 Advertisement | ||||||||
 | ||||||||
|
||||||||
 |
 |
 |
 |
 |
 |
|
||
Electronic letters to:
|
|
Electronic letters published:
![[Read eLetter]](/icons/misc/sectionDOWN.gif){kind=icon}
Authors respond to Friesen and Allen's eletters
Relevance of case-control study in practice
PPIs and Clopidogrel: Addressing the data and the bigger clinical question
Diabetes, not a drug interaction, accounts for the recurrent myocardial infarction
Pantoprazole not proven to be different from other PPIs
Issues that limit study conclusions
|
|
|||
|
David Juurlink Department of Medicine, University of Toronto, Toronto, Ont.
Send letter to journal:
dnj{at}ices.on.ca David Juurlink
|
[Three of the authors respond:] Aaron Tejani and Michael Wasdell postulate a type II error in our pantoprazole analysis.<1> Our observations in this regard derive from 171 subjects exposed to pantoprazole and yield an effect size near unity (adjusted odds ratio 1.02, 95% confidence interval 0.70–1.47); thus, a type II error is an exceedingly unlikely explanation for the finding. The 12 cases and 53 controls noted in their letter represent patients whose prescriptions for a proton pump inhibitor were dispensed long before the index date. These exposures (between 6 and 12 months before the date of recurrent myocardial infarction) were so remote that they added no useful information to our study findings and were not presented in Table 2. Consequently, the worst-case scenario analysis proposed by Tejani and Wasdell is uninterpretable. We agree with the comments of Tejani and Wasdell regarding the potential overuse of proton pump inhibitors. Recent consensus guidelines recommend a proton pump inhibitor for patients receiving aspirin in combination with another antiplatelet agent: in other words, virtually every patient taking clopidogrel for coronary disease would receive one.<2> If these guidelines are followed indiscriminately, millions of patients worldwide may not receive the full antiplatelet effect of clopidogrel. Pending further data on the clinical significance of drug interactions between proton pump inhibitors and clopidogrel, we believe that pantoprazole should be preferentially prescribed to patients who require both drugs. There is presently no evidence that pantoprazole increases the risk of adverse events in patients receiving clopidogrel, whereas such evidence exists for other proton pump inhibitors. David N. Juurlink, MD PhD Jack V. Tu, MD PhD Muhammad M. Mamdani, PharmD MPH Department of Medicine University of Toronto Toronto, Ont. REFERENCES 1. Juurlink DN, Gomes T, Ko DT, et al. A population-based study of the drug interaction between proton pump inhibitors and clopidogrel. CMAJ 2009;180:713-8. 2. Bhatt DL, Scheiman J, Abraham NS, et al. ACCF/ACG/AHA 2008 expert consensus document on reducing the gastrointestinal risks of antiplatelet therapy and NSAID use: a report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents. Circulation 2008;118:1894-909. Conflict of Interest:None declared |
|||
|
|
|||
|
Brian Y L Wong Sudbury Regional Hospital
Send letter to journal:
brianylwong{at}sympatico.ca Brian Y L Wong
|
Juurlink et al. (1) found a higher prevalence of current proton pump inhibitor (PPI) use among patients who were readmitted with reinfarction within 90 days of the index event. This led them to speculate a loss of beneficial effect of clopidogrel due to inhibition of its bioactivation pathway by selected PPI. Notwithstanding their provocative findings, the following points were not specifically addressed by the authors but are important considerations when putting these results in clinical context. First, the use of clopidogrel reduces but not eliminates the risk of reinfarction. Based on clinical trial data (2), the addition of clopidogrel to standard care is expected to prevent one out of five adverse events. Whether concurrent use of PPI leads to a higher than expected rate of reinfarction among clopidogrel-treated patients remains unknown in this case-control study. Second, information on indication of PPI use is unavailable. Anecdotally, it is not uncommon for angina to be perceived as, or misdiagnosed to be indigestion. Thus, current PPI use may in fact be a marker of post-infarction angina rather than mediator of thrombotic event. Third, patients who either developed reinfarction (1) or were prescribed PPI (3) were sicker cohorts and with higher prevalence of such distinctive factors as diabetes, renal failure, and heart failure; factors that were associated with higher risk of restenosis, stent thrombosis, and natural disease progression. Therefore patients identified in such manner harbored more advanced coronary disease, making multivariable adjustment inappropriate or irrelevant. References: 1)Juurlink DN, Gomes T, Ko DT, et al. A population-based study of the drug interaction between proton pump inhibitors and clopidogrel. CMAJ 2009;180(7):713-8. 2)Yusuf S, Zhao F, Mehta SR, et al. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST segment elevation. N Engl J Med 2001;345:494-502. 3)Ho PM, Maddox TM, Wang L, et al. Risk of adverse outcomes associated with concomitant use of clopidogrel and proton pump inhibitors following acute coronary syndrome. JAMA 2009;301(9):937-944. Conflict of Interest:None declared |
|||
|
|
|||
|
Aaron M Tejani Therapeutics Initiative/Fraser Health
Send letter to journal:
aaron.tejani{at}fraserhealth.ca Aaron M Tejani
|
Written by Aaron M Tejani (Therapeutics Initiative/Fraser Health) and Michael Wasdell (Fraser Health): The results and conclusions reported by Jurrlink et al.(1), if taken at face value, could have significant implications for best practices, policies and procedures at the level of the regional health authority pharmacy departments. Previous letters have addressed some potential limitations of the study and we present here other concerns that are relevant to decisions for practice change based on the study results. First, the sum of the numbers reported in Table 2 of the study (1) does not represent the entire study sample of 734 cases and 2057 controls; there are 12 cases and 53 controls that are not accounted for based on the information provided in Table 2. We elected to conduct a "worse case scenario" analysis in which all missing cases were assumed to not have had an exposure to a PPI, and the missing controls are assumed to have had a PPI exposure. This scenario deliberately aims to bias the resulting odds ratio in favour of no association between MI and PPI exposure. In this scenario, we have assumed the same reference group of patients with no prescription for a PPI (448 cases and 1317 controls) and current PPI exposed cases and controls (194 current PPI exposed cases, 424 current PPI exposed controls) as noted in the Jurrlink et al. paper(1). Thus, in our worst case scenario, there would be 460 cases without PPI exposure (448 reference cases plus 12 missing cases) and 194 cases with current PPI exposure. For controls, the scenario would include the 1317 unexposed reference control patients and 477 PPI exposed controls (424 PPI exposed controls plus 53 missing controls). An unadjusted odds ratio and 95% CI using these values would then be 1.16 (0.95 -1.42) which is not significant. We feel that this missing data is extremely important as the worse case scenario tells us that our conclusions may change considerably if this data was available. As we have also been unable to reproduce any of the unadjusted odds ratios based on data provided, it is important that the data reported in the tables be rechecked for accuracy. Second, Juurlink et al. have reported no increased risk of myocardial infarction with Pantoprazole. We feel that this may not be the case as it is entirely possible that by analyzing the Pantoprazole data separately a Type II error could have occurred (i.e an increased risk with Pantoprazole could be present but was missed due to the relatively small numbers of data points compared to the "other PPI" analysis). To further support this notion that there may be no difference in risk between Pantoprazole and "other PPIs" one can easily see (see Figure 2 of the Juurlink study(1)) that the confidence intervals for each of these groups of data overlap suggesting that the effect in one group (e.g Pantoprazole) is very similar to the other (e.g. "other PPI"). Finally, the recommendation (ACC/AHA 2007 guidelines (2)) to use PPIs in unstable angina/non-ST segment elevation MI (UA/NSTEMI) patients on dual antiplatelet therapy is not based on strong evidence. The guidelines quote 2 studies (3,4) and 1 systematic review (5) all of which do not look specifically at coronary artery disease patients on dual antiplatelets, but all include any patients who have had bleeding ulcers previously. Perhaps we should not worry about potential drug-drug interactions, but rather we should be concerned that using PPIs at all in these patients is not supported by evidence. Indeed the recent study by Ho et al (6) suggests an increased risk of adverse outcomes in patients who use PPIs with clopidogrel. The Rx Files (7) have rightly stated that, when faced with a coronary artery disease patient on dual antiplatelet therapy, one should first assess whether a PPI is warranted. In our opinion, we have not come across any definitive evidence that coronary artery disease patients on dual antiplatelet therapy should be routinely treated with PPIs in the first place. Further to this, the unwarranted use of PPI in a large patient population such as those with coronary artery disease does not come without serious public health risks. Specifically, there is evidence to suggest that the risk of clostridium difficile diarrhea and community-acquired pneumonia rises with the use of PPIs (8,9,10). We are not convinced that current evidence demonstrates that the benefits outweigh the risks when PPIs are used for the sole purpose of reducing the risk of bleeding complications in patients on dual antiplatelet therapy. 1. Juurlink DN, Gomes T, Ko DT, et al. A population-based study of the drug interaction between proton pump inhibitors and clopidogrel. CMAJ 2009;180(7). DOI:10.1503/cmaj.082001. 2. Anderson JL, Adams CD, Antman EM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non–ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non–ST-Elevation Myocardial Infarction). Circulation. 2007;116:e148–e304. 3. Lai KC, Lam SK, Chu KM, et al. Lansoprazole for the prevention of recurrences of ulcer complications from long-term low-dose aspirin use. N Engl J Med 2002;346:2033– 8. 4. Chan FK, Ching JY, Hung LC, et al. Clopidogrel versus aspirin and esomeprazole to prevent recurrent ulcer bleeding. N Engl J Med 2005;532:238–44. 5. Leontiadis GI, Sharma VK, Howden CW. Systematic review and meta-analysis of proton pump inhibitor therapy in peptic ulcer bleeding. BMJ 2005;330:568. 6. Ho PM, Maddox TM, Wang L, et al. Risk of adverse outcomes associated with concomitant use of clopidogrel and proton pump inhibitors following acute coronary syndrome JAMA. 2009;301(9):937-944 (doi:10.1001/jama.2009.261). 7. Schuster P, Regier L, Jorgensen D, et al. PPIs may reduce the efficacy of clopidogrel (Plavix™). January 2009: http://www.rxfiles.ca/rxfiles/uploads/documents/Clopidogrel-PPI- interaction-QandA.pdf 8. Dial S, Alrasadi K, Manoukian C, et al. Risk of Clostridium difficile diarrhea among hospital inpatients prescribed proton pump inhibitors: cohort and case–control studies. CMAJ 2004;171(1):33-8. Conflict of Interest:None declared |
|||
|
|
|||
|
Doson Chua Department of Pharmacy
Send letter to journal:
dchua{at}providencehealth.bc.ca Doson Chua
|
To the Editor: The article by Juurlink and colleagues highlights a potential drug interaction of proton pump inhibitors with clopidogrel in their case- control study.(1) While the results do show a clinically significant association of recurrent myocardial infarction with concurrent proton pump inhibitor and clopidogrel use, the patient population studied may account for this increased risk. The case group had a significantly higher percentage of diabetic patients compared to the matched control group (28.3% vs. 19.9%, p<0.001). Diabetic patients demonstrate poorer outcomes post myocardial infarction compared to non diabetic patients due platelet and endothelial dysfunction.(2) Diabetics have also been shown to have an attenuated response to clopidogrel mediated platelet inhibition and show enhanced platelet reactivity compared to non diabetic patients.(3-5) This poor response to antiplatelet therapy predisposes diabetic patients to recurrent myocardial infarction. Thus, it is plausible that the higher risk of recurrent myocardial infarction seen in the case group was due to the greater number of diabetic patients as compared to the matched control group and not an interaction between proton pump inhibitors and clopidogrel. 1. Juurlink DN, Gomes, T, Ko DT, Szmitko PE, Austin PC, Tu JV, Henry DA et al. A population-based study of the drug interaction between proton pump inhibitor and clopidogrel 2. Haffner SM, Lehto S, Ronnemaa T, Pyorala K, Laakso M. Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic patients with and without prior myocardial infarction. New Engl J Med 1998;339:229-234 3. Angiolillo DJ, Fernande-Ortiz A, Bernardo E, Ramierz C, Sabate M, Jimenez-Quevedo P et al. Platelet function profiles in patients with type 2 diabetes and coronary artery disease on combined aspirin and clopidogrel treatment. Diabetes 2005;54:2430-2435 4. Geisler T, Anders N, Paterok M, Langer H Konstantinos S, Lindemann S et al. Platelet response to clopidogrel is attenuated in diabetic patients undergoing coronary stent implantation. Diabetes Care 2007;30:372 -374 5. Angiolillo DJ, Shoemaker SB, Desai B, Yuan H, Charlton RK, Bernardo E et al. Randomized comparision of a high clopidogrel maintenance dose in patients with diabetes mellitus and coronary artery disease: results of the optimizing antiplatelet therapy in diabetes mellitus (OPTIMUS) study. Circ 2007;115:708-716 Doson Chua, PharmD, BCPS(AQ) Clinical Pharmacy Specialist Department of Pharmacy St. Paul's Hospital Vancouver, BC Angela Lo, BSc(Pharm), MSc, MHA Regional Medication Use Management Pharmacist Providence Health Care/Vancouver Coastal Health Authority Vancouver, BC Conflict of Interest:None declared |
|||
|
|
|||
|
Michael J Allen Dalhousie University CME
Send letter to journal:
michael.allen{at}dal.ca Michael J Allen
|
Juurlink et al (1) show that among patients receiving clopidogrel following an acute myocardial infarction, current users of proton pump inhibitors (PPIs) had an increased risk of reinfarction (odds ratio 1.27, 95% confidence interval [CI] 1.03-1.57). A further analysis by type of PPI showed that this effect was not seen with pantoprazole (odds ratio 1.02, 95% CI 0.70-1.47) but only with other PPIs (odds ratio 1.40, 95% CI 1.10- 1.77). The authors concluded that PPIs other than pantoprazole negate the therapeutic effect of clopidogrel in such patients. However, the difference in the effect of pantoprazole compared to other PPIs is not statistically significant. The point estimate of the effect of other PPIs (odds ratio 1.40) lies within the 95% confidence interval associated with the effect of pantoprazole (95% CI 0.70 to 1.47). A formal test for heterogeniety of these odds ratios (2) also shows no statistically significant difference between the effect of pantoprazole and the other PPIs (Chi-square 2.99, df = 1, P=0.08). While this study indicates concurrent use of PPIs and clopidogrel is associated with increased risk of recurrent myocardial infarction, it does not contain definitive evidence to make a distinction between the clopidogrel-inhibiting effects of pantoprazole and other PPIs. More research is needed to elucidate the mechanism of action of the clopidogrel/PPI interaction and determine whether some PPIs are safer than others. 1. Juurlink DN, Gomes T, Ko DT et al. A population-based study of the drug interaction between proton pump inhibitors and clopidogrel. CMAJ 2009;180(7). DOI:10.1503/cmaj.082001 2. Breslow NE, Day NE. Statistical methods in cancer research: Vol. I - The analysis of case-control studies. Lyon: IARC; 1980: Scientific Publ. No 32. Pamela McLean-Veysey, BSc (Pharm) Team Leader, Drug Evaluation Unit Capital Health - Pharmacy Department Halifax, NS Conflict of Interest:None declared |
|||
|
|
|||
|
Mark H Friesen Clinical Pharmacy, Health Sciences Centre Winnipeg
Send letter to journal:
mhfriesen{at}hsc.mb.ca Mark H Friesen
|
The study by Juurlink et. al. is a significant contribution to the awareness of potential interactions with clopidogrel and proton pump inhibitors. However there are a number of important issues that were not addressed in this article.(1) These issues limit the general applicability of the study to clinical practice. First, the authors did not sufficiently address the limitations of their case-control study design. Statements regarding causation do not seem to be warranted based on results from this type of study. The authors make statements such as: “we estimated that about 7.4% of readmissions because of reinfarction… occurred as the result of concomitant therapy with these agents”.(1) Although this study does suggest an association between concomitant therapy and reinfarction rates, the authors did not demonstrate causation as their statement would imply. In addition, given the flaws inherent in a case-control design, higher odds ratios (e.g. >2) are generally required before the results can be seen as indicative of real risk.(2) Secondly, the authors do not address studies that have reported contradictory results to their own. Although they correctly reference Simon et al. as demonstrating the effect of genetic polymorphisms on clopidogrel, they fail to mention the lack of association between proton pump inhibitors and major cardiovascular events in this study.(3)Although they correctly reference abstract 3998 from the American Heart Association Scientific Sessions as demonstrating an increased risk with proton pump inhibitors,(4) they fail to reference abstract 3999 which found that the benefit of clopidogrel was not diminished by baseline proton pump inhibitor use.(5) This report is of particular interest since it was an analysis of a randomized control trial which would have better control of confounders than their case-control study. Thirdly, the authors do not address whether all three of the implicated proton pump inhibitors (omeprazole, lansoprazole, and rabeprazole) equally contribute to the reported odds ratio of 1.40. Of particular interest would be the results for rabeprazole. A reasonable case could be made that rabeprazole is sufficiently different from the other two agents to justify separate analysis. Both lansoprazole and omeprazole have been shown to inhibit the antiplatelet activity of clopidogrel in vivo.(6,7) To date no such study has been reported for rabeprazole. In contrast with omeprazole and lansoprazole, rabeprazole is not a potent in vitro inhibitor of CYP2C19.(8) In contrast with omeprazole, rabeprazole showed no inhibition in the metabolism of CYP2C19 substrate diazepam.(9) Although the metabolite rabeprazole thioether is a potent in vitro inhibitor of CYP2C19, its concentration in vivo is approximately 30% of its parent compound.(8,10) Given the differences cited above it is far from clear that rabeprazole would exhibit comparable interaction potential with clopidogrel as omeprazole and lansoprazole. Thus it would be important to know whether the increased reinfarction rate seen in this study was consistent with all the 3 PPI’s implicated. This was not addressed in the article. 1. Juurlink DN, Gomes T, Ko DT et al.A population-based study of the drug interaction between proton pump inhibitors and clopidogrel. CMAJ 2009;180(7). DOI:10.1503/cmaj.082001 2. Levine M, Waiter S, Lee H: et al. Users Guides to the Medical Literature IV. How to use an article about harm. JAMA 1994;271:1615-1619. 3. Simon T, Verstuyft C, Mary-Krause M et al. Genetics determinants of response to clopidogrel and cardiovascular response. N Engl J Med2009;360-75. 4. Aubert RE, Epstein RS, Teagarden JR et al. Abstract 3998: Proton pump inhibitors effect on clopidogrel effectiveness: the Clopidogrel Medco Outcomes Study. Circulation 2008;118:S_815. 5. Dunn SP, Macaulay TE, Brennan DM et al. Abstract 3999 :Baseline proton pump inhibitors use is associated with increased cardiovascular events with and without the use of clopidogrel in the CREDO trial. Circulation 2008;118:S_815. 6. Small DS, Farid NA, Payne CD, et al. Effects of the proton pump inhibitor lansoprazole on the pharmacokinetics and pharmacodynamics of prasugrel and clopidogrel. J Clin Pharmacol 2008;48:475-84. 7. Gilard M, Arnaud B, Cornily JC, et al. Influence of omeprazole on the antiplatelet action of clopidogrel associated with aspirin: the randomized, double-blind OCLA (Omeprazole CLopidogrel Aspirin) study. J Am Coll Cardiol 2008;51:256-60. 8. Li XQ, Andersson TB, Ahlstrom M, et al. Comparison of inhibitory effects of the proton pump-inhibiting drugs omeprazole, esomeprazole, lansoprazole, pantoprazole, and rabeprazole on human cytochrome P450 activities. Drug Metab Dispos 2004;32:821-7. 9. Ishizaki T, Chiba K, Manabe K et al. Comparison of the interaction potential of a new proton pump inhibitor, E3810, versus omeprazole with diazepam in extensive and poor metabolizers of S- mephenytoin 4’-hydroxylation. Clin Pharmacol Ther 1995;58:155-64. 10. Setoyama T, Drijfhout WJ, van de Merbel NC et al. Mass balance study of [14C] rabeprazole following oral administration in healthy subjects. Int J Clin Pharmacol Ther 2006;32:466-73. Conflict of Interest:None declared |
|||
| HOME | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | HELP | SEARCH |
|
|||||
|
Copyright 1995-2010, Canadian Medical Association. All rights reserved. ISSN 1488-2329
(e) 0820-3946 (p)
All editorial matter in CMAJ represents the opinions of the authors and not necessarily those of the Canadian Medical Association. |
|||||
eLetters: