Share this article on:

Immunogenicity, Safety and Reactogenicity of a Mammalian Cell-Culture–Derived Influenza Vaccine in Healthy Children and Adolescents Three to Seventeen Years of Age

Vesikari, Timo MD*; Block, Stan L. MD; Guerra, Fernando MD, MPH; Lattanzi, Maria MD, PhD§; Holmes, Sandra PhD, MHA; Izu, Allen MS; Gaitatzis, Nicolaos PhD; Hilbert, Anne Katrin PhD; Groth, Nicola MD§

The Pediatric Infectious Disease Journal: May 2012 - Volume 31 - Issue 5 - p 494–500
doi: 10.1097/INF.0b013e31824bb179
Vaccine Reports

Background: The safety and immunogenicity of the cell-culture–derived seasonal trivalent influenza vaccine ([CCIV]; Optaflu) has been reported previously in adults and the elderly. In this study, we compared the safety, reactogenicity and immunogenicity of CCIV with a conventional egg-derived trivalent influenza vaccine (TIV) in a healthy pediatric population.

Methods: A total of 3604 subjects were randomized to receive 2 doses of CCIV or TIV (3–8 years, n = 2630) at a 28-day interval or a single vaccination (9–17 years, n = 974). Antibody levels on days 1, 29 and 50 were measured by hemaglutination inhibition assay using egg-derived and cell-derived test antigens. Adverse reactions were solicited via memory aids for 7 days after each injection, and unsolicited adverse events/serious adverse events were collected for 6 months postvaccination.

Results: Noninferiority of CCIV versus TIV was demonstrated for most immunogenicity measures, particularly by using cell-derived antigen in the hemaglutination inhibition assay. In 3- to 8-year-olds (the primary objective), both CCIV and TIV met all 3 Committee for Medicinal Products for Human Use immunogenicity criteria for A/H1N1 and A/H3N2 strains. Lower immune responses were observed against the B strain, fulfilling Committee for Medicinal Products for Human Use criteria only for geometric mean ratio (TIV, CCIV) and seroconversion rate (TIV, CCIV [cell-derived antigen]). Both CCIV and TIV were safe and well tolerated, with no differences in local and systemic solicited reactions or in unsolicited adverse events/serious adverse events.

Conclusion: CCIV produced in mammalian cell culture is a safe, well-tolerated and immunogenic alternative to conventional egg-derived influenza vaccines for children and adolescents.

*University of Tampere Medical School, Tampere, Finland

Kentucky Pediatric/Adult Research, Bardstown, KY

San Antonio Metropolitan Health District Immunization, San Antonio, TX

§Novartis Vaccines and Diagnostics, Siena, Italy

Novartis Vaccines and Diagnostics, Cambridge, MA

Novartis Vaccines and Diagnostics, Marburg, Germany

Accepted for publication January 17, 2012.

This project has been funded in part with Federal funds from the Office of Public Health Emergency Preparedness, Office of Research and Development Coordination, under Contract No: HHSO100200600012C to Novartis Vaccines and Diagnostics, Inc. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products or organizations imply endorsement by the US Government.

M.L., S.H., A.I., N.Ga., A.K.H. and N.Gr. are Novartis employees. S.L.B., F.G. and T.V. received a research grant from Novartis Vaccines and Diagnostics. The authors have no other funding or conflicts of interest to disclose. Identifier:NCT00645411.

Address for correspondence: Timo Vesikari, MD, University of Tampere Medical School, Biokatu 10, 33520 Tampere, Finland. E-mail:

Influenza leads to substantial morbidity and mortality in children.1–4 Young children also play a major role in the transmission and spread of influenza in households and the community during epidemics.1,5,6 Vaccination of children against influenza may bring health benefits to the children themselves and also to persons in other age groups, by helping to control the spread of influenza.1,6–8

Conventional seasonal influenza vaccines are produced in embryonated hens’ eggs, a 50-year-old method with inherent limitations, with respect to contamination and relative inflexibility in production. Recently, the declining fraction of H3N2 viruses that can be isolated in eggs has threatened to compromise the selection of the most appropriate strains for inclusion in seasonal vaccines.9 Cell lines remain sensitive and their use also provide advantages in scalability, sterility and fewer excipients, for example, thimerosal-free, antibiotic-free, gelatin-free and formalin-free production.10,11 Cell-culture–derived antigens have a hypothetical advantage of providing a more structurally and antigenically faithful antigen.12–15 It is obvious that allergy to eggs is not an issue with cell-derived vaccines.

Madin-Darby Canine Kidney (MDCK) cells are widely used in clinical diagnosis of influenza and research. Recent studies, including a large phase III clinical efficacy trial in adults,16 demonstrated that a licensed MDCK cell-culture–derived seasonal trivalent influenza vaccine (CCIV) was comparable with the conventional egg-derived vaccine with respect to safety, reactogenicity, immunogenicity and vaccine efficacy in adult and elderly subjects.16–19 We report the results of a combined phase II/III trial to assess the safety, reactogenicity and immunogenicity of that CCIV (Optaflu, Novartis Vaccines and Diagnostics, Marburg, Germany), compared with a conventional US-licensed egg-derived subunit trivalent influenza vaccine (TIV) (Fluvirin, Novartis Vaccines and Diagnostics, Liverpool, UK) in a healthy pediatric population 3–17 years of age.

Back to Top | Article Outline


Study Design

This combined phase II/III, observer-blind, randomized study was conducted between October 2007 and July 2008 at multiple study sites in the United States (16 sites) and Europe (Finland [14 sites], Croatia [9 sites], Hungary [8 sites], Lithuania [6 sites], Italy [5 sites] and Romania [2 sites]). The study was conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki and was approved by institutional review boards or ethics committees in each area. Written informed consent was obtained from parents or legal guardians and the participant, if applicable.

Two investigational subunit influenza study vaccines were used in the study, a novel CCIV produced in MDKC cells (Optaflu), and a US-licensed conventional egg-derived influenza vaccine (Fluvirin). Optaflu was approved by the European Medicines Agency in June 2007 for individuals ≥18 years of age. Fluvirin was approved in the United States in 1988 and is indicated for persons ≥4 years of age. Each vaccine contained purified viral neuraminidase and hemagglutinin antigens (HA) with 15 µg of HA for each strain (A/Solomon Islands/3/2006[H1N1]-like, A/Wisconsin/67/2005[H3N2]-like and B/Malaysia/2506/2004-like), recommended for the 2007 to 2008 influenza season in the northern hemisphere. Vaccination consisted of 0.5-mL doses of CCIV or TIV; 1 dose for cohorts 1 and 2, and 2 doses administered 4 weeks apart for cohort 3. Vaccines were administered intramuscularly, preferably in the deltoid muscle of the nondominant arm, by an unblinded study vaccine administrator.

The primary objective of the study was to demonstrate noninferior immunogenicity of CCIV compared with a licensed conventional egg-derived influenza vaccine in children 3–8 years of age in terms of postvaccination (day 50) geometric mean titers (GMTs) and seroconversion according to the Center for Biologics Evaluation and Research criteria. The 3–8 years of age group was considered more challenging for the noninferiority comparison, and therefore a larger sample was enrolled than for the 9–17 years of age group. A secondary immunogenicity objective was to evaluate the immunogenicity of the study vaccines according to the Committee for Medicinal Products for Human Use (CHMP) immunogenicity criteria.20 Safety and tolerability objectives included the evaluation of solicited local and systemic reactogenicity and spontaneously reported adverse events (AEs) and serious adverse events (SAEs).

Main exclusion criteria included history of any serious disease, history of adverse reactions to vaccine components, impaired or altered immune function, history of Guillain-Barré syndrome, history of bleeding diathesis, planned surgery during the study period, laboratory-confirmed influenza within 6 months before vaccination, receipt of an influenza vaccine within 6 months before study start, routine childhood vaccines within 2 weeks (inactivated vaccines) or 4 weeks (live vaccines) before vaccination and fever (≥38°C) or any acute illness within 3 days before study vaccination. Children 3–8 years of age who previously received 2 doses of an influenza vaccine in the same influenza season were excluded from enrollment.

A total of 3604 healthy children and adolescents were enrolled in this study. The 3 study cohorts are illustrated in Figure 1. It is important to note that the 3–8 years of age group included more subjects than the 9–17 years of age group.



Randomization was performed using an interactive voice response system, which was accessed by designated unblinded study staff at each site. An unblinded staff member prepared and administered the vaccines and had no further contact with subjects or access to the data. The investigator and all other study staff were blinded throughout the trial.

Back to Top | Article Outline

Immunogenicity Assessment

Blood samples were obtained at day 1 (prevaccination) and day 29 for subjects in cohort 1, and immediately before each vaccination (days 1 and 29) and at day 50 for subjects in the immunogenicity subset from cohort 3. Hemagglutination inhibition (HI) assay using both egg-derived and cell-derived HA antigens in the vaccine were performed at the Novartis Vaccines Clinical Serology Laboratory in Marburg, Germany, according to standard methods.21 HI antibody responses were expressed as geometric mean titers (GMTs) and geometric mean ratios (GMRs) of the postvaccination to prevaccination titer (day 29/day 1 titer and day 50/day 1 titer); seroprotection rates were defined as the percentage of subjects with HI titers ≥40, and seroconversion rates were defined as the percentage of subjects per group achieving at least a 4-fold increase in HI titer from a seropositive prevaccination titer (≥10) or a rise from <10 to ≥40 in those who were originally seronegative.

Back to Top | Article Outline

Safety and Tolerability Assessment

The frequency and severity of all solicited and unsolicited events and the use of nonprescription medication (with the exception of vitamins and minerals) were collected for 7 days after vaccination. Solicited local reactions included pain, ecchymosis, erythema, induration and swelling. Solicited systemic reactions included chills, malaise, myalgia, arthralgia, headache, sweating, fatigue and fever (axillary temperature ≥38°C, severe ≥40°C). Diameter measurements of local reactions (ecchymosis, erythema, swelling and induration) were categorized as none, 1 to ≤10 mm, 11 to ≤25 mm, 26 to ≤50 mm, 51 to ≤100 mm and >100 mm (severe).

Reports of all unsolicited AEs and SAEs and associated medications were collected for the 29-day (cohorts 1 and 2) or the 50-day study period (cohort 3). Safety analyses up to study termination (days 29–181: cohorts 1 and 2; days 50–202: cohort 3) included the collection of SAEs and AEs leading to study withdrawal and concomitant medications associated with these events. The severity of AEs was categorized as mild, moderate or severe, if they resulted in no limitation, some limitation or inability to perform normal daily activities, respectively. Assessments of the causal relationship of unsolicited AEs to the vaccination were classified by the investigator as not related, possibly related or probably related.

Back to Top | Article Outline

Statistical Analysis

Immunogenicity analyses were run on the per-protocol set, which included all enrolled subjects who received the vaccine correctly, provided evaluable serum samples at relevant time points and had no major protocol deviations. Safety was analyzed for all subjects exposed to study vaccines. A sample size of at least 600 evaluable subjects per vaccine group was estimated to provide sufficient power for the primary study objective, to demonstrate noninferiority of CCIV to egg-derived subunit influenza vaccine in children 3–8 years of age (cohort 3, immunogenicity subset). Specifically, the Center for Biologics Evaluation and Research criteria for noninferiority for each of the 3 strains was defined as the lower limits of the 95% confidence intervals for the ratio (CCIV:TIV) of day 50 GMTs >0.667 and the difference (CCIV – TIV) in the percentage of subjects achieving seroconversion >−10%, according to the Miettinen-Nurminen method.22

Immunogenicity data also were evaluated according to the HI licensure criteria established by the CHMP.20 As there are no specific criteria for immunogenicity established for children, standards for persons 18–60 years of age were applied: (i) the percentage with seroconversion or at least 4-fold increase in HI antibody is >40%; (ii) the percentage achieving seroprotection, or an HI titer ≥40, is >70%; and (iii) the GMR is >2.5. Log10-transformed antibody titers were modeled using analysis of variance for each strain, and GMTs, their ratios, percentages reflecting the immunogenicity endpoints and corresponding 2-sided 95% confidence intervals were calculated.

Safety and tolerability data were summarized by vaccine group providing the number and percentage of subjects reporting an event.

Back to Top | Article Outline



The disposition of study subjects is summarized in Figure 1. Among the 9- to 17-year-olds in cohort 1, 96% and 97% of subjects in the CCIV and TIV groups, respectively, completed the study and in cohort 2, 99% of subjects within both vaccine groups completed the study. Among the 3- to 8-year-olds, 91% and 90% of subjects in the CCIV and TIV groups, respectively, completed the study. Baseline characteristics of enrolled subjects per age and vaccine group were similar (data not shown).

Back to Top | Article Outline


In children 3–8 years of age, CCIV was noninferior to TIV for 5 of the 6 immunogenicity measures using cell-derived HA antigens in the HI assay () Fig. 2. For the A/H3N2 strain, responses to CCIV were noninferior to TIV as measured by the percentage of subjects achieving seroconversion, but not for GMTs at day 50. Using egg-derived antigen in the HI assay, postvaccination responses of CCIV versus TIV were noninferior for 2 of the 6 measures ( Fig. 2). Noninferiority of CCIV to TIV was demonstrated for day 50 GMTs and for the percentage of subjects achieving seroconversion to the A/H1N1 strain, but not to the A/H3N2 and B strains.



Evaluation of seroprotection, seroconversion rates and GMRs for A/H1N1, A/H3N2 and B strains, according to CHMP licensure criteria across age and vaccine groups are presented in Table 1. For children 3–8 years of age, strong immune responses to both vaccines were observed for the A strains, and all 3 CHMP criteria were fulfilled at days 29 and 50 when assessed with cell-derived antigen. For the B strain, lower responses to both vaccines were observed, and only 1 of 3 (GMR), and 2 of 3 criteria (GMR, seroconversion) were met for CCIV at days 29 and 50, respectively, and 2 of 3 criteria (GMR, seroconversion) were met for TIV at days 29 and 50, when assessed by the HI assay using cell-derived antigen.



In children and adolescents 9–17 years of age, strong antibody responses to both vaccines met all 3 CHMP criteria for all strains, using both cell-derived and egg-derived antigens at day 29 postvaccination.

Back to Top | Article Outline

Safety and Tolerability

In the 3–8 years of age cohort, any local reactions were reported in 38% (CCIV) and 35% (TIV), and 35% (CCIV) and 34% (TIV), after first and second vaccinations, respectively. Any systemic reactions were reported in 23% (CCIV) and 26% (TIV), and 15% (CCIV) and 19% (TIV), after first and second injections, respectively. In the 9–17 years of age cohort, 42% (CCIV) and 45% (TIV) of subjects reported any local reaction and 29% (CCIV) and 30% (TIV) of subjects reported any systemic reaction after 1 vaccination (Figs. 3 and 4). Severe local (pain only) and severe systemic (chills, malaise, myalgia, headache, fatigue, sweating and arthralgia) solicited reactions were reported rarely, with only ≤1% of any reaction classified as severe, equally divided across age and vaccine groups.





Spontaneously reported AEs were reported in 32% (CCIV) to 34% (TIV) and 18% (CCIV) to 20% (TIV) in children 3–8 years of age after the first and second vaccinations, respectively, of which 5–8% were considered at least possibly related to vaccine. In the age cohort 9–17 years, 19–20% of subjects reported unsolicited AEs, of which 3% were considered at least possibly related. During the follow-up period, spontaneously reported AEs occurred in 1–4% across age and vaccine groups, of which 0% to <1% were considered at least possibly related to vaccine.

Throughout the study period, a total of 28 SAEs (4 during postvaccination period, 24 during safety follow-up period) were observed, none of which was considered to be related to the study vaccines.

Back to Top | Article Outline


This study comparing cell-culture and egg-derived influenza vaccines demonstrated that, in 9- to 17-year-olds, the CCIV met all European criteria. In the 3- to 8-year-olds, which was the primary objective, CCIV was noninferior to TIV for A/(H1N1) and for the B strain, while for the A(H3N2) strain noninferiorly was demonstrated for seroconversion only. The CCIV met all 3 regulatory criteria for the A(H1N1) and A(H3N2) strains, but, in line with previous studies, responses to influenza B virus were lower for both CCIV and TIV, failing the seroprotection criterion. Somewhat better responses were seen when the source of antigen in the HI test was itself cell-derived than when egg-grown virus was used, but irrespective of the origin of the antigen, each of the virus strains in CCIV and TIV induced responses that fulfilled the overall criteria as set by the CHMP, which includes seroprotection rate, seroconversion rate and a rise in GMTs.

Previously, trends toward higher titers using cell-derived antigen versus egg-derived antigen in CCIV recipients have been detected but not to the extent seen here.10,18,19 It is likely that the adaptation of the World Health Organization (WHO)-specified reference viruses to cells to produce CCIV on the one hand, and to eggs for egg-derived vaccines on the other, may have led to changes in the respective hemagglutinins that are detectable in the HI reactions of CCIV recipients who use the respective antigens. This phenomenon of viral adaptation to host cells is well established and is a basis for speculating that a fully mammalian cell-culture–derived vaccine could be a better antigenic match to the wild-type virus (derived from humans) than the egg-based avian cell-derived vaccine.23 In the current WHO system of vaccine virus derivation, all clinical isolates are passaged through eggs, leading potentially to egg-adaptive changes in master seed viruses that are used in vaccine production—whether that be in cells or in eggs.24 Therefore, the differences seen here between HI tests using cell-derived and fully egg-derived antigens could be even greater with a fully cell-derived antigen. Changes to the WHO system to allow for direct isolation of clinical isolates in cells have been proposed, and fully cell-derived viruses for cell-culture manufacturing could be available within a few years. That possibility could allow for observations in humans to finally compare the clinical effectiveness of fully mammalian cell-derived and egg-derived vaccines.

More important is the trend toward a reduced sensitivity of eggs for the primary isolation of clinical isolates, particularly of H3N2 viral strains. Currently, as few as 5% of H3N2 viruses submitted to WHO collaborating centers can be isolated in eggs, severely limiting the optimal choice of strains that can be included in the recommended vaccine formulation. In contrast, 80–90% of H3N2 strains can be isolated in MDCK cells.9 Between 2003 and 2004, the unavailability of an egg-derived H3N2 strain that best matched the emerging circulating strain forced the WHO to delay their recommendation and, finally, to select an alternate suboptimal strain, resulting in a significant mismatch between the vaccine and circulating strains in the subsequent season in an outbreak causing significant morbidity and mortality.25 The introduction of certified cells into the process for primary isolation of influenza viruses from clinical specimens should help avoid such frustrations and public health consequences in the future.

The results of the present study covering 6 months of follow-up postvaccination demonstrate that both CCIV and TIV were safe and well tolerated in a healthy population 3–17 years of age, with no differences in local and systemic solicited reactions or in spontaneously reported AEs and SAEs between cell-derived and egg-derived influenza vaccines in either frequency or severity. The results obtained in this study of a pediatric population are consistent with previous trials of the MDCK cell-derived influenza vaccine, Optaflu, in adult and elderly populations.10,16,18,19 In addition, with respect to clinical practice, the absence of egg-derived proteins in the vaccine provides a convenience in immunizing egg-allergic children. This cell-culture–derived vaccine also is produced without preservatives and antibiotics, as the cell-culture production process is contained and more easily controlled for external contamination than the egg process, which is open to the environment. Improving the speed of vaccine production, which is essential to an effective response to a pandemic, is being tackled at each of the various steps between the initial identification of the pandemic strain and vaccine production.24 A promising approach is the synthesis of viral genes from their sequences and their introduction into master viruses, developed to produce high yields of hemagglutinin in the manufacturing cell line. The use of such modern cell-culture–based technology for the production of influenza vaccines promises an increase in speed and other important advantages over conventional egg-based vaccines for production of both seasonal and pandemic influenza vaccines.

Back to Top | Article Outline


The authors are grateful to the children and parents/guardians for participating in this trial. The authors acknowledge all investigators for their invaluable participation in the study, and they thank the clinical and serologic teams of Novartis Vaccines and Diagnostics for their contribution to this trial. The authors thank Patricia de Groot, PhD (CHC Europe) and Pinki Rajeev, PhD (Novartis Vaccines and Diagnostics) for their writing assistance in the preparation of the manuscript.

Back to Top | Article Outline


1. Heikkinen T, Silvennoinen H, Peltola V, et al. Burden of influenza in children in the community. J Infect Dis. 2004;190:1369–1373
2. Poehling KA, Edwards KM, Weinberg GA, et al. The underrecognized burden of influenza in young children. N Engl J Med. 2006;355:31–40
3. Thompson WW, Shay DK, Weintraub E, et al. Influenza-associated hospitalizations in the United States. JAMA. 2004;292:1333–1340
4. Thompson WW, Shay DK, Weintraub E, et al. Mortality associated with influenza and respiratory syncytial virus in the United States. JAMA. 2003;289:179–186
5. World Health Organization.. Influenza vaccinesWHO position paper. Wkly Epidemiol Rec.. 2005;33:279–287
6. Reichert TA, Sugaya N, Fedson DS, et al. The Japanese experience with vaccinating schoolchildren against influenza. N Engl J Med. 2001;344:889–896
7. Basta NE, Chao DL, Halloran ME, et al. Strategies for pandemic and seasonal influenza vaccination of schoolchildren in the United States. Am J Epidemiol. 2009;170:679–686
8. Jordan R, Connock M, Albon E, et al. Universal vaccination of children against influenza: are there indirect benefits to the community? A systematic review of the evidence. Vaccine. 2006;24:1047–1062
9. Ampofo WK, Baylor N, Cobey S, et al. Improving influenza vaccine virus selectionReport of a WHO informal consultation held at WHO headquarters, Geneva, Switzerland, 14–16 June 2010. Influenza Other Respi Viruses.. 2012;6:142–152
10. Ambrozaitis A, Groth N, Bugarini R, et al. A novel mammalian cell-culture technique for consistent production of a well-tolerated and immunogenic trivalent subunit influenza vaccine. Vaccine. 2009;27:6022–6029
11. Barrett PN, Mundt W, Kistner O, et al. Vero cell platform in vaccine production: moving towards cell culture-based viral vaccines. Expert Rev Vaccines. 2009;8:607–618
12. Neuzil KM, Bright RA. Influenza vaccine manufacture: keeping up with change. J Infect Dis. 2009;200:835–837
13. Ulmer JB, Valley U, Rappuoli R. Vaccine manufacturing: challenges and solutions. Nat Biotechnol. 2006;24:1377–1383
14. Govorkova EA, Kodihalli S, Alymova IV, et al. Growth and immunogenicity of influenza viruses cultivated in Vero or MDCK cells and in embryonated chicken eggs. Dev Biol Stand. 1999;98:39–51 discussion 73.
15. Kessler N, Thomas-Roche G, Gérentes L, et al. Suitability of MDCK cells grown in a serum-free medium for influenza virus production. Dev Biol Stand. 1999;98:13–21 discussion 73
16. Reisinger KS, Block SL, Izu A, et al. Subunit influenza vaccines produced from cell culture or in embryonated chicken eggs: comparison of safety, reactogenicity, and immunogenicity. J Infect Dis. 2009;200:849–857
17. Frey S, Vesikari T, Szymczakiewicz-Multanowska A, et al. Clinical efficacy of cell culture–derived and egg-derived inactivated subunit influenza vaccines in healthy adults. Clin Infect Dis. 2010;51:997–1004
18. Groth N, Montomoli E, Gentile C, et al. Safety, tolerability and immunogenicity of a mammalian cell-culture-derived influenza vaccine: a sequential Phase I and Phase II clinical trial. Vaccine. 2009;27:786–791
19. Szymczakiewicz-Multanowska A, Groth N, Bugarini R, et al. Safety and immunogenicity of a novel influenza subunit vaccine produced in mammalian cell culture. J Infect Dis. 2009;200:841–848
20. The Committee for Proprietary Medicinal Products (CHMP). Note for guidance on hamonisation of requirements for influenza vaccines. CHMP/BWP/214/96. Available at Accessed September 21, 2011.
21. Rowe T, Abernathy RA, Hu-Primmer J, et al. Detection of antibody to avian influenza A (H5N1) virus in human serum by using a combination of serologic assays. J Clin Microbiol. 1999;37:937–943
22. Miettinen O, Nurminen M. Comparative analysis of two rates. Stat Med.. 1985;4:213–226
23. Robertson JS. An overview of host cell selection. Dev Biol Stand. 1999;98:7–11 discussion 73
24. Minor PD. Vaccines against seasonal and pandemic influenza and the implications of changes in substrates for virus production. Clin Infect Dis. 2010;50:560–565
25. World Health Organization. Recommendations for influenza vaccine composition. February 28, 2003. Available at Accessed September 21, 2011

vaccine; cell culture; immunogenicity; safety; children

© 2012 Lippincott Williams & Wilkins, Inc.