Epidemiology

For over 100 years, major epidemics of meningococcal disease have occurred every few years within the African meningitis belt, which runs across the continent from Senegal to Ethiopia (Lapeyssonie, 1963). These epidemics are very disruptive, requiring the establishment of emergency treatment centres, and placing a severe strain on routine health services. The reason for the susceptibility of this region of Africa to major epidemics of meningococcal disease is at least in part related to its climatic features, with outbreaks occurring mainly in the hot, dry season (Sultan et al., 2005). Most epidemics have been due to Neisseria meningitidis serogroup A (NmA), and some have been due to serogroups W, X and C, but there has been a conspicuous absence of outbreaks due to serogroups B and Y. The hypervirulent clonal complexes ST-5 (mainly serogroup A) and ST-11 (mainly serogroup W) have accounted for most epidemics in this region. NmW, in particular, has been responsible for several epidemics in the past 10 years (e.g. in Burkina Faso, Ghana and Niger), but the dynamic of these NmW outbreaks appears to differ from those due to NmA. Based on district-level data from 2002 to 2012 in Burkina Faso, Chad, Niger and Nigeria, and defining an epidemic as crossing a weekly threshold of 10 cases/100 000 population, there were fewer NmW epidemics (36) than NmA epidemics (177) during this period. On average, NmW epidemics were 78% the size of NmA epidemics, with median cumulative attack rates of 109/100 000 (interquartile range [IQR] 79–134) for NmW and of 139/100 000 (IQR 99–230) for NmA.

Since 2010, countries in the extended meningitis belt (Figure 1) have started to progressively introduce a new serogroup A meningococcal conjugate vaccine (MenAfriVac) through mass campaigns (WHO, 2013). This preventive measure is expected to confer both long-lasting individual protection and herd immunity (Kristiansen et al., 2013; Novak et al., 2012; Sow et al., 2011). With the support of the Global Alliance for Vaccines and Immunization (GAVI), all but one country in the meningitis belt have, since 2000, introduced Haemophilus influenzae type b (Hib) vaccines, and many have already introduced Streptococcus pneumoniae (Spn) conjugate vaccines; hence, the incidence of bacterial meningitis due to non-meningococcal pathogens is also evolving.

Figure 1

Countries of the extended African meningitis belt.

The epidemiological pattern changed after the introduction of MenAfriVac, as demonstrated in Table 1. NmA declined to a low proportion of confirmed cases after 2010, and there was an accompanying rise in the proportion of cases due to NmW and, to a lesser extent, to NmX. Also, the total number of reported cases declined in the last 3 years. The pathogen distribution partly reflects the laboratory sampling and confirmation capacity, which is higher in Burkina Faso and Niger than in other countries in the belt. The proportion of suspected cases with laboratory confirmation across the belt has been rising over the past 10 years, from 2–3% in 2003 to 6–7% in 2013, but is still at a relatively low level.

Table 1

Reported and confirmed meningitis cases and pathogen distribution, countries of the African meningitis belt, 2003–2013.

Current strategy

The current WHO epidemic control strategy (WHO, 2000) is based on the dynamics of epidemics before the introduction of MenAfriVac (Lewis et al., 2001). The main elements of the strategy are as follows (WHO, 2010):

• During the dry season, the epidemic risk is assessed at district level by monitoring the number of cases reported in a given population.
• Based on this risk assessment, operational thresholds have been defined that trigger the reinforcement of surveillance (the alert threshold), and the launch of vaccination campaigns and the use of a specific antibiotic treatment protocol (the epidemic threshold).
• Once the alert threshold is crossed, a sufficient number of cerebrospinal fluid (CSF) samples are taken to confirm the occurrence of a meningococcal meningitis outbreak, and to identify the responsible serogroup to inform the choice of vaccine.
• Once a meningococcal meningitis epidemic is confirmed, single-dose antibiotic treatment is recommended for case management, together with mass vaccination campaigns using an appropriate polysaccharide vaccine.

Because polysaccharide vaccines offer short-term protection only, they are used in outbreak response but not in routine immunization programmes.

Need for review of strategy

The current thresholds have been appraised as being both sensitive and specific for detection of NmA epidemics (Kaninda et al., 2000; Leake et al., 2002; Lewis et al., 2001). Following the introduction of MenAfriVac, NmW is now the predominant serogroup; hence, these threshold incidence rates and the population base for interventions may no longer be valid, and the recommendations for management of epidemic meningitis in the meningitis belt need to be reviewed.

Rapid diagnostic tests (RDTs) support urgent decision-making for outbreak management. Latex agglutination tests and dipsticks, such as those developed by the Centre de Recherche Médicale et Sanitaire (CERMES) in Niger (Chanteau et al., 2006), show variable sensitivity and specificity in field conditions. RDTs can discriminate, for example, between the various bacterial species causing meningitis and, for Nm, between serogroups A, C, W and Y, but not X. With the changing epidemiology, identifying the predominant bacterial agent and meningococcal serogroup in outbreaks has become increasingly important.

The recommended presumptive treatment of bacterial meningitis is administration of ceftriaxone for at least 5 days (WHO, 2010). During epidemics of meningococcal meningitis, to ensure rapid and effective treatment at first contact, single-dose regimens of ceftriaxone or chloramphenicol have been recommended (WHO, 2007). This protocol has been shown to provide effective treatment for meningococcal meningitis (Nathan et al., 2005); it is simple and cheap, and the necessary antibiotics can be made available at the most peripheral level. However, single-dose regimens are not effective against other pathogens such as Hib or Spn, which require longer courses of treatment (Brouwer et al., 2010). With fewer NmA outbreaks, a higher proportion of cases of meningitis due to Hib and Spn is expected, especially where vaccines against these pathogens have not yet been introduced.

Chemoprophylaxis is recommended for household contacts of sporadic cases of meningococcal disease in Africa, but not during epidemics (WHO, 1998). In European countries, no distinction is made between sporadic cases and outbreaks in terms of recommendations for prophylaxis (antibiotics and vaccination, where relevant), although the scale of outbreaks in the WHO European Region is much smaller than in the meningitis belt. Chemoprophylaxis of household contacts may be effective in settings outside the African meningitis belt (Purcell et al., 2004), and vaccination, where relevant, may offer additional protection (Hoek et al., 2008), but evidence from the meningitis belt is scarce. One randomized trial in Nigeria during an epidemic of serogroup A in the 1970s showed benefit from vaccination of household contacts in the absence of chemoprophylaxis (Greenwood et al., 1978). With the expected reduction in frequency and extent of epidemics in the meningitis belt, it is timely to review this policy.

The objectives of this review were to revise WHO guidelines on control of epidemic meningitis in sub-Saharan Africa concerning:

1. operational thresholds and vaccination responses
2. use of RDTs
3. use of single-dose antibiotic regimens
4. prophylaxis for household contacts.

The outcomes of reviews of these four questions are presented in this guideline. Operational procedures on implementation of the guideline will be produced separately. The guideline only applies to countries of the extended meningitis belt (Figure 1), and does not include guidance on vaccines, which is covered by a WHO position paper (WHO, 2011).

The target audience includes ministries of health, nongovernment organizations (NGOs), WHO regional and country offices, and public health professionals working in the extended African meningitis belt; WHO collaborating centres; manufacturers of vaccines, RDTs and antibiotics; and funding agencies.