Introduction Bronchiolitis is a common infectious respiratory condition of infancy. Evidence of any correlation between the infecting pathogen, length of stay and severity of bronchiolitis continues to be sparse. This study evaluated the virology in bronchiolitis and suggests a new severity index to predict the severity of bronchiolitis.

Methods We conducted a retrospective study of children admitted with bronchiolitis at our children’s hospital between October to December 2015.Data was collected using a structured proforma. Nasopharyngeal aspirates were performed using a standardised protocol and analysed using viral PCR.

We suggest the Bronchiolitis severity index (BSI) below:

BSI was calculated for all children admitted with bronchiolitis where nasopharyngeal aspirates for viral PCR were sent.

Results 279 infants were admitted with a diagnosis of bronchiolitis during the study period. 163 were males.The mean length of stay was 4.52 days.

Of the 228 infants who had nasopharyngeal aspirates done for viral PCR, 165 infants (72.3%) had RSV infection, 27 (11.8%) had RSV co-infection with another respiratory virus and 36 (15.7%) had non-RSV infection or negative virology. There were no significant differences in ages, length of stay and need for respiratory support among these groups. We found a significant statistical difference (p 0.04) in the severity index between the above groups with the BSI for RSV co-infection with another respiratory virus in bronchiolitis being significantly higher than the other groups.

Conclusion Our study shows that BSI has good clinical utility and is easy to use in a busy clinical setting. Larger prospective studies to validate BSI needs to be conducted.