Comment in which the possibility of emergence of resistance to the different neuraminidase inhibitors was compared. 20, 21 Six children in Japan had viruses with mutations of arginine to lysine at aminoacid 292 of the neuraminidase molecule. This arginine forms part of the active site of the enzyme, where the neuraminidase inhibitors also bind, blocking the enzyme’s activity. The change to lysine prevents the movement of other aminoacids in the active site that create a pocket into which the bulky side-chain of oseltamivir fits. If the pocket cannot be formed, oseltamivir cannot fit, leading to resistance. 22 The new mutation at position 294, described by Kiso and colleagues, might also affect movement of these aminoacids. Viruses from two other children had mutations at glutamic acid 119. This mutation, in a residue that forms part of the neuraminidase active-site framework, has previously been recorded in oseltamivir-treated patients. If the high percentage of mutants in Kiso’s study is a general occurrence, then mutations could occur in children far more frequently than has been observed. 19 Are the oseltamivir-resistant viruses transmissible and pathogenic? If we are very lucky, they may have a growth disadvantage, or, for other reasons, be less virulent or transmissible. This is an important question left open by Kiso that needs to be addressed. If resistant variants are transmissible and pathogenic, then the widespread use of oseltamivir in a pandemic situation raises concerns. Generally, neuraminidase mutations have led to viruses with reduced pathogenicity in animal models22, 23 because the mutations cause defects in an important enzyme. There is no documented transmission of an oseltamivir-resistant virus in human beings, but the frequent emergence of resistance mutations in Kiso’s study suggests that this is only a matter of time. What can be done? In view of the threat of a pandemic, we need to heighten awareness of the critical importance of neuraminidase inhibitors for our arsenal. Stockpiles of these drugs are a key piece of preparedness and would be critical to an effective response. For these reasons, it is vital to understand more about which features of neuraminidase inhibitors might discourage the emergence of resistance. Whereas zanamivir seems preferable in terms of development of resistance, the route of delivery could be problematic for some populations. We need more information on the emergence of resistance, especially in oseltamivir-treated patients, and we urgently need to know whether resistant variants, such as those identified in Kiso’s study, are transmissible. The development of the neuraminidase inhibitors has been a true success story: protein structural analysis directly applied to preventing and treating a major infectious threat. Let us take Kiso and colleagues’ study as an energising mandate to learn more about the incidence and mechanisms of resistance to the neuraminidase inhibitors, so that appropriate strategies can be developed for their use during the next pandemic.