ARSENIC
                                    r
                                    c

Agency for Toxic Substances and Disease Registry

U.S. Public Health Service
9-
e
3

-------
                                                       ATSDR/TP-88/02
           TOXICOLOGICAL PROnLE FOR
                     ARSENIC
            Date Published — March 1989
                     Prepared by:

                  Life Systems, Inc.
             under Contract No. 68-02-4228

                         for

Agency for Toxic Substances and Disease Registry (ATSDR)
               U.S. Public Health Service

                 in collaboration with

       U.S. Environmental Protection Agency (EPA)
        Technical editing/document preparation by:

             Oak Ridge National Laboratory
                        under
      DOE Interagency Agreement No. 1857-B026-A1

-------
                          DISCLAIMER

Mention of company name or product does not constitute endorsement by
the Agency for Toxic Substances and Disease Registry.

-------
                                FOREWORD

     The Superfund Amendments and Reauthorization Act of 1986 (Public
Law 99-499) extended and amended the Comprehensive Environmental
Response, Compensation, and Liability Act of 1980 (CERCLA or Superfund).
This public law (also known as SARA) directed the Agency for Toxic
Substances and Disease Registry (ATSDR) to prepare toxicological
profiles for hazardous substances which are most commonly found at
facilities on the CERCLA National Priorities List and which pose the
most significant potential threat to human health, as determined by
ATSDR and the Environmental Protection Agency (EPA). The list of the 100
most significant hazardous substances was published in the Federal
Register on April 17, 1987.
     Section 110 (3) of SARA directs the Administrator of ATSDR to
prepare a toxicological profile for each substance on the list. Each
profile must include the following content:
     "(A)  An examination, summary, and interpretation of available
     toxicological information and epidemiologic evaluations on a
     hazardous substance in order to ascertain the levels of significant
     human exposure  for the substance and the associated acute,
     subacute. and chronic health effects.
     (B)  A determination of whether adequate information on the health
     effects of each substance is available or in the process of
     development to  determine levels of exposure which present a
     significant risk  to human health of acute, subacute, and chronic
     health effects.
     (C)  Where appropriate, an  identification of toxicological testing
     needed to identify the  types or levels of exposure  that may present
    .significant risk  of adverse health effects  in humans."
     This  toxicological profile  is  prepared in accordance with
guidelines developed by ATSDR and  EPA. The guidelines were  published  in
the Federal Register on April 17,  1987. Each  profile will be revised  and
republished as necessary,  but no less  often than every  three years, as
required by SARA.
     The ATSDR  toxicological profile  is  intended to characterize
succinctly the  toxicological and health  effects  information for the
hazardous  substance  being described.  Each profile identifies and  reviews
the key literature  that  describes  a hazardous substance's  toxicological
properties. Other  literature is  presented but described in less detail
than the key  studies.  The profile  is not  intended to be an exhaustive
document;  however,  more  comprehensive sources of specialty information
are  referenced.
                                                                      ill

-------
 Foreword
      Each toxicological profile begins with a public health statement,
 which describes in nontechnical language a substance's relevant
 toxicological properties.  Following the statement  is material that
 presents levels of significant human exposure and,  where  known,
 significant health effects.  The adequacy of information to  determine a
 substance's health effects is described in a health effects summary.
 Research gaps in toxicologic and health effects  information are
 described in the profile.  Research gaps that are of significance  to
 protection of public health  will be identified by ATSDR,  the National
 Toxicology Program of the  Public Health Service, and EPA. The focus of
 the  profiles is on health  and toxicological  information;  therefore,  we
 have included this information in the front  of the  document.

      The principal audiences for the  toxicological  profiles  are health
 professionals at the federal,  state,  and local levels,  interested
 private  sector organizations and groups,  and members  of the  public.  We
 plan to  revise these documents in response  to public  comments and as
 additional data become  available;  therefore,  we  encourage comment that
 will make the toxicological  profile series of the greatest use.

      This profile  reflects our assessment of all relevant toxicological
 testing  and information that has  been peer reviewed.  It has  been
 reviewed by scientists  from  ATSDR, EPA,  the  Centers for Disease Control,
 and  the  National Toxicology  Program.  It  has  also been reviewed by a
 panel of nongovernment  peer  reviewers and was  made available  for  public
 review.  Final responsibility for  the  contents  and views expressed in
 this  toxicological profile resides with ATSDR.
                                    James 0. Mason, M.D.,  Dr. P.H.
                                    Assistant Surgeon General
                                    Administrator, ATSDR
iv

-------
                                CONTENTS

FOREWORD 	     ill

LIST OF FIGURES 	     ix

LIST OF TABLES 	    xi

 1.   PUBLIC HEALTH STATEMENT 	      1
     1.1  WHAT IS ARSENIC? 	     1
     1.2  HOW MIGHT I BE EXPOSED TO ARSENIC? 	     1
     1.3  HOW DOES ARSENIC GET INTO MY BODY? 	     2
     1.4  HOW CAN ARSENIC AFFECT MY HEALTH? 	     2
     1.5  IS THERE A MEDICAL TEST TO DETERMINE IF I HAVE BEEN
          EXPOSED TO ARSENIC? 	     3
     1.6  WHAT LEVELS OF EXPOSURE HAVE RESULTED IN HARMFUL
          HEALTH EFFECTS? 	     3
     1.7  WHAT RECOMMENDATIONS HAS THE FEDERAL GOVERNMENT
          MADE TO PROTECT HUMAN HEALTH? 	     7

 2.   HEALTH EFFECTS SUMMARY 	     9
     2.1  INTRODUCTION 	     9
     2.2  LEVELS OF SIGNIFICANT HUMAN EXPOSURE 	    10
          2.2.1  Key Studies 	    10
                 2.2.1.1  Oral 	     11
                 2.2.1.2  Inhalation	       17
                 2.2.1.3  Dermal  	    25
          2.2.2  Biological Monitoring as a Measure of
                 Exposure and Effects 	        25
                 2.2.2.1  Blood arsenic 	      25
                 2.2.2.2  Urinary arsenic  	      26
                 2.2.2.3  Hair and nails  	    26
                 2.2.2.4  Electromyography  	    26
          2.2.3  Environmental Levels as  Indicators of
                 Exposure and Effects 	    27
                 2.2.3.1  Levels  found  in the environment  	      27
                 2.2.3.2  Human exposure  potential  	      27
     2.3  ADEQUACY OF DATABASE  	      27
          2.3.1  Introduction  	        27
          2.3.2  Health Effect End Points 	      28
                 2.3.2.1  Introduction  and graphic  summary 	      28
                 2.3.2.2  Description of  highlights of  graphs  ...     31
                 2.3.2.3  Summary of relevant ongoing research        31
          2.3.3  Other Information Needed for Human
                 Health Assessment  	          31
                 2.3.3.1  Pharmacokinetics and mechanism of
                          action  	        31
                 2.3.3.2  Monitoring of human biological samples      33
                 2.3.3.3  Environmental considerations  	        33

-------
 Concents
      CHEMICAL AND PHYSICAL INFORMATION
      3.1
      3.2  PHYSICAL AND CHEMICAL PROPERTIES
                                                                       35
      CHEMICAL IDENTITY	     35
      TOXICOLOGICAL DATA 	
      4.1  OVERVIEW 	
      4.2  INORGANIC ARSENIC 	
           4.2.1   Overview 	
           4.2.2   Toxicoklnecics  	
                  4.2.2.1  Overview ...
                  4.2.2.2  Absorption  .
                  4.2.2.3  Distribution
                  4.2.2.4
                  4.2.2.5
     4.3
                     Metabolism  	
                     Excretion 	
     4.2.3  Toxicity 	
            4.2.3.1  Lethality 	
            4.2.3.2  Systemic/targe: organ toxicity
            4.2.3.3  Developmental coxicity 	
            4.2.3.4  Reproductive toxicity 	
            4.2.3.5  Genotoxicity 	
            4.2.3.6  Carcinogenicity 	
     4.2.4  Interactions with Other Chemicals 	
     ORGANIC ARSENIC 	
     4.3.1  Overview 	
     4.3.2  Toxicokinetics 	
            4.3.2.1  Methanearsonates 	
                     Phenylarsonates 	
                     Fish arsenic 	
                     Arsine and methylarsines 	
          4.3.3
4.3.2.2
4.3.2.3
4.3.2.4
Toxicity
4.3.3.1
                 4,
                 4.
                 4.
                 4.
                3.2
                3.3
                3.4
              3.3.5
Methanearsonates 	
Phenylarsonates 	
Fish arsenic 	
Arsine and methylarsines
Summary 	
     MANUFACTURE, IMPORT, USE, AND DISPOSAL
     5.1  OVERVIEW 	
     5.2  PRODUCTION 	
          IMPORT 	
          USES 	
          DISPOSAL 	
5.3
5.4
5.5
     ENVIRONMENTAL FATE 	
     6.1  OVERVIEW 	
     6.2  RELEASES TO THE ENVIRONMENT
          6.2.1  Anthropogenic 	
          6.2.2  Natural 	
     6 . 3  ENVIRONMENTAL FATE 	
          6.3.1  Atmosphere 	
          6.3.2  Surface Water 	
          6.3.3  Groundwater 	
          6.3.4  Soil 	
          6.3.5  Biota 	
 35
 41
 41
 41
 41
 42
 42
 42
 44
 47
 49
 50
 50
 50
 54
 54
 55
 56
 60
 61
 61
 62
 62
 62
 63
 63
 63
63
65
66
66
67

69
69
69
69
69
 70

71
71
71
 71
71
74
74
74
74
75
75
vi

-------
                                                               Cone en cs

 7 .   POTENTIAL FOR HUMAN EXPOSURE .........................         77
     7 . 1   OVERVIEW ...........................................       77
     7.2   LEVELS  MONITORED OR ESTIMATED  IN THE  ENVIRONMENT  .....   77
          7.2.1  Water .......................................    77
          7.2.2  Air  .............................................   78
          7.2.3  Soil  ...........................................   78
          7.2.4  Biota and Food ...............................      78
          7.2.5  Resulting Background Exposure  Levels   ..........   78
     7 . 3   OCCUPATIONAL EXPOSURES  .............................    81
     7.4   POPULATIONS  AT HIGH RISK .............................   81
          7.4.1  Above -Average Exposure  ........................     81
          7.4.2  Above-Average Sensitivity  .......................   82

 8 .   ANALYTICAL METHODS ................................ ' ...........   83
     8 . 1   ENVIRONMENTAL MEDIA ...................................    83
          8.1.1  Air  ..............................................   83
          8.1.2  Water ...........................................   83
          8.1.3  Soil  ............................................    85
          8.1.4  Food  .............................................   85
     8 . 2   BIOMEDICAL SAMPLES ....................................     85
          8.2.1  Fluids and Exudates  ............................   85
          8.2.2  Tissues ..........................................   85

 9 .   REGULATORY AND ADVISORY STATUS ...............................   87
     9 . 1   INTERNATIONAL ...........................................   87
     9 . 2   NATIONAL ................................................   87
          9.2.1  Regulations ......................................   87
                 9.2.1.1  Air ....................................    87
                 9.2.1.2  Water .................................     90
                 9.2.1.3  Reportable  quantities ..............        90
                 9.2.1.4  Waste disposal ......................       90
                 9.2.1.5  Pesticide ............................     91
          9.2.2  Advisory Guidance ............................      91
                 9.2.2.1  Air levels  ........................      91
                 9.2.2.2  Water levels ........................       91
          9.2.3  Data  Analysis ...................................    92
                 9.2.3.1  Reference dose ......................       92
                 9.2.3.2  Carcinogenic potency ..................    92
     9.3   STATE .................................................     93
          9.3.1  Regulations ....................................     93
          9.3.2  Advisory Guidance .............................      93

10 .   REFERENCES ................................................     95
11.  GLOSSARY ..................................................
APPENDIX:  PEER REVIEW ........................................    125
                                                                     VLl

-------
                            LIST OF FIGURES
1.1  Health effects from breathing inorganic arsenic 	    4
1.2  Health effects from ingesting inorganic arsenic 	    5
2.1  Effects of inorganic arsenic — oral exposure 	   13
2.2  Levels of significant exposure for inorganic arsenic — oral ...   14
2.3  Dose-response relationship for arsenic-induced skin cancer
     in humans 	   16
2.4  Effects of inorganic arsenic--inhalation exposure 	   19
2.5  Levels of significant exposure for inorganic arsenic--
     inhalation 	   20
2.6  Dose-response relationship for lung cancer in 	
     occupationally exposed workers 	   23
2.7  Availability of information on health effects of arsenic
     (human data)  	   29
2.8  Availability of information on health effects of arsenic
     (animal data) 	   3°
                                                                       ix

-------
                             LIST OF TABLES
2.1  Summary of key studies on oral toxicity of arsenic
     in humans 	   12
2.2  Summary of key human studies on exposure to
     airborne arsenic 	   18
2.3  Summary of lung cancer risk estimates 	   24
2.4  Summary of ongoing research 	    32
3.1  Chemical identity of arsenic and selected inorganic
     arsenic compounds 	   36
3.2  Chemical identity of arsine and selected organic
     arsenic compounds 	   37
3.3  Physical and chemical properties of arsenic and selected
     inorganic arsenic compounds 	   38
3.4  Physical and chemical properties of arsine and selected
     organic arsenic compounds 	   39
6.1  Arsenic releases to the environment in 1979 	   72
6.2  Estimates of arsenic emissions from  natural sources 	   73
7.1  Arsenic levels in foods 	   79
7.2  Summary of estimated levels of human exposure
     to arsenic 	   80
8.1  Analytical methods for arsenic in environmental samples 	   84
8.2  Analytical methods for arsenic in biological samples 	    86
9.1  Regulations and guidelines applicable to arsenic 	   88

-------
                      1.  PUBLIC HEALTH STATEMENT
1.1  WHAT IS ARSENIC?
     Arsenic is a naturally occurring element in the earth's crust.  Pure
arsenic is a gray-colored metal, but this form is not common in the
environment. Rather, arsenic is usually found combined with one or more
other elements such as oxygen, chlorine, and sulfur. Arsenic combined
with these elements is referred to as inorganic arsenic, whereas arsenic
combined with carbon and hydrogen is referred to as organic arsenic.
Many arsenic-containing substances, both inorganic and organic, are
naturally occurring, while others are man-made. It is important to
maintain a distinction between inorganic and organic arsenic, since  the
organic forms are usually less toxic than the inorganic forms.

1.2  HOW MIGHT I BE EXPOSED TO ARSENIC?

     Arsenic is very widely distributed in the environment, and all
humans are exposed to low levels of this element. For most people, food
constitutes the largest source of arsenic intake (about 25 to 50
micrograms per day - a microgram is one millionth of a gram), with lower
amounts coming from drinking water and air. Some edible fish and
shellfish contain elevated levels of arsenic, but this is predominantly
in an organic form ("fish arsenic") that has low toxicity. Above-average
levels of exposure are usually associated with one or more of the
following situations:

   • Natural mineral deposits in some geographic areas contain large
     quantities of arsenic, and this may result in elevated  levels of
     inorganic arsenic in water. If this water is used for drinking,
     high exposures may result.

   • Some waste-chemical disposal sites contain large quantities of
     arsenic, although the chemical form (inorganic or organic) is often
     unknown. If the material is not properly stored or contained at the
     site, arsenic may escape into the water, increasing  the chances
     that nearby residents might be exposed.

   • Elevated levels of arsenic in soil (due either to natural mineral
     deposits or to contamination from human activities)  may lead to
     exposure from ingesting soil. This is of particular  concern  for
     small children who swallow small amounts of soil while  playing

   • Manufacturing (smelting) of copper and other metals  often releases
     inorganic arsenic into the air. Thus, workers  in metal  smelters and
     nearby residents are exposed to elevated arsenic levels.

-------
 2   Section 1

    • Low levels of arsenic are found in most fossil fuels (oil. coal,
      gasoline, and wood), so burning of these materials (in power
      stations, furnaces, stoves, automobiles, etc.) results in low
      levels of inorganic arsenic emissions into the air.  There are also
      low levels of arsenic in cigarette smoke.

    • The main use of arsenic in this country is for pesticides. Some
      products, mostly weed killers, use organic arsenic as the active
      ingredient.  Other pesticides use inorganic forms of arsenic to kill
      plants,  insects, or rodents, or to preserve wood.  Persons who
      manufacture or use these pesticides or who handle  treated wood may
      be exposed to arsenic if adequate  safety procedures  are not
      followed. Widespread application of pesticides (e.g.,  in orchards
      and fields and along roadways) may lead to water or  soil
      contamination,  creating the possibility for more widespread
      exposure  of residents in the area.

    •  In the past,  inorganic  arsenic was contained in household products
      such as paints,  dyes, and rat poisons,  and in medicines for
      diseases  such as asthma and psoriasis.  However,  these products are
      no longer in general use;  therefore,  exposure from these sources is
      now unlikely.

 1.3   HOW DOES  ARSENIC GET INTO NT BODYT

      Arsenic enters  the body principally through the  mouth,  either in
 food  or in  water.  Most ingested arsenic  is  quickly absorbed  through the
 stomach and intestines and enters the blood stream, although this  varies
 somewhat for different chemical forms of arsenic.  Arsenic which is
 inhaled is  also well-absorbed through the  lungs  Into  the blood stream.
 Small amounts  of arsenic may  enter-the body through the skin,  but  this
 is not  usually an  important consideration.

     Most arsenic  that is absorbed  into  the body is converted by the
 liver to a  less-toxic  form that is  efficiently excreted in the urine.
 Consequently,  arsenic  does not  have a strong  tendency to accumulate  in
 the body, except at high exposure  levels.

 1.4  HOW CAN ARSENIC AFFECT NT  HEALTH?

     Inorganic arsenic  has been recognized  as a human poison since
 ancient  times, and large  doses  can  produce death.  Lower levels  of
 exposure may produce  injury in  a  number of different body tissues  or
 system: these are called "systemic" effects. When  taken by  mouth,  a
 common effect  is irritation of  the  digestive  tract, leading  to  pain.
 nausea, vomiting, and diarrhea. Other effects typical of exposure by
 mouth include decreased production  of red and white blood cells,
 abnormal heart function, blood vessel damage, liver and/or kidney
 injury, and impaired nerve function causing a "pins and needles" feeling
 in the feet and hands.  There  is evidence from animal studies that high
oral doses during pregnancy may be damaging to the fetus,  but  this has
not been well studied in humans.

     Perhaps the single most characteristic systemic effect of oral
exposure to inorganic arsenic is a pattern of skin abnormalities

-------
                                             Public Health Statement   3

Including the appearance of dark and light specs on the skin,  and small
"corns" on the pains,  soles, and trunk. While these skin changes are not
considered to be a health concern in their own right,  some of the corns
may ultimately progress to skin cancer. In addition, arsenic ingestion
has been reported to increase the risk of cancer inside the body,
especially in the liver, bladder, kidney, and lung.
     Inhalation exposure to inorganic arsenic dusts or fumes sometimes
produces the same types of systemic health effects produced by oral
exposure. However, this is not c6mmon, and the effects are usually mild.
Of much greater concern is the ability of inhaled arsenic to increase
the risk of lung cancer. This has been observed mostly in humans exposed
to high levels of airborne arsenic in or around smelters, but lower
levels may increase lung cancer risk as well.
     Direct dermal contact with arsenic compounds, frequently from
inorganic arsenic dusts in air. may result in mild to severe irritation
of the skin, eyes or and throat.
     Despite all  the adverse health effects associated with arsenic
exposure, there is some evidence that  low levels of exposure may be
beneficial to good health. Animals maintained on a diet with unusually
low concentrations of  arsenic did not  gain weight normally, and  they
became pregnant less frequently  than  animals maintained on a diet
containing a more normal  (but low) concentration of arsenic. Also,  the
offspring from these animals tended to be smaller  than normal, and  some
died at  an early  age.  The  estimated daily dose  of  arsenic that  is
beneficial is quite small  (about the  same as normally supplied  in the
diet), and no cases of arsenic deficiency in humans have  been found.

1.5  IS  THERE A MEDICAL TEST TO  DETERMINE IF  I  HAVE BEEN
     EXPOSED TO ARSENICT
     Several different ways exist  for testing people  for arsenic
exposure.  Measuring  the levels  of arsenic  in urine is the best way to
determine exposures that occurred  within the  last one to two days.
However, some  common  tests do not  distinguish nontoxic forms such as
 fish arsenic  from other forms,  so  a high concentration of arsenic in
urine  may not  necessarily indicate that a health problem exists.
Measurement of arsenic in hair or  fingernails is sometimes used to
detect chronic exposures, but this method is not very reliable for
 detecting low levels  of arsenic exposure.

 1.6.   WHAT LEVELS OP EXPOSURE HAVE RESULTED IN HARMFUL HEALTH EFFECTS?

      The amount of arsenic intake that is required to cause a harmful
 effect depends on the chemical and physical form of  the  arsenic. In
 general, inorganic forms of arsenic are more toxic than  organic forms.
 and forms that dissolve easily  in water (soluble forms of arsenic) tend
 to be more toxic than those that dissolve poorly  in  water. Also,
 toxicity depends somewhat  on the electric charge  (the oxidation state or
 valence) of the  arsenic.
      The graphs  on the  following pages  (Figs.  1.1 and 1.2) show the
 relationship between  exposure to soluble forms of inorganic arsenic and
 known health effects.  In the  first set  of graphs  labeled "Health effects

-------
     Sacelon  1
        SHORT-TERM EXPOSURE
    (LESS THAN OR EQUAL TO 14 DAYS)
          LONG-TERM EXPOSURE
         (GREATER THAN 14 DAYS)
EFFECTS
IN
ANIMALS
CONG IN
AIR
(ng/m3)
EFFECTS
IN
HUMANS
EFFECTS
IN
ANIMALS
CONC IN
AIR
(ng/m3)
EFFECTS
IN
HUMANS
ESTIMATED
DEATH LEVEL
INJURY TO	
FETUS
IMMUNE
SYSTEM
EFFECTS"
              100.000
                                                        100.000
               10.000
                                                        10.000
               1000
                                                         1000
IMMUNE
SYSTEM
EFFECTS
                100
                                                         100
                                                                  SKIN
                                                                  DISORDERS
                10
                                                          10
                1.0
                                                         1 0
               Flf. 1.1. Hetftfc effect! froa brcattaf taorfuk

-------
                                               Public  Wealth Scacemenc
    SHORT-TERM EXPOSURE
(LESS THAN OR EQUAL TO 14 DAYS)
                              LONG-TERM EXPOSURE
                             (GREATER THAN 14 DAYS)
EFFECTS
IN
ANIMALS

DOSE
(ng/kg/day)
EFFECTS
IN
HUMANS
EFFECTS
IN
ANIMALS

DOSE
(iig/kg/day)
EFFECTS
IN
HUMANS
 DEATH •<
         r  100.000
        ^  10.000
            1000
             100
                    DEATH
> SYSTEMIC
  EFFECTS
             10
             1.0
                                      NO OBVIOUS
                                      EFFECTS IN
                                      DOGS OR —
                                      MONKEYS
                                   100.000
                                    10.000
                                          BENEFICIAL
                                     1000
100
SYSTEMIC
EFFECTS/
SKIN LESIONS
                                      10
                                     1.0
              Fig. 1.2.  Health effects from ingesting inorganic arsenic.

-------
     Section 1
 from breathing inorganic arsenic," exposure is measured in raicrograras of
 arsenic per cubic meter of air Oig/m3).  In all graphs,  effects in
 animals are shown on the left side,  effects in humans on the right.

      In the second set of graphs,  the same relationship is represented
 for the known "Health effects from ingesting inorganic  arsenic."
 Exposures are measured in micrograms of  arsenic per kilogram of body
 weight per day (^g/kg/day) .

      As shown in the figures, most studies indicate that humans are  more
 sensitive to arsenic than animals, which means that studies in animals
 are of limited utility in predicting exposure  levels affecting humans.

      Studies in humans indicate  that there is  considerable variation
 among different individuals,  and it  is difficult to identify with
 certainty the exposure ranges of concern.  For  example,  some humans can
 ingest over 150 ^gAg/day without  any apparent ill-effects,  while more
 sensitive individuals in exposed populations often  begin to display  one
 or  more of the characteristic signs  of arsenic toxicity at oral doses  of
 around 20 ^gAg/day  (about 1000  to 1500  jig/day for  an adult).  Effects
 are usually mild at  this exposure  level, becoming more  severe  as  doses
 become higher.  Doses of 600  to 700 /igAg/day (around 50,000 ng/day in  an
 adult or 3,000 A»g/day in an  infant)  have caused death in some  cases.
 When exposure is  from contaminated water,  concentrations  of around 100
 to  200 micrograms per liter  (/*g/L) do not  seem to produce  significant
 noncancer health  risks,  while typical signs of arsenic  toxicity have
 been reported in  several populations  drinking  water  with  400 /jg/L of
 arsenic or more.  The levels of arsenic that most  people  ingest  in food
 or  water (around  50  j»g/day) are not  usually considered  to  be of health
 concern.
     For  inhalation exposure, air concentrations of around 200 Mg/m3 are
associated with  irritation to nose, throat and exposed skin, and higher
levels may occasionally lead to mild signs of systemic toxicity similar
to that seen with oral exposure.

     Direct skin contact with arsenic compounds can cause mild to severe
skin irritation, but no reliable dose estimates are available on the
exposure .levels at which these effects begin to appear.

     Because it is believed that cancer -causing agents can increase risk
even at very low exposures, Figs. 1.1 and 1.2 do not identify dose
ranges for skin cancer or lung cancer. From available data in humans.
the EPA has calculated that lifelong ingestion of 1 ng/kg/day (around 50
to 100 Mg/day in an adult) is associated with a risk of skin cancer of
about 0.1% (1/1000). This dose level is comparable to drinking water
containing 25 to 50 ^g/L for a lifetime.  Lifelong inhalation of air
containing 1 /*g/m3 is estimated by EPA to cause a lung cancer risk of
about 0.4% (4/1,000). Since there is considerable uncertainty in the
cancer risk assessment process,  quantitative estimates of cancer risk
such as these are intentionally conservative.  That is, the actual risks
of cancer could be lower,  but are unlikely to be higher.

-------
                                             Public Health Seacement   7

1.7  WHAT RECOMMENDATIONS HAS THE FEDERAL GOVERNMENT MADE TO
     PROTECT HUMAN HEALTH?
     The federal government has taken a number of steps to protect
humans from arsenic. The Environmental Protection Agency (EPA) has
established limits on the amount of arsenic that can be released into
the environment from factories which manufacture or use arsenic. EPA has
also restricted or cancelled many of the uses of arsenic in pesticides
and is considering further restrictions. The EPA has established a
Maximum Contaminated Level (MCL) of 50 ng/L for arsenic in drinking
water; this value is presently undergoing review by the Agency as part
of a rulemaking to establish a new MCL for arsenic. The Occupational
Safety and Health Administration (OSHA) has established a maximum
permissible airborne exposure limit of 10 A«g/m3 for inorganic arsenic
and 500 ^g/m3 for organic arsenic in various workplaces where arsenic is
used.

-------
                       2.  HEALTH EFFECTS SUMMARY
2.1  INTRODUCTION
     This section summarizes and graphs data on Che health effects
concerning exposure to arsenic. The purpose of this section is to
present levels of significant exposure for arsenic based on key
toxicological studies, epidemiological investigations, and environmental
exposure data. The information presented in this section is critically
evaluated and discussed in Sect. 4, Toxicological Data, and Sect. 7,
Potential for Human Exposure.

     This Health Effects Summary section comprises two major parts.
Levels of Significant Exposure (Sect. 2.2) presents brief narratives  and
graphics for key studies in a manner that provides public health
officials, physicians, and other interested individuals and groups with
(1) an overall perspective of the toxicology of arsenic and (2) a
summarized depiction of significant exposure levels associated with
various adverse health effects. This section also includes information
on the levels of arsenic that have been monitored in human fluids and
tissues and information about levels of arsenic found in environmental
media and their association with human exposures.
     The significance of the exposure levels shown on the graphs may
differ depending on the user's perspective. For example, physicians
concerned with the interpretation of overt clinical findings in exposed
persons or with the identification of persons with the potential to
develop such disease may be interested in levels of exposure associated
with frank effects (Frank Effect Level, FEL). Public health officials
and project managers concerned with response actions at Superfund sites
may want information on levels of exposure associated with more subtle
effects in humans or animals (Lowest-Observed-Adverse-Effect Level,
LOAEL) or exposure levels below which no adverse effects (No-Observed-
Adverse-Effect Level, NOAEL) have been observed. Estimates of  levels
posing minimal risk to humans  (Minimal Risk Levels) are of interest to
health professionals and citizens alike.
     Adequacy of Database (Sect. 2.3) highlights the availability of key
studies on exposure to arsenic in the scientific literature and displays
these data in three-dimensional graphs consistent with  the format in
Sect. 2.2. The purpose of this section is to suggest where there might
be insufficient information to establish levels of significant human
exposure. These areas will be considered by the Agency  for Toxic
Substances and Disease Registry (ATSDR), EPA, and the National
Toxicology Program (NTP) of the U.S. Public Health Service in  order co
develop a research agenda for arsenic.

-------
 10   Section 2

 2.2  LEVELS OF SIGNIFICANT HUMAN EXPOSURE

      To help public health professionals  address  the  needs  of persons
 living or working near hazardous waste  sites,  the  toxicology data
 summarized in this section are organized  first by  route  of  exposure--
 inhalation, ingestion, and dermal--and  then by toxicological end points
 that are categorized into six general areas — lethality,  systemic/target
 organ toxicity,  developmental toxicity, reproductive  toxicity,  genetic
 toxicity,  and careinogenieity.  The  data are discussed in terms  of three
 exposure periods--acute,  intermediate, and chronic.

      Two kinds of graphs  are  used to depict the data.  The first  type is
 a  "thermometer"  graph. It provides  a graphical  summary of the human and
 animal toxicological end  points (and levels of  exposure) for each
 exposure route for which  data are available.  The ordering of effects
 does not reflect the exposure duration or species of  animal  tested. The
 second kind of graph shows Levels of Significant Exposure (LSE)  for each
 route and exposure duration.  The  points on the  graph  showing NOAELs and
 LOAELs reflect the actual doses (levels of exposure)  used in the  key
 studies.  Where appropriate, adjustments for exposure  duration or
 intermittent exposure protocol  were made.

      Adjustments reflecting the uncertainty of extrapolating animal data
 to  man,  intraspecies variations,  and differences between experimental vs
 actual human exposure conditions  were considered when  estimates of
 levels posing minimal risks to  human health were made  for noncancer end
 points.  These minimal risk levels were derived  for the most  sensitive
 noncancer  end point  for each  exposure duration by applying uncertainty
 factors. These levels are shown on the graphs as a broken line starting
 from the actual  dose (level of  exposure) and ending with a concave -
 curved line at its terminus.  Although methods have been established to
 derive these  minimal risk levels  (Barnes et al. 1987), shortcomings
 exist  in  the  techniques that  reduce the confidence in  the projected
 estimates.  Also  shown on  the  graphs under the cancer end point are low-
 level  risks (10'4  to 10-7)  reported by EPA. In addition,  the actual dose
 (level  of  exposure)  associated  with the tumor incidence is plotted.

 2.2.1   Key Studies

      Investigation and analysis of the toxicity of arsenic are
 complicated by the finding that different chemical forms of  arsenic are
 not equally toxic. In particular, methylated derivatives and more
 complex organic  derivatives such as "fish arsenic" (arsenobetaine) are
 considerably  less  toxic than  inorganic arsenic. For example, oral LDso
values  in  animals  range from  10 to 300 mg/kg for inorganic arsenic
 compounds,  while LDso values  for  the monomethyl and dimethyl derivatives
 of arsenic  acid  range  from 600  to 2600 mgAg.  Arsenobetaine  is even less
 toxic,  producing no  clear  symptoms at doses of 10,000 mg/kg. For  this
 reason, this  section will  focus on studies that address the  toxicity of
 inorganic  arsenic. A summary  of available toxicity data on the most
 important  organic  arsenic  compounds is presented in Sect. 4.3.

     Among inorganic  arsenic  compounds,  those that are easily soluble in
water  tend to  have more acute toxicity than the poorly soluble ones,  and
arsenic(III)  compounds are generally observed to be somewhat more toxic
 than arsenic(V)  compounds. While  it is possible to investigate the

-------
                                             Health Effaces Summary   11

toxicity of individual arsenic compounds in animals, information on
arsenic valence and chemical form is often not available for human
studies, and exposure must be expressed in terms of total inorganic
arsenic.
     An additional difficulty is raised by the observation that most
studies indicate that animals are less sensitive to the toxic effects of
arsenic than are humans. For example, reported lethal doses in animals
(10 to 300 mg/kg) are significantly higher than lethal doses reported in
humans (0.6 to 2 mg/kg). Similarly, chronic oral exposure of humans to
inorganic arsenic at doses of 50 to 100 /ig/kg/day is frequently
associated with neurological or hematological signs of arsenic toxicity,
but no characteristic neurological or hematological signs of arsenism
were detected in young monkeys exposed to 3700 MgAg/day of arsenate for
one year (Heywood and Sortwell 1979), in dogs exposed to 3700 jig/kg/day
of arsenite or arsenate for two years (Byron et al. 1967), or in rats
given intraperitoneal injections of As203 at an average dose of 700
^g/kg/day for 18 months (Schaumberg 1980). Moreover, while there is good
evidence that arsenic is carcinogenic in humans by both the oral and
inhalation routes, evidence of arsenic -induced carcinogenicity in
animals is mostly negative. For these reasons, dose-response data from
animals are not judged to be reliable for determining levels of
significant human exposure and will not be considered except in the
absence of any human data.

2.2.1.1  Oral

     Many reports describe the symptoms and course of inorganic arsenic
poisoning in humans, but only a few of these studies provide sufficient
information on exposure levels to permit accurate estimates of the
ingested dose. These studies are summarized in Table 2.1, and the data
from this table are displayed graphically in Figs. 2.1 and 2.2. While no
one study by itself provides a full description of the dose -response
relationships, taken together, these studies do provide an adequate
basis for estimating NOAEL values  for the principal end points of
arsenic toxicity  in humans.
     Lethality.  Reports by Armstrong et al.  (1984) and Hamamoto  (1955)
indicate that acute or subacute lethality may occur at doses of around
600 MgAg/day or higher. Lethality has not been associated with doses of
100 MAg/day or  less.
     Systemic/target organ  toxic Lty.   Although  some humans  can  ingest up
to 150 pgAg/day without  apparent  ill-effects  (Bencko  1987).  a  number of
studies indicate that,  in more  sensitive  individuals,  doses as  low  as 20
to 60 pgAg/day  (roughly  1  to 4 mg/day in an adult) may produce one or
more of the characteristic  signs of arsenic  toxicity,  including gastro-
intestinal irritation,  anemia,  neuropathy, skin lesions,  vascular
lesions, and hepatic or renal injury (Tay and  Seah 1975,  Mizuta et  al
1956. Silver and Wainman  1952.  Huang et al.  1985,  Borgono et al.  1980,
Tseng et al. 1968). The severity of symptoms in affected individuals
generally tends  to increase as  a function of exposure  duration  (Tseng
1977), although  in some individuals,  effects may occur after relatively
brief exposure periods  (e.g., see  Mizuta et  al. 1956). In most  cases of
subchronic or chronic exposure, many or all  of the signs of arsenic

-------
                                                                       ToMrll  fMiry ol k«y nttjlka «n Mil loikfc. «l

Chcmicil
lorn.
"

A.O.
"*
AMIII)
*
AncnK
Sulfide

*

AMIII)






*

*
V
*
'
*

E.p«,,.
DDK
Source («|/k|/dr DimunB
Wtui 400-1000 1 2 rat*
(lOOnf/D

Milk po»dei 700 11 dan
IIS liO|/k|)
Wiicr 60 140 Chronic
(06a|/L)
Fo»kf-| JO- 100 Svbchronic
Saluiun
W.ier 6010 2-llyun
(06-Olmt/L)
Madicia.li )) Snbchfouc

Soy Suet M 2 20 dayi

»»_i>i t JO 1 1 21 rnonlhe


Wiiei 40 II ycnn
(0601/L)
W.ier 10-40 Chronic
(OI-04UI/L)
Wiicr 20-10 Chronic
(04-06 OH/L)
W.ur 20 Ckronic
(01-1 IBI/L)
Wiier 20 Ckronic
(OJ Of/LI
Wnla 20 Chrome
(04n|/L)
Wua 10 Chronic
(0001-0 4 a|/L)
Wua 6 Chronic
(02B|/U
Witti 6 Chronic
(02ai|/U
Witer 0 7 Cnranic
(0-002 ni/U
Wild 0 2 Chronic
(000) of /I)
Noacuanofcnic u|ni of loucily obtcmd C.ruaofcnic effect.
Gl Hcaulalaiic NCWIIU/ Skin HCD.IK/ CVS Skin Oihcr
UlhiUl, diurtu .boonnilun oc.rop.lhy la.,., leM, t^,, CMlccf lumorl Slu(1, |l
* * + * Rciidcnu of
Aniul.|iUi thik
+ »•»%) +(21/262) 262Adulu
*'12*' 2 7.000 children
+ (21%) +(2J%) +01%) +(97%) +(19%) +(6/74) +(4/74) 74 .dull,

+ (70%) +(!)%) +(]%) +(20%) +(60%) +(10%) 220»dy|ii

+ + + A
* 1 piiicnl (uihnu)

+ (12%) +(16%) +(4S%) 116«Kkcr.ui
pUni
+ 129%) looo
icnool children
+ «!•%) +(09%) +(428/40.421) 40000 Tii.inac
+(SMR-)M-6)2C +(SMR- 170 2009) 140.000 Tii.uac
+ (AOR-27-19)« 279cuc,indl6»
oonuob
+ <"«») +(4/296) WOrcudcouof
Mcucu 'UUfc
Nff NS "S Groupi of l)io)2
in 6 U S cilia
NS "S NS NS »0 Mormon.
NS NS NS 211 Alukini
NS NS NS 7)00 Tniouac
NS 300 raidcnu of
Hciiun »ill>|c
	 	
Reference*
Arnuironi
ci .1 1914
ll.ouaialo 19))
/.Idivir 1974
ricn 196)
Bor|.no tnd
Greiber 1972
l.tiodSuh
197)
Milul. cl .1
19)6

Silver .nd
Wnnm.n 19)2
Hu.nf
ci •! 191)
Bm|.no cl .1
1910
1961
Chen ci .1
191)
Chen cl •!
1916
Ccbrun cl il
191)
Vikniinc ci nl
191)
Souihxick a ml
1911
Huiinflaa
cl .4 1971
Ticnf ci .1
1961
Ccbrun cl .1
191)
N
to
n
n
























•U ma reported Mbcrvw by .ylhar or by U>A IM7b cikvliled by uioninf 2 L/day ol »«lei lor . 70-k| odull or I  I /diy lor i lO-kf child
1 hen**! lam unknown twi prawmod la be inar|inn .nenic
SMN - tund.rdind non.liiy i.liu
           |nifit«ally diflcrcol Iron coalrol or cipctlcd

-------
                                                          Health  Effects  Summary    13
  ANIMALS
100.000 i- O MOUSE. DEVELOPMENTAL (As V)
        • MOUSE. HAMSTER DEVELOPMENTAL (A* 111)
        O MOUSE. HAMSTER. DEVELOPMENTAL (A* III)
 10.000 -
  1000
   100
        O DOG. MONKEY. ANEMM. NEUROPATHY. CHRONIC
        O MOUSE. REPRODUCTION. 3 GENERATIONS
     1 I—
                                                           HUMANS
1 00.000 1-
                                                         10.000
                                                          1000
                                                           100
                                                            10
                                                                  LETHALITY (ADULTS)
                                                                A LETHALITY (INFANTS)
                                                                A ANEMM. SUBCHRONIC
                                                                A Gl DISTRESS. NEUROPATHY CHRONIC
                                                                A SKIN. VASCULAR LESIONS CHRONIC
                                                               - A SKIN. BLOOD. NERVE
                   • LOAEL FOR ANIMALS   A LOAEL FOR HUMANS
                   ONOAEL FOR ANIMALS   A NOAEL FOR HUMANS
                     Fig. 2.1.  Effects of inorganic arsenic—oral exposure.

-------
14    Section 2
                   ACUTE
                 (SI 4 DAYS)
                    DEVELOP-  TARGET
         LETHALITY   MENTAL   ORGAN
(ng/Vg/day)
 100.000
  10.000
[I
                                     INTERMEDIATE
                                     (15-364 DAYS)

                                   TARGET   REPRO-
                                   ORGAN   DUCTION
                                               CHRONIC
                                              (2365 DAYS)

                                            TARGET
                                            ORGAN   CANCER
             r. m. h
O m(As V)
• h. m (As III)
O h. m (As III)
   1000
                                                    O m (3 GENERATIONS)
    100
    10
    10
                        I
             (Gl. NERVE)   A (Gl. NERVE
                        I BLOOD)

                        t
                        I
                        I
                       vis
I
                                                                  I
                                                                 vlx
(Gl. NERVE
BLOOD, SKIN)

     A (SKIN
       CANCER)
    01
   001
  0001
 00001  l-
            r  RAT
            m MOUSE
            h  HAMSTER
                   • LOAEL FOR ANIMALS
                   O NOAEL FOR ANIMALS

                   A LOAEL FOR HUMANS
                   A NOAEL FOR HUMANS
                            I   MINIMAL RISK
                            '   LEVEL FOR
                            J   EFFECTS OTHER
                            Ox  THAN CANCER
                                                                  10~4 —
                                                                  10~5  —
                                                                  10-6  —
                                                                  io-7—I
                                                             ESTIMATED
                                                             UPPER-BOUND
                                                             HUMAN
                                                             CANCER
                                                             RISK LEVELS
              Fig. 2.2. Levcb of significant exposure for ioorguk arsenic—oraL

-------
                                             Wealth Effects Summary   15

coxiclty are detected together, indicating that the dose-response
relationships for the various systemic end points are fairly similar.  If
any one effect is most sensitive, it is probably the appearance of the
characteristic pigmentation pattern and hyperkeratotic lesions. Although
it is difficult to identify a no-effect level with certainty, doses of
around 10 jjgAg/day (about 0.7 mg/day in an adult) are not generally
expected to cause measurable signs of arsenic intoxication (e~.g.,
Valentine et al. 1985, Southwick et al. 1981).
     Developmental toxicIty.  Studies in animals have revealed that very
high oral doses of sodium arsenate (in excess of 100 mg/kg) may be
teratogenic and fetotoxic, while oral doses of 60 to 100 mg/kg/day have
no significant effect (Hood et al. 1977, 1978). Sodium arsenite appears
to be somewhat more toxic, causing increased malformations and prenatal
mortality in mice and hamsters dosed by gavage at levels of 25 to 40
mg/kg/day (Baxley et al. 1981. Hood and Harrison 1982. Willhite and Ferm
1984). These researchers observed no effect at 20 mgAg/day. It should
be noted that these dose levels may cause maternal lethality in exposed
animals and are considerably higher than levels which may cause
lethality in humans (0.6 mg/kg/day). On this basis, it seems likely that
developmental end points are not of primary concern at exposure levels
lower than those which cause maternal toxicity.
     Reproductive toxicity.  The effects of arsenic exposure on
reproductive parameters have not been well studied. Limited data in mice
suggest that ingestion of water containing 5 mg/L of arsenite  (about 0 7
mg/kg/day) for three generations does not significantly impair
reproductive success (Schroeder and Mitenner 1971).
     Genotoxicity.  The genotoxicity of various inorganic arsenic
compounds has been investigated in a number of prokaryotic and
eukaryotic test systems. Although data are not entirely consistent, che
weight of evidence indicates that arsenic is clastogenic  (i.e., causes
chromosomal breaks and aberrations) and induces sister chromatid
exchange (SCE) in cultured mammalian cells (Jacobson-Kram and  Montalbano
1985).  Trlvalent arsenic compounds (NaAs02. As203, AsCl3, and NaAs02)
tend to be more potent in causing chromosomal aberrations than
pentavalent compounds (Na2HAs04, H3As04, and As205) (Nakamuro  and  Sayaco
1981). Several studies have reported increased frequencies of  SCEs and
chromosome aberrations in lymphocytes  from exposed humans, but these
studies have limitations and must be interpreted with caution  (EPA
1984a). Despite these cytogenetic effects, arsenic appears  to  be either
inactive or extremely weak  for the  induction of specific  gene  mutations
in vivo.
     CarcinogenicIty.  The  study  of Tseng et al.  (1968) generally
provides the best available description of the dose-response
relationship for skin cancer.  In  this  study,  individuals  were  classified
into one of three exposure  groups on the basis of  the concentration of
arsenic in their drinking water:  low - 0 to 0.29 mg/L; medium -  0  3 co
0.59 mg/L; high - 0.6 mg/L  or  more. EPA  (1987a) estimated that the
average daily  intake of  arsenic by  these groups was  10.8,  29.9.  and 50 9
MgAg/day  for  males and  6,  8,  18.8, and  32.0 pgAg/day  for  females.
respectively.  As shown  in  Fig. 2.3. skin cancer prevalence  rates in
these  groups were proportional to arsenic  exposure level.  This study  Is

-------
16   Section 2
    240
~  210
o
o
UJ
(0
<
o

UJ


oc
UJ
o
z
UJ
Ul
tr
0.
    150
    120
     90
     60
     30
                  0 TO 0.29       0.30 TO 0.59    0.60 OR MORE

                   DRINKING WATER CONCENTRATION (mg/L)
   Fig. 2.3. Dwe-respoue relationship for weak-induced skin cancer in

 from Tseng et al. 1968.
                                                           Source: Adapted

-------
                                             Health Effects Summary   17

consistent: with a number of other studies (see Table 2 1) that detected
increased frequency of skin cancer and/or internal cancer in individuals
exposed to water containing 0.3 mg/L or more. Failure to detect
significant increases at lower doses (e.g., Harrington et al. 1978,
Southwick et al. 1981) may be due to a lack of statistical power in che
studies (Andelman and Barnett 1983), or it could suggest that arsenic-
induced cancers have a threshold dose.  Although toxicokinetic data
regarding arsenic methylation (see Sect. 4.2.2 on toxicokinetics)
provide some support for this concept, the EPA has judged that the
evidence is not adequate at present to conclude that arsenic-induced
cancer has a nonzero threshold (EPA 1987b).

     Based on the data of Tseng, EPA (1984a) originally calculated that
a dose of 1 ^ig/kg/day corresponded to a skin cancer risk of 1.58 x 10"2
More recently, EPA (1987b) has refined these calculations, having
estimated that a dose of 1 /ig/kg/day corresponds to a risk of 1.5 x 10 "^

     As with any cancer risk calculation, there is considerable
uncertainty in this value, especially when extrapolated to very low
exposure levels. In addition, there are a number of limitations to che
Tseng study that may introduce uncertainty concerning the applicability
of this risk estimate to the U.S. population. These uncertainties
include (1) possible exposure of the Taiwanese subjects to arsenic from
sources other than drinking water, (2) an above-average death rate in
the exposed population from Blackfoot disease, (3) differences in diet
between the Taiwanese and U.S. populations, (4) exposure to other
chemicals besides arsenic, (5) lack of blinding in the researchers who
collected data on exposure levels and symptoms in the study populations.
and (6) use of prevalence rates to estimate cumulative cancer incidence
rates (EPA 1987b).  Although these uncertainties may cast some doubt on
the precise quantitative value calculated for the oral cancer potency
factor for arsenic, they do not challenge the conclusion that arsenic
ingestion does increase the risk of skin cancer.
     None of the available studies provide adequate dose-response daca
to calculate the risk of internal cancers following exposure to arsenic
(EPA 1987b).

2.2.1.2  Inhalation
     Host information on human exposure to arsenic dusts and fumes is
derived from occupational settings such as smelters and chemical planes
It should be noted that significant oral and dermal exposures are  likely
under these conditions, and that exposure to other metals and chemicals
is also common. Table 2.2 summarizes available reports on health effects
in humans exposed to airborne arsenic. Studies with quantitative
estimates of exposure are shown graphically  in Figs. 2.4 and 2 5

     Lethality.  Inhalation exposure is not usually associated normally
with acute lethality in animals or humans. Webb et al.  (1986) reported
that the maximally tolerated nonlethal dose of As203 given by
intratracheal instillation to rats was 17 mg/kg.  If a rat  inhales  about
1 mVkg/day (Guyton 1947) . and approximately 40%  of the airborne arsenic
is deposited in the lung  (Holland et al.  1959), this would correspond  co
an air concentration of over 40 rag/m^.

-------
   18     Section  2
                       TaMe 2.2.  Summary of key human undies on exposure to airborne aneaic
Study group
Copper smelter
workers


1 276 smelter
workers
348 smelter
workers

Sodium anenite
factory
Chem workers
Mamt./packers
Controls
Children near
gold smelter
1 1 smelter
workers

Women working
in or living
near smelter
8000 white male
smelter workers
1800 male
smelter workers
2800 male
smelter workers
1900 pesticide
plant workers
2802 male
smelter worken
Airborne
exposure data
High variable;
very high
(over 50 mg/mj)
in some locations
Up to 7 mg/m1
02'




04- 1.0 mg/m3
0.06-0. 16 mg/m1
Not reported
None
None


None


3000-36.000
Mg/m'-years
250-16.000
wg/m'-yean
90-4000
pg/m'-yean*
40-19.500
Mg/m'-years
<750-45,000
Mg/m'-years
Results
Dermatitis, mostly on exposed areas of skin; more
common in work areas of high exposure


Rhinopharyngolaryngitis. tracheobronchitis
One possible case of systemic effects: no
hyperpigmentation; primary observation was
dermatitis, conjunctivitis, pharyngitis
HyperkeraUMCs riiiiniillini

9/31 (29%) 28/31 (90%)
1/32(3%) 12/32(37%)
2/56(4%) 10/56(18%)
Dermatosea. mostly on face and neck
Preclinical neurological effects detected by
EMC; correlation with arsenic levels in hair and
urine
Increased spontaneous abortion rate, decreased
birth weight

Exposure-dependent increase in lung cancer
mortality (overall SMR - 285)
Exposure-dependent increase in lung cancer
mortality (high dose SMR - 548)
Exposure-dependent increase in lung cancer
mortality (overall SMR - 189)
— . . .
cxposuremepenoent increase in lung cancer
mortality (high dose SMR - 694)
Exposure-dependent increase (nonlinear) in lung
cancer mortality (high dote SMR - 477)
References
Holmqvist 19SI



Lundgren >54
Pinto and
McGdl 1953

Perry et al
1948



Birmingham
et al 1965
Landau et al
1977

Nordstrom et al
I978a.b.c,d

Lee-Feldstein
1983
Higgins et al
1982
Enterlme and
Marsh (1982)
On et al (1974)
Enterlme
etal 1987
'ACGIH (1986); bated on urinary arsenic levels.
'Estimated from urinary arsenic levels.

-------
  ANIMALS
100 000
 10000
  1000
   100
    10
   101-
           I RAT (MAXIMUM TOLERATED DOSE ACUTE

           I MOUSE. FETOTOXICITY GESTATION DAYS 9-19
          O MOUSE. FETOTOXICITY GESTATION DAYS 9-12
          • MOUSE IMMUNOTOXICITY 1-20 DAYS

          O MOUSE IMMUNOTOXICITY 1-20 DAYS
          • LOAEL FOR ANIMALS   O NOAEL FOR ANIMALS
          A LOAEL FOR HUMANS
                                                              Health  Effects  Summary    19
                                                                    HUMANS
                                                                  100 000 1—
                                                                   10000 -
                                                                    1000 -
                                                                     100
                                                                      10
                                                                     10
  MILD SKIN DISORDERS
  (HYPERPIGMENTATION
A HYPERKERATOSE5)

A DIRECT DERMAL  IRRITATION
                   Fig. 2.4.  Effects of inorganic arsenic—inhalation exposure.

-------
  20    Section  2
                   ACUTE
                 (SI 4 DAYS)
 (WJrtn3)

 100.000



  10.000
                    DEVELOP-   TARGET
         LETHALITY   MENTAL    ORGAN
• r
                                INTERMEDIATE
                                (15-364 DAYS)

                                   TARGET
                                   ORGAN
                                       CHRONIC
                                      (2365 DAYS)

                                    TARGET
                                    ORGAN   CANCER
   1000
    100
     10
    10
         I m (FETOTOXICITY)
m (IMMUNOTOXICITY)  m (IMMUNOTOXICITY)

    I                I
                                                     I

                                                    &
                                                                (LUNG CANCER.
                                                              A 5 YEARS)

                                                              A (LUNG CANCER.
                                                                20 YEARS)
    01
   001
  0001
 00001
0 00001
             r  RAT       • LOAEL FOR ANIMALS
             m MOUSE     O NOAEL FOR ANIMALS
                                    MINIMAL RISK
                                    LEVEL FOR
                                 I   EFFECTS OTHER
             A LOAEL FOR HUMANS  six  THAN CANCER
             A NOAEL FOR HUMANS
                                             io-7j
                                                                    ESTIMATED
                                                                    UPPER-BOUND
                                                                    HUMAN
                                                                    CANCER
                                                                    RISK LEVELS
             F1f.2J. Letcb of rignifkaat expoon for horfufc

-------
                                             Health Effects Summary   21

     Systemic/target organ effecta.  There is a general agreement among
studies of occupationally exposed workers that signs of systemic
toxicity (nausea, neuropathy, and hyperkeratoses) are infrequent and
generally mild. Rather, the primary noncarcinogenic health risk
associated with exposure to airborne arsenic is irritation of the skin
and mucous membranes (dermatitis, conjunctivitis, pharyngitis, and
rhinitis) (Pinto and McGill 1953, Vallee et al. 1960, Stokinger 1981.
ACGIH 1986).
     Very few studies provide quantitative information on the
concentrations of arsenic in air that may cause dermatitis or systemic
effects. The data of Holmqvist (1951) and Lundgren (1954) reveal dermal
and mucosal irritation in workers who have been exposed to high levels
of arsenic Ln air, but they do not identify a threshold for these
effects. Perry et al. (1948) reported that chemical workers exposed to
about 0.4 to 1.0 mg/m3 inorganic arsenic in air manifested gross
pigmentation and an elevated frequency of hyperkeratinization, whereas
maintenance workers and packagers with a lower level of exposure
(estimated to be about 0.06 to 0.16 mg/m3 of air) displayed only a
slight tendency toward mild hyperpigmentation. Pinto and McGill (1953)
reported that dermatitis was common in smelter workers whose urinary
arsenic levels exceeded 0.4 mg/L, a value which probably corresponds to
an airborne exposure level of about 0.1 to 0.2 mg/m3 (Pinto et al. 1976,
ACGIH 1986). No cases of hyperpigmentation or hyperkeratoses were
observed in these workers. Based on these limited observations, it
appears that chronic exposure to concentrations of 0.4 mg/m3 or higher
in workplace air may result in characteristic systemic effects in skin,
and that exposure to levels of 0.1 to 0.2 mg/m3 may cause direct dermal
irritation. Levels below 0.1 mg/m3 do not appear likely to cause
significant noncarcinogenic effects. Because these data are derived from
the workplace where exposure is for 8 h/day 240 days/year, these values
must be divided by a factor of 4.56 to yield values appropriate for
continuous exposure (EPA 1984a).
     Aranyi et al. (1985) have investigated the effect of inhalation
exposure on pulmonary antibacterial defenses in mice. Animals were
exposed to aerosols containing As203, 3 h/day  for 1 to 20 days.
Concentrations of 270 Mg/»3 arsenic or higher  resulted in significantly
increased mortality following Infection with active pneumonia-causing
bacteria, while concentrations of  125 Mg/m3 or lower had no significant
effect. This action of arsenic did not appear  to be cumulative, since
the sane response was  seen  following  20 days of  exposure as after  1 day
of exposure.
     Developmental toxicity.  There  are very  few reports dealing with
developmental  effects  of  arsenic by  the  inhalation  route. Limited  data
suggest  that women who work in  or  live nearby  smelters may have higher
than normal abortion  rates  and  lower  than normal birth weights.
Nordstrom et al.  (1978c)  reported  that  the prevalence  rate of congenital.
birth defects  was about  the same  (3.0%)  in Swedish  women working  in a
smelter  as  in  women  living  in the  same  area but  not working  in the
smelter. However, there  was a decrease  in the  average  birth  weight and
an  increase  In the  incidence of spontaneous abortions  in the female
workers, with  the highest rate  (17%)  in women who had  been employed
during  their pregnancy (Nordstrom  et al.  1978a,b).  The incidence  of

-------
22   Section 2

spontaneous abortion  in women  living near the plant was also higher than
average, although many of  the  women who lived nearby also worked in the
smelter  (Nordstrom et al.  1978d). These observations should be
interpreted with caution,  because these studies were designed to
evaluate the effects of smelter pollutants in general. This prevents
making conclusive statements about the specific effects of arsenic from
these findings. NagymaJtenyi et al. (1985) reported that inhalation
exposure of pregnant mice  to 28.5 mg/m3 of As2<>3 (4 h/day on days 9 Co
12 of gestation) caused fetotoxic effects, while concentrations of 2.9
or 0.26 mg/m3 caused no changes except for a slight decrease in fetal
weight (9.9% and 3.1%, respectively).

     Reproductive toxicity.  No information was located on the
reproductive effects of inhalation exposure to arsenic in animals or
humans.

     Genotoxicity.  As described above for oral exposure,  the weight of
evidence indicates that arsenic induces SCEs and chromosome aberrations,
but does not induce detectable gene mutations.

     Carcinogenicity.  Many studies report above-average lung cancer
rates in groups of people with above-average exposure to airborne
arsenic, but only a few contain adequate quantitative data on exposure
levels and durations to permit derivation of dose-response
relationships. EPA (1981,  1984a) has reviewed the available data and
calculated dose-response curves for cohorts exposed at the Anaconda
smelter in Montana (Lee and Fraumeni 1969; Lee-Feldstein 1983;  Higgins
et al. 1982; Brown and Chu 1983a,b),  the ASARCO smelter in Tacoma,
Washington (Enterline and Harsh 1982. Pinto et al.  1977),  and a
pesticide plant (Ott et al. 1974).  In most cases,  cumulative exposure
(the time-weighted average of air concentration times years of exposure
expressed as Mg/m3-years) was taken to be the most  appropriate index of
exposure, although in some cases (e.g., Enterline and Marsh 1982,  Welch
et al. 1982),  exposure duration did not appear to be as important as
exposure level. The calculations of exposure are quite complex in some
cases, and the interested reader is referred to the EPA documents (1981,
1984a) for a detailed description. Figure 2.6 shows one typical data set
(Lee-Fel'dstein 1983), which illustrates the approximately  linear
increase in relative risk  (the frequency of lung cancer in the exposed
group divided by the frequency of lung cancer in the control group) as a
function of increasing exposure. Relative risk was  significantly
increased even at the lowest cumulative doses (290  jig/m3 for an average
of 20 years or 580 Mg/m3 for an average of 5.3 years) (EPA 1987a).
Adjusted for intermittent worker exposure (8 h/day, 240 days/year),
these concentrations correspond to continuous exposure levels of 64 and
127 pg/n3. respectively. Based on these analyses, EPA (1984a) concluded
that the most reliable estimates of the dose-response curve were
provided by the reports of Lee-Feldstein (1983), Higgins et al.  (1982).
Brown and Chu (1983a,b), and Enterline and Marsh (1982). Table 2.3
presents the estimates of unit risk (the increased  risk of lung cancer
associated with lifetime exposure to 1 pg/m3) derived from these
studies. By calculating the geometric mean of these estimates.  EPA
(1984a) derived an overall unit risk of 4.3 x 10*3.

-------
                                                      Wealth  Effects Summary   23
CO

oc

UJ
UJ
cc
                         CUMULATIVE DOSE (mQ/m  -YEARS)
    Fig. 2.6. Dose-response relationship for lung cancer in occupationally exposed workers. Source

  Adapted from Lee-Feldstein 1983.

-------
24    Section  2
                    Table 2.3.  Summary of lung cancer risk estimates
                             Study                Unit risk"
                    Lee-Feldstem 1983          2.8 X 10~3
                    Higgmsetal. 1982          49 X I0~]
                    Brown and Chu  1983a,b     1.2 X 10~3
                    Enterline and Marsh 1982   72 X 10~3

                        "Unit risk is the increased risk of cancer
                    associated with lifetime exposure to 1 Mg/mJ.
                        Adapted from EPA 1984a.

-------
                                             Health Effects Summary   25

     The Occupational Safety and Health Administration (OSHA) also
conducted a detailed risk assessment for lung cancer from inhalation of
arsenic (OSHA 1983). After review of the epidemiological data of Lee and
Frauoeni (1969). Pinto et al. (1977), Ott et al. (1974), Hill and
Fanning (1948), Lee-Feldstein (1983), Enterline and Marsh (1982),
Hlggins et al.  (1982), Mabuchi et al. (1979) and Lubin et al  (1981),
OSHA concluded  that the risk of lung cancer from a working lifetime of
exposure to inorganic arsenic at an exposure level of 10 Aig/m-* ranged
from 2.2 to 29  excess deaths per 1000 exposed employees, with the
preferred estimate being 8 deaths/1000 employees.
     More recently, Enterline et al. (1987) reexamlned the dose-response
relationship between inhalation exposure to arsenic and risk of lung
cancer, using historical records of airborne arsenic levels in the
smelters, along with records of urinary arsenic levels in exposed
workers. The researchers concluded that arsenic Is a more potent lung
carcinogen then previously believed, with a dose-response relationship
concaved downward at exposure levels below 10,000 Mg/nVyear. In
contrast, the relationship between lung cancer and urinary arsenic
levels was linear, suggesting that bloavailability and lung absorption
of arsenic tend to be proportionately greater at low exposure levels
than at high exposure levels.

2.2.1.3  Dermal
     As discussed above (see Table 2.2), many reports Indicate that
dermal exposure to inorganic arsenic compounds leads to dermatitis.
However, none of these reports provide quantitative information on
dose-duration relationships. No reports indicating that dermal exposure
is associated with increased risk of cancer were located.

2.2.2  Biological Monitoring as a Measure of Exposure and Effects
     Arsenic  levels In blood, urine, hair, and nails have all been
Investigated  and used as biological  indicators of exposure  to arsenic  A
discussion of the utility and the limitations of each of the  indicators
for human biomonitoring is provided  below.

2.2.2.1  Blood  arsenic
     Most arsenic  is  cleared from blood within  a  few hours  (Tarn  et  al
1979,  Vahter  1983), so measurements  of blood arsenic reflect  exposures
only within the very  recent  past. Typical  values  in nonexposed
individuals range  from 1  to  5 /*g/L  (Heydorn 1970, Valentine et  al   1979.
Hindmarsh and McCurdy 1986).  Consumption  of medicines  containing
arsenic  is associated with blood values of 100  to  250 jig/L.  while  levels
in acutely toxic  and  fatal cases may be 1000 jig/L or higher (Driesback
1980). However, blood levels do not appear to be  reliable  indicators  of
chronic  exposure  to  low  levels  of arsenic.  For  example,  Valentine  et  al
(1979,  1981)  measured the  concentration of arsenic  in whole blood of
residents  in  several  U.S   communities where arsenic  levels in water
ranged from 0.6 to 393 ^g/L  They  found that  In groups  with water levels
ranging  from  6  to 123 ng/L,  average blood levels did not increase in
proportion  to Increased  exposure. Consequently,  measurement of blood
arsenic  is not generally  considered to  be a reliable means of monicqring
human populations for arsenic  exposure.

-------
 26   Section 2

 2.2.2.2  Urinary arsenic

      Because absorbed arsenic is rapidly  and efficiently excreted in
 urine [mostly as monomethylarsonic acid (MMA)  and dimethylarsinic acid
 (DMA)],  analysis of urinary arsenic levels  is  useful  as  an indicator of
 recent exposure. Normal urinary levels  generally  range from 2  to  100
 Mg/L, with values of 20 to 50 ;*g/L being  typical  (Pinto  et al   1977
 Wagner et al.  1979,  Foa et al.  1984). Values of 150 /ig/L or higher are
 common in industrially exposed populations  (Pinto et  al.  1976,  1977-
 Enterline and Marsh 1982;  Beckett et al.  1986) and in populations
 exposed to elevated arsenic levels in drinking water  (Southwick et al
 1981, Valentine et al.  1981,  Harrington et  al. 1978). Some workers have
 found that urinary arsenic levels are linearly related to  airborne
 concentrations (Pinto et al.  1977,  Vahter et al.  1986). but Borgono  et
 al.  (1980) did not find urinary levels  to be well  correlated with
 clinical  signs of arsenic  toxicity.  Elevated levels of arsenic  in urine
 may  occur following ingestion of the  nontoxic forms of arsenic  present
 in fish or seafood (Hindmarsh and McCurdy 1986),  illustrating that
 increased total urinary arsenic levels  may  not necessarily reflect
 exposures of toxicological significance. Chemical  analyses  which  dis-
 tinguish  between inorganic arsenic, MMA, DMA, and  arsenobetaine in the
 urine help to  solve  this difficulty  (Foa et  al. 1984, Lovell and  Farmer
 1985).

 2.2.2.3   Hair  and nails

      Arsenic tends to accumulate  in hair and nails, and measurement  of
 levels in these tissues may be  a  useful indicator  of chronic arsenic
 exposure.  Normal levels in nails  are 0.5 ppm or less, whereas typical
 levels in hair range  from  0.02  to  1.0 ppm (Gordon  1985,  Takagi et  al.
 1986). These values may increase  from several-fold to over  100-fold
 following chronic arsenic  exposure  (Landau et al.   1977,  Valentine  et al
 1979,  Southwick et al.  1981,  Valentine et al. 1985, Bencko et al.   1986)
 The greatest problem  with  the use of these tissues is that both hair and
 nails  appear to adsorb  and strongly retain arsenic from external  sources
 as well as  from internal deposition. Thus, elevated levels  in hair or
 nails may not  be definite  evidence that a significant dose of arsenic
 has been  absorbed, and  could  lead to an overestimation of potential
 health risks.

 2.2.2.4   ELectromyography

      Since peripheral neuropathy  is one of the characteristic results  of
 chronic arsenic  intoxication, electromyographic measurement of nerve
 conduction velocity and amplitude has been evaluated as a means of
 detecting preclinical signs of neuropathy in exposed individuals.  While
 electromyographic  abnormalities have been detected in some exposed
populations  (Hindmarsh  et  al. 1977, Landau et al.  1977,  Valentine et al
 1981), no significant effects were detected in other populations with
moderately elevated exposure  in drinking water (Southwick et al.  1981.
Kreiss et al.  1983). Thus,  this approach does not  appear to be
 sufficiently sensitive  to be useful as a biological indicator of
exposure  (Hindmarsh and McCurdy 1986).

-------
                                             Health Effects Summary   27

2.2.3  Environmental Levels as Indicators of Exposure and Effects

2.2.3.1  Levels found in the environment

     Arsenic is very widely distributed in the environment. Levels in
water typically range from 0.2 to 10 jig/L. with more than 99% of all
drinking water systems having values lower than SO pg/L. Most
epidemiological studies indicate that water levels of 400 jjg/L or higher
may lead to signs of arsenic toxicity in the population, whereas water
concentrations of 100 ng/L are not usually observed to produce
significant noncarcinogenic effects.

     Typical concentrations of arsenic in ambient air range from 2 to
10 ng/m3, with over 99% falling below 80 ng/m3 (Akland 1983). These
levels are not usually associated with any systemic effects, although
they may represent a small increase in the risk of lung cancer.

     For most humans, the diet normally represents the largest source of
arsenic exposure (typically about 25 to 50 pg/day) (EPA 1982b). Levels
in different types of food typically vary between 0.01 and 0.10 ppm,
with higher levels in rice (0.4 ppm), chicken (0.5 ppm), and fish or
shellfish (20 ppm). As previously noted, the organic form of arsenic in
seafood is rapidly excreted in urine and is not of toxicological
concern.

2.2.3.2  Human exposure potential

     Because arsenic is nearly ubiquitous in air, food, water, and soil,
all humans are routinely exposed at low levels via these routes. Outside
the occupational setting, the greatest potential for exposure of
significant health concern is through ingestion of contaminated water
This could result from natural mineral deposits, wells contaminated wich
arsenical pesticides, or groundwater contaminated by a nearby waste
site. Typically, groundwater will contain a mixture of arsenate and
arsenite, depending on pH and oxidation potential, usually with arsenace
predominating. Another pathway that may be of concern in some cases is
ingestion of contaminated soil. All persons ingest low levels of soil
(around 10 mg/day); however, small children may ingest 200 mg/day or
more (Calabrese et al. 1987). Significant inhalation exposure is likely
to occur only around industrial sites, such as copper smelters or
pesticide factories.

2.3  ADEQUACY OF DATABASE

2.3.1  Introduction

     Section 110 (3) of SARA directs the Administrator of ATSDR to
prepare a toxicological profile for each of the 100 most significant
hazardous substances found at facilities on the CERCLA National
Priorities List. Each profile must include the following content

    "(A)  An examination, summary, and interpretation of available
          toxicological information and epidemiologic evaluations on a
          hazardous substance in order to ascertain the levels of
          significant human exposure for the substance and  the
          associated acute, subacute, and chronic health effects.

-------
 28   Section 2

      (B)  A determination of whether adequate information on the health
           effects of each substance is available or in the process of
           development to determine levels of exposure which present a
           significant risk to human health of acute, subacute,  and
           chronic health effects.

      (C)  Where appropriate, an identification of toxicological testing
           needed to identify the types or levels of exposure that may
           present significant risk of adverse health effects in humans.

      This section identifies gaps in current knowledge relevant to
 developing levels of significant exposure for arsenic.  Such gaps are
 identified for certain health effect "end points" (lethality.
 systemic/target organ toxicity,  developmental toxicity,  reproductive
 toxicity, and carcinogenicity)  reviewed in Sect.  2.2 of this profile in
 developing levels of significant exposure for arsenic,  and for  other
 areas such as human biological  monitoring and mechanisms  of toxicity.
 The present section briefly summarizes  the availability of existing
 human and animal data,  identifies data  gaps,  and summarizes research in
 progress that may fill  such gaps.

      Specific research  programs  for  obtaining data  needed to develop
 levels  of significant exposure  for arsenic will  be  developed by ATSDR
 NTP,  and EPA in the  future.

 2.3.2  Health Effect End Points

 2.3.2.1  Introduction and graphic  summary

      The availability of data for  health  effects  in  humans  and  animals
 is  depicted  on  bar graphs in Figs. 2.7  and 2.8,  respectively.

      The bars of full height indicate that there  are data  to meet at
 least one of  the following criteria:

 1.   For noncancer health end points, one  or more studies are available
      that meet  current scientific  standards and are  sufficient  to define
      a  range of  toxicity from no effect levels (NOAELs) to  levels that
      cause effects (LOAELs or FELs).

 2.   For human carcinogenicity, a  substance is classified as either a
      "known human carcinogen" or "probable human carcinogen" by both EPA
      and the International Agency  for Research on Cancer  (IARC)
      (qualitative),  and  the data are sufficient to derive a cancer
      potency factor  (quantitative).

 3.   For  animal  carcinogenicity. a substance causes a statistically
      significant number  of tumors  in at least one species, and the data
      are  sufficient  to derive a cancer potency factor.

 4.   There are studies which show that the chemical does not cause this
      health effect via this exposure route.

      Bars of half height indicate that "some" information for the end
point exists, but does not meet any of these criteria.

-------
                                           HUMAN  DATA
                                                                                                                      SUFFICIENT
                                                                                                                     INFORMATION*
                                                                                                                J
                                                                                                                         SOME
                                                                                                                     INFORMATION
                                                                                                                           NO
                                                                                                                     INFORMATION
                                                                                                     INHALATION
                                                                                                DERMAL
LETHALITY        ACUTE      INTERMEDIATE     CHRONIC   DEVELOPMENTAL  REPRODUCTIVE  CARCINOOENtCITV
            ^	/    TOXICITY        TOXICITV
as
n>
to
t—
3-
to
                      SVSTEMIC TOXICITY

                       'Sufficient information exists to meet at least one of the criteria for cancer or noncancer end points.

                            Fig. 2.7. Availability of information on health effects of arsenic (human data).

-------
                                       ANIMAL DATA
                                                                                                                             ft
                                                                                                                             O
                                                                                                                             r»
                                                                                                     SUFFICIENT
                                                                                                    INFORMATION*
                                                                                                        SOME
                                                                                                    INFORMATION
                                                                                                         NO
                                                                                                    INFORMATION
                                                                                         INHALATION
LETHALITY       ACUTE    INTERMEDIATE    CHRONIC   DEVELOPMENTAL  REPRODUCTIVE  CARCINOOENIOTY
          /                             y    TOXICITV       TOXICITV
                   SYSTEMIC TOXICITY

                    'Sufficient information exists to meet at least one of the criteria for cancer or noncancer end points.


                       Kig. 2.8.  Availability of information <     Ith effects of arsenic (animal data).

-------
                                             Health Effaces Summary   31

     The absence of a  column  indicates  chat no information exists for
that end point and route.  In  some cases,  information for one route of
exposure may not be applicable  for developing levels of significant
human exposure, even if  it is available,  and this is indicated by a
fully shaded cell on the graph.

2.3.2.2  Description of highlights of graphs

     Because arsenic occurs naturally in  groundwater and because arsenic
has been used for many years  as a poison, a therapeutic agent, and an
industrial material, there is a broad database available on the human
health effects of arsenic  exposure. Most  quantitative information
available is for oral  exposure, although  risk of lung cancer from
inhalation exposure has been well studied. Reproductive and
developmental effects  have not been adequately studied in humans,
although limited data  from animal studies suggest this is not an effect
of major concern. Little quantitative information exists on risks from
dermal exposure. While dermal irritation  may be significant, systemic
effects following dermal exposure are rare.

     Arsenic toxicity  in animals has been reasonably well studied by the
oral route, but only limited data are available for inhalation toxicity
in animals. As previously  noted, animals  appear to be considerably less
sensitive to arsenic than  humans, and several of the most characteristic
effects of arsenic toxicity in humans (neuropathy, skin lesions,  and
anemia) have not been  observed in animal  models at doses where humans
would be affected. Thus, animal data are  of limited utility in
estimating significant levels of human exposure.

2.3.2.3  Summary of relevant ongoing research

     Table 2.4 summarizes  ongoing research projects related to arsenic
that are presently funded  by the National Institutes of Health (NIH).
These projects may be  expected to produce valuable new information on
the toxicokinetics, mechanism of action,  and dose-response relationships
for arsenic. Significant research on arsenic is also being conducted in
Taiwan and Mexico.

2.3.3  Other Information Needed for Human Health Assessment

2.3.3.1  Phannacoklnetlcs  and mechanism of action

     There is good evidence in animal and human studies that inorganic
arsenic is metabolized by  methylation. Additional research is needed co
further clarify the relationship between  absorbed dose, methylation
capacity, and levels of inorganic arsenic in blood and tissues. In
addition, a more thorough  evaluation of the toxicity of the metabolites
(MMA, DMA) and transitory  intermediates (e.g., arsenoxy radicals) is
needed. A detailed understanding of these relationships will help in che
selection of appropriate no-effect values for noncarcinogenic effects.
and may also influence the evaluation of  mutagenic and carcinogenic risk
from arsenic. Of special importance is the need for studying differences
in methylating capacity between individuals and for distinguishing
"methylators" from "nonmethylators" (Table 2.4).

-------
32     Section  2
                                      Table 2.4. Summary of ongoing research*
               Pnncipal
             investigator
      Institutional
       affiliation
        Description
        of research
                                                            Data gap
           Carter. D. E.
University of Arizona
           Fowler. B. E.
           Landolph, J. R.
NIEHS. NIH
University of
Southern California
 Investigate pulmonary
 toxicity (lung fibrous)
 of gallium arsmide
 (GaAs). a semiconductor
 with growing industrial
 use

 Investigate molecular
 mechanisms which
 regulate mtracellular
 availability of metals

 Molecular biology of
 transformation of mouse
 fibroblasts in vitro by
 arsenic and other metals

 Epidemiological studies
 of environmental
 toxicants, including
 arsenic

 Absorbtion and clearance
 of metals, including
 arsenic, adsorbed to
 airborne particulates
 such as flyash

 Improved statistical
 means for controlling
 the -healthy worker'
effect in epidemic-
 logical studies; applied
to mortality data for
 8000 arsenic workers

Mutagenisis of metals;
 improved techniques for
detecting mutations in
bacterial, viral, and
eukaryotic systems
          Snyder. C. A.      New York University      Respiratory carcinogenicity
                                                      of Ai,O, in rodents
          Upton. A. C.       New York University      Inhalation carcinogenicity
                                                      of arsenic in rats
           Landngan, P. J    Mount Sinai. New York
           Menzel. D  B.      Duke University
          Robins, J. M.      Harvard University
          Rossman. T. G.    New York University
                                                                                   Toxicity of this
                                                                                   arsenic compound
                                                                                   has not been
                                                                                   studied
                                                                                   Arsenic
                                                                                   toxicokinetics
Mechanism of
arsenic-induced
carcinogenicity

Dose-response
data for arsenic-
induced effects
in humans
Inhalation
toxicokinetics,
role of
paniculate sue
and composition
Better ability
to detect cause-
effect relations
in exposed
humans
                                                      Clarification of
                                                      the co-mutagenic
                                                      activity of
                                                      arsenic, and

                                                      solubility
                                                      Dose-response curve
                                                      for lung cancer

                                                      Dose-response curve
                                                      for lung cancer
               •Adapted from NIH CRISP database.

-------
                                              Healch Effects  Summary    33

2.3.3.2  Monitoring  of  human biological  samples

     A number  of  biomonicorlng options are  available  for  estimating
short-term  and long-term arsenic  exposure  levels  in humans.  Arsenic
levels in blood,  urine,  hair,  and nails  all tend  to increase with
increasing  exposure,  but there is a wide range in "normal" levels, and
values in people  with moderate levels of exposure are sometimes not
distinguishable from typical "background"  levels.  In  addition, in  the
low to moderate exposure range of chief  concern,  there  is not a strong
correlation between  arsenic  levels in these fluids or tissues and  the
onset of arsenic-induced toxicity. Thus, these biomarkers are of limited
utility in  judging the  health  risk to populations with  low to
intermediate levels  of  arsenic exposure.

     This difficulty is  not  primarily related to  analytical  methodology,
but to intra-human variability in toxicokinetics  and sensitivity.
Additional  research  on  other potential biomarkers of exposure (skin or
tissue levels, levels in cells, enzyme activity,  adducts with proteins
or other molecules,  etc.) might lead to  a more sensitive and predictive
indicator.

2.3.3.3  Environmental  considerations

     The environmental  fate  of arsenic is very complex, involving
processes such as oxidation, reduction,  organification, volatilization.
solubllization, and  adsorption. While these processes are understood in
general terms  (EPA 1984a), there  are presently inadequate data to allow
quantitative modeling of arsenic  fate in the environment. Development of
rate constants and identification of the key variables  in rate processes
would permit a more  sophisticated analysis  of arsenic fate than is
presently possible.

-------
                                                                      35
                  3. CHEMICAL AND PHYSICAL INFORMATION

3.1 CHEMICAL IDENTITY

     Arsenic is a naturally occurring metalloid element (atomic number
33). Tables 3.1 and 3.2 list the common name, the Chemical Abstracts
Service (CAS) number, molecular formula, synonyms, and identification
numbers for arsenic and a number of arsenic salts, oxides, and organic
derivatives. These arsenic compounds were selected because their
toxicity and/or presence in the environment identified them as compounds
of concern.

3.2 PHYSICAL AND CHEMICAL PROPERTIES

     Pertinent physical and chemical properties of arsenic and the
selected arsenic compounds are listed in Tables 3.3 and 3.4. The
inorganic compounds of arsenic are solids at normal temperatures and are
not likely to volatilize. The solubility of these compounds in water
ranges from quite soluble (sodium arsenite and arsenic acid) to
practically insoluble (arsenic trisulfide). Some organic arsenic
compounds are gases or low-boiling liquids at normal temperatures.
Except for the organic arsonic acid compounds, they are not readily
soluble in water.

-------

i me J.i. «.




IO
Identification numbers o\
Chemical name
Anenic




Arsenic acid


Arsenic penloxide


Arsenic irwxidc




Arsenic trisulfide


Calcium ancnale


Gallium arsenide
Sodium anenaie

Sodium anemic


Synonym(i) Formula
Arsemc-75 As
Metallic arsenic
Ancnic black
Colloidal anenic

Onhoanenic acid H,AsO4


Arsenic (V) oxide A*,O,
Arsenic acid anhydride
Diarscnic penloxide
Arsenic oxide As2O,
Arscnous acid anhydride
White arsenic
Arsenolue
Claudetitc
Orpimcnl ASjSj
Arsenic sulfidc
Anenic yellow
Calcium onboarsenate Ca,(AsO4),


Gallium monoarsemic GaAs
Duodium ancnatc Na,HAsO4
Sodium bianenatc
Sodium mclaarscnitc NaAsO,


Wiswcsser NIOSH
line notation CAS registry RTECS
AS4 7440-38-2 CG0525000




H3 AS-O4 7778-39-4 CG0700000


AS2O5 1303-28-2 CG2275000


AS2O3 1327-53-3 CG332SOOO




.AS2S3 1303-33-9 CG2638000


CA3AS-O4*2 7778-44-1 CG0830000


ND 1303-00-0 LW8800000
NA2 AS-O-Q3 7778-43-0 CG0875000

NA AS-02 7784-46-5 CG3675000


EPA
hazardous
waste
D004
K084
KIOI
KI02

POIO


POI1


POI2




P038


ND


ND
ND

ND


DOT/UN/'
NA/IMCO
OHM-TADS' shipping STCC" HSDB*
ND' UN 1558 4923204 509
NA28II 4923207
IMCO 6 1
UN 2760
UN 2759
7217393 UN 1554 4923106 431
UN 1553 49 231 OS
IMCO 6 1
7217408 IMCO 61 49231 12 429
UN 1559

7800005 UN 1561 4923209 419
IMCO 61



7800007 NA 1557 49 232 22 428


7216623 UN 1573 4923217 1433
UN 1574
IMCO 61
ND ND ND 4376
ND ND ND 1675

7800057 UN 1686 49 232 91 693
UN 2027
IMCO 61
n
h—
0
3
to


























'Adapted from HSDB 1987. NLM 1987
'Oil and Hazardous Materials/Technical Assistance Data System
 Department of Transportation/International Maritime Dangerous Goods Code
'Standard Irantpurlalion Commodity Code
'Hd      »  Subjlanixj Data Bank
'Ni      nmed. as reported in HSDB 1987

-------

IWHC J.4. V.



Identification numbcri


Chemical aame SynooyrnU)
Arsanilic acid 4-Aminopncnylariooic
acid
Arwnobclainc' Fuh arsenic
A rune Arwnic hydride
Arsenic trihydrtde
Dimelhyliriinic Cacodylic acid
acid Dimclhylaricnic acid
DMA
|)imcih»Uriinc- Catodyl hydride
Uuudium methane- USMA
afwnale
Mcihancariunic 4tid Melh|rlarienic acid
MoDomclbylarsooK acid
MMA
McthyUrunc' Aninomelhane
3-Nnro-4-hydroiy- Roiaraonc (
phcnyi.ffoesc Kti Ns'repfceao! erntatc
acid
Sodium araaoilalc Aloiyl
Sodium dimethyl- Sodium cacodylalc
•rainatc
Sodium methane- MSMA
arionaU
Trimclhylamnc' Aricnic Irimelhyl
Gouoga*
•Adapted from HSDB 1987. NLM 1987
•Oil and Hazardous Materials/Technical Aaaiala


Formula
C.H.AiNO,

(CH^At'CH^OO
AaH,

(CH,).HA>O,


(CH.^Atll
CH.Na.AtO,

CH.H.AsO,


CH.AUI,
C.H.AINO.


C.H,AiNO,Na
(CH.hNaAlO,

CH.NaHAkO,

(CH,),Ai


incc Data Syilcm
hPA
Wuwcucr IMIOSH hazardous
line notation CAS registry RTECS waalc
ZR D-AS-QQO 98 50-0 CF7875000 NlX

{ 64436-13-1 g t
AS H3 7784-42-1 CC6475000 ND

Q-AS-OAIAI 75-60-5 CH7525000 ND


I S93-S7-7 g g
Q-AS-QOAIA-NA-2 144-21-8 PA2275000 ND

Q-AS-QOAI 124-58-3 PA 1575000 K03I


( 593-52-2 g I
WNR BQ E-AS-QQO 121-19-7 CY5250000 ND


ND 127-8)-} CF9625000 ND
Q-AS-OAIAIA-NA- 124-65-2 CH7700000 ND

Q-AS-QOAIA-NA- 2163-80-6 PA2625OOO ND

f 593 88-4 f f



DOT/UN/'
NA/IMCO
OHM-TADS* shipping STCC* HSDB*
ND ND ND 432

* tit
ND UN 2188 49 201 35 510

ND ND ND 360


( lit
ND ND ND 1701

ND ND ND 845


( t t I
ND ND ND 4296


ND NU ND 5189
ND IMCO/UN ND 731
6 1/1688
ND ND ND 754

f ( 1 t



'Department of Traruportalion/lnlcrnalioaal Manltme Coodi Code
'Standard Transportation Commodity Code
•Hazardous Subalancea Data Bank
'Not determined, u icporled in HSDB 1987
'Information not available
•Data from lleilbron Daiabaic 1987
































3
n>
B
}-•
r>
to
§
»3
a
•XJ
,*
M
n
to

a1
o1

to
n
K—
o
a


-------
                                lafttsJJ. Physical and rhimlrsl propertlss of arsenic and i
Molecular Vakno
fhcmical name miatit sletA

Arsenic 74.92 0
Arsenic Mad 141.95 -1-5
Ancoic pcfllfttidf 229 M + 5
Ancnic inoiide 197.82 +3
Arsenic msulfide 246.00 +3
Calcium anenatc 398.08 +5
Gallium arsenide 144.64 +3
Sodium arsenatc 185.91 +5
Sodium anemic 130.92 +3
i Melting
point (°C)
817 (28 aim)
35.5
DOQOQlfKMCS Bl
312.3
300-325
1455
1238
57
ND
Boding
pout (eC)
613
(sublimes)
160
315
465
707
ND
ND
ND
ND
Density
(g/cm1)
5.727
2.2
432
3.738
3.46
3.620
5.31
1.87
1.87
Solubility
Slate
Solid
Solid
Amorphous
solid
Solid
Solid
Amorphous
solid
Solid
Solid
Solid
Color
Silver-gray
While
While
Colorless
Yellow or
orange
While
Dark gray
b
White
Water (g/ 100 cm')
Insoluble
302
Freely soluble
2.1
Practically
insoluble
0.013
ND
Very soluble
Freely soluble
Organic solvents
0
Freely soluble in
alcohol and glyccrol
Freely soluble in
alcohol
Soluble in glycerin;
practically
insoluble in chloro-
form and ether
Soluble in alcohol
6
ND
Slightly soluble in
alcohol, soluble in
glyccrol
Slightly soluble in
alcohol
n
rt
t-
Flammabdity O
Dust flammable
when exposed
to beat or
flame
ND
ND
Not flammable
ND
ND
ND
ND
ND
•Adapted from HSDB 1987. Weast 1985
'Information not available

-------
                          Takk3.4. Physical aod chesilcal propcrtk* of ante ud selected orguk araeak
Chemical name
Ananihc acid

Ancnobclaine*
Arsine

Dimethylarsmic
acid

Dimcihylarnne*

Duodium melhane-

arsonalc
Meihaneanonic acid
Methylanine'

3-Nitro-4-hydro*y-
phenylanonic acid
Sodium arsamlale


Sodium dimelbyl-
snmate

Sodium melhane-

anonaic
Trimethylarsme*


Molecular
weight
217.04

178.06
77.93

13801

10600

18595


13998
9197

263.03
239.05


159.98


16196


12003


Melting Boiling
pout CC) point CO
232 ND

204-210 c
-117 -62.5

195-196 ND

c 36

>355 ND


161 ND
-143 2

ND ND
ND ND


200 ND


119 ND


c 70


Density
(g/cm1)
1.9571

c
2.695

ND

1 213

ND


ND


ND
ND


ND


1 57


c


Solubility
Stale
Solid

Solid
Gas

Solid

Liquid

Solid


Solid
Gas

Solid
Solid


Solid


Solid


Liquid


Color
White

c
Colorless

Colorless

c

Colorless


White


Pale yellow
White


Colorless
to light
yellow
White


c


Water (g/ 100 cm1)
Slightly
soluble

20mL/IOOg

Soluble in
0 5 part water

Insoluble

100


Freely soluble
Insoluble

Slightly
soluble
Soluble


ND


57


c


Organic solvents
Slightly soluble
in alcohol
Soluble in alcohol
Soluble in chloro-
form and benzene
Soluble in alcohol;
practically
insoluble in ether
c

Slightly soluble
in alcohol

Soluble in alcohol
c

Soluble in alcohol and
acetone
Slightly soluble in
alcohol

ND


Soluble in
methanol

c


Flammability
ND

c
ND

ND

Flammable in
air
ND


ND
Ignites
spontaneously
in air
ND
ND


ND


ND


Spontaneously
flammable in
air













_
if
B
I—
n
0)

§
a
•a
a-
^^
(n
n
fti
i—

a
Q
|M
R
o
n
•Adapted fiom HSDB I9K7. Weail 1985
'Data from Meilbron Database 1987
' Information not available

-------
                         4.  TOXICOLOGICAL DATA
4.1  OVERVIEW
     Arsenic is an element that forms a variety of inorganic and organic
compounds, of which the toxicity may vary considerably from compound co
compound. In general, soluble inorganic compounds of arsenic(III) are
considered to be the principal toxic species. These compounds are
relatively well absorbed from both the gastrointestinal tract and the
lungs, and are distributed widely throughout the body. Although the
mechanism of action is not known with certainty, it is generally
considered that arsenic(III) exerts its toxicity by reacting with
sulfhydryl groups of cellular proteins, thereby destroying their
activity (Harvey 1970, Knowles and Benson 1984). Soluble inorganic
compounds of arsenic(V) are also toxic, although usually somewhat less
than arsenic(III). The toxicity of this species may be mediated in pare
by in vivo reduction of arsenic(V) to arsenic(III),  or may be due to the
ability of arsenate to function as an analogue of phosphate, thereby
interfering with normal cellular energy metabolism (Harvey 1970).
Section 4.2 summarizes toxicological data on inorganic forms of arsenic

     In contrast to the inorganic arsenic species, most organic
derivatives have relatively low toxicity. MMA and DMA are formed in vivo
by enzymic methylation of arsenic(III) in the liver.  Because mechylacion
reduces toxicity and increases urinary excretion of arsenic, this
metabolic pathway represents an effective detoxification mechanism,
especially at low doses. Higher doses may exceed the methylating
capacity of the liver, leading to a relative increase in the percent of
the dose present in blood and in tissues as inorganic arsenic.  Section
4.3 presents a summary of toxicological data for MMA, DMA, and other
organic derivatives of potential health concern.

4.2  INORGANIC ARSENIC

4.2.1  Overview

     Because inorganic arsenic (both the trivalent and pentavalenc
forms) are general cellular poisons, it is not surprising that many
tissues are affected by arsenic exposure. By the oral route, the
principal acute effect is irritation of the gastrointestinal tract
Long-term oral exposure may lead to anemia, peripheral neuropathy,
hepatotoxicity, nephrotoxicity, cardiotoxicity, and a group of skin
disorders characterized by hyperpigmentation and hyperkeratoses. Chronic
oral arsenic exposure is also associated with increased risk of skin
cancer and possibly internal cancer as well.

-------
 42    Section 4

      Inhalation expc> ire to dusts  or  aerosols containing  inorganic
 arsenic may lead to  che appearance of some of the  same  symptoms  as  seen
 following oral exposure,  but this  is  rarely of health consequence in
 humans.  Rather,  the principal health  concern following  inhalation
 exposure is increased  risk of lung cancer.

      Dermal contact with arsenic compounds may produce  moderate  to
 severe  skin irritation but is not  generally associated  with any  systemic
 effects.                                                         y

      Despite the variety of adverse effects associated  with arsenic
 exposure,  there  is limited evidence that low levels of  oral intake may
 be beneficial  or essential to animals. If so, arsenic is also likely to
 be beneficial  to humans.  However,  the  estimated required daily Intake is
 quite low (similar to  the typical  daily intake from the diet), and no
 cases of arsenic deficiency have been  recognized in humans.

 4.2.2  Tozlcokinetics

 4.2.2.1   Overview

      Most  toxicokinetic  studies of  inorganic arsenicals in humans and
 animals  have been conducted with readily soluble or moderately soluble
 salts or  oxides.  In general,  these compounds are well absorbed, both in
 the gastrointestinal tract and the lung,  and are widely distributed
 throughout  the body. Trivalent arsenicals are metabolized in the liver
 by methylation to form primarily dimethylarsinic acid. This compound is
 less  toxic  than  the parent inorganic compounds and is efficiently
 excreted  in urine.  Pentavalent arsenicals undergo reduction to the
 trivalent form before methylation may occur.  At low-to-moderate doses of
 inorganic arsenic, methylation and urinary excretion prevent marked
 accumulation in  the body,  but high doses  may lead to the deposition of
 inorganic arsenic in some  tissues.

     Analysis of  the absorption, distribution,  metabolism, and excretion
of inorganic arsenic compounds is complicated by important differences
between compounds. Of primary importance  in absorption is the solubility
of the-compound, with solutions of arsenic and readily soluble compounds
being absorbed more efficiently than solids and poorly soluble compounds
 (Harrison et al.  1958,  EPA 1984b).  Other  variables that may influence
absorption are dosing rate  and (for solid materials) particle size.

     Another important factor in toxicokinetic studies is species
variation. Although some animals appear to absorb and metabolize
 inorganic arsenicals similarly to humans,  there are a number of cases
where this does not happen.  For example,  hamsters have low
gastrointestinal absorption, rats have excessive binding in red blood
cells, and marmoset monkeys have low methylating capacity. For these
reasons, toxicokinetic data  from animals  must be applied to humans with
caution.

4.2.2.2  Absorption

     Oral.  Absorption of  inorganic arsenic is dependent upon the water
solubility of the arsenic  compound. There is  general agreement that
soluble trivalent (arsenic(III)] or pentavalent (arsenic(V)] arsenic Is

-------
                                                 ToxicoLogLcaL Data   U}>

almost completely absorbed (more than 90%) from the gastrointestinal
tract of laboratory animals (Vahter 1983, EPA 1984a).  Hamsters appear co
be the exception, since they only excreted 30 to 40% of an oral dose of
sodium arsenite in the urine and about 50% in the feces (Charbonneau et
al. 1980a, Marafante and Vahter 1987). When sodium arsenate was
administered to hamsters, substantial absorption, as indicated by
urinary excretion of 75% of the dose, was found (Marafante and Vahter
1987). Studies on mice indicate that absorption does not depend on dose
in the range of 0.4 to 4.0 mg/kg body weight (Vahter and Norin 1980)
Aqueous suspensions of arsenic trioxide were absorbed only to the extent
of 40 and 30% in rabbits and rats, respectively, because of its limited
water solubility (Ariyoshi and Ikeda 1974).

     Humans also appear able to absorb extensively arsenic compounds
from the gastrointestinal tract. Greater than 95% of inorganic arsenic
taken orally by man appeared to be absorbed since <5%  appeared in the
feces (EPA 1984a).  Buchet and coworkers (1981a) administered an oral
dose of 500 jig and found 46% in the urine after 4 days. When multiple
doses were administered, a steady state of 60% of the  dose was excreted
in the urine (Buchet et al.  1981b).  Feces were not analyzed in these
studies, but it is assumed that the urinary excretion is representative
of extensive arsenic absorption. When insoluble arsenic selenide was
taken orally, there was no absorption as indicated by no increase in
urinary arsenic excretion (Mappes 1977).

     Inhalation.  Absorption of arsenic from the lungs depends in large
part on particle size. Large particles (e.g., 10 urn) tend to be
deposited in the upper airway and are cleared by ciliary movement with
only limited absorption across lung tissue. In contrast, small particles
(e.g., 0.1 to 1 ^m) tend to penetrate deep into the lung and are
deposited in alveoli where absorption occurs across the respiratory
epithelium (EPA 1984a). Absorption of arsenic that has been deposited in
the lungs will also depend on the chemical form of the arsenic. Animal
studies indicate that water-soluble compounds such as arsenic trioxide
(arsenic(III) ) , sodium arsenate (arsenic(V)], and dimethylarsinic acid
(Dutkiewicz 1977, Stevens et al. 1977a, Rhoads and Sanders 1985,
Marafante and Vahter 1987) are rapidly absorbed. The clearance half-Life
from the lung of rats given an intratracheal instillation of an aqueous
solution of arsenic trioxide was calculated to be 31 min (Rhoads and
Sanders 1985). Despite this very rapid clearance, small fractions of the
arsenic remained in the lungs for several weeks after exposure, with a
half-life of about 75 days. Less soluble arsenic compounds (e.g., lead
arsenate) are not cleared so rapidly, with 45% of an intratracheal  dose
remaining in the lung 3 days after instillation  (Marfante and Vahter
1987). Similar evidence of slow lung absorption was reported by
Pershagen et al. (1982) for calcium arsenate and by Webb et al.  (1984)
for gallium arsenide.
     Limited information about pulmonary deposition and absorption  has
been reported for humans. Holland and coworkers  (1959) studied  the
deposition and absorption of arsenic from  arsenite-containing cigarettes
and from arsenic-containing aerosols in  lung cancer patients.
Approximately 40% was deposited, and 75  to 85% of the  deposited  arsenic
was absorbed from the lungs within 4 days. Other human studies have
examined the relationship between airborne arsenic concentrations and

-------
 44   Section 4

 urinary excretion of arsenic (Pinto et al.  1976,  Smith et al  1977
 Vahter et al. 1986). All studies showed a good correlation between'
 arsenic levels In air and urinary excretion of inorganic arsenic
 metabolites. Vahter and coworkers (1986) estimated that 42% of the
 inhaled arsenic was excreted daily.

      Dermal.  Insufficient Information exists  to  evaluate dermal
 absorption of the arsenic compounds.  Vahter (1983) has reviewed the
 clinical reports of arsenic toxlcity  after  accidents  where the only
 exposure was through the skin,  but the data are not quantitative.

      Transplacental.  Transfer  of Inorganic arsenic to the placenta  and
 fetus Is rapid after parenteral or oral administration.  Hood et al.
 (1987)  administered sodium arsenate to pregnant mice  by gavage
 (40 mg/kg) or intraperitoneal injection (20 mg/kg). Levels in placenta
 and fetus followed a similar time course of accumulation and clearance
 in both cases,  reaching maximum values around  1 to 2  h after exposure,
 and then declining to near control levels within  24 h.  In both cases,
 levels  (expressed as ^g arsenic/ng tissue)  were about 2  to 3 times
 higher  in placenta than in the  fetus.  During the  first several hours,
 most of the fetal arsenic was inorganic, but DMA  became  the  predominant
 form within 4 to 6 h.  Similar results  have  been observed for trlvalent
 arsenic,  which  was transferred  across  the placenta  in pregnant rats  and
 was detected In newborn rats  when the  dams  were fed arsenic  trloxide in
 their diets (Perm 1977,  Hanlon  and Fern 1977).  Hanlon and Perm (1987)
 exposed pregnant hamsters to  arsenate,  using a  subcutaneous  osmotic
 minipump and observed that  tissue  levels in placentas  were  1.2  to 2.0
 times higher than levels in maternal blood.  Most  of the  placental
 arsenic was bound to macromolecules. while  that in  maternal  blood was
 free. Lindgren  et al.  (1984)  reported  that  the  marmoset  monkey  showed a
 lower rate  of placental  transfer  of arsenate and  arsenite  than  mice.
 They suggested  that  this was  probably  related to  the  lack  of methylating
 ability in  the  marmoset  which resulted in greater tissue binding in  the
 mother.

 4.2.2.3  Distribution

      In  most animals, all but a small  fraction  of systemic arsenic is
 rapidly  cleared  from tissues. The  rat  is an  exception, since arsenic is
 avidly bound by  rat  red  blood cells (RBC). The  rat  Is  a poor model for
 human toxicokinetic  data for  this  reason.

      Blood.   In  studies  reviewed by Vahter  (1983), humans, dogs, mice.
 and  rabbits  cleared  arsenic from blood  in a biexponential, or possibly
 triexponential, curve. The major part of the blood arsenic (>90%) was
 cleared at  a  high  rate,  the half-life being  1 to 2 h.  The half-lives of
 the  second  and third phases have been estimated to be  about  30 and
 200  h, respectively. A more detailed study measured the clearance of
 arsenate after intravenous administration of 0.4 mg/kg to rabbits
 (Marafante  et al.  1985). The plasma concentration of arsenic(V)
 decreased with a  first-order half-life of about 1  h. Some arsenic(V)  was
 rapidly reduced to arsenic(III); by 15 min,  10% of the plasma arsenic
was  in the  form of arsenic(III)   About 30% of plasma arsenic was as
 arsenic(V). and about 60% was bound to plasma protein. The clearance of
arsenic(III) was biexponential.  wich half-lives of about 10 min and 2 h

-------
                                                 lexicological Data   £o

 The kinetics  of  protein-bound arsenic  were  also biphasic with half-lives
 of 15 min  and 2.5  h.  In  RBC,  there was more arsenic(III) than
 arsenic(V), and  arsenic(III)  was  approximately equal  to its plasma
 concentration, while  arsenic(V) was  only  10% of its concentration  The
 concentration of protein-bound arsenic increased up to 1 h and then
 began a slow  decline. This  suggests  an irreversible binding to RBC
 protein that  was not  present  with plasma  protein.

     Vahter and  Marafante  (1985)  examined the blood distribution of
 0.4 rag/kg  arsenic  from sodium arsenate following intravenous injection
 in marmoset monkeys,  a species which cannot methylate arsenic. Arsenic
 was rapidly cleared from the  plasma, and  the relative amount of
 arsenic(III)  in  the plasma  increased from only a few percent 30 min
 after injection  to about 50%  at 6 h. In RBC, arsenic(III),  arsenic(V),
 and a protein-bound arsenic were  found. Arsenic(V) was cleared from RBC
 more rapidly  than arsenic(III), but  both  were completely cleared by 72
 h. The protein-bound  arsenic  remained  in  RBC.

     Arsenic  in  blood is also  rapidly  cleared in humans (Ducoff et al
 1948, Mealey  et  al. 1959, Tarn  et  al. 1979). Hunter et al. (1942)
 reported nearly  complete removal  of  blood arsenic in humans 24 h after a
 subcutaneous  injection of potassium  arsenite, and Mealey et al.  (1959)
 reported that more than  90% of an intravenous dose of arsenic is removed
within several hours. Tarn et  al.  (1979) reported that clearance of
arsenic from blood in humans  fit  a 3-exponential model with half-lives
of 1, 5,  and  35  h.

     Heydorn  (1969) reported concentrations of arsenic in whole blood,
RBC,  and plasma  for normal healthy individuals from Denmark and for a
number of population  groups in Taiwan  with high arsenic exposure levels
In general, the  concentration of  arsenic  (expressed as Mg/D was
somewhat higher  (1.1- to 3.3-fold) in  RBC than in plasma, with whole
blood having an  intermediate value.  This  may reflect a binding of
arsenic to red-cell proteins.

     Other tissues.  There is general  agreement that exposure of various
animal species to either trivalent or  pentavalent arsenic leads to che
initial accumulation of  arsenic in liver, kidney,  lung, spleen,  aorta.
skin, hair, and upper gastrointestinal tract (EPA 1984a). These tissues
are cleared rapidly except for skin  and hair where the sulfhydryl groups
of keratin may promote tight arsenic(III) binding. Arsenic is apparently
retained in the brain of experimental  animals with slow clearance
reported (Crema  1955).

     More detailed studies have been published recently which have
focused on the tissue and subcellular  distribution of different arsenic
species.  Vahter and Norin (1980)  found that levels of arsenic in kidney,
liver,  bile,  brain, skeleton, skin,  and blood were always greater
(twofold to tenfold)  in mice given oral doses of arsenic(III) as
compared to arsenic(V),   and this  difference was greater at the higher
dose.  Similar results were reported  in Syrian golden hamsters by Cikrc
et al.  (1980). An autoradiographic study  comparing trivalent and
pentavalent arsenic in pregnant mice and  monkeys indicated that
arsenate,  but not arsenite. showed affinity for calcified areas of che
skeleton (Lindgren et al. 1984)   This  is  in apparent disagreement wich
earlier studies by Vahter and Norin  (1980) but is logical since arsenace

-------
 46   Seccion 4

 is an analog of phosphate,  a normal component  of bone.  After oral
 administration of 4.5 mg/kg arsenic trloxlde to  hamsters,  Yamauchi  and
 Yamamura (1985) reported that MMA concentrations were higher than DMA In
 organs and tissues and that DMA tended to be detected after  the
 appearance of MMA. Inorganic arsenic disappeared rapidly  from the brain.
 liver, lung,  and spleen,  although concentrations In  the hair,  skin,
 muscle,  and kidney remained slightly elevated  at 120 h  after
 administration.

      The cellular distribution of the arsenlcals has also  been examined.
 Fischer  et al.  (1985) reported that mouse flbroblasts did  not  take up
 arsenic(V) as well as arsenic(lll).  Similar findings were  reported by
 Lerman and Clarkson (1983)  in rat liver hepatocytes. Vahter  and
 Marafante (1983)  incubated  arsenlte,  arsenate, and DMA  with  homogenates
 of mouse and  rabbit liver,  lung,  and kidney and  showed  that  arsenite  was
 the main form of arsenic  bound to tissues.

      There are  very few studies which have examined  the tissue
 distribution  of arsenic after prolonged exposure. Vahter (1983) gave
 arsenic(III)  and arsenic(V)  orally  3  times per week  for 12 weeks  to mice
 and followed  the  increase of arsenic  concentration with time. Skin and
 hair  showed the greatest  increase while concentrations  In  liver,  kidney,
 lung,  and intestinal  mucosa  increased much less. Accumulation was
 significantly higher  after administration of arsenic(III)  than after
 arsenic(V).

      Tissue distribution  of  arsenic  in humans has been  studied using
 autopsy  and dosing data.  Kadowaki (1960) determined  that nails contained
 0.89  ppm;  hair, 0.18  ppm; bone, 0.07  ppm to 0.12 ppm; teeth, 0.08 ppm;
 and skin,  0.06  ppm.   The  researcher  found these  tissues had  the highest
 concentrations  of arsenic in the  body of a Japanese population. Heart.
 kidney,  liver,  and lung contained somewhat lower concentrations in the
 range  of 0.04 to  0.05 ppm, brain  tissue only slightly less (0.03  ppm).
 Liebscher and Smith (1968) analyzed  soft tissues from nonexposed  persons
 from  Scotland and found that their  lungs had the highest concentration
 (0.09  ppm), compared  with levels  of 0.03 ppm in  liver and kidney.
Absolute  levels were  highest In hair  (0.46 ppm), nails  (0.28 ppm). and
 skin  (0.08 ppm).  Other  tissues, including bone and teeth, contained 0.06
ppm or less.

     Brune et al.  (1980)  analyzed samples of liver, kidney,  and lung
 taken  at  autopsy  from workers who had retired from a nonferrous metal
refinery  and  smelter. Levels  of arsenic in kidney and liver were  not
significantly higher  than in controls, but levels in lung tissue  tended
to be about six times greater than  in controls. This was true both for
workers who had been  exposed recently (within 1.5 years of their  death)
and those who had not been exposed  for 2 to 19 years prior to death.
This suggested  to  Brune et al. that arsenic had a long half-life  in the
lung,  but  this  interpretation is  complicated by the fact that exposure
levels may have been  significantly higher In workers exposed some years
ago, and  this group may originally have had higher tissue burdens of
arsenic  in the  lung.

-------
                                                 Tox Leo logical Data   47

4.2.2.4  Metabolism
     Substantial information has appeared in the recent literature on
arsenic metabolism. This has been triggered by the recent development of
analytical methods which can separate the metabolites of inorganic
arsenic [arsenic(III),  arsenic(V), MMA, and DMA] and distinguish them
from the arsenic compounds found in shellfish (arsenobetaine and
arsenocholine).
     Animals.  There is general agreement on the products of arsenic
metabolism in mice, rats, rabbits, hamsters, and humans (Klaassen 1974,
Vahter 1983; EPA 1984a; Hanlon and Perm 1986a,b; Marafante and Vahcer
1984, 1986, 1987; Marafante et al. 1985; Yamauchi and Yamamura 1984)
The following general statements can be concluded from in vivo and in
vitro studies:

   • The major site of methylation is the liver.

   • DMA is the major metabolite in most animals and humans, and it
     appears mainly in the urine.

   • MMA is most often a secondary metabolite and its appearance in
     urine varies with the anima-l species.

   • MMA can be partially methylated to DMA, but neither species is
     significantly demethylated to inorganic arsenic.

   • Methylation results in a detoxification of inorganic arsenic (about
     1 order of magnitude per methyl group) and increases the rate of
     arsenic excretion.

   • Trivalent arsenic is the substrate for methylation, and arsenic(V)
     must be reduced to arsenic(III) before methylation can occur.

   • Methylation is dependent on dose level. The percentage of DMA  in
     the urine decreases with increasing  inorganic arsenic dose level,
     while the amount of retained arsenic increases.

   • Earlier data on arsenic metabolism,  toxicokinetics, and  toxic icy
     must be reassessed in light of current knowledge about methylation
     in different species.
     Several studies have focused on understanding the methylation
process in vivo and  in vitro. Rat liver  in vitro only accepts  trivalenc
arsenic as a substrate for methylation and  requires  reduced glutathione
for activity  (Buchet and Lauwerys 1985).  Similar results were  obtained
by Lerman et al. (1983), who studied methylation of  arsenic(III)  and
arsenic(V)  in cultured hepatocytes. They  observed that arsenite was
converted to DMA, but arsenate was not  taken up by the cells  and  no
metabolism could be  detected. The authors postulated that  this  may  be
due  to the  fact  that at physiological  pH, arsenite is not  ionized.
whereas arsenate is  ionized.
     Hanlon  and  Perm (1986a.b) examined the chemical species  of arsenic
present in  the blood after constant  infusion or intraperitoneal

-------
48    Section  4

injection  of  hamsters  with  64.2 /*molAg of sodium arsenace. The major
species  in plasma was  arsenic(V), with much smaller amounts of
arsenic(III)  and DMA.  No  protein binding was found in plasma. Similar
values were observed in RBC, except that a fraction of the arsenic  (13%
of  the total  at 1 h) was  bound to cellular proteins (probably
hemoglobin).

      In  vivo  methylation  in mice and rabbits can be blocked by treatment
with  100 ^molAg periodate-oxidized adenosine, and that treatment causes
an  increase in the amount of arsenic retained in tissue (Marafante  and
Vahter 1984,  Marafante et al. 1985). A choline-deficient diet in rabbits
decreases  the methylation of arsenic and increases the retention in
liver, lung,  and skin  (Marafante and Vahter 1986). These maneuvers  to
decrease the  methylating  capacity in animals with concomitant increased
retention  of  arsenic in tissues may be considered to be a model for
humans who  have low methylating capacity,  due either to genetic or
dietary  factors. Arsenic  pretreatment of mouse fibroblast cells made che
cells more  resistant to arsenic toxicity and increased the methylation
of  inorganic  arsenic to MMA and DMA, suggesting the enzymic methylating
system may be inducible (Fischer et al.  1985).

     Humans.  Several  studies indicated that inorganic arsenic is
methylated  to MMA and  DMA in humans, much as it is in animals. For
example, Buchet et al.  (1980) reported that urinary excretion was about
60% DMA, 20%  MMA,  and  20% inorganic arsenic,  both in smelter workers
exposed  to As203 and in individuals with typical dietary exposure to
arsenic. Similar proportions of inorganic and organic urinary
metabolites in humans  have been reported by Vahter (1983)  and Buchet et
al. (1981a,b). Ingestion  of MMA is accompanied by limited methylation
(about 13%) to DMA. while no metabolism of ingested DMA occurs (Buchet
et al. 1981a).

     In  animals, high  doses of arsenic appear to saturate  the
methylation system. For humans,  Love11 and Farmer (1985)  studied urinary
excretion of  arsenic in patients who ingested very high doses of As203
for the  purpose of suicide.  Shortly after the exposure,  most arsenic in
urine (about  90%)  was  inorganic,  and this fell to about 30% after 100 h,
accompanied by a concomitant increase in the level of organic arsenic
(MMA and DMA), from about 10 to 70%. Buchet et al.  (1981b)  found that
doses of 125  to 500 /ig/day of sodium arsenite were methylated
approximately equally  (about 80%) in human volunteers,  but that the
percentage of the dose methylated began to decrease at 1000 Mg/day,
accompanied by a slowed clearance of arsenic.  This suggests that the
enzyaic  systems of the liver may begin to become saturated at doses
between  500 to 1000 pg/day  (EPA 1987b).

     Foa et al.  (1984)  observed a broad variation between individuals in
the percentage of methylated arsenic in urine,  both in workers in a
glass factory where exposure to As203 was elevated,  and in a group of
people with normal exposures through food. This indicates  that some
individuals may have lower methylating capacity than others.   The
researchers selected a group of five glass workers with high urinary
arsenic  concentrations and suspended their exposure for one month.
Urinary  concentrations of inorganic arsenic and its methylated
metabolites decreased with  time nearly to that of the control

-------
                                                 Toxicologies! Data   49

populacion. When occupational exposure was resumed, only a moderace
increase was seen for urinary inorganic.arsenic and its methylated
metabolites. Two months after exposure resumed, urinary concentrations
of total arsenic were still diminished relative to previous levels. This
suggests that methylating capacity may adapt in proportion to exposure,
but that full methylation capacity for high exposures takes several
months to build up and that any accommodation the body had made to very
high arsenic levels is lost rapidly.

4.2.2.5  Excretion

     The major route of arsenic excretion is in the urine for all
chemical species of arsenic that are absorbed. Biliary excretion may be
significant, but enteric reabsorption is such that little arsenic is
excreted in feces.
     Urinary.  Mice, rabbits, swine, dogs, and monkeys usually excrete
>70% of injected trivalent and pentavalent arsenic in the urine within
24 h (Ducoff et al. 1948, Crema 19SS. Ginsberg and Lotspeich 1963,
Peoples 1964, Munro et al. 1974, Lakso and Peoples 1975, Tarn et al.
1978,  Charbonneau et al. 1978a). Urinary excretion of arsenic in the rat
is much slower, with about 10% of a parenteral dose excreted in 4 days
(Gregus and Klaassen 1986). Excretion of DMA administered to rabbits or
mice by injection or by the oral route was essentially complete within
24 h (Vahter and Marafante 1983, Vahter et al. 1984).

     Several studies have compared the rates of excretion of trivalent
arsenic and pentavalent arsenic (Vahter and Norin 1980, Vahter 1981,
Marafante et al. 1985). Differences in urinary excretion rates appeared
to relate to the relative degree of methylation of the inorganic
arsenic, with rapid formation of DMA enhancing the urinary excretion
Further, differences in excretion rates were dose-dependent. Marafante
et al. (1985) examined the urinary excretion within the first 6 h after
intravenous sodium arsenate administration to rabbits. It was found that
arsenic(V) was excreted immediately and that DMA excretion required 2 to
3 h before appreciable urinary excretion occurred.

     Mappes (1977) administered a single oral dose of sodium arsenite to
a human and found maximal renal excretion at 3 h and about 25% of the
dose in the urine by 1 day after exposure. Crecelius (1977) studied the
urinary excretion of arsenic from a person who had ingested wine
containing 50 /jg arsenic(III) and 13 pg arsenic(V) or water containing
200 pg of mainly arsenic(V). About 80% of the arsenic ingested with wine
was recovered in urine in 61 h, and 50% of the arsenic ingested with
water was found within 70 h. Mealey and coworkers  (1959) measured
urinary arsenic in a patient administered trivalent arsenic
intravenously and found -60% of the dose in the urine within 24 h.

     Buchet et al. (1981b) administered oral sodium arsenite doses of
125 to 1000 Mg for 5 days and found that a steady state was reached and
that 60% of the administered dose of arsenic appeared in the urine each
day. Vahter et al. (1986) studied urinary excretion of arsenic in
smelter workers exposed to airborne levels of 1 to 194 pg/m^. The amount
of arsenic (inorganic, MMA, and DMA) in urine increased in proportion to
exposure level, and the relationship was well described by a straight -
line equation.

-------
 50    Section  4

      Fecal.   Little  arsenic  can be  recovered in human  feces after either
 oral  doses  or parenteral  administration of  inorganic arsenic  (Vahter
 1983).  Fecal  arsenic levels  for rabbits administered arsenite
 intraperitoneally  were  -10%  of the  dose in  4 days  (Bertolero et al.
 1981).  Hamsters  administered arsenite by intraperitoneal  injection
 excreted about 5%  of the  dose in  the bile within 24 h  (Cikrt et al.
 1980).  About  25% of  arsenic  trichloride given intravenously to rats was
 excreted in bile within 2 h,  but  <10% appeared in  feces over a 7-day
 period  (Klaassen 1974.  Gregus and Klaassen  1986).  Rats excreted arsenic
 in  the  bile 40 times faster  than  rabbits and 800 times faster than dogs.
 No  biliary  excretion data exist for humans. Although biliary excretion
 may be  significant and  highly variable between species, it will not
 contribute  extensively  to elimination because of reabsorption from the
 intestines  (Klaassen 1974, Cikrt  et al. 1980).

      Other  routes.   There is  no evidence that expired air is a route of
 excretion for arsenic.  Profuse sweating may eliminate 2 Mg/h. and
 desquamation may account  for  small quantities of arsenic  (Vahter 1983).
 Hair  and nails have  been  considered as an excretory route for arsenic,
 but their contribution  would  be minor compared to  other routes of
 excretion.

 4.2.3  Toxicity

 4.2.3.1  Lethality

      Several workers have  reported that oral doses of about SO to 300 mg
 of  inorganic arsenic may  be fatal to adults (Vallee et al. 1960,
 Hindmarsh and McCurdy 1986, Armstrong et al. 1984,  Zaloga et al.  1985).
 Subchronic oral exposure  to only about 3 mg/day was fatal in a number of
 infants exposed to arsenic via contaminated milk (Hamamoto 1955).  On
 this basis,  the acute and subacute lethal dose  in humans may be
 estimated to be about 0.6  mg/kg/day or higher.

      In animal studies, oral  LDso values for various inorganic arsenic
 compounds have been  reported  to range from 10 to 300 mg/kg.  with soluble
compounds being more toxic than poorly soluble  forms (NAS 1977,  EPA
 1984a).  For example,  Harrison et al. (1958) reported that the acute oral
LDSO  for As203 given to several different strains of mice by gavage
 ranged from 26 to 47 mg/kg- A value of 15 mg/kg was reported for rats
dosed by gavage,  but the value was much higher  (145 mg/kg) when given in
 food.  Intraperitoneal LDso and LD75 values range from 4 to 20 mg/kg.
These results suggest that animals are not as sensitive to arsenic as
humans,  and that this difference is not due entirely to differences in
 gastrointestinal absorption.

      Inhalation and dermal exposure are not normally associated with
 acute lethality in humans  or animals.

 4.2.3.2  Systemic/target organ toxicity

     Gastrointestinal disturbances.   Oral exposure of humans to arsenic
 often produces a range  of  gastrointestinal signs,  with nausea, vomiting,
 diarrhea, and thirst being most common (NAS 1977,  EPA 1980c, 1984a).
Armstrong et al.  (1984) described an incident in which a family of
 8 suffered severe  arsenic  toxic icy from ingestion of water containing

-------
                                                  ToxicologicaL Oaca   5L

 108  mg/L of arsenic.  All of che exposed Individuals  suffered marked
 symptoms of gastroenteritis.  Vomiting and diarrhea were  noted in  7/8 and
 6/8  of  the  individuals,  respectively,  with swollen throat  and abdominal
 pain in 3/8.  Similar  gastrointestinal signs may  also occur at much lower
 doses.  Tay  and Seah (1975)  noted gastrointestinal involvement in  17 of
 74 people ingesting an arsenic-containing herbal preparation at an
 estimated dose of  3 to 10 mg/day.  Zaldivar (1974) noted  diarrhea  and
 anorexia in residents of Antofagasta,  Chile, where drinking water
 contained about 0.4 mg/L (corresponding to a dose of about 0.5 to
 1 mg/day).

      Gastrointestinal signs are chiefly associated with  oral exposure
 and  are  rarely noted  following inhalation or dermal  contact. The most
 likely mechanism of these disturbances  is direct toxicity  to the
 epithelial  cells of the  gastrointestinal tract,  with resulting
 irritation,  injury, and  abnormal function.

      Hematological  effects.   The hematopoietic system is affected by
 both  short-  and long-term arsenic  exposure.  Effects  include anemia,
 leukopenia,  and eosinophilia  (EPA 1984a).  Hamamoto (1955)  reported these
 effects  in  infants  exposed  to about  3.5  mg/day in contaminated milk for
 33 days,  and Mizuta et al.  (1956)  reported anemia and leukopenia  in
 adults ingesting about 3  mg/day in contaminated  soy  sauce.  Similar signs
 of impaired hematopoiesis are also commonly observed in humans
 chronically exposed to arsenic in  water,  medicine, or the  workplace (EPA
 1984a). Woods  and Fowler  (1977,  1978) reported that  arsenate exposure of
 rats  (20  to 85  mg/L in water  for 6 weeks  or more) resulted in decreased
hemoglobin production accompanied  by decreased activity of several
 enzymes  (hepatic w-aminolevulinic  acid  synthetase and ferrochelatase)
 required  for heme biosynthesis.

     Cardiological  and vascular effects.   Large  oral  exposures to
arsenic have been reported  to cause  injury and abnormal function in
cardiac tissue. Rosenberg (1974) reported that autopsy of  5 children
exposed to high levels of arsenic  (up to  0.8 mg/L) in drinking water
supplies  in Antofagasta,  Chile,  revealed evidence of myocardial
 Infarction in  2 cases and arterial thickening in all cases, and Zaldivar
 (1974) reported several cases of myocardial infarction and arterial
 thickening in children consuming water containing about 0.6 mg/L.
Altered electrocardiograms  (prolonged Q-T intervals  and abnormal T-
waves) have been described  in several studies of humans exposed to
arsenic (NIOSH  1975.  HAS  1977,  EPA 1984a,  Zaloga et  al. 1985,  Hindmarsh
and McCurdy 1986).

     Peripheral vascular  disease leading to gangrene of the toes and
feec ("Blackfoot disease") has  been reported to  occur in association
with chronic arsenic  exposure from contaminated  drinking water in Taiwan
 (Tseng et al. 1968, Tseng 1977). The overall incidence of  the disease
was 0.9% in a population  of 40,000 persons  consuming water  containing an
average of about 0.4  to 0.6 mg/L of arsenic, while no cases were
observed in a population  of 7500 people  consuming water containing <0 02
mg/L. Prevalence rates tended to  increase  as a function of age (i.e ,
duration of exposure)  and with  the arsenic  content of the  water.  For
example, when exposure was stratified into  three  levels  (low - 0  - 0 29
mg/L, medium. - 0.30 -  0.59 mg/L. high -  0.60 mg/L and above),  the

-------
 52    Section 4

 prevalence rates (cases/1000)  were  4.5,  13.2, and  14.2 for  20-  to  39-
 year-olds,  10.5, 32.0,  and 46.9  for 40-  to  59 -year-olds, and 20.3.  32 2,
 and  61.4  for those  60 years old  and older,  respectively. Severity  of the
 disease was also related directly to duration of exposure.  Blackfooc
 disease was often fatal,  with  an overall mortality rate close to 50%.

      Similar peripheral vascular lesions have been reported in  vintners
 exposed to arsenical  pesticides  (NAS 1977), in persons in Chile
 consuming water  containing about 0.8 mg/L of arsenic (Borgano and
 Greiber 1972), and  in patients from a region in Mexico where arsenic
 toxicity  is endemic (Salcedo et  al.  1984).  Peripheral vascular  disease
 is not normally  associated with  inhalation  exposure to arsenic, although
 Lager kvist et al.  (1986)  reported evidence  of altered blood vessel
 function  in copper  smelter workers  exposed  to about 50 ^g/m3 in air.
     Some  researchers have questioned the role of arsenic in Blackfoot
disease. For example, Kuo and Chen  (1969) reported that the prevalence
rate for Blackfoot disease varied widely (from 1.8 to 26.1 cases per
1000) in Taiwanese villages where drinking water levels of arsenic were
quite similar  (380 to 850 /ig/L) . Yu (1984) and Yu et al. (1984) noted
that vascular  disease was not observed in a number of studies where oral
arsenic exposure was elevated, and- proposed that the occurrence of
vascular lesions in Taiwan may be related to the presence of a
fluorescent arsenic -containing compound of unknown structure which is
present in water where Blackfoot disease is endemic. Ko (1986) noted
that the incidence of Blackfoot disease increased in Taiwan after steps
were taken to  reduce arsenic exposure through groundwater and,
therefore, concluded that arsenic may not be the factor causing this
disease.

     Neurological effects.  Both peripheral and central neuropathy have
been observed  in humans following arsenic exposure.  Very high oral doses
may produce acute encephalopathy (Armstrong et al.  1984, Danan et al.
1984, Beckett  et al. 1986). Other indicators of central damage include
mental retardation in arsenic -exposed children (Hamamoto 1955), hearing
loss, and  abnormal electroencephalograms (EPA 1984a) .

     Arsenic -induced peripheral neuropathy is typically characterized by
paresthesia, hyperesthesia. and neuralgia, with muscle pain and
weakness.  Such effects have been noted in patients exposed to doses of
3 to 10 mg/day for periods ranging from several weeks (Mizuta et al .
1956) up to several years (Silver and Wainman 1952).

     Arsenic -induced peripheral neuropathy can sometimes be detected by
electromyographic (EMC) techniques.  Characteristic changes include
decreased  nerve conduction amplitude with little change in nerve
conduction velocity (NCV) (Donofrlo and Wilbourn 1985). Hindmarsh et al
(1977), studying a group of patients exposed to elevated levels of
inorganic  arsenic in drinking water (above 100 Mg/L) ,  observed EMC
abnormalities  in about half of the cases. Valentine  et al.  (1981)
measured NCVs  in residents of several western communities where drinking
water levels of arsenic ranged from 50 to 387 ng/L.  Although most
findings were  in the normal range, a significant decrease in the ulnar
sensory nerve  function was noted in both males and females. Landau et
al. (1977) reported level- and duration- dependent decreases in EMG
measurements in smelter workers exposed to arsenic primarily through

-------
                                                 Toxicologies! Data   53

inhalation. Similar  results were  reported by Blom et al. (1985), who
concluded  that workplace exposure  to air levels below 50 /*g/m3 did noc
result in  any clinically significant neuropathy. In a study by Southwick
et al. (1981), no  significant differences in NCV measurements were noced
between a  population consuming water with high  (0.2 mg/L) or low
(0.02 mg/L) levels of  inorganic arsenic.

     Arsenic neuropathy is classified as a distal axonopathy with axonal
degeneration, especially of large  myelinated fibers (Hindmarsh and
McCurdy 1986). Both  sensory and motor neurons are involved, usually bi-
laterally. Recovery  of nerve function following cessation of exposure is
often slow and incomplete (EPA 1984a).

     Chronic exposure  to doses of  up to 10 mg/kg/day (given by injection
once a week for 18 months) did not produce any evidence of neuropathy in
rats (Schaumburg 1980). This exposure level (an average of around
1.4 mg/kg/day) would almost certainly be expected to produce neuropathy,
if not lethality,  in humans. This  study lends further evidence to the
view that  animals are not appropriate models for arsenic toxicity in
humans.

     Dermatological  effects.  Chronic oral arsenic exposure produces a
characteristic group of dermatological manifestations, including
hyperkeratoses on the palms and soles and hyperpigmentation around the
eyelids,  temple, neck, nipples, and groin. Usually the appearance is
mottled,  like raindrops on a dusty road, but in severe cases the
pigmentation may extend broadly over the chest, back,  and abdomen
(Hindmarsh and McCurdy 1986, EPA 1984a. EPA 1987b).  Skin disorders of
this type have been  observed in a  number of epidemiological studies.
with effects being readily apparent in populations consuming drinking
water containing arsenic at levels of 0.4 mg/L or higher (Cebrian et al
1983, Zaldivar 1974, Borgono et al. 1977, Tseng et al. 1968, Huang et
al. 1985). Similar findings of hyperkeratoses and hyperpigmentation were
reported in retrospective studies  of patients who had been treated with
Fowler's  solution (Fierz 1965) and groups of workers exposed to airborne
arsenic in a pharmaceutical plant  (Vatrous and McCaughy 1945) or a
sheep-dip  factory (Perry et al. 1948). As described in greater detail in
the section on careinogenieity (Sect. 4.2.3.6), chronic oral arsenic
exposure  is also strongly associated with increased risk of skin cancer.
The present consensus  is that some hyperkeratinized lesions, which
appear as  small corn-like elevations, may develop into squamous-cell
carcinomas. Areas of hyperpigmentation are not thought to be
precancerous (EPA 1987b).

     Direct effects  on skin and mucous membranes.  Direct dermal contact
with arsenic compounds may result  in local inflammation and vesiculation
(NAS 1977, Zaloga et al. 1985). Chronic inhalation exposure results in
irritation to the mucous membranes of the eyes and nasopharynx (Vallee
et al. 1960). A.hoarse voice is a  common sign in arsenic workers, and a
perforated nasal septum may occur  following prolonged exposures (ACGIH
1986). No  such effects, however, were noted in workers exposed to
0.2 mg/m3  in air (ACGIH 1986).

     Holmqvist (1951) reported evidence that arsenic may act as a
contact allergen, causing an increased response in dermal patch tests of
workers who were chronically exposed to arsenic dusts in a copper

-------
54   Section 4

smelter. However, Wahlberg  and Boman  (1986) reported that neither
Na2HAs04 nor NaAs02  induced contact allergy in guinea pigs.

     Hepatic and renal  injury.  High  oral exposures to arsenic have been
observed to injure both the liver and the kidneys. In the incident
involving infants poisoned  with arsenic-contaminated milk, Hamamoto
(1955) observed swollen liver in all  of the victims, and necrosis and
fatty infiltration were noted at autopsy of infants who died. Similar
signs of hepatic fatty  infiltration,  central necrosis, and eventual
cirrhosis have been  described in patients employing Fowler's solution
(Franklin et al. 1950).  Other studies have reported an association
between chronic oral arsenic exposure and hepatic cirrhos'is and portal
hypertension (Datta  1976, Viallet et  al.  1972, Morris et al. 1974).
Signs of renal injury reported by Hamamoto (1955) included hematuria,
leukocyturia, and glycosuria. Cortical necrosis also has been reported
in some cases (Gerhardt et  al. 1978).

     Beneficial effects.  There are several studies in animals which
indicate that low levels of arsenic in the diet are beneficial or
essential. Schwartz  (1977)  reported that racs fed diets containing
<0.05 ppm arsenite failed to gain weight normally, while animals
receiving 0.5 to 2.0 ppm did gain weight. Anke et al.  (1976, 1978) did
not observe differences in  growth of  goats and minipigs fed a low
arsenic diet (<0.05 ppm), but conception rate was depressed in both
species, and offspring  had  decreased birth weight and elevated perinatal
mortality. More recently, Anke et al. (1987) reported the appearance of
ultrastruetural changes in  the mitochondria of cardiac tissue from goats
fed an arsenic-deficient diet (0.035 ppb).  Uthus et al.  (1983) noted
low-arsenic diets (<0.03 ppm) led to growth depression and decreased
fertility in rats. These workers proposed that arsenic plays a role in
arginine metabolism.

     While these observations suggest that low levels of arsenic (about
0.5 ppm in the diet) may be essential or  beneficial to animals,  several
researchers consider the weight of evidence inadequate to conclude this
with certainty (Solomons 1984, Hindmarsh  and McCurdy 1986).  EPA (1987b)
performed a detailed review of the evidence, concluding that
essentiality, although  not  rigorously established, is plausible.

     If arsenic is essential or beneficial to animals,  then it could be
important to humans as  well. If so,  the daily requirement for humans
probably lies somewhere between 10 and 50 pg/day (NAS 1977.  EPA 1987b).
This level of arsenic intake is usually provided in a normal diet, and
no cases of arsenic-deficiency in humans  have ever been reported.

4.2.3.3  Developmental  toxicity

     Information on the teratogenic and fetotoxic potential of arsenic
is derived mainly from  studies in animals.  Parenteral administration of
10 to 45 mg/kg/day of sodium arsenate to  rats, mice, or hamsters during
gestation has been reported to increase the frequency of a number of
fetal malformations  (Ferm and Carpenter 1968. Perm et al. 1971,  Hood and
Bishop 1972, Burk and Beaudoin 1977. Willhite 1981, Ferm an-. Hanlon
1985).  By the oral route, arsenate is ouch less toxic.  Dose  af
120 mg/kg administered  by gavage co mice  during gestation i,ased
decreased birth weight  and  increased prenatal mortality, but no

-------
                                                 Toxicologlcal Data   55

fecotoxic or teracogenic effects were seen with oral doses up to 100
BgAg (Hood et al. 1977, 1978).

     Similar results have been reported for sodium arsenlte. although
the teratogenlc and fetotoxlc potential of arsenlte appears to be
somewhat greater than for arsenate. Baxley et al. (1981) administered
single oral doses of sodium arsenite to pregnant mice and observed no
discernable teratogenic or maternal toxicity at doses of 20 rag/kg. Doses
of 40 or 45 mg/kg resulted in both maternal lethality and fetotoxicicy
(decreased fetal weight and increased fetal resorptions),  with a low
incidence of gross malformations. Similar findings were reported by Hood
and Harrison (1982) who administered single oral doses of sodium
arsenite by gavage to pregnant hamsters. Doses of 20 mg/kg given on days
9, 10, or 11 produced no significant fetotoxicity, while doses of
25 mg/kg given on day 8 or day 12 lead to Increased prenatal mortality
and decreased fetal weight. Animals exposed on day 8 also showed
evidence of increased fetal malformations, although this was not
significant. The authors concluded that acute oral exposure to arsenite
was not likely to be teratogenic or fetotoxic at doses that are
tolerated by the dam.

     The effects of continuous maternal arsenic exposure on the
developing fetus have not been well studied. Perm and Hanlon (1985)
implanted osmotic minipumps containing sodium arsenate in pregnant
hamsters and found time- and/or dose-dependent decreases in the size and
number of living fetuses in dams receiving doses of around
5 to 9 mg/kg/day. Teratogenicity was associated with the exposure level
on day 8 (the critical stage of embryogenesis), and a dose-dependent
increase in malformations was seen at all doses tested (5 to 9
mg/kg/day).

     Evidence of developmental effects of arsenic in humans is lacking
Epidemiological studies of smelter workers in Sweden (Nordstrom et al.
1978a,b,c,d) provide limited evidence of decreased birth weight and
increased abortion rate in women working in the smelter or living
nearby, but these data are not adequate to implicate arsenic as the
responsible agent.

4.2.3.4  Reproductive toxicity

     The effect of arsenic exposure on reproductive parameters has not
been thoroughly investigated. Schroeder and Mitenner (1971) exposed mice
to drinking water containing 5 mg/L of arsenite  (about 0.7 mg/kg/day)
There were no effects on survival over three generations, and the only
effects noted were an increase in the ratio of males to females in the
F3 generation and a small decrease in the average litter size.

4.2.3.3  Genotoxic ity

     Gene mutation studies.  Arsenic(III) and arsenic(V) compounds have
been tested for gene mutations in a number of systems. Nearly all
results have been negative. One positive report  in bacteria (Nishioka
1975) could not be confirmed (Rossman et al. 1980), and other studies  in
bacteria have been negative (Leonard and Lauwerys 1980, Lofroth and Ames
1978). Arsenite and arsenate were also reported  to be inactive in
gene-specific mutation assays  in yeast  (Singh 1983) and cultured

-------
 56   Section A

 mammalian cells (Amacher and Patllet  1980, Rossman et  al.  1980,  Oberly
 ec al.  1982. Lee et al.  1985).  After  reviewing  the data,  EPA believes
 the weight of evidence supports the conclusion  that arsenic is either
 inactive or extremely weak for  induction of gene mutations  (Jacobson-
 Krara and Montalbano 1985) .

      Cytogenetic studies.   In contrast  to the negative results in  gene
 mutation tests,  both arsenate and arsenite have been found  to result in
 chromosome aberrations and sister chromatid exchanges  (SCEs) in  cultured
 animal  and human cells tested in vitro  (Jacobson-Kram and Montalbano
 1985).  For example,  Larramendy  et al. (1981) incubated Syrian hamster
 embryo  cells and human peripheral lymphocytes with concentrations  of
 around  10'5 H NaAs02 and Na2HAs04 and observed highly significant
 increases in chromosome  aberration frequency with both compounds in both
 cell types.  Arsenite was about  tenfold more effective than arsenate
 Dose-related increases in  SCE were also reported. Similar results  have
 been reported by a  number  of  other groups for a variety of arsenic(III)
 and arsenic(V)  compounds (Nakamuro and Sayato 1981, Nordenson et al
 1981, Wan et al.  1982, Lee  et al. 1985).

      Several studies have  described a positive association between
 arsenic  exposure and chromosome  aberrations or SCEs in humans.  For
 example,  Burgdorf et al. (1977)  reported about a threefold increase in
 SCE in  lymphocytes  from  patients who had taken Fowler's solution for
 periods  of 4 to  27 years,  although no difference in the frequency  of
 chromosomal  aberrations  was noted. Conversely,  Nordenson et al.   (1979)
 reported increased  frequencies of chromosomal aberrations but similar
 frequencies  of  SCE  in patients who had been exposed to total doses of
 300 to  1200  mg of arsenic  taken  in Fowler's solution over many years
 However,  these and other similar studies in humans are limited by
 methodological difficulties,  small sample numbers,  and likely exposure
 of  subjects  to other clastogenic agents (EPA 1984a),  so these results
 must be  interpreted  with caution. No increase in chromosomal aberrations
 or  SCE was  observed  in a recent  study of residents in Fallen.  Nevada.
 where drinking water contains about 0.1 mg/L of arsenic (Vig et al.
 1984). Poma  et al.  (1981) examined bone marrow cells  and spermatogenia
 from mice  treated with 4 to 12 mg/kg of As203 and observed no
 significant  increase  in  chromosomal aberrations in either cell type

 4.2.3.6  Carcinogenicity

     Skin cancer.  There is clear evidence that chronic oral exposure co
 elevated levels  of arsenic  increases the risk of skin cancer.  The most
 common cancerous  lesions are  squamous cell carcinomas which appear to
 develop  from some of  the hyperkeratinized corns described earlier.
 although most of  these corns  remain benign for decades. In addition,
 multiple basal cell  carcinomas may occur,  typically arising from cells
 not associated with  hyperkeratinization. Although dermal lesions of chis
 sort can be  removed  surgically,   they may be fatal if  left untreated.

     The largest  study of arsenic-induced skin cancer was described by
Tseng et al. (1968).  The study focused on a large population in Taiwan
where arsenic levels  in  deep wells used for drinking  water ranged  from
0.001 to 1.82 mg/L, with average levels  of around 0.4 to 0.6 mg/L.  Based
on  examination of over 40.000 people In the area,  the skin cancer  race

-------
                                                 Toxicological Daca   57

was found to be  10.6/1000. Typical arsenic blood levels In residents of
this area were around  15 ng/L  (Heydorn 1970), somewhat higher than
typical values (2  to 5 ^g/L) in Denmark or the United States (Heydorn
1970, Valentine  et al. 1979).  There was a strong relationship between
the occurrence of  skin cancer  and other signs of arsenic intoxication
(hyperpigmentation, hyperkeratinization, and Blackfoot disease) .  and
skin cancer incidence was correlated with arsenic levels in the water
No cases of skin cancer or other signs of arsenic poisoning were
reported in a control population of 7500 people consuming water with low
arsenic content  (<17
     The relevance of this study to skin cancer risk in the United
States occasionally has been questioned, based on concerns that there
may have been significant exposure to arsenic from sources other than
the well water (EPA 1987b) , and that the dietary and socloeconomic
characteristics of the exposed population are quite different from those
of average U.S. citizens (EPA 1984a) . Although these considerations may
call the precise dose-response relationship observed in this study into
question, they do not alter the conclusion that chronic arsenic
ingestion is associated with increased risk of skin cancer.

     Cebrian et al. (1983) reported a 3. 6 -fold elevation in the
incidence of ulcerative skin lesions (compatible with a diagnosis of
epidermoid or basal cell carcinomas) in the residents of a Mexican town
where drinking water contained 0.4 mg/L of arsenic, compared to a
similar town where water contained 0.005 mg/L.  Similarly, elevated
incidences of skin cancer have been reported in studies of humans who
had used Fowler's solution as a medicine (Scanners and McManus 1953,
Fierz 1965, Cuzick et al. 1982).

     Several epidemiological studies performed in the United States have
not detected an increased frequency of skin cancer in small populations
consuming water containing arsenic at levels of around 0.1 to 0.2 mg/L
(Goldsmith et al. 1972, Morton et al. 1976, Harrington et al. 1978,
Southwick et al.  1981). These studies suggest that arsenic-associated
skin cancer is not a common problem in this country, but they lacked
sufficient statistical power to conclude that arsenic exposures of this
sort do not increase the risk of skin cancer (Andelman and Barnett
1983).

     Internal cancers.  A number of studies suggest that the incidence
of some types of internal malignancies may also be increased by chronic
oral exposure to arsenic (EPA 1987b) , but the data are not adequate to
draw a firm conclusion (EPA 198&a, Philipp 1985). For example, Sommers
and McManus (1953) reported that 10 of 27 patients with skin cancer
following arsenic exposure also had an internal cancer. A similar
increase in the incidence of internal cancer in patients with arsenical
keratoses was noted by Reyraann et al. (1978). Additional support for an
association between arsenic exposure and internal tumors was reported by
Dobson et al. (1965), who noted that palmar keratoses (indicative of
arsenic poisoning) were common in patients with internal tumors.

     Several types of internal tumor have been observed in association
with oral arsenic exposure. A number of studies have noted that hepatic
angiosarcoma, a rare tumor in the general population, occurs at
increased frequency in persons exposed to Fowler's solution  (Regelson

-------
 58    Section 4

 et  al.  1967;  Lander  et al.  L975;  Falk  et al.  1981a,b; Roat et  al.  1982;
 Kasper  et  al. 1984).  and Roth (1958) also noted angiosarcoma in vintner
 exposed to arsenical  pesticides.  Chen  et al.  (1985) studied the
 correlation between  mortality from  several types of internal cancer and
 exposure to arsenic  through ingestion  of water containing 0.35 to  1.14
 mg/L of arsenic.  The  standardized mortality ratios (SMR) for bladder,
 kidney,  lung, and liver tumors were 1100, 772, 320, and 170 in males and
 2009, 1119,  413,  and  229 in females, respectively. A positive dose-
 response relationship with  arsenic  exposure was noted for bladder, lung,
 and liver  cancer.  In  a follow-up  case-control study, Chen et al. (1986)
 found that the  odds  ratio for bladder, lung, and liver cancers for those
 who had used well water containing arsenic for 40 years of more were
 3.90, 3.39,  and 2.67,  respectively, compared to people who had never
 used arsenic-contaminated well water.

      Limited  observations suggest arsenic might be associated with
 several  other types of cancer as well, including cancer of the mammary
 gland (Knoth  1966), cancer  of the lymphatic tissues (Ott et al. 1974),
 leukemia (Axelson et  al.  1978), and renal adenocarcinoma (Sommers and
 NcManus  1953).

      Lung  cancer.  A  number  of epidemiological studies have been
 performed  to  determine if there is an association between inhalation
 exposure to arsenic and increased risk of lung cancer.  Although some of
 these studies have not detected an association (Snegireff and Lombard
 1951, Pinto and Bennett 1963, Nelson et al.  1973,  Greaves et al.  1981,
 Rom  et al.   1982), most have  observed an above-average incidence of lung
 cancer in  exposed populations (OSHA 1983,  EPA 1984a,  EPA 1986a).  As is
 often the  case,  these  studies are to some degree limited by confounding
 factors  such as smoking and  exposure to other chemicals;  however,  the
 weight of  evidence that arsenic is a risk factor for lung cancer is,
 nevertheless, convincing.

     Lee and Fraumeni  (1969)  reported an exposure-dependent increase In
 frequency  of respiratory  cancer in workers in a large copper smelter in
Montana. Observed SMR  values were 239,  478,  and 667 in low,  medium, and
high exposure groups,  respectively.  Based on industrial hygiene reports,
 the average levels of  air concentration for these three groups  were 0.4,
 7, and 62  mg/m3 (Lee-Feldstein 1986).  Follow-up studies of this cohort
 (Lubin et  al. 1981, Higgins  et al. 1982,  Welch et al.  1982,  Lee-
Feldstein  1983,  Brown  and Chu 1983b) obtained additional evidence of a
 level- and duration-dependent increase in lung-cancer risk.

     Similar results have been observed in workers at another large
copper smelter In Tacoma, Washington.  Studies by Pinto et al.  (1977.
 1978) and  Enterline and Marsh (1982) npted that death rates from lung
cancer In  this cohort  were several times  higher than expected.  Airborne
concentrations of arsenic ranged from 3 to 295 Mg/m3 (averaging around
 53 pg/m3),  however, urinary  arsenic excretion levels were Judged to be
 the best measure of exposure  in each individual.  A time-weighted index
of cumulative exposure  calculated from urinary arsenic excretion levels
was linearly related to  lung cancer mortality and ranged from an SMR of
 111 at the  lowest cumulative exposure to  one of 832 at the highest
cumulative  exposure. Based on a reanalysis of the exposure data,
Enterline  et al.  (1987) concluded chat the risk of lung cancer

-------
                                                 Toxicological Daca   59

associated with airborne arsenic is even greater than previously
believed, with SMR values >200 at a cumulative exposure of 10 mg/m-*-
years.  The dose-response relationship was concaved downward, suggesting
that risk at low levels of exposure is greater than expected based on
observations at high levels of exposure.
     Increased lung cancer rates in workers at other smelters have been
described by Rencher et al. (1977) and Axelson et al.  (1978). Similar
elevations in respiratory cancer rates have been noted in other types of
occupational exposure involving arsenic dusts or fumes (Ott et al  197&,
Blejer and Wagner 1976, Mabuchi et al. 1979, Baetjer et al. 1975, Roth
1958).
     Several epidemiological studies suggest that there is also an
increased risk of lung cancer for nonoccupationally exposed individuals
living within several kilometers of arsenic-emitting industries.
Matanoski et al. (1981) noted an elevated lung cancer rate in male
residents around a large pesticide plant in Baltimore. Brown et al.
(1984)  reported a 60% increase in lung-cancer incidence in the vicinity
of a smelter in Pennsylvania, and Pershagen (1985) reported a relative
risk of 2.0 for lung cancer in men who lived within 20 km of a large
copper smelter in Sweden.

     Carcinogenicity studies in animals.  Most attempts to induce tumors
in laboratory animals following oral exposure to arsenic have been
inconclusive or negative (EPA 1986a). The reasons for this inability to
observe a clear carcinogenic response in animals are not known, but
these studies are not considered to refute the positive associations
between exposure and cancer observed in humans.
     Some animal studies have produced suggestive evidence for arsenic -
induced lung cancer. In most of these studies, animals were exposed to
arsenic by intratracheal instillation and then observed for their
lifetimes. Ivankovik et al. (1979) reported that a single intratracheal
instillation of an arsenate-containing pesticide mixture caused lung
tumors in rats, and Ishinishi et al. (1983) reported that intratracheal
instillation of arsenic trioxide for 15 weeks caused lung tumors in
hamsters. Pershagen et al. (1984a,b) reported that intratracheal
instillation of As203 along with a carrier dust (charcoal carbon) and
sulfuric acid produces pulmonary carcinomas in hamsters. More recently,
Pershagen (1985) and Pershagen and Bjorklund (1985) reported that
intratracheal instillation of calcium arsenate alone causes an increase
in lung tumors in hamsters.
     Mechanism of carcinogenicity.  Compounds which are carcinogenic but
which do not cause gene damage directly may do so indirectly through a
variety of mechanisms  (Barrett and Shelby 1986),  including  inhibition of
one or more enzymes involved in DNA replication or repair.  Several
studies provide evidence that this may be the case with arsenic
(Nordberg and Anderson 1981, Rossman 1981, Jacobson-Kram and Montalbano
1985, Okui and Fujiwara 1986). This mechanism of  action is  also
consistent with the view that arsenic acts primarily as a promoter of
lung cancer rather than as an initiator (Hindmarsh and HcCurdy  1986)   In
addition, Lee et al. (1986)  found that although arsenic alone  is a very
weak mutagen, it greatly increases the mutagenicity of other direct-

-------
 60   Section 4

 acting mutagens  (UV radiation,  alkylating agents, cross-linking agents
 etc.).

      Another possible mechanism of arsenic-induced carcinogenicity  is
 incorporation of arsenate into  DNA in place of phosphate. This concept
 is consistant with observations that arsenate must be present during DNA
 synthesis  in order to be  effective. It would explain why arsenic  is
 clastogenic  (the arsenate-phosphate bond would be weaker than the normal
 phosphodiester)  but does  not  cause gene mutations (Jacobson-Kram  and
 Montalbano 1985).

 4.2.4  Interactions with  Other  Chemicals

      Arsenic is  known to  interact with selenium, and the relationship
 between these two  chemicals has been studied both in vivo and in vitro
 Most  studies indicate that arsenic and selenium are mutually
 antagonistic,  each chemical reducing the effects caused by the other.
 For example,  high  levels  of selenium in the diet are toxic to livestock
 and poultry,  and addition of  arsenic to the diet or to drinking water
 reduces  the  extent of the selenium-induced injury (Levander 1977,  Cabe
 et al.  1979,  Hill  1975, Sky-Peck 1985).  Likewise, low doses of selenium
 have  an  anticarcinogenic  effect in animals and humans, and concomitant
 arsenic  exposure reduces  this beneficial effect of selenium and
 increases  tumor  rates (Schrauzer 1987,  Schrauzer et al. 1978). With
 respect  to the effect of  selenium on arsenic toxicity, selenium has been
 shown to protect against  arsenic-induced chromosome aberrations and
 sister chromatid exchanges in cultured human lymphocytes (Beckman and
 Nordenson  1986,  Sweins 1983), against arsenic-induced cytotoxicity in
 bovine pulmonary macrophages  (Fischer et al. 1986),  and against
 arsenic-induced  teratogenesis in hamsters (Holmberg and Perm 1969).
 Gerhardsson  et al.  (1985) noted that selenium had a possible protective
 effect against lung cancer in smelter workers exposed to a variety of
 airborne carcinogens,  including arsenic.  The mechanism by which arsenic
 and selenium influence each other is not known,  but each chemical  tends
 to increase  the biliary excretion of the other,  suggesting that they may
 react in the  liver to form a conjugate  (Hill 1975,  Levander 1977).

      Some  interactions between  arsenic  and other chemicals have been
 documented.  Pershagen et  al.   (1981) reported that smoking and arsenic
 inhalation had a multiplicative effect  on lung cancer mortality in
 smelter workers. There are also evidences of a positive interaction
 between arsenic and benzo(a)pyrene in induction of lung tumors in
 hamsters (Pershagen et al. 1984a).  Mahaffey and Fowler (1977) reported
 that weight  gain In young rats was reduced more by cadmium and arsenic
 given together than expected by  the sum of the effects of each agent
 alone. In a  follow-up study.  Mahaffey et al. (1981)  found that
 interactions between several metallic elements affect both toxicity and
 tissue concentrations of  the metals.  Lead and arsenic in the diet
 produced an  additive  effect on  coproporphyrln excretion in male rats,
 and arsenic  produced a significant increase in renal copper
 concentration. In  addition, cadmium exposure reduced tissue
 concentrations of  arsenic. The mechanisms of these interactions have noc
been determined.

-------
                                                 ToxicoLogical Data   61

     In a study of classroom behavior. Marlowe et al. (1985) reported
that arsenic and other toxic metals increased the neurotoxic effects of
lead in children as measured by aggressive behavior in the classroom and
reading and spelling achievement. Interaction of arsenic and aluminum
also increased aggressive behavior. Arsenic in the diet of rats has been
reported to have a slight goiterogenic effect, causing increases in
thyroid weight accompanied by decreases in iodine concentration in the
thyroid, even with increased iodine intake (Sharpless and Metzger 1941,
NIOSH 1975) .

4.3  ORGANIC ARSENIC

4.3.1  Overview

     Arsenic can form stable bonds with carbon, forming a large number
of organic derivatives. The chemical and physical properties of the most
common organic derivatives of arsenic are summarized in Table 3.4.  As
with inorganic arsenic, there are significant toxicokinetic and
toxicological differences between various organic arsenicals and animal
species as well. From the perspective of likely exposure and potential
risk to human health, four groups of derivatives merit special
attention:

   • Methyl derivatives of arsenic acid. The most important members of
     this group are MMA,  DMA, and their salts. These compounds have been
     widely used as pesticides (primarily as herbicides and defoliants)
     In the early part of this century, DMA and its sodium salt were
     employed as medicine for a variety of diseases, including syphilis
     By the oral route, these compounds produce symptoms of
     gastrointestinal irritation and renal and hepatic injury similar to
     some of the effects produced by inorganic arsenic,  but the potency
     of these methyl derivatives is much lower. As described above (see
     Sect.  4.2.2 on toxicokinetics),  the methyl derivatives are also
     formed in the body by metabolism of inorganic arsenic, and this is
     generally viewed as a detoxification pathway.

   • Phenyl derivatives of arsenic acid. Several phenyl arsenates (e.g.,
     arsanilic acid) and their salts have been widely used as feed
     additives to improve weight gain and prevent enteric disease in
     poultry and swine. The mechanism of these beneficial effects is noc
     known but may be due to effects on intestinal microorganisms rather
     than a direct effect on the animal. Exposure of animals to high
     levels of phenylarsonate compounds results primarily in sensory and
     peripheral nerve injury. The toxicity of these compounds in humans
     has noc been investigated extensively.

   • "Fish arsenic." Fish and shellfish often accumulate rather high
     tissue levels of arsenic. Depending on the species, most of this
     accumulation exists in the form of arsenobetaine or arsenocholine
     Available data indicate that these organic derivatives have low
     toxicity, and ingestion of arsenic in this form is not generally
     considered to be of health concern.

-------
62   Seccion 4

    • Methyl derivatives of arsine. Arsine (not a true organic arsenical
     but considered here  for convenience) and its methyl derivatives ar
     not widely used  in industry but might be encountered at some
     hazardous waste  sites. In addition, arsine and its methyl
     derivatives may  be formed in the environment from other arsenic
     compounds by the action of microbial organisms or by inadvertent
     chemical reactions that generate strong reducing conditions. These
     compounds are highly irritating gases or volatile liquids and
     possess varying  degrees of hemolytic potential.
4.3.2  Toxicokinetics

4.3.2.1  Methanearsonates

     Both MMA and DMA are well absorbed in animals and humans.  Buchet et
al. (1981a) reported  that gastrointestinal absorption was at least 75 to
80% in humans given oral doses of MMA or DMA. Similar values (about 70%)
have been reported for DMA in rats (Stevens et al.  1977b) and hamsters
(Yamauchi and Yamamura 1984).  The extent (92%) and rate (half-time of 2
min) of DMA absorption following intratracheal instillation in rats is
somewhat greater than in gastrointestinal absorption (Stevens et al.
1977b).

     Stevens et al.  (1977b) reported that administration of DMA to rats
initially lead to high concentrations in whole blood,  muscle, kidney,
liver, and lung. Soon thereafter, however,  levels decreased rapidly in
all tissues except in whole blood. Vahter et al.  (1984) administered DMA
orally to rats and mice and observed that the highest initial
concentrations of arsenic were found in the kidney,  lung,
gastrointestinal tract, and testes.  Tissues showing the longest
retention of arsenic  were lung, thyroid, intestinal wall, and lens.
Marafante and coworkers (1985) reported high concentrations of DMA in
the lung after intraperitoneal injections to rabbits,  and attributed
that to rapid uptake  by the lung followed by rapid clearance.

     Both MMA and DMA are excreted primarily in urine.  Buchet et al.
(1981a) administered  oral doses (500 pg arsenic)  of MMA and DMA to human
volunteers. After four days,  the excretion of urinary arsenic was 78 and
75% of'the dose of MMA and DMA, respectively. Analysis of the excretion
productions revealed  that DMA was excreted unchanged,  while 13% of the
MMA was methylated to DMA. In rats,  DMA was not excreted in feces after
intravenous administration, and fecal excretion of arsenic after oral
administration of DMA probably resulted from incomplete absorption
(Stevens et al.  1977a).

4.3.2.2  Phenylarsenates

     Phenylarsonic compounds are not well absorbed from the intestinal
tract in humans and animals (NAS 1977).  Following oral administration of
74As-arsanilic acid to human volunteers, 74% was excreted in the feces
(Calesnick et al.  1966).

     Phenylarsonates  do not appear to undergo significant metabolism in
the body.  Cristau et  al. (1975) administered sodium arsanilate to rats
and guinea pigs, detecting no arsenic-containing compounds in the urine
except the parent arsanilate.  Similar evidence indicates that chickens

-------
                                                 ToxicologLcal Data   63

do not metabolize  arsanilic  acid,  acetylarsonic  acid, or  3-nitro-4-
hydroxy-phenylarsonic  acid (roxarsone)  (Moody  and Williams  1964a,b).

     Fhenylarsonates are  not retained  in  the body but are excreted in
urine. Following parenteral  administration of  phenylarsonic compounds to
animals, most  of the dose is excreted  in  the urine within 24 to 48 h
(NAS 1977).  In chickens fed  50 ppm of  roxarsone, only slight increases
in arsenic levels  could be detected in skin, muscle, liver, and kidney
(Buck et al. 1973).

4.3.2.3  Fish  arsenic

     In humans, gastrointestinal absorption of organic.arsenic in fish
and seafood  is at  least 70%  (Vestoo and Rydalv 1972, Charbonneau et al.
1980b). Tarn  et al.  (1982)  reported that in humans, only 0.3% of the
arsenic ingested in fish  was  excreted  in  the feces, indicating
essentially  complete absorption. Similar  results have been obtained in
animals. Charbonneau et al.  (1978b)  reported that monkeys fed a
homogenate of  fish absorbed  over 90% of the arsenic present. Vahter et
al. (1983) and Marafante  et  al. (1984)  synthesized radiolabeled
arsenobetaine  and  arsenocholine and reported that 90 to 98% of oral
doses given by gavage  to  mice or rats was absorbed.

     Both arsenobetaine and  arsenocholine are  rapidly excreted in the
urine.  Tarn et  al.  (1982)  reported  about 50% of the arsenic in fish was
excreted in urine  by humans within 1 day, and  Luten et al. (1982)
reported 70  to 85%  urinary excretion of fish arsenic by humans within
5 days. Vahter et  al.  (1983)  reported that over 99% of an intravenous
dose of arsencbetaine  in  mice was  cleared within 4 days, with a half-
time of 12 h. A small  fraction of  the dose (about 0.2%) was retained and
cleared within a half-time of about 60  days. Only unchanged
arsenobetaine  could be detected in urine  (Vahter et al. 1983,  Cannon et
al. 1983,  Kaise et  al. 1985).

     Arsenocholine  is  cleared somewhat  more slowly than arsenobetaine.
with a half-time of about 12 days  in mice. The main urinary product is
arsenobetaine,  and metabolic conversion of arsenocholine to
arsenobetaine  appears  to  be  the rate-limiting  step (Marafante et al.
1984)'.

4.3.2.4  Arslna and methylarsines

     No quantitative data on  the toxicokinetics of arsine or
methylarsines were  located. The principal exposure route of concern is
inhalation, and it  seems  likely that these compounds would be well
absorbed from  the  lung. Once arsenic is inhaled, it subsequently breaks
down, releasing inorganic  arsenic  into  the blood (NIOSH 1979). Elevated
levels of arsenic  have been observed in the tissues and urine of humans
following arsine exposure  (ACGIH 1986).

4.3.3  Toxicity

4.3.3.1  Methanearsonates

     Lethality.  Acute oral exposure of animals to methanearsonates
results mainly in  signs of gastrointestinal irritation (diarrhea and

-------
 64   SecCion 4

 vomiting), accompanied by  listlessness or hematuria  (Palmer 1972).  In
 extreme cases, sequelae may  include stupor, convulsions, paralysis, and
 death  (Weed Society Handbook 1967). In rats, estimates of the acute oral
 LD50 for DMA, its monosodium salt  (NaDMA), and the monosodium salt of
 MMA (NaMMA) range between  600 and  2600 mg/kg. Calves appear to be more
 sensitive than rats, with  acute oral LD5Q values for these compounds
 ranging from 100 to 450 mg/kg (Hood 1985). Available evidence is not
 adequate to determine  if humans are more sensitive than animals to these
 organic arsenic compounds.

     Acute inhalation  exposure to methanearsonates usually produces only
 mild signs of respiratory  distress (Stevens et al.  1977b).  Exposure of
 rats to aerosols of DMA for  2 h caused labored respiration,  rhinorrhea,
 irritation of the eyes,  and  diarrhea,  with an estimated LC5Q of
 4300 mg/m3. The estimated  acute inhalation LC50 for DMA in mice was
 >6400 mg/m3 (Stevens et al.  1977b).

     Dermal exposure of rabbits to NaDMA at a dose of about 500 mg/kg
 did not cause lethality (IBT  1976), while application of 2800 mg/kg of
 NaMMA caused death within  2  to 4 days (Nees 1969).

     Systemic effects.   In humans exposed to methanearsonates in
 occupation-related incidents, symptoms most often reported were
vomiting,  diarrhea, abdominal pain, eye irritation, and dermatitis.  No
 long-lasting sequelae  of these exposures were observed (Peoples et al
 1979).

     Early in this century, NaDMA was investigated as a medicine for
syphilis,  skin disease,  tuberculosis,  and anemia (Simon 1932).  Typical
doses were 25 to 150 mg/day.  Clinical experience eventually led to the
conclusion that DMA was  ineffective in this role.  Adverse symptoms were
 rarely observed after  oral dosing, although intravenous administration
 led to renal injury in a few cases (Cole 1916).

     Data on methylarsonate toxicity in animals are mostly  derived from
studies in cattle,  sheep,  and chickens.   Palmer (1972)  conducted an
extensive investigation of the toxicity of DMA.  NaMMA,  and  the  disodium
 salt of MMA (Na2MMA) in  these species and found that multiple doses of 5
 to 10 mg/kg/day were without  ill effects in cattle  and sheep.  Higher
doses often resulted in diarrhea,  anorexia,  and weight loss.  In
chickens,  doses of 100  to  250 mg/kg/day did not affect normal weight
gain.  In a 90-day feeding  trial in rats and dogs,  no effects were
observed at dietary levels of 30 ppm (roughly 1 mg/kg/day)  of DMA or MMA
 (Ansul Co.  1971). Siewicki (1981)  fed rats diets containing 42  ppm
 (about 2 mg/kg/day) of  DMA for 42  days and observed no effects  on body
weight, organ weight, hematology,  or urinary excretion of ALA or
coproporphyrin,  even though significant levels of arsenic were  present
 in liver,  spleen, and  erythrocytes.

     Developmental and  reproductive effects.   Oral  administration of
high doses (40 to 100  mg/kg)  of DMA has been reported to cause
 fetotoxicity (reduced  fetal weight, delayed ossification,  and cleft
palate) in mice,  but this was associated with maternal  toxicity and
mortality (Chernoff and Rogers 1975).  In a later study, Rogers  et ai.
 (1981)  reported maternal toxicity  and skeletal and/or palate anomalies
 in fetuses following oral  administration of DMA to  mice (400 mg/kg/day)

-------
                                                  Toxicologies! Data   65

 or rats (30 mg/kg/day).  Developmental defects were observed at doses
 below those that appeared to cause maternal toxicity in the rat,  but  noc
 in the mouse (Hood 1985). Harrison et al.  (1980)  and Hood (1985)
 reported maternal deaths and fetal malformations  in hamsters after
 intraperitoneal administration of NaDMA at 900 to 1000 mg/kg and  Na2MMA
 at 500 to 1500 mg/kg.  These doses are as much as  100 times greater than
 the doses of inorganic arsenic required to cause  developmental effects
 in mice (Hood and Bishop 1972, Hood 1972).

      Mutagenicity/carcinogenicity.   Results of mutagenicity tests for
 DMA have been negative in bacterial systems (Simmon et al.  1977,
 Andersen et al.  1972,  Felkner 1980).  In yeast (S.  cerevisiae),  DMA
 yielded positive results in tests involving gene  conversion,  reverse
 mutation,  and mitotic  crossing over (Mortelmans et al.  1980,  Simmon et
 al.  1977).

      Innes  et al.  (1969) administered DMA  orally  to mice  at
 46.4 mgAg/day for three weeks and then at 121 ppm in the diet (about
 20 mgAg/day)  for 18 months.  Histopathologic examinations did not reveal
 any evidence  of increased tumor frequency.  In vitro cell  transformation
 tests with  DMA and NaMMA also gave  negative results (Moore  1976)
 Johansen et al.  (1984) exposed partially hepatectomized rats  to DMA in
 drinking water (80 mg/L,  corresponding to  about 20 mg/kg/day)  for six
 months and  observed no increase in  hepatic tumors  or preneoplastic  foci
 There was a suggestive increase in  tumors  and foci in animals  treated
 with diethy1  nitrosoamine (DENA)  as an initiator,  indicating  that DMA
 might have  promoted activity,  but the  data were too limited to draw firm
 conclusions.

      It  should be  remembered  that most carcinogenicity  tests of
 inorganic arsenic  in animals  have been negative, even though  inorganic
 arsenic  appears  to  be a  human carcinogen by both the  oral and  inhalation
 routes. Hood  (1985) concluded that  available data  are  inadequate  to
 assess the human carcinogenic potential  of the  methylarsonates.

 4.3.3.2  Phenylarsonates

      Lethality.  Phenylarsonic  compounds appear to  be more  toxic  than
 methanearsonates. Reported oral LD50 values  range  from  44 to 216  mg/kg
 for  the rat, and subcutaneous  LDso  values  were  75  mg/kg for the rat and
 400 mg/kg for  the mouse.  Intraperitoneal LDso values  for  the rat  range
 from  18.8 to 66 mg/kg (RTECS  1983,  Kerr  et  al.  1963). Symptoms of acute
 poisoning include  loss of coordination,  inability  to  control limb
movements, ataxia, blindness,  and paralysis  (NAS 1977).

     Systemic  effects.   Arsanilic acid derivatives  have been used as
human  therapeutic agents  to treat syphilis  and  trypanisomiasis. Adverse
 effects that have been noted  in association  with this use include opcic
neuritis and loss of peripheral vision,  similar to  that observed  with
other organometals such as organotin or methyl  mercury. In  some cases,
 injury may progress to loss of central vision and  optic atrophy (Crane
 1974, Potts and Gonasun  1980)

     Toxic effects of chronic excess phenylarsonate  ingestlon  in  swine.
 include blindness, partial paralysis of  the  extremities, and poor weight
gain  (NAS 1977). Histopathologic  changes observed  include optic nerve.

-------
66   Section 4

optic tract, and peripheral nerve damage (Buck et al.  1973). No toxic
effects were reported in pigs administered 100 ppm arsanilic acid in the
diet (about 4 mg/kg/day) for three generations (Frost et al. 1962).
However, chronic poisoning of swine has been reported after
administration of 100 ppm roxarsone in the diet for two months (Buck
1969). Kennedy et al. (1986) reported that pigs fed a diet containing
187 ppm of roxarsone (about 7 mg/kg/day) developed a nervous disorder in
10 days. The main symptom was clonic convulsive episodes brought on by
exercise. This was accompanied by histological evidence of myelin and
axonal degeneration. Similar doses of arsanilic acid did not produce
this effect (Rice et al. 1985).
     The mechanism of this neurotoxicity is not known, but it is evident
that roxarsone affects copper metabolism in exposed animals. Roxarsone
decreases levels of copper in liver and other tissues both in swine
(Edmonds and Baker 1986) and chickens (Czarnecki et al. 1984a).  At
high-dose levels, roxarsone and copper Interact synergistically, causing
more weight loss and toxicity than either agent alone (Czarnecki and
Baker 1984, Edmonds and Baker 1986). Concomitant feeding of roxarsone
and cysteine also increases toxicity, perhaps by a reduction of arsenic
from the pentavalent to the more toxic trivalent form (Czarnecki et al.
1984b).
     Reproductive/developmental effects.  No reports on reproductive or
developmental effects of phenylarsonates were located.
     Mutagenicity/carcinogenicity.  Only one study was located which
evaluated the carcinogenic potential of arsanilic acid. Frost et al.
(1962) found no increase in the tumor incidence of rats fed 100 ppm
arsanilic acid in the diet for 116 weeks.

4.3.3.3  Fish arsenic
     Arsenobetaine is the principal organic arsenic compound contained
in the flesh of fish, shellfish, and crustaceans. Studies on this
compound indicate that it has very low oral toxicity.  Cannon et al.
(1983) administered arsenobetaine by intraperitoneal injection to mice
Doses up.to 500 mg/kg produced no symptoms of toxicity. Similarly, Kaise
et al. (1985) administered oral doses of arsenobetaine as high as 10,000
mg/kg to mice and observed no toxic symptoms. No evidence of
mutagenicity was detected with arsenobetaine in the Ames test or the
hypoxanthine guanine phosphoribosyl transferase (HGPRT) forward mutation
assay, and no increase in SCE was observed in Chinese hamster cells
exposed Co concentrations as high as 10,000 mg/L (Cannon et al. 1983,
Jongen et al. 1985).

4.3.3.4  Arsine and methylarsines

     Arslne is a colorless gas that is a powerful local and pulmonary
irritant. High concentrations (10 ppm or above) of this gas may be
lethal within hours (ACGIH 1986). The chief health concern is hemolysis
of red blood cells (NIOSH 1979. Sittig 1985). The characteristic
symptoms of arsine poisoning include discolored urine  (hemoglobinurea),
jaundice, and anemia, accompanied in some cases by renal damage
secondary to clogging the nephrons with hemolytic debris (NIOSH 1979,
Sittig 1985).

-------
                                                 Toxicological Daca   67

     The hemolytic effects appear  to be  related primarily to a marked
reduction in reduced glutathione levels  and oxidation of hemoglobin in
red cells (Foa and Bertolero  1983, Sax 1984).  Intermittant exposure co
concentrations of 0.5  to 2.0  ppm have been observed to cause blood
effects in a few weeks  (ACGIH 1986). An  average concentration of 0 5 ppm
(0.2 mg/m3) is considered acceptable in  the workplace (NIOSH 1979. ACGIH
1986).

     Like arsine, the methylarsines (monomethylarsine, dimethylarsine,
and crimethylarsine) are strong irritants, but these compounds are less
powerful than arsine as hemolytic  agents  (NIOSH 1983). No other
quantitative information on dose-response relationships for the
methylarsines was located.

4.3.3.5  Summary

     In general, organic derivatives of  arsenic are less toxic than
inorganic forms. The apparent  order of toxicity is phenylarsonates >
methylarsonates > fish arsenic (arsenobetaine). The most characteristic
effect of the phenylarsonates  is neurotoxicity, with effects occurring
at oral exposure levels of about 4 mg/kg/day or higher. Hethylarsonaces
are primarily associated with  irritation of the gastrointestinal tract
or the skin, usually at exposure levels  in excess of 10 mg/kg/day.
Arsenobetaine has not been found to cause toxicity in animals even at
very high doses (10,000 mg/kg). Arsine is a powerful hemolytic agent
that breaks down to inorganic  arsenic in the body. The methylated
derivatives of arsine are less toxic than arsine itself.

-------
                                                                      69
                5.   MANUFACTURE,  IMPORT,  USE,  AND DISPOSAL

 5.1  OVERVIEW

     Arsenic  is  produced  primarily  as  a  by-product  from  the operation of
 nonferrous  smelters.  Currently,  there  are  no  producers in  the United
 States, and all  raw materials  for the  production of arsenic-containing
 products must be imported.  The major uses  of  arsenic  in  this country are
 as wood preservatives or  as agricultural products.  Due to  recent
 regulatory  activities by  EPA's Office  of Toxic Substances  and Office of
 Pesticide Programs, these uses will probably  be  restricted or curtailed
 in the future.

 5.2  PRODUCTION

     Arsenic  is  produced  commercially  (primarily as a by-product) from
 the flue dust of copper and lead smelters. Arsenic  trioxide is
 concentrated  in  these dusts, which are roasted with pyrite or galena to
 yield an arsenic trioxide that is 90 to  95% pure. The trioxide is
 reduced with  carbon to produce arsenic metal  (EPA 1982b).

     In recent years,  arsenic  trioxide was produced in the United States
 only at the ASARCO smelter  in  Tacoma, Washington. Annual production was
 7300 megagrams (Mg) in 1983, but production decreased to 2200 Mg in
 1985.  In 1985, the ASARCO smelter ceased operation, and arsenic is no
 longer produced  in this country  (Bureau  of Mines  1988).

 5.3  IMPORT

     In 1979,  the United States  imported 8940 Mg  of arsenic in the form
of arsenic metal or inorganic  arsenic compounds  (EPA 1982). By 1985,
 imports had risen to  19,000 Mg and increased  to nearly 28,000 Mg in 1986
after domestic production ceased. Thus,  the United States  is completely
dependent on  foreign  suppliers for arsenic (Bureau of Mines 1988).

5.4  USES

     The United  States made use  of approximately  23,000 Mg of arsenic in
1987.  Most of this (74%) was used in wood preservatives, with 19% used
in agricultural  chemicals (principally herbicides and desiccants), 3% in
glass,  2% in nonferrous alloys,  and 2% in other uses (Bureau of Mines
1988).  The use of arsenic (as  gallium arsenide)  in semiconductors is
increasing,  but  total  usage for  this purpose  (about 5 tons) is still
small  compared with other uses.

-------
70   Section 5

5. 5  DISPOSAL

     The principal waste product of arsenic production is slag.  In 1979,
1200 Mg of arsenic in che form of slag were disposed of on land.  Since
arsenic production is a dry operation, only small quantities of the
chemical were discharged in wastewater (0.4 Mg in 1979) (EPA 1982).
There is essentially no recycling of arsenic from its principal uses as
wood preservatives or agricultural chemicals (Bureau of Mines 1988).

-------
                                                                      71
                          6.   ENVIRONMENTAL FATE

6.1  OVERVIEW

     Arsenic enters  the  environment both  as the result of natural forces
(volcanos and weathering of  arsenic-containing rocks) and human activity
(metal smelting, glass manufacturing, pesticide production and
application, and fossil-fuel burning).

     Arsenic in the  environment may undergo a complex cycle of chemical
interconversions and transfers between media. Atmospheric emissions,
which are usually adsorbed to particulate  matter, may undergo oxidation
before being returned to the surface by wet or dry deposition. Arsenic
in water may undergo either  reduction or oxidation, depending on pH, the
electrochemical oxidation-reduction potential (Eh), and other ions
present. Soluble forms of arsenic tend to  be quite mobile in water,
while less soluble species adsorb to clay  or soil particles.
Microorganisms in soils,  sediments, and water can reduce and methylate
arsenic to yield methyl  arsines, which volatilize and re-enter the
atmosphere. These forms  then undergo oxidation to become methyl arsonic
acids and ultimately transform back to inorganic arsenic.

6.2  RELEASES TO THE ENVIRONMENT

6.2.1  Anthropogenic

     Table 6.1 summarizes releases of arsenic into the environment in
1979 as a result of  human activity (EPA 1982b). The total amount of
arsenic released was 53,400  Mg, most of which (81%) was deposited on
land. The three largest  sources of emissions to air and soil are fossil
fuel consumption, pesticide  use, and copper smelting. These three
sources accounted for 35, 26, and 19%, respectively, of total air
emissions in 1979, and similar quantities  (32, 19, and 19%) of land
emissions. The largest sources of arsenic  in surface water are urban
runoff (37%), pesticide  application (25%),  and zinc production (20%).

6.2.2  Natural

     Table 6.2 summarizes the quantities of arsenic estimated to be
released to the environment  from natural sources. In the northern
hemisphere, the single largest source is volcanic emissions, accounting
for 88% of the total  releases. Other important natural sources include
weathering of arsenic-containing minerals  and ores (160 Mg/year), forest
fires (110 Mg/year),  and the  terrestrial biosphere (170 Mg/year).

-------
72
SacCion 6
                   Table 6.1.  Arsenic releases to the environment in 1979"
                Source
                                  Estimated environmental arsenic releases (Mg/year)
                             Air
Land
                                                          Water
Surface   POTW*     Total
       Production
         ASARCO, Tacoma         210     1,200        <1       Neg<     1,410

       Use
         Pesticides                 1,500     8,100       720        NA*   10,000
         Wood preservatives         Neg      Neg       Neg        NA      Neg
         Glass manufacture            10      Neg       Neg        NA        10
         Alloys                        e        e         e        NA      NA
         Other                        2       10       NA        <50        62
Other sources of releases
Fossil fuel consumption
Copper production
Lead production
Zinc production
Iron and steel production
Aluminum production
Boron production
Phosphorous production
'Manganese production
Antimony production
Cotton ginning
POTW
Urban runoff
Total

2,000
1,100
230
280
55
NA
NA
NA
10
NA
300
NA
NA
5.700

14,000
8,100
1,100

5,700
NA
2,200
640
1,400
Neg
580
20
NA
43,000

150
38
Neg
560
9
180
4
160
NA
NA
NA
NA
1.050
2,870

NA
NA
NA
1
6
NA
NA
NA
NA
NA
NA
1,800
NA
1,857

17,000
9,300
1,300
830
5,700
180
2,200
800
1,440
Neg
880
1,820
1,050
53,400
           "Adapted from EPA 1982b.
           *PubUcly owned treatment works.
           "Negligible.
           *Not available in reference.
           'Included in other sources of releases.

-------
                                      Environmental  Face
73
    Table 6.2. Estimates of arsenic emissions
             from natural sources"

Source
Ocean
Bubble bursting
Gas exchange
Earth's crust
Particle weathering
Direct volatilization
Volcanoes
Forest wildfires
Terrestrial biosphere
Natural source total
Global
arsenic
emission
(Mg/year)
28
84
240
0.7
7000
160
260
7800
Northern
hemisphere
arsenic
emission
(Mg/year)
12
SO
160
0.5
3500
110
170
4000
    "Adapted from EPA 1984c, after Walsh et al.
1979.

-------
74   Section 6

6.3  ENVIRONMENTAL FATE

6.3.1  Atmosphere

     Arsenic released  Co the atmosphere as a gas vapor or adsorbed Co
particulate matter may be  transported to other media via wet or dry
deposition, making the atmosphere an importanC route of arsenic cransfer
co ocher media. Trivalenc  arsenic may undergo oxidation in che air, and
arsenic in che atmosphere  is usually a mixture of the trivalent and
pentavalent forms (EPA 1984a).

     Most arsenic in air is adsorbed Co particulate matcer, especially
small diameter particles (e.g., less than 2 urn in diameter) (Coles et
al. 1979, as cited in EPA  1982b).  The residence time of particulate-
bound arsenic in the air depends on particle size, but a typical value
is about 9 days (Walsh et  al. 1979, as cited in EPA 1982b). Arsenic may
persist longer under conditions of limited atmospheric mixing or low
precipitation.

     Photolysis is not considered an important fate process for arsenic
compounds (Callahan et al. 1979).

6.3.2  Surface Water

     Arsenic in surface water can undergo a complex pattern of
transformations, including oxidation-reduction reactions,  ligand
exchange, biotransformation, precipitation, and adsorption (Callahan et
al. 1979). This complexity results in extremely mobile behavior in
aquatic systems, with much of che arsenic entering rivers  and eventually
transported to oceans (Callahan et al.  1979).  Rate constants for these
various reactions are noc  readily available,  but the factors most
strongly influencing intramedium fate processes in surface water include
Eh, pH, metal sulfide and  sulfide ion concentrations,  iron
concentration, presence of phosphorus minerals, temperature,  salinity,
and distribution and composition of biota (Callahan et al. 1979).

     Sorption onto clays,  iron oxides,  manganese compounds, and organic
material is an important fate of arsenic in surface water  (Callahan et
al. 1979, EPA 1982b), and  sediment serves as a reservoir for much of the
arsenic entering surface waters. Sediment-bound arsenic
(arsenate/arsenite),  which has been methylated by aerobic  and anaerobic
microorganisms,  may be released back to the water column (EPA 1982b)

6.3.3  Groundvater

     Soluble forms of arsenic interact with soil and travel with the
groundwater mass with which they are associated.  Shifts in oxidation
state may occur in either direction, depending on the  particular
physical and chemical characteristics of the soil and  groundwater.

     Volatilization of methylated forms from groundwater is possible
Nonporous soil and heavy vegetation cover are expected to  impede
volatilization,  and oxidation may transform volatile forms into
nonvolatile species or species that will adsorb to clay, organic matter.
and iron and aluminum complexes.

-------
                                                 Environmental Face    75

 6.3.4  Soil

     Arsenic  occurs  in soil  predominantly in  an  insoluble adsorbed form
 (EPA 1982b).  Clay  with high  anion exchange  capacity  (e.g., high
 kaolinite  content)  is  particularly effective  at  adsorbing arsenate via
 anion exchange.  Complexation and  chelation  by organic material, iron, or
 calcium are also important processes  fixing arsenic  in  insoluble form
 (Cooper et al.  1932, as cited in  EPA  1982b).

     Red and  yellow  podzols,  latosols, arid and  limestone soils, and
 subsoils high  in clay  and iron oxides have  greater holding capacity for
 arsenic than  other  types of  soil  (Hiltbold  et al. 1974,.as cited in EPA
 1982b). A  rise  in  pH in high iron soil, a drop in pH in lime soil, or a
 change in  redox  potential may lead to resolubilization of fixed arsenic

     Leaching of arsenic is  usually important only in the top 30 cm of
 soil (EPA  1982b).  Leaching carries  arsenic  deeper in sandy soils than in
 clay or loam soils,  although EPA  (1982b)  reports that no leached arsenic
 could be detected below 90 cm in  any of the studies.

     While arsenate  dominates in  aerobic  soils,  arsenite is the
 predominant form in  slightly reduced soils  (e.g., temporarily flooded
 soil),  and arsine, methylated arsenic, and  elemental arsenic predominate
 in very reduced  conditions (e.g.,  swamps  and  bogs) (EPA 1982b).

 6.3.5  Biota

     As noted above, arsenic  in water and soil may be reduced and
methylated by fungi, yeasts,  algae, and bacteria, and these forms may
volatilize and escape  into the air  (Wood  1974). The rate of
volatilization may vary considerably, depending on soil conditions
 (oxygenated or anaerobic). The pH value of  the soil and microbes present
also influence the rate of volatilization.  For example, a report by the
PAX company (1973) estimated  that  50% of  an applied dose might
volatilize in one year,  while Woolson (1976)  reported only 1 to 2%
volatilization over  a  period  of several months.

     Bioconcentration  of arsenic occurs in  aquatic organisms, primarily
in algae and lower invertebrates. Biomagnification in aquatic food
chains  does not appear  to be  significant  (EPA 1982b, Callahan et al.
1979),  although some fish and invertebrates contain high levels of
arsenic compounds which are  relatively inert  toxicologically (EPA
1984a).

     Plants may accumulate arsenic  via root uptake from soil solution,
and certain species may accumulate  substantial levels (EPA 1982b). In
addition to species differences,  the amount of arsenic taken up depends
on soil arsenic concentration, soil characteristics, and other factors.

-------
                                                                      77
                     7.   POTENTIAL FOR HUMAN EXPOSURE

7 . 1  OVERVIEW

     Arsenic is widely distributed in the environment, and all humans
are exposed to low levels via air, water, and food.  Typical
"background" exposure levels range from 20 to 70 jjg/day, with most of it
coming from food.

     Higher levels of exposure that may lead to significant human health
consequences are most often associated with drinking water contaminated
from natural mineral deposits, pesticide use, or improper disposal of
arsenic chemicals.  Emissions from metal smelters or arsenical pesticide
plants may result in significant exposure of workers and nearby
residents .

7.2  LEVELS MONITORED OR ESTIMATED IN THE ENVIRONMENT

7.2.1  Water

     A number of surveys of arsenic levels in drinking water, surface
water,  and groundwater have been performed (Greathouse and Craun 1978,
EPA 1980d,  Irgolic 1982, EPA 1980c, EPA 1982b,  Whitnack and Martens
1971,  McCabe et al. 1970, Francis et al. 1982.  Wentworth 1983).  The
main findings of these surveys may be summarized as follows:

   • Most drinking water supplies in this country provide water
     containing arsenic at levels well below the present standard of
     50
   •  Over 90% of all surface water contains 10 Mg/L arsenic or less.
     Concentrations above 10 Mg/L are most often encountered in the
     following major river basins:  Missouri River, Lower Mississippi.
     Colorado River, Western Gulf, Pacific Northwest, and Great Basin.

   •  Average arsenic concentrations in well water are generally
     <20 Mg/L. with highest maximum concentrations occurring in the Ohio
     River and Lake Erie basins.

   •  The average concentration of arsenic in U.S. drinking water
     supplies is about 2 Mg/L (Greathouse and Craun 1978).
     The chemical form of arsenic in drinking water is predominantly
     inorganic and typically contains a mixture of arsenate and arsenite
     (Irgolic 1982).

-------
78   Section 7

7.2.2  Air
     Average 24-h ambient air arsenic levels in Che United States,  based
on National Air Sampling Network data,  ranged from a low of 2.6 ng/m3 in
1980 to a high of 10.9 ng/m3 in 1978 (Akland 1983).  In all years,  the
99th percentile values (the concentration below which 99% of all
measurements fall) were less than 78 ng/m3.
     Suta (1980) reported that the highest levels of arsenic detected in
air (up to 1000 ng/m3) were found in the vicinity of copper smelters,
cotton gins, lead and zinc smelters, and glass manufacturers.   Arsenic
air levels were lowest (0.4 ng/m3) in remote areas.
     Studies by Andreae (1980, 1983), Attrep and Anirudhan (1977),
Crecelius (1974), and Johnson and Braman (1975) indicate that trivalent
and pentavalent forms of inorganic arsenic predominate in the air.

7.2.3  Soil
     The natural arsenic content of virgin soils varies between 0.1 and
80 ppm, with an average around 5 to 6 ppm (Walsh and Keenly 1975).   The
amount of arsenic in soil depends on geologic inputs from mineral
weathering processes (NAS 1977), atmospheric deposition, and residue
from pesticide application.  Soil arsenic is usually bound to clay
surfaces, and the extent to which arsenic is mobile in soil depends on
soil type, pH, and content of phosphate, aluminum, and iron (EPA 1984a).

7.2.4  Biota and Food
     Arsenic is found in most organisms that have been sampled, both
animal and plant. Table 7.1 displays typical arsenic levels in several
types of food. Arsenic content in plants typically varies from 0.01  to  5
ppm (NAS 1977). This is due mostly  to uptake of arsenic from the soil
(see Sect. 7.2.3 above), but some arsenic may also be present on the
surface as a residue from atmospheric deposition or pesticide
application (EPA 1982b).  Marine plants, crustaceans, and some fish
often contain naturally high concentrations of arsenic  (EPA 1984a),  but
this is in an organic form which has very low toxicity.

7.2.5  Resulting Background Exposure Levels
     Table 7.2 summarizes typical levels and probable forms involved in
human exposure to arsenic.  Typical  inhalation exposures from airborne
arsenic  (principally As203) are estimated at 0.06  ^g/day, while maximum
exposures may reach 0.6 /*g/day  in the general population.   Suta  (1980)
in EPA  (1982b) provides an estimate  of  6 pg/day  as the  upper limit of
nonoccupational  subpopulation exposure  to airborne arsenic, while
populations may be exposed to as much as 20 /ig/day of arsenic downwind
of copper smelters and  cotton gins.
     Ingestion of arsenic via drinking  water  is  expected  to occur  at
levels between  5  and  4000 /ig/day, with  5 pg/day  being typical.   The
highest  exposure  levels are expected in isolated wells  in  areas  of
naturally high background  arsenic  levels.   Arsenate is  the  predominant
form in  most  drinking water exposure situations.

-------
                                     Potential for Human  Exposure    79
                  Table 7.1.  Arsenic levels in foods"
          Food group
  Arsenic concentration (mg/kg)6
 Range of
mean values        Maximum
Meats, eggs, and milk            0.01-0.03
Vegetables and fruits             0 01-0.03
Cereal, nuts, and sugar products   0.01-0.04
Fmfish and shellfish              0.07-1.47
              0.5 (chicken)
              0.3 (potato products)
              0.4 (nee)
              19.1 (fin fish)
    "Adapted from EPA 1982b, after Jelinek and Corneliussen 1977.
    *Arsenic levels are reported as concentrations of As2O3, but this
does not imply that arsenic exists in this form in the food samples.

-------
80     Section  7
                        Table 7.2. Summary of estimated kreb of
                                          expooorc to metis?
             Route
                             Probable
                               form
                  Exposure («ig/day)

                 Typical    Maximum
                                Assumptions
       Surface water
       sources
                          Arsenate
       Groundwater
       sources
       Food-total
       diet
       High fish
       consumption
       Wine
       consumption
       Soil mgestion
       (children)
       General
       atmosphere
Arsenite from
25-100%.
MMA. DMAA.
some arsenate
in less
reduced water,
arsine has
been detected
under very
reduced
conditions

All forms;
large part may
be organically
bound arsenic

Arsenobetame,
MMA. possible
arsenite and
others
Arsenite
predominately
and arsenate
Arsenate or
organic
arsemcals
Arsenic
tnoxide
       Cigarette smoking   Arsenic
                          tnoxide
             200   Typical Most levels (99 6% of D W
                   survey) in U S. <10 pg/L.
                   mean-2 5 «ig/L, consumption of
                   2 L/day
                   Maximum. Maximum level in drinking
                   water supply 100 tig/L, consumption
                   of 2 L/day
            4.000   Typical Small sample of average
                   groundwater levels at 10 «ig/L  or
                   less, consumption of 2 L/day; there
                   are many incidences of higher
                   groundwater levels

                   Maximum: Maximum levels of 2000 pg/L
                   in naturally contaminated
                   supplies, consumption of 2 L/day
  21         190    Typical- FDA total diet study
                    estimate

                    Maximum. Total diet with seafood
1,000       10.000    Typical Fish or shellfish
                    containing 10 mg/kg, consumption
                    of 100 g fish/day

                    Maximum. Fish or shellfish
                    containing 100 mg/kg As, consumption
                    of 100 g fish/day
             500    Typical. Wine containing 100 pg/L,
                    consumption of 28 mL/day

                    Maximum. Wine containing maximum
                    levels of 500 Mg/L, consumption of
                    1 L/day
 0.02          20    Typical. Soil  containing 2.1 mg/kg
                    consumption of 10 mg soil/day

                    Maximum: Soil containing 2100 mg/kg,
                    consumption of 10 mg soil/day

     Inkaladoa

 0 06          06    Typical- Average ambient
                    concentration of 0.003 Mg/m1.
                    respiratory flow of 20 mj/day

                    Maximum. Typical urban
                    concentration in cities (containing
                    smelters) of 0.03 «ig/mj

              90    Arsenic concentration of
                    12 Mg/cigarette, 15% volatilized;
                    consumption  of 50 cigarettes/day
                                             0004
            •Sourer Adapted from EPA I982b.

-------
                                       Potential for Human Exposure   81

     Estimates of arsenic exposure from soil ingestion depend on Che
assumed concentration of arsenic in soil and on the amount of soil
ingested per day. EPA (1982) estimated intake from soil to range from
0.02 to 20 jig/day, based on a daily intake of 10 mg of soil. More recenc
estimates of soil intake suggest that children ingest about 200 mg of
soil per day (Calabrese et al. 1987). For soil containing an average of
5 ppm of arsenic (Walsh and Keeney 1975), the average arsenic intake per
day would be about 1 Mg-

     Arsenic exposure from dietary intake is estimated to range from 21
to 190 Mg/day (EPA 1982), with a typical value of around 45 to 50 jig/day
(Gartrell et al. 1986). Some individuals may experience higher
exposures, based on the intake of foods that are high in arsenic content
(e.g., fish, shellfish, and some wines).  Some of the arsenic ingested in
food (especially that in seafood) may be organic derivatives that are
less toxic.

7.3  OCCUPATIONAL EXPOSURES

     The two industries associated with significant risk of arsenic
exposure are metal smelting (especially copper) and arsenical pesticide
manufacture and application.

     In the past, air levels of arsenic in metal smelters ranged from
0.2 to 1500 /Jg/m3 (EPA 1981), and this was commonly associated with eye,
nose, throat, and skin irritation.   More recently, reductions in
emissions and improved industrial hygiene practices have reduced
occupational exposures substantially.

     Another occupation that can lead to significant arsenic exposure is
in the manufacture or use of arsenic-treated wood. The most common
arsenic-based wood preservatives are chromated copper arsenate (CCA) and
ammonium copper arsenate (ACA). Rosenberg et al. (1980) reported that
workers in three plants that prepare treated wood have increased arsenic
exposure as determined by urinary excretion levels, although no
prominent signs of arsenic-induced toxicity were noted in these workers.
In contrast, marked signs of arsenic toxicity have been reported in
individuals who saw or burn CCA-treated wood (Peters et al. 1964, 1986).
The principal exposure pathway is probably through the inhalation of
arsenic-contaminated dust (from sawing) or smoke (from burning),
although dermal contact with treated wood may also be a source of
exposure (ECI 1981).

7.4  POPULATIONS AT HIGH RISK

7.4.1  Above-Average Exposure

     Populations relying on groundwater or surface water near geologic
or man-made sources of arsenic are likely to receive higher than typical
exposures.  These areas include industrialized areas and areas where
large quantities of arsenic are disposed of in landfills (e.g.,
Pennsylvania, southern New York, Ohio, Indiana, and Washington); areas
of high historical pesticide use, with soil low in available ferrous and
aluminum hydroxides; and areas of high natural levels of arsenic-
containing mineral deposits (e.g., Western United States).

-------
82   Section 7

     Populations in Che area of copper and other types of metal smelter
may be exposed to above-average levels both through the air and as a
result of atmospheric deposition in soil and water.

7.4.2  Above-Average Sensitivity

     Inorganic arsenic is detoxified in humans by enzymic methylacion to
MMA and DMA, a process carried out mostly in the liver.  There may be
differences in the level of the activity of these enzymes between
individuals, and, if so, those with low activity ("nonmethylators") may
be more sensitive to arsenic than those with high enzymic activity.  In
addition, individuals with protein-poor diets or choline deficiency may
also be more sensitive to arsenic.

-------
                                                                      83
                         8.  ANALYTICAL METHODS

     Atomic absorption spectrometry  (AAS) is the most common analytical
procedure for measuring arsenic concentrations in environmental and
biological samples. Samples may be prepared for AAS in a variety of
ways. Most often the gaseous hydride procedure is employed, in which
arsenic in the sample is reduced  to arsine (arsenic H3),  a gas which is
trapped and introduced into the flame. This approach measures total
inorganic arsenic; however, it may not detect all organic forms unless a
digestion step is included.

     If concentrations of specific arsenic species [arsenic(III),
arsenic(V), MMA, DMA, fish arsenic, etc.] are to be determined,
separation procedures must be used prior to introduction of sample
material into a detection system. Various types of chromatography or
electrophoresis-separation systems are commonly used. Alternately, a gas
chromatography-multiple ion detection system following a hydride
generation/heptane cold trap system can be used to measure specific
organic forms of arsenic (Odanaka et al. 1983).

     The following sections briefly describe methods that are often
employed for measuring arsenic in environmental and biological samples

8.L  ENVIRONMENTAL MEDIA

     Representative methods appropriate for measuring arsenic in various
environmental media are listed in Table 8.1.

8.1.1  Air

     The American Conference of Government Industrial Hygienists (ACGIH)
Method 803 measures total particulate arsenic in air (APHA 1977).  The
method involves filter collection of air samples, arsine generation, and
silver diethyldithiocarbamate (SDDC) colorimetry. This method is similar
to NIOSH Method 7900, except that with Method 7900, flame atomic
absorption is used for the quantification of arsenic (NIOSH 1984). If
As203 fumes are present (as might occur in or near a smelter),  then
NIOSH Method 7901 (which uses a sodium carbonate impregnated filter) is
most appropriate. NIOSH Method 5022 is applicable for the quantification
of particulate organoarsenic compounds (NIOSH 1985).

8.1.2  Vater

     The atomic absorption graphite furnace technique (EPA 1983, EPA
1986b) is often used for measurement of total arsenic in water. It also
has been standardized by EPA.  Techniques to compensate for chemical and
matrix interferences are described in EPA (1983). Irgolic (1982),
Edwards et al. (1975),  and Brown and Button (1979) describe methods for
identifying species of arsenic in water samples.

-------
84    Seccion  8
               Table 8.1. Analytical methods for arsenic in environmental samples
Sample
matrix
Air
Water
or soil
Soil
Food
Sample
preparation
Filter
collection
and acid
digestion
Acid
digestion
Hydride
generation-
heptane
cold trap
Dry ashing
or acid
digestion
Analytical
method
Arstne generation-
colorimetric
(Method 803)
Atomic absorption
furnace technique
(Method 206.2)
Gas chromatography-
multiple ion
detection mass
spectrometry
Hydride generation-
AAS
Sample
detection
limit
0.4 /ig/m3
total As'
Ug/L
total As
0.2 ppb
total As
5 ppb
total As
Accuracy

85 to 95%
recovery
76 to 102%
recovery
79 to 117%
recovery
References
APHA 1977
EPA 1983,
EPA 1986b
Odanaka et al.
1983
Tarn and LaCroix
1982
     "4 to 9% average deviation (precision).

-------
                                                 Analytical Methods   85
8.1.3  Soil
     Inorganic arsenic and methylarsenic species can be identified and
measured in soil, using gas chromatography and multiple ion-detection
mass spectrometry after hydride generation and application of a heptane
cold trap to collect material for analysis (Odanaka et al. 1983).
Alternately, acid digestion of the sample followed by direct-furnace AAS
(EPA 1986b) or by hydride generation and AAS can be used for total
arsenic determination.

8.1.4  Food
     Dry ashing and wet digestion, followed by hydride generation and
AAS, are often employed for measuring total arsenic in foods. Narasaki
(1985) has successfully used oxygen bomb combustion, followed by hydride
generation-AAS, for arsenic in fatty foods.

8.2  BIOMEDICAL SAMPLES
     Methods for measuring arsenic in biological samples are listed in
Table 8.2.

8.2.1  Fluids and Exudates
     Total arsenic in blood and urine is usually measured using hydride
generation-AAS techniques (Foa et al. 1984). Norin and Vahter (1981)
described a procedure for determining specific forms of arsenic in body
fluids and exudates.

8.2.2  Tissues
     Mushak et al. (1977) described a furnace AAS technique  for total
arsenic, and also furnace AAS and gas-liquid chromatography  (GLC)
techniques for measuring chemical forms of arsenic in soft mammalian
tissues. Instead of hydride generation, these authors used chelation-
extraction via iodide derivatives to measure chemical forms  of arsenic
in these matrices.

-------
86    Section 8
                 Table 8.2.  Analytical methods for arsenic in biological samples
Sample
matrix
Blood

Urine



Adipose

Hair

Liver,
kidney.
other
soft
tissues
Liver,
kidney,
other
soft
tissues
Sample
preparation
Dry ashing

Dry ashing and
ion-exchange
chromatography

Oxygen bomb
combustion
Acid digestion

Acid digestion,
chloroform
extraction,
and iodide
chelation
Acid digestion,
chelation, and
benzene
extraction

Analytical
method
Hydride generation-
AAS
Hydride generation-
AAS


Hydride generation-
AAS
Hydride generation-
AAS
Furnace AAS for
total As



Gas-liquid
chromatography,
electron capture
detection

Sample
detection
limit
0.5 Mg/L
total As
0.5 Mg/L
for each
chemical
form
Sppb
total As
0.06 ppm
total As
0.2 ppm
or better



0.9 ppm
or better*
for chemi-
cal forms

Accuracy
95 to 102%
recovery
98 to 105%
recovery


90 to 102%
recovery
93% recovery

79.8%
recovery



98 to 108%
recovery



References
Foa et al.

Foa et al.



Narasaki

Curatola
1978
1984

1984



1985

et al.

Mushak et al.
1977



Mushak
1977







etal.



     "Sample detection limits and accuracy estimates for analysis of butter samples.
     *Data not available, lowest analytical spike levels listed.

-------
                                                                      87
                   9.  REGULATORY AND ADVISORY STATUS

     Table 9.1 summarizes regulations and guidelines that apply to
arsenic and inorganic arsenic compounds. These regulations and
guidelines have been established by a number of U.S. and international
advisory or regulatory agencies.

9.1  INTERNATIONAL
     The World Health Organization (WHO) recommends a drinking water
guideline value of 0.05 mg/L, based on the human health effects of
arsenic.

9.2  NATIONAL

9.2.1  Regulations
     Regulations in Table 9.1 are values that address air emissions,
occupational exposure concentrations, drinking-water levels, presence Ln
food, spill quantities, presence ir. hazardous wastes, and pesticide
usage.

9.2.1.1  Air
     The EPA Office  of Air Quality Planning and Standards (OAQPS), under
authorization  from Section 112  of the Clean Air Act, lists  inorganic
arsenic as a hazardous air pollutant  (HAP). Hazardous air pollutants are
those substances which may cause an  increase  in mortality or  serious
illness in humans following  significant exposure.
     EPA promulgated National Emissions Standards  for Hazardous Air
Pollutants (NESHAPS) for  three  stationary  source categories known to
emit inorganic arsenic: primary copper  smelters, glass  manufacturing
plants, and arsenic  plants.  These regulations, promulgated  in 1986,
cover equipment and  design specifications, work practices,  emission
limits, inspection and maintenance plans,  and monitoring requirements
for  ambient arsenic  concentrations near the  plant.
     The Occupational  Safety and Health Administration  (OSHA) sets
permissible exposure limits  (PELs)  for  occupational exposures to
chemicals based on  the  recommendations  of the National  Institute  for
Occupational  Safety  and  Health  (NIOSH). The OSHA  PEL for arsenic  is  10
Mg/m*  in workplace  air for  a time-weighted average (TWA) (8 h/day. 40
h/week).  EPA  has also  established a  TWA permissible exposure limit of  10
/*g/m3  which applies  to wood preservative application facilities.

-------
88    Section 9
                            Table 9.1.
          Agency
                             Description
Value
                 Ref<
                      Guideline for drinking water
WHO
       EPA OAQPS    Hazardous air pollutant (ambient air)
                                                       0.05 mg/L
                                                       NA'
       OSHA
       EPA OWRS
      EPAODW
      FDA
      EPAOERR
               NESHAPt-Ioorganic aneuc from primary    NA
               copper smelters, glass manufacturing
               plants, and arsenic plants

               Permissible TWA* workplace exposure limit    10 Mg Ai/mJ
               for inorganic arsenic
Permissible TWA workplace
exposure limit for organic arsenic

General permits under the National
Pollutant Discharge Elimination System
(NPDES)

Criteria aad standards for the NPDES

General pretreatmeat regulations for
existing aad aew sources of pollution

Maximum anmamimut level (MCL) in
drinking water


Permissible level* in food
  Muscle meats
  Edible meat by-products
  Eggs

Reparable quantity (RQ)
  Arsenic
  Arsenic disulflde
  Arsenic penioxide
  Aneaic trichloride
  Anofiic tnoudo
         triralftfe
                       Calcium aneaate
                       Calcium anentte
                       Cupric aoetoaneahe
                       Dietbytafww
                       Sodium aneaate
                       Sodium anenite

                     RQ (proposed)
                       Aneafe
                       AiMne djautfids
                       Arsenic peatoxido
                       Ancaic UUJUQC
                       Aneaie trifulflde
                       Chalet*"*1 aneaate
                       Calcium arsenite
                       Cupric acetoaneahe
                       Diethylaniae
                                                      500 «ig Ai/m1


                                                      NA



                                                      NA

                                                      NA


                                                      0.050 mg/L
                                                             0.5 |
                                                             t.Oppm
                                                             0.5 I
                       Sodium arsenite
                                                      1 Ib
                                                      5.000 Ib
                                                      5.000 Ib
                                                      5.000 Ib
                                                      5.000 Ib
                                                      5.000 Ib
                                                      1.000 Ib
                                                      1.000 Ib
                                                      lOOIb
                                                      I Ib
                                                      5.000 Ib
                                                      1.000 Ib
                                                      1.000 Ib


                                                        Ib
                                                        Ib
                                                        Ib
                                                        Ib
                                                        Ib
                                                        Ib
                                                        Ib
                                                        Ib
                                                        Ib
                                                        Ib
                                                        Ib
                                                        Ib
                                                        w
              WHO 1984
             48 FR 37886
             (06/05/80)

             40CFR6I
             51 FR 27956
             (04/04/86)

             29CFR 1910.1018

             43 FR 19584
             (05/05/78)

             29 CFR 1910.1000


             40 CFR 122.28
             40 CFR 125

             40 CFR 403


             40 CFR 141.11
             40 FR 59566
             (12/24/75)

             HSDB  1987
                                                                                 40 CFR 3014
                                                                                 50 FR 13456
                                                                                 (04/04/85)
             52 FR 8140
             (03/16/87)

-------
                                         Regulatory  and Advisory Status
89
                                Tasat9.1 (<
Agency

EPAOSW

EPAOPP

NIOSH
ACGIH
EPAODW
EPAOWRS
IARC
EPA
EPA
OSHA

Washington
Sute
Depeiuiieat
of Ecology
Sute
Agenciei
Detcnptioo
Extremely hazardous substances
Threshold planning quantity (TPQ)
Arsenic pentoude
Arsenous oxide
Arseaous trichloride
Anuw
Calcium aneaate
Potassium arseniu
Sodium anenate
Sodium anenite
Listing as a hazardous waste
constituent (Appendix VIII)
Arsenic and compounds (not other-
wise specified)
Arsenic aod
Arsenic pentoude
Arsenic trioxide
Restricted use pesticide— Inorganic
anemcals for wood preservative use*
Notice of preliminary determination to
cancel registration— Inorganic
anenkals for non-wood preservative uses
National GeAMsMi
Recommend exposure limit for
occupational exposure: ceiling
Threshold limit value (TLV-TWA)
Maximum «""«"•'••"« level goal (MCLG)
(proposed)
Ambient Water Quality Criteria to
iMioti?fi hvTTBn htjahh
Ingesting water and organisms
Ingesting organisms only
Group 1 (carcinogenic rank)
Group A (carcinogenic rank)
Unit rifk (Sofc.i.tiMi, ' r
Unh risk (oral. 1 ng/kg/day)
Point estimate for excess lung
cancer risk for working lifetime
exposure at 10 pg/m'
State HijalnflBBi
Interim community exposure standard
24-h Ambient air concentration
Annual average ambient air
concentration
Water quality standards
Value
100/10.000 Ib
100/10.000 Ib
500 Ib
100 Ib
500/10.000 Ib
500/10.000 Ib
1.000/10.000 Ib
500/10.000 Ib
NA

10 Mg/m1
(8-h TWA)
NA
i
2 
-------
 90   Section 9

 9.2.1.2  Water

      The discharge of arsenic in industrial wastewater  is  regulated  by
 EPA under the Clean Water Act National  Pollutant Discharge Elimination
 System (NPDES) and General Pretreatment Regulations. Regulatory
 limitations  (Effluent Guidelines)  for arsenic ami arsenic  compounds  have
 been established for 13 different  industrial point source  categories
 under 40 CFR 415-469.

      The EPA promulgated an Interim Maximum Contaminant Level  (IMCL)  for
 arsenic of 0.050 mg/L to protect the public health to the  extent
 feasible using technologies,  treatment  techniques, and other means chac
 are  generally available.  The  EPA is in  the process of setting  a new  MCL
 for  arsenic  and,  to that end,  has  proposed a Maximum Contaminant Level
 Goal (MCLG)  of 0.050 mg/L.  The MCLG is  a nonenforeeable goal,  based  onlv
 on the  consideration of health effect data. The MCL is set  as  close  to
 the  MCLG as  possible,  taking  cost  and feasibility into consideration.

 9.2.1.3  Reportable quantities

      The Comprehensive  Environmental Response, Compensation and
 Liability Act of 1980  (CERCLA) requires that persons in charge of
 vessels or facilities  from  which a hazardous substance has  been released
 (except where permitted)  in quantities equal to or greater  than its
 reportable quantity (RQ)  immediately notify the National Response Center
 of the  release.  The reportable quantities for arsenic and  several
 arsenic  compounds  set by  the  EPA Office of Emergency and Remedial
 Response (OERR)  are presented in Table 9.1. EPA has proposed decreasing
 the  RQ  to 1  Ib for  each arsenic  compound.

     Under the Superfund  Amendments and Reauthorization Act of 1986
 (SARA),  EPA  published a final  rule (EPA 1987a) listing extremely
 hazardous  substances and  corresponding threshold planning quantities
 (TPQs)  for those  substances.  The TPQs are intended to help  communities
 focus on the  substances and facilities of the moat immediate concern  for
 emergency planning  and  response  in case of accidental spills or releases
 to the  environment.  Several arsenic compounds are included  on  the
 extremely  hazardous  substances list and are listed with their TPQs in
 Table 9.1. Some  of  these  compounds are solids and,  therefore, have two
 TPQs; the  first,  for solids in forms which potentially can  result in an
 airborne  release,  the second,  for  solids in any other form.

 9.2.1.4  Waste disposal

     Chemicals are  included on the Resource Conservation and Recovery
Act  (RCRA) Appendix VIII  list of hazardous constituents (40 CFR Part
 261) If  they  have  toxic,  carcinogenic,  mutagenic,  or teratogenic effects
 on humans  or  other  life forms. Arsenic compounds are included on this
 list (see Table  9.1). Wastes containing arsenic are subject to the RCRA
 regulations promulgated by  the EPA Offlea of Solid Waste (OSW). These
regulations address generation,  transport, treatment,  storage, and
disposal of hazardous wastes.

-------
                                     Regulatory and Advisory Status   91

9.2.1.5  Pesticide
     The EPA Office of Pesticide Programs (OPP) is responsible for the
registration of all pesticide products sold in the United States.  The
OPP may cancel or modify the terms of registration whenever it is
determined that the pesticide causes unreasonable adverse effects  on the
environment. The OPP has restricted the use of inorganic arsenic for
pressure treating wood. It has proposed cancellation of all registered
uses of inorganic arsenic for non-wood preservative use; however,  the
use of calcium arsenate as a turf herbicide, lead arsenate as a
grapefruit growth regulator, sodium arsenite as a grape fungicide, and
arsenic acid as a dessicant are still under review. All copper
acetoarsenite and arsenic acid herbicide registrations have been
voluntarily cancelled by the manufacturers.

9.2.2  Advisory Guidance
     Advisory guidance levels are environmental concentrations
recommended by regulatory agencies protective of human health or aquatic
life. While not enforceable, these levels may be used as the basis for
judging acceptable and unacceptable levels of arsenic in the environment
or the workplace. Advisory guidance for arsenic is summarized in Table
9.1 and includes a recommended standard for occupational exposure, a
threshold limit value  (TLV), the MCLG for drinking water, health
advisories, and ambient water quality criteria.

9.2.2.1  Air levels
     NIOSH sets its recommended exposure limit  (REL) for occupational
exposure to arsenic in air at  2 jig/m3 for a 15-minute  ceiling, based on
classification of arsenic as a potential human  carcinogen.

     ACGIH recommends  a Threshold Limit Value  (TLV) TWA of 0.2 mg/m3
arsenic for soluble arsenic compounds, based on epidemiological evidence
of human health effects from inhaled arsenic.

9.2.2.2  Water levels
     As mentioned earlier,  the EPA ODW has  proposed an  MCLG  for arsenic
of O.OSO mg/L. The ODW assumes that for carcinogens,  there  is no
threshold below which  adverse effects will  not occur  and,  therefore,  the
MCLG for carcinogens  is normally  zero. Although arsenic has  been
classified  as a human  carcinogen  by EPA,  the MCLG was not  proposed as
zero because  increased risks of cancer due to  drinking  water exposure
have not been found  in the  United States,  and  there is  evidence
suggesting  that  arsenic may be an essential element for humans  (50 FR
46960). This  proposal  is  currently undergoing  Agency  review.

     The ODW  prepared Health Advisories  (HAs)  for numerous drinking
water  contaminants.  The HAs describe  concentrations of contaminants in
drinking water  at which  adverse  effects  would not be  anticipated  to
occur  and  include a margin of  safety  to  protect sensitive members of the
population.  The  HAs are  calculated for 1-day,  10-day,  longer-term, and
 lifetime exposures.  For arsenic,  the  EPA proposes that all HAs be
0.05 mg/L,  based on specific recommendations of the National Academy of
 Sciences.

-------
  92   Section  9
      Ambient water quality  criteria are guidelines set by the EPA Office
 of Water Regulations and  Standards (OWRS) to protect human health from
 potential adverse effects from  the ingestion of water and/or edible
 organisms (fish and shellfish)  from surface water sources. Since it is
 assumed that for carcinogens, there is no level at which risk is
 eliminated, the ambient water quality criteria are given for incremental
 lifetime cancer risks of  10'5,  10'6,  and 10''. The values for arsenic
 for an incremental increased lifetime cancer risk of 10*6 are 2.2 x 10'6
 mg/L for ingesting both water and organisms and 1.75 x 10'5 mg/L for
 ingesting organisms only  (EPA 1980a).

 9.2.3  Data Analysis

 9.2.3.1  Reference dose

      An oral reference dose (RfD) for  inorganic arsenic  of 1.0 x 10'3
 mgAg/day has been established by the  EPA in 1988.  This  value was based
 on the study of Tseng (1977) that identified a no-effect level of 1
 0gAg/day in chronically exposed humans.

 9.2.3.2  Carcinogenic potency

      Arsenic is classified by the EPA  as  a Group A  carcinogen (a known
 human carcinogen), based on evidence of increased lung cancer mortality
 in populations  exposed primarily through  inhalation and  on increased
 skin  cancer  incidence  in several populations consuming drinking  water
 with  high  arsenic  concentrations.

      Based on the  study by Tseng et al. (1968)  (see Sect.  4.2.3  6 on
 carcinogenicity),  EPA  (1987b)  calculated  that  the unit risk  for  skin
 cancer  (the  increased  risk of  developing  skin  cancer after lifetime
 ingestion  of water containing  1  jig/L) was between 3 and  7  x  10'5.  These
 values were  derived using  the  generalized multistage model with  both
 linear and quadratic terms and the maximal likelihood  method.  The  EPA
 has selected 5  x 10'3 as the most appropriate  single estimate  of oral
 unit risk  for skin cancer  (EPA 1988).

     For lung cancer. EPA  (1984a) used a linear  absolute-risk model  to
 calculate maximal likelihood estimates of the  slope of the dose-response
 relationship in the epidemiological studies  of Lee-Feldstein  (1983)
 ^ffj"8 6t al>  (1982>- Brown and Ch" (1983b), and Enterline and  Marsh
 (1982). From these slope estimates, unit risk values (the  additional
 risk of lung cancer associated with lifetime exposure  to 1 Mg/m3 of
 arsenic) were calculated for each study.  Since the  studies by Lee-
 Feldstein (1983), Higgins  et al. (1982),  and Brown and Chu (1983b) were
 all on the same worker population, the individual unit risks from these
 studies were combined, and the resulting value was  then combined with
 the estimate from the Enterline  and Marsh (1982) study to yield  an
 overall geometric mean unit  risk value of 4.29 x 10'3.

     Using similar epidemiological data on lung cancer incidence in
 arsenic-exposed workers, OSHA calculated that chronic exposure of a
worker to inorganic arsenic  in air at concentrations of 10 /*g/m3
corresponded to an excess cancer risk of 2.2 to 29 deaths per 1000
exposed employees.  This corresponds to a unit risk for occupational
exposure of 0.22 to 2.9 x  10'3.

-------
                                     Regulatory and Advisory Status   93

     No quantitative risk estimates have been performed for the risk of
internal cancers associated with oral exposure to arsenic.

9.3  STATE

9.3.1  Regulations

     The State of Washington Department of Ecology (DOE) adapted interim
community ambient air exposure standards for arsenic in 1984. The
interim standards are 2.0 pg/m3 arsenic for a 24-h period and 0.3 pg/m3
arsenic as an annual average concentration. The DOE plans to adopt
permanent standards after evaluating the sources of high ambient arsenic
concentrations in the Tacoma, Washington, area (51 FR 28012). No other
state ambient air standards for arsenic were located.

     State water quality standards are water quality criteria applied to
waters specified for designated uses. Twenty-seven of the 50 states have
established specific water quality standards for arsenic for waters
designated for general and/or domestic use. Host states set the water
quality standard for arsenic at 0.05 mg/L; however, for a few states,
the values range from 0.005 to 1.0 mg/L (Environment Reporter). In
addition, several states cite the Interim MCL and/or the EPA Water
Quality Criteria as being the guidance for the water quality standards
for toxic pollutants.

9.3.2  Advisory Guidance

     (Advisory guidelines from the states were still being compiled at
the time of printing.)

-------
                                                                       95
                             10.  REFERENCES


 ACGIH. (American Conference of Governmental Industrial Hygienists)
 1986.  Committee on Threshold Limit Values: Documentation of Threshold
 Limit Values and Biological Exposure Indices.  5ch ed.  Cincinnati, OH.

 Akland G.  1983.  Memo to D.  Sivulka, dated February 23,  1983.  Available
 from Environmental Protection Agency,  Research Triangle Park  NC
 Project file ECAO-HA-79-5.

 Amacher DE.  Paillet SC.  1980.  Induction of trifluorothymidine-resistant
 mutants by metal in L5178Y/TK cells.  Mutat Res 78:279-288.

 Andelman JB,  Barnett M.  1983.   Feasibility Study to Resolve  Questions on
 the  Relationship of Arsenic in Drinking Water  to Skin  Cancer.  Center
 for  Environmental Epidemiology,  University of  Pittsburgh.   Final report
 Submitted  to Office of Research and Development,  Environmental
 Protection Agency,  Washington,  DC.

 Andersen KJ,  Leighty EG,  Takahashi  MK.  1972. Evaluation of herbicides
 for  possible  mutagenic properties.  J Agric Food Chem 20:649-656.

 Andreae  MO.  1980.  Arsenic in rain and  the  atmospheric mass balance of
 arsenic. J Geophys Res 85:4512-4518.

 Andreae  MO.  1983.  Biotransformation of  arsenic  in  the marine
 environment.  In:  Lederer  W,  Fensterheim R,  eds. Arsenic: Industrial,
 Biomedical, and  Environmental  Perspectives. New York: Van Nostrand
 Reinhold.

 Anke M. Grun M,  Partschefeld M.  1976. The  essentiality  of arsenic for
 animals. In: Hempnill  DD. ed. Trace Substances  in  Environmental  Health
 Vol. 10. Columbia,  MO: University of Missouri,  pp. 403-409.

Anke M, Grun M,  Partschefeld M, Groppel  B,  Hennig A. 1978. Essentiality
 and  function of  arsenic.  In: Kirchgessner  M, ed. Trace  Element
Metabolism in Man and Animals. Vol. 3. Munich:  Freising-Weihenstephen
Technical University, pp. 248-252.
*Key studies.

-------
 96   Section 10

 Anke M, Krause U, Groppel D. 1987. The essentiality of arsenic  In-
 Hemphill DD, ed. Trace Substances in Environmental Health (in press).

 Ansul Company. 1971. Comments in support of continued registration of
 organic arsenical herbicides. Response to the Federal Register arsenic
 and lead notice. Fed Regist 36:  12709.

 APHA (American Public Health Association).  1977.  Methods of Air Sampling
 and Analysis: Arsenic.  American Public Health Association.  Washineton
 DC, pp. 435-438.                                                 6

 * Aranyi C, Bradof JN,  O'Shea WJ,  Graham JA,  Miller FJ.  1985.  Effects of
 arsenic trioxide inhalation exposure on pulmonary antibacterial defenses
 in mice. J Toxicol Environ Health  15:163-172.

 Ariyoshi T, Ikeda T. 1974.  On the  tissue distribution and the  excretion
 of arsenic in rats and  rabbits of  administration  with arsenical
 compounds.  J Hyg Chem 20:290-295.

 Armstrong CW,  Stroube RB,  Rubio  T, et al. 1984. Outbreak of  fatal
 arsenic poisoning caused  by contaminated drinking water.  Arch  Environ
 Health  39:276-279.

 Attrep  M, Anirudhan  M.  1977.  Atmospheric  inorganic and organic  arsenic
 Trace Subst Environ  Health  11:365-369.

 Axelson 0,  Dahlgren  E.  Jansson C-D,  Rehnlund SO.  1978. Arsenic  exposure
 and mortality: A  case reference  study from a Swedish  copper  smelter.  Br
 J  Ind Med 35:8-15.

 Baetjer  AM.  Lilienfeld  AM, Levin ML.  1975. Cancer  and occupational
 exposure to inorganic arsenic. In: Abstracts, 18th International
 Congress on Occupational Health, Brighton, England, September 14-19
 1975. Organizing  Committee, Brighton, England, p.  393.

 Barnes D et al. 1987. Reference  dose  (RfD):  Description and use in
 health risk assessment.  Appendix A in Integrated Risk  Information
 Systems  Supportive Documentation. Vol. 1. Office of Health and
 Environmental Assessment, Environmental Protection Agency, Washington,
 DC. EPA/600/8-86-032a.

 Barrett  JC,  Shelby MD.  1986. Mechanisms of multistep carcinogenesis:
 Keys to  developing in vitro approaches for assessing the carcinogenicity
 of chemicals. Fed Chem Toxicol 24(6/7):657-661.

 Baxley MH, Hood RD, Vedel GC, Harrison UP, Szczech CM. 1981. Prenatal
 toxicity or orally administered  sodium arsenite in mice.  Bull Environ
Contain Toxicol 26:749-756.

Beckett WS, Moore JL, Keogh JP,  Bleecker ML.  1986. Acute encephalopathy
due to occupational exposure to arsenic.  Br  J Ind Med 43:66-67.

-------
                                                         References   97

Beckman L, Nordenson" 1. 1986. Interaction between some common genotoxic
agents. Hum Hered 36:397-401.

Bencko V, Gelst T, Arbetova D, Dharmadlkari DM, Svandova E. 1986.
Biological monitoring of environmental pollution and human exposure to
some trace elements. J Hyg Epidemiol Microbiol Immunol 30:1-10.

Bencko V. 1987. Arsenic. Adv Mod Environ Toxicol 11:1-30.

Bertolero F, Marafante E, Edel Rade J, Pietra R, Sabbioni E. 1981.
Biotransformation and intracellular binding of arsenic in tissues  of
rabbits after intraperitoneal administration of As-74 labelled arsenlte.
Toxicology 20:35-44.

Birmingham DJ, Key MM, Doladay DA, Perone VB. 1965. An outbreak of
arsenical dermatoses in a mining community. Arch Dermato1 91:457-464.

Blejer HP, Wagner V. 1976. Case study 4: Inorganic arsenic-ambient level
approach to the control of occupational cancerigenic exposures.  In:
Occupational Carcinogenesis, Proceedings of a Conference, New York
Academy of Sciences, New York, March 24-27, 1975. Ann N Y Acad Sci
271:179-186.

Blom S, Lagerkvist B, Linderholm H. 1985. Arsenic exposure to smelter
workers: Clinical and neurophysiological studies. Scand J Work Environ
Health 11:265-269.

Borgono JM, Greiber R. 1972. Epidemiological study of arsenism in the
city of Antofagasta. In: Hemphill DC, ed.  Trace Substances in
Environmental Health-V, Proceedings of the 5th Annual Conference,
University of Missouri, Columbia, pp. 13-24.

Borgono JM, Venturino H, Vincent P. 1980. Clinical and epidemiological
study of arsenism in northern Chile (1977). Rev Med Chile 108:1039-1048.

Borgono JM, Vincent P, Venturino H, Infante A. 1977. Arsenic in the
drinking water of the city of Antofagasta: Epidemiological and clinical
study before and after installation of a treatment plant. Environ Health
Perspect 19:103-105.

Brown CC, Chu KC. 1983a. Approaches to epidemiologic analysis of
prospective and retrospective studies: Example of lung cancer and
exposure to arsenic. In: Proceedings of the SIMS Conference on
Environmental Epidemiology: Risk Assessment, June 28-July 2, 1982. Alta,
UT. SIAM Publication.

Brown CC, Chu KC. 1983b. Implications of the multistage  theory of
carcinogenesis applied to occupational arsenic exposure. J Natl Cancer
Inst 70:455-463.

Brown EJ, Button DK. 1979. A  simple method of  arsenic speciation.  Bull
Environ Contain Toxicol 21:37-42.

-------
98   Section 10

Brown LH, Pottarn LM, Blot WJ. 1984. Lung cancer in relation to
environmental pollutants emitted from industrial sources. Environ Res
34:250-261.

Brune 0, Nordberg G, Wester P-0. 1980. Distribution of 23 elements in
kidney, liver, and lung of a control group in Northern Sweden and of
exposed workers from a smelter and refinery. Sci Total Environ 16:13-35.

Buchet JP, Lauwerys R. 1985. Study of inorganic arsenic methylation by
rat liver in vitro: Relevance for the interpretation of observations in
man. Arch Toxicol 57:125-129.

Buchet JP, Lauwerys R, Roels H. 1980. Comparison of several methods for
the determination of arsenic compounds in water and in urine.  Int Arch
Occup Environ Health 46:11-29.

Buchet JP, Lauwerys R, Roels H. 1981a. Comparison of the urinary
excretion of arsenic metabolites after a single oral dose of sodium
arsenite, monomethyl arsonate, or dimethyl arsinate in man. Int Arch
Occup Environ Health 48:71-79.

Buchet JP, Lauwerys R, Roels H. 1981b. Urinary excretion of inorganic
arsenic and its metabolites after repeated ingestion of sodium meta
arsenite by volunteers. Int Arch Occup Environ Health 48:111-118.

Buck VB. 1969. Untoward reactions encountered with medicated feeds. In:
The Use of Drugs in Animal Feeds. Proceedings of a symposium.
Washington. DC: National Academy of Sciences. HAS Publ 1679, pp. 196-
217.

Buck WB, Osweiler GD, Van Gelder GA. 1973. Clinical and Diagnostic
Veterinary Toxicology. Dubuque, IA: Kendall-Hunt Publishing.

Bureau of Mines. 1988. Mineral Commodity Summaries. Washington, DC: U.S.
Department of the Interior, Bureau of Mines.

Burgdorf W, Kurvlnk K, Cervenka J.  1977. Elevated sister chromatid
exchange rate in lymphocytes of subjects treated with arsenic. Hum Genet
36:69-72.

Burk D, Beaudoin AR. 1977. Arsenate-induced renal agenesis in rats.
Teratology 16:247-260.

* Byron WR, Bierbower GW, Brouwer JB, Hansen WH. 1967. Pathological
changes in rats and dogs from two-year feeding of sodium arsenite or
sodium arsenate. Toxicol Appl Pharmacol 10:132-147.

Cabe PA, Carmichael NG, Tilson HA.  1979. Effects of selenium, alone and
in combination with silver or arsenic, in rats. Neurobehav Toxicol
1:275-278.

-------
                                                         References   99

Calabrese EJ, Kostecki PT, Gilbert CE. 1987. How much soil do children
eat? An emerging consideration for environmental health risk assessment.
Comments Toxicol 1:229-241.

Calesnick B, Wase A, Overby LR. 1966. Availability during human
consumption of the arsenic in tissue of chicks fed arsanilic-74As acid
Toxicol Appl Pharmacol 9:27-30.

Callahan MA, Slimak MW, Gable NV. et al. 1979. Water-Related
Environmental Fate of 129 Priority Pollutants. Washington, DC.  EPA-
440/4-79-029.

* Cannon JR, Saunders JB, Toia RF. 1983. Isolation and preliminary
toxicological evaluation of arsenobetaine - the water-soluble arsenical
constituent from the hepatopancreas of the western rock lobster. Sci
Total Environ 31:181-185.

Cebrian ME, Albores A, Aguilar M, Blakely E. 1983. Chronic arsenic
poisoning in the north of Mexico. Hum Tozicol 2:121-133.

Charbonneau SM, Hollins JG, Tarn GKH, Bryce F, Ridgeway JM, Willes RF.
1980a. Whole body retention, excretion, and metabolism of As-75 arsenic
acid in the hamster. Toxicol Lett 5:175-182.

Charbonneau SM, Spencer K, Bryce F, Sandi E. 1978a. Arsenic excretion by
monkeys dosed with arsenic-containing fish or with inorganic arsenic.
Bull Environ Contain Toxicol 20:470-477.

Charbonneau SM, Tarn GKH, Bryce F, Sandi E. 1980b. In: Proceedings of the
19th Annual Meeting of the Society of Toxicology, Washington, DC.. March
9-13, 1980. Abstract  362.  p. A121.

Charbonneau SM, Tarn GKH, Bryce F, Collins B.  1978b. Pharmacokinetics and
metabolism of  inorganic arsenic  in the  dog. Trace Subst  Environ Health
12:276-283.

* Chen CJ, Chuang YC,  Lin  TM,  Wu HY.  1985. Malignant neoplasms  among
residents of a Blackfoot disease-endemic area in Taiwan:  High-arsenic
artesian well  water and cancers. Cancer Res 45:5895-5899.

* Chen CJ, Chuang YC,  You  SL,  Lin TM, Wu HY.  1986. A retrospective  study
on  malignant neoplasms of  bladder,  lung, and  liver in  a Blackfoot
disease-endemic  area  in Taiwan.  Br J  Cancer 53:399-405.

Chernoff N,  Rogers  EH.  1975.  Effects  of cacodylic  acid on the prenatal
development  of rats and mice.  Substitute chemical  program;  the first
year  of progress.  Proceedings of a  symposium.  Vol. II.  Toxicological
Methods and Genetic Effects  Workshop.  Environmental  Protection Agency.
pp. 197-204.

Cikrt MV,  Bencko V, Tich M,  Benes B.  1980.  Biliary excretion of 74As and
 its distribution in the  golden hamster after administration of 74As
 (III)  and  74As (V). J Hyg Epidemiol Microbiol 24:384-388.

-------
 100   Section 10

 Cole HN. 1916. A study of sodium cacodylate In the treatment of
 syphilis. J Am Med Assoc 67:2012-2013.

 Coles DC, Ragaini RC, Ondov JM,  Fisher  GL,  Silberman D,  Prentice BA.
 1979. Chemical studies of stack fly ash from a coal-fired power plant
 Environ Sci Technol 13(4):455-459.

 Cooper H, Paden WR, Hall EE,  et al.  1932. Soils differ markedly in their
 response to addictions of calcium arsenate.  South  Carolina Agric Exp
 Stan Ann Rep 45:23-27 (as cited in EPA  1982).

 Crecelius EA.  1974. The Geochemistry of Arsenic and Antimony in Puget
 Sound and Lake Washington,  Washington.  Thesis:  University of Washington
 Seattle.

 Crecelius EA.  1977. Changes in  the  chemical  speciation of arsenic
 following ingestion by man.  Environ Health  Perspect 19:147-150.

 Crema A.  1955.  Distribution et  elimination  de  1'arsenic  76 chez  la
 souris normale et cancereuse. Arch  Int  Pharmacodyn 103:57-70.

 Cristau B,  Chabas ME,  Placldl M.  1975.  Is p-arsanillc acid
 biotransformed in the rat and guinea pig? Ann  Pharm Fr 33(1):37-41.

 Curatola  CJ, Grunder FI,  Moffltt  AE.  1978. Hydride generation atomic
 absorption  spectrophotometry for  determination of  arsenic  in hair  Am
 Ind  Hyg Assoc  J  39:933-938.

 Cuzick J, Evans  S,  Gillman  M, Evans  D.  Price.  1982. Medicinal arsenic
 and  internal malignancies.  Br J Cancer  45:904-911.

 Czarnecki GL,  Baker DH.  1984. Feed additive  interactions  in  the  chicken.
 Reduction of tissue copper  deposition by dietary roxarsone  In healthy
 and  In Elmeria acervalina-infected or Eimeria  teneela-infected chicks
 Poult  Sci 63(7):1412-1418.

 Czarnecki GL,  Baker DH, Garst JE. 1984b. Arsenic-sulfur amino acid
 interactions in  the chick. J Anim Sci 59(6):1573-1581.

 Czarnecki GL,  Edmonds MS, Izqulerdo OA,  Baker DH.  1984a.  Effect  of 3-
 nitro-4-hydroxylphenylarsonic acid on copper utilization by  the  pig,
 rat, and chick. J Anim Sci  59(4):997-1002.

 Danan M, Dally S, Conso F.  1984. Arsenic-induced encephalopathy  (letter
 to the  editor). Neurology 34:1524.

 Datta DV. 1976. Arsenic and non-cirrhotic portal hypertension. Lancet
 1:433.

Dickinson JO. 1975.  Toxicity of cacodylic acid sllvicide in cattle. West
Vet 3:1-4.

-------
                                                        References   101

Dobson RL, Young MR, Pinto JS. 1965. Palmar keratoses and cancer. Arch
Dermatol 92:553-556.

Donofrio PD, Wilbourn AJ. 1985. Subacute arsenic toxicity presenting on
clinical and electromyographic examination as Guillain-Barre syndrome.
110th Annual Meeting of  the American Neurological Association, Chicago,
IL, October 2-5, 1985. Ann Neurol 18:156-157.

Driesback RH. 1980. Handbook of Poisoning: Prevention, Diagnosis, and
Treatment. 10th ed. Los Altos, CA: Lange Medical Publications, pp 215-
218.

Ducoff HS, Neal WB, Straube RL, et al. 1948. Biological studies with
arsenic-76. II. Excretion and tissue localization. Proc Soc Exp Biol Med
69:548-554.

Dutkiewicz T. 1977. Experimental studies on arsenic absorption routes in
rats. Environ Health Perspect 19:173-177.

ECI (Enviro Control, Inc). 1981. Assessment of Exposure to Arsenic from
CCA-Treated Wood. Final report. Prepared for Environmental Protection
Agency, Exposure Assessment Group, Office of Health and Environmental
Assessment, Washington, DC.

Edmonds MS, Baker DH. 1986. Toxic effects of supplemental copper and
roxarsone when fed alone or in combination to young pigs. J Anim Sci
63(2):533-537.

Edwards T, Merilees HL, McBride BC. 1975. Rapid, sensitive method for
the separation and detection of arsenic compounds in biological systems.
J Chromatogr 106:210-212.

* Enterline PE, Henderson VL, Marsh GM. 1987. Exposure to arsenic and
respiratory cancer. A reanalysis. Am J Epidemiol 125(6):929-938.

Enterline PE, Marsh GM. 1982. Mortality among workers exposed to arsenic
and other substances in a copper smelter. Am J Epidemiol 116:895-910.

EPA (Environmental Protection Agency). 1980a. Ambient Water Quality
Criteria for Arsenic. Springfield, VA: National Technical Information
Service. PB81-117327.

EPA (Environmental Protection Agency). 1980b. STORET. Washington, DC:
Monitoring and Data Support Division.

EPA (Environmental Protection Agency). 1981. The Carcinogen Assessment
Group's Final Risk Assessment on Arsenic. Washington, DC: Environmental
Protection Agency, Office of Health and Environmental Assessment.  PB
81-206013.

EPA (Environmental Protection Agency). 1982. An Exposure and Risk
Assessment for Arsenic. Washington, DC: Environmental Protection Agency,
Office of Water Regulations and Standards. EPA 440/4-85-005.

-------
102   Section 10

EPA  (Environmental Protection Agency). 1983. Methods for Chemical
Analysis of Water and Wastes.  Office of Research and Development,
Environmental Monitoring and Support Laboratory.  Cincinnati, OH. EPA
600/4-79-020.

* EPA (Environmental Protection Agency).  1984a. Health Assessment
Document for Inorganic Arsenic. Final report. Research Triangle Park,
NC:  Environmental Protection Agency. EPA 600/8-83-021F.

EPA  (Environmental Protection Agency). 1984b. Health Effects Assessment
for Arsenic. Washington, DC: Environmental Protection Agency, Office of
Emergency and Remedial Response. EPA/540/1-86/020.

EPA  (Environmental Protection Agency). 1984c. Occurrence of Arsenic in
Drinking Water, Food, and Air. Washington, DC: Environmental Protection
Agency, Office of Water Regulations and Standards.

EPA  (Environmental Protection Agency). 1985. Health Advisory for
Arsenic. Draft. Washington, DC: Environmental Protection Agency, Office
of Drinking Water.

EPA  (Environmental Protection Agency). 1986a.  Evaluation of the
Potential Carcinogenicity of Arsenic (7940-38-2) in Support of
Reportable Quantity Adjustments Pursuant to CERCLA Section 102.
Washington, DC: Environmental Protection Agency, Office of Health and
Environmental Assessment.  OHEA C-073-S.

EPA  (Environmental Protection Agency). 1986b. Statement of Work for
Inorganics Analysis. EPA Contract Laboratory Program, October 1986.

EPA  (Environmental Protection Agency). 1987a. Extremely hazardous
substances list and threshold planning quantities;  emergency planning
and release notification requirements; final rule.  Fed Regist 52:12278,
April 22, 1987.

* EPA (Environmental Protection Agency).  1987b. Risk Assessment Forum.
Special Report on Ingested Inorganic Arsenic: Skin Cancer; Nutritional
Essentiality. Draft report dated November 1987.  EPA/625/3-87/013.

EPA  (Environmental Protection Agency). 1988. Recommended Agency policy
on the carcinogenicity risk associated with the ingestion of inorganic
arsenic. Memorandum from the Administrator to Assistant Administrators,
June 21, 1988.

Falk H, Caldwell GG, Ishak KG, et al. 1981a. Arsenic-related hepatic
angiosarcoma. Am J Ind Med 2:43-50.

Falk H, Herbert JT, Edmonds L, et al. 1981b. Review of four cases of
childhood hepatic angiosarcoma--elevated environmental arsenic exposure
in one case. Cancer 47:383-391.

Felkner 1C. 1980. Annual Report to EPA. Quoted in EPA 1981
(unpublished).

-------
                                                        References   103

Perm VH. 1977. Arsenic as a teratogenic agent. Environ Health Perspect
19:215-217.

Perm VH, Carpenter S. 1968. Malformations induced by sodium arsenate  J
Reprod Fertil 17:199-201.

Fern VH, Hanlon DP. 1985. Constant rate exposure of pregnant hamsters to
arsenate during early gestation. Environ Res 37:425-432.

Perm VH, Saxon A, Smith BM. 1971. The teratogenic profile of sodium
arsenate in the golden hamster. Arch Environ Health 22:557-560.

Fierz U. 1965. Katamnestische untersuchungen uber die nebenwirkungen der
therapie mit anorganischem arsen bei hautkrankheiten. Dermatologies
131:141.

Fischer AB, Buchet JP, Lauwerys RR. 1985. Arsenic uptake, cytotoxicity,
and detoxification studied in mammalian cells in culture. Arch Toxicol
57:168-172.

Fisher GL, McNeill KL, Democko CJ. 1986. Trace element interactions
affecting pulmonary macrophage cytotoxicity. Environ Res 39:164-171.

Foa V, Colombi A, Maroni M, Buratti M, Calzaferri G. 1984. The
speciation of the chemical forms of arsenic in the biological monitoring
of exposure to inorganic arsenic. Sci Total Environ 34:241-259.

Foa V, Bertolero P. 1983. Arsines. Encyclopedia of Occupational Health
and Safety. NIOSH/00148981, pp. 183-184.

Francis J, Brower B, Graham W, Larson 0, McCaull J, Vigorita H.  1982.
National Statistical Assessment of Rural Water Conditions (Vol.  II).
Department of Rural Sociology, Cornell University.

Franklin M, Bean W, Harden RC. 1950. Fowler's solution as an etiologic
agent in cirrhosis. Am J Med Sci 219:589-596.

Frost DV, Perdue HS, Main BT, et al. 1962. Further considerations on the
safety of arsanilic acid for feed use. In: Proceedings,  12th World's
Poultry Congress, Sydney, Australia. Section papers, pp. 234-237.

Gartrell MJ, Craun JC, Podrebarac DS, Gunderson EL. 1986.  Pesticides,
selected elements, and other chemicals in adult total diet samples,
October 1980-March 1982. J Assoc Off Anal Chem 69:146-161.

Gerhardsson L, Brune D, Nordberg IGF, Wester PO.  1985.  Protective effect
of selenium on lung cancer in smelter workers. Br J  Ind  Med 42:617-626.

Gerhardt R, Hudson J, Rao R, Sobel R. 1978. Chronic  renal insufficiency
from cortical necrosis induced by arsenic poisoning. Arch Intern Med
138:1267-1269.

-------
104   Section 10

Ginsberg JM, Lotspeich UD. 1963. Interrelations of arsenate and
phosphate transport in the dog kidney. Am J Physiol 205:707-714.

Goldsmith JR, Dean -M,' Thorn J, Gentry G. 1972.  Evaluation of health
implications of elevated arsenic in well water. Water Res 6:1133-1136.

Gordon GF. 1985. Sex and age related differences in trace element
concentrations in hair. Sci Total Environ 42:133-147.

Grant WM. 1974. Toxicology of the Eye. Springfield, IL:  Charles C.
Thomas, pp. 51-57.

Greathouse DG, Craun GF. 1978. Cardiovascular  disease study--occurrence
of inorganics in household tap water and relationships to cardiovascular
mortality rates. In: Hemphill DD, ed.  Trace Substances  in Environmental
Health-V, Proceedings of the 5th Annual Conference, University of
Missouri, Columbia 12:31-39.

Greaves W, Rom W, Lyon J. 1981. Relationship between lung cancer and
distance of residence from non-ferrous smelter stack effluent. Am J Ind
Med 2:15-23.

Gregus Z, Klaassen CD.  1986. Disposition of metals in rats: A
comparative study of fecal, urinary, and biliary excretion of eighteen
metals. Toxicol Appl Pharmacol 85:24-38.

Guyton AC. 1947. Measurement of respiratory volume of laboratory
animals. Am J Physiol 150:70-77.

Hamamoto E. 1955. Infant arsenic poisoning by  powdered milk. Jap Med J
1649:2-12.

Hanlon DP, Perm VH. 1977. Placental permeability of arsenate ion during
early embryogenesis in the hamster. Experientia 33:1221-1222.

Hanlon DP, Ferm VH. 1986a. Concentration and chemical status of arsenic
in the blood of pregnant hamsters during critical embryogenesis. I.
Subchronic exposure to arsenate utilizing constant rate administration.
Environ Res 40:372-379.

Hanlon DP, Ferm VH. 1986b. Concentration and chemical status of arsenic
in the blood of pregnant hamsters during critical embryogenesis. II.
Acute exposure. Environ Res 40:380-390.

* Hanlon DP. Ferm VH. 1987. The concentrations and chemical status of
arsenic in the early placentas of arsenate-dosed hamsters. Environ Res
42:546-552.

* Harrington JM, Middaugh JP, Morse DL, Housworth J. 1978. A survey of a
population exposed to high arsenic in well water in Fairbanks, Alaska.
Am J Epidemiol 108:377-385.

-------
                                                        References   105

 Harrison VP.  Frazier JC,  Mazzanti EN,  Hood RD.  1980. Teratogenlclty of
 disodiua methanearsonate  and sodiua dimethylarslnate (sodium cacodylace)
 in mice.  Teratology 21:43A.

 Harrison JV,  Packman EW,  Abbott DD.  1958.  Acute oral toxicity and
 chemical and  physical properties of arsenic trioxides. Arch Ind Health
 17:118-123.

 Harvey  SC.  1970.  Heavy metals.  In:  Goodman LS, Oilman A, eds. The
 Pharmacological Basis of  Therapeutics. Toronto: Collier-Macmillan,
 pp. 958-965.

 Heilbron Database.  1987.  Computer printout.  Palo Alto, CA: Dialog
 Information Services.

 Heydorn K.  1970.  Environmental  variation of arsenic levels in human
 blood determined  by neutron  activation analysis. Clin Chim Acta 28:349-
 357.

 * Heyvood R,  Sortwell  RJ. 1979.  Arsenic intoxication in the rhesus
 monkey. Toxicol Lett 3:137-144.

 Higgins  I, Welch  K,  Burchfiel C.  1982. Mortality of Anaconda Smelter
 Workers  in Relation to Arsenic  and Other Exposures. Ann Arbor, MI:
 Department of Epidemiology,  University of  Michigan.

 Hill AB.  Fanning  EL.  1948. Studies on  the  incidence of cancer in a
 factory handling  inorganic compounds of arsenic. I. Mortality experience
 in the  factory. Br  J  Ind Ned 5:1-6.

 Hill CH.  1975. Interrelationships of selenium with other trace elements.
 Fed Proc  34:2096-2100.

 Hiltbold AE,  Hajek  AE, Buchanan GA. 1974.  Weed Sci 22(3):272-275. In:
 Sandberg GR,  Allen  IK. 1975. A  proposed arsenic cycle in an agronomic
 ecosystem. Arsenical pesticides.  ACA Symposium Series 7:124-147.

 Hindmarsh JT, McCurdy  RF. 1986.  Clinical and environmental aspects of
 arsenic toxicity. CRC  Grit Rev  Clin Lab Sci  23:315-347.

 Hindmarsh JT, McLetchie OR,  Heffernan  LP.  et al.  1977.
 Electromyographic abnormalities  in chronic  environmental arsenicalism.  J
Anal Toxicol  1:270-276.

Holland RH, McCall  MS, Lanz  HC.  1959. A study of Inhaled arsenic-74 in
man.  Cancer Res 19:1154-1156.

Holmberg RE,  Ferm VH.  1969.  Interrelationshiops of selenium, cadmium,
and arsenic in mammalian teratogenesis. Arch Environ Health 18:873-877.

-------
106   Section 10

Holmqviat I. 1951. Occupational arsenical dermatitis. A study among
employees at a copper ore smelting work, including investigations of
skin reactions to contact with arsenic compounds.  Acta Derm Venereol
31:1-214.

Hood RD. 1972. Effects of sodium arsenite on fetal development.  Bull
Environ Contain Toxicol 7:216-222.

* Hood RD. 1985. Cacodylic Acid: Agricultural Uses, Biological Effects.
and Environmental Fate. Washington, DC: U.S.  Government Printing Office.

Hood RD, Bishop SL. 1972. Teratogenic effects of sodium arsenate in
mice. Arch Environ Health 24:62-65.

Hood RD, Harrison VP. 1982. Effects of prenatal arsenite exposure in
hamster. Bull Environ Contam Toxicol 29:671-678.

Hood RD, Harrison VP, Vedel GC. 1982. Evaluation of arsenic metabolites
for prenatal effects in the hamster. Bull Environ Contam Toxicol
29:679-687.

* Hood RD, Thacker GT, Patterson BL. 1977. Effects in the mouse and rat
of prenatal exposure to arsenic. Environ Health Perspect 19:219-222.

Hood RD, Thacker GT, Patterson BL, Szczech GM. 1978. Prenatal effects of
oral versus intraperitoneal sodium arsenate in mice. J Environ Pathol
Toxicol 1:857-864.

Hood RD, Vedel-Macrander GC, Zaworotko HJ, Tatum FM, Meeks RG. 1987.
Distribution, metabolism, and fetal uptake of pentavalent arsenic in
pregnant mice following oral and interaperitoneal administration.
Teratology 35:19-25.

HSDB (Hazardous Substances Data Bank). 1987.  Record for Arsenic.
Computer printout. National Library of Medicine, April 9, 1987.

Huang YZ, Qian XC, Vang GQ, et al. 1985. Endemic chronic arsenism in
Xinjiang. Chin Med J (Eng) 98:219-222.

Hunter FT, Kip AF, Irvine V. 1942. Radioactive tracer studies on arsenic
injected as potassium arsenite. J Pharmacol Exp Ther 76:207.

IBT  (Industrial Bio-Test Laboratories). 1976. Acute Toxiclty Studies
with Phytar-560 in Albino Rabbits. Report to  the Ansul Company,
Marinette, VI.

Innes JRM et al. 1969. Bioassay of pesticides and  industrial chemicals
for  tumorigenicity in mice. A preliminary note. J  Natl Cancer Inst
42:1101-1114.

Irgolic K. 1982. Speciation of Arsenic Compounds in Vater Supplies.
Environmental Protection Agency, Health Effects Research Laboratory,
Cincinnati, OH. EPA  600/51-82-010.

-------
                                                        References   107

Ishinishi N, Yamamoto A, Hisanaga A, Inamasu T. 1983. Tumorigenicicy of
arsenic trioxide to the lung in Syrian golden hamsters by intermittent
instillations. Cancer Lett 21:141-147.
                •   ••
Ivankovic S, Eisenbrand G, Preussmann R. 1979. Lung carcinoma induction
in BD rats after a single intratracheal instillation of an arsenic-
containing pesticide mixture formerly used in vineyards. Int J Cancer
24:786-788.

Jacobson-Kram D, Montalbano 0. 1985. The reproductive effects assessment
group's report on the mutagenicity of inorganic arsenic. Environ
Hutagenesis 7:787-804.

Jelinek CF, Corneliussen PE. 1977. Levels of arsenic in the United
States food supply. Environ Health Perspect 19:83-87.

Johansen MG, McGowan JP, Tu SH, Shirachi DY. 1984. Tumorigenic effect of
dimethylarsinic acid in the rat. Proc Vest Pharmacol Soc 27:289-291.

Johnson DL, Braman RS. 1975. Alky1-and inorganic arsenic in air samples.
Chemosphere 6:333-338.

Jongen WMF, Cardinaals JM, Bos PMJ. 1985. Genotoxicity testing of
arsenobetaine, the predominant form of arsenic in marine fishery
products. Food Chem Toxicol 23(7):669-673.

Kadowaki K. 1960. Studies on the arsenic contents in organ tissues of
the normal Japanese. Osaka City Med J 9:2083-2099 (in Japanese with
English summary).

* Kaise T, Vatanabe S, Itoh K. 1985. The acute toxicity of
arsenobetaine. Chemosphere 14(9):1327-1332.

Kasper ML, Schoenfield L. Strom RL, Theologides A. 1984. Hepatic
angiosarcoma and bronchiolalveolar carcinoma induced by Fowler's
solution. J Am Med Assoc 252:3407-3408.

Kennedy S, Rice DA, Cush PF. 1986. Neuropathology of experimental  3-
nitro-4-hydroxyphenylarsonic acid toxicosis in pigs. Vet Pathol 23:454-
461.

Kerr KB, Cavett JW, Thompson OL.  1963. Toxicity of an organic arsenical,
3-nitro-4-hydroxylphenylarsonic  acid. I. Acute and subacute  toxicity.
Toxicol Appl Pharmacol 5:507-525.

Klaassen C. 1974. Biliary excretion of arsenic in rats, rabbits, and
dogs. Toxicol Appl Pharmacol 29:447-457.

Knoth W. 1966. Arsenbehandlung.  Arch Klin  Exp Derm 227:228-234.

Knowles FC, Benson AA. 1984. The  enzyme  inhibitory form of arsenic. Z
Gesante Hyg Grenzgeb  30:625-626.

-------
108   Section 10

Ko YC. 1986. A critical  review of epidemiologic studies on Blackfoot
disease. Sangyo Ika Daigaku Zasshi 8:339-353.

Kreiss K, Feldman RG,  Landrigan PJ, Boyd MH, Cox DH. 1983. Neurological
evaluation of a population exposed to arsenic in Alaskan well water.
Arch Environ Health 38:116.

Kuo TP, Chen MH. 1969. Follow-up study of surgical treatment of endemic
spontaneous gangrene  in  tropical Taiwan. J Formosan Med Assn 68:275-289.

Lagerkvist B, Linderholm H, Nordberg GF. 1986. Vasospastic tendency and
Raynaud's phenomenon  in  smelter workers exposed to arsenic. Environ Res
39:465-474.

Lakso JU, Peoples SA.  1975. Methylation of inorganic arsenic by mammals.
J Agric Food Chem 23:674-676.

Landau ED, Thompson DJ,  Feldman RG, Coble GJ, Dixon WJ. 1977. Selected
Noncarcinogenic Effects  of Industrial Exposure to Inorganic Arsenic.
Environmental Protection Agency, Washington, DC. EPA 569/6-77-018.

Lander JJ, Stanley RJ, Sumner HW, Boswell DC, Aach RD.   1975.
Angiosarcoma of the liver associated with Fowler's solution (potassium
arsenite). Gastroenterol 68:1582-1586.

Larramendy ML, Popescu NC, DiPaolo J. 1981. Induction by Inorganic metal
salts of sister chromatid exchanges and chromosome aberrations in human
and Syrian hamster strains. Environ Mutagenesis 3:597-606.

Lee AM, Fraumeni JF, Jr. 1969. Arsenic and respiratory cancer in man: An
occupational study. J Natl Cancer Inst 42:1045-1052.

Lee T-C,  Lee KG, Tzena Y J, Huang RY,  Jan KY. 1986. Sodium arsenite
potentiates the clastogenicity and mutagenicity of DNA crosslinking
agents. Environ Mutagenesis 8:119-128.

Lee T-C,  Oshimura M, Barrett JC. 1985. Comparison of arsenic-induced
cell transformation, cytotoxicity,  mutation, and cytogenetic effects in
Syrian hamster embryo cells in culture.  Carcinogenesis (London) 6:1421-
1426.

* Lee-Feldstein A.  1983. Arsenic and respiratory cancer in man: Follow-
up of an occupational study.  In: Lederer U, Fensterheim R, eds.
Arsenic:  Industrial, Biomedical, and Environmental Perspectives.  New
York: Van Nostrand Reinhold.

Lee-Feldstein A. 1986. Cumulative exposure to arsenic and its
relationship to respiratory cancer among copper smelter employees. J
Occup Med 28:296-302.

Leonard A, Lauwerys RR.  1980. Careinogenieity, teratogenicity, and
mutagenicity of arsenic. Mutat Res 75:49-62.

-------
                                                        References   109

Lerman S, Clarkson TV.  1983. The metabolism of arsenite and arsenate by
the rat. Fund Appl Toxlcol  3:309-314.

Lerman SA, Clarkson TV, Gerson RJ. 1983. Arsenic uptake and metabolism
by liver cells is dependent on arsenic oxidation state. Chem Biol
Interact 45:401-406.

Levander OA. 1977. Metabolic interrelationships between arsenic and
selenium. Environ Health Perspect 19:159-164.

Liebscher K, Smith H. 1968. Essential and nonessential trace elements:  A
method of determining whether an element is essential or nonessential in
human tissue. Arch Environ Health 17:881-890.

Lindgren A, Danielsson  BRG, Dencker L, Vahter M. 1984. Embryotoxicity of
arsenite and arsenate:  Distribution in pregnant mice and monkeys and
effects on embryonic cells  in vitro. Acta Pharmacol Toxicol 54:311-320.

Lofroth G, Ames BN. 1978. Mutagenicity of Inorganic compounds in
Salmonella typhiourium: Arsenic, chromium, and selenium. Hutat Res
54:65-66.

Love11 MA, Farmer JG. 1985. Arsenic speciation in urine from humans
intoxicated by inorganic arsenic compounds. Hum Toxicol 4:203-214.

Lubin JH, Pottern LM, Blot VJ, et al. 1981. Respiratory cancer among
copper smelter workers: Recent mortality statistics. J Met 23:779-784.

Lubin JH, Pottern LM, Blot VJ, Tokudome S, Stone BJ, Fraumeni JF. 1981.
Respiratory cancer among copper smelter workers: Recent mortality
statistics. J Occup Med 23:779-784.

Lundgren KD. 1954. Damage to respiratory organs in workers in a smelting
plant. Nord Hyg Tidskr  3:66-82.

Luten JB, Riekwel-Booy  G, Rauchbaar A. 1982. Occurrence of arsenic in
plaice (Pleuronectes platessa), nature of organoarsenic compound
present, and its excretion by man. Environ Health Perspect 45:165-170.

Mabuchi K, Lilienfeld A, Snell L. 1979. Lung cancer among pesticide
workers exposed to inorganic arsenicals. Arch Environ Health 34:312-320.

Mahaffey KR, Capar SG,  Gladen BC, Fowler BA. 1981. Concurrent exposure
to lead, cadmium, and arsenic. Effects on toxicity and tissue metal
concentrations in the rat. J Lab Clin Med 98:463-481.

Mahaffey KR. Fowler BA. 1977. Effects of concurrent administration of
lead, cadmium, and arsenic  in the rat. Environ Health Perspect 19:165-
171.

Nappes R. 1977. (Experiments on excretion of arsenic in urine.) Versuche
zur Ausscheidung von Arsen  in Urin. Int Arch Occup Environ Health
40:267-272.

-------
110   Section 20

Marafante E, Vahter M. 1984. The effect of methyltransferase inhibition
on the metabolism of As-74 arsenite in mice and rabbits. Chem Biol
Interact 50:49-57.

Marafante E, Vahter H. 1986. The effect of dietary and chemically
induced methylation deficiency on the metabolism of arsenate in the
rabbit. Acta Pharmacol Toxicol (Copenhagen) 59 (Suppl 7):35-38.

Harafante E, Vahter M. 1987. Solubility, retention, and metabolism of
intratracheally and orally administered inorganic arsenic  compounds in
the hamster. Environ Res 42:72-82.

Marafante E, Vahter M, Dencker L. 1984. Metabolism of arsenocholine in
mice, rats, and rabbits. Sci Total Environ 34:223-240.

Marafante E, Vahter M, Envall J.  1985. The role of the methylation in
the detoxification of arsenate in the rabbit.  Chem Biol Interact
56:225-238.

Marlowe M, Stellern J, Moon C, Errera J. 1985. Main and interaction
effects of metallic toxins on aggressive classroom behavior.  Aggressive
Behav 11:41-48.

Matanoski G, Landau E, Towascia J, et al.  1981.  Lung Cancer Mortality
in Proximity to a Pesticide Plant. Washington, DC: American Public
Health Association, Environmental Protection Agency, Office of Toxic
Substances. PB80-207376.

McCabe LJ, Symons JM, Lee RD, Robeck GG. 1970. Survey of community water
supply systems. J Am Water Works Assoc 62:670-687.

Mealey J, Jr, Browne11 GL, Sweet WH. 1959. Radioarsenic in plasma,
urine, normal tissues, and intracranial neoplasms. Arch Neurol Psychiatr
81:310-320.

* Mizuta N, Mizuta M, Ita F, et al.  1956. An outbreak of acute arsenic
poisoning caused by arsenic-contaminated soy sauce (shoye). A clinical
report of 220 cases. Bull Yamaguchi Med Sch 4:131-150.

Moody JP, Williams RT. 1964a. The fate of arsanilic acid and
acetylarsanilic acid in hens. Food Cosmet Toxicol 2:687-693.

Moody JP, Williams RT. 1964b. The fate of 4-nitrophenylarsonic acid in
hens. Food Cosmet Toxicol 2:695-706.

Moore LO. 1976. Mutagenesis and Carcinogenesis Tests with Monosodium
Methanearsenate, Cacodylic Acid, and Arsenic Trioxide by Bacteria and
Cell Culture Tests. Ueslaco, TX: The Ansul Company, Weslaco Technical
Center.

Morris JS, Schmid M, Newman S, Scheuer PJ, Sherlock S. 1974.  Arsenic and
noncirrhotic portal hypertension. Gastroenterology 64:86-94.

-------
                                                         References    111

 Mortelmans KE,  Riccio ES, Sheppard GF.  1980.  In vitro detection of
 mltotic crossing-over, mitotic gene conversion,  and reverse  mutation
 with S. cerevisiae D7 for 7 pesticides.  EPA Contract 68-02-2947.

 Morton W,  Starr G,  Pohl D,  Stoner J,  Wagner S,  Weswig P.  1976.  Skin
 cancer and water arsenic in Lane County,  Oregon.  Cancer 37:2523-2532.

 Munro 1C,  Charbonneau SM, Sandi E,  Spencer K,  Bryce F,  Grice HC.  1974.
 Biological availability of arsenic  from  fish.  Toxicol Appl Pharmacol
 29:111.

 Mushak P,  Dessauer  K,  Walls E.  1977.  Flaoeless  atomic absorption  (FAA)
 and  gas-liquid  chromatography studies in  arsenic  bioanalysis. Environ
 Health Perspect 19:5-10.

 * NagymaJtenyi  L, Selypes A,  Berencsi G.  1985. Chromosomal aberrations
 and  fetotoxic effects  of atmospheric  arsenic exposure in  mice  J Appl
 Toxicol  5:61-63.

 Nakamuro M, Sayato  Y.  1981.  Comparative studies of  chromosomal
 aberrations induced by trivalent and  pentavalent  arsenic. Mutat Res
 88:73-80.

 Narasaki H. 1985. Determination of  arsenic and selenium in fat materials
 and petroleum products by oxygen bomb combustion  and automated atomic
 absorption speccrometry with hydride  generation.  Anal Chem 57:2481-2486.

 NAS  (National Academy  of Sciences). 1977.  Arsenic.  Prepared  by Committee
 on Medical and  Biological Effects of  Environmental  Pollutants, Division
 of Medical Sciences, National Research Council, Washington,  DC.

 Nees PO. 1969.  Ansar 529, Acute Dermal Toxicity Study.  WARF  report to
 the Ansul Company,  Marinette, WI.

 Nelson WC,  Lykens MH,  Mackey J,  Newill VA,  Finkles  JF,  Hammer DI. 1973.
 Mortality among orchard workers exposed to lead arsenate  spray: A cohort
 study. J Chronic Dis 26:105-118.

 NIOSH  (National Institute for Occupational Safety and Health). 1975.
 Criteria for a Recommended  Standard:  Occupational Exposure to Inorganic
Arsenic. New Criteria-1975.  Washington, DC: Department  of Health,
 Education,  and Welfare. HEW  Publ  N.NIOSH  75-149.

NIOSH  (National Institute for Occupational Safety and Health). 1979.
Arsine (arsenic hydroxide) poisoning  in the workplace.  Current
 Intelligence Bulletin  32. Cincinnati, OH.

NIOSH  (National Institute for Occupational Safety and Health). 1984.
NIOSH Manual of Analytical Methods. 3rd ed. Cincinnati, OH: U.S.
Department  of Health and  Human  Services.

-------
112   Section 10

NIOSH (National Institute for Occupational Safety and Health). 1985.
1985 Supplement to NIOSH Manual of Analytical Methods. 3rd ed.
Cincinnati, OH: U.S. Department of Health and Human Services.

Nishioka H. 1975. Mutagenic activities of metal compounds in bacteria.
Mutat Res 31:185-189.

NLM (National Library of Medicine). 1987. Computer printout. Chemline
Data Base Record for Arsenic. Washington, DC: National Library of
Medicine.

Nordberg GF, Andersen 0. 1981. Metal interactions in carcinogenesis:
Enhancement, Inhibition. Environ Health Perspect 40:65-81.

Nordenson I, Salmonsson S, Brun E, Bechman G. 1979.  Chromosome
aberrations in psoriatic patients treated with arsenic. Hum
Genet 48:1-8.

Nordenson I, Sweins A, Beckman L. 1981. Chromosome aberrations in
cultured human lymphocytes exposed to trivalent and pentavalent arsenic.
Scand J Work Environ Health 7:277-281.

Nordstrom S, Beckman L, Nordenson .1. 1978a. Occupational and
environmental risks in and around a smelter in northern Sweden. I.
Variations in birth weight. Hereditas 88:43-46.

Nordstrom S, Beckman L, Nordenson 1. 1978b. Occupational and
environmental risks in and around a smelter in northern Sweden. III.
Frequencies of spontaneous abortion. Hereditas 88:51-54.

Nordstrom S, Beckman L, Nordenson I. 1978c. Occupational and
environmental risks in and around a smelter in northern Sweden. V.
Spontaneous abortion among female employees and decreased birth weight
in their offspring. Hereditas 90:291-296.

Nordstrom S, Beckman L, Nordenson I. 1978d. Occupational and
environmental risks in and around a smelter in northern Sweden. VI.
Congenital malformations. Hereditas 90:297-302.

Norin H, Vahter M. 1981. A rapid method for the selective analysis of
total urinary metabolites of inorganic arsenic. Scand J Work Environ
Health 7:38-44.

Oberly TJ, Piper CE, McDonald DS. 1982. Mutagenicity of metal salts in
the L5178Y mouse lymphoma assay. Toxicol Environ Health 9:367-376.

Odanaka Y, Tsuchiya N, Matano 0, Goto S. 1983. Determination of
inorganic arsenic and methylarsenic compounds by gas chromatography and
multiple ion detection mass spectrometry after hydride generation  -
heptane cold trap. Anal Chem 55:929-932.

-------
                                                        References   113

Okul T, Fujlwara Y.  1986.  Inhibition of human excision DNA repair by
inorganic arsenic and the  co-mutagenic effect in V79 Chinese hamster
cells. Mutat Res 172:69-76.

* OSHA (Occupational- Safety  and Health Administration). 1983.
Occupational exposure to inorganic  arsenic: Supplemental statement of
reasons for final rule. Fed  Regist  48: 1894 (01/14/83).

Ott MG, Holder BB, Gordon  HI.  1974.  Respiratory cancer and occupational
exposure to arsenicals. Arch Environ Health 29:250-255.

Palmer JS. 1972. Toxicity  of 45 Organic Herbicides to Cattle, Sheep, and
Chickens. USDA Production  Research  Report 137.

PAX. 1973. PAX Company Arsenic Advisory Committee. Report of the PAX
Company Advisory Committee to the Environmental Protection Agency. Salt
Lake City: PAX Company Arsenic Advisory Committee. Available by NTIS PB
265 964.

Peoples SA. 1964. Arsenic  toxicity  in cattle. Ann N Y Acad Sci 111:644-
649.

Peoples SA. et al. 1979. Occupational exposures to pesticides containing
organoarsenicals in  California. Vet Hum Toxicol 21:417-421.

Perry K, Bowler RG,  Buckell  HM, Druett HA. Schilling RSF. 1948. Studies
in the incidence of  cancer in a factory handling inorganic compounds of
arsenic--II: Clinical  and  environmental investigations. Br J Ind Med
5:6-15.

Pershagen G. 1985. Lung cancer mortality among men living near an
arsenic-emitting smelter.  Am J Epidemiol 122:684-696.

Pershagen G, BJorklund NE. 1985. On the pulmonary tumorigenicity of
arsenic trisulfide and calcium arsenate in hamsters. Cancer Lett
27:99-104.

Pershagen G, Lind B,  BJorklund NE.  1982. Lung retention and toxicity of
some inorganic arsenic compounds. Environ Res 29:425-434.

Pershagen G. Nordberg G, Bjorkland  NE. 1984a. Carcinomas of the
respiratory tract in hamsters given arsenic trloxide and/or benzo-a-
pyrene by the pulmonary route. Environ Res 37:425-432.

Pershagen G, Nordberg G, Bjorkland  NE. 1984b. Experimental evidence on
the pulmonary carcinogenicity of arsenic trioxide. Arch Toxicol, Suppl,
ISS Dis Metab Reprod Toxic Response Drugs Other Chem 7:403-404.

Pershagen G, Vail S,  Taube A, Linnman L. 1981. On the interaction
between occupational arsenic exposure and smoking and its relationship
to lung cancer. Scand J Work Environ Health 7:302-309.

-------
114   Section 10

Peters HA, Croft WA, WooIson EA, Darcey BA, Olson MA.   1984.  Seasonal
arsenic exposure froa burning chromium-copper-arsenate-treated wood.  J
An Had Assoc 251:2393-2396.

Peters HA, Croft WA, Woolson EA, Darcey BA, Olson MA.   1986.
Hematological, denial, and neuropsychological disease  from burning and
saving chromium-copper-arsenic (CCA)-treated wood. Acta Pharmacol
Toxicol (Copenhagen) S9(Suppl 7):39-43.

Philipp R. 1985. Arsenic exposure: Health effects and  the risk of
cancer. Rev Environ Health 5:27-57.

Pinto SS, Bennett BM. 1963. Effect of arsenic trioxide exposure on
mortality. Arch Environ Health 7:583-591.

Pinto SS, Enterline PE, Henderson V, Varner MO. 1977.  Mortality
experience in relation to a measured arsenic trioxide  exposure. Environ
Health Perspect 19:127-130.

Pinto SS, Henderson V, Enterline PE. 1978. Mortality experience of
arsenic-exposed workers. Arch Environ Health 33:325-331.

Pinto SS, McGill CM. 1953. Arsenic trioxide exposure in industry. Ind  -
Med Surg 22:281-287.

Pinto SS, Varner MO, Nelson Ktf, Labbe AL, White LD. 1976. Arsenic
trioxide absorption and excretion in industry. J Occup Med 18:677-680.

Poma K, Degraeve N, Kirsch-Volders M, Susanne C. 1981. Cytogenetic
analysis of bone marrow cells and spermatogonia of male mice after in
vivo treatment with arsenic. Experientia 37:129-130.

Potts AM, Gonasun LM. 1980. Toxic responses of the eye. In: Casarett LJ,
Ooull J, eds. Casarett and Doull's Toxicology. The Basic Science of
Poisons. New York: Macmillan Publishing, pp. 301-302.

Regelson W, Kim U, Osplna J, Holland JF. 1967.  Hemangioendothelial
sarcoma of liver froa chronic arsenic  intoxication by Fowler's solution.
Cancer 21:514-522.

Rencher AC, Carter MW, McKee DW.  1977. A retrospective  epidemiological
study of mortality at a large western  copper smelter. J Occup  Med
19:754-758.

Reymaim F. Moller R, Nielsen A.  1978.  Relationship between arsenic
intake and Internal malignant neoplasms. Arch  Dermatol  114:378-381.

Rhoads K. Sanders CL.  1985. Lung clearance,  translocation, and acute
toxlclty  of arsenic, beryllium,  cadmium,  cobalt,  lead,  selenium,
vanadiua, and ytterbium oxides  following deposition in  rat lung.  Environ
Res 36:359-378.

-------
                                                        References   115

Rice DA, Kennedy S, McMurray CH, Blanchflower WJ. 1985. Experimental 3-
nitro-4-hydroxyphenylarsonic acid toxicosis in pigs. Res Vet Sci 39-47-
51.

Roat JW, Vald A, Mendelow H, Pataki KI. 1982. Hepatic angiosarcoma
associated with short-term arsenic ingestion. Am J Med 73:933-936.

Rogers EH, Chernoff N, Kaylock RJ. 1981. The teratogenic potential of
cacodylic acid in the rat and mouse. Drug Chem Toxicol 4:49-61.

Rom V, Varely G, Lyon J. 1982. Lung cancer among residents living near
the El Paso smelter. Br J Ind Med 39:269-272.

Rosenberg HG. 1974. Systemic arterial disease and chronic arsenicism in
infants. Arch Pathol 97:360.

Rosenberg MJ, Landrigan PJ, Crowley S. 1980. Low-level arsenic exposure
in wood processing plants. Am J Ind Med 1:99-107 (1980).

Rossman TG. 1981. Enhancement of UV-mutagenesis by low concentrations of
arsenite in E. coll. Mutat Res 91:207-211.

Rossman TG, Stone D, Molina M, Troll W. 1980. Absence of arsenite
mutagenicity in EM coli and Chinese hamster cells. Environ Mutagenesis
2:371-379.

Roth F. 1958. The sequelae of chronic arsenic poisoning in Moselle
Vintners. Ger Med Mon 2:172-175.

RTECS (Registry of Toxic Effects of Chemical Substances). 1983.
Cincinnati, OH: National Institute for Occupational Safety and Health.

Salcedo JC, Portales A, Landecho E, Diaz R. 1984. Transverse study of a
group of patients with vasculopathy from chronic arsenic poisoning in
communities of the Francisco de Madero and San Pedro Districts,
Coahuila, Mexico. Revista de la Facultad de Medicina de Torreon 12:16.

Sax NI. 1984. Dangerous Properties of Industrial Materials. 6th ed. New
York, NY: Van Nostrand Reinhold.

Schaunburg HA. American Public Health Association. 1980. Failure to
Produce Arsenic Neurotoxicity in the Rat. An Experimental Study.
Washington, DC: Environmental Protection Agency, Office of Toxic
Substances.  EPA 560/11-80-022. PB 80-209505.

Schrauzer GN. 1987. Effects of selenium antagonists on cancer
susceptibility: New aspects of chronic heavy metal toxicity. Sangyo Ika
Daigaku Zasshi 9(Suppl):208-215.

Schrauzer GN, White DA. McGlnness JE. Schneider CJ, Bell LJ. 1978.
Arsenic and cancer: Effects of joint administration of arsenite and
selenite on the genesis of adenocarcinoma in inbred female C3H/St mice.
Biolnorg Chen 9:245-253.

-------
116   Section 10

Schroeder HA, Kitchener H. 1971. Toxic effects of trace elements on the
reproduction of mice and rats. Arch Environ Health 23:102-106.

Schwartz K. 1977. Essentiality versus toxicity of metals.  In:  Brown SS,
ed. Clinical Chemistry and Chemical Toxicology of Metals.  Vol.  1.
Amsterdam: Elsevier, pp. 3-22.

Sharpless GR. Metzger M. 1941. Arsenic and goiter. J Nutr 21:341-346.

Siewicki TC. 1981. Tissue retention of arsenic in rats fed with flounder
or cacodylic acid. J Nutr 111:602-609.

* Silver AS, Uainman PL. 1952. Chronic arsenic poisoning following use
of an asthma remedy. J Am Ned Assoc 150:584.

Simmon VF, Mitchell AD, Jorgensen TA. 1977. Evaluation of Selected
Pesticides as Chemical Mutagens. "In vitro" and "in vivo" studies. Palo
Alto, CA: Stanford Research Institute.

Simon I. 1932. Studii comparativi fra il dimetilarsinato di sodio ed il
manometilarsinato bisodico. I. Tossicita del metilarsinato bisodico
(arrhenal o metarsolo) ed alterazioni anatomo-patologiche
deliawelananento. Arch Int Pharmacodyn Ther 45:142-159.

Singh I. 1983. Induction of reverse mutation and mitotic gene conversion
by some metal compounds in Saccharomyces cerevLsLae. Mutat Res 117:149-
152.

Sittig M. 1985. Handbook of Toxic and Hazardous Chemicals and
Carcinogens. 2nd ed. Park Ridge, NJ: Noyes Publications.

* Sky-Peck HH. 1985. Trace metals and neoplasia. Clin Physiol Biochem
4:99-111.

Smith TJ, Crecelius EA, Reading JC. 1977. Airborne arsenic exposure and
excretion of methylated arsenic compounds. Environ Health Perspect
19:89-93.

Snegireff LS, Lombard OM. 1951. Arsenic and cancer. Observations  in the
metallurgical industry. Arch  Ind Hyg Occup Med 4:199-205.

Solomons NW. 1984. The other  trace minerals manganese, molybdenum,
vanadium, nickel, silicon, and arsenic. Nutrition and Disease.
Absorption and Malabsorption  of Mineral Nutrients.  New York: Alan R.
Liss. ISBN 0-8451-1611-8, Chap. 12.

* Sommers SC, McManus RG. 1953. Multiple arsenical cancers of the skin
and  internal organs. Cancer 6:347-359.

* Southwick JW,  Western AE, Beck MM,  et ml.   1981. Community Health
Associated with  Arsenic in Drinking Water  in  Millard County, Utah.
Health  Effects Research Laboratory, Cincinnati, OH: Environmental
Protection Agency.   EPA-600/1-81-064. NTIS  PB82-108374.

-------
                                                        References   117

* Stevens JT,  DiPasquale  LC.  Farmer JD.  1977b. The Evaluation of the
Acute Inhalation Toxicology of  Technical Grade Organoarsenical
Pesticides.  I.  Inhalation Toxicology.  Environmental Protection Agency.

Stevens JT,  Hall LL,  Farmer JD,  DiPasquale LC, Chemoff N, Durham UF.
1977a. Disposition of ^C and/or 74As-cacodylic acid in rats after
intravenous,  intratracheal, or  peroral administration. Environ Health
Perspect 19:151-157.

Stokinger HE.  1981. The metals.  In: Clayton CD, Clayton FE, eds. Patty's
Industrial Hygiene and Toxicology. 3rd ed.  Vol. 2A. Toronto: John Wiley
and Sons, pp.  1517-1530.

Suta BE. 1980.  Human  Exposure to Atmospheric Arsenic. Final report.
Stanford Research Institute CRESS Rep  50. EPA 68-01-4314 and 68-02-2835.

Sveins A. 1983. Protective effect of selenium against arsenic-induced
chromosomal  damage in cultured  human lymphocytes. Hereditas 98:249-252.

Takagi Y, Imai  S, Ohmori  Y, et  al. 1986. Trace elements in human hair:
An international comparison. Bull Environ Contain Toxicol 36:793-800.

Tarn GKH, Charbonneau  SM,  Bryce  F, Sandi  E. 1982. Excretion of a single
oral dose of  fish-arsenic in man. Bull Environ Contain Toxicol 28(6):
669-673.

Tarn GKH, Lacroix G. 1982.  Dry ashing,  hydride generation, atomic
absorption spectrometric  determination of arsenic and selenium in foods.
J Assoc Off Anal Chem 65:647-650.

Tarn KH,  Charbonneau SM, Bryce F,  Lacroix G. 1978. Separation of arsenic
metabolites  in  dog plasma and urine following intravenous injection of
74As.  Anal Biochem 86:505-511.

Tarn KH,  Charbonneau SM, Bryce E,  Lacroix G. 1979. Metabolism of
inorganic arsenic (74As)  in humans following oral ingestion. Toxicol
Appl Pharmacol  50:319-322.

* Tay C-H, Seah C-S.  1975. Arsenic poisoning from anti-asthmatic herbal
preparations. Med J Aust  2:424-428.

* Tseng VP, Chu HM. How SU, Fong JM, Lin CS, Yeh S. 1968. Prevalence of
skin cancer  in  an endemic area  of chronic arsenism in Taiwan. J Natl
Cancer Inst 40:453-463.

Tseng WP. 1977. Effects and dose- response relationship of skin cancer
and Blackfoot disease with arsenic. Environ Health Perspect 19:109-119.

Uthus E, Cornatzer W,  Nielsen F.  1983. Consequences of arsenic
deprivation  in  laboratory animals. In: Lederer V, Fensterheim R, eds.
Arsenic: Industrial,  Blomedical,  and Environmental Perspectives. New
York:  Van Nostrand Reinhold.

-------
118   Section 10

Vahter H. 1981. Blotransformation of crivalent and pentavalent inorganic
arsenic in mice and rats. Environ Res 25:286-293.

Vahter M. 1983. Metabolism of arsenic.  In:  Fowler  BA,  ed.  Biological and
Environmental Effects of Arsenic. New York:  Elsevier,  pp.  171-198.

Vahter M, Friberg L, Rahnster B, Nygren A,  Nolinder P.  1986.  Airborne
arsenic and urinary excretion of metabolites of inorganic arsenic among
smelter workers. Int Arch Occup Environ Health 57:79-91.

Vahter M, Marafante E. 1983. Intracellular interaction and metabolic
fate of arsenite and arsenate in mice and rabbits. Chem Biol Interact
47:29-44.

Vahter M, Marafante E. 1985. Reduction and binding of arsenate in
marmoset monkeys. Arch Toxicol  57:119-124.

Vahter M, Marafante E, Dencker  L. 1983. Metabolism of arsenobetaine in
mice, rats, and rabbits. Sci Total Environ 30:197-211.

Vahter M, Marafante E, Dencker  L. 1984. Tissue distribution and
retention of 74As-dimethylarsinic acid in mice and rats. Arch Environ
Contain Toxicol 13:259-264.

Vahter M, Norin H.  1980. Metabolism  of 74As-labelled trivalent and
pentavalent inorganic arsenic in mice. Environ Res 21:446-457.

Valentine JL,  Campion DS,  Schluchter MD, Massey FJ. 1981. Arsenic
effects  on human nerve conduction. In: Howell JC, Hawthorne JM.  White  L,
eds. Trace Element  Metabolism in Man and Animals. Canberra: Australian
Academy  of Science, pp.  409-411.

Valentine J, Rang H,  Spivery G. 1979. Arsenic  levels  in human blood,
urine, and hair  in  response to  exposure via drinking water.  Environ Res
20:24-32.

* Valentine JL,  Reisbord LS, Kang HK,  Schluchter  MD.  1985. Arsenic
effects  on population health histories.  In: Mills CF,  Bremner I,
Chesters JK, eda. Trace  Elements in Man and Animals  -  TEMA5.  Proceedings
of the 5th  International Symposium.  Slough, UK: Commonwealth
Agricultural Bureau.

Vallee BL. Ulmer DD,  Wacker UEC. 1960.  Arsenic toxicology and
biochemistry.  Arch Ind Health 21:132-151.

Viallet A,  Guillauae I,  Cote J, Legare A,  Lavoie  P.  1972. Presinusoidal
 portal hypertension following chronic arsenic intoxication.
 Gastroenterology 62:177.

 Vig BK,  Figueroa ML, Cornforth MN.  Jenkins SH. 1984.  Chromosome studies
 in human subjects chronically  exposed to arsenic in drinking water. Am J
 Ind Med 6:325-338.

-------
                                                        References   119

Wagner SL, Maliner JS, Morton WE, Branan RS. 1979. Skin cancer and
arsenic intoxication  from well water. Arch Dermatol 115:1205-1207.

Wahlberg JE, Boman A. 1986. Contact sensitivity to arsenical compounds:
Clinical and experimental studies. Derm Beruf Umwelt 34:10-12.

Walsh LM, Keeney DR.  1975. Behavior and phytotoxiclty of inorganic
arsenicals in soils.  In: Arsenic Pesticides. ACS Symposium Series 7.
Washington, DC: American Chemical Society, p. 35.

Walsh PR, Duce RA, Fasching JL. 1979. Considerations of the enrichment,
sources, and flux of  arsenic in the troposphere. J Geophys Res
84(C4):1719-1726.

Wan B, Christian RT,  Soukup SW. 1982. Studies of cytogenetic effects of
sodium arsenicals on  mammalian cells in vitro. Environ Mutagenesis
4:493-498.

Watrous RM, McCaughey MB. 1945. Occupational exposure to arsenic--In the
manufacture of arsphenamine and related compounds. Ind Med 14:639-646.

Weast RC. ed. 1985. CRC Handbook of Chemistry and Physics. 66th ed. Boca
Raton, FL: CRC Press.

Webb DR, Sipes IG, Carter DE. 1984. In vitro solubility and in vivo
toxicity of gallium arsenide. Toxicol Appl Pharmacol 76:96-104.

Webb DR. Wilson SE, Carter DE. 1986. Comparative pulmonary toxicity of
gallium arsenide, gallium (III) oxide, or arsenic (III) oxide
intratracheally instilled into rats. Toxicol Appl Pharmacol 82:405-416.

Weed Society of America. 1967. Herbicide Handbook of the Weed Society of
America. Geneva, NY:  W. F. Humphrey, pp. 134-137.

Welch K, Higgins I, Oh M, Burchfiel C. 1982. Arsenic exposure, smoking,
and respiratory cancer in copper smelter workers. Arch Environ Health
37:325-335.

Wentworth W. 1983. Compliance Monitoring Data from Federal Reporting
Data System. Environmental Protection Agency, Office of Drinking Water.
State Programs Division.

Westoo G. Rydalv M. 1972. Arsenic levels in foods. Var Foda 24:21-40 (in
Swedish with English  summary).

Whitnack GC, Martens  BH. 1971. Arsenic in potable desert groundwater: An
analysis problem. Science 171(1969):383-385.

WHO (World Health Organization). 1984. Guidelines for Drinking Water
Quality. Vol. 2. Health Criteria and Other Supporting Information.
Geneva, Switzerland.

-------
120   Section 10

Willhite CC. 1981. Arsenic - induced axial skeletal (dysraphic) disorders.
Exp Mol Pathol 34:145-158.

Willhite CC, Perm VH. 1984. Prenatal and developmental toxicology of
arsenicals. Adv Exp Med Biol 177:205-228.

Wood JH. 1974. Biological cycles for toxic elements in the environment.
Science 188:1049-1052.

Woods JS, Fowler BA. 1977. Effects of chronic arsenic exposure on
hematopoietic function in adult mammalian liver.  Environ Health Perspect
19:209-213.

Woods JS, Fowler BA. 1978. Altered regulation of mammalian hepatic heme
biosynthesis and urinary porphyrin excretion during prolonged exposure
to sodium arsenate. Toxicol Appl Pharmacol 43:361-371.

Woolson EA. 1976. Generation of dimethyl arsine from soil. Paper
presented at 16th meeting. Weed Science Society of America.

Yamauchi H, Yamamura Y. 1984. Metabolism and excretion of orally
administered dimethylarsinic acid in the hamster. Toxicol Appl Pharmacol
74:134-140.

Yamauchi H, Yamamura Y. 1985. Metabolism and excretion of orally
administered arsenic trioxide in the hamster. Toxicology 34:113-121.

Yu HS. 1984. Blackfoot disease and chronic arsenism in southern Taiwan.
Int J Dennatol 23:258-260.

Yu HS, Sheu HM, Ko  SS, et al. 1984. Studies on Blackfoot disease and
chronic arsenism  in southern Taiwan: With special reference  to skin
lesions and fluorescent substances. J Dermatol (Tokyo) 11:361-370.

* Zaldivar R. 1974. Arsenic contamination of drinking water  and
foodstuffs causing  endemic chronic poisoning. Beitr Path  151:384-400.

Zaloga GP, Deal J,  Spurling T, Richter J, Chernow B.  1985. Unusual
manifestations of arsenic  intoxication. Am J Med Sci  289:210-214.

-------
                                                                     121
                             11.  GLOSSARY

Acute Exposure -- Exposure to a chemical for a duration of 14 days or
less, as specified in the Toxicological Profiles.

Bioconcentration Factor (BCF)--The quotient of the concentration of a
chemical in aquatic organisms at a specific time or during a discrete
time period of exposure divided by the concentration in the surrounding
water at the same time or during the same time period.

Carcinogen--A chemical capable of inducing cancer.

Ceiling value (CL)--A concentration of a substance that should not be
exceeded, even instantaneously.

Chronic Exposure--Exposure to a chemical for 365 days or more, as
specified in the Toxicological Profiles.

Developmental Toxlclty--The occurrence of adverse effects on the
developing organism that may result from exposure to a chemical prior to
conception (either parent), during prenatal development, or postnatally
to the time of sexual maturation. Adverse developmental effects may be
detected at any point in the life span of the organism.

Embryotoxicity and Fetotoxicity--Any toxic effect on the conceptus as a
result of prenatal exposure to a chemical; the distinguishing feature
between the two terms is the stage of development during which the
insult occurred. The terms, as used here, include malformations and
variations, altered growth, and in utero death.

Frank Effect Level (FED--That level of exposure which produces a
statistically or biologically significant increase in frequency or
severity of unmistakable adverse effects, such as irreversible
functional impairment or mortality, in an exposed population when
compared with its appropriate control.

EPA Health Advisory—An estimate of acceptable drinking water levels for
a chemical substance based on health effects information. A health
advisory is not a legally enforceable federal standard, but serves as
technical guidance to assist federal, state, and local officials.

Immediately Dangerous to Life or Health (IDLH)--The maximum
environmental concentration of a contaminant from which one could escape
within 30 min without any escape-impairing symptoms or irreversible
health effects.

-------
122   Section 11

Intermediate Exposure--Exposure to a chemical for a duration of 15-364
days, as specified in the Toxicolbgircal Profiles.

Immunologic Toxicity--The occurrence'of adverse effects on the immune
system that may result from exposure to 'environmental agents such as
chemicals.

In vitro--Isolated from the riving organism and artificially maintained,
as in a test tube.

In vivo—Occurring within the living organism.

Key Study--An animal or human toxicological study that best illustrates
the nature of the adverse effects produced and the doses associated with
those effects.

Lethal Concentration(LO) (LCLO)--The lowest concentration of a chemical
in air which has been reported to have caused death in humans or
animals.

Lethal Concentration(50) (LCSO)--A calculated concentration of a
chemical in air to which exposure for a specific length of time is
expected to cause death in 50% of a defined experimental animal
population.

Lethal Dose(LO) (LDLO)--The lowest dose of a chemical introduced by a
route other than inhalation that is expected to have caused death in
humans or animals.

Lethal Doso(SO) (LDSO)--The dose of a chemical which has been calculated
to cause death in 50% of'a defined experimental animal population.

Lowest-Observed-Adverse-Effect Level (LOAEL)--The lowest dose of
chemical-in~a study or group of studies which produces statistically or
biologically significant increases in frequency or severity of adverse
effects between the exposed population and its appropriate control.

Lovest-Observed-Effect Level (LOEL)--The lowest dose of chemical in a
study or group of studies which produces statistically or biologically
'significant increases in frequency or severity of effects between the
exposed population and its appropriate control.

Halfornations--Permanent structural changes that may adversely affect
survival, development, or'function.

Minia&l Risk Level--An estimate of daily human exposure to a chemical
that is likely to be without an appreciable risk of deleterious effects
(nohcancerous)' over a specified duration of exposure.

Mut«gon--A substance that causes mutations. A mutation is a change in
the genetic material in a body cell. Mutations can lead to birth
defects, miscarriages, or cancer.

-------
                                                          Glossary   123

Reurotoxicity--The occurrence of adverse effects on the nervous system
following exposure to a> chemical-.

No-Observed-Adverse-Effect Level (NOAEL)--That dose of chemical at which
there are no statistically oc biologically significant increases in
frequency or severity of adverse effects seen between the exposed
population and its appropriate control. Effects may be produced at this
dose, but they are not considered to be adverse.

No-Observed-Effect Level (NOEL)--That dose of chemical at which there
are no statistically or biologically significant increases in frequency
or severity of effects seen between the exposed population and its
appropriate control.

Permissible Exposure Limit (PEL)--An allowable exposure level in
workplace air averaged over an 8-h shift.

q*--The upper-bound estimate of the low-dose slope of the dose-response
curve as determined by the multistage procedure. The q^* can be used to
calculate an estimate of carcinogenic potency, the incremental excess
cancer risk per unit of exposure (usually pg/L for water, mgAg/day for
food, and Mg/«3 for air),.

Reference Dose (RfD)--An estimate (with uncertainty spanning perhaps an
order of magnitude) of the daily exposure of the human population to a
potential hazard that is likely to be without risk of deleterious
effects during a lifetime. The RfD la operationally derived from the
NOAEL (from animal and human studies; by a consistent application of
uncertaincy factors that reflect various types of data used to estimate
RfDs and an additional modifying .factor  which  is based on a
professional judgment of the entire^database on the chemical. The RfDs
are not applicable to nonthreshold effects such as cancer.

Reportable Quantity (RQ)--The quantity of a hazardous .substance that  is
considered reportable under CERCLA. Reportable  quantities are:  (1)  1  Ib
or greater or  (2) for selected substances, an amount established by
regulation either under CERCLA or under  Sect. 311 of the Clean Water
Act. Quantities are measured over a 24-h period.

Reproductive Toxicity--The occurrence  of adverse.effects on the
reproductive system that may result from exposure; to.a chemical. The
toxicity may be directed to the  reproductive organs and/or the  related
endocrine system. The manifestation of such  toxicity may be noted  as
alterations In sexual behavior,  fertility, pregnancy outcomes,  or
modifications  in other  functions  that  are dependent on  tne integrity  of
this system.

Short-Term Exposure Limit  (STEL)--The  maximumr.concentration  to which
workers can be exposed  for up  to 15 mln continually. No more  than four
excursions are allowed  per day,  and  there must  be  at  least 60 min
between exposure periods. The  daily TLV-TWA may not.be  exceeded.

-------
 124   Section 11

 Target Organ Toxiclty--This term covers a broad range of adverse effects
 on target organs or physiological systems (e.g.,  renal,  cardiovascular)
 extending from thos* vising through a single  limited exposure to those
 assumed over a lifetime of exposure to a chemical.

 Teratogen--A chemical that causes structural defects  that affect the
 development  of an organism.

 Threshold Limit Value (TLV)-.A concentration of a substance  to which
 most workers can be exposed without adverse effect. The  TLV  mav be
 expressed as a TWA,  as  a STEL,  or as a CL

 Time-weighted Average (WA)-.-An allowable exposure concentration
 averaged  over a normal  8-h workday or 40-h workweek.

 Uncertainty  Factor  (UF)--A factor used "in, operationally  deriving the RfD
 from experimental data.  UFs  are intended  to account for  (1)  the
variation in sensitivity among,-the  members of the human  population
 (2) the uncertainty  in-extrapolating animal data  to the  case of  humans
 (3) the..uncertainty  In  extrapolating from data obtained  in a study that
TJL«  I688       lifetime  exposure,  and-(4) the uncertainty in  using
LOAEL data rather than  NOAEL data;  Usually each of these  factors  is set
equal to  10.

-------
                                                                     125
                         APPENDIX:  PEER REVIEW

     A peer review panel was assembled for arsenic. The panel consisted
of the following members: Dr. J.B. Andelman, University of Pittsburg;
Dr. T.W. Clarkson, University of Rochester; and Dr. R.D. Hood,
University of Alabdma. These experts collectively have knowledge of
arsenic's physical and chemical properties, toxicokinetics. key health
end points, mechanisms of"action, human and animal exposure, and
quantification of risk to humans. All reviewers were selected in
conformity with the ponditions for peer review specified.* in the
Superfund Amendments and Reauthorization Act of 1986, Section 110.

     A Joint panel of scientists from ATSDR and EPA has reviewed the
peer reviewers' comments and.determined.which comments will be included
in the profile. A listing of the peer reviewers' comments not
incorporated in the profile  with- a brief explanation of the rationale
for their exclusion, exists as part of the administrative record for
this compound. A list of databases reviewed and a list of unpublished
documents cited are also included in the administrative record.

     The citation of the peer review panel should not be understood to
imply their approval of the profile's final content. The responsibility
for the content of this profile lies with the Agency for Toxic
Substances and Disease Registry.

-------