Asthma and the Risk of Invasive Pneumococcal Disease: A Meta-analysis Jose A. Castro-Rodriguez, MD, PhD,a Katia Abarca, MD, MPH,b Erick Forno, MDc CONTEXT: Invasive pneumococcal disease (IPD) and pneumonia are a leading cause of morbidity abstract and mortality throughout the world, and asthma is the most common chronic disease of childhood. OBJECTIVE: To evaluate the risk of IPD or pneumonia among children with asthma after the introduction of pneumococcal conjugate vaccines (PCVs). DATA SOURCES: Four electronic databases were searched. STUDY SELECTION: We selected all cohorts or case-control studies of IPD and pneumonia in populations who already received PCV (largely 7-valent pneumococcal conjugate vaccine), but not 23-valent pneumococcal polysaccharide, in which authors reported data for children with asthma and in which healthy controls were included, without language restriction. DATA EXTRACTION: Two reviewers independently reviewed all studies. Primary outcomes were occurrence of IPD and pneumonia. Secondary outcomes included mortality, hospital admissions, hospital length of stay, ICU admission, respiratory support, costs, and additional medication use. RESULTS: Five studies met inclusion criteria; of those, 3 retrospective cohorts (∼26 million person-years) and 1 case-control study (N = 3294 children) qualified for the meta-analysis. Children with asthma had 90% higher odds of IPD than healthy controls (odds ratio = 1.90; 95% confidence interval = 1.63–2.11; I2 = 1.7%). Pneumonia was also more frequent among children with asthma than among controls, and 1 study reported that pneumonia-associated costs increased by asthma severity. LIMITATIONS: None of the identified studies had information of asthma therapy or compliance. CONCLUSIONS: Despite PCV vaccination, children with asthma continue to have a higher risk of IPD than children without asthma. Further research is needed to assess the need for supplemental 23-valent pneumococcal polysaccharide vaccination in children with asthma, regardless of their use of oral steroids. aDivision of Pediatrics, Department of Pediatric Pulmonology and Cardiology and bDepartment of Pediatric Infectious Diseases and Immunology, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile; and cDivision of Pulmonary Medicine, Department of Pediatrics, School of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania Dr Castro-Rodriguez conceptualized and designed the study, reviewed and collected the data, drafted the initial manuscript, and reviewed and revised the manuscript; Dr Abarca made a substantial contribution to the analysis and interpretation of data and critically revised the manuscript for important intellectual content; Dr Forno reviewed and collected the data, conducted and interpreted the meta-analyses, and reviewed and revised the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work. DOI: https://doi.org/10.1542/peds.2019-1200 Accepted for publication Aug 28, 2019 Address correspondence to Jose A. Castro-Rodriguez, MD, PhD, Division of Pediatrics, Department of Pediatric Pulmonology and Cardiology, School of Medicine, Pontificia Universidad Católica de Chile, Lira 44, 1er Piso, Casilla 114-D, Santiago, Chile. E-mail: [email protected] To cite: Castro-Rodriguez JA, Abarca K, Forno E. Asthma and the Risk of Invasive Pneumococcal Disease: A Meta-analysis. Pediatrics. 2020;145(1):e20191200 Downloaded from www.aappublications.org/news by guest on September 25, 2021 PEDIATRICS Volume 145, number 1, January 2020:e20191200 REVIEW ARTICLE Streptococcus pneumoniae is still one Asthma is the most common chronic populations who have already of the most frequent causes of disease of childhood.6 It affects .6.5 received PCV7, 10-valent invasive disease, such as sepsis and million children in the United States pneumococcal conjugate vaccine, or meningitis, and a frequent cause of alone,7 with millions more around the PCV13, but not PPSV23; with (4) no bacterial pneumonia, acute otitis world, and its prevalence steadily language restriction. The exclusion media, and rhinosinusitis.1 Although increased from the 1980s to at least criteria were (1) no specific data for these diseases can occur in both the 2000s.7 In a previous study in children with asthma reported in the healthy children and those with which authors reported increased population analysis, (2) IPD chronic underlying diseases, their risk of IPD in children with asthma morbidity or mortality description in incidence and severity are and adults involved a period of time children with asthma but in the significantly higher in those with (1995–2002) in which most children absence of a control group, and (3) chronic underlying disease.2 would not have received PCV,8 and reviews, letters, abstracts, or articles a systematic review of IPD in asthma lacking sufficient information in Invasive pneumococcal disease (IPD) included only 1 pediatric study in the English for data synthesis or analysis. is a leading cause of morbidity and PCV era.9 Here, we aim to evaluate The primary outcomes were the mortality throughout the world. In current evidence on the risk of IPD in occurrence of IPD, defined as above, 2000, before the introduction of the children with asthma after the and pneumococcal pneumonia. 7-valent pneumococcal conjugate introduction of PCV. Secondary outcomes, if available, vaccine (PCV7), an estimated 14.5 were hospital admissions, mortality, million episodes of IPD occurred METHODS length of hospital stay, admission to among children ,5 years of age, the ICU, need for invasive respiratory resulting in an estimated 826 000 Search and Selection Criteria support, additional medication use deaths (11% of all deaths in that age (ie, in addition to the patient’s 3 We searched 4 electronic databases group). Pneumococcal conjugate baseline), all-cause pneumonia, and (Medline, the Cochrane Collaboration vaccine (PCV) PCV7 was introduced costs associated with disease. clinical trials register, Latin American in the United States in 2000, and 13- and Caribbean Health Sciences Data Abstraction and Assessment of valent pneumococcal conjugate Literature, and Cumulative Index to Risk of Bias vaccine (PCV13) replaced it in 2010. Nursing and Allied Health Literature) Current guidelines from the US Titles, abstracts, and citations were up to October 2018. The search was Centers for Disease Control and independently analyzed by 2 conducted by using the following Prevention (CDC)4 and the American independent investigators (J.A.C.-R. keywords: “(((pneumococcal Academy of Pediatrics (AAP)5 and E.F.), and any disagreements infections) OR (invasive recommend 4 doses of PCV13 (at 2, 4, were resolved by consensus after pneumococcal disease) OR 6, and 12–15 months of age) and discussion. The reviewers (pneumococcal pneumonia)) AND a dose of the 23-valent pneumococcal independently assessed the full text ((asthma OR wheezing))),” as well as polysaccharide vaccine (PPSV23) at of all studies for inclusion on the the corresponding Medical Subject 2 years of age for children with basis of the criteria for population Headings terms, restricted to children conditions considered high risk for intervention, study design, and (birth to 18 years of age). We also IPD or as soon as possible after outcomes. After obtaining full reports searched the references of included a diagnosis of chronic illness is made from potentially relevant studies, they publications as well as other after the age of 2 years. High-risk independently reassessed eligibility. If nonbibliographic data sources such as conditions include cerebrospinal fluid the information was incomplete, we pharmaceutical industry Web sites. leak or cochlear implants; diabetes; attempted to contact the authors. The HIV infection or immunodeficiencies The inclusion criteria were (1) risk of bias from including certain (congenital, acquired, or secondary to cohorts or case-control studies studies was assessed according to the medications); anatomic or functional including children with and without Newcastle-Ottawa Scale.10,11 asplenia; sickle cell and other asthma; (2) assessment of IPD Data Analysis hemoglobinopathies; neoplasms; and (defined as the isolation of S chronic diseases including chronic pneumoniae from a normally sterile When feasible, we calculated pooled heart, lung, kidney, or liver diseases. fluid [eg, blood, cerebrospinal fluid, odds ratios (ORs) with 95% Currently, PPSV23 vaccination is pleural fluid, peritoneal fluid, confidence intervals (CIs). recommended for patients with pericardial fluid, surgical aspirate, Heterogeneity was assessed by using asthma only if they are treated with bone or joint fluid by any laboratory the I2 test (#25% absence of bias; high-dose oral corticosteroid diagnosis test]) with association of 26%–39% unimportant; 40%–60% therapy.4,5 morbidity or mortality; (3) in moderate; 60%–100% substantial Downloaded from www.aappublications.org/news by guest on September 25, 2021 2 CASTRO-RODRIGUEZ et al bias).12 To address the variability the inclusion criteria for the studies did the authors report across studies for each outcome of qualitative synthesis, of which 413–16 separately the fluid site from which interest, a fixed-effects meta-analysis also fulfilled criteria for the the pneumococcal infection was was used when low heterogeneity quantitative synthesis or meta- isolated. was present (I2 ,40%), and analysis (Table 1). The other 6 Among the 4 studies included for the a random-effects