Two thousand six hundred thirty-eight of 3803 eligible children were enrolled from January 2010 through June 2012 in a prospective study of children younger than 18 years old requiring hospitalization at three children’s hospitals in the United States (Memphis, Nashville, and Salt Lake City). Children with recent hospitalization or severe immunosuppression history were excluded. Blood and respiratory specimens were collected for pathogen detection by traditional cultures and PCR. Chest X-rays were independently reviewed by a panel of study radiologists.

Eighty-nine percent of children hospitalized had radiographic evidence of pneumonia. The mean age was 2 years old. Twenty-one percent of children required admission to intensive care units, 7% required mechanical ventilation, and 3 children (1%) died. Thirty-three percent of children had underlying asthma or reactive airway disease, and 21% of children younger than 2 years old had a history of preterm birth. Among the 2222 children with radiographic evidence of pneumonia, a viral or bacterial pathogen was detected in 81%, one or more viruses in 66%, bacteria in 8%, and both bacterial and viral pathogens in 7%.The overall incidence of CAP requiring hospitalization in children was 15.7 cases/10,000 children and the highest rate was in children younger than 2 years, in whom the rate was 62.2 cases/10,000 children.

Respiratory syncytial virus (RSV) was more common in children younger than 5 years of age than in older children (37% vs. 8%), as were adenovirus (15% vs. 3%), and human metapneumovirus (HMPV) (15% vs. 8%). Together, HMPV, adenovirus, parainfluenza virus, and coronavirus accounted for one-third of pathogens detected, with the highest rates seen in children younger than 5 years old. As expected, RSV peaked sharply in the winter months. Human rhinovirus was detected in 27% of children with pneumonia. Bacterial pathogens were detected in 15% of children with pneumonia. Streptococcus pneumoniae was detected in just 79 cases (2%) and was roughly equal in younger and older children.Mycoplasma pneumoniae was more common in children older than 5 years of age than in younger children (19% vs. 3%).


This paper presents a nice update on the etiology of CAP in children requiring hospitalization and used modern sensitive laboratory methods to reveal an etiology of infection in a high percentage of patients studied. The study reinforces the importance of CAP requiring hospitalization being of much higher incidence in very young children. Of note in this study was the predominance of viral vs. bacterial pathogens identified in this study (71% vs. 15%), with some overlap in cases in which both viral and bacterial pathogens were identified. The predominance of viral pathogens, especially in young children, was striking and likely reflects both the direct effects (and herd immunity effects) of the use of both pneumococcal conjugate vaccine and HiB vaccine. It should also be noted that this study took place after the 2009-2010 pandemic of Influenza A (H1N1), which would have made the incidence of influenza virus more common than was seen in this study.