Vol. 86, Issue 6, 1984-1993, June 1999
Longitudinal distribution of chlorine absorption in human airways:
comparison of nasal and oral quiet breathing
Vladislav
Nodelman and
James S.
Ultman
Biological Transport Dynamics Laboratory, Department of Chemical
Engineering, Pennsylvania State University, University Park,
Pennsylvania 16802
 |
ABSTRACT |
The fraction of
an inspired chlorine (Cl2) bolus
absorbed during a single breath (
) was measured as a function of
bolus penetration (VP) into the
respiratory system of five male and five female nonsmokers during both
nasal and oral breathing at a quiet respiratory flow of 250 ml/s. The
correspondence between VP and
specific anatomic landmarks was found for each subject by a combination
of acoustic reflection and nitrogen washout measurements. For both
nasal and oral breathing, reached ~0.95 at the distal end of the
upper airways and reached 1.00 within the lower conducting airways. The
values of a regional mass transfer parameter computed from the
-VP data indicated that the
resistance to Cl2 diffusion in the
airway mucosa was negligible compared with the diffusion resistance in
the respired gas. Changing the peak inhaled
Cl2 concentration from 0.5 to 3.0 parts/million did not significantly affect the distribution of
Cl2 absorption, suggesting that
the underlying mass transport and chemical reaction processes were
linear with respect to Cl2 concentration.
air pollution; inhalation toxicology; lung dosimetry; regional
uptake; mass transfer coefficient; conducting airways
| |
INTRODUCTION |
CHLORINE (Cl2)
is a reactive oxidant gas used in the large-scale production of
chlorinated hydrocarbon solvents, polyvinyl chloride plastics, and
pharmaceuticals, in the bleaching of paper, and in the disinfection of
drinking and swimming pool water. The average
Cl2 level in contemporary
occupational settings is mandated to be 0.5 parts/million (ppm) or less
over an 8-h workshift (1). This is less than the 1-3 ppm range in
which people first perceive irritation of their eyes or respiratory
tract (18). This is also below the
Cl2 concentrations at which
respiratory effects have been observed in short-term laboratory
exposures of healthy people. For example, during 8-h controlled
exposures of young adults to 1 ppm
Cl2, there was clearly a decrement
in forced expired lung function and in airway resistance that could
persist for as long as 24 h. These responses were not nearly so
definite during exposure to 0.5 ppm
Cl2, however (22). In a more
recent laboratory investigation in which healthy adults were exposed to
Cl2 at concentrations of 0.5 ppm
or less for 6 h on 3 consecutive days, no changes were found in forced
expired lung function or in biochemical markers of airway inflammation
and epithelial permeability (5).
Over their entire lifetime, people might experience long-term
cumulative Cl2 exposure leading to
irreversible tissue damage and functional impairment, even at
concentrations below 0.5 ppm. To investigate this possibility, female
and male B6C3F1 mice as well as F344 rats were exposed to
Cl2 concentrations of 0.4-2.5 ppm for 6 h/day, 5 days/wk (3 days/wk for female rats) for up to 2 yr
(28). Exposure-dependent nonneoplastic lesions were detected at all
Cl2 concentrations, were confined
to the nasal cavities of both rodents, and were most severe in the
anterior nose. Respiratory and olefactory epithelial degeneration,
septal fenestration, and secretory cell metaplasia were some of the
Cl2-induced histological changes
that were observed even at an exposure concentration of 0.4 ppm. In a
similarly designed study in which rhesus monkeys were exposed to
Cl2 concentrations of 0.1-2.3
ppm for up to 1 yr (11), nonneoplastic nasal lesions were only observed
at the highest exposure concentration. As in the rodents, the severity of the lesions decreased with distal distance into nasal cavities. Although lesions in the monkeys were milder than in the rodents, the
lesions in the monkeys penetrated beyond the nose into the trachea.
To utilize these observations in the prediction of health effects on
people, one must understand the factors that influence the severity and
spatial distribution of
Cl2-induced tissue damage. One of
these factors is the regional pattern of
Cl2 absorption. Because
Cl2 is a highly soluble gas, its
principal site of absorption is most likely the upper airways, the same
airways in which the majority of
Cl2-induced lesions have been
observed in animal studies. Moreover, a progressive loss of
Cl2 from the inhaled gas stream as
it passes over more and more airway surface is a logical explanation for the anterior-to-posterior attenuation of lesion severity observed in monkeys as well as in rodents. The deeper penetration and less severe tissue responses in monkeys may have resulted from slower Cl2 absorption in their
larger-diameter airways and may also have been related to the fact that
primates can breathe in an oronasal fashion whereas rodents are
obligate nasal breathers.
The purpose of the present study was to observe the longitudinal
distribution of Cl2 absorption in
intact human airways during quiet breathing by employing the
noninvasive bolus inhalation method that was previously developed for
ozone (O3), another oxidant gas
that can pollute inhaled air (8). This first required that the
inhalation apparatus be modified so that it could deliver Cl2 boluses and could continuously
monitor Cl2 concentration. By
using the modified apparatus, bolus measurements were compared during
nasal and oral breathing to determine whether the site of air access
influenced the penetration of Cl2
beyond the upper airways. To investigate the possibility of nonlinear
chemical reaction effects in the airway mucosa, bolus measurements were also made at different peak inhaled
Cl2 concentrations. The resulting Cl2 distribution data were
referenced to specific anatomic landmarks by using upper airway and
total conducting airway volumes measured in each subject with acoustic
reflection and nitrogen-washout methods, respectively.
| |
METHODS |
Subject population.
Five healthy male and five healthy female nonsmokers participated in
this investigation. After reading an explanation of the study, each
subject completed an informed consent form, a medical questionnaire,
and a standard spirometry test to determine his or her forced vital
capacity (FVC) and forced expired volume in 1 s
(FEV1). Each subject was
selected without regard to body size and was included in the study only
if he or she met the following criteria: was between 18 and 40 yr old;
had not smoked within the past 3 yr; had no history of hay fever,
asthma, allergic rhinitis, chronic respiratory disease, or
cardiovascular disease; had not used medication within 1 wk of the
experiment; was not exposed to air pollution on a daily basis; did not
swim more than once a week in a chlorinated swimming pool; and had an
FEV1-to-FVC ratio >75% of the
predicted value (12). Each female participant provided menstrual
information, was administered a human chorionic gonadotropin-urine
pregnancy test (QuickVue, Quidel) at the beginning of the session, and
was excluded from the study if this information suggested that she was
pregnant. All procedures were approved by the Institutional Review
Board of the Pennsylvania State University.
Instrument development.
The design and performance of the bolus inhalation system as configured
for O3 measurements were described
in detail in previous publications (8-10). The apparatus
originated from a breathing assembly that contained an injection port
connected to an O3 bolus generator, a pneumotachometer to monitor respired flow, and a sampling
port connected to a fast-responding
O3 analyzer. For the present
study, a new breathing assembly,
Cl2 bolus generator, and
fast-responding Cl2 analyzer were
constructed. To minimize absorption of
Cl2, the internal surfaces of
these devices were fabricated of Teflon, wherever possible.
The breathing assembly (Fig. 1,
1-3) consisted of a specially
machined Teflon tube that incorporated a mild Venturi-type constriction (Fig. 1, 1). The distal end of the
breathing assembly could be fitted with one of two breathing fixtures
(Fig. 1, 2). One fixture was a
flexible plastic mouthpiece, and the other was a nasal cannula. The
nasal cannula was composed of a rigid plastic manifold holding a pair
of flexible rubber connectors that provided a comfortable but tight fit
with both nostrils. The internal volume of each breathing fixture was
20 ml. Located near the distal end of the breathing assembly was a
1/8-in. Swagelock fitting connected to the inlet metering valve of the
Cl2 analyzer. Located near the proximal end of the breathing assembly was a 1/4-in. Swagelock fitting
that was connected to the Cl2
bolus generator. The proximal end of the breathing assembly was fitted
with a pneumotachometer (Fig. 1, 3)
for monitoring respiratory flow.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 1.
Bolus apparatus. 1-3, Breathing
assembly: 1, custom-made PTFE (Teflon)
breathing tube; 2, custom-made PTFE breathing
fixture adapter connected to either a flexible mouthpiece (Hans
Rudolph) or a nasal cannula (Nellcor Puritan Bennett);
3, linear screen pneumotachometer
(R4500C, Hans Rudolph). 4-9,
Bolus generator: 4A and
4B, 3-way subminiature PTFE solenoid
valves (series 1, General Valve); 5A
and 5B, pressure regulators (series
18, Ralph Hiller); 6, flowmeter with
needle valve (series VFB, Dwyer Instruments);
7, custom-made 1.5-inner-diameter × 5-in.-long PTFE diffusion chamber;
8,
Cl2 permeation device,
0.2-10-cm-long (Vici Metronics);
9, 1/8-in.-inner-diameter PTFE hold-up
tube (Berghoff/America), respectively.
10, Data-acquisition system:
analog-to-digital interface and relay control (570DAS, Keithley) and
computer workstation (386SX, Datel).
|
|
The bolus generator (Fig. 1,
4-9) utilized a diffusion
chamber (Fig. 1, 7) containing a
Cl2 permeation tube (Fig. 1,
8) to produce a chlorinated
airstream. The outlet of the diffusion chamber was connected to a pair
of three-way solenoid valves (Fig. 1, 4A and
4B) that were separated by a Teflon
hold-up tube (Fig. 1, 9). When these
valves were in the electrically "off" position, as shown in Fig.
1, chlorinated air continuously flowed through the hold-up tube to an
exhaust. To produce a bolus, the valves were simultaneously energized
by a data-acquisition system (Fig. 1,
10) such that clean air bypassed the
diffusion chamber and all the chlorinated air in the hold-up tube was
rapidly propelled into the subject's breathing tube (Fig. 1,
1). To precisely measure the
Cl2 distribution within the
airways, boluses with a small volume of ~10 ml and reproducible
Cl2 concentration distributions were necessary. This was achieved by employing a volume of 10 ml for
the hold-up tube, a duty cycle of 20 ms for energizing the solenoid
valves, and a pressure of 30 psi (Fig. 1,
5B) for propelling the bolus. The
peak Cl2 concentration of the
boluses could be conveniently regulated by adjusting the flow rate of clean air (Fig. 1, 6) entering the
diffusion chamber.
Because the inner surfaces of the mouthpiece and nasal cannula were
constructed of plastic, they might absorb
Cl2 before the inhaled bolus
reached the respiratory tract. To check this possibility, the breathing
assembly was alternatively fitted with the oral and the nasal fixture,
and clean air was supplied to the open end of the pneumotachometer at a
flow of 150 ml/s. The concentration of
Cl2 was monitored immediately
proximal to the breathing fixture as six boluses were separately
injected into the breathing assembly. The
Cl2 concentration was then
monitored just distal to the fixture as six additional boluses were
injected. By integrating the resulting Cl2 concentration curves, the mass
of Cl2 entering (proximal
sampling) and the mass exiting (distal sampling) the fixture were
determined. A two-tailed t-test
indicated no significant difference (P > 0.05) between these two sets of measurements for either fixture.
Therefore, Cl2 absorption by
either the mouthpiece or nasal cannula was unlikely, even at the
relatively low airflow of 150 ml/s that exaggerated the contact time of
Cl2 boluses with the breathing fixtures.
The fast-responding Cl2 analyzer
developed for this study (19) was based on the principle of thermionic
ionization. The commercial detector used in the analyzer consisted of
an electrically heated catalytic cathode that ionized
Cl2 to
Cl
and a metallic anode
that functioned as an electron collector when biased at a positive
voltage relative to the cathode. At a sample inlet flow of 600 ml/min,
the analyzer had a sensitivity of 3 pA ppm, a minimum detection limit
of 0.04 ppm, a 10-90% step-response time of 80 ms, and a delay
time of 100 ms. The static calibration of the instrument was linear for
Cl2 concentrations from 0.03 to
4.0 ppm and was insensitive to variations in temperature,
CO2 concentration, and humidity
that normally occur in respired air.
In addition to Cl2, the thermionic
detector could ionize chlorinated vapors such as chloroform, methyl
chloride, and carbon tetrachloride (17). Therefore, chlorinated vapors
originating from either endogenous
Cl2 metabolism (4, 13, 23) or the reaction of inhaled Cl2 with
airway mucosa could be wrongly interpreted as expired
Cl2. To check this possibility,
the chemical composition of expired breath was measured in three male
and two female subjects while they breathed clean air or chlorinated
air. In each measurement, 1 liter of slowly expired air was collected
in a 9 × 9-in. Teflon bag (F5009009, Eagle Picher). The bag was
then connected to a gas chromatograph mass spectrometer (5890IIGC/5792A
MSD, Hewlett-Packard) with a Nafion drying tube (MD-070-48F, Perma
Pure). The chromatograph utilized a capillary column with a high
affinity for halogenated organic compounds (2-4154, Supelco).
Cryofocusing at the sample inlet of the instrument and again at the
head of the chromatographic column concentrated the condensable
components of the expired breath sample to achieve a minimum detectable
limit of ~0.001 ppm (8533 Cryo-concentrator, Graseby-Nutech).
One expired breath sample was collected immediately after clean air
breathing for 2 min, and three replicate breath samples were obtained
at the end of a 2-min exposure to 0.5 ppm
Cl2. Chlorinated vapors were found
in only 2 of the 20 expired breath samples collected from the 5 subjects. In both cases, the concentration of the chlorinated compound
in the breath sample was on the order of 0.002 ppm. Because the minimum
detection limit of the thermionic ionization analyzer was 10 times
larger than this, expired chlorinated compounds were unlikely to affect
the Cl2 bolus measurements.
Bolus inhalation measurements.
All 10 subjects participated in a 2- to 4-h session in which bolus
measurements were made during nasal and oral quiet breathing. The
subject was seated comfortably on a stool, wore noseclips during oral
breathing, and maintained a closed mouth during nasal breathing. To
carry out a bolus test breath, the subject donned the mouthpiece or
nasal cannula, activated the inhalation apparatus by depressing a
handheld switch, and inhaled beginning at functional residual capacity
while viewing a computer monitor on which the integrated
pneumotachometer signal (i.e., the respired volume) was displayed in
real time. The subject controlled his or her breathing so that the
respired volume signal followed a predrawn pattern corresponding to
equal inspiratory and expiratory flows of 250 ml/s and a
tidal volume of 500 ml. At a predetermined time during inhalation, the
data-acquisition system automatically injected a 10-ml
Cl2 bolus into the inspired airflow.
Penetration of the bolus into the respiratory system could be
systematically varied from breath to breath by changing the bolus
injection time relative to the time that the subject was supposed to
switch from inhalation to exhalation. The earlier the injection time
relative to the end of inhalation, the greater the penetration of the
bolus distal to the airway opening. Throughout a test breath, the
Cl2 analyzer and pneumotachometer
voltage outputs were continuously recorded on a computerized
data-acquisition system at a sampling rate of 200 Hz. The system was
also used for triggering the bolus injection valves, integrating the
pneumotachometer output to obtain respired volume, and displaying the
respired volume on the breathing monitor. The subject took 2-3
test breaths/min, and a collection of 50-70 breaths between bolus
penetrations of 0-200 ml constituted a complete experiment.
Three experiments were conducted during the bolus inhalation session:
oral breathing with a peak inhaled
Cl2 concentration (Cmax) of 3.0 ppm; nasal
breathing with a Cmax of 3.0 ppm;
and nasal breathing with a Cmax of
0.5 ppm. To avoid possible systematic errors associated with
Cl2 preexposure, the sequence in
which the experiments were carried within a session was randomized for each subject. In addition, bolus test breaths were arbitrarily carried
out at progressively increasing penetration in some experiments but at
progressively decreasing penetration in others. In all experiments, a
test breath was deemed acceptable if the subject could maintain an
average respiratory flow within ±15% of 250 ml/s.
Anatomic measurements.
The anatomy of the respiratory system of each subject was characterized
by measurements of FVC by using a forced spirometer (model 110 automated spirometer, CDX), and dead space
(VD) by using a previously
described nitrogen-washout apparatus (3). In addition, the nasal volume
(VNS), oral volume
(VOR), and pharyngeal volume
(VPH) of each subject were
determined by an acoustic reflection apparatus. The value of FVC
represented the average of the highest two of three forced expired
measurements, VD was the average
of 5-7 washout measurements, and
VNS,
VOR, and
VPH were each determined as the
average of six acoustic reflection measurements.
The commercially available acoustic reflection apparatus (Eccovision
Acoustic Rhinometry-Pharyngometry System, Hood Laboratories) consisted
of two acoustic wave tubes, one fitted with a rubber mouthpiece and
used to obtain oropharyngeal geometry and the other fitted with a
rubber nosepiece and used to obtain unilateral nasopharyngeal geometry.
During an acoustic measurement, a subject was seated comfortably on a
stool and either grasped the mouthpiece or placed the nosepiece
adjacent to one nostril. In the case of the nasal measurements, a small
amount of petroleum jelly was applied to the tip of the nosepiece to
ensure a good seal with the external nares. The subject was instructed
to maintain an upright posture and perform a slow expiration while
relaxing the glottis. In each measurement, a pulse of white noise was
generated in the wave tube, propagated into the subject's airways, and
was partially reflected because of changes in the cross-sectional area
of the airway. The incident and reflected sound were monitored with
microphones and then converted to a cross-sectional area vs. distance
function by using a proprietary computer algorithm.
The length of the nasal cavity and the length of the nasopharynx were
estimated from two points of minimum cross section in the nasal
area-distance function as previously illustrated by Swift and Proctor
(24) (Fig.
2A).
Similarly, the length of the oral cavity including the oropharynx and
the length of the hypopharynx were estimated from two points of minimum
cross section in the oral area-distance function, as previously shown
by Fredberg and associates (7) (Fig.
2B). The volumes of the left and
right nasal cavities were determined by integrating each nasal
area-distance function over the length of the nasal cavity. The volume
of the nasopharynx was calculated as an average of the values obtained by integrating the two nasal area-distance functions over the length of
the nasopharynx. The value of VNS
was then calculated as the sum of the volumes of the nasal cavities and
the nasopharynx. The value of VOR
was obtained by integrating the oral area-distance function over the
length of the oral cavity including the oropharynx, and the value of
VPH was obtained by integrating
the oral area-distance function over the length of the hypopharynx.
Because VD measurements were
made during oral breathing, the volume of a subject's lower conducting
airways, VLA, was computed as
(VD 
VOR VPH).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 2.
Representative acoustic reflection measurements. Area-distance profiles
of left nasopharyngeal airways (A)
and oropharyngeal airways (B) in 1 subject.
|
|
Analysis of bolus inhalation data.
The integrals (MI and
ME), means (VP and VB),
and variances (
2I, 2E) of
the inhaled and the exhaled portion of each test breath were obtained
by numerical integration of the
Cl2 concentration data with
respect to respired volume as illustrated in Fig.
3. These moments were then
interpreted as follows: absorbed fraction (; i.e., 1 ME/MI)
was the amount of pollutant absorbed during a single respiratory cycle
relative to the inhaled amount; penetration volume
(VP) represented the mean airway
volume that would be reached by inhaled
Cl2 molecules relative to the
gas-sampling point if no absorption had occurred; breakthrough volume
(VB) symbolized the mean airway
volume traversed by unabsorbed Cl2
molecules that reached the gas-sampling point during expiration; and
dispersion (2; i.e.,
2E 2I) was a measure of the
longitudinal mixing of the unabsorbed
Cl2 molecules. A more quantitative
definition of these variables can be found elsewhere (8).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 3.
Representative bolus test breath. Respired volume axis is obtained by
integrating repiratory flow relative to end inspiration. The physical
interpretation of the integrals (MI,
ME), means (VP, VB), and
variances (2I,
2E) of the inhaled and exhaled
portions of the curve are given in the text.
|
|
By viewing VP as an independent
variable that characterized the spatial excursion of an inhaled bolus
into the respiratory system, the distributions of ,
VB, and
2 with respect to
VP were considered to be the
primary dose-distribution information retrieved with the bolus
inhalation method. The -VP distribution for a particular subject during one of the three experimental conditions is illustrated in Fig.
4. To determine the pooled
-VP,
VB-VP,
and
2-VP
distributions for all subjects at each experimental condition, the ,
VB, and
2 values collected from all
test breaths were sorted into 10-ml increments of
VP and averaged. The overall SE of
each average, including both between-subject and within-subject
variations, was computed as previously specified by Kabel and
associates (10).
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 4.
Regression of diffusion model to Cl2 distribution data
obtained during nasal breathing in 1 subject. Each point represents
absorption fraction () obtained from a bolus test breath; smooth
curve, splined regression of these data according to
Eq. 1. Because of extensive absorption
in nasal cavity and pharynx, data were insufficient to determine a mass
transfer parameter in lower conducting airways. VP 0,
VP 1, and VP 2: boundaries of
airway compartments as specified in Fig. 5.
|
|
Compartmental analysis and the overall mass transfer coefficient.
In modeling Cl2 absorption, the
respiratory system was subdivided into nasal-oral (N/O), pharyngeal
(PH), lower airway (LA), and respiratory air space (RA) compartments
(Fig. 5). Longitudinal position was
specified by the penetration VP of
a bolus distal to the Cl2-sampling
point. The N/O compartment was bounded at its proximal end by the
volume of the breathing fixture
(VP 0 
VBF) consisting of the nasal
cannula during nasal breathing and the mouthpiece during oral
breathing. The PH compartment was bounded at its proximal end by
VP 1 (VBF
+VN/O), where
VN/O VNS during nasal breathing and
VN/O VOR during oral breathing. The LA
compartment was bounded at its proximal end by
VP 2 (VBF + VN/O + VPH), and the RA compartment was
bounded at its proximal end by
VP 3 (VBF + VN/O + VPH + VLA).
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 5.
Compartment airway model. Conducting airways are divided into 3 compartments, and longitudinal position within this model is specified
by VP (definitions of VBF, VN/O,
VPH, and VLA appear above
abbreviations).
|
|
The transport of a solute from respired gas to adjacent tissue may be
characterized by an overall mass transfer coefficient (K) representing the absorption rate
normalized by the solute concentration in the gas phase and by the
surface area through which the solute is absorbed (25). In situations
where the surface area is not known, a surface-to-volume ratio
a is customarily introduced such that
Ka is the absorption rate normalized by solute concentration and the volume of the gas-filled conduit. The diffusion model published by Hu and associates (9) was utilized to estimate the
values of Ka in the N/O, PH, and LA compartments. The model predicts a -VP distribution
given by the following formula
![ln (1 − &Lgr;) = − (2/<A><AC>V</AC><AC>˙</AC></A>)[(<IT>Ka</IT>)<SUB>N/O</SUB>(V<SUB>P</SUB> − V<SUB>P0</SUB>) + <IT>I</IT><SUB>1</SUB>(&Dgr;<IT>Ka</IT>)<SUB>PH</SUB>(V<SUB>P</SUB> − V<SUB>P1</SUB>) + <IT>I</IT><SUB>2</SUB>(&Dgr;<IT>Ka</IT>)<SUB>LA</SUB>(V<SUB>P</SUB> − V<SUB>P2</SUB>)]](/content/vol86/issue6/fulltext/1984/img001.gif)
|
(1)
|
where
is the respiratory flow rate,
(
Ka)PH [(Ka)PH (Ka)N/O] represents the change in
Ka between the PH and N/O compartments, and
(Ka)LA [(Ka)LA (Ka)PH] represents the change in
Ka between the LA and PH compartments. The
I1 and
I2 indicator
variables are defined as follows:
I1 is unity if
VP > VP 1 and is zero otherwise, and I2 is unity
if VP > VP 2 and is zero otherwise.
The principal assumptions in this model were that
Cl2 absorption is a
quasi-steady-state process, Ka is constant within a
compartment, and that longitudinal transport of
Cl2 in the gas phase by
longitudinal dispersion can be neglected in comparison with transport
by longitudinal convection. The factor of two multiplying the
right-hand side of Eq. 1 accounts for
the fact that the Cl2 bolus must
pass over the airway surface twice, once during inhalation and again
during exhalation.
To estimate the values of (Ka)N/O,
(Ka)PH, and
(Ka)LA, a splined least squares regression
of each subject's -VP data to Eq. 1 was separately performed for
each of the three experiments (Fig. 4). The parameters
, VNS,
VOR,
VPH, and
VLA were fixed at those values
measured for the subject of interest. Values of > 0.96 were not
used in the regression because they resulted from measurements of
expired concentration that were near the resolution of the
Cl2 analyzer. Furthermore, when
increased above 0.96 before a bolus could penetrate halfway through
a compartment, the corresponding Ka value was disregarded.
Once the regressions were completed, the fraction of inhaled
Cl2 absorbed within each of the
four compartments () was determined. In particular, values of at the proximal boundaries of the PH, LA, or RA compartments were
computed by using Eq. 1 with the estimated values of
Ka. The value of was then calculated as the
difference between at the distal and proximal boundaries of the
compartment of interest.
Unpaired comparisons of anthropometric characteristics between the male
and female subjects, of compartmental volumes, and of values
were performed by using two-tailed Student's
t-tests. Paired comparisons of
Ka at different experimental conditions were made when
values were available in the same group of subjects. Otherwise, the
comparisons of Ka were unpaired. In all cases, the
difference between two parameter values were considered to be
significant if the probability that they were the same was <0.05.
| |
RESULTS |
Characteristics of subjects.
All the subjects were physically fit individuals; their anthropometric
characteristics are summarized in Table
1. Although the men were
generally older, taller, heavier, had larger FVC, and had larger
VD than did the women, only the
difference in weight was statistically significant
(P = 0.007). Compartment volumes of
the subjects are summarized in Table 2.
Mean values of VNS were similar,
VOR was somewhat smaller, and
VPH was considerably larger
in the men than in the women. The mean value of
VNS in the subject population as a whole was
somewhat smaller than the corresponding value of
VOR. Of these diferences in average
compartment volumes, only the gender difference in
VPH was statistically significant (P < 0.001).
There was a large intersubject variability in all the compartment
volumes listed in Table 2. As one would expect,
VD was a good predictor of
VLA in the different subjects
(coefficient of determination adjusted for number of subjects:
r2 = 0.71) but
was not a predictor of VNS,
VOR, and
VPH
(r2 = 0.00, 0.00, and 0.03, respectively). Whereas
VD as well as VNS were somewhat correlated with
weight (r2 = 0.28 and 0.27, respectively), VOR was
not correlated with weight (r2 = 0.00). The
value of VD was strongly
correlated with height (r2 = 0.81), but
neither VNS nor
VOR was correlated with height
(r2 = 0.03 and
0.00, respectively). Finally, there was no correlation between
VNS and
VOR
(r2 = 0.00).
Cl2 absorption.
The -VP distributions pooled
for all participants at each of the three experimental conditions
(i.e., oral breathing, Cmax = 3.0 ppm; nasal breathing, Cmax = 3.0 ppm; nasal breathing, Cmax = 0.5 ppm) are given in Fig. 6. In general, the
inhaled Cl2 boluses were
completely absorbed at a VP of
~80 ml, which was immediately distal to the upper airways.
Considering the magnitude of the SE bars on these graphs, there appears
to be little difference between the
-VP distributions during oral
and nasal breathing. On the other hand, decreasing
Cmax from 3.0 to 0.5 ppm appears to increase the absorbed fraction of
Cl2 at
VP below 60 ml, which is within
the hypopharynx. The
VB-VP
and
2-VP
distributions pooled for all participants during oral and nasal
breathing at Cmax = 3.0 ppm are
given in Fig. 7. The relationship between
VB and
VP is similar for both modes of
breathing, with oral values of VB
being somewhat larger than nasal values at
VP > 10 ml. Values of
2 appear relatively insensitive
to VP, with oral values being
somewhat larger than nasal values at
VP < 70 ml.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 6.
Pooled -VP distributions for 10 subjects. A: nasal and oral quiet
breathing at a peak inhaled Cl2
concentration of 3.0 parts/million (ppm).
B: nasal quiet breathing at peak
inhaled Cl2 concentrations of 3.0 and 0.5 ppm. Vertical bars, overall SE because of within-subject and
between-subject variations.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 7.
Pooled
VB-VP
and
2-VP
distributions for 10 subjects. Nasal and oral quiet breathing at a peak
inhaled Cl2 concentration of 3.0 ppm. Error bars are defined as in Fig. 6.
|
|
Table 3 summarizes the mass transfer
parameters that were calculated from an individual regression of each
subject's -VP distribution.
The (Ka)NS averaged for all subjects was
significantly larger (P = 0.04) than
(Ka)OR. On the other hand, there was not a
significant difference (P = 0.97) between the average
(Ka)NS obtained when
Cmax was 0.5 ppm and that when
Cmax was 3.0 ppm. Although the
data were insufficient to calculate (Ka)PH
in some of the subjects, there was nevertheless a consistent trend of negative values at all three experimental conditions, indicating that
Ka decreases between the nasal or oral cavity and the
hypopharynx. The values of (Ka)PH determined
by summing the average values of (Ka)PH and
(Ka)N/O were 1.5 s1 during
oral breathing of 3 ppm Cl2
boluses; 1.1 s1 during
nasal breathing of 3 ppm Cl2
boluses; and 1.6 s1 during
nasal breathing of 0.5 ppm Cl2
boluses. This suggests that mass transfer in the hypopharynx was not
markedly affected by the mode of breathing or the inhaled
Cl2 concentration. Because of
extensive absorption of Cl2 in the
NS and OR compartments, there were insufficient
-VP data in the LA compartment
to evaluate (Ka)LA in any of the subjects.
View this table:
[in this window]
[in a new window]
|
Table 3.
Compartmental mass transfer parameters in individual subjects during
quiet oral and nasal breathing with peak inspired chlorine
concentration of 0.5 and 3.0 ppm
|
|
The for Cl2 in the three
conducting airway compartments is given in Fig.
8. Nearly all of the inspired
Cl2 was absorbed in the upper
airways, and ~90% of the inspired
Cl2 was absorbed in the nose or
mouth of all the subjects. This result was independent of the mode of
breathing and of Cmax.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 8.
Pooled compartmental Cl2
absorption for 10 subjects. Each histogram level () represents
the difference between on the corresponding compartment boundaries.
Values of for respiratory air space compartment were 0 for all
subjects. Error bars are defined as in Fig. 6.
|
|
| |
DISCUSSION |
One purpose of this research was to determine the principal site of
Cl2 uptake in the respiratory
system. From the bolus inhalation measurements, it is clear that nearly
all of the Cl2 inhaled during quiet breathing is absorbed in the upper airways, whether the nose or
the mouth is the site of air access. In reaching this conclusion, we
relied on measurements of VOR,
VNS, and
VPH measured by acoustic
reflection. The average volume of the nasal cavity, between the nares
and the entrance to the nasopharynx, was found to be 23 ml. This is
within the range of 20-32 ml reported for the nasal cavity volume
by two other research groups who used computerized tomography and
magnetic resonance imaging scans in a total of eight subjects (16). The
average ± SD volume of the oral cavity between the lips and the
entrance to the hypopharynx was found to be 54 ± 20 ml, which is
consistent with the values of 41 ± 9 and 50 ± 10 ml previously
reported for acoustic reflection and oropharyngeal cast measurements,
respectively (10).
Another purpose of this study was to compare
Cl2 uptake by the nose and the
mouth. Because (Ka)NS is ~25% larger than
(Ka)OR, the absorption rate of
Cl2 per unit volume of airway
lumen is somewhat larger for the nose than for the mouth (Table 3). The nose has a smaller volume than the mouth (Table 2), however, so that
(KaV)NS is only ~5% larger than
(KaV)OR. It can be concluded that the total
absorption rates for the nose and mouth are similar. A third purpose of
this research was to examine the influence of inhaled
Cl2 concentration on airway
absorption during nasal breathing. The fact that
(Ka)NS retained the same value when
Cmax was changed from 0.5 to 3.0 ppm (Table 3) indicates that the dissolution, diffusion, and chemical
reactions governing Cl2 uptake from respired gas to the nasal mucosa are all linear processes.
The continuous inhalation of Cl2
is the theoretical equivalent of inhaling a train of
Cl2 boluses that is introduced
between the beginning and the end of inspiration. The absorption
distribution that applies during continuous exposure can therefore be
estimated by integrating the
-VP bolus distribution. To a
first approximation, the change in bolus absorption is an
estimate of Cl2 absorption within
a particular airway compartment during continuous exposure. This
approximation is valid when tidal volume is much larger than VP at the proximal end of the
compartment of interest. This is always the case for the N/O and PH
compartments and is usually true for the LA compartment. Judging from
the results of this study, 95% or more of continuously inhaled
Cl2 is absorbed in the upper
airways that consist of the N/O and PH compartments (Fig. 8).
Cl2 undergoes a rapid and
reversible hydrolysis in aqueous solutions according to the following
chemical reaction
|
(2)
|
This
reaction has an equilbrium constant with the following value at
25°C (27)
![10<SUP>−pH</SUP>[Cl<SUP>−</SUP>][HOCl]/[Cl<SUB>2</SUB>] = 5 × 10<SUP>−4</SUP>(mol/l)<SUP>2</SUP>](/content/vol86/issue6/fulltext/1984/img003.gif)
|
(3)
|
where
[Cl] and
[HOCl] indicate the molar concentrations of the chloride
ion and hypochlorous acid in the aqueous phase, respectively, and
[Cl2] is the
concentration of Cl2 in dissolved
form. For mucus that has a
[Cl] of ~0.16
mol/l (14) and a pH of ~6.6 (2), Eq. 3 indicates that the concentration of
Cl2 in hydrolyzed form
([HOCl]) is 120,000 times
[Cl2]. In other words,
the effective solubility of Cl2
between the respired gas and mucus phase is five orders of magnitude
larger than the physical solubility. This large effective solubility explains why >95% of inhaled
Cl2 was absorbed in the upper
airways. Moreover, the linearity of the equilibrium relationship
between [HOCl] and
[Cl2] that occurs when
pH and [Cl] are
fixed is consistent with the observation that
(Ka)NS did not depend on the peak inhaled
Cl2 concentration.
As Cl2 absorbs into an airway, it
encounters a diffusion resistance created by a respiratory gas boundary
layer and a second diffusion resistance imposed by the surrounding
mucosa. Because of the large effective solubility of
Cl2, the diffusion resistance in
mucus and tissue should be minimal, and gas absorption should be
limited primarily by the convection and diffusion processes in the
respired gas phase. If this is true, then the overall mass transfer
parameter Ka should be similar in value to the gas-phase mass transfer parameter kga. To make
this comparison, the gas-phase mass transfer coefficient
kg was computed
by applying the heat-mass transfer analogy to an empirical correlation
developed from heat transfer measurements on casts of the upper airways
(20)
|
(4)
|
where
m and
n are empirical constants that have
respective inspiratory values of 0.028 and 0.854 in the nose and 0.035 and 0.804 in the mouth and respective expiratory values of 0.0045 and
1.08 in the nose and 0.0006 and 1.269 in the mouth;
Dg is the binary
diffusivity of Cl2 in air, which
was estimated to be 0.13 cm2/s
(25); d is tracheal diameter, which is
~1.8 cm (26); and A is the mean
tracheal cross-sectional area, which is ~2.6
cm2 (26). From previously
published anatomic data, the surface-to-volume ratio
a was estimated to be 9.2 cm1 in the nose (16) and
1.7 cm1 in the mouth (21).
On the basis of these estimates, the predicted
kga averaged
over inspiration and expiration at a of 250 ml/s
were 7.7 s1 in the nose and
1.0 s1 in the mouth. The
kga value predicted for the nose is
within the intrasubject range of (Ka)NS values
from the bolus experiments (i.e., 5.0-8.1
s1), supporting the
conclusion that gas-phase diffusion is the rate-limiting step in the
overall absorption process. However, the
kga predicted for the mouth is much
below the range of (Ka)OR deduced from the bolus study (i.e., 2.9-6.5
s1). This result cannot
be explained by the influence of a mucosal resistance that would cause
kga to be larger, not smaller, than (Ka)OR. The result is possibly due to a
difference in the mouth opening and tongue placement of the cadaver
casts used to obtain kg and
a compared with the mouth and tongue
orientations of the subjects who breathed through a mouthpiece during
the bolus experiments.
Previous bolus measurements indicated that
O3 had a value of 0.8 in the
nasal cavity and 0.5 in the oral cavity (10). The value for
Cl2 in the present investigation
was ~0.9 in both the nasal and oral cavities. The smaller absorbed
fractions of O3 relative to
Cl2 may be due to the fact that
O3 does not hydrolyze in mucus. In
other words, the solubility of O3
may be sufficiently low that the diffusion resistance of
O3 through the mucosa influences the overall absorption process. In addition, because
Cl2 absorption is similar in the
mouth and nose but O3 absorption
is less in the mouth than in the nose, it appears that the mucosal
diffusion resistance of O3 is
greater in the mouth than in the nose.
The toxicity of Cl2 is mediated
locally by HOCl, which may disrupt the integrity and increase the
permeability of the epithelium (6), and by an increase in
H+, which may decrease blood pH,
provided that a sufficient dose of
Cl2 is absorbed (29).
Cl2 also reacts with the
sulfhydryl groups of the amino acid cysteine, thereby inhibiting
various enzymes (6, 15). Because this study revealed that almost all
inhaled Cl2 was absorbed in the
nose during nasal breathing and the mouth during oral breathing, it is
reasonable to conclude that the upper airways are the most likely site
of long-term Cl2-induced tissue
damage in humans. This is consistent with the findings in laboratory
animals that lesions due to the chronic inhalation of
Cl2 are confined primarily to the
nasal cavities (11, 28).
One of the general limitations of the bolus inhalation method is a lack
of spacial resolution because of the finite width of the bolus. The
bolus is most narrow when it enters the respiratory system but becomes
progressively more dispersed by longitudinal mixing throughout the test
breath. The fact that VP 0,
the intercept of the -VP
distribution with the abscissa, is less than the 20-ml volume of the
nonabsorbing breathing fixture (Table 3) is a symptom of this
limitation. In other words, if the width of the bolus had been zero,
then would rise above zero only when
VP was greater than
VBF. Because the bolus had a
finite width, however, a portion of the bolus penetrated into the
respiratory system even when its center of mass was located at values
of VP < VBF. This artifact can lead to an
incorrect interpretation of the
-VP distribution. For example,
it appears in Fig. 6A that
Cl2 absorption at low VP is somewhat more efficient
during oral than nasal breathing. In fact, because bolus dispersion is
greater during oral than nasal breathing (Fig.
7A),
VP 0 is less during oral
than during nasal breathing (Table 3), and the
-VP distribution for oral breathing is shifted to the left of that for nasal breathing. Judging
from the values of (Ka)NS and
(Ka)OR that are based on the slopes of the
-VP data, it appears that the
absorption efficiency in the nose is actually greater than in the
mouth. The same logic applies to the effect of
Cmax on
Cl2 absorption. From Fig.
6B, it appears that absorption is more
efficient at a Cmax, of 0.5 ppm
than at a Cmax of 3.0 ppm, whereas
the values of (Ka)NS indicate that there is
actually no significant difference.
An important consideration in any dosimetric process is whether the
initial uptake of a pollutant causes some disturbance in the chemical
or physiological state of the respiratory system that induces a
subsequent change in the uptake. For example, exposure to
Cl2, whether continuously or by
multiple bolus breaths, can result in the accumulation of
Cl2 in mucus, with a concomitant decrease in the absorption rate. To test for such an effect, one subject nasally inhaled a series of 3-ppm boluses once every five breaths for 1 h at a VP target
level of 40 ml, corresponding to the posterior nasal cavity. A linear
regression of against time had a small coefficient of determination
(r2 = 0.03) and a
slope that was not significantly different from zero. It can thus be
concluded that the bolus inhalation method provides a time-invariant
measure of Cl2 absorption.
Although no statistically significant gender differences in the values
of Ka and the values of were uncovered, this study lacked sufficient power to conclude with high probability that Cl2 absorption is the same for men
and women. In particular, the probability of falsely concluding that
there was no gender difference was ~0.9. To reduce this probability
of making a type II error to a more reasonable level of 0.2 would have
required that 10 female and 10 male subjects be tested. Nevertheless,
the apparent lack of a gender difference in
Cl2 absorption is consistent with the previous finding for O3
absorption that Ka was related to gender only insofar as the
smaller conducting airway volumes of the women produced larger values
of Ka than for the men (3). In the case of
Cl2 that was primarily absorbed
into the N/O compartment, one would expect
(Ka)N/O to be negatively correlated with
VN/O, irrespective of gender. This
expectation was met for the mouth but not for the nose (Fig.
9). It may be that the shape of the oral
cavity was more similar from person to person than was the shape of the
nasal cavity and/or that flow patterns in the mouth were more similar
among different subjects than in the nose.
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 9.
Correlation of mass transfer parameter Ka for individual
subjects against their VN/O
values. Data were obtained at a maximum inspired
Cl2 concentration of 3.0 ppm
during quiet oral (A;
r2 = 0.71) and nasal
(B; r2 = 0.15)
breathing. Nasal inhalation measurements obtained at a maximum inspired
Cl2 concentration of 0.5 ppm (data
not shown) showed a lack of correlation similar to
B.
|
|
The breakthrough volume VB and
dispersion 2 are indirectly
affected by the absorption process. Because it represents the average airway volume from which unabsorbed
Cl2 molecules originate during expiration, VB should
progressively increase as VP
increases. However, absorption is very efficient in the small airways
where a is exceedingly large, so that
VB should level off once
VP is larger than some critical
value. This is exactly the type of
VB-VP behavior that has been observed for
Cl2 boluses in the present study
(Fig. 7A) as well as for
O3 boluses in a previous study (10). As was the case for O3, the
VB for
Cl2 tends to be greater during
oral breathing than during nasal breathing and is not markedly affected
by a change in Cmax. The value of
2, which characterizes the
volume of air in which the test gas becomes distributed by longitudinal
mixing, will also tend to level off because boluses are truncated by
the highly efficient absorption occurring in the small airways. As was
previously observed for O3,
2 for
Cl2 is relatively independent of
VP and
Cmax and is greater during oral
breathing than during nasal breathing.
To summarize, measurements of the
-VP distribution of
Cl2 during nasal as well as oral
quiet breathing in five men and five women indicated that >95% of
inspired Cl2 was absorbed in the upper airways of all subjects, whereas the dose delivered to the respiratory air spaces was negligible. Although there were no statistically significant gender differences in the results, individual values of (Ka)OR were inversely correlated
with individual values of VOR.
Representative overall mass transfer coefficients estimated in the nose
were in good agreement with gas-phase mass transfer coefficients
calculated from established correlations. This suggested that
diffusional resistance in the nasal mucosa was negligible relative to
diffusional resistance in the respired gas. Both the high absorptivity
of Cl2 in the upper airways and
the domination of the gas-phase diffusion resistance were attributable
to the rapid hydrolysis of Cl2 in
the mucosa.
| |
ACKNOWLEDGEMENTS |
The authors are grateful to Janice Schneider of the Applied
Research Laboratory and John Showalter for assistance with the chromatographic analyses of expired breaths and to Abdelaziz Ben-Jebria for insightful suggestions.
| |
FOOTNOTES |
This work was funded by the Chlorine Institute through a subcontract
with the Chemical Industry Institute of Toxicology. Clinical support
was provided by the General Clinical Research Center through funding by
National Heart, Lung, and Blood Institute Grant M01 RR-10732.
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Address for reprint requests and other correspondence: J. Ultman, Dept.
of Chemical Engineering, The Pennsylvania State Univ., 106 Fenske Lab,
University Park, PA 16802 (E-mail: jsu{at}psu.edu).
Received 24 September 1998; accepted in final form 18
February 1999.
| |
REFERENCES |
1.
American Conference of Government Industrial Hygienists.
1996-1997 Threshold Limit Values for Chemical Substances, and Physical Agents and Biological Exposure Indices. Cincinnati, OH: American Conference of Government Industrial Hygienists, 1996.
2.
Bodem, C. R.,
L. M. Lampton,
D. P. Miller,
E. F. Tarka,
and
E. D. Everett.
Endobronchial pH. Relevance to aminoglycoside activity in gram-negative bacillary pneumonia.
Am. Rev. Respir. Dis.
127:
39-41,
1983[Medline].
3.
Bush, M. L.,
P. T. Asplund,
K. A. Miles,
A. Ben-Jebria,
and
J. S. Ultman.
Longitudinal distribution of ozone absorption in the lung: gender differences and intersubject variability.
J. Appl. Physiol.
81:
1651-1657,
1996[Abstract/Free Full Text].
4.
Conkle, J. P.,
B. J. Camp,
and
B. E. Welch.
Trace composition of human respiratory gas.
Arch. Environ. Health
30:
290-295,
1975[Medline].
5.
Emmen, H. H.,
E. M. G. Hoogendijk,
P. J. A. Borm,
and
R. Schins.
Nasal Inflammation and Respiratory Parameters in Human Volunteers During and After Repeated Exposures to Chlorine. Zeist, The Netherlands: Netherlands Organization for Applied Scientific Research, 1997. (TNO Nutrition and Food Research Institute, Netherlands Organization for Applied Scientific Research, Rep. V97.517)
6.
<Environmental Health Criteria 21. Chlorine and
Hydrogen Chloride. Geneva, Switzerland: World Health
Organization, 1982, p. 3.
7.
Fredberg, J. J.,
M. E. B. Wohl,
G. M. Glass,
and
H. L. Dorkin.
Airway area by acoustic reflections measured at the mouth.
J. Appl. Physiol.
48:
749-758,
1980[Abstract/Free Full Text].
8.
Hu, S. C.,
A. Ben-Jebria,
and
J. S. Ultman.
Longitudinal distribution of ozone absorption in the lung: quiet respiration in healthy subjects.
J. Appl. Physiol.
73:
1655-1661,
1992[Abstract/Free Full Text].
9.
Hu, S. C.,
A. Ben-Jebria,
and
J. S. Ultman.
Longitudinal distribution of ozone absorption in the lung: effects of respiratory flow.
J. Appl. Physiol.
77:
574-583,
1994[Abstract/Free Full Text].
10.
Kabel, J. R.,
A. Ben-Jebria,
and
J. S. Ultman.
Longitudinal distribution of ozone absorption in the lung: comparison of nasal and oral quiet breathing.
J. Appl. Physiol.
77:
2584-2592,
1994[Abstract/Free Full Text].
11.
Klonne, D. R.,
C. E. Ulrich,
M. G. Riley,
T. E Hamm, Jr.,
K. T. Morgan,
and
C. S. Barrow.
One-year inhalation toxicity study of chlorine in rhesus monkeys (Macaca mulata).
Fundam. Appl. Toxicol.
9:
557-572,
1987[Medline].
12.
Knudsen, R. J.,
R. C. Slatin,
M. D. Lebowitz,
and
B. Burrows.
The maximal expiratory flow-volume curve.
Am. Rev. Respir. Dis.
113:
587-600,
1976[Medline].
13.
Krotoszynski, B. K.,
G. M. Bruneau,
and
H. J. O'Neill.
Measurement of chemical inhalation exposure in urban population in the presence of endogenous effluents.
J. Anal. Toxicol.
3:
225-234,
1979.
14.
Matthews, L. W.,
S. Spector,
J. Lemm,
and
J. Potter.
Studies on pulmonary secretions. 1. The overall chemical composition of pulmonary secretions from patients with cystic fibrosis, bronchiectasis and laryngectomy.
Am. Rev. Respir. Dis.
88:
199-204,
1963[Medline].
15.
McNulty, M. J.,
J. C. F. Chang,
C. S. Barrow,
M. C. Schmitz,
and
H. d'A. Heck.
Sulfhydryl oxidation in rat nasal mucosal tissues after chlorine inhalation.
Toxicol. Lett.
17:
241-246,
1983[Medline].
16.
Menache, M.,
L. M. Hanna,
E. A. Gross,
S. R. Lou,
S. J. Zinreich,
D. A. Lepold,
A. M. Jarabek,
and
F. J. Miller.
Upper respiratory tract surface areas and volumes of laboratory animals and humans: considerations for dosimetry models.
J. Toxicol. Environ. Health
50:
475-506,
1997[Medline].
17.
Mitra, S.,
W. Li,
and
B. Kebbekus.
Evaluation of a thermionic ionization detector for selective detection of oxygenated volatile organic compounds.
J. Chromatogr. Sci.
33:
405-409,
1995.
18.
National Research Council, Committee on Medical, and Biological Effects of Environmental Pollutants.
Chlorine and Hydrogen Chloride. Washington, DC: National Academy of Sciences, 1976, p. 116.
19.
Nodelman, V.,
A. Ben-Jebria,
and
J. S. Ultman.
Fast-responding thermionic chlorine analyzer for respiratory applications.
Rev. Sci. Instrum.
69:
3978-3983,
1998.
20.
Nuckols, M. L.
Heat and Water Vapor Transfer in the Human Respiratory System at Hyperbaric Conditions (PhD dissertation). Durham, NC: Duke University, 1981.
21.
Olson, D. E.,
M. F. Sudlow,
K. Horsfield,
and
G. F. Filley.
Convective patterns of flow during inspiration.
Arch. Intern. Med.
131:
51-57,
1973[Medline].
22.
Rotman, H. H.,
M. J. Fliegelman,
T. Moore,
R. G. Smith,
D. M. Anglen,
C. J. Kowalski,
and
J. G. Weg.
Effects of low concentrations of chlorine on pulmonary function in humans.
J. Appl. Physiol.
54:
1120-1124,
1983[Abstract/Free Full Text].
23.
Shaw, J. W. Indoor air quality of swimming pool
enclosures. New Zealand J. Sports Med.
55-58, 1987.
24.
Swift, D. L.,
and
D. F. Proctor.
Access of air to the respiratory tract.
In: Respiratory Defense Mechanisms, edited by J. D. Brain,
D. F. Proctor,
and L. H. Reid. New York: Dekker, 1977, vol. 5, pt. I, chapt. 3, p. 63-93. (Lung Biol. Health Dis. Ser.)
25.
Treybal, R. E.
Mass-Transfer Operations (3rd ed.). New York: McGraw-Hill, 1980, chapts. 3 and 5, p. 45-88 and 104-136.
26.
Weibel, E. R.
Morphometry of the Human Lung. New York: Academic, 1963, chapt. 11, p. 136-148.
27.
Whitney, R. P.,
and
J. E. Vivian.
Solubility of chlorine in water.
Ind. Eng. Chem.
33:
741-744,
1941.
28.
Wolf, D. C.,
K. T. Morgan,
E. A. Gross,
C. Barrow,
O. R. Moss,
R. A. James,
and
J. A. Popp.
Two-year inhalation exposure of female and male B6C3F1 mice and F344 rats to chlorine gas induces lesions confined to the nose.
Fundam. Appl. Toxicol.
24:
111-131,
1995[Medline].
29.
Wood, B. R.,
J. L. Colombo,
and
B. E. Benson.
Chlorine inhalation toxicity from vapors generated by swimming pool chlorinator tablets.
Pediatrics
79:
427-430,
1987[Abstract/Free Full Text].
J APPL PHYSIOL 86(6):1984-1993
8570-7587/99 $5.00
Copyright © 1999 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
