Meningococcal and pneumococcal meningitis in Northern Ghana

Despite improvements in technology, treatments and understanding of how bacterial meningitis
develops, the disease remains a potentially life-threatening emergency capable of causing significant
morbidity and mortality. N. meningitidis, S. pneumoniae and H. influenzae type b, which are
commensally normal human nasopharyngeal flora, are the most important and common causes of
bacterial meningitis. N. meningitidis (especially, serogroup A) is well known for its association with
epidemics in the meningitis belt of sub-Saharan Africa. This nearly always starts during the dry
season and stops during the onset of the rains and occurs every 8-12 years in the “meningitis belt”
with attack rates sometimes exceeding 1% during these epidemics. H. influenzae type b and S.
pneumoniae are mostly endemic affecting certain risk groups. N. meningitidis serogroup W135,
traditionally known to cause isolated cases, has raised general concern in recent years due to
outbreaks in Burkina Faso since 2002 attributed to it.
Following a major meningococcal meningitis epidemic in Northern Ghana in 1996/7 the Navrongo
Health Research Centre in collaboration with the Swiss Tropical Institute in 1998 initiated a longterm
colonization and disease study in the Kassena Nankana District (KND), with the aim of
contributing to the understanding of the epidemiology, pathogenesis, improved intervention and
early detection of bacterial meningitis epidemics in the “meningitis belt”. As part of this long term
study, this thesis focuses on meningococcal colonization and invasive disease surveillance
(pneumococcal and meningococcal), burden of pneumococcal meningitis and the relationship
between environmental factors and the risk of meningococcal and pneumococcal meningitis.
From 1998 to 2005 clonal waves of nasopharyngeal colonization with pathogenic and nonpathogenic
meningococcal genoclouds were observed in the KND through the longitudinal
meningococcal colonization study of residents of 37 randomly selected compounds. These
meningococci were not only less diverse and unstable in composition with rare non-groupable
strains, but they were also mostly made up of predominantly hyperinvasive strains (up 71%) with
constant microevolution. In 1998 serogroup A meningococci ST5 caused an outbreak of
meningococcal meningitis in the KND with persistent carriage up to 1999, disappearing in 2001. In
2000 serogroup X ST571 meningococci emerged with high carriage rates and few cases. Carriage of
this serotype persisted until 2001 when it was replaced by serogroup A ST7 which only disappeared
at the latter part of 2005 after causing outbreaks between 2002 and 2004.
Although N. meningitidis serogroup W135 has been the cause of epidemics in neighbouring Burkina
Faso since 2002, only sporadic cases (4) were reported in Ghana from 2003 to 2004. The disease
isolates were very similar to the Burkinabe epidemic strains by Pulse Field Gel Electrophoresis
analysis. Colonization surveys over a one-year period in one of the patient home communities
(which has semi-closed features) showed an initial high carriage rate of 17.5% and persistence of
carriage with rapid microevolution.
Between 2000 and 2004 there was an outbreak of pneumococcal meningitis (PCM) caused by a S.
pneumoniae serotype 1 clonal complex in the KND with features (seasonality, clonality and broad
age spectrum of the patients) characteristic of meningococcal meningitis (MCM). This hypervirulent
serotype is repeatedly being isolated in various parts of sub-Saharan Africa.
A two-year survival analysis comparing 67 PCM cases recorded at the War Memorial Hospital
(WMH), Navrongo, Ghana, identifiable on a demographic surveillance system, with equal numbers
of MCM and community controls, showed profound excess mortality of the PCM compared with
both MCM and community controls. A case-control study of sequelae (using a structured disability
questionnaire, neuropsychological and audiometric examinations of both cases and controls),
matching for age, sex and geographical location, including 46 traceable survivors of PCM (cases),
46 community controls (CC) and 34 survivors of MCM, showed that hearing and speech impairment
as well as psychiatric disorders are much more frequent and severe in PCM than MCM.
Epidemics of MCM and PCM are closely related to climate. A time series analysis of weekly
meteorological data (humidity, rain fall, dust, wind speed, temperature and sunshine) from the local
weather station and the corresponding reported epidemiological data (confirmed meningococcal and
pneumococcal cases) from 1998 - 2004 from the WMH microbiology database was carried out using
negative binomial regression and Bayesian methods. The aim of these micro epidemiological
analyses was to describe as well as provide an early warning system for the short-term prediction of
likely meningococcal and pneumococcal meningitis outbreaks in the KND.
The environmental factors that influence the incidence of PCM and MCM were found to be similar
but not always the same. The duration of a preceding absence of rainfall appears to be the best
predictor of both PCM and MCM outbreaks. Outbreaks of MCM are best predicted by concurrent
decrease in rainfall with increase in weekly mean maximum temperature. Those of PCM are
influenced by concurrent decrease in rainfall.
The natural variations in the predominance of different pharyngeal meningococcal serotypes and
serogroups over time might contribute to meningococcal meningitis epidemics in the African
meningitis belt. The future epidemiological trend of meningococcal and pneumococcal meningitis
will be influenced by changes in the use of antibiotics, immune status, aging of the global population
and technology. The introduction of carbohydrate-conjugate or common protein vaccines to routine
immunization schedules, together with maternal immunization and enhanced disease (and/or
colonization) surveillance, could make pneumococcal and meningococcal diseases of less public
health importance.