Epidemiology and clinical features of pneumonia according to radiographic findings in Gambian children

OBJECTIVE To assess the effect of vaccines against pneumonia in Gambian children. METHODS Data from a randomized, controlled trial of a 9-valent pneumococcal conjugate vaccine (PCV) were used. Radiographic findings, interpreted using WHO definitions, were classified as primary end point pneumonia, 'other infiltrates/abnormalities' pneumonia and pneumonia with no abnormality. We calculated the incidence of the different types of radiological pneumonia, and compared clinical and laboratory features between these groups. RESULTS Among children who did not receive PCV, the incidence of pneumonia with no radiographic abnormality was about twice that of 'other infiltrates' pneumonia and three times that of primary endpoint pneumonia. Most respiratory symptoms, reduced feeding and vomiting occurred most frequently in children with primary endpoint pneumonia. These children were more likely to be malnourished, to have bronchial breath sounds or invasive bacterial diseases, and to die within 28 days of consultation than children in the other groups. Conversely, a history of convulsion, diarrhoea or fast breathing, malaria parasitaemia and isolation of salmonellae were commoner in children with pneumonia with no radiographic abnormality. Lower chest wall indrawing and rhonchi on auscultation were seen most frequently in children with 'other infiltrates/abnormalities' pneumonia. CONCLUSION Primary endpoint pneumonia is strongly associated with bacterial aetiology and severe pneumonia'. Since this category of pneumonia is significantly reduced after vaccination with Hib and pneumococcal vaccines, the risk-benefit of antimicrobial prescription for clinical pneumonia for children with increased respiratory rate may warrant re-examination once these vaccines are in widespread use.