The role of respiratory virus infection in suspected pertussis: A prospective study.

Background: Infections caused by Bordetella pertussis are frequent and responsible for cases of huge severity in unvaccinated young infants. However, clinical manifestations vary and mimic other respiratory diseases as respiratory viruses.

Methods: A prospective cohort study was performed with infants under 1 old, hospitalized with suspected pertussis. All infants were submitted to etiological research to identify Bordetella pertussis (nasopharynx swab for culture and/or PCR) and respiratory viruses (nasopharyngeal aspirate for indirect immunofluorescence). Clinical and demographic data were collected.

Results: Among 59 infants, an etiological agent was identified in 37 (62.8%). Respiratory virus was identified in 19 (32%) and Bordetella pertussis in 14 (23.7%) as sole agent. Codetection was found in 4 (7%). Younger age, absence of fever, lack of BP immunization, leukocytosis > 20,000/mm, lymphocytosis >10,000/mm were associated to a greater chance of pertussis. Wheezing and living with siblings were associated with viral infection. After adjustment for confounders, the most important predictors were presence of wheezing for respiratory virus and leukocytosis for pertussis. The severity of infections by RV and BP were similar.

Conclusion: Respiratory virus infections are frequent in cases of clinical suspicion of pertussis and may actually exceed the prevalence of BP. Clinical/laboratory characteristics may suggest the etiology, but they are not pathognomonic, which stresses the need for respiratory virus and Bordetella pertussis research in this clinical situation.