Meta-analysis Evaluates Evidence for Clinical Impact of PPIs in Patients on DAPT

Observational studies and RCTs assessing whether proton pump inhibitors (PPIs)—some of which may interfere with clopidogrel activation—worsen cardiovascular outcomes when used in conjunction with dual antiplatelet therapy (DAPT) yield conflicting results, according to a meta-analysis published in the January 2015 issue of Circulation: Cardiovascular Quality and Outcomes.

Take Home: Meta-analysis Evaluates Evidence for Clinical Impact of PPIs in Patients on DAPT

Given that modeling and propensity-score matching may not fully resolve the possibility of selection biases in observational studies, there remains a need for future randomized studies “combining the assessment of pharmacodynamic parameters and their association with clinical outcomes,” investigators say.

Reviewing the medical literature between 1995 and July 2012, investigators led by Chiara Melloni, MD, MHS, of the Duke Clinical Research Institute (Durham, NC), analyzed 35 studies reporting outcomes of patients with unstable angina or NSTEMI. Five (4 RCTs and 1 observational study) assessed the effect of omeprazole or esomeprazole (Nexium; AstraZeneca) added to DAPT compared with DAPT alone, while the other 30 (all observational) looked at the effect of PPIs as a class compared with no PPI use in the setting of DAPT.

Mixed Bag of Evidence

To accommodate the heterogeneity of the studies, random-effects models were used for all outcomes. Meta-analyses of the studies assessing PPIs as a class consistently reported higher 1-year event rates in patients receiving the drugs, while data from RCTs evaluating omeprazole vs placebo showed no difference in ischemic outcomes. According to the authors, the strength of evidence was low for most outcomes and only moderate for combined all-cause mortality/nonfatal MI and all-cause mortality alone (table 1).

 Table 1. Summary Results of Study Outcomes at About 1 Year: PPI vs No PPI Use

In addition, 3 of 4 RCTs found less upper GI bleeding at 6 months with omeprazole vs no PPI use, while 4 observational studies reported conflicting results for both in-hospital and longer-term GI bleeding. Given the differences in duration of follow-up in both study types, a meta-analysis was not performed for this endpoint. However, the investigators say, overall there was moderate evidence favoring use of PPIs for reduction of GI bleeding.

Dr. Melloni and colleagues observe that although they attempted to control for certain variables, “it is highly probable that unmeasurable confounding still remains, and the effect of PPIs analyzed as a class in this analysis is not based on strong evidence.”

They note that the American Heart Association/American College of Cardiology guidelines now suggest that physicians evaluate the need for starting or continuing PPIs in patients taking clopidogrel. The document also describes the potential risks and benefits of coadministration.

“Our findings also support a cautious approach to PPI use with DAPT therapy,” the investigators say. “Further research—preferably additional RCTs of specific PPI agents or prospective propensity score–matched cohort studies—is warranted on whether the detrimental effect of PPIs is [attributable to] comorbid conditions of the patient population, to the specific PPI agent used, or to genetic predisposition for reduced clopidogrel responsiveness.”

Evidence Points to Confounding

In an accompanying editorial, Peter B. Berger, MD, of the Geisinger Medical Center (Danville, PA), applauds the authors for their effort to assess the quality of evidence.

He not only endorses their conclusion that unmeasured confounders likely explain the discordant findings between RCTs and observational studies but also cites other “lines of evidence” that argue against an adverse clinical impact of PPIs:

Interactions have been described between clopidogrel and other medications that share enzymes required to activate them, such as statins and calcium channel blockers—yet it is now accepted that these are not clinically relevant.

The magnitude of the negative effect of PPIs in certain studies is such that the drugs appear not just to neutralize the benefit of clopidogrel but actually to trigger thrombosis on their own—which is highly improbable.

The fact that several drugs shown to increase platelet inhibition have not reduced thrombotic events should help alleviate concern about the opposite effect—increased thrombosis—when tests show diminished platelet inhibition.

Spotlight Should Be on GI Bleeding

Lost in the “hubbub” about potential interactions, Dr. Berger writes, are randomized findings of reduced GI bleeding—and less stoppage of antiplatelet therapy—in patients on PPIs. The more appropriate concern, he notes, is that PPIs are too infrequently coadministered with clopidogrel alone, DAPT, or triple therapy with an anticoagulant.

In a telephone interview with TCTMD, Deepak L. Bhatt, MD, MPH, of Brigham and Women’s Hospital (Boston, MA), called the editorial’s observations “right on the mark.” He said the real value of the paper is to draw attention to the need to reduce GI bleeding in cardiovascular patients, especially those on potent antithrombotic medications.

Even though worry about the coadministration of clopidogrel and PPIs is unwarranted—at least for standard doses—questions remain about how to optimize the GI benefits of PPIs, he added. These include which patients should be on a prophylactic regimen and for how long, and whether the type of PPI used matters.

Are Label Warnings Relevant?

Concern over the adverse effect of certain PPIs on clopidogrel metabolism prompted the FDA to require label cautions against use of omeprazole in 2009 and esomeprazole in 2011. According to Dr. Bhatt, physicians’ worries have waned in recent years. Nonetheless, if there are no cost issues and the patient is just starting a PPI, he advises avoiding these particular agents. However, he recommends against switching patients who are already on omeprazole or esomeprazole to another PPI.

In his editorial, Dr. Berger goes further, advocating removal of label warnings since, he claims, more patients have been harmed by not filling prescriptions for clopidogrel or a PPI than have been benefited by the added caution.

The PPI-clopidogrel story is another illustration of the idea that extrapolating from drug mechanisms to clinical effects can be misleading, Dr. Bhatt commented, noting that the majority of the variability in clopidogrel response remains unaccounted for. A multitude of factors are at work, from genetics to clinical risk and drug-drug interactions, he said, but each one in isolation plays a very small role.




Sources:1. Melloni C, Washam JB, Jones WS, et al. Conflicting results between randomized trials and observational studies on the impact of proton pump inhibitors on cardiovascular events when coadministered with dual antiplatelet therapy: systematic review. Circ Cardiovasc Qual Outcomes. 2015;Epub ahead of print.
2. Berger PB. Should proton pump inhibitors be withheld from patients taking clopidogrel? Issue that has been giving me heartburn [editorial]! Circ Cardiovasc Qual Outcomes. 2015;Epub ahead of print.




Related Stories:



  • Dr. Melloni reports no relevant conflicts of interest.
  • Dr. Berger reports that his institution has received research grants for which he served as PI from Bristol-Meyers Squibb, Eli Lilly, and Sanofi Aventis.
  • Dr. Bhatt reports receiving research funding from AstraZeneca, Bristol-Myers Squibb, Eisai, and Sanofi-Aventis.

We Recommend