Proton pump inhibitor use and risk of adverse cardiovascular events in aspirin treated patients with first time myocardial infarction: nationwide propensity score matched study

We appreciate the comments by Fernandez-Fernandez et al. to our
article and find their discussion regarding differences in the effect of
individual H2-blockers interesting. However, due to the relatively small
number of patients receiving H2-blockers in our study, we were not able to
include an analysis of differences between individual H2-blockers.

We were interested to read the paper by Charlot et al. (1) reporting
that proton pump inhibitors (PPI) increase the risk of cardiovascular
events in patients taking aspirin (ASA). They are to be congratulated for
the well conducted study and the thoughtful discussion; however we
disagree with their interpretation of the data.

In any epidemiological study it is important to be cautious in
assuming that the association found is causal as it could be due to bias
or confounding factors. The current spate of articles that associate PPIs
with harm is a good example of this. PPI therapy has been shown to be
associated with an increased risk of hip fracture (2), pneumonia (3),
infective diarrhoea (4), as well as increased cardiac events in ischaemic
heart disease (IHD) patients taking clopidogrel (5, 6). PPI therapy may
cause all these diseases but it is likely that, in at least some cases,
the observed association is due to residual confounding.

The common theme in all studies evaluating the possible adverse
effects of PPI therapy is that sicker patients tend to be prescribed these
medications. Patients on PPI therapy were statistically significantly
more likely to be have chronic obstructive pulmonary disease, heart
failure, coronary artery bypass surgery, previous myocardial infarction,
peripheral vascular disease, renal disease and even cancer (3, 5). Whilst
authors did adjust for these confounding factors it is likely with such
imbalances between groups other confounding factors that were not known or
not measured were also present. This is particularly true if the observed
association was relatively modest with an odds ratio or relative risk of
less than two as is the case with most of these studies.

The paper by Charlot et al. may be another example of this issue,
particularly as direct prospective studies on the interaction of ?ASA and
PPI show no detrimental effect on absorption of aspirin (7, 8). They did
conduct propensity matching which is a better approach to control for
confounding factors but there are still serious concerns regarding
residual confounding with this design, which is why we rely on the
randomized controlled trial (RCT). The authors dismiss this as an
unlikely possibility because the effect was not seen with H2 receptor
antagonists, which they state have identical indications. However,
firstly only 661 patients filled a script for H2 receptor antagonists
compared with 4306 being prescribed PPI therapy. Any lack of association
with H2 receptor antagonists may well relate to a lack of power of
detecting an association with these drugs. Secondly the indications are
not as identical as the authors claim. Patients with severe reflux and
Barrett's oesophagus are much more likely to be prescribed PPI therapy and
we have shown that Barrett's patients are more likely to die from IHD (9).

It is appropriate that an editorial in this issue of BMJ (10)
discusses the putative interaction between clopidogrel and PPI therapy.
The editorial did point out that the COGENT RCT (11) had failed to show
any harm in prescribing PPI therapy. In this RCT all patients were given
ASA as well as clopidogrel and randomized to PPI therapy or placebo.
There was no difference in IHD events between those randomized to PPI
therapy and placebo. Admittedly this study was underpowered but the
hazard ratio for any IHD event was 0.99 (95% confidence intervals (CI) =
0.68 to 1.44). The upper limit of the 95% CI of the hazard ratio was
therefore below the lower 95% CI given by the propensity matched study
reported by Charlot et al. Furthermore the COGENT study did show an
increase risk of gastrointestinal bleeding in those not allocated to PPI
therapy (11). It is interesting that the study by Charlot et al. did not
show any benefit of PPI therapy in reducing GI adverse events when other
RCTs have shown benefit, again suggesting unknown confounders are present.

Patients should not be prescribed any medication without careful
consideration of potential harms as well as benefits. However patients on
ASA have an increased risk of GI bleeding and if there are other serious
risk factors such as concomitant nonsteroidal anti-inflammatory or
anticoagulant therapy then it may be appropriate to prescribe PPI therapy
to prevent GI bleeding (12).

References

1. Charlot M, Grove EL, Hansen PR, Olesen JB, Ahlehoff O, Selmer C et
al. Proton pump inhibitor use and risk of adverse cardiovascular events in
aspirin treated patients with first time myocardial infarction: nationwide
propensity score matched study. BMJ 2011; 342: d2690.

2. Ngamruengphong S, Leontiadis GI, Rahdi S, Dentino A, Nugent K. Proton
pump inhibitors and risk of fracture: a systematic review and meta-
analysis of observational studies. American Journal of Gastroenterology
2011; advance on line publication doi: 10.1038/ajg.2011.113.

3. Sarkar M, Hennessy S, Yang YX. Proton-pump inhibitor use and the risk
for community-acquired pneumonia. Annals of Internal Medicine 2008; 149:
391-8.

4. Leonard J. Marshall JK. Moayyedi P. Systematic review of the risk of
enteric infection in patients taking acid suppression. American Journal
of Gastroenterology 2007; 102: 2047-56.

5. Ho PM, Maddox TM, Wang L, Fihn SD, Jesse RL, Peterson ED, Rumsfeld JS.
Risk of adverse outcomes associated with concomitant use of clopidogrel
and proton pump inhibitors following acute coronary syndrome. JAMA 2009;
301: 937-44.

6. Laine L, Hennekens C. Proton pump inhibitor and clopidogrel
interaction: fact or fiction. American Journal of Gastroenterology 2010;
105: 34-41.

7. Nefesoglu FZ, Ayanoglu-Dulger G, Ulusoy NB, Imeryuz N. Interaction of
omeprazole with enteric-coated salicylate tablets. International Journal
of Clinical Pharmacology & Therapeutics 1998; 36: 549-53.

8. Niazi M, Andersson T, Naucler E, Sundin M, Naesdal J. Evaluation of
the pharmacokinetic interaction between esomeprazole (40 mg) and
acetylsalicylic acid (325 mg) in healthy volunteers. International Journal
of Clinical Pharmacology & Therapeutics 2009; 47: 564-9.

9. Moayyedi P., Burch N., Akhtak-Danesh N., Enaganti SK., Harrison R.,
Talley NJ., Jankowski J. Mortality rates in patients with Barrett's
esophagus. Alimentary Pharmacology and Therapeutics 2008; 27: 316-20.

10. Small GR, Chow BJW, So DFY. Coprescription of clopidogrel and proton
pump inhibitors. BMJ 2010; 341: b4351.

11. Bhatt DL, Cryer BL, Contant CF, Cohen M, Lanas A, Schnitzer TJ, et
al. Clopidogrel with or without omeprazole in coronary artery disease. N
Engl J Med 2010; 363: 1909-17.

12. Rostom A, Moayyedi P, Hunt R; Canadian Association of
Gastroenterology Consensus Group. Canadian consensus guidelines on long-
term nonsteroidal anti-inflammatory drug therapy and the need for
gastroprotection: benefits versus risks. Aliment Pharmacol Ther. 2009; 29:
481-96.