*>EPA
EPA/690/R-23/001F | November 2022 | FINAL
United States
Environmental Protection
Agency
Provisional Peer-Reviewed Toxicity Values for
Sodium and Potassium Salts of Inorganic Phosphates
(Multiple CASRNs)
PRO1*
supERFU[\|D
U.S. EPA Office of Research and Development
Center for Public Health and Environmental Assessment
-------
A United $ta»s
Environmental Protection
%#UI r% Agency
EPA/690/R-23/001F
November 2022
https://www.epa.gov/pprtv
Provisional Peer-Reviewed Toxicity Values for
Sodium and Potassium Salts of Inorganic Phosphates
(Multiple CASRNs)
Center for Public Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
CHEMICAL MANAGERS
Daniel D. Petersen, MS, PhD, DABT, ATS, ERT
Center for Public Health and Environmental Assessment, Cincinnati, OH
Allison L. Phillips, PhD
Center for Public Health and Environmental Assessment, Cincinnati, OH
Kathleen Newhouse, MS, DABT
Center for Public Health and Environmental Assessment, Washington, DC
DRAFT DOCUMENT PREPARED BY
SRC, Inc.
7502 Round Pond Road
North Syracuse, NY 13212
PRIMARY INTERNAL REVIEWERS
Joyce Donohue, PhD
Office of Water, Washington, DC
Jay Zhao, PhD, DABT
Center for Public Health and Environmental Assessment, Cincinnati, OH
PRIMARY EXTERNAL REVIEWERS
Eastern Research Group, Inc.
110 Hartwell Avenue
Lexington, MA 02421-3136
PPRTV PROGRAM MANAGEMENT
Teresa L. Shannon
Center for Public Health and Environmental Assessment, Cincinnati, OH
J. Phillip Kaiser, PhD, DABT
Center for Public Health and Environmental Assessment, Cincinnati, OH
Questions regarding the content of this PPRTV assessment should be directed to the U.S. EPA
Office of Research and Development (ORD) Center for Public Health and Environmental
Assessment (CPHEA) website at https://ecomments.epa.gov/pprtv.
iii Na/K Salts of Inorganic Phosphates
-------
TABLE OF CONTENTS
COMMONLY USED ABBREVIATIONS AND ACRONYMS v
BACKGROUND 1
QUALITY ASSURANCE 1
DISCLAIMERS 2
QUESTIONS REGARDING PPRTVs 2
1. INTRODUCTION 3
2. REVIEW OF POTENTIALLY RELEVANT DATA (NONCANCER AND CANCER) 12
2.1. HUMAN STUDIES 21
2.1.1. Oral Exposures 21
2.1.2. Inhalation Exposures 61
2.2. ANIMAL STUDIES 61
2.2.1. Oral Exposures 61
2.2.2. Inhalation Exposures 79
2.3. OTHER DATA (SHORT-TERM TESTS, OTHER EXAMINATIONS) 79
2.3.1. Genotoxicity 80
2.3.2. Supporting Animal Studies 85
2.3.3. Metabolism/Toxicokinetics 95
2.3.4. Mode-of-Action/Mechanistic Studies 96
3. DERIVATION 01 PROVISIONAL VALUES 98
3.1. DERIVATION OF PROVISIONAL REFERENCE DOSES 98
3.1.1. Derivation of the Subchronic Provisional Reference Dose 100
3.1.2. Derivation of the Chronic Provisional Reference Dose 104
3.1.3. Consideration of Human Data 107
3.2. DERIVATION OF PROVISIONAL REFERENCE CONCENTRATIONS 109
3.2.1. Derivation of the Subchronic Provisional Reference Concentration 109
3.2.2. Derivation of the Chronic Provisional Reference Concentration 109
3.3. SUMMARY OF NONCANCER PROVISIONAL REFERENCE VALUES 110
3.4. CANCER WEIGHT-OF-EVIDENCE DESCRIPTOR 110
3.5. DERIVATION OF PROVISIONAL CANCER RISK ESTIMATES Ill
iv Na/K Salts of Inorganic Phosphates
-------
COMMONLY USED ABBREVIATIONS AND ACRONYMS
a2u-g
alpha 2u-globulin
IVF
in vitro fertilization
ACGIH
American Conference of Governmental
LC50
median lethal concentration
Industrial Hygienists
LD50
median lethal dose
AIC
Akaike's information criterion
LOAEL
lowest-observed-adverse-effect level
ALD
approximate lethal dosage
MN
micronuclei
ALT
alanine aminotransferase
MNPCE
micronucleated polychromatic
AR
androgen receptor
erythrocyte
AST
aspartate aminotransferase
MOA
mode of action
atm
atmosphere
MTD
maximum tolerated dose
ATSDR
Agency for Toxic Substances and
NAG
7V-acetyl-P-D-glucosaminidase
Disease Registry
NCI
National Cancer Institute
BMC
benchmark concentration
NOAEL
no-observed-adverse-effect level
BMCL
benchmark concentration lower
NTP
National Toxicology Program
confidence limit
NZW
New Zealand White (rabbit breed)
BMD
benchmark dose
OCT
ornithine carbamoyl transferase
BMDL
benchmark dose lower confidence limit
ORD
Office of Research and Development
BMDS
Benchmark Dose Software
PBPK
physiologically based pharmacokinetic
BMR
benchmark response
PCNA
proliferating cell nuclear antigen
BUN
blood urea nitrogen
PND
postnatal day
BW
body weight
POD
point of departure
CA
chromosomal aberration
PODadj
duration-adjusted POD
CAS
Chemical Abstracts Service
QSAR
quantitative structure-activity
CASRN
Chemical Abstracts Service registry
relationship
number
RBC
red blood cell
CBI
covalent binding index
RDS
replicative DNA synthesis
CHO
Chinese hamster ovary (cell line cells)
RfC
inhalation reference concentration
CL
confidence limit
RfD
oral reference dose
CNS
central nervous system
RGDR
regional gas dose ratio
CPHEA
Center for Public Health and
RNA
ribonucleic acid
Environmental Assessment
SAR
structure-activity relationship
CPN
chronic progressive nephropathy
SCE
sister chromatid exchange
CYP450
cytochrome P450
SD
standard deviation
DAF
dosimetric adjustment factor
SDH
sorbitol dehydrogenase
DEN
diethylnitrosamine
SE
standard error
DMSO
dimethylsulfoxide
SGOT
serum glutamic oxaloacetic
DNA
deoxyribonucleic acid
transaminase, also known as AST
EPA
Environmental Protection Agency
SGPT
serum glutamic pyruvic transaminase,
ER
estrogen receptor
also known as ALT
FDA
Food and Drug Administration
SSD
systemic scleroderma
FEVi
forced expiratory volume of 1 second
TCA
trichloroacetic acid
GD
gestation day
TCE
trichloroethylene
GDH
glutamate dehydrogenase
TWA
time-weighted average
GGT
y-glutamyl transferase
UF
uncertainty factor
GSH
glutathione
UFa
interspecies uncertainty factor
GST
g 1 ut a t h i o nc - V-1 ra n s fc ra sc
UFC
composite uncertainty factor
Hb/g-A
animal blood-gas partition coefficient
UFd
database uncertainty factor
Hb/g-H
human blood-gas partition coefficient
UFh
intraspecies uncertainty factor
HEC
human equivalent concentration
UFl
LOAEL-to-NOAEL uncertainty factor
HED
human equivalent dose
UFS
subchronic-to-chronic uncertainty factor
i.p.
intraperitoneal
U.S.
United States of America
IRIS
Integrated Risk Information System
WBC
white blood cell
Abbreviations and acronyms not listed on this page are defined upon first use in the
PPRTV document.
v Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
PROVISIONAL PEER REVIEWED TOXICITY VALUES FOR SODIUM AND
POTASSIUM SALTS OF INORGANIC PHOSPHATES (MULTIPLE CASRNS)
BACKGROUND
A Provisional Peer-Reviewed Toxicity Value (PPRTV) is defined as a toxicity value
derived for use in the Superfund program. PPRTVs are derived after a review of the relevant
scientific literature using established U.S. Environmental Protection Agency (U.S. EPA)
guidance on human health toxicity value derivations.
The purpose of this document is to provide support for the hazard and dose-response
assessment pertaining to chronic and subchronic exposures to substances of concern, to present
the major conclusions reached in the hazard identification and derivation of the PPRTVs, and to
characterize the overall confidence in these conclusions and toxicity values. It is not intended to
be a comprehensive treatise on the chemical or toxicological nature of this substance.
Currently available PPRTV assessments can be accessed on the U.S. EPA's PPRTV
website at https://www.epa.gov/pprtv. PPRTV assessments are eligible to be updated on a 5-year
cycle and revised as appropriate to incorporate new data or methodologies that might impact the
toxicity values or affect the characterization of the chemical's potential for causing adverse
human-health effects. Questions regarding nomination of chemicals for update can be sent to the
appropriate U.S. EPA eComments Chemical Safety website at
https://ecomments.epa.gov/chemicalsafetv/.
QUALITY ASSURANCE
This work was conducted under the U.S. EPA Quality Assurance (QA) program to ensure
data are of known and acceptable quality to support their intended use. Surveillance of the work
by the assessment managers and programmatic scientific leads ensured adherence to QA
processes and criteria, as well as quick and effective resolution of any problems. The QA
manager, assessment managers, and programmatic scientific leads have determined under the
QA program that this work meets all U.S. EPA quality requirements. This PPRTV was written
with guidance from the CPHEA Program Quality Assurance Project Plan (PQAPP), the QAPP
titled Program Quality Assurance Project Plan (PQAPP) for the Provisional Peer-Reviewed
Toxicity Values (PPRTVs) and Related Assessments/Documents (L-CPAD-0032718-QP), and the
PPRTV development contractor QAPP titled Quality Assurance Project Plan—Preparation of
Provisional Toxicity Value (PTV) Documents (L-CPAD-0031971-QP). As part of the QA
system, a quality product review is done prior to management clearance. A Technical Systems
Audit may be performed at the discretion of the QA staff.
All PPRTV assessments receive internal peer review by at least two CPHEA scientists
and an independent external peer review by at least three scientific experts. The reviews focus on
whether all studies have been correctly selected, interpreted, and adequately described for the
purposes of deriving a provisional reference value. The reviews also cover quantitative and
qualitative aspects of the provisional value development and address whether uncertainties
associated with the assessment have been adequately characterized.
1 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
DISCLAIMERS
The PPRTV document provides toxicity values and information about the adverse effects
of the chemical and the evidence on which the value is based, including the strengths and
limitations of the data. All users are advised to review the information provided in this document
to ensure that the PPRTV used is appropriate for the types of exposures and circumstances at the
site in question and the risk management decision that would be supported by the risk
assessment.
Other U.S. EPA programs or external parties who may choose to use PPRTVs are
advised that Superfund resources will not generally be used to respond to challenges, if any, of
PPRTVs used in a context outside of the Superfund program.
This document has been reviewed in accordance with U.S. EPA policy and approved for
publication. Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
QUESTIONS REGARDING PPRTVS
Questions regarding the content of this PPRTV assessment should be directed to the
U.S. EPA ORD CPHEA website at https://ecomments.epa.gov/pprtv.
2 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
1. INTRODUCTION
In deriving provisional reference doses (p-RfDs) for sodium and potassium (Na/K) salts
of inorganic phosphates, two notable departures from the traditional approach to contaminants
found on contaminated sites exist: (1) the background concentrations found in humans is not
zero, and (2) the background concentrations found in contaminated sites is not zero. As discussed
below, phosphorus, an essential nutrient commonly found in various phosphates, exhibits a
"U-shaped" dose-response curve, wherein doses above deficiency and doses below toxicity may
overlap for some population subgroups. Thus, care must be taken in the interpretation and risk
communication of extra-dietary exposures above those occurring in a standard American diet
that may result from exposures at contaminated sites.
Phosphorus (P) is most commonly found in nature in its pentavalent form in combination
with oxygen, known as phosphate or orthophosphate anion (PO43 ). Phosphorus is an essential
constituent of all living organisms, and its content is quite uniform across most plant and animal
tissues. Orthophosphoric acid (H3PO4) is the basic unit for all phosphates. Condensed (oligo-,
pyro-, meta-, and other polyphosphates) are formed when two or more orthophosphoric acid
molecules condense into a single molecule. Pyrophosphates refer to compounds with two
condensed orthophosphates (P2O7)4 , and higher number polymers are termed polyphosphates,
sometimes followed by a suffix indicating the number (e.g., [HPCbjn; thus, tri- and
tetrapolyphosphates have three and four condensed phosphates, respectively). The term
"metaphosphates" is used when phosphoric acid moieties form a cyclic (ring) structure.
Inorganic phosphates (both ortho- and condensed polyphosphate anions) can be grouped into
classes on the basis of their cations: monovalent cations (sodium [Na+], potassium [K+],
ammonium [NH4+], and hydrogen [H+]); bivalent calcium (Ca2+) and magnesium (Mg2+); and
trivalent aluminum (Al3+). The phosphoric acids have been grouped with the other monovalent
cations on the basis of valence state of the cation.
Organic phosphates are commonly found in the form of esters as nucleotides or
deoxynucleotides (e.g., adenosine monophosphate [AMP], adenosine diphosphate [ADP],
adenosine triphosphate [ATP], or deoxyadenosine triphosphate [dATP]) and in deoxyribonucleic
acid (DNA) and ribonucleic acid (RNA). Free orthophosphate anions can be released by the
hydrolysis of the phosphoanhydride bonds in ATP or ADP. These phosphorylation and
dephosphorylation reactions are the immediate storage and source of energy for many metabolic
processes. ATP and ADP are often referred to as high-energy phosphates, as are the phosphagens
in muscle tissue. Similar reactions exist for the other nucleoside diphosphates and triphosphates.
Various other organic phosphates, such as phosphoproteins, are important in kinase cascades in
signal transduction pathways, structurally as phospholipids, or as sugar-phosphates in
intermediary metabolism. Plants also contain poorly absorbed phytates (inositol
hexaphosphates), which are storage forms of phosphorus principally found in seeds.
This document addresses the available data on the toxicity of phosphoric acids and
monovalent (sodium and potassium) salts of inorganic phosphates (see Table 1A). Aluminum,
ammonium, and the divalent salts (calcium [Ca] and magnesium) are not included in this
assessment because they are expected to have differing toxicity, chemistry, or toxicokinetics.
3 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Specifically, ammonium ions are expected to exert toxic effects that are independent of
the monohydrogen phosphate moiety, which would confound the hazard identification for other
inorganic phosphates. In addition, ammonium phosphates exhibit different chemistry and
different toxicity than Na/K phosphates: the ammonium salts are relatively unstable because
ammonium hydroxide is a weaker base than metal hydroxides and ammonia can escape as a gas
(Gard, 2005). Ca phosphates are much less soluble than sodium or potassium phosphates (Gard,
2005). In addition, interactions between phosphate and Ca occur both in the intestine and in the
kidney. In the intestines, the phosphate ion present will transition from the monovalent
dihydrogenphosphate in areas near the stomach to the divalent hydrogen phosphate ion as the pH
of the intestinal contents approaches pH 7 (where calcium phosphate formation reduces
absorption of both nutrients) and in the kidney (where phosphate decreases urinary Ca excretion)
[as reviewed by EFSA (2015)1. These interactions do not occur with monovalent (sodium and
potassium) phosphates. There is evidence that the ratio of calcium:phosphate is an important
determinant of phosphate toxicity (particularly regarding bone health); thus, administration of
calcium phosphate is expected to yield different toxicological effects than would administration
of sodium or potassium phosphate. Finally, although magnesium phosphates are somewhat more
soluble than calcium phosphates (Gard, 2005). both are most appropriately considered with other
divalent salts.
Na/K salts of inorganic phosphates are used in a wide variety of applications; a few
examples include in fertilizers, soaps, detergents, pH-regulating agents, antifreezing agents,
adsorbents in baking powders, food nutrient supplements, and acid cleansers; in electroplating,
dyeing and tanning, wastewater treatment, pottery glazing, and soldering and brazing; as a scale
inhibitor in boiling water treatment; and as a fireproofing agent (Nl.M. 2021; OHCD. 2009). In
general, Na/K salts of inorganic phosphates are soluble in water (see Table IB), and the
phosphate moiety may persist indefinitely in natural waters (Nl .M. 2021). Polyphosphates
decompose ultimately to orthophosphate in water, sodium tripolyphosphate (NasP^Oio)
hydrolyzes to phosphonic acid (H2PO3), and tetrasodium phosphate (Na4P20?) slowly hydrolyzes
to orthophosphate (Nl .M. 2021). Sodium polyphosphates ([NaPOj]«) and sodium
hexametaphosphate ([NaPCbje) depolymerize in aqueous solution to form sodium
trimetaphosphate ([NaPO.^) and sodium orthophosphates (NLM. 2021; CIR Expert Panel.
2016).
Human exposure to Na/K salts of inorganic phosphates is likewise ubiquitous,
particularly through their use as food additives and in water treatment. Although little
information was located regarding the proportion of total dietary P load from monovalent salts of
inorganic phosphate additives, one source estimated the contribution as -500 mg P/day (Calvo
and Uribarri. 2013 as cited in Trautvetter et al.. 2018). Phosphate blends, which may include any
of the compounds in Table 1 A, are also used in municipal water treatment for a variety of
purposes, including to control scale and corrosion; to reduce iron, lead, and copper
concentrations in drinking water; and to reduce staining, scale, and objectionable tastes and
odors (Willhite et al.. 2013); human exposure may occur through any of these uses. Application
of monovalent salts of inorganic phosphates in fertilizers and use in consumer products, such as
toothpaste, water softeners, and detergents, also provide opportunities for human exposure.
Finally, sodium phosphate (typically as a mixture of mono- and disodium phosphate) has been
used as a bowel cleansing agent for colonoscopy preparation and as a therapeutic treatment for
constipation. Its use in bowel cleansing was significantly curtailed by the U.S. Food and Drug
Administration (FDA) in 2008 in response to adverse event reports (FDA. 2008).
4 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
The empirical formulas for selected Na/K salts of inorganic phosphates are shown in
Table 1 A. Tables 1A and IB show the nomenclature and physicochemical properties
(respectively) of the selected salts.
5 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 1A. Nomenclature, Chemical Formula, and Molecular Weight of Selected Na/K Salts of Inorganic Phosphates
Compound
CASRN
Empirical Formula
MW
(g/mol)a
MW Ratio P:
Compoundb
Physical State3
Orthophosphoric acid
7664-38-2
H3PO4
97.995
0.31608
Colorless crystals or viscous liquid
Polyphosphoric acid
8017-16-1
(HPO3),,: molecular weights vary
Clear, viscous liquid
Sodium dihydrogen orthophosphate
(monosodium phosphate)
7558-80-7
NaH2P04
119.977
0.25817
White or colorless crystalline powder
Sodium monohydrogen orthophosphate
(disodium phosphate)
7558-79-4
Na2HP04
141.96
0.21819
Colorless crystals or white hygroscopic powder
Sodium orthophosphate
(trisodium phosphate)
7601-54-9
Na3PC>4
163.94
0.18893
Colorless crystals
Sodium dihydrogen pyrophosphate
(disodium diphosphate)
7758-16-9
Na2H2P2C>7
221.94
0.27912
White crystalline powder
Sodium pyrophosphate
(tetrasodium diphosphate)
7722-88-5
Na4P2C>7
265.94
0.23294
Colorless transparent crystals or white powder
Sodium tripolyphosphate
(pentasodium triphosphate)
7758-29-4
NasPaOi 0
367.86
0.25260
White powder
Sodium trimetaphosphate
7785-84-4
(NaP03)3
305.882d
0.30378
White crystals or white crystalline powder0
Sodium polyphosphate
68915-31-1
(NaPOs,),,. where n = 10-15; molecular weights vary
Clear hygroscopic glass
Sodium hexametaphosphate
10124-56-8
(NaP03)6
611.763
0.30378
Clear hygroscopic glass
Potassium dihydrogen phosphate
(monopotassium phosphate)
7778-77-0
KH2PO4
136.085
0.22761
Colorless crystals or white granular powder
Potassium monohydrogen phosphate
(dipotassium phosphate)
7758-11-4
K2HPO4
174.18
0.17783
White crystals or colorless granules or powder
Potassium phosphate
(tripotassium phosphate)
7778-53-2
K3PO4
212.265d
0.14389
Colorless or white hygroscopic crystals or granules'1
Potassium pyrophosphate
(tetrapotassium diphosphate)
7320-34-5
K4P2O7
330.33°
0.18753
Colorless crystals or white, very hygroscopic powderd
6
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 1A. Nomenclature, Chemical Formula, and Molecular Weight of Selected Na/K Salts of Inorganic Phosphates
Compound
CASRN
Empirical Formula
MW
(g/mol)a
MW Ratio P:
Compoundb
Physical State3
Potassium tripolyphosphate
(pentapotassium triphosphate)
13845-36-8
K5P3O10
448.403d
0.20723
White, very hygroscopic powder or granules'1
aNLM (2021) unless otherwise specified.
bMW of P — 30.974 g/mol (NLM. 20211.
°CIR Expert Panel (2016).
dPubChem (2018).
MW = molecular weight; Na/K = sodium and potassium; P = phosphorus.
Table IB. Physicochemical Properties of Selected Na/K Salts of Inorganic Phosphates3'b
Compound
Chemical
Formula
Melting Point
(°C)
Density
(g/cm3 at 25°C)
pH
(unitless)
pKa
(unitless)
Solubility in Water
(mg/L at 25°C)
Orthophosphoric acid
H3PO4
42.4
1.87
1.5
(0.1 N aqueous solution)
pKi =2.15,
pK2 = 7.09,
pK3 = 12.32
5.48 x 106 at 20°C
Polyphosphoric acid
(HPO,)„
Varies
Soluble
Sodium dihydrogen orthophosphate
(monosodium phosphate)
NaH2P04
200 (decomposes)
2.36
4.5
(1% aqueous solution)
6.8-7.2
9.49 x 105
Sodium monohydrogen orthophosphate
(disodium phosphate)
Na2HP04
35 (dodecahydrate)
1.7
(approximate)
9.1
(1% aqueous solution at 25°C)
pKi =2.15
1.18 x 105
Sodium orthophosphate
(trisodium phosphate)
Na3P04
1,583 (anhydrous)b
75 (dodecahydrate)0
2.54
11.5-11.9
(0.1-1% solution)
NV
1.45 x 105
Sodium dihydrogen pyrophosphate
Na;H;P.O-
220 (decomposes)
1.86 (hexahydrate)
NV
NV
3.5 x 105 at 40°C
Tetrasodium pyrophosphate
(tetrasodium diphosphate)
Na4P 2O7
988 (anhydrous)
79.5 (decahydrate)0
2.534
10.2
(1% solution)
NV
6.7 x 104
Sodium tripolyphosphate
(pentasodium triphosphate)
NasPaOiu
622
NV
9.7-9.8
(1% solution at 25°C)
NV
2.0 x 105
7
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table IB. Physicochemical Properties of Selected Na/K Salts of Inorganic Phosphates"'b
Compound
Chemical
Formula
Melting Point
(°C)
Density
(g/cm3 at 25°C)
pH
(unitless)
pKa
(unitless)
Solubility in Water
(mg/L at 25°C)
Sodium trimetaphosphate
(NaP03)3
53 (hexahydrate)
1.786 (hexahydrate)
NV
NV
2.2 x 105
Sodium polyphosphate
(NaP03)„
Varies
Soluble in water"1
Sodium hexametaphosphate
(NaP03)6
628
NV
NV
NV
Slowly soluble in water
Potassium dihydrogen orthophosphate
(monopotassium phosphate)
KH2PO4
253
2.34
4.4-4.7
NV
2.5 x 105
Potassium monohydrogen
orthophosphate
(dipotassium phosphate)
K2HPO4
NV/decomposes
0.6 to ~0.7e
8.8
(1% solution)
NV
1.68 x 106
Potassium phosphate
(tripotassium phosphate)
k3po4
NV
NV
11.5-12.3
(1% solution/
NV
Freely soluble in water'
Tetrapotassium pyrophosphate
(tetrapotassium diphosphate)
K4P2O7
1,090
NV
10.0-10.8
(1% solution/
NV
1.87 x 106f
Potassium tripolyphosphate
(pentapotassium triphosphate)
KsP3Oio
NV
NV
9.2-10.5
(1% solution/
NV
Very soluble in waterf
aOctanol-water partition coefficient, Henry's law constant, soil adsorption coefficient, atmospheric OH rate constant, and atmospheric half-life are not applicable to
inorganic phosphates.
bNLM (2021). unless otherwise specified.
"U.S. EPA (2012).
dCIR Expert Panel (2016).
eOECD (2006). potassium.
'PubChetti (2018).
Na/K = sodium and potassium; NV = not available; pKa = acid dissociation constant.
8
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
A summary of available subchronic and chronic toxicity values for phosphorus from
U.S. EPA and other agencies/organizations is provided in Table 2.
Table 2. Summary of Available Subchronic or Chronic Toxicity Values and
Adequate Intake Values for Na/K Salts of Inorganic Phosphates (Multiple
CASRNs)
Source
(parameter)3'b
Phosphorus
Form
Value (applicability)
Notes
Reference0
Noncancer
IRIS (RfC)
Phosphoric
acid (PA)
0.01 mg/m3
Basis: Bronchiolar fibrosis in a
13-wk rat study. This RfC is for
aerosols of PA and P oxidation
products and does not apply to
elemental P or other forms of P,
such as phosphorus salts.
U.S. EPA (1995)
HEAST
NA
NV
NA
U.S. EPA
(2011a)
DWSHA
NA
NV
NA
U.S. EPA (2018)
ATSDR
NA
NV
NA
ATSDR (2018)
EFSA (AI)
P
AI by age:
7-11 mo: 200 mg/d
I-3 yr: 300 mg/d
4-10 yr: 600 mg/d
II-17 yr: 800 mg/d
Adults and pregnant/lactating
women: 700 mg/d
Phosphorus from all sources,
expressed as P.
EFSA (2015)
IOM (UL)
P
Children: 3,000 mg/d
Adults <70 yr: 4,000 mg/d
Adults >70 yr: 3,000 mg/d
Pregnant women: 3,500 mg/d
Lactating women: 4,000 mg/d
The maximum level of daily
nutrient intake that is likely to
pose no risk of adverse effects.
The UL represents total intake
from food, water, and
supplements.
IOM (1997)
WHO (MTDI)
P
70 mg/kg
Maximum intake of P across all
sources; based on
nephrocalcinosis in rats.
WHO (1982)
C,a.1F,PA (REL,
chronic
inhalation)
PA
7 (ig/m3
Basis: Bronchiolar fibrosis in
rats.
CalEPA (2017.
2008)
OSHA (PEL)
PA
1 mg/m3
8-h TWA for general industry,
construction, and shipyard
employment.
OSHA (2020.
2017a. 2017b)
NIOSH (REL)
PA
1 mg/m3
TWA for up to a 10-h workday
during a 40-h workweek.
NIOSH (2016)
ACGIH
(TWA-TLV)
PA
1 mg/m3
Basis: Upper respiratory tract,
eye, and skin irritation.
ACGIH (2016)
USAPHC
(air-MEG)
PA
1-yr negligible: 0.068 mg/m3
Based on IRIS.
U.S. APHC
(2013)
9 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 2. Summary of Available Subchronic or Chronic Toxicity Values and
Adequate Intake Values for Na/K Salts of Inorganic Phosphates (Multiple
CASRNs)
Source
(parameter)3'b
Phosphorus
Form
Value (applicability)
Notes
Reference0
USAPHC
(soil-MEG)
PA
1-yr negligible:
1,000,000 mg/kg
Basis: Noncancer.
U.S. APHC
(2013)
Cancer
IRIS
NA
NV
NA
U.S. EPA (2020)
HEAST
NA
NV
NA
U.S. EPA
(2011a)
DWSHA
NA
NV
NA
U.S. EPA (2018)
NTP
NA
NV
NA
NTP (2016)
CalFPA
NA
NV
NA
CalEPA (2020.
2019)
ACGIH
NA
NV
NA
ACGIH (2020)
aSources: ACGIH = American Conference of Governmental Industrial Hygienists; ATSDR = Agency for Toxic
Substances and Disease Registry; CalEPA = California Environmental Protection Agency; DWSHA = Drinking
Water Standards and Health Advisories; EFSA = European Food Safety Authority; HEAST = Health Effects
Assessment Summary Tables; IOM = Institute of Medicine; IRIS = Integrated Risk Information System;
NIOSH = National Institute for Occupational Safety and Health; NTP = National Toxicology Program;
OSHA = Occupational Safety and Health Administration; USAPHC = U.S. Army Public Health Command;
WHO = World Health Organization.
Parameters: AI = adequate intake; MEG = military exposure guideline; MTDI = maximum tolerable daily intake;
PEL = permissible exposure limit; REL = recommended exposure limit; RfC = reference concentration;
TLV = threshold limit value; TWA = time-weighted average; UL = tolerable upper intake level.
°Reference date is the publication date for the database and not the date the online source was accessed.
NA = not applicable; Na/K = sodium and potassium; NV = not available; P = phosphorus; PA = phosphoric acid.
Non-date-limited literature searches were conducted in October 2017 and updated most
recently in February 2022 for studies relevant to the derivation of provisional toxicity values for
inorganic phosphates, including phosphoric acid (PA) and selected Na/K salts (names and
CASRNs shown in Table 1 A). The literature search updated conducted in February 2022, post
External Peer Review, did not identify any studies which would affect the quantitative
conclusions of the PPRTV assessment.
Searches were conducted using U.S. EPA's Health and Environmental Research Online
(HERO) database of scientific literature. HERO searches the following databases: PubMed,
TOXLINE (including TSCATS1), and Web of Science. Secondary key words were applied in
EndNote to help prioritize results for screening. EndNote was used to manually screen items
published since the previous version of this PPRTV in 2009. Studies were then imported into
SWIFT Review software to identify those references most likely to be applicable to a human
health risk assessment, and resulting studies were imported to EndNote. In brief, SWIFT Review
has preset literature search strategies ("filters") developed by information specialists that can be
applied to identify studies that are more likely to be useful for identifying human health content
10 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
from those that likely do not (e.g., environmental fate). Only studies meeting the populations,
exposures, comparators, and outcomes (PECO) criteria are cited. The following databases were
searched outside of HERO for health-related values: American Conference of Governmental
Industrial Hygienists (ACGIH), Agency for Toxic Substances and Disease Registry (ATSDR),
California Environmental Protection Agency (CalEPA), European Centre for Ecotoxicology and
Toxicology of Chemicals (ECETOC), European Chemicals Agency (ECHA), U.S. EPA
Chemical Data Access Tool (CDAT), U.S. EPA ChemView, U.S. EPA Health Effects
Assessment Summary Tables (HEAST), U.S. EPA Integrated Risk Information System (IRIS),
U.S. EPA Office of Water (OW), International Agency for Research on Cancer (IARC), Japan
Existing Chemical Data Base (JECDB), National Institute for Occupational Safety and Health
(NIOSH), National Pesticide Information Retrieval System (NPIRS), National Toxicology
Program (NTP), Organisation for Economic Co-operation and Development (OECD) Existing
Chemicals Database, OECD Screening Information Data Set (SIDS) High Production Volume
Chemicals via International Programme on Chemical Safety (IPCS) INCHEM, Occupational
Safety and Health Administration (OSHA), and World Health Organization (WHO).
11 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
2. REVIEW OF POTENTIALLY RELEVANT DATA
(NONCANCER AND CANCER)
Tables 3A and 3B provide overviews of the relevant noncancer and cancer evidence
bases, respectively, for Na/K salts of inorganic phosphates, and include selected potentially
relevant acute, short-term, subchronic, and chronic studies as well as reproductive and
developmental toxicity studies. These tables include studies for which
no-observed-adverse-effect levels (NOAELs)/lowest-observed-adverse-effect levels (LOAELs)
could be identified (principal study is identified in bold). All NOAELs/LOAELs were identified
by the U.S. EPA unless noted otherwise. Studies that could not be used quantitatively (for
various reasons such as study quality issues, form of P not specified, Ca intake not measured,
etc.) but may be useful for hazard identification, are presented in Tables 4-10. The phrase
"statistical significance" and term "significant," used throughout the document, indicate a
p-value of < 0.05 unless otherwise specified.
12
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 3A. Summary of Potentially Relevant Noncancer Data for Na/K Salts of Inorganic Phosphates
(Multiple CASRNs)
Category3
Number of Male/Female, Strain,
Species, Study Type, Reported
Doses, Study Duration
Dosimetryb
Critical Effects
NOAELb
LOAELb
Reference
(comments)
Notes0
Human
1. Oral (mg/kg-d)
Acute
Retrospective cohort study of
9,799 patients given OSP or PEG
purgative prior to endoscopy, who
had available serum CR levels
within 1 yr of the procedure.
0 (PEG), 164
Use of an OSP purgative was associated
with increased odds of acute kidney injury
(OR = 2.35, 95% CI = 1.51-3.66)
compared with use of PEG purgative.
NDr
164
Hurst et al.
(2007); same
LOAEL as
FDA (2008)
warning
PR
Acute
Male and female children (>5 yr
old) or adults exposed for 1-3 d to
1,220-7,340 mgP/d for
constipation relief.
40-105
Dose above which dehydration,
electrolyte imbalances, and renal and
cardiovascular effects that are fatal in
some cases have occurred.
NDr
40
FDA (2014)
PR
Acute
16 M/0 F, meal containing
supplement, three different meals
(1 wk between meals) containing
400, 800, or 1,200 mg P; breakfasts
and dinners on these days each
contained 400 mg P/meal.
Participants served as their own
controls. Total intake: 1,200, 1,600,
or 2,000 mg P/d on 1 d.
19.87, 26.49, or
33.11 (including
diet)
Significantly decreased percent FMD, a
measure of endothelial function) at 1, 2,
and 4 h after the 800 and 1,200 mg P test
meals; no difference in percent FMD 20 h
after the meal. Blood pressure and other
serum chemistry measures were not
affected by treatment.
19.87
26.49
Nishi et al.
(2015)
PR
Acute
11 healthy male volunteers, mean
age 24.6 yr; double-blinded
crossover design. Participants
consumed a single meal containing
either 400 or 1,200 mg P
(supplemental P provided as
neutralized phosphate supplement).
Intake at other meals was not
reported. Participants served as
their own controls.
6.62 or 19.87
Percent FMD was significantly decreased,
and serum P was significantly increased
2 h after the high-dose meal (compared
with premeal measures); FMD returned to
normal within 24 h. Blood pressure and
other serum chemistry measures were not
affected by treatment.
NDr
19.87
Shuto et al.
(2009)
PR
13
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 3A. Summary of Potentially Relevant Noncancer Data for Na/K Salts of Inorganic Phosphates
(Multiple CASRNs)
Category3
Number of Male/Female, Strain,
Species, Study Type, Reported
Doses, Study Duration
Dosimetryb
Critical Effects
NOAELb
LOAELb
Reference
(comments)
Notes0
Short-term
31 adults (21 M and 10 F) mean
age 66.0 yr with early CKD (eGFR
>45 mL/min/1.73 m2 and urine
albumin:creatinine ratio >17 mg/g
in men or 25 mg/g in women);
randomized double-blind crossover
trial. Supplemental P given as
commercial diet soda or breakfast
bars.
15.38 or 24.87
No effect on frequency of self-reported GI
symptoms. Intake of P additives did not
significantly alter urinary albumin or
serum P or FGF-23. Blood pressure was
not altered by exposure.
24.87
NDr
Chang et al.
(2017)
PR
Short-term
10 healthy women aged 23-29 yr,
mean body weight of 60 kg, with
typical P intakes between 1,060 and
1,810 mg P/d based on 14-d dietary
records; crossover design. After a
4-wk control period (with mean
intakes of 1,700 mg P/d and
1,500 mg Ca/day), participants
were given sodium phosphate
(NaH2P04) tablets (containing
620 mg P) and orange juice
containing 975 mg P (bringing their
total daily intake to 3,008 mg P) for
6 wk; this period was followed by a
4-wk washout period in which
intakes were similar to the control
period.
28.33 (control
period) or
50.13 (including
diet)
Participants reported diarrhea, soft stools,
and intestinal disturbances (not further
described) during supplementation.
NDr
50.13
Grimm et al.
(2001)
PR
14
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 3A. Summary of Potentially Relevant Noncancer Data for Na/K Salts of Inorganic Phosphates
(Multiple CASRNs)
Category3
Number of Male/Female, Strain,
Species, Study Type, Reported
Doses, Study Duration
Dosimetryb
Critical Effects
NOAELb
LOAELb
Reference
(comments)
Notes0
Short-term
43 (35F/8M) patients with irritable
bowel syndrome-constipation
predominant; recruited by
researchers who were also
gastroenterologists; open-label dose
ranging study. Patients were
divided in two groups. Group A
had a mean age of 46 yr, while
Group B had a mean age of 49 yr.
Group A was 15 F/3 M, while
group B was 20 F/5 M. Mean body
weights were 155 (70.4 kg) lb for
Group A and 162 lb (73.6 kg) for
Group B. Supplemental P provided
as sodium phosphate, 1.5 g tablets,
(sodium phosphate (NaH2P04) and
disodium phosphate (Na2HP04) in
a ratio of 2.67:1)
Approximate doses
were 855 mg P
(25.2 mg P/kg-d) or
1.770 gP (48.2 g
P/kg-d), but
patients were
allowed to increase
or decrease dose
based on
symptoms. High
dose group (eight
tablets) received
3.44 gP/d
Symptoms reported in the low-dose group
included nausea, diarrhea, and incomplete
evacuation (1/18 each). In the high-dose
group, nausea and diarrhea were reported
by 4/25 and 3/25, respectively, and
additional symptoms included bloating
(2/25) and cramping, headache, lower
back pain, migraine, and lower quadrant
pain (1/25 each). No effects on body
weight, heart rate, or blood pressure were
observed. The only serum chemistry
change that was considered noteworthy
and related to treatment was occasional
hypokalemia in five patients (in both dose
groups). The study authors concluded that
the low dose was well tolerated and
suggested that a starting dose of
2-4 tablets/d (equivalent to 885-1,770 mg
P/d) would be appropriate.
25.2
48.1
Medoff et al.
(2004)
PR
2. Inhalation (mg/m3)
ND
15
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 3A. Summary of Potentially Relevant Noncancer Data for Na/K Salts of Inorganic Phosphates
(Multiple CASRNs)
Category3
Number of Male/Female, Strain,
Species, Study Type, Reported
Doses, Study Duration
Dosimetryb
Critical Effects
NOAELb
LOAELb
Reference
(comments)
Notes0
Animal
1. Oral (mg/kg-d)
Subchronic
10 M/0 F per dose group, Wistar
rat, 0.6 (as CaHP04), 1.2, or 1.8%
P (as CaHP04 and KH2PO4) in diet
for 8 wk.
510 (referent), 980,
or 1,400
Statistically significantly decreased body
weight (11% less than control at the end
of exposure) and food intake; significantly
decreased bmc and bmd; alterations in
several measures of bone
histomorphometry; nonsignificant
decrease in bone strength.
NDr
980
Huttunen et al.
(2007)
PR
Subchronic
0 M/5 F per dose group, Wistar rat,
0.5, 1.5% P (as KH2PO4) in diet for
6 wk.
270 (referent) or
800
Significantly decreased bmd, significantly
increased serum osteocalcin, and urinary
deoxypy ridinoline.
NDr
800
Koshihara et al.
(2005)
PR
Subchronic
0 M/6-16 F per dose group, Wistar
RIV:TOX rat, 0.4 or 0.6% P (as
NaH2P04 dihydrate) in diet for
4 wk.
390 (referent) or
580 (Experiment 1);
410 (referent) or
580 (Experiment 2)
Renal effects (increased urinary albumin,
increased relative kidney weight, and
nephrocalcinosis).
NDr
580
Ritskes-
Hoitinga et al.
(1989)
PR
Subchronic
8 M/0 F per dose group, NZW
rabbit, 0.45 or 0.88% P (as
NaEhPCh dihydrate) in diet for
8 wk.
150 (referent) or
290
Nephrocalcinosis, based on significantly
increased kidney Ca (and P), and
significantly increased severity scores
for cortical calcifications.
NDr
290
Ritskes-
Hoitinsa et al.
(2004)
PR, PS
16
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 3A. Summary of Potentially Relevant Noncancer Data for Na/K Salts of Inorganic Phosphates
(Multiple CASRNs)
Category3
Number of Male/Female, Strain,
Species, Study Type, Reported
Doses, Study Duration
Dosimetryb
Critical Effects
NOAELb
LOAELb
Reference
(comments)
Notes0
Subchronic
6 M/0 F per dose group, Wistar rat,
0.3,0.6,0.9, 1.2, or 1.5% P (as
KH2PO4) in diet for 4 wk.
250 (referent), 450,
670, 920, or 1,000
Decreased body-weight gain and body-
weight gain normalized to intake
(decreased 38 and 26% relative to
comparison group, respectively). Renal
Npt Ha mRNA and brush border protein
decreased at high dose.
920
1,000
Tani et al.
(2007)
PR
Subchronic
6 M/0 F per dose group, Sprague
Dawley rat, 0.6 or 1.2% P (as
KH2PO4) in diet for 4 or 14 wk.
530 (referent) or
1,100
Decreased terminal body weight (-14%
less than control; shown graphically) after
14 wk; decreased visceral fat mass and
plasma leptin, and increased serum BUN
and absolute kidney weights (-12%
higher than comparison group; shown
graphically) after 4 wk.
NDr
1,100
Abuduli et al.
(2016)
PR
Subchronic
20 M/20 F, Birmingham-Wistar rat,
control and 3% (526 or 1,650 mg
P/100 g) diet (as Tetron K,
analyzed as 97.5% Na |P;0- and
2.5% Na3PC>4) in diet for 16 wk.
Additional dose groups were
reported, but without P content of
the diet, precluding dose
estimation.
280 (referent) or
860 (M);
350 (referent) or
1,100 (F)
Decreased body-weight gain (females);
increased relative heart, stomach and
kidney weights (both sexes); increased
relative testes weight (males) and
increased relative intestine weight
(females); impaired kidney function
(concentration test, both sexes) at 860 or
1,100 mg P/kg-d. Increased relative
kidney weight, renal histopathology, and
impaired kidney function (males) also
occurred in animals receiving intermediate
doses for which the analytical P
concentrations were not given. For this
reason, effect levels could not be
determined.
NDr
NDr
Datta et al.
(1962)
PR
17
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 3A. Summary of Potentially Relevant Noncancer Data for Na/K Salts of Inorganic Phosphates
(Multiple CASRNs)
Category3
Number of Male/Female, Strain,
Species, Study Type, Reported
Doses, Study Duration
Dosimetryb
Critical Effects
NOAELb
LOAELb
Reference
(comments)
Notes0
Subchronic
20 M/20 F, Birmingham-Wistar rat,
control (referent) and 5% (526 or
1,598 mg P/100 g diet (as
Na2HP04) in diet for 16 wk.
280 (referent) or
840 (M);
350 (referent) or
1,000 (F)
Increased relative kidney weight (17 and
39% in males and females, respectively);
impaired kidney function (concentration
test, both sexes); and increased incidence
of renal histopathology (medullary
calcification and necrosis; tubular casts;
hemorrhages/exudate; chronic
inflammatory changes).
NDr
840
Datta et al.
(1962)
PR
Subchronic to
chronic
12 M/0 F, Wistar rat, 0.43 or 1.3%
P in diet (as K2HPO4) or 0.46 or
1.2% P in diet (as |NaPCh|r,) for
60 or 150 d (8.5 or 21 wk).
380 (referent) or
1,100 (K2HPO4);
400 (referent) or
1,100 ([NaP03]e)
There was a 12% increase in body weight
gain in the high K2HPO4 group and in the
high (NaP03)6 group; relative testes
weight was increased by 11% compared
with the low (NaP03)6 group.
1,100
NDr
Dvmsza et al.
PR
(1959)
Chronic
No studies met inclusion criteriad.
Reproductive
No studies met inclusion criteriad.
Developmental
No studies met inclusion criteriad.
18
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 3A. Summary of Potentially Relevant Noncancer Data for Na/K Salts of Inorganic Phosphates
(Multiple CASRNs)
Number of Male/Female, Strain,
Species, Study Type, Reported
Reference
Category3
Doses, Study Duration
Dosimetryb
Critical Effects
NOAELb
LOAELb
(comments)
Notes0
2. Inhalation (mg/m3)
ND
aDuration categories are defined as follows: Acute = exposure for <24 hours; short-term = repeated exposure for 24 hours to <30 days; long term (subchronic) = repeated
exposure for >30 days <10% life span for humans (>30 days up to approximately 90 days in typically used laboratory animal species); and chronic = repeated exposure
for >10% life span for humans (>~90 days to 2 years in typically used laboratory animal species) (U.S. EPA. 2002).
bDosimetry: All doses are reported in mg P/kg-day. Doses are presented as ADDs in (mg P/kg-day) units for oral noncancer effects.
°Notes: PR = peer reviewed; PS = principal study.
inclusion criteria are detailed in the text. Studies that did not meet inclusion criteria are considered supporting data and discussed briefly in the text or presented in
tables.
ADD = adjusted daily dose; bmc = bone mineral content; bmd = bone mineral density; BUN = blood urea nitrogen; Ca = calcium; CI = confidence interval;
CKD = chronic kidney disease; CR = creatinine; eGFR = estimated glomerular filtration rate; F = female(s); FGF-23 = fibroblast growth factor-23; FMD = flow
mediated dilation; GI = gastrointestinal; KH2PO4 = monopotassium phosphate; K2HPO4 = dipotassium phosphate; LOAEL = lowest-observed-adverse-effect level;
M = male(s); mRNA = messenger ribonucleic acid; (NaP03)6 = sodium hexametaphosphate; Na/K = sodium and potassium; Na2HP04 = disodium phosphate;
Na3PC>4 = sodium orthophosphate; Na4P207 = sodium pyrophosphate; NaH2P04 = monosodium phosphate; ND = no data; NDr = not determined;
NOAEL = no-observed-adverse-effect level; NPT Iia = type II sodium-dependent phosphate transporter; NZW = New Zealand White; OR = odds ratio; OSP = oral
sodium phosphate; P = phosphorus; PEG = polyethylene glycol.
19
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 3B. Summary of Potentially Relevant Cancer Data for Na/K Salts of Inorganic Phosphates (Multiple CASRNs)
Category
Number of Male/Female,
Strain, Species, Study Type,
Reported Doses, Study
Duration
Dosimetry
Critical Effects
Reference
(comments)
Notes
Human
1. Oral (mg/kg-d)
ND
2. Inhalation (mg/m3)
ND
Animal
1. Oral (mg/kg-d)
ND
2. Inhalation (mg/m3)
ND
ND = no data.
20
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
2.1. HUMAN STUDIES
2.1.1. Oral Exposures
Human data pertinent to the hazard assessment of oral exposure to monovalent inorganic
phosphates can be grouped into three main categories: human epidemiological studies on
associations between dietary P intake and health outcomes, randomized controlled trials of
humans exposed to Na/K salts of inorganic phosphates for acute or short-term durations, and
human studies of renal toxicity or gastrointestinal (GI) symptoms after acute exposure to oral
sodium phosphate (OSP; a combination of monosodium phosphate and disodium phosphate) for
bowel cleansing or constipation treatment.
The database of human studies for oral exposure to P and monovalent inorganic
phosphates is extensive. To select the studies most relevant to the assessment of inorganic
phosphate toxicity, and particularly for dose-response assessment, several overall criteria were
established:
• Human studies published before the comprehensive IOM (1997) review were excluded,
as these studies would have been considered in establishing the recommended daily
intake (RDI) and tolerable upper intake levels (UL) derived at that time. Due to lack of
suitable dose-response data to derive tolerable upper intakes directly, the Institute of
Medicine (IOM) instead used the high end of the normal adult serum P range as the basis
for the dietary UL of 4 g P/day for adults and 3 g P/day for adults >70 years old (12.5 and
9.4 g/d respectively, as monohydrogen phosphate).
• Studies that did not explicitly report an administered dose or estimated intake of
phosphate were excluded. This group included studies of dietary intake that reported
exposure qualitatively (e.g., high, medium, low) and therefore could not be used for
dose-response assessment.
• Studies that used serum P, urinary P, or another potential marker (such as parathyroid
hormone [PTH]) to assess exposure were not included as they are not considered a
reliable measure of phosphate exposure or intake [as reviewed by EFSA (2015)1. as both
absorption and excretion of P are tightly regulated in the human body by homeostatic
mechanisms (see further discussion in Section 2.3.3).
• Abstracts of conference presentations or posters, correspondence, and foreign language
studies were excluded, unless they identified a health outcome not identified in the
remaining literature.
The remaining studies were sorted into dietary intake studies, controlled exposure
studies, and studies of acute exposure to OSP for colonoscopy preparation. These studies were
then screened to determine whether they met the criteria listed below. Studies meeting the
criteria were tabulated and evaluated for their potential to support a quantitative evaluation of
dose-response. The studies retained include those that met the following conditions:
• Controlled exposure to Na3P04:
1) The study must evaluate a noncancer toxicological endpoint associated with oral
intake. Studies evaluating whether phosphate supplementation improves athletic
performance, those assessing beneficial effects of P supplementation on patients with
hypophosphatemia, or those evaluating cotreatment with another agent were not
included.
21 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
• Monitored dietary phosphate intake:
1) The study must use cohort or case-control design. Cross-sectional studies of dietary
intake and health outcomes were not generally considered sufficiently robust to use in
dose-response assessment because intake and effect are assessed at approximately the
same time, making them susceptible to recall bias and unable to support a temporal
relationship between outcomes and exposure. However, these studies may provide
information on potential indicators of hazard. Thus, they are discussed broadly but
not included in the studies evaluated for dose-response.
2) The study must evaluate a noncancer or cancer endpoint associated with the oral
exposure route. Studies assessing only biomarkers of exposure, such as serum
electrolytes, PTH, fibroblast growth factor-23 (FGF-23), or Klotho protein were
excluded. The toxicological significance of changes in the levels of these biomarkers
is unclear.
• Acute exposure to a sodium phosphate bowel preparation:
1) The study must use cohort or randomized prospective design. Case reports and case
series were not tabulated. They are discussed qualitatively in the text.
2) The study must evaluate sensitive noncancer toxicological outcomes, such as serum
creatinine. Studies focused on bowel cleansing efficacy that included only
patient-reported symptoms (e.g., nausea, vomiting) were not included.
3) The study must include a comparison group not exposed to sodium phosphate
(e.g., studies comparing oral versus enema administration of sodium phosphate were
not included). The comparison group may be exposed to another bowel cleansing
agent (e.g., polyethylene glycol [PEG]).
4) Studies describing the occurrence of aphthous (stomatitis) lesions in the intestinal
mucosa after sodium phosphate administration are not tabulated, because they did not
have information on the condition of the mucosa prior to sodium phosphate exposure.
However, these lesions and their potential relationship to sodium phosphate exposure
are discussed in the text.
Appendix A provides a flow chart showing the disposition of human studies from the
literature searches.
In addition to the European Food Safety Authority (EFSA) literature search results
(HFSA. 2015. 2005), recent published reviews assessing the human health effects of elevated
dietary P intake (Chang and Anderson. 2017: Brown and Razzaque. 2015: Hong et at.. 2015:
Nadkarni and Uribarri. 2014: Anderson. 2013: Calvo and Tucker. 2013: Gutierrez. 2013: Jain
and El saved. 2013: Menon and Ix. 2013: Uribarri. 2013) were consulted to provide context for
the results from the qualified human data and relevant information on associations between
health endpoints and serum P or other biomarkers.
Most of the human studies selected for qualitative or quantitative consideration in the
assessment of phosphate noncancer effects, including all of the bowel preparation studies,
describe associations between exposure and renal effects; these studies are discussed in the
following subsection. Later subsections discuss studies of cardiovascular, skeletal/bone, and
other (miscellaneous) noncancer effects. Studies of cancer endpoints are discussed under
subsections for prostate cancer and other cancers. Each subsection is followed by one or more
tables summarizing key elements of the studies that met selection criteria. The tabulated
22 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
information includes study design features (population, exposure, and outcome assessment),
results (with covariates considered or included), and some strengths and limitations of each
study.
Renal Effects in Humans
Tables 4 (dietary intake) and 5 (acute exposure for bowel preparation) show studies of
renal effects that met criteria for dose-response consideration.
Controlled Exposure Studies
Limited information on renal effects was obtained from controlled exposure studies. In a
controlled exposure study described below in the "Cardiovascular Effects and All-Cause
Mortality in Humans" section, Chang et al. (2017) did not observe a statistically significant
relationship between P supplementation (2,206 mg P/day during supplementation versus
1,364 mg P/day during control period, including background) for 3 weeks and urinary albumin in
a randomized crossover trial of 31 patients with early stage chronic kidney disease (CKD). In
another cross-over design study in which urinary levels of microalbumin, al-microglobulin, and
P2-microglobulin were measured after 6 weeks of P supplementation (as sodium phosphate
tablets and orange juice fortified with P), the only observed change was a nonsignificant
decrease in microalbumin (Grimm et al.. 2001) (see "Skeletal/Bone Effects" section and
Table 7).
Dietary Intake Studies
Only one dietary intake study met selection criteria for dose-response consideration
(see Table 4). No cross-sectional studies examining the association between renal effects and
dietary P were found in the literature reviewed. Yoon et al. (2017) evaluated incident CKD
among participants aged 40-69 years in the Korean Genome and Epidemiology Study. The study
authors assessed glomerular filtration rate (GFR) and proteinuria every 2 years in 873 subjects
with diabetes mellitus and 5,846 subjects without diabetes; the mean follow-up time was 8 years.
Intake of P was assessed by semiquantitative 1-day recall food frequency questionnaire (FFQ)
administered by trained interviewers at baseline and 4 years later. Yoon et al. (2017) did not
report the number of items considered in the FFQ. They did indicate that it had been validated as
a general nutrient exposure assessment tool. The mean P intake in the participants was
959 mg P/day.
The study authors analyzed the association between incident CKD and P intake in
quartiles of P dietary density (g P/total daily calorie intake) in subjects with and without diabetes
mellitus. P intakes in the first through fourth quartiles of dietary density were 753, 903, 1,022,
and 1,185 mg P/day. No association between P intake and incident CKD was seen in subjects
without diabetes. In subjects with diabetes mellitus, the hazard ratio (HR) for CKD exhibited a
nonsignificant increase in Quartiles 2 and 3 of dietary P density, when compared with the first
quartile, whereas a significantly increased risk of CKD was observed in the highest quartile
(HR = 1.68, 95% confidence interval [CI] = 1.08-2.63. The study authors controlled for several
covariates in their analysis but did not control for concurrent Ca intake, or for a dietary source of
P (organic or inorganic; plant, animal, or additive).
Identification of effect levels from this study was complicated by the fact that the HR for
CKD was increased in Quartiles 2-4 for dietary P intakes when compared with Quartile 1. Thus,
the P intakes that represent the LOAEL and NOAEL are not immediately clear. Because the HR
23 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
was increased statistically significantly in the fourth quartile, an approximate LOAEL of
1,185 mg P/day is identified on the basis of P intake in the fourth quartile. This intake estimate
corresponds to an approximate dose of 16.93 mg P/kg-day based on a default human body
weight of 70 kg (mean body weight of subjects was not reported). A NOAEL is approximated on
the basis of the intake in the third quartile of 1,022 mg P/day, corresponding to 14.6 mg/kg-day.
Serum Phosphorus Studies
Several studies have reported an association between serum or urine P levels and
progression of CKD or development of end-stage renal disease [as reviewed by Chang and
Anderson (2017); Nadkarni and Uribarri (2014)1. These studies are limited by the use of serum
or urine biomarkers to assess exposure to P (due to tightly controlled homeostatic mechanisms),
as well as potential for reverse causation. Progression of renal disease impairs excretion of P,
leading itself to higher serum P. In addition, there is evidence that, as kidney function declines in
patients with CKD, they decrease their protein intake, with consequent decreases in phosphate
intake [as reviewed by Chang and Anderson (2017)1; thus, increases in serum P may not reflect
increased intake, but rather decreased excretion. There is some support in the literature indicating
that restriction of dietary P intake limits the progression of CKD [as reviewed by Nadkarni and
Uribarri (2014)1. This finding likely forms the basis for medical recommendations to restrict
phosphate intake and the use of phosphate binder therapeutic agents, in patients with CKD.
Studies of Oral Sodium Phosphate Use for Bowel Preparation or Constipation
Beginning in the 1990s, OSP preparations (e.g., Fleet Phospho-Soda) were used for
bowel cleansing in preparation for colonoscopy. OSP preparations were often better tolerated by
patients than were the other bowel cleansing options at the time (e.g., PEG solution), which
required consumption of large volumes (4 L) of fluid. The OSP dosing regimen used typically
consisted of two 45 mL solutions, each administered with 8 ounces of water, 10-12 hours apart
(the night before and the morning of scoping) (Marshall 2014; Markowitz and Perazeiia. 2009).
The total 90 mL dose included 43.2 g of NaH2P04 and 16.2 g of Na2HP04, containing a total of
1 1.6 g of P (Markowitz and Perazeiia. 2009). which is equivalent to a dose of 164 mg P/kg in
1 day for a 70-kg adult. Contraindications against use of OSP preparations included advanced
age, renal insufficiency, and other conditions that could influence fluid and electrolyte balance
(Marshall 2014).
Case reports and case series describing acute kidney injury (acute phosphate
nephropathy) after use of OSP preparations for bowel cleansing began to be published as their
use became more common (Connor et al. 2008; Ori et al. 2008; Slee et al. 2008; Bcvca et al.
2007; Ma et al. 2007; Markowitz et al. 2005; Markowitz et al. 2004; Orias et al. 1999). In
December 2008, after receiving several adverse event reports, the FDA required a "black box"
warning to be included on Fleet Phospho-Soda over-the-counter products and warned that it
should no longer be used for colonoscopy preparation (Marshall 2014; Markowitz and Perazetta.
2009). The bases for the FDA action included the case reports of acute kidney injury and
evidence from biopsies (calcium phosphate deposits) that supported a causal relationship.
In a meta-analysis that reported a nonsignificant pooled odds ratio (OR) for kidney
injury, Brunelli et al. (2009) suggested that instances of acute kidney injury resulted from OSP
use in persons with pre-existing kidney disease (a contraindicated use) or in individuals who did
not follow hydration recommendations during use. Although compliance with contraindications
24 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
and recommended hydration is important in assessing safety of OSP for medical purposes, these
concerns do not apply to assessment of hazard to humans exposed to sodium phosphates in the
environment.
A number of epidemiological studies published after the 2008 FDA action (Lee et al..
2017; Lav ton. et al.. 2014; Kan et al.. 2012; Scot et al.. 2010) (see Table 5) reported no increased
risk of acute kidney injury with OSP use. However, as Table 5 shows, the study populations
evaluated in these investigations excluded susceptible populations (those with inadequate renal
function or other contraindications to use), which may explain the lack of toxicological outcomes
in the studies published after the FDA restrictions were enacted.
Sodium phosphate preparations (both oral and enema versions) were also available over
the counter for use to relieve constipation. A typical oral product contained 21.6 g NaH2P04 and
8.1 g ofNa2HP04, P in one adult dose (3 tablespoons in 24 hours), containing a total of 7.34 g of
P (equivalent to a dose of 105 mg P/kg-day for an adult weighing 70 kg). For children aged
5-9 years, the recommended dose was 0.5 tablespoons in 24 hours, equivalent to 1.22 g P or
approximately 40 mg P/kg-day for a 30 kg child. In 2014, FDA published an additional warning
pertaining to the use of oral and rectal sodium phosphate preparations for constipation relief.
FDA (2014) warned that exceeding the recommended dose (40-105 mg P/kg-day in people
>5 years of age) or use of these products in children under 5 years of age (except under physician
direction) can lead to renal or cardiovascular injury or death via dehydration and electrolyte
imbalances. The warning noted the following particularly susceptible populations: young
children; individuals older than 55 years; patients who are dehydrated; patients with kidney
disease, bowel obstruction, or inflammation of the bowel; and patients who are using
medications that may affect kidney function (diuretics or water pills; blood pressure medications
including angiotensin converting enzyme inhibitors [ACEIs] and angiotensin receptor blockers
[ARBs]; and nonsteroidal anti-inflammatory drugs such as aspirin, ibuprofen, and naproxen).
25
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 4. Selected Cohort Studies Evaluating Associations between Phosphorus Intake and Renal Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations
Yoonetal. (2017) (Korea):
873 subjects with DM and
5,846 without (non-DM);
participants in Korean Genome
and Epidemiology Study;
recruited from June 2001 on;
aged 40-69 yr. There were
454 males (52%) and
419 females. Body weights
were not reported. The mean
age was 55.6 yr. Health status
reported 28.8% hypertension,
4.4% cardiovascular disease,
16.2% gastritis/ulcer, and 1.5%
malignancies. Other health
outcomes were not reported.
Intake assessed using
validated semiquantitative
1-d dietary recall FFQ
(number of items not
reported) administered by
trained interviewers, once at
baseline and again 4 yr later.
P content estimates were
obtained from 2011 nutrient
database of Korean
Nutrition Society.
Mean intake: 959 mg P/d.
Incident CKD was defined as
a composite of estimated
glomerular filtration rate
<60 mL/min/1.73 m2 or the
development of proteinuria,
evaluated biennially from
2001 to 2014.
P density in the highest quartile (in which the
mean intake was 1,185 mg P/d was
associated with an increased risk of CKD
(compared with lowest quartile) in DM
subjects; HR = 1.68 (1.08-2.63). No
association was seen in non-DM subjects.
Covariates in final model: Age, sex,
waist-to-hip ratio, average protein intake,
education, income, marital status, smoking
status, history of hypertension, fasting
glucose, serum albumin, and HDL
cholesterol.
Strengths: Relatively long
follow-up (mean 8 yr).
Limitations: Short-term
food frequency estimates;
highest quartile of intake
was less than typical U.S.
intake; does not distinguish
between organic and
inorganic P sources; did not
control for Ca intake.
Ca = calcium; CKD = chronic kidney disease; DM = diabetes mellitus; eGFR = estimated glomerular filtration rate; FFQ = food frequency questionnaire;
HDL = high-density lipoprotein; HR = hazard ratio; P = phosphorus.
26
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 5. Studies Evaluating Associations between Acute Oral Sodium Phosphate (OSP) Intake for Bowel Preparation
and Renal Endpoints
Citation (location); Size and Description of
Population
P Intake
Methods for
Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations
Abaskharotm et al. (2007) (Canada):
Retrospective cohort study of 767 patients
(324 females/443 males) given either an OSP
(/-/ = 618) or PEG (n = 149) purgative prior to
colonoscopy, who had available serum CR
levels. No eligibility restriction for baseline
kidney function. Body weight was not
reported. Mean age was 55.1 yr. 9.8% had
diabetes, 11.1% had cardiovascular disease,
3.4% had peripheral vascular disease, 1.8%
had kidney stones, 0.3% had CKD, and
26.2% had hypertension. Other health
outcomes were not reported.
Standard protocol
was two doses of
45 mL each,
yielding total dose
of 11.5 g P.
Intake = 11,500 mg
P in 1 d.
Renal function
assessed by serum CR
and estimated CR
clearance
immediately before
colonoscopy and
again 3 mo to 9 yr
later (mean 3.7 yr in
OSP group and 1.0 yr
in PEG group) at
repeat colonoscopy.
No significant difference was observed
between the two groups in the proportion
of patients developing renal insufficiency
(serum CR greater than the upper limit of
normal) or in the change in CR
clearance. Further, multivariate analysis
did not show type of purgative as an
independent predictor of renal
insufficiency.
Covariates in multivariate analysis: Age,
sex, comorbid conditions, and
medications.
Strengths: Moderately large
sample size.
Limitations: Study funded by OSP
manufacturer. Comparison group
exposed to PEG; more women in
OSP group, and more patients
with chronic disease and taking
medications in PEG group; time to
repeat CR measurement differed
significantly between groups, and
included some as long as 9 yr after
exposure; laboratory changed
upper limit of normal CR level
between 80 and 100 |imol/L.
Brunelli et al. (2007) (United States):
Nested case-control study of 116 cases of
kidney injury (93 females/39 males) and
349 control colonoscopy patients
(227 females/171 males) who did not meet
criteria as cases, with data on the purgative
used prior to the procedure. Mean ages were
65 yr (cases) and 63 yr (controls). Eligibility
was restricted to those with baseline serum
CR <1.5 mg/dL. Body weights were not
reported. Diabetes was noted in 38.4%,
congestive heart failure was noted in 14.6%.
Other health outcomes were not reported.
Standard protocol
was two doses of
45 mL each,
yielding total dose
of 11.5 g P.
Intake = 11,500 mg
P in 1 d.
Kidney injury
(defined as rise in
serum CR of
>0.5 mg/dL or 25%
between values
obtained up to 6 mo
before and up to 6 mo
after colonoscopy).
The odds of using OSP were not higher
among cases than by controls (adjusted
OR = 0.70, 95% CI = 0.44-1.11).
Covariates in multivariate analysis: Age,
race, sex, site, diabetes, congestive heart
failure, ACEI/ARB, diuretics, and
indication (screening or diagnostic
colonoscopy).
Strengths: Control for important
covariates.
Limitations: Small population
size. Some controls exposed to
other purgatives; more cases than
controls were female, had
congestive heart failure, and had
been exposed to diuretics.
27
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 5. Studies Evaluating Associations between Acute Oral Sodium Phosphate (OSP) Intake for Bowel Preparation
and Renal Endpoints
Citation (location); Size and Description of
Population
P Intake
Methods for
Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations
Hurst et al. (2007) (United States):
Retrospective cohort study of 9,799 patients
given OSP or PEG purgative prior to
endoscopy, who had available serum CR
levels within 1 yr of the procedure. Eligibility
restricted to subjects receiving screening
colonoscopy, age >50 yr, and without end
stage renal disease. No additional eligibility
restriction for baseline kidney function. Mean
age was 62.9 yr. 54.1% were male. Body
weight was not reported. Diabetes was noted
in 24.1%, hypertension in 64.0%,
cardiovascular disease in 14.0%, congestive
heart failure in 3.5%, and CKD in 9.6%.
Other health outcomes were not reported.
Standard protocol
was two doses of
45 mL each,
yielding total dose
of 11.5 g P.
Intake = 11,500 mg
P in 1 d.
Acute kidney injury
defined as increase of
>50% in baseline
serum CR (values
before and after
procedure and nearest
in date to the
procedure were used).
Use of an OSP purgative was associated
with increased odds of acute kidney
injury (OR = 2.35, 95% CI = 1.51-3.66)
compared with use of PEG purgative.
Covariates in final model: Age,
diabetes, hypertension, atherosclerotic
cardiovascular disease, ACEI or ARB
use, diuretic use, and factors suspected to
be associated with acute kidney injury
(e.g., NSAID use, congestive heart
failure, CKD, proteinuria, intravenous
contrast agent exposure).
Strengths: Large sample size;
includes sensitive subpopulations
(patients with chronic kidney
disease); reliable dose estimate;
control for important covariates;
studv performed prior to FDA
(2008) warnine on OSPs.
Limitations: Comparison group
exposed to PEG; differences in
comorbidities in the two groups.
Kan et al. (2012) (Taiwan):
Retrospective cohort study of 2,270 healthy
patients (1,107 females/1,663 males) given
either an OSP or PEG purgative prior to
colonoscopy, who had available electrolyte
levels. No patient with known conflicts and
with contraindications was included. Mean
age was not reported. Body weight was not
reported. Health conditions were not reported.
Standard protocol
was two doses of
45 mL each,
yielding total dose
of 11.5 g P.
Intake = 11,500 mg
P in 1 d.
Renal function
indicators (serum CR,
eGFR, serum uric
acid, urine specific
gravity, urine protein)
and serum electrolyte
levels (Ca, P, sodium,
potassium); timing of
blood and urine
collection was not
reported.
Use of an OSP purgative was associated
with significantly (p < 0.001) higher
prevalence of hyperuricemia (>9 mg/dL),
hypocalcemia (<8.8 mg%),
hypernatremia (145-152 mEq/L),
hyperphosphatemia (>4.7 mg%), and
high Ca x P product (>55 mg2/dL2).
In multivariate analysis, use of an OSP
purgative was associated with increased
odds of hyperphosphatemia (OR = 9.49,
95% CI = 7.12-12.72). Male sex reduced
the odds of hyperphosphatemia.
Covariates in final model: Laxative, sex,
age, and BMI.
Strengths: Large sample size;
reliable dose estimate.
Limitations: Comparison group
exposed to PEG; healthy
population; limited control for
covariates.
28
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 5. Studies Evaluating Associations between Acute Oral Sodium Phosphate (OSP) Intake for Bowel Preparation
and Renal Endpoints
Citation (location); Size and Description of
Population
P Intake
Methods for
Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations
Kkirana et al. (2008) (United States):
Retrospective cohort study of 286 patients
(103 males/183 females) with normal serum
CR (<1.5 mg/dL) given OSP purgative for
colonoscopy or flexible sigmoidoscopy,
compared with control group of 125 patients
with similar baseline comorbidities
(hypertension, diabetes, use of ACEI, ARB,
or diuretics). Mean ages were 68 and 69 yr in
OSP and control groups, respectively.
Eligibility restricted to those with baseline
serum CR <1.5 mg/dL. Body weights were
not reported. Diabetes was identified in 45%
of cases and 50% of controls. Hypertension
was identified in 95% of cases and 94% of
controls. Other health conditions were not
reported.
Standard protocol
was two doses of
45 mL each,
yielding total dose
of 11.52 g P.
Intake = 11,520 mg
P in 1 d.
Renal function (serum
CR and GFR) at 6 mo
and 1 yr after
enrollment.
Use of an OSP purgative was associated
with significantly higher serum CR and
lower GFR at 6 mo and 1 yr.
Multivariate linear regression analysis
showed significant (p < 0.008)
associations between OSP use and:
(1) increase in CR level and (2) decrease
in GFR (from baseline to 6 mo).
Covariates in final models: Age, sex,
race, ACEI or ARB use, diuretic use,
diabetes mellitus, and hypertension (for
analysis of CR); and ACEI or ARB use,
diuretic use, diabetes mellitus, and
hypertension (for analysis of GFR).
Strengths: Includes some sensitive
subpopulations; unexposed
comparison group; reliable dose
estimate; long-term follow-up.
Limitations: More women in
control than in OSP group.
29
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 5. Studies Evaluating Associations between Acute Oral Sodium Phosphate (OSP) Intake for Bowel Preparation
and Renal Endpoints
Citation (location); Size and Description of
Population
P Intake
Methods for
Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations
Lavton et al. (2014) (United States):
Retrospective cohort study of patients given
either OSP (n = 121,266) or PEG
(n = 429,430) within 1 mo of a screening
colonoscopy; age 50-75 yr. Population
sampled before FDA warning on OSPs.
Analysis also performed on
1:1 propensity-matched (matched on
predicted probability of initiating OSP or
PEG) subcohorts of 121,203 patients (each,
OSP and PEG). Subjects who had used either
medication in the prior year were excluded, as
were repeat colonoscopies on the same
patient. Eligibility was restricted to
individuals without acute kidney injury,
end-stage renal disease, unspecified renal
failure, rhabdomyolysis, dialysis, or renal
transplantation in the baseline year. Age of
patients was not reported, except a statement
that the PEG group was slightly older. Weight
of patients was not reported. Sex was not
reported, except a statement that there was a
slight preponderance of females in the PEG
group. Health status was not reported other
than a statement that the PEG group had
higher prevalence of diabetes, CKD,
hypertension, and cardiovascular disease.
Exposure estimated
from pharmacy
dispensing claims
during 30 d before
colonoscopy.
Standard protocol
was two doses of
45 mL each,
yielding total dose
of 11.52 g P.
Intake = 11,520 mg
P in 1 d.
Inpatient or outpatient
diagnosis of acute
renal failure in the
6 mo following
colonoscopy that
resulted in insurance
claim recorded in
TruvenMarketScan, a
U.S. administrative
claims database.
Mean follow-up time
was 170.7 d.
Use of OSP was not associated with
increased risk of acute renal failure
(compared with PEG) before or after
adjustment for potential confounders
(adjusted HR = 0.86, 95%
CI = 0.75-0.99). There was also no
association in subgroups with increased
risk of acute renal failure or in analyses
using the propensity-matched
subcohorts.
Covariates considered in analysis (final
covariates were not reported; assessed by
administrative claims in the baseline
year): Diagnoses of kidney stones,
hypercalciuria, diabetes mellitus,
hypertension, hyperlipidemia, ischemic
heart disease, heart failure, liver disease,
kidney disease, atrial fibrillation,
systemic lupus erythematosus, or
metabolic disorders.
Strengths: Large population;
samoline time was prior to FDA
(2008) warnine on OSPs.
Limitations: Comparison group
exposed to PEG; administrative
claims for baseline and impaired
renal function may be insensitive
measures; pharmacy dispensing
claims are indirect estimates of
exposure and do not account for
over-the-counter OSP use.
30
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 5. Studies Evaluating Associations between Acute Oral Sodium Phosphate (OSP) Intake for Bowel Preparation
and Renal Endpoints
Citation (location); Size and Description of
Population
P Intake
Methods for
Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations
Lee et al. (2017) (Korea):
Retrospective cohort study of patients given
OSP (n = 109, 57 males/52 females) PEG
with ascorbic acid (n = 163, 101 males/
62 females) or sodium picosulfate
(Ci8Hi3Nna208S2) magnesium citrate (n = 93,
56M/37F) purgative for colonoscopy.
Eligibility restricted to patients with eGFR
>60 mL/min/1.73m2. Body weight was not
reported. Mean ages were 47.6 yr in the OSP
group, 49.6 yr in the PEG ascorbic acid
group, and 50.7 yr in the sodium picosulfate
magnesium citrate group. Diabetes was
identified in 9.2% of the PEG ascorbate
group, 8.6% of the sodium picosulfate
magnesium citrate group, and 1.8% of the
OSP group. Hypertension was identified in
27, 25.8, and 16.5% in the same groups,
respectively.
32 OSP tablets
taken in two doses
10 h apart.
Renal function (serum
CR and eGFR) and
serum electrolytes
(sodium, chlorine, P,
and Ca) evaluated just
prior to colonoscopy
and compared with
baseline established
within 1 mo before.
Use of OSP was not associated with
significant difference in eGFR or serum
CR. The group given OSP had
significantly (p < 0,001) higher serum P
and significantly lower serum Ca than
both of the other treatment groups.
Covariates considered in analysis: Age,
sex, smoking, alcohol, diabetes mellitus,
hypertension, and BMI.
Strengths: Control for important
covariates.
Limitations: Small sample size;
comparison groups exposed to
PEG or sodium picosulfate
magnesium citrate; brief
follow-up; primary outcome was
bowel cleansing efficacy.
31
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 5. Studies Evaluating Associations between Acute Oral Sodium Phosphate (OSP) Intake for Bowel Preparation
and Renal Endpoints
Citation (location); Size and Description of
Population
P Intake
Methods for
Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations
Russmaniiet al. (2007) (United States):
Retrospective cohort study of patients of a
specific health maintenance organization,
given either OSP (n = 2,083,
1,158 males/925 females) or PEG (n = 269,
137 males/132 females) prior to colonoscopy,
who had serum CR measures within the
12 mo before and 6 mo after colonoscopy.
Patients with preexisting renal disease (during
prior 12 mo) were excluded (eligible patients
had eGFR >60 mL/min/1.73m2). Mean age
was not reported. Mean weight was not
reported. Hypertension was identified in
61.3% of the OSP group and 66.5% of the
PEG group. Diabetes was identified in 24.2%
of the OSP group and 25.3% of the PEG
group. Congestive heart failure was identified
in 3.6% of the OSP group and 14.9% of the
PEG group. Liver cirrhosis was identified in
1.6% of the OSP group and 3.0% of the PEG
group.
Standard protocol
was two doses of
45 mL each,
yielding total dose
of 11.52 g P.
Intake = 11,520 mg
P in 1 d.
Incident renal
impairment (GFR
<60 mL/min and
change in GFR of
>10 mL/min or
>twofold increase in
serum CR between
measurements before
and after
colonoscopy). Cases
reviewed by blinded
investigators for other
causes of renal
dysfunction and
excluded if another
cause was likely.
Use of OSP was not associated with
increased risk of incident renal
impairment in the 6 months following
colonoscopy (adjusted RR= 1.07, 95%
CI = 0.51-2.23), even when propensity
scoring methodology was used to control
for potential confounding.
Covariates in final model: Age, sex,
African American race, hospitalization
within 12 months prior to colonoscopy,
hypertension, baseline GFR >60 and
<90 mL/min, and current use of ACEI,
ARB, thiazide, or loop diuretics.
Strengths: Large population size.
Limitations: Study funded by OSP
manufacturer. Size of comparison
group was small. Comparison
group exposed to PEG. Serum CR
was not determined soon after
colonoscopy for all patients, so
transient changes in GFR were not
captured.
32
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 5. Studies Evaluating Associations between Acute Oral Sodium Phosphate (OSP) Intake for Bowel Preparation
and Renal Endpoints
Citation (location); Size and Description of
Population
P Intake
Methods for
Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations
Seol et al. (2010) (South Korea):
Retrospective cohort study of patients given
OSP (n = 224, (149 males/75 females) mean
age 47 yr) or PEG {n = 113,
(81 males/32 females) mean age 51 yr) for
screening colonoscopy, compared with
control group of 672 (475 males/197 females)
age-matched patients (mean age 47 yr) who
did not undergo colonoscopy. Eligibility
restricted to those with baseline serum CR
<1.5 mg/dL. Body weights were not reported.
Hypertension was identified in 22% of the
OSP group, 30% of the PEG group and 28%
of controls. Diabetes was identified in 13%
(OSP), 20% (PEG), and 20% in controls.
Standard protocol
was two doses of
45 mL each,
yielding total dose
of 11.5 g P.
Intake = 11,500 mg
P in 1 d.
Renal function (serum
CR and GFR at
baseline health
evaluation without
colonoscopy and at
follow-up evaluation
without colonoscopy
12-24 mo later).
Mean follow-up time
was not reported.
Use of OSP was not associated with
significant difference in baseline or
follow-up serum CR or GFR, or in
change in CR or GFR from baseline to
follow-up.
Covariates in final model: Baseline CR,
group, age, sex, medication for
hypertension, medication for diabetes
mellitus, BMI, and baseline phosphate
level.
Strengths: Long follow-up.
Limitations: Small exposed
population size.
33
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 5. Studies Evaluating Associations between Acute Oral Sodium Phosphate (OSP) Intake for Bowel Preparation
and Renal Endpoints
Citation (location); Size and Description of
Population
P Intake
Methods for
Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations
Sineal et al. (2008) (United States):
Exposure was
determined from
chart review.
Standard protocol
was two doses of
45 mL each,
yielding total dose
of 11.5 g P.
Intake = 11,500 mg
P in 1 d.
Renal function (serum
CR within 12 mo
after colonoscopy,
obtained by chart
review).
OSP use was associated with slight
increase in serum CR that was
significantly different from the change
induced by PEG (0.04 mg/dL vs.
-0.05 mg/dL for PEG; p = 0.005).
Likewise, OSP use resulted in an
increase in percentage change in serum
CR that differed from that induced by
PEG (5 vs. -3% for PEG; p = 0.01). In
forward logistic regression analysis, OSP
use was significantly associated with a
>25% increase in serum CR (p = 0.003).
Covariates in final model: NSAID use
and baseline CR.
Limitations: Small population
size, very few women (more
sensitive sex); few covariates
considered.
Retrospective cohort study of 311 patients
(>96% men) given either OSP (n = 157,
(151 males/6 females) mean age 66 yr) or
PEG (n = 154, (153 males/1 females) mean
age 69 yr) for colonoscopy. Patients without
pre- and postprocedure serum CR levels, or
with levels >1.5 mg/dL before procedure
were excluded. Body weights were not
reported. Hypertension was identified in
68.2% of the PEG group and 55.4% of the
OSP group. Diabetes was identified in 27.9%
in the PEG group and 28% in the OSP group.
Coronary artery disease was identified in
29.9 and 22.3%, respectively.
ACEI = angiotensin converting enzyme inhibitor; ARB = angiotensin receptor blocker; BMI = body mass index; Ca = calcium; CI = confidence interval; CKD = chronic
kidney disease; CR = creatinine; eGFR = estimated glomerular filtration rate; FDA = U.S. Food and Drug Administration; GFR = glomerular filtration rate; HR = hazard
ratio; NSAID = nonsteroidal anti-inflammatory drug; OR = odds ratio; OSP = oral sodium phosphate; P = phosphorus; PEG = polyethylene glycol; RR = relative risk.
34
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Cardiovascular Effects and All-Cause Mortality in Humans
Table 6 shows three dietary cohort studies and three controlled exposure studies that
evaluated cardiovascular effects and all-cause mortality (dietary cohorts only) and met criteria
for dose-response consideration. The controlled exposure studies examined effects of phosphate
additives or supplements on blood pressure, heart rate, and endothelial function.
Controlled Exposure Studies
Two controlled exposure studies provided direct evidence for an effect of acute increased
phosphate intake on endothelial function. Shuto et al. (2009) and Nishi et al. (2015) measured
flow-mediated dilation (FMD), a measure of endothelial function, and blood pressure in healthy
male volunteers, before and after meals containing additional P in the form of neutralized P
supplement. Both studies used a double-blinded crossover design. In both experiments,
significant decreases in percent FMD were observed 1-4 hours after the meals; FMD had
returned to normal 20-24 hours after the meals. Blood pressure was not affected in either
experiment. Nishi et al. (2015) observed that serum P levels were significantly increased at the
same time points when percent FMD was significantly decreased. While these studies used small
numbers of subjects, the endpoint (FMD) is sensitive, and is predictive of chronic outcomes, as
FMD has been shown to correlate with the severity and extent of coronary atherosclerosis
(Raitakari and Cclermaier. 2000). Acute LOAELs from Shuto et al. (2009) and Nishi et al.
(2015) were 19.87 and 26.49 mg P/kg-day, respectively, including dietary intake.
Chang et al. (2017) observed no change in blood pressure in 3 1 patients with early CKD
who were exposed for 3 weeks to food or beverages with or without phosphate additives (998 mg
P/day) in a randomized, double-blind crossover trial. In a study evaluating the use of sodium
phosphate tablets for chronic constipation (see "Other Noncancer Effects in Humans" section
below), 43 patients with irritable bowel syndrome received 2-6 or 4-12 tablets (total doses of
1,770 or 3,541 mg P/day, respectively, excluding dietary contributions) for 28 consecutive days
(Medoff et al.. 2004). In addition to target endpoints, the investigators measured heart rate and
blood pressure for comparison with pretreatment levels. No significant difference in either
parameter was observed.
Dietary Intake Cohort Studies
Dietary intake of P was not associated with hypertension in a prospective cohort study of
13,444 subjects from two U.S. cohorts (Alonso et al.. 2010). The investigators evaluated the risk
of incident hypertension with higher dietary P intake during 11 years of follow-up. Participants
were members of the Atherosclerosis Risk in Communities (ARIC) and Multi-Ethnic Study of
Atherosclerosis (MESA) cohorts and were 45-64 and 45-84 years of age, respectively. Dietary
intake of P was assessed at baseline using validated FFQs. Systolic and diastolic blood pressure
were measured at baseline, and a cross-sectional analysis of the relationship of baseline blood
pressure with dietary intake was performed. The subjects were then followed for 11 years, and
cases of incident hypertension were identified by measured blood pressure or current use of
hypertensive medication. At baseline, higher dietary P was associated with lower blood pressure
in cross-sectional analyses. In the prospective study, the risk of incident hypertension decreased
with increasing intake of P from dairy sources but not from other sources.
35 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
A large (n = 9,686) prospective cohort study of healthy adults recruited from the third
National Health and Nutrition Examination Survey (NHANES 111) (Chang et al.. 2014) did not
observe a statistically significant increase in cardiovascular mortality associated with dietary
intake of P; however, increased density of P in the diet was significantly associated with
increased cardiovascular mortality at densities >0.35 mg P/kcal diet (adjusted HR = 3.39; 95%
CI = 1.43-8.02 per 0.1-unit increase in P density in units of mg P/kcal diet [data not shown]). In
addition, dietary P intake >1,400 mg P/day was associated with a significant increase in all-cause
mortality after adjustment for a range of covariates (adjusted HR = 2.23; 95% CI = 1.09-4.55 per
1-unit increase in natural-log transformed P intake in mg P/day). Consideration of serum P level
as a covariate had little to no effect on the magnitude of the association or its confidence limits.
Noori et al. (2010a) also reported an increased risk of all-cause mortality (adjusted HR = 2.37;
95% CI = 1.01-6.32) associated with higher dietary P intake in 224 hemodialysis patients;
however, interpretation of this finding is complicated by the fact that P homeostasis is impaired
in hemodialysis patients, and because 69% of the patients in the study were taking phosphate
binders to inhibit absorption of dietary P.
Dietary Intake Cross-Sectional Studies
Cross-sectional studies in which dietary intake of P was measured evaluated associations
with blood pressure, coronary artery calcification, coronary artery intima media thickness (IMT),
a risk factor for cardiovascular disease, left ventricular hypertrophy, and all-cause mortality. No
significant associations between dietary P and blood pressure were observed (after adjustment
for covariates) in a study of 4,680 men and women in four different countries (data not shown)
(Elliott et al.. 2008).
Coronary artery calcification scores were not associated with dietary P intake in a
cross-sectional study of 25,652 Korean subjects (data not shown) (Kwak et al.. 2014). In a study
of 546 healthy male and female subjects in Finland, Itkonen et al. (2013) reported no significant
association between total dietary intake of P or P intake from food additives and carotid artery
IMT when assessed by quintiles of intake. However, there were significant (p < 0.05) trends
between energy-adjusted total phosphate intake and increased IMT (in all subjects), and between
food additive-derived P and increased IMT (Itkonen et al.. 2013).
In an analysis of 4,494 participants in the MESA cohort (data not shown), a significant
association was observed between dietary P intake and left ventricular mass measured by
magnetic resonance imaging (MRI) (Yamamoto et al.. 2013). The study authors estimated that,
after adjustment for covariates, a 20% increase in dietary phosphate intake was associated with
an increase in left ventricular mass (LVM) of 1.06 g (95% CI = 0.50-1.62 g,p< 0.001).
Yamamoto et al. (2013) also evaluated risk of left ventricular hypertrophy (LVH) with dietary
phosphate intake. LVH, a risk factor for heart attack and stroke, was defined in the study as
LVM indexed to body surface area (>85.3 g/m2 for women or 107.8 g/m2 for men). The study
authors observed higher adjusted odds of LVH among female participants (adjusted
Ors -1.5-3 across quintiles of intake, statistically significant in Quintiles 4 and 5, on the basis of
data shown graphically), but not male participants (data not shown) (Yamamoto et al.. 2013).
Murtaugh et al. (2012) did not observe a statistically significant association between dietary P
intake and all-cause mortality in 1,105 subjects with moderate (nondialysis) CKD in a
cross-sectional study of NHANES participants (data not shown).
36 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Serum Phosphorus Studies
Much of the epidemiological literature on cardiovascular effects of P consists of studies
measuring associations with serum P. Recent reviews of these relationships and their potential
mechanisms (Brown and Razzaque. 2015; Gross et al.. 2014; Anderson. 2013; Gutierrez. 2013;
Jain and El saved. 2013) point to increased risks of vascular calcification and atherosclerosis,
LVH, general cardiovascular disease, and mortality from cardiovascular disease with increases in
serum P. Vascular calcification reflects the deposition of calcium phosphate (usually as apatite)
in the cardiovascular system (blood vessels, valves, and myocardium) [as reviewed by Gross et
al. (2014)1. Several studies, including large prospective cohort studies, have demonstrated
associations between higher serum P and increased Ca in the coronary arteries in people with
moderate and advanced CKD, but also in healthy adults [as reviewed by Gross et al. (2014)1.
Although dietary P intake was not a predictor of coronary artery calcification scores in the
cross-sectional study reported by Kwak et al. (2014) (discussed under dietary intake), serum Ca,
serum P, and Ca x P product were significantly associated with increased calcification scores in
the three highest quartiles (compared with the lowest quartile). Other studies have also shown
that elevated serum P is associated with increased arterial stiffness and carotid vessel disease [as
reviewed by Gutierrez (2013)1 and an increased risk of LVH or increased LVM [as reviewed by
Brown and Razzaque (2015); Gutierrez (2013)1. Additional support for the relationship between
serum P and cardiovascular disease comes from studies showing higher frequencies of
cardiovascular events or higher mortality from cardiovascular disease with higher levels of
serum P (Brown and Razzaque. 2015; Anderson. 2013; Jain and HI saved. 2013). In contrast,
when urinary P excretion or the ratio of P to creatinine in urine was used as a biomarker for
phosphate homeostasis, no association with cardiovascular disease mortality was observed
(Dominguez et al.. 2013 as cited in Gutierrez. 2013).
The serum studies, supported by mechanistic evidence for effects of circulating P on
cardiovascular health, provide a relatively robust link between cardiovascular effects of
increased serum P. However, as there is a tenuous link between serum P and intake of P, studies
using serum P as a biomarker of exposure are less robust. Thus, these data are of limited utility
for assessment of inorganic phosphate toxicity, and specifically for Na/K salts.
37
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 6. Studies Evaluating Associations between Phosphorus Intake and Cardiovascular Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and
Limitations3
Dietary cohort studies
Alonso et al. (2010)
(United States):
13,444 participants in two
population-based cohorts: ARIC
(,n = 9,785 for cross-sectional
analysis and n = 8,208 for
longitudinal analysis) and
MESA (n = 3,659 for
cross-sectional analysis and
n = 2,901 for longitudinal
analysis); age at baseline
45-64 yr (ARIC) and 45-84 yr
(MESA). Sex distribution was
not reported. Age and weight
reported by phosphorus intake
quintile. Age averages were
52.8, 53.4, 53.5, 53.9, and
53.7 yr. Average body weights
were 75.2, 76.0, 75.7, 76.5, and
76.8 kg.
Intake was assessed by
validated FFQ (66 items
for ARIC and 120 items
for MESA); nutrient
content source was not
specified for ARIC; source
for MESA was the
Nutrition Data Systems for
Research Database.
Systolic and diastolic
blood pressure at baseline;
incident hypertension,
based on blood pressure
measurement (at follow up
visits every 2-3 yr) or
current use of
antihypertensive
medication. Follow-up
was 11 yr (ARIC) or 7 yr
(MESA).
In cross-sectional analyses, higher dietary P
intake was associated with decreased baseline
systolic and diastolic blood pressures in both
cohorts and in pooled analysis. In longitudinal
analysis, the HR for incident hypertension
decreased across quintiles of P intake from
dairy products, but the relationship did not
hold for P intake from nondairy sources.
Covariates in final model: Age, race, sex,
BMI, waist circumference, eGFR, education,
income, physical activity, cigarette smoking,
study site, alcohol intake, and energy intake.
Strengths: Longitudinal
design; large and
diverse populations;
validated exposure
assessment method;
clearly defined
outcome assessment;
consideration of known
confounders;
consideration of P
source.
Limitations: Diet
assessed only at
baseline.
38
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 6. Studies Evaluating Associations between Phosphorus Intake and Cardiovascular Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and
Limitations3
Chang et al. (2014)
Intake was assessed by
FFQ (24-h dietary recall,
recorded by trained
interviewers); content
estimates were obtained
from USD A Survey
Nutrient Database.
Exposure was assessed
both as absolute P intake
and P density.
Median intake = 1,166 mg
P/d.
All-cause and
cardiovascular mortality
abstracted from
NHANES III mortality
file through December 31,
2006; cardiovascular
mortality definition from
ICD, 10th edition.
P intake >1,400 mg P/day was associated with
increased all-cause mortality (adjusted
HR = 2.23; 95% CI = 1.09-4.55 per 1-unit
increase in natural-log-transformed P intake in
mg P/d). The association remained when
serum P was added as a covariate, and a
similar association was seen for P density
>0.35 mg P/kcal. P intake (mg P/kg-d) was not
associated with increased cardiovascular
mortality, while P density (mg P/kg food) was.
Covariates in final model: Age, sex, race,
ethnicity, poverty income ratio, total energy
intake, BMI, systolic blood pressure, current
and former smoking, physical activity,
non-HDL cholesterol, log albumin: creatinine
ratio, eGFR, and low vitamin D concentration.
Strengths: Longitudinal
design; large
population size; clearly
defined outcome
assessment;
consideration of known
confounders.
Limitations: Diet
assessed by 24-h recall;
no consideration of P
source (e.g., organic or
inorganic).
(United States):
9,686 nonpregnant healthy
adults without diabetes, cancer,
or kidney or cardiovascular
disease, recruited from
NHANES III, 1988-1994; aged
20-80 yr at baseline. Sex
distribution was not reported.
Age and health status
(hypertension only) were
reported by phosphorus intake
quartiles. Average body weights
were not reported. Age averages
were 42.5, 42.6, 41.0, and
38.5 yr. Hypertension incidence
was 19.6, 17.6, 16.0, and
15.8%.
39
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 6. Studies Evaluating Associations between Phosphorus Intake and Cardiovascular Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and
Limitations3
Controlled exposure studies
C ha tie etal. (2017)
(United States):
31 adults (21 males and
10 females) with early CKD
(eGFR>45 mL/min/1.73 m2
and urine albumin: creatinine
ratio >17 mg/g in men or
25 mg/g in women);
randomized double-blind
crossover trial. Body weights
were not reported. In the two
groups (low to high and high to
low), the mean ages were
64.2 and 68.0 yr. Females were
25% in the first group and 40%
of the second. Hypertension was
identified in 81 and 93%
respectively. Diabetes was
identified in 50 and 47%.
Dyslipidemia was identified in
50 and 67%. Coronary artery
disease was identified in 13 and
7%.
After a 2-wk run-in
period, participants were
given a breakfast bar and
65 ounces of diet
beverages with or without
phosphate additives
(additives provided
998 mg P/d) for 3 wk.
After the first period and a
2-wk washout period,
subjects crossed over to
the other treatment for an
additional 3 wk. Ca
content of the P foods was
higher than that of the
foods without P
(1,050 vs. 800 mg Ca/d).
Total intake: 1,364 mg P/d
in additive-free period or
2,206.1 mg P/d in additive
period.
Blood pressure was
measured, and blood and
24-h urine samples were
collected at the end of
each period; blood was
analyzed for P, intact
PTH, and FGF-23; urine
was analyzed for albumin,
P, and Ca.
Participants reported GI symptoms at
approximately the same frequency in both
treatment periods. Intake of P additives did not
significantly alter urinary albumin or serum P
or FGF-23. Blood pressure was not altered by
exposure. Intact PTH was significantly
increased after the additive exposure period
(compared with additive-free period). Urinary
excretion of P and Ca was significantly higher
during the P additive period compared with the
additive-free period.
NOAEL = 24.87 mg P/kg-d (including diet)
based on reported mean body weight of
88.7 kg during additive period. Alterations in
serum intact PTH and urinary P and Ca were
not considered to be toxicological.
Strengths: Robust
study design; duration
of 3 wk.
Limitations: Small
sample size;
background diet
changed during
treatment; limited
endpoints evaluated.
40
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 6. Studies Evaluating Associations between Phosphorus Intake and Cardiovascular Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and
Limitations3
Nishi et al. (2015) (Japan):
16 healthy male volunteers with
a mean age of 23.4 yr; double
blinded crossover design. Body
weight was not reported.
Participants consumed
meals at three different
times (1 wk between
meals) containing 400,
800, or 1,200 mg P
(supplemental P provided
as neutralized phosphate
supplement); breakfasts
and dinners on these days
contained 400 mg P/meal.
Participants served as their
own controls.
Total intake:
1,200, 1,600, or 2,000 mg
P/d on 1 d.
Blood samples were
obtained, and blood
pressure and FMD (a
measure of endothelial
function) were measured
immediately before the
test meal and 1, 2, 4, and
20 h after the test meal.
Serum levels of minerals,
intact PTH, glucose,
insulin, high, sensitive
C-reactive protein,
monocyte/macrophage
chemoattractant protein-1,
and FGF-23 were
measured.
Percent FMD was significantly decreased
(compared with 400 mg P/d meal) after the
800 and 1,200 mg P/d meals at 1, 2, and 4 h
after the test meal, but not 20 h after the meal.
The degree of change in percent FMD did not
differ between the two dose groups. Serum P
levels were significantly increased at the same
time points. Systolic and diastolic blood
pressure and other serum chemistry endpoints
were not affected by the test meal.
LOAEL = 26.49 mg P/kg-d (including diet) for
transient decrease in percent FMD, based on
reported mean body weight of 60.4 kg.
Strengths: Robust
study design; Ca intake
held constant across
exposure groups.
Limitations: Small
number of subjects;
single exposure.
41
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 6. Studies Evaluating Associations between Phosphorus Intake and Cardiovascular Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and
Limitations3
Noori et al. (2010a)
(United States):
224 patients on maintenance
hemodialysis, participants in the
NIED cohort; mean age 55 yr;
69% of patients were taking P
binders. Average body weight
was not reported. Age, sex, and
health status (diabetes only)
were reported by phosphorus
intake tertile. The mean ages
were 54.0, 57.7, and 54.4 yr.
Female sex was 57, 46, and
41%. Diabetes was identified in
57, 64, and 60%, respectively,
in the three tertiles.
Intake for prior 6-12 mo
was assessed by validated
questionnaire (Block Food
Frequency Questionnaire;
107 items) administered
by trained interviewers;
P content estimates
obtained from USD A food
ingredient data.
Intake ranged between
-250 and 2,000 mg P/d.
All-cause mortality;
methods not reported.
Highest tertile of P intake was associated with
increased risk of death compared with lowest
tertile; adjusted HR = 2.37 (95%
CI = 1.01-6.32) after adjustment for
covariates. The association remained after
adjustment for additional confounders
including serum cholesterol (data not shown).
Covariates in final model: Age, sex,
race/ethnicity, diabetes mellitus, dialysis
vintage, insurance, marital status, modified
Charlson comorbidity score, dialysis dose
(Kt/V), intake of P binders, residual urine,
energy, protein, and potassium intake, serum
concentrations of albumin, creatinine,
bicarbonate, ferritin, Ca, and P, blood levels of
Hb, WBC, and lymphocyte percent; and
normalized protein catabolic rate, nPCR, BMI,
and averaged doses of erythropoietin and
injected active vitamin D; serum
concentrations of c-reactive protein,
interleukin-6, and tumor necrosis factor-a.
Strengths: Control for
important covariates.
Limitations: Small
sample size; intakes
reported only as tertiles
without absolute
values; brief follow-up
(5 yr); methods for
mortality assessment
not reported; most
participants were
taking P binders.
42
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 6. Studies Evaluating Associations between Phosphorus Intake and Cardiovascular Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and
Limitations3
Shuto et al. (2009) (Japan):
11 healthy male volunteers,
mean age 24.6 yr; double
blinded crossover design. Body
weight was not reported.
Participants consumed a
single meal containing
either 400 or 1,200 mg P
(supplemental P provided
as neutralized phosphate
supplement). Intake at
other meals was not
reported. Participants
served as their own
controls.
Intake: 400 or 1,200 mg
P/d on 1 d (not including
diet).
Blood samples were
obtained, and blood
pressure and FMD were
measured immediately
before the test meal and
2 h after the test meal.
Serum minerals, intact
PTH, glucose, cholesterol,
and triglycerides were
measured.
Percent FMD was significantly decreased, and
serum P was significantly increased 2 h after
the meal (compared with premeal measures);
FMD returned to normal within 24 h. Blood
pressure and other serum chemistry measures
were not affected by treatment.
LOAEL = 19.87 mg P/kg-d (not including
diet) for transient decrease in percent FMD,
based on reported mean body weight of
60.4 kg.
Strengths: Robust
study design.
Limitations: Small
number of subjects;
single exposure; P
intake at other meals
not reported.
"Nutrient databases in general are believed to underestimate P intake [as reviewed by McClure et al. (2017): EFSA (2015)1: this is a limitation of all the dietary
cohort studies.
ARIC = Atherosclerosis Risk in Communities; BMI = body mass index; Ca = calcium; CI = confidence interval; CKD = chronic kidney disease;
eGFR = estimated glomerular filtration rate; FFQ = food frequency questionnaire; FGF-23 = fibroblast growth factor-23; FMD = flow-mediated dilation;
GI = gastrointestinal; Hb = hemoglobin; HDL = high-density lipoprotein; HR = hazard ratio; ICD = International Classification of Diseases;
LOAEL = lowest-observed-adverse-effect level; MESA = Multi-Ethnic Study of Atherosclerosis; NHANES III = Third National Human and Nutrition
Examination Survey; NIED = Nutritional and Inflammatory Evaluation in Dialysis; NOAEL = no-observed-adverse-effect level; nPCR = normalized protein
catabolic rate; P = phosphorus; PTH = parathyroid hormone; USDA = U.S. Department of Agriculture; WBC = white blood cell.
43
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Skeletal/Bone Effects in Humans
Table 7 summarizes studies that examined potential associations between dietary intake
of phosphate and effects on the skeleton/bone. Five prospective cohort studies met inclusion
criteria; of these, one study was of adults and the remaining four were birth cohort studies. The
available data suggest a potential association between P intake and improved bone health, but
several important confounding factors limit the conclusions that can be drawn.
In a cohort study of 46- to 68-year-old men (Eimstahl et al.. 1998). the odds of a bone
fracture were decreased with increasing dietary P intake during 2.4 years of follow-up.
The first birth cohort study (Bounds et al.. 2005) evaluated the association between
children's P intake and bone development in a cohort of 52 children. P intake was evaluated by
FFQs completed by the children's mothers on eight occasions between the ages of 2.3 and
8 years. In addition to the maternal exposure studies. Bounds et al. (2005) also observed a
positive association between P intake and bone mineral content (bmc) at age 8 years, and no
association with bone mineral density (bmd). Three additional birth cohort studies (Jones et al..
2000 as cited in EFSA. 2015; Heppe et al.. 2013; Yin et al.. 2010; Tobias et al.. 2005) examined
whether maternal intake of P during pregnancy influenced bmd or bmc in children between 8 and
16 years of age. In these studies, maternal intake of P was either positively associated with
measures of bmc or bmd.
One controlled exposure study that met inclusion criteria also examined bone endpoints.
Grimm et al. (2001) used a cross-over design to evaluate the effects of 6 weeks of
P supplementation (as sodium phosphate tablets and orange juice fortified with P) on markers of
bone metabolism in young adult women (23-39 years old). The supplementation period was
preceded by a 4-week period in which mean P intake was controlled at 1,700 mg P/day.
Compared with the control period, supplementation with mean P at 3,008 mg P/day did not
induce significant differences in serum minerals or hormones, although serum osteocalcin was
significantly lower in the washout period than in the control period. Levels of urinary
pyridinoline and deoxypyridinoline were higher during supplementation, but marked
interindividual variability was noted, and differences from the control period were not
statistically significant.
In general, cross-sectional studies (Lee and Clio. 2015; Lee et al.. 2014; Haraikawa et al..
2012; I to et al.. 2011; Nakamura et al.. 2004; Mendez et al.. 2002; Whiting et al.. 2002; Brot et
al .. 1999) showed that P intake was either not associated with measures of bone health, or a
positive association was noted between higher P intake and improved bone health.
Evaluation of the impact of P intake on bone health is confounded by the influence of Ca.
Brot et al. (1999) observed positive associations between Ca:P intake ratio and bmc and bmd in
peri-menopausal women, implying that a low Ca:P ratio is associated with lower bmc and bmd.
Thus, increasing P levels while Ca intake remains unchanged could have negative effects on
bone, but the available data are inadequate to support this conclusion. Finally, inorganic
phosphate is a significant source of inorganic acid load in the diet [as reviewed by Calvo and
Tucker (2013)1. Dietary acid load was associated with bone catabolism in some studies [as
reviewed by Calvo and Tucker (2013)1. suggesting one possible mechanism by which higher
intake of inorganic phosphates could adversely affect bone status.
44 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
In summary, available human data do not provide clear evidence for an association
between dietary intake of P and toxicological effects on bone health. Interpretation of the
available data is confounded by the role of adequate Ca intake, dietary acid load, and varying
bioavailabilities of different dietary sources of P.
45 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 7. Selected Studies Evaluating Associations between Phosphorus Intake and Bone Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations"
Dietary cohort studies of adults
Elmstahl et al. (1998)
(Sweden):
6,576 randomly selected male
residents of Malmo, aged
46-68 yr, recruited between
1991 and 1994 into the
Malmo Diet and Cancer
cohort. Body weight was not
reported. Health conditions
were not reported.
Intake for prior year was
assessed using validated FFQ
and 7-d menu book. Content
estimates obtained from the
Swedish Food Data Base from
the National Food
Administration.
Incident fractures
occurring between 1991
and 1995 and verified
by x-ray were identified
from registry in the one
local hospital with a
radiology and
orthopedics department.
Lower P intake (<1,357 mg P/d) was
associated with an increased odds of
bone fracture (RRs ranged between
0.5 and 0.67 in Quintiles 2-5
compared with quintile 1).
Covariates in final model: Age,
education, marital status, ethnicity,
physical activity at work, history of
myocardial infarction, stroke or
hypertension (yes/no), smoking,
intakes of energy, fat, vitamin D, and
Ca, and of zinc and P by quintiles.
Strengths: Food frequency
estimate for prior year;
control for nutrient covariates
potentially related to
fractures.
Limitations: Brief follow-up
period (mean 2.4 yr); no
consideration of P source.
Birth cohort studies of dietary intake
Bounds et al. (2005)
(United States):
52 healthy children (25 males
and 27 females) and their
mothers, participants in 8-yr
longitudinal study of
children's diet and growth.
The mean ages were 8.08 yr
for male children, 8.14 yr for
female children, and 38.0 yr
for the mothers. The mean
body weights were 30.5 kg
(boys), 27.9 kg (girls), and
69.0 kg (mothers).
FFQ (two food records and one
24-h recall) completed by
mothers at nine in-home
interviews (when children
were ages 2.3, 2.8, 3.5, 4, 4.5,
5, 6, 7, and 8 yr). Mothers
were trained in estimating
portion sizes and recording
intake. Frequencies averaged
across 3 d and mean of nine
averages was used as
longitudinal estimate of intake.
Content estimate sources not
reported (reported elsewhere).
Mean intake across
time = 1,063 mg P/d.
Children's bmc and
bmd were measured by
DXA scan at 8 yr of
age.
Children's dietary intake of P was
positively associated with bmc
(P = 0.11;/? = 0.01) but not bmd, at age
8 yr.
Covariates considered: Intakes of
energy, Ca, vitamin D, P, protein,
magnesium, vitamin C, vitamin K,
zinc, iron, and caffeine; children's
level of sedentary activity; children's
sex, height, weight, BMI, and age; and
mothers' total bmc or bmd.
Strengths: Longitudinal food
frequency estimates.
Limitations: Small sample
size; did not distinguish
between organic and
inorganic P sources; serum P
not measured.
46
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 7. Selected Studies Evaluating Associations between Phosphorus Intake and Bone Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations"
Henoe et al. (2013)
(Netherlands):
2,819 mothers and their
children (embedded in the
Generation R study); mean
maternal age at recruitment
was 31 yr; children were
assessed at median age of
6 yr. Body weights were not
reported. The mean ages were
31.6 and 31.5 yr for the
Generation R males and
females, respectively. Health
conditions were not reported.
Maternal intake was assessed
using validated
semi-quantitative FFQ
(293 items) administered at
enrollment (median 13.5 wk of
gestation) and covering diet
patterns over the previous
3 mo; content estimates were
obtained from the 2006 Dutch
Food-Composition Table.
Mean intake = 1,443 mg P/d.
Children's total body
bmd, bmc, and bone
area were measured by
DXA scan at 6 yr of
age. Analyses used bmd
and bmc for total body
minus head.
Increased P intake during pregnancy
was associated with higher childhood
bmc and bmd (/Mrcnd across quintiles
<0.001 for both). Quintile intake
values were not reported.
Strengths: Food frequency
estimate for prior 3 mo.
Limitations: Intake quintiles
were not reported so it is
uncertain whether the intake
in the reference group was
adequate or deficient. Did not
consider children's Ca, P, or
other nutrient intake after
birth.
Tobias et al. (2005)
(England):
4,451 mothers and their
children, members of
ALSPAC birth cohort;
children assessed at mean age
of 9-10 yr. The study had
two stages. Male mean
weights were 33.5 kg at
116 mo in stage 1 and 35.7 kg
at 116 mo in stage 2. Female
mean weights were 31.4 kg at
116 mo in stage 1 and 36.9 kg
at 116 mo in stage 2. Health
conditions were not reported.
Maternal intake was assessed
using FFQ (75 items)
administered at 32-wk
gestation; P content estimates
were obtained from British
food tables.
Mean intake = 1,339 mg P/d.
Children's total body
and spine-only bmd,
bmc, and area-adjusted
bmc were measured by
DXA scan at ~9 yr of
age. Spine data
available for
2,466 children.
Analyses used bmc for
total body minus head
and areal bmd.
In univariate analysis, increased P
intake during pregnancy was
associated with higher total body and
spine bmc and bmd, but not
area-adjusted bmc. In multivariate
models, P intake was not associated
with these metrics.
Covariates in final model: Age at DXA
scan, sex, pubertal stage (girls only),
housing, maternal and paternal
education, and social class.
Strengths: Control for
important covariates.
Limitations: Did not consider
children's Ca, P, or other
nutrient intake after birth.
47
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 7. Selected Studies Evaluating Associations between Phosphorus Intake and Bone Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations"
Jones et al. (2000) as cited in
EFSA (2015); Yin etal.
(2010) (Australia):
216 mothers and their
children, participants in birth
cohort established in
1988-1989. 70% of the
children in the study were
male. The average body
weight was 67.9 kg. Health
conditions were not reported.
Maternal intake during third
trimester assessed using one of
two self-administered FFQs
(151 and 179 items) shortly
after birth; P content estimates
were obtained from 1991
Australian Tables of Food
Composition.
Mean intake = 2,314 mg P/d.
Children's femoral
neck, lumbar spine, and
total body bmd were
measured by DXA scan
at ases 8 (Jones et al..
2000 as cited in EFSA.
2015) and 16 vr l( Yin et
al.. 2010)1.
Femoral neck and lumbar spine bmd
were positively associated with
maternal P diet density at age 8 yr; no
measure of bmd at age 16 yr was
associated with maternal P diet
density.
Covariates in final model (Yin et al..
2010): Sex. weisht at ase 16 vr.
sunlight exposure in winter at age
16 yr, sports participation, child's
current Ca intake, Tanner stage, ever
breast-fed, smoking during pregnancy,
maternal age at the time of childbirth.
Strengths: Adjusted for
child's Ca intake.
Limitations: Mean
self-reported intake of P
much higher than average in
Australian pregnant women;
did not consider children's P
or other nutrient intake after
birth.
48
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 7. Selected Studies Evaluating Associations between Phosphorus Intake and Bone Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and Limitations"
Controlled exposure studies
Grimm et al. (2001)
(Germany):
10 healthy women aged
23-29 yr, (mean 25), mean
body weight of 51-69 kg
(mean 60 kg), with typical P
intakes between 1,060 and
1,810 mg P/d based on 14-d
dietary records; crossover
design. Health conditions
were not reported.
After a 4-wk control period
(with mean intakes of
1,700 mg P/d and 1,500 mg
Ca/d), participants were given
NaH2P04 tablets (containing
620 mg P) and orange juice
containing 975 mg P (bringing
their total daily intake to
3,008 mg P) for 6 wk; this
period was followed by a 4-wk
washout period in which
intakes were similar to the
control period. Ca intake
during supplementation was
higher (1,995 mg Ca/d) during
supplementation than during
the control periods (-1,500 mg
Ca/d).
Total mean intake during
supplementation: 3,008 mg
P/d.
Symptoms were
recorded during
supplementation. Blood
and urine samples were
collected at the end of
each period and halfway
through the
supplementation period.
Serum minerals and
bone-related hormones
were measured, as were
urinary levels of
pyridinium crosslinks
(markers of bone
resorption),
microalbumin,
al-microglobulin, and
(^-microglobulin.
Participants reported diarrhea, soft
stools, and intestinal disturbances
during supplementation. No significant
differences in serum minerals or
hormones, although serum osteocalcin
was significantly lower in the washout
compared with the control period.
There was a tendency toward higher
levels of urinary pyridinoline and
deoxypyridinoline during
supplementation, but there was
significant interindividual variability.
A decrease in urinary microalbumin
during treatment was not statistically
significant.
LOAEL = 50.13 mg P/kg-d (including
diet) for GI distress, based on reported
mean initial body weight of 60 kg.
Strengths: Duration of 6 wk.
Limitations: Small sample
size; change in Ca intake
during supplementation.
"Nutrient databases in general are believed to underestimate P intake [as reviewed by McClure et al. (2017): EFSA (2015)1: this is a limitation of all the dietary
cohort studies.
ALSPAC = Avon Longitudinal Study of Parents and Children; bmc = bone mineral content; bmd = bone mineral density; BMI = body mass index;
Ca = calcium; DXA = dual-energy x-ray bone densitometry; FFQ = food frequency questionnaire; GI = gastrointestinal;
LOAEL = lowest-observed-adverse-effect level; NaH2P04 = monosodium phosphate; P = phosphorus; RR = relative risk.
49
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Other Noncancer Effects in Humans
Early Menarche
In a prospective study examining whether dietary intake of P was associated with
premature menarche (defined as menarche before 12 years of age), Ramezani Tehrani et al.
(2013) followed 134 prepubertal girls for a median duration of 6.5 years, assessing their pubertal
status at clinical interviews every 3 years (Table 8). Dietary intake of P was estimated using
FFQs administered at baseline. In this study, higher dietary intake of P (>647 mg P/day) was
associated with higher odds of premature menarche (OR = 3.43, 95% CI = 1.45-8.13). This
study had several important limitations, especially the very small population size (n = 134), the
wide range of ages at entry into the cohort (4-12 years at baseline), and differences in age at
baseline between children exhibiting early menarche (mean age 8 years at baseline) and those not
exhibiting early menarche (mean age 10 years). Because follow-up examinations took place
every 3 years, many of the girls who entered the cohort at mean age 10 years were not followed
at all until after the age cutoff for early menarche (12 years). No other studies evaluating onset of
menarche and dietary intake of P were identified.
Gastrointestinal Tract
OSP preparations have been used therapeutically to treat constipation and for bowel
cleansing prior to colonoscopy; thus, their laxative effects are well established. Additional GI
symptoms, including nausea, vomiting, GI distress, and diarrhea (Cheng et al.. 2016; Haas et al..
2014; Manukvan et al.. 2011; Seo et al.. 2011b; Yakut et al.. 2010; Patel et al.. 2009; Belooseskv
et al .. 2003; Fine et al.. 1998). have been reported in patients using preparations for bowel
cleansing. The LOAEL associated with these symptoms after colonoscopy preparation is the
same (164 mg P/kg-d) as that identified for risk of acute renal failure under these circumstances
(see "Renal Effects in Humans" section above).
For acute OSP use (1-3 days) to treat constipation, a LOAEL of 40 mg P/kg-d can be
identified based for laxative effects in children ages 5-9 years (see additional information in the
"Renal Effects in Humans" section above). Although laxation is the desired effect in persons
voluntarily using sodium phosphate medications, it would be considered to be a toxicological
effect in persons unwittingly exposed.
GI symptoms were evaluated in three longer duration controlled exposure studies (Chang
et al.. 2017; Medoff et al.. 2004; Grimm et al.. 2001). In a randomized, double-blind crossover
trial, GI symptoms occurred at about the same frequency in the treated and untreated periods in
31 patients with early CKD who were treated for 3 weeks with food or beverages with or without
phosphate additives (998 mg P/day; see Table 6) (Chang et al.. 2017). When 10 healthy female
volunteers were given supplemental P (as sodium phosphate tablets containing 620 mg P and
orange juice containing 975 mg P) for 6 weeks after a 4-week control period, participants
reported higher incidences of GI disturbances during the supplementation period compared with
the control period (Grimm et al.. 2001) (see Table 7). Finally, in 43 subjects with irritable bowel
syndrome presenting primarily as constipation, higher incidences of GI symptoms were reported
in the group receiving between 4 and 12 sodium phosphate tablets per day (3,541 mg P/day
based on initial target of 8 tablets/day) compared with the group receiving 2-6 tablets/day
(1,770 mg P/day based on initial target of 4 tablets/day) (Medoff et al.. 2004) (Table 8).
50 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Colonic mucosal abnormalities believed to be attributable to the oral use of sodium
phosphate for bowel cleansing have been reported (Coton et al.. 2011; also reviewed by Betsey.
2008; Atkinson et al.. 2005; Reichrt et al.. 2004; Atkinson and Hunter. 2002; Watts et al.. 2002;
Driman and Preiksaitis. 1998). These abnormalities are described as shallow aphthous
ulcerations or raised nonulcerated rings [as reviewed by Belsev (2008)1. Their cause has not been
established. The incidence of such lesions in patients using sodium phosphate for bowel
preparation is reported to range between 2.6 and 24.5% (data not shown)[as reviewed by Belsev
(2008)1. In one analysis, Reichrt et al. (2004) observed aphthous lesions potentially associated
with sodium phosphate in 21 of 730 colonoscopy patients (data not shown). To rule out
inflammatory bowel disease as the cause of the lesions, the study authors followed the patients
for 3 years after colonoscopy. None of the affected patients were diagnosed with inflammatory
bowel disease during that time (Reichrt et al. (2004). providing support for attributing the cause
of the lesions to sodium phosphate ingestion. Further, the incidence of the lesions was much
higher after sodium phosphate use than bowel preparation with PEG in a randomized controlled
trial (/was et al.. 1996 as cited in Belsev. 2008). However, in the absence of colonoscopy or
histopathology data from the same patients prior to sodium phosphate use, it is difficult to assign
causation. Importantly, Belsev (2008) indicated that these lesions did not have clinical
consequences and typically resolved without any medical intervention.
Anemia
In a cross-sectional study, Samuel et al. (2013) observed a significantly higher prevalence
of anemia in 366 pregnant women consuming dietary P at the highest tertile of intake
(>1,243.3 mg P/day), compared with the lowest tertile (<1,147.1 mg P/day) (data not shown). No
additional information pertaining to the potential relationship between anemia and inorganic
phosphate intake in humans was identified in the literature reviewed.
51
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 8. Selected Studies Evaluating Associations between Phosphorus Intake and Other Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and
Limitations3
Dietary cohort studies
Ramezani Teltrani et al.
(2013) (Iran):
134 prepubertal girls; aged
4-12 yr (mean 8.9 yr) at
baseline (participants in
Tehran Lipid and Glucose
Study). Body weights were
not reported. Health
conditions were not reported.
Children's intake was assessed
using two nonconsecutive 24-h
dietary recall FFQs
administered by dieticians;
P content estimates were
obtained from nutritionist III
software modified for the
Iranian food consumption table.
Median intake = 647 mg P/d.
Early (defined as
<12 yr of age)
menarche determined
by clinical interview at
follow-up
examinations every
3 yr (median follow-up
6.5 yr).
Higher P intake (>647 mg P/d) was
associated with increased odds of reaching
menarche <12 yr of age; OR = 3.43
(1.45-8.13).
Covariates in final model: Energy and protein
intake at baseline, interval between the age at
study initiation and the age of menarche,
maternal age at menarche, BMI Z-score at
baseline, and height Z-score at baseline.
Strengths: Control for
important covariates.
Limitations: Small
population size; age at
baseline differed
between children with
and without early
menarche (8 vs. 10 yr,
p = 0.001); dietary
intake assessed only at
baseline; short-term
intake estimates used;
socioeconomic status
not considered but
known to be related to
age at menarche in
Iran.
52
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 8. Selected Studies Evaluating Associations between Phosphorus Intake and Other Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates
Considered
Strengths and
Limitations3
Controlled exposure studies
Medoff et al. (2004)
(United States):
43 patients with irritable
bowel syndrome (IBS),
constipation predominant;
recruited by researchers who
were also gastroenterologists;
open-label dose ranging
study. Patients were grouped
for treatment. Group A was
15 females/3 males with a
mean age of 46 yr and a mean
body weight of 155 lb.
Group B was
20 females/5 males with a
mean age of 49 yr and a mean
body weight of 162 lb. Health
conditions other than IBS
were not reported.
Participants were allocated to
two groups, receiving either
four or eight sodium phosphate
tablets (each containing 1.102 g
NaH2P04 monohydrate and
0.398 g Na2HPC>4 anhydrous)
for 28 d. The tablets each
provided 443 mg P, yielding
doses of 1,770 or 3,541 mg P/d
when four or eight tablets
(respectively) were given.
Patients were allowed to
increase or decrease the dose
depending on symptoms, and in
the end, the groups had
consumed between 2 and 6 or
between 4 and 12 tablets/d.
Patients consumed their normal
diets during treatment (P and Ca
content not reported).
Intake: 1,770 or 3,541 mgP/d
(not including diet).
Efficacy was assessed
by relief of
constipation; safety
was assessed by
monitoring of serious
toxicological events,
patient-reported
symptoms, and
measurement of body
weight, heart rate,
blood pressure, and
serum chemistry.
Symptoms reported in the low-dose group
included nausea, diarrhea, and incomplete
evacuation (1/18 each). In the high-dose
group, nausea and diarrhea were reported by
4/25 and 3/25, respectively, and additional
symptoms included bloating (2/25) and
cramping, headache, lower back pain,
migraine, and lower quadrant pain
(1/25 each). No effects on body weight, heart
rate, or blood pressure. The only serum
chemistry change that was considered
noteworthy and related to treatment was
occasional hypokalemia in five patients (in
both dose groups). The study authors
concluded that the low dose was well
tolerated, suggesting that a starting dose of
two to four tablets/day (equivalent to
885-1,770 mg P/d) would be appropriate.
LOAEL = 48.2 mg P/kg-d (not including
diet) for GI distress and hypokalemia, based
on reported mean body weights of 70.3 and
73.5 kg in low- and high-dose groups
(respectively).
Strengths: Duration of
4 wk.
Limitations: Open
label design may
confound symptom
reporting; actual doses
varied from initial
targets; dietary P and
Ca intake not reported.
"Nutrient databases in general are believed to underestimate P intake [as reviewed by McClure et al. (2017): EFSA (2015)1: this is a limitation of all the dietary
cohort studies.
BMI = body mass index; Ca = calcium; Na2HP04 = disodium phosphate; FFQ = food frequency questionnaire; GI = gastrointestinal;
LOAEL = lowest-observed-adverse-effect level; NaH2P04 = monosodium phosphate; OR = odds ratio; P = phosphorus.
53
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Cancer in Humans
Tables 9 and 10 show dietary cohort and case-control studies (respectively) of cancer that
met criteria for dose-response consideration.
Prostate Cancer
The most robust study of prostate cancer risk and dietary phosphate intake was published
by Wilson et al. (2015). In this large, prospective cohort study of 47,885 men in the Health
Professionals Follow-Up Study, dietary phosphate intake was assessed by FFQ every 4 years
during the 24-year follow-up period. Incident prostate cancer was evaluated every 2 years and
confirmed by review of medical records and pathology reports. After control for covariates that
included known confounders as well as dietary sources of P (dairy and animal protein intake),
the risk of incident prostate cancer was significantly increased in the fourth and fifth quintiles of
P intake (mean intake 1,783 mg P/day), compared with the first (relative risk [RR] = 1.13,
95% CI = 1.00-1.27), with a significant (p = 0.04) dose-response trend (data not shown). In
addition, in analyses stratified by cancer stage, dietary P intake in the highest quintile was
associated with elevated risks of lethal cancer (resulting in death or distant metastases;
RR = 1.43, 95% CI = 1.02-1.99), advanced-stage cancer (RR = 1.45, 95% CI = 1.08-1.94), and
high-grade cancer (based on Gleason scores; RR = 1.51, 95% CI = 1.06-2.17) after adjustment
for confounders (data not shown). A significant dose-response trend was seen only for high-
grade cancer (p = 0.01).
Three earlier prospective cohort studies of dietary phosphate intake did not observe
significant associations with incident prostate cancer (Kesse et al.. 2006; Chan et al.. 2000) or
incident invasive prostate cancer (Tseng et al.. 2005). Unlike Wilson et al. (2015). these studies
suffered several limitations (see Table 9), most notably the relatively brief follow-up duration.
Follow-up averaged approximately 8 years in studies by Kesse et al. (2006) and Tseng et al.
(2005). and was between 5 and 8 years in the study by Chan et al. (2000). compared with
24 years in the study by Wilson et al. (2015). In addition, the earlier studies evaluated smaller
populations, and did not account for the varying dietary P sources and their bioavailability
differences.
In a case-control study (n = 1,294 cases and 1,451 controls) of prostate cancer and dietary
P intake (see Table 10), Tavani et al. (2005) did not observe a significant association between the
odds of prostate cancer and higher dietary phosphate intake (OR = 1.20, 95% CI = 0.79-1.84),
when the highest quintile of intake (>1,897 mg P/day) was compared with the lowest
(<1,204.66 mg P/day). Case-control studies can be subject to recall bias if exposure estimates are
affected by diagnosis. However, recall bias is less likely in this study because phosphate intake
was through an FFQ rather than by direct measurement. It is likely that this study suffers from
misclassification of intake because the subjects were asked to remember their diets for the 2-year
period prior to disease onset; random misclassification could bias the result toward the null.
No studies examining the association between serum P and prostate cancer were
identified in the literature search or published reviews that considered cancer (Brown and
Razzaque. 2015; Anderson. 2013; Jain and HI saved. 2013).
54 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Colorectal Cancer
As shown in Tables 9 and 10, one prospective cohort study Kesse et al. (2005) and one
case-control study van l.ee et al. (2011) each observed decreased risk of colorectal adenoma or
cancer with higher dietary P intake. The adjusted RR for colorectal adenoma in the highest
quartile of dietary P intake (>1,633.84 mg P/day) was 0.70 (95% CI = 0.54-0.90) compared with
the lowest quartile (<1,141.86 mg P/day) in a cohort of 73,034 women (Kesse et al.. 2005).
However, the follow-up period was brief (2-7 years), and dietary intake was assessed only once.
van Lee et al. (2011) observed a nonsignificant decrease in the adjusted OR for colorectal cancer
(OR = 0.78, 95% CI = 0.58-1.05) in the highest quintile of intake (>1,755.84 mg P/day)
compared with lowest quintile (<1,335.29 mg P/day). Trend analysis showed a significant
(p = 0.016) trend for decreased OR with increasing P intake. As with the case-control study of
prostate cancer discussed above (Tavani et al.. 2005). this study is more likely to suffer from
random misclassifi cation of exposure than recall bias; in this study (van Lee et al.. 201 1). cases
and controls were asked to fill out FFQs designed to recount diets 10 years earlier.
Bladder Cancer
Another analysis of data from the Health Professionals Follow-Up Study, Michaud et al.
(2000) evaluated the risk of incident bladder cancer in 47,909 male cohort members. Dietary
intake of P was evaluated by FFQ at baseline (in 1986) and again 4 years later (1990). During
12 years of follow-up, the risk of bladder cancer was not significantly affected (adjusted
RR = 0.85, 95% CI = 0.57-1.21) in the highest quintile of P intake (median intake 1,728 mg
P/day) compared with the lowest quintile (median intake 1,101 mg P/day). In addition, the test
for dose-response trend was not significant (p = 0.40).
55
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 9. Selected Cohort Studies Evaluating Associations between Dietary Phosphorus Intake and Cancer Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates Considered
Strengths and
Limitations3
Prostate cancer
Wilson etal. (2015)
(United States):
Prospective cohort study of
47,885 male participants of
the HPFS (recruited in 1986;
aged 40-75 yr at baseline).
Intake was assessed by
semiquantitative FFQ
(130 items); administered
every 4 yr between 1986
and 2006 and nutrient
content of each food.
Quintiles with mean
intakes of 1,079, 1,248,
1,365, 1,499, and
1,783 mgP/d.
Incident prostate cancer
between entry and 2010
(24 yr of follow up); was
initially determined by
self-report (participant or
next of kin) on biennial
questionnaire and
confirmed by review of
medical records and
pathology reports. Deaths
were determined by
questionnaire and National
Death Index; cause of death
was determined by review
of all available data,
including death certificate.
Dietary P intake was associated with an increased risk
of prostate cancer (all and high-grade subcategory);
p for dose-response trend = 0.04; adjusted RR
>1.11 and significant atp< 0.05 in fourth and
fifth quintiles compared with first quintile. In
addition, in analyses stratified by cancer stage, dietary
P intake in the highest quintile was associated with
elevated risks of lethal cancer (resulting in death or
distant metastases; RR = 1.43, 95% CI = 1.02-1.99),
advanced-stage cancer (RR = 1.45, 95%
CI = 1.08-1.94), and high-grade cancer (based on
Gleason scores; RR = 1.51, 95% CI = 1.06-2.17) after
adjustment for confounders (data not shown). A
significant dose-response trend was seen only for
high-grade cancer (p = 0.01).
Covariates in final model: Age in months; calendar
time; race; height; BMI at age 21 yr; current BMI;
vigorous physical activity; smoking; diabetes; family
history of prostate cancer; intakes of tomato sauce,
a-linolenic acid, supplemental vitamin E, and alcohol;
energy intake; multivitamin use; and history of
prostate-specific antigen testing; and Ca, dairy, and
animal protein intake.
Strengths: Collection of
multiple FFQs over time,
high follow-up rates,
large sample size,
consideration of known
confounders,
consideration of
P source; 24-yr
follow-up.
56
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 9. Selected Cohort Studies Evaluating Associations between Dietary Phosphorus Intake and Cancer Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates Considered
Strengths and
Limitations3
Chanet al. (2000) (Finland):
Prospective cohort study of
27,062 male participants in
ATBC (randomized
2x2 trial of
alpha-tocopherol and
beta-carotene for prevention
of lung cancer among
smokers); randomly selected
from residents of
southwestern Finland
between 1985 and 1988;
median age was between
56.5 and 57.6 yr for quintiles
of energy-adjusted Ca intake.
Intake was assessed by
FFQ (276 items)
administered at baseline
and asking about intake
over the previous 12 mo;
content estimates were
obtained from database at
National Public Health
Institute of Finland.
Incident prostate cancer
between 1985 and 1993,
initially identified via
Finnish Cancer Registry
and Register of Causes of
Death; diagnosis, and stage
verified by review of
medical records and
histopathology and
cytology specimens.
Dietary P intake was not independently associated
with prostate cancer risk (adjusted RR = 0.8, 95%
CI = 0.4-1.5 comparing highest to lowest quintile of
intake: p for trend = 0.11).
Covariates in final model: Age, smoking, BMI, total
energy intake, education, and supplementation group
(alpha-tocopherol, beta-carotene, both, or placebo).
Strengths: Large sample
size.
Limitations: Dietary
intake evaluated only at
baseline; relatively brief
follow-up; did not
account for family
history.
Kesse et al. (2006) (France):
Prospective cohort study of
2,776 male participants in the
SU.VI.MAX trial; mean ages
of cases and controls were
57.1 and 53.3 yr,
respectively.
Intake was assessed by
24-h FFQ (number of
items not reported)
administered every 2 mo
for a total of 5 times in the
first 18 mo of the study;
content estimates were
obtained from 2005 food
composition table.
Quartiles of
energy-adjusted intake
were <1,167,
1,167-1,291,
1,291-1,434, and
>1,434 mgP/d.
Incident prostate cancer
during 7.7 yr (median) of
follow-up; determined by
self-report at annual
follow-up or by death
certificate. Cases were
verified by independent
review of pathology report.
Dietary P intake was not significantly associated with
increased prostate cancer risk in any group (adjusted
RR comparing highest quartile of intake with lowest
quartile = 1.83, 95% CI = 0.89-3.73), although the
trend for RR across quartiles was marginally
significant (p for trend = 0.04). A significant
interaction was seen between P and Ca intake (p for
interaction, n = 0.02).
Covariates in final model: Occupation, group of
treatment, smoking status, overall physical activity
level, energy from fat, energy from other sources,
alcohol intake, BMI, and family history of prostate
cancer in first-degree relative.
Strengths: Collection
of multiple FFQs over
time, consideration of
known confounders.
Limitations: Relatively
small size and brief
follow-up.
57
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 9. Selected Cohort Studies Evaluating Associations between Dietary Phosphorus Intake and Cancer Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates Considered
Strengths and
Limitations3
Tseng et al. (2005)
(United States):
Prospective cohort study of
3,612 male participants in
National Health and Nutrition
Examination Epidemiologic
Follow-up Study; recruited
between 1982 and 1984
during follow-up for
longitudinal study begun
1971-1975; mean age at
baseline was 57.8 yr.
Intake was assessed by
FFQ (105 items)
administered at baseline
(1982-1984); source for
content estimates was not
reported.
Mean intake was
1,317 mg P/d; median
intakes by tertile were:
984.0, 1,218.9, and
1,443.3 mg P/d.
Incident invasive prostate
cancer between 1982 and
1984 and 1992 (mean
7.7 yr of follow-up). Cases
were determined by:
(a) self-report at interviews
in 1986, 1987, or 1992;
(b) at least one hospital
stay with diagnosis coded
as invasive prostate cancer;
or (c) death certificate with
underlying or
nonunderlying cause of
death coded as invasive
prostate cancer.
Dietary P intake was not significantly associated with
prostate cancer risk after adjustment for covariates
including Ca (p for trend = 0.77; adjusted RR
comparing highest to lowest tertile of P intake = 0.9,
95% CI = 0.5-1.6).
Covariates in full model with Ca: Age, race, energy
intake, and design variables; U.S. region; rural, urban,
or suburban residence; education; recreational sun
exposure; recreational and usual level of physical
activity; smoking status; current alcohol intake; and
Ca intake.
Strengths: Consideration
of many known
confounders.
Limitations: Dietary
intake evaluated only at
baseline; relatively small
size; relatively brief
follow-up; did not
account for family
history.
Other cancers
Kesse et al. (2005) (France):
Prospective cohort study of
73,034 female participants in
the E3N-EPIC cohort
established in 1990; mean
ages at baseline ranged
between 53 and 57 yr.
Adenoma study included
516 cases (including
175 with high-risk adenomas)
and 4,804 polyp-free
noncases; colorectal cancer
study included 172 cases and
67,312 noncases.
Intake assessed by
validated FFQ (208 items)
administered once
between 1993 and 1995;
content estimates obtained
from French national
database.
Intakes by quartile were:
<1,141.86,
1,141.86-1,374.86,
1,374.86-1,633.83, and
>1,633.83 mg P/d.
Incident colorectal
adenoma between 1993 and
1995 and December 1997,
or incident colorectal
cancer between 1993-1995
and June 2000 (2-7 yr of
follow-up); initially
determined by self-report
and verified by pathology
report.
Dietary P intake was inversely associated with risk of
adenoma (p for trend = 0.005; adjusted RR comparing
highest to lowest quartile = 0.70, 95%
CI = 0.54-0.90). Dietary P intake was not associated
with risk of high-risk adenoma or colorectal cancer
(p for trend = 0.23 and 0.11, respectively).
Covariates in final model: Education level, current
smoking status, family history of colon cancer, BMI,
physical activity level and energy, and alcohol intake
(at time of dietary questionnaire).
Strengths: Large
population size.
Limitations: Dietary
intake assessed only
once; brief (2-7 yr)
follow-up; did not
account for source of
dietary P.
58
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 9. Selected Cohort Studies Evaluating Associations between Dietary Phosphorus Intake and Cancer Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates Considered
Strengths and
Limitations3
Michaud et al. (2000)
(United States):
Prospective cohort study of
47,909 male participants in
HPFS; recruited in 1986;
ages 40-75 yr at baseline;
men with cancers diagnosed
before 1986 were excluded.
Intake was assessed by
FFQ (131 items)
administered at baseline
(1986) and again in 1990;
questionnaire asked about
dietary patterns over prior
year; content estimates
obtained from USDA.
Median intakes by quintile
were: 1,101, 1,250, 1,364,
1,495, and 1,728 mgP/d.
Incident bladder cancer
between baseline (1986)
and January 31, 1998;
initially determined by
self-report and confirmed
by review of medical
records when possible;
unreported bladder cancers
identified by review of
National Death Index were
also included.
Dietary P intake was not associated with risk of
bladder cancer after adjustment for covariates (p for
trend = 0.40; adjusted RR comparing highest to lowest
quintile of P intake = 0.85, 95% CI = 0.57-1.21).
Covariates in final model: Age, pack-years of
smoking history, current smoking status, geographic
region of the United States, cruciferous vegetable
intake, and total fluid intake.
Strengths: Large
population size; 12-yr
follow-up.
Limitations: Dietary
intake evaluated only
twice.
"Nutrient databases in general are believed to underestimate P intake [as reviewed by McClure et al. (2017): EFSA (2015)1: this is a limitation of all the dietary cohort
studies.
ATBC = Alpha-Tocopherol Beta-Carotene Cancer Prevention Study; BMI = body mass index; Ca = calcium; CI = confidence interval; E3N-EPIC = European
Prospective Investigation into Cancer and Nutrition; FFQ = food frequency questionnaire; HPFS = Health Professionals Follow-up Study; P = phosphorus; RR = relative
risk; SU.VI.MAX = Supplementation en Vitamines et Mineraux Anti-oxydants; USDA = U.S. Department of Agriculture.
59
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 10. Selected Case-Control Studies Evaluating Associations between Dietary Phosphorus Intake and Cancer
Endpoints
Citation (location); Size and
Description of Population
Methods for P Intake
Estimation
Methods for Outcome
Assessment
Summary of Results and Covariates Considered
Strengths and
Limitations3
Tavani et al. (2005) (Italv):
1,294 men with incident
prostate cancer and
1,451 controls with
nonneoplastic disease or
injury; recruited in one of
four areas of Italy between
1991 and 2002; mean ages of
cases and controls were
66 and 63 yr, respectively.
Intake assessed by FFQ
(78 items); content
estimates obtained from
Italian food composition
databases.
Quintile upper cut
points = 1,204.66,
1,413.51, 1,609.22,
1,897.00, >1,897 mgP/d.
Eligible cases had incident
diagnosis of histologically
confirmed prostate cancer and
were admitted to one of the
major teaching and general
hospitals in one of the four
areas under study. Controls
were from the same area and
admitted to same network of
hospitals with nonneoplastic
disease or injury.
P intake was not associated with risk of prostate
cancer (adjusted OR = 1.20, 95% CI = 0.79-1.84
for highest quintile compared with lowest; p for
trend = 0.39).
Covariates in final model: Age, center, education,
body mass index, tobacco smoking, physical
activity, total energy, and family history of prostate
cancer.
Strengths: Large size.
Limitations: Potential for
recall bias (intake
assessed by questionnaire
after diagnosis).
van Lee etal. (2011)
(Australia):
577 cases of left-sided CRC,
277 cases of right-sided
CRC, and 958 age- and
sex-matched controls from
the Western Australian
Bowel Health Study
(recruited between 2005 and
2007); 59-61% men; mean
ages of cases and controls
were 64.94 and 64.60 yr,
respectively.
Intake was assessed by
FFQ (74 items) modified
to assess diet 10 yr earlier.
Content estimates were
obtained from Australian
Food Composition Tables.
Energy-adjusted intakes:
Mean= 1,606.9 mgP/d.
Quintile cut
points = <1,335.29,
<1,467.33, <1,606.80,
<1,755.84, and
>1,755.84 mgP/d.
Eligible cases had incident
diagnosis of adenocarcinoma
of the colon or rectum
reported to the Western
Australia Cancer Registry
between 2005 and 2007.
Tumor site was verified by
review of histology reports in
the Registry. Controls were
obtained from random sample
of the electoral roll in the area,
and frequency matched to age
and sex.
P intake was associated with a decreased risk of
colorectal cancer (adjusted OR = 0.78, 95%
CI = 0.58-1.05 for highest quintile compared with
lowest; p for trend = 0.016). When stratified by
location, the association was more evident for
right-sided cancers than for left-sided cancers.
Covariates in final model: Sex, age, BMI, smoking,
cholecystectomy, alcohol consumption, SEIFA,
and aspirin use.
Strengths: Large size.
Limitations: Potential for
recall bias (intake
assessed by questionnaire
after diagnosis).
"Nutrient databases in general are believed to underestimate P intake [as reviewed by McClure et al. (2017): EFSA (2015)1: this is a limitation of all the dietary studies.
BMI = body mass index; CI = confidence interval; CRC = colorectal cancer; FFQ = food frequency questionnaire; OR = odds ratio; P = phosphorus;
SEIFA = Socio-Economic Indexes for Areas.
60
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
2.1.2. Inhalation Exposures
Two occupational health studies with potential relevance to the hazard assessment of
inhaled inorganic phosphates were identified: Kheiifi et al. (2014) examined a small group of
phosphate mine workers, and Yiin et al. (2016) evaluated mortality in phosphate fertilizer plant
workers. Neither of these studies included any quantitative or qualitative estimates of exposure to
airborne inorganic phosphates. Both populations likely had substantial exposure to other
hazardous substances; Yiin et al. (2016) noted the following coexposures common to phosphate
mining and processing: uranium and radon daughters; sulfuric acids and acid mists; inorganic
mists containing sulfate, fluoride, and ammonium; and aerosols containing technically enhanced
naturally occurring radioactive materials (TENORMs). Some of the workers at the phosphate
fertilizer plant evaluated by Yiin et al. (2016) had been involved in uranium extraction during the
1950s. Kheiifi et al. (2014) reported that phosphate miners were also exposed to silica dusts and
often exhibited silicosis.
Kheiifi et al. (2014) reported significantly higher white blood counts and serum levels of
inflammatory markers (including IL-ip, IL-6, IL-8, MIP-ip, and LTB-4) in 12 male Tunisian
phosphate mine workers, when compared with levels in 8 unexposed male controls. The effects
were more pronounced in the subset of workers who smoked (data not shown). Yiin et al. (2016)
observed significantly increased standardized mortality ratios (SMRs) for all-cause mortality
(SMR = 1.07, 95% CI= 1.02-1.13), all-cancer mortality (SMR= 1.16, 95% CI = 1.06-1.28),
lung cancer (SMR = 1.32, 95% CI = 1.13-1.53), and leukemia (SMR = 1.74, 95%
CI = 1.11-2.62) in a cohort of 3,199 workers at a Florida facility producing phosphate fertilizers
(data not shown). However, due to the potential coexposures and lack of qualitative or
quantitative phosphate exposure estimates, it is not possible to ascertain the potential
contribution of inorganic phosphates to the observed effects.
2.2. ANIMAL STUDIES
2.2.1. Oral Exposures
The database of animal studies used to examine effects of oral intake of Na/K salts of
inorganic phosphates is substantial and highly varied. To prioritize studies for consideration in
the dose-response assessment, the following criteria were used to select the most relevant
studies:
1) The experimental animal was a common laboratory species for toxicology studies
(i.e., rat, mouse, hamster, rabbit, guinea pig, dog, monkey). Studies of pigs, cows,
sheep, and cats were not considered.
2) A suitable control/referent group was included.
3) The phosphate compound administered and the dose(s) (either in terms of compound
or in terms of P) were clearly and unambiguously reported.
4) Experimental animals used in the study were not genetically modified or pretreated to
induce a disease or injured state.
5) Report or publication is in English. Studies reported in foreign languages but with an
English abstract, tables, or summary in a secondary source are discussed briefly in
Section 2.3.2.
61 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
6) Concentrations of P and Ca in the baseline diet administered to the referent or
"control" and comparison groups were reported. The P data are necessary to ensure
that the total phosphate dose can be accurately characterized, and the Ca data are
needed to evaluate potential effects of altered Ca:P ratio, since Ca and P are
coordinately controlled, and Ca content influences P absorption and excretion.
Studies that did not provide the concentrations of Ca in the diet or did not report P
concentration in the referent/control group, are discussed in Section 2.3.2.
7) At least two exposure groups received diets with Ca and P levels that met minimum
nutritional requirements. As Ca and P are essential nutrients, intake below
requirements can result in health effects. P and Ca requirements1 for rats
(3-3.7 g P/kg diet and 5-6.3 g Ca/kg diet and) and guinea pigs (4 g P/kg diet and 8 g
Ca/kg diet) were obtained from MRC (1995). For studies in which the group reported
as the control received P or Ca below nutritional requirements, other exposure groups
were examined to determine whether another group could be designated as the
referent group. In these instances, quantitative health outcome information was
reanalyzed for statistical significance by comparing the exposure group(s) to the
designated referent group.
8) Ca intake was constant across the referent/control and all relevant exposure groups.
This requirement ensures any health outcomes observed can be attributed to changes
in P dose alone.
Appendix A provides a flow chart showing the disposition of animal studies from the
literature searches. In addition to the literature search results, the 2015 Cosmetic Ingredient
Review (CIR) Safety assessment of phosphoric acid and simple salts as used in cosmetics, EFSA
(HFSA. 2015. 2005), OECD S1DS Initial Assessment Reports for phosphoric acid and
dipotassium hydrogen phosphate (K2HPO4) (OECD. 2009. 2006). and published reviews by
Willhitc et al. (2013) and Wcincr et al. (2001) were consulted for animal studies on the subject of
monovalent (specifically sodium or potassium) salts of inorganic phosphates, including
unpublished studies.
Short-Term Studies
Hitchman et al. (1979)
In a study designed to evaluate nephrocalcinosis, female weanling Wistar rats (8/group)
were administered diets containing 0.2-1% P and 0.5 or 1% Ca for 6 weeks. Because the
minimum Ca and P requirements for rats are 0.5-0.63% (5-6.3 g Ca/kg diet) and 0.3-0.37%)
(3-3.7 g P/kg diet), respectively (MRC. 1995). only the groups administered 0.5%> P (as calcium
phosphate; designated as referent group) or 1% P (as a mixture of CaHPCU and NaFhPC^;
designated treated group) and 0.5% Ca (both groups) were considered for this analysis. The
proportions of calcium phosphate and NaH2P04 administered to the treated group were not
reported, but assuming that the additional phosphate in this group was derived from NaH2P04 is
reasonable. Based on reference food consumption (0.0164 kg/day) and body weight (0.156 kg)
for female Wistar rats in a subchronic study (U.S. EPA. 1988). 0.5 and 1% in the diet are
equivalent to approximately 530 and 1,100 mg P/kg-day. At sacrifice, kidneys were weighed and
1A definitive reference for Ca and P requirements in rabbits was not located, as discussed further in the summary of
the Ritskes-Hoitinga et al. (2004) rabbit study.
62 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
analyzed for P and Ca content. Kidney histopathology was examined (dose groups were not
specified).
In rats treated at 1,100 mg P/kg-day for 6 weeks, the study authors reported that renal Ca
concentrations and kidney weights were significantly higher than in the 530 mg P/kg-day group
(by 6.3-fold and 16%, respectively; p < 0.05 based on a two-sided /-test performed for this
review) (see Table B-l) (Hitchman et al.. 1979). The study authors noted "severe renal
calcification" in rats administered 1,100 mg P/kg-day (presumably based on Ca concentrations
only; histopathological results for this dose group were not reported). A LOAEL of 1,100 mg
P/kg-day is identified based on increased kidney weights and kidney Ca levels in female rats.
The LOAEL of 1,100 mg P/kg-day corresponds to a LOAEL (human equivalent dose [HED]) of
240 mg P/kg-day.2
Huttunen et al. (2007)
Effects on bone development were evaluated in male Wistar rats (10/group; 1 month old
at study initiation) administered 0.6 (referent), 1.2, or 1.8% inorganic phosphate in the diet for
8 weeks. The diet of all groups of rats (including referents) contained 20.21% CaHP04 (as the
only source of Ca; constituting 0.6% of the total diet). The diets of the 1.2- and 1.8%-groups
were created by adding 14.5 and 40.9% monopotassium phosphate (KH2PO4), respectively, to
the referent diet. Measured food consumption reportedly was 0.0184, 0.0177, and 0.0164 kg/day
for the referent, 1.2-, and 1.8%-groups, respectively. Based on the food consumption above and
reference (U.S. HP A. 1988) body weights (0.217 kg) for male Wistar rats, the U.S. EPA
estimates that the concentrations of 0.6, 1.2, and 1.8% in the diet are equivalent to approximately
510, 980, and 1,400 mg P/kg-day.3 Actual body-weight data were not presented in the published
study.
The study authors estimated food consumption on the basis of the weight of leftover
food; animals were offered 0.020 kg food/day, consistent with reference value food intakes for
rats (U.S. HP A. 1988). Although the results section of the study states that total food
consumption over the course of the 8-week study was 10.3, 9.9, and 9.2 kg for the referent, 1.2-,
and 1.8%-groups, respectively, these values translate to the daily consumption rates (in kg/day)
reported in the paragraph above around 10-fold greater than 0.020 kg/day; therefore, the total
food intakes reported in the study were apparently mistakenly inflated 10-fold, and the true
values were actually 1.03, 0.99, and 0.92 kg, respectively. These values were divided by 56 days
to get the daily food consumption rates used in the dosimetry calculations.
Body weights were measured four times during the study (Days 0, 28, 42, and 52). After
8 weeks, blood was collected to evaluate serum Ca, P, and PTH concentrations. Prior to study
initiation and at study termination, right femur bone area, bmc, and bmd were measured using
dual-energy x-ray bone densitometry (DXA). Bone labeling (with tetracycline) was performed
12 and 2 days prior to sacrifice; labeling was used to measure mineral apposition rate (MAR) in
2As outlined in the U.S. EPA's Recommended use of body weight3'4 as the default method in derivation of the oral
reference dose (U.S. EPA. 201151. the LOAEL was converted by the U.S. EPA to an HED of 240 mg/kg-day using
a dosimetric adjustment factor (DAF) of 0.22 (HED = adjusted daily dose [ADD] x DAF). The DAF was calculated
as follows: DAF = (BWa1/4 ^ BWh14). Quantitative body-weight data for rats (0.156 kg) and reference body weights
for humans (70 kg) recommended by U.S. EPA (1988) were used to calculate the DAF.
3Reported dietary intakes (% P in food) were converted to ADDs using the following equation: ADD = (% P in
food x 10,000 [mg P/kg food] x food intake [kg food/day]) average body weight.
63 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
longitudinal frontal sections of the distal (left) femur. Left femurs were subsequently processed
for histomorphometry (measurements of trabecular bone area, total trabecular perimeter,
trabecular width, osteoblast perimeter, and osteoclast count); right femurs were analyzed using
peripheral quantitative computed tomography (to measure bmd, bmc, and cross-sectional area
[CSA] of total bone, trabecular bone, and cortical bone for the distal metaphysis and midshaft).
The length of the right femur was measured; six right tibias were evaluated using microcomputed
tomography (to evaluate the trabecular total volume, bone volume, and bone surface; trabecular
separation and number, structure model index [SMI; a measure of plate:rod ratio in bone
architecture], connective structures per unit volume, and degree of anisotropy; cortical
cross-sectional area, thickness, and porosity). Right femoral necks and tibial shafts were
subjected to mechanical testing (including measurements of strength, stiffness, toughness, and
yield point).
Although the study indicated that 30 rats were randomized into three groups, tabular
results are presented for 9 (rather than 10) referent group rats; no explanation was provided
(Huttunen et al.. 2007). Rats treated at 980 and 1,400 mg P/kg-day reportedly showed significant
reductions in food consumption (4-11% lower than referent group) and body weights (reported
by the study authors to be 11 and 29% lower than controls; data not presented numerically) over
the course of the study.
The study authors reported that serum PTH (but not Ca or P) was significantly increased
in 1,400 mg P/kg-day rats (sevenfold change relative to referent group). Based on DXA
measurements, femur length was significantly decreased in rats treated at 1,400 mg P/kg-day
(8% shorter than referent); rats treated at 980 and 1,400 mg P/kg-day showed decreased femur
bmc or bmd (9-20% lower than referent; see Table B-2). Analysis of histomorphometric
parameters showed trabecular-related endpoints (bone area, width, and perimeter) were
significantly decreased (by about 27-72%), and osteoclast number and MAR were significantly
increased (by about 1.4- to 3.3-fold) in treated rats compared with those in the referent group;
these effects were dose-related (data were shown graphically).
In the distal metaphysis (of right femurs), the study authors reported that trabecular and
cortical bmc and CSA were significantly decreased at 980 and 1,400 mg P/kg-day; changes were
not strictly dose-related (see Table B-2). The study authors reported that these effects plus
reductions in cortical bmd and thickness (3 and 10% lower than referents, respectively, at
1,400 mg P/kg-day) were seen in the midshaft. Analyses of (right) tibial bones showed that
treatment significantly affected trabecular bone volume, bone surface, and SMI (all negatively)
on the basis of analysis of variance (ANOVA) only; no post hoc analysis was performed. The
study authors reported that the cortical cross-sectional area was significantly decreased.
Mechanical testing of femurs (femoral neck) showed treatment at 1,400 mg P/kg-day
significantly decreased strength and yield point (by 24 and 36%, respectively). The study authors
reported that treatment negatively affected all parameters of mechanical competence in tibial
shafts (on the basis of an ANOVA of strength, yield point, stiffness, and toughness). A LOAEL
of 980 mg P/kg-day is identified on the basis of decreased body weight and effects on bone
parameters (decreased femur bmd and alterations in bone structure) in rats treated for 8 weeks (a
64 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
NOAEL was not identified). The LOAEL of 980 mg P/kg-day corresponds to a LOAEL (HED)
of 240 mg P/kg-day.4
Koshihara et al (2005)
Female Wistar rats (5/group) were administered 0.5% (referent) or 1.5% inorganic
phosphate (as KH2PO4) for 6 weeks. On the basis of average reported food-consumption
(0.0136 kg/day) and body-weight (average of initial and final weights; 0.256 and 0.254 kg for
referents and treated rats, respectively) data provided in the study, concentrations of 0.5 and
1.5% in the diet are equivalent to approximately 270 and 800 mg P/kg-day as calculated by the
U.S. EPA. An additional group of rats was administered a low-phosphate diet (0.15% in the diet
or approximately 74 mg P/kg-day). The composition of the diets otherwise remained constant
(Ca was maintained at 0.5% of the diet). Food consumption and body weights were monitored
(time points not specified). The feces and urine of all rats were collected for 3 days prior to study
termination; the excretion of Ca, P, creatinine, or deoxypyridinoline (DPD; a marker of bone
resorption) were quantified. Serum was collected at study termination to determine Ca, P, PTH,
and osteocalcin (a marker of bone formation) concentrations. Fifth lumbar vertebra (L5) samples
were also collected; bmc and bmd were measured, and compression load of L5 was determined.
Results for the 800 mg P/kg-day group were compared with the 270 mg P/kg-day group
(as referent) (Koshihara et al.. 2005). Food consumption and body weights were reported to be
unaffected by treatment. The study authors reported that Ca absorption was significantly
decreased (61% lower than referents), whereas P absorption and excretion (in the urine) were
significantly increased (3.8- to 3.9-fold higher than referents; see Table B-3). No significant,
treatment-related effects on serum concentrations of Ca, P, or PTH were observed; however,
serum osteocalcin was 39% higher in rats treated at 800 mg P/kg-day compared with referents.
The study authors reported that urinary excretion of DPD also was significantly elevated at
800 mg P/kg-day (27% higher than controls). Data from the low-phosphate (0.15%) group
showed increased P absorption. Although bmc and the compression load of L5 were not
significantly affected by treatment, the bmd in vertebrae of rats treated for 6 weeks decreased.
The LOAEL of 800 mg P/kg-day is based on decreased bmd in vertebrae bmd that was
significantly lower (by 8%) in 800 mg P/kg-day rats relative to referents. A LOAEL is identified
on the basis of these data of 800 mg P/kg-day, which corresponds to a LOAEL (HED) of 200 mg
P/kg-day.5
4As outlined in the U.S. EPA's Recommended use of body weight3/4 as the default method in derivation of the oral
reference dose (U.S. EPA. 201151. the LOAEL was converted by the U.S. EPA to an HED of 240 mg P/kg-day
using a DAF of 0.24 (HED = ADD x DAF). The DAF was calculated as follows: DAF = (BWa1/4 ^ BWh1'4).
Quantitative body-weight data were not reported; therefore, reference body weights recommended by U.S. EPA
(1988) were used to calculate the DAF: 70 kg for humans and 0.217 kg for male Wistar rats in a subchronic study.
5As outlined in the U.S. EPA's Recommended use of body weight3'4 as the default method in derivation of the oral
reference dose (U.S. EPA. 2011b). the LOAEL was converted by the U.S. EPA to an HED of 200 mg P/kg-day
using a DAF of 0.25 (HED = ADD x DAF). The DAF was calculated as follows: DAF = (BWa1/4 - BWh1/4).
Quantitative body-weight data for rats (0.254 kg) and reference body weights for humans (70 kg) recommended by
U.S. EPA (1988) were used to calculate the DAF.
65 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Ritskes-Hoitinga et al. (1989)
Female Wistar RIV:TOX rats (6 or 16 females/group) were administered 0.4 (control) or
0.6% P (as NaH2PC>4 dihydrate) in the diet for 28 days. The concentration of Ca in the diet
remained constant (0.48-0.50%, administered as calcium carbonate). On the basis of measured
food consumption (0.0119 and 0.0124 kg/day for the control and treated groups, respectively)
and body weights (average of initial and final weights; 0.126 and 0.129 kg, respectively), 0.4 and
0.6% in the diet are equivalent to approximately 390 and 580 mg inorganic phosphate/kg-day
(respectively). Food consumption was measured continuously, while body weights were
measured at study initiation and at study termination. Data on water consumption, and urine or
feces were collected on Study Days 0-2, 13-15, and 26-28; urine and feces were analyzed for P,
Ca, and magnesium levels (Days 13-15 and 26-28). Serum was evaluated for creatinine, urea,
and osmolality (Study Day 28); urine was evaluated for the same parameters as for volume, pH,
and albumin (Study Days 0-2, 13-15, and/or 26-28). At sacrifice, kidney weights were
recorded. The right kidney of each animal was homogenized and analyzed for P, Ca, and
magnesium content. The left kidney, heart, liver, thoracic aorta, parathyroid, stomach, and lung
were examined grossly and microscopically (hematoxylin and eosin [H&E] staining). Additional
sections of the kidney, stomach, and lung were processed (using von Kossa staining) to detect
P-containing deposits. The severity of nephrocalcinosis was scored on a scale of 0-3 (on the
basis of the average scores of two blinded assessors).
No significant, treatment-related effects on food and water consumption or body weights
were observed (see Table B-4) (Ritskes-Hoitinga et al.. 1989). The study authors reported that
urinary P excretion was significantly increased relative to controls on Study Days 13-15 and
26-28 (2.1- to 2.4-fold) (data not shown); magnesium retention was transiently increased (42%
on Days 13-15 only). The study authors reported observing no significant changes in serum
parameters; however, urine pH was significantly decreased (8—17% on Days 0-2, 13-15, and
26-28) and urinary albumin was significantly increased (1.8- to 2.8-fold on Days 13-15 and
26-28) at 580 mg/kg-day. Levels of urea in the urine were increased on Days 13-15 only (the
toxicological significance was uncertain). At study termination, relative kidney weights were
reported to be 27% higher in treated rats than in controls. Kidney concentrations of Ca, P, and
magnesium also were reported to be significantly higher in treated rats; the magnitude of this
effect was most pronounced for Ca (increased 14-fold, compared with about twofold for P and
magnesium). Reported histopathological abnormalities were confined to the kidneys; deposits
were seen in the cortex, corticomedullary junction, and papilla. Because deposits in the cortex
and papilla were birefringent and showed no specific localization, the study authors considered
them artifacts of the fixation process (likely caused by the dissolution of corticomedullary
deposits). Therefore, only calculi in the corticomedullary area were factored into
nephrocalcinosis scores. The reported incidence of nephrocalcinosis was 16/16 in rats treated at
580 mg/kg-day (mean severity = 2.7) compared with 2/6 in controls (mean severity = 0.5). The
difference in the distribution of histological scores among groups was statistically significant
(p < 0.05). Extreme cases of nephrocalcinosis reportedly were accompanied by interstitial
fibrosis and focal tubuli with regenerated epithelia (incidence data not provided).
In a follow-up experiment designed to determine the time course of these effects
(i.e., increased urine albumin and nephrocalcinosis), two additional groups of rats
(12 females/group) were administered 0.4% (control) or 0.6% P (as NaFhP04) in the diet for
28 days. The concentration of Ca in the diet remained constant (0.48-0.50%), administered as
calcium carbonate). Based on measured food consumption (0.0125 and 0.0119 kg/day for the
66 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
control and treated groups, respectively) and body weights (average of initial and final weights;
0.126 and 0.124 kg, respectively), 0.41 and 0.61% in the diet are equivalent to approximately
410 and 590 mg P/kg-day (respectively). Food consumption was measured continuously, while
body weights were measured at study initiation and at study termination. Urinary parameters
(volume, pH, and albumin only) were evaluated from samples collected on Study Days 0-2,
13-15, and 26-28. Serum determinations (of P, Ca, and magnesium) were conducted on six
rats/group prior to sacrifice on Study Days 14 and 28. Relative kidney weights were recorded.
Kidney P, Ca, and magnesium concentrations were measured. Histopathological examinations
were performed (kidneys, parathyroid, stomach, and lungs).
No significant, treatment-related effects on food and water consumption or body weights
were observed (see Table B-5) (Ritskes-Hoitinga et al.. 1989). The study authors reported that
urinary pH was significantly decreased at all time points (7-10%); urine albumin was increased
1.7- to 3.4-fold on Study Days 13-15 and 26-28 (after an initial decrease on Days 0-2); effects
were reported to be statistically significant on Study Days 0-2 and 13-15. Serum P was
reportedly not significantly affected by treatment, but serum Ca and magnesium levels were
significantly reduced in treated rats relative to controls (by 6 and 18%, respectively). Although
not statistically significantly increased, the study authors reported that relative kidney weights in
rats treated at 580 mg/kg-day were 24% higher than controls by Day 28, which is considered by
this review to represent a biologically significant change. In general, kidney concentrations of
Ca, P, and magnesium were significantly increased except P on Day 28, with the effect on Ca
being most pronounced (at least fourfold higher than controls). Most rats (treated group and
controls) showed evidence of nephrocalcinosis (Days 14 and 28); however, the distribution of
severity scores among groups was statistically significantly different (i.e., more severe at
580 mg/kg-day). The study authors observed mineral deposits at the corticomedullary junction
only, and, unlike Experiment 1, no deposits occurred in the cortex or papilla (the study authors
attributed the cortical and papillary deposits seen in Experiment 1 to be artifacts of fixation with
Bouin's solution, and Experiment 2 used formalin for fixation). As in Experiment 1, the study
authors reported that severe nephrocalcinosis cases showed interstitial fibrosis and tubular
epithelial regeneration. No other histopathological effects were reported. A LOAEL of
580 mg/kg-day is identified on the basis of the data from these two experiments for kidney
effects (increases in urine albumin and relative kidney weights; evidence of nephrocalcinosis).
The LOAEL of 580 mg/kg-day corresponds to a LOAEL (HED) of 120 mg/kg-day.6
Ritskes-Hoitinga et al. (2004)
In a study designed to more rigorously determine the optimal dietary P concentration for
rabbits, New Zealand White (NZW) rabbits (eight males/group) were administered P (as
NaH2P04 dihydrate) at four nominal dietary concentrations (0.1, 0.2, 0.4, and 0.8% P in the diet)
for 8 weeks. As noted by the study authors, few data on the P requirements of rabbits are
available. NRC (1977) recommended 0.22% P in diet, for growth, with higher requirements
(0.37-0.5%)) for pregnant or lactating rabbits; these values were based on studies using
predominantly natural diets in which P bioavailability may have been limited. Clarke et al.
(1977) recommended a P concentration of 4 g P/kg diet (0.4%) and Ca concentration of 5 g
6As outlined in the U.S. EPA's Recommended use of body weight3'4 as the default method in derivation of the oral
reference dose (U.S. EPA. 201151. the LOAEL was converted by the U.S. EPA to an HED of 120 mg/kg-day using
a DAF of 0.21 (HED = LOAEL x DAF). The DAF was calculated as follows: DAF = (BWa1/4 - BWh1/4).
Quantitative body-weight data for rats (0.129 or 0.124 kg) and reference body weights for humans (70 kg)
recommended by U.S. EPA (1988) were used to calculate the DAF.
67 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Ca/kg diet (0.5%) for laboratory rabbits but cited no published sources as the basis for these
recommendations.
Diets used by Ritskes-Hoitinga et al. (2004) contained relatively constant Ca
concentrations of 0.45-0.46%. Clarke et al. (1977) recommended a minimum concentration of
5 g Ca/kg diet (0.5%) for laboratory rabbits, but with no supporting citations. NRC (1977)
suggested that dietary Ca needs in rabbits depend on dietary P, citing data suggesting that at a
dietary P level of 0.37%, optimal growth was achieved at 0.22% Ca, but optimal bone
calcification required 0.34-0.40% Ca (suggesting a Ca:P ratio requirement of -1). NRC (1977)
further noted that in rabbits, serum Ca levels generally correspond to dietary Ca intake; in
contrast, serum Ca is regulated within a narrow range in other mammalian species. Ritskes-
Hoitinga et al. (2004) concluded on the basis of their experiments 0.2% should be a maximum
dietary P level for rabbits, based on increasing incidences of renal calcifications.
In the absence of a definitive minimum dietary P requirement for rabbits, and evidence
that Ca and P requirements appear to be interdependent in this species, the referent group
selected by the study authors for this study was the concentration that provided an approximate
Ca:P ratio of 1:1 (i.e., 0.4% P). This group also had Ca and P intakes corresponding to the
recommendations of Clarke et al. (1977).
Measured concentrations of P in the referent and exposed group diets were 0.45 and
0.88%), respectively. On the basis of measured food consumption (0.0731 and 0.0689 kg/day for
the referent and treated groups, respectively) and body weights (average of initial and final
weights; 2.22 and 2.09 kg, respectively), 0.45 and 0.88% P in the diet are equivalent to
approximately 150 and 290 mg P/kg-day as calculated by the U.S. EPA. Food consumption
(based on the weight of leftover food) and body weights (initial and final) were recorded.
Phosphorus, Ca, and magnesium concentrations in the urine and feces were measured from
samples collected on Study Days 20-23 and 48-51; levels in the serum were quantified on Study
Days 28 and 56. At sacrifice on Day 57, right kidney weights were recorded; kidneys were
analyzed for P, Ca, and magnesium content. Left kidneys were subjected to histopathological
examinations (H&E and von Kossa staining); the severity of nephrocalcinosis in the cortex and
the medulla was graded on a scale of 0 (absence of nephrocalcinosis) to 3 (severe
nephrocalcinosis). Femur parameters (volume, length, circumference) were evaluated; two parts
of the femur (the medial diaphysis and the epiphysis) were also processed for mineral content.
No significant, treatment-related effects on food consumption or body weights were
observed among rabbits administered 150 and 290 mg P/kg-day (Ritskes-Hoitinga et al.. 2004).
The study authors observed numerous effects on mineral balance win rabbits treated at 290 mg
P/kg-day (see Table B-6); changes consistently observed on Study Days 20-23 and 48-51 were
decreased urinary pH (8.04 vs 9.35 at the latter time point) and decreased Ca (25- to 34-fold) and
increased P (2.2- to 3.3-fold) in the urine. The study authors reported that serum P, but not serum
Ca or magnesium, was significantly increased on Study Days 28 and 56 in rats treated at 290 mg
P/kg-day (17.9-19.7%) higher than referents). The study authors reported that kidney weights
(absolute and relative) were not statistically significantly affected by treatment; however,
concentrations of Ca and P in the kidney were increased (Ca statistically) at 290 mg P/kg-day
(412 and 29.2% higher than referents, respectively); these effects were considered by the study
authors to be indicative of nephrocalcinosis. In agreement, histopathological evaluations showed
significantly increased incidence and severity scores for cortical calcifications using both
68 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
staining methods (differences in calcinosis were not observed in the medulla). At 150 and
290 mg/P/kg-day, the incidence of any calcification (i.e., severity scores 1-3) in the cortex was
2/6 vs. 8/8 with H&E staining and 3/8 vs. 8/8 in the von-Kossa-stained group. The study authors
noted that femur parameters were not significantly affected, other than significantly increased
magnesium content in both the medial diaphysis (14-16% higher than referents) and epiphysis
(26% higher than referents). A LOAEL of 290 mg P/kg-day is identified on the basis of
histological evidence of nephrocalcinosis in rabbits. The LOAEL of 290 mg P/kg-day
corresponds to a LOAEL (HED) of 120 mg P/kg-day.7
Tani et al. (2007)
Wistar rats (six males/group) were administered P (as KH2PO4) at 0.3 (control), 0.6, 0.9,
1.2, or 1.5% in the diet for 4 weeks. On the basis of measured food consumption
(0.015-0.018 kg/day, calculated by dividing the reported 4-week intakes by 28 days) and
reference body weights for male Wistar rats (0.217 kg) (U.S. EPA. 1988). concentrations of 0.3,
0.6, 0.9, 1.2, and 1.5% in the diet are equivalent to approximately 250, 450, 670, 920, and
1,000 mg P/kg-day. The composition of the diets otherwise remained constant (Ca was
maintained at 0.6% of the diet). Food consumption and body weights were measured twice
weekly. Urine and feces were collected on the last 3 days of study to evaluate Ca and P content.
At study termination, blood was collected to measure serum levels of Ca, P, PTH, and vitamin D.
Kidney samples were attained to determine messenger ribonucleic acid (mRNA) (total kidney)
and protein (in brush border membrane vesicles of the renal cortex) levels of the
sodium-dependent phosphate transporter (Npt Iia).
Food consumption was significantly decreased in all groups of treated rats relative to the
referent group (8-17% lower than referents and not strictly dose-related; Table B-7) (Tani et al..
2007). Body weights were not reported, but, according to the study authors, body-weight gain
(over the course of 4 weeks) was significantly decreased at 670 and 1,000 mg P/kg-day (but not
920 mg/kg-day); body-weight gain was only 62% of referent values at 1,000 mg P/kg-day.
Body-weight gain expressed per 100 g of food intake was significantly decreased at 1,000 mg
P/kg-day only (26% lower than referents). Tani et al. (2007) reported that urinary (at >450 mg
P/kg-day) and fecal (at >670 mg P/kg-day) excretion of P was significantly increased in a
dose-related manner; net absorption also increased with dose (5.4-fold higher than in referents at
the 1,000 mg P/kg-day dose). Urinary (but not fecal) excretion of Ca was significantly affected
(decreased at >450 mg/kg-day), according to the study authors; net absorption was decreased
only at 1,000 mg P/kg-day only (7.4-fold lower than referents). Tani et al. (2007) reported that
levels of Ca and P in the serum were not significantly altered by treatment. Although PTH levels
tended to be increased in all treatment groups, the study authors reported that serum PTH was
significantly increased only at 1,000 mg P/kg-day (9.6 times higher than in referents). Vitamin D
levels were significantly increased at 670 mg P/kg-day only (and not at higher doses) (Tani et al..
2007). The study authors also reported that expression of Npt Iia was markedly decreased at
1,000 mg P/kg-day; mRNA and protein levels were 57 and 21% of referent levels, respectively
(data not shown). NOAEL and LOAEL values of 920 and 1,000 mg P/kg-day, are identified
respectively, based on decreased body-weight gain and body-weight gain normalized to intake.
7As outlined in the U.S. EPA's Recommended use of body weight3'4 as the default method in derivation of the oral
reference dose (U.S. EPA. 201151. the LOAEL was converted by the U.S. EPA to an HED of 120 mg P/kg-day
using a DAF of 0.42 (HED = LOAEL x DAF). The DAF was calculated as follows: DAF = (BWa1/4 - BWh1/4).
Quantitative body-weight data for rabbits (2.09 kg) and reference body weights for humans (70 kg) recommended
by U.S. EPA (1988) were used to calculate the D AF.
69 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
The NOAEL and LOAEL of 920 and 1,000 mg P/kg-day correspond to NOAEL (HED) and
LOAEL (HED) values of 220 and 240 mg P/kg-day, respectively.8
Subchronic toxicity studies
A huduli et ill. (2016)
In a study designed to assess effects on glucose and lipid metabolism, male
Sprague Dawley rats (six/group) were administered 0.6% (referent) or 1.2% inorganic phosphate
(as 2.53 or 5.11% KH2PO4 in the diet, respectively) for 4 or 14 weeks. Based on reference (U.S.
EPA. 1988) food consumption (0.0234 kg/day) and body weight (0.267 kg) for male
Sprague Dawley rats in a subchronic study (food consumption and body-weight data from the
study were not presented numerically), concentrations of 0.6 and 1.2% in the diet are equivalent
to approximately 530 and 1,100 mg P/kg-day. An additional group of rats was administered a
low-phosphate diet (0.2% in the diet or approximately 180 mg P/kg-day). The composition of the
diets otherwise remained constant (Ca was maintained at 0.6% of the diet).
Rats treated for 14 weeks were evaluated for food consumption (measured every other
day), body weights (measured daily), and liver triglycerides only. In rats administered inorganic
phosphate for 4 weeks, food consumption and body weights were recorded regularly. Fat
deposits were measured and categorized as epididymal (eWAT), mesenteric (mWAT), or
retroperitoneal (rpWAT) white adipose tissue (WAT). Oxygen consumption (V02) and carbon
dioxide production (Vccc) were evaluated continuously at 10-minute intervals over a 3-day
period; the respiratory quotient (RQ; the ratio of O2 consumption to CO2 production) was
calculated. Locomotor activity was measured over a 24-hour period (using an infrared-based
device). Glucose tolerance tests (to measure serum glucose and insulin levels before and after
administration of a bolus dose of glucose) were performed after 3 weeks of treatment; the
homeostasis model assessment of insulin resistance (HOMA-IR) was determined on the basis of
fasting glucose and insulin concentrations.
Prior to terminal sacrifice, blood was collected to evaluate biochemical or clinical
chemistry parameters, including P and Ca levels; creatinine and blood urea nitrogen (BUN), total
cholesterol and triglycerides, leptin and adiponectin levels, nonesterified fatty acids (NEFA),
1,25-dihydroxy-vitamin D3, and the concentrations of PTH, FGF-23, triiodothyronine (T3), or
thyroxine (T4); hepatic lipids were also isolated to measure total cholesterol and triglyceride
levels. Selected tissues were weighed (liver, kidney, heart, muscle, and brown adipose tissue
(BAT); interscapular fat, white adipose tissue (WAT) and BAT were examined histologically
(H&E staining). RNA was extracted from the liver, BAT, and WAT to assess gene expression
related to lipogenesis and lipolysis. Three additional groups of male rats were administered the
experimental diets for 4 weeks to evaluate metabolic activity in the BAT (measured as
[18F]-fludeoxyglucose [FDG] uptake).
8As outlined in the U.S. EPA's Recommended use of body weight3'4 as the default method in derivation of the oral
reference dose (U.S. EPA. 201151. the NOAEL and LOAEL were converted by the U.S. EPA to HEDs of 220 and
240 mg P/kg-day using a DAF of 0.24 (HED = LOAEL x DAF). The DAF was calculated as follows:
DAF = (BWa1/4 ^ BWh1/4). Reference body weights for male Wistar rats (0.217 kg) and humans (70 kg)
recommended by U.S. EPA (1988) were used to calculate the DAF.
70 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Results for the 1,100 mg P/kg-day group are compared with the 530-mg P/kg-day group
only (as referent) (Abuduli et al.. 2016). In rats evaluated after treatment for 4 weeks, the study
authors observed no significant treatment-related effects on body weight (see Table B-8).
However, the study authors reported that rats treated at 1,100 mg P/kg-day showed decreased
visceral fat accumulation (mWAT and rpWAT content were about 25% lower than in referents,
based on data presented graphically). No significant differences in V02 or Vco2 were observed,
however, the RQ was lower in the 1,100-mg P/kg-day group than in referents (slightly but
significantly affected during the dark cycle only; <10% lower based on visual inspection of data
presented graphically) (Abuduli et al.. 2016). The study authors reported that locomotive activity
was similar among groups. No significant differences between the referent group and the
1,100-mg P/kg-day group were observed with respect to blood glucose levels, fasting blood
glucose, or insulin levels, or HOMA-IR. The administration of a diet containing 1,100 mg
P/kg-day did not significantly alter the levels of P, Ca, creatinine, cholesterol or triglycerides,
1,25-dihydroxy-vitamin D3, adiponectin, or thyroid hormones in the blood (Abuduli et al..
2016). However, rats treated at 1,100 mg P/kg-day showed significantly increased blood levels
of BUN (32%) higher than referents) and PTH (~6 times higher than referents), and significantly
decreased leptin (—15% lower than referents based on data presented graphically) and NEFA
(28% lower than referents) (Abuduli et al.. 2016); see Table B-8. FGF-23 levels tended to be
increased at 1,100 mg P/kg-day (77% higher than referents), but this effect was not statistically
significant according to the study authors. Liver cholesterol and triglycerides were also reported
to be unaffected by treatment.
Although the study authors observed no significant differences in food consumption
throughout the 14-week study period, rats administered 1,100 mg P/kg-day in the diet showed
decreased body weights from Week 8 onward (up to approximately 14% lower than referents,
based on data presented graphically). Likewise, lower liver triglyceride levels were observed in
rats treated at 1,100 mg P/kg-day (1.36 ± 0.35 mg/g in the high-phosphate group compared with
2.57 ± 0.51 mg/g in referents; the study authors' statistical analyses were based on comparison to
the 180 mg P/kg-day group only; number of animals not reported for independent analysis).
Decreased triglycerides in 1,100 mg P/kg-day rats is not considered a toxicologically relevant
effect. A LOAEL of 1,100 mg P/kg-day is identified based on decreased body weights in rats
administered inorganic phosphate in the diet for 14 weeks. A NOAEL cannot be identified as the
next lowest dose is the referent dose. The LOAEL of 1,100 mg P/kg-day corresponds to a
LOAEL (HED) of 270 mg P/kg-day.9
No statistically significant effects on organ weights occurred based upon inspection of the
data presented graphically; however, on the basis of the data contained in the study report,
kidney weights were approximately 12% higher (based on data digitized using Grab It™) in rats
treated at 1,100 mg P/kg-day relative to the referent group, which is considered biologically
relevant. Histological analysis showed decreased lipid content in BAT of rats administered
1,100 mg P/kg-day. In the liver and WAT, no statistically significant changes in gene expression
were observed; however, the expression of genes related to lipid oxidation (namely uncoupling
9As outlined in the U.S. EPA's Recommended use of body weight3'4 as the default method in derivation of the oral
reference dose (U.S. EPA. 201151. the LOAEL was converted by the U.S. EPA to an HED of 270 mg P/kg-day
using a DAF of 0.25 (HED = LOAEL x DAF). The DAF was calculated as follows: DAF = (BWa1/4 - BWh1/4).
Quantitative body-weight data were not reported; therefore, reference body weights recommended by U.S. EPA
(1988) were used to calculate the DAF: 70 kg for humans and 0.267 kg for male Sprague Daw ley rats in a
subchronic study.
71 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
protein 1 [UCP1] and peroxisome proliferator-activated receptor-y coactivator-la [PGC-la]) was
significantly increased in BAT of rats treated at 1,100 mg P/kg-day relative to referents. No
significant differences in [18F]-FDG uptake were noted among treatment groups. Effects on lipid
metabolism (e.g., reduced fat accumulation) were not considered toxicologically relevant by the
U.S. EPA. However, a LOAEL of 1,100 mg P/kg-day is identified for the 4-week study on the
basis of evidence of kidney effects (increases in BUN and kidney weight). The administered
dose of 1,100 mg P/kg-day corresponds to an HED of 270 mg P/kg-day.
Datta el nl. (1962)
Birmingham-Wistar rats (20/sex/group) were administered inorganic phosphate as
Na4P2C>7 (97.5% with 2.5% orthophosphate) at 0 (referent), 1, 2.5, or 5% in the diet for
16 weeks. The test material was also reported by the trade name, Tetron K; however, the specific
forms of sodium phosphates corresponding to the reported material could not be determined.
Analysis of the diet showed P constituted 0.526% of the referent diet and 1.65% of the high-dose
diet, so these values were used to calculate doses. The P contents of the 1 and 2.5% diets were
not provided; therefore, doses (in mg P/kg-day) cannot be reliably estimated for these groups.
The concentration of Ca in the diets was relatively constant (0.45-0.64%) of the diet). Based on
data provided in the study report for food consumption (0.0117 and 0.0106 kg/day for referent
males and females, respectively, and 0.0108 and 0.0094 kg/day for 5% males and females,
respectively; calculated by dividing total intakes reported by study authors by 112 days) and
mean body weights (0.220 and 0.159 kg for referent males and females, respectively, and
0.209 and 0.147 kg for 5% males and females, respectively; calculated as average of initial and
final weights, assuming initial body weight of 115 g for males and 90 g for females on the basis
of range reported by study authors), U.S. EPA estimates that 0.526% P in the diet is equivalent to
approximately 280 mg P/kg-day (males) and 350 mg P/kg-day (females), and 1.65% P in the diet
is equivalent to approximately 860 mg P/kg-day (males) and 1,100 mg P/kg-day (females). An
additional group of rats (20/sex) was administered 5% Na2HPC>4; P composed 1.598% of that
diet. Based on food consumption (0.0122 and 0.0104 kg/day for males and females, respectively)
and mean body weights (0.232 and 0.160 kg for males and females, respectively), 1.598% P in
the diet is equivalent to approximately 840 mg P/kg-day (males) and 1,000 mg P/kg-day
(females).
In each dose group, 10 rats/sex were housed individually; remaining rats were grouped
5 per cage. Food consumption and body weights were recorded weekly (for rats housed
individually only) for the first 100 days (about 14 weeks) of the study. Urine and feces were
collected after 3 weeks (five referent males and five males treated with 5% Na4P20?) and
8 weeks (five males each from the 5% Na4P2C>7 and orthophosphate groups); levels of Ca,
phosphate, and pyrophosphate were measured. Urine was also analyzed for pH, and for sodium,
potassium, and ammonium concentrations. At study termination, hematology parameters (red
blood cell [RBC] and differential white blood cell [WBC] counts; hemoglobin [Hb]) were
examined in five rats/sex/group. Liver function (based on bromosulphalein [BSP] clearance) and
kidney function (phenol red excretion test, concentration test, albumin, and cellular contents)
were evaluated in five rats/sex/group after 110 days. At sacrifice, organ weights (of the heart,
liver, spleen, stomach, intestines, adrenals, kidneys, and testes) were recorded; rats were
subjected to gross and microscopic examinations (10 rats/sex/group that were housed
individually; tissues not specified).
72 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Results are summarized in Table B-9. In groups exposed to Na4P2C>7, the study authors
reported reduced food consumption and body-weight gain in rats receiving 5%; but only
body-weight gain in females was statistically significantly lower compared with the referent
group (Datta et al.. 1962). Additional statistical analyses (covariance using the method of
Crampton) showed food consumed per unit of body-weight gain was similar between the 5%
Na4P2C>7 group (both sexes) and referents (data not shown). Ca absorption and excretion were
similar among (male) rats treated with 5% Na4P2C>7 and referents, but the percentage of
phosphate excreted in the urine was markedly increased (Datta et al.. 1962). The study authors
reported that other than urinary pH (7.5 in the both the pyrophosphate and Na2HPC>4 5%-groups,
compared with 6.7 in referents), urinary endpoints were unaffected by treatment. Pyrophosphate
was not detected in the urine or feces of treated animals, suggesting pyrophosphate was
hydrolyzed to orthophosphate prior to or after absorption.
The study authors observed no hematological effects considered toxicologically relevant
in groups exposed to the pyrophosphate, and BSP clearance was unaffected by treatment.
Although no significant effects were observed on other tests of kidney function, the
concentration test showed significantly decreased urine specific gravity in the 2.5%- and
5%-groups in males and the 5%-group in females (Datta et al.. 1962). Relative kidney weight
was significantly increased in female rats treated with 2.5 and 5% Na4P2C>7 (24 and 49% higher
than referents, respectively) and in males exposed to 5% (21% higher) (Datta et al.. 1962). Rats
treated at 5% Na4P2C>7 (860 mg P/kg-day for males and 1,100 mg P/kg-day for females) also
showed significantly increased relative heart (12—21% higher than referents), stomach (—60%
higher), and testes (28% higher) weights; intestinal weight was increased in females only (27%
higher) (Datta et al.. 1962). The study authors noted an apparent increased incidence (although
not statistically significant) of gross pathological changes in the stomach of female rats treated
with 2.5 and 5% Na4P2C>7 (hypertrophy and hemorrhage); gross kidney effects (pale kidneys and
calcification) were also noted in these groups. In males, increased incidence of histological
effects were only seen in the 5%-groups (Datta et al.. 1962). Histopathological changes were
confined to the kidney, and (on the basis of data for the combined sexes) included significantly
increased incidences of hemorrhages and exudate (all treatment groups); medullary calcification
(2.5- and 5%-groups); and medullary necrosis, tubular casts, and chronic inflammation
(5%-groups only) (Datta et al.. 1962). Cortical effects noted in the study report (atrophy, hyaline
degeneration, and calcification) were not dose-related. The study report noted that no
histopathological changes were observed in the stomach. Because effects in rats administered
phosphate as pyrophosphate occurred at doses that could not be quantified, no effects levels are
identified for the pyrophosphate experiment.
In the group exposed to Na2HP04, no treatment-related change in Ca absorption or
excretion was seen, but the urinary excretion of P was markedly increased (Datta et al.. 1962).
As noted above, the urinary pH was higher in the Na2HP04 exposed group than in referents
(7.5 compared with 6.7 in referents), while other urinary parameters did not differ. Exposure to
Na2HP04 resulted in slightly, but statistically significantly increased RBC count in females but
not males (Datta et al.. 1962). The study authors reported that hemoglobin was slightly (6%) but
statistically significantly increased in both sexes. BSP clearance was unaffected by treatment
(Datta et al.. 1962). As with the pyrophosphate, exposure to Na:HP04 resulted in significantly
decreased specific gravity in the concentration test in both sexes, while other kidney function
tests showed no significant effects( Datta et al.. 1962). Significantly increased relative kidney
weight was observed by the study authors in Na2HP04-treated rats of both sexes (17—39% higher
73 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
than referents). Increased incidence of gross changes were not statistically significant and were
observed only in the kidney and consisted of pale kidneys and macroscopically observed
calcification. Increased incidences (both sexes combined) of renal damage, hemorrhages and
exudate, chronic inflammation, medullary calcification, medullary necrosis, and tubular casts
were reported (Datta et al.. 1962). A LOAEL of 840 mg P/kg-day is identified for males on the
basis of evidence of kidney effects (changes in specific gravity, increased relative kidney weight,
gross or microscopic evidence of kidney damage); a NOAEL could not be established as there
were effects at all calculable doses above referent. The LOAEL of 840 mg P/kg-day corresponds
to a LOAEL (HED) of 200 mg P/kg-day.10
Dymsza et al (1959)
Male Wistar rats (up to 12/group) were administered a referent diet (0.87% K2HPO4, with
P comprising 0.43% of the total diet) for up to 150 days. Additional groups of rats received diets
containing "normal" orthophosphate (0.87% K2HPO4 and 0.43% P) or "high" orthophosphate
(5.1%) K2HPO4 and 1.3% P), "normal" metaphosphate (0.93% [NaP03]6 and 0.46% P), or "high"
metaphosphate (3.5% [NaPOsje and 1.2% P). Other components of the diet were held relatively
constant (Ca comprised 0.47-0.56%) of the diet). Differences occurred between the referent and
normal ortho- and metaphosphate diets unrelated to phosphate administration; the normal
ortho- and metaphosphate diets contained 5—6% more mineral salts (not further characterized)
and higher sodium and chlorine concentrations. Based on measured food consumption (ranging
from 0.01825 to 0.0200 kg/day; calculated by dividing 60-day intakes reported by study authors
by 60 days) and body weights (ranging from 0.190 to 0.202 kg; calculated as average of initial
and final weights, assuming initial body weight of 55 g for males on the basis of range reported
by study authors) for the first 60 days of the study, these concentrations in the diet are equivalent
to approximately 380 (referent), 420 (normal K2HPO4), 1,100 (high K2HPO4), 470 (normal
[NaP03]6), or 1,100 (high [NaP03]6) mg P/kg-day. Food consumption and body-weight gain
were recorded (time points not specified); food and protein efficiency were calculated. After
50 days of the study, 7-day Ca and P balance studies were conducted using nine rats/group (by
comparing the amounts consumed to amounts excreted in the feces and the urine). After 60 days
on the test, five rats/group were sacrificed. Hematological (Hb concentration) and clinical
chemistry (serum levels of Ca and P) parameters were evaluated. The remaining animals (6 or
7/group) were kept on the study for 150 days. Based on measured food consumption
(0.020-0.0235 kg/day) and body weights (0.243-0.271 kg) for the 150-day study, doses were
approximately 330 (referent), 380 (normal K2HPO4), 1,100 (high K2HPO4), 400 (normal
[NaP03]6), or 1,100 (high [NaP03]6) mg P/kg-day. At sacrifice, the same hematological and
clinical chemistry parameters were evaluated (and including RBC count); organ weights (of the
liver, heart, kidneys, spleen, and testes) were recorded. All animals were subjected to
histopathological examinations (of the heart and kidneys). The right and left femur were
measured (length); femurs and carcasses were analyzed for Ca and P content.
10As outlined in the U.S. EPA's Recommended use of body weight3'4 as the default method in derivation of the oral
reference dose (U.S. EPA. 201151. the LOAEL was converted by the U.S. EPA to an HED of 200 mg/kg-day using
a DAF of 0.24 (HED = LOAEL x DAF). The DAF was calculated as follows: DAF = (BWa1/4 - BWh1/4).
Quantitative body-weight data for rats (0.232 kg) and reference body weights for humans (70 kg) recommended by
U.S. EPA (1988) were used to calculate the DAF.
74 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Differences between the referent group and the normal ortho- and metaphosphate groups
were noted by the study authors (nonsignificantly increased kidney weight in the "normal"
phosphate groups), despite similar P doses (320, 380, and 400 mg P/kg-day, respectively)
(Dvmsza et al.. 1959). Because the referent and normal K2HPO4 groups both received K2HPO4 at
the same concentration (0.87%), these differences may be attributable to other elements of the
diets (mineral salts and corresponding mineral levels noted above) that varied. Thus, for the
purposes of this assessment, effects in the high K2HPO4 and high (NaP03)6 groups were
compared with the normal K2HPO4 and (NaP03)6 groups, respectively.
In the group sacrificed after 60 days, the study authors reported that no significant
differences occurred between the normal and high phosphate groups with respect to body-weight
gain, food or protein consumption or efficiency, or Hb levels. Although the high K2HPO4 group
was exposed to a higher dose of P, serum P levels were lower (although not statistically
significant) in this group (7.2 mg/100 mL) than in the normal K2HPO4 group (8.3 mg/100 mL).
No explanation for this phenomenon was provided by the study authors.
After 150 days of treatment, total body-weight gain was significantly higher (12%) in the
high K2HPO4 group compared with the normal K2HPO4 group (see Table B-10); differences
between the normal and high (NaPOj)h groups were not significant (Dvmsza et al.. 1959).
Animals administered high K2HPO4 or (NaP03)6 did not exhibit differences in food or protein
consumption, femur length, ash or Ca percent (dry) of femur bone, Hb levels, or RBC counts,
when compared by the study authors with the normal groups. As in the group exposed for only
60 days, serum P was significantly higher in the normal K2HPO4 group compared with the high
K2HPO4 group (7.8 mg/100 mL in normal and 6.0 mg/100 mL in high); the study authors offered
no explanation. There were no significant differences between normal and high phosphate
groups with respect to relative weights of liver or spleen (Dvmsza et al.. 1959). However, study
authors reported kidneys in high dose animals (orthophosphate or metaphosphate) were
significantly heavier (p = 0.05) than control rats. Rats administered the high (NaP03)6 dose
(1,100 mg P/kg-day as metaphosphate) showed significantly increased (11% higher than referent
group) relative testes weights compared with the normal (NaPOj)h group (Dvmsza et al.. 1959);
the toxicological significance of this effect is uncertain. ANOAEL of 1,100 mg P/kg-day is
identified for this study. The NOAEL of 1,100 mg P/kg-day corresponds to a NOAEL (HED) of
270 mg P/kg-day.11
Reproductive and Developmental Studies
The majority of reproductive and developmental toxicity studies of sodium or potassium
salts of inorganic phosphates are technical reports (including those commissioned by FDA) and
unpublished studies (many with limited information published in secondary sources). One
published study of reproductive or developmental effects in laboratory animals was identified in
the literature searches: Hodge (1964). Both the published and unpublished studies/technical
reports lack information on concurrent Ca intake, and in most cases, dietary P levels. Table 11
summarizes the available information.
nThe NOAEL was converted to an HED of 270 mg/kg-day using a DAF of 0.25 (HED = NOAEL x DAF). The
DAF was calculated as follows: DAF = (BWa1/4 ^ BWh1/4). Quantitative body-weight data for rats (0.271 kg) and the
reference body weight for humans (70 kg) recommended by U.S. EPA (1988) were used to calculate the D AF.
75 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 11. Summary of Unpublished Reproductive and Developmental Toxicity Studies of Na/K Salts of Inorganic
Phosphates
Ingredient
Strain/Species/
Sex/Number
Mode and
Duration
Doses
(mg P/kg-d)
Results
Reference
Combined repeat dose and reproductive/developmental screening studies
PA
Sprague Dawley
rats,
13/sex/group
Gavage
42-54 d (2 wk
before, during,
and after
mating in
males; females
continued
through
gestation to
PND 4)
NR (referent),
39.5,79, or 158
At 158 mg P/kg-d, two females died; GI distension was observed. Soft, mucosal
stool and a dirty nose were observed in one male at 158 mg P/kg-d. No clinical signs
were observed, no differences in body-weight gain or food efficiency, and no
differences in hematology, urinalysis, or neurobehaviors. Female rats exhibited
decreased absolute kidney weight at all doses and decreased relative uterine weight
at 158 me P/ke-d. OECD (2009) reported a 2-wk. duration for this studv but
characterized it as an OECD Test Guideline 422 studv. ECHA (2008) and CIR
Enoch Panel (2016) both reported treatment durations consistent with this Test
Guideline (as shown under "Duration") and are assumed to report the correct
duration. No reproductive or developmental effects were observed, including
mating, conception, parturition, fetal survival, and body weight.
CIR Expert
Panel (2016);
OECD (2009);
ECHA (2008)
K2HPO4
Sprague Dawley
rats,
16/sex/group
Gavage
42-54 d (2 wk
before, during,
and after
mating in
males; females
continued
through
gestation to
PND 4)
NR (referent)
or 180
No effects were observed on mating, male reproductive organ weights, or
histopathology in male or female parents (tissues examined were not specified). No
effects were observed on the numbers of corpora lutea, implantations, loss rate, birth
rate, survival rate, or sex ratio. Exposure was reported as 2 wk. premating, 2 wk.
mating, and 2 wk. post mating for males or 4 d parturition for females: 42 d (males)
and 42-54 d (females). ECHA (2005) reported number of animals as 22. whereas
CIR reported 16 animals/group.
OECD (2006):
ECHA (2005)
Developmental toxicity studies
NasPaOiu.
anhydrous
Dutch belted
rabbits,
17-21 F/group
Gavage
GDs 6-18
NR (referent),
0.6,2.9, 13.6,
or 63.1
No maternal toxicity and no effects were observed on pregnancy, numbers of
corpora lutea, implantations or resorptions, or fetal survival or fetal weight. No
treatment-related abnormalities in soft or skeletal tissues were noted.
FDRL (1973a)
NasPaOiu.
anhydrous
CD-I mice,
21-24 F/group
Gavage
GDs 6-15
NR (referent),
0.6,2.8, 13.1,
or 60.1
No effects were observed on maternal survival, numbers of implantations, fetal
survival, or numbers of soft tissue or skeletal abnormalities were reported. Dose
assumed to be reported as NasP.iOm.
CIR Expert
Panel (2016);
FASEB/LSRO
(1975)
76
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 11. Summary of Unpublished Reproductive and Developmental Toxicity Studies of Na/K Salts of Inorganic
Phosphates
Ingredient
Strain/Species/
Sex/Number
Mode and
Duration
Doses
(mg P/kg-d)
Results
Reference
NasPaOiu.
anhydrous
Wistar rats,
19-23 F/group
Gavage
GDs 6-15
NR (referent),
0.4,2.0, 9.3, or
42.9
No effects on maternal survival, numbers of implantations, fetal survival, or
numbers of soft tissue or skeletal abnormalities were reported. Dose assumed to be
reported as NasP^Oiu.
CIR Expert
Panel (2016);
FASEB/LSRO
(1975)
NasPaOiu.
anhydrous
Golden
hamsters,
19-23 F/group
Gavage
GDs 6-10
NR (referent),
0.4, 1.6, 7.6, or
35.6
No effects on maternal survival, numbers of implantations, fetal survival, or
numbers of soft tissue or skeletal abnormalities were observed. Dose assumed to be
reported as NasP^Oiu.
CIR Expert
Panel (2016);
FASEB/LSRO
(1975)
Na2H2P2C>7
Wistar rats,
21-24 F/group
Gavage
GDs 6-15
0.5,2.6, 12.0,
or 47.2
No treatment-related maternal toxicity and no effects on pregnancy, numbers of
implantations or resorptions, or fetal survival were observed. No treatment-related
increases in abnormalities in soft or skeletal tissue.
FDRL (1973b)
Na2H2P207
CD-I mice,
24-25 F/group
Gavage
GDs 6-15
NR (referent),
0.9,4.4, 20.2,
or 93.5
There was no treatment-related maternal toxicity and no effects on pregnancy,
numbers of implantations or resorptions, or fetal survival or weight. No
treatment-related increases in abnormalities in soft or skeletal tissue were noted.
FDRL (1973b)
Na2H2P207
Golden
hamsters,
22-24 F/group
Gavage
GDs 6-10
0.5,2.2, 10, or
46.3
No treatment-related maternal toxicity, and no effects on pregnancy, numbers of
implantations or resorptions, or fetal survival were observed. No treatment-related
increases in soft or skeletal tissue abnormalities were noted.
FDRL (1973b)
Na2H2P207
Dutch belted
rabbits,
9-12 F/group
Gavage
GDs 6-18
0.4, 1.7, 7.7, or
35.7
No treatment-related maternal toxicity and no effects on pregnancy, numbers of
corpora lutea, implantations, or resorptions, or fetal survival. No test-related
increases in abnormalities in soft or skeletal tissue.
FDRL (1973b)
Na4P207,
anhydrous
CD-I mice,
24-25 F/group
Gavage
GDs 6-15
NR (referent),
0.3, 1.4, 6.5, or
30.3
No treatment-related maternal toxicity and no effects on pregnancy, numbers of
corpora lutea, implantations, or resorptions, or fetal survival. No treatment-related
abnormalities in fetal skeletal or soft tissue were observed. One pup in the 1.4
mg/kg-d treatment group had a fused/split rib and one pup in the 6.5 mg/kg-d
exhibited soft tissue exophthalmos and encephalomeningocele, but neither
malformation occurred in the high-dose group.
FDRL (1974a)
Na4P2C>7,
anhydrous
Wistar rats,
23-25 F/group
Gavage
GDs 6-15
NR (referent),
0.3, 1.5,6.9, or
32.1
No treatment-related maternal toxicity and no effects on pregnancy, numbers of
corpora lutea, implantations, or resorptions, or fetal survival were observed. No
treatment-related increases in soft tissue or skeletal abnormalities were noted.
FDRL (1974a)
77
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 11. Summary of Unpublished Reproductive and Developmental Toxicity Studies of Na/K Salts of Inorganic
Phosphates
Ingredient
Strain/Species/
Sex/Number
Mode and
Duration
Doses
(mg P/kg-d)
Results
Reference
(NaP03)6
CD-I mice,
19-21 F/group
Gavage
GDs 6-15
1.1, 5.2, 24.3,
or 112.4
No treatment-related maternal toxicity and no effects on pregnancies or numbers of
corpora lutea or implantations were observed. An increased incidence of resorptions
was observed at the lowest dose only. Incidences of dead fetuses were 1/215 and
1/219 in the negative and positive control groups (respectively) and 2/213, 3/235,
3/224, and 4/228 at increasing doses; the differences from control were not
statistically significant. No changes in numbers of abnormalities in soft or skeletal
tissue were noted.
FDRL (1974b)
(NaP03)6
Wistar rats,
20-25 F/group
Gavage
GDs 6-15
0.7,3.4, 15.7,
or 72.9
No treatment-related maternal toxicity; no effects on pregnancy, corpora lutea,
implants or resorptions; no effects on litter parameters or fetal survival and no
changes in number of abnormalities in soft or skeletal tissue were observed.
FDRL (1974b)
KH2PO4
CD-I mice,
20-22 F/group
Gavage
GDs 6-15
0.7,3.4, 15.7,
or 72.8
No treatment-related maternal toxicity; no effects on pregnancy, corpora lutea,
implants or resorptions; and no effects on litter parameters or fetal survival were
observed. No treatment related changes in soft or skeletal tissue abnormalities were
noted.
FDRL (1975)
KH2PO4
Wistar rats,
21-28 F/group
Gavage
GDs 6-15
0.6,3.0, 13.8,
or 64.2
No treatment-related maternal toxicity; no effects of pregnancy, corpora lutea,
implantation or resorptions; and no effects on litter parameters or fetal survival or
fetal weight were observed. No changes in abnormal fetal development of soft or
skeletal tissue compared with the sham control.
FDRL (1975)
CIR = cosmetic ingredient review; F = female(s); GD = gestation day; GI = gastrointestinal; K2HPO4 = dipotassium phosphate; KH2PO4 = monopotassium phosphate;
NR = not reported; OECD = Organisation for Economic Co-operation and Development; (NaP03)6 = sodium hexametaphosphate; Na2H2P207 = sodium dihydrogen
pyrophosphate; Na4P2C>7 = tetrasodium phosphate; NasPsOio = sodium tripolyphosphate; PA = phosphoric acid; PND = postnatal day.
78
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
In a published paper, Hodge (1964) summarized the results of unpublished
three-generation reproductive toxicity studies of NasPsOio, sodium trimetaphosphate (NasP^Oio),
and (NaPO?)«,. The summaries lacked details of study design and results. Hodge (1964) reported
that for each compound, a dietary concentration of 0.5% administered to male and female rats
had no effect on fertility, litter size, growth or survival of offspring, or organ weights or
histopathology in F3 offspring. This concentration is roughly estimated to provide doses in the
range of 100-200 mg P/kg-day for adult animals, depending on compound.
Two combined repeat-dose and reproductive/developmental toxicity screening studies
were performed; available information on these studies is from secondary sources (HCHA. 2018;
CIR Expert Panel. 2016; OECD. 2009. 2006). as the original reports were not available.
Administration of K2HPO4 at a dose of 180 mg P/kg-day for 42-54 days did not affect
reproductive or developmental endpoints (see Table 11). In a comparable study of PA, however,
two females exposed to 158 mg P/kg-day died prematurely; in addition, this dose was associated
with reduced absolute kidney weight and relative uterine weight in females. No reproductive or
developmental effects were observed, and no treatment-related findings were noted in males
apart from soft stool and dirty nose in one high-dose rat.
In developmental toxicity studies of rats and mice administered PA, K2HPO4, NasP^Oio,
sodium dihydrogen pyrophosphate (Na2H2P2C>7), Na4P2C>7, (NaPCbX or KH2PO4 on gestation
days (GDs) 6-15, no maternal toxicity, developmental toxicity, or teratogenicity was observed
(OECD. 2009. 2006; FDRL. 1975. 1974a. b, 1973b). Similarly, in rabbits administered
Na2H2P2C>7 or NasP^Oio on GDs 6-18, and in hamsters administered Na2H2P2C>7 or NasP^Oio on
GDs 6-10, no maternal toxicity was observed, and neither developmental toxicity nor
teratogenicity was observed (CIR Expert Panel. 2016; FASEB/LSRO. 1975; FDRL. 1973a. b).
Taken together, these data suggest the sodium or potassium salts of inorganic phosphates
are unlikely to produce maternal, developmental, or teratogenic toxicity. However, the lack of
information on concurrent Ca intake and (in several cases) dietary contribution to P intake limits
the confidence in these findings.
2.2.2. Inhalation Exposures
No relevant animal studies of inorganic phosphate inhalation were identified in the
literature searches.
2.3. OTHER DATA (SHORT-TERM TESTS, OTHER EXAMINATIONS)
Section 2.3.1 provides an overview of genotoxicity studies (see Table 12) of Na/K salts
of inorganic phosphates, and Sections 2.3.2-2.3.4 provide summaries of other supporting studies
of Na/K salts of inorganic phosphates, including studies that did not report levels of Ca or P in
the referent group diet (see Table 13); foreign language studies (see Table 14); and studies in
which the exposure duration was <28 days (see Table 15).
79 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
2.3.1. Genotoxicity
Tests for mutagenicity in Salmonella typhimurium or Escherichia coli were negative in
the presence or absence of activation for PA, sodium phosphate salts (NaH2P04, Na2HP04,
Na2H2P2C>7, Na4P2C>7, and [NaPCbje), and potassium phosphate salts (KH2PO4, K2HPO4, and
potassium pyrophosphate [K4P2O7]) (Table 12). NaH2P04 also tested negative in the SOS
chromotest (without activation) (Quillardet et al.. 1982 and Olivier and Mar/in. 1987 as cited in
CIR Expert Panel. 2016). Tests for mutagenicity in Saccharomyces cerevisiae (strains D3 or D4)
produced negative results using NaH2P04, Na2H2P207, Na4P207, (NaP03)6, K2HPO4, and
K4P2O7 (FDA. 1975 and Weiner et al.. 2001 as cited in CIR Expert Panel 2016; Litton
Bionetics. 1975b. c). No clastogenic effects were observed in Chinese hamster lung (CHL) cells
or WI-38 human lung cells treated with PA, Na2HP04, Na2H2P207, NasP^Oio, or K2HPO4
(Ishidate et al.. 1984 as cited in CIR Expert Panel. 2016; NIER. 2005 as cited in both OECD.
2009; and OECD. 2006; Litton Bionetics. 1975a). Deoxyribonucleic acid (DNA) damage was
observed in human lymphocytes treated with PA in a comet assay (Yilmaz et al.. 2014). the only
positive result in the database.
In vivo genotoxicity assays were negative, including tests for dominant lethal mutations
in rats and host-mediated mutation assays in mice (and evaluated in S. typhimurium and
S. cerevisiae) for Na2H:P:07 and Na.sP.iOio (FDA. 1975 (PB262651) as cited in CIR Expert
Panel. 2016; Litton Bionetics. 1975a). No effect on the frequency of translocations was seen in
mice treated for 7 weeks with Na2H2P20? (FDA. 1975 as cited in CIR Expert Panel. 2016).
Chromosomal aberrations (CAs) were not significantly induced in the bone marrow of rats
treated with Na.sPsOm (Litton Bionetics. 1975a).
80 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 12. Summary of Sodium or Potassium Salts of Inorganic Phosphates Genotoxicity
Endpoint
Substance
Test System
Doses/
Concentrations
Tested3
Results
without
Activationb
Results
with
Activationb
Comments
References
Genotoxicity studies in prokaryotic organisms
Mutation
PA
Salmonella typhimurium
strains TA98, TA100,
TA1535, and TA1537, and
Escherichia coli strain WP2
uvrA
Up to
5,000 ng/plate
OECD 471 Guideline study.
Cytotoxicity was noted at
5,000 ng/plate in TA100 (+S9)
NIER (2008) as cited in
OECD (2009)
Mutation
PA
S. typhimurium strains TA97,
TA98, TA100, TA102, and
TA1535
NR
pHs ranged from 4 to 9. This
study was deemed not assignable
(lacking sufficient experimental
details) in OECD (2009).
Cioollaro et al. (1986) as
cited in CIR Expert Panel
(2016); and OECD (2009)
Mutation
PA
S. typhimurium strains TA97,
TA98, TA100, and TA104
0, 0.5, 1, 2 |iL/platc
An appropriate positive control
group was not used. This study
was deemed not reliable in
OECD (2009).
Al-Ani and Al-Lami
(1988) as cited in CIR
Expert Panel (2016); and
OECD (2009)
Mutation
NaH2P04;
anhydrous
S. typhimurium TA1535,
TA1537, and TA1538
0, 0.625, 1.25, 2.5%
—
—
Plate tests conducted at 1.25%;
suspension tests up to 2.5%.
Litton Bionetics (1975c)
Mutation
Na2HP04
S. typhimurium strains TA92,
TA94, TA98, TA100,
TA1535, and TA1537
Up to 100 mg/plate
No additional information was
reported.
Ishidate et al. (1984) as
cited in CIR Expert Panel
(2016)
Mutation
Na2HP04
S. typhimurium strains TA98,
TA100, TA1535, and
TA1537
Up to
10,000 ng/plate
No additional information was
reported.
Haworth et al. (1983) as
cited in CIR Expert Panel
(2016)
Mutation
Na2H2P20-
S. typhimurium strains TA92,
TA94, TA98, TA100,
TA1535, and TA1537
Up to 10 mg/plate
No additional information was
reported.
Ishidate et al. (1984) as
cited in CIR Expert Panel
(2016)
Mutation
Na2H2P2C>7
S. typhimurium strains TA97,
TA98, TA100, TA102, and
TA1535
5% (w/v)
No additional information was
reported.
FDA (1975) as cited in
CIR Expert Panel (2016)
Mutation
Na4P207
S. typhimurium strains
TA1535, TA1537, and
TA1538
0,0.05, 0.1% (w/v)
Plate and suspension tests.
Litton Bionetics (1975b)
81
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 12. Summary of Sodium or Potassium Salts of Inorganic Phosphates Genotoxicity
Endpoint
Substance
Test System
Doses/
Concentrations
Tested3
Results
without
Activationb
Results
with
Activationb
Comments
References
Mutation
Na4P207
S. typhimurium strains TA98,
TA100, TA1535, and
TA1537; E. coli strain WP2
uvrA
0, 301.25, 602.5,
1,205,2,410,
4,820 ng/plate
Positive and negative controls
responded appropriately.
Kimetal. (2010)
Mutation
(NaP03)6
S. typhimurium strains
TA1535, TA1537, and
TA1538
0, 0.018, 0.035%
Plate and suspension tests.
Litton Biottetics (1975d)
Mutation
KH2PO4
S. typhimurium strains
TA1535, TA1537, and
TA1538
Up to 5% (w/v)
No additional information was
reported.
FDA (1975) as cited in
CIR Expert Panel (2016)
Mutation
K2HPO4
S. typhimurium strains TA98,
TA100, TA1535, TA1537,
and TA1538
Up to 5 |iL/platc
No additional information was
reported. This study was deemed
not assignable (lacking sufficient
experimental details) in OECD
(2006).
Weiner et al. (2001) as
cited in CIR Expert Panel
(2016)
Mutation
K2HPO4
S. typhimurium strains TA97
and TA102
Up to
10,000 ng/plate
-
-
Positive and negative controls
responded appropriately.
Fuiita and Sasaki (1994)
as cited in OECD (2006)
Mutation
K4P2O7
S. typhimurium strains TA98,
TA100, TA1535, TA1537,
and TA1538
Up to 5 |iL/platc
No additional information was
reported.
Weiner et al. (2001) as
cited in CIR Expert Panel
(2016)
DNA
damage
(SOS
chromotest)
NaH2P04
E. coli WP2 uvrA
10-100,000 nM/mL
NDr
No additional information was
reported.
Ouillardet et al. (1982) and
Olivier and Marzin (1987)
as cited in CIR Expert
Panel (2016)
Genotoxicity studies in nonmammalian eukaryotic organisms
Mutation
NaH2P04;
anhydrous
Saccharomyces cerevisiae
strain D4
0, 2.5, 5, 10%
—
—
Suspension tests.
Litton Biottetics (1975c)
Mutation
Na2H2P2C>7
S. cerevisiae (strain not
specified)
NR
—
—
No additional information was
reported.
FDA (1975) as cited in
CIR Expert Panel (2016)
82
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 12. Summary of Sodium or Potassium Salts of Inorganic Phosphates Genotoxicity
Endpoint
Substance
Test System
Doses/
Concentrations
Tested3
Results
without
Activationb
Results
with
Activationb
Comments
References
Mutation
Na4P207
S. cerevisiae strain D4
0, 1,13,2.25%
(w/v)
—
—
Suspension tests.
Litton Biottetics (1975b)
Mutation
(NaP03)6
S. cerevisiae strain D4
0, 2.5, 5%
-
-
Suspension tests.
Litton Biottetics (1975d)
Mutation
K2HPO4
S. cerevisiae strain D4
Up to 5 |iL/platc
No additional information was
reported. This study was deemed
not assignable (lacking sufficient
experimental details) in OECD
(2006).
Weiner et al. (2001) as
cited in CIR Expert Panel
(2016)
Mutation
K4P2O7
S. cerevisiae strain D4
Up to 5 |iL/platc
No additional information was
reported.
Weiner et al. (2001) as
cited in CIR Expert Panel
(2016)
Genotoxicity studies in mammalian cells—in vitro
CA
PA
CHL cells
0, 112.5,225,
450 iig/mL
—
—
OECD 473 Guideline study.
NIER (2005) as cited in
OECD (2009)
CA
Na2HP04
CHL fibroblasts
Up to 2 mg/mL
NDr
No additional information was
reported.
Ishidate et al. (1984) as
cited in CIR Expert Panel
(2016)
CA
Na2H2P2C>7
CHL fibroblasts
Up to 0.5 mg/mL
NDr
No additional information was
reported.
Ishidate et al. (1984) as
cited in CIR Expert Panel
(2016)
CA
NasPaOi u
WI-38 human lung cells
0,0.1, 1.0,
10 |ig/mL
-
NDr
100 anaphase cells per dose were
evaluated.
Litton Biottetics (1975a)
CA
K2HPO4
CHL cells
0, 1,250, 2,500,
5,000 ng/mL
—
—
OECD 473 Guideline study.
NIER (2005) as cited in
OECD (2006)
DNA
damage
(comet
assay)
PA
Human lymphocytes
0, 25, 50, 100,
200 |ig/mL
+
NDr
Mean tail intensity and length
significantly increased at all test
concentrations. Cell viability
>95%.
Yilttiaz et al. (2014)
83
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 12. Summary of Sodium or Potassium Salts of Inorganic Phosphates Genotoxicity
Endpoint
Substance
Test System
Doses/
Concentrations
Tested3
Results
without
Activationb
Results
with
Activationb
Comments
References
Genotoxicity studies—in vivo
Dominant
lethal
mutations
Na2H2P2C>7
Rats (exposure NR)
Up to 720 mg/kg
NA
No additional information was
reported.
FDA (1975) as cited in
CIR Expert Panel (2016)
Dominant
lethal
mutations
NasPaOi u
Male rats were exposed via
gavage for 1 d (acute) or 5 d
(subacute) and mated to
unexposed females
0, 1,100 (subacute),
or 2,500 mg/kg
(acute)
NDr
Increased numbers of resorbed
implants (acute study) or
decreased fertility (subacute
study) were noted.
Litton Bionetics (1975a)
Mutation
(host-
mediated
assay)
Na2H2P2C>7
Mice (exposure NR)
Up to 1,400 mg/kg
NDr
Responses evaluated in
S. typhimurium strain TA1530
and S. cerevisiae strain D3.
FDA (1975) as cited in
CIR Expert Panel (2016)
CA
(heritable
translocation
assay)
Na2H2P2C>7
Male mice were exposed in
the diet for 7 wk at two dose
levels
Up to 1,400 mg/kg
NA
No additional information was
reported.
FDA (1975) as cited in
CIR Expert Panel (2016)
CA (bone
marrow)
NasPaOi u
Male rats were exposed via
gavage for 1 d (acute) or 5 d
(subacute)
0, 1,100 (subacute),
or 2,500 mg/kg
(acute)
NDr
No significant induction of CAs
at 1 or 5 d was noted.
Litton Bionetics (1975a)
CA = chromosomal aberration; DNA = deoxyribonucleic acid; K2HPO4 = dipotassium phosphate; K4P2O7 = potassium pyrophosphate; KH2PO4 = monopotassium
phosphate; (NaP03)6 = sodium hexametaphosphate; NA = not applicable; Na2H2P207 = sodium dihydrogen pyrophosphate; Na2HP04 = disodium phosphate;
Na4P2C>7 = tetrasodium phosphate; NasPsOio = sodium tripolyphosphate; NaH2P04 = monosodium phosphate; NDr = not determined; OECD = Organisation for
Economic Co-operation and Development; PA = phosphoric acid.
84
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
2.3.2. Supporting Animal Studies
Studies That Do Not Report Basal P and Ca Levels
Studies that did not provide information regarding P or Ca levels in the basal diet are
summarized (by study type) in Table 13. Because these studies did not report the amount of P in the
diet, dose estimates could not be calculated.
The most prominent effects identified in these studies pertain to the kidneys. In rats and
dogs administered inorganic phosphates (as PA, NaH2P04, Na2HPC>4, trisodium phosphate
[Na3PC>4], Na4P2C>7, NasP^Oio, NasP^Oio, [NaPCbje, and K2HPO4), increased kidney weights and
histopathological evidence of kidney damage (including nephrocalcinosis, tubular damage,
degeneration, and necrosis) were observed (Seo et al.. 2011a; Pelham et al.. 2009; Shibata et al..
1993; Schneider et al.. 1981; Hodge. 1964; MacKav and Oliver. 1935). Decreased body weights
(Pelham et al .. 2009; Hodge. 1964) and effects on bone (decreased femur length and pathological
effects) (Hodge. 1964) were also observed in rats administered inorganic phosphates. No effects on
fertility or development in rats administered inorganic phosphates as NasP^Oio, NasP^Oio, or
(NaPO?)«, were observed for three generations (Hodge. 1964).
85
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 13. Supporting Studies: >28-Day Exposure but without Basal Intake of P or Ca
Strain/Species/
Sex/Number
Mode/
Duration
Dose or Dietary
Concentration3
Results
Reference
(notes)
Short-term (>28 d) and subchronic studies
Crl:CD (SD)
rats;
10/sex/group
Gavage, 28 d
NR (control) or
5,130 mg/kg-d
NaH2P04 and
Na2HP04 in Fleet
Phospho-Soda (not
clear if mg Fleet or
mgP)
Significant mortality occurred (eight males and seven females) starting in Week 2. Clinical signs of
toxicity included dermal atonia, hypoactivity, impaired equilibrium, tremors, decreased respiration,
and diarrhea. Food consumption and body weights were significantly decreased at 0-3 wk; surviving
animals weighed 17% less than referents on Day 27. Absolute (but not relative) thymus weights
were decreased. Histopathological effects in treated rats (surviving and dead) were nephrocalcinosis
and tubular degeneration (kidney), mineralization (aorta and stomach), and degeneration and
necrosis (heart and liver).
Pelhatn et
al. (2009)
Sprague Dawley
rats,
10/sex/group
Gavage,
5 d/wk for
90 d
0, 250, 500, or
1,000 mg
Na4P2C>7/kg-d
No treatment-related mortality occurred. Increased WBC count (both sexes) and decreased RBC
count (males only) were noted at 1,000 mg Na4P207/kg-d. Liver weights were significantly increased
at >500 mg Na4P2C>7/kg-d. Kidney lesions were reported at 1,000 mg Na4P207/kg-d.
Seo et al.
(2011a)
Albino rats,
18-25 F/group
Diet, 44 d
0 or 2.94% PA, 5.50%
NaH2P04, 6.53%
Na2HP04 or 4.92% in
diet
Rats administered Na2HP04 showed decreased final body weights relative to referents. All groups of
treated rats showed increased kidney weights; this effect was accompanied by pathological signs of
kidney damage (disorganization of the outer stripe of the outer zone of the medulla, cystic dilatation
or collapse of tubules, and cell infiltration in the cortex).
MacKav
and Oliver
(1935)
Rats (strain not
specified),
5 M/group
Diet, 1 mo
0 or 5% PA
0, 0.2, 2, or 10%
NasPsOio, or
(NaP03)6
Decreased growth was reported in rats treated at 10% NasPsOio or (NaP03)6 in the diet. Rats
administered 10% as (NaP03)6 showed pale, swollen kidneys. Relative kidney weights were
increased in rats treated at 10% in the diet (as NasPsOio or |NaPO;,|r,). Tubular necrosis was
observed in rats treated at 5% as PA, and at 10% as NasPsOio, NasPsOio, or (NaP03)6; this effect was
especially apparent in NasPsOio- and (NaP03)6-treated rats. Some rats administered 2% in the diet
exhibited acute inflammatory changes of the renal pelvis or tubular lesions.
Hodee
(1964)
Chronic studies
F344 rats,
15 M/group
Diet, 32 wk
0 or 3% NaH2P04 or
Na;,POi
Body weights were not significantly affected by treatment. Changes in urinary pH and the urinary
concentrations of P (increased) and Ca (decreased) were noted. Absolute (but not relative) bladder
weights were increased in Na3P04-treated rats; absolute kidney weights were increased in both dose
groups. Na3P04-treated rats showed significantly increased incidences of hyperplasia in the urinary
bladder and calcification of the renal pelvis.
Shibata et
al. (1993)
Beagle dogs,
5 M/group
Gavage,
22 wk
0 or 1,330 mg
Na2HP04/kg-d or
1,700 mg
KzHPOVkg-d (TWA)
One K2HP04-treated dog died (Week 12); autopsy revealed enlarged, yellow kidneys and diffuse
calcification. Vomiting was noted, especially during the first week of treatment. Nephrocalcinosis
and disseminated atrophy of the proximal tubule were reported. Kidney effects were confined to the
cortex in Na2HP04-treated rats; however, lesions were observed in both the cortex and medulla of
K2HP04-treated rats.
Schneider
et al.
(1981)
86
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 13. Supporting Studies: >28-Day Exposure but without Basal Intake of P or Ca
Strain/Species/
Sex/Number
Mode/
Duration
Dose or Dietary
Concentration3
Results
Reference
(notes)
Rochester rats,
50/sex/group
Diet, 2 yr
0,0.05, 0.5, or 5%
NasPsOio or
(NaP03)6;
0,0.1, 1.0, or 10%
NasPsOiu
The incidence of mortality was high; however, deaths were most frequently attributed to respiratory
infection or pericarditis/peritonitis. Decreased body weights were reported at 5 and 10% in the diet
as NasPsOio or (NaP03)6. Kidney weights were increased in rats at 5% NasPsOio; femur length was
decreased. In general, analyses of these parameters were confounded by infection or reductions in
growth. Histopathological examinations revealed changes consistent with chronic tubular
nephropathy (5% as NasPsOio) and calcification in the tubules of the kidneys (5% as |NaPO;,|r,). No
evidence of carcinogenicity was observed.
Hodee
(1964)
Reproductive and developmental studies
Rats (strain not
specified), 16 F
and 8 M/group
Diet,
3 generations
0 or 0.5% NasPsOio,
(NaP03)3, or
(NaP03)6
No significant, treatment-related effects on fertility, litter size, or the survival and growth of
offspring were observed. Third-generation rats (sacrificed at 100 d of age) showed no changes in
organ weights or gross or microscopic pathology.
Hodee
(1964)
aBecause these studies did not report P content of the baseline diet/control group, doses in terms of mg P/kg-day were not calculated.
Ca = calcium; K2HPO4 = dipotassium phosphate; KH2PO4 = monopotassium phosphate; F = female(s); M = male(s); (NaP03)3 = sodium trimetaphosphate;
(NaP03)6 = sodium hexametaphosphate; Na2HP04 = disodium phosphate; Na3PC>4 = trisodium phosphate; Na4P2C>7 = tetrasodium phosphate; NasPsOm = sodium
tripolyphosphate; NaH2P04 = monosodium phosphate; NR = not reported; P = phosphorus; PA = phosphoric acid; RBC = red blood cell; TWA = time-weighted average;
WBC = white blood cell.
87
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Foreign Language Studies
Information about foreign language studies (Table 14) was obtained mainly from secondary
sources (CIR Expert Panel. 2016; WHO. 1982). These studies identify the kidney as a primary
target of toxicity following exposure to inorganic phosphates (NaH2P04, Na2HPC>4, NaJ^O?,
NasPjOio, [NaP03]«, K2HPO4, and K4P2O7) (Shimoii et al.. 1988; Schneider et al.. 1980a. 1980b;
Hahn. 1961; Hahn and Seifen. 1959; Hahn et at.. 1958; Hahn et at.. 1956; Nishii. 1993 all as cited in
CIR Expert Panel. 2016; or WHO. 1982). The most common renal effects identified included
increased weight and evidence of calcification (Shimoii et al .. 1988; Schneider et al.. 1980a. 1980b;
Hahn. 1961; Hahn and Seifen. 1959; Hahn et al.. 1958; Hahn et al.. 1956; Nishii. 1993 all as cited in
CIR Expert Panel. 2016; or WHO. 1982); less common effects were renal failure (Shimoii et al..
1988 as cited in CIR Expert Panel. 2016). nephropathy (Nishii et al.. 1993 as cited in CIR Expert
Panel. 2016). tubular atrophy (Schneider et al .. 1980b. 1980a as cited in CIR Expert Panel. 2016). or
necrosis (Shimoii et al.. 1988 as cited in CIR Expert Panel. 2016). Owing to the limited information
available in the secondary sources, the basal level of phosphate in the diet (i.e., phosphate exposure
in the referent group), and the units of the reported concentrations (percent compound or percent P
in diet), was generally not known; thus, doses were not estimated for this review.
88
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 14. Supporting Studies Published in Foreign Languages with English Summary Available
Strain/Species/
Sex/Number
Mode/
Duration
Dietary Concentrations3
(as reported in
secondary source)
Results
Reference
Rats (strain and
sex not
specified),
34-36/group
Diet, 24 wk
0, 1.8, 3, or 5%
NaH2P04, Na2HP04,
Na4P2C>7, or \;i 7 growth was adversely affected
at 5%. NasPsOm induced nephrocalcinosis at 3 and 5%, while calcification was slight
or absent and kidney weights were unaffected at 1.8%. Nephrocalcinosis was
reported at all doses of NasPsOio; damage was less prevalent at pH 5 than at pH 9.5.
Halm and Seifen
(1959) as cited in CIR
Expert Panel (2016);
or WHO (1982)
Rats (number,
strain, and sex
not specified)
Diet, 39 wk
0 or l.l%NaH2P04;
0, 1.1, 1.8, 3, or 5%
Na2HP04, Na4P2C>7, or
NasPsOm
Slight kidney calcification was reported in rats treated at 1.1% NaH2P04 or
Na;HPOi. Effects of Na iP;0- and NasPsOio were not reported. According to CIR
Expert Panel (2016). the LOEL was 495 me/ke-d (based 011 food consumption and
body weight values of 0.018 kg/d and 0.35 kg, respectively); whether the dose was
as P or as the compound was unclear.
Halm and Seifen
(1959) as cited in CIR
Expert Panel (2016);
or WHO (1982)
Male Wistar rats
(number not
specified)
Diet, 8 wk
0 or 10% K2HPO4
Kidney toxicity was reported in rats treated at 10% in the diet.
Nishii et al. (1993) as
cited in CIR Expert
Panel (2016)
Diet (duration
not specified)
0 or 5% K2HPO4
Renal calcification and severe nephropathy were reported.
Beagle dogs,
n= 15
Diet, 14 or
38 wk
0 or 800 mg/kg-d as
K2HPO4 (not clear if mg
K2HPO4 or mg P)
Evidence of kidney damage was found in all dogs (severity was greater at 38 wk
compared with 14 wk). Renal damage included disseminated tubular atrophy
(predominantly affecting the proximal tubules), focal scar tissue, and
nephrocalcinosis.
Schneider et al.
(1980a. 1980b) as
cited in CIR Expert
Panel (2016)
F344 rats,
60/sex/group
Diet (duration
not specified)
0.6, 1.25, 2.5, 5, or 10%
as K4P2O7; whether the
0.6%-group was the study
control was unclear
Three rats treated at 10% in the diet died; deaths were attributed to renal failure.
Histopathological effects noted at 2.5 and 5% included necrosis and calcification of
renal tubules, ulceration or granuloma formation in the tongue mucosa, and
hypertrophy of the salivary glands.
Shimoii et al. (1988)
as cited in CIR Expert
Panel (2016)
aCIR Expert Panel (2016) did not clearly indicate whether the dietary concentrations were reported as percent compound or percent P in diet. For the purpose of this
table, the percentages were assumed to be as the compound. In addition, because the CIR Expert Panel (2016) did not report P content of the baseline diet/control group,
doses in terms of mg P/kg-day were not calculated.
K4P2O7 = potassium pyrophosphate; KH2PO4 = monopotassium phosphate; LOEL = lowest-observed-effect level; Na2HP04 = disodium phosphate;
Na4P2C>7 = tetrasodium phosphate; NasPsOio = sodium tripolyphosphate; NaH2P04 = monosodium phosphate.
89
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Short-Term Studies
Short-term studies (typically 14-28 days in duration) are summarized in Table 15. The
following types of effects were identified in these studies:
• Decreased growth. Rats administered diets containing NaH2P04, NasP^Oio, KH2PO4, or
potassium tripolyphosphate (K5P3O10) at >960 mg P/kg-day showed decreased body weights
relative to referents (Katsumata et al.. 2015; Katsumata et al.. 2005; Matsuzaki et al.. 2002;
Matsuzaki et al.. 1999).
• Changes in serum chemistry associated with kidney function. Rats treated with
phosphates at >960 mg P/kg-day as NaH2P04 or K5P3O10 showed significantly increased
BUN (Matsuzaki et al.. 1999; Matsuzaki et al.. 1997).
• Increased kidney weights. In most studies, rats treated with phosphates in the diet as
NaH2PC>4, NasPsOlo, KH2PO4, or K5P3O10 showed significantly increased kidney weights.
Effects were seen at approximately >1,000 mg P/kg-day (Katsumata et al.. 2015; Matsuzaki
et al.. 2010; Matsuzaki et al.. 2002; Matsuzaki et al.. 1999).
• Evidence of kidney damage. Increased kidney weights were accompanied by kidney
lesions, most frequently, nephrocalcinosis (Katsumata et al.. 2015; Matsuzaki et al.. 2010;
Matsuzaki et al.. 2002; Matsuzaki et al.. 1999). In one study (Matsuzaki et al.. 1997).
changes to the proximal tubules (vacuoles, lysosomes, swelling of microvilli, hydroxyapatite
deposits) were noted in rats treated at 1,400 mg P/kg-day as K5P3O10.
• Bone effects. Two studies by Katsumata et al. (2015) and Koshihara et al. (2005) showed
increased markers of bone turnover, decreased compression or bending load, and decreased
bmc or bmd (femur, tibia, or lumbar vertebra) in rats administered >830 mg P/kg-day as
KH2PO4.
90
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 15. Supporting Studies of Acute or Short-Term (<28 Days) Exposure Duration
Strain/Species/
Sex/Number
Mode/
Duration
Dietary P
Concentration
(%)
Dietary Ca
Concentration
(%)
Results
Reference
Wistar rats,
5 males/group
Diet, 21 d
0.3 (referent) or
1.4% (as
NaH2P04,
calculated for
this review)
0.5% (both
groups)
Rats treated at 1.4% in the diet showed significantly decreased food consumption
and body weights (23% lower than referents). Serum BUN was significantly
increased. Kidney concentrations of P and Ca and relative kidney weights were also
significantly increased. All animals showed evidence of nephrocalcinosis (mean
severity score = 1.6; maximum score = 4). Based on measured food consumption
(0.0178 and 0.0131 kg/d) and body weights (0.207 and 0.173 kg), 0.3 and 1.4% in
the diet are equivalent to approximately 260 and 1,100 mg P/kg-d.
Matsuzaki et al.
(1999)
Wistar rats,
5 males/group
Diet, 21 d
0.3 (referent) or
1.4% (as
NasPaOiu.
calculated for
this review)
0.5% (both
groups)
Rats treated at 1.4% in the diet showed significantly decreased food consumption
and body weights (35% lower than referents). Urinary albumin and NAG activity
were significantly increased. Kidney concentrations of P and Ca and relative kidney
weights were also significantly increased. All animals showed evidence of
nephrocalcinosis (mean severity score = 3.8; maximum score = 4). Based on
measured food consumption (0.0183 and 0.0108 kg/d) and body weights (0.206 and
0.157 kg), 0.3 and 1.4% in the diet are equivalent to approximately 270 and
960 mg P/kg-d.
Matsuzaki et al.
(1999)
Wistar rats,
5 males/group
Diet, 14 d
0.3 (referent) or
1.2% (as
KH2PO4)
0.5% (both
groups)
Food consumption and body weights were not significantly affected by treatment.
Rats treated at 1.2% in the diet showed significantly decreased levels of magnesium
and Ca and significantly increased levels of P, PTH, osteocalcin, and C-terminal
telopeptide of type I collagen (CTx) in the serum. No significant effects were
observed on femoral mineral content (with respect to P, Ca, or magnesium). Based
on measured food consumption (0.0219 and 0.0210 kg/d) and body weights
(0.169 and 0.162 kg), 0.3 and 1.2% in the diet are equivalent to approximately
390 and 1,600 mg P/kg-d.
Matsuzaki et al.
(2010)
91
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 15. Supporting Studies of Acute or Short-Term (<28 Days) Exposure Duration
Strain/Species/
Sex/Number
Mode/
Duration
Dietary P
Concentration
(%)
Dietary Ca
Concentration
(%)
Results
Reference
Wistar rats,
6 males/group
Diet, 21 d
0.3 (referent),
0.9, or 1.5% (as
KH2PO4)
0.5% (both
groups)
Food consumption and body weights were statistically significantly decreased at
1.5% in the diet. Relative to referents, rats treated at 0.9 and 1.5% in the diet showed
significant increases in serum P, excretion of CTx (a marker of bone turnover), bone
formation rate (lumbar vertebra), osteoclast number (tibia and lumbar vertebra), and
mRNA expression of receptor activator of NF-kB ligand (RANKL; femur); bmc and
bmd (femur and lumbar vertebra) and ultimate compression load (lumbar vertebra)
were significantly decreased. Additional effects at 1.5% in the diet included
significantly decreased serum Ca, increased serum PTH and osteocalcin, and
decreased ultimate bending load (femur). Based on measured food consumption
(0.017, 0.017, and 0.012 kg/d) and body weights (0.187, 0.184, and 0.154 kg), 0.3,
0.9, and 1.5% in the diet are equivalent to approximately 270, 830, and
1,200 mg P/kg-d.
Katsumata et al.
(2005)
Wistar rats,
6 males/group
Diet, 21 d
0.3 (referent) or
1.5% (as
KH2PO4)
0.5% (both
groups)
Food consumption and body weights were statistically significantly decreased at
1.5% in the diet. Relative to referents, rats treated at 1.5% in the diet showed
significantly decreased serum Ca and increased serum P, PTH, and osteocalcin;
increased urinary albumin, CTx, NAG activity, and (32-microglobulin excretion;
increased kidney Ca and P and increased relative kidney weight; decreased bmc and
bmd (femur, tibia, and lumbar); and increased RANKL expression (femur). Based
on measured food consumption (0.017 and 0.014 kg/d) and body weights (0.195 and
0.172 kg), 0.3 and 1.5% in the diet are equivalent to approximately 260 and
1,200 mg P/kg-d.
Katsumata et al.
(2015)
Wistar rats,
5 males/group
Diet, 21 d
0.3 (referent) or
1.4% (as
KH2PO4,
calculated for
this review)
0.5% (both
groups)
Rats treated at 1.4% in the diet showed significantly decreased food consumption
and body weights (21% lower than referents). Serum BUN was significantly
increased. Relative kidney weights were also statistically significantly increased. All
animals showed evidence of nephrocalcinosis (mean severity score = 1.2; maximum
score = 4). Based on measured food consumption (0.0177 and 0.0143 kg/d) and
body weights (0.202 and 0.173 kg), 0.3 and 1.4% in the diet are equivalent to
approximately 260 and 1,200 mg P/kg-d.
Matsuzaki et al.
(1999)
92
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 15. Supporting Studies of Acute or Short-Term (<28 Days) Exposure Duration
Strain/Species/
Sex/Number
Mode/
Duration
Dietary P
Concentration
(%)
Dietary Ca
Concentration
(%)
Results
Reference
Wistar rats,
5/sex/group
Diet, 21 d
0.3 (referent),
0.6, 0.9, 1.2, or
1.5% (as
KH2PO4)
NR
Food consumption and body weights were adversely affected in rats treated at 1.5%
in the diet. Creatinine clearance was higher in rats of both sexes treated at 1.5% in
the diet; urinary albumin was significantly increased at >0.9% in both sexes. Kidney
concentrations of Ca and P were significantly increased in males at 1.2 or 1.5%, and
in all groups of treated females. Relative kidney weights were statistically
significantly increased in 1.5% males and 1.2 and 1.5% females. Based on measured
food consumption (0.018-0.021 kg/d in males and 0.014-0.017 kg/d in females)
and body weights (0.197-0.224 kg in males and 0.156-0.177 kg in females), 0.3,
0.6, 0.9, 1.2, and 1.5% in the diet are equivalent to approximately 280, 550, 820,
1,100, and 1,400 mg P/kg-d in males and 290, 560, 860, 1,100, and 1,400 mg P/kg-d
in females.
Matsuzaki et al.
(2002)
Wistar rats,
42 males/group
Diet, 21 d
(sacrificed
atO, 1, 3,
5, 7, 14,
and 21 d)
0.5 (referent) or
1.5% (as
K5P3O10)
0.5% (both
groups)
Rats treated at 1.5% in the diet showed significantly increased serum BUN (21 d),
decreased urinary pH (21 d), and increased urinary albumin (3-21 d), NAG activity
(1-21 d), and (^-microglobulin excretion (14-21 d). After 21 d, kidney
concentrations of P and Ca and kidney weights were statistically significantly
increased. Nephrocalcinosis was noted in 4/6 rats after 1 d and in 6/6 rats at all other
time points; the severity of kidney effects increased overtime. Changes in the
proximal tubules included vacuoles, lysosomes, and swelling of the microvilli (1 d);
giant lysosomes with Ca deposits and hydroxyapatite deposition in mitochondria
(3 d); hydroxyapatite deposition in microvilli (5 d); and necrotic cells (21 d). Based
on reference values for food consumption (0.0204 kg/d) and body weights
(0.217 kg) for male Wistar rats, 0.5 and 1.5% in the diet are equivalent to
approximately 470 and 1,400 mg P/kg-d.
Matsuzaki et al.
(1997)
93
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 15. Supporting Studies of Acute or Short-Term (<28 Days) Exposure Duration
Strain/Species/
Sex/Number
Mode/
Duration
Dietary P
Concentration
(%)
Dietary Ca
Concentration
(%)
Results
Reference
Wistar rats,
5 males/group
Diet, 21 d
0.3 (referent) or
1.4% (as
K5P3O10,
calculated for
this review)
0.5% (both
groups)
Rats treated at 1.4% in the diet showed significantly decreased food consumption
and body weights (43% lower than referents). Serum BUN and urinary albumin and
NAG activity were significantly increased. Kidney concentrations of P and Ca and
relative kidney weights were also significantly increased. All animals showed
evidence of nephrocalcinosis (mean severity score = 3.2; maximum score = 4).
Based on measured food consumption (0.0189 and 0.0101 kg/d) and body weights
(0.208 and 0.146 kg), 0.3 and 1.4% in the diet are equivalent to approximately
270 and 970 mg P/kg-d.
Matsuzaki et al.
(1999)
bmc = bone mineral content; bmd = bone mineral density; BUN = blood urea nitrogen; Ca = calcium; K5P3O10 = potassium tripolyphosphate; KH2PO4 = monopotassium
phosphate; mRNA = messenger RNA; NAG = N-acetyl-P-D-glucosaminidase; NasPsOio = sodium tripolyphosphate; NaH2P04 = monosodium phosphate; NR = not
reported; P = phosphorus; PTH = parathyroid hormone.
94
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
2.3.3. Metabolism/Toxicokinetics
Inorganic phosphate is absorbed from the GI tract, with net P absorption ranging from
55 to 80% of intake in adults and from 60 to 95% of intake in infants and children (EFSA. 2015;
TOM. 1997). Intestinal absorption occurs by passive absorption at tight junctions between
intestinal cells and by sodium-dependent active transport, with the relative contribution of each
mechanism dependent on the luminal concentration of phosphate (Chang and Anderson. 2017;
EFSA. 2015; Lee and Marks. 2015; Marks et at.. 2010). Multiple sodium-dependent transporters
are proposed to mediate intestinal phosphate absorption in rats, including the intestinal type II
sodium phosphate cotransporter, NaPi-llb, and Type 111 transporters, PiTl and PiT2 (Candeat et
at.. 2017; Marks et at.. 2010). NaPi-llb plays a key role in intestinal absorption of inorganic
phosphate at low dietary concentrations or during fasting; however, studies in NaPi-llb knockout
mice suggest the sodium-independent pathway is also important, especially when dietary
phosphate levels are elevated (Chang and Anderson. 2017; Lee and Marks. 2015). Intestinal
phosphate absorption by active transport is physiologically regulated by dietary phosphate, PTH,
1,25-dihydroxy-vitamin D3, epidermal growth factor, glucocorticoids, estrogen, metabolic
acidosis, and phosphatonins including FGF-23 (EFSA. 2015; Lee and Marks. 2015).
The oral bioavailability of inorganic phosphates depends largely on the amount of
coingested Ca, which can bind with high affinity to phosphate in the digestive tract and prevent
its absorption (EFSA. 2015). A study of Ca and P balance in healthy adults revealed increased
dietary Ca intake without a corresponding increase in phosphate intake reduced the absorption of
phosphate from the intestine (Heanev and Nordin. 2002). In dietary studies, oral bioavailability
also depends on the food source (animal or plant derived) and the organic or inorganic form of
the P-containing compound (EFSA. 2015; Catvo and Tucker. 2013). Organic forms of phosphate
esters require enzymatic hydrolysis by phosphatases in the intestinal lumen to produce phosphate
that can be absorbed (EFSA. 2015; Catvo and Tucker. 2013). This slows the phosphate
absorption rate and reduces the efficiency of phosphate absorption from organic sources. In
addition, humans lack the enzyme needed to digest phytic acid, which is the storage form of P in
plants. Intestinal bacteria contain phytase, however, which can hydrolyze phytic acid and
increase the bioavailability of phosphate from plant sources (EFSA. 2015). Food additives and
supplements contain inorganic salts of phosphate, which dissociate readily in the gut and are
rapidly absorbed and highly bioavailable (EFSA. 2015; Catvo and Tucker. 2013). Noori et at.
(2010b) estimated the GI absorption of P to be 10-30% for plant-based proteins, compared with
40-60%) from animal protein and 80-100%) from additives and preservatives.
Phosphate ions (i.e., HPO42 and H2PO4 ) are transported in blood plasma, with
approximately 85-90% existing as free serum phosphate and 10-15%) bound to protein (EFSA.
2015). Phosphorus is stored in bones and teeth, primarily as a hydroxyapatite complex with Ca
(EFSA. 2015). Total body P in adults exists as approximately 85% hydroxyapatite, 14%> as
components of cells in soft tissues, and 1% in serum (Chang and Anderson. 2017). Phosphate is
present in breast milk and crosses the placenta during pregnancy using sodium-dependent
transporters (i.e., NaPi-llb) and is maintained in fetal serum at higher concentrations than found
in the maternal circulation (EFSA. 2015).
Inorganic phosphates are eliminated in both urine and feces (EFSA. 2015). Under normal
dietary conditions, approximately 10-20%) of the P filtered by the kidney is ultimately excreted,
with the remainder being reabsorbed in the proximal tubule (Chang and Anderson. 2017; EFSA.
2015; Marks et at.. 2010). Sodium-phosphate cotransporters, NaPi-lla, NaPi-llc and PiT2, are
95 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
responsible for the reabsorption of phosphate in renal tubules (Chang and Anderson. 2017;
EFSA. 2015). Inorganic phosphates found in feces arise from pancreatic, biliary, and intestinal
secretions (EFSA, 2015).
Serum phosphate levels are maintained within relatively narrow limits by interactions or
crosstalk between the intestine, kidney, parathyroid glands, and bone (Chang and Anderson.
2017; Rittcr and Slatopolskv. 2016; EFSA. 2015; Anderson. 2013). PTH and FGF-23 (produced
by bone osteoblasts and osteocytes) reduce serum phosphate levels by downregulating the renal
cotransporters NaPi-IIa and NaPi-IIc, resulting in decreased phosphate reab sorption and
increased excretion by the kidney (Chang and Anderson. 2017; EFSA. 2015; Lee and Marks.
2015; Anderson. 2013; Marks et at.. 2010). Increased serum phosphate concentrations also
produce a reduction in renal 1,25-dihydroxy-vitamin D3 synthesis, resulting in decreased
intestinal absorption of phosphate (Brown and Razzaque. 2015; EFSA. 2015; Lee and Marks.
2015; Marks et at.. 2010). Conversely, a reduction in serum phosphate levels decreases PTH and
FGF-23 release, resulting in decreased renal excretion of P. 1,25-Dihydroxy-vitamin D3
synthesis in the kidney also increases when serum phosphate is low, resulting in enhanced
intestinal absorption of phosphate (EFSA. 2015). Klotho is a transmembrane protein that is
necessary for FGF-23 receptor binding and signal transduction in the kidney and parathyroid
glands. Experiments with Klotho knockout mice indicate that this protein is a critical cofactor in
regulation of phosphate homeostasis by FGF-23 (Erben and Andrukhova. 2017; Rittcr and
Slatopolskv. 2016; Gutierrez. 2013).
Serum phosphate concentrations are not considered a reliable biomarker for dietary
exposure because these homeostatic mechanisms maintain serum phosphate levels within a
narrow range, even in the presence of wide variations in intake (EFSA. 2015). Urinary phosphate
excretion is similarly regulated by homeostatic mechanisms and thus has limitations as a
potential biomarker for oral exposure (EFSA. 2015).
2.3.4. Mode-of-Action/Mechanistic Studies
Disruption of phosphate homeostasis leads to renal and cardiovascular toxicity and
decreased bone health. Altered endocrine communication between bone-derived FGF-23 and
kidney-derived a-Klotho leads to increased serum phosphate levels (Brown and Razzaque. 2015;
Gutierrez. 2013). a-Klotho changes cellular Ca homeostasis, by both increasing the expression
and activity of TRPV5 (a Ca channel protein, decreasing phosphate reabsorption in the kidney)
and decreasing that of TRPC6 (a receptor-activated Ca channel, decreasing phosphate absorption
from the intestine). a-Klotho increases kidney Ca reabsorption by stabilizing TRPV5 (Huang.
2010). Mice lacking either FGF-23 or the a-Klotho enzyme display premature aging due to
hyperphosphatemia. Many of these symptoms can be alleviated by feeding the mice a
low-phosphate diet (Kuro. 2019). Elevated serum phosphate combines with free ionized serum
Ca to form a Ca-P product that can be deposited in tissues in a process referred to as ectopic
calcification (Brown and Razzaque. 2015). Ectopic calcification in the renal parenchyma leads to
tubule damage and interstitial fibrosis (Chang and Anderson. 2017; Brown and Razzaque. 2015).
Ectopic calcification also occurs in vascular endothelial cells, leading to arteriosclerosis,
hypertension, LVH, and aortic valve disease (Brown and Razzaque. 2015; Anderson. 2013).
Increasing serum Ca-P product upregulates osteocalcin, a bone matrix protein, which promotes
further vascular calcification (Brown and Razzaque. 2015). Studies in cultured vascular smooth
muscle cells show that high phosphate concentrations promote expression of osteochondrogenic
96 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
differentiation markers and extracellular matrix calcification (Menon and Ix. 2013). Other
mechanisms that may contribute to hypertension include impaired vasodilation in endothelial
cells, high serum PTH levels, and increased renin-angiotensin activity (Brown and Razzaque.
2015; Anderson. 2013). Cardiovascular disease is commonly found in patients with
hyperphosphatemia resulting from CKD (Chang and Anderson. 2017; Rittcr and Slatopolskv.
2016; Nadkarni and Uribarri. 2014).
The Ca:P ratio in the diet influences bone health (Brown and Razzaque. 2015; EFSA.
2015). Experimental animal studies have demonstrated that a high P intake combined with a low
Ca intake results in bone resorption, low peak bone mass, and increased bone fragility (EFSA.
2015; Calvo and Tucker. 2013). Cross-sectional studies in humans suggest an association
between the dietary Ca:P molar ratio and bmd (EFSA. 2015; Calvo and Tucker. 2013). Calvo
and Tucker (2013) suggested that phosphate-induced dysregulation of PTH, FGF-23, and
1,25-dihydroxy-vitamin D3 may contribute to progressive bone loss with age (i.e., osteoporosis).
Increased serum P also causes hyperplasia of the parathyroid gland, leading to secondary
hyperparathyroidism, followed by high-turnover bone disease and increased risk of fracture
(Nadkarni and Uribarri. 2014).
Little information is available on potential mechanisms of phosphate-induced cancer. A
review by Brown and Razzaque (2018) hypothesizes that phosphate is a mitogenic factor that can
enhance tumor cell growth, with excessive phosphate stimulating growth promoting cell
signaling and neovascularization. Wilson et al. (2015) postulated a role for P-induced increases
in PTH influencing the metastasis of prostate tumors to bone. These study authors noted that
PTH promotes bone remodeling, and that prostate cancer is more likely to metastasize to bone
when remodeling activity is higher. This sequence of events was suggested as a possible
explanation for the shorter latency to prostate cancer diagnosis observed in their study (Wilson et
al.. 2015).
97
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
3. DERIVATION OF PROVISIONAL VALUES
3.1. DERIVATION OF PROVISIONAL REFERENCE DOSES
The human and animal toxicity literature for dietary P and Na/K salts of inorganic
phosphates, coupled with mechanistic information, clearly indicates that impaired P homeostasis
is associated with renal toxicity (acute phosphate nephropathy/nephrocalcinosis) and GI
symptoms, is potentially associated with cardiovascular effects (including increased risk of
mortality or cardiovascular events), and may alter bone composition.
As discussed earlier, oral data pertinent to the hazard assessment of sodium or potassium
salts of inorganic phosphates can be grouped into four main categories: human epidemiological
studies on associations between dietary P intake and health outcomes; controlled trials of humans
exposed to sodium or potassium salts of inorganic phosphates for acute or short-term durations;
human studies of renal toxicity or GI symptoms after acute exposure to OSP for bowel cleansing
or constipation treatment; and short-term and subchronic animal toxicity studies. Sections 2.1
and 2.2 above provide the rationales for selecting human and animal data for consideration in the
assessment, and these details will not be reiterated here.
All the above sources were considered relevant to hazard assessment. However, several
of these sources suffer from limitations that render them less useful for dose-response
assessment. Specifically, the human dietary intake studies have uncertainties that preclude their
use for dose-response assessment. These include potential underestimation of dose estimates and
varying bioavailabilities of different P sources. In addition, although the acute colonoscopy
preparation studies provide important hazard information and identify clear effect levels, the
considerable uncertainty involved in extrapolating from a single-day exposure to subchronic or
chronic exposure limits value of these data for dose-response assessment.
Dietary intake of phosphates is difficult to quantify. EFSA (2015) noted that food product
processing and formulations change continuously, making it difficult for food composition
databases to keep pace with the changes. Further, P-containing food additives are used in a large
variety of foods including baked goods, meats, and beverages. To capture the inputs from all of
these sources for an epidemiological study requires that the FFQ include all of these foods and be
updated over time to reflect changing content. Whether an FFQ administered twice in 4 years
correlates with long-term dietary intake patterns is uncertain. Several reviews (McCture et al..
2017; EFSA. 2015) have suggested that dietary intake of P is usually underestimated for the
reasons outlined above.
The bioavailability of P from different sources varies widely, as discussed in
Section 2.3.3. Oral bioavailability of dietary P depends on the food source (animal or plant
derived) and the organic or inorganic form of the P-containing food: bioavailability is lowest for
plant sources (10-30%), higher for animal sources (40-60%), and highest for food additives
(80-100%)), which includes phosphoric acid and several (Na/K) phosphates. Most of the dietary
intake studies did not account for different sources of P in the diet, and none of the studies
provided estimates of intake for subcategories of dietary P that may be relevant to the assessment
of sodium or potassium salts of inorganic phosphates.
98 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
As a result of the uncertainties in doses and differing bioavailability of P forms inherent
in the dietary intake studies, the studies considered for dose-response assessment for sodium or
potassium salts of inorganic phosphates were restricted to those in which a clearly defined test
material was administered. These include the controlled human exposure studies, the human
colonoscopy preparation studies, and most (but not all) of the animal studies.
A few short-term, controlled-exposure human studies measuring adverse outcomes
(Chang et al.. 2017; Medoff et al.. 2004; Grimm et al.. 2001) were identified. These studies
ranged in duration from 3 to 6 weeks. Of the three studies, two (Medoff et al.. 2004; Grimm et
al.. 2001) reported that the test material administered to the volunteers, while Chang et al. (2017)
reported only that the supplemented P was in the form of food additives provided as
commercially available beverages and breakfast bars. Grimm et al. (2001) administered a
combination of calcium phosphate (CasfPO^OH) in orange juice and NaH2P04 tablets. Because
the supplemental Ca provided by the calcium phosphate may limit the absorption of phosphate,
this study also has limitations for dose-response assessment. Finally, Medoff et al. (2004)
administered NaH2P04 tablets to a group of patients with chronic constipation, but allowed the
patients to vary the dose to improve their response or mitigate side effects. Assigning a clear
LOAEL for this study is difficult. For these reasons, the controlled exposure studies are not
considered to provide reliable dose-response information and are not considered for use in the
derivation.
The human colonoscopy preparation studies appear to offer several advantages. First, the
test material (sodium phosphate) is clearly defined, consistent across exposed persons, and of
predictable bioavailability. In addition, unlike the dose estimates in dietary intake studies, the
prescribed dose of sodium phosphate taken in preparation for colonoscopy is well characterized.
The primary disadvantage of the human colonoscopy preparation studies for dose-response
assessment is that exposure to inorganic phosphates occurred for only a single day; thus, using
these data to derive a provisional reference dose (p-RfD) would necessitate extrapolation from a
single-day exposure to subchronic or chronic exposure. Some evidence suggests that
intake-related, short-term spikes in serum P may lead to acute insults (e.g., transient decreases in
endothelial dysfunction) (Nishi et al.. 2015; Shu to et al.. 2009) that could accumulate over time
and lead to permanent damage or functional changes, despite homeostatic mechanisms that
regulate serum P. However, due to the uncertainty in extrapolating from effects of a single day to
subchronic or chronic exposure, these data were not selected for use in p-RfD derivation.
Given the limitations articulated above in the human toxicity database, these studies were
not used for toxicity value derivation, which instead relies on animal toxicity data. The human
toxicity information is used as supporting information: specifically, the estimated effect levels
defined by the colonoscopy preparation studies, FDA warnings, and controlled exposure studies
were compared with the p-RfDs derived from the animal studies to ensure that the p-RfDs will
adequately protect against effects observed in humans exposed for acute and short-term
durations.
99 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
3.1.1. Derivation of the Subchronic Provisional Reference Dose
The published, peer-reviewed rabbit study by Ritskes-Hoitinga et al. (2004) was selected
as the principal study for deriving the subchronic p-RfD for Na/K salts of inorganic phosphates.
Several subchronic animal studies met criteria for inclusion in the assessment
(see Table 3 A). Among these studies, the lowest LOAEL was 290 mg P/kg-day (HED
120 mg/kg-day) for nephrocalcinosis in rabbits (Ritskes-Hoitinga et al.. 2004). LOAELs in the
other studies in the database were as high as 1,100 mg P/kg-day (HED 270 mg/kg-day) based on
renal, bone, and body-weight effects in rats (Abuduli et al.. 2016; Huttunen et al.. 2007; Tani et
al.. 2007; Koshihara et al.. 2005; Ritskes-Hoitinga et al.. 1989; Hitchman et al.. 1979; Datta et
al.. 1962; Dvmszaetal.. 1959). One of the two subchronic experiments reported by Datta et al.
(1962) lacked information needed to identify a LOAEL, so this experiment was not considered.
From the remaining experiments, the five identifying the lowest LOAELs were considered for
use in the dose-response assessment: Huttunen et al. (2007). Koshihara et al. (2005). Ritskes-
Hoitinga et al. (2004). Ritskes-Hoitinga et al. (1989). and the remaining experiment reported by
Datta et al. (1962).
All these studies were limited in examination of toxicological endpoints, mostly focusing
on renal, cardiovascular, or bone effects. Candidate points of departure (PODs) from these
studies, reported in HEDs (see calculations in study descriptions in Sections 2 and 2.2), are
shown in Table 16. As discussed above and shown in the table, all the animal studies included a
referent ("control") group with nonzero phosphate intake, because P is an essential nutrient, and
a lack of P in the diet is harmful to health. The doses administered to the referent groups are
reported in the table but should not be interpreted as NOAELs (due to the lack of a comparison
group) and were not considered as candidate PODs.
100
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 16. Candidate PODs for Noncancer Dose-Response Assessment of
Na/K Salts of Inorganic Phosphates
Species and
Study Type
Endpoint
LOAEL
(HED)
(mgP/kg-d)
Referent Group
Dose (HED)
(mg P/kg-d)
Reference
Rabbit; dietary
exposure to
NaH2P04
dihydrate for
8 wk
Nephrocalcinosis, on the basis of
significantly increased kidney Ca and P,
and significantly increased incidence and
severity scores for cortical calcifications
120
63
Ritskes-Hoitinga
et al. (2004)
Rat; dietary
exposure to
NaH2P04
dihydrate for
4 wk
Nephrocalcinosis, on the basis of
statistically or biologically significant
increases in urine albumin and relative
kidney weights
120
86
Ritskes-Hoitinga
et al. (1989)
Rat; dietary
exposure to
KH2PO4 for 6 wk
Significantly bmd (fifth lumbar vertebra),
significantly increased serum osteocalcin
and urinary deoxypyridinoline
200
68
Koshihara et al.
(2005)
Rat; dietary
exposure to
Na2HPC>4 for
16 wk
Increased relative kidney weight
(17-39%); impaired kidney function;
increased incidence of renal histopathology
(medullary calcification and necrosis,
tubular casts, hemorrhages/exudate,
chronic inflammatory changes)
200
67
Datta et al.
(1962)
Rat; dietary
exposure to
CaHP04 and
KH2PO4 for 8 wk
Significantly decreased body weight (11%)
and food intake; significantly decreased
bmc and bmd; alterations in several
measures of bone histomorphometry;
nonsignificant decrease in bone strength.
240
120
Huttunen et al.
(2007)
bmc = bone mineral content; bmd = bone mineral density; Ca = calcium; CaHP04 = calcium phosphate;
HED = human equivalent dose; KH2PO4 = monopotassium phosphate; LOAEL = lowest-observed-adverse-effect
level; Na/K = sodium and potassium; Na2HP04 = disodium phosphate; NaH2P04 = monosodium phosphate;
P = phosphorus; POD = point of departure.
The lowest LOAELs (HED) (120 mg P/kg-day) are based on nephrocalcinosis in male
rabbits exposed to NaH:P04 in the diet for 8 weeks (Ritskes-Hoitinga et al.. 2004) and also in
female rats exposed to NaH:P04 in the diet for 4 weeks (Ritskes-Hoitinga et al. 1989). Renal
effects were also the basis of the LOAEL (HED) (200 mg P/kg-day) for the subchronic study
(Datta ct al.. 1962). in which rats received Na:HP04 for 16 weeks. The renal effects seen in these
studies (nephrocalcinosis, increased kidney weights, impaired kidney function) are similar to the
acute and chronic kidney injury (e.g., acute phosphate nephropathy) seen in some human studies
after exposure to OSP compounds for colonoscopy preparation. The fact that this potential POD
is identical in rats and rabbits adds confidence for nephrocalcinosis as the critical effect and 120
mg/kg-d P as the POD.
101 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
LOAEL (HED) values for the other two studies shown in Table 16 (Huttunen et al.. 2007;
Koshihara et al.. 2005) were 240 and 200 nig P/kg-day (respectively) based on decreases in bind
and related parameters. Unlike for renal effects, there is less evidence for parallel changes in
bone metabolism in humans exposed to higher dietary intakes of phosphates, although the
available literature examining these endpoints suffers from the same limitations as other dietary
intake studies. Available information in humans appears to point to improved bone health with
higher dietary P intake (Jones et al.. as cited in HFSA. 2015; Heppe et al.. 2013; Yin et al.. 2010;
Tobias et al.. 2005; Elmstahl et al.. 1998). Because the bone effects in animals are not clearly
reflected in effects seen in humans while renal effects are, and given the lower LOAELs for renal
effects, neither Koshihara et al. (2005) nor Huttunen et al. (2007) was considered further for use
in the derivation.
Ritskes-Hoitinga et al. (2004). Ritskes-Hoitinga et al. (1989). and Datta et al. (1962) did
not include more than one nonreferent dose with suitable data, so these data sets were not
amenable to benchmark dose (BMD) modeling. Both Ritskes-Hoitinga et al. (2004) and Ritskes-
Hoitinga et al. (1989) identified a LOAEL (HED) value of 120 nig P/kg-day, which is lower than
the LOAEL (HED) value from Datta et al. (1962). Thus, these studies were selected for use in
deriving the subchronic p-RfD for Na/K salts of inorganic phosphates, and the LOAEL (HED) of
120 mg P/kg-day for increased incidence of nephrocalcinosis in rabbits exposed to NaH2P04
dihydrate for 8 weeks via the diet (Ritskes-Hoitinga et al.. 2004) was selected as the POD for
derivation of the subchronic p-RfD.
The subchronic p-RfD for extra-dietary exposure to Na/K salts of inorganic phosphates is
derived by applying a composite uncertainty factor (UFc) of 30 (reflecting an interspecies
uncertainty factor [UFa] of 3, an intraspecies uncertainty factor [UFh] of 3, and a
LOAEL-to-NOAEL uncertainty factor [UFl] of 3) to the selected POD (HED) of 120 mg
P/kg-day.
Subchronic p-RfD = POD (HED) UFc
= 120 mg P/kg-day -^30
= 4 mg P/kg-day as sodium or potassium
salts of inorganic phosphates
Table 17 summarizes the uncertainty factors for the subchronic p-RfD for Na/K salts of
inorganic phosphates. Uncertainty factors were applied in accordance with applicable guidance
and methodology (U.S. EPA. 201 1 b. 2002).
102
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 17. Uncertainty Factors for the Subchronic p-RfD for Na/K Salts of
Inorganic Phosphates
UF
Value
Justification
UFa
3
A UFa of 3 (10°5) is applied to account for uncertainty in characterizing the toxicokinetic or
toxicodynamic differences between rats and humans following Na/K phosphate exposure. The
toxicokinetic uncertainty has been accounted for by calculating an HED through application of a DAF
as outlined in the U.S. EPA's Recommended Use of Body Weight4 as the Default Method in
Derivation of the Oral Reference Dose ('U.S. EPA. 201 lb).
UFd
1
A UFr> of 1 is applied in accordance with U.S. EPA (2002). because reproductive and developmental
toxicity studies in rats, rabbits, mice, and hamsters are available (see Table 11) and show no
indication of effects at doses below those in subchronic and chronic studies. Additionally, numerous
short-term, subchronic, and chronic oral studies in several species are available that investigated a
variety of health effects. The evidence base also comprises numerous epidemiological studies
including controlled exposure studies, cohort and case-control dietary intake studies, and randomized
studies of OSP use for bowel preparation or constipation that report health effects.
UFh
3
A UFh of 3 (10°5) is applied to account for low human variability in susceptibility to P, due to tightly
regulated homeostatic mechanisms in average individuals. Populations with pre-existing kidney
disease may be more susceptible. CKD is known to increase susceptibility to inorganic phosphate
toxicity, and its prevalence in the United States is around 14% (NIDDK. 2016). However, the
differences in susceptibility are modest and are captured in the available human data. For instance, the
prospective studv bv Yooti et al. (2017) found no effect of increased P intake on incidence of CKD in
nondiabetic adults followed for 8 yr but did see an increased incidence in CKD in the highest P
exposed quartile of diabetics. Support for using a UFH of 3 is provided by the fact that the p-RfD
based on Ritskes-Hoitinea et al. (2004) is more protective than a p-RfD derived based on kidney
effects observed in the susceptible (diabetic) population in Yooti et al. (2017).
UFl
3
A UFl of 3 (10°5) is applied because the POD is a LOAEL. The difference between LOAEL and
referent group dose in the study was -twofold, and the incidence of nephrocalcinosis in the referent
group was not significantly different than the next lower dose (indicating the referent group
approximates a NOAEL). Therefore, a full 10-fold uncertainty factor is not considered appropriate.
UFS
1
A UFS of 1 is applied because the subchronic POD is from a subchronic (8-wk) study.
UFC
30
Composite UF = UFA x UFD x UFH x UFL x UFS.
CKD = chronic kidney disease; DAF = dosimetric adjustment factor; HED = human equivalent dose;
LOAEL = lowest-observed-adverse-effect level; Na/K = sodium and potassium;
NOAEL = no-observed-adverse-effect level; OSP = oral sodium phosphate; P = phosphorus; POD = point of
departure; p-RfD = provisional reference dose; UF = uncertainty factor; UFA = interspecies uncertainty factor;
UFC = composite uncertainty factor; UFD = database uncertainty factor; UFH = intraspecies uncertainty factor;
UFl = LOAEL-to-NOAEL uncertainty factor; UFS = subchronic-to-chronic uncertainty factor;
U.S. EPA = U.S. Environmental Protection Agency.
Confidence in the subchronic p-RfD for Na/K salts of inorganic phosphates is medium, as
described in Table 18.
103
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 18. Confidence Descriptors for the Subchronic p-RfD for Na/K Salts
of Inorganic Phosphates
Confidence
Categories
Designation
Discussion
Confidence in
study
M
Confidence in the orincioal studv is medium. Ritskes-Hoitinga et al. (2004)
exposed small groups (eight per dose) of male rabbits for 8 wk to diets
containing four different P concentrations, but only two contained at least the
minimum level of P (0.4%) recommended for rabbits (Clarke et al.. 1977).
However, the study used a well-characterized diet and controlled for dietary Ca.
In addition, the renal endpoint (nephrocalcinosis) and POD are supported by a
similar studv in rats (Ritskes-Hoitinga et al.. 1989). Endooints tested in these
studies, although limited compared with a comprehensive subchronic toxicity
study, included sensitive endpoints known or believed to be associated with
phosphate intake. Endpoints included body weight/growth and feed intake, urine
and fecal mineral content, weight and histology of the kidney, and femur
dimensions and mineral content.
Confidence in
database
M
Confidence in the database is medium. The database comprises numerous
epidemiological studies including controlled exposure studies, cohort and case-
control dietary intake studies, and randomized studies of OSP use for bowel
preparation or constipation that report a variety of health effects. Although the
database includes many human studies, these suffer from several limitations,
including uncertainty in doses and relevance of dietary P (i.e., organic and
inorganic sources of P). The database also contains short-term, subchronic, and
chronic oral studies in rats, rabbits, and dogs that report a variety of health
effects. Although animal data are extensive, many studies are limited by a lack
of information on concurrent Ca intake, a key determinant of phosphate toxicity.
Reproductive and developmental toxicity screening studies in rats, rabbits, mice,
and hamsters are also available (see Table 11), although many studies are
technical reports or reported only in secondary publications.
Confidence in
subchronic p-RfD
M
Overall confidence in the subchronic p-RfD is medium.
Ca = calcium; L = low; M = medium; Na/K = sodium and potassium; OSP = oral sodium phosphate;
P = phosphorus; p-RfD = provisional reference dose.
3.1.2. Derivation of the Chronic Provisional Reference Dose
As shown in Table 3A, none of the available chronic animal studies of Na/K salts of
inorganic phosphates met selection criteria. Thus, the chronic p-RfD for extra-dietary exposure
to Na/K salts of inorganic phosphates is derived from the same POD as the subchronic p-RfD—
the LOAEL (HED) of 120 mg P/kg-day for increased incidence of nephrocalcinosis in rabbits
exposed to NaH:P04 dihydrate for 8 weeks via the diet (Ritskes-Hoitinga et al.. 2004)—by
applying a UFc of 100 (reflecting a UFa of 3, a UFh of 3, a UFl of 3 for using a LOAEL, and a
subchronic-to-chronic extrapolation uncertainty factor [UFs] of 3 for use of a subchronic
LOAEL as a POD).
104 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Chronic p-RfD = POD (HED) - UFC
= 120 mg P/kg-day -MOO
= 1 mg P/kg-day as sodium or potassium
salts of inorganic phosphates
Table 19 summarizes the uncertainty factors for the chronic p-RfD for Na/K salts of
inorganic phosphates. Uncertainty factors were applied in accordance with applicable guidance
and methodology (U.S. EPA. 2011b. 2002).
105
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 19. Uncertainty Factors for the Chronic p-RfD for Na/K Salts of
Inorganic Phosphates
UF
Value
Justification
UFa
3
A UFa of 3 (10°5) is applied to account for uncertainty in characterizing the toxicokinetic or
toxicodynamic differences between rats and humans following Na/K phosphate exposure. The
toxicokinetic uncertainty has been accounted for by calculating an HED through application of a DAF
as outlined in the U.S. EPA's Recommended Use of Body Weight4 as the Default Method in
Derivation of the Oral Reference Dose ('U.S. EPA. 201 lb).
UFd
1
A UFd of 1 is applied as reproductive and developmental toxicity studies in rats, rabbits, mice, and
hamsters are available (see Table 11), including a three-generation reproductive toxicity study, which
do not signal concern for effects on fertility or development at levels below the POD for kidney effects
in the principal study. The evidence base also contains short-term, subchronic, and chronic oral studies
in rats and rabbits that investigate a variety of health effects. Regarding human data, numerous
epidemiological studies have been conducted that investigate a variety of health effects, although
methodological concerns limit their use for estimating effect levels for Na/K inorganic phosphates.
UFh
3
A UFh of 3 (10°5) is applied to account for low human variability in susceptibility to increased P, due
to tightly regulated homeostatic mechanisms in average individuals. Populations with pre-existing
kidney disease may be more susceptible. CKD is known to increase susceptibility to inorganic
phosphate toxicity, and its prevalence in the United States is around 14% (NIDDK. 2016). However,
the differences in susceptibility are modest and are captured in the available human data. For instance,
the prospective studv bv Yooti et al. (2017) found no effect of increased P intake on incidence of CKD
in nondiabetic adults followed for 8 yr but did see an increased incidence in CKD in the highest P
exposed quartile of diabetics. Support for using a UFH of 3 is provided by the fact that the p-RfD
based on Ritskes-Hoitinea et al. (2004) is more protective than a p-RfD derived based on kidnev
effects observed in the susceptible (diabetic) population in Yooti et al. (2017).
UFl
3
A UFl of 3 (10°5) is applied because the POD is a LOAEL. The difference between LOAEL and
referent group dose in the key study was -twofold, and the incidence of nephrocalcinosis in the
referent group was not significantly higher than the next lower dose (indicating the referent group
approximates a NOAEL). Therefore, a full 10-fold UF was not considered appropriate.
UFS
3
A UFS of 3 (100 5) was applied to address the uncertainty in extrapolating from data obtained in a study
with less-than-lifetime exposure to lifetime exposure. The POD was derived from an 8-wk study in
rabbits (Ritskes-Hoitinea et al.. 2004). which observed increased Ca deposits in the kidnev. It is
possible that longer exposure durations could result in increased severity of kidney effects or effects
seen at lower doses. However, a full uncertainty factor of 10 was not considered appropriate due to the
narrow dose range between the POD for kidney toxicity (0.88% P) and the recommended (lower
bound) P intake in rabbits to meet nutritional needs (0.22% P)(NRC. 1977).
UFC
100
Composite UF = UFA x UFD x UFH x UFL x UFS.
Ca = calcium; CKD = chronic kidney disease; DAF = dosimetric adjustment factor; HED = human equivalent dose;
LOAEL = lowest-observed-adverse-effect level; Na/K = sodium and potassium;
NOAEL = no-observed-adverse-effect level; POD = point of departure; p-RfD = provisional reference dose;
UF = uncertainty factor; UFA = interspecies uncertainty factor; UFC = composite uncertainty factor; UFD = database
uncertainty factor; UFH = intraspecies uncertainty factor; UFL = LOAEL-to-NOAEL uncertainty factor;
UFS = subchronic-to-chronic uncertainty factor; U.S. EPA= U.S. Environmental Protection Agency.
Confidence in the chronic p-RfD for the Na/K salts of inorganic phosphates is medium,
as described in Table 20.
106 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 20. Confidence Descriptors for the Chronic p-RfD for Na/K Salts of
Inorganic Phosphates
Confidence
Categories
Designation
Discussion
Confidence in
study
M
Confidence in the orincioal studies is medium. Ritskes-Hoitinea et al. (2004)
exposed small groups (eighty per dose) of male rabbits to diets containing four
different P concentrations, but only two contained at least the minimum level of P
(0.4%) recommended for rabbits (Clarke et al.. 1977). However, the studv used a
well-characterized diet and controlled for dietary Ca. In addition, the renal endpoint
(neohrocalcinosis) and POD is suDDorted bv a similar studv in rats bv Ritskes-
Hoitinea et al. (1989). Endooints tested in the studies, althoueh limited compared
with a comprehensive toxicity study, included sensitive endpoints known or
believed to be associated with phosphate intake. Endpoints included body
weight/growth and feed intake, urine and fecal mineral content, weight and
histology of the kidney, and femur dimensions and mineral content.
Confidence in
database
M
Confidence in the database is medium. The database comprises numerous
epidemiological studies including controlled exposure studies, cohort and case-
control dietary intake studies, and randomized studies of OSP use for bowel
preparation or constipation that report a variety of health effects. Although the
database includes many human studies, these suffer from several limitations,
including uncertainty in doses and relevance of dietary P (i.e., organic or inorganic
P). The database also contains short-term, subchronic, and chronic oral studies in
rats, rabbits, and dogs that report a variety of health effects. Although animal data
are extensive, many studies are limited by a lack of information on concurrent Ca
intake, a key determinant of phosphate toxicity. Reproductive and developmental
toxicity screening studies in rats, rabbits, mice, and hamsters are also available
(see Table 11), although many studies are technical reports or reported only in
secondary publications.
Confidence in
chronic p-RfD
M
Overall confidence in the chronic p-RfD is medium.
Ca = calcium; L = low; M = medium; Na/K = sodium and potassium; OSP = oral sodium phosphate;
P = phosphorus; p-RfD = provisional reference dose.
3.1.3. Consideration of Human Data
As discussed above, information from human studies and FDA warnings, while
inadequate for dose-response assessment, provides clear not-to-exceed dose estimates that may
be used to assess whether the derived p-RfDs will be adequately protective for effects observed
after acute and short-term durations. In addition, the relevance of the Recommended Daily Intake
(RDI) of P, typical intake in the United States, and human subpopulations known to be more
susceptible to phosphate toxicity warrant discussion.
Table 21 provides a comparison between the subchronic and chronic p-RfDs for Na/K
salts of inorganic phosphates and effect levels identified in 1-day and short-term exposure
studies in humans exposed to sodium or potassium phosphates. The table shows the study,
exposure duration, estimated LOAEL, and margins of safety calculated as the ratios of the
LOAEL to the subchronic p-RfD. As the table demonstrates, the p-RfD values derived from
animal toxicity data provide adequate protection against effects seen in humans after acute or
short-term exposure.
107 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 21. Comparison between Subchronic p-RfD and LOAELs from Acute
and Short-Term Human Exposures to Na/K Salts of Inorganic Phosphates
Reference and Study Type
Exposure
Duration
Endpoint
LOAEL
(mg P/kg-d as
inorganic
phosphate)
Margin of Safety
(human
LOAEL/subchronic
p-RfD)
Nishi et al. (2015);
controlled exposure
Id
Transient decrease in
vascular endothelial function
compared with pre-exposure
levels
26.49
7
FDA (2014); warnine re:
constipation therapy
1-3 d
Laxative effect, with severe
effects occurring at higher
doses
40-105 (not
including
dietary intake)
>10
Grimm et al. (2001);
controlled exposure
4-6 wk
GI disturbances
50.13
13
Colonoscopy preparation3:
Hurst et al. (2007)
prospective cohort study,
FDA (2008) warnine. and
numerous other studies and
case reports
Id
Increased risk of acute renal
failure/acute phosphate
nephropathy; GI distress
(nausea, diarrhea, bloating,
vomiting) during bowel
preparation for colonoscopy
164
41
Standard dose is administered in <24 hours (usually 10-12 hours apart). Dose is assumed to reflect all P sources
during the day of preparation, but small amounts of P may also be consumed if allowed foods (e.g., gelatin
desserts) contain P.
GI = gastrointestinal; LOAEL = lowest-observed-adverse-effect level; Na/K = sodium and potassium;
P = phosphorus; p-RfD = provisional reference dose.
FDA (2014) and FDA (2008) issued health warnings pertaining to the use of OSP
preparations for bowel cleansing and treatment of constipation. The 2008 warning pertained to
acute kidney injury associated with bowel cleansing, and this effect and its LOAEL are shown in
Table 21 with corresponding margins of safety. The 2014 FDA warning pertained to severe
dehydration, electrolyte imbalances, and serious effects on the heart and kidneys, sometimes
leading to death in individuals (including children) who exceed recommended daily doses of
sodium phosphate preparations for constipation therapy. As discussed further in Section 2.1, the
therapeutic recommendations correspond to P intakes of-40-100 mg P/kg-day (for up to 3 days)
as sodium phosphate for children >5 years of age and adults. The derived p-RfD values are well
below these intake levels. FDA (2014) recommended against any use of sodium phosphate
colonoscopy therapeutics for infants and children under 5 years of age unless under physician
supervision; further discussion of these and other susceptible populations is presented after the
discussion of nutritional requirements below.
Phosphorus is an essential nutrient that exhibits a U-shaped dose-response curve: doses
below physiological requirements may lead to deleterious effects, as can doses that exceed
physiological requirements. The RDI value for P is 700 mg P/day or 10 mg P/kg-day for a 70-kg
adult (HFSA. 2015; IOM. 1997). Note however, that the RDI includes P from less bioavailable
and organic plant and animal sources in addition to other poorly absorbed inorganic phosphate
108 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
sources. Furthermore, dietary intake of P in the U.S. population is more than adequate; based on
NHANES data between 2001 and 2014, McClure et al. (2017) estimated mean dietary P intake
over the entire period to be 1,373 mg P/day (range of means by year was 1,324-1,414 mg
P/day), or -20 mg P/kg-day for a 70-kg adult. Although little information regarding the
proportion of total dietary P load from inorganic phosphate additives (the source most relevant to
this assessment) was located, one source estimated the contribution to be -500 mg P/day (Calvo
et al.. 2013 as cited in Trautvettcr et al.. 2018). Given that the RDI of P includes both organic
and inorganic sources, and extra-dietary environmental exposure to (highly bioavailable) Na/K
salts of inorganic phosphate will increase P intake over and above a dietary intake that is more
than adequate, the RDI is not considered to be a lower bound on the p-RfD for the Na/K salts of
inorganic phosphates.
A significant proportion of the U.S. population may exhibit conditions or characteristics
that increase their susceptibility to phosphate toxicity. For example, CKD impairs the body's
ability to excrete excess P, leading to hyperphosphatemia. As a result, people with CKD may be
prescribed phosphate binders or instructed to reduce their intake of dietary P below the values
recommended for healthy persons. Approximately 14% of the U.S. population has CKD
(NIDDK. 2016). Populations on dialysis are particularly susceptible to increases in P, with one
study indicating disruption of P homeostasis and activation of markers of bone resorption in
patients with increased dietary intake of 100 mg/d (Tsai et al.. 2021). Additional characteristics
that may increase susceptibility to phosphate toxicity include female sex, low body mass index
(BMI), dehydration, use of ACEIs or diuretics, and older age (FDA. 2014). Finally, as noted
above, FDA recommends against the use of sodium phosphate compounds to treat constipation
in infants and children under 5 years of age, unless under physician supervision, due to their
enhanced susceptibility.
In deriving the p-RfDs presented herein, every effort was made to ensure that the
resulting values would provide adequate protection against phosphate toxicity while working
within the limitations of available data and U.S. EPA methodologies. However, Na/K salts of
inorganic phosphates present unique challenges in toxicity value derivation, among them the
large number of potentially susceptible individuals and high background intake of both P and
inorganic phosphate additives in the U.S. population.
3.2. DERIVATION OF PROVISIONAL REFERENCE CONCENTRATIONS
3.2.1. Derivation of the Subchronic Provisional Reference Concentration
Relevant data with which to derive a subchronic provisional reference concentration
(p-RfC) for Na/K salts of inorganic phosphates were not identified in the available literature. As
noted earlier, the studies of workers involved in phosphate mining and fertilizer production (Yiin
et al .. 2016; Kheliti et al .. 2014) are confounded by coexposure to a host of other hazardous and
radioactive substances.
3.2.2. Derivation of the Chronic Provisional Reference Concentration
For phosphoric acid, a chronic inhalation reference concentration (RfC) of 0.01 mg/m3 is
available in U.S. EPA's IRIS database (U.S. EPA. 1995). This RfC is based on a subchronic
study (Aranvi et al .. 1988) of rats exposed to an aerosol of combustion products from burning
95% red phosphorus and 5% butyl rubber. Uncertainty with respect to the toxicology of the
exposure mixture precludes using that study as the basis for a p-RfC. No relevant human or
109 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
animal studies of inorganic phosphate inhalation published since the 1995 assessment were
identified in the literature searches or secondary sources reviewed.
3.3. SUMMARY OF NONCANCER PROVISIONAL REFERENCE VALUES
A summary of the noncancer provisional reference values is shown in Table 22.
Table 22. Summary of Noncancer Risk Estimates for Na/K Salts of
Inorganic Phosphates (Multiple CASRNs)
Toxicity Type
(units)
Species/
Sex
Critical Effect
p-Reference
Value
POD
Method
POD
(HED)
UFc
Principal Study
Subchronic p-RfD
(mg P/kg-d as Na/K salts
of inorganic phosphates)
Rabbit/M;
Nephro-
calcinosis
4 x 10°
LOAEL
120
30
Ritskes-Hoitinea
et al. (2004):
Chronic p-RfD
(mg P/kg-d as Na/K salts
of inorganic phosphates)
Rabbit/M;
Nephro-
calcinosis
1 x 10°
LOAEL
120
100
Ritskes-Hoitinea
et al. (2004):
Subchronic p-RfC
(mg P/m3)
NDr
Chronic p-RfC
(mg P/m3)
An RfC for phosphoric acid is available on IRIS (U.S. EPA. 1995).a
aBased on bronchiolar fibrosis in rats exposed by inhalation for 13 weeks.
HED = human equivalent dose; IRIS = Integrated Risk Information System;
LOAEL = lowest-observed-adverse-effect level; M = male; Na/K = sodium and potassium; NDr = not determined;
P = phosphorus; p-RfC = provisional inhalation reference concentration; p-RfD = provisional oral reference dose;
POD = point of departure; UFC = composite uncertainty factor.
3.4. CANCER WEIGHT-OF-EVIDENCE DESCRIPTOR
Human data pertinent to the assessment of Na/K salts of inorganic phosphate
carcinogenicity are limited to cohort and case-control studies of dietary P intake (see Tables 9
and 10). These studies suffer from many of the same limitations noted in dietary intake studies of
noncancer endpoints, including variability in the bioavailability of different P sources and lack of
data on intake for subcategories of dietary P that may be relevant to the assessment of Na/K salts
of inorganic phosphates. A well-conducted, large prospective cohort study (Wilson et al.. 2015)
observed increased risk of prostate cancer, especially lethal and high-grade cancers, with higher
dietary intake of P, after controlling for Ca, dairy, and animal protein intake. Three other dietary
intake cohort studies and a case-control study examined associations with prostate cancer but did
not find any significant associations. All three negative cohort studies were limited by very brief
follow-up times ( -8 years on average) compared with Wilson et al. (2015) (24 years).
No studies examined cancer endpoints in animals exposed to sodium or potassium salts
of inorganic phosphates by oral or inhalation routes. Available genotoxicity data indicate that
sodium or potassium salts of inorganic phosphates are not mutagenic and do not induce CAs or
DNA damage in vitro, nor mutations in vivo (see Section 2.3.1). Wilson et al. (2015) suggested
that P could influence prostate cancer progression via increases in PTH that increase bone
110 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
remodeling, leading to an increase in the metastasis of prostate tumors to bone, but confirmatory
or supporting data were not identified in the literature.
Although the study by Wilson et al. (2015) addressed several limitations of previous
human studies, its relevance to the carcinogenicity of Na/K salts of inorganic phosphates remains
uncertain, as the proportion of the dietary intake corresponding to inorganic phosphates relevant
to the compounds under assessment is not known. In addition, the increases in RR of prostate
cancer observed by Wilson et al. (2015) were small; adjusted RRs were <1.13 for all prostate
cancers and <1.51 for lethal and high-grade cancers. Finally, animal and mechanistic data
supporting an association with prostate cancer are lacking. Thus, the available data are not
considered adequate to assess the carcinogenicity of Na/K salts of inorganic phosphates.
The cancer weight-of-evidence (WOE) descriptor for Na/K salts of inorganic phosphates
is shown in Table 23.
Table 23. Cancer WOE Descriptor for Na/K Salts of Inorganic Phosphates
Possible WOE Descriptor
Designation
Route of Entry
(oral, inhalation,
or both)
Comments
"Carcinogenic to Humans "
NS
NA
The available data do not support this
descriptor.
"Likely to Be Carcinogenic to
Humans "
NS
NA
The available data do not support this
descriptor.
"Suggestive Evidence of
Carcinogenic Potential"
NS
NA
The available data do not support this
descriptor.
"Inadequate Information to
Assess Carcinogenic Potential"
Selected
Both
Existing information is inadequate to
evaluate the carcinogenicity of Na/K salts
of inorganic phosphate compounds in
humans or animals.
"Not Likely to Be Carcinogenic to
Humans "
NS
NA
The available data do not support this
descriptor.
NA = not applicable; Na/K = sodium and potassium; NS = not selected; WOE = weight of evidence.
3.5. DERIVATION OF PROVISIONAL CANCER RISK ESTIMATES
Derivation of quantitative estimates of cancer risk for Na/K salts of inorganic phosphates
is precluded by the lack of data demonstrating carcinogenicity associated with exposure
(see Table 24).
Ill Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table 24. Summary of Cancer Risk Estimates for Na/K Salts of Inorganic
Phosphates (Multiple CASRNs)
Toxicity Type (units)
Species/Sex
Tumor Type Cancer Risk Estimate Principal Study
p-OSF (mg/kg-d 1
NDr
p-IUR (mg/m3) 1
NDr
Na/K = sodium and potassium; NDr = not determined; p-IUR = provisional inhalation unit risk;
p-OSF = provisional oral slope factor.
112
Na/K Salts of Inorganic Phosphates
-------
EPA 690 11-23 OOJF
APPENDIX A. LITERATURE SCREENING RESULTS
¦"Treatment of hypophosphatemia, effects of phosphate restriction, cotreatment with other agent, pathological hyperphosphatemia, no comparison group, inappropriate route, performance enhancement.
"""Includes studies discussed in text, in addition to those tabulated and formally evaluated. Not all included studies are cited.
Figure A-l. Literature Screening Results: Human Studies
113
Na/K Salts of Inorganic Phosphates
-------
EPA 690 11-23 OOJF
*No animal studies using inhalation exposure were identified.
"""Includes studies discussed in text in addition to those tabulated and formally evaluated. All included studies are cited.
Figure A-2. Literature Screening Results: Animal Studies
114
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
APPENDIX B. DATA TABLES
Table B-l. Significant Effects in Female Weanling Wistar Rats
Administered Inorganic Phosphate (as Ca[H2P04]2 Alone or with NaH2P04)
in the Diet for 6 Weeks"
Effects (units)
Dose in mg P/kg-d (% P in diet)
530
(0.5% P in diet; referent)
1,100
(1% P in diet)
Kidney Ca concentration (|ig/g)
460 ± 167b
3,350 ± 470* (+628%)°
Kidney wet weight (g)
1.87 ±0.04
2.16 ±0.11* (+16%)
aHitchman et al. (1979).
bMean ± SE.
°Value in parentheses is percent change relative to control = [(treatment mean - control mean) + control
mean] x 100.
* Significantly different from referent group at p< 0.05, based on two-sided /-test performed for this review.
Ca = calcium; Ca[H2P04]2 = Calcium phosphate; NaH2P04 = monosodium phosphate; P = phosphorus;
SE = standard error.
Table B-2. Significant Effects in Male Wistar Rats Administered Inorganic
Phosphate (as CaHP04 and KH2PO4) in the Diet for 8 Weeks"
Effects (units)
Dose in mg P/kg-d (% P in diet)
510
(0.6% P in diet; referent)
980
(1.2% P in diet)
1,400
(1.8% P in diet)
Serum PTH (units not specified)
300 ±166b
583 ± 272 (+94%)°
2,190 ±821** (+630%)°
Femur length (cm)
3.7 ±0.14
3.6 ±0.19 (-3%)
3.4 ±0.15** (-8%)
Femur bmc; final (g)
0.35 ±0.03
0.33 ± 0.03 (-6%)
0.28 ±0.03** (-20%)
Femur bmd; final (g/cm3)
0.23 ± 0.02
0.21 ±0.02** (-9%)
0.19 ±0.02** (-17%)
Femur; distal metaphysis
Total bone bmc (g)
11.3 ±0.90
9.6 ± 0.92** (-15%)
9.8 ± 1.12** (-13%)
Total bone CSA (mm2)
19.9 ±2.3
17.9 ± 1.3* (-10%)
17.3 ± 2.4** (-13%)
Cortical bmc (g)
8.04 ± 0.46
7.34 ± 0.64* (-9%)
7.03 ± 0.76** (-13%)
Cortical CSA (mm2)
7.7 ±0.3
7.1 ±0.2** (-8%)
7.6 ±0.3* (-1%)
Femur; midshaft
Total bone bmc (g)
8.5 ±0.67
7.9 ± 0.54* (-7%)
7.0 ± 0.69** (-18%)
Total bone CSA (mm2)
10.8 ± 1.3
9.8 ± 0.8* (-9%)
9.4 ± 0.7** (-13%)
Cortical bmd (g/cm3)
1,289.2 ± 12.5
1,291.0 ± 15.6 (+0%)
1,250.1 ±19.2** (-3%)
Cortical bmc (g)
7.74 ±0.56
7.18 ±0.49* (-7%)
6.32 ± 0.69** (-18%)
Cortical CSA (mm2)
6.0 ±0.5
5.6 ± 0.4* (-7%)
5.0 ± 0.5** (-17%)
Cortical thickness (mm)
0.72 ±0.43
0.70 ± 0.35 (-3%)
0.65 ± 0.46** (-10%)
115 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-2. Significant Effects in Male Wistar Rats Administered Inorganic
Phosphate (as CaHP04 and KH2PO4) in the Diet for 8 Weeks"
Effects (units)
Dose in mg P/kg-d (% P in diet)
510
(0.6% P in diet; referent)
980
(1.2% P in diet)
1,400
(1.8% P in diet)
Femur; mechanical competence
Ultimate strength (N)
117 ±8.6
104 ± 7.4 (-10%)
88.3 ± 18** (-24%)
Yield point
100 ± 9.2
80 ± 20 (-20%)
64 ± 25** (-36%)
Tibia; trabecular bone
Bone volume (mm3)
2.5±0.3f
1.2 ± 0.2 (-52%)
1.7 ±0.4 (-32%)
Bone surface (mm2)
141 ± 8. If
75 ± 14 (-47%)
97 ± 18 (-31%)
Structure model index
2.1 ±0.10f
2.1 ±0.03 (+0%)
1.8 ±0.05 (-14%)
Connective structures per
unit volume (mm-3)
46.4 ± 14
37.9 ± 0.79 (-18%)
69.9 ±2.28 (+51%)
Tibia; cortical bone
Total cross-sectional area
(mm2)
5.3 ±0.24}
4.5 ±0.14 (-15%)
4.0 ± 0.06 (-25%)
Tibia; mechanical competence
Ultimate strength (N)
118 ± 18}
104 ± 8.7 (-12%)
83 ± 14 (-30%)
Stiffness (N/mm)
332 ±51}
295 ±23 (-11%)
251 ± 49 (-24%)
Toughness (N-M x 10 3)
103 ± 17}
71 ± 18 (-32%)
62 ± 23 (-40%)
Yield point
74 ± 19f
81 ± 18 (+9%)
56 ± 20 (-24%)
aHuttunen et al. (2007).
bMean± SD.
°Value in parentheses is percent change relative to control = [(treatment mean - control mean) + control
mean] x 100.
* Significantly different from the referent group at p< 0.05, based on pairwise analyses performed by the study
authors.
**p < 0.01.
f Significant difference across all groups by ANOVA at p< 0.05, as reported by the study authors.
J/? <0.01.
ANOVA = analysis of variance; bmc = bone mineral content; CaHP04 = calcium phosphate; CSA = cross-sectional
area; KH2PO4 = monopotassium phosphate; N = newtons; N-M = newton meters (units of toughness);
P = phosphorus; PTH = parathyroid hormone; SD = standard deviation.
116
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-3. Significant Effects in Female Wistar Rats Administered
Inorganic Phosphate (as KH2PO4) in the Diet for 6 Weeks"
Effects (units)
Dose in mg P/kg-d (% P in diet)
270 (0.5% P in diet; referent)
800 (1.5% P in diet)
Body-weight gain (g)
32.27 ± 2.69b
28.65 ±2.88 (-11%)C
Ca absorption (mg/d)
14.29 ± 1.52
5.57 ±3.23* (-61%)
P balance
Absorption (mg/d)
38.78 ± 1.30
148.40 ±5.41* (+283%)
Urinary excretion (mg/d)
29.36 ±5.83
115.59 ±6.86* (+294%)
Serum osteocalcin (ng/mL)
20.05 ± 1.50
27.88 ± 2.53* (+39%)
Urinary deoxypyridinoline
(nmol/mmol creatinine)
70.76 ± 5.44
89.60 ±7.16* (+27%)
Fifth lumbar bmd (mg/cm2)
78.02 ± 0.77
71.87 ±0.88* (-8%)
aKoshihara et al. (2005).
bMean ± SEM for n = five per group.
°Value in parentheses is percent change relative to control = [(treatment mean - control mean) + control
mean] x 100.
* Significantly different from referent group at p< 0.05, based on analyses performed by the study authors.
bmd = bone mineral density; Ca = calcium; KH2PO4 = monopotassium phosphate; P = phosphorus;
SEM = standard error of the mean.
117
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-4. Effects in Female Wistar RIVrTOX Rats Administered Inorganic
Phosphate in the Diet (as NaH2P04 Dihydrate) for 28 Days (Experiment l)a
Effects (units)
Dose in mg P/kg-d (% P in diet)
390
(0.41% P in diet; referent)
580
(0.60% P in diet)
Body weight (g)
89.4 ± 10.4b
90.7 ± 8.4 (+1%)C
Urine pH
Study Days 0-2
9.2 ±0.1
8.5 ± 0.8** (-8%)
Study Days 13-15
8.3 ±0.7
7.5 ± 0.6* (-10%)
Study Days 26-28
8.8 ±0.6
7.3 ±0.3** (-17%)
Urine albumin (mg/d)
Study Days 13-15
0.52 ±0.17
1.45 ±0.71** (+179%)
Study Days 26-28
0.56 ±0.29
0.99 ± 0.40* (+77%)
Urine urea (mg/100 g body weight x d)
Study Days 13-15
157.8 ±24.6
185.3 ± 33.9* (+17%)
Study Days 26-28
177.7 ±21.8
189.4 ± 28.5 (+7%)
Relative kidney weight (g/100 g body weight)
0.41 ±0.04
0.52 ± 0.08** (+27%)
Kidney P (%)
1.5 ±0.2
3.7 ± 1.3** (+2%)
Kidney Ca (%)
0.4 ±0.2
5.7 ±3.0** (+5%)
Kidney Mg (%)
0.10 ±0.00
0.23 ±0.07** (+0.1%)
Nephrocalcinosis
2/6d (33%)
16/16 (100%)
Mean severity
0.5
2.7f
aRitskes-Hoitinga et al. (1989).
bMean± SD.
°Value in parentheses is percent change relative to control = [(treatment mean - control mean) + control
mean] x 100, or for data presented as percent, it is percent difference = (treatment mean % - control mean %).
dNumber affected/number examined (% incidence).
* Significantly different from referent group at p< 0.05, based on analyses performed by the study authors.
**p < 0.01.
fThe distribution of histological scores was significantly different from that of the reference group at p< 0.05,
based on analyses (Mann-Whitney U test) performed by the study authors.
Ca = calcium; Mg = magnesium; NaH2P04 = monosodium phosphate; P = phosphorus; SD = standard deviation.
118
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-5. Effects in Female Wistar RIVrTOX Rats Administered Inorganic
Phosphate in the Diet (as NaH2P04 Dihydrate) for 28 Days (Experiment 2)a
Effects (units)
Dose in mg P/kg-d (% P in diet)
410
(0.41% P in diet; referent)
580
(0.61% P in diet)
Body weight (g)
87.2 ± 4.1b
88.5 ± 6.5 (+1%)C
Urine pH
Study Days 0-2
8.9 ±0.4
8.2 ± 0.4** (-8%)
Study Days 13-15
8.6 ±0.6
7.7 ± 0.7** (-10%)
Study Days 26-28
8.5 ±0.8
7.9 ± 0.9 (-7%)
Urine albumin (mg/d)
Study Days 0-2
0.15 ±0.07
0.10 ±0.03* (-33%)
Study Days 13-15
0.22 ±0.11
0.74 ± 0.55** (+236%)
Study Days 26-28
0.35 ±0.15
0.60 ±0.33 (+71%)
Serum Ca (mg/100 mL)
10.21 ±0.14
9.63 ± 0.45* (-6%)
Serum Mg (mg/100 mL)
2.04 ±0.15
1.66 ±0.19* (-18%)
Relative kidney weight (g/100 g body weight)
Study Day 14
0.45 ±0.02
0.49 ± 0.07 (+9%)
Study Day 28
0.37 ±0.03
0.46 ±0.14 (+24%)
Kidney P (%)
Study Day 14
1.2 ±0.0
2.7 ± 1.1** (+2%)
Study Day 28
1.3 ±0.2
2.4 ± 1.1 (+1%)
Kidney Ca (%)
Study Day 14
0.1 ±0.0
3.4±1.9** (+3%)
Study Day 28
0.9 ±-0.8
3.5 + 2.1* (+3%)
Kidney Mg (%)
Study Day 14
0.09 ±0.00
0.18 + 0.09* (+0.09%)
Study Day 28
0.08 ±0.00
0.13 + 0.04* (+0.05%)
119
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-5. Effects in Female Wistar RIVrTOX Rats Administered Inorganic
Phosphate in the Diet (as NaH2P04 Dihydrate) for 28 Days (Experiment 2)a
Effects (units)
Dose in mg P/kg-d (% P in diet)
410
(0.41% P in diet; referent)
580
(0.61% P in diet)
Nephrocalcinosis
Study Day 14
5/6d (83%)
6/6d (100%)
Mean severity
1.0
2.8f
Study Day 28
5/6 (83%)
6/6 (100%)
Mean severity
1.5
2.5f
aRitskes-Hoitinga et al. (1989).
bMean± SD.
°Value in parentheses is percent change relative to control = [(treatment mean - control mean) + control
mean] x 100, or for data presented as percent, it is percent difference = (treatment mean % - control mean %).
dNumber affected/number examined (% incidence).
* Significantly different from referent group at p< 0.05, based on analyses performed by the study authors.
**p < 0.01.
fThe distribution of histological scores was significantly different from that of the reference group at p< 0.05,
based on analyses performed by the study authors.
Ca = calcium; Mg = magnesium; NaH2P04 = monosodium phosphate; P = phosphorus; SD = standard deviation.
Table B-6. Significant Effects in Male NZW Rabbits Administered
Inorganic Phosphate (as NaH2P04 Dihydrate) in the Diet for 8 Weeks"
Effects (units)
Dose in mg P/kg-d (% P in diet)
77
(0.2% P in diet)
150
(0.45% P in diet)
290
(0.85% P in diet)
Initial body weight (kg)
1.77 ±0.10
1.77 ± 0.10b,c
1.77 ± 0.10 (+0%)
Final body weight (kg)
2.42 ±0.41
2.66 ±0.27
2.41 ± 0.29 (-9%)
Feed intake (g/d)
66.9 ±9.4
73.1 ±7.6
68.9 ± 9.5 (-6%)
Urine pH Days 20-23
9.42b ± 0.16
9.36 ±0.15
8.20 ± 0.35* (-l,450%)d
Urine pH Days 48-51
9.23b ± 0.34
9.35+0.23
8.04 + 0.30* (-l,620%)d
Urinary Ca
96.8+16.7
49.6 +23.5
1.48 +0.43* (-3,350%)
Urinary P
22.4 + 5.8
73.4 + 12.6
238 +23* (+326%)
Serum P (Day 28/Day 56)
1.92 + 0.19/1.80 + 0.28
2.06+0.17/1.84 + 0.12
2.51 +0.63/2.29 + 0.57
(+17.9%/+19.7%)
Serum Ca (Day 28/Day 56)
2.97 + 0.34/3.16 + 0.09
2.81+0.29/3.21+0.41
2.55 + 0.23/2.71 +0.20
(-9.0%/-16.6%)
Serum Mg (Day 28/Day 56)
0.63 +0.09/0.64 + 0.01
0.68+0.07/0.63 +0.18
0.65 + 0.23/0.64 + 0.18
(-4.4%/-2.7%)
Kidney Ca concentration (% dry
weight)
0.07 ± 0.04
0.34 ±0.65
1.40 ± 1.51* (+412%)
120 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-6. Significant Effects in Male NZW Rabbits Administered
Inorganic Phosphate (as NaH2P04 Dihydrate) in the Diet for 8 Weeks"
Dose in mg P/kg-d (% P in diet)
Effects (units)
77
(0.2% P in diet)
150
(0.45% P in diet)
290
(0.85% P in diet)
Kidney P concentration (% dry
weight)
1.50 + 0.10
1.61+0.15
2.08 + 0.60 (+29.2%)
Absolute kidney wet weight (g)
7.34+1.41
8.71 + 1.63
8.12 + 1.17 (-6.77)
Absolute kidney dry weight (g)
1.44 + 0.19
1.60 + 0.25
1.46 + 0.16 (-9%)
Relative kidney weight (% body
weight)
0.31 + 0.05
0.33 + 0.06
0.34 + 0.05 (+3%)
Diaphysis of femur Mg concentration
% dry weight
0.40 + 0.02
0.42 + 0.01
0.48 + 0.04* (+14%)
% ash
0.52 + 0.03
0.55 + 0.02
0.64 + 0.05* (+16%)
Epiphysis of femur Mg concentration
% dry weight
0.24 + 0.02
0.23 + 0.04
0.29 + 0.03* (+26%)
% ash
0.47 + 0.02
0.47 + 0.03
0.59 + 0.07* (+26%)
mg/cm3
2.26 + 0.28
2.30 + 0.42
2.90 + 0.76* (+26%)
Incidence and severity6 of renal Ca deposits in the cortex (von Kossa stained)
0
6/8 (75%)
5/8f (63%)
0/8e (0%)f
1
2/8 (25%)
2/8 (25%)
3/8 (38%)f
2
0/8 (0%)
1/8 (13%)
4/8 (50%)f
3
0/8 (0%)
0/8 (0%)
1/8 (13%)f
Incidence and severity6 of renal Ca deposits in the cortex (hematoxylin-eosin stained)
0
6/8 (75%)
6/8f (75%)
0/8e (0%)f
1
2/8 (25%)
1/8 (13%)
3/8 (38%)f
2
0/8 (0%)
1/8 (13%)
4/8 (50%)f
3
0/8 (0%)
0/8 (0%)
1/8 (13%)f
Incidence and severity6 of renal Ca deposits in the medulla (von Kossa stained)
0
5/8 (63%)
0/8 (0 %)
0/8 (0%)
1
2/8 (25%)
1/8 (13%)
1/8 (13%)
2
1/8 (13%)
6/8 (75%)
7/8 (88%)
3
0/8 (0%)
0/8 (0%)
0/8 (0%)
121 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-6. Significant Effects in Male NZW Rabbits Administered
Inorganic Phosphate (as NaH2P04 Dihydrate) in the Diet for 8 Weeks"
Effects (units)
Dose in mg P/kg-d (% P in diet)
77
(0.2% P in diet)
150
(0.45% P in diet)
290
(0.85% P in diet)
Incidence and severity6 of renal Ca deposits in the medulla (hematoxylin-eosin stained)
0
3/8 (38%)
0/8 (0%)
0/8 (0%)
1
3/8 (38%)
3/8 (38%)
3/8 (38%)
2
2/8 (25%)
5/8 (63%)
5/8 (63%)
3
0/8 (0%)
0/8 (0%)
0/8 (0%)
aRitskes-Hoitinga et al. (2004).
bMean ± SD for n = eight per group.
0Value in parentheses is percent change relative to control = ([treatment mean - control mean] control
mean) x 100.
Significantly different from the two other groups (p < 0.05), based on analyses performed by the study authors
(Tukey B test).
"Calcification score: 0 (no calcium deposits); 1 (a few calcified deposits); 2 (multiple deposits); 3 (band of
calcification throughout entire section).
fNumber affected/number examined (% incidence).
* Significantly different from the referent group at p< 0.05, based on analyses performed by the study authors.
+The distribution of histological scores was significantly different from the reference group at p< 0.05, based on
analyses performed by the study authors.
Ca = calcium; SP = monosodium phosphate; P = phosphorus; SD = standard deviation; NZW = New Zealand White
122
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-7. Significant Effects in Male Wistar Rats Administered Inorganic
Phosphate (as KH2PO4) in the Diet for 4 Weeks"
Effects (units)
Dose in mg P/kg-d (% P in diet)
250
(0.3% P in
diet; referent)
450
(0.6% P in
diet)
670
(0.9% P in
diet)
920
(1.2% P in
diet)
1,000
(1.5% P in
diet)
Body-weight gain (g/4 wk)
95 ± 5.1b
82.3 ±0.9
(-13%)°
73.3 ±4.4**
(-23%)°
84.0 ±4.4
(-12%)°
58.7 ±6.0**
(-38%)°
Body-weight gain/100 g
intake
18.78 ± 1.24
18.21 ±0.48
(-3%)
16.21 ±0.63
(-14%)
18.01 ± 1.14
(-4%)
13.88 ± 1.3*
(-26%)
Intake (g/4 wk)
507.9 ± 13.9
453.8 ± 15.6*
(-11%)
451.1 ± 13.4*
(-11%)
467.0 ±7.1*
(-8%)
421.4 ±4.6**
(-17%)
P balance
Fecal (mg/d)
23.8 ±2.9
30.6 ±3.7
(+29%)
32.4 ±2.3*
(+36%)
45.2 ±6.2*
(+90%)
47.5 ±3.9**
(+100%)
Urinary (mg/d)
4.2 ±0.4
56.5 ±5.8**
(+1,245%)
102.1 ±9.0**
(+2,331%)
179.4 ±0.1**
(+4,171%)
274.9 ± 15.6**
(+6,445%)
Net absorption (mg/d)
28.0 ±2.2
60.2 ±4.8**
(+115%)
99.0 ±6.8**
(+254%)
137.6 ±7.1**
(+391%)
150.1 ± 12.3**
(+436%)
Balance (mg/d)
23.8 ±2.3
3.7 ±5.0**
(-4%)
-3.1 ± 12.1
(-113%)
-41.9 ±7.1**
(-276%)
-124.8 ±21.3**
(-624%)
Ca balance
Urinary (mg/d)
0.53 ±0.04
0.23 ±0.02**
(-57%)
0.17 ±0.01**
(-68%)
0.10 ±0.00**
(-81%)
0.09 ±0.02**
(-83%)
Net absorption (mg/d)
34.9 ±5.7
31.5 ± 7.1
(-10%)
28.7 ±7.4
(-18%)
23.3 ±8.5
(-33%)
4.7 ±6.6**
(-87%)
Balance (mg/d)
34.3 ±5.7
31.3 ± 7.1
(-9%)
28.5 ±7.4
(-17%)
23.2 ±8.5
(-32%)
4.7 ±6.6**
(-86%)
Serum chemistry
PTH (pg/mL)
7.82 ±3.81
33.76 ± 11.43
(+332%)
32.53 ± 19.48
(+316%)
25.06 ± 10.35
(+220%)
75.12 ±0.28**
(+861%)
l,25(OH)2D3 (pg/mL)
65.4 ±4.7
102.6 ±25.9
(+57%)
115.8 ± 12.4**
(+77%)
124.9 ±51.4
(+91%)
112.3 ± 19.0
(+72%)
aTani et al. (2007).
bMean ± SEM for n = 6 per group.
°Value in parentheses is percent change relative to control = [(treatment mean - control mean) + control
mean] x 100.
* Significantly different from the referent group at p< 0.05, based on analyses performed by the study authors.
**p < 0.01.
l,25(OH)2D3 = 1,25-dihydroxy-vitamin D3; Ca = calcium; KH2PO4 = monopotassium phosphate; P = phosphorus;
PTH = parathyroid hormone; SEM = standard error of the mean.
123 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-8. Significant Effects in Male Sprague Dawley Rats Administered
Inorganic Phosphate (as KH2PO4) in the Diet for 4 Weeks"
Parameter (units)
Dose in mg P/kg-d (% P in diet)
530 (0.6% P in diet; referent)
1,100 (1.2% P in diet)
Body weight (g)
352 ± 7.5b
345 ± 8.0 (-2%)c
NEFA (mEq/L)
0.47 ±0.03
0.34 ± 0.02* (-28%)
PTH (pg/mL)
40 ± 12.6
256.4 ±75.7* (+541%)
FGF-23 (pg/mL)
231 ±47.6
408 ± 82.6 (+77%)
BUN (mg/dL)
12.8 ±0.5
16.9 ± 1.6* (+32%)
aAbuduli et at (2016).
bMean ± SE for six animals/group.
°Value in parentheses is percent change relative to control = [(treatment mean - control mean) + control
mean] x 100.
* Significantly different from referent group at p< 0.05 based on analyses performed by the study authors.
BUN = blood urea nitrogen; FGF-23 = fibroblast growth factor-23; KH2PO4 = monopotassium phosphate;
NEFA = nonesterified fatty acids; P = phosphorus; PTH = parathyroid hormone; SE = standard error.
124
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-9. Significant Effects in Birmingham-Wistar Rats Administered
Inorganic Phosphate as Na4P207 or Na2HP04 in the Diet for 16 Weeks"
Parameter (units)
Percent in Diet
0.5%
(referent)
1% (as
Na4P207)
2.5% (as
Na4P207)
5% (as Na4P207)
5% (as
NajHPO-t)
Males
Dose (mg P/kg-d)b
280
396 (estimated)
570 (estimated)
860
840
Body-weight gain
(g)
210 ± 20°
245 ± 16
(+17%)d
229 ± 15
(+9%)d
187 ± 15
(-ll%)d
234 ± 14
(+ll%)d
Food consumption
(g)
1,309 ±28
1,359 ±27
(+4%)
1,305 ± 24
(+0%)
1,213 ±48
(-7%)
1,363 + 18
(+4%)
RBCs
(millions/mm3)
7.51 ±0.20
7.95 ±0.25
(+6%)
7.75 ±0.17
(+3%)
7.67 ±0.18
(+2%)
8.25 + 0.31
(+10%)
Hb (g/100 mL)
13.1 ±0.17
12.8 ±0.24
(-2%)
12.4 ±0.21*
(-5%)
12.5 ±0.25
(-5%)
13.9 + 0.29*
(+6%)
Specific gravity
1.06 ±0.003
1.05 ±0.005
(-1%)
1.04 ±0.006*
(-2%)
1.04 ±0.005*
(-2%)
1.05 + 0.002**
(-1%)
Relative organ weights (mg/100 g live weight)
Heart
244 ±4
255 ± 6 (+5%)
251 ± 77 (+3%)
295 ± 19* (+21%)
256 + 7 (+5%)
Stomach
378 ± 12
365 ± 15
(-3%)
388 ± 12
(+3%)
608 ±39**
(+61%)
407 + 14
(+8%)
Intestines
2,229 ± 112
2,108 ±111
(-5%)
1,964 ± 83
(-12%)
2,716 ±239
(+22%)
1,940 + 99
(-13%)
Kidneys
634 ±21
598 ± 17
(-6%)
648 ± 24
(+2%)
769 ±34*
(+21%)
741+15**
(+17%)
Testes
962 ± 42
900 ± 29
(-6%)
905 ± 42
(-6%)
1,233 ±79*
(+28%)
1,005 + 37
(+4%)
Gross pathology
Pale, pitted;
kidneys
0/10d (0%)d
0/10d (0%)d
0/10d (0%)d
6/10e,f (60%)d
6/10e,f (60%)d
Calcification;
kidneys
0/10 (0%)
0/10 (0%)
0/10 (0%)
5/10f (50%)
6/10f (60%)
Hypertrophy;
stomach
0/10 (0%)
0/10 (0%)
0/10 (0%)
4/10 (40%)
0/10 (0%)
Kidney concentration test
Males
1.0622 ±0.0026
1.0547 ±0.0047
1.0429 ±0.0061*
1.0432 + 0.0051*
1.0488 + 0.0018**
Females
Dose (mg P/kg-d)b
350
500 (estimated)
725 (estimated)
1,100
1,000
Body-weight gain
(g)
137 ±7
129 ± 6 (-6%)
129 ± 7 (-6%)
114 ±7* (-17%)
140 + 8 (+2%)
Food consumption
(g)
1,184 ±28
1,110 ±28
(-6%)
1,116 ± 16
(-6%)
1,055 ±26
(-11%)
1,168 + 23
(-1%)
RBCs
(millions/mm3)
7.30 ±0.18
7.21 ±0.15
(-1%)
7.00 ±0.19
(-4%)
6.91 ±0.53
(-5%)
7.92 + 0.13*
(+8%)
125 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-9. Significant Effects in Birmingham-Wistar Rats Administered
Inorganic Phosphate as Na4P207 or Na2HP04 in the Diet for 16 Weeks"
Parameter (units)
Percent in Diet
0.5%
(referent)
1% (as
Na4P207)
2.5% (as
Na4P207)
5% (as Na4P207)
5% (as
NajHPO-t)
Hb (g/100 mL)
12.4 ±0.51
12.0 ±0.46
(-3%)
12.9 ±0.23
(+4%)
12.1 ±0.36
(-2%)
13.2 + 0.54*
(+6%)
Specific gravity
1.06 ±0.002
1.06 ±0.003
(0%)
1.06 ±0.002
(0%)
1.04 ±0.001**
(-1%)
1.05 + 0.003*
(-1%)
Relative organ weights (mg/100 g live weight)
Heart
278 ±4
273 ± 5 (-2%)
295 ± 10 (+6%)
311 ±9* (+12%)
285 + 9 (+3%)
Stomach
451 ± 16
459 ± 15
(+2%)
506 ± 22
(+12%)
721 +63**
(+60%)
486 + 23
(+8%)
Intestines
2,521 ±73
2,928 ± 149
(+16%)
2,626 ± 85
(+4%)
3,203 + 149**
(+27%)
2,696 + 103
(+7%)
Kidneys
601 ± 12
629 ±21
(+5%)
743 ± 24**
(+24%)
893 +42**
(+49%)
838 + 35**
(+39%)
Gross pathology
Pale, pitted;
kidneys
0/10 (0%)
0/10 (0%)
2/10 (20%)
5/10f (50%)
7/10f (70%)
Calcification;
kidneys
0/10 (0%)
0/10 (0%)
2/10 (20%)
5/10f (50%)
5/10f (50%)
Hypertrophy;
stomach
0/10 (0%)
0/10 (0%)
3/10 (30%)
6/10f
(60%)
0/10
(0%)
Kidney concentration test
Females
1.0592 ±0.0024
1.0628 ±0.0026
1.0561 + 0.0015
1.0445 + 0.0014**
1.0513 +0.0025*
Combined sexes
P excretion; 3 wk
(%)
11.4 ± 0.18
NR
NR
82.5 +3.12f
(+71%)
NR
P excretion; 8 wk
(%)
NR
NR
NR
74.2 + 3.03
75.2 + 0.93
Histopathology
Renal damage
5/20 (25%)
19/20f (95%)
20/20 (100%)
20/20f (100%)
20/2 (100%)
Cortical atrophy
3/20 (15%)
12/20f (60%)
12/20f (60%)
3/20 (15%)
3/20 (15%)
Cortical hyaline
degeneration
2/20 (10%)
1 l/20f (55%)
1 l/20f (55%)
4/20 (20%)
2/20 (10%)
Medullary
calcification
0/20 (0%)
1/20 (5%)
6/20f (30%)
14/20f (70%)
15/20f (75%)
Medullary
necrosis
0/20 (0%)
1/20 (5%)
2/20 (10%)
12/20f (60%)
13/20f (65%)
Tubular casts
1/20 (5%)
0/20 (0%)
6/20 (30%)
1 l/20f (55%)
1 l/20f (55%)
Hemorrhages and
exudate
1/20 (5%)
9/20f (45%)
10/20f (50%)
12/20f (60%)
12/20f (60%)
126 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-9. Significant Effects in Birmingham-Wistar Rats Administered
Inorganic Phosphate as Na4P207 or Na2HP04 in the Diet for 16 Weeks"
Parameter (units)
Percent in Diet
0.5%
(referent)
1% (as
Na4P207)
2.5% (as
Na4P207)
5% (as Na4P207)
5% (as
NajHPO-t)
Chronic
inflammatory
changes
0/20
(0%)
0/20
(0%)
1/20
(5%)
14/20f
(70%)
14/20f
(70%)
aDatta et at (1962).
' P content of the diet was not reported by Datta et at (1962) for the 1 and 2.5% dose groups for either males or
females. However, doses can be reasonably estimated in units of mg P/kg-day for these groups. For males, the
inclusion of Na 1P2O- in the diet at 5% (860 mg P/kg-day) corresponds to an increase of 580 mg P/kg-day above the
referent dose (0.5%, 280 mg P/kg-day). Assuming linearity, this amount was scaled for the 1 and 2.5% dose groups
(116 and 290 mg P/kg-day, respectively) and added to the referent dose to generate reasonably estimated doses in
units of mg P/kg-day. Thus, the inclusion of Na4P2C>7 in the diet at 1 and 2.5% correspond to estimated doses of
396 and 570 mg P/kg-day, respectively, in males. For females, the inclusion of Na4P2C>7 in the diet at 5% (1,100 mg
P/kg-d) corresponds to an increase of 750 mg P/kg-day above the referent dose (0.5%, 350 mg P/kg-day).
Assuming linearity, this amount was scaled for the 1 and 2.5% dose groups (150 and 375 mg P/kg-day,
respectively) and added to the referent dose to generate reasonably estimated doses in units of mg P/kg-day. Thus,
the inclusion of Na4P2C>7 in the diet at 1 and 2.5% correspond to estimated doses of 500 and 725 mg P/kg-day,
respectively, in females.
°Mean ± SE.
dValue in parentheses is percent change relative to control = [(treatment mean - control mean) control
mean] x 100.
eNumber affected/number examined (% incidence).
Significantly different from referent group at p< 0.05, based on two-sided /-test (continuous data) or Fisher's exact
test (incidence data) performed for this review
* Significantly different from referent group at p< 0.05, based on analyses performed by the study authors.
**p < 0.01.
Hb = hemoglobin; Na2HP04 = disodium phosphate; Na4P2C>7 = sodium pyrophosphate; NR = not reported;
P = phosphorus; RBC = red blood cell; SE = standard error.
127 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Table B-10. Significant Effects in Male Wistar Rats Administered Inorganic
Phosphate (as K2HPO4 or [NaP03]6) in the Diet for 150 Days"
Parameter (units)
Dose in mg P/kg-d (% P in diet)
380
(0.43% P in diet as
K2HPO4 referent)
1,100
(1.3% P in diet as
K2HPO4)
400
(0.46% P in diet as
INaPO^I#; referent)
1,100
(1.2% P in diet as
[NaPOsJs)
Body-weight gain (g)
385 ± llb
432 ±9**
(+12%)°
403 ± 9b
376 ±9
(-7%)°
Relative kidney weight
(g/100 gbody weight)
0.71 ±0.03
0.73 ±0.04
(+3%)
0.68 ±0.03
0.73 ± 0.04
(+7%)
Relative testes weight
(g/100 gbody weight)
0.79 ±0.03
0.75 ±0.19
(-5%)
0.72 ±0.02
0.80 ±0.01**
(+11%)
aDvmsza et al. (1959).
bMean ± measure of variance (not specified).
°Value in parentheses is percent change relative to control = [(treatment mean - control mean) + control
mean] x 100.
* Significant difference between groups administered the same compound (high K2HPO4 vs. low K2HPO4 or high
[NaP03]6 vs. low [NaP03]6) at p< 0.05, based on analyses performed by the study authors.
**p < 0.01.
K2HPO4 = dipotassium phosphate; [NaP03]6 = sodium hexametaphosphate; P = phosphorus.
128
Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
APPENDIX C. REFERENCES
Abaskharoun. R; Depew. W; Yanner. S. (2007). Changes in renal function following
administration of oral sodium phosphate or polyethylene glycol for colon cleansing
before colonoscopy. Can J Gastroenterol 21: 227-231.
Abuduli. M; Ohminami. H; Otani. T; Kubo. H; Ueda. H; Kawai. Y; Masuda. M; Yamanaka-
Okumura. H; Sakaue. H; Yamamoto. H; Takeda. E; Taketani. Y. (2016). Effects of
dietary phosphate on glucose and lipid metabolism. Am J Physiol Endocrinol Metab 310:
E526-E538. http://dx.doi.Org/10.l 152/aipendo.00234.2015
ACGIH (American Conference of Governmental Industrial Hygienists). (2016). Phosphoric acid
[TLV/BEI], In 2016 TLVs and BEIs: Based on the documentation of the threshold limit
values for chemical substances and physical agents & biological exposure indices (pp.
49). Cincinnati, OH. https://www.acgih.org/forms/store/ProductFormPublic/2016-tlvs-
and-beis
ACGIH (American Conference of Governmental Industrial Hygienists). (2020). 2020 TLVs and
BEIs: Based on the documentation of the threshold limit values for chemical substances
and physical agents & biological exposure indices. Cincinnati, OH.
A Ion so. A; Ncttleton. JA; I\. JH; de Boer. IH; Folsom. AR; Bidulescu. A; Kestcnbaum. BR;
Chambless. LE; Jacobs. DR. Jr. (2010). Dietary phosphorus, blood pressure, and
incidence of hypertension in the atherosclerosis risk in communities study and the multi-
ethnic study of atherosclerosis. Hypertension 55: 776-784.
http://dx.doi. org/10.1161/HYPERTEN SIONAHA. 109.143461
Anderson. JJ. (2013). Potential health concerns of dietary phosphorus: cancer, obesity, and
hypertension [Review], Ann NY Acad Sci 1301: 1-8.
http://dx.doi.org/10.1 1 11/nvas. 12208
Aranvi. C; Henry. MC; Vana. SC; Gibbons. RP; Iverson. WO. (1988). Effects of multiple
intermittent inhalation exposures to red phosphorus/butyl rubber obscurant smokes in
Sprague-Dawley rats. Inhal Toxicol 1: 65-78.
Atkinson. RJ; Hunter. JO. (2002). Aphthous ulceration in the colon after sodium phosphate
bowel preparation for colonoscopy [Abstract], Gut 50: A72-A72.
Atkinson. RJ; Save. V; Hunter. JO. (2005). Colonic ulceration after sodium phosphate bowel
preparation [Letter], Am J Gastroenterol 100: 2603-2605.
http://dx.doi.org/10.1111/i. 1572-0241.2005.00305 11 x
ATSDR (Agency for Toxic Substances and Disease Registry). (2018). Minimal risk levels
(MRLs). June 2018. Atlanta, GA: Agency for Toxic Substances and Disease Registry
(ATSDR).
Belooseskv. Y; Grinblat. J; Weiss. A; Grosman. B; Gafter. U; Chagnac. A. (2003). Electrolyte
disorders following oral sodium phosphate administration for bowel cleansing in elderly
patients. Arch Intern Med 163: 803-808. http://dx.doi.org/10.1001/archinte. 163.7.803
Belsev. J. (2008). Acute phosphate nephropathy—comment on Sica et al [Letter], Am J
Gastroenterol 103: 800. http://dx.doi.Org/10.l 111/i. 1572-0241.2007.01612 2.x
Bevea. A; Block. C; Schned. A. (2007). Acute phosphate nephropathy following oral sodium
phosphate solution to cleanse the bowel for colonoscopy. Am J Kidney Dis 50: 151-154.
http://dx.doi.Org/10.1053/i.aikd.2007.03.005
Bounds. W; Skinner. J; Carruth. BR; Ziegler. P. (2005). The relationship of dietary and lifestyle
factors to bone mineral indexes in children. J Am Diet Assoc 105: 735-741.
http://dx.doi.Org/10.1016/i.iada.2005.02.046
129 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Brot. C; Jorgcnsen. N; Mad sen. OR; Jensen. LB; Sorcnscn. OH. (1999). Relationships between
bone mineral density, serum vitamin D metabolites and calcium:phosphorus intake in
healthy perimenopausal women. J Intern Med 245: 509-516.
Brown. RB; Razzaque. MS. (2015). Dysregulation of phosphate metabolism and conditions
associated with phosphate toxicity. Chemosphere 4: 705.
http://dx.doi.org/10.1038/bonekev.2Q15.74
Brown. RB; Razzaque. MS. (2018). Phosphate toxicity and tumorigenesis. Biochim Biophys
Acta Rev Cancer 1869: 303-309. http://dx.doi.Org/10.1016/i.bbcan.2018.04.007
Brunetti. SM; Feldman. HI; Latif. SM; Gupta. M; Wcincr. MG; Lewis. JD. (2009). A
comparison of sodium phosphosoda purgative to polyethylene glycol bowel preparations
prior to colonoscopy. Fam Med 41: 39-45.
Brunetti. SM; Lewis. IP; Gupta. M; Latif. SM; Wcincr. MG; Feldman. HI. (2007). Risk of
kidney injury following oral phosphosoda bowel preparations. J Am Soc Nephrol 18:
3199-3205. http://dx.doi.org/10.1681/ASN.200704044Q
CatEPA (California Environmental Protection Agency). (2008). TSD for noncancer RELs -
Appendix D.3 Chronic RELs and toxicity summaries using the previous version of the
Hot Spots Risk Assessment guidelines (OEHHA 1999). Sacramento, CA: California
Office of Environmental Health Hazard Assessment:: OEHHA.
http://oehha.ca.gov/media/downloads/crnr/appendixd3final.pdf
CalEPA (California Environmental Protection Agency). (2017). OEHHA chemical database.
Sacramento, CA: Office of Environmental Health Hazard Assessment. Retrieved from
http s://oehha.ca. gov/chemi cal s
CalEPA (California Environmental Protection Agency). (2019). Consolidated table of
OEHHA/ARB approved risk assessment health values (September 19, 2019 ed.).
Sacramento, CA: California Air Resources Board.
https://www.arb.ca.gov/toxics/healthval/contable.pdf
CalEPA (California Environmental Protection Agency). (2020). OEHHA chemical database.
Sacramento, CA: Office of Environmental Health Hazard Assessment. Retrieved from
http s://oehha.ca. gov/chemi cal s
Calvo. MS; Tucker. KL. (2013). Is phosphorus intake that exceeds dietary requirements a risk
factor in bone health? [Review], Ann N Y Acad Sci 1301: 29-35.
http://dx.doi.org/10.1 1 11/nvas. 12300
Candeal. E; Caldas. YA; Guillen. N; Levi. M; Sorribas. V. (2017). Intestinal phosphate
absorption is mediated by multiple transport systems in rats. Am J Physiol Gastrointest
Liver Physiol 312: G355-G366. http://dx.doi.org/10.1152/aipgi.00244.2016
Chan. JM; Pietinen. P; Virtanen. M; Mali la. N; Tangrea. J; Albanes. D; Virtamo. J. (2000). Diet
and prostate cancer risk in a cohort of smokers, with a specific focus on calcium and
phosphorus (Finland). Cancer Causes Control 11: 859-867.
http://dx.doi.Org/10.1023/a:1008947201132
Chang. AR; Anderson. C. (2017). Dietary phosphorus intake and the kidney [Review], Annu Rev
Nutr 37: 321-346. http://dx.doi.org/10.1146/annurev-nutr-071816-064607
Chang. AR; Lazo. M; Appel. LJ; Gutierrez. OM; Grams. ME. (2014). High dietary phosphorus
intake is associated with all-cause mortality: results from NHANES III. Am J Clin Nutr
99: 320-327. http://dx.doi.Org/10.3945/aicn.l 13.073148
130 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Chang. AR; Miller. ER; Anderson. CA; Juraschek. SP; Moser. M; White. K; Henry. B; Krekei
C; Oh. S; Charleston. J; Appel. LJ. (2017). Phosphorus additives and albuminuria in early
stages of CKD: A randomized controlled trial. Am J Kidney Dis 69: 200-209.
http://dx.doi.Org/10.1053/i.aikd.2016.08.029
Cheng. J; Tao. K; Shuai. X; Gao. J. (2016). Sodium phosphate versus polyethylene glycol for
colonoscopy bowel preparation: an updated meta-analysis of randomized controlled trials
[Review], Surgical Endoscopy 30: 4033-4041. http://dx.doi.org/10.1007/sQ0464-015-
4716-6
CIR Expert Panel (Cosmetic Ingredient Review Expert Panel). (2016). Safety assessment of
phosphoric acid and simple salts as used in cosmetics. Cosmetic Ingredient Review.
https://www.cir-safetv.org/suppiementaidoc/safetv-assessment-phosphoric-acid-and-
simple-salts-used-cosmetics-0
Clarke. HE; Coates. ME; Eva. JK; Ford. DJ; Milner. CK; O'Donoghue. PN; Scott. PP; Ward. RJ.
(1977). Dietary standards for laboratory animals: report of the Laboratory Animals
Centre Diets Advisory Committee. Lab Anim 11: 1-28.
Connor. A; Svkes. L; Roberts. IS; Weston. CE. (2008). Acute phosphate nephropathy after
sodium phosphate preparations. BMJ 337: a 182. http://dx.doi.org/10.1136/bmi.a 182
Coton. T; Luciano. L; Marcelko. A. (2011). Colic lesions induced by Colokit (R) preparation.
Clinics and Research in Hepatology and Gastroenterology 35: 509-510.
http://dx.doi.Org/10.1016/i.clinre.2011.04.003
Datta. PK; Frazer. AC; Sharratt. M; Sammons. HG. (1962). Biological effects of food additives:
II. Sodium pyrophosphate. J Sci Food Agric 13: 556-566.
http://dx.doi.org/10.1002/isfa.27401311Q2
Driman. DK; Preiksaitis. HG. (1998). Colorectal inflammation and increased cell proliferation
associated with oral sodium phosphate bowel preparation solution. Hum Pathol 29: 972-
978. http://dx.doi.org/10.1016/S0046-8177(98)90203-9
Dvmsza. HA; Reussner. G. Jr; Thiessen. R. Jr. (1959). Effect of normal and high intakes of
orthophosphate and metaphosphate in rats. J Nutr 69: 419-428.
http://dx.doi.Org/10.1093/in/69.4.419
ECHA (European Chemicals Agency). (2005). Dipotassium hydrogenorthophosphate: Screening
for reproductive/developmental toxicity. 001 Supporting | Experimental result.
https://echa.curopa.cu/rcgistration-dossier- registered-dossier/14974/1
ECHA (European Chemicals Agency). (2008). Orthophosphoric acid: Toxicological information:
Short-term repeated dose toxicity: Oral. 003 Key | Experimental result (JS member).
(231-633-2). https://echa.europa.eu/cs/registration-dossier/-/registered-
dossier/15531/7/6/2/?documentUUID=10e6b289-cbcb-4644-acc2-8d387f3a70fl#
ECHA (European Chemicals Agency). (2018). European chemicals agency. Available online at
https ://echa. europa. eu/
EFSA (European Food Safety Authority). (2005). Opinion of the scientific panel on dietetic
products, nutrition and allergies on a request from the commission related to the tolerable
upper intake level of phosphorus. EFSA J 233: 1-19.
EFSA (European Food Safety Authority). (2015). Scientific opinion on dietary reference values
for phosphorus. EFSA panel on dietetic products, nutrition and allergies (NDA). EFSA J
13:1-54.
131 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Elliott. P; Kesteloot. H; Appcl. LJ; Dyer. AR; Ueshima. H; Chan. Q; Brown. I J; Zhao. L;
Stamler. J; Group. ICR. (2008). Dietary phosphorus and blood pressure: international
study of macro- and micro-nutrients and blood pressure. Hypertension 51: 669-675.
http://dx.doi.Org/10.l 161/hypertensionaha. 107.103747
Elmstahl. S; Gullberg. B; Janzon. L; Johnett. O; Elmstahl. B. (1998). Increased incidence of
fractures in middle-aged and elderly men with low intakes of phosphorus and zinc.
Osteoporos Int 8: 333-340. http://dx.doi.org/10.1007/s001980050Q72
Erben. RG; Andrukhova. O. (2017). FGF23-Klotho signaling axis in the kidney [Review], Bone
100: 62-68. http://dx.doi.Org/10.1016/i.bone.2016.09.010
FASEB/LSRO (Federation of American Societies for Experimental Biology. Life Sciences
Research Office). (1975). Evaluation of the health aspects of phosphates as food
ingredients. (ETICBACK/5777). Washington, DC: Food and Drug Administration.
FDA (U.S. Food and Drug Administration). (2008). Information for healthcare professionals:
oral sodium phosphate (OSP) products for bowel cleansing (marketed as Visicol and
OsmoPrep, and oral sodium products available without a prescription). Available online
at https://www.fda.gov/Drugs/DrugSafetv/ucm 103354.htm
FDA (U.S. Food and Drug Administration). (2014). FDA drug safety communication. FDA
warns of possible harm from exceeding recommended dose of over-the-counter sodium
phosphate products to treat constipation. U.S. FDA.
https://www.fda.gov/Drugs/DrugSafetv/ucm380757.htm
FDRL (Food and Drug Research Laboratories Inc.). (1973a). Teratologic evaluation of FDA 71-
46 (sodium tripolyphosphate, anhydrous) (pp. 14 PP). (ETICBACK/6186). Washington,
DC: Food and Drug Administration.
FDRL (Food and Drug Research Laboratories Inc.). (1973b). Teratologic evaluation of FDA 71-
61 (sodium acid pyrophosphate). (ETICBACK/6191). Washington, DC: Food and Drug
Administration.
FDRL (Food and Drug Research Laboratories Inc.). (1974a). Teratologic evaluation of FDA 73-
1, tetrasodium pyrophosphate, anhydrous, in mice and rats. (PB-245 534). Washington,
DC: Food and Drug Administration.
FDRL (Food and Drug Research Laboratories Inc.). (1974b). Teratologic evaluation of FDA 73-
3 (sodium hexametaphosphate) in mice and rats. (ETICBACK/6206). Washington, DC:
Food and Drug Administration.
FDRL (Food and Drug Research Laboratories Inc.). (1975). Teratologic evaluation of FDA 73-
65, monopotassium phosphate, in mice and rats. (ETICBACK/6130). Washington, DC:
Food and Drug Administration.
Fine. KD; Qgunji. F; Florio. R; Porter. J; Ana. CS. (1998). Investigation and diagnosis of
diarrhea caused by sodium phosphate. Dig Dis Sci 43: 2708-2714.
http://dx.doi.Org/10.1023/A:1026611629624
Gard. DR. (2005). Phosphoric acids and phosphates. In Kirk-Othmer encyclopedia of chemical
technology. John Wiley & Sons, Inc.
https://onlinelibrarv.wilev.eom/doi/10.1002/0471238961.1608151907011804.a01.pub2
Grimm. M; Muller. A; Hein. G; Funfstuck. R; Jahreis. G. (2001). High phosphorus intake only
slightly affects serum minerals, urinary pyridinium crosslinks and renal function in young
women. Eur J Clin Nutr 55: 153-161. http://dx.doi.org/10.1038/si.eicn.1601131
Gross. P; Six. I; Kamel. S; Massy. MP. (2014). Vascular toxicity of phosphate in chronic kidney
disease-beyond vascular calcification. Circ J 78: 2339-2346.
132 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Gutierrez. OM. (2013). The connection between dietary phosphorus, cardiovascular disease, and
mortality: where we stand and what we need to know [Review], Adv Nutr 4: 723-729.
http://dx.doi.org/10.3945/an.113.0Q4812
Haas. S; Andersen. LM; Sommer. T. (2014). Randomized controlled trial comparing Moviprep®
and Phosphoral® as bowel cleansing agents in patients undergoing colonoscopy. Tech
Coloproctol 18: 929-935. http://dx.doi.org/10.1007/sl0151-014-1198-9
Haraikawa. M; Tanabe. R; Sogabe. N; Sugimoto. A; Kawamura. Y; Michigami. T; Hosoi. T;
Goseki-Sone. M. (2012). A study of the association between serum bone-specific alkaline
phosphatase and serum phosphorus concentration or dietary phosphorus intake. J Nutr Sci
Vitaminol 58: 442-445.
Heanev. RP; Nordin. BE. (2002). Calcium effects on phosphorus absorption: Implications for the
prevention and co-therapy of osteoporosis. J Am Coll Nutr 21: 239-244.
http://dx.doi.org/10.1080/07315724.20Q2.10719216
Heppe. DH; Medina-Gomez. C; Hofman. A; Franco. OH; Rivadeneira. F; Jaddoe. VW. (2013).
Maternal first-trimester diet and childhood bone mass: the Generation R Study. Am J
Clin Nutr 98: 224-232. http://dx.doi.Org/10.3945/aicn.l 12.051052
Hitchman. AJ; Hasanv. SA; Hitchman. A; Harrison. JE; Tam. C. (1979). Phophate-induced renal
calcification in the rat. Can J Physiol Pharmacol 57: 92-97.
http://dx.doi.Org/10.l 139/v79-013
Hodge. HC. (1964). Toxicity studies on phosphates. Food Cosmet Toxicol 2: 147-154.
Hong. SH; Park. SJ; Lee. S; Kim. S; Cho. MH. (2015). Biological effects of inorganic phosphate:
potential signal of toxicity [Review], J Toxicol Sci 40: 55-69.
http://dx.doi.org/10.2131/its.40.55
Huang. CL. (2010). Regulation of ion channels by secreted Klotho: Mechanisms and
implications. Kidney Int 77: 855-860. http://dx.doi.org/10.1038/ki.2010.73
Hurst FP: Bohen. EM: Osgard. EM: Oliver. DK: Das. NP: Gao. SW: Abbott KC. (2007).
Association of oral sodium phosphate purgative use with acute kidney injury. J Am Soc
Nephrol 18: 3192-3198. http://dx.doi.org/10.1681/ASN.200703Q349
Huttunen. MM: Tillman. I: Viliakainen. HT; Tuukkanen. J: Peng. Z; Pekkinen. M; Lamberg-
Allardt. CJ. (2007). High dietary phosphate intake reduces bone strength in the growing
rat skeleton. J Bone Miner Res 22: 83-92. http://dx.doi.org/10.1359/ibmr.061009
IOM (Institute of Medicine). (1997). Dietary reference intakes for calcium, phosphorus,
magnesium, vitamin D, and fluoride. Washington, DC: National Academy Press.
http://www.nap.edu/catalog.php7record id 5776
Itkonen. ST: Karp. HI: Kemi. YE: Kokkonen. EM: Saarnio. EM: Pekkinen. MH: Karkkainen.
MU: Laitinen. EK; Turanlahti. MI: Lamberg-Allardt. CJ. (2013). Associations among
total and food additive phosphorus intake and carotid intima-media thickness-a cross-
sectional study in a middle-aged population in Southern Finland. Nutr J 12: 94.
http://dx.doi.org/10.1186/1475-2891-12-94
Ito. S: Ishida. H; Uenishi. K; Murakami. K; Sasaki. S. (2011). The relationship between habitual
dietary phosphorus and calcium intake, and bone mineral density in young Japanese
women: a cross-sectional study. Asia Pac J Clin Nutr 20: 411-417.
Jain. N: Elsaved. EF. (2013). Dietary phosphate: What do we know about its toxicity? [Review],
J Nephrol 26: 856-864. http://dx.doi.org/10.5301/in.500Q242
133 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Kan. WC; Wang. HY; Chi en. CC; Tan. CK; Lin. CY; Su. SB. (2012). Intermediate bioelectrolyte
changes after phospho-soda or polyethylene glycol precolonoscopic laxatives in a
population undergoing health examinations. Nephrol Dial Transplant 27: 752-757.
http://dx.doi.org/10.1093/ndt/gfrl89
Katsumata. S; Masuvama. R; Uehara. M; Suzuki. K. (2005). High-phosphorus diet stimulates
receptor activator of nuclear factor-kappaB ligand mRNA expression by increasing
parathyroid hormone secretion in rats. Br J Nutr 94: 666-674.
http://dx.doi.org/10.1079/BM20Q51552
Katsumata. S; Matsuzaki. H; Uehara. M; Suzuki. K. (2015). Effects of dietary calcium
supplementation on bone metabolism, kidney mineral concentrations, and kidney
function in rats fed a high-phosphorus diet. J Nutr Sci Vitaminol 61: 195-200.
http://dx.doi.org/10.3177/insv.61.195
Kesse. E; Bertrais. S; Astorg. P; Jaouen. A; Arnault N; Galan. P; Hercberg. S. (2006). Dairy
products, calcium and phosphorus intake, and the risk of prostate cancer: results of the
French prospective SU.VI.MAX (Supplementation en Vitamines et Mineraux
Antioxydants) study. Br J Nutr 95: 539-545.
Kesse. E; Boutron-Ruault MC; Norat. T; Riboli. E; Ctavet-Chapeton. F; Group. EN. (2005).
Dietary calcium, phosphorus, vitamin D, dairy products and the risk of colorectal
adenoma and cancer among French women of the E3N-EPIC prospective study. Int J
Cancer 117: 137-144. http://dx.doi.org/10.10Q2/iic.21148
Khetifi. M; Zarrouk. A; Nurv. T; Hamed. H; Saguem. S; Salah. RB; Riedinger. JM; Lizard. G.
(2014). Cytokine and eicosanoid profiles of phosphate mine workers. J Toxicol Sci 39:
465-474.
Khurana. A; Mclean. L; Atkinson. S; Foulks. CJ. (2008). The effect of oral sodium phosphate
drug products on renal function in adults undergoing bowel endoscopy. Arch Intern Med
168: 593-597. http://dx.doi.Org/10.1001/archinte.168.6.593
Kim. SJ; Rim. KT; Kim. HY; Yang. IS. (2010). Mutagenicity of octane and tetrasodium
pyrophosphate in bacterial reverse mutation (Ames) test [Letter], J Toxicol Sci 35: 555-
562.
Koshihara. M; Katsumata. S; Uehara. M; Suzuki. K. (2005). Effects of dietary phosphorus intake
on bone mineralization and calcium absorption in adult female rats. Biosci Biotechnol
Biochem 69: 1025-1028. http://dx.doi.org/10.1271/bbb.69.1025
Kuro. OM. (2019). The Klotho proteins in health and disease. Nature Revisions, Nephrology 15:
27-44. http://dx.doi.org/10.1038/s41581-018-0078-3
Kwak. SM; Kim. IS; Choi. Y; Chang. Y; Kwon. MJ; Jung. JG; Jeong. C; Ahn. J; Kim. HS; Shin.
H; Ryu. S. (2014). Dietary intake of calcium and phosphorus and serum concentration in
relation to the risk of coronary artery calcification in asymptomatic adults. Arterioscler
Thromb Vase Biol 34: 1763-1769. http://dx.doi.Org/10.l 161/ATVBAHA. 114.303440
Lavton. JB; Klemmer. PJ; Christiansen. CF; Bomback. AS; Baron. JA; Sandler. RS; Kshirsagar.
AY. (2014). Sodium phosphate does not increase risk for acute kidney injury after routine
colonoscopy, compared with polyethylene glycol. Clin Gastroenterol Hepatol 12: 1514-
1521 .el5 13. http://dx.doi.Org/10.1016/i.cgh.2014.01.034
Lee. AW; Cho. SS. (2015). Association between phosphorus intake and bone health in the
NHANES population. Nutr J 14: 28. http://dx.doi.Org/10.l 186/sl2937-015-0017-0
Lee. GJ; Marks. J. (2015). Intestinal phosphate transport: a therapeutic target in chronic kidney
disease and beyond? [Review], Pediatr Nephrol 30: 363-371.
http://dx.doi.org/10.1007/sQ0467-014-2759-x
134 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Lee. KJ; Kim. KS; Kim. HN; Sco. JA; Song. SW. (2014). Association between dietary calcium
and phosphorus intakes, dietary calcium/phosphorus ratio and bone mass in the Korean
population. Nutr J 13: 114. http://dx.doi.org/10.1186/1475-2891-13-114
Lee. SW; Bang. CS; Park. TY; Suk. KT; Baik. GH; Kim. DJ. (2017). Split-dose bowel
preparation for colonoscopy: 2 liters polyethylene glycol with ascorbic acid versus
sodium picosulfate versus oral sodium phosphate tablets. Korean J Gastroenterol 70: 89-
95. http://dx.doi.Org/10.4166/kig.2017.70.2.89
Litton Bionetics. (1975a). Mutagenic evaluation of compound FDA 71-46, sodium
tripolyphosphate. (EMICBACK/29961).
Litton Bionetics. (1975b). Mutagenic evaluation of compound FDA 73-1, tetrasodium
pyrophosphate. (EMICBACK/29943).
Litton Bionetics. (1975c). Mutagenic evaluation of compound FDA 73-2, monosodium
phosphate anhydrous powdered, FCC grade. (EMICBACK/29919).
Litton Bionetics. (1975d). Mutagenic evaluation of compound FDA 73-3, sodium
hexametaphosphate. (PB-245486).
Ma. HL; Qin. MJ; Qi. LW; Wu. G; Shu. P. (2007). Improved quality evaluation of Radix Salvia
miltiorrhiza through simultaneous quantification of seven major active components by
high-performance liquid chromatography and principal component analysis. Biomed
Chromatogr 21: 931-939. http://dx.doi.org/10.1002/bmc.836
MacKav. EM; Oliver. J. (1935). Renal damage following the ingestion of a diet containing an
excess of inorganic phosphate. J Exp Med 61: 319-334.
http://dx.doi.Org/10.1084/iem.61.3.319
Manukvan. MN; Tolan. K; Severge. U; Attaallah. W; Kebudi. A; Cingi. A. (2011). Prospective
randomized comparison of oral sodium phosphate and sennoside A+B calcium lavage for
colonoscopy preparation. Surg Laparosc Endosc Percutan Tech 21: 90-93.
http://dx.doi.org/10.1097/SLE.0b013e3182Qe8793
Markowitz. GS; Nasr. SH; Klein. P; Anderson. H; Stack. JI; Alterman. L; Price. B;
Radhakrishnan. J; D'Agati. YD. (2004). Renal failure due to acute nephrocalcinosis
following oral sodium phosphate bowel cleansing. Hum Pathol 35: 675-684.
http://dx.doi.Org/10.1016/i.humpath.20Q3.12.0Q5
Markowitz. GS; Perazetta. MA. (2009). Acute phosphate nephropathy [Review], Kidney Int 76:
1027-1034. http://dx.doi.org/10.1038/ki.2009.3Q8
Markowitz. GS; Stokes. MB; Radhakrishnan. J; D'Agati. YD. (2005). Acute phosphate
nephropathy following oral sodium phosphate bowel purgative: an underrecognized
cause of chronic renal failure. J Am Soc Nephrol 16: 3389-3396.
http://dx.doi.org/10.1681/ASN.200505Q496
Marks. J; Debnam. ES; Unwin. RJ. (2010). Phosphate homeostasis and the renal-gastrointestinal
axis [Review], Am J Physiol Renal Physiol 299: F285-F296.
http://dx.doi.Org/10.l 152/aiprenal.00508.2009
Marshall. JB. (2014). Should oral sodium phosphate solution return as the first-line preparation
for colonoscopy? [Editorial], Clin Gastroenterol Hepatol 12: 1522-1524.
http://dx.doi.Org/10.1016/i.cgh.2014.02.015
Matsuzaki. H; Fuchigami. M; Miwa. M. (2010). Dietary magnesium supplementation suppresses
bone resorption via inhibition of parathyroid hormone secretion in rats fed a high-
phosphorus diet. Magnes Res 23: 126-130. http://dx.doi.org/10.1684/mrh.2010.0212
135 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Matsuzaki. H; Katsumata. S; Masuvama. R; Uehara. M; Nakamura- K; Suzuki. K. (2002). Sex
differences in kidney mineral concentrations and urinary albumin excretion in rats given
high-phosphorus feed. Biosci Biotechnol Biochem 66: 1737-1739.
http://dx.doi.org/10.1271/bbb.66.1737
Matsuzaki. H; Kikuchi. T; Kaiita. Y; Masuvama. R; Uehara. M; Goto. S; Suzuki. K. (1999).
Comparison of various phosphate salts as the dietary phosphorus source on
nephrocalcinosis and kidney function in rats. J Nutr Sci Vitaminol 45: 595-608.
Matsuzaki. H; Uehara. M; Suzuki. K; Liu. QL; Sato. S; Kanke. Y; Goto. S. (1997). High
phosphorus diet rapidly induces nephrocalcinosis and proximal tubular injury in rats. J
Nutr Sci Vitaminol 43: 627-641.
McClure. ST; Chang. AR; Selvin. E; Rebholz. CM; Appel. LJ. (2017). Dietary sources of
phosphorus among adults in the United States: Results from NHANES 2001-2014.
Nutrients 9. http://dx.doi.org/10.3390/nu9020Q95
Medoff. J; Katz. S; Malik. P; Pambianco. D; Pruitt. R; Poulos. J; Rank. J; Rose. M. (2004).
Open-label, dose-ranging pilot study of 4 weeks of low-dose therapy with sodium
phosphate tablets in chronically constipated adults. Clin Ther 26: 1479-1491.
http://dx.doi.Org/10.1016/i.clinthera.2004.09.007
Mendez. RO; Gomez. MA; Lopez. AM; Gonzalez. H; Wvatt. CJ. (2002). Effects of calcium and
phosphorus intake and excretion on bone density in postmenopausal women in
Hermosillo, Mexico. Ann Nutr Metab 46: 249-253. http://dx.doi.Org/10.l 159/000066497
Menon. MC; Ix. JH. (2013). Dietary phosphorus, serum phosphorus, and cardiovascular disease
[Review], Ann NY Acad Sci 1301: 21-26. http://dx.doi.Org/10.l 111/nvas. 12283
Michaud. DS; Spiegetman. D; Clinton. SK; Rimm. EB; Willctt. WC; Giovannucci. E. (2000).
Prospective study of dietary supplements, macronutrients, micronutrients, and risk of
bladder cancer in US men. Am J Epidemiol 152: 1145-1153.
http://dx.doi.org/10.1093/aie/152.12.1145
Murtaugh. MA; Filipowicz. R; Baird. BC; Wei. G; Greene. T; Beddhu. S. (2012). Dietary
phosphorus intake and mortality in moderate chronic kidney disease: NHANES III.
Nephrol Dial Transplant 27: 990-996. http://dx.doi.org/10.1093/ndt/gfr367
Nadkarni. GN; Uribarri. J. (2014). Phosphorus and the kidney: What is known and what is
needed [Review], Adv Nutr 5: 98-103. http://dx.doi.Org/10.3945/an.l 13.004655
Nakamura. K; Hori. Y; Nashimoto. M; Okuda. Y; Mivazaki. H; Kasai. Y; Yamamoto. M.
(2004). Dietary calcium, sodium, phosphorus, and protein and bone metabolism in
elderly Japanese women: a pilot study using the duplicate portion sampling method.
Nutrition 20: 340-345. http://dx.doi.Org/10.1016/i.nut.2003.12.004
NIDDK (National Institute of Diabetes and Digestive and Kidney Diseases). (2016). Kidney
disease statistics for the United States. Available online at
https://www.niddk.nih.gov/health-information/health-statistics/kidnev-disease
NIOSH (National Institute for Occupational Safety and Health). (2016). Phosphoric acid. NIOSH
pocket guide to chemical hazards. Atlanta, GA: Centers for Disease Control and
Prevention, U.S. Department of Health, Education and Welfare.
http s: //www, cdc. gov/ni osh/np g/np gdO5 06. html
Nishi. T; Shuto. E; Ogawa. M; Ohva. M; Nakanishi. M; Masuda. M; Katsumoto. M; Yamanaka-
Okumura. H; Sakai. T; Takeda. E; Sakaue. H; Taketani. Y. (2015). Excessive dietary
phosphorus intake impairs endothelial function in young healthy men: a time- and dose-
dependent study. J Med Invest 62: 167-172. http://dx.doi.org/10.2152/imi.62.167
136 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
NLM (National Library of Medicine). (2021). Hazardous substances data bank [Database],
Bethesda, MD. Retrieved from https://www.nlm.nih.gov/databases/download/hsdb.html
Noori. N; Kalantar-Zadeti. K; Kovesdv. CP; Bross. R; Benner. D; Kopple. JD. (2010a).
Association of dietary phosphorus intake and phosphorus to protein ratio with mortality
in hemodialysis patients. Clin J Am Soc Nephrol 5: 683-692.
http://dx.doi.org/10.2215/CJN.086012Q9
Noori. N; Sims. JJ; Kopple. JD; Shah. A; Col man. S; Shinaberger. CS; Bross. R; Mehrotra. R;
Kovesdv. CP; Kalantar-Zadeh. K. (2010b). Organic and inorganic dietary phosphorus and
its management in chronic kidney disease. Iran J Kidney Dis 4: 89-100.
NRC (National Research Council). (1977). Nutrient requirements of rabbits: second revised
edition, 1977. Washington, DC: The National Academies Press.
http ://dx.doi.org/10.17226/35
NRC (National Research Council). (1995). Nutrient requirements of laboratory animals (Fourth
revised ed.). Washington, DC: The National Academies Press.
http://dx.doi.org/10.17226/4758
NTP (National Toxicology Program). (2016). 14th Report on carcinogens. Research Triangle
Park, NC. https://ntp.niehs.nih.gov/pubhealth/roc/index-1.html
OECD (Organisation for Economic Co-operation and Development). (2006). S1DS initial
assessment report for SIAM 23. Dipotassium hydrogenphosphate (DKP).
OECD (Organisation for Economic Co-operation and Development). (2009). S1DS initial
assessment profile. Phosphoric acid.
Ori. Y; Herman. M; Tobar. A; Chernin. G; Gafter. U; Chagnac. A; Izhak. OB; Korzets. A.
(2008). Acute phosphate nephropathy-an emerging threat. Am J Med Sci 336: 309-314.
http://dx.doi. org/10.1097/MAJ. ObO 13 e3 18167410c
Orias. M; Mahnensmith. RL; Perazetta. MA. (1999). Extreme hyperphosphatemia and acute
renal failure after a phosphorus-containing bowel regimen. Am J Nephrol 19: 60-63.
OSHA (Occupational Safety & Health Administration). (2017a). Air contaminants: Occupational
safety and health standards for shipyard employment, subpart Z, toxic and hazardous
substances. (OSHA Standard 1915.1000). Washington, DC: U.S. Department of Labor.
https://www.osha.gov/pls/oshaweb/owadisp.show document?p table STANDARDS&p
id10286
OSHA (Occupational Safety & Health Administration). (2017b). Table Z-l: Limits for air
contaminants. Occupational safety and health standards, subpart Z, toxic and hazardous
substances. (OSHA standard 1910.1000, 29 CFR). Washington, DC: U.S. Department of
Labor.
http://www.osha.gov/pls/oshaweb/owadisp.show document?p table STANDARDS&p
id=9992
OSHA (Occupational Safety & Health Administration). (2020). Safety and health regulations for
construction: Occupational health and environmental controls: Gases, vapors, fumes,
dusts, and mists: Appendix A. Available online at
http://www.osha.gov/pls/oshaweb/owadisp.show document?p table STANDARDS&p
id10629
Patel. V; Nicar. M; Emmett. M; Asplin. J; Maguire. JA; Santa Ana. CA; Fordtran. IS. (2009).
Intestinal and renal effects of low-volume phosphate and sulfate cathartic solutions
designed for cleansing the colon: Pathophysiological studies in five normal subjects. Am
J Gastroenterol 104: 953-965. http://dx.doi.org/10.1038/aig.2008.124
137 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Pel ham. RW; Russell RG; Padgett EL; Reno. FE; Cleveland. M. (2009). Safety of oral sulfates
in rats and dogs contrasted with phosphate-induced nephropathy in rats. Int J Toxicol 28:
99-112. http://dx.doi.org/10.1177/10915818Q9335124
PubChem. (2018). PubChem. Available online at https://pubchem.ncbi.nlm.nih.gov/
Raitakari. OT; Cclcrmaicr. DS. (2000). Flow-mediated dilatation. Br J Clin Pharmacol 50: 397-
404.
Ramezani Tehrani. F; Moslehi. N; Asghari. G; Gholami. R; Mirmiran. P; Azizi. F. (2013). Intake
of dairy products, calcium, magnesium, and phosphorus in childhood and age at
menarche in the Tehran Lipid and Glucose Study. PLoS ONE 8: e57696.
http://dx.doi.org/10.1371/iournal.pone.0Q57696
Reichrt. S; Bures. J; Sirokv. M; Kopacova. M; Stezak. L; Langr. F. (2004). A prospective,
observational study of colonic mucosal abnormalities associated with orally administered
sodium phosphate for colon cleansing before colonoscopy. Gastrointest Endosc 59: 651-
654. http://dx.doi.org/10.1016/S0016-5107(04)00158-0
Ritskes-Hoitinga. J; Grooten. FIN; Wienk. KJ; Peters. M; Lemmens. AG; Bevnen. AC. (2004).
Lowering dietary phosphorus concentrations reduces kidney calcification, but does not
adversely affect growth, mineral metabolism, and bone development in growing rabbits.
Br JNutr 91: 367-376. http://dx.doi.org/10.1079/BJN20041065
Ritskes-Hoitinga. J; Lemmens. AG; Danse. I.H; Bevnen. AC. (1989). Phosphorus-induced
nephrocalcinosis and kidney function in female rats. J Nutr 119: 1423-1431.
http://dx.doi.Org/10.1093/in/l 19.10.1423
Ritter. SC; Slatopolskv. E. (2016). Phosphate toxicity in CKD: The killer among us. Clin J Am
SocNephrol 11: 1088-1100. http://dx.doi.Org/10.2215/CJN.l 1901115
Russmann. S; Lamerato. L; Marfatia. A; Motsko. SP; Pezzutto. JC; Olds. G; Jones. JK. (2007).
Risk of impaired renal function after colonoscopy: A cohort study in patients receiving
either oral sodium phosphate or polyethylenglycol [Abstract], Pharmacoepidemiol Drug
Saf 16: S230-S231. http://dx.doi.org/10.1002/pds.1461
Samuel. TM; Thomas. T; Finkelstein. J; Bosch. R; Raiendran. R; Virtanen. SM; Srinivasan. K;
Kurpad. AY; Duggan. C. (2013). Correlates of anaemia in pregnant urban South Indian
women: A possible role of dietary intake of nutrients that inhibit iron absorption. Public
Health Nutr 16: 316-324. http://dx.doi.org/10.1017/S136898001200119X
Schneider. P; Ober. KM; Ueberberg. H. (1981). Contribution to the phosphate-induced
nephropathy in the dog. Comparative light and electron microscopic investigations on the
proximal tubule after oral application of K2HP04, Na2HP04, KC1 and NaCl. Exp Pathol
19: 53-65.
Seo. DS; Kwon. M; Sung. HJ; Park. CB. (2011a). Acute oral or dermal and repeated dose 90-day
oral toxicity of tetrasodium pyrophosphate in Spraque [sic] Dawley (SD) rats.
Environmental Health and Toxicology 26.
http://dx.doi.org/10.5620/eht.2011.26.e2011014
Seo. EH; Kim. TO; Kim. TG; Joo. HR; Park. MJ; Park. J; Park. SH; Yang. SY; Moon. YS.
(201 lb). Efficacy and tolerability of split-dose PEG compared with split-dose aqueous
sodium phosphate for outpatient colonoscopy: a randomized, controlled trial. Dig Dis Sci
56: 2963-2971. http://dx.doi.org/10.1007/s 10620-01 1-1772-1
Seol. DC; Hong. SN; Kim. JH; Sung. IK; Park. HS; Lee. JH; Shim. CS. (2010). Change in renal
function after sodium phosphate preparation for screening colonoscopy. World J
Gastroenterol 16: 2010-2016.
138 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Shibata. MA; Sano. M; Shirai. T; Irnaida. K; Fukushima. S. (1993). Epithelial hyperplasia in the
renal papilla and pelvis but not the urinary-bladder of male F344 rats associated with
dietary-sodium phosphates after uracil exposure. Toxicol Appl Pharmacol 121: 303-312.
http://dx.doi.Org/10.1006/taap.1993.l 158
Stiuto. E; Taketani. Y; Tanaka. R; Harada. N; Isshiki. M; Sato. M; Nashiki. K; A mo. K;
Yamamoto. H; Higashi. Y; Nakava. Y; Takeda. E. (2009). Dietary phosphorus acutely
impairs endothelial function. J Am Soc Nephrol 20: 1504-1512.
http://dx.doi. org/10.1681/ASN.2008101106
Singal. AK; Rosman. AS; Post JB; Bauman. WA; Spungen. AM; Korsten. MA. (2008). The
renal safety of bowel preparations for colonoscopy: a comparative study of oral sodium
phosphate solution and polyethylene glycol. Aliment Pharmacol Ther 27: 41-47.
http://dx.doi.org/10.1 1 11/i. 1365-2036.2007.03558.X
Slee. TM; Vleming. LJ; Valentijn. RM. (2008). Renal failure due to acute phosphate
nephropathy. Neth J Med 66: 438-441.
Tani. Y; Sato. T; Yamanaka-Okumura. H; Yamamoto. H; Arai. H; Sawada. N; Genjida. K;
Taketani. Y; Takeda. E. (2007). Effects of prolonged high phosphorus diet on phosphorus
and calcium balance in rats. J Clin Biochem Nutr 40: 221-228.
http://dx.doi.org/10.3164/icbn.40.221
Tavani. A; Bertuccio. P; Bosetti. C; Talamini. R; Negri. E; Franceschi. S; Montella. M; La
Vecchia. C. (2005). Dietary intake of calcium, vitamin D, phosphorus and the risk of
prostate cancer. Eur Urol 48: 27-33. http://dx.doi.org/10.1016/i.eururo.2005.03.023
Tobias. JH; Steer. CD; Emmett PM; Tonkin. RJ; Cooper. C; Ness. AR. (2005). Bone mass in
childhood is related to maternal diet in pregnancy. Osteoporos Int 16: 1731-1741.
http://dx.doi.org/10.1007/s00198-0Q5-1912-6
Trautvetter. U; Ditscheid. B; Jahreis. G; Glei. M. (2018). Habitual intakes, food sources and
excretions of phosphorus and calcium in three german study collectives. Nutrients 10:
171. http://dx.doi.org/10.3390/nu 10020171
Tsai. TY; Hsu. PF; Wu. CH; Yang. YL; Chen. SC; Huang. SS; Chan. WL; Lin. SJ; Chen. JW;
Pan. JP; Charng. MJ; Chen. YH; Wu. TC; Lu. TM; Huang. PH; Cheng. HM; Huang. CC;
Sung. SH; Lin. YJ; Leu. HB. (2021). Association between phosphate and long-term
outcome in CAD patients underwent coronary intervention. Sci Rep 11: 20080.
http://dx.doi. org/10.103 8/s41598-021-995 18-z
Tseng. M; Breslow. RA; Graubard. BI; Ziegler. RG. (2005). Dairy, calcium, and vitamin D
intakes and prostate cancer risk in the National Health and Nutrition Examination
Epidemiologic Follow-up Study cohort. Am J Clin Nutr 81: 1147-1154.
http://dx.doi.Org/10.1093/aicn/81.5.l 147
U.S. APHC (U.S. Army Public Health Command). (2013). Environmental health risk assessment
and chemical exposure guidelines for deployed military personnel. (Technical guide 230,
2013 revision). Aberdeen Proving Ground, MD.
https://phc.amedd.armv.mil/PHC%20Resource%20Librarv/TG230-DeplovmentEHRA-
and-MEGs-2013 -Revi si on. pdf
U.S. EPA (U.S. Environmental Protection Agency). (1988). Recommendations for and
documentation of biological values for use in risk assessment [EPA Report],
(EPA600687008). Cincinnati, OH.
http://cfpub. epa.gov/ncea/cfm/recordisplav. cfm?deid=34855
139 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
U.S. EPA (U.S. Environmental Protection Agency). (1995). Integrated Risk Information System
(IRIS) chemical assessment summary for phosphoric acid. Washington, DC: Office of
Research and Development, National Center for Environmental Assessment.
https://cfpub.epa.gov/ncea/iris/iris documents/documents/subst/0697 summarv.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2002). A review of the reference dose and
reference concentration processes. (EPA630P02002F). Washington, DC.
https://www.epa.gov/sites/production/files/2014-12/documents/rfd-final.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2011a). Health effects assessment summary
tables (HEAST) for superfund [EPA Report], Washington, DC. https://epa-
heast. ornl. gov/heast. php
U.S. EPA (U.S. Environmental Protection Agency). (2011b). Recommended use of body weight
3/4 as the default method in derivation of the oral reference dose. (EPA100R110001).
Washington, DC. https://www.epa.gov/sites/production/files/2013-
09/documents/recommended-use-of-b w3 4.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2012). Estimation Programs Interface
Suite™ for Microsoft® Windows, v 4.11 [Computer Program], Washington, DC.
Retrieved from https://www.epa.gov/tsca-screening-tools/epi-suitetm-estimation-
program-interface
U.S. EPA (U.S. Environmental Protection Agency). (2018). 2018 Edition of the drinking water
standards and health advisories tables [EPA Report], (EPA822F18001). Washington, DC:
U.S. Environmental Protection Agency, Office of Water.
https://www.epa.gOv/svstem/files/documents/2022-01/dwtable2018.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2020). Integrated risk information system.
IRIS assessments [Database], Washington, DC. Retrieved from http://www.epa.gov/iris/
Uribarri. J. (2013). Dietary phosphorus and kidney disease [Review], Ann N Y Acad Sci 1301:
11-19. http://dx.doi.org/10.1 1 11/nvas. 12201
van Lee. L; Hevworth. J; Mcnaughton. S; lacopetta. B; ('lavforth. C; Fritschi. L. (2011). Selected
dietary micronutrients and the risk of right- and left-sided colorectal cancers: a case-
control study in Western Australia. Ann Epidemiol 21: 170-177.
http://dx.doi.Org/10.1016/i.annepidem.2010.10.005
Watts. DA; Lcssclls. AM; Penman. ID; Ghosh. S. (2002). Endoscopic and histologic features of
sodium phosphate bowel preparation-induced colonic ulceration: Case report and review.
Gastrointest Endosc 55: 584-587. http://dx.doi.org/10.1067/mge.20Q2.122582
Weiner. ML; Salminen. WF; Larson. PR; Barter. RA; Kranetz. JL; Simon. GS. (2001).
Toxicological review of inorganic phosphates [Review], Food Chem Toxicol 39: 759-
786. http://dx.doi.org/10.1016/S0278-6915(01 )00028-X
Whiting. SJ; Bovle. JL; Thompson. A; Mirwatd. RL; Faulkner. RA. (2002). Dietary protein,
phosphorus and potassium are beneficial to bone mineral density in adult men consuming
adequate dietary calcium. J Am Coll Nutr 21: 402-409.
http://dx.doi.org/10.1080/07315724.20Q2.10719242
WHO (World Health Organization). (1982). Phosphoric acid and phosphate salts. In
Toxicological evaluation of certain food additives and contaminants. (WHO Food
Additive Series 17). Geneva, Switzerland: International Programme on Chemical Safety.
Willhite. CC; Ball. GL; Bhat. VS. (2013). Emergency do not consume/do not use concentrations
for blended phosphates in drinking water. Hum Exp Toxicol 32: 241-259.
http://dx.doi.org/10.1177/0960327112459207
140 Na/K Salts of Inorganic Phosphates
-------
EPA/690/R-23/001F
Wilson. KM; Shut IM; Mucci. LA; (jiovannucci. E. (2015). Calcium and phosphorus intake and
prostate cancer risk: a 24-y follow-up study. Am J Clin Nutr 101: 173-183.
http://dx.doi.Org/10.3945/aicn.l 14.088716
Yakut M; Cinar. K; Seven. G; Cetinkava. H; Bahar. K. (2010). The efficacy and safety of
colonoscopy preparation with oral sodium phosphate in elderly patients. Turk J
Gastroenterol 21: 140-145. http://dx.doi.org/10.4318/tig.2010.0Q72
Yamamoto. KT; Robinson-Cohen. C; de Oliveira. MC; Kostina. A; Nettleton. JA; Ix. JH;
Nguyen. H; Eng. J; Lima. JA; Siscovick. DS; Weiss. NS; Kestenbaum. B. (2013). Dietary
phosphorus is associated with greater left ventricular mass. Kidney Int 83: 707-714.
http://dx.doi.org/10.1038/ki.2012.303
Yiin. JH; Daniels. RD; Kubale. TL; Dunn. KL; Stavner. LT. (2016). A study update of mortality
in workers at a phosphate fertilizer production facility. Am J Ind Med 59: 12-22.
http://dx.doi.org/10.10Q2/aiim.22542
Yilmaz. S; Unal. F; Yiizbasioglu. D; Celik. M. (2014). DNA damage in human lymphocytes
exposed to four food additives in vitro. Toxicol Ind Health 30: 926-937.
http://dx.doi.org/10.1177/0748233712466132
Yin. J; Dwver. T; Rilev. M; Cochrane. J; Jones. G. (2010). The association between maternal
diet during pregnancy and bone mass of the children at age 16. Eur J Clin Nutr 64: 131-
137. http://dx.doi.Org/10.1038/eicn.2009.l 17
Yoon. CY; Park. IT; Jhee. JH; Noh. J; Kee. YK; Seo. C; Lee. M; Cha. MU; Kim. H; Park. S;
Yun. HR; Jung. SY; Han. SH; Yoo. TH; Kang. SW. (2017). High dietary phosphorus
density is a risk factor for incident chronic kidney disease development in diabetic
subjects: a community-based prospective cohort study. Am J Clin Nutr 106: 311-321.
http://dx.doi.Org/10.3945/aicn.l 16.151654
141 Na/K Salts of Inorganic Phosphates
------- |