Parasitol Res (2002) 88: 1040–1043
DOI 10.1007/s00436-002-0712-6
O R I GI N A L P A P E R
Jean-François Faucher Æ Edouard Ngou-Milama
Michel Anoumou Missinou Æ Raphaël Ngomo
Maryvonne Kombila Æ Peter G. Kremsner
The impact of malaria on common lipid parameters
Received: 2 April 2002 / Accepted: 10 June 2002 / Published online: 23 July 2002
Ó Springer-Verlag 2002
Abstract No data are available in the literature to in-
dicate whether low-level Plasmodium falciparum infec-
tions induce lipid parameter changes. We hypothesized
that low-level P. falciparum infections induce significant
changes in common lipid parameters. We retrospectively
selected samples from a malaria prophylaxis study to
measure the impact of sustained parasite clearance on
common lipid parameters [total cholesterol (TChol),
high-density lipoprotein (HDL)-cholesterol (HDL-c),
low-density lipoprotein (LDL)-cholesterol (LDL-c) and
triglycerides (TG)] in 47 apparently healthy schoolchil-
dren whose P. falciparum parasitemia was initially below
1000/ll. After parasite clearance, mean values were
significantly increased for Tchol (P<0.001) and HDL-c
(P<0.001), unlike LDL-c (P=0.93); and TG were sig-
nificantly decreased from the baseline (P=0.004). No
significant change was found in a control group. This is
the first study showing significant lipid changes related
to low-level P. falciparum infections. Further studies are
J.-F. Faucher (&)
Service des Maladies Infectieuses et Tropicales,
Hôpital St Jacques, CHU de Besançon,
25030 Besançon cedex, France
E-mail: jffaucher@chu-besancon.fr
Tel.: +33-381-218533
Fax: +33-381-218772
J.-F. Faucher Æ M.A. Missinou Æ P.G. Kremsner
Medical Research Unit, Albert Schweitzer Hospital,
Lambaréné, Gabon
E. Ngou-Milama Æ R. Ngomo
Laboratoire de Biochimie, Faculté de Médecine et des
Sciences de la Santé, Libreville, Gabon
M.A. Missinou Æ P.G. Kremsner
Department of Parasitology,
Institute for Tropical Medicine, University of Tübingen,
Tübingen, Germany
M. Kombila
Laboratoire de Parasitologie,
Faculté de Médecine et des Sciences de la Santé,
Libreville, Gabon
needed to explore the relevance of this finding at the
population level in hyperendemic malaria areas.
Introduction
Very little is known about lipid changes related to ma-
laria. Hypocholesterolemia and hypertriglyceridemia
were observed in the setting of both uncomplicated and
complicated malaria (Das et al. 1996; Davis et al. 1993;
Mohanty et al. 1992; Nilsson-Ehle and Nilsson-Ehle
1990; Ngou-Milama et al. 1995) and in the setting of
many other acute infections (Bentz and Magnette 1998;
Samra et al. 1996). The magnitude of these changes
seems related to the severity of malaria (Das et al. 1996;
Mohanty et al. 1992). In one study conducted in non-
immune patients, however, no correlation was found
between the severity of malaria attacks and the extent of
high-density lipoprotein (HDL)-cholesterol decrease
(Kittl et al. 1992).
Population studies on common lipid parameters
indicate that cholesterol values are currently lower in
Africa, where malaria is endemic, than in many other
parts of the world (Brotons et al. 1997). Thus, yet
unknown genetic and environmental factors take part in
the trend to cholesterol decrease in Africa and are
therefore to be studied. The impact of environmental
factors on cholesterolemia is illustrated by the rise in
cholesterolemia observed in Africans living in urban
centers, in comparison to Africans living traditionally in
rural areas (Walker and Sareli 1997). Environmental
factors include lifestyle changes and changes in the
prevalence of infections, the main one being malaria.
Evaluating the impact of malaria on lipid parameters at
the population level is a difficult challenge. At any given
time, low-level Plasmodium falciparum parasitemia is
highly prevalent in areas hyperendemic for malaria.
However, no data are available in the literature to in-
dicate whether lipid parameter changes are related to this
condition. We hypothesized that low-level P. falciparum

infections induce significant changes in common lipid
parameters. We designed a retrospective analysis of
plasma samples obtained while performing a malaria
prophylaxis study, in order to explore the role played by
low level parasitemia on common lipid parameters in
Gabonese children attending school.
Materials and methods
Samples were retrospectively selected from a prophylaxis study on
atovaquone-proguanil (A-P; Malarone, Glaxo Wellcome, Becken-
ham, UK) which took place in the Lalala Public school in
Lambaréné, Gabon, between January and June 2000. Lambaréné is
in the tropical rain forest in an area hyperendemic for Plasmodium
falciparum malaria (Wildling et al. 1995); and the parasites in this
region are resistant to chloroquine (Kremsner et al. 1994). Healthy
schoolchildren aged 4–16 years, who did not receive drugs with
antimalarial activity within 7 days before enrolment in the trial,
participated in screening. We assessed baseline health by a medical
examination. Written informed consent was obtained from the
parents of the children. The study was approved by the ethics
committee of the International Foundation of the Albert
Schweitzer Hospital in Lambaréné.
At baseline screening, we collected venous blood samples for
hematological and clinical chemistry tests and if appropriate, for a
pregnancy test. Because we expected a high rate of parasitemia in
these children at baseline, all the children received an initial cu-
rative treatment course of artesunate (Plasmotrim Lactab, Me-
pha) once daily for 3 days. A thick blood smear was taken at the
start and at the end of this initial curative treatment course. The
smears were stained with Giemsa stain and parasite counts were
measured as previously described, with a detection limit that
corresponded to 5 parasites/ll (Kremsner et al. 1988). For the
chemosuppression phase, volunteers received a placebo or a daily
dose of A-P. Thick blood smears were taken weekly. The
chemosuppression phase began 7 days after the initial treatment
and continued for 12 weeks. Blood samples were again collected
5 weeks after the initial clearance treatment in all children re-
maining in the study.
Sample selection
Plasma was obtained for each child at the start of clearance
treatment (D0) and 5 weeks later (D35). Since all chidren received
at least one study drug (artesunate) and in order to rule out a study
drugs-induced effect on lipids parameters, we selected two sets of
samples according to the following criteria:
1. Low parasitemia (LP) group : (1) a positive P. falciparum par-
asitemia below 1000 parasites/ll at the start of clearance treat-
ment followed by parasite clearance within 7 days and (2) a
further negativity of weekly thick blood smears during 6 weeks.
2. Control (C) group: (1) A negative P. falciparum parasitemia at
the start of clearance treatment and (2) a further negativity of
weekly thick blood smears during 6 weeks. Because the number
of such samples largely exceeded the number of group LP
samples and for logistic reasons, we randomly selected a number
of samples that was equal to the LP group samples (see Results)
in each treatment group (A-P, placebo).
Sample analysis
Each of the common lipid parameters [total cholesterol (Tchol),
HDL-cholesterol (HDL-c), low-density lipoprotein (LDL)-choles-
terol (LDL-c) and triglycerides (TG)] was measured using a
Biomérieux kit and an enzymatic procedure on an autoanalyser
RA 1000 System.
1041
Statistical analysis
Statistical analysis was done with Statview software (SAS Insti-
tute). Differences between D0 data and D35 data for each pa-
rameter were assessed using the Wilcoxon signed rank test. The P
values given are two-sided.
Results
Of the total 330 volunteers, 112 (34%) had a positive
blood smear at the start of clearance treatment: 107 had
Plasmodium falciparum (median = 620/ll), 3 had
P. malariae and 2 had P. ovale. A total of 58 chidren
(18% of volunteers) had a P. falciparum parasitemia
below 1000/ll. Among these children, 47 (21 allocated
to receive a placebo, 26 allocated to receive A-P) both
cleared parasitemia within 7 days and remained negative
for at least 6 weeks (median = 120/ll).
Among the total volunteers, 218 (66%) had a nega-
tive blood smear at the start of clearance treatment; and
194 remained negative for at least 6 weeks. In this
group, we randomly selected samples from 21 volunteers
allocated to receive a placebo and 26 allocated to receive
A-P.
Table 1 shows the demographic and laboratory
characteristics of volunteers from groups LP and C at
study enrolment.
In the LP group (Table 2), from D0 until D35, a
highly significant increase in both Tchol (P<0.001) and
HDL-c (P<0.001) values were observed, while LDL-c
values did not change (P=0.93). TG values decreased
significantly between D0 and D35 (P=0.005).
In the C group (Table 3), there was no significant
change between D0 and D35 for Tchol (P=0.6), HDL-c
(P=0.05), LDL-c (P= 0.17) and TG (P=0.26).
A further analysis was performed in a subset of the
47 children of the LP group, restricted to 21 healthy
schoolchildren whose P. falciparum parasitemia was
initially below 100/ll [very low parasitemia (VLP)
group]. In the VLP group, mean values on D35 were
significantly increased for Tchol (P=0.03) and HDL-c
Table 1. Demographic and laboratory characteristics of volunteers
in the low parasitemia (LP) and control (C) groups at enrolment in
a prophylaxis trial comparing atovaquone-proguanil (A-P) against
a placebo. In the LP group, the male:female ratio for the total
107 children with an initial Plasmodium falciparum positive smear
was 60:47. In the C group, the male:female ratio for the total
218 children with a initially negative smear was 89:129. In both
groups, the A-P:placebo ratio was 1.24:1.00 (26:21). Values given
are means (±SD)
LP group C group
(47 children) (47 children)
Male:female (n) 25:22 20:27
Age (years) 9.1 (±2.7) 8.9 (±2.3)
Weight (kg) 26 (±7) 25 (±7)
Hb (g/dl) 10.2 (±1.4) 10.9 (±1.1)
Creatinine (lmol/l) 66.0 (±6.6) 67.0 (±9.5)
ALAT (units/ll) 30 (±10) 34 (±23)
Bilirubin (units/ll) 16.0 (±8.1) 14.0 (±8.1)