Patterns of "Plasmodium falciparum" infection and morbidity in a rural community in Northern Ghana

Malaria continues to be of major public health importance, especially in sub-Saharan
Africa. At a minimum, 900 million acute febrile episodes occur yearly resulting in 1-3
million deaths yearly, mostly in African children below 5 years of age. Less than 20% of
these deaths come to the attention of any formal health system. The World Bank (1993)
ranks malaria as the leading cause of lost disability-adjusted life years (DALYs) in Africa
with an estimated 35 million future life-years lost from disability and premature death.
Without effective malaria control programme(s), the massive burden of malaria
morbidity and mortality is expected to at least double in the next 20 years due mainly to
the growing spread of Plasmodium falciparum resistance to Chloroquine and other antimalarials.
Lack of development of new affordable drugs; the financial constraints on
health services in most countries; and lack of expertise to plan properly for malaria
control are also sources of worry. Hope has been re-kindled by the initiation of the Roll
Back Malaria (RBM) whose goal is to reduce malaria by half in 10 years. The goal of the present study is to characterise the epidemiology of Plasmodium
falciparum parasites, patterns of seasonality in infection and morbidity and their
relationship to radical cure in the Kassena-Nankana district (KND) in northern Ghana.
Such information prior to clinical malaria intervention trials will be important for optimal
design and implementation. In the KND, malaria transmission is holoendemic with
seasonal peaks and troughs mirroring the wet and dry seasons in the district. It is the
leading cause of both morbidity and mortality, accounting for 60% of hospital
admissions, and 35% of all deaths
Cohorts of either non-immune children or semi-immune adults or all age groups were
randomly selected using a cluster sampling approach that was facilitated by the availability
of the Navrongo Demographic Surveillance System (NDSS). In total, between 200 and
300 inhabitants constituted the cohort for each of the studies. With all eligible people
within the district included in the sample frame, and using the Stata program, 16 "index"
compounds were randomly selected. Potential volunteers were recruited sequentially from nearby compounds in order of proximity to the "index" compounds until the
required number of volunteers was made up. Participants screened based on criteria predetermined
were enrolled to participate in the study. Those studied for incidence of
infection and first clinical malaria were treated with anti-malarials to clear them of all
malaria parasitaemia prior to follow-up. They were then followed prospectively for at least
four months to determine re-infection, and clinical malaria. The selection procedure in
those studies for parasite genotypes and severe anaemia were the same, but these
studies were carried out as cross-sectional surveys. The main findings were:
Malaria transmission was found to be intense throughout the year, with seasonal peaks
and troughs. The overall prevalence of P. falciparum was 70% by microscopy and 82%
by PCR with the highest parasite rates among 5-9 year olds and highest parasite
density (geometric mean 1,922/μl blood) in 1-2 year olds. PCR-RFLP typing of the P.
falciparum msp2 gene revealed a mean multiplicity of 3.4 (range: 1 to 8) genotypes with
the two msp2 allelic families (FC27 and 3D7) in almost equal proportions. The
correlation between parasite density and msp2 multiplicity was highest in infants, and
decreased with age to a minimum by 10 years, then start to increase again from this
age into adolescence and adulthood.
The incidence density of P. falciparum infection in both infants/young children and
adults was the same (ID = 7.0 cases/person-year) in the wet season, and only
decreased slightly (5.0 cases/person-year) in the dry season. The cumulative incidence
of infection profiles in both age groups indicated the same rapid rise with over 90% reinfection
rate within 12 weeks post-treatment.
The risk of developing febrile parasitaemia of >5,000/μl, >10,000/μl, >20,000/μl and
>50,000/μl during the wet season was 1.92, 1.93, 2.45 and 4.33 times that of the dry
season with p-values always less than 0.0025. Malaria parasites were cleared prior to following up on the cohorts in order to determine
incidence of P. falciparum infections. This resulted in 49% of volunteers in the treatment
group experiencing clinical attacks of malaria compared to 38% in the untreated group
[RR (95% CI)=1.29 (1.03-1.61)]. Clinical malaria in the treated group was associated
with significantly more symptoms and lower parasitemia. Severe anaemia defined as Hb<6.0g/dL, at the end of the wet season (November 1996)
was 22.1% compared to 1.4% at the end of the dry season (April 1997), [OR (95%
CI)=20.1 (7.1 – 55.3)]. Nutritional and hookworm anaemia appeared to have little impact
upon this seasonal difference since anthropometric indices were comparable with no
hookworm infection among 6-24 months aged children. A repeat survey in November
2000 confirmed that the proportion of severely anaemic children and overall mean
haemoglobin levels, in the 2000 sample were significantly improved over those of the
1996 (17.5 % vs. 26.4%, P = 0.03; Hb 7.5 g/dL vs. 6.9 g/dL, P = 0.002). Relative to
children with Hb≥6.0 g/dL, those with severe anaemia (Hb<6.0 g/dL) were older, more
frequently parasitaemic [OR (95% CI)=1.60 (1.08-2.35)], more often febrile [OR (95%
CI)= 2.44 (1.71-3.48)], and predominantly male [OR (95% CI)=1.50 (1.05-2.13)]. These findings bear upon the design of malaria drug and vaccine trials in holoendemic
areas such as the KND. Optimal design of malaria intervention trials, ie sample sizes
calculations and follow-up could be borne out of these findings. Changes in the
multiplicity of infections based on msp2 genotyping can be used for assessing the
outcome of malaria clinical intervention trials. The evidence also suggests that dramatic
peaks and troughs of severe anaemia are regular and possibly predictable events that
may be used to assess malaria clinical intervention trials in areas similar to the KND.
Evaluation of some clinical malaria intervention trials will require clearance of
parasitaemia prior to follow-up. The interpretation of such intervention trials need to be
carried out with a lot of caution as clinical malaria occurring after clearance of malaria
parasitaemia may be distinctly different from "natural" disease and this may have
significance for the design and interpretation of intervention trials.