[25] Maramatsu, Y.; M. Sumiya and Y. Ohmomo, Stable iodine contents in human milk related to dietary algae consumption. Hoken Busuri 18, 113-117 (1983).

[26] Stoeppler, M., and K. Brandt, Rapid determination of Cd in whole blood by electrochemical atomization. Ref. [20], 183–186. [27] Le Baron, J. G., PhD thesis, University of Waterloo, Canada (1977). [28] Peerboom, J. W. C.; P. de voogt, B. V. Hattum, W. V. D. Velde, and J. H. W. C. Peerboom-Stegeman. The use of human placenta as a biological indicator for Cd exposure. Ref. [20], 8-10. [29] Underwood, E. J., Trace Element Nutrition in Human and Animal Nutrition. Academic Press: New York (1977). [30] Iyengar, G. V., Elemental composition of Human and Animal Milk: A Review. IAEA-TCDOC-269, IAEA: Vienna (1982). [31] Heaton, F. W., and A. Hodgkinson, External factors affecting diurnal variation in electrolyte excretion with particular reference to Ca and Mg. Clin. Chim Acta. 8, 246–54 (1963). [32] Battacharya, R. D., Circadian rhythmic aspects of urinary Zn excretion in presumably healthy subjects. Pan Med. Torino 21, 201-203 (1979).

[33] Stengle, J. M., and A. L. Schade, Circadian variation of Fe in plasma. Br. J. Haematol. 3, 117-20 (1957).

[34] Werkmann, H. P.; J. M. F. Trijbels, and E. D. A. M. Schretten, Short term Fe rhythm in serum of children. Clin. Chim. Acta. 53, 65-68 (1974).

[35] Subramanian, K. S., and J. E. Meranger, Diurnal variations in the concentration of Cd in urine. Clin. Chem. 39, 1110 (1984). [36] Kasperek, K.; H. Schicha, V. Siller, L. E., Feinendegen, and A. Hoeck, Trace element concentrations in human serum: diagnostic implications. Proc. IAEA Symp. Nuclear Activation Techniques in the Life Science. IAEA: Vienna, 517-526 (1972).

[37] Heydorn K.; E. Damsgaard, N. A. Larson and B. Nielsen, Sources of variability of trace element concentrations in human serum. Ref. [36], 129–42.

[38] Airey, D., Total Hg concentrations in human hair from 13 countries in relation to fish consumption and location. Sci. Total Environ. 31, 157–180 (1983).

[39] Henry, R. W., and M. E. Elmes, Plasma Zn in acute starvation. Br. Med. J. 4, 625–626 (1975).

[40] Behne, D., and H. Juergensen, Determination of trace elements in human blood serum and in the standard reference material bovine liver by instrumental neuton activation analysis. J. Radioanal. Chem. 42, 447-453 (1978).

[41] Juswigg, T.; R. Batres, N. W. Solomons, O. Pineda, and D. B. Milne, The effect of temporary venous occlusion on trace mineral concentrations in plasma. Am. J. Clin. Nutr. 36, 354-58 (1982).

[42] Osborn, D., Seasonal changes in the fat, protein and metal content of the liver of the starling Sturnus Vulgaris. Environ. Pollut. 19, 145-55 (1979).

[43] Iyengar, G. V., Autopsy sampling and elemental analysis: errors arising from post mortem changes. J. Path. 134, 173-80 (1981).

[44] Iyengar, G. V., Post mortem changes of the elemental composition of autopsy specimens: variations of K, Na, Mg, Ca, C., Fe, Zn, Cu, Mn and Rb in rat liver. Sci. total Environ. 15, 217-36 (1980).

[45] Iyengar, G. V.; K. Kasperek and L. E. Feinendegen, Post mortem changes of the elemental concentration of autopsy specimens: variations of Co, Cs and Se in rat liver. J. Radioanal. Chem. 69, 463-72 (1982).

[46] Livingstone, H. D., Distribution of Zn, Cd and Hg in human kidneys. Proc. Trace Substances in Environmental Health-V (Hemphill, D. D. ed.) University of Columbia, Mo, pp. 399-411 (1971).

[47] Molokhia, M. M., and B. Portnaoy, Neutron activation analysis of trace elements in skin. IV Regional variations in Copper, Manganese and Zinc. Br. J. Derm. 82, 254–255 (1970).

1.2 U.S. Pilot National Environmental

Specimen Bank Program

The historical development in the United States during the 1970's of the concept of a National Environmental Specimen Bank (NESB) has been reviewed by Goldstein [6,7]. The purpose of such a national system is 1) to detect changes in the environment on a real-time basis (i.e., monitoring) using bioaccumulators as indicators, 2) to distinguish such changes from natural inputs, and 3) to provide a "bank" of well preserved and documented environmental samples for retrospective analyses in future years as analytical techniques improve or as new pollutants are identified.

Since 1975, the National Bureau of Standards (NBS), in conjunction with the U.S. Environmental Protection Agency (EPA), has been involved in research relating to the establishment of a National Environmental Specimen Bank. The initial plans and results of the EPA/ NBS effort for the NESB in research and methodology evaluation have been described [8–13].

In 1979, a special "clean" laboratory/storage facility [14] was completed at NBS to initiate a Pilot National Environmental Specimen Bank Program. This pilot program is designed to evaluate the feasibility of a national program by providing actual working experience in all aspects of specimen banking, i.e., specimen collection, processing, storage, analysis, and data management. The major goals of this pilot study are: 1) to develop analytical protocols for sampling, processing, and storing four types of environmental accumulators and biomonitors; 2) to evaluate and improve analytical methods for the determination of both trace element and organic pollutants in biological matrices; 3) to evaluate the feasibility of long-term storage under various conditions; and 4) to provide a "bank" of samples for retrospective analyses in future years. The experience gained during the pilot study will be the basis for evaluating the feasibility of establishing a National (or International) Environmental Specimen Bank.

Four types of accumulators and biomonitors were selected for inclusion in the NBS pilot program as a result of an EPA/NBS Workshop on "Recommendations and Conclusions on the National Environmental Specimen Bank" held in 1976 [8]. These accumulators are: 1) human soft tissue-liver, 2) aquatic accumulator-marine bivalues and sediments, 3) food monitortotal diet composite, 4) air pollutant accumulator. Human liver was selected as the first sample type for the inclusion in the pilot specimen banking program. Sample types 2 and 3 are currently incorporated into the program with continued collection, storage, and analysis of the previous sample type. Consequently, the

the development of the human liver as a monitor tissue. This contribution summarizes this experience with specific emphasis on sampling and analysis by nuclear methods.

2. Development of the Human Monitor Tissue 2.1 Criteria Determining the Selection of

Monitor Tissues

For the selection and development of a monitor tissue the following aspects have to be considered: 1) tissues are selected because they are known either to respond to pollutant exposure or to accumulate pollutants; 2) a choice has to be made between biopsy and autopsy samples; 3) the quantity of material must be sufficient for baseline determination of constituents and storage for future analysis; 4) it must be possible to collect biologically or anatomically well described samples; and 5) analytical methods must be available to determine the pollutants. Table 1 lists a selection of possible monitor tissues and their evaluation according to the above considerations.

The general advantage of monitor tissue taken by biopsy versus autopsy is the possibility to monitor pollutant trends in the very same individual. The individual would reflect his own habitat over an extended time period. However, a wide variety of tissues is not available for sequential collection throughout the lifetime of an individual. Most of the biopsy tissue samples are collected in connection with surgical incisions, which usually provide only small samples, or one time events (e.g., placenta) where ethical barriers restrict the degree of sampling. Some monitor tissues can be collected on a routine basis without the above restrictions. Nonintrusive sampling includes hair, nail clippings, and various excretions. Requiring only minimal intrusion, blood has been demonstrated to be useful as an effective monitor that responds directly to the environment. In the U.S., the large-scale screening of blood from children helped to verify the accumulation of lead in blood and consequently resulted in the initiatives to lower the risk of exposure, e.g., by banning the use of lead based paints and reducing the lead content of gasoline [15,16]. The disadvantage of blood as a monitor is the fact that it reflects more acute exposure rather than being an accumulator. The natural dilution effect resulting from the blood's function as a distributor means that the concentrations of most pollutants are very low and difficult to measure. The danger of contamination during collection is very high (because of the low concentrations), although this problem can be minimized [17]. A more serious problem is the preservation of the blood sample and its long term storage without change, e.g., coagulation.

2.2 Human Liver as Monitor

2.2.1 Biological and Anatomical Aspects. The liver exhibits several features which are advantageous to its use as a human monitor tissue. The liver is among the few organs which are significantly exposed to most of the pollutants that enter the body. Practically all substances which are assimilated by the human body are transported through the blood stream and consequently pass through the liver. The liver's function is to detoxify, store, and regulate trace substances in the body. Comparative studies on biological half-times [18] of trace substances in tissues support the storage function of the liver. Though some toxic elements have shown more affinity to calcareous tissue, biological half-times of trace substances in liver are so long that liver can be regarded as a general accumulator. In addition, the liver accumulates lipophilic organic trace substances because of its fat content. Therefore the liver is possibly the most universal accumulator tissue.

For the purpose of sample collection and specimen banking, the liver offers a biologically and anatomically well defined tissue, which also can be obtained in large quantities. The collection of the left lobes of the liver alone provides 200-400 g per sample depending on the physical stature of the subject. The liver is homogeneous in its function and in its macroscopic structure. During an autopsy, only a few cuts of ligaments and blood vessels are necessary to remove the liver from the corpse. The integrity of the sample can thus be pre

served and subsequent sampling can be carried out under controlled conditions.

2.2.2 Sampling Protocol Development. Because of the extremely low concentrations of trace element and organic pollutants found in most environmental samples, extreme caution must be exercised during sample collection and sample processing to avoid contamination. A detailed sampling protocol, designed to provide samples suitable for both trace element and organic analysis, has been developed and implemented for the collection of human liver samples. For the complete detailed protocol see Harrison et al., [5,19]. This liver sampling protocol was developed in conjunction with those individuals performing the autopsies, and its implementation within the bounds of practicality required periods of education and close cooperation. The need for careful communication was evident from perceived meanings of such terms as "clean," i.e., interpreted as "sterile" by autopsy personnel vs. "non-contaminated chemically" by analytical chemists.

The sampling protocol was designed to avoid possible contamination of the sample by either inorganic or organic constituents. Teflon supplies (e.g., sheets, bags, and storage jars) were selected as the most suitable materials to prevent contamination of the sample from inorganic and organic constituents and from the diffusion of water [7,15]. The protocol specifies the use of such non-contaminating items as non-talced vinyl gloves, pre-cleaned dust-free Teflon FEP sheets and bags, highpurity water, and a titanium/Teflon TFE knife. These

items are provided by NBS at each collection site to insure uniformity in sampling materials. A special knife with a titanium blade and a Teflon TFE handle was designed and constructed at NBS for use during dissecting of the specimen. This special knife is used to avoid trace element contamination by various constituents associated with a regular stainless-steel scalpel/ knife (e.g., Ni and Cr) and to limit the possible contamination by an element of little environmental interest, namely titanium. The liver samples are sealed in Teflon FEP bags, frozen in liquid nitrogen (LN2), and shipped to NBS in a special biological shipper at LN2 vapor temperature.

To eliminate potentially infectious liver samples from the specimen bank, a blood sample is removed from the donor at the time of the autopsy to be used for hepatitis B screening. In addition, liver specimens from the right lobe are removed for preparation of histological slides. These slides are examined by a pathologist to verify the absence of infectious diseases and then stored at NBS. They are also intended to serve for possible future refer

ence.

A data form, sent to NBS with each liver sample, contains information about the donor, e.g., date of birth, sex, residence, ethnic group, height, weight, smoking history, occupation (if known), date and time of death and autopsy, diagnosis of autopsy, and liver specimen weight (see ref. [18,19]).

During the first 18 months, 300 liver samples from three geographical locations (Baltimore, MD; Minneapolis, MN; and Seattle, WA) were collected using this protocol. During the remaining years of the pilot study, liver samples were received predominately from one location (Seattle) at a rate of approximately 50 per year, bringing the number of banked specimens to about 500. After the first year of human liver collection, the sample collection protocol was evaluated with respect to such items as: initial set-up costs for each site, sample procurement costs, transport time, time required for receipt of hepatitis results and histological slides, and suitability of the donor selection criteria [20].

A conclusion from the experience of the first year of collection was that a technician was needed at the collection site whose primary responsibility was to work closely with autopsy personnel to select, remove, and prepare the liver samples. Since these technicians were responsible for implementing the sampling protocol, NBS personnel worked closely with them to stress the importance of following the protocol precisely. NBS must have confidence that the samples stored in the specimen bank are collected exactly as prescribed in the protocol.

2.2.3 Analytical Aspects. The U.S. Pilot NESB Program, and possibly any biomonitoring program, has

to operate under constraints regarding the amount of available material from which the desired set of data will be obtained. In addition, modern analytical techniques require only small subsamples for analysis. Since significant trace element inhomogeneities have been found within human livers [21,22], homogenization of the bulk sample is required before analysis. The necessary reduction of a bulk sample to a laboratory sample suitable for analytical techniques can introduce errors due to contamination and heterogeneity. These errors may become the limiting factors for achieving the goal of precise and accurate analysis [23]. As the analytical test portion becomes smaller, better homogeneity is required so that the sample is representative of the bulk. Consequently, a major research effort of the pilot program has focused on homogenization procedures to provide uncontaminated analytical increments which represent the sample and which are homogeneous for subsampling of the analytical test portions.

Though the technique of cryogenic homogenization (brittle fracture technique) was first introduced and evaluated by Iyengar [24], considerable upgrading and evaluation was required before it could be routinely applied in the pilot program. The initial sample capacity of about 20 g was too small for application to the NESB program. Sample sizes for the human liver specimens in the pilot program are 120-180 g; and sample sizes up to 1000 g are anticipated for future specimens. As part of the pilot program, NBS has developed and evaluated the required technology for homogenizing larger samples [25].

NBS has designed two larger Teflon ball mills with capacities of 60 g and 150 g. A performance evaluation of the ball mills, based on particle size distribution and mixing ability, led to the design of the disk mill which has been determined to give superior performance and which is adequate for our needs (sec. 3.1). These experimental designs and their features are listed in table 2. Extensive evaluation for liver tissue showed that the cryogenic homogenization procedure with Teflon disk mills can be readily applied to soft tissues. Applications to other tissues resulted in equally fine particulate homogenates, using sieves for quick assay of the results of the homogenization process (fig. 1). These results were confirmed with the determination of sampling constants (sec. 3.1).

An important criterion for a comprehensive biomonitoring program is the availability of analytical techniques which can detect and determine all substances of concern at reasonable cost. The human liver, owing to its function as an accumulator, has higher concentrations of many elements and compounds of interest than most other tissues and therefore has less need for ultratrace level techniques. For most of the elements of