Compounds lacking oral activity may be delivered intranasally to treat influenza virus infections in mice. However, intranasal treatments greatly enhance the virulence of such virus infections. This can be partially compensated for by giving reduced virus challenge doses. These can be 100- to 1,000-fold lower than infections without such treatment and still cause equivalent mortality. We found that intranasal liquid treatments facilitate virus production (probably through enhanced virus spread) and that lung pneumonia was delayed by only 2 days relative to a 1,000-fold higher virus challenge dose not accompanied by intranasal treatments. In one study, zanamivir was 90 to 100% effective at 10 mg/kg/day by oral, intraperitoneal, and intramuscular routes against influenza A/California/04/2009 (H1N1) virus in mice. However, the same compound administered intranasally at 20 mg/kg/day for 5 days gave no protection from death although the time to death was significantly delayed. A related compound, Neu5Ac2en (N-acetyl-2,3-dehydro-2-deoxyneuraminic acid), was ineffective at 100 mg/kg/day. Intranasal zanamivir and Neu5Ac2en were 70 to 100% protective against influenza A/NWS/33 (H1N1) virus infections at 0.1 to 10 and 30 to 100 mg/kg/day, respectively. Somewhat more difficult to treat was A/Victoria/3/75 virus that required 10 mg/kg/day of zanamivir to achieve full protection. These results illustrate that treatment of influenza virus infections by the intranasal route requires consideration of both virus challenge dose and virus strain in order to avoid compromising the effectiveness of a potentially useful antiviral agent. In addition, the intranasal treatments were shown to facilitate virus replication and promote lung pathology.