- Email: [email protected]
Quantitative assessment of geophagous behaviour as a potential source of exposure to geohelminth infection
TRANSACTKINS OF THE ROYAL .%~IETY OF TROPICAL ~~EDIUNE AND HYGIENE (1988) 82, 621-625
Quantitative assessment exposure to geohelminth
of geophagous infection
behaviour
as a potential
621
source of
M. s. wong”, D. A. P. Bundyz” and M. H. N. Golden3 ‘Department of Zoology, Mona Campus, University of the WestIndies, Kingston 7, Jamaica; ‘Parasite Epidemiology ResearchGroup, Department of Pure & Applied Biology, Imperial College, Prince Consort Road, London, SW7 2BB, UK; 3Tropical Metabolism Research Unit, Faculty of Medicine, Universi& of the West Indies, Kingston 7, Jamaica Abstract The most common form of pica, geophagy, has direct adverse nutritional effects and also exposes children to soil-borne infection. Existing methods for assessinggeophagy are either inappropriate for field use (radiology) or unreliable (reporting). A new method is described, based on the measurement of soil-derived silica in stools. More than 90% of silica is excreted within one gut transit period of ingestion. The amount excreted-is proportional to the amount ingested. Faecal levels of dietarv silica (<2% drv wt stool) can be distinguished from levels due to geophagy (Up to 25% dry wt stool). Studies in 2 children’s homes in Jamaica showed that 33% and 66% of children were geophagous, ingesting up to 10 p; soil dav-‘. The KeoDhatzvof t20% of the &i&en accounted for 56&/, ;f the total soil ingested. This overdispersion of exposure to soil-borne infection may contribute to the observed aggregation of geohelminth infection. Illtrodllctioll Pica is a perversion of appetite, characterized by persistent and purposeful ingestion of unsuitable substancesof no apparent nutrient value (LANZKOWSKY, 1959). The most common manifestation of pica is geophagy (soil, dirt or clay eating), perhaps because soil is the material most readilv available to the behaviourally affected patient {HALSTEAD, 1968; VERMEER& FRATE 1979). The major health consequences of geophagy are generally considered to result from direct nutritional effects, such as nutrient substitution and chelation of trace elements (LANZKOWSKY,1959.; REINHOLD et al., 1974; VESSALet al., 1975). The mgestion of soil also has signif&mt indirect consequencesfor health, since it exposes the affected patient to the risk of soil-borne infection, particularly with geohehninths (LARIWBREet al., 1965; BUNDY, 1986). It has been shown, for example, that toxocariasis and severe trichuriasis are statistically associatedwith a reported history of geophagy (CLICKMANet al., 1981; GILMAN et al., 1983). Attempts accurately to assessthese risks, however, have been frustrated by the lack of a quantitative technique for determining individual soil ingestion rates. Radiographic techniques provide an objective diagnosis of geophagy (GARDNER& TEVETOGLU, 1957; VESSAL et al., 1973, but are too cumbersome for use in disease endemic communities. ‘Present address: Parasite Control Programme, Glendon Hospital, Plymouth, Montserrat, West Indies. Author for correspondence.
Patient histories are readily collected, but are notoriously unreliable for assessingbehavioural characteristicicdeee;marly when the reported habit is socially The aim of the present study was to develop and test a method for assessing geophagy which is objective, quantitative and appropriate to field applications. The method is based on determining the level of soil-derived silica in stool as an indirect indicator of the quantity of soil ingested. This method assumes that soil-derived silicates are qualitatively and quantitatively unchanged during gut transit and therefore that the amount excreted is in direct proportion to the amount ingested. The method also assumes that normal (i.e, dietary) levels of silica in stool may be differentiated from levels resulting from non-dietary sources. The paper describes the validation of these assumptions, the standardization of the methodology, and the use of the technique to determine the rates of geophagous activity in children living in Jamaica. Materials and Methods Method for estimating the concentration of silicates in stool This method is adapted from the incineration acid wash (ICW) procedure for determining the crude silica content of soil (HESSE,1971). based on the removal of organic material by high temperature incineration and dissolving inorganic materials in concentrated acid, leaving a residue of acid-insoluble inert silicates. This procedure has been developed and standardized to ensure optimal efficiency (WONG,1988). 1 g of faeces is dried to constant weight at 105X, and then incinerated in a muffle furnace at 800°C for 1 h. The ash is cooled in a desiccator, covered with 6~ HCI for 12 h, then filtered under vacuum on to Whatman 541 filter pads in a Millipore apparatus. The residue of inert silicates is then dried to constant weight under desiccation at 50°C. EJiciency of rewwy of silicates added to stool This experiment sought to determine the efficiency with which known amounts of silicate added to stool can be recovered by the above method. A standard preparation of silicates was prepared from a soil sample using established procedures (HESSE, 1971). 2 g specimens of faeces from a non-geophagous volunteer were dried to constant weight at 105°C and a known weight of standard silicates added to provide 9 silicate-stool mixtures,
622 with proportions ranging from
to estimate for each child the mean proportion of silicates in stool and the mean quantity of silicate excreted in a 24 h period. Soil samples were- collected from the unsurfaced play areasof both homes and the crude silica content determined by the ICW procedure. Assuming direct proportionality between rate of silica ingestion and excretion it was therefore possible to estimate the mean quantity of soil ingested by each child in a 24 h
Efficiency of recovery of ingested silica
period.
This experiment examined the relationship between the quantity of silicate ingested and the amount of silicate recoverable from stool. 2 non-geophagous volunteers (males aged 26 years) each ingested gelatin capsules containing approximately 4 g of soil-derived silicate (prepared using standard procedures; HESSE, 1971) and O-2 g of carmine dye. All stools passedwere collected until no carmine die was present. 5 replicate aliquots of each stool were examined using the ICW procedure, and their silicate content compared with that of a control faecal sample collected from each volunteer before the experiment. Estimation of stool silicate concentration due to dietary sources
This experiment sought to determine the normative concentration of silicates in the stool of non-geophagous West Indian children receiving a normal diet. In order to ensure the absence of geophagous behaviour the investigation was conducted on children conlined to bed in hospital. Under such conditions even persistent geophagous individuals exhibit no evidence of soil in the colon 3-10 d after admission (GARDNER& TEVETOGLU,1957). 30 patients were recruited to the study from the paediatric wards of the University Hospital of the West Indies and the Bustamante Hosnital for children in Kingston? Jamaica. Recruitment was based on the following crtteria: no history of pica; confined to bed while in hospital for more than 7 d; admitted for reasons other than nutritional or gastroenterological disease; free of geohehninth infection at the time of the study. The children received a diet judged by the dietitians to be qualitatively similar to that obtaining in the wider community. A single stool sample was collected from each child and the stool silicate concentration determined by the ICW procedure. Estimation of geophagy in children
This study estimated the amounts of soil ingested by children and examined the prevalence and population distribution of geophagous behaviour in 2 ‘Places-of-Safety’ in urban Kingston, Jamaica. Although intended as temporary refuges for abandoned or abused children, these institutions function in practice as long-stay residential children’s homes. Children were recruited to the study on the following criteria: age range 6 months-14 years; absence of overt clinical disease;no reported history of behavioural abnormality; resident at the home for at least 3 months. Stools were collected from each child over a 24 h period on 4 sequential occasions. The total weight of each 24 h stool collection was determined and the silicate concentration estimated from duplicate analysesusing the ICW procedure. By taking the mean of the 4 replicate stool collections it was possible
This study was conducted with the agreement of the Ministrv of Health and the Child Welfare Division of the M&try of Justice. All investigations were made with the consent of the subjects or, in the caseof young children, their guardians. The subjects were under the care of a physician independent of the study team throughout the course of the investigations. Results Efficiency of recovery of silicates added to stool
Table 1 shows the efficiency of recovery of silicates added in various proportions to faecal samples. An overall mean of 91*2+ 12.2% of added silicate was recoverable by the ICW procedure. Efficiency of recovery of ingested silica
Volunteer 1 uassed4 stools in a 44 h oeriod after ingesting the capsules containing silica and carmine dye (Table 2). Dye was apparent in the second and third stools. The base-line stool specimen contained 1.35% silicate by dry weight of stool. Silicate concentrations in excessof this presumed dietary level were observed in the second, third and fourth stools, with
Table l-Efficiency various proportions
of recovery of silica added in to faecal samples’
Silicate in mixture (%)
Efficiency of recovery of silicate (%)
81.6 24.1 2.1 1.7 1.0 ;:;
1:*: 87-O
97.7 91.2 82.6 69.4 63.6 8:: 125.6 “Each sample contained a mean of 0.13% silica from dietary sources. Table 2-Efficiency volunteers
of recovery of 4 g silica ingested by
Volunteer 1 Stools Time after Silica analysed ingestion(h) contenta 1 (base-line) 0.19 ; 20.5 12: 0.89 3.48 4
5
44
0.50
6 7
‘Percentageof wet weight.
Volunteer 2 Time after Silica ingestion(h) content’ 0.08 16O 2.54 1.55 z 88 104 112
0.50 0.17
0.0s 0.08
623 Estimation of stool silicate concentration due to dietary sources
The mean proportion of silica in the stools of 30 control children in hospital was 1.45+0*720/oof the dry weight of stool. This was taken to be the normative value for subsequent analyses.
Estimation of geophagy in children
Fig. 1. ProportionaI rcrxwery of 4 g of ingested silica from stools passedby volunteers 1 (0) and 2 (0). In both cases,approximately w% of the silica weg recovered from the !irst 2 stools passed,lessthan 48 h after ingestion.
Fig. 2. Frequency distributioo of soil ingestion rate at children’s home no. 1 61 and no. 2 a. The distribution is overdispcrscd, with mo.st children ingesting zero or ~01 g soil per day and a few practisiag more overt gcopbagy.
the largest proportion in the second stool passed 12.5 h after ingestion (Table 2). A total of 4.16 g of silicate in excessof that attributed to dietary sources was recovered from the 4 stools, a theoretical recovery efficiency of 104.1%. The carmine dye was dissolved from the stool during the ICW procedure and did not contribute to the weight of silica residue recovered from the stools. More than 90% of the total silica recovered was excreted in the second and third bowel motions, less than 24 h after ingestion (Fig. 1). Volunteer 2 passed7 stools in a 112 h period after ingestion (Table 2). Dye was apparent in the second and third stools. The base-line stool specimen contained 05% silicate. Silicate concentrations in excess of this dietary level were observed in the second, third, fourth and fifth stools, the largest proportion being in the third stool passed 40 h after ingestion (Table 2). A total of 6.34 g of silica in excessof that attributed to dietary sources was recovered from the stools, a theoretical recovery efficiency of 158.4%. Approximately 90% of the total silica recovered was excreted in the second and third stools, less than 48 h after ingestion (Fig. 1).
Four 24 h stool collections were analysed from 27 children (mean age 7.2 years) at home no. 1, and 24 children (mean age 3.1 years) at home no. 2. The mean proportion of silicate per g of dry stool ranged from 0.4% to 249%. Using the normative value for dietary silica estimated as described above, it was possible to estimate the number of children whose stools contained a higher proportion of silica than would be expected from dietary sourcesalone. On this basis, 33% of children at home no. 1 and 66% of children at home no. 2 were geophagous. The mean proportion of silica excreted in a 24 h period was estimated from the product of the mean proportion of non-dietary silica in stool and the mean weight of 24 h stool excreted. This value, divided by the proportion of silica present in the soil at the homes, provided an estimate of the quantity of soil ingested, assuming a direct relationship between soil ingestion rate and silica excretion rate. The soil at the homes contained 89.8% silica by dry weight. The estimated mean rate of soil ingestion was O-06 g day-’ for the children at home no. 1 and 0.47 g day-’ at home no. 2. The maximum amount of soil ingested by an individual on any one day of observation was 10.3 g day-‘. The frequency distribution of soil ingestion rates is shown in Fig. 2. Both distributions are overdispersed and are adequately described by the negative binomial probability distribution. The value of the negative binomial exponent k, which varies inversely with the severity of aggregation, was 0.88 for home no. 1 and 064 for home no. 2. For the populations at both homes, more than 60% of the total amount of soil ingested was ingested by less than 20% of the population. Discussion
The method developed for measuring silicates in faeces was highly efficient. On average, more than 90% of silica artificially added to faeceswas detectable by the ICW procedure. No specific attempt was made to determine the lower limits of detectability, but the method was sufficiently sensitive to detect changesin silica concentration representing only 0.4% of the dry weight of stool. The volunteer experiments indicated that this efficiency was maintained when silica was artificially added to stool in rvlvo by ingestion. The apparent excess of silica recovered corn ared to the amount ingested is attributable to the dikculties of estimating the dietary level of silica for an individual; only a single stool was used to estimate the normal dietary level of silica. Any inaccuracy in estimating this value, or any change in diet during the course of the experiment, would influence the apparent quantity of non-dietary silica present in stool. The volunteer experiments showed that approximately 90% of the ingested silica was excreted during the second and third bowel motions, between 24 and 48 h after ingestion. This time period corresponds to
624
normal gut transit times, assessedby BURKITT (1972) and CUMMINGS (1983) and indicated bv the rate of carmine dye excretion’in the present study. The small amounts of silica (
ably in the extent of their practice of geophagy: most children are only marginally geophagous while a few ingest disproportionately large amounts of soil. This overdiswrsed
distribution
of the behaviour results iu
over 6&!! of the total amount of soil ingested by the woulation being inrrested bv less than 20% of the m&duals. Tl& &plies that the more geophagous children are disproportionately exposed to soil-borne infection. Studies of the distribution of geohelminth infection intensity in human populations
have repe-
atedlv shown overdispersion, such that more than 65% of the worm population is harboured by less than 15% of the host nonulation (BIJNDY et al., 1985; ANDERSON& MEDLEY, 1985): The proportions are
tantalizingly similar to those observed for geophagy, and may indicate that the observed overdispersion of
infection is a consequence of overdispersion of exposure to the soil-borne infective stages. This relationship is unlikely to be simple, however, since the degreeof exposure is dependent not merely on the quantity
of soil ingested but also on the density of
infective stages in the soil and hence the spatial distribution
of contamiuation.
An alternative hypothesis is that the more geophagous children are more susceptible to infection due to nutritional effects. Geophagy has the effect of chelating zinc, reducing its bioavailability, and potentially causing a zinc deficiency syndrome which includes
immunosuppression (REINHOLDet al., 1974; BEISEL, 1982; GOLDEN, 1982). It has been shown that children deficient in zinc have more intense geohelrniuth infections (BUNDY& G OLDEN,1987), though not that the deficiency was due to geophagy. In conclusion, this study has shown that geophagy may be quantitatively assessedon the basis of stool silica excretion rate: the rate of silica excretion is proportional to the rate of ingestion, and normal dietary silica levels are sufficiently low to be differentiated from stool silica levels due to geophagia. It
has further been shown that geophagy is common in
children in 2 commuuities, and that the frequency of the practice has au overdispersed distribution. Further studies are under way to determine whether the infection exposure resulting from this pattern of
geophagousbehaviour is sufficient to account for the overdispersed distribution of geohehninth infections. Acknowledgements We gratefully acknowledge the cooperation of the staff of Reddies and Glenhope Places-of-Safety, andthe assistance of the Ministries of Health and Justice (Jamaica). In addition we thank Mr S. Venugopal (UHWI) and MS C. Robotham (UWI) for their advice and assistance. The study was supported by grants from the University of the West Indies and the Wellcome Trust. References Anderson, R. M. & Medley, G. F. (1985). Community control of helminth infections of man by mass and selective chemotherapy. Parasia@v, 90, 629460. Beisel, W. R. (1982). Single nutrients and immunity. American Journal of Clinical Nutrition,
35, 417-468.
Bundy, D. A. P. (1986). Epidemiological aspectsof Trichuti and trichuriasis in Caribbean communities. Transactions of the Royal Societyof Tropical Medicine and Hygiene, 80, 706-718. Bundy, D. A. P. & Golden, M. H. N. (1987). The impact of host nutrition on gastrointestinal helminth populations. Parasiwlogv, 95, 623-635. Bundy, D. A. P., Thompson, D. E., Golden, M. H. N., Cooper, E. S., Anderson, R. M. & Harlsnd, P. S. E. (1985). Population distribution of Trichuris tmhiura in a community of Jamaican children. Tramactiom of the Royal Society of Tropical Medicine and Hygiene, 79,
232-237. Bnrkitt, D. P. (1972). Varicose veins, deep vein thrombosis and haemorrhoids: e idemiology and suggested aetiology. British Medical P our&, ii, 556-561. Cummings, J. H. (1983). Dietary fibre: progress report. Gut, 14, 69-81. Gardner, J. E. & Teventoglu, F. (1957). The roentgenographic diagnosis of geophagia (dirt-eating) in children. 3ournal of Paediatrics, 51, 667-671. Gilman, R. H., Chong, Y. H., Davis, C., Greenberg, B., Virik, H. K. & Dixon H. B. (1983). The adverse
625 consequencesof heavy Trichuris infection. Transacrions of the Royal Society of Tropical Medicine and Hygiene, 77,
of cellulose consumption upon the metabolism of zinc, calcium and phosphorous in man. In: Proceedingsof the hternational Symposium on Trace Elements in Human Disease, Detroit, Michigan. New York: Academic Press, pp. 121-124. Singhi, S., Singhi, P. & Adwani, G. B. (1981). Role of psychosocial stressin the causeof pica. Clinical Paediatrics, 20, 783-785. Vermeer, D. E. & Frate, D. A. (1979). Geophagia in rural Misstssippi: environmental and cultural contexts and nutritional implications. American Journal of Clinical Nutrition, 32, 2129-2135. Vessal, K., Ronaghy, H. A. & Zarabi, M. (1975). Radiological changes in pica. American Journal of Clinical Nutrition, 28, 1095-1098. Wong, M. S. (1988). The role of enviwnmenta 1 and host behavioural factors in determining exposure w infection evith Ascaris lumbricoides and Trichuris trichiura. Ph.D. Thesis, Faculty of Natural Sciences, University of the West Indies.
432-438. Gliclunan, L. T., Chaudry, I. U., Constantine, J., Clack, F. B., Cypess, R. H. & Winslow, L. (1981). Pica patterns, toxocariasis, and elevated blood lead in children. AmericanJouma1 of Tropical Medicine and Hygiene, 30, 77-80.
Golden, M. H. N. (1982). Trace elements in human nutrition. Human Nutririon: Clinical Nunirion, 36C, 185-202. Halstead, J. A. (1968). Geophagia in man: its nature and nutritional effects. AmericanJournal of Clinical Nutrition, 21, 1384-1393. Hesse, L. (1971). A Textbook of Soil Chemical Analysis. London: John Murray. Lanxkowsky, P. (1959). Investigation into the aetiology and treatment of pica. Archiwes of Disease in Childhood, 34, 140-148. Lariviere? M., Satge, P. & Dan, V. (1%5). Les parasitoses intestmales de l’enfant africain au Senegal. I. Milieu urbain. Afiique Medicale, 32, 441-446. Reinhold, J. G., Farad& B., Abadi, P. & Ismail-Beigl, F. (1974). Binding of xmc fiber and other solids of whole meal bread with a prehminary examination of the effects
1 Announcement
Received Februay
8 February 1988
1988; accepted
for publication 24
1
Fit
Annual
Course in Infectious
Bethesda, Maryland, USA-28 Sponsored by the Armed Forces Institute Further information from: American Registry of Pathology, Armed Forces Institute of Pathology, Washington, DC 20306-6000, USA.
and Parasitic
Diseases
November - 3 December1988
of Pathology and directed by Dr B. 0. L. Duke.
Quantitative assessment exposure to geohelminth
of geophagous infection
behaviour
as a potential
621
source of
M. s. wong”, D. A. P. Bundyz” and M. H. N. Golden3 ‘Department of Zoology, Mona Campus, University of the WestIndies, Kingston 7, Jamaica; ‘Parasite Epidemiology ResearchGroup, Department of Pure & Applied Biology, Imperial College, Prince Consort Road, London, SW7 2BB, UK; 3Tropical Metabolism Research Unit, Faculty of Medicine, Universi& of the West Indies, Kingston 7, Jamaica Abstract The most common form of pica, geophagy, has direct adverse nutritional effects and also exposes children to soil-borne infection. Existing methods for assessinggeophagy are either inappropriate for field use (radiology) or unreliable (reporting). A new method is described, based on the measurement of soil-derived silica in stools. More than 90% of silica is excreted within one gut transit period of ingestion. The amount excreted-is proportional to the amount ingested. Faecal levels of dietarv silica (<2% drv wt stool) can be distinguished from levels due to geophagy (Up to 25% dry wt stool). Studies in 2 children’s homes in Jamaica showed that 33% and 66% of children were geophagous, ingesting up to 10 p; soil dav-‘. The KeoDhatzvof t20% of the &i&en accounted for 56&/, ;f the total soil ingested. This overdispersion of exposure to soil-borne infection may contribute to the observed aggregation of geohelminth infection. Illtrodllctioll Pica is a perversion of appetite, characterized by persistent and purposeful ingestion of unsuitable substancesof no apparent nutrient value (LANZKOWSKY, 1959). The most common manifestation of pica is geophagy (soil, dirt or clay eating), perhaps because soil is the material most readilv available to the behaviourally affected patient {HALSTEAD, 1968; VERMEER& FRATE 1979). The major health consequences of geophagy are generally considered to result from direct nutritional effects, such as nutrient substitution and chelation of trace elements (LANZKOWSKY,1959.; REINHOLD et al., 1974; VESSALet al., 1975). The mgestion of soil also has signif&mt indirect consequencesfor health, since it exposes the affected patient to the risk of soil-borne infection, particularly with geohehninths (LARIWBREet al., 1965; BUNDY, 1986). It has been shown, for example, that toxocariasis and severe trichuriasis are statistically associatedwith a reported history of geophagy (CLICKMANet al., 1981; GILMAN et al., 1983). Attempts accurately to assessthese risks, however, have been frustrated by the lack of a quantitative technique for determining individual soil ingestion rates. Radiographic techniques provide an objective diagnosis of geophagy (GARDNER& TEVETOGLU, 1957; VESSAL et al., 1973, but are too cumbersome for use in disease endemic communities. ‘Present address: Parasite Control Programme, Glendon Hospital, Plymouth, Montserrat, West Indies. Author for correspondence.
Patient histories are readily collected, but are notoriously unreliable for assessingbehavioural characteristicicdeee;marly when the reported habit is socially The aim of the present study was to develop and test a method for assessing geophagy which is objective, quantitative and appropriate to field applications. The method is based on determining the level of soil-derived silica in stool as an indirect indicator of the quantity of soil ingested. This method assumes that soil-derived silicates are qualitatively and quantitatively unchanged during gut transit and therefore that the amount excreted is in direct proportion to the amount ingested. The method also assumes that normal (i.e, dietary) levels of silica in stool may be differentiated from levels resulting from non-dietary sources. The paper describes the validation of these assumptions, the standardization of the methodology, and the use of the technique to determine the rates of geophagous activity in children living in Jamaica. Materials and Methods Method for estimating the concentration of silicates in stool This method is adapted from the incineration acid wash (ICW) procedure for determining the crude silica content of soil (HESSE,1971). based on the removal of organic material by high temperature incineration and dissolving inorganic materials in concentrated acid, leaving a residue of acid-insoluble inert silicates. This procedure has been developed and standardized to ensure optimal efficiency (WONG,1988). 1 g of faeces is dried to constant weight at 105X, and then incinerated in a muffle furnace at 800°C for 1 h. The ash is cooled in a desiccator, covered with 6~ HCI for 12 h, then filtered under vacuum on to Whatman 541 filter pads in a Millipore apparatus. The residue of inert silicates is then dried to constant weight under desiccation at 50°C. EJiciency of rewwy of silicates added to stool This experiment sought to determine the efficiency with which known amounts of silicate added to stool can be recovered by the above method. A standard preparation of silicates was prepared from a soil sample using established procedures (HESSE, 1971). 2 g specimens of faeces from a non-geophagous volunteer were dried to constant weight at 105°C and a known weight of standard silicates added to provide 9 silicate-stool mixtures,
622 with proportions ranging from
to estimate for each child the mean proportion of silicates in stool and the mean quantity of silicate excreted in a 24 h period. Soil samples were- collected from the unsurfaced play areasof both homes and the crude silica content determined by the ICW procedure. Assuming direct proportionality between rate of silica ingestion and excretion it was therefore possible to estimate the mean quantity of soil ingested by each child in a 24 h
Efficiency of recovery of ingested silica
period.
This experiment examined the relationship between the quantity of silicate ingested and the amount of silicate recoverable from stool. 2 non-geophagous volunteers (males aged 26 years) each ingested gelatin capsules containing approximately 4 g of soil-derived silicate (prepared using standard procedures; HESSE, 1971) and O-2 g of carmine dye. All stools passedwere collected until no carmine die was present. 5 replicate aliquots of each stool were examined using the ICW procedure, and their silicate content compared with that of a control faecal sample collected from each volunteer before the experiment. Estimation of stool silicate concentration due to dietary sources
This experiment sought to determine the normative concentration of silicates in the stool of non-geophagous West Indian children receiving a normal diet. In order to ensure the absence of geophagous behaviour the investigation was conducted on children conlined to bed in hospital. Under such conditions even persistent geophagous individuals exhibit no evidence of soil in the colon 3-10 d after admission (GARDNER& TEVETOGLU,1957). 30 patients were recruited to the study from the paediatric wards of the University Hospital of the West Indies and the Bustamante Hosnital for children in Kingston? Jamaica. Recruitment was based on the following crtteria: no history of pica; confined to bed while in hospital for more than 7 d; admitted for reasons other than nutritional or gastroenterological disease; free of geohehninth infection at the time of the study. The children received a diet judged by the dietitians to be qualitatively similar to that obtaining in the wider community. A single stool sample was collected from each child and the stool silicate concentration determined by the ICW procedure. Estimation of geophagy in children
This study estimated the amounts of soil ingested by children and examined the prevalence and population distribution of geophagous behaviour in 2 ‘Places-of-Safety’ in urban Kingston, Jamaica. Although intended as temporary refuges for abandoned or abused children, these institutions function in practice as long-stay residential children’s homes. Children were recruited to the study on the following criteria: age range 6 months-14 years; absence of overt clinical disease;no reported history of behavioural abnormality; resident at the home for at least 3 months. Stools were collected from each child over a 24 h period on 4 sequential occasions. The total weight of each 24 h stool collection was determined and the silicate concentration estimated from duplicate analysesusing the ICW procedure. By taking the mean of the 4 replicate stool collections it was possible
This study was conducted with the agreement of the Ministrv of Health and the Child Welfare Division of the M&try of Justice. All investigations were made with the consent of the subjects or, in the caseof young children, their guardians. The subjects were under the care of a physician independent of the study team throughout the course of the investigations. Results Efficiency of recovery of silicates added to stool
Table 1 shows the efficiency of recovery of silicates added in various proportions to faecal samples. An overall mean of 91*2+ 12.2% of added silicate was recoverable by the ICW procedure. Efficiency of recovery of ingested silica
Volunteer 1 uassed4 stools in a 44 h oeriod after ingesting the capsules containing silica and carmine dye (Table 2). Dye was apparent in the second and third stools. The base-line stool specimen contained 1.35% silicate by dry weight of stool. Silicate concentrations in excessof this presumed dietary level were observed in the second, third and fourth stools, with
Table l-Efficiency various proportions
of recovery of silica added in to faecal samples’
Silicate in mixture (%)
Efficiency of recovery of silicate (%)
81.6 24.1 2.1 1.7 1.0 ;:;
1:*: 87-O
97.7 91.2 82.6 69.4 63.6 8:: 125.6 “Each sample contained a mean of 0.13% silica from dietary sources. Table 2-Efficiency volunteers
of recovery of 4 g silica ingested by
Volunteer 1 Stools Time after Silica analysed ingestion(h) contenta 1 (base-line) 0.19 ; 20.5 12: 0.89 3.48 4
5
44
0.50
6 7
‘Percentageof wet weight.
Volunteer 2 Time after Silica ingestion(h) content’ 0.08 16O 2.54 1.55 z 88 104 112
0.50 0.17
0.0s 0.08
623 Estimation of stool silicate concentration due to dietary sources
The mean proportion of silica in the stools of 30 control children in hospital was 1.45+0*720/oof the dry weight of stool. This was taken to be the normative value for subsequent analyses.
Estimation of geophagy in children
Fig. 1. ProportionaI rcrxwery of 4 g of ingested silica from stools passedby volunteers 1 (0) and 2 (0). In both cases,approximately w% of the silica weg recovered from the !irst 2 stools passed,lessthan 48 h after ingestion.
Fig. 2. Frequency distributioo of soil ingestion rate at children’s home no. 1 61 and no. 2 a. The distribution is overdispcrscd, with mo.st children ingesting zero or ~01 g soil per day and a few practisiag more overt gcopbagy.
the largest proportion in the second stool passed 12.5 h after ingestion (Table 2). A total of 4.16 g of silicate in excessof that attributed to dietary sources was recovered from the 4 stools, a theoretical recovery efficiency of 104.1%. The carmine dye was dissolved from the stool during the ICW procedure and did not contribute to the weight of silica residue recovered from the stools. More than 90% of the total silica recovered was excreted in the second and third bowel motions, less than 24 h after ingestion (Fig. 1). Volunteer 2 passed7 stools in a 112 h period after ingestion (Table 2). Dye was apparent in the second and third stools. The base-line stool specimen contained 05% silicate. Silicate concentrations in excess of this dietary level were observed in the second, third, fourth and fifth stools, the largest proportion being in the third stool passed 40 h after ingestion (Table 2). A total of 6.34 g of silica in excessof that attributed to dietary sources was recovered from the stools, a theoretical recovery efficiency of 158.4%. Approximately 90% of the total silica recovered was excreted in the second and third stools, less than 48 h after ingestion (Fig. 1).
Four 24 h stool collections were analysed from 27 children (mean age 7.2 years) at home no. 1, and 24 children (mean age 3.1 years) at home no. 2. The mean proportion of silicate per g of dry stool ranged from 0.4% to 249%. Using the normative value for dietary silica estimated as described above, it was possible to estimate the number of children whose stools contained a higher proportion of silica than would be expected from dietary sourcesalone. On this basis, 33% of children at home no. 1 and 66% of children at home no. 2 were geophagous. The mean proportion of silica excreted in a 24 h period was estimated from the product of the mean proportion of non-dietary silica in stool and the mean weight of 24 h stool excreted. This value, divided by the proportion of silica present in the soil at the homes, provided an estimate of the quantity of soil ingested, assuming a direct relationship between soil ingestion rate and silica excretion rate. The soil at the homes contained 89.8% silica by dry weight. The estimated mean rate of soil ingestion was O-06 g day-’ for the children at home no. 1 and 0.47 g day-’ at home no. 2. The maximum amount of soil ingested by an individual on any one day of observation was 10.3 g day-‘. The frequency distribution of soil ingestion rates is shown in Fig. 2. Both distributions are overdispersed and are adequately described by the negative binomial probability distribution. The value of the negative binomial exponent k, which varies inversely with the severity of aggregation, was 0.88 for home no. 1 and 064 for home no. 2. For the populations at both homes, more than 60% of the total amount of soil ingested was ingested by less than 20% of the population. Discussion
The method developed for measuring silicates in faeces was highly efficient. On average, more than 90% of silica artificially added to faeceswas detectable by the ICW procedure. No specific attempt was made to determine the lower limits of detectability, but the method was sufficiently sensitive to detect changesin silica concentration representing only 0.4% of the dry weight of stool. The volunteer experiments indicated that this efficiency was maintained when silica was artificially added to stool in rvlvo by ingestion. The apparent excess of silica recovered corn ared to the amount ingested is attributable to the dikculties of estimating the dietary level of silica for an individual; only a single stool was used to estimate the normal dietary level of silica. Any inaccuracy in estimating this value, or any change in diet during the course of the experiment, would influence the apparent quantity of non-dietary silica present in stool. The volunteer experiments showed that approximately 90% of the ingested silica was excreted during the second and third bowel motions, between 24 and 48 h after ingestion. This time period corresponds to
624
normal gut transit times, assessedby BURKITT (1972) and CUMMINGS (1983) and indicated bv the rate of carmine dye excretion’in the present study. The small amounts of silica (
ably in the extent of their practice of geophagy: most children are only marginally geophagous while a few ingest disproportionately large amounts of soil. This overdiswrsed
distribution
of the behaviour results iu
over 6&!! of the total amount of soil ingested by the woulation being inrrested bv less than 20% of the m&duals. Tl& &plies that the more geophagous children are disproportionately exposed to soil-borne infection. Studies of the distribution of geohelminth infection intensity in human populations
have repe-
atedlv shown overdispersion, such that more than 65% of the worm population is harboured by less than 15% of the host nonulation (BIJNDY et al., 1985; ANDERSON& MEDLEY, 1985): The proportions are
tantalizingly similar to those observed for geophagy, and may indicate that the observed overdispersion of
infection is a consequence of overdispersion of exposure to the soil-borne infective stages. This relationship is unlikely to be simple, however, since the degreeof exposure is dependent not merely on the quantity
of soil ingested but also on the density of
infective stages in the soil and hence the spatial distribution
of contamiuation.
An alternative hypothesis is that the more geophagous children are more susceptible to infection due to nutritional effects. Geophagy has the effect of chelating zinc, reducing its bioavailability, and potentially causing a zinc deficiency syndrome which includes
immunosuppression (REINHOLDet al., 1974; BEISEL, 1982; GOLDEN, 1982). It has been shown that children deficient in zinc have more intense geohelrniuth infections (BUNDY& G OLDEN,1987), though not that the deficiency was due to geophagy. In conclusion, this study has shown that geophagy may be quantitatively assessedon the basis of stool silica excretion rate: the rate of silica excretion is proportional to the rate of ingestion, and normal dietary silica levels are sufficiently low to be differentiated from stool silica levels due to geophagia. It
has further been shown that geophagy is common in
children in 2 commuuities, and that the frequency of the practice has au overdispersed distribution. Further studies are under way to determine whether the infection exposure resulting from this pattern of
geophagousbehaviour is sufficient to account for the overdispersed distribution of geohehninth infections. Acknowledgements We gratefully acknowledge the cooperation of the staff of Reddies and Glenhope Places-of-Safety, andthe assistance of the Ministries of Health and Justice (Jamaica). In addition we thank Mr S. Venugopal (UHWI) and MS C. Robotham (UWI) for their advice and assistance. The study was supported by grants from the University of the West Indies and the Wellcome Trust. References Anderson, R. M. & Medley, G. F. (1985). Community control of helminth infections of man by mass and selective chemotherapy. Parasia@v, 90, 629460. Beisel, W. R. (1982). Single nutrients and immunity. American Journal of Clinical Nutrition,
35, 417-468.
Bundy, D. A. P. (1986). Epidemiological aspectsof Trichuti and trichuriasis in Caribbean communities. Transactions of the Royal Societyof Tropical Medicine and Hygiene, 80, 706-718. Bundy, D. A. P. & Golden, M. H. N. (1987). The impact of host nutrition on gastrointestinal helminth populations. Parasiwlogv, 95, 623-635. Bundy, D. A. P., Thompson, D. E., Golden, M. H. N., Cooper, E. S., Anderson, R. M. & Harlsnd, P. S. E. (1985). Population distribution of Trichuris tmhiura in a community of Jamaican children. Tramactiom of the Royal Society of Tropical Medicine and Hygiene, 79,
232-237. Bnrkitt, D. P. (1972). Varicose veins, deep vein thrombosis and haemorrhoids: e idemiology and suggested aetiology. British Medical P our&, ii, 556-561. Cummings, J. H. (1983). Dietary fibre: progress report. Gut, 14, 69-81. Gardner, J. E. & Teventoglu, F. (1957). The roentgenographic diagnosis of geophagia (dirt-eating) in children. 3ournal of Paediatrics, 51, 667-671. Gilman, R. H., Chong, Y. H., Davis, C., Greenberg, B., Virik, H. K. & Dixon H. B. (1983). The adverse
625 consequencesof heavy Trichuris infection. Transacrions of the Royal Society of Tropical Medicine and Hygiene, 77,
of cellulose consumption upon the metabolism of zinc, calcium and phosphorous in man. In: Proceedingsof the hternational Symposium on Trace Elements in Human Disease, Detroit, Michigan. New York: Academic Press, pp. 121-124. Singhi, S., Singhi, P. & Adwani, G. B. (1981). Role of psychosocial stressin the causeof pica. Clinical Paediatrics, 20, 783-785. Vermeer, D. E. & Frate, D. A. (1979). Geophagia in rural Misstssippi: environmental and cultural contexts and nutritional implications. American Journal of Clinical Nutrition, 32, 2129-2135. Vessal, K., Ronaghy, H. A. & Zarabi, M. (1975). Radiological changes in pica. American Journal of Clinical Nutrition, 28, 1095-1098. Wong, M. S. (1988). The role of enviwnmenta 1 and host behavioural factors in determining exposure w infection evith Ascaris lumbricoides and Trichuris trichiura. Ph.D. Thesis, Faculty of Natural Sciences, University of the West Indies.
432-438. Gliclunan, L. T., Chaudry, I. U., Constantine, J., Clack, F. B., Cypess, R. H. & Winslow, L. (1981). Pica patterns, toxocariasis, and elevated blood lead in children. AmericanJouma1 of Tropical Medicine and Hygiene, 30, 77-80.
Golden, M. H. N. (1982). Trace elements in human nutrition. Human Nutririon: Clinical Nunirion, 36C, 185-202. Halstead, J. A. (1968). Geophagia in man: its nature and nutritional effects. AmericanJournal of Clinical Nutrition, 21, 1384-1393. Hesse, L. (1971). A Textbook of Soil Chemical Analysis. London: John Murray. Lanxkowsky, P. (1959). Investigation into the aetiology and treatment of pica. Archiwes of Disease in Childhood, 34, 140-148. Lariviere? M., Satge, P. & Dan, V. (1%5). Les parasitoses intestmales de l’enfant africain au Senegal. I. Milieu urbain. Afiique Medicale, 32, 441-446. Reinhold, J. G., Farad& B., Abadi, P. & Ismail-Beigl, F. (1974). Binding of xmc fiber and other solids of whole meal bread with a prehminary examination of the effects
1 Announcement
Received Februay
8 February 1988
1988; accepted
for publication 24
1
Fit
Annual
Course in Infectious
Bethesda, Maryland, USA-28 Sponsored by the Armed Forces Institute Further information from: American Registry of Pathology, Armed Forces Institute of Pathology, Washington, DC 20306-6000, USA.
and Parasitic
Diseases
November - 3 December1988
of Pathology and directed by Dr B. 0. L. Duke.