Epidemiology of Acute Respiratory Infections in Children of Developing Countries Stephen Berman
From the Department of Pediatrics, University of Colorado Health Sciences Center, Denver, Colorado
In developing countries every 7 seconds a child under 5 years of age dies because of an acute respiratory infection (ARI), usually pneumonia. ARI is responsible for the deaths of four and a half million children each year, accounting for 30 % of all deaths in childhood . Pneumonia unassociated with measles causes 70% of these deaths; post-measles pneumonia, 15%; pertussis, 10 %; and bronchiolitis or croup, 5 % . This paper reviews current information on the etiology of ARI in developing countries and the influence of several host and environmental risk factors on the incidence and severity of ARI. Interpretation of the available data is compromised by the difficulty in establishing whether differences among studies represent true geographic differences rather than methodologic differences related to patient selection, study design, and/or laboratory methods.
disease. Most viral infections are mild, self-limited illnesses involving the upper respiratory tract. Bacterial agents cause upper respiratory infections such as acute otitis media and Streptococcus pyogenes pharyngitis, as well as pneumonia. While bacterial pneumonia is less common than viral lower respiratory infection (LRI), the risk of death is far greater with bacterial pneumonia. The estimated case-fatality rate for bacterial pneumonia due to Streptococcus pneumoniae and Haemophilus infiuenzae in developing countries is >50 times higher than the case-fatality rate for infection due to respiratory syncytial virus (RSV) or parainfluenza viruses, yet the estimated number of deaths due to bacterial pneumonia is only 2.7 times higher than the number due to the viral infections . It now seems clear that both viral and bacterial infections are important contributors to the excessive ARI mortality among children living in developing countries.
Etiology Viral and bacterial agents are responsible for most cases of ARI. Viral agents cause the majority of ARI, but only a small percentage of these infections result in severe or fatal
Grant support: Grant no. RR-69 from the General Clinical Research Center Program of the Division of Research, National Institutes of Health. Reprints and correspondence: Dr. Stephen Berman, Department of Pediatrics, University of Colorado Health Sciences Center, Box C230, 4200 East Ninth Avenue, Denver, Colorado 80262. Reviews of Infectious Diseases 1991;13(Suppl 6):8454-62 © 1991 by The University of Chicago. All rights reserved. 0162-0886/91/1303-0062$02.00
Bacterial Agents The best available etiologic information comes from 14 studies that reported on the bacteria isolated from lung aspirates obtained from children hospitalized for pneumonia who had not received antibiotics (table 1) [4-17]. The 14 studies are summarized in relation to agent-specific isolation rates for S. pneumoniae, H. infiuenzae, and Staphylococcus aureus. Methodologic variations limit the meta-analysis of these data. S. pneumoniae, the most frequently isolated pathogen, was isolated from >30% of the patients in 60% of the studies. H. inftuenzae was isolated from 11%-50% of the patients in all of the studies. S. aureus usually was isolated in
Downloaded from http://cid.oxfordjournals.org/ at University of Sussex on October 12, 2012
Acute respiratory infections cause four and a half million deaths among children every year, the overwhelming majority occurring in developing countries. Pneumonia unassociated with measles causes 70% of these deaths; post-measles pneumonia, 15%; pertussis, 10%; and bronchiolitis and croup syndromes, 5%. Both bacterial and viral pathogens are responsible for these deaths. The most important bacterial agents are Streptococcus pneumoniae, Haemophilus influenzae, and Staphylococcus aureus. The data on bacterial etiology of pneumonia during the first 3 months of life are limited, and almost no information on the role of chlamydia and pertussis in this age period is available. The distribution of viral pathogens in developing countries can be summarized as follows: respiratory syncytial virus, 15%-20%; parainfluenza viruses, 7%-10%; and influenza A and B viruses and adenovirus, 2%-4%. Mixed viral and bacterial infections occur frequently. Risk factors that increase the incidence and severity of lower respiratory infection in developing countries include large family size, lateness in the birth order, crowding, low birth weight, malnutrition, vitamin A deficiency, lack of breast feeding, pollution, and young age. Effective interventions for prevention and medical case management are urgently needed to save the lives of many children predisposed to severe disease.
Epidemiology of Pediatric ARI
RID 1991;13 (Suppl 6)
Table 1. Review of 14 studies that reported isolation results for 1,096 lung aspirates from hospitalized children before administration of antibiotics. Bacterial pathogens (no. of studies) Total pathogens (14)
S. pneumoniae (10) H. infiuenzae (10) S. aureus (8) NOTE.
Percentage of positive isolations
62 27 27 17
0 0 0 3
1 4 8 4
2 4 2 1
11 2 0 0
No. of studies with indicated isolation rates
Data are from [4-17].
Table 2. Review of selected studies that reported isolation results for lung aspirates according to antibiotic (A) history.
Location of study [reference] Santiago, Chile  Cali, Colombia  Pune, India  Benin, Nigeria 
No. of children
Percentage with positive isolation
Age of children (y)
0-2 0-14 0-11 0-12
66 31 20 27
160 71 50 46
22.7 12.9 30.0 22.2
56.8 21.1 60.0 73.9
Table 3. Correlation of bacterial isolation from blood cultures and lung aspirates.
Location of study [reference] Nigeria  The Gambia 
No. (%) of bacterial isolations from indicated site(s) No. of cases
Lung and blood
4 (13) 11 (27)
Lung alone Blood alone 28 (87) 25 (61)
serotypes (table 5). The three most common types isolated in Papua New Guinea were types 6, 14, and 19; in The Gambia, types 5, 1, and 6; and in Pakistan, types 19, 31, and 16. No type 31 strains were isolated in The Gambia or Papua New Guinea, and no type 1 strains were isolated in Pakistan or Papua New Guinea. Limited information is available on blood culture isolates in ARI in early infancy «2 months of age) from developing countries (table 6). Community studies in the BOSTID program [21, 23-25] reported few cases with positive blood cultures. There were seven LRI -associated cases of bacteremia among the 340 newborns followed during their first 5 months of life in Cali, Colombia . The rate of bacteremic LRI is higher in studies carried out in hospital emergency rooms and inpatient units. The BOSTID study carried out in Islamabad was remarkable for the very high isolation rates from blood cultures . A retrospective review of hospitalized young infants in Cali identified 17 cases in infants