News Daily News

Details on Clopidogrel Discontinuation Remain Elusive, Unclear

By L.A. McKeown

Download this article's factoid (PDF & PPT for Gold subscribers) 

Data on cardiovascular events associated with discontinuation of clopidogrel for surgery or other reasons are conflicting and questionable. In addition, it is unclear whether there is a time threshold for discontinuation, after which event rates are negligible, according to a review published online July 30, 2011, ahead of print in the European Heart Journal.

Michael A. Gaglia Jr, MD, MSc, and Ron Waksman, MD, of Washington Hospital Center (Washington, DC), searched the medical literature from January 2000 through March 2010 for articles reporting adverse clinical events following discontinuation of clopidogrel. The review included:

  • 13 nonrandomized studies of perioperative cessation of clopidogrel or ticlopidine
  • 2 randomized trials comparing different durations of clopidogrel therapy
  • 16 nonrandomized studies of premature clopidogrel discontinuation without reference to surgery

Perioperative Cessation Appears Hazardous

For patients undergoing noncardiac surgery, the 13 studies documenting perioperative thienopyridine cessation were mostly observational, but suggest a hazard for adverse cardiac events. Overall in-hospital mortality ranged from 4.9% to 20%. Perioperative stent thrombosis rates ranged from 2.0% in a study of 481 patients to 2 in 5 patients who discontinued clopidogrel in a 56-patient case series. Similarly, MACE incidence ranged from 0 in a study of 38 patients (41% of whom were taking clopidogrel at the time) to 30.7% in 192 patients who discontinued clopidogrel and underwent surgery less than 30 days after PCI.

In the 1 study that included multivariable analysis, MACE risk was increased within 30 days of cardiac or noncardiac surgery in a cohort of 481 patients who had received BMS or DES a mean of 1.1 years earlier, but neither aspirin nor clopidogrel within 24 hours before surgery was protective for these patients.

Drs. Gaglia and Waksman said it is “readily apparent” that there are few data to guide perioperative management of patients on clopidogrel, “with only small nonrandomized case series available that are almost invariably without adequate multivariable adjustment.”

Other obstacles include inconsistent and incomplete reporting of details regarding clopidogrel/aspirin, cessation, incidence of stent thrombosis, and suggested strategies for restarting clopidogrel. The available data also do not differentiate between DES and BMS patients, Drs. Gaglia and Waksman add, nor do they help clarify whether there is a safe window for surgery and/or discontinuation of clopidogrel.

Despite the ambiguity, guidelines issued by the European Society of Cardiology in 2009 on the timing of noncardiac surgery in patients who have undergone PCI recommend a minimum of 6 weeks, but optimally 3 months for BMS patients and a minimum of 12 months for DES patients. The American College of Cardiology and American Heart Association have not issued formal guidelines about perioperative discontinuation of clopidogrel, although a science advisory was issued in 2007 cautioning against premature cessation in general.

Drs. Gaglia and Waksman say the guidelines have been issued primarily “in order to bridge the considerable gap between evidence and clinical practice.”

Ambiguity Surrounding Premature Discontinuation

The studies regarding premature discontinuation of clopidogrel without reference to surgery were primarily retrospective and statistically flawed. Like the surgery studies, these studies are not uniform in terms of the proportion of patients undergoing DES and BMS.

In an early report from 2004, the incidence of stent thrombosis was 1.1% in a population of 652 patients with sirolimus-eluting stents (SES) and over half (57%) of these were in patients who had discontinued clopidogrel. In addition, all stent thromboses occurred within 13 days of PCI. But in the DIABETES trial, published in 2007, the stent thrombosis rate was 3.8% when clopidogrel was discontinued at 1 year after SES implantation. And among the 746 patients enrolled in the BASKET-LATE trial who received BMS or DES and stopped clopidogrel at 6 months, the stent thrombosis risk was higher in the DES group (2.6% vs. 1.6%).

The few retrospective studies on clopidogrel cessation have not found any effect on stent thrombosis. In one, the study authors conclude that among 15,157 patients with SES, despite the occurrence of 126 stent thromboses, clopidogrel compliance had no effect on risk. Another study of 8,146 patients who received DES, meanwhile, found the absence of clopidogrel was not a risk factor for stent thrombosis. Furthermore, a study of 10,778 patients with SES showed that discontinuation of dual antiplatelet therapy, but not thienopyridines alone, increased risk. A landmark analysis of patients who were event-free at 6 months did not show a benefit from thienopyridine therapy beyond this time point.

Then there is the issue of how premature discontinuation of clopidogrel is defined among these various studies. According to Drs. Gaglia and Waksman, while some studies used cut points of 1 year, 6 months, 3 months, or 30 days, others reported the cessation time point as patient-dependent or simply did not provide the information.

‘Conspicuous’ Lack of Evidence

“The absence of adequate randomized studies comparing different durations of clopidogrel therapy following PCI, and especially DES, should now be conspicuous,” they write.

Some studies suggest that the hazard of clopidogrel discontinuation, although highest within the first 30 days, never disappears, Drs. Gaglia and Waksman note. But details specifically regarding the risks involved in stopping aspirin are lacking in many reports, they add, and continued therapy with aspirin is often assumed but rarely verified.

Of the 2 studies that were randomized, one found no benefit from continuing clopidogrel in patients who had gone event-free for 12 months; at a median follow-up of 19 months, the rates of MI or cardiac death were similar with or without clopidogrel (1.8% vs. 1.2%; P = 0.17). The other, which was greatly underpowered, found a lower rate of death, MI, or stroke at 6 months in patients who received 180 days of clopidogrel following BMS implantation compared with those who received the antiplatelet only for 30 days (1.7% vs. 5.0%; P = 0.01).

Drs. Gaglia and Waksman conclude that future trials following patients for longer durations on dual antiplatelet therapy may help sort out some of the ongoing ambiguities regarding discontinuation. However, the introduction of prasugrel to the antiplatelet armamentarium complicates the situation further, they say, because the only available data, which come from TRITON-TIMI 38, followed patients on prasugrel for up to 15 months but do not address the issue of optimum duration.

Ongoing Studies Somewhat Limited

In an email communication with TCTMD, Ajay J. Kirtane, MD, SM, of Columbia University Medical Center (New York, NY), said the paper is an excellent review of the current literature surrounding clopidogrel discontinuation. 

“The complexities of clinical decision-making make it very difficult to design prospective/randomized studies in this area, which is why so many clinicians utilize empirically based recommendations,” he said.

Although fears of ‘clopidogrel rebound’ have largely abated, Dr. Kirtane added, the optimal duration of clopidogrel therapy remains unknown, with bleeding and cost factors counterbalancing the potential anti-ischemic benefits of continued clopidogrel therapy. 

“Even the current ongoing studies may be somewhat limited in their eventual impact given more recent advances in stent and pharmacologic therapy,” he added.

Source:
Gaglia MA, Waksman R. Systematic review of thienopyridine discontinuation and its impact upon clinical outcomes. Eur Heart J. 2011;Epub ahead of print.

Related Stories:

Disclosures
  • Drs. Gaglia and Waksman report no relevant conflicts of interest.

Comments

CAD Pharma