Increasing evidence that treatment with a proton-pump inhibitor can reduce the efficacy of clopidogrel in patients with coronary disease in a clinically meaningful and dangerous way may prompt cardiology groups to rethink current recommendations on using the two drugs together.
Although data from “a series of observational reports … are not ideal for clinical decision making, I think we are going to have to readdress the issue,” said Dr. Robert A. Harrington, an interventional cardiologist and professor of medicine at Duke University, Durham, N.C.
He was one of several cardiologists who quickly conferred last November, after a talk at the American Heart Association's scientific sessions reported the first clinical evidence of an increased risk for ischemic events in patients with coronary disease on clopidogrel.
Representatives of the American College of Cardiology, the American College of Gastroenterology, and the AHA later that same day issued a statement saying that the evidence from that single study did not provide “sufficient evidence to change clinical practice.” But findings in two later reports, the most recent of which appeared in JAMA in early March, changed the minds of some experts.
“The idea of prescribing a PPI as routine practice [in patients on clopidogrel] has to be questioned given the combination of pharmacologic as well as now clinical data. I'm not doing this routinely now without thinking, Does the patient really need the PPI?” said Dr. Eric D. Peterson, also professor of medicine at Duke, and a coauthor of the JAMA report.
Preliminary evidence also suggests that for patients who clearly need a PPI while taking clopidogrel, pantoprazole may be the safest PPI because, unlike other PPIs, it does not inhibit the hepatic enzyme responsible for converting clopidogrel to the active form with an antiplatelet effect.
A special danger may also come from the OTC availability of omeprazole, which patients could start on their own without consulting their physicians. “The burden is on physicians to tell patients that just because a drug is available OTC doesn't make it safe,” said Dr. Harrington, also director of the Duke Clinical Research Institute, in an interview.
Physicians often prescribe a PPI to patients taking clopidogrel (Plavix), a mainstay of treatment for patients who have received a coronary stent or have had a recent acute coronary syndrome (ACS) event, because clopidogrel has been linked to gastric bleeding.
Last October, an expert consensus document from the ACC, ACG, and AHA endorsed using a PPI in patients at high risk for gastrointestinal bleeding when on clopidogrel or other agents that can cause duodenal bleeds such as aspirin or NSAIDs (Circulation 2008;118:1894–909).
High-risk patients include those with a history of an ulcer and those on an additional medication that would boost their bleeding risk, such as warfarin or a corticosteroid, said Dr. Deepak L. Bhatt, a cardiologist at Brigham and Women's Hospital in Boston and cochair of the expert panel that wrote the recommendations last October. But many physicians go beyond this recommendation and prescribe a PPI to patients on clopidogrel who are not at high risk.
The new report in JAMA came from an observational study of patients hospitalized for an acute myocardial infarction or unstable angina during October 2003-January 2006 at any of 127 Veterans Affairs hospitals. At discharge, 8,205 of the patients filled a prescription for clopidogrel from a VA pharmacy; 64% also took a PPI at discharge, and the other 36% did not receive a PPI prescription and were presumed not taking one of these drugs. Perhaps as many as 60% of the patients prescribed a PPI at discharge received it without having a bleeding indication and so probably got the drug as prophylaxis only, said Dr. P. Michael Ho, lead author of the study and a cardiologist at the Denver Veterans Affairs Medical Center, in an interview.
During a median follow-up of 521 days, the rate of death or rehospitalization for an ACS event was 21% in patients on clopidogrel only and 30% in those on both clopidogrel and a PPI. In a multivariant analysis that controlled for baseline demographic and clinical variables, patients on both drugs were 25% more likely to die or be rehospitalized for ACS, 86% more likely to be rehospitalized for ACS, and 49% more likely to need a revascularization procedure, compared with patients on clopidogrel only (JAMA 2009;301:937–44). All three increased relative risks were statistically significant.
The results suggest that in such patients prescribed a PPI for prophylaxis or for reflux symptoms, an alternative gastroprotective drug should be considered, such as a histamine H2-receptor antagonist, sucralfate, or misoprostol, said Dr. Ho, who acknowledged that each of these alternatives may be less effective than a PPI.