key: cord-011718-hcyluzkx
authors: Gaglani, Manjusha; Spencer, Sarah; Ball, Sarah; Song, Juhee; Naleway, Allison; Henkle, Emily; Bozeman, Sam; Reynolds, Sue; Sessions, Wendy; Hancock, Kathy; Thompson, Mark
title: Antibody Response to Influenza A(H1N1)pdm09 Among Healthcare Personnel Receiving Trivalent Inactivated Vaccine: Effect of Prior Monovalent Inactivated Vaccine
date: 2014-06-01
journal: J Infect Dis
DOI: 10.1093/infdis/jit825
sha: 
doc_id: 11718
cord_uid: hcyluzkx

Background. Few data are available on the immunogenicity of repeated annual doses of influenza A(H1N1)pdm09-containing vaccines. Methods. We enrolled healthcare personnel (HCP) in direct patient care during the autumn of 2010 at 2 centers with voluntary immunization. We verified the receipt of A(H1N1)pdm09-containing monovalent inactivated influenza vaccine (MIIV) and 2010–2011 trivalent inactivated vaccine (TIV). We performed hemagglutination inhibition antibody (HI) assays on preseason, post-TIV, and end-of-season serum samples. We compared the proportion of HCPs with HI titer ≥40 against A(H1N1)pdm09 per receipt of prior-season MIIV, current-season TIV, both, or neither. Results. At preseason (n = 1417), HI ≥ 40 was significantly higher among those who received MIIV (34%) vs those who did not (14%) (adjusted relative risk [ARR], 3.26; 95% confidence interval [CI], 2.72–3.81). At post-TIV (n = 865), HI ≥ 40 was lower among HCP who received MIIV and TIV (66%) than among those receiving only TIV (85%) (ARR, 0.93 [95% CI, .84–.997]). At end-of-season (n = 1254), HI ≥ 40 was 40% among those who received both MIIV and TIV and 67% among those receiving only TIV (ARR, 0.76 [95% CI, .65–.88]), 52% among those who received MIIV only, and 12% among those receiving neither. Conclusions. HCP immunization programs should consider effects of host immune response and vaccine antigenic distance on immunogenicity of repeated annual doses of influenza vaccines.

performed a prospective cohort study during 2010-2011 at 2 medical centers offering voluntary HCP immunization to examine factors that influence immunogenicity against laboratory-confirmed influenza. The administration of monovalent vaccines in 2009-2010 presented an opportunity to examine the HI response against the 2009 influenza A(H1N1) pandemic virus [A(H1N1)pdm09] among HCP who may have received vaccine containing the same antigen. Our hypothesis was that receipt of 2010-2011 seasonal A(H1N1)pdm09-containing vaccines should produce an HI response similar to that reported in clinical trials [16] [17] [18] [19] [20] , regardless of receipt of prior-season vaccine.

A prospective cohort was enrolled after the 2009-2010 influenza A(H1N1) pandemic season in autumn 2010 at Scott & White Healthcare (SWH) in Temple/Round Rock, Texas, and Kaiser Permanente Northwest (KPNW) in metropolitan Portland, Oregon. Eligible HCP were aged 18-65 years, working ≥32 hours per week, receiving care from site for >12 months, and providing direct patient care. Details on cohort recruitment are presented in Thompson et al [21] . Announcements regarding a "respiratory illness in healthcare workers" study were sent to all employees before and during voluntary immunizations. Study protocol was approved by the sites' institutional review boards.

Participants completed an Internet-based questionnaire at enrollment/preseason that included demographics, health, occupation, and work setting. We assessed self-rated health status with a 5-level rating of overall health [22, 23] , and calculated body mass index. We extracted data from electronic health records to characterize participants with high-risk conditions during the prior year [24] . We confirmed 2010-2011 and prior-season influenza immunization from medical/employee health records. Blood was drawn at 3 time points: at enrollment from all participants ( preseason serum), approximately 30 days after receipt of 2010-2011 vaccine, ( post-trivalent inactivated vaccine [TIV] serum), and approximately 7 months after enrollment from all participants (end-of-season serum).

From 18 December 2010 to 30 April 2011, participants completed weekly Internet or computer-assisted telephone surveys assessing for acute respiratory illness (ARI; cough and fever/ feverishness/chills, onset ≤7 days). Internet reports were common (78%). Noncompliant participants received email/telephone reminders to complete surveys. Health records were monitored daily for ARI diagnoses (International Classification of Diseases, Ninth Revision, Clinical Modification codes 460-466, 480-488). Identified ARIs prompted a visit at home or the clinic where nasal, nasopharyngeal, and oropharyngeal swabs were collected, then tested for influenza A and B using real-time reverse transcription polymerase chain reaction (rRT-PCR) assay, with primers, probes, and reagents from the Centers for Disease Control and Prevention. A positive result from any swab was accepted.

Each site offered both TIV and live attenuated influenza vaccine (LAIV) during both seasons. The monovalent A(H1N1)pdm09 vaccines contained A/California/7/2009 (H1N1)-like virus, which was the predominant virus in 2009-2010 at both sites [25] . The 2010-2011 seasonal trivalent vaccines also contained A/California/7/2009 (H1N1)-like virus. The predominant circulating viruses at the sites in 2010-2011 were similar to that in the United States: 74% of all positive results were influenza A, A(H3N2) 62% of all subtyped, and the rest A(H1N1)pdm09 [26] .

At KPNW, a single lot (A) of TIV from manufacturer W was administered to 88% of vaccinated HCPs for whom all 3 serum samples were taken; remaining lots were from W and X. At SWH, 2 lots (B and C) from manufacturer Y accounted for 91%; remaining lots were from manufacturer Z. In 2009-2010, both sites had administered many lots of monovalent inactivated influenza vaccine (MIIV) from W, X, Z, or unknown manufacturers. Of 489 HCP who received 2009-2010 MIIV and 2010-2011 TIV, 146 of 242 (60%) from KPNW received a single lot of MIIV and 193 of 247 (78%) from SWH received 2 lots from the same manufacturer Z.

Serum HI assays for each HCP were run in duplicate simultaneously using standard technique at the Battelle laboratory (Aberdeen, Maryland), after completing a Centers for Disease Control and Prevention (CDC) proficiency panel [27, 28] . A standard turkey red blood cell (RBC) suspension was prepared, and serum samples were treated with receptor-destroying enzyme to remove nonspecific inhibitors. Nonspecific agglutinins were removed by serum adsorption with packed RBCs. Serum was diluted 2-fold starting from 1:10. The HI titer was the reciprocal of the serum dilution in the last well with complete hemagglutination inhibition. The geometric mean titer (GMT) from duplicate results was reported; HI < 10 was considered to be 5 for GMT calculation.

We performed the primary analysis at 3 time points. For time point 1, we included 1417 HCP who had a serum specimen at preseason. Because the predominant vaccine used at each site during both seasons was inactivated and we were examining the effect of second annual homologous revaccination on A(H1N1)pdm09 HI titers, we excluded monovalent LAIV recipients. For time point 2, we included HCP who had a specimen at preseason, received 2010-2011 TIV, and had a specimen approximately 30 days (from 14 to 63 days only) post-TIV. Some members of this group had received A(H1N1)pdm09 MIIV and others had not. For time point 3, we included HCP who had serum samples taken at both preseason and end-ofseason. This included some who received only 2009-2010 MIIV or 2010-2011 TIV, some who received both, and some who received neither. For the end-of-season analysis, we excluded H1N1pdm09-infected HCP during 2010-2011 evidenced by a positive rRT-PCR test or seroconversion (between preseason and end-of-season serum for unvaccinated, and between post-TIV and end-of-season serum for TIV recipients). Seroconversion was defined as preseason HI < 10 and post-TIV/end-of-season HI ≥ 40; or preseason HI ≥ 10 and a minimum 4-fold rise for post-TIV/end-of-season serum (US Food and Drug Administration definition).

For the 3 primary analyses, we examined HCP characteristics: demographics, health status, vaccination history, timing of serum sampling, site, proxies for influenza exposure, and HI titer preseason. Means were compared utilizing the Wilcoxon rank-sum or 2-sample t test, and proportions by χ 2 or Fisher exact test. A P value of <.05 indicated statistically significant difference. We constructed logistic regression models for each analysis with outcome variable A(H1N1)pdm09 HI ≥ 40, adjusting for HCP characteristics. As preseason HI titers correlate with post-TIV titers, controlling for baseline/preseason HI titer is recommended for single-season studies [29] . However, because we examined effects over 2 seasons, we did not adjust for preseason titers, because of a lack of true baseline before receipt of all vaccines (MIIV receipt was prior to 2010-2011 preseason titer) and doing so would have removed prior H1N1 infection and/or MIIV effect.

An HI ≥ 40 is a generally accepted laboratory correlate for at least 50% protection against influenza infection [30] [31] [32] . We thus compared the proportion of HCP with HI ≥ 40 for preseason, post-TIV, and end-of-season serum samples by vaccination history. We also assessed HI response for post-TIV and end-ofseason serum samples by preseason HI < 40 or HI ≥ 40; by HI < 10, HI 10 to <40, and HI ≥ 40; and by vaccination history.

Because the outcome variable HI ≥ 40 was common, the odds ratio (OR) is not a good approximation to the relative risk (RR), so we converted adjusted ORs to adjusted relative risks (ARRs) using the following equation, where P 0 = the incidence of disease in the nonexposed group: RR = OR/([1 -P 0 ] + [OR* P 0 ]) [33] . This conversion is recommended for cohort designs such as ours; however, we also validated the ARR conversion using Poisson regression with robust covariance, adjusting for covariates.

We conducted 4 sets of secondary analyses to confirm our findings. First, we ran logistic regression models for each site for the 3 primary analyses. Second, we compared HCP characteristics and HI titers for those who received the 3 most common lots of vaccine at our 2 sites. Third, we compared descriptive characteristics for TIV-vaccinated HCP by receipt or not of prior MIIV. Fourth, as a sensitivity analysis for the subgroup of vaccinated HCP with all 3 serum samples, we applied a linear mixed effects model. For this analysis, HI titers were first rounded down to 2-fold reciprocal serum dilutions. Only a small number of GMTs that fell in between levels were rounded down (eg, GMT of 7 or 28 was rounded down to 5 or 20, respectively). For assessment as a continuous variable, HI titers were then converted to ordinal HI levels from 0 to 9 for each 2fold increase from 5 to 2560 by log transformation: log 2 (HI titer/5) [32, 34] . The linear mixed-effects model assessed ordinal HI level as primary outcome, controlling for time and other covariates. Post-TIV days were represented with linear and quadratic time terms. We considered different errorcovariance structures to account for within-subject correlation over time and selected a first-order autoregressive with randomintercept model because it had the smallest Bayesian information criterion. We evaluated the effect of prior receipt of MIIV on ordinal HI levels for the 3 serum samples. SAS version 9.2 (SAS Institute) was used for analyses.

Supplementary Figure 1 shows that 1417 HCP in direct patient care were enrolled and gave preseason sera. Of these, 936 (66%) received TIV and 865 (59%) had a blood specimen approximately 30 days post-TIV; the specimen was drawn between 14 and 42 days for 822 (95%) and between 43 and 63 days for 43 (5%). When compared with HCP included preseason (Table 1) , those post-TIV were more likely to be aged 50-65 years, have high-risk conditions, and receive 2009-2010 seasonal vaccines, and less likely to work in the emergency department. Of 1417 HCP from preseason, 1254 (89%) had end-of-season serum samples assessed for A(H1N1)pdm09 HI titers. The proportion of HCP who had prior-season MIIV was similar among those providing preseason (42%) and end-of-season (43%) serum samples, but higher among those providing post-TIV sera (57%) ( Table 1 ). There were modest but statistically significant differences between the 2 study sites in HCP characteristics at enrollment (Supplementary Table 1 ).

In the time 1/preseason analysis, an HI ≥ 40 for A(H1N1) pdm09 was found in 34% of those receiving prior-season MIIV vs 14% for those without (ARR = 3. 26 . For post-TIV-serum samples, HI ≥ 40 was associated with female sex and SWH site and inversely associated with age and direct patient care hours (data not shown).

In the time 3/end-of-season analysis, we found a statistically significant interaction between receipt of MIIV and TIV and therefore stratified the analysis into 4 groups ( Table 2) . Among all 1254 HCPs without lab-confirmed A(H1N1)pdm09 infection in 2010-2011, HI ≥ 40 was maintained in 40% among those who received both MIIV and TIV, 67% among those . Among those with preseason HI < 40, end-of-season HI ≥ 40 was much less common in those who received both TIV and MIIV (15%) vs those receiving only TIV (63%) ( Table 2) .

Among the 834 vaccinated HCP with all 3 sera, the mean foldchange in HI titers was highest among those with preseason HI < 10 (Table 3) . Post-TIV serum HI ≥ 40 was acquired in more than half of HCP with preseason HI < 10 (61%) and HI 10 to <40 (79%). However, at end-of-season, these proportions dropped to 35% and 43%, respectively. In contrast, 96% of those with preseason HI ≥ 40 remained so through end-ofseason (Table 3) . Among these 834 vaccinated HCP, history of receipt of MIIV was associated with lower HI GMTs for post-TIV and end-of-season regardless of preseason HI GMTs ( Figure 1 ). This finding was seen at both sites but was more pronounced at SWH (Supplementary Figure 2) . Table 4 presents the descriptive characteristics of 865 HCP who received TIV by receipt or not of prior MIIV, including high-risk conditions and PCR-confirmed H1N1 infections in 2010-2011. The preseason HI GMTs of those with no prior MIIV were significantly lower than those receiving 2009-2010 MIIV. When we controlled for preseason HI titers in the logistic regression models, even though HI ≥ 40 post-TIV and endof-season was associated with preseason baseline titers, the effect of receipt of 2009-2010 MIIV on HI titers noted in Table 2 increased (data not shown).

The linear mixed-effects model included 2502 HI titers, 3 for each of 834 vaccinated HCP (who provided all 3 serum samples). The model confirmed results from our primary analyses (data not shown). After controlling for covariates, those receiving MIIV had higher preseason ordinal HI level (GMT ratio estimate: 2.31) than those who had not (Table 5 ). However, they also had lower ordinal HI level (GMT ratio estimates: 0.57 and 0.74 for post-TIV and end-of-season, respectively), than those who had not. There was no significant difference in preseason HI titers for HCP receiving 3 major TIV lots (at KPNW: lot A, n = 238; at SWH: lot B, n = 291 and lot C, n = 221 When HI titers were measured at end-of-season, compared to those who had never received any A(H1N1)pdm09-containing vaccine, HCP who were vaccinated both seasons had significantly higher probability for maintaining HI ≥ 40. However, those vaccinated with TIV in 2010-2011 but not MIIV in 2009-2010 ended the season with significantly higher odds of HI ≥ 40 than HCP vaccinated both seasons. Thus, HCP who were naive to the A(H1N1)pdm09 vaccine antigen in 2010-2011 had the best HI response post-TIV and at end-of-season.

We noted similar trends for seroconversion post-TIV, and these effects were expected and noted at both study sites (data not shown). Periodically, there have been reports of reduced immunogenicity and effectiveness associated with consecutive annual vaccination [9] [10] [11] [12] [13] [14] [15] . Nabeshima et al [15] found lower immunogenicity of revaccination in 2003 among Japanese HCP compared with those unvaccinated in 2002, unrelated to prevaccination HI titers [15] . Other studies have reported better serologic response among previously unvaccinated adults [13, [35] [36] [37] although findings regarding revaccination and vaccine effectiveness have been mixed [38, 39] . antibody and effector and memory B-cell responses were greater in TIV recipients than in LAIV recipients. Prior-season TIV recipients had significantly higher preseason HI titers, but lower HI response after vaccination with either TIV or LAIV in the study season compared with those who were not vaccinated during the previous season. These subjects also had a lower effector B-cell (antibody-secreting cell) response to new TIV but not LAIV. A possible mechanism is that some of the injected hemagglutinin protein in TIV could form antigen-antibody complexes with preexisting HI antibodies, which could reduce the amount of HI antigen available for stimulating B cells [40] . In our study, we observed differences in HI antibody response based on receipt of MIIV even in participants with preseason HI < 10, suggesting that HCP with HI < 10 despite prior MIIV were primary vaccine nonresponders who were also less likely to respond to repeat vaccination. We speculate that this nonresponse may be related to exhaustion of memory B cells from prior influenza infections. Another explanation is the "antigenic distance hypothesis" [41] . By comparing the predictions from a computer model to 7 influenza outbreaks from 2 studies [9, 42] , Smith et al [41] accurately predicted year-to-year variations in vaccine efficacy, and specifically predicted that revaccination would negatively interfere with serologic response when the antigenic distance between strains in consecutive vaccines is small. Because MIIV and 2010-2011 TIV contained an identical A(H1N1)pdm09 antigen, this theory would predict lower response among those receiving a second annual vaccination with A(H1N1)pdm09, which fits what we observed.

In a recent report of T-cell and antibody responses against influenza A(H1N1)pdm09, similar CD8 recall T-cell responses to H1N1 from 1934 and 2009 implied cross-reactive T-cell responses [43] . Almost 60% of a Toronto cohort had crossreactive memory T-cell responses to influenza virus at 1 year after the pandemic. The size of the long-lived pandemic H1N1reactive memory T-cell pool was not different between infected, vaccinated, and unvaccinated individuals, suggesting that the memory T-cell response increased only transiently postinfection and was not boosted by adjuvanted MIIV. On the basis of the findings in a single donor, the authors postulate that these T cells could expand significantly postinfection. Also, 46% of vaccinated and 15% of unvaccinated donors from their seroprevalence cohort had HI ≥ 40 during summer 2010, suggesting that antibody levels were not maintained at high levels postvaccination/infection. Similar to our HCP cohort, those vaccinated with 2009-2010 MIIV were more likely to have preseason HI ≥ 40 compared with those unvaccinated in their entire cohort [43] .

HCP characteristics and TIV vaccines used differed at the 2 sites. The major TIV lot from KPNW was associated with lower mean fold-change postvaccination for those with low or intermediate HI titers at preseason compared with the response to the 2 lots from SWH. This effect was at least partially mediated by KPNW HCP characteristics and does not impact the interpretation of our results overall. This is so because when we stratified analyses by site, the same inverse association between receipt of MIIV and HI response to subsequent TIV was found, statistically significant for SWH albeit not for KPNW.

The proportion of HCP with preseason HI ≥ 40 against A(H1N1)pdm09 was 14% among 819 HCP not receiving MIIV, suggesting that some had past infection (or cross-reactive antibody). Our finding is similar to previous studies reporting 19% of emergency department providers who had HI ≥ 40 after the first wave [44] and to an estimated 20% of the US population infected with A(H1N1)pdm09 prior to the 2010-2011 season [45] .

Randomized clinical trials reported that 90%-100% of healthy adults had HI ≥ 40 post-MIIV [16] [17] [18] [19] [20] . For our cohort, the proportion of HCP with HI ≥ 40 after A(H1N1)pdm09containing TIV was 74% (95% CI, 71%-77%) overall; 85% (95% CI, 81%-88%) among those who had only TIV and 66% (95% CI, 61%-70%) among those who had both MIIV and TIV. However, the overall proportion we observed is higher than that reported for HCP in Hong Kong [46] of 54% (95% CI, 44%-63%) and in Japan [47] of 38% (95% CI, 33.2%-42.9%) after receipt of 1 unadjuvanted MIIV dose. What we observed is closest to HI ≥ 40 of 80% among HCP in the Netherlands, after 1 MF-59 adjuvanted MIIV dose [48] . Strengths of our study include its large sample size, and the ability to compare findings across study sites and major vaccine lots. There are also several limitations to our study. First, although HI ≥ 40 is considered a surrogate marker of protection for licensure of influenza vaccines [29, 34] , its association with vaccine effectiveness is limited and the clinical meaningfulness of the differences we observed is unknown. Second, the observational nature of this study introduced differences in timing of blood draws, site, and participant characteristics, which could only be adjusted for statistically in multivariate models and by subsetting our data by site. Because HCP aged ≥50 years and those with high-risk conditions are more likely to be revaccinated, we cannot rule out residual confounding. Although we included medically attended acute respiratory illness as a covariate in models, we are limited by variable prior exposure to influenza viruses and vaccines, and underlying health and immunologic status of our participants. Other potentially serious limitations include selection bias, as MIIV or TIV recipients were not randomized, and possible variable potency of many lots of 2009-2010 MIIV. There are possibly other unknown biases inherent to any nonrandomized study.

Additional study is needed to ascertain if our finding of lower immunogenicity to A(H1N1)pdm09 antigen with second annual inactivated vaccination applies to other antigens or LAIV. Whether or not lower HI titers have clinical significance also requires further research, through vaccine-effectiveness studies, especially among HCP. Finally, although annual vaccination for HCP remains a safe and effective prevention strategy [2, 24, 49] , the imperfect immune response and afforded protection [38, 39] imply that HCP should remain vigilant for wintertime respiratory illness irrespective of vaccination history to limit transmission of influenza to their patients.

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.

Incidence of influenza in healthy adults and healthcare workers: a systematic review and metaanalysis

Influenza vaccination of health-care personnel: recommendations of the Healthcare Infection Control Practices Advisory Committee (HICPAC) and the Advisory Committee on Immunization Practices (ACIP)

Interim results: influenza A (H1N1) 2009 monovalent and seasonal influenza vaccination coverage among health-care personnel-United States

Effectiveness of seasonal influenza vaccination in healthcare workers: a systematic review

The effect of giving influenza vaccination to general practitioners: a controlled trial

Serum antibodies against circulating influenza strains among vaccinated and unvaccinated general practitioners during two consecutive years

Effectiveness of influenza vaccine in health care professionals: a randomized trial

Randomized, placebo-controlled double blind study on the efficacy of influenza immunization on absenteeism of health care workers

Assessment of inactivated influenza-A vaccine after three outbreaks of influenza A at Christ's Hospital

Different protection rates in various groups of volunteers given subunit influenza virus vaccine in 1976

Responses of elderly subjects to monovalent A/USSR/77 (H1N1) and trivalent A/ USSR/77 (H1N1)-A/TEXAS/77 (H3N2)-B/Hong Kong/72 vaccines

Immune response of adults to sequential influenza vaccination

Influenza A(H1N1)pdm09 HI Titers in HCP •

HI antibody kinetics in adult volunteers immunized repeatedly with inactivated trivalent influenza vaccine in 1990-1992

Paradoxical response to a novel influenza virus vaccine strain: the effect of prior immunization

Antibody response to influenza vaccine in adults vaccinated with identical vaccine strains in consecutive years

Phase 2 assessment of the safety and immunogenicity of two inactivated pandemic monovalent H1N1 vaccines in adults as a component of the U.S. pandemic preparedness plan in 2009

Metaanalysis of the immunogenicity and tolerability of pandemic influenza A 2009 (H1N1) vaccines

A single dose of unadjuvanted novel 2009 H1N1 vaccine is immunogenic and well tolerated in young and elderly adults

Immune response after a single vaccination against 2009 influenza A H1N1 in USA: a preliminary report of two randomised controlled phase 2 trials

Response to a monovalent 2009 influenza A (H1N1) vaccine

The expected emotional benefits of influenza vaccination strongly affect pre-season intentions and subsequent vaccination among healthcare personnel

Self-rated health and mortality: a review of twenty-seven community studies

Comparing social determinants of self-rated health across the United States and Canada

Prevention and control of influenza with vaccines: recommendations of the Advisory Committee on Immunization Practices (ACIP)

Update: influenza activity-United States, 2009-10 season

Update: influenza activity-United States, 2010-2011 season, and composition of the 2011-12 influenza vaccine

Methods and techniques in virology

World Health Organization Global Influenza Surveillance Network. Serological diagnosis of influenza by haemagglutination inhibition testing. In: Manual for the laboratory diagnosis and virological surveillance of influenza

Seroprotection rate, mean fold increase, seroconversion rate: which parameter adequately expresses seroresponse to influenza vaccination?

The role of serum haemagglutination-inhibiting antibody in protection against challenge infection with influenza A2 and B viruses

Immunity to attenuated influenza virus WRL 105 infection induced by heterologous, inactivated influenza A virus vaccines

Osterhaus AD. Haemagglutination-inhibiting antibody to influenza virus

What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes

On the relationship between mean antibody level, seroprotection and clinical protection from influenza

Efficacy of sequential annual vaccination with inactivated influenza virus vaccine

Effects of repeated annual influenza vaccination on vaccine sero-response in young and elderly adults

Kinetics of humoral antibody response to trivalent inactivated split influenza vaccine in subjects previously vaccinated or vaccinated for the first time

Immunogenicity and safety of inactivated influenza vaccines in primed populations: a systematic literature review and meta-analysis

Protection against influenza after annually repeated vaccination: a meta-analysis of serologic and field studies

Influence of prior influenza vaccination on antibody and B-cell responses

Variable efficacy of repeated annual influenza vaccination

Efficacy of repeated annual immunization with inactivated influenza virus vaccines over a five year period

Humoral and cell-mediated immunity to pandemic H1N1 influenza in a Canadian cohort one year post-pandemic: implications for vaccination

H1N1 hemagglutinin-inhibition seroprevalence in emergency department health care workers after the first wave of the 2009 influenza pandemic

Fry AM; H1N1 Serosurvey Working Group. Prevalence of seropositivity to pandemic influenza A/ H1N1 virus in the United States following the 2009 pandemic

Seroprevalence of pandemic H1N1 antibody among health care workers in Hong Kong following receipt of monovalent 2009 H1N1 influenza vaccine

Immunogenicity of a monovalent pandemic influenza A H1N1 vaccine in health-care workers of a university hospital in Japan

Immunogenicity, boostability, and sustainability of the immune response after vaccination against influenza A virus (H1N1) 2009 in a healthy population

Prevention and control of influenza with vaccines: recommendations of the Advisory Committee on Immunization Practices (ACIP)-United States, 2012-13 influenza season

Acknowledgments. The authors thank the following persons at the Influenza Division, National Center for Immunization and Respiratory Diseases, CDC, for critical review of this manuscript: Joe Miller, PhD; Jackie Katz, PhD; David Shay, MD, MPH; Alicia Fry, MD, MPH; Joe Breese, MD; and Jerry Tokars, MD. We appreciate the research teams at Scott & White Healthcare, Kaiser Permanente Center for Health Research, and Abt Associates. We also thank the healthcare personnel in direct patient care at the study sites who volunteered to participate in this study.Disclaimer. The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the CDC, Abt Associates, Inc, Kaiser Permanente Center for Health Research, or Scott & White Healthcare.Financial support. This work was supported by the CDC (contract 200-2010-F-33396 to Abt Associates Inc). This research was supported in part by an appointment to the Research Participation Program at the CDC administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the US Department of Energy and the CDC.Potential conflicts of interest. M. G. has received research funding from MedImmune and Novartis, and A. N. has received research funding from GlaxoSmithKline. All other authors report no potential conflicts.All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.