10 December 2009
A large observational study finds no conclusive evidence for a clinically relevant drug interaction between clopidogrel and PPIs in elderly patients after percutaneous coronary intervention or hospitalisation for acute coronary syndrome.
Level of evidence:
Level 2 (limited quality patient-oriented evidence) according to the SORT criteria.
This study adds to the evidence regarding the clinical significance of the interaction between clopidogrel and proton pump inhibitors (PPIs), but by itself should not change practice.
- The need for PPI therapy in patients who are also taking clopidogrel should be reviewed at their next appointment: avoid concomitant use of these medicines unless considered essential
- Prescribe PPIs in line with their licensed indications where possible
- Check whether patients who are taking clopidogrel are buying over-the-counter omeprazole and consider whether another gastrointestinal therapy would be more suitable.
No doubt European and UK regulatory authorities are examining this recent data and considering it alongside the rest of the evidence on this topic. Given the interest in this topic, if there are further developments we will produce another MeReC Stop Press or Rapid Review.
What is the background to this?
As reviewed in MeReC Extra 41, a CHMP review in May 2009 concluded that data supports a possible clinically significant interaction between clopidogrel and PPIs, that makes clopidogrel less effective. The CHMP recommended that product information for clopidogrel should be amended to discourage concomitant use of a PPI and clopidogrel unless considered absolutely necessary; hence, the MHRA advice given above.
The present study provides information obtained from three large cohorts (one in Canada, two in the US), on the risk of cardiovascular outcomes and mortality associated with the prescribing of clopidogrel and a PPI to elderly patients after percutaneous coronary intervention (PCI) or hospitalisation for acute coronary syndrome (ACS).
What does this study claim?
This study of 18,565 users of clopidogrel aged 65 years or older identified that 2.6% of those people who were initiated on clopidogrel with a PPI were hospitalised for myocardial infarction [MI], 1.5% died, and 3.4% underwent revascularisation. This compared with 2.1%, 0.9%, and 3.1%, respectively, for clopidogrel-users who were non-users of a PPI. The difference in the primary end point of MI or death was not statistically significantly different (adjusted rate-ratio [RR] 1.22; 95% confidence interval [CI] 0.99 to 1.51).
How does this relate to other studies?
The evidence for a clinically relevant drug interaction of clopidogrel and PPIs comes largely from observational studies (see the July Drug Safety Update). We previously blogged an analysis of two subsequent studies which found that, although PPIs attenuated the in-vitro antiplatelet effects of clopidogrel and prasugrel▼, the combination was not associated with an increased risk of adverse cardiovascular outcomes. These findings were consistent regardless of which PPI was used, or whether an H2-receptor antagonist was used.
Recently, the FDA has issued a Public Health Advisory that ‘new data’ has shown that when clopidogrel and omeprazole are taken together, the effectiveness of clopidogrel is reduced, and advises patients taking clopidogrel and omeprazole to consult their healthcare provider. The data on which this advisory is based comes from pharmacokinetic studies demonstrating reductions in the blood level of clopidogrel’s active metabolite when clopidogrel was taken with omeprazole, regardless of whether the omeprazole was taken at the same or different time of day. Some limited further information is provided for healthcare professionals. This states that the FDA are aware there are studies, such as the COGENT study (as yet unpublished) that might provide information about the effect of this interaction on clinical outcomes. However, the FDA considered that the applicability of data from this study is limited because of the study design and follow-up.
Despite the pharmacokinetic evidence of an interaction between clopidogrel and omeprazole, as recently reported by the FDA Public Health Advisory, not all studies measuring clinical outcomes have been able to demonstrate that PPIs significantly reduce the clinical effect of clopidogrel. This has resulted in some US cardiologists questioning the recommendations made by the FDA in response to the pharmacokinetic data.
In the present observational study, which used both traditional and advanced techniques to account for confounders, a higher proportion of patients suffered cardiovascular events or death if taking a PPI in addition to clopidogrel compared with clopidogrel alone. However, the difference was not statistically significant at the normally accepted probability level of 95%. The study did not rule out the possibility, taking 95% confidence intervals into account for the primary endpoint, that there could be as much as a 51% greater relative risk, as opposed to only as little as a 1% decreased relative risk, of patients who took a PPI and clopidogrel suffering a cardiovascular event or death compared with those taking clopidogrel alone. This would seem somewhat at odds with the authors’ concluding statement that the risk, if it exists, is “unlikely to exceed a 20% risk increase”.
This study by itself is inconclusive, but it does provide useful additional data that needs to be considered alongside that of other studies to assess the clinical significance of the clopidogrel/PPI interaction.
Rassen JA, et al. Cardiovascular outcomes and mortality in patients using clopidogrel with proton pump inhibitors after percutaneous coronary intervention or acute coronary syndrome. Circulation 2009;120:2322–29
Observational study of three cohorts (Pennsylvania and New Jersey, US; British Columbia , Canada) identified from health insurance programs.
Patients (n=18,565) aged 65 years or older prescribed clopidogrel initially who underwent PCI or were hospitalised for ACS between 2001 and 2005.
Intervention and comparison
Rate ratios [RR] compared the incidence of the study outcomes among clopidogrel users who also used PPIs and those that were non-users of PPIs. Multivariate adjusted RRs were estimated by Cox proportional-hazards regression for each cohort and for the pooled cohorts. Patients were followed up for a median of 29 to 30 days among the PPI users and non-users (maximum 180 days).
Outcomes and results
On a pooled basis, 2.6% of those people who were initiated on clopidogrel with a PPI were hospitalised for MI compared with 2.1% for non-users; figures for death and revascularisation were 1.5% vs. 0.9% and 3.4% vs. 3.1%, respectively. The propensity score-adjusted RR for the primary endpoint of hospitalisation for MI or death was 1.22; 95%CI 0.99 to 1.51). There were also no significant differences between PPI users and non-users for the pooled cohort with respect to death (RR 1.20 [95%CI 0.84 to 1.70]) or revascularisation (RR 0.97 [95%CI 0.79 to 1.21]).
The trial was publicly funded.
Further information on the use of clopidogrel is available on the cardiovascular section of NPC.
Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you