Impact of proton pump inhibitor treatment on gastrointestinal bleeding associated with non-steroidal anti-inflammatory drug use among post-myocardial infarction patients taking antithrombotics: nationwide study
BMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h5096 (Published 19 October 2015) Cite this as: BMJ 2015;351:h5096All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
It's amazing and pleasantly surprising to go through a study involving more than a hundred thousand patients having an acute event. The extra ordinary efforts and hard work of the researchers to track such a big cohort for such a prolonged time deserve special compliment and applause from our side.
Nevertheless, we request the authors to provide us a little extra information, not included in the article, and a little clarification about a few points we are interested in after we discussed the article.
A) Under the title 'Methods" it's written that only those patients were included who were admitted as a case of first Myocardial Infarction (MI). In a box on the first page of the article, there is box with the title "WHAT THIS STUDY ADDS". Here conclusions are presented in such a way so that there is no indication of aforementioned inclusion criteria. Hence there is an apparent discrepancy between what this study looks at and what conclusion it draws. Please clarify the point.
B) Patients admitted to a Coronary Care Unit (CCU) have high CV mortality. But even among these patients some patients have a higher mortality than others. These higher risk patients are those having recurrent ischemia (1) . These patients are diagnosed by their clinical characteristics and specialized investigations. These features include recurrent chest pain (in symptoms) , transient hemodynamic instability, transient ominous arrhythmia on cardiac monitor, transient murmur of MR or MVP on cardiac auscultation (in physical examination), dynamic ST - T changes, persistent advanced heart block, new IVCD (in routine ECG), reversible perfusion defect on myocardial perfusion imaging.
As these patients are at high risk of CV mortality, they are likely to get aggressive anti-platelet therapy, including dual drugs (both Aspirin and Clopidogrel). Therefore these patients are likely to bleed more. We are interested to know whether these patients derived a greater benefit (relative risk reduction in statistical terms) from bleeding after the usage of PPIs. If there is a subgroup analysis of such patients, please provide this to us.
C) Figure 1 of the article classifies 45,796 patients as excluded from the study. The figure says " Death from other causes during 30 days after discharge from hospital (n = 22 518)". What we understand is that 22 518 patients died shortly after discharge from hospital, after they recovered from MI. In simple terms, 1 out of every 6th patient discharged from the hospital died. To our experience this high and scary statistics. 1 Year mortality rate of MI is about 15% (2) . We are curious to know these "other" causes causing so many deaths.
D) The same figure also says that 15 353 patients got " no anti thrombotic treatment". In our experience, shortly after discharge, patients are too scared about their life - and - death issue and due to experience of severe pain recently to avoid regularly taking their treatment. Then why did these 12% of the cohort population not get anti-thrombotic treatment, we want to know. If there are physician factors, please let us know.
E) There are only 4.1% diabetics ( with complications); 7.7% received glucose lowering drugs, 0.8% with COPD in the cohort, as provided in a table under the title ' Baseline characeristics". On the other side, as per the report of the Institute for Health Metrics and Evaluation, University of Washington (2013), in high - income countries, among the leading causes of disease - burden (2010), COPD is the 6th and Diabetes Mellitus the8th leading cause (3). Patients with these comorbidities are regularly admitted in our CCUs, some with a diagnosis of first MI. Patients with chronic renal failure are also frequently seen here. Why are patients with these common comorbidities infrequent in this study cohort, we are left wondering.
F) The same table states that only 0.3% patients presented with shock. Whereas we regularly see patients getting admitted with shock in our setup. Judith S. Hochman and David H. Ingbar describe an incidence of shock to range between 5-7 % in the first decade of the millennium in acute MI ( 4). We are interested to know that why is there a lesser number of patients presenting with shock here.
G) After discharge from a hospital, patients with MI are recommended Aspirin in a dose of 75 to 162 mg.( 5). We'd like to know that in what dose is Aspirin prescribed to the Danish cohort, who are subsequently also prescribe NSAIDs and then PPIs too.
There is an age-old saying that well designed and properly conducted research opens doors to several questions and their answers to even more questions. In this way the light of knowledge spreads far and wide. We too believe that answers to these posers will carry forward the message of this article to even farther distances.
References -
(1) J Anderson et al : J Am Coll Cadiol 61 : e 179, 2013
(2) Antman E M , Loscalzo J , ST Segment Elevation Myocardial Infarction , in Harrison's Principles of Internal Medicine 2015 : 19; 1599
(3) Institute for Health Metrics and Evaluation , University of Washington ( 2013 ) data are available through healthmetricsandevaluations.org/gbd/visualizations/country
(4) Hochman J S , Ingbar D H , Cardiogenic Shock and Pulmonary Edema , in Harrison's Principles of Internal Medicine 2015: 19 ; 1759
(5) Antman E M , Loscalzo J, ST Segment Elevation Myocardial Infarction , in Harrison's Principles of Internal Medicine 2015 : 19; 1603
Competing interests: No competing interests
Re: Impact of proton pump inhibitor treatment on gastrointestinal bleeding associated with non-steroidal anti-inflammatory drug use among post-myocardial infarction patients taking antithrombotics: nationwide study
We appreciate the interest in our study from Dr. Gupta and Dr. Verma.
A) Our colleagues rightly note that in the accompanying word-limited information box, the condensed description of the study does not mention that the population under investigation was first-time myocardial infarction. The abstract, full text and Figure 1 each state clearly that the study population was comprised of patients who had suffered a first myocardial infarction. We apologize if some readers had the impression that patients with multiple coronary ischemic events were studied.
B) Unfortunately, the databases used for this study do not hold information on the degree of severity of admitted patients with first-time myocardial infarction.
C-D) Interesting points raised by our colleagues. Performing studies using administrative registries and databases, we experience that from linkage and extraction of data to manuscript writing and data presentation, information on subjects not fulfilling the study inclusion criteria is not easily interpreted without specific study in its own right. Our focus in this study was the cohort of patients with first-time myocardial infarction who survived at least 30 days and who claimed at least one prescription of an antiplatetelet agent (i.e. likely to be discharged in a condition where life-saving treatment can be initiated). The proportions excluded because of death and not claiming antithrombotic treatment are not readily interpretable and cannot be compared with those in observational studies based on non-administrative-based cohorts or in randomized controlled trials. Investigation of early mortality and of discharges without anti-thrombotic treatments would be separate exercises.
E-F) We agree that the numbers with diabetes/taking oral glucose-lowering medicine, chronic renal failure, (COPD see below) and shock seem low but bear in mind that the population studied was comprised of relatively young patients who had suffered a first myocardial infarction. Studies of all-comer patients with myocardial infarction will include those with recurrent as well as first myocardial infarctions and are likely to represent a somewhat sicker overall population in whom greater prevalence of co-morbidities might be expected. In addition, the listed co-morbidities are derived from administrative coding. As a result, some complications are potentially not registered in full (as other and possibly more acute reasons for hospitalization results in other codes being used).
Concerning COPD, the authors have drawn our attention to an error in the table that we did not notice previously - the numbers illustrate codes for pulmonary oedema (and not COPD). We have alerted the journal. As stated we used variables incorporated in the Ontario Acute Myocardial Infarction Mortality Prediction Rules (1) where COPD is not included (but pulmonary oedema is). For diabetes, the percentage is slightly lower than expected compared to other studies from our group. We were rather restrictive in our definition of other pharmacological treatment in the present study when only including prescriptions claimed in the 30-day quarantine period. Including use 30 days prior to the beginning of the quarantine period (i.e. total of 60 days prior to inclusion) a further 3,082 patients claimed prescription of oral glucose-lowering medicine resulting in 11.4%. As we are aware of this type of limitation, we also used (which can be found in the manuscript) a model which incorporated knowledge of subsequent hospitalizations/prescription during the entire follow-up period. Our results and conclusions were unaffected.
G) Only aspirin dosages of 75 mg daily were used as recommended by national guidelines. There is no tradition in Denmark for higher dosages following myocardial infarction.
We thank the writers for their valuable comments and their curiosity on aspects of the study. To paraphrase Sam Johnson, curiosity is one of the permanent and certain characteristics of a vigorous intellect.
Reference
(1) Tu JV, Austin PC, Walld R, Roos L, Agras J, McDonald KM. Development and validation of the Ontario acute myocardial infarction mortality prediction rules. J Am Coll Cardiol 2001;37:992-7.
Competing interests: No competing interests