Postweaning Exposure to Aflatoxin Results in Impaired Child Growth:

[Guest guest]

http://ehp.niehs.nih.gov/realfiles/members/2004/6954/6954.html

Children's Health Article

---------------------------------------------------------------------

-----------

Environmental Health Perspectives Volume 112, Number 13, September

2004

Postweaning Exposure to Aflatoxin Results in Impaired Child Growth:

A Longitudinal Study in Benin, West Africa

Yunyun Gong,1 Assomption Hounsa,2 Sharif Egal,2 C. ,1

Anne E. Sutcliffe,1 J. Hall,3 Kitty Cardwell,2 and

P. Wild1

1Molecular Epidemiology Unit, Centre for Epidemiology and

Biostatistics, Leeds Institute of Genetics Health and Therapeutics,

Faculty of Medicine and Health, University of Leeds, Leeds, United

Kingdom; 2International Institute of Tropical Agriculture, Cotonou,

Benin, West Africa; 3London School of Hygiene and Tropical Medicine,

London, United Kingdom

Introduction

Materials and Methods

Results

Discussion

Full Article in PDF

EHP-in-Press

Abstract

Aflatoxins are dietary contaminants that are hepatocarcinogenic and

immunotoxic and cause growth retardation in animals, but there is

little evidence concerning the latter two parameters in exposed

human populations. Aflatoxin exposure of West African children is

known to be high, so we conducted a longitudinal study over an 8-

month period in Benin to assess the effects of exposure on growth.

Two hundred children 16-37 months of age were recruited from four

villages, two with high and two with low aflatoxin exposure (50

children per village). Serum aflatoxin-albumin (AF-alb) adducts,

anthropometric parameters, information on food consumption, and

various demographic data were measured at recruitment (February) and

at two subsequent time points (June and October). Plasma levels of

vitamin A and zinc were also measured. AF-alb adducts increased

markedly between February and October in three of the four villages,

with the largest increases in the villages with higher exposures.

Children who were fully weaned at recruitment had higher AF-alb than

did those still partially breast-fed (p < 0.0001); the major weaning

food was a maize-based porridge. There was no association between AF-

alb and micronutrient levels, suggesting that aflatoxin exposure was

not accompanied by a general nutritional deficiency. There was,

however, a strong negative correlation (p < 0.0001) between AF-alb

and height increase over the 8-month follow-up after adjustment for

age, sex, height at recruitment, socioeconomic status, village, and

weaning status; the highest quartile of AF-alb was associated with a

mean 1.7 cm reduction in growth over 8 months compared with the

lowest quartile. This study emphasizes the association between

aflatoxin and stunting, although the underlying mechanisms remain

unclear. Aflatoxin exposure during the weaning period may be

critical in terms of adverse health effects in West African

children, and intervention measures to reduce exposure merit

investigation. Key words: aflatoxin, biomarkers, child growth,

dietary exposure, longitudinal study, weaning. Environ Health

Perspect 112:1334-1338 (2004). doi:10.1289/ehp.6954 available via

http://dx.doi.org/ [Online 27 April 2004]

---------------------------------------------------------------------

-----------

Address correspondence to C.P. Wild, Molecular Epidemiology Unit,

Centre for Epidemiology and Biostatistics, Leeds Institute of

Genetics Health and Therapeutics, Faculty of Medicine and Health,

University of Leeds, Leeds, UK LS2 9JT. Telephone: 0113-343-6601.

Fax: 0113-343-6603. E-mail: c.p.wild@...

We thank the following people who participated in the fieldwork: A.

Agognon, Z. Assani, G. Ayeni, M. Elegbede, A. Gandjeto, M. Koube,

and J. Djossou.

This study was supported by Gesellschaft für Technische

Zusammenarbeit project 98.7860.4-001.00 and National Institute of

Environmental Health Sciences grant ES06052.

The authors declare they have no competing financial interests.

Received 6 January 2004; accepted 27 April 2004.

Introduction

Aflatoxins are fungal metabolites that contaminate dietary staple

foods such as groundnuts and maize in agroecologies where hot, humid

climates combine with poor food storage conditions to facilitate

fungal growth and toxin production [international Agency for

Research on Cancer (IARC) 2002]. Aflatoxins are proven

hepatocarcinogens in many animal species. In populations in parts of

Africa and Southeast Asia, exposure is associated with an increased

risk of hepatocellular carcinoma, particularly in individuals with

chronic hepatitis B virus infection (Hall and Wild 1994; IARC 2002;

Wild and 2002). In addition to their carcinogenic properties,

aflatoxins can cause growth retardation and impairment of immune

function in animals (Raisuddin et al. 1993). However, to date there

has been little investigation of these latter parameters in exposed

human populations. In one study of Gambian children, et al.

(2003) found evidence of a reduced level of salivary immunoglobulin

A (IgA) in exposed individuals but no effect on antibody titers to

pneumococcal and rabies vaccines.

Aflatoxin exposure cannot be measured accurately at the individual

level through a combination of questionnaire-based approach and food

analysis, primarily because the heterogeneity of toxin distribution

within a particular food product makes representative sampling

impractical. Exposure biomarkers have been developed to circumvent

this problem, including serum aflatoxin-albumin (AF-alb) adducts

that reflect recent past exposure (previous 2-3 months) (Wild and

2002). In a cross-sectional study in Benin and Togo, young

children showed a consistently high prevalence and level of AF-alb,

with detection of the marker in 99% of children [geometric mean

(GM), 32.8 pg/mg; 95% confidence interval (CI), 25.3-42.5]. Exposure

was significantly related to weaning status in children 1-3 years of

age, with mean AF-alb levels approximately 2-fold higher in fully

weaned children compared with those receiving a mixture of breast

milk and solid foods. Furthermore, the level of AF-alb was strongly

associated with growth faltering, particularly stunting (Gong et al.

2002, 2003). Although breast milk may contain aflatoxins (Zarba et

al. 1992), these are generally less toxic metabolites (AFM1) than

are the parent toxins found in the diet (AFB1, AFG1), and they occur

at lower levels. Thus, breast-feeding provides a period of

relatively low aflatoxin exposure in a population whose primary

weaning foods, particularly maize, are at high risk of

contamination. Toxin exposure during the postweaning period may be a

critical factor in young children in determining the adverse health

effects of aflatoxins in terms of growth, immune status, and

eventually liver cancer risk.

Our earlier study of aflatoxin in relation to weaning and growth was

of cross-sectional design (Gong et al. 2002, 2003). The study

reported here is of longitudinal design over 8 months examining

these associations with repeat measures of aflatoxin exposure and

anthropometry.

Materials and Methods

Subject recruitment and survey time. Fifty children (16-37 months of

age) from each of the four villages (Bagbe, Sedje, Djidja, and Dovi-

Cogbe) in Benin were recruited into the study in February 2001 and

were followed up in June and October 2001. Bagbe and Sedje are

located in the coastal savannah (CS), the southernmost zone in the

country, and were expected to have lower aflatoxin exposure. Djidja

and Dovi-Cogbe are in southern Guinea savannah (SGS), the zone

immediately to the north of CS, and were expected to have higher

aflatoxin exposure. Rainfall and humidity decrease from south to

north (Hell et al. 2000a). The two agroecologic zones each have two

maize-growing seasons per year. SGS was the zone with the highest

aflatoxin exposure in the country in our previous study (Gong et al.

2002, 2003). Ethical approval was obtained from the Ministry for

Health in Benin. The head of household and the mother of the chosen

child were informed about the nature of the study and, if they

agreed to participate, signed a statement of informed consent.

A questionnaire, administered by trained interviewers to the mothers

of children recruited to the study, obtained information on the

child, namely, age, sex, food consumption (including frequency of

maize and groundnut consumption during 3 days before blood

sampling), weaning status, weaning foods, and general health status.

Information was also obtained at each of the later two survey

points. Additional data were gathered concerning the economic status

of the household and the mother. These were used to generate an

index of relative socioeconomic status (SES) based upon actual

material belongings and potential for income generation.

Questionnaires were administered at each of the three survey

periods. Only the mother's SES was used in the analysis because it

was considered more relevant to the child's diet (Gong et al. 2002,

2003). The mother's mean SES calculated from the February and

October surveys was used for the analysis. Data collected at the

February survey with regard to personal information (parent's

religion and ethnicity and mother's education and body mass index)

were used in the analysis.

Aflatoxin exposure assessment. A 5-mL blood sample was obtained from

each child at each survey date. The serum was separated and the

samples stored at -20°C in Benin before shipment on dry ice to the

University of Leeds for analysis. The levels of AF-alb adduct were

determined after albumin extraction, digestion, and enzyme-linked

immunosorbent assay (ELISA) as previously described (Chapot and Wild

1991). The detection limit was 3 pg of aflatoxin-lysine equivalents

per milligram albumin. Controls included three positive and one

negative control analyzed alongside batches of samples. Samples were

measured in ELISA in quadruplicate on at least two occasions on

separate days.

Blood micronutrients. Plasma vitamin A was measured by reverse-phase

high-performance liquid chromatography by the method of Thurnham et

al. (1988) with the minor modification that hexane was used for

extraction instead of heptane. Zinc was measured by atomic

absorption spectroscopy.

Anthropometry. Child and mother's body weight and height were

measured at all three survey dates, using accurately calibrated

instruments [electronic scales: Soehnle (BCB Ltd., Cardiff, UK), 20

kg maximum weight, accurate to 10 g; height measurement: Schorr

(Olney, land, USA)]. Field workers, trained to maximize

repeatability, made all height and weight measurements. Height for

age Z-score (HAZ), weight for age Z-score (WAZ), and weight for

height Z-score (WHZ) were calculated at the end of the study

(October) as described previously (Gong et al. 2002, 2003),

according to the World Health Organization reference population (WHO

1986).

Statistical analysis. The AF-alb adduct data were not normally

distributed and were natural log transformed for statistical

analysis. The mean AF-alb level from all three surveys for a given

individual was calculated and used as a measure of aflatoxin

exposure in some of the analyses. Growth velocity was calculated

either as the height difference between two survey points or the

difference over the whole 8-month period. The difference between

means was tested by t-test or analysis of variance (ANOVA).

Significant variables of age, village, and mother's SES were entered

into multivariable models to analyze effects of weaning status on AF-

alb level and AF-alb on growth velocity. All the analyses were

performed using Stata version 8.0 software (StataCorp., College

Station, TX, USA). GMs for AF-alb with 95% CIs are reported in the

tables and text unless otherwise stated.

Results

Table 1.

Figure 1. AF-alb adduct GM level across the four villages at the

three survey time points. The " Total " bars show the adduct level of

all the villages at three survey points, with the overall adduct

level at the October survey being significantly higher than the

other two survey points (p < 0.01). Djidja and Dovi-Cogbe had

significantly higher AF-alb than did Bagbe and Sedje at both the

October and June survey points (p < 0.01 for both). However, at the

February survey, the AF-alb adduct levels are different across all

the villages (p < 0.01).

Demographic data for the 200 children at recruitment (February) are

presented by village in Table 1. There were no significant

differences in age and sex distribution by village. The majority

religion was Christian in three of the four villages, with Voodoo

being the most common in Djidja. Dovi-Cogbe had the lowest mean

measure of mother's SES, whereas Djidja had the highest. In terms of

the major dietary sources of aflatoxin during the period of the

study, most of the children (> 80%) in all four villages had

consumed maize (including in weaning foods) daily in the 3 days

before recruitment in February, and this pattern was maintained in

the latter two surveys. In contrast to the almost uniform

consumption of maize, the frequencies of groundnut consumption in

February in the four villages did differ significantly (p < 0.0001).

Groundnut consumption was more common at this time in Djidja and

Dovi-Cogbe than in the other villages (Table 1). The same general

pattern was also found at the two later survey points, except for

somewhat increased groundnut consumption in Bagbe in June (data not

shown).

AF-alb levels. AF-alb was detected in almost every individual at all

time points, with a prevalence of 98, 99.5, and 100% in February,

June, and October respectively. At the individual level, AF-alb

showed a highly significant positive correlation between the three

survey points (r = 0.6253 for February vs. June, 0.5624 for February

vs. October, and 0.5398 for June vs. October; p < 0.0001 in all

cases), suggesting that individuals track over time in terms of

their exposure level, although this was predominantly a feature of

the higher exposure villages (data not shown). There was no

association between AF-alb adduct levels and sex of the child or

mother's SES, body mass index, or level of education. Although there

were some differences in adduct level in relation to religious

group, these differences were not significant in multivariable

analysis (data not shown). Mother's SES was included in the

multivariable analysis because of the relatively strong rationale

behind this parameter affecting the child's diet and hence aflatoxin

exposure and, more generally, as a way of controlling for unmeasured

dietary confounding. In addition, plasma vitamin A was measured in

February and June and zinc in June to assess whether dietary

deficiency in these nutrients has a confounding effect on the

association between aflatoxin exposure and growth. However, neither

nutrient was correlated with AF-alb level (data not shown).

More frequent groundnut consumption was correlated with higher AF-

alb in a univariate analysis (Spearman correlation, p < 0.0001).

However, groundnut consumption did not make a significant

contribution to AF-alb level after adjusting for age, weaning

status, village, and SES (p = 0.256). There was no significant

correlation between maize consumption and AF-alb, which is probably

explained by the relatively uniform consumption frequency.

The AF-alb levels among the four villages showed a similar pattern

to that predicted from earlier work, in that Dovi-Cogbe and Djidja

had the highest exposures. Longitudinally, AF-alb levels did not

differ between February and June (GMs across all four villages, 37.4

vs. 38.7 pg/mg, respectively) but were markedly higher in October

(GM, 86.8 pg/mg, p < 0.0001) compared with both February and June.

This pattern was in general observed for each of the villages

individually, although the increase was particularly marked in the

two higher-exposure villages, whereas in Sedje there was little

variation in AF-alb over the 8-month period (Figure 1).

Village was one of the strongest determinants of AF-alb in the

present study together with the time of sampling. To analyze the

contribution of both the village and the timing of sampling to the

AF-alb, a repeated ANOVA model was used. This analysis showed that

village made the major contribution to AF-alb (F = 89.7, p < 0.0001)

followed by the survey time point (F = 46.8, p < 0.0001).

Weaning food and weaning practice. In addition to the data on foods

consumed by the children during this 8-month study, we also obtained

information concerning the introduction of weaning foods for each

child. Because when the children entered the study they were 16-37

months of age, these data were retrospective for most of them. None

of the children were receiving exclusively breast milk at the time

of the longitudinal study.

The mean age at which children were fully weaned was 22 months, with

the youngest recorded age for complete weaning being 9 months. By 3

years of age, all but eight children were completely weaned. In

terms of weaning foods, 95% of the children were given a maize-based

porridge, although in Bagbe this maize-based porridge was less

frequently consumed (only 64%) than in the other three villages,

with millet and sorghum used as an alternative. The porridge was

introduced quite early in life, with 25% of the children starting

from 3 months of age and almost all children (96%) eating this food

to some degree by 7 months.

In addition to the porridge or other specified weaning foods,

different types of family foods are introduced to the child's diet

between 5 and 12 months of age. Specifically, at 5 months only 6% of

children were reported to be receiving family foods additional to

the specific weaning foods, whereas by 12 months of age 90% of the

children were consuming such foods. The data in Table 1 for maize

and peanut consumption refer to all children, both partially and

fully weaned. Patterns of weaning across villages were generally

similar, although Bagbe had a lower frequency of weaned children at

recruitment (Table 1; 44%) even though the age range did not differ

from the other villages.

Weaning and AF-alb. We examined whether increases in AF-alb with age

can be explained by the change in weaning status. As expected in a

cohort of this age group, the percentage of fully weaned children

increased over the three survey dates from 64% (February) to 79%

(June) to 96% (October). When examining the relationship between age

and AF-alb, there was a strong positive correlation at recruitment

(February), but this became progressively less significant over time

and was no longer significant at the end of the 8-month follow-up (p

= 0.001, 0.033, and > 0.05 for the February, June, and October time

points, respectively).

Figure 2. Adjusted mean AF-alb level in weaned and partially weaned

children. Data used in this figure are from the February survey

point, and the mean AF-alb levels (95% CI) are adjusted for age and

SES. The differences between partially breast-fed and fully weaned

children are significant in ANOVA, p = 0.0001.

Table 2.

Table 3.

To separate the effect of age on AF-alb from that of weaning status,

we dichotomized the children in to fully weaned or partially breast-

fed groups and examined the age effect in these two separate groups.

In this analysis, we found no correlation between age and AF-alb

within either group alone (data not shown). Given the small numbers

of children still partially breast-fed at the later two survey

points, we were able to conduct this analysis only with the February

data. When considering the fully weaned group of children compared

with those partially breast-fed from all villages in February, we

found 2.7-fold higher GM AF-alb level in the former group (53.5 vs.

19.5 pg/mg). The mean AF-alb adduct levels at recruitment in fully

weaned and partially breast-fed groups after adjustment for age and

SES are shown in Figure 2, revealing a higher mean AF-alb in fully

weaned children in each of the villages, even when absolute levels

of exposure are significantly different.

To take advantage of the information on change in weaning status

over time in individual children, we further categorized the

children into four groups. Two groups were partially breast-fed at

recruitment (February) but were fully weaned by June or October,

respectively, one small group of children (n = 7) were partially

breast-fed at recruitment and remained so throughout the study, and

one group were fully weaned throughout the whole study period. In

this analysis (adjusted for age at recruitment, village, and SES),

we expressed the results for each child as a ratio of the AF-alb

level in October compared with February. The increase in AF-alb was

significantly different between the four groups over the 8-month

period (F = 4.50, p = 0.0046; Table 2); however, among the four

groups, the greatest increase occurred in children who were fully

weaned between February and June and, to a lesser extent, in those

fully weaned between June and October. This is further evidence that

the change from partial breast-feeding to fully weaned is associated

with an increase in aflatoxin exposure.

Growth and AF-alb. When AF-alb levels, either in February or the

mean level from the three survey points, were analyzed by quartiles,

there was a significant inverse correlation with HAZ and WHZ score

but not WAZ score. After adjustment for age, sex, height, weaning

status (all data from the February point), SES, and village, a

significant correlation remained between HAZ and both measures of AF-

alb (p = 0.009 for February AF-alb, p < 0.0001 for mean AF-alb over

three survey points), but there was no significant correlation

between WHZ and AF-alb.

Height and weight were measured at each of the survey dates. The

increase in height and weight between the first and last time point

(8 months apart) was calculated and compared with mean AF-alb

(represented as quartiles) for each individual over the three survey

points or the level at recruitment (February). There was a

significant inverse association between mean AF-alb at the three

survey points and increase in height but not weight (data not shown)

from February to October (Table 3). This association remains highly

significant after adjustment for age, height, weaning status (all at

recruitment), SES, and village (p < 0.0001). The retardation in

height increase was 1.7 cm over the 8-month period between the

highest and lowest quartiles of aflatoxin exposure (Table 3). In

addition, when AF-alb at entry into the study was used as the

measure of exposure, the results were quite similar to those found

when exposure was integrated over the whole period (Table 3; p =

0.003).

Discussion

The present study confirmed that children in Benin have

exceptionally high aflatoxin exposure, with some individual levels

of AF-alb (> 1,100 pg aflatoxin-lysine equivalents per milligram

albumin) being higher than we have observed in any other population.

This biomarker has permitted studies of the health effects of

aflatoxin exposure that were previously precluded because of the

inability to accurately estimate individual exposure by dietary

assessment. Together with previous studies of children in other

parts of West Africa using this biomarker, a picture emerges of

consistently high levels of exposure in this part of the world

( et al. 1992; et al. 2000, 2003). The present data from

Benin are consistent with the previous cross-sectional study in

Benin and Togo with regard to both high levels and the fact that

villages in the SGS (i.e., Djidja and Dovi-Cogbe) were found to have

the highest exposure (Gong et al. 2002, 2003). In fact, the AF-alb

levels are characterized by marked geographic variations with Dovi-

Cogbe, the village with highest levels, having a mean AF-alb 10

times that in Bagbe. Overall, village was the strongest determinant

of AF-alb level. Temperature and humidity are two factors that favor

growth of Aspergillus species and production of the associated

aflatoxins as secondary metabolites, and these will vary

geographically. Harvest and storage practices also differ from

village to village, and these will influence the susceptibility of

crops to fungal infestation and toxin production (Hell et al.

2000a).

Seasonal changes in aflatoxin exposure have been reported in

previous work, presumably as a result of toxin accumulation during

storage under hot, humid conditions, often complicated by insect

infestation (Hell et al. 2000a; et al. 2000; Wild and Hall

2000). In the present study, only minor changes in AF-alb were

observed between February and June, but there was a substantial

increase between June and October in all but one village (Sedje).

These dynamics are difficult to relate directly to any specific

factor because of variations in annual climatic conditions, the fact

that there are two maize harvests per annum, and that maize can be

stored for > 1 year. Consequently, the variations in toxin level are

more complex than they are for groundnuts, a crop that tends to be

eaten within the year following harvest. The AF-alb level will also

be influenced by the frequency, quality, and quantity of maize and

groundnut consumption throughout the year due to availability of

these and alternative food sources.

The observations from this longitudinal study confirmed our earlier

report (Gong et al. 2002, 2003) that weaning onto family foods

represents a period of increasing aflatoxin exposure. Although age

was significantly correlated with AF-alb at recruitment, this

association became weaker over the study period. In further

analysis, it was apparent that weaning status was the underlying

cause of this observation, for the following reasons. First, the

correlation at the February survey between age and AF-alb level

disappeared when the correlation was considered separately in

children categorized as fully weaned or partially breast-fed.

Second, when grouping the children according to their change in

weaning status over time, we found that it was the change from

partial breast-feeding to complete weaning that was correlated with

the largest increase in AF-alb. Nevertheless, it is noteworthy that

even in those children who continued to receive some breast milk

throughout the follow-up, there was a modest increase in AF-alb,

possibly reflecting the increasing proportion of total food

consumption coming from the weaning and family foods as the child

becomes older.

The most likely source of aflatoxin exposure during the weaning

period in this population is maize. The main source of aflatoxins

will vary by region, and in other parts of West Africa groundnuts

are the major contributor to exposure ( et al. 2002); in the

present study the precision of our dietary analysis does not permit

us to completely exclude groundnuts as a contributing factor to

aflatoxin exposure, but groundnuts are eaten less frequently and in

smaller amounts than is maize. Maize is one of the main dietary

staples frequently contaminated with aflatoxins worldwide, including

in West Africa (Hell et al. 2000a, 2000b; Jelinek et al. 1989;

Setamou et al. 1997). Levels of aflatoxins in maize in Benin have

previously been reported to range up to 532 ppb (Hell et al. 2003).

Maize-based porridge was found to be the principle weaning food in

all four villages, and AF-alb levels increased when this food

replaced breast milk, probably because of the lower toxin levels in

milk compared with foods. The fact that maize-based porridge as a

weaning food is consumed less frequently in Bagbe, the village with

the lowest AF-alb level, is consistent with this interpretation that

the maize porridge is a major source of aflatoxin. In Bagbe

alternative weaning foods were sorghum and millet, and the

prevalence of fully weaned children was somewhat lower than in the

other villages. Estimated carryover of aflatoxin from dietary intake

to milk in animals is around 1%, and similar estimates were made

from studies measuring intakes versus excretion in individual women

in The Gambia (Zarba et al. 1992). An alternative hypothesis for the

increase in AF-alb after weaning is that breast milk could have an

effect on the intestinal absorption of aflatoxin or on its

metabolism to reactive metabolites once ingested. However, this

hypothesis is not supported by any experimental data so far to our

knowledge.

Fetal and early childhood environment is considered critical for

growth and disease risk in later life (Terry and Susser 2001).

Aflatoxin has been shown to cause both immune suppression and growth

impairment in animals (Hall and Wild 1994; Raisuddin et al. 1993).

Exposure has been linked to kwashiorkor, a severe protein-energy-

deficient disease in African children (Hendrickse et al. 1982);

however, this association awaits confirmation in appropriately

designed epidemiologic studies (Hall and Wild 1994). In a previous

cross-sectional study in Benin and Togo, we found an inverse

association between HAZ score and AF-alb adduct level in 480

children 1-5 years of age (Gong et al. 2002). Comparatively, growth

velocity is more persuasive than a cross-sectional measure in

clarifying the growth impairment associated with aflatoxin. In this

longitudinal study, the reduction in height increase was

significantly correlated both with higher AF-alb level at

recruitment and with high mean AF-alb level over the three time

points studied. This association was present after adjusting for

age, weaning status, height at recruitment, SES, and village.

Categorizing the children into quartiles for mean AF-alb over the

three time points in the study, there is a mean 1.7-cm reduction in

height gain in the highest versus lowest quartile of exposure over

just an 8-month period. This corresponded to a difference in GM of

160.2 pg/mg (174.2-14.0 pg/mg) in AF-alb over the 8-month period

between the lowest and highest quartiles of exposure. It should be

noted that all the levels of aflatoxin exposure in this study are

high and chronic in nature compared with developed countries. If the

effects on growth were compared with children infrequently exposed

to negligible toxin levels, the observations may appear even more

striking.

The strong association between aflatoxin exposure and impaired

growth may have significant effects on other aspects of child

health, such as immunity and susceptibility to infectious diseases.

Nevertheless, the underlying biology to explain the effect of

aflatoxin on growth is not understood and is important to

investigate. Recently, we reported a reduction in salivary IgA in

Gambian children exposed to aflatoxin ( et al. 2003). If

aflatoxins can alter mucosal barriers and affect resistance to

intestinal infections, for example, then this would provide one

mechanism for the observations we have made on growth impairment. It

is also recognized that mycotoxins occur commonly as mixtures; most

notably for the present study, aflatoxins would be expected to co-

occur with fumonisins in maize (IARC 2002), and the role of possible

interactions between these co-contaminants in the underlying

mechanisms of growth impairment is of interest. It might be argued

that AF-alb is a surrogate marker for food of poor nutritional

quality and that reduced dietary intakes of nutrients are the

underlying cause of the association between AF-alb and impaired

growth. Evidence that this is not the case comes from the fact that

blood micronutrient levels (vitamin A and zinc) were not correlated

with AF-alb levels in this study. However, we did not have a measure

of consumption of other dietary components or of total energy

intake. In fact, to fully distinguish the effects of the toxin from

other confounding factors in the diet would require a randomized

intervention study where the impact of lowering aflatoxin exposure

on child immunity, growth, and disease susceptibility can be

assessed. This would also permit a better understanding of the

relative contribution of aflatoxin to growth impairment in relation

to other important determinants in these communities. Given the

potential adverse health effects on West African children of this

ubiquitous dietary toxin, it is important to evaluate intervention

strategies appropriate to these populations (Wild and Hall 2000).

References

SJ, Wild CP, Wheeler JG, Riley EM, Montesano R, S, et

al. 1992. Aflatoxin exposure, malaria, and hepatitis B infection in

rural Gambian children. Trans R Soc Trop Med Hyg 86:426-430.

Chapot B, Wild CP. 1991. ELISA for quantification of aflatoxin-

albumin adducts and their application to human exposure assessment.

In: Techniques in Diagnostic Pathology , Vol 2 (Warhol M, van Velzen

D, Bullock GR, eds). San Diego, CA:Academic Press, 135-155.

Gong YY, Cardwell K, Hounsa A, Egal S, PC, Hall AJ, et al.

2002. Dietary aflatoxin exposure and impaired growth in young

children from Benin and Togo: cross sectional study. Br Med J 325:20-

21.

Gong YY, Egal S, Hounsa A, PC, Hall AJ, Cardwell KF, et al.

2003. Determinants of aflatoxin exposure in young children from

Benin and Togo, West Africa: the critical role of weaning. Int J

Epidemiol 32:556-562.

Hall AJ, Wild CP. 1994. Epidemiology of aflatoxin related disease.

In: The Toxicology of Aflatoxins: Human Health, Veterinary and

Agricultural Significance (Eaton DA, Groopman JD, eds). San Diego

CA:Academic Press, 233-258.

Hell K, Cardwell KF, Poehling HM. 2003. Relationship between

management practices, fungal infection and aflatoxin for stored

maize in Benin. J Phytopathol 151:690-698.

Hell K, Cardwell KF, Setamou M, Poehling H-M. 2000a. The influence

of storage practices on aflatoxin contamination in maize in four

agroecological zones of Benin, West Africa. J Stored Prod Res 36:365-

382.

Hell K, Cardwell KF, Setamou M, Schulthess F. 2000b. Influence of

insect infestation on aflatoxin contamination of stored maize in

four agroecological regions in Benin. Afr Entomol 8:169-177.

Hendrickse RG, Coulter JB, Lamplugh SM, Macfarlane SB, TE,

Omer MI, et al. 1982. Aflatoxins and kwashiorkor: a study in

Sudanese children. Br Med J [Clin Res] 285:843-846.

IARC. 2002. Some traditional herbal medicines, some mycotoxins,

naphthalene and styrene. IARC Monogr Eval Carcinog Risks Hum 82:1-

556.

Jelinek CF, Pohland AE, Wood GE. 1989. Worldwide occurrence of

mycotoxins in foods and feeds--an update. J Assoc Off Anal Chem

72:223-230.

Raisuddin S, Singh KP, Zaidi SI, BN, Ray PK. 1993.

Immunosuppressive effects of aflatoxin in growing rats.

Mycopathologia 124:189-194.

Setamou M, Cardwell KF, Schulthes F, Hell K. 1997. Aspergillus

flavus infection and aflatoxin contamination of preharvest maize in

Benin. Plant Disease 81:1323-1327.

Terry MB, Susser E. 2001. Commentary: the impact of fetal and infant

exposures along the life course. Int J Epidemiol 30:95-96.

Thurnham DI, E, Flora PS. 1988. Concurrent liquid-

chromatographic assay of retinol, alpha-tocopherol, beta-carotene,

alpha-carotene, lycopene, and beta-cryptoxanthin in plasma, with

tocopherol acetate as internal standard. Clin Chem 34:377-381.

PC, Mendy M, Whittle H, Fortuin M, Hall AJ, Wild CP. 2000.

Hepatitis B infection and aflatoxin biomarker levels in Gambian

children. Trop Med Int Health 5:837-841.

PC, SE, Hall AJ, Prentice AM, Wild CP. 2003.

Modification of immune function through exposure to dietary

aflatoxin in Gambian children. Environ Health Perspect 111:217-220.

PC, Sylla A, Diallo MS, Castegnaro JJ, Hall AJ, Wild CP.

2002. The role of aflatoxins and hepatitis viruses in the

etiopathogenesis of hepatocellular carcinoma: a basis for primary

prevention in Guinea-Conakry, West Africa. J Gastroenterol Hepatol

17:S441-S448.

WHO. 1986. WHO working group: Use and interpretation of

anthropometric indicators of nutritional status. Bull WHO 64:929-

941.

Wild CP, Hall AJ. 2000. Primary prevention of hepatocellular

carcinoma in developing countries. Mutat Res 462:381-393.

Wild CP, PC. 2002. The toxicology of aflatoxins as a basis

for public health decisions. Mutagenesis 17:471-481.

Zarba A, Wild CP, Hall AJ, Montesano R, Hudson GJ, Groopman JD.

1992. Aflatoxin M1 in human breast milk from The Gambia, West

Africa, quantified by combined monoclonal antibody immunoaffinity

chromatography and HPLC. Carcinogenesis 13:891-894.

---------------------------------------------------------------------

-----------

Last Updated: August 10, 2004

FAIR USE NOTICE:

This site contains copyrighted material the use of which has not

always been

specifically authorized by the copyright owner. We are making such

material

available in our efforts to advance understanding of environmental,

political, human rights, economic, democracy, scientific, and social

justice

issues, etc. We believe this constitutes a 'fair use' of any such

copyrighted material as provided for in section 107 of the US

Copyright Law.

In accordance with Title 17 U.S.C. Section 107, the material on this

site is

distributed without profit to those who have expressed a prior

interest in

receiving the included information for research and educational

purposes.

For more information go to:

http://www4.law.cornell.edu/uscode/17/107.html

If you wish to use copyrighted material from this

site for purposes of your

own that go beyond 'fair use', you must obtain permission from the

copyright

owner.