Conclusion
The lack of clear superiority of respirators over facemasks in the studies of Loeb et al. and MacIntyre et al. may result from poor respirator face seals, poor compliance due to discomfort, lack of recognition of infectious patients and consequent inappropriate non-use of respirators, infection arising from infectious co-workers, trans-ocular infection despite appropriate respirator use but no eye protection, or infection from sources outside the healthcare setting.
Regardless of the reason for failure, the high rate of infection in both of the groups in the Loeb study is impressive and reinforces the need to consider how protection can be strengthened. In relation to aerosol-generating pro- cedures the results of a recent review concluded that, although there are a number of these procedures listed under this heading, few have sufficient evidence to confirm they actually do produce - therefore further research in this area is warranted. The view that cross-infection may be reduced by placing masks on potentially infectious patients, supported by bench and clinical studies, opens up an additional approach to protection. The demonstration of trans-ocular infection by aerosols needs further investigation and suggests that eye protection may be required as a component of respiratory and facial protection, not only to mitigate risks associated with direct splash or splatter contamination, but also to prevent aerosol exposure. Finally, as pointed out by Srinivasan and Perl, and also by a recent Department of Health scientific review, the use of masks and respirators should be considered as the last line of defence in the hierarchy of infection prevention measures. These include vaccination (when available), hand hygiene (always), environmental mea- sures including sufficient ambient ventilation, the provision of single occupancy rooms, and administrative practices that emphasize early recognition of infectious patients and their removal from others.
