Aims To evaluate the effect of influenza vaccination on the coronary events in patients with confirmed coronary artery disease (CAD).
Methods and results Randomized, double-blind, placebo controlled study. We included 658 optimally treated CAD patients; 477 men, mean age 59.9±10.3 years. Three hundred and twenty-five patients received the influenza vaccine, and 333 patients placebo. Median follow-up was 298 (interquartile range 263–317) days. Primary endpoint was the cardiovascular death. Its estimated 12-month cumulative event rate was 0.63% in the vaccine vs. 0.76% in controls (HR 1.06 95% CI: 0.15–7.56, P = 0.95). There were two secondary composite endpoints: (i) the MACE (cardiovascular death, myocardial infarction, coronary revascularization) tended to occur less frequently in the vaccine group vs. placebo with the event rate 3.00 and 5.87%, respectively (HR 0.54;95% CI: 0.24–1.21, P = 0.13). (ii) Coronary ischaemic event (MACE or hospitalization for myocardial ischaemia) estimated 12-month event rate was significantly lower in the vaccine group 6.02 vs. 9.97% in controls (HR 0.54; 95% CI: 0.29–0.99, P = 0.047).
Conclusion In optimally treated CAD patients influenza vaccination improves the clinical course of CAD and reduces the frequency of coronary ischaemic events. Large-scale studies are warranted to evaluate the effect of influenza vaccination on cardiovascular mortality. (ClinicalTrials.gov: NCT 00371098).
A possible relation between influenza and higher mortality from cardiovascular events was first noticed in the 1900s, after influenza pandemic in Europe and the USA.1 Acute myocardial infarction and stroke have peak incidence in winter and the risk of both events increases significantly after respiratory and urinary tract infections.2–5 Inflammation and infection as trigger factors of atherosclerotic plaque rupture are an attractive explanation of acute coronary syndromes (ACS).6 Retrospective analysis and epidemiological studies on large populations have indicated that influenza vaccination might be associated with reduced rates of death, myocardial, and brain infarction.3,5,7–12 In the only randomized, but not double-blind clinical trial, Gurfinkel et al.13 showed reduced rates of mortality and ischaemic events in coronary artery disease (CAD) patients who had received influenza vaccination, in comparison to controls.
The current guidelines are inconsistent. American Heart Association and American College of Cardiology guidelines recommend influenza vaccination in all patients with coronary and other atherosclerotic vascular diseases.14 In contrast, European Society of Cardiology guidelines found no documented evidence of the effects of influenza immunization either on the clinical course of chronic heart failure or stable angina, and demanded placebo-controlled trials.15,16
We performed a prospective, randomized, double-blind, placebo-controlled study to evaluate the effect of influenza vaccination on the incidence of coronary ischaemic events in optimally treated patients with CAD confirmed by coronary angiography.
This study was designed and performed as a single centre, randomized, double-blind, placebo-controlled study. The randomization was 1:1 (influenza or placebo vaccine). The randomization sequence was computer generated by an independent statistician. Identical syringes with vaccine or placebo were labelled with randomization numbers according to the code by persons not involved in the study, then delivered to the site. A study nurse received the syringes in blocks of 50 each, and after each vaccination the syringe number (randomization number) was noted in a patient’s file. All study personnel contacting patients was blinded till the database was closed. The randomization code remained with the independent ethics committee till the database was closed, and the protocol of its opening is included in the study files. The study is registered with ClinicalTrials.gov NCT 00371098 and was done in accordance with the Helsinki Declaration and Good Clinical Practice.
Inclusion and exclusion criteria
Patients were eligible, if they were aged 30–80 years, and had CAD confirmed by angiography with at least 50% stenosis of at least one large epicardial coronary artery. We excluded patients with congestive heart failure NYHA III/IV, planned cardiovascular surgery within 6 months, evolving renal failure, neoplastic disease, psycho-organic disorder or any factor impeding follow-up, and contraindication to vaccination.
Demographic, clinical, and biochemical variables listed in Table 1 were prospectively collected in the study patients.
Baseline characteristics of the study population (n = 658)
Vaccine (n = 325)
Placebo (n = 333)
Male sex, n (%)
Living in town
Influenza vaccination within previous year
Influenza vaccination in a family within previous year
Recent intervention patients, n (%)
Primary PCI for ACS
Ambulatory CAD patients, n (%)
Hypertension, n (%)
Diabetes, n (%)
Smoking, n (%)
Total cholesterol (mmol/L)
Myocardial infarction, n (%)
PCI, n (%)
Coronary bypass, n (%)
Atrial fibrillation, n (%)
Stroke, n (%)
Heart failure, n (%)
Aspirin, n (%)
Thienopiridines, n (%)
Statins, n (%)
β-blockers, n (%)
ACE inhibitors, n (%)
Calcium blockers, n (%)
Nitrates, n (%)
Diuretics, n (%)
Oral hypoglycaemic, n (%)
Insulin, n (%)
Inflammatory markers and infectious burden
n = 124
n = 139
IgG antibodies against Helicobacter pyroli (U/mL)
IgG antibodies against Cytomegalovirus (AU/mL)
Results are presented as median (range) for continuous data, and as percentages for categorical data. CAD, coronary artery disease; PCI, percutaneous coronary intervention; ACS, acute coronary syndrome.
aBNP evaluated in 123 vaccine and 131 placebo patients.
Timing of the study
The study was performed between October 2004 and December 2005. Patients were recruited and vaccinated from October 2004 through February 2005. The follow-up was planned for one influenza vaccine season and was censored for all participants on 31 October 2005.
The study was approved by local Ethics Committee (IK-NP-0021-2/812/04). All participating patients gave the written informed consent.
Vaccination and laboratory tests
Patients randomly received intramuscular single inactivated subunit influenza vaccine containing 0.5 mL dose (15 µg) hemagglutinin of each of the following strains: A/NewCaledonia/20/99 (H1N1), A/ Christchurch/28/03 (H3N2), B/Jiangsu/10/03, or placebo containing all vaccine compounds except viral antigens. Patients hospitalized for percutaneous coronary interventions (PCIs) were vaccinated after the procedure before discharge from the hospital, and outpatients were vaccinated during office visits to cardiologists. In 174 patients, the antibody levels against viral antigens of the vaccine were measured by hemaggutination inhibition test and ELISA test, before the vaccination and 8–10 weeks after, to confirm vaccine effectiveness and seroconversion (the findings will be presented in a separate paper).
Laboratory tests before vaccination: lipid profile, creatinine, hs-CRP, BNP, IgG antibodies against Helicobacter pylori, and Cytomegalovirus were done to evaluate risk factors, inflammatory markers, and infection burden in the study patients.
Influenza epidemiology during the 2004/2005 season
According to the National Influenza Center, the peak of influenza activity in our country is usually registered in March and even in the beginning of April. In the epidemic season analysed, the highest incidence of influenza-like illness was observed between July 2005 and November 2005 and ranged from 415.5 cases per 100 000 to 641.7 cases per 100 000. The highest number of laboratory confirmations was registered in February and March 2005.
In the 2004/2005 season, almost 60% (59.2%) of the influenza virus strains isolated in Poland were antigenically similar to the vaccine strains, i.e A/NewCaledonia/20/99 (H1N1) or A/Wyoming/3/2003 or, B/Jiangsu/10/03 recommended by the European Agency for the Evaluation of Medicinal Products (EMEA) based on WHO recommendations for the use in that season.17 Nevertheless, 40.8% of the isolated strains were characterized as type B similar to B/Hong Kong/330/2001 (B/Victoria/2/87 lineage) a strain distinguishable from the vaccine influenza B strain. The B/Hong Kong/330/2001 strain was included into the influenza vaccine for the previous epidemic season, i.e. 2003/2004 but almost half of the strains of type B isolated in the whole of Europe in the 2004/2005 epidemic season were similar to this strain.18 Using epidemiological data available at the National Institute of Hygiene, Warsaw, it was clear that the incidence of influenza-like illness in Poland in the 2004/2005 epidemic season was lower than in the previous season, i.e. 2003/2004 (1807.9/100 000 when compared with 2366.9/100 000). The numbers above may indicate that there was no serious clinical impact of the mismatch between Polish influenza isolates of type B and vaccine strains of type B recommended for the 2004/2005 season.
Follow up and endpoints
The follow-up data were obtained by telephone, mailed questionnaire, National Citizen Registry consultation, and home visits. All cardiovascular events and study endpoints were confirmed.
The primary endpoint was 12-month cardiovascular death. There were two secondary endpoints: major adverse cardiac event (MACE), which was the composite of: cardiovascular death, acute myocardial infarction (MI), or coronary revascularization (PCI or coronary bypass), and coronary ischaemic event, defined as a combination of MACE or hospitalization for myocardial ischaemia, both assessed at 12-month.
Acute MI was diagnosed when chest pain lasting 20 min or longer was associated with CK-MB or troponin I level twice above normal range. ST-elevation MI was diagnosed when additionally ST-elevation ≥2 mm was present in at least two contiguous ECG leads. Non ST-elevation MI was diagnosed when MI criteria were associated with ST-depression or T-wave inversion.
Coronary revascularization was defined as PCI or coronary bypass, resulting from myocardial ischaemia.
Hospitalization for myocardial ischaemia was defined as at least 24 h hospitalization resulting from chest pain with new ST–T changes in at least two standard ECG leads without markers of myocardial necrosis.
Influenza-like illness was defined as an acute onset of upper respiratory tract infection with temperature above 37.8°C, and two of the following symptoms: cough, sore throat, nasal symptoms, myalgia, headache, or malaise.
All analyses were done on a intention-to-treat basis. There were no crossovers. The criteria of the sample-size choice were based on 1 year MACE incidence after primary PCI in our Institution. Under the assumption of 10% frequency of events in the controls, and 4% of events in the active arm, we estimated the sample size of each group to 298 patients (for significance level of α = 5% and power 1-β = 80%). Results are presented as median, range or 25th and 75th percentile, for continuous variables, and as percentages for categorical data. For all continuous variables, the test of normality was performed by Shapiro–Wilk test. Groups were compared (all tests two-sided) by unpaired t-test or one-way analysis of variance for normally distributed variables or Mann–Whitney–Wilcoxon test, where data were not normally distributed. Frequencies were compared using chi-square or Fisher’s exact test, where appropriate. Survival curves were assessed by Kaplan–Meier method and compared using a log-rank test. Hazard ratios (HR) and 95% CI were calculated by the Cox proportional hazards model. Appropriateness of the proportional hazards assumption was assessed by examination of interaction between time and variables included into the model (log(–log[survival]) curves which showed no important departures from linearity proportionality). Estimated 12-months cumulative rate of the study endpoints in vaccinated vs. placebo patients was presented as percentages based on Kaplan–Meier method. In measuring the time to an event for cases in which a patient had multiple endpoints, the first event was taken into account.
All baseline variables listed in Table 1 were assessed with univariable Cox regression. Multivariable Cox regression model included major causal risk factors for CAD: age, sex, smoking, arterial hypertension, elevated LDL-cholesterol (≥3.12 mmol/L), low HDL-cholesterol (≤0.9 mmol/L), diabetes along with influenza vaccination, and recent ACS.19–21 Non-normally distributed continuous variables were log-transformed prior to inclusion in regression model. The prognostic discriminatory capacity of the resulting model was measured with the c-statistic representing the area under the receiver operating characteristic curves for prediction of the coronary ischaemic event.
Data were analysed using the SAS package, ver. 8.2. P-values below 0.05 were considered significant.
The authors had full access to the data and take responsibility for its integrity. All authors have read and agreed to the manuscript as written.
Total number of 658 Caucasian patients (477 men) aged 59.9 ± 10.3 years were included and vaccinated. Figure 1 shows the trial profile and patients flow. Our patients had optimally treated CAD in a regional reference centre. More than 90% patients received aspirin, statins, ACE- inhibitors, and beta-blockers. Detailed characteristics and comparison of study groups are listed in Table 1. One hundred and fifty-seven patients had primary PCI in ACS or unstable angina, 131 patients had elective PCI in stable angina, and 370 outpatients had stable angina and CAD confirmed by coronary angiography.
Median follow-up time was 298 (IR 263–317) days, range 4–370 (measured to the first event, or to the end of the observation period). No patient was lost to the follow-up.
Three hundred and twenty-five patients received influenza vaccine and 333 patients received placebo vaccine. No severe complications were observed following vaccination. Minor, local, or generalized, transient adverse effects occurred in 13 patients (4.0%) in active arm, and in 10 patients (3.0%) in the placebo arm (P = 0.50). During the follow-up, vaccinated patients reported influenza-like illness significantly less frequently than those who received placebo: 26 patients (8.1%) vs. 43 patients (12.9%), respectively, (P = 0.042).
Study endpoints and overall mortality
The occurrence of study endpoints is summarized in Table 2 and shown graphically in Figure 2.
Kaplan–Meier curves of event free survival for composite study endpoints. (A) major adverse cardiac event-free survival; (B) coronary ischaemic event-free survival
The primary endpoint, cardiovascular death, occurred in two patients in the vaccine and in two patients in the placebo group. The estimated 12-month cumulative cardiovascular death rate was 0.63% in vaccinated and 0.76% in placebo patients (unadjusted HR for the vaccine group 1.06; 95% CI, 0.15–7.56; P = 0.95). Two other patients died of non-cardiovascular cause: one in the active group of gastrointestinal haemorrhage, and one in the controls of lung cancer diagnosed after enrolment, so the overall mortality rate was three patients in the vaccine group and three patients in controls.
The MACE tended to occur less frequently in the vaccine group in comparison to the placebo group, namely the estimated 12-month cumulative event rate was 3.00 vs. 5.87%, respectively, with a HR of 0.54 (95% CI: 0.24–1.21; P = 0.13).
For coronary ischaemic event estimated 12-month cumulative event rate was significantly lower in the vaccine group (6.02%) than in the placebo group (9.97%), with a HR of 0.54 (95% CI: 0.29–0.99, P = 0.047) (Table 2).
Cardiac events during influenza season
We compared the distribution of coronary ischaemic events in controls and vaccinated group during the influenza season. According to the National Influenza Center, the influenza viral circulation period ended on the 30 April 2005.18 By the end of the influenza season the risk of coronary ischaemic events had tended to be lower in vaccinated than placebo patients with HR = 0.46, 95% CI: 0.20–1.05; P = 0.065.
Predictors of coronary ischaemic events
All of demographic, clinical, biochemical, and angiographic variables listed in Table 1 were analysed in search for predictors of coronary ischaemic event during the follow-up. By univariate analysis seven of them: gender, influenza vaccination, history of heart failure, primary PCI for ACS, elective PCI for stable angina (both within a week before vaccination), thienopiridines, and nitrates treatment were associated with coronary ischaemic events (Table 3). Influenza vaccination was a negative predictor associated with significant reduction of the risk of coronary ischaemic event (Table 3).
By multivariable analysis, recent ACS (HR = 2.93, 95% CI: 1.51–5.65; P = 0.0014), female gender (HR = 2.15, 95% CI: 1.11–4.15; P = 0.0235), and influenza vaccination (HR = 0.38, 95% CI: 0.19–0.78; P = 0.0086), emerged as independent predictors of coronary ischaemic event (Table 4). Other examined factors: age, smoking, arterial hypertension, elevated LDL-cholesterol, low HDL-cholesterol, and diabetes were not significantly related to the outcome. The area under the receiver operating curves for the multivariable Cox model, in prediction of the coronary ischaemic event suggested good model discrimination (c-statistic = 0.69, P = 0.001).
Predictors of coronary ischaemic events at follow-up by multivariable analysis
Hazard ratio (95% CI)
Primary PCI for ACS
PCI, percutaneous coronary intervention; ACS, acute coronary syndrome. Other examined factors; age, smoking, arterial hypertension, elevated LDL-cholesterol, low HDL-cholesterol, and diabetes were not significantly related to the outcome.
In contrast to previous, non-placebo-controlled and retrospective studies, we did not find any significant influence of influenza vaccination on the cardiovascular mortality and the incidence of MI.3,5,7,8,12 However, we found a trend to better MACE free survival in vaccinated patients in comparison to controls. Furthermore, in the presented study, vaccination against influenza significantly reduced the risk of coronary ischaemic events (MACE or hospitalization for myocardial ischaemia) in comparison to placebo group. In multivariable analysis, influenza vaccination emerged as an independent factor associated with lower incidence of coronary ischaemic events.
Naghavi et al.8 were the first who reported a reduction of 67% (95% CI: 18–87%) in the risk of subsequent MI in the case-control study on 218 CAD patients. Siscovick et al. in another case-control study reported reduced risk of primary cardiac arrest (odds ratio = 0.51, 95% CI: 0.33–0.79) in influenza vaccinated patients vs. controls.10 In contrast, Jackson et al.22 in an observational study of first MI survivors, found no significant association between influenza vaccination and risk of recurrent coronary event.
Gurfinkel et al.13 in a randomized, non-placebo-controlled Fluvacs study found, in 6-month follow-up, 2% of cardiovascular deaths in vaccine vs. 8% in placebo group (relative risk 0.25; 95% CI:0.07–0.86; P = 0.01). In Fluvacs study, patients received conservative treatment: less than half of patients were administered statins, and most of MI patients had no reperfusion therapy. Of interest, in our results, the effect of influenza vaccination turned out to be more important in multivariable analysis (HR = 0.38, risk reduction of 62%) than in univariate analysis (HR = 0.54, risk reduction of 46%). There is no obvious explanation for this fact. However, one may speculate that the vaccine effect was attenuated by PCI. It is in accordance with Gurfinkel et al.13 results, who found in Fluvacs study that differences between vaccinated vs. control patients were statistically more significant in a conservative treatment, than in a PCI group.
While analysing our results, we did not find any early hazard in the intervention arm. During the first 20 days after vaccination, we had three events in controls and only one in the vaccine group (Figure 2B). Our observations, similar to the previous study on prophylactic flu vaccination early after emergency or planned PCI, show no increase in cardiovascular events after vaccination.3
There are several potential mechanisms which may explain the protective effect of influenza vaccination on the clinical course of CAD. The first is a reduction of acute flu complications. It may be considered influenza unspecific, but characteristic of consequences of any severe infection with fever, tachycardia, and dehydratation. The results of a large study carried out by Smeeth et al.3 showed an increased risk of MI after acute respiratory and also urinary tract infections, which supports this hypothesis.
There is also growing evidence suggesting special relationship between influenza, immune response, and CAD. Van Lenten et al.23,24 reported that influenza impaired anti-inflammatory properties of HDL-cholesterol and increased macrophages’ infiltration into the arterial wall.
Madjid and coworkers25,26 showed that influenza A infection results in smooth muscle cell proliferation, fibrin deposition, platelets aggregation, and thrombosis mimicking changes in coronary plaques after fatal MI. The authors postulated that influenza affects presumably already present plaques and in this way may trigger angina destabilization.26
Molecular mimicry and antigenic cross-reactivity between influenza viral antigens and plaque self-antigens were postulated as a cause of autoimmune reaction responsible for the progression of atherosclerosis and for plaque destabilization.26,27 Azambuya and Duncan28 hypothesized that 1918 influenza pandemia might have played a role in CAD epidemics of the 20th century by initiating immune response. In our study, the beneficial effect of influenza vaccination was not restricted to the flu season (Figure 2). A higher rate of events in unvaccinated patients was observed till the end of the follow-up, similarly to Gurfinkel et al.11 results. This interesting finding favours the hypothesis that influenza vaccination, not only prevents acute flu complications, but can also modify immune and inflammatory component of atherosclerosis resulting in the plaque stabilization.23–26 However, in our opinion in optimally treated patients, a similar plaque stabilizing effect may be obtained by PCI, high doses of statins, ACE-inhibitors, and antiplatelets. Interestingly, statins might play important role in prophylaxis and treatment of the next influenza pandemic due to their cardioprotective, anti-inflammatory, and immunomodulatory effect.29
Limitations of the study
The important limitation is a small number of cardiac events in our population, due to a small sample size. The study may be considered as underpowered to detect a statistical difference in mortality and MACE rates. On the basis of our results, we calculated post hoc needed sample size. For MACE under the assumption of our events rate, we estimated the sample size of 2240 patients to reach statistical significance. Another limitation is that the effect of flu vaccination on restenosis is unknown. An increased number of events in the placebo group at the end of the follow-up period may be theoretically related to a higher rate of restenosis, which remains one of the leading clinical issues in post angioplasty patients.11 As routine angiography was not planned in this study, we could not resolve this question clearly. Out of seven patients with coronary ischaemic events after 250 days of follow-up (Figure 2), four underwent angiography and two cases of restenosis were confirmed. Finally, the results of a single centre study may be influenced by patient selection biases. PCI patients were recruited after successful angioplasty, and NYHA class III/IV patients were excluded by protocol. Thus, PCI group represents only low risk patients and contributes to a lower event rate than originally expected.
In conclusion, our study did not prove that in CAD patients treated according to recent guidelines influenza vaccination significantly reduced either cardiovascular mortality or myocardial infarction, although it tends to reduce the incidence of MACE.
Influenza vaccination added to optimal medical and invasive treatment improves the clinical course of CAD, reducing the occurrence of coronary ischaemic events defined as combined frequency of MACE or hospitalization for myocardial ischaemia. The mechanism of the beneficial effect of influenza vaccination in CAD is not clear, but it extends over the period of viral circulation. A blockade of immune and inflammatory component of atherosclerosis may be as important as is the reduction of acute flu complications. A large scale, double-blind, multicentre study is still needed to evaluate life saving and MI-protecting properties of influenza vaccination in optimally treated coronary patients.
Conflict of interest: A.C.’s participation at several scientific congresses has been financially supported by Solvay Pharma. Other authors have declared no conflict of interests.
The study was financed by the Grant of Polish Ministry of Education and Science No. 2P05B 01627. Solvay Pharmaceuticals B.V. provided influenza vaccine and placebo vaccine free of charge.
. Seasonal variation in the occurrence of stroke in a Finnish adult population: The FINMONICA Stroke Register. Finnish Monitoring Trends and Determinants in Cardiovascular Disease. Stroke 1996;27:1774-1779.
. Influenza epidemics and acute respiratory disease activity are associated with a surge in autopsy-confirmed coronary heart disease death: results from 8 years autopsies in 34 892 subjects. Eur Heart J 2007;28:1205-1210.
. Influenza vaccine pilot study in acute coronary syndromes and planned percutaneous coronary interventions: the FLU Vaccination Acute Coronary Syndromes (FLUVACS) Study. Circulation 2002;105:2143-2147.
AHA/ACC guidelines for secondary prevention for patients with coronary and other atherosclerotic vascular disease: 2006 update: endorsed by the National Heart, Lung, and Blood Institute. J Am Coll Cardiol 2006;47:2130-2139.
Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (update 2005): The Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Eur Heart J 2005;26:1115-1140.
Guidelines on the management of stable angina pectoris: executive summary: the Task Force on the Management of Stable Angina Pectoris of the European Society of Cardiology. Eur Heart J 2006;27:1341-1381.
. Similarities in mortality patterns from influenza in the first half of the 20th century and the rise and fall of ischemic heart disease in the United States: a new hypothesis concerning the coronary heart disease epidemic. Cad Saude Publica 2002;18:557-577.