Vaccine 24 (2006) 6615–6619
An influenza A/H3 outbreak during the 2004/2005 winter in elderly
vaccinated people living in a nursing home
A.M. Iorioa,∗, M. Neria, E. Lepria, B. Camillonia, M. Basileoa, N. Sigismondib,
C. Fabianic, L. Calzolettic, S. Puzellic, I. Donatellic
aDepartment of Hygiene, University of Perugia, Via del Giochetto, I-06100 Perugia, Italy
bNursing Home Bartolomeo Castori, Foligno, Italy
cDepartment of Infectious, Parasitic and Immuno-Mediated Diseases, Istituto Superiore di Sanit` a, Roma, Italy
Available online 5 June 2006
This study examined the antibody response against the three vaccine antigens and the epidemic A/H3N2 drift variant (A/California) and
the prevention of laboratory diagnosed influenza infections in a group of elderly institutionalized people vaccinated with the 2004/2005
influenza vaccine. Antibody titres were measured by hemagglutination inhibition (HI) in sera collected before and 1 month after vaccination.
Laboratory diagnosis was done examining throat swabs (RT-PCR or MDCK cell culture) or by serology (seroconversion comparing HI titres
in sera collected 1 and 5 months after vaccination).
Results obtained showed that influenza vaccination induced an adequate immune response against the three vaccine antigens and the
epidemic A/H3N2 variant, however it was not capable of preventing an influenza outbreak due to the new A/H3N2 (A/California) variant.
© 2006 Elsevier Ltd. All rights reserved.
Keywords: Influenza; Vaccination; Antigenic mismatch
Influenza is a major cause of morbidity and mortality
worldwide, with most of the serious complications occur-
ring in the elderly population and particularly in individuals
living in high-density environments, such as nursing homes.
venting influenza and reducing the risk of complications in
those who are most vulnerable. However, vaccine efficacy
was found to be lower in elderly, especially in infirm per-
age-related decrease in immunoresponse to influenza vacci-
nation [1–4]. Moreover, vaccine efficacy is usually expected
to vary according to the match between the vaccine and epi-
demic strains [5,6].
Since the 2004/2005 winter season was characterized
by a prevalent circulation of an A/H3N2 drift variant
(A/California/7/04) from the vaccine strain (A/Fujian/
∗Corresponding author. Tel.: +39 075585 7313; fax: +39 075585 7317.
E-mail address: firstname.lastname@example.org (A.M. Iorio).
411/02), the aim of this study was to investigate the antibody
immune response, induced by immunization with a trivalent
influenza vaccine, against the three vaccine antigens, as
well as the epidemic A/H3N2 variant drift, in a group
of elderly people living in a nursing home. Moreover
we examined the ability of influenza vaccine to prevent
laboratory diagnosed influenza infection over the epidemic
2.1. Study design and vaccination
Sixty-seven volunteers living in a nursing home in
Umbria, a region of central Italy, received intramuscu-
larly one dose of commercially available trivalent influenza
2004/2005 MF-59 adjuvanted vaccine (FLUAD, Chiron) in
November 2004. Each dose of vaccine contained 15?g of
A/Fujian/411/02 (H3N2), A/New Caledonia/20/99 (H1N1)
and B/Shanghai/361/02. Blood samples were drawn before,
0264-410X/$ – see front matter © 2006 Elsevier Ltd. All rights reserved.
A.M. Iorio et al. / Vaccine 24 (2006) 6615–6619
1 month and, in most instances, 5 months after vaccination.
All subjects, or their relatives, gave informed consent to par-
ticipate in the trial.
2.2. Antibody assay
Titres of hemagglutination inhibiting (HI) antibodies to
the three influenza vaccine antigens and to the A/H3N2 drift
variant epidemic strain (A/California/7/04) were determined
simultaneously for serum samples taken from the same sub-
ject and frozen at −20◦C by a standard microtitre method
using 0.5% turkey erythrocytes and egg-grown influenza
viruses. All sera were treated with receptor-destroying
enzyme and heat-inactivated at 56◦C for 30min to remove
2.3. Vaccine immunogenicity
Vaccine immunogenicity was evaluated by comparing
titres in blood samples collected before and 1 month after
ered 1:5 for GMT calculation), mean fold increase (MFI) of
GMT (ratio of post-immunization titre to pre-immunization
titre), protection rate (numbers of volunteers showing HI
influenza infection) , positive responses (subjects with a
four-fold or greater increase in titre in prevaccination serum
positive people or from <1:10 to ≥1:20 in seronegative vol-
2.4. Laboratory diagnosis of influenza infection
The laboratory diagnosis of influenza was made either
on the basis of virus identification in throat swabs (RT-PCR
or cultivation in MDCK cells) or by serology, comparing
HI titres in pre-epidemic and post-epidemic sera, i.e. sera
collected respectively 1 and 5 months after vaccination. RT-
PCR was done using primers capable of amplifying A and B
nucleoprotein, A/H3 and A/H1 viral genes.
2.5. Antigenic characterization of isolated influenza
Antigenic characterization of influenza viruses grown in
MDCK cells and isolated from throat swabs was performed
by HI test using virus specific ferret anti-sera.
2.6. Phylogenetic analysis of influenza viruses
Amplified segments obtained by RT-PCR from some of
the influenza viruses isolated were purified and used for the
tor Cycle Sequencing” (Applied Biosystems). Phylogenetic
analysis of the HA1 domain of HA was determined, using
software package MEGA2, 2.1 version , comparing the
sequences of the isolated viruses with those found in the spe-
cific data bank (GenBank). “Kimura-2” distance method and
the “Neighbor-Joining” algorithm were used for the phylo-
genetic tree reconstruction.
3.1. Characteristics of the study subjects
females at risk for influenza due to age (mean age 84 years,
range 60–99) and health condition status. In fact, a history of
ditions) and of associated chronic use of drugs was reported
in 97% of these patients. All had received influenza vaccine
in the previous year.
3.2. HI antibody response to the three influenza vaccine
antigen components and to the A/H3N2 epidemic drift
Baseline and 30 days post-vaccination HI antibody titres
days after vaccination a statistically significant increase of
the values of GMT and of seroprotected volunteers was
observed not only against the three vaccine antigens, but also
HI antibody response in 67 elderly institutionalized volunteers vaccinated with 2004/2005 trivalent influenza vaccine against the epidemic drift H3N2 variant
(A/California) and the three vaccine antigens (A/Fujian, H3N2; A/New Caledonia, H1N1; B/Shanghai)
AntigenGMT %HI≥1:40 (N)
%Positive response (N)
PrePost (MFI) Post
A/California (H3N2 epidemic)
A/Fujian (H3N2 vaccine)
A/New Caledonia (H1N1)
ap<0.01 comparing A/California and A/Fujian (Student’s t- and χ2-tests).
**p<0.01 comparing pre- and post-vaccination values.
A.M. Iorio et al. / Vaccine 24 (2006) 6615–6619
cinated elderly subjects in the winter 2004/2005
aAntigenically equivalent to vaccine A/Fujian/411/02.
against the A/California/7/04 (H3N2) epidemic drift variant
circulating among the population in the world in 2004/2005
winter season. Comparing HI titres against A/H3N2 vac-
cine (A/Fujian) and epidemic strain (A/California) before
vaccination, GMT values and percentages of people with
protective levels (≥1:40) of HI antibodies against the vac-
epidemic strain (36.3 versus 30.9 and 56.7% versus 40.3%,
respectively, Table 1). After vaccination, statistically sig-
nificant higher values of GMT and of positive responses
were found against the vaccine strain as compared with epi-
demic antigen (252.1 versus 125.2 and 73.1% versus 55.2%,
respectively, Table 1). The percentages of seroprotected peo-
ple against vaccine and epidemic A/H3 strain were similar
(95.5% versus 89.5%, Table 1).
ability of influenza vaccination in elderly subjects (post-
response rate ≥30%) were in general satisfied against the
three vaccine strains and the epidemic A/H3 virus. The posi-
tive responses against the A/H1N1 vaccine component were
slightly lower than the requested 30% (28.3%, Table 1).
3.3. Laboratory confirmed influenza illness with virus
identification and characterization in some of the
In mid of February 2005, the occurrence of influenza like
syndrome was reported in many of the vaccinated elderly
people living in the nursing home. Throat swabs from 12
of the 67 vaccinated people were collected. Eight samples
were positive for A/H3 influenza virus when examined by
RT-PCR and five of them grew in MDCK cells. The viruses
isolated in MDCK cells were antigenically and genetically
The results of HI test using ferret specific antisera are
reported in Table 2. All the five viruses were antigenically
closely related to recent reference strains and, particularly,
to the new A/California/7/04 strain. However the values of
HI titres were not the same for the five viruses. Moreover,
all the five viruses showed cross-reactivity also against the
alent antigenic strain in the vaccine composition respect to
Fig. 1. Phylogenetic analysis of the HA1 nucleotide sequences of influenza A/H3N2 viruses isolated in the 2003/2004 and 2004/2005 winter seasons.
A.M. Iorio et al. / Vaccine 24 (2006) 6615–6619
The phylogenetic analysis of the HA1 nucleotide
sequences of two (A/Perugia/17/05 and A/Perugia/18/05)
of the five viruses grown in MDCK is reported in Fig. 1.
The two viruses showed a high genetic affinity with the
A/California/7/04 strain. Both the isolates were character-
common to other isolates collected during the 2004/2005
influenza season in different parts of Italy and reported in
italics in Fig. 1.
3.4. Serological confirmed influenza virus infection in
elderly vaccinated people
Patients were considered positive for serologically diag-
nosed influenza infection after vaccine administration if they
had a seroconversion on comparing HI antibody titres found
in sera collected 1 and 5 months after vaccination. Fifty-
seven of the 67 volunteers vaccinated were examined for HI
antibodies against the three vaccine antigens and the epi-
demic A/H3N2 strain. No seroconversion was found using
vaccine A/H1N1 and B antigens (data not shown). Twenty-
three volunteers (40%) were positive for influenza A/H3N2
virus infection (Table 3). All patients showed seroconver-
sion using A/H3 epidemic strain (A/California) as antigen,
whereas only 11 of the 23 volunteers showed significant
HI antibody increases using A/H3 vaccine strain (A/Fujian)
(Table 3, patients nos. 1–11).
The present report describes an outbreak of influenza
occurred in the 2004/2005 winter season, characterized
by a mismatch between the A/H3N2 vaccine component
(A/Fujian/411/02) and the major epidemic A/H3N2 strain
(A/California/7/04), in a group of vaccinated elderly peo-
ple living in a nursing home. The results were obtained
studying 67 elderly institutionalized and chronically
ill female volunteers vaccinated with the 2004/2005
MF-59 adjuvanted trivalent influenza vaccine (FLUAD,
According to previous data [9,10] and in spite of the pres-
impairing immune responses, administration of MF-59 adju-
vanted influenza vaccine to the elderly volunteers induced
an adequate HI antibody response (Table 1). Significant
increases in GMT values and in numbers of people with pro-
tective antibody titres were found 1 month after vaccination
and the responses generally satisfied the requirements of the
European Commission, not only against the three vaccine
antigens, but also against the A/H3N2 (A/California) drift
variant circulating in the 2004/2005 winter. After vaccina-
tion, protective antibody titres ranged from 76.1 to 95.5%
Seroconversions found in elderly vaccinated people comparing HI titres in
A/California epidemic H3N2 strains
PatientAntigen HI titre: 1 month HI titre: 5 months
A.M. Iorio et al. / Vaccine 24 (2006) 6615–6619 Download full-text
In spite of the satisfactory vaccine induced immune
response, an outbreak of influenza occurred in the winter
months following vaccination in the nursing home residents.
Influenza A/H3 viruses were identified in eight vaccinated
elderly people with influenza symptoms and the antigenic
and genetic characterization of some of the viruses showed
demic variant (Table 2 and Fig. 1). Moreover, using paired
serum samples collected in 57 of the 67 volunteers 1 and 5
months after vaccination, seroconversion against the A/H3
epidemic virus was found in 23 subjects (Table 3).
These results suggest that A/H3 vaccine antigen
(A/Fujian) was not capable of inducing protective immunity
against the A/H3 circulating influenza strain (A/California),
confirming data previously reported in other winter seasons
with a mismatch between vaccine and epidemic influenza
strains [5,6]. However HI titres considered to be protective
1 month after vaccination (20/23, 87%, Table 3), suggesting
the possibility that this level of antibody might not be able to
biologically neutralise virus infectivity of the A/H3N2 epi-
demic variant. Indeed, if the titre of 1:160, regarded as more
protective , is considered, only 6 of the 23 infected peo-
body titres for inducing protection was also suggested by the
fact that the HI GMT values found 1 month after vaccina-
tion in the 23 people with serologically diagnosed infection
in the 34 people not showing seroconversion (72.9 versus
159.0, p≤0.01) (data not shown).
The antigenic differences between A/H3N2 vaccine
A/Fujian strain and A/California epidemic strain, found
using ferret antisera (Table 2), were confirmed by the
statistically significant lower values found after vaccina-
tion of HI GMT increase and of positive response rate
against the A/California epidemic strain as compared to the
A/Fujian vaccine strain (Table 1). Indeed, analysing the HA1
nucleotide sequence of A/California/7/04 virus and compar-
ing results with A/Fujian/411/02, some amino acid changes
were found (Fig. 1) which might have influenced the anti-
genic cross reactivity between vaccine and epidemic A/H3
viruses. Interestingly, the asparagine residue at position 145
(K145N), which creates an additional glycosylation site in
antigenic site A, could have been responsible of a reduc-
tion in the antigenic cross reactivity with antisera to vaccine
In accordance with our results of low ability of the
A/H3N2 California strain, WHO has defined A/H3N2 Cal-
ifornia strain as an epidemic strain, although the latter only
shows minor genetic changes. This new virus variant has
been included in the antigenic components of the 2005/2006
winter season vaccine .
 Beyer WEP, Palache AM, Baljet M, Masurel N. Antibody induction
by influenza vaccines in the elderly: a review of the literature. Vaccine
 Gross PA, Hermogenes AW, Sacks HS, Lau J, Levandowski RA. Effi-
of the literature. Ann Intern Med 1995;123:518–27.
 Vu T, Farish S, Jenkins M, Kelly H. A meta-analysis of effectiveness
of influenza vaccine in persons aged 65 years and over living in the
community. Vaccine 2002;20:1831–6.
and effectiveness of influenza vaccines in elderly people: a systematic
review. Lancet 2005;366:1165–74.
 Mullooloy J, Bennett M, Hornbrook M, Barker W, Williams W, Patri-
arca, et al. Cost-effectiveness of influenza vaccination programs in an
HMO: the experience of Kaiser Permanente, North-west Region. In:
Hannoun C, Kendal AP, Klenk HD, Ruben FL, editors. Options for the
control of influenza II. Amsterdam: Elsevier; 1992. p. 53–62.
 De Jong JC, Beyer WEP, Palache AM, Rimmelzwaan GF, Osterhaus
ADME. Mismatch between the 1997/1998 influenza vaccine and the
major epidemic A(H3N2) virus strain as the cause of an inadequate
vaccine-induced antibody response to this strain in the elderly. J Med
 Palache AM. Dose-comparative serological studies: critical evaluation
of the parameters to assess the antibody response to influenza vaccina-
tion. In: Palache AM, editor. Influenza vaccination. The effect of dose
and age on the antibody response. Amsterdam: Broos Amsterdam BV;
1991. p. 83–119.
 Kumar S, Tamura K, Jakobsen IB, Nei M. MEGA: molecular evolu-
tionary genetic analysis, version 2.1. Pennsylvania State University,
University Park, PA 16802; 2001.
M, et al. Increased immunogenicity of the MF-59 adjuvanted influenza
Eur J Epidemiol 2001;17:135–40.
 Podda A. The adjuvanted influenza vaccines with novel adju-
vants: experience with the MF-59-adjuvanted vaccine. Vaccine
 Weekly Epidemiological Record. Recommended composition of
influenza virus vaccine for use in the 2005–2006 influenza season. No.
8, 2005; 80:65–76.