Background: The major route of transmission of rotavirus is fecal-oral. However, the short incubation period, explosive nature of rotavirus outbreaks, frequency of respiratory symptoms in rotavirus-infected patients and detection of rotavirus in respiratory secretions suggest that transmission by small particle aerosol may occur. To evaluate this possibility, air samples from hospital rooms of infants and children (ages 2 weeks to 6 years) with rotavirus infection, defined by presence of rotavirus antigen in stool by EIA, were tested for rotavirus RNA by PCR.

Methods: The primers for PCR were specific for a portion of the gene coding for the VP6 protein of group A rotaviruses. Air samples from hospital rooms were passed through protected 0.45 μm cellulose filters placed 38 to 70 cm from the patient and attached to wall suction by a flow controller for 1 to 23 hours. Material trapped on the filter was eluted in 500μl of sterile water, RNA was extracted, reverse transcribed and subjected to 30 cycles of PCR amplification. PCR products were detected by gel electrophoresis. Negative specimens underwent a second round of 20 cycles of amplification. Control experiments indicated that one amplification detected 3600,000 rotavirus particles while the second amplification improved the sensitivity 100 fold.

Results: Rotavirus RNA was detected in 46 (75%) of 61 air samples from rooms with rotavirus-infected patients. The detection of rotavirus RNA was unrelated to the time from onset of disease, the presence of concomitant respiratory symptoms, severity of disease or duration of sampling. The mean age of children with rotavirus positive air samples (18.1± 2.2 mo) was older than that of children with negative air samples (10.2±1.9 mo) (p=0.05).

Conclusions: PCR has the sensitivity to detect rotavirus in air and to allow studies of airborne spread of rotavirus in the hospital environment. These findings suggest that aerosol transmission of rotavirus may occur.