SEARCH

SEARCH BY CITATION

Keywords:

  • meningitis;
  • Neisseria meningitidis;
  • epidemic;
  • African meningitis belt;
  • Niger

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Population and geographical characteristics
  6. Clinical and epidemiological records
  7. Bacteriological diagnosis
  8. Case definition and statistical analysis
  9. Results
  10. Discussion
  11. References

In the African meningitis belt, the recurrent meningococcal meningitis epidemics are generally caused by serogroup A. In the past 20 years, other serogroups have been detected, such as X or W135, which have caused sporadic cases or clusters. We report here 134 meningitis cases caused by Neisseria meningitidis serogroup X that occurred in Niamey between 1995 and 2000. They represented 3.91% of the meningococcal isolates from all CSF samples, whereas 94.4% were of serogroup A. Meningococcal meningitis cases were detected using the framework of the routine surveillance system for reportable diseases organized by the Ministry of Public Health of Niger. The strains were isolated and determined by the reference laboratory for meningitis in Niamey (CERMES) and further typed at the WHO collaborating center of the Pharo in Marseille and at the National Reference Center for the Meningococci at the Institut Pasteur. Reference laboratories in Marseille and Paris characterized 47 isolates having the antigenic formula (serogroup:serotype:sero-subtype) X:NT:P1.5. Meningitis cases due to meningococcus serogroup X did not present any clinical or epidemiological differences to those due to serogroup A. The seasonal incidence was classical; 93.3% of the cases were recorded during the dry season. The mean age of patients was 9.2 years (±6 years). The sex ratio M/F was 1.3. Case fatality rate was 11.9% without any difference related to age or sex. The increasing incidence of the serogroup X was not related to the decrease of serogroup A, but seemed cyclic, and evolved independently of the recurrence of both serogroups A and C.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Population and geographical characteristics
  6. Clinical and epidemiological records
  7. Bacteriological diagnosis
  8. Case definition and statistical analysis
  9. Results
  10. Discussion
  11. References

In the African meningitis belt, recurrent outbreaks of meningococcal meningitis were mostly caused by serogroup A Neisseria meningitidis, less frequently to serogroup C. Other serogroups, such as X or W135, were considered as rare until the emergence of serogroup W135 at the end of the 2001 epidemic in Niger and Burkina Faso (Taha et al. 2002), during the 2002 epidemic in Burkina Faso (WHO 2002) and in early 2003 also in Burkina Faso (WHO 2003). We report here the occurrence of 134 cases due to N. meningitidis serogroup X in Niamey between 1995 and 2000. This serogroup represented 3.91% of the meningococcal strains isolated from cerebrospinal fluid (CSF) samples during this period, whereas serogroup A represented 94.40%.

Niamey, the capital of Niger, is located in the center of the African meningitis belt (Lapeyssonnie 1963). In this African region, between the isohyets 300 and 1100 mm, large outbreaks occur periodically. Since 1981, serogroup A was identified in more than 85% isolates studied at the Centre de Recherche Médicale et Sanitaire (CERMES) of Niamey (Campagne et al. 1999b). Serogroup C has been involved in one outbreak that occurred between 1990 and 1993. Serogroups B, W135, Y and X were found occasionally. We describe here the epidemiological and clinical profiles of epidemic meningitis cases because of serogroup X between 1996 and 2000.

Population and geographical characteristics

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Population and geographical characteristics
  6. Clinical and epidemiological records
  7. Bacteriological diagnosis
  8. Case definition and statistical analysis
  9. Results
  10. Discussion
  11. References

Niamey is located in the sub-Saharan Africa near the river Niger. The city has grown rapidly since independence in 1960, and the growth of the population once more increased after the Touareg's rebellion in 1990. Several districts developed anarchically, especially in the suburbs. In 2002, the population was estimated at approximately 700 000 inhabitants, based on the 1999 census.

Clinical and epidemiological records

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Population and geographical characteristics
  6. Clinical and epidemiological records
  7. Bacteriological diagnosis
  8. Case definition and statistical analysis
  9. Results
  10. Discussion
  11. References

Every suspected case of meningitis is admitted to the infectious disease unit at the National Hospital. After lumbar puncture, CSF samples are stored at 37 °C until shipment to the Meningitis Reference Laboratory of the CERMES. Each clinical case of serogroup X meningococcal meningitis occurring between 1995 and 2002 was documented.

Bacteriological diagnosis

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Population and geographical characteristics
  6. Clinical and epidemiological records
  7. Bacteriological diagnosis
  8. Case definition and statistical analysis
  9. Results
  10. Discussion
  11. References

Each CSF sample was analysed on receipt at the CERMES, by cytological examination and cell counts, Gram staining and latex agglutination for the detection of soluble antigens of Streptococcus pneumoniae, N. meningitidis serogroups A, B and C and Haemophilus influenzae type b (Slidex méningite-kit 5®, bioMérieux, Lyon, France). Bacteria were identified according to standard methods (Riou & Guibourdenche 1992). Cultures were performed on chocolate agar supplemented with PolyVitex (bioMérieux) for 24 h at 37 °C in 5–10% CO2. Typical colonies were identified by using Api® NH kits (bioMérieux). Serogroups were determined by slide agglutination with specific immune sera to serogroups A, B, C, Y, W135 and X from the W.H.O. Collaborating Center of Marseille and/or the Institut Pasteur. Neisseria meningitidis isolates were stored at −80 °C until further determination of their serotype and subtype (Riou & Guibourdenche 1992).

Antibiotic susceptibility testing was performed by disk-diffusion test and E-test, according to Nicolas et al. (1998). Multilocus sequence typing was used to characterize two serogroup X strains isolated in 1997 and two other isolated in 1998. DNA PCR-amplified fragments of approximately 450 base-pairs from seven house-keeping genes (abcZ, adk, aroE, fumC, gdh, pdhC et pgm) were sequenced and their allelic profiles were compared with the existing alleles available on the internet Multilocus sequence typing data bank (http://www.mlst.net/) to determine their sequence type (Maiden et al. 1998).

Case definition and statistical analysis

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Population and geographical characteristics
  6. Clinical and epidemiological records
  7. Bacteriological diagnosis
  8. Case definition and statistical analysis
  9. Results
  10. Discussion
  11. References

A case of serogroup X meningococcal meningitis X was defined as a patient suffering from clinical signs of meningitis and from whose CSF a strain of N. meningitidis serogroup X was isolated. Statistical significance was assessed by using the χ2-test (significance threshold P ≥ 0.05 for given degree of freedom, d.o.f.) for independent analysis of epidemic or non-epidemic years. For serogroup A, we used the WHO threshold, which defines an epidemic year on the basis of an overall incidence over 100 cases per 100 000 inhabitants (Campagne et al. 1999b). This incidence was never reached for serogroup X. However, we considered that in 1997 the incidence of meningitis X reaching 15.1 per 100 000 habitants was unacceptable and imposed emergency measures (WHO 1999).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Population and geographical characteristics
  6. Clinical and epidemiological records
  7. Bacteriological diagnosis
  8. Case definition and statistical analysis
  9. Results
  10. Discussion
  11. References

Between 1982 and 2001, 164 N. meningitidis serogroup X strains from meningitis cases were identified at CERMES (Table 1; Figure 1). However, our study was limited to the 134 cases occurring between 1995 and 2001, including 83 cases in 1997, because of the lack of clinical and epidemiological data for previously reported cases. Ninety-three per cent occurred during the dry season between January and May, with a peak in March with 53 cases corresponding to 40% of the total number of cases recorded between 1995 and 2000 (Figure 2). Ninety-four per cent of the patients were under 20 years of age (mean 9.9 ± 6, P < 0.05, median 8 years), roughly the same as for serogroup A meningococcal meningitis cases (Table 2). However, in contrast to serogroup A cases which occur both in neonates and older children (Campagne et al. 1999a), serogroup X was not isolated from infants. The male to female sex ratio was 1.3 (76/58), mean and median age were similar in both sexes, and the case fatality rates were 13.2 and 10.3, respectively (χ2 = 0.25; d.f. = 1; P > 0, 5), with an overall case fatality rate for serogroup X not significantly different from that of serogroup A (11.9 and 11.7, respectively). The age distribution between serogroup A or X cases appeared slightly different but statistical comparison was difficult because of the small size of the age groups for X meningitis (Tables 2 and 3). Case fatality rates due to serogroups X and A are compared in Table 3. However, the complete lack of X cases below 1 year has to be considered. The mean duration of hospitalization was 6.6 ± 4 days (P = 0.05), independent of sex and age. The geographical distribution of the cases in Niamey during the 1997 epidemic showed that 68% occurred in poor and precarious traditional districts (Table 4; Figure 3). Only 5% of the patients lived in modern districts.

Table 1.  Serogroups of Neisseria meningitidis isolated from CSF at CERMES from 1995 to 2002
YearsACXBW135Not determined
  1. * From January to April 2002.

19951883120021
1996218212000
1997168183200
199858127000
19994216000
200070034004
2001169100156
2002*4501130
Total32831013531831
image

Figure 1. Incidence of meningitis A and X between 1982 and 2001.

Download figure to PowerPoint

image

Figure 2. Seasonal distribution of meningitis A and X cases.

Download figure to PowerPoint

Table 2.  Mean annual incidence of A and X meningitis according to age (per 100 000 people belonging to this age group)
Age group (years)Serogroup A*Serogroup A†Serogroup X‡
  1. * 1995 Outbreak (n = 1883).

  2. † Inter-epidemic period (1981–1994, n = 1868).

  3. ‡ 1997 Outbreak (n = 83).

<1638.437.40
1–4489.832.217.9
5–9468.843.335.7
10–14476.337.218.5
15–19475.830.93.1
≥2074.76.614.7
Table 3.  Case fatality rate of meningitis A and X meningitis cases and deaths according to age
Age group (years)Serogroup A (%)*Serogroup A (%)†X casesX deaths (%)Lost to follow up
  1. * 1995 Outbreak (n = 1883).

  2. † Inter-epidemic period (1981–1994, n = 1868).

<129.119.3000
1–412.812.0265 (19)2
5–98.49.0547 (13)4
10–1410.67.6351 (3)2
15–198.15.2111 (1)1
≥2020.128.972 (29)1
Table 4.  Population density and incidence of meningitis cases during the 1997 outbreak related to the type of quarters
Quarter typesApproximate populationPopulation density (inhab./km2)X casesIncidence rateIncidence in infected quarters
Modern110 000750098.222.5
Rehabilitated125 00085001915.237.3
Traditional385 00010 0005814.922.1
Total620 00090008613.924.4
image

Figure 3. Distribution of meningitis X cases in Niamey during the 1997 outbreak.

Download figure to PowerPoint

Forty-seven N. meningitidis serogroup X strains were isolated between 1995 and 2001, all with the same antigenic formula (serogroup:serotype:sero-sub-type) X:NT:P1.5. MLST of the two strains isolated in 1997 and the two isolated in 1998 showed that they had the sequence-type ST-181. All strains were susceptible to penicillin G, amoxicillin, ceftriaxone, chloramphenicol and rifampicin.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Population and geographical characteristics
  6. Clinical and epidemiological records
  7. Bacteriological diagnosis
  8. Case definition and statistical analysis
  9. Results
  10. Discussion
  11. References

Although the incidences of serogroup X meningococcal meningitis cases reported here are far below the epidemic threshold, the high number of grouped cases, all apparently because of the same clone, recorded during the period of the study in the same area justify considering the phenomenon as an outbreak due to the emergence of serogroup X N. meningitidis, even if taking into account the improved techniques for aetiological diagnosis of acute bacterial meningitis in Niamey since the creation of the national reference center at the CERMES. However, this sudden increase of cases caused by N. meningitidis X occurring between two severe epidemics of serogroup A meningococcal meningitis, in 1995 with an incidence of 300/100 000 and in 2000 with an incidence of 290/100 000, is exceptional. Etienne et al. (1990) reported a similar outbreak in 1990 with 22 bacteriologically confirmed cases occurring between February and April.

Clinical and epidemiological features of serogroup X meningococcal meningitis do not differ from those due to serogroup A excepted for the lack of X cases before 1 year of age. The geographical distribution, with most of the cases occurring in districts of higher population densities and poverty, all factors of enhanced host to host direct transmission, was similar to that reported by Etienne et al. (1990). The incidence has been higher in central and some peripheral quarters, the latter probably being more rural. Population density is quite similar in the different districts (modern, rehabilitated and traditional). But the incidence rate was lower in modern quarters and obviously higher in rehabilitated infected quarters but numerous traditional quarters were not concerned by the outbreak. Statistical comparison was not feasable because of the small number of cases in each district type. On the contrary, demographical and geographical data were not perfectly linked because there is no precise address or quarters delimitations leading to comprehensive description of the case distribution.

Serogroup X outbreaks did not reach the magnitude of serogroup A epidemics, with tenths of cases per year rather than thousands, respectively. Serogroup X outbreaks resolved spontaneously and seemed self limited whereas large outbreaks of serogroup A occurred in the following years. Whether serogroup X is less transmissible than serogroup A which has spread through sub-Saharan Africa for decades remains to be elucidated (Riou et al. 1996; Gagneux et al. 2002b). A recent report on high prevalence of serogroup X carriage among children in Northern Ghana is also an alert for the need of careful surveillance of this potentially new epidemic variant (Gagneux et al. 2002a). It remains questionable whether the emergence of serogroup X could be linked to the decrease of serogroup A diseases and carriage as it was observed after the intense serogroup A epidemic in Niamey in 1990 and 1997 and in Ghana in 2000 (Gagneux et al. 2002a). The dramatic reduction in circulation of serogroup A after a large epidemic can be explained by the acquired immunity among the exposed population (herd immunity) and by the effects of reactive vaccination campaigns. Therefore, an ecological niche can be occupied by a new antigenic variant escaping the specific herd immunity to the former epidemic serogroup. This hypothesis underlines the constant risk for selecting new meningococcal antigenic variants against which the exposed host has no immune defenses, because of the extraordinary polymorphism potential of several genetic determinants in N. meningitidis (Linz et al. 2000). The selective role of the A and C vaccine-induced immunity cannot be obviously retained in our study, as we did not observe an inverse correlation between the high rate of A and C vaccination in some districts and the subsequent incidence of serogroup X. Moreover, the large and effective A and C vaccination campaign of 2000 in Niamey (Chabalier et al. 2001) did not result in the emergence of serogroups X. It seems therefore that serogroup X recurrent outbreaks in Niamey between 1995 and 2000 occurred independently of the dynamics of the epidemic waves due to serogroups A.

Meningococcal disease epidemics are due to the expansion of one single clone (Achtman 1995; Nicolas & Debonne 2002). In the 2000 survey in Ghana, Gagneux et al. (2002a) by molecular typing of serogroup X strains showed that they belonged to ST-751. In our study, MLST of four strains from two different outbreaks were ST-181, a sequence type yet found in Chad in 1995 and in Mali from 1970 to 1990 (Gagneux et al. 2002b). Pulsed-field gel electrophoresis patterns of genomic DNA from these clones show few differences. The emergence of serogroup X meningococcal clones in various countries of the African meningitis belt threatens a potential epidemic in replacement or addition to serogroup A epidemics. The global expansion of serogroup W135 from the clonal complex ET-37, ST-11 since the 2000 Hajj pilgrimage (Taha et al. 2000) and in the 2001 epidemics in Burkina Faso and Niger in 2001 (Taha et al. 2002) followed by the 2002 epidemic (WHO 2002) and 2003 cases in Burkina Faso (WHO 2003) is an example of the sudden emergence of a new epidemic clone formerly found only in sporadic cases in various countries. This underlines the urgent need for reinforcing the bacteriological diagnosis capacities in the laboratories in Africa with the aim to identify and type rapidly any causative agent of acute bacterial meningitis in order to adapt the most appropriate curative and preventive strategies. New epidemic or potentially epidemic clones of N. meningitidis, such as those of the serogroup X, have been poorly studied and no vaccine has been evaluated. Further studies are needed on their molecular pathogenesis and immunogenicity to design new vaccines to prevent any predictable epidemic spread in Africa.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Population and geographical characteristics
  6. Clinical and epidemiological records
  7. Bacteriological diagnosis
  8. Case definition and statistical analysis
  9. Results
  10. Discussion
  11. References
  • Achtman M (1995) Global epidemiology of meningococcal disease. In: Meningococcal Disease (ed. KACartwright), John Wiley & Sons, Chichester, pp. 159175.
  • Campagne G, Chippaux JP, Djibo S, Issa O & Garba A (1999a) Epidémiologie et contrôle des méningites bactériennes chez les nourrissons à Niamey (Niger). Bulletin de la Société de Pathologie Exotique 92, 118122.
  • Campagne G, Djibo S, Schuchat A, Ousseini A, Cisse L & Chippaux JP (1999b) Epidemiology of bacterial meningitis in Niamey, Niger 1981–1996. Bulletin of the World Health Organization 77, 499508.
  • Chabalier de F, Garba A & Chippaux JP (2001) Enquête de couverture vaccinale par sondage en grappes après une campagne de vaccination antiméningococcique de masse, Niamey (Niger). Cahiers Santé 11, 173176.
  • Etienne J, Sperber G, Adamou A, Picq JJ (1990) Notes épidémiologiques: les méningites à méningocoques du sérogroupe X à Niamey (Niger). Médecine Tropicale 50, 227229.
  • Gagneux S, Hodgson A, Smith TA et al. (2002a) Prospective study of a serogroup X Neisseria meningitidis outbreak in northern Ghana. Journal of Infectious Diseases 185, 618626.
  • Gagneux S, Wirth T, Hodgson A et al. (2002b) Clonal groupings in serogroup X Neisseria meningitidis. Emerging Infectious Diseases 8, 462466.
  • Lapeyssonnie L (1963) La méningite cérébro-spinale en Afrique. Bulletin of the World Health Organization 28, 1114.
  • Linz B, Schenker M, Zhu P & Achtman M (2000) Frequent interspecific genetic exchange between commensal Neisseriae and Neisseria meningitidis. Molecular Microbiology 36, 10491058.
  • Maiden MCJ, Bygraves JA, Feil E et al. (1998) Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms. Proceedings of the National Academy of Sciences of the United States of America 95, 31403145.
  • Nicolas P & Debonne JM (2002) Infections à méningocoques. In: Maladies infectieuses, 8-013-A-10, Pédiatrie/Maladies infectieuses, 4-250-A-30. Encyclopédie Médico-chirurgicale (EMC), Paris, p. 23.
  • Nicolas P, Cavallo JD, Fabre R & Martet G (1998) Standardisation de l’étude de la sensibilité de Neisseria meningitidis par dilution en milieu solide, E-test et la méthode des disques, pour la pénicilline, l'amoxicilline, le céfotaxime, la ceftriaxone, le chloramphénicol et la rifampicine. Détection des souches relativement résistantes à la pénicilline. Bulletin of the World Health Organization 76, 393400.
  • Riou JY & Guibourdenche M (1992) Laboratory Methods Neisseria and Branhamella. Institut Pasteur Edition, Paris.
  • Riou JY, Djibo S, Sangare L et al. (1996) A predictable comeback: the second pandemic of infections due to Neisseria meningitidis serogroup A subgroup III-1 in Africa in 1995. Bulletin of the World Health Organization 74, 181187.
  • Taha MK, Achtman M, Alonso JM et al. (2000) Serogroup W135 meningococcal disease in Hajj pilgrims. Lancet, 356, 2159.
  • Taha MK, Parent du Chatelet I, Schlumberger M et al. (2002) Neisseria meningitidis serogroup W135 and A were equally prevalent among meningitis cases occurring at the end of the 2001 epidemics in Burkina Faso and Niger. Journal of Clinical Microbiology 40, 10831084.
  • WHO (1999) Control of Epidemic Meningococcal Disease. WHO Practical Guidelines, 2nd edn. OMS, Geneva.
  • WHO (2002) Meningococcal disease, serogroup W135, Burkina Faso: preliminary report 2002. Weekly Epidemiological Records 77, 152155.
  • WHO (2003) Meningococcal disease, Burkina Faso. Weekly Epidemiological Records 78, 33.