Introduction
A change in the epidemiological pattern of meningococcal disease was observed
in Spain and other European countries in the mid-1990s [1,2]. The median
incidence of meningococcal disease was 2.9/100 000 for the five year period
1991-1996 and serogroup B was the most frequent among all the confirmed
infections. In the epidemiological year 1996/1997, however, serogroup
C became the predominant group (63%) in almost all Spanish regions, with
a consequent increase in incidence and mortality. The incidence reached
5.8/100 000 with incidence due to serogroup C at 2.3/100 000. To reduce
this incidence, a nationwide vaccination campaign (which included 16 out
of the 19 autonomous Spanish regions) using anti-meningococcal A+C polysaccharide
vaccine was launched in autumn 1997. The overall estimated coverage was
76.3%. This strategy reduced the national incidence of meningococcal disease
by 45%. A reduction was seen in all age groups, and the most important
reduction was found in 2 to 19 year olds, the target group of the intervention.
In this group, the number of serogroup C cases fell by 76% in comparison
with the year before vaccination was introduced. However, the incidence
of meningococcal disease caused by serogroup C continued to increase in
the years following vaccination [3], a foreseeable circumstance given
the limitations in the immunogenicity of polysaccharide vaccine [4,5].
The new conjugate vaccine became available in Spain in 2000, and was included
in the infant vaccination schedule. A catch-up campaign was carried out
aimed at the most vulnerable group: children under the age of six years.
In three of the 19 national regions, this group was extended to include
all those under 19 years of age over the next three years. This study
analyses meningococcal disease surveillance data from the three epidemiological
years from 2000/2001 and 2002/2003 following the introduction of the meningococcal
serogroup C (MenC) conjugate vaccine in Spain, and includes data on incidence
in the different age groups and on the characterisation of strains isolated
in clinical cases. In this sense, the main interest in characterising
these strains lies in verifying whether the vaccination with the MenC
conjugate vaccine carried out during the last months of 2000 led to the
selection of new antigenic variants as a result of recombination or shift
of group B or C percentages, as well as an increase in the number of cases
caused by serogroups other than B and C.
Methods
Epidemiological surveillance of meningococcal disease in Spain is based
on a passive notification system. The weekly reporting of cases diagnosed
is compulsory and physicians must complete a questionnaire for every case
notified with the patient's demographic, clinical and epidemiological
data. The epidemiologist updates the information on outcome and vaccination
status a few days after the case is notified.
Data for the calculation of global and specific incidences by age during
the years 1999/2000 to 2002/2003 were obtained from the cases notified
to the Sistema de Enfermedades de Declaraci?n Obligatoria (EDO, Compulsory
Disease Reporting System). For notification purposes, a probable case
is defined as a patient who presents with clinical symptoms compatible
with the disease and a presumptive analytical test (such as the presence
of intracellular Gram negative diplococci in cerebrospinal fluid or other
biochemistry analyses). In 2000, the definition of a confirmed case was
modified to include both isolation of Neisseria meningitidis at
a normally sterile site and the presence of meningococcal DNA or the detection
of meningococcal antigen in the appropriate samples.
In this study, the epidemiological year (also known as 'season') runs
from week 27 of a particular year until week 26 of the next. Calculation
of national and age-specific meningococcal disease incidences was made
using population estimates made mid-year by the National Statistics Institute
(Instituto Nacional de Estadística). The risk of suffering from
the disease during the previous epidemiological year in the study was
calculated, (using relative risk and a 95% confidence interval,) by comparing
the current season with each of the previous epidemiological seasons.
Vaccine failure cases were studied from 1 January 2001 to 31 December
2003. A confirmed vaccine failure was defined as a confirmed case of serogroup
C disease with onset more than 14 days after the last dose of vaccine
scheduled for that age group.
To analyse the specific characteristics of the meningococcal strains in
all isolates, serogroup, serotype and serosubtype were determined using
specific monoclonal antibodies according to techniques described elsewhere
[6]. Furthermore, in order to study the possible appearance of capsular
genetic interchange phenomena, multilocus sequence typing (MLST) [7] was
used to define precisely the clone lines to which the suspected strains
belonged.
Results
In the epidemiological year 2002/2003, 948 cases of meningococcal disease
were reported (2.3/100 000), of which 76% were confirmed cases. The
percentage of confirmed cases during the study's final season was higher
than that of the previous three seasons (71% on average), and 468 serogroup
B (1.2/100 000) and 175 serogroup C cases (0.4/100 000) were notified
[TABLE 1]. The number of serogroup C cases notified fell during the
three seasons after the introduction of the vaccine, except for a slight
increase in the rates both for this serogroup and for the others in
the season 2001/2002 [FIGURE 1, TABLE 1].
The risk of suffering from serogroup C disease during the 2002/2003
season was 25% less than the previous season and 58% less compared with
the season before vaccination. The risk of suffering from serogroup
C disease was lower compared with that observed during each of the previous
seasons except for 2000/2001. For the other serogroup categories, although
the point estimates indicated a decrease in risk when comparing the
last epidemiological season with each of the previous ones, the upper
confidence interval is compatible with increasing risks. On the other
hand, the incidence of non-groupable cases increased gradually throughout
the study period and the risk was 59% higher in the season 2002/2003
if we compare it with the season 1999/2000. This increase is statistically
significant [TABLES 1 and 2].
A decrease in the number of serogroup C cases [FIGURE 2] in children
under ten years was observed during the last three seasons in this study.
These children were either born after the conjugate vaccine was included
in the routine vaccination schedule or were part of the target group
for the catch-up campaign. In the epidemiological year 2002/2003, 38
cases due to serogroup C (1.0/100 000) were reported in children under
-ten years, compared with 254 cases (6.6/100 000) during the season
before Menc C conjugate vaccine was introduced, which represents an
85% reduction in the incidence at this age. On the other hand, although
not statistically significant, there was an increase in the incidence
in 10 to 14 -year olds.
The incidence of serogroup B disease in children under one year old
in the 2000/2001 season decreased significantly in relation to the previous
epidemiological year [FIGURE 3]. The distribution by age of serogroup
B disease reflects the usual age distribution pattern with a large incidence
for children under five years.
Information on the clinical presentation of the disease (sepsis, meningitis,
or both) was registered for 95% of cases during the season 2002/2003.
Clinical sepsis was present in 46.9% of cases, and clinical meningitis
in 38.4% (p = 0.01). Sepsis and meningitis together were present in
10% of the cases. The percentage of cases with sepsis was greater in
serogroup C cases (52%) than in serogroup B cases (49%), but the difference
is not statistically significant.
Outcome is known for more than 95% of cases. In the epidemiological
year 2002/2003, 88 deaths due to meningococcal disease were reported.
During the same period, the number of deaths and the case fatality rate
(CFR) for the most important serogroups were 33 and 7.5% for serogroup
B, and 29 and 16.6% for serogroup C. Table 3 shows the number of deaths
due to serogroup C disease by age before and after the introduction
of the MenC conjugate vaccine. There was a continuous decrease in the
mortality caused by serogroup C disease after vaccination, except in
2001/2002. Three deaths due to serogroup C occurred in the age groups
targeted for vaccination in 2002/2003, compared with 30 deaths that
occurred in the same age groups in the season prior to the 2000 vaccine
campaign was launched. One of the three deaths was in a child who had
received the complete course of three doses of conjugate vaccine in
the routine vaccination programme. In the epidemiological year 2002/2003,
there was a 36% reduction in CFR compared with that registered before
vaccination.
Coverage in the catch-up and routine immunisation programmes in 2001,
2002 and 2003 was estimated on the basis of coverage data from nine
Spanish autonomous regions, which together account for 60% of the total
Spanish population. The proportion of the vaccinated population in the
catch-up displayed considerable homogeneity for children born between
1995 and 2000, with values above 92% for these birth cohorts. For the
2001, 2002 and 2003 routine childhood immunisation programmes, coverage
values ranged from 90% to 95%.
During the epidemiological years after the introduction of the conjugate
vaccine in the national schedule, 111 cases of serogroup C disease were
reported throughout the country in patients in the vaccine target group
(including those regions where the catch-up was extended to adolescents).
Twenty three of them had received the vaccine, 78 had not, and the vaccination
status was unknown in 10 cases. The number of cases for which the vaccination
status was unknown fell during the last three seasons and status information
is available for all cases during the 2002/2003 season. Of the 23 vaccinated
cases, 15 were classified as confirmed vaccine failures (accomplished
complete vaccination) and eight as probable failures (incomplete vaccination
or fewer than 14 days between vaccination and first symptoms) [TABLE
4].
Eleven of the 15 confirmed vaccination failures occurred during the
season 2002/2003. Ten of these children were between one and four years
of age, and only one of these cases had previously received the A+C
polysaccharide vaccine during the 1997 vaccination campaign. Of the
15 vaccination failures, seven had received the vaccination during the
catch-up campaign in 2000. The eight remaining cases had been vaccinated
according to the routine vaccination schedule established, and all but
one had received the vaccination in 2001. Failures were distributed
throughout the country. According to figures provided by pharmaceutical
companies on the number of doses of conjugate vaccine sold in Spain,
we estimate that during the period 2000-2003, 2.6 vaccine confirmed
failures per one million doses occurred.
During the study period, 2113 strains of Neisseria meningitidis isolated
from clinical cases with symptoms of meningitis and/or sepsis were received
for characterisation by the national reference laboratory. The proportion
of serogroup B cases increased gradually from 58.3% in 2000 to 68% in
2001 and 2002, reaching 72.9% during the first nine months of 2003.
The frequency of serogroup C strains fell from 38.5% in 2000, to 27%
in 2001, 22.6% in 2002, and finally to 20% in 2003. The percentage for
2003 corresponds to that observed in Spain during the 1980s, before
the increase in serogroup C disease. Only slight increases were observed
in the frequency of serogroups Y and W135, with the former rising from
1.2% in the year 2000 to 2.8% in 2003, and the latter from 1% to 1.9%.
These increases are not statistically significant.
The serosubtypes which appear in the strains of serogroup C have undergone
a considerable change. C:2a increased from around 20% in 2000 to 56%
in 2003, a trend which began in 1999 before the new vaccine was introduced.
The serosubtypes associated with serogroup B have not undergone modifications.
4:P1.15 strains have predominated (25%). Increases in non-typable strains
(42%) were observed. This high percentage of non-typable strains is
increasing, although non-serosubtypable strains only reach around 14%.
The isolation in the north of Spain of several B:2a:P1.5 strains (ST-11
lineage) is worthy to note. These may have their origin in C:2a:P1.5
strains which, after undergoing genetic recombination at the capsular
operon level, express serogroup B capsule.
Discussion
The incidence of meningococcal disease, especially serogroup C, has
fallen sharply during the last three epidemiological seasons in Spain
covered by this study. It has been estimated that the risk of contracting
the disease of this serogroup fell by 58% if we compare the incidence
of the last epidemiological year in the study with that of the season
before the conjugate vaccine was introduced. Nevertheless, we must bear
in mind that the rates for the season 1999/2000, especially those of
serogroup C, had already fallen considerably due to the vaccination
campaign with polysaccharide vaccine in 1997. The increased incidence
in the 10-14 year age group, although not statistically significant,
deserves special mention, given that only about 35% of the population
in this group has been vaccinated. No increase was observed among others
adolescents or in young adults. Nevertheless, a continuous monitoring
of the incidence in these age groups is needed in order to evaluate
the current vaccination policy.
The proportion of vaccine failures detected is similar to that observed
in other countries [8]. However, a large number of the confirmed vaccination
failures occurred during the epidemiological year 2002/2003. Also, 8
out of 15 (53%) confirmed vaccine failures occurred in children who
had been routinely immunised in infancy. These aspects could be related
to a loss of protection with time since vaccination. The study currently
being carried out on vaccine effectiveness may help clarify the issue.
No relationship has been found between vaccination failures and a previous
history of having received the polysaccharide vaccine.
The incidence of meningococcal disease caused by serogroup B and other
serogroups has remained stable during the last three seasons. This circumstance
is compatible with the cyclical evolution of the disease and suggests
that we are in an interepidemic period. It can also be concluded that
the analysis of the situation immediately after the introduction of
this new conjugate vaccine, which could have led to the appearance of
a new bacterial equilibrium, shows that there is no evidence of alterations
in the populations of circulating meningococci. However, special mention
must be made of the isolation of B:2a:P1.5 strains, previously mentioned
in the results section. These strains could have relevant epidemic potential
and, after two years of evolution, they are now being isolated in more
areas of Spain, although only in small numbers. Future evolution of
this strain will enable us to analyse the real importance of these processes
of active immunisation in the selection of this type of antigenic recombinants.
We do not have any explanation for the increase in the number of non-groupable
cases. The current use of microbiological techniques cannot be evaluated
here, as we do not have information on the techniques used in the diagnosis
of each case.
Finally, we must stress the importance of maintaining epidemiological
surveillance of this disease, as well as on improving the quality of
the information collected from each case. This will enable us to observe
changes in the presentation pattern of the disease, and in the identification
of vaccination failures with a view to reviewing the functioning of
current prevention programmes.
Acknowledgements
B. Alcalá was a postdoctoral fellowship at the Instituto de Salud
Carlos III. Fellowships from Wyeth España supported C. Salcedo
and A. Larrauri.
We would like to extend thanks to the epidemiologists and public health
officials responsible for the epidemiological surveillance and control
in the 19 Spanish autonomous regions who collected and reported the data
necessary for accomplishing the task which results we present in this
paper.
|