DRAFT - DO NOT CITE OR QUOTE
EPA/635/R-11/013A
www.epa.gov/iris
Toxicological Review of Ammonia
(CAS No. 7664-41-7)
In Support of Summary Information on the
Integrated Risk Information System (IRIS)
June 2012
NOTICE
This document is an External Review draft. This information is distributed solely for the purpose
of pre-dissemination peer review under applicable information quality guidelines. It has not been
formally disseminated by EPA. It does not represent and should not be construed to represent any
Agency determination or policy. It is being circulated for review of its technical accuracy and
science policy implications.
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
-------
Toxicological Review of Ammonia
DISCLAIMER
This document is a preliminary draft for review purposes only. This information is
distributed solely for the purpose of pre-dissemination peer review under applicable
information quality guidelines. It has not been formally disseminated by EPA. It does not
represent and should not be construed to represent any Agency determination or policy.
Mention of trade names or commercial products does not constitute endorsement of
recommendation for use.
This document is a draft for review purposes only and does not constitute Agency policy.
ii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
CONTENTS
AUTHORS | CONTRIBUTORS | REVIEWERS vi
PREFACE viii
PREAMBLE TO IRIS TOXICOLOGICAL REVIEWS xi
EXECUTIVE SUM MARY xxiii
LITERATURE SEARCH STRATEGY | STUDY SELECTION xxvii
1. HAZARD IDENTIFICATION 1-1
1.1. Synthesis of Evidence 1-1
1.1.1. Respiratory Effects 1-1
1.1.2. Gastrointestinal Effects 1-12
1.1.3. Reproductive and Developmental Effects 1-18
1.1.4. Immune System Effects 1-19
1.1.5. Other Systemic Effects 1-24
1.1.6. Carcinogenicity 1-32
1.2. Summary and Evaluation 1-35
1.2.1. Effects Other than Cancer 1-35
1.2.2. Carcinogenicity 1-36
1.2.3. Susceptible Populations and Lifestages 1-36
2. DOSE-RESPONSE ANALYSIS 2-1
2.1. Oral Reference Dose for Effects Other than Cancer 2-1
2.2. Inhalation Reference Concentration for Effects Other than Cancer 2-2
2.2.1. Identification of Candidate Principal Studies and Critical Effects 2-2
2.2.2. Methods of Analysis 2-5
2.2.3. Derivation of the Reference Concentration 2-5
2.2.4. Uncertainties in the Derivation of the Reference Concentration 2-7
2.2.5. Confidence Statement 2-8
2.2.6. Previous IRIS Assessment: Reference Concentration 2-8
2.3. Cancer Risk Estimates 2-9
REFERENCES R-l
This document is a draft for review purposes only and does not constitute Agency policy.
iii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
TABLES AND FIGURES
Table ES-1. Summary of reference concentration (RfC) derivation xxiv
Table LS-1. Details of the literature search strategy xxviii
Table 1-1. Evidence pertaining to respiratory effects in humans following inhalation
exposure 1-6
Table 1-2. Evidence pertaining to respiratory effects in animals following inhalation
exposure 1-9
Table 1-3. Evidence pertaining to gastrointestinal effects in animals following oral
exposure 1-15
Table 1-4. Evidence pertaining to reproductive and developmental effects in animals
following inhalation exposure 1-18
Table 1-5. Evidence pertaining to immune system effects in animals following inhalation
exposure 1-21
Table 1-6. Evidence pertaining to other systemic effects in humans following inhalation
exposure 1-26
Table 1-7. Evidence pertaining to other systemic effects in animals following oral exposure .. 1-26
Table 1-8. Evidence pertaining to other systemic effects in animals following inhalation
exposure 1-27
Table 1-9. Evidence pertaining to cancer in animals following oral exposure 1-34
Figure LS-1. Literature search and study selection strategy for ammonia xxix
Figure 1-1. Exposure-response array of respiratory effects following inhalation exposure 1-11
Figure 1-2. Exposure-response array of gastrointestinal effects following oral exposure 1-16
Figure 1-3. Exposure-response array of immune system effects following inhalation
exposure 1-23
Figure 1-4. Exposure-response array of systemic effects following inhalation exposure 1-31
Figure 2-1. Exposure-response array of toxicological effects following inhalation exposure 2-3
This document is a draft for review purposes only and does not constitute Agency policy.
iv DRAFT—DO NOT CITE OR QUOTE
-------
ABBREVIATIONS
Toxicological Review of Ammonia
ACGIH American Conference of Governmental NHs
Industrial Hygienists NfV
ALT alanine aminotransferase NIOSH
AST aspartate aminotransferase
ATSDR Agency for Toxic Substances and Disease NOAEL
Registry NRC
BCG bacillus Calmette-Guerin ORD
BMC benchmark concentration
BMD benchmark dose PBPK
CAC cumulative ammonia concentration PEF
CCRIS Chemical Carcinogenesis Research PEFR
Information System POD
CPU colony forming unit PPD
EPA Environmental Protection Agency RDso
FDA Food and Drug Administration RfC
FEVi forced expiratory volume in 1 second RfD
FVC forced vital capacity RTECS
HERO Health and Environmental Research
Online TLV
HSDB Hazardous Substances Data Bank TSCATS
IgE immunoglobulin E
IgG immunoglobulin G UF
IRIS Integrated Risk Information System UFA
LCso 50% lethal concentration UFH
LD50 50% lethal dose UFL
LOAEL lowest-observed-adverse-effect level UFS
MAO monoamine oxidase UFo
MNNG N-methyl-N'-nitro-N-nitrosoguanidine VEh
MRM murine respiratory mycoplasmosis
NCEA National Center for Environmental VEho
Assessment
ammonia
ammonium ion
National Institute for Occupational
Safety and Health
no-observed-adverse-effect level
National Research Council
EPA's Office of Research and
Development
physiologically based pharmacokinetic
peak expiratory flow
peak expiratory flow rate
point of departure
purified protein derivative
50% response dose
reference concentration
reference dose
Registry of Toxic Effects of Chemical
Substances
threshold limit value
Toxic Substance Control Act Test
Submission Database
uncertainty factor
interspecies uncertainty factor
intraspecies uncertainly factor
LOAEL to NOAEL uncertainty factor
subchronic-to-chronic uncertainty factor
database deficiencies uncertainty factor
human occupational default minute
volume
human ambient default minute volume
This document is a draft for review purposes only and does not constitute Agency policy.
v DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
AUTHORS | CONTRIBUTORS | REVIEWERS
Assessment Team
Audrey Galizia, Dr. PH (Chemical
Manager)
James Ball, Ph.D.
Keith Salazar, Ph.D.
Christopher Sheth, Ph.D.
Louis D'Amico, Ph.D.
Christopher Brinkerhoff, Ph.D.
U.S. EPA
Office of Research and Development
National Center for Environmental Assessment
Edison, NJ
U.S. EPA
Office of Research and Development
National Center for Environmental Assessment
Washington, DC
Oak Ridge Institute for Science and Education
fellow
Scientific Support Team
Vincent Cogliano, Ph.D.
Samantha Jones, Ph.D.
Jamie Strong, Ph.D.
Ted Berner, MS
Jason Fritz, Ph.D.
Martin Gehlhaus, MPH
John Stanek, Ph.D.
U.S. EPA
Office of Research and Development
National Center for Environmental Assessment
Washington, DC
U.S. EPA
Office of Research and Development
National Center for Environmental Assessment
Research Triangle Park, NC
Production Team
Maureen Johnson
Vicki Soto
Ellen F. Lorang, MA
U.S. EPA
Office of Research and Development
National Center for Environmental Assessment
Washington, DC
U.S. EPA
Office of Research and Development
National Center for Environmental Assessment
Research Triangle Park, NC
Contractor Support
SRC, Inc., Chemical, Biological and Environmental Center, Syracuse, NY
Amber Bacom, MS
Fernando Llados, Ph.D.
Julie Stickney, Ph.D.
Executive Direction
Vincent Cogliano, Ph.D. U.S. EPA
Lynn Flowers, Ph.D. Office of Research and Development
Samantha Jones, Ph.D. National Center for Environmental Assessment
Susan Rieth, MPH Washington, DC
This document is a draft for review purposes only and does not constitute Agency policy.
vi DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Internal Review Team
Marian Rutigliano, MD
John Whalan
Amanda S. Persad, Ph.D.
Paul Reinhart, Ph.D.
U.S. EPA
Office of Research and Development
National Center for Environmental Assessment
Washington, DC
U.S. EPA
Office of Research and Development
National Center for Environmental Assessment
Research Triangle Park, NC
Reviewers
This assessment was provided for review to scientists in EPA's Program and Regional Offices.
Comments were submitted by:
Office of Policy, Washington, DC
Office of Water, Washington, DC
Office of Children's Health Protection, Washington, DC
Office of Transportation and Air Quality in the Office of Air and Radiation, Ann Arbor, Michigan
Office of Air Quality and Planning Standards in the Office of Air and Radiation, Washington, DC
Region 2, New York, New York
This assessment was provided for review to other federal agencies and the Executive Office of the
President Comments were submitted by:
Agency for Toxic Substances and Disease Registry, Centers for Disease Control and
Prevention, Department of Health & Human Services
Council on Environmental Quality, Executive Office of the President
Food Safety and Inspection Service, U.S. Department of Agriculture
This document is a draft for review purposes only and does not constitute Agency policy.
vii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
PREFACE
This Toxicological Review critically reviews the publicly available studies on ammonia in
order to identify its adverse health effects and to characterize exposure-response relationships.
The assessment covers gaseous ammonia (NH3) and ammonia dissolved in water (ammonium
hydroxide, NH4OH). It was prepared under the auspices of EPA's Integrated Risk Information
System (IRIS) program.
Ammonia and ammonium hydroxide are listed as hazardous substances under the
Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (CERCLA) and
ammonia is found at about 8% of hazardous waste sites on the National Priorities List (ATSDR,
2004]. Ammonia is subject to reporting requirements for the Toxics Release Inventory under the
Emergency Planning and Community Right-to-Know Act of 1986 and to emergency planning
requirements under section 112(r) of the Clean Air Act
This assessment updates a previous IRIS assessment of ammonia that was developed in
1991. The previous assessment included only an inhalation reference concentration for effects
other than cancer. New information has become available, and this assessment reviews
information on all health effects by all exposure routes.
This assessment was conducted in accordance with EPA guidance, which is cited and
summarized in the Preamble to IRIS Toxicological Reviews. The findings of this assessment and
related documents produced during its development are available on the IRIS website
(http://www.epa.gov/iris/). Appendices for chemical and physical properties, the toxicity of
ammonium salts, toxicokinetic information, summaries of toxicity studies and other information
are provided as Supplemental Information to this assessment (see Appendices A to D).
Portions of this Toxicological Review were adapted from the Toxicological Profile for
Ammonia developed by the Agency for Toxic Substances and Disease Registry (ATSDR, 2004] under
a Memorandum of Understanding that encourages interagency collaboration, sharing of scientific
information, and more efficient use of resources.
On December 23, 2011, The Consolidated Appropriations Act, 2012, was signed into law1.
The report language included direction to EPA for the IRIS Program related to recommendations
provided by the National Research Council (NRC) in their review of EPA's draft IRIS assessment of
formaldehyde. The NRC's recommendations, provided in Chapter 7 of their review report, offered
suggestions to EPA for improving the development of IRIS assessments. The report language
included the following:
'Pub. L. No. 112-74, Consolidated Appropriations Act, 2012.
This document is a draft for review purposes only and does not constitute Agency policy.
viii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
The Agency shall incorporate, as appropriate, based on chemical-specific datasets
and biological effects, the recommendations of Chapter 7 of the National Research
Council's Review of the Environmental Protection Agency's Draft IRIS Assessment of
Formaldehyde into the IRIS process...For draft assessments released in fiscal year
2012, the Agency shall include documentation describing how the Chapter 7
recommendations of the National Academy of Sciences (NAS) have been
implemented or addressed, including an explanation for why certain
recommendations were not incorporated.
Consistent with the direction provided by Congress, documentation of how the recommendations
from Chapter 7 of the NRC report have been implemented in this assessment is provided in
Appendix E. This documentation also includes an explanation for why certain recommendations
were not incorporated.
For additional information about this assessment or for general questions regarding IRIS,
please contact EPA's IRIS Hotline at 202-566-1676 (phone), 202-566-1749 (fax), or
hotline.iris@epa.gov.
Chemical Properties and Uses
Ammonia is a corrosive gas with a pungent odor. It is highly soluble in water (up to
482 g/L) and is a weak base fLide. 2008: O'NeiletaL 2006: Eggeman. 2001: Dean. 19851
Additional information on the chemical and physical properties of ammonia is presented in
Appendix A.
About 80% of commercially produced ammonia is used in agricultural fertilizers. Ammonia
is also used as a corrosion inhibitor, in water purification, as a household cleaner, as an
antimicrobial agent in food products, as a refrigerant, as a stabilizer in the rubber industry, as a
source of hydrogen in the hydrogenation of fats and oils, and as a chemical intermediate in the
production of Pharmaceuticals, explosives, and other chemicals. Ammonia is also used to reduce
nitrogen oxide emissions from combustion sources such as industrial and municipal boilers, power
generators, and diesel engines (HSDB, 2012: Johnson etal., 2009: Eggeman, 2001).
Ammonia is a component of the global nitrogen cycle and is essential to many biologic
processes. Nitrogen-fixing bacteria convert atmospheric nitrogen to ammonia that is available for
uptake into plants. Organic nitrogen released from biota can be converted to ammonia. Ammonia
in water and soil can be converted to nitrite and nitrate through the process of nitrification.
Ammonia is also endogenously produced in humans and other mammals, where it is an essential
metabolite used in nucleic acid and protein synthesis, is necessary for maintaining acid-base
balance, and is an integral part of nitrogen homeostasis (Nelson and Cox, 2008: Socolow, 1999:
Rosswall. 1981). This assessment compares endogenous levels of ammonia in humans to the
toxicity values that it derives.
This document is a draft for review purposes only and does not constitute Agency policy.
ix DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Consideration of Ammonium Salts for Inclusion in This Assessment
EPA considered whether to include ammonium salts (e.g., ammonium acetate, chloride, and
sulfate) in this assessment. These salts readily dissolve in water through dissociation into an
ammonium cation (NH4+) and an anion. Oral toxicity studies on ammonium chloride and
ammonium sulfate suggest that these salts may differ in toxicity (see Appendix B for a summary of
subchronic/chronic toxicity information for selected ammonium salts), but it is not clear whether
this reflects differences between the salts or in the effects that were studied. If the toxicity of the
salts is affected by the anion, then it would not be correct to attribute toxic effects to the ammonium
cation. ATSDR considered this question and concluded, "... that it would be inappropriate to
extrapolate findings obtained with ammonium chloride (or any ammonium salt) to equivalent
amounts of ammonium, but derived from a different salt" (ATSDR. 2004). Similarly, the World
Health Organization considered ammonium chloride-induced kidney hypertrophy and observed
that the extent to which it results from ammonium chloride-induced acidosis or from a direct effect
of the ammonium ion is not clear (IPCS, 1986). Thus, in light of the uncertain influence of the anion
on toxicity, ammonium salts were not used in the identification of effects or in the derivation of
reference values for ammonia and ammonium hydroxide.
Assessments by Other National and International Health Agencies
Toxicity information on ammonia has been evaluated by ATSDR, the National Research
Council, the American Conference of Governmental Industrial Hygienists, the National Institute for
Occupational Safety and Health, and the Food and Drug Administration. The results of these
assessments are presented in Appendix C. It is important to recognize that these earlier
assessments were prepared for different purposes using different methods and could consider only
the studies that were available at the time.
This document is a draft for review purposes only and does not constitute Agency policy.
x DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
PREAMBLE TO IRIS TOXICOLOGICAL REVIEWS
1. Scope of the IRIS Program
Soon after EPA was established in 1970, it was
at the forefront of developing risk assessment as a
science and applying it in decisions to protect
human health and the environment. The Clean Air
Act, for example, mandates that EPA provide "an
ample margin of safety to protect public health";
the Safe Drinking Water Act, that "no adverse
effects on the health of persons may reasonably be
anticipated to occur, allowing an adequate margin
of safety." Accordingly, EPA uses information on
the adverse effects of chemicals and on exposure
levels below which these effects are not anticipated
to occur.
IRIS assessments critically review the publicly
available studies to identify adverse health effects
from long-term exposure to chemicals and to
characterize exposure-response relationships. An
assessment may cover a single chemical, a group of
structurally or lexicologically related chemicals, or
a complex mixture. Exceptions are chemicals
currently used exclusively as pesticides, ionizing
and non-ionizing radiation, and criteria air
pollutants listed under section 108 of the Clean Air
Act (carbon monoxide, lead, nitrogen oxides, ozone,
particulate matter, and sulfur oxides; EPA's
Integrated Science Assessments evaluate the effects
from these pollutants in ambient air).
Periodically, the IRIS Program asks other EPA
programs and regions, other federal agencies, state
government agencies, and the general public to
nominate chemicals and mixtures for future
assessment or reassessment. These agents may be
found in air, water, soil, or sediment. Selection is
based on program and regional office priorities and
on availability of adequate information to evaluate
the potential for adverse effects. IRIS may assess
other agents as an urgent public health need arises.
IRIS also reassesses agents as significant new
studies are published.
2. Process for developing and peer-reviewing
IRIS assessments
The process for developing IRIS assessments
(revised in May 2009) involves critical analysis of
the pertinent studies, opportunities for public
input, and multiple levels of scientific review. EPA
revises draft assessments after each review, and
external drafts and comments become part of the
public record (U.S. EPA. 2 009).
Step 1. Development of a draft Toxicological
Review (usually about 11-1/2 months
duration). The draft assessment considers all
pertinent publicly available studies and applies
consistent criteria to evaluate the studies,
identify health effects, weigh the evidence of
causation for each effect, identify mechanistic
events and pathways, and derive toxicity
values.
Step 2. Internal review by scientists in EPA
programs and regions (2 months). The draft
assessment is revised to address comments
from within EPA.
Step 3. Interagency science consultation with
other federal agencies and the Executive
Offices of the President (1-1/2 months). The
draft assessment is revised to address the
interagency comments. The science
consultation draft, interagency comments, and
EPA's response to major comments become
part of the public record.
Step 4. External peer review, after public
review and comment (3-1/2 months or more,
depending on the review process). EPA
releases the draft assessment for public review
and comment, followed by external peer
review. The peer review meeting is open to the
public and includes time for oral public
comments. The peer reviewers also receive the
written public comments. The peer reviewers
assess whether the evidence has been
assembled and evaluated according to
guidelines and whether the conclusions are
justified by the evidence. The peer review
draft, peer review report, and written public
comments become part of the public record.
Step 5. Revision of draft Toxicological Review
and development of draft IRIS summary
(2 months). The draft assessment is revised to
reflect the peer review comments, public
comments, and newly published studies that
are critical to the conclusions of the
assessment The disposition of peer review
This document is a draft for review purposes only and does not constitute Agency policy.
xi DRAFT—DO NOT CITE OR QUOTE
-------
comments and public comments becomes part
of the public record.
Step 6. Final EPA review and interagency
science discussion with other federal
agencies and the Executive Offices of the
President [1-1/2 months). The draft
assessment and summary are revised to
address EPA and interagency comments. The
science discussion draft, written interagency
comments, and EPA's response to major
comments become part of the public record.
Step 7. Completion and posting (1 month). The
Toxicological Review and IRIS summary are
posted on the IRIS website (http://
www.epa.gov/iris/).
The remainder of this Preamble addresses step
1, the development of a draft Toxicological Review.
IRIS assessments follow standard practices of
evidence evaluation and peer review, many of
which are discussed in EPA guidelines [U.S. EPA.
2005a. b, 2000b. 1998. 1996. 1991. 1986a. b) and
other methods [U.S. EPA. 2011b. 2006a. b, 2002.
2000a. 1994). Transparent application of scientific
judgment is of paramount importance. To provide a
harmonized approach across IRIS assessments, this
Preamble summarizes concepts from these
guidelines and emphasizes principles of general
applicability.
3. Identifying and selecting pertinent studies
3.1. Identifying studies
Before beginning an assessment, EPA conducts
a comprehensive search of the primary scientific
literature. The literature search follows standard
practices and includes the PubMed and ToxNet
databases of the National Library of Medicine and
other databases listed in EPA's HERO system
(Health and Environmental Research Online,
http://hero.epa.gov/). Each assessment specifies
the search strategies, keywords, and cut-off dates
of its literature searches. EPA posts the results of
the literature search on the IRIS website and
requests information from the public on additional
studies and ongoing research.
EPA also considers studies received through
the IRIS Submission Desk and studies (typically
unpublished) submitted under the Toxic
Substances Control Act Material submitted as
Confidential Business Information is considered
only if it includes health and safety data that can be
publicly released. If a study that may be critical to
the conclusions of the assessment has not been
peer-reviewed, EPA will have it peer-reviewed.
Toxicological Review of Ammonia
EPA also examines the toxicokinetics of the
agent to identify other chemicals (for example,
major metabolites of the agent) to include in the
assessment if adequate information is available, in
order to more fully explain the toxicity of the agent
and to suggest dose metrics for subsequent
modeling.
In assessments of chemical mixtures, mixture
studies are preferred for their ability to reflect
interactions among components. The literature
search seeks, in decreasing order of preference
(U.S. EPA.2000b. 1986b):
Studies of the mixture being assessed.
Studies of a sufficiently similar mixture. In
evaluating similarity, the assessment considers
the alteration of mixtures in the environment
through partitioning and transformation.
Studies of individual chemical components of
the mixture, if there are not adequate studies
of sufficiently similar mixtures.
3.2. Selecting pertinent epidemiologic studies
Study design is the key consideration for
selecting pertinent epidemiologic studies from the
results of the literature search.
Cohort studies and case-control studies
provide the strongest epidemiologic evidence,
as they collect information about individual
exposures and effects.
Ecologic studies (geographic correlation
studies) relate exposures and effects by
geographic area. They can provide strong
evidence if there are large exposure contrasts
between geographic areas, relatively little
exposure variation within study areas, and
population migration is limited.
Case reports of high or accidental exposure
lack definition of the population at risk and the
expected number of cases. They can provide
information about a rare effect or about the
relevance of analogous results in animals.
The assessment briefly reviews ecologic
studies and case reports but reports details only if
they suggest effects not identified by other
epidemiologic studies.
3.3. Selecting pertinent experimental studies
Exposure route is a key design consideration
for selecting pertinent experimental studies from
the results of the literature search.
Studies of oral, inhalation, or dermal exposure
involve passage through an absorption barrier
This document is a draft for review purposes only and does not constitute Agency policy.
xii DRAFT—DO NOT CITE OR QUOTE
-------
and are considered most pertinent to human
environmental exposure.
Injection or implantation studies are often
considered less pertinent but may provide
valuable toxicokinetic or mechanistic
information. They also may be useful for
identifying effects in animals if deposition or
absorption is problematic (for example, for
particles and fibers).
Exposure duration is also a key design
consideration for selecting pertinent experimental
studies.
Studies of effects from chronic exposure are
most pertinent to lifetime human exposure.
Studies of effects from less-than-chronic
exposure are pertinent but less preferred than
studies of chronic exposure.
Short-duration studies involving animals or
humans may provide toxicokinetic or mechanistic
information. Research involving human subjects is
considered only if conducted according to ethical
principles.
For developmental toxicity and reproductive
toxicity, irreversible effects may result from a brief
exposure during a critical period of development.
Accordingly, specialized study designs are used for
these effects fU.S. EPA.2006b. 1998.1996.19911
4. Evaluating the quality of individual studies
4.1. Evaluating the quality of epidemiologic
studies
The assessment evaluates design and
methodologic aspects that can increase or decrease
the weight given to each epidemiologic study in the
overall evaluation [U.S. EPA, 2005a, 1998, 1996,
1994.19911:
Documentation of study design, methods,
population characteristics, and results.
Definition and selection of the study and
comparison populations.
Ascertainment of exposure and the potential
for misclassification.
Ascertainment of disease or effect and the
potential for misclassification.
Duration of exposure and follow-up and
adequacy for assessing the occurrence of
effects, including latent effects.
Characterization of exposure during critical
periods.
Toxicological Review of Ammonia
Sample size and statistical power to detect
anticipated effects.
Participation rates and the resulting potential
for selection bias.
Potential confounding and other sources of
bias are identified and addressed in the study
design or in the analysis of results. The basis
for consideration of confounding is a
reasonable expectation that the confounder is
prevalent in the population and is related to
both exposure and outcome.
For developmental toxicity, reproductive
toxicity, neurotoxicity, and cancer there is further
guidance on the nuances of evaluating
epidemiologic studies of these effects [U.S. EPA.
2005a. 1998.1996.1991].
4.2. Evaluating the quality of experimental
studies
The assessment evaluates design and
methodologic aspects that can increase or decrease
the weight given to each experimental study in the
overall evaluation [U.S. EPA. 2005a. 1998. 1996.
1994.1991]:
Documentation of study design, animals or
study population, methods, basic data, and
results.
Relevance to humans of the animal model and
experimental methods.
Characterization of the nature and extent of
impurities and contaminants of the
administered chemical or mixture.
Characterization of dose and dosing regimen
(including age at exposure) and their adequacy
to elicit adverse effects, including latent effects.
Sample sizes and statistical power to detect
dose-related differences or trends.
Ascertainment of survival, vital signs, disease
or effects, and cause of death.
Control of other variables that could influence
the occurrence of effects.
The assessment uses statistical tests to
evaluate whether the observations may be due to
chance. The standard for determining statistical
significance of a response is a trend test or
comparison of outcomes in the exposed groups
against those of concurrent controls. In some
situations, examination of historical control data
from the same laboratory within a few years of the
study may improve the analysis. For an uncommon
effect that is not statistically significant compared
with concurrent controls, historical controls may
This document is a draft for review purposes only and does not constitute Agency policy.
xiii DRAFT—DO NOT CITE OR QUOTE
-------
show that the effect is unlikely to be due to chance.
For a response that appears significant against a
concurrent control response that is unusual,
historical controls may offer a different
interpretation [U.S. EPA. 2005a).
For developmental toxicity, reproductive
toxicity, neurotoxicity, and cancer there is further
guidance on the nuances of evaluating
experimental studies of these effects [U.S. EPA.
2005a. 1998. 1996. 1991]. In multi-generation
studies, agents that produce developmental effects
at doses that are not toxic to the maternal animal
are of special concern. Effects that occur at doses
associated with mild maternal toxicity are not
assumed to result only from maternal toxicity.
Moreover, maternal effects may be reversible,
while effects on the offspring may be permanent
[U.S. EPA. 1998.1991).
4.3. Reporting study results
The assessment uses evidence tables to
summarize details of the design and key results of
pertinent studies. There may be separate tables for
each site of toxicity or type of study.
If a large number of studies observe the same
effect, the assessment considers the study
characteristics in this section to identify the
strongest studies or types of study. The tables
report details from these studies, and the
assessment explains the reasons for not reporting
details of other studies or groups of studies that do
not add new information. Supplemental material
provides references to all studies considered,
including those not summarized in the tables.
The assessment discusses strengths and
limitations that affect the interpretation of each
study. If the interpretation of a study in the
assessment differs from that of the study authors,
the assessment discusses the basis for the
difference.
As a check on the selection and evaluation of
pertinent studies, EPA asks peer reviewers to
identify studies that were not adequately
considered.
5. Weighing the overall evidence of each effect
5.1. Weighing epidemiologic evidence
For each effect, the assessment evaluates the
evidence from the epidemiologic studies as a whole
to determine the extent to which any observed
associations may be causal. Positive, negative, and
null results are given weight according to study
quality. This evaluation considers aspects of an
association that suggest causality, discussed by Hill
[1965) and elaborated by Rothman and Greenland
Toxicological Review of Ammonia
f!998) fU.S. EPA.2005a: CDC. 2004: U.S. EPA. 2002.
1994).
Strength of association: The finding of a large
relative risk with narrow confidence intervals
strongly suggests that an association is not due
to chance, bias, or other factors. Modest
relative risks, however, may reflect a small
range of exposures, an agent of low potency, an
increase in an effect that is common, exposure
misclassification, or other sources of bias.
Consistency of association: An inference of
causality is strengthened if elevated risks are
observed in independent studies of different
populations and exposure scenarios.
Reproducibility of findings constitutes one of
the strongest arguments for causality.
Discordant results sometimes reflect
differences in study design, exposure, or
confounding factors.
Specificity of association: As originally intended,
this refers to one cause associated with one
effect. Current understanding that many agents
cause multiple effects and many effects have
multiple causes make this a less informative
aspect of causality, unless the effect is rare or
unlikely to have multiple causes.
Temporal relationship: A causal interpretation
requires that exposure precede development of
the effect.
Biologic gradient (exposure-response relation-
ship): Exposure-response relationships
strongly suggest causality. A monotonic
increase is not the only pattern consistent with
causality. The presence of an exposure-
response gradient also weighs against bias and
confounding as the source of an association.
Biologic plausibility: An inference of causality is
strengthened by data demonstrating plausible
biologic mechanisms, if available.
Coherence: An inference of causality is
strengthened by supportive results from
animal experiments, toxicokinetic studies, and
short-term tests. Coherence may also be found
in other lines of evidence, such as changing
disease patterns in the population.
"Natural experiments": A change in exposure that
brings about a change in disease frequency
provides strong evidence of causality, for
example, an intervention to reduce exposure in
the workplace or environment that is followed
by a reduction of an adverse effect.
This document is a draft for review purposes only and does not constitute Agency policy.
xiv DRAFT—DO NOT CITE OR QUOTE
-------
Analogy: Information on structural analogues or
on chemicals that induce similar mechanistic
events can provide insight into causality.
These considerations are consistent with
guidelines for systematic reviews that evaluate the
quality and weight of evidence. Confidence is
increased if the magnitude of effect is large, if there
is evidence of an exposure-response relationship,
or if an association was observed and the plausible
biases would tend to decrease the magnitude of the
reported effect. Confidence is decreased for study
limitations, inconsistency of results, indirectness of
evidence, imprecision, or reporting bias [Guyatt et
al.. 2008a: Guyatt et al.. 2008b].
To make clear how much the epidemiologic
evidence contributes to the overall weight of the
evidence, the assessment may choose a descriptor
such as sufficient evidence, suggestive evidence,
inadequate evidence, or evidence suggestive of no
causal relationship to characterize the
epidemiologic evidence of each effect [CDC. 2004].
5.2. Weighing experimental animal evidence
For each effect, the assessment evaluates the
evidence from the animal experiments as a whole
to determine the extent to which they indicate a
potential for effects in humans. Consistent results
across various species and strains increase
confidence that similar results would occur in
humans. Several concepts discussed by Hill (1965]
are pertinent to the weight of experimental results:
consistency of response, dose-response
relationships, strength of response, biologic
plausibility, and coherence [U.S. EPA. 2005a. 2002.
1994].
In weighing evidence from multiple
experiments, [U.S. EPA. 2005a] distinguishes
Conflicting evidence (that is, mixed positive and
negative results in the same sex and strain
using a similar study protocol] from
Differing results (that is, positive results and
negative results are in different sexes or strains
or use different study protocols].
Negative or null results do not invalidate positive
results in a different experimental system. EPA
regards all as valid observations and looks to
methodological differences or, if available,
mechanistic information to reconcile differing
results.
It is well established that there are critical
periods for some developmental and reproductive
effects. Accordingly, the assessment determines
whether critical periods have been adequately
investigated (U.S. EPA. 2006b. 2005a. b, 1998.
1996. 1991]. Similarly, the assessment determines
Toxicological Review of Ammonia
whether the database is adequate to evaluate other
critical sites and effects.
In evaluating evidence of genotoxicity:
Demonstration of gene mutations,
chromosome aberrations, or aneuploidy in
humans or experimental mammals (in vivo)
provides the strongest evidence.
This is followed by positive results in lower
organisms or in cultured cells (in vitro) or for
other genetic events.
Negative results carry less weight, partly
because they cannot exclude the possibility of
effects in other tissues (IARC. 2006].
For germ-cell mutagenicity, EPA has defined
categories of evidence, ranging from positive
results of human germ-cell mutagenicity to
negative results for all effects of concern (U.S. EPA.
1986a].
5.3. Characterizing modes of action
For each effect, the assessment discusses the
available information on its modes of action and
associated key events (key events being empirically
observable, necessary precursor steps or biologic
markers of such steps; mode of action being a series
of key events involving interaction with cells,
operational and anatomic changes, and resulting in
disease]. Pertinent information may also come
from studies of metabolites or of compounds that
are structurally similar or that act through similar
mechanisms. Information on mode of action is not
required for a conclusion that an effect is causally
related to an agent (U.S. EPA. 2005a].
The assessment addresses several questions
about each hypothesized mode of action (U.S. EPA.
2005a].
(1] Is the hypothesized mode of action
sufficiently supported in test animals?
Strong support for a key event being necessary
to a mode of action can come from
experimental challenge to the hypothesized
mode of action, in which studies that suppress
a key event observe suppression of the effect.
Support for a mode of action is meaningfully
strengthened by consistent results in different
experimental models, much more so than by
replicate experiments in the same model. The
assessment may consider various aspects of
causality in addressing this question.
(2] Is the hypothesized mode of action relevant
to humans? The assessment reviews the key
events to identify critical similarities and
differences between the test animals and
humans. Site concordance is not assumed
This document is a draft for review purposes only and does not constitute Agency policy.
xv DRAFT—DO NOT CITE OR QUOTE
-------
between animals and humans, though it may
hold for certain effects or modes of action.
Information suggesting quantitative
differences in doses where effects would occur
in animals or humans is considered in the
dose-response analysis but is not used to
determine relevance. Similarly, anticipated
levels of human exposure are not used to
determine relevance.
(3) Which populations or lifestages can be
particularly susceptible to the hypothesized
mode of action? The assessment reviews the
key events to identify populations and
lifestages that might be susceptible to their
occurrence. Quantitative differences may result
in separate toxicity values for susceptible
populations or lifestages.
The assessment discusses the likelihood that
an agent operates through multiple modes of
action. An uneven level of support for different
modes of action can reflect disproportionate
resources spent investigating them [U.S. EPA.
2005aj. It should be noted that in clinical reviews,
the credibility of a series of studies is reduced if
evidence is limited to studies funded by one
interested sector [Guyatt et al.. 2008aj.
For cancer, the assessment evaluates evidence
of a mutagenic mode of action to guide
extrapolation to lower doses and consideration of
susceptible lifestages. Key data include the ability
of the agent or a metabolite to react with or bind to
DNA, positive results in multiple test systems, or
similar properties and structure-activity
relationships to mutagenic carcinogens [U.S. EPA.
2005a).
5.4. Characterizing the overall weight of the
evidence
After weighing the epidemiologic and
experimental studies pertinent to each effect, the
assessment may select a standard descriptor to
characterize the overall weight of the evidence. For
example, the following standard descriptors
combine epidemiologic, experimental, and
mechanistic evidence of carcinogenicity [U.S. EPA.
2005al
Carcinogenic to humans: There is convincing
epidemiologic evidence of a causal association
(that is, there is reasonable confidence that the
association cannot be fully explained by
chance, bias, or confounding); or there is
strong human evidence of cancer or its
precursors, extensive animal evidence,
identification of key precursor events in
Toxicological Review of Ammonia
animals, and strong evidence that they are
anticipated to occur in humans.
Likely to be carcinogenic to humans: The
evidence demonstrates a potential hazard to
humans but does not meet the criteria for
carcinogenic. There may be a plausible
association in humans, multiple positive
results in animals, or a combination of human,
animal, or other experimental evidence.
Suggestive evidence of carcinogenic potential:
The evidence raises concern for effects in
humans but is not sufficient for a stronger
conclusion. This descriptor covers a range of
evidence, from a positive result in the only
available study to a single positive result in an
extensive database that includes negative
results in other species.
Inadequate information to assess carcinogenic
potential: No other descriptors apply.
Conflicting evidence can be classified as
inadequate information if all positive results
are opposed by negative studies of equal
quality in the same sex and strain. Differing
results, however, can be classified as suggestive
evidence or as likely to be carcinogenic.
Not likely to be carcinogenic to humans: There is
robust evidence for concluding that there is no
basis for concern. There may be no effects in
both sexes of at least two appropriate animal
species; positive animal results and strong,
consistent evidence that each mode of action in
animals does not operate in humans; or
convincing evidence that effects are not likely
by a particular exposure route or below a
defined dose.
Multiple descriptors may be used if there is
evidence that carcinogenic effects differ by dose
range or exposure route [U.S. EPA. 2005aj.
EPA is investigating and may on a trial basis
propose standard descriptors to characterize the
overall weight of the evidence for effects other than
cancer.
6. Selecting studies for derivation of toxicity
values
For each effect where there is credible
evidence of an association with the agent, the
assessment derives toxicity values if there are
suitable epidemiologic or experimental data. The
decision to derive toxicity values may be linked to
the weight-of-evidence descriptor. For example,
EPA typically derives toxicity values for agents
classified as carcinogenic to humans or as likely to
be carcinogenic [U.S. EPA. 2005aj.
This document is a draft for review purposes only and does not constitute Agency policy.
xvi DRAFT—DO NOT CITE OR QUOTE
-------
Dose-response analysis requires quantitative
measures of dose and response. Then, other factors
being equal [U.S. EPA. 2005a. 1994]:
Epidemiologic studies are preferred over
animal studies, if quantitative measures of
exposure are available and effects can be
attributed to the agent.
Among experimental animal models, those that
respond most like humans are preferred, if the
comparability of response can be determined.
Studies by a route of human environmental
exposure are preferred, although a validated
toxicokinetic model can be used to extrapolate
across exposure routes.
Studies of longer exposure duration and
follow-up are preferred, to minimize
uncertainty about whether effects are
representative of lifetime exposure.
Studies with multiple exposure levels are
preferred for their ability to provide
information about the shape of the exposure-
response curve.
Studies with adequate power to detect effects
at lower exposure levels are preferred, to
minimize the extent of extrapolation to levels
found in the environment.
Studies with non-monotonic exposure-
response relationships are not necessarily excluded
from the analysis. A diminished effect at higher
exposure levels may be satisfactorily explained by
factors such as competing toxicity, saturation of
absorption or metabolism, exposure
misclassification, or selection bias.
If a large number of studies are suitable for
dose-response analysis, the assessment considers
the study characteristics in this section to focus on
the most informative data. The assessment explains
the reasons for not analyzing other groups of
studies. As a check on the selection of studies for
dose-response analysis, EPA asks peer reviewers to
identify studies that were not adequately
considered.
7. Deriving toxicity values
7.1. General framework for dose-response
analysis
EPA uses a two-step approach that
distinguishes analysis of the observed dose-
response data from inferences about lower doses
[U.S. EPA. 2005a).
Within the observed range, the preferred
approach is to use modeling to incorporate a wide
range of data into the analysis. The modeling yields
Toxicological Review of Ammonia
a point of departure (an exposure level near the
lower end of the observed range, without
significant extrapolation to lower doses) (sections
7.2-7.3).
Extrapolation to lower doses considers what is
known about the modes of action for each effect
(sections 7.4-7.5). When response estimates at
lower doses are not required, an alternative is to
derive reference values, which are calculated by
applying factors that account for sources of
uncertainty and variability to the point of
departure (section 7.6).
For a group of agents that induce an effect
through a common mode of action, the dose-
response analysis may derive a relative potency
factor for each agent. A full dose-response analysis
is conducted for one well-studied index chemical in
the group, then the potencies of other members are
expressed in relative terms based on relative toxic
effects, relative absorption or metabolic rates,
quantitative structure-activity relationships, or
receptor binding characteristics (U.S. EPA. 2005a.
200Qb).
Increasingly, EPA is basing toxicity values on
combined analyses of multiple data sets or multiple
responses. EPA also considers multiple dose-
response approaches when they can be supported
by robust data.
7.2. Modeling dose
The preferred approach for analysis of dose is
toxicokinetic modeling because of its ability to
incorporate a wide range of data. The preferred
dose metric would refer to the active agent at the
site of its biologic effect or to a close, reliable
surrogate measure. The active agent may be the
administered chemical or a metabolite. Confidence
in the use of a toxicokinetic model depends on the
robustness of its validation process and on the
results of sensitivity analyses (U.S. EPA, 2006a,
2005a,1994).
Because toxicokinetic modeling can require
many parameters and more data than are typically
available, EPA has developed standard approaches
that can be applied to typical data sets. These
standard approaches also facilitate comparison
across exposure patterns and species.
Intermittent study exposures are standardized
to a daily average over the duration of
exposure. For chronic effects, daily exposures
are averaged over the lifespan. Exposures
during a critical period, however, are not
averaged over a longer duration (U.S. EPA.
2005a. 1998.1996.19911
This document is a draft for review purposes only and does not constitute Agency policy.
xvii DRAFT—DO NOT CITE OR QUOTE
-------
Doses are standardized to equivalent human
terms to facilitate comparison of results from
different species.
Oral doses are scaled allometrically using
mg/kg3/4-d as the equivalent dose metric
across species. Allometric scaling pertains
to equivalence across species, not across
lifestages, and is not used to scale doses
from adult humans or mature animals to
infants or children [U.S. EPA. 2011b.
2005a].
Inhalation exposures are scaled using
dosimetry models that apply species-
specific physiologic and anatomic factors
and consider whether the effect occurs at
the site of first contact or after systemic
circulation [U.S. EPA. 1994].
It can be informative to convert doses across
exposure routes. If this is done, the assessment
describes the underlying data, algorithms, and
assumptions (U.S. EPA. 2005a].
In the absence of study-specific data on, for
example, intake rates or body weight, EPA has
developed recommended values for use in dose-
response analysis [U.S. EPA. 1988].
7.3. Modeling response in the range of
observation
Toxicodynamic ("biologically based"] modeling
can incorporate data on biologic processes leading
to an effect. Such models require sufficient data to
ascertain a mode of action and to quantitatively
support model parameters associated with its key
events. Because different models may provide
equivalent fits to the observed data but diverge
substantially at lower doses, critical biologic
parameters should be measured from laboratory
studies, not by model fitting. Confidence in the use
of a toxicodynamic model depends on the
robustness of its validation process and on the
results of sensitivity analyses. Peer review of the
scientific basis and performance of a model is
essential fU.S. EPA. 2005a].
Because toxicodynamic modeling can require
many parameters and more knowledge and data
than are typically available, EPA has developed a
standard set of empirical ("curve-fitting"] models
(http://www.epa.gov/ncea/bmds/] that can be
applied to typical data sets, including those that are
nonlinear. EPA has also developed guidance on
modeling dose-response data, assessing model fit,
selecting suitable models, and reporting modeling
results (U.S. EPA. 2000a]. Additional judgment or
alternative analyses are used when the procedure
fails to yield reliable results, for example, if the fit is
poor, modeling may be restricted to the lower
Toxicological Review of Ammonia
doses, especially if there is competing toxicity at
higher doses (U.S. EPA. 2005a].
Modeling is used to derive a point of departure
(U.S. EPA. 2005a. 2000a]. (See section 7.6 for
alternatives if a point of departure cannot be
derived by modeling.]
For dichotomous responses, the point of
departure is often the 95% lower bound on the
dose associated with a 10% response, but a
lower response that falls within the observed
range may be used instead. For example,
reproductive or developmental studies often
have power to detect a 5% response;
epidemiologic studies, 1% or lower.
For continuous responses, the point of
departure is ideally the dose where the effect
becomes biologically significant. In the absence
of such definition, both statistical and biologic
factors are considered.
7.4. Extrapolating to lower doses
The purpose of extrapolating to lower doses is
to estimate responses at exposures below the
observed data. Low-dose extrapolation is typically
used for known and likely carcinogens. Low-dose
extrapolation considers what is known about
modes of action (U.S. EPA, 2005a].
(1] If a biologically based model has been
developed and validated for the agent,
extrapolation may use the fitted model below
the observed range if significant model
uncertainty can be ruled out with reasonable
confidence.
(2] Linear extrapolation is used if the dose-
response curve is expected to have a linear
component below the point of departure. This
includes:
Agents or their metabolites that are DNA-
reactive and have direct mutagenic
activity.
Agents or their metabolites for which
human exposures or body burdens are
near doses associated with key events
leading to an effect.
Linear extrapolation is also used if the
evidence is insufficient to establish a mode of
action.
The result of linear extrapolation is described
by an oral slope factor or an inhalation unit risk,
which is the slope of the dose-response curve
at lower doses or concentrations, respectively.
(3] Nonlinear extrapolation is used if there are
sufficient data to ascertain the mode of action
This document is a draft for review purposes only and does not constitute Agency policy.
xviii DRAFT—DO NOT CITE OR QUOTE
-------
and to conclude that it is not linear at lower
doses, and the agent does not demonstrate
mutagenic or other activity consistent with
linearity at lower doses. If nonlinear
extrapolation is appropriate but no model is
developed, an alternative is to calculate
reference values.
If linear extrapolation is used, the assessment
develops a candidate slope factor or unit risk for
each suitable data set. These results are arrayed,
using common dose metrics, to show the
distribution of relative potency across various
effects and experimental systems. The assessment
then derives an overall slope factor and an overall
unit risk for the agent, considering the various
dose-response analyses, the study preferences
discussed in section 6, and the possibility of basing
a more robust result on multiple data sets.
7.5. Considering susceptible populations and
lifestages
The assessment analyzes the available
information on populations and lifestages that may
be particularly susceptible to each effect. A tiered
approach is used (U.S. EPA. 2005a).
(1) If an epidemiologic or experimental study
reports quantitative results for a susceptible
population or lifestage, these data are analyzed
to derive separate toxicity values for
susceptible individuals.
(2) If data on risk-related parameters allow
comparison of the general population and
susceptible individuals, these data are used to
adjust the general-population toxicity values
for application to susceptible individuals.
(3) In the absence of chemical-specific data, EPA
has developed age-dependent adjustment
factors for early-life exposure to suspected
carcinogens that have a mutagenic mode of
action. There is evidence of early-life
susceptibility to various carcinogenic agents,
but most epidemiologic studies and cancer
bioassays do not include early-life exposure. To
address the potential for early-life
susceptibility, EPA recommends [U.S. EPA.
2005b1:
10-fold adjustment for exposures before
age 2 years.
3-fold adjustment for exposures between
ages 2 and 16 years.
7.6. Reference values and uncertainty factors
An oral reference dose or an inhalation
reference concentration is an estimate of an
Toxicological Review of Ammonia
exposure (including in susceptible subgroups) that
is likely to be without an appreciable risk of
adverse health effects over a lifetime [U.S. EPA.
2002]. Reference values are typically calculated for
effects other than cancer and for suspected
carcinogens if a well characterized mode of action
indicates that a necessary key event does not occur
below a specific dose. Reference values provide no
information about risks at higher exposure levels.
The assessment characterizes effects that form
the basis for reference values as adverse,
considered to be adverse, or a precursor to an
adverse effect. For developmental toxicity,
reproductive toxicity, and neurotoxicity there is
guidance on adverse effects and their biologic
markers fU.S. EPA. 1998.1996.19911
To account for uncertainty and variability in
the derivation of a lifetime human exposure where
effects are not anticipated to occur, reference
values are calculated by applying a series of
uncertainty factors to the point of departure. If a
point of departure cannot be derived by modeling,
a no-observed-adverse-effect level or a lowest-
observed-adverse-effect level is used instead. The
assessment discusses scientific considerations
involving several areas of variability or uncertainty.
Human variation. A factor of 10 is applied to
account for variation in susceptibility across
the human population and the possibility that
the available data may not be representative of
individuals who are most susceptible to the
effect. This factor is reduced only if the point of
departure is derived specifically for susceptible
individuals (not for a general population that
includes both susceptible and non-susceptible
individuals) (U.S. EPA. 2002.1998.1996.1994.
1991).
Animal-to-human extrapolation. A factor of 10 is
applied if animal results are used to make
inferences about humans. This factor is often
regarded as comprising toxicokinetics and
toxicodynamics in equal parts. Accordingly, if
the point of departure is based on toxicokinetic
modeling, dosimetry modeling, or allometric
scaling across species, a factor of 101/2
(rounded to 3) is applied to account for the
remaining uncertainty involving toxicodynamic
differences. An animal-to-human factor is not
applied if a biologically based model adjusts
fully for toxicokinetic and toxicodynamic
differences across species (U.S. EPA. 2011b.
2002.1998.1996.1994.1991).
Adverse-effect level to no-observed-adverse-
effect level. If a point of departure is based on
a lowest-observed-adverse-effect level, the
assessment must infer a dose where such
This document is a draft for review purposes only and does not constitute Agency policy.
xix DRAFT—DO NOT CITE OR QUOTE
-------
effects are not expected. This can be a matter of
great uncertainty, especially if there is no
evidence available at lower doses. A factor of
10 is applied to account for the uncertainty in
making this inference. A factor other than 10
may be used, depending on the magnitude and
nature of the response and the shape of the
dose-response curve [U.S. EPA. 2002. 1998.
1996.1994.1991].
Subchronic-to-chronic exposure. If a point of
departure is based on subchronic studies, the
assessment considers whether lifetime
exposure could have effects at lower levels of
exposure. A factor of 10 is applied to account
for the uncertainty in using subchronic studies
to make inferences about lifetime exposure.
This factor may also be applied for
developmental or reproductive effects if
exposure covered less than the full critical
period. A factor other than 10 may be used,
depending on the duration of the studies and
the nature of the response [U.S. EPA. 2002.
1998.1994].
Incomplete database. If an incomplete database
raises concern that further studies might
identify a more sensitive effect, organ system,
or lifestage, the assessment may apply a
database uncertainty factor [U.S. EPA. 2002.
1998.1996.1994.1991]. The size of the factor
depends on the nature of the database
deficiency. For example, EPA typically follows
the suggestion that a factor of 10 be applied if
both a prenatal toxicity study and a two-
generation reproduction study are missing and
a factor of 101/2 if either is missing [U.S. EPA,
20021.
In this way, the assessment derives candidate
reference values for each suitable data set and
effect that is credibly associated with the agent.
These results are arrayed, using common dose
metrics, to show where effects occur across a range
of exposures [U.S. EPA. 1994]. The assessment then
selects an overall reference dose and an overall
reference concentration for the agent to represent
lifetime human exposure levels where effects are
not anticipated to occur.
The assessment may also report reference
values for each effect. This would facilitate
subsequent cumulative risk assessments that
consider the combined effect of multiple agents
acting at a common site or through common
mechanisms [U.S. EPA. 2002].
Toxicological Review of Ammonia
7.7. Confidence and uncertainty in the
reference values
The assessment selects a standard descriptor
to characterize the level of confidence in each
reference value, based on the likelihood that the
value would change with further testing.
Confidence in reference values is based on quality
of the studies used and completeness of the
database, with more weight given to the latter. The
level of confidence is increased for reference values
based on human data supported by animal data
fU.S. EPA. 1994].
High confidence: The reference value is not likely
to change with further testing, except for
mechanistic studies that might affect the
interpretation of prior test results.
Medium confidence: This is a matter of judgment,
between high and low confidence.
Low confidence: The reference value is especially
vulnerable to change with further testing.
These criteria are consistent with guidelines
for systematic reviews that evaluate the quality of
evidence. These also focus on whether further
research would be likely to change confidence in
the estimate of effect [Guyatt et al.. 2008a].
All assessments discuss the significant
uncertainties encountered in the analysis. EPA
provides guidance on characterization of
uncertainty [U.S. EPA. 2005a]. For example, the
discussion distinguishes model uncertainty (lack of
knowledge about the most appropriate
experimental or analytic model] and parameter
uncertainty (lack of knowledge about the
parameters of a model]. Assessments also discuss
human variation (interpersonal differences in
biologic susceptibility or in exposures that modify
the effects of the agent].
References
CDC (Centers for Disease Control and Prevention].
(2004]. The health consequences of smoking: A
report of the Surgeon General. Washington, DC:
U.S. Department of Health and Human Services.
http://www.surgeongeneral.gov/library/smok
ingconsequences/.
Guyatt. GH: Oxman. AD: Vist. GE: Kunz. R: Falck-
Ytter. Y: Alonso-Coello. P: Schiinemann. HI.
(2008a]. GRADE: An emerging consensus on
rating quality of evidence and strength of
recommendations. BMJ 336: 924-926.
http://dx.doi.org/10.1136/bmj.39489.470347.
AD.
Guyatt. GH: Oxman. AD: Kunz. R: Vist. GE: Falck-
Ytter. Y: Schiinemann. HI. (2008b]. GRADE:
This document is a draft for review purposes only and does not constitute Agency policy.
xx DRAFT—DO NOT CITE OR QUOTE
-------
What is "quality of evidence" and why is it
important to clinicians? BMJ 336: 995-998.
http://dx.doi.org/10.1136/bmj.39490.551019.
BE.
Hill. AB. (1965). The environment and disease:
Association or causation? Proc R Soc Med 58:
295-300.
IARC (International Agency for Research on
Cancer). (2006). Preamble to the IARC
monographs. Lyon, France.
http://monographs.iarc.fr/ENG/Preamble/.
Rothman. Kl: Greenland. S. (1998). Modern
epidemiology (2nd ed.). Philadelphia, PA:
Lippincott, Williams, & Wilkins.
U.S. EPA (U.S. Environmental Protection Agency).
(1986a). Guidelines for mutagenicity risk
assessment [EPA Report]. (EPA/630/R-
98/003). Washington, DC.
http://www.epa.gov/iris/backgrd.html.
U.S. EPA (U.S. Environmental Protection Agency).
(1986b). Guidelines for the health risk
assessment of chemical mixtures [EPA Report]
(pp. 34014-34025). (EPA/630/R-98/002).
Washington, DC.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cf
m?deid=22567.
U.S. EPA (U.S. Environmental Protection Agency).
(1988). Recommendations for and
documentation of biological values for use in
risk assessment. (EPA/600/6-87/008).
Cincinnati, OH: U.S. Environmental Protection
Agency, Environmental Criteria and
Assessment Office.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cf
m?deid=34855.
U.S. EPA (U.S. Environmental Protection Agency).
(1991). Guidelines for developmental toxicity
risk assessment [EPA Report]. (EPA/600/FR-
91/001). Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment Forum.
http://www.epa.gov/iris/backgrd.html.
U.S. EPA (U.S. Environmental Protection Agency).
(1994). Methods for derivation of inhalation
reference concentrations and application of
inhalation dosimetry. (EPA/600/8-90/066F).
Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of
Research and Development, Office of Health
and Environmental Assessment,
Environmental Criteria and Assessment Office.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cf
m?deid=71993.
U.S. EPA (U.S. Environmental Protection Agency).
(1996). Guidelines for reproductive toxicity
risk assessment [EPA Report]. (EPA/630/R-
96/009). Washington, DC: U.S. Environmental
Toxicological Review of Ammonia
Protection Agency, Risk Assessment Forum.
http://www.epa.goV/raf/publications/pdfs/R
EPR051.PDF.
U.S. EPA (U.S. Environmental Protection Agency).
(1998). Guidelines for neurotoxicity risk
assessment. (EPA/630/R-95/001F).
Washington, DC: U.S. Environmental Protection
Agency, Risk Assessment Forum.
http://www.epa.gOV/raf/publications/pdfs/N
EUROTOX.PDF.
U.S. EPA (U.S. Environmental Protection Agency).
(2000a). Benchmark dose technical guidance
document [external review draft].
(EPA/630/R-00/001). Washington, DC: U.S.
Environmental Protection Agency, Risk
Assessment Forum.
http://www.epa.gov/raf/publications/benchm
ark-dose-doc-draft.htm.
U.S. EPA (U.S. Environmental Protection Agency).
(2000b). Supplementary guidance for
conducting health risk assessment of chemical
mixtures [EPA Report]. (EPA/630/R-00/002).
http://cfpub.epa.gov/ncea/cfm/recordisplay.cf
m?deid=20533.
U.S. EPA (U.S. Environmental Protection Agency).
(2002). A review of the reference dose and
reference concentration processes.
(EPA/630/P-02/002F). Washington, DC.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cf
m?deid=51717.
U.S. EPA (U.S. Environmental Protection Agency).
(2005a). Guidelines for carcinogen risk
assessment. (EPA/630/P-03/001F).
Washington, DC.
http: //www. epa. go v/cancer guidelines/.
U.S. EPA (U.S. Environmental Protection Agency).
(2005b). Supplemental guidance for assessing
susceptibility from early-life exposure to
carcinogens. (EPA/630/R-03/003F).
Washington, DC: U.S. Environmental Protection
Agency, Risk Assessment Forum.
http: //www. epa. go v/cancer guidelines/guideli
nes-carcinogen-supplementhtm.
U.S. EPA (U.S. Environmental Protection Agency).
(2006a). Approaches for the application of
physiologically based pharmacokinetic (PBPK)
models and supporting data in risk assessment
(Final Report). (EPA/600/R-05/043F).
Washington, DC: U.S. Environmental Protection
Agency, Office of Research and Development.
U.S. EPA (U.S. Environmental Protection Agency).
(2006b). A framework for assessing health risk
of environmental exposures to children.
(EPA/600/R-05/093F). Washington, DC.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cf
m?deid=158363.
This document is a draft for review purposes only and does not constitute Agency policy.
xxi DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
U.S. EPA (U.S. Environmental Protection Agency).
(2009). EPAs Integrated Risk Information
System: Assessment development process.
Washington, DC.
http://epa.gov/iris/process.htm.
U.S. EPA (U.S. Environmental Protection Agency).
(2011a). Recommended use of body weight
3/4 as the default method in derivation of the
oral reference dose. (EPA/100/R11/0001).
Washington, DC.
http://www.epa.gov/raf/publications/intersp
ecies-extrapolation.htm.
May 2012 version
This document is a draft for review purposes only and does not constitute Agency policy.
xxii DRAFT—DO NOT CITE OR QUOTE
-------
EXECUTIVE SUMMARY
3
4
5 Occurren ce an d Health Effects
6
7 Ammonia occurs naturally in air, soil and water and is produced by humans
8 and other animals as part of normal biological processes. Ammonia is also used as
9 an agricultural fertilizer. Exposure to ammonia occurs primarily through breathing
10 air containing ammonia gas, and may also occur via diet or direct skin contact
11 Health effects observed at levels exceeding naturally-occurring
12 concentrations are generally limited to the site of direct contact with ammonia
13 (skin, eyes, respiratory tract, and digestive tract). Short-term exposure to high
14 levels of ammonia can cause irritation and serious burns on the skin and in the
15 mouth, lungs, and eyes. Chronic exposure to airborne ammonia can increase the
16 risk of respiratory irritation, cough, wheezing, tightness in the chest, and reduction
17 in the normal function of the lung. Studies in experimental animals similarly
18 suggest that breathing ammonia at sufficiently high concentrations can result in
19 effects on the respiratory system. Animal studies also suggest that exposure to high
20 levels of ammonia in air or water may adversely affect other organs, such as the
21 stomach, liver, adrenal gland, kidney, and spleen. There is inadequate information
22 to evaluate the carcinogenicity of ammonia.
23
24
25 Effects Other Than Cancer Observed Following Oral Exposure
26 There are few oral toxicity studies for ammonia. Gastric toxicity may be a hazard for
27 ammonia based on evidence from case reports in humans and mechanistic studies in experimental
28 animals. Evidence in humans is limited to case reports of individuals suffering from
29 gastrointestinal effects from ingesting household cleaning solutions containing ammonia or biting
30 into capsules of ammonia smelling salts. The experimental animal toxicity database for ammonia
31 lacks standard toxicity studies that evaluate a range of tissues/organs and endpoints. In rats,
32 gastrointestinal effects, characterized as increased epithelial cell migration in the mucosa of the
33 stomach leading to decreased thickness of the gastric mucosa, were reported following short-term
34 and subchronic exposures to ammonia via ingestion [Hataetal.. 1994: Tsujii etal.. 1993: Kawano et
35 al., 1991]. While these studies provide consistent evidence of changes in the gastric mucosa
36 associated with exposure to ammonia in drinking water, the investigators reported no evidence of
37 microscopic lesions of the stomach, gastritis, or ulceration in the stomachs of these rats.
38 Given the limited scope of toxicity testing of ingested ammonia and questions concerning
39 the adversity of the gastric mucosal findings in rats, the available oral database for ammonia was
This document is a draft for review purposes only and does not constitute Agency policy.
xxiii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Toxicological Review of Ammonia
considered insufficient to characterize toxicity outcomes and dose-response relationships, and an
oral reference dose (RfD) for ammonia was not derived.
Effects Other Than Cancer Observed Following Inhalation Exposure
Respiratory effects have been identified as a hazard following inhalation exposure to
ammonia. Evidence for respiratory toxicity associated with inhaled ammonia comes from studies
in humans and animals. Cross-sectional occupational studies involving chronic exposure to
ammonia have consistently demonstrated an increased prevalence of symptoms consistent with
respiratory irritation and decreased lung function [Rahman et al., 2007: Alietal., 2001: Ballaletal.,
1998: Holness etal.. 1989). Cross-sectional studies of livestock farmers exposed to ammonia,
controlled volunteer studies of ammonia inhalation, and case reports of injury in humans with
inhalation exposure to ammonia provide additional, consistent support for the respiratory system
as a target of ammonia toxicity. Additionally, respiratory effects were observed in several animal
species following short-term and subchronic inhalation exposures to ammonia.
The experimental toxicology literature for ammonia also provides evidence that inhaled
ammonia may be associated with toxicity to target organs other than the respiratory system,
including the liver, adrenal gland, kidney, spleen, heart, and immune system, at concentrations
higher than those associated with respiratory system effects. Less evidence exists for these effects
than for respiratory effects.
Inhalation Reference Concentration (RfC) for Effects Other Than Cancer
Table ES-1. Summary of reference concentration (RfC) derivation
Critical effect
Decreased lung function and increased
respiratory symptoms
Occupational epidemiology study
Holness etal. (1989)
Point of departure3
NOAELADJ: 3.1 mg/m3
UF
10
Chronic RfC
0.3 mg/m3
25
26
27
28
29
30
aBecause the study involved workplace exposure conditions, the NOAEL of 8.8 mg/m was adjusted for
continuous exposure based on the ratio of VEho (human occupational default minute volume of 10 m3 breathed
during an 8-hour workday) to VEh (human ambient default minute volume of 20 m3 breathed during the entire
day) and an exposure of 5 days out of 7 days.
NOAEL = no-observed-adverse-effect level; UF = uncertainty factor.
The study of ammonia exposure in workers in a soda ash plant by Holness etal. [1989] was
identified as the principal study for RfC derivation. Respiratory effects, characterized as increased
respiratory symptoms (including cough, phlegm, chronic bronchitis, wheeze, chest tightness, and
dyspnea) and decreased lung function, observed in workers exposed to ammonia were selected as
the critical effect. Holness etal. [1989] found no differences in the prevalence of respiratory
This document is a draft for review purposes only and does not constitute Agency policy.
xxiv DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 symptoms or lung function between workers in any of the three exposure categories, including the
2 high-exposure category (>8.8 mg/m3), and the control group. The Holness etal. [1989] study in
3 conjunction with a second occupational study by Rahman etal. [2007] collectively provide
4 information useful for examining the relationship between chronic ammonia exposure and
5 increased prevalence of respiratory symptoms and decreased lung function. Both studies reported
6 either the presence or absence of respiratory effects in workers exposed to ammonia over a range
7 of concentrations (approximately 4-18 mg/m3], with the no-observed-adverse-effect level [NOAEL]
8 of 8.8 mg/m3 from the Holness etal. [1989] study falling between the NOAEL and lowest-observed-
9 adverse-effect level [LOAEL] (4.9 and 18.5 mg/m3, respectively] from the Rahman et al. [2007]
10 study. The NOAEL of 8.8 mg/m3 (NOAELADj = 3.1 mg/m3, i.e., adjusted to continuous exposure]
11 from the Holness etal. [1989] study was used as the point of departure (POD] for RfC derivation.
12 An RfC of 0.3 mg/m3 was calculated by dividing the POD (adjusted for continuous
13 exposure, i.e., NOAELADj] by a composite uncertainty factor (UF] of 10 to account for potentially
14 susceptible individuals in the absence of data evaluating variability of response to inhaled ammonia
15 in the human population.
16
17 Confidence in the Chronic Inhalation RfC
18
19 Study - medium
20 Database - medium
21 RfC - medium
22
23 Under EPA's Methods for Derivation of Inhalation Reference Concentrations and Application
24 of Inhalation Dosimetry [U.S. EPA. 1994]. the overall confidence in the RfC is medium and reflects
25 medium confidence in the principal study (adequate design, conduct, and reporting of the principal
26 study; limited by small sample size and identification of a NOAEL only] and medium confidence in
27 the database, which includes occupational and volunteer studies and studies in animals that are
28 mostly of subchronic duration. There are no studies of developmental toxicity and studies of
29 reproductive and other systemic endpoints are limited; however, reproductive, developmental, and
30 other systemic effects are not expected at the RfC because it is well documented that ammonia is
31 endogenously produced in humans and animals, ammonia concentrations in blood are
32 homeostatically regulated to remain at low levels, and ammonia concentrations in air at the POD
33 are not expected to alter homeostasis.
34
35 Evidence of Carcinogenicity
36 Under EPA's Guidelines for Carcinogen Risk Assessment [U.S. EPA, 2005a], there is
37 "inadequate information to assess carcinogenic potential" for ammonia, based on the absence
38 of ammonia carcinogenicity studies in humans and a single lifetime drinking water study of
39 ammonia in mice [Toth, 1972] that showed no evidence of carcinogenic potential. There is limited
40 evidence that ammonia may act as a cancer promoter [Tsujiietal.. 1995: Tsujiietal.. 1992a]. The
This document is a draft for review purposes only and does not constitute Agency policy.
xxv DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 available genotoxicity studies are inadequate to characterize the genotoxic potential of ammonia. A
2 quantitative cancer assessment for ammonia was not conducted.
o
J
4 Susceptible Populations and Lifestages
5 Hyperammonemia is a condition of elevated levels of circulating ammonia that can occur in
6 individuals with severe diseases of the liver or kidney or with hereditary urea [CO(NH2)2] cycle
7 disorders. These elevated ammonia levels can predispose an individual to encephalopathy due to
8 the ability of ammonia to cross the blood-brain barrier; these effects are especially marked in
9 newborn infants. Thus, individuals with disease conditions that lead to hyperammonemia may be
10 more susceptible to the effects of ammonia from external sources, but there are no studies that
11 specifically support this susceptibility.
12 Studies of the toxicity of ammonia in children or young animals compared to other
13 lifestages that would support an evaluation of childhood susceptibility have not been conducted.
14
15 Key Issues Addressed in the Assessment
16 Endogenous Ammonia
17 Ammonia, which is produced endogenously, has been detected in the expired air of healthy
18 volunteers. Ammonia concentrations in breath exhaled from the mouth or oral cavity (0.085-
19 2.1 mg/m3) are higher and more variable than concentrations measured in breath exhaled from the
20 nose and trachea (0.013-0.078 mg/m3). See Appendix D, Section D.3 (Elimination) and Table D-l
21 for further discussion of studies that examined ammonia in exhaled breath. Concentrations exhaled
22 from the mouth and oral cavity are largely attributed to the production of ammonia via bacterial
23 degradation of food protein in the oral cavity or gastrointestinal tract, and can be influenced by
24 factors such as diet, oral hygiene, and age. In contrast, the lower ammonia concentrations
25 measured in breath exhaled from the nose and trachea more likely reflect levels of ammonia
26 circulating in the blood. These levels are lower than the ammonia RfC of 0.3 mg/m3 by a factor of at
27 least fourfold. Although the RfC falls within the range of concentrations measured in the mouth or
28 oral cavity, exhaled ammonia is rapidly diluted in the larger volume of ambient air and would not
29 contribute significantly to ammonia exposure. Further, occupational epidemiology studies served
30 as the basis for the ammonia RfC; the worker populations in these studies would have been exposed
31 to ammonia that also included endogenously produced ammonia, and as such the RfC accounts for
32 ammonia exposures from endogenous sources.
33
This document is a draft for review purposes only and does not constitute Agency policy.
xxvi DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
2 LITERATURE SEARCH STRATEGY | STUDY
3 SELECTION
4
5
6 The primary, peer-reviewed literature pertaining to ammonia was identified through a
7 keyword search of the databases listed in Table LS-1. References from health assessments
8 developed by other national and international health agencies and review articles were also
9 examined. EPA requested the public submit additional data on December 21, 2007 [U.S. EPA.
10 2007]: no submissions were received. The last search was conducted in March 2012.
11 Figure LS-1 depicts the literature search and study selection strategy and the number of references
12 obtained at each stage of literature screening. Approximately 22,400 references were identified
13 with the initial keyword search. Based on a secondary keyword search followed by a preliminary
14 manual screen of titles or abstracts by a toxicologist, approximately 1,022 references were
15 identified that provided information potentially relevant to characterizing the health effects or
16 physical and chemical properties of ammonia. A more detailed review of titles, abstracts, and/or
17 papers pared this to 32 epidemiological studies (i.e., studies of occupational or livestock worker
18 populations or short-term exposure studies in volunteers), 43 case reports, 62 oral or inhalation
19 animal studies, 104 other studies (e.g., studies that provided supporting information on physical
20 and chemical properties, mechanisms, and toxicokinetics). The majority of the toxicokinetics
21 studies came from the ATSDR (2004] Toxicological Profile of Ammonia2 or were identified based on
22 a focused keyword search (e.g., for studies on ammonia in exhaled breath or ammonia in fetal
23 circulation].
24
25
2Portions of this Toxicological Review were developed under a Memorandum of Understanding with the
Agency for Toxic Substances and Disease Registry (ATSDR) and were adapted from the Toxicological Profile
for Ammonia (ATSDR. 2004) and the references cited in that document as part of a collaborative effort in the
development of human health lexicological assessments for the purposes of making more efficient use of
available resources and to share scientific information.
This document is a draft for review purposes only and does not constitute Agency policy.
xxvii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Table LS-1. Details of the literature search strategy
Database
Keywords
Pubmed
Toxcenter
Toxline
Current Contents
(2008 & 2010 only)
Chemical names (CASRN): ammonia (7664-41-7); ammonium hydroxide (1336-21-6)
Synonyms: spirit of hartshorn; aquammonia
Initial keyword search
Standard toxicology search
Toxicity (including duration, effects to children and occupational exposure); development;
reproduction; teratogenicity; exposure routes; pharmacokinetics; toxicokinetics; metabolism;
body fluids; endocrinology; carcinogenicity; genotoxicity; antagonists; inhibitors
Chemical-specific keywords
Respiration; metabolism; breath tests; inhalation; air; breath; exhalation; biological markers;
analysis
Secondary keyword search0
reproductive; developmental; teratogen; gastrointestinal; stomach; gastric AND mucosa,
cancer OR tumor; genotoxicity; kidney OR spleen AND toxicity; exhaled breath; respiratory
irritation, symptom OR disease, including dyspnea, bronchitis, pneumonitis, asthma; lung;
pulmonary function; chest tightness; inflammation; congestion; edema; hemorrhage;
discharge; epithelium; immune; immunosuppression; hypersensitivity; skin lesion; erythema;
host resistance; bacterial colonization; T-cell; liver function OR toxicity; fatty liver; clinical
chemistry; adrenal; heart AND toxicity; myocardium; lacrimation; ocular symptoms; blood
pH; brain AND amino acid; neurotransmitter.
The following terms were used to filter out reference not relevant to the evaluation of the
health effects of ammonia: hyperammonemia; ammonemia; hepatic coma; liver failure; Reye
syndrome; hepatic encephalopathy; cirrhosis; fish; daphnia; crustaceans; amphibians.
TSCATS
Searched by chemical names (including synonyms) and CASRNs
ChemID
Chemfinder
CCRIS
HSDB
GENETOX
RTECS
aThe use of certain keywords in a given database was contingent on number and type of results. The large number
of search results required restriction search terms to filter out references not relevant to evaluation of ammonia
health effects and limiting metabolism results to studies in animals and humans.
bAs discussed in the Preface, literature on ammonium salts were not included in this review because of the
uncertainty as to whether the anion of the salt can influence the toxicity of the ammonium compound (see also
Appendix B, Table B-l).
Secondary keywords were selected from an understanding of the targets of ammonia toxicity gained from review
of papers identified in literature searches conducted at the start of document development and relevant review
documents.
This document is a draft for review purposes only and does not constitute Agency policy.
xxviii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Referencesidentified based on initial keyword search (see Table LS-1): ~22,400
Referencesexcluded based on secondary keyword search (see
Table LS-1): ~13,270
Referencesidentified based on secondary keyword search (see Table LS-1): ~9,130
Reference excluded based on preliminary manual screen of
titles/abstracts: ~8,110
Reasons/or excluding references included the following:
• Topics not relevant to ammonia toxicity
• Co-exposure with other chemicals
Referencesconsidered for inclusion in the Toxicological Review
Human studies: 220
• Epidemiologic studies: 93
• Case reports: 127
Animal studies (oral & inhalation): 203
Other supporting studies: 599
Including:
• Reviews
• Background and physical/chemical properties
• Animal studies by routes other than oral & inhalation
• Studies of H. pylori and ammonia
• Studies related to mode of action
Referencesexcluded based on manual reviewof
papers/abstracts: 781
Types of papers evaluated and not considered further for
inclusion in the Toxicological Review:
• Concernsabout ethical conduct (Kalandarov et al., 1984)
• Not relevant to ammonia toxicity
• Inadequate information to characterize exposure
• Exposure route not relevant
• Co-exposure with other chemicals
• Nonstandard animal model (e.g., nonmammalian species,
cattle, etc.)
• Pathogenic effects of H. pylori infection
• Review paper
• Abstract
• Not available in English
• Duplicate
Referencescited in the Toxicological
Human studies/reports: 75
• Epidemiologic studies: 32
• Occupational studies (5)
• Studiesin volunteers (12)
• Studiesin livestock workers
(15)
• Case reports: 43
Review
Animal studies: 62
• Oral: 13
• Acute (3)
• Subchronic(7)
• Chronic (3)
• Inhalation: 49
• Acute/short-term (33)
• Subchronic(9)
• Reproductive/
developmental (1)
• Immunotoxicity (6)
Other supporting studies: 104
• Background and physical & chemical
properties: 15
• Studies related to mode of action,
including genotoxicity: 14
• Toxicokinetic studies: 70
• Miscellaneous: 5
Assessments bv others: 7
Guidances/notices: 27
Figure LS-1. Literature search and study selection strategy for ammonia.
This document is a draft for review purposes only and does not constitute Agency policy.
xxix DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 Selection of studies for inclusion in the Toxicological Review was based on consideration of
2 the extent to which the study was informative and relevant to the assessment and general study
3 quality considerations. In general, the scientific quality of the available studies was evaluated as
4 outlined in the Preamble and in EPA guidance (i.e., A Review of the Reference Dose and Reference
5 Concentration Processes [U.S. EPA. 2002] and Methods for Derivation of Inhalation Reference
6 Concentrations and Application of Inhaled Dosimetry [U.S. EPA. 1994]].
7 The majority of the human studies consisted of case reports involving acute ammonia
8 exposure; because case reports are generally anecdotal and thereby provide little information that
9 would be useful for characterizing chronic health hazards. These studies were only briefly
10 reviewed, and representative citations from this collection of literature are provided as
11 supplemental information in Appendix D, Section D.2.
12 The references considered for inclusion, as well as those cited in this document, including
13 bibliographic information and abstracts, can be found on the Health and Environmental Research
14 On-line (HERO] website3 (http://hero.epa.gov/ammonia].
15
16
3HERO is a database of scientific studies and other references used to develop EPA's risk assessments aimed
at understanding the health and environmental effects of pollutants and chemicals. It is developed and
managed in EPA's Office of Research and Development (ORD) by the National Center for Environmental
Assessment (NCEA). The database includes more than 300,000 scientific articles from the peer-reviewed
literature. New studies are added continuously to HERO.
This document is a draft for review purposes only and does not constitute Agency policy.
xxx DRAFT—DO NOT CITE OR QUOTE
-------
1
2 1. HAZARD IDENTIFICATION
4 1.1. Synthesis of Evidence
5 1.1.1. Respiratory Effects
6 The respiratory system is the primary target of toxicity of inhaled ammonia in humans and
7 experimental animals. Four cross-sectional occupational epidemiology studies [Rahman etal.,
8 2007: Ali etal., 2001: Holness etal., 1989] examined the association between inhaled ammonia and
9 prevalence of respiratory symptoms and changes in lung function. The association between
10 ammonia exposure and respiratory effects suggested by these cross-sectional studies is also
11 informed by studies of livestock farmers, volunteer studies involving acute exposures to inhaled
12 ammonia, human case reports, and subchronic inhalation toxicity studies in various experimental
13 animal species. The evidence of respiratory effects in humans and experimental animals exposed to
14 ammonia is summarized in Tables 1-1 and 1-2, respectively, and as an exposure-response array in
15 Figure 1-1 atthe end of this section.
16
17 Respiratory Symp toms
18 Ammonia is an upper respiratory tract irritant in humans. Respiratory symptoms
19 (including cough, chest tightness, stuffy/runny nose, sneezing, phlegm, wheezing, dyspnea, chronic
20 bronchitis, and asthma) were reported in two cross-sectional studies of industrial worker
21 populations exposed to ammonia [Rahman et al., 2007: Ballal etal., 1998] (see Table 1-1 atthe end
22 of this section]. Rahman etal. (2007]4 found up to a 4.1-fold higher prevalence of respiratory
23 symptoms (cough, chest tightness, stuffy nose, runny nose, and sneezing] in workers exposed to a
24 mean ammonia concentration of 18.5 mg/m3 (high-exposure group] for about 16 years compared
25 to a control group (administration building workers]; the prevalence of cough and chest tightness
26 were statistically significantly elevated in the high-exposure group compared to the control group.
27 The prevalences of respiratory symptoms in the low-exposure group exposed to a mean ammonia
28 concentration of 4.9 mg/m3 were up to threefold higher than those in the control group, but none
29 were statistically significantly different from control.
30 Significantly higher relative risks (ranging from 1.6- to 4.7-fold] for cough, phlegm,
31 wheezing, dyspnea, chronic bronchitis, and asthma were also observed in workers from another
32 cross-sectional study (Ballal etal.. 1998] with ammonia exposure concentrations higher than the
4Rahman et al. (2007) examined respiratory effects in workers from two plants in a urea fertilizer factory.
Workers in the urea plant were exposed to higher concentrations of ammonia (arithmetic mean =
18.5 mg/m3) than workers in the ammonia plant (arithmetic mean = 4.9 mg/m3). Therefore, the urea plant
workers represented the high-exposure group, and the ammonia plant workers represented the low-
exposure group. Exposure to dusts and other contaminants, except for nitrogen dioxide, were not measured;
however, based on information about the production process and previous literature, the authors considered
ammonia to be the major exposure agent in this work environment.
This document is a draft for review purposes only and does not constitute Agency policy.
1-1 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 American Conference of Governmental Industrial Hygienists [ACGIH] threshold limit value [TLV] of
2 18 mg/m3 compared with workers exposed to levels below the TLV. Distribution of respiratory
3 symptoms by cumulative ammonia concentration (CAC, mg/m3-years) also showed significantly
4 higher relative risks for respiratory symptoms among workers with higher CAC (>50 mg/m3-years)
5 compared to those with a lower CAC (<50 mg/m3-years) [Ballal etal.. 1998). Only Ballal et al.
6 [1998] evaluated respiratory symptoms in terms of cumulative ammonia exposure.
7 In a third cross-sectional study of ammonia-exposed male workers, no differences were
8 observed in the prevalence of respiratory symptoms, eye irritation, or odor detection threshold
9 between the ammonia-exposed workers (concentrations were relatively lower than those in
10 Rahman etal. [2007] and Ballal etal. [1998]) and the control group fHolness etal.. 19891 when
11 evaluating all ammonia-exposed workers as one group or when stratifying them into three
12 exposure categories: high = >8.8 mg/m3, medium = 4.4-8.8 mg/m3, or low = <4.4 mg/m3. Although
13 respiratory irritation prevalence was similar across groups, the exposed workers reported that
14 exposure in the plant aggravated some of their reported respiratory symptoms (cough, sputum,
15 chronic bronchitis, wheeze, chest tightness, dyspnea, chest pain, rhinitis); however, no further
16 information was provided as to how the authors evaluated aggravation of symptoms. Co-exposures
17 to dust and inorganic gases such as nitrogen dioxide and sulfur dioxide were possible in these
18 cross-sectional studies; however, except for the low levels of nitrogen dioxide identified in the
19 Rahman etal. (2007] study, these workplace exposures were not measured or reported.
20 Overall, the cross-sectional occupational epidemiology studies that evaluated the
21 prevalence of respiratory symptoms provide consistent estimates of the effect level associated with
22 exposure to ammonia. Rahman etal. (2007] observed that exposure to 18.5 mg/m3 ammonia
23 increased the prevalence of respiratory symptoms (up to 4.1-fold]. This is consistent with the
24 observation by Ballal etal. (1998] that workers in a factory with ammonia concentrations
25 exceeding the TLV of 18 mg/m3 had significantly higher relative risks (up to 4.7-fold] for
26 respiratory symptoms. The prevalence of respiratory symptoms was not increased following
27 occupational exposures at lower workplace concentrations; i.e., >8.8 mg/m3 (Holness etal., 1989]
28 and 4.9 mg/m3 ammonia (Rahman et al., 2007].
29 Elevated prevalence of respiratory symptoms, including cough, phlegm, wheezing, chest
30 tightness, and eye, nasal, and throat irritation, have been reported in livestock farmers and stable
31 workers compared to controls (Melbostad and Eduard. 2001: Preller etal.. 1995: Choudatetal..
32 1994: Zeida etal.. 1994: Crook etal.. 1991: Heederik etal.. 19901: fMonso etal.. 20041 (see
33 Appendix D, Section D.2 and Table D-7 for more detailed information]. Additionally, bronchial
34 hyperreactivity to methacholine or histamine challenge (tests used to assist in the diagnosis of
35 asthma by provoking bronchoconstriction] was increased in farmers exposed to ammonia
36 compared to control workers (Vogelzangetal., 2000: Vogelzangetal., 1997: Choudatetal., 1994],
37 indicating that exposure to ammonia and other air contaminants in farm settings may contribute to
38 chronic airway inflammation. In addition to ammonia, these studies also documented exposures to
39 airborne dust, bacteria, fungal spores, endotoxin, and mold—agents that could also induce
40 respiratory symptoms and airway effects. The release of other volatiles on livestock farms is likely,
This document is a draft for review purposes only and does not constitute Agency policy.
1-2 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 but measurements for other volatile chemicals were not conducted. Therefore, while several
2 studies have reported associations between ammonia exposure in livestock farmers or stable
3 workers and an increase in respiratory symptoms, these findings are of limited use because of
4 exposures to other constituents in air that likely confound this association.
5 Reports of irritation and hyperventilation in volunteers acutely exposed to ammonia at
6 concentrations ranging from 11 to 354 mg/m3 ammonia for durations up to 4 hours under
7 controlled exposure conditions [Petrovaetal., 2008: Smeets etal., 2007: Altmannetal., 2006: Ihrig
8 etal.. 2006: Verberk, 1977: Silverman et al., 1949] provide support for ammonia as a respiratory
9 irritant (see Appendix D, Section D.2 and Table D-8 for more detailed information, including
10 documentation of human subjects research ethics procedures). Two controlled-exposure studies
11 report habituation to eye, nose, and throat irritation in volunteers after several weeks of ammonia
12 exposure [Ihrig etal.. 2006: Ferguson et al.. 1977). Numerous case reports document the acute
13 respiratory effects of inhaled ammonia, ranging from mild symptoms (including nasal and throat
14 irritation and perceived tightness in the throat) to moderate effects (including pharyngitis,
15 tachycardia, dyspnea, rapid and shallow breathing, cyanosis, transient bronchospasm, and rhonchi
16 in the lungs) to severe effects (including burns of the nasal passages, soft palate, posterior
17 pharyngeal wall, and larynx, upper airway obstruction, bronchospasm, dyspnea, persistent,
18 productive cough, bilateral diffuse rales and rhonchi, mucous production, pulmonary edema,
19 marked hypoxemia, and necrosis of the lung) (see Appendix D, Section D.2, for more detailed
20 information and references).
21 Experimental studies in laboratory animals also provide consistent evidence that repeated
22 exposure to ammonia can affect the respiratory system (see Appendix D, Section D.3 for more
23 detailed information). The majority of available animal studies did not look at measures of
24 respiratory irritation (in contrast to the majority of human studies), but rather examined
25 histopathological changes of respiratory tract tissues. Histopathological changes in the nasal
26 passages were observed in Sherman rats after 75 days of exposure to 106 mg/m3 ammonia or in
27 F344 rats after 35 days of exposure to 177 mg/m3 ammonia, with respiratory and nasal epithelium
28 thicknesses increased 3-4 times that of normal (Brodersonetal., 1976). Thickening of nasal and
29 tracheal epithelium (50-100%) was also observed in pigs exposed to 71 mg/m3 ammonia
30 continuously for 1-6 weeks (Doig and Willoughby, 1971). Nonspecific inflammatory changes (not
31 further described) were reported in the lungs of Sprague-Dawley and Long-Evans rats continuously
32 exposed to 127 mg/m3 ammonia for 90 days and rats and guinea pigs intermittently exposed to
33 770 mg/m3 ammonia for 6 weeks; continuous exposure to 455 and 470 mg/m3 ammonia increased
34 mortality in rats (Coon etal., 1970). Focal or diffuse interstitial pneumonitis was observed in all
35 Princeton-derived guinea pigs, New Zealand white rabbits, beagle dogs, and squirrel monkeys
36 exposed to 470 mg/m3 ammonia (Coon etal., 1970). Additionally, under these exposure conditions,
37 dogs exhibited nasal discharge and other signs of irritation (marked eye irritation, heavy
38 lacrimation). Nasal discharge was observed in 25% of rats exposed to 262 mg/m3 ammonia for 90
39 days (Coonetal.. 1970).
This document is a draft for review purposes only and does not constitute Agency policy.
1-3 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 At lower concentrations, approximately 50 mg/m3 and below, the majority of studies of
2 inhaled ammonia show that ammonia does not produce respiratory effects in laboratory animals.
3 Lung congestion, edema, and hemorrhage were observed in guinea pigs and mice exposed to 14
4 mg/m3 ammonia for 42 days [Anderson et al. [1964]. However, no increase in the incidence of
5 respiratory or other diseases common to young pigs were observed after continuous exposure to
6 ammonia and inhalable dust at concentrations representative of those found in commercial pig
7 farms (< 26 mg/m3 ammonia) for 5 weeks [Done etal., 2005]. No gross or histopathological
8 changes in the turbinates, trachea, and lungs of pigs were observed after continuous exposure to 35
9 or 53 mg/m3 ammonia for up to 109 days [Curtis etal., 1975]. No signs of toxicity in rats or dogs
10 were observed after continuous exposure to 40 mg/m3 ammonia for 114 days or after intermittent
11 exposure (8 hours/day] to 155 mg/m3 ammonia for 6 weeks [Coon etal., 1970].
12
13 Lung Function
14 Decreased lung function in ammonia-exposed workers has been reported in two cross-
15 sectional studies of industrial worker populations [Rahman etal., 2007: Alietal., 2001] that
16 measured lung function [Rahman et al., 2007: Ali etal., 2001: Holness etal., 1989]. Ammonia
17 exposure was correlated with a significant decline in lung function over the course of a work shift
18 (cross-shift] as measured by forced vital capacity (FVC] and forced expiratory volume in 1 second
19 (FEVi % predicted] in the high-exposure worker group (mean ammonia concentration of 18.5
20 mg/m3] in a fertilizer factory (Rahman etal.. 2007]. In a second study (Alietal.. 2001]. the FVC%
21 predicted was higher in fertilizer factory workers exposed to ammonia than in controls (4.6%
22 increase, p < 0.002]; FEVi % predicted was higher (1.5%] in the exposed workers but the difference
23 was not statistically significant When Alietal. (2001] based their analysis on measures of
24 cumulative exposure, workers with cumulative exposure >50 mg/m3-years had significantly lower
25 FVC% predicted (5.4% decrease, p < 0.030] and FEVi% predicted (7.4% decrease, p < 0.006] than
26 workers with cumulative ammonia exposure <50 mg/m3-years, but had similar FEVi/FVC%. The
27 authors did not explain the inconsistent findings across the analyses of noncumulative and
28 cumulative exposures.
29 Lung function did not appear to be affected in worker populations chronically exposed to
30 ammonia at concentrations below approximately 18 mg/m3. Baseline lung function, based on
31 spirometry (test measuring lung function volume and flow] conducted at the beginning and end of
32 the work shift, differed very slightly relative to control in workers exposed to ammonia
33 concentrations ranging from <4.4 to >8.8 mg/m3 in a cross-sectional study of male workers in a
34 soda ash plant (Holness etal., 1989], but was not statistically significant Additionally, no changes
35 in lung function were observed over either work shift (days 1 or 2] or over the work week in the
36 exposed group compared with controls. Similarly, measures of lung function (FVC, FEVi, and PEFR
37 [peak expiratory flow rate]] in workers exposed to a mean concentration of 4.9 mg/m3 ammonia
38 (low-exposure group] in a urea [CO(NH2]2] fertilizer factory showed no significant cross-shift
39 changes.
This document is a draft for review purposes only and does not constitute Agency policy.
1-4 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 Decreased lung function (e.g., measured as decreased FEVi, FVC) was reported in farmers
2 with ammonia exposure from animal waste (Monso et al., 2004: Cormier et al., 2000: Donham etal.,
3 2000: Vogelzangetal.. 1998: Reynolds etal.. 1996: Donham etal.. 1995: Preller etal.. 1995: Crook
4 etal., 1991: Heederiketal., 1990] (see Appendix D, Section D.2 and Table D-7). These findings are
5 of limited use because of the failure of these studies to account and control for exposures to other
6 constituents in air (including respirable dust, bacteria, fungal spores, endotoxin, and mold) that can
7 affect lung function, and likely confound the association between exposure to ammonia and
8 decreased lung function observed in these study populations.
9 Changes in lung function following acute exposure to ammonia have been observed in some,
10 but not all, controlled exposure studies conducted in volunteers (see Appendix D, Section D.2 and
11 Table D-8). Cole etal. (1977] reported reduced lung function as measured by reduced expiratory
12 minute volume and changes in exercise tidal volume in volunteers exposed for a half-day in a
13 chamber at ammonia concentrations >106 mg/m3, but not at 71 mg/m3. Bronchoconstriction was
14 reported in volunteers exposed to ammonia through a mouthpiece for 10 inhaled breaths of
15 ammonia gas at a concentration of 60 mg/m3 (Douglas and Coe, 1987]: however, there were no
16 bronchial symptoms reported in volunteers exposed to ammonia at concentrations of up to 35
17 mg/m3 for 10 minutes in an exposure chamber (MacEwen etal.. 1970]. Similarly, no changes in
18 bronchial responsiveness or lung function (as measured by FVC and FEVi] were reported in healthy
19 volunteers exposed to ammonia at concentrations up to 18 mg/m3 for 1.5 hours during exercise
20 (Sundbladetal., 2004]. There were no changes in lung function as measured by FEVi in 25 healthy
21 volunteers and 15 mild/moderate persistent asthmatic volunteers exposed to ammonia
22 concentrations up to 354 mg/m3 ammonia for up to 2.5 hours (Petrovaetal., 2008], or in 6 healthy
23 volunteers and 8 mildly asthmatic volunteers exposed to 11-18 mg/m3 ammonia for 30-minute
24 sessions (Sigurdarsonetal.. 2004].
25 Lung function effects following ammonia exposure were not evaluated in the available
26 animal studies.
27
28
This document is a draft for review purposes only and does not constitute Agency policy.
1-5 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1-1. Evidence pertaining to respiratory effects in humans following
itinn evnnsiire
Table
inhalation exposure
Study design and reference
Results
Respiratory symptoms
Cross-sectional occupational study of soda ash
plant workers in Canada; 58 exposed workers and
31 controls (from stores and office areas of
plant)3
Low: <6.25 ppm (<4.4 mg/m3); n = 34
Medium: 6.25-12.5 ppm (4.4-8.8 mg/m3); n = 12
High: >12.5 ppm (>8.8 mg/m3); n = 12
Average exposure: 12 y
Holnessetal. (1989)
Cross-sectional occupational study of urea
fertilizer factory in Bangladesh; 63 ammonia plant
workers, 77 urea plant workers, and 25 controls
(administration building staff)
Low-exposure group (ammonia plant) : 6.9 ppm
(4.9 mg/m3)
High-exposure group (urea plant)b: 26.1 ppm
(18.5 mg/m3)
Mean employment duration: 16 y
Rahman et al. (2007)
Cross-sectional occupational study of two urea
fertilizer factories in Saudi Arabia; 161 exposed
workers and 355 unexposed controls0
Exposures were stratified > or < the ACGIH TLV of
18 mg/m3
Mean employment duration: 51.8 mo (exposed
workers) and 73.1 mo (controls)
Ballaletal. (1998)
Percentage of workers reporting symptoms (%):
Control Exposed
(n = 31) (n = 58) p-value
Flu 3 7 0.63
Cough 10 16 0.53
Sputum 16 22 0.98
Bronchitis 19 22 0.69
Wheeze 10 10 0.91
Chest tightness 6 3 0.62
Dyspnea 13 7 0.05
Chest pain 6 2 0.16
Rhinitis 19 10 0.12
Throat 3 7 0.53
Percentage of workers reporting symptoms (%):
Low exposed High exposed
Control (p-value)1 (p-value)2
(n = 25) (n = 63) (n = 77)
Cough 8 17(0.42) 28 (0.05) (0.41)
Chest tightness 8 17 (0.42) 33 (0.02) (0.19)
Stuffy nose 4 12(0.35) 16 (0.17) (1.0)
Runny nose 4 4(1.0) 16 (0.17) (0.28)
Sneeze 8 0(0.49) 22 (0.22) (0.01)
Vvalue for ammonia plant compared to control
2p-value for urea plant compared to control and for urea plant
compared to ammonia plant
Relative risks for those exposed to ammonia at concentrations
>TLV (>18 mg/m3) as compared to those exposed at levels
-------
Toxicological Review of Ammonia
Table 1-1. Evidence pertaining to respiratory effects in humans following
inhalation exposure
Study design and reference
Lung function
Cross-sectional occupational study of soda ash
plant workers in Canada; 58 exposed workers and
31 controls (from stores and office areas of
plant)3
Low: <6.25 ppm (<4.4 mg/m3); n = 34
Medium: 6.25-12.5 ppm (4.4-8.8 mg/m3); n = 12
High: >12.5 ppm (>8.8 mg/m3); n = 12
Average exposure: 12 y
Holnessetal. (1989)
Cross-sectional occupational study of urea
fertilizer factory in Bangladesh; 63 ammonia plant
workers, 77 urea plant workers, and 25 controls
(staff from administration building)
Low-exposure group (ammonia plant)b: 6.9 ppm
(4.9 mg/m3)
High-exposure group (urea plant) : 26.1 ppm
(18.5 mg/m3)
Mean employment duration: 16 y
Rahman etal. (2007)
Cross-sectional occupational study of a urea
fertilizer factory in Saudi Arabia— follow-up of
Ballal et al. (1998); 73 exposed workers and 348
unexposed controls
Exposures were stratified < or > the ACGIH TLV of
18 mg/m3
Mean employment duration: not reported
AN etal. (2001)
Results
Control Exposed
(n = 31) (n = 58) p-value
Lung function (% predicted values):
FVC 98.6 96.8 0.094
FEVi 95.1 94.1 0.35
FEVi/FVC 96.5 97.1 0.48
Change in lung function over work shift:
FVCdayl -0.9 -0.8 0.99
day 2 +0.1 -0.0 0.84
FEVidayl -0.2 -0.2 0.94
day 2 +0.5 +0.7 0.86
Pre-shift Post-shift p-value
Ammonia plant (low-exposure group); n = 24 of 63 ammonia
plant workersd
FVC 3.308 3.332 0.67
FEVi 2.627 2.705 0.24
PEFR 8.081 8.313 0.22
Urea plant (high-exposure group); n = 64 of 77 urea plant
workersd
FVC 3.362 3.258 0.01
FEVi 2.701 2.646 0.05
PEFR 7.805 7.810 0.97
p-value reflects the comparison of pre- and post-shift values.
Control Exposed
(n = 348) (n = 73) p-value
FEV^/o predicted 96.6 98.1 NS
FVC% predicted 101.0 105.6 0.002
FEVi/FVC% 83.0 84.2 NS
<50 mg/m3-y >50 mg/m3-y p-value
FVCi% 100.7 93.4 0.006
predicted
FVC% 105.6 100.2 0.03
predicted
FEV!/FVC°/o 84.7 83.4 NS
NS = not significant (p-values not provided by study authors)
aAt this plant, ammonia, carbon dioxide, and water were the reactants used to form ammonium bicarbonate,
which in turn was reacted with salt to produce sodium bicarbonate and subsequently processed to form sodium
carbonate. Ammonia and carbon dioxide were recovered in the process and reused.
This document is a draft for review purposes only and does not constitute Agency policy.
1-7 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Table 1-1. Evidence pertaining to respiratory effects in humans following
inhalation evnnsnre
inhalation exposure
Study design and reference
Results
bExposure concentrations were determined by both the Dra'ger tube and Dra'ger PAC III methods. Using the Dra'ger
tube method, concentrations of ammonia in the ammonia and urea plants were 17.7 and 88.1 mg/m , respectively;
using the Dra'ger PAC III method, ammonia concentrations were 4.9 and 18.5 mg/m3, respectively (Rahman et al.
(2007). The study authors observed that their measurements indicated only relative differences in exposures
between workers and production areas, and that the validity of the exposure measures could not be evaluated
based on their results. Based on communication with technical support at Dra'ger Safety Inc (telephone
conversations and e-mails dated June 22, 2010, from Michael Yanosky, Dra'ger Safety Inc., Technical Support
Detection Products to Amber Bacom, SRC, Inc., contractor to NCEA, ORD, U.S. EPA), EPA considered the PAC III
instrument to be a more sensitive monitoring technology than the Dra'ger tubes. Therefore, more confidence is
attributed to the PAC III air measurements of ammonia for the Rahman et al. (2007) study.
°The process of fertilizer production involved synthesis of ammonia from natural gas, followed by reaction of the
ammonia and carbon dioxide to form ammonium carbamide, which was then converted to urea.
dLung function testing was not performed on all workers; only the morning shift was chosen for data collection for
practical reasons and workers who planned to have less than a 4-hr working day were excluded.
1
This document is a draft for review purposes only and does not constitute Agency policy.
1-8 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Table 1-2. Evidence pertaining to respiratory effects in animals following
inhalation exposure
Study design and reference
Results
Effects on the lungs
Squirrel monkey (Saimiri sciureus); male; 3/group
Beagle dog; male; 2/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Sprague-Dawley & Long-Evans rat; male and female; 15-51/group
0,155, or 770 mg/m3 8 hrs/d, 5 d/wk for 6 wks
Coon etal. (1970)
Gross necropsies were normal; focal
pneumonitis in one of three monkeys at
155 mg/m3.
Nonspecific lung inflammation observed in
guinea pigs and rats but not in other species
at 770 mg/m3.3
Squirrel monkey (5. sciureus); male; 3/group
Beagle dog; male; 2/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Coon etal. (1970)
Focal or diffuse interstitial pneumonitis in all
animals. Calcification of bronchial
epithelium observed in several animals.
Hemorrhagic lung lesion in one of two dogs;
moderate lung congestion in two of three
rabbits.3
Sprague-Dawley or Long-Evans rat; male and female; 15-51/group
0 or 40 mg/m3 for 114 d or 127, 262 or 470 mg/m3 for 90 d or 455
mg/m3 for 65 d
Coon etal. (1970)
Dyspnea (mild) at 455 mg/m . Focal or
diffuse interstitial pneumonitis in all
animals, and calcification of bronchial
epithelium observed in several animals at
470 mg/m3.3'b
Guinea pig (strain not specified); male and female; 2/group
0 or 20 ppm (0 or 14 mg/m3) for 7-42 d or 50 ppm (35 mg/m3) for
42 d
Anderson etal. (1964)
Lung congestion, edema and hemorrhage
observed at 14 and 35 mg/m3 after 42 d.3
Swiss albino mouse; male and female; 4/group
0 or 20 ppm (0 or 14 mg/m3) for 7-42 d
Anderson etal. (1964)
Lung congestion, edema, and hemorrhage
observed at 14 mg/m3 after 42 d.3
Pig (several breeds); sex not specified; 24/group
0, 0.6,10,18.8, or 37 ppm (0, 0.4, 7,13.3, or 26 mg/m3) and 1.2,
2.7, 5.1, or 9.9 mg/m3 inhalable dust for 5 wks
(Exposure to ammonia and inhalable dust at concentrations
commonly found at pig farms)
Done etal. (2005)
No increase in the incidence of respiratory
or other diseases.
Pig (crossbred); sex not specified; 4-8/group
0, 50, or 75 ppm (0, 35, or 53 mg/m3 for 109 d)
Curtis etal. (1975)
Turbinates, trachea, and lungs of all pigs
were classified as normal.
Effects on the upper respiratory tract
Sherman rat; 5/sex/group
10 or 150 ppm (7 or 106 mg/m3) from bedding for 75 d
Brodersonetal. (1976)c
1" thickness of the nasal epithelium (3-4
times) and nasal lesions at 106 mg/m3.3
This document is a draft for review purposes only and does not constitute Agency policy.
1-9 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Table 1-2. Evidence pertaining to respiratory effects in animals following
inhalation exposure
Study design and reference
Results
F344 rat; 6/sex/group
0 or 250 ppm (0 or 177 mg/m3) in an inhalation chamber for 35 d
Brodersonetal. (1976)c
'f thickness of the nasal epithelium (3-4
times) and nasal lesions at 177 mg/m3.a
Yorkshire-Landrace pig; sex not specified; 6/group
0 or 100 ppm (0 or 71 mg/m3) for 6 wks
Doig and Willoughby (1971)
'f thickness of nasal and tracheal
epithelium (50-100% increase).3
Squirrel monkey (5. sciureus); male; 3/group
Beagle dog; male; 2/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Sprague-Dawley & Long-Evans rat; male and female; 15-51/group
0,155, or 770 mg/m3 8 hrs/d, 5 d/wk for 6 weeks
Coon etal. (1970)
Dyspnea in rats and dogs exposed to 770
mg/m3 during week 1 only; no indication of
irritation after week 1; nasal tissues not
examined for gross or histopathologic
changes.
Beagle dog; male; 2/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Coon etal. (1970)
Nasal discharge at 470 mg/m .a
Sprague-Dawley or Long-Evans rat; male and female; 15-51/group
0 or 40 mg/m3 for 114 d or 127, 262 or 470 mg/m3 for 90 d or
455 mg/m3 for 65 d
Coon etal. (1970)
Nasal irritation in all animals at
455 mg/m3.a'b
White albino mouse; male; 50
Ammonia vapor of 0 or 12% ammonia solution for 15 min/d,
6 d/wk, for 8 wks
Gaafaretal. (1992)
Histological changes in the nasal mucosa.
Duroc pig; both sexes; 9/group
12, 61,103,145 ppm (8, 43, 73, or 103 mg/m3) for 5 wks
Stombaughetal. (1969)
Excessive nasal, lacrimal, and mouth
secretions and /T" frequency of cough at 73
and 103 mg/m .a
Incidence data not provided.
bExposure to 455 and 470 mg/m3 ammonia increased mortality in rats.
°The Broderson et al. (1976) paper includes a number of experiments in rats designed to examine whether
ammonia at concentrations commonly encountered in laboratory cage environments plays a role in the
pathogenesis of murine respiratory mycoplasmosis caused by the bacterium Mycoplasma pulmonis. The
experiments conducted without co-exposure to M. pulmonis are summarized in this table; the results of
experiments involving co-exposure to M. pulmonis are discussed in Section 1.1.4, Immune System Effects.
This document is a draft for review purposes only and does not constitute Agency policy.
1-10 DRAFT—DO NOT CITE OR QUOTE
-------
1000
CO
in
o
100 -
I 10 ^
8
c
o
o
§
in
o
Q.
X
LLI
1 -
0.1
A T
A I
Respiratory symptoms & decreased
J£j 1C lung function (male occupational);
2 <— Holnesset al. (1989)
5 I
m if
l/i ^i Respiratory symptoms & decreased
lung function (occupational);
Rahman et al. (2007)
1 1 M . In
T i
A
A • ^
" " I
"
* * ^.w v ^, v ^ if)
I gf f || 1| I-? | ||
1§ J§ Hi! |l| fg 11 |1 =1
CH^H 'o1" £^^3-"5j OO ^3-- 1^3^ 2^(D
— ^ ^"'0. ^E ;H ^: g o ^3 ;-; o) Q- ai'a? ~ Q
Z0° oro ^-^go "- .0 ^ gjS g=s w
"° zc -S"E ro'l|"' °<= °E o
| I I"3 | |1 p |" 1
| ]s a z
& EXPERIMENTAL A
Effects on the lung
A
• LOAEL
ANOAEL range of concentrations
in study.
• Additional
concentrations
* exposures were intermittent : 8 h rs/d , 5 d/wk
0)2 0)2 oo '"dj
iS m iS m iE Q z^
!c !c
1— 1—
NIMAL STUDIES
Effects on the upper respiratory tract
Figure 1-1. Exposure-response array of respiratory effects following inhalation exposure.
This document is a draft for review purposes only and does not constitute Agency policy,
1-11
DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 Mode-of-Action Analysis—Respiratory Effects
2 Data regarding the potential mode of action for respiratory effects associated with chronic
3 exposure to ammonia are limited. However, acute exposure data demonstrate that injury to
4 respiratory tissues is primarily due to ammonia's alkaline (i.e., caustic) properties from the
5 formation of hydroxide ion when it comes in contact with water and is solubilized. Ammonia
6 readily dissolves in the moisture on the mucous membranes, forming ammonium hydroxide, which
7 causes liquefactive necrosis of the tissues. Specifically, ammonia directly denatures tissue proteins
8 and causes saponification of cell membrane lipids, which leads to cell disruption and death
9 (necrosis). In addition, the cellular breakdown of proteins results in an inflammatory response,
10 which further damages the surrounding tissues (Amshel etal.. 2000: Milleaetal.. 1989: Tarudi and
11 Golden. 1973).
12
13 Summary of Respiratory Effects
14 Evidence for respiratory toxicity associated with exposure to ammonia comes from studies
15 in humans and animals. Cross-sectional occupational studies involving chronic exposure to
16 ammonia have consistently demonstrated an increased prevalence of symptoms consistent with
17 respiratory irritation (Rahman etal.. 2007: Ballal etal.. 1998) and decreased lung function (Rahman
18 etal.. 2007: Ali etal.. 2001) (see Appendix D, Section D.2 and Tables D-3 to D-6). Cross-sectional
19 studies of livestock farmers exposed to ammonia, controlled volunteer studies of ammonia
20 inhalation, and case reports of injury in humans with inhalation exposure to ammonia provide
21 additional and consistent support for the respiratory system as a target of ammonia toxicity when
22 inhaled (see Appendix D, Section D.2 and Tables D-7 and D-8).
23 Short-term and subchronic animal studies show histopathological changes of respiratory
24 tissues in several animal species (lung inflammation in guinea pigs and rats; focal or interstitial
25 pneumonitis in monkeys, dogs, rabbits, and guinea pigs; pulmonary congestion in mice; thickening
26 of nasal epithelium in rats and pigs; nasal inflammation or lesions in rats and mice) across different
27 dose regimens (Gaafar etal., 1992: Brodersonetal., 1976: Doig and Willoughby, 1971: Coon etal.,
28 1970: Anderson etal.. 1964) (see Appendix D, Section D.3). In general, responses in respiratory
29 tissues increased with increasing ammonia exposure concentration. The evidence of observed
30 respiratory effects seen across multiple human and animal studies identifies respiratory system
31 effects as a hazard from ammonia exposure via inhalation.
32
33 1.1.2. Gastrointestinal Effects
34 Reports of gastrointestinal effects of ammonia in humans are limited to case reports
35 involving intentional or accidental ingestion of household cleaning solutions or ammonia inhalant
36 capsules (Dworkinetal.. 2004: Rosenbaum etal.. 1998: Christesen. 1995: Wasonetal.. 1990: Lopez
37 etal.. 1988: Klein etal.. 1985: Klendshoi and Reient. 19661 (see Appendix D, Section D.2). Clinical
38 signs of gastrointestinal effects reported in these case studies include stomachache, nausea,
39 diarrhea, drooling, erythematous and edematous lips, reddened and blistered tongues, dysphagia,
This document is a draft for review purposes only and does not constitute Agency policy.
1-12 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 vomiting, oropharyngeal burns, laryngeal and epiglottal edema, erythmatous esophagus with
2 severe corrosive injury, and hemorrhagic esophago-gastro-duodeno-enteritis.
3 The experimental animal toxicity database for ammonia lacks standard toxicity studies that
4 evaluate a range of tissues/organs and endpoints. Exposure to ammonia in drinking water has,
5 however, been associated with effects on the gastric mucosa. Evidence for this association comes
6 from animal studies [Hataetal.. 1994] designed to investigate the mechanisms by which the
7 bacterium Helicobacter pylori, which produces a potent urease that increases ammonia production,
8 may have a significant role in the etiology of chronic atrophic gastritis (see Appendix D, Section
9 D.3). Statistically significant decreases of 40-60% in the thickness of the antral gastric mucosa
10 were reported in Sprague-Dawley rats administered 0.01% ammonia in drinking water for
11 durations of 2-8 weeks [Tsujii etal., 1993: Kawano etal., 1991]: estimated doses were 22 mg/kg-
12 day [Kawano etal.. 1991] and 33 mg/kg-day [Tsujii etal.. 1993]. The magnitude of the decrease in
13 gastric mucosal thickness increased with dose and duration of ammonia exposure [Tsujii etal..
14 1993: Kawano etal., 1991]. Further, the effect was more prominent in the mucosa of the antrum
15 region of the stomach than in the body region of the stomach.5 Antral gastric mucosal thickness
16 decreased significantly (by 56-59% of the tap water control] at 4 and 8 weeks of exposure to
17 0.01% ammonia in drinking water, but there was no significant effect on the thickness of the body
18 gastric mucosa. Similarly, the height of fundic and pyloric glands in the gastric mucosa was
19 decreased by approximately 30% in Donryu rats exposed to ammonia in drinking water for up to
20 24 weeks at concentrations of 0.02 and 0.1% (estimated doses of 28 and 140 mg/kg-day,
21 respectively] (Hataetal.. 1994].
22 Mucosal cell proliferation and migration (as measured by 5-bromo-2'-deoxyuridine
23 labeling] were also significantly increased in rats exposed to ammonia (Tsujii etal., 1993]. The
24 authors observed that it was not clear whether mucosal cell proliferation was primarily stimulated
25 directly by ammonia or indirectly by increased cell loss followed by compensatory cell
26 proliferation. Cell proliferation in the gastric mucosa was also affected in the 24-week drinking
27 water study in Donryu rats (Hataetal.. 1994]. although the pattern differed from that reported by
28 Tsujii etal. (1993]. The labeling index in gastric mucosal glands was increased at earlier time
29 points (up to week 1 for fundic glands and up to week 4 for pyloric glands], suggesting enhanced
30 cell cycling subsequent to repeated erosion and repair. At later time points (up to 24 weeks of
31 exposure], however, the labeling index was decreased, a finding the authors' attributed to reduced
32 capability of the generative cell zone of the mucosal region.
33 The gastric changes observed by Kawano etal. (1991]. Tsujii etal. (1993]. and Hata etal.
34 (1994] were characterized by the study authors as consistent with changes observed in human
35 atrophic gastritis; however, Kawano etal. (1991] and Tsujii etal. (1993] observed that no mucosal
36 lesions were found macroscopically or microscopically in the stomachs of rats after exposure to
37 ammonia in drinking water for 4-8 weeks, and Hataetal. (1994] reported that there was no
5The body is the main, central region of the stomach. The antrum is the distal part of the stomach near the
pyloric sphincter and adjacent to the body.
This document is a draft for review purposes only and does not constitute Agency policy.
1-13 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 evidence of ammonia-induced gastritis or ulceration in rats following 24-weeks of exposure to
2 0.1% ammonia in drinking water.
3 A relationship between ammonia ingestion and gastrointestinal effects is supported by
4 findings from three acute oral studies in rats following gavage administration of ammonium
5 hydroxide [Nagyetal.. 1996: Takeuchietal.. 1995: Murakami etal.. 1990). Takeuchi et al. [1995]
6 reported hemorrhagic necrosis of the gastric mucosa in male Sprague-Dawley rats that received a
7 single gavage dose of ammonium hydroxide (concentration >!%]. Nagyetal. [1996] observed
8 severe hemorrhagic mucosal lesions in female Sprague-Dawley rats 15 minutes after exposure to an
9 estimated dose of 48 mg/kg ammonium hydroxide via gavage. Lesions of the gastric mucosa,
10 including necrosis, were observed in male Sprague-Dawley rats 15 minutes after being given 1 mL
11 of ammonia by intubation at concentrations of 0.5-1%, but not at concentrations of 0.025-0.1%
12 [Murakami etal.. 1990].
13 The evidence of gastrointestinal effects in experimental animals following oral exposure to
14 ammonia is summarized in Table 1-3 and as an exposure-response array in Figure 1-2.
15
This document is a draft for review purposes only and does not constitute Agency policy.
1-14 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Table 1-3. Evidence pertaining to gastrointestinal effects in animals following
oral exposure
Study design and references
Results
Histopathologic changes of the gastric mucosa
Sprague-Dawley rat; male; 6/group
0, 0.01 or 0.1% in drinking water (0, 22, or
220 mg/kg-d)b for 2 or 4 wks
Kawanoetal. (1991)
% change in thickness of mucosa compared to control:
Antrum Body
Wk2: 0, -1, 3%
Wk4: 0,-22,-30*%
Wk2: 0,-5,-20*%
Wk4: 0,-38*,-61*%
Sprague-Dawley rat; male; 36/group
0 or 0.01% in drinking water (0 or 33 mg/kg-
d)c for 3 d or 1, 2, 4, or 8 wks; tap water
provided for the balance of the 8-wk study
Tsujiietal. (1993)
% change in thickness of mucosa compared to control (at d 3, wks 1,
2, 4, and 8):
Antrum Body
D3: 0,8% D3: 0,5%
Wk 1: 0, -4% Wk 1: 0,1%
Wk2: 0,6% Wk2: 0,4%
Wk4: 0,-44%* Wk4: 0,-1%
Wk8: 0,-41%* Wk8: 0,-5%
(extracted from Figure 3 of Tsujii et al., 1993)
Donryu rat; male; 6/group and time point
0, 0.02, or 0.1% in drinking water (0, 28, or
140 mg/kg-d)c for 1, 3, or 5 days and 1, 4, 8,
12, or 24 weeks
Hataetal. (1994)
% change in gland height compared to control (week 24):
Fundic region: 0, -18*, -34*%
Pyloric region: 0,-17*,-26*%
(estimated from Figure 3 of Hata et al., 1994)
% change in labeling index compared to control (week 24):
Fundic region: 0,-35*,-27*%
Pyloric region: 0,-17*,-11*%
a% change compared to control calculated as: (treated value - control value)/control value x 100.
bDoses were estimated based on a body weight of 230 g for male rats and an estimated drinking water intake of
50 mL/d (as reported by study authors).
cDoses were estimated based on an initial body weight of 150 g and an estimated drinking water intake of 50
mL/d (as reported by study authors).
dBody weights and drinking water intakes were not provided by the authors. Doses were estimated assuming a
body weight of 267 g (subchronic value for a male Sprague-Dawley rat, Table 1-2, (U.S. EPA, 1988)) and a drinking
water intake of 37 mL/d (subchronic value for a male Sprague-Dawley rat, Table 1-5, (U.S. EPA, 1988)).
*Statistically significantly different from the control (p < 0.05).
This document is a draft for review purposes only and does not constitute Agency policy,
1-15 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1000
1
2
o
J
4
5
6
7
9
10
11
12
13
14
15
16
17
18
19
20
21
TJ
OuO
.§100
0)
i/i
o
Q
10
T
• LOAEL Vertical lines show
range of doses in
ANOAEL study.
• Additional doses
I
1
thickness of gastric mucosa 4, thickness of gastric mucosa; 4, height of gastric mucosal
(rat); increased cell migration (rat); glands; suppressed cell cycle
Kawanoetal. (1991) Tsujii et al. (1993) (rat);
Hataetal. (1994)
Gastric mucosa
Figure 1-2. Exposure-response array of gastrointestinal effects following oral
exposure.
Mode-of-Action Analysis—Gastrointestinal Effects
The alkalinity of the ammonia solution does not seem to play a direct role in the gastric
effects associated with ammonia. An ammonia solution (pH 10.3) produced dose-related acute
macroscopic mucosal lesions, whereas a glycine-sodium hydroxide buffer (pH 10.3) or ammonium
chloride (pH 4.5) did not [Tsujii etal., 1992b). Rather, the ability of ammonia to damage the gastric
mucosa may be related to its ionization state. Ammonia (NHs) can easily penetrate cell membranes,
subsequently reacting to form NH4+ and OH- in the interior of the membrane [Tsujii etal., 1992b).
The finding that antral and body regions of the rat stomach mucosa responded differently following
administration of 33 mg/kg-day ammonia in drinking water for 8 weeks [Tsujii etal.. 1993) is
consistent with the influence of ionization. The hydrogen chloride secreted by the mucosa in the
body of the stomach resulted in a lower pH in the body mucosa and a corresponding decrease in the
ratio of ammonia to ammonium ion. In contrast, in the antral mucosa (a nonacid-secreting area),
the pH was higher, the ratio of ammonia to ammonium ion was increased, and measures of gastric
mucosal changes were increased compared to those observed in the stomach body where there was
relatively higher exposure to NH4+.
This document is a draft for review purposes only and does not constitute Agency policy,
1-16 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 Several specific events that may contribute to the induction of gastric mucosal changes by
2 ammonia have been proposed. Increased cell vacuolation and decreased viability of cells were
3 associated with increasing ammonia concentration in an in vitro system [Megraudetal.. 1992): the
4 effect was not linked to pH change because of the high buffering properties of the medium. Using
5 an in situ rat stomach model, hemorrhagic mucosal lesions induced by ammonia were associated
6 with the rapid release and activation of cathepsins—which are mammalian cysteine proteases that
7 are released from lysosomes or activated in the cytosol and can be damaging to cells, tissues, or
8 organs [Nagy etal., 1996]. Ammonia also appears to inhibit cellular and mitochondrial respiration,
9 possibly by elevating intracellular or intraorganelle pH or by impairing adenosine triphosphate
10 synthesis [Tsujiietal.. 1992b]. Mori etal. [1998] proposed a role for increased release of
11 endothelin-1 and thyrotropin-releasing hormone from the gastric mucosa in ammonia-induced
12 gastric mucosal injury based on findings in rats given ammonia intragastrically. Although a specific
13 mechanism(s) by which ammonia may induce cellular toxicity has not been established, the
14 available evidence suggests that ammonia-related acceleration of mucosal cell desquamation and
15 stimulation of cell proliferation occurs via a compensatory mechanism [Tsujiietal.. 1992a].
16
17 Summary of Gastrointestinal Effects
18 Evidence that oral exposure to ammonia causes gastrointestinal effects is based on human
19 case reports and studies in rats that focused on mechanistic understandings of effects of ammonia
20 on the gastric mucosa. Acute gastric toxicity observed in case reports involving intentional or
21 accidental ingestion of cleaning solutions or ammonia inhalant capsules appears to reflect the
22 corrosive properties of ammonia. Whether these acute effects are relevant to toxicity following
23 chronic low-level ammonia exposure is not known. Indirect evidence for the biological plausibility
24 of gastric tissue as a target of ammonia toxicity is provided by the association between the
25 bacterium H. pylori, which produces urease that catalyzes urea into ammonia, and human diseases
26 of the upper gastrointestinal tract (including chronic gastritis, gastric ulcers, and stomach cancer).
27 Three mechanistic studies in male rats [Hataetal., 1994: Tsujii etal., 1993: Kawano etal.,
28 1991] provide consistent evidence of changes in the gastric mucosa associated with exposure to
29 ammonia in drinking water, including decreased thickness or gland height These gastric changes
30 did not correlate, however, with other lesions in the stomach; no evidence of other microscopic
31 lesions, gastritis, or ulceration was found in the stomachs of these rats. It is also interesting to note
32 that chronic toxicity studies of other ammonia compounds have not identified the gastrointestinal
33 tract as a target of ammonia toxicity. For example, no treatment-related changes in the stomach or
34 other parts of the gastrointestinal tract were observed in Wistar rats exposed to ammonium
35 chloride in the diet for 130 weeks at doses up to 1,200 mg/kg-day [Lina and Kuijpers. 2004] or in
36 F344 rats exposed to ammonium sulfate for 104 weeks at a dose up to 1,371 mg/kg-day [Ota etal.,
37 2006] (see Appendix B, Table B-l]. Therefore, while drinking water studies with a mechanistic
38 focus provide evidence for ammonia-related changes in rat gastric mucosa, adverse changes of the
39 gastrointestinal tract were not identified in standard toxicity bioassays of ammonia compounds.
This document is a draft for review purposes only and does not constitute Agency policy.
1-17 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Toxicological Review of Ammonia
Mechanistic studies in rodent models support the biological plausibility that ammonia
exposure may be associated with gastric effects in humans. Conditions that favor the un-ionized
form of ammonia (pH>9.25) facilitate penetration of the cell membrane and are associated with
greater gastric cytotoxicity. Given the evidence primarily from human case reports as supported by
mechanistic studies in experimental animals, gastric effects may be a hazard from ammonia
exposure.
1.1.3. Reproductive and Developmental Effects
No changes in reproductive or developmental endpoints were found between two groups of
female pigs (crossbred gilts) exposed to ammonia via inhalation for 6 weeks at mean
concentrations of 5 or 25 mg/m3 and then mated [Diekman et al., 1993] in the only study of the
reproductive and developmental toxicity of ammonia. Age at puberty did not differ significantly
between the two groups. Gilts exposed to 25 mg/m3 ammonia weighed 7% less (p < 0.05) at
puberty than those exposed to 5 mg/m3; however, body weights of the two groups were similar at
gestation day 30. Conception rates in the mated females were similar between the two groups
(94.1 versus 100% in low- versus high-exposure groups). At sacrifice on day 30 of gestation, there
were no significant differences between the two exposed groups in body weights of the pregnant
gilts, number of corpora lutea, number of live fetuses, or weight and length of the fetuses. The
strength of the findings from this study are limited by the absence of a control group and possible
confounding by exposures to bacterial and mycoplasm pathogens. The evidence of reproductive
and developmental effects in experimental animals exposed to ammonia is provided in Table 1-4.
Table 1-4. Evidence pertaining to reproductive and developmental effects in
animals following inhalation exposure
Study Design and Reference
Results
Crossbred gilts (female pigs); 4.5 months old;
40/group
7 ppm (5 mg/m3), range 4-12 ppm (3-8.5 mg/m3) or
35 ppm (25 mg/m3), range 26-45 (18-32 mg/m3) for
6wksa
Diekman et al. (1993)
No change in any of the reproductive or developmental
parameters measured (age at puberty, conception rates,
body weight of pregnant gilts, number of corpora lutea,
number of live fetuses, and weight or length of fetuses).
aA control group was not included. Prior to exposure to ammonia, pigs were also exposed naturally in
conventional grower units to Mycoplasma hypopneumoniae and Pasteurella multocida, which cause pneumonia
and atrophic rhinitis, respectively.
23
This document is a draft for review purposes only and does not constitute Agency policy,
1-18 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 Summary of Reproductive/Developmental Effects
2 No studies of the potential reproductive or developmental toxicity of ammonia in humans
3 are available, and only one animal study that examined the reproductive effects of ammonia in the
4 pig has been conducted. This study did not use a conventional test species and did not include a
5 control group with no ammonia exposure. Further, animals were exposed naturally to bacterial
6 and mycoplasm pathogens.
7 Although the reproductive and developmental toxicity database for ammonia is limited,
8 information on the endogenous formation of ammonia can inform the potential for ammonia to
9 present a reproductive and developmental hazard. Ammonia is endogenously produced in humans
10 and animals during fetal and adult life, and concentrations in blood are homeostatically regulated to
11 remain at low levels. Studies in humans and animals demonstrate that ammonia is present in fetal
12 circulation. In vivo studies in several animal species and in vitro studies of human placenta
13 demonstrate that ammonia is produced within the uteroplacenta and released into the fetal and
14 maternal circulations [Bell etal., 1989: Johnson etal., 1986: Hauguel etal., 1983: Meschiaetal.,
15 1980: Remesar etal.. 1980: Holzmanetal.. 1979: Holzmanetal.. 1977: RubaltelliandFormentin.
16 1968: Luschinsky, 1951]. Tozwiketal. [2005] reported that ammonia levels in human fetal blood
17 (specifically, umbilical arterial and venous blood] at birth were 1.0-1.4 ug/mL, compared to 0.5
18 ug/mL in the mothers' venous blood. Ammonia was also present in human umbilical arterial and
19 venous blood collected at delivery (range of 25-43 weeks of gestation], with umbilical arterial
20 ammonia concentrations significantly higher than venous concentrations (DeSanto etal. (1993]:
21 there was no correlation between umbilical ammonia level and gestational age. In sheep,
22 uteroplacental tissues are a site of ammonia production, with outputs of ammonia into both the
23 uterine and umbilical circulations (Tozwiketal., 1999]. In late-gestation pregnant sheep that were
24 catheterized to allow measurement of ammonia exposure to the fetus, concentrations of ammonia
25 in umbilical arterial and venous blood and uterine arterial and venous blood ranged from
26 approximately 0.39 to 0.60 ug/mL (Tozwiketal.. 2005: Tozwiketal.. 1999]. Thus, the developing
27 fetus and reproductive tissues are normally exposed to ammonia in blood, and external
28 concentrations that do not alter homeostasis would not be expected to pose a developmental or
29 reproductive hazard. Experimental animal data suggest that ammonia exposures below 18 mg/m3
30 will not increase blood ammonia levels (Manninenetal., 1988: Schaerdeletal., 1983: see also
31 Appendix D. 1, Metabolism]; however, information is not available to identify air concentrations of
32 ammonia that could alter homeostasis.
33
34 1.1.4. Immune System Effects
35 A limited number of studies have evaluated the immunotoxicity of ammonia in human
36 populations and in experimental animal models. Immunological function was evaluated in two
37 independent investigations of livestock farmers exposed to ammonia via inhalation;
38 immunoglobulin G- (IgG] and E-specific (IgE] antibodies for pig skin and urine (Crook etal.. 1991].
39 elevated neutrophils from nasal washes, and increased white blood cell counts (Cormier etal.,
This document is a draft for review purposes only and does not constitute Agency policy.
1-19 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 2000] were reported. These data are suggestive of immunostimulatory effects; however, the test
2 subjects were also exposed to a number of other respirable agents in addition to ammonia, such as
3 endotoxin, bacteria, fungi, and mold that are known to stimulate immune responses. Data in
4 humans following exposure to ammonia only are not available.
5 Animal studies that examined ammonia immunotoxicity were conducted using short-term
6 inhalation exposures and were measured by three general types of immune assays, namely host
7 resistance, T cell proliferation, and delayed-type hypersensitivity. Immunotoxicity studies of
8 ammonia using measures of host resistance provide the most relevant data for assessing immune
9 function since they directly measure the immune system's ability to control microorganism growth.
10 Other available studies of ammonia employed assays that evaluated immune function. Changes in
11 immune cell populations without corresponding functional data are considered to be the least
12 predictive, and studies that looked only at these endpoints [Gustinetal.. 1994: Neumann etal..
13 1987] were excluded from the hazard identification for ammonia.
14 Several host resistance studies utilized lung pathogens to assess bacterial clearance
15 following ammonia exposure; however, these studies were not designed to discriminate between
16 direct immunosuppression associated with ammonia exposure or immune effects secondary to
17 damage to the protective mucosal epithelium of the respiratory tract Further, the available studies
18 do not correlate increased bacterial colonization with reduced immune function. Lung lesions, both
19 gross and microscopic, were positively correlated with ammonia concentration in F344 rats
20 continuously exposed to ammonia in an inhalation chamber for 7 days prior to inoculation with
21 Mycoplasma pulmonis (WB colony forming units [CPU]] followed by up to 42 days of ammonia
22 exposure post inoculation [Broderson etal., 1976]. (Inoculation with the respiratory pathogen M.
23 pulmonis causes murine respiratory mycoplasmosis (MRM] characterized by lung lesions.] The
24 incidence of lesions was significantly increased at ammonia concentrations >35 mg/m3, suggesting
25 that ammonia exposure decreased bacterial clearance resulting in the development of M. pulmonis-
26 induced MRM. However, increasing ammonia concentration was not associated with increased CPU
27 of M. pulmonis isolated from the respiratory tract The high number of inoculating CPU could have
28 overwhelmed the innate immune response and elicited a maximal response that could not be
29 further increased in immunocompromised animals.
30 Conversely, significantly increased CPU of M. pulmonis bacteria isolated in the trachea, nasal
31 passages, lungs, and larynx were observed in F344 rats continuously exposed to 71 mg/m3
32 ammonia for 7 days prior to M. pulmonis (104-106 CPU] inoculation and continued for 28 days post
33 inoculation [Schoeb etal.. 1982]. This increase in bacterial colonization indicates a reduction in
34 bacterial clearance following exposure to ammonia. Lesions were not assessed in this study.
35 OF1 mice exposed to 354 mg/m3 ammonia for 7 days prior to inoculation with a 50% lethal
36 dose (LDso] ofPasteurella multocida exhibited significantly increased mortality compared to
37 controls (86% versus 50%, respectively]; however, an 8-hour exposure was insufficient to affect
38 mortality (Richard etal., 1978b]. The authors suggested that the irritating action of ammonia
This document is a draft for review purposes only and does not constitute Agency policy.
1-20 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 destroyed the tracheobronchial mucosa and caused inflammatory lesions thereby increasing
2 sensitivity to respiratory infection with prolonged ammonia exposure.
3 Pig studies support the findings observed in the rodent studies that ammonia exposure
4 increases the colonization of respiratory pathogens. Andreasen et al. [2000] demonstrated that 63
5 days of ammonia exposure increased the number of bacterial positive nasal swabs following
6 inoculation with P. multocida and Mycoplasma hyopneumoniae; however, the effect was not dose
7 responsive and did not result in an increase in pulmonary lesions. Additional data obtained from
8 pigs suggest that ammonia exposure eliminates the commensal flora of the nasal cavities, which
9 allows for increased colonization of P. multocida; however, this effect abates following cessation of
10 ammonia exposure [Hamilton etal.. 1999: Hamilton et al.. 1998).
11 Suppressed cell-mediated immunity and decreased T cell proliferation was observed
12 following ammonia exposure. Using a delayed-type hypersensitivity (DTH) test to evaluate cell-
13 mediated immunity, Hartley guinea pigs were vaccinated with Mycobacterium bovis bacillus
14 Calmette-Guerin (BCG) and exposed to ammonia followed by intradermal challenge with a purified
15 protein derivative (PPD). Dermal lesion size was reduced in animals exposed to 64 mg/m3
16 ammonia indicating immunosuppression [Targowski et al., 1984]. Blood and bronchial
17 lymphocytes harvested from naive guinea pigs treated with the same 3-week ammonia exposure
18 and stimulated with phytohaemagglutinin or concanavalin A demonstrated reduced T cell
19 proliferation [Targowski etal.. 1984). Bactericidal activity in alveolar macrophages isolated from
20 ammonia-exposed guinea pigs was not affected. Lymphocytes and macrophages isolated from
21 unexposed guinea pigs and treated with ammonia in vitro showed reduced proliferation and
22 bactericidal capacity only at concentrations that reduced viability, indicating nonspecific effects of
23 ammonia-induced immunosuppression [Targowski etal., 1984]. These data suggest that! cells
24 may be the target of ammonia since specific macrophage effects were not observed.
25 The evidence of immune system effects in experimental animals exposed to ammonia is
26 summarized in Table 1-5 and as an exposure-response array in Figure 1-3.
27
Table 1-5. Evidence pertaining to immune system effects in animals following
inhalation exposure
Study design and reference
Results
Host resistance
F344 rat; male and female; 11-12/sex/ group
<5 (control), 25, 50, 100, or 250 ppm (<3.5 [control], 18, 35,
71, or 177 mg/m3). 7 d (continuous exposure) pre-
inoculation/28-42 d post-inoculation with M. pulmonis
Broderson etal. (1976)
% of animals with gross lesions: 16 , 46, 66*, 33, and
83%
No effect on CPU.
This document is a draft for review purposes only and does not constitute Agency policy,
1-21 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Table 1-5. Evidence pertaining to immune system effects in animals following
inhalation exposure
Study design and reference
F344 rat; 5-15/group (sex unknown)
<2 or 100 ppm (<1.4 [control] or 71 mg/m3), 7 d
(continuous exposure) pre-inoculation/28 d post-
inoculation with M. pulmonis
Schoebetal. (1982)
OF1 mouse; male; 99/group
0 or 500 ppm (0 or 354 mg/m3), 8 hrs or 7 d (continuous
exposure), prior to infection with P. multocida
Richard etal. (1978b)
Landrace X large white pigs; 10/group (sex unknown)
<5 (control), 50, 100 ppm (3.5, 35, 71 mg/m3), 63 d
(continuous exposure) inoculated with M. hyopneumoniae
on day 9 and P. multocida on d 28, 42, 56
Andreasen et al. (2000)
Large white pigs; 4-7/group (sex unknown)
0 or 20 ppm (0 or 14 mg/m3), 14 d (continuous exposure),
inoculated with P. multocida on d 0
Hamilton etal. (1998)
Large white pigs; 5/group (sex unknown)
0 or 50 ppm (0 or 35 mg/m3), 1 week pre-inoculation with
P. multocida, 3 weeks post-inoculation
Hamilton etal. (1999)
Results
1" bacterial colonization (as a result of reduced
bacterial clearance).
% Mortality: 50 and 86%*
% of animals with positive day 49 nasal swab:
24, 100*, 90%*
/T" bacterial colonization
/T" bacterial colonization
Bacteria isolated from nasal cavities: 3.18 and 4.30*
CPU
T cell proliferation
Hartley guinea pig; 8/group (sex unknown)
<15, 50 or 90 ppm (<11 (control), 35 or 64 mg/m3), 3 wks
(continuous exposure)
Targowski et al. (1984)
•^ proliferation in blood and bronchial T cells.
Delayed-type hypersensitivity
Hartley guinea pig, BCG immunized; 8/group (sex unknown)
<15, 50 or 90 ppm (<11 [control], 35 or 64 mg/m3), 3 wks
(continuous exposure) followed by PPD challenge
Targowski etal. (1984)
Mean diameter of dermal lesion (mm): 12, 12.6 and
8.7*
*Statistically significantly different from the control (p < 0.05).
This document is a draft for review purposes only and does not constitute Agency policy,
1-22 DRAFT—DO NOT CITE OR QUOTE
-------
1000
100 -
o
c
CD
-------
Toxicological Review of Ammonia
1 Summary of Immune System Effects
2 The evidence for ammonia immunotoxicity is based on two epidemiological studies and
3 seven animal studies. Available epidemiological studies that addressed immunological function are
4 confounded by exposures to a number of other respirable agents that have been demonstrated to
5 be immunostimulatory. Single-exposure human studies of ammonia evaluating immune endpoints
6 are not available. Therefore, human studies provide little support for ammonia immunotoxicity.
7 Animal studies provide consistent evidence of elevated bacterial growth following ammonia
8 exposure. This is supported by observations of lung lesions [Brodersonetal., 1976], elevated CPU
9 [Schoeb etal.. 1982]. and increased mortality [Richard etal., 1978b] in rats or mice exposed to
10 ammonia; however, the findings from the Brodersonetal. [1976] study (which described the
11 percent of animals with gross lesions] were not dose-responsive, and the other studies used single
12 concentrations of ammonia and therefore did not provide information on dose-response. One
13 study suggested that T cells are inhibited by ammonia [Targowskietal., 1984], but the data were
14 not dose responsive.
15 Mechanistic data are not available that would support a biologically plausible mechanism
16 for immunosuppression. Because ammonia damages the protective mucosal epithelium of the
17 respiratory tract, it is unclear if elevated bacterial colonization is the result of damage to this
18 barrier or the result of suppressed immunity. Overall, the evidence in humans and animals
19 indicates that ammonia exposure may be associated with these effects, but does not support the
20 immune system as a sensitive target of ammonia toxicity.
21
22 1.1.5. Other Systemic Effects
23 Although the majority of information suggests that ammonia induces effects in and around
24 the portal of entry, there is limited evidence that ammonia can produce effects on organs distal
25 from the portal of entry, including the liver, adrenal gland, kidney, spleen, and heart Alterations in
26 liver function, based on elevated mean levels of aspartate aminotransferase (AST], alanine
27 aminotransferase (ALT], and blood urea, decreased hemoglobin, and inhibition of catalase and
28 monoamine oxidase (MAO] activities, were observed in workers exposed to ammonia over an
29 average exposure duration of 12 years at an Egyptian urea production plant; measurements of
30 workplace exposure concentrations were not provided (Hamid and El-Gazzar. 1996].
31 Evidence of hepatotoxicity in animals comes from observations of histopathological
32 alterations in the liver. Fatty changes in liver plate cells were consistently reported at exposure
33 concentrations >470 mg/m3 ammonia in rats, guinea pigs, rabbits, dogs, and monkeys following
34 identical subchronic inhalation exposure regimens (Coon etal., 1970]. Congestion of the liver was
35 observed in guinea pigs following subchronic and short-term inhalation exposure to 35 and
36 120 mg/m3 (Anderson etal., 1964: Weatherby, 1952]: no liver effects were observed in similarly
37 exposed mice at 14 mg/m3 (Anderson etal.. 1964: Weatherby. 1952].
38 No histopathological or hematological effects were observed in rats, guinea pigs, rabbits,
39 dogs, or monkeys when these animals were repeatedly, but not continuously, exposed to ammonia
40 even at high concentrations (e.g., 770 mg/m3 for 8 hours/day, 5 days/week; see Table 1-8 for
This document is a draft for review purposes only and does not constitute Agency policy.
1-24 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 additional information); suggesting that animals can recover from intermittent exposure to
2 elevated ammonia levels [Coonetal., 1970]. In addition, no effects on organs distal from the
3 respiratory system were observed in mice exposed to 14 mg/m3 for up to 6 weeks [Anderson etal..
4 19641.
5 Adrenal effects were observed in animals following subchronic and short-term exposure to
6 ammonia. Increased mean adrenal weights and fat content of the adrenal gland, as well as
7 histological changes in the adrenal gland (enlarged cells of the zona fasiculata of the adrenal cortex
8 that were rich in lipid), were observed in rabbits exposed via gavage to ammonium hydroxide for
9 durations ranging from 5.5 days to 17 months [Fazekas, 1939]. The strength of these findings is
10 limited by inadequate reporting and study design. A separate study identified early degenerative
11 changes in the adrenal glands of guinea pigs exposed to 120 mg/m3 ammonia by inhalation for 18
12 weeks [Weatherby. 1952]. providing additional limited evidence for effects on the adrenal gland.
13 Evidence that inhaled ammonia can affect the kidney and spleen is limited to studies in
14 experimental animals. Nonspecific degenerative changes in the kidneys (not further described] of
15 rats exposed 262 mg/m3 ammonia were reported (Coonetal., 1970]. Histopathological evaluation
16 of other animal species in the same study exposed to 470 mg/m3, an ammonia concentration that
17 induced a high rate of mortality in rats, consistently showed alterations in the kidneys (calcification
18 and proliferation of tubular epithelium; incidence not reported]. Exposure of guinea pigs to inhaled
19 ammonia at a concentration of 120 mg/m3 for 18 weeks (but not 6 or 12 weeks] resulted in
20 histopathological alterations (congestion] of the kidneys and spleen, although incidence was not
21 reported (Weatherby, 1952]. Enlarged and congested spleens were reported in guinea pigs
22 exposed to 35 mg/m3 ammonia for 6 weeks in a separate study (Anderson etal., 1964].
23 Myocardial fibrosis was observed in monkeys, dogs, rabbits, guinea pigs, and rats following
24 subchronic inhalation exposure to 470 mg/m3 ammonia; no changes were observed at lower
25 concentrations (Coonetal., 1970]. At the same concentration, ocular irritation (characterized as
26 heavy lacrimation, erythema, discharge, and ocular opacity of the cornea] was also reported by
27 Coonetal. (1970] in dogs and rabbits, but was not observed in similarly treated monkeys and rats.
28 Additionally, there is limited evidence of biochemical or metabolic effects of acute or short-
29 term ammonia exposure. Evidence of slight acidosis, as indicated by a decrease in blood pH, was
30 reported in rats exposed to 18 or 212 mg/m3 ammonia for 5 days; study authors stated that
31 differences in pH leveled off at 10 and 15 days (Manninenetal.. 1988]. In another study, blood pH
32 in rats was not affected by exposure to ammonia at concentrations up to 818 mg/m3 for up to 24
33 hours (Schaerdel etal.. 1983]. Oxygen partial pressure (p02] in rats exposed to 11 and 23 mg/m3
34 ammonia were statistically significantly increased, but remained within the normal range; exposure
35 to 219 and 818 mg/m3 over the same time period resulted in no change in p02 (Schaerdel etal.,
36 1983]. No explanation for a change in pC>2 only at the lower exposure concentrations was provided.
37 Encephalopathy related to ammonia may occur following disruption of the body's normal
38 homeostatic regulation of the glutamine and urea cycles resulting in elevated ammonia levels in
39 blood, e.g., as a result of severe liver or kidney disease (Minanaetal., 1995: Souba, 1987]. Acute
40 inhalation exposure studies have identified alterations in amino acid levels and neurotransmitter
This document is a draft for review purposes only and does not constitute Agency policy.
1-25 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
Toxicological Review of Ammonia
metabolism (including glutamine concentrations) in the brain of rats and mice [Manninen and
Savolainen. 1989: Manninen et al., 1988: Sadasivudu etal., 1979: Sadasivudu and Radha Krishna
Murthy. 1978). It has been suggested that glutamate and y-aniino butyric acid play a role in
ammonia-induced neurotoxicity [Tones, 2002]. There is no evidence, however, that ammonia is
neurotoxic in humans or animals following chronic exposures.
The evidence of systemic toxicity in humans and experimental animals exposed to ammonia
is summarized in Tables 1-6 to 1-8 and as an exposure-response array in Figure 1-4.
Table 1-6. Evidence
inhalation exposure
pertaining to other systemic effects in humans following
Study design and reference
Results
Occupational study workers in an Egyptian urea plant;
30 exposed and 30 control subjects
No measurement of exposure concentrations
Average employment time: 12 yrs
Hamid and EI-Gazzar (1996)
AST, ALT, and blood urea in exposed workers;
hemoglobin and inhibition of catalase and MAO.
9
10
Table 1-7. Evidence pertaining to other systemic effects in animals following
oral exposure
Study design and reference
Results
Adrenal effects
Rabbits (strain and sex not specified); 16-33/group
50-80 mL of a 0.5 or 1.0% ammonium hydroxide
solution by gavage; initially every other day, later daily;
duration ranged from 5.5 d to 17 mo; estimated dose:
61-110 mg/kg-d and 120-230 mg/kg-d, respectively3
Fazekas (1939)
Mean adrenal weight — response relative to control:
Fat content of adrenal gland—response relative to
control: 4.5-fold 1\
95%
11
12
aAmmonia doses estimated using assumed average default body weight of 3.5-4.1 kilograms for adult rabbits
(U.S. EPA, 1988).
This document is a draft for review purposes only and does not constitute Agency policy.
1-26 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Table 1-8. Evidence pertaining to other systemic effects in animals following
inhalation exposure
Study design and reference
Results
Liver effects
Guinea pig (strain not specified); male; 6-12/group
0 or 170 ppm (0 or 120 mg/m3) for 6 h/d, 5 d/wk for 6,12 or 18
wks
Weatherby (1952)
Congestion of the liver at 18 wks, not observed
at earlier times.3
Guinea pig (strain not specified); male and female; 2/group
0 or 20 ppm (0 or 14 mg/m3) for 7-42 d or 50 ppm (35 mg/m3
for 42 d
Anderson etal. (1964)
Congestion of the liver at 35 mg/m for 42 d.a
Swiss albino mouse; male and female; 4/group
0 or 20 ppm (0 or 14 mg/m3) for 7-42 d
Anderson etal. (1964)
No visible signs of liver toxicity.
Squirrel monkey (5. sciureus); male; 3/group
Beagle dog; male; 2/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Sprague-Dawley and Long-Evans rat; male and female; 15-
51/group
0, 155, or 770 mg/m3 8 hrs/d, 5 d/wk for 6 weeks
Coon et al. (1970)
No histopathologic changes observed.
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Sprague-Dawley and Long-Evans rat; male and female; 15-
51/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Coon etal. (1970)
Fatty liver changes in plate cells at 470 mg/m .a
Sprague-Dawley or Long-Evans rat; male and female; 15-
51/group
0 or 40 mg/m3 for 114 d or 127, 262 or 470 mg/m3 for 90 d
Coon etal. (1970)
Fatty liver changes in plate cells at
470 mg/m3.a'b
Adrenal gland effects
Guinea pig (strain not specified); male; 6-12/group
0 and 170 ppm (0 and 120 mg/m3) 6 hrs/d, 5 d/wk for 6, 12, or
18 wks
Weatherby (1952)
"Early" degenerative changes in the adrenal
gland (swelling of cells, degeneration of the
cytoplasm with loss of normal granular
structure) at 18 wks, not observed at earlier
times.3
This document is a draft for review purposes only and does not constitute Agency policy.
1-27 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Table 1-8. Evidence pertaining to other systemic effects in animals following
inhalation exposure
Study design and reference
Results
Kidney and spleen effects
Squirrel monkey (5. sciureus); male; 3/group
Beagle dog; male; 2/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Sprague-Dawley and Long-Evans rat; male and female; 15-
51/group
0,155, or 770 mg/m3 8 hrs/d, 5 d/wk for 6 wks
Coon etal. (1970)
No histopathologic changes observed.
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Coon etal. (1970)
Calcification and proliferation of renal tubular
epithelium at 470 mg/m3.3
Sprague-Dawley or Long-Evans rat; male and female; 15-
51/group
0 or 40 mg/m3 for 114 d or 127, 262, or 470 mg/m3 for 90 d
Coon etal. (1970)
Calcification and proliferation of renal tubular
epithelium at 470 mg/m3.a'b
Guinea pig (strain not specified); male; 6-12/group
0 or 170 ppm (0 or 120 mg/m3) 6 hrs/d, 5 d/wk for 6,12, or 18
wks
Weatherby (1952)
Congestion of the spleen and kidneys.3
Guinea pig (strain not specified); male and female; 2/group
0 or 20 ppm (0 or 14 mg/m3) for 7-42 d or 50 ppm (35 mg/m3
for 42 d
Anderson etal. (1964)
Enlarged and congested spleens at 35 mg/m .a
Swiss albino mouse; male and female; 4/group
0 or 20 ppm (0 or 14 mg/m3) for 7-42 d
Anderson etal. (1964)
No visible signs of toxicity.
Myocardial effects
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Sprague-Dawley and Long-Evans rat; male and female; 15-
51/group
0,155, or 770 mg/m3 8 hrs/d, 5 ds/wk for 6 wks
Coon etal. (1970)
No histopathologic changes observed.
This document is a draft for review purposes only and does not constitute Agency policy.
1-28 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Table 1-8. Evidence pertaining to other systemic effects in animals following
inhalation exposure
Study design and reference
Squirrel monkey (5. sciureus); male; 3/group
Beagle dog; male; 2/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Coon etal. (1970)
Sprague-Dawley or Long-Evans rat; male and female; 15-
51/group
0 or 40 mg/m3 for 114 d or 127, 262, or 470 mg/m3 for 90 d
Coon etal. (1970)
Results
Myocardial fibrosis at 470 mg/m3.3
Myocardial fibrosis at 470 mg/m3.a'b
Ocular effects
Beagle dog; male; 2/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Coon etal. (1970)
New Zealand albino rabbit; male; 3/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Coon etal. (1970)
Squirrel monkey (S. sciureus); male; 3/group
Princeton-derived guinea pig; male and female; 15/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Coon etal. (1970)
Sprague-Dawley and Long-Evans rat; male and female; 15-
51/group
0 or 40 mg/m3 for 114 d or 127, 262 or 470 mg/m3 for 90 d
Coon etal. (1970)
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Sprague-Dawley and Long-Evans rat; male and female; 15-
51/group
0, 155, or 770 mg/m3 8 hrs/d, 5 d/wk for 6 wks
Coon etal. (1970)
Heavy lacrimation at 470 mg/m3.3
Erythema, discharge and ocular opacity over %
to % of cornea at 470 mg/m3.3
No ocular irritation observed.
No ocular irritation observed.
No ocular irritation observed.
Blood pH changes
Wistar rat; female; 5/group
0, 25 or 300 ppm (0, 18, or 212 mg/m3) 6 hrs/d for 5, 10 or 15 d
Manninen etal. (1988)
•\, blood pH at 5 days; pH differences "leveled
off at later time points (data not shown)".
Blood pH (day 5): 7 A3, 7.34*, 7.36*
This document is a draft for review purposes only and does not constitute Agency policy.
1-29 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Table 1-8. Evidence pertaining to other systemic effects in animals following
inhalation exposure
Study design and reference
CrhCOBS CD(SD) rat; male; 32 and 70
15, 32, 310, 1157 ppm (11, 23, 219, 818 mg/m3) for 0, 8, 12, 24
hrs, 3 and 7 d
Schaerdeletal. (1983)
Results
-t pO2 at 11 and 23 mg/m3 for 8, 12 and 24 hrs;
no change at higher concentrations; no change
in blood pH.
Percent change in pO2from time 0 (at 24 hours
of exposure)': 20*, 17*, 1, -2%
Amino acid levels and neurotransmitter metabolism in the brain
Wistar rat; female; 5/group
0, 25 or 300 ppm (0, 18, or 212 mg/m3) 6 hrs/d for 5 d
Manninen and Savolainen (1989)
Wistar rat; female; 5/group
0, 25 or 300 ppm (0, 18, or 212 mg/m3) 6 hrs/d for 5, 10 or 15 d
Manninen et al. (1988)
% change compared to control:6
Brain glutamine: 42*, 40*%
% change compared to control at 212 mg/ms:d
Blood glutamine (5, 10, 15 d): 44*, 13, 14%
Brain glutamine (5, 10, 15 d): 40*, 4, 2%
Incidence data not provided.
bExposure to 470 mg/m3 ammonia increased mortality in rats.
Measurements at time zero were used as a control; the study did not include an unexposed control group.
d% change compared to control calculated as: (treated value - control value)/control value x 100.
*Statistically significantly different from the control (p < 0.05).
This document is a draft for review purposes only and does not constitute Agency policy.
1-30 DRAFT—DO NOT CITE OR QUOTE
-------
1000
M
l/l
o
S 100
c
o
u
Ol
in
O
Q.
X
10
1
•;
^
^
^
4
1 •
^
'
)
L
•;
^ 1
;
1 •
' _ L
•;
^ i
•ir
A A •
t
4)
•
• LOAEL Vertical lines show range of
ANOAEL concentrations in study.
A
exposures were intermittent
1& f^- If ^ |
fs 1 2 J.5. f-g |
0.0" 's •— - ' j3 ~ra ra cu * ~t/T
to cu ~Xra cu cu u.Ep"oj
& -c S -^ t t- ra =3 "^ u
bo ^_, .Qoi ojo u bo cn ^
S| So £ £ I*- *
5 tlO t" > ^ T^ ~m O
cu • •• £ cu ' — r; t~ ra .1-1 ^
.> _bO g -a o < ~ >- cu '-
— ' o_ s_S ^ o_tiO£=cu
CU *^ O O O tlO
2 - ifu ^
o .3
Z d O)
0 >
-§ t.
ra
LJ_
Liver effects
o- ^
rv cu
en u ^_^
~ s s
-d -5 en
ra Q. T— 1
cu .E -
:= to ra
O CU 4->
O tlO CU
u. ™ c
-^ r- O
00 u 0
'EL >- u
ra ^ ~
E -IS
• — >, ^_
°? IS
^
_Q
ra
bS
o
-a
ra
c *
S! TN-
~a LH
ra cn
CU T~'
£ "?
changes in
; Weather!
01 "55
•s -ci
ra nj
cu
S 1
CU 3
bo bo
01 -—
•a ~a
>. ^
ra ~bb
LU
Adrenal
gland
effects
A A
6or8hrs/d,5d/wk
ra 1— •— r^- cu *£) to t^
^- JSQ- -sen c?7 ccn tt °1
^
fj
"ra l
o"
^= '
,» ^ra cji -ain j"jH. ^^3
™ -Qai 01^ ^cn _o)~—- tt-
?s-l '
|S ^:
5 _: —
3 ra ra
- CU 14— -;
^ c ^o
5) 0"S c^-atu ^oj
j- ^C CU ^£= QJc
^S"'S° 7^-c a39 "o. o
^rv raO S"-^ bo£ tfl^
TI<:naJt~) CUra £=CU nCU
3 ra ^_ O -—- °O . ^ £= . fc
si ?! ri P u
— O nm t/ini ^^ra _i~i
>• ^2i Q o '~~ _c cu c — — cu ^ c
j ^ (— < i "K 4-1 bo- — Cr— c
tr 4-1 (- **— UJ ,*". bO
o ^ c '{")!_ *"* *~-"^
'u
ra _•
1! u E
- 5
cu
Kidney and spleen effects
bS #
0 —. M
-a 0 o
>:& ^
-Sd £T
c ^ _*
o ra c
E « o
£i E.
Ǥ -g
ra U 0
-C -^ J3
^ ra c
ra *- -=
.y M .-
bo •— -a
O Q- i-
o S S
£ c °
ra '^ ^
°- hD ^
O ^.
tn ."^
'— _Q
_Q
o ra
^ *~
• Additional concentrations
o" -1*
3 lo
T— 1 J— '
— to r^-
'to cn
ra o TH_
QJ -Q ^
j_ «= ra
8 ™ «
(_) "C5 d
~ ra °
f o5
>-
ra "^
cu ^
c
^
bo
0
_Q
Myocardia effects
"5B — ° if ' — T
0 • — • S & "^ ° 1
If cn o^ I — j
.0 ^n u ra -S ^ -i
~ o2 " -a ra
3 S" *
• & ,s!5
] JH. o cn
3 15 "J-
: ^ 1 ra
1! 11 ll 1| I|
§• § .« -j? i V, ^ ~ "S u
ra1-1 ~° si OM ^ra £ -if
01 ra~ J3 E 'a. ^
1 £ -S. .•? s
^ ^ ° M
o
- ,b- cu ra
15
c Q-
o ra
•;-; cu
ra _E
• — 3
•z. Jg
ra
Ocular Irritation
Figure 1-4. Exposure-response array of systemic effects following inhalation exposure.
This document is a draft for review purposes only and does not constitute Agency policy,
1-31
DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 Summary of Other Systemic Effects
2 Effects of ammonia exposure on organs distal from the portal of entry are based on
3 evidence in animals and, to a more limited extent, in humans. One occupational epidemiology study
4 of ammonia-exposed workers reported changes in serum enzymes indicative of altered liver
5 function [Hamid and El-Gazzar. 1996]. Because the study population was small and measurements
6 of workplace ammonia concentrations were not provided, the evidence for liver effects in humans
7 associated with ammonia exposure is weak.
8 Effects on various organs, including liver, adrenal gland, kidney, spleen, and heart, were
9 observed in several studies that examined responses to ammonia exposure in a number of
10 laboratory animal species. While effects on many of these organs were observed in multiple
11 species, including monkey, dog, rabbit, guinea pig, and rat, effects were not consistent across
12 exposure protocols. For example, Coonetal. [1970] reported fatty liver and calcification and
13 proliferation of renal tubular epithelium in monkeys, dogs, rabbits, and guinea pigs exposed
14 continuously to ammonia for 90 days at a concentration of 470 mg/m3, but no histopathological
15 changes in these organs were observed in the same species following intermittent exposure
16 (8 hours/day, 5 days/week for 6 weeks) to concentrations as high as 770 mg/m3. It could be
17 speculated that these differences in response reflect recovery from short-term (i.e., 8-hour)
18 exposures, but the reason for the inconsistent findings is not known.
19 Studies of ammonia toxicity that examined other systemic effects were all published in the
20 older toxicological literature. The only oral study of ammonium hydroxide was published in 1939
21 [Fazekas. 1939). and three subchronic inhalation studies were published between 1952 and 1970
22 [Coonetal.. 1970: Anderson etal., 1964: Weatherby, 1952). In general, the information from these
23 studies is limited by small group sizes, minimal characterization of some of the reported responses
24 (e.g., "congestion," "enlarged," "fatty liver"), insufficiently detailed reporting of study results, and
25 incomplete, if any, incidence data. In addition, Weatherby (1952), Anderson et al. (1964), and some
26 of the experiments reported by Coonetal. (1970) used only one ammonia concentration in addition
27 to the control, so no dose-response information is available from the majority of experimental
28 studies to inform the evidence for systemic effects of ammonia.
29 As discussed in Section 1.1.3, ammonia is endogenously produced in all human and animal
30 tissues, and concentrations in all physiological fluids are homeostatically regulated to remain at low
31 levels (Souba. 1987). Thus, tissues are normally exposed to ammonia, and external concentrations
32 that do not alter homeostasis would not be expected to pose a hazard for systemic effects. Overall,
33 the evidence in humans and animals indicates that ammonia exposure may be associated with
34 effects on organs distal from the portal of entry, but does not support the liver, adrenal gland,
35 kidney, spleen, or heart as sensitive targets of ammonia toxicity.
36
37 1.1.6. Carcinogenicity
38 No information is available regarding the carcinogenic effects of ammonia in humans
39 following oral or inhalation exposure. The carcinogenic potential of ammonia by the inhalation
40 route has not been assessed in animals, and animal carcinogenicity data by the oral route of
This document is a draft for review purposes only and does not constitute Agency policy.
1-32 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 exposure are limited. Toth [1972] concluded that tumor incidence was not increased in Swiss mice
2 exposed for their lifetime (exact exposure duration not specified) to ammonium hydroxide in
3 drinking water at concentrations up to 0.3% (equivalent to 410 and 520 mg/kg-day in female and
4 male mice, respectively) or in C3H mice exposed to ammonium hydroxide in drinking water at a
5 concentration of 0.1% (equivalentto 214 and 191 mg/kg-day in female and male mice,
6 respectively). With the exception of mammary gland tumors in female C3H mice, concurrent
7 control tumor incidence data were not reported and, therefore, comparison of tumor incidence in
8 exposed and control mice could not be performed. The general lack of concurrent control data
9 limits the ability to interpret the findings of this study.
10 The incidence of gastric cancer and the number of gastric tumors per tumor-bearing rat
11 were statistically significantly higher in rats exposed to 0.01% ammonia solution in drinking water
12 (equivalent to 10 mg/kg-day) for 24 weeks following pretreatment (for 24 weeks) with the
13 initiator, N-methyl-N'-nitro-N-nitrosoguanidine (MNNG), compared with rats receiving only MNNG
14 and tap water (Tsujiietal., 1992a). An ammonia-only exposure group was not included in this
15 study. In another study with the same study design, Tsujiietal. (1995) reported similar increases
16 in the incidence of gastric tumors in rats following exposure to MNNG and 10 mg/kg-day ammonia.
17 Additionally, the size and penetration to deeper tissue layers of the MNNG-initiated gastric tumors
18 were enhanced in the rats treated with ammonia (Tsujii etal., 1995). The investigators suggested
19 that ammonia administered in drinking water may act as a cancer promoter (Tsujii etal.. 1995:
20 Tsujii etal..!992a).
21 The evidence of carcinogenicity in experimental animals exposed to ammonia is
22 summarized in Table 1-9.
23
This document is a draft for review purposes only and does not constitute Agency policy.
1-33 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Table 1-9. Evidence pertaining to cancer in animals following oral exposure
Study design and reference
Results
Carcinogenesis studies
Swiss mouse; 50/sex/group
0.1, 0.2, and 0.3% ammonium hydroxide in
drinking water for their lifetime [250, 440, and
520 mg/kg-d (males); 240, 370, and 410 mg/kg-d
(females)]3
Toth(1972)
Tumor incidence not increased in ammonia-exposed mice;
however, concurrent control tumor incidence data were not
reported.
C3H mouse; 40/sex/group
0.1% ammonium hydroxide in drinking water for
their lifetime [191 (males) and 214 mg/kg-d
(females)]"
Toth(1972)
Tumor incidence not increased in ammonia-exposed mice;
however, with the exception of mammary gland tumors in
female mice, concurrent control tumor incidence data were
not reported.
Mammary gland adenocarcinoma: 76, 60%
Initiation-promotion studies
Sprague Dawley rat; male; 40/group
0 or 0.01% ammonia in drinking water (0 or 10
mg/kg-d)c for 24 wks; both groups pretreated for
24 wks with the tumor initiator, MNNG; no
ammonia-only group
Tsujiietal. (1992a)
Gastric tumor incidence: 31, 70*%
# of gastric tumors/tumor-bearing rat: 1.3, 2.1*
Sprague-Dawley rat; male; 43-44/group
0 or 0.01% ammonia in drinking water (0 or 10
mg/kg-d)c for 24 wks; both groups pretreated for
24 wks with the tumor initiator, MNNG; no
ammonia-only group
Tsujiietal. (1995)
Gastric tumor incidence: 30, 66*%
Penetrated muscle layer or deeper: 12, 22*%
Size (mm): 4.4, 5.3*
aAmmonium hydroxide doses estimated based on reported average daily drinking water intakes of 9.2, 8.2, and
6.5 mL/day for males and 8.3, 6.5, and 4.8 mL/day for females in the 0.1, 0.2, and 0.3% groups, respectively,
and assumed average default body weights of 37.3 and 35.3 g for males and females, respectively (U.S. EPA.
1988).
bAmmonium hydroxide doses estimated based on reported average daily drinking water intakes of 7.9 and 8.4
mL/day for males and females, respectively, and assumed average default body weights of 37.3 and 35.3 g for
males and females, respectively (U.S. EPA, 1988).
cAmmonia doses estimated based on reported drinking water intake of 50 mL/day and assumed average default
body weight of 523 g for male Sprague-Dawley rats during chronic exposure (U.S. EPA, 1988).
1
2
3
4
5
6
*Statistically significantly different from the control (p < 0.05).
A limited number of genotoxicity studies are available for ammonia vapor, including one
study in exposed fertilizer factory workers in India that reported chromosomal aberrations and
sister chromatid exchanges in lymphocytes [Yadav and Kaushik, 1997], two studies that found no
evidence of DNA damage in rabbit gastric mucosal or epithelial cell lines [Suzuki etal., 1998: Suzuki
etal.. 1997]. mutation assays in Salmonella typhimurium (not positive] and Escherichia coli
This document is a draft for review purposes only and does not constitute Agency policy.
1-34 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 (positive) [Shimizu etal., 1985: Demerec et al., 1951], a micronucleus assay in mice (positive)
2 (Yadav and Kaushik. 1997). one positive and one negative study in Drosophila melanogaster
3 (Auerbach and Robson. 1947: Lobasov and Smirnov. 1934). and a positive chromosomal aberration
4 test in chick fibroblast cells in vitro (Rosenfeld, 1932] (see Appendix D, Section D.4, Tables D-13
5 and D-14). The finding of chromosomal aberrations and sister chromatid exchanges in human
6 lymphocytes (Yadav and Kaushik. 1997] was difficult to interpret because of the small number of
7 samples and confounding in the worker population by smoking and alcohol consumption. In
8 addition, the levels of ammonia in the plant were low compared to other fertilizer plant studies,
9 raising questions about the study's exposure assessment. Positive findings in in vitro studies with
10 nonhuman cell lines were difficult to interpret because of the presence of a high degree of toxicity
11 (Demerec etal., 1951: Lobasov and Smirnov, 1934] or inadequate reporting (Rosenfeld, 1932]. It is
12 noteworthy that four of the eight available genotoxicity studies were published between 1932 and
13 1951. In two of the more recent studies, ammonia exposure did not induce DNA damage in rabbit
14 gastric mucosal or epithelial cell lines in vitro (Suzuki etal., 1998: Suzuki etal., 1997). Overall, the
15 available genotoxicity literature is inadequate to characterize the genotoxic potential of ammonia.
16
17 1.2. Summary and Evaluation
18 1.2.1. Effects Other than Cancer
19 The respiratory system is the primary and most sensitive target of inhaled ammonia
20 toxicity in humans and experimental animals. Evidence for respiratory system toxicity in
21 humans comes from cross-sectional occupational studies that demonstrated an increased
22 prevalence of respiratory symptoms consistent with irritation and changes in lung function. The
23 findings of respiratory effects in cross-sectional studies of livestock farmers, controlled exposures
24 in volunteers, and case reports of injury following acute exposure provide additional and consistent
25 evidence that the respiratory system is a target of inhaled ammonia. Short-term and subchronic
26 animal studies show respiratory effects in several animal species across different dose regimens.
27 Thus, the weight of evidence of observed respiratory effects seen across multiple human and
28 animal studies identifies respiratory system effects as a hazard from ammonia exposure.
29 Evidence for an association between inhaled ammonia exposure and effects on other organ
30 systems distal from the portal of entry, including the immune system, liver, adrenal gland, kidney,
31 spleen, and heart, is less compelling than for the respiratory system. The two epidemiological
32 studies that addressed immunological function are confounded by exposures to a number of other
33 respirable agents that have been demonstrated to be immunostimulatory and provide little support
34 for ammonia immunotoxicity. Animal studies provide consistent evidence of elevated bacterial
35 growth following ammonia exposure. It is unclear, however, whether elevated bacterial
36 colonization is the result of suppressed immunity or damage to the barrier provided by the mucosal
37 epithelium of the respiratory tract Overall, the weight of evidence does not support the
38 immune system as a sensitive target for ammonia toxicity. Findings from animal studies
39 indicate that ammonia exposure may be associated with effects in the liver, adrenal gland,
This document is a draft for review purposes only and does not constitute Agency policy.
1-35 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 kidney, spleen, and heart; however, the weight of evidence indicates that these organs are
2 not sensitive targets for ammonia toxicity.
3 A limited experimental toxicity database indicates that oral exposure to ammonia may be
4 associated with effects on the stomach mucosa. Increased epithelial cell migration in the antral
5 gastric mucosa leading to a statistically significant decrease in mucosal thickness was reported in
6 male Sprague-Dawley rats exposed to ammonia in drinking water for durations up to 8 weeks
7 [Tsujiietal.. 1993: Kawano etal., 1991]. Similarly, decreases in the height and labeling index of
8 gastric mucosa glands were reported in Donryu rats exposed to ammonia in drinking water for up
9 to 24 weeks [Hataetal., 1994]. The gastric mucosal effects observed in rats were reported to
10 resemble mucosal changes in human atrophic gastritis [Tsujiietal.. 1993: Kawano etal.. 1991]:
11 however, the investigators also reported an absence of microscopic lesions, gastritis, or ulceration
12 in the stomach of these rats. Evidence that oral exposure to ammonia is associated with
13 gastrointestinal effects in humans is limited to case reports of individuals suffering from
14 gastrointestinal effects (e.g., stomach ache, nausea, diarrhea, distress, and burns along the digestive
15 tract] from intentionally or accidentally ingesting household cleaning solutions containing
16 ammonia or biting into capsules of ammonia smelling salts. Mechanistic studies in rodent models
17 support the biological plausibility that ammonia exposure may be associated with gastric effects.
18 Given the weight of evidence from human, animal, and mechanistic studies, gastric effects
19 may be a hazard from ammonia exposure.
20 Studies of the potential reproductive or developmental toxicity of ammonia in humans are
21 not available. No reproductive effects were associated with inhaled ammonia in the only animal
22 study that examined the reproductive effects of ammonia (i.e., a limited-design inhalation study in
23 the pig]. Toxicokinetic information provides support for the conclusion that exposures to
24 ammonia at levels that do not alter homeostasis (i.e., that do not alter normal blood or tissue
25 ammonia levels) would not be expected to pose a developmental or reproductive hazard to
26 the developing fetus and reproductive tissues.
27
28 1.2.2. Carcinogenicity
29 The available information on carcinogenicity following exposure to ammonia is limited to
30 oral animal studies. There was inadequate reporting in studies in Swiss or C3H mice administered
31 ammonium hydroxide in drinking water for a lifetime (Toth. 1972]. There is limited evidence that
32 ammonia administered in drinking water may act as a cancer promoter (Tsujii etal., 1995: Tsujii et
33 al.. 1992a]. The genotoxic potential cannot be characterized based on the available genotoxicity
34 information. Thus, under the Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005a], there is
35 "inadequate information to assess carcinogenic potential" of ammonia.
36
37 1.2.3. Susceptible Populations and Lifestages
38 Studies of the toxicity of ammonia in children or young animals compared to other
39 lifestages that would support an evaluation of childhood susceptibility have not been conducted.
40 Hyperammonemia is a condition of elevated levels of circulating ammonia that can occur in
This document is a draft for review purposes only and does not constitute Agency policy.
1-36 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 individuals with severe diseases of the liver or kidney, organs that biotransform and excrete
2 ammonia, or with hereditary urea cycle disorders [Cordoba etal., 1998: Schubiger etal., 1991:
3 Gilbert. 1988: Teffers etal.. 1988: Souba. 1987). The elevated ammonia levels that accompany
4 human diseases such as acute liver or renal failure can predispose an individual to encephalopathy
5 due to the ability of ammonia to cross the blood-brain barrier; these effects are especially marked
6 in newborn infants [Minanaetal.. 1995: Souba. 1987). Thus, individuals with disease conditions
7 that lead to hyperammonemia may be more susceptible to the effects of ammonia from external
8 sources, but there are no studies that specifically support this hypothesized susceptibility.
9 Because the respiratory system is a target of ammonia toxicity, individuals with respiratory
10 disease (e.g., asthmatics) might be expected to be a susceptible population; however, controlled
11 human studies that examined both healthy volunteers and volunteers with asthma exposed to
12 ammonia, as well as cross-sectional studies of livestock farmers exposed to ammonia [Petrovaetal..
13 2008: Monso etal.. 2004: Sigurdarsonetal.. 2004: Vogelzangetal.. 2000: Vogelzangetal.. 1998:
14 Vogelzangetal.. 1997: Preller etal., 1995], generally did not demonstrate greater respiratory
15 sensitivity after exposure to ammonia in populations with underlying respiratory disease.
16
This document is a draft for review purposes only and does not constitute Agency policy.
1-37 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
2. DOSE-RESPONSE ANALYSIS
4 2.1. Oral Reference Dose for Effects Other than Cancer
5 The RfD (expressed in units of mg/kg-day) is defined as an estimate (with uncertainty
6 spanning perhaps an order of magnitude) of a daily oral exposure to the human population
7 (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects
8 during a lifetime. It can be derived from a no-observed-adverse-effect level (NOAEL), lowest-
9 observed-adverse-effect level (LOAEL), or the 95 percent lower bound on the benchmark dose
10 (BMDL), with uncertainty factors (UFs) generally applied to reflect limitations of the data used.
11 The available data are inadequate to derive an oral RfD for ammonia. Human data involving
12 oral exposure to ammonia are limited to case reports of gastrointestinal effects involving
13 intentional or accidental ingestion of household cleaning solutions or ammonia inhalant capsules.
14 Human data were not considered for derivation of the RfD because, although case reports can
15 indicate the nature of acute endpoints in humans and inform hazard identification, they are
16 inadequate for dose-response analysis and for subsequent derivation of a chronic reference value
17 due to short duration of exposure and incomplete or missing quantitative exposure information.
18 The experimental animal toxicity database for ammonia lacks standard toxicity studies that
19 evaluate a range of tissues/organs and endpoints. Repeat-exposure animal studies of the
20 noncancer effects of ingested ammonia are limited to three studies designed to investigate
21 mechanisms by which ammonia can induce effects on rat gastric mucosa (Hataetal.. 1994: Tsujii et
22 al., 1993: Kawano etal., 1991). While these studies provide consistent evidence of changes in the
23 gastric mucosa associated with exposure to ammonia in drinking water (see Section 1.1.2), the
24 investigators reported no evidence of microscopic lesions of the stomach, gastritis, or ulceration in
25 the stomachs of these rats. In addition, the gastrointestinal tract has not been identified as a target
26 of ammonia toxicity in chronic toxicity studies of ammonium compounds, including ammonium
27 chloride and sulfate (see Section 1.1.2).
28 Given the limited scope of toxicity testing of ingested ammonia and questions concerning
29 the adversity of the gastric mucosal findings in rats, the available oral database for ammonia was
30 considered insufficient to characterize toxicity outcomes and dose-response relationships.
31 Accordingly, an RfD for ammonia was not derived.
32
33 Previous IRIS Assessment: Reference Dose
34 No RfD was derived in the previous IRIS assessment for ammonia.
35
This document is a draft for review purposes only and does not constitute Agency policy.
2-1 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
i 2.2. Inhalation Reference Concentration for Effects Other than Cancer
2 The RfC (expressed in units of mg/m3) is defined as an estimate (with uncertainty spanning
3 perhaps an order of magnitude) of a continuous inhalation exposure to the human population
4 (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects
5 during a lifetime. It can be derived from a NOAEL, LOAEL, or the 95 percent lower bound on the
6 benchmark concentration (BMCL), with UFs generally applied to reflect limitations of the data used.
7
8 2.2.1. Identification of Candidate Principal Studies and Critical Effects
9 Figure 2-1 is an exposure-response array comparing effect levels for inhaled ammonia
10 across a range of toxicological effects. As discussed in Section 1.2, the respiratory system is the
11 primary and most sensitive target of inhaled ammonia toxicity in humans and experimental
12 animals, and respiratory effects have been identified as a hazard following inhalation exposure to
13 ammonia. The experimental toxicology literature for ammonia provides some evidence that
14 inhaled ammonia may be associated with toxicity to target organs other than the respiratory
15 system, including the liver, adrenal gland, kidney, spleen, heart, and immune system. The evidence
16 for these associations is weak; therefore, they were not considered as the basis for RfC derivation.
This document is a draft for review purposes only and does not constitute Agency policy.
2-2 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1000
T
A
A
*-^ *-^
"ro "ro
.° .° — -
°3 ro c ro <-,
. ,£. ~ro
g (D _. in c 4-"
>. TO ^ ^" Q
i? E v o 'c s=
^" •*— - QJ -I— » U fD
o c y, £ § J
.™ -g _E ^" 'oo ce
1/1 § I ^ =j
QJ tjS —
^ oo ->
^
HUMAN STUDIES
I I 1
A A L
I I ....
' 1 ' '
[ ]
^ L
• <
I <
; '
»
» ^
1
\
• LOAEL Vertical lines show
range of concentrations
ANOAEL jnstudy
• Additional concentrations
* exposures were intermittent : 6 hrs/d, 5 d/wk
M—
O TO
•£<=•$ £ °a
4-» O TO u 00 - — .
£ '-i3 — • — . <" '5 I-H
QJ 3 C ° 4= O ^
4-> .y c r-> ^ 1-1 a,
c H= .=J 01 15 ^ ^H
j_ _U — TH to tlO — '
— •*— ' m ,. ~ t~
3 oa E ™ OEM
O w QJ o; bo ^ ^
<*~ '^ ~m C ^ — ^
nT 'c O 'p QJ —
g 1"5 5 ^f ^
& = | 1 &
~ «T
ro r^
M_ ^, Ol
o „ ^
QJ C
^ .2 ro
| H tu
0 M §
~ 1 12
<- — QJ
^ y, -o
0 0
oo co
? i- =
-C ' ^
°i r\i QJ c ro ro
•— LO u . — . o — —
Bo 2 S 5 '^ P «
— — mm 2^— '«
°- "^^ -^ 1
tt i ~ -^ °^3, f
tie ~£ QJ -i-J Q. Q. ' TO
§S C* c 03^"^ T3
u > ™ o ^
i- . u ° °
~ 4-J
5 QJ 3
3 r- n
>-5S -u ™ "i S t-
'Jj '5. ^ 1? ~ -^ (3 ^
i_ "u ^
f " 5
ul 4
EXPERIMENTAL ANIMALSTUDIES
Respiratory effects
Immune
System effects
U
5
^^
. . 5
0 -Q
P*N ro
01 s~
ri ob
o
ro S
ou c
= -B
O CD
o •*-•
U s-
s_
s_
ro
^
u
O
0
r-~
01
T-H
TO
4-J
QJ
C
o
o
u
Other Systemic effects
00
•§100
-------
Toxicological Review of Ammonia
1 Respiratory effects, characterized as increased prevalence of respiratory symptoms and
2 decreased lung function, have been observed in worker populations exposed to ammonia
3 concentrations >18.5 mg/m3 [Rahman etal.. 2007: Alietal.. 2001: Ballaletal.. 1998). Effects,
4 including changes in lung function parameters and increased prevalence of wheezing, chest
5 tightness, and cough/phlegm, have been identified as adverse respiratory health effects by the
6 American Thoracic Society [ATS. 2000] and are similarly noted as adverse in the EPA's Methods for
1 Derivation of Inhalation Reference Concentrations and Application of Inhalation Dosimetry [U.S. EPA,
8 1994]. As shown in Figure 2-1, respiratory effects were also observed in animals, but at
9 concentrations higher than those associated with respiratory effects in humans and in studies
10 involving exposure durations (up to 114 days] shorter than those in occupational studies.
11 Human data are preferred over animal data for deriving reference values when possible
12 because the use of human data is more relevant in the assessment of human health and avoids the
13 uncertainty associated with interspecies extrapolation introduced when animal data serve as the
14 basis for the RfC. In the case of ammonia, the available human occupational studies provide data
15 adequate for quantitative analysis of health outcomes considered relevant to potential general
16 population exposures. Further, human data provide a more sensitive measure of respiratory effects
17 than do data from animal studies. Therefore, data on respiratory effects in humans were
18 considered for derivation of the RfC and the respiratory effects in animals were not further
19 considered.
20 Of the available human data, two occupational studies—Rahman etal. [2007] and Holness
21 etal. [1989]—provide information useful for examining the relationship between chronic ammonia
22 exposure and increased prevalence of respiratory symptoms and decreased lung function. Both
23 studies reported either the presence or absence of respiratory effects in workers exposed to
24 ammonia over a range of concentrations (approximately 4-18 mg/m3]. These studies are coherent,
25 with the NOAEL of 8.8 mg/m3 from the Holness etal. [1989] study falling between the NOAEL and
26 LOAEL values (4.9 and 18.5 mg/m3, respectively] from the Rahman etal. (2007) study. These
27 studies are considered as candidate principal studies for RfC derivation. Other occupational
28 epidemiology studies [Ali etal., 2001: Ballal etal., 1998] did not provide exposure information
29 adequate for dose-response analysis and were thus not considered useful for RfC derivation.
30 Higher confidence is associated with the analytical methods used by Holness etal. [1989]
31 than Rahman etal. [2007]. Rahman et al. [2007] used two analytical methods for measuring
32 ammonia concentrations in workplace air (Drager PAC III and Drager tube); concentrations
33 measured by the two methods differed by four- to fivefold, indicating some uncertainty in these
34 measurements, although ammonia concentrations measured by the two methods were strongly
35 correlated (correlation coefficient of 0.8). In contrast, the Holness etal. [1989] study used an
36 established analytical method for measuring exposure to ammonia recommended by the National
37 Institute for Occupational Safety and Health (NIOSH) that involved the collection of air samples on
38 acid-treated silica gel (ATSG) absorption tubes.
39 In light of the greater confidence in the ammonia measurements in Holness etal. [1989] and
40 considering the range of NOAELs and LOAELs reported in both studies [with a higher NOAEL being
This document is a draft for review purposes only and does not constitute Agency policy.
2-4 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 reported by Holness etal. [1989]], the occupational study of ammonia exposure in workers in a
2 soda ash plant by Holness etal. (1989) was identified as the principal study for RfC derivation
3 and respiratory effects as the critical effect.
4
5 2.2.2. Methods of Analysis
6 The highest occupational exposure in the Holness etal. [1989] study, a NOAEL of 8.8
7 mg/m3, was used as the POD for RfC derivation.
8 Because the RfC is a measure that assumes continuous human exposure over a lifetime, the
9 POD was adjusted to account for the noncontinuous exposure associated with occupational
10 exposure (i.e., 8-hour workday and 5-day workweek]. The duration-adjusted POD was calculated
11 as follows:
12
13 NOAELADj = NOAEL x VEho/VEh x 5 days/7 days
14 = 8.8 mg/m3 x 10 m3/20 m3 x 5 days/7 days
15 = 3.1 mg/m3
16 Where:
17 VEho = human occupational default minute volume (10 m3 breathed during the 8-hour
18 workday, corresponding to a light to moderate activity level] [U.S. EPA, 2011b]
19 VEh = human ambient default minute volume (20 m3 breathed during the entire day].
20
21 2.2.3. Derivation of Reference Concentration
22 Under EPA's Review of the Reference Dose and Reference Concentration Processes [U.S. EPA,
23 2002: Section 4.4.5], also described in the Preamble, five possible areas of uncertainty and
24 variability were considered. A composite UF of 10 was applied to the selected duration-adjusted
25 POD of 3.1 mg/m3 to derive an RfC. An explanation of the five possible areas of uncertainty and
26 variability follows:
27
28 • An intraspecies uncertainty factor, UFH, of 10 was applied to account for potentially
29 susceptible individuals in the absence of data evaluating variability of response to inhaled
30 ammonia in the human population;
31
32 • An interspecies uncertainty factor, UFA, of 1 was applied to account for uncertainty in
33 extrapolating from laboratory animals to humans because the POD was based on human
34 data from an occupational study;
35
36 • A subchronic to chronic uncertainty factor, UFs, of 1 was applied because the occupational
37 exposure period in the principal study [Holness etal., 1989], i.e., mean number of years at
38 present job for exposed workers, of approximately 12 years was considered to be of chronic
39 duration;
40
41 • An uncertainty factor for extrapolation from a LOAEL to a NOAEL, UFu of 1 was applied
42 because a NOAEL was used as the POD; and
43
This document is a draft for review purposes only and does not constitute Agency policy.
2-5 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 • A database uncertainty factor, UFD, of 1 was applied to account for deficiencies in the
2 database. The ammonia inhalation database consists of epidemiological studies and
3 experimental animal studies. The epidemiological studies include industrial worker
4 populations, cross sectional studies in livestock farmers exposed to inhaled ammonia and
5 other airborne agents, controlled exposure studies involving volunteers exposed to
6 ammonia vapors for short periods of time, and a large number of case reports of acute
7 exposure to high ammonia concentrations (e.g., accidental spills/releases) that examined
8 irritation effects, respiratory symptoms, and effects on lung function. Studies of the toxicity
9 of inhaled ammonia in experimental animals include subchronic studies in a number of
10 species, including rats, guinea pigs, and pigs, that examined respiratory and other systemic
11 effects of ammonia, several immunotoxicity studies, and one limited, reproductive toxicity
12 study in young female pigs. (See Chapter 1 for more details regarding available studies.)
13 The database lacks developmental and multigeneration reproductive toxicity studies.
14
15 As noted in EPA's A Review of the Reference Dose and Reference Concentration Processes (U.S.
16 EPA, 2002], "the size of the database factor to be applied will depend on other information
17 in the database and on how much impact the missing data may have on determining the
18 toxicity of a chemical and, consequently, the POD." While the database lacks
19 multigeneration reproductive and developmental toxicity studies, these studies would not
20 be expected to impact the determination of ammonia toxicity at the POD. Therefore, a
21 database UF to account for the lack of these studies is not considered necessary. This
22 determination was based on the observation that ammonia is endogenously produced and
23 homeostatically regulated in humans and animals during fetal and adult life. Uteroplacental
24 tissues produce ammonia, and ammonia concentrations in human umbilical vein and artery
25 blood (at term) of healthy individuals have been shown to be higher than concentrations in
26 maternal blood (Jozwiketal., 2005). Human fetal umbilical blood levels of ammonia at
27 birth were not influenced by gestational age based on deliveries ranging from gestation
28 week 25 to 43 (DeSanto etal. (1993). This evidence provides some assurance that
29 endogenous ammonia concentrations in the fetus are similar to other lifestages, and that
30 baseline ammonia concentrations would not be associated with developmental toxicity.
31 Additionally, evidence in animals (Manninenetal.. 1988: Schaerdeletal.. 1983] suggests
32 that exposure to ammonia at concentrations up to 18 mg/m3 does not alter blood ammonia
33 levels (see Appendix D, Section D.I, for a more detailed discussion of ammonia distribution
34 and elimination). Accordingly, exposure at the duration-adjusted POD (3.1 mg/m3) would
35 not be expected to alter ammonia homeostasis nor result in measureable increases in blood
36 ammonia concentrations. Thus, the concentration of ammonia at the POD for the RfC would
37 not be expected to result in systemic toxicity, including reproductive or developmental
38 toxicity.
39
40 The RfC for ammonia6 was calculated as follows:
41
42 RfC = NOAELADj - UF
43 =3.1 mg/m3 -H 10
44 = 0.31 mg/m3 or 0.3 mg/m3 (rounded to one significant figure)
45
6 Due to uncertainty concerning the possible influence of anions on the toxicity of ammonium, information on
ammonium salts was not used to characterize the effects for ammonia and ammonium hydroxide. Therefore,
the RfC derived in this assessment is applicable to ammonia and ammonium hydroxide, but not ammonium
salts.
This document is a draft for review purposes only and does not constitute Agency policy.
2-6 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 2.2.4. Uncertainties in the Derivation of the Reference Concentration
2 As presented earlier in this section and in the Preamble, EPA standard practices and RfC
3 guidance [U.S. EPA. 2002.1995.1994] were followed in applying an UF approach to a POD (from a
4 NOAEL) to derive the RfC. Specific uncertainties were accounted for by the application of UFs (i.e.,
5 in the case of the ammonia RfC, a factor to address the absence of data to evaluate the variability in
6 response to inhaled ammonia in the human population). The following discussion identifies
7 additional uncertainties associated with the quantification of the RfC for ammonia.
8
9 Use of a NOAEL as a POD
10 Data sets that support BMD modeling are generally preferred for reference value derivation
11 because the shape of the dose-response curve can be taken into account in establishing the POD.
12 For the ammonia RfC, no decreases in lung function or increases in the prevalence of respiratory
13 symptoms were observed in the worker population studied by Holness etal. (1989], i.e., the
14 principal study used to derive the RfC, and as such, the data from this study did not support dose-
15 response modeling. Rather, a NOAEL from the Holness etal. (1989] study was used to estimate the
16 POD. The availability of dose-response data from a study of ammonia, especially in humans, would
17 increase the confidence in the estimation of the POD.
18
19 Endogenous Ammonia
20 Ammonia, which is produced endogenously, has been detected in breath exhaled from the
21 nose and trachea (range: 0.013-0.078 mg/m3] (Smith etal.. 2008: Larson etal.. 1977]. Higher and
22 more variable ammonia concentrations are reported in breath exhaled from the mouth or oral
23 cavity, with the majority of ammonia concentrations from these sources ranging from 0.085 to
24 2.1 mg/m3 fSmith etal.. 2008: Spanel etal.. 2007a. bj Turner etal.. 2006: Diskin etal.. 2003: Smith
25 etal., 1999: Norwood etal., 1992: Larson etal., 1977]. Ammonia in exhaled breath from the mouth
26 or oral cavity is largely attributed to the production of ammonia via bacterial degradation of food
27 protein in the oral cavity or gastrointestinal tract (Turner etal., 2006: Smith etal., 1999: Vollmuth
28 and Schlesinger. 1984]. and can be influenced by factors such as diet, oral hygiene, and age. In
29 contrast, ammonia concentrations measured in breath exhaled from the nose and trachea are lower
30 (range: 0.013-0.078 mg/m3] (Smith etal.. 2008: Larson etal.. 1977] and more likely reflect
31 systemic levels of ammonia (i.e., circulating levels in the blood] (Smith etal.. 2008].
32 Ammonia concentrations measured in breath exhaled from the nose and trachea (i.e.,
33 concentrations expected to more closely correlate with circulating levels of ammonia in blood] are
34 lower than the ammonia RfC of 0.3 mg/m3 by a factor of fourfold or more; however, the RfC does
35 fall within the more variable range of breath concentrations collected from the mouth or oral cavity.
36 Although the RfC falls within the range of breath concentrations collected from the mouth or oral
37 cavity, ammonia in exhaled breath is expected to be rapidly diluted in the much larger volume of
38 ambient air and not contribute significantly to overall ammonia exposure. Further, occupational
39 epidemiology studies served as the basis for the ammonia RfC; the worker populations in these
This document is a draft for review purposes only and does not constitute Agency policy.
2-7 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 studies would have been exposed to any endogenously produced ammonia, and as such the RfC
2 accounts for ammonia exposures from endogenous sources.
o
J
4 2.2.5. Confidence Statement
5 A confidence level of high, medium, or low is assigned to the study used to derive the RfC,
6 the overall database, and the RfC itself, as described in Section 4.3.9.2 of EPA's Methods for
7 Derivation of Inhalation Reference Concentrations and Application of Inhalation Dosimetry [U.S. EPA,
8 1994]. Confidence in the principal study (Holness etal.. 1989) is medium. The design,
9 conduct, and reporting of this occupational exposure study were adequate, but the study was
10 limited by a small sample size and by the fact that workplace ammonia concentrations to which the
11 study population was exposed were below those associated with ammonia-related effects (i.e., only
12 a NOAEL was identified). However, this study is supported in the context of the entire database,
13 which includes the NOAEL and LOAEL values identified in the Rahman etal. [2007] occupational
14 exposure study, other occupational epidemiology studies, multiple studies of acute ammonia
15 exposure in volunteers, and the available inhalation data from animals.
16 Confidence in the database is medium. The inhalation ammonia database includes one
17 study of reproductive toxicity and no studies of developmental toxicity. Normally, confidence in a
18 database lacking these types of studies is considered to be lower due to the uncertainty
19 surrounding the use of any one or several studies to adequately address all potential endpoints
20 following chemical exposure at various critical lifestages. Unless a comprehensive array of
21 endpoints is addressed by the database, there is uncertainty as to whether the critical effect chosen
22 for the RfC derivation is the most sensitive or appropriate. However, reproductive, developmental,
23 and other systemic effects are not expected at the RfC because it is well documented that ammonia
24 is endogenously produced in humans and animals, ammonia concentrations in blood are
25 homeostatically regulated to remain at low levels, and ammonia concentrations in air at the POD
26 are not expected to alter homeostasis. Thus, confidence in the database, in the absence of these
27 types of studies, is medium. Reflecting medium confidence in the principal study and medium
28 confidence in the database, the overall confidence in the RfC is medium.
29
30 2.2.6. Previous IRIS Assessment: Reference Concentration
31 The previous IRIS assessment for ammonia (posted to the database in 1991) presented an
32 RfC of 0.1 mg/m3 based on co-principal studies—the occupational exposure study of workers in a
33 soda ash plant by Holness etal. [1989] and the subchronic study by Brodersonetal. [1976] that
34 examined the effects of ammonia exposure in F344 rats inoculated on day 7 of the study with the
35 bacterium M. pulmonis. The NOAEL of 6.4 mg/m3 (estimated as the mean concentration of the
36 entire exposed group) from the Holness etal. [1989] study (duration adjusted: NOAELADj =
37 2.3 mg/m3) was used as the POD.7
7In this document, the lower bound of the high exposure category from the Holness et al. [1989] study
(8.8 mg/m3, adjusted for continuous exposure to 3.1 mg/m3) was identified as the POD because workers in
This document is a draft for review purposes only and does not constitute Agency policy.
2-8 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1 The previous RfC was derived by dividing the exposure-adjusted POD of 2.3 mg/m3 (from a
2 NOAEL of 6.4 mg/m3) by a composite UF of 30: 10 to account for the protection of sensitive
3 individuals and 3 for database deficiencies to account for the lack of chronic data, the proximity of
4 the LOAEL from the subchronic inhalation study in the rat [Broderson et al., 1976] to the NOAEL,
5 and the lack of reproductive and developmental toxicity studies. A UFo of 3 (rather than 10) was
6 applied because studies in rats (Schaerdel etal.. 1983] showed no increase in blood ammonia levels
7 at an inhalation exposure to 32 ppm (22.6 mg/m3) and only minimal increases at 300-1,000 ppm
8 (212-707 mg/m3), suggesting that no significant distribution is likely to occur at the human
9 equivalent concentration. In this document, a UFD of one was selected because a more thorough
10 investigation of the literature on ammonia homeostasis and literature published since 1991 on
11 fetoplacental ammonia levels provides further support that exposure to ammonia at the POD would
12 not result in a measureable increase in blood ammonia, including fetal blood levels.
13
14 2.3. Cancer Risk Estimates
15 The carcinogenicity assessment provides information on the carcinogenic hazard potential
16 of the substance in question and quantitative estimates of risk from oral and inhalation exposure
17 may be derived. Quantitative risk estimates may be derived from the application of a low-dose
18 extrapolation procedure. If derived, and unless otherwise stated, the oral slope factor is a plausible
19 upper bound on the estimate of risk per mg/kg-day of oral exposure. Similarly, an inhalation unit
20 risk is a plausible upper bound on the estimate of risk per [J.g/m3 air breathed.
21 As discussed in Section 1.2, there is "inadequate information to assess carcinogenic
22 potential" of ammonia. Therefore, a quantitative cancer assessment was not conducted and
23 cancer risk estimates were not derived for ammonia.
24 The previous IRIS assessment also did not include a carcinogenicity assessment
this high exposure category, as well as those in the two lower exposure categories, showed no statistically
significant increase in the prevalence of respiratory symptoms or decreases in lung function.
This document is a draft for review purposes only and does not constitute Agency policy.
2-9 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
REFERENCES
All. BA: Ahmed. HO: Ballal. SG: Albar. AA. (2001). Pulmonary function of workers exposed to ammonia: a
study in the Eastern Province of Saudi Arabia. Int J Occup Environ Health 7:19-22.
Altmann. L: Berresheim. H: Krull. H: Fricke. H: Van Thriel. C: Schaper. M: Merget. R: Bruning. T. (2006). Odor
and sensory irritation of ammonia measured in humans [Abstract]. Naunyn Schmiedebergs Arch
Pharmacol372:99.
Amshel. CE: Fealk. MH: Phillips. Bl: Caruso. DM. (2000). Anhydrous ammonia burns case report and review of
the literature. Burns 26: 493-497.
Anderson. DP: Beard. CW: Hanson. RP. (1964). The adverse effects of ammonia on chickens including
resistance to infection with Newcastle disease virus. Avian Dis 8: 369-379.
Andreasen. M: Baekbo. P: Nielsen. IP. (2000). Lack of effect of aerial ammonia on atrophic rhinitis and
pneumonia induced by Mycoplasma hyopneumoniae and toxigenic Pasteurella multocida. J Vet Med
847:161-171.
ATS (American Thoracic Society). (2000). What constitutes an adverse health effect of air pollution? This
official statement of the American Thoracic Society was adopted by the ATS Board of Directors, July
1999. Am J Respir Grit Care Med 161: 665-673.
ATS PR (Agency for Toxic Substances and Disease Registry). (2004). Toxicological profile for ammonia
[ATSDR Tox Profile]. Atlanta, GA: U.S. Department of Health and Human Services, Public Health
Service, http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=ll&tid=2.
Auerbach. C: Robson. IM. (1947). Tests of chemical substances for mutagenic action. Proc Roy Soc Edinb B
Biol 62: 284-291.
Ballal. SG: Ali. BA: Albar. AA: Ahmed. HO: al-Hasan. AY. (1998). Bronchial asthma in two chemical fertilizer
producing factories in eastern Saudi Arabia. Int J Tuberc Lung Dis 2: 330-335.
Bell. AW: Kennaugh. IM: Battaglia. FC: Meschia. G. (1989). Uptake of amino acids and ammonia at mid-
gestation by the fetal lamb. Q J Exp Physiol 74: 635-643.
Broderson. IR: Lindsey. IR: Crawford. IE. (1976). The role of environmental ammonia in respiratory
mycoplasmosis of rats. Am J Pathol 85:115-130.
CDC (Centers for Disease Control and Prevention). (2004). The health consequences of smoking: A report of
the Surgeon General. Washington, DC: U.S. Department of Health and Human Services.
http://www.surgeongeneral.gov/librarv/smokingconsequences/.
Choudat. D: Goehen. M: Korobaeff. M: Boulet. A: Dewitte. ID: Martin. MH. (1994). Respiratory symptoms and
bronchial reactivity among pig and dairy farmers. Scand J Work Environ Health 20: 48-54.
Christesen. HB. (1995). Prediction of complications following caustic ingestion in adults. Clin Otolaryngol
Allied Sci 20: 272-278.
Cole. TI: Cotes. IE: lohnson. GR: Martin Hde. V: Reed. IW: Saunders. Ml. (1977). Ventilation, cardiac frequency
and pattern of breathing during exercise in men exposed to 0-chlorobenzylidene malononitrile (CS)
and ammonia gas in low concentrations. Q J Exp Physiol 62: 341-351.
Coon. RA: lones. RA: lenkins. LI. Ir: Siegel. I. (1970). Animal inhalation studies on ammonia, ethylene glycol,
formaldehyde, dimethylamine, and ethanol. Toxicol Appl Pharmacol 16: 646-655.
Cordoba. I: Gottstein. I: Blei. AT. (1998). Chronic hyponatremia exacerbates ammonia-induced brain edema in
rats after portacaval anastomosis. J Hepatol 29: 589-594.
Cormier. Y: Israel-Assayag. E: Racine. G: Duchaine. C. (2000). Farming practices and the respiratory health
risks of swine confinement buildings. Eur Respir J15: 560-565.
Crook. B: Robertson. IF: Glass. SA: Botheroyd. EM: Lacey. I: Topping. MD. (1991). Airborne dust, ammonia,
microorganisms, and antigens in pig confinement houses and the respiratory health of exposed farm
workers. Am Ind Hyg Assoc J 52: 271-279. http://dx.doi.org/10.1080/15298669191364721.
Curtis. SE: Anderson. CR: Simon. I: lensen. AH: Day. PL: Kelley. KW. (1975). Effects of aerial ammonia,
hydrogen sulfide and swine-house dust on rate of gain and respiratory-tract structure in swine. J
Anim Sci 41: 735-739.
This document is a draft for review purposes only and does not constitute Agency policy.
R-l DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Dean. IA. (1985). Lange's handbook of chemistry. New York, NY: McGraw-Hill.
Demerec. M: Bertani. G: Flint. I. (1951). A survey of chemicals for mutagenic action on E coli. Am Nat 85:119-
135.
DeSanto. IT: Nagomi. W: Liechty. EA: Lemons. IA. (1993). Blood ammonia concentration in cord blood during
pregnancy. Early Hum Dev 33:1-8.
Diekman. MA: Scheldt. AB: Sutton. AL: Green. ML: Clapper. IA: Kelly. DT: Van Alstine. WG. (1993). Growth and
reproductive performance, during exposure to ammonia, of gilts afflicted with pneumonia and
atrophic rhinitis. Am J Vet Res 54: 2128-2131.
Diskin. AM: Spanel. P: Smith. D. (2003). Time variation of ammonia, acetone, isoprene and ethanol in breath: a
quantitative SIFT-MS study over 30 days. Physiol Meas 24:107-119.
Doig. PA: Willoughby. RA. (1971). Response of swine to atmospheric ammonia and organic dust. J Am Vet Med
Assoc 159: 1353-1361.
Done. SH: Chennells. PI: Gresham. AC: Williamson. S: Hunt. B: Taylor. LL: Bland. V: lones. P: Armstrong. D:
White. RP: Demmers. TG: Teer. N: Wathes. CM. (2005). Clinical and pathological responses of weaned
pigs to atmospheric ammonia and dust. Vet Rec 157: 71-80.
Donham. Kl: Reynolds. SI: Whitten. P: Merchant. IA: Burmeister. L: Popendorf. Wl. (1995). Respiratory
dysfunction in swine production facility workers: dose-response relationships of environmental
exposures and pulmonary function. Am J Ind Med 27: 405-418.
Donham. Kl: Cumro. D: Reynolds. SI: Merchant. IA. (2000). Dose-response relationships between occupational
aerosol exposures and cross-shift declines of lung function in poultry workers: recommendations for
exposure limits. J Occup Environ Med 42: 260-269.
Douglas. RB: Coe. IE. (1987). The relative sensitivity of the human eye and lung to irritant gases. Ann Occup
Hyg 31: 265-267.
Dworkin. MS: Patel. A: Fennell. M: Vollmer. M: Bailey. S: Bloom. I: Mudahar. K: Lucht. R. (2004). An outbreak of
ammonia poisoning from chicken tenders served in a school lunch. J Food Prot 67:1299-1302.
Eggeman. T. (2001). Ammonia. In Kirk-Othmer encyclopedia of chemical technology. New York, NY: John
Wiley & Sons.
Fazekas. IG. (1939). Experimental suprarenal hypertrophy induced by ammonia. Endokrinologie 21: 315-337.
Ferguson. WS: Koch. WC: Webster. LB: Gould. IR. (1977). Human physiological response and adaption to
ammonia. J Occup Environ Med 19: 319-326.
Gaafar. H: Girgis. R: Hussein. M: el-Nemr. F. (1992). The effect of ammonia on the respiratory nasal mucosa of
mice. A histological and histochemical study. Acta Otolaryngol 112: 339-342.
Gilbert. Gl. (1988). Acute ammonia intoxication 37 years after ureterosigmoidostomy. South Med J 81:1443-
1445.
Gustin. P: Urbain. B: Prouvost. IF: Ansay. M. (1994). Effects of atmospheric ammonia on pulmonary
hemodynamics and vascular permeability in pigs: interaction with endotoxins. Toxicol Appl
Pharmacol 125:17-26. http://dx.doi.org/10.1006/taap.1994.1044.
Guyatt. GH: Oxman. AD: Vist. GE: Kunz. R: Falck-Ytter. Y: Alonso-Coello. P: Schimemann. HI. (2008a). GRADE:
An emerging consensus on rating quality of evidence and strength of recommendations. BMJ 336:
924-926. http://dx.doi.org/10.1136/bmi.39489.470347.AD.
Guyatt. GH: Oxman. AD: Kunz. R: Vist. GE: Falck-Ytter. Y: Schimemann. HI. (2008b). GRADE: What is "quality of
evidence" and why is it important to clinicians? BMJ 336: 995-998.
http://dx.doi.org/10.1136/bmj.39490.551019.BE.
Hamid. HA: El-Gazzar. RM. (1996). Effect of occupational exposure to ammonia on enzymatic activities of
catalase and mono amine oxidase. J Egypt Public Health Assoc 71: 465-475.
Hamilton. TD: Roe. IM: Hayes. CM: Webster. Al. (1998). Effects of ammonia inhalation and acetic acid
pretreatment on colonization kinetics of toxigenic Pasteurella multocida within upper respiratory
tracts of swine. J Clin Microbiol 36:1260-1265.
Hamilton. TD: Roe. IM: Hayes. CM: lones. P: Pearson. GR: Webster. Al. (1999). Contributory and exacerbating
roles of gaseous ammonia and organic dust in the etiology of atrophic rhinitis. Clin Diagn Lab
Immunol 6:199-203.
Hata. M: Yamazaki. Y: Ueda. T: Kato. T: Kohli. Y: Fujiki. N. (1994). Influence of ammonia solution on gastric
mucosa and acetic acid induced ulcer in rats. Eur I Histochem 38: 41-52.
This document is a draft for review purposes only and does not constitute Agency policy.
R-2 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Hauguel. S: Challier. 1C: Cedard. L: Olive. G. (1983). Metabolism of the human placenta perfused in vitro:
glucose transfer and utilization, 02 consumption, lactate and ammonia production. Pediatr Res 17:
729-732.
Heederik. D: van Zwieten. R: Brouwer. R. (1990). Across-shift lung function changes among pig farmers. Am J
IndMed 17: 57-58.
Hill, AB. (1965). The environment and disease: Association or causation? Proc R Soc Med 58: 295-300.
Holness. PL: Purdham. IT: Nethercott. IR. (1989). Acute and chronic respiratory effects of occupational
exposure to ammonia. AIHA J 50: 646-650. http://dx.doi.org/10.1080/15298668991375308.
Holzman. IR: Lemons. IA: Meschia. G: Battaglia. FC. (1977). Ammonia production by the pregnant uterus
(39868). Proc Soc Exp Biol Med 156: 27-30.
Holzman. IR: Philipps. AF: Battaglia. FC. (1979). Glucose metabolism, lactate, and ammonia production by the
human placenta in vitro. Pediatr Res 13:117-120.
HSDB (Hazardous Substances Data Bank). (2012). Ammonia. Washington, DC: National Library of Medicine.
IARC (International Agency for Research on Cancer). (2006). Preamble to the IARC monographs. Lyon,
France. http://monographs.iarc.fr/ENG/Preamble/.
Ihrig. A: Hoffmann. I: Triebig. G. (2006). Examination of the influence of personal traits and habituation on the
reporting of complaints at experimental exposure to ammonia. Int Arch Occup Environ Health 79:
332-338. http://dx.doi.org/10.1007/s00420-005-0042-y.
IPCS (International Programme on Chemical Safety). (1986). Environmental health criteria: Ammonia. (EHC
54). Geneva, Switzerland: World Health Organization.
http://www.inchem.org/documents/ehc/ehc/ehc54.htm.
larudi. NI: Golden. B. (1973). Ammonia eye injuries. J Iowa Med Soc 63: 260-263.
leffers. LI: Dubow. RA: Zieve. L: Reddy. KR: Livingstone. AS: Neimark. S: Viamonte. M: Schiff. ER. (1988).
Hepatic encephalopathy and orotic aciduria associated with hepatocellular carcinoma in a
noncirrhotic liver. Hepatology 8: 78-81.
lohnson. PR: Bedick. CR: Clark. NN: McKain. PL. (2009). Design and testing of an independently controlled
urea SCR retrofit system for the reduction of NOx emissions from marine diesels. Environ Sci Technol
43: 3959-3963. http://dx.doi.org/10.1021/es900269p.
lohnson. RL: Gilbert. M: Block. SM: Battaglia. FC. (1986). Uterine metabolism of the pregnant rabbit under
chronic steady-state conditions. Am J Obstet Gynecol 154:1146-1151.
lones. EA. (2002). Ammonia, the GABA neurotransmitter system, and hepatic encephalopathy. Metab Brain
Dis 17: 275-281.
lozwik. M: Teng. C: Meschia. G: Battaglia. FC. (1999). Contribution of branched-chain amino acids to
uteroplacental ammonia production in sheep. Biol Reprod 61: 792-796.
lozwik. M: Pietrzycki. B: Chojnowski. M: Teng. C: Battaglia. FC. (2005). Maternal and fetal blood ammonia
concentrations in normal term human pregnancies. Biol Neonate 87: 38-43.
http://dx.doi.org/10.1159/000081702.
Kawano. S: Tsujii. M: Fusamoto. H: Sato. N: Kamada. T. (1991). Chronic effect of intragastric ammonia on
gastric mucosal structures in rats. Dig Dis Sci 36: 33-38.
Klein. I: Olson. KR: McKinney. HE. (1985). Caustic injury from household ammonia. Am J Emerg Med 3: 320.
Klendshoj. NC: Rejent. TA. (1966). Tissue levels of some poisoning agents less frequently encountered. J
Forensic Sci 11: 75-80.
Larson. TV: Covert. PS: Frank. R: Charlson. Rl. (1977). Ammonia in the human airways: Neutralization of
inspired acid sulfate aerosols. Science 197:161-163.
Lide. PR (Ed.). (2008). CRC handbook of chemistry and physics (88th ed.). Boca Raton, FL: CRC Press.
Lina, BAR: Kuijpers, MHM. (2004). Toxicity and carcinogenicity of acidogenic or alkalogenic diets in rats;
effects of feeding NH(4)C1, KHCO(3) or KC1. Food Chem Toxicol 42:135-153.
Lobasov. M: Smirnov. F. (1934). Nature of the action of chemical agents on mutational process in Drosophila
melanogaster: II. The effect of ammonia on the occurrence of lethal transgenations. C R Biol 3:174-
176.
Lopez. GP: Dean. BS: Krenzelok. EP. (1988). Oral-exposure to ammonia inhalants: A report of 8 cases
[Abstract]. Vet Hum Toxicol 30: 350.
Luschinsky. HL. (1951). The activity of glutaminase in the human placenta. Arch Biochem Biophys 31: 132-
140.
This document is a draft for review purposes only and does not constitute Agency policy.
R-3 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
MacEwen. ID: Theodore. I: Vernot. EH. (1970). Human exposure to EEL concentrations of
monomethylhydrazine. In Proceedings of the first annual conference on environmental toxicology.
Wright-Patterson Air Force Base, OH: Aerospace Medical Research Laboratory.
Manninen. A: Anttila. S: Savolainen. H. (1988). Rat metabolic adaptation to ammonia inhalation. Proc Soc Exp
BiolMed 187: 278-281.
Manninen. ATA: Savolainen. H. (1989). Effect of short-term ammonia inhalation on selected amino acids in rat
brain. Pharmacol Toxicol 64: 244-246.
Megraud. F: Neman-Simha. V: Briigmann. D. (1992). Further evidence of the toxic effect of ammonia produced
by Helicobacter pylori urease on human epithelial cells. Infect Immun 60:1858-1863.
Melbostad. E: Eduard. W. (2001). Organic dust-related respiratory and eye irritation in Norwegian farmers.
Am JIndMed 39: 209-217.
Meschia. G: Battaglia. FC: Hay. WW: Sparks. IW. (1980). Utilization of substrates by the ovine placenta in vivo.
Fed Proc 39: 245-249.
Millea. TP: Kucan. 10: Smoot. EC. (1989). Anhydrous ammonia injuries. Journal of Burn Care and
Rehabilitation 10: 448-453.
Minana. MD: Marcaida. G: Grisolia. S: Felipo. V. (1995). Prenatal exposure of rats to ammonia impairs NMDA
receptor function and affords delayed protection against ammonia toxicity and glutamate
neurotoxicity. J Neuropathol Exp Neurol 54: 644-650.
Monso. E: Riu. E: Radon. K: Magarolas. R: Danuser. B: Iversen. M: Morera. I: Nowak. D. (2004). Chronic
obstructive pulmonary disease in never-smoking animal farmers working inside confinement
buildings. Am J Ind Med 46: 357-362. http://dx.doi.org/10.1002/ajim.20077.
Mori. S: Kaneko. H: Mitsuma. T: Hayakawa. T: Yamaguchi. C: Uruma. M. (1998). Implications of gastric topical
bioactive peptides in ammonia-induced acute gastric mucosal lesions in rats. Scand J Gastroenterol
33: 386-393.
Murakami. M: Yoo. IK: Teramura. S: Yamamoto. K: Saita. H: Matuo. K: Asada. T: Kita. T. (1990). Generation of
ammonia and mucosal lesion formation following hydrolysis of urea by urease in the rat stomach. J
Clin Gastroenterol 12: S104-S109.
Nagy. L: Kusstatscher. S: Hauschka. PV: Szabo. S. (1996). Role of cysteine proteases and protease inhibitors in
gastric mucosal damage induced by ethanol or ammonia in the rat. J Clin Invest 98:1047-1054.
http://dx.doi.org/10.1172/ICI118865.
Nelson. PL: Cox. MM. (2008). Amino acid oxidation and the production of urea. In Lehninger principles of
biochemistry (5th ed.). New York, NY: W.H. Freeman & Co.
Neumann. R: Mehlhorn. G: Leonhardt. W: Kastner. P: Willig. R: Schimmel. D: lohannsen. U. (1987).
[Experimental studies on the effect on unweaned pigs of aerogenous toxic gas stress, using ammonia
of differing concentrations. 1. Clinical picture of ammonia-exposed piglets under the influence of
Pasteurella multocida infection, with and without exposure to thermo-motor stress]. J Vet Med B 34:
183-196. http://dx.doi.0rg/10.llll/j.1439-0450.1987.tb00386.x.
Norwood. DM: Wainman. T: Lioy. PI: Waldman. IM. (1992). Breath ammonia depletion and its relevance to
acidic aerosol exposure studies. Arch Environ Occup Health 47: 309-313.
http://dx.doi.org/10.1080/00039896.1992.9938367.
O'Neil. Ml: Heckelman. PE: Koch. CB: Roman. Kl. (2006). Ammonia. In The Merck index: An encyclopedia of
chemicals, drugs, and biologicals (14th ed.). Whitehouse Station, NJ: Merck & Co., Inc.
Ota. Y: Hasumura. M: Okamura. M: Takahashi. A: Ueda. M: Onodera. H: Imai. T: Mitsumori. K: Hirose. M.
(2006). Chronic toxicity and carcinogenicity of dietary administered ammonium sulfate in F344 rats.
Food Chem Toxicol 44: 17-27. http://dx.doi.Org/10.1016/j.fct.2005.06.001.
Petrova, M: Diamond, I: Schuster, B: Dalton, P. (2008). Evaluation of trigeminal sensitivity to ammonia in
asthmatics and healthy human volunteers. Inhal Toxicol 20:1085-1092.
http://dx.doi.org/10.1080/08958370802120396.
Preller. L: Heederik. D: Boleij. ISM: Vogelzang. PFI: Tielen. MIM. (1995). Lung function and chronic respiratory
symptoms of pig farmers: Focus on exposure to endotoxins and ammonia and use of disinfectants.
Occup Environ Med 52: 654-660.
Rahman. MH: Bratveit. M: Moen. BE. (2007). Exposure to ammonia and acute respiratory effects in a urea
fertilizer factory. Int J Occup Environ Health 13: 153-159.
Remesar. X: Arola. L: Palou. A: Alemany. M. (1980). Activities of enzymes involved in amino-acid metabolism
in developing rat placenta. Eur J Biochem 110: 289-293.
This document is a draft for review purposes only and does not constitute Agency policy.
R-4 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Reynolds. SI: Donham. Kl: Whitten. P: Merchant. I A: Burmeister. LF: Popendorf. Wl. (1996). Longitudinal
evaluation of dose-response relationships for environmental exposures and pulmonary function in
swine production workers. Am J Ind Med 29: 33-40. http://dx.doi.org/10.1002/(SICi)1097-
0274ri99601129:l<:33::AID-AIIM5>:3.0.CO:2-#.
Richard, D; Jouany, JM; Boudene, C. (1978a). [Acute inhalation toxicity of ammonia in rabbits]. C R Acad Sci
Hebd Seances Acad Sci D 287: 375-378.
Richard. D: Bouley. G: Boudene. C. (1978b). [Effects of ammonia gas continuously inhaled by rats and mice].
Bull Europ Physiol Resp 14: 573-582.
Rosenbaum. AM: Walner. PL: Dunham. ME: Holinger. LD. (1998). Ammonia capsule ingestion causing upper
aerodigestive tract injury. Otolaryngol Head Neck Surg 119: 678-680.
Rosenfeld. M. (1932). [Experimental modification of mitosis by ammonia]. Archiv fur experimentelle
Zellforschung, besonders Gewebeziichtung 14:1-13.
Rosswall. T. (1981). The biogeochemical nitrogen cycle. In Some perspectives of the major biogeochemical
cycles. Chichester, England: John Wiley & Sons.
Rothman. Kl: Greenland. S. (1998). Modern epidemiology (2nd ed.). Philadelphia, PA: Lippincott, Williams, &
Wilkins.
Rubaltelli. FF: Formentin. PA. (1968). Ammonia nitrogen, urea and uric acid blood levels in the mother and in
both umbilical vessels at delivery. Biol Neonat 13:147-154. http://dx.doi.org/10.1159/000240142.
Sadasivudu. B: Radha Krishna Murthy. C. (1978). Effects of ammonia on monoamine oxidase and enzymes of
GABA metabolism in mouse brain. Arch Physiol Biochem 86: 67-82.
Sadasivudu. B: Rao. TI: Murthy. CR. (1979). Chronic metabolic effects of ammonia in mouse brain. Arch
Physiol Biochem 87: 871-885.
Schaerdel. AD: White. Wl: Lang. CM: Dvorchik. BH: Bohner. K. (1983). Localized and systemic effects of
environmental ammonia in rats. J Am Assoc Lab Anim Sci 33: 40-45.
Schoeb. TR: Davidson. MK: Lindsey. IR. (1982). Intracage ammonia promotes growth of Mycoplasma
pulmonis in the respiratory tract of rats. Infect Immun 38: 212-217.
Schubiger. G: Bachmann. C: Barben. P: Colombo. IP: Tonz. 0: Schiipbach. D. (1991). N-acetylglutamate
synthetase deficiency: diagnosis, management and follow-up of a rare disorder of ammonia
detoxication. Eur J Pediatr 150: 353-356.
Shimizu. H: Suzuki. Y: Takemura. N: Goto. S: Matsushita. H. (1985). The results of microbial mutation test for
forty-three industrial chemicals. Sangyo Igaku 27: 400-419.
Sigurdarson. ST: O'Shaughnessy. PT: Watt. IA: Kline. IN. (2004). Experimental human exposure to inhaled
grain dust and ammonia: towards a model of concentrated animal feeding operations. Am J Ind Med
46: 345-348. http://dx.doi.org/10.1002/ajim.20055.
Silverman. L: Whittenberger. IL: Muller. I. (1949). Physiological response of man to ammonia in low
concentrations. J Ind HygToxicol 31: 74-78.
Smeets. MA: Bulsing. PI: van Rooden. S: Steinmann. R: de Ru. IA: Ogink. NW: van Thriel. C: Dalton. PH. (2007).
Odor and irritation thresholds for ammonia: a comparison between static and dynamic olfactometry.
Chem Senses 32:11-20. http://dx.doi.org/10.1093/chemse/bjl031.
Smith. D: Spanel. P: Davies. S. (1999). Trace gases in breath of healthy volunteers when fasting and after a
protein-calorie meal: a preliminary study. J Appl Physiol 87:1584-1588.
Smith. D: Wang. T: Pysanenko. A: Spanel. P. (2008). A selected ion flow tube mass spectrometry study of
ammonia in mouth- and nose-exhaled breath and in the oral cavity. Rapid Commun Mass Spectrom
22: 783-789. http://dx.doi.org/10.1002/rcm.3434.
Socolow. RH. (1999). Nitrogen management and the future of food: lessons from the management of energy
and carbon. PNAS 96: 6001-6008.
Souba. WW. (1987). Interorgan ammonia metabolism in health and disease: a surgeon's view. JPEN J Parenter
Enteral Nutr 11: 569-579.
Spanel. P: Dryahina. K: Smith. D. (2007a). Acetone, ammonia and hydrogen cyanide in exhaled breath of
several volunteers aged 4-83 years. J Breath Res 1: 011001. http://dx.doi.org/10.1088/1752-
7155/1/1/011001.
Spanel. P: Dryahina. K: Smith. D. (2007b). The concentration distributions of some metabolites in the exhaled
breath of young adults. J Breath Res 1: 026001. http://dx.doi.Org/10.1088/1752-7155/l/2/026001.
Stombaugh. DP: Teague. HS: Roller. WL. (1969). Effects of atmospheric ammonia on the pig. J Anim Sci 28:
844-847.
This document is a draft for review purposes only and does not constitute Agency policy.
R-5 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
Sundblad. BM: Larsson. BM: Acevedo. F: Ernstgard. L: lohanson. G: Larsson. K: Palmberg. L. (2004). Acute
respiratory effects of exposure to ammonia on healthy persons. Scand J Work Environ Health 30:
313-321.
Suzuki. H: Mori. M: Suzuki. M: Sakurai. K: Miura. S: Ishii. H. (1997). Extensive DNA damage induced by
monochloramine in gastric cells. Cancer Lett 115: 243-248.
Suzuki. H: Seto. K: Mori. M: Suzuki. M: Miura. S: Ishii. H. (1998). Monochloramine induced DNA fragmentation
in gastric cell line MKN45. Am J Physiol 275: G712-G716.
Takeuchi. K: Ohuchi. T: Harada. H: Okabe. S. (1995). Irritant and protective action of urea-urease ammonia in
rat gastric mucosa. Different effects of ammonia and ammonium ion. Dig Dis Sci 40: 274-281.
Targowski. SP: Klucinski. W: Babiker. S: Nonnecke. Bl. (1984). Effect of ammonia on in vivo and in vitro
immune responses. Infect Immun 43: 289-293.
Toth. B. (1972). Hydrazine, methylhydrazine and methylhydrazine sulfate carcinogenesis in Swiss mice.
Failure of ammonium hydroxide to interfere in the development of tumors. Int J Cancer 9:109-118.
Tsujii. M: Kawano. S: Tsuji. S: Nagano. K: Ito. T: Hayashi. N: Fusamoto. H: Kamada. T: Tamura. K. (1992a).
Ammonia: a possible promotor in Helicobacter pylori-related gastric carcinogenesis. Cancer Lett 65:
15-18.
Tsujii. M: Kawano. S: Tsuji. S: Fusamoto. H: Kamada. T: Sato. N. (1992b). Mechanism of gastric mucosal
damage induced by ammonia. Gastroenterology 102:1881-1888.
Tsujii. M: Kawano. S: Tsuji. S: Ito. T: Nagano. K: Sasaki. Y: Hayashi. N: Fusamoto. H: Kamada. T. (1993). Cell
kinetics of mucosal atrophy in rat stomach induced by long-term administration of ammonia.
Gastroenterology 104: 796-801.
Tsujii. M: Kawano. S: Tsuji. S: Takei. Y: Tamura. K: Fusamoto. H: Kamada. T. (1995). Mechanism for ammonia-
induced promotion of gastric carcinogenesis in rats. Carcinogenesis 16: 563-566.
Turner. C: Spanel. P: Smith. D. (2006). A longitudinal study of ammonia, acetone and propanol in the exhaled
breath of 30 subjects using selected ion flow tube mass spectrometry, SIFT-MS. Physiol Meas 27:
321-337. http://dx.doi.Org/10.1088/0967-3334/27/4/001.
L (U.S. Environmental Protection Agency). (1986a). Guidelines for mutagenicity risk assessment [EPA
Report]. (EPA/630/R-98/003). Washington, DC. http://www.epa.gov/iris/backgrd.html.
L (U.S. Environmental Protection Agency). (1986b). Guidelines for the health risk assessment of
chemical mixtures [EPA Report] (pp. 34014-34025). (EPA/630/R-98/002). Washington, DC.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=22567.
, (U.S. Environmental Protection Agency). (1988). Recommendations for and documentation of
biological values for use in risk assessment. (EPA/600/6-87/008). Cincinnati, OH: U.S.
Environmental Protection Agency, Environmental Criteria and Assessment Office.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=34855.
U.S. EPA (U.S. Environmental Protection Agency). (1991). Guidelines for developmental toxicity risk
assessment [EPA Report]. (EPA/600/FR-91/001). Washington, DC: U.S. Environmental Protection
Agency, Risk Assessment Forum, http://www.epa.gov/iris/backgrd.html.
U.S. EPA (U.S. Environmental Protection Agency). (1994). Methods for derivation of inhalation reference
concentrations and application of inhalation dosimetry. (EPA/600/8-90/066F). Research Triangle
Park, NC: U.S. Environmental Protection Agency, Office of Research and Development, Office of Health
and Environmental Assessment, Environmental Criteria and Assessment Office.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=71993.
U.S. EPA (U.S. Environmental Protection Agency). (1995). The use of the benchmark dose approach in health
risk assessment [EPA Report]. (EPA/630/R-94/007). Washington, DC.
http://www.epa.gov/raf/publications/useof-bda-healthrisk.htm.
U.S. EPA (U.S. Environmental Protection Agency). (1996). Guidelines for reproductive toxicity risk assessment
[EPA Report]. (EPA/630/R-96/009). Washington, DC: U.S. Environmental Protection Agency, Risk
Assessment Forum. http://www.epa.gov/raf/publications/pdfs/REPR051.PDF.
U.S. EPA (U.S. Environmental Protection Agency). (1998). Guidelines for neurotoxicity risk assessment.
(EPA/630/R-95/001F). Washington, DC: U.S. Environmental Protection Agency, Risk Assessment
Forum. http://www.epa.gov/raf/publications/pdfs/NEUROTOX.PDF.
U.S. EPA (U.S. Environmental Protection Agency). (2000a). Benchmark dose technical guidance document
[external review draft]. (EPA/630/R-00/001). Washington, DC: U.S. Environmental Protection
Agency, Risk Assessment Forum, http://www.epa.gov/raf/publications/benchmark-dose-doc-
drafthtm.
This document is a draft for review purposes only and does not constitute Agency policy.
R-6 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
U.S. EPA (U.S. Environmental Protection Agency). (2000b). Supplementary guidance for conducting health
risk assessment of chemical mixtures [EPA Report]. (EPA/630/R-00/002).
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=2 0533.
U.S. EPA (U.S. Environmental Protection Agency). (2002). A review of the reference dose and reference
concentration processes. (EPA/630/P-02/002F). Washington, DC.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=51717.
U.S. EPA (U.S. Environmental Protection Agency). (2005a). Guidelines for carcinogen risk assessment.
(EPA/630/P-03/001F). Washington, DC. http: //www. ep a. go v/cancer guidelines/.
U.S. EPA (U.S. Environmental Protection Agency). (2005b). Supplemental guidance for assessing susceptibility
from early-life exposure to carcinogens. (EPA/630/R-03/003F). Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment Forum, http://www.epa.gov/cancerguidelines/guidelines-
carcinogen-supplementhtm.
U.S. EPA (U.S. Environmental Protection Agency). (2006a). Approaches for the application of physiologically
based pharmacokinetic (PBPK) models and supporting data in risk assessment (Final Report).
(EPA/600/R-05/043F). Washington, DC: U.S. Environmental Protection Agency, Office of Research
and Development.
U.S. EPA (U.S. Environmental Protection Agency). (2006b). A framework for assessing health risk of
environmental exposures to children. (EPA/600/R-05/093F). Washington, DC.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=158363.
U.S. EPA (U.S. Environmental Protection Agency). (2007). Integrated Risk Information System (IRIS):
Announcement of 2008 program. Fed Reg 72: 72715-72719.
U.S. EPA (U.S. Environmental Protection Agency). (2009). EPAs Integrated Risk Information System:
Assessment development process. Washington, DC. http://epa.gov/iris/process.htm.
U.S. EPA (U.S. Environmental Protection Agency). (2011a). Recommended use of body weight 3/4 as the
default method in derivation of the oral reference dose. (EPA/100/R11/0001). Washington, DC.
http://www.epa.gov/raf/publications/interspecies-extrapolation.htm.
U.S. EPA (U.S. Environmental Protection Agency). (2011b). Exposure factors handbook 2011 edition (final).
(EPA/600/R-09/052F). http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=236252.
Verberk. MM. (1977). Effects of ammonia in volunteers. Int Arch Occup Environ Health 39: 73-81.
http://dx.doi.org/10.1007/BF00380887.
Vogelzang. PF: van der Gulden. IW: Preller. L: Tielen. Ml: van Schayck. CP: Folgering. H. (1997). Bronchial
hyperresponsiveness and exposure in pig farmers. Int Arch Occup Environ Health 70: 327-333.
Vogelzang. PF: van der Gulden. IW: Folgering. H: van Schayck. CP. (1998). Longitudinal changes in lung
function associated with aspects of swine-confinement exposure. J Occup Environ Med 40:1048-
1052.
Vogelzang. PF: van der Gulden. IW: Folgering. H: Heederik. D: Tielen. Ml: van Schayck. CP. (2000).
Longitudinal changes in bronchial responsiveness associated with swine confinement dust exposure.
Chest 117: 1488-1495.
Vollmuth. TA: Schlesinger. RB. (1984). Measurement of respiratory tract ammonia in the rabbit and
implications to sulfuric acid inhalation studies. Toxicol Sci 4: 455-464.
Wason. S: Stephan. M: Breide. C. (1990). Ingestion of aromatic ammonia 'smelling salts' capsules. Am J Dis
Child 144: 139-140.
Weatherby. IH. (1952). Chronic toxicity of ammonia fumes by inhalation. Exp Biol Med 81: 300-301.
Yadav. IS: Kaushik. VK. (1997). Genotoxic effect of ammonia exposure on workers in a fertilizer factory. Indian
J Exp Biol 35: 487-492.
Zejda. IE: Barber. E: Dosman. IA: Olenchock. SA: McDuffie. HH: Rhodes. C: Hurst. T. (1994). Respiratory health
status in swine producers relates to endotoxin exposure in the presence of low dust levels. J Occup
Med 36: 49-56.
This document is a draft for review purposes only and does not constitute Agency policy.
R-7 DRAFT—DO NOT CITE OR QUOTE
------- |