"Preventing the spread of influenza within families is an essential part of influenza management, regardless of the strain. This study shows that there is a clear benefit to be gained by giving antivirals to people who have been exposed to the virus to prevent the onset of symptomatic illness," said lead author M. Elizabeth (Betz) Halloran, M.D., D.Sc., a Hutchinson Center-based biostatistician.
"While the efficacy of antivirals to protect against influenza is critical, the effect of these drugs on infectiousness also has important public-health consequences. Further studies to determine antiviral efficacy for reducing infectiousness would therefore be of great value," said Halloran, a member of the Hutchinson Center's Public Health Sciences Division and a professor of biostatistics at the University of Washington School of Public Health and Community Medicine.
The report evaluated four household-based, randomized, placebo-controlled clinical trials, conducted from 2000 to 2004, which were designed to estimate the effect of post-exposure antiviral treatment on preventing influenza within households. Two of the trials were conducted with zanamivir and two with oseltamivir, permitting comparisons to be made between the two. The trials covered a total of 1,475 households. The majority of first household cases (53 percent to 70 percent) had influenza A (H3N2 or H1N1).
The authors reviewed the effects of each antiviral on infectiousness and pathogenicity ?the ability of the influenza virus to cause overt disease. Pathogenicity in controls ranged from 44 percent to 66 percent. Efficacy in reducing pathogenicity for zanamivir was 52 percent and 56 percent in the two zanamivir studies; for oseltamivir, it was 56 percent and 79 percent. The researchers found that both drugs were highly preventive against influenza illness, with oseltamivir at 81 percent efficacy and zanamivir at 75 percent. Thus, the preventive use of either product reduced by 75 percent to 81 percent the chance that an exposed person would become ill with influenza.
The researchers' analysis found a significant reduction in infectiousness, as defined by reductions in influenza illnesses in household contacts, when oseltamivir was used for treatment of ill persons, but not when zanamivir was administered. Although these results are of interest, the authors stress that the numbers were small and combined estimates from two studies for each drug were used in both instances. Furthermore, oseltamivir treatment of ill persons did not appear to reduce the frequency of influenza infection in their contacts.
In exploring the reasons for oseltamivir's greater effects on infectiousness, the authors speculated that the different modes of administration ?oral for oseltamivir and inhaled for zanamivir ?and resultant upper-respiratory viral levels could be the key, with zanamivir not reaching or affecting influenza virus in the nose. While both antivirals reduced cough, in earlier studies inhaled zanamivir did not significantly reduce nasal symptoms. In the paper the authors discussed the possibility that infectious droplets inhaled and exhaled from the nose may be important in viral transmission, thus orally administered oseltamivir might have an advantage in limiting spread.
In addressing the limitations to their research, the authors call for further stu dies into antiviral efficacy.
"Trials in non-household settings where influenza readily spreads ?such as schools, homes for the elderly and workplaces ?would be beneficial not only for planning management of seasonal influenza but also would be invaluable in pandemic planning. Additionally, simple solutions such as standardized randomization and study criteria would greatly facilitate trial comparison in the future. Such further studies will help us ensure we are better prepared not only for annual influenza seasons but also for the ultimate pandemic," said co-author Ira M. Longini, Jr., Ph.D., also a member of the Hutchinson Center's Public Health Sciences Division and a biostatistics professor at the University of Washington.