Researchers Accurately Predict Who Will Discontinue Clopidogrel

By identifying key predictors such as lack of health insurance, low salary level, marriage status, and diagnosis of acute coronary syndromes (ACS), researchers have developed a risk score that can accurately identify patients undergoing percutaneous coronary intervention (PCI) who are likely to discontinue thienopyridine therapy within 30 days. The results were reported online December 24, 2010, ahead of print in the American Journal of Cardiology.

Renato D. Lopes, MD, PhD, of the Duke Clinical Research Institute (Durham, NC), and colleagues from the Institute of Cardiology of Rio Grande do Sul (Rio Grande do Sul, Brazil), looked at 400 consecutive PCI patients treated from November 2007 to March 2008 at the Brazilian center who were instructed to remain on thienopyridine therapy for at least 30 days after discharge (≥ 30 days for patients given BMS, ≥ 1 year for those given DES).

Importance of Antiplatelet Therapy Not Getting Through

At 30 days, 66 patients (16.5%) reported they had discontinued clopidogrel since the procedure. An additional 102 patients (25%) reported problems with adherence. The median time of discontinuation was 10 days. The most common reasons given for discontinuing the antiplatelet included:

• Cost of the medication, which was equivalent to US $50 per month (41 patients)
• Insufficient information about importance of dual antiplatelet regimen (11 patients)
• Advice by another doctor to stop the drug (10 patients)

Patients who discontinued thienopyridine therapy were almost twice as likely to state they were unaware of the importance of using clopidogrel compared with those who maintained treatment (50% vs. 26%; P < 0.001).

On multiple logistic regression analysis, several factors including unmarried status, lack of health insurance, ACS presentation, and absence of diabetes increased the risk of noncompliance (table 1).

Table 1. Factors Associated with Clopidogrel Discontinuation

A risk score ranging from 0 to 14 points was subsequently developed from these variables:

• Unmarried: 1
• ACS: 1
• Absence of diabetes: 1
• Lack of private health insurance: 4
• Salary
  • < 2 times minimum wage: 7
  • 2-3 times minimum wage: 3

The risk score predicted discontinuation rates with a high degree of accuracy (table 2).

Table 2. Clopidogrel Discontinuation Rates According to Risk Score^a

^a P < 0.0001 for association between risk score and discontinuation.

In addition, the risk score showed a high predictive probability based on 1,000 samples for adherence to clopidogrel (P < 0.001) as well as a strong correlation with complete adherence as assessed by the Morisky questionnaire, a standardized instrument that also evaluates drug adherence (P < 0.001).

The researchers conclude that “we have identified patients at risk for premature discontinuation of thienopyridines using variables obtained before stent implantation and developed a risk score that accurately predicts premature thienopyridine discontinuation.”

In an e-mail communication with TCTMD, Sunil V. Rao, MD, of Duke University Medical Center (Durham, NC), who was not involved in the study, noted that the investigators identified predictors such as low income that have previously been associated with poor medication adherence. “The value of this study is that it extends [such research] beyond the United States or Europe and looks at Latin America, which is a part of the cardiology world we have not focused on previously,” he said.

Risk Scores Nice, But Not Often Used

In terms of actual clinical utility, however, “risk scores are always tricky,” Dr. Rao said. “They are scientifically interesting, but almost no one uses them in practice. We probably should be using risk scores more often.” A drawback of the current risk score is that the researchers did not account for noncardiac procedures following PCI, such as colonoscopy or dental work, he pointed out, adding, “This is a major issue for discontinuation in most practices.”

In a telephone interview with TCTMD, Dr. Lopes noted that the main contribution of the paper was to hopefully “serve as a wake up call for a much bigger problem that we sometimes don’t appreciate in terms of non-adherence,” he said. “Then, secondly, if you have a score, maybe you can start to identify certain groups of patients with poor compliance and start to build strategies around them to improve compliance.”

Dr. Rao stressed that education is key. “This paper shows that many patients do not understand why they are on dual antiplatelet therapy,” he said. “Patient education is a big part of what we should be doing and educating patients about the need for dual antiplatelet therapy after PCI should be started in the pre-PCI setting and continued through the post-PCI period.”

Extrapolating to US Practice

Another issue is whether Dr. Lopes and colleagues’ results translate to the US PCI population. For instance, due to health insurance policies in Brazil, which are much more restrictive over stenting practices, the DES implantation rate in the study was only 3%, much lower than that in the United States.

According to Dr. Rao, the paper is relevant to US practice despite the differences in stenting rates. “We know that there is a large population of patients in the United States who cannot afford their medicines, and if they can, they tend to change the way they take them. For example, taking a BID medicine once a day gets you twice the mileage out of a prescription,” he said. “In that sense, I think this is more of a medication adherence paper than a DES paper.”

Dr. Lopes added that even with the DES issues, the paper remains highly applicable. “You cannot extrapolate from one culture and health system to another, but if you look at all the US studies, the variables (cost, insurance, education) are all the same, so the situation is similar,” he said. “And for tertiary hospitals that receive a lot of referral patients in whom we don’t know their background, we’re trained to put BMS in unless there’s a clear angiographic indication for a DES. These results might apply in such cases.”

Knowing the Patient on the Table

Beyond that, he added, “if these troubles might be happening with BMS, to extrapolate [to DES] where clopidogrel should be given for at least a year, these problems may be at least as important.”

But what should physicians do with a patient who is at risk for poor clopidogrel adherence? “There are several strategies that should be tested,” Dr. Lopes said, noting that family members can be enlisted to help patients take pills or set alarms as reminders.

Dr. Rao suggested an alternative. “I think it would be reasonable to say that a patient who is at risk for medication nonadherence may be better off with a bare metal stent rather than a DES,” he said, adding that a key message from the study “is to know the patient on the cath lab table before you pick the stent!”

Study Details

Mean patient age was 61.0 ± 10.4 years. Slightly under half (45%) had ACS. More patients who discontinued clopidogrel had chronic heart failure, emergency procedures, and ACS compared with those who stayed on thienopyridine therapy. Patients who discontinued also more often lacked private health insurance, had lower salaries, and were unmarried.

 

---

Source:
Quadros AS, Welter DI, Camozzatto FA, et al. Identifying patients at risk for premature discontinuation of thienopyridine after coronary stent implantation. Am J Cardiol. 2010;Epub ahead of print.

 

 

Related Stories:

• Stopping Antiplatelet Therapy Often Unrelated to Major Bleeding, Surgery
• Delay in Filling Clopidogrel Prescriptions May Increase Death, MI Risk After DES
• Antiplatelet Compliance After PCI Suboptimal