jlCDA
M %
EPA/635/R-16/098a
Final Agency/Interagency Draft
www.epa.gov/iris
Toxicological Review of Ammonia
Noncancer Inhalation
(CASRN 7664-41-7]
June 2016
NOTICE
This document is a Final Agency Review/Interagency Science Discussion 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.
Integrated Risk Information System
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 or
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ii DRAFT—DO NOT CITE OR QUOTE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 ix
PREAMBLE TO IRIS TOXICOLOGICAL REVIEWS xii
EXECUTIVE SUMMARY ES-1
LITERATURE SEARCH STRATEGY | STUDY SELECTION AND EVALUATION LS-1
1. HAZARD IDENTIFICATION 1-1
1.1. OVERVIEW OF CHEMICAL PROPERTIES AND TOXICOKINETICS 1-1
1.1.1. Chemical Properties 1-1
1.1.2. Toxicokinetics 1-2
1.2. SYNTHESIS OF EVIDENCE 1-3
1.2.1. Respiratory Effects 1-3
1.2.2. Immune System Effects 1-20
1.2.3. Other Systemic Effects 1-25
1.3. SUMMARY AND EVALUATION 1-33
1.3.1. Weight of Evidence for Effects Other than Cancer 1-33
1.3.2. Susceptible Populations and Lifestages 1-34
2. DOSE-RESPONSE ANALYSIS 2-1
2.1. INHALATION REFERENCE CONCENTRATION FOR EFFECTS OTHER THAN CANCER 2-1
2.1.1. Identification of Studies and Effects for Dose-Response Analysis 2-1
2.1.2. Methods of Analysis 2-5
2.1.3. Derivation of the Reference Concentration 2-7
2.1.4. Uncertainties in the Derivation of the Reference Concentration 2-8
2.1.5. Confidence Statement 2-11
2.1.6. Previous IRIS Assessment 2-12
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
Table ES-1. Summary of reference concentration (RfC) derivation ES-3
Table LS-1. Inclusion/exclusion criteria for inhalation health effect/toxicity studies (pre-peer
review) LS-4
Table LS-2. Summary of epidemiology database LS-13
Table LS-3. Considerations and relevant experimental information for evaluation of
experimental animal studies LS-14
Table LS-4. Summary of experimental animal database LS-16
Table 1-1. Chemical and physical properties of ammonia 1-1
Table 1-2. Evidence pertaining to respiratory effects in humans following inhalation exposure
in industrial settings 1-8
Table 1-3. Evidence pertaining to respiratory effect in humans following inhalation exposure in
cleaning settings 1-12
Table 1-4. Evidence pertaining to respiratory effects in animals 1-16
Table 1-5. Evidence pertaining to immune system effects in animals 1-22
Table 1-6. Evidence pertaining to other systemic effects in animals 1-27
FIGURES
Figure LS-1. Summary of literature search and screening process for ammonia LS-3
Figure 1-1. Exposure-response array of respiratory effects following inhalation exposure to
ammonia 1-18
Figure 1-2. Exposure-response array of immune system effects following inhalation exposure
to ammonia 1-24
Figure 1-3. Exposure-response array of systemic effects following inhalation exposure to
ammonia 1-31
This document is a draft for review purposes only and does not constitute Agency policy.
iv DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
ABBREVIATIONS
ALT
alanine aminotransferase
NCEA
National Center for Environmental
AST
aspartate aminotransferase
Assessment
ATSDR
Agency for Toxic Substances and Disease
NHs
ammonia
Registry
NH4+
ammonium ion
BCG
bacillus Calmette-Guerin
NIOSH
National Institute for Occupational
BMCL
95% lower bound on the benchmark
Safety and Health
concentration
NOAEL
no-observed-adverse-effect level
BMDL
95% lower bound on the benchmark
NRC
National Research Council
dose
ORD
EPA's Office of Research and
CAC
cumulative ammonia concentration
Development
CCRIS
Chemical Carcinogenesis Research
PEFR
peak expiratory flow rate
Information System
pO 2
oxygen partial pressure
CERCLA
Comprehensive Environmental
POD
point of departure
Response, Compensation, and Liability
PPD
purified protein derivative
Act
RfC
reference concentration
CFU
colony forming unit
RfD
reference dose
CI
confidence interval
RTECS
Registry of Toxic Effects of Chemical
DAP
diammonium phosphate
Substances
EPA
Environmental Protection Agency
TSCATS
Toxic Substance Control Act Test
FEVi
forced expiratory volume in 1 second
Submission Database
FVC
forced vital capacity
UF
uncertainty factor
HERO
Health and Environmental Research
UFa
interspecies uncertainly factor
Online
UFh
intraspecies uncertainty factor
HSDB
Hazardous Substances Data Bank
UFl
LOAEL to NOAEL uncertainty factor
IgE
immunoglobulin E
UFs
subchronic-to-chronic uncertainty factor
IgG
immunoglobulin G
UFd
database deficiencies uncertainty factor
IRIS
Integrated Risk Information System
VEh
human occupational default minute
LDso
50% lethal dose
volume
LOAEL
lowest-observed-adverse-effect level
VEho
human ambient default minute volume
MAO
monoamine oxidase
MNNG
N-methyl-N'-nitro-N-nitrosoguanidine
MRM
murine respiratory mycoplasmosis
This document is a draft for review purposes only and does not constitute Agency policy.
v DRAFT—DO NOT CITE OR QUOTE
-------
1
3 AUTHORS | CONTRIBUTORS | REVIEWERS
4
5
Assessment Team
Audrey Galizia, Dr. PH (Assessment U.S. EPA/ORD/NCEA
Manager)
James Ball, Ph.D. (no longer with EPA)
Glinda Cooper, Ph.D. (no longer with
EPA)
Louis D'Amico, Ph.D.
Keith Salazar, Ph.D.
Christopher Sheth, Ph.D. (no longer with
EPA)
Christopher Brinkerhoff, Ph.D. ORISE Postdoctoral Fellow at the U.S. EPA
Washington, DC (currently with the Office of
Chemical Safety and Pollution Prevention)
Scientific Support Team
Vincent Cogliano, Ph.D. U.S. EPA/ORD/NCEA
Samantha Jones, Ph.D.
Jamie Strong, Ph.D. (currently with the
Office of Water)
Ted Berner, MS
Jason Fritz, Ph.D.
Martin Gehlhaus, MPH (currently with
EPA Region 3)
John Stanek, Ph.D.
Production Team
Maureen Johnson U.S. EPA/ORD/NCEA
Ellen F. Lorang, MA (retired)
Vicki Soto
Contractor Support
Amber Bacom, MS SRC, Inc., Syracuse, NY
Fernando Llados, Ph.D.
Julie Stickney, Ph.D.
Executive Direction
Kenneth Olden, Ph.D., Sc.D., L.H.D. (Center Director) U.S. EPA/ORD/NCEA
Lynn Flowers, Ph.D., DABT (Associate Director for Health,
currently with the Office of Science Policy)
John Vandenberg, Ph.D. (HHRA National Program Director)
Vincent Cogliano, Ph.D. (IRIS Program Director)
Gina Perovich, M.S. (IRIS Program Deputy Director)
Samantha Jones, Ph.D. (IRIS Associate Director for Science)
Susan Rieth, MPH (Branch Chief)
This document is a draft for review purposes only and does not constitute Agency policy.
vi DRAFT—DO NOT CITE OR QUOTE
-------
1
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Internal Reviewers
Marian Rutigliano, MD U.S. EPA/ORD/NCEA
Amanda S. Persad, Ph.D.
Paul Reinhart, Ph.D.
John Whalan
Reviewers
This assessment was provided for review to scientists in EPA's Program and Regional Offices.
Comments were submitted by:
Office of the Administrator/Office of Children's Health Protection
Office of Air and Radiation/Office of Air Quality and Planning Standards
Office of Air and Radiation/Office of Transportation and Air Quality
Office of Policy
Office of Water
Region 2, New York, NY
This assessment was provided for review to other federal agencies and the Executive Office of the
President. Comments were submitted by:
Department of Agriculture/Food Safety and Inspection Service
Department of Health and Human Services/Agency for Toxic Substances and Disease
Registry
Executive Office of the President/Council on Environmental Quality
This assessment was released for public comment on June 8, 2012 and comments were due on
August 7, 2012. A summary and EPA's disposition of the comments received from the public is
included in Appendix G of the Supplemental Information to the Revised External Review draft of the
Toxicological Review. Comments were received from the following entities:
The American Chemistry Council Washington, DC
The Fertilizer Institute Washington, DC
This assessment was peer reviewed by independent expert scientists external to EPA convened by
EPA's Science Advisory Board (SAB), the Chemical Assessment Advisory Committee Augmented for
the IRIS Ammonia Assessment. A peer review meeting was held on July 14 to 16, 2014. The report
ofthe SAB's review ofEPA's Draft Toxicological Review of Ammonia, dated August 6, 2015, is
available on the IRIS website.
Dr. Michael Dourson (chair), Toxicology Excellence for Risk Assessment, Cincinnati, OH
Dr. Daniel Acosta, National Center for Toxicological Research, U. S. Food and Drug
Administration, Jefferson, AR
Dr. Henry Anderson, Wisconsin Division of Public Health, Madison, WI
Dr. Scott Bartell, Program in Public Health, University of California - Irvine, Irvine, CA
Dr. Arthur Cooper, Department of Biochemistry and Molecular Biology, New York Medical
College, Valhalla, NY
Dr. David Eastmond, Department of Cell Biology and Neuroscience and Environmental
Toxicology Graduate Program, University of California, Riverside, CA
This document is a draft for review purposes only and does not constitute Agency policy.
vii DRAFT—DO NOT CITE OR QUOTE
-------
Dr. William Michael Foster, Pulmonary and Critical Care Medicine, Duke University Medical
Center, Durham, NC
Dr. Russ Hauser, Department of Environmental Health, Harvard T. H. Chan School of Public
Health, Harvard University, Boston, MA
Dr. Abby A. Li, Health Science Practice, Exponent Incorporated, San Francisco, CA
Dr. Maria Morandi, Independent Consultant, Houston, TX
Dr. Victoria Persky, Epidemiology & Biostatistics Program, School of Public Health, University of
Illinois at Chicago, Chicago, IL
Dr. Kenneth Ramos, Precision Health Sciences and Professor of Medicine, Arizona Health
Sciences Center, Tucson, AZ
Dr. Alan H. Stern, New Jersey Department of Environmental Protection, Trenton, NJ, and
Department of Environmental and Occupational Health, Rutgers University School of Public
Health, NJ
Dr. I. David Weiner, Division of Nephrology, Hypertension & Renal Transplantation, University
of Florida, Gainesville, FL
Designated Federal Officer: Dr. Suhair Shallal
This document is a draft for review purposes only and does not constitute Agency policy.
viii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
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 the Environmental Protection Agency's
(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).
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 (RfC) for
effects other than cancer. This assessment provides an updated review of information on all
noncancer health effects by the inhalation route only.
This assessment was conducted in accordance with EPA guidance; relevant EPA guidance
documents can be found on the IRIS website (http://www.epa. go v/iris A The findings of this
assessment and related documents produced during its development are also available on the IRIS
website f http: / /www, ep a. gov/ iris A Appendices for other health toxicity values, details of the
literature search strategy and study selection and evaluation, supporting information for hazard
identification and dose response, and other information are provided as Supplemental Information
to this assessment (see Appendices A to C).
Portions of this Toxicological Review were adapted from the Toxicological Profile for
Ammonia developed by the Agency for Toxic Substances and Disease Registiy (ATSDR. 20041 under
a Memorandum of Understanding that encourages interagency collaboration, sharing of scientific
information, and more efficient use of resources.
The IRIS program released this assessment for public comment and peer review in June
2012, as it was beginning to implement systematic review. The approach to implementation is to
use procedures and tools available at the time, without holding assessments until new methods
become available. Accordingly, the IRIS program edited this assessment to increase transparency
and clarity and to use more tables and figures. It conducted literature searches and evaluated
studies using tools and documentation standards then available. Problem formulation materials
and protocol development began with assessments started in 2015, after this assessment was well
This document is a draft for review purposes only and does not constitute Agency policy.
ix DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
into peer review. This assessment addresses peer review comments and retains the structure of
the peer review draft, to maintain fidelity with what the peer reviewers saw. Implementation of
systematic review is a process of continuous improvement subject to periodic review by the
Chemical Assessment Advisory Committee of the U.S. EPA's Science Advisory Board. This
assessment represents a step in the evolution of the IRIS program.
Assessments by Other National and International Health Agencies
Toxicity information on ammonia has been evaluated by ATSDR, the National Research
Council (NRC), the National Institute for Occupational Safety and Health, and the Food and Drug
Administration. The results of these assessments are presented in Appendix A of the Supplemental
Information. It is important to recognize that these assessments may have been prepared for
different purposes and may utilize different methods, and that newer studies may be included in
the IRIS assessment.
Overview of Uses, Sources, and Environmental Exposure
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, in the
pulp and paper and metallurgy industries, 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 fHSDB. 2012: lohnson et
al.. 2009: Eggeman. 20011.
Major sources of ammonia gas include leaks and spills during commercial synthesis,
production, storage, processing, or transporting of ammonia; refrigeration equipment failure;
decaying manure from livestock; application of fertilizers; sewage or wastewater effluent; burning
of coal, wood or other natural products; volcanic eruptions, forest fires and the decomposition of
nitrogenous compounds. Ammonia from agricultural and other sources, along with sulfate and
nitrate salts, is an important contributor to fine inorganic particulate matter (PM2.5) mass (e.g.,
see Paulotand lacob f201411 This literature on airborne particular matter is reviewed and
evaluated in EPA's Integrated Science Assessment for Particulate Matter (PM ISA) fU.S. EPA.
2009b).
Environmental exposures to ammonia in the air vary widely. Average ambient
concentrations of ammonia in the United States range from 0.28-15 |J.g/m3, as measured in 2012 by
the National Atmospheric Deposition Program's Ammonia Monitoring Network fAMoN. 20121.
Indoor residential ammonia concentrations can vary widely; one survey reported ammonia
concentrations in homes in Connecticut and southwest and central Virginia ranging from 0.09-
This document is a draft for review purposes only and does not constitute Agency policy.
x DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
166 |J.g/m3, depending on the season, use of air conditioning, type of heating, and other factors
fLeaderer etal.. 19991.
Ammonia is found naturally in the environment and is a component of the global nitrogen
cycle; it is essential to many biological 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
fNelson and Cox. 2008: Socolow. 1999: Rosswall. 19811.
Scope of this Assessment
This assessment presents an evaluation of the noncancer health effects of ammonia by the
inhalation route of exposure. To address peer-review recommendations to expand the scope of the
oral toxicity literature to include ammonium salts and to allow expeditious completion of the
assessment of inhaled ammonia, ingested ammonia, including consideration of ammonium salts,
will be the focus of a separate assessment. Because carcinogenicity studies of ammonia have been
performed by the oral route of exposure only, the cancer assessment will be moved into the
separate oral assessment
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.
This document is a draft for review purposes only and does not constitute Agency policy.
xi DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
PREAMBLE TO IRIS TOXICOLOGICAL REVIEWS
The Preamble summarizes the objectives and scope of the IRIS program, general principles
and systematic review procedures used in developing IRIS assessments, and the overall
development process and document structure.
1. Objectives and Scope of the IRIS 43
Program 44
Soon after the EPA was established in 1970, it
45
46
was at the forefront of developing risk assessment^
48
as a science and applying it in support of actions
to protect human health and the environment. The
49
EPA's IRIS program1 contributes to this endeavor^
by identifying adverse health effects of chemicals ^
in the environment and characterizing exposure- ^
response relationships. IRIS assessments cover ^
the hazard identification and dose-response step^
of risk assessment. Exposure assessment and risk^
characterization are outside the scope of IRIS ^
assessments, as are political, economic, and
technical aspects of risk management.
An IRIS assessment may cover one chemical, a^
group of structurally or toxicologically related ^
chemicals, or a chemical mixture. Exceptions 60
outside the scope of the IRIS program are 61
radionuclides, chemicals used only as pesticides,
and the "criteria air pollutants" (particulate
matter, ground-level ozone, carbon monoxide,
sulfur oxides, nitrogen oxides, and lead].
Enhancements to the IRIS program are
improving its science, transparency, and
productivity. To improve the science, the IRIS
program is adapting and implementing principles
of systematic review (i.e., using explicit methods
to identify, evaluate, and synthesize study
findings]. To increase transparency, the IRIS
program releases a problem formulation and
other materials during draft development and
discusses key science questions with the scientific^
community and the public. External peer review,
62
63
64
65
66
67
68
69
70
71
72
73
:f:„74
75
independently managed and in public, improves
both science and transparency. Increased
productivity requires that assessments be concise,
focused on EPA's needs, and completed without
undue delay.
This assessment was conducted in accordance
with EPA guidance.2 This Preamble summarizes
and does not change IRIS operating procedures or
EPA guidance.
Periodically, the IRIS program asks for
nomination of agents for future assessment or
reassessment. Selection depends on EPA's
priorities, relevance to public health, and
availability of pertinent studies. The IRIS
multiyear agenda3 lists upcoming assessments.
The IRIS program may also assess other agents in
anticipation of public health needs.
2. Planning an Assessment: Scoping,
Problem Formulation, and Protocols
Early attention to planning ensures that IRIS
assessments meet EPA's needs and properly
frame science questions.
Scoping refers to the first step of planning
where the IRIS program consults with EPA's
program and regional offices to ascertain their
needs. Scoping specifies the agents an assessment
will address, routes and durations of exposure,
susceptible populations and lifestages, and other
questions of interest to the EPA.
Problem formulation refers to the science
questions an assessment will address and includes
input from the scientific community and the
public. A preliminary survey of secondary sources
1 IRIS program website: http: //www.epa.gov/iris/
2 EPA guidance documents: http://www.epa.gov/iris/basic-information-about-integrated-risk-information-
svstem# guidance /
3 IRIS multiyear agenda: https: //www.epa.gov/iris/iris-agenda
This document is a draft for review purposes only and does not constitute Agency policy.
xii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
(e.g., assessments by national and international 46
health agencies and comprehensive review 47
articles] identifies potential health outcomes and 48
science questions. It also identifies related 49
chemicals (e.g., toxicologically active metabolites 50
and compounds that metabolize to the chemical o6l
interest], 52
Each IRIS assessment comprises multiple 53
systematic reviews for multiple health outcomes. 54
It also evaluates hypothesized mechanistic 55
pathways and characterizes exposure-response 56
relationships. An assessment may focus on 57
important health outcomes and analyses rather 58
than expand beyond what is necessary to support59
EPA's needs. 60
Protocols refer to the systematic review 61
procedures planned for use in an assessment. 62
They include strategies for literature searches, 63
criteria for study inclusion or exclusion, 64
considerations for evaluating study methods and 65
quality, and approaches to extracting data. As an 66
assessment progresses, additional science 67
questions may emerge and protocols may change.^
3. Identifying and Selecting Pertinent69
Studies 70
IRIS assessments conduct systematic literature^
searches with criteria for inclusion and exclusion.72
The objective is to retrieve the pertinent primary 73
studies (i.e., studies with original data on health 74
outcomes or their mechanisms], PECO statements75
(Populations, Exposures, Comparisons, Outcomes^
govern the literature searches and screening 77
criteria. "Populations" and animal species 73
generally have no restrictions. "Exposures" refers79
to the agent and related chemicals identified gQ
during scoping and problem formulation and mayg^
consider route, duration, or timing of exposure. g2
"Comparisons" means studies that allow gj
comparison of effects across different levels of
exposure. "Outcomes" may become more specific g^
(e.g., from "toxicity" to "developmental toxicity" tc^
"hypospadias"] as an assessment progresses. g7
For studies of absorption, distribution, gg
metabolism, and elimination, the first objective is gg
to create an inventory of pertinent studies.
Subsequent sorting and analysis facilitates
characterization and quantification of these
processes.
Studies on mechanistic events can be
numerous and diverse. Here, too, the objective is
to create an inventory of studies for later sorting
to support analyses of related data. The inventory
also facilitates generation and evaluation of
hypothesized mechanistic pathways.
IRIS assessments go beyond standard practices
of systematic review in including pertinent
studies. After posting search strategies on its
website and adding search results to the EPA's
HERO database,4 the IRIS program encourages the
scientific community and the public to provide
information on additional studies and ongoing
research. Assessments also consider data
submitted under the Toxic Substances Control Act
and the Federal Insecticide, Fungicide, and
Rodenticide Act. Even during the review process,
IRIS assessments consider late-breaking studies
that would impact the credibility of the
conclusions.5
4. Evaluating Study Methods and
Quality
IRIS assessments evaluate study methods and
quality, using uniform approaches for each group
of similar studies. The objective is that subsequent
syntheses can weigh study results on their merits.
Key concerns are bias (factors that affect the
magnitude or direction of an effect] and sensitivity
(factors that limit the ability of a study to detect a
true effect].
For human and animal studies, the evaluation
of study methods and quality considers study
design, exposure characterization, outcome
assessment, data analysis, and selective reporting.
For human studies, this evaluation also considers
selection of participant and referent groups and
potential confounding. Emphasis is on discerning
bias that would substantively change an effect
estimate, considering also the expected direction
of the bias. Low sensitivity is a bias towards the
null.
4 Health and Environmental Research Online: https: //hero.epa.gov/hero/
5 IRIS "stopping rules": https: //www.epa.gov/sites/production/files/2014-06/documents/
iris stoppingrules.pdf
This document is a draft for review purposes oniy and does not constitute Agency poiicy.
xiii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Study-evaluation considerations are specific tc49
each study design, agent, and health effect. 50
Subject-matter experts evaluate each group of 51
studies to identify characteristics that would make2
results more or less informative. For 53
carcinogenicity, neurotoxicity, reproductive 54
toxicity, and developmental toxicity, there is EPA 55
guidance for study evaluation. As subject-matter 56
experts examine a group of studies, additional 57
methodologic concerns may emerge and a second58
pass become necessary. 59
Assessments use evidence tables to summarizaSO
the design and results of pertinent studies. If 61
tables become too numerous or unwieldy, they 62
may focus on effects that are more important or 63
studies that are more informative. 64
The IRIS program posts on its website the 65
study-evaluation considerations and table entries66
for illustrative studies, then considers public inpu67
on these approaches as it completes study 68
evaluation and data extraction. 69
70
5. Integrating the Evidence of 71
Causation for Each Health Outcome^
Synthesis within lines of evidence. For each 73
health outcome, IRIS assessments synthesize the 74
human evidence and the animal evidence, 75
augmenting each with informative subsets of 76
mechanistic data. Each synthesis considers 77
aspects of an association that may suggest 78
causation: consistency, exposure-response 79
relationship, strength of association, temporal 80
relationship, biological plausibility, coherence, 81
and "natural experiments" in humans. 82
Each synthesis seeks to reconcile ostensible 83
inconsistencies between studies, taking into 84
account differences in study methods and quality.85
This leads to a distinction between conflicting 86
evidence (unexplained positive and negative 87
results in similarly exposed human populations 0188
in similar animal models] and differing results 89
(mixed results attributable to differences betweerDO
human populations, animal models, or exposure 91
conditions], 92
Each synthesis of human evidence explores 93
alternative explanations (e.g., chance, bias, or 94
confounding] and determines whether they 95
satisfactorily explain the results. Each synthesis of>6
animal evidence explores the potential for 97
analogous results in humans. Coherent results
across multiple species increase confidence that
the animal results are relevant to humans.
Mechanistic data are useful to augment the
human or animal evidence with information on
precursor events, to evaluate the human relevance
of animal results, or to identify susceptible
populations and lifestages. An agent may operate
through multiple mechanistic pathways, even if
one hypothesis dominates the literature.
Integration across lines of evidence. For
each health outcome, IRIS assessments integrate
the human, animal, and mechanistic evidence to
answer the question: What is the nature of the
association between exposure to the agent and the
health outcome?
For cancer, the EPA includes a standardized
hazard descriptor in characterizing the strength of
the evidence of causation. The objective is to
promote clarity and consistency of conclusions
across assessments.
Carcinogenic to humans: convincing epidemiologic
evidence of a causal association; or strong
human evidence of cancer or its key
precursors, extensive animal evidence,
identification of mode-of-action and its key
precursors in animals, and strong evidence
that they are anticipated in humans.
Likely to be carcinogenic to humans: evidence that
demonstrates a potential hazard to humans.
Examples include a plausible association in
humans with supporting experimental
evidence, multiple positive results in animals, a
rare animal response, or a positive study
strengthened by other lines of evidence.
Suggestive evidence of carcinogenic potential:
evidence that raises a concern for humans.
Examples include a positive result in the only
study, or a single positive result in an extensive
database.
Inadequate information to assess carcinogenic
potential: no other descriptors apply. Examples
include little or no pertinent information,
conflicting evidence, or negative results not
sufficiently robust for not likely.
Not likely to be carcinogenic to humans: robust
evidence to conclude that there is no basis for
concern. Examples include no effects in well-
This document is a draft for review purposes only and does not constitute Agency policy.
xiv DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
conducted studies in both sexes of multiple 48
animal species, extensive evidence showing 49
that effects in animals arise through modes-of-50
action that do not operate in humans, or 51
convincing evidence that effects are not likely 52
by a particular exposure route or below a 53
defined dose. 54
55
If there is credible evidence of carcinogenicity,56
an assessment determines whether the mode-of- 57
action involves mutagenicity, because this 58
influences the approach to dose-response ^
assessment and subsequent application of
adjustment factors for exposures early in life. 60
The EPA is discussing the potential use of 61
hazard descriptors for noncancer outcomes in 62
IRIS assessments. 63
64
6. Selecting Studies for Derivation of 65
Toxicity Values 66
The purpose of toxicity values (i.e., slope 67
factors, unit risks, reference doses, reference 68
concentrations; see section 7] is to estimate 69
exposure levels likely to be without appreciable 70
risk of adverse health effects. The EPA uses these 71
values to support its actions to protect human 72
health. 73
The health outcomes considered for derivatiori74
of toxicity values may depend on the hazard 75
descriptors. For example, IRIS assessments 76
generally derive cancer values for agents that are 77
carcinogenic or likely to be carcinogenic, and 78
sometimes for agents with suggestive evidence. 79
Derivation of toxicity values begins with a new80
evaluation of studies, as some studies used 81
qualitatively for hazard identification may not be 82
useful quantitatively for exposure-response 83
assessment. Quantitative analyses require 84
quantitative measures of exposure and response. 85
An assessment weighs the merits of the human 86
and animal studies, of various animal models, and87
of different routes and durations of exposure. 88
Study selection is not reducible to a formula, and 89
each assessment explains its approach. 90
Other biological determinants of study quality 91
include appropriate measures of exposure and 92
response, investigation of early effects that 93
precede overt toxicity, and appropriate reporting 94
of related effects (e.g., combining effects that
comprise a syndrome, or benign and malignant
tumors in a specific tissue].
Statistical determinants of study quality
include multiple levels of exposure (to
characterize the shape of the exposure-response
curve] and adequate exposure range and sample
sizes (to minimize extrapolation and maximize
precision].
Studies of low sensitivity tend to
underestimate toxicity and may be less useful.
7. Deriving Toxicity Values
General approach. EPA guidance describes a
two-step approach to dose-response assessment:
analysis in the range of observation, then
extrapolation to lower levels. The analysis
considers studies by the exposure route of interest
and may include studies by other routes if dose
conversion is possible.
IRIS assessments derive a candidate value from
each suitable data set Consideration of candidate
values yields a toxicity value for each organ or
system. Consideration of the organ/system-
specific values results in the selection of an overall
toxicity value to cover all health outcomes. The
organ/system-specific values are useful for
subsequent cumulative risk assessments that
consider the combined effect of multiple agents
acting at a common anatomical site.
Analysis in the range of observation. Within
the observed range, the preferred approach is
modeling to incorporate a wide range of data.
Toxicokinetic modeling has become increasingly
common for its ability to support target-dose
estimation, cross-species adjustment, or
exposure-route conversion. If data are too limited
to support toxicokinetic modeling, there are
standardized approaches to estimate daily
exposures and scale them from animals to
humans.
For human studies, an assessment may
develop exposure-response models that reflect
the structure of the available data. For animal
studies, the EPA has developed a set of empirical
("curve-fitting"] models6 that can fit typical data
sets. Such modeling yields a point of departure,
defined as a dose near the lower end of the
6 Benchmark Dose Software: http: //www.epa.gov/bmds/
This document is a draft for review purposes only and does not constitute Agency policy.
xv DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
observed range, without significant extrapolation49
to lower levels (e.g., the estimated dose associatecfcO
with an extra risk of 10% for animal data or 1% 51
for human data, or their 95% lower bounds], 52
With complex data, an assessment may 53
develop specialized exposure-response models if 54
compatible with the scope of the assessment. 55
Toxicodynamic ("biologically based"] modeling is 56
possible if data are sufficient to ascertain the key 57
events of a mode-of-action and to estimate their 58
parameters. For a group of agents that act at a 59
common site or through common mechanisms, an60
assessment may derive relative potency factors 61
based on relative toxicity, rates of absorption or 62
metabolism, quantitative structure-activity 63
relationships, or receptor-binding characteristics.64
Extrapolation: slope factors and unit risks. 65
An oral slope factor or an inhalation unit risk 66
facilitates subsequent estimation of human cancei67
risks at low levels of exposure. They presuppose a68
linear component to the dose-response curve 69
below the point of departure (e.g., if the mode-of- 70
action involves mutagenicity], or there may be no71
established mode-of-action. Extrapolation 72
proceeds linearly (i.e., risk proportional to dose] 73
from the point of departure to the levels of 74
interest 75
Differences in susceptibility may warrant 76
derivation of multiple slope factors or unit risks. 77
For early-life exposure to known or likely 78
carcinogens whose mode-of-action involves 79
mutagenicity, the EPA has developed default age- 80
dependent adjustment factors for agents without ^
chemical-specific susceptibility data.
If data are sufficient to ascertain the key events
of the mode-of-action and to conclude that they 83
are not linear at low levels, extrapolation may use84
the reference-value approach. 85
Extrapolation: reference values. An oral 86
reference dose or an inhalation reference 87
concentration is an estimate of human exposure 88
(including in susceptible populations] likely to be 89
without appreciable risk of adverse health effects 90
over a lifetime. Reference values generally cover 91
effects other than cancer. They are also 92
appropriate for cancer if a well-characterized 93
mode-of-action indicates that a necessary key 94
event does not occur below a specific dose. 95
96
Calculation of reference values starts with a
point of departure, generally for an early effect
that precedes overt toxicity. To account for
different sources of uncertainty and variability, an
assessment applies uncertainty factors (each
typically 1, 3, or 10] to the point of departure.
Human variation: An uncertainty factor covers
susceptible populations and lifestages that may
respond at lower levels, unless the data
originate from a susceptible study population.
Animal-to-human extrapolation: For reference
values based on animal results, an uncertainty
factor reflects cross-species differences, which
may cause humans to respond at lower levels.
Subchronic-to-chronic exposure: For reference
values based on subchronic studies, an
uncertainty factor reflects the likelihood that a
lower level over a longer duration may induce
a similar response. This factor may not be
necessary for reference values of shorter
duration.
Adverse-effect level to no-observed-adverse-effect
level: For reference values based on a lowest-
observed-adverse-effect level, an uncertainty
factor reflects a level judged to have no
observable adverse effects.
Database deficiencies: If there is concern that
additional studies may identify a more
sensitive effect, target organ, population, or
lifestage, a database uncertainty factor reflects
the nature of the database deficiency.
8. Process for Developing and Peer-
Reviewing IRIS Assessments
The IRIS process (revised in 2009 and
enhanced in 2013] involves extensive public
engagement and multiple levels of scientific
review.
Step 1: Draft development. As outlined in
section 2 of this Preamble, IRIS program
scientists specify the scope of an assessment
and formulate science questions for discussion
with the scientific community and the public.
Next, they release protocols for the systematic
review procedures planned for use in the
assessment. IRIS program scientists then
develop a first draft, using structured
approaches to identify pertinent studies,
This document is a draft for review purposes only and does not constitute Agency policy.
xvi DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
87
evaluate study methods and quality, integrate 44
the evidence of causation for each health 45
outcome, select studies for derivation of 46
toxicity values, and derive toxicity values, as 47
outlined in Preamble sections 3-7. 48
Step 2: Agency review. Health scientists across 49
the EPA review the draft assessment. 50
Step 3: Interagency science consultation. 0ther51
federal agencies and the Executive Office of thd52
President review the draft assessment. 53
Step 4: Public comment, followed by external 54
peer review. The public reviews the draft 55
assessment. IRIS program scientists address 56
the public comments, then release a revised 57
draft for independent external peer review. 58
The peer reviewers consider whether the draft59
assessment assembled and evaluated the 60
evidence according to EPA guidance and 61
whether the evidence justifies the conclusions.62
Step 5: Revise assessment. IRIS program 63
scientists revise the assessment to address the64
comments from the peer review. 65
Step 6: Final agency review and interagency 66
science discussion. The IRIS program 67
discusses the revised assessment with EPA's 68
program and regional offices and with other 69
federal agencies and the Executive Office of th£70
President. 71
Step 7: Post final assessment. The IRIS program72
posts the completed assessment and a 73
summary on its website. 74
75
9. General Structure of IRIS ?6
Assessments 77
Main text. IRIS assessments generally 78
comprise two major sections: (1] Hazard 79
Identification and (2] Dose-Response AssessmentSO
Section 1.1 briefly reviews chemical properties 81
and toxicokinetics to describe the disposition of 82
the agent in the body. This section identifies 83
related chemicals and summarizes their health 84
outcomes, citing authoritative reviews. If an 85
assessment covers a chemical mixture, this sectiol86
discusses environmental processes that alter the
mixtures humans encounter and compares them
to mixtures studied experimentally.
Section 1.2 includes a subsection for each
major health outcome. Each subsection discusses
the respective literature searches and study
considerations, as outlined in Preamble sections 3
and 4, unless covered in the front matter. Each
subsection concludes with evidence synthesis and
integration, as outlined in Preamble section 5.
Section 1.3 links health hazard information to
dose-response analyses for each health outcome.
One subsection identifies susceptible populations
and lifestages, as observed in human or animal
studies or inferred from mechanistic data. These
may warrant further analysis to quantify
differences in susceptibility. Another subsection
identifies biological considerations for selecting
health outcomes, studies, or data sets, as outlined
in Preamble section 6.
Section 2 includes a subsection for each
toxicity value. Each subsection discusses study
selection, methods of analysis, and derivation of a
toxicity value, as outlined in Preamble sections 6
and 7.
Front matter. The Executive Summary
provides information historically included in IRIS
summaries on the IRIS program website. Its
structure reflects the needs and expectations of
EPA's program and regional offices.
A section on systematic review methods
summarizes key elements of the protocols,
including methods to identify and evaluate
pertinent studies. The final protocols appear as an
appendix.
The Preface specifies the scope of an
assessment and its relation to prior assessments.
It discusses issues that arose during assessment
development and emerging areas of concern. The
Preface also identifies assessment-specific
approaches that may differ from the general
approaches outlined in this Preamble.
May 2016
This document is a draft for review purposes only and does not constitute Agency policy.
xvii DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
EXECUTIVE SUMMARY
Occurrence and Health Effects
Ammonia occurs naturally in air, soil, and water. Ammonia is also produced
by humans and other animals as part of normal biological processes.
Ammonia is used as an agricultural fertilizer and in many cleaning products.
Exposure to ammonia occurs primarily through breathing air containing ammonia
gas, and may also occur via diet, drinking water, or direct skin contact.
Concentrations of ammonia measured in ambient outdoor air range from 0.28-
15 |J.g/m3 and in indoor air from 0.09-166 |ig/m3.
Health effects of inhaled ammonia observed at levels exceeding naturally-
occurring concentrations are generally limited to the respiratory tract, the site of
direct contact with ammonia. Short-term inhalation exposure to high levels of
ammonia in humans can cause irritation and serious burns in the mouth, lungs, and
eyes. Chronic exposure to airborne ammonia can increase the risk of respiratory
irritation, cough, wheezing, tightness in the chest, and reduction in the normal
function of the lung in humans. Studies in experimental animals similarly indicate
that breathing ammonia at sufficiently high concentrations can result in effects on
the respiratory system. Animal studies also suggest that exposure to high levels of
ammonia in air may adversely affect other organs, such as the liver, kidney, and
spleen.
Chemical Properties
Ammonia (NH3) is a colorless alkaline gas with a pungent odor. In solution, ammonia exists
as ammonium hydroxide, a weak base that is only partially ionized in water according to the
following equilibrium fATSDR. 20041: NH3 + H2O *± NH4+ + OH". A decrease in pH results in an
increase in the concentration of ammonium ion (NH4+) and a decrease in the concentration of the
un-ionized form (NH3). At physiological pH (7.4), this equilibrium favors the formation of NH4+.
Toxicokinetics
Inhaled ammonia is almost completely retained in the upper respiratory tract Ammonia
produced endogenously in the intestines through the use of amino acids as an energy source and by
bacterial degradation of nitrogenous compounds from ingested food is largely absorbed. At
physiological pH, 98.3% of ammonia is present in the blood as the ammonium ion (NH4+). Given its
importance in amino acid metabolism, the urea cycle, and acid-base balance, ammonia is
homeostatically regulated to remain at low concentrations in the blood. Ammonia is present in
fetal circulation and in human breast milk as a source of nonprotein nitrogen. Ammonia production
This document is a draft for review purposes only and does not constitute Agency policy.
ES-1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
occurs endogenously by catabolism of amino acids by glutamate dehydrogenase or glutaminase
primarily in the liver, renal cortex and intestines, but also in the brain and heart Ammonia is
metabolized to glutamine via glutamine synthetase in the glutamine cycle or incorporated into urea
as part of the urea cycle. The principal means of excretion of ammonia is as urinary urea; lesser
amounts are eliminated in the feces, through sweat production, and in expired air.
Effects Other Than Cancer Observed Following Inhalation Exposure
Respiratory effects have been identified as a human health hazard following inhalation
exposure to ammonia. This hazard determination is based on findings from multiple epidemiology
studies in human populations exposed to ammonia in different settings (workers in industrial,
cleaning and agricultural settings, volunteers exposed for up to 6 hours under controlled
conditions, and case reports) and animals (short-term and subchronic studies in several species
and across different exposure regimes).
Cross-sectional occupational studies involving chronic exposure to ammonia in industrial
settings provide evidence of an increased prevalence of respiratory symptoms f Rah man etal..
2007: Ballal etal.. 1998) and decreased lung function f Rah man etal.. 2007: Ali etal.. 2001: Ballal et
al.. 1998: Bhatand Ramaswamv. 19931. Other studies of exposure to ammonia when used as a
disinfectant or cleaning product, for example in health care workers, provide additional evidence of
effects on asthma, asthma symptoms, and pulmonaiy function, using a variety of study designs
(Casas etal.. 2013: Arif and Delclos. 2012: Dumas etal.. 2012: Lemiere etal.. 2012: Vizcavaetal..
2011: Zock etal.. 2007: Medina-Ramon et al.. 2006: Medina-Ramon etal.. 2005). Additional
evidence of respiratory effects of ammonia is seen in studies of pulmonary function in an
agricultural setting, specifically in the studies that accounted for effects of co-exposures to other
agents such as endotoxin and dust f Don ham et al.. 2000: Reynolds etal.. 1996: Donham etal..
1995: Preller etal.. 1995: Heederik etal.. 1990) and in one study that did not control for co-
exposures fLoftus etal.. 20151. Despite the variation in population characteristics, level and
pattern of exposure, and potential confounders across these three settings of epidemiology studies,
respiratory effects were consistently observed in these studies. Further, but more limited, support
for the respiratory system as a target of ammonia toxicity comes from controlled human exposure
studies of ammonia inhalation and case reports of injury in humans with inhalation exposure to
ammonia. Additionally, respiratory effects were observed in several animal species following short-
term and subchronic inhalation exposures to ammonia.
Overall, there are suggestions in experimental animals that ammonia exposure may be
associated with effects on organs distal from the portal of entry, but there is inadequate
information to draw conclusions about the liver, kidney, spleen, or heart as sensitive targets of
ammonia toxicity.
This document is a draft for review purposes only and does not constitute Agency policy.
ES-2 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Inhalation Reference Concentration (RfC) for Effects Other Than Cancer
Table ES-1. Summary of reference concentration (RfC) derivation
Critical effect
Point of departure3
UF
Chronic RfC
Decreased lung function and respiratory symptoms
Occupational epidemiology studies
Holness et al. (1989), supported bv Rahman et al.
(2007), Ballal et al. (1998), and Ali et al. (2001)
NOAELadj: 4.9 mg/m3
10
0.5 mg/m3
aAn estimate of the 95% lower confidence bound of the mean exposure concentration in the high-exposure
group of the Holness et al. (1989) study was used as the NOAEL. Because the study involved workplace
exposure conditions, the NOAEL of 13.6 mg/m3 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 ).
with support from three studies in urea fertilizer plants by Rahman et al. (2007). Ballal et al.
(1998). and Ali etal. (2001). was identified as the principal study for RfC derivation. Respiratory
effects, characterized as increased respiratory symptoms based on self-report (including cough,
wheezing, and other asthma-related symptoms) and decreased lung function in workers exposed to
ammonia, were selected as the critical effect Holness etal. T19891 found no differences in the
prevalence of respiratory symptoms or lung function between workers (mean exposure 6.5 mg/m3)
and the control group, and no differences when stratified by exposure level (highest exposure
group, >8.8 mg/m3). Rahman et al. (2007) observed an increased prevalence of respiratory
symptoms and decreased lung function in workers exposed in a plant with a mean ammonia
concentration of 18.5 mg/m3, but not in workers in a second plant exposed to a mean concentration
of 4.9 mg/m3. Ballal etal. (1998) observed an increased prevalence of respiratory symptoms
among workers in one factory with exposures ranging from 2 to 27.1 mg/m3,7 but no increase in
another factory with exposures ranging from 0.02-7 mg/m3. A companion study by Ali etal. (2001)
also observed decreased lung function among workers exposed to higher cumulative ammonia
levels (>50 mg/m3-years), with an approximate 5-7% decrease in FVC % predicted and FEVi%
predicted.
These four studies addressed smoking by a variety of methods (e.g., adjustment for
smoking, exclusion of smokers, or stratification of the results by smoking status). Two of the
studies—Rahman et al. f2007) and Holness etal. f 19891—addressed other potential confounders
as appropriate. In particular, a high level of control of exposures in the facility studied by Holness
7This concentration range does not include exposures in the urea store (number of employees = 6; range of
ammonia concentrations = 90-130.4 mg/m3) because employees in this area were required to wear full
protective clothing, thus minimizing potential exposure.
This document is a draft for review purposes only and does not constitute Agency policy.
ES-3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
etal. (1989) was reported, suggesting a low potential for co-exposures. As discussed in more detail
in the Literature Search Strategy/Study Selection and Evaluation section, confounding by other
workplace exposures, although a potential concern, was unlikely to be a major limitation of these
studies.
Considerations in selecting the principal study for RfC derivation include the higher
confidence placed in the measures of ammonia exposure in Holness et al. (19891. evaluation of both
respiratory symptoms and lung function parameters in this study, and the fact that the estimate of
the no-observed-adverse-effect level (NOAEL) for respiratory effects of 13.6 mg/m3 from Holness et
al. (1989) was the highest of the studies with adequate exposure-response information. The
synthesis of findings from the full body of evidence demonstrates that there is a relationship
between ammonia exposure and respiratory effects. Although Holness etal. (19891 do not report
associations between ammonia exposure and respiratory effects, it is included in the body of
epidemiologic studies of industrial settings because it is informative of the levels below which
ammonia causes effects. These epidemiology studies include those with higher workplace
ammonia concentrations associated with respiratory effects (i.e., higher concentrations relative to
those reported by Holness etal. (198911 and for which LOAELs could be identified. The Holness et
al. Q9891 study is identified as the principal study for RfC derivation based on the quality of the
exposure data and other factors, as stated above.
In summary, the study of ammonia exposure in workers in a soda ash plant by Holness et al.
(19891 was identified as the principal study for RfC derivation, with support from Rahman et al
(20071. Ballal etal. (19981. and Ali etal. (20011. and respiratory effects were identified as the critical
effect The NOAEL, represented by an estimate of the 95% lower confidence bound of the mean
exposure concentration in the high-exposure group from the Holness etal. (19891 study, or
13.6 mg/m3, was used as the point of departure (POD) for RfC derivation. The NOAEL adjusted to
continuous exposure (NOAELadj) was 4.9 mg/m3.
An RfC of 0.5 (rounded) mg/m3 was calculated by dividing the POD (adjusted for
continuous exposure, i.e., NOAELadj) by a composite uncertainty factor (UF) of 10 to account for
potentially susceptible individuals in the absence of data evaluating variability of response to
inhaled ammonia in the human population.
Confidence in the Chronic Inhalation RfC
Study - medium
Database - medium
RfC - medium
Consistent with EPA's Methods for Derivation of Inhalation Reference Concentrations and
Application of Inhalation Dosimetry fU.S. EPA. 19941. the overall confidence in the RfC is medium
and reflects medium confidence in the principal study (adequate design, conduct, and reporting of
the principal study; limited by small sample size and identification of a NOAEL only) and medium
This document is a draft for review purposes only and does not constitute Agency policy.
ES-4 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
confidence in the database, which includes occupational, cleaner, agricultural, and human exposure
studies and studies in animals that are mostly of subchronic duration. There are no studies of
developmental toxicity, and studies of reproductive and other systemic endpoints are limited;
however, the likelihood of reproductive, developmental, and other systemic effects at the RfC is
considered small because it is well documented that ammonia is endogenously produced in humans
and animals, and any changes in blood ammonia levels at the POD would be small relative to
normal blood ammonia levels. Further, EPA is not aware of any mechanisms by which ammonia
can exert effects at the point of contact (i.e., respiratory system) that could directly or indirectly
impact tissues or organs distal to the point of contact
Susceptible Populations and Lifestages
Studies of the toxicity of ammonia in children that would support an evaluation of
childhood susceptibility are limited. Casas etal. (2013) and Loftus etal. (2015) reported evidence
of an association between ammonia exposure and decrements in lung function in children; however
these studies did not report information that would allow a comparison of children and adults.
A limited number of studies provides inconsistent evidence of greater respiratory
sensitivity to ammonia exposure in asthmatics f Loftus etal.. 2015: Petrova etal.. 2008: Sigurdarson
etal.. 2004: Preller etal.. 1995). Loftus etal. (2015) reported no increase in asthma symptoms and
medication use in asthmatic children living near animal feeding operations; however, ammonia
exposure was associated with lower FEVi.
Hyperammonemia is a condition of elevated levels of circulating ammonia that can occur in
individuals with severe diseases of the liver or kidney or with hereditary urea [CO(NH2)2] cycle
disorders. These elevated ammonia levels can predispose an individual to encephalopathy due to
the ability of ammonia to cross the blood-brain barrier; these effects are especially marked in
newborn infants. Thus, individuals with disease conditions that lead to hyperammonemia may be
more susceptible to the effects of ammonia from external sources, but there are no studies that
specifically support this susceptibility.
Key Issues Addressed in This Assessment
Comparison of Exhaled Ammonia to the RfC
Ammonia is generated endogenously in multiple organs and plays central roles in nitrogen
balance and acid-base homeostasis (Weiner etal.. 2014: Weiner and Verlander. 2013). Given its
important metabolic role, free ammonia is homeostatically regulated to remain at low
concentrations in blood (Souba. 1987). Elimination of ammonia occurs primarily in urine and
exhaled breath. Consideration was given to the presence of ammonia in exhaled air because the
range of ammonia concentrations in exhaled breath (0.009-2 mg/m3) overlaps the ammonia RfC
(0.5 mg/m3).
In general, the higher and more variable ammonia concentrations (0.03 to 2 mg/m3) are
reported in human breath exhaled from the mouth or oral cavity f Schmidt etal.. 2013: Smith etal..
2008: Spanel etal.. 2007a. b; Turner etal.. 2006: Diskin etal.. 2003: Smith etal.. 1999: Norwood et
This document is a draft for review purposes only and does not constitute Agency policy.
ES-5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
al.. 1992: Larson etal.. 19771. Ammonia concentrations measured in breath derived from oral
breathing largely reflect the production of ammonia via bacterial degradation of food protein in the
oral cavity or gastrointestinal tract, and can be influenced by diet, oral hygiene, age, and saliva pH.
In contrast, concentrations of ammonia in breath exhaled from the nose and trachea of humans
(0.0092-0.1 mg/m3) are lower than those in air exhaled from the mouth fSchmidt et al..
2013: Smith etal.. 2008: Larson etal.. 19771. and are generally lower than the RfC by a factor of five
or more. Concentrations in breath exhaled from the nose appear to better represent levels at the
alveolar interface of the lung and are more relevant to understanding systemic levels of ammonia
than breath exhaled from the mouth fSchmidt etal.. 2013: Smith etal.. 20081: however, neither
concentrations in breath from the mouth or nose can be used to predict blood ammonia
concentration or previous exposure to environmental (ambient) concentrations of ammonia.
Regardless of the source of expired ammonia (mouth or nose), the level of ammonia in
breath, even at concentrations that exceed the RfC, does not necessarily raise questions about the
appropriateness of the RfC. The exhalation of ammonia is a clearance mechanism for a product of
metabolism that is otherwise toxic in the body at sufficiently high concentrations. Thus, ammonia
concentrations in exhaled breath may be higher than inhaled concentrations. However, the
presence of ammonia in exhaled breath is not considered an uncertainty in the RfC.
This document is a draft for review purposes only and does not constitute Agency policy.
ES-6 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
LITERATURE SEARCH STRATEGY | STUDY
SELECTION AND EVALUATION
Literature Search and Screening Strategy
The literature search for ammonia was conducted in six online scientific databases,
including PubMed, Toxline, the Toxic Substances Control Act Test Submissions (TSCATS) database,
Web of Science (WOS), HERO8, and Toxcenter. The initial search was performed in March 2012
(PubMed, Toxline, TSCATS, HERO, and Toxcenter) and literature search updates were conducted in
March 2013 (PubMed, Toxline, TSCATS, HERO, and WOS) and September 2015 (PubMed, Toxline,
TSCATS, and WOS). Toxcenter is a database in which titles may be viewed for free after a fee-based
search, but full citations and abstracts are purchased. The use of Toxcenter was discontinued in
2013. No unique relevant hits were returned in the 2013 update search of HERO; therefore, this
search was not repeated in 2015. The detailed search approach, including the query strings, is
presented in Appendix B, Table B-l. This search of online databases identified approximately
~28,000 unique citations (after electronically eliminating duplicates).
The core computerized database searches were supplemented by a review of citations in
other national and international health agency documents (see Table B-2). The ATSDR (2004)
Toxicological Profile of Ammonia9 was used to identify toxicokinetic studies for ammonia. A search
of online chemical assessment-related websites was performed in 2012 and 2015; links to the
websites that were searched are provided in Table B-2. An additional focused search strategy was
also employed to obtain studies of cleaning and hospital workers to address a new area of research
identified during the 2013 literature search update. This strategy involved a manual reference list
review of several seminal studies published in 2012 (see Appendix B, Table B-2). In addition,
electronic forward searches were conducted in WOS in 2013 and 2015, using a methods paper
describing the development of a job exposure matrix focusing on asthma as a health outcome
8Health and Environmental Research Online (HERO) is a database of scientific studies and other references
used to develop EPA 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 1.6 million
scientific references, including articles from the peer-reviewed literature. New studies are added
continuously to HERO. For each IRIS assessment, a HERO project page is created that stores all citations
identified from that chemical-specific literature search. These citations may be organized using various tags
to indicate if the citations are used in the assessment and how they are categorized.
9Portions 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 f ENREF 121 and the references cited in that document, as part of a collaborative effort in the
development of human health toxicological assessments for the purposes of making more efficient use of
available resources and sharing scientific information.
This document is a draft for review purposes only and does not constitute Agency policy.
LS-1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
(Kennedy et al.. 20001. The disposition of studies obtained from the manual backward and
electronic forward searches is presented in Table B-3.
In Federal Register notices announcing annual IRIS agendas and on the IRIS website, EPA
encouraged the public to submit information on IRIS chemicals throughout the assessment
development process, and specifically requested that the public submit additional data to support
development of the ammonia assessment on December 21, 2007 and November 2, 2009 fU.S. EPA.
2009a. 20071. No public submissions were received in response to these calls for data.
Figure LS-1 depicts a summary of the literature search and screening process and the
number of references included or excluded at each step. In 2012, the initial literature search was
conducted in core computerized databases. These citations were electronically screened in an
EndNote database using a set of terms intended to prioritize "on-topic" references for title and
abstract review. The electronic screening process created two broad categories: one of all citations
that contain (in title, abstract, or keywords) at least one inclusion term related to health outcomes,
epidemiological or toxicological study design, absorption/distribution/metabolism/excretion
(ADME) or toxicokinetics, or mechanistic information (see Appendix B, Table B-4), and one that did
not contain any of the terms. Some of the electronic inclusion terms listed in Table B-4 are generic
(i.e., not chemical specific) and are intended to capture health effect studies of any type. Other
terms are specific to ammonia and are based on previous knowledge of health effects and possible
mechanisms of toxicity summarized in other health agency review documents (see Appendix A).
Citations that did not contain at least one inclusion term in Table B-4 (i.e., excluded by the
electronic screening) were subjected to a quality control check to verify that relevant references
were not missed. Specifically, a random sample (approximately 10%) of the electronically excluded
citations were subjected to title and abstract review by a toxicologist to confirm that the electronic
screening process produced acceptable results (i.e., no relevant citations were inadvertently
missed). Relevant items were added to the HERO project page for ammonia and retrieved for full-
text review. The results from the updated literature searches performed in March 2013 and
September 2015 were not screened electronically in EndNote. All titles and abstracts obtained
from these search updates were reviewed manually by a toxicologist
This document is a draft for review purposes only and does not constitute Agency policy.
LS-2 DRAFT—DO NOT CITE OR QUOTE
-------
PubMed
n=16,644
Toxline
n=2,550
Database Searches
(see Table B-l for query strings and date limits)
/-
TSCATS
n=58
r
wos
n=3,816
Toxcenter
n=2,591
HERO
n=5,410
n=~28,000 (After duplicates removed electronically and manually)
Additional Search Strategies
(see Table B-2 for methods and results)
Excluded by electronic screen
n=~13,000
Combined Dataset
n=~15,000
>
Supplementary Studies
Excluded (n=~14,750)
See exclusion criteria in Table LS-1
Secondary Literature
9 Reviews and regulatory documents
Title, Abstract, and Full Text Screening
Health Effect/Toxicity Studies
(see inclusions criteria in Table LS-1)
26 Human studies (including studies in
occupational, agricultural, and cleaner
settings)
17 Animal studies
Supplementary Health Effect/
Toxicity Studies
(see inclusion criteria in Table LS-1)
35 Case reports
13 Controlled human exposure studies
27 Inhalation acute and short-term studies
Physical-Chemical/Mechanistic/
Toxicokinetic Studies
13 Physical-chemical property studies
2 Mode of action studies
76 Toxicokinetic studies
Figure LS-1. Summary of literature search and screening process for
ammonia.
This document is a draft for review purposes only and does not constitute Agency policy,
LS-3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Manual screening of titles/abstracts and full text was accomplished using a set of inclusion
and exclusion criteria to identify sources of primary human health effects data and sources of
primary data that supplement the assessment of ammonia health effects (i.e., bottom boxes in
Figure LS-1). The inclusion/exclusion criteria that were used prior to peer review are presented in
Table LS-1. Manual screening of the post-peer review literature search update (i.e., September
2015) was performed using more stringent inclusion and exclusion criteria to capture studies that
would impact the credibility of the assessment's conclusions consistent with EPA's IRIS Stopping
Rules (http://www.epa.gov/sites/production/files/2014-06/documents/iris_stoppingrules.pdf].
For ammonia, those references identified in the post-peer review literature search that were
considered for inclusion in hazard identification were in vivo animal toxicity and epidemiology
studies. No additional in vivo animal toxicity studies of ammonia were identified in the post-peer
review search. The disposition of epidemiology studies obtained from the post-peer review
literature search update (i.e., September 2015) is provided in Table B-5.
Specific inclusion/exclusion criteria were not applied in identifying sources of mechanistic
and toxicokinetic data. Because ammonia is produced endogenously and serum ammonia levels are
measured in certain disease states, the toxicokinetics literature is large and complex; relevant
toxicokinetic studies for ammonia were initially identified using the ATSDR (2004) Toxicological
Profile of Ammonia and supplemented by more recent studies identified in literature search
updates. The number of mechanistic studies identified for ammonia was not large, and therefore all
mechanistic studies were included.
Table LS-1. Inclusion/exclusion criteria for inhalation health effect/toxicity
studies (pre-peer review)*
Inclusion criteria
Exclusion criteria
Population
• Humans, including occupational workers,
livestock workers and those in close
proximity to agricultural operations,
hospital workers/cleaners and volunteers
• Standard mammalian animal models,
including rat, mouse, hamster, rabbit,
guinea pig, monkey, dog
• Pigs
• Ecological species/ecosystem effects
• Nonmammalian species
• Agricultural species/livestock (except pigs)
Exposure
• Exposure is to ammonia by the inhalation
route (any duration)
• Exposure is measured as a concentration
in air
• Exposure is in vivo
• Not chemical specific (i.e., not ammonia-
specific)
• Animal studies: exposure is to a mixture
only
• Human studies: exposure is inferred but
not measured (e.g., some cleaning and
hospital worker studies)
• Exposure by oral, dermal, injection or
instillation routes
• Studies of quaternary ammonia
Outcome
• One or more of the following health effect
endpoints is evaluated: effects on the
• No health outcome evaluated
• Pathogenic effects of H. pylori infection
This document is a draft for review purposes only and does not constitute Agency policy.
LS-4 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Inclusion criteria
Exclusion criteria
cardiovascular, dermal/integumentary,
endocrine, gastrointestinal, immune,
musculoskeletal, nervous, reproductive,
respiratory, hepatic, or renal (urinary)
systems; effects on the eyes, survival,
growth, or development
Other
• Review article or abstract only (i.e., no
primary data)
• Environmental fate and transport of
ammonia
• Analytical methods for measuring
ammonia in environmental media, and
use in sample preparations and assays
• Study of physical-chemical properties
• Study of in vitro or in vivo toxicokinetics
• Study of in vitro or in vivo mechanistic
endpoints
• Other studies not on topic and not
captured by other exclusion criteria
* Reviews and regulatory documents were retained as Secondary Literature. Studies that provided primary
information on the physical-chemical properties, mode of action, or toxicokinetics of ammonia were also
retained, but were not screened as sources of health effect/toxicity information for ammonia.
The results of the pre- and post-peer review literature screening are described below and
graphically in Figure LS-1:
• 43 references (including 26 human studies and 17 animal studies) were identified as
studies with health effects data and were considered for data extraction to evidence
tables and exposure-response arrays.
• Supplementary health effect/toxicity studies included 35 case reports, 13 acute-
duration controlled human exposure studies, and 27 acute or short-term animal studies.
Information from these studies was not extracted into evidence tables; however, these
studies were considered as supplementary studies for assessing ammonia health effects.
• 91 studies were identified as physical-chemical, mode of action, or toxicokinetic studies,
including 13 studies of physical-chemical properties, 2 studies providing mode of action
information, and 76 toxicokinetic studies. Information from these studies was not
extracted into evidence tables; however, these studies were considered as
supplementary studies for assessing ammonia health effects (e.g., consideration of
toxicokinetic information in assessing the health effects literature).
• Nine reviews or regulatory/health assessment documents were identified as secondary
literature. These references were retained as additional resources in developing the
Toxicological Review.
• More than 27,000 references were identified as not pertinent to an evaluation of the
inhalation health effects of ammonia. Approximately 13,000 were excluded by
This document is a draft for review purposes only and does not constitute Agency policy.
LS-5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
electronic screening (see Table B-4) and approximately 14,750 were excluded by
manual screening (see Table LS-1 for exclusion criteria).
Study Selection and Evaluation
Selection of studies for inclusion in the Toxicological Review was based on consideration of
the extent to which the study was informative and relevant to the assessment and general study
quality considerations. In general, the relevance and scientific quality of the available studies was
evaluated as outlined in the Preamble and in EPA guidance (i.e., A Review of the Reference Dose and
Reference Concentration Processes (U.S. EPA. 20021 and Methods for Derivation of Inhalation
Reference Concentrations and Application of Inhaled Dosimetry fU.S. EPA. 199411. The scientific
considerations used to evaluate and select studies and the relevance of these studies to the
assessment are described in the section below.
Considerations for evaluation of epidemiology studies
Case reports are often anecdotal and describe unusual or extreme exposure situations,
providing little information that would be useful for characterizing chronic health hazards.
Ammonia case studies were only briefly reviewed; representative citations from the collection of
case reports are provided as supplemental information in Appendix C, Section C.2.4. Similarly,
acute controlled human exposure studies would not be useful for characterizing chronic health
effects; these studies were therefore briefly reviewed and are provided as supplemental
information in Appendix C, Section C.2.3.
Epidemiology studies of chronic exposure to ammonia have primarily focused on industrial
worker populations, workers exposed to ammonia as a cleaning or disinfectant product, and those
exposed in an agricultural setting. There is considerable variation in population characteristics,
level and pattern of exposure, and potential confounders across the three categories of studies.
Evaluations of the observational epidemiology studies of industrial worker populations and
workers exposed to ammonia as a cleaning or disinfectant product identified in Figure LS-1 (i.e., the
studies considered most informative for evaluating ammonia toxicity from chronic exposure) are
provided in Appendix B (Tables B-6 to B-8). The process used to evaluate these studies addressed
aspects relating to the selection of study participants, exposure parameters, outcome measurement,
confounding, and statistical analysis. As discussed below, studies of populations exposed in
agricultural settings were considered to be supporting material because of the variety of potential
co-exposures in these studies (including dust, endotoxin, mold, and disinfectant products). The
process for evaluating studies in an agricultural setting considered the same five aspects (selection
of study participants, exposure parameters, outcome measurement, confounding, and statistical
analysis); however, specific study evaluation tables were not provided in Appendix B for this set of
studies.
For study evaluation purposes, EPA differentiated between "major" limitations, defined as
biases or deficiencies that could materially affect the interpretation of the study, and "minor"
limitations, defined as limitations that are not likely to be severe or to have a substantive impact on
This document is a draft for review purposes only and does not constitute Agency policy.
LS-6 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
the results. These categories are similar to the "serious risk of bias" and "moderate risk of bias"
categories, respectively, described by Stearne etal. f 20141 in the Cochrane Collaborative
Assessment Tool for non-randomized studies of clinical interventions. Identification of major
limitations in the epidemiology studies of populations exposed in industrial, cleaning, and
agricultural settings is included in the broader evaluation of study quality below. Uninformative
studies are also noted.
Studies of Industrial Settings
Selection of study participants
All of the studies were cross-sectional analyses in occupational settings. The workers were
healthy enough to remain in the work area for a considerable time; with one exception, mean
duration ranged from 52 months to 16 years. One study (Bhatand Ramaswamv. 19931 grouped
workers into those exposed for up to 10 years and those with more than 10 years of exposure; a
minimum exposure duration was not provided. As in inherent property of occupational studies,
these designs may result in a "healthy worker" bias. In addition, the workers in these studies are
not representative of the general population, as they do not include children and only one study of
ammonia exposure in hair salons included women fNemer etal.. 20151. These aspects of the study
design may result in an underestimate of the risk of health effects of ammonia exposure, as the
worker population may not exhibit health effects (such as decreased lung function or increased
prevalence of respiratory symptoms) to the same degree that would be seen in the general
population under the same conditions. In addition to the "healthy worker" effect, the Nemer etal.
(20151 study exhibited a potential selection bias in the controls due to differences in recruitment
(self-selected based on interest) or workload.
Exposure parameters
Exposure methods differ across these occupational studies, which makes comparison of
ammonia measurements among the studies difficult Spectrophotometric absorption measures of
areas samples (Ali etal.. 2001: Ballal etal.. 19981 are not directly comparable to direct-reading
diffusion methods Rahman et al. (20071 or electrochemical sensors methods (Nemer etal.. 20151
used to analyze personal samples. Nor are they comparable to the NIOSH-recommended protocol
for personal sampling and analysis of airborne contaminants fHolness etal.. 19891. In the study
by Rahman et al. (20071. exposure concentrations were determined by both the Drager tube and
Drager PAC III methods. The Drager tube method yielded concentrations of ammonia in the two
plants studied that were approximately fourfold higher than the concentrations obtained by the
Drager PAC III method; a strong correlation between measurements by the two methods was
reported. Rahman etal. (20071 stated that their measurements indicated only relative differences
in exposures between workers and production areas, and did not identify one analytical measure as
the more valid of the two. Based on communication with technical support at Drager Safety Inc.
(Bacom and Yanoskv. 20101. EPA considered the PAC III instrument to be a more sensitive
monitoring technology than the Drager tubes. Ammonia concentrations based on the PAC III
This document is a draft for review purposes only and does not constitute Agency policy.
LS-7 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
method were also in line with concentrations reported in other studies. Therefore, exposure levels
based on PAC III air measurements of ammonia were used in the current health assessment to
characterize the exposure-response relationship in the Rahman et al. f20071 study.
In the Abdel Hamid and El-Gazzar (1996) study, no direct measurement of ammonia
exposure was made; blood urea was used as a surrogate measure of ammonia exposure. The
correlation of blood urea with ammonia is not reported by the authors. EPA considered this a
major limitation of this study, based on other data indicating no correlation between ammonia
levels in air and serum urea levels in a study of six groups of workers with varying types of
exposure (Giroux and Ferrieres. 19981. No exposure measurements of ammonia were used in the
study by Bhat and Ramaswamv Q9931. EPA considered the lack of exposure measure in this study
to be a major limitation. In the Nemer etal. (2015) study, the measurement device had limited
specificity for measuring ammonia relative to other gases and therefore could have produced false
positive results in the presence of other gases. In addition, few exposure measurements were made
in the Nemer etal. (2015) study. EPA considered the limited specificity for measuring ammonia,
the limited number of exposure measurements, as well as possible misclassification of exposure in
the Nemer etal. (2015) study to be major limitations.
Outcome measurement
Assessment of respiratory symptoms in Rahman etal. (2007). Ballal etal. f 19981 Holness et
al. (1989). and Nemer etal. (2015) was based on four different questionnaires; each of these,
however, is a standardized, validated questionnaire. Self-reporting of types and severity of
respiratory symptoms could be biased by the knowledge of exposure, for example, in studies
comparing factory workers to office workers. EPA evaluated this non-blinded outcome assessment
as a potential bias. In each of these studies, comparisons were made across exposure categories
among the exposed; EPA concluded that the non-blinded outcome assessment as a potential bias is
unlikely in these types of comparisons. One study also compared exposed to nonexposed, and
observed little differences in symptom prevalence between these groups (Holness etal.. 1989).
Thus, EPA concluded that the non-blinded outcome assessment was not a major bias in this analysis
either. Assessment of lung function was performed by standard spirometry protocols in five
studies (Nemer etal.. 2015: Rahman et al.. 2 0 0 7: Ali etal.. 2001: Bhat and Ramaswamv.
1993: Holness etal.. 19891. EPA did not consider any of these procedures for assessing lung
function to be a source of bias.
Confounding
Co-exposures to other ambient chemicals in urea fertilizer factories included inorganic
gases (nitrogen dioxide and sulfur dioxide) and dust In one of these studies f Rah man et al.. 20071.
nitrogen dioxide was measured concurrently with ammonia and found to be below detection limits
for all areas (urea plant, ammonia plant, and administration area). The other urea fertilizer studies
(Ali etal.. 2001: Ballal etal.. 1998: Abdel Hamid and El-Gazzar. 1996) did not describe potential co-
exposures. [It appears from the exposure measurements that the plant in Ali etal. (2001) is
This document is a draft for review purposes only and does not constitute Agency policy.
LS-8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
"Factory A" in Ballal etal. (1998)]. In the fertilizer plant in Bhat and Ramaswamv (1993). co-
exposures are not discussed, but the workers are grouped based on different parts of the plant
(ammonia, urea, and diammonium phosphate); effects observed with respect to lung function tests
were similar in magnitude, albeit slightly stronger, in the ammonia plant workers compared with
the urea plant workers. One study was conducted in a soda ash production plant fHolness etal..
19891. No measurements of co-exposures were described in this study, but the authors note the
high level of control of exposures (resulting in low ammonia levels) in this facility. Because of the
lack of demonstration of co-exposures correlated with ammonia levels in these studies, and lack of
demonstration of stronger associations between potential co-exposures and respiratory outcomes,
EPA concluded that confounding by other workplace exposures, although a potential concern, was
unlikely to be a major limitation for the urea plant and soda ash plant studies. However, in a study
of ammonia exposure among hairdressers fNemer etal.. 20151. co-exposures to other workplace
contaminants (such as persulfates and paraphenylenediamine) were not measured or controlled
for in the analysis; therefore, possible confounding is considered to be a limitation in this study.
The analyses of respiratory symptoms and lung function may also be confounded by
smoking. In six studies, analyses accounted for smoking as follows: the analysis included either an
adjustment for smoking (Rahman et al.. 2 0 0 7: Holness et al.. 19891. the exclusion of smokers
(Nemer etal.. 2015: Bhat and Ramaswamv. 1993). or stratification of the results by smoking status
fAli etal.. 2001: Ballal etal.. 19981. Thus, EPA did not consider potential confounding by smoking
to be a major limitation of these studies.
Ammonia is present in both tobacco and cigarette smoke fCallicuttetal.. 2006). Typical
concentrations of ammonia in commercial U.S. tobacco blends range from 0.02-0.4% (Seeman and
Carchman. 2008). Thus, there is some potential for additional exposure to ammonia associated
with use of ammonia-containing tobacco products and/or inhalation of tobacco smoke. This finding
reinforces the importance of controlling for smoking in the analyses of the respiratory symptoms
and lung function. EPA did not consider potential confounding by smoking of ammonia-containing
tobacco or by inhaling tobacco smoke to be a major limitation of these studies because smoking as a
potential confounder was adequately addressed in the studies that examined effects on the
respiratory system.
Information on smoking habits and use of alcohol (an exposure potentially affecting liver
function tests) was not documented in the study of liver function by Abdel Hamid and El-Gazzar
(1996). The lack of information and potential failure to control for these confounders is considered
a major limitation.
Statistical analysis
EPA considered the statistical analysis in the epidemiological studies fNemer etal..
2015: Rahman etal.. 2007: Ali etal.. 2001: Ballal etal.. 1998: Abdel Hamid and El-Gazzar.
1996: Bhatand Ramaswamv. 1993: Holness etal.. 19891 to be adequate and appropriate. Although
the type of statistical testing was not specified in Abdel Hamid and El-Gazzar (1996). the results
were presented in sufficient detail to allow interpretation of the data and analysis. Sample size, an
This document is a draft for review purposes only and does not constitute Agency policy.
LS-9 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
important consideration with respect to statistical power, was also considered. EPA noted the
small number of exposed workers and low levels of exposure in the study by Holness etal. (1989)
as limitations that could result in "false negative" results (i.e., a test result indicating a lack of
association, whereas a positive association between exposure and a health effect exists).
Identification of uninformative studies
The study by Abdel Hamid and El-Gazzar (1996) was determined to have major limitations.
Air concentrations of ammonia were not directly measured, and the use of blood urea has not been
established as a reliable surrogate of ammonia exposure. Further, the lack of information on
smoking and alcohol use, factors that could affect liver function, in a study intended to examine the
association between liver function and ammonia exposure, was considered a significant flaw.
Therefore, Abdel Hamid and El-Gazzar T19961 was not further considered in this assessment
Major limitations were also identified in the Nemer etal. (2015) study: potential selection
bias in the control group due to differences in recruitment (self-selected based on interest in the
study) or workload; limited specificity of the analytical method used to measure ammonia (i.e.,
potential for false positives from other gases); and failure to control for confounders. In addition,
the study used small sample sizes and only a single measurement of ammonia for each location
(which may not have been representative of workplace exposures). Therefore, the Nemer etal.
T2015) study was deemed to be uninformative and was not further considered in this assessment.
Studies of Health Care and Cleaning Settings
Selection of study participants
EPA also evaluated the studies that examined exposure to ammonia when used as a
cleaning or disinfectant product EPA noted the potential for the "healthy worker" bias arising from
movement out of jobs by affected individuals in most of these studies (Le Moual et al.. 2008). This
issue was less of a concern in the study by Zocketal. (2007). which was conducted in a general
(non-occupational) population sample, focusing on cleaning activities in the home. In a birth cohort
that evaluated the association between exposure to cleaning products and children's respiratory
health (Casas etal.. 2013). 35% of the recruited population were excluded because information on
the use of cleaning products and/or respiratory tests was not available, representing a potential
study limitation. However, the authors of this study noted that the children included were not
different from those excluded regarding most study characteristics (sex atopy, asthma, parental
asthma and parental smoking).
Exposure parameters
None of these studies used a direct measure of ammonia exposure in the analysis,
precluding interpretation of the results in relation to an absolute level of exposure. The limited
data available concerning exposure levels in cleaning scenarios found median exposures of 0.6 to
5.4 ppm (0.4 to 3.8 mg/m3), with peaks exceeding 50 ppm (35 mg/m3), in a small study (n = 9)
using personal samples during a domestic cleaning session (Medina-Ramon etal.. 2005). Although
This document is a draft for review purposes only and does not constitute Agency policy.
LS-10 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
an absolute level of exposure is not available, the relative ranking of exposure used in these studies
does allow examination of risk by relative levels of exposure.
Key considerations regarding the validity of the exposure measures are the specificity of the
classification and the extent to which classification could be influenced by knowledge of the disease
or symptoms under study. Methodological research has reported underestimation of self-reported
exposure to specific products by health care workers, and differential reporting by disease status
(i.e., asthma) for self-reported use of cleaning products in patient care, but not in instrument
cleaning or building materials (Donnav etal.. 2011: Delclos etal.. 2009: Kennedy et al.. 20001. Two
of these studies used an exposure assessment protocol that incorporated an independent, expert
review, blinded to disease status fDumas etal.. 2012: Lemiere etal.. 20121: one study collected
exposure information using a 2-week daily diary (Medina-Ramon et al.. 20061 and one study (Casas
etal.. 20131 developed a composite exposure score based on an interviewer-led questionnaire
concerned with the frequency of use and number of products used. EPA considered these to be the
strongest of the exposure protocols used within this set of studies.
Outcome measures
Six of the studies in this set of studies used standard protocols for the assessment of
respiratory symptoms in epidemiological studies (Casas etal.. 2013: Arif and Delclos. 2012: Dumas
etal.. 2012: Vizcava etal.. 2011: Zock etal.. 2007: Medina-Ramon etal.. 20051. and one study
included a clinical assessment protocol designed specifically for the assessment of occupational
asthma (Lemiere etal.. 20121. Details of the specific questions were provided, and EPA did not
consider any of these methods to be a limitation in terms of specificity of the outcome. The study
by Medina-Ramon et al. (20061 collected information on daily respiratory symptoms in a two-week
diary, and also trained the participants to measure peak expiratory flow three times daily. A
potential limitation in the Casas etal. (20131 study was the lack of information about the reliability
of the pulmonary function measures.
Confounding
All of these studies addressed the potential for smoking to act as a confounder in the
analysis. Two of the studies reported relatively weak correlations between ammonia and other
products assessed fZock etal.. 2007: Medina-Ramon etal.. 20051 and one study reported stronger
associations with ammonia than with bleach (Dumas et al.. 20121. Based on this information, EPA
did not consider potential confounding to be a major limitation of this set of studies.
Statistical analysis
EPA considered the statistical analysis in this set of studies to be appropriate. One study,
however, was limited in terms of the level of detail provided pertaining to the results for ammonia
from multivariate models (Medina-Ramon etal.. 20051.
This document is a draft for review purposes only and does not constitute Agency policy.
LS-11 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Studies of Agricultural Settings
Selection of study participants
EPA also evaluated a set of studies conducted among livestock farmers and one study of
asthmatic children in close proximity to animal feeding operations fLoftus etal.. 20151. As with the
other occupational studies discussed above, the selection of sensitive individuals out of the
workforce ("healthy worker bias") would be a potential bias in cross-sectional studies of livestock
farmers.
Exposure parameters
Among the studies examining pulmonary function, one study collected 24-hour air sampling
from 14 ammonia monitoring devices located outside the home of a subset of the participants every
6 days for at least 3 months during the air monitoring period fLoftus etal.. 20151. two studies used
area-based exposure sampling in animal confinement buildings fMonso etal.. 2004: Zeida etal..
19941. one study used area samples taken in conjunction with specific tasks and calculated a
personal exposure measure taking into account duration spent in specific locations and tasks
(Heederik etal.. 19901. four studies collected personal samples over a workshift fDonham etal..
2000: Reynolds etal.. 1996: Preller etal.. 19951. or an unspecified time period fDonham etal..
19951. and two studies used colorimetric tubes, which are generally less precise, to measure
ammonia exposure fMonso etal.. 2004: Zeida etal.. 19941. EPA considered the use of the area-
based samples without consideration of exposure duration to be limitations of the studies by Zeida
etal. (19941 andMonso etal. (20041.
Outcome measures
All of the studies reported using a standard spirometric technique; one study fLoftus etal..
20151 used twice daily home lung function measurements taken by the test subject; four studies
compared two measures per individual (i.e., pre- and post-shift) fMonso etal.. 2004: Donham etal..
2000: Reynolds etal.. 1996: Heederik etal.. 19901: and two studies used a single pulmonary
function measure, adjusted for height, age, and smoking variables (Preller etal.. 1995: Zeida etal..
19941. EPA did not consider any of these outcome measures to be limitations in these studies,
although the self-administered spirometry testing in the Loftus etal. (20151 study is a potential
limitation.
Confounding
Six of these studies addressed confounding in some way. Four studies controlled for co-
exposures (e.g., endotoxin, dust, disinfectants) (Melbostad andEduard. 2001: Reynolds etal..
1996: Donham etal.. 1995: Preller etal.. 19951. one study noted only weak correlations (i.e.,
Spearman r < 0.20) between ammonia and dust or endotoxin fDonham et al.. 20001. and one study
observed associations with ammonia but not with endotoxin or dust measures f Heederik etal..
19901. Three studies did not address confounding fLoftus etal.. 2015: Monso etal.. 2004: Zeida et
al.. 19941.
This document is a draft for review purposes only and does not constitute Agency policy.
LS-12 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Based on these considerations, EPA considered the studies by Reynolds etal. (19961. Preller
etal. (19951. Donham etal. (20001. Donham etal. (19951. and Heederik etal. (19901 to be the
methodologically strongest studies of this set
Based on the evaluation of the epidemiology studies of ammonia in terms of selection of
study participants, exposure parameters, outcome measurement, confounding and statistical
analysis, the studies listed in Table LS-2 were selected for data extraction into evidence tables in
Chapter 1.
Table LS-2. Summary of epidemiology database
Study setting
Reference
Industrial
Rahman et al. (2007)
AN etal. (2001)
Ballaletal. (1998)
Bhat and Ramaswamv (1993)
Holness et al. (1989)
Cleaning
Casas et al. (2013)
Arif and Delclos (2012)
Dumas et al. (2012)
Lemiere (2012)
Vizcaya (2011)
Zock (2007)
Medina-Ramon et al. (2006)
Medina-Ramon et al. (2005)
Agricultural
Loftus et al. (2015)
Monso et al. (2004)
Melbostad and Eduard (2001)
Donham et al. (2000)
Reynolds et al. (1996)
Donham et al. (1995)
Preller et al. (1995)
Choudat et al. (1994)
Zeida et al. (1994)
Crook et al. (1991)
Heederik et al. (1990)
Considerations for evaluation of animal studies
Repeat-exposure toxicity studies of ammonia in experimental animals were evaluated using
the study quality considerations outlined in the Preamble and discussed in various U.S. EPA
guidance documents (U.S. EPA. 2005. 2002.19941. including consideration of aspects of design,
This document is a draft for review purposes only and does not constitute Agency policy.
LS-13 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
conduct, or reporting that could affect the interpretation of results, overall contribution to the
synthesis of evidence, and determination of hazard potential. The objective was to identify the
stronger, more informative studies based on a uniform evaluation of quality characteristics across
studies of similar design.
Additionally, a number of general questions, presented in Table LS-3, were considered in
evaluating the animal studies. Much of the key information for conducting this evaluation can be
determined based on study methods and how the study results were reported.
Table LS-3. Considerations and relevant experimental information for
evaluation of experimental animal studies
Methodological
feature
Considerations
(relevant information extracted into evidence tables)
Test animal
Suitability of the species, strain, sex, and source of the test animals
Experimental design
Suitability of animal age/lifestage at exposure and endpoint testing; periodicity and
duration of exposure (e.g., hrs/day, days/week); timing of endpoint evaluations; sample
size and experimental unit (e.g., animals; dams; litters)
Exposure
Characterization of test article source, composition, purity, and stability; suitability of
the control (e.g., vehicle control); documentation of exposure techniques (e.g.,
chamber type); verification of exposure levels (e.g., consideration of homogeneity,
stability, analytical methods)
Endpoint evaluation
Suitability of specific methods for assessing the endpoint(s) of interest
Results presentation
Data presentation for endpoint(s) of interest (including measures of variability) and for
other relevant endpoints needed for results interpretation (e.g., decrements in body
weight in relation to organ weight)
Information relevant to study evaluation is reported in evidence tables and was considered
in the synthesis of evidence. Discussion of study strengths and limitations (that ultimately
supported preferences for the studies and data relied upon) were included in the text where
relevant. The general finings of this evaluation are presented in the remainder of this section.
Study evaluation considerations that are outcome specific are discussed in the relevant hazard
section in Section 1.2.
Test animal
The ammonia database consists of toxicology studies conducted in rats (F344, Sprague-
Dawley, Long-Evans, Sherman, Wistar), mice (0F1, Swiss albino), New Zealand white rabbits,
guinea pig (Princeton-derived, Hartley), beagle dog, squirrel monkey, and pig (several strains). The
species and strains of animals used are consistent with those typically used in laboratory studies,
and all were considered relevant to assessing the potential human health effects of ammonia. The
species, strain, and sex of the animals used in the experimental studies were recorded in the
evidence tables. The Anderson et al. (1964a) and Weatherbv (1952) guinea pig studies provided no
information on the strain of the test animal; this is considered a minor limitation of these studies.
This document is a draft for review purposes only and does not constitute Agency policy.
LS-14 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Experimental design
General aspects of study design and experimental design were evaluated to determine if
they were appropriate for evaluation of specific endpoints. Key features of the experimental
design, including the periodicity and duration of exposure and sample sizes, were summarized in
the evidence tables in Chapter 1.
A single exposure group was used in a number of the general toxicity studies (Gaafar etal..
1992: Broderson et al.. 1976: Doigand Willoughbv. 1971: Anderson et al.. 1964a: Weatherbv.
19521. and in about half of the studies that examined immune endpoints fHamilton et al..
1999: Hamilton etal.. 1998: Schoeb etal.. 1982: Richard et al.. 19781. Use of a single exposure
group limits the extent to which conclusions about a dose-response relationship can be drawn.
Sample size was not a basis for excluding a study from consideration, as studies with small
numbers of animals can still inform the consistency of effects observed for a specific endpoint.
Nevertheless, the following studies with small sample sizes were considered relatively less
informative: Anderson etal. f!964al studies in the mouse (4 animals/exposure interval) and
guinea pigs (2 animals/exposure interval); the Weatherbv (19521 study in guinea pigs (2 control
and 4 exposed animals/exposure interval); and the Coon etal. Q9701 studies in the rabbit (3
animals/group), monkey (3 animals/group), and dog (2 animals/group).
Exposure
Because inhalation toxicity studies can be technically difficult to perform, particular
attention was paid to each study's exposure methods and documentation for assurance that the
animals were properly exposed to gaseous ammonia. Exposure evaluation focused on those studies
that reported effects on the respiratory system. Of the studies evaluated for exposure quality, six
provided information on generation method, analytical method used to measure ammonia
concentrations, analytical chamber concentrations, and chamber type; exposure characterization
for these studies was considered robust (Broderson etal.. 1976: Coon etal.. 19701 (Done etal..
2005: Diekman etal.. 1993: Doigand Willoughbv. 1971: Stombaugh etal.. 19691. Studies
by Anderson et al. (1964a) and Curtis et al. f 19751 failed to report analytical chamber
concentrations, but otherwise exposures were considered to be adequately characterized.
Exposure characterization in two studies fGaafar etal.. 1992: Weatherbv. 19521 was considered
poor because the studies failed to report analytical chamber concentrations, analytical method, and
the type of inhalation chamber used. One of these two studies f Gaafar etal.. 19921 also failed to
describe how gaseous ammonia was generated from a 12% "ammonia solution."
Endpoint evaluation
Respiratory system and other noncancer effects were largely evaluated based on clinical
signs (in the case of respiratory system effects) and histopathologic examination. All studies
identified the tissues taken for histopathologic examination; however, the extent to which
histopathologic methods were described varied across studies. Because histopathology is
This document is a draft for review purposes only and does not constitute Agency policy.
LS-15 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
considered a relatively routine measure, limited reporting of methodologic details was not
considered a significant study deficiency.
Essentially all studies examined tissues from the lung and approximately half of the studies
examined upper respiratory tissues. This is a concern because the highest exposure would have
been to the upper respiratory tract due to the fact that ammonia is both water soluble and highly
reactive. Gaafar etal. f19921 examined only the nasal mucosa. Tissues from other organs remote
from the point of entry were inconsistently examined. Coon etal. (1970) examined sections from
the heart, lung, liver, kidney, and spleen from all surviving monkeys, dogs, and rabbits, but from
approximately half of the surviving guinea pigs and rats only; this incomplete histopathological
investigation of guinea pigs and rats is considered a limitation. Anderson etal. (1964a) examined
only the liver and spleen from exposed mice and guinea pigs. Broderson etal. (1976) examined
sections from the liver, kidney, adrenal gland, pancreas, testicle, spleen, mediastinal nodes, and
thymus. Curtis etal. (1975) noted that "visceral organs" were taken at necropsy for subsequent
histopathologic examination, but provided no further details. Weatherbv (1952) examined the
heart, liver, stomach, small intestines, spleen, kidney, and suprarenal gland, but only reported
limited incidence and severity information for the exposed and control guinea pigs. The extent of
histopathological examination of the tissues was taken into consideration in evaluating animal
findings.
Methodological considerations related to immune-specific endpoints are discussed in
Section 1.2.2.
Results presentation
The majority of studies reported only limited qualitative results. With the exception
of Broderson etal. Q976I none provided information on the incidence of histopathologic lesions.
In summary, relatively few repeat-dose toxicity studies of inhaled ammonia in experimental
animals are available. The majority of these studies come from the older toxicological literature
and were generally limited in terms of study design (e.g., small group sizes), documentation of
methods, and reporting of results. Nevertheless, no study was considered sufficiently flawed as to
be uninformative. Therefore, all in vivo animal toxicity studies, as listed in Table LS-4, were
considered in hazard identification and data extraction to evidence tables.
Table LS-4. Summary of experimental animal database
Reference and study description (duration, route, species/strain)
Done et al. (2005) — 5-week inhalation study in pigs (several breeds)
Andreasen et al. (2000b) -- 63-day inhalation study in Landrace X large white pigs
Hamilton et al. (1999) - 4-week inhalation study in large white pigs
Hamilton et al. (1998) - 14-day inhalation study in large white pigs
Diekman et al. (1993) - 6-week inhalation study in crossbred gilts (female pigs)
This document is a draft for review purposes only and does not constitute Agency policy.
LS-16 DRAFT—DO NOT CITE OR QUOTE
-------
Reference and study description (duration, route, species/strain)
Gaafar et al. (1992) - 8-week inhalation study in white albino mice
Gustin (1994) - 6-day inhalation study in pigs
Manninen and Savolainen (1989) - 5-day inhalation study in Wistar rats*
Manninen et al. (1988) - 15-day inhalation study in Wistar rats*
Neumann (1987) - 35-day inhalation study in unweaned piglets
Targowski et al. (1984) - 3-week inhalation study in Hartley guinea pigs
Schaerdel et al. (1983a) -- 24-hour inhalation study in CrkCOBS CD(SD) rats *
Schoeb et al. (1982) - 35-day study in F344 rats
Richard (1978) - 7-day study in OF1 mice
Broderson et al. (1976) - 35- to 75-day inhalation studies in Sherman rats and F344 rats
Curtis et al. (1975) - 109-day inhalation study in crossbred pigs
Doig and Willoughbv (1971) - 6-week inhalation study in Yorkshire-Landrace pigs
Coon et al. (1970) - 42- to 90-day inhalation studies in Sprague-Dawley and Long-Evans rats, New Zealand albino
rabbits, Princeton-derived guinea pigs, squirrel monkeys, and beagle dogs
Stombaugh et al. (1969) - 5-week inhalation study in Duroc pigs
Anderson et al. (1964b) - 7- to 42-day inhalation studies in Swiss albino mice and guinea pigs (strain not specified)
Weatherbv (1952) - 6- to 18-week inhalation study in guinea pigs (strain not provided)
*These studies were not identified as health effect/toxicity studies in Figure LS-1, but were included in Table 1-6
(evidence pertaining to other system effects in animals) as studies that provided useful quantitative information
on the biochemical/metabolic effects of ammonia.
The references considered and cited in this document, including bibliographic information
and abstracts, can be found on the Health and Environmental Research On-line (HERO) website
f http://hero.epa.gov/ammonial.
This document is a draft for review purposes only and does not constitute Agency policy.
LS-17 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Toxicological Review of Ammonia
1. HAZARD IDENTIFICATION
1.1. OVERVIEW OF CHEMICAL PROPERTIES AND TOXICOKINETICS
1.1.1. Chemical Properties
Ammonia (NH3) is a colorless alkaline gas with a pungent odor. Ammonia is very soluble in
water (NRC, 2008); in solution, it exists as ammonium hydroxide. Ammonium hydroxide is a weak
base that is partially ionized in water according to the following equilibrium fATSDR. 20041:
NH3 + H20?±NH4+ + OH"
Ammonium hydroxide ionizes with a dissociation constant of 1.77 x 10 5 at 25°C that
increases slightly with increasing temperature (Read. 19821. A decrease in pH results in an
increase in the concentration of ammonium ion (NH4+ or protonated form), a decrease in the
concentration of the un-ionized form (NH3), and an increase in solubility of ammonia in water. At
pH 9.25, half of the ammonia will be ionized (NH4+) and half will be un-ionized (NH3). AtpH values
of 8.25 and 7.25, 90% and 99%, respectively, of ammonia will be ionized (NH4+) fATSDR. 20041.
Thus, at physiological pH (7.4), the equilibrium between NH3 and NH4+ favors the formation of
NH4+. Chemical and physical properties of ammonia are listed in Table 1-1.
Table 1-1. Chemical and physical properties of ammonia
Parameter
Value
Reference
Chemical name
Ammonia3
Synonym(s)
AM-Fol; anhydrous ammonia; ammonia gas;
Nitro-sil; R 717; Spirit of hartshorn
NLM (2012)
Structure
H
hAh
NLM (2012)
Chemical formula
NHb
NLM (2012)
CASRN
7664-41-73
NLM (2012)
Molecular weight
17.031
Lide (2008). pp. 4.46-4.48. 8.40
Form
Colorless gas; corrosive
O'Neiletal. (2006)
Melting point
-77.73°C
Lide (2008). pp. 4.46-4.48. 8.40
Boiling point
-33.33°C
Lide (2008), pp. 4.46-4.48, 8.40
Odor threshold
53 ppm (37 mg/m3)
2.6 ppm (2 mg/m3)
O'Neiletal. (2006)
Smeets et al. (2007)
Density
0.7714 g/Lat 25°C
O'Neiletal. (2006)
This document is a draft for review purposes only and does not constitute Agency policy.
1-1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Table 1-1. Chemical and physical properties of ammonia
Parameter
Value
Reference
Vapor density
0.5967 (air = 1)
O1 Neil et al. (2006)
pKa (ammonium ion)
9.25
Lide (2008), pp. 4.46-4.48, 8.40
Solubility:
Water
Organic solvents
4.82 x 105 mg/L at 24°C
Soluble in ethanol, chloroform, and ether
Lange and Dean (1985), dd. 10-3,
10-23;
Lide (2008), dd. 4.46-4.48,
8.40; O1 Neil etal. (2006)
Vapor pressure
7.51 x 103 mm Hg at 25°C
(AlChE, 1999)
Henry's law constant
1.61 x 10"5 atm-m3/mol at 25°C
Betterton (1992)
Conversion factors
ppm to mg/m3
mg/m3 to ppm
1 ppm = 0.707 mg/m3
1 mg/m3 = 1.414 ppm
Verschueren (2001)
aAmmonia dissolved in water is sometimes referred to as ammonium hydroxide (CASRN 1336-21-6). Ammonium
hydroxide does not exist outside of solution.
1.1.2. Toxicokinetics
Ammonia is absorbed by the inhalation route of exposure. Most inhaled ammonia is
retained in the upper respiratory tract and is subsequently eliminated in expired air. Ammonia (as
NH4+) is produced endogenously in the human intestines through the use of amino acids as an
energy source (glutamine deamination) and by bacterial degradation of nitrogenous compounds
from ingested food is largely absorbed. At physiological pH, 98.3% of ammonia is present in the
blood as the ammonium ion (NH4+). Given its importance in amino acid metabolism, the urea cycle,
and acid-base balance, ammonia is homeostatically regulated to remain at low concentrations in the
blood. Ammonia is present in fetal, as well as adult, circulation, and is also present in human breast
milk as one of the sources of nonprotein nitrogen. Ammonia is produced endogenously by
catabolism of amino acids by glutamate dehydrogenase or glutaminase primarily in the liver, renal
cortex and intestines, but also in the brain and heart. Ammonia is metabolized to glutamine via
glutamine synthetase in the glutamine cycle or incorporated into urea as part of the urea cycle. The
liver removes an amount of ammonia from circulation equal to the amount added by the intestines
at metabolic steady state, such that the gut does not contribute significantly to systemic ammonia
release under normal conditions. Renal elimination via the kidney is a major contributor to
ammonia homeostasis; however, the kidneys are themselves a source of systemic ammonia. The
principal means of excretion of ammonia is as urinary urea; lesser amounts are eliminated in the
feces, through sweat production, and in expired air. A more detailed summary of ammonia
toxicokinetics is provided in Appendix C, Section C.l.
This document is a draft for review purposes only and does not constitute Agency policy.
1-2 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
1.2. SYNTHESIS OF EVIDENCE
Section 1.2 provides a synthesis and evaluation of the literature on the health effects of
inhaled ammonia in humans and experimental animals organized by organ/system. Evidence for
ammonia health effects is also summarized in organ/system-specific evidence tables, which present
key study design information and results, and graphically in exposure-response arrays. More
detailed study design information and results are provided in individual study summaries in
Appendix C in the Supplemental Information.
1.2.1. Respiratory Effects
The respiratory system is the primary target of toxicity of inhaled ammonia in humans and
experimental animals. Five cross-sectional occupational epidemiology studies in industrial settings
(Rahman etal.. 2007: Ali etal.. 2001: Ballal etal.. 1998: Bhatand Ramaswamv. 1993: Holness etal..
19891 examined the association between inhaled ammonia and prevalence of respiratory
symptoms or changes in lung function (Table 1-2). Another set of studies examined pulmonary
function or asthma symptoms in relation to ammonia exposure in health care workers and
domestic cleaners fArifand Delclos. 2012: Dumas etal.. 2012: Lemiere etal.. 2012: Vizcava etal..
2011: Zocketal.. 2007: Medina-Ramon etal.. 2006: Medina-Ramon etal.. 20051 (Table 1-3). The
association between ammonia exposure and respiratory effects indicated by these studies is also
informed by studies of pulmonary function in individuals in agricultural settings and subchronic
inhalation toxicity studies in various experimental animal species (Table 1-4). The evidence of
respiratory effects in humans and experimental animals exposed to ammonia is summarized in an
exposure-response array in Figure 1-1 at the end of this section.
Respiratory Symptoms
Respiratory symptoms (including cough, wheezing, and other asthma-related symptoms)
were reported in two cross-sectional studies of industrial worker populations exposed to ammonia
at levels greater than or equal to approximately 18 mg/m3 f Rah man etal.. 2007: Ballal etal.. 19981
(Table 1-2). One of these studies also examined frequency of respiratory symptoms by cumulative
ammonia concentration (CAC, mg/m3-years) and observed significantly higher relative risks (2.4-
5.3) with higher CAC (>50 mg/m3-years) compared to those with a lower CAC (<50 mg/m3-years)
f Ballal etal.. 19981. In three studies examining lower exposure settings (Rahman etal.. 2007: Ballal
etal.. 1998: Holness etal.. 19891 (Table 1-2), no differences were observed in the prevalence of
respiratory symptoms between ammonia-exposed workers and controls. Ammonia concentrations
reported in these lower exposure settings included a mean ammonia concentration of 6.5 mg/m3
and a high-exposure group defined as >8.8 mg/m3 in Holness et al. (19891. an exposure range of
0.2—7 mg/m3 in "Factory B" of Ballal et al. (19981. and a mean concentration of 4.9 mg/m3
in Rahman et al. f20071. The primary limitation noted in all of these studies was the potential
under-ascertainment of effects inherent in the study of a long-term worker population (i.e., "healthy
worker" effect) (see Literature Search Strategy | Study Selection and Evaluation section and Table
B-6 in the Supplemental Information). Confounding by other workplace exposures, although a
This document is a draft for review purposes only and does not constitute Agency policy.
1-3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
potential concern, was unlikely to be a major limitation affecting the interpretation of the pattern of
results seen in these studies, given the lack of nitrogen dioxide measurements above the detection
limit in one study f Rah man et al.. 20071 and the high level of control of exposures in another study
fHolness etal.. 19891.
Studies of health care workers or hospital workers fArif and Delclos. 2012: Dumas etal..
20121 (Table 1-3) provide evidence that exposure to ammonia as a cleaning or disinfectant product
is associated with increased risk of asthma or asthma symptoms. Use of ammonia as a cleaning
product in other settings has also been associated with asthma and respiratory symptoms fCasas et
al.. 2013: Vizcava etal.. 2011: Zock etal.. 2007: Medina-Ramon et al.. 20051 (Table 1-3).
Occupational exposure to ammonia was associated with work-exacerbated asthma (compared to
non- work related asthma) in a study at two occupational asthma specialty clinics by Lemiere et al.
f20121 (Table 1-3). Six studies, from Europe, Canada, and the United States, observed elevated
odds ratios, generally between 1.5 and 2.0, with varying degrees of precision. These studies were
conducted using a variety of designs, including a prospective study (Zock etal.. 20071 and two
nested case-control studies (Medina-Ramon et al.. 2006: Medina-Ramon etal.. 20051. Criteria used
to define current asthma or asthma symptoms were generally well defined and based on validated
methods. A major limitation of this collection of studies is the lack of direct measures of ammonia
exposure. Two of the studies included expert assessment of exposure (blinded to case status);
expert assessment improves reliance on self-reported exposure fDumas etal.. 2012: Lemiere etal..
20121. Confounding by other cleaning products is an unlikely explanation for these results, as two
of the studies noted only weak correlations between ammonia and other product use (Zock etal..
2007: Medina-Ramon et al.. 20051. and another study observed stronger associations with
ammonia than with bleach (Dumas et al.. 20121. All of the studies addressed smoking as a potential
confounder.
Studies in populations exposed in agricultural settings, including swine and dairy farmers,
that analyzed for prevalence of respiratory symptoms (including cough, phlegm, wheezing, chest
tightness, and eye, nasal, and throat irritation) in relation to ammonia exposure provided generally
negative results (Loftus etal.. 2015: Melbostad andEduard. 2001: Preller etal.. 1995: Zeida etal..
19941 (Appendix C, Table C-7). Two other studies that measured ammonia, but did not present an
analysis in relation to variability in ammonia levels, reported an increased prevalence of
respiratory symptoms in pig farmers exposed to ammonia from animal waste fChoudatetal..
1994: Crook etal.. 19911 (Appendix C, Table C-8). With the exception of the Loftus etal. (20151
study, all studies involving exposure in agricultural settings documented exposures to compounds
in addition to ammonia, such as airborne dust, endotoxin, mold, and disinfectants: Loftus et al.
(20151 did not analyze for other contaminants.
Reports of irritation and hyperventilation in volunteers acutely exposed to ammonia at
concentrations ranging from 11 to 354 mg/m3 ammonia for durations up to 4 hours under
controlled exposure conditions fPetrova et al.. 2008: Smeets etal.. 2007: Ihrigetal.. 2006: Verberk.
1977: Silverman etal.. 19491 provide support for ammonia as a respiratory irritant (Appendix C,
Section C.2.3 and Table C-9). Two controlled-exposure studies provide some evidence of
This document is a draft for review purposes only and does not constitute Agency policy.
1-4 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
habituation to eye, nose, and throat irritation in volunteers after repeated ammonia exposure.
Following exposure to ammonia at concentrations ranging from 7 to 35 mg/m3 for 4 hours/day on
five consecutive days, Ihrig etal. (2006) reported higher mean intensities for irritative, olfactory,
and respiratory symptoms in male volunteers unfamiliar with ammonia when compared to male
chemical company workers exposed to ammonia vapor for several years in a urea department;
differences were statistically significant only for olfactory symptoms; however the sample size was
small. In a more limited study with only four male volunteers each exposed to 18, 35, or 71 mg/m3
ammonia (exposure to each concentration was for one week, 2-6 hour/day, 5 days/week), fewer
occurrences of irritation occurred upon the second weekly exposure to the same
concentration Ferguson etal. f 19771
Numerous case reports document the acute respiratory effects of inhaled ammonia, ranging
from mild symptoms (including nasal and throat irritation and perceived tightness in the throat) to
moderate effects (including pharyngitis, tachycardia, dyspnea, rapid and shallow breathing,
cyanosis, transient bronchospasm, and rhonchi in the lungs) to severe effects (including burns of
the nasal passages, soft palate, posterior pharyngeal wall, and larynx, upper airway obstruction,
bronchospasm, persistent, productive cough, bilateral diffuse rales and rhonchi, mucous
production, pulmonary edema, marked hypoxemia, and necrosis of the lung) (Appendix C, Section
C.2.3).
Experimental studies in laboratory animals also provide consistent evidence that repeated
exposure to ammonia can affect the respiratory system (Table 1-4 and Appendix C, Section C.3).
The majority of available animal studies did not look at measures of respiratory irritation, in
contrast to the majority of human studies, but rather examined histopathological changes of
respiratory tract tissues. Histopathological changes in the nasal passages were observed in
Sherman rats after 75 days of exposure to 106 mg/m3 ammonia and in F344 rats after 35 days of
exposure to 177 mg/m3 ammonia, with respiratory and nasal epithelium thicknesses increased 3-4
times that of normal fBroderson etal.. 19761. Thickening of nasal and tracheal epithelium (50-
100%) was also observed in pigs exposed to 71 mg/m3 ammonia continuously for 1-6 weeks (Doig
and Willoughby. 1971). Nonspecific inflammatory changes (not further described) were reported
in the lungs of Sprague-Dawley and Long-Evans rats and guinea pigs intermittently exposed to
770 mg/m3 ammonia for 6 weeks; continuous exposure to 455 and 470 mg/m3 ammonia increased
mortality in rats fCoonetal.. 19701. Focal or diffuse interstitial pneumonitis was observed in all
Princeton-derived guinea pigs, New Zealand white rabbits, beagle dogs, and squirrel monkeys
exposed to 470 mg/m3 ammonia fCoon etal.. 19701. Additionally, under these exposure conditions,
dogs exhibited nasal discharge and other signs of irritation (marked eye irritation, heavy
lacrimation). Nasal discharge was observed in 25% of rats exposed to 262 mg/m3 ammonia for
90 days (Coon etal.. 1970).
At lower concentrations, approximately 50 mg/m3 and below, the majority of studies of
inhaled ammonia did not identify respiratory effects in laboratory animals exposed to ammonia.
No increase in the incidence of respiratory or other diseases common to young pigs was observed
after continuous exposure to ammonia and inhalable dust at concentrations representative of those
This document is a draft for review purposes only and does not constitute Agency policy.
1-5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
found in commercial pig farms (<26 mg/m3 ammonia) for 5 weeks (Done etal.. 20051. No gross or
histopathological changes in the turbinates, trachea, and lungs of pigs were observed after
continuous exposure to 35 or 53 mg/m3 ammonia for up to 109 days fCurtis etal.. 19751. No signs
of toxicity in rats or dogs were observed after continuous exposure to 40 mg/m3 ammonia for 114
days or after intermittent exposure (8 hours/day) to 155 mg/m3 ammonia for 6 weeks fCoonetal..
19701. Only one study reported respiratory effects at concentrations <50 mg/m3 (i.e., lung
congestion, edema, and hemorrhage in guinea pigs and mice exposed to 14 mg/m3 ammonia for up
to 42 days; Anderson et al. (1964all. but confidence in the findings from this study is limited by
inadequate reporting and the small numbers of animals tested.
Lung Function
Decreased lung function in ammonia-exposed workers has been reported in three of the
four studies examining this outcome measure (Rahman etal.. 2007: Ali etal.. 2001: Bhat and
Ramaswamv. 19931: the exception is the study by Holness etal. (19891 (Table 1-2) in which no
significant changes in lung function were observed in workers exposed to ammonia in an industrial
setting with relatively low ammonia exposure levels (Table 1-2). These effects were observed in
short-term scenarios (i.e., cross-work shift changes in lung function) in fertilizer factory workers
(mean ammonia concentration of 18.5 mg/m3) compared with administrative staff controls
f Rah man et al.. 20071. and in longer-term scenarios, in workers with a cumulative exposure of
>50 mg/m3-years when compared with workers with a lower cumulative exposure of <50 mg/m3-
years (with an approximate 5-7% decrease in FVC% predicted and FEVi% predicted) (Ali etal..
20011. There were no decrements in the percent of predicted lung function values when comparing
the total exposed group to a control group of office workers in the latter study, in the relatively low
exposure scenario examined in Holness etal. Q9891 (mean ammonia concentration of 6.5 mg/m3
and high-exposure group defined as >8.8 mg/m3), or in the low-exposure group (mean ammonia
concentration of 4.9 mg/m3) in Rahman et al. f20071. Another study of ammonia plant fertilizer
workers reported statistically significant decreases in forced expiratory volume (FEVi) and peak
expiratory flow rate (PEFR/minute) in workers compared to controls (Bhat and Ramaswamv.
19931: however, measurements of ammonia levels were not included in this study. As discussed
previously in the summary of respiratory symptoms studies, the primary limitation within this set
of studies is the potential under-ascertainment of effects in these studies of long-term worker
populations.
One of the studies of domestic cleaning workers described in Table 1-3 included a measure
of pulmonary function (Medina-Ramon et al.. 20061. Ammonia use was associated with a decrease
in peak expiratory flow (PEF) (-9.4 [95% CI, -17, -2.3]). A limitation of this study was the use of
lung function measurements conducted by the participant; the reliability of this procedure has not
been established. In a study by Casas etal. T2 0131 on the effects of cleaning product use on the
respiratory health of children, ammonia exposure was associated with decreased lung function
(FEVi: -28 [95% CI -131, 76]) (Table 1-3).
This document is a draft for review purposes only and does not constitute Agency policy.
1-6 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Impaired respiratory function (e.g., decreased FEVi and/or forced vital capacity [FVC]) in
an agricultural setting was associated with ammonia exposure in six of the eight studies that
included pulmonary function measures fLoftus etal.. 2015: Monso etal.. 2004: Donham etal..
2000: Reynolds etal.. 1996: Donham etal.. 1995: Preller etal.. 1995: Zeida etal.. 1994: Heederik et
al.. 19901 (Appendix C, Table C-7). In general, EPA considered these eight studies to be the
strongest with respect to methodology, based on considerations of exposure assessment and
assessment of potential confounding (see Literature Search Strategy | Study Selection and
Evaluation section).
Changes in lung function following acute exposure to ammonia have been observed in some,
but not all, controlled human exposure studies conducted in volunteers (Appendix C, Section C.2.3
and T able C-9). Cole etal. (1977) reported reduced lung function as measured by reduced
expiratory minute volume and changes in exercise tidal volume in volunteers exposed for a half-day
in a chamber at ammonia concentrations >106 mg/m3, but not at 71 mg/m3. Bronchoconstriction
was reported in volunteers exposed to ammonia through a mouthpiece for 10 inhaled breaths of
ammonia gas at a concentration of 60 mg/m3 fDouglas and Coe. 19871: however, there were no
bronchial symptoms reported in volunteers exposed to ammonia in an exposure chamber at
concentrations of up to 35 mg/m3 for 10 minutes fMacEwen etal.. 19701. Similarly, no changes in
bronchial responsiveness or lung function (as measured by FVC and FEVi) were reported in
healthy volunteers exposed to ammonia at concentrations up to 18 mg/m3 for 1.5 hours during
exercise (Sundblad etal.. 2004). There were no changes in lung function as measured by FEVi in 25
healthy volunteers and 15 mild/moderate persistent asthmatic volunteers exposed to ammonia
concentrations up to 354 mg/m3 ammonia for up to 2.5 hours (Petrova etal.. 2008). or in 6 healthy
volunteers and 8 mildly asthmatic volunteers exposed to 11-18 mg/m3 ammonia for 30-minute
sessions fSigurdarson etal.. 20041.
Lung function effects following ammonia exposure were not evaluated in the available
animal studies.
This document is a draft for review purposes only and does not constitute Agency policy.
1-7 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-2. Evidence pertaining to respiratory effects in humans following
inhalation exposure in industrial settings
Study design and reference
Results
Respiratory symptoms
Rahman et al. (2007) (Bangladesh)
Urea fertilizer factory worker (all men); 24 ammonia
plant workers, 64 urea plant workers, and 25
controls (staff from administration building). Mean
employment duration: 16 years
Exposure: Personal samples (2 methods3;
correlation = 0.80)
Low-exposure group (ammonia plant), mean: 6.9
ppm (4.9 mg/m3); range: 2.8-11.1 ppm (2-8
mg/m3)
High-exposure group (urea plant), mean: 26.1 ppm
(18.5 mg/m3); range: 13.4-43.5 ppm (9-31 mg/m3)
Outcome: Respiratory symptoms (5 point scale for
severity over last shift), based on Optimal Symptom
Score Questionnaire
Percentage of workers reporting symptoms (p-value):
Low exposed High exposed
Controls (n = 24) (n = 64)
(n = 25) (p-value)1 (p-value)2 (p-value)3
Cough
Chest tightness
Stuffy nose
Runny nose
Sneeze
17 (0.42)
17 (0.42)
12 (0.35)
4 (1.0)
0 (0.49)
28 (0.05)
33 (0.02)
16 (0.17)
16 (0.17)
22 (0.22)
(0.41)
(0.19)
(1.0)
(0.28)
(0.01)
^-value for ammonia plant compared to control
2p-value for urea plant compared to control
3p-value for urea plant compared to ammonia plant
Ballal et al. (1998) (Saudi Arabia)
Urea fertilizer factory workers (two factories) (all
men); 161 exposed workers and 355 unexposed
controls'5. Mean employment duration: 51.8 months
(exposed workers) and 73.1 months (controls)
Exposure: Area monitors (3 sets in each work
section taken at least 3 months apart, mean 16
measures per set).
Factory A (high-exposure factory): 2-1301 mg/m3
(mid-point = 66 mg/m3); geometric mean <18
mg/m3, except for urea packaging and store areas
(geometric means = 18.6 and 115 mg/m3,
respectively)
Factory B (low-exposure factory): 0.02-7 mg/m3;
geometric mean <18 mg/m3
Cumulative exposure calculated based on exposure
and duration; dichotomized to high and low at 50
mg/m3-years
Outcome: Respiratory symptoms based on British
Medical Research Council questionnaire
Relative risk (95% CI), compared with controls
Factory B2 Factory A2
(0.02-7 mg/m3; n = 77) (2-27.1 mg/m3; n = 78)1
Cough
Phlegm
Wheezing
Dyspnea
No cases
No cases
0.97 (0.21, 4.5)
0.45 (0.11, 1.9)
2.0 (0.38, 10.4)
2.0 (0.38, 10.4)
3.4(1.2,9.5)
1.8 (0.81, 4.2)
Relative risk (95% CI), compared with lower exposure setting
(<18 mg/m3 [n = 138] or <50 mg/m3-years [n = 130])
Cumulative
>18 mg/m3 >50 mg/m3-years
(n = 17) (n = 30)
Cough
Phlegm
Wheezing
Dyspnea
Asthma
Chronic
bronchitis
3.5 (1.8, 6.6)
3.8(2.0. 7.1)
5.0 (2.4, 10.6)
4.6 (2.4, 8.8)
4.3 (2.1, 9.0)
2.3 (0.31, 17)
2.8 (1.6, 5.0)
3.0 (1.7, 5.5)
5.2 (2.9, 9.5)
2.6 (1.3, 5.4)
2.4 (1.1, 5.4)
5.3 (1.7, 16)
^he ammonia concentration range in Factory A is
better represented as 2-27.1 mg/m3. This range
excludes the employees in the urea store (n = 6;
range of ammonia concentrations = 90-130.4
mg/m3) who were required to wear full protective
clothing, thus minimizing potential exposure.
Number of workers in Factory A excluding urea
store workers = 78.
2Factory-specific analyses stratified by smoking status;
results presented here are for non-smokers. Similar patterns
seen in other smoking categories.
Approximate 1.3-1.5 relative risk (p < 0.05) per unit increase
in ammonia concentration for cough, phlegm, wheezing, and
asthma, adjusting for duration of work, cumulative exposure,
smoking, and age.
This document is a draft for review purposes only and does not constitute Agency policy.
1-8 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-2. Evidence pertaining to respiratory effects in humans following
inhalation exposure in industrial settings
Study design and reference
Results
Holness et al. (1989) (Canada)
Percentage of workers reporting symptoms (%):
Soda ash plant workers (all men); 58 exposed
Control
Exposed
workers and 31 controls (from stores and office
(n = 31)
(n = 58)
p-value
areas of plant)0. Average exposure: 12.2 years
Cough
10
16
0.53
Exposure: Personal samples, one work-shift per
Sputum
16
22
0.98
person, mean 8.4 hours
Bronchitis
19
22
0.69
Low: <6.25 ppm (<4.4 mg/m3); n = 34
Wheeze
10
10
0.91
Medium: 6.25-12.5 ppm (4.4-8.8 mg/m3); n = 12
Chest tightness
6
3
0.62
High: >12.5 ppm (>8.8 mg/m3); n = 12
Dyspnea
13
7
0.05
All exposed workers (mean): 6.5 mg/m3
(shortness of
Outcome: Respiratory symptoms based on
breath)
American Thoracic Society questionnaire
Chest pain
6
2
0.16
Rhinitis (nasal
19
10
0.12
complaints)
Throat irritation
3
7
0.53
No increased risk seen in analyses stratified by exposure
group.
Lung function
Rahman et al. (2007) (Bangladesh)
Pre-shift
Post-shift
p-value
Urea fertilizer factory workers (all men); 24
Ammonia plant (low-exposure group, 4.9 mg/m3]
; n = 24
ammonia plant workers, 64 urea plant workers, and
ammonia plant workers
25 controls (staff from administration building).
FVC
3.308
3.332
0.67
Mean employment duration: 16 years
FEVi
2.627
2.705
0.24
Exposure: Personal samples (2 methods3;
PEFR
8.081
8.313
0.22
correlation = 0.80)
Low-exposure group (ammonia plant), mean: 6.9
Urea plant (high-exposure group, 18.5 mg/m3); n
= 64 urea
ppm (4.9 mg/m3); range: 2.8-11.1 ppm (2-8
plant workers
mg/m3)
FVC
3.362
3.258
0.01
High-exposure group (urea plant), mean: 26.1 ppm
FEVi
2.701
2.646
0.05
(18.5 mg/m3); range: 13.4-43.5 ppm (9-31 mg/m3)
PEFR
7.805
7.810
0.97
Outcome: Lung function (standard spirometry)
p-value reflects the comparison of pre- and post-shift values.
Multiple regression model (data from 23 ammonia and urea
plant workers with concurrent measurements of ammonia
exposure and lung function):
- Concentration of ammonia and exposure duration (yrs of
employment as proxy for duration) were significantly
correlated with percentage cross-shirt decrease in FEVi%
(AFEVi%).
- Each year of work in a production section was associated
with a decrease in AFEVi% of 0.6%. [Limitation of analysis:
failure to explore the age parameter; age and years of work
were highly correlated (Pearson correlation coefficient 0.97).
This document is a draft for review purposes only and does not constitute Agency policy.
1-9 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-2. Evidence pertaining to respiratory effects in humans following
inhalation exposure in industrial settings
Study design and reference
Results
Ali et al. (2001) (Saudi Arabia)
<50 mg/m3-y >50 mg/m3-y
Urea fertilizer factory workers (all men)—(additional
(n = 45) (n = 28)
p-value
studv of "Factorv A" in Ballal et al. (1998)); 73
FVCi% 100.7
93.4
0.006
exposed workers and 348 unexposed controls.
predicted
Mean employment duration: not reported
FVC% 105.6
100.2
0.03
Exposure: 4-hour measurements. Cumulative
predicted
exposure calculated based on exposure and
FEVi/FVC% 84.7
83.4
NS
duration; dichotomized to high and low at 50
NS = not significant (p-values not provided by study authors)
mg/m3-years
Outcome: Lung function (standard spirometry;
morning measurement)
Bhat and Ramaswamv (1993) (India)
Ammonia
Fertilizer chemical plant workers; 30 diammonium
Controls DAP plant
Urea plant
plant
phospate (DAP) plant workers, 30 urea plant
(n = 68) (n = 30)
(n = 30)
(n = 31)
workers, 31 ammonia plant workers, and 68
FVC 3.4 ±0.21 2.5 ±0.06*
3.3 ±0.11
3.2 ±0.07
controls (people with comparable body surface area
FEVi 2.8 ±0.10 2.1 ±0.08*
2.7 ±0.10
2.5 ±0.1*
chosen from the same socio-economic status and
PEFR 383 ±7.6 228 ± 18*
307 ±19*
314 ± 20*
sex as exposed workers)
*p < 0.05
Exposure: Measurements not reported; duration
dichotomized as <10 and >10 years
Outcome: Lung function (standard spirometry)
Holness et al. (1989) (Canada)
Control
Exposed
Soda ash plant workers (all men); 58 exposed
(n = 31)
(n = 58)
p-valuea
workers and 31 controls (from stores and office
Lung function (% predicted values
b:
areas of plant)0. Average exposure: 12.2 years
FVC 98.6 ± 11.3
96.8 ± 11.0
0.094
Exposure: Personal samples, one work-shift per
FEVi 95.1 ±12.5
94.1 ± 12.9
0.35
person, mean 8.4 hours
FEVi/FVC 96.5 ±6.1
97.1 ±7.1
0.48
Low: <6.25 ppm (<4.4 mg/m3); n = 34
Medium: 6.25-12.5 ppm (4.4-8.8 mg/m3); n = 12
Change in lung function over work shift:
High: >12.5 ppm (>8.8 mg/m3); n = 12
FVC dayl -0.9
-0.8
0.99
All exposed workers (mean): 6.5 mg/m3
day 2 +0.1
-0.0
0.84
Outcome: Lung function (standard spirometry;
FEVi day 1 -0.2
-0.2
0.94
beginning and end of shift, at least two test days per
day 2 +0.5
+0.7
0.86
worker)
ap-valuefor difference between exposed and control workers calculated by
using actual baseline values and correcting for age, height, and pack-years
smoked determined by multiple regression analysis.
'Percentage of the subject's predicted value (% predicted) has been widely
adopted as follows: % predicted = recorded value x 100/predicted value);
this value is now calculated on automated spirometers based on sex, race,
age and height.
This document is a draft for review purposes only and does not constitute Agency policy.
1-10 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-2. Evidence pertaining to respiratory effects in humans following
inhalation exposure in industrial settings
Study design and reference
Results
FEVi = forced expiratory volume in 1 second; FVC = forced vital capacity; PEFR = peak expiratory flow rate.
aExposure concentrations were determined by both the Drager tube and Drager PAC III methods. Using the Drager
tube method, concentrations of ammonia in the ammonia and urea plants were 17.7 and 88.1 mg/m3, respectively;
using the Drager 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 Drager Safety Inc (telephone
conversations and e-mails dated June 22, 2010, from Michael Yanosky, Drager 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 Drager tubes. Therefore, higher confidence is
attributed to the PAC III air measurements of ammonia for the Rahman et al. (2007) study.
bThe 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.
cAt 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.
1-11 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-3. Evidence pertaining to respiratory effect in humans following
inhalation exposure in cleaning settings
Study design and reference
Results
Asthma or asthma symptoms
Dumas et al. (2012) (France)
Hybrid design, hospital workers, drawn from
population-based case-control study; 179 hospital
workers (136 women), 545 other workers (333 women).
Exposure: Asthma-specific job exposure matrix plus +
expert review (blinded), ever exposed, 18 specific
products, based on all jobs held at least 3 months;
ammonia prevalence 23% in female hospital workers
Outcome: Current asthma: Asthma attack, respiratory
symptoms or asthma treatment in the last 12 months
(based on standardized questionnaire)
Odds ratio (95% CI), current asthma
Women: 3.05 (1.19, 7.82)
Men: no associations with any specific products
(prevalence low)
Adjusted for age and smoking, and accounting for
familial dependence (due to sampling of cases and first
degree relatives)
Arif and Delclos (2012) (United States, Texas)
Population survey of 3,650 health care workers
(physicians, nurses, respiratory therapists, occupational
therapists), (total n = 5,600, response rate 66%)
Exposure: Structured questionnaire—frequency of use
of products for longest job held; ever contact with list of
28 products; ammonia prevalence 23%
Outcome: Structured questionnaire
• Work-related asthma symptoms: wheezing/whistling
at work or shortness of breath at works that gets
better away from work or worse at work
• Work-exacerbated asthma: onset before began work
• Occupational asthma: onset after began work)
Odds ratio (95% CI) [n cases]
Work-related asthma symptoms [n = 132]
2.45(1.28, 4.69)
Work-exacerbated asthma [n = 41]
1.58 (0.56, 4.43)
Occupational asthma [n = 33]
1.86 (0.49, 7.13)
Adjusted for age, sex, race/ethnicity, body mass index,
seniority, atopy, and smoking status
Lemiere et al. (2012) (Quebec, Canada)
Case-control study, workers seen at two tertiary care
centers specializing in occupational asthma. Asthma
(defined below) based on reversible airflow limitation
or airway hyper-responsiveness tests; referent group =
non-work related asthma (NWRA) seen at same clinics
but symptoms did not worsen at work (n = 33).
Exposure: Structured interview focusing on last/current
job, combined with expert review (blinded); ammonia
prevalence 19/153 = 12%
Outcome: Diagnoses made based on reference tests
• Occupational asthma if specific inhalation challenge
test was positive
• Work-exacerbated asthma if specific inhalation test
was negative but symptoms worsened at work
Odds ratio (95% CI) [n cases]
Work exacerbation [n = 53]
8.4 (1.1, 371.7)
Occupational asthma [n = 67]
3.7 (0.4,173.4)
Age, smoking, occupational exposure to heat, cold,
humidity, dryness, and physical strain assessed as
confounders.
[Wide confidence intervals reflect sparseness in
referent group, with only 1 of the 33 classified as
exposed to ammonia]
This document is a draft for review purposes only and does not constitute Agency policy.
1-12 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-3. Evidence pertaining to respiratory effect in humans following
inhalation exposure in cleaning settings
Study design and reference
Results
Vizcava et al. (2011) (Spain)
Survey of cleaning service workers (n = 917) from 37
businesses (19% response rate to questionnaire
distributed through the employers); 761 current
cleaners, 86 former cleaners, 70 never cleaners;
referent group = never cleaners and current cleaners
who have not used any of the specified cleaning
products in last year (n = 161)
Exposure: Structured questionnaire, use of cleaning
tasks and 12 products; ammonia prevalence 66%
Outcome: Structured questionnaire
• Current asthma: in past 12 months, woken by an
attack of shortness of breath, had an attack of
asthma or currently taking any asthma medications
(including inhalers, aerosols or tablets)
• Asthma score: Sum of "yes" answers to 5 symptoms
in last 12 months (wheeze with breathlessness,
woken up with chest tightness, attack of shortness
of breath at rest, attack of shortness of breath after
exercise, woken by attack of shortness of breath)
Odds ratio (95% CI) (among current cleaners) [n]
Current asthma 1.4 (0.6, 3.2) [81]
Wheeze without having a cold 2.1 (0.9, 4.7) [83]
Chronic cough 1.6 (0.8, 3.3) [95]
Asthma score 1.6 (1.0, 2.5)
[mean 0.59, SD 1.12]
Adjusted for age, country of birth (Spanish versus non-
Spanish), sex, and smoking status
Zock et al. (2007) (Europe, 22 sites)
Longitudinal study, n = 3,503, 9-year follow-up of
European Community Respiratory Health Survey,
population-based sample, ages 20-44 years. Excluded
764 individuals with asthma at baseline; limited to
individuals reporting doing the cleaning or washing in
their home.
Exposure: Structured interview at follow-up; frequency
of use of 15 products
Outcome: Structured interview at follow-up
• New onset (since baseline survey) current asthma,
defined by asthma attack or nocturnal shortness of
breath in the past 12 months or current use of
medication for asthma
• Current wheeze defined as wheezing or whistling in
the chest in last 12 months when not having a cold
• New onset physician-diagnosed asthma, asthma
defined as above with confirmation by a physician
and information on age or date of first attack
Odds ratio (95% CI) [n]
Current asthma 1.4 (0.87, 2.23) [199]
Current wheeze 1.3 (0.81, 2.13) [226]
Physician-diagnosed asthma 0.92 (0.33, 2.59) [71]
Adjusted for sex, age, smoking, employment in a
cleaning job during follow-up, and study center;
heterogeneity by center also assessed. Correlations
among products generally weak (Spearman rho < 0.3)
This document is a draft for review purposes only and does not constitute Agency policy.
1-13 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-3. Evidence pertaining to respiratory effect in humans following
inhalation exposure in cleaning settings
Study design and reference
Results
Medina-Ramon et al. (2005) (Spain)
Nested case-control, cleaning workers; case (n = 40;
74% participation rate) based on asthma and/or
bronchitis at both assessments. Controls (n = 155, 69%
participation rate)—no history of respiratory symptoms
in preceding year and no asthma at either assessment.
Exposure: Structured interview; frequency of use of 22
products; ammonia prevalence 16% undiluted, 56%
diluted
Outcome: Asthma: asthma attack or being woken by
attack or shortness of breath in past 12 months;
Chronic bronchitis: regular cough or regular bringing up
phlegm for at least 3 months each year
Odds ratio (95% CI) (unadjusted), >12 compared with
<12 times per year
Undiluted 3.1(1.2,8.0)
Diluted 0.8 (0.4,1.7)
FeNO and pulmonary function
Casas et al. (2013) (Spain)
Adiusteda associations of FeNO, FVC and FEVi-
Population based cross sectional birth cohort study; n =
432 infants enrolled; n = 295 total number of individuals
recruited that completed the 10-year follow up visit and
the cleaning products questionnaire and performed the
FeNO and/or lung function test; 35% of recruited
population were excluded because information on use
of cleaning products and/or respiratory tests was not
available; only 46 individuals reported use of ammonia
Exposure: Interviewer-led questionnaire; frequency of
use of 10 different cleaning products (bleach, ammonia,
polishes or waxes, acids, solvents, furniture sprays,
glass cleaning sprays, degreasing sprays, air freshening
sprays, and air freshening plug- ins); exposure score
developed based on frequency of use and number of
products used
Outcome: Questionnaires on wheezing asthma,
treatment and allergies were administered by mother
from birth to age 10; at age 10-13 FeNO and lung
function tests were carried out
with weekly use of ammonia (n=46; 16%)
FeNOcppb FVC mL FEVi mL
GM ratio (95% CI) P (95% CI) P (95% CI)
0.86 (0.66 to 1.12) 3 (-127 to 133) -28 (-131 to 76)
GM: geometric mean
a adjusted for sex, age, asthma medication, season of
respiratory measurement, maternal education and
parental smoking; FVC and FEVi models were
additionally adjusted for height and weight
b change in FeNO, FVC and FEVi per interquartile range
increase of the score (interquartile range = 6.5 d of
product use per week).
c FeNO (fraction of exhaled nitric oxide) is used to
characterize asthma or other conditions associated
with airway inflammation; it is measured in a breath
test.
Medina-Ramon et al. (2006) (Spain)
Panel study, sample selected from participants in
nested case-control studv bv Medina-Ramon et al.
(2005). Current asthma svmptoms or chronic bronchitis
in 2000-2001 survey; n = 51 of 80 (64%); 8 excluded for
possible recording errors, outliers, learning effects
Exposure: Daily diary of use of products
Outcome: Respiratory symptoms based on 2-week daily
diary (7 symptoms, 5 point intensity scale); summed
score for upper respiratory symptoms (blocked nose,
throat irritation, watery eyes) and lower respiratory
symptoms (chest tightness, wheezing, shortness of
breath, and cough); PEF measured with mini-Wright
peak flow meter (with training and written
Diluted and Diluted
undiluted only
OR (95% CI)
Upper
respiratory 1.8 (0.7,4.9) 1.3 (0.3,5.0)
symptoms
Lower 1.6(0.6,4.4) 3.0(1.0,9.1)
respiratory
symptoms
Beta (95% CI)
PEF at night -9.4 (-17, -2.3) -10.3 (-18,-2.7)
PEF,
following -1.2 (-8.5, 6.2) -2.9 (-11, 6.2)
morning
This document is a draft for review purposes only and does not constitute Agency policy.
1-14 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-3. Evidence pertaining to respiratory effect in humans following
inhalation exposure in cleaning settings
Study design and reference
Results
instructions); measured morning, lunchtime, night (3
measurements each; highest recorded)
Adjusted for respiratory infection, use of maintenance
medication, and age; daily number of
cigarettes smoked, years of employment in domestic
cleaning, and/or weekly working hours in domestic
cleaning also assessed as potential confounders
This document is a draft for review purposes only and does not constitute Agency policy.
1-15 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-4. Evidence pertaining to respiratory effects in animals
Study design and reference
Results
Effects on the lungs
Coon et al. (1970)
Sprague-Dawley and Long-Evans rat; male and female; 15/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Squirrel monkey (Saimiri sciureus); male; 3/group
Beagle dog; male; 2/group
0,155, or 770 mg/m3 for 8 hrs/d, 5 d/wk for 6 wks
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.a
Coon et al. (1970)
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
At 470 mg/m3, 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 (This exposure was lethal
to ~25% of the guinea pigs).
Coon et al. (1970)
Sprague-Dawley or Long-Evans rat; male and female; 15-51/group
0 or 40 mg/m3 for 114 d, 127, 262 or 470 mg/m3 for 90 d, or 455
mg/m3 for 65 d
Focal or diffuse interstitial pneumonitis in all
animals, and calcification of bronchial
epithelium observed in several animals at
470 mg/m3, an exposure that was lethal to
most of the rats.3
Anderson et al. (1964a)
Swiss albino mouse; male and female; 4/exposure interval
0 or 20 ppm (0 or 14 mg/m3) for 7,14, 21, 28, or 42 d
Lung congestion, edema, and hemorrhage
observed at 14 mg/m3 after 42 d.3
Anderson et al. (1964a)
Guinea pig (strain not specified); male and female; 2/exposure
interval at 20 ppm, 6/exposure interval at 50 ppm
0 or 20 ppm (0 or 14 mg/m3) for 7,14, 21, 28, or 42 d or 50 ppm
(35 mg/m3) for 42 d
Lung congestion, edema, and hemorrhage
observed at 14 and 35 mg/m3 after 42 d.3
Done et al. (2005)
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)
No increase in the incidence of respiratory
or other diseases.
Curtis et al. (1975)
Pig (crossbred); sex not specified; 4-8/group
0, 50, or 75 ppm (0, 35, or 53 mg/m3 for 109 d)
Turbinates, trachea, and lungs of all pigs
were classified as normal.
Effects on the upper respiratory tract
Coon et al. (1970)
Sprague-Dawley and Long-Evans rat; male and female; 15/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
0,155, or 770 mg/m3 for 8 hrs/d, 5 d/wk for 6 wks
Dyspnea in rabbits and dogs exposed to
770 mg/m3 during wk 1 only; no indication
of irritation after wk 1; nasal tissues not
examined for gross or histopathologic
changes.
Broderson et al. (1976)b
Sherman rat; 5/sex/group
10 or 150 ppm (7 or 106 mg/m3) from bedding for 75 d
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.
1-16 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-4. Evidence pertaining to respiratory effects in animals
Study design and reference
Results
Broderson et al. (1976)b
F344 rat; 6/sex/group
0 or 250 ppm (0 or 177 mg/m3) in an inhalation chamber for 35 d
1" thickness of the nasal epithelium (3-
4 times) and nasal lesions at 177 mg/m3.a
Coon et al. (1970)
Sprague-Dawley or Long-Evans rat; male and female; 15-51/group
0 or 40 mg/m3 for 114 d, 127, 262, or 470 mg/m3 for 90 d, or
455 mg/m3 for 65 d
Nasal discharge at 262 mg/m3 (25% of rats).
Dyspnea and nasal irritation/discharge in all
animals at 455 and 470 mg/m3, an exposure
that was lethal to the majority of the rats.3
Gaafar et al. (1992)
White albino mouse; male; 50
Ammonia vapor of 0 or 12% ammonia solution for 15 min/d,
6 d/wk, for 8 wks
Histological changes in the nasal mucosa.3
Doig and Willoughbv (1971)
Yorkshire-Landrace pig; sex not specified; 6/group
0 or 100 ppm (0 or 71 mg/m3) for 6 wks
1" thickness of nasal and tracheal
epithelium (50-100% increase).3
Stombaugh et al. (1969)
Duroc pig; both sexes; 9/group
12, 61,103,145 ppm (8, 43, 73, or 103 mg/m3) for 5 wks
Excessive nasal, lacrimal, and mouth
secretions and T* frequency of cough at
73 and 103 mg/m3.3
Coon et al. (1970)
Beagle dog; male; 2/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Nasal discharge at 470 mg/m3.3
incidence data not provided.
bThe 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.
1-17 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1000
100
u>
E.
C
o
£ 10
C
q;
u
C
O
u
QJ
\A
O
Q.
X
0.1
A
i
A
A
A
A
A
A
I t
....
A
A
A
ILOAEL ANOAEL •Additional concentrations
Vertical lines show range of concentrations in study,
t Highest concentration (470 mg/m3) and LOAEL (455
mg/m3) cannot be distinguished on this figure
* Exposures were intermittent: 8 hrs/d, 5 d/wk
flJ
03 2 ~
£ « B
O 9r O
a u Ci
c u
§. O. 75
oj
¦= c
o
1 £ j=
oj c
CC 3
E ^ ^
'a. ¦
oj >.
c °
I &
co d
E -•
£ 2
g
aj aj
- c
c o
— a
_ -C
(13 E QD
£ J S
O aj =
£?i g
- o3
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Mode-of-Action Analysis—Respiratory Effects
Data on the potential mode of action for respiratory effects associated with chronic
exposure to ammonia are limited. However, acute exposure data demonstrate that injury to
respiratory tissues is primarily due to ammonia's alkaline (i.e., caustic) properties from the
formation of hydroxide ion when it comes in contact with water and is solubilized. Ammonia
readily dissolves in the moisture on the mucous membranes, forming ammonium hydroxide, which
causes liquefactive necrosis of the tissues. Specifically, ammonia directly denatures tissue proteins
and causes saponification of cell membrane lipids, which leads to cell disruption and death
(necrosis). In addition, the cellular breakdown of proteins results in an inflammatory response,
which further damages the surrounding tissues fAmshel etal.. 2000: Millea et al.. 1989: larudi and
Golden. 19731
Summary of Respiratory Effects
Evidence for respiratory toxicity associated with exposure to ammonia comes from studies
in humans and animals. Multiple occupational studies involving chronic exposure to ammonia in
industrial settings provide evidence of an increased prevalence of respiratory symptoms (Rahman
etal.. 2007: Ballal etal.. 19981 and decreased lung function f Rah man et al.. 2 0 0 7: Ali etal..
2001: Bhatand Ramaswamv. 1993) (Table 1-2 and Appendix C, Section C.2.1). An increase in
respiratory effects was reported both with higher workplace ammonia concentrations f Rah man et
al.. 2007: Ballal etal.. 1998) and with greater cumulative ammonia concentration (expressed in
mg/m3-years) (Ali etal.. 2001: Ballal etal.. 1998). Evidence of respiratory effects is provided by
studies of asthma, asthma symptoms, and pulmonary function in workers and others exposed to
cleaning agents containing ammonia, in a variety of study designs and populations (Casas etal..
2013: Arif and Delclos. 2012: Dumas etal.. 2012: Lemiere etal.. 2012: Vizcavaetal.. 2011: Zocket
al.. 2007: Medina-Ramon etal.. 2006: Medina-Ramon et al.. 20051 (Table 1-3). Additional evidence
of respiratory effects of ammonia is seen in studies of pulmonary function in an agricultural setting,
specifically in livestock farmer studies that accounted for effects of co-exposures to other agents
such as endotoxin and dust f Don ham et al.. 2000: Reynolds etal.. 1996: Donham etal.. 1995: Preller
etal.. 1995: Heederik etal.. 19901. and in one study of asthmatic children that lived near animal
feeding operations that did not control for co-exposures fLoftus etal.. 2015) (Appendix C, Table
C-7). The livestock farmer studies, however, do not provide evidence of associations between
ammonia and respiratory symptoms. Controlled human exposure studies of ammonia inhalation
and case reports of injury in humans with inhalation exposure to ammonia provide additional
support for the respiratory system as a target of ammonia toxicity when inhaled (Appendix C,
Section C.2.3). Overall, the consistency of findings across three categories of epidemiological
studies (industrial, cleaner, and agricultural settings) that differed in population characteristics,
level and pattern of exposure, and potential confounders, and support from studies of acute
exposures, adds strength to the evidence for an association between respiratory effects and
ammonia exposure.
This document is a draft for review purposes only and does not constitute Agency policy.
1-19 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Evidence from animal studies supports an association between inhaled ammonia and
respiratory effects. Short-term and subchronic animal studies show histopathological changes of
respiratory tissues in several animal species (lung inflammation in guinea pigs and rats; focal or
interstitial pneumonitis in monkeys, dogs, rabbits, and guinea pigs; pulmonary congestion in mice;
thickening of nasal epithelium in rats and pigs; nasal inflammation or lesions in rats and mice)
across different dosing regimens fGaafar etal.. 1992: Broderson etal.. 1976: Doig and Willoughbv.
1971: Coon etal.. 1970: Anderson etal.. 1964a) (Table 1-4 and Appendix C, Section C.3). In general,
responses in respiratory tissues increased with increasing ammonia exposure concentration.
Based on evidence of respiratory effects in multiple human and animal studies (including
epidemiological studies in different settings and populations), respiratory system effects are
identified as a hazard associated with inhalation exposure to ammonia.
1.2.2. Immune System Effects
A limited number of studies have evaluated the immunotoxicity of ammonia in human
populations and in experimental animal models. Immunological function was evaluated in two
independent investigations of livestock farmers exposed to ammonia via inhalation.
Immunoglobulin G- (IgG) and E-specific (IgE) antibodies for pig skin and urine fCrook etal.. 19911.
elevated neutrophils from nasal washes, and increased white blood cell counts fCormier etal..
20001 were reported. These data on immunological function are suggestive of immunostimulatory
effects; however, the test subjects were also exposed to a number of other respirable agents in
addition to ammonia, such as endotoxin, bacteria, fungi, and mold that are known to stimulate
immune responses. Data in humans following exposure to ammonia only are not available.
Animal studies that examined ammonia immunotoxicity were conducted using short-term
inhalation exposures and were measured by three general types of immune assays: host resistance,
T cell proliferation, and delayed-type hypersensitivity. Immunotoxicity studies of ammonia using
measures of host resistance provide the most relevant data for assessing immune function since
they directly measure the ability of the immune system to control microorganism growth. Other
available studies of ammonia employed assays that evaluated immune function. Changes in
immune cell populations without corresponding functional data are considered to be the least
predictive, and studies that looked only at these endpoints fGustin etal.. 1994: Neumann etal..
19871 were considered less informative and not further considered in evaluating the immune
system effects of ammonia.
Several host resistance studies utilized lung pathogens to assess bacterial clearance
following ammonia exposure; however, these studies were not designed to discriminate between
direct immunosuppression associated with ammonia exposure or immune effects secondary to
damage to the protective mucosal epithelium of the respiratory tract. The available studies also do
not correlate increased bacterial colonization with reduced immune function. Lung lesions, both
gross and microscopic, were positively correlated with ammonia concentration in F344 rats
continuously exposed to ammonia in an inhalation chamber for 7 days prior to inoculation with 108
This document is a draft for review purposes only and does not constitute Agency policy.
1-20 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
colony forming units [CFU] of Mycoplasma pulmonis followed by up to 42 days of ammonia
exposure post inoculation fBroderson etal.. 19761. (Inoculation with the respiratory pathogen
M. pulmonis causes murine respiratory mycoplasmosis [MRM] characterized by lung lesions.) The
incidence of lung lesions was significantly increased at ammonia concentrations >35 mg/m3,
suggesting that ammonia exposure decreased bacterial clearance resulting in the development of M.
pulmonis-induced MRM. However, increasing ammonia concentration was not associated with
increased CFU of M. pulmonis isolated from the respiratory tract The high number of inoculating
CFU could have overwhelmed the innate immune response and elicited a maximal response that
could not be further increased in immunocompromised animals.
Conversely, significantly increased CFU of M. pulmonis bacteria isolated in the trachea, nasal
passages, lungs, and larynx were observed in F344 rats continuously exposed to 71 mg/m3
ammonia for 7 days prior to M. pulmonis (104-106 CFU) inoculation and continued for 28 days post
inoculation fSchoeb etal.. 19821. This increase in bacterial colonization indicates a reduction in
bacterial clearance following exposure to ammonia. Lesions were not assessed in this study.
0F1 mice exposed to 354 mg/m3 ammonia for 7 days prior to inoculation with a 50% lethal
dose (LD50) of Pasteurella multocida exhibited significantly increased mortality compared to
controls (86% versus 50%, respectively); however, an 8-hour exposure was insufficient to affect
mortality (Richard et al.. 19781. The authors suggested that the irritating action of ammonia
destroyed the tracheobronchial mucosa and caused inflammatory lesions thereby increasing
sensitivity to respiratory infection with prolonged ammonia exposure.
Pig studies support the findings observed in the rodent studies that ammonia exposure
increases the colonization of respiratory pathogens. Andreasen et al. (2000a) demonstrated that
63 days of ammonia exposure increased the number of bacterial positive nasal swabs following
inoculation with P. multocida and Mycoplasma hyopneumoniae; however, the effect was not dose
responsive and did not result in an increase in lung lesions. Additional data obtained from pigs
suggest that ammonia exposure eliminates the commensal flora of the nasal cavities, which allows
for increased colonization of P. multocida; however, this effect abates following cessation of
ammonia exposure fHamilton et al.. 1999: Hamilton et al.. 1998).
Suppressed cell-mediated immunity and decreased T cell proliferation was observed
following ammonia exposure. Using a delayed-type hypersensitivity test to evaluate cell-mediated
immunity, Hartley guinea pigs were vaccinated with Mycobacterium bovis bacillus Calmette-Guerin
(BCG) and exposed to ammonia followed by intradermal challenge with a purified protein
derivative (PPD). Dermal lesion size was reduced in animals exposed to 64 mg/m3 ammonia,
indicating immunosuppression fTargowski etal.. 19841. Blood and bronchial lymphocytes
harvested from naive guinea pigs treated with the same 3-week ammonia exposure and stimulated
with phytohaemagglutinin or concanavalin A demonstrated reduced T cell proliferation (Targowski
etal.. 1984). Bactericidal activity in alveolar macrophages isolated from ammonia-exposed guinea
pigs was not affected. Lymphocytes and macrophages isolated from unexposed guinea pigs and
treated with ammonia in vitro showed reduced proliferation and bactericidal capacity only at
This document is a draft for review purposes only and does not constitute Agency policy.
1-21 DRAFT—DO NOT CITE OR QUOTE
-------
1 concentrations that reduced viability, indicating nonspecific effects of ammonia-induced
2 immunosuppression fTargowski etal.. 19841. These data suggest that T cells may be the target of
3 ammonia exposure since specific macrophage effects were not observed.
4 The evidence of immune system effects in experimental animals exposed to ammonia is
5 summarized in Table 1-5 and as an exposure-response array in Figure 1-2.
6
Table 1-5. Evidence pertaining to immune system effects in animals
Study design and reference
Resu Its
Host resistance
Broderson et al. (1976)
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
% of animals with gross lung lesions: 16, 46, 66*, 33,
and 83%
No effect on CFU.
Schoeb et al. (1982)
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
1" bacterial colonization (as a result of reduced
bacterial clearance).
Richard et al. (1978)
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
% Mortality: 50 and 86%*
Andreasen et al. (2000a)
Landrace X large white pigs; 10/group (sex unknown)
<5 (control), 50, or 100 ppm (3.5, 35, or 71 mg/m3), 63 d
(continuous exposure) inoculated with M. hyopneumoniae
on day 9 and P. multocida on d 28, 42, and 56
% of animals with positive day 49 nasal swab:
24,100*, and 90%*
Hamilton et al. (1998)
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
1" bacterial colonization
Hamilton et al. (1999)
Large white pigs; 5/group (sex unknown)
0 or 50 ppm (0 or 35 mg/m3), 1 wk pre-inoculation with P.
multocida, 3 wks post-inoculation
1" bacterial colonization
Bacteria isolated from nasal cavities: 3.18 and 4.30*
CFU
T cell proliferation
Targowski et al. (1984)
Hartley guinea pig; 8/group (sex unknown)
<15, 50, or 90 ppm (<11 [control], 35, or 64 mg/m3), 3 wks
(continuous exposure)
4/ proliferation in blood and bronchial T cells.
This document is a draft for review purposes only and does not constitute Agency policy.
1-22 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-5. Evidence pertaining to immune system effects in animals
Study design and reference
Resu Its
Delayed-type hypersensitivity
Targowski et al. (1984)
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
Mean diameter of dermal lesion (mm): 12,12.6, and
8.7*
^Statistically significantly different from the control (p < 0.05).
1
This document is a draft for review purposes only and does not constitute Agency policy.
1-23 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review of Ammonia
1000
E
"m 100
E
I/)
C
o
as
I—
c
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Summary of Immune System Effects
The evidence for ammonia immunotoxicity is based on epidemiological and animal studies.
Available epidemiological studies that addressed immunological function are confounded by
exposures to a number of other respirable agents that have been demonstrated to be
immunostimulatory. Single-exposure human studies of ammonia evaluating immune endpoints are
not available. Therefore, human studies are not particularly informative for evaluating whether
ammonia has immunotoxic properties.
Animal studies provide consistent evidence of elevated bacterial growth following ammonia
exposure. This is supported by observations of lung lesions (Broderson etal.. 19761. elevated CFU
fSchoeb et al.. 19821. and increased mortality (Richard etal.. 19781 in rats or mice exposed to
ammonia; however, the findings from the Broderson etal. (19761 study (which described the
percent of animals with gross lesions) were not dose-responsive, and the other studies used single
concentrations of ammonia and therefore did not provide information on dose-response. A single
study suggested that T cells are inhibited by ammonia (Targowski etal.. 19841. but the data were
not dose responsive.
Overall, there are suggestions that ammonia exposure may be associated with
immunotoxicity, but it is unclear if elevated bacterial colonization is the result of damage to the
protective mucosal epithelium of the respiratory tract or the result of suppressed immunity.
Therefore, there is inadequate information to draw a conclusions about the immune system as a
potential hazard of ammonia exposure.
1.2.3. Other Systemic Effects
The majority of information suggests that ammonia induces effects in and around the portal
of entry. As discussed below, there is limited evidence from experimental animals that ammonia
can produce effects on organs distal from the portal of entry, including the liver, kidney, spleen, and
heart.
Evidence of liver toxicity in animals comes from observations of histopathological
alterations in the liver. Histopathologic changes described as "fatty changes of the liver plate cells"
were reported at an exposure concentration of 470 mg/m3 ammonia in rats, guinea pigs, rabbits,
dogs, and monkeys following the same subchronic inhalation exposure regimens (Coon etal..
19701: this concentration was lethal to approximately 25% of exposed guinea pigs and the majority
of exposed rats. Congestion of the liver was reported in guinea pigs following inhalation exposure
to 35 mg/m3 for 42 days and 120 mg/m318 weeks fAnderson et al.. 1964a: Weatherbv. 19521: no
liver effects were observed in similarly exposed mice at 14 mg/m3 (Anderson etal.. 1964a).
Experimental animal studies provide some evidence that inhaled ammonia can affect the
kidney and spleen. Alterations in the kidneys (calcification and proliferation of tubular epithelium)
were reported in rats, rabbits, guinea pigs, monkeys, and dogs exposed to 470 mg/m3, an ammonia
concentration that was lethal to rats and guinea pigs fCoon etal.. 19701. "Congestion" of the
kidneys and spleen was reported in four guinea pigs exposed to 120 mg/m3 ammonia for 18 weeks
(but not 6 or 12 weeks) fWeatherbv. 19521. Enlarged and "congested" spleens were reported in
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
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
guinea pigs exposed to 35 mg/m3 ammonia for 6 weeks (Anderson et al.. 1964a). None of these
studies provided incidence of histopathologic lesions.
Myocardial fibrosis was observed in monkeys, dogs, rabbits, guinea pigs, and rats following
subchronic inhalation exposure to 470 mg/m3 ammonia, a concentration lethal to exposed guinea
pigs and rats; no changes were observed at lower concentrations fCoon etal.. 19701. At the same
concentration, ocular irritation (characterized as heavy lacrimation, erythema, discharge, and
ocular opacity of the cornea) was also reported by Coon et al. (1970) in small numbers of dogs and
rabbits, but was not observed in similarly exposed monkeys or rats.
"Early degenerative changes" in the adrenal gland were reported in four guinea pigs
exposed to 120 mg/m3 ammonia by inhalation for 18 weeks, but not in guinea pigs exposed for 6 or
12 weeks (Weatherbv. 1952). With the exception ofBroderson etal. (1976). no other investigators
examined effects on the adrenal gland following exposure to inhaled ammonia, and Broderson etal.
(1976) did not describe effects on nonrespiratory tissues. These limited findings are insufficient to
draw conclusions about possible effects of ammonia on the adrenal gland.
As discussed above, Coon etal. (1970) reported effects on the liver, kidney, and heart
following continuous exposure to 470 mg/m3; however, no histopathological changes were
observed in rats, guinea pigs, rabbits, dogs, or monkeys when these animals were repeatedly, but
not continuously, exposed to ammonia even at high concentrations (e.g., 770 mg/m3 for
8 hours/day, 5 days/week; Table 1-6). These findings suggest that animals can recover from
intermittent exposure to elevated ammonia levels (Coon etal.. 1970). although the evidence to
support this observation is limited.
Additionally, there is limited evidence of biochemical or metabolic effects of acute or short-
term ammonia exposure. Evidence of slight acidosis, as indicated by a decrease in blood pH, was
reported in rats exposed to 18 or 212 mg/m3 ammonia for 5 days; the study authors stated that
differences in pH leveled off at 10 and 15 days fManninen etal.. 1988). In another study, blood pH
in rats was not affected by exposure to ammonia at concentrations up to 818 mg/m3 for up to
24hours (Schaerdel etal.. 1983b).
Encephalopathy related to ammonia may occur in humans following disruption of the
body's normal homeostatic regulation of the glutamine and urea cycles, e.g., due to severe liver
disease resulting in elevated ammonia levels in blood (Minana etal.. 1995: Souba. 1987). Acute
inhalation exposure studies have identified alterations in amino acid levels and neurotransmitter
metabolism (including glutamine concentrations) in the brain of rats and mice fManninen and
Savolainen. 1989: Manninen et al.. 1988: Sadasivudu etal.. 1979: Sadasivudu and Radha Krishna
Murthv. 1978). It has been suggested that glutamate and y-amino 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 inhalation exposure.
In the only study of the reproductive and developmental toxicity of ammonia, 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 fDiekman et al.. 19931. A control group without ammonia exposure was
This document is a draft for review purposes only and does not constitute Agency policy.
1-26 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
not evaluated. 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 5- versus 25-mg/m3
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 with no ammonia exposure and possible confounding by
exposures to bacterial and mycoplasm pathogens.
The evidence of systemic toxicity in experimental animals exposed to ammonia is
summarized in Table 1-6 and as an exposure-response array in Figure 1-3.
Table 1-6. Evidence pertaining to other systemic effects in animals
Study design and reference
Resu Its
Liver effects
Coon et al. (1970)
Sprague-Dawley and Long-Evans rat; male and female;
15/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
0, 155, or 770 mg/m3 for 8 hrs/d, 5 d/wk for 6 wks
No histopathologic changes observed.
Coon et al. (1970)
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
"Fatty changes of the liver plate cells" in several
animals of each species at 470 mg/m3.a
Coon et al. (1970)
Sprague-Dawley or Long-Evans rat; male and female; 15-
51/group
0 or 40 mg/m3 for 114 d, 127, 262, or 470 mg/m3 for 90 d, or
455 mg/m3 for 65 days
"Fatty changes of the liver plate cells" in several
rats at 470 mg/m3, an exposure that was lethal
to the majority of the rats.3
Anderson et al. (1964a)
Swiss albino mouse; male and female; 4/exposure interval
0 or 20 ppm (0 or 14 mg/m3) for 7,14, 21, 28, or 42 d
No visible signs of liver toxicity.
Weatherbv (1952)
Guinea pig (strain not specified); male; 2 control and 4
exposed/exposure interval
0 or 170 ppm (0 or 120 mg/m3) for 6 hrs/d, 5 d/wk for 6, 12 or
18 wks
Congestion of the liver at 18 wks, not reported
at earlier times.3
Anderson et al. (1964a)
Guinea pig (strain not specified); male and female; 2/exposure
interval at 20 ppm, 6/exposure interval at 50 ppm
0 or 20 ppm (0 or 14 mg/m3) for 7,14, 21, 28, or 42 d or 50 ppm
(35 mg/m3) for 42 d
Congestion of the liver at 35 mg/m3 for 42 d.3
This document is a draft for review purposes only and does not constitute Agency policy.
1-27 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-6. Evidence pertaining to other systemic effects in animals
Study design and reference
Resu Its
Adrenal gland effects
Weatherbv (1952)
Guinea pig (strain not specified); male; 2 control and 4
exposed/exposure interval
0 and 170 ppm (0 and 120 mg/m3) 6 hrs/d, 5 d/wk for 6,12, or
18 wks
"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
Kidney and spleen effects
Coon et al. (1970)
Sprague-Dawley and Long-Evans rat; male and female;
15/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
0,155, or 770 mg/m3 for 8 hrs/d, 5 d/wk for 6 wks
No histopathologic changes reported.
Coon et al. (1970)
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Calcification and proliferation of renal tubular
epithelium at 470 mg/m3.a (This exposure was
lethal to ~25% of guinea pigs.)
Coon et al. (1970)
Sprague-Dawley or Long-Evans rat; male and female; 15-
51/group
0 or 40 mg/m3 for 114 d, 127, 262, or 470 mg/m3 for 90 d, or
455 mg/m3 for 65 d
Calcification and proliferation of renal tubular
epithelium at 470 mg/m3, an exposure that was
lethal to the majority of the rats.3
Anderson et al. (1964a)
Swiss albino mouse; male and female; 4/exposure interval
0 or 20 ppm (0 or 14 mg/m3) for 7,14, 21, 28, or 42 d
No visible signs of toxicity.
Weatherbv (1952)
Guinea pig (strain not specified); male; 2 control and 4
exposed/exposure interval
0 or 170 ppm (0 or 120 mg/m3) 6 hrs/d, 5 d/wk for 6,12, or
18 wks
Congestion of the spleen and kidneys.3
Anderson et al. (1964a)
Guinea pig (strain not specified); male and female; 2/exposure
interval at 20 ppm, 6/exposure interval at 50 ppm
0 or 20 ppm (0 or 14 mg/m3) for 7,14, 21, 28, or 42 d or 50 ppm
(35 mg/m3) for 42 d
Enlarged and congested spleens at 35 mg/m3.3
This document is a draft for review purposes only and does not constitute Agency policy.
1-28 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-6. Evidence pertaining to other systemic effects in animals
Study design and reference
Resu Its
Myocardial effects
Coon et al. (1970)
Sprague-Dawley and Long-Evans rat; male and female;
15/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
0,155, or 770 mg/m3 for 8 hrs/d, 5 d/wk for 6 wks
No histopathologic changes reported.
Coon et al. (1970)
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Myocardial fibrosis at 470 mg/m3.a (This
exposure was lethal to ~25% of guinea pigs.)
Coon et al. (1970)
Sprague-Dawley or Long-Evans rat; male and female; 15-
51/group
0 or 40 mg/m3 for 114 d, 127, 262, or 470 mg/m3 for 90 d, or
455 mg/m3 for 65 d
Myocardial fibrosis at 470 mg/m3, an exposure
that was lethal to the majority of the rats.3
Ocular effects
Coon et al. (1970)
Princeton-derived guinea pig; male and female; 15/group
Squirrel monkey (S. sciureus); male; 3/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
No ocular irritation reported.
Coon et al. (1970)
Sprague-Dawley and Long-Evans rat; male and female;
15/group
New Zealand albino rabbit; male; 3/group
Princeton-derived guinea pig; male and female; 15/group
Squirrel monkey (S. sciureus); male; 3/group
Beagle dog; male; 2/group
0,155, or 770 mg/m3 for 8 hrs/d, 5 d/wk for 6 wks
No ocular irritation reported.
Coon et al. (1970)
Sprague-Dawley and Long-Evans rat; male and female; 15-
51/group
0 or 40 mg/m3 for 114 d, 127, 262, or 470 mg/m3 for 90 d, or
455 mg/m3 for 65 d
No ocular irritation reported.
Coon et al. (1970)
New Zealand albino rabbit; male; 3/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Erythema, discharge, and ocular opacity over
Zi-Zi of cornea at 470 mg/m3.a
Coon et al. (1970)
Beagle dog; male; 2/group
0 or 40 mg/m3 for 114 d or 470 mg/m3 for 90 d
Heavy lacrimation at 470 mg/m3.a
This document is a draft for review purposes only and does not constitute Agency policy.
1-29 DRAFT—DO NOT CITE OR QUOTE
-------
Table 1-6. Evidence pertaining to other systemic effects in animals
Study design and reference
Resu Its
Blood pH changes
Manninen et al. (1988)
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
4/ blood pH at 5 days; pH differences "leveled
off at later time points (data not shown)".
Blood pH (day 5): 7.43, 7.34*, 7.36*
Schaerdel et al. (1983b)
CrkCOBS CD(SD) rat; male; 8/group [blood pC>2 based on n = 5]
15, 32, 310, or 1,157 ppm (11, 23, 219, or 818 mg/m3) for
0 (control), 8,12, or 24 hrs
1" blood pO2 at 11 and 23 mg/m3 at 8-, 12-, and
24-hr time points; no change at higher
concentrations; no change in blood pH.
Percent change in pC>2 from time 0 (at 24 hours
of exposure0-. 20*, 17*, 1, -2%
Amino acid levels and neurotransmitter metabolism in the brain
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 d
% change compared to control:c
Brain glutamine: 42*, 40*%
Manninen et al. (1988)
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
% change compared to control at 212 mg/m3:c
Blood glutamine (5,10,15 d): 44*, 13,14%
Brain glutamine (5,10,15 d): 40*, 4, 2%
Reproductive and developmental effects
Diekman et al. (1993)
Crossbred gilt (female pig); 4.5 mo 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 6 wksd
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).
incidence data not provided.
bMeasurements at time zero were used as a control; the study did not include an unexposed control group.
cPercent change compared to control calculated as: (treated value - control value)/control value x 100.
dA 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.
^Statistically significantly different from the control (p < 0.05).
1
This document is a draft for review purposes only and does not constitute Agency policy.
1-30 DRAFT—DO NOT CITE OR QUOTE
-------
Exposure concentrations (mg/m3)
3
to'
Ei-
O
n
c
CX)
3
Co
S3
Ci-
I
-s
3
"Q
~3
00 S
^ §
Q
3
Ci-
o
to
3
O
o
O
o
3
to
C
(?
CQ
CD
3
"Q
O
t
H
o
o
z
o
H
c~>
H
m
o
73
" 5
Early degenerative changes in the adrenal
gland (guinea pig); Weatherby (1952) *
No histopathological changes (rat, rabbit,
guinea pig, monkey, dog); Coon et al. (1970) *
Calcification & proliferation renal tubular epithelium
(rabbit, guinea pig, monkey, dog); Coon etal. (1970)
Calcification & proliferation of renal
tubularepithelium (rat); Coon etal. (1970)
No kidney orspleeneffects(mouse);
Anderson et al. (1964)
Congestion ofthespleen & kidneys (guinea pig);
Weatherby (1952)*
Enlarged & congested spleen (guinea pig);
Anderson et al. (1964)
No histopathological changes (rat, rabbit,
guinea pig, monkey, dog); Coon et al. (1970) *
Myocardial fibrosis (rabbit, guinea pig,
monkey, dog); Coon et al.(1970)
Myocardial fibrosis (rat);
Coon et al. (1970)
No irritation observed (guinea pig, monkey);
Coon et al. (1970)
No irritation observed (rat, rabbit, guinea pig,
monkey, dog); Coon et al.(1970) *
No irritationobserved (rat);
Coon et al. (1970)
Erythema, discharge & corneal opacity (rabbit);
Coon etal. (1970)
Heavy lacrimation(dog);
Coon et al. (1970)
O
>
O
>
>
0-
E>-
-E>
|
8
sr
i>
n>
s
05'
0
1
§
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Summary of Other Systemic Effects
Effects of ammonia exposure on organs distal from the portal of entry (systemic effects) are
based on evidence in animals. Effects on various organs, including liver, kidney, spleen, and heart,
were observed in several studies that examined responses to ammonia exposure in a number of
laboratory animal species. While effects on many of these organs were observed in multiple
species, including monkey, dog, rabbit, guinea pig, and rat, effects were not consistent across
exposure protocols. Evidence of ocular irritation in experimental animals was inconsistently
observed, and then only at high ammonia concentrations (470 mg/m3).
Studies of ammonia toxicity that examined other systemic effects were all published in the
older toxicological literature. Three subchronic inhalation studies were published between 1952
and 1970 (Coon etal.. 1970: Anderson et al.. 1964a: Weatherbv. 19521. In general, the information
from these studies is limited by small group sizes, minimal characterization of reported
histopathological changes (e.g., "congestion," "enlarged," "fatty liver"), insufficiently detailed
reporting of study results, and incomplete, if any, incidence data. In addition, Weatherbv
(1952). Anderson et al. (1964a). and some of the experiments reported by Coon etal. (1970) used
only one ammonia concentration in addition to the control, so no dose-response information is
available from the majority of experimental studies to inform the evidence for systemic effects of
ammonia. Finally, exposure characterization in Weatherbv (1952) was considered poor.
Overall, there are suggestions in experimental animals that ammonia exposure may be
associated with effects on organs distal from the portal of entry, but there is inadequate
information to draw conclusions about the liver, kidney, spleen, or heart as sensitive targets of
ammonia toxicity.
Given the inadequacies of the available toxicology literature for other systemic effects, the
potential toxicity of inhaled ammonia at sites distal from the respiratory system was evaluated by
considering ammonia levels normally present in blood. As discussed in more detail in Appendix C,
Section C.1.2, ammonia is produced endogenously in all human and animal tissues during fetal and
adult life. In adults, the normal range of ammonia in venous blood is 0.1-0.8 |ig/ml. Concentrations
in fetal circulation are higher than maternal blood concentrations; two studies reported that mean
umbilical concentrations of ammonia in venous blood at delivery were 50% to threefold higher
than mean concentrations in maternal blood, with umbilical concentrations ranging from
approximately 0.5-5 ng/ml flozwik etal.. 2005: DeSanto etal.. 19931. Human fetal umbilical blood
levels of ammonia at birth were not influenced by gestational age based on deliveries ranging from
gestation week 25 to 43 f DeSanto etal.. 19931.
At external concentrations that do not measurably change normal (baseline) levels of
ammonia, the likelihood is low that exposures would pose a hazard for systemic effects. In rats,
exposure to ammonia concentrations <18 mg/m3 did not produce a statistically significant change
in blood or brain ammonia co nee ntratio ns f M a n nine n etal.. 1988: Schaerdel etal.. 1983bl. Higher
external ammonia concentrations (>212 mg/m3) were associated with elevated blood ammonia
levels, but even at these relatively high concentrations, experimental findings in rats indicate that
compensation readily occurs fManninen etal.. 19881. In a 24-hour exposure duration study, blood
This document is a draft for review purposes only and does not constitute Agency policy.
1-32 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
ammonia concentrations at 12 hours of exposure to >219 mg/m3 ammonia in air were lower than
at 8 hours; in a second 15-day exposure duration study, blood ammonia concentrations that were
elevated on day 5 of exposure to 212 mg/m3 ammonia in air were not significantly different from
control values on days 10 and 15 of exposure fSchaerdel etal.. 1983a). See Appendix C, Section
C.1.3, Metabolism/Endogenous Production of Ammonia, for a more detailed summary of the
available literature that describes the relationship between environmental ammonia
concentrations and blood ammonia levels. Therefore, the available experimental data suggest that
any changes in blood ammonia at external concentrations <18 mg/m3 would be small relative to
levels normally present in blood. The potential for systemic effects (i.e., on tissues/organs distal
from the respiratory system), including reproductive and developmental effects, at these
concentrations cannot be ruled out, but the likelihood of such effects is considered small.
Because the health effects literature identified the respiratory system as the primary target
of ammonia toxicity, EPA also considered the possibility that point of contact effects could translate
into effects on tissues or organs distal from the respiratory system. EPA is not aware of any
mechanisms by which point of contact effects could directly or indirectly impact distal tissues or
organs.
1.3. SUMMARY AND EVALUATION
1.3.1. Weight of Evidence for Effects Other than Cancer
The respiratory system is the primary and most sensitive target of inhaled ammonia toxicity
in humans and experimental animals. Evidence for respiratory system toxicity in humans comes
from cross-sectional occupational studies in industrial settings that reported changes in lung
function and an increased prevalence of respiratory symptoms. The findings of respiratory effects
in workers exposed to ammonia as a disinfectant or cleaning product (primarily studies of asthma
or asthma symptoms), studies in agricultural settings (primarily lung function studies), controlled
human exposure studies, and case reports of injury following acute exposure provide additional
evidence that the respiratory system is a target of inhaled ammonia. Short-term and subchronic
animal studies show respiratory effects in several animal species across different dose regimens.
Thus, the weight of evidence of observed respiratory effects observed across multiple human and
animal studies identifies respiratory system effects as a hazard from ammonia exposure.
Evidence for an association between inhaled ammonia exposure and effects on other organ
systems distal from the portal of entry is less compelling than for the respiratory system. Overall,
there are suggestions in experimental animals that ammonia exposure may be associated with
effects on the liver, kidney, spleen, or heart, but the available information is inadequate to draw
conclusions. The two epidemiological studies that addressed immunological function are
confounded by exposures to a number of other respirable agents that have been demonstrated to
be immunostimulatory and provide little support for ammonia immunotoxicity. Animal studies
provide consistent evidence of elevated bacterial growth following ammonia exposure. It is
unclear, however, whether elevated bacterial colonization is the result of suppressed immunity or
This document is a draft for review purposes only and does not constitute Agency policy.
1-33 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
damage to the barrier provided by the mucosal epithelium of the respiratory tract. Overall, the
weight of evidence does not support the immune system as a target of ammonia toxicity.
Studies of the potential reproductive or developmental toxicity of ammonia in humans are
not available. Reproductive effects were not associated with inhaled ammonia in the only animal
study that examined the reproductive effects of ammonia (i.e., a limited-design inhalation study in
the pig). As discussed in Section 1.2.3, ammonia is produced endogenously in human and animal
tissues during fetal and adult life. Although the potential for effects on reproduction and the
developing fetus cannot be ruled out at external concentrations that do not alter normal blood or
tissue ammonia levels, there is no evidence that raises concerns for the developing fetus or
reproduction or to other distal tissues/organs.
1.3.2. Susceptible Populations and Lifestages
Studies of the toxicity of ammonia in children or young animals that would support an
evaluation of childhood susceptibility are limited. Casas etal. (2013) found evidence of airway
inflammation (as indicated by increased exhaled nitric oxide) and decreased lung function in
school-age children exposed to cleaning products.
Because the respiratory system is a target of ammonia toxicity, individuals with respiratory
disease (e.g., asthmatics) might be expected to be a susceptible population. Loftus etal. (2015)
reported no increase in asthma symptoms and medication use in asthmatic children living near
animal feeding operations; however, ammonia exposure was associated with lower FEVi.
Controlled human exposure studies that examined both healthy adult volunteers and volunteers
with asthma (Petrova etal.. 2008: Sigurdarson et al.. 2004) did not demonstrate greater respiratory
sensitivity in asthmatics than healthy volunteers after acute exposure to ammonia. Under longer-
term exposure conditions, however, as seen among livestock farmers, one study observed
associations between ammonia exposure and decreased lung function among workers with chronic
respiratory symptoms, but not among the asymptomatic workers fPreller etal.. 19951. Additional
research focusing on the question of susceptibility and variability in response to ammonia exposure
in these populations is needed.
Individuals with disease conditions that lead to hyperammonemia, a condition of elevated
levels of circulating ammonia, may be more susceptible to the effects of ammonia from external
sources. Hyperammonemia can occur in individuals with severe diseases of the liver (e.g.,
cirrhosis) or kidney, organs that biotransform and excrete ammonia, urea cycle disorders, and
other conditions such as fatty acid oxidation defects and Reye syndrome fBiirki etal.. 2015: Auron
and Brophv. 2012: Romero-Gomez etal.. 2004: Cordoba etal.. 1998: Davies etal.. 1997: Schubiger
etal.. 1991: Gilbert. 1988: leffers etal.. 1988: Souba. 1987). Elevated ammonia levels can
predispose an individual to encephalopathy as a result of the ability of ammonia to cross the blood-
brain barrier and subsequent disturbances in amino acid synthesis and alterations in
neurotransmission systems. Neonates and infants are particularly susceptible to the neurological
effects of elevated levels of ammonia; hyperammonemia can cause irreparable damage to the
developing brain fMinana etal.. 1995: Souba. 19871 fAuron and Brophv. 20121. While patients with
This document is a draft for review purposes only and does not constitute Agency policy.
1-34 DRAFT—DO NOT CITE OR QUOTE
-------
1 hyperammonemia could plausibly be considered a susceptible population, there are no studies that
2 specifically support this hypothesized susceptibility.
3
4
5
This document is a draft for review purposes only and does not constitute Agency policy.
1-35 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
2. DOSE-RESPONSE ANALYSIS
2.1. INHALATION REFERENCE CONCENTRATION FOR EFFECTS OTHER
THAN CANCER
The RfC (expressed in units of mg/m3) is defined as an estimate (with uncertainty spanning
perhaps an order of magnitude) of a continuous inhalation exposure to the human population
(including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects
during a lifetime. It can be derived from a NOAEL, LOAEL, or the 95% lower bound on the
benchmark concentration (BMCL), with UFs generally applied to these PODs to reflect limitations of
the data used.
2.1.1. Identification of Studies and Effects for Dose-Response Analysis
As discussed in Section 1.2, the respiratory system is the primary and most sensitive target
of inhaled ammonia in humans and experimental animals, and respiratory effects have been
identified as a hazard following inhalation exposure to ammonia. The experimental toxicology
literature for ammonia provides evidence that inhaled ammonia may be associated with toxicity to
target organs other than the respiratory system, including the liver, kidney, spleen, heart, and
immune system. Effects in these other (nonrespiratory) target organs were not considered as the
basis for RfC derivation because the evidence for these associations is weak relative to that for
respiratory effects.
Respiratory effects, characterized as increased prevalence of respiratory symptoms or
decreased lung function, have been observed in worker populations exposed to ammonia
concentrations >18.5 mg/m3 (Rahman etal.. 2007: Ali etal.. 2001: Ballal etal.. 1998). Decrements
in lung function parameters and increased prevalence of respiratory symptoms, such as wheezing,
chest tightness, and cough/phlegm, have been identified as adverse respiratory health effects by
the American Thoracic Society (ATS. 2000) and are similarly noted as adverse in the EPA's Methods
for Derivation of Inhalation Reference Concentrations and Application of Inhalation Dosimetry (U.S.
EPA. 1994). At the population level, ATS (2000) stated that "any detectable level of permanent
pulmonary function loss attributable to air pollution exposure should be considered as adverse"
and that
It should be emphasized that a small but significant reduction in a population mean
FEVi or FEV0.75 is probably medically significant, as such a difference may indicate
an increase in the number of persons with respiratory impairment in the
population. In other words, a small part of the population may manifest a marked
change that is medically significant to them, but when diluted with the rest of the
population the change appears to be small fATS. 20001.
This document is a draft for review purposes only and does not constitute Agency policy.
2-1 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Thus, even small changes in the average (mean) of a distribution of pulmonaiy function parameters
is considered adverse for purposes of deriving an RfC.
In general human data are preferred over animal data for deriving reference values because
these data are more relevant for assessing human health effects than animal studies and avoid the
uncertainty associated with interspecies extrapolation when animal data serve as the basis for the
RfC. In the case of ammonia, the available occupational studies provide adequate data for the
quantitative analysis of health outcomes considered relevant to potential general population
exposures. Respiratory effects have also been observed in animals, but at ammonia concentrations
higher than those associated with respiratory effects in humans and in studies involving exposure
durations (up to 114 days) shorter than those in occupational studies (Section 1.2.1). Therefore,
data on respiratory effects in humans were used for the derivation of the RfC and respiratory
effects in animals were not further considered.
Of the available human data, associations between ammonia exposure and respiratory
effects have been examined in epidemiology studies of industrial worker populations (Table 1-2), in
studies of ammonia exposure in a cleaning setting (Table 1-3), and in studies of populations in
agricultural settings. Studies using ammonia as a cleaning product provide evidence of an
association between ammonia exposure and increased risk of asthma; however, these studies did
not measure ammonia concentrations and thus are not useful for dose-response analysis. Studies
in agricultural settings also support an association between ammonia exposure and decreased
pulmonary function; however, because of co-exposures to other agents (including dust, endotoxin,
mold, and disinfectant products) and the availability of studies with fewer co-exposures, studies in
agricultural settings were considered to be supportive of the association between ammonia
exposure and respiratory effects but were not carried forward for dose-response analysis. In
addition, several controlled-exposure studies in volunteers evaluated the effects of ammonia on
irritation and lung function following acute exposures. These human exposure studies have several
methodological strengths compared to epidemiological studies of worker populations, including
well characterized exposures and resistance to confounding; however, the short exposure
durations used in these studies (i.e., 15 seconds to 6 hours) make them inappropriate for evaluating
the effects of chronic exposure to ammonia.
Of the available studies of ammonia exposure in industrial settings, four cross-sectional
epidemiology studies of industrial worker populations—three studies in urea fertilizer plants
by Rahman et al. (2007). Ballal etal. (1998). and Ali etal. (2001). and a study in a soda ash plant
by Holness etal. f 19891—provide information useful for examining the relationship between
chronic ammonia exposure and increased prevalence of respiratory symptoms and/or decreased
lung function. Bhat and Ramaswamv (1993) evaluated lung function in ammonia plant workers,
but did not measure ammonia concentrations in workplace air. Therefore, this study was not
considered useful for RfC derivation.
In general, these four cross-sectional occupational studies provide a coherent set of
estimated NOAELs and effect levels, and are considered candidate principal studies for RfC
derivation. A brief description of these studies and the contribution of each to the understanding of
This document is a draft for review purposes only and does not constitute Agency policy.
2-2 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
the dose-response relationship between ammonia exposure and respiratory effects follows. More
study details are provided in the Supplemental Information, Section C.2.1 and in Table 1-2, and
evaluation of the strengths and limitations are more fully considered in the Literature Search
Strategy | Study Selection and Evaluation section.
• Rahman etal. (2007) observed an increased prevalence of respiratory symptoms
(coughing, chest tightness) in urea fertilizer plant workers (mean employment
duration: 16 years) exposed to a mean ammonia concentration of 18.5 mg/m3 (range:
9-31 mg/m3), but not in workers in a second plant exposed to a mean ammonia
concentration of 4.9 mg/m3 (range: 2-8 mg/m3). Decrements in lung function (FVC and
FEVi) between pre- and post-shift in the high-exposure group (2-3%) were statistically
significant Exposure was measured by personal samples using two different analytical
methods.
• Ballal etal. (1998) observed an increased prevalence of respiratory symptoms (cough,
phlegm, wheezing, and dyspnea) among urea fertilizer factory workers (mean
employment duration: 4.3 years) in one factory (Factory A) with ammonia exposures
ranging from 2-27.1 mg/m3,10 but no increase in symptoms in another factory (Factory
B) with exposures ranging from 0.02-7 mg/m3. Lung function was not measured.
• A companion study by Ali etal. (2001) examined lung function among workers in
Factory A from Ballal etal. (1998): respiratory symptoms were not evaluated. Workers
with cumulative exposure >50 mg/m3-years had significantly lower lung function values
(declines of 5-7% in FVC% predicted and FEVi% predicted) than workers with
cumulative exposure <50 mg/m3-years. In this and the Ballal etal. f 19981 study,
exposure was measured by air monitors.
• Holness etal. f 19891 found no differences in the prevalence of respiratory symptoms or
lung function between soda ash plant workers (mean exposure 6.5 mg/m3; mean
exposure duration of 12.2 years) and the control group, and also no differences in
respiratory symptoms or lung function when workers were stratified by ammonia
exposure level (lowest exposure group, <4.4 mg/m3; middle exposure group, 4.4-
8.8 mg/m3; highest exposure group, >8.8 mg/m3). Exposure was measured by personal
samples. EPA identified the concentration range for the high-exposure group (i.e., >8.8
mg/m3) as the NOAEL from this study. The authors stated that 3 of the 12 workers in
the high-exposure group were exposed to concentrations >17.7 mg/m3; therefore, the
majority of workers in the high-exposure group (9 of 12) would have been exposed to
ammonia concentrations in the range of 8.8-17.7 mg/m3.
10This concentration range does not include exposures in the urea store (number of employees = 6; range of
ammonia concentrations = 90-130.4 mg/m3) because employees in this area were required to wear full
protective clothing, thus minimizing potential exposure.
This document is a draft for review purposes only and does not constitute Agency policy.
2-3 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
In selecting the principal study for RfC derivation, consideration was given to exposure
measures, assessment of outcomes, potential for co-exposures, and the value of the NOAEL. Of the
four candidate principal studies, higher confidence was associated with the exposure measures
from Holness etal. (1989). Both Holness etal. (1989) and Rahman etal. (2007) collected personal
air samples, but confidence in the analytical method used by Holness etal. T19891 is higher than
that used by Rahman etal. f2007I Rahman etal. f20071 used two analytical methods for
measuring ammonia concentrations in workplace air (i.e., Drager PAC III and Drager tube);
concentrations measured by the two methods differed by four- to fivefold, indicating some
uncertainty across the two measurement methods, although ammonia concentrations measured by
the two methods were strongly correlated (correlation coefficient of 0.8). In contrast, the Holness
etal. (1989) study used an established analytical method for measuring exposure to ammonia
recommended by the National Institute for Occupational Safety and Health (NIOSH) that involved
the collection of air samples on acid-treated silica gel absorption tubes. Ballal etal. (1998) used
area monitors rather than personal air sampling methods; the latter method provides a better
estimate of an individual's exposure.
As discussed in the Literature Search Strategy | Study Selection and Evaluation section,
assessment of respiratory symptoms in all studies that measured this outcome was based on self-
reporting by questionnaire, and assessment of lung function was performed using standard
spirometry protocols. While considered unlikely, non-blinded outcome assessments of respiratory
symptoms could introduce bias. Therefore, both Holness etal. (1989) and Rahman et al. (2007). the
two studies of industrial populations that examined both respiratory symptoms and lung function,
provide stronger evidence of respiratory effects than studies that evaluated symptoms data only
(notably Ballal etal. (1998)).
Also as discussed in the Literature Search Strategy | Study Selection and Evaluation section,
confounding by other workplace exposures is a potential concern, although not likely to be a major
limitation of the studies considered for dose-response analysis. Only Rahman etal. f20071
measured another workplace chemical (nitrogen dioxide; below detection limits); other studies did
not describe potential co-exposures. Therefore, a more rigorous examination of the potential for
confounding by co-exposure to other workplace chemicals could not be performed. Holness et al.
(1989) noted the high level of control of exposures in the facility used in their study, resulting in
low ammonia levels.
Three of the four occupational studies supported the identification of a NOAEL (or, more
correctly, an exposure range not associated with an increase in respiratory effects). Rahman et al.
(2007) did not observe a change in respiratory effects in workers exposed to a mean ammonia
concentration of 4.9 mg/m3 (range: 2-8 mg/m3). Holness etal. (1989) found no differences in
respiratory effects in soda ash plant workers when compared to a control group or when workers
were stratified by exposure level (low, medium, and high); the concentration range for the high-
exposure group (i.e., >8.8 mg/m3) was identified as the NOAEL. Ballal etal. f 19981 reported no
increase in respiratory symptoms in a factory with exposures ranging from 0.02-7 mg/m3.
Because Ali etal. (2001). the companion study to Ballal etal. f 19981. evaluated only workers in a
This document is a draft for review purposes only and does not constitute Agency policy.
2-4 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
second factory with higher exposures, study findings did not support identification of an estimated
NOAEL.
In light of the above considerations, overall confidence in the Holness etal. T19891 study as
the principal study for RfC derivation was higher than other candidate studies in terms of:
measurement of ammonia exposure, evaluation of both respiratory symptoms and lung function
parameters, smaller potential for co-exposures to other workplace chemicals, and the fact that the
estimated NOAEL for respiratory effects of >8.8 mg/m3 was the highest of the NOAELs estimated
from the candidate principal studies. The Holness etal. (1989) study does not demonstrate a
relationship between ammonia exposure and respiratory effects. The relationship between
ammonia exposure and respiratory effects is based on the body of evidence, and the Holness etal.
(1989) study is identified as the principal study for derivation of the RfC for the reasons given
above.
In summary, the occupational study of ammonia exposure in workers in a soda ash plant
by Holness et al. (1989) was identified as the principal study for RfC derivation, with support
from Rahman etal. f2 007). Ballal etal. f!998). and Ali et al. f2001). and respiratory effects
were identified as the critical effect.
2.1.2. Methods of Analysis
A NOAEL of 13.6 mg/m3, or an estimate of the lower confidence bound of the mean
exposure concentration in the high-exposure group of the Holness (1989) study, was used as
the point of departure (POD) for RfC derivation. The point of departure (POD) for respiratory
effects was based on the NOAEL representing the high-exposure group in Holness etal. f 19891 The
individual subject data from this study were no longer available (call from S. Rieth, U.S. EPA, to C.
Clayton, administrative assistant to Dr. Holness, St Michael's Hospital, Center for Research
Expertise in Occupational Health, Toronto, Canada, February 11, 2015), so that the mean exposure
in the high-exposure group could not be calculated precisely based on the data. Therefore, the
mean was estimated assuming that the data in the study followed a skewed probability distribution,
specifically the lognormal distribution. The frequency distribution provided in Holness etal.
(1989) (see Table 2-1) was used to estimate the parameters (log-scale mean and standard
deviation) of the lognormal distribution that best fit the data.
This document is a draft for review purposes only and does not constitute Agency policy.
2-5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Table 2-1. Frequency distribution of ammonia exposure from Holness (1989)
Exposure group
Interval of exposures
(mg/m3)
Interval of exposures
(ppm)
Number of exposed
workers
Low
0-4.4
0-6.25
34
Medium
4.4-8.8
6.25-12.5
12
High3
8.8-17.7
12.5-25
9
>17.7
>25
3
aEPA divided the high-exposure group into two subgroups based on the statement in Holness et al. (1989):
"Three workers were exposed to TWA concentrations of ammonia in excess of 25 ppm, the current exposure
guideline."
Lognormal parameter estimates were obtained by applying the maximum likelihood
method to this frequency distribution. Using the estimated distribution defined by these parameter
estimates, the estimated mean exposure in the high-exposure group and 95% lower confidence
bound on this mean were calculated as follows. See Appendix C, Section C.4 for detailed
documentation of this calculation.
mean exposure estimate (high-exposure group) = 17.9 mg/m3
95% lower confidence bound on this mean (high-exposure group) = 13.6 mg/m3
The lower confidence bound of 13.6 mg/m3 was used as the POD for respiratory effects.
Because the RfC assumes continuous human exposure over a lifetime, the POD was adjusted
to account for the noncontinuous exposure associated with occupational exposure (i.e., 8-hour
workday and 5-day workweek). Cross-shift data for FVC and FEVi from the Rahman et al. (2007)
study provide some evidence of an immediate effect of ammonia exposure on lung function11, which
could argue against adjustment from noncontinuous to continuous exposure; however, Rahman et
al. f20071 also reported that duration of exposure (using years of employment as a proxy for
exposure duration) was significantly associated with percentage cross-shift decrease in FEVi%. In
addition, Ballal etal. (1998) found a significant correlation between respiratory symptoms (cough,
phlegm, and wheezing) and duration of service (a proxy for exposure duration). In the absence of
clear evidence that respiratory effects in occupationally-exposed populations are an acute
response, and given evidence for contributions of exposure duration (cumulative exposure) to the
respiratory effects of ammonia, the standard adjustment to continuous exposure was applied. The
duration-adjusted POD was calculated as follows:
"Rahman et al. (2007) reported that mean preshift FVC and FEVi values in ammonia and urea plants workers were
similar, suggesting similar lung function in low- and high-exposure workers upon arrival at work. Cross-shift
changes in FVC and FEVi were statistically significant decreased in the urea plant (more highly-exposed) workers
only.
This document is a draft for review purposes only and does not constitute Agency policy.
2-6 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
NOAELadj = NOAEL x VEho/VEh x 5 days/7 days
= 13.6 mg/m3 x 10 m3/20 m3 x 5 days/7 days
= 4.9 mg/m3 or 5 mg/m3 (rounded)
Where:
VEho = human occupational default minute volume (10 m3 breathed during an 8-hour
workday) fU.S. EPA. 19941. This inhalation rate corresponds to more current
inhalation rates for light to moderate activity levels from U.S. EPA (2009c). as cited
in U.S. EPA (2011). An occupational inhalation rate of 10.8 m3 for an 8-hour
workday, similar to the default value from U.S. EPA (1994). can be derived as an
average of activity-specific inhalation rates for males, in age groups from 21-60
years, for combined light and moderate activity from Table 6-17 of U.S. EPA (2011).
The average inhalation rate of 1.3 m3/hour (0.022 m3/min) can be multiplied by 8
hours to obtain an inhalation rate of 10.8 m3/8-hour workday.
VEh = human ambient default minute volume (20 m3 breathed during the entire day) (U.S.
EPA. 1994). This value is consistent with the average of the daily average inhalation
rates for males, in age groups from 21-60 years, of 20.2 m3/day, from U.S. EPA
(2009c). as summarized in Table 6-14 of U.S. EPA f 20 111.
2.1.3. Derivation of the Reference Concentration
Consistent with EPA's A Review of the Reference Dose and Reference Concentration Processes
(U.S. EPA. 2002: Section 4.4.5). also described in the Preamble, five possible areas of uncertainty
and variability were considered when deriving the RfC. A composite UF of 10 was applied to the
selected duration-adjusted POD of 4.9 mg/m3 to derive the RfC of 0.5 mg/m3. An explanation of the
five possible areas of uncertainty and variability follows:
• An intraspecies uncertainty factor, UFh, of 10 was applied to account for potentially
susceptible individuals in the absence of data evaluating variability of response to inhaled
ammonia in the human population;
• An interspecies uncertainty factor, UFa, of 1 was applied to account for uncertainty in
extrapolating from laboratory animals to humans because the POD was based on human
data from an occupational study;
• A subchronic to chronic uncertainty factor, UFs, of 1 was applied because the occupational
exposure period in the principal study fHolness etal.. 1989). defined as the mean number of
years at the present job for exposed workers, of approximately 12 years was considered to
be of chronic duration;
• An uncertainty factor for extrapolation from a LOAEL to a NOAEL, UFl, of 1 was applied
because a NOAEL was used as the POD; and
• A database uncertainty factor, UFd, of 1 was applied to account for deficiencies in the
database. As discussed in Section 1.2, available epidemiological studies include studies of
workers exposed in industrial settings, in agriculture, or through use of cleaning products.
This document is a draft for review purposes only and does not constitute Agency policy.
2-7 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
There are also controlled human exposure studies involving short-duration exposure to
ammonia vapors, and many case reports of acute exposures to high concentrations.
Available animal studies include subchronic studies that investigated respiratory and
systemic effects in rats, guinea pigs, and pigs. There are also several immunotoxicity
studies, and one limited reproductive toxicity study in young female pigs. The database
lacks developmental and multigenerational reproductive toxicity studies. The EPA's review
of RfD and RfC processes fU.S. EPA. 20021 states,
"If data from the available toxicology studies raise suspicions of
developmental toxicity and signal the need for developmental data on
specific organ systems (e.g., detailed nervous system, immune system,
carcinogenesis, or endocrine system), then the database factor should take
into account whether or not these data are available and used in the
assessment and their potential to affect the POD ..."
Although the database lacks developmental and multigenerational reproductive toxicity
studies, there are no data or suspicions of developmental toxicity at levels below the POD.
The available studies identify the respiratory system as the principal target of toxicity for
inhaled ammonia and do not suggest a likelihood of developmental or reproductive effects
at lower levels (see Sections 1.2.3 and 1.3.1).
The RfC for ammonia was calculated as follows:
RfC = NOAELadj - UF
= 4.9 mg/m3 4- 10
= 0.49 mg/m3 or 0.5 mg/m3 (rounded to one significant figure)
2.1.4. Uncertainties in the Derivation of the Reference Concentration
As presented earlier in this section and in the Preamble, EPA standard practices and RfC
guidance (U.S. EPA. 2002.1995.1994) were followed in applying an UF approach to a POD (from a
NOAEL) to derive the RfC. Specific uncertainties were accounted for by the application of UFs (i.e.,
in the case of the ammonia RfC, a factor to address the absence of data to evaluate the variability in
response to inhaled ammonia in the human population). The following discussion identifies
additional uncertainties associated with the quantification of the RfC for ammonia.
Use of a NOAEL as a POD
Data sets that support benchmark dose modeling are generally preferred for reference
value derivation because the shape of the dose-response curve can be taken into account in
establishing the POD. For the ammonia RfC, no decreases in lung function or increases in the
prevalence of respiratory symptoms were observed in the worker population studied by Holness et
al. Q9891. i.e., the principal study used to derive the RfC, and as such, the data from this study did
not support dose-response modeling. Rather, a NOAEL from the Holness etal. (1989) study was
used to estimate the POD. The availability of dose-response data from a study of ammonia,
especially in humans, would increase the confidence in the estimation of the POD.
Comparison of Exhaled Ammonia to the RfC
This document is a draft for review purposes only and does not constitute Agency policy.
2-8 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Ammonia is generated endogenously in multiple organs, including the liver, kidneys,
intestines, brain, and skeletal muscle, as a product of amino acid catabolism. Ammonia plays
central roles in nitrogen balance and acid-base homeostasis fWeiner etal.. 2014: Weiner and
Verlander. 20131. Given its important metabolic role, free ammonia is homeostatically regulated to
remain at low concentrations in blood fSouba. 19871. Elimination of ammonia occurs primarily in
urine and exhaled breath. (See Appendix C, Section C.1.3 for additional information on production
and regulation of endogenous ammonia.)
Further consideration was given to the presence of ammonia in exhaled air because the
range of ammonia concentrations in exhaled breath overlaps the ammonia RfC. Specifically,
ammonia has been measured in exhaled breath at concentrations ranging from 0.009-2 mg/m3 (see
Appendix C, Table C-l), a range that exceeds the RfC of 0.5 mg/m3. This section reviews
information related to the exhalation of ammonia that provides context for this comparison.
In general, the higher and more variable ammonia concentrations are reported in human
breath exhaled from the mouth or oral cavity. Investigators reported concentrations ranging from
0.03 to 2 mg/m3, with the majority of concentrations >0.2 mg/m3 f Schmidt etal.. 2013: Smith etal..
2008: Spanel etal.. 2007a. b; Turner etal.. 2006: Diskin etal.. 2003: Smith etal.. 1999: Norwood et
al.. 1992: Larson etal.. 19771. Ammonia concentrations measured in breath derived from oral
breathing largely reflect the production of ammonia via bacterial degradation of food protein in the
oral cavity or gastrointestinal tract fTurner etal.. 2006: Smith etal.. 1999: Vollmuth and
Schlesinger. 19841. Ammonia concentrations from exhaled breath can be influenced by factors such
as diet, oral hygiene, and age (Solga etal.. 2013: Spanel etal.. 2007a. b; Turner etal.. 2006: Diskin et
al.. 2003: Norwood et al.. 19921. Schmidt etal. (20131 reported that ammonia concentrations in
breath from the mouth strongly depended on saliva pH.
Concentrations of ammonia in breath exhaled from the nose and trachea of humans
(0.0092-0.1 mg/m3) are lower than those in air exhaled from the mouth fSchmidt et al..
2013: Smith etal.. 2008: Larson etal.. 19771. Whereas the upper end of the range of ammonia
concentrations in mouth breath exceeds the RfC of 0.5 mg/m3, concentrations from the nose and
trachea are generally lower than the ammonia RfC by a factor of five or more. Ammonia
concentrations in breath exhaled from the nose appear to better represent levels at the alveolar
interface of the lung and are thought to be more relevant to understanding systemic levels of
ammonia than breath exhaled from the mouth fSchmidt etal.. 2013: Smith etal.. 20081.
Nevertheless, the relationship between nose ammonia concentrations and systemic levels is
complicated by the possibility that nose ammonia concentrations are still influenced by the oral
cavity (e.g., in individuals with the soft palate incompletely closed), and tracheobronchial fluids
that, like saliva, can influence the airway concentration of ammonia. Further, measurements of
exhaled ammonia reported in the literature were generally not conducted in ammonia-free
environments, and thus the ammonia in inhaled air may account for some of the ammonia
measured in exhaled air (e.g., see Spanel etal. T201311.
Thus, ammonia concentrations in exhaled breath, and particularly those exhaled through
the mouth, are not correlated with blood ammonia; factors identified as influencing exhaled
This document is a draft for review purposes only and does not constitute Agency policy.
2-9 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
ammonia concentrations include bacterial populations in the oral cavity, salivary pH, diet, oral
hygiene, and age (see Appendix C, Section C.1.4). Concentration in breath cannot be used to predict
blood ammonia concentration or previous exposure to environmental (ambient) concentrations of
ammonia.
Regardless, the level of ammonia in breath, even at concentrations that exceed the RfC, does
not necessarily raise questions about the appropriateness of the RfC. The exhalation of ammonia is
a clearance mechanism for a product of metabolism that is otherwise toxic in the body at
sufficiently high concentrations. Ammonia concentrations in exhaled breath may be higher than
inhaled concentrations, particularly when compared to exhaled air from the mouth or oral cavity.
However, the fact that humans may exhale ammonia at concentrations higher than 0.5 mg/m3 (i.e.,
the RfC) is not considered an uncertainty in the RfC.
Consideration of Tolerance and the Healthy Worker Effect on Selection of the POD
As discussed in Section 1.2.1, two controlled-exposure studies provide some evidence of
habituation to eye, nose, and throat irritation in volunteers after repeated ammonia exposure.
Following exposure to ammonia at concentrations ranging from 7 to 35 mg/m3 for 4 hours/day on
five consecutive days, Ihrig etal. (2006) reported higher mean intensities for irritative, olfactory,
and respiratory symptoms in male volunteers unfamiliar with ammonia when compared to male
chemical company workers exposed to ammonia vapor for several years in a urea department;
differences were statistically significant only for olfactory symptoms. In a more limited study with
only four male volunteers each exposed to 18, 35, or 71 mg/m3 ammonia (exposure to each
concentration was for one week, 2-6 hours/day, 5 days/week; individuals were exposed to each
concentration twice), fewer occurrences of irritation were reported during week 2 than during
week 1 at the same exposure concentration Ferguson etal. f!977I However, in the same Ferguson
etal. (1977) study, the occurrences of irritation in two individuals exposed to 50 ppm for 6
hours/day, 5 days/week for 6 weeks was variable from week to week and did not show any clear
trend. The study by Ihrig etal. (2006). and to a lesser extent the study by Ferguson etal. (1977).
provide some evidence of decreased irritation following repeated exposure; the results of Ihrig et
al. (2006) may also be influenced by attrition out of the workforce of those most affected by the
irritation symptoms. These studies raise the possibility that repeated exposure could lead to the
development of tolerance to ammonia (i.e., to decreased sensory responsiveness). It is possible,
therefore, that industrially-exposed populations considered in deriving the RfC for ammonia
(i.e., Holness etal. Q989I Rahman etal. f2007I Ballal etal. f 19981. and Ali etal. (2001)) may have
developed some degree of tolerance to ammonia, and may underpredict responses to ammonia that
would be observed in the general population. The magnitude of tolerance, if any, cannot be
estimated from the available studies.
In addition, as discussed in the Literature Search Strategy | Study Selection and Evaluation
section, the workers in the cross-sectional occupational studies used to derive the RfC were healthy
enough to remain in the plant for a considerable time; mean employment duration ranged from 52
months to 18 years. In general, studies in these populations may result in a "healthy worker
This document is a draft for review purposes only and does not constitute Agency policy.
2-10 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
survivor" bias and in an underestimate of the risk of health effects of ammonia exposure, as a
healthy worker population may not exhibit health effects (such as decreased lung function or
increased prevalence of respiratory symptoms) to the same degree that would be seen in the
general population under the same conditions.
Therefore, there is potential for tolerance development in populations exposed
occupationally to ammonia and "healthy worker" bias, both of which may result in underestimation
of the general population response. However, the evidence is limited and not conclusive, and thus
does not warrant increasing the intraspecies uncertainty factor.
2.1.5. Confidence Statement
A confidence level of high, medium, or low is assigned to the study used to derive the RfC,
the overall database, and the RfC itself, as described in Section 4.3.9.2 of EPA's Methods for
Derivation of Inhalation Reference Concentrations and Application of Inhalation Dosimetry (U.S. EPA.
19941. Confidence in the principal study fHolness etal.. 19891 is medium. The design, conduct, and
reporting of this occupational exposure study were adequate, but the study was limited by a small
sample size and by the fact that workplace ammonia concentrations to which the study population
was exposed were below those associated with ammonia-related effects (i.e., only a NOAEL was
identified). However, the results from the principal study are supported by the results from other
cross-sectional studies of workers in industrial settings, studies of ammonia exposure in a cleaning
setting, studies in agricultural settings, multiple studies of acute ammonia exposure in volunteers,
and the available inhalation data from animals.
Confidence in the database is medium. The inhalation ammonia database includes one
limited study of reproductive and developmental toxicity in pigs that did not examine a complete
set of reproductive or developmental endpoints. Normally, confidence in a database lacking these
types of studies is considered to be lower due to the uncertainty surrounding the use of any one or
several studies to adequately address all potential endpoints following chemical exposure at
various critical lifestages. Unless a comprehensive array of endpoints is addressed by the database,
there is uncertainty as to whether the critical effect chosen for RfC derivation is the most sensitive
or appropriate. However, the likelihood of reproductive, developmental, and other systemic effects
at the RfC is considered small because it is well documented that ammonia is endogenously
produced in humans and animals, and any changes in blood ammonia levels at the POD would be
small relative to normal blood ammonia levels. Further, EPA is not aware of any mechanisms by
which effects at the point of contact (i.e., respiratory system) could directly or indirectly impact
tissues or organs distal to the point of contact. Thus, confidence in the database, in the absence of
these types of studies, is medium.
Reflecting medium confidence in the principal study and medium confidence in the
database, the overall confidence in the RfC is medium.
This document is a draft for review purposes only and does not constitute Agency policy.
2-11 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
2.1.6. Previous IRIS Assessment
The previous IRIS assessment for ammonia (posted to the database in 1991) presented an
RfC of 0.1 mg/m3 based on co-principal studies—the occupational exposure study of workers in a
soda ash plant by Holness etal. (1989) and the subchronic study by Broderson etal. (1976) that
examined the effects of ammonia exposure in F344 rats inoculated on day 7 of the study with the
bacterium M. pulmonis. The NOAEL of 6.4 mg/m3 (estimated as the mean concentration of the
entire exposed group) from the Holness etal. (1989) study (duration adjusted: NOAELadj =
2.3 mg/m3) was used as the POD.12
The previous RfC was derived by dividing the exposure-adjusted POD of 2.3 mg/m3 (from a
NOAEL of 6.4 mg/m3) by a composite UF of 30: 10 to account for the protection of sensitive
individuals and 3 for database deficiencies to account for the lack of chronic data, the proximity of
the LOAEL from the subchronic inhalation study in the rat fBroderson et al.. 19761 to the NOAEL,
and the lack of reproductive and developmental toxicity studies. A UFd of 3 (rather than 10) was
applied because studies in rats (Schaerdel etal.. 1983b) showed no increase in blood ammonia
levels at an inhalation exposure up to 32 ppm (22.6 mg/m3) and only minimal increases at 300-
1,000 ppm (212-707 mg/m3), suggesting that no significant distribution is likely to occur at the
human equivalent concentration.
12In this document, the lower confidence bound of the estimated mean exposure concentration in the high-
exposure group from the Holness etal. (1989) study (13.6 mg/m3, adjusted for continuous exposure to 4.9
mg/m3) was identified as the POD because workers in this high-exposure group, as well as those in the two
lower-exposure groups, 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.
2-12 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
REFERENCES
{, , 783307@@hidden;, , 783306@@hidden;, , 8079@@hidden;, , 998996@@hidden;, ,
626138@@hidden;, , 989022@@hidden;, , 993989@@hidden;, , 992265@@hidden;, ,
752972@@hidden;, , 86237@@hidden;, , 194568@@hidden;, , 194567@@hidden;, ,
786546@@hidden;,, 1578548@@hidden}
Multiple references published in the same year by the same author(s) have been assigned a
letter (e.g., 1986a, 1986b) based on author(s), year, and title. Those same letters have been
retained for the appendices.
Abdel 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.
AIChE (American Institute of Chemical Engineers). (1999). Ammonia H3N. In TE Daubert; RP
Danner; HM Sibul; CC Stebbins (Eds.), Physical and Thermodynamic Properties of Pure
Chemicals: Evaluated Process Design Data. Philadelphia, PA: Taylor & Francis.
Ali. 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. http: / /dx.doi.or g/10.1179 /oeh. 2 0 01.7.1.19
AMoN (Ammonia Monitoring Network). (2012). Ammonia Monitoring Network. Champaign, IL:
National Atmospheric Deposition Program (NADP). http: //nadp.sws.uiuc.edu/amon/
Amshel. CE: Fealk. MH: Phillips. BT: Caruso. DM. (2000). Anhydrous ammonia burns case report and
review of the literature [Review], Burns 26: 493-497.
Anderson. DP: Beard. CW: Hanson. RP. (1964a). The adverse effects of ammonia on chickens
including resistance to infection with Newcastle disease virus. Avian Dis 8: 369-379.
Anderson. DP: Beard. CW: Hanson. RP. (1964b). The adverse effects of ammonia on chickens
including resistance to infection with Newcastle disease virus. Avian Dis 8: 369-379.
http://dx.doi.org/10.2307/1587967
Andreasen. M: Baekbo. P: Nielsen. TP. (2000a). Lack of effect of aerial ammonia on atrophic rhinitis
and pneumonia induced by Mycoplasma hyopneumoniae and toxigenic Pasteurella
multocida. J Vet Med B 47: 161-171.
Andreasen. M: Baekbo. P: Nielsen. TP. (2000b). Lack of effect of aerial ammonia on atrophic rhinitis
and pneumonia induced by Mycoplasma hyopneumoniae and toxigenic Pasteurella
multocida. 47: 161-171. http://dx.doi.org/10.1046/i.l439-0450.2000.00335.x
Arif. AA: Delclos. GL. (2012). Association between cleaning-related chemicals and work-related
asthma and asthma symptoms among healthcare professionals. Occup Environ Med 69: 35-
40. http://dx.doi.org/10.1136/oem.2011.Q64865
ATS (American Thoracic Society). (2000). What constitutes an adverse health effect of air pollution?
Am J Respir Crit Care Med 161: 665-673. http://dx.doi.org/10.1164/airccm. 161.2,ats4-00
ATSDR (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=1 l&tid=2
Auron. A: Brophv. PP. (2012). Hyperammonemia in review: Pathophysiology, diagnosis, and
treatment [Review], Pediatr Nephrol 27: 207-222. http://dx.doi.org/10.1007/s00467-011-
1838-5
Bacom. A: Yanoskv. M. (2010). E-mails dated June 22, 2010, from Michael Yanosky, Drager Safety
Inc., Technical Support Detection Products to Amber Bacom, SRC, Inc., contractor to NCEA,
ORD, U.S. EPA. Available online
This document is a draft for review purposes only and does not constitute Agency policy.
R-l DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
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.
Betterton. EA. (1992). Henry's law constants of soluble and moderately soluble organic gases:
Effects in aqueous phase chemistry. In JO Nriagu (Ed.), Gaseous pollutants: Characterization
and cycling (pp. 1-50). New York, NY: Wiley.
Bhat. MR: Ramaswamv. C. (1993). Effect of ammonia, urea and diammonium phosphate (DAP) on
lung functions in fertilizer plant workers. Indian J Physiol Pharmacol 37: 221-224.
Broderson. IR: Lindsev. IR: Crawford. IE. (1976). The role of environmental ammonia in respiratory
mycoplasmosis of rats. Am J Pathol 85: 115-130.
Biirki. R: Mohebbi. N: Bettoni. C: Wang. X: Serra. AL: Wagner. CA. (2015). Impaired expression of
key molecules of ammoniagenesis underlies renal acidosis in a rat model of chronic kidney
disease. Nephrol Dial Transplant 30: 770-781. http://dx.doi.org/10.1093/ndt/gfu384
Callicutt. CH: Cox. RH: Hsu. F: Kinser. RD: Laffoon. SW: Lee. PN: Podraza. KF: Sanders. EB: Seeman.
JX (2006). The role of ammonia in the transfer of nicotine from tobacco to mainstream
smoke. Regul Toxicol Pharmacol 46: 1-17. http://dx.doi.Org/10.1016/i.yrtph.2006.05.008
Casas. L: Zock. IP: Torrent. M: Garci'a-Esteban. R: Gracia-Lavedan. E: Hvvarinen. A: Sunver. I. (2013).
Use of household cleaning products, exhaled nitric oxide and lung function in children
[Letter], Eur Respir J 42: 1415-1418. http://dx.doi.org/l 0.1183/09031936.00066313
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. http:/ /dx.doi.org/lQ.5271/siweh.1429
Cole. TI: Cotes. IE: lohnson. GR: Martin Hde. V: Reed. IW: Saunders. MI. (1977). Ventilation, cardiac
frequency and pattern of breathing during exercise in men exposed to O-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.
http://dx.doi.Org/10.1016/0041-008Xf70190069-4
Cordoba. I: Gottstein. I: Blei. AT. (1998). Chronic hyponatremia exacerbates ammonia-induced brain
edema in rats after portacaval anastomosis. J Hepatol 29: 589-594.
http:/ /dx.doi.org/10.1016/S0168-8278f98180154-2
Cormier. Y: Israel-Assavag. E: Racine. G: Duchaine. C. (2000). Farming practices and the respiratory
health risks of swine confinement buildings. Eur Respir J 15: 560-565.
Crook. B: Robertson. IF: Glass. SA: Botherovd. EM: Lacev. I: Topping. MP. (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: Kellev. 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. http://dx.doi.org/10.2134/iasl975.413735x
Pavies. S: Spanel. P: Smith. P. (1997). Quantitative analysis of ammonia on the breath of patients in
end-stage renal failure. Kidney Int 52: 223-228. http://dx.doi.org/10.1038/ki.1997.324
Pelclos. GL: Gimeno. P: Arif. AA: Benavides. FG: Zock. IP. (2009). Occupational exposures and
asthma in health-care workers: Comparison of self-reports with a workplace-specific job
exposure matrix. Am J Epidemiol 169: 581-587. http://dx.doi.org/10.1093/aie/kwn387
PeSanto. IT: Nagomi. W: Liechtv. EA: Lemons. IA. (1993). Blood ammonia concentration in cord
blood during pregnancy. Early Hum Pev 33: 1-8. http://dx.doi.org/10.1016/0378-
3782f93190168-T
Piekman. MA: Scheidt. AB: Sutton. AL: Green. ML: Clapper. IA: Kelly. PT: 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.
This document is a draft for review purposes only and does not constitute Agency policy.
R-2 PRAFT—PO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
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.
http://dx.doi.Org/10.1088/0967-3334/24/l/308
Doig. PA: Willoughbv. 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: Iones. P:
Armstrong. D: White. RP: Demmers. TG: Teer. N: Wathes. CM. (2005). Clinical and
pathological responses of weaned pigs to atmospheric ammonia and dust. VetRec 157: 71-
80.
Don ham. KI: 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.
Don ham. KT: Reynolds. ST: Whitten. P: Merchant. 1A: Burmeister. L: Popendorf. WT. (1995).
Respiratory dysfunction in swine production facility workers: Dose-response relationships
of environmental exposures and pulmonary function. Am J Ind Med 27: 405-418.
http://dx.doi.org/10.1002/aiim.47002703Q9
Donnav. C: Denis. MA: Magis. R: Fevotte. I: Massin. N: Dumas. 0: Pin. I: Choudat. D: Kauffmann. F: Le
Moual. N. (2011). Under-estimation of self-reported occupational exposure by
questionnaire in hospital workers. Occup Environ Med 68: 611-617.
http://dx.doi.org/10.ri36/oem.2010.061671
Douglas. RB: Coe. IE. (1987). The relative sensitivity of the human eye and lung to irritant gases.
Ann Occup Hyg 31: 265-267. http://dx.doi.Org/10.1093/annhvg/31.2.265
Dumas. 0: Donnav. C: Heederik. PI: Herv. M: Choudat. D: Kauffmann. F: Le Moual. N. (2012).
Occupational exposure to cleaning products and asthma in hospital workers. Occup Environ
Med 69: 883-889. http://dx.doi.org/10.1136/oemed-2012-100826
Eggeman. T. (2001). Ammonia. In Kirk-Othmer Encyclopedia of Chemical Technology. New York,
NY: John Wiley & Sons, Inc.
Ferguson. WS: Koch. WC: Webster. LB: Gould. IR. (1977). Human physiological response and
adaption to ammonia. J Occup 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.
http:/ /dx.doi.org/10.1080/00016489.1992.11665429
Gilbert. GI. (1988). Acute ammonia intoxication 37 years after ureterosigmoidostomy. South Med J
81: 1443-1445.
Giroux. M: Ferrieres. I. (1998). Serum nitrates and creatinine in workers exposed to atmospheric
nitrogen oxides and ammonia. Sci Total Environ 217: 265-269.
http://dx.doi.Org/10.1016/S0048-9697f98100190-9
Gustin. P: Urbain. B: Prouvost. IF: Ansav. 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
Hamilton. TP: Roe. IM: Hayes. CM: Iones. P: Pearson. GR: Webster. AT. (1999). Contributory and
exacerbating roles of gaseous ammonia and organic dust in the etiology of atrophic rhinitis.
Clin Piagn Lab Immunol 6: 199-203.
Hamilton. TP: Roe. IM: Hayes. CM: Webster. AT. (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.
Heederik. P: van Zwieten. R: Brouwer. R. (1990). Across-shift lung function changes among pig
farmers. Am JInd Med 17: 57-58. http://dx.doi.org/10.1002/aiim.4700170110
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
This document is a draft for review purposes only and does not constitute Agency policy.
R-3 PRAFT—PO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
HSDB (Hazardous Substances Data Bank). (2012). Ammonia. Washington, DC: National Library of
Medicine.
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
larudi. NI: Golden. B. (1973). Ammonia eye injuries. J Iowa Med Soc 63: 260-263.
leffers. LI: Dubow. RA: Zieve. L: Reddv. 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.
http://dx.doi.Org/10.1002/hep.1840080116
lohnson. PR: Bedick. CR: Clark. NN: McKain. PL. Ir. (2009). Pesign and testing of an independently
controlled urea SCR retrofit system for the reduction of NOx emissions from marine diesels.
Environ Sci Techno 1 43: 3959-3963. http://dx.doi.org/10.1021/es900269p
lones. EA. (2002). Ammonia, the GABA neurotransmitter system, and hepatic encephalopathy
[Review], Metab Brain Pis 17: 275-281. http://dx.doi.Org/10.1023/A:1021949616422
lozwik. M: lozwik. M: Pietrzvcki. B: Choinowski. M: Teng. C: lozwik. M: Battaglia. FC. (2005).
Maternal and fetal blood ammonia concentrations in normal term human pregnancies. Biol
Neonate 87: 38-43. http://dx.doi.org/l 0.1159/000081702
Kennedy. SM: Le Moual. N: Choudat. P: Kauffmann. F. (2000). Pevelopment of an asthma specific job
exposure matrix and its application in the epidemiological study of genetics and
environment in asthma (EGEA). Occup Environ Med 57: 635-641.
http://dx.doi.org/10.ri36/oem. 57.9.635
Lange. NA: Pean. 1A. (1985). Lange's handbook of chemistry (13th ed.). New York, NY: McGraw-Hill.
Larson. TV: Covert. PS: Frank. R: Chatison. Rl. (1977). Ammonia in the human airways:
Neutralization of inspired acid sulfate aerosols. Science 197: 161-163.
http://dx.doi.org/10.ri26/science.877545
Le Moual. N: Kauffmann. F: Eisen. EA: Kennedy. SM. (2008). The healthy worker effect in asthma:
Work may cause asthma, but asthma may also influence work [Review], Am J Respir Crit
Care Med 177: 4-10. http://dx.doi.org/10.ri64/rccm.2007Q3-415PP
Leaderer. BP: Naeher. L: lankun. T: Balenger. K: Hoi ford. TR: Toth. C: Sullivan. 1: Wolfson. 1M:
Koutrakis. P. (1999). Indoor, outdoor, and regional summer and winter concentrations of
PM10, PM25, S042-, H+, NH4+, N03-, NH3, and nitrous acid in homes with and without
kerosene space heaters. Environ Health Perspect 107: 223-231.
Lemiere. C: Begin. P: Camus. M: Forget. A: Boulet. LP: Gerin. M. (2012). Occupational risk factors
associated with work-exacerbated asthma in Quebec. Occup Environ Med 69: 901-907.
http://dx.doi.org/10.ri36/oemed-2012-100663
Lide. PR. (2008). CRC handbook of chemistry and physics: A ready-reference book of chemical and
physical data. In PR Lide (Ed.), (89th ed.). Boca Raton, FL: CRC Press.
Loftus. C: Yost. M: Sampson. P: Torres. E: Arias. G: Breckwich Vasquez. V: Hartin. K: Armstrong. I:
Tchong-French. M: Vedal. S: Bhatti. P: Karr. C. (2015). Ambient ammonia exposures in an
agricultural community and pediatric asthma morbidity. Epidemiology 26: 794-801.
http://dx.doi.Org/10.1097/EPE.0000000000000368
MacEwen. IP: 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 Biol Med 187: 278-281. http://dx.doi.org/10.3181/00379727-187-42664
Manninen. ATA: Savolainen. H. (1989). Effect of short-term ammonia inhalation on selected amino
acids in rat brain. Pharmacol Toxicol 64: 244-246. http://dx.doi.org/10.1111/1.1600-
0773.1989.tb00639.x
This document is a draft for review purposes only and does not constitute Agency policy.
R-4 PRAFT—PO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Medina-Ramon. M: Zock. IP: Kogevinas. M: Sunver. 1: Basagana. X: Schwartz. 1: Burge. PS: Moore. V:
Anto. 1M. (2006). Short-term respiratory effects of cleaning exposures in female domestic
cleaners. Eur Respir J 27: 1196-1203. http://dx.doi.org/10.1183/09031936.06.000854Q5
Medina-Ramon. M: Zock. IP: Kogevinas. M: Sunver. 1: Torralba. Y: Borrell. A: Burgos. F: Anto. 1M.
(2005). Asthma, chronic bronchitis, and exposure to irritant agents in occupational
domestic cleaning: A nested case-control study. Occup Environ Med 62: 598-606.
http://dx.doi.org/10.1136/oem.2004.017640
Melbostad. E: Eduard. W. (2001). Organic dust-related respiratory and eye irritation in Norwegian
farmers. Am J Ind Med 39: 209-217. http://dx.doi.org/10.1002/1097-
0274f200102139:2<209::A1D-A11M1008>3.0.CQ:2-5
Millea. TP: Kucan. 10: Smoot. EC. (1989). Anhydrous ammonia injuries. Journal of Burn Care and
Rehabilitation 10: 448-453.
Minana. MP: 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. 1: 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 /aiim.20Q77
Nelson. PL: Cox. MM. (2008). Amino acid oxidation and the production of urea. In DL Nelson; AL
Lehninger; MM Cox (Eds.), Lehninger Principles of Biochemistry (5th ed., pp. 656-689). New
York, NY: W.H. Freeman & Co.
Nemer. M: Sikkeland. LI: Kasem. M: Kristensen. P: Niiem. K: Biertness. E: Skare. 0: Bakke. B:
Kongerud. I: Skogstad. M. (2015). Airway inflammation and ammonia exposure among
female Palestinian hairdressers: a cross-sectional study. Occup Environ Med 72: 428-434.
http://dx.doi.org/10.ri36/oemed-2014-102437
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.org/10.rm/i.l 439-
0450.1987.tb003 86.x
NLM (National Institutes of Health, National Library of Medicine). (2012). ChemlDplus: Ammonia.
Bethesda, MD: National Library of Medicine.
http://chem.sis.nlm.nih.gov/chemidplus/name/startswith/amonia
Norwood. DM: Wainman. T: Liov. PI: Waldman. IM. (1992). Breath ammonia depletion and its
relevance to acidic aerosol exposure studies. Arch Environ Health 47: 309-313.
http://dx.doi.Org/10.1080/00039896.1992.9938367
O'Neil. MI: Heckelman. PE: Koch. CB: Roman. KI. (2006). Ammonia. In The Merck Index: An
Encyclopedia of Chemicals, Drugs, and Biologicals (14th ed.). Whitehouse Station, NJ: Merck
& Co., Inc.
Paulot. F: lacob. PI. (2014). Hidden cost of u.s. Agricultural exports: particulate matter from
ammonia emissions. Environ Sci Technol 48: 903-908.
http://dx.doi.org/10.1021/es4034793
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: Boleii. 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.
http://dx.doi.org/10.ri36/oem.52.10.654
This document is a draft for review purposes only and does not constitute Agency policy.
R-5 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
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.
http://dx.doi.Org/10.1179/oeh.2007.13.2.153
Read. AT. (1982). Ionization constants of aqueous ammonia from 25 to 250C and to 2000 bar.
Journal of Solution Chemistry 11: 649-664. http://dx.doi.org/10.1007/bf0065Q397
Reynolds. SI: Donham. KT: Whitten. P: Merchant. IA: Burmeister. LF: Popendorf. WI. (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/l 0.1002/fSICIVl 097-02 74fl 99601129:1 <33 ::AID-AIIM5>3.0.CQ:2-#
Richard. D: Boulev. G: Boudene. C. (1978). [Effects of ammonia gas continuously inhaled by rats and
mice]. Bull Europ Physiol Resp 14: 573-582.
Romero-Gomez. M: Ramos-Guerrero. R: Grande. L: de Teran. LC: Corpas. R: Camacho. I: Bautista. ID.
(2004). Intestinal glutaminase activity is increased in liver cirrhosis and correlates with
minimal hepatic encephalopathy. J Hepatol 41: 49-54.
http://dx.doi.Org/10.1016/i.ihep.2004.03.021
Rosswall. T. (1981). The biogeochemical nitrogen cycle. In GE Likens (Ed.), Some Perspectives of
the Major Biogeochemical Cycles (pp. 25-49). Chichester, England: John Wiley & Sons.
http://dge.stanford.edu/SCOPE/SCOPE 17/SCOPE 17.html
Sadasivudu. B: Radha Krishna Murthv. 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: Murthv. CR. (1979). Chronic metabolic effects of ammonia in mouse brain.
Arch Physiol Biochem 87: 871-885.
Schaerdel. AD: White. WI: Lang. CM: Dvorchik. BH: Bohner. K. (1983a). Localized and systemic
effects of environmental ammonia in rats. Lab Anim Sci 33: 40-45.
Schaerdel. AD: White. WI: Lang. CM: Dvorchik. BH: Bohner. K. (1983b). Localized and systemic
effects of environmental ammonia in rats. J Am Assoc Lab Anim Sci 33: 40-45.
Schmidt. FM: Vaittinen. 0: Metsala. M: Lehto. M: Forsblom. C: Groop. PH: Halonen. L. (2013).
Ammonia in breath and emitted from skin. J Breath Res 7: 017109.
http://dx.doi. org/10.1088/1752-7155/7/1/017109
Schoeb. TR: Davidson. MK: Lindsev. 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.
Seeman. II: Carchman. RA. (2008). The possible role of ammonia toxicity on the exposure,
deposition, retention, and the bioavailability of nicotine during smoking [Review], Food
Chem Toxicol 46: 1863-1881. http://dx.doi.Org/10.1016/i.fct2008.02.021
Sigurdarson. ST: O'Shaughnessv. PT: Watt. IA: Kline. IN. (2004). Experimental human exposure to
inhaled grain dust and ammonia: Towards a model of concentrated animal feeding
operations. Am JInd Med 46: 345-348. http://dx.doi.org/10.1002 /aiim.20055
Silverman. L: Whittenberger. IL: Muller. I. (1949). Physiological response of man to ammonia in low
concentrations. J Ind Hyg Toxicol 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/bilQ31
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: Pvsanenko. 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/10.10Q2/rcm.3434
Socolow. RH. (1999). Nitrogen management and the future of food: lessons from the management of
energy and carbon. PNAS 96: 6001-6008.
This document is a draft for review purposes only and does not constitute Agency policy.
R-6 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
Solga. SF: Mudalel. M: Spacek. LA: Lewicki. R: Tittel. F: Loccioni. C: Russo. A: Risbv. TH. (2013).
Factors influencing breath ammonia determination. J Breath Res 7: 037101.
http://dx.doi.Org/10.1088/1752-7155/7/3/037101
Souba. WW. (1987). Interorgan ammonia metabolism in health and disease: A surgeon's view
[Review], JPEN J Parenter Enteral Nutr 11: 569-579.
http://dx.doi. org/10.1177/0148607187011006569
Spanel. P: Drvahina. 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/l/l/011001
Spanel. P: Drvahina. 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/
10.1088/1752-7155/1/2/026001
Spanel. P: Drvahina. K: Smith. D. (2013). A quantitative study of the influence of inhaled compounds
on their concentrations in exhaled breath. J Breath Res 7:017106.
http://dx.doi.Org/10.1088/1752-7155/7/l/017106
Stearne. 1AC: Higgins. 1PT: Reaves. BC. (2014). A Cochrane risk of bias assessment tool: For non-
randomized studies of interventions (ACROBAT-NRSI) (Version 1.0.0). Retrieved from
http://www.riskofbias.info
Stombaugh. DP: Teague. HS: Roller. WL. (1969). Effects of atmospheric ammonia on the pig. J Anim
Sci 28: 844-847.
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.
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.
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
U.S. EPA (U.S. Environmental Protection Agency). (1994). Methods for derivation of inhalation
reference concentrations (RfCs) and application of inhalation dosimetry [EPA Report],
(EPA/600/8-90/066F). Washington, DC: 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/630/R-94/007). Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment Forum.
http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=3 0004WBL.txt
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: U.S.
Environmental Protection Agency, Risk Assessment Forum.
http://www.epa.gov/osa/review-reference-dose-and-reference-concentration-processes
U.S. EPA (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk assessment
[EPA Report], (EPA/630/P-03/001F). Washington, DC: U.S. Environmental Protection
Agency, Risk Assessment Forum, http: //www2.epa.gov/osa/guidelines-carcinogen-risk-
assessment
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). (2009a). Integrated Risk Information System
(IRIS); announcement of availability of literature searches for IRIS assessments [FRL89741;
Docket ID No. EPAHQORD20070664], Fed Reg 74: 56611-56612.
This document is a draft for review purposes only and does not constitute Agency policy.
R-7 DRAFT—DO NOT CITE OR QUOTE
-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
U.S. EPA (U.S. Environmental Protection Agency). (2009b). Integrated science assessment for
particulate matter [EPA Report], (EPA/600/R-08/139F). Research Triangle Park, NC: U.S.
Environmental Protection Agency, National Center for Environmental Assessment
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm7deich216546
U.S. EPA (U.S. Environmental Protection Agency). (2009c). Metabolically derived human ventilation
rates: A revised approach based upon oxygen consumption rates (final report) [EPA
Report], (EPA/600/R-06/129F). Washington, DC.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=202543
U.S. EPA (U.S. Environmental Protection Agency). (2011). Exposure factors handbook: 2011 edition
(final) [EPA Report], (EPA/600/R-090/052F). Washington, DC: U.S. Environmental
Protection Agency, Office of Research and Development, National Center for Environmental
Assessment. http://cfpub.epa.gov/ncea/cfm/recordisplav.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
Verschueren. K. (2001). Ammonia. In Handbook of environmental data on organic chemicals. New
York, NY: John Wiley & Sons.
Vizcava. D: Mirabelli. MC: Anto. IM: Orriols. R: Burgos. F: Ariona. L: Zock. IP. (2011). A workforce-
based study of occupational exposures and asthma symptoms in cleaning workers. Occup
Environ Med 68: 914-919. http://dx.doi.org/10.ri36/oem.2010.063271
Vollmuth. TA: Schlesinger. RB. (1984). Measurement of respiratory tract ammonia in the rabbit and
implications to sulfuric acid inhalation studies. Fundam Appl Toxicol 4: 455-464.
http://dx.doi. org/10.1016/0272-0590f84190203-3
Weatherbv. IH. (1952). Chronic toxicity of ammonia fumes by inhalation. Proc Soc Exp Biol Med 81:
300-301. http://dx.doi.org/10.3181/00379727-81-19857
Weiner. ID: Mitch. WE: Sands. IM. (2014). Urea and Ammonia Metabolism and the Control of Renal
Nitrogen Excretion. Clin J Am Soc Nephrol 10: 1444-1458.
http://dx.doi.org/10.2215/CIN.10311013
Weiner. ID: Verlander. IW. (2013). Renal ammonia metabolism and transport [Review], Compr
Physiol 3: 201-220. http://dx.doi.org/10.1002/cphv.cl20010
Zeida. 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.
Zock. IP: Plana. E: larvis. D: Anto. IM: Kromhout. H: Kennedy. SM: Kiinzli. N: Villani. S: Olivieri. M:
Toren. K: Radon. K: Sunver. I: Dahlman-Hoglund. A: Norback. D: Kogevinas. M. (2007). The
use of household cleaning sprays and adult asthma: An international longitudinal study. Am
J Respir Crit Care Med 176: 735-741. http://dx.doi.org/10.ri64/rccm.200612-17930C
This document is a draft for review purposes only and does not constitute Agency policy.
R-8 DRAFT—DO NOT CITE OR QUOTE
------- |