Case-based surveillance of bacterial meningitis following MenAfrivac™ introduction at Sourô Sanou University Hospital, Bobo-Dioulasso, Burkina Faso, 2011-2021

Highlights

• •The large sample size strengthened the robustness of the study and emphasized the need for multivalent vaccines against non-A serogroups (W, X, C, Y) that continue to drive epidemics in Burkina Faso and the African meningitis belt.
• •The introduction of reverse transcription-polymerase chain reaction has significantly enhanced the sensitivity and specificity of bacterial meningitis diagnosis compared with conventional methods, such as Gram stain, latex agglutination, and culture.
• •The implementation of metagenomic analyses is proposed to resolve cases with undetermined serogroups.
• •Findings also support the relevance of using pneumococcal vaccines in the country.

Abstract

Objectives

To describe the distribution of Neisseria meningitidis serogroups in bacterial meningitis cases diagnosed by polymerase chain reaction at the Sourô Sanou University Hospital, Bobo-Dioulasso, from 2011 to 2021.

Methods

A descriptive cross-sectional study was conducted using retrospective data from cerebrospinal fluid samples collected in Bobo-Dioulasso and surrounding districts and analyzed at the National Reference Laboratory for bacterial meningitis.

Results

Among 3823 suspected cases, 3341 samples underwent polymerase chain reaction, confirming 1186 cases (35.5%). The highest number of samples was recorded in 2012 (32.9%), mainly from Dafra district (20.2%). Streptococcus pneumoniae was the leading pathogen (51.9%), followed by N. meningitidis (46.1%) and Haemophilus influenzae type b (1.9%). Most N. meningitidis cases (69.5%) occurred in 2012, declining thereafter, with no cases by 2021. Serogroup W predominated (69.7%), while serogroups A and B were absent.

Conclusions

The results emphasize the need to strengthen epidemiological surveillance and adapt vaccination strategies to achieve meningitis elimination in Africa.

Introduction

Bacterial meningitis remains a major public health concern. Approximately one in 10 affected individuals dies, while one in five develops severe complications such as cognitive impairment, hearing loss, or limb paralysis [1,2]. According to the World Health Organization, bacterial meningitis causes around 1.2 million cases and 135,000 deaths annually worldwide [3], with the highest incidence and mortality rates occurring in low- and middle-income countries [4].

In Africa, more than 500 million people are at risk of seasonal meningitis epidemics [5]. These recurrent large-scale outbreaks mainly occur within the so-called “meningitis belt” of sub-Saharan Africa [6]. In 2020 alone, countries of the meningitis belt reported 19,552 new cases and 885 deaths [3].

Burkina Faso, located at the center of the meningitis belt, regularly experiences major epidemics. The most severe occurred in 1996, with 42,967 reported cases caused by Neisseria meningitidis serogroup A [5]. During the 2012 epidemic, 7022 suspected cases were reported, including 739 deaths [7], while in 2019, 1841 cases with 130 deaths were recorded [8].

Meningitis is largely preventable through vaccination [2]. Before 2010, N. meningitidis serogroup A was responsible for nearly 90% of epidemic cases, despite the availability of non-conjugated polysaccharide vaccines [9]. The introduction of the MenAfriVac® conjugate vaccine dramatically reduced the burden of serogroup A meningitis and reshaped the epidemiological landscape. Since then, epidemics have increasingly involved serogroups C, W, X, and Streptococcus pneumoniae [3].

To more accurately assess the impact of the MenAfriVac® conjugate vaccine, case-based surveillance of bacterial meningitis was established in the countries of the African meningitis belt. This surveillance system is supported by reference laboratories, which are responsible for confirming suspected cases using standard diagnostic methods such as Gram staining, latex agglutination tests, and culture. To enhance the epidemiological surveillance, multiplex gene amplification techniques, including real-time reverse transcription-polymerase chain reaction (RT-PCR), are also employed. RT-PCR was implemented in October 2011 at the Sourô Sanou University Hospital (CHUSS) in Bobo-Dioulasso, with the objective of strengthening case-by-case meningitis surveillance activities [5].

This study aimed to describe the distribution of N. meningitidis serogroups in bacterial meningitis cases detected by direct PCR at the Bacteriology-Virology Laboratory of Sourô Sanou University Hospital, Bobo-Dioulasso, from 2011 to 2021, following the introduction of the meningococcal A conjugate vaccine in 2010.

Methods

Type and setting of study

The study was conducted in Bobo-Dioulasso, an urban commune in Houet Province, located in the Hauts-Bassins region of southwestern Burkina Faso. CHUSS, the regional referral center, also serves as a surveillance site for meningitis cases. We carried out a cross-sectional, descriptive, and observational study based on retrospective data collected from January 2011 to December 2021.

Population

Patient samples included in this study were obtained from the health districts of Bobo-Dioulasso, particularly the urban districts of Dafra and Do, as well as surrounding areas. All cerebrospinal fluid (CSF) samples with incomplete or non-compliant reports were excluded. Bacteriological analyses were performed at the Bacteriology and Virology Laboratory of CHUSS, designated as a National Level Laboratory for case-based surveillance of bacterial meningitis. This laboratory is affiliated with more than a dozen health districts and receives CSF samples collected in cryotubes and transported under cold chain conditions (+4 to +8°C) for RT-PCR testing.

Source and type of data

Burkina Faso relies on two complementary population-based meningitis surveillance systems. The first is the routine notifiable disease surveillance, which includes meningitis and is conducted through the Weekly Official Letter Telegram (TLOH). All health facilities submit weekly reports, which are then aggregated at the regional level. The second is an enhanced surveillance system for diseases with epidemic potential. For meningitis, this system operates on a case-by-case basis, established after the introduction of the MenAfriVac^Ⓡ conjugate vaccine. It involves the systematic detection and confirmation of all suspected meningitis cases by a reference laboratory.

For each suspected case, epidemiological, clinical, and laboratory data are collected from notification forms, and a CSF sample is obtained and sent to the reference laboratory for confirmation [5]. For the present study, we used data from the case-by-case surveillance system, which included information on the patient’s age, sex, occupation, medical history, the reason for requesting the test, and the results of RT-PCR analysis of the CSF.

Operational definitions

We defined a suspected case as a sudden onset of fever, headache with stiff neck, or bulging fontanel in infants.

A confirmed case of bacterial meningitis was defined as a suspected case in which one of the following three bacterial species, N. meningitidis, H. influenzae type b, or S. pneumoniae, was identified by PCR.

Analysis of samples at the CHUSS National Level Laboratory

Real-time PCR of CSF from suspected cases was performed with both positive and negative controls, using specific primers and reference strains (Streptococcus pneumoniae ATCC 49136, Neisseria meningitidis serogroup A ATCC 13077, and Haemophilus influenzae type b ATCC 10211). Species identification relied on the detection of the lytA gene for S. pneumoniae, sodC for N. meningitidis, and hpd for H. influenzae (CDC Pneumococcus Laboratory Resources and Protocols [6]). The PCR procedures were standardized and provided by the National Meningitis Reference Laboratory (Charles De Gaulle Pediatric University Hospital, Ouagadougou).

Statistical analysis

Qualitative variables are presented as counts and percentages, while quantitative variables are described using means and standard deviations. The distribution and temporal trends of the samples were illustrated graphically. Data were analyzed using Stata 17.

Results

A total of 3823 suspected cases of acute bacterial meningitis were included during the study period. Of these, 3341 samples were analyzed by RT-PCR, with 1186 testing positive, yielding a confirmation rate of 35.5% (1186/3341) for bacterial meningitis (Figure 1).

Figure 1.

Flow diagram of CSF samples received at the University Hospital laboratory, CSF: cerebrospinal fluid; PCR, polymerase chain reaction.

The temporal distribution of samples submitted to the CHUSS National Reference Laboratory showed a pronounced peak in 2012, accounting for 32.9% of all samples collected during the study period. In terms of geographical origin, the Dafra health district was the main contributor, providing 20.2% of the total submissions (Figure 2).

Figure 2.

Distribution of samples received at the Hospital Bacteriology and Virology Laboratory of CHUSS for suspected bacterial meningitis by the patient's health district of origin (n = 3823)

Among patients with molecularly confirmed bacterial meningitis, 794 (66.9%) were male, corresponding to an M/F sex ratio of 2.02. Children aged 0-5 years constituted the most affected group (42.9%). The temporal distribution of confirmed cases revealed a peak in 2012 (37.2%), followed by a progressive decline, reaching the lowest frequency in 2021 (Figure 3).

Figure 3.

Curve of the annual evolution of confirmed cases of bacterial meningitis from 2011 to 2021.

The highest proportions of confirmed cases were reported in the Dafra (16.3%) and Dandé (14.1%) health districts (Figure 4).

Figure 4.

Distribution of confirmed cases of bacterial meningitis by health district of origin.

S. pneumoniae was the leading causative agent (51.9%), followed by N. meningitidis (46.1%) and H. influenzae type b (1.9%). The temporal distribution of N. meningitidis meningitis cases showed a marked peak in 2012, with 369 cases (69.5%), followed by a progressive decline from 2013 onwards.

Among the 547 confirmed N. meningitidis cases, molecular characterization revealed a predominance of serogroup W (69.7%), no detection of serogroups A or B, and 12.8% of cases with undetermined serogroups. Yearly analysis highlighted the predominance of serogroups W and X in 2011 and 2012, with respective frequencies of 71.4% and 35.2%. No cases of N. meningitidis meningitis were reported in 2021 (Figure 5).

Figure 5.

Annual distribution of Neisseria meningitidis serogroups from 2011 to 2021.

Discussion

This retrospective cross-sectional study analyzed cases of acute bacterial meningitis identified through case-based surveillance between January 2011 and December 2021. Several limitations were encountered, including missing data on patient age, sex, and health district of origin, as well as the absence of PCR testing for some samples. These constraints may have introduced selection bias. Nevertheless, the study achieved its primary objective of monitoring the dynamics of N. meningitidis serogroups since the introduction of the meningococcal A conjugate vaccine in 2010 at the CHUSS National Reference Laboratory.

The extended surveillance period in this region of Burkina Faso following the successful mass vaccination with MenAfriVac™ provides robust evidence of the disappearance of serogroup A. This outcome is likely attributable to the rapid establishment of herd immunity induced by the conjugate vaccine, which offers longer-lasting protection compared with previously used non-conjugate vaccines. However, our findings also underscore the persistent risk of meningitis outbreaks caused by non-A serogroups, particularly W and X, which have emerged following the decline of serogroup A [7,8].

The most affected age group was children aged 0-5 years (42.9%), consistent with the 40% reported by MacNeil et al. [10] in Burkina Faso in 2014 for the same age group. According to the literature, this high susceptibility is explained by the immaturity of the immune system, increased exposure to infections, and pathogen-specific characteristics, all of which predispose young children to bacterial meningitis [11,12]. A marked peak in sample submissions and confirmed cases was observed in 2012, corresponding to the meningitis epidemic declared in Burkina Faso that year. The epidemic, spanning weeks 8 to 17, accounted for 7022 suspected cases and 739 deaths, with a national incidence of 42 cases per 100,000 population 17 points higher than the previous year. The most affected health districts included Hauts-Bassins, Cascades, and Boucle du Mouhoun [10,13]. No cases of N. meningitidis serogroup A were detected during the 11 years of surveillance. This disappearance mirrors findings from Niger (Collard et al. [14]) and Burkina Faso (Ky-Ba et al. [15]), and confirms the major impact of MenAfriVac™ not only in reducing meningitis A incidence but also in decreasing asymptomatic carriage, enhancing individual protection, and generating herd immunity [16,17].

However, post-MenAfriVac™ surveillance has revealed an increasing burden of epidemics caused by non-A serogroups, particularly W, X, and C. The emergence of these serogroups may reflect ecological replacement following the elimination of serogroup A [18]. In our study, four N. meningitidis serogroups were identified: W (69.7%), X (16.6%), C (0.7%), and Y (0.2%). The predominance of W, X, and C has also been highlighted by Laforce et al. [16] and the World Health Organization reports [17]. To address this evolving epidemiology, recommendations have included targeted immunization of children born after the MenAfriVac™ campaigns at 9 months of age, as well as strategies to reach adults over 29 years who were not included in the initial mass vaccination. Furthermore, the circulation of multiple non-A serogroups underscores the urgent need for polyvalent conjugate vaccines adapted to the African meningitis belt.

During the peak of serogroup W meningitis in Burkina Faso, reactive vaccination campaigns using the tetravalent polysaccharide vaccine (A, C, Y, W) were implemented in affected health districts. This intervention led to a sharp decline in W cases, and since 2018, no new cases of serogroup W meningitis have been recorded in the region. Given that 12.8% of N. meningitidis cases in this study were of undetermined serogroup, it would be valuable to perform a metagenomic analysis of the CSF samples stored in our laboratory. This approach could improve our understanding of meningitis epidemiology and allow the inclusion of additional probes in the reagent panel used for case-by-case surveillance.

With the disappearance of serogroup A N. meningitidis, S. pneumoniae has become the leading cause of major meningitis epidemics, accounting for approximately one-quarter to one-third of all laboratory-confirmed endemic cases in the African meningitis belt [18,19]. In the present study, S. pneumoniae was identified in 51.9% of cases, followed by N. meningitidis (46.1%) and H. influenzae type b (1.9%). This predominance of S. pneumoniae has also been reported in Burkina Faso by Soeters et al. in 2019 (97.9%) [20] and by Bagaya (44.9%) [21].

The continued emergence of pneumococcal meningitis despite the widespread use of the PCV13™ conjugate vaccine since October 2013 [20] may be explained by the limited serotype coverage of the vaccine, which includes only 13 serotypes. Notably, serotype 12F, which is a major cause of pneumococcal meningitis after serotype 1 (included in PCV13™), is not covered.

The H. influenzae type b vaccine was introduced in Burkina Faso on january 1, 2006. An evaluation conducted by Kaboré et al. between 2004 and 2008 demonstrated a substantial reduction in H. influenzae type b meningitis from 42.3% to 11.8% of cases [22].

Conclusion

This study highlights a shift in the epidemiology of bacterial meningitis over the 10 years following the introduction of the meningococcal A conjugate vaccine in Burkina Faso. Although the elimination of serogroup A is evident, our data underscore the ongoing risk of epidemics caused by non-A meningococcal serogroups in the African meningitis belt. These findings emphasize the need for a comprehensive preventive strategy, including polyvalent meningococcal vaccination, robust surveillance systems, and rapid laboratory confirmation to effectively control and prevent meningitis outbreaks.

Declaration of competing interest

The authors have no competing interests to declare.