusses how perioperative physicians are currently faced with unique challenges when providing care for surgical patients who have undergone recent percutaneous transluminal coronary angioplasty with drug-eluting stent (DES) placement. Despite adhering to the currently recommended antiplatelet regimens, these patients may still be at risk for sustaining a myocardial infarction secondary to stent thrombosis during the perioperative period. Given the high morbidity and mortality rates associated with perioperative myocardial infarction, it is important that the perioperative care of surgical patients with previous DES placement should be discussed, evaluated, and clarified by all practitioners who may be involved in their care.

Prev Cardiol. 2010;13:8–13.©2009 Wiley Periodicals, Inc.

The introduction of percutaneous transluminal coronary angioplasty (PTCA) for the treatment of coronary artery disease (CAD) was initially met with great enthusiasm as early studies have indicated that long-term survival in the interventional group (PTCA) was similar to the surgical group (coronary artery bypass grafting). However, the patients in the bypass group had better symptom control (angina) while patients in the PTCA group required more frequent revascularization procedures. Restenosis of the opened vessel and the need for repeated revascularization in the interventional group were addressed with the advent of metal intracoronary stents (bare metal stents [BMS]). However, it was soon noted that the use of BMS was also associated with a restenosis rate of 20% to 30% in the stented region secondary to the vascular injury. Therefore, drug-eluting stents (DES) were subsequently introduced into practice after animal and clinical studies reported a dramatic reduction in the incidence of stent restenosis when compared with BMS’.1,2 DES are impregnated with compounds (eg, paclitaxel, sirolimus) aimed at suppressing neointimal proliferation, which is the main cause of stent stenosis. Unfortunately, DES also delay re-endothelialization, thereby increasing the risk of stent thrombosis especially within the proinflammatory–prothrombotic milieu associated with the perioperative stress response. The transition from BMS to DES occurred very rapidly in the nonsurgical population and soon expanded to unapproved indications. However, patients with DES now present unique challenges in terms of their preoperative management. We report an unusual case of immediate postoperative myocardial infarction (MI) in a patient who had 2 drug-eluting stents placed more than 3 years ago. This case is unusual because the patient sustained a postoperative MI despite having been on dual antiplatelet therapy for more than 3 years.

Case Report

Our patient was a 69-year-old man with prostate cancer who was scheduled to undergo robotic laparoscopic prostatectomy. His medical history was significant for hypertension, hyperlipidemia, and CAD for which he had undergone angioplasty with DES placement. The DES (paclitaxel) had been placed in both the left anterior descending artery (LAD) and right coronary artery (RCA), respectively, more than 3 years prior to his prostatectomy. Post–coronary stent placement, the patient was started on aspirin and clopidogrel as part of his dual antiplatelet regimen to prevent stent thrombosis. The patient stated that he had been compliant with this regimen since the placement of his intracoronary stents. Since the coronary revascularization, the patient reported no chest pain or shortness of breath and stated that he had improved exercise tolerance.

Preoperatively, the patient was seen by his private cardiologist for cardiac evaluation, perioperative risk assessment, and medical optimization prior to his scheduled prostatectomy. Given the significant time period the patient had been taking a dual-antiplatelet regimen, it was decided to discontinue clopidogrel 7 days prior to surgery, as well as the aspirin 4 days prior to surgery. The patient was also taking simvastatin, which was discontinued preoperatively.

The patient received general anesthesia as his primary anesthetic with the standard American Society of Anesthesiologists monitors as well as an arterial line for the purpose of close monitoring of the patient’s hemodynamics. The surgical procedure was uncomplicated and the patient’s hemodynamic status was extremely stable throughout the procedure. Approximately 5 minutes after his admission to the post–anesthesia care unit, the patient started to complain of chest pain. Given the patient’s significant cardiac history, 12-lead electrocardiography was performed. The findings revealed ST elevation in leads II, III, and AVF confirming the diagnosis of an ST-elevation MI (STEMI) most likely in the RCA territory. The patient was immediately put on 100% oxygen; his heart rate was controlled with metoprolol. Additionally, the patient was administered a 325-mg aspirin tablet and started on nitroglycerin. Furthermore, cardiac enzymes were drawn, which soon demonstrated an enzyme leak and elevated troponin level. The patient was immediately seen by a cardiologist and was transferred to the intensive care unit for closer monitoring. Because stent thrombosis was the most likely etiology for his MI, we had the patient expeditiously transferred to the catheterization laboratory so that the stents could be recanalized.

Cardiac catheterization showed diffuse CAD with some stenosis of the LAD and obtuse marginal branch of the circumflex artery. In addition, he was found to have a thrombosis of his RCA stent. The stent thrombosis of the RCA was considered to be the culprit lesion and the patient underwent successful PTCA of this vessel with no complications.

After PTCA, the patient was transferred to the intensive care unit. Although the patient had undergone prostatic surgery the same day as his cardiac catheterization, considering the risk of re-thrombosis, the patient was started on both aspirin and a heparin infusion. Because the risk of bleeding was also high, the patient’s physical examination, hemodynamics, serial hematocrit values, and drainage outputs were all closely monitored. Eventually the patient was transitioned to oral medications and discharged from the hospital on a β-blocker, statin, aspirin, and clopidgrel.


Although the patient made a full recovery and was discharged home, this case report highlights the serious consequences of routine discontinuation of antiplatelet medications during the perioperative period in a patient with previously placed DES. Furthermore, despite careful planning, the physicians involved in the care of this patient were lulled into a sense of complacency given the patient’s long history of having been on dual antiplatelet therapy. Therefore, they did not adhere to the current science advisory guidelines and discontinued both antiplatelet agents, which led the patient to sustaining a perioperative stent thrombosis and STEMI.

The risk of perioperative stent thrombosis has been well documented in previous work. For instance, in a recent prospective observational cohort study of 2229 patients who had DES placement, Iakovou and colleagues3 reported that the incidence of subacute or late thrombosis was 1.3% and that the case fatality rate among patients who developed stent thrombosis was an alarming 45%. The review suggested that premature discontinuation of antiplatelet therapy, renal failure, bifurcation lesions, diabetes, and lower ejection fraction are the most significant contributory factors for subacute or late thrombosis. Additionally, the science advisory guidelines mentioned advanced age, multiple lesions, overlapping stents, small vessels, and prior brachytherapy as predictors of DES thrombosis.4 Of note, our patient had multiple stents.

Premature discontinuation of antiplatelet therapy is the most critical predictor of stent thrombosis3 (Figure 1). As duration of optimal antiplatelet therapy after DES placement has continued to evolve, we emphasize the recent American Heart Association/American College of Cardiology/Society for Cardiovascular Angiography and Intervention/American College of Surgeons/American Dental Association science advisory recommendation of increasing dual antiplatelet therapy for 12 months after patients have had a DES.4 Whether clopidogrel would be needed after a year would depend on the physician’s judgment, taking into consideration the risk-benefit ratio for each individual patient. This, therefore, highlights the importance and consideration for the continuation of antiplatelet therapy during the perioperative period to prevent stent thrombosis in high-risk patients. Even patients adhering to the recommended antiplatelet regimen may still be prone to thrombotic complications due to inadequate response to the antiplatelet medications as a result of genetic polymorphism, receptor up-regulation, underdosing, and drug interactions.5,6 Consequently, this emphasizes the need to identify certain subsets of patients who need to continue dual antiplatelet therapy for even a longer period than the current practice guidelines. Furthermore, our patient had a paclitaxel-eluting stent, and given the fact that the elution of the drug in paclitaxel stents can be for an indefinite period of time,7 it is not unreasonable to consider keeping these patients on dual antiplatelets for possibly a longer duration than even currently recommended.

Figure 1.

 Effect of thienopyridine interruption on mortality of noncardiac surgery within 3 weeks after percutaneous coronary intervention (PCI). Reprinted with permission from Satler LF. Recommendations regarding stent selection in relation to the timing of noncardiac surgery postpercutaneous coronary intervention. J Catheter Cardiov Interv. 2004;63:146–147.

If the surgery cannot be postponed or deferred in a patient who recently had a DES placed, as is frequently the case with cancer patients, continuation of aspirin throughout the entire perioperative period should be strongly considered (Figure 2). However, continuing antiplatelet therapy during the perioperative period may increase bleeding complications. To date, most of the published literature on the effects of continued antiplatelet medication during noncardiac surgery suggests an increased risk of bleeding, although randomized controlled trials are lacking.8–10 Chu and colleagues11 and Chen and colleagues12 have independently reported that patients taking dual antiplatelet therapy who undergo elective coronary artery bypass grafting surgery have more clinical bleeding and subsequently more blood transfusions than patients not taking antiplatelet therapy. Conversely, Shim and colleagues 13 and Kennedy and colleagues 14 independently reported that perioperative use of aspirin did not lead to increased transfusion requirements in patients undergoing off-pump coronary artery bypass graft and hip replacement procedures. In their meta-analysis of perioperative aspirin use and bleeding complications, Burger and colleagues8 looked at 41 studies that included 49,590 patients. They reported that aspirin did increase the risk of bleeding by 1.5 times but did not lead to more severe bleeding complications. Thus, they concluded that aspirin should only be discontinued if the risks of bleeding and the consequences thereof were at least “comparable” to the cardiovascular risks associated with the withdrawal of antiplatelet therapy.8

Figure 2.

 Proposed approach to the management of patients with previous percutaneous coronary intervention (PCI) who require noncardiac surgery, based on expert opinion. Reprinted with permission from Fisher et al. ACC/AHA guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2007;116(17):1971–1996.

Although the debate about when and whether to continue antiplatelet therapy in surgical patients will go on and continue to challenge physicians providing perioperative care, the benefits of long-term dual antiplatelet therapy has been elegantly demonstrated by Steinhubl and colleagues.15 These authors reported that dual antiplatelet therapy with aspirin and clopidogrel is associated with a 27% relative reduction in the combined risk of death, acute MI, and stroke. However, balancing the risk of stent thrombosis by deviating from the recommended antiplatelet regimen with the hemorrhagic consequences in the perioperative period is a double-edged sword.16 In that context, it is reasonable to choose monotherapy with aspirin alone in the perioperative period in the case of a patient with a DES unless the patient is scheduled for a neurosurgical or retinal procedure where even minor bleeding could be catastrophic. This is despite the fact that dual antiplatelet therapy is superior to monotherapy in the prevention of stent thrombosis.17,18 Alternatively, a group from Wake Forest 19 has advocated the use of glycoprotein IIb/IIIa platelet inhibitors and heparin as bridging therapy to help reduce the incidence of perioperative stent thrombosis in patients with DES. Although this novel idea seems appealing because it utilizes the short half-life of both drugs, it actually places an enormous cost burden on both the patient and the hospital. This regimen has not been sufficiently studied in controlled trials so as to be recommended based on scientific evidence. Furthermore, both heparin and glycoprotein IIb/IIIa platelet inhibitors have not been shown to be consistently effective in reducing the incidence of stent thrombosis in patients who discontinued antiplatelet medications.18,20

Despite the risk of thrombosis, many patients with coronary stents are advised to discontinue all antiplatelet medications in the preoperative period. However, it is clear that recent aspirin withdrawal is a risk factor for acute coronary syndrome and is associated with higher 30-day rates of acute MI and death.21 In instances where antiplatelet therapy was discontinued in the preoperative period, efforts should be made to reinitiate antiplatelet therapy as soon as possible in order to reduce the incidence of perioperative stent thrombosis. In the Clopidogrel for the Reduction of Events During Observation (CREDO) trial, all patients randomized to clopidrogrel pretreatment demonstrated a 3.0% absolute reduction (P=.02) in the composite of death, MI, and stroke after a year.15 Therefore, thought should be given to preloading high-risk patients with 300-mg clopidogrel in the immediate postoperative period, since this strategy not only reduces the time to the maximal inhibition of platelets but also may reduce the hyper-responsiveness sometimes seen in patients with activated platelets.22

Patients with gastrointestinal cancers and solid tumors have abnormalities in all the components of Virchow’s triad (abnormalities in blood flow, blood constituents, and vessel wall). For example, many of these cancers produce procoagulant substances (such as tissue factor), induce increased platelet activity and turnover, reduce the activity of thrombolytic factors (eg, plasminogen activating inhibitor), and display abnormal tumor angiogenesis, all of which lead to a prothrombotic state.23,24 Similarly, a damaged and or dysfunctional endothelium due to adjuvant chemoradiation also significantly contributes to a hypercoagulable state.25 Furthermore, the procoagulant, proinflammatory effects of the surgical stress response adds to the above-mentioned factors to predispose the surgical cancer patient to an even higher risk for procoagulant and thrombotic complications. Therefore, it is conceivable that the diagnosis of cancer played a pivotal role in development of STEMI in our patient. An additional factor for the occurrence of perioperative MI in our patient could be the abrupt preoperative withdrawal of long-term statin therapy. Abrupt perioperative discontinuation of statin therapy is considered an important contributor to postoperative cardiac morbidity and mortality. The proposed mechanism for the adverse effects of statin withdrawal is an overshoot translocation and activation of Rho causing downregulation of e-NOS production below baseline values. These combined effects are proposed to lead to both a decrease in the vasodilatory mechanisms and expression of increased markers of inflammation and oxidative stress.26,27 Although numerous other publications have addressed the various contributing factors of perioperative MI in the stent population, to date no studies or case reports have addressed the predicament posed by coronary stents and perioperative MI in surgical cancer patients.

In their meta-analysis, Bavry and colleagues 28 showed that there was a 4- to 5-fold increased risk of late thrombosis with DES relative to BMS (Figure 3). Other multiple reports addressing the same issue have reached similar conclusions.29–31 Therefore, given the currently available stent technology (sirolimus, everolimus, and paclitaxel) and literature on late-stent thrombosis it seems prudent to recommend BMS instead of DES for patients requiring coronary revascularization and who are known to require surgery in the near future. Additionally, patient compliance is unfortunately often poor, whether from financial constraints or inadequate patient education. Therefore, this fact underscores the need for patient education and suggests that at least some patients may benefit from receiving a BMS rather than a DES, as this obviates the need for prolonged dual antiplatelet therapy. Alternatively, newer stents that use nonpermanent scaffolding, such as bioabsorbable magnesium stents, might be a viable option in the near future. These stents utilize a magnesium alloy that appears to have minimum inflammatory changes, complete absorption within 2 months, and rapid endothelization and low neointima proliferation, thus demonstrating a promising outlook for future intracoronary interventions.32

Figure 3.

 Temporal sequences of re-endothelialization in bare metal stents (BMS) and drug-eluting stents (DES). Reprinted with permission from Head DE et al. A tale of two stents: perioperative management of patients with drug-eluting coronary stents. J Clin Anesth. 2007;19(5):386–396.

Along those lines, Brilakis and colleagues 33 reported 3 adverse events (1 perioperative death and 2 nonfatal perioperative MIs) among 350 patients who underwent balloon angioplasty as the sole means of coronary revascularization prior to noncardiac surgery. This amounted to an adverse event rate of 0.9%, which is lower than patients who had stenting prior to noncardiac surgery (3.9%–32%). Therefore, it can be argued that balloon angioplasty alone is a reasonable option for the patient requiring coronary revascularization who may need a surgical procedure in the very near future.

Physicians are frequently faced with the situation of a patient presenting to the preoperative anesthesia assessment center or the operating room with the history of having had a DES. Although we can discuss the appropriateness of preoperative coronary intervention vs medical management and the potential benefits of BMS vs DES, we rarely have a chance to influence that decision since the patients present to us just before the surgical procedure. Clearly, routine preoperative discontinuation of all antiplatelet medications in patients who have received recent DES has to be questioned. Current literature suggests that we need to rethink our perioperative strategy of managing patients with coronary stents. Particular caution and emphasis need to be placed on reducing the risks for perioperative stent thrombosis and balancing those concerns with hemorrhagic complications. For instance, certain institutions have implemented protocols on the perioperative management of DES; namely, they have made arrangements to have the patient emergently transferred to the catheterization laboratory if the suspicion of DES thrombosis is considered high. In this manner, the strategy is to reduce the delay in door-to-(balloon) inflation time to be <90 minutes. Furthermore, these patients are not considered for neuraxial blocks, since the management becomes quite complicated given the fact these patients may require anticoagulation status post-diagnosis of a thrombosed stent.

In addition, further studies need to be conducted on the risk-benefit ratio of continuing statin therapy during the perioperative period. Patients with metabolically active cancers can have a predisposition to a prothrombotic state in the perioperative period. Therefore, we need to lead the discussion on education, awareness and perioperative strategy in managing these patients so that our concerns for stent thrombosis can be understood by all care providers. We also need to better inform our surgical colleagues that the early discontinuation of antiplatelet therapy may lead to 29% of patients getting acute perioperative stent thrombosis, for which the mortality rate ranges from 20% to 45%.3,4 In addition, we propose that a multidisciplinary board of experts should convene to help stratify the risk/benefits of continuing perioperative antiplatelet therapy use with that of stent thrombosis in different surgical procedures. For example, for low-risk procedures such as colonoscopies, patients need not stop their antiplatelet therapy, whereas for neurosurgical procedures, in which bleeding would be significantly detrimental, patients should discontinue all antiplatelet therapy. Therefore, it is our belief that only through education, awareness, and cooperation among the different disciplines can the catastrophic consequences of coronary stent thrombosis be prevented.