This Guidance document represented current agency thinking in regards to the available science at the time it was issued. It no longer represents the current state of science and is presented here for the historical record only.
United
States Food and Drug Administration
200 C St., S.W.
Washington,
D.C. 20204
January 1993
i. AUTHORS AND CONTRIBUTERS
ii. EXECUTIVE SUMMARY
iii. FORWARD
I. STATUTORY AUTHORITY
II. PUBLIC HEALTH STATEMENT
III. SAMPLING OF AND TRACE ELEMENT ANALYSIS IN SHELLFISH
1. Sampling
1.1 Field Sampling Procedure
1.2 Laboratory Sample Preparation
2. Analytical Methodology
2.1 Mineralization
2.2 Determinative Techniques
3. Quality Assurance/Quality control (QA/QC)
IV. CONSUMPTION AND EXPOSURE ASSESSMENT
1. Shellfish Intake
2. Arsenic Concentrations in Shellfish
3. Arsenic Exposure form Shellfish
4. Background and Relative Source Contribution
to Arsenic Exposure
4.1 Diet
4.2 Water
4.3 Air
4.4 Soil
5. Select Subpopulations
V. HAZARD ASSESSMENT
VI. LEVELS OF CONCERN
VII. REFERENCES
Michael A. Adams, Ph.D
Chemistry Review Branch, HFS-247
Division of Product Manufacture and Use
Office of Pre-Market Approval
Michael Bolger, Ph.D., D.A.B.T.
Contaminants Standards Monitoring and Programs Branch, HFS-308
Division of Programs and Enforcement Policy
Office of Plant and Dairy Foods and Beverages
Clark D. Carrington, Ph.D., D.A.B.T.
Contaminants Standards Monitoring and Programs Branch, HFS-308
Division of Programs and Enforcement Policy
Office of Plant and Dairy Foods and Beverages
Curtis E. Coker
Special Assistant to the Director, HFS-301
Office of Plant and Dairy Foods and Beverages
Gregory M. Cramer, Ph.D.
Policy and Guidance Branch, HFS-416
Division of Programs and Enforcement Policy
Office of Seafood
Michael J. DiNovi, Ph.D.
Chemistry Review Branch, HFS-247
Division of Product Manufacture and Use
Office of Pre-Market Approval
Scott Dolan
Elemental Research Branch, HFS-338
Division of Pesticides and Industrial Chemicals
Office of Plant and Dairy Foods and Beverages
This document has been developed to satisfy requests from local and state officials for federal guidance regarding the public health significance of arsenic in shellfish. (The term "shellfish" is used in a general sense throughout this document and is meant to include both molluscan bivalves and crustacea.) This document is designed to assist local and state health officials in their deliberations concerning the possible need to either issue consumption advisories or to close waters for fishing because of excessive arsenic contamination. The contents of this document include sections on FDA's statutory authority, a public health statement, sampling techniques and trace element analysis, consumption and exposure information, hazard assessment, and a discussion of estimating levels of concern for local consumption advisories or water closures.
Arsenic and its compounds are widely distributed in nature
primarily in two oxidation states, arsenite (trivalent) and
arsenate (pentavalent). Inorganic and organic arsenic compounds
used as pesticides, plant defoliants, and herbicides may
accumulate in agricultural and horticultural soils and plants.
Traces of arsenic are found in most foods, with the highest
concentrations found in seafood, particularly in shellfish, at
total arsenic levels up to 30 
g/g wet weight. Nearly all of
the arsenic present in seafood is organic arsenic which is
considered to be much less toxic than inorganic arsenic. It is
estimated that in the U.S. the mean total arsenic intake
from all food (excluding shellfish) is approximately
30 g/person/day.
Inorganic arsenic can have acute, subacute, and chronic effects.
Based on the provisional maximum tolerable weekly adult intake
for inorganic arsenic recommended by the WHO/FAO (15
g/kg/week) for short term or chronic
exposures, this document
recommends a tolerable daily intake of inorganic arsenic of 130
g.
Surveys of contaminants in shellfish conducted by FDA and the
National Marine Fisheries Service have found mean total arsenic
levels 1.1 ppm to 30 ppm (wet weight basis). FDA has combined
these survey results with nationally representative shellfish
consumption information to estimate the range of total arsenic
exposures that are possible among shellfish consumers.
For individuals who chronically consume an average of 15 g/day of
molluscan bivalves (90th percentile average intake over 14-days
for consumers among survey population) that have mean
total arsenic levels of 3.8 ppm, total arsenic
intake will average 57 g/person/day.
For people consuming
an average of 17 g/day of crustacean shellfish (90th percentile
average intake over 14-days for consumers among survey
population) that contain mean total arsenic levels of
10.6 ppm, total arsenic intake will average
180 g/person/day.
Although the tolerable daily intake for arsenic is based on exposure to inorganic arsenic, most arsenic present in shellfish is in an organic form (which is relatively non-toxic) and most monitoring methods determine total arsenic. To use traditional monitoring results to evaluate acceptable levels of shellfish consumption or acceptable levels of arsenic contamination, an estimation procedure that assumes that inorganic arsenic accounts for only 10% of the arsenic in shellfish is proposed for converting measurements of total arsenic to estimates of inorganic arsenic. It is proposed that this estimation procedure be used in lieu of monitoring results that are specific for inorganic arsenic.
Local patterns of shellfish consumption and/or the levels of
arsenic contamination may vary from national averages. Hence,
arsenic exposures from shellfish in particular regions may also
vary from values estimated using national figures. In order to
decide whether local arsenic exposure levels are of concern, it
is suggested that the maximum tolerable daily intake for
inorganic arsenic (130 g/person/day) be
used to calculate
Levels of Concern, either maximum permitted amounts of chronic
shellfish consumption or maximum permitted levels of arsenic
contamination.
Although local figures for shellfish consumption or shellfish contamination would be most appropriate for evaluating local situations, reference to national figures may also prove useful. As an example, if it is assumed that total arsenic exposure is derived solely from shellfish, it is calculated that the arsenic level of concern for individuals consuming molluscan bivalves on a chronic basis at the 90th percentile average among eaters (15 g/person/day) would be 86 ppm. The corresponding consumption level of concern for individuals consuming molluscan bivalves with total arsenic levels equivalent to the highest average found in one of the national surveys (3.8 ppm) is 340 g/person/day. If other sources of inorganic arsenic exposure are to be considered (e.g., relative source contribution including other dietary sources of arsenic), then corresponding adjustments in the levels of concern will need to be made.
Purpose
Over the years, and particularly of late, concern has been raised regarding the public health significance of the presence of contaminants in shellfish in both fresh water and marine species. Local and state officials have repeatedly sought guidance at the federal level on these matters; in particular, the advice and position of the U.S. Food and Drug Administration (FDA) has often been requested. FDA is developing a series of guidance documents in an attempt to satisfy the requests for federal guidance. They are designed to assist local and state health officials in their deliberations concerning contaminants in shellfish and the possible need to either issue fish consumption advisories or to close waters for fishing. The term "shellfish" is used in a general sense throughout these documents and is meant to include both molluscan bivalves and crustacea.
Standards have been developed for those contaminants which have been identified as national problems (e.g. methylmercury). Guidance documents have been and will continue to be developed for contaminants which could cause health effects, based on the toxicity of the contaminant and potential exposure. These health effects generally have a low probability of occurring. In some localities, however, adverse health effects may occur under extreme conditions of exposure. Should such conditions exist or occur the information in this guidance document indicates how tolerable levels of shellfish consumption or contamination might be determined; however, it does not dictate an approach or decision regarding a particular contaminant in shellfish. These guidance documents have no statutory authority. They merely present the relevant scientific information on each contaminant. How this information is used is entirely up to public health officials who consult them and may in large part be determined by the particular circumstances of each case.
Based on conclusions reached from a meeting with representatives of state public health organizations, the decision was made not to develop a single advisory level. Instead, a broad guidance approach was developed to address the relevant information on a contaminant so that the information can be used as part of the overall determination of the public health significance of contaminants in shellfish.
Selection of Contaminants
The first guidance documents (for arsenic, cadmium, chromium, lead, and nickel) address contaminants chosen because they are most likely to occur and frequent concerns have been raised regarding their presence in shellfish. Concern about these elements is to be expected, since shellfish tend to accumulate elemental contaminants present in the environment.
Relevant Statutory Authority
The guidance documents were developed with a clear understanding of the statutory responsibilities of the Environmental Protection Agency (EPA) (e.g. Clean Water Act), the FDA, and the states. There is no intent to circumvent the statutory authority of any state or federal authority.
Scope
These guidance documents have been designed for use by public health officials at the federal, state and local level, other interested parties and members of the general public. These documents are being issued to present FDA's assessment of the current state of knowledge on specific contaminants and they will be revised when important new information becomes available. The agency requests and strongly encourages anyone with relevant information to add to these documents to contact the Office of Seafood at the address shown on the cover of this document.
Each guidance document begins with a public health statement much like that given in the toxicological profiles published by the Agency for Toxic Substances and Disease Registry (ATSDR). The sections that follow present information about sampling shellfish, analytical procedures for analysis of elemental contaminants in shellfish, consumption and exposure assessments, and conclude with a summary of toxicological and adverse effects of these contaminants. Each document identifies and summarizes the key literature describing the information, including the work of other health organizations (e.g toxicological profiles - ATSDR and environmental health criteria -International Program on Chemical Safety/World Health Organization).
The purpose of the Federal Food, Drug, and Cosmetic Act (FD&C Act) is to ensure a safe and wholesome food supply. The FD&C Act and other related statutes, including the Public Health Service Act, provide the regulatory framework under which the Food and Drug Administration assesses the effects of environmental contaminants on the safety of consumption of fish and shellfish (molluscan bivalves and crustacea). FDA has jurisdiction over foods that are intended for introduction into, or have been shipped in interstate commerce.
The basic provision of the Act dealing with poisonous or deleterious substances in food is section 402(a)(1). This section states:
(a)(1) if it bears or contains any poisonous
or deleterious substance which may render it
injurious to health; but in case the
substance is not an added substance such food
shall not be considered adulterated under
this clause if the quantity of such substance
in such food does not ordinarily render it
injurious to health;
This section applies to all poisonous or deleterious substances, whether they are naturally occurring or added.
Section 402(a) (1) of the Act distinguishes between naturally occurring and "added" substances and sets different standards for each. Naturally occurring substances would adulterate a food only if they render the food "ordinarily" injurious to health. Added substances would adulterate the food if the food "may" be injurious to health.
The term "added" in this legal context is important. An added poisonous or deleterious substance is a substance that does not occur naturally in foods. A "naturally occurring poisonous and deleterious substance" is a substance that is an inherent, natural constituent of a food and is not the result of environmental, agricultural, industrial, or other types of contamination. When a naturally occurring, poisonous, or deleterious substance is increased to abnormal levels through mishandling or other intervening acts, it is regarded as an added substance.
Under conditions where an unavoidable contaminant occurs at a level that is of concern to FDA, FDA may establish a formal tolerance limiting the extent of allowable contamination of a food. Compliance with a tolerance is strictly determined on the basis of the results of analysis of commingled representative subsamples of edible portions taken from a fish shipment, not on the basis of a single fish. However, when toxicological data are scanty or conflicting, when additional data are being developed, or when other conditions are rapidly changing, FDA may choose not to establish a tolerance. Nevertheless, it may still be appropriate to take some regulatory action to control exposure to a contaminant. In such circumstances, FDA may consider developing guidelines or regulatory limits.
The authority of the FD&C Act extends only to fish and shellfish that are moving in interstate commerce. Accordingly, tolerances and guidelines are typically established on a national basis when it is judged that a national problem exists for a particular contaminant. Thus, guidelines are generally, but not always, tailored to national needs and national patterns of consumption. For example, consumption levels of fish on a national per capita basis are generally considerably less than those typical of sports fishermen or of individuals living near lake shores or coastal regions of the U.S. Nonetheless, the toxicity information {Reference Dose, (RFD) or Acceptable Daily Intake (ADI)} that is considered in the setting of national guidelines, may be useful to the states in establishing local controls or advisories on local shellfish consumption. If a potential local health threat exists, a state or locality may wish to issue warnings or provide guidance on the quantity of contaminated shellfish that may be safely consumed. Local authorities can utilize the available federal toxicity information and compare it with the level of the contaminant found locally and local shellfish consumption patterns to establish acceptable levels.
Arsenic and its compounds are widely distributed in nature. Arsenic and its compounds exist primarily in two oxidation states, arsenite (+III) and arsenate (+V). Arsenic trioxide is produced as a by-product of copper, lead, and nickel smelting. Dimethylated arsenic compounds, the most abundant form found in the environment, arises from the conversion of arsenic by microorganisms. This is the predominant form which accumulates in fish. Arsenic compounds are used as defoliants in cotton fields, as a fungicide, herbicide, insecticide, algicide, wood preservative, and for hardening and corrosion resistance in the metallurgy of alloys of copper, lead, and bronze, semiconductor technology, the manufacture of pigments and anti-fouling paints, and therapeutic and veterinary medicine. Arsenic compounds are released to the atmosphere from both natural and industrial sources. The major sources of airborne arsenic emissions are the smelting of metals, burning of fossil fuels, and application of pesticides.
Inorganic and organic arsenic compounds used as pesticides, plant
defoliants, and herbicides may accumulate in agricultural and
horticultural soils and plants. Traces of arsenic are found in
most foods, with the highest concentrations found in seafood,
particularly shellfish, at total arsenic levels ranging up to
30 g/g
wet weight. In sporadic cases lobster total arsenic levels
may exceed 100 g/g because they
specifically store arsenic
(LeBlanc and Jackson, 1973; Bohn, 1975; Benson and Summons,
1981). Nearly all of the arsenic present in seafood is organic
arsenic which is considered to be much less toxic than inorganic
arsenic salts. It is estimated that in the U.S. the mean total
arsenic intake from all food is approximately
30 g/person/day.
Only with high levels of consumption of fish, lobsters and other
seafood might larger quantities be ingested. Human arsenic
exposure in the U.S. (about 50 g/day) occurs mainly from foods
with much smaller exposures from drinking water, cigarette smoke
and ambient air.
The absorption of arsenic in the body is affected by the type of arsenic compound present, its solubility, and its physical form. Arsenic (+V) is more readily absorbed than arsenic (+3) and inorganic forms are more readily absorbed than organic forms. Arsenic (+V) compounds are excreted more rapidly than arsenic (+3) compounds and organic arsenic compounds are excreted more rapidly than inorganic arsenic compounds.
Recent data on the relationship of arsenic oxidation state and exposure indicates that the levels of arsenic in the kidney, liver, bile, brain, skeleton, skin, and blood are 2-25 times higher for arsenic (+3) than for arsenic (+V). In humans, dimethylated arsenic represents 75% of the total arsenic excretion, whereas monomethylated arsenic compounds are excreted in lesser amounts. Arsenic transport from the blood appears to conform to a three-component model indicating , in part, a biomethylation of arsenic. The biological half-lives of inorganic arsenic and methylated forms of arsenic are about 10 and 30 hours, respectively. The normal body burden in adults is 0.01-0.46 mg/kg arsenic.
All soluble arsenic compounds are considered to be poisonous to humans. The trioxide form of arsenic is one of the oldest poisons used by man, dating back to pre-Christian times. The levels of intoxication from arsenic depend upon one's age, state of health, nutritional status, possible accustomization, and the time-dose relationship. Arsenic containing compounds vary in their mammalian toxicity depending on the oxidation state of the arsenic, whether it is in the organic or inorganic form, its physical state, and other factors such as solubility, particle size, and rates of absorption and elimination of the compound. With few exceptions, inorganic arsenic is more toxic than organic arsenic. Toxicity of arsenic (+3) is typically greater than that of arsenic (+5).
Inorganic arsenic can have acute, subacute and chronic effects which may be either local or systemic. Acute toxicity occurs in two forms, the paralytic and the gastrointestinal forms. Chronic toxicity results from the application of sub-toxic doses of arsenic over a long period of time and usually occurs through occupational exposure rather than from a food source. Arsenicals may act as skin contact allergens. Peripheral vascular disorders such as Raynaud's syndrome and mesenteric thrombosis have been reported in children exposed to arsenic through drinking water. Epidemiological studies have demonstrated an evident causal relationship between environmental, occupational, and medicinal exposure of humans to inorganic arsenic and cancer of the skin and lungs. The majority of animal studies did not demonstrate the carcinogenicity of arsenic compounds. A few observations of increased incidence of leukemia and lung cancers suggest that inorganic arsenicals should be considered as cancer promoters rather than as initiators. Arsenicals do not induce gene mutations in microorganisms or eukaryotes. Inorganic arsenicals, after parenteral administration at relatively high exposure levels, are teratogenic in a number of animal species, although after oral administration they did not produce any notable effects on reproduction and development.
The World Health Organization (WHO) recommends a tolerable daily intake of 0.05 mg As/kg body weight from food and no more than 20 g/L in the drinking water (WHO, 1983). The tolerable daily intake figure was developed to be applied to short term and chronic exposures and was not meant to be applied to incidences involving single daily exposures. The low levels of arsenic in the environment are reflected in the low levels in food and water. The consumption of seafood, especially shrimp and lobster, may lead to increased arsenic ingestion; however, most of the arsenic compounds found in shrimp and lobster are methylated, and therefore are relatively low in toxicity.
In this section methods for the determination of elemental contaminants in shellfish are examined. Necessary steps for shellfish analysis are outlined without repeating published methodologies. Specific analytical procedures are cited.
1. Sampling
Collecting, processing, preserving, and shipping primary samples are the critical steps required for generating valid analytical data. A detailed sampling plan for proposed specific sample collection sites should be developed before actual collection is started. The plan should include site locations, target species, target contaminants, sampling times, types of samples, and sample replication. Samples must be representative of the site or lot being studied. In addition, these steps must be conducted in a manner that preserves the integrity of the analyte(s) of interest and precludes inadvertent contamination (EPA, 1991). If sampling of processed shellfish products is required, specific sampling procedures can be found in the FDA Inspection Operations Manual (FDA, 1990).
1.1 Field Sampling Procedure
A minimum of 3 replicate composite samples from each sampling location is recommended for a representative site or lot. The shellfish collected should satisfy any legal requirements for harvestable size or weight or be of consumable or "market" size when no legal harvestable requirements are in effect. Selection of the largest animals is recommended if a worst-case exposure sampling is desired (EPA, 1991).
Animals used in composited samples should be of similar size with the total length (size) of the smallest animal having no less than 75 percent of the total length (size) of the largest animal. Composites should be composed of equal numbers of from 10 to 50 animals (depending on the species and size class being sampled) for a total minimum weight of 500 g of tissue homogenate. The relative difference (in percent) between the overall mean length (size) of replicate composites and the mean length (size) of any individual composite should be no greater than 10 percent (EPA In progress).
1.2 Laboratory Sample Preparation
Shellfish should be preserved with ice if they are not processed immediately. If shellfish are shipped or stored more than 24 h before processing, dry ice must be used. All laboratory samples should be prepared and analyzed, or frozen, as soon as possible after collection, preferably within 48 h. All instruments, work surfaces, and containers used in processing laboratory samples must be composed of materials that can be easily cleaned and that are not themselves potential sources of contamination. To avoid cross-contamination, all equipment used in sample handling should be cleaned before each test sample (EPA, 1991). Cleaning plastic equipment for trace element analysis has been studied and guidelines are given by Moody and Lindstrom (1977).
Equipment should be chosen to minimize contamination. Contamination may be minimized by using titanium, aluminum, quartz, TFE (tetraflouroethylene), polypropylene, or polyethylene utensils, and a glass blender or food processor with titanium or stainless steel blades and Teflon, polyethylene, or polystyrene gaskets (FDA, 1975). Stainless steel that is resistant to corrosion may be used if necessary. Stainless steel scalpels have been found not to contaminate mussel samples (Stephenson et al., 1979). However, other biological tissues (e.g., fish muscle) containing low concentrations of heavy metals may be contaminated significantly by any exposure to stainless steel (EPA, 1991). The predominant metal contaminants from stainless steel are chromium and nickel. Titanium utensils have been used successfully and are available commercially (Zeisler et al.,1988; Private Communication 1). Aluminum is also an option to consider, especially if it is not an element being studied. Titanium and aluminum each have the advantage of contaminating the sample with only one element. Quartz utensils are ideal but expensive. The analyst must know the elements of interest and their approximate levels in the species to be analyzed in order to make an informed decision about utensil selection.
Before shellfish are "shucked" (opened), the external shell surface should be thoroughly cleaned with brush and water to remove all sand, dirt, etc., adhering to the shell to prevent contamination of composite samples. After shellfish are shucked, the meat and liquor are carefully removed from the shell and collected to minimize incorporation of shell fragments and sand into the test sample. The presence of shell fragments or sand will complicate mineralization. Separation can be accomplished by transferring the meat and decanting the liquor, leaving sand and shell fragments behind. For analysis of canned, frozen, or otherwise processed products, the test sample should include meats, liquor, and any breading. Thaw frozen products prior to homogenization (McMahon and Hardin, 1968). Once shellfish are frozen, thawing of shellfish samples should be kept at a minimum during tissue removal procedures to avoid loss of liquids.
Comminute entire test sample in a glass blender or food processor to obtain a homogeneous mixture. When a food processor is used, adding dry ice successfully freezes the suspension and minimizes water separation. Precaution: If dry ice is used, check dry ice for interferences and store frozen test samples loosely covered overnight to allow for dissipation of carbon dioxide. Composite test samples should be homogenized to a paste-like consistency before portions are taken for analysis. Store prepared test samples frozen in glass jars with Teflon-lined lids or in Teflon containers. If analysis is not started immediately after homogenization, appropriate analytical portions may be weighed and stored frozen. This will eliminate the need to rehomogenize a thawed test sample.
2. Analytical Methodology
Review of the literature reveals that many inorganic analytical studies of shellfish have been prompted largely by a concern that abnormally high levels of toxic elements contaminate some of the world's shellfish growing areas (Marcus and Thompson 1986; Lytle and Lytle, 1982; Watling and Watling, 1982; May and McKinney, 1981; Micallef and Tyler,1989; Balogh, 1988; Eisenberg and Topping, 1984; Gault et al., 1983; Meeus-Verdinne, 1983; Favretto et al., 1981; Cooper et al., 1982).
This concern has focused mainly on elements of biological interest including antimony, arsenic, cadmium, chromium, copper, lead, manganese, mercury, molybdenum, nickel, selenium, tin, and zinc. Various techniques have been applied to analysis of shellfish; however, in the laboratories surveyed (Private Communications 2-5) routine shellfish analysis is conducted usually by using only a few methods. Most methods of shellfish analysis require mineralization of test portions. The digestion procedures can then be coupled with the determinative techniques presented in 2.2 of this section to form complete analytical methods.
2.1 Mineralization
Digestion procedures fall into one of two general categories, dry ash or wet ash. Dry ash digestions use a long, slow ashing step, usually performed overnight in a muffle furnace. The ashing process is completed by the addition of a small quantity of inorganic acid to the residue and evaporation on a hot plate. The residue is redissolved in acid and the solution is brought to volume with distilled or deionized water. Dry ash procedures can be found in the Association of Official Analytical Chemists (AOAC) method 972.23 (Helrich, 1990). Element loss during the ashing process is greater for dry ash procedures than for wet ash procedures. "Ashing aids" are often employed to prevent loss of the more volatile elements like arsenic and selenium (Maher, 1983; May, 1982).
Wet ash digestions are characterized by short ashing times (normally 3 to 4 h) and the use of acids (HCl, HNO3, H2SO4, HClO4). Wet ashing is performed by adding acid or an acid mixture to the test portion and boiling until digestion is complete. A variety of procedures have been developed using different types of digestion apparatus to retain volatile elements. The most common procedure is performed in beakers on a hot plate. The analytical portion is boiled with HNO3 or a mixture of HNO3 and HCl until digestion is complete (FDA, 1975; ASTM, 1986). Combinations of HNO3, HClO4, and H2SO4 have also been used for more efficient oxidation. Perchloric acid requires constant operator attention, specialized equipment (HClO4 hoods, fume traps, etc.), and safety precautions (Michie et al., 1978).
"Hot-block" digesters have been used for improved temperature control and their reflux capabilities (Agemian et al., 1980). "PTFE (polytetrafluoroethylene) bomb" digesters have been used because digestion can take place in a sealed vessel, thereby retaining volatile elements (Welz and Melcher, 1985).
Another wet ash approach that has been applied to fish analysis is microwave digestion (Schnitzer et al., 1988; Nadkarni, 1984; McCarthy and Ellis, 1991). This technique has been subject to extensive research and development since microwave technology was first applied to digestions (Abu-Samra et al., 1975; Barrett et al., 1978; Nadkarni, 1984; White and Douthit,1985; Kingston and Jassie, 1986, 1988; Burguera et al., 1986; Fernando et al., 1986). Current systems incorporate many features to improve digestion time, element recoveries, contamination, and safety. Improved recoveries can be achieved with systems operating under power-regulated control of pressure because these systems allow digestions to be completed without venting volatile elements. Sealed digestion vessels also reduce acid consumption and, in combination with Teflon construction, reduce contamination problems. Microwave digestion methods developed for shellfish do not require HClO4 to complete mineralization. Microwave digestion has been compared to some of the other digestion procedures (Blust et al., 1988).
Regardless of the digestion procedure selected, appropriate quality assurance and quality control guidelines need to be followed. These include the use of contamination control, digestion blanks, spiked test portions, replicate analyses, an appropriate standard reference material (e.g., National Institute of Standards and Technology: Oyster or Bovine Liver), and recovery calculations (Boyer and Horwitz, 1986; Jones, 1988).
2.2 Determinative Techniques
Selection of a determinative technique is often dictated by available equipment and quantitation requirements. Ease of operation, number of analyses, and familiarity of personnel with techniques are also influential factors. For these reasons only a few methods developed for shellfish analysis are used for routine testing. Spectrometric techniques are most commonly used and include flame atomic absorption spectrometry (FAAS), graphite furnace atomic absorption spectrometry (GFAAS), and inductively coupled plasma-atomic emission spectrometry (ICP-AES) (see Table 1).
FAAS is often used to determine many elements in shellfish analysis because it is easy and affordable (Private Communications 2-5). The American Society for Testing and Materials (ASTM) method D3559 (ASTM, 1986) and the American Public Health Association (APHA) method 303 (APHA, 1986) both use FAAS. FAAS has been combined with high-performance liquid chromatography (HPLC) for the study of metal ion speciation in biological systems. Information about the distribution of metal ions in biological extracts can be used to estimate exposure of an animal to chemical toxins, and to estimate the relationship between chemical form and bioavailability (Guy, 1985).
FAAS has been applied to two methods to improve detection of some elements. First, measurement of arsenic and lead has been improved by on-line generation of volatile hydrides with hydride generation-AAS (HG-AAS) (Holak, 1969). In applications of hydride generation to fish analysis, (Agemian and Thompson, 1980; Julshamn et al., 1982; Cutter, 1985; Maher, 1982, 1986; Madrid et al., 1988) the enhancement of detection limits ranged from a factor of 5 to a factor of over 500 (see Table 1). Hydride generation (HG) has also been coupled with GFAAS (Willie et al., 1986; Sturgeon et al., 1986), direct current plasma-AES (DCP-AES) (Panaro and Krull, 1984), and ICP-AES (de Oliveira et al., 1983; Montaser and Golightly, 1987). HG-AAS has been coupled with HPLC for determination of inorganic and methylated arsenic species in marine samples (Maher, 1981).
GFAAS has been used for measurement of selected elements when FAAS determination limits were insufficient for shellfish analysis (Cruz et al., 1980;Serra and Serrano, 1984). Clearly, GFAAS is more sensitive than FAAS although GFAAS analysis time is longer. The main problems with GFAAS shellfish analysis are matrix and spectral interferences. Deuterium arc and Zeeman-effect background corrections have been applied to reduce spectral interference in shellfish analysis. While deuterium arc background correction is plagued by spectral interferences, Zeeman-effect background correction appears to be satisfactory for routine analysis of marine biological tissue (Welz and Schlemmer, 1986; McMahon et al., 1985). Various "matrix modifiers" (organic and inorganic reagents) have been used to reduce matrix effects. Molybdenum, lanthanum, and ammonium dihydrogen phosphate have been used as matrix modifiers for shellfish analysis (Poldoski, 1980; Mckie, 1987). Solvent extraction is another approach taken to minimize matrix effects (Stephenson and Smith, 1988; Harms, 1985).
Solid sampling has been developed for GFAAS to increase relative sensitivity by avoiding the large dilution factors required in the sample dissolution. Solid sampling also eliminates contamination and/or loss of analyte resulting from dissolution. Solid sampling has not been widely used for shellfish analysis because of possible interference by very large background matrix absorbance (Chakrabarti, 1979; May and Brumbaugh, 1982). A complete analysis by GFAAS can be found in ASTM method D3559 (ASTM, 1986).
ICP-AES has also been used for routine elemental analysis of shellfish (Grant and Ellis, 1988). Sequential or simultaneous mode ICP-AES allows rapid analysis, dramatically improving throughput. Detection limits in ICP-AES are generally the same as those in FAAS except for elements that are difficult to atomize. Such elements have lower ICP-AES detection limits because of the higher temperature of the plasma (see Table 1). A complete analysis by ICP-AES can be found in the EPA's water method 200.7 (EPA, 1979).
Although not widely used for routine shellfish analysis, techniques have been developed using voltammetry for trace element determination (ASTM, 1986; Adeloju and Bond, 1983; Breyer and Gilbert, 1987), polarography (Lemly, 1982), ICP-MS (Munro et al., 1986; Beauchemin et al., 1988; Park et al., 1987), HPLC with electrochemical and spectrophotometric detection (Bond and Nagaosa, 1985), and neutron activation analysis (NAA) (Zeisler et al., 1988; LaTouche et al., 1981; DeSilva and Chatt, 1982; Chisela and Bratter, 1986).
Regardless of the analytical method chosen, sound scientific analytical principles must be followed. All methods used by a laboratory must be validated by the laboratory prior to routine sample analysis. The detection and quantitation limits, and accuracy and precision of each method must be assessed and documented to be sufficient for reliable quantitation at or below the level of interest for each metal. All analytical methods used routinely for the analysis of shellfish should be documented thoroughly; formal standard operating procedures (SOP's) are preferred. A published method may serve as an analytical SOP only if the analysis is performed exactly as described. Analytical SOP's should be followed as written (EPA, 1991).
3. Quality Assurance/Quality Control (QA/QC)
Initial demonstration of laboratory capability and routine analysis of appropriate QA/QC samples to document data quality and to demonstrate continued acceptable performance is essential. QA/QC requirements should be based on specific performance criteria, or control limits, for data quality indicators, such as accuracy and precision. Typically, control limits for accuracy are based on the historical mean recovery plus or minus three standard deviation units, and control limits for precision are based on the historical standard deviation or coefficient of variation (or mean relative percent difference for duplicate samples) plus three standard deviation units. Some recommended QA/QC procedures include: instrument calibration and calibration checks, assessment of method detection and quantitation limits, assessment of method accuracy and precision, routine monitoring of interferences and contamination, regular external QA assessment of analytical performanceinterlaboratory comparison programs, appropriate documentation, and reporting of data (including QA/QC data) (EPA, 1991).
Table 1. Quantitation Limits (
ICP-AES FAAS GFAAS(f) Solution Hydride Solution Hydride Element neb.(b) gen.(c) neb.(d) gen.(e)
As 0.5 0.008 10.0 0.03 0.01 Cd 0.05 0.007 0.0005 Cr 0.3 0.20 0.003 Ni 0.2 0.3 0.02 Pb 0.8 0.03 0.3 0.003 0.005 (a) Quantitation limits (10 sigma) were estimated from instrumental detection limits (3 sigma), and a 5 g test portion in 50 ml of test solution was assumed. (b) Values for pneumatic nebulizer/ICP-AES are taken from Applied Research Laboratories literature on Model 3580 OES. (c) Values are reported in Montaser and Golightly (1987). (d) Values are taken from Varian literature on Models AA-1275 and AA-1475 best value was selected for either C2H2-air or C2H2-N2O flame. (e) Value for arsenic are reported in Robbins and Caruso (1979); value for lead in Ikeda et al. (1982). (f) Values are taken from Varian literature on Model Spectra AA-300 (nominal 20 uL injection).
The following sections provide estimates of chronic shellfish intake as well as estimates of arsenic exposures resulting from chronic shellfish consumption. In addition, estimates of arsenic exposure are provided for background sources, both dietary (i.e., non-seafood) and non-dietary sources.
1. Shellfish Intake
The frequency of shellfish eating occasions has been tabulated in the Market Research Corporation of America (MRCA) 14-day survey (5-Year Menu Census, 1982-87) (MRCA, 1988). The MRCA reports that only 13% of the surveyed population consumed crustaceans and only 4.8% of the surveyed population (25,726 individuals, 2+ years) consumed molluscan bivalves. Using standard portion sizes from the USDA's 3-day National Food Consumption Survey (NFCS, 1977-78) (Pao et al., 1982), we estimate the 14-day-average mean and 90th percentile daily intakes of molluscan bivalves. These are presented in Table 2. The intakes for crustacean shellfish are presented in Table 3.
Table 2. 14-Day-average intake of molluscan bivalves,
grams/person/day, for eaters-only.
Age Group Mean 90th
Percentile
2+ years (all ages) 10 15
2-5 years 4 8*
(male/female)
18-44 years** 12 18
(male/female)
* Estimated value. Reliable data are not available in the MRCA
survey. The 90th percentile value is estimated to be twice
the mean (WHO, 1983)
** USDA portion size for 33-44 year age group used in the
calculation. This age subgroup has the highest consumption of
any subgroup in the 18-44 year range (Pao et al., 1982)
Table 3. 14-Day-average-intake of crustacean shellfish,
grams/person/day, for eaters-only.
Age Group Mean 90th
Percentile
2+ years (all ages) 9 17
2-5 years 5 10
(male/female)
18-44 years* 9 19
(male/female)
* USDA portion size for 33-44 year age group used in the
calculation. This age subgroup has the highest consumption of
any subgroup in the 18-44 year range (Pao et al., 1982).
2. Arsenic Concentrations in Shellfish
The recent National Oceanic and Atmospheric Administration (NOAA) Mussel Watch project progress report (NOAA, 1989) indicates that none of the mussels or oysters in the 169 sites examined in 1988 exhibited an average total arsenic concentration in excess of 14 ppm (wet weight). This conclusion was reached by applying the following factors to convert the dry weight concentrations reported by NOAA to wet weight values: Crassostrea virginica, 0.124; Mytilus edulis, 0.121; Mytilus californianus, 0.140; Ostrea sandvicensis, 0.146 (Private Communication, 1990).
In 1985-86 the FDA surveyed the levels of arsenic in softshell clams (Mya arenaria), hardshell clams (Mercenaria mercenaria), Eastern oysters (Crassostrea virginica), and Pacific oysters (C. gigas) (S. Capar, FDA, Division of Contaminants Chemistry, unpublished data). The shellfish samples were harvested from approved waters in 20 coastal states (all coastal states except Alaska and New Hampshire). The results of that survey are presented in Table 4.
Table 4. Results of the 1985-86 FDA shellfish survey for arsenic.
Mean arsenic level(1) Range Species (
Clam, hardshell 2.7 0.82-6.6 Clam, softshell 1.2 N/A(2) Oyster, Eastern 1.1 0-2.8 Oyster, Pacific 1.8 1.4-2.1 Clams, overall 2.0 Oysters, overall 1.5 (1) Wet weight basis. (2) Not available.
In 1978, the National Marine Fisheries Service (NMFS) published the results of a survey on trace elements in fish (Hall et al., 1978). The arsenic levels in several species of shellfish were reported. The results are presented in Table 5.
Table 5. Results of the NMFS survey of trace elements in the
fishery resource for selected shellfish.
Range of mean arsenic levels(1,2)
Species (g/g)
Clam, hardshell 3.0 - 4.0
Clam, softshell 2.0 - 3.0
Oyster, Eastern 3.0 - 4.0
Oyster, Pacific 3.0 - 4.0
Range of mean arsenic levels for 2.8 - 3.8
molluscan bivalves
Crab, blue, body/claw 3.0 - 4.0
Crab, dungeness, body/claw 5.0 - 6.0
Lobster, American, claw/tail 10.0 - 20.0
Lobster, spiny, Atlantic, tai 10.0 - 20.0
Lobster, spiny, Pacific, tail 20.0 - 30.0
Shrimp, ocean 3.0 - 4.0
Shrimp, pink (northern) 9.0 - 10.0
Range of mean arsenic levels for 8.6 - 10.6
crustaceans
(1) Wet weight basis.
(2) Shellfish from a single survey site formed a sample. Arsenic
content was averaged for each sample. Each species was sampled
at many sites. This table reports the range of the mean arsenic
levels that encompassed all samples.
Arsenic levels reported in the NMFS survey are consistently higher than those reported in the FDA survey. No clear explanation for this is available. In general, greater numbers of each species were sampled in the NMFS survey. The FDA survey data, however, were gathered seven years or more after the NMFS survey. Perhaps levels of available arsenic in the environment had dropped as the result of abatement actions to such an extent that marine organisms sampled in 1985 had lower arsenic levels than in 1978. Neither the FDA nor the NMFS values for molluscan bivalves exceed the maximum arsenic concentrations found by the NOAA Mussel Watch program. The NMFS is beginning a resurvey of trace metals in marine organisms and the results of this survey may delineate any trend in arsenic contamination in marine organisms. Until such data are available to corroborate the FDA study, we believe it is prudent to use the NMFS data to estimate consumer exposure.
3. Arsenic Exposure from Shellfish
Using information on chronic shellfish consumption and typical
levels of arsenic in shellfish, it is possible to estimate
arsenic intake on a chronic basis for consumers of shellfish.
According to the MRCA survey, only 13% of the population eat
crustacean shellfish, and only 4.8% of the population eat
molluscan bivalves. Assuming that mean arsenic levels range from
2.8 g/g (ppm) to 3.8 g/g for molluscan bivalves and from
8.6 g/g to 10.6 g/g for crustacean shellfish, consumer
exposures (for eaters only) are presented in Tables 6 and 7, below.
Table 6. Molluscan bivalves - chronic arsenic intake for
specified levels of contamination, g/person/day.
Contam. at 2.8 ppm Contam. at 3.8 ppm
Age Group, Yrs. Mean 90th Mean 90th
Percentile Percentile
2+ (all ages) 28 42 38 57
(male/female)
2-5 11 22* 15 30*
(male/female)
18-44 34 50 46 68
(male/female)
* Estimated value. Reliable data are not available in the MRCA
survey. The 90th percentile value is estimated to be twice the
mean (WHO, 1983)
Table 7. Crustacean shellfish - chronic arsenic intake for
specified levels of contamination, g/person/day.
Contam. at 8.6 ppm Contam. at 10.6 ppm
Age Group, Yrs. Mean 90th Mean 90th
Percentile Percentile
2+ (all ages) 77 146 95 180
(male/female)
2-5 43 86* 53 106*
(male/female)
18-44 77 163 95 201
(male/female)
* Estimated value. Reliable data are not available in the
MRCA survey. The 90th percentile value is estimated to be
twice the mean (WHO, 1983).
4. Background and Relative Source Contribution to Arsenic Exposure
4.1 Diet
Data from the FDA Total Diet Study (TDS) (7-Year Summary, 4/82 - 2/89) (E. Gunderson, FDA, Division of Contaminants Chemistry, private communication) suggest that the mean chronic exposure to total arsenic (no distinction between organic and inorganic forms of arsenic) from all food is 32 g/p/d (averaging male and female lifetime total exposures). The data from the individual TDS food category contributions to this total show that approximately 90% of the total is derived from cod/haddock filets, shrimp, fish sticks, and tuna (these are the only seafood categories in the FDA Total Diet Study).
A Canadian group examined the intake of arsenic by studying the diets of 24 individuals in 5 cities (Dabbeka et al., 1987). The average arsenic intake was 16.7 g/day, lower than the FDA-TDS results, but only 32% of the intake was attributed to seafood. Rice, another food containing relatively high amounts of arsenic, was consumed by many of the participants, contributing up to 18% of the total arsenic intake.
The inorganic arsenic content of seafood is generally quite low (Chapman, 1926; Lunde, 1973, 1977, 1983; Edmonds et al., 1977; Cooney et al., 1978; Edmonds and Francesconi, 1981; Brooke and Evans, 1981; Flanjak, 1982; Maher, 1983). Approximately 80% - 99% of the arsenic in seafood is present in an organic form (fish arsenics) which is not toxic (Nriagru and Simmons, 1990). There is no information readily available regarding the organic arsenic fraction of nonseafood arsenic; therefore, a conservative position would be to assume that all of the nonseafood arsenic is inorganic.
Arsenic from nondietary, environmental (water, soil, and air) sources does not contribute significantly to background arsenic intake except for those individuals living in the immediate vicinity of arsenic-treated materials (e.g., cotton gin trash), incinerators, or smelting operations (Ellen, 1990). Regardless, we shall examine the environmental contributions to arsenic exposure in more detail below.
Data are available from the EPA on the relative contributions to arsenic exposure from these sources. Additional data have been taken from the Agency for Toxic Substances and Disease Registry's toxicological profile for arsenic (ATSDR/TP-88/02, 1989). Of the environmental sources, only drinking water, due to its high relative consumption, contributes to nonseafood arsenic exposure.
4.1 Water
A 1978 survey (Greathouse and Croun, (1978)) has shown that the
average level of arsenic in drinking water is approximately 2
g/L, with 99.6% of all samples
containing less than 10 g/L
(ATSDR/TP-88/02, 1989). The EPA maximum contaminant level (MCL)
for arsenic in drinking water is 50
g/L. Tap water intake has
been estimated at 1193 mL/p/d by the Life Sciences Research
Office of the Federation of American Societies for Experimental
Biology [includes water consumed directly as a beverage and added
to foods in preparation (Ershow and Cantor, 1989)|. The average
water contribution to background arsenic exposure would thus be
approximately 2.4 g/p/d. We note that
the FDA-TDS, from which
our nonseafood food contribution to arsenic exposure was derived,
includes tap water as a beverage explicitly in its sample diet,
as well as in water used in the preparation of food. Therefore,
the 2.4 g/day estimate of water's
contribution to arsenic
exposure has been taken into consideration in our previous
estimate of background arsenic exposure.
4.2 Air
The EPA has estimated that the average inhalation rate for an adult is 20 m3/day (ATSDR/TO-88/02, 1989). The typical concentration of arsenic in the air has been reported to be 0.003 g/m3 (EPA, 1989), resulting in a daily exposure from air intake of 0.06 g. Concentration levels can be as much as 10 times higher in the vicinity of smelting operations or near sites where arsenic-treated trash is burned (Suta, 1980).
4.3 Soil
Walsh and Keeney (1975) reported that average soil levels for
arsenic are between 5 and 6 ppm. Calabrese et al. (1987) has
recently summarized the results of studies on soil ingestion in
children. He estimates 10 mg/day of soil ingestion for adults
and children over the age of five years old, with an
"intermediate tendency" to ingest soil [see also EPA Exposure
Factors Handbook (EPA, 1989)|. Typical soil ingestion levels for
children below age six have been reported to be approximately 200
mg/day [see Walsh and Keeney (1975), Part II, p. 1-18|. The
Coalition for Safe Ceramicware, in its report on lead intake to
FDA (1989), estimates soil ingestion for adults of 1 mg/day. The
contribution of nonseafood exposure to arsenic for adults will
thus be taken as 0.005 g/day
(5 ppm x 1 mg soil/day) because
the data for the nonseafood food contribution (from the FDA-TDS)
were derived from the intake of individuals over their lifetimes.
The background exposures that have been derived are mean values, meant for those persons not living near sources of arsenic contamination. The background arsenic exposures for those living or working near high concentration sources of the contaminants would have to be evaluated using information specific to those sites.
5. Select Subpopulations
The preceding exposure estimates for arsenic intake from shellfish consumption are based on nationally representative food consumption surveys. These surveys may not be suitable for estimating exposures to particular subpopulations or to individuals residing in specific regions of the country. State or local organizations will want to tailor estimates to their specific needs.
In addition to the nationwide food consumption surveys mentioned previously [(MRCA, 1988) 14-day Survey; USDA-NFCS, 3-Day|, the National Purchase Diary survey (NPD Group) presents a 14-day survey covering the continental United States.
Several other food consumption databases are also available for use in preparing estimates of exposure. Although none have the national character of the MRCA, NPD Group, or USDA-NFCS, they do address either the consumption of specific types of food (e.g., fish, shellfish), or they reflect the consumption patterns and habits of regional areas or of specific groups (e.g., fishermen). Some available databases and selected shellfish consumption values are listed in Table 8 below.
Table 8: Shellfish Consumption
Seafood Consumption Level Population Source of the Data Category (g/day)
All 5 (availability(a)) Entire U.S. NMFS "Fisheries of the U.S." Shellfish Series (1986)(b) 5 (average) Persons Earning NMFS (1981)(c) More than $15,000 per Year 9 (average) Shellfish-Eaters 1977-1978 Menu Census(d) 16 (90th percentile) Shellfish-Eaters 1977-1978 Menu Census(d) 30 (availability(a)) Shellfish-Eaters NMFS "Fisheries of the U.S." Series (1986)(b) Clams 1 (average) New Englanders NMFS (1981)(c) 22 (95th percentile) Texans 1973-1974 Tuna Institute Survey(e) Oysters 0.5 (average) Persons Earning NMFS (1981)(c) more than $15,000 per year 21 (95th percentile) Californians 1973-1974 Tuna Institute Survey(e) (a) Total availability of edible shellfish "meat" (excluding bones, viscera, shells), fresh, frozen and canned. (b) C. Lewis, FDA, in memo to I. Boyer, FDA, "Estimating a Level of Concern for Cadmium in Shellfish," 10-26-88. (c) Hu, T.-W., (1985). (d) Market Research Corporation of America (MRCA) as reported by C. Lewis, FDA, in memo to I. Boyer, FDA, "Estimating a Level of Concern for Cadmium in Shellfish," 10-26-88. (e) Tuna Research Foundation (1975), Seafood Consumption Study, National Purchase Diary Panel, Inc., Schaumburg, IL.
It is important to note that the level of consumption, and thus the level of contaminant intake, varies from survey to survey. The variation is controlled by several factors, among which are:
Finally, we note that in some circumstances it may be difficult or impossible to find a survey that adequately addresses a particular public health problem. In this case, the best choice may be to initiate a survey that will be specifically designed to address the question.
Methyl transferase enzymes are important in the methylation of arsenic in mammals. The effect of dietary deficiencies and genetic variability on methylating capacity may have important implications for tissue distribution and individual susceptibility to arsenic toxicity. Experimental animals fed protein-deficient diets while exposed to high levels of arsenic exhibited a decreased methylating capacity, leading to increased deposits of arsenic in the liver, lung, and other organ sites and presumably to arsenic toxicity. There is evidence that arsenic is an essential element in some animal species, and possibly also in humans. However, chronic inorganic arsenic exposure in humans has also been associated with certain characteristic lesions including, most commonly, symmetrical palmar and plantar hyperkeratosis, warts, and melanosis of the skin (Buchanan, 1962; Nakamura et al., 1973; Borgono and Greiber, 1972; Borgono et al., 1977; Yeh, 1963; Yeh Borgono and Greiber, 1972; Borgono, 1977 Yeh, 1963; Yeh and How, 1963; Tseng, 1968; 1977; Pershagen, 1981). Epidemiological studies suggest that a causal relationship exists between skin cancer and heavy exposure to inorganic arsenic via medication, contaminated well water, or occupational exposure (Borgono et al., 1977; Ishinishi et al., 1977; Tseng, 1977; Tsuchiya, 1977; Penrose et al., 1974; Buchanan, 1962; Pinto and McGill, 1953; Hill and Faning, 1948). In particular, Tseng (1968) established that a strong geographical correlation appears to exist between inorganic arsenic levels in drinking water and the skin cancer rate. However, attempts to induce cancer with arsenic in laboratory animals have been uniformly unsuccessful. Thus, it has been suggested that inorganic arsenic compounds are tumor promoters rather than carcinogenic initiators (Axelson et al., 1978; Mabuchi et al., 1979).
Extensive reports in the literature document the in vivo methylation of inorganic arsenic to monomethyl and dimethyl arsenic derivatives (the latter being the major methylated metabolite) in humans and animals. In humans, dimethylated arsenic represents approximately 75% of the total arsenic excretion, monomethylated arsenic being excreted in lesser amounts. This biotransformation is dose-dependent and high exposure will give relatively more of the monomethyl form. In humans, arsenic (+5) undergoes reduction in vivo to arsenic (+3). A portion of the reduced arsenic species is then biomethylated predominantly in the liver to methylarsenic acid and dimethylarsenic acid. This methylation process may represent detoxification since the metabolites exhibit less acute toxicity in experimental lethality studies.
Like inorganic arsenic, organoarsenic compounds in seafoods are readily absorbed (>80%) from the gastrointestinal tract in animals and humans (Bettley and O'Shea, 1975; Charbonneau et al, 1978 a, b; Coulson et al., 1935; Westoo and Rydalv, 1972; Munro, et al. 1974; Munro, 1976; Urakubo et al., 1975; Dutkiewitz, 1977; Maranfante, 1984). However, more than 80% of the ingested arsenic-in-fish is excreted in the human urine within a few days after ingestion (Chapman, 1926; Coulson et al., 1935; Crecelius, 1977; Luten et al, 1982; Tam et al., 1982). Canon et al., (1983) showed that arsenobetaine, an organoarsenic compound found in seafood, was excreted unchanged in the urine within a few days after ingestion (Chapman, 1926; Coulson , 1935; Crecelius, 1977; Luten et al, 1982; Tam , 1982). Canon et al. (1983) also found that the arsenobetaine compound was excreted unchanged in the urine of volunteers eating fishery products.
Accordingly, arsenobetaine administered to mice by i.p. injection at doses as high as 500 mg/kg body weight was rapidly excreted, unchanged, and no signs of toxicity were produced (Canon et al., 1983). The arsenic levels in the carcasses of the treated mice were indistinguishable from those of the controls within 7 days after the dose was administered. The low acute toxicity of arsenobetaine and related compounds (e.g. arsenocholine) has been demonstrated in rats and chicks as well as in mice (Welch and Landau, 1942; Welch and Welch, 1938). Coulsen et al. (1935) reported that rats ingesting shrimp containing 17.7 ppm arsenic ("shrimp arsenic") for 12 months exhibited no alterations in growth, physical appearance and activity. The rapid elimination of the ingested "shrimp arsenic" in these rats was comparable to that observed in volunteers consuming 980-1180 g arsenic in boiled shrimp. Only 0.7% of the arsenic from the ingested shrimp accumulated in the bodies of the exposed rats during the first three months of the study. In contrast, more than 18% of the inorganic arsenic accumulated in the rats when arsenic (+3) oxide (17.7 ppm As203) was added to either a stock diet or to a shrimp diet with low (2 ppm) "shrimp arsenic" content. Dietary inorganic arsenic accumulated to yield 50-65 times more arsenic in the carcasses of the exposed rats than the controls. Similarly, dietary inorganic arsenic produced liver arsenic concentrations which were more than 100 times the concentrations in the control livers.
In accord with the epidemiological studies mentioned above, inorganic arsenic has been shown to be mutagenic in animal test systems (Leonard and Lauwerys, 1980; Ishinishi et al., 1980; Leonard, 1984). In contrast, arsenobetaine exhibited no mutagenicity in the Ames test (Salmonella typhimurium) in the presence or absence of metabolic activators (Cannon, 1983). Also, Jongen (1985) showed that arsenobetaine does not appear to be mutagenic in assays for sister chromatid exchange and forward mutation of the HGPRT gene. In addition, arsenobetaine was added to the positive controls in the various test systems used. The results suggest that arsenobetaine and related organoarsenic compounds lack initiating and promoting activities.
There have been no reports of ill effects among the many regional
and ethnic populations throughout the world who consume large
quantities of seafood that result in exposures to as much as 50
g organic As/kg/day (3,000 g organic
As/day for a 60 kg
individual). This observation supports the conclusion that the
organic forms of arsenic in seafood present little or no hazard
to health. In contrast, exposure to the inorganic forms of
arsenic in food and drinking water may produce adverse health
consequences.
Using information summarized above, the WHO/FAO (1989) has
suggested a provisional maximum tolerable weekly adult intake
(PTWI) for inorganic arsenic of 0.015 mg/kg of body weight.
Thus, the WHO/FAO provisional maximum tolerable intake is about
130 g inorganic As/day for a 60 kg
individual (15 g/kg/week x
60 kg / 7 days/week = 128.6 g/day).
This tolerable daily intake
figure is meant to apply to short term and chronic exposures and
is not meant to apply to incidences involving acute exposures.
The following section illustrates how levels of concern, either permitted amounts of chronic shellfish consumption or permitted levels of contamination, can be determined using information on the tolerable daily intake for arsenic. For the purpose of this illustration, it is assumed that total arsenic exposure is derived solely from shellfish. This approach leads to estimates of the maximum permitted levels of either shellfish intake or levels of arsenic contamination in the shellfish. If other sources of arsenic exposure are to be considered (e.g., relative source contribution), then corresponding adjustments in the levels of concern will need to be made.
It is recommended that the molluscan bivalve and crustacean shellfish intakes be considered separately in the preparation of guidance levels. As can be seen from data presented in the CONSUMPTION AND EXPOSURE ASSESSMENT section, these shellfish have very different contaminant levels. Therefore, use of a total shellfish intake figure when deriving a guidance level for arsenic may lead to bias against one or the other types of shellfish.
The tolerable daily intake for arsenic is based on exposure to inorganic arsenic. Although nearly all of the arsenic present in seafood is in an organic form (approximately 80-99%), most published studies and most monitoring of arsenic levels in shellfish employ an analytical method like the one presented in this document that determines total arsenic. To use traditional monitoring results to evaluate acceptable levels of shellfish consumption or acceptable levels of arsenic contamination, an estimation procedure that accounts for the low fraction of inorganic arsenic present in shellfish should be used. Available information indicates that inorganic arsenic levels will on the average account for less than about 10% of the arsenic in shellfish. In lieu of analytical methods that specifically measure inorganic arsenic levels in shellfish, an estimate of inorganic arsenic concentrations in shellfish can be obtained using this percentage factor and total arsenic levels in shellfish.
The following equations illustrate how the tolerable daily intake
for inorganic arsenic (130 g/day
for a 60 kg person) can be
combined with information on either shellfish consumption or
total arsenic levels to estimate the corresponding levels of
concern of either total arsenic contamination levels or shellfish
intake.
Total As Level of Concern =
{Tolerable Daily Intake} / {Daily intake of Shellfish x 10%}
As an example, the Total As Level of Concern is estimated for individuals in the 2+ year age group consuming molluscan bivalves on a chronic basis at the 90th percentile level (15 g/p/d, see Table 2).
Total As Level of Concern = {130 g/p/d} / {15 g/p/d x 10%} = 86 g/g
For molluscan bivalves with total arsenic levels corresponding to the highest mean concentration reported in the 1978 NMFS survey (see Table 5), the corresponding molluscan bivalve Consumption Level of Concern is calculated as follows:
Consumption Level of Concern = {130 g/p/d} / {3.8 g/g x 10%} = 340 g/p/d
The following tables present a number of plausible levels of concern, either consumption or concentration, that could be used in the assessment of the potential impact of arsenic on human health through exposure from shellfish. For comparison, the following values should be compared with intake and arsenic contamination figures presented in Section IV.
Table 9. Consumption Levels of Concern for Inorganic Arsenic
Consumption Levels of Shellfish Concern (g/day)*
Mollusks 340 Crustaceans 120 * Derived from the tolerable daily intake for a 60 kg person, the highest mean concentrations for total arsenic (see Table 5) reported in 1978 NMFS survey (3.8 ppm and 10.6 ppm, for mollusks and crustaceans, respectively), and the 10% adjustment factor for inorganic arsenic levels in shellfish.
Table 10. Molluscan Shellfish: Concentration Levels of Concern for Total Arsenic
Levels of Concern (
2+ years (all ages) 130 86 2-5 years 110** 55** (male/female) 18-44 years 110 72 (male/female) * Derived from the tolerable daily intake, the consumption figures for molluscan shellfish presented in Table 2, and the 10% adjustment factor for inorganic arsenic levels in shellfish. ** Assumes body weight of 20 kg for children 2-5 years.
Table 11. Crustacean Shellfish: Concentration Levels of Concern for Total Arsenic
Levels of Concern (
Abu-Samra, A., Morris, J. S. and Koirtyohann, S.R. (1975) Anal. Chem. 47:1475.
Adeloju, S.B. and Bond A.M. (1983) Anal. Chim. Acta 148:59.
Agemian, H., Sturtevant, D.P. and Austen, K.D. (1980) Analyst, 105:125.
Agemian, H. and Thomson R. (1980) Analyst 105:902.
ASTM (American Society for Testing and Materials) (1986) Annual Book of ASTM Standards, 11.01, D3559-85, ASTM, Philadelphia, PA 19103.
APHA (American Public Health Association) (1986) Standard Methods for the Examination of Water and Waste Water, 16th Ed., APHA, 1015 15th St., N.W., Washington, DC 20005.
ASTDR/TP-88/02 (Agency for Toxic Substances and Disease Registry), (1989) "Toxicological Profile for Arsenic." under contract No. 68-02-4228. Prepared by Life Systems Inc. for ASTDR, US Public Health Service (PHS), and the US Environmantal Protection Agency.
Axelson, O., Dahlgren, E., Jansson, C.-D. and Rehnlund, S.O., (1978) Br. J. Ind. Med. 35:8.
Balogh, K.V. (1988) Bull. Environ. Contam. Toxicol. 41(6):910.
Barrett, P., Davidowski, L.J., Penaro, K.W. and Copeland, T.R. (1978) Anal. Chem. 50:1021.
Beauchemin, D., McLaren, J.W. and Berman, S.S. (1988) J. Anal. At. Spectrom. 3:775.
Benson, A.A. and Summons, R.E., (1981) Science 211:482.
Bettley, F.R. and O'Shea, J.A., (1975) Br. J. Dermatol. 92:563.
Blust, R., Van der Linden, A., Verheyen, E. and Decleir, W.J. (1988) J. Anal. At. Spectrom. 3:387.
Bohn, A. (1975) Mar. Pollut. Bull. 6:87.
Bond, A.M. and Nagaosa, Y. (1985) Anal. Chim. Acta 178:197.
Borgono, J.M. and Greiber, R. (1972), Trace Sustances in Environmental Health-V (D. Hemphill, Ed.), 13-24 University of Missouri Press, Columbia, MO.
Borgono, J.M., Vincent P., Venturino, H. and Infante, A. (1977), Environ. Health Perspect. 19:103.
Boyer, I.W. and Horwitz, W. (1986) Environmental Carcinogens-Selected Methods of Analysis I.K. O'Neil, P. Schuller and L. Fishbein, Eds., International Agency for Research on Cancer, Lyon, France 8, 191.
Breyer, P.H. and Gilbert, B.P. (1987) Anal. Chim. Acta 201:23.
Brooke P.J. and Evans, W.H., (1981) Analyst (Lond.) 106:514.
Buchanan, W.D. (1962) Elsevier Monographs on Toxic Agents, Browning, E., Ed. Elsevier/North Holland, New York.
Burguera, M., Burguera, J.L. and Alarcon, O.M. (1986) Anal. Chim. Acta 179:351.
Calabrese, E.J., Kostecki, P.T. and Gilbert, C.E. (1987) Comments Toxicol. 1:229-242.
Cannon, J.R., Saunders, J.B. and Toia, R.F. (1983) Sci. Total Environ. 31:181.
Chakrabarti, C.L., Wan, C.C. and Li, W.C. (1979) Spectrochim. Acta 35B:547.
Chapman, A.C. (1926) Analyst (Lond.) 51:548.
Charbonneau, S.M., Spencer, K., Bryce, F., and Sandi, E. (1978a) Bull. Environ. Contam. Toxicol. 20:470.
Charbonneau, S.M., Tam, G.K.H., Bryce, F., and Collins, B. (1978b) Trace Subst. Environ. Health 12:276.
Chisela, F. and Bratter, P.(1986) Anal. Chim. Acta 188:85.
Cooney, R.V., Munna, R.O. and Benson, A.A. (1978) Proc. Natl. Acad. Sci. 75:4262.
Cooper, R.J., Langlois, D. and Olley, J. (1982) J. Appl. Toxicol. 2(2):99.
Coulsen, E.J., Remington, R.E. and Lynch, K.M. (1935) Nutr. 10:255.
Crecelius, E.A. (1977) Environ. Health Perspect. 19:147.
Cruz, R.B., Lorouso, C., George, S., Thomassen, Y., Kinrade, J.D., Butler, L.R.P., Lye, J. and Van Loon, J.C. (1980) Spectrochim. Acta 35B:775.
Cutter, G.A. (1985) Anal. Chem. 57:2951.
de Oliveira, E., McLaren, J.W. and Berman, S.S. (1983) Anal. Chem. 55:2047.
DeSilva, K.N. and Chatt, A. (1982) J. Trace Microprobe Tech. 1(3):307.
Dutkiewitz, T. (1977). Environ. Health Perspect. 19:173.
Edmonds, J.S. and Francesconi, K.A. (1981) Nature (Lond.) 289:602.
Edmonds, J.S., Francesconi, K.A., Cannon, J.R., Ratson, C.L., Skelton, B.W., and White, A.H. (1977) Tetrahedron. Lett. 18:1543.
Eisenberg, M. and Topping, J.J. (1984) J. Environ. Sci. Health Part B. 19(7):649.
EPA (U.S. Environmental Protection Agency) (1989) "Exposure Factors Handbook" EPA/600/8-89/043. p.3.2 to 3.8.
EPA (U.S. Environmental Protection Agency) (1979) Methods for Chemical Analysis of Water and Waste, EPA-600/47902, EPA, Cincinnati, OH 45268.
EPA (U.S. Environmental Protection Agency) (1991) Draft Fish Sampling and Analysis: A Guidance Document for Issuing Fish Advisories Washington, DC 20460.
Ershow, A.G. and Cantor, K.P (1989) Total water and tapwater intake in the United States: Population-based estimates of quantitities and sources. National Cancer Institute Order #263-MD-810264, Life Sciences Research Office, FASEB, 9650 Rockville Pike, Bethesda, MD 20814.
Favretto, L.G., Marletta, G.P. and Favretto, L. (1981) Recent Developments in Food Analysis: Proceedings of the First European Conference on Food Chemistry EURO FOOD CHEM I, Vienna, Austria, 372.
Fernando, L.A., Heavner, W.D. and Gabrielli, C.C. (1986) Anal. Chem. 58:511.
Flanjak, J. (1982) J. Sci. Food Agric. 33:579.
FDA (U.S. Food and Drug Administration) (1990) Inspection Operations Manual, Rockville, MD.
Gault, N.F.S., Tolland, E.L.C. and Parker, J.G. (1983) Mar. Biol. 77(3):307.
Grant, W.A. and Ellis, P.C. (1988) J. Anal. At. Spectrom. 3:815.
Greathouse, D.G. and Croun, G.F. In Trace Substances in Environmental Health-V, Proceedings of the 5th Annual Conference, University of Missouri, Columbia, MO. 12:31-39.
Guy, R.D. (1985) Anal. Chim. Acta 174:269.
Hall, R.A., Zook, E.G. and Meaburn, G.M. (March, 1978) National Marine Fisheries Service Survey of Trace Elements in the Fishery
Resource. NOOA Technical Report NMFS SSRF-721, U.S. Department of Commerce, National Oceanic and Atmospheric Administration, National Marine Fisheries Service.
Harms, U. (1985) Fresenius Z. Anal. Chem. 322:53.
Helrich, K., Ed. (1990) Official Methods of Analysis of the Association of Official Analytical Chemists, 15th Ed., Vol. I, Chap. 9, Method 972.23, AOAC, Arlington, VA.
Hill A.B. and Faning, E.L. (1948) Bar. J. Ind. Med. 5:1.
Holak, W. (1969) Anal. Chem. 41:1712.
Hu, T.-W. (1985) Analysis of Seafood Consumption in the U.S.: 1970, 1974, 1978, 1981. NTIS PB86-135043
Ikeda, M., Nishibe, J., Hamada, S. and Tujino, R. (1982) Anal. Chim. Acta. 125:109.
Ishinishi, N., Kodama, Y., Nobutomo, K., Inamasu, T., Kunitake, E, and Suenaga, Y. (1977) Environ. Health Perspect. 19:121.
Ishinishi, N., Tschiya, K., Vahter, M. and Fowler, B.A. (1986) Handbook on the Toxicology of Metals, second edition, Vol. II Friberg, L., Nordberg, G.F., and Vouk, V.B., Eds. Elsevier, New York, pp. 43-83.
Jones, J.W. (1988) Quantitative Trace Analysis of Biological Materials H. A. McKenzie and L. E. Smythe, Eds., Elsevier Science, New York p. 353.
Jongen, W.M.F., Carninaals, J.M., Bos, P.M.J. and Hagel, P. (1985) Food Chem. Toxicol. 23:669.
Julshamn, K., Ringdal, O., Slinning, K.E. and Braekkan, O.R. (1982) Spectrochim. Acta 37B(6):473.
Kingston, H.M. and Jassie, L.B. (1986) Anal. Chem. 58:2534.
Kingston, H.M. and Jassie, L.B. (1988) Introduction to Microwave Sample Preparation, ACS, Washington, DC.
LaTouche, Y.D., Bennett, C.W. and Mix, M.C. (1981) Bull. Environ. Contam. Toxicol. 26:224.
Le Blanc, P.F. and Jackson, A.L. (1973) Mar. Pollut. Bull. 4:88.
Lemly, A.D. (1982) Environ. Technol. Lett. 3:497.
Leonard, A. (1984) Toxicol. Environ. Chem. 7:241.
Leonard, A. and Lauwerys, R.R. (1980) Mutat. Res. 75:49.
Lunde, G. (1975) J. Sci. Food Agri. 26:1257.
Lunde, G. (1973) J. Sci. Food Agri. 24:1021.
Lunde, G. (1977) Environ. Health Perspect. 19:45.
Luten, J.B., Riekwel-Booy, G. and Rauchbaar, A. (1982) Environ. Health Perspect. 45:165.
Lytle, T.F. and Lytle, J.S. (1982) Bull. Environ. Contam. Toxicol. 29:50.
Mabuchi, K., Lilienfeld, A.M., and Snell, L.M. (1979) Arch. Environ. Health 34:312.
Madrid, Y., Bonilla, M. and Camara, C. (1988) J. Anal. At. Spectrom. 3:1097.
Maher, W.A. (1981) Anal. Chim. Acta 126:157.
Maher, W. (1982) Anal. Chim. Acta 138:365.
Maher, W. (1986) Anal. Lett. 19:295.
Maher, W.A., (1983). Mar. Pollut. Bull. 14:308.
Marafante, E., Vahter, M., and Dencker, L. (1984) Sci. Total Environ. 34:223.
Marcus, J.M. and Thompson, A.M. (1986) Bull. Environ. Contam. Toxicol. 36:587.
May, T.W. (1982) J. Assoc. Off. Anal. Chem. 65:1140.
May, T.W. and Brumbaugh, W.G. (1982) Anal. Chem. 54:1032.
May, T.W. and McKinney, G.L. (1981) Pestic. Monit. J. 15(1):14.
McCarthy, H. and Ellis, P.C. (1991) J. Assoc. Off. Anal. Chem. 74:566.
Mckie, J.C. (1987) Anal. Chim. Acta 187:303.
McMahon, J.W., Docherty, A.E., Judd, J.M.A. and Gentner, S-R. (1985) Int. J. Environ. Anal. Chem. 24:297.
McMahon, B.M. and Hardin, B.F. Eds. (1968) Pesticide Analytical Manual, Vol. I, 2nd Ed. and updates, U.S. Food and Drug Admin istration, Rockville, MD.
Meeus-Verdinne, K., Van Cauter, R. and De Borger, R. (1983) Mar. Pollut. Bull. 14(5):198.
Micallef, S. and Tyler, P.A. (1989) Bull. Environ. Contam. Toxicol. 42(3):344.
Michie, N.D., Dixon, E.J. and Bunton, N.G. (1978) J. Assoc. Off. Anal. Chem. 61:48.
Montaser, A. and Golightly, D.W. Eds. (1987) Inductively Coupled Plasmas in Analytical Atomic Spectrometry, VCH, New York, NY pp. 487-511.
MRCA (Market Research Corporation of America) (1988) 14-day survey (5-Year Menu Census, 1982-87) Northbrook, ILL.
Munro, I.C. (1976) Clin. Toxicol. 9:647
Munro, I.C., Charbonneau, S.M., Sandi E., Spencer, K., Bryce, F., and Grice, F. (1974) Toxicol. Appl. Pharmacol. 29:111.
Munro, S., Ebdon, L. and McWeeny, D.J. (1986) J. Anal. At. Spectrom. 1:211.
Nadkarni, R.A. (1984) Anal. Chem. 56:2233.
Nakamura, I., Arao, T., Inoue, K., Ono, T., Ishii, Y., Sato, T., Kawakami, K. and Kuwahara, H. (1973). Study of the Effect of Arsenic on Human Bodies, Part I. Japanese Public Association (Translated for EPA by SCITRAN, Santa Barbara, Calif. EPA translation No. TR-120).
NOAA (National Oceanic and Atmospheric Administration) (1989) A summary of data on tissue contamination from the first three years (1986-1988) of Mussel Watch Project NOAA Technical Memorandum NOS OMA 49, Rockville, Md.
Nriagu, J.O. and Simmons, M.S. Eds. (1990) Food Contamination from Environmental Sources, J. Wiley and Sons, New York, NY.
Panaro, K.W. and Krull, I.S. (1984) Anal. Lett. 17:157.
Pao, E.M., Fleming K.H., Guenther, P.M., et al. (1982) Foods commonly eaten by individuals: amount per day and per eating occasion. U.S. Department of Agriculture. Home Economics Report No. 44.
Park, C.J., Van Loon, J.C, Arrowsmith, P. and French, J.B. (1987) Anal. Chem. 59:2191.
Pennington, J.A.T. (1983). J. Am. Diet. Assoc. 82:166.
Penrose, W.R. (1974). CRC Crit. Rev. Environ. Control 4:465.
Penrose, W.R., Conacher, H.B.S., Black, R., Meranger, J.C., Miles, W., Cunningham, H.M. and Squires, W.R. (1977). Environ. Health Perspect. 19:53.
Pershagen, G. (1981) Environ. Health Perspect. 40:93.
Pinto, S.S. and McGill, C.M. (1953) Ind. Med. Surg. 22:281.
Poldoski, J.E. (1980) Anal. Chem. 52:1147.
Private Communication 1, Hans Kurner, Post FACH 171, D-8200 Rosenheim, Federal Republic of Germany
Private Communication 2, Commonwealth of Virginia, Department of General Services, Division of Consolidated Laboratory Services, Commonwealth of Virginia, 1 North 14th St., Richmond, VA 23219.
Private Communication 3, P. C. Ellis, Rhode Island Dept. of Health, Division of Laboratories, 50 Orms St., Providence, RI 02904.
Private Communication 4, J. Anderson, Bureau of Chemistry, Division of Consolidated Laboratories, Commonwealth of Virginia, 1 North 14th St., Richmond, VA 23219.
Private Communication 5, K. Wiles, Texas Department of Health, Shellfish Sanitation Branch, 1100 W. 49th Street, Austin, Texas 78756.
Private Communication 6, E. Feerst, State of New Jersey, Department of Environmental Protection, Marine Water Classification and Analysis, P.O. Box 405, Stoney Hill Road, Leeds Point, NJ 08220.
Private Communication 7, Letter from G. Lauenstein, NOAA Ocean Assessments Division, to I. Boyer, FDA, February 12, 1990.
Robbins, W.B. and Caruso, J.A. (1979) Anal Chem. 51:889A.
Schnitzer, G., Pellerin, C. and Clouet, C. (1988) Lab. Pract. 37(1):63.
Serra, T.M. and Serrano, J.F.B. (1984) J. Assoc. Off. Anal. Chem. 67(1):186.
Stephenson, M.D., Martin, M., Lange, S.E., Flegal, A.R. and Martin, J.H. (1979) California mussel watch 1977-78. Volume II: Trace metals concentrations in the California mussel, Mytilus californianus. State Water Resources Control Board, Water Quality Monitoring Report, No. 79-22, Sacramento, CA.
Stephenson, M.D. and Smith, D.R. (1988) Anal. Chem. 60:696.
Sturgeon, R.E., Willie, S.N. and Berman, S.S. (1986) J. Anal. At. Spectrom. 1:115.
Suta, B.E. (1980) Human Exposure to Atmospheric Arsenic, Final Report. Stanford Research Institute CRESS Rep. 50 EPA 68-01-4314 and 68-02-2835.
Tam, G.K.H., Charbonneau, S.M., Bryce, F. and Sandi, E. (1982). Bull. Environ. Contam. Toxicol. 28:669.
Tseng, W.-P. (1977) Environ. Health Perspect. 19:109.
Tseng, W.-P., Chu, H.M., How, S.W. (1968) J. Natl. Cancer. Inst. 40:453-463.
Tsuchiya, K. (1977) Environ. Health Perspect. 19:35.
Urakubo, G., Hasegawa, A. and Nakaura, S. (1975) J. Food Hyg. Soc. Jap. 16:334.
Watling, H.R. and Watling, R.J. (1982) Bull. Environ. Contam. Toxicol. 28:460.
Walsh, L.M. and Keeney, D.R. (1975) "Behavior and Phytotoxicity of Inorganic Arsenic in Soils" In Arsenical Pesticides, ACS Symposium Series No. 7. American Chemical Society Washington, DC.
Welch, A.D. and Landau, R.L. (1942) J. Biol. Chem. 144:581.
Welz, B. and Melcher, M. (1985) Anal. Chem. 57:427.
Welch, A.D. and Welch, M.S. (1938) Proc. Soc. Exp. Biol. Med. 39:7.
Welz, B. and Schlemmer, G. (1986) J. Anal. At. Spectrom. 1:119.
Westoo, G. and Rydlav, M. (1972) Var Foda. 24:21.
White, R.T. Jr. and Douthit, G.E. (1985) J. Assoc. Off. Anal. Chem. 68:766.
Willie, S.N., Sturgeon, R.E. and Berman, S.S. (1986) Anal. Chem. 58:1140.
WHO (World Health Organization) (1983) "Guidelines for the Study of Dietary Intakes of Chemical Contaminants" In Global Environmental Monitoring System, WHO, Geneva, Switzerland, pp. 49-50.
WHO/FAO (World Health Organization/Food and Agriculture Organization) (1989) "Evaluation of certain food additives and contaminants." Thirty third report of the joint WHO/FAO expert committee on food additives, Technical Report Series 776, World Health Organization.
Yeh, S. (1963) Natl. Cancer Inst. Monogr. 10:81-107.
Yeh, S. and How, S.W. (1963) Rep. Inst. Pathol. Natl. Taiwan Univ. 14:25-73.
Zeisler, R., Stone, S.F. and Sanders, R.W. (1988) Anal. Chem. 60:2760.
This Guidance document represented current agency thinking in regards to the available science at the time it was issued. It no longer represents the current state of science and is presented here for the historical record only.