No Association between 2008–09 Influenza Vaccine and Influenza A(H1N1)pdm09 Virus Infection, Manitoba, Canada, 2009

Salaheddin M. Mahmud

, Paul Van Caeseele, Gregory Hammond, Carol Kurbis, Tim Hilderman, and Lawrence Elliott

Author affiliations: University of Manitoba, Winnipeg, Manitoba, Canada (S.M. Mahmud, P. Van Caeseele, G. Hammond, T. Hilderman, L. Elliott); Winnipeg Regional Health Authority, Winnipeg (S.M. Mahmud, C. Kurbis); Cadham Provincial Laboratory, Winnipeg (P.V. Caeseele); Manitoba Health, Winnipeg (T. Hilderman, L. Elliott)

Abstract

We conducted a population-based study in Manitoba, Canada, to investigate whether use of inactivated trivalent influenza vaccine (TIV) during the 2008–09 influenza season was associated with subsequent infection with influenza A(H1N1)pdm09 virus during the first wave of the 2009 pandemic. Data were obtained from a provincewide population-based immunization registry and laboratory-based influenza surveillance system. The test-negative case–control study included 831 case-patients with confirmed influenza A(H1N1)pdm09 virus infection and 2,479 controls, participants with test results negative for influenza A and B viruses. For the association of TIV receipt with influenza A(H1N1)pdm09 virus infection, the fully adjusted odds ratio was 1.0 (95% CI 0.7–1.4). Among case-patients, receipt of 2008–09 TIV was associated with a statistically nonsignificant 49% reduction in risk for hospitalization. In agreement with study findings outside Canada, our study in Manitoba indicates that the 2008–09 TIV neither increased nor decreased the risk for infection with influenza A(H1N1)pdm09 virus.

The nature of the relationship between receipt of the 2008–09 seasonal inactivated trivalent influenza vaccine (TIV) and the risk for infection with the pandemic (H1N1) 2009 virus strain, hereafter referred to as A(H1N1)pdm09, remains unclear. A case–control study in Canada that used data from a network of sentinel physicians monitoring influenza vaccine effectiveness in the provinces of Alberta, British Colombia, Ontario, and Quebec found an increased risk for influenza A(H1N1)pdm09 virus infection among persons who received the 2008–09 TIV (odds ratio [OR] 1.7, 95% CI 1.0–2.7); an increased risk for severe illness was not detected (1). In addition, 3 other studies conducted by the same team found a 1.4- to 2.5-fold increased risk for infection (laboratory confirmed) with influenza A(H1N1)pdm09 virus among persons who received the 2008–09 TIV (1). The 3 studies were 1) a test-negative case–control study in Ontario; 2) a household transmission cohort study in Quebec; and 3) a conventional case–control study using population controls in Quebec.
The results of these studies in Canada were not confirmed by studies conducted elsewhere. In fact, several studies using different designs found that TIV partially prevented or had no effect on infections with the pandemic strain (2–16). It has been suggested that the finding in Canada of an increased risk for influenza A(H1N1)pdm09 virus infection among persons who received the 2008–09 TIV might be unique to Canada; the increased risk might be related to the use of the domestically manufactured vaccine (1) or to the timing of the pandemic in relation to the most recent influenza season and the types of circulating influenza strains during that season (17,18). At the time of the pandemic, the Canadian province of Manitoba was not part of the Canadian vaccine effectiveness monitoring network. However, the availability of a provincewide, population-based immunization registry and laboratory-based influenza surveillance system provided a unique opportunity to investigate these issues in Manitoba.
In the first wave of the pandemic (May–August 2009), Manitoba was more severely affected than any other Canadian province, accounting for 50% of hospital intensive care unit admissions attributable to the virus in Canada (19,20). TIVs used in Manitoba during the 2008–09 influenza season were identical to those used elsewhere in Canada; they included 15 μg hemagglutinin each of A/Brisbane/59/2007 (H1N1)–like virus, A/Brisbane/10/2007 (H3N2)–like virus, and B/Florida/4/2006-like virus. These were the 3 strains recommended that year by the World Health Organization for influenza vaccines in the Northern and Southern Hemispheres (21). In Manitoba, as in other provinces, ≈75% of the administered seasonal influenza vaccine doses were manufactured domestically (Fluviral; GlaxoSmithKline, Mississauga, Ontario, Canada); imported vaccines, predominantly Vaxigrip (Sanofi Pasteur Ltd, Toronto, Ontario, Canada), comprised the remaining 25%. The live attenuated influenza vaccine was not available in Canada during the 2008–09 season.
To investigate whether use of TIV in Manitoba was associated with influenza A(H1N1)pdm09 virus infection during the first wave of the pandemic, we conducted a population-based case–control study using data from Cadham Provincial Laboratory (CPL) and the Manitoba Immunization Monitoring System. The test-negative case–control design used in this study is similar to the design of the Ontario study (1).