AJCN Tufts Nutrition Symposium, Boston Sept 24-26
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow An erratum has been published
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nyakeriga, A. M
Right arrow Articles by Williams, T. N
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nyakeriga, A. M
Right arrow Articles by Williams, T. N
Agricola
Right arrow Articles by Nyakeriga, A. M
Right arrow Articles by Williams, T. N
American Journal of Clinical Nutrition, Vol. 80, No. 6, 1604-1610, December 2004
© 2004 American Society for Clinical Nutrition

ORIGINAL RESEARCH COMMUNICATION

Malaria and nutritional status in children living on the coast of Kenya1,2,3

Alice M Nyakeriga, Marita Troye-Blomberg, Alex K Chemtai, Kevin Marsh and Thomas N Williams

1 From the KEMRI/Wellcome Trust Program, Center for Geographic Medicine Research, Coast, Kilifi District Hospital, Kilifi, Kenya (AMN, KM, and TNW); the Department of Immunology, Wenner-Gren Institute, Stockholm University, Stockholm (AMN and MT-B); and the Faculty of Health Sciences, Moi University, Eldoret, Kenya (AMN and AKC)

2 Supported by UNDP/World Bank/WHO Special Program for Research and Training in Tropical Diseases, the Swedish Agency for Research Cooperation with developing countries, and the Wellcome Trust, United Kingdom.

3 Reprints not available. Address correspondence to AM Nyakeriga, KEMRI/Wellcome Trust Program, Center for Geographic Medicine Research, Coast, PO Box 230, Kilifi, Kenya. E-mail: nyakeriga{at}imun.su.se.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: The relation between malnutrition and malaria is controversial. On the one hand, malaria may cause malnutrition, whereas on the other hand, malnutrition itself may modulate susceptibility to the disease.

Objective: The objective was to investigate the association between Plasmodium falciparum malaria and malnutrition in a cohort of Kenyan children.

Design: The study involved the longitudinal follow-up of children aged 28–60 mo for clinical malaria episodes and anthropometric measurements through 4 cross-sectional surveys. We used Poisson regression analysis to investigate the association between malaria and nutritional status.

Results: The crude incidence rate ratios (IRRs) for malaria during the 6-mo period before assessment in children defined as malnourished on the basis of low height-for-age or low weight-for-age z scores (<–2) were 1.17 (95% CI: 0.91, 1.50; P = 0.21) and 0.94 (0.71, 1.25; P = 0.67), respectively, which suggests no association between malaria and the subsequent development of protein-energy malnutrition. However, we found that age acted as an effect modifier in the association between malaria episodes and malnutrition on prospective follow-up. The IRR for malaria in children aged 0–2 y, who were subsequently characterized as underweight, was 1.65 (1.10, 2.20; P = 0.01), and a significant overall relation between malaria and stunting was found on regression analysis after adjustment for the interaction with age (IRR: 1.91; 1.01, 3.58; P = 0.04).

Conclusion: Although children living on the coast of Kenya continue to experience clinical episodes of uncomplicated malaria throughout the first decade of life, the effect of malaria on nutritional status appears to be greatest during the first 2 y of life.

Key Words: Anthropometric measures • nutrition • protein-energy malnutrition • PEM • Plasmodium falciparum • malaria • immunoglobulins • Kenya • children


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Both Plasmodium falciparum malaria and malnutrition are major causes of child morbidity and mortality (1, 2), yet their precise interaction remains unknown. Although many studies have shown that P. falciparum infection can result in acute weight loss (3), whether recurrent malaria has a sustained effect on growth remains unclear. Support for such an effect largely comes from intervention studies involving malaria control. For example, improvements in growth and other anthropometric indexes have been described in children protected from malaria by both chemoprophylaxis (4–8) and insecticide-treated bed nets (9–11). Nevertheless, this experience has not been universal (12), and the benefits of malaria control have been most apparent in younger children (4, 13). Although the reasons for an age-specific effect remain unknown, possible explanations include age-related issues of compliance and effectiveness, physiologic factors such as breastfeeding and the presence of fetal hemoglobin, and the acquisition of malaria-specific immunity. In the current study we analyzed data from an ongoing cohort study in children living on the coast of Kenya, with a view to further exploring the relation between malaria and anthropometric indexes.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Study area
The study was conducted in the Ngerenya area of Kilifi District on the Kenyan coast. The pattern of malaria transmission and the characteristics of the study population were described previously (14, 15). Protein-energy malnutrition (PEM) is common in this part of Kenya and affects >30% of children aged <5 y (16).

Study design
The current study was nested within a rolling cohort study evaluating the natural history and acquisition of immunity to malaria, which was described in detail previously (17). Briefly, the study began in 1998 when a cohort was recruited that consisted of all residents of randomly selected households that were identified during a prestudy census of a prescribed area (17). In November 2001 the cohort was reduced in size to include only children aged <8 y. The study design allowed for the inclusion and exclusion of children on the basis of eligibility criteria. In particular, children born into the study households during the study period were recruited to the study at birth and children were dropped after their eighth birthday. The present study was conducted during a 2-y period between November 2001 and October 2003 and involved a total of 340 children. Ethical permission for this study was granted by the Kenya Medical Research Institute National Ethical Review Committee. Individual written informed consent was obtained from the parents of all study participants.

Anthropometric assessment and cross-sectional surveys
Four cross-sectional surveys were conducted 6, 12, 18, and 24 mo after the study began, in May and October 2002 and in May and October 2003 (Figure 1Go). At the time of each survey, all children were weighed with Seca digital scales (model 835; CMS Instruments, Oxford, United Kingdom). Heights were measured in children aged >1 y with a Leicester height measure (CMS Instruments), and lengths were measured in younger children with the use of a length board (AHRTAG design as modified by Nicoll and Ulijaszec) (18). Weight-for-age (WAZ), height-for-age (HAZ), and weight-for-height (WHZ) z scores were calculated with reference to National Center for Health Statistics standards with the use of EPI INFO (version 6; Centers for Disease Control and Prevention, Atlanta). Underweight and wasting (indexes of acute malnutrition) were defined as a WAZ and a WHZ <–2, respectively, and stunting (an index of chronic malnutrition) as an HAZ <–2. At each survey, additional routine data were collected regarding a range of clinical indexes, including axillary temperature, clinical symptoms of fever, and the presence of malaria parasites on a blood film. In addition, a venous blood sample was collected for immunologic and biochemical assays during the first 2 surveys. All children involved in the study received routine health and nutritional advice, irrespective of their nutritional status at each point of clinical contact.



View larger version (13K):
[in this window]
[in a new window]
  		FIGURE 1. Study schedule.

 
Morbidity surveillance
Cohort children were seen once weekly by trained fieldworkers. At each follow-up visit, a standard morbidity questionnaire was completed that detailed each child’s state of health, any clinical symptoms, and any medications taken recently. Axillary temperature was measured at each visit with a digital thermometer. Thick and thin blood smears were prepared for all children that had either a history of fever during the preceding 48 h or an axillary temperature of >37.5°C. All slides were read within 24 h, appropriate management being informed by the results. For the purposes of this study, malaria was defined according to 2 alternative definitions. "malaria definition 1" applied to those subjects with either a measured fever (axillary temperature >37.5°C) at the time the slide was taken or a clinical history of fever within the preceding 48 h in conjunction with a positive test result for blood-stage asexual P. falciparum parasites at any density. "Malaria definition 2" applied to those subjects with a measured fever in conjunction with P. falciparum parasites at any density (for children aged <1 y) or at a density of >2500/µL (for children aged ≥1 y). Although definition 1 is the most sensitive definition for malaria, definition 2—which is derived by multiple logistic regression as previously described (19)—is more specific (>80%) (17). Nonmalaria fever was defined as fever in the absence of parasites on microscopy. Children whose slide results were positive for malaria parasites were treated according to standard Government of Kenya guidelines. First-line treatment was with sulfadoxine/pyrimethamine (Falcidin, Cosmos, Kenya); in the event of treatment failure, children were treated with amodiaquine. Children born in study households during the study period were recruited at birth. Children exited from the study on their 8th birthday, if their parents withdrew informed consent, if they moved out of the study area for >2 mo, or if they died.

Laboratory procedures
Blood films were stained with Giemsa and examined for malaria parasites by standard microscopy. Parasite densities were recorded as a ratio of parasites to white blood cells (read from thick smears) or to red blood cells (from thin smears) for heavier infections. Densities (parasites/µL whole blood) were then calculated assuming a white blood count of 8 x 103/µL or a red cell count of 5 x 106/µL. Hemoglobin typing was conducted by cellulose acetate electrophoresis. We assayed the concentrations of total malaria-specific immunoglobulin (Ig) G and total and malaria-specific IgE by enzyme-linked immunosorbent assay as previously described (20). These immunoglobulins were chosen because, in previous studies, we have found them to be elevated in persons living in malaria-endemic areas (21, 22).

Statistical methods
The principal question we wished to explore in this analysis was the relation between malaria infection and the subsequent development of malnutrition. Our hypothesis a priori was that malnutrition would be related to the incidence of malaria during the preceding 6-mo period. As such, our outcome of primary interest was the IRR of malaria in malnourished compared with well-nourished children, both before and after adjustment for potential confounders. We first explored our data for potential confounders, factors that significantly affected the risk of uncomplicated clinical malaria, using Poisson regression. We identified age (in 1-y categories), sickle cell status (HbAS or HbAA), season (categorized in 13-wk blocks), and ethnic group as significant associations. We checked for interactions and included these in models where indicated. We then compared the incidence of malaria during the period preceding anthropometric assessment in children categorized as well nourished or malnourished on the basis of HAZ and WAZ. Because only a small proportion of children had a low WHZ, this analysis was not conducted using a low WHZ. We constructed a separate regression model to investigate the possibility that nutritional status at the beginning of a follow-up interval affected the risk of malaria during the subsequent 6-mo period.

For the purpose of this analysis, children were ascribed their cross-sectional nutritional status for the entire 6-mo period after a cross-sectional survey. Their status was refreshed at the next follow-up survey if they attended or they were dropped from further analysis if absent. This model included WAZ or HAZ (each categorized as <–2 or ≥–2) as dependent variables and malaria incidence as the independent variable. For both analyses, children were considered not at risk and were dropped from both numerator and denominator populations for 21 d after receiving treatment with an antimalarial drug. Children were withdrawn from the study on attainment of their eighth birthday, if for any reason they failed to attend follow-up for a total of 8 consecutive wk, or if they died. All analyses were conducted with and without adjustment for the effect of the confounding variables age group, sickle cell status, season, and ethnic group and are expressed as incidence rate ratios (IRRs). We used the sandwich estimator (23) to calculate CIs and significance values in our regression models to account for the possibility that individual children might contribute disproportionately to our results. This estimator accounts for within-person clustering of events by relaxing the assumption of independence of events in the calculation of variance.

Proxy indicators of malaria incidence, including the proportion of children with an HbAS genotype and the mean concentration of malaria-specific immunoglobulin were outcomes of secondary interest. Immunoglobulin data were normalized by log transformation before analysis. Biological data were compared with the use of unpaired Student’s t tests, except for proportion data, which were compared with the use of Fisher’s exact tests. All data were analyzed with STATA (version 8; Stata Corp, Timberlake, TX).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
A total of 340 children participated in the study. The prevalence of underweight and stunting and the characteristics of the cohort according to nutritional category at each survey are summarized in Table 1Go. The prevalence of underweight ranged from 22% to 26%, of stunting from 32% to 38%, and of wasting from 2% to 5%. These values did not vary significantly between surveys. Similarly, no significant differences were seen with regard to the biological characteristics of the malnourished compared with the well-nourished children. Because the subjects were a rolling cohort, allowing for the inclusion of new members who satisfied the inclusion criteria and the exclusion of children who no longer met the inclusion criteria, not all children contributed equally to the data. Of the 340 children who participated in our study, 279 took part in 3 surveys, and 179 took part in all 4 surveys. Of the 179 children, 16% were underweight throughout the 4 surveys, 24% remained stunted throughout the 4 surveys, and 25% had variations in their nutritional status throughout the 4 surveys (ie, were underweight or stunted or both at any one survey depending on the characteristic but did not retain either characteristic of malnourishment throughout the 4 surveys). The remaining percentage of subjects were neither underweight nor stunted (ie, were well nourished throughout the 4 surveys).


View this table:
[in this window]
[in a new window]
  		TABLE 1 Biological characteristics of the subjects by nutritional category1

 
The adjusted IRR (95% CI) for malaria by definition 1 in HbAS children (compared with baseline HbAA children) was 0.59 (95% CI: 0.40, 0.86; P < 0.01) and for definition 2 was 0.52 (0.31, 0.85; P < 0.01). Nevertheless, no consistent trend was seen toward a lower representation of children with the HbAS genotype in the worst-nourished than in the best-nourished categories, as might be expected if malaria was a major cause of malnutrition. Similarly, no trend was seen toward a better nutritional status in children with the HbAS genotype (Table 2Go). On the contrary, the trend was opposite that expected, WHZs being significantly lower in HbAS children in 2 of the 4 surveys. WAZs were also consistently lower in HbAS children although not significantly so in any individual survey. No significant differences were seen in immunoglobulin concentrations between children in the different nutritional groups.


View this table:
[in this window]
[in a new window]
  		TABLE 2 Mean anthropometric measurements by hemoglobin genotype1

 
The incidence of malaria in both malnourished and nonmalnourished children during the interval preceding assessment is compared in Table 3Go and Figure 2Go. In an overall crude analysis, no significant differences were seen in the incidence of malaria by subsequent anthropometric status (Table 3Go). However, when the data were stratified by age, an association was seen between the incidence of malaria and subsequent malnutrition in both the youngest and oldest age strata (Figure 2Go). An elevated incidence of malaria was seen in children aged <2, who were subsequently found to be either underweight or stunted; a reduced incidence was seen in the oldest children. These effects were significant for low WAZ in the children aged 1–2 y (IRR: 1.98; 95% CI: 1.28, 2.68; P = 0.002) and in the children aged <2 y (1.65; 1.10, 2.20; 0.01). We therefore investigated the possibility that age might have been acting as an effect modifier in the association between malaria and malnutrition by fitting interaction terms to our Poisson regression models. We found no evidence for an interaction between age (in y) and the association between malaria and malnutrition over the full range of ages. However, in a comparison of models including and excluding an interaction between age (in tertiles) and nutrition status, we found evidence for a significant interaction between age tertile and both low HAZ (likelihood ratio {chi}2 = 11.6, P = 0.003) and low WAZ (likelihood ratio {chi}2 = 10.36, P = 0.006). Our final model, therefore, included sickle cell status (HbAA or HbAS), age (in y), ethnic group, season, and an interaction term between low HAZ or low WAZ and age tertile. The results of these analyses are summarized in Table 3Go. These analyses were repeated to include nonmalaria febrile episodes instead of or as well as malaria. No significant associations were identified (Table 3Go).


View this table:
[in this window]
[in a new window]
  		TABLE 3 Poisson regression analysis modeling the incidence of malaria in well-nourished and malnourished children before the anthropometric assessments were conducted1

 


View larger version (10K):
[in this window]
[in a new window]
  		FIGURE 2. Age-stratified incidence rate ratios (IRRs) for malaria by malaria definition 1 in children subsequently categorized as having a low height-for-age z score (HAZ; A) or a low weight-for-age z score (WAZ; B). Significant interactions were found between age (in tertiles) and both a low HAZ (likelihood ratio {chi}2 = 11.6, P = 0.003) and a low WAZ (likelihood ratio {chi}2 = 10.36, P = 0.006) score. The results of analyses for malaria by malaria definition 2 were not significantly different. The reference lines represent an IRR of 1.00.

 
No difference was seen in the incidence of malaria in malnourished children (characterized as either underweight or stunted) during the interval after the anthropometric assessment, which suggests that malnutrition did not modulate the risk of malaria in our cohort (Table 4Go). No significant effects of HbAS genotype on any nutrition index were observed (data not shown).


View this table:
[in this window]
[in a new window]
  		TABLE 4 Poisson regression analysis modeling the incidence of malaria in well-nourished and malnourished children after the anthropometric assessments were conducted1

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The relation between malaria and malnutrition is complex. Although increasing evidence suggests an association between nutrition, both in terms of anthropometric (24) and micronutrient (25–27) status, and susceptibility to malaria, our study focused on the role of malaria in the etiology of malnutrition. Both PEM and malaria are major risk factors for child morbidity and mortality in Africa. In a recent study that aimed to partition the global burden of disease into specific risk categories, childhood and maternal underweight accounted for the single largest portion (9.5%) of the global total of disability adjusted life years (1), whereas a second study recently showed a strong correlation between the prevalence of P. falciparum and all-cause mortality in children aged <5 y (2). Although largely unknown, the etiology of mild-to-moderate malnutrition in most tropical countries is probably multifactorial, not only involving an inadequate or inappropriate diet but also intercurrent infection (28). As such, the relation between malaria and malnutrition is important: if malaria is a major cause of malnutrition, it may be responsible for considerable indirect mortality that could potentially be reversed by effective malaria control. Indeed, this argument has been advanced in conjunction with a recent study of malaria control conducted in The Gambia, where the combination of insecticide-impregnated bed nets and targeted chemoprophylaxis for malaria led to a nearly 33% reduction in all-cause mortality (29), an effect that far exceeded that which could be attributed to malaria alone. However, the role of malaria in the etiology of PEM remains unclear. Some role appears likely from observations made during malaria-control programs. For example, small improvements in growth and anthropometric indexes were seen in children given chemoprophylaxis with chloroquine in both The Gambia (4) and Nigeria (6) and in children protected with bed nets in 2 parts of Kenya (9, 10). Nevertheless, malaria control has not resulted in nutritional benefit in all studies or in all children within individual studies. For instance, no effect was seen in a successful vector-control program at Pare in Tanzania (12), and the nutritional benefits observed have tended to be greatest in younger children (4, 13). This observation has generally been attributed to immunologic factors: the effect of malaria on nutrition decreasing as children acquire natural immunity. Important questions, therefore, still remain regarding the precise interaction between these 2 important conditions.

We investigated the association between malaria and PEM in a well-defined cohort of children living on the coast of Kenya. An important attribute of our study site is that malaria transmission is relatively low at {approx}10 infective bites per person per year (30) and that the acquisition of clinical immunity is therefore relatively slow (Figure 3Go). As a consequence, we anticipated that any effect of malaria on the nutritional status of children would be observed throughout the age range of the study participants. However, on the basis of multiple lines of evidence, we found little evidence for a sustained role of uncomplicated malaria in the etiology of malnutrition throughout the first 8 y of life. On crude analysis we found no evidence of an elevated incidence of malaria in children subsequently classified as malnourished on the basis of either HAZ or WAZ. Indeed, in the case of low WAZ, the trend was opposite that expected; the IRRs for malaria by both definitions were <1.00 in the group of children found to have low WAZs (<–2) at the subsequent survey compared with the baseline group (WAZ ≥–2). Next, we looked at the cross-sectional plasma concentrations of malaria-specific IgG in children classified by nutritional status. The log concentration of this immunoglobulin was strongly associated with the incidence of clinical malaria during the preceding 6-mo period (IRR for malaria by definition 1: 2.90; 95% CI: 2.10, 3.99; P < 0.0001 adjusted for age and ethnic group) and therefore acted as an alternative tool for investigating the association between malaria exposure and subsequent nutritional status. Nevertheless, there was no trend toward an elevated immunoglobulin concentration in malnourished children. Although this finding could be explained by a reduced antibody response in malnourished children, we found no difference in the concentration of the other immunoglobulins measured in our study. Taken together, these observations did not appear to support a role for malaria in the etiology of malnutrition in our population. An obvious issue regarding this conclusion is whether our study was sufficiently powered to detect an association if one existed. However, on the basis of simulations constructed from our observed data, our study appeared to be well powered: we estimated that our study had roughly 90% power to detect an IRR for malaria of 1.2 at the 5% significance level on the assumption that the effect of malaria was sustained throughout the age range studied (data not shown). We wondered whether the lack of an obvious association between malaria and nutritional status might simply reflect the fact that our study was too short to see such an effect. Although malaria has been found to result in acute weight loss in some studies, it is possible that malnutrition only results from repeated episodes over a longer period than we could observe during our study.



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 3. Age-specific incidence of malaria in the study population by malaria definition 1.

 
One way of investigating the longer-term effects of malaria on nutritional status was by studying the nutritional status of children with the sickle cell trait. Sickle cell trait was associated with a high degree of protection against uncomplicated malaria in our cohort, protection being constant at {approx}50% throughout the full age-range studied (data not shown). We reasoned that if malaria were a significant cause of malnutrition, children with an HbAS genotype should be better nourished and that, furthermore, this improvement in nutritional status might increase with age, a marker of cumulative malaria incidence. We found no evidence of such an effect (Tables 2Go and 3Go). Although this suggests to us that malaria is not a major cause of malnutrition at the population level, it must be stated that important differences may exist between persons with an HbAS and HbAA genotype in terms of their response to malaria infections, which might nullify our hypothesis. Nevertheless, on considering these results in the round it is tempting to conclude that malaria has no role to play in the high rates of malnutrition seen in our cohort. However, when we investigated the age-specific association between malaria and nutritional status we found evidence for an association between malaria and subsequent nutritional status in the youngest age groups. On further analysis it was evident that age was acting as an effect modifier in this association. When this interaction was taken into account, the association between malaria and nutritional status became manifest—an observation that is consistent with the conclusion that malaria is an important cause of malnutrition only in certain subgroups of children on the basis of age.

Although an age-specific effect has been seen in previous studies (4, 13), it has generally been thought to reflect a lower burden of malaria in older children because of acquired immunity. Although this was true in the oldest children in our study (aged >4 y), this does not appear to explain the lack of association between malaria and nutritional status in children aged 2–4 y, an age at which the incidence of malaria was still rising (Figures 2Go and 3Go). It is therefore interesting to speculate further regarding possible causes for this age-specific effect. Malaria might result in malnutrition through a range of different mechanisms, including a reduced intake through anorexia or through the induction of a catabolic state from the elaboration of proinflammatory cytokines such as tumor necrosis factor (TNF) (31). It is possible that the particular effect in early life may reflect age-specific differences in immune responsiveness profiles. For example, the immune response matures with age, with preferential secretion of TNF in younger children. Alternatively, it may simply reflect the fact that this is the time when linear growth is maximal or result from physiologic or behavioral characteristics that are peculiar to younger children. Future studies should aim to address these issues more specifically.

To date, the most convincing data supporting a role for malaria in the etiology of malnutrition has come from intervention studies. We have used a cohort approach to address this question from a different perspective. We recognize that both approaches have their problems. For example, it is impossible to differentiate cause and effect from a cohort study: although not supported by our prospective analysis or the prevailing hypothesis that malnutrition reduces malaria risk (24), we cannot exclude the possibility that malnutrition is a cause of increased malaria risk as opposed to being caused by it. Moreover, by conducting a prospective study, we necessarily had an effect on the course of both malaria infections—children received effective treatment more promptly than they might have done in the absence of the study—and malnutrition. This might be expected to reduce our ability to detect an effect when one existed. Nevertheless, our study does suggest a role for malaria in the etiology of malnutrition, which supports conclusions drawn from intervention studies and argues for further studies aimed at confirming this relation and exploring potential mechanisms. In this regard it is interesting to note that in a previous study of similar design, conducted in the South Pacific islands of Vanuatu, we found a strong association between the incidence of P. vivax malaria and subsequent underweight (IRR: 2.6; 1.5, 4.4; P < 0.0001) but no significant effect of P. falciparum (IRR: 1.1; 0.57, 2.1; P = 0.8) (32). Such studies should, therefore, account for the epidemiologic patterns of malaria and interactions with other species.

In conclusion, our data support a role for malaria in the etiology of PEM in younger children living on the coast of Kenya. Future studies should aim to confirm our findings and investigate the reasons for the age-specificity of this effect.


   ACKNOWLEDGMENTS
 
We thank Hedvig Perlmann, Margreta Hargstedt, and Brett Lowe for technical support; Neal Alexander for statistical advice; and the study volunteers, fieldworkers, and clinical staff for their support with this study.

AMN helped design the study, conducted the laboratory assays, and cowrote the manuscript. MT-B was the doctoral advisor to AMN and contributed to the writing of the manuscript. AKC provided intellectual support. KM was jointly responsible for the design of the study and cowrote the grant that funded the work. TNW was the senior investigator and had overall responsibility for the study and assisted with the analysis and manuscript preparation. AMN was supported by TDR/WHO. TNW was supported by a Career Development Award (052595) and KM was supported by a Senior Fellowship (062372), both from the Wellcome Trust United Kingdom. MT-B was supported by SIDA/Swedish Agency for Research Development with Developing Countries. The authors declare that they had no competing financial interests. This paper was published with the permission of the Director of KEMRI.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Ezzati M, Lopez AD, Rodgers A, Vander Hoorn S, Murray CJ. Selected major risk factors and global and regional burden of disease. Lancet 2002;360:1347-60.[Medline]
  2. Snow RW, Korenromp EL, Gouws E. Pediatric mortality in Africa: Plasmodium falciparum malaria as a cause or risk? Am J Trop Med Hyg 2004;71(suppl)64-70.[Abstract/Free Full Text]
  3. McGregor IA. Malaria: nutritional implications. Rev Infect Dis 1982;4:798-804.[Medline]
  4. McGregor IA, Gilles HM, Walters JH, Davies AH, Pearson FA. Effects of heavy and repleted malarial infections on Gambian infants and children. Effects of erythrocyte parasitization. BMJ 1956;2:686-92.
  5. Molineaux L, Gramiccia G. The Garki Project. Research on the epidemiology and control of malaria in the Sudan Savanna of West Africa. Geneva: WHO, 1980:217-29.
  6. Bradley-Moore AM, Greenwood BM, Bradley AK, Kirkwood BR, Gilles HM. Malaria chemoprophylaxis with chloroquine in young Nigerian children. III. Its effect on nutrition Ann Trop Med Parasitol 1985;79:575-84.[Medline]
  7. Colbourne M. The effect of malaria suppression in a group of Accra school children. Trans R Soc Trop Med Hyg 1955;49:556-69.[Medline]
  8. Harland PS, Frood JD, Parkin JM. Some effects of partial malaria suppression in Ugandan children during the first 3 years of life. Trans R Soc Trop Med Hyg 1975;69:261-2.[Medline]
  9. Snow RW, Molyneux CS, Njeru EK, et al. The effects of malaria control on nutritional status in infancy. Acta Trop 1997;65:1-10.[Medline]
  10. ter Kuile F, Terlouw DJ, Kariuki S, et al. Impact of permethrin-treated bed nets on malaria, anemia, and growth in infants in an area of intense perennial malaria transmission in western Kenya. Am J Trop Med Hyg 2003;68:68-77.[Abstract/Free Full Text]
  11. Shiff C, Checkley W, Winch P, Premji Z, Minjas J, Lubega P. Changes in weight gain and anaemia attributable to malaria in Tanzanian children living under holoendemic conditions. Trans R Soc Trop Med Hyg 1996;90:262-5.[Medline]
  12. Draper KC, Draper CC. Observations on growth of African infants with special reference to the effects of malaria control. Am J Trop Med Hyg 1960;63:165-s71.
  13. Friedman JF, Phillips-Howard PA, Hawley W, et al. Impact of permethrin-treated bed nets on growth, nutritional status, and body composition of primary school children in Western Kenya. Am J Trop Med Hyg 2003;68:78-85.[Abstract/Free Full Text]
  14. Snow RW, Omumbo JA, Lowe B, et al. Relation between severe malaria morbidity in children and level of Plasmodium falciparum transmission in Africa. Lancet 1997;349:1650-4.[Medline]
  15. Snow RW, Bastos de Azevedo I, Lowe BS, et al. Severe childhood malaria in two areas of markedly different falciparum transmission in east Africa. Acta Trop 1994;57:289-300.[Medline]
  16. Kenya Go. Kenya Demographic and Health Survey, 1993. Nairobi: Central Bureau of Statistics, 1994:278.
  17. Mwangi TW. Clinical epidemiology of malaria under differing levels of transmission. PhD thesis. Open University, Milton Keynes, United Kingdom, 2003:326.
  18. Nicoll A, Ulijaszek S. Modifications of the AHRTAG child length measurer. Trop Doct 1987;17:129-31.[Medline]
  19. Smith T, Schellenberg JA, Hayes R. Attributable fraction estimates and case definitions for malaria in endemic areas. Stat Med 1994;13:2345-58.[Medline]
  20. Perlmann H, Helmby H, Hagstedt M, et al. IgE elevation and IgE anti-malarial antibodies in Plasmodium falciparum malaria: association of high IgE levels with cerebral malaria. Clin Exp Immunol 1994;97:284-92.[Medline]
  21. Perlmann P, Perlmann H, Flyg BW, et al. Immunoglobulin E, a pathogenic factor in Plasmodium falciparum malaria. Infect Immun 1997;65:116-21.[Abstract]
  22. Calissano C, Modiano D, Sirima BS, et al. IgE antibodies to Plasmodium falciparum and severity of malaria in children of one ethnic group living in Burkina Faso. Am J Trop Med Hyg 2003;69:31-5.[Abstract/Free Full Text]
  23. Armitage P, Berry G, Matthews JNS. Using STATA’s robust cluster command as appropriate. Statistical methods in medical research.: Oxford, United Kingdom: Blackwell Scientific Publications, 2001.
  24. Genton B, Al-Yaman F, Ginny M, Taraika J, Alpers MP. Relation of anthropometry to malaria morbidity and immunity in Papua New Guinean children. Am J Clin Nutr 1998;68:734-41.[Abstract]
  25. Gera T, Sachdev HP. Effect of iron supplementation on incidence of infectious illness in children: systematic review. BMJ 2002;325:1142.[Abstract/Free Full Text]
  26. Oppenheimer SJ. Iron and its relation to immunity and infectious disease. J Nutr 2001;131:616S-33S; discussion 633S–5S.[Abstract/Free Full Text]
  27. Nussenblatt V, Semba RD. Micronutrient malnutrition and the pathogenesis of malarial anemia. Acta Trop 2002;82:321-37.[Medline]
  28. Scrimshaw NS, Taylor CE, Gordon JE. Interactions of nutrition and infection. World Health Organ Monogr Ser 1968;57:3-329.[Medline]
  29. Alonso PL, Lindsay SW, Armstrong Schellenberg JR, et al. A malaria control trial using insecticide-treated bed nets and targeted chemoprophylaxis in a rural area of The Gambia, west Africa. 6. The impact of the interventions on mortality and morbidity from malaria. Trans R Soc Trop Med Hyg 1993;87(suppl 2):37-44.[Medline]
  30. Mbogo CN, Snow RW, Kabiru EW, et al. Low-level Plasmodium falciparum transmission and the incidence of severe malaria infections on the Kenyan coast. Am J Trop Med Hyg 1993;49:245-53.[Abstract/Free Full Text]
  31. Tracey KJ, Cerami A. Tumor necrosis factor in the malnutrition (cachexia) of infection and cancer. Am J Trop Med Hyg 1992;47:2-7.[Medline]
  32. Williams TN, Maitland K, Phelps L, et al. Plasmodium vivax: a cause of malnutrition in young children. QJM 1997;90:751-7.[Abstract]
Received for publication February 5, 2004. Accepted for publication August 5, 2004.




This article has been cited by other articles:


Home page
 Am J Trop Med HygHome page
S. Gwer, C. R.J.C. Newton, and J. A. Berkley
Over-Diagnosis and Co-Morbidity of Severe Malaria in African Children: A Guide for Clinicians
Am J Trop Med Hyg, December 1, 2007; 77(6_Suppl): 6 - 13.
[Abstract] [Full Text] [PDF]

Home page
 Infect. Immun.Home page
B. C. Urban, D. Cordery, M. J. Shafi, P. C. Bull, C. I. Newbold, T. N. Williams, and K. Marsh
The Frequency of BDCA3-Positive Dendritic Cells Is Increased in the Peripheral Circulation of Kenyan Children with Severe Malaria
Infect. Immun., December 1, 2006; 74(12): 6700 - 6706.
[Abstract] [Full Text] [PDF]

Home page
 Am J Trop Med HygHome page
B. C. URBAN, M. J. SHAFI, D. V. CORDERY, A. MACHARIA, B. LOWE, K. MARSH, and T. N. WILLIAMS
FREQUENCIES OF PERIPHERAL BLOOD MYELOID CELLS IN HEALTHY KENYAN CHILDREN WITH {alpha}+ THALASSEMIA AND THE SICKLE CELL TRAIT
Am J Trop Med Hyg, April 1, 2006; 74(4): 578 - 584.
[Abstract] [Full Text] [PDF]

Home page
 Am J Trop Med HygHome page
J. F. FRIEDMAN, A. M. KWENA, L. B. MIREL, S. K. KARIUKI, D. J. TERLOUW, P. A. PHILLIPS-HOWARD, W. A. HAWLEY, B. L. NAHLEN, Y. P. SHI, and F. O. T. KUILE
MALARIA AND NUTRITIONAL STATUS AMONG PRE-SCHOOL CHILDREN: RESULTS FROM CROSS-SECTIONAL SURVEYS IN WESTERN KENYA
Am J Trop Med Hyg, October 1, 2005; 73(4): 698 - 704.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow An erratum has been published
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nyakeriga, A. M
Right arrow Articles by Williams, T. N
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nyakeriga, A. M
Right arrow Articles by Williams, T. N
Agricola
Right arrow Articles by Nyakeriga, A. M
Right arrow Articles by Williams, T. N

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS