Reduction in Neisseria meningitidis infection in Italy after Meningococcal C conjugate vaccine introduction: A time trend analysis of 1994–2012 series

Abstract

The incidence of invasive meningococcal disease (IMD) in Italy is among the lowest in Europe. Meningococcal C conjugate vaccine (MCC) was introduced in 2005 for 12 months old infants. The aim of this study was to describe the epidemiology of IMD in Italy from 1994 to 2012 and to evaluate the impact of MCC introduction. Data about Neisseria meningitidis (N. meningitidis) cases were drawn from the National Surveillance of Invasive Bacterial Diseases. The average incidence of IMD during 1994–2012 in Italy was 0.36 per 100,000 (95%CI 0.30; 0.40). N. meningitidis B was the most frequent serogroup and infants less than 12 months old were the most affected. Joinpoint analysis showed a statistically significant reduction in the incidence of N. meningitidis C related IMD after MCC introduction: the Annual Percentage Change declined from 21.8 (95%CI 15.1; 28.9) in 1994–2005 to −19.9 (95%CI −28.2; −10.7) afterwards. No changes were observed with respect to N. meningitidis B related IMD. Poisson regression showed a statistically significant reduction in the incidence of IMD both associated to N. meningitidis C (Incidence Rate Ratio 0.33; 95%CI 0.29; 0.37) and due to all serogroups (Incidence Rate Ratio 0.70; 95%CI 0.65; 0.75) in the post-vaccination period compared to the pre-vaccination one. On the other hand, the incidence of N. meningitidis B related IMD did not decrease. Our results suggest that MCC had an impact in decreasing the incidence of N. meningitidis C related IMD. However, data on typing are incomplete and efforts are needed to make them available for studying the need and the impact of other meningococcal vaccines.

Keywords:

• epidemiology
• meningococcal C conjugate Vaccine
• Neisseria meningitidis
• surveillance
• time analysis

Abbreviation

APC	=	Annual Percent Change
ECDC	=	European Center for Disease Prevention and Control
IMD	=	Invasive meningococcal disease
IRR	=	Incidence Rates Ratio
ISTAT	=	Italian National Institute of Statistics
JP	=	Joinpoint
MCC	=	meningococcal C conjugate vaccine
Neisseria meningitides	=	N. meningitidis
SIMI	=	Informative System of Infectious Diseases
STATA	=	Data Analysis and Statistical Software for Professionals.

Introduction

Invasive meningococcal disease (IMD) is an acute bacterial infection caused by Neisseria meningitidis (N. meningitidis) that may be considered rare in Europe but often severe and potentially life-threatening.¹

In Europe the overall notification rate for confirmed cases was 0.68 per 100,000 in 2012.¹ The majority of cases were due to serogroups C and B but the latter was dominant. The highest notification rates were reported in infants less than 12 months old (11.4 cases per 100,000) and in children from 1 to 4 y of age (3.7 cases per 100,000).¹

A decreasing trend in the incidence of IMD has been observed in Europe over the last 10 years, partly attributable to the introduction of meningococcal C conjugate vaccine (MCC) in the universal immunization schedules.²

In Italy, at the end of the 80s and in 90s the incidence of IMD was 0.5 per 100,000 per year in the general population, serogroup B was predominant and children from 1 to 4 y of age were the most affected.^3-7 A change in the epidemiological pattern of IMD was observed in Italy in the early 2000s, when a rapid increase of serogroup C related cases occurred between 2000 and 2005.⁸ A reduction of serogroup C related IMD was reported following the inclusion of MCC in the Italian National Immunization Plan. MCC was for the first time introduced in the Italian National Immunization Plan in 2005 and is currently provided in a single dose at 13–15 months of age in the whole country, with a catch-up at 11–18 y of age. ⁹ Notwithstanding this, universal vaccination campaigns started at different points in time. In 2010, the incidence of IMD in Italy was among the lowest in Europe (0.24 cases per 100,000) with infants under one year of age and children from 1 to 4 y of age showing the highest values (4.6 and 1 per 100,000 per year respectively) followed by teenagers and young adults.¹⁰

In recent years several countries in Europe have introduced MCC in their immunization schedules¹¹ and have subsequently experienced a substantial decline in the incidence of serogroup C related IMD.¹²

The objective of this study was to describe the epidemiology of IMD in Italy in 1994–2012 and to evaluate the impact of MCC introduction.

Results

The incidence of IMD during 1994–2012 ranged from 0.23 to 0.59 per 100,000. The total number of IMD cases was 3,928, with a minimum of 138 cases in 2012 and a maximum of 343 cases in 2004.

The overall mean incidence per year during the study period was 0.36 per 100,000 (95%CI 0.30; 0.40) (Table 1). Infants less than 12 months old were the most affected (4.29 per 100,000; 95%CI 3.70; 4.82), followed by children in the 1–4 age group (Table 1).

Table 1. - N. meningitidis mean incidence (per 100,000 inhabitants) by age and serogroup in the time period from 1994 to 2012

Age group	All serogroup (95% CI)	Serogroup B (95% CI)	Serogroup C (95% CI)	Other^* (95% CI)	Untyped (95% CI)
Less than 1 year	4.29 (3.70–4.82)	1.97 (1.60–2.33)	0.59 (0.32–0.91)	0.02 (0.00–0.04)	1.71 (1.29–2.14)
1–4 years	1.84 (1.51–2.16)	0.67 (0.57–0.75)	0.42 (0.21–0.62)	0.01 (0.00–0.02)	0.75 (0.55–0.94)
5–9 years	0.75 (0.63–0.87)	0.28 (0.23–0.33)	0.10 (0.05–0.13)	0.01 (0.00–0.02)	0.37 (0.28–0.47)
10–14 years	0.52 (0.42–0.62)	0.13 (0.09–0.16)	0.12 (0.07–0.19)	0.01 (0.00–0.02)	0.26 (0.21–0.32)
15–24 years	0.64 (0.52–0.75)	0.24 (0.14–0.54)	0.15 (0.09–0.19)	0.00 (0.00–0.00)	0.25 (0.19–0.31)
25–64 years	0.16 (0.13–0.19)	0.05 (0.04–0.05)	0.03 (0.02–0.04)	0.00 (0.00–0.00)	0.08 (0.06–0.09)
≥65 years	0.11 (0.09–0.13)	0.03 (0.02–0.04)	0.02 (0.01–0.03)	0.00 (0.00–0.00)	0.07 (0.05–0.08)
All age groups	0.36 (0.30–0.40)	0.12 (0.10–0.13)	0.07 (0.04–0.09)	0.01 (0.00–0.02)	0.15 (0.13–0.19)

* average incidence (per 100,000 inhabitants) for serogroups A, W135, X, Y in the time period from 2005 to 2012.

The time trend of IMD is shown in Figure 1 which demonstrates that the highest incidence was observed in infants less than 12 months old throughout the study period. A decrease in incidence of IMD started in 2004 in the youngest age groups, especially in those from 1 to 4 y of age which are the primary target of MCC vaccination campaign. The mean incidence decreased from 2.13 per 100,000 per year in 1994–2005 to 1.33 per 100,000 per year in 2006–2012.

Figure 1. incidence of IMD (per 100,000) by age group in Italy - 1994–2012.

With reference to serogroups, N. meningitidis B was more common than N. meningitidis C in all age groups, with a mean 2-fold higher incidence (Table 1).

From 1994 to 2012, a total of 2,248 (57%) meningococcal strains were typed. Among them, 1,371 (61%) belonged to serogroup B, 760 (34%) to serogroup C and 117 (5%) to “other,” i.e. A, W135, X, Y. The proportion of cases due to serogroup C increased gradually reaching the highest value in 2005 (n = 116 ; 35.5%) and decreased following the MCC introduction (Fig. 2) whereas the proportion of cases due to serogroup B has increased throughout time so that serogroup B is currently the most represented. Data on “other” serogroups related IMD were available since 2005. The trend of “other” serogroups related IMD has increased steadily and in 2011 the number of cases exceeded those due to serogroup C (22 vs. 21). The incidence of N. meningitidis by serogroup during the study period is shown in Figure 3. Comparing the mean incidence per year before (1994–2005) and after (2006–2012) the MCC introduction, a decrease was observed from 0.40 per 100,000 (95%CI 0.33; 0.46) to 0.28 per 100,000 (95%CI 0.24; 0.30) for all cases, from 0.07 per 100,000 (95%CI 0.03; 0.11) to 0.05 per 100,000 (95%CI 0.03; 0.08) for serogroup C related IMD and from 0.20 per 100,000 (95%CI 0.18–0.22) to 0.07 per 100,000 (95%CI 0.04–0.10) for untyped cases. On the contrary, the incidence remained stable for serogroup B (0.12 per 100,000 95%CI 0.10; 0.14 in both periods).

Figure 2. number of cases of IMD by serogroup in Italy - 1994–2012.

Figure 3. incidence of IMD (per 100,000) by serogroup in Italy - 1994–2012.

The Joinpoint (JP) analysis by age and serogroup (Table 2) showed a joinpoint in 2005 for N. meningitidis C in all age groups, with children from 1 to 4 y of age that registered the greatest reduction: the Annual Percentage Change (APC) decreased from 24.0 (95%CI 8.7; 41.3; p < 0.01) in 1994–2005 to −24.1 (95% CI −41.5; −1.6; p < 0.01) afterwards. On the contrary, no joinpoints were found for serogroup B related IMD with a no significant steady increase in all age groups (APC 0.9; 95%CI −1.3; 3.2; p = 0.3). Considering all serogroups, joinpoints were found with respect to the whole population and with age groups 1–4; 10–14; 15–24. Children from 1 to 4 y of age showed the greatest reduction: APC changed from 6.7 (95%CI 2.9; 10.7; p < 0.01) in 1994–2005 to −13.3 (95%CI −18.2; −8.1; p < 0.01) afterwards.

Table 2. – Results of the JP regression by age and serogroup

Age group	Serogroup	Joinpoint	Years range	APC (95%CI)	p-value
Less than 1 year	B	No JP	1994–2012	3.1 (−1.5; 7.8)	0.17
	C	2005	1994–2005	19.5 (6.2; 34.3)	<0.01
			2005–2012	−18.8 (−35.6;−2.5)	0.01
	Untyped	No JP	1994–2012	−8.8 (−13.0; 4.5)	0.27
	All serogroup	No JP	1994–2012	−1.5 (−3.9; 1.0)	0.20
1–4 years	B	No JP	1994–2012	1.3 (−1.2; 3.8)	0.28
	C	2005	1994–2005	24.0 (8.7; 41.3)	<0.01
			2005–2012	−24.1 (−41.5; −1.6)	<0.01
	Untyped	No JP	1994–2012	−5.8 (−9.7; 1.7)	0.08
	All serogroup	2004	1994–2004	6.7 (2.9; 10.7)	<0.01
			2004–2012	−13.3 (−18.2; −8.1)	<0.01
5–9 years	B	No JP	1994–2012	2.4 (−0.8; 5.6)	0.12
	C	2005	1994–2005	17.6 (8.6; 27.4)	<0.01
			2005–2012	−15.3 (−19.8; −7.4)	<0.01
	Untyped	No JP	1994–2012	−7.0 (−11.2; 2.5)	0.21
	All serogroup	No JP	1994–2012	−1.1 (−4.1; 2.1)	0.48
10–14 years	B	No JP	1994 – 2012	4.8 (−0.1; 10.0)	0.05
	C	2005	1994–2005	4.5 (2.0; 12.5)	0.02
			2005 – 2012	−18.4 (−23.4; −8.8)	0.02
	Untyped	No JP	1994 – 2012	−5.9 (−9.2; 13.6)	0.09
	All serogroup	2004	1994–2004	5.5 (−2.2; 13.9)	0.14
			2004–2012	−10.3 (−24.7; −6.8)	<0.01
15–24 years	B	No JP	1994–2012	1.0 (−2.6; 4.8)	0.54
	C	2005	1994–2005	16.8 (6.7; 27.8)	<0.01
			2005 – 2012	−16.1 (−29.8; −0.4)	0.04
	Untyped	No JP	1994–2012	−17.6 (−25.8; 8.4)	0.12
	All serogroup	2004	1994–2004	8.4 (2.3; 14.9)	0.01
			2004–2012	−10.1 (−19.9; −3.5)	<0.01
25–64 years	B	No JP	1994–2012	1.8 (−2.0; 5.9)	0.29
	C	2005	1994–2005	34.1 (17.6; 52.9)	<0.01
			2005–2012	−12.5 (−27.2; 5.1)	0.14
	Untyped	No JP	1994–2012	−18.7 (−25.4; 11.4)	0.24
	All serogroup	No JP	1994–2012	−10.7 (−16.5; 4.5)	0.18
≥ 65 years	B	No JP	1994–2012	5.2 (0.7; 10.0)	0.02
	C	No JP	1994–2012	3.6 (−4.4; 12.1)	0.34
	Untyped	No JP	1994–2012	−5.6 (−9.8; 1.1)	0.07
	All serogroup	No JP	1994–2012	−0.9 (−4.6; 3.1)	0.64
All age groups	B	No JP	1994–2012	0.9 (−1.3; 3.2)	0.35
	C	2005	1994–2005	21.8 (15.1; 28.9)	<0.01
			2005–2012	−19.9 (−28.2; −10.7)	<0.01
	Untyped	No JP	1994–2012	−10.1 (−26.1; 13.6)	0.15
	All serogroup	2004	1994–2004	5.5 (1.5; 9.6)	<0.01
			2004–2012	−9.6 (−14.4; −4.6)	<0.01

APC: annual percentage change; CI: confidence interval; JP: joinpoint.

The results of Poisson regression (Table 3) showed a statistically significant decrease in the incidence of N. meningitidis IMD after MCC introduction in all age groups. In particular, considering all serogroups and all age groups, N. meningitidis decreased by 30% (IRR 0.70; 95%CI 0.65; 0.75; p < 0.01) after MCC vaccination introduction. As far as serogroup C related IMD, reduction in the incidence was estimated 70% in infants less than 12 months old (IRR 0.30; 95%CI 0.20; 0.45; p < 0.01) and in children from 1 to 4 y of age (IRR 0.30; 95%CI 0.23; 0.43; p < 0.01).

Table 3. – Results of the Poisson regression by age and serogroup

Age group	Serogroup	Incidence Rate Ratio (95%CI)	p-value
Less than 1 year	B	1.26 (0.95 – 1.66)	0.09
	C	0.30 (0.20 – 0.45)	<0.01
	Untyped	0.73 (0.59 – 0.89)	0.02
	All serogroup	0.69 (0.39 – 0.80)	0.03
1–4 years	B	1.04 (0.81 – 1–32)	0.32
	C	0.30 (0.23 – 0.43)	<0.01
	Untyped	0.66 (0.47 – 0.91)	0.01
	All serogroup	0.62 (0.53 – 0.73)	<0.01
5–9 years	B	1.18 (0.85–1.65)	0.3
	C	0.38 (0.27–0.55)	<0.01
	Untyped	0.67 (0.54 – 0.83)	<0.01
	All serogroup	0.48 (0.24–0.94)	0.03
10–14 years	B	1.43 (0.90 – 2.29)	0.12
	C	0.50 (0.34–0.75)	<0.01
	Untyped	1.05 (0.63–1.74)	0.2
	All serogroup	0.87 (0.68 – 0.92)	0.03
15–24 years	B	0.96 (0.75–1.23)	0.7
	C	0.44 (0.33–0.58)	<0.01
	Untyped	0.89 (0.65–1–22)	0.4
	All serogroup	0.73 (0.62–0.85)	<0.01
25–64 years	B	0.89 (0.71–1–13)	0.3
	C	0.47 (0.38 – 0.58)	<0.01
	Untyped	1.27 (0.96–1.69)	0.08
	All serogroup	0.76 (0.66–0.87)	<0.01
≥ 65 years	B	1.82 (0.99–2.99)	0.08
	C	0.79 (0.55–0.99)	0.04
	Untyped	0.78 (0.40–1.55)	0.4
	All serogroup	0.75 (0.57–0.99)	0.04
All age groups	B	1.04 (0.93–1.15)	0.4
	C	0.33 (0.29–0.37)	<0.01
	Untyped	0.84 (0.72–0.98)	0.02
	All serogroup	0.70 (0.65–0.75)	<0.01

On the other hand, the incidence of N. meningitidis B related IMD did not decrease after MCC introduction.

Discussion

This study showed a significant decrease of IMD in all age groups after MCC introduction. In particular, the reduction of N. meningitidis C serogroup related IMD was demonstrated from 2005 onward and especially in children from 1 to 4 y of age who are the primary target of vaccination campaign in Italy. The reduction is likely due to the effect of MCC introduction but may also be explained partly by the epidemiological natural fluctuations of the disease. It is well known that the distribution of meningococcal serogroups changes over time. With respect to serogroup C, an increasing predominance was observed in late 90s and early 2000s also in other European countries.^11-14 As also observed by Mooney et al.¹⁴, the reversal of the increasing trend of serogroup C related IMD was simultaneous with the timing of MCC introduction and, therefore, likely attributable to it. The impact of MCC was investigated in many other European and Italian regional studies ^14-25 which showed an important decrease in N. meningitidis C related IMD and a vaccine effectiveness as high as 100% depending on the age group and on the time elapsed from vaccination. A downward trend in serogroup C related IMD was also confirmed by the European Center for Disease Prevention and Control (ECDC) in countries with MCC vaccination. On the contrary, a stable trend was reported for countries without MCC vaccination.¹

Another Italian study performed in Emilia Romagna Region²² highlighted a 70% reduction in the incidence of N. meningitidis C related IMD after MCC introduction. This result is in line with that yielded by our Poisson model and bears the attention to the importance of attaining good vaccination coverage in order to further reduce the incidence of N. meningitidis IMD. In this respect, vaccination campaign could be revised. Although Regions started to deliver MCC from 2005, the vaccination became free of charge from 2009–2010.²⁶ Therefore, vaccination coverage is deeply heterogeneous among Italian Regions, ranging from 37.4% to 93.2% in 24 months old children in 2012.²⁶ Results are very far from the 95% coverage goal set by the 2012–2014 National Immunization Plan.⁹ In this light, catch up programs could be strengthened because it is well known that the impact of vaccine is also linked to its efficacy against carriage in older age groups and therefore to herd immunity.^27-30

As far as the other serogroups are concerned, our study showed that the incidence of serogroup B related IMD remained stable in the pre and post vaccination era even though the relative proportion of cases due to N. meningitidis B has increased over time. The incidence of IMD due to serogroups A, W135, X, Y has increased during 2005–2012.

The increase of non-B non-C serogroups after MCC introduction was reported also in studies performed in the Netherlands and UK,^31,32 but may also be due to the improvement of surveillance systems. It should be observed that also the ECDC reported an increase of notification rates for serogroup Y and W.¹ With respect to serogroup B related IMD, the ECDC reported an overall steadily decreasing trend in notification rates between 2008 and 2012.¹ Since the distribution of serogroups varied between countries, not all the European countries reported a similar trend. For example, the incidence of serogroup B related IMD has slightly increased between 2010 and 2012 in Sweden.³³

A limit of our study is represented by the high proportion of untyped cases. The problem of non-typing is common to other surveillance systems of EU Member States and reported by literature.^11,15 In Italy, following the adoption of the new protocol for the National Surveillance of Invasive Bacterial Diseases in 2007, the completeness of data on typing has improved. The average number of untyped cases dropped from 50.9% in 1994–2007 to 23.4% in 2008–2012. Notwithstanding, the incompleteness of typing information implies that the incidence of N. meningitidis by serogroup is underestimated even though we may expect that the distribution of different N. meningitidis serogroups among untyped cases resembles each year the known one. The availability of vaccines against N. meningitidis B and A, C, W135, Y requires much attention on typing in order to properly assess their need and impact. Considering epidemiological data from Italy, it appears that a relevant benefit could be expected with the implementation of meningococcal B vaccination. In fact, meningococcal B related IMD did not decrease over time and this epidemiological trend was also observed in other countries which introduced MCC.¹⁶ Mooney et al.¹⁴ particularly suggested that the observed gradual increase in the incidence of serogroup B related IMD is not likely associated to MCC introduction but it is more a reflection of a natural fluctuation and, as a consequence, could benefit of vaccine introduction. However, because of the different impact of protein-based and conjugate polysaccharide vaccines on carriage, it is unlikely that IMD will be completely eradicated.³⁴

Another limitation of the study is represented by the lack of data on MCC coverage which are necessary to better evaluate the impact of the vaccine. In this respect, the literature on immunization coverage of children 12–24 months old is still scant^26,35,36 but the periodic collection of data would be necessary. Our study did not investigate the difference in MCC impact across Regions which have and have not implemented catch up programs. This analysis was not done because of the well-known problem of underreporting and under ascertainment across Regions.

In conclusion, the results of our study, which is the first at the Italian national level to evaluate the time trend of IMD and the impact of MCC using 2 different statistical approaches, suggest that MCC had an impact on the epidemiology of IMD in Italy. The work contributes to the growing body of evidence about the reduced incidence of IMD since the implementation of vaccination programs. The study also highlights that further efforts are needed to increase the proportion of typed cases and to provide regularly update on vaccination coverage. This information would enable better assessment of changes in the distribution of serogroups and the impact of vaccination campaigns, especially now that a vaccine against serogroup B is available.³⁷

Materials and Methods

In Italy, the surveillance of invasive diseases due to N. meningitidis derives from 2 sources of data: the Informative System of Infectious Diseases (SIMI) since 1991, and the National surveillance of Bacterial Meningitis since 1994.³⁸ SIMI receives the statutory notifications of meningitis due to N. meningitidis. Cases are notified by health professionals to the Local Health Units that transfer data to the regional and national authorities. The National surveillance of Bacterial Meningitis is a parallel surveillance system which requires all hospitals to report any confirmed case of meningitis from S. pneumoniae, N. meningitidis and H. influenzae. Limited to meningococcal diseases, the National surveillance of Bacterial Meningitis has also included sepsis from the beginning (1994). In 2007, in order to expand the surveillance to all forms of invasive bacterial diseases, the protocol for data collection was amended and the system was renamed National Surveillance of Invasive Bacterial Diseases.³⁸ The system currently collects data about confirmed case of invasive diseases from S. pneumoniae, N. meningitidis or H. influenzae. As for case definition, Italy has adopted the 2008 EU case definition³⁹ and has recently moved to using that of 2012.⁴⁰ Cases are defined according to clinical and laboratory criteria. A confirmed case is defined as a patient with a compatible clinical illness and a laboratory confirmation such as the detection - through microscopic direct examination, culture or polymerase chain reaction of N. meningitidis from a normally sterile site or the identification of the polysaccharide antigen in the cerebrospinal fluid.

Information about IMD was drawn from the National Surveillance of Invasive Bacterial Diseases which comprises cases from 1994 onward stratified by age groups (<1 ; 1–4; 5–9; 10–14; 15–24; 25–64; ≥ 65 ) and serogroups. Hospital laboratories are entrusted in typing and samples are sent to the National Institute for Health (Istituto Superiore di Sanità) for further microbiological characterization. Information on typing was available only for serogroups B and C until 2005. Since 2005, cases due to non-B and non-C serogroups were reported under “other.”

With respect to population at risk for N. meningitidis, data were collected from the Italian National Institute of Statistics (ISTAT) which provides resident population at the beginning of the year stratified by age groups. The incidence of IMD was computed dividing the number of cases by the population at risk; incidence was calculated with respect to age groups (<1 ; 1–4; 5–9; 10–14; 15–24; 25–64; ≥ 65 ) and serogroups. The detection of significant change in the incidence of IMD was performed by JP regression according to Kim's method.⁴¹ Time changes were expressed in terms of Annual Percent Change (APC) with 95% Confidence Intervals (95%CI). The null hypothesis was tested using a maximum of 3 changes in the slope with an overall significance level of 0.05 divided by the number of joinpoints in the final model. Joinpoint Regression Program version 4.0.1 was used in order to carry out the analysis. The impact of MCC was evaluated using Poisson regression model. Considering the lack of data for MCC coverage, a binary covariate, intended as absence/presence of MCC vaccination program, was used: absence for 1995–2005; presence for 2006–2012. The impact of vaccination program was assumed to be not immediate (2005) because Italian Regions started to provide MCC at different point in time. The output of Poisson regression model was represented by the IMD log relative risk as a linear function of the values assumed by independent variables.^42-44 Results were expressed in terms of Incidence Rates Ratio (IRR) with 95%CI. Poisson regression was performed by STATA. JP and Poisson analyses were carried out with respect to all age groups and serogroups together as well as stratifying for them. Significance level was set at 0.05.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Ethical Statements

The study was carried out according to the Helsinki Declaration of 1975. Ethical approval was not required because the study was based on data routinely collected.

Acknowledgments

We would like to thank Esther Vitto for language revision.