Background
Neisseria meningitidis is an obligate human commensal bacteria that colonizes the nasopharynx [
1]. Colonization is a prerequisite for transmission and for developing invasive meningococcal disease (IMD). However, IMD is a rare consequence of invasion and in most individuals the colonization leads to a period of asymptomatic carriage after which
N. meningitidis is naturally cleared, determining a serogroup-specific antibody response that acts as transitory immunization event [
2].
Asymptomatic meningococcal carriage is recognized as an age-dependent phenomenon, with prevalence increasing through childhood from 4.5% in infants to a peak of 23.7% in 19-year olds and subsequently decreasing in adulthood to 7.8% in 50-year olds [
3]. However, in closed/semi-closed settings and during outbreak-situations, the carriage rate may be higher [
2].
The duration of the carriage state varies; it may be chronic, lasting for several months, intermittent or transient, and depends on the properties of the colonizing-strain [
2]. One strain, the sequence-type 11 (ST11) clonal-complex (cc11) meningococci bearing serogroups C or W polysaccharide capsules, is recognized to have a high transmission and recovery rate to balance its short duration of carriage and inability to create a commensal relation with the host [
4].
In Italy, the incidence rate (IR) of IMD is among the lowest in Europe [
5,
6]. Until 2014, serogroup B was the main cause of IMD in the country; from 2015, serogroup C began predominating, determining the 44 and 43% of all IMD-cases in Italy in 2015 and 2016, respectively.
This sharp increase was driven by the Tuscany region that contributed for 48.4% (31/64) and 37.5% (30/80) of all IMD serogroup C cases (MenC) at national level in 2015 and 2016, respectively [
6,
7]. The increased incidence of MenC reported by Tuscany was due to a clonal expansion of the hyper-virulent
N. meningitidis strain C:P1.5–1,10–8:F3–6:ST-11(cc11) [
8], which mainly affected the age group 20–30 years, men who have sex with men, and the area located between the cities of Firenze, Prato and Empoli, with discos and gay-venues associated-clusters [
9]. As a consequence, in May 2015, a reactive immunization campaign offering a single dose of meningococcal serogroup C conjugate vaccine (MCC) or quadrivalent meningococcal conjugate vaccine (ACWY) to people aged 11–45 years, was implemented in Tuscany with the support of the Italian Ministry of Health (MoH). In addition, a targeted immunization programme for MSM was performed in late 2016.
Moreover, with the support of Italian National Institute of Health (Istituto Superiore di Sanità - ISS), the Tuscany Region conducted an outbreak investigation [
9], and a cross-sectional carriage survey, whose results are presented in this paper. The primary objectives of the survey were to estimate the meningococcal carriage prevalence during the outbreak and to identify related risk factors, in order to address public health interventions.
Results
A total of 2285 oropharyngeal samples were collected (95.2% of the planned sample size).
The overall prevalence of N. meningitidis carriage was 4.8% (n = 110; 95% CI 2.0–5.3%), and the majority carried nonencapsulated meningococci (2.3%; 95% CI 1.7–3.1%). Among encapsulated meningococci, serogroup B was the most prevalent (1.8%; 95% CI 1.3–2.4%; n = 41), followed by serogroup Y (0.5%; 95% CI 0.2–0.9%; n = 11) and serogroup C (0.2%; 95% CI 0.0–0.4; n = 4). One carrier of serogroup E and one of serogroup Z were also found (0.04%; 95% CI 0.0–0.2%).
At the MLST analysis performed on the 4 N. meningitidis serogroup C isolates, three individuals from the city of Empoli (outbreak-area) were found to carry the outbreak strain, C:ST-11 (cc11), and one form the city of Grosseto (control-area) carried a C:ST-41/44 clonal complex meningococci.
Additional file
1: Table S1 describes the study-population showing the meningococcal carriage prevalence by serogroups, demographic characteristics and risk factors.
Each study centre contributed equally to the sample size, with serogroup C most prevalent in Empoli (0.5%). The 4 serogroup C carriers were all aged 11–19 years, students, and attended disco/clubs/parties in the month before swab collection.
Participants who shared drinks in the month before swab collection had a higher serogroups B and C carriage prevalence (2.7 and 0.4%, respectively) compared with participants who did not (1.2 and 0.0%, respectively).
Higher carriage prevalence was observed for all serogroups among illicit-drugs users, with serogroup Y most prevalent (11.4%).
The majority of smokers carried nonencapsulated meningococci (7.2%), while the 4 serogroup C carriers were all exposed to second-hand smoke.
Those who in the month before swab collection had same-sex intercourses showed a higher meningococcal carriage prevalence than those who had heterosexual or had not sexual intercourses, with nonencapsulated meningococci (76.2%) and N. meningitidis serogroups B and Y most prevalent (76.2, 4.8 and 4.8%, respectively).
Additional file
2: Table S2 shows the results of the univariate and multivariate analyses of factors associated with meningococcal carriage status.
At the multivariate analysis including variables with p < 0.20 at the unviariate analysis, factors statistically significant associated with the risk of being meningococcal carriage were: illicit-drugs consumption (AOR 6.30; 95% CI 2.44–16.32; p < 0.01), active smoking (AOR 2.78; 95% CI 1.37–5.63; p = 0.01), disco/clubs/parties attendance (AOR 2.06; 95%CI 1.04–4.23; p = 0.04), the age-group (p < 0.01) but with a significant association only for those aged 20–30 years (AOR 3.08; 95% CI 1.97–9.79), as well as having had sexual intercourse (p < 0.01) but with a significant increased risk only for those who had same-sex intercourses (AOR 6.69; 95% CI 2.16–20.70).
Similar results were observed at the multivariate analysis with backward elimination at 10% level, where the same variables remained correlated with meningococcal carriage status in the same direction of association.
Discussion
This study shows meningococcal carriage prevalence and related risk factors during an outbreak due to the hypervirulent
N. meningitidis serogroup C ST-11 (cc11) strain occurred in Tuscany, Italy, during 2015–2016 [
8,
9]. Despite the outbreak, a low prevalence (0.2%) of serogroup C carriers was found. This finding is consistent with those of previous carriage surveys conducted during MenC-outbreaks, where serogroup C carriage prevalence found not to exceed 1%, being 0.3% in Canada [
27], 1.0% in The Netherlands [
28], 0.6% in Spain [
29], 1.0% in Denmark [
30], and 0.2% in Brazil [
31]. This low serogroup C carriage prevalence during MenC-outbreaks, could be partially explained both by the hypothesis of Trotter et al. [
32] that serogroup C strain have a high recovery and transmission rate, moving quickly through populations while maintaining a low prevalence, and by the hypothesis of Caugant et al. [
4] that MenC-outbreaks, in particular due to hypervirulent strains (as ST11/cc11), follow their rapid spread in a specific-population in terms of asymptomatic carriers of short duration, as suggested by fact that these outbreaks are characterized by the occurrence of MenC-cases in different locations of restricted geographical areas over short periods of time, without apparent epidemiological links.
Both hypotheses imply that during MenC-outbreaks, there are population-pockets where transmission occurs with people becoming serogroup C carrier for a short period of time. These population-pockets, are difficult to detect through prevalence-surveys because this study-design provides a limited snapshot of a population, that also due to biases, in particular selection-bias, may not be reliable of the real carriage prevalence [
1].
This limit is present in our study, because the convenience sample with recruitment at vaccination clinics, may have introduced selection biases by recruiting participants who were not in the affected population-pockets.
Moreover, the reactive vaccination campaign with MMC/ACWY, started 1-year before the study may have reduced the serogroup C meningococcal carriage prevalence, since meningococcal serogroup C conjugate vaccines have demonstrated a positive impact on reducing the carrier rate [
33]. Despite this, it should be of note, that before, during and after this study, the MenC-outbreak in Tuscany continued to occur [
9], suggesting that the bacterium was still circulating within specific population-pockets not surveyed.
On the other hand, a higher carriage prevalence due to nonencapsulated and serogroups B, Y meningococci was found, as in the aforementioned studies conducted during MenC-outbreak situations [
27‐
31], indicating a more commensal behaviour and a longer duration of the carriage status, that facilitates the identification of people carrying these strains during carriage surveys [
4,
32].
Differences in the meningococcal prevalence by serogroups, demographic characteristics and risk factors of the study population were observed.
In particular, according to the 2015–2016 MenC outbreak-investigation conducted by Miglietta et al. in Tuscany [
9], where clusters were recognized among teenagers attending discos, the 4 serogroup C carriers were all aged 11–19 years and attended discos/club/parties during the month before swab collection. On the other hand, in contrast with the result of the outbreak-investigation, where an involvement of the MSM-community was also identified, none of the serogroup C carriers reported MSM-behaviour. This mismatch may be due to the low number of serogroup C carriers found that did not allow to clearly identify risk factors for serogroup C carriers during the outbreak, also determining the need to group all the serotypes together for the risk factors analyses.
Higher serogroup C carriage prevalence was observed in the city of Empoli; in line with this finding, this city was the one with the higher attack-rate during the 2015–2016 MenC-outbreak [
9,
34]. In addition, the fact that the 3 carriers of the outbreak-strain (C:ST-11/cc11) were all from the city of Empoli and the one carrying a C:ST-41/44 clonal complex meningococci was found in Grosseto (control-area), strongly confirm the circulation of the outbreak-strain in the affected area.
Overall, risk behaviours identified through multivariate analyses confirmed what reported in the literature from carriage surveys. The age group 20-to-30 years was at increased risk for meningococcal carriage during outbreaks occurred in The Netherlands, Turkey, Wales, and in a Review of carriage studies conducted in Europe, and this is likely due to the convergence of risk behaviours among young adults [
2,
15,
27,
28].
IMD-outbreaks were reported among illicit-drugs users in the United States (US) [
35,
36] and this was a risk behaviour for meningococcal carriage reported in other surveys [
14,
17]. The mechanisms through which illicit-drug use may contribute to carriage, meningococcal transmission, or development of IMD are unknown: sharing illicit-drugs and cannabis-cigarettes use, as well as prolonged close contact among groups of illicit-drug users (that may represents population-pockets of meningococcal carriers), is likely to increase the risk [
36]. Active smoking is a well-recognized risk factor for meningococcal carriage as well as for other infections since it causes structural changes in the respiratory tract and a decrease in immune response [
2,
4,
13,
17].
Meningococcal outbreaks linked to discos attendance are reported in the literature [
37,
38] and it is also indicated as a risk factor for meningococcal carriage [
2,
4,
13,
17]. According to these studies, in addition to the crowded conditions, several risk behaviours may converge and cumulate in these places, from smoke to illicit-drug and alcohol use, drinks and cigarettes sharing, as well as intimate kissing with multiple partners.
Higher meningococcal carriage prevalence among MSM is reported in the literature [
39‐
41], and recently, several MenC-outbreaks involved gay, bisexual and other MSM in Europe and US [
16,
42]. What promote the spread of meningococci among this population-group is partially explained by the high prevalence of risk behaviours reported in the literature among MSM that facilities frequent close-contacts, but also to an unique mechanism of sexual-transmission or a specific-susceptibility [
16,
42].
No significant difference in the risk of being meningococcal carriage was found by place and month of swab collection. On this regard, the literature shows contrasting results [
27‐
31], and a clear relation between meningococcal carriage and IMD-IR has not yet been established.
Other risk factors for meningococcal carriage reported in the literature, as alcohol-consumption, drinks-sharing, bar attendance, upper respiratory infections, antibiotics consumption, occupation and gender [
2,
4,
14‐
17], were instead not found associated in our study or associated only at the univariate analysis.
Limits of this cross-sectional carriage survey are intrinsic to the study-design [
4,
32]; in particular the convenience sampling at vaccination clinics may have introduced selection biases, as the general population might be at lower risk of carriage compared to close contacts and/or population-pockets where transmission occurs; however, the swabbing of close-contacts of MenC-cases in other studies did not find higher carriage rates [
43,
44], supporting the hypothesis reported by Trotter et al. of the high recovery rate of
N. meningitidis serogroup C strain [
31]. Another limit of the study is represented by response biases that may have been introduced in sensitive questions. In addition the low number of serogroup C carriers found did not allow to identify specific risk factors and to clearly compare them with the results of the outbreak-investigation [
9].
On the other hand, strengths of this study are represented by the good stability and fit of the data of multivariate models as indicated by the very similar results produced. In addition, the combined use of PCR and culture allowed maximizing the sensitivity.
Acknowledgments
The authors would like to thank the Public Health Nurses (Assistenti Sanitari/Infermieri) who recruited participants at vaccination clinics.
Meningococcal Carriage Study Group
Alessandra Bagnoli, Giuseppe Boncompagni, Francesco Cipriani, Paolo Filidei, Giorgio Garofalo, Marinella Chiti, Gabriele Mazzoni, Astrid Mercone, Giovanna Mereu, Sabrina Novelli, Maria Grazia Santini, Maurizio Spagnesi.