Temporal relationships between colds, upper respiratory viruses detected by polymerase chain reaction, and otitis media in young children followed through a typical cold season.

Winther, Birgit; Alper, Cuneyt M; Mandel, Ellen M; Doyle, William J; Hendley, J Owen

INTRODUCTION Otitis media is a frequent complication of a viral upper respiratory tract infection, and the reported co-incidence of those diseases increases with assay sensitivity and sampling density. We determined the incidence of otitis-media complications in young children when referenced to cold-like illnesses and to concurrent virus recovery from the nasopharynx. METHODS A total of 60 children from 24 families were followed from October 2003 through April 30, 2004, by daily parental recording of illness signs, weekly pneumatic otoscopic examinations, and periodic polymerase chain reaction assay of collected nasal fluids for common viruses. RESULTS One hundred ninety-nine cold-like illnesses were observed, but a sample for virus assay was not collected concurrent with 71 episodes. Of the remainder, 73% of cold-like illnesses were temporally related to recovery of 1 or a combination of the assayed viruses, with rhinovirus predominating. For non-cold-like illness periods, 54 (18%) of 297 assays were positive for virus, and the virus frequency distribution was similar to that for cold-like illnesses. There were 93 diagnosed otitis-media episodes; 65 (70%) of these occurred during a cold-like illness. For the 79 otitis-media episodes with available nasal samples, 61 (77%) were associated with a positive virus result. In this population, the otitis-media complication rate for a cold-like illness was 33%. CONCLUSIONS A cold-like illness was not a prerequisite for polymerase chain reaction detection of viruses in the nose and nasopharynx of young children. Viral detection by polymerase chain reaction in the absence of a cold-like illness is associated with complications in some subjects. Otitis media is a complication of viral infection both with and without concurrent cold-like illnesses, thus downwardly biasing coincidence estimates that use cold-based illnesses as the denominator.