&EPA
EPA/600/R-11/067
United States
Environmental Protection
Agency
STATUS REPORT:
Advances in Inhalation Dosimetry for Gases
with Lower Respiratory Tract and Systemic
Effects
September 2011
U.S. Environmental Protection Agency
Washington, DC
-------
DISCLAIMER
This report has been reviewed in accordance with U.S. Environmental Protection Agency Policy and approved for
publication. Mention of trade names or commercial products does not constitute endorsement or recommendation
for use.
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
AUTHORS
John J. Stanek
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Research Triangle Park, NC
Eva D. McLanahan
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Research Triangle Park, NC
REVIEWERS
This document has been reviewed by EPA scientists and has undergone a formal peer letter review
performed by independent scientists external to the EPA. Comments from reviewers were evaluated
carefully and considered by the Agency during preparation of this final report. A summary of significant
comments made by the external peer reviewers, and EPA responses, is included in Appendix A.
INTERNAL EPA CONTRIBUTORS AND REVIEWERS
Robert Benson
Region 8, U.S. Environmental Protection Agency
U.S. Environmental Protection Agency
Denver, CO
Rebecca Dzubow
Office of Children's Health Protection
U.S. Environmental Protection Agency
Washington, DC
Ernest Falke
Office of Pollution Prevention and Toxics
U.S. Environmental Protection Agency
Washington, DC
Brenda Foos
Office of Children's Health Protection
U.S. Environmental Protection Agency
Washington, DC
Connie Meacham
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC
Suril Mehta
Office of Children's Health Protection
U.S. Environmental Protection Agency
Washington, DC
Deirdre Murphy
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, NC
m
-------
Reeder Sams
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Research Triangle Park, NC
Paul Schlosser
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Washington, DC
Ravi Subramanian
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Washington, DC
John Vandenberg
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Research Triangle Park, NC
Debra Walsh
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Research Triangle Park, NC
EXTERNAL REVIEWERS
Phillip Worth Longest, Jr
Virginia Commonwealth University, Richmond, VA
Jeffry Schroeter
The Hamner Institute, Research Triangle Park, NC
James Ultman
Pennsylvania State University, University Park, PA
TECHNICAL SUPPORT
James S. Lucy - Student Services Authority
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Research Triangle Park, NC
Ken Breito-Senior Environmental Employment Program
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Research Triangle Park, NC
Deborah Wales
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Research Triangle Park, NC
Barbara Wright - Senior Environmental Employment Program
National Center for Environmental Assessment
U.S. Environmental Protection Agency
Research Triangle Park, NC
CONTRACT SUPPORT
Gary L. Foureman (Preparation)
ICF International; 9300 Lee Highway; Fairfax, VA 22031
Julie Kimball (Internal Reviewer)
ICF International; 9300 Lee Highway; Fairfax, VA 22031
IV
-------
TABLE OF CONTENTS
AUTHORS, CONTRIBUTORS, AND REVIEWERS.
GLOSSARY VII
ABBREVIATIONS AND ACRONYMS IX
EXECUTIVE SUMMARY 1-1
1 INTRODUCTION AND PURPOSE 1-4
2 INTERSPECIES GAS DOSIMETRY IN THE RFC - THE REGIONAL GAS DOSE RATIO
(RGDR) AS A DOSIMETRIC ADJUSTMENT FACTOR (DAF) 2-1
2.1 Current Applications of the DAFs - RGDRTB, RGDRPU, and Hb/g 2-2
Table 2-1 Surface Areas for the Tracheobronchial (TB) and Pulmonary (PU) Regions
of the Respiratory Tract in Various Species 2-2
Table 2-2 Intercept (bo) and Coefficient (b-i) Values Used to Calculate Default Minute
Volumes Based on Body Weight3 2-2
Table 2-3 Some Example Blood:Air Partition Coefficients (Hb/g) in Humans and Rats
Expressed as a Ratio, Animal/Human 2-3
2.2 Assumptions for the Current Application of RGDR Procedures 2-3
2.2.1 DAF for TB and PU Effects - Uniformity of Flow and Deposition 2-3
2.2.2 DAF for Extrarespiratory Effects - Blood:Gas Partition Coefficients (Hb/g) 2-3
2.3 The RGDR for the Tracheobronchial (TB) Region - RGDRTB 2-4
2.4 The RGDR for the Pulmonary (PU) Region - RGDRPU 2-5
2.5 The DAF for the Extrarespiratory Region (ER) Region - Hb/g 2-7
2.6 Children's Dosimetry 2-8
3 ADVANCES 3-1
3.1 Inhalation Dosimetry - Concepts, Design, and Results 3-1
Table 3-1 Modeled Predictions of Amount of O3 and SO2 Absorbed at Various Sites in
the Airways of Three Species 3-5
3.2 Inhalation Rates 3-6
Table 3-2 Distribution Percentiles of Physiological Daily Inhalation Rates for
Free-Living Normal-Weight Males and Females Aged 2.6 Months to 96
Years 3-8
Table 3-3 Distribution Percentiles of Physiological Daily Inhalation Rates on a per
Body Weight Basis for Free-Living Normal-Weight Males and Females
Aged 2.6 Months to 96 Years 3-10
Table 3-4 Probabilistic 24-hr Breathing Rate Estimates 3-13
Table 3-5 Comparison of Mean 24-hour Inhalation Rates (nrVday) Determined Using
Time-Activity-Ventilation (TAV), Metabolic Energy Conversion (MEC) or
Doubly Labeled Water (DLW) Approaches 3-14
3.3 TB Region 3-14
3.3.1 Flow and Deposition 3-14
Figure 3-1 Distributions of Deposition Enhancement Factor (DEF) for MTBE Vapor
with Qin = 30 L/min in the Bifurcation Airway Models 3-17
Figure 3-2 Distributions of Deposition Enhancement Factor (DEF) for Ethanol Vapor
with Qin = 30 L/min in the Bifurcation Airway Models 3-17
3.3.2 Advances in TB Inhalation Dosimetry Modeling 3-20
3.4 PU Region 3-21
3.4.1 Flow and Deposition 3-21
Figure 3-3 Dynamic Ventilation 3He MRI After Inhalation of Hyperpolarized 3He Gas 3-22
Figure 3-4 Simulated Flow Velocities from CFD Solutions in an Alveolar Sac Model 3-23
3.4.2 Advances in Quantitation in Lung Geometry 3-24
Table 3-6 Estimates of Right, Left, and Total Lung Volumes in Male Wistar Rats 3-25
Table 3-7 Summary Data on Human Lung Alveolar Number and Volume 3-26
Table 3-8 Summary Table of Measures from Right Lungs of Human Cadavers 3-27
Table 3-9 Functional and Morphological Features of the Developing Male Rat Lung 3-28
-------
3.5 Extrarespiratory (ER) Region 3-28
3.5.1 Methods and Advances for Estimating Blood:Air Partition Coefficients 3-28
Table 3-10 Mean Ratios and Standard Deviations (Rmean ± SD) of Predicted to
Experimentally Derived Human and Rat Blood:Air Partition Coefficients, P,
and Mean Absolute Differences Between Predicted and Experimental
Values of log P (E) for a Reference Set of Unreactive Volatile Organic
Chemical Vapors 3-30
3.5.2 Quantitation using Inhalation PBPK Models 3-31
Table 3-11 Compilation of Blood:Air Partition Coefficients used in Inhalation PBPK
Models for Animal to Human Interspecies Extrapolation 3-33
3.5.3 HECER Derivation - PBPK and Hb/g (RfC Methods) Comparison 3-34
Table 3-12 Estimations from Inhalation PBPK Models of Human Equivalent
Concentrations (HECs) from Effect Levels and Internal Dose Measures in
Laboratory Animals 3-35
Table 3-13 Comparison of Approaches for Calculating Human Equivalent
Concentrations (HECs) for Several Gases with Effects in the
Extrarespiratory Region (ER) 3-37
3.6 Children's Inhalation Dosimetry 3-37
3.6.1 Introduction and Focus 3-37
3.6.2 PBPK Models 3-40
Table 3-14 Human kinetic adjustment factors (UFH-TK) obtained for inhalation exposure
in each population group using a dose surrogate of 24 hour AUCpc 3-42
Table 3-15 Air Concentration of Chloroform at Various Ages and Genders
Corresponding to Threshold of Damage in Human Liver and Kidney 3-43
Table 3-16 Age-Dependent and Gender-Specific Dose Metric Comparison of Inhaled
Isopropanol 3-45
Table 3-17 Tissue Concentrations in Various Compartments Expressed as Adult/Child
(1 to 2 years old) Ratios for 8 Different Gases 3-46
3.6.3 Respiratory Tract Flow Models 3-48
Table 3-18 Summary Listing of Findings on Morphometry and Gas Flow/Uptake
Simulations for Human Nasal Cavities 3-49
Table 3-19 Selected Morphologic and Simulated Modeling Results of Hydrogen Sulfide
Dosimetry in Casts of Human Nasal Cavities 3-50
3.6.4 Respiratory Tract Growth 3-52
3.6.5 Child Data Collations 3-56
Figure 3-5 Alveoli count per lung as a function of age 3-57
Table 3-18 Lung Weights (Right and Left) of Males and Females from Birth to
Adulthood 3-58
4 SUMMARY AND CONCLUSIONS 4-1
Table 4-1 Respiratory Tract Surface Areas and Volumes for Various Species
comparing the 1994 RfC Methods and this Report 4-3
5 REFERENCES 5-1
APPENDIX A. SUMMARY AND DISPOSITION OF INDEPENDENT EXTERNAL PEER
REVIEW COMMENTS A-1
APPENDIX B. DEFINITIONS FOR VENTILATORY VOLUMES B-1
APPENDIX C. LITERATURE SEARCH STRATEGY C-1
VI
-------
GLOSSARY
Chronic Exposure - Multiple exposures occurring over an extended period of time, or a significant fraction of the
animal's or the individual's lifetime.
Computational fluid dynamics (CFD) - (Three-dimensional) - A branch of fluid mechanics that uses numerical
methods and algorithms to solve and analyze problems of fluid flows. Flows may apply to liquid and gases,
including inspired and expired air, and are thus applicable to solving flows within the respiratory tract. The
fundamental bases of any CFD problem are the Navier-Stokes equations, which define any single-phase fluid flow.
These equations can be simplified by removing terms describing viscosity to yield the Euler equations.
Diffusivity (gas) - The transport of matter from one point to another by random molecular motions to become
equalized with respect to concentration. For gases, rates of diffusion increase with the temperature and are inversely
proportional to the pressure. The interdiffusion coefficients of gas mixtures are almost independent of the
composition. Kinetic theory shows that diffusion of a pure gas is inversely proportional to both the square root of the
molecular weight and the square of the molecular diameter.
Dosimetric Adjustment Factor (DAF) - A multiplicative factor used to adjust observed experimental or
epidemiological data to human equivalent concentration (HEC) for assumed ambient scenario. See also regional gas
dose ratio (RGDR).
Flux - The rate of flow of energy, gas or particles across a given surface.
Gas - Term referring to a compressible fluid phase of a substance. Fixed gases are gases for which no liquid or solid
can form at the temperature of the gas, such as air at ambient temperatures.
Henry's Law Constant - The law can be expressed in several equivalent forms, a convenient form being: Cg = HC1
where Cg and Cl are the gas-(g) and liquid-(l) phase concentrations. The constant (H) is the ratio at equilibrium of
the gas phase concentration to the liquid-phase concentration of the gas (i.e., moles per liter in air/moles per liter in
solution).
Identical Path Model - (One- or two-dimensional) - An anatomical mathematical model where all paths from the
nose or mouth entrance to the alveolar sacs are identical.
Inhalation Reference Concentration (RfC) - An estimate (with uncertainty spanning perhaps an order of
magnitude) of a continuous inhalation exposure to the human population (including sensitive subgroups) that is
likely to be without an appreciable risk of deleterious noncancer health effects during a lifetime. The inhalation
reference concentration is for continuous inhalation exposures and is appropriately expressed in units of mg/m3.
Kg - The overall mass transfer coefficient describing movement of gas from the air phase into the liquid phase of the
respiratory tract (see also MTC).
kg - The gas phase mass transfer coefficient describing movement of gas from the gas phase to liquid/tissue
boundary (see also MTC).
Lowest-Observed-Adverse-Effect Level (LOAEL) - The lowest exposure level at which there are statistically
and/or biologically significant increases in frequency or severity of adverse effects between the exposed population
and its appropriate control group.
Mass Transfer Coefficient (MTC) - A diffusion rate constant that relates the mass transfer rate, mass transfer area,
and concentration gradient as driving force between and through phases. These coefficients may also be viewed in
terms of resistance to flow and movement. For purposes of this report (with phases of gas and solid) MTC requires
units of mass, time, distance, and concentration: mol/(s-m2), mol/m3, or m/s. Examples of MTCs used in this report
relate to movement of gases in the respiratory tract. They include the MTC designated for the gas phase only, kg, and
an overall MTC inclusive of both the gas and liquid phases, Kg.
VII
-------
No-Observed-Adverse-Effect Level (NOAEL) - An exposure level at which there are no statistically and/or
biologically significant increases in the frequency or severity of adverse effects between the exposed population and
its appropriate control. Some effects may be produced at this level, but they are not considered as adverse, nor
immediate precursors to specific adverse effects. In an experiment with several NOAELs, the assessment focus is
primarily on the highest one for a given critical effect, leading to the common usage of the term NOAEL as the
highest exposure without adverse effect.
Physiologically-Based Pharmacokinetic (PBPK) Modeling - (Zero-dimensional) - A mathematical modeling
technique for predicting the absorption, distribution, metabolism and excretion of a compound in humans and other
animal species. PBPK models strive to be mechanistic by mathematically transcribing anatomical, physiological,
physical, and chemical descriptions of the phenomena involved in complex pharmacokinetic processes. These
models have an extended domain of applicability compared to that of classical, empirical function based,
compartmental pharmacokinetic models.
Portal-of-Entry Effect - A local effect produced at the tissue or organ of first contact between the biological system
and the toxicant.
Regional Gas Dose (RGDr) - The gas dose (mg/cm2) of respiratory tract surface area (per minute) calculated for the
respiratory tract region of interest (r) as related to the observed toxicity (e.g., calculated for the tracheobronchial
region for an adverse effect in the conducting airways). Regions of interest may be the extrarespiratory (ER),
tracheobronchial (TB), or pulmonary (PU).
Regional Gas Dose Ratio (RGDRr) - The ratio of the deposited gas dose in a respiratory tract region (r) for the
laboratory animal species of interest to that of humans. This ratio is used to adjust the observed gas exposure level
for interspecies dosimetric differences.
Sherwood Number (Sh) - A dimensionless term for the ratio of convective to diffusive forces. The air-phase mass
transfer coefficient can be defined in terms of the Sherwood number.
Uncertainty Factor (UF) - One of several, generally 3- to 10-fold factors, used in operationally deriving the
inhalation reference concentration (RfC) from experimental data. UFs are intended to account for (1) the variation in
sensitivity among the members of the human population, (2) the uncertainty in extrapolating laboratory animal data
to humans, (3) the uncertainty in extrapolating from data obtained in a study that is of less-than-lifetime exposure,
(4) the uncertainty in using LOAEL data rather than NOAEL data, and (5) the inability of any single study to
adequately address all possible adverse outcomes in humans. The RfC Methods (U.S. EPA. 1994) use 3 for the UF
for interspecies extrapolation due to the incorporation of default dosimetric adjustments.
Vapor - A term referring to a gas phase at a temperature below the critical temperature of the substance where the
same substance can also exist in the liquid or solid state. If the gas is in contact with the liquid or solid phase, the
two phases will be in a state of equilibrium. This report is intended to consider those agents present as gaseous
vapors at ambient temperatures.
Vlll
-------
ABBREVIATIONS AND
ACRONYMS
1,1,1-TCE 1,1,1-trichloroethane
1,2,4-TMB 1,2,4-trimethylbenzene
2-BE 2-butoxyethanol
2-ME ethylene glycol monomethyl ether
2D two dimensional
3D three dimensional
A overall or summation hydrogen
bond acidity
ADAM aerosol-derived airway
morphometry
ADC apparent diffusion coefficient
AMET amount metabolized per 24 hour
period
ASPM axisymmetric single path model
AUC area under the curve
AUCpC area under the parent compound's
arterial blood concentration vs.
time curve
AV alveolar volumes
B overall or summation hydrogen
bond basicity
bb bronchioles
BB tracheobronchial
BW body weight
CA arterial blood concentration
CATE carbon tetrachloride
CFD computational fluid dynamic
CFDM computational fluid dynamic
modeling
Cmax maximum concentration
CO2 carbon dioxide
CT computed tomography
CV venous blood concentration
d day
D diffusivity
D2O deuterium oxide
DAF dosimetric adjustment factor
DEF deposition enhancement factor
DF deposition fraction
DLCO diffusion capacity of carbon
monoxide
DLW doubly labeled water
E solute excess molar refractivity
with units of (dm3 mol"1 )/10
EAD effective air space dimension
EBZ ethylbenzene
ECG energy cost of growth
EPA Environmental Protection Agency
ER extrarespiratory or systemic
F flux fraction
fp fractional penetration
FQPA Food Quality Protection Act
FVC forced vital capacity
g gram
GCMS gas chromatography mass
spectrometry
GSH glutathione
3He hyperpolarized helium-3
Hb/g blood:air or blood:gas partition
coefficient
(Hb/g)A animal blood:gas partition
coefficient
(Hb/g)H human blood:gas partition
coefficient
H,/g tissue: gas partition coefficient
HEC human equivalent concentration
HP hyperpolarized
hr hour
K absorption parameter
kg kilogram
kg gas-phase mass-transport
coefficient
Kg overall mass transfer coefficient
k\ liquid/tissue phase mass transport
coefficient
L log of the gas-hexadecane partition
coefficient (unitless) at 25 °C
LFER linear free energy relationship
Mel methyl iodide
mM millimolar
MR magnetic resonance
MRI magnetic resonance imaging
MTBE methyl tertiary butyl ether
NAS National Academy of Science
O2 oxygen
O3 ozone
PBPK physiologically-based
pharmacokinetic
PCE perchloroethylene
PD pharmacodynamic
PDIR physiological daily inhalation rate
PGME propylene glycol methyl ether
PGMEA propylene glycol methyl ether
acetate
PK pharmacokinetic
POD point of departure
PODadj point of departure duration adjusted
ppb parts-per-billion
ppm parts-per-million
PU pulmonary
Qaiv alveolar ventilation rate
Qb regional blood flow
R radius of the airway
RfC reference concentration
RGDR regional gas dose ratio
S solute dipolarity/ polarizability
SA surface area
Sh Sherwood number
SO2 sulfur dioxide
Sp available surface area
STP standard temperature and pressure
tin half-life
IX
-------
TAV time-activity-ventilation
TB tracheobronchial
TCE trichloroethylene
TDEE total daily energy expenditure
TLC total lung capacity
UBA upper bronchial airway
UFH uncertainty factor for
interindividual human
variability
URT upper respiratory tract
v viscosity
Vd volume of distribution
VE ventilation rate or minute volume
VLDfrans volume of gas required to reach
transitional bronchioles into the
lung
VQ ventilator equivalent ratio
wk week
XYL m-xylene
yr year
-------
EXECUTIVE SUMMARY
The purpose of this report is to evaluate new scientific developments and advancements
in gas dosimetry focusing on tracheobronchial (TB), pulmonary (PU), and
systemic/extrarespiratory (ER) inhalation dosimetry related to the U.S. EPA's 1994
Methods for Derivation of Inhalation Reference Concentrations and Applications of
Inhalation Dosimetry (U.S. EPA. 1994). hereafter RfC Methods. Particular emphasis is
given on animal to human extrapolation and inhalation dosimetry in children. The goal of
this project is to provide information necessary for ensuring that methods and guidance
used and implemented by EPA in inhalation risk assessment reflects the
state-of-the-science. Overall, the scientific advances support and, in some cases, actually
build further upon the approaches of the current methodology for gas dosimetry in the
tracheobronchial (TB), pulmonary (PU) and extrarespiratory (ER) regions. With regards
to gas dosimetry, there appears to be insufficient quantitative evidence to modify the RfC
Methods specifically for children; however, in some cases, chemical-specific information
may warrant consideration of alternative approaches or adjustments to account for this
lifestage. It is anticipated that information will continue to become available to further
inform this issue.
This report summarizes the status of specific inhalation dosimetry procedures for gases as
outlined in RfC Methods, and reviews recent scientific advances in gas dosimetry related
to these procedures. These procedures are used predominately for interspecies
extrapolation, typically from laboratory animal inhalation exposures to humans. The
specific procedures addressed in this report are those used for the tracheobronchial (TB)
and pulmonary (PU) regions of the respiratory tract and the procedure used for the
systemic or extrarespiratory (ER) region. In addition, this report presents, reviews and
discusses information and data on inhalation dosimetry throughout the respiratory tract of
children pertaining to the derivation of an RfC. For the purposes of this report the
scientific literature was searched from 1985 (about 10 years prior to the issuance of RfC
Methods) to April 30, 2011. This report is a source document for evaluation and potential
future revisions to and updating of the gas dosimetry procedures outlined in the RfC
Methods, as they relate to the state-of-the-science. However, it is intended neither to
displace the current procedures for gas dosimetry in RfC Methods nor to reflect on use of
these procedures for assessments currently posted in IRIS.
The studies identified in this update addressing overall concepts and approaches for
portal-of-entry gas dosimetry in the TB and PU regions of the airways support the
principles and procedures in RfC Methods. In some cases these studies suggest and
provide examples of further refinement within the existing dosimetry modeling
framework of the RfC Methods through development and application of mass transfer
coefficients as regional measures of gas uptake. Alternative gas dosimetry procedures
1-1
-------
published using simplified airway models inclusive of the TB and PU regions arrive at
tissue metrics that support the approach of the RfC Methods. Conceptually, the methods
for extension of state-of-science flow models to the TB and PU areas promise further
refinement and resolution for inhalation gas dosimetry.
The default procedures of RfC Methods for TB and PU gas dosimetry employ the basic
components of inhalation, regional surface areas of the TB and PU regions, and air flow
to these surface areas. Recently, refined methods for measurement of inhalation rates in
humans have been developed. The advent of the doubly labeled water (DLW) technique
in estimation of physiological daily inhalation rates (PDIR) has provided resolutions to
concerns regarding inhalation patterns of free-living individuals across all age groups
including children. DLW-based PDIR values are currently included in the Child-Specific
Exposure Factors Handbook (U.S. EPA, 2008). and are being proposed for inclusion in
other key Agency documents, including the external review draft of the Exposure Factors
Handbook (U.S. EPA. 2009a). for all ages including children.
Recent advances in understanding the airflow to the TB and PU regions have been made.
Human flows in the PU region generally support the assumption of uniformity as
methodological advances and increased resolution of several in vivo imaging techniques
indicate highly uniform and homogenous flows in the alveolar regions for humans. On
the other hand, examination of the TB regions with human models and advanced dynamic
fluid flow programs reveal a degree of nonuniformity of flow for this region although
apparently not to the extent that has been documented for the upper airway. Marked
advances in morphometry of these regions for both animals and humans are being
achieved with the development and application of stereology. These techniques,
described as the estimation of higher dimensional information from lower dimensional
samples, have and continue to provide more accurate estimates of flow to regions of the
respiratory tract. A detraction regarding these advancements, however, is that most apply
to humans and comparable information in the critical comparative component of
interspecies extrapolation, the laboratory animal, lags.
The principal determinative component for dosimetry of the ER region is the highly
chemical-specific blood:air partition coefficient (Hb/g). The Hb/g is also a sensitive
component of physiologically based pharmacokinetic (PBPK) models, models that are of
ever increasing utility to the risk assessment community. Different techniques and
approaches have been proposed to derive these values for both human and laboratory
animals. A set of key reviews (Abraham et al.. 2005; Payne and Kenny. 2002). compiling
and analyzing results from several of these approaches, makes several conclusions
relevant to dosimetry and risk assessment, including that there appears to be no difference
between human and laboratory values for a prominent subgroup of toxic gases, the
volatile organics. Examination and compilation of Hb/gs in published inhalation PBPK
models configured for interspecies comparisons was also undertaken. These findings also
1-2
-------
provide evidence that the current default dosimetry approach of RfC Methods that uses
Hb/gS as a basis of dosimetry for the ER region, or systemic toxicity, remains valid.
Recent research relevant to inhalation gas dosimetry in children was found to closely
follow the recommendations and guidance of the National Academy of Sciences (NAS)
on children's risk (NRC. 1993). These recommendations include use of PBPK models to
explore and evaluate potential child susceptibility as well as the related effort to generate
accurate measurements and parameters to be used in these models. A number of studies
were reviewed that followed from these activities including development of PDIRs,
morphometry of conducting airways and lung tissue using advanced state-of-science
techniques, as well as respiratory tract function using new highly refined in vivo analyses
of airway function. Sophisticated flow models that use these refined measures and that
are capable of examining uptake differences of gases in the upper airways of both adults
and children are also presented and discussed. Several PBPK models have been
configured and parameterized with results from these newer techniques to consider child
versus adult dosimetry. Although few data sets and models pertaining to gas dosimetry in
children exist, the spectrum of methods and approaches is robust. In some scenarios, the
available methods and approaches are fairly uniform in their indications of potential
higher inhaled doses in young children, which may be in the range of up to 2-fold more
than adult. Individual instances exceeding this range are also found but no apparent
pattern appears to be associated with these occurrences. This range is within that built
into RfC Methods using the human interindividual uncertainty factor (UFH) to account for
pharmacokinetic and pharmacodynamic variability and for consideration of potential
sensitive populations and lifestages including children. It should be noted that this finding
is very similar to that of the NAS (1993).
1-3
-------
1 INTRODUCTION AND PURPOSE
The purpose of this report is to evaluate new scientific developments and advancements
in gas dosimetry focusing on tracheobronchial (TB), pulmonary (PU), and
systemic/extrarespiratory (ER) inhalation dosimetry related to the current methodology
used by EPA. Particular emphasis is given on animal to human extrapolation and
inhalation dosimetry in children. The goal of this project is to provide information
necessary for ensuring that methods and guidance used and implemented by EPA in
inhalation risk assessment reflects the state-of-the-science.
The current guidance, Methods for Derivation of Inhalation Reference Concentrations
and Application of Inhalation Dosimetry [(U.S. EPA. 1994), hereafter RjC Methods], was
made publicly available in 1994. RfC Methods is used by EPA in developing reference
concentrations (RfCs) for the Agency's IRIS (Integrated Risk Information System) public
database. RfC Methods addresses broad areas of risk assessment but focuses especially on
inhalation dosimetry and provides methods for converting inhalation exposures in
laboratory animals to human equivalent exposure concentrations (HECs). Sections
devoted to inhalation dosimetry are extensive including information on respiratory tract
function and anatomy, physiology, and pathology in humans and typical laboratory
animals. Other sections explore the properties of inhaled agents (e.g., particles and
gases). In critical areas where important observations and application processes were not
yet available, reasoned approaches based on scientific theory were given. These data,
theories, and empirical observations were then synthesized into methods applicable to
RfC derivation. These methods are also discussed in A Review of the Reference Dose and
Reference Concentration Processes (U.S. EPA. 2002).
In 2009, the document Status Report: Advances in Inhalation Dosimetry of Gases and
Vapors with Portal of Entry Effects in the Upper Respiratory Tract (U.S. EPA. 2009b)
was completed. The purpose of this first status document (hereafter Status I) was to
evaluate scientific developments since 1994 and advancements in the area of gas
dosimetry, focusing on the upper respiratory tract, and to determine how this information
might inform our approach to gas dosimetry. The focus of the evaluation was based on
the results from an expert panel assembled in 2005 and tasked with reviewing the
state-of-the-science of inhalation dosimetry in relationship to the RfC Methods. Status I
focused on portal-of-entry effects in the upper respiratory tract which, according to RfC
Methods, comprises only the nasal tract or extrarespiratoy region (ER). While Status I did
not consider advances in the area of gas dosimetry related to susceptible lifestages and
populations, it is addressed in this report.
The current report (hereafter Status II) serves a purpose similar to Status I (U.S. EPA.
2009b) but focuses on the remaining regions comprising the lower respiratory tract or
thoracic (TH) region as designated by RfC Methods, the TB and PU regions. Also
1-4
-------
included in this report is the dosimetry used by RfC Methods for effects of gases in all
sites other than the respiratory tract, referred to by RfC Methods as the extrarespiratory
(ER) region. The ER region is considered to be other tissues that may be exposed to the
gas once it enters systemic circulation via respiratory tract uptake.
This report, Status II, also evaluates new data and approaches for inhalation dosimetry of
gases in children. This area was included in recognition of the Agency's commitment to
ensuring that EPA actions are protective of children, given the potential for sensitivity of
early lifestages to some environmental exposures. RfC Methods currently considers
children within the intraspecies variability uncertainty factor intended to account for
intrahuman variability in response among sensitive populations and lifestages within the
population but devotes no further analysis to the matter. The final portion of this report
focuses on information and investigations that follow the recommendations of the
National Academy of Sciences (NRC. 1993) to more thoroughly characterize
susceptibility of children in relation to the adult population.
Status //is a source document for evaluation and potential future revisions to and
updating of the gas dosimetry procedures outlined in the RfC Methods, as they relate to
the state-of-the-science. However, it is intended neither to displace the current procedures
for gas dosimetry in RfC Methods nor to reflect on use of these procedures for
assessments currently posted in IRIS.
1-5
-------
2 INTERSPECIES GAS DOSIMETRY IN THE RFC -
THE REGIONAL GAS DOSE RATIO (RGDR) AS
A DOSIMETRIC ADJUSTMENT FACTOR (DAF)
The purpose of dosimetry is to calculate an internal dose metric (e.g., target tissue dose,
steady-state blood concentration) that results from an experimentally applied laboratory
animal dose (or concentration) and estimate a human exposure dose (or concentration)
that would result in an equivalent dose metric. Below, the steps currently used for the
dosimetric adjustment procedure for deposition in the TB and PU regions, as well as the
ER adjustment, are reviewed.
Equation 2-1 is a general equation that may be applied to estimate a human equivalent
concentration (FiEC) from an animal point of departure (POD). The POD corresponds to
an exposure concentration at which a particular health effect is observed. The subscript
"ADJ" refers to an adjustment that converts the POD from the actual exposure
concentration to an average daily exposure concentration for a continuous exposure
scenario. This adjustment will not be considered further as it is not a focus of this report.
PODnec (mg/m3) = PODADj (mg/m3) x DAFr
Equation 2-1
The DAFr is the dosimetric adjustment factor for a respiratory tract region, where r in this
report refers to TB, PU, or ER. As can be seen here, the DAF is a factor used to convert
an average exposure concentration for a particular laboratory species to a constant
exposure concentration estimate for humans that would result in the same delivered dose,
the HEC. When evaluating toxicity following inhalation exposure, in particular, dose
refers to the mass of toxicant absorbed across an airway surface per unit surface area.
Also, for such portal-of-entry (e.g., TB and PU) effects, the DAF is termed the regional
gas dose ratio (RGDR) and depends on animal to human ratios of two important
parameters: minute volume or ventilation rate (VE), and surface area (SA) of the target
region. When evaluating ER effects, the DAF depends on the ratio of animal and human
blood:gas partition coefficients (Hb/g).
2-1
-------
2.1 Current Applications of the DAFs - RGDRTB, RGDRPU, and Hb/g
RfCMethods provides species-specific values or algorithms to generate values for the
parameters required to derive the RGDR for the TB and PU regions. The listing of SA
values and the coefficients used to generate VE given in RfC Methods are presented below
in Table 2-1 and Table 2-2. It is these values and algorithms that are to be part of the
evaluation of new evidence and advances offered by this document. Also provided below
is an example listing of animal and human Hb/g values as well as their A/H ratio
(Table 2-3), this being the basis for the DAF applied for the ER region.
Table 2-1 Surface Areas for the Tracheobronchial (TB) and Pulmonary (PU) Regions of the
Respiratory Tract in Various Species
Species
Human
Mouse
Hamster
Rat
Guinea pig
Rabbit
TB (cm2)
3,200
3.5
20.0
22.5
200.0
300.0
Source
Mercer et al. (1994b)
Mercer et al. (1994b)
Yu and Xu (1987)
Mercer et al. (1994b)
Schreider and Hutchens
(1980)
Kliment (1973)
PU (cm2)
540,000
500
3,000
3,400
9,000
59,000
Source
Mercer et al. (1994a)
Geelhaar and Weibel (1971);
Mercer et al. (1994a)
Lechner (1978)
Mercer etal. (1994a)
Tenney and Remmers (1963)
Gehr et al. (1981)
Source: (U.S. EPA, 1994)
Table 2-2 Intercept (b0) and Coefficient (bt) Values Used to Calculate Default Minute Volumes
Based on Body Weight3
Species
Rat
Mouse
Hamster
Guinea pig
Rabbit
'Calculation of default minute volume
Source: (U.S. EPA, 1994)
bo
-0.578
0.326
-1.054
-1.191
-0.783
based on body weight is conducted using the following algorithm:
bi
0.821
1.050
0.902
0.516
0.83
((log VE) = b0 + b, x log (BW))
2-2
-------
Table 2-3 Some Example BloockAir Partition Coefficients (Hb/g) in Humans and Rats Expressed
as a Ratio, Animal/Human
Chemical Human (Hb/g)
Chloroform
Dichloromethane
Carbon tetrachloride
Chlorodibromomethane
Chloroethane
1,1-Dichloroethane
1,2-Dichloroethane
1,1,1-Trichloroethane
1,1,2-Trichloroethane
1,1,1,2-Tetrachloroethane
1,1,2,2-Tetrachloroethane
Hexachloroethane
Methylchloride
Source: Reprinted with permission of Elsevier©; Gargas et al.
6.85
8.94
2.73
52.7
2.69
4.94
19.5
2.53
35.7
30.2
116
52.4
2.48
(1989)
Animal (rat, Hb/g)
20.8
19.4
4.52
116
4.08
11.2
30.4
5.67
58
41.7
142
62.7
2.47
Animal/Human
3.0
2.2
1.7
2.2
1.5
2.3
1.6
2.2
1.6
1.4
1.2
1.2
1.0
2.2 Assumptions for the Current Application of RGDR Procedures
2.2.1 DAF for TB and PU Effects - Uniformity of Flow and Deposition
As outlined in the RfCMethods and described below, the dosimetric adjustment
procedure for both the TB and PU regions are basically identical. The default procedure
is based on the following assumptions: (1) the flow of inspired air (i.e., VE) into the
region of interest (TB or PU) is uniform and even, (2) the surface area in the region of
interest is uniform, and (3) as a consequence of these paired assumptions, there is
uniform deposition of gas over the surface areas of these regions in the respiratory tract.
2.2.2 DAF for Extrarespiratory Effects — Blood:Gas Partition
Coefficients (Hb/g)
As outlined in the RfC Methods, a number of assumptions exist for the application of this
DAF. In addition to making the assumption that differences will exist between species for
the basic biological component of Hb/g, blood, assumptions also are made that similarities
will exist between species. These assumptions include that (1) chronic laboratory animal
exposure scenarios are equivalent to human lifetime exposures, (2) the human
time-integrated arterial blood concentration is less than or equal to that of the exposed
laboratory animal such that (3) laboratory animal time-averaged arterial blood
2-3
-------
concentration is equal to the human equilibrium arterial blood concentration. It is also
emphasized in RfC Methods that the equilibrium referred to here is that portion of the
exposure period that is under conditions of "periodicity", i.e., the periodic steady state
concentration versus time profile is the same for every week. RfC Methods further states
that conditions of periodicity should be met during "most" (elsewhere indicated as 90%)
of the exposure duration.
2.3 The RGDR for the Tracheobronchial (TB) Region - RGDRTB
The DAF for the TB region is the "regional gas dose, tracheobronchial" ratio (RGDRTB).
It is constructed with species-specific ventilation rates (or minute volumes) and surface
areas for the TB region (U.S. EPA. 1994).
The equation for deriving a default RGDRTB for reactive and water soluble (e.g.,
Category 1) gases as it appears in RfC Methods (U.S. EPA. 1994) (Equation 4-22 and
Appendix I, Equation 1-24) is shown in Equation 2-2, where VE is the minute ventilation
rate (mL/min) and SATO the surface area (cm2) of the TB region for laboratory animals
(A) or humans (H).
fpE?A
Equation 2-2
The term to the right is the A/H ratio of the fractional penetration (fp) of the gas into the
extrarespiratory (ER) region. This term provides an estimate of the uptake of the inhaled
gas through this upstream region in animals relative to humans. As detailed in the RfC
Methods, fp values depend on the overall mass transfer coefficient, Kg (cm/min), surface
area (SA), and minute volume (VE) according to Equation 2-3:
KgSA
fp = e~T
Equation 2-3
Chemical and species-specific Kg values in biological settings are rare. Therefore, when
KgET is not known or cannot be reasonably approximated with data for either species, the
RfC Methods default assumes that the actual value of this term is 1, such that the left term
determines the RGDRTO, according to Equation 2-4:
2-4
-------
(V?./ \
or no -\ /SATB/
KbL)KTB--^
Equation 2-4
This report addresses advancements in dosimetry to the TB region including overall
concepts and approaches to dosimetry, and in specifics for this region including minute
ventilation (VE), surface area (SA), and mass transfer coefficients (Kg). Shown below is
an example calculation of the DAF for the TB region using Equation 2-4 for a rat to
human extrapolation assuming a rat VE of 0.250 L/min and SA of 22.5 cm2 and a human
VE of 13.8 L/min and SA of 3,200 cm2.
fO.25 L/min/ \
_( /22.5cm2)A __,
mm
'/3200cm2jH
Equation 2-4 (example)
The calculation using these default parameters (U.S. EPA. 1994) results in a RGDRTB of
2.6 indicating that rats receive nearly 3 times more dose in the TB region on a per SA
unit basis than humans.
2.4 The RGDR for the Pulmonary (PU) Region - RGDRPU
The DAF for the pulmonary region is the "regional gas dose ratio, pulmonary" ratio
(RGDRpu). It is constructed with species-specific ventilation values and surface areas for
the PU region.
The equation for deriving a default RGDRPU for reactive and water soluble
(e.g., Category 1) gases as it appears in RfCMethods (U.S. EPA. 1994) (Equations 4-23,
4-25 and 4-28 and Appendix I Equations 1-35,1-43 and 1-46) is shown below as Equation
2-5 where Qaiv is the alveolar ventilation rate (mL/min) and SAPU the surface area of the
pulmonary region for laboratory animals (A) or humans (H) (cm2), and Kg for this region
of interest (PU) (cm/min). Mass transfer coefficients are rarely available and are
reasonably assumed to be very large for reactive and soluble gases, and Equation 2-5
reduces to Equation 2-6.
RGDRP,,= -^—?I 2. ^ '•'""",; _1±^ -l^ii
KgpybApu \ p;iiys!i _ ) rpET|j rpTBll
PI)SAp[i+ Qalv/j|
Equation 2-5
2-5
-------
RGDR -
Equation 2-6
The right two terms in Equation 2-5 and Equation 2-6 are the A/H ratios of the fractional
penetration (fp) of the gas into the extrarespiratory (ER) and tracheobronchial (TB)
regions. As explained above, these terms provide an estimate of the uptake of the inhaled
gas in these two upstream regions (prior to entrance into the PU region).
If the fp to ET and TB regions are unknown due to lack of data on Kg, and the Kgs in
animals are assumed to be equal to humans, and/or assumed to be equal to 1, the far left
term in Equation 2-6, determines the RGDRPU. By combining these default scenarios for
the fp terms in Equation 2-6, the RGDRPU can be computed from alveolar ventilation and
pulmonary surface areas according to Equation 2-7:
RGDRPU =
Equation 2-7
Alveolar ventilation (Qaiv) in the RGDRPU equations, is the appropriate parameter to use
as it refers to the gas that reaches the alveoli and takes part in gas exchange and excludes
that which does not, often referred to as alveolar dead space. In practice, however,
alveolar ventilation and alveolar dead space are difficult to measure and values are not
readily available even in the clinical setting (Appendix B). Minute volume, VE, is readily
measured and typically reported in epidemiological and laboratory animal studies. Thus,
the equation to determine the RGDRPU has been simplified through usage to the form
presented in Equation 2-8.
RrnR _
RGDR-
p,,
I /SAp,,JH
Equation 2-8
As noted above, because chemical and species-specific Kg values in biological settings
are rare, Equation 2-8 is the default utilized for gas exposure with effects in the PU
region to obtain the DAF and used in calculation of the HEC.
Several recent reports in the literature, however, have investigated, derived, and applied
Kgs for various compounds and for different regions of the respiratory tract. These reports
include Kg values for the PU and TB regions for formaldehyde (Overton et al., 2001), Kg
values for ethanol, nitric oxide, and water vapor in the lung (Condorelli and George.
2-6
-------
1999). derivation and comparison of Kg values in contrasting geometries (Madasu. 2007).
estimation of Kgs in terms of the Sherwood number and correlation with Reynolds and
Schmidts numbers (Zhang and Kleinstreuer. 2011). and an approach for derivation for
soluble and reactive vapors in lung tissues derived by combining the overall air-phase Kg
with analytical expressions for tissue-phase Kgs (Asgharian et al.. 2011). several of which
are discussed later in this report. Thus, incorporation of Kgs for both regional and fp
determinations for more refined and precise inhalation gas dosimetry may be possible in
the future.
This report addresses advancements in dosimetry to the PU region including overall
concepts and approaches to dosimetry, and in specifics such as minute ventilation (VE),
surface area (SA), and mass transfer coefficients (Kg). Shown below is an example
calculation of the DAF for the PU region using Equaiton 2-8 for a rat to human
extrapolation assuming a rat VE of 0.250 L/min and SA of 0.34 m2 and a human VE of
13.8 L/min and SA of 54 m2.
(0.25 L/min /
'
0.34m2 /A
- — =
1 /13.8 L/min, \
( '54 myH
Equation 2-8 (example)
The calculation using these default parameters (U.S. EPA. 1994) results in a RGDRPU of
2.9 indicating that rats receive nearly 3 times more dose to the PU region on a per SA
unit basis than humans.
2.5 The DAF for the Extrarespiratory Region (ER) Region - Hb/g
Gases with physicochemical properties that lessen their potential for effects in the
respiratory tract (e.g., nonreactivity and higher lipid versus water solubility) may at the
same time exhibit potential for significant uptake and accumulation in the blood where
they can cause toxicity at systemic or remote (extrarespiratory, ER) sites. Based on these
properties and other kinetic properties governing how such gases may be expected to
distribute in the body, RfCMethods posits a fundamentally different DAF for gases that
have little or no potential for reactivity in the respiratory tract. This DAF is based on
assumptions of dose-response that are consistent with basic principles of kinetics and
toxicity applied to the scenario of systemic toxicity from an inhaled toxicant: toxicity is
directly related to the concentration of the agent at the target site; the concentration of the
agent at the target site is related to the concentration of the agent in the arterial blood at
equilibrium1; arterial blood concentration at equilibrium is related to its concentration in
the inspired air. The last link in this process, the partitioning of the agent from the
The gas or its concentration multiplied by time (C x T)
2-7
-------
inspired air into the blood at the alveolar endothelial interface, is determined by the
blood:air partition coefficient, Hb/g. Further, it is reasonably anticipated that as properties
of blood differ between species so will the partition coefficient itself. Thus, the DAF for
ER sites is based upon the species-specific (animal / human) ratio of the blood:air
partition coefficient (Hb/g) at equilibrium shown here in Equation 2-9:
"(HVi)H
Equation 2-9
Appendix J of the RfC Methods provides a mathematical derivation and application of
this procedure as well as a case study employing a physiologically-based
pharmacokinetic (PBPK) model parameterized for interspecies extrapolation.
In the RfC Methods, the DAF derivation for ER effects is based more on science policy
than on an empirical procedure. Further, this policy is bi-level; (1) where if Hb/g values
are unknown the default value for (Hb/g)A / (Hb/g)H = 1; (2) if (Hb/g)A is greater than (Hb/g)H
then a default value of 1 is also used. These procedures are justified by RfC Methods on
the animal human datasets that were available at the time (Gargas et al.. 1989). Gargas et
al. (1989). reported that for an appreciable number of volatile and nonvolatile agents the
(Hb/g)A was greater than the corresponding (Hb/g)H. These values are also shown above in
Table 2-3. This report, Status II, adds new evidence and advances from sources
identified.
2.6 Children's Dosimetry
Consideration of children's dosimetry in RfC Methods is discussed as a component of the
intraspecies uncertainty factor (UFH) that accounts for unknown pharmacokinetic and
pharmacodynamic differences. The default value of this UFH is 10 and is applied to
account for uncertainty and potential variations in susceptibility within the human
population (interhuman variability) and the possibility that the available database is not
representative of the population groups that may be most sensitive to the health hazards.
Early lifestages (including (1) embryo, fetus, and neonate and (2) young children) are
also listed in Table 2-4 of the RfC Methods as 2 of the 5 sensitive populations and
lifestages who, based on empirical observations or compromised physiological functions,
are assumed susceptible to toxicity elicited by certain groups of chemicals. It is discussed
further that certain populations and lifestages may be differentially susceptible, e.g.,
elderly individuals could be more susceptible to some chemicals and children to others.
RfC Methods acknowledged that very little is known about this important area of
population sensitivity and that guidance should be developed concerning the prevalence
2-8
-------
of sensitive populations and lifestages and the range of sensitivities in the general
population exposed to inhaled toxicants.
The Food Quality Protection Act (FQPA) of 1996 contains several requirements (directed
primarily toward the evaluation of pesticides) related to a new standard described in the
Act as "reasonable certainty of no harm." One of the specific requirements identified was
that the EPA considers the specific risk pesticides might have for infants and children. In
general, the manner in which this was to be accomplished was through application of
uncertainty factors based on an evaluation of information relevant to children. However,
this requirement engendered considerable interest, including interest in inhalation
dosimetry in children. On the whole, these evaluations, including conclusions by the
NAS (1993). indicate that for most chemicals the very large majority of people, including
children, respond sufficiently similarly so that the 10-fold intraspecies uncertainty factor
is adequate to cover any variability that may exist in the human population. However,
there are chemicals for which some humans may display a greater range of variability and
sometimes that variability appears age-related, with children exhibiting a greater degree
of sensitivity than adults. Further considerations of these matters are included in the
section on children's dosimetry (Section 3.6).
2-9
-------
3 ADVANCES
RfC Methods was a state-of-the-art document for inhalation dosimetry of gases in 1994.
Perhaps because of RfC Methods, further investigations into virtually all aspects relating
to inhalation dosimetry of gases have been undertaken. This section presents and
discusses many of these studies, results, and concepts that provide information to advance
our knowledge of gas dosimetry in the TB, PU, and ER regions.
3.1 Inhalation Dosimetry - Concepts, Design, and Results
Hanna et al. (2001) provided an update on the RfC Methods elaborating upon the
physicochemical framework and the mass transport theory that underlies the dosimetric
approach for gases. The authors made clear the distinction between the equilibrium
assumption in which gas transport is dependent on regional blood flow (e.g., the
endothelial blood flow of the lungs or PU region) or dynamic models in which gas
transport occurs across a concentration gradient near the transport barrier which is
essentially the membrane or boundary of another phase (e.g., as in the ET and TB
regions) and how this distinction was considered in formulating the dosimetric models in
RfC Methods.
For gases where transport is controlled by regional blood flow, it is the ventilation rate
that determines the alveolar gas-phase concentration; the equilibrium partition coefficient
of the gas then establishes the blood concentration in equilibrium with the alveolar gas
concentration. Transport to systemic (or remote) sites is then determined by tissue
specific blood flow. For a membrane barrier-limited transport model, equilibrium will not
be established between the air and the tissue and the tissue and blood. Instead,
concentration gradients exist in the air, the airway liquid lining layer, and the underlying
tissue. Depending upon the physicochemical properties, the absorption of an inhaled gas
may be controlled by gas-phase transport alone or the gas-phase in combination with the
transport through the liquid tissue phase. An individual mass transfer coefficient through
the gas phase (kg, cm/s) can be defined as the proportionality constant relating the flux of
the toxicant through the liquid phase, Ng (mass transported per surface area per second),
and the concentration difference between the central gas stream and that at the interface
of the gas/liquid interface (ACg) is described in Equation 3-1.
Equation 3-1
To incorporate the blood phase with the gas and liquid + tissue phase into an analysis of
transport, one defines the Kg. Kg is the proportionality constant between the absorptive
3-1
-------
flux at the gas/liquid interface and the overall concentration difference between the
central gas stream and the blood. It can be written in terms of the individual mass transfer
coefficients and the blood flow (Qb) as shown in Equation 3-2. Here, Fis the flux fraction
(to account for the fraction reacted in the previous transport phase), Sp the available
surface area (cm2), Ht/g the tissue to gas partition coefficient (Henry's Law value of the
gas, unitless), Hb/g is the blood to gas partition coefficient (unitless), ki is the mass
transfer coefficient in the surface-liquid/tissue phase (cm/s), and Qb the regional blood
flow (cm/s).
Ill FSp
— = — + -—-+ '
ke Ht/g^i Hh/8Qb
Equation 3-2
The development and availability of these mass-transport coefficients is necessary for the
full development of the procedures given in RfC Methods. In particular, mass transport
coefficients are important for realistic determination of (1) the gas dose to a region and
(2) the fractional penetration (fp) of gases through the various regions of the respiratory
tract. Also, use of an overall mass transport coefficient could retain a spatial
concentration distribution within the analysis rather than assuming well-stirred models in
which concentrations are spatially uniform. Such an approach provides a more realistic
basis from which to assess critical dose metrics, including peak concentration or peak
flux.
Other considerations regarding the development of these values exist. Whereas the kg can
be scaled to any gas of interest simply through a scaling of the gas diffusivity; kg is
highly dependent on the complexity of the airway morphometry and the airflow patterns
induced by the morphometry. For example, based on major differences in the upper
airway structures, one should expect important differences between rat and human mass
transfer coefficients in the TB region. In other words, the overall mass-transport
coefficient (Kg) can be highly species specific depending upon the nature of the gas.
Hanna et al. (2001) summarized and emphasized that development of robust data on
physicochemical and toxicokinetic properties for specific gases or classes of gases is
necessary to develop appropriate DAFs. Furthermore, obtaining anatomical and
physiological parameters for both sexes at various ages in both laboratory animal species
and humans could further extend the understanding of the determinants of the DAF, their
variability, and potential uncertainties when used to extrapolate across species.
Several approaches to estimate kgs have recently been reported (Asgharian et al.. 2011;
Zhang and Kleinstreuer. 2011; Schroeter et al.. 2010; Longest and Kleinstreuer. 2005;
Condorelli and George. 1999). In general, these approaches utilized the Sherwood
number (Sh), a dimensionless term of the ratio of convective to diffusive forces. The kg
3-2
-------
can be estimated using the relationship between the Sherwood number (Sh), diffusivity
coefficient (D), and radius of the airway (R):
ShxD
2R
Equation 3-3
All of approaches used to estimate kgs, utilized variations of the classical Ranz and
Marshal (1952a. b) approach, an empirical relationship between the Reynolds and
Schmidt numbers. The use of Sherwood numbers to calculate Kgs may allow for the
calculation of air-phase concentrations.
Flux-based dosimetry estimates for formaldehyde gas to the TB and PU regions were
developed by Overton and coworkers (2001). These estimates were inclusive of
calculations for overall mass transport coefficients for the lower respiratory tract. The
anatomical model for the lower airways was that of Weibel with ventilation and
respiratory tract dimensions (23 generations, including the upper airways) for a male of
specified age, height, and weight at functional residual capacity. The dosimetry model
employed a single path symmetric anatomical model versus a fully descriptive multi-path
model. The model developed by Overton and coworkers (2001) is referred to as an
"identical-path" model because all routes from the airway entrance to the terminal
airspaces are modeled identically. Assumptions for the single path symmetric model
included (1) all paths are identical, (2) for a given generation, the dimensions of one
single airway or airspace and the number of airways or airspaces in the generation
completely define the characteristics of that model generation, and (3) each airway or
airspace in the same generation or model segment has the same transport characteristics,
such as airflow rate, effective dispersion coefficient, and Kg. These assumptions made it
feasible to use the dosimetry model to estimate the flux of formaldehyde to tissue in each
airway passage, and airspace of the respiratory tract.
Formaldehyde transport and uptake for the generations comprising the TB and PU
regions were all approximated by a one-dimensional (ID) convection-dispersion equation
that accounted principally for molecular diffusion and absorption at the air-liquid surface.
The mass transfer coefficients in the nasal cavity were estimated by matching (within
0.2%) the percent uptake predicted by an existing three-dimensional (3D) computational
fluid dynamic (CFD) model of transport during inspiratory flow through an anatomically
accurate reconstruction of the nasal passages of an adult human male. The resulting
overall identical-path nasal airway mass transfer coefficients multiplied by the nasal
surface area, corresponding to minute volumes of 7.5, 9.0, 25, and 50 L/min (nasal
steady-state inspired flows rates of 15, 18, 50, and 46 L/min) were 1.68, 1.78, 2.98, and
2.83 cm/s, respectively. The Kg for the lower airways was calculated with extensive
consideration given to the kg component. The ID equation of mass transport was then
3-3
-------
applied to each generation airway and airway passage of a symmetric, bifurcating
respiratory tract anatomical model to provide predictions of local formaldehyde surface
fluxes (dose). The results obtained included the following: (1) more than 95% of the
inhaled formaldehyde is predicted to be retained by the respiratory tract for all activity
states simulated (a total of 4 different minute volumes); (2) in the lower respiratory tract,
surface flux (dose) is predicted to increase for several generations and then decrease
rapidly, (3) compared to first pulmonary generation fluxes, the first few tracheobronchial
generations fluxes are over 1,000 times larger; and (4) there is essentially no flux in the
alveolar sacs. The authors stated that the predicted fluxes based on the ID model for
those lower regions of the respiratory tract can be used in dose-response modeling. This
work provided information on mass transfer coefficients for the PU and TB regions
including their derivation, and demonstrating their use in a dosimetry model for these
regions.
Tsujino et al. (2005) developed a simplified mathematical airway model to simulate the
transport of gases (ozone [O3] and sulfur dioxide [SO2]) in airways of laboratory animals
(rats and dogs) and humans. The aim of the study was to examine through model
simulations how interspecies anatomical and physiological differences influence the
transport of the inhaled gases throughout the airways and alveoli. This comparison could
potentially provide an interspecies comparison of gas dosimetry in airways. The authors
acknowledge and document that nearly all input parameters used were assumed or scaled,
albeit with reasonable assumptions and allometry. Upper airways were assumed to be
straight tubes with length, diameter, volume, and surface area all mathematically derived
and scaled from existing information. The total length of the cylindrical upper airways
was obtained from the sum of the widths of multiple cross sections of the upper airways,
including the nasal cavity, pharynx, and larynx, of rats, dogs, and a human child
(described as 3 years old and weighing 13.6 kg). The dimensions from the human child
were subsequently converted to correspond to the values of a human adult weighing 70
kg for the simulations. The movement and velocities of gases within the airway were all
modeled based on convection with bulk flow simplified by assuming conservation of gas
volume without solving incompressible Navier-Stokes equations. Gas absorption at the
surface of the airways was determined by mathematical formulations incorporating the
basic elements of diffusivity and absorption constants (which included the absorption rate
at the airway surface) that were scaled to each gas. The basis for this scaling was actual
absorption data and concentration differences for these gases obtained by direct
measurements in dog airways. Real-time changes in gas concentrations were simulated at
three airway sites in each species: (1) the upper airway, (2) the lower airways consisting
of the 5th or 10th bronchial generation and (3) the alveolar region. The amount of O3 and
SO2 absorbed (modeled assuming a 10% concentration) at the airway surface was then
calculated. Interspecies comparison was also performed for the amount of gas absorbed
per body weight (g/BW), and for the corrected amount of gas absorbed per unit of airway
surface area (g/cm2). The results obtained for O3 and SO2 are shown in Table 3-1 below.
3-4
-------
Table 3-1 Modeled Predictions of Amount of O3 and SO2 Absorbed at Various Sites in the
Airways of Three Species
Parameter
Rats
Dogs
Humans
Ozone
Total absorbed amount (g/kg BW)
Upper airways (% of total)
Lower airways (% of total)
Alveolar region (% of total)
1.1 xicr7
73.9
23.4
2.7
1.46x1Q-7
80.7
16.3
3.0
0.847 x1Q-7
34.4
60.7
4.9
Absorbed amount per SA/unit time
Upper airways (g/cm2/ min)
Lower airways (g/cm2/ min )
Alveolar region (g/cm2/ min)
1.76x10'7
3.52x1(r8
1.56x10'13
0.89x10'7
1.29x1Q-8
1.23x10'13
1.31x10'7
7.58x1Q-8
1.40x10'13
Sulfur dioxide
Total absorbed amount (g/kg BW)
Upper airways (% of total)
Lower airways (% of total)
Alveolar region (% of total)
Source: Reprinted with permission of Informa
1.77x10'7
98.6
1.4
0.0
Healthcare©; Tsujino et al. (2005)
3.24x10'7
99.4
0.6
0.0
1.61x10'7
96.5
3.5
0.0
These simulations indicate that the amount of O3 absorbed per body weight throughout
the airways was lowest in humans (Table 3-1). However, the amount of absorbed O3 per
surface area in each airway were fairly equivalent in the upper airways and alveolar
regions, and were higher in humans in the lower airways, over 2 times that of rats. This
trend was noted also for SO2. Concentrations of SO2 in the lower airways and alveoli
were low in all species, which reflects the predicted rapid absorption of the gas in the
upper airway. Also, these simulations were for short periods of inhalation and relatively
high concentrations of these agents. It should be noted that many simplifications and
assumptions were necessary in order to accomplish the simulations. Some of these were
application of a simple three-compartment model of the airways and alveoli, without
specific consideration of the effects of different branching patterns on the airway surface
areas. Human airways are considered to be the least asymmetrical among mammalian
species especially with regard to daughter tube diameter ratio and daughter branch angle
ratio. Advances in respiratory tract imaging techniques can provide a more accurate
airway model. Coaxial diffusion of gas molecules was not taken into account in the
simulations, as it is well known that gas molecules in airways are transported by both
bulk flow and diffusion. Thus the modeled gas concentrations might not accurately
reflect actual concentrations, particularly in the peripheral airways and in the alveoli.
Nonetheless this study is of considerable value for further hypothesis testing regarding
the variations in the kinetics of inhaled gases among experimental animals and humans. It
numerically demonstrated that interspecies variations in anatomy and respiratory patterns
3-5
-------
cause significant differences in gas transport in the airways and alveoli of rats, dogs, and
humans.
Minard et al. (2006) proposes concepts on how the application of CFD modeling may be
developed for the lower areas of the respiratory tract, the TB and PU regions. CFD
models are developed through a sequence of steps, each requiring data intensive input.
Major components of this process are development of the graphical grid or mesh that
mirrors the surface over which the fluid flow is to be modeled. Initial mesh developments
were based on histological sections or digitized tissue sections and required immense
effort to complete. Also, such approaches have fundamental limitations in using an
inherently 2D (tissue sections) approach for compiling 3D descriptions of airway
architecture. Minard et al. (2006) investigated the feasibility of employing magnetic
resonance (MR) or computed tomography (CT) for this purpose. Ideally these methods
would allow a more detailed view of areas such as airway architecture in a digital format
that is inherently 3D and therefore ideal for subsequent computational analysis. These
authors reported results from visualization of airway architecture in the rat using proton
(1H) MR image data and describe computational techniques that could reduce the time
from image analysis to complete mesh to a few days. In addition, 3D 1H MR imaging of
rat pulmonary casts is reported as a step toward extending techniques to the lower regions
of the respiratory tract. The authors noted that in vivo pulmonary imaging with MR or CT
does not yet provide sufficient detail for visualizing more than the uppermost (~4-7
generations) pulmonary airways. Thus imaging of pulmonary casts by 1H MR is
proposed to provide an important alternative with the capacity to capture the actual
geometries for the development of CFD in a 3D setting. This report also explored the
feasibility of validating CFD predictions of velocity distributions with MR by imaging
hyperpolarized (HP) helium (3He) in a straight pipe with a diameter comparable to the rat
trachea at physiological flow rates. The comparison showed significant differences with
the features in the MR image showing considerable blurring in the axial flow pattern as
compared to the CFD solution. Thus, these proposals require more examination before
sound recommendations can be made. Future work should continue to explore, adapt and
refine imaging techniques to facilitate development of the concept of applying CFD
analysis to the lower respiratory tract airways.
3.2 Inhalation Rates
Regardless of the inhalation dosimetry approach considered, inspired and/or expired air
can be determinative components of an airborne dose to airway tissues for both humans
and laboratory animals. The dual subcomponents of the inspired air that relate to tissue
exposure are volume of inspired air (liters) and rate of breathing (e.g., frequency/min)
which are typically considered together as a factor in the parameter of minute volume,
VE, in liters/min. VE is the measure most often reported in a clinical setting under very
3-6
-------
brief conditions. For purposes of risk assessment, which typically consider long-term
exposures, knowledge of this parameter and variability in nonclinical settings, e.g.
free-living, is required. Major advancement in the area of inhalation rate measurement in
humans has been the application of methods used in free-living conditions, which would
allow for more accurate measures inclusive of this variability.
Two prominent approaches for inhalation rate measurement in a nonclinical setting
include (1) activity pattern questionnaires where oxygen consumption is calculated from
daily activity patterns/energy intake and (2) differential dilution of isotopes in water
administered orally as a bolus, usually over a two-week period. This latter method, the
doubly labeled water (DLW) method, measures oxygen lost through carbon dioxide
production. This is accomplished by administering oral doses of water that is radiolabeled
with both O18 and deuterium oxide (D2O). O18 is lost in time both through carbon dioxide
production and body water loss whereas D2O is lost only through body water. After a
period of 7-21 days (usually 14 days in humans), urine is collected and analyzed by
GCMS. The disappearance rate of deuterium reflects water output and the disappearance
of O18 water reflects water output plus carbon dioxide production rates. The carbon
dioxide production rate is then calculated by the difference between the two
disappearance rates.
The DLW method was used to calculate the physiological daily inhalation rates (PDIR)
for 2,210 individuals aged 3 weeks to 96 years (Brochu et al., 2006b) (Table 3-2). Total
daily energy expenditures (TDEEs; kcal/day) were determined from CO2 production rates
using classic respirometry formulas, in which values for the respiratory quotient
(CO2 produced /O2 consumed) were derived from the composition of the diet during the period
of time of each study. The DLW method also allows for measurement of the energy cost
of growth (ECG; in kcal /day) for children. TDEE and ECG measurements were
converted into PDIRs using the relationship developed by Layton (1993).
PDIR=(TDEE+ECG)xHxVQxlO;
Equation 3-4
where H is the O2 uptake factor (volume of O2 consumed at standard temperature and
pressure (STP) to produce 1 kcal of energy expended = 0.21 L O2) and VQ is the
ventilatory equivalent ratio of the minute volume (VE) at body temperature pressure
saturation to the O2 uptake rate (VO2 at standard temperature and pressure, dry air) = 27
(unitless) with the conversion factor of 10"3. The aggregate period of monitoring and
analysis for this study was more than 30,000 person days.
3-7
-------
Table 3-2 Distribution Percentiles of Physiological Daily Inhalation Rates for Free-Living
Normal-Weight Males and Females Aged 2.6 Months to 96 Years
Age Group
(years)
N
Body
Weight (kg)a
Mean ± SD
Physiological Daily Inhalation
Rate" (m3/day)
Percentilec
Mean + SD
5th
10th
25th
50th
75th
90th
95th
99th
Males
0.22 to < 0.5
0.5 to < 1
1to<2
2to<5
5to<7
7 to < 11
11 to < 23
23 to < 30
30 to < 40
40 to < 65
65 to < 96
32
40
35
25
96
38
30
34
41
33
50
6.7±1.0
8.8±1.1
10.6 ±1.1
15.3 ±3.4
19.8 ±2.1
28.9 ±5.6
58.6 ±13.9
70.9 ±6.5
71. 5 ±6.8
71.1 ±7.2
68.9 ±6.7
3.38 ±0.72
4.22 ±0.79
5.12 ±0.88
7.60 ±1.28
8.64 ±1.23
10.59 ±1.99
17.23 ±3.67
17.48 ±2.81
16.88 ±2.50
16.24 ±2.67
12.96 ±2.48
2.19
2.92
3.68
5.49
6.61
7.32
11.19
12.86
12.77
11.84
8.89
2.46
3.21
3.99
5.95
7.06
8.04
12.53
13.88
13.68
12.81
9.79
2.89
3.69
4.53
6.73
7.81
9.25
14.75
15.59
15.20
14.44
11.29
3.38
4.22
5.12
7.60
8.64
10.59
17.23
17.48
16.88
16.24
12.96
3.87
4.75
5.71
8.47
9.47
11.94
19.70
19.38
18.57
18.04
14.63
4.30
5.23
6.25
9.25
10.21
13.14
21.93
21.08
20.09
19.67
16.13
4.57
5.51
6.56
9.71
10.66
13.87
23.26
22.11
21.00
20.64
17.03
5.06
6.05
7.16
10.59
11.50
15.22
25.76
24.02
22.70
22.46
18.72
Females
0.22 to < 0.5
0.5 to < 1
1to<2
2to<5
5to<7
7 to < 11
11 to < 23
23 to < 30
30 to < 40
40 to < 65
65 to < 96
53
63
66
36
102
161
87
68
59
58
45
6.5 ±0.9
8.5±1.0
10.6±1.3
14.4 ±3.0
19.7 ±2.3
28.3 ±4.4
50.0 ±8.9
59.2 ±6.6
58.7 ±5.9
58.8 ±5.1
57.2 ±7.3
3.26 ±0.66
3.96 ±0.72
4. 78 ±0.96
7.06±1.16
8.22 ±1.31
9.84 ±1.69
13.28 ±2.60
13.67 ±2.28
13.68 ±1.76
12.31 ±2.07
9.80 ±2.17
2.17
2.78
3.20
5.15
6.06
7.07
9.00
9.91
10.78
8.91
6.24
2.41
3.05
3.55
5.57
6.54
7.68
9.94
10.74
11.42
9.66
7.02
2.81
3.48
4.13
6.28
7.34
8.70
11.52
12.13
12.49
10.92
8.34
3.26
3.96
4.78
7.06
8.22
9.84
13.28
13.67
13.68
12.31
9.80
3.71
4.45
5.43
7.84
9.11
10.98
15.03
15.21
14.87
13.70
11.27
4.11
4.88
6.01
8.54
9.90
12.00
16.61
16.59
15.94
14.96
12.58
4.36
5.14
6.36
8.97
10.38
12.61
17.56
17.42
16.58
15.71
13.37
4.81
5.63
7.02
9.76
11.27
13.76
19.33
18.98
17.78
17.12
14.85
'Measured body weight. Normal-weight individuals defined according to the body mass index (BMI) cutoffs.
'Physiological daily inhalation rates were calculated using the following equation: (TDEE + ECG) x H x (VE/V02) x10"3, where H = 0.21 Lof 02/Kcal,
VE/V02 = 27 (Lavton, 1993), TDEE = total daily energy expenditure (kcal/day) and ECG = stored daily energy cost for growth (kcal/day). Also see
accompanying text.
"Percentiles based on a normal distribution assumption for age groups.
N = number of individuals
SD = standard deviation
Source: Reprinted with permission of Taylor & Francis©; (Brochuetal., 2006b)
The study results were also expressed and presented in terms of m3/kg-day (Table 3-3). In
terms of m3/kg-day, these results may be considered to be related to intake of toxic agents
present in air inhaled for a typical daily activity pattern and thus as a basic surrogate for
an overall or systemic dose normalized to body weight. On this basis, average PDIRs
exhibited a decrease from birth (e.g., 0.509 m3/kg-day for males) to adulthood (e.g.,
0.247 for males), a factor of nearly 2. As shown in Table 3-3, the average rate continued
to decrease in an age-related manner such that the average rate in the most elderly group
3-8
-------
tested (65- 96 years) was 0.188 ± 0.031 m3/kg-day for males and 0.172 ± 0.037
m3/kg-day for females, also nearly a factor of 2, compared to the 11-23 year age group.
However, the authors' distributional analysis of the data further extended the range of
values that were present in the sampling. Specifically, the rates at the 99th percentile for
males and females at birth, 0.725 and 0.721 m3/kg-day, respectively, were approximately
2-times the corresponding values for adult males and females (0.410 and 0.383
m3/kg-day, respectively).
3-9
-------
Table 3-3 Distribution Percentiles of Physiological Daily Inhalation Rates on a per Body Weight
Basis for Free-Living Normal-Weight Males and Females Aged 2.6 Months to 96 Years
Age Group
(years)
N
Body
Weight (kg)a
MeaniSD
Physiological Daily Inhalation Rateb (m3/kg-day)
Percentilec
Mean + SD
5th
10th
25th
50th
75th
90th
95th
99th
Males
0.22 to < 0.5
0.5 to < 1
1to<2
2to<5
5to<7
7 to < 11
11 to < 23
23 to < 30
30 to < 40
40 to < 65
65 to < 96
32
40
35
25
96
38
30
34
41
33
50
6.7 ±1.0
8.8±1.1
10.6±1.1
15.3 ±3.4
19.8 ±2.1
28.9 ±5.6
58.6 ±13.9
70.9 ±6.5
71. 5 ±6.8
71.1 ±7.2
68.9 ±6.7
0.509 ±0.093
0.479 ±0.071
0.480 ±0.059
0.444 ±0.042
0.41 5 ±0.047
0.372 ±0.062
0.300 ±0.047
0.247 ±0.039
0.237 ±0.034
0.230 ±0.042
0.188 ±0.031
0.356
0.363
0.383
0.375
0.337
0.270
0.222
0.183
0.181
0.161
0.137
0.390
0.389
0.405
0.391
0.354
0.293
0.239
0.198
0.193
0.176
0.149
0.447
0.432
0.441
0.416
0.383
0.330
0.268
0.221
0.214
0.202
0.168
0.509
0.479
0.480
0.444
0.415
0.372
0.300
0.247
0.237
0.230
0.188
0.571
0.526
0.520
0.472
0.446
0.413
0.331
0.273
0.260
0.258
0.209
0.627
0.570
0.556
0.497
0.475
0.451
0.360
0.297
0.281
0.284
0.228
0.661
0.595
0.578
0.512
0.492
0.474
0.377
0.311
0.293
0.299
0.239
0.725
0.644
0.618
0.541
0.524
0.516
0.410
0.338
0.317
0.328
0.260
Females
0.22 to < 0.5
0.5 to < 1
1to<2
2to<5
5to<7
7 to < 11
11 to < 23
23 to < 30
30 to < 40
40 to < 65
65 to < 96
53
63
66
36
102
161
87
68
59
58
45
6.5 ±0.9
8.5 ±1.0
10.6±1.3
14.4±3.0
19.7 ±2.3
28.3 ±4.4
50.0 ±8.9
59.2 ±6.6
58.7 ±5.9
58.8 ±5.1
57.2 ±7.3
0.504 ±0.093
0.463 ±0.064
0.451 ±0.077
0.441 ±0.071
0.395 ±0.048
0.352 ±0.062
0.269 ±0.049
0.233 ±0.042
0.235 ±0.035
0.211 ±0.036
0.172 ±0.037
0.351
0.358
0.325
0.323
0.315
0.251
0.189
0.163
0.178
0.151
0.112
0.385
0.382
0.353
0.350
0.333
0.273
0.207
0.179
0.191
0.164
0.125
0.442
0.421
0.399
0.393
0.362
0.311
0.236
0.204
0.212
0.186
0.148
0.504
0.463
0.451
0.441
0.395
0.352
0.269
0.233
0.235
0.211
0.172
0.566
0.506
0.502
0.489
0.427
0.393
0.302
0.261
0.258
0.235
0.197
0.623
0.545
0.549
0.532
0.457
0.431
0.331
0.287
0.279
0.257
0.220
0.657
0.568
0.577
0.559
0.474
0.453
0.349
0.302
0.292
0.270
0.233
0.721
0.612
0.630
0.607
0.507
0.496
0.383
0.331
0.316
0.295
0.258
'Measured body weight. Normal-weight individuals defined according to the body mass index (BMI) cutoffs.
'Physiological daily inhalation rates per body weight were calculated using the following equation: (TDEE + ECG) x H x (VE/V02) x 10"3 divided by the age
group-specific body weight, where H = 0.21 L of 02/Kcal, VE/V02 = 27 (Lavton, 1993), TDEE = total daily energy expenditure (kcal/day) and ECG = stored
daily energy cost for growth (kcal/day). Also see accompanying text.
"Percentiles based on a normal distribution assumption for age groups.
N = number of individuals
Source: Reprinted with permission of Taylor & Francis©; (Brochuetal., 2006b)
The current value used by U.S. EPA in the RfCMethods is 0.286 m3/kg-day based on the
accepted indices of 20 m3/day (13.8 L/min) and 70 kg body weight. This value (0.286
m3/kg-day) appears to be near the upper end of the statistical bounds for the Brochu et al.
(2006a) average value of the adult male age 23-30 population (0.247 ± 0.039 m3/kg-day,
Table 3-3). The younger aged populations (2.6 months to 11 years) consistently exceeded
this value, whereas older aged populations (greater than 65 years) were consistently
lower (Brochu et al.. 2006b) (Table 3-2). When viewed in a cross-sectional body
weight-normalized perspective, these results indicate that (1) age groups younger than the
3-10
-------
11-23 age group may potentially be exposed more and (2) age groups older than the
11-23 age group may potentially be exposed less based on daily inhalation rates. This
applies to both genders. When viewed from a longitudinal perspective (i.e. over a lifetime
comprising these age categories), the potential for exposure is highest at a young age and
decreases with increasing age. With either view, younger ages would likely be exposed at
a higher rate per body weight and, if this were the sole consideration for dosimetry,
would be considered more vulnerable than adults. It should also be noted that the higher
inhalation rate per body weight found in younger age groups compared to adults in this
study is consistent with related processes such as increased metabolic demands and
oxygen consumption in growing individuals; in particular lung growth where alveoli are
being added and gas-exchange surfaces in the pulmonary area are increasing
(Section 3.6).
Brochu and coworkers (2006c) undertook a systematic examination of the various
methods, including the DLW method to estimate PDIRs. The first part of the study
provided a critical review of traditional approaches to systematically reveal and
understand sources of bias in each of the several approaches identified. Consideration of
the often used time-activity-ventilation (TAV) approach showed this approach often
considers only those activities that can be specifically accounted for thereby ignoring all
others. Also time-activity-patterns are not well or equally-known for all age and gender
groups and adequate statistical data are rarely available. These liabilities cause the PDIR
estimates from this approach to be mostly underestimated. Energy expenditure
approaches may be evaluated either as energy expenditures or food-energy intakes.
Energy expenditures may be expressed as a mean of the energy expended for the activity
(kcal/min) or as a function of the basal metabolic rate (BMR) using metabolic equivalent
(METs) values. As with the TAV approach, adequate data are not available for all
age/gender groups and do not take into account energy cost/expenditure for all groups,
such as thermogenesis growth and spontaneous gesturing. Use and research of these
methods has also indicated that they may provide overestimates of PDIR, especially
among younger individuals. Daily-food-energy intake approaches sought to simplify the
array and interaction of biases by using a single input parameter (i.e., food intake), which
is then converted into PDIRs. However, sources of bias and error are now known for this
approach. Daily food-energy loss in feces, intestinal gases and urine is not taken into
account. It is now widely recognized that there is substantial underreporting of food
intake surveys with the authors suggesting as much as 10-45%. Further, this
J OO O ?
underreporting is not uniform in the population but more pronounced in female
adolescents and in overweight/obese individuals.
In the second part of this study (Brochu et al.. 2006c). the magnitude of under- and
overestimations of published inhalation rates derived from the non-DLW approaches was
described by a comparison with new sets of PDIRs and distribution percentile values
based on TDEEs measured by the DLW method and expressed both as mVday and as
3-11
-------
m3/kg-day. The DLW-based TDEEs were derived from an aggregate period of over
20,000 days for unrestrained free-living normal-weight individuals aged 2.6 months to 96
years (n = 1,252). Rates were also produced with Monte Carlo simulations on the main
input parameters for the various other approaches. The results indicated that few of the
Monte Carlo simulation percentiles based on traditional approaches (57 out of 253) were
close to physiological values within a gap of ± 5% or less. Relative to DLW-based
TDEEs, outputs from time-activity-ventilation approaches were overestimated, whereas
most of those using the metabolic equivalent approach were underestimated. Food-intake
approaches led to underestimated rates due to biases discussed above. The most accurate
daily inhalation rates were those based on DLW measurements with an error of about ±
5%, as calculated in previous studies for free-living males and females aged 1 month to
96 years during real-life situations in their normal surroundings (Table 3-2 and
Table 3-3). Aggregate errors in all estimates (mVday and m3/kg-day) for the traditional
approaches varied from -52% to +126%.
Allan and Richardson (1998) used Monte Carlo simulations to derive probability density
functions of daily air inhalation rates for six age groups of Canadians. The objective was
to use these functions to describe inhalation rates in probabilistic health risk assessments.
A collection of time-activity and breathing rate studies were reviewed in order to define
random variables describing probable durations that North Americans (Canadians and
Americans) spend at various levels of activity and their probable inhalation rates while at
each level of activity. The general approach to developing the VE probability density
functions involved compiling suitable minute volume data from the literature, organizing
it by age, gender and degree of physical activity of the subjects, and estimating
appropriate mean values, standard deviations and distribution shapes based on the
reported data. Only data for "normal" subjects were included; for example, data
pertaining to people with asthma or other disorders known to affect inhalation rates were
excluded. These random variables were combined in a Monte Carlo simulation (10,000
trials) to empirically generate probability density functions describing 24-hour inhalation
rates for each age group. The simulations suggested that all except one age groups'
24-hour inhalation rates could be represented with log-normal probability density
functions. The distribution of 24-hour inhalation rates for infants was found to be better
represented by a normal distribution than a log-normal distribution. Arithmetic mean
values, standard deviations, and coefficients of variation were approximated for these
distributions (Table 3-4).
3-12
-------
Table 3-4 Probabilistic 24-hr Breathing Rate Estimates
Age Group
Infants (0-6 months)
Toddler (7 months-4 yr)
Children (5-11 yr)
Teenagers (12-1 Syr)
Adults (20-59 yr)
Seniors (60+yr)
Source: Reprinted with permission of Taylor &
Gender
combined
male
female
combined
male
female
combined
male
female
combined
male
female
combined
male
female
combined
Francis©; Allan and
Distribution (m3/day)
2.1 ±0.58
9.7 ±.2.5
8.8 ±.2.3
9.3 ±.2.4
15.1 ±.3.1
14.0 ±.2.8
14.6 ±.3.0
17.7 ±3.8
14.0 ±2.7
15.8 ±3.7
17.5 ±4.0
14.9 ±3.1
16.2 ±3.8
15.6 ±3.5
12.8 ±2.5
14.2 ±3.3
Richardson (1998)
Coefficient of Variation
0.27
0.28
0.27
0.28
0.22
0.21
0.22
0.23
0.2
0.25
0.23
0.21
0.24
0.23
0.2
0.24
Allan et al. (2008) discussed the methodology of the three principal methods to
determining physiological daily inhalation; TAV, metabolic energy conversion (MEC),
and DLW. Probability density functions (PDFs) derived from recent data using the TAV
method were also given and compared with PDFs developed based on earlier analyzed
TAV and published MEC and DLW approaches. The sensitivity and influence of number
of activity categories employed in the TAV method was discussed along with the
inherent disadvantage in subjective judgment of activity levels. Similar critical analysis is
given to the MEC approaches, whether based on daily food intake, on average daily
energy expenditures to BMR, or as time weighted average (TWA) of energy expenditure
across activity levels; each of these have inherent uncertainty due to judgment and
subjective estimates. The DLW method is well suited for direct study in free-living
scenarios and does not require invasive sampling; however, the administration and
analysis of DLW can be cost prohibitive. The authors did not address the fact that
judgment and subjective issues are prominent with TAV and MEC questionnaire-based
approaches, issues that are totally absent from the DLW method. Table 3-5 shows a
selected portion of the age-grouped data. In discussing their results, Allan et al. (2008)
compared and interpreted both the MEC and DLW findings to those of the TAV
approach as a benchmark. This comparison showed that the estimated mean inhalation
rates determined by the MEC approach were 8-30% lower than TAV mean inhalation
rates. Compared to the TAV approach, mean inhalation rates determined by the DLW
3-13
-------
approach were 56% higher for infants, 38% lower for toddlers and children, and 1-21%
lower for teenagers, adults and seniors.
The authors concluded that, with the exception of infants, mean inhalation rates derived
by the MEC and DLW approaches were generally lower than those developed by the
TAV approach. Rather than to reflect on the possibility of unresolved influences related
to the known issues of judgment and subjective estimation to the accuracy of the
approach, the authors propose that the higher estimates of TAV approach "... appear to
offer a level of conservatism to human health risk assessments by being more
representative of highly exposed members of the population...". This statement cannot be
related either to precision or to accuracy.
Table 3-5 Comparison of Mean 24-hour Inhalation Rates (m3/day) Determined Using
Time-Activity-Ventilation (TAV), Metabolic Energy Conversion (MEC) or Doubly
Labeled Water (DLW) Approaches
Age Group TAV MEC DLW
Infants (0-6 mo) 2.18 3.90 3.39
Toddlers (7 mo-4 yr) 8.31 7.60 5.17
Children (5-11 yr) 14.52 11.53 9.01
Teenagers (12-19 yr) 15.57 14.47 15.49
Adults (20-59 yr) 16.57 12.77 15.37
Seniors (60» 15.02 11.35 11.86
Source: Reprinted with permission of Taylor & Francis©; Allan et al. (2008)
3.3 TB Region
3.3.1 Flow and Deposition
Taylor et al. (2007) examined the pattern of lung injury resulting from exposure to ozone.
The distribution of ozone uptake was studied in a single, symmetrically branched airway
bifurcation using CFD. Separate simulations for inspiratory and expiratory flows were
conducted at Reynolds numbers ranging from 100 to 500, corresponding to laminar flow
conditions, to examine the effect of flow rate on uptake. The simulations demonstrated
the total rate of ozone uptake increased with increasing flow rate during both inspiration
and expiration and that flux progressively decreased along the parent branch. In addition,
hotspots of ozone flux were observed at the carina of the bifurcation for all simulated
flow rates. However, at the lowest simulated flow rate, the location of maximum flux
shifted to the outer wall of the daughter branch. Compared to a straight tube with a
similar surface area, the presence of branching resulted in a enhancement of overall
3-14
-------
uptake. The results of these simulations may be applicable to transport and uptake of any
gas with similar properties and are useful in revealing hotspots that may lead to focal
regions of airway tissue injury.
Padaki et al. (2009) used CFD modeling to simulate the transport and uptake of ozone for
comparison between an idealized model of the larynx, trachea, and first bifurcation and a
"control" model in which the larynx was replaced by an equivalent, cylindrical tube
segment. This comparison was performed in order to examine the effect of laryngeal
geometry on flow behavior. Inlet Reynolds numbers from 140 to 4,300 were used in the
simulations. The results revealed a strong laryngeal jet with a reattachment point in the
proximal trachea. Jet turbulence occurred only at the high Reynolds numbers and was
attenuated by the first bifurcation. Hotspots previously reported at the first carina were
confirmed by the local fractional uptake data; additional hotspots at the glottis and
proximal trachea were also observed. These laryngeal effects were dependent on
Reynolds number, with maximal effects (-15% enhancement of uptake efficiency)
occurring at the highest flow rate. Although the increase in regional uptake subsided by
the end of the model (i.e. the first bifurcation), the effect of the larynx on cumulative
uptake persisted further downstream. Together, these results suggested that with
prolonged exposure to a reactive gas entire regions of the larynx and proximal trachea
could show signs of tissue exposure.
Zhang et al. (2006) employed a representative human upper airway model to describe
uptake and deposition of MTBE and ethanol vapors. This description was accomplished
using CFD approach. Model simulations were done under varying conditions, including 3
inspiratory flow rates (Qm = 15, 30, and 60 L/min). The airway model utilized was
created from a human cast consisting of two parts: the oral airway, including oral cavity,
pharynx, larynx and trachea; and a symmetric triple bifurcation representing generations
GO (trachea) to G3 (referred to in the report and herein as the "upper bronchial airway" or
UBA). This study therefore isolated the extrarespiratory (ER) region from the lower TB
region (UBA). To attain representative modeling of airflow in such a model, a
low-Reynolds-number model was selected (to assure laminar flow and constant fluid
motion) and adapted to the laminar-to-turbulent flow regimes that are likely to occur in
the human airway during inhalation at the flow rates employed in the simulations. The
deposition of vapors in each airway segment was described by the deposition fraction
(DF), which was calculated with the regional mass balance and the sum of local wall
mass flux. An uptake parameter (K) was also calculated for both ethanol and MTBE
using available values of diffusivity of vapor in air and liquid mucus phase and
equilibrium partition coefficients in gas and liquid interfaces. The respiratory mass
transfer coefficient (called hmby the authors) was also estimated.
The simulations showed that flow rate had a strong effect on vapor deposition; the lower
the flow rate, the higher the deposition fraction due to the extended vapor residence times
at low flow rates. Results showed that as the flow rate decreased from 60 L/min to 15
3-15
-------
L/min, DF for MTBE increased from 2.5% to 7.7% in the UBA. The simulation showed
further that the DFs increased in a nearly linear fashion with the distance into the airway,
indicating consistent deposition efficiency along the airway passage. Compared with
MTBE, DF values of ethanol were approximately three to six times greater in the oral
airways and two to five times greater in the UBA in the range of flow rates used. The
higher deposition of ethanol vapor may be attributed not only to its higher diffusivity but,
more importantly, to its higher solubility in the mucus layer as indicated by the value of
K for ethanol (413) compared to MTBE (11). Vapors that pass through the upper airway
may further penetrate into and deposit partly in the lower airway and alveolar regions.
This suggests that compared to ethanol MTBE may penetrate further and deposit in the
lower airways.
Simulations based on the mesh were analyzed by the authors on a more refined scale.
Local vapor deposition patterns were quantified in terms of a deposition enhancement
factor (DEF), which is defined as the ratio of local to average deposition densities, DEF
therefore being an indication and representation of vapor deposition "hotspots" in a given
region. Figure 3-1 and Figure 3-2 show the distributions of these DEFs in the airway
components of the model. These deposition patterns were clearly not homogeneous and
were nonuniform for ethanol, which is relatively highly absorbed in the UBA, and for
MTBE, which is not highly absorbed. The maximum DEF was ~ 1.5 for MTBE in the
UBA with the value reaching 7.8 in the UBA model for ethanol. The low maximum DEF
values for MTBE indicated that deposition of MTBE vapor was relatively uniformly
distributed in the upper airways with relatively little absorbed by the airway walls
whereas the opposite appears to be the case for ethanol with the greater overall absorption
allowing for more contrasting differences and higher DEF "hotspots."
In the bifurcation airway model, enhanced deposition occurred mainly at the carina
ridges and the inside walls around the carina ridges, due to the complicated airflows and
large concentration gradients in these regions. When the absorption parameter (K)
increases above the typical value, however, deposition of MTBE increases with
deposition patterns not changing much. With increasing absorption, however, the
locations of enhanced deposition receive even greater deposition and the maximum DEF
values increase.
3-16
-------
60-3
Figure 3-1 Distributions of Deposition Enhancement Factor (DEF) for
MTBE Vapor with Qin = 30 L/min in the Bifurcation Airway
Models
Source: Reprinted with permission of Informa Healthcare©; Zhang et al. (2006)
GQ-3
Figure 3-2 Distributions of Deposition Enhancement Factor (DEF) for
Ethanol Vapor with Qin = 30 L/min in the Bifurcation Airway
Models
Source: Reprinted with permission of Informa Healthcare©; Zhang et al. (2006)
These simulations utilized a three dimensional computational fluid dynamic simulation
method and provided detailed local deposition patterns for both MTBE and ethanol,
agents widely disparate in uptake, transport and deposition. These deposition patterns
showed clearly that tissue burdens at local sites may exceed by many times the average
dose of the airways, i.e. they are highly nonuniform. Whereas flow rates greatly affected
deposition fractions, deposition patterns were not much altered. The localized deposition
pattern suggested that the uptake pathway may have a preferential route along which
local tissues are subjected to heavy exposure of vapors much the same as has been
demonstrated for formaldehyde in the ET region of rats (e.g.. Kimbell et al.. 1997). Thus,
3-17
-------
this enhanced deposition at local sites in this lower region of the respiratory tract may
also result in tissue damage or other adverse biological responses at local sites in the first
four generations of the human tracheobronchial tree.
Typically, animal lung models are based on the geometry of the human lung especially in
the TB region. In human lung models, termed typical path morphometric models, the
bronchial airway geometry is represented in terms of atypical path lung model, in which
all airways of a given airway generation have identical diameters, lengths, and branching
angles. Such geometry does not apply to murine species, including rats and mice. Thus,
Madl et al. (2010) undertook the task of developing a stochastic morphometric model of
the bronchial tree of the BALB/c mouse. Morphometric data on the TB geometry of three
in situ lung casts of the BALB/c mouse were analyzed in terms of probability density
functions and correlations among the different airway parameters. The results of this
statistical analysis revealed significant differences in diameters and branching angles
between major and minor progeny branching off from the same parent airway at a given
airway bifurcation. Significantly, the number of bronchial airways generations along a
given path, expressed by the termination probability, branching angles, and
daughter-to-parent diameter ratios indicated that the location of an airway with defined
linear airway dimensions within the lung was more appropriately identified by its
diameter (or its parent diameter) than by an assigned generation number as per typical
path models. This conclusion of classification by airway diameter rather than by
generation number was the same one posited for the rat (Koblinger and Hofmann. 1988).
Application of the techniques and procedures resulted in a single datum for the
bronchial/bronchiolar volume of 0.114 cm3, which was defended by the authors as being
in agreement with the few other historical estimates available. These findings for particle
deposition in both rat and mouse indicates that these data support a stochastic rather than
typical path selection. The computed distributions of the geometric airway parameters
and their correlations may be used for random pathway selection of inhaled particles in
subsequent Monte Carlo-like deposition calculations. The significance of these findings
for gas dosimetry, especially for estimates for critical parameters such as surface area and
character of the gas flow within this anatomy relative to humans remains to be elucidated.
Madasu (2007) compared approaches to modeling inhaled dose in the pulmonary airways
that can characterize the axial nature of dose and injury known to occur with various
reactive agents. These authors' comparison was based on contrasting models representing
the lower airways of the lung. The basis for comparative measurement of flow and
absorption characteristics in these contrasting models is a representative 3-generational
bifurcating unit of pulmonary airway. The first model reported was composed of Weibel
geometry of repeated symmetric bifurcating tube geometry to which a 3D computational
fluid dynamic model (CFDM) was applied (under conditions of steady expiratory flow)
and typical contour plots of velocity and concentration fields obtained. The second model
was a two-dimensional model with geometry consisting of a series of rigid cylindrical
3-18
-------
tubes of decreasing diameter representing the branches in a generation; the tubes were
connected by "leaky" nodes of flow in conical transition regions that represent the
bifurcations. This model was termed by the authors as an axisymmetric single path model
(ASPM). The basis of the comparison of these two models was their mass transfer
coefficients for formaldehyde obtained from the designated gas characteristics and the
gas flow conditions applied. The mass transfer coefficient (Kg) was defined to represent
an overall coefficient of uptake or absorption. Numerical results were compared for two
different inlet flow rates, wall mass transfer coefficients, and bifurcation angles. The
results of these model simulations indicated that the mass transfer coefficients from the
ASPM representation compared well with CFDM qualitatively and quantitatively. In
general, the mass transfer coefficients from both models were noted to increase with
bifurcation angle, inlet flow, and wall mass transfer coefficient. Further, the change in
mass transfer coefficients at each bifurcation unit was also closely predicted, and the
average concentration variation axially was qualitatively the same in both the predictions
from the CFDM and ASPM models with quantitative differences observed likely due to
the differences in flow characteristics in the branches. The authors concluded that these
results indicated that the "simplified" ASPM was very useful in predicting mass transfer
coefficients, flux at the walls, and hence injury sites as accurately as the "complex"
CFDM in symmetric lung systems where it was not possible to measure them. Similar
observations were made by Madasu et al. (2008; 2007).
Many vapor absorption models (compartmental, ID, and 3D) operate on the basis of the
assumption of steady state mass transport fluxes across the mucus and tissue barriers.
That is, the concentration profiles in the mucus and tissue are immediately linear and a
constant flux is assumed through the wall in the absence of reactions. The validity of this
assumption has recently been examined in studies by Longest and coworkers that
evaluated the mass transport of acetaldehyde and benzene, considered highly soluble and
moderately soluble in mucus, repsectively as sample vapors through a simple multilayer
system composed of mucus, tissue, and blood components on a transient basis.
Specifically, Tian and Longest (2010c) showed that transient wall absorption
significantly influences uptake during cyclic inhalation over multiple breaths in the upper
airways. For example, Tian and Longest (2010c) found that the uptake of highly and
moderately soluble compounds modeled with transient fluxes varied from steady state
flux estimates by a factor of approximately 30 in nasal and upper TB models. As a result,
it appears important to consider a time-dependent (or transient) flux value when
estimating absorption into the respiratory airway walls. The results of Tian and Longest
(2010c) were based on analytical and numerical solutions of transient absorption into a
simple mucus-tissue-blood wall representation. However, a constant vapor concentration
was assumed at the air-mucus interface and the air phase transport was neglected. They
then extended this analysis to an air-mucus-tissue-blood (AMTB) system and developed
a boundary condition to predict transient absorption and desorption fluxes in a CFD
model (Tian and Longest 201 Ob). Results of the AMTB wall model verified that
3-19
-------
absorption was highly time dependent over the timescale of an inhalation cycle
(approximately 1 to 2 s). Application of this boundary condition to CFD simulations of a
nasal-laryngeal geometry showed, as expected, that transient absorption significantly
affected total deposition fractions in the mucus, tissue, and blood for highly and
moderately soluble compounds. Moreover, transient absorption was also shown to
significantly affect the local deposition patterns and maximum local DBFs. Based on
these previous studies, it can be concluded that transient mass absorption significantly
affects uptake in individual wall layers for moderately and highly soluble compounds.
Based on their preliminary results, Tian and Longest (2010a) evaluated the effects of both
transient flow fields and transient mass absorption on the uptake of highly and
moderately soluble compounds in an upper airway model. In this study, a boundary
condition that represents transient absorption into the airway walls was applied. A new
dosimetry program, named Transient Absorption of Chemical Species (TAOCS) 1.0, was
developed and implemented to determine the coefficients needed for the transient
boundary condition expression and to apply the boundary condition to the CFD model.
Results indicated that implementation of the transient absorption boundary condition was
critical to predict local deposition characteristics for even highly soluble compounds. Use
of the TAOCS program simplified the implementation of the complex transient
absorption condition making the CFD simulation process more efficient.
3.3.2 Advances in TB Inhalation Dosimetry Modeling
Recently, Morris and Hubbs (2009) characterized the inhalation dosimetry of diacetyl, a
component of butter flavoring vapors, through development of a CFD-PBPK hybrid
model. Upper respiratory tract (URT) uptake of diacetyl was measured experimentally
and used to validate the model. Model simulations were then performed to estimate tissue
(anterior and posterior) and airborne concentrations of diacetyl for the URT (i.e. nasal)
and trachea in rats and humans. At an exposure concentration of 100 ppm, tissue
concentrations in the nose were estimated to be 1.6 and 1.4 mM in rats and 1.4 and 1.2
mM in humans, and in the trachea were estimated to be 1.2 and 1.1 mM in rats and 1.2
mM in humans. The air exiting the URT was estimated to be 67 ppm in rats and 82 ppm
in humans, and in the trachea was estimated to be 61 ppm in rats and 79 ppm in humans.
When the human model was run for mouth breathing only, the tissue concentrations in
the trachea were predicted to be 1.5 mM and the air exiting this region to be 96 ppm.
These results demonstrated that target tissue concentrations of diacetyl in the trachea
were highly similar in rats and humans and that diacetyl may penetrate to the lower
airways of humans to a greater degree than in rats. The authors concluded that based on
these relationships and differences in uptake efficiencies upper airway injury in the rat
may be predictive of lower airway injury in humans.
3-20
-------
3.4 PURegion
3.4.1 Flow and Deposition
Kauczor et al. (2002) provided a thorough review of techniques that are used for
functional imaging of lung function. Advances in magnetic resonance imaging (MRI)
now allow for functional imaging and direct visualization of lung ventilation. Utilizing
hyperpolarized noble gases with MRI was a recent approach developed for ventilation
imaging. Imaging lung tissue with either hyperpolarized xenon (Xe)-129 or Helium-3
(3He) has had considerable success. These agents can be inhaled in relatively large
quantities without substantial risks as they have no known toxic side effects and are not
absorbed by lung tissues. Compounds containing a relatively large number of fluorine
atoms per molecule also can be used as gaseous or liquid contrast agents for ventilation;
inert fluorinated gases, such as tetrafluoromethane, hexafluoroethane, or sulfur
hexafluoride have been utilized. Even more recently oxygen (at high concentrations) is
utilized for direct visualization of ventilation. Although molecular oxygen is weakly
paramagnetic, its effect in the lung is significant due to the enormous surface area of the
lung and the large difference in oxygen partial pressure between breathing room air (20%
oxygen) and pure (100%) oxygen.
3He MRI has been especially used to visualize dynamic ventilation during both
inspiration and expiration of ventilation in normal individuals (Kauczor et al., 2002).
Application of this technique indicates that normal ventilation in healthy lungs is
represented by a completely homogeneous distribution at the level of resolution of 3He
signal. Figure 3-3 illustrates the in-life rapid and homogenous filling of the airspaces
bilaterally (the numbers correspond to the sequence imaging times). In volunteers the
inflow of 3He was shown to be very rapid with the discernable signal appearing almost
simultaneously in the upper, middle and lower portions of the lung with a uniform
wash-in and wash-out of the gas also observed. Further advances, involving echo-planar
imaging of axial slices having rapid temporal resolution times of 122 ms, are able to
demonstrate in supine individuals preferential ventilation of the posterior lung zones,
again through visualization of areas of nonhomogenous flow in the lung. Further
demonstrations of the resolution of the 3He- imaging is the capacity to observe even small
(2 cm) transient ventilation defects in the lungs of smokers that appear as
nonhomogeneous flow and distribution. In clinically healthy smokers even markedly
smaller ventilation defects leading to nonhomogenous flow, such as those thought to
correspond to chronic inflammation and obstruction of small airways caused by smoking,
can be detected with 3He MRI. Thus, these techniques provide an approach to acquire
regional information on lung morphology and pulmonary function.
3-21
-------
Figure 3-3 Dynamic Ventilation 3He MRI After Inhalation of
Hyperpolarized 3He Gas
Source: Reprinted with permission of Springer Berlin/Heidelberg©; Kauczor et al. (2002)
Whole-lung dosimetry models do not account for the flow field to the level of inside the
alveoli and therefore may not accurately describe alveolar flow or deposition. To better
understand the fluid characteristics at this level of the lung, Harding and Robinson (2010)
employed CFD to a model of a terminal air sac much in the manner that it has been
applied to other respiratory tract regions, notably the extrarespiratory (ER) region. An
expanding terminal alveolar sac using truncated spheres to represent individual alveoli
was modeled numerically, based on dimensions from human lung casts. The flow field is
quantified for a breathing cycle derived from pulmonary function test measurements. The
alveolar sac model was considered representative of a terminal air unit in humans that
could be present in Weibull generations 19 and below based upon dimensions from
literature. The wall motion of the alveolar sac model (full expansion of 15.6%) was
obtained in vivo using a spirometer for a 21-year-old female breathing normally in the
sitting position. Model output was obtained for detailed regional flow rates, alveolar
mouth to depth flow rate ratio, and penetration depth of residual air. Figure 3-4
demonstrates the directionality and range of regional flow velocities as well as their
extent of incursion into the sac, all obtained from the model (Harding and Robinson.
2010). Examination of the flow field in the alveoli revealed no recirculation during any
/ O J
point in the breathing cycle. Other parameters addressed with the model included the
flow rate ratios of alveolar mouth to duct flow that were noted in the range of 0.18-0.36.
Penetration depths were less than 33% into the air sac during inhalation, decreasing in
length for air inside the sac to zero near the wall. These results indicated dominance of
diffusive motion over convective motion and flow at the level of the alveoli.
3-22
-------
The authors provided further comment on the modeled alveolar flow rate ratios obtained
in that they were intermediate between two other wide-ranging estimates of 0.057
(Kumar et al., 2009) and 1.0 (Sznitman et al., 2007). They could not provide an
explanation for this wide range, and indicated that more studies are needed before
quantification of flow fields in the alveolar region can be clearly understood as the ratios
that are present in vivo are also unknown.
Velocity Magnitude
(mis)
1.2E-04
1.1E-04
1 .OE-04
9.6E-05
8.8E-05
8.0E-05
7.2E-05
6.4E-05
5.6E-05
4.8E-05
4.0E-05
3.2E-05
2.4E-05
1.6E-05
8.0E-06
Point A
0.000
-0.200
-0.400
-0.600
-0.200-0.1000.000 0.100 0.200
X(mm)
Figure 3-4 Simulated Flow Velocities from CFD Solutions in an
Alveolar Sac Model
Source: Reprinted with permission of Informa Healthcare©; Harding and Robinson (2010)
In an earlier study, Tsuda et al. (2002) observed flow patterns of different colored
polymerizable fluids, representing tidal and residual air, injected into rat lungs in a
manner to simulate inhalation of tidal air. These authors concluded that the swirls seen in
the solidified cast in the large, medium, and alveolar airways were characteristic of
chaotic flow. They observed swirl patterns in alveoli that became more intense with
increasing number of cycles, which were not seen by Harding and Robinson (2010). who
utilized a model of terminal air sacs. Although these authors did not indicate if their
observations were from a terminal sac or a respiratory bronchiole, it is possible that the
patterns observed by Tsuda et al. (2002) occurred higher up in respiratory bronchioles
where the flow rate ratio was large enough to cause irreversibility.
It is clear that more studies are needed on pulmonary fluid flow to better understand the
nature of tidal and residual air mixing and the conditions under which mixing occur. It is
apparent from these disparate results that more corroborating evidence is needed before
3-23
-------
actual flow fields in the terminal air sacs are understood. In addition, the occurrence of
significant localized deposition cannot be excluded without additional studies.
3.4.2 Advances in Quantitation in Lung Geometry
The estimation of alveolar number in the lung has traditionally been done by assuming a
specific geometric shape. These geometries are then applied to small sampled volumes of
pulmonary tissue. However, the realizations that there exists a diversity of alveolar
shapes and that statistical error from small sample size and bias may be considerable,
have led to alternative approaches. Hyde et al. (2004) synthesized recent approaches and
technologies that were designed to be less prone to error and bias and to therefore
produce more reliable counts. These authors employed the following for the counting of
alveoli in the lungs of monkeys and rats: a fractionator which allows for systematic
random sampling from blocks of variable slab thickness (thereby minimizing the
inaccuracy inherent in using section sampling fractions based on the average thickness of
sections of variable thicknesses); use of the Euler characteristic of the net of alveolar
openings to estimate alveolar number; the disector principal (usually a counting probe for
isolated objects) as a sampling probe of the Euler characteristic. The Euler characteristic
of structure (an integer)2 applies to any level of topological complexity and is not biased
toward any specific geometry (as have other attempts to count alveoli).
Lung tissues from four male and one female rhesus macaques (Macaca mulatto) ranging
in age from 28 to 157 months and in body weight from 3.4 to 11.6 kg, as well as tissue
from five male Wistar rats with age not given and varying in body weight from 503 to
625g were used for this study. Using this approach on these tissues indicated the number
of alveoli in the two left lung lobes in the monkey ranged from 48.8 x 106 to 67.1 x 106
with a mean of 57.7 x 106. The average number of alveoli in the rat lung ranged from
17.3 x 106 to 24.6 x 106, with a mean of 20.1 x 106. The coefficient of error due to
stereological sampling was 0.06 in both monkeys and rats and the biological variation
(coefficient of variance between individuals) was 0.13 in monkey and 0.15 in rat (left
lobe, only). Between subdivisions (left/right in rat and cranial/caudal in monkey) there
was an increase in variation, most markedly in the rat. With age (2-13 years) the alveolar
volume increased 3-fold (as did parenchymal volume) in monkeys, but the alveolar
number was unchanged. The lung volumes as estimated in rats are presented in Table 3-6.
2 The Euler characteristic is a number that describes a shape or structure regardless of its orientation or the manner
in which it may be bent. For simple structures it may be determined from the formula j = 7 - £ + F where % is
the Euler characteristic, V the vertices, E the edges, and F the faces of a polyhedron shape. For a tetrahedron, for
example, the Euler characteristic from this formula is 4 - 6 + 4 = 2.
3-24
-------
Table 3-6 Estimates of Right, Left, and Total Lung Volumes in Male Wistar Rats
Animal # Body Weight (g)
R5 503
R3 528
R4 573
R1 595
R2 625
Mean 565, (CV 0.09)
Right Lung
10.6
8.0
11.4
10.2
12.3
10.5, (CV 0.15)
Lung Volumes (cm3)
Left Lung
8.8
4.2
5.5
5.3
6.1
6.0, (CV 0.29)
Total Lung
19.4
12.2
16.9
15.5
18.4
16.5, (CV 0.17)
CV = coefficient of variation
Source: Reprinted with permission of John Wiley and Sons©; Hyde et al. (2004)
Ochs et al. (2004) performed advanced stereologic analysis of human lungs for the
purpose of evaluating the number of alveoli present in the total lung (Table 3-7). The
stereologic method for the estimation of alveoli utilized the Euler number as the basis for
quantification, eliminating assumptions and the resultant bias about the shape, the size, or
the spatial orientation or distribution of alveoli. Alveolar number was estimated using
light microscopic sections and concentrating on alveolar lumens, using their appearance
or disappearance in a physical disector as counting events. Lungs for analysis were
obtained from six cases of single lung transplantation, four females and two males. After
fixation the whole lung was cut into horizontal slices of 2 cm thickness, starting with a
random apical position between 0 and 2 cm. Alveoli were defined as alveolar lumen and
septae, excluding respiratory bronchioles and alveolar ducts. Because of the design used,
alveolar number estimation became completely independent of the orientation
distribution of the lung sections sampled, making it effortless to combine the method with
all other stereologic estimators of interest for lung quantitation. In six adult human lungs,
the mean alveolar number determined by these procedures was 480 million (240 million
x 2 to account for both right and left lungs), with a range of 274-790 million and the
coefficient of variation 37% (Table 3-7). Alveolar number was observed to be closely
related to total lung volume, with larger lungs having considerably more alveoli. The
mean size of a single alveolus was rather constant with 4.2 x 106 (im3 (range: 3.3 x 106to
4.8 x 106(im3; coefficient of variation 10%), irrespective of the lung size. One cubic
millimeter lung parenchyma was calculated to contain around 170 alveoli. No further
attempts were made by the authors to obtain estimates for other parameters including
surface areas, although such calculations were feasible.
3-25
-------
Table 3-7 Summary Data on Human Lung Alveolar Number and Volume
Parameter
Gender (age)
Lung analyzed
N (alv), 106
V(lung), cm3
Lung 1
Female
(31)
Left
137
1,031
Lung 2
Female
(41)
Right
226
1,273
Lung 3
Female
(18)
Right
220
1,509
Lung 4
Female
(37)
Left
185
1,103
Lung 5
Male
(24)
Right
275
1,917
Lung 6
Male
(20)
Left
395
2,317
Mean Value
240 ± 89
1,534 ±521
Source: Reprinted with permission of American Thoracic Society©; Ochs et al. (2004)
Wiebe and Laursen (1995) compared a stereological morphometric method with a
standard fluid displacement method for determination of volume of right human lungs
obtained from 4 cadavers. Comparison showed that the two methods were in very close
agreement (Table 3-8). These authors then completed a stereological estimation of
alveolar surface area of these same lungs. Specifically sampled sections of lung tissue
(vertical and isotropic uniform random, IUR) were evaluated by specific counting
techniques related to a test line in a reference space whereas the volume of the section
was evaluated with the Cavalieri principle3. The capillary length and length density was
also estimated by similar stereological procedures. Point counting also showed 87.5% of
the lung is parenchyma, 5.4% is vessel volume, and 7.1% is bronchial volume. The
authors also estimated that of the total variation encountered in the processes only,
approximately 2%, was due to the stereological variation whereas approximately 98%
was due to the biological variation on the sections themselves. In evaluating their
estimates of lung surface areas by these techniques, the authors compared their results
with other known determinations of lung surface area (Table 3-8).
3For a 3-dimensional case, the Cavalieri principle is: Suppose two regions (3-dimensional solids) are included
between two parallel planes. If every plane parallel to these two planes intersects both regions in cross-sections of
equal area, then the two regions have equal volumes. This provides an unbiased and efficient estimate of the volume
of a solid object of arbitrary shape using systematic stereologic sectioning.
3-26
-------
Table 3-8 Summary Table of Measures from Right Lungs of Human Cadavers
Lung Measure
Case#
Mean ± SD
Reference
Volume (L)
Fluid
1.9
1.7
1.9
2.0
1.9±0.13
Wiebe and Laursen (1995)
Cavalier!
1.7
2.2
2.2
2.1 ± 0.25 Wiebe and Laursen (1995)
Capillary length (m xl05)
Vertical slices
12.3
5.6
7.5
6.3
7.9 ±3.0
Wiebe and Laursen (1995)
IUR
11.6
6.1
9.6
6.6
8.5 ±2.6
Wiebe and Laursen (1995)
SA (m2;
Vertical section
50.3
35.0
49.4
38.5
43.3 ±7.7
Wiebe and Laursen (1995)
IUR section
49.9
32.0
49.1
35.3
41.6 ±9.3 Wiebe and Laursen (1995)
Total SA (m2;
40-973
Thurlbeck(1967
78.4-81.6" Wiebe and Laursen (1995)
'Internal surface area range for 25 pairs of lungs, free from acute or chronic disease, from patients ranging from 25 to 70 years of age.
"Calculated by authors using right lung SA mean measurements of Vertical section 43.3/0.53 = 81.6 m2 and of IUR section 41.6/0.53 = 78.4 m2.
Source: Reprinted with permission of John Wiley and Sons©; Wiebe and Laursen (1995)
Knust et al. (2009) employed advanced stereological morphometric techniques in
measuring lung parameters in adult female CL57B6 mice (20.6 g average weight; no N
given). Capillary length was measured using the harmonic mean of the surface weighted
diameter. The Euler characteristic was applied in the physical fractionator with varying
but known sampling fractions and enabled the estimation of alveolar number. The
estimation of volume fractions of different lung compartments was carried out by point
counting. All values were corrected for tissue shrinkage. The following measures were
obtained for adult mice lungs (mean, CV): total values for alveolar number of 2.31 x 106
(0.23); alveolar surface area of 82.2 cm2 (0.17), alveolar air spaces of 138 mm3 (0.29);
capillary surface area of 124 cm2 (0.13), and capillary length of 1.13 km (0.13).
Bolle et al. (2008) examined functional and morphological characteristics in the
developing rat lung. Groups of specific pathogen-free Wistar-Kyoto (WKY) rats were
used for the examinations. Measures recorded included lung volume, respiratory
mechanics (intrapulmonary gas mixing, and gas exchange) and structural (alveolar
surface area, mean linear intercept length, and alveolar septal thickness) at 7-90 days.
Four males were sacrificed at each age for analysis. A selected set of measurements are
presented from this report in Table 3-9.
3-27
-------
Table 3-9 Functional and Morphological Features of the Developing Male Rat Lung
Parameter (n = 4)
Body weight (g)
Surface area (cm2)
Total lung capacity (ml)
Alveolar wall thickness (urn)
7 Days 14 Days
22 ±1.4 34 ±6.5
744 ±20 1,175 ±114
1.54±0.07 1.9 ±0.46
13.4±1.8 8.1 ±0.6
21 Days
76 ±8.5
1,648 ±188
4.6 ±2.6
5.4 ±0.4
35 Days
165 ±13.3
3,571 ±490
7.8 ±0.83
5.5 ±0.8
90 Days
41 7 ±22.6
6,536 ± 488
16.7 ±2. 46
6.4±1.0
Source: Reprinted with permission of The American Physiology Society©; Bolle et al. (2008)
3.5 Extrare s p irato ry (ER) Re g io n
3.5.1 Methods and Advances for Estimating Blood:Air Partition
Coefficients
The importance of blood:air partition coefficients (Hb/g) for PBPK models, and lack
thereof, prompted several approaches and strategies to enhance their development and
availability. Payne and Kenny (2002) reviewed, evaluated, and conducted a comparative
analysis of several predictive methods and models utilized to calculate Hb/g. As a first step
in their analysis, these authors gathered principal resources and approaches to derive Hb/g
(Meulenberg and Vijverberg. 2000; DeJongh et al.. 1997; Poulin and Krishnan. 1995;
Abraham and Weathersby. 1994; GargasetaL 1989; Abraham et al.. 1985).
The basis of the comparative analysis was a test set of 12 well characterized and
documented experimentally determined reference human Hb/g values for chemicals with a
wide range of lipophilicity (Payne and Kenny, 2002). In order of increasing octanol-water
partition coefficient the 12 were: acetone, isopropanol, diethyl ether, dichloromethane,
benzene, trichloroethylene, 1,1,1-trichloroethane, toluene, cyclohexane, «-pentane,
«-hexane, and «-heptane. The corresponding rat Hb/g values were also examined for all of
these compounds except for «-pentane, for which data were not available. The Hb/g value
for each compound was calculated from the various predictive models and ratios for
predicted:experimental values were determined. The mean (Rmean) and standard deviation
of the predicted: experimental ratio was calculated for each model using data for the 12
chemicals. The variability was characterized through calculation of the average error (E),
which was defined as the arithmetic mean of the absolute values of the logarithm of the
ratio (R) of predicted to experimental values where E = Mean (| log 10 R \) (for example,
a value of E of 0.301 corresponds to an average error of the predicted Hb/g equivalent to a
factor of IO0301 = 2 higher or 2 lower than the experimental reference Hb/g value).
Together this information was used to assess the accuracy of the various predictive
models.
3-28
-------
For humans, the most accurate methods for estimating Hb/g (within a factor of 2 of
experimental values; value of E < 0.301) were the empirical equations of Meulenberg and
Vijverberg (2000) and salvation equation of Abraham and Weathersby (1994). The
Paterson and Mackay (1989) approach also performed moderately well (Table 3-10). For
rats, the statistical estimates obtained indicated that use of solubilities in vegetable oil
rather than octanol gave better agreement with experimental values, which may reflect a
greater chemical similarity of vegetable oil to rat (or human) lipid. Low ratios observed
for rats suggest that protein binding makes a predominant contribution to partitioning
especially for chemicals of moderate or high lipophilicity. Overall, the simple linear
equation of Meulenberg and Vijverberg (2000) was the most accurate in estimating Hb/g
in rats. The human and rat equations of Gargas et al. (1989) gave reasonable accuracy for
moderately lipophilic chemicals but only over a restricted range of lipophilicity (range
not specified in the text). The approach and equations of Abraham and Weathersby
(1994). which contain consideration for protein binding and lipophilicity, fared well in
the analysis. However, none of the approaches were considered adequate to satisfactorily
cover the full range of lipophilicity. Therefore, the choice of method for the potential use
in PBPK models needs to take into account the species, tissue, and chemical lipophilicity.
Payne and Kenny (2002) generally concluded that partitioning into blood was determined
by solubility in water and lipid components, and by protein binding. For humans, the
accuracy of predictions in general was greater than for rat. It is speculated that this is
likely due to a considerably lesser influence of protein than with rats. The effects of
protein binding appeared extensive in rat blood such that they can result in as much as a
10-fold increase in Hb/g. For human blood such effects are smaller but still were estimated
to be in the range of 2 or more over that expected from lipid and water content alone
(data were not available to confirm these estimates). For humans, approximate estimates
of human Hb/g (within a factor of 2 of reference values) can be made over restricted
ranges of lipophilicity in the case of Meulenberg and Vijverberg (2000). Abraham and
Weathersby (1994). and Gargas et al. (1989). These authors did not conclude that rat
values were higher or lower than humans. Table 3-10 shows a summary of the results
from the comparative analysis of Hb/g in humans and rats.
3-29
-------
Table 3-10 Mean Ratios and Standard Deviations (Rmean ± SD) of Predicted to Experimentally
Derived Human and Rat Blood:Air Partition Coefficients, P, and Mean Absolute
Differences Between Predicted and Experimental Values of log P (E) fora Reference
Set of Unreactive Volatile Organic Chemical Vapors
Measure
Poulin &
Krishnan (1995)
(oil)
Poulin &
Krishnan (1995)
(octanol)
Paterson &
Mackay (1989)
Study
Gargas et al. (1989)
Meulenberg &
Vijverberg (2000)
Abraham et al.
(1985)
Abraham &
Weathersby (1994)
Human3
Rmean ± SD
E(logP)
0.83 ±0.38
0.175
0.76 ±0.39
0.218
0.86 ±0.52
0.224
1.11 ±0.69
0.325
1.10±0.46
0.156
0.76 ±0.42
0.276
0.93 ±0.38
0.166
Ratb
Rmean ± SD
E(logP)
0.40 ±0.33
0.537
0.38 ±0.34
0.590
No data
No data
1.13±1.00
0.337, 0.209C
0.79 ±0.50
0.236
No data
No data
No data
No data
Note: Actual values from which ratios were developed were not available.
aTest set included 12 chemicals.
bTest set included 11 chemicals. Data on n-pentane were not available for experimental values.
"Excluding acetone and isopropanol since they are more hydrophilic compounds with a log Pm < 0.5.
Source: Reprinted with permission of Taylor & Francis©; Payne and Kenny (2002)
In another study by Abraham et al. (2005). 155 human Hb/g values and 127 rat Hb/g values
for volatile organic compounds were collected to conduct a correlative analysis. One goal
of the analysis was to derive an equation that could be used to predict log Hb/g for rats,
humans, or both. The general method used for the correlation and prediction of log Hb/g
values used in this study (Abraham et al.. 2005) is the solvation equation for linear free
energy relationship (LFER) where SP is log Hb/g.
SP-c+e • E+s - S+a • A+b -B+l-L
Equation 3-5
The compound descriptors following all relate to unique chemical properties of the VOC:
E, solute excess molar refractivity with units of (dr^mo!"1 )/10; S, solute dipolarity/
polarizability; A and B, the overall or summation hydrogen bond acidity and basicity; L,
the log of the gas-hexadecane partition coefficient (unitless) at 25 °C. Multiple linear
regression analysis is used to evaluate the coefficients (lowercase italicized letters) in this
equation.
The initial step in the analysis was to determine the level of random and measurement
error between available rat and human Hb/g. For the random error evaluation, 86
compounds for which both rat and human log Hb/g were available were compared
(Abraham et al.. 2005). The average error from this comparison was 0.124 log units
indicating that the ratio between Hb/g (rat) and Hb/g (human) was about 1.3 with N=86.
This value is slightly less than that reported by Gargas et al. (1989) who concluded that
rat Hb/g was generally larger than human Hb/g by a factor of 1.5-2 (N=36). Further, this
3-30
-------
magnitude of error was found to be lower than estimates of interlaboratory variation
albeit only 3 agents were available for the comparison. Nonetheless, the relationship
identified by Abraham et al. (2005) indicated that the data on Hb/g (human) and Hb/g (rat)
can be combined in a correlative analysis.
Correlative analyses with LFER first solved separately for rat and humans, indicated that
the rat and human equations were comparable. This finding allowed for averaging the
data available for both rat and human to yield values for 196 compounds. This dataset
was then divided into two equal sets of 98 each with one set being used as a training set
to obtain one LFER equation applicable to either rat or human Hb/g. This test set equation
was then be applied to the second dataset (as an independent test set) for predictive and
error assessment. These sets were then reversed with predictive and error assessment
repeated. From this error assessment, the authors determined there was no bias in the
predictions (Abraham et al.. 2005). This finding led the authors to combine the entirety of
the data and provide an LFER equation for prediction of Hb/g values that could be applied
to either rat or human4. This equation was robust with 282 data points for 196 compounds
and a correlation coefficient (R2) of 0.927.
The authors concluded that this evaluation would allow for the prediction of Hb/g values
applicable to either rat or human, accurate to 0.33 log units (Abraham et al.. 2005). They
also point out that the descriptors required for the LFER method are available for some
3,000 volatile organic compounds, far more available than information for partition
coefficients based on, for example, air to oil and air to saline. Lastly, it is noted that this
was the first predictive assessment of calculations for (log) Hb/g made using the method of
training and test sets.
3.5.2 Quantitation using Inhalation PBPK Models
Physiologically-based pharmacokinetic (PBPK) models are biological, integrated
functioning systems of flow, volumes, and partitioning processes, with the purpose to
predict the time course distribution of a chemical in the body. The robustness of such
models is demonstrated by their ability to predict empirical observations.
When model simulations successfully predict empirical results, typically obtained
independent of the model, it is an indication that both the model and the sensitive critical
parameters within the model have predictive validity. For example, when models that are
parameterized and configured to predict interspecies dose extrapolation (e.g., between
rats and humans) are successful in their predictions, the model and its parameters are both
considered adequate. As referred to above, partition coefficients and in particular
airblood partition coefficients (Hb/g), are among these critical determinative parameters.
4 The final equation with coefficients (per Equation 3-5) is:
log Hb/g (human or rat) = -1.062 + 0.460 E +1.067 S + 3.777 A + 2.556 B + 0.375 L
3-31
-------
It then follows that inhalation PBPK models that (1) are parameterized and configured for
interspecies extrapolation and (2) are successful in predicting empirical results in animals
and humans would be a source of representative Hb/g for both humans and animals. It is
the ratio of Hb/g between animals and humans that is the basis for RfC Methods inhalation
gas dosimetry for effects in the extrarespiratory or ER region (see Section 2).
Consequently, validated inhalation PBPK models were obtained and examined for these
critical parameters which were extracted and constructed as a ratio in accordance with the
RfC Methods. The results of this investigation are presented in Table 3-11. This table
includes the PBPK model reference, chemical modeled, animal gender, species, and
strain when available, the method used to determine the Hb/g employed in the model, and
the A/H Hb/g ratio. Based on this analysis, in 3 instances the A/H ratios were less than 1
(e.g., 0.7, 0.6, and 0.6). For 2-BE and 2-ME, the rat values were assumed to be equal to
human Hb/g values; and for napthalene and n-butanol, the human values were assumed to
be equal to the rat Hb/g values.
3-32
-------
Table 3-11 Compilation of BloockAir Partition Coefficients used in Inhalation PBPK Models for
Animal to Human Interspecies Extrapolation
Chemical3
(Reference)
PCE
(Dallas etal., 1995)
TCE
(Croninetal., 1995)
Toluene
(Tardif etal., 1997)
Xylene
(Tardif etal., 1997)
EBZ
(Tardif etal., 1997)
Ethanol
(Pasting etal., 1997)
Toluene
(Benignusetal., 1998)
2-BE
(Lee etal. ,1998)
2-ME
(Gargasetal., 2000)
Naphthalene
(Willems etal. ,2001)
Ethylene glycol
(Corlev etal., 2005)
n-Butanol
(Teeguardenetal., 2005)
PGME
(Corlev etal., 2005)
PGMEA
(Corlev etal., 2005)
n-Decane
(Hissink etal. ,2007)
1,2,4-TMB
(Hissink etal. ,2007)
Chloroform
(Liao etal. ,2007)
Hb/g
18.9
14.3
13.2
18
46
42.7
2,140
1,244
18
7,965
7,965
32,800
571
17,901
1,160
4,866
1,251
21
148
20.8
91 3
Species/
Strain
3SD rat
$ Mouse
3 Mouse
Rat
Rat
Rat
Rat
Mouse
Rat
Rat
Mouse
Pregnant SD rat
Rat
$ SD & Wistar rat
Rat
Rat
Rat
Rat
Rat
Rat
MDIIRP
Animal
Method
In vivo tissue cone -time course
Not stated1
Not stated1
Sealed vial
Sealed vial
Sealed vial
Sealed vial'
Sealed vial8
In vivo
Not stated"
Not stated"
Sealed vial"
Calculated
Sealed vial
Sealed vial
Sealed vial
Sealed vial
Sealed vial
Sealed vial
Sealed vialh
SpqlpH viq|h
Hb/g
10.3
Q 9
15.6
26.4
28.0
1,265
1,265
15.0
7,965
32,800
571 c
17,542
1,160C
7,107
609
37
85
7.43
743
Human
Method
Sealed vial
Not stated1
Sealed vial
Sealed vial
Sealed vial
Sealed vial9
Sealed vial9
-
Sealed vial skin: air
Sealed vial
Calculated
Sealed vial
-
Sealed vial
Sealed vial
Sealed vial
Sealed vial
Not Stated'
Mnt Statprf
- A/H
Ratio
1.8
1.6
1.4
1.1
1.7
1.5
1.7
1.0
1.2
1"
1"
1"
1C
1.0
1C
(0.7)
2.0
(0.6)
1.7
2.8
99
3-33
-------
Animal
Human
Chemical9
(Reference)
1,1,1-TCE
(Lu et al.. 2008)
Mel
(Sweeney et al.. 2009)
Hb/g
5.76
39.3
16
12
Species/
Strain
Rat
Rat
Rabbit (adult)
Rabbit (fetal)
Method
Sealed viald
In vivo, sealed vial
In vivo, sealed vial
In vivo, sealed vial
Hb/g
2.53
18
(male)
17.1
(female)
17.6
(fetal)
Method
Sealed viald
Sealed vial
Sealed vial
Sealed vial
A/H
Ratio
2.3
2.2
1.0
(0.6)
"Chemical abbreviations: ethylene glycol monomethyl ether (2-ME); 2-butoxyethanol (2-BE); propylene glycol methyl ether (PGME);
propylene glycol methyl ether acetate (PGMEA); trichloroethylene (TCE); perchloroethylene (PCE); 1,2,4-trimethylbenzene
(1,2,4-TMB); ethylbenzene (EBZ); methyl iodide (Mel), 1,1,1-trichloroethane (1,1,1-TCE).
bRat values were assumed to be equal to human Hb/g values in this model.
°Human values were assumed to be equal to the rat Hb/g values in this model.
Experiments and values first reported by Reitz et al. (1988).
Experiments and values first reported by Pastino et al. (1996).
'Experiments and values first reported by Kaneko et al. (1994).
Experiments and values first reported for whole blood by Fiserova-Bergerova and Diaz (1986).
Experiments and values first reported by Gargas et al. (1989).
Values first reported by Fisher and Allen (1993).
Values first reported by Steward et al. (1973)
3.5.3 HECERDerivation - PBPK and Hb/g (RfC Methods) Comparison
Inhalation PBPK models use air and blood flows, predicted or measured absorption rates,
various biological rate processes (e.g., metabolism) and partitioning overtime, and a
range of external exposure air concentrations to a given toxicant to predict dose metrics.
As explained above, the Hb/g, is a key, and often determinative, parameter.
A dose metric is the internal tissue concentration of a toxicant, or a form of that toxicant
such as a metabolite, associated with the external exposure to a toxicant. For a tissue that
is a focus of toxicity (i.e., a target tissue), the concentration of a toxicant in the tissue is
considered to be the ultimate determinant of risk. The dose metric may be a concentration
overtime (e.g., area under the curve, AUC), a maximum concentration achieved (Cmax),
or a steady-state concentration. Examples of dose metrics are Cmax of parent compound
in the liver, AUC of a metabolite in the brain, or circulating blood concentration of parent
compound at steady state. The concentration in the blood is often used instead of the
concentration in a target tissue because blood concentrations are more readily measured,
allowing for model calibration and validation, and average or steady-state tissue
concentrations are expected to vary in proportion to blood levels.
PBPK models may be developed for a variety of purposes, one of which is interspecies
extrapolation, the general subject of this report. The manner in which this is performed is
to first use the animal model to estimate a dose metric (internal dose) associated with a
given level of toxicity or response and then use the human model to estimate the external
concentration for humans that yields the same internal tissue dose metric. The human
3-34
-------
estimate of the external concentration that produces that same internal dose metric is the
human equivalent concentration or HEC.
Several of the studies listed in Table 3-11 developed inhalation models for purposes of
interspecies extrapolation. Table 3-12 below presents specific descriptions of the dose
metric and the modeling estimates of the human equivalent concentration that
corresponds to the same internal dose metric calculated for the laboratory animal based
on the animal exposure scenario.
Table 3-12 Estimations from Inhalation PBPK Models of Human Equivalent Concentrations
(HECs) from Effect Levels and Internal Dose Measures in Laboratory Animals
Chemical3
(Reference)
Isopropanol
(Gentry etal.,
2002)
n-Butanol
(Teeguarden et
al.,2005)
PGME
(Kirman etal.,
2005b)
White spirits
(Hissinket al.,
2007)
2-ME
(Gargasetal.,
2000)
Ethylene glycol
(Corlevetal.,
2005)
1,1,1-TCE
(Lu etal. ,2008)
Level and
Effect
NOAEL
2,500 ppm
renal tissue of
female rats
LOAEL
3,500 ppm
developmental
NOAEL
500 ppm
weight gain
NOAEL
3,000 ppm
neurotoxicity
NOAEL
3,000 ppm
presence of
sedation
NOEL
600 mg/m3
neurotoxicity
LOEL
2,400 mg/m3
neurotoxicity
NOEL
10 ppm
developmental
LOEL
50 ppm
developmental
11 ppm
(28 mg/m3)b
developmental
NOAEL
1,500 ppm
liver effects
Dose Metric
Arterial blood
Arterial blood
concentrations, AUC
Arterial blood
concentrations, AUC
Cmax, richly perfused
tissues
Brain concentration of
_1,2,4-TMBordecane
determined in rats exposed
for 6 hr/day
Blood Cmax or average
daily
AUu tor z-MAA (acetic
acid; metabolite of 2-ME)
in rats exposed for
6 hr/day, 5 days/wk
Cmax for
glycolic acid (GA) in blood
Average daily venous
blood, AUC. Calculated in
rats exposed for 6 hr/day,
5 days/wk
Comments
HEC derived from Table 4 (in (Gentry etal., 2002)) by applying uncertainty
factor of 30: (159.8 x 30 = 4,767 ppm); 189.8 ppm x 30 = 5,700 ppm.
Animals exposed for 6 hr/day, 5 days/wk. Continuous exposure modeled in
humans.
Weekly average blood cone, estimated for rats at 6 hr/day, 5 days/wk and
continuous for humans. Model estimates compared against human blood
levels from 30 min inhalation exposure. Tables and equations are provided
for HEC calculation over wide range of butanol concentrations.
Model simulations estimated NOAEL internal dose metric values in rodents
ranging from 2,300-5,000 mg/L for exposures from 3,000 ppm for 1-78 wks
of exposure (6 hr/day, 5 days/wk). The arithmetic mean of the NOAEL was
4,036 mg/L. This value was used to estimate an HEC for a continuous 24 hr
exposure.
Model and 4-hr HEC based on main components of WS, 1,2,4-TMB and
decane. Estimates are for acute exposure CNS effects. Human model
validated with blood and alveolar air kinetics.
The model was used to calculate an HEC for pregnant women exposed for
8 hr/day, 5 days/wk for 270 days at various 2-ME. Human validation
information from urinary excretion rates of 2-MAAfrom volunteers exposed
to 5 ppm 2-ME
Model was used to generate a dose-response comparison of internal dose
surrogates (Cmax for GA in blood) in female Sprague-Dawley rats and in
humans (Figure 10B in (Corlev etal., 2005)). Several controlled rat and
human metabolism studies were used to validate the PBPK model.
Table 5 (in (Luetal., 2008)) shows HEC calculations over a wide range of
exposures concentrations for continuous human exposure. Four human
data sets were used in evaluating model selection.
PBPK
Derived
HEC
4,767 ppm
5,700 ppm
169 ppm
1,066 ppm
560 ppm
344-721 c
mg/m3
1 RRQ
4,431°
mg/m3
12 ppm
60 ppm
-79 ppm
(-200
mg/m )
640 ppm
'Chemical abbreviations: ethylene glycol monomethyl ether (2-ME); propylene glycol methyl ether (PGME); 1,1,1-trichloroethane (1,1,1-TCE),
1,2,4-trimethylbenzene (1,2,4-TMB), 2-methoxyacetic acid (2-MAA), glycolic acid (GA), white spirit (WS).
bThe threshold blood concentration for developmental effects of 2 mM is not attainable in humans based on the modeling and maximum tolerated inhalation
exposures reported in this paper (Corlevetal., 2005). The maximum vapor concentration for EG is only 79 ppm (-200 mg/m3) due to low volatility
(0.06 mm Hg at 20°C) (Corlevetal., 2005). Therefore, for this comparison, the human Cmax at the maximum vapor concentration (200 mg/m3) was
estimated by the model to be -6.5 uM. The exposure concentration predicted by the model that would yield the same Cmax in the rat is -28 mg/m3.
"Range of values is presented because exposure concentrations were estimated that yielded brain concentrations equivalent to observed values for
1,2,4-TMB or decane. Values at the lower end of the range correspond to WS estimates based on 1,2,4-TMB brain concentrations, while the higher
values are based on decane brain concentrations.
3-35
-------
Table 3-13 combines data from Table 3-11 and Table 3-12 to present examples
comparing approaches in estimating HEC from laboratory animal data for systemic
effects. For example, with n-butanol an extrarespiratory effect level of 500 ppm in the
laboratory animal study is duration and dosimetrically adjusted to an HEC using the RfC
Methods default approach (a DAF of 1; see Section 2) to yield 90 ppm. The neighboring
column to the right shows the HEC derived using the PBPK model at 169 ppm. The ratio
of these HECs are then compared to indicate the extent and direction of difference, such
that the n-butanol default HEC is two-times less than estimated by the PBPK model. For
further comparison, the actual A/H Hb/g ratio is also given, here shown for n-butanol
which in this case is the same as the RfC Methods default.
As can be seen, the extent of difference encountered between the default and PBPK HEC
values is quite wide, spanning nearly 10 times (e.g., isopropanol default method gives an
HEC of 446 ppm and PBPK method gives 4,767 ppm) even for this small set of example
chemicals. In all cases, the default RfC Method provides a lower HEC than those derived
using PBPK modeling, except for PGME which is nearly equal. No general trend can be
discerned to explain this range of differences, either between the default and PBPK HEC
or between the actual Hb/g and the PBPK HEC. It may be that other covariates, such as
concentration-dependent metabolism may need to be further explored and evaluated. In
application of PBPK models, it may also be necessary to thoroughly evaluate the
origination of model parameters, including the Hb/g.
3-36
-------
Table 3-13 Comparison of Approaches for Calculating Human Equivalent Concentrations (HECs)
for Several Gases with Effects in the Extrarespiratory Region (ER)
Chemical
(Reference)
n- Butanol
(Teeguarden et al.
(2005)
1,1,1-TCE(Luetal.
(2008)
PGME Kirman et al.
(2005a)
2-ME (Gargas et al.
(2000)
Isopropanol (Gentry
et al. (2002)
Ethylene glycol
(Corlev et al. (2005)
D^t Dnn •
(Table 3-1 2)
500 ppm
1,500ppm
3,000 ppm
10 ppm
2,500 ppmc
3,500 ppmd
11 ppm
PODadj
90 ppm
270 ppm
540 ppm
1.8 ppm
446 ppm
907 ppm
--
RfC Method
(Hb/gU / (Hb/g)H -. . p
(Table 3-11)
1.0 1
2.3 1
0.7 1
1 1
.0 1
1 1
HECa
90 ppm
270 ppm
540 ppm
1.8 ppm
446 ppm
907 ppm
11 ppm
. HEC-PBPK
Method
(Table 3-12)
169 ppm
640 ppm
560 ppm
2.9 ppmb
4,767 ppm
5,700 ppm
79 ppm
PBPK/RfC HEC
Ratio
1.88
2.4
1.04
1.61
10.7
6.28
7.18
aHEC derived by default RfC Methods: PODadj * DAF = HEC where the PODadj is the POD adjusted for duration of exposure in the animal study and a
default DAF of 1 is applied for (Hb/g)A / (Hb/g)H. (e.g., for n-butanol, the PODadj = 500 ppm x 6 hr/24 hr x 5 days/7days = 90 ppm.)
bln the PBPK model for 2-ME, the HEC was cafculated for a discontinuous exposure and was therefore adjusted for duration (8hr/24 hr x 5 days/7 days).
"Based on renal effects.
dBased on developmental effects.
3.6 Children's Inhalation Dosimetry
3.6.1 Introduction and Focus
This section is focused on identification and preliminary evaluation of data, evidence, and
information relating directly to gas dosimetry in children.
Although not as explicitly as with the fetus (developmental studies) or sexual maturity
and function (reproductive studies), the 1994 RfC Methods considers lifestages and
children in the intraspecies uncertainty factor that is designed to incorporate the range of
response variability in human populations. This uncertainty factor is typically considered
to have two components, pharmacodynamics and pharmacokinetics, with the latter
component being the basis of dosimetry. It is within the kinetic portion of this uncertainty
factor that susceptible lifestages, including children, are considered.
The 1996 Food Quality Protection Act (FQPA), refocused interest in matters of child
risk. Title III of this act specifically tasked the Agency in their assessments under the
FQPA to "... ensure that there is a reasonable certainty that no harm will result to infants
and children
Although this Act was directed at oral ingestion of pesticides, specifically those used on
foodstuffs, the Agency considered its implications both with regard to pesticide risk
assessments and more broadly to EPA methodology. For example, EPA developed
3-37
-------
approaches for interpretation and implementation of the requirements specifically to
FQPA-required pesticide assessments (e.g., see
http://www.epa.gov/oppfeadl/trac/science/). and additionally implemented a full review
of the Agency's RfC/RfD processes to insure they appropriately considered the potential
for increased childhood susceptibility (U.S. EPA. 2002).
This Act eventually affected many organizations and resulted in a spectrum of
implementation actions and strategies. One of the most prominent was completed by the
state of California in implementing their Children's Environmental Health Protection Act
(Senate Bill 25) of 1999. The state's Technical Support Document for the Derivation of
Noncancer Reference Exposure Levels (OEHHA. 2008) provided extensive information
on children as a population of concern and on pharmacodynamic and pharmacokinetic
differences between children and adults. Appendix E of that document included an
extensive analysis of children related data and models, including PBPK models, that
provided insight into the range of interindividual variability in general, but focus
extensively on the differences among infants, children and adults. This report does not
intend to reflect on these activities or these reports, but only note them as examples of the
movement of the risk assessment community towards further consideration of children in
dose-response toxicity and, in this case, of children's dosimetry.
In 1993, the NAS published its findings regarding chemical toxicity in children compared
to adults (NRC, 1993). The report addressed both specific findings and recommendations.
Conclusions of the committee included that infants and children may be more, or less,
susceptible than adults depending upon the chemical and the age of the subject. It was
acknowledged that substantial changes occur in organ size, structure, and function from
infancy through puberty; such changes could substantially affect the pharmacokinetics
and pharmacodynamics of chemicals. Accordingly, there may be periods, or lifestages of
vulnerability, when developing tissues are much more sensitive to toxicants than later in
life. The NAS report (1993) also stresses the importance of recognizing that the younger
the individual, the more pronounced his/her structural and functional anomalies and thus
the greatest differences from adults in susceptibility to chemical toxicity can be
anticipated, with continuous diminishment of those differences thereafter. The report also
stated the need for scientifically defensible means to deal with toxic agents that cannot be
directly studied in children. A specific recommendation in the report following from this
realized the need to use PBPK models. PBPK models can be used both to simulate the
time course of parent compounds and bioactive metabolites in blood and tissues of adult
animals and humans and to predict target organ doses of toxic chemicals/metabolites for
different exposure scenarios in children of different ages.
A recommendation following from the potential use of PBPK models was that they be
reliably developed by obtaining accurate measurements of respiratory parameters,
circulation, metabolism, tissue and fat volumes, and partition coefficients. These
parameters can be measured in primates or in children of different ages by noninvasive
3-38
-------
procedures. The parameters would be used in PBPK models, which could then be utilized
to better estimate the concentration time course of chemicals/metabolites in potential
target organs. It is the recommendations and statements from the NAS report (1993) that
guides the structure and content of this section of this report. There is an additional body
of literature on risk to children and children's' dosimetry that reviewed the need,
feasibility, adequacy (or inadequacy) and data gaps of existing methods and procedures
(Ginsberg et al., 2010; Firestone et al., 2008; Foos et al., 2008; Foos and Sonawane,
2008; Ginsberg et al.. 2008). The details of these reports are not considered here since
they do not directly inform the state-of-the-science related to gas dosimetry in children.
Recognizing that young children have a greater ventilation rate per body weight or per
surface area in the respiratory tract compared with adults, Ginsberg et al. (2005) analyzed
the outcomes of gas (and particle) dosimetry approaches of RfC Methods utilizing child
(3 mo) and adult male values available from various sources for the principal
determinants of VE, SAETPU, and BW. The TB region was characterized differently from
RfC Methods as comprising two separate regions termed tracheobronchial (BB) and
bronchioles (bb) by the authors. Dosimetry was estimated for 3-month-old children and
adults for reactive and nonreactive gases. Estimations of comparative dosimetry were
made using a reasonable range of assumed values for Kgs thereby allowing for direct
calculation of regional dosimetry as well as adjustment for fractional penetration through
the various regions (see Section 2). The authors use the same Kg values for both children
and adults indicating that no basis exists for assuming a difference. The modeling results
suggested similar dosimetry of gases for children and adults for the ET and BB regions.
Dosimetry for the bb region generally showed a higher dose of gases in adults than in
children. It was also noted that, based on the value of the Kg, dosimetry for adults versus
children in the PU region could be slightly different, either higher or lower but not greater
than 2-fold different. There were no cases in which gas dose was substantially greater in
the respiratory regions of 3-month-old children compared to adults. Estimates of systemic
doses of nonreactive gases were greater in 3-month-old children than in adults, especially
for liver doses (up to 2-fold) of metabolites for rapidly metabolized gases. Overall, these
results suggest the potential for a 2-fold greater inhalation dose in children (based on data
from 3-month-old children) than in adults, although there are cases in which this
differential could be greater or less.
As PBPK models configured for elucidating dose to children and infants were
recommended in the NAS report and are prominent in the current literature, they will be
featured in this section. Studies that provided insight and data for parameters needed for
these models and/or for general knowledge about development in children related to
aspects of dosimetry are also presented. These include reports on stages of alveoli
growth, lung parenchyma development, age-related airway diameters and surface areas
and volumes of upper airways. Information on inhalation rates in children have been
3-39
-------
presented earlier (Table 3-5). In addition to the TB and PU regions, information on the
ET regions is also included in this section.
3.6.2 PBPK Models
Firestone et al. (2008) reported results based on analyses conducted by the Office of
Environmental Health Hazard Assessment (OEHHA) that investigated the potential
differences between adult and child (0-18 yr) internal doses resulting from inhalation
exposure to a toxicant. Modified PBPK models for 24 compounds were used to assess
child/adult ratios for at least three dose metrics. Detailed methods, equations, and model
parameters were not included in the manuscript; however, the chemicals were classified
into one of three categories pertaining to the intrahuman uncertainty factor for
toxicokinetic variability (UFH_TK, default value = 3.16): UFH_TK < 3.16; UFH_TK > 3.16 to
9.9; and UFH_TK > 10-0- Twelve of the compounds examined had child/adult ratios <3.16,
eight had ratios between 3.16 and 9.9, while four had ratios greater than 10. The authors
found that majority of the higher ratios were in infants (< 1 yr) and child vs. adult
metabolic differences likely account for this observation.
In addition, as reported in Firestone et al. (2008). OEHHA applied modeling to evaluate
alternative methods for interspecies extrapolation of gas dosimetry in a limited number of
test chemicals. Limited information on the model structures and parameters employed
were provided; however, detailed methods, equations, and model parameters were not
described. Blood Cmax and AUC for parent and metabolite and amount metabolized
were the dose metrics modeled for a 24 hr simulation. Chemical-specific principal effects
(i.e. POE vs. systemic) and thus potential target-tissue doses were not modeled. In
general, the DAFs calculated for this set of chemicals were lower in adults (Gmean =
1.85) and higher in children (Gmean = 1.94) compared to the current default methods.
With the exception of one case (amount of ethylbenzene metabolized), the child/adult
DAF ratios were within a 2-fold range.
Ginsberg et al. (2008) analyzed ozone gas dosimetry in the TB region using a
mathematical model for uptake. The TB model consisted of 15 generations of
symmetrically-branched airway bifurcations. Air was modeled starting at the entrance of
the trachea and thus did not simulate reactions possible in the ET region. The numerical
simulations of reactive gas uptake utilized airway and ventilatory parameters specific to
children of different ages (0-18 yr). The model was exercised to examine the uptake
distribution of ozone along the gas-mucus and mucus tissue interfaces of these children at
a constant inhalation concentration of 0.1 ppm. The results demonstrated that for all ages
and all airway generations, the controlling resistance to uptake was the mucus layer and
the overall Kg was not significantly different across ages. In addition, there were no
significant differences in the predicted flux of ozone to the mucus and tissue for children
3-40
-------
of different ages. The authors concluded that although the analysis was conducted for
ozone, other gases would qualitatively behave in a similar way. However, more highly
soluble gases, such as chlorine and formaldehyde, would likely be absorbed more in the
URT and more proximally in the lungs than less soluble gases such as ozone.
More recently, Valcke and Krishnan (2011) examined the impact of exposure route on
the UFH.TK- A multiroute, steady-state, PBPK model was modified from the literature and
used to compute the internal dose metrics of the area under the parent compound's
arterial blood concentration vs. time curve (AUCpc) and amount metabolized per 24 hours
(AMET). Dose metrics were computed for adults (18-64 yr), neonates (10-30 d), children
(1-3 yr), elderly (65-90 yr) and pregnant women (15-44 yr) for a 24 hour inhalation
exposure scenario to chloroform, bromoform, tri- or per-chloroethylene (TCE or PERC).
The inhalation exposure scenarios were performed at a concentration of 5 (ig/m3
representative of alow, environmental level. Monte Carlo simulations were performed
and the UFH.TK was calculated as the ratio of the 95th percentile value of internal dose
metrics in the various population groups to 50th percentile value in adults. On the basis of
AUCpC, the highest UFH_TK values were demonstrated in neonates for each scenario
compound. The highest UFH_TK computed was 3.6 for bromoform, but in all other cases
the UFH_TK values ranged from 1.2 to 2.2. A synthesis of the results from this study are
presented in Table 3-14. These results are in agreement with those presented by Firestone
et al. (2008) for PERC and chloroform; however, TCE was categorized as having a
UFH_TK> 10 by Firestone et al. (2008) and < 3.16 by Valcke and Krishnan (2011). The
reason for this difference cannot be determined from the limited information provided in
the Firestone et al. (2008) report.
3-41
-------
Table 3-14 Human kinetic adjustment factors (UFH.TK) obtained for inhalation exposure in each
population group using a dose surrogate of 24 hour AUCpc
Substance
Chloroform
Bromoform
Trichloroethylene
Perchloroethylene
Adults (41, 18-64yr)b
median
95th percentile
UFH-TK
15.8
20.2
1.3
25.7
37.5
1.5
21.8
28.8
1.3
37.3
47.2
1.3
Neonates(14, 0-30 d)
95th percentile
UFH-TK
33.4
2.1
93.1
3.6
48.4
2.2
66.6
1.8
Children (2, 1-3 yr)
95th percentile
UFH-TK
25.2
1.6
51.7
2.1
35.1
1.6
58.8
1.6
Elderly (78, 65-90 yr)
95th percentile
UFH-TK
20.4
1.3
37.6
1.5
28.8
1.3
45.8
1.2
Pregnant women (29, 15-44 yr)
95th percentile
UFH-TK
22.9
1.5
44.4
1.7
30.6
1.4
46.4
1.3
Note: AUCpc, area under the arterial blood concentration vs. time curve (ug 24 hr/L)
bShown in parentheses are the median age, range for each population group.
cBolded values indicate the population group with the greater UFH-TK for corresponding internal dose surrogate for each compound.
Source:Reprinted with permission of Elsevier©; Valcke and Krishnan (2011)
Both pharmacokinetics and pharmacodynamics were considered by Liao et al. (2007) in a
study of chloroform toxicity and carcinogenicity. A PBPK and a pharmacodynamic (PD)
model were developed and then linked to produce a hybrid PBPK/PD model to
investigate chloroform toxicity and carcinogenicity. The PBPK model was configured for
rats, mice, and humans with the human configuration expanded to consider different age
groups (1 month, 3 month, 6 month, 1 year, 5 year, and 25 year old) with the age-specific
physiological values being obtained from documented literature sources. The PD model
was used to quantitatively estimate rates for mode-of-action processes known to be
prominently involved in the toxicity of chloroform (metabolism, reparable cell damage,
cell death, and regenerative cellular proliferation). A PD model with chloroform
parameters previously developed for female mice was modified to simulate both hepatic
and renal data in male and female mice and rats. This study used a Bayesian approach
(Markov Chain Monte Carlo algorithm) to analyze and address parameter uncertainty in
both the PD and PBPK models. The human PBPK/PD model was developed using the
rodent PD parameters from Bayesian analysis together with human physiological,
partitioning, and metabolism PK parameters all of which are reasonable and are
3-42
-------
documented within the study. The critical PD parameters for metabolism in children were
based on the adult values with an age-dependent adjustment for the maximum rate of
metabolism (available and documented from literature).
The hypothesized mode-of-action used as the basis of the PD model by the authors was
that carcinogenicity was related to regenerative hyperplasia that, in turn, occurs in
response to cytolethality which occurs when the rate of generation of toxic metabolites
exceeds the capacity of cellular protective and repair mechanisms (Liao et al.. 2007).
Thus the absence of regenerative hyperplasia (as measured by cellular labeling indices,
LI) was a key point of control of a tumorigenic response. The human model was used to
estimate internal doses at steady state over a range of inhalation (and oral) concentrations
for different age groups to identify the threshold for LI below which no cytolethality
would be expected in each age group. The simulations presented in Table 3-15 indicated
that for liver effects, a young child (< 5 years) was more sensitive than adults by a factor
of about 2. For renal effects, however, the results indicated age-related increases in
sensitivity to the toxicity of chloroform with 1-month-old infants nearly 7- to 8-fold less
sensitive than adults, 1-year-olds about 3-fold less sensitive than adults, and no difference
in concentration corresponding to kidney effects between adults and 5-year-old children.
Table 3-15 Air Concentration of Chloroform at Various Ages and Genders Corresponding to
Threshold of Damage in Human Liver and Kidney
Male
1 Month
Female
Male
Female
Male
Female
Male
1 Ymr
Female
Male
5 Year
Female
Male
Arii lit
Female
Air Concentration (ppm)
Liver
5.16a
4.86
4.80
4.79
5.13
4.90
6.07
5.66
6.61
6.81
9.24
12.7
Note: Values generated from model simulations of a PBPK-PD model.
'Results given as point values only, as estimates of variability were problematic in the absence of data on cell proliferation in human
Source: Reprinted with permission of John Wiley and Sons©; Liao et al. (2007)
Kidney
7.56
8.08
2.60
2.85
2.19
2.29
3.17
3.00
1.18
1.35
0.887
1.06
liver and kidneys.
Sarangapani et al. (2003) used a PBPK model to evaluate the effect of age- and
gender-specific lung morphology and ventilation rate on the inhalation dosimetry of
several gases. The gases were selected on the basis of their potential range of reactivity
within the respiratory tract, from reactive and soluble (ozone and isopropanol) to
3-43
-------
relatively insoluble and nonreactive (styrene, vinyl chloride, and perchloroethylene). Ten
age-specific PBPK models were run for males and females from 1 month of age to 75
years. Model structure was typical of PBPK models but simplified to three main axial
compartments of the respiratory tract: the ET, TB, and PU, with the ET and TB each
divided into three lateral subcompartments from airway lumen to circulating blood. The
parameter sources were varied but well documented. Age-dependent changes in
physiological parameters were developed based on sources available in the literature.
Information on BW and VE for various ages for both males and females 1 month to 75
years were based on data reported in U.S. EPA's Exposure Factors Handbook (1997) and
NHANES data (CDC. 1995). Age-specific data collected in the literature also expressed
lung volume and SA on BW allometry across age groups as a relatively constant fraction
of body weight across lifestages. Richly perfused tissue was modeled as 84% of the total
BW minus the volume of the other tissues with the rest of the body (16%) assumed to be
nonperfused tissue. Alveolar ventilation (Qaiv) was assumed to be 67% of VE.
Age-dependent changes in ET airway dimensions were estimated based on in vivo
measurements of growth patterns in children and adults using CT scans where the ET
dimensions in infants and children were computed by scaling from adult values on a
proportional basis. Similar scaling from other literature sources was also done for
children's values of both TB and PU measures. The same Hb/g was used for all age
groups. Biochemical parameters were varied with age (e.g., relative activity of CYP2E1
26.1% at 1 month to 90% at 15 years; and alcohol dehydrogenase (ADH) 24.9% at 1
month to 83.6% at 25 years). Dose metrics evaluated included parent and metabolite
concentrations in blood, liver, and lung. Results for the dose metrics were expressed
relative to the young adult (25-year-old) model which were all set at unity.
The results from the Sarangapani et al. (2003) model indicated that tissue dose metrics at
any age generally fell within a factor of 2 of the young adult values for parent ozone,
vinyl chloride, styrene, isopropanol, and perchloroethylene. Little variability due to
gender was apparent at any age for any of the gases or metrics examined. The only
exceptions were those observed in early childhood (either gender), where dose metrics
(especially for metabolites) were as much as 12 times higher for a 1-month-old child than
young adult values, declining to 2 times by age 5-10 years, for these same compounds.
This is shown for the parent isopropanol and its water soluble metabolites (Table 3-16).
3-44
-------
Table 3-16 Age-Dependent and Gender-Specific Dose Metric Comparison of Inhaled Isopropanol
Parent Chemical Concentration
Aac
Male Female
1 Month 1.75 1.74
3 Month 1.77 1.78
6 Month 1.77 1.75
1Year 1.54 1.54
SYear 1.25 1.18
10 Year 1.05 1.03
15Year 1.09 1.14
25 Year 1 1
50 Year 0.94 1.00
75 Year 1.04 1.03
Metabolite Concentration
Male
8.02
6.68
5.70
4.12
1.98
1.53
1.46
1
0.80
0.89
Female
11.44
9.14
8.01
5.96
2.55
2.04
1.70
1
0.82
0.93
Note: Comparisons presented as % ratio of metric at a specific age to the 25-yr-old adult set at 100%.
Source: Reprinted with permission of Informa Healthcare©; Sarangapani et al. (2003)
Pelekis et al. (2001) developed a PBPK model for adults of low (50 kg) and high (90 kg)
body weights and for a 10 kg child (1 or 2 years old). The model was applied to
inhalation exposures of dichloromethane, tetrachloroethylene, toluene, m-xylene, styrene,
carbon tetrachloride, chloroform, and trichloroethylene. The parent compound
concentrations in arterial blood (CA) and venous blood (CV), and tissues (Ctlssue) (but no
metabolites) were evaluated. The values (and ranges) of the physiological and
mechanistic parameters were obtained from literature cited in the study for all of the
gases studied, including ranges applied for the blood:air partition coefficient and
ventilation. Metabolism was described and limited to the liver only. The intent of the
study was to characterize various concentration metrics in model simulations in which the
mechanistic parameters varied between low and high adult values. Data were unavailable
for children to determine high and low estimates for these parameters, thus average
values for the child (childaverage) were used for comparison. The ratios of the metrics from
these different runs characterize the pharmacokinetic behavior of the child relative to the
adult (e.g., adulthigh/childaverage). The exposure scenario simulated was 1 ppm continuous
for 720 hrs (30 days).
The simulation results (i.e., the concentration of the parent compound in various tissues
and compartments) for all of the gases were expressed as the adult^gh /childaverage ratio.
These ratios indicated that the estimation of concentrations in children's blood were
about the same as for the adult. With other tissues metrics, however, values were
considerably higher in a few instances. For example, the adulthlgh /childaverage ratio for the
concentration in the liver (which was dependent on metabolism) was predicted as 0.033
for styrene, 0.037 for m-xylene, 0.061 for trichloroethylene, 0.092 for dichloromethane,
and 0.11 for chloroform. These predictions indicate up to 30-fold higher concentrations
of the VOC chemicals in child liver than in adult liver. The average adulthlgh /childaverage
3-45
-------
ratios for the various dose metrics estimated for the composite runs by Pelekis et al.
(2001) are shown in Table 3-17.
Table 3-17 Tissue Concentrations in Various Compartments Expressed as Adult/Child (1 to 2
years old) Ratios for 8 Different Gases
Gas
Dichloromethane
Tetrachloroethylene
Toluene
m-Xylene
Styrene
Carbon tetrachloride
Chloroform
Trichloroethylene
Average ± SD
Adulthigh/Childaverage Ratios of Concentrations3
Venous Blood
0.70
1.61
0.86
0.50
0.34
1.81
0.78
0.77
0.92 ±0.52
Arterial Blood
0.91
1.74
0.98
0.63
0.45
2.20
1.02
0.97
1.11±0.58
Fat
0.25
0.47
0.27
0.17
0.12
0.60
0.28
0.27
0.30 ±0.16
Liver
0.092
0.75
0.34
0.037
0.033
0.57
0.11
0.061
0.25 ±0.28
'Steady-state concentration ratios for 1 ppm continuous exposures.
Note: Initial values are all from PBPK simulations.
Source: Reprinted with permission of Elsevier©; Pelekis et al. (2001)
In an effort to evaluate the potential effects in the nasal cavity of inhaled methyl iodide
(Mel) exposure, a PBPK model was developed, complete with parameters for sensitive
populations and lifestages, including children (Sweeney et al.. 2009). For the human child
(3 months to 15 yr), the basic measures of the total nasal surface area and the total nasal
volume were obtained from literature. Further subdivision of these measures into nasal
tissue types (olfactory and respiratory epithelium) in the child age categories was based
on literature sources. Other requisite parameters, such as the thickness of the nasal tissues
to the capillary beds were also obtained from the literature (Inagi. 1992). Breathing rates
were based on ICRP data. The modeled point-of-departure for the effect of Mel in the
nasal tract was a decrease (either 25% or 50% decrement from untreated levels) in
glutathione (GSH) concentrations in the olfactory epithelium. The modeled turnover rates
of GSH determined for the adult rat were used for all species and lifestages. Fetal human
tissue GSH concentrations were identified from literature sources given in the study. This
adult human model indicated that depletion of GSH in the dorsal olfactory epithelium to
50% of control would be achieved after 24 hours of exposure to 72 ppm Mel. For
workers exposed for 8 hrs, 50% GSH depletion would be achieved by the end of the shift
at an exposure concentration of 110 ppm. At a target POD of 25% GSH depletion at 24
hrs, the 24-hr adult value was 36 ppm and the 8-hr (worker) value was 50 ppm. When
configured for the 3-month-old child the corresponding 24-hr concentration for 25%
depletion of olfactory GSH was 8.2 ppm under these conditions. No other age-related
results were given in the study. This concentration differential for the POD, 36 ppm for
the adult and 8.2 ppm for the 3-month-child, indicates differential sensitivity of 3-4 fold
3-46
-------
resulting from a combination of biochemical (e.g., GSH turnover) and physiological (e.g.,
respiration rate) factors. However, it should be noted that the equivalent rat exposure
concentration (associated with a 50% depletion of GSH) upon which the adult and child
modeled HECs were based was 3.8 ppm (21 ppm for 6 hr/day, 5 d/wk), indicating that
humans including children would be less sensitive than rats to this effect.
Clewell et al. (2004) constructed a PBPK lifestage model specifically to evaluate age-
and gender-specific differences in tissue dosimetry for oral, dermal, and inhalation
exposures to a range of chemicals with various physical and toxic properties. The model
was mostly parameterized using equations that described various age-dependent
alterations derived from U.S. EPA (1997). which was also the source for ventilation rates
(m3/day); pulmonary ventilation for various ages were converted to alveolar ventilation
based on the assumption that alveolar ventilation is approximately two-thirds of
pulmonary ventilation. The results for the isopropanol inhalation model are the only ones
discussed here; however, the predictions of this age-dependent model were only able to
be validated against human kinetic data for the adult. The arterial blood concentrations of
isopropanol and acetone (the principal metabolite of isopropanol), were estimated for a 1
ppb continuous inhalation exposure and summarized in age-group ranges of birth to 6
months, 6 months to 5 years, 5 to 25 years, and 25 to 75 years. In general, the model
estimations for the average internal concentration of inhaled isopropanol and its
metabolite acetone varied 2 to 4-fold across the range of lifestages. The highest dose ratio
(constructed from the lifestage/average daily inhalation dose for a 25-year-old adult)
among the lifestages was 2.0 for isopropanol (birth-6 months) and 3.9 (birth-6 months)
for acetone.
Ginsberg et al. (2002) investigated child/adult pharmacokinetic differences through
analysis of pharmacokinetic (PK) data from 45 different chemicals, nearly all therapeutic
drugs and all administered by routes other than inhalation. In an initial metabolic
evaluation, the drugs were classified as to their excretion: unchanged in urine, CYP
(various) metabolism, glucuronidation, sulfation, GSH conjugation or unclassified. The
infants/children were classified in age as premature neonates (< 1 week), full-term
neonates (< 1 week), newborns (1 week-2 months), early infants (2-6 months), toddlers (6
months-2 years), preadolescents (2-12 years), adolescents (12-18 years) and adults. There
were data from 118 adults and 248 infants/children. The kinetic parameters evaluated
included AUC, clearance, Cmax, half-life (ti/2), and volume of distribution (Vd).
Relationships between age groups and the kinetic parameters were evaluated by
regression analysis.
The combined results showed that, for those chemicals with clearance data (27
substrates), premature to 2-6 months of age infants showed significantly lower clearance
(P<0.01) whereas 6-month-old to 12-year-old children had significantly higher clearance
(P<0.0001) than adults. The combined results (40 substrates) indicated also that the
youngest age groups (premature neonates, full-term neonates, and newborn infants up to
3-47
-------
2 months) tended to have longer half lives (average 2-to-4-fold) than adults whose levels
were attained by infants 2-6 months of age. Other results included those for the chemicals
identified as CYP1A2 substrates (caffeine and theophylline) in which neonates to infants
2 months of age showed about 4 to 9-fold longer half-lives than adults while older age
groups 6-months to 12 years had significantly shorter half-lives than adults. A similar
pattern was observed with those chemicals thought to be metabolized primarily through
CYP3A.
These data are for drugs orally administered, rather than from toxics being inhaled, but
nonetheless are relevant to situations involving dosimetry in the extrarespiratory (ER)
region of children versus adults, and thus indicate a potential for greater susceptibility in
children. These data also demonstrate empirically the prominent feature and likely
mechanism of children's susceptibility, decreased clearance functions.
3.6.3 Respiratory Tract Flow Models
Garcia et al. (2009) obtained the MRI or CT head scans of seven individuals including
those of two children, a male (7 years) and a female (8 years) and five adults in a
vanguard study to examine inter-individual variability of nasal air flows in human
subjects. Several prior studies had shown that actual airflow patterns in the nasal tract of
both animals and (adult) humans is highly non-uniform with highly localized areas of
flow that have been correlated with (at least in laboratory animals) areas of focal
pathology in air exposures to reactive gases. Thus, these scans allowed for an initial
evaluation of the extent of variability of these flows at the level of the individual. (All
subjects or their representatives signed a consent form agreeing with the use of their
scans.) These scans were successfully utilized to conduct anatomical measurements and
to construct mesh configurations necessary for conducting CFD of the main nasal
chamber (i.e., anterior to the nasopharyngeal region) for each individual. Breathing rates
for the flow simulations were set at 5.5 L/min for the 7-year-old boy and 5.8 L/min for
the 8-year-old girl with flows for the adults each allometrically adjusted with a final
range of between 6.8 and 9.0 L/min. Simulations of steady-state inspiratory airflow were
conducted using commercially-available CFD software. Simulations of nasal uptake of
inhaled gas (concentration in ambient air defined to be 1 ppm by volume) were conducted
under one of two boundary conditions, one to simulate a maximum gas uptake and a
second boundary condition to simulate moderate uptake (approximately 80% of
maximum) at the nasal tract walls. Results of the study are along several lines. The
simulations predicted that, under both boundary conditions, gas was rapidly absorbed by
the nasal mucosa once it entered the nostrils. At the end of the nasal septum, gas
concentration in the inspired air had dropped to -13% and -29% of the inlet
concentration for the maximum and moderate uptake scenarios, respectively. The spatial
distribution of wall fluxes, especially under the maximum uptake boundary condition,
3-48
-------
were shown to be highly non-uniform for all scans including those of the two children.
Further analysis of the subjects showed that the extent of the non-uniform flows (where
areas of non-uniformity were divided into categories of increasing mass flux) was not
appreciably different among the subjects, including between adults and the two children
(the minimal number of subjects precluded any statistical analysis). Additional analysis
also showed that the overall rate of uptake in the nasal region, although highly
non-uniform under localized internal conditions as shown by this study, was very similar
from one individual to the next with no apparent differences between adults and the two
children. Importantly, delivered dose estimated in terms of maximum (99th percentile) or
average flux was not different between adults and children. These principal results from
the maximum uptake condition, including some of the first available ET surface areas for
children, are shown in Table 3-18.
Table 3-18 Summary Listing of Findings on Morphometry and Gas Flow/Uptake Simulations for
Human Nasal Cavities
Parameter (units)
Gender
Age (years)
ET area (cm2)
ET volume (ml)
Total gas uptake, maximum conditions (%)
Average flux, left cavity (10"8 kg /sm2)b
Maximum fluxc, left cavity (10"8 kg/sm2)b
Subjects
Male3
53
20,085
18.0
93.5
1.8
10.8
Male
NA
23,219
26.5
93.1
1.6
11.0
Adults
Female
NA
16,683
15.4
92.4
1.5
10.8
Children
Female
NA
20,688
23.8
89.2
1.2
10.6
Female3
37
17,752
18.7
91.5
1.4
10.8
Male
7
12,093
10.7
92.0
1.9
11.8
Female
8
13,027
13.7
88.2
1.6
12.3
3Data obtained from repaired casts.
bGas absorption rate
"The 99th percentile flux (i.e., the flux value below which 99% of flux values fall)
NA= data not available
Source: Reprinted with permission of Informa Healthcare©; Garcia et al. (2009)
In a follow-on study from Garcia et al. (2009). Schroeter et al. (2010) utilized the reactive
gas hydrogen sulfide (H2S) to expand the investigation of interhuman variability of nasal
dosimetry and anatomically accurate CFD models of the nasal passages of five adults and
two children generated from MRI or CT scan data. Preceding studies showed that
inter-individual differences in nasal anatomy affect the distribution of airflow and
(simulated) uptake from the airflow inside the nose, such that highly localized areas with
potential for widely varying deposition were seen among individuals (e.g.. Garcia et al..
2009). It has also been established that H2S is a potent toxicant of the respiratory tract
with particular and specific effects in olfactory tissue. The aim of this study was to
characterize the variability of H2S dose to the olfactory region of humans arising from
inter-individual differences in nasal anatomy, airflow, and inspiratory uptake patterns.
3-49
-------
This study used essentially the same conditions of modeling with the same computational
meshes used by Garcia et al. (2009). Several of the meshes were, however, repaired and
further refined to provide more complete olfactory regions representative of normal or
decongested nasal anatomy. Olfactory regions were mapped into the nasal models of all
subjects as consistently as possible based on the prior descriptions of the extent of
olfactory epithelial in humans. Steady-state inspiratory airflow rates were applied based
on body weight allometry. The H2S specific kinetic parameters used were previously
estimated by the authors by fitting in vivo uptake data in rats, then allometrically scaled
to humans based on nasal surface areas. The air phase mass transfer was solved
numerically under an assumed equilibrium condition across the air-tissue interface. Flows
were simulated at three different concentrations, 1, 5 and 10 ppm. Comparisons among
individuals were made for the 99th percentile flux (i.e., the flux value below which 99%
of flux values fall) and average flux in the olfactory regions at an exposure concentration
of 1 ppm. The olfactory surface area in each model was partitioned into bins based on
levels of H2S tissue flux to examine the distribution of surface area by wall flux levels.
Results included morphological measurements in human adults and children of nasal
cavity surface areas and estimates of olfactory epithelia and airflow apportionment. The
modeling results in terms of average flux, maximum flux, and distribution of flux ranges
within the target area of olfactory epithelium showed uniform responses despite the
morphological ranges characterized. Differences in nasal anatomy and ventilation among
adults and children were not predicted to have a significant effect on H2S dosimetry in
the olfactory region (Table 3-19). The 99th percentile flux ranged from 153.1 to 170.1 in
adults compared to 149.2 and 159 in children, while the average flux ranged from 12.2 to
13.6 in adults compared to 11.8 and 12.1 in children.
Table 3-19 Selected Morphologic and Simulated Modeling Results of Hydrogen Sulfide
Dosimetry in Casts of Human Nasal Cavities
Parameter (units)
Gender
Age (years)
Surface area of main nasal cavity (cm2)
Surface area of olfactory region (cm2)
Olfactory airflow allocation (%)
99th percentile flux (pg/cm2-s) @ 1ppm
Average flux (pg/cm2-s) @ 1 ppm H2S
Male3
53
198.7
14.4
4.8
167.7
13.6
aData obtained from repaired casts.
NA= data not available
Source: Reprinted with permission of Informa Healthcare©; Schroeter et al.
Male
NA
231.5
11.5
5.5
170.1
13.5
(2010)
Adults
Female
NA
167.3
10.5
7.9
158.9
12.7
Subject
Children
Female
NA
207.9
9.9
2.6
161.3
12.8
Female3
37
177.0
11.2
4.9
153.1
12.2
Male
7
118.9
9.1
16.2
149.2
12.1
Female
8
135.1
9.6
1.6
159.0
11.8
3-50
-------
Allen et al. (2004) studied the mechanisms of gas and aerosol transport in the human
respiratory system in the upper airways of a pediatric subject (male, age 5 years).
(Approval was granted by the appropriate Institutional Review Board to utilize blinded
MRI images.) An in vitro reconstruction of the subject's anatomy was produced from
MRI images with axial MRI slices every 3 mm. The model included the oral cavity,
oropharynx, larynx, trachea, and carinal bifurcation. The computational model was
directly imported into a CFD software program; an unstructured tetrahedral mesh was
fitted to the topology (26,852 nodes total). Airflow was then simulated using this
computational mesh in the CFD software. A computational model of an adult upper
airway was derived from the cadaver of a female, aged 84 and constructed in an identical
fashion (15,096 nodes total). The CFD airflow simulations in the adult model were
validated through comparisons made between numerical simulations and experimental
measurements (determined indirectly from the centerline velocity) using Phase Doppler
Interferometry. Flow fields were solved under steady inhalation at 6.4 and 8 liter/min,
and the resulting computational data were compared to experimental results obtained
with this adult model to determine CFD solution validity. The numerical simulations
provided an accurate representation of axial velocities and turbulence intensity.
Simulation results on flow resistance, axial velocities, secondary velocity vectors, and
turbulent kinetic energy and intensity are presented for the pediatric case.
Particularly detailed information was gathered on the nature of the laryngeal jet, which
was expected to have the dominant effect on flow features in the upper airways. The
development of the laryngeal jet could be clearly seen in the anticipated immediate
post-epiglottal area at both flow rates studied. Contours of axial velocity, secondary
velocity vectors, and turbulent kinetic energy and intensity of the modeled airway were
all characterized in the study. The highest turbulent intensity was noted to occur
immediately downstream of the glottic restriction followed by areas at the carinal
bifurcation. Other key features of the modeled flow included skewed velocity profiles
due to bends in the airways and recirculation zones due to abrupt changes in cross
sectional area. The authors stated that these results all qualitatively agreed with the
computational outcomes of the adult model. An unanticipated outcome between the
pediatric and adult models was observed when the laryngeal jet from an adult model was
compared to the laryngeal jet in the pediatric model based on the same tracheal Reynolds
number. This comparison showed higher axial velocities, recirculation and turbulent
kinetic energy in the adult than in the pediatric case. Although the intensity of turbulence
in the laryngeal jet was comparable in the adults and in the child, it should be recognized
that adults have much higher tracheal Reynolds numbers than children. A similar trend
was noted for axial velocities which were higher in the pediatric model than would be
expected from measurements in adults at similar tracheal Reynolds numbers. The authors
speculated that the critical range for the dependence of turbulence on tracheal Reynolds
numbers appears to differ between adults and children for reasons that are not clear.
However, that the authors did not discuss the fact that the laryngeal area is flexible and
3-51
-------
that the larynxes of the models used may not be in comparable states of openness. The
authors concluded that there was reasonable agreement between the location of flow
structures between adults and children, suggesting that an unknown length scale
correlation factor could exist that would produce acceptable predictions of pediatric
velocimetry based off of adult datasets.
3.6.4 Respiratory Tract Growth
It has been well established that the human respiratory system passes through several
distinct stages of maturation and growth that involve branching morphogenesis and
cellular differentiation during the first several years of life and into adolescence
(Pinkerton and Joad. 2000). The proportion of surface area to ventilation volume may be
markedly different during these developmental stages. The significance of these
disproportions with regard to toxicant exposure overall or to the sites of active cellular
differentiation have yet to be elucidated.
The major proposed processes in human lung growth and development are (1) an increase
in numbers of alveoli via septation of elementary saccules, followed by (2) increases in
dimensions of all of the lung structures, including alveolar size and, most prominently,
the diameter of airways, followed by (3) distension of lung due to changes in the
mechanical properties of the chest wall. These changes are postulated to result in a
relative under-distension of the lung followed by a relative over-distension (Zeltner et al..
1987 for general review). De Jong et al. (2003) postulated that indications of these
processes could be determined through in situ scanning and visualization techniques.
Therefore an institutionally sanctioned study was conducted where the CT scans of 35
children (age range from 15 days to 17.6 years of age; 17 males, 18 females) were
obtained and examined for these indications of growth and development. CT analysis
allowed for accurate calculation of lung volume and estimation of lung density and
weight (lung density x volume). Lung expansion may also be obtained by subtracting the
inverse of the density of tissue from the inverse of the CT-measured lung density. All
data were then plotted with respect to age (or body length) and analyzed. The data on
volume showed a decline from birth to 2 years of age and an increase thereafter. This
finding would be anticipated as added alveoli are of uniform size and divisions of
existing airspace into smaller units via septation would cause the gas volume to fall. The
subsequent increase in gas volume of tissue from age 2-8 years would be consistent with
expansion in the size of alveoli in combination with a gradual increase in functional
residual capacity (FRC) due to changes in the mechanical properties of the lung and chest
wall.
In a companion study de Jong et al. (2006) used CT scans from a group of 50 young
individuals (age range 0 - 17.2 years) to obtain estimates of various lung dimensions also
3-52
-------
through the period of growth. Clinical CT scans were performed and analyzed as above
for lung weight, gas volume, lung expansion, lung surface/volume ratio, airway wall area,
airway lumen area, airway lumen perimeter, arterial area and airway surface length/area
ratio. The authors discussed the nature of these ratios in relation to length and growth of
the individual but did not give specifically determined estimates of measures such as
surface areas. For example, lung alveolar surface area to total lung volume ratio (S/V)
was calculated using the lung expansion values at total lung capacity (TLC) per the
following equation:
jj/y ]unp_e6:84(0:32xlung expansion at TLC)
Equation 3-6
The regression of these ratios against other growth parameters, such as body length,
suggested that the relationship between these various measures was closely linked.
Collectively these results provide functional indications of lung growth processes using
noninvasive methods and demonstrate that CT scans can be used to provide valuable
information about normal lung growth in addition to the more typical application of
diagnosis of lung disease.
Rao and coworkers (2010) evaluated lung growth and development in vivo in infants and
toddlers using multi-slice CT. The developmental process is thought to be sequential in
terms of the alveoli, with new alveoli being added until about 24 months of age followed
by alveolar expansion with no new alveoli added after 24 months. The high resolution
capability of CT was applied to a group of 38 subjects (14 male, 24 female) of ages in
this range (17 to 142 weeks; 4 to ~ 36 months). Subjects were excluded if they were born
at <37 weeks gestation, had congenital cardiorespiratory abnormalities or histories of
wheezing, were hospitalized for respiratory illness, or used asthma medications. The
study was approved by an institutional review board and signed consent was obtained
from the parents. These high-resolution scans provided a basis for significant measures.
Lung volume was calculated by summing the voxels (i.e. a volume element in 3D space)
within the lung, while lung density was calculated from the X-ray attenuation values, and
lung tissue weight was calculated by multiplying lung volume by lung density. Tissue
volume was calculated as lung or tissue weight divided by tissue density, which was
assumed to be 1.065 g/mL; air volume was obtained as lung volume minus tissue volume.
The measures were regressed against body length and showed that total parenchymal
lung volume, parenchymal air volume, and parenchymal tissue volume increased
significantly with increasing body length. This in vivo assessment suggests that the
growth of the lung parenchyma in infants and toddlers occurs with a constant relationship
between air volume and lung tissue, which is consistent with lung growth occurring
primarily by the addition of alveoli rather than the expansion of alveoli. In addition, the
central conducting airways grow proportionately in infants and toddlers.
3-53
-------
The pulmonary growth sequence in early life of alveolar septation followed by alveolar
expansion was examined by Balinotti et al. (2009) with pulmonary function testing. The
basis of the hypothesis relates to the ratio between pulmonary diffusion capacity of
carbon monoxide (DLCO) and alveolar volume (VA). During the process of
alveolarization, usually considered to be in the first two years of life, this ratio would
remain constant whereas during alveolar expansion, i.e., in children older than 2 years, it
would decrease. The authors measured DLCO and VA using single breath-hold
maneuvers at elevated lung volumes in 50 sleeping infants and toddlers less than two
years, between the ages of 3 and 23 months. There were an equal number of males and
females. No subjects had cardio-respiratory malformations, and their respiratory history
was negative for wheezing, asthma, treatment with asthma medications, or hospitalization
for a respiratory illness. The study was approved by appropriate institutional review
boards. Both alveolar volume and pulmonary diffusing capacity increased with increasing
age in both male and female children. Significantly, ratio of pulmonary diffusing capacity
to alveolar volume remained constant in this age group. The constant ratio for DLCO/VA
in infants and toddlers is consistent with lung growth in this age occurring primarily by
the addition of alveoli rather than the expansion of volume.
Zeman and Bennett (2006) employed in vivo methodology, aerosol-derived airway
morphometry (ADAM), to measure the age-related changes in air space caliber of the
small airways and alveolar dimensions. The principal of ADAM related to predictable
gravitational settling of small inhaled particles to infer the vertical distance or effective
air space dimension, (BAD), that the particles must have settled to become lost to the
airway wall. ADAM involves individuals inhaling to TLC a particle aerosol of known
size characteristics followed by breath-holds for 0-10 seconds and (non-deposited)
particle recovery upon exhalation. EADs would then be associated with their volumetric
depth into the lung by the exhaled volume through the principle of first in, last out. Those
EADs calculated from the aerosol at the beginning of the exhalation would be assumed to
be representative of the proximal airways at that depth, and those calculated from the
aerosol toward the end of the exhalation would be representative of distal airways.
Certain EADs are closely associated with morphological correlates: EADmm is related to
alveolar diameter, BAD^^ is related to transitional bronchiole caliber, VED^^ (volume
of gas required to reach transitional bronchioles into the lung) is related to anatomical
dead space.
The subjects recruited from the general local population included 53 children (6-22
years) and 59 adults (23-80 years). Subjects had no smoking history, no history of lung
disease, and no recent history of acute respiratory infection or viral illness within the
previous 4 weeks. Informed consent was obtained from each volunteer. This technique
was employed to derive EADs that was then used to estimate alveolar diameters,
transitional bronchiolar caliber, and the volume of conducting airways anatomical dead
space. Data were collected, then regressed according to age. Alveolar diameters were
3-54
-------
found to increase with age, from 184 (im at age 6 to 231 (im at age 22 based on the
regression equations derived. This observation would account for the increase in TLC
observed over this age range. The caliber of transitional bronchioles did not increase with
TLC (average 572 (im), but did increase with subject age and height when the entire age
range of 6-80 years was included (Zeman and Bennett. 2006). The anatomical dead space
scaled linearly with lung volume, but relative to TLC did not change with age, averaging
7.04 ± 1.55% of TLC. The authors concluded that from childhood (6 years) to adulthood
a constant number of respiratory units is maintained; however, both the smallest
bronchioles and alveoli expand in size to produce the increased lung volume with
increased age and height.
It is believed that humans grow new alveoli from a few weeks before term birth until
approximately 8 years of age, after which the alveoli are thought to enlarge as the lungs
increase in volume or size with no new alveoli formed. To this end, Altes and coworkers
(2004a) examined the apparent diffusion coefficient (ADC) with a gaseous contrast agent
for MRI, hyperpolarized helium-3 (3He), in a cohort of twelve individuals. An increase in
ADC is a measure of volume maturation. It was expected that in the pediatric age group,
the increase in alveolar size with increasing age will be reflected in an increase in 3He
ADC with age. The age range of the 12-member cohort was 7 to 29 years (mean 15.6,
standard deviation 6.9 years). All 12 of the subjects had homogenous appearing ADC
maps. Comparing the mean ADC with other measures of maturation or lung volume gave
correlation coefficients of 0.74 with height, 0.64 with weight, 0.76 with forced vital
capacity (FVC) in liters, 0.81 with the predicted FVC based on the subject's age and
height, and 0.34 with the percent predicted FVC. In summary, it was found that the mean
ADC increased with age in the pediatric population and that the mean ADC was lower in
the pediatric age group than in young adults. These observations suggest that the pediatric
subjects had smaller airspaces than the young adults. Further, the variability of the
airspace structure, as measured by the standard deviation of the ADC values, did not
change with age, as expected. Thus 3He diffusion MRI of lung appears to be able to
detect this normal maturation process of increased lung volume via increases in the size
of the functioning alveoli.
Altes et al. (2004b) used advanced imaging techniques to detect age-related development
in lung microstructure that relate to both lung volume and surface area. 3He diffusion
magnetic resonance scanning produces in vivo images of tissues weighted as to water
diffusion through local microstructure. MRIs were acquired for each of 29 individuals (2
separate trials for each), aged four to 30 years, and used to determine the mean ADC and
lung volume for each subject. The mean ADC was reported to increase with increasing
subject age (r = 0.8; P < 0.001), with a 55% increase in mean ADC from the youngest (4
years) to oldest (30 years) subject. The lung volumes measured on MRI were highly
repeatable for the two acquisitions (r = 0.980) and also reflected increased volumes
3-55
-------
concordant with the ADC. These advanced imaging results gave functional indications
that alveoli increase in size rather than number during childhood.
3.6.5 Child Data Collations
Menache et al. (2008) generated quantitative whole-lung models from silica casts of the
lungs from 11 subjects between 3 months and 21 years of age. The models were based on
a combination of cast data and published information on distal airway dimensions and
were inclusive of the conducting airways (trachea through terminal bronchioles), the
respiratory bronchioles, and the alveolar airways, which include alveolar ducts and sacs.
Parameters evaluated from the data included airway generation number count, length and
diameter of terminal bronchioles and alveolar ducts, acinar length and alveolar
dimensions (assumed spherical), and total alveolar number. Further estimates from these
parameters and reasonable assumptions were made for alveolar volumes and the
physiological volumes of TLC and FRC. Model dimensions for the conducting airways,
as well as the estimated dead space, for all children fell within the range of the limited
published information. The assumptions and estimates used produced results that were
reasonably consistent with available physiological data for children 8 years and older.
The predicted TLC for the older individuals (aged 8 to 21 yr) fell within or near the range
arising from published scaling equations. However, the models for children 3 years of
age and younger resulted in predicted TLCs well below those predicted using these same
equations by as much as an order of magnitude (data not shown). Another unexpected
result was the total number of model calculated alveoli compared to the published
number of alveoli as a function of age. As shown in Figure 3-5, the calculated number of
alveoli increased linearly as a function of age in contrast to the data of Dunnill (1962) and
Thurlbeck (1982). This suggested that the fixed relationship between respiratory airway
volumes and alveolar volumes assumed for all ages was incorrect and that the
relationship must be different in the younger children. These differences might be
explained by growth in early childhood when the alveolar region is growing more than
the airways. The airways show symmetric growth since they are complete, while the
alveoli are increasing in both number and size. These results suggest that the geometry
model airway dimensions for all ages are appropriate for use with dosimetry models;
however, they also point out a need for a greater understanding of lung development for
children 3 years of age and under.
3-56
-------
500
._, 400
g
I 300
o
1 200
100
0
Dunnill
Thurlbeck (1982) (male)
Thurlbeck (1982) (female)
Model Values
0 5 10 15 20
Age, years
Figure 3-5 Alveoli count per lung as a function of age
25
Source: Reprinted with permission of Informa Healthcare©; Menache et al. (2008): using data from Dunnill
(1962) and Thurlbeck (1982)
Ogiu et al. (1997) presented detailed physical mass measurements of various organs in
4,667 Japanese subjects, aged 0-95 years, including 3,023 males and 1,644 females.
Analyses of age-dependent changes in weights of the brain, heart, lung, kidney, spleen,
pancreas, thymus, thyroid gland, and adrenal gland and also of correlations between
organ weights and body height, weight, or surface area were carried out. It was concluded
that organ weights, including lung, in the growing generation (under 19 years) generally
increased with a coefficient expressed as (body height) * body weight0 5. Specific
coefficients were derived for both right and left lungs and for both males and females. It
was also noted that adult males had heavier lungs than adult females, and that the
male:female lung weight ratios were nearly the same, 1.27 for the right lung and 1.28 for
the left lung. The age-specific weights presented in this study for lungs only, 0-15 years
of age, are shown in Table 3-18.
3-57
-------
Table 3-18 Lung Weights (Right and Left) of Males and Females from Birth to Adulthood
Age
0
1 mo
2
3
4
5
6
7
8
9
10
11
1yr
2
3
4
5
6
7
8
9
10
11
12
13
14
15
N
39
5
11
3
11
4
8
6
7
1
4
1
15
7
17
11
8
4
13
10
10
12
10
12
8
10
14
20-24 68
Source:
Left Lung
Average wt
(g±SD)
22.3 ±5.7
42.1 ±12.7
48.4 ±6.6
46.3 ±6.4
51.1 ±9.7
51.8±16.1
55.5 ±12.1
72.2 ±8.5
66.5±8.9
66.0
71. 9 ±24.9
50.0
83.9 ±20.2
100.5 ±28.4
108.4 ±28.3
118.5 ±38.5
138.1 ±38.2
194.8 ±16.0
170.8 ±61. 6
169.8 ±54.7
232.5 ±75.5
245.6 ±71. 9
254.8 ±77.1
370.8 ±130.1
248.5 ±131. 9
402.5 ±146.1
442.0 ±155.6
363.8 ±129.1
Males
N
40
6
12
3
11
4
8
6
8
2
4
1
15
7
16
10
9
4
14
10
9
13
10
12
8
10
13
61
Females
Right Lung
Average wt
(g±SD)
28.4 ±8.0
49.3 ±16.1
56.6 ±11.0
62.7 ±11. 7
62.7 ±11. 5
58.0 ±18.6
68.3 ±12.4
86.7 ±12.1
82.2 ±19.0
108.0 ±31.1
77.7 ±27.6
62.0
93.9 ±21. 9
101.4±21.2
129.4 ±36.0
122.8 ±32.2
159.7 ±34.3
230.3 ±18.4
186.9 ±63.7
204.4 ±63.7
243.3 ±60.0
255.2 ±96.9
298.5 ±78.1
398.8 ±129.6
383.4 ±131. 8
467.0 ±203.0
500.3 ±127.7
444.9 ±164.6
Left Lung
N
54
4
7
6
5
7
6
1
5
8
3
-
22
14
11
4
4
3
8
8
9
5
10
6
4
6
8
37
Average wt
(g±SD)
23.1 ±7.1
38.7 ±7.7
45.7 ±9.9
50.2 ±8.1
51.5±12.8
48.3 ±10.1
62.1 ±6.9
55.0
62.0 ±10.2
67.6 ±12.9
53.3 ±10.4
-
76.8 ±23.7
94.8 ±26.7
112.5±21.0
117.3 ±23.6
113.0 ±50.6
143.3 ±12.6
163.8 ±44.2
215.0 ±50.1
208.3 ±62.8
314.2 ±42.0
287.5 ±77.4
289.2 ±89.4
269.5 ±93.2
339.3 ±54.4
297.6 ±191. 9
343.9 ±118.1
N
52
4
7
6
4
8
6
1
5
8
3
-
23
13
11
5
5
4
8
8
9
4
9
6
4
6
7
41
Right Lung
Average wt
(g±SD)
29.1 ±8.3
43.8 ±8.1
52.2 ±8.7
66.3 ±15.6
61.6±14.5
58.9 ±9.6
70.2 ±6.8
68.0
74.8 ±16.8
81.3±16.1
67.3 ±9.3
-
87.4 ±30.4
107.7 ±32.0
117.9 ±25.8
158.4 ±43.2
128.6 ±43.2
197.5 ±62.9
200.0 ±47.8
242.5 ±70.3
247.9 ±81. 2
368.0 ±85.4
300.6 ±90.1
280.0 ±107.9
303.5 ±62.9
389.2 ±72.3
344.0 ± 224.8
363.6 ±122.8
Reprinted with permission of Lippincott Williams & Wilkins©; Ogiu et al. (1997)
In a translated
Japanese
study, Inagi (1992)
described
the collection
and measurement of
the heights of the mucous membrane in the human nasal septum from 74 cadavers,
including 5 males and 4 females classified as "fetal/infant," and 5 males and 3 females
aged 1 to 19 years referred to as the "infant/adolescent" group, as well as older aged
groups. The purpose of the study was to examine histological changes in mucosal tissues
although measurements were made in relation to age including heights of the mucous
membrane, including both the epithelium and the underlying lamina propria. The average
height for the epithelium of the "fetal/infant" group was estimated to be ~0.4 (im with a
range of-0.35 - 0.5 (im. For the remainder of the groups, the average and range of height
3-58
-------
was estimated to be -0.7 (im with a range of-0.4 - 0.9 (im. Estimation of the lamina
propria heights (described and given as being from the convex and concave sides of the
nasal septum) yielded: average height for fetal/infant group -500 (im with a range of
-300 - 700 (im; for the remainder of the groups the average and range of height was
estimated to be -900 (im with a range of-400 - 1,500 (im. Such data and results may
have utility in gas dosimetry as they give a basis for diffusion distance in mass transport
processes, in this case across age groups including the very young.
3-59
-------
4 SUMMARY AND CONCLUSIONS
The new studies dealing with overall gas dosimetry in the airways support many of the
principles and approaches of dosimetry in RfCMethods. Hanna et al. (2001) discussed
advancements within the existing dosimetry modeling framework through development
and application of mass transfer coefficients. This direction is evidenced by the recent
publication of several studies utilizing mass transfer coefficients to calculate dose to
respiratory tract tissues from inhaled vapors (Asgharian et al., 2011; Madasu. 2007;
Overton et al.. 2001): however, still data gaps exist with regards to fractional penetration
of gases and localized deposition, from a predictive and biological standpoint, in the TB
and PU regions. Although the use of a simplified geometric model of the airways limited
the breadth of their conclusions, the tissue metric for the alveolar area arrived at by
Tsujino et al. (2005). g/cm2/min, is similar to that used in the RfC Methods. The
proposals and methods for extension of computational flow evaluation to the lower
airways of Minard et al. (2006) should provide refinement and further resolution to flow
and dose in the lower airways as has been done extensively for the upper airways.
Additional studies need to encompass CFD simulations in the rat and human lower
respiratory tracts to be able to compare gas uptake rates between species, similar to what
has been done for the URT.
The advent of the doubly labeled water (DLW) technique in estimation of physiological
daily inhalation rates (PDIR) have provided resolutions to concerns not only regarding
inhalation patterns of free-living individuals that are ideal for applying to long-term risk
assessments involving inhalation exposures but also addresses lingering concerns of
subjectiveness and bias to which virtually all time-activity-ventilation (TAV) methods are
especially susceptible. Through a systematic consideration of aggregate errors, the
method comparison studies available show that the DLW technique is by far the most
accurate and reliable technique examined which included a number of TAV approaches.
The capabilities and accuracy of DLW can reasonably be inferred over the entire range of
its use which includes the very young (infants) and the very old. DLW-based PDIR
values are currently included in the Child-Specific Exposure Factors Handbook (U.S.
EPA. 2008). and being proposed for inclusion in other key Agency documents, including
the external review draft of the Exposure Factors Handbook (U.S. EPA. 2009a). for all
ages including children.
The studies and information relating directly to dosimetry of the tracheobronchial
(TB) and pulmonary (PU) regions generally support the dosimetric approaches and
assumptions of RfC Methods. Methodological advances and increased resolution of
several in vivo imaging techniques indicate highly homogenous and uniform flows in the
alveolar regions. On the other hand, examination of the tracheobronchial (TB) region
with human models and advanced dynamic fluid flow programs reveal a degree of
4-1
-------
non-uniformity of flow for this region although apparently not to the extent that has been
documented for the upper airway. As discussed in Status I (U.S. EPA. 2009b). these
assumptions and thus, the default dosimetric procedure for the ET region were not
supported as studies consistently demonstrated highly non-uniform airflow and
deposition to airway surfaces, and advance kinetic models clearly demonstrated the
animal/human dose to be > 1.
Marked advances in morphometry of these regions are being achieved with the
development and application of stereology. These techniques, described as the estimation
of higher dimensional information from lower dimensional samples, have and continue to
provide more accurate estimates of measures and vital parameters such as alveoli number
and size characteristics, volumes and surface areas in both humans (e.g.. Ochs et al..
2004) and laboratory animals (Knust et al., 2009). all of which may influence and refine
inhalation dosimetry of gases. Table 4-1 summarizes the various respiratory tract surface
areas and volumes that have become available and that are documented in this report.
The significance of the blood:air partition coefficient (Hb/g) to the advanced PBPK
models have apparently been responsible for the generation of a number of direct and
surrogate approaches for providing these values, both animal and human. The critical and
comprehensive analyses of Payne and Kenny (2002) and Abraham et al. (2005) of human
and animal (rat) Hb/g for a large number of volatile organics from several sources and
approaches made several conclusions. A major indirect conclusion affecting interspecies
dosimetry was that for VOCs there was no difference between rat and human Hb/g. The
other strategy to evaluate the Hb/g for purposes of interspecies dosimetry involved
inspection of published inhalation PBPK models that were configured for interspecies
extrapolation, and therefore had Hb/gs that were validated with simulations compared to
relevant human empirical data. These results give indications the current dosimetry
approach of RfCMethods that uses ratios of animal to human Hb/g as a basis of
dosimetry for the extrarespiratory (ER) region may result in human equivalent
concentrations that are less than estimated by PBPK models.
4-2
-------
Table 4-1 Respiratory Tract Surface Areas and Volumes for Various Species comparing the
1994 RfC Methods and this Report
Surface Area
This Report'
Surface Area RfC Methods9
Volume'
Mouse*9
Total alveolar
airspace 0.082 m2
0.05 m2
138mm3(CV,0.29)
Pulmonary
Ratb'g
NA
0.34m2
16.5cm3, CV
0.17
(PU)
Human0'9
78.4 -81. 6m2; 43.3
m2SD7.7
(Vertical section);
41.6m2SD9.3
(IUR section)
54m2
1,534±521cm3(n =
6)
Mouse*9
3.5cm2
3.5cm2
0.114cm3
Tracheobronchial
Rat
NAe
22.5 cm2
NAe
(TB)
Human
NAe
3200 cm2
NAe
'Values for a C57B6 mouse weighing 20.6g from Knust et al. (2009).
"Values from Hyde et al. (2004).
"Surface area data from Wiebe and Laursen (1995): volume data from Ochs et al. (2004).
dValue for a BALB/c mouse weighing approximately 25 g from Madl et al. (2010).
el\IA=data not available.
'New information provided in this report.
'Default values presented in the RfC Methods (U.S. EPA, 1994)
An overview of the literature available on children's dosimetry closely follows the
recommendations and guidance of the NAS on children's risk (NRC. 1993). These
recommendations include the proposal to use PBPK models to explore and evaluate
potential child susceptibility. A recommendation linked to the development and
utilization of models is the need to generate accurate measurements and parameters to be
used in these models. Accordingly there exist a number of studies examining various
parameters essential to inhalation modeling including physiological daily inhalation rates,
lung tissue and lower airway measures and function. A compelling dataset (orally
administered therapeutics) documents the generally slower clearance rate in children
(Ginsberg et al.. 2002). Flow models are available that examine uptake differences of
gases in the upper airways of both adults and children. Also, several PBPK models that
are configured to specifically consider child versus adult dosimetry have been developed.
Although the actual number of datasets and models relating to gas dosimetry in children
is not yet plentiful, a number of methods and approaches are available. These methods
and approaches indicate child vulnerability related to inhalation dosimetry is
typically in the range of 1 to 2-fold more than adult, but can be more or less. This
range is within that built into RfC Methods using the human interindividual uncertainty
factor (UFH) to accommodate pharmacokinetic and pharmacodynamic variability and for
consideration of potential sensitive population and lifestages including children. It may
also be noted that this finding is very similar to that of the NAS (1993). Consequently,
with regards to gas dosimetry, there appears to be insufficient quantitative evidence
to modify the RfC Methods specifically for children; however, in some cases,
chemical-specific information may warrant consideration of alternative approaches
or adjustments to account for this lifestage. It is anticipated that information will
continue to become available to further inform this issue.
4-3
-------
5 REFERENCES
Abraham. M: Kamlet M: Taft R: Dohertv. R: Weathersbv. P. (1985). Solubility properties in polymers and
biological media. 2. The correlation and prediction of the solubilities of nonelectrolytes in biological tissues
and fluids. J Med Chem 28: 865-870. http://dx.doi.org/10.1021/im00145a004.
Abraham. M and Weathersbv. P. (1994). Hydrogen bonding. 30. Solubility of gases and vapors in biological liquids
and tissues. J Pharm Sci 83: 1450-1456. http://dx.doi.org/10.1002/ips.2600831017.
Abraham. MH: Ibrahim. A: Acree WE. Jr. (2005). Air to blood distribution of volatile organic compounds: A linear
free energy analysis. Chem Res Toxicol 18: 904-911. http://dx.doi.org/10.1021/tx050066d.
Allan. M and Richardson. GM. (1998). Probability density functions describing 24-hour inhalation rates for use in
human health risk assessment. Hum Ecol Risk Assess 4: 379-408.
http://dx.doi.org/10.1080/10807039891284389.
Allan. M: Richardson. GM: Jones-Otazo. H. (2008). Probability density functions describing 24-hour inhalation
rates for use in human health risk assessments: An update and comparison. Hum Ecol Risk Assess 14: 372-
391. http://dx.doi.org/10.1080/10807030801934796.
Allen. G: Shortall B: Gemci T: Corcoran. T: Chigier. N. (2004). Computational simulations of airflow in an in vitro
model of the pediatric upper airways. J Biomech Eng 126: 604-613. http://dx.doi.0rg/10.1115/l.1800554.
Altes. T: Rehm P: Harrell F: Salerno. M: Daniel T: De Lange. E. (2004a). Ventilation imaging of the lung:
Comparison of hyperpolarized helium-3 MR imaging with Xe-133 scintigraphy. AcadRadiol 11: 729-734.
http://dx.doi.0rg/10.1016/i.acra.2004.04.001.
Altes. TA: Mata. J: de Lange. EE: Brookeman. JR: Mugler JP. I. (2004b). Assessment of lung development using
hyperpolarized helium-3 diffusion MR imaging. JMagnReson Imaging 24: 1277-1283.
http://dx.doi.org/10.1002/imri.20723.
Asgharian. B: Price. O: Schroeter. J: Kimbell. J: Jones. L: Singal. M. (2011). Derivation of mass transfer
coefficients for transient uptake and tissue disposition of soluble and reactive vapors in lung airways. Ann
Biomed Eng 39: 1788-1804. http://dx.doi.org/10.1007/sl0439-011-0274-9.
Balashazy. I: Hofmann. W: Heistracher. T. (1999). Computation of local enhancement factors for the quantification
of particle deposition patterns in airway bifurcations. J Aerosol Sci 30: 185-203.
Balinotti. JE: Tiller. CJ: Llapur. CJ: Jones. MH: Kimmel. RN: Coates. CE: Katz. BP: Nguyen. JT: Tepper. RS.
(2009). Growth of the lung parenchyma early in life. Am J Respir Crit Care Med 179: 134-137.
http://dx.doi.org/10.1164/rccm.200808-1224OC.
Benignus. V: Boyes. W: Bushnell. P. (1998). A dosimetric analysis of behavioral effects of acute toluene exposure
in rats and humans. Toxicol Sci 43: 186-195. http://dx.doi.org/10.1006/toxs.1998.2458.
Bolle. I: Eder. G: Takenaka. S: Ganguly. K: Karrasch. S: Zeller. C: Neuner. M: Kreyling. WG: Tsuda. A: Schulz. H.
(2008). Postnatal lung function in the developing rat. J Appl Physiol 104: 1167-1176.
http://dx.doi.org/10.1152/iapplphvsiol.00587.2007.
Brochu. P: Ducre-Robitaille. J-F: Brodeur. J. (2006a). Physiological daily inhalation rates for free-living individuals
aged 1 month to 96 years, using data from doubly labeled water measurements: A proposal for air quality
criteria, standard calculations and health risk assessment. Hum Ecol Risk Assess 12: 675-701.
http://dx.doi.org/10.1080/10807030600801550.
Brochu. P: Ducre-Robitaille. JF: Brodeur. J. (2006b). Physiological daily inhalation rates for free-living individuals
aged 2.6 months to 96 years based on doubly labeled water measurements: Comparison with time-activity-
ventilation and metabolic energy conversion estimates. Hum Ecol Risk Assess 12: 736-761.
http://dx.doi.org/10.1080/10807030600801626.
5-1
-------
Brochu. P: Ducre-Robitaille. J-F: Brodeur. J. (2006c). Physiological daily inhalation rates for free-living pregnant
and lactating adolescents and women aged 11 to 55 years, using data from doubly labeled water
measurements for use in health risk assessment. Hum Ecol Risk Assess 12: 702-735.
http://dx.doi.org/10.1080/10807030600801592.
CDC. (Centers for Disease Control and Prevention). (1995). Third National Health and Nutrition Examination
Survey, 1988-1991. Selected laboratory and mobile examination center data. Atlanta, GA: Centers for
Disease Control and Prevention, National Health and Nutrition Examination Survey.
http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=2825#Download.
Clewell HJ: Gentry. PR: Covington. TR: Sarangapani. R: Teeguarden. JG. (2004). Evaluation of the potential
impact of age- and gender-specific pharmacokinetic differences on tissue dosimetry. Toxicol Sci 79: 381-
383. http://dx.doi.org/10.1093/toxsci/kfhl09.
Condorelli. P and George. SC. (1999). Theoretical gas phase mass transfer coefficients for endogenous gases in the
lungs. AnnBiomedEng27: 326-339.
Corlev. RA: Bartels. MJ: Carney. EW: Weitz. KK: Soelberg. JJ: Gies. RA: Thrall KD. (2005). Development of a
physiologically based pharmacokinetic model for ethylene glycol and its metabolite, glycolic acid, in rats
and humans. Toxicol Sci 85: 476-490. http://dx.doi.org/10.1093/toxsci/kfill9.
Cronin. W: Oswald. E: Shelley. M: Fisher. J: Flemming. C. (1995). A trichloroethylene risk assessment using a
Monte Carlo analysis of parameter uncertainty in conjunction with physiologically-based pharmacokinetic
modeling. Risk Anal 15: 555-565.
Dallas. CE: Chen. XM: Muralidhara. S: Varkonyi. P: Tackett RL: Bruckner. JV. (1995). Physiologically based
pharmacokinetic model useful in prediction of the influence of species, dose, and exposure route on
perchloroethylene pharmacokinetics. J Toxicol Environ Health 44: 301-317.
http://dx.doi.org/10.1006/taap. 1994.1179.
de Jong. PA: Nakano. Y: Lequin. MH: Merkus. PJ: Tiddens. HA: Hogg. JC: Coxsoa HO. (2003). Estimation of
lung growth using computed tomography. Eur Respir J 22: 235-238. http://dx.doi.org/10.11837
09031936.03.00089702.
de Jong. PA: Long. FR: Wong. JC: Merkus. PJ: Tiddens. HA: Hogg. JC: Coxson. HO. (2006). Computed
tomographic estimation of lung dimensions throughout the growth period. Eur Respir J 27: 261 -267.
http://dx.doi.org/10.1183/09031936.06.00070805.
DeJongh. J: Verhaar. H: Hermens. J. (1997). A quantitative property-property relationship (QPPR) approach to
estimate in vitro tissue-blood partition coefficients of organic chemicals in rats and humans. Arch Toxicol
72: 17-25.
Dunnill. MS. (1962). Postnatal growth of the lung. Thorax 17: 329-333.
Firestone. M: Sonawane. B: Barone. S: Salmon. AG: Brown. JP: Hattis. D: Woodruff. T. (2008). Potential new
approaches for children's inhalation risk assessment. J Toxicol Environ Health A 71: 208-217.
http://dx.doi.org/10.1080/15287390701597905.
Fiserova-Bergerova. V and Diaz. ML. (1986). Determination and prediction of tissue-gas partition coefficients. Int
Arch Occup Environ Health 58: 75-87. http://dx.doi.org/10.1007/BF00378543.
Fisher. JW and Allen. BC. (1993). Evaluating the risk of liver cancer in humans exposed to trichloroethylene using
physiological models. Risk Anal 13: 87-95.
Foos. B: Marty. M: Schwartz. J: Bennett. W: Moya. J: Jarabek. AM: Salmon. AG. (2008). Focusing on children's
inhalation dosimetry and health effects for risk assessment: An introduction. J Toxicol Environ Health A
71: 149-165. http://dx.doi.org/10.1080/15287390701597871.
Foos. B and Sonawane. B. (2008). Overview: Workshop on children's inhalation dosimetry and health effects for
risk assessment. J Toxicol Environ Health A 71: 147-148. http://dx.doi.org/10.1080/15287390701597855.
5-2
-------
Garcia. GJ: Schroeter. JD: Segal. RA: Stanek. J: Foureman. GL: Kimbell. JS. (2009). Dosimetry of nasal uptake of
water-soluble and reactive gases: A first study of interhuman variability. Inhal Toxicol 21: 607-618.
http://dx.doi.org/10.1080/08958370802320186.
Gargas. ML: Burgess. RJ: Voisard. DE: Cason. GH: Andersen. ME. (1989). Partition coefficients of low-molecular-
weight volatile chemicals in various liquids and tissues. Toxicol Appl Pharmacol 98: 87-99.
Gargas. ML: Tyler. TR: Sweeney. LM: Corlev. RA: Weitz. KK: Mast. TJ: Paustenbach. DJ: Hays. SM. (2000). A
toxicokinetic study of inhaled ethylene glycol monomethyl ether (2-ME) and validation of a
physiologically based pharmacokinetic model for the pregnant rat and human. Toxicol Appl Pharmacol
165: 53-62. http://dx.doi.org/10.1006/taap.2000.8928.
Geelhaar. A and Weibel ER. (1971). Morphometric estimation of pulmonary diffusion capacity III The effect of
increased oxygen consumption in Japanese Waltzing mice. Respir Physiol Neurobiol 11: 354-366.
http://dx.doi.org/10.1016/0034-5687(71)90009-0.
Gehr. P: Mwangi. DK: Ammann. A: Maloiv. GM: Taylor. CR: Weibel ER. (1981). Design of the mammalian
respiratory system. V. Scaling morphometric pulmonary diffusing capacity to body mass: wild and
domestic mammals. Respir Physiol 44: 61-86.
Gentry. PR: Covington. TR: Andersen. ME: Clewell HJI. (2002). Application of a physiologically based
pharmacokinetic model for isopropanol in the derivation of a reference dose and reference concentration.
Regul Toxicol Pharmacol 36: 51-68.
Ginsberg. G: Hattis. D: Sonawane. B: Russ. A: Banati. P: Kozlak. M: Smolensk!. S: Goble. R. (2002). Evaluation of
child/adult pharmacokinetic differences from a database derived from the therapeutic drug literature.
Toxicol Sci 66: 185-200.
Ginsberg. G: Foos. B: Dzubow. RB: Firestone. M. (2010). Options for incorporating children's inhaled dose into
human health risk assessment. Inhal Toxicol 22: 627-647. http://dx.doi.org/10.3109/08958371003610958.
Ginsberg. GL: Foos. BP: Firestone. MP. (2005). Review and analysis of inhalation dosimetry methods for
application to children's risk assessment. J Toxicol Environ Health A 68: 573-615.
http://dx.doi.org/10.1080/15287390590921793.
Ginsberg. GL: Asgharian. B: Kimbell JS: Ultman. JS: Jarabek. AM. (2008). Modeling approaches for estimating
the dosimetry of inhaled toxicants in children. J Toxicol Environ Health A 71: 166-195.
http://dx.doi.org/10.1080/15287390701597889.
Hanna. LM: Lou. S-R: Su. S: Jarabek. AM. (2001). Mass transport analysis: inhalation RfC methods framework for
interspecies dosimetric adjustment. Inhal Toxicol 13: 437-463. http://dx.doi.org/10.1080/08958370119198.
Harding. EM and Robinson. RJ. (2010). Flow in a terminal alveolar sac model with expanding walls using
computational fluid dynamics. Inhal Toxicol 22: 669-678. http://dx.doi.org/10.3109/08958371003749939.
Hissink. AM: Kriise. J: Kulig. BM: Verwei M: Muijser. H: Salmon. F: Leenheers. LH: Owen. DE: Lammers. JH:
Freidig. AP: McKee. RH. (2007). Model studies for evaluating the neurobehavioral effects of complex
hydrocarbon solvents III. PBPK modeling of white spirit constituents as a tool for integrating animal and
human test data. Neurotoxicology 28: 751-760. http://dx.doi.0rg/10.1016/i.neuro.2007.03.005.
Hyde. D: Tyler. N: Putney. L: Singh. P: Gundersen. H. (2004). Total number and mean size of alveoli in mammalian
lung estimated using fractionator sampling and unbiased estimates of the Euler characteristic of alveolar
openings. AnatRec 277: 216-226. http://dx.doi.0rg/10.1002/ar.a.20012.
Inagi. K. (1992). [Histological study of mucous membranes in the human nasal septum]. Nippon Jibiinkoka Gakkai
Kaiho95: 1174-1189.
Kaneko. T: Wang. P-Y: Sato. A. (1994). Partition coefficients of some acetate esters and alcohols in water, blood,
olive oil, and rat tissues. Occup Environ Med 51: 68-72.
Kauczor. HU: Hanke. A: Van Beek. EJ. (2002). Assessment of lung ventilation by MR imaging: Current status and
future perspectives. EurRadiol 12: 1962-1970. http://dx.doi.org/10.1007/s00330-002-1379-l.
5-3
-------
Kimbell JS: Gross. EA: Richardson. RB: Conolly. RB: Morgan. KT. (1997). Correlation of regional formaldehyde
flux predictions with the distribution of formaldehyde-induced squamous metaplasia in F344 rat nasal
passages. Mutat Res 380: 143-154. http://dx.doi.org/10.1016/S0027-5107(97)00132-2.
Kirman. CR: Gargas. ML: Marsh. GM: Strother. DE: Klaunig. JE: Collins. JJ: Deskin. R. (2005a). Cancer dose--
response assessment for acrylonitrile based upon rodent brain tumor incidence: use of epidemiologic,
mechanistic, and pharmacokinetic support for nonlinearity. Regul Toxicol Pharmacol 43: 85-103.
http://dx.doi.0rg/10.1016/i.vrtph.2005.06.007.
Kirman. CR: Sweeney LM: Corlev. R: Gargas. ML. (2005b). Using physiologically-based pharmacokinetic
modeling to address nonlinear kinetics and changes in rodent physiology and metabolism due to aging and
adaptation in deriving reference values for propylene glycol methyl ether and propylene glycol methyl
ether acetate. Risk Anal 25: 271 - 284. http://dx.doi.0rg/10.llll/i.1539-6924.2005.00588.x.
Kliment V. (1973). Similarity and dimensional analysis, evaluation of aerosol deposition in the lungs of laboratory
animals and man. Folia Morphol (Warsz) 21: 59-64.
Knust J: Ochs. M: Gundersen. H: Nyengaard. J. (2009). Stereological estimates of alveolar number and size and
capillary length and surface area in mice lungs. Anat Rec 292: 113-122. http://dx.doi.org/10.1002/ar.20747.
Koblinger. L and Hofmann. W. (1988). Stochastic morphological model of the rat lung. Anat Rec 221: 533-539.
http://dx.doi.org/10.1002/ar.1092210109.
Kumar. H: Tawhai. M: Hoffman. E: Lin. C. (2009). The effects of geometry on airflow in the acinar region of the
human lung. J Biomech 42: 1635-1642. http://dx.doi.0rg/10.1016/i.jbiomech.2009.04.046.
Lavton. DW. (1993). Metabolically consistent breathing rates for use in dose assessments. Health Phys 64: 23-36.
Lechner. AJ. (1978). The scaling of maximal oxygen consumption and pulmonary dimensions in small mammals.
RespirPhysiol Neurobiol 34: 29-44. htto://dx.doi.org/10.1016/0034-5687(78)90047-6.
Lee. KM: Dill. JA: Chou. BJ: Rovcroft JH. (1998). Physiologically based pharmacokinetic model for chronic
inhalation of 2-butoxyethanol. Toxicol Appl Pharmacol 153: 211-226.
Liao. KH: Tan. YM: Conolly. RB: Borghoff. SJ: Gargas. ML: Andersen. ME: III. CH. (2007). Bayesian estimation
of pharmacokinetic and pharmacodynamic parameters in a mode-of-action-based cancer risk assessment for
chloroform. Risk Anal 27: 1535-1551. http://dx.doi.0rg/10.llll/i.1539-6924.2007.00987.x.
Longest. PW and Kleinstreuer. C. (2005). Computational models for simulating multicomponent aerosol
evaporation in the upper respiratory airways. Aerosol Sci Technol 39: 124-138.
http://dx.doi.org/10.1080/027868290908786.
Lu. Y: Rieth. S: Lohitnaw. M: Dennison. J: El-Masri. H: Barton. HA: Bruckner. J: Yang. RS. (2008). Application
of PBPK modeling in support of the derivation of toxicity reference values for 1,1,1-trichloroethane. Regul
Toxicol Pharmacol 50: 249-260. http://dx.doi.0rg/10.1016/i.vrtph.2007.12.001.
Madasu. S. (2007). Gas uptake in a three-generation model geometry during steady expiration: Comparison of
axisymmetric and three-dimensional models. Inhal Toxicol 19: 199-210.
http://dx.doi.org/10.1080/08958370601067855.
Madasu. S: Borhan. A: Ultman. J. (2007). Gas uptake in a three-generation model geometry with a flat inlet velocity
during steady inspiration: Comparison of axisymmetric and three-dimensional models. Inhal Toxicol 19:
495-503. http://dx.doi.org/10.1080/0895837070127104.
Madasu. S: Ultman. JS: Borhan. A. (2008). Comparison of axisymmetric and three-dimensional models for gas
uptake in a single bifurcation during steady expiration. J Biomech Eng 130: 011013.
http://dx.doi.0rg/10.1115/l.2838041.
Madl P: Hofmann. W: Oldham. MJ: Asgharian. B. (2010). Stochastic morphometric model of the BALB/c mouse
lung. Anat Rec 293: 1766-1775. http://dx.doi.org/10.1002/ar.21208.
Martin. L. (1987). PCO2 and Alveolar Ventilation
5-4
-------
Pulmonary Physiology in Clinical Practice: The Essentials for Patient Care and Evaluation. St Louis, MO: Mosby-
Year Book.
Menache. MG: Hofmann. W: Ashgarian. B: Miller. FJ. (2008). Airway geometry models of children's lungs for use
indosimetry modeling. Inhal Toxicol 20: 101-126. http://dx.doi.org/10.1080/08958370701821433.
Mercer. RR: Russell ML: Crapo. JD. (1994a). Alveolar septal structure in different species. J Appl Physiol 77:
1060-1066.
Mercer. RR: Russell ML: Roggli VL: Crapo. JD. (1994b). Cell number and distribution in human and rat airways.
Am J Respir Cell Mol Biol 10: 613-624.
Meulenberg. C and Vijverberg. H. (2000). Empirical relations predicting human and rat tissue: Air partition
coefficients of volatile organic compounds. Toxicol Appl Pharmacol 165: 206-216.
http://dx.doi.org/10.1006/taap.2000.8929.
Minard. KR: Einstein. PR: Jacob. RE: Kabilan. S: Kuprat AP: Timchalk. CA: Trease. LL: Corlev. RA. (2006).
Application of magnetic resonance (MR) imaging for the development and validation of computational
fluid dynamic (CFD) models of the rat respiratory system. Inhal Toxicol 18: 787-794.
http://dx.doi.org/10.1080/08958370600748729.
Morris. JB and Hubbs. AF. (2009). Inhalation dosimetry of diacetyl and butyric acid, two components of butter
flavoring vapors. Toxicol Sci 108: 173-183. http://dx.doi.org/10.1093/toxsci/kfn222.
NRC. (National Research Council). (1993). Pesticides in the diets of infants and children. Washington, DC: National
Academy Press.
Ochs. M: Nyengaard. JR: Jung. A: Knudsen. L: Voigt. M: Wahlers. T: Richter. J: Gundersen. HJ. (2004). The
number of alveoli in the human lung. Am J Respir Crit Care Med 169: 1-10.
OEHHA. (California Office of Environmental Health Hazard Assessment). (2008). Air toxics hot spots program
technical support document for the derivation of noncancer reference exposure levels. Oakland, CA:
Office of Environmental Health Hazard Assessment; California Environmental Protection Agency.
Ogiu. N: Nakamura. Y: Ijiri. I: Hiraiwa. K: Ogiu. T. (1997). A statistical analysis of the internal organ weights of
normal Japanese people. Health Phys 72: 368-383.
Overton. JH: Kimbell. JS: Miller. FJ. (2001). Dosimetry modeling of inhaled formaldehyde: The human respiratory
tract. Toxicol Sci 64: 122-134.
Padaki. A: Urtman. JS: Borhan. A. (2009). Ozone uptake during inspiratory flow in a model of the larynx, trachea
and primary bronchial bifurcation. Chem Eng Sci 64: 4640-4648.
http://dx.doi.0rg/10.1016/i.ces.2009.05.017.
Pasting. GM: Sultatos. LG: Flynn. EJ. (1996). Development and application of a physiologically based
pharmacokinetic model for ethanol in the mouse. Alcohol Alcohol 31: 365-374.
Pasting. GM: Asgharian. B: Roberts. K: Medinsky. MA: Bond. JA. (1997). A comparison of physiologically based
pharmacokinetic model predictions and experimental data for inhaled ethanol in male and female B6C3F1
mice, F344 rats, and humans. Toxicol Appl Pharmacol 145: 147-157.
Paterson. S and Mackav. D. (1989). Correlation of tissue, blood, and air partition of volatile organic chemicals.
Occup Environ Med 46: 321-328.
Payne. MP and Kenny. LC. (2002). Comparison of models for the estimation of biological partition coefficients. J
Toxicol Environ Health A 65: 897-931. http://dx.doi.org/10.1080/00984100290071171.
Pelekis. M: Gephart. LA: Lerman. SE. (2001). Physiological-model-based derivation of the adult and child
pharmacokinetic intraspecies uncertainty factors for volatile organic compounds. Regul Toxicol Pharmacol
33: 12-20. http://dx.doi.org/10.1006/rtph.2000.1436.
Phalen. RF and Oldham. MJ. (1983). Tracheobronchial airway structure as revealed by casting techniques. Am Rev
Respir Dis 128: S1-S4.
5-5
-------
Phalen. RF: Oldham. MJ: Nel. AE. (2006). Tracheobronchial particle dose considerations for in vitro toxicology
studies. Toxicol Sci 92: 126-132. http://dx.doi.org/10.1093/toxsci/kfi 182.
Pinkerton. KE and Joad. JP. (2000). The mammalian respiratory system and critical windows of exposure for
children's health. Environ Health Perspect 108: 457-462.
Poulin. P and Krishnan. K. (1995). A biologically-based algorithm for predicting human tissue: Blood partition
coefficients of organic chemicals. HumExp Toxicol 14: 273-280.
http://dx.doi.org/10.1177/096032719501400307.
Ranz. WE and Marshall. WR. (1952a). Evaporation from drops: Part 1. Chemical Engineering Progress 48: 141-146.
Ranz. WE and Marshall. WR. (1952b). Evaporation from drops: Part 2. Chemical Engineering Progress 48: 173-
180.
Rao. L: Tiller. C: Coates. C: Kimmel R: Applegate. KE: Granroth-Cook. J: Denski. C: Nguyen. J: Yu. Z: Hoffman.
E: Tepper. RS. (2010). Lung growth in infants and toddlers assessed by multi-slice computed tomography.
AcadRadiol 17: 1128-1135. http://dx.doi.0rg/10.1016/i.acra.2010.04.012.
Reitz. RH: McDougal. JN: Himmelstein. MW: Nolan. RJ: Schumann. AM. (1988). Physiologically based
pharmacokinetic modeling with methylchloroform: Implications for interspecies, high dose/low dose, and
dose route extrapolations. Toxicol Appl Pharmacol 95: 185-199. http://dx.doi.org/10.1016/0041-
008X(88)90155-X.
Sarangapani. R: Gentry. PR: Covington. TR: Teeguarden. JG: Clewell HJ. I. (2003). Evaluation of the potential
impact of age- and gender-specific lung morphology and ventilation rate on the dosimetry of vapors. Inhal
Toxicol 15: 987-1016.
Schreider. JP and Hutchens. JO. (1980). Morphology of the guinea pig respiratory tract. Anat Rec 196: 313-321.
http://dx.doi.org/10.1002/ar.1091960307.
Schroeter. JD: Garcia. GJ: Kimbell JS. (2010). A computational fluid dynamics approach to assess interhuman
variability in hydrogen sulfide nasal dosimetry. Inhal Toxicol 22: 277-286.
http://dx.doi.org/10.3109/08958370903278077.
Steward. A: Allott PR: Cowles. AL: Mapleson. WW. (1973). Solubility coefficients for inhaled anaesthetics for
water, oil and biological media. Br J Anaesth 45: 282-293. http://dx.doi.Org/10.1093/bja/45.3.282.
Sweeney. LM: Kirman. CR: Gannon. SA: Thrall KD: Gargas. ML: Kinzell. JH. (2009). Development of a
physiologically based pharmacokinetic (PBPK) model for methyl iodide in rats, rabbits, and humans. Inhal
Toxicol 21: 552-582. http://dx.doi.org/10.1080/08958370802601569.
Sznitman. J: Heimsch. F: Heimsch. T: Rusch. D: Rosgen. T. (2007). Three-dimensional convective alveolar flow
induced by rhythmic breathing motion of the pulmonary acinus. J Biomech Eng 129: 658-665.
http://dx.doi.0rg/10.1115/l.2768109.
Tardif. R: Charest-Tardif. G: Brodeur. J: Krishnan. K. (1997). Physiologically based pharmacokinetic modeling of a
ternary mixture of alkyl benzenes in rats and humans. Toxicol Appl Pharmacol 144: 120-134.
Taylor. AB: Borhan. A: Ultman. JS. (2007). Three-dimensional simulations of reactive gas uptake in single airway
bifurcations. AnnBiomed Eng 35: 235-249. http://dx.doi.org/10.1007/sl0439-006-9195-4.
Teeguarden. JG: Deisinger. PJ: Poet TS: English. JC: Faber. WD: Barton. HA: Corlev. RA: Clewell HJI. (2005).
Derivation of a human equivalent concentration for n-butanol using a physiologically based
pharmacokinetic model for n-butyl acetate and metabolites n-butanol and n-butyric acid. Toxicol Sci 85:
429-446. http://dx.doi.org/10.1093/toxsci/kfil03.
Tennev. SM and Remmers. JE. (1963). Comparative quantitative morphology of the mammalian lung: diffusing
area. Nature 197: 54-56.
Thurlbeck. WM. (1967). The internal surface area of nonemphysematous lungs. Am Rev RespirDis 95: 765-773.
Thurlbeck. WM. (1982). Postnatal human lung growth. Thorax 37: 564-571.
5-6
-------
Tian. G and Longest PW. (2010a). Application of a new dosimetry program TAOCS to assess transient vapour
absorption in the upper airways. Inhal Toxicol 22: 1047-1063.
http://dx.doi.org/10.3109/08958378.2010.521783.
Tian. G and Longest PW. (2010b). Development of a CFD boundary condition to model transient vapor absorption
in the respiratory airways. J BiomechEng 132: 051003. http://dx.doi.0rg/10.1115/l.4001045.
Tian. G and Longest PW. (2010c). Transient absorption of inhaled vapors into a multilayer mucus-tissue-blood
system. 38: 517-536. http://dx.doi.org/10.1007/sl0439-009-9808-9.
Tsuda. A: Rogers. RA: Hydron. PE: Butler. JP. (2002). Chaotic mixing deep in the lung. PNAS 99: 10173-10178.
http://dx.doi.org/10.1073/pnas.102318299.
Tsuiino. I: Kawakami. Y: Kaneko. A. (2005). Comparative simulation of gas transport in airway models of rat, dog,
and human. Inhal Toxicol 17: 475-485. http://dx.doi.org/10.1080/08958370590964476.
U.S. EPA. (U.S. Environmental Protection Agency). (1994). Methods for derivation of inhalation reference
concentrations and application of inhalation dosimetry. (EPA/600/8-90/066F). Research Triangle Park, NC:
U.S. Environmental Protection Agency, Office of Research and Development, Office of Health and
Environmental Assessment, Environmental Criteria and Assessment Office.
http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=71993.
U.S. EPA. (U.S. Environmental Protection Agency). (1997). Exposure factors handbook (final report). (EPA/600/P-
95/002Fa-c). Washington, DC: U.S. Environmental Protection Agency, Office of Research and
Development, National Center for Environmental Assessment.
http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=12464.
U.S. EPA. (U.S. Environmental Protection Agency). (2002). A review of the reference dose and reference
concentration processes. (EPA/630/P-02/0002F). Washington, DC.
http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=51717.
U.S. EPA. (U.S. Environmental Protection Agency). (2006). Peer review handbook (3rd edition). (EPA/100/B-
06/002). Washington, DC: U.S. Environmental Protection Agency, Science Policy Council.
http://www.epa.gov/peerreview/pdfs/peer review handbook 2006.pdf.
U.S. EPA. (U.S. Environmental Protection Agency). (2008). Child-specific exposure factors handbook.
(EPA/600/R-06/096F). Washington, DC: U.S. Environmental Protection Agency, National Center for
Environmental Assessment. http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=199243.
U.S. EPA. (U.S. Environmental Protection Agency). (2009a). Exposure factors handbook: 2009 update [external
review draft]. (EPA/600/R-09/052A). Washington, DC: U.S. Environmental Protection Agency, National
Center for Environmental Assessment, http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=209866.
U.S. EPA. (U.S. Environmental Protection Agency). (2009b). Status report: Advances in inhalation dosimetry of
gases and vapors with portal of entry effects in the upper respiratory tract. (EPA/600/R-09/072). Research
Triangle Park, NC: U.S. Environmental Protection Agency, National Center for Environmental
Assessment. http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=212131.
Valcke. M and Krishnan. K. (2011). Evaluation of the impact of the exposure route on the human kinetic adjustment
factor. Regul Toxicol Pharmacol 59: 258-269. http://dx.doi.0rg/10.1016/i.yrtph.2010.10.008.
Wiebe. BM and Laursen. H. (1995). Human lung volume, alveolar surface area, and capillary length. Microsc Res
Tech 32: 255-262. http://dx.doi.org/10.1002/iemt.1070320308.
Willems. B: Melnick. R: Kohn. M: Portier. C. (2001). A physiologically based pharmacokinetic model for inhalation
and intravenous administration of naphthalene in rats and mice. Toxicol Appl Pharmacol 176: 81-91.
http://dx.doi.org/10.1006/taap.2001.9269.
Yu. CP and Xu. GB. (1987). Predictive models for deposition of inhaled diesel exhaust particles in humans and
laboratory species. (Research Report No. 10). Cambridge, MA: Health Effects Institute.
http://pubs.healtheffects.org/view.php?id= 138.
5-7
-------
Zeltner. TB: Caduff. JH: Gehr. P: Pfenninger. J: Burtj PH. (1987). The postnatal development and growth of the
human lung. I. Morphometry. Respir Physiol 67: 247-267. http://dx.doi.org/10.1016/0034-5687(87)90057-
o.
Zeman. KL and Bennett WD. (2006). Growth of the small airways and alveoli from childhood to the adult lung
measured by aerosol-derived airway morphometry. J Appl Physiol 100: 965-971.
http://dx.doi.org/10.1152/iapplphvsiol.00409.2005.
Zhang. Z: Kleinstreuer. C: Kim. CS. (2006). Transport and uptake of MTBE and ethanol vapors in a human upper
airway model. Inhal Toxicol 18: 169-184. http://dx.doi.org/10.1080/08958370500434172.
Zhang. Z and Kleinstreuer. C. (2011). Computational analysis of airflow and nanoparticle deposition in a combined
nasal-oral-tracheobronchial airway model. J Aerosol Sci 42: 174-194.
http://dx.doi.0rg/10.1016/i.iaerosci.2011.01.001.
5-8
-------
APPENDIX A. SUMMARY AND DISPOSITION OF
INDEPENDENT EXTERNAL PEER REVIEW
COMMENTS
The "Status Report: Advances in Inhalation Dosimetry for Gases with Effects on the
Lower Respiratory Tract and the Body" (Status IT) has undergone a formal external peer
letter review performed by scientists in accordance with EPA guidance on peer review
(U.S. EPA. 2006). The reviewers were tasked with providing written answers to charge
questions on both general and specific scientific aspects of the report. A summary of
significant comments made by the external reviewers to these charge questions and
EPA's responses to these comments arranged by charge question follow. Editorial
comments were considered and incorporated directly into the document as appropriate.
A.I External Peer Reviewer Comments -Comments and Response
to Charge:
A.I.I Charge Question 1
The primary focus of this report relates to the pharmacokinetic component of interspecies
gas dosimetry for portal of entry effects in the lower respiratory tract. Issues related to
pharmacodynamics, including variability in response, are specifically excluded from this
report. Is the scope and primary focus of this report clear?
Comments:
All three reviewers commented that the scope and primary focus of the report are clear.
However, one reviewer thought the organization was somewhat choppy and provided
editorial suggestions to help improve the organization and flow of the text. One reviewer
commented further that is was made very clear that this report focuses on
pharmacokinetics and does not include any pharmacodynamic issues.
Response:
The majority of the editorial changes suggested by the reviewer were incorporated in the
revised document.
A-l
-------
A.I.2 Charge Question 2
Have the principal studies examining interspecies gas dosimetry for effects in the TB,
PU, and ER regions that have been reported since the issuance of the 1994 RfC Methods
been identified in this report? Please identify and provide a rationale for any other key
studies that should be considered for inclusion.
Comments:
One reviewer provided suggested text for four additional references (Tian and Longest.
2010a. b, c; Zhang et al., 2006) pertaining to the assumption of steady-state mass
transport fluxes across mucus and tissue barriers. One reviewer provided three additional
references (Zhang and Kleinstreuer. 2011; Longest and Kleinstreuer. 2005; Ranz and
Marshall. 1952a. b) that provide more information on mass transfer coefficients and the
Sherwood number. One reviewer suggested including a reference regarding air-phase
mass transfer coefficients (Condorelli and George. 1999) and a few papers related to the
uptake of ozone (Taylor et al.. 2007).
Response:
The text and references suggested by the reviewers were evaluated, and EPA agrees that
the text summarizing these additional references should be included in this report. The
work by Tian and Longest (Tian and Longest 2010a. b, c) and Zhang, Kleinstreuer, and
Kim (Zhang et al.. 2006) was included in Section 3.3.1. Additional citations (Zhang and
Kleinstreuer. 2011; Condorelli and George. 1999) were provided in Section 2.3 on
gas-phase mass transfer coefficients, and information on the use of the Sherwood number
to estimate gas-phase mass transfer coefficients (Asgharian et al., 2011; Zhang and
Kleinstreuer. 2011; Longest and Kleinstreuer. 2005; Condorelli and George. 1999; Ranz
and Marshall 1952a. b) was added to Section 3.1. The suggested references (Padaki et
al.. 2009; Taylor et al.. 2007) regarding ozone modeling and uptake provided useful
information on flow and deposition and thus were added to Section 3.3.1.
A.I.3 Charge Question 3
The state-of-the-science pertaining to the focus of this report is primarily presented in
Section 3. Is the description of those studies in this report concerned with gas dosimetry
appropriate and accurate? Are the analyses and evaluations of the scientific evidence
supported by the studies cited?
Comments:
One reviewer questioned whether the current DBFs are inside the range of the adjustment
factors currently used, and whether these adjustment factors account for local deposition.
This reviewer also suggested two references on particle DBFs that discuss the impact of
A-2
-------
surface area size considered on the calculation of the DBF (Sweeney et al., 2009; Phalen
et al.. 2006; Balashazy et al.. 1999). Two reviewers agreed that the studies included in the
report adequately describe what is known about alveolar gas transport; however, one
reviewer noted that future studies are needed before the occurrence of significant
localized deposition can be ruled out. Two reviewers suggested correcting the description
of the gas-phase mass transport coefficient (kg) to indicate that it does not depend on the
reactivity of the gas. One reviewer suggested including more general descriptions of the
mathematical models that may be used for inhalation dosimetry (e.g., PBPK, identical
path, and CFD) and their advantages and limitations.
Response:
No changes were made to the document in response to one reviewers comment with
regards to whether the range of adjustment factors accounts for the possibility of the
various DBFs. It was inferred that this comment pertained more to the uncertainty factor
application to derivation of RfCs, which is not a focus of this report. The Balashazy et al.
(1999) and Phalen et al. (2006) papers were evaluated for inclusion in this report;
however, these papers described particle deposition that was not directly related to
inhalation gas dosimetry in the lower respiratory tract and lacked information needed for
comparative inter- and intra-species extrapolations, thus these papers were not included.
Text was added to the end of Section 3.4.1. to clarify that additional studies are needed to
rule out the possibility of localized deposition ("hotspots"). The description of the
gas-phase mass transport coefficient was updated as suggested by the reviewers. General
descriptions of the mathematical models commonly used in inhalation dosimetry were
included in the glossary for this report; however, significant discussion of the advantages
and limitations of these models for application in inhalation dosimetry was not included
as this is outside the scope of this current work.
A.I.4 Charge Question 4
The state-of-the-science pertaining to children's inhalation dosimetry is presented in
Section 3.6. Is the description of those studies in this report, as they pertain to inhalation
gas dosimetry, appropriate and accurate? Are the analyses and evaluations of the scientific
evidence supported by the studies cited? Are there additional evidence-based studies and
information specific to children's inhalation dosimetry that should be considered for
inclusion that contribute to the science and understanding of inhalation gas dosimetry in
children?
Comments:
One reviewer commented that the available studies are included and appropriately
presented. Another reviewer commented that description of the studies pertaining to
A-3
-------
children's dosimetry is appropriate and is not aware of any additional studies that could
be useful for the RfC methodology in children. This reviewer also commented that the
two most extensive studies that directly applicable to RfC derivation appear to be the
Ginsberg and Sarangapani studies, and that these studies were appropriately evaluated.
The conclusion was important since this variability is included in the uncertainty factor
for interindividual variability. The third reviewer thought that this section clearly
indicates new information that can impact on future refinements of dosimetry modeling
and default, but suggested that the subsections could be reordered.
Response:
Subsection titles were revised to more clearly indicate the content in each section. Some
text was moved as suggested to improve the flow of the document; however, the general
structure of the document was retained such that the overall document was arranged by
anatomy (e.g., TB, PU, or ET region) and within each section the specific advances and
modeling/quantitation approaches were described. A separate section pertaining to
inhalation dosimetry in children was retained.
A.I.5 Charge Question 5
This report provides new information on the pharmacokinetic component of interspecies
gas dosimetry for effects in the TB, PU, and ER regions. Is this report successful in
presenting and evaluating the state-of-the-science relating to this focus?
Comments:
All three reviewers were in agreement that this report adequately presented and evaluated
the state-of-the-science related to this area. One reviewer further commented that this
report nicely summarizes some new key studies related to inhalation gas dosimetry, but
these studies do not seem to have much of a direct application to the refinement of the
RfC Methods.
Response:
No response necessary.
A-4
-------
A.I.6 Charge Question 6
Please comment on the effectiveness of the Summary and Commentary in describing
advances in the state of the science since publication of the RfC Methods document (U.S.
EPA. 1994) (http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=71993. Please
identify any additions, deletions or changes that would improve the effectiveness of the
draft review document in summarizing the state of the science.
Comments:
One reviewer commented that the report does an excellent job in describing the RfC
Methods related to the purpose of the report. However, the Summary should add points
concerning future studies that could address some the data gaps present. This reviewer
provided some additional studies that could be referenced in this regard. One reviewer
commented that this section provides the necessary concluding remarks of the studies
presented to reach the conclusion presented in this report. One reviewer thought that this
section was generally well-written and was in agreement with the conclusions reached.
Response:
As the reviewer suggested, additional text was added to the Summary and Conclusions
Section (Section 4) regarding studies that could be conducted to help fill known data gaps
in inhalation dosimetry methodologies.
A-5
-------
APPENDIX B. DEFINITIONS FOR VENTILATORY
VOLUMES
B.I Ventilation
Ventilation is a general term for the movement of air into and out of the lungs; without a
preceding adjective, such as alveolar or minute, the term does not have any more specific
meaning. The symbol for ventilation is ; V stands for volume and the dot for "per unit
time".
B.2 Minute Ventilation
Minute or total ventilation the amount of air moved in or out of the lungs per minute.
Quantitatively, the amount of air breathed in per minute (Vi) is slightly greater than the
amount expired per minute (VE). Clinically this difference is not important, and by
convention minute ventilation is always measured on an expired sample and symbolized
VE. It is useful to remember that VE is the breathing frequency (f) per minute times the
tidal volume (VT, volume of tidal breath):
VE = f x VT
Equation B-l
VE is also the sum of two other ventilations, alveolar ventilation and dead space
ventilation.
B.3 Alveolar Ventilation
Alveolar ventilation (VA) is the volume of air breathed in per minute that (1) reaches the
alveoli and (2) takes part in gas exchange. Alveolar ventilation is often misunderstood as
representing only the volume of air that reaches the alveoli. Physiologically, VA is the
volume of alveolar air/minute that takes part in gas exchange (transfer of oxygen and
carbon dioxide) with the pulmonary capillaries. Air that reaches the alveoli, but for one
reason or other does not take part in gas exchange, is not considered part of VA (for
example, air that goes to an unperfused alveolus). Such alveolar regions lacking gas
exchange constitute alveolar dead space. Clinically, the terms hyperventilation and
hypoventilation apply to alveolar ventilation only.
B-l
-------
B.4 Dead Space Ventilation
Dead space ventilation is that part of minute ventilation that does not take part in gas
exchange; it is also referred to as "wasted ventilation." Dead space ventilation (VD)
includes (1) air that enters only conducting airways (referred to as anatomic dead
space) and (2) air that reaches alveoli but does not exchange carbon dioxide or oxygen
with the capillary blood. The combined volume of these two areas is often referred to
as physiologic dead space.
Based on these definitions:
VB = VAx VD
Equation B-2
or
VA = VE ~ VD
Equation B-3
In actual practice, VE is relatively easy to measure with a spirometer (or any device that
can measure tidal volume). However, neither VA nor VD is measured in the clinical
setting; they are difficult to measure, and knowing their absolute value is not considered
all that helpful.
Source: Martin (1987)
B-2
-------
APPENDIX C. LITERATURE SEARCH STRATEGY
Literature searches were conducted on sources published from 1985 through March 30
2011, for studies relevant to inhalation dosimetry utilizing the search terms listed below.
Searches were conducted using U.S. EPA's Health and Environmental Research Online
(HERO) evergreen database of scientific literature. HERO searches the following
databases: AGRICOLA; American Chemical Society; BioOne; Cochrane Library; DOE:
Energy Information Administration, Information Bridge, and Energy Citations Database;
EBSCO: Academic Search Complete; GeoRef Preview; GPO: Government Printing
Office; Informaworld; IngentaConnect; J-STAGE: Japan Science & Technology; JSTOR:
Mathematics & Statistics and Life Sciences; NSCEP/NEPIS (EPA publications available
through the National Service Center for Environmental Publications (NSCEP) and
National Environmental Publications Internet Site (NEPIS) database); PubMed:
MEDLINE and CANCERLIT databases; SAGE; Science Direct; Scirus; Scitopia;
SpringerLink; TOXNET (Toxicology Data Network): ANEUPL, CCRIS, ChemlDplus,
CIS, CRISP, DART, EMIC, EPIDEM, ETICBACK, FEDRIP, GENE-TOX, HAPAB,
KEEP, HMTC, HSDB, IRIS, ITER, LactMed, Multi-Database Search, NIOSH, NTIS,
PESTAB, PPBIB, RISKLINE, TRI, and TSCATS; Virtual Health Library; Web of
Science (searches Current Content database among others); World Health Organization;
and Worldwide Science. The following databases outside of HERO were searched for
risk assessment values: ACGIH, ATSDR, CalEPA, U.S. EPA IRIS, U.S. EPA HEAST,
U.S. EPA KEEP, U.S. EPA OW, U.S. EPA TSCATS/TSCATS2, NIOSH, NTP, OSHA,
and RTECS. Additionally cited reference searches were conducted utilizing the
references listed below.
Search Terms (Proposed)
Blood partitioning AND gas AND/OR equilibrium
Comparative dosimetry
Dosimetry AND children (AND gas OR particle OR aerosol)
Inhalation AND model AND pulmonary
Inhalation AND model AND pulmonary AND gas
Inhalation AND model AND tracheobronchial
Inhalation AND dosimetry AND tracheobronchial AND gas
Inhalation AND model AND tracheobronchial AND gas
Inhalation AND dosimetry AND pulmonary AND gas
Inhalation AND model AND systemic AND gas
Inhalation AND modeling AND systemic AND gas
Inhalation AND PBPK
Inhalation AND CFD AND pulmonary AND gas
C-l
-------
Inhalation AND CFD AND pulmonary AND gas
Inhalation AND CFD AND tracheobronchial AND gas
Inhalation rates AND children
Neonatal AND dosimetry
Lung AND morphometry
Lung AND lesions AND regional AND pattern
Trachea AND morphometry
Trachea AND lesions AND pattern
Pulmonary AND lesions AND pattern
Respiratory tract morphometry
(Some) Citation Search References
Gargas ML, Burgess RJ, Voisard DE, Cason GH, Andersen ME. (1989). Partition
coefficients of low-molecular-weight volatile chemicals in various liquids and tissues.
Toxicol Appl Pharmacol. Mar 15; 98(l):87-99
Kimbell JS, Gross EA, Richardson RB, Conolly RB, Morgan KT. (1997). Correlation of
regional formaldehyde flux predictions with the distribution of formaldehyde-induced
squamous metaplasia in F344 rat nasal passages. Mutat Res. Oct 3 l;380(l-2): 143-54.
Phalen, R. F.; Oldham, M. J. (1983). Tracheobronchial airway structure as revealed by
casting techniques. Am. Rev. Respir. Dis. 128: S1-S4
C-2
------- |