NATAP: PPIs & CVD

[Guest guest]

Subject: NATAP: PPIs & CVDPLATO: PPI analysis "raises flag" but should not change practice

theheart.org January 24, 2012

Nainggolan

Toronto, ON- A new analysis of the PLATO

study has found that use of a proton-pump inhibitor (PPI) was independently associated with a higher rate of cardiovascular events but

that this was apparent in patients taking ticagrelor (Brilinta, Astra Zeneca) as well those on clopidogrel [1]. The results are reported online January 18, 2012 in Circulation.Lead author Dr Shaun Goodman (St 's Hospital, Toronto, ON) told heartwire

that the finding is likely due to confounding—"ie, the fact that you are on a PPI is a marker for increased risk, rather than [being] a true cause-and-effect relationship." Hence, these new observational findings should

not alter anything. "The observational studies—ours included—raise a yellow flag about the potential interaction on a clinical basis of a PPI

and particularly clopidogrel. But I don't think that any of these studies by themselves should change clinical practice."

In my clinical practice, where there's a strong indication for both—an ADP-receptor antagonist and a PPI—I continue to use those.

He says he sympathizes with certain regulatory

agencies that have issued warnings on the concomitant use of PPIs and clopidogrel "because they always want to err on the cautious side. But I

would place more weight on a randomized trial than I would on any observational study, including ours." The one randomized study that has been conducted on this subject, COGENT, found no clinical interaction in terms of cardiovascular events between the PPI omeprazole

and clopidogrel, he notes, "so there wasn't any rip-roaring signal for harm from the CV standpoint, and, consistent with other studies, there certainly was an apparent advantage from a gastroprotection side of things. Certainly in my clinical practice, where there's a strong indication for both—an ADP-receptor antagonist and a PPI—I continue to use those."

Bleeding findings support conclusions of confounding Goodman and colleagues say that although it is

well established that PPIs interfere with the metabolism of clopidogrel, it is still unclear what the clinical impact of this is, and there is much confusion as to whether PPIs and clopidogrel should be

coprescribed. So they decided to look at this issue in the large acute coronary syndrome (ACS) population of PLATO. They examined the relationship between use of PPIs and the primary end point, one-year cardiovascular events (CV death, MI, or stroke) in these patients, the main aim of PLATO being to compare the effects of clopidogrel and the newer antiplatelet agent ticagrelor in ACS patients. PPIs are not known to interfere with the metabolism of ticagrelor.Those on PPIs (n=6539) at the time of randomization in PLATO were more likely to suffer CV events at one year,

compared with those not on a PPI (n=12 060), regardless of whether they

were assigned to the clopidogrel (13.0% vs 10.9%; adjusted hazard ratio

1.20) or ticagrelor (11.0% vs 9.2%; HR 1.24) arms. "Even after attempting to adjust for differences between those who were on and not on a PPI, we still found an association between PPI use and adverse events. We saw this in both the clopidogrel and the ticagrelor-treated patients," Goodman told heartwire. "Yet ticagrelor—unlike clopidogrel—does not require in vivo biotransformation, so the fact that CV events are generally higher in the PPI-treated ticagrelor patients (just like in the PPI-treated clopidogrel patients) suggests that it isn't the PPI per

se—which shouldn't 'interfere' with ticagrelor's action—but rather the association between PPI use and a higher-risk patient."

It isn't the PPI per se—which shouldn't 'interfere' with ticagrelor's action—but rather the association between PPI use and a higher-risk patient.

This reasoning is further supported by the fact that the PPI-treated patients had higher rates of bleeding in this analysis, he says. "If PPIs interfered with clopidogrel and ticagrelor, we would expect a potentially lower antiplatelet effect and therefore lower rates of bleeding; in fact, we see the opposite."As with any of the nonrandomized (PPI vs no PPI) observational studies, ours included, there may be other unmeasured

factors that aren't included in attempts at 'adjusting'—a good reminder

that observational association can never replace randomized cause-and-effect studies," he continues.And he adds that although they looked at different PPIs individually and found higher CV event rates with omeprazole—which, despite the COGENT findings, has been identified as probably the 'worst offender'—"this is a subgroup of nonrandomized subgroups, so we wouldn't want to overemphasize this particular finding."

Goodman has received research grant support from AstraZeneca, Bristol-Myers Squibb, Daiichi, Eisai, Eli Lilly, Merck, Sanofi-Aventis, and the Medicines Company and speaker/consulting honoraria from AstraZeneca, Bristol-Myers Squibb, Eli Lilly, Novartis, Sanofi-Aventis, Merck, Teva, and the Medicines Company. Disclosures for the coauthors are listed in the paper.

Source

Goodman SG, Clare R, Pieper KS, et al. Association of proton pump inhibitor use on cardiovascular outcomes with clopidogrel

and ticagrelor: Insights from PLATO. Circulation 2012; DOI:CIRCULATIONAHA.111.032912. Available at: http://circ.ahajournals.org.