&EPA
                                                             EPA/635/R-13/139b
                                                      Revised External Review Draft
                                                               www.epa.gov/iris
                     Toxicological Review of Ammonia

                             (CASRN 7664-41-7)

                In Support of Summary Information on the
                Integrated Risk Information System (IRIS)

                        Supplemental Information
                                 August 2013
                                   NOTICE

This document is a Revised External Review draft. This information is distributed solely for the
purpose of pre-dissemination peer review under applicable information quality guidelines.  It has
not been formally disseminated by EPA. It does not represent and should not be construed to
represent any Agency determination or policy. It is being circulated for review of its technical
accuracy and science policy implications.
                    National Center for Environmental Assessment
                        Office of Research and Development
                        U.S. Environmental Protection Agency
                                Washington, DC

-------
                                                 Supplemental Information—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                                           ii             DRAFT—DO NOT CITE OR QUOTE

-------
                                             Supplemental Information—Ammonia
CONTENTS
APPENDIX A. ASSESSMENTS BY OTHER NATIONAL AND INTERNATIONAL HEALTH AGENCIES	A-l
APPENDIX B. CHEMICAL AND PHYSICAL PROPERTY INFORMATION FOR AMMONIA	B-l
APPENDIX C. TOXICITY INFORMATION FOR SELECTED AMMONIUM SALTS	C-l
APPENDIX D. ADDITIONAL DETAILS OF LITERATURE SEARCH STRATEGY | STUDY SELECTION
           AND EVALUATION	D-l
APPENDIX E. INFORMATION IN SUPPORT OF HAZARD IDENTIFICATION AND DOSE-RESPONSE
           ANALYSIS	E-l
       E.I. TOXICOKINETICS	E-l
       E.2. HUMAN STUDIES	E-15
       E.3. ANIMAL STUDIES	E-34
       E.4. OTHER PERTINENT TOXICITY INFORMATION	E-53
APPENDIX F. DOCUMENTATION OF IMPLEMENTATION OF THE 2011 NATIONAL RESEARCH
           COUNCIL RECOMMENDATIONS	F-l
APPENDIX G. SUMMARY OF EXTERNAL PEER REVIEW AND PUBLIC COMMENTS AND EPA's
           DISPOSITION	G-l
REFERENCES FOR APPENDICES	R-l
        This document is a draft for review purposes only and does not constitute Agency policy.
                                        iii            DRAFT—DO NOT CITE OR QUOTE

-------
                                                   Supplemental Information—Ammonia
TABLES
Table A-l.   Assessments by other national and international health agency assessments for
            ammonia	A-l
Table B-l.   Chemical and physical properties of ammonia	B-l
Table C-l.   Summary of repeat dose studies of selected ammonium salts following oral
            exposure	C-2
Table D-l.   Literature search strings*	D-l
Table D-2.   Evaluation of epidemiology studies summarized in Table 1-1 (industrial D-
            settings/respiratory measures)	D-5
Table D-3.   Evaluation of epidemiology studies summarized in Table 1-2 (use in
            cleaning/disinfection settings)	D-9
Table D-4.   Evaluation of epidemiology study summarized in Table 1-6 (industrial
            setting/serum chemistry measures)	D-13
Table E-l.   Ammonia levels in exhaled breath of volunteers	E-8
Table E-2.   Symptoms and lung function results of workers exposed to different levels of
            TWA ammonia concentrations	E-17
Table E-3.   The prevalence of respiratory symptoms and disease in urea fertilizer workers
            exposed to ammonia	E-18
Table E-4.   Logistic regression analysis of the relationship between ammonia concentration
            and respiratory symptoms or disease in exposed urea fertilizer workers	E-19
Table E-5.   Prevalence of respiratory symptoms and cross-shift changes in lung function
            among workers exposed to ammonia in a urea fertilizer factory	E-21
Table E-6.   Comparison of lung function parameters in ammonia plant workers with controls .. E-22
Table E-7.   Evidence pertaining to respiratory effects in humans in relation to ammonia
            exposure in livestock farmers	E-23
Table E-8.   Studies of respiratory effects in livestock farmers without direct analysis of
            ammonia exposure	E-26
Table E-9.   Controlled human exposure studies of ammonia inhalation	E-27
Table E-10.  Effect of ammonia in drinking water on the thickness of the gastric antral and
            body mucosa of the rat stomach	E-35
Table E-ll.  Effect of ammonia in drinking water on gastric antral and body mucosa in the
            stomach of Sprague-Dawley rats administered 0.01% ammonia in drinking water... E-36
Table E-12.  Summary of histological changes observed in pigs exposed to ammonia for 6
            weeks	E-42
Table E-13.  Acute and short-term inhalation toxicity studies of ammonia in animals	E-47
Table E-14.  Summary of in vitro studies of ammonia genotoxicity	E-53
Table E-15.  Summary of in vivo studies of ammonia genotoxicity	E-55
Table F-l.   The EPA's implementation of the National Research Council's recommendations
            in the ammonia assessment	F-3
Table F-2.   National Research Council recommendations that the EPA is generally
            implementing in the long term	F-8


FIGURES

Figure E-l.   Glutamine cycle	G-4
Figure E-2.   The urea cycle showing the compartmentalization of its steps within liver cells	G-5

         This document is a draft for review purposes only and does not constitute Agency policy.
                                             iv              DRAFT—DO NOT CITE OR QUOTE

-------
ABBREVIATIONS
                                                      Supplemental Information—Ammonia
ACGIH  American Conference of Governmental          LOAEL
       Industrial Hygienists                         MAO
AEGL  Acute Exposure Guideline Level                MMEF
ALP    alkaline phosphatase                         MNNG
ALT    alanine aminotransferase                     MRL
ANOVA analysis of variance                          NHs
AST    aspartate aminotransferase                   NH4+
ATSDR Agency for Toxic Substances and Disease        NIOSH
       Registry
BMI    body mass index                            NOAEL
BrDU  bromodeoxyuridine                          NRC
BUN   blood urea nitrogen                          OR
CAC    cumulative ammonia concentration            OSHA
CI     confidence interval
COPD  chronic obstructive pulmonary disease          PAS
DAP    diammonium phosphateEU                   PEF
       endotoxin unit                              PEFR
FDA    Food and Drug Administration                PEL
FEF    forced expiratory flow                        RDso
FEVi   forced expiratory volume in 1 second           REL
FVC    forced vital capacity                          SD
GABA  gamma-aminobutyric acid                    SIFT-MS
HERO  Health and Environmental Research
       Online                                     TLV
IgE    immunoglobulin E                           TWA
IgG    immunoglobulin G                           UF
IRIS    Integrated Risk Information System            U.S. EPA
LCso    50% lethal concentration
lowest-observed-adverse-effect level
monoamine oxidase
mean midexpiratory flow
N-methyl-N'-nitro-N-nitrosoguanidine
minimal risk level
ammonia
ammonium ion
National Institute for Occupational
Safety and Health
no-observed-adverse-effect level
National Research Council
odds ratio
Occupational Safety and Health
Administration
periodic acid-Schiff
peak expiratory flow
peak expiratory flow rate
Permissible Exposure Limit
50% response dose
Recommended Exposure Limit
standard deviation
selected ion flow tube mass
spectrometry
threshold limit value
time-weighted average
uncertainty factor
U.S. Environmental Protection Agency
          This document is a draft for review purposes only and does not constitute Agency policy.
                                             v              DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
4
5
6
7
    APPENDIX A.   ASSESSMENTS  BY OTHER NATIONAL
    AND INTERNATIONAL HEALTH AGENCIES
       Toxicity values and other health-related regulatory limits for ammonia that have been
developed by other national and international health agencies are summarized in Table A-l.
            Table A-l. Assessments by other national and international health agency
            assessments for ammonia
             Organization
                                                    Toxicity value
     Agency for Toxic Substances and
     Disease Registry (ATSDR, 2004)
                            Chronic inhalation MRL = 0.1 ppm (0.07 mg/m )
                            Basis: Lack of significant alterations in lung function in chronically exposed
                            workers (Holnesset al., 1989) and a composite UF of 30 (10 for human
                            variability and a modifying factor of 3 for the lack of reproductive and
                            developmental studies).
      National Advisory Committee for
      Acute Exposure Guideline Levels
      for Hazardous Substances (NRC,
      2008)
                            AEGL-1 (nondisabling) = 30 ppm (21 mg/m ) for exposures ranging from
                            10 mins to 8 hrs to protect against mild irritation
                            Basis: mild irritation in human subjects (MacEwen et al., 1970)
                            AEGL-2 (disabling) = 220 ppm (154 mg/m3) for a 10-min exposure to 110 ppm
                            (77 mg/m3) for an 8-hr exposure
                            Basis: irritation (eyes and throat; urge to cough) in human subjects (Verberk,
                            1977)
                            AEGL-3 (lethal) = 2,700 ppm (1,888 mg/m3) for a 10-min exposure to 390 ppm
                            (273 mg/m3) for an  8-hr exposure
                            Basis: lethality in the mouse (Kapeghian et al., 1982; MacEwen and Vernot,
                            1972)
     American Conference of
     Governmental Industrial
     Hygienists (ACGIH, 2001)

     TLV established in 1973
                            TLV = 25 ppm (17 mg/m )a TWA for an 8-hr workday and a 40-hr work week
                            Basis: To protect against irritation to eyes and the respiratory tract. ACGIH
                            stated that irritation is the prime hazard to workers, but that systemic effects
                            cannot be  ruled out based on the findings of reduced feed consumption and
                            body weight loss in pigs exposed to 103 and 145 ppm ammonia. References
                            cited in  support of the TLV included papers from the primary literature for the
                            years up to 1973; no specific reference served as the basis for the TLV.
      National Institute for
      Occupational Safety and Health
      (NIOSH, 2010)

      REL established in 1992
                            REL = 25 ppm (18 mg/m )  TWA for up to a 10-hr workday and a 40-hr work
                            week
                            Basis: To project against respiratory and eye irritation. References cited in
                            support of the REL included review documents for the years up to 1992; no
                            specific reference served as the basis for the REL.
              This document is a draft for review purposes only and does not constitute Agency policy.
                                                A-l            DRAFT—DO NOT CITE OR QUOTE

-------
                                                     Supplemental Information—Ammonia

       Table A-l. Assessments by other national and international health agency
       assessments for ammonia
Organization
Occupational Safety and Health
Administration (OS HA, 2006)
PEL established in early 1970s
Food and Drug Admistration
(FDA, 2011a, b)

Toxicity value
PEL for general industry = 50 ppm (35 mg/m3) TWA for an 8-hr workday
Basis: The 1968 ACGIH TLV was promulgated as the OSHA PEL soon after
adoption of the Occupational Safety and Health Act in 1970. The ACGIH TLV
from 1968 was intended to protect against irritation of ammonia in humans;
no specific reference served as the basis for the 1968 TLV.
Ammonium hydroxide: direct food substance affirmed as generally recognized
as safe (21 CFR 184.1139); substance generally recognized as safe when used
in accordance with good manufacturing or feeding practices (21 CFR
582.1139).
aACGIH andr NIOSH used slightly different ppm to mg/m3 conversion factors.

AEGL = Acute Exposure Guideline Level; MRL = minimal risk level; PEL = Permissible Exposure Limit;
REL = Recommended Exposure Limit; TWA = time weighted average; UF = uncertainty factor
         This document is a draft for review purposes only and does not constitute Agency policy.
                                            A-2            DRAFT—DO NOT CITE OR QUOTE

-------
                                                 Supplemental Information—Ammonia
 1

 2

 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
APPENDIX B.  CHEMICAL AND  PHYSICAL  PROPERTY
INFORMATION FOR AMMONIA
      Many physical and chemical properties of ammonia (NHs) are related to the pH of ammonia
in solution (ammonium hydroxide). Ammonium hydroxide is a weak base that is partially ionized
in water with a dissociation constant of 1.77 x 1Q-5 at 25°C that increases slightly with increasing
temperature [Read. 19821. AtapH of 8.25, 90% of ammonia will be protonated. At a pH of 7.25,
99% of ammonia will be protonated. Thus, a decrease in pH would result in an increase in the
ammonium ion (NH4+) concentration and an increase in solubility of ammonia in water. At
physiological pH (7.4), the equilibrium between NHs and NH4+ favors the formation of NH4+.
Chemical and physical properties of ammonia are listed in Table B-l.

       Table B-l. Chemical and physical properties of ammonia
Chemical name
Synonym(s)
Structure
Chemical formula
CASRN
Molecular weight
Form
Melting point
Boiling point
Odor threshold
Density
Vapor density
pKa (ammonium ion)
Solubility:
Water
Organic solvents
Vapor pressure
Henry's law constant
Ammonia3
AM-Fol; anhydrous ammonia; ammonia gas;
Nitro-sil; R 717; Spirit of hartshorn
H
H'%
NH3
7664-41-73
17.031
Colorless gas; corrosive
-77.73°C
-33.33°C
53 ppm (37 mg/m3)
2.6 ppm (2 mg/m3)
0.7714 g/L at 25°C
0.5967 (air =1)
9.25
4.82 x 105 mg/L at 24°C
Soluble in ethanol, chloroform, and ether
7.51xl03mmHgat25°C
1.61 x 10"5 atm-m3/mol at 25°C

NLM (2012)

NLM (2012)



NLM (2012)
NLM (2012)
Lide (2008, pp. 4.46-4
48, 8.40)
O'Neiletal. (2006)
Lide (2008, pp. 4.46-4
Lide (2008, pp. 4.46-4
O'Neiletal. (2006)
Smeetsetal. (2007)
48, 8.40)
48, 8.40)

O'Neiletal. (2006)
O'Neiletal. (2006)
Lide (2008, pp. 4.46-4
Dean (1985, pp. 10-3,
48, 8.40)
10-23);
Lide (2008, pp. 4.46-4.48, 8.40);
O'Neiletal. (2006)

(AlChE, 1999)
Betterton (1992)
             This document is a draft for review purposes only and does not constitute Agency policy.
                                         B-l           DRAFT—DO NOT CITE OR QUOTE

-------
                                                     Supplemental Information—Ammonia

       Table B-l. Chemical and physical properties of ammonia
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.
         This document is a draft for review purposes only and does not constitute Agency policy.
                                           B-2            DRAFT—DO NOT CITE OR QUOTE

-------
                                                       Supplemental Information—Ammonia

 1

 2   APPENDIX C.  TOXICITY  INFORMATION  FOR

 3   SELECTED AMMONIUM SALTS	
 4
 5
 6         Because of uncertainty concerning the possible influence of anions on the toxicity of
 7   ammonium, information on ammonium salts was not used to characterize the effects or to derive
 8   reference values for ammonia or ammonium hydroxide.  A summary of the subchronic and chronic
 9   toxicity of selected ammonium salts is presented here as supplemental information.
10         The toxicology literature for ammonium salts includes 13-, 78-, and 130-week ammonium
11   chloride dietary studies in male and female Wistar rats [Lina and Kuijpers, 2004], a 47-week
12   ammonium chloride drinking water study in Sprague-Dawley rats [Barzel and Towsey, 1969], and
13   52- and 104-week ammonium sulfate dietary studies in male and female F344 rats [Otaetal.,
14   2006]. No inhalation toxicity studies of ammonium salts were found.
15         Ammonium chloride in the diet or drinking water of rats consistently altered the acid-base
16   balance in the body [Lina and Kuijpers. 2004: Barzel and Towsey. 1969] causing a dose-related
17   hyperchloremic metabolic acidosis in rats as evidenced by increased plasma chloride levels and
18   decreases in blood pH, base excess, and bicarbonate concentration. Ammonium chloride
19   administered in the diet for 130 weeks was also associated with zona glomerulosa hypertrophy of
20   the adrenal gland [Lina and Kuijpers, 2004]. Kidney weights were not significantly affected by
21   exposure to ammonium chloride for 78 or 130 weeks [Lina and Kuijpers. 2004]: liver weights were
22   not reported in this study.
23         Dietary administration of ammonium sulfate to rats has not been associated with metabolic
24   acidosis, but this endpointwas not specifically evaluated in the 52- or 104-week studies by Ota et
25   al. [2006].  Unlike  ammonium chloride, no histopathologic changes in the adrenal gland were
26   observed following ammonium sulfate exposure [Otaetal., 2006]. The dose-related effects in male
27   and female rats associated with 52-week exposure to ammonium sulfate were increased liver and
28   kidney weights [Ota etal.. 2006].  See Table C-l  for study details.
              This document is a draft for review purposes only and does not constitute Agency policy.
                                              C-l            DRAFT—DO NOT CITE OR QUOTE

-------
                                                            Supplemental Information—Ammonia
        Table C-l. Summary of repeat dose studies of selected ammonium salts
        following oral exposure
     Study design and reference
                           Results
Ammonium chloride
Wistar rat (10/sex/group)

0,1,590, or 3,050 mg/kg-d (males); 0,
1,800, or 3,700 mg/kg-d (females)
administered in diet for 13 wks

(Una and Kuijpers, 2004; Barzel and
Jowsey, 1969)
Body weight: 4> (6-17% in males; 11-19% in females)
Liver weight: not reported
Kidney weight (relative): 1" (both dose levels, both sexes, 7-28%)
Adrenal weight (relative):  1" (high-dose males, 18%)
Metabolic acidosis*: observed in males and females; severity
increased with dose
ALP activity: 1" at high dose, no change at lower doses
Wistar rat (15/sex/group)

0, 481, or 1,020 mg/kg-d (males); 0,
610, or 1,370 mg/kg-d (females)
administered in diet for 78 wks
(Una and Kuijpers, 2004; Barzel and
Jowsey, 1969)
Body weight:  no significant change
Liver weight: not reported
Kidney weight (relative): no significant change
Adrenal weight (relative): no significant change
Metabolic acidosis3: observed in males and females; severity
increased with dose
ALP activity: not measured
Wistar rat (50/sex/group)

0, 455, or 1,000 mg/kg-d (males); 0,
551, or 1,200 mg/kg-d (females)
administered in diet for 130 wks

(Una and Kuijpers, 2004; Barzel and
Jowsey, 1969)
Body weight: no significant change
Liver weight: not reported
Kidney weight (relative): no significant change
Adrenal weight (relative): no significant change
Metabolic acidosis*: observed in males and females; severity
increased with dose
ALP activity: not measured
Hypertrophy of the adrenal glomerulosa: /T" incidence (both doses
in males, high dose only in females)
Chronic progressive nephrosis: 4, incidence in males at the highest
dose
Sprague-Dawley rat (11 males/group)
0 or 1,800 mg/kg-d administered in
drinking water for 47 wks
(Una and Kuijpers, 2004; Barzel and
Jowsey, 1969)
Body weight: 4> (13-20% with regular and low-calcium diets,
respectively)
Kidney weight (relative): not measured
Kidney weight (absolute): no change
Adrenal weight (relative): not measured
Femur weight (relative): 4,
Femur calcium: -^
Metabolic acidosis: was inferred from measurements of reduced
blood pH and plasma carbon dioxide
ALP activity: not measured
Ammonium sulfate
F344 rat (10/sex/group)

0, 42, 256, or 1,527 mg/kg-d (males);
0, 48, 284, or 1,490 mg/kg-d (females)
administered in diet for 52 wks

(Ota etal., 2006)
Body weight: no significant change in males and females
Liver weight (relative): T* in males (7%); T* in females (7%)
Kidney weight (relative): T in males  (10%); T* in females (10%)
Adrenal weight (relative): no significant change in males and
females
Metabolic acidosis3: not measured
ALP activity: not significantly changed (except in females at
intermediate dose, 284 mg/kg, % change compared to control ALP
activity was-19%)
            This document is a draft for review purposes only and does not constitute Agency policy.
                                                 C-2               DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
        Table C-l. Summary of repeat dose studies of selected ammonium salts
        following oral exposure
     Study design and reference
                          Results
F344 rat (50/sex/group)
0, 564, or 1,288 mg/kg-d (males); 0,
650, or 1,371 mg/kg-d (females)
administered in diet for 104 wks
(Ota etal., 2006)
Body weight: not measured
Liver weight (relative): not measured
Kidney weight (relative): not measured
Adrenal weight (relative): not measured
Metabolic addosis*: not measured
ALP activity: not measured
Hypertrophy of the adrenal glomerulosa: no change in incidence
Chronic nephropathy: /T" incidence in male rats over control (1/48,
5/49, and 3/48 in the control, mid, and high dose); increase was
statistically significant only at the mid-dose
aMetabolic acidosis was assessed as decreased base excess in blood, decreased urinary pH, and increased
urinary net acid excretion.

ALP = alkaline phosphatase
           This document is a draft for review purposes only and does not constitute Agency policy.
                                               C-3               DRAFT—DO NOT CITE OR QUOTE

-------
                                   Supplemental Information—Ammonia
5
6
   APPENDIX D.  ADDITIONAL DETAILS OF
   LITERATURE SEARCH STRATEGY |  STUDY
   SELECTION AND EVALUATION
       Table D-l. Literature search strings*
Database
Set#
Terms
Hits
Initial strategy
PubMed
Date range:
1950'sto
present
Search date:
3/26/2012

ToxLine
Date range:
1907-present
Search date:
3/26/2012
1A1
1A2
IB
(("Ammonia"[MeSH Terms] OR "ammonium hydroxide" [Supplementary
Concept]) AND (("ammonia/adverse effects"[MeSH Terms] OR
"ammonia/antagonists and inhibitors"[MeSH Terms] OR
"ammonia/blood"[MeSH Terms] OR "ammonia/cerebrospinal fluid"[MeSH
Terms] OR "ammonia/pharmacokinetics"[MeSH Terms] OR
"ammonia/poisoning"[MeSH Terms] OR "ammonia/toxicity"[MeSH Terms]
OR "ammonia/urine"[MeSH Terms]) OR ("hydroxides/adverse
effects" [MeSH Terms] OR "hydroxides/antagonists and inhibitors"[MeSH
Terms] OR "hydroxides/blood"[MeSH Terms] OR "hydroxides/cerebrospinal
fluid"[MeSH Terms] OR "hydroxides/pharmacokinetics"[MeSH Terms] OR
"hydroxides/poisoning"[MeSH Terms] OR "hydroxides/toxicity"[MeSH
Terms] OR "hydroxides/urine"[MeSH Terms]) OR
(("ammonia/metabolism"[MeSH Terms] OR
"hydroxides/metabolism" [MeSH Terms]) AND (animals[MeSH Terms] OR
humans[MeSH Terms])) OR (ci[Subheading] OR "environmental
exposure"[MeSH Terms] OR "endocrine system "[MeSH Terms] OR
"hormones, hormone substitutes, and hormone antagonists"[MeSH Terms]
OR risk[MeSH Terms] OR cancer[sb]) OR ((ammonia[majr] OR "ammonium
hydroxide"[Supplementary Concept]) AND (dose-response relationship,
drug[MeSH Terms] OR pharmacokinetics[MeSH Terms] OR
metabolism[MeSH Terms]) AND (humans[MeSH Terms] OR
mammals[MeSH Terms]))) OR ((Ammonia [Title] OR "Ammonium
hydroxide"[Title] OR "Spirit of hartshorn"[Title] OR Aquammonia[Title])
NOTmedline[sb])
Additional Search on Exhaled Breath
(inhal* OR (air OR breath OR exhal* OR respiration) OR (biological
markers[MeSH Terms] AND (air OR breath OR exhal* OR respiration)) OR
("air pollutants"[MeSH Terms] AND (breath OR exhal*)) OR breath OR
(analysis[Subheading] AND breath) OR (respiration[MeSH Terms] OR breath
tests[MeSH Terms] OR exhalation[MeSH Terms])) AND (7664-41-7[rn] OR
1336-21-6[rn])
limited to ammon* in title. This covered all synonyms listed to both
ammonia and ammonium hydroxide with the exception of "spirit of
hartshorn" which found no results when limited to the title.
Original: 13,012
Update: 410
Original: 1,600
Update: 50
Original: 2,417
Update: 100
         This document is a draft for review purposes only and does not constitute Agency policy.
                             D-l        DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
        Table D-l. Literature search strings*
Database
Set#
Terms
Hits
TSCATS1,
TSCATS2, TSCA
recent notices
Date range: no
limit
Search date:
3/26/2012
1C
7664-41-7
1336-21-6
Original:
50TSCATS1
7 TSCATS2
1 recent notices

Update: 0
Toxcenter
Date range:
1907-present
Search date:
3/27/2012
1D1
((7664-41-7 OR 1336-21-6) not (patent/dt OR tscats/fs)) and (chronic OR
immunotox? OR neurotox? OR toxicokin? OR biomarker? OR neurolog? OR
pharmacokin? OR subchronic OR pbpk OR epidemiology/st,ct, it) OR acute
OR subacute OR Id50# OR Ic50# OR (toxicity OR adverse OR
poisoning)/st,ct,it OR inhal? OR pulmon? OR nasal? OR lung? OR respir? OR
occupation? OR workplace? OR worker? OR oral OR orally OR ingest? OR
gavage? OR diet OR diets OR dietary OR drinking(w)water OR (maximum
and concentration? and (allowable OR permissible)) OR (abort? OR
abnormalit? OR embryo? OR cleft? OR fetus? OR foetus? OR fetal? OR
foetal? OR fertil? OR malform? OR ovum OR ova OR ovary OR placenta? OR
pregnan? OR prenatal OR perinatal? OR postnatal? OR reproduc? OR steril?
OR teratogen? OR sperm OR spermac? OR spermag? OR spermati? OR
spermas? OR spermatob? OR spermatoc? OR spermatog? OR spermatoi?
OR spermatoi? OR spermator? OR spermatox? OR spermatoz? OR
spermatu? OR spermi? OR spermo? OR neonat? OR newborn OR
development OR developmental? OR zygote? OR child OR children OR
adolescen? OR infant OR wean? OR offspring OR age(w)factor? OR dermal?
OR dermis OR skin OR epiderm? OR cutaneous?  OR carcinog? OR
cocarcinog? OR cancer? OR precancer? OR neoplas? OR tumor? OR
tumour? OR oncogen? OR lymphoma? OR carcinom? OR genetox? OR
genotox? OR mutagen? OR genetic(w)toxic? OR  nephrotox? OR hepatotox?
OR endocrin? OR estrogen? OR androgen? OR hormon?) AND (((biosis/fs
AND py>1999 AND (hominidae/ct,st,it OR human/ct,st,it OR humans/ct,st,it
OR mammals/ct,st,it OR mammal/ct,st,it OR mammalia/ct,st,it)) OR ipa/fs
OR (caplus/fs AND 4-?/cc) OR ammonia/ti OR "ammonium hydroxide"/ti OR
"spirit of hartshorn"/ti OR aquammonia/ti)
Dupicates were removed; Biosis subfile results were date limited to avoid
extensive overlap with Toxline
Original: 2,591

Update: No
access
               1D2
Additional Search on Exhaled Breath
 (7664-41-7 OR 1336-21-6) AND (breath OR exhale? OR "expired air")
                                                                     81
HERO
Date range: -
present
Search date:
3/27/2012
IE
ammonia OR ammonium hydroxide
Original: 5,295

Update: 115
(this represents
all of 2012 and
2013; not
limited to
March 2012 to
March 2013)
            This document is a draft for review purposes only and does not constitute Agency policy.
                                               D-2              DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
        Table D-l. Literature search strings*
Database
Set#
Terms
Hits
Combined
Reference Set
(duplicates eliminated through electronic screen)
                                                                      Original: 22,400

                                                                      Update:
                                                                      duplicates
                                                                      eliminated
                                                                      directly by
                                                                      HERO
Secondary refinement
Combined
reference set
with additional
terms applied
     (gastrointestinal OR gastro-intestinal OR digestive tract OR stomach* OR
     (gastric AND (mucosa* OR cancer* OR tumor* OR tumour* OR neoplas*))
     OR (ammoni*[title] AND intestin*[title or keyword]) OR genotox* OR
     (genetic* AND toxic*) OR ames assay* OR ames test* OR aneuploid* OR
     chromosom*[title] OR clastogen* OR cytogen* OR dominant lethal OR
     genetic*[title] OR genotox* OR hyperploid* OR micronucle* OR mitotic* OR
     mutagen*[title] OR mutat*[title] OR recessive lethal OR sister chromatid OR
     ((kidney* OR renal) AND (toxic* OR poisoning OR adverse OR congestion OR
     calcif*)) OR nephrotox* OR ((spleen* OR splenic) AND (toxic* OR poisoning
     OR adverse OR congestion OR enlarged)) OR absorption OR distribution OR
     metabolism[title or keywords] OR excret* OR PBPK OR toxicokinetic* OR
     pharmacokin* OR exhal* OR breath OR (expired AND air) OR (respiratory
     AND (irritation OR symptom* or disease* OR adverse OR chemically
     induced)) OR lung* OR (pulmonary AND (irritation* OR function*)) OR FVC
     OR Forced  vital capacity OR Forced expiratory volume OR FEV OR FEV1 OR
     inflammation OR congest* OR edema* OR hemorrhag* OR discharge* OR
     phlegm* OR cough* OR wheez* OR dyspnea OR bronchitis OR pneumonitis
     OR asthma* OR nose OR nasal OR throat OR trachea* OR bronchial OR
     airway* OR (chest AND tightness) OR epithelium* OR epithelia* OR
     immune OR immun*[title] OR antibod* OR antigen* OR autoimmun* OR
     cytokine* ORgranulocyte* OR interferon OR interleukin* OR leukocyte* OR
     lymph* OR lymphocyt* OR monocyt*  OR immunosupress* OR
     immunotox* OR (immun* AND toxic*) OR hypersensitivity OR ((dermal OR
     skin) AND lesion*) OR erythema* OR host resistance OR ((bacterial OR
     bacteria) AND coloniz*) ORT cell* ORT-Lymphocyte* OR thymocyte* OR
     ((liver* OR hepatic) AND (function* OR congest* OR toxic* OR poisoning OR
     adverse)) OR hepatotox* OR fatty liver OR clinical chemistry OR adrenal OR
     ((heart* OR cardiac) AND (toxic* OR adverse OR poisoning)) OR cardiotox*
     OR myocardium OR myocardial OR lacrimation OR ocular OR (eye* AND
     discharge*) OR opacity OR blood pH OR neurotransmitter* OR (amino acid*
     AND brain) OR reproduct*[title] OR reproductive OR developmental[title or
     keywords]  OR terato* OR (ammoni*[title] AND (abort* OR cleft* OR
     embryo* OR fertilit* OR fetal OR fetus* OR foetal OR foetus* OR gestation*
     OR infertilit* OR malform* OR neonat* OR newborn* OR ova OR ovaries OR
     ovary OR ovum OR perinatal OR placenta* OR postnatal OR pregnan* OR
     prenatal OR sperm OR sterilit* OR zygote*)) NOT (hyperammon* OR
     ammonemia OR ammonaemia OR hepatic coma OR liver failure OR (reye
     AND syndrome) OR ((hepatic OR liver OR portosystemic OR portal-systemic)
     AND (encephalopathy OR cirrhosis)) OR fish OR fishes OR carp OR catfish
     OR crayfish OR jellyfish OR daphnia OR shrimp OR frog OR frogs OR
     amphibians OR bivalve OR bivalves OR clam OR Crustacea OR crustaceans)
                                                                Original: 9,130

                                                                Update:
                                                                Further
                                                                narrowing of
                                                                database not
                                                                deemed
                                                                necessary;
                                                                small enough
                                                                numbers to do
                                                                a manual
                                                                screen
            This document is a draft for review purposes only and does not constitute Agency policy.
                                               D-3             DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
           Table D-l. Literature search strings*
     Database
Set# Terms
Hits
     *The literature search was updated through March 2013 using the same search strategies as previously used;
     PubMed, ToxLine TSCATS1, TSCATS2, and TSCA recent notices and HERO databases were searched; Toxcenter was
     not searched because it was not accessible. The number of hits is indicated for both the original search and the
     updated search (from March 2012 through March 2013).
 1
 2
 3    Additional Search Strategy Focused in Cleaning and Hospital Worker Literature
 4           The updated literature search (through March 2013) identified papers published in 2012
 5    that included information on ammonia exposure in health care workers. Because this represented
 6    an area of research that had not been previously identified, EPA conducted additional searches
 7    focusing on ammonia use in cleaning scenarios. The references in Dumas etal. [2012] and in a
 8    review paper by Zocketal. [2010] that was cited in Dumas etal. [2012] were reviewed looking for
 9    data on ammonia exposure; references in each newly identified publication were also reviewed. In
10    addition, a forward search was conducted using a methods paper describing the development of a
11    job exposure matrix focusing on asthma as a key reference [Kennedy et al., 2000], as this work has
12    been instrumental in developing this area of research from a focus on job titles to specific tasks and
13    then to specific products.  This updated and augmented search process led to the identification of
14    seven additional references [Arif andDelclos. 2012: Dumas etal.. 2012: Lemiere etal.. 2012:
15    Vizcayaetal..2011: Zocketal.. 2007: Medina-Ramon etal.. 2006: Medina-Ramon etal.. 2005] that
16    were included in the Toxicological Review.
17
               This document is a draft for review purposes only and does not constitute Agency policy.
                                                D-4             DRAFT—DO NOT CITE OR QUOTE

-------
                                                                                  Supplemental Information—Ammonia
Table D-2. Evaluation of epidemiology studies summarized in Table 1-1 (industrial settings/respiratory
measures)


Reference

Study setting/
participant selection


Exposure parameters

Outcome
measured

Consideration of
confounding


Statistical analysis
Comments regarding
potential major
limitations
Respiratory symptoms
Rahman et
al. (2007)


















Ballaletal.
(1998)













Bangladesh, urea fertilizerfactory;
cross sectional study
Exposed: n =88 (24 ammonia plant
workers and 64 urea plant workers)
Controls: n = 25
Exposed: production operators in
ammonia (low exposure; 24 out of 63
workers participated)3 and urea (high
exposure, 64 out of 77 workers
participated)11 plants, 5-9 out of 15-
19 per shift selected. Excluded if
planned to have less than a four-hour
work day. Mean age ~40 yrs, mean
duration ~18 yrs; never smoked
~52%.
Controls: from administration
building, 4-7 per day over 5 days
selected. Mean age ~43 yrs, mean
duration ~16yrs; never smoked
~72%.
Saudi Arabia; two urea fertilizer
factories; cross sectional study; all
males
Exposed: n = 161
Factory A: n = 84
Factory B: n = 77
Controls: n = 355
Exposed: 20% of workers selected
(systematic sample representing
different workplaces using payroll
lists); 100% participation rate. Mean
age 30 yrs, mean duration 51.8
months; never smoked ~59%.
Controls: administrative staff from
other companies in the area (same
Personal airborne levels of
ammonia exposure by two
direct-reading methods:
Dra'ger diffusion tube and
Dra'ger PAC III monitoring
instrument'; 1 worker per
day per measure.
Correlation between
methods; r = 0.80, but
higher absolute values (by
four- to fivefold) using
Dra'ger diffusion tubes0
Concentrations based on
PAC III monitoring:
Low-exposure group
(ammonia plant): 6.9
ppm (4.9 mg/m3)
High-exposure group
(urea plant): 26.1 ppm
(18.5 mg/m3)
Area monitors (3 sets in
each work section taken at
least 3 months apart,
mean 16 measures per
set); spectrophotometric
absorption measure.
Computed geometric
mean concentration per
section and cumulative
ammonia concentration (a
function of both exposure
intensity and duration of
service) assigned to each
worker.

Respiratory
symptoms (5 point
scale for severity
over last shift),
based on Optimal
Symptom Score
Questionnaire)













Prevalence of
respiratory
symptoms and
conditions based on
the British Medical
Research Council
questionnaire








Nitrogen dioxide
(measured by Drager
tubes) was below
detection limit in all areas
(urea plant, ammonia
plant and administration
area); other workplace
exposures not assessed.
Exposure analysis adjusted
for current smoking and
duration









Authors stated no other
pollutants in workplace.
Stratified or adjusted for
smoking











Fisher's exact test;
repeated excluding 33
current smokers or
workers with history of
previous respiratory
disease














Contingency tables
(stratified by smoking);
logistic regression of
exposure measures,
adjusted for duration,
smoking (yes, no)









Study population and
design: "healthy" workers;
long duration— potential
for lack of complete
ascertainment of effect

Differences in exposure
measurement methods
(Dra'ger diffusion tube and
Dra'ger PAC III monitoring
instrument) considered
limitation for quantitation
of exposure-response
relationship but not a
limitation for hazard
identification due to
uncertainty in the absolute
value, but not the relative
ranking, of exposure

Study population and
design: "healthy" workers;
long duration— potential
for lack of complete
ascertainment of effect










                   This document is a draft for review purposes only and does not constitute Agency policy.
                                                      D-5                         DRAFT—DO NOT CITE OR QUOTE

-------
                                                                                  Supplem en tal Inform ation —Amm on/a
Table D-2. Evaluation of epidemiology studies summarized in Table 1-1 (industrial settings/respiratory
measures)


Reference




Holness et al.
(1989)


















Study setting/
participant selection
sampling system as exposed);
participation rate 100%. Mean age 34
yrs, mean duration 73 months; never
smoked ~49%.
Canada, sodium carbonate (soda ash)
production plant; cross sectional
study
Exposed: n = 58
Controls: n = 31
Exposed: 52 of 64 available
production workers (82%) and 6
maintenance workers; all males.
Mean age 39 yrs, mean duration 14.4
yrs, nonsmokers ~29%.
Controls from stores and office
workers in the plant; excluded if
previous ammonia exposure.
Participation rate not reported. Mean
age 43 yrs, mean duration 12.2 yrs;
nonsmokers ~39%.
Indication of self-selection of exposed
out of workplace based on atopy
(lower prevalence of hay fever).


Exposure parameters




Airborne levels of
ammonia (mean = 6.5
mg/m3 for exposed; mean
= 0.2 mg/m3 for controls)
using NIOSH-
recommended protocol
for personal sampling and
analysis (measured over
one work-shift per person,
mean 8.4 hours)










Outcome
measured




Prevalence of self-
reported symptoms
and conditions
obtained through
questionnaire based
on American
Thoracic Society
questionnaire












Consideration of
confounding




Adjusted for smoking
(pack-yrs); other
workplace exposures not
assessed, but study
authors note high level of
control of exposures in the
plant














Statistical analysis




Comparison between
groups by logistic
regression. Also analyzed
by three categories of
exposure.














Comments regarding
potential major
limitations




Study population and
design: "healthy" workers;
long duration— potential
for lack of complete
ascertainment of effect

Relatively small sample
size— potential of not
being able to detect a
difference between
controls and exposed
when one might exist

Low exposure
concentrations— potential
that an effect level may
not have been reached


Lung function
Rahman et
al. (2007)










Bangladesh, urea fertilizer factory;
cross sectional study
Exposed: n = 88 (24 ammonia plant
workers and 64 urea plant workers);
production operators in ammonia
(low exposure; 24 out of 63 workers
participated)3 and urea (high
exposure, 64 out of 77 workers
participated)11 plants, 5-9 out of 15-
19 per shift selected. Excluded if
planned to have less than a four-hour
work day. Mean age ~40 yrs, mean
Personal airborne levels of
ammonia exposure by two
direct-reading methods:
Dra'ger diffusion tube and
Dra'ger PAC III monitoring
instrument'; 1 worker per
day per measure.
Correlation between
methods; r = 0.80, but
higher absolute values (by
four- to fivefold) using
Dra'ger diffusion tubes.0
Spirometry by
standard protocol,
beginning and end of
shift








Nitrogen dioxide
(measured by Dra'ger
tubes) was below
detection limit in all areas
(urea plant, ammonia
plant, and administration
area); other workplace
exposures not assessed.
Exposure analysis adjusted
for current smoking and
duration.

Paired t-tests compared
cross shift differences in
lung function within and
between plants; analyses
repeated excluding
workers with previous
respiratory diseases.
Multiple linear regression
analyzed exposure level
and change in lung
function for n = 23 with
both concurrent measure
Study population and
design: "healthy" workers;
long duration-potential
for lack of complete
ascertainment of effect

Differences in exposure
measurement methods
(Dra'ger diffusion tube and
Dra'ger PAC III monitoring
instrument) considered
limitation for quantitation
                   This document is a draft for review purposes only and does not constitute Agency policy.
                                                      D-6                         DRAFT—DO NOT CITE OR QUOTE

-------
                                                                                  Supplemental Information—Ammonia
Table D-2. Evaluation of epidemiology studies summarized in Table 1-1 (industrial settings/respiratory
measures)


Reference








All et al.
(2001)















Bhat and
Ramaswamy
(1993)










Study setting/
participant selection
duration ~18 yrs; never smoked
~52%.






Saudi Arabia; urea fertilizer factory;
cross sectional study (appears to be
same as Factory A in Ballal et al.
(1998)
Exposed: n = 73
Controls: n = 348
Exposed: 20% of workers selected
(systematic sample representing
different workplaces using payroll
lists); 95% participation rate. Mean
age 30 yrs, mean duration 51.8
months; nonsmokers~49%.
Controls: administrative staff from 4
industrial groups (same sampling
system as exposed); participation
rate 98%. Mean age 34 yrs;
nonsmokers ~42%.
Mangalore; fertilizer chemical plant;
cross sectional study
Exposed: n =91
Controls: n = 68
Exposed: 30 urea plant workers, 30
DAP plant workers, and 31 ammonia
plant workers; sex of workers not
reported; age, sex, height, weight,
and duration of exposure were
recorded but not reported; duration
of exposure dichotomized into two
groups (up to 10 yrs and more than


Exposure parameters
Concentrations based on
PAC III monitoring:
Low-exposure group
(ammonia plant): 6.9
ppm (4.9 mg/m3)
High-exposure group
(urea plant): 26.1 ppm
(18.5 mg/m3)
Ammonia concentration in
air determined by
sampling pump with a
flow rate of 1 L/min for 4
hours for each
measurement and
spectrophotometry (i.e.,
by absorption techniques
and comparison to a
standard). Computed
cumulative ammonia
concentration (a function
of both exposure level and
duration of service)
assigned to each worker,
dichotomized to high and
low at 50 mg/m3-yrs
No measurement of
exposure made











Outcome
measured








Spirometry by
standard protocol,
morning
measurement, 3 or
more replicates












Spirometry by
standard protocol, 3
replicates with
highest reading
retained for
calculation







Consideration of
confounding








Stratified by smoking
status















All smokers excluded from
study.
Other workplace
exposures not assessed.










Statistical analysis








T-tests and Chi-square
tests for comparisons
between groups and by
exposure level among
exposed












Paired t-test for
comparisons between
exposed and controls









Comments regarding
potential major
limitations
of exposure-response
relationship but not a
limitation for hazard
identification due to
uncertainty in the absolute
value, but not the relative
ranking, of exposure

Study population and
design: "healthy" workers;
long duration — potential
for lack of complete
ascertainment of effect












Study population and
design: "healthy" workers;
long duration — potential
for lack of complete
ascertainment of effect







                   This document is a draft for review purposes only and does not constitute Agency policy.
                                                      D-7                         DRAFT—DO NOT CITE OR QUOTE

-------
                                                                                                                 Supplemental Information—Ammonia
         Table D-2.  Evaluation of epidemiology studies summarized in Table 1-1 (industrial settings/respiratory
         measures)


Reference







Holness et al.
(1989)



















Study setting/
participant selection
10 yrs); smokers excluded.
Controls: people having comparable
body surface area chosen from the
same socio-economic status and sex;
smokers excluded; no other
information provided on participant
selection.
Canada, sodium carbonate (soda ash)
production plant; cross sectional
study
Exposed: n=58
Controls: n=31
Exposed: 52 of 64 available
production workers (82%) and 6
maintenance workers; all males,
mean age 39 yrs, mean duration 14.4
yrs; nonsmokers~29%.
Controls from stores and office
workers in the plant; excluded if
previous ammonia exposure.
Participation rate not reported. Mean
age 43 yrs, mean duration 12.2 yrs;
nonsmokers~39%.
Indication of self-selection of
exposed out of workplace based on
atopy (lower prevalence of hay
fever).


Exposure parameters







Airborne levels of
ammonia (mean = 6.5 mg/
m3 for exposed; mean =
0.2 mg/m3 for controls)
using NIOSH-
recommended protocol
for personal sampling and
analysis (measured over
one work-shift per person,
mean 8.4 hours)











Outcome
measured







Spirometry by
standard protocol,
beginning and end of
shift, 3-6 replicates,
each worker
measured on two
test days














Consideration of
confounding







Adjusted for smoking
(pack-yrs); other
workplace exposures not
assessed


















Statistical analysis







Baseline lung function
compared between
groups using linear
regression, adjusting for
age, height, and pack-yrs
(linear regression).
Unpaired t-tests
compared change in lung
function overworkshift
between groups. Percent
predicted lung function
at baseline and change in
lung function also
analyzed by three
categories of exposure.





Comments regarding
potential major
limitations







Study population and
design: "healthy" workers;
long duration— potential
for lack of complete
ascertainment of effect

Relatively small sample
size— potential of not
being able to detect a
difference between
controls and exposed
when one might exist

Low exposure
concentrations— potential
that an effect level may
not have been reached



aAmmonia plant workers checked temperature, pressure, and concentration of ammonia and checked the pumps, prepared solutions, and checked the revolutions per minute of various motors.
These are considered the low-exposure group.
bUrea plant workers purged solution and washed pipelines, operated various pumps, and washed and cleaned the cooling fluidized bed in the production area. These are considered the high-
exposure group.
cBased on communication with technical support at Drager Safety Inc. (Bacom and Yanosky, 2010), the U.S. Environmental Protection Agency (U.S. EPA) considered the PAC III instrument to be a
more sensitive monitoring technology than the Drager tubes. Therefore, more confidence is attributed to the PAC III air measurements of ammonia forthe Rahman et al. (2007) study.
                                 This document is a draft for review purposes only and does not constitute Agency policy.
                                                                             D-8                                DRAFT—DO NOT CITE OR QUOTE

-------
                                                                                 Supplemental Information—Ammonia
Table D-3. Evaluation of epidemiology studies summarized in Table 1-2 (use in cleaning/disinfection settings)


Reference
Dumas et al.
(2012)
























Arif and
Delclos (2012)










Study setting/
participant selection
France. Nested case-control study of
adult asthma cases recruited from
pulmonary clinics in 1991-1995;
follow-up in 2003-2007. Drawn from
the Epidemiological study on the
Genetics and Environment in Asthma
(EGEA) study (included first degree
relatives of cases and population
control group). Study base = 1,355:
included if had occupation data,
excluded if asthma at baseline or and
missing data on smoking. Selected if
ever worked in hospital (exposure
group) and referent group
Hospital workers: 179 (43 men, 136
women)
Referent group: 545 (212 men, 333
women)
Smoking history and age similarfor
women; smoking history similarfor
men, but mean age approximately 5
yrs higher in hospital workers)
Possible "healthy worker" bias, with
underestimation of associations from
movement out of jobs or avoidance of
specific jobs by affected individuals
United States (Texas). Survey of 3,650
licensed health care professionals
(physicians, nurses, respiratory
therapists, occupational therapists.
Response rate 66% (3,650 out of
5,600)







Exposure measure
Exposure to specific agents
based on three methods (ever
exposed, based on all jobs held
at least 3 months):
• Self-report: two job
exposure questionnaire
modules for health care
workers (including frequency of
use of specific products)
[possible underestimate of
exposure]
• Expert assessment- hospital
workers (probability,
frequency, intensity; 18
products)
• Asthma-specific job
exposure matrix (22 agents)
with expert review
Control group: "Never exposed
to cleaning/disinfecting
products" based on each of the
methods described above, plus
expert review of additional
(broader) information from
main occupation questionnaire

For longest job held: frequency
of use of specific products
(never/once a month,
at least once a week, more than
once a day, every
day) (for 2,049 of the 3,650,
current/most recent job was
longest held job)
For all jobs: ever been in contact
with list of 28 products at least
once a month for a period of 6


Outcome measured
Asthma attack, respiratory
symptoms or asthma
treatment in the last 12
months (based on
standardized
questionnaire)




















• Four outcomes, based
on structured
questionnaire
• Work Related Asthma
Symptoms (WRAS):
wheezing/whistling at
work or shortness of
breath at works that gets
better away from work or
worse at work
• Work Related Asthma

Consideration of
confounding
Adjusted for age and
smoking status.
Additional adjustment
for body mass index
tested.
Association with
ammonia strongerthan
that seen with bleach
(OR 1.87 and 0.93,
respectively, for
ammonia and bleach)















Adjusted for age, sex,
race/ethnicity, body
mass index, seniority,
atopy and smoking
status.







Statistical
analysis
Products analyzed
if 5 or more
exposed cases.
Analyses stratified
by sex (small n in
men so focused on
women). Familial
dependence in
data accounted for
by generalized
estimating
equations.














Multinomial
logistic regression
with four asthma
outcome
categories: WRAS,
WEA, OA and
none.
Oversampling
nurses and
physicians was
accounted for with
Comments
regarding potential
major limitations


























Limited exposure
assessment (i.e., "ever
exposed")








                  This document is a draft for review purposes only and does not constitute Agency policy.
                                                     D-9                          DRAFT—DO NOT CITE OR QUOTE

-------
                                                                                 Supplemental Information—Ammonia
Table D-3. Evaluation of epidemiology studies summarized in Table 1-2 (use in cleaning/disinfection settings)


Reference










Lemiere et al.
(2012)















Vizcava et al.
(2011)










Study setting/
participant selection










Quebec. Case-control study.
Workers with work-related asthma
(WRA) seen at two tertiary care
centers; WRA based on specific
inhalation challenges (SIC); reversible
airflow limitation or airway hyper-
responsiveness (provocative
concentration of methacholine
inducing a 20% fall in FEVi equal or
lower than 8 mg/ml.
Controls: Non-work related asthma
(NWRA) seen at same clinics but
symptoms did not worsen at work.

Total n = 153 (33 controls, 120 work
related asthma)

Barcelona, Spain
Survey of 1,018 cleaning services to
find companies willing to participate;
286 (28%) not eligible (no longer in
business); 37 agreed to participate (n
workers ranged from 6 to >1,000).
4,993 questionnaires distributed by
company representatives to
employees; 950 (19%) completed; 33
excluded because of missing data.
Total n = 917. Two companies


Exposure measure
months or longer (ammonia part
of general cleaning factor in
factor analysis)







Structured interview about
last/current job (including job
title, tasks, machines, materials),
work environment, protective
equipment. This information
used in conjunction with other
material (e.g., technical and
material safety data sheets,
occupational hygiene literature,
data bases and web sites) for
expert review and classification
of exposure to 41 specific
agents, blinded to case status.
Semiquantitative
estimate (low=l, medium=2,
high=3) for intensity, frequency,
and confidence.
Standardized questionnaire
about cleaning tasks and
products used in the last yr
Reference group = never
cleaners AND current cleaners
who had not used bleach,
degreasers, multi-purpose
cleaners, glass cleaners,
perfumed products, air
fresheners, mop products,
hydrochloric acid, ammonia,


Outcome measured
(WRA): same as above and
physician-diagnosed
asthma (n = 74)
• Work exacerbated
asthma (WEA): onset
before began work (n =
41)
• Occupational asthma
(OA): onset after began
workfn = 33)
• Diagnoses made based
on reference tests
• Occupational asthma
(OA) if specific inhalation
challenge test was
positive (n = 67);
• Work exacerbated
asthma (WEA) if specific
inhalation test was
negative but symptoms
worsened at work (n = 53)






• Current asthma based
on structured
questionnaire (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

Consideration of
confounding










Assessed confounding
effects of age, smoking,
occupational exposureto
heat, cold, humidity,
dryness and physical
strain; not included in
final models because
none acted as
confounders of
exposures under study







Adjusted for age,
country of birth (Spanish
vs non-Spanish), sex,
and smoking status








Statistical
analysis
post-stratification
weights








Logistic regression
















Asthma: logistic
regression
Asthma score:
Negative binomial
regression (to
account for over-
dispersion in the
data)



Comments
regarding potential
major limitations



























Exposure assessment
limited (use in past year;
no frequency data)








                  This document is a draft for review purposes only and does not constitute Agency policy.
                                                    D-10                         DRAFT—DO NOT CITE OR QUOTE

-------
                                                                                 Supplemental Information—Ammonia
Table D-3. Evaluation of epidemiology studies summarized in Table 1-2 (use in cleaning/disinfection settings)


Reference











Zocket al.
(2007)


















Medina-
Ramon et al.
(2006)





Study setting/
participant selection
completed non-responder survey (sex,
age, nationality, job position); no
major differences with responders.
Selection bias unlikely.







Europe (22 sites in 10 countries).
Longitudinal study. Random
population sample, ages 20-44 yrs
(the European Community Respiratory
Health Survey), 9-yrfollow-up period.
Excluded 764 individuals with asthma
at baseline. Analysis limited to
individuals reporting doing the
cleaning or washing in their home (n =
3,503).










Cornelia, Spain. Two-week diary and
pulmonary function study, 2001-
2002. Female domestic cleaners aged
31-66 yrs with a history of obstructive
lung disease, recruited from
participants in a nested case-control
based on population survey from


Exposure measure
polishes or waxes, solvents, or
carpet cleaners in the last yr









At follow-up, standardized
interview about use of 15
cleaning products in the home
(frequency never, <1 day/week,
1 to 3 days/week, 4 to 7
days/week)
Reference group: did not use the
product or used <1 day/week












2-week diary recordeddaily use
of cleaning products and cleaning
tasks (checklist of cleaning
exposures, number of hours
cleaning in each house).




Outcome measured
"yes" answers to five
questions on asthma
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
• Incident (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
• Incident physician-
diagnosed asthma,
defined as above with
confirmation by a
physician and information
on age or date of first
attack
• Incident (since baseline
survey) current wheeze,
defined as wheezing or
whistling in the chest in
last 12 months when not
having a cold.
• Respiratory symptoms
based on 2-week daily
diary (7 symptoms, 5
point intensity scale);
summed score for upper
respiratory symptoms
(blocked nose, throat

Consideration of
confounding











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)











Adjusted for
respiratory infection,
use of maintenance
medication and age;
daily number of
cigarettes smoked, yrs
of employment in

Statistical
analysis











Incident asthma
and wheeze: log-
binomial
regression
Incident physician
diagnosed asthma:
Cox proportional
hazards
regression, with
date on onset
defined as
reported date of
first attack.
Referent category =
used product never
or <1 day/week




Respiratory
symptom scores
dichotomized as >
and <2 for use in
logistic regression.
PEF analysis based
on night time and
Comments
regarding potential
major limitations











Referent group included
some exposure (to the
product, and to other
products); could
underestimate risk;
although it is an incident
study, the exposure
information was
collected at follow-up so
may not reflect pre-
disease patterns (if
practices changed
because of symptoms) or
could be influenced by
knowledge of outcome





Pulmonary function
measured by participant;
validation of method not
reported.
Potential for knowledge
of exposure to affect
reporting of symptoms
                  This document is a draft for review purposes only and does not constitute Agency policy.
                                                    D-ll                         DRAFT—DO NOT CITE OR QUOTE

-------
                                                                                 Supplemental Information—Ammonia
Table D-3. Evaluation of epidemiology studies summarized in Table 1-2 (use in cleaning/disinfection settings)


Reference

















Medina-
Ramon et al.
(2005)















Study setting/
participant selection
2000-2001 (see Medina-Ramon et al.
(2005), below). Selected if reported
current asthma symptoms or chronic
bronchitis in 2000-2001 survey
(standard definitions). Excluded if
illiterate or unable to complete diary
(n = 57). 80 met eligibility criteria; 51
(64%) completed diary. Participants
and non-participants similar except
for higher prevalence of bronchial
hyperresponsiveness and shorter
duration of domestic cleaning
employment among responders




Cornelia, Spain. Nested case-control
study in 2001-2002 of 650 cleaning
workers drawn from population-
based survey in 2000-2001, 4,521
women ages 30-65 yrs.
Cases: 160 identified, 117 still
employed in domestic cleaning, 87
(74%) agreed to participate, 40 met
final case definition
Controls: 386 identified, 281 still
employed in domestic cleaning, 194
(69%) agreed to participate, 155 met
final control definition






Exposure measure

















Job-specific questionnaire for
cleaning workers, frequency of
use of 22 specific products
(times per week, month, oryr);
summed across each home and
personal home and divided into
two groups (cut-point = 12 times
peryr). Also assessed accidental
exposures (e.g., spills)
Measurements taken in 10
cleaning sessions to obtain data
on exposure to chlorine and
ammonia during specific tasks
and with specific products
(ammonia used in kitchen
cleaning; median 0.6-6.4 ppm;
peaks >50 ppm)


Outcome measured
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
instructions); measured
morning, lunchtime, night
(3 measurements each;
highest recorded).
• Occupational asthma
based on analysis of PEF
patterns by occupational
asthma system (OASYS)
Case based on asthma
and/or bronchitis at both
assessments. 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 yr.
Controls: no history of
respiratory symptoms in
preceding year and no
asthma at either
assessment.


Consideration of
confounding
domestic cleaning
and/or weekly working
hours in domestic
cleaning also assessed
and included as
necessary











Correlations among
tasks/products reported
to be generally weak
(but specific values for
ammonia and other
products not reported).
Multivariate model
adjusted for age tertile
and smoking status (but
results for ammonia in
this model only reported
as "not statistically
significant" — no
information on effect
estimate/variability)



Statistical
analysis
the next morning
values; linear
regression














Logistic regression
















Comments
regarding potential
major limitations

















Results of adjusted
model not reported in
detail, but confounding
unlikely major factor if
correlations weak.












                  This document is a draft for review purposes only and does not constitute Agency policy.
                                                    D-12                         DRAFT—DO NOT CITE OR QUOTE

-------
                                                                                             Supplemental Information—Ammonia
        Table D-4.  Evaluation of epidemiology study summarized in Table 1-6 (industrial setting/serum chemistry
        measures)
Reference
Hamid and
EI-Gazzar
(1996)













Study setting/
participant selection
Egypt, urea fertilizer production
plant; cross sectional study.
Exposed: n = 30
Controls: n = 30
Exposed: workers selected randomly
(process not described). Mean age
36 yrs, mean duration 12 yrs.
Controls from administrative
departments with no known history
of ammonia exposure; matched to
exposed by age, educational status,
and socioeconomic status. Mean age
35 yrs



Exposure
parameters
No direct measurement
of ammonia exposure;
blood urea was used as
a surrogate measure
(ammonia is detoxified
mainly through the
formation of urea in
the liver)
Mean (±SD) mg/dl(p<
0.01)
Exposed: 31.9 (± 7.6)
Controls: 20.3 (±5.1)
The reliability of blood
urea and correlation
with ammonia
exposure not reported
Outcome
measured
Fasting blood sample
for AST, ALT
(measures of liver
function),
hemoglobin, catalase
enzyme activity as
mediator of cell
membrane
permeability and
serum monoamine
oxidase enzyme
activity as mediator
of effects on nervous
system


Consideration of
confounding
No information on
exposure to other
contaminants; no
information on smoking
status











Statistical analysis
Type of statistical test
not reported (EPA
assumes to be t-test);
data presented as group
means ± SD, with p-
value.










Comments regarding
major limitations
Study population and
design: "healthy" workers;
long duration — potential
for lack of complete
ascertainment of effect

Lack of information on
smoking, and alcohol
use— potential for
possible confounding for
liverfunction measures;
uncertain affect on
enzyme measures



ALT = alanine aminotransferase; AST = asparate aminotransferase; SD = standard deviation
                           This document is a draft for review purposes only and does not constitute Agency policy.
                                                               D-13                           DRAFT—DO NOT CITE OR QUOTE

-------
                                                      Supplemental Information—Ammonia
 1
 2   APPENDIX  E.  INFORMATION IN SUPPORT OF
 3   HAZARD  IDENTIFICATION  AND DOSE-RESPONSE
 4   ANALYSIS
 6   E.I. TOXICOKINETICS
 7   Overview
 8         Ammonia can be absorbed by the inhalation and oral routes of exposure. There is less
 9   certainty regarding absorption through the skin, although absorption through the eye has been
10   documented. Most of the inhaled ammonia is retained in the upper respiratory tract and is
11   subsequently eliminated in expired air. Ammonia that reaches systemic circulation is widely
12   distributed to all body compartments, although substantial first-pass metabolism occurs in the
13   liver, where biotransformation into urea and glutamine occurs. Ammonia exists in the blood as
14   ammonium ion (NH4+). Ammonia is transported in the circulatory system primarily via glutamine
15   and alanine, amino acids that are used to transport ammonia to and from tissues. When
16   transported to the liver and kidney, the amide moiety is hydrolyzed via glutaminase forming
17   glutamatic acid (glutamate) and NH4+, which is synthesized into urea and excreted in the urine.
18   Ammonia or NH4+ reaching the tissues is utilized for glutamate production, which participates in
19   transamination and other reactions. The principal means of excretion of absorbed ammonia in
20   mammals is as urinary urea; minimal amounts are excreted in the feces and in expired air.
21         Ammonia is endogenously produced in humans and animals. It is an essential mammalian
22   metabolite used in nucleic acid and protein synthesis, is necessary for maintaining acid-base
23   balance, and is an integral part of nitrogen homeostasis. Given its important metabolic role,
24   ammonia exists in a homeostatically regulated equilibrium in the body.
25
26   E.I.I. Absorption
27   Inhalation Exposure
28         Experiments  with volunteers1 show that ammonia, regardless of its tested concentration in
29   air (range, 40-354 mg/m3), is almost completely retained in the nasal mucosa (83-92%) during
30   short-term acute exposure  (i.e., up to 120 seconds] (Landahl and Herrmann, 1950]. However,
31   longer-term acute exposure (10-27 minutes) to a  concentration of 354 mg/m3 resulted in lower
32   retention (4-30%), with expired breath concentrations of 247-283 mg/m3 observed by the end of
     :The human toxicokinetic studies cited in this section did not provide information on the human subjects' research
     ethics procedures undertaken in the studies; however, there is no evidence that the conduct of the research was
     fundamentally unethical or significantly deficient relative to the ethical standards prevailing at the time the research
     was conducted.
              This document is a draft for review purposes only and does not constitute Agency policy.
                                            E-l            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   the exposure period [Silverman et al., 1949], suggesting saturation of absorption into the nasal
 2   mucosa. Nasal and pharyngeal irritation, but not tracheal irritation, suggests that ammonia is
 3   retained in the upper respiratory tract Unchanged levels of blood urea nitrogen (BUN), nonprotein
 4   nitrogen, urinary urea, and urinary ammonia following these acute exposures are evidence of low
 5   absorption into the blood. Exposure to a common occupational limit of ammonia in air (18 mg/m3),
 6   assuming 30% uptake into blood, would yield an increase in blood ammonia concentration of
 7   0.09 [J.g/mL (calculated by IPCS, 1986]. This calculated rise would likely be indistinguishable from
 8   the observed baseline levels of 0.1-1.0 [ig/mL (Monsen. 1987: Conn. 1972: Brown etal.. 1957] for
 9   healthy controls.
10          Data in rabbits and dogs provide supporting evidence for high-percentage nasal retention,
11   resulting in a lower fraction of the inhaled dose reaching the lower respiratory tract (Egle, 1973:
12   Dalhamn, 1963: Boydetal., 1944].  Continuous exposure of rats to up to 23 mg/m3 for 24 hours did
13   not result in a statistically significant increase in blood ammonia levels (0.1 |J.g/mL above
14   preexposure levels], whereas exposures to 219-818 mg/m3 led to significantly increased blood
15   concentrations of ammonia within 8 hours of exposure initiation; blood ammonia returned to
16   preexposure values within 12 hours of continuous exposure (Schaerdeletal.. 1983].
17
18   Oral Exposure
19          Case reports of human ingestion of household ammonia (ammonium hydroxide] provide
20   evidence of oral absorption, but few quantitative data are available. For example, in a fatal case of a
21   man who drank an unknown amount of a 2.4% solution of ammonium hydroxide, analysis of the
22   contents of the stomach and blood showed NH4+ levels of 15.3 mg and 33 |J.g/mL, respectively
23   (Klendshoj andRejent. 1966]. This blood concentration is about 30-fold higher than the
24   concentration of 1 |J.g/mL in fasting volunteers, as reported by Conn (1972].
25          NH4+ is endogenously produced in the human digestive tract, much of it arising from the
26   bacterial degradation of nitrogenous compounds from ingested food. Approximately 4,200 mg of
27   ammonia are produced each day, with >70% of that amount liberated from fecal contents within
28   the colon (Summerskill and Wolpert, 1970].  About99% of the total amount produced (4,150 mg] is
29   systemically absorbed. Evidence suggests that fractional absorption of ammonia increases as the
30   lumen pH increases, and that active transport occurs at lower pH levels (absorption has been
31   detected at a pH as low as 5] (Castell and Moore.  1971: Mossbergand Ross. 1967]. NH4+absorbed
32   from the gastrointestinal tract travels via the hepatic portal vein directly to the liver where, in
33   healthy individuals, most of it is converted to urea and glutamine.
34
35   Dermal Exposure
36          Quantitative data on absorption from exposure by the dermal route are not available. One
37   report of five case histories of workers exposed to anhydrous ammonia via a burst gas pipe
38   indicated that there was systemic toxicity (vomiting, renal congestion, and delirium], suggesting
39   dermal absorption; however, the fractional dose from dermal exposure could not be determined
40   (Slot, 1938]. IPCS (1986] concluded that systemic effects from skin and eye exposure are not

               This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-2             DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   quantitatively important. Ammonia is readily absorbed into the eye, and it was found to diffuse
 2   within seconds into the cornea, lens, drainage system, and retina [Beare etal.. 1988: Tarudi and
 3   Golden. 1973). However, amounts absorbed were not quantified, and absorption into systemic
 4   circulation was not investigated.
 5
 6   E.1.2. Distribution
 7          The range of mean ammonia concentrations in humans as a result of endogenous
 8   production was reported as 0.1-0.6 |J.g/mL in arterial blood and 0.2-1.7 |J.g/mL in venous blood
 9   [Huizengaetal.. 1994). Other baseline levels observed in volunteers range from 1 to 5.5 |J.g/mL
10   [Conn. 1972: Brown etal.. 1957). Ammonia is homeostatically regulated to remain at low
11   concentrations, with 95-98% existing in the blood (at physiological pH) as NH4+ [da Fonseca-
12   Wollheim. 1995: Souba. 19871.
13          Ammonia is present in fetal circulation. In vivo studies in several animal species and in
14   vitro studies of human placenta suggest that ammonia is produced within the uteroplacenta and
15   released into the fetal and maternal circulations [Bell etal., 1989: Tohnsonetal., 1986: Hauguel et
16   al.. 1983: Meschiaetal.. 1980: Remesar etal.. 1980: Holzmanetal.. 1979: Holzmanetal.. 1977:
17   Rubaltelli and Formentin. 1968: Luschinsky. 1951). Tozwiketal. [2005] reported that ammonia
18   levels in human fetal blood (specifically umbilical arterial and venous blood) at birth were 1.0-
19   1.4 [J.g/mL, compared to 0.5 |J.g/mL in the mothers' venous blood. DeSanto etal. (1993] similarly
20   collected human umbilical arterial and venous blood at delivery and found that umbilical arterial
21   ammonia concentrations were significantly higher than venous concentrations; there was no
22   correlation between umbilical ammonia levels and gestational age (range of 25-43 weeks of
23   gestation]. In sheep, the uteroplacental tissue is the main site of ammonia production, with outputs
24   of ammonia into both the uterine and umbilical circulations (Tozwiketal.. 1999].  In late-gestation
25   pregnant sheep that were catheterized to allow measurement of ammonia exposure to the fetus,
26   concentrations of ammonia in umbilical arterial and venous blood and uterine arterial and venous
27   blood ranged from approximately 0.39 to 0.60 |J.g/mL (Tozwiketal., 2005: Tozwiketal., 1999].
28          Ammonia is present in human breast milk as one of the sources of nonprotein nitrogen
29   (Atkinson etal.. 1980].
30
31   Inhalation Exposure
32          Little information was found in the available literature on the distribution of inhaled
33   ammonia. Information on the distribution of endogenously produced ammonia suggests that any
34   ammonia absorbed through inhalation would be distributed to all body compartments via the
35   blood, where it would be used in protein synthesis as a buffer, reduced to normal concentrations by
36   urinary excretion, or converted by the liver to glutamine and urea (Takagaki etal., 1961].  Rats
37   inhaling 212 mg/m3 ammonia 6 hours/day for 15  days exhibited increased blood ammonia (200%]
38   and brain glutamine (28%] levels at5 days of exposure, butnotatlO or 15 days (Manninenetal..
39   1988], demonstrating transient distribution of ammonia to the brain (metabolic adaptation].
40

               This document is a draft for review purposes only and does not constitute Agency policy.
                                              E-3             DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information — Ammonia
 1   Oral Exposure
 2          Human oral exposure data indicate that ammonia readily enters the portal circulation and is
 3   delivered to the liver, as has been shown to be the case for endogenously produced ammonia [Pitts.
 4   1971: Summerskill and Wolpert, 1970]. Un-ionized ammonia is freely diffusible, whereas the NH4+
 5   is less so, and is relatively confined to the extracellular compartment [Stabenauetal., 1959].
 6
 7   Dermal Exposure
 8          No quantitative data on distribution of ammonia from dermal exposure were located in the
 9   available literature.
10
11   E.1.3. Metabolism
12          Endogenously, ammonia is produced by catabolism of amino acids by glutamate
13   dehydrogenase primarily in the liver and renal cortex, but also in the brain and heart [Souba, 1987].
14   In skeletal muscle, ammonia may be produced by metabolism of adenosine monophosphate via
15   adenylate deaminase.  Information on the metabolism of exogenously-introduced ammonia was not
16   found in the available literature.  Ammonia and NH4+ are metabolized to glutamine mainly in the
17   liver via glutamine synthetase in the glutamine cycle (Figure E-l], or incorporated into urea as part
18   of the urea cycle as observed in the hepatic mitochondria and cytosol (Figure E-2] (Nelson and Cox,
19   2008]. Ammonia can be rapidly converted to glutamine in the brain as well (Takagaki etal., 1961].
20   van de Poll etal. (2008] reported that the liver removes an amount of ammonia from circulation
21   equal to the amount added by the intestines at metabolic steady state, indicating that the gut does
22   not contribute significantly to systemic ammonia release.
23
                                              Glutamate     ATP

                                                                  ADP
                                        utaminase                  ~
                                      (in li^
                                      mitochondria)
                                 Glutamine             y-Glutamyl
                                                       phosphate
24
25
26   Adapted from: Nelson and Cox f20081
27
28   Figure E-l. Glutamine cycle.
29
               This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-4             DRAFT—DO NOT CITE OR QUOTE

-------
                                                           Supplemental Information—Ammonia
 1
 2
 o
 6
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
                                        CO, +NH/
                                      H2O
                                     2 ATP
                                      3H+
                                                  MlTOCHONDKJAI>
                                                  MATRIX
                                              Carbamoyl
                                              phosphate
                                              synthase I
                                          Carbamoyl
                                          phosphate
                                       Ornithine      Citrulline
                                              Omithine
                                            transcarbamoylase
                           Urea
                                 Ornithine
                                    Arginase
                            H,O
                                                            ATP
                                             Argininosuccinate j^Aspartate
                                                synthase
                                                                  AMP + PPj
                                   Arginine            Argininosuccinate
                                          Argininosuccinate
                                              lyase
                                                             CYTOSOL
                                   Fumarate
Adapted from: Nelson and Cox [2008].
Figure E-2. The urea cycle showing the compartmentalization of its steps within liver
cells.
       Given its important metabolic role, ammonia exists in a homeostatically regulated
equilibrium in the body.  In particular, free ammonia has been shown to be homeostatically
regulated to remain at low concentrations, with 95-98% of body burden existing in the blood (at
physiological pH) as NH4+ [da Fonseca-Wollheim. 1995: Souba. 1987). Two studies in rats
[Manninen etal.. 1988: Schaerdeletal.. 1983] provide evidence that ammonia concentrations
<18 mg/m3 in air do not alter blood ammonia concentrations. Schaerdel etal. [1983] exposed rats
to ammonia for 24 hours at concentrations of 11-818 mg/m3. Exposure to 11 mg/m3 ammonia did
not increase blood ammonia concentrations after 24 hours; concentrations of >23 mg/m3 caused an
exposure-released increase in blood ammonia, but concentrations at 12- and 24-hour sampling
periods were lower than at 8 hours, suggesting compensation by increasing ammonia metabolism
through conversion to urea, pyrimidine and polyamine synthesis, incorporation into amino acid
substrates, and metabolism in nervous system tissue. Rats inhaling 18 mg/m3 ammonia
6 hours/day for 5 days did not exhibit blood or brain ammonia or glutamine levels that were
different from controls; however, rats inhaling 212 mg/m3 for the same  daily exposure exhibited
statistically significantly increased levels of blood ammonia (threefold] and brain glutamine
(approximately 40%] at 5 days of exposure, but not at 10 or 15 days (Manninen et al., 1988]. The
return of blood and brain ammonia and glutamine levels to control levels with time is consistent
               This document is a draft for review purposes only and does not constitute Agency policy.
                                                E-5              DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1   with metabolic adaptation, and these data suggest that animals have a large capacity to handle high
 2   concentrations of inhaled ammonia.
 3          Various disease states can affect the rate of glutamine uptake and catabolism and thereby
 4   affect the blood and tissue levels of ammonia. Abnormally elevated levels of ammonia are
 5   indicative of end-stage renal failure [Davies etal., 1997]. Acute renal failure can result in increased
 6   renal glutamine consumption and ammonia production with a decreased capability of eliminating
 7   urea in the urine  [Souba, 1987]. End-stage liver failure due to fulminant hepatitis or hepatic
 8   cirrhosis may result in decreased ureagenesis and increased levels of ammonia in blood
 9   (hyperammonemia], leading to increased uptake into the brain and the onset of hepatic
10   encephalopathy.  The increased metabolic alkalosis associated with hepatic encephalopathy may
11   result in a shift in the NH4+/NH3 ratio in the direction of ammonia, which could pass through the
12   blood-brain barrier [Katayama, 2004]. In patients with liver cirrhosis and acute clinical hepatic
13   encephalopathy, the observed trapping of [13N]-ammonia in the brain appeared to be related to a
14   fivefold increase of ammonia permeability across the blood-brain barrier relative to healthy
15   controls [Keidingetal., 2010: Keiding et al., 2006].  Furthermore, S0rensenetal. [2009]
16   demonstrated greater unidirectional clearance of ammonia from the blood to brain cells than
17   metabolic clearance of ammonia from the blood both in healthy controls and in cirrhotic patients
18   with and without hepatic encephalopathy.
19
20   E.1.4. Elimination
21          Absorbed ammonia, as well as endogenously produced ammonia, is excreted by the kidneys
22   as urea [Summerskill and Wolpert, 1970: Gay etal., 1969: Muntwyler et al., 1956: Davies and
23   Yudkin. 1952: Van Slyke etal.. 19431 and is a component of sweat fGuyton. 1981: Wands. 19811
24   Acidosis-stimulated renal excretion of ammonia is mediated by intercalated cell-specific Rh B
25   glycoprotein expression in mice [Bishop etal., 2010: Lee etal., 2010: Lee etal., 2009].  In rat kidney,
26   NH4+ is secreted into the lumen of the outer medullary collecting duct via H+ secretion and parallels
27   ammonia diffusion [Flessner etal., 1992]. The inner medullary collecting duct exhibits a Na+- and
28   K+-independent NH4+/H+ exchange activity that may be mediated by an Rh C glycoprotein
29   [Handlogtenetal., 2005], which is also expressed in human kidneys [Han etal., 2006].
30          Additionally, ammonia is known to be present in the expired air of all humans [Manolis.
31   1983]. Three investigators specifically measured ammonia in breath exhaled from the nose
32   [Schmidt etal.. 2013: Smith etal.. 2008: Larson etal.. 1977]. Smith etal.  [2008] reported median
33   ammonia concentrations of 0.059-0.078 mg/m3 in exhaled breath from the nose of three healthy
34   volunteers (with  samples collected daily over a 4-week period]; these concentrations were similar
35   to or slightly higher than the mean laboratory air level of ammonia reported in this study of
36   0.056 mg/m3. In another study of 20 health volunteers, the mean ammonia concentration in
37   exhaled breath from the nose was 0.032 mg/m3 (range: 0.0092-0.1 mg/m3] fSchmidtetal.. 20131
38   Larson etal. [1977] reported that the median concentration of ammonia collected from air samples
39   exhaled from the nose ranged from 0.013 to 0.046 mg/m3.  One sample collected from the trachea
               This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-6             DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1   via a tube inserted through the nose of one subject was 0.029 mg/m3—a concentration within the
 2   range of that found in breath exhaled through the nose [Larson etal.. 1977).
 3          Higher and more variable ammonia concentrations are reported in breath exhaled from the
 4   mouth or oral cavity than in breath exhaled from the nose.  In studies that reported ammonia in
 5   breath samples from the mouth or oral cavity, ammonia concentrations were commonly found in
 6   the range of 0.085-2.1 mg/m3 (Schmidtetal.. 2013: Smith etal.. 2008: Spaneletal.. 2007a. bj
 7   Turner etal.. 2006: Diskin etal.. 2003: Smith etal.. 1999: Norwood etal.. 1992: Larson etal.. 19771.
 8   and strongly correlated with saliva pH (Schmidt etal., 2013]. These higher concentrations are
 9   largely attributed to the production of ammonia by bacterial degradation of food protein in the oral
10   cavity or gastrointestinal tract (Turner etal.. 2006: Smith etal.. 1999: Vollmuth and Schlesinger.
11   1984]. This source of ammonia in breath was demonstrated by Smith etal. (1999], who observed
12   elevated ammonia concentrations in the expired air of six healthy volunteers following the
13   ingestion of a protein-rich meal.
14          Other factors that can affect ammonia levels in breath exhaled from the mouth or oral cavity
15   include diet, oral hygiene, age, living conditions, and disease state. Norwood etal. (1992] reported
16   decreases in baseline ammonia levels (0.085-0.905 mg/m3] in exhaled breath following tooth
17   brushing (<50% depletion], a distilled water oral rinse (<50% depletion], and an acid oral rinse
18   (80-90% depletion]. These findings are consistent with ammonia generation in the oral cavity by
19   bacterial and/or enzymatic activity. Several investigators have reported that ammonia in breath
20   from the mouth and oral cavity increases with age (Spaneletal., 2007a, b; Turner etal., 2006:
21   Diskin etal., 2003], with ammonia concentrations increasing on average about 0.1 mg/m3 for each
22   10 years of life (Spaneletal., 2007a]. Turner etal. (2006] reported that the age of the individual
23   accounts for about 25% of the variation observed in mean breath ammonia levels, and the
24   remaining 75% is due to factors other than age. Certain disease states can also influence ammonia
25   levels in exhaled breath. Ammonia is greatly elevated in the breath of patients in renal failure
26   (Spanel etal., 2007a: Davies etal., 1997]. These studies are further described in Table E-l.
27          Because ammonia measured in samples of breath exhaled from the mouth or oral cavity can
28   be generated in the oral cavity and may thus be substantially influenced by diet and other factors,
29   ammonia levels measured in mouth or oral cavity breath samples do not likely reflect systemic
30   (blood] levels of ammonia.  Ammonia concentrations in breath exhaled from the nose appear to
31   better represent levels at the alveolar interface  of the lung and are thought to be more relevant to
32   understanding systemic levels of ammonia (Schmidt etal., 2013: Smith etal., 2008].
33          Ammonia has also been detected in the expired air of animals.  Whittaker et al. (2009]
34   observed a significant association between ambient ammonia concentrations and increases in
35   exhaled ammonia in stabled horses. Analysis of endogenous ammonia levels in the expired air of
36   rats showed concentrations of 0.007-0.250 mg/m3 (mean = 0.06 mg/m3] (Barrow and Steinhagen,
37   1980]. Larson etal. (1980] reported ammonia concentrations measured in the larynx of dogs
38   exposed to  sulfuric acid ranging between 0.02 and 0.16 mg/m3 following mouth breathing and
39   between 0.04 and 0.16 mg/m3 following nose breathing.
40

               This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-7             DRAFT—DO NOT CITE OR QUOTE

-------
        Table E-l. Ammonia levels in exhaled breath of volunteers
                                                                                                        Supplemental Information—Ammonia
     Test subjects
       Breath samples
 Levels of ammonia in exhaled breath
     Methods
          Comments
 Reference
Breath samples from the nose and trachea
20 healthy volunteers
(13 males and 7 females
aged 22-6lyrs)
Subjects fasted overnight and
refrained from exercise in the
morning before sampling;
samples collected between
8 and 11 AM; end-tidal breath
samples collected from the
nose; subjects breathed
continuously into the sampling
piece for 3-5 min to obtain
stable sample; samples also
collected after an acidic mouth
wash
Concentrations in exhaled breath from
the nose (mg/m3):
Range = 0.0092-0.10
Mean = 0.032 (95% Cl: 0.021-0.042)
Median = 0.024

Concentrations following acidic mouth
wash (mg/m3):
Range = 0.011-0.027
Mean = 0.016 (95% Cl: 0.014-0.018)
Median = 0.015
Commercial cavity
ring-down
spectrometer
Ammonia concentrations in
outdoor air were down to
0.0004 g/m3, in indoor air were
0.002-0.004 mg/m3, and in
indoor air in the presence of
humans were 0.006-
0.007 mg/m3
Schmidt et
al. (2013)
Three healthy male
volunteers (>30 yrs of
age)
Ammonia levels measured in
nose-exhaled breath of test
subjects each morning about
2 hrs after eating a regular
breakfast; samples collected
daily over a 4-wk period
Volunteer A = 0.0728 ± 0.000848 mg/nr
Volunteer B = 0.0777 ± 0.000919 mg/m3
Volunteer C = 0.0587 ± 0.000848 mg/m3

(median ammonia levels estimated as
geometric mean ± geometric SD)
SIFT-MS analysis
Mean ambient air level of
ammonia was 0.056 ±
0.0071 mg/m3

The authors indicated that
ammonia measured in mouth-
exhaled breath may be
generated in the oral cavity and
suggested that concentrations in
nose-exhaled breath may better
represent systemic conditions
(such as metabolic disease)
Smith et al.
(2008)
Sixteen healthy subjects
(9 males aged 25-63 yrs
and 7 females aged 23-
41 yrs); subgroups
tested were all male
Breath samples collected during
quiet nose breathing, and direct
sampling during a deep
inspiration followed by breath-
holding with the glottis closed
Ammonia concentrations ranged from
0.013 to 0.046 mg/m3 during nose
breathing (median 0.025 mg/m )
(five male subjects), and 0.029 mg/m3
from an air sample collected from the
trachea (collected from a tube inserted
into one male subject's nose and into
the trachea)
Chemiluminescence
                             Larson et al.
                                    This document is a draft for review purposes only and does not constitute Agency policy.
                                                                      E-8                                 DRAFT—DO NOT CITE OR QUOTE

-------
        Table E-l. Ammonia levels in exhaled breath of volunteers
                                                                                                       Supplemental Information—Ammonia
     Test subjects
       Breath samples
 Levels of ammonia in exhaled breath
     Methods
          Comments
 Reference
Breath samples from the mouth and oral cavity
20 healthy volunteers
(13 males and 7 females
aged 22-6lyrs)
Subjects fasted overnight and
refrained from exercise in the
morning before sampling;
samples collected between
8 and 11 AM; end-tidal breath
samples collected from the
mouth; subjects breathed
continuously into the sampling
piece for 3-5 min to obtain
stable sample; samples also
collected after an acidic mouth
wash
Concentrations in exhaled breath from
the mouth (mg/m3):
Range = 0.28-1.5
Mean = 0.55 (95% Cl: 0.42-0.68)
Median = 0.49

Concentrations following acidic mouth
wash (mg/m ):
Range = 0.010-0.027
Mean = 0.015 (95% Cl: 0.014-0.018)
Median = 0.015
Commercial cavity
ring-down
spectrometer
Ammonia concentrations in
outdoor air were down to
0.0004 mg/m3, in indoor air were
0.002-0.004 mg/m3, and in
indoor air were 0.006-
0.007 mg/m3
Schmidt et
Three healthy male
volunteers (>30 yrs of
age)
Ammonia levels measured in
mouth-exhaled breath and in
the closed mouth cavity of test
subjects each morning about
2 hrs after eating a regular
breakfast; samples collected
daily over a 4-wk period
Via mouth:
Volunteer A = 0.769 ± 0.000919 mg/m3
Volunteer B = 0.626 ± 0.000919 mg/m3
Volunteer C = 0.604 ± 0.000919 mg/m3

Via oral cavity:
Volunteer A = 1.04 ± 0.000990 mg/m3
Volunteer B = 1.52 ± 0.00106 mg/m3
Volunteer C = 1.31 ± 0.000919 mg/m3
                                                  (median ammonia levels estimated as
                                                  geometric mean ± geometric SD)
SIFT-MS analysis
Mean ambient air level of
ammonia was 0.056 ±
0.0071 mg/m3

The authors indicated that
ammonia measured in mouth-
exhaled breath may be
generated in the oral cavity and
suggested that concentrations in
nose-exhaled breath may better
represent systemic conditions
(such as metabolic disease)
Smith et al.
                                   This document is a draft for review purposes only and does not constitute Agency policy.
                                                                      E-9                                DRAFT—DO NOT CITE OR QUOTE

-------
                                                                                    Supplemental Information—Ammonia
Table E-l. Ammonia levels in exhaled breath of volunteers
Test subjects
Four healthy children
(two males and two
females, 4-6 yrs old)

Thirteen senior
volunteers (11 males
and 2 females, 60-
83 yrs old); four had
type-2 diabetes mellitus
with onset at ages
between 50 and 70 yrs,
and controlled by diet
All subjects had their
regular breakfast
without any specific
restrictions
Twenty-six secondary
school students
(10 males and
16 females, 17-18 yrs
old and one 19-yr-old)



Breath samples
Breath samples collected in
morning at least 1 hr after
breakfast and at least 1 hr prior
to lunch; each volunteer
performed two
exhalation/inhalation cycles
(both about 5-10 sec in
duration)








Three sequential breath
exhalations collected over
5 min following the students
listening to a 1-hr presentation
(at least 1 hr following
breakfast and before lunch);
alveolar portion measured
(identified using humidity)
Levels of ammonia in exhaled breath
Children = range 0.157-0.454 mg/m3
Seniors = 0.224-1.48 mg/m3














Median values reported for:
17-yr-olds = 0.165 mg/m3
18-yr-olds = 0.245 mg/m3





Methods
SIFT-MS analysis















SIFT-MS analysis







Comments
Ammonia breath levels
significantly increased with age

Some seniors reported diabetes

Measured ammonia level in
breath reported for each subject









Significant differences in
ammonia levels in exhaled
breath between 17- and 18-yr-
olds (p < 10"8) were reported
(statistical test not reported)



Reference
Spanel et al.
(2007a)














Spanel et al.
(2007b)






                        This document is a draft for review purposes only and does not constitute Agency policy.
                                                      E-10                            DRAFT—DO NOT CITE OR QUOTE

-------
Table E-l. Ammonia levels in exhaled breath of volunteers
                                                                                    Supplemental Information—Ammonia
Test subjects
Thirty healthy
volunteers (19 males
and 11 females, 24-
59 yrs, 28 Caucasian,
1 African, and 1 mixed
race); volunteers were
instructed to maintain
their normal daily
routines and to not
rinse out their mouths
prior to providing a
breath sample
Five subjects (two
females, three males;
age range 27-65 yrs)
Six normal nonsmoking
male volunteers (24-
61 yrs old), fasted for
12 hrs prior to testing
Breath samples
Breath samples collected in the
morning prior to lunch at
approximately weekly intervals
for about 6 mo; some
volunteers provided samples
more frequently than others;
480 samples collected and
analyzed for ammonia
Breath samples collected
between 8 and 9 AM in three
sequential breath exhalations
on multiple days (12-30 d) over
the course of a month
Baseline breath sample
obtained; breath samples
collected 20, 40, and 60 min
and 5 hrs following the
ingestion of a liquid protein-
calorie meal
Levels of ammonia in exhaled breath
Geometric mean and geometric
SD = 0.589 ± 0.00114 mg/m3
Median = 0.595 mg/m3
Range = 0.175-2.08 mg/m3
Ammonia concentrations were 0.298-
1.69 mg/m3
Premeal levels were 0.2-0.4 mg/m3;
Postmeal levels at 30 min were
0.1 mg/m3 increasing to maximum
values at 5 hrs of 0.4-1.3 mg/m3
Methods
SIFT-MS analysis
SIFT-MS analysis
SIFT-MS analysis
Comments
Ammonia breath levels were
shown to increase with age
Background levels in the testing
laboratory were typically around
0.28 mg/m3
Differences in ammonia breath
levels between individuals were
significant (p < 0.001; ANOVA
test)
A biphasic response in breath
ammonia concentration was
observed after eating
Reference
Turner et al.
(2006)
Diskin et al.
(2003)
(Smith etal..
1999)

                        This document is a draft for review purposes only and does not constitute Agency policy.
                                                      E-ll                            DRAFT—DO NOT CITE OR QUOTE

-------
Table E-l. Ammonia levels in exhaled breath of volunteers
                                                                                    Supplemental Information—Ammonia
Test subjects
Fourteen healthy,
nonsmoking subjects
(age range 21-54 yrs)
performed one or more
of the following hygiene
maneuvers:
(1) acidic oral rinse
(PH2.5)
(2) tooth brushing
followed by acidic oral
rinse
(3) tooth brushing
followed by distilled
water rinse
(4) distilled water rinse
Sixteen healthy subjects
(nine males aged 25-
63 yrs and seven
females aged 23-
41 yrs); subgroups
tested were all male
Breath samples
Subjects fasted for 8 hrs prior to
baseline measurement,
refrained from oral hygiene
after their most recent meal,
refrained from heavy exercise
for 12 hrs, and had no liquid
intake for several hours; initial
breath ammonia was measured
between 8 and 10 AM, then
subjects performed one or
more of the hygiene measures
listed (at 30-min intervals for a
total 90-min period; samples
collected over 5 min)

Breath samples collected during
quiet mouth breathing




Levels of ammonia in exhaled breath
Baseline levels varied from 0.085 to
0.905 mg/m3













Ammonia concentrations ranged from
0.029 to 0.52 mg/m3 during mouth
breathing (median of 0.17 mg/m3)



Methods
Nitrogen oxide
analyzer with an
ammonia
conversion channel
(similar to chemi-
luminescence)









Chemiluminescence





Comments
An 80-90% depletion of volatile
ammonia emissions was seen
within 10 min of acid rinsing;
<50% depletion of ammonia was
seen following tooth brushing or
distilled water rinse; gaseous
ammonia levels increased after
all rinse procedures over time







The oral cavity appears to be a
source of breath ammonia; no
attempt was made to control the
diet of subjects or standardize
the interval between the last
meal and the measurement
Reference
Norwood et
al. (1992)













Larson et al.
(1977)




                        This document is a draft for review purposes only and does not constitute Agency policy.
                                                      E-12                            DRAFT—DO NOT CITE OR QUOTE

-------
Table E-l. Ammonia levels in exhaled breath of volunteers
                                                                                    Supplemental Information—Ammonia
Test subjects
Breath samples
Levels of ammonia in exhaled breath
Methods
Comments
Reference
Breath samples: source (nose/mouth/oral cavity) not specified
Sixteen healthy,
nonsmoking subjects
(4 females and
12 males, 29 ± 7 yrs); no
significant differences in
mean age, height,
weight, BMI, or time
since last oral intake;
10 subjects tested in
each experiment










Experiment 1: single whole-
breath samples collected from
each subject (same samples
immediately reanalyzed within
<10 sec to assess instrument
specific variability)

Experiment 2: three repeat
breath samples collected from
each subject (to evaluate intra-
subject differences); this
experiment evaluated
differences based on
standardization of expiratory
pressure and flow
Experiment 3: two mixed
breath samples and two bag
alveolar breath samples
collected in short succession
from each subject
Experiment 1: 0.843 ± 0.0601 mg/m3
(median ± measurement error)

Experiment 2:
Nonstandardized = 0.712 ± 0.130 mg/m3
(median ±SD)
Standardized = 1.01 ± 0.113 mg/m3
(median ±SD)

Experiments:
Mixed = 0.860 ± 0.585 mg/m3 (median ±
SD)
Alveolar = 0.920 ± 0.559 mg/m3 (median
±SD)






SIFT-MS analysis

This study
established that
SIFT-MS analysis is
reliable and
repeatable













Relatively small number of
healthy subjects used

Did not address the breath of
those with disease

Intra- and inter-day repeatability
were not investigated












Boshier et
al. (2010)


















                        This document is a draft for review purposes only and does not constitute Agency policy.
                                                      E-13                            DRAFT—DO NOT CITE OR QUOTE

-------
       Table E-l. Ammonia levels in exhaled breath of volunteers
                                                                                                Supplemental Information—Ammonia
Test subjects
Eight healthy subjects
(average age
39.8 ± 9.6 yrs)











Three groups of children
were used as test
subjects:
(1) 68 asthmatic
children residing in a
National Park in the
mountains (mean age
10 yrs, 48 boys, 20 girls)
(2) 52 asthmatic
children in an urban
area (mean age 9 yrs,
35 boys, 17 girls)
(3) 20 healthy children
from the same urban
area as a control group
(mean age 10 yrs,
12 boys, 8 girls)
Breath samples
Subjects fasted for 6 hrs prior to
samples being collected;
subjects breathed normally into
collection device for 5 min










Subjects performed a 5-sec
breath-hold and exhaled slowly
into collection device














Levels of ammonia in exhaled breath
Mean breath ammonia = 0.35 ±
0.17 mg/m3












Asthmatic children from National Park =
0.0040 ± 0.0033 mg/m3

Asthmatic urban children:
Mean NH3 = 0.0101 ± 0.00721 mg/m3

Urban children control group:
Mean NH3 = 0.0105 ± 0.00728 mg/m3









Methods
Fiberoptic sensor













Chemiluminescence
















Comments
This study measured ammonia
levels in healthy volunteers
compared to Helicobacter pylori
positive individuals (five
subjects) (data not shown); the
experiment also included a
challenge with a 300 mg urea
capsule to evaluate the urease
activity of healthy versus
infected individuals (data not
shown); the authors concluded
that breath ammonia
measurement may be feasible as
a diagnostic test for H. pylori
Both groups of asthmatic
children had some subjects on
glucocorticoids, often combined
with histamine antagonists
and/or b2 agonists, while others
were left untreated; ammonia
concentrations in exhaled breath
appeared to be correlated with
exposure to urban air








Reference
Kearney et
al. (2002)












Giroux et al.
(2002)















ANOVA= analysis of variance; BMI = body mass index; Cl = confidence interval; SD = standard deviation; SIFT-MS = selected ion flow tube mass spectrometry
                                 This document is a draft for review purposes only and does not constitute Agency policy.
                                                                E-14                              DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1   Physiologically Based Pharmacokinetic Models
 2          No physiologically based pharmacokinetic models have been developed for ammonia. An
 3   expanded one-compartment toxicokinetic model in rats was developed by Diack and Bois [2005].
 4   which used physiological values to represent first-order uptake and elimination of inhaled
 5   ammonia (and other chemicals).  The model is not useful for dose-response assessment of ammonia
 6   because: (1] it cannot specify time-dependent amounts or concentrations of ammonia in specific
 7   target tissues, (2) it has not been verified against experimental data for ammonia, glutamate, or
 8   urea levels in tissues, and (3) it cannot extrapolate internal doses of ammonia between animals and
 9   humans.
10

11   E.2.  HUMAN STUDIES
12          More detailed summaries are provided of epidemiology studies of workers in industrial
13   exposure settings that examined respiratory parameters; information from these studies was used
14   as the basis for the RfC.
15
16   E.2.1. Occupational Studies in Industrial Worker Populations
17   Holness et al. (1989)
18          Holness etal. [1989] conducted a cross-sectional study of workers in a soda ash (sodium
19   carbonate] plant2 who had chronic, low-level exposure to ammonia. The cohort consisted of
20   58 workers and 31 controls from stores and office areas of the plant  All workers were males
21   (average age 43 years], and the average exposure duration for the exposed workers at the plant
22   was 12 years. The mean time-weigh ted average (TWA] ammonia exposure of the exposed group
23   based on personal sampling over one work shift (mean sample collection time  8.4 hours] was
24   9.2 ppm (6.5 mg/m3] compared to 0.3 ppm (0.2 mg/m3] for the control group.  The average
25   concentrations of ammonia to which workers were exposed were determined using the procedure
26   recommended by the National Institute for Occupational Safety and Health (NIOSH], which involves
27   the collection of air samples on sulfuric acid-treated silica gel adsorption tubes (NIOSH. 1979].
28          No statistically significant differences were observed in age, height, years worked,
29   percentage of smokers, or pack-years smoked for exposed versus control workers. Exposed
30   workers weighed approximately 8% (p < 0.05] more than control workers. Information regarding
31   past occupational exposures, working conditions, and medical and smoking history, as well as
32   respiratory symptoms and eye and skin complaints was obtained by means of a questionnaire that
33   was based on an American Thoracic Society  questionnaire (Ferris. 1978].  Each participant's sense
34   of smell was evaluated at the beginning and  end of the work week using several concentrations of
35   pyridine (0.4, 0.66, or 10 ppm]. Lung function tests were conducted atthe beginning and end of the
36   work shift on the first and last days of their work week (four tests administered].  Differences in
      At 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.
                                               E-15            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   reported symptoms and lung function between groups were evaluated using the actual exposure
 2   values with age, height, and pack-years smoked as covariates in linear regression analysis. Exposed
 3   workers were grouped into three exposure categories (high = >12.5 ppm [>8.8 mg/m3], medium =
 4   6.25-12.5 ppm [4.4-8.8 mg/m3], and low = <6.25 ppm [<4.4 mg/m3]) for analysis of symptom
 5   reporting and lung function data.
 6          Endpoints evaluated in the study included sense of smell, prevalence of respiratory
 7   symptoms (cough, bronchitis, wheeze, dyspnea, and others), eye and throat irritation, skin
 8   problems, and lung function parameters (forced vital capacity [FVC], forced expiratory volume in
 9   1 second [FEVi], FEVi/FVC, forced expiratory flow [FEF50], and FEF75). No statistical differences in
10   the prevalence of respiratory irritation or eye irritation were evident between the exposed and
11   control groups (Table E-2).
12          There was a statistically significant increase (p < 0.05) in the prevalence of skin problems in
13   workers in the lowest exposure category (<4.4 mg/m3) compared to controls; however, the
14   prevalence was not increased among workers in the two higher exposure groups.  Workers also
15   reported that exposure at the plant had aggravated specific symptoms including coughing,
16   wheezing, nasal complaints,  eye irritation, throat discomfort, and skin problems. Odor detection
17   threshold and baseline lung functions were similar in the exposed and control groups.  No changes
18   in lung function were demonstrated over either work shift (days 1 or  2) or over the work week in
19   the exposed group compared with controls. No relationship  was demonstrated between chronic
20   ammonia exposure and baseline lung function changes either in terms of the level or duration of
21   exposure. Study investigators noted that this finding was limited by the lack of adequate exposure
22   data collected over time, precluding development of a meaningful index accounting for both level
23   and length of exposure. Based on the lack of exposure-related differences in subjective
24   symptomatology, sense of smell, and measures of lung function, EPA identified 8.8 mg/m3 as the no-
25   observed-adverse-effect level (NOAEL). A lowest-observed-adverse-effect level (LOAEL) was not
26   identified for this study.
27
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-16            DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
            Table E-2. Symptoms and lung function results of workers exposed to
            different levels of TWA ammonia concentrations
Parameter
Ammonia concentration
Control
0.2 mg/m3
Exposed
<4.4 mg/m3
Exposed
4.4-8.8 mg/m3
Exposed
>8.8 mg/m3
Symptom
Cough
Sputum
Wheeze
Chest tightness
Shortness of breath
Nasal complaints
Eye irritation
Throat irritation
Skin problems
3/31 (10)a
5/31 (16)
3/31 (10)
2/31 (6)
4/31 (13)
6/31 (19)
6/31 (19)
1/31 (3)
2/31 (6)
6/34 (18)
9/34 (26)
5/34 (15)
2/34 (6)
3/34 (9)
4/34 (12)
2/34 (6)
2/34 (6)
10/34* (29)
1/12 (8)
3/12 (25)
1/12 (8)
0/12 (0)
1/12 (8)
2/12 (17)
2/12 (17)
1/12 (8)
1/12 (8)
2/12 (17)
1/12 (8)
0/12 (0)
0/12 (0)
0/12 (0)
0/12 (0)
1/12 (8)
1/12 (8)
1/12 (8)
Lung function (% predicted)
FVC
FEVi
FEF50
FEF75
98.6
95.1
108.4
65.2
96.7
93.7
106.9
71.0
96.9
93.9
106.2
67.8
96.8
95.3
111.2
78.8
     al\lumber affected/number examined. The percentage of workers reporting symptoms is indicated in parentheses.
     *Significantly different from controls, p < 0.05, by Fisher's exact test performed for this review.
     Source: Holness et al. (1989).
 1
 2    Ballaletal. (1998)
 3           Ballal etal. [1998] conducted a cross-sectional study of male workers at two urea fertilizer
 4    factories in Saudi Arabia3. The cohort consisted of 161 exposed subjects (84 from factory A and
 5    77 from factory B) and 355 unexposed controls. Workers in factory A were exposed to air ammonia
 6    levels of 2-130 mg/m3, and workers in factory B were exposed to levels of 0.02-7 mg/m3. Mean
 7    duration of employment was 51.8 months for exposed workers and 73.1 months for controls.
 8    Exposure levels were estimated by analyzing a total of 97 air samples collected over 8-hour shifts
 9    close to the employee's work site. The prevalence of respiratory symptoms and diseases was
10    determined by administration of a questionnaire. The authors stated that there were no other
11    chemical pollutants in the workplace that might have affected the respiratory system. Smoking
12    habits were similar for exposed workers and controls.
13           Stratifying the workers by ammonia exposure levels (above or below the American
14    Conference of Governmental Industrial Hygienists [ACGIH] threshold limit value [TLV] of
     3The 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.

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                                E-17            DRAFT—DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
                                                         Supplemental Information — Ammonia
     18 mg/m3) showed that those exposed to ammonia concentrations higher than the TLV had 2.2- to
     fourfold higher relative risks for cough, phlegm, wheezing, dyspnea, and asthma than workers
     exposed to levels below the TLV (Table E-3). The relative risk for wheezing was also elevated
     among those exposed to ammonia levels at or below the TLV. Distribution of symptoms by
     cumulative ammonia concentration (CAC, mg/m3-years) also showed 2- to 4.8-fold higher relative
     risk for all of the above symptoms among those with higher CAC (Table E-3). Results of the logistic
     regression analysis showed that ammonia concentration was significantly related to cough, phlegm,
     wheezing with and without shortness of breath, and asthma (Table E-4).

            Table E-3. The prevalence of respiratory symptoms and disease in urea
            fertilizer workers exposed to ammonia
Respiratory
symptom/disease
Cough
Wheezing
Phlegm
Dyspnea
Chronic bronchitis
Bronchial asthma
Chronic bronchitis and
bronchial asthma
Relative risk (95% Cl)
Exposure category
ACGIHTLV
(18 mg/m3)
(n = 17)
3.48 (1.84-6.57)
5.01 (2.38-10.57)
3.75(1.97-7.11)
4.57 (2.37-8.81)
2.32 (0.31-17.28)
4.32 (2.08-8.98)
6.96 (0.76-63.47)
CACa(mg/m3-yrs)
<50
(n = 130)
0.72 (0.38-1.35)
1.86 (1.04-3.32)
0.63 (0.31-1.26)
1.19 (0.66-2.17)
0.61 (0.13-2.77)
1.22 (0.66-2.28)
1.82 (0.31-10.77)
>50
(n = 30)
2.82 (1.58-5.03)
5.24 (2.85-9.52)
3.03 (1.69-5.45)
2.59 (1.25-5.36)
5.32 (1.72-16.08)
2.44(1.10-5.43)
8.38(1.37-45.4)
     a = one missing value

     Source: Ballal et al. (1998).
10
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                              E-18            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
            Table E-4. Logistic regression analysis of the relationship between ammonia
            concentration and respiratory symptoms or disease in exposed urea fertilizer
            workers
Respiratory symptom/disease
Cough
Phlegm
Shortness of breath with wheezing
Wheezing alone
Dyspnea on effort
Diagnosis of asthma
OR (95% Cl)
1.32 (1.08-1.62)*
1.36 (1.10-1.67)*
1.26 (1.04-1.54)*
1.55 (1.17-2.06)*
0.83 (0.68-1.02)
1.33 (1.07-1.65)*
      *p < 0.05.
      OR = odds ratio
      Source: Ballal et al. (1998).
 1
 2   AM etal. (20011
 3          Results from limited spirometry testing of workers from factory A were reported in a
 4   followup study [Ali etal., 2001]. The lung function indices measured in 73 ammonia workers and
 5   348 control workers included FEVi and FVC. Prediction equations for these indices were developed
 6   for several nationalities (Saudis, Arabs, Indians, and other Asians), and corrected values were
 7   expressed as the percentage of the predicted value for age and height The FVC% predicted was
 8   higher in exposed workers than in controls (4.6% increase, p < 0.002); however, workers with
 9   cumulative exposure >50 mg/m3-years had significantly lower FEVi% predicted (7.4% decrease,
10   p < 0.006) and FVC% predicted (5.4% decrease, p < 0.030) than workers with cumulative exposure
11   <5 0 mg/m3-years. A comparison between symptomatic and asymptomatic exposed workers
12   showed that FEVi% predicted and FEVi/FVC% were significantly lower among symptomatic
13   workers (9.2% decrease in FEVi% predicted, p < 0.001, and 4.6% decrease in FEVi/FVC%,
14   p<0.02).
15
16   Rahman etal. (20071
17          Rahman etal.  (2007) conducted a cross-sectional  study of workers at a urea fertilizer
18   factory in Bangladesh that consisted of an ammonia plant and a urea plant The exposed group
19   consisted of 24 participants of the 63 operators in the ammonia plant and 64 participants of the 77
20   operators in the urea plant; 25 individuals from the administration building served as a control
21   group. Mean duration of employment exceeded 16 years  in all groups. Personal ammonia
22   exposures were measured by two different methods (Drager PAC III and Drager tube) in five to nine
23   exposed workers per day for 10 morning shifts in the urea plant (for a total of 64 workers) and in
24   five to nine exposed workers per day for 4 morning shifts from the ammonia plant (for a total of 24
25   workers). Four to seven volunteer workers per day were selected from the administration building
26   as controls, for a total of 25 workers over a 5-day period.  Questionnaires were administered to

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-19            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   inquire about demographics, past chronic respiratory disease, past and present occupational
 2   history, smoking status, respiratory symptoms (cough, chest tightness, runny nose, stuffy nose, and
 3   sneezing), and use of protective devices. Lung function tests (FVC, FEVi, and peak expiratory flow
 4   rate [PEFR]) were administered preshift and postshift (8-hour shifts) to the 88 exposed workers
 5   after exclusion of workers who had planned to have less than a 4-hour working day; lung function
 6   was not tested in the control group. Personal ammonia exposure and lung function were measured
 7   on the same shift for 28 exposed workers.  Linear multiple regression was used to analyze the
 8   relationship between workplace and the percentage cross-shift change in FEVi (AFEVi%) while
 9   adjusting for current smoking.
10          Mean exposure levels at the ammonia plant determined by the Drager tube and Drager PAC
11   III methods were 25.0 and 6.9 ppm (17.7 and 4.9 mg/m3), respectively; the corresponding means in
12   the urea plant were 124.6 and 26.1 ppm (88.1 and 18.5 mg/m3) (Rahman etal.. 2007). Although
13   the Drager tube  measurements indicated ammonia levels about 4-5 times higher than levels
14   measured with the PAC III instrument, there was a significant correlation between the ammonia
15   concentrations measured by the two methods (p = 0.001). No ammonia was detected in the control
16   area using the Drager tube (concentrations less than the measuring range of 2.5-200 ppm [1.8-
17   141 mg/m3]). The study authors observed that their measurements indicated only relative
18   differences in exposures between workers and production areas, and that the validity of the
19   exposure measures could not be evaluated based on their results. Based on an evaluation of the
20   two monitoring  methods and communication with technical support at Drager Safety Inc. (Bacom
21   and Yanosky, 2010), EPA considered the PAC III instrument to be a more sensitive monitoring
22   technology than the Drager tubes.  Therefore, the PAC III air measurements were considered the
23   more reliable measurement of exposure to ammonia for the Rahman etal. (2007) study.
24          The prevalence of respiratory irritation and decreased lung function was higher in the urea
25   plant than in the ammonia plant or in the administration building. Comparison between the urea
26   plant and the administration building showed that cough and chest tightness were statistically
27   higher in the former; a similar comparison of the ammonia plant and the administration building
28   showed no statistical difference in symptom prevalence between the two groups (Table E-5).
29   Preshift measurement of FVC, FEVi, and PEFR did not differ between urea plant and ammonia plant
30   workers.  Significant cross-shift reductions in FVC and FEVi were reported in the urea plant (2 and
31   3%, respectively, p < 0.05), but not in the ammonia plant.  When  controlled for current smoking,  a
32   significant decrease in AFEVi% was observed in the urea plant (p < 0.05). Among 23 workers with
33   concurrent measurements of ammonia and lung function on the same shift, ammonia exposure was
34   correlated with a cross-shift decline in FEVi of 3.9% per unit of log-transformed ammonia
35   concentration in ppm.  EPA identified a NOAEL of 4.9 mg/m3 and a LOAEL of 18.5 mg/m3 in the
36   Rahman etal. (2007) study based on increased prevalence of respiratory symptoms and a decrease
37   in lung function.
38
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-20           DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
            Table E-5.  Prevalence of respiratory symptoms and cross-shift changes in
            lung function among workers exposed to ammonia in a urea fertilizer factory
Parameter
Ammonia plant
(4.9 mg/m3)3
Urea plant
(18.5 mg/m3)3
Administration building
(concentration not
determined)13
Respiratory symptoms
Cough
Chest tightness
Stuffy nose
Runny nose
Sneeze
4/24 (17%)c
4/24 (17%)
3/24 (12%)
1/24 (4%)
0/24 (0%)
18/64 (28%)*
21/64 (33%)*
10/64 (16%)
10/64 (16%)
14/64 (22%)
2/25 (8%)
2/25 (8%)
1/25 (4%)
1/25 (4%)
2/25 (8%)
Lung function parameters (cross-shift percentage change)d'e
FVC
FEVi
PEFR
0.2 ±9.3
(Pre-shift: 3.308;
Post-shift: 3.332)
3.4 ±13.3
(Pre-shift: 2.627;
Postshift: 2.705)
2.9 ±11.1
(Pre-shift: 8.081;
Post-shift: 8.313)
-2.3 ±8.8
(Pre-shift: 3.362;
Post-shift: 3.258)
-1.4 ±8.9
(Pre-shift: 2.701;
Post-shift: 2.646)
-1.0 ±16.2
(Pre-shift: 7.805;
Post-shift: 7.810)
No data
No data
No data
     aMean ammonia concentrations measured by the Dra'ger PAC III method.
     bConcentrations in the administration building were rejected by study authors due to relatively large drift in the
     zero levels.
     "Values are presented as incidence (prevalence expressed as a percentage).
     Calculated as ([postshift - preshift]/preshift) x 100.
     eValues are presented as mean ± standard deviation (SD).

     *Statistically significant (p < 0.05) by Fisher's exact test, comparing exposed workers to administrators.
     Source: Rahman etal.  (2007).
 1
 2    Bhat and Ramaswamy (19931
 3        A cross-sectional study of workers exposed to fertilizer chemicals in a plant in Mangalore, India
 4    [Bhat and Ramaswamy, 1993] showed significant reduction in lung function parameters
 5    (PEFR/min and FEVi) compared to a control group. The exposed group consisted of 91 workers
 6    who underwent lung function testing, and included 30 urea plant workers, 30 diammonium
 7    phosphate (DAP) plant workers, and 31 ammonia plant workers. The controls were a group of 68
 8    people having comparable body surface area and were chosen from the same socioeconomic status
 9    and sex. All smokers were excluded from the study to avoid the effect of smoking on lung function.
10    Other workplace exposures were not assessed. The duration of exposure was dichotomized into
11    two groups (<10 and >10 years), but no exposure measurements were made.
12           Lung function parameters (FVC, FEVi, and PEFR/minute) were measured by a standard
13    spirometry protocol for all workers in the study, and the highest of three replicates were retained

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                                E-21           DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1   for calculation. A comparison of FVC, FEVi, and PEFR/minute was made between controls and
 2   fertilizer workers as a whole and also between controls and urea workers, DAP workers, and
 3   ammonia workers individually. The ammonia plant workers showed a significant decrease in FEVi
 4   (p < 0.05) and PERF/minute (p < 0.001) when compared to controls, but no significant decrease in
 5   FVC (Table E-6).  PEFR/minute, a measure of airflow in the bronchi, was reduced in all plant
 6   workers (urea, DAP, and ammonia), indicating that these fertilizer chemicals affected the larger
 7   airways. The reduction of FEVi, a measure of the amount of air that can be exhaled in 1 second, in
 8   ammonia plant workers suggested that ammonia can enter into the smaller bronchioles and cause
 9   bronchospasm. NOAEL and LOAEL values were not identified by the authors of this study or by
10   EPA due to the lack of exposure concentration measurements in this study.
11
             Table E-6. Comparison of lung function parameters in ammonia plant
             workers with controls
Parameter
FVC
FEVi
PEFR/min
Controls (n = 68)
(mean ± standard error)
3.43 ±0.21
2.84 ±0.10
383.3 ±7.6
Ammonia Plant (n = 31)
(mean ± standard error)
3. 19 ±0.07
2.52 ±0.1*
314 ±19.9**
      * Significantly different from controls (p < 0.05); paired t-test.
      **Significantly different from controls (p < 0.001); paired t-test.
      Source: Bhat and Ramaswamy [1993].
12
13    E.2.2. Studies in Livestock Farmers Exposed to Inhaled Ammonia
14          Several studies have investigated respiratory health and other outcomes in livestock
15    farmers exposed to ammonia. These and other studies have also demonstrated respiratory effects
16    associated with exposure to other constituents in farm worker air (e.g., respirable dust, endotoxin).
17    Ammonia exposure was associated with a decrease in lung function measures in five of the seven
18    studies (Mons6etal..20Q4: Donhametal.. 2000: Reynolds etal.. 1996: Donhametal.. 1995: Preller
19    etal.. 1995: Zeida etal.. 1994: Heederik etal.. 1990) examining this outcome (Table E-7). These five
20    studies controlled for co-exposures (e.g., endotoxin, dust, disinfectants) (Reynolds etal., 1996:
21    Donhametal., 1995: Preller etal.,  1995], noted only weak correlations (i.e., Spearman r < 0.20)
22    between ammonia and dust or endotoxin (Donhametal.. 2000], or observed associations with
23    ammonia but not with endotoxin or dust measures (Heederik etal.. 1990). and are the studies EPA
24    considered to be methodologically strongest (see Literature Search Strategy | Study Selection and
25    Evaluation section). In summary, this set of studies provides relatively consistent evidence of an
26    association between ammonia exposure and reduced lung function among livestock farmers,
27    accounting for endotoxin and dust.
28          Some of these farm worker studies also included analyses of respiratory outcomes in
29    relation to exposure, based on ammonia measurements. The studies analyzing prevalence of

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-22            DRAFT—DO NOT CITE OR QUOTE

-------
                                                             Supplemental Information—Ammonia
1

2
o
J

4

5

6

7
respiratory symptoms (including cough, phlegm, wheezing, chest tightness, and eye, nasal, and
throat irritation) in relation to ammonia provide generally negative results [Melbostad and Eduard.
2001: Preller etal.. 1995: Zejdaetal.. 1994). Two other studies reported an increased prevalence of
respiratory symptoms in pig farmers [Choudatetal., 1994: Crook etal., 1991].  The authors of these
studies measured air ammonia, but did not include a direct analysis of respiratory symptoms in
relation to ammonia (Table E-8).


        Table E-7. Evidence pertaining to respiratory effects in humans in relation to
        ammonia exposure in livestock farmers
                  Study design and reference
                                                                      Results
     Lung function
     Monso et al. (2004)
     105 never-smoking farmers (84 males, 21 females) working
     inside animal confinement buildings; sampled from the
     European Farmers' Study; mean age 45 yrs
     Exposure: Area samples (confinement building, morning)
                            Median
      ammonia              10 ppm (7 mg/m3)
      total dust              5.6 mg/m3
      total endotoxin         687.1 units/m3
     Outcome: Lung function (standard spirometry, before and
     after shift; chronic obstructive pulmonary disease (COPD)
     defined as FEVi <70 (n = 18; 17%).
                                                    COPD, Odds ratio (95% Cl), by quartile of
                                                    ammonia (1st and 2nd groups = referent)
                                                                   OR     (95%CI)
                                                    0 to 10        1.0      (referent)
                                                    >10-17        0.73     (0.17,3.20)
                                                    >17-60        1.32     (0.34,5.12)
                                                    Adjusted for age, gender, types of farming
                                                   Monso et al. (2004)
     Donham et al. (2000) (United States, Iowa)
     257 poultry workers (30% women, 70% men); 150 controls
     (42% women, 58% men; postal workers and electronics
     plant)
     Exposure: Personal samples (workshift)
                                    Mean
                           18.4 ppm (13 mg/m3)
                                                    OR (95%CI) for 3% or greater cross-shift
                                                    decline in FEN/i, by quartile of ammonia
                                                     ppm         OR       (95%CI)
 ammonia
 total dust
 respirable dust
 total endotoxin
 respirable endotoxin
Outcome: Lung function (standard spirometry, before and
after work shift)
                                                    >0 to <5
                                                    5 to <12
                                                    12 to <25
                                                    >25
               1.88
               1.93
               4.25
               2.45
(0.68,5.14)
(0.72,5.17)
(1.60, 11.2)
(0.88, 6.85)
                           6.5 mg/m
                           0.63 mg/m3
                           1,589 EU/m3(0.16 ng/m3)
                           58.9 EU/m3 (0.006 Hg/m3)
Adjusted for age, years worked in poultry industry,
gender, smoking status, education.
In linear regression, ammonia was statistically
significant predictor of 5% decline in FEF2s-75 (p =
0.045; Beta not reported)
Correlations between ammonia and other exposures
relatively weak (Spearman r < 0.20).
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                                  E-23            DRAFT—DO NOT CITE OR QUOTE

-------
                                                           Supplemental Information—Ammonia

        Table E-7. Evidence pertaining to respiratory effects in humans in relation to
        ammonia exposure in livestock  farmers
              Study design and reference
                                                  Results
Reynolds et al. (1996) (United States, Iowa)
151 men >18 yrs of age employed at swine farms and spent
time in swine confinement buildings (mean years of
employment = 12.4); a farm comparison group
(nonconfinement production) was included (number not
given). Follow-up study of Donham et al. (1995).
Exposure: Personal samples (workshift)
                     Geometric Mean (Time 2)
 ammonia                5.15 ppm (4 mg/m3)
 total dust                3.45 mg/m3
 respirable dust           0.26 mg/m3
 total endotoxin           176.12 EU/m3
 respirable                11.86  EU/m3
 endotoxin
 Ammonia levels similar at time 1 (5.65 ppm), but total
 dust and respirable dust higher at time 1 than time 2
Outcome: Lung function (standard spirometry, before and
after work shift at two times, two years apart (same season)
                              Correlation between cross-shift decline in FEVj and
                              ammonia: Spearman r = 0.18 (p < 0.05); strongest for
                              0-6 and 10-13 yrs duration
                              Predictive model relating ammonia to cross-shift
                              change in FEVi developed at baseline was
                              corroborated by Time 2 data; dust and endotoxin did
                              not add to the  significance of ammonia as predictor
Donham et al. (1995) (United States, Iowa)
201 men >18 yrs of age employed at swine farms and spent
time in swine confinement buildings (mean years of
employment = 9.6); a farm comparison group
(nonconfinement production) was included (number not
given)
Exposure: Personal samples
                        Geometric Mean
 ammonia
 total dust
 respirable dust
 total endotoxin
 respirable endotoxin        16.59 EU/m"*
Outcome: Lung function (standard spirometry, before shift
and then after a minimum of 2 hrs of exposure)
    5.64 ppm (4 mg/m3
    4.53 mg/m3
    0.23 mg/m3
    202.35 EU/m3
Ammonia was significant predictor of cross-shift
decline in lung function (included with age, duration,
smoking, total dust, respirable dust, and total
endotoxin in the models, as well as interaction terms)
Positive correlations were associated with changes in
lung function and exposure to total dust, respirable
dust, respirable endotoxin, and ammonia; dust was
related to all lung function measures; ammonia
results more variable across measures and duration
strata—strongest for 7-9 yrs duration); exposure to
ammonia concentrations of >7.5 ppm (5 mg/m3)
were predictive of a >3% decrease in FEVi
Heederiketal. (1990) (Nethelands)
27 pig farmers (mean age of 29 yrs; 43% current smokers)
Exposure: Area samples, used in conjunction with duration
of specific tasks to calculate an individual exposure measure
                        Mean
 ammonia
 total dust
 total endotoxin         24 ng/m;
Outcome: Lung function (standard spirometry, before and
after work shift, taken on Monday, Tuesday, and Friday)
                               Change (ml) in cross-shift lung function per
                               5 mg/m3 increase in ammonia
                                        Beta (SE)      (p-value)
                               FVC
5.6 mg/m
1.57 mg/m3
       3
  -3  (35)
-112 (38)
-330(131)
-170 (335)
-505 (300)
-404 (215)
-70 (179)
(< 0.05)
(< 0.05)
 MMEF
 PEF
 MEF75    -505 (300)     (< 0.05)
 MEF50    -404 (215)     (< 0.05)
 MEF25
Results from Tuesday measures presented;
other days reported to be similar patterns
but not as strong
No association between dust or endotoxins with the
lung function variables.
             This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-24             DRAFT—DO NOT CITE OR QUOTE

-------
                                                           Supplemental Information—Ammonia

        Table E-7. Evidence pertaining to respiratory effects in humans in relation to
        ammonia exposure in livestock farmers
              Study design and reference
                                                                         Results
Lung function and respiratory symptoms
Preller et al. (1995) (Netherlands)
194 swine farmers (94 with chronic respiratory symptoms,
100 without symptoms); 106 with complete data for lung
function analysis.
Exposure: Personal samples (two workshifts; winter and
summer)
                       Mean
 ammonia             2 mg/m3
 total dust             2.7 mg/m3
 total endotoxin        112 ng/m3
 Long-term average exposure derived based on measured
 values and model based on farm characteristics and tasks
Outcome: Lung function (standard spirometry, single
measure); standardized questionnaire for respiratory
symptoms
                                                      Association between ammonia and lung
                                                      function (n = 106)
                                                                  Beta   (SE)      (p-value)
                                                      FVC (I)
                                                      FEVi (I)
                                                      MMEF(l/s)
                                                      PEF (l/s)
-0.05
-0.27
-0.68
-0.77
(0.13)
(0.13)
(0.23)
(0.43)
(0.36)
(0.022)
(0002)
(0.039)
                                                      Adjusted for age, height, smoking,
                                                      endotoxin, disinfection variables
                                                     Stronger patterns seen in symptomatic group (n =
                                                     55).
                                                     No association with respiratory symptoms (chronic
                                                     cough, chronic phlegm, wheezing, shortness of
                                                     breath, chest tightness)
Zeida et al. (1994)
54 male swine producers (mean age = 36.3 yrs; mean
duration of employment = 10.7 yrs)
Exposure: Area samples
                       Mean
 ammonia             11.3 ppm (8 mg/m3)
 total dust             2.93 mg/m3
 respirable dust        0.13 mg/m3
 total endotoxin        11,332 units/m3
Exposure measures categorized into tertiles (cut-points
10.2 and 12.7 ppm) for some analyses.
Outcome: Lung function (standard spirometry, single
measure); respiratory symptoms-based on standardized
questionnaire (cough, phlegm, chest wheeze, chest
tightness)
                                                      Correlation coefficients (Spearman r) with
                                                      ammonia
                                                                                 with hr/day
                                                                                 interaction
                                                      FVC (% predicted)     0.18        -0.13
                                                      FEVi (% predicted)    0.18        -0.16
                                                      FEVi/FVC            0.00        -0.06
                                                      FEF (% predicted)     0.08        -0.09
                                                      Adjusted for age, height, and smoking

                                                     Some symptoms associated with ammonia
                                                     exposure—hours/day interaction but it is difficult to
                                                     distinguish these effects from the other exposures
                                                     and interactions in the analyses (particularly
                                                     endotoxin)
Respiratory symptoms (without lung function measures)
Melbostad and Eduard (2001)
Survey of 8,482 farmers and spouses; exposure study
conducted in 102 farmers
Exposure: personal samples
                        Range
                       0 to 8.2 ppm (0-6 mg/m3)
 ammonia
 total dust
 total endotoxin
 fungal spores
 bacteria
Outcome: Respiratory symptoms (standard questionnaire);
eye, nose, and throat irritation, cough, chest tightness, and
wheezing.
                       0.4-5.1 mg/m
                       500-28000 EU/m3
                       0.02-2.0 106/m3
                       0.2-48 106/m3
                                                     Negative correlation (r = -0.64) with total symptom
                                                     prevalence
ED = endotoxin unit (10 EU/ng)
             This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-25             DRAFT—DO NOT CITE OR QUOTE

-------
                                                     Supplemental Information—Ammonia
       Table E-8. Studies of respiratory effects in livestock farmers without direct
       analysis of ammonia exposure
Subjects
29 farm workers;
48 electronic factory
workers (controls)














102 pig farmers (mean
age 39.7 yrs; mean
duration of
employment of
15.7 yrs) who worked
at least half-time in a
swine confinement
building; 51 male dairy
farmers (mean age
40.1 yrs; mean
duration of
employment of
20.3 yrs) and 81 male
dairy industry workers
(controls; mean age
38.5 yrs; mean
duration of
employment of
15.7 yrs)
The use of nonpig
farmers as a reference
group is debatable
since they may be
exposed to various
airborne contaminants
Methods
20 pig houses were
monitored for dust and
ammonia concentrations;
respiratory symptoms
were determined by
questionnaire; lung
function tests were
performed; 24 subjects
provided blood samples to
determine IgE and IgG
antibody levels
No mention of controlling
for dust and other
airborne contaminant
exposures in the statistical
evaluation of ammonia

Lung function tests were
given to subjects before
and after a methacholine
challenge; respiratory
symptoms were
determined by
questionnaire

Co-exposures to other
airborne contaminants not
controlled for














Exposure conditions
Mean airborne
ammonia
concentrations ranged
from 1.5 to 13.23 ppm
(1-9 mg/m3) and mean
dust concentrations
ranged from
approximately 2 to
21 mg/m3. Mean
concentrations of
airborne dust and
ammonia increased
significantly in winter
due to restricted
ventilation


Mean total dust level of
2.41 mg/m3; mean
airborne ammonia
concentration of
8.5 mg/m3; mean
personal ammonia
exposure of
3.23 mg/m3; carbon
dioxide— range of 1,000
to 5,000 ppm















Results
Respiratory symptoms
included chest tightness,
wheeze, nasal and eye
irritation (23/29 farm
workers); 3/29 farm
workers had impaired
lung function
(decreased FEN/! and
FVC); 3 farmers had IgE
antibodies to pig
squames or urine;
specific IgG antibodies
were found in
14 workers to pig
squames, and 9 to pig
urine, suggesting an
allergic response
Pig and dairy farmers
had higher prevalence
of reported cough and
morning phlegm;
bronchial
hyperreactivity to
methacholine was
higher for pig and dairy
farmers compared to
controls















Reference
Crook et al.
(1991)















Choudatet
al. (1994)























ED = endotoxin unit (10 EU/ng); MMEF = mean midexpiratory flow; COPD = chronic obstructive pulmonary disease.
            This document is a draft for review purposes only and does not constitute Agency policy.
                                           E-26           DRAFT—DO NOT CITE OR QUOTE

-------
                                                               Supplemental Information—Ammonia
1

2
o
J

4

5
E.2.3.  Controlled Human Inhalation Exposure Studies
        Controlled exposure studies conducted with volunteers to evaluate irritation effects and
changes in lung function following acute inhalation exposure to ammonia are summarized in
Table E-9.


        Table E-9. Controlled human exposure studies of ammonia inhalation
           Subjects
                        Exposure conditions
               Results
   Reference
     25 healthy volunteers
     (mean age 29.7yrs),
     and 15
     mild/moderate
     persistent asthmatic
     volunteers (mean age
     29.1 yrs)
                    2-500 ppm (1-354 mg/m )
                    (ocular and nasal exposure)
                    for various durations lasting
                    up to 2.5 hrs; baseline lung
                    function was recorded prior
                    to exposure
Irritation threshold, odor intensity, and
annoyance were not significantly different
between healthy volunteers and
asthmatics; nasal irritation threshold = 129
ppm (91 mg/m3); ocular irritation
threshold = 175 ppm (124 mg/m3); there
were no changes in lung function
for subjects in either group
Petrova et al.
     24 healthy female
     volunteers age 18-45
     yrs (mean age 29.9
     yrs)
                    0.03-615.38 ppm (0.02-
                    435 mg/m3) (nasal exposure)
                    for a maximum of 2 sec; pre-
                    exposure measurements
                    included rhinoscopic exam,
                    screening for chemical
                    sensitivities, allergies,
                    respiratory disease, general
                    health, and  prior chemical
                    exposure by personal
                    interview
Both the static and dynamic methods
showed similar averages for detection
thresholds for the odor and irritancy of
ammonia; mean odor detection threshold
of 2.6 ppm (2 mg/m3) (both static and
dynamic) and mean irritation thresholds of
31.7 or 60.9 ppm (22 or 43 mg/m3) for
static and dynamic methods, respectively
Smeets et al.
     43 healthy male
     volunteers age 21-47
     yrs; one group of 30
     men not familiar with
     the smell of ammonia
     and 10 men exposed
     to ammonia regularly
     at the workplace
                    0,10, 20, 20+ 2 peak
                    exposures at 40 and 50 ppm
                    (0,7,14,14+2 peak
                    exposures at 28 and
                    35 mg/m3) on  5 consecutive
                    days for 4 hrs/d in an
                    exposure chamber
Subjects familiar to ammonia reported
fewer symptoms than na'i've subjects; at
concentrations <14 mg/m3, there were no
significant differences in symptoms
reported between the groups; the
perceived intensity of symptoms was
concentration-dependent in both groups
Ihrig et al.
     12 healthy volunteers
     (7 females, 5 males)
     21-28 yrs old
                    5 and 25 ppm (4 and
                    18 mg/m3) for three separate
                    exposures in inhalation
                    chamber for 1.5 hrs resting
                    and 1.5 hrs exercising on a
                    stationary bike; 1-4
                    volunteers were exposed on
                    each occasion; lung function
                    and nasal lavage were
                    performed before and after
                    exposure
Reported discomfort in eyes, detection of
solvent smell, headache, dizziness, and
feeling of intoxication were significantly
increased at 4 mg/m3; there were no
changes in lung function  or exhaled nitric
oxide levels in exposed individuals;
exposure did not result in upper-airway
inflammation or bronchial responsiveness
Sundblad et al.
                  This document is a draft for review purposes only and does not constitute Agency policy.
                                                    E-27             DRAFT—DO NOT CITE OR QUOTE

-------
                                                           Supplemental Information—Ammonia

       Table E-9. Controlled human exposure studies of ammonia inhalation
      Subjects
    Exposure conditions
                Results
   Reference
Six healthy volunteers
(two males and four
females, 25-45 yrs
old) and eight
volunteers with mild
asthma (four males
and four females, 18-
52 yrs old)
16-25 ppm (11-18 mg/m )
for 30-min sessions with 1 wk
between sessions; lung
function was measured
before and after exposure
No significant changes in lung function in
healthy subjects at any concentration;
decreased FEVi and increased bronchial
hyperreactivity were reported in
asthmatics exposed to dust and ammonia,
but not to ammonia alone; exposure to
dust alone caused similar effects,
suggesting that dust was responsible for
the effects
Sigurdarson et
Eight healthy male
volunteers (23-28 yrs
old)
Exposed for 4 hrs at 1-wk
intervals to swine
confinement buildings; mean
airborne ammonia
concentration of 20.7 ppm
(15 mg/m3); also exposed to
airborne dust, bacteria,
endotoxin, and  molds
Decreased expiratory flows
increased neutrophils in the nasal wash,
and increased white blood cell count

The relationship between environmental
and human variables was evaluated. The
only significant correlation (r = 0.74; p <
0.04) was between ammonia and
interleukin. Thus, changes in lung function
may not be caused by ammonia exposure
only.
Cormier et al.
 2000)
Unspecified number
of volunteer subjects
Acute exposure up to 15 sec,
1 time/d at unspecified
concentrations; also a
separate exposure of
10 inhaled breaths via
mouthpiece at unspecified
concentrations; there was no
mention of pre-exposure
examinations
The lachrymatory threshold was 55 ppm
(39 mg/m3) and bronchoconstriction was
seen at 85 ppm (60.1 mg/m3)
Douglas and
Coe (1987)a
18 healthy
servicemen
volunteers, 18-39 yrs
old
50-344 mg/m for a half-day
(session day 2); sessions on
days land 3 acted as
controls; all volunteers
underwent a preliminary
examination prior to exposure
No effect at 71 mg/m ; reduced expiratory
minute volume at concentrations ranging
from 106 to 235  mg/m3 compared to
controls (not dose-dependent); exercise
tidal volume was increased at 106 mg/m3,
but reduced at higher concentrations in a
dose-dependent manner
Cole et al.
Six male and female
volunteers, 24-46 yrs
old
25, 50, and 100 ppm (18, 35,
and 71 mg/m3) ammonia for
6 hrs/d, 1 time/wk over
6 wks; occasional brief
exposure to 150-200 ppm
(106-141 mg/m3); there was
no mention of pre-exposure
examinations
Habituation to eye, nose, and throat
irritation after 2-3 wks with short-term
adaption; there were no significant
differences for common  biological
indicators, physical exams, or in normal
job performance when compared to
control subjects; continuous exposure to
71 mg/m3 became easily tolerated and
had no effect on general health after
acclimation occurred;  brief exposure to
106-141 mg/m3 produced lacrimation and
transient discomfort
Ferguson et al.
(1977)a
             This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-28             DRAFT—DO NOT CITE OR QUOTE

-------
                                                             Supplemental Information—Ammonia

             Table E-9. Controlled human exposure studies of ammonia inhalation
           Subjects
    Exposure conditions
               Results
   Reference
      15 volunteers, 18-53
      yrs old
50, 80,110, and 140 ppm (35,
57, 78, and 99 mg/m3) for 2
hrs in an exposure chamber;
there was no mention of
preexposure examinations
No effect on vital capacity or
99 mg/m3 caused severe irritation and
could not be tolerated; reported eye
irritation increased with concentration
Verberk (1977)a
      Seven male
      volunteers with an
      average age of 31 yrs
30, 50, and 90 ppm (21, 35,
and 64 mg/m3) for 10 min in
an inhalation chamber;
physical and neurological
examinations were conducted
prior to exposure
Increased eye erythema at 64 mg/m
compared to 21 and 35 mg/m3 exposure;
64 mg/m3 did not produce significant
bronchiospasm or severe lacrimation;
intensity of odor perception was reported
as higher at 21 and 35 mg/m3 than at
64 mg/m3
MacEwen et al.
      Seven male
      volunteers
500 ppm (354 mg/m ) for
30 min from masked
breathing apparatus for nose
and throat inhalation; there
was no mention of
preexposure examinations
Hyperventilation (50-250% increase above
controls) characterized by increased
breathing rate and expiratory minute
volume (i.e., volume of air exhaled in 1
min); no coughing was induced, excessive
lacrimation occurred in two subjects; two
subjects reported nose and throat
irritation that lasted 24 hrs after exposure;
no changes were reported in nitrogen
metabolism or in blood or urine urea,
ammonia, or nonprotein nitrogen
Silverman et al.
 1949 )a
      aThis controlled-exposure study did not provide information on the human subjects research ethics procedures
      undertaken in the study; however, there is no evidence that the conduct of the research was fundamentally
      unethical or significantly deficient relative to the ethical standards prevailing at the time the research was
      conducted.
      blnvestigators reported the use of ethical standards involving informed consent by volunteers and/or study
      approval by an Institutional Review Board or other ethics committee.
 1
 2           Twelve healthy volunteers exposed to 4 and 18 mg/m3 ammonia on three different
 3    occasions for 1.5 hours in an exposure chamber while exercising on a stationary bike reported
 4    discomfort in the eyes and odor detection at 4 mg/m3 [Sundbladetal., 2004]. Eye irritation was
 5    also shown to increase  in a concentration-dependent manner in 15 volunteers exposed to ammonia
 6    for 2 hours in an exposure chamber at concentrations of 35, 57, 78, and 99 mg/m3; ammonia
 7    concentrations of 99  mg/m3 caused severe and intolerable irritation [Verberk. 1977). The
 8    lachrymatory threshold was determined to be 39 mg/m3 in volunteers  exposed to ammonia gas
 9    inside tight-fitting goggles for an acute duration of up to 15 seconds [Douglas and Coe. 1987). In
10    contrast,  exposures to up to 64 mg/m3 ammonia gas did not produce severe lacrimation in seven
11    volunteers after 10 minutes in an exposure chamber, although increased eye erythema was
12    reported [MacEwen et al., 1970].  Exposure to 354 mg/m3 of ammonia gas for 30 minutes through a
13    masked nose and throat inhalation apparatus resulted in two of seven volunteers reporting
14    lacrimation and two of seven reporting nose and throat irritation that lasted up to 24 hours after
                  This document is a draft for review purposes only and does not constitute Agency policy.
                                                  E-29            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   exposure [Silverman et al., 1949].
 2          Petrovaetal. [2008] investigated irritation threshold differences between 25 healthy
 3   volunteers and 15 mild-to-moderate persistent asthmatic volunteers exposed to ammonia via the
 4   eyes and nose at concentrations of 1-354 mg/m3 for durations lasting up to 2.5 hours.  Irritation
 5   threshold, odor intensity, and annoyance were not significantly different between the two groups.
 6   The nasal and eye irritation thresholds were reported to be 91 and 124 mg/m3, respectively.
 7   Smeets etal. [2007] investigated odor and irritation thresholds for ammonia vapor in 24 healthy
 8   female volunteers at concentrations of 0.02-435 mg/m3. This study found a mean odor detection
 9   threshold of 2 mg/m3 and a mean irritation threshold of 2 2 or 43 mg/m3, depending on the
10   olfactometry methodology followed (static versus dynamic, respectively]. Irritation thresholds may
11   be higher in people who have had prior experience with ammonia exposure [Ihrigetal., 2006].
12   Thirty male volunteers who had not experienced the smell of ammonia and 10 male volunteers who
13   had regular workplace exposure to ammonia were exposed to ammonia vapors at concentrations of
14   0, 7,14, and 35 mg/m3 on 5 consecutive days (4 hours/day] in an exposure chamber; an additional
15   group was exposed to 14 mg/m3 plus two peak exposures to 28 mg/m3 for 30 minutes.  Volunteers
16   in the group familiar to the smell of ammonia reported fewer symptoms than the nonhabituated
17   group, but at a concentration of 14 mg/m3, there were no differences in perceived symptoms
18   between the groups. However, the perceived intensity of symptoms was concentration-dependent
19   in both groups, but was only significant in the group of volunteers not familiar with ammonia
20   exposure [Ihrigetal., 2006]. Ferguson et al. [1977] reported habituation to eye, nose, and throat
21   irritation in six male and female volunteers after 2-3 weeks of exposure to ammonia concentrations
22   of 18, 35, and 71 mg/m3 during a 6-week study (6 hours/day, 1 time/week]. Continuous exposure
23   to even the highest concentration tested became easily tolerated with no general health effects
24   occurring after acclimation.
25          Several studies evaluated lung functions following acute inhalation exposure to ammonia.
26   Volunteers exposed to ammonia (lung only] through a mouthpiece for 10 inhaled breaths of gas
27   experienced bronchioconstriction at a concentration of 60 mg/m3 [Douglas and Coe, 1987]:
28   however, there were no bronchial symptoms reported in seven volunteers exposed to ammonia at
29   concentrations of 21, 35, and 64 mg/m3 for 10 minutes in an exposure chamber [MacEwen etal.,
30   1970]. Similarly, 12 healthy volunteers exposed to ammonia on three separate occasions to 4 and
31   18 mg/m3 for 1.5 hours in an exposure chamber while exercising on a stationary bike did not have
32   changes in bronchial responsiveness, upper airway inflammation, exhaled nitric oxide levels, or
33   lung function as measured by vital capacity and FEVi [Sundbladetal., 2004].  In another study,
34   18 healthy servicemen volunteers were placed in an exposure chamber for 3 consecutive half-day
35   sessions. Exposure to ammonia at concentrations of 50-344 mg/m3 occurred on the second
36   session, with sessions 1 and 3 acting as controls [Cole etal., 1977]. The no-effect concentration was
37   determined to be 71 mg/m3. Exercise tidal volume was increased at 106 mg/m3, but then
38   decreased at higher concentrations in a concentration-dependent manner [Cole etal.. 1977].
39   Decreased FEVi and FVC were reported in eight healthy male volunteers exposed to a mean
40   airborne ammonia concentration of 15 mg/m3 in swine confinement buildings for 4 hours at

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-30           DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   1-week intervals; however, swine confinement buildings also include confounding exposures to
 2   dust, bacteria, endotoxin, and molds, thereby making measurement of effects due to ammonia
 3   uncertain in this study [Cormier et al.. 2000).
 4          Differences in lung function between healthy and asthmatic volunteers exposed to ammonia
 5   were evaluated in several studies. There were no changes in lung function as measured by FEVi in
 6   25 healthy volunteers and 15 mild/moderate persistent asthmatic volunteers after ocular and nasal
 7   exposure to 1-354 mg/m3 ammonia at durations lasting up to 2.5 hours [Petrova et al., 2008].  In
 8   another study, six healthy volunteers and eight mildly asthmatic volunteers were exposed to 11-
 9   18 mg/m3 ammonia, ammonia and dust, and dust alone for 30-minute sessions, with 1 week
10   between sessions [Sigurdarsonetal.. 2004). There were no significant changes in lung function as
11   measured by FEVi in the healthy volunteers for any exposure. A decrease in FEVi was reported in
12   asthmatics  exposed to dust and ammonia, but not to ammonia alone; similarly, increased bronchial
13   hyperreactivity was reported in asthmatics after exposure to dust and ammonia, but not to
14   ammonia alone. Exposure to dust alone caused similar effects, suggesting that dust was responsible
15   for decreased lung function [Sigurdarsonetal., 2004].
16          In summary, volunteer studies demonstrate that eye irritation can occur following acute
17   exposure to ammonia at concentrations as low as 4 mg/m3. Irritation thresholds may be higher in
18   people who have had prior experience with ammonia exposure, and habituation to  eye, nose, and
19   throat irritation occurs over time. Lung function was not affected in workers acutely exposed to
20   ammonia concentrations as high as 71 mg/m3. Studies comparing the lung function of asthmatics
21   and healthy volunteers exposed to ammonia do not suggest that asthmatics are more  sensitive to
22   the lung effects of ammonia.
23
24   E.2.4. Case Reports of Human Exposure to Ammonia
25          Oral exposure to ammonia most commonly involved ingestion of household cleaning
26   solutions or biting into the capsules of ammonia smelling salts, which are commonly found in first
27   aid kits. Young children, generally <4 years old, have been reported as "biting into" or ingesting
28   smelling salts capsules.  The acute effects included drooling, erythematous and edematous lips,
29   reddened and blistered tongues, dysphagia, vomiting, and oropharyngeal burns [Robertson etal.,
30   2010: Rosenbaumetal.. 1998: Wasonetal.. 1990: Lopez etal.. 1988). Delayed effects were not
31   noted in these cases. Gilbert [1988] reported ammonia intoxication characterized by lethargy,
32   restlessness, irritability, and confusion in a 37-year-old man following surgery. Most other cases of
33   ammonia ingestion involved household cleaning solutions and detergents. Many cases were
34   intentional; however, not all were fatal.  Klein etal. [1985] described two cases of ingestion of
35   approximately 30 mL and "two gulps" of Parson's sudsy ammonia (ammonia 3.6%;  pH 11.5],
36   respectively. The first case resulted in a white and blistered tongue and pharynx, and esophageal
37   burns with  friable, boggy mucosa; and in the second case, several small esophageal lesions with
38   mild to moderate ulceration and some bleeding were reported.  There were no oropharyngeal
39   burns in the second case and no delayed complications in either case. Christesen [1995] reported
40   that of the 11 cases involving accidental or intentional ingestion of ammonia water by adults

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-31            DRAFT—DO NOT CITE  OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1   (>15 years old), 2 cases exhibited acute respiratory obstruction and 1 case developed an
 2   esophageal stricture 3 months postinjury. In cases involving fatalities, evidence of laryngeal and
 3   epiglottal edema, erythmatous esophagus with severe corrosive injury, and hemorrhagic esophago-
 4   gastro-duodeno-enteritis was noted [Klein etal., 1985: Klendshoj andRejent, 1966]. Dworkin et al.
 5   [2004] reported a case of ingestion of contaminated chicken tenders, prepared and served in a
 6   school cafeteria, by approximately 157 students and 6 teachers. The onset of acute symptoms
 7   occurred within an hour of ingestion, and included headache, nausea, vomiting, dizziness, diarrhea,
 8   and burning mouth. In a case of forced ingestion of an unknown quantity of dilute ammonia [Dilli et
 9   al.. 2005]. a 14-year-old boy presented with difficulty speaking, ataxic gait, isochoric pupils, and
10   evidence of brain edema. There were no burns to the eyes or mouth and no indication of gastric
11   pathology. It was only after the patient was able to communicate that ammonia was involved that
12   appropriate treatment, followed by a satisfactory outcome, was achieved. In general, these acute
13   gastrointestinal exposures produce effects that reflect the corrosive nature of ammonia. The
14   relevance of these acute effects to effects associated with chronic  low-level exposure to ammonia is
15   unclear.
16          Inhalation is the most frequently reported route  of exposure and cause of morbidity and
17   fatality, and often occurs in conjunction with dermal and ocular exposures.  Acute effects from
18   inhalation have been reported to range from mild to severe, with  mild symptoms consisting of nasal
19   and throat irritation, sometimes with perceived tightness in the throat [Price and Watts, 2008:
20   Prudhomme etal.. 1998: Weiser and Mackenroth. 1989:  Yang etal.. 1987: O'Kane. 1983: Ward etal..
21   1983: Caplin, 1941]. Moderate effects are described as moderate to severe pharyngitis;
22   tachycardia; frothy, often blood-stained sputum; moderate dyspnea; rapid, shallow breathing;
23   cyanosis; some vomiting; transient bronchospasm; edema and some evidence of burns to the lips
24   and oral mucosa; and localized to general rhonchi in the  lungs [Weiser and Mackenroth. 1989: Yang
25   etal.. 1987: O'Kane. 1983: Ward etal.. 1983: Couturier etal.. 1971: Caplin. 1941]. Severe effects
26   include second- and third-degree burns to the nasal passages, soft palate, posterior pharyngeal
27   wall, and larynx; upper airway obstruction; loss of consciousness; bronchospasm, dyspnea;
28   persistent, productive cough; bilateral diffuse rales and rhonchi; production of large amounts of
29   mucous; pulmonary edema; marked hypoxemia; local necrosis  of the lung; deterioration of the
30   whole lung; and fatality.  Delayed effects of acute exposure to high concentrations of ammonia
31   include bronchiectasis; bronchitis; bronchospasm/asthma; dyspnea upon exertion and chronic
32   productive cough; bronchiolitis; severe pulmonary insufficiency; and chronic obstructive
33   pulmonary disease [Lalicetal., 2009: Leducetal., 1992:  Bernstein and Bernstein, 1989: Fluryetal.,
34   1983: Ward etal.. 1983: Stroud. 1981: Close etal.. 1980:  Taplin etal.. 1976: Walton. 1973: Kass et
35   al.. 1972: Slot. 1938].
36          Respiratory effects were also observed following chronic occupational exposure to
37   ammonia. After 18 months and 1 year on the job, respectively,  two men developed cough, chest
38   tightness, and wheezing, typically after 2-6 hours from the beginning of each work day, but not on
39   weekends or holidays. In another case, progressive deterioration of the clinical condition of a
40   68-year-old male was documented for 4 years, and development of diffuse interstitial and severe

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-32             DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1   restrictive lung disease was reported following long-term repetitive occupational exposure to
 2   ammonia at or above the odor recognition level of 3-50 ppm [Brautbar etal., 2003]. Lee et al.
 3   [1993] reported a case of a 39-year-old man who developed occupational asthma 5 months after
 4   beginning a job requiring the polishing of silverware. The room in which he worked was poorly
 5   ventilated.  The product used contained ammonia and isopropyl alcohol and the measured
 6   ammonia concentration in the breathing zone when using this product was found to be 6-
 7   11 mg/m3.
 8          Acute dermal exposure to anhydrous (liquid] ammonia and ammonia vapor has resulted in
 9   caustic burns of varying degrees to the skin and eyes. There are numerous reports of exposures
10   from direct contact with anhydrous ammonia in which first-, second-, and third-degree burns
11   occurred over as much as 50% of the total body surface [Lalic etal., 2009: Pirjavec etal., 2009:
12   Arwood et al.. 1985]. Frostbite injury has also been reported in conjunction with exposure to
13   sudden decompression of liquefied ammonia, which is typically stored at -33°F [George etal.. 2000:
14   Sotiropoulos etal., 1998: Arwood etal., 1985]. However, direct contact is not a prerequisite for
15   burn injury. Several reports have indicated that burns to the skin occurred with exposure to
16   ammonia gas or vapor. Kass etal. [1972] reported one woman with chemical burns to her
17   abdomen, left knee, and forearm and another with burns to the feet when exposed to anhydrous
18   ammonia gas released from a derailed train in the vicinity. Several victims at or near the scene of
19   an overturned truck that had been carrying 8,000 gallons of anhydrous ammonia were reported as
20   having second- and third-degree burns over exposed portions of the body [Burns etal.. 1985: Close
21   etal., 1980: Hattonetal., 1979]. In a case involving a refrigeration leak in a poorly ventilated room,
22   workers located in an adjacent room reported a "burning skin" sensation [de la Hoz et al., 1996],
23   while in another case involving the sudden release  of ammonia from a pressure valve in a
24   refrigeration unit, one  victim received burns to the leg and genitalia [O'Kane. 1983].
25          In addition to the skin, the eyes are particularly vulnerable to ammonia burns due to the
26   highly water-soluble nature of the chemical and the ready dissociation of ammonium hydroxide to
27   release hydroxyl ions.  When ammonia or ammonia in solution has been splashed or sprayed into
28   the face (accidently or  intentionally], immediate effects include temporary blindness,
29   blepharospasm, conjunctivitis, corneal burns, ulceration, edema, chemosis, and loss of corneal
30   epithelium [George etal.. 2000: Helmers etal.. 1971: Highman. 1969: McGuiness. 1969: Levy etal..
31   1964: Abramovicz. 1925]. The long-term effects included photophobia, progressive loss of
32   sensation, formation of bilateral corneal opacities and cataracts, recurrent corneal ulcerations,
33   nonreactive pupil, and gradual loss of vision [Yang  etal.. 1987: Kass etal.. 1972: Helmers etal..
34   1971: Highman. 1969:  Osmond and Tallents. 1968: Levy etal.. 1964: Abramovicz. 1925]. White et
35   al. [2007] reported a case with acute bilateral corneal injury that developed into bilateral uveitis
36   with stromal vascularization and stromal haze and scarring, and pigmented keratic precipitates
37   that resulted in legal blindness.  An increase in intraocular pressure, resembling acute-angle closure
38   glaucoma, was reported by Highman [1969] following ammonia intentionally sprayed into the eyes
39   during robbery attempts.
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-33            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 i   E.3.  ANIMAL STUDIES
 2   E.3.1. Oral Exposure
 3   Hataetal. (1994)
 4          In a study designed to look at the effects of ammonia on gastric mucosa histology and cell
 5   kinetics, Hataetal. [1994] exposed groups of male Donryu rats (6 rats/group/time interval) to
 6   drinking water containing 0, 0.02, or 0.1% ammonia for durations up to 24 weeks. Based on an
 7   assumed body weight of 267 g and daily water intake of 37 mL (subchronic values for male
 8   Sprague-Dawley rat [U.S. EPA. 198811: the doses were estimated to be 0, 28, or 140 mg/kg-day.
 9   After 1, 3, and 5 days and 1, 4, 8,12, and 24 weeks from the start of exposure, the gastric mucosa in
10   the fundic gland region and the antrum was examined histologically. In addition, the labeling index
11   of gastric mucosal tissue was measured using either a double labeling technique with
12   bromodeoxyuridine (BrDU) and 3H-thymidine (weeks 8 and 24) or the flash labeling technique
13   with BrDU (other weeks).
14          A dose-related decrease in the height of the glandular ducts of the gastric mucosa was
15   observed in the fundic region (by week 4) and in the pyloric region (by week 8). There was a
16   decrease in periodic acid-Schiff (PAS)-positive mucus only in the early stages of ammonia exposure
17   (through day 3 of exposure). The labeling index in gastric mucosa glands was increased at earlier
18   time points (up to week 1 for fundic glands and to week 4 for pyloric glands), indicating enhanced
19   cell cycling subsequent to repeated erosion and repair; however, at later time points up to 24 weeks
20   of exposure, the labeling index was decreased, consistent with  reduced capability of the generative
21   cell zone of the mucosal region. The authors reported that there was no ammonia-induced gastritis
22   or ulceration. Based on histological changes in the gastric mucosa, EPA identified a LOAEL of 0.02%
23   ammonia in drinking water; a NOAEL was not identified.
24
25   Kawano etal. (19911: Tsujiietal. (19931
26          Kawano etal.  (1991) investigated the hypothesis that the bacterium Helicobacter pylori,
27   which produces a potent urease that increases ammonia production, plays a significant role in the
28   etiology of chronic atrophic gastritis.  Male Sprague-Dawley rats (6/group) were given tap water or
29   0.01 or 0.1% ammonia ad libitum  for 2 or 4 weeks. The daily dose of 0.01 and 0.1% ammonia in
30   drinking water, based on a weight of 230 g for male rats and a water consumption of 50 mL/day,
31   was estimated to be 22 and 220 mg/kg-day, respectively.  The effect of ammonia on the antral
32   mucosa was estimated by three measurements of the thickness of the mucosa about 175 um from
33   the pyloric ring in the antral mucosa.  The parietal cell number per gland was determined at three
34   locations in the oxyntic glandular area.
35          Mucosal lesions were not observed macro- or microscopically.  There was a statistically
36   significant decrease in mean antral mucosal thickness with increasing dose and duration of
37   exposure (Table E-10).  Parietal cell number per oxyntic gland  decreased in a statistically
38   significant dose- and time-dependent fashion. The index of PAS Alician blue positive intracellular
39   mucin was significantly lower in the antral and body mucosa with 0.1% ammonia; the index was
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-34            DRAFT—DO NOT CITE OR QUOTE

-------
                                                           Supplemental Information—Ammonia
 1
 2
 o
 J
 4
 5
significantly lower only for the antral mucosa with 0.01% ammonia. The authors suggested that
administration of ammonia in drinking water causes gastric mucosal atrophy.  Based on the
reduction in antral mucosal thickness, EPA identified a LOAEL of 22 mg/kg-day; a NOAEL was not
identified.
             Table E-10. Effect of ammonia in drinking water on the thickness of the
             gastric antral and body mucosa of the rat stomach
Length of treatment
Thickness of mucosa (urn); mean ± standard error of the mean
Control (tap water)
Percent ammonia in drinking water
0.01%
0.1%
Antral mucosa
2 wks
4 wks
270 ± 18
276 ± 39
258 ± 22
171 ±22*
217 ± 40*
109 ±12**'***
Body mucosa
2 wks
4 wks
574 ± 116
618 ± 154
568 ± 159
484 ± 123
591 ± 183
440 ±80*'***
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
*Statistically significant by Student's t-test; (p < 0.05) versus control group.
**Statistically significant by Student's t-test; (p < 0.01) versus control group.
***Statistically significant by Student's t-test; (p < 0.01) versus 2-week treatment group.
Source: Kawano etal. (1991).

       In a follow-up study of the effect of ammonia produced from H. pylori, Tsujiietal. [1993]
studied the subchronic effect of ammonia in drinking water on the cell kinetics of the gastric
mucosa of the stomach. Six groups of male Sprague-Dawley rats (36 rats/group) were given 0.01%
ammonia in drinking water for 3 days, or 1, 2, 4, or 8 weeks; ammonia solutions were changed
daily. Tap water was provided for the balance of the 8-week study. A control group was given tap
water for 8 weeks. Based on the initial body weight (150 g) and estimated daily water intake
(50 mL),  the daily dose at a drinking water concentration of 0.01% ammonia was estimated to be
33 mg/kg-day. Cellular migration was measured by labeling cells with BrDU at different time
periods and measuring the incorporation of this modified nucleoside  with a histochemical
technique using anti-BrDU monoclonal antibodies. Antral and body mucosa thickness was
measured as described in Kawano etal. (1991].  The measurement of cell proliferation in the
gastric mucosa was estimated using the labeling index in gastric pits (ratio of labeled nuclei to total
nuclei in  the proliferation zone].
       As in Kawano etal. (1991]. no mucosal lesions were found macroscopically or
microscopically. The antral mucosal thickness decreased significantly at 4 and 8 weeks of
treatment (Table E-l 1], but there was no effect on the body mucosa.  Cell migration preceded the
decrease  in thickness of the antral mucosa. The rate of cell migration (cells/day] toward the
mucosal surface was significantly greater for 0.01% ammonia-treated rats compared to the control
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                                E-35            DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1
 2
 o
 J
 4
 5
 6
at 4 and 8 weeks of treatment Cell proliferation, as estimated from the labeling index, was
significantly increased after 1 week for the antral and body mucosa. The authors concluded that
0.01% ammonia increased epithelial cell migration in the antrum leading to mucosal atrophy. EPA
identified a LOAEL of 33 mg/kg-day based on decreased thickness of the gastric antrum; a NOAEL
was not identified.
            Table E-ll. Effect of ammonia in drinking water on gastric antral and body
            mucosa in the stomach of Sprague-Dawley rats administered 0.01% ammonia
            in drinking water
Length of treatment
Control (tap water only)
3d
Iwk
2 wks
4 wks
8 wks
Thickness of mucosa (u.m)a
Antral mucosa
283 ± 26
305 ± 45
272 ± 31
299 ± 26
159 ± 29*
168 ± 26*
Body mucosa
534 ± 27
559 ± 50
542 ± 28
555 ± 37
531 ±32
508 ± 29
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Extracted from Figure 3 of Tsujii etal. (1993); mean ± SD.
*Statistically significant by Student's t-test. (p < 0.05) versus control (tap water only) group.
Source: Tsujii etal. (1993).

Fazekas (19391
       Fazekas [1939] administered ammonium hydroxide to 51 rabbits (strain and sex not
specified) via gavage every other day initially and, later, daily in increasing amounts of 50-80 mL as
either a 0.5 or 1.0% solution.  The exact duration of the study is not reported, but it is clear from the
data that by the end of the experiment, some rabbits received only three or four  doses before dying
as a result of intoxication in 5.5 days, and other rabbits received over 80 doses and survived for up
to 17 months. The daily dose (mg/kg-day) was estimated using the weight of adult rabbits from
standard growth curve for rabbits (3.5-4.1 kg) (U.S. EPA, 1988).  Based on a daily gavage volume of
50-80 mL, daily doses for the rabbits receiving 0.5 and 1.0% ammonia solutions were
approximately 61-110 and 120-230 mg/kg-day, respectively. Toxicological endpoints evaluated
included fluctuations in body weights, changes in blood pressure measured at the central artery of
the ear in 10 rabbits after lengthy treatment, and changes in the weight, fat, and  cholesterol content
of adrenals. For comparison purposes, the weight of the adrenals from 41 healthy rabbits of similar
age and body weight were also determined.  The average weight of adrenals from these 41 control
rabbits was 400.0 ± 13.4 mg.
       Fazekas (1939) reported that differences in mean adrenal weight in ammonium hydroxide-
treated animals were significant, although there was no description of the statistical analysis
performed in this study. Chemical evaluation of the adrenals from treated rabbits revealed fat and

            This document is a draft for review purposes only and does not constitute Agency policy.
                                          E-36            DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1   cholesterol content 4.5 and 6.5 times greater than controls, respectively. At the beginning of the
 2   experiment, a greater weight loss was observed among those rabbits receiving ammonium
 3   hydroxide more frequently (daily) at higher doses. Body weights fluctuated among treated rabbits
 4   and generally decreased initially and gradually increased in the later months, only to drop again a
 5   few weeks before death. Body weights for controls were not reported. Thirteen rabbits exhibited
 6   weight increases after the initial loss that persisted until the end of the experiment. Dissection of
 7   these rabbits revealed enlarged adrenals (800-1,340 mg) and fatty tissue surrounding the kidneys,
 8   mesentery, and pericardium. This fat accumulation was not observed in untreated controls.
 9   Histology revealed enlarged cells of the zona fasciculata of the adrenal cortex that were rich in lipid.
10   The blood pressure of rabbits before dosing ranged from 60 to 74 mm Hg and dropped with initial
11   exposure (during the first5-10 minutes that lasted up to 7  hours) to 20-30 mm Hg. Following
12   several months of ammonium hydroxide treatment, a moderate elevation in blood pressure of 10-
13   30 mm Hg was found in 8/10 rabbits. In the other two rabbits, the blood pressure increased from
14   the initial values of 62 and 65-90 mm Hg during the first 7 months of treatment and remained
15   almost unchanged at this level until sacrifice.
16          In summary, Fazekas (1939) concluded that initial decreases in blood pressure and effects
17   of emaciation in rabbits following gavage treatment with ammonium hydroxide is associated with
18   the hypofunction of the cortical or medullary substance of the adrenal gland.  The authors also
19   concluded that the subsequent increases in blood pressure  and body weight could be attributed to
20   hypertrophy of the adrenal cortex.  This study is limited by lack of reporting detail and inadequate
21   study design.  EPA did not identify a NOAEL or LOAEL from this study.
22
23   Toth(19721
24          Toth (1972)  evaluated whether hydrazine, methylhydrazines, and ammonium hydroxide
25   play a role in tumorigenesis in mice. Solutions of hydrazine (0.001%), methyl hydrazine (0.01%),
26   methyl hydrazine sulfate (0.001%), and ammonium hydroxide (0.1,  0.2, and 0.3%) were
27   administered continuously in the drinking water of 5- and 6-week-old randomly bred Swiss mice
28   (50/sex) for their entire lifetime. For ammonium hydroxide, the study authors reported the
29   average daily drinking water intakes for the 0.1, 0.2, and 0.3% groups as 9.2, 8.2, and 6.5 mL/day
30   for males, respectively, and 8.3, 6.5, and 4.8 mL/day for females, respectively. Given these rates and
31   assuming average default body weights of 37.3 and 35.3 gfor males  and females, respectively (U.S.
32   EPA, 1988), the approximate continuous doses for ammonium hydroxide  are  250, 440, and
33   520 mg/kg-day for males and 240,  370, and 410 mg/kg-day for females. Additionally, groups of
34   C3H mice (40/sex) were exposed to ammonium hydroxide  in the drinking water at a concentration
35   of 0.1% for their lifetime. Average daily water consumption for these mice was reported as 7.9 and
36   8.4 mL/day for males and females, respectively. The approximate equivalent doses for these mice
37   assuming the  same default body weights as above (U.S. EPA. 1988) are 191 and 214 mg/kg-day for
38   males and females, respectively. Data were not reported for a concurrent control group. Mice were
39   monitored weekly for changes in body weights, and gross pathological changes were recorded. The
40   animals were either allowed to die  or were killed when found in poor condition. Complete

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-37            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   necropsies were performed on all mice, and the liver, kidney, spleen, lung, and organs with gross
 2   lesions were processed for histopathological examination. Data on body weights were not
 3   reported.
 4          For Swiss mice, tumor incidence at the 0.3% ammonium hydroxide concentration was as
 5   follows: malignant lymphomas: 3/50 (males), 9/50 (females); and lung adenoma or
 6   adenocarcinoma: 7/50 (males), 4/50 (females).  Tumor incidence atthe 0.2% ammonium
 7   hydroxide concentration was: malignant lymphomas: 7/50 (males), 10/50 (females); lung adenoma
 8   or adenocarcinoma: 5/50 (males), 8/50 (females); and breast tumors: 4/50 (females). Tumor
 9   incidence at the 0.1% ammonium hydroxide concentration was: malignant lymphomas:
10   4/50 (males), 10/50 (females); lung adenoma or adenocarcinoma: 5/50 (males), 12/50 (females);
11   and breast tumors: 1/50 (females). The denominators were not adjusted for survival, and
12   concurrent control data were not provided.  For a second strain of mice (C3H) that received 0.1%
13   ammonium hydroxide in drinking water, the incidence of adenocarcinomas of the mammary gland
14   in female mice was 60%. The incidence of breast tumors in the corresponding untreated control
15   mice was 76%. Other tumors were identified in treated mice, but we re of low incidence. Toth
16   (1972) concluded that ammonium hydroxide was not carcinogenic in either strain of mouse.
17   Because concurrent control tumor incidence was not provided other than the incidence of breast
18   tumors in C3H female mice, the incidence of tumors in treated mice cannot be independently
19   compared to control tumor incidence.
20
21   Tsuiii etal. (19951: Tsuiii etal. f!992bl
22          Tsujiietal. (1992b) and Tsujiietal. (1995) evaluated the role of ammonia in H. pylori-
23   related gastric carcinogenesis. H. pylori is a bacterium that produces a potent urease, which
24   generates ammonia from urea in the stomach, and has been implicated in the development of
25   gastric cancer. Tsujiietal. (1992b) and Tsujiietal. (1995) pretreated groups of 40-44 male
26   Sprague-Dawley rats with the initiator N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) in the
27   drinking water for 24 weeks before administering 0.01% ammonium solution as a drinking fluid for
28   24 weeks. Based on an average body weight of 523 g for male Sprague-Dawley rats during chronic
29   exposure (U.S. EPA, 1988) and a reported water consumption rate of 0.05 L/day, the approximate
30   continuous dose administered to these rats is 10 mg/kg-day.  In each study, an additional group of
31   40-43 rats given tap water for 24 weeks following pretreatment with MNNG served as controls.
32   The study protocol did not include a dose group that received ammonia only in drinking water.
33   Stomachs from rats surviving beyond 45 weeks were examined histologically for evidence of ulcers,
34   lesions, and tumors. Tsujiietal. (1995) also evaluated serum gastrin levels from blood collected at
35   30 and 46 weeks and mucosal cell proliferation in animals surviving to 48 weeks by calculating the
36   labeling index (percentage ratio of labeled nuclei to total number of nuclei in the proliferation zone)
37   and the proliferation zone index (fraction of the  gastric pit occupied by the proliferation zone).
38          Tsujiietal. (1992b) and Tsujiietal. (1995) observed a significantly greater incidence of
39   gastric cancers among rats receiving ammonia after pretreatment with MNNG compared to rats
40   receiving only MNNG and tap water (p < 0.01, x2 test). Seventy percent of MNNG+ammonia-treated

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-38            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   rats versus 31% of control rats developed gastric tumors in the first study [Tsujiietal., 1992b].
 2   The number of gastric cancers per tumor-bearing rat in this study was 2.1 ± 1.4 among treated rats
 3   and 1.3 ± 0.6 among control rats (p < 0.01, x2 test).
 4          In the second study, 66% of rats dosed with ammonia and pretreated with MNNG developed
 5   gastric cancers compared to 30% of the control rats [Tsujiietal. [1995].  The numbers of gastric
 6   tumors per rat in this study were also significantly higher among MNNG+ammonia-exposed rats
 7   compared to controls (p < 0.001, Mann-Whitney test), suggesting that ammonia was a promoter. In
 8   the absence of an ammonia-only treatment group, however, it is not possible to distinguish with
 9   certainty between possible promotion and initiator activity.  The degree of differentiation of
10   adenocarcinomas in control and ammonia-treated rats was significantly different Ammonia-
11   treated rats also demonstrated a significantly higher incidence of larger tumors (5.3 mm compared
12   to 4.4 mm for controls) and of gastric cancers penetrating the muscularis propria or deeper
13   (p < 0.01, 22% compared to 12% of controls). In this study, the labeling index and the proliferation
14   zone index were statistically significantly elevated in ammonia-exposed rats compared to  controls
15   in the fundic mucosa and antral mucosa.
16          Tsujii etal. [1995] explored the hypothesis that ammonia might increase intragastric pH,
17   leading to an increase in serum gastrin, a trophic hormone in the gastric fundus mucosa and a
18   possible proliferating factor in gastric epithelial cells. The investigators found no significant effects
19   on serum gastrin levels and concluded that serum gastrin does not appear to play a significant role
20   in ammonia-induced promotion.
21
22   E.3.2. Inhalation Exposure
23   Anderson etal. (19641
24          Anderson et al. [1964] exposed a group of 10 guinea pigs (strain not given) and 10 Swiss
25   albino mice of both sexes continuously to 20 ppm (14 mg/m3) ammonia vapors for up to 6 weeks
26   (anhydrous ammonia, purity not reported). Controls (number not specified) were maintained
27   under identical conditions except for the exposure to ammonia. An additional group of six guinea
28   pigs was exposed to 50 ppm (35 mg/m3) for 6 weeks. The animals were observed daily for
29   abnormal signs or lesions. At termination, the mice and guinea pigs were sacrificed  (two per group
30   at 1, 2, 3, 4, and 6 weeks of exposure), and selected tissues (lungs, trachea, turbinates, liver, and
31   spleen) were examined for  gross and microscopic pathological changes. No significant effects were
32   observed in animals exposed for up to 4 weeks, but exposure to 14 mg/m3 for 6 weeks caused
33   darkening, edema, congestion, and hemorrhage in the lung. Exposure of guinea pigs to 35  mg/m3
34   ammonia for 6 weeks caused grossly enlarged and congested spleens, congested livers and lungs,
35   and pulmonary edema.
36
37   Coon  etal. (19701
38          Coon etal. [1970] exposed groups of male and female Sprague-Dawley and Long-Evans rats,
39   male and female Princeton-derived guinea pigs, male New Zealand rabbits, male  squirrel monkeys,
40   and purebred male beagle dogs to 0,155, or 770 mg/m3 ammonia for 8 hours/day, 5 days/week for

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-39            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   6 weeks (anhydrous ammonia, >99% pure). The investigators stated that a typical loaded chamber
 2   contained 15 rats, 15 guinea pigs, 3 rabbits, 3 monkeys, and 2 dogs. Blood samples were taken
 3   before and after the exposures for determination of hemoglobin concentration, packed erythrocyte
 4   volume, and total leukocyte counts. Animals were routinely checked for clinical signs of toxicity. At
 5   termination, sections of the heart, lung, liver, kidney, and spleen were processed for microscopic
 6   examination in approximately half of the surviving rats and guinea pigs and all of the surviving dogs
 7   and monkeys. Sections of the brain, spinal cord, and adrenals from dogs and monkeys were also
 8   retained, as were sections of the thyroid from the dogs. The nasal passages were not examined in
 9   this study.
10          Exposure to 155 mg/m3 ammonia did not result in any deaths or adverse clinical signs of
11   to xicity in any of the animals. Hematological values were within normal limits for the laboratory
12   and there were no significant gross alterations in the organs examined. Microscopic examination
13   showed evidence of focal pneumonitis in the lung of one of three monkeys. Exposure to 770 mg/m3
14   caused initial mild to moderate lacrimation and dyspnea in rabbits and dogs.  However, these
15   clinical signs disappeared by the second week of exposure. No significant alterations were
16   observed in hematology tests or upon gross or microscopic examinations at the highest dose.
17   However, consistent nonspecific inflammatory changes (not further described) that were more
18   extensive than in control animals (incidence not reported) were observed in the lungs from rats
19   and guinea pigs in the high-dose group.
20          Coonetal. (1970) also exposed rats (15-5I/group) continuously to ammonia (anhydrous
21   ammonia, >99% pure) atO, 40,127, 262, 455, or 470 mg/m3 for 90-114 days. Fifteen guinea pigs,
22   three rabbits, two dogs, and three monkeys were also exposed continuously under similar
23   conditions to ammonia at either 40 or 470 mg/m3.  No significant effects were reported in any
24   animals exposed to 40 mg/m3 ammonia. Exposure of rats to 262 mg/m3 ammonia caused nasal
25   discharge in 25%; nonspecific circulatory and degenerative changes in the lungs and kidneys were
26   also demonstrated (not further described, incidence not reported), which the authors stated were
27   difficult to relate to ammonia inhalation. A frank effect level at 45 5 mg/m3 was observed due to
28   high mortality in the rats (50/51). Thirty-two of 51 rats died by day 25 of exposure; no
29   histopathological examinations were conducted in these rats.  Exposure to 470 mg/m3 caused death
30   in 13/15 rats and 4/15 guinea pigs and marked eye irritation in dogs and rabbits. Dogs
31   experienced heavy lacrimation and nasal discharge, and corneal opacity was noted in rabbits.
32   Hematological values did not differ significantly from controls in animals exposed to 470 mg/m3
33   ammonia. Histopathological evaluation of animals exposed to 470 mg/m3 consistently showed
34   focal or diffuse interstitial pneumonitis in all animals and alterations in the kidneys (calcification
35   and proliferation of tubular epithelium), heart (myocardial fibrosis), and liver (fatty change) in
36   several animals of each species (incidence not reported). The study authors did not determine a
37   NOAEL or LOAEL concentration from this study. EPA identified a NOAEL of 262 mg/m3 and a
38   LOAEL of 455 mg/m3 based on nonspecific inflammatory changes  in the lungs and kidneys in rats
39   exposed to ammonia for 90 days.
40

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-40            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   Stombaugh etal. (19691
 2          Stombaugh etal. [1969] exposed groups of Duroc pigs (9/group) to measured
 3   concentrations of 12, 61,103, or 145 ppm ammonia (8, 43, 73, or 103 mg/m3) continuously for
 4   5 weeks (anhydrous ammonia, purity not reported). Endpoints evaluated included clinical signs,
 5   food consumption (measured 3 times/week), weight gain (measured weekly), and gross and
 6   microscopic examination of the respiratory tract at termination. A control group was not included.
 7   In general, exposure to ammonia reduced food consumption and body weight gain, but because a
 8   control group was not used, it could not be determined whether this reduction was statistically
 9   significant  Food efficiency (food consumed/kg body weight gain) was not affected. Exposure to
10   >73 mg/m3 ammonia appeared to cause excessive nasal, lacrimal, and mouth secretions and
11   increased the frequency of cough (incidence data for these effects were not reported). Examination
12   of the respiratory tract did not reveal any significant exposure-related alterations. The study
13   authors did not identify a NOAEL or LOAEL concentration from this study.
14
15   Doig and Willoughby (19711
16          Doig and Willoughby (1971) exposed groups of six specific-pathogen-free derived Yorkshire
17   Landrace pigs to 0 or 100 ppm ammonia (0 or 71 mg/m3) continuously for up to 6 weeks.  The
18   mean concentration of ammonia in the control chamber was 8 ppm (6 mg/m3). Additional groups
19   of pigs were exposed to similar levels of ammonia as well as to 0.3 mg/ft3 of ground corn dust to
20   simulate conditions on commercial farms. Pigs were monitored daily for clinical signs and changes
21   in behavior. Initial and terminal body weights were measured to determine body weight gain
22   during the exposure period. Blood samples were collected prior to the start of each experiment and
23   at study termination for hematology (packed cell volume, white blood cell, differential leukocyte
24   percentage, and total serum lactate dehydrogenase). Two pigs (one exposed and one control) were
25   necropsied at weekly intervals, and tracheal swabs for bacterial and fungal culture were taken.
26   Histological examination was conducted on tissue samples from the lung, trachea, and bronchial
27   lymph nodes.
28          During the first week of exposure, exposed pigs exhibited slight signs of conjunctival
29   irritation including photophobia and excessive lacrimation. These irritation effects were not
30   apparent beyond the first week. Measured air concentrations in the exposure chambers increased
31   to more than 150 ppm (106 mg/m3) on two occasions. Doig and Willoughby (1971) reported that,
32   at this concentration, the signs of conjunctival irritation were more pronounced in all pigs. No
33   adverse effects on body weight gain were apparent. Hematological parameters and gross pathology
34   were comparable between exposed and control pigs. Histopathology revealed epithelial thickening
35   in the trachea of exposed pigs and a corresponding decrease in the numbers of goblet cells (see
36   Table E-12). Tracheal thickening was characterized by thinning and irregularity of the ciliated
37   brush border and an increased number of cell layers.  Changes in bronchi and bronchioles,
38   characterized as lymphocytic cuffing, were comparable between exposed and control pigs.
39   Similarly, intraalveolar hemorrhage and lobular atelectasis were common findings in both exposed
40   and control pigs. Pigs exposed to both ammonia and dust exhibited similar reactions as those pigs

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-41            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1
 2
 o
 J
 4
exposed only to ammonia, although initial signs of conjunctival irritation were more severe in these
pigs, and these pigs demonstrated lesions in the nasal epithelium similar to those observed in the
tracheal epithelium of pigs exposed only to ammonia.
            Table E-12. Summary of histological changes observed in pigs exposed to
            ammonia for 6 weeks
Duration of exposure (wks)
1
2
3
4
5
6
Mean ±SD
Thickness of tracheal epithelium (u.m)
Control
15.7
20.4
20.4
21.8
19.3
18.9
19.4 ±2.1
71 mg/m3 NH3
21.0
29.3
36.6
36.2
33.2
41.6
32.9 ±7.2
Number of tracheal goblet cells (per
500 urn)
Control
13.6
22.7
18.9
18.3
20.2
20.0
18.9 ±3.0
71 mg/m3 NH3
24.0
10.3
7.3
10.7
10.0
1.3
10.6 ±7.5
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Source: Doig and Willoughby (1971).

       Doig and Willoughby [1971] concluded that ammonia exposure at 71 mg/m3 may be
detrimental to young pigs. The authors suggested that although the structural damage to the upper
respiratory epithelium was slight, such changes may cause severe functional impairment The
study authors did not identify a NOAEL or LOAEL concentration from this study. EPA identified a
LOAEL of 71 mg/m3 based on damage to the upper respiratory epithelium. A NOAEL could not be
identified from this single-concentration study.

Broderson etal. (1976)
       Broderson et al. [1976] exposed groups of Sherman rats [5/sex/dose] continuously to 10 or
150 ppm ammonia (7 or 106 mg/m3, respectively] for 75 days (anhydrous ammonia, purity not
reported]. The 7 mg/m3 exposure level represented the background ammonia concentration
resulting from cage bedding that was changed 3 times/week. The 106 mg/m3 concentration
resulted from cage bedding that was replaced occasionally, but never completely changed.  F344
rats (6/sex/group] were exposed to ammonia in an inhalation chamber at concentrations of 0 or
250 ppm (177 mg/m3] continuously for 35 days.  Rats were sacrificed atthe end of the exposure
period, and tissues were prepared for histopathological examination of nasal passages, middle ear,
trachea, lungs, liver, kidneys, adrenal, pancreas, testicle,  mediastinal lymph nodes, and spleen.
       Histopathological changes were observed in the nasal passage of rats exposed to
106 mg/m3 for 75 days (from bedding] or 177 mg/m3 for 35 days (inhalation chamber]. Nasal
lesions were most extensive in the anterior portions of the nose compared with posterior sections
of the nasal cavity. The respiratory and olfactory mucosa was similarly affected with a three- to
fourfold increase in the thickness of the epithelium. Pyknotic nuclei and eosinophilic cytoplasm
           This document is a draft for review purposes only and does not constitute Agency policy.
                                         E-42            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   were observed in epithelial cells located along the basement membrane.  Epithelial cell hyperplasia
 2   and formation of glandular crypts were observed, and neutrophils were located in the epithelial
 3   layer, the lumina of submucosal glands, and the nasal passages.  Dilation of small blood vessels and
 4   edema were observed in the submucosa of affected areas. Collagen replacement of submucosal
 5   glands and the presence of lymphocytes and neutrophils were also observed. No histopathological
 6   alterations were seen in control rats (7 mg/m3 from bedding or 0 mg/m3 from the inhalation
 7   chamber). Brodersonetal. [1976] did not identify a NOAEL or LOAEL from this study. EPA
 8   identified a NOAEL of 7 mg/m3 and a LOAEL of 106 mg/m3 based on nasal lesions in rats exposed to
 9   ammonia (from bedding) for 75 days.
10
11   Gaafaretal. (19921
12          Gaafaretal. [1992] exposed 50 adult male white albino mice under unspecified conditions
13   to ammonia vapor derived from a 12% ammonia solution (air concentrations were not reported)
14   for 15 minutes/day, 6 days/week for up to 8 weeks. Twenty-five additional mice served as
15   controls. Starting the fourth week, 10 exposed and 5 control mice were sacrificed weekly.
16   Following sacrifice, the nasal mucosa was removed and examined histologically. Frozen sections of
17   the nasal mucosa were subjected to histochemical analysis (succinic dehydrogenase, nonspecific
18   estrase, acid phosphatase, and alkaline phosphatase [ALP]).  Histological  examination revealed a
19   progression of changes in the nasal mucosa of exposed rats from the formation of crypts and
20   irregular cell arrangements at 4 and 5 weeks; epithelial hyperplasia, patches of squamous
21   metaplasia, and loss of cilia at 6 weeks; and dysplasia in the nasal epithelium at 7 weeks. Similar
22   changes were exaggerated in the  nasal mucosa of rats sacrificed at 8 weeks.  Neoplastic changes
23   included a carcinoma in situ in the nostril of one rat sacrificed at 7 weeks, and an invasive
24   adenocarcinoma in one rat sacrificed at 8 weeks. Histochemical results revealed changes in
25   succinic dehydrogenase, acid phosphatase, and ALP in exposed mice compared to controls
26   (magnitude of change not reported), especially in areas of the epithelium characterized by
27   dysplasia. Succinic dehydrogenase and acid phosphatase changes were largest in the superficial
28   layer of the epithelium, although the acid phosphatase reaction was stronger in the basal and
29   intermediate layers in areas of squamous metaplasia. The presence of ALP was greatest in the
30   goblet cells from the basal part of the epithelium and basement membrane.
31          In summary, Gaafar et al.  [1992] observed that ammonia exposure induces histological
32   changes in the nasal mucosa of male mice that increase in severity over longer exposure periods.
33   Corresponding abnormalities in histochemistry suggest altered cell metabolism and energy
34   production, cell injury, cell proliferation, and possible chronic inflammation and neoplastic
35   transformation. The study authors did not determine a NOAEL or LOAEL concentration from this
36   study.  EPA did not identify a NOAEL or LOAEL because air concentrations were not  reported in the
37   study.
38
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-43            DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1   Done etal. (20051
 2          Done etal. [2005] continuously exposed groups of 24 weaned pigs of several breeds in an
 3   experimental facility to atmospheric ammonia at 0, 0.6,10,18.8, or 37 ppm (0, 0.4, 7,13.3, or
 4   26 mg/m3) and 1.2, 2.7, 5.1, or 9.9 mg/m3 inhalable dust for 5 weeks (16 treatment combinations).
 5   The concentrations of ammonia and dust used were representative of those found commercially. A
 6   split-plot design was used in which one dust concentration was allocated to a "batch" (which
 7   involved five lots of 24 pigs each) and the four ammonia concentrations were allocated to the four
 8   lots within that batch. The fifth lot served as a control. Each batch was replicated.
 9
           2 x  [4 dust concentrations x 4 ammonia concentrations + 4 controls] =  40 lots total
10
11          In total, 960 pigs  (460 males and 500 females) were used in the study; 560 pigs were given
12   postmortem examinations.  Blood was collected from 15 sows before the start of the experiment
13   and tested for porcine reproductive and respiratory syndrome virus and swine influenza. Five
14   sentinel pigs were sacrificed at the start of each batch, and lung, nasal cavity, and trachea, together
15   with material from any lesions, were examined postmortem and subjected to bacteriological
16   examination.
17          Postmortem examination involved examination of the pigs' external surfaces for condition
18   and abnormalities, examination of the abdomen for peritonitis and lymph node size, internal gross
19   examination of the stomach for abnormalities, and gross examination of the nasal turbinates,
20   thorax, larynx, trachea, tracheobronchial lymph nodes, and lung. Pigs were monitored for clinical
21   signs (daily), growth rate, feed consumption, and feed conversion efficiency (frequency of
22   observations not specified). After 37 days of exposure, eight pigs from each lot were sacrificed.
23   Swabs of the nasal cavity and trachea were taken immediately after death for microbiological
24   analysis, and the pigs were grossly examined postmortem.  On day 42, the remaining pigs were
25   removed from the exposure facility and transferred to a naturally ventilated building for a recovery
26   period of 2 weeks.  Six pigs from each lot were assessed for evidence of recovery and the remaining
27   10 pigs were sacrificed and examined postmortem.
28          The pigs in this study demonstrated signs of respiratory infection and disease common to
29   young pigs raised on a commercial farm (Done etal. (2005). The different concentrations of
30   ammonia and dust did not have a significant effect on the pathological findings in pigs or on the
31   incidence of pathogens. In summary, exposure to ammonia and inhalable dust at concentrations
32   commonly found at pig farms was not associated with increase in the incidence of respiratory or
33   other disease.  The study authors did not identify a NOAEL or LOAEL concentration from this study.
34   EPA identified a NOAEL of 26 mg/m3, based on the lack of respiratory or other disease following
35   exposure to ammonia in  the presence of respirable dust
36
37   Weatherby (19521
38          Weatherby (1952) exposed a group of 12 guinea pigs (strain not reported) to a target
39   concentration of 170 ppm (120 mg/m3) 6 hours/day, 5 days/week for up to 18 weeks (anhydrous

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-44            DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   ammonia, purity not reported). The actual concentration measured in the exposure chamber
 2   varied between 140 ppm (99 mg/m3) and 200 ppm (141 mg/m3). A control group of six guinea
 3   pigs was exposed to room air. All animals were weighed weekly. Interim sacrifices were conducted
 4   at intervals of 6 weeks (four exposed and two control guinea pigs), and the heart, lungs, liver,
 5   stomach and small intestine, spleen, kidneys, and adrenal glands were removed for microscopic
 6   examination; the upper respiratory tract was not examined.
 7          No exposure-related effects were observed in guinea pigs sacrificed after 6 or 12 weeks of
 8   exposure. However, guinea pigs exposed to ammonia for 18 weeks showed considerable
 9   congestion of the spleen, liver, and kidneys, and early degenerative changes in the adrenal gland.
10   The most severe changes occurred in the spleen and the least severe changes occurred in the liver.
11   The spleen of exposed guinea pigs contained a large amount of hemosiderin, and kidney tubules
12   showed cloudy swelling with precipitated albumin in the lumens and some urinary casts
13   (cylindrical structures indicative of disease). The incidence of histopathological  lesions was not
14   reported. EPA identified the ammonia concentration of 120 mg/m3 to be a LOAEL based on
15   congestion of the spleen, liver, and kidneys and early degenerative changes in the adrenal gland.  A
16   NOAEL could not be identified in this single-concentration study.
17
18   Curtis etal. (19751
19          Curtis etal. (1975) exposed groups of crossbred pigs (4-8/group) to 0, 50,  or 75 ppm
20   ammonia (0, 35, or 53 mg/m3) continuously for up to 109 days (anhydrous ammonia, >99.9%
21   pure). Endpoints evaluated included clinical signs and body weight gain. At termination, all pigs
22   were subjected to a complete gross examination and representative tissues from the respiratory
23   tract, the eye and its associated structures, and the visceral organs (not specified) were taken for
24   subsequent microscopic examination. Weight gain was not significantly affected by exposure to
25   ammonia, and the results of the evaluations of tissues and organs were unremarkable. The
26   turbinates, trachea, and lungs of all pigs were classified as normal. The study authors did not
27   identify a NOAEL or LOAEL from this study. EPA identified a NOAEL of 53 mg/m3 based on the
28   absence of effects occurring in pigs  exposed to ammonia; a LOAEL was not identified from this
29   study.
30
31   E.3.3. Reproductive/Developmental Studies
32   Diekman etal. (1993)
33          Diekman etal. (1993) reared 80 crossbred gilts (young female pigs) in a conventional
34   grower from 2 to 4.5 months of age; pigs were exposed naturally during that time to Mycoplasma
35   hypopneumoniae and Pasteurella multocida, which causes pneumonia and atrophic rhinitis,
36   respectively. At 4.5 months of age, the pigs were transferred to environmentally regulated rooms
37   where they were exposed continuously to a mean concentration of ammonia of 7 ppm (range, 4-
38   12 ppm) (5 mg/m3; range, 3-8.5 mg/m3) or 35 ppm (range, 26-45 ppm) (25 mg/m3; range, 18-
39   32 mg/m3) for 6 weeks (Diekman etal., 1993). A control group was not included in this study. The
40   low concentration of ammonia was obtained by the flushing of manure pits weekly and the higher

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                              E-45           DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   concentration of ammonia was maintained by adding anhydrous ammonia (purity not reported) to
 2   manure pits that were not flushed. After 6 weeks of exposure, 20 gilts from each group were
 3   sacrificed, and sections of the lungs and snout were examined for gross lesions.  In addition, the
 4   ovaries, uterus, and adrenal glands were weighed. The remaining 20 gilts/group were mated with
 5   mature boars and continued being exposed to ammonia until gestation day 30, at which time they
 6   were sacrificed. Fetuses were examined for viability, weight, and length, and the number of copora
 7   lutea were counted.
 8          Gilts exposed to 2 5 mg/m3 ammonia gained less weight than gilts exposed to 5 mg/m3
 9   during the first 2 weeks of exposure (7% decrease, p < 0.01), but growth rate recovered thereafter.
10   Mean scores for lesions in the lungs and snout were not statistically different between the two
11   exposure groups, and there were no differences in the weight of the ovaries, uterus, and adrenals.
12   Age at puberty did not differ significantly between the two groups, but gilts exposed to 25 mg/m3
13   ammonia weighed 7% less (p < 0.05) at puberty than those exposed to 5 mg/m3. In gilts that were
14   mated, conception rates were similar between the two groups (94.1 versus 100% in low versus
15   high exposure, respectively). At sacrifice on day 30 of gestation, body weights were not
16   significantly different between the two groups. In addition, there were no significant differences
17   between the two groups regarding percentage of lung tissue  with lesions and mean snout grade.
18   Number of corpora lutea, number of live fetuses, and weight and length  of the fetuses on day 30 of
19   gestation were not significantly different between treatment groups. Diekmanetal. (1993) did not
20   identify NOAEL or LOAEL concentrations for maternal or fetal effects in this study. EPA did not
21   identify NOAEL or LOAEL values from this study due to the absence of a control  group and due to
22   confounding exposures to bacterial and mycoplasm pathogens.
23
24   E.3.4. Acute and Short-term Inhalation Toxicity Studies
25          Table E-13 provides information on animal studies of acute and  short-term inhalation
26   exposure to ammonia.
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-46            DRAFT—DO NOT CITE OR QUOTE

-------
                                                                              Supplemental Information—Ammonia
Table E-13. Acute and short-term inhalation toxicity studies of ammonia in animals
Animal
Ammonia
concentration
(mg/m3)
Duration
Parameters
examined
Results
Reference
Rats
Female Porton rats
(16/group)
Male OFA rats
(27/group)
Male and female
Wistar rats
(5/sex/group)
MaleCrhCOBSCD
Sprague-Dawley rats
(8/group)
MaleCrhCOBSCD
Sprague-Dawley rats
(14/group)
0 or 141
0 or 354
9,898-37,825; no
mention of control
group
11, 23, 219, and 818;
arterial blood
collected prior to
exposure served as
control
3, 17, 31, 117, and
505; arterial blood
collected prior to
exposure served as
control
Continuous
exposure for 4,
8, or 12 d
Continuous
exposure for 1-
8 wks
10, 20, 40, or
60 min
24hrs
3 and 7 d
Histology of the trachea
Body weight, organ
weights, airway structure,
cell population, alveolar
macrophages
Clinical signs, pathology,
LC50
Clinical signs, histology,
blood pH, blood gas
measurement
Hepatic cytochrome P450
content and
ethylmorphine-
N-demethylase activity
4 d: transitional-stratified appearance of the
epithelium
8 d: gross change with disappearance of cilia
and stratification on luminal surface
12 d: increased epithelial thickness
No deaths occurred; decreased food
consumption and body weight gain; increased
lung and kidney weights; at 3 wks, nasal
irritation and upper respiratory tract
inflammation, but no effect on lower airways;
slight decrease in alveolar macrophages; no
histopathological effects seen at 8 wks,
suggesting adaptation to exposure
Eye irritation, eye and nasal discharge,
dyspnea; hemorrhagic lungs on necropsy;
10-min LC50 = 28,492 mg/m3
20-min LC50 = 20,217 mg/m3
40-min LC50 = 14,352 mg/m3
60-min LC50 = 11,736 mg/m3
No clinical signs of toxicity, no histologic
differences in tracheal or lung sections, no
change in blood pH or pCO2, minor changes in
p02
No dose-related change in P450 content or
enzyme activity
Gamble and dough
(1976)
Richard etal. (1978a)

Appelman et al.
(1982)
Schaerdel etal. (1983)

Schaerdel etal. (1983)

                      This document is a draft for review purposes only and does not constitute Agency policy.
                                                     E-47                          DRAFT—DO NOT CITE OR QUOTE

-------
                                                                              Supplemental Information—Ammonia
Table E-13. Acute and short-term inhalation toxicity studies of ammonia in animals
Animal
Male Long-Evans
rats (4/group)
Female Wistar rats
(5/group)
Female Wistar rats
(5/group)
Female albino rats
(8/group)
Male Sprague-
Dawley rats
(number/group not
given)
Male rats
(10/group)
Ammonia
concentration
(mg/m3)
70 and 212; results
were compared to
"control", but it was
not clear if the
authors were
referring to historical
or concurrent controls
0, 18, or 212
0, 18, or 212
OQ/IQ 1 rtfis

Air concentration not
given; ammonia vapor
added to inspiratory
line of ventilator;
controls exposed to
same volume of room
air
0, 848-1,068 at the
beginning and end of
the exposure period
Duration
6hrs
6 hrs/d for 5, 10,
orlSd
6 hrs/d for 5 d
3hrs
20 sec
3hrs
Parameters
examined
Clinical signs, behavioral
observation
Blood ammonia, urea,
glutamine, and pH; brain
ammonia, glutamine;
histopathology of lungs,
heart, liver, and kidneys
(light and electron
microscopy)
Plasma and brain
ammonia and amino acid
analysis
Mortality, respiratory
movement, and O2
consumption
Activity of upper thoracic
spinal neurons
Oxygen consumption
Results
Decreased running, decreased activity
Brain and blood glutamine increased; slight
acidosis (i.e., decreased blood pH) at
212 mg/m3; lung hemorrhage observed in some
exposed rats
Increase in brain and plasma glutamine
concentrations; increased brain/plasma ratio of
threonine
No deaths reported; inhibition of external
respiration and decreased O2 consumption
Lower airway irritation, activation of vagal
pulmonary afferents and upper thoracic spinal
neurons receiving pulmonary sympathetic input
Decreased O2 consumption
Reference
Tepperetal. (1985)

Manninen et al.
(1988)
Manninen and
Savolainen (1989)

Reiniuketal. (2007)

Qinetal. (2007a, b)

Reiniuketal. (2008)

                      This document is a draft for review purposes only and does not constitute Agency policy.
                                                     E-48                          DRAFT—DO NOT CITE OR QUOTE

-------
                                                                              Supplemental Information—Ammonia
Table E-13. Acute and short-term inhalation toxicity studies of ammonia in animals
Animal
Male Wistar rats
(4/group)
Ammonia
concentration
(mg/m3)
OQ9 1 9A3' tho
preexposure period
was used as the
control for each
animal
Duration
45 min
Parameters
examined
Airway reflexes by the
changes in respiratory
patterns elicited by
ammonia in either dry,
steam-humidified, or
aqueous aerosol-
containing atmospheres
Results
Ammonia-induced upper respiratory tract
sensory irritation is not affected to any
appreciable extent by wet atmospheres (with
or without aerosol) up to 1,243 mg/m
Reference
Li and Pauluhn (2010)

Mice
Mice (20/group,
species, sex not
specified)
Male Swiss albino
mice (4/group)
Albino mice (sex not
specified; 6/dose)
Male Swiss-Webster
mice
(4/group)
6,080-7,070; no
controls
5,050-20,199; no
controls
Air concentration not
measured; results
were compared to
"control", but it was
not clear if the
authors were
referring to historical
or concurrent controls
Concentrations not
given; baseline levels
established prior to
exposure
10 min
30-120 min
Continuously for
2or5d
10 min
LC50
LC50
Regional brain metabolism
(cerebral cortex,
cerebellum, brainstem);
MAO, enzymes of
glutamate and gamma-
aminobutyric acid (GABA)
metabolism, and (Na+-K+)-
ATPase; amino acid levels
in the brain
Reflex decrease in
respiratory rate was used
as an index of sensory
irritation; RD50 = the
concentration associated
with a 50% decrease in the
respiratory rate
LC50 = 7,056 mg/m3
LC50 (30 min) = 15,151 mg/m3
Altered activities of MAO, glutamate
decarboxylase, ALT, GABA-transaminase, and
(Na+-K+)-ATPase; increased alanine and
decreased glutamate
RD50 = 214 mg/m3
Silver and McGrath
(1948)
Hiladoetal. (1977)

Sadasivudu et al.
(1979); Sadasivudu
and Radha Krishna
Murthy(1978)

Kane etal. (1979)

                      This document is a draft for review purposes only and does not constitute Agency policy.
                                                     E-49                          DRAFT—DO NOT CITE OR QUOTE

-------
                                                                              Supplemental Information—Ammonia
Table E-13. Acute and short-term inhalation toxicity studies of ammonia in animals
Animal
Male albino ICR
mice (12/dose)
Male Swiss-Webster
mice (16-24/group)
Male albino ICR
mice (12/dose)
Male albino ICR
mice (12/dose)
Male Swiss mice
(6/dose)
Mice (sex not
specified; 4/group)
Ammonia
concentration
(mg/m3)
0-3,436
0 or 216
0, 954, 3,097, or 3,323
0, 81, or 233
71 and 212; data
collected during the 2
d separating each
ammonia exposure
served as the control
baseline
3, 21, 40, or 78,
lowest measured
concentration was the
nominal control group
Duration
1 hr (14-d
followup)
6 hrs/d for 5 d
4hrs
4 hrs/d for 4 d
6hrs
2d
Parameters
examined
Clinical signs, body weight,
organ weight,
histopathology, LC50
Respiratory tract
histopathology
Hexobarbitol sleeping
time, microsomal protein
content, liver microsomal
enzyme activity
Microsomal protein
content, liver microsomal
enzyme activity
Clinical signs, behavioral
observation
Responses to atmospheric
ammonia in an
environmental preference
chamber with four
chambers of different
concentrations of
ammonia
Results
Eye and nose irritation, dyspnea, ataxia,
seizures, coma, and death; decreased body
weight and increased liver to body weight ratio
in mice surviving to 14 d; effects in the lung
included focal pneumonitis, atelectasis, and
intralveolar hemorrhage; liver effects included
hepatocellular swelling and necrosis, vascular
congestion; LC50 = 2,990 mg/m3
Lesions in the nasal respiratory epithelium
(moderate inflammation, minimal necrosis,
exfoliation, erosion, or ulceration); no lesions in
trachea or lungs
Increased hexobarbitol sleeping time
(3,097 mg/m3), increased microsomal protein
content and aminopyrene-N-deethylase and
aniline hydroxylase activities (3,323 mg/m3)
No dose-dependent effects on microsomal
enzymes
Decreased running, decreased activity
No distinguishable preference for, or aversion
to, different ammonia concentrations
Reference
Kapeghian et al.
(1982)
Buckley et al. (1984)

Kapeghian et al.
(1985)
Kapeghian et al.
(1985)
Tepperetal. (1985)

Green et al. (2008)

                      This document is a draft for review purposes only and does not constitute Agency policy.
                                                     E-50                          DRAFT—DO NOT CITE OR QUOTE

-------
                                                                              Supplemental Information—Ammonia
Table E-13. Acute and short-term inhalation toxicity studies of ammonia in animals
Animal
Male OF1 mice
(4/group)
Ammonia
concentration
(mg/m3)
OQ9 1 9A3' tho
preexposure period
was used as the
control for each
animal
Duration
45 min
Parameters
examined
Airway reflexes by the
changes in respiratory
patterns elicited by
ammonia in either dry,
steam-humidified, or
aqueous aerosol
containing atmospheres
Results
Ammonia-induced upper respiratory tract
sensory irritation is not affected to any
appreciable extent by wet atmospheres (with
or without aerosol) up to 1,243 mg/m
Reference
Li and Pauluhn (2010)

Rabbits
Female New
Zealand White
rabbits (7-9/dose)
Rabbits (species,
sex, number/dose
not specified)
New Zealand White
rabbits (sex not
specified; 16 total;
8/dose)
0, 35, or 71
07fl7 1A 1Afl

Peak concentrations:
24,745-27,573;
concurrent controls
tested
9 c o n hrc

15-180 min
4 min
Lung function
Lung function, death
Lung function, heart rate,
blood pressure, blood
gases
Decreased respiratory rate at both
concentrations
Bradycardia at 1,768 mg/m3; arterial pressure
variations and blood gas modifications (acidosis
indicated by decreased pH and increased pCO2)
at 3,535 mg/m3; death occurred at
4,242 mg/m3
Lung injury was evident after 2-3 min
(decreased pO2 increased airway pressure)
Mayan and Merilan
(1972)
Richard etal. (1978b)

Sioblometal. (1999)

Cots
Mixed breed stray
cats (sex not
specified; 5/group)
0 or 707
10 min
Lung function, lung
histopathology on 1, 7, 21,
and 35 d postexposure
Lung function deficits were correlated with lung
histopathology; acute effects were followed by
chronic respiratory dysfunction (secondary
bronchitis, bronchiolitis, and
bronchopneumonia)
Dodd and Gross
(1980)
                      This document is a draft for review purposes only and does not constitute Agency policy.
                                                     E-51                          DRAFT—DO NOT CITE OR QUOTE

-------
                                                                              Supplemental Information—Ammonia
Table E-13. Acute and short-term inhalation toxicity studies of ammonia in animals
Animal
Ammonia
concentration
(mg/m3)
Duration
Parameters
examined
Results
Reference
Pigs
Young pigs (sex not
specified; 2/group)
Male and female
Belgian Landrace
pigs (4/group)
Belgian Landrace
pigs (sex not
specified; 4/group)
Landrace-Yorkshire
pigs (sex not
specified; 4/group)
Hybrid gilts (White
synthetic Pietrain,
white Duroc,
Landrace, Large
White)
(14 pigs/group)
0, 35, 71, or 106
0, 18, 35, or 71
0, 18, 35, or 71
Oor42
<4 (control) or 14
Continuous
exposure for
4 wks
6d
6d
15 min/d for
8 wks
15 wks
Clinical signs, food
consumption, body
weight, gross necropsy,
organ weight,
histopathology
Clinical signs, body weight,
lung function
Clinical signs, body weight,
neutrophil count, and
albumin in nasal lavage
fluid
Thromboxane A2 (TXA2),
leukotriene C4 (LTC4), and
prostaglandin (PGI2)
production
Salivary cortisol, adrenal
morphometry, body
weight, food conversion
efficiency, general health
scores, play behavior;
reaction to light and noise
intensity tested
concurrently
Lethargy and histopathological alterations in
the tracheal and nasal epithelium were
observed at 71 and 106 mg/m3; decreased body
weight occurred at all concentrations (7-19%
decrease from control)
Lethargy and decreased body weight gain (all
concentrations); no effect on lung
microvascular hemodynamics and permeability
Nasal irritation (increased neutrophils in nasal
lavage fluid) and decreased body weight gain at
all concentrations
Significant increases in TXA2 and LTC4, no
significant effect on PGI2 production
Decreased salivary cortisol, larger adrenal
cortices, less play behavior, no measurable
impact on productivity or physiological
parameters
Drummond et al.
(1980)
Gustinetal. (1994)

Urbainetal. (1994)

Chaung et al. (2008)

O'Connor etal. (2010)

                      This document is a draft for review purposes only and does not constitute Agency policy.
                                                     E-52                          DRAFT—DO NOT CITE OR QUOTE

-------
                                                   Supplemental Information—Ammonia
1
2
3
4
5
E.4. OTHER PERTINENT TOXICITY INFORMATION
Genotoxicity Studies
      Information on in vitro and in vivo ammonia genotoxicity studies is presented in
Tables E-14 and E-15, respectively.

      Table E-14. Summary of in vitro studies of ammonia genotoxicity
Endpoint
Test system
Concentration3
Results"
Without
activation
With
activation
Comments
Reference
Genotoxicity studies in prokaryotic organisms
Reverse
mutation
Reverse
mutation,
streptomycin
resistance
Salmonella
typhimuhum
(TA98, TA100,
TA1535,
TA1537,
TA1538);
Escherichia coli
(WP2 uvrA)
E. coli (B/SD-4
strains)
25,000 ppm
(17,675 mg/m3)
ammonia vapor
0.25%
ammonia

+ (T)d
c
No data
Plate incorporation
assay with ammonia
vapor
Plate incorporation
assay
Shimizu et
al. (1985)
Demerec et
al. (1951)
Genotoxicity studies in nonmammalian eukaryotic organisms
Chromosomal
aberrations
Chick
fibroblasts
Not available
+
No data
Cultures immersed in
buffered ammonia
solution
Rosenfeld
(1932)
Genotoxicity studies in mammalian systems
DNA double
strand breaks
DNA
fragmentation
Chromatin
condensation
Rabbit gastric
mucosal or
KATO III cells
Rabbit gastric
mucosal cells
Rabbit gastric
mucosal or
KATO III cells
0.1mMNH3in
solution
0.1mMNH3in
solution
0.1mMNH3in
solution
No data
No data
No data

-

15-min incubation with
0.1 mM ammonia

15-min incubation with
0.1 mM ammonia
Suzuki et
al. (1997)
Suzuki et
al. (1997)
Suzuki et
al. (1997)
               This document is a draft for review purposes only and does not constitute Agency policy.
                                          E-53          DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia

       Table E-14. Summary of in vitro studies of ammonia genotoxicity
Endpoint
DNA
fragmentation
Test system
Gastric
epithelial cell
line MKN45
Concentration3
0.001 mM NH3
in solution
Results"
Without
activation
No data
With
activation
e
Comments
5-hr incubation;
cytoplasmic levels of
mono- and
oligonucleosomes
measured
Reference
Suzuki et
al. (1998)
aLowest effective dose for positive results; highest dose tested for negative or equivocal results.
b+ = positive; - = negative; (T) = toxicity reported.
Exogenous metabolic activation used; S9 liver fractions from male Sprague-Dawley rats pretreated with
pentachlorobiphenyl (KC500).
dOnly positive in treatments using toxic levels of ammonia (98% lethality).
Comparison was to elevated mono- and oligonucleosomes levels associated with monochloramine (NH2CI);
control (untreated) value not reported.
            This document is a draft for review purposes only and does not constitute Agency policy.
                                              E-54            DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia

        Table E-15. Summary of in vivo studies of ammonia genotoxicity
Endpoint
Test system
Dose/
concentration3
Results"
Comments
Reference
Genotoxicity studies in mammalian systems
Chromosomal
aberrations
Sister chromatid
exchange
Micronucleus
formation
Sex-linked
recessive lethal
mutations
Dominant lethal
mutations
Dominant lethal
mutations
Human
lymphocytes
Human
lymphocytes
Swiss albino mice
Drosophila
melanogaster
D. melanogaster
D. melanogaster
88.28 |jg/m3
88.28 |jg/m3
12.5-50 mg/kg
Not available
Not available
Not available
+c
+c
+
-m
-m
+ md
22 healthy workers
occupationally exposed
to ammonia in an Indian
fertilizer factory (ambient
concentration of
0.0883 mg/m3);
42 nonexposed factory
staff served as control
subjects
Intraperitoneal injections
for 24-48-hr expression
times
Inhalation exposure to
ammonia as vapor at a
concentration killing the
majority of flies
Inhalation exposure to
ammonia as vapor at a
concentration killing the
majority of flies
Dominant lethal assay;
inhalation exposure up to
318 mg/m3 ammonia,
6 hrs/d for 5 d
Yadavand
Kaushik(1997)
Yadavand
Kaushik(1997)

Yadavand
Kaushik(1997)

Auerbach and
Robson (1947)

Auerbach and
Robson (1947)

Lobasovand
Smirnov
(1934)
aLowest effective dose for positive results; highest dose tested for negative or equivocal results.
b+ = positive; - = negative; (T) = toxicity reported.
frequencies of chromosomal aberrations, sister chromatid exchanges, and mitotic index all increased with
increased duration of exposure. This study is difficult to interpret because of small samples sizes and confounding
factors of smoking and alcohol consumption. In addition, the levels of ammonia in the plant seemed low
compared to other fertilizer plant studies (see, for example, Section 1.1; Rahman et al., 2007; AN et al., 2001; Ballal
et al., 1998); the accuracy and reliability of the sampling and measurement could not be determined.
Survival after exposure was <2%.
             This document is a draft for review purposes only and does not constitute Agency policy.
                                               E-55             DRAFT—DO NOT CITE OR QUOTE

-------
                                                     Supplemental Information—Ammonia
     APPENDIX F.  DOCUMENTATION OF
     IMPLEMENTATION OF THE 2011 NATIONAL
     RESEARCH COUNCIL RECOMMENDATIONS
 3         Background: On December 23, 2011, The Consolidated Appropriations Act, 2012, was
 4   signed into law [U.S. Congress. 2011). The report language included direction to EPA for the
 5   Integrated Risk Information System (IRIS) Program related to recommendations provided by the
 6   National Research Council (NRC) in their review of EPA's draft IRIS assessment of formaldehyde
 7   [NRC, 2011]. The report language included the following:
 8
 9         The Agency shall incorporate, as appropriate, based on chemical-specific data sets
10         and biological effects, the recommendations of Chapter 7 of the National Research
11         Council's Review of the Environmental Protection Agency's Draft IRIS Assessment of
12         Formaldehyde into the IRIS process. ..For draft assessments released in fiscal year
13         2012, the Agency shall include documentation describing how the Chapter 7
14         recommendations of the National Academy of Sciences (NAS) have been
15         implemented or addressed, including an explanation for why certain
16         recommendations were not incorporated.
17
18         The NRC's recommendations, provided in Chapter 7 of the  review report, offered
19   suggestions to EPA for improving the development of IRIS assessments. Consistent with the
20   direction provided by Congress, documentation of how the recommendations from Chapter 7 of the
21   NRC report have been implemented in this assessment is provided in the tables below. Where
22   necessary, the documentation includes an explanation for why certain recommendations were not
23   incorporated.
24         The IRIS Program's implementation of the NRC recommendations is following a phased
25   approach that is consistent with the NRC's "Roadmap for Revision" as described in Chapter 7 of the
26   formaldehyde review report. The NRC stated that, "the committee recognizes that the changes
27   suggested would involve a multi-year process and extensive effort by the staff at the National
28   Center for Environmental Assessment and input and review by the EPA Science Advisory Board and
29   others."
30         Phase 1 of implementation has focused on a subset of the short-term recommendations,
3 1   such as editing and streamlining documents, increasing transparency and clarity,  and using more
32   tables, figures, and appendices to present information and data in assessments. Phase 1 also
33   focused on assessments near the end of the development process and close to final posting. The
34   IRIS ammonia assessment is the first in Phase 2 of implementation, which addresses all of the
               This document is a draft for review purposes only and does not constitute Agency policy.
                                            F-l          DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1    short-term recommendations from Table F-l. The IRIS Program is implementing all of these
 2    recommendations but recognizes that achieving full and robust implementation of certain
 3    recommendations will be an evolving process with input and feedback from the public,
 4    stakeholders, and external peer review committees. Chemical assessments in Phase 3 of
 5    implementation will incorporate the longer-term recommendations made by the NRC as outlined
 6    below in Table F-2, including the development of a standardized approach to describe the strength
 7    of the evidence for noncancer effects.  On May 16, 2012, EPA announced [U.S. EPA, 2012c] that as a
 8    part of a review of the IRIS Program's assessment development process, the NRC will also review
 9    current methods for weight-of-evidence analyses and recommend approaches for weighing
10    scientific evidence for chemical hazard identification. This effort is included in Phase 3 of EPA's
11    implementation plan.
12
13
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                                F-2            DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
       Table F-l. The EPA's implementation of the National Research Council's
       recommendations in the ammonia assessment
       NRC recommendations that EPA is
        implementing in the short term
    Implementation in the ammonia assessment
General recommendations for completing the IRIS formaldehyde assessment that EPA will adopt for all IRIS
assessments (see p. 152)
1.  To enhance the clarity of the document, the draft
   IRIS assessment needs rigorous editing to reduce
   the volume of text substantially and address
   redundancies and inconsistencies.  Long
   descriptions of particular studies should be replaced
   with informative evidence tables. When study
   details are appropriate, they could be provided in
   appendices.
Implemented. The overall document structure has been
revised in consideration of this NRC recommendation.  The
new structure includes a concise Executive Summary and
an explanation of the literature review search strategy,
study selection criteria, and methods used to develop the
assessment. The main body of the assessment has been
reorganized into two sections,  Hazard Identification and
Dose-Response Analysis, to help reduce the volume of text
and redundancies that were a part of the previous
document structure. Section 1 provides evidence tables
and a concise synthesis of hazard information organized by
health effect. More detailed summaries of the most
pertinent epidemiology and experimental animal studies
are provided in Appendix E. Information on chemical and
physical properties and toxicokinetics is now provided  in
Appendices B and E.I, respectively. The main text of the
Toxicological Review is approximately 50 pages, which  is a
major reduction from previous IRIS assessments.
Technical and scientific edits were performed to eliminate
any redundancies or inconsistencies.
             This document is a draft for review purposes only and does not constitute Agency policy.
                                                F-3             DRAFT—DO NOT CITE OR QUOTE

-------
                                                           Supplemental Information—Ammonia
        Table F-l. The EPA's implementation of the National Research Council's
        recommendations in the ammonia assessment
       NRC recommendations that EPA is
        implementing in the short term
    Implementation in the ammonia assessment
2.  Chapter 1 needs to be expanded to describe more
   fully the methods of the assessment, including a
   description of search strategies used to identify
   studies with the exclusion and inclusion criteria
   articulated and a better description of the
   outcomes of the searches and clear descriptions of
   the weight-of-evidence approaches used for the
   various noncancer outcomes.  The committee
   emphasizes that it is not recommending the
   addition of long descriptions of EPA guidelines to
   the introduction, but rather clear concise
   statements of criteria used to exclude, include, and
   advance studies for derivation of the RfCs and  unit
   risk estimates.
Implemented.  Chapter 1 has been replaced with a
Preamble that describes the application of existing EPA
guidance and the methods and criteria used in developing
the assessment. The term "Preamble" was chosen to
emphasize that these methods and criteria are being
applied consistently across IRIS assessments. The new
Preamble includes information on identifying and selecting
pertinent studies, evaluating the quality of individual
studies, weighing the overall evidence of each effect,
selecting studies for derivation of toxicity values, and
deriving toxicity values. These topics correspond directly
to the five steps that the NRC identified in Figure 7-2 of
their 2011 report.
    A new section, Literature Search Strategy | Study
Selection and Evaluation,  provides detailed information on
the search strategy used to identify health effect studies,
search outcomes, and selection of studies for hazard
identification; the complete search string is provided in
Appendix D. This information is chemical-specific and has
been designed to provide enough information that an
independent literature search would  be able to replicate
the results. This section also includes information on how
studies were selected to be included  in the document and
provides a link to EPA's Health and Environmental
Research Online (HERO) database (www.epa.gov/hero)
that contains the references that were cited in the
document, along with those that were considered but not
cited.
3.  Standardized evidence tables for all health
   outcomes need to be developed.  If there were
   appropriates tables, long text descriptions of studies
   could be moved to an appendix of deleted.
Implemented.  In the new document template,
standardized evidence tables that present key study
findings that support how toxicological hazards are
identified for all major health effects are provided in
Section 1.1. More detailed summaries of the most
pertinent epidemiology and experimental animal studies
are provided in Appendix E.
             This document is a draft for review purposes only and does not constitute Agency policy.
                                                 F-4             DRAFT—DO NOT CITE OR QUOTE

-------
                                                            Supplemental Information—Ammonia
        Table F-l.  The EPA's implementation of the National Research Council's
        recommendations in the ammonia assessment
       NRC recommendations that EPA is
         implementing in the short term
    Implementation in the ammonia assessment
4.  All critical studies need to be thoroughly evaluated
   with standardized approaches that are clearly
   formulated and based on the type of research, for
   example, observational epidemiologic or animal
   bioassays.  The findings of the reviews might be
   presented  in tables to ensure transparency.
Partially implemented.  Information in Section 4 of the
Preamble provides an overview of the approach used to
evaluate the quality of individual studies. The evaluation
of epidemiology and animal studies of ammonia, including
consideration of the extent to which studies were
informative and relevant to the assessment, is provided in
the Literature Search Strategy | Study Selection and
Evaluation section, and tables to support the evaluation of
study quality for epidemiology studies are provided in
Appendix D. Consistent with findings of the study quality
review, study design information and results of ammonia
studies are included in the evidence tables in Section 1.1.
Additional information on study characteristics is found in
Appendix E.  Summaries of individual studies for ammonia
are presented in text format only.  EPA is developing
standardized study summary tables that will replace
written study summaries to clearly present more detailed
study summary information and key study characteristics.
As more rigorous systematic review processes are
developed, they will be utilized in future assessments.
5.  The rationales for the selection of the studies that
   are advanced for consideration in calculating the
   RfCs and unit risks need to be expanded.  All
   candidate RfCs should be evaluated together with
   the aid of graphic displays that incorporate selected
   information on attributes relevant to the  database.
Implemented. The Dose-Response Analysis section of the
new document structure provides a clear explanation of
the rationale used to select and advance studies that were
considered for calculating toxicity values. Rationales for
the selection of studies advanced for reference value
derivation are informed by the weight of evidence for
hazard identification as discussed in Section 1.2. Graphical
displays that describe the database (by health endpoint)
are provided in Section 1. In the case of ammonia, the
database did not support development of multiple
candidate RfC's. Such values have been developed
previously for other chemicals and will be developed in
future assessments, where the data allow.
6.  Strengthened, more integrative, and more
   transparent discussions of weight of evidence are
   needed. The discussions would benefit from more
   rigorous and systematic coverage of the various
   determinants of weight of evidence, such as
   consistency.
Partially implemented.  The new Hazard Identification
(Section 1) provides a strengthened and more integrated
and transparent discussion of the weight of the available
evidence. This section includes both standardized
evidence tables to present the key study findings that
support how potential toxicological hazards are identified
and exposure-response arrays for each potential
toxicological effect. Weight-of-evidence discussions are
provided for each major effect (Section 1.1.1 Respiratory
Effects, Section 1.1.2 Gastrointestinal Effects, Section  1.1.3
Immune System Effects, and Section 1.1.4 Other Systemic
Effects). A more rigorous and formalized approach for
characterizing the weight of evidence will be developed as
a part of Phase 3 of the implementation process.
             This document is a draft for review purposes only and does not constitute Agency policy.
                                                 F-5             DRAFT—DO NOT CITE OR QUOTE

-------
                                           Supplemental Information—Ammonia
Table F-l. The EPA's implementation of the National Research Council's
recommendations in the ammonia assessment
NRC recommendations that EPA is
implementing in the short term
Implementation in the ammonia assessment
General Guidance for the Overall Process (see p. 164)
1. Elaborate an overall, documented, and quality-
controlled process for IRIS assessments.
8. Ensure standardization of review and evaluation
approaches among contributors and teams of
contributors; for example, include standard
approaches for reviews of various types of studies
to ensure uniformity.
9. Assess disciplinary structure of teams needed to
conduct the assessments.
Implemented. EPA has created Chemical Assessment
Support Teams to formalize an internal process to provide
additional overall quality control for the development of
IRIS assessments. This initiative uses a team approach to
making timely, consistent decisions about the
development of IRIS assessments across the Program. This
team approach has been utilized for the development of
the ammonia assessment. Additional objectives of the
teams are to help ensure that the necessary disciplinary
expertise is available for assessment development and
review, provide a forum for identifying and addressing key
issues prior to external peer review, and monitor progress
in implementing the NRC recommendations.
Evidence Identification: Literature Collection and Collation Phase (see p. 164)
10. Select outcomes on the basis of available evidence
and understanding of mode of action.
11. Establish standard protocols for evidence
identification.
12. Develop a template for description of the search
approach.
13. Use a database, such as the Health and
Environmental Research Online (HERO) database, to
capture study information and relevant quantitative
data.
Partially implemented. A new section, Literature Search
Strategy | Study Selection and Evaluation, contains
detailed information on the search strategy used for the
ammonia assessment, including keywords used to identify
relevant health effect studies. A complete search string is
provided in Appendix D. Figure LS-1 depicts the study
selection strategy and the number of references obtained
at each stage of literature screening. This section also
includes information on how studies were selected to be
included in the document and provides a link to an
external database (www.epa.gov/hero) that contains the
references that were cited in the document, along with
those that were considered but not cited. Each citation in
the Toxicological Review is linked to HERO such that the
public can access the references and abstracts to the
scientific studies used in the assessment.
Section 3 of the Preamble summarizes the standard
protocols for evidence identification that are provided in
EPA guidance. For each potential toxicological effect
identified for ammonia, the available evidence is informed
by the mode of action information as discussed in Section
1.1. As more rigorous systematic review processes are
developed, they will be utilized in future assessments.
     This document is a draft for review purposes only and does not constitute Agency policy.
                                  F-6           DRAFT—DO NOT CITE OR QUOTE

-------
                                           Supplemental Information—Ammonia
Table F-l. The EPA's implementation of the National Research Council's
recommendations in the ammonia assessment
NRC recommendations that EPA is
implementing in the short term
Implementation in the ammonia assessment
Evidence Evaluation: Hazard Identification and Dose-Response Modeling (see p. 165)
14. Standardize the presentation of reviewed studies in
tabular or graphic form to capture the key
dimensions of study characteristics, weight of
evidence, and utility as a basis for deriving
reference values and unit risks.
15. Develop templates for evidence tables, forest plots,
or other displays.
16. Establish protocols for review of major types of
studies, such as epidemiologic and bioassay.
Implemented. Standardized tables have been developed
that provide summaries of key study design information
and results by health effect. The inclusion of all positive
and negative findings in each health effect-specific
evidence table supports a weight-of-evidence analysis. In
addition, exposure-response arrays are utilized in the
assessment to provide a graphical representation of points
of departure for various effects resulting from exposure to
ammonia. The exposure-response arrays inform the
identification of doses associated with specific effects and
the weight of evidence for those effects.
Implemented. Templates for evidence tables and
exposure-response arrays have been developed and are
utilized in Section 1.1.
Partially implemented. General principles for reviewing
epidemiologic and experimental animal studies are
described in Section 4 of the Preamble. Standardized
systematic review is an ongoing process.
Selection of Studies for Derivation of Reference Values and Unit Risks (see p. 165)
17. Establish clear guidelines for study selection.
a. Balance strengths and weaknesses.
b. Weigh human vs. experimental evidence.
c. Determine whether combining estimates
among studies is warranted.
Implemented. EPA guidelines for study selection,
including balancing strengths and weaknesses and
weighing human vs. experimental evidence, are described
in the Preamble (Sections 3-6). These guidelines have
been applied in Section 2 of the ammonia assessment to
inform the evaluation of the weight-of-evidence across
health effects and the strengths and weaknesses of
individual studies considered for reference value
derivation. In the case of ammonia, the database did not
support the combination of estimates across studies. In
future assessments, combining estimates across studies
will be routinely considered.
Calculation of Reference Values and Unit Risks (see pp. 165-166)
18. Describe and justify assumptions and models used.
This step includes review of dosimetry models and
the implications of the models for uncertainty
factors; determination of appropriate points of
departure (such as benchmark dose, no-observed-
adverse-effect level, and lowest observed-adverse-
effect level), and assessment of the analyses that
underlie the points of departure.
19. Provide explanation of the risk-estimation modeling
processes (for example, a statistical or biologic
model fit to the data) that are used to develop a
unit risk estimate.
Implemented as applicable. The rationale for the
selection of the point of departure (a no-observed-
adverse-effect level; NOAEL) for the derivation of the
inhalation reference value for ammonia is transparently
described in Section 2. No modeling was applied in the
derivation of the reference value. An oral reference value
was not derived.
Not applicable. The ammonia assessment concludes that
there is inadequate information to assess the carcinogenic
potential. Therefore, a unit risk estimate for cancer was
not derived.
     This document is a draft for review purposes only and does not constitute Agency policy.
                                  F-7           DRAFT—DO NOT CITE OR QUOTE

-------
                                                           Supplemental Information—Ammonia
        Table F-l. The EPA's implementation of the National Research Council's
        recommendations in the ammonia assessment
       NRC recommendations that EPA is
         implementing in the short term
    Implementation in the ammonia assessment
20. Provide adequate documentation for conclusions
   and estimation of reference values and unit risks.
   As noted by the committee throughout the present
   report, sufficient support for conclusions in the
   formaldehyde draft IRIS assessment is often lacking.
   Given that the development of specific IRIS
   assessments and their conclusions are of interest to
   many stakeholders, it is important that they provide
   sufficient references and supporting documentation
   for their conclusions. Detailed appendixes, which
   might be made available only electronically, should
   be provided when appropriate.
Implemented.  The new template structure that has been
developed in response to the NRC recommendations
provides a clear explanation of the literature search
strategy, study selection criteria, and methods used to
develop the ammonia assessment. It provides for a clear
description of the decisions made in developing the hazard
identification and dose-response analysis. Information
contained in the Preamble and throughout the document
reflects the guidance that has been utilized in developing
the assessment. As recommended, supplementary
information is provided in the accompanying appendices.
        Table F-2. National Research Council recommendations that the EPA is
        generally implementing in the long term
   NRC recommendations that EPA is generally
         implementing in the long term
    Implementation in the ammonia assessment
Weight-of-Evidence Evaluation: Synthesis of Evidence
for Hazard Identification (see p. 165)
1. Review use of existing weight-of-evidence
   guidelines.
2. Standardize approach to using weight-of-evidence
   guidelines.
3. Conduct agency workshops on approaches to
   implementing weight-of-evidence guidelines.
4. Develop uniform language to describe strength of
   evidence on noncancer effects.
5. Expand and harmonize the approach for
   characterizing uncertainty and variability.
6. To the extent possible, unify consideration of
   outcomes around common modes of action rather
   than considering multiple outcomes separately.
As indicated above, Phase 3 of EPA's implementation
plan will incorporate the longer-term recommendations
made by the NRC. On May 16, 2012, EPA announced
(U.S.  EPA, 2012c) that as a part of a review of the IRIS
Program's assessment development process, the NRC will
also review current methods for weight-of-evidence
analyses and recommend approaches for weighing
scientific evidence for chemical hazard identification. In
addition, EPA will hold a workshop on August 26, 2013,
on issues related to weight of evidence to inform future
assessments.
Calculation of Reference Values and Unit Risks (see
pp. 165-166)
7. Assess the sensitivity of derived estimates to model
   assumptions and endpoints selected.  This step
   should include appropriate tabular and graphic
   displays to illustrate the range of the estimates and
   the effect of uncertainty factors on the estimates.
As discussed in Section 1.2, the respiratory system is the
primary and most sensitive target of inhaled ammonia
toxicity. There is some evidence that inhaled ammonia
may be associated with toxicity to target organs other than
the respiratory system, but the evidence for these
associations is weak. Therefore, these endpoints were not
considered appropriate for the development of candidate
or alternative reference values.  In addition, no modeling
was performed in this assessment. Assessing the
sensitivity of the inhalation reference value to model
assumptions and endpoint selection was not possible.
             This document is a draft for review purposes only and does not constitute Agency policy.
                                                F-8             DRAFT—DO NOT CITE OR QUOTE

-------
                                                      Supplemental Information—Ammonia

 1

 2   APPENDIX G.  SUMMARY OF  EXTERNAL PEER

 3   REVIEW AND PUBLIC COMMENTS AND EPA's

 4   DISPOSITION	
 5
 6
 7   RESOLUTION OF PUBLIC COMMENTS ON DRAFT TOXICOLOGICAL
 s   REVIEW (dated June 2012)
 9         The Toxicological Review of Ammonia was released for a 60-day public comment period on
10   June 8, 2012.  Public comments on the assessment were submitted to EPA by the American
11   Chemistry Council (ACC; dated August 6, 2012), the Fertilizer Institute (TFI; dated August 7, 2012),4
12   and a private individual (Unger; dated June 8, 2012). The submission by Unger was a request for
13   information related to a specific site and did not contain comments on the Toxicological Review. A
14   summary of major public comments provided in these submissions and EPA's response to these
15   comments follow. The comments have been synthesized and paraphrased and are organized to
16   follow the order of the Toxicological Review.  The reviewers made several editorial suggestions to
17   clarify specific portions of the text. These changes were incorporated in the document as
18   appropriate and are not discussed further. The full submissions by public commenters are
19   available on the docket at http://www.regulations.gov (Docket ID No. EPA-HQ-ORD-2012-0399).
20
21   Comments on the Preface
22
23   Comment: The ACC recommended that EPA expand the Preface of the Toxicological Review to
24   include:
25       •   All the factors that can prompt a chemical review (e.g., EPA statutory, regulatory, or
26          program-specific implementation needs; availability of new scientific information or
27          methodology that might significantly change the current IRIS information) and list the
28          factors that led to the initiation of the ammonia review (e.g., availability of new studies).
        "American Chemistry Council (ACC). (2012) Re: Request for Public Comment on the EPA's Draft
     Toxicological Review of Ammonia: In Support of the Summary Information in the Integrated Risk Information
     System (IRIS). Docket #EPA-HQ-ORD-2012-0399; FRL-9683-8. Submitted by Kimerbly Wise, Ph.D., Senior
     Director, Chemical Products & Technology Division, ACC, on behalf of the Center for Advancing Risk
     Assessment Science and Policy, managed by ACC. Dated August 6, 2012.

        Fertilizer Institute (TFI). (2012) Re: Comments on external review draft human health assessment titled
     'Toxicological Review of Ammonia: In Support of the Summary Information on the Integrated Risk
     Information System (IRIS)" (EPA/635/R-11/013A). Docket ID No. EPA-HQ-ORD-2012-0399. Submitted by
     William C. Herz, Vice President of Scientific Programs, The Fertilizer Institute. Dated August 7, 2012.

                This document is a draft for review purposes only and does not constitute Agency policy.
                                         G-l               DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1       •   Description of the scope and limitations of an IRIS assessment and how any derived
 2           toxicity values should be used, especially in conjunction with exposure information to
 3           make informed risk management determinations. (ACC, p. 2)
 4
 5   EPA Response: Some of this information is found in the Preface; other information is in the
 6   Preamble. The Preface serves as a brief introduction to the assessment; it is the Assessment
 7   Manager talking directly to the reader. The Preamble, on the other hand, describes the scope of the
 8   IRIS program and the process for developing IRIS assessments, and provides a brief overview of
 9   EPA guidance and methods.
10          Accordingly, the factors that led to the initiation of the ammonia review, including the
11   availability of new studies, are described in the Preface. The Preface also discusses EPA's interest in
12   an assessment of ammonia (e.g., listings under the Comprehensive Environmental Response,
13   Compensation, and Liability Act [CERCLA] and Toxics Release Inventory [TRI]). More general
14   information not specific to ammonia is provided in the Preamble.
15
16   Comment: The ACC stated that the Preface could be improved by including information relating to
17   any cooperative agreements, contracts, or memoranda of understanding that the Agency has in
18   place that may have informed the development of the assessment.  (ACC, p. 2)
19
20   EPA Response: EPA agrees that information relating to cooperative agreements, contracts, or
21   memoranda of understanding is important, and that is why information on the Memorandum of
22   Understanding with the Agency for Toxic Substances and Disease Registry (ATSDR) had been
23   present in the Preface (p. viii) and in the Literature Search Strategy | Study Selection and Evaluation
24   section (p. xxvii) of the public comment draft It has been retained in the external review draft
25
26   Comment: The ACC recommended that the Preface present the  findings of other regulatory
27   agencies and discuss why conclusions and toxicity values in other agency assessments were similar
28   to or different than the draft IRIS assessment. In particular, ACC suggested that it would be useful
29   to explain how the processes used to evaluate ammonia by the ATSDR and EPA differed. (ACC, p. 2)
30
31   EPA Response: Information summarizing other assessments, specifically that of ATSDR (2004],
32   was provided in Table A-l.  This information was provided in the Preface of the public comment
33   draft  The Preface also states that assessments prepared by other health agencies were prepared
34   for different purposes using different methods and could consider only the studies that were
35   available at the time that those assessments were developed. It is beyond the scope of an individual
36   IRIS assessment to provide a general critique of methodological differences across assessment
37   programs  in different agencies at different times.
38
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            G-2                DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1   Comments on the Preamble
 2
 3   Comment: The ACC observed that the Preamble provides an abbreviated view of EPA policies,
 4   guidance, and standard practices that omits critical information and may unduly lead readers to
 5   incorrectly interpret EPA guidance. ACC did not consider it appropriate to use the Preamble as a
 6   means to communicate to the public new criteria, guidance, or approaches that have not been
 7   properly peer reviewed, and stated that the adoption of new approaches should be done through an
 8   open and robust process that involves peer review and stakeholder participation. ACC identified a
 9   number of specific examples of the above.  (ACC, p. 3)
10
11   EPA Response: EPA appreciates the comments on the Preamble.  In response to these comments,
12   revisions were made throughout the Preamble to make sure that this section provides a clear
13   overview of the application of existing Agency guidance and the methods and criteria used in
14   developing IRIS assessments. Among these revisions are:
15       •  Clarification that IRIS assessments  cover the hazard identification and dose-response
16          sections of the risk assessment process
17       •  Inclusion of public meetings as part of the process for IRIS assessment development
18       •  Expansion of the types of human studies considered (to include population-based surveys)
19          when evaluating epidemiological evidence
20       •  Expanded discussion of the evaluation of individual study quality
21       •  Revised discussion of Agency guidance related to evaluating overall weight of evidence,
22          including the use of standard descriptors and consideration of mechanistic information or
23          methodological differences to explain differing results
24       •  Expanded discussion of the use of mechanistic data to identify adverse outcome pathways
25          and modes of action
26       •  Additional discussion of approaches used to derive a point of departure
27       •  Clarification of the Agency practice for applying uncertainty factors to account for human
28          variation
29       •  Inclusion of organ- or system-specific reference values and a corresponding rationale for
30          each of their derivations
31
32   Comments on the Literature Search
33
34   Comment: The ACC recommended improving the transparency of the literature search by including
35   in Figure LS-1 more detailed information regarding the criteria used by EPA to include or exclude
36   studies from consideration in the assessment and a breakdown of the number of studies excluded
37   in each exclusion category, by generating separate figures with study selection criteria for human,
38   animal, and supporting studies, and by explaining what is meant by conducting a literature search
39   using "standard practices." (ACC, p. 9)
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            G-3                DRAFT—DO NOT CITE OR QUOTE

-------
                                                           Supplemental Information—Ammonia
 1
 2   EPA Response: The Literature Search Strategy | Study Selection and Evaluation section, including
 3   Figure LS-1, represents one of EPA's initial efforts to increase the level of detail in and transparency
 4   of the literature search strategy and output. EPA recognizes that documentation of the ammonia
 5   literature search is not fully consistent with a systematic review approach, and is working to more
 6   fully implement systematic review practices in other ongoing assessments. To provide further
 7   details of the ammonia literature search, the search string used in the literature search and other
 8   details of the search strategy were added in a new appendix to the Toxicological Review (Appendix
 9   D, Table D-l). Although additional figures to detail study selection criteria for human, animal, and
10   other supporting studies were not added, the text in the Literature Search Strategy | Study Selection
11   and Evaluation section was expanded to describe study selection in greater detail.  To ensure that
12   all key references on ammonia toxicity have been identified and considered, external peer
13   reviewers will be asked, as part of their charge, to identify any missing studies relevant to the
14   assessment
15
16   Comment: The ACC recommended that the Toxicological Review provide a clear correlation as to
17   how the data (evidence) tables connect to the literature search strategy, and specific information as
18   to how and why studies were selected from the literature search for further consideration. More
19   specifically, the ACC noted that of the 75 human studies identified in the literature search, only
20   three were included in evidence tables.  (ACC, p. 10)
21
22   EPA Response: In general, the more informative studies for evaluating the health effects of chronic
23   exposure to a chemical are carried forward into evidence tables. EPA appreciates the comment on
24   study selection, and the text in the Literature Search Strategy | Study Selection and Evaluation
25   section was expanded to describe the study selection process in more detail, and in particular the
26   study quality considerations that informed study selection. Briefly, six occupational epidemiology
27   studies involving industrial exposure to ammonia (identified in Figure LS-1) are summarized in
28   evidence tables (i.e., Tables 1-1 and 1-6). An additional seven epidemiology studies of workers
29   exposed to ammonia when used as a cleaning product or disinfectant were identified through a
30   literature search update (March 2012-March 2013); documentation of these studies was added to
31   Figure LS-1 and results of the studies were summarized in a new evidence table (Table 1-2).
32   Studies of ammonia-associated effects in livestock farmers (n = 10), controlled-exposure
33   (volunteer) studies involving exposures ranging up to four hours in duration (n = 12), and human
34   case reports (n = 44) were considered less informative than studies of workers exposed to
35   ammonia in industrial settings or through the use of cleaning products and were not included in
36   evidence tables; however, findings from these studies were summarized as supporting evidence in
37   the text of Section 1.1 and in more detail in Appendix E.2. The numbers of studies in Figure LS-1
38   were updated consistent with the updated literature search.
39
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                             G-4                DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1   Comments on the Evidence Tables
 2
 3   Comment: The ACC recommended expanding the evidence tables to include the specific statistical
 4   tests used by study authors to obtain p-values, confidence in exposure measurements (low,
 5   medium, high),  and narrative about the exposure quantification provided in the text.  The ACC also
 6   suggested that the entries in the table entitled, Evidence pertaining to respiratory effects in animals
 7   following inhalation exposure, and the accompanying exposure-response array (Figure 1-1) be
 8   ordered in terms of adversity, occurrence within the mode of action, and/or test species. (ACC, p.
 9   10)
10
11   EPA Response: Evidence tables are used to summarize the design and results of the most
12   informative studies. The evidence table and synthesis text are meant to be complementary, not
13   redundant. To be an effective tool, the entries in an evidence table are focused on information that
14   describes the relationship between the exposure (dose) and an outcome. In general, other
15   information important to understanding the results of individual studies in the context of the
16   available literature for that health endpoint are included in the accompanying synthesis text
17          In the ammonia assessment, the specific statistical tests used by the study authors were
18   identified in study summaries in Appendix E.2 and E.3 when that information was available; these
19   tests were not repeated in the evidence tables.  In a few instances where the name of the statistical
20   test had not been identified in the study summary, the appendix was revised to identify the test.
21   Study evaluation tables for epidemiology studies were added to new Appendix D (Tables D-2, D-3,
22   and D-4); statistical analyses and  additional exposure information that would inform an evaluation
23   of the confidence in exposure measurements was included in these tables and discussed in the
24   Literature Search Strategy  | Study Selection and Evaluation section.  Consistent with the National
25   Research Council (NRC) recommendations to reduce the volume of text and address redundancies,
26   additional narrative on exposure  quantification and confidence in exposure measures was  not
27   added to the evidence tables.
28          The EPA agrees that an appropriate grouping of entries in an evidence table can be helpful
29   in understanding and integrating  the available health effects information. Studies of the respiratory
30   effects of ammonia in Table 1-3 (Evidence pertaining to respiratory effects in animals following
31   inhalation exposure) and the accompanying exposure-response array (Figure 1-1) were organized
32   by location of the effect in the respiratory tract (i.e., lung versus upper respiratory tract) in the
33   public comment draft The available information on ammonia respiratory effects does not support
34   further ordering by level of adversity or mode of action. EPA agrees, however, that more consistent
35   organization by species would be  appropriate.  The order of entries in Tables 1-3 and 1-7 and
36   Figures 1-1 and 1-4 were revised  to provide a more consistent grouping by species.
37
38   Comment: The ACC recommended that the RfC be added to Figure 1-1 to illustrate where the RfC
39   falls relative to the lowest-observable-adverse-effect levels or the no-observed-adverse-effect levels
40   noted in the relevant scientific studies.  (ACC, p. 10)

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                             G-5                DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1
 2   EPA Response: Figure 1-1 is part of the hazard identification for ammonia in Chapter 1, Hazard
 3   Identification, of the Toxicological Review, and is intended to provide a graphical representation of
 4   qualitative evidence of respiratory effects associated with inhalation exposure to ammonia.
 5   Because derivation of the RfC is not presented until Chapter 2, Dose-Response Analysis, the
 6   addition of the RfC to Figure 1-1 would be out of sequence and potentially confusing.
 7
 8   Comments on Hazard Identification
 9
10   Comment: TFI recommended that the discussion of acute gastrointestinal health effects of
11   intentional or accidental ingestion of household cleaning solutions or ammonia inhalant capsules be
12   limited to an appendix or eliminated altogether from the Toxicological Review. (TFI, p. 2)
13
14   EPA Response: EPA agrees that the synthesis of evidence for gastrointestinal effects of ammonia
15   would benefit from additional discussion of the acute nature of the gastrointestinal findings in
16   humans. Therefore, the discussion of acute gastrointestinal health effects of intentional or
17   accidental ingestion of ammonia or ammonia-containing solutions (Section 1.1.2 and Appendix E.2)
18   was revised to provide more context for these findings, i.e., that the acute effects appear to reflect
19   the corrosive properties of ammonia and their relevance to effects associated with chronic low-
20   level exposure to ammonia is unclear.
21
22   Comment: TFI requested that the Hazard Identification section of the Toxicological Review include
23   some qualitative discussion regarding potential confounding factors, such as co-exposure to other
24   ambient chemicals, particulates or dust, that may be associated with ammonia exposure  in urea
25   production areas and in sodium carbonate production areas and a qualitative statement that the
26   exposures and NOAEL are expected to be underestimates of the ammonia inhalation exposure.
27   (TFI, p. 2-3)
28
29   EPA Response: EPA appreciates this comment Consideration of potential confounding was
30   addressed more fully in Tables D-2, D-3, and D-4 on the evaluation of epidemiology studies (see
31   Appendix D), and in text in the Literature Search Strategy | Study Selection and Evaluation section
32   of the external review draft.  Consideration of co-exposure to other agents in the livestock farmer
33   studies was  also addressed in Appendix E and Tables E-7 and E-8. Section 2.2.1 was revised to
34   clarify the rationale for selection of the NOAEL from Holness etal. (1989] as the POD for the
35   ammonia RfC.
36
37   Comment: The ACC stated that the draft assessment needs to provide sufficient detailed
38   information concerning how the ammonia literature was used to derive toxicity values and how a
39   study's strengths or weaknesses were used to inform the weight of evidence.  The ACC
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                             G-6                DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1   recommended that EPA add a table that specifically denotes the strength and weaknesses of a study
 2   and the reasons for excluding studies. (ACC, p. 11)
 o
 J
 4   EPA Response: For epidemiology studies, study evaluation tables were added to a new Appendix D
 5   (Tables D-2, D-3, and D-4); these tables were used to supportEPA's evaluation of the extentto
 6   which a study was considered informative and relevant to the assessment in the section Literature
 7   Search Strategy | Study Selection and Evaluation. Because the animal studies were, in general, from
 8   the older toxicological literature, limited in terms of study design and reporting of results, and not
 9   carried forward for RfC derivation, a table was not necessary to convey the limitations of animal
10   studies. Additional text describing the body of animal toxicology literature was, however, added to
11   the Literature Search Strategy |  Study Selection and Evaluation section.
12
13   Comment: TFI commented that undue emphasis was placed on a handful of recent studies at the
14   expense of a substantive database of studies on the relationship between the effects of ammonia
15   and human health and as such does not provide adequate context for hazard identification.  (TFI, p.
16   2)
17
18   EPA Response: EPA appreciates the comment but wishes to point out that all available human and
19   experimental animal studies were considered in assessing the hazards of ammonia exposure.
20   Based on a study evaluation process described in the Literature Search Strategy | Study Selection
21   and Evaluation section and synthesis of the hazard information in Section 1.1 of the Toxicological
22   Review, EPA concluded that the most informative studies for dose-response analysis were the
23   studies by Holnessetal. fl9891 Rahman etal. f20071 Ballal etal. fl9981 and Alietal. f20011
24   These four studies, which were  published over the last 2 decades, provided data most suitable for
25   dose-response analysis.
26
27   Comments on Dose-Response Analysis
28
29   Comment: The ACC observed that although the narrative on page 2-2 of the draft assessment
30   indicates that the evidence for associations of ammonia with toxicity to target organs other than the
31   respiratory system is weak, Figure 2-1 does not give any indication as to why the immune system
32   effects or other systemic effects were not selected for dose-response analysis. (ACC, p. 11)
33
34   EPA Response: EPA appreciates the comment  The original purpose of this figure was to compare
35   graphically effect levels for ammonia across a range of target organs, including the respiratory
36   system, liver, kidney, heart, eyes, and the immune system. As discussed in Section 1.2.1, however,
37   the hazard potential for the immune system and other systemic targets is weak compared to the
38   hazard potential for the respiratory system. Because Figure 2-1 does not capture the strength of
39   evidence for a given organ system and because the available literature identifies only respiratory
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            G-7               DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia
 1   effects as a hazard from inhaled ammonia, EPA recognizes that the information presented in this
 2   figure may be misleading. Accordingly, this figure was removed from the Toxicological Review.
 o
 J
 4   Comment: The ACC commented that the selection of the critical study [Holnessetal., 1989] was
 5   not clearly supported because [1] no statistically significant differences were noted between the
 6   control and exposed groups for respiratory irritation, (2) no changes in lung function were
 7   observed between control and exposed groups, and (3) no relationship between level or duration of
 8   ammonia exposure and lung function changes was demonstrated. The ACC also noted that the
 9   Holness etal. [1989] study was often mischaracterized as part of a body of literature that
10   consistently demonstrates an increased prevalence of symptoms. (ACC, p. 11}
11
12   EPA Response: EPA recognizes that the Holness etal. [1989] study did not find a significant
13   association between level or duration of exposure to ammonia and respiratory symptoms or
14   changes in lung function under the conditions of exposure in that plant The choice of Holness etal.
15   [1989] as the principal study was made only in the context of the entire database, including studies
16   of workers exposed to higher workplace concentrations of ammonia than in the Holness et al.
17   [1989] study, where a relatively high level of control of exposures resulted in relatively low
18   ammonia levels in the plant Specifically, the study by Holness etal. [1989] was selected as the
19   principal study only with support from the findings from three other cross-sectional occupational
20   studies by Rahman etal. [2007]. Ali etal. [2001]. and Ballal etal. [1998].  Holnessetal. [1989] was
21   chosen as the principal study over Rahman et al. [2007] and Ballal etal. [1998] because confidence
22   in the exposure measures used by Holness etal. [1989] were higher, because Holness etal. [1989]
23   evaluated both respiratory symptoms and lung function, and because the estimate of the NOAEL
24   from Holness etal. T19891 was higher.  Alietal. f20011 a companion study to Ballal etal. fl9981
25   examined lung function in workers in only one of the two plants studied by Ballal etal. [1998] and
26   was less useful for RfC derivation. Clarifying text was added to Section 2.2.1 of the Toxicological
27   Review.
28          EPA regrets that there were a couple of instances where the Holness etal. [1989] study was
29   incorrectly cited as one  of the studies that reported an increased prevalence of respiratory
30   symptoms associated with ammonia exposure. Those citations have been removed.
31
32   Comment: TFI requested that EPA select either 50 ppm (35.4 mg/m3] or 25 ppm (17.7 mg.m3] as
33   the POD for derivation of the RfC (as opposed to 8.8 mg/m3], or that the actual range of data in the
34   "highest occupational exposure" category from the Holness etal. [1989] study  be retrieved to
35   determine a representative and justifiable POD value from the referenced study. TFI also suggested
36   that the NOAEL selected for RfC derivation should be  consistent with the Acute Exposure Guideline
37   Level (AEGL]-1 value  of 21 mg/m3.  (TFI, p. 3-4]
38
39   EPA Response: The rationale for selecting 8.8 mg/m3 from the Holness etal. [1989] study as the
40   NOAEL was expanded in Section 2.2.1.

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            G-8                DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia
 1          In general, an acute emergency response value, such as an AEGL-1, is not a scientifically
 2   supported basis for deriving an RfC, defined as an estimate of a continuous inhalation exposure to
 3   the human population (including sensitive subgroups) that is likely to be without an appreciable
 4   risk of deleterious effects during a lifetime. AEGLs are applicable to emergency exposure periods
 5   from 10 minutes to 8 hours. The AEGL-1 of 21 mg/m3 for ammonia is based on a study in which 2
 6   of 6 human volunteers experienced faint irritation (confined only to the upper respiratory tract)
 7   after exposure to 21 mg/m3 for 10 minutes. Thus, the AEGL-1 for ammonia does not provide a
 8   scientifically sound point of departure for the chronic RfC.
 9
10   Other Key Issues:
11
12   Comment: The ACC noted that the discussion of endogenous production of ammonia was not
13   adequate and considered the rationale used to justify setting an RfC at a level equivalent to the
14   internal human breath level to be unclear. The ACC recommended that clear justification for setting
15   an RfC that is within the range of natural human breath levels be provided.  (ACC, p. 11)
16
17   EPA Response: The RfC is not at the level of internal human breath.  The RfC is several fold above
18   ammonia concentrations in breath exhaled from the nose and trachea. Concentrations in breath
19   exhaled from the nose and trachea are expected to correlate with levels at the alveolar interface of
20   the lung or in the tracheo-bronchial region.  These concentrations are thought to be more relevant
21   to understanding systemic levels of ammonia than ammonia in breath exhaled from the mouth or
22   oral cavity, which largely reflect production of ammonia via bacterial degradation of food protein.
23   This information was provided in Section 2.2.4 and as a key issue in the Executive Summary. To
24   ensure that this issue is adequately addressed in the Toxicological Review, external peer reviewers
25   will be asked, as  part of their charge, whether the discussion of endogenous ammonia in the
26   Toxicological Review is scientifically supported and clearly described.
27
28
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                             G-9               DRAFT—DO NOT CITE OR QUOTE

-------
                                                        Supplemental Information—Ammonia
     REFERENCES FOR APPENDICES
 3
 4
 5          Multiple references published in the same year by the same author(s) have been assigned a
 6   letter (e.g., 1986a, 1986b) based on author(s), year, and title. Those same letters have been
 7   retained for the appendices.
 8
 9   Abramovicz. I. (1925). Ocular injury caused by liquid ammonia. Br J Ophthalmol 9: 241-242.
10   ACGIH (American Conference of Governmental Industrial Hygienists). (2001). Ammonia [TLV/BEI].
11          In Documentation of the threshold limit values and biological exposure indices. Cincinnati,
12          OH.
13   AIChE (American Institute of Chemical Engineers). (1999). Ammonia H3N. In TE Daubert; RP
14          Banner; HM Sibul; CC Stebbins (Eds.), Physical and thermodynamic properties of pure
15          chemicals: Evaluated process design data. Philadelphia, PA: Taylor & Francis.
16   Ali. BA: Ahmed. HO: Ballal. SG: Albar. AA. (2001).  Pulmonary function of workers exposed to
17          ammonia: a study in the Eastern Province of Saudi Arabia. IntJ Occup Environ Health 7: 19-
18          22.
19   Anderson, DP: Beard, CW: Hanson, RP. (1964). The adverse effects of ammonia on chickens
20          including resistance to infection with Newcastle disease virus. Avian Dis 8: 369-379.
21   Appelman, LM: ten Berge, WF: Reuzel, PGT. (1982). Acute inhalation toxicity study of ammonia in
22          rats with variable exposure periods. Am Ind Hyg Assoc J 43: 662-665.
23          http://dx.doi.org/10.1080/15298668291410387
24   Arif. AA: Delclos. GL. (2012). Association between cleaning-related chemicals and work-related
25          asthma and asthma symptoms among healthcare professionals. Occup Environ Med 69: 35-
26          40. http://dx.doi.org/10.1136/oem.2011.064865
27   Arwood. R: Hammond. I: Ward.  GG. (1985). Ammonia inhalation. J Trauma 25: 444-447.
28   Atkinson. SA: Anderson. GH: Bryan. MH. (1980). Human milk: comparison of the nitrogen
29          composition in milk from mothers of premature and full-term infants. Am J Clin Nutr 33:
30          811-815.
31   ATSDR (Agency for Toxic Substances and Disease Registry). (2004). Toxicological profile for
32          ammonia [ATSDR Tox Profile]. Atlanta, GA: U.S. Department of Health and Human Services,
33          Public Health Service, http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=ll&tid=2
34   Auerbach, C: Robson, TM. (1947). Tests of chemical substances for mutagenic action. Proc Roy Soc
35          EdinbBBiol 62: 284-291.
36   Bacom, A: Yanosky, M. (2010). E-mails dated June 22, 2010, from Michael Yanosky, Drager Safety
37          Inc., Technical Support Detection Products to Amber Bacom, SRC, Inc., contractor to NCEA,
38          ORD, U.S. EPA. Available online
39   Ballal. SG: Ali. BA: Albar. AA: Ahmed. HO: al-Hasan. AY. (1998). Bronchial asthma in two chemical
40          fertilizer producing factories in eastern Saudi Arabia. IntJ Tuberc Lung Dis 2: 330-335.
41   Barrow, CS: Steinhagen, WH. (1980). NH3 concentrations in the expired air of the rat: importance to
42          inhalation toxicology. ToxicolAppl Pharmacol 53: 116-121.
43   Barzel, US: Towsey, I. (1969). The effects of chronic acid and alkali administration on bone turnover
44          in adult rats. Clin Sci (Lond) 36: 517-524.
45   Beare, ID: Wilson, RS: Marsh, RT. (1988). Ammonia burns of the eye: an old weapon in new hands
46          [Editorial]. Br Med J (Clin Res Ed) 296: 590.

                This document is a draft for review purposes only and does not constitute Agency policy.
                                           R-l                DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia

 1   Bell, AW: Kennaugh, TM: Battaglia, FC: Meschia, G. (1989). Uptake of amino acids and ammonia at
 2          mid-gestation by the fetal lamb. Q J Exp Physiol 74: 635-643.
 3   Bernstein, IL: Bernstein, PI. (1989). Reactive airways disease syndrome (RADS) after exposure to
 4          toxic ammonia fumes [Abstract]. J Allergy Clin Immunol 83: 173.
 5   Betterton, EA. (1992). Henry's law constants of soluble and moderately soluble organic gases:
 6          Effects in aqueous phase chemistry. In JO Nriagu (Ed.), Gaseous pollutants: Characterization
 7          and cycling (pp. 1-50). New York, NY: Wiley.
 8   Bhat, MR: Ramaswamy, C. (1993). Effect of ammonia, urea and diammonium phosphate (DAP) on
 9          lung functions in fertilizer plant workers. Indian J Physiol Pharmacol 37: 221-224.
10   Bishop. TM: Verlander. TW: Lee. HW: Nelson. RD: Weiner. AT: Handlogten. ME: Weiner. ID. (2010).
11          Role of the Rhesus glycoprotein, Rh B glycoprotein, in renal ammonia excretion. Am J
12          Physiol Renal Physiol 299: F1065-F1077. http://dx.doi.org/10.1152/aiprenal.00277.2010
13   Boshier, PR: Marczin, N: Hanna, GB. (2010). Repeatability of the measurementof exhaled volatile
14          metabolites using selected ion flow tube mass spectrometry. J Am Soc Mass Spectrom 21:
15          1070-1074. http://dx.doi.0rg/10.1016/i.iasms.2010.02.008
16   Boyd, EM: MacLachlan, ML: Perry, WF. (1944). Experimental ammonia gas poisoning in rabbits and
17          cats. J IndHygToxicol 26: 29-34.
18   Brautbar, N: Wu, MP: Richter, ED. (2003). Chronic ammonia inhalation and interstitial pulmonary
19          fibrosis: A case report and review of the literature [Review]. Arch Environ Health 58: 592-
20          596. http://dx.doi.Org/10.3200/AEOH.58.9.592-596
21   Broderson. JR.: Lindsey. TR: Crawford. IE. (1976). The role of environmental ammonia in respiratory
22          mycoplasmosis of rats. AmJPatholSS: 115-130.
23   Brown. RH: Duda. GD: Korkes. S: Handler. P. (1957). A colorimetric micromethod for determination
24          of ammonia; the ammonia content of rat tissues and human plasma. Arch Biochem Biophys
25          66: 301-309. http://dx.doi.org/10.1016/S0003-986ir57180005-8
26   Buckley, LA: Tiang, XZ: Tames, RA: Morgan, KT: Barrow, CS. (1984). Respiratory tract lesions induced
27          by sensory irritants atthe RD50 concentration. Toxicol Appl Pharmacol 74: 417-429.
28          http://dx.doi.org/10.1016/0041-008Xf84190295-3
29   Burns, TR: Mace, ML: Greenberg, SD: Tachimczyk, TA.  (1985). Ultrastructure of acute ammonia
30          toxicity in the human lung. Am J Forensic Med Pathol 6: 204-210.
31   Caplin, M. (1941). Ammonia-gas poisoning: forty-seven cases in a London shelter. Lancet 26: 95-96.
32   Caste 11, DO: Moore, EW. (1971). Ammonia absorption from the human colon. The role of nonionic
33          diffusion. Gastroenterology 60: 33-42.
34   Chaung, H, -C: Hsia, L, -C: Liu, S, -H. (2008). The effects of vitamin A supplementation on the
35          production of hypersensitive inflammatory mediators of ammonia-induced airways of pigs.
36          Food and Agricultural Immunology 19: 283-291.
37          http://dx.doi.org/10.1080/09540100802471546
38   Choudat. D: Goehen. M: Korobaeff. M: Boulet. A:  Dewitte. ID: Martin. MH. (1994). Respiratory
39          symptoms and bronchial reactivity among pig and dairy farmers. Scand J Work Environ
40          Health 20: 48-54.
41   Christesen, HB. (1995). Prediction of complications following caustic ingestion in adults. Clin
42          Otolaryngol Allied Sci 20: 272-278.
43   Close, LG: Catlin, FI: Cohn, AM. (1980). Acute and chronic effects of ammonia burns of the
44          respiratory tract. Eur Arch Otorhinolaryngol 106: 151-158.
45   Cole. TT: Cotes. IE: Tohnson. GR: Martin Hde. V: Reed.  TW: Saunders. Ml. (1977). Ventilation, cardiac
46          frequency and pattern of breathing during exercise in men exposed to 0-chlorobenzylidene
47          malononitrile (CS) and ammonia gas in low concentrations. Q J Exp Physiol 62: 341-351.


                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            R-2                DRAFT—DO NOT CITE OR QUOTE

-------
                                                          Supplemental Information—Ammonia

 1   Conn, HO. (1972). Studies of the source and significance of blood ammonia. IV. Early ammonia
 2          peaks after ingestion of ammonium salts. Yale J Biol Med 45: 543-549.
 3   Coon, RA: Tones, RA: Jenkins, LI, Tr: Siegel, 1. (1970). Animal inhalation studies on ammonia, ethylene
 4          glycol, formaldehyde, dimethylamine, and ethanol. Toxicol Appl Pharmacol 16: 646-655.
 5          http://dx.doi.org/10.1016/0041-008xf70j90069-4
 6   Cormier, Y: Israel-Assayag, E: Racine, G: Duchaine, C. (2000). Farming practices and the respiratory
 7          health risks of swine confinement buildings. Eur Respir J15: 560-565.
 8   Couturier, Y: Barbotin, M: Bobin, P: Derrien, IP. (1971). [3 cases of toxic lung caused by ammonia
 9          vapors and sulfureted hydrogen]. Bulletin Soc Med Afr Noire LangFr 16: 250-252.
10   Crook, B: Robertson, IF: Glass, SA: Botheroyd, EM: Lacey, 1: Topping, MD. (1991). Airborne dust,
11          ammonia, microorganisms, and antigens in pig confinement houses and the respiratory
12          health of exposed farm workers. Am IndHygAssocJ 52: 271-279.
13          http://dx.doi.org/10.1080/15298669191364721
14   Curtis. SE: Anderson. CR: Simon. 1: Tensen. AH: Day. PL: Kelley. KW. (1975). Effects of aerial
15          ammonia, hydrogen sulfide and swine-house dust on rate of gain and respiratory-tract
16          structure in swine. J Anim Sci 41: 735-739.
17   da Fonseca-Wollheim. F. (1995). The influence of pH and various anions on the distribution of
18          NH4+ in human blood. Eur J Clin Chem Clin Biochem 33: 289-294.
19   Dalhamn. T. (1963). Effect of ammonia alone and combined with carbon particles on ciliary activity
20          in the rabbit trachea in vivo, with studies of the absorption capacity of the nasal cavity. Air
21          Water Pollut 7: 531-539.
22   Davies, BM: Yudkin, 1. (1952). Studies in biochemical adaptation; the origin or urinary ammonia as
23          indicated by the effect of chronic acidosis and alkalosis on some renal enzymes in the rat
24          Biochem J 52: 407-412.
25   Davies, S: Spanel, P: Smith, D. (1997). Quantitative analysis of ammonia on the breath of patients in
26          end-stage renal failure. Kidney Int 52: 223-228.
27   de la Hoz, RE: Schlueter, DP: Rom, WN. (1996). Chronic lung disease secondary to ammonia
28          inhalation injury: a reporton three cases. Am J Ind Med 29: 209-214.
29          http://dx.doi.org/10.1002/rSICni097-0274ri99602129:2<:209::AID-
30          ATIM12>:3.0.CO:2-7
31   Dean. TA. (1985). Lange's handbook of chemistry. New York, NY: McGraw-Hill.
32   Demerec. M: Bertani. G: Flint. 1. (1951). A survey of chemicals for mutagenic action on E coli. Am Nat
33          85:119-135.
34   DeSanto. IT: Nagomi. W: Liechty. EA: Lemons. TA. (1993). Blood ammonia concentration in cord
35          blood during pregnancy. Early Hum Dev 33: 1-8.
36   Diack. C: Bois. FY. (2005). Pharmacokinetic-pharmacodynamic models for categorical toxicity data.
37          Regul Toxicol Pharmacol 41: 55-65. http://dx.doi.Org/10.1016/i.yrtph.2004.09.007
38   Diekman. MA: Scheldt. AB: Sutton. AL: Green. ML: Clapper. TA: Kelly. DT: Van Alstine. WG. (1993).
39          Growth and reproductive performance, during exposure to ammonia, of gilts afflicted with
40          pneumonia and atrophic rhinitis. Am J Vet Res 54: 2128-2131.
41   Dilli, D: Bostanci, I: Tiras, U: Hatipoglu, N: Dallar, Y. (2005). A non-accidental poisoning with
42          ammonia in adolescence. Child Care Health Dev 31: 737-739.
43          http://dx.doi.0rg/10.llll/i.1365-2214.2005.00552.x
44   Diskin, AM: Spanel, P: Smith, D. (2003). Time variation of ammonia, acetone, isoprene and ethanol
45          in breath: a quantitative SIFT-MS study over 30 days. Physiol Meas 24: 107-119.
46   Dodd, KT: Gross, DR. (1980). Ammonia inhalation toxicity in cats: A study of acute and chronic
47          respiratory dysfunction. Arch Environ Health 35:6-14.


                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            R-3               DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia

 1   Doig, PA: Willoughby, RA. (1971). Response of swine to atmospheric ammonia and organic dust J
 2          Am VetMedAssoc 159: 1353-1361.
 3   Done. SH: Chennells. DT: Gresham. AC: Williamson. S: Hunt. B: Taylor. LL: Bland. V: Tones. P:
 4          Armstrong. D: White. RP: Demmers. TG: Teer. N: Wathes. CM. (2005). Clinical and
 5          pathological responses of weaned pigs to atmospheric ammonia and dust Vet Rec 157: 71-
 6          80.
 7   Donham, KT: Cumro, D: Reynolds, ST: Merchant, TA. (2000). Dose-response relationships between
 8          occupational aerosol exposures and cross-shift declines of lung function in poultry workers:
 9          recommendations for exposure limits. J Occup Environ Med 42: 260-269.
10   Donham. KT: Reynolds. ST: Whitten. P: Merchant. TA: Burmeister. L: Popendorf. WT. (1995).
11          Respiratory dysfunction in swine production facility workers: dose-response relationships
12          of environmental exposures and pulmonary function. Am J Ind Med 27: 405-418.
13   Douglas. RB: Coe. IE. (1987). The relative sensitivity of the human eye and lung to irritant gases.
14          Ann Occup Hyg 31: 265-267.
15   Drummond.  TG: Curtis. SE: Simon. 1: Norton. HW. (1980). Effects of aerial ammonia on growth and
16          health of young pigs. JAnimSci 50: 1085-1091.
17   Dumas. 0: Donnay. C: Heederik. DT: Hery. M: Choudat. D: Kauffmann. F: Le Moual. N. (2012).
18          Occupational exposure to cleaning products and asthma in hospital workers. Occup Environ
19          Med  69: 883-889. http://dx.doi.org/10.1136/oemed-2012-100826
20   Dworkin. MS: Patel. A: Fennell. M: Vollmer. M: Bailey. S: Bloom. 1: Mudahar. K: Lucht. R. (2004). An
21          outbreak of ammonia poisoning from chicken tenders served in a school lunch. J Food Prot
22          67:1299-1302.
23   Egle, TL, Tr. (1973). Retention of inhaled acetone and ammonia in the dog. Am Ind Hyg Assoc J 34:
24          533-539. http://dx.doi.org/10.1080/0002889738506894
25   Fazekas, IG. (1939). Die durch ammoniak hervorgerufene experimentelle
26          nebennierenhypertrophie. Endokrinologie 21: 315-337.
27   FDA. Direct food substances affirmed as generally recognized as safe (GRAS): Ammonium
28          hydroxide, 21 CFR§ 184.1139 (2011a).
29          http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=l 84.1139
30   FDA. Substances generally recognized as safe: General purpose food additives: Ammonium
31          hydroxide. 21 CFR5 582.1139 f2011bl
32          http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=582.1139
33   Ferguson. WS: Koch. WC: Webster. LB: Gould. JR. (1977). Human physiological response and
34          adaption to ammonia. J Occup Environ Med 19: 319-326.
35   Ferris, BG. (1978). Epidemiology standardization project (American Thoracic Society). Am Rev
36          RespirDis 118: 1-120.
37   Flessner, MF: Wall, SM: Knepper, MA. (1992). Ammonium and bicarbonate transport in rat outer
38          medullary collecting ducts. Am J Physiol 262: F1-F7.
39   Flury. KE: Dines. DE: Rodarte. TR: Rodgers. R. (1983). Airway obstruction due to inhalation of
40          ammonia. Mayo Clin Proc 58: 389-393.
41   Gaafar, H: Girgis, R: Hussein, M: el-Nemr, F. (1992). The effect of ammonia on the respiratory nasal
42          mucosaof mice. Ahistological and histochemical study. Acta Otolaryngol 112: 339-342.
43   Gamble, MR: Clough, G. (1976). Ammonia build-up in animal boxes and its effect on rat tracheal
44          epithelium. Lab Anim 10: 93-104.
45   Gay, WM: Crane, CW: Stone, WD. (1969). The metabolism of ammonia in liver disease: a comparison
46          of urinary data following oral and intravenous loading of [15N]ammonium lactate. Clin Sci
47          (Lond) 37: 815-823.


                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            R-4                DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia

 1   George, A: Bang, RL: Lari, AR: Gang, RK: Kanjoor, TR. (2000). Liquid ammonia injury. Burns 26: 409-
 2          413.
 3   Gilbert, GT. (1988). Acute ammonia intoxication 37 years after ureterosigmoidostomy. South Med J
 4          81: 1443-1445.
 5   Giroux, M: Bremont, F: Salles, IP: Rev, E: Delia Massa, IP: Ferrieres, J. (2002). Exhaled NH3 and
 6          excreted NH4+ in children in unpolluted or urban environments. Environ Int 28:  197-202.
 7   Green, SM: Machin, R: Cresser, MS. (2008). Effect of long-term changes in soil chemistry induced by
 8          road salt applications on N-transformations in roadside soils. Environ Pollut 152: 20-31.
 9          http://dx.doi.0rg/10.1016/i.envpol.2007.06.005
10   Gustin, P: Urbain, B: Prouvost, IF: Ansay, M. (1994). Effects of atmospheric ammonia on pulmonary
11          hemodynamics and vascular permeability in pigs: interaction with endotoxins. Toxicol Appl
12          Pharmacol 125: 17-26. http://dx.doi.org/10.1006/taap.1994.1044
13   Guyton, AC. (1981). The body fluids and kidneys. In Textbook of medical physiology (6th ed.).
14          Philadelphia, PA: WB Saunders Company.
15   Hamid, HA: El-Gazzar, RM. (1996). Effect of occupational exposure to ammonia on enzymatic
16          activities of catalase and mono amine oxidase. J Egypt Public Health Assoc 71: 465-475.
17   Han. KH: Croker. BP: Clapp. WL: Werner. D: Sahni. M: Kim. 1: Kim. HY: Handlogten. ME: Weiner. ID.
18          (2006). Expression of the ammonia transporter, Rh C glycoprotein, in normal and neoplastic
19          human kidney. J Am Soc Nephrol 17:  2670-2679.
20          http://dx.doi.org/10.1681/ASN.2006020160
21   Handlogten. ME: Hong. SP: Westhoff. CM: Weiner. ID. (2005). Apical ammonia transport by the
22          mouse inner medullary collecting duct cell (mIMCD-3). Am J Physiol Renal Physiol 289:
23          F347-F358. http://dx.doi.org/10.1152/aiprenal.00253.2004
24   Hata. M: Yamazaki. Y: Ueda. T: Kato. T: Kohli. Y: Fujiki. N. (1994). Influence of ammonia solution on
25          gastric mucosa and acetic acid induced ulcer in rats. Eur J Histochem 38:  41-52.
26   Hatton. DV: Leach. CS: Beaudet. AL: Dillman.  RO: Pi Ferrante. N. (1979). Collagen breakdown and
27          ammonia inhalation. Arch Environ Occup Health 34: 83-87.
28   Hauguel, S: Challier, 1C: Cedard, L: Olive, G. (1983). Metabolism of the human placenta perfused in
29          vitro: glucose transfer and utilization, 02 consumption, lactate and ammonia production.
30          Pediatr Res 17: 729-732. http://dx.doi.org/10.1203/00006450-198309000-00009
31   Heederik, D: van Zwieten, R: Brouwer, R. (1990). Across-shift lung function changes among pig
32          farmers. Am JInd Med 17: 57-58.
33   Helmers, S: Top, FH, Sr: Knapp, LW, Jr. (1971). Ammonia injuries in agriculture. J Iowa Med Soc 61:
34          271-280.
35   Highman. VN. (1969). Early rise in intraocular pressure after ammonia burns. Br Med J1: 359-360.
36   Hilado, CJ: Casey, CT: Furst, A. (1977). Effect of ammonia on Swiss albino mice. J Combustion Toxicol
37          4:385-388.
38   Holness, PL: Purdham, IT: Nethercott, TR. (1989). Acute and chronic respiratory effects of
39          occupational exposure to ammonia. AIHA J 50: 646-650.
40          http://dx.doi.org/10.1080/15298668991375308
41   Holzman. IR: Lemons. TA: Meschia. G: Battaglia. FC. (1977). Ammonia production by the pregnant
42          uterus (39868). Proc Soc Exp Biol Med 156: 27-30.
43   Holzman. IR: Philipps. AF: Battaglia. FC. (1979). Glucose metabolism, lactate, and ammonia
44          production by the human placenta in vitro. Pediatr Res 13: 117-120.
45   Huizenga. TR: Tangerman. A: Gips. CH. (1994). Determination of ammonia in biological fluids
46          [Review]. Ann Clin Biochem 31 (Pt 6): 529-543.
47   Ihrig. A: Hoffmann. J: Triebig. G. (2006). Examination of the influence of personal traits and

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            R-5               DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia

 1          habituation on the reporting of complaints at experimental exposure to ammonia. Int Arch
 2          Occup Environ Health 79: 332-338. http://dx.doi.org/10.1007/s00420-005-0042-y
 3   IPCS (International Programme on Chemical Safety). (1986). Environmental health criteria:
 4          Ammonia. (EHC 54). Geneva, Switzerland: World Health Organization.
 5          http://www.inchem.org/documents/ehc/ehc/ehc54.htm
 6   Tarudi, NI: Golden, B. (1973). Ammonia eye injuries. J Iowa Med Soc 63: 260-263.
 7   Tohnson, RL: Gilbert, M: Block, SM: Battaglia, FC. (1986). Uterine metabolism of the pregnant rabbit
 8          under chronic steady-state conditions. Am J Obstet Gynecol 154: 1146-1151.
 9   Tozwik, M: Tozwik, M: Pietrzycki, B: Chojnowski, M: Teng, C: Tozwik, M: Battaglia, FC. (2005).
10          Maternal and fetal blood ammonia concentrations in normal term human pregnancies. Biol
11          Neonate 87: 38-43. http://dx.doi.org/10.1159/000081702
12   Tozwik, M: Teng, C: Meschia, G: Battaglia, FC. (1999). Contribution of branched-chain amino acids to
13          uteroplacental ammonia production in sheep. Biol Reprod 61: 792-796.
14   Kane, LE: Barrow, CS: Alarie, Y. (1979). A short-term test to predict acceptable levels of exposure to
15          airborne sensory irritants. Am Ind Hyg Assoc J 40: 207-229.
16          http://dx.doi.org/10.1080/15298667991429516
17   Kapeghian, 1C: Tones, AB: Waters, IW. (1985). Effects of ammonia on selected hepatic microsomal
18          enzyme activity in mice. Bull Environ Contain Toxicol 35: 15-22.
19   Kapeghian, 1C: Mincer, HH: Tones, AB: Verlangieri, AT: Waters, IW. (1982). Acute inhalation toxicity
20          of ammonia in mice. Bull Environ Contam Toxicol 29: 371-378.
21   Kass, I: Zamel, N: Dobry, CA: Holzer, M. (1972). Bronchiectasis following ammonia burns of the
22          respiratory tract: A review of two cases [Review]. Chest 62: 282-285.
23   Katayama, K. (2004). Ammonia metabolism and hepatic encephalopathy. Hepatology Research SOS:
24          73-80. http://dx.doi.0rg/10.1016/i.hepres.2004.08.013
25   Kawano, S: Tsujii, M: Fusamoto, H: Sato, N: Kamada, T. (1991). Chronic effect of intragastric
26          ammonia on gastric mucosal structures in rats. Dig Dis Sci 36: 33-38.
27   Kearney, DT: Hubbard, T: Putnam, D. (2002). Breath ammonia measurement in Helicobacter pylori
28          infection. Dig Dis Sci 47: 2523-2530.
29   Keiding. S: S0rensen. M: Bender. D: Munk. PL: Ott. P: Vilstrup. H. (2006). Brain metabolism of 13N-
30          ammonia during acute hepatic encephalopathy in cirrhosis measured by positron emission
31          tomography. Hepatology 43: 42-50. http://dx.doi.org/10.1002/hep.21001
32   Keiding. S: S0rensen. M: Munk. PL: Bender. D. (2010). Human (13)N-ammonia PET studies: The
33          importance of measuring (13)N-ammonia metabolites in blood. Metab Brain Dis 25: 49-56.
34          http://dx.doi.org/10.1007/sll011-010-9181-2
35   Kennedy, SM: Le Moual, N: Choudat, D: Kauffmann, F. (2000). Development of an asthma specific job
36          exposure matrix and its application in the epidemiological study of genetics and
37          environment in asthma (EGEA). Occup Environ Med 57: 635-641.
38   Klein, I: Olson, KR: McKinney,  HE. (1985). Caustic injury from household ammonia. Am J Emerg Med
39          3:320.
40   Klendshoj, NC: Rejent, TA. (1966). Tissue levels of some poisoning agents less frequently
41          encountered. J Forensic Sci 11: 75-80.
42   Lalic, H: Djindjic-Pavicic, M: Kukuljan, M. (2009). Ammonia intoxication on workplace-case report
43          and a review of literature [Review]. Coll Antropol 33: 945-949.
44   Landahl, HP: Herrmann, RG. (1950). Retention of vapors and gases in the human nose and lung.
45          Arch Environ Occup Health 1: 36-45.
46   Larson, T: Frank, R: Covert, D: Holub, D: Morgan, M. (1980). The chemical neutralization of inhaled
47          sulfuric acid aerosol. Am J Ind Med 1: 449-452.

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            R-6               DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia

 1   Larson, TV: Covert, PS: Frank, R: Charlson, RT. (1977). Ammonia in the human airways:
 2          Neutralization of inspired acid sulfate aerosols. Science 197: 161-163.
 3   Leduc, D: Gris, P: Lheureux, P: Gevenois, PA: De Vuyst, P: Yernault, 1C.  [1992]. Acute and long term
 4          respiratory damage following inhalation of ammonia. Thorax 47: 755-757.
 5   Lee. HS: Chan. CC: Tan. KT: Cheong. TH: Ghee. CBE: Wang. YT. (1993).  Burnisher's asthma - a case
 6          due to ammonia from silverware polishing. Singapore Med J 34: 565-566.
 7   Lee. HW: Verlander. TW: Bishop. TM: Igarashi. P: Handlogten. ME: Weiner. ID. (2009). Collecting
 8          duct-specific Rh C glycoprotein deletion alters basal and acidosis-stimulated renal ammonia
 9          excretion. Am J Physiol Renal Physiol 296: F1364-F1375.
10          http://dx.doi.org/10.1152/aiprenal.90667.2008
11   Lee. HW: Verlander. TW: Bishop. TM: Nelson. RD: Handlogten. ME: Weiner. ID. (2010). Effect of
12          intercalated cell-specific Rh C  glycoprotein deletion on basal and metabolic acidosis-
13          stimulated renal ammonia excretion. Am J Physiol Renal Physiol 299: F369-F379.
14          http://dx.doi.org/10.1152/aiprenal.00120.2010
15   Lemiere. C: Begin. D: Camus. M: Forget. A: Boulet. LP: Gerin. M. (2012). Occupational risk factors
16          associated with work-exacerbated asthma in Quebec. Occup Environ Med 69: 901-907.
17          http://dx.doi.org/10.1136/oemed-2012-100663
18   Levy. DM: Divertie. MB: Litzow. TT: Henderson. TW. (1964). Ammonia burns of the face and
19          respiratory tract JAMA 190: 873-876.
20   Li, WL: Pauluhn, 1.  (2010). Comparative assessment of the sensory irritation potency in mice and
21          rats nose-only exposed to ammonia in dry and humidified atmospheres. Toxicology 276:
22          135-142. http://dx.doi.0rg/10.1016/i.tox.2010.07.020
23   Lide. DR. (2008). CRC handbook of chemistry and physics. In DR Lide  (Ed.), (88th ed.). Boca Raton,
24          FL: CRC  Press.
25   Lina, BAR: Kuijpers, MHM. (2004). Toxicity and carcinogenicity of acidogenic or alkalogenic diets in
26          rats; effects of feeding NH(4)C1, KHCO(3) or KC1. Food ChemToxicol 42: 135-153.
27   Lobasov, M: Smirnov, F. (1934). Nature of the action of chemical agents on mutational process in
28          Drosophila melanogaster: II. The effect of ammonia on the occurrence of lethal
29          transgenations. C R Biol 3: 174-176.
30   Lopez. GP: Dean. BS:  Krenzelok. EP. (1988). Oral-exposure to ammonia inhalants: A report of 8
31          cases [Abstract]. Vet Hum Toxicol 30: 350.
32   Luschinsky. HL.  (1951). The activity of glutaminase in the human placenta. Arch Biochem Biophys
33          31:132-140.
34   MacEwen. ID: Theodore. I: Vernot. EH. (1970). Human exposure to EEL concentrations of
35          monomethylhydrazine. In Proceedings of the first annual conference on environmental
36          toxicology. Wright-Patterson Air Force Base, OH: Aerospace Medical Research Laboratory.
37   MacEwen. ID: Vernot. EH. (1972). Toxic hazards research unit annual report: 1972. (AMRL-TR-72-
38          62). Wright-Patterson Air Force Base, OH: Aerospace Medical  Research Laboratory.
39   Manninen, A: Anttila, S: Savolainen, H. (1988). Rat metabolic adaptation to ammonia inhalation.
40          ProcSocExp Biol Med 187: 278-281.
41   Manninen, ATA: Savolainen, H. (1989). Effect of short-term ammonia inhalation on selected amino
42          acids in  rat brain. Pharmacol Toxicol 64: 244-246.
43   Manolis, A. (1983). The diagnostic potential of breath analysis [Review]. Clin Chem 29: 5-15.
44   Mayan. MH: Merilan. CP. (1972). Effects of ammonia inhalation on respiration rate of rabbits. J Anim
45          Sci 34: 448-452.
46   McGuiness. R. (1969). Ammonia in the eye [Letter]. Br Med J1: 575.
47   Medina-Ramon. M: Zock. IP:  Kogevinas. M: Sunyer. 1: Basagana. X: Schwartz. I: Burge. PS: Moore. V:

                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            R-7                DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia

 1          Anto, TM. (2006). Short-term respiratory effects of cleaning exposures in female domestic
 2          cleaners. Eur Respir J 27: 1196-1203. http://dx.doi.org/10.1183/09031936.06.00085405
 3   Medina-Ramon, M: Zock, IP: Kogevinas, M: Sunyer, 1: Torralba, Y: Borrell, A: Burgos, F: Anto, TM.
 4          (2005). Asthma, chronic bronchitis, and exposure to irritant agents in occupational
 5          domestic cleaning: A nested case-control study. Occup Environ Med 62: 598-606.
 6          http://dx.doi.org/10.1136/oem.2004.017640
 7   Melbostad, E: Eduard, W. (2001). Organic dust-related respiratory and eye irritation in Norwegian
 8          farmers. Am JInd Med 39: 209-217.
 9   Meschia. G: Battaglia. FC: Hay. WW: Sparks. TW. (1980). Utilization of substrates by the ovine
10          placenta in vivo. Fed Proc 39: 245-249.
11   Monsen, ER. (1987). The journal adopts SI units for clinical laboratory values. J Am Diet Assoc 87:
12          356-358.
13   Monso, E: Riu, E: Radon, K: Magarolas, R: Danuser, B: Iversen, M: Morera, 1: Nowak, D.  (2004).
14          Chronic obstructive pulmonary disease in never-smoking animal farmers working inside
15          confinement buildings. Am J Ind Med 46: 357-362. http://dx.doi.org/10.1002/aiim.20077
16   Mossberg, SM: Ross, G. (1967). Ammonia movement in the small intestine: Preferential transport by
17          the ileum. J Clin Invest 46: 490-498. http://dx.doi.org/10.1172/TCI105551
18   Muntwyler, E: lacobellis, M: Griffin, GE. (1956). Kidney glutaminase and carbonic anhydrase
19          activities and renal electrolyte excretion in rats. Am J Physiol 184: 83-90.
20   Nelson. PL: Cox. MM. (2008). Amino acid oxidation and the production of urea. In DL Nelson; AL
21          Lehninger; MM Cox (Eds.), Lehninger principles of biochemistry (5th ed., pp. 680, 683). New
22          York, NY: W.H. Freeman & Co.
23   NIOSH (National Institute for Occupational Safety and Health). (1979). NIOSH manual of analytical
24          methods: Second edition, volume 5. (DHEW (NIOSH) Publication No. 79-141).  Cincinnati,
25          OH.
26   NIOSH (National Institute for Occupational Safety and Health). (2010). NIOSH pocket guide to
27          chemical hazards: Ammonia, http://www.cdc.gov/niosh/npg/npgd0028.html
28   NLM (National Institutes of Health, National Library of Medicine). (2012). ChemlDplus: Ammonia.
29          Bethesda, MD: National Library of Medicine, http://chem.sis.nlm.nih.gov/chemidplus/
30   Norwood, DM: Wainman, T: Lioy, PI: Waldman, TM. (1992). Breath ammonia depletion and its
31          relevance to acidic aerosol exposure studies. Arch Environ Occup Health 47: 309-313.
32          http://dx.doi.org/10.1080/00039896.1992.9938367
33   NRC (National Research Council). (2008). Acute exposure guideline levels for selected airborne
34          chemicals: Volume 6. Washington, DC: The National Academies Press.
35          http://www.nap.edu/catalog.php7record id=12018
36   NRC (National Research Council). (2011). Review of the Environmental Protection Agency's draft
37          IRIS assessment of formaldehyde. Washington, DC: National Academies Press.
38          http://www.nap.edu/catalog/13142.html
39   O'Connor. EA: Parker. MO: McLeman. MA: Demmers. TG: Lowe. 1C: Cui. L: Davey. EL: Owen. RC:
40          Wathes, CM: Abeyesinghe, SM. (2010). The impact of chronic environmental stressors on
41          growing pigs, Sus scrofa (part 1): Stress physiology, production and play behaviour. Animal
42          4: 1899-1909. http://dx.doi.org/10.1017/S1751731110001072
43   O'Kane,  GT. (1983). Inhalation of ammonia vapour: A report on the management of eight patients
44          during the acute stages. Anaesthesia 38: 1208-1213.
45   O'Neil. Ml: Heckelman. PE: Koch. CB: Roman. KT. (2006). Ammonia. In The Merck index: An
46          encyclopedia of chemicals, drugs, and biologicals (14th ed.). Whitehouse Station, NJ: Merck
47          & Co., Inc.
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            R-8                DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia

 1   OSHA. Table Z-l: Limits for air contaminants. 29 CFR § 1910.1000 [2006].
 2          http://www.osha.gov/pls/oshaweb/owadisp.show document?p  table=STANDARDS&p id=
 3          9992
 4   Osmond. AH: Tallents. CT. [1968]. Ammonia attacks [Letter]. BrMedJ 3: 740.
 5   Ota, Y: Hasumura, M: Okamura, M: Takahashi, A: Ueda, M: Onodera, H: Imai, T: Mitsumori, K: Hirose,
 6          M. (2006). Chronic toxicity and carcinogenicity of dietary administered ammonium sulfate
 7          in F344 rats. Food Chem Toxicol 44: 17-27. http://dx.doi.Org/10.1016/i.fct.2005.06.001
 8   Petrova, M: Diamond, T: Schuster, B: Dalton, P. (2008). Evaluation of trigeminal sensitivity to
 9          ammonia in asthmatics and healthy human volunteers. Inhal Toxicol 20: 1085-1092.
10          http://dx.doi.org/10.1080/08958370802120396
11   Pirjavec, A: Kovic, I: Lulic, I: Zupan, Z. (2009). Massive anhydrous ammonia injury leading to lung
12          transplantation. J Trauma 67: E93-E97. http://dx.doi.org/10.1097/TA.Ob013e31817fd93f
13   Pitts, RF. (1971). The role of ammonia production and excretion in regulation of acid-base balance
14          [Review]. N Engl T Med 284: 32-38. http://dx.doi.org/10.1056/NETM197101072840110
15   Preller, L: Heederik, D: Boleij, ISM: Vogelzang, PFT: Tielen, MTM. (1995). Lung function and chronic
16          respiratory symptoms of pig farmers: Focus on exposure to endotoxins and ammonia and
17          use of disinfectants. Occup Environ Med 52: 654-660.
18   Price. S: Watts. 1C. (2008). Ammonia gas incident [Letter]. Anaesthesia 63: 894-895.
19          http://dx.doi.0rg/10.llll/i.1365-2044.2008.05625.x
20   Prudhomme. 1C: Shusterman. DT: Blanc. PP. (1998). Acute-onset persistent olfactory deficit
21          resulting from multiple overexposures to ammonia vapor at work. J Am Board Fam Pract
22          11:66-69.
23   Oin. C: Foreman. RD: Farber. IP. (2007a). Afferent pathway and neuromodulation of superficial and
24          deeper thoracic spinal neurons receiving noxious pulmonary inputs in rats. Auton Neurosci
25          131: 77-86. http://dx.doi.Org/10.1016/i.autneu.2006.07.007
26   Qin, C: Foreman, RD: Farber, IP. (2007b). Inhalation of a pulmonary irritant modulates activity of
27          lumbosacral spinal neurons receiving colonic input in rats. Am J Physiol Regul Integr Comp
28          Physiol 293: R2052-R2058. http://dx.doi.org/10.1152/aipregu.00154.2007
29   Rahman, MH: Bratveit, M: Moen, BE. (2007). Exposure to ammonia and acute respiratory effects in a
30          urea fertilizer factory. Int J Occup Environ Health 13: 153-159.
31   Read, AT. (1982). lonization constants of aqueous ammonia from 25 to 250C and to 2000 bar.
32          Journal of Solution Chemistry 11: 649-664. http://dx.doi.org/10.1007/BF00650397
33   Rejniuk, VL: Schafer, TV: Ivnitsky, IT. (2008). Ammonia potentiates the lethal effect of ethanol on
34          rats. Bull Exp Biol Med 145: 741-743.
35   Rejniuk. VL: Schafer. TV: Ovsep'yan. RV:  Ivnitsky. IT. (2007). Effect of atmospheric ammonia on
36          mortality rate of rats with barbiturate intoxication. Bull Exp Biol Med 143: 692-704.
37   Remesar. X: Arola. L: Palou. A: Alemany.  M. (1980). Activities of enzymes involved in amino-acid
38          metabolism in developing rat placenta. Eur J Biochem 110: 289-293.
39          http://dx.doi.0rg/10.llll/i.1432-1033.1980.tb04867.x
40   Reynolds. ST: Donham. KT: Whitten. P: Merchant. TA: Burmeister. LF: Popendorf. WT. (1996).
41          Longitudinal evaluation of dose-response relationships for environmental exposures and
42          pulmonary function in swine production workers. Am J Ind Med 29: 33-40.
43          http://dx.doi.org/10.1002/(SICni097-0274(199601)29:l<:33::AID-ATIM5>:3.0.CO:2-#
44   Richard, D: Bouley, G: Boudene, C. (1978a). [Effects of ammonia gas continuously inhaled by rats
45          and mice]. Bull Europ Physiol Resp 14: 573-582.
46   Richard, D: Touany, TM: Boudene, C. (1978b). [Acute inhalation toxicity of ammonia in rabbits]. C R
47          Acad Sci Hebd Seances Acad Sci D 287: 375-378.
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            R-9               DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia

 1   Robertson, TL: Kolmakova-Partensky, L: Miller, C. (2010). Design, function and structure of a
 2          monomeric C1C transporter [Letter]. Nature 468: 844-847.
 3          http://dx.doi.org/10.1038/nature09556
 4   Rosenbaum, AM: Walner, PL: Dunham, ME: Holinger, LD. (1998). Ammonia capsule ingestion
 5          causing upper aerodigestive tract injury. Otolaryngol Head Neck Surg 119: 678-680.
 6   Rosenfeld, M. (1932). [Experimental modification of mitosis by ammonia]. Archiv fur
 7          experimentelle Zellforschung, besonders Gewebeziichtung 14: 1-13.
 8   Rubaltelli, FF: Formentin, PA. (1968). Ammonia nitrogen, urea and uric acid blood levels in the
 9          mother and in both umbilical vessels at delivery. Biol Neonat 13: 147-154.
10          http://dx.doi.org/10.1159/000240142
11   Sadasivudu, B: Radha Krishna Murthy, C. (1978). Effects of ammonia on monoamine oxidase and
12          enzymes of GABA metabolism in mouse brain. Arch Physiol Biochem 86: 67-82.
13   Sadasivudu, B: Rao, TI: Murthy, CR. (1979). Chronic metabolic effects of ammonia in mouse brain.
14          Arch Physiol Biochem 87: 871-885.
15   Schaerdel. AD: White. WT: Lang. CM: Dvorchik. BH: Bohner. K. (1983). Localized and systemic effects
16          of environmental ammonia in rats. J Am Assoc Lab Anim Sci 33: 40-45.
17   Schmidt. FM: Vaittinen. 0: Metsala. M: Lehto. M: Forsblom. C: Groop. PH: Halonen. L. (2013).
18          Ammonia in breath and emitted from skin. J Breath Res 7: 017109.
19          http://dx.doi.Org/10.1088/1752-7155/7/l/017109
20   Shimizu. H: Suzuki. Y: Takemura. N: Goto. S: Matsushita. H. (1985). The results of microbial
21          mutation test for forty-three industrial chemicals. Sangyo Igaku 27: 400-419.
22   Sigurdarson. ST: O'Shaughnessy. PT: Watt. TA: Kline. IN. (2004). Experimental human exposure to
23          inhaled grain dust and ammonia: towards a model of concentrated animal feeding
24          operations. Am J Ind Med 46: 345-348. http://dx.doi.org/10.1002/ajim.20055
25   Silver. SD: McGrath. FP. (1948). A comparision of acute toxicities of ethylene imine and ammonia to
26          mice. J Ind Hyg Toxicol 30: 7-9.
27   Silverman. L: Whittenberger. TL: Muller. 1. (1949). Physiological response of man to ammonia in low
28          concentrations. J Ind Hyg Toxicol 31:  74-78.
29   Sioblom. E: Hoier. I: Kulling. PE: Stauffer. K: Suneson. A: Ludwigs. U. (1999). A placebo-controlled
30          experimental study of steroid inhalation therapy in ammonia-induced lung injury. J Toxicol
31          Clin Toxicol 37: 59-67.
32   Slot, GMT. (1938). Ammonia gas burns: An account of six cases. Lancet 10: 1356-1357.
33   Smeets. MA: Bulsing. PI: van Rooden. S: Steinmann. R: de Ru. TA: Ogink. NW: van Thriel. C: Dalton.
34          PH. (2007). Odor and irritation thresholds for ammonia: a comparison between static and
35          dynamic olfactometry. Chem Senses 32: 11-20. http://dx.doi.org/10.1093/chemse/bjl031
36   Smith, D: Spanel, P: Davies, S. (1999). Trace gases in breath of healthy volunteers when fasting and
37          after a protein-calorie meal: a preliminary study. J Appl Physiol 87: 1584-1588.
38   Smith, D: Wang, T: Pysanenko, A: Spanel, P. (2008). A selected ion flow tube mass spectrometry
39          study of ammonia in mouth- and nose-exhaled breath and in the oral cavity. Rapid Commun
40          Mass Spectrom 22: 783-789. http://dx.doi.org/10.1002/rcm.3434
41   S0rensen. M: Munk. PL: Keiding. S. (2009). Backflux of ammonia from brain to blood in human
42          subjects with and without hepatic encephalopathy. Metab Brain Dis 24: 237-242.
43          http://dx.doi.org/10.1007/sll011-008-9126-l
44   Sotiropoulos. G: Kilaghbian. T: Dougherty. W: Henderson. SO. (1998). Cold injury from pressurized
45          liquid ammonia: a report of two cases. J Emerg Med 16: 409-412.
46   Souba. WW. (1987). Interorgan ammonia metabolism in health and disease: a surgeon's view
47          [Review]. JPEN J Parenter Enteral Nutr 11: 569-579.


                 This document is a draft for review purposes only and does not constitute Agency policy.
                                           R-10               DRAFT—DO NOT CITE  OR QUOTE

-------
                                                         Supplemental Information—Ammonia

 1   Spanel, P: Dryahina, K: Smith, D. (2007a). Acetone, ammonia and hydrogen cyanide in exhaled
 2          breath of several volunteers aged 4-83 years. J Breath Res 1: 011001.
 3          http://dx.doi.0rg/10.1088/1752-7155/l/l/011001
 4   Spanel, P: Dryahina, K: Smith, D. (2007b). The concentration distributions of some metabolites in
 5          the exhaled breath of young adults. J Breath Res 1: 026001.
 6          http://dx.doi.Org/10.1088/1752-7155/l/2/026001
 7   Stabenau, JR.; Warren, KS: Rail, DP. (1959). The role of pH gradient in the distribution of ammonia
 8          between blood and cerebrospinal fluid, brain and muscle. J Clin Invest 38: 373-383.
 9          http://dx.doi.org/10.1172/TCI103811
10   Stombaugh, DP: Teague, HS: Roller, WL. (1969). Effects of atmospheric ammonia on the pig. J Anim
11          Sci 28: 844-847.
12   Stroud, S. (1981). Ammonia inhalation - a case report Grit Care Nurse 1: 23-26.
13   Summerskill. WH: Wolpert. E. (1970). Ammonia metabolism in the gut Am J Clin Nutr 23: 633-639.
14   Sundblad, BM: Larsson, BM: Acevedo, F: Ernstgard, L: Tohanson, G: Larsson, K: Palmberg, L. (2004).
15          Acute respiratory effects of exposure to ammonia on healthy persons. Scand J Work Environ
16          Health 30: 313-321.
17   Suzuki, H: Mori, M: Suzuki, M: Sakurai, K: Miura, S: Ishii, H. (1997). Extensive DNA damage induced
18          by monochloramine in gastric cells. Cancer Lett 115: 243-248.
19   Suzuki. H: Seto. K: Mori. M: Suzuki. M: Miura. S: Ishii. H. (1998). Monochloramine induced DNA
20          fragmentation in gastric cell line MKN45. Am J Physiol 275: G712-G716.
21   Takagaki. G: Berl. S: Clarke. DP: Purpura. DP: Waelsch. H.  (1961). Glutamic acid metabolism in brain
22          and liver during infusion with ammonia labelled with nitrogen-15. Nature 189: 326.
23   Taplin, GV: Chopra, S: Yanda, RL: Elam, D. (1976). Radionuclidic lung-imaging procedures in the
24          assessment of injury due to ammonia inhalation. Chest 69: 582-586.
25   Tepper, IS: Weiss, B: Wood, RW. (1985). Alterations in behavior produced by inhaled ozone or
26          ammonia. Toxicol Sci 5: 1110-1118.
27   Toth, B. (1972). Hydrazine, methylhydrazine and methylhydrazine sulfate carcinogenesis in Swiss
28          mice. Failure of ammonium hydroxide to interfere in the development of tumors. Int J
29          Cancer 9:109-118.
30   Tsujii, M; Kawano, S; Tsuji, S; Fusamoto, H; Kamada, T;  Sato, N. (1992a). Mechanism of gastric
31          mucosal damage induced by ammonia. Gastroenterology 102: 1881-1888.
32   Tsuiii. M: Kawano. S: Tsuii. S: Nagano. K: Ito. T: Hayashi. N: Fusamoto. H: Kamada. T: Tamura. K.
33          (1992b). Ammonia: a possible promotor in Helicobacter pylori-related gastric
34          carcinogenesis. Cancer Lett 65: 15-18.
35   Tsujii, M: Kawano, S: Tsuji, S: Ito, T: Nagano, K: Sasaki, Y: Hayashi, N: Fusamoto, H: Kamada, T.
36          (1993). Cell kinetics of mucosal atrophy in rat stomach induced by long-term
37          administration of ammonia. Gastroenterology 104: 796-801.
38   Tsujii, M: Kawano, S: Tsuji, S: Takei, Y: Tamura, K: Fusamoto, H: Kamada, T. (1995). Mechanism for
39          ammonia-induced promotion of gastric carcinogenesis in rats. Carcinogenesis 16: 563-566.
40   Turner, C:  Spanel, P: Smith, D. (2006). A longitudinal study of ammonia, acetone and propanol in the
41          exhaled breath of 30 subjects using selected ion flow tube mass spectrometry, SIFT-MS.
42          Physiol Meas 27: 321-337. http://dx.doi.Org/10.1088/0967-3334/27/4/001
43   U.S. Congress. (2011). Consolidated Appropriations Act, 2012. (Pub. L. No. 112-74; 125 STAT. 786).
44          112th U.S. Congress. http://www.gpo.gov/fdsys/pkg/PLAW-112publ74/pdf/PLAW-
45          112publ74.pdf
46   U.S. EPA (U.S. Environmental Protection Agency). (1988). Recommendations for and documentation
47          of biological values for use in risk assessment [EPA Report]. (EPA/600/6-87/008).


                 This document is a draft for review purposes only and does not constitute Agency policy.
                                           R-ll               DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia

 1          Cincinnati, OH. http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=34855
 2   U.S. EPA (U.S. Environmental Protection Agency). (2012a). Advances in inhalation gas dosimetry for
 3          derivation of a reference concentration (rfc) and use in risk assessment [EPA Report].
 4          (EPA/600/R-12/044). Washington, DC.
 5          http://cfpub.epa.gov/ ncea/cfm/recordisplay.cfm?deid=2 44650
 6   U.S. EPA (U.S. Environmental Protection Agency). (2012b). Benchmark dose technical guidance.
 7          (EPA/100/R-12/001). Washington, DC: Risk Assessment Forum.
 8          http://www.epa.gov/raf/publications/pdfs/benchmark dose guidance.pdf
 9   U.S. EPA (U.S. Environmental Protection Agency). (2012c). EPA announces NAS'review of IRIS
10          Assessment development process. Available online at
11          http://yosemite.epa.gOV/opa/admpress.nsf/0/lce2a7875daf093485257a000054df5470pe
12          nDocument
13   Urbain. B: Gustin. P: Prouvost. IF: Ansay. M. (1994). Quantitative assessment of aerial ammonia
14          toxicity to the nasal mucosa by use of the nasal lavage method in pigs. Am J Vet Res 55:
15          1335-1340.
16   van de Poll. MC: Ligthart-Melis. GC: Olde Damink. SW: van Leeuwen. PA: Beets-Tan. RG: Deutz. NE:
17          Wigmore, SI: Soeters, PB: Dejong, CH. (2008). The gut does not contribute to systemic
18          ammonia release in humans without portosystemic  shunting. Am J Physiol  Gastrointest
19          Liver Physiol 295: G760-G765. http://dx.doi.org/10.1152/aipgi.00333.2007
20   Van Slyke. DP: Phillips. RA: Hamilton. PB. (1943). Glutamine as source material for urinary
21          ammonia. J Biol Chem 150: 481-482.
22   Verberk. MM. (1977). Effects of ammonia in volunteers. Int Arch Occup Environ Health 39: 73-81.
23          http://dx.doi.org/10.1007/BF00380887
24   Verschueren. K. (2001). Ammonia. In Handbook of environmental data on organic chemicals. New
25          York, NY: John Wiley & Sons.
26   Vizcaya. D: Mirabelli. MC: Anto. TM: Orriols. R: Burgos.  F: Arjona. L: Zock. IP. (2011). A workforce-
27          based study of occupational exposures and asthma symptoms in cleaning workers. Occup
28          Environ Med 68: 914-919. http://dx.doi.org/10.1136/oem.2010.063271
29   Vollmuth. TA: Schlesinger. RB. (1984). Measurement of respiratory tract ammonia  in the rabbit and
30          implications to sulfuric acid inhalation studies. Toxicol Sci 4: 455-464.
31   Walton. M. (1973). Industrial ammonia gassing. Occup Environ Med 30: 78-86.
32   Wands. RC.  (1981). Alkaline materials. In GD Clayton; FE Clayton (Eds.), Patty's industrial hygiene
33          and toxicology: Volume 2B: Toxicology (3rd rev. ed., pp. 3045-3070). New York, NY: John
34          Wiley & Sons.
35   Ward, K: Murray, B: Costello, GP. (1983). Acute and long-term pulmonary sequelae of acute
36          ammonia inhalation. Ir Med J 76: 279-281.
37   Wason, S: Stephan, M: Breide, C. (1990). Ingestion of aromatic ammonia 'smelling salts' capsules
38          [Letter]. Am J Dis Child 144: 139-140.
39          http://dx.doi.org/10.1001/archpedi.1990.02150260017009
40   Weatherby, TH. (1952). Chronic toxicity of ammonia fumes by inhalation. Proc Soc Exp Biol Med 81:
41          300-301.
42   Weiser, JR.; Mackenroth, T. (1989). Acute inhalatory mass ammonia intoxication with fatal course.
43          Exp Pathol 37: 291-295.
44   White. CE: Park. MS: Renz. EM: Kim. SH: Ritenour. AE:  Wolf.  SE: Cancio. LC. (2007).  Burn center
45          treatment of patients with severe anhydrous ammonia injury: case reports and literature
46          review [Review]. J Burn Care Res 28: 922-928.
47          http://dx.doi.org/10.1097/BCR.Ob013e318159a44e
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                            R-12               DRAFT—DO NOT CITE OR QUOTE

-------
                                                         Supplemental Information—Ammonia

 1    Whittaker, AG: Love, S: Parkin, TD: Duz, M: Hughes, KT. (2009). Stabling causes a significant increase
 2           in the pH ofthe equine airway. Equine VetJ 41: 940-943.
 3           http://dx.doi.org/10.2746/042516409X474383
 4    Yadav, IS: Kaushik, VK. (1997). Genotoxic effect of ammonia exposure on workers in a fertilizer
 5           factory. Indian J Exp Biol 35: 487-492.
 6    Yang, GY: Tominack, RL: Deng, IF. (1987). An industrial mass ammonia exposure. Vet Hum Toxicol
 7           29:476-477.
 8    Zejda. IE: Barber. E: Dosman. TA: Olenchock. SA: McDuffie. HH: Rhodes. C: Hurst. T. (1994).
 9           Respiratory health status in swine producers relates to endotoxin exposure in the presence
10           of low dust levels. JOccupMed 36: 49-56.
11    Zock. IP: Plana. E: Tarvis. D: Anto. TM: Kromhout. H: Kennedy. SM: Kunzli. N: Villani. S: Olivieri. M:
12           Toren. K: Radon. K: Sunyer. 1: Dahlman-Hoglund. A: Norback. D: Kogevinas. M. (2007). The
13           use of household cleaning sprays and adult asthma: An international longitudinal study. Am
14           J Respir Grit Care Med 176: 735-741. http://dx.doi.org/10.1164/rccm.200612-17930C
15    Zock. IP: Vizcaya. D: Le Moual. N. (2010). Update on asthma  and cleaners [Review]. Curr Opin
16           Allergy Clin Immunol 10: 114-120. http://dx.doi.org/10.1097/ACI.Ob013e32833733fe
17
18
                 This document is a draft for review purposes only and does not constitute Agency policy.
                                           R-13               DRAFT—DO NOT CITE OR QUOTE

-------