LETTER TO THE EDITOR Response to influenza vaccine in people with non-protective HI antibody titers Vincenzo Baldo 1 , Tatjana Baldovin 1 , Annarosa Floreani 2 , Michele Minuzzo 1 and Renzo Trivello 1 1 Department of Environmental Medicine and Public Health, Institute of Hygiene, University of Padua, Via Loredan 18, 35131, Padova, Italy; 2 Department of Surgical and Gastroenterological Sciences, University of Padova, Padova, Italy Received: 22 August 2005/Accepted in revised form 2 October 2006 Abstract. The purpose of the study was to determine which factors correlate directly with response to vaccination in such a group of subjects with non- protective HI antibody titers before vaccination. Two vaccines were used, a subunit virus vaccine adju- vanted with MF59 and a split virus vaccine. The analysis indicated that immunization with vaccine adjuvanted with MF59 was an independent variable for immune response against A/H3N2 (OR: 3.51; 95% CI: 1.81–6.79) and B (OR: 2.31; 95% CI: 1.37– 3.89). The results suggest that antibody response to vaccine is satisfactory in elderly people previously lacking a protective antibody titer, and that the adjuvanted vaccine reveals a better immunogenicity. Key words: Antibody response, Influenza, MF59, Vaccine Influenza infection is a major cause of illness, mor- bidity and mortality over the world. The World Health Organization estimates that influenza affects 5–15% of the globeÕs population each year. The groups at high risk of influenza complications mainly include the elderly and patients with cardiovascular or pulmonary disorders, and metabolic disease. The institutionalized population is also considered at risk, since viruses are more easily transmitted in such environments. The influenza vaccine is effective in reducing influenza-related illness and death among people aged 65 years and over who live in the community [1]. The results of the haemagglutination-inhibition (HI) antibody test for influenza virus antibody in human sera closely match those produced by virus neutral- ization assays and are indicative of protection [2]. Resistance to influenza infection in humans has been correlated with global humoral immune response, which is less efficient in the elderly [3]. The goal of the present study was to determine which factors correlate directly with response to vaccination in such a group (with age >64 years) of subjects with non-protective HI antibody titers before vaccination. The present study was a randomised, double-blind trial. Five hundred subjects were enrolled in the Mirano Public health District (North East Italy) during the influenza vaccination campaign. A 10 ml blood sample was obtained from each person before immunization and then again 4 weeks later. The sera were stored at )20 °C until the laboratory determi- nation of HI antibody titers, as described elsewhere [4]. The present study only considers subjects who were found unprotected prior to vaccination (anti-HI antibody ‡1:40). Counselling were administered and informed consent was obtained from each partici- pant. Subjects were not included in the study group if they had evidence of acute disease at time of immu- nisation, known allergy to any vaccine component or avian proteins and previous severe reactions to im- munisation against influenza. Blinding was main- tained until all study data were collected and analysed. Two vaccines were used, i.e. a subunit virus vaccine adjuvanted with MF59, Fluad Ò , Chiron Vaccines, Siena, Italy (sub/MF59), and a split virus vaccine, Mutagrip Ò , Pasteur Me´rieux MSD, Lyon, France (SVV). The vaccines contained the influenza strains recommended by the WHO (A/H1N1/New Caledonia/20/99; A/H3N2 Moscow/10/99; B/Sichu- an/379/99). Subjects received a single dose of vaccine in the deltoid muscle. The parameters used to mea- sure humoral immune response were: (i) the geo- metric means of HI antibody titers (GMTs); (ii) the number of people showing protective HI antibody titers (a titer of 1:40 was assumed to provide pro- tection); and (iii) the number of people with a positive response after vaccination (i.e. a four-fold or greater rise in HI antibody titer in people who were serum- positive before vaccination or a rise from <1:10 to ‡1:40 in those who were serum-negative). The v 2 test was used to analyze differences between proportions. Significance between pre- and post- vaccination titers was calculated using the StudentÕs paired t-test. Comparisons of different vaccine groups were drawn using StudentÕs t-test for unpaired data. Logistic regression was conducted out for each virus strain to assess which variables were significant in a European Journal of Epidemiology (2006) 21:843–845 Ó Springer 2006 DOI 10.1007/s10654-006-9071-4