4th International Conference On Building Energy, Environment
Quantitative risk assessment of transient inhalation exposure
using PBPK-CFD hybrid model with computer simulated person
S. Yoo1 and K. Ito11Faculty of Engineering Sciences, Kyushu
University, Kasuga, Fukuoka 816-8580, Japan
SUMMARY This study focused on the comprehensive estimation of
indoor air quality targeting micro-climate around human body and
respiratory area by applying PBPK (physiologically-based
pharmacokinetic)-CFD (computational fluid dynamics) hybrid analysis
with unsteady breathing condition. The overarching objective of
this study is to predict the contaminant concentration
distribution, doses of inhaled contaminant in the respiratory
tissue, which could be applied for establishing a threshold
concentration or a reference concentration (RfC). In this study,
formaldehyde was assumed as a target contaminant, and flow field,
sensible/latent heat and contaminant transfer analysis were
conducted by using CFD simulation coupled with CSP (computer
simulated person). Finally, inhaled formaldehyde concentration,
maximum/peak concentration and adsorption flux at airway wall
surface were estimated by PBPK-CFD analysis.
INTRODUCTION Recently, the indoor environmental quality (IEQ)
greatly influences on the public health. Especially, indoor air
quality (IAQ) problem, i.e. sick house syndrome, directly affects
human health. To create a healthy, comfortable and productive
indoor environment with minimizing negative health impact, an
accurate prediction method of indoor air quality problem and
respiratory exposure risks is required at the architectural design
stage.
From the viewpoint of risk assessment of respiratory exposure,
it is necessary to estimate a quantitative effect of inhaled
contaminants on the human body. In this case, physiologically-based
pharmacokinetic (PBPK) model could be used for the prediction of
drug/contaminant transport phenomenon inside the human body.
Pharmacokinetic models and their application in inhaled exposure
based on numerical analysis (in silico) was first suggested on the
basis of in vivo and in vitro experiments for analyzing the effect
of medicinal substances on the human body and then the PBPK model
was developed to estimate the human health risk caused by
respiratory exposure. Subsequently, a PBPK-CFD hybrid model that
can estimate the inhalation of aldehydes was suggested by Corley et
al.
A number of researches have reported the development and
application of the human model or respiratory tract model coupled
with CFD simulation for predicting contaminant distributions and
their impact on the human body. However, a comprehensive analytical
method that took into consideration indoor spaces and the
respiratory tract, and that fully integrated the physiological
model into the human body model with the respiratory tract has not
been established.
With this background, our previous study aimed to develop a
comprehensive prediction method of indoor environmental
quality based on the CFD analysis using computer simulated
person (CSP) with numerical respiratory tract.
The purpose of this study is to develop a hybrid model for
assessment of health risk occurred by respiratory exposure. This
hybrid model is established by integrating PBPK model with CFD
analysis which was coupled with CSP. By applying PBPK model,
contaminant adsorption flux at airway wall surface and
reaction/diffusion phenomenon inside the tissue could be
estimated.
Figure 1. Outline of the computer simulated person
Figure 2. Outline of the PBPK-CFD hybrid model
(a) Model room (b) Grid designFigure 3. Outline and grid design
of the analytical domain
ISBN: 978-0-646-98213-7 COBEE2018-Paper146 page 417
4th International Conference On Building Energy, Environment
METHODS Previously, we have developed computer simulated person
by integrating 2 models: a virtual manikin (numerical human body
model), a virtual airway (numerical respiratory tract model), for
assessment of indoor environmental quality. Figure 1 shows the
outline of the CSP. In order to apply CSP into the estimation of
indoor thermal environmental quality, thermoregulation model
(2-node model) was also adopted into CSP.
Furthermore, a PBPK–CFD hybrid model shown in Figure 2 was
adopted into virtual airway which was integrated into CSP. PBPK
model is widely used for assessment of health risk of human body
caused by respiratory exposure. Here, two-compartment type PBPK
model, epithelium+mucus and sub-epithelium compartments, was
applied into virtual respiratory tract. Finally, contaminant
concentration distribution and airway tissue dosimetry inside human
airway were estimated. Contaminant sorption/ transfer phenomenon
from the airway lumen to a human body was also predicted
quantitatively.
Figure 3 shows the outline and grid design of the analytical
domain of this study. The analytical domain including CSP and a
simple model room was designed to enable comprehensive analysis
targeting continuous area from indoor space to respiratory area via
nasal/oral cavity. Inflow and outflow boundary of the target domain
were set to reproduce displacement ventilation method which is
widely used in indoor spaces. In this study, formaldehyde which is
representative contaminant in indoor spaces was used as a target
contaminant, and it was generated from floor material. Contaminant
generation rate was set in accordance with the perfect mixing
concentration which was intended (100.0μg/m3 in this study).
(a) (b) (c) Figure 4. Flow field, temperature, and formaldehyde
concentration distribution around the CSP
(a) (b) Figure 5. Flow field and formaldehyde concentration
distritrbution in human airway(at the moment of peak inhalation in
the breathing cycle)
Figure 6. Time-series of formaldehyde concentration at airway
wall surface(0~3sec, [μg/m3])
Figure 7. Formaldehyde (HCHO) concentration profile inside the
airway tissue(peak concentration at the representative point in
nasal cavity)
Table 1. Summary of the numerical and boundary conditions in the
PBPK-CFD hybrid analysis
Target contaminant Formaldehyde (HCHO)
Contaminant condition
Fixed flux at the floor (Perfectly mixed concentration Cout =
100.0μg/m3)
Algorithm SIMPLE(Unsteady calculation)
Scheme Convection, Scalar transport equation : Second order
upwind
Boundary condition
(PBPK side)
Sub-epithelial surface : Gradient zero Side wall of the air,
epithelium and sub-epithelium : free-slip
Diffusion coefficient of
Formaldehyde
Da= 0.15 10-4 [m2/s] (Formaldehyde in Air) Dt= 8.08 10-10 [m2/s]
(Tissue) Db = 1.62 10-9 [m2/s] (Blood)
Metabolism Km1 = 2.01 105 [μg/ m3], Vmax1C = 0.196 [μg/
m3/s]
Kf =1.8 10-2 [s-1]
Non-specific binding Kb = 1.07 10
-7 [s-1]
Blood flow Qb = 9.86 10-5 [m3/s] (= 5920.6 [mL/min])
Compartment 1 Vb = 3.4479 10-3 [m3] (Mucus + epithelium)
Compartment 2 Vb = 0.7896 10-3 [m3] (Sub-epithelium)
Thermoregulation Sensible and latent heat transfer analysis by
2-node model
ISBN: 978-0-646-98213-7 COBEE2018-Paper146 page 418
4th International Conference On Building Energy, Environment
In this study, unsteady breathing cycle model (breathing flow
rate of 7.5L/min) proposed by Gupta et al. was adopted into CSP in
order to reproduce breathing process of the human.
We meticulously followed the benchmark test guidelines for
indoor CFD application to maintain quality control. We reported the
validation results for the flow and convective heat transfer rate
from our CSP using the findings of a wind tunnel test in a previous
study. Grid independence of the flow field inside the airway for a
numerical respiratory tract model was also carefully checked, and
the prediction accuracy of the CFD analysis was also validated
using experimental data (PIV results). However, direct validation
procedure of our PBPK analysis results targeting human airway
tissue has much difficulty because of ethical problem/limitations
of experimental work for health risk assessment using real human
body. For this reason, we carefully checked and applied each model
parameter to the PBPK analysis, to which accuracies were
validated.
RESULTS AND DISCUSSION Figure 4 shows the flow field,
temperature, and formaldehyde concentration distribution around the
CSP. In this result, heat generation at skin surfaces and thermal
plume near the CSP were found as a result of the thermoregulatory
response. In addition, non-uniform distribution of the formaldehyde
in indoor space was observed.
Figure 5 indicates the time series analysis result of
formaldehyde concentration distribution in the human airway and
breathing zone. It was found that contaminant inhalation and
exhalation could be reproduced by unsteady breathing cycle model.
The peak concentration inhaled formaldehyde was approximately
75.3μg/m3, and it was reached into lungs via trachea and bronchus
with the concentration of 33.8μg/m3. Figure 6 shows the time-series
concentration distribution of inhaled formaldehyde at airway wall
surface. It was revealed that predominant adsorption of the inhaled
formaldehyde is located in the nasal cavity.
Finally, formaldehyde concentration distribution in airway
tissue was predicted by PBPK analysis shown in Figure 7. It was
revealed that formaldehyde concentration at air-tissue interface
boundary was fluctuating with the range of 0 ~ 5.3×10-2 [μg/m3] by
the unsteady breathing cycle. We also found that almost the whole
formaldehyde concentration was reacted at the epithelium+mucus
layer, very low concentration of formaldehyde was reached to the
subepithelial layer. As shown in Table 2, we revealed that
saturable metabolic clearance was the predominant mechanism of the
reduction of formaldehyde concentration in tissue.
Through a demonstrative case study, the PBPK-CFD-CSP hybrid
analysis proposed in this study was confirmed to be able to provide
fundamental and quantitative information for health risk
assessments of respiratory exposure in the design stage of indoor
environment.
CONCLUSIONS This study presents a comprehensive prediction
method that integrates PBPK-CFD hybrid model with CSP, and its
application to estimation of inhalation exposure in indoor spaces.
The formaldehyde concentration distributions, doses of inhaled
formaldehyde in the epithelial/sub-epithelial tissue was estimated
in this study. We believe this PBPK-CFD-CSP hybrid model could
contribute toward establishing a comprehensive prediction method
for airway tissue dosimetry, and could applied for setting
threshold concentration or a reference concentration (RfC) for
indoor environmental design.
ACKNOWLEDGEMENT This research was partly supported by a
Grant-in-Aid for Scientific Research (JSPS KAKENHI, 15H04086). We
express special thanks to the funding source.
REFERENCES Yoo SJ and Ito K. 2017. “Numerical Prediction of
Tissue
Dosimetry in Respiratory Tract using Computer Simulated Person
integrated with physiologically based pharmacokinetic
(PBPK)-computational fluid dynamics (CFD) Hybrid Analysis”, Indoor
and Built Environment, in Press (DOI: 10.1177/1420326X17694475)
Corley RA et al. 2015. “Comparative Risks of Aldehyde
Constituents in Cigarette Smoke Using Transient Computational Fluid
Dynamics/Physiologically Based Pharmacokinetic Models of the Rat
and Human Respiratory Tracts”, Toxicological Sciences, 1, 1–24
Gupta JK, Lin CH, Chen Q. 2009. “Characterizing exhaled airflow
from breathing and talking”, Indoor Air, 20, 31-39
Li C, Ito K. 2014. “Numerical and experimental estimation of
convective heat transfer coefficient of human body under strong
forced convective flow”, Journal of Wind Engineering &
Industrial Aerodynamics. 126, 107-117
Ito K, Inthavong K, Kurabuchi T, Ueda T, Endo T, Omori T, Ono H,
Kato S, Sakai K, Suwa Y, Matsumoto H, Yoshino H, Zhang W, Tu J.,
2015. “CFD Benchmark Tests for Indoor Environmental Problems: Part
1, Isothermal/Non-Isothermal Flow in 2D and 3D Room Model”,
International Journal of Architectural Engineering Technology. 2,
1-22.
Ito, K., Inthavong, K., Kurabuchi, T., Ueda, T., Endo, T.,
Omori, T., Ono, H., Kato, S., Sakai, K., Suwa, Y., Matsumoto, H.,
Yoshino, H., Zhang, W., Tu, J., 2015. “CFD Benchmark Tests for
Indoor Environmental Problems: Part 3, Numerical Thermal Manikins”,
International Journal of Architectural Engineering Technology. 2,
50-75.
Phuong N.L., Ito, K., 2015. “Investigation of flow pattern in
upper human airway including oral and nasal inhalation by PIV and
CFD”, Building and Environment. 94, 504-515
Phuong, N.L., Yamashita, M., Yoo, SJ., Ito, K., 2016.
“Prediction of convective heat transfer coefficient of human upper
and lower airway surfaces in steady and unsteady breathing
conditions”, Building and Environment. 100, 172-185
Table 2. Dose of inhaled formaldehyde and contribution of each
reaction term inside tissue(at the moment of peak inhalation,
averaged by volume of whole airway tissue)
Epithelium+mucus Subepithelium
Saturable metabolic clearance
1st order reaction
1st order reaction
Blood perfusion
Reduction rate of HCHO [μg/m3 s] 8.55×10
3 1.58 < 10-5 0.05
ISBN: 978-0-646-98213-7 COBEE2018-Paper146 page 419