Impact of the antipneumococcal conjugate vaccine on the occurrence of infectious respiratory diseases and hospitalization rates in children

Introduction: In 2010, to reduce the occurrence of serious pneumococcal disease, the Ministry of Health in Brazil incorporated the 10-valent pneumococcal vaccine in the immunization schedule of children younger than two years of age. The objective of this study was to evaluate the impact of vaccination on the incidence of infectious respiratory diseases in infants before and after the introduction of the 10-valent pneumococcal vaccine. Methods: This cross-sectional study involved primary care and hospital networks from a city in Minas Gerais State, Brazil, between 2009 and 2012. Results: A 40% reduction in the prevalence of community-acquired pneumonia (CAP) was observed after introducing the pneumococcal conjugate vaccine. Male children were 28% more likely to develop the disease. The prevalence ratio ([PR] = 1.96, 95% CI: 1.52 to 2.53, p < 0.05) suggested that not being vaccinated was associated with the occurrence of pneumonia. The prevalence of CAP was 70% lower (PR 0.30, 95% CI: 0.24 to 0.37, p<0.05) in children vaccinated as recommended compared to children with delayed vaccination, suggesting that the updated vaccine schedule improves protection. Conclusions: Immunization with the 10-valent pneumococcal vaccine appeared to reduce the number of pneumonia cases in children during the study period. Prospective studies are needed to confi rm the effi cacy of the vaccine against the occurrence of pneumococcal pneumonia.

Since the early twentieth century, research has sought to contain and prevent respiratory diseases such as pneumonia, otitis media, sinusitis, and meningitis caused by Streptococcus pneumoniae.In 2010, the 10-valent pneumococcal conjugate vaccine (10VPCV) was introduced to the Brazilian basic immunization calendar to reduce morbidity and mortality in children under fi ve years of age; pneumococcal diseases, primarily community-acquired pneumonia (CAP), signifi cantly impact morbidity in this age group (1) (2) (3) .
According to a report by Rodrigues (4) , Brazilian CAP-related deaths in children younger than one year of age and between one and four years of age decreased by 74% and 56%, respectively, between 1991 and 2007.Of an estimated 120 million CAP episodes in children worldwide in 2011, 14 million progressed to severe disease, with 1.3 million deaths, 81% in children younger than two years of age (5) (6) .A recent publication by Black (7) stated that pneumonia remains the most commonly experienced complication, accounting for 18% (1.5 million) of the estimated 8.7 million deaths in children.However, the proportion of deaths attributed to CAP varies widely among regions with different health systems (8) .
Pneumococci can lead to asymptomatic infections of the respiratory tract and cause invasive diseases, such as otitis, sinusitis, meningitis, and pneumonia.Global pneumococcal resistance to antibiotics is approximately 19%, according to the international criteria of the Clinical and Laboratory Standards Institute (CLSI) (9) .In Brazil, the reported resistance rate ranges from 21.2% to 55% for children 8-71 months of age and 19% to 43% in the general population (10) , with a reported pneumococcal colonization rate of 21% and 35.8% in children in States of São Paulo and Goiás, respectively (11) (12) .Other publications have reported colonization rates of 40% in children younger than 2 years old, with highest prevalence during the fi rst year of life (10) (11) (13)   .The increased resistance of airway-colonizing pathogens due to indiscriminate antibiotic use affects the entire community and complicates treatment (14) .Of 91 identifi ed S. pneumoniae serotypes, 10 are responsible for 60-80% of disease, depending on the geographical region.

METHODS
Prevention is the most effective method of reducing pneumococcal disease, and immunization reduces bacterial dissemination.Vaccination is one of a set of procedures offered by primary health care services because health is an unquestionable right of citizenship (15) .Due to the good seroconversion results achieved with the vaccine, the national immunization program (NIP) has implemented the 10VPCV as part of the immunization schedule for children younger than two years old.
This study quantitatively evaluated children up to 24 months of age diagnosed with infectious diseases of the upper and lower airways in a small town before and after introduction of the 10VPCV by the NIP.This cross-sectional study was conducted in the primary care system of Guaranésia and in the hospitals of Guaranésia and Guaxupé, State of Minas Gerais, in southern Brazil.Data were collected from hospitalized children and from children treated as outpatients for respiratory diseases.
For hospitalized children, a survey of the charts of children 0-24 months of age was conducted from 2009 to 2012 in Santa Casa, Guaranésia, and Santa Casa, Guaxupé; the target population was all children in this age group who lived in the city during that time period.For children who were treated on an outpatient basis, the data and procedures performed during primary care visits were collected only in the municipality of Guaranésia.A survey was conducted of consultations between 2009-2012, with fi lters for patient age, attending physician, ambulatory care unit, and family health care unit.A records search was then conducted to identify pneumonia, acute otitis media, and sinusitis clinical diagnoses.Data recorded using the International Classifi cation of Diseases (ICD) diagnosis codes of A40.3, J13, J15.8, J15.9, J18.9, or J18.0 were analyzed; all other admissions were discarded.Patient name, age, sex, diagnosis, address, and affi liation were collected from these records.
Records of children 0-24 months of age with comorbidities related to cardiac diseases, central nervous system diseases, birth defects, gastroesophageal refl ux, or aspiration pneumonia were excluded.Using diagnostic confi rmation and patient information, each child's vaccine status was checked at the time of admission and after consulting the records stored in consultation room fi les.

Statistical analysis
To analyze the association between variables, an association estimator, the prevalence ratio (PR), was used with a 95% confidence interval.To evaluate the differences between central tendency measurements, Student's t test was used for independent samples (averages), and the Wilcoxon test was used for medians.In the analysis of difference between proportions, the chi-squared test was used.All calculations used a 5% signifi cance level.A sample size calculation was not necessary because we included all vaccinated individuals in the reported period.

Ethical considerations
The research project was submitted to the Ethics Committee of the University Research Center at the Educational Guaxupé Foundation [Universidade da Fundação Educacional Guaxupe (UNIFEG)] and approved at its 54 th regular meeting held on September 18, 2012 under Protocol 258.
This study compared the hospitalization rates and outpatient diagnoses of CAP, acute otitis media (AOM), and sinusitis in children in a city within Minas Gerais (MG) in 2009-2012.The comparison evaluated differences in cases before and after introduction of the 10VPCV to the NPI calendar recommended by the Ministry of Health.A total of 377 records (198 boys and 179 girls) were evaluated, including hospitalized children or children treated on an outpatient basis for one of three respiratory diseases.The general characteristics of the children included in this study are presented in Table 1.
The median ages of boys and girls were 12 and 13 months, respectively.Among children diagnosed with CAP, the median age was nine months, which was signifi cantly younger than those children who did not have CAP (14 months, p<0.05).The median age of children diagnosed with otitis media was 13 months, compared to a median age of 11.5 months for those who did not present with the disease (p<0.05).Children with sinusitis had a median age of 18.5 months, compared to 11 months for those without sinusitis (p<0.05).
We observed a greater incidence of CAP among boys (Table 3) when comparing disease occurrence before (2009-2010) and after (2011-2012) introduction of the 10VPCV.Before introduction of the vaccine, boys displayed a 28% higher CAP prevalence than girls (PR=1.28;95% CI: 1.02 to 1.60; p<0.05).
An analysis of AOM cases after vaccine introduction revealed a decrease in the number of cases of the disease in 2011; however, there was a marked increase in cases in 2012.Our data indicate an increase in otitis media cases after the introduction of the 10VPCV (Table 4) and an association between vaccination and disease occurrence.The percentage of female children diagnosed with otitis media was higher than that of males in 2009 and 2010.After introducing immunobiological data, we observed an equal trend in otitis media susceptibility, indicating no gender-based prevalence.
A high number of outpatient sinusitis cases were observed in 2009 and 2010 among children 12-24 months of age.After introduction of the vaccine in 2011, we observed a decrease in the number of outpatient cases.However, in 2012, the 12    Table 5 shows a comparison of children who received the vaccination as recommended and children with delayed vaccinations revealed a prevalence of CAP among children with correctly timed vaccination (PR=0.30;95% CI: 0.24 to 0.37; p<0.05).
However, the comparison of the incidence of otitis media between non-vaccinated and vaccinated individuals yielded a PR of 0.47 (95% CI: 0.35 to 0.64; p<0.05).
Brook and Gober (16) observed a reduction in oropharynx and nasal colonization by S. pneumoniae in children after the introduction of pneumococcal conjugate vaccines, with a relative increase in the presence of nontypeable Haemophilus infl uenza; the reported cases of sinusitis may have been caused by these pathogens.A surveillance study conducted by Pelton et al. (17) from 2000-2003 in two Massachusetts communities revealed that colonization by vaccine serotypes decreased from 22% to 2%.Meanwhile, the prevalence of several vaccine serotypes increased from 7% to 16%.Veenhoven et al. (18) noted that there was no reduction in the number of otitis media episodes in a group of individuals who received the pneumococcal conjugate vaccine and that, while vaccine serotypes were reduced in the nasopharynx after vaccination, there was immediate and complete replacement by non-vaccine serotypes.Félix (19) reported that the heptavalent pneumococcal vaccine did not cause a signifi cant reduction in the overall incidence of otitis media and described microbiological profi le changes.These fi ndings confi rm those of Block (20) , who reported an increased incidence of otitis media caused H. infl uenzae and Moraxella catarrhalis during the post-vaccine period accompanied by reduced pneumococcal disease.A study conducted in Norway reported that after introducing the 7-valent pneumococcal conjugate vaccine (7VPCV), the prevalence of nasopharyngeal colonization in children remained unchanged after widespread vaccination (77.7% versus 80.2%).This fi nding may explain the steady incidence levels of sinusitis among vaccinated individuals in this study.The decrease in 7VPCV serotypes among carriers was completely compensated by an increase in non-7VPCV serotypes, similar to reports in other locations 21 .
The evaluation of hospitalization records and outpatient care visits during the 10VPCV post-introduction period revealed a 40% reduction in CAP prevalence, a signifi cant decrease (PR: 0.60; 95% CI: 0.46 to 0.78; p<0.05) that suggests the effectiveness of 10VPCV in reducing CAP prevalence in the period after introduction of the vaccine (2011-2012).According to Tsai (22) , after introducing the vaccine, the US hospitalization rate for pneumonia decreased by 66% among children younger than two years of age, while in Brazil, there was a 31% decrease in the number of pneumonia cases with viral isolation (23) (24) .
Seasonal incidence changes were detected in this study, with increased hospitalizations occurring between April and June.While the number of occurrences dropped after introducing the 10VPCV, hospitalization rates remained higher in this period compared to other times of the year.There were no major differences in the seasonal trend between the vaccinated and unvaccinated time periods.Seasonal variation is a relevant factor that may lead to serious consequences and a higher number of deaths, particularly in developing countries (25) (26) .In 2013, Tamayo and Bastard (27) reported a slight predominance of respiratory diseases in children during the winter months, a fi nding that aligns with the data in this study.
Our data also indicate an increased number of otitis media cases after introduction of the 10VPCV, an observation that was also reported by Block et al. (20) .This finding can be explained by replacement by H. infl uenzae and M. catarrhalis during the post-vaccine period, accompanied by a reduction in pneumococcal otitis media.The higher prevalence in other studies (28) (29) in children between six and 24 months of age confl icts with the fi ndings of this study.In 2009 and 2010, there was outpatient care for children with otitis media in all age groups, predominantly among those 12-24 months of age.After introducing the vaccine in 2011, there was a decrease (30) in care required for children younger than six months of age; however, the prevalence remained highest among children between 12 and 24 months of age and increased in other strata compared to previous years.Felix (19) reported that the 7VPCV did not cause a signifi cant reduction in the overall incidence of otitis media but that the microbiological profi le of this disease changed.
Statistical analysis suggested that non-vaccination is a risk factor for CAP (PR=1.96;95% CI: 1.52 to 2.53; p<0.05).A comparison of CAP prevalence in children vaccinated as recommended and children with delayed vaccinations revealed a PR of 0.30 (95% CI: 0.24 to 0.37; p<0.05) among children with correctly timed vaccinations, indicating that updated vaccinations may be a protective factor against disease occurrence.Similar comparisons of AOM among unvaccinated and vaccinated children confi rmed that the pneumococcal vaccine protects against AOM (PR=0.47,95% CI: 0.35 to 0.64; p<0.05).Finally, an analysis of sinusitis prevalence revealed no association between vaccination status and disease occurrence (PR=1.05,95% CI: 0.65 to 1.70; p>0.05).
This study observed a significant reduction in CAP occurrence rates in children younger than two years old (infants) after the 10VPCV was introduced to the NIP immunization schedule.However, no reduction in the prevalence of otitis media and sinusitis was observed, suggesting that most cases (particularly for otitis media) are caused by other pathogens, particularly viruses, as well as other strains that are not found in the vaccine.
Despite these results, this study had several limitations.In this cross-sectional study, temporality, one of the main criteria of causality, is not guaranteed.However, the fi ndings are in accordance with the literature, thus minimizing this potential limitation.In addition, disease diagnosis was based on medical records from hospitals (hospitalizations) and outpatient records (queries), which could lead to potential bias due to errors inherent in medical practice.However, for cases of CAP, otitis media and sinusitis, the suspicion and confi rmation of diagnoses are essentially clinical and occasionally require radiographic supplementation, particularly in pneumonia cases and, more rarely, in sinusitis cases.Thus, we believe that the diagnostic criteria and confirmations cannot be neglected here because the children in this study were treated by doctors experienced in pediatric care.
In conclusion, disease prevention and health promotion are a primary goal, and vaccination programs can contribute to improvements in population health standards.In this decade, vaccines implemented by immunization programs, including the pneumococcal conjugate vaccines, have been demonstrated to be safe, effi cient, and effective in reducing illness and death caused by S. pneumoniae.In countries where vaccination has been implemented and maintained with high coverage, the reduction of invasive pneumococcal disease has been signifi cant.
Similarly, this study demonstrated a reduction in the number of children affected by CAP.However, to maximize vaccine effects, homogeneous coverage should be the goal of all primary care units.

DISCUSSION Rev Soc Bras Med Trop 48(1):44-49, Jan-Feb, 2015
-24 month age group had the highest number of cases compared to other age groups.There was no gender difference in the prevalence of sinusitis either before (2009-2010) or after (2011-2012) the vaccine was introduced (PR=0.96;95% CI: 0.60 to 1.55; p>0.05).Children's attendance data at