Impact of prior use or recent withdrawal of oral antiplatelet agents on acute coronary syndromes

[Is withdrawal of antiplatelet therapy necessary prior to skin cancer surgery?]

AIM

The aim of this study was to assess the value of discontinuing antiplatelet treatment prior to surgery for skin cancer with loss of skin of up to 10 cm in diameter. The study hypothesis postulated equivalence between discontinuation of antiplatelet therapy and continuation of these drugs during the perioperative period with regard to risk of cutaneous complications.

PATIENTS AND METHODS

This was a prospective, randomised study comparing two groups of patients on long-term antiplatelet treatment scheduled for surgery for skin carcinoma. In the first group, patients underwent surgery without change to their antiplatelet therapy while in the second, antiplatelet treatment was discontinued and substituted with either flurbiprofen or isocoagulant fractionated heparin or calcium heparin. Sixty patients took part in the study. Statistical analysis using ANOVA and Mann-Whitney non-parametric tests was performed to compare the following variables: age, sex, number of pre- and post-operative consultations, complications (haemorrhage, haematoma, separation and infection) and quality of wound healing.

RESULTS

No statistically significant difference was seen between the two groups regarding either complications or quality of wound healing.

CONCLUSION

There appear to be no rational grounds for discontinuing antiplatelet therapy in advance of skin cancer surgery involving loss of skin of up to 10 cm in diameter.

[Spinal cord ischaemia and preoperative clopidogrel withdrawal in an arteriosclerotic patient]

We report the case of a motor impairment associated with bladder dysfunction several days after clopidogrel withdrawal in an arteriosclerotic woman scheduled for thoracotomy under general and thoracic epidural anaesthesia. Even if spinal artery syndrome may have a lot of aetiologies, we believe in a direct link between clopidogrel withdrawal and medulla ischaemia.

Acute thrombosis of a mesenteric artery drug-eluting stent following clopidogrel cessation

LEARNING OBJECTIVE

To describe thrombosis of sirolimus-coated mesenteric arterial stents following cessation of clopidogrel therapy.

BACKGROUND

Cardiac drug-eluting stent thrombosis following cessation of antiplatelet therapy with clopidogrel has been associated with increased mortality. The application of such stents in the mesenteric arterial system and the subsequent need for clopidogrel therapy has not been studied. This is the first case report of acute thrombosis of a drug-coated stent in the mesenteric circulation.

CLINICAL FINDINGS

Acute mesenteric ischemia secondary to thrombosis of a mesenteric arterial stent following clopidogrel cessation is described.

CONCLUSION

Drug-eluting stents represent an option for mesenteric revascularization in the surgically complicated abdomen. As in the setting of cardiac stenting, acute thrombosis of these devices following cessation of clopidogrel therapy is a concern. Indefinite clopidogrel therapy following deployment of drug-coated stents should be considered.

Failure of aspirin to prevent myocardial infarction and adverse outcome during follow-up - a large series of all-comers

BACKGROUND

Aspirin has been used for secondary prevention of myocardial infarction (MI) in individuals with coronary disease. Although supported by several large controlled trials, aspirin prevents only a portion of recurrent events.

AIM AND METHOD

We set out to study the prevalence of long-term aspirin use prior to admission for MI and its significance for medium-term event-free survival; 998 consecutive patients with acute MI admitted to a tertiary center were included in the study.

RESULTS

Nearly half (42.4%) of all patients reported long-term use of low-dose aspirin prior to the index event. Prior aspirin use was associated with a 50% increase in the risk of both the combined end point of recurrent unstable angina, recurrent myocardial infarction, stroke, or death (OR 1.49; 95% CI 1.12-2.00, P=0.006) and mortality (OR 1.50; 95% CI 1.03-2.17, P=0.03) during 10-month follow-up. Prior aspirin use was not associated with an increased frequency of added antithrombotic therapy at discharge.

CONCLUSIONS

We have found that prior aspirin use is common in patients hospitalized for acute MI. Individuals already on aspirin had increased risk of recurrent ischemic events and all-cause mortality during the 10-month follow-up after their index MI.

Perioperative Management of Patients Undergoing Non-cardiac Vascular Surgery

Patients undergoing non-cardiac vascular surgery have arterial disease affecting more than one vascular bed and commonly have multiple significant co-morbidities. The surgical and anaesthetic teams are asked to address pre-, peri- and postoperative management issues relating not only to the surgery but arising from these co-morbidities. Here we review the strategies and rationale for the optimisation of these high risk patients.

[Management of antiplatelet therapy: a case report]

We report a personal case of myocardial infarction with flurbiprofen used instead of clopidogrel before an operation. We re-examined the literature concerning the management of the antiagregant treatment of the patient with angioplasty and/or stent who needs surgery. This management proves to be delicate and all the questions that it raises have not so far been answered in particular with regard to the problem of resistance to aspirin. Anyway this personal observation supports our view that contrary to certain recomendations flurbiprofène is not a satisfactory alternative to aspirin either with or without thiénopyridine.

[Perioperative therapy with antiplatelet agents]

Antiplatelet therapy is used to prevent thromboembolic complications in patients with artherothrombotic disease. In clinical routine an assessment of the risk of perioperative bleeding must be weighed against the risk of thromboembolic complications in the event of discontinuation of antiplatelet agents. The indications as well as the timepoint for the termination of antiplatelet therapy are important issues to consider. The premature interruption of oral antiplatelet therapy with recovery of platelet function may lead to recurrence of thromboembolic events. Especially patients with intracoronary stents are at threat when the dual platelet therapy is interrupted before complete endothelialization of the stent.

Perioperative antiplatelet therapy: the case for continuing therapy in patients at risk of myocardial infarction

Recent clinical data show that the risk of coronary thrombosis after antiplatelet drugs withdrawal is much higher than that of surgical bleeding if they are continued. In secondary prevention, aspirin is a lifelong therapy and should never be stopped. Clopidogrel is regarded as mandatory until the coronary stents are fully endothelialized, which takes 3 months for bare metal stents, but up to 1 yr for drug-eluting stents. Therefore, interruption of antiplatelet therapy 10 days before surgery should be revised. After reviewing the data on the use of antiplatelet drugs in cardiology and in surgery, we propose an algorithm for the management of patients, based on the risk of myocardial ischaemia and death compared with that of bleeding, for different types of surgery. Even if large prospective studies with a high degree of evidence are still lacking on different antiplatelet regimens during non-cardiac surgery, we propose that, apart from low coronary risk situations, patients on antiplatelet drugs should continue their treatment throughout surgery, except when bleeding might occur in a closed space. A therapeutic bridge with shorter-acting antiplatelet drugs may be considered.

Aspirin and bleeding in dentistry: an update and recommendations

Aspirin use in the United States remains high because of its diverse and beneficial activities. In adults at risk for cardiovascular thrombotic events, low-dose aspirin is an excellent preventive agent; however, its antiplatelet properties have contributed to a perceived increased risk for bleeding after dental extractions. This article discusses recent evidence regarding the thrombotic risks associated with discontinuing aspirin use in patients who take aspirin daily and presents a new recommendation for continuing low-dose aspirin during invasive dental procedures.

Very late coronary stent graft thrombosis after aspirin cessation

We report a case of late stent thrombosis two years after implantation of a coronary stent graft (Jomed Covered Stent, CSG) and a week after cessation of aspirin therapy due to circumcision. Caution in adherence to anti-thrombotic therapy is warranted years following implantation of coronary stent grafts.

A tale of two stents: perioperative management of patients with drug-eluting coronary stents

Drug-eluting stents were introduced into clinical practice to decrease coronary stent restenosis rates. Though remarkably effective in reducing this complication, recent data reveal that drug-eluting stents pose a significant risk for late stent thrombosis, an event strongly correlated with discontinuation of anti-platelet therapy. Because anti-platelet agents are often discontinued perioperatively, patients with DES are at risk for perioperative stent thrombosis and myocardial infarction. Along with a review of the recent literature, we present two cases of patients with drug-eluting stents scheduled for renal transplantation. Two distinct antithrombotic management strategies illustrate the risk of either approach-bleeding and transfusion versus stent thrombosis and myocardial infarction.

Coronary artery stents and non-cardiac surgery

The utility of interventional cardiology has developed significantly over the last two decades with the introduction of coronary angioplasty and stenting, with the associated antiplatelet medications. Acute coronary stent occlusion carries a high morbidity and mortality, and the adoption of therapeutic strategies for prophylaxis against stent thrombosis has major implications for surgeons and anaesthetists involved in the management of these patients in the perioperative period. Currently, there is limited published information to guide the clinician in the optimal care of patients who have had coronary stents inserted when they present for non-cardiac surgery. This review examines the available literature on the perioperative management of these patients. A number of key issues are identified: the role of surgery vs percutaneous coronary intervention for coronary revascularization in the preoperative period; the different types of coronary stents currently available; the emerging issues related to drug-eluting stents; the pathophysiology of coronary stent occlusion; and the recommended antiplatelet regimes that the patient with a coronary stent will be receiving. The role of preoperative platelet function testing is also discussed, and the various available tests are listed. Appropriate management by all the clinicians involved with patients with coronary stents undergoing a variety of non-cardiac surgical procedures is essential to avoid a high incidence of postoperative cardiac mortality and morbidity. The review examines the evidence available for the perioperative strategies aimed at reducing adverse outcomes in a number of different clinical scenarios.

Perioperative use of anti-platelet drugs

Performing a surgical procedure on a patient undergoing anti-platelet therapy raises a dilemma: is it safer to withdraw the drugs and reduce the haemorrhagic risk, or to maintain them and reduce the risk of myocardial ischaemic events? Based on recent clinical data, this review concludes that the risk of coronary thrombosis on anti-platelet drugs withdrawal is much higher than the risk of surgical bleeding when maintaining them. In secondary prevention, aspirin is a lifelong therapy and should never be stopped. Clopidogrel is mandatory as long as the coronary stents are not fully endothelialized, which takes 6-24 weeks depending on the technique used, but might be required for a longer period.

Drug‐eluting coronary stents — a note of caution

There are two types of coronary stents: bare-metal stents (BMS) that cost about $800 each, and drug-eluting stents (DES) that cost about $3,300 each. DES reduce the rate of restenosis but have a higher incidence of late stent thrombosis, particularly if dual antiplatelet therapy with aspirin and clopidogrel is interrupted. Stent thrombosis has a myocardial infarction rate of 70% and a mortality rate of 31%-45%. Randomised studies of BMS versus DES show no increase in myocardial infarction or death with DES in simple coronary lesions, but in clinical practice, DES are mainly used in complex coronary disease where the rate of stent thrombosis is higher. Registry data suggest an increased rate of death and myocardial infarction of 0.5%-1.0% per annum with DES. Clinicians need to be aware of the risks associated with prematurely ceasing dual antiplatelet therapy in patients with DES.

[Coronary stents and anaesthesia: it is time to have national data]

We report 13 cases of coronary stent patients, undergoing a non cardiac surgery. Despite an heterogenous perioperative management of antiplatelet agents, none of these patients developed any significant complications. Recently, several case reports of postoperative drug eluting stent thrombosis have been reported. However, the actual incidence of this dramatic event is not known. This confirms the need to perform prospective studies or registries of patients with coronary stents undergoing non cardiac surgery, in order to propose evidence-based recommendations on perioperative antiplatelet management in such patients.

Late thrombosis of drug-eluting stents: a meta-analysis of randomized clinical trials

PURPOSE

Drug-eluting stents are commonly used for percutaneous coronary intervention. Despite excellent clinical efficacy, the association between drug-eluting stents and the risk for late thrombosis remains imprecisely defined.

METHODS

We performed a meta-analysis on 14 contemporary clinical trials that randomized 6675 patients to drug-eluting stents (paclitaxel or sirolimus) compared with bare metal stents. Eight of these trials have reported more than a year of clinical follow-up.

RESULTS

The incidence of very late thrombosis (>1 year after the index procedure) was 5.0 events per 1000 drug-eluting stent patients, with no events in bare metal stent patients (risk ratio [RR]=5.02, 95% confidence interval [CI], 1.29 to 19.52, P=.02). Among sirolimus trials, the incidence of very late thrombosis was 3.6 events per 1000 sirolimus stent patients, with no events in bare metal stent patients (RR=3.99, 95% CI, .45 to 35.62, P=.22). The median time of late sirolimus stent thrombosis was 15.5 months, whereas with bare metal stents it was 4 months. Among paclitaxel trials, the incidence of very late thrombosis was 5.9 events per 1000 paclitaxel stent patients, with no events in bare metal stent patients (RR=5.72, 95% CI, 1.08 to 32.45, P=.049). The median time of late paclitaxel stent thrombosis was 18 months, whereas it was 3.5 months in bare metal stent patients.

CONCLUSIONS

Although the incidence of very late stent thrombosis more than 1 year after coronary revascularization is low, drug-eluting stents appear to increase the risk for late thrombosis. Although more of this risk was seen with paclitaxel stents, it remains possible that sirolimus stents similarly increase the risk for late thrombosis compared with bare metal stents.

Late clinical events after clopidogrel discontinuation may limit the benefit of drug-eluting stents: an observational study of drug-eluting versus bare-metal stents

OBJECTIVES

We sought to define the incidence of late clinical events and late stent thrombosis in patients treated with drug-eluting (DES) versus bare-metal stents (BMS) after the discontinuation of clopidogrel as well as their timing and outcome.

BACKGROUND

There is growing concern that delayed endothelialization after DES implantation may lead to late stent thrombosis and related myocardial infarction (MI) or death. However, event rates and outcomes after clopidogrel discontinuation versus BMS are unknown.

METHODS

A consecutive series of 746 nonselected patients with 1,133 stented lesions surviving 6 months without major events were followed for 1 year after the discontinuation of clopidogrel. Patients were assigned randomly 2:1 to DES versus BMS in BASKET (Basel Stent Kosten Effektivitäts Trial). The primary focus of this observation was cardiac death/MI.

RESULTS

Rates of 18-month cardiac death/MI were not different between DES and BMS patients. However, after the discontinuation of clopidogrel (between months 7 and 18), these events occurred in 4.9% after DES versus 1.3% after BMS implantation. Target vessel revascularization remained lower after DES, resulting in similar rates of all clinical events for this time period (DES 9.3%, BMS 7.9%). Documented late stent thrombosis and related death/target vessel MI were twice as frequent after DES versus BMS (2.6% vs. 1.3%). Thrombosis-related events occurred between 15 and 362 days after the discontinuation of clopidogrel, presenting as MI or death in 88%.

CONCLUSIONS

After the discontinuation of clopidogrel, the benefit of DES in reducing target vessel revascularization is maintained but has to be balanced against an increase in late cardiac death or nonfatal MI, possibly related to late stent thrombosis.

Interférence des médicaments cardiovasculaires avec l'anesthésie et la chirurgie

Decisions about chronic treatments during the perioperative period must be made at the presurgical anesthesia consultation. It is increasingly rare to stop treatment during this period, because: This new rule is applied particularly to patients with cardiovascular disorders. Beta blockers have a special role in preventing the onset of postoperative cardiovascular events. The role of statins requires further precision but they appear to fit into the same preventive approach. Interruption of antiplatelet agents appears to be associated with a risk of arterial thrombosis in patients with coronary conditions, notably those with conventional stents and most especially those with drug-eluting stents.

[Antithrombotic drugs in urological surgery. Treatment interfering with haemostasis in urology?]

In storing or withholding a treatment that interferes haemostasis should be considered after assessing the benefit-risk ratio which requires the understanding of the efficacy, safety and pharmacology of heamostatic agents. This benefit to risk assessment is the key stone for the prevention of venous thrombosis, and peri-operative management of antiptatelet agents and vitamin K antagonists.

Prior use of antithrombotic agents and neurological functional outcome at discharge in patients with ischemic stroke

BACKGROUND

Studies in experimental animals have suggested that antithrombotic agents may have a neuroprotective effect after an ischemic injury. The aim of this study was to analyze the effect of prior use of antithrombotic agents (antiplatelets or anticoagulants) on neurological functional outcome in patients with acute ischemic stroke.

SUBJECTS AND METHODS

Consecutive patients included in the Perugia Stroke Registry were considered for this analysis. Neurological functional outcome was evaluated at discharge using the modified Rankin Scale (mRS >or= 3 disabling stroke).

RESULTS

Of the 1921 patients included in the analysis (mean age 76.3 +/- 12.5 years; 53% males), 662 (34.5%) were on antithrombotic treatment (581 antiplatelets, 71 anticoagulants and 10 antiplatelets associated with anticoagulants). One hundred and twenty-two patients (6.4%) died in hospital; at discharge 712 patients (37.1%) were disabled and 1,087 patients (56.6%) were non-disabled. Fifty-four (44.3%) of the deceased patients and 270 (37.9%) of disabled patients were on antithrombotic treatment, while 338 (31.1%) non-disabled patients were taking antithrombotic agents. From multivariate analysis, age and stroke severity were associated with an adverse outcome. Male gender, dyslipidemia, stroke due to small vessel disease and no history of previous stroke were associated with an improved outcome, while no correlation was found between prior use of antithrombotic agents and outcome (mortality odds ratio; OR = 1.32, 95% confidence interval; CI 0.85-2.04; P = 0.20, mortality or disability OR = 0.95, 95% CI 0.72-1.25; P = 0.80).

CONCLUSION

Prior use of antithrombotic agents does not improve the functional outcome in patients with acute ischemic stroke.

Paradoxical rebound platelet activation after painkillers cessation: missing risk for vascular events?

BACKGROUND

Several reliable reports strongly indicate that the use of nonsteroidal anti-inflammatory drugs (NSAIDs) and cyclooxygenase-2 (COX-2) inhibitors is associated with an increased risk of cardiovascular events. Considering the key role of platelets in coronary atherosclerosis and the fact that antiplatelet therapy with aspirin (and more recently, clopidogrel) has been associated with reduced vascular mortality, we sought to determine the effect of therapy and withdrawal of NSAIDs and COX-2 inhibitors on platelet activity.

METHODS

Platelet characteristics from 34 aspirin-naive volunteers who were receiving NSAIDs and COX-2 inhibitors were compared with 138 drug-free controls. Platelets were assessed twice at baseline (at least 1 month of NSAIDs and COX-2 inhibitors) and after a 14-day washout. We used adenosine diphosphate-induced conventional aggregometry, the point-of-care Ultegra analyzer (Ultegra Accumetrics, San Diego, Calif), and whole blood flow cytometry.

RESULTS

Platelet activity during therapy with NSAIDs and COX-2 inhibitors was similar and unremarkable between groups. However, there was a highly significant increase of platelet activity as assessed by conventional aggregometry (P=.0003), Ultegra analyzer readings (P=.03), and expression of GPIIb/IIIa (P=.02), P-selectin (P=.03), and platelet endothelial cell adhesion molecule-1 (P=.001) after withdrawal from NSAIDs and COX-2 inhibitors.

CONCLUSIONS

These data suggest that drug cessation, rather than continuous therapy with NSAIDs and COX-2 inhibitors, may be associated with rebound platelet activation, which may predispose one to a higher risk of vascular events. This hypothesis requires intensive testing in crossover randomized studies and may justify more aggressive antiplatelet regimens in patients after discontinuation of therapy with NSAIDs and COX-2 inhibitors.

Clopidogrel withdrawal is associated with proinflammatory and prothrombotic effects in patients with diabetes and coronary artery disease

Inhibition of the P2Y12 pathway by the platelet antagonist clopidogrel is associated with a marked reduction in platelet reactivity. Recent reports have shown that P2Y12 inhibition has anti-inflammatory effects as well. However, whether clopidogrel withdrawal is associated with proaggregatory and proinflammatory effects has not yet been explored. Since diabetic subjects are characterized by a prothrombotic and proinflammatory status, we hypothesize that these patients may be more vulnerable to these effects. A total 54 patients with diabetes on long-term (12 months) dual antiplatelet therapy (aspirin plus clopidogrel) were studied. Platelet aggregation (following 6 and 20 micromol/l ADP stimuli) and inflammatory markers (C-reactive protein and P-selectin expression) were assessed before and 1 month following clopidogrel withdrawal. Following clopidogrel withdrawal, aspirin responsiveness using platelet function analyzer-100 was determined as well. A significant increase in all the assessed platelet (P < 0.0001 for 6 and 20 micromol/l ADP-induced aggregation) and inflammatory (P < 0.05 for C-reactive protein, P < 0.001 for P-selectin expression in resting platelets, and P < 0.0001 for P-selectin expression in ADP-stimulated platelets) biomarkers was observed following clopidogrel withdrawal. Low responders to aspirin had increased platelet aggregation profiles (P < 0.05 for 6 and 20 micromol/l ADP-induced aggregation) but no differences in inflammatory markers. In conclusion, clopidogrel withdrawal is associated with an increase in platelet and inflammatory biomarkers in diabetic patients, supporting pleiotropic effects coupled with P2Y12 receptor antagonism.

Anesthésie et réanimation pour chirurgie réglée de l'anévrisme de l'aorte abdominale

OBJECTIVES

Patient scheduled for infrarenal abdominal aortic aneurysm surgery carries a high risk of cardiac or respiratory comorbidity. To outline the perioperative management for these patients.

METHODS

Review of the literature using MesH Terms "abdominal aortic aneurysm", "anesthesia", "analgesia" "critical care" and/or "surgery" in Medline database.

RESULTS

Cardiac preoperative evaluation and management have recently been reviewed. Intermediate and high-risk patients should undergo non-invasive cardiac testing to decide between a preoperative medical strategy (using betablocker+/-statin and aspirin) and an interventional strategy (coronary angioplasty or cardiac surgery). Perioperative myocardial ischaemia should also be investigated by clinical, electrocardiographic and biologic monitoring such as plasmatic troponin Ic dosage. Specific score could also assess the respiratory failure risk preoperatively. Epidural analgesia decreases this risk. There is no evidence that a pharmacological treatment decreases the incidence of acute renal failure after aortic surgery. Endovascular repair is actually recommended for older, higher-risk patients or patients with a hostile abdomen or other technical factors that may complicate standard open repair.

Conduite à tenir concernant le traitement médicamenteux des patients adressés pour chirurgie programmée

This review focuses on potential drug interactions between anaesthetic drugs or techniques and chronic medications in patients scheduled for surgery. The vast majority of therapeutics can be continued until the morning of surgery. However, for some drugs such as ACE inhibitors, there is strong evidence to recommend their discontinuation prior to surgery. When juged necessary, interruption of chronic therapeutic needs to be anticipated and planned. In the other hand, for other drugs such as beta-blockers or L-Dopa, acute withdrawal is associated with documented adverse outcome. As a result, such drugs have to be continuing throughout the operative period. Although a general consensus exists for many medications, there are still controverses as to the management of antithrombotic drugs and some central nervous system agents. Advances in anaesthesia include knowledge on the mechanisms involved in drug interactions, which allows us to improve the preoperative management of chronic therapeutics.

Thrombosis of sirolimus-eluting coronary stent in the postanesthesia care unit

A 44-yr-old woman with a drug-eluting coronary stent placement two weeks before surgery suffered a myocardial infarction in the postanesthesia care unit immediately after hysterectomy. She had missed only one dose of aspirin and clopidogrel preoperatively. Early recognition of subacute stent thrombosis and urgent percutaneous coronary intervention probably prevented her death. In this case report, we highlight perioperative coronary stent issues and discuss their implications.

Coronary syndromes following aspirin withdrawal: a special risk for late stent thrombosis

OBJECTIVES

We sought to determine whether aspirin withdrawal is an encountered situation in coronary disease patients who relapsed.

BACKGROUND

Despite the recognized benefits of aspirin in coronary disease, and because of the threat of bleeding or poor compliance, aspirin intake is sometimes stopped. It is not known whether withdrawal of aspirin can be harmful in coronary-disease patients.

METHODS

Between September 1999 and April 2002, a total of 1,236 patients hospitalized for acute coronary syndrome (ACS) were questioned in order to determine whether aspirin intake had been interrupted.

RESULTS

Fifty-one of these ACSs occurred within 1 month after aspirin withdrawal. This represents 4.1% of all coronary events but 13.3% of recurrences. Among those patients who relapsed, the incidence of ST-segment elevation ACS was higher in those who stopped aspirin when compared to the 332 patients who did not stop aspirin (39% vs. 18%; p = 0.001). Ten (20%) cases involved a thrombosis of an uncoated stent implanted on average 15.5 +/- 6.5 months previously. Mean delay between aspirin withdrawal and the acute coronary event was 10 +/- 1.9 days. Reasons for aspirin withdrawal included minor surgery in 7 cases, fibroscopy in 8 cases, dental treatment in 13 cases, bleeding in 3 cases, and patient non-compliance in 20 cases.

CONCLUSIONS

Our results support the hypothesis that aspirin withdrawal in coronary patients may represent a real risk for the occurrence of a new coronary event. Many cases involved late uncoated-stent thrombosis. Assessment of the exact incidence of coronary recurrences after aspirin withdrawal will need prospective studies.