This double blind, multicentre study performed over a single winter season recruited 1291 household contacts (>5 years old) of 485 patients presenting with an influenza-like illness (occurring when influenza was known to be circulating locally). Household contacts were randomised to receive zanamivir 10 mg or placebo for 10 days within 36 hours of symptom onset in the index case (who did not receive antiviral treatment). Symptomatic laboratory confirmed influenza (confirmed by a fourfold rise in anti-influenza antibody, viral culture or PCR) during the period of prophylaxis occurred in at least one household contact in 4% of households treated with zanamivir compared with 19% of households given placebo (p<0.001), representing an 81% protective efficacy. Similar efficacy was achieved with both influenza A (78%) and B (85%). There was no evidence of development of resistance in isolates and there were no episodes of bronchospasm noted among the 661 contacts receiving zanamivir, although only 11% had an underlying respiratory condition and severe persistent asthma was an exclusion criteria.

Influenza vaccination remains the mainstay of prophylaxis, but this study suggests that prophylactic zanamivir has a role particularly in households with at-risk individuals where vaccination uptake is low.