TKANSACTIONS OFTHEROYALSOCIETY OFTROPICAL MEDICINE ANUHYC;IE~.E (1990)

Quantitative lumbricoides

assessment of contamination and Trichuris trichiura

84, 567-570

567

of soil by the eggs of Ascaris

M. S. Wang” and D. A. P. Bund++ ‘Departmen t of Zoology, University of the West Indies, Mona Campus, Kingston 7, Jamaica; ‘Parasite Epidetnioloey Research Group, Department of Pure and Applied Biology, Imperial College, Prince Consort Road, London, SW7 2BB, UK

Abstract

This study used a method of retrieving eggs from soil to examine the spatial and temporal dynamics of soil contamination with geohelminth eggs. The level of soil contamination in two children’s homes in Jamaica was determined before and after further soil contamination was prevented by chemotherapy. The home which had higher human infection levels also had a higher prevalence and density of eggs in soil. The spatial distribution of the eggs in soil was overdispersed in the home with higher levels of infection, and underdispersed in the other, perhaps due to the low density of eggs. At both localities, the proportion of soil samples containing eggs and the density of eggs in soil declined over a two-month period. The results suggestthat geohelminth eggsare rapidly depleted from the surface of tropical soils in the absenceof continuing sources of contamination. Introduction

The population dynamics of human infection with geohelminths have been the subject of extensive investigation in recent years, in an attempt to develop new strategies for control (reviewed by ANDERSON& MAY, 1985; ANDERSON& MEDLEY, 1985). These studies suggest that patterns of established infection are determined by a balance between rates of exposure to infection and rates of loss of infection due to host resistance. Whereas the immune component of the latter process has been a popular subject for study (reviewed by WAKELIN, 1985), few studies have attempted @antication of exposure to infection (BUNDY. 1988: BUNDY & BLUMENTHAL. 1990). Since infect&n is acquired by ingestion bf infhctive eggs from contaminated soil, rates of exposure to geohelminthiasis are a compound of the rate of ingestion of soil and the level of soil contamination. Methods for estimating soil ingestion rates have been proposed (BEAVER,1975; WONGet al., 1988), but few quantitative estimates of soil contamination levels have been attempted (BROWN, 1927; BEAVER, 1952; GIBODA, 1981; ISMID et al., 1983). The aims of the present study were to standardize a method that has comparable levels of helminth egg recovery from different soil types (thus making it possible to compare results from geologically different areas), to use this method to determine the density and spatial distribution of eggsin areascontaminated by human populations with either high or low intensities of infection, and to examine the persistence ‘Present address: Parasite Control Programme, Glendon Hospital, Plymouth, Montserrat, West Indies. +Author for correspondence.

of hehninth eggsin these environments in the absence of further contamination. Material Method

and Methods recovering geohelminth eggs from soil

for

Reviews of existing methods for the recovery of helminth eggs from soil indicated that the procedure of DADA (1979) has the highest recovery efficiency (67%) (WHO, 1964; WONG, 1988). This flotation procedure relies on the difference between the specific gravity of the eggs (A. lumbricoides=l*13; T. trichiura= l-15: CHEESBROUGH, 1981) and that of saturated zinc sulphate (l-45 kO.25). The following modified Dada method (WON?, 1988) was used in the present study. A one-gram sol1 sample was placed in a 20 ml test-tube which was then partly filled with water. After adding 3 drops of the detergent Triton-X100 (which reduces the adhesion between eggs and soil particles) the suspension was vortex-mixed for l-2 min (or shaken for 5 min), and then centrifuged (using a bench or hand centrifuge) at 1000 rpm for 5 min. The supernatant was then discarded, and the sediment re-suspended in sufficient zinc sulphate to give a positive meniscus. The tube was then sealed with a coverslip before centrifugation at 1000 rpm for 5 min. After standing for 5 min, the coverslip was removed, placed on a drop of glycerol on a microscope slide (to prevent crystallization), and systematically scanned for geohelminth eggs using 400~ magnification. This procedure was repeated so that eggs were recovered from the same soil sample by 3 sequential flotations. Full details of the procedure are given by WONG (1988). Efficiency of recovery of geohelminth eggs from soil

The efficiency of the modified Dada method was examined in relation to egg density, soil type, sample size and the use of pretreatment with detergent. Standardization experiments were carried out using only the eggs of T. trichiura since the eggsof both A. lumbricoides and T. trichiuru have similar specific gravities. Geohehninth eggs were obtained from human faecesusing the formoi-ether method (CHEESBROUGH,1981), and the density was estimated by examining ten 10 ml aliquots of vortex-mixed suspension. The 3 most common soil types from the study areas of urban Kingston were used for the standardization: sandy silt (particle diameter: 0.063-0.09 mm), mediim sand-(0.25-O-5 mm) and coarse sand (0.52.0 mm). Standard soils were orepared bv the Botanv Depart&ent of the University of-the W&t Indies by differential sieving through certified meshes.

568

Each home had an unsurfaced play area which was nredominantlv sandv silt at home no. 1 (total bea= rn’- and hoderate-coarse sand at home no. 2 (43 m23 1. These areas were divided into one metre square quadrats from each of which 2 replicate soil samples(each l-2 g sample taken from an areaof 0.1 m2 to a depth of 1 cm) were collected and analysed, once a month for 2 months after anthelmintic treatment. Due to logistic constraints, 38 randomly selectedquadrats (76 samples)were processedeach month from home no. 1. At home no. 2 all 43 quadrats were sampled monthly (86 samples). The eggs recovered from each soil sample were identified and the mean density of eggsper gram (epg) estimated from the 2 replicates. The surface soil rather than the deeper strata was examined becauseit was observed that the children interacted only with these superficial layers during play and episodes of geophagic behaviour (WONG, 1988).

The standardization experiments involved seeding 10 g of soil with a known densitv of T. trichiuru eas, and- determining the proportion of the inocuiurn which was recovered -from a subsample by the modified Dada method. Three exneriments were conducted for eachof the soil types: experiment (A), 5 replicates of 1 g soil samples, without detergent pretreatment; experiment (B), 10 one-gram replicates with pretreatment; and experiment (C), six 9 g replicates with pretreatment. Field study

The modilied Dada method was used to examine soil contamination at two places-of-safety located in residential areasof urban Kineston, Iamaica. Due to local socio-economic circumst&ces these tend to be long-term refuges, functioning in practice as children’s homes, and hence the child populations tend to be relativelv stable over a neriod of vears. Geohelminthiasis is a’ common feature of such institutions in Jamaica (BUNDY et al., 1985). At the beginning of the study, the prevalence of infection in the resident populations was determined coprologically, and infected children were offered 100 me mebendazole (Vermox? lanssen Pharmaceut~ca~~ Belgium) twice daily fo; 3 d. It was considered unlikely that any infections reacquired after treatment would become patent during the 2-month study period.

Results

Efficiency of recovety of geohelminth eggs from soil

The results of the standardization exneriments indicate that a pretreatment agent greatly-increased recovery efficiency (Table 1). Comparison between experiments B and C (Table 1) shows that the efficiency tends to increase with egg density, but is acceptableeven at densities less than one epg of soil. Increasing the sample size from 1 g to 9 g of soil did not significantly enhance recovery efficiency. The consistent finding of over 100% recovery efficiency under optimal conditions suggests a source of overestimation intrinsic to the standardization method; perhaps non-homogeneity in dispersion of eggsin the seeded soil sample, or inaccuracy in measuring the density of eggs in the aliquots used to seed the soil. The values of recovery efficiency should therefore be considered as of relative rather than absolute significance.

Table 1. Efficiency of recovery of Trichris tricbiura eggs from different soil types, using 1 g or 9 g soil samples, with and without detergent (T&on-X100) pretreatment. Efficiency is expressed as the percentage recovery of inoculated eggs+95% confidence limits Soil SXllPk

Sandy silt

Eat den%+

A. Without pretreatment lg 10 Ig 100 B. With pretreatment It? 3 lg 103 C. With pretreatment 9g 0.4 9g 11

Percentage recovery MdiUll Coarse sand sand -2or10

65+ 18

1229 20_+6

44*17

17_+6

82225 135+27

87f20 121+19

85+15 118+20

77rf 16 119+21

81k27 116+27

114k21

Field stuay

The prevalence of infection in the child populations, and monthly estimatesof the proportion of soil samples containing eggs and the density of geohelminth eggs in the soil, are given in Table 2 for both localities. The prevalence of human infection is positively associatedwith the initial proportion of soil samples positive and the density of eggs in soil.

77+37

“Eggsper gram Table

2. Soil contamination

and prevalence

of human infection

Ascaris lumbricoides

Time’

Human prevalence (%‘1

Soil contamination Percentage wsitive

Home no. 1 (~~76 soil sam$s/month) 0 : ::

Human prevalence

-

229

in Kingston,

Jamaica

Trichuris m‘chiura

Soil contamination

Percentage oositive

Dens@

Densit@

s=/ ?iF

(%I

l-6 128.1 0.6 O-7 /21.9 110.2

11.8 11.1 36

86

62 2

4.0 170.3 2.2 2-o 149.1 113.9

23.7 18-8 5.8

33

26

0.09 0.5

0.23

-

13 8

0.02/ o-5 owl 0.4

0.33 0.24

H;me no. 2 (ng=86soil sam$s/month) 0.071 0.7 :

at two places-of-safety

;j;;*

*Months. bEggsper gram of soil: mean/maximum. Tariancelmean.

8’;.

0.3 ;::

Figure.Initial frequencydistributionsof eggdensityin soilat two places-of-safety in Kingston,Jamaica.Arcaris lumbncmdes at (a) homeno. 1and (b) homeno. 2, andTrichuti trichiwa at (c) homeno. 1and(d) homeno. 2. The distributionsareoverdispersed at homeno. i andunderdispersed at home no. 2. The distribution of the eggs in soil (as measured by the variance/mean ratio) is given in Table 2 and graphically presented in thi Figure. These data indicate that the eggs were aggregated in the soil of home no. 1, but had an underdispersed distribution at home no. 2. During the course of the study the proportion of soil samples containing eggs remained relatively constant at home no. 1 and declined at home no. 2, while the density of the eggs of both species of helminths showed a significant decline at both homes (Table 2). Over the 2-month period of observation, during which further contamination by eggs had been limited by anthelmintic therapy, the egg density fell by approximately 50%. The spatial distribution of eggs, in contrast, remained relatively unchanged, being overdispersed at home no. 1, and underdispersed at home no. 2 at all times of observation. Discussion The modified Dada method had an adequate recovery efficiency in most common soil types and at a wide range of egg densities, and was sufficiently simple to be used under field conditions. The field study indicated that higher levels of soil contamination occurred in the home where the population also had a higher prevalence and intensity of infection. Whether these factors are causally related was not assessed in the present study, although it was apparent that the home which had the higher levels of

both contamination and infection a!so had a lower standard of sanitation. The distribution of the parasite eggs in soil was overdispersed in home no. 1, where there was a high density of eggs in soil. This aggregation may reflect the dynamics of soil contamination. Eggs are randomly distributed within the original faecal mass (MARTIK, 1975) which, after deposition on the soil surface, is broken down by environmental processes (BEAVER, 1975). The eggs may thus be initially aggregated at the deposition site and become dispersed radially over time. In contrast, the spatial distribution of eggs in the soil of home no. 2 was underdispersed. This apparent lack of aggregation may be a sampling artefact due to the low density of eggs at this site. For very sparse populations the probability of individuals occurring in any sampling unit may be so low that the distribution is effectively random (SOUTHWOOD, 19783. This statistical phenomenon may lead to the anomaly of populations which apparently exhibit underdispersed distributions at low density and contagious distributions at high density, as was observed in the present study. During the 2 months after the treatment intervention, when eggs were unlikely to be entering the environment, there was a decline in the density of both A. lumbricoides and T. trichiura eggs at both localities. The rate of decline under these tropical conditions was more rapid than that previouslv

570

observed in temperate areas (BURDENet al., 1976). This decline mav have been due to 3 causes. First. due to environmental factors, the eggs might have become concentrated into fewer quadrats. This does not appearto be the casein the present study since the spatial distribution remained essentially unchanged. Second, the eggs may have become damaged, destroyed or otherwise undetectable due to the combined effects of ageing and environmental factors (BROWN, 1927; FASSATIOVA& LYSEK, 1982). Previous studies (BURDENet al., 1976; KATAKURAet al., 1986) have shown that the eggsof A. lumbricoides and T. trickiura remain intact and recognizable over a period of at least several months; however, the possibility does remain that exposure to tropical heat and sunlight may reduce persistenceon the soil surface. Finally, the eggsmight have percolated into deeper strata due to the leaching effects of rain (STOREY& PHILLIPS, 1985), leaving fewer eggs in the surface soil. Vertical movement of eggswas not assessedin the present study but may be of relevance since the sandy soil of both local&es would facilitate leaching (FOTH. 1978: BROWN.1927). While the egg densi& heclined at both study- sites, the proportion of soil samplescontaining eggsshowed no significant change at home no. 1 but declined markedly at home no. 2. These contrasting observations may be a mathematical consequence of the relationship between the proportion of soil samples positive and the mean density of eggs. For overdispersed distributions, as at home no. 1, there may be a significant decline in density with only a modest decline in the proportion of positive soil samples, provided that the degree of aggregation is severe (ANDERSON,1986). For underdispersed distributions, however, there is an approximately linear relationship between the proportion of soil samples containing eggsand mean egg density, so that both variables tend to change proportionally, as observed for home no. 2. Whatever the underlying mechanisms, the density of geohelminth eggsin surface soil declined markedly over a period of 2 months. Since it is the soil surface which is the main target for childhood geophagic behaviour (WONG. 1988). this imnlies that. even in endemic areas, human exposure to infection may be significantly reduced once additional soil contamination is prevented by chemotherapy. Acknowledgements

This study was conducted with the agreement of the staff

of Reddiesand Glenhopeplaces-of-safety and the Ministries of Justice (Child Welfare division) andHealth, Jamaica.We

thank Claudette Robotham and Dr M. H. N. Golden (Tropical Medical Research Unit, University of the West Indies) for assistanceand advice.The study wassupported by the University of the West Indies. D. A. P. B. acknowledges the generous assistance of the Wellcome Trust.

References

Anderson, R. M. (1986). The population dynamics and epidemiology of intestinal nematode infections. Transactionsof the Royal Societyof Tropical Medicine and Hygiene, 80, 686-696. Anderson, R. M. & Medley, G. R. (1985). Community control of helminth infections of man by mass and selective chemotherapy. Parasitology, 90, 629-660. Beaver, P. C(1952). Observations on the epidemiology of ascariasis in a region of high hookworm endemicity. Journal

of Parasitology,

38, 445453.

Beaver, P. C. (1975). Biology of soil-transmitted helminths: the massive infection. Health Laboratory Science, 12, 116-12s. Brown, H. W. (1927). Studies on the rate of development and viability of the eggs of Ascaris lumbricoides and Trichuris trichiura under field conditions. Journal of Parasitology, 14, 1-15. Bundy, D. A. I’. (1988). Population ecology of intestinal hehninth infections in human communities. Philosophical Transactions of the Royal Society, B, 321, 405-420.

Bundy, D. A. I’. & Blumenthal, U. (1990). Human behaviour and the epidemiology of helminth infection. In: Parasitism and Host Behaviour, Barnard, C. J. & Behnke, J. M. (editors). London: Taylor and Francis, in Bunpdr~D. A. P., Thompson, D., Golden, M. H., Cooper, E., Anderson, R. M. & Harland, P. (1985). Population distribution of Trichuris trichiura in a community of Jamaican children. Transactions of the Royal Sociery of Tropical Medicine and Hygiene, 79, 232-237.

Burden, D. J., Whitehead, A., Green, E., McFadzean, J. & Beer, R. (1976). The treatment of soil infested with the human whipworm, Trichuris trichiura. Journal of Hygiene, 77. 377-382. Cheesbrough, M. (1981). Medical Laboratory Manual for Tropical Countries, vol. 1. Hertford: Stephen Austin & Sons. Dada, B. J. 0. (1979). A new technique for the recovery of T’rcy; eggs from sod. Journal of Helminthology, 53, Fassatiova, 6. & Lysek, H. (1982) Ovicidal fungi in the soil ecological system. Acta Univer&atis Carolinae, Bioloeica. II 9, 297-334. Foth, H. D. (1978). Fundamentals of Soil Science, 6th edition. Toronto: John Wiley & Sons. Giboda, M. (1981). Conditions of occurrence of geohelminths in the children population of east Slovakia. Helminthologia,

18, 99-l 12.

Ismid, I. S., Margono, S. S., Anggarini, S., Mahfundin, H., Rasad, R., Rasidi, R. & R&mono, B. (1983). The effect of mass treatment in different target groups on the dispersion of Ascaris eggs in the soil. In: Collected Papers on the Control of Soil-transmitted Helminthiasis, vol. 2, Tokyo: Asian Parasite Control Organization, p. 231. Katakura, K., Hamada, A. & Kobayashi, A. (1986). The fate of Ascaris eggs applied to the soil under various conditions. Japanese Journal of Parasitology, 35, 109. Martin, L. K. (1965). Randomnessof particle distribution in human fecesand the resulting influence on helminth exe counting. American Journal-of Tropical Medicine a2 Hygiene, 14, 747-759.

Southwood, T. R. E. (1978). Ecological Methods, 2nd edition. London: Chapman & Hall, pp. 35-36. Storey, G. W. & Phillips, R. A. (1985). The survival of parasite eggsthroughout the soil profile. Parasitology, 91, 585-590. Wakelin, D. (1985). Genetic control of immunity to hehninth infections. Parasitology Today, 1, 17-23. WHO (1964). Soil-transmitted Helminths. Geneva: World Health Organization, Technical Report Series, no. 277. Wonn. M. S. (1988). The role of environmental and host bihavioural factors in determining exposure to infection with Ascaris lumbricoides and Trichuris trichiura. PhD Thesis, Faculty of Natural Sciences, University of the West Indies. Wang, M. S., Bundv, D. A. I’. & Golden. M. H. N. (1988). Qiantitative assessment of geophagic behaviour as ‘a potential source of exposure to geohelminth infection. Transactions of the Royal Society of Tropical Medicine and Hygiene, 82, 621-625.

Received 16 May 1989; revised IS March for publication 1.5 March 1989

1989; accepted