A Comparison of 2 Influenza Vaccine Schedules in 6- to 23-Month-Old Children
http://www.100md.com
《小儿科》
Division of Pediatric Infectious Diseases, Allergy, and Rheumatology, University of Washington and Children's Hospital and Regional Medical Center, Seattle, Washington
Department of Pediatrics, Duke University Medical Center, Durham, North Carolina
Department of Pediatrics, Madigan Army Medical Center, Tacoma, Washington
School of Public Health, University of Michigan, Ann Arbor, Michigan
Division of Allergy and Infectious Diseases, Department of Medicine, University of Washington School of Medicine, Seattle, Washington
ABSTRACT
Background. Trivalent inactivated influenza vaccine (TIV) is recommended for all children ages 6 to 23 months. Delivering 2 doses of TIV at least 4 weeks apart to young children receiving this vaccine for the first time is challenging.
Methods. We compared the immunogenicity and reactogenicity of the standard 2-dose regimen of TIV administered in the fall with an early schedule of a single spring dose followed by a fall dose of the same vaccine in healthy toddlers 6 to 23 months of age. Children were recruited in the spring to be randomized into either the standard or early schedule. An additional group was also enrolled in the fall as part of a nonrandomized standard comparison group. The 2002–2003 licensed TIV was administered in the spring; the fall 2003–2004 vaccine contained the same 3 antigenic components. Reactogenicity was assessed by parental diaries and telephone surveillance. Blood was obtained after the second dose of TIV for all children. The primary outcome measure was antibody response to influenza A/H1N1, A/H3N2, and B after 2 doses of vaccine, as determined by hemagglutination-inhibition titers 1:32 and geometric mean titer (GMT).
Results. Two hundred nineteen children were randomized to receive either the standard or early TIV schedule; 40 additional children were enrolled in the fall in the nonrandomized standard group. Response rates in the combined standard versus early groups were similar overall: 78% (GMT: 48) vs 76% (GMT: 57) to H1N1, 89% (GMT: 115) vs 88% (GMT: 129) to H3N2, and 52% (GMT: 24) vs 60% (GMT: 28) to B. Reactogenicity after TIV in both groups of children was minimal and did not differ by dose, age, or time between doses. Reaction rates were higher in those receiving TIV and concomitant vaccines compared with those receiving TIV alone. Overall rates of fever >38°C axillary and injection-site pain, redness, or swelling were 5.4%, 3.1%, 0.9%, and 1.1%, respectively.
Conclusions. When the spring and fall influenza vaccines had the same 3 antigenic components, the early vaccine schedule resulted in similar immunogenicity and reactogenicity compared with the standard schedule. When the vaccine components do not change between years, initiating influenza vaccine in the spring at the time of routine office visits would facilitate full immunization of children against influenza earlier in the season.
Key Words: inactivated influenza vaccine children
Abbreviations: TIV, trivalent inactivated influenza vaccine CI, confidence interval HAI, hemagglutination-inhibition GMT, geometric mean titer
Trivalent inactivated influenza vaccine (TIV) has been demonstrated to be safe and effective in children and is licensed for those 6 months of age.1–6 Data describing risks of influenza disease5–11 suggest that children ages 6 to 23 months are at substantially increased risk for influenza-related hospitalizations. In 2004, the Centers for Disease Control and Prevention's Advisory Committee on Immunization Practices6 recommended routine immunization for all children aged 6 to 23 months, as well as their close contacts.
Previously unvaccinated and influenza-naive young children, unlike adults and older children, need 2 doses of vaccine to achieve an optimal antibody response.6 Therefore, 2 doses of influenza vaccine are recommended for children between the ages of 6 and 23 months during a 3-month period (between September and November). Although influenza vaccine is likely to reduce the burden of disease in this age group, implementation of the standard influenza vaccination schedule in young children creates logistic problems for health care providers and concern about increased number of visits and injections.
It is not known if the antigen content of each of the 2 doses of influenza vaccine must be identical to optimally immunize children in this age group or if the 2 doses must be spaced 4 weeks apart, as is routinely recommended. We hypothesized that an alternate vaccination schedule could be used to enhance the delivery of 2 doses of influenza vaccine to young children. Specifically, we examined whether the administration of the previous year's influenza vaccine for the first dose in the spring would adequately prime for the second dose of vaccine in the fall in children who had not received TIV previously. This alternate vaccination regimen would allow a much broader time period for administration of the 2 doses of influenza vaccine. Coincidental to the initiation of this study, the formulation of TIV did not change between the fall of 2002 and the fall of 2003. Thus, the goal of our study was to compare reactogenicity and immunogenicity of 2 doses of the influenza vaccine with the same vaccine components administered according to the standard schedule compared with that using a spring/fall schedule in young children.
METHODS
Study Design
This study was a prospective, randomized, open-label, clinical trial conducted at 3 clinics near Seattle, Washington (Skagit Valley Pediatrics [Children's Hospital and Regional Medical Center], Madigan Army Medical Center, and Virginia Mason Medical Center) and 2 pediatric practices in Durham, North Carolina (Duke Children's Primary Care and Durham Pediatrics), with children enrolled between April 1 and October 15, 2003. Each institutional review board approved the study protocol, and informed consent was obtained from a parent or guardian of all study participants. Site-specific randomization was allocated by an investigator not involved in patient recruitment or patient care and performed in blocks of 10 by using random-number tables. Individual patient-randomization codes were placed in sealed envelopes opened by the study nurse only after informed consent was obtained. In this open-label trial, children enrolled in the spring were randomized to either the "early" or "standard" groups; additional children enrolled in the fall ("nonrandomized standard" group) were not randomized but prospectively followed. Those randomized to the early group received their first dose of 2002–2003 TIV in the spring and 2 doses of 2003–2004 TIV in the fall (a total of 3 doses), whereas children in both standard groups received no vaccine in the spring and 2 doses of 2003–2004 TIV in the fall (Table 1). Thus, all children received 2 doses of 2003–2004 TIV in the fall, in accordance with Advisory Committee on Immunization Practices guidelines.6 A limit of 2 blood draws per child was included in the study design. To achieve the primary outcome, blood samples were obtained 4 weeks after the second dose of influenza vaccine in both groups. Blood samples were drawn also before the first dose of vaccine in the early group and 4 weeks after the first dose of vaccine in the standard group.
The primary outcome of this study was to demonstrate noninferiority of the antibody response after an early-dosing schedule compared with the standard dosing schedule. This noninferiority was based on the proportion of subjects in each group achieving a titer of at least 1:32 after vaccination to 2 of the 3 antigens (H1N2, H3N2, and B) contained in the 2003–2004 vaccine. For each antigen, the antibody response was proposed to be noninferior if the upper bound of the 95% confidence interval (CI) of the difference between the proportion of children in the standard and early groups with postvaccination titers 1:32 was <15%. A sample size of 235 per group was initially calculated for this study to have 80% power in favor of noninferiority between the standard and early groups. We assumed a 1-sided of .05 and expected seroprotection rates in the standard group of 80% and <5% lower rates in the early group for each included antigen, with a 15% maximum margin.
Secondary study objectives included determining the immune status of children before any dose of TIV (early group only); comparing the proportion of seroresponders after 1 injection (standard group) versus 2 injections (standard group); comparing the proportion of seroresponders between the early group and the standard group after the first dose of vaccine in the fall; characterizing the seroconversion rate in children receiving 2 doses of vaccine separated by up to 6 months (early group only); and comparing reactogenicity in the 2 groups.
Population
Healthy children born between November 1, 2001, and December 31, 2002, were eligible for enrollment. Children were enrolled if parents gave informed consent, planned to be available for all study visits, and had telephone access. Subjects with acute febrile illnesses were eligible for enrollment, but immunization was deferred for 24 hours after the last temperature >38°C axillary. Children were excluded from enrollment if there was a history of premature birth before 36 weeks' gestation, previous receipt of influenza vaccine of any kind, allergy to eggs or egg products, history of Guillain-Barre syndrome, immunosuppression as a result of underlying illness or treatment, any acute or chronic condition that in the opinion of the investigator would render vaccination unsafe or ineffective, history of receiving immunoglobulin or other blood product within 3 months before enrollment, receipt of a live-virus vaccine (eg, measles-mumps-rubella, varicella) within the preceding 4 weeks, or need to obtain a live-virus vaccine within the upcoming consecutive 4 weeks. Simultaneous administration of a live-virus vaccine was permitted.6
Vaccine
Single lots of licensed 2002–2003 and 2003–2004 preservative-free TIV provided by Aventis-Pasteur (Swiftwater, PA) (A/New Caledonia/20/99 [H1N1], A/Panama/2007/99 [H3N2], and B/HongKong/1434/2002) were used throughout the trial. Vaccine was prepackaged in 0.25-mL syringes and administered intramuscularly in the thigh with a 25-gauge needle using standard sterile technique.
Immunogenicity
Sera were stored frozen at –20°C or less until analyzed at the University of Michigan. Hemagglutination-inhibition (HAI) antibody titers were determined in duplicate, running all paired specimens in the same test. Antigens were provided by the Centers for Disease Control and Prevention (A/Panama/2007/99 [H3N2], A/New Caledonia/20/99 [H1N1], and B/Hong Kong/330/01 ether extracted). Sera were treated with receptor-destroying enzyme (Denke Seiken Co Ltd, Tokyo, Japan). To inactivate the receptor-destroying enzyme, they were heated to 56°C for 30 minutes. After the treatment, the sera were at a 1:8 dilution and subjected to twofold serial dilutions. Twenty-five microliters of the diluted sera were incubated with an equal volume of antigen diluted to contain 4 to 8 hemoglutinen units, and 50 μL of a 0.5% suspension of chicken red blood cells then was added to the mixture. Antibody response was defined as an HAI titer of 1:32.12,13
Reactogenicity
Prospective evaluation of reactogenicity was obtained by using a parental diary. Parents were requested to record daily axillary temperatures, any local reactions (pain, tenderness, redness, swelling at the site of TIV), and systemic reactions (irritability, alteration in sleep behavior, emesis, change in appetite) for 5 days after vaccination. In addition, parents were contacted by telephone between 3 and 5 days after vaccination to confirm temperatures and any adverse reactions. Parents were contacted again 6 months after the last dose of vaccine to inquire about any serious adverse reactions and parental preferences regarding vaccine schedules.
Statistics
Descriptive and exploratory analyses were used to evaluate demographic characteristics stratified by different vaccine-regimen groups. Univariate analyses were performed to assess the associations between reactogenicity, concomitant vaccines, and groups. Antibody titers were expressed as log2, and geometric mean titers (GMTs) were reported. Any titer <1:8 was assigned a minimum value of 4. Vaccination titers 1:32 were considered positive. Seroconversion rate was defined as number of subjects with a fourfold rise divided by the number of paired preimmunization and post-second-dose subjects. Correlation coefficients by Spearman method were reported between seroconversion rate and positive postvaccination titers. All comparisons were performed using 2 test or Fisher's exact test when appropriate for contingency tables and t-test for continuous variables. Noninferior tests were performed using StatXact version 6.0 (Cytel Corporation, Cambridge, MA), and the remaining analyses were performed using SAS version 9.1 (SAS Institute, Cary, NC).
RESULTS
Subjects
All children in the early and standard group were enrolled and randomized between April 2003 and June 2003. Overall, 284 children between the ages of 6 and 24 months of age were enrolled in the study; 118 in the early group, 126 in the standard group, and 40 in the nonrandomized standard group. Twenty-five subjects withdrew from the study before receiving the first dose of vaccine; these subjects were excluded from additional analysis, leaving a total of 259 evaluable subjects (Table 2). At enrollment, children in the early, standard, and nonrandomized standard control group were similar in terms of gender and race (65.8%, 66.7%, and 60% were white, with 16.7%, 10.5, and 15% reporting black, and 8.8%, 8.6%, and 17.5% Hispanic, respectfully). Nine children (6 early, 3 standard) withdrew from the study or became lost to follow-up after the first dose but before the second dose of vaccine, and 6 children in the early group and 11 children in 1 of the standard groups withdrew from the study after the second TIV but before a second blood sample was drawn. Altogether, 233 children received 2 immunizations (102 early, 99 standard, 32 late standard), and 102 children in the early group received 3 immunizations.
A blood sample after the second dose of vaccine was obtained from 212 children, and paired blood samples were available from 192 subjects (83 early, 80 standard, 29 late standard). A total of 107 children in the early group and 136 children in the combined standard groups (99 randomized, 37 nonrandomized) were evaluated for immunogenicity outcomes.
All children with data captured after TIV are included in the reactogenicity analysis. Although 611 doses of TIV were administered, reactogenicity data after 577 immunizations in 259 children were recorded and analyzed. The number of children with evaluable reactogenicity data for each dose of vaccine is listed in Table 2. Because the number of subjects with blood samples of sufficient volume for analysis or returned diaries after each dose of vaccine differed slightly, the number of subjects analyzed for immunogenicity or reactogenicity at each time point is not consistent (Table 2).
Of the 259 evaluable subjects, 49% were female, 65% were white, 14% were black, 10% were Hispanic, and 11% reported other racial or ethnic backgrounds. No significant differences between gender or race/ethnicity were noted between or among any group. The age at the time of the first dose of TIV was significantly younger in the early compared with the standard and nonrandomized standard groups (P < .001), in accordance with the study design. The age at first dose of TIV was similar for the randomized and nonrandomized standard group (P = .289). The mean age at first TIV was 11.4 months for the early group, 15.1 months for the standard group, and 14.1 months for the nonrandomized standard group (P < .001 comparing early versus combined standard plus nonrandomized standard group). Over half (56%) of the children in the early group received their first vaccine at <1 year of age compared with 25% of those in the standard group.
By the time the children received their second dose of TIV, 80% of the early group, 83% of the standard group, and 90% of the nonrandomized standard group were >1 year old. The mean age at the time of the second dose of influenza vaccine was 15.1 months for the early group and 16.0 months for the combined standard and nonrandomized standard groups (P = .041, comparing early versus combined standard groups). Because there were no statistically significant differences between the standard and the nonrandomized standard groups based on gender, race, date of birth, age at first TIV dose, and age at second TIV dose, those 2 groups have been combined into 1 group (combined standard group) for the purposes of this article.
Reactogenicity
Complete reactogenicity data were available from 94% of children immunized in this study (577 of 611). Of these, data were captured from 247 of 259 children who received 1 dose, 236 of 250 children who received 2 doses, and 94 of 102 who received 3 doses of TIV. Overall, TIV was well tolerated, with relatively low rates of reactions (Table 2). Temperatures >38°C axillary during the first 3 days after vaccination were reported in 5.2% of children overall, with no difference between groups, and only 1 child (0.2%) reported a fever of >39.5°C. Rates of overall moderate to severe pain, redness, or swelling were 3%, 1%, and 1%, respectively, and these rates did not differ based on the age of the child or number of previous doses of TIV (Fig 1). Rates of fever of >37.8°C axillary were 7% when TIV was given concomitantly with pneumococcal conjugate vaccine (Prevnar; N = 96), 4% when given with any diphtheria-tetanus toxoid-acellular pertussis combination vaccine (N = 75), and 5% when given alone (N = 361), but there was no statistically significant association between fever and specific concomitant vaccines or between fever with TIV alone and any concomitant vaccine.
Reactions after vaccinations were also compared between those who received TIV alone and those who received TIV in combination with any other vaccine, including diphtheria-tetanus toxoid-acellular pertussis combination vaccine, Haemophilus influenzae type b vaccine, measles-mumps-rubella vaccine, and varicella vaccine. Children receiving TIV alone demonstrated significantly less irritability (20% vs 11%; P = .004), tenderness (7% vs 3%; P = .015), and swelling (3% vs 0.3%; P = .023) but no significant differences in sleeping, appetite, or pain. Increased rates of fever, pain, redness, swelling, or irritability were not seen after the second shot regardless of whether the doses were separated by 1 month in the combined standard groups or by 4 to 6 months in the early group. Increased reactogenicity was not seen after the third dose of TIV (early group only), with reported rates of moderate to severe fever, pain, redness, and swelling of 5%, 1%, 0%, and 0%, respectively. No significant increase in fever, redness, or swelling was seen based on increasing age at vaccination. Other reactions assessed, including changes in appetite, sleep-pattern disturbances, and vomiting, did not differ significantly by vaccine group or number of doses of TIV received. Rates for moderate to severe changes in appetite, sleep-pattern disturbances, and vomiting were reported at <15% across all vaccine groups after all vaccine doses. No serious adverse events possibly or probably related to TIV were reported
Immunogenicity
Primary Outcome
For influenza A/H1N1, 76% of the early group and 78% of the standard group had antibody levels 1:32, and for A/H3N2, 88% and 89% of the same groups had antibody levels 1:32. The antibody response to influenza B was lower in both groups, with 59.8% of the early group and 51.7% of the combined standard groups having antibody levels 1:32 (Fig 2A). Response rates to both H1N1 and H3N2 antigens in the early group were noninferior to the standard group (P = .002 and P < 0.001, respectively). The seroresponse rate to influenza B was greater in the early group and statistically not noninferior to the seroresponse rate in the combined standard group (P = .155). However, the seroresponse rate in the early group was not significantly greater (P = .24). Therefore, the primary outcome of noninferiority was not met with respect to influenza B antibody response. GMTs to influenza A/H1N1, A/H3N2, and B antigens after 2 doses of vaccine were 57.2 ± 4.2, 129 ± 3.7, and 28.1 ± 3.9 in the early group and 47.7 ± 3.1, 114.6 ± 3.3, and 24.3 ± 3.9 in the combined standard group, respectively (Fig 2B) and were not statistically different between the early and standard combined groups. Although we did not meet our estimated enrollment goal, power calculations based on the actual seroprotection rates showed that we had 75% power to show noninferiority for the A/H1N1 antigen and >90% power to show noninferiority for the A/H3N2 and B antigens given our enrolled sample size.
Secondary Outcomes
Antibody levels were determined before any influenza vaccine in 98 children in the early group. In these children (mean age: 11.4 months; range: 6–19.5 months), very little preexisting antibody to any influenza antigen was present: only 13%, 5%, and 1% of children had levels of HAI antibody 1:32 to H1N1, H3N2, and B, and the GMT of HAI antibody to H1N1, H3N2, and B was 6.3, 5.3, and 4.3, respectively. Antibody titers were determined 4 weeks after 1 dose of influenza vaccine in the combined standard group only. These children, with a mean age of 14.5 months at first vaccination, also had low HAI antibody titers to all 3 of the vaccine antigens: only 16% and 8% of children had antibody titers of 1:32 to H1N1 and B, respectively, and the GMTs were 10.6 and 5.9, respectively. Higher antibody to H3N2 was seen after 1 dose of TIV in this group, with nearly half (46%) having titers >1:32 and a GMT of 21, possibly reflecting past exposure to H3N2 either during the previous influenza season or early in the season of the study. Despite the relatively low seroconversion rates to both A/H1N1 and B in these children after 1 dose of TIV, the GMT antibody to all 3 vaccine antigens was greater in these children compared with the baseline antibody in the early group.
The fourfold seroconversion rate comparing preimmunization and post-second dose was 84%, 90%, and 70% (measured in the early group only) against A/H1N1, A/H3N2, and B antigens, respectively. Because of the low preimmunization titers in these groups, the fourfold seroconversion rate correlated well with the percentage of children with postvaccination titers of 1:32. The correlation coefficients were 0.5, 0.3, and 0.7 against the 3 vaccine antigens, and all P values were <.001.
Beginning in early September, all families received regular reminders via letters, phone calls, and appointment reminders to have the child immunized. Despite active encouragement by study personnel to initiate timely vaccination in all study participants and availability of TIV in early September, weeks earlier than routinely available in the participating clinics, receipt of the second TIV dose remained significantly delayed in the standard group (Fig 3). Ninety-five percent of the children in the early group had received their second dose of TIV by October 28, but 95% of children in the standard group did not receive their second dose until December 2, a difference of 35 days. Based on surveillance at the University of Washington Clinical Virology Laboratory and the state of North Carolina, the influenza season began approximately on November 15, 2003. The goal of influenza-vaccination programs is to complete the vaccination series at least 2 weeks before influenza season begins, thereby allowing time for the development of a protective antibody response. On November 2, 2 weeks before influenza season was widespread at both sites, only 60% of children in the combined standard group had received 2 doses of vaccine, whereas 97% of children in the early group had received their second dose. Children in the early group were more likely to have a protective antibody level by the time the epidemic became widespread.
Parents were surveyed 6 months after study completion to determine preference regarding influenza vaccine schedule. Responses were obtained from 51% of parents of study subjects. Two thirds of parents (66.4%) preferred their child to receive 1 shot in the spring and the second dose in the fall, with 26% preferring 2 shots in the fall and the remainder expressing no preference. The vast majority (83%) preferred their children to receive the influenza vaccine at the same time as other vaccines; only 10% wanted their child to receive the influenza vaccine as the sole injection at an office visit. The majority of parents who preferred their children to be vaccinated in the spring-fall schedule and the fall schedule also preferred the influenza vaccine to be administered with other childhood vaccines (83.9% and 76.5%, respectively).
DISCUSSION
In this study evaluating influenza-vaccine schedules in young children, we demonstrated that when vaccine components do not change between years, administering the initial dose of TIV in the spring followed by a single fall dose produced antibody responses similar to those achieved after 2 sequential TIV doses in the fall. Furthermore, children given the initial "priming" dose of TIV in the spring at the time of a routinely scheduled visit were potentially protected against influenza at the onset of the 2003–2004 season, an important consideration during a season that started early in most areas of the country and was categorized as moderately severe by Centers for Disease Control and Prevention mortality criteria.14 It is important to note that initial immunization with TIV in young children at routine office visits was also preferred by parents in our study.
Prospective studies evaluating efficacy and immunogenicity of influenza vaccine in children have been published,1,2,4,12,15–17 but relatively limited data from clinical trials are available in children <2 years of age.17–22 Previous studies have demonstrated that children between 2 and 6 years who do not have detectable HAI antibody levels have lower antibody responses to vaccine,2 and that antibody responses in young children to influenza B antigens after vaccine17,23 or wild-type influenza B virus infection7 can be substantially lower than responses to A antigens. Thus, the low antibody responses seen in infants in both groups to influenza B was not surprising. In our study, both the initial and repeated doses of TIV were well tolerated, with low rates of fever, pain, redness, or swelling and no significant change in reactogenicity after subsequent doses. The reactogenicity data presented here after 568 doses of TIV in healthy children <2 years of age provides 1 of the largest data sets thus far and further demonstrates the excellent safety profile of this vaccine. However, our study does not provide sufficient power to detect less common serious adverse events.
A survey of primary care physicians reported that only 50% of pediatricians and 40% of family physicians considered the implementation of routine influenza vaccine to young children down to 6 months of age feasible because of issues related to cost, vaccine-safety issues, and the inability to identify eligible children.17 The practical advantages of requiring only 1 fall dose of vaccine in infants are therefore important to consider. Although recent guidelines recommend routine administration of influenza vaccine for children between the ages of 6 and 23 months and their contacts,6 providing 2 doses of vaccine to large numbers of young children within 2 to 3 months is problematic for many clinics and pediatric practices, and often logistically difficult for caretakers as well. Vaccination campaigns are recommended to begin in October6 due to uncertain availability of vaccine before that time. During recent years, supply-related issues have resulted in an even shorter time period in which to administer 2 doses of vaccine to children. Vaccine shortages disproportionately affect this age group because when vaccine supplies are limited, less vaccine becomes available for the second dose in the younger children. In our study, vaccination was available early, without cost, and at a time convenient to parents, and personalized reminders and phone calls were consistently used, but completion of 2 doses of vaccine still occurred surprisingly late in the season. Based on our immunogenicity data, receipt of the second dose of TIV seems to be necessary to elicit potentially protective levels of influenza-specific antibody, but receipt of the second dose of influenza vaccine in the standard group was substantially delayed. Therefore, a spring-fall schedule would benefit children by providing immunity by the time at which influenza might begin to circulate. It is important to note that parents in our study preferred obtaining influenza vaccine at regularly scheduled visits, which included routine visits in the spring.
Our data confirm the recommendation that 2 doses of TIV are needed in previously unvaccinated infants. Seroconversion after 1 dose of vaccine in young children was minimal, but 2 doses of vaccine, regardless of when the first dose was administered, resulted in significantly higher antibody levels compared with baseline levels or levels after 1 dose. The vaccine used in the spring group in this study was a commercially available, licensed vaccine used within the scheduled usage period, a vaccine that otherwise would likely have been discarded. If a spring-fall strategy were to be implemented, the prior season's remaining vaccine could potentially be used to prime children in the spring for the upcoming fall season.
Our study demonstrates excellent priming of infants for influenza antibody response even with an increased time interval between doses and demonstrates no advantage in waiting until the child is older to administer TIV. The major limitation of our study was that vaccine antigens did not change between the spring and fall, a situation that had not occurred since 1985–1986. Therefore, we are unable to assess the impact of an early schedule when vaccine antigen(s) change from 1 year to another. A study to assess the impact of early priming with a vaccine in which 2 of the 3 antigens differ is currently underway.
ACKNOWLEDGMENTS
This study was funded by an unrestricted grant from Aventis Pasteur, Inc (Swiftwater, PA).
We gratefully acknowledge the assistance of the physicians and staff at Skagit Valley Pediatrics (Dr Frances Chalmers and colleagues); Virginia Mason Pediatrics/Federal Way (Dr Jon Almquist and colleagues); Well Child Clinic at Madigan Army Medical Center (Dr Elizabeth Hasert and colleagues); Duke Children's Primary Care (Dr Elizabeth Landolfo and colleagues); and Durham Pediatrics (Dr Martha Gagliano and colleagues). We also thank our research nurses in this study (Lynn Harrington, Lorna Imbruglio, Diane Kinnunen, Laurel Laux, Lisa Pulley, and Leslie Walker), research assistants (Leigh Ellen Floyd and Sheda Moori), and the participating families and their children.
FOOTNOTES
Accepted Jan 5, 2005.
This work was presented in part at the 42nd Infectious Disease Society of America Meeting; October 2, 2004; Boston, MA. Abstracts 1018–1019.
The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of Defense.
Conflict of interest: Dr Englund received research support from Aventis Pasteur and MedImmune, Inc, and Drs Walter, Fairchok, Monto, and Neuzil received research support from Aventis-Pasteur.
REFERENCES
Neuzil KM, Dupont WD, Wright PF, Edwards, KM. The efficacy of inactivated and cold-adapted vaccines against influenza A infection, 1985 yo1990: the pediatric experience. Pediatr Infect Dis J. 2001;20 :733 –740
Hurwitz ES, Haber M, Chang A. Studies of the 1996–1997 inactivated influenza vaccine among children attending daycare: immunologic response, protection against infection, and clinical effectiveness. J Infect Dis. 2000;182 :1218 –1221
Griffin MR, Neuzil KM. Global implications of influenza in Hong Kong. N Engl J Med. 2002;347 :2159 –2162
American Lung Association Asthma Clinical Research Centers. Safety of inactivated influenza vaccine in adults and children with asthma. N Engl J Med. 2001;345 :1529 –1536
Neuzil KM. Influenza vaccine for children with asthma: why no progress J Pediatr. 2001;138 :301 –303
Harper SA, Fukuda K, Uyeki TM, Cox NJ, Bridges CB. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP) [published correction appears in MMWR Recomm Rep. 2004;53:743]. MMWR Recomm Rep. 2004;53(RR-6) :1 –40
Neuzil KM, Zhu Y, Griffin MR, et al. Burden of interpandemic influenza in children younger than 5 years: a 25-year prospective study. J Infect Dis. 2002;185 :147 –152
Neuzil KM, Mellen BG, Wright PF, Mitchel EF Jr, Griffin MR. The effect of influenza on hospitalizations, outpatient visits, and courses of antibiotics in children. N Engl J Med. 2000;342 :225 –231
Neuzil KM, Wright PF, Mitchel EF Jr, Griffin MR. The burden of influenza illness in children with asthma and other chronic medical conditions. J Pediatr. 2000;137 :856 –864
Glezen WP, Keitel WA, Taber LH, Piedra PA, Clover RD, Couch RB. Age distribution of patients with medically-attended illnesses caused by sequential variants of influenza A/H1N1: comparison to age-specific infection rates, 1978–1979. Am J Epidemiol. 1991;133 :296 –304
Chiu SS, Lau YL, Chan KH, Wong WH, Peiris JS. Influenza-related hospitalizations among children in Hong Kong. N Engl J Med. 2002;347 :2097 –2103
Hobson D, Curry RL, Beare AS, Ward-Gardner A. The role of serum haemagglutination-inhibiting antibody in protection against challenge infection with influenza A2 and B viruses. J Hyg (Lond). 1972;70 :767 –777
Dowdle WR, Coleman MT, Mostow SR, Kaye HS, Schoenbaum SC. Inactivated influenza vaccines: 2, laboratory indices of protection. Postgrad Med J. 1973;49 :159 –163
Centers for Disease Control and Prevention. Update: influenza activity-United States and worldwide, 2003–2004 season, and composition of the 2004–2005 influenza vaccine. MMWR Morb Mortal Wkly Rep. 2004;53 :547 –552
Reichert TA, Sugaya N, Fedson DS, Glezen WP, Simonsen L, Tashiro M. The Japanese experience with vaccinating schoolchildren against influenza. N Engl J Med. 2001;344 :889 –896
Clements DA, Langdon L, Bland C, Walter E. Influenza A vaccine decreases the incidence of otitis media in 6- to 36-month-old children in day care. Arch Pediatr Adolesc Med. 1995;149 :1113 –1117
Hoberman A, Greenberg DP, Paradise JL, et al. Effectiveness of inactivated influenza vaccine in preventing acute otitis media in children: a randomized controlled trial. JAMA. 2003;290; 1608 –1616
Kanra G, Marchiso P, Feiterna-Sperling C, et al. Comparison of immu-nogenicity and tolerability of a virosome-adjuvanted and a split influenza vaccine in children. Pediatr Infect Dis J. 2004;23 :300 –306
Groothuis JR, Levin MJ, Rabalais GP, Meiklejohn G, Lauer BA. Immu-nization of high-risk infants younger than 18 months with split product influenza vaccine. Pediatrics. 1991;87 :823 –828
Piedra PA, Glezen WP, Mbawuike I, et al. Studies on reactogenicity and immunogenicity of attenuated bivalent cold recombinant influenza type A (CRA) and inactivated trivalent influenza virus (TI) vaccines in infants and young children. Vaccine. 1993;11 :718 –724
Terebuh P, Uyeki T, Fukuda K. Impact of influenza on young children and the shaping of United States influenza vaccine policy. Pediatr Infect Dis J. 2003;22(suppl) :S231 –S235
Neuzil KM, Edwards KM. Influenza vaccines in children. Semin Pediatr Infect Dis. 2002;13 :174 –181
Davies JR, Grilli EA. Natural or vaccine-induced antibody as predictor of immunity in the fact of natural challenge with influenza viruses. Epidemiol Infect. 1989;102 :325 –333
Humiston SG, Szilagyi PG, Iwane MK, et al. The feasibility of universal influenza vaccination for infants and toddlers. Arch Pediatr Adolesc Med. 2004;158 :867 –874(Janet A. Englund, MD, Emm)
Department of Pediatrics, Duke University Medical Center, Durham, North Carolina
Department of Pediatrics, Madigan Army Medical Center, Tacoma, Washington
School of Public Health, University of Michigan, Ann Arbor, Michigan
Division of Allergy and Infectious Diseases, Department of Medicine, University of Washington School of Medicine, Seattle, Washington
ABSTRACT
Background. Trivalent inactivated influenza vaccine (TIV) is recommended for all children ages 6 to 23 months. Delivering 2 doses of TIV at least 4 weeks apart to young children receiving this vaccine for the first time is challenging.
Methods. We compared the immunogenicity and reactogenicity of the standard 2-dose regimen of TIV administered in the fall with an early schedule of a single spring dose followed by a fall dose of the same vaccine in healthy toddlers 6 to 23 months of age. Children were recruited in the spring to be randomized into either the standard or early schedule. An additional group was also enrolled in the fall as part of a nonrandomized standard comparison group. The 2002–2003 licensed TIV was administered in the spring; the fall 2003–2004 vaccine contained the same 3 antigenic components. Reactogenicity was assessed by parental diaries and telephone surveillance. Blood was obtained after the second dose of TIV for all children. The primary outcome measure was antibody response to influenza A/H1N1, A/H3N2, and B after 2 doses of vaccine, as determined by hemagglutination-inhibition titers 1:32 and geometric mean titer (GMT).
Results. Two hundred nineteen children were randomized to receive either the standard or early TIV schedule; 40 additional children were enrolled in the fall in the nonrandomized standard group. Response rates in the combined standard versus early groups were similar overall: 78% (GMT: 48) vs 76% (GMT: 57) to H1N1, 89% (GMT: 115) vs 88% (GMT: 129) to H3N2, and 52% (GMT: 24) vs 60% (GMT: 28) to B. Reactogenicity after TIV in both groups of children was minimal and did not differ by dose, age, or time between doses. Reaction rates were higher in those receiving TIV and concomitant vaccines compared with those receiving TIV alone. Overall rates of fever >38°C axillary and injection-site pain, redness, or swelling were 5.4%, 3.1%, 0.9%, and 1.1%, respectively.
Conclusions. When the spring and fall influenza vaccines had the same 3 antigenic components, the early vaccine schedule resulted in similar immunogenicity and reactogenicity compared with the standard schedule. When the vaccine components do not change between years, initiating influenza vaccine in the spring at the time of routine office visits would facilitate full immunization of children against influenza earlier in the season.
Key Words: inactivated influenza vaccine children
Abbreviations: TIV, trivalent inactivated influenza vaccine CI, confidence interval HAI, hemagglutination-inhibition GMT, geometric mean titer
Trivalent inactivated influenza vaccine (TIV) has been demonstrated to be safe and effective in children and is licensed for those 6 months of age.1–6 Data describing risks of influenza disease5–11 suggest that children ages 6 to 23 months are at substantially increased risk for influenza-related hospitalizations. In 2004, the Centers for Disease Control and Prevention's Advisory Committee on Immunization Practices6 recommended routine immunization for all children aged 6 to 23 months, as well as their close contacts.
Previously unvaccinated and influenza-naive young children, unlike adults and older children, need 2 doses of vaccine to achieve an optimal antibody response.6 Therefore, 2 doses of influenza vaccine are recommended for children between the ages of 6 and 23 months during a 3-month period (between September and November). Although influenza vaccine is likely to reduce the burden of disease in this age group, implementation of the standard influenza vaccination schedule in young children creates logistic problems for health care providers and concern about increased number of visits and injections.
It is not known if the antigen content of each of the 2 doses of influenza vaccine must be identical to optimally immunize children in this age group or if the 2 doses must be spaced 4 weeks apart, as is routinely recommended. We hypothesized that an alternate vaccination schedule could be used to enhance the delivery of 2 doses of influenza vaccine to young children. Specifically, we examined whether the administration of the previous year's influenza vaccine for the first dose in the spring would adequately prime for the second dose of vaccine in the fall in children who had not received TIV previously. This alternate vaccination regimen would allow a much broader time period for administration of the 2 doses of influenza vaccine. Coincidental to the initiation of this study, the formulation of TIV did not change between the fall of 2002 and the fall of 2003. Thus, the goal of our study was to compare reactogenicity and immunogenicity of 2 doses of the influenza vaccine with the same vaccine components administered according to the standard schedule compared with that using a spring/fall schedule in young children.
METHODS
Study Design
This study was a prospective, randomized, open-label, clinical trial conducted at 3 clinics near Seattle, Washington (Skagit Valley Pediatrics [Children's Hospital and Regional Medical Center], Madigan Army Medical Center, and Virginia Mason Medical Center) and 2 pediatric practices in Durham, North Carolina (Duke Children's Primary Care and Durham Pediatrics), with children enrolled between April 1 and October 15, 2003. Each institutional review board approved the study protocol, and informed consent was obtained from a parent or guardian of all study participants. Site-specific randomization was allocated by an investigator not involved in patient recruitment or patient care and performed in blocks of 10 by using random-number tables. Individual patient-randomization codes were placed in sealed envelopes opened by the study nurse only after informed consent was obtained. In this open-label trial, children enrolled in the spring were randomized to either the "early" or "standard" groups; additional children enrolled in the fall ("nonrandomized standard" group) were not randomized but prospectively followed. Those randomized to the early group received their first dose of 2002–2003 TIV in the spring and 2 doses of 2003–2004 TIV in the fall (a total of 3 doses), whereas children in both standard groups received no vaccine in the spring and 2 doses of 2003–2004 TIV in the fall (Table 1). Thus, all children received 2 doses of 2003–2004 TIV in the fall, in accordance with Advisory Committee on Immunization Practices guidelines.6 A limit of 2 blood draws per child was included in the study design. To achieve the primary outcome, blood samples were obtained 4 weeks after the second dose of influenza vaccine in both groups. Blood samples were drawn also before the first dose of vaccine in the early group and 4 weeks after the first dose of vaccine in the standard group.
The primary outcome of this study was to demonstrate noninferiority of the antibody response after an early-dosing schedule compared with the standard dosing schedule. This noninferiority was based on the proportion of subjects in each group achieving a titer of at least 1:32 after vaccination to 2 of the 3 antigens (H1N2, H3N2, and B) contained in the 2003–2004 vaccine. For each antigen, the antibody response was proposed to be noninferior if the upper bound of the 95% confidence interval (CI) of the difference between the proportion of children in the standard and early groups with postvaccination titers 1:32 was <15%. A sample size of 235 per group was initially calculated for this study to have 80% power in favor of noninferiority between the standard and early groups. We assumed a 1-sided of .05 and expected seroprotection rates in the standard group of 80% and <5% lower rates in the early group for each included antigen, with a 15% maximum margin.
Secondary study objectives included determining the immune status of children before any dose of TIV (early group only); comparing the proportion of seroresponders after 1 injection (standard group) versus 2 injections (standard group); comparing the proportion of seroresponders between the early group and the standard group after the first dose of vaccine in the fall; characterizing the seroconversion rate in children receiving 2 doses of vaccine separated by up to 6 months (early group only); and comparing reactogenicity in the 2 groups.
Population
Healthy children born between November 1, 2001, and December 31, 2002, were eligible for enrollment. Children were enrolled if parents gave informed consent, planned to be available for all study visits, and had telephone access. Subjects with acute febrile illnesses were eligible for enrollment, but immunization was deferred for 24 hours after the last temperature >38°C axillary. Children were excluded from enrollment if there was a history of premature birth before 36 weeks' gestation, previous receipt of influenza vaccine of any kind, allergy to eggs or egg products, history of Guillain-Barre syndrome, immunosuppression as a result of underlying illness or treatment, any acute or chronic condition that in the opinion of the investigator would render vaccination unsafe or ineffective, history of receiving immunoglobulin or other blood product within 3 months before enrollment, receipt of a live-virus vaccine (eg, measles-mumps-rubella, varicella) within the preceding 4 weeks, or need to obtain a live-virus vaccine within the upcoming consecutive 4 weeks. Simultaneous administration of a live-virus vaccine was permitted.6
Vaccine
Single lots of licensed 2002–2003 and 2003–2004 preservative-free TIV provided by Aventis-Pasteur (Swiftwater, PA) (A/New Caledonia/20/99 [H1N1], A/Panama/2007/99 [H3N2], and B/HongKong/1434/2002) were used throughout the trial. Vaccine was prepackaged in 0.25-mL syringes and administered intramuscularly in the thigh with a 25-gauge needle using standard sterile technique.
Immunogenicity
Sera were stored frozen at –20°C or less until analyzed at the University of Michigan. Hemagglutination-inhibition (HAI) antibody titers were determined in duplicate, running all paired specimens in the same test. Antigens were provided by the Centers for Disease Control and Prevention (A/Panama/2007/99 [H3N2], A/New Caledonia/20/99 [H1N1], and B/Hong Kong/330/01 ether extracted). Sera were treated with receptor-destroying enzyme (Denke Seiken Co Ltd, Tokyo, Japan). To inactivate the receptor-destroying enzyme, they were heated to 56°C for 30 minutes. After the treatment, the sera were at a 1:8 dilution and subjected to twofold serial dilutions. Twenty-five microliters of the diluted sera were incubated with an equal volume of antigen diluted to contain 4 to 8 hemoglutinen units, and 50 μL of a 0.5% suspension of chicken red blood cells then was added to the mixture. Antibody response was defined as an HAI titer of 1:32.12,13
Reactogenicity
Prospective evaluation of reactogenicity was obtained by using a parental diary. Parents were requested to record daily axillary temperatures, any local reactions (pain, tenderness, redness, swelling at the site of TIV), and systemic reactions (irritability, alteration in sleep behavior, emesis, change in appetite) for 5 days after vaccination. In addition, parents were contacted by telephone between 3 and 5 days after vaccination to confirm temperatures and any adverse reactions. Parents were contacted again 6 months after the last dose of vaccine to inquire about any serious adverse reactions and parental preferences regarding vaccine schedules.
Statistics
Descriptive and exploratory analyses were used to evaluate demographic characteristics stratified by different vaccine-regimen groups. Univariate analyses were performed to assess the associations between reactogenicity, concomitant vaccines, and groups. Antibody titers were expressed as log2, and geometric mean titers (GMTs) were reported. Any titer <1:8 was assigned a minimum value of 4. Vaccination titers 1:32 were considered positive. Seroconversion rate was defined as number of subjects with a fourfold rise divided by the number of paired preimmunization and post-second-dose subjects. Correlation coefficients by Spearman method were reported between seroconversion rate and positive postvaccination titers. All comparisons were performed using 2 test or Fisher's exact test when appropriate for contingency tables and t-test for continuous variables. Noninferior tests were performed using StatXact version 6.0 (Cytel Corporation, Cambridge, MA), and the remaining analyses were performed using SAS version 9.1 (SAS Institute, Cary, NC).
RESULTS
Subjects
All children in the early and standard group were enrolled and randomized between April 2003 and June 2003. Overall, 284 children between the ages of 6 and 24 months of age were enrolled in the study; 118 in the early group, 126 in the standard group, and 40 in the nonrandomized standard group. Twenty-five subjects withdrew from the study before receiving the first dose of vaccine; these subjects were excluded from additional analysis, leaving a total of 259 evaluable subjects (Table 2). At enrollment, children in the early, standard, and nonrandomized standard control group were similar in terms of gender and race (65.8%, 66.7%, and 60% were white, with 16.7%, 10.5, and 15% reporting black, and 8.8%, 8.6%, and 17.5% Hispanic, respectfully). Nine children (6 early, 3 standard) withdrew from the study or became lost to follow-up after the first dose but before the second dose of vaccine, and 6 children in the early group and 11 children in 1 of the standard groups withdrew from the study after the second TIV but before a second blood sample was drawn. Altogether, 233 children received 2 immunizations (102 early, 99 standard, 32 late standard), and 102 children in the early group received 3 immunizations.
A blood sample after the second dose of vaccine was obtained from 212 children, and paired blood samples were available from 192 subjects (83 early, 80 standard, 29 late standard). A total of 107 children in the early group and 136 children in the combined standard groups (99 randomized, 37 nonrandomized) were evaluated for immunogenicity outcomes.
All children with data captured after TIV are included in the reactogenicity analysis. Although 611 doses of TIV were administered, reactogenicity data after 577 immunizations in 259 children were recorded and analyzed. The number of children with evaluable reactogenicity data for each dose of vaccine is listed in Table 2. Because the number of subjects with blood samples of sufficient volume for analysis or returned diaries after each dose of vaccine differed slightly, the number of subjects analyzed for immunogenicity or reactogenicity at each time point is not consistent (Table 2).
Of the 259 evaluable subjects, 49% were female, 65% were white, 14% were black, 10% were Hispanic, and 11% reported other racial or ethnic backgrounds. No significant differences between gender or race/ethnicity were noted between or among any group. The age at the time of the first dose of TIV was significantly younger in the early compared with the standard and nonrandomized standard groups (P < .001), in accordance with the study design. The age at first dose of TIV was similar for the randomized and nonrandomized standard group (P = .289). The mean age at first TIV was 11.4 months for the early group, 15.1 months for the standard group, and 14.1 months for the nonrandomized standard group (P < .001 comparing early versus combined standard plus nonrandomized standard group). Over half (56%) of the children in the early group received their first vaccine at <1 year of age compared with 25% of those in the standard group.
By the time the children received their second dose of TIV, 80% of the early group, 83% of the standard group, and 90% of the nonrandomized standard group were >1 year old. The mean age at the time of the second dose of influenza vaccine was 15.1 months for the early group and 16.0 months for the combined standard and nonrandomized standard groups (P = .041, comparing early versus combined standard groups). Because there were no statistically significant differences between the standard and the nonrandomized standard groups based on gender, race, date of birth, age at first TIV dose, and age at second TIV dose, those 2 groups have been combined into 1 group (combined standard group) for the purposes of this article.
Reactogenicity
Complete reactogenicity data were available from 94% of children immunized in this study (577 of 611). Of these, data were captured from 247 of 259 children who received 1 dose, 236 of 250 children who received 2 doses, and 94 of 102 who received 3 doses of TIV. Overall, TIV was well tolerated, with relatively low rates of reactions (Table 2). Temperatures >38°C axillary during the first 3 days after vaccination were reported in 5.2% of children overall, with no difference between groups, and only 1 child (0.2%) reported a fever of >39.5°C. Rates of overall moderate to severe pain, redness, or swelling were 3%, 1%, and 1%, respectively, and these rates did not differ based on the age of the child or number of previous doses of TIV (Fig 1). Rates of fever of >37.8°C axillary were 7% when TIV was given concomitantly with pneumococcal conjugate vaccine (Prevnar; N = 96), 4% when given with any diphtheria-tetanus toxoid-acellular pertussis combination vaccine (N = 75), and 5% when given alone (N = 361), but there was no statistically significant association between fever and specific concomitant vaccines or between fever with TIV alone and any concomitant vaccine.
Reactions after vaccinations were also compared between those who received TIV alone and those who received TIV in combination with any other vaccine, including diphtheria-tetanus toxoid-acellular pertussis combination vaccine, Haemophilus influenzae type b vaccine, measles-mumps-rubella vaccine, and varicella vaccine. Children receiving TIV alone demonstrated significantly less irritability (20% vs 11%; P = .004), tenderness (7% vs 3%; P = .015), and swelling (3% vs 0.3%; P = .023) but no significant differences in sleeping, appetite, or pain. Increased rates of fever, pain, redness, swelling, or irritability were not seen after the second shot regardless of whether the doses were separated by 1 month in the combined standard groups or by 4 to 6 months in the early group. Increased reactogenicity was not seen after the third dose of TIV (early group only), with reported rates of moderate to severe fever, pain, redness, and swelling of 5%, 1%, 0%, and 0%, respectively. No significant increase in fever, redness, or swelling was seen based on increasing age at vaccination. Other reactions assessed, including changes in appetite, sleep-pattern disturbances, and vomiting, did not differ significantly by vaccine group or number of doses of TIV received. Rates for moderate to severe changes in appetite, sleep-pattern disturbances, and vomiting were reported at <15% across all vaccine groups after all vaccine doses. No serious adverse events possibly or probably related to TIV were reported
Immunogenicity
Primary Outcome
For influenza A/H1N1, 76% of the early group and 78% of the standard group had antibody levels 1:32, and for A/H3N2, 88% and 89% of the same groups had antibody levels 1:32. The antibody response to influenza B was lower in both groups, with 59.8% of the early group and 51.7% of the combined standard groups having antibody levels 1:32 (Fig 2A). Response rates to both H1N1 and H3N2 antigens in the early group were noninferior to the standard group (P = .002 and P < 0.001, respectively). The seroresponse rate to influenza B was greater in the early group and statistically not noninferior to the seroresponse rate in the combined standard group (P = .155). However, the seroresponse rate in the early group was not significantly greater (P = .24). Therefore, the primary outcome of noninferiority was not met with respect to influenza B antibody response. GMTs to influenza A/H1N1, A/H3N2, and B antigens after 2 doses of vaccine were 57.2 ± 4.2, 129 ± 3.7, and 28.1 ± 3.9 in the early group and 47.7 ± 3.1, 114.6 ± 3.3, and 24.3 ± 3.9 in the combined standard group, respectively (Fig 2B) and were not statistically different between the early and standard combined groups. Although we did not meet our estimated enrollment goal, power calculations based on the actual seroprotection rates showed that we had 75% power to show noninferiority for the A/H1N1 antigen and >90% power to show noninferiority for the A/H3N2 and B antigens given our enrolled sample size.
Secondary Outcomes
Antibody levels were determined before any influenza vaccine in 98 children in the early group. In these children (mean age: 11.4 months; range: 6–19.5 months), very little preexisting antibody to any influenza antigen was present: only 13%, 5%, and 1% of children had levels of HAI antibody 1:32 to H1N1, H3N2, and B, and the GMT of HAI antibody to H1N1, H3N2, and B was 6.3, 5.3, and 4.3, respectively. Antibody titers were determined 4 weeks after 1 dose of influenza vaccine in the combined standard group only. These children, with a mean age of 14.5 months at first vaccination, also had low HAI antibody titers to all 3 of the vaccine antigens: only 16% and 8% of children had antibody titers of 1:32 to H1N1 and B, respectively, and the GMTs were 10.6 and 5.9, respectively. Higher antibody to H3N2 was seen after 1 dose of TIV in this group, with nearly half (46%) having titers >1:32 and a GMT of 21, possibly reflecting past exposure to H3N2 either during the previous influenza season or early in the season of the study. Despite the relatively low seroconversion rates to both A/H1N1 and B in these children after 1 dose of TIV, the GMT antibody to all 3 vaccine antigens was greater in these children compared with the baseline antibody in the early group.
The fourfold seroconversion rate comparing preimmunization and post-second dose was 84%, 90%, and 70% (measured in the early group only) against A/H1N1, A/H3N2, and B antigens, respectively. Because of the low preimmunization titers in these groups, the fourfold seroconversion rate correlated well with the percentage of children with postvaccination titers of 1:32. The correlation coefficients were 0.5, 0.3, and 0.7 against the 3 vaccine antigens, and all P values were <.001.
Beginning in early September, all families received regular reminders via letters, phone calls, and appointment reminders to have the child immunized. Despite active encouragement by study personnel to initiate timely vaccination in all study participants and availability of TIV in early September, weeks earlier than routinely available in the participating clinics, receipt of the second TIV dose remained significantly delayed in the standard group (Fig 3). Ninety-five percent of the children in the early group had received their second dose of TIV by October 28, but 95% of children in the standard group did not receive their second dose until December 2, a difference of 35 days. Based on surveillance at the University of Washington Clinical Virology Laboratory and the state of North Carolina, the influenza season began approximately on November 15, 2003. The goal of influenza-vaccination programs is to complete the vaccination series at least 2 weeks before influenza season begins, thereby allowing time for the development of a protective antibody response. On November 2, 2 weeks before influenza season was widespread at both sites, only 60% of children in the combined standard group had received 2 doses of vaccine, whereas 97% of children in the early group had received their second dose. Children in the early group were more likely to have a protective antibody level by the time the epidemic became widespread.
Parents were surveyed 6 months after study completion to determine preference regarding influenza vaccine schedule. Responses were obtained from 51% of parents of study subjects. Two thirds of parents (66.4%) preferred their child to receive 1 shot in the spring and the second dose in the fall, with 26% preferring 2 shots in the fall and the remainder expressing no preference. The vast majority (83%) preferred their children to receive the influenza vaccine at the same time as other vaccines; only 10% wanted their child to receive the influenza vaccine as the sole injection at an office visit. The majority of parents who preferred their children to be vaccinated in the spring-fall schedule and the fall schedule also preferred the influenza vaccine to be administered with other childhood vaccines (83.9% and 76.5%, respectively).
DISCUSSION
In this study evaluating influenza-vaccine schedules in young children, we demonstrated that when vaccine components do not change between years, administering the initial dose of TIV in the spring followed by a single fall dose produced antibody responses similar to those achieved after 2 sequential TIV doses in the fall. Furthermore, children given the initial "priming" dose of TIV in the spring at the time of a routinely scheduled visit were potentially protected against influenza at the onset of the 2003–2004 season, an important consideration during a season that started early in most areas of the country and was categorized as moderately severe by Centers for Disease Control and Prevention mortality criteria.14 It is important to note that initial immunization with TIV in young children at routine office visits was also preferred by parents in our study.
Prospective studies evaluating efficacy and immunogenicity of influenza vaccine in children have been published,1,2,4,12,15–17 but relatively limited data from clinical trials are available in children <2 years of age.17–22 Previous studies have demonstrated that children between 2 and 6 years who do not have detectable HAI antibody levels have lower antibody responses to vaccine,2 and that antibody responses in young children to influenza B antigens after vaccine17,23 or wild-type influenza B virus infection7 can be substantially lower than responses to A antigens. Thus, the low antibody responses seen in infants in both groups to influenza B was not surprising. In our study, both the initial and repeated doses of TIV were well tolerated, with low rates of fever, pain, redness, or swelling and no significant change in reactogenicity after subsequent doses. The reactogenicity data presented here after 568 doses of TIV in healthy children <2 years of age provides 1 of the largest data sets thus far and further demonstrates the excellent safety profile of this vaccine. However, our study does not provide sufficient power to detect less common serious adverse events.
A survey of primary care physicians reported that only 50% of pediatricians and 40% of family physicians considered the implementation of routine influenza vaccine to young children down to 6 months of age feasible because of issues related to cost, vaccine-safety issues, and the inability to identify eligible children.17 The practical advantages of requiring only 1 fall dose of vaccine in infants are therefore important to consider. Although recent guidelines recommend routine administration of influenza vaccine for children between the ages of 6 and 23 months and their contacts,6 providing 2 doses of vaccine to large numbers of young children within 2 to 3 months is problematic for many clinics and pediatric practices, and often logistically difficult for caretakers as well. Vaccination campaigns are recommended to begin in October6 due to uncertain availability of vaccine before that time. During recent years, supply-related issues have resulted in an even shorter time period in which to administer 2 doses of vaccine to children. Vaccine shortages disproportionately affect this age group because when vaccine supplies are limited, less vaccine becomes available for the second dose in the younger children. In our study, vaccination was available early, without cost, and at a time convenient to parents, and personalized reminders and phone calls were consistently used, but completion of 2 doses of vaccine still occurred surprisingly late in the season. Based on our immunogenicity data, receipt of the second dose of TIV seems to be necessary to elicit potentially protective levels of influenza-specific antibody, but receipt of the second dose of influenza vaccine in the standard group was substantially delayed. Therefore, a spring-fall schedule would benefit children by providing immunity by the time at which influenza might begin to circulate. It is important to note that parents in our study preferred obtaining influenza vaccine at regularly scheduled visits, which included routine visits in the spring.
Our data confirm the recommendation that 2 doses of TIV are needed in previously unvaccinated infants. Seroconversion after 1 dose of vaccine in young children was minimal, but 2 doses of vaccine, regardless of when the first dose was administered, resulted in significantly higher antibody levels compared with baseline levels or levels after 1 dose. The vaccine used in the spring group in this study was a commercially available, licensed vaccine used within the scheduled usage period, a vaccine that otherwise would likely have been discarded. If a spring-fall strategy were to be implemented, the prior season's remaining vaccine could potentially be used to prime children in the spring for the upcoming fall season.
Our study demonstrates excellent priming of infants for influenza antibody response even with an increased time interval between doses and demonstrates no advantage in waiting until the child is older to administer TIV. The major limitation of our study was that vaccine antigens did not change between the spring and fall, a situation that had not occurred since 1985–1986. Therefore, we are unable to assess the impact of an early schedule when vaccine antigen(s) change from 1 year to another. A study to assess the impact of early priming with a vaccine in which 2 of the 3 antigens differ is currently underway.
ACKNOWLEDGMENTS
This study was funded by an unrestricted grant from Aventis Pasteur, Inc (Swiftwater, PA).
We gratefully acknowledge the assistance of the physicians and staff at Skagit Valley Pediatrics (Dr Frances Chalmers and colleagues); Virginia Mason Pediatrics/Federal Way (Dr Jon Almquist and colleagues); Well Child Clinic at Madigan Army Medical Center (Dr Elizabeth Hasert and colleagues); Duke Children's Primary Care (Dr Elizabeth Landolfo and colleagues); and Durham Pediatrics (Dr Martha Gagliano and colleagues). We also thank our research nurses in this study (Lynn Harrington, Lorna Imbruglio, Diane Kinnunen, Laurel Laux, Lisa Pulley, and Leslie Walker), research assistants (Leigh Ellen Floyd and Sheda Moori), and the participating families and their children.
FOOTNOTES
Accepted Jan 5, 2005.
This work was presented in part at the 42nd Infectious Disease Society of America Meeting; October 2, 2004; Boston, MA. Abstracts 1018–1019.
The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of Defense.
Conflict of interest: Dr Englund received research support from Aventis Pasteur and MedImmune, Inc, and Drs Walter, Fairchok, Monto, and Neuzil received research support from Aventis-Pasteur.
REFERENCES
Neuzil KM, Dupont WD, Wright PF, Edwards, KM. The efficacy of inactivated and cold-adapted vaccines against influenza A infection, 1985 yo1990: the pediatric experience. Pediatr Infect Dis J. 2001;20 :733 –740
Hurwitz ES, Haber M, Chang A. Studies of the 1996–1997 inactivated influenza vaccine among children attending daycare: immunologic response, protection against infection, and clinical effectiveness. J Infect Dis. 2000;182 :1218 –1221
Griffin MR, Neuzil KM. Global implications of influenza in Hong Kong. N Engl J Med. 2002;347 :2159 –2162
American Lung Association Asthma Clinical Research Centers. Safety of inactivated influenza vaccine in adults and children with asthma. N Engl J Med. 2001;345 :1529 –1536
Neuzil KM. Influenza vaccine for children with asthma: why no progress J Pediatr. 2001;138 :301 –303
Harper SA, Fukuda K, Uyeki TM, Cox NJ, Bridges CB. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP) [published correction appears in MMWR Recomm Rep. 2004;53:743]. MMWR Recomm Rep. 2004;53(RR-6) :1 –40
Neuzil KM, Zhu Y, Griffin MR, et al. Burden of interpandemic influenza in children younger than 5 years: a 25-year prospective study. J Infect Dis. 2002;185 :147 –152
Neuzil KM, Mellen BG, Wright PF, Mitchel EF Jr, Griffin MR. The effect of influenza on hospitalizations, outpatient visits, and courses of antibiotics in children. N Engl J Med. 2000;342 :225 –231
Neuzil KM, Wright PF, Mitchel EF Jr, Griffin MR. The burden of influenza illness in children with asthma and other chronic medical conditions. J Pediatr. 2000;137 :856 –864
Glezen WP, Keitel WA, Taber LH, Piedra PA, Clover RD, Couch RB. Age distribution of patients with medically-attended illnesses caused by sequential variants of influenza A/H1N1: comparison to age-specific infection rates, 1978–1979. Am J Epidemiol. 1991;133 :296 –304
Chiu SS, Lau YL, Chan KH, Wong WH, Peiris JS. Influenza-related hospitalizations among children in Hong Kong. N Engl J Med. 2002;347 :2097 –2103
Hobson D, Curry RL, Beare AS, Ward-Gardner A. The role of serum haemagglutination-inhibiting antibody in protection against challenge infection with influenza A2 and B viruses. J Hyg (Lond). 1972;70 :767 –777
Dowdle WR, Coleman MT, Mostow SR, Kaye HS, Schoenbaum SC. Inactivated influenza vaccines: 2, laboratory indices of protection. Postgrad Med J. 1973;49 :159 –163
Centers for Disease Control and Prevention. Update: influenza activity-United States and worldwide, 2003–2004 season, and composition of the 2004–2005 influenza vaccine. MMWR Morb Mortal Wkly Rep. 2004;53 :547 –552
Reichert TA, Sugaya N, Fedson DS, Glezen WP, Simonsen L, Tashiro M. The Japanese experience with vaccinating schoolchildren against influenza. N Engl J Med. 2001;344 :889 –896
Clements DA, Langdon L, Bland C, Walter E. Influenza A vaccine decreases the incidence of otitis media in 6- to 36-month-old children in day care. Arch Pediatr Adolesc Med. 1995;149 :1113 –1117
Hoberman A, Greenberg DP, Paradise JL, et al. Effectiveness of inactivated influenza vaccine in preventing acute otitis media in children: a randomized controlled trial. JAMA. 2003;290; 1608 –1616
Kanra G, Marchiso P, Feiterna-Sperling C, et al. Comparison of immu-nogenicity and tolerability of a virosome-adjuvanted and a split influenza vaccine in children. Pediatr Infect Dis J. 2004;23 :300 –306
Groothuis JR, Levin MJ, Rabalais GP, Meiklejohn G, Lauer BA. Immu-nization of high-risk infants younger than 18 months with split product influenza vaccine. Pediatrics. 1991;87 :823 –828
Piedra PA, Glezen WP, Mbawuike I, et al. Studies on reactogenicity and immunogenicity of attenuated bivalent cold recombinant influenza type A (CRA) and inactivated trivalent influenza virus (TI) vaccines in infants and young children. Vaccine. 1993;11 :718 –724
Terebuh P, Uyeki T, Fukuda K. Impact of influenza on young children and the shaping of United States influenza vaccine policy. Pediatr Infect Dis J. 2003;22(suppl) :S231 –S235
Neuzil KM, Edwards KM. Influenza vaccines in children. Semin Pediatr Infect Dis. 2002;13 :174 –181
Davies JR, Grilli EA. Natural or vaccine-induced antibody as predictor of immunity in the fact of natural challenge with influenza viruses. Epidemiol Infect. 1989;102 :325 –333
Humiston SG, Szilagyi PG, Iwane MK, et al. The feasibility of universal influenza vaccination for infants and toddlers. Arch Pediatr Adolesc Med. 2004;158 :867 –874(Janet A. Englund, MD, Emm)