Comparison of conventional viral cultures with direct fluorescent antibody stains for diagnosis of community-acquired respiratory virus infections in hospitalized children

Abstract

Objective: Because of the widespread availability of rapid viral antigen testing, many institutions never adopted a routine practice of ordering viral cultures to detect community-acquired respiratory viruses (CRVs). The ease of performing complete viral studies in our on site laboratory allowed us to assess the clinical implications of the absence of conventional culture results in previously healthy hospitalized children with CRV infections.

Methods: From June 1997 through May 2000, the results of direct immunofluorescence assay (DFA) of 1069 nasopharyngeal swab (NP) specimens were compared with simultaneously inoculated conventional tube cell cultures for detection of CRVs. In addition the medical records of 140 previously healthy infants and children hospitalized for management of lower respiratory tract infections caused by culture-proved CRVs were reviewed.

Results: Viruses were isolated or detected by DFA or viral culture or both in 468 (30%) of the 1557 NP samples evaluated. The most common CRV isolated was respiratory syncytial virus (49%), followed by parainfluenza viruses (15%), influenza A viruses (14%), rhinoviruses (8%), adenoviruses (4%), enteroviruses (4%) and influenza B viruses (1%). Of the 1069 NP specimens for which both viral culture and rapid antigen testing were performed, 190 specimens were DFA-positive and culture-positive, 7 specimens were DFA-positive and culture-negative, 35 specimens were DFA-negative and culture-positive and 837 specimens were DFA-negative and culture-negative. The overall sensitivity, specificity, positive predictive value and negative predictive value of DFA were 84, 99, 96 and 96%, respectively. Of the 140 hospitalized patients with culture-proved viral cultures (89 respiratory syncytial virus, 22 influenza A, 20 parainfluenza virus and 9 adenovirus), the mean duration of hospital stay was 3.6 days, and the mean time for viral cultures to become positive was 7.7 days (P < 0.001, signed rank test). One hundred twenty (86%) viral cultures did not become positive until after the patient had been discharged from the hospital. In no case was the clinical decision regarding the patient's treatment or discharge from the hospital based on the results of viral culture.

Conclusions: We conclude that positive viral cultures have no impact on clinical decision making and management of healthy children during hospitalization for illness attributable to community-acquired respiratory viruses.