Am Fam Physician. 2006;73(11):2019
Clinical Question: Are antiviral medications effective in preventing influenza or treating healthy adults with influenza?
Setting: Outpatient (any)
Study Design: Systematic review
Synopsis: The authors searched multiple databases for randomized controlled trials of antiviral medications used for treating healthy adults 16 to 65 years of age with influenza. They also sought unpublished studies. Two authors independently decided which studies to include and assessed the quality of the included studies, and a third member of the team settled discrepancies.
They analyzed 20 prophylaxis studies and 13 treatment trials using the M2 ion blockers amantadine (Symmetrel) and rimantadine (Flumadine). Additionally, they analyzed 19 studies of the neuraminidase inhibitors zanamivir (Relenza) and oseltamivir (Tami-flu) as prophylaxis, treatment, and postexposure prophylaxis.
In the prophylaxis studies, amantadine prevented 61 percent of influenza A cases but only 25 percent of influenza-like illnesses. Adverse effects caused more patients to stop taking amantadine than placebo. Rimantadine was no better than placebo in preventing influenza and influenza-like illness and also was more likely to cause adverse effects that resulted in medication cessation. The neuraminidase inhibitors were no better than placebo in prophylaxis against influenza-like illness. Oseltamivir prevented 54 percent of influenza cases and zanamivir prevented 43 percent. In a single study, the neuraminidase inhibitors prevented lower respiratory tract infections in confirmed influenza cases but not in influenza-like illnesses. The authors were unable to find any rigorous studies of oseltamivir on cases of avian influenza. The authors do not provide enough detail to calculate numbers needed to treat or to harm.
Bottom Line: Antiviral agents are only slightly effective in preventing confirmed influenza or influenza-like illness. When given in the first few days of illness, the M2 ion blockers and neuraminidase inhibitors reduce the duration of illness by approximately one day. (Level of Evidence: 1a)