OBJECTIVE: This study evaluated the immunogenicity, safety, and tolerability of a MF59-adjuvanted H5N1 vaccine in a population 6 months through 17 years of age.
METHODS: Healthy subjects 6 to <36 months, 3 to <9 months, and 9 to <18 years of age were assigned randomly to receive 2 doses of either a MF59-adjuvanted H5N1 vaccine (7.5 μg/dose) or a MF59-adjuvanted trivalent seasonal influenza control vaccine (15 μg/dose for each antigen). Immunogenicity against the A/Vietnam/1194/2004-like vaccine strain was measured before and 3 weeks after the 2-dose primary series, through hemagglutination inhibition (HI), single radial hemolysis (SRH), and microneutralization. Local and systemic reactions were recorded.
RESULTS: A total of 335 subjects received the H5N1 vaccine, and 137 subjects received the seasonal vaccine. Rates of seroprotection (HI titer of ≥40) against the H5N1 vaccine antigen were 97% for children 6 to 36 months and 3 to 9 years of age and 89% for older children. All subjects seroconverted in the SRH assay. Microneutralization titers of ≥40 were achieved by 99% of subjects, and ≥98% of subjects, respectively. Local reactions, particularly injection site pain in older children, were common, generally mild to moderate in nature, and transient and resolved spontaneously. Up to 5% of participants. There were no vaccine-related serious adverse events in either group.
CONCLUSIONS: In this pediatric population, MF59-adjuvanted H5N1 vaccine was highly immunogenic, had a good safety profile, reactogenicity comparable with that of an adjuvanted seasonal influenza control vaccine.
- H5N1 influenza vaccine
- prepandemic influenza vaccine
- adjuvanted influenza vaccine
- avian influenza
WHAT'S KNOWN ON THIS SUBJECT:
Prepandemic vaccines offer an important option for reducing the impact of pandemic influenza; children and adolescents represent primary vaccination targets. A MF59-adjuvanted, inactivated, subunit, H5N1 influenza virus vaccine has been licensed in Europe for pandemic vaccination of adults.
WHAT THIS STUDY ADDS:
The MF59-adjuvanted H5N1 study vaccine was highly immunogenic in a pediatric population. The adjuvanted H5N1 vaccine had a good safety profile; reactogenicity was comparable to that of an adjuvanted seasonal influenza control vaccine.
The first outbreak of A/H5N1 avian influenza in humans occurred in Hong Kong in 1997, and the virus reemerged in humans in 2003.1 Between 2003 and April 2010, 495 human cases of H5N1 infection were identified and 292 people died, which yielded a mortality rate of 59%.2 There is some evidence of human-to-human transmission or “humanization” of H5N1 influenza virus strains; therefore, H5N1 remains a pandemic threat.3,–,5 The H1N1 pandemic and possible recombination of H1N1 and H5N1, with associated high mortality rates, require preparedness for H5N1 to continue.
Pandemic preparedness plans include development, licensing, and production of specific monovalent vaccines against the H5N1 influenza virus strains.6 The availability of safe, effective, pandemic vaccines will play a crucial role in efforts to combat this pandemic threat.7,–,9 Difficulties in producing quantities of influenza vaccine sufficient to conduct mass vaccination programs during a pandemic are magnified because very few people have been exposed to the H5 hemagglutinin or “primed” through previous H5 infection or vaccination. Clinical experience with current H5N1 vaccines confirms that 1 dose does not induce a protective immune response in H5-naive individuals and that larger or multiple doses of influenza surface antigen will be necessary.8 One clinical study showed that 2 vaccinations with 90 μg of H5 hemagglutinin were necessary to achieve seroprotection against a H5N1 influenza virus strain, that is, 12 times the 15-μg dose included in seasonal influenza vaccines.10
Adjuvants increase immunogenicity, allowing reduction of the quantity of antigen per vaccine dose and potentially leading to increased numbers of doses available during a pandemic. The use of adjuvants thus can address limitations in manufacturing capacity through dose-sparing. A trivalent, inactivated, seasonal, influenza vaccine (Fluad [Novartis Vaccine and Diagnostics, Siena, Italy]) adjuvanted with MF59, an oil/water emulsion containing squalene as the oil phase, was licensed in 1997 and is marketed in the European Union, Asia, and Latin America for patients ≥65 years of age. MF59, as used in this influenza vaccine, has been shown to increase the immune response to vaccination in elderly patients11,–,14 and other at-risk populations15,–,17 A recent pediatric trial showed that MF59 increased the immunogenicity of seasonal influenza vaccine in 6- to 36-month-old infants and children without increasing side effects.18 An MF59-adjuvanted, inactivated, subunit, H5N1 influenza virus vaccine containing hemagglutinin surface antigen from A/Vietnam/1194/2004 (NIBRG-14; Focetria/Aflunov [Novartis Vaccines]) has been licensed in Europe for pandemic vaccination of adults and is being assessed for prepandemic use.
Children have high rates of morbidity resulting from influenza and are at increased risk of hospitalization for treatment of influenza and its complications.19 Because children also play a role in the transmission of influenza, vaccination of children would be expected to reduce pandemic disease burden both in vaccinees and in the general population, through a herd effect.20 Therefore, children are among the priority groups for vaccination in case of an influenza pandemic.21 For the same reasons, vaccination of children could be considered an important prepandemic strategy. The earlier that vaccination of children begins, the greater would be the expected impact on viral attack rates; therefore, it would be advantageous to have stocks of prepandemic vaccines on hand.22 This study evaluated the immunogenicity, safety, and tolerability of 2 doses of a MF59-adjuvanted, prepandemic, H5N1 vaccine, administered 3 weeks apart, in a population 6 months through 17 years of age.
Study Design and Objectives
This phase II, randomized, controlled, observer-blinded, single-center study was conducted at 15 clinics of the Vaccine Research Center of the University of Tampere Medical School (Tampere, Finland) from September through November 2007. The study was approved by the ethics committee of the Pirkanmaa University Hospital District and was conducted in accordance with the principles of the Declaration of Helsinki and good clinical practice. Subjects, parents, or legal guardians gave written informed consent before study enrollment. The primary study objective was to determine the immunogenicity of two 0.5-mL intramuscular injections (7.5 μg of hemagglutinin) of a MF59-adjuvanted, H5N1 influenza vaccine (Aflunov [Novartis Vaccines]), administered 3 weeks apart. The secondary objective was to determine the tolerability of the adjuvanted H5N1 vaccine in comparison with a seasonal influenza control vaccine (Fluad [Novartis Vaccines]).
Healthy subjects who were 6 months to 17 years of age and were able to comply with the study procedures were eligible. Subjects were excluded if they were participating or had participated (within the previous 3 months) in another clinical trial; had received a pandemic candidate vaccine, a MF59-adjuvanted vaccine, a 2007/2008 seasonal influenza vaccine, or another vaccine within 3 weeks of a study vaccination; had an acute febrile disease or were receiving antibiotics or antiviral drugs; had immunodeficiency or were receiving immunosuppressive therapy; had an allergy to eggs or any vaccine component or a history of anaphylactic shock; or had a chronic disease, such as cancer, autoimmune disease, diabetes mellitus, asthma, or hepatic or renal disease.
Subjects were separated into 3 age groups, that is, toddlers 6 to <36 months of age, children 3 to <9 years of age, and adolescents 9 to <18 years of age. With the use of lists of random numbers, participants in each age group were assigned in a 3:1 ratio to receive two 0.5-mL doses of a MF59-adjuvanted H5N1 vaccine or a MF59-adjuvanted seasonal influenza control vaccine (two 0.25-mL doses of seasonal vaccine for participants <3 years of age). The vaccines were administered 3 weeks apart, in the deltoid muscle of the nondominant arm or the anterolateral thigh, by study staff members who had access to the randomization code but no subsequent contact with study subjects. Other study staff members and the study subjects were blinded to the vaccine group assignments.
The study vaccine (Aflunov; lot 060601A), a MF59-adjuvanted, inactivated, subunit vaccine against clade 1 A/Vietnam/1194/2004 (NIBRG-14), contained 7.5 μg of H5N1 hemagglutinin surface antigen in each 0.5-mL dose. It was propagated in embryonated hens' eggs by manufacturing processes similar to those used for a licensed, MF59-adjuvanted, seasonal influenza vaccine (Fluad).19,23 Each 0.5-mL dose of the control vaccine (Fluad; lot 078402) contained 15 μg of hemagglutinin from each influenza virus strain recommended by the World Health Organization/European Union for the 2007/2008 season, that is, A/Solomon Island/3/2006 (H1N1)-like, A/Wisconsin/67/2005 (H3N2)-like, and B/Malaysia/2506/2004-like.
Blood samples were collected before the first vaccination (day 1), before the second vaccination (day 22), and 3 weeks after the second vaccination (day 43) for determination of antibody titers against the vaccine H5 hemagglutinin surface antigen, through hemagglutination inhibition (HI), microneutralization, and single radial hemolysis (SRH). The HI and microneutralization assays were performed at the clinical serology laboratory of Novartis Vaccines (Marburg, Germany); SRH assays were performed at the molecular epidemiology laboratory of the University of Siena (Siena, Italy). The HI, microneutralization, and SRH assays used were described previously.24,–,26
HI titers were expressed as the reciprocal of the highest dilution at which hemagglutination was still completely inhibited. For HI, sera were tested at an initial dilution of 1:8 and those that yielded negative results were assigned a titer of 4. For microneutralization, sera were tested at an initial dilution of 1:20 and those that yielded negative results were assigned a titer of 10. The reciprocals of twofold dilutions that achieved ≥50% neutralization of virus were considered positive. A seroprotective neutralizing antibody titer against H5N1 influenza strains has not been agreed upon but was defined here as ≥40, with seroconversion as a fourfold increase in microneutralization titer, both of which are in line with values used in other studies.10,27,–,31
The European Agency for Evaluation of Medicinal Products, Committee for Medicinal Products for Human Use, immunogenicity criteria for influenza vaccines require that ≥1 of 3 conditions be met for subjects 18 to 60 years of age, that is, (1) seroconversion rate of >40%, (2) geometric mean ratio of >2.5, or (3) seroprotection rate of >70.32 No criteria are available for subjects <18 years of age. In this trial, seroprotection was defined as a HI titer of ≥40 or a SRH area of >25 mm2. Seroconversion was defined as reaching a postvaccination titer of ≥40 for HI or an area of ≥25 mm2 for SRH for subjects with negative prevaccination titers or at least a fourfold increase in titer for HI or a 50% increase in area for SRH for subjects with positive prevaccination titer results.32
Reactogenicity and Safety
Study participants were monitored for 30 minutes after each injection, for immediate reactions. Parents or participants recorded solicited local and systemic reactions on diary cards for 7 days after each vaccination. Local reactions included ecchymosis, erythema, induration, swelling, and tenderness for subjects <3 years of age plus pain for subjects ≥3 years of age. Systemic reactions included sleepiness, diarrhea, vomiting, irritability, change in eating habits, shivering, unusual sweating, and unusual crying for subjects <3 years of age; chills, malaise, myalgia, arthralgia, headache, sweating, nausea, vomiting, diarrhea, and fatigue for subjects ≥3 years of age; and fever (axillary temperature of ≥38.0°C) for all subjects. Unsolicited adverse events and serious adverse events (SAEs) were recorded for the duration of the study.
Geometric mean titers and areas with 95% confidence intervals (CIs) were calculated for the HI, microneutralization, and SRH results within each vaccine group and age cohort on days 1, 22, and 43, by using the Clopper-Pearson method. Postvaccination/prevaccination geometric mean ratios also were calculated. Seroprotection (HI titer of ≥40, SRH area of ≥25 mm2, or microneutralization titer of ≥40) rates and seroconversion rates from before vaccination to days 22 and 43, with 95% CIs, were calculated for each vaccine group and age cohort. To evaluate reactogenicity, numbers and proportions of subjects who experienced each event were calculated for each symptom and severity was graded. The sample size for subjects in the H5N1 vaccine groups was chosen to demonstrate the immunogenicity of the test vaccine on the basis of an expected seroprotection rate of ≥70% and a seroconversion rate of ≥40%. The study was not powered to detect significant differences between vaccines or groups, and no hypotheses were tested; therefore, immunogenicity and reactogenicity objectives were analyzed descriptively.
A total of 472 subjects were enrolled, including 335 in the MF59-adjuvanted H5N1 vaccine group and 137 in the seasonal influenza control vaccine group. Enrollment numbers, baseline characteristics, and subject disposition are shown in Table 1. The mean ages for subjects within each age cohort were similar in the H5N1 and control vaccine groups. Gender imbalances among children given seasonal vaccine and adolescents given H5N1 vaccine were judged unlikely to introduce any experimental bias. Two percent of toddlers, 6% to 13% of children, and 2% to 6% of adolescents had received influenza vaccinations in previous seasons. The withdrawal rate was very low in this study (0%–6% in the various groups). One adolescent subject withdrew voluntarily before receiving any vaccination; 334 subjects received the H5N1 vaccine and 137 received the control vaccine. Eight toddlers, 3 children, and 4 adolescents in the H5N1 group and 2 children and 1 adolescent in the control group withdrew voluntarily before completing the study. No subjects were lost to follow-up monitoring. No withdrawals occurred because of adverse events.
All subjects who received 1 dose of H5N1 vaccine and provided evaluable serum samples before and after vaccination were included in the immunogenicity analysis and constituted the full analysis set. The full analysis set included 314 (94%) of the 334 subjects given the H5N1 vaccine for the HI assay and 318 subjects for both the SRH and microneutralization assays. Differences in the numbers of subjects for some assays occurred because of insufficient amounts of some serum samples. Before the first dose of vaccine, anti-HI geometric mean titers were ∼1:5 in all age groups (Table 2); 1 toddler and 1 adolescent had seroprotective HI titers of ≥40. Three percent of toddlers and 4% of adolescents had seroprotective geometric mean areas of ≥25 mm2 (Table 3). Microneutralization titers of ≥40 were observed for 1% to 2% of subjects.
Three weeks after the first dose, HI seroconversion or fourfold titer increases were observed for 47% of adolescents, 39% of children, and 33% of toddlers (Table 2). The corresponding SRH responses were 57% for adolescents, 56% for children, and 44% for toddlers, and the microneutralization responses were 49% for adolescents, 34% for children, and 27% for toddlers (Tables 2 and 3). Three weeks after the second vaccination, seroprotective HI antibody titers were reached by 89% of adolescents and 97% of children and toddlers. With the SRH assay, 100% of the subjects had seroprotective antibody levels. Ninety-nine percent of the subjects in all age groups had microneutralization titers of ≥40 (Tables 2 and 3), and ≥98% had microneutralization titers of ≥80 (Table 4). The full analysis set results, which included data for subjects with partially missing data, showed no difference from the per-protocol analysis results, for which data were excluded only if the subject incurred a major protocol deviation or did not have complete immunogenicity data.
Reactogenicity and Safety
The reactogenicity profiles of the H5N1 vaccine and the seasonal control vaccine were similar. There was a slight increase in reactogenicity with increasing age. Fewer subjects experienced solicited reactions after the second dose than after the first dose (Fig 1). Most solicited local reactions occurred within 2 days after vaccination and were of short duration. The most-frequent local reactions in toddlers were erythema (21%–33%), tenderness (21%–29%), swelling (4%–12%), and induration (2%–11%), with <1% of any reactions being reported as severe (Fig 1A). The most-common local reactions among children were pain (36%–53%), erythema (26%–36%), and induration (6%–13%) (Fig 1B), and the most-common local reactions among adolescents were pain (63%–78%), erythema (16%–23%), and swelling (11%–18%) (Fig 1C). Up to 5% of pain reactions and up to 3% of other local reactions reported by children and adolescents were severe.
The majority of solicited systemic reactions in toddlers in both vaccine groups were mild or moderate, occurred within 3 days after vaccination, and resolved by day 7. For most reactions, occurrence rates were lower after the second vaccination than after the first. The most-frequently reported reactions in toddlers were irritability (20%–39%), unusual crying (16%–30%), and sleepiness (14%–23%) (Fig 1A). Fever of ≥38°C was reported for 7% of toddlers after the first dose of H5N1 vaccine and 4% after the second dose. The corresponding rates for toddlers given the seasonal influenza control vaccine were 12% and 9%. No cases of fever of ≥40°C were reported.
The most-frequent solicited systemic reactions in children were fatigue (15%–33%), headache (9%–23%), and myalgia (6%–21%) (Fig 1B). Fever of ≥38°C, which was reported for 2% to 5% of subjects, was less frequent in children than in toddlers, and no cases of fever of ≥40°C were reported. In adolescents, the most-frequent solicited systemic reactions were headache (22%–59%), myalgia (18%–37%), and fatigue (11%–49%) (Fig 1C). Fever of ≥38°C was reported for only 2% of adolescents after the first dose of control vaccine and 1% after the second dose of H5N1 vaccine. No cases of fever of ≥40°C were reported. For each solicited systemic symptom, severe reactions were reported by up to 2% of the H5N1 vaccine recipients and 5% of the control vaccine recipients (ie, by 1 or 2 subjects).
Unsolicited Adverse Events
Approximately 50% of subjects in the H5N1 vaccine group and 60% in the seasonal vaccine group experienced unsolicited adverse events during the entire study period; the most frequent were rhinitis, cough, otitis media, and pyrexia. Most cases were mild to moderate and were assessed as being unrelated to the vaccine. There were 2 SAEs. A case of renal injury in an 8-year-old subject was being treated at study end, and a case of acute pyelonephritis in an 11-year-old subject resolved after treatment. Both subjects required hospitalization, and neither case was assessed as being related to vaccination. There were no withdrawals because of adverse events. No deaths occurred during the study.
The immune responses to primary vaccination seen in this study compared favorably with the results of previous studies in adults using up to 90 μg of nonadjuvanted10 or 30 to 45 μg of aluminum-adjuvanted27,28 H5N1 vaccine. These results confirm, in a pediatric population, the antigen-sparing benefit for H5N1 influenza vaccines attributable to the MF59 adjuvant.29 The immune responses of this pediatric population exceeded earlier clinical trial data in which 81% of adults and 76% of elderly subjects had neutralizing antibody titers of ≥40; 73% of adults and 67% of elderly subjects had seroprotective HI titers of ≥40 after priming with two 7.5-μg doses of the same MF59-advuvanted vaccine formulation.30 A third dose of the adjuvanted H5N1 vaccine given 6 months after priming resulted in seroprotection of 83% of adults and 92% of elderly subjects.30 A similar follow-up study of antibody persistence and booster response was conducted recently with the participants in this study and will be reported separately.
Previous studies in adults with this MF59-adjuvanted H5N1 vaccine showed cross-reactivity with a number of drifted H5N1 influenza strains. Subjects who were primed with two 7.5-μg doses of the study vaccine and received a third dose 17 to 18 months later developed strong immune responses not only against the vaccine strain but also against a heterologous A/turkey/Turkey/05-like clade 2 strain after both the second and third doses.30 Protective cross-reactive titers against A/Indonesia/5/2005 (clade 2.1), A/Anhui/1/2005 (clade 2.3), and A/Turkey/15/2006 (clade 2.2) viruses developed within 7 days after a single dose of the study vaccine in subjects who had been primed 6 years earlier with 2 doses of a clade 0 A/duck/Singapore/97 (H5N3) vaccine.31 Studies of heterologous antibody responses are planned for the participants in the present pediatric study after administration of a booster dose.
There were no vaccine-related SAEs in either group and no withdrawals from the study because of an adverse event, which adds to the evidence that the MF59 adjuvant has a good safety profile in children. The overall safety of the MF59 adjuvant was documented in clinical trials that enrolled 219 000 participants, of whom 119 000 received ≥1 dose of a MF59-adjuvanted vaccine. Postmarketing surveillance of >45 million MF59-adjuvanted seasonal vaccine doses has not indicated an increase of rare events over the expected baseline levels.11,33,34 The experience with children is still limited, because the first clinical trials in young children began in only 2006.18 Currently, ∼2000 children have received a MF59-adjuvanted seasonal or H1N1 2009 pandemic influenza vaccine in clinical trials, with no unexpected SAEs being observed.18,35,–,37 However, the present study confirmed previous experience in Finland, in that local solicited reactions were common in children; most were mild to moderate in nature and resolved spontaneously within few days. Injection site pain was reported by the majority of adolescents. Solicited systemic reactions were mostly mild to moderate in nature, but some were graded as severe. Fever occurred in only 2% to 5% of subjects and occurred less frequently with increasing age, and no cases of high fever (≥40°C) were reported.
The results of this pediatric study, together with findings from earlier studies with H5N1 influenza vaccines that show cross-reactivity and suggest memory responses to booster doses even 6 to 7 years after heterologous priming,23 provide a scientific rationale for a strategy of prepandemic vaccination. Heterotypic immunity, or priming that induces immune memory with cross-protection, might allow single-dose “booster” vaccination to a pandemic strain in response to a pandemic situation.31,38 Prepandemic priming would help to overcome the problems of overstretched health care resources and logistic support to distribute vaccine quickly at the start of a pandemic, yielding efficacy of strain-matched pandemic vaccine after the first dose. Vaccination of children would indirectly protect high-risk populations and maximize the herd effect of vaccination by slowing influenza virus circulation through a group that plays an important role in the transmission of influenza in the community.
This study provides the first findings on the immunogenicity and safety of a MF59-adjuvanted H5N1 vaccine in a pediatric population. The immune responses to a 2-dose primary vaccination were strong in all age groups, including toddlers, and exceeded all Committee for Medicinal Products for Human Use criteria for all subjects, as measured with 3 different serological assays. Both the adjuvanted H5N1 vaccine and the adjuvanted seasonal influenza vaccine had good safety profiles but, as expected for adjuvanted vaccines, induced solicited local reactions in a majority of study participants. Local and systemic reactions after vaccination were mostly mild or moderate in severity and of short duration.
This trial was supported by Novartis Vaccines and Diagnostics.
- Accepted June 9, 2010.
- Address correspondence to Timo Vesikari, MD, PhD, University of Tampere Medical School, Vaccine Research Center, FM3, Biokatu 10, 33520 Tampere, Finland. E-mail:
FINANCIAL DISCLOSURE: Dr Vesikari has received consultancy fees from GSK, MedImmune and Novartis and speaker fees from MedImmune, Novartis, and Crucell for meetings on influenza vaccination. Drs Borkowski, Banzhoff, and Clemens and Ms Tilman are employed by Novartis Vaccines and Diagnostics.
- CI =
- confidence interval •
- HI =
- hemagglutination inhibition •
- SAE =
- serious adverse event •
- SRH =
- single radial hemolysis
- 1.↵World Health Organization. Avian influenza. Available at: http://www.who.int/csr/disease/avian_influenza/en. Accessed July 27, 2010
- 2.↵World Health Organization. Cumulative number of confirmed human cases of avian influenza A/(H5N1) reported to WHO, 23 April 2009. Available at: http://www.who.int/csr/disease/avian_influenza/country/cases_table_2009_04_23/en/index.html. Accessed May 4, 2009
- 3.↵World Health Organization. Avian Influenza: Assessing the Pandemic Threat. Geneva, Switzerland: World Health Organization; 2005. Publication WHO/CDS/2005.29. Available at: http://www.who.int/csr/disease/influenza/H5N1-9reduit.pdf. Accessed July 27, 2010
- Melidou A,
- Gioula G,
- Exindari M,
- Chatzidimitriou D,
- Diza-Mataftsi E
- 6.↵World Health Organization. WHO Strategic Action Plan for Pandemic Influenza 2006–2007. Geneva, Switzerland: World Health Organization; 2006. Publication WHO/CDS/EPR/GIP/2006.2. Available at: http://www.who.int/csr/resources/publications/influenza/WHO_CDS_EPR_GIP_2006_2c.pdf. Accessed July 27, 2010
- Iorio AM,
- Francisci D,
- Camilloni B,
- et al
- 19.↵American Academy of Pediatrics, Committee on Infectious Diseases. Prevention of influenza: recommendations for influenza immunization of children, 2008–2009. Pediatrics. 2008;122(5):1135–1141
- 21.↵US Department of Health and Human Services. Draft Guidance on Allocating and Targeting Pandemic Influenza Vaccine 2007. Washington, DC: US Department of Health and Human Services; 2007. Available at: www.flu.gov/individualfamily/vaccination/allocatinguidance.pdf. Accessed July 27, 2010
- Galli G,
- Hancock K,
- Hoschler K,
- et al
- 32.↵European Committee for Medicinal Products for Human Use. Note for Guidance on Harmonisation of Requirements for Influenza Vaccines. Brussels, Belgium: European Commission on Enterprise and Industry; 1997. Publication CPMP/BWP/214/96. Available at: www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2009/09/WC500003945.pdf. Accessed July 27, 2010
- Banzhoff A,
- Lattanzi M,
- Clemens R
- Black S,
- Vesikari T,
- Della Cioppa G,
- Pellegrini M,
- Nicolay U
- Copyright © 2010 by the American Academy of Pediatrics