Preliminary adult soil ingestion estimates: Results of a pilot study

REGULATORY TOXICOLOGY AND PHARMACOLOGY 12, 88-95

(1990)

Preliminary Adult Soil Ingestion Estimates: Results of a Pilot Study EDWARD J. CALABRESE,* E D W A R D J. STANEK,'~ CHARLES E. GILBERT,* AND R A M O N M . BARNES~

*Environmental Health Sciences Program and t Biostatistics and Epidemiology Program, School of Public Health, University of Massachusetts, Amherst, Massachusetts 01003-0081; and ~Department of Chemistry, University of Massachusetts, Amherst, Massachusetts 01003-0035

Received December 23, 1989

Six adults were evaluated for the extent to which they ingest soil. This study was originally part of a large childhood soil ingestion study and was used to validate part of the analytical methodology used in that larger study. The adult study followed the soil tracer methodology of the children's study. The principal findings of the adult study revealed that the four most reliable tracer elements based on recovery studies in a mass balance assessment [aluminum (A1), silicon (Si), yttrium (Y), and zirconium (Zr)] yielded the following mean daily adult soil ingestion estimations: A1 (77 rag), Si (5 mg), Y (53 rag), and Zr (22 rag). The median daily soil ingestion estimations based on these same tracer elements were: A1 (57 rag), Si (1 mg), Y (65 rag), and Zr ( - 4 mg). These findings represent the first published quantitative estimates of soil ingestion by adults. 9 1990 Academic Press, lnc.

1. INTRODUCTION Regulatory and public health agencies have become concerned that consumption of contaminated soil by children may be a significant public health problem. For example, soil levels of lead that range from 1500 to 7500 #g/g in certain sections of Boston (Spittler, 1986) are suspected of contributing to elevated blood lead levels as a result of soil ingestion. Also widely discussed has been the dioxin contamination of Times Beach, Missouri. The Centers for Disease Control (CDC) derived a theoretical cancer risk associated with levels of dioxin in soil (Kimbrough et al., 1984). A major concern was the assumed consumption by children of soil containing dioxin, a contaminant that is relatively tightly bound to soil. While the issue of soil contamination has correctly focused on young children, it is important to note that the U.S. EPA has developed guidance recommendations for how much soil adults ingest. In a 1989 communication (Porter, 1989) EPA concluded that 100 mg of soil per day is assumed to be ingested by adults. To our knowledge no paper has yet been published concerning how much soil adults ingest, although several extensive studies quantifying how much soil children eat (Calabrese et al., 1989; Binder et al., 1986; Clausing et al., 1987) exist. In the course of conducting a large88 0273-2300/90 $3.00 Copyright 9 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.

ADULT SOIL INGESTION ESTIMATES

89

scale study of soil ingestion by children, (Calabrese et al., 1989), part of the analytical methodology was validated via exposure of a limited number of adults to known quantities of a well-characterized soil. While this investigation was not designed to estimate the amount of soil normally ingested by adults it did offer potential and limited insights as to how much soil these adults ingested. In light of the new initiative by EPA (Porter, 1989) to consider adult soil ingestion in risk assessment and the paucity of data on this topic it is felt that these findings deserve broader dissemination. 2. M E T H O D O L O G Y

2.1. Participants and Soil Capsules Participants in the study were six healthy adults, three males, three females, 25-41 years old, who did not have chronic illness such as diabetes, heart disease, or gastrointestinal disorders. Each of the volunteers ingested: (1) one empty gelatin capsule at breakfast and one with dinner Monday, Tuesday, and Wednesday during the first week of the study; (2) 50 mg of sterilized soil within a gelatin capsule at breakfast and dinner for a total of 100 mg of sterilized soil per day for the 3 days during Week 2; (3) 250 mg of sterilized soil in a gelatin capsule at breakfast and dinner for a total of 500 mg of soil per day over the 3 days during Week 3.

2.2. Soil Preparation and Security Soil for the adult ingestion study was selected from a soil library maintained by the Department of Plant and Soil Sciences on the University of Massachusetts, Amherst campus. The soil was selected because it was previously characterized as noncontaminated (Calabrese et al., 1989) and the tracer elements, aluminum, barium, manganese, silicon, titanium, vanadium, yttrium, and zirconium were of sufficient concentration to be detected during analysis of the volunteer's excretory samples. A formal security system was employed following the selection of this soil to prevent tampering with the soil and soil-filled capsules. The soil was sterilized by autoclaving, oven dried, and evaluated microbiologically to detect the presence of anaerobic and aerobic organisms and fungi. Human subjects review required that the soil be chemically analyzed for the metal lead and the U.S. EPA's extractable priority pollutants consisting of approximately 100 compounds including chlorinated hydrocarbon insecticides, PCBs, numerous polynuclear aromatic hydrocarbons, and phenolic compounds. The concentrations of these compounds in the soil were less than the detection limits: all compounds except polychlorinated biphenyls, 1 ~g per gram; polychlorinated biphenyls, 5 #g per gram.

2.3. Duplicate Meal Duplicate meal samples, food and beverage, collected from the six adults, included all food ingested from breakfast Monday, through the evening meal Wednesday during each of the 3 weeks to reflect total dietary intake. All medications and vitamins

90

CALABRESE ET AL. TABLE 1 VARIANCE COMPONENTSFOR FREEZE-DRIED WEIGHT OF FOOD INGESTED BY SIX SUBJECTS OVER 3 DAYS FOR 3 WEEKS

Source

Variancecomponent gsq/day

Percentof variance

Subject Week Day

63071 17071 17826

64 17 18

ingested by the adults were included in the duplicate meals or were addressed separately during chemical analysis and the values added to the food total for that day. The adults prepared two meals each containing the same types and quantities of food. One meal was eaten by the volunteer; the second meal was used as the duplicate meal. All uneaten food on the volunteer's plate was consolidated and compared to the duplicate meal. The food portions in the duplicate meal were separated to represent ingested food and uneaten food. The estimated amount of eaten food was placed into a polyethylene storage container for study analysis. The polyethylene containers were enclosed in plastic bags and transported in an insulated cooler bag with a reusable refrigerant.

2.4. Excretory Output Total excretory output, feces and urine, were collected from Monday noon through Friday midnight (4 days) over 3 consecutive weeks. Urine and stool samples were collected in polyethylene commodes. Toilet paper or other wipes were not included in the commode and not collected by the research team. Used commodes were enclosed in plastic bags and placed into insulated bags containing portable refrigerants.

3. RESULTS Complete data were recorded on all six subjects for the entire 3-week period of the study. We describe results for food ingestion, soil capsule ingestion, and fecal output for the adults, and then use these results to estimate the amount of soil ingested by the adults (methodology previously described, Calabrese et aL, 1989).

3.1. Food Ingestion Daily freeze-dried weights of food samples varied from 97 to 913 g. There was large variability in food intake between study subjects, with less variability between days of the week or weeks (Table l).

91

ADULT SOIL INGESTION ESTIMATES TABLE 2 MEAN DAILY TRACERSIN FOOD INGESTEDPER DAY FOR 3 DAYSFOR SIX ADULTS FOR 3 CONSECUTIVEWEEKSAND OVERALL

Week

Freezedried wt g

A1 (mg)

Ba (#g)

Mn (rag)

Si (mg)

Ti (mg) ,

1 2 3

553 461 504

2.06 3.66 1.32

759 547 631

3.11 1.77 2.59

34.1 34.3 28.8

Mean

506

2.35

646

2.49

32.4

V (tzg)

Y (#g)

Zr (#g)

0.94 2.46 4.05

9.97 65.09 8.66

1.59 1.52 0.77

2.51 6.97 3.01

2.48

27.91

1.29

4.16

The mean amount of food ingested per subject per day was calculated for each week. These amounts were then averaged per week and overall. The average amounts ingested per week and overall are given in Table 2. Table 3 summarizes components of variability in food ingestion for each tracer element. Variance components are given between subjects, between weeks, and between days of the week. The variance attributed to each component is also indicated. The components of variance indicate that most variability in amount of elements ingested occurred from day to day or week to week. The magnitude of variability from subject to subject in amount of tracer element ingested was low relative to these other sources. 3.2. Soil Concentrations Soil was homogenized and the concentrations of each of the eight study elements were analyzed prior to filling the soil capsules. A total of 300 mg of soil was ingested in the second week of the study, and 1.5 g of soil was ingested in the third week of

TABLE 3 COMPONENTSOF VARIANCE(AND PERCENT) FOR AMOUNTOF TRACER ELEMENTSINGESTED IN FOOD FOR SIX ADULTS

Element

Unit (squared)

Subject a

Weeka

Day a

A1 Ba Mn Si Ti V Y Zr

mg #g mg mg mg ug ug t~g

0.1 (0.4) 13427 (9.2) 0.04 (2.1) 0 (0.0) 1.3 (4.7) 2078 (16.3) 0.04 (2.5) 18.9 (27.3)

0.8 (3.6) 6576 (9.2) 0.8 (40.4) 0 (0.0) 15.5 (55.7) 2548 (20.0) 0.2 (12.9) 2.0 (2.8)

21.8 (96.0) 51358 (72.0) 1.1 (57.5) 993.0 (100.0) 11.0 (39.7) 8116 (63.7) 1.3 (84.6) 48.3 (69.8)

a Values in parentheses are percentages.

92

CALABRESE ET AL. TABLE 4 TOTAL SOILTRACER AMOUNTIN CAPSULEINGESTEDBY ADULTSIN EACHWEEK

Week

A1 (mg)

Ba (vg)

Mn (mg)

Si (mg)

Ti (nag)

V (t~g)

Y (pg)

Zr 0zg)

1 2 3

0.00 24.36 121.80

0 177 885

0.00 0.24 1.23

0.0 83.4 417.0

0.00 1.56 7.80

0 42 210

0.0 7.2 36.0

0 54 270

the study. The total amount of each element ingested by week from soil capsules is given in Table 4. Since fecal samples were collected for 4 days in each week, we also express weight in milligrams or micrograms of each element ingested per day over a 4-day period for comparison with the fecal sample results (Table 5). The average daily capsule ingestions reported in Table 5 for each element can be compared with the ingestion from food (Table 2) to quantify the experimental dose relative to background intake. For A1, Si, Y, and Zr, maximal daily capsule dose was more than four times average daily ingestion from food.

3.3 Excretory Concentrations Fecal and urine samples were collected from the Monday first morning void through Friday midnight of each week. The number of samples per day ranged from one to four for subjects. The most common number of excretory samples per day was one or two, and the average number of excretory samples per day for each week was two.

Samples were pooled on a given day, freeze dried, and weighed. Concentrations of each element were then measured for each pooled sample. Daily excretory freezedried weight varied from 28.1 to 141.4 g. There was no apparent association between excretory weight and number of excretory samples. There was marked variability in freeze-dried weight between subjects; 73% of the variability in freeze-dried weight was attributable to differences between subjects, with the remaining variability attributable to week-to-week differences. The mean freeze-dried excretory weight per day for the six subjects was 58.4 g for the first week, 66.0 g for the second week, and 71.2 g for the third week.

TABLE 5 DAILY (OVER4 DAYS) SOILTRACERAMOUNTIN CAPSULEINGESTEDBY ADULTSBY WEEK

Week

A1 (mg)

Ba (pg)

Mn (mg)

Si (mg)

Ti (mg)

V (#g)

Y (#g)

Zr (pg)

1 2 3

0.00 6.09 30.45

0.00 44.25 221.25

0.00 0.06 0.31

0.00 20.85 104.25

0.00 0.39 1.95

0.0 10.5 52.5

0.0 1.8 9.0

0.0 13.5 67.5

93

ADULT SOIL INGESTION ESTIMATES TABLE 6 AVERAGEDAILY AMOUNTOF TRACERSEXCRETEDIN 4 DAYSFOR SIX SUBJECTSBY WEEK

Week

A1 (nag)

Ba (#g)

Mn (mg)

Si (mg)

Ti (nag)

V (/~g)

Y (#g)

Zr (ug)

1 2 3

7.98 15.18 33.30

660.35 1341.40 919.67

3.24 2.79 3.48

43.5 59.6 125.8

4.21 2.94 9.07

100.4 78.4 95.6

3.08 3.82 11.34

27.8 31.2 56.6

Concentrations for each element in the excretory samples were evaluated and used to estimate the amount of each element excreted per day in each of the study weeks. The average amount excreted by week is given in Table 6.

4. Ingestion Estimates Table 7 provides the mean and median values of soil ingestion for each element by week, while Table 8 provides daily soil ingestion estimates by subject averaged over the three weeks of the study. These resultant values in Tables 7 and 8 have had tracer quantities in ingested food and capsules subtracted from amounts excreted. Table 8 also provides the estimated median and mean soil ingestion values for the six subjects averaged over the 3 weeks of the study. The median values for the eight tracers ranged from 1 mg (Si) to 754 mg (Mn). The mean values ranged from 5 mg (Si) to 828 mg (Mn). On the basis of sample percentage recovery values, AI, Si, Y, and Zr were considered the most valid tracers (Calabrese et al., 1989). The median values of these four tracers are: AI, 57 rag; Si, 1 mg; Y, 65 mg; and Zr, - 4 mg. The mean values are A1, 77 rag; Si, 5 rag; Y, 53 mg; and Zr, 22 mg. The average of the four mean tracer values is 39.25 mg.

TABLE 7 ADULT DAILY SOIL INGESTION ESTIMATESa BY WEEK AND TRACER ELEMENT AFFER SUBTRACTING FOOD AND CAPSULE INGESTION, BASEDON MEDIAN AMHERST SOIL CONCENTRATIONS: MEANS AND MEDIANSOVER SUBJECTS Week

A1

Ba

Mn

Si

Means 1 2 3

0.110 0.098 0.028

-0.232 2.265 0.201

0.330 1.306 0.790

0.030 0.014 -0.023

Medians 1 2 3

0.060 0.085 0.066

-0.071 0.597 0.386

0.388 1.368 0.831

0.031 0.015 -0.027

a In grams.

Ti

V

Y

Zr

0.071 0.025 0.896

1.288 0.043 0.532

0.063 0.021 0.067

0.134 0.058 -0.074

0.102 0.112 0.156

t. 192 0.150 0.047

0.044 0.035 0.060

0.124 0.065 -0.144

94

CALABRESE ET AL. TABLE8

ADULT DAILY SOIL INGESTION ESTIMATESa BY SUBJECTAND TRACERELEMENT,AFTERSUBTRACTING FOOD AND CAPSULEINGESTION, BASEDON MEDIAN AMHERSTSOIL CONCENTRATIONSAVERAGEDOVER WEEKS

ID

A1

Ba

Mn

Si

Ti

V

Y

Zr

1 2 3 4 5 6

0.068 0.046 0.037 0.001 0.173 0.134

0.218 0.097 3.968 -0.229 0.617 -O.053

0.681 0.626 0.826 0.542 1.411 0.883

-0.006 -0.070 -0.016 0.007 0.014 0.099

0.908 -0.126 0.317 -0.054 0.104 1.113

-1.824 1.474 0.027 0.712 0.118 2.183

0.111 0.075 -0.035 0.054 0.027 0.084

-0.172 -0.025 0.148 -0.052 0.017 0.216

Mean Median SD

0.077 0.057 0.065

0.770 0.158 1.593

0.828 0.754 0.312

0.005 0.001 0.055

0.377 0.211 0.518

0.448 0.415 1.384

0.053 0.065 0.051

0.022 -0.004 0.141

a In grams.

5. DISCUSSION The results of this pilot study on adult soil ingestion indicate that soil ingestion values based on both mean and median estimates for the four most reliable tracers (A1, Si, Y, and Zr) are less than the EPA guidance figure of 100 mg/day and more closely approximate a value near 50 mg/day. The preliminary nature of these findings should be emphasized given the small sample size (n = 6) and the fact that the adult data were originally designed to validate the children's ingestion study (Calabrese et al., 1989) and not to determine adult soil ingestion. Nonetheless, these findings provide initial estimates of adult soil ingestion and should provide assistance in the development of methodological approaches including study design, tracer selection, massbalance consideration, analytical processes, and statistical evaluation. Despite the limited nature of the adult soil ingestion data present here, comparison with the children ingestion data (Table 9) is warranted in light of EPA regulatory guidance procedures which address both children and adults (Porter, 1989). While

TABLE 9 COMPARISONOF SOIL INGESTION VALUESIN CHILDRENAND ADULTS

Adult

Children/adult ratio

Children a

Tracer elements

Mean (mg)

Median (mg)

Mean (mg)

Median (rag)

Mean

Median

A1 Si Y Zr

77 5 53 22

57 1 65 -4

153 154 85 21

29 40 9 16

2.0 30.8 1.6 1.0

0.5 40.0 0.1 >16.0

Data from Calabrese et al. (1989).

ADULT SOIL INGESTION ESTIMATES

95

the factors which contribute to enhancing the occurrence of h u m a n soil ingestion remain to be more comprehensively and quantitatively assessed, it is generally believed that the greater hand-to-mouth behavior observed in young children c o m p a r e d with that of older children and adults would contribute to their displaying greater soil ingestion than older age groups. The EPA guidance values appear to reflect this conceptual difference with the agency assuming that adults would ingest about onehalf as m u c h soil as young children. In the absence o f data it is certainly reasonable to assume that adults should ingest less soil than y o u n g children. However, the issue is how much less. The present data reveal the mean children values for soil ingestion were generally higher than the adult values with the exception o f Zr where adult and children values were similar. In contrast, the comparison o f children versus adult soil ingestion values using the median revealed no consistent relationship a m o n g the four tracers (Table 9). Conclusions based on these observations must be tempered by the fact that the adult study was of a m u c h more limited nature (six subjects) than the children's study, which involved 64 children and was not designed to determine adult soil ingestion.

ACKNOWLEDGMENT This research was supported in part by Syntex Agribusiness, Inc., and the Gradient Corp. REFERENCES BINDER, S., SOKAL,D., AND MAUGHAN,D. (1986). Estimating soil ingestion: The use of tracer elements

in estimating the amount of soil ingested by youngchildren. Arch. Environ. Health 41, 341-345. CALABRESE,E. J., PASTIDES,H., BARNES,R., EDWARDS,C., KOSTECKI,P., STANEK,E., VENEMAN,P., AND GILBERT,C. E. (1989). How much soil do young children ingest: An epidemiologic study. Reg. Toxicol. Pharmacol. 10, 1-15. CLAUSING,P., BRUNEKREEF,B., ANDVANWIJNEN,J. H. (1987). A method for estimating soil ingestion by children. Int. Arch. Occup. Environ. Health 59, 73-82. KIMBROUGH,R. D., FALK,n., STEHR,P., ANDFIRES,G. (1984). Health implications of 2,3,7,8-tetrachlorodibenzodioxin (TCDD) contamination of residential soil. J. Toxicol. Environ. Health 14, 47-93. PORTER,J. W. (US EPA Officeof Solid Waste and EmergencyResponse). (January 27, 1989). Memorandum to RegionalAdministrators, Region I-X, regarding interim final guidance on soil ingestion rates. SPITTLER, T. (September 24, 1986). Technical Resource Committee Meeting. Minnesota Pollution Control Agency,Roseville, MN.