Vaccines for preventing influenza in healthy adults: a Cochrane review Clinical

Clinical question What are the effects (efficacy, effectiveness and harm) of vaccines against influenza in healthy adults? 2 Source: Jefferson T, Di Pietrantonj C, Rivetti A, Bawazeer GA, Al-Ansary LA, Ferroni E. Vaccines for preventing influenza in healthy adults. Cochrane Database of Systematic Reviews 2010, Issue 7. Art. No.: CD DOI: / CD pub4. Vaccines for preventing influenza in healthy adults

Context Influenza is seen as an important threat to health worldwide. Vaccines are the mainstay of its prevention. Each year, extensive vaccination campaigns cover an increasing number of age groups. These campaigns are costly in money and time. 3

Methods This is an update of a Cochrane review, which was first published in It was updated previously in 2004 and The specialised registers of the Cochrane Acute Respiratory Infections (ARI) Group and the Cochrane Vaccines Field were used to identify trials. The authors also searched the following bibliographic databases, most recently in June 2010: Cochrane Central Register of Controlled Trials, MEDLINE and EMBASE. A random effects model was used for the meta-analyses to calculate the risk ratio (RR). The vaccine efficacy (VE) was estimated using the formula VE =(1-RR)%. The risk difference (RD) and number needed to vaccinate (NNV) were also estimated from the meta-analyses. 4

PICO(S) to assess eligible studies Participants: Healthy individuals aged 16 to 65 years. Intervention: Live, attenuated or killed vaccines administered by any route. Comparison: Placebo or standard care. Primary outcomes: Symptomatic influenza and influenza-like illness (ILI) and systemic adverse effects. Secondary outcomes: Local adverse effects, time off work, hospital admissions and transmission of influenza. Studies: Randomized and quasi-randomized trials for efficacy or effectiveness, and comparative non-randomised studies for harms.

Description of eligible studies Fifty eligible reports were identified and included, an increase of 2 over the previous version of the review. 40 trials (with 59 sub-studies, and more than 70,000 people) reported data on effectiveness, efficacy and harms of influenza vaccines. 20 (25 sub studies) of the trials tested unlicensed aerosol vaccines, which had a very modest performance and were not included in the meta-analyses. 8 comparative non-randomised studies investigated serious harms. Two of these could not be included in the analysis. The main findings of the review are based on 26 clinical trials of the effectiveness of inactivated vaccines and 10 studies of the harms of the same vaccines.

Results: inactivated influenza vaccines The risk ratio (RR) for influenza symptoms for vaccines in which the WHO recommended content matched the circulating strain compared to no vaccine or placebo was 0.27 (95% CI 0.16 to 0.46). This gives a vaccine efficacy (VE) of 73% (95% CI 54% to 84%). The VE decreased to 44% (95% CI 23% to 59%) when matching of the vaccine was incomplete or not reported (RR 0.56, 95% CI 0.41, 0.77). The risk difference (RD) in the case of complete matching of the vaccine is –0.03 (95% CI –0.05 to –0.02). This corresponds to a number needed to vaccinate (NNV) of 33 (ie an average of 33 healthy adults need to be vaccinated for one fewer to develop influenza symptoms). With incomplete matching or if matching was not reported, the RD was –0.01 (95% CI –0.03 to 0.00), with a NNV of

8 Efficacy of influenza vaccine in healthy adults (vaccine content completely matching circulating strain) Based on 8 clinical trials, 11,285 participants 100 healthy adults, vaccinated 100 healthy adults, not vaccinated 1 develops influenza 4 develop influenza

9 Efficacy of influenza vaccine in healthy adults (vaccine content incompletely matching circulating strain or not reported) Based on 6 clinical trials with 10,331 participants 100 healthy adults, vaccinated 100 healthy adults, not vaccinated 1 develops influenza 2 develop influenza

Results: systemic adverse effects Inactivated vaccines caused myalgia (RR 1.54, 95% CI 1.12 to 2.11, based on research involving 2676 people). Several non-randomised studies assessed the association between inactivated influenza vaccines and a variety of rare potential harms, such as cutaneous melanoma, primary cardiac arrest, demyelinating diseases, pulmonary function, oculo-respiratory syndrome, Guillan Barrè Syndrome and Bell’s Palsy, and the last two of these appear to be related to inactivated vaccines. The occurrance of Guillan Barrè Syndrome was estimated as 1.6 to 10 extra cases per million vaccinated (based on surveys of more than 21 million people). Bell’s Palsy was associated with administration of an intranasal inactivated virosomal influenza vaccine (odds ratio 84.0, 95% CI 20.1 to 351.9) within 1 to 91 days after vaccination. The product was withdrawn. 10

Results: other outcomes Based on single studies, inactivated vaccines were not effective in preventing pneumonia (RR 0.59, 95% CI 0.04 to 9.43) or hospitalisation (RR 0.37, 95% CI 0.12 to 1.12). They had a minimal effect on working days lost, with a standardized mean difference of –0.21 (95% CI –0.36 to –0.05). No studies were identified which tested the effect on interruption of transmission. 11

Results: local harms Inactivated vaccines caused erythema (RR 4.01, 95% CI 1.91 to 8.41, based on 3388 observations) and tenderness or soreness (RR 3.11, 95% CI 2.08 to 4.66, based on 6833 observations). There was no statistically significant effect on arm stiffness (RR 1.62, 95% CI 0.54 to 4.83, based on 50 observations) nor induration (RR 2.24, 95% CI 0.48 to 10.59, based on 543 observations). 12

Conclusions Influenza vaccines have a modest effect in reducing influenza symptoms and working days lost among healthy adults. There is no evidence that they affect complications (such as pneumonia), or transmission. The dramatic difference between the apparent size of the effect with the risk ratio (RR) and the risk difference (RD) is because RR is a relative measure and RD is an absolute measure taking account of the baseline risk of influenza. This baseline risk is relatively low. During the winter months there will be 7 episodes of ILI among 100 adults, but fewer than one of these is caused by influenza. 13

14 How many influenza-like illnesses a year? How many of these are influenza? Based on 274 studies of influenza vaccines and 28 epidemiological studies from 1966 to 2007, with more than 3,000,000 participants 100 people Influenz a A or B 7 people with flu episodes (only 1 has influenza)

Useful links Cochrane Journal Club discussion points Vaccines for preventing influenza in healthy adults Vaccines for preventing influenza in healthy adults