Influenza vaccination to prevent myocardial infarction
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To identify, retrieve and assess all randomised controlled studies evaluating the effects of influenza vaccines on acute coronary syndromes in terms of mortality and cardiovascular events. The following hypotheses will be tested: in comparison between groups intended for influenza vaccination compared with control groups, there is no difference in mortality and the number cardiovascular events.
Background
Several risk factors are recognised to contribute to coronary heart disease (CHD), but together they do not explain the total burden of disease. At present about one third of all cardiovascular events occur in individuals who do not have identifiable risk factors (Ridker 2000). Additional and potentially modifiable risk factors for the development of coronary heart disease have to be detected and treated. Recent studies have suggested that influenza may be one of these risk factors. Infection with influenza may put patients transiently in danger of an acute cardiovascular event. The incidence of myocardial infarction (MI) and the risk of death from cardiovascular disease (CVD) have a seasonal variation, with peak rates during the winter (Enquselassie 1993; Sheth 1999). Although many factors could contribute to this association including temperature variation (Danet 1999), seasonal variation in fibrinogen, platelets and factor VII (Crawford 2003; Woodhouse 1994) and season dependent behaviour, it has also been proposed that influenza activity could be a possible causal link (Bainton 1978, Meier 1998). This is underlined by the several observations: First
the number of deaths from cardiovascular disease increases during and after epidemics of influenza (Glezen 1987; Scragg 1985); Second several reports describe that influenza infection precipitates acute coronary syndrome (ACS). Spodick showed in a case‐control study that patients with myocardial infarction were more likely to have signs of an acute respiratory infection in the two weeks before the event (Spodick 1984); and third, vaccination against influenza infections contribute to decreased mortality of overall causes (Voordouw 2003) and may prevent cardiac complications in elderly patients. In 2000 Naghavi and Siscovick reported independently that influenza vaccination was associated with a reduction in the risk of myocardial infarction in the subsequent influenza season (Naghavi 2000; Siscovick 2000).
Recently, two large cohort studies have reported conflicting data on the effect of vaccination on acute coronary syndromes. In a large observational study, in which more than 140.000 patients were followed for two years, vaccination against influenza was associated with a reduction in the risk of hospitalisation for cardiac causes (P < 0,001) (Nichol 2003). These findings are conflicting with a study by Jackson et al (Jackson 2002). Jackson described a cohort of survivors from a first myocardial infarction, in which after 3267 person‐years there was no evidence that influenza vaccination protects against recurrent coronary events. In a small, unblinded, randomised controlled trial, influenza vaccination was associated with lower rates of death, myocardial infarction, or revascularization compared to no vaccination among persons with an acute coronary syndromes and those scheduled to undergo percutaneous coronary intervention with a beneficial effect up to two years (Gurfinkel 2004). Influenza is postulated as a risk factor of acute cardiovascular events. Rather than initiating or aggravating a chronic vascular response, influenza may trigger destabilization of already present vulnerable plaques. Different mechanisms could contribute to this process including pronounced expression of inflammatory cytokines which may induce local inflammatory reaction with plaque rupture (Naghavi 2003) and a pro thrombotic state (Visseren 2000). Influenza as a risk factor for acute cardiovascular events could be modified since effective influenza vaccines are available.
The aim of this systematic review is to assess the efficacy of influenza vaccination in the primary and secondary prevention of acute coronary syndrome.
Objectives
To identify, retrieve and assess all randomised controlled studies evaluating the effects of influenza vaccines on acute coronary syndromes in terms of mortality and cardiovascular events. The following hypotheses will be tested: in comparison between groups intended for influenza vaccination compared with control groups, there is no difference in mortality and the number cardiovascular events.
Methods
Criteria for considering studies for this review
Types of studies
Any randomised clinical trials comparing influenza vaccination with placebo or no treatment and with data on one of the outcomes will be included. Studies with control vaccine or no intervention or comparing types, doses or schedules of influenza vaccine will be included if placebo or control is added in the study.
Types of participants
People of all ages who are randomised to either influenza vaccination or placebo or no treatment will be included. Participants will be included if influenza vaccination is given as a routine prevention program, as primary prevention of myocardial infarction and as secondary prevention. Participants may have or have not a myocardial infarction. Participants may be of either sex. Participants on medication will be included.
Types of interventions
Influenza vaccines
Patients must be randomised to either influenza vaccines administered by any route, at any dosage or placebo. As a comparison group either a placebo or no vaccination will be accepted. Influenza vaccines may include inactivated whole virus, detergent‐treated split products, or purified hemagglutinin and neuramidase surface antigen formulation of the three influenza virus strains (influenza A, B or C). Studies in which co‐interventions are used will not be rejected.
Types of outcome measures
The following clinical outcome will be considered during 1 year follow up.
(1) myocardial infarction or re‐infarction;
(2) unstable angina;
(3) death from cardiovascular causes;
(4) death from any cause;
(5) any combination of the above mentioned outcomes;
(6) all adverse events.
Search methods for identification of studies
We aim to identify as many relevant randomised controlled trials as possible by
(1) Searching electronic, medical, bibliographic databases; The Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE (through Ovid) and EMBASE (through Ovid);
(2) Extensive manual searching with check of references from original articles and pertinent reviews;
(3) Searching web sites for recent or ongoing trials including (http://www.epi.bris.ac.uk/Cochrane/cardi.html) and "digital dissertations" (wwwlib.umi.com/dissertations).
(4) Contacting suppliers of influenza vaccine for information on published and unpublished relevant trials.
No language restriction will be applied.
Search strategies for electronic bibliographic databases
We will search CENTRAL using the specific search strategy given below. This search strategy has been developed in accordance with the Cochrane Heart Group guidelines using a combination of free text words and thesaurus terms. CENTRAL incorporates all controlled trials from EMBASE and MEDLINE with exception of the most recent years. To avoid duplicates, we will search EMBASE and MEDLINE only for those years in which controlled trials would not have been included in CENTRAL We will use search strategies as reported in Table 1 for MEDLINE and Table 2 for EMBASE using appropriate filters to identify RCTS (Dickersin 1994; Lefebvre 1996).
| MEDLINE (supplier OVID) will be searched with a standard filter for RCT (Dickersin 1994) |
| #1 Influenza Vaccine/ |
| EMBASE (at http://gateway1.ovid.com) with a standard filter for RCT (Lefebvre 1996) |
| #1 exp Influenza Vaccine/ |
Search Strategy for CENTRAL on The Cochrane Library
#1 INFLUENZA VACCINE
#2 (influenza* near vaccin*)
#3 (influenza* near immuni*)
#4 (flu near immuni*)
#5 (flu near vaccin*)
#6 (#1 or #2 or #3 or #4 or #5)
#7 CARDIOVASCULAR DISEASES
#8 heart
#9 coronary
#10 cardiac
#11 myocardial
#12 cardiovascular
#13 angina
#14 (#7 or #8 or #9 or #10 or #11 or #12 or #13)
#15 (#6 and #14)
Data collection and analysis
Study selection
All studies identified by the search strategy will be scanned. Selection of studies which will be independently performed by two reviewers (TTK and ATAM). The studies will be rejected if the reviewer can determine from the title and abstract that the study is not a randomised controlled trial or controlled clinical trial. Selected trials will be compared and any disagreement will be solved by discussion and consensus. When a abstract cannot be excluded with certainty the full text of the report will be obtained for further evaluation. The full text is assessed by two reviewers independently. A predefined "in‐out" form will be used for evaluation. If a trial is excluded after this point a record of the article and the reason for exclusion will be kept.
Quality assessment
The methodological quality of selected studies will be assessed based on Section 6 of The Cochrane Reviewers' Handbook (Alderson 2004) and will consist the following factors:
ASSESSMENT OF STUDY QUALITY
(1) Method of randomisation ‐ each factor will be marked as 'done', 'not done' or 'unclear'.
Can the study be described as randomised (including use of words such as "random", "randomly" and "randomisation"). Does the study describe the method of randomisation and was it an appropriate method. Methods of allocation using date of birth, date of admission, hospital numbers, or alternation should not be regarded as appropriate, because it does not allowed each study participant to have the same chance of receiving each intervention;
(2) Concealment of allocation ‐ it will be scored A (adequate), B (unclear), C (inadequate), following criteria adopted from the Cochrane Reviewer's Handbook (Alderson 2004) and Schulz et al (Schulz 1995).
A ‐ Adequate measures to conceal allocations such as central randomisation; serially numbered, opaque, sealed envelopes; or other descriptions with convincing concealment.
B ‐ Unclearly concealed trials, in which the authors either did not report allocation concealment at all, or reported an approach that did not fall into one of the categories in (A).
C ‐ Inadequately concealed trials, in which the method of allocation was not concealed, such as alteration methods or use of case record numbers;
(3) Blinding ‐ each factor will be marked as 'done', 'unclear' or 'not done'
Whether participants and investigators were blinded for the treatment will be marked as done (both blinded), unclear (blinding methods are unclear, only the participant is blinded) and not done (no blinding at all);
(4) Intention‐to‐treat analysis ‐ will be marked as 'done', 'not done' or 'unclear'
Whether an intention‐to‐treat analysis possible on all patients from the published data and the number of patients who were lost to follow‐up. If there were no withdrawals it should be stated in the article, otherwise it will be marked as 'unclear'.
Based on these criteria, studies will be subdivided into the following three categories:
A ‐ all quality criteria met: low risk of bias.
B ‐ one or more of the quality criteria only partly met: moderate risk of bias.
C ‐ one or more criteria not met: high risk of bias.
Trials will be independently assessed by two reviewers (TTK and ATAM) and disagreement will be resolved by discussion with the third reviewer. A judgement will be made based on consensus. To avoid selection bias no study will be rejected because of methodological characteristics or any subjective quality criteria. However differences in study methods will be taken into account in a sensitivity analysis. Additionally, we will explore the influence of individual quality criteria in a sensitivity analysis.
Data extraction
The same reviewers will independently extract data from each trial with a predefined review form. Differences in interpretation will be resolved by discussion. The third reviewer will be consulted when differences cannot be resolved.
Abstracted data will include the following:
(1) general information: published/unpublished, title, authors, source, country year of publication, duplicate publications;
(2) trial characteristics: design, duration, randomisation (and method), allocation concealment (and method), blinding (outcome assessors), checking of blinding;
(3) intervention: type of vaccination, method of vaccination;
(4) participants: exclusion criteria, total number and number in comparison groups, gender/age, similarity of groups at baseline, withdrawals/losses to follow‐up;
(5) outcome: myocardial infarction or reinfarction, unstable angina, death from cardiovascular causes, death from any cause.
Data analysis
Data will be analysed with RevMan version 4.2.2. Primary measures of interest will be the effect of influenza vaccination on:
(1) myocardial infarction or reinfarction;
(2) unstable angina;
(3) death from cardiovascular causes;
(4) death from any cause.
Quantitative analysis of outcome will be based on an intention‐to‐treat principle. Measure of effect for each study will be the relative risk with 95% confidence interval (CI). The overall treatment effect will be estimated by the pooled relative risk with 95% CI using a fixed model (Mantel‐Haenszel) or a random effect model (DerSimonian and Laird). Tests for heterogeneity will be performed by a standard chi‐squared test. If the chi‐squared test is significant (P>0.1) and heterogeneity is present, an attempt will be made to explain the differences. This will be based on patient's clinical characteristics and intervention and characteristics of included studies.
Funnel plots will be used to assess for evidence of bias. Sensitivity analysis will be used to take into account the influence of various factors e.g. (a) study quality, (b) exclusion from the analysis of any unpublished studies, (c) exclusion of large trials on effect size. Subgroup analyses will be performed, where feasible, to assess whether particular groups of subjects could obtain more benefit from intervention than other groups e.g. diabetics, those with diagnosed heart disease, those with risk factors for heart disease.
| MEDLINE (supplier OVID) will be searched with a standard filter for RCT (Dickersin 1994) |
| #1 Influenza Vaccine/ |
| EMBASE (at http://gateway1.ovid.com) with a standard filter for RCT (Lefebvre 1996) |
| #1 exp Influenza Vaccine/ |
