Assignment: Evidence-Based Project, Part 4: Critical Appraisal of Research

Vaccine 35 (2017) 2775–2780

Contents lists available at ScienceDirect

Vaccine

journal homepage: www.elsevier .com/locate /vacc ine

Review

Efficacy and safety of high-dose influenza vaccine in elderly adults: A systematic review and meta-analysis

http://dx.doi.org/10.1016/j.vaccine.2017.03.092 0264-410X/� 2017 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Department of Community Health Sciences, S113 Medical Services Building - 750 Bannatyne Ave, University of Manitoba, Winnipeg 0W3, Canada.

E-mail address: wilkinsk@myumanitoba.ca (K. Wilkinson).

Krista Wilkinson a,b,⇑, Yichun Wei b, Andrea Szwajcer c, Rasheda Rabbani a,d, Ryan Zarychanski a,d,e,f, Ahmed M. Abou-Setta a,d, Salaheddin M. Mahmud a,d

aDepartment of Community Health Sciences, College of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada b Public Health Branch, Manitoba Health, Healthy Living and Seniors, Winnipeg, Manitoba, Canada cUniversity of Manitoba Libraries, Winnipeg, Manitoba, Canada dGeorge & Fay Yee Center for Healthcare Innovation, University of Manitoba/Winnipeg Regional Health Authority, Winnipeg, Manitoba, Canada eDepartment of Internal Medicine, Section of Critical Care, University of Manitoba, Winnipeg, Manitoba, Canada fDepartment of Haematology and Medical Oncology, CancerCare Manitoba, Winnipeg, Manitoba, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Received 19 August 2016 Received in revised form 8 March 2017 Accepted 31 March 2017 Available online 18 April 2017

Keywords: Influenza Influenza vaccines High-dose Randomized control trial Systematic review Meta-analysis

Introduction: Older adults are prioritized for influenza vaccination but also have lowered antibody responses to the vaccine. Higher-doses of influenza antigen may increase immune response and thus be more effective. Our objectives were to compare the efficacy and safety of the high-dose influenza vac- cine to the standard-dose influenza vaccine in the elderly (age > 65). Methods: Data sources: Randomized trials (RCTs) from Medline (Ovid), EMBASE (Ovid), Cochrane Library (Wiley), ClinicalTrials.gov, reference lists of relevant articles, and gray literature. Study selection: Two reviewers independently identified RCTs comparing high-dose influenza vaccine (60 lg of hemagglutinin per strain) to standard-dose influenza vaccine (15 lg of hemagglutinin per strain) in adults over the age of 65 years. Data extraction: Two reviewers independently extracted trial-level data including population character- istics, interventions, outcomes, and funding sources. Risk of bias was assessed using the Cochrane Risk of Bias tool. Results: We included seven eligible trials; all were categorized as having a low (n = 3) or unclear (n = 4) risk of bias. Patients receiving the high-dose vaccine had significantly less risk of developing laboratory- confirmed influenza infections (Relative Risk 0.76, 95%CI 0.65 to 0.90; I2 0%, 2 trials, 41,141 patients). Post-vaccination geometric mean titres and seroprotection rates were also higher in high-dose vaccine recipients. There were no protocol-defined serious adverse events in the included trials in either group. Conclusions: In elderly adults, the high-dose influenza vaccine was well-tolerated, more immunogenic, and more efficacious in preventing influenza infections than the standard-dose vaccine. Further prag- matic trials are needed to determine if the higher efficacy translates into higher vaccine effectiveness in adults over the age of 65.

� 2017 Elsevier Ltd. All rights reserved.

Contents

1. Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2776 2. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2776

2.1. Populations, interventions, comparators, outcome measures, settings, and study designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2776 2.2. Search strategy for identification of studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2776 2.3. Study selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2776 2.4. Data abstraction and management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2777

MB R3E

2776 K. Wilkinson et al. / Vaccine 35 (2017) 2775–2780

2.5. Assessment of potential risk of bias. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2777 2.6. Measures of treatment effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2777 2.7. Subgroup analyses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2777

3. Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2777

3.1. Primary outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2777 3.2. Subgroup analysis for primary outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2777 3.3. Secondary outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2777 3.4. Subgroup analyses for secondary outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2778

4. Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2779 5. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2780

Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2780 Appendix A. Supplementary material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2780 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2780

1. Background

Influenza has a high morbidity and mortality; it has been esti- mated that in Canada, there are an average of 4000 deaths and 12,200 hospital admissions associated with influenza on an annual basis[1,2]. The risk of a severe outcome associated with seasonal influenza infection increases with age and adults over the age of 65 years account for the majority of influenza-associated hospital- izations and deaths in Canada [3].

Vaccination against influenza is recommended annually as a key prevention strategy with older adults targeted as a high-risk population. There is evidence that seasonal influenza vaccine effec- tiveness (VE) is lower in adults over the age of 65 than in healthy adults 18–64 years old; recent meta-analysis estimated VE against influenza in older adults at about 49% (95% CI 33,62) while effec- tiveness was closer to 59% (95% CI 51, 67) for healthy younger adults [4,5]. This reduction in influenza VE may be partially explained by a reduction in immune response to influenza immu- nization as adults age [6].

Influenza vaccines must be updated and administered annually as the effectiveness of the seasonal vaccine depends on the match between the circulating virus strains and the antigens included in the vaccine. Effectiveness is also dependent on the immune response of the vaccine recipient and several methods have been proposed to improve the efficacy of the conventional influenza vac- cines; adding adjuvants, administering vaccine through routes other than the intramuscular standard, or using live-attenuated influenza vaccine instead of the inactivated form [7].

Another strategy to enhance antibody response in the elderly has been the use of high-dose antigen influenza vaccines. These vaccines deliver higher doses of influenza virus antigen than the standard-dose vaccine (typically 60 lg of hemagglutinin per strain compared to 15 lg in standard dose vaccines) to induce a stronger immune response. The increase in antibody in serum is expected to be correlated with an increase in vaccine effectiveness [8].

There have been no previous systematic reviews conducted comparing the vaccine effectiveness of the high-dose influenza vaccine to the standard-dose vaccine in older adults.

The purpose of this systematic review was to identify, critically appraise, and meta-analyze data from prospective randomized controlled trials comparing high-dose trivalent inactivated influ- enza vaccine to standard-dose trivalent influenza vaccine in adults over the age of 65.

2. Methods

We conducted our systematic review using methodological approaches outlined in the Cochrane Handbook for Systematic

Reviewers [9] and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) criteria [10]. The study protocol was registered in PROSPERO – Interna- tional prospective register of systematic reviews (CRD42016039387).

2.1. Populations, interventions, comparators, outcome measures, settings, and study designs

We included only randomized, controlled trials of adults over the age of 65 years old. The primary research question was, ‘‘In elderly adults (over the age of 65) is the high-dose influenza vac- cine, compared to the standard-dose vaccine, associated with pre- vention of laboratory-confirmed influenza infections, influenza- associated hospitalizations, influenza-associated deaths and seri- ous adverse events?”

The main outcome measure was laboratory-confirmed influ- enza infection. Secondary outcomes were influenza-associated hospitalizations and deaths, and immune response (immunogenic- ity and seroprotection). Serious adverse events following immu- nization were included as safety outcomes.

2.2. Search strategy for identification of studies

We searched Medline (Ovid), EMBASE (Ovid), and Cochrane Library (Wiley) from inception to present using individualized search strategies prepared for each database with the Cochrane Highly Sensitive Search Strategy as a model [9]. The search strategy for Medline and EMBASE is presented in Appendix 4. We searched the World Health Organization’s International Clinical Trials Regis- try Platform and ClinicalTrials.gov and hand-searched relevant conference proceedings for the preceding 5 years to identify planned, ongoing, or recently completed but unpublished trials of high-dose influenza vaccine. The reference lists of included trials were hand-searched for relevant citations. No language, publica- tion date, or publication status restrictions were imposed. We per- formed reference management in EndNote (version X7.2.1, Thomson Reuters).

2.3. Study selection

We used a 2-stage process for study screening and selection using standardized and piloted screening forms. Two reviewers independently screened the titles and abstracts of search results to determine whether a citation met the inclusion criteria (Appen- dix 1). The full text of citations classified as include or unclear were reviewed independently with reference to the predetermined inclusion and exclusion criteria. Discrepancies between the two

K. Wilkinson et al. / Vaccine 35 (2017) 2775–2780 2777

reviewers were resolved through consensus by discussion with a third reviewer, as required.

2.4. Data abstraction and management

Two reviewers independently extracted data, using standard- ized and piloted data extraction forms, from included trial reports. Discrepancies between the two reviewers were resolved through consensus in discussion with a third reviewer, as required. Extracted data included demographics of the enrolled patient pop- ulation, interventions, study outcomes, and funding sources. Data management was performed using Microsoft Excel 2010 (Excel version 14, Microsoft Corp).

2.5. Assessment of potential risk of bias

We assessed internal validity using the Cochrane Collabora- tion’s Risk of Bias Tool [11]. This tool consists of seven domains and a categorization of the overall risk of bias. We also assessed the source(s) of funding for each study and its potential influence on the outcome measures. We used information from the risk of bias assessment to guide sensitivity analyses and explore sources of heterogeneity.

2.6. Measures of treatment effect

We analyzed data from the included studies using Review Man- ager (RevMan, version 5.3, the Nordic Cochrane Center, the Cochrane Collaboration). A formal meta-analysis was conducted to evaluate if the data were statistically and clinically homoge- neous. We expressed pooled continuous effect measures as mean differences with 95% CIs. Pooled dichotomous data were expressed as risk ratios (RR) with 95% CIs. We used the random effects model for all analyses. Each vaccine year was included as a separate study in the analysis for trials contributing more than one year of data as vaccine composition changes annually.

Statistical heterogeneity of the data was explored and quanti- fied, using the I2 test which describes the percentage of variation in study outcomes due to heterogeneity and not chance [12]. If sig- nificant heterogeneity was detected (I2 > 50%), subgroup analysis was conducted. All tests of statistical inference reflect a 2-sided a of 0.05.

2.7. Subgroup analyses

Subgroup analyses were determined a priori and included clin- ical (patient population) and methodological (risk of bias, source of study funding) considerations. Subgroup analyses were performed to determine summary effect estimates of high-dose influenza vac- cine in adults over the age of 65 years relative to standard-dose influenza vaccine. Analyses were dependent on the number of

Table 1 Characteristics of included trials.

Trial Vaccine year(s) Population

Keitel et al. (2006) [17] 2001–2002 Ambulatory, medically stable Couch et al. (2007) [8] 2004–2005 Ambulatory, medically stable Falsey et al. (2009) [18] 2006–2007 Ambulatory, medically stable DiazGranados et al. (2013) [14] 2009–2010 Ambulatory, medically stable DiazGranados et al. (2014) [19] 2011–2013 Ambulatory, medically stable Nace et al. (2014) [13] 2011–2013 Frail residents of LTCF Tsang et al. (2014) [20] 2007–2008 Ambulatory, medically stable Cate et al. (2010) [16] 2004-2005a Ambulatory, medically stable

LTCF = Long term care facility, SD = standard deviation. a Companion paper to Couch et al. [8]

studies included and the availability of appropriate outcomes and covariates.

3. Results

Of the 2955 citations identified from electronic and hand- searches, we included 7 unique trials with 101 to 31,803 patients (median 634; interquartile range [IQR], 300 to 6429) and one com- panion paper (Table 1, Fig. 1). Trials were published between 2006 and 2014, and all were published in peer-reviewed journals. Most trials were from North America and the populations were ambula- tory, medically-stable adults in all but the trial by Nace et al. in which the population was frail residents of long-term care facilities [13]. The mean age of study patients ranged from 72 to 87 years. Four trials were adjudicated to be of unclear risk of bias and three trials were considered to have low risk of bias (Fig. 2).

3.1. Primary outcomes

Two trials, with 41,141 patients in the full-analysis sets, reported cases of laboratory-confirmed influenza [14,15]; the pooled RR for confirmed cases of influenza in the high-dose vaccine group compared to the standard-dose vaccine group was 0.76 (95% CI, 0.65–0.90; I2, 0%) (Fig. 3), corresponding to 24% greater vaccine efficacy in the high-dose group as compared to the standard-dose group.

One trial [15] contributed data for influenza seasons in which the vaccine components were classified as well-matched or not well-matched to circulating influenza virus strains by genetic sequencing. For well-matched seasons, the RR of laboratory- confirmed influenza for the high-dose influenza vaccine recipients compared to the standard-dose vaccine recipients was 0.65 (95%CI, 0.48–0.87). In influenza seasons when the vaccine components were not well-matched to circulating influenza virus strains, the pooled RR of laboratory-confirmed influenza for the high-dose influenza vaccine recipients compared to standard-dose vaccine recipients was 0.83 (95%CI, 0.67 to 1.02; I2, 0%).

3.2. Subgroup analysis for primary outcome

Subgroup analyses according to a priori defined clinical consid- erations (patient population) or methodological considerations (risk of bias, study sponsor) were not done as both trials reporting primary outcomes [14,15] were conducted by the same authorship group and I2 was 0%.

3.3. Secondary outcomes

Influenza-associated hospitalizations and deaths – none of the included studies reported on these secondary outcomes.

Overall age, Mean (max; min or SD), y High-dose (n) Standard-dose (n)

72 (65; 88) 50 51 74 (65; 95) 206 208 73 (65; 97) 2575 1262 73 (64; 100) 6013 3008 73 (5.8) 15,892 15,911 87 (6) 89 98 73 (6) 317 317 74 (65; 95) 206 208

Records iden�fied through database searching and other

sources (n=2955)

Unique records screened for eligibility (n=2201)

Full-text ar�cles assessed for eligibility (n=39)

Trials included in review (n=8) Primary studies (n=7)

Companion studies (n=1)

Duplicate records excluded (n=754)

Screened records excluded (n=2162)

Full text ar�cles excluded (n=32): Inappropriate study design

(n=12) Inappropriate interven�ons

(n=8) Inappropriate study popula�on

(n=10) Inappropriate outcomes (n=2)

Fig. 1. Study flow diagram of included trials.

2778 K. Wilkinson et al. / Vaccine 35 (2017) 2775–2780

Immunogenicity – For the subset of patients (n = 18,215) that were assessed for immunogenicity, the high-dose vaccine was associated with higher hemagglutination inhibition (HAI) assay geometric mean titres (GMT); the pooled mean difference in GMT for the H1 component of the vaccine was 86.2 (95% CI, 47.6 to 124.8; I2, 98%) (Table 2). The pooled mean difference in GMT for the 18,323 patients assessed for the H3 component of the vaccine was 183.2 (95% CI, 84.6 to 285.7; I2 99%) and the pooled mean difference in GMT for the 18, 214 patients assessed for the B component of the vaccine was 24.7 (95% CI, 14.1 to 35.3; I2, 95%).

The companion paper by Cate et al. was the only trial that assessed serum anti-neuraminidase (NA) titres [16] . For the 414 individuals included, the high-dose vaccine resulted in greater GMT in the high-dose group for both the N1 and N2 antigens.

Seroprotection – For the H1 component of the vaccine, a subset of 17,996 patients was assessed for seroprotection (the proportion of participants with an HAI titre � 1:40 or HAI titre � 1:32); the pooled RR for the high-dose vaccine recipients was 1.09 (95%CI, 1.05 to 1.14; I2, 92%) (Table 2). For the 17, 997 patients assessed for the H3 component of the vaccine, the pooled RR for the high- dose recipients was 1.03 (95% CI, 1.02 to 1.04; I2, 61%) and the pooled RR for the 17,995 patients assessed for the B component of the vaccine was 1.14 (95% CI, 1.09 to 1.19; I2 81%).

Safety – None of the included trials reported any cases of vaccine-associated mortality, Guillain-Barré Syndrome, or anaphy- laxis in either the high-dose or the standard-dose vaccine groups. One study reported a case of Bell’s Palsy in the standard-dose group that was considered by the investigator to be related to the vaccination [14].

3.4. Subgroup analyses for secondary outcomes

We detected significant statistical heterogeneity in the compar- isons for both immunogenicity and seroprotection. Subgroup anal- ysis by funding source (industry versus other) for immunogenicity indicated that industry-sponsored trials showed larger mean dif- ferences in GMT between the high-dose and standard-dose vaccine recipients than the differences observed in non-industry sponsored studies; however the direction of the effect in both subgroups significantly favoured the high-dose vaccine recipients across all vaccine components (Supplementary data; Appendices 5–7). Sub- group analysis by funding source for the outcome measure of sero- protection showed that the subgroups were statistically different with respect to the H1 component; the trials not sponsored by industry showed no difference in seroprotection between the high-dose and standard-dose groups for both the H3 and B vaccine components (Supplementary data; Appendices 8–10).

Fig. 2. Risk of bias summary for included trials. Judgement about each risk item for each included study. Circleswith embedded plus sign reflect a judgement of low risk of bias. Circles with embedded question mark reflect a judgement of unclear risk of bias.

K. Wilkinson et al. / Vaccine 35 (2017) 2775–2780 2779

Subgroup analysis by patient population (ambulatory and med- ically stable versus frail LTCF residents) shows that the subgroups were statistically different with respect to mean GMT difference across all three vaccine components; for the H1 component, the frail population showed no difference in GMT between the high- dose and the standard-dose influenza vaccine whereas a significant difference was observed in the ambulatory, medically stable sub- group. Subgroup analysis by patient population showed a signifi- cant difference in seroprotection between the medically stable and frail populations for both the H3 and B vaccine components; although both subgroups favoured the high-dose vaccine.

Fig. 3. Laboratory-confirmed influenza infection in patients randomized to high-dose in represent point estimates, varying in size according to the weight in the analysis and 9

Subgroup analysis for both immunogenicity and seroprotection by risk of bias (low versus unclear) demonstrated no major differences.

4. Discussion

In the two trials that assessed the primary outcome of laboratory-confirmed illness, the high-dose influenza vaccine was found to be associated with a 24% reduction in the risk of laboratory-confirmed influenza compared to the standard-dose influenza vaccine [14,15]. Immunogenicity and seroprotection were also greater in the high-dose influenza vaccine group com- pared to the standard-dose group. There was one protocol- defined safety outcome (Bell’s Palsy) observed in the standard- dose vaccine group [14].

This systematic review highlights evidence gaps that will need to be addressed before the benefit of the high-dose vaccine can be established. The two studies by DiazGranados et al. that pro- vided estimates of vaccine efficacy were done in ambulatory, med- ically stable populations with a mean age of 73 [14,15]. There is no evidence as to the efficacy of the high-dose influenza vaccine in older, immune-suppressed adults.

Although influenza-associated hospitalizations and deaths are important clinical outcomes, they were not included in any of the trials in this systematic review. Given that there are an esti- mated 12,200 influenza-related hospital admissions annually in Canada [1], this may be a more relevant outcome to decision- makers than measures of immunogenicity.

Strengths of this systematic review include the comprehensive search methodology including searching multiple citation data- bases and trial registries and hand searching of the gray literature. We also used an a priori protocol and followed establishedmethod- ological guidelines in the conduct and reporting of this review.

Our systematic review has limitations. Although attempts were made to limit clinical and methodological heterogeneity during study design, there was a high degree of statistical heterogeneity observed in the estimates for the secondary outcomes for both immunogenicity and seroprotection. This heterogeneity was only partially explained by subgroup analysis by patient population (ambulatory and medically stable versus frail residents of LTCF).

An important feature of influenza is its year-to-year variability in severity due to genetic changes in the virus or changes in circu- lating strains. The annual seasonal differences in influenza vaccine might be contributing to the heterogeneity seen in the pooled esti- mates; however, subgroup analysis by year was not possible since there were only two studies with overlapping vaccine seasons [13,15]. In one study that spanned two influenza seasons, the authors reported that the high-dose influenza vaccine produced superior responses for all strains except for the H1 component in the 2012–2013 influenza season, providing further evidence that differences in strains included in the annual vaccine may be an important factor in immunogenicity [13].

fluenza vaccine versus standard-dose influenza vaccine. Boxes and horizontal lines 5% CIs. M-H = Mantel Haenszel.

Table 2 Secondary outcome measures.a

Outcome Trialsb Participants (n) Effect estimate I2

HD SD (95% CI)

Immunogenicityc

H1 component 8 10,253 7962 MD:86.2 (47.6, 124.8) 98% H3 component 8 10,254 7962 MD:162.9 (86.1, 239.8) 99% B component 8 10,252 7962 MD:24.7 (14.1, 35.3) 96%

Seroprotectiond

H1 component 8 10,142 7854 RR:1.09 (1.05, 1.14) 92% H3 component 8 10,143 7854 RR:1.03 (1.02, 1.04) 61% B component 8 10,141 7854 RR:1.14 (1.09, 1.19) 81%

HD = high-dose influenza vaccine, SD = standard-dose influenza vaccine, RR = risk ratio, MD = mean difference. a Influenza-associated hospitalizations and deaths were not reported. b Trials spanning multiple vaccine seasons are treated as separate studies by year. c Refs. [13–15,17–19]. d Refs. [8,14,15,17,20].

2780 K. Wilkinson et al. / Vaccine 35 (2017) 2775–2780

5. Conclusions

There is limited evidence that the high-dose trivalent, inacti- vated influenza vaccine in ambulatory, medically stable patients over the age of 65 is associated with decreased rates of laboratory-confirmed influenza infection compared with the standard-dose vaccine. The overall impact remains uncertain due to the dearth of RCTs demonstrating efficacy against influenza infection and lack of evidence for clinically-relevant outcomes. Longer-term pragmatic trials will be needed to show if this superi- ority lasts in real-world settings.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.vaccine.2017.03. 092.

References

[1] Schanzer DL, McGeer A, Morris K. Statistical estimates of respiratory admissions attributable to seasonal and pandemic influenza for Canada. Influenza Other Respir Viruses 2013;7(5):799–808.

[2] Schanzer DL, Sevenhuysen C, Winchester B, Mersereau T, Nishiura H. Estimating influenza deaths in Canada, 1992–2009. PLoS ONE 2013;8(11).

[3] Mitchell R, Taylor G, McGeer A, et al. Understanding the burden of influenza infection among adults in Canadian hospitals: a comparison of the 2009–2010 pandemic season with the prepandemic and postpandemic seasons. Am J Infect Control 2013;41(11):1032–7.

[4] Beyer WE, Nauta JJ, Palache AM, Giezeman KM, Osterhaus AD. Immunogenicity and safety of inactivated influenza vaccines in primed populations: a systematic literature review and meta-analysis. Vaccine 2011;29 (34):5785–92.

[5] Osterholm MT, Kelley NS, Sommer A, Belongia EA. Efficacy and effectiveness of influenza vaccines: a systematic review and meta-analysis. Lancet Infect Dis 2012;12(1):36–44.

[6] Goodwin K, Viboud C, Simonsen L. Antibody response to influenza vaccination in the elderly: a quantitative review. Vaccine 2006:1159–69.

[7] Parodi V, de Florentiis D, Martini M, Ansaldi F. Inactivated influenza vaccines: recent progress and implications for the elderly. Drugs Aging 2011;28 (2):93–106.

[8] Couch R, Winokur P, Brady R, et al. Safety and immunogenicity of a high dosage trivalent influenza vaccine among elderly subjects. Vaccine 2007;25 (44):7656–63.

[9] Higgins JPT, Green, S., eds. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011] <http://handbook.cochrane. org>; 2008 [accessed February 2016].

[10] Moher DL, Liberati A, Telzlaff J, Altman DG. Group P. Preferred reporting items for systematic review and meta-analyses: the PRISMA statement. Ann Intern Med 2009;151(4):264–9.

[11] Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011;343:d5928.

[12] Higgins JPT, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21(11):1539–58.

[13] Nace D, Lin C, Ross T, Saracco S, Churilla R, Zimmerman R. Randomized, controlled trial of high-dose influenza vaccine among frail residents of long- term care facilities. J Infect Dis 2015;211(12):1915–24.

[14] DiazGranados C, Dunning A, Jordanov E, Landolfi V, Denis M, Talbot H. High- dose trivalent influenza vaccine compared to standard dose vaccine in elderly adults: safety, immunogenicity and relative efficacy during the 2009–2010 season. Vaccine 2013;31(6):861–6.

[15] DiazGranados C, Dunning A, Kimmel M, et al. Efficacy of high-dose versus standard-dose influenza vaccine in older adults. N Engl J Med 2014;371 (7):635–45.

[16] Cate TR, Rayford Y, Niño D, et al. A high dosage influenza vaccine induced significantly more neuraminidase antibody than standard vaccine among elderly subjects. Vaccine 2010;28(9):2076–9.

[17] Keitel W, Atmar R, Cate T, et al. Safety of high doses of influenza vaccine and effect on antibody responses in elderly persons. Arch Intern Med 2006;166 (10):1121–7.

[18] Falsey A, Treanor J, Tornieporth N, Capellan J, Gorse G. Randomized, double- blind controlled phase 3 trial comparing the immunogenicity of high-dose and standard-dose influenza vaccine in adults 65 years of age and older. J Infect Dis 2009;200(2):172–80.

[19] DiazGranados CA, Dunning AJ, Kimmel M, et al. Efficacy of high-dose versus standard-dose influenza vaccine in older adults. N Engl J Med 2014;371 (7):635–45.

[20] Tsang P, Gorse G, Strout C, et al. Immunogenicity and safety of Fluzone(�) intradermal and high-dose influenza vaccines in older adults ?65 years of age: a randomized, controlled, phase II trial. Vaccine 2014;32(21):2507–17.