Article Text

Download PDFPDF

Clinical pharmacology
Antiplatelet therapy during surgery
  1. Helge Möllmann,
  2. Holger M Nef,
  3. Christian W Hamm
  1. Department of Cardiology, Kerckhoff Heart Center, Bad Nauheim, Germany
  1. Correspondence to Dr Helge Möllmann, Department of Cardiology, Kerckhoff Heart Center, Benekestrasse 2-8, 61231 Bad Nauheim, Germany; h.moellmann{at}

Statistics from

Antiplatelet agents are one of the cornerstones of primary and secondary prevention of stable and unstable coronary artery disease (CAD), and are especially important after percutaneous coronary interventions (PCIs).1 More than two million PCIs are performed annually in western Europe and North America. Currently, over 90% of all PCIs involve the placement of coronary stents, and drug eluting stents are used for a variety of indications. The prevalence of CAD and thereby also of acute coronary syndromes (ACS) is expected to increase dramatically in the next decade due to the progressive ageing of the population.

Stent thromboses are rare but potentially fatal complications after coronary stent implantation.2 Different factors have been identified that independently correlate with the risk for stent thromboses, such as PCI in small vessels or coronary bifurcations and comorbidities like diabetes mellitus or renal failure.2 3 The most important risk factor, however, is premature cessation of dual antiplatelet therapy with aspirin and a thienopyridine.2 The latter has gained in importance considerably in recent years with the use of drug eluting stents, since when increased rates of late stent thromboses (<30 days after implantation) have been reported.

Recent reports suggest that 5% of patients who underwent PCI will also have to undergo non-cardiac or cardiac surgery within the first year after coronary stenting.4 This subset of patients presents cardiologists, surgeons, and anaesthetists with the problem of having to decide between the risk of increased blood loss when continuing the antiplatelet agents in the perioperative period, and the risk of stent thrombosis if the drugs are stopped abruptly and prematurely.

This article aims to balance the individual risks of these patients and to offer a practical approach for daily clinical practice.

Aspirin and clopidogrel: pharmacological mechanisms and indications

Aspirin exerts its antiplatelet effects by irreversibly acetylating serine 529 of cyclooxygenase-1 (COX-1), resulting in inhibition of thromboxane A2 (TXA2). TXA2 plays a pivotal role in the final phase of platelet aggregation. Inhibition of COX-1 is irreversible in platelets due to the lack of nuclei, and therefore lasts during the platelet's circulating lifespan (around 7 days).5 Several trials examined the required dosage of aspirin that exerts a sufficient antiplatelet effect on the one side and minimises bleeding risks on the other. The initial loading dose should be between 300–500 mg followed by 75–162 mg daily for long term treatment.6

Clopidogrel acts by covalently binding to a cysteine residue of the purinergic adenosine diphospate (ADP) receptor P2Y12 and thereby attenuates platelet aggregation in response to ADP released from activated platelets. As with aspirin, the clopidogrel induced inhibition of platelets is irreversible during the platelet's lifespan. Clopidogrel is a prodrug and requires hepatic metabolisation to become active.5 It is recommended to apply a loading dose of 300–600 mg in order to achieve a more rapid and pronounced antiplatelet effect, followed by 75 mg daily for long term medication.1 7

Both drugs are recommended for a broad spectrum of cardiovascular diseases to reduce thrombotic events. Whereas aspirin is usually prescribed as long term or even life long therapy, the duration of clopidogrel medication depends on the underlying disease if not given as an alternative to aspirin. Important indications and recommended treatment durations are summarised in table 1.1 7

Table 1

Important indications for antiplatelet therapy according to current guidelines

Bleeding risk

Dual antiplatelet therapy increases the risk of spontaneous severe bleeding events by around one third.6 8 In reference to its propensity to aggravate surgical bleeding, it has been referred to as the ‘surgeon's headache’. Consequently, in the past the fear of excessive bleeding led to the generally accepted policy of withdrawing antiplatelet drugs 5–10 days before surgical procedures, despite the fact that such a policy goes against current guidelines on antithrombotic therapy.

Unfortunately, studies on perioperative haemorrhagic risk of antiplatelet therapy, although numerous, are mostly statistically underpowered and there are only few if any large prospective randomised trials. These have been performed mainly in orthopaedic (hip replacement) and cardiac surgery (coronary artery bypass grafting (CABG)). Nevertheless, the analysis of these studies reveals interesting facts about the impact of aspirin and clopidogrel on surgical outcome.


A large meta-analysis on the impact of low dose aspirin on surgical blood loss revealed that patients on aspirin alone have on average an increase in perioperative bleeding complications of 50%. These bleeding complications, however, did not translate into an increase in mortality or morbidity.9 For vascular surgery, an increase in bleeding complications of only 2.46% has been described.10 Aspirin seems to be associated with a rise in bleeding only in specific procedures. A meta-analysis of 50 studies examining cardiac surgery procedures showed an average increase in blood loss of 300 ml per patient,11 which certainly does not outbalance the beneficial effects of aspirin in this subset of patients. In patients undergoing intracranial neurosurgery, perioperative aspirin treatment seems to be associated with an increased risk of postoperative intracerebral haematoma. However, this risk has been a contributing factor for a fatal outcome only in sporadic cases.12 In a double blind, placebo controlled study examining the impact of aspirin treatment in patients undergoing carotid artery surgery, surgeons were not able to differentiate patients on aspirin from patients on placebo based on the observed intraoperative bleeding.13

Dual antiplatelet therapy

It would be expected that the addition of clopidogrel to aspirin would increase surgical haemorrhage. Indeed, blood loss is effectively further increased on average by 30–50%, but most of the studies have been performed in cardiac surgery with full intraoperative heparinisation for cardiopulmonary bypass. In a study with around 400 matched pairs, dual antiplatelet therapy during the last 4 days before CABG was an independent predictor of transfusion requirement (odds ratio (OR) 4.2, 95% confidence interval (CI) 1.79 to 9.95),14 and need for redo surgery (OR 4.9, 95% CI 2.63 to 8.97).15 Nevertheless, surgical outcome and operative mortality were not different from patients without dual antiplatelet therapy. A large prospective observational study on 1628 consecutive patients in whom the same surgical and anaesthesia team performed CABG surgery, Karabulut et al failed to find any association between the perioperative use of clopidogrel and significantly increased bleeding, transfusion requirements, need for surgical re-exploration or intensive care unit stay.16 Likewise, a subgroup analysis of the CURE trial examining patients who underwent CABG with (n=1011) or without (n=1061) clopidogrel and aspirin found a minor, but not significant, difference in the incidence of major or life threatening bleeding events.17

The effect of clopidogrel in non-cardiac surgery is evaluated only in a few smaller studies. In patients undergoing vascular, orthopaedic or visceral surgery shortly after coronary stenting, the need for blood transfusions was 42.6% in patients on dual antiplatelet therapy and 38.5% in patients on aspirin alone. This difference, which was not statistically significant, was most probably due to the rather small study population of 207 patients.18 A further study showed that in patients who underwent non-cardiac surgery and were treated with aspirin and clopidogrel during the first year after coronary stent insertion, there was no major morbidity from bleeding. Smaller clinical series or case reports in visceral or vascular surgery have shown a clear trend towards an increased surgical blood loss and transfusion rate in patients treated with dual antiplatelet therapy, but failed to demonstrate a significant impact on mortality.

As with aspirin alone, there are some specific procedures such as intracranial neurosurgery in which catastrophic events can be anticipated when performed under dual antiplatelet therapy—an assumption precluding appropriate studies.

Reduction of thrombotic events

In several landmark studies, both aspirin and clopidogrel have been proven to cause a substantial reduction in cardiovascular events and mortality after ACS and after PCI.6 These properties have resulted in class I recommendations for dual antiplatelet therapy for up to 1 year or even life-long in the respective guidelines of the European Society of Cardiology.1 7

For aspirin this recommendation is based on the meta-analysis of the Antithrombotic Trialists' Collaboration that showed a 46% reduction in the rate of vascular events.6 Numerous trials demonstrated the advantage of dual antiplatelet therapy after routine coronary stent implantation. In the CURE trial a significant risk reduction for death from vascular causes, non-fatal myocardial infarctions or stroke was observed for patients treated with dual antiplatelet therapy after ACS.8

The increased risk of surgery early after coronary stenting was first described in 2000: eight of 25 patients undergoing non-cardiac surgery within the first 2 weeks after coronary stenting died, mostly due to acute myocardial infarction caused by stent thrombosis.19 Although further and larger trials could not reproduce this dramatic rate of fatalities in the setting of surgery early after percutaneous interventions, it is still well documented that these patients are faced with an especially high risk. Patients undergoing non-cardiac surgery during the early post-interventional period have a risk for myocardial infarction and mortality 5–10 times higher than matched patients undergoing the same operation under maximal medical treatment or after appropriate delay.20 These findings underline the importance of continuous dual antiplatelet therapy until the implanted stent is properly re-endothelialised. It is widely accepted that the re-endothelialisation takes 4–6 weeks after bare metal stents and 6–12 months or longer after the implantation of drug eluting stents.7

Acute withdrawal of antiplatelet agents is associated with a progressive recovery of platelet function and may thereby produce a deleterious rebound, with prothrombotic effects overcoming the physiological balance. Excessive TXA2 production and decreased fibrinolytic activity have been noted after aspirin treatment is stopped.

In the clinical setting, this rebound effect was documented in a study of 2229 patients having received a drug eluting stent. Premature clopidogrel cessation was the most significant independent predictor of stent thrombosis with a high hazard ratio of 57.13 (p<0.001) and a mortality rate linked to stent thrombosis of 45%.2 Two thirds of late drug eluting stent thromboses were linked to early discontinuation of clopidogrel. Patients who stopped the drug within the first month after stent drug eluting stent implantation had a 10 times higher risk of death (7.5% vs 0.7%, p<0.0001) during the following 11 months compared with patients who had taken clopidogrel continuously.21 The question of whether late stent thromboses occur more frequently in drug eluting stents in comparison to bare metal stents is not fully clarified yet. Nonetheless, the safety of drug eluting stents largely depends on consequent treatment with dual antiplatelet therapy during the first year.

The interruption of aspirin alone is also associated with a higher risk and may lead to stent thrombosis even more than 1 year after percutaneous intervention. A large meta-analysis including over 50 000 patients treated with aspirin for secondary prevention of CAD showed that the cardiac complication rate was increased threefold after aspirin withdrawal (OR 3.14, p<0.0001). This risk was even considerably higher after coronary stent implantation (OR 89.78, p<0.0001).22 Aspirin interruption accounts for around 5% of all ACS. The average time delay between stopping aspirin and the acute coronary event has been reported to be 8.5 days.23 This delay corresponds well to platelet half-life and is therefore a further strong argument for the importance of an uninterrupted aspirin therapy.

The consequences

The above mentioned data clearly demonstrate the necessity to reconsider the widely used approach to generally withdraw antiplatelet drugs before all surgical procedures. The relative risks—increased bleeding rates in the case of maintaining antiplatelet drugs, and the risk of cardiac events in the case of interrupting the medication—have to be carefully balanced:

On the one hand, maintaining antiplatelet therapy during surgery leads to an average increase in haemorrhagic risk of 2.5–20% with aspirin alone, or 30–50% in case of dual antiplatelet therapy. However, no study has yet demonstrated an increase in surgical morbidity or mortality linked to the increased blood loss (with the exception of intracranial neurosurgery), although the rate of blood transfusions is increased by one third. The complication rate of red blood cell transfusion has been calculated to be only 0.4%, with all types of complications included.24

On the other hand, there are well documented specific risks of prematurely withdrawing antiplatelet agents, which leads to a rebound effect with increased platelet adhesiveness. The average postoperative myocardial infarction rate due to stent thrombosis during the re-endothelialisation phase of coronary stents due to stent thrombosis is 35%. This is associated with an average mortality of 20–45%.2 19 Taken together, it appears that the risks of prematurely withdrawing antiplatelet agents before each and every surgical intervention is generally higher than those of maintaining this vital medication at the cost of increased haemorrhage.25

As a consequence, in a science advisory from the American Heart Association/American College of Cardiology (AHA/ACC) and guidelines of the American College of Chest Physicians (ACCP) the potential sequelae of premature cessation of antiplatelet drugs in the perioperative period are clearly characterised. Accordingly, a multidisciplinary approach has been proposed for the management of patients on dual antiplatelet therapy facing cardiac or non-cardiac surgery.

‘How to’ strategy

Two distinct scenarios can be distinguished. First, the patient is deemed a candidate for percutaneous revascularisation but already knows that he/she has to undergo surgery in the near future. Second, the patient undergoes stent implantation, not suspecting that a surgical intervention within the first year after PCI will be necessary.

Planned surgery

Of course, the first scenario with already anticipated surgery is considerably easier to handle, given the broader spectrum of potential treatment regimens. It should be critically discussed whether preoperative revascularisation procedures can be avoided and, if not, which type of revascularisation may be suitable in this setting. A considerable number of patients do not benefit from preoperative revascularisation procedures at all—even with angiographically diagnosed CAD. This has been clearly documented in the CARP (Coronary Artery Revascularization Prophylaxis) trial that enrolled 510 patients with stable CAD facing major vascular surgery. Patients were randomised to either preoperative revascularisation or no revascularisation. Revascularisation was carried out with both modalities, CABG (40%) and PCI (60%). Both groups—revascularisation or no revascularisation—had identical incidences of perioperative myocardial infarction (8.4% vs 8.4%, p=NS) and survival after a median follow-up of 27 months (78% vs 77%).26 These results suggest that preoperative revascularisation may not be mandatory unless the patients present with unstable coronary syndromes or other high risk characteristics. However, preoperative revascularisation is often regarded as being imperative by both cardiologists and anaesthetists in patients who have had an ACS or after detection of ischaemic regions in stress tests.

Although contemporary practice during PCIs includes placement of stents, a sole balloon angioplasty may be safely performed in this subgroup of patients. This approach has been tested in a trial, which enrolled 350 patients undergoing non-cardiac surgery within 2 months after a primarily successful balloon angioplasty. Only one patient (0.3%) died in the perioperative period, and two patients (0.6%) suffered a myocardial infarction. These promising results propose that this approach may be safer than (drug eluting) stent placement before planned surgery. Of course, these patients should be advised to be regularly seen by a cardiologist or to undergo a second angiography after surgery to allow eventually for optimal treatment of the coronary lesion, given the considerably high restenosis rate after balloon angioplasty alone in comparison to stent implantation.

Sometimes stenting cannot or appears not appropriate to be avoided during PCI. In this case the selection of stent type greatly influences the optimal timing for anticipated surgery. If surgery has to be performed within 1 year after stent placement, then bare metal stents are preferable to drug eluting stents, owing to their favourable earlier re-endothelialisation and thus the reduced risk of stent thrombosis. This is of particular relevance if dual antiplatelet therapy cannot be continued during the perioperative period. If surgery is elective and can safely be postponed for more than 1 year, nothing argues against implantation of drug eluting stents. In every case it is most important to aim for the best possible procedural result with optimally deployed stents without unnecessary overlapping, in order to minimise the risk of stent thrombosis. Intravascular ultrasound may be a valuable tool in these patients.

Unplanned surgery

A completely different—and by far more difficult to handle—situation is found in the setting of unplanned surgery in patients who have had an ACS and/or who have been treated with PCI, and are therefore in need of antiplatelet drug therapy. In these patients the decision between continuing antiplatelet therapy or withdrawing it prematurely has to be made on an individual basis. Nonetheless, we propose an algorithm to help guide cardiologists, surgeons and anaesthetists through this difficult decision making process (figures 1 and 2).

Figure 1

Algorithm for perioperative management of patients receiving antiplatelet therapy with aspirin. Cardiological risk assessment according to table 2. High bleeding risk includes, for example, intracranial neurosurgery or spinal canal surgery.

Figure 2

Algorithm for perioperative management of patients receiving antiplatelet therapy with aspirin and clopidogrel. Cardiological risk assessment according to table 2. High bleeding risk includes, for example, intracranial neurosurgery or spinal canal surgery.

Patients on aspirin

Aspirin is often prescribed for primary prevention in coronary high risk patients. There are no studies providing evidence that its interruption may be harmful in this subset of patients. Therefore, in this situation aspirin may be safely withdrawn during the perioperative period. With respect to platelet half-life, aspirin should be stopped 5–7 days before surgery and restarted as soon as possible after postoperative bleeding can be excluded. In patients receiving aspirin as secondary prevention following stroke, ACS or any type of revascularisation, this treatment is lifelong and should therefore not be stopped before surgery. Surgery with an extensively high risk of bleeding in a closed space such as intracranial neurosurgery or spinal canal surgery is a clear exception. In these cases, in which even small postoperative haemorrhage may have catastrophic effects, preoperative withdrawal of aspirin given for secondary prevention is most probably associated with a smaller risk than maintaining it (figure 1).

Patients on clopidogrel and aspirin

Patients with a need for dual antiplatelet therapy with clopidogrel and aspirin should be divided into high risk and low risk groups (table 2). Low risk in this setting includes an interval of >6 weeks after bare metal stent implantation, stroke, or uncomplicated myocardial infarction. In this group clopidogrel should be stopped 5–7 days before surgery and restarted as soon as possible with a de novo loading dose of 600 mg. Since aspirin acts as secondary prevention, it should be given continuously. High risk situations comprise an interval of <6 weeks after myocardial infarction, percutaneous intervention, and stroke or <12 months after implantation of a drug eluting stent. In general, surgery should be postponed in these patients unless it is vital. Although it is well known that surgery in patients being treatment with clopidogrel and aspirin significantly increases perioperative bleeding and complications, it still appears safer to continue dual antiplatelet therapy during most surgical procedures which cannot be postponed. Of course, there are several exceptions such as neurosurgery, for example. If intracranial surgery or surgery in the medullary canal has to be performed, clopidogrel should be stopped before surgery and restarted with a loading dose as soon as perioperative bleeding can be excluded. However, in these cases, in which it is deemed mandatory to pause aspirin and clopidogrel treatment, surgery should only be performed in centres with the possibility of immediate PCI in order to minimise the risk of potential stent thromboses (figure 2).

Table 2

Characteristics of patients being at high or low risk for suffering stent thromboses after withdrawal of antiplatelet drugs

Patients on prasugrel and aspirin

The recently launched new antiplatelet drug prasugrel, which is usually prescribed in combination with aspirin, has proven to result in a better outcome for patients with ACS in comparison to the combination of clopidogrel and aspirin. However, the reduction of thrombotic events has been achieved—at least in part—at the cost of further increased bleeding complications. Since prasugrel treatment led to a significant increase in bleeding complications in patients undergoing urgent CABG, it can be anticipated that an equally increased bleeding rate will occur also in other surgical procedures.27 Therefore, a potentially acceptable compromise in cases of mandatory surgical procedures may be to replace prasugrel during the perioperative period with clopidogrel.

Alternative approaches

There are several approaches which attempt to overcome the drawbacks of perioperative antiplatelet therapy. Some authors, for example, have proposed a bridging with low molecular weight heparins during the perioperative period. However, this approach is not supported by studies—on the contrary, in a small observational study, a bridging with low molecular weight heparins was associated with an increased cumulative blood loss.28

Another approach may be bridging with glycoprotein IIb/IIIa inhibitors. Tirofiban and eptifibatide have the advantage of a considerably shorter half-life, given their reversible mode of action, and can therefore be given safely until a few hours before surgery. Until now, however, no studies have been performed proving the theoretical advantage of this approach.

A major advantage in this context can be anticipated for the next generation of antiplatelet drugs, such as cangrelor or ticagrelor. Due to their reversibility, they can be withdrawn shortly before surgery, thereby minimising both the interval of suboptimal platelet inhibition and the rate of perioperative bleeding. Both compounds are currently being tested in large randomised studies and are expected to be launched soon.

A completely different approach is to test for re-endothelialisation after coronary stent implantation by optical coherence tomography, before the patient undergoes surgical procedures. Once complete re-endothelialisation after stenting has been achieved and can be unequivocally demonstrated, dual antiplatelet therapy can most probably be safely withdrawn during the perioperative period. As with the other alternative approaches mentioned above, this attractive method needs further evidence through controlled clinical trials.


Patients requiring antiplatelet therapy who have to undergo surgical interventions are at high risk. If the antiplatelet therapy is withdrawn, they face an increased risk of thrombotic cardiovascular events. If dual antiplatelet therapy is continued during the perioperative phase, they face an increased rate of bleeding complications. Therefore, these patients deserve special interdisciplinary attention and anaesthetists, surgeons, and cardiologists should collaborate closely in order to identify the best approach to take for each patient on an individual basis.

Antiplatelet therapy during surgery: key points

  • Antiplatelets drugs are a cornerstone in the primary and secondary prevention of coronary artery disease. They are especially important after coronary stent implantation and acute coronary syndromes.

  • About 5% of patients who undergo percutaneous coronary intervention will have to undergo surgery within the first year. In these patients the problem arises of how to manage antiplatelet therapy in the perioperative period.

  • A continuation of these antiplatelet drugs in order to reduce cardiovascular events is associated with increased perioperative bleeding.

  • Withdrawing antiplatelet agents in order to reduce surgical haemorrhage leads to a significant increase in cardiovascular morbidity and mortality, especially in patients who have received drug eluting stents.

  • There is only very limited evidence from controlled trials to be used for guidelines.

You can get CPD/CME credits for Education in Heart

Education in Heart articles are accredited by both the UK Royal College of Physicians (London) and the European Board for Accreditation in Cardiology—you need to answer the accompanying multiple choice questions (MCQs). To access the questions, click on BMJ Learning: Take this module on BMJ Learning from the content box at the top right and bottom left of the online article. For more information please go to:

Please note: The MCQs are hosted on BMJ Learning—the best available learning website for medical professionals from the BMJ Group. If prompted, subscribers must sign into Heart with their journal's username and password. All users must also complete a one-time registration on BMJ Learning and subsequently log in (with a BMJ Learning username and password) on every visit.


  1. Important study examining risk factors for stent thromboses in patients who have received drug eluting stents.

  2. Trial highlighting the risk for patients undergoing surgery shortly after stent implantation.

  3. Landmark study on the beneficial effects of dual antiplatelet therapy in patients with acute coronary syndromes.

  4. Article on the problem of perioperative blood loss in patients treated with dual antiplatelet therapy.

  5. First study reporting the high risk of stent thromboses when surgery is performed early after stent implantation.

  6. The CARP trial investigated the effect of preoperative coronary artery revascularisation in comparison to a conservative approach.

View Abstract


  • Competing interests In compliance with EBAC/EACCME guidelines, all authors participating in Education in Heart have disclosed potential conflicts of interest that might cause a bias in the article. The authors have no competing interests.

  • Provenance and peer review Commissioned; not externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.