Antibiotics for secondary prevention of coronary heart disease
This is the protocol for a review and there is no abstract. The objectives are as follows:
To assess the beneficial and harmful effects of antibiotics given as secondary prevention to patients with CHD.
Description of the condition
Coronary heart disease (CHD) is the narrowing or blockage of the coronary arteries, usually caused by atherosclerosis due to build-up of fatty material and plaque. The Atherosclerotic pathological process in the walls of blood vessels is complex and develops over many years. CHD is the leading cause of death worldwide, particularly in low-middle income and high income countries, with an estimated 3.8 million men and 3.4 million women dying each year (WHO 2004; WHO 2011). Common risk factors of CHD include tobacco use, physical inactivity, overweight and obesity, hypertension, diabetes, hyperlipidaemia, advancing age, gender, and inherited (genetic) disposition (WHO 2011).
Infections may be involved in the pathogenesis of atherosclerosis (Fabricant 1978). It has been suggested that the pathogenic mechanism of the various infectious agents might be an increased inflammatory response seen in connection with infections (Aceti 1996). A number of studies have assessed the association between CHD and various infectious agents, notably Chlamydia pneumoniae (C. pneumoniae), Helicobacter pylori (H. pylori), and cytomegalovirus, but the results are equivocal.
There appears to be an association of C. pneumoniae and CHD based on studies demonstrating the presence of C. pneumoniae in atheromatous plaques (Shor 1992; Kuo 1993; Muhlestein 1996; Ramirez 1996), on retrospective seroepidemiological studies demonstrating an increased amount of antibodies in patients with acute or chronic CHD (Saikku 1988; Leinonen 1990; Thom 1991; Thom 1992; Linnanmaki 1993; Melnick 1993; Patel 1995; Dahlen 1995; Kazar 2005), on a prospective seroepidemiological study (Romano Carratelli 2006) , and on a meta-analysis of seroepidemiological studies (Danesh 1997). However, the results are equivocal and two prospective studies (Ridker 1999; Danesh 2000a; Danesh 2000b, Danesh 2002) and a meta-analysis of prospective seroepidemiological studies (Danesh 2000a; Danesh 2000b, Danesh 2002) could not find any strong association between C. pneumoniae immunoglobulin G nor immunoglobulin A titres and CHD.
Some epidemiological studies have shown a modest association between antibodies to H. pylori and CHD (Danesh 1997), and a recent prospective seroepidemiological study (Lenzi 2006) showed a higher overall prevalence of H. Pylori in patients with CHD. Larger studies and a meta-analysis of prospective studies, however, could not demonstrate any association (Danesh 1998), nor could two prospective studies (Folsom 1998; Roivainen 2000).
A recent meta-analysis on the risk of CHD in relation to cytomegalovirus infection demonstrated an increased in CHD risk for people exposed to cytomegalovirus infection in both prospective studies and retrospective studies (Ji 2012).
Description of the intervention
Antibiotics are of different classes, for example penicillins, macrolides or tetracyclines. A few randomized clinical trials have assessed the effects of macrolide antibiotics given as secondary prevention to patients with CHD and known C. pneumoniae seropositivity (Gupta 1997; Anderson 1999), C. pneumoniae and/or H. pylori seropositivity (Torgano 1999), or unknown antibody status (Gurfinkel 1999; Neumann 2001). The results are equivocal. A statistically significant reduction in the incidence of coronary events in the group receiving antibiotics compared with the group receiving placebo were reported by one trial at 18 months follow-up (Gupta 1997), and at the end of 30 days treatment but not at six months follow-up by another trial (Gurfinkel 1999). Finally, a third trial found no significant difference in coronary events at the end of treatment or at 24 months follow-up (Anderson 1999). The results of these three trials are all limited by the small number of patients included (from 60 to 302). There is a great risk of both false positive responses, i.e. the reduction in the number of events in the antibiotic-treated group occurred by chance, and false negative responses, i.e. the trials were too small and underpowered to demonstrate a clear benefit or harm. Although C. pneumoniae has been the main target and macrolide antibiotics have been the principal antibiotic treatment in most studies, C. pneumoniae is susceptible to a variety of antibiotics of which some have been assessed in randomized trials (Sinisalo 1998). Further, various antibiotic treatments might be used against other bacterial agents that may be associated with atherosclerosis, e.g. H. pylori. Moreover, antibiotics may have general anti-inflammatory effects that could affect the atherosclerotic lesions (Agen 1993).
How the intervention might work
The mechanism by which antibiotics may prevent CHD is not established. It is assumed that it is through an antibacterial activity. However, macrolides also exert anti-inflammatory activity (Agen 1993).
Why it is important to do this review
There have been several attempts to meta-analyze randomized trials assessing antibiotics for secondary prevention of CHD (Andraws 2005; Jespersen 2006; Gluud 2008). Andraws 2005 analyzed 11 trials with 19,217 participants and could not demonstrate a significant benefit. We meta-analyzed 13 trials with 24,063 participants and could not demonstrate benefit or harm, but the estimate showed a relative increase of mortality of 9% in the antibiotics group (Jespersen 2006). When we updated the analysis with 17 trials and 25,271 patients, there was a significant increased mortality of 10% in the patients receiving antibiotics (Gluud 2008). These analyses were not conducted as systematic reviews taking into consideration the risk of systematic errors and the risk of random errors (Wetterslev 2008; Keus 2010; Higgins 2011). In the present systematic review, we will collect and present current evidence of antibiotic treatment given as secondary prevention for CHD.The review will attempt to assess if the efficacy of antibiotics in general and macrolide antibiotics in particular is related to C. pneumoniae, H. pylori, or cytomegalovirus titres.
To assess the beneficial and harmful effects of antibiotics given as secondary prevention to patients with CHD.
Criteria for considering studies for this review
Types of studies
We will include all secondary prevention randomized clinical trials, irrespective of trial design or setting, blinding, publication status, publication year, or language. we will include unpublished trials if the methodology and the data of the trial can be accessed in written form. We will also include non-randomized studies (e.g. quasi-randomized studies and observational studies) for the assessment of harms. We will use such studies as we identify them, but we will not conduct specific searches to identify them.
Types of participants
Patients with a diagnosis of acute or chronic CHD (i.e. acute or previous myocardial infarction; stable or unstable angina) according to the definitions of the individual trials. We will also include trials with no explicit definitions. Patients can be of any age, sex, and have known or unknown antibody status for C. pneumoniae, H. pylori, and cytomegalovirus.
Types of interventions
Any kind of antibiotics in any dose, rout of administration, or duration versus placebo or no intervention. We will accept collateral experimental interventions if given to both intervention groups of the trial.
Types of outcome measures
We will consider all outcomes at the end of treatment and at the maximum follow-up according to the individual trial.
Serious adverse events, defined according to the ICH guidelines as any event that would increase mortality; was life-threatening; required inpatient hospitalisation or prolongation of existing hospitalisation; resulted in persistent or significant disability; or any important medical event, which may have jeopardized the patient or required intervention to prevent it (ICH-GCP 1997).
Any adverse event defined as any untoward medical occurrence in a patient, which did not necessarily have a causal relationship with the treatment, but resulted in a dose reduction or discontinuation of treatment.
Search methods for identification of studies
We will search the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library.We will supplement this by searches of MEDLINE Ovid, EMBASE Ovid and ISI Web of Science. The preliminary search strategy that is given in Appendix 1 will be used to search MEDLINE and adapted appropriately for use with other databases. The search will include the addition of a standard randomized control trial filter for MEDLINE and EMBASE(Higgins 2011).
Searching other resources
We will search for ongoing trials on:
The reference lists of relevant publications will be checked for any unidentified randomized trials. We will contact authors of included studies, and major pharmaceutical companies, by post or email asking for unpublished randomized trials.
We will apply no language restrictions.
Data collection and analysis
Selection of studies
Two authors (MS, BG) will independently assess trial eligibility without blinding of the trial authors, investigators, institution, source, or results. Disagreements will be resolved by discussion with a third author (CG). we will list excluded trials with the reason for exclusion. Study selection will be displayed in an adapted flow diagram as per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Moher 2009).
Data extraction and management
Two authors (MS, BG) will extract all data independently. Disagreements will be resolved by discussion with a third author (CG or JW or AH). We will assess duplicate publications and companion papers of a trial together in order to evaluate all available data simultaneously (maximize data extraction, correct bias assessment). Cost and recourse data, if available, will be extracted, but not analyzed. we will approach the authors of the trials to specify any of the following data, had they not been reported sufficiently in the publication:
Bias risks components (as defined below); trial design (parallel, factorial, or crossover); number of intervention arms; length of follow-up; estimation of sample size; use of intention-to-treat analysis; relevant inclusion and exclusion criteria.
Participant characteristics and diagnosis
Baseline characteristics:number of participants; age range (mean or median) and sex ratio; type of CHD (acute or chronic); presence of cardiovascular risk factors (i.e. diabetes mellitus, hypertension, hyperlipidaemia, or smoking); method for,detection and presence of antibodies for C. pneumoniae, H. pylori, and cytomegalovirus.
Type of antibiotics; dose of intervention; duration of therapy; mode of administration; type and dose of collateral intervention(s).
All outcome measures will be extracted from each randomized clinical trial, and harms from non-randomized studies, and we will identify incomplete outcomes and selective outcome reporting.
Assessment of risk of bias in included studies
There is a risk of overestimation of beneficial intervention effects in randomized trials with unclear or inadequate methodological quality (Schulz 1995; Jadad 1996; Moher 1998; Jüni 2001; Kjaergard 2001). We will asses the influence of risk of bias on the results by evaluation of the following domains (Table 1):
Table 1. The Cochrane Collaboration's tool for assessing risk of bias
|Domain||Description||Review authors judgement|
|Allocation sequence generation||Describe the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.||Was the allocation sequence adequately generated?|
|Allocation sequence concealment||Describe the method used to conceal the allocation sequence in sufficient detail to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment.||Was allocation adequately concealed?|
|Blinding of participants, personnel and outcome assessors Assessments should be made for each main outcome (or class of outcomes)||Describe all measures used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. Provide any information relating to whether the intended blinding was effective.||Was knowledge of the allocated intervention adequately prevented during the study?|
|Complete outcome data reporting Assessments should be made for each main outcome (or class of outcomes)||Describe the completeness of outcome data for each main outcome, including attrition and exclusions from the analysis. State whether attrition and exclusions were reported, the numbers in each intervention group (compared with total randomized participants), reasons for attrition/exclusions where reported, and any re-inclusions in analysis performed by the review authors.||Were incomplete outcome data adequately addressed?|
|Selective outcome reporting||State how the possibility of selective outcome reporting was examined by the review authors, and what was found.||Are reports of the study free of suggestion of selective outcome reporting?|
|Other sources of bias|
State any important concerns about bias not addressed in the other domains in the tool.
If particular questions/entries were pre-specified in the review's protocol, responses should be provided for each question/entry.
|Was the study apparently free of other problems that could put it at a high risk of bias?|
allocation sequence generation;
allocation sequence concealment;
blinding (participants, personal and outcome assessors);
complete outcome data reporting;
selective outcome reporting; and
other apparent biases (threats to validity).
The risk of bias in the included studies will be assessed as ether "low", "high" or "unclear" according to the criteria in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).
Assessment of reporting biases
Potential existence of bias will be addressed by funnel plots (Egger 1997), both visual and subsequent appropriate test for funnel plot asymmetry, according to the recommendations by the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).
Unit of analysis issues
Unit of analysis is randomized trials. For trials using cluster randomization, if data is presented without proper control for clustering we will try to obtain individual patient data. If there is insufficient information to control for clustering, the unit of analysis used will be the cluster (Higgins 2011), and bias in this approach will be examined with sensitivity analysis. For trials using crossover design, only data from the first period will be included (Elbourne 2002; Higgins 2011). This in order to minimize any possibility of carry-over effect.
Statistical analysis will be performed using the Cochrane Collaboration's statistical software, Review Manager (Review Manager 2012), and the software Trial Sequential Analysis (TSA 2011). Meta-analysis will be performed according to the intention-to-treat principle using the last reported observed response (carry forward). Binary outcomes will be expressed as relative risks (risk ratios, RRs) and 95% confidence intervals (CIs). Continuous outcomes will be analyzed using the differences in means or standardized mean difference (SMD).
For the assessment of harms, we will include and present non-randomized studies separately. Potential biases are expected to be greater for non-randomized studies, as well as heterogeneity due to different study designs for a combined analysis.
A random-effects and a fixed-effect model will be used (DerSimonian 1986; DeMets 1987). If both analyses show similar results, the random-effect model will be presented. Rare events will be estimated by Peto odds ratio (Deeks 1999). Should a meta-analysis not be possible, or inappropriate we will aim for a narrative assessment.
We will perform trial sequential analysis (TSA) on the primary outcomes. It calculates a diversity adjusted required information size, reduce the risk of random error and prevent premature statement of superiority of either experimental or control intervention (Brok 2008; Wetterslev 2008; Wetterslev 2009; Thorlund 2011). We will assume a type I error of 5%, type II error of 20%, event proportion as estimated in the control group (CEP) of all trials, and an anticipated number needed to treat (NNT) or number needed to harm (NNH) of 200 (for all-cause mortality and cardiovascular death ). As the CEP is a priori very uncertain we use the NNT or NNH as our a priori intervention effect anticipations in the TSA's allowing for quite different relative risk reductions or increase pending the CEP. Further, we will assume a type I error of 5%, type II error of 20%, CEP of all trials, as estimated in the traditional meta-analysis, and number needed to harm of 100 (for number of serious adverse events). We will perform sensitivity analysis in which the estimated required information size will be based upon the highest and lowest possible anticipated intervention effect and I2 from the 95% confidence intervals of the traditional meta-analysis intervention effect and I2, as well as the point estimate of these parameters.
Subgroup analysis and investigation of heterogeneity
Subgroup analysis will be performed analyzing the primary outcomes (all-cause mortality and cardiovascular death) according to methodological quality (high risk of bias or low risk of bias), class of antibiotic, form of CHD (acute or chronic), antibody status, and duration of treatment.
Heterogeneity will be identified by visual inspection of the forest plots, using Chi2 test and a significance level of 0.1 (this test has low power when studies have small sample size, or are few in number), as well as I2 statistics with levels over 50% indicating substantial level of heterogeneity (Higgins 2011). In case significant heterogeneity will be found, the potential reason for it will be explored by performing sensitivity analysis. For the adjustment of the estimation of required information size we will apply the heterogeneity measure D2 as the intuitively obvious adjustment with I2 underestimate information size estimation (Wetterslev 2009)
Cochrane Heart Group is thanked for their expert assistance in creating the search strategy.
Appendix 1. Search strategy
1 exp Coronary Disease/
2 exp Myocardial Ischemia/
3 (MYOCARD* adj3 INFARCT*).tw.
4 (CORONARY adj3 DISEASE*).tw.
5 (MYOCARD* adj3 ISCHEMI*).tw.
6 (MYOCARD* adj3 ISCHAEM*).tw.
7 (ISCHEMIC adj3 HEART).tw.
8 (ISCHAEMIC adj3 HEART).tw.
9 exp Atherosclerosis/
13 Anti-Bacterial Agents/
22 exp Penicillins/
23 exp lactams/ or beta-lactams/
24 exp Tetracyclines/
25 exp Aminoglycosides/
26 exp Macrolides/
27 exp Clindamycin/
28 exp Chloramphenicol/
29 Fusidic Acid/
32 exp Polymyxins/
33 (antibiotic* or anti-bacterial* or anti-infective*).tw.
35 12 and 34
36 randomized controlled trial.pt.
37 controlled clinical trial.pt.
40 drug therapy.fs.
44 36 or 37 or 38 or 39 or 40 or 41 or 42 or 43
45 exp animals/ not humans.sh.
46 44 not 45
47 35 and 46
|11 May 2012||New citation required and major changes||Authorship changed with new author and new contact person. The methods have been updated in line with more recent Collaboration guidelines.|
Protocol first published: Issue 2, 2002
|11 August 2008||Amended||Authorship changed with new author and new contact person.|
|12 June 2008||Amended||Converted to new review format.|
Contributions of authors
Maria Skoog drafted the protocol based on a previous version authored by Tea Monk-Hansen, Eva Prescot, Bodil Als-Nielsen, Asbjørn Hróbjartsson, Jørn Wetterslev and Christian Gluud.
Asbjørn Hróbjartsson, Jørn Wetterslev, Berit Grevstad and Christian Gluud amended the protocol.
Declarations of interest
Berit Grevstad, Asbjørn Hróbjartsson, and Jørn Wetterslev have no potential conflict of interest. Maria Skoog, and Christian Gluud are involved in a randomized trial (CLARICOR) in which the intervention drug (Klacid Uno®) and placebo were donated by Abbott. This systematic review is not supported by any pharmaceutical company.
Sources of support
The 1991 Pharmacy Foundation, Denmark.
The Danish Medical Research Council, Denmark.