Influenza Deaths in Spite of Immunization and Prophylaxis

SIR—We would like to add evidence for the multimodal influenza prevention strategies so ably advanced by Monto et al. [1] and by Bridges and Harper [2]. In a long-term care facility (Laguna Honda Hospital, San Francisco, California) with a census of 1055, we documented 29 patients with cases of influenza A by rapid EIA during the 2001–2002 influenza season. Of these 29 patients, 26 received timely influenza immunizations, as did 92% of all facility residents.

Four patients died, all of whom had received timely influenza vaccinations and had received at least 5 days of prophylaxis with rimantadine before the appearance of their first symptoms. All 4 patients had their treatment switched to oseltamivir after the onset of symptoms—3 patients within 1 day after the onset of symptoms and 1 patient within 2 days after the onset of symptoms. One of the deaths occurred within 24 h after the onset of symptoms. The rapid course of the illness and the autopsy findings of hemorrhagic tracheitis and bronchopneumonia were consistent with primary viral pneumonia. In addition to those 4 deaths, another death occurred in a patient who had negative EIA results but whose illness was clinically consistent with influenza. This brought our total number of cases to 30, for an attack rate of 3%. The average age of the 30 patients involved in the cases was 72 years; the average age at death for the 5 patients was 80 years. We checked one specimen for virus subtype and found that it was H2N3. Two nursing facilities in our area experienced >50% attack rates with subtype H2N3.

These deaths did not diminish our enthusiasm for immunization or antiviral prophylaxis; the benefit of both is well demonstrated. Rather, with this reminder of the lethal potential of influenza, we redoubled our efforts in all components of influenza prevention, including patient and staff immunization, aggressive surveillance with rapid-testing followed by more-sensitive testing for specimens with negative results, oseltamivir treatment within 24 h after the onset of symptoms for patients with documented infection, prophylaxis with antivirals (rimantadine, in our case), and implementation of infection control measures. The staff immunization rate increased from 26% in 2001–2002 to 37% in 2002–2003 and 48% in 2003–2004. Improvements in tracking and follow-up will likely yield better rates during the 2004–2005 season.

For most nursing facilities, the most difficult prevention component will be to perform rapid tests and to implement treatment and prophylaxis protocols within 24–36 h after the onset of symptoms. Even partial implementation of laboratory testing has payoffs, however, because laboratory-confirmed diagnoses can galvanize nurses, administrators, and physicians into action. Aggressive viral testing of appropriate nursing facility patients in hospital emergency departments could improve identification of influenza outbreaks and help to leverage system changes. Testing also reveals the presence of other lethal viruses. In particular, the impact of respiratory syncytial virus infection in our facility has again motivated us to strengthen our infection control procedures.

As Bridges and Harper [2] suggest, it is unrealistic to think that most nursing facilities can plan and implement all of these components without outside assistance. We support their call for all stakeholders to move to a new level of collaboration.

Acknowledgments

Potential conflicts of interest.All authors: no conflicts.