Wednesday, January 10, 2007

Foundation of Breath Testing Cannot be True

The Impact of Breathing Pattern and Lung Size on the Alcohol Breath
Test

MICHAEL P. HLASTALA
1,2,3 and JOSEPH C. ANDERSON4

1Department of Physiology and Biophysics, University of Washington, Box 356522, Seattle, WA 98195-6522, USA; 2Department
of Medicine, University of Washington, Box 356522, Seattle, WA 98195-6522, USA; 3Division of Pulmonary and Critical Care
Medicine, University of Washington, Box 356522, Seattle, WA 98195-6522, USA; and 4Department of Bioengineering,
University of Washington, Seattle, WA 98195-5061, USA
(Received 21 March 2006; accepted 29 September 2006)

Abstract—Highly soluble gases exchange primarily with the
bronchial circulation through pulmonary airway tissue.
Because of this airway exchange, the assumption that
end-exhaled alcohol concentration (EEAC) is equal to
alveolar alcohol concentration (AAC) cannot be true.
During exhalation, breath alcohol concentration (BrAC)
decreases due to uptake of ethanol by the airway tissue. It
is therefore impossible to deliver alveolar gas to the mouth
during a single exhalation without losing alcohol to the
airway mucosa. A consequence of airway alcohol exchange
is that EEAC is always less than AAC. In this study, we
use a mathematical model of the human lung to determine
the influence of subject lung size on the relative reduction
of BrAC from AAC. We find that failure to inspire a
full inspiration reduces the BrAC at full exhalation, but
increases the BrAC at minimum exhalation. In addition, a
reduced inhaled volume and can lead to an inability to
provide an adequate breath volume. We conclude that
alcohol exchange with the airways during the singleexhalation
breath test is dependent on lung size of the
subject with a bias against subjects with smaller lung
size.

Keywords—Ethyl alcohol, Ethanol, Bronchial circulation,
Airway gas exchange.

INTRODUCTION

An assumption used in the development of the
alcohol breath test (ABT) is that the ethanol concentration
in the last part of the exhaled breath is equal to
that in the alveolar gas. This long-held assumption is
the basis for justifying the ABT1 as an accurate
measure of blood alcohol concentration (BAC).
However, under normal circumstances, a singleexhalation
alcohol breath test shows a gradually and
continually increasing breath alcohol concentration
(BrAC) if the subject exhales at a constant rate
(Fig. 1). The end-exhaled alcohol concentration
(EEAC) is always lower than the alveolar alcohol
concentration (AAC). As more volume is exhaled the
BrAC continues to increase. It has recently been shown
that EEAC is less than AAC due to the exchange of
alcohol in the airways during both inspiration and
expiration.2,3,8
Earlier studies have examined the assumption of
equality between end-exhaled and AAC by comparing
ABT values with blood measurements and found a
considerable amount of variation in the ratio of EEAC
to BAC. For further evidence regarding the lack of
end-exhaled and alveolar equality, two studies10,13
have shown that EEAC is approximately 15–20%
lower than AAC on average (obtained using isothermal
rebreathing). The explanation for this variation
has been discussed before.2,8 The physiological
importance of the discrepancy between EEAC and
AAC are the subject of this study.
Two recent studies have demonstrated a relationship
between the blood:breath2 ratio (BBR) for alcohol and
body weight14 or gender11 in normal subjects. Thus, it
may be possible that the BBR for alcohol is dependent
on physiological or anatomic differences among individual
subjects.9 One anatomical feature, lung size,
depends on body size, age, gender and ethnicity.
When an ABT is performed, subjects are not
required to control either the volume inhaled or the
Address correspondence to Michael P. Hlastala, Division of
Pulmonary and Critical Care Medicine, University of Washington,
Box 356522, Seattle, WA 98195-6522, USA. Electronic mail: hlastala@
u.washington.edu
1 A list of abbreviations used in this paper is shown in Table 1.
2 The blood:breath ratio is equal to the ratio of end-exhaled alcohol
concentration divided by blood alcohol concentration (EEAC/BAC).
Annals of Biomedical Engineering ( 2006)
DOI: 10.1007/s10439-006-9216-3
 2006 Biomedical Engineering Society
volume exhaled. Under normal resting conditions, a
subject inhales and exhales a tidal volume (VT)
beginning from a functional residual capacity (FRC)
(Fig. 2). When administering an ABT, the subject is
asked to inhale ambient air and exhale into the breath
test instrument as far as possible. Although the subject
is asked to take a full inhalation, he/she is not required
to inhale to total lung capacity (TLC). Because it takes
some effort to inhale from FRC to TLC, a volume
known as inspiratory capacity (IC), it is most likely
that a subject’s lung size is less than TLC at the time
exhalation is initiated (gray line in Fig. 2). Some subjects
may exhale after inhaling only a very small volume.
The expiratory volume also varies naturally
between tests. To obtain a valid ABT, a subject can
exhale any amount between the minimum exhaled
volume required by the particular breath test instrument
(usually either 1.1 or 1.5 l)5 and the maximum
exhaled volume of the lungs, which is limited by the
vital capacity (VC), the difference between TLC and
residual volume (RV). The exhaled volume depends on
the mechanical limitations of the lungs and the relative
effort of the subject, which may vary from time to
time. For the calculations below, we assume that an
average exhaled volume is the average of the minimum
volume and the VC.
Lung volume varies substantially among individual
human subjects (both normal and with lung disease). In
1991, the American Thoracic Society (ATS) compiled
data from three international societies (the ATS, the
European Community for Coal and Steel, and the
European Society for Clinical Respiratory Research)
and published a summary document of lung volumes in
normal, non-smoking, human subjects for clinical use
in interpretation of pulmonary function tests.1 Collectively,
the summary of data (Table 2) shows that, in
adults, lung volumes increases with body height and
decreases with age. Lung volumes are smaller in African
Americans, both males and females, than their Caucasian
height-, age-, and gender- matched counterparts.
For either racial group, females have smaller vital
capacities than males. Because individuals with smaller
lung size must exhale a greater fraction of their lung
volume to fulfill any minimum volume requirement for
a valid sample, we reasoned that a subject with a smaller
lung volume would exhale farther along the increasing
exhaled partial pressure profile before an end-exhaled
sample is taken (see Fig. 3). Consequently, the alcohol
breath test would tend to overpredict the BAC for
individuals with small lung volumes.
We use a mathematical model2 to explore the
dependence of BrAC on lung size (a function of height,
age, gender, and race), inspiratory volume, and expiratory
volume. We hypothesize that BBR will depend
on the subject physical characteristics as well as the
level of cooperation.
FIGURE 2. Lung volume tracing for a single exhalation
maneuver. A subject breathes tidal volumes (VT) at functional
residual capacity (FRC) and then expands his lungs to total
lung capacity (TLC) by inhaling a volume equal to the inspiratory
capacity (IC). The subject exhales his vital capacity (VC)
at a constant flow rate, which causes his lung volume to
approach residual volume (RV). The gray tracing shows the
lung volume dimensions if the subject only inhales 50% of IC
during the prolonged inhalation.
BrAC
AAC
0.2
0.4
0.6
0.8
1.0
I
III
II
0 1 3 4 5
Exhaled Volume (Liters)
0.0
2
FIGURE 1. Exhaled ethanol concentration, normalized by
alveolar alcohol concentration, over a full exhalation at a
constant flow (From4).
TABLE 1. Glossary of abbreviations.
AAC Alveolar alcohol concentration
ABT Alcohol breath test
ATS American Thoracic Society
BAC Blood alcohol concentration
BBR Blood:breath ratio
BrAC Breath alcohol concentration
EEAC End-exhaled alcohol concentration
FRC Functional residual capacity
IC Inspiratory capacity
RR Respiratory rate
RV Residual volume
TLC Total lung capacity
VC Vital capacity
VI Volume of inspiration
VT Tidal volume
M.P. HLASTALA AND J.C. ANDERSON
METHODS
Mathematical Model
A detailed description of the model has been published
previously.2,4,15 Only the essential features will
be described here. The airway tree has a symmetric
bifurcating structure through 18 generations. The
respiratory bronchioles and alveoli are lumped
together into a single well-mixed alveolar unit. Axially,
the airways are divided into 480 control volumes.
TABLE 2. Predicted forced vital capacity for healthy, Non-smoking subjects: Caucasian and African American, male and female.
Predicted vital capacity (l)
Caucasian African-American
Height (in) Height (m) Age (Year) Male Female Male Female
51 1.30 20 2.587 2.137 2.866 2.244
51 1.30 40 2.195 1.721 2.430 1.810
51 1.30 60 1.803 1.305 1.994 1.376
55 1.40 20 3.178 2.560 3.191 2.541
55 1.40 40 2.786 2.144 2.755 2.107
55 1.40 60 2.394 1.728 2.319 1.673
59 1.50 20 3.770 2.984 3.517 2.838
59 1.50 40 3.378 2.568 3.081 2.404
59 1.50 60 2.986 2.152 2.645 1.970
63 1.60 20 4.361 3.407 3.842 3.135
63 1.60 40 3.969 2.991 3.406 2.701
63 1.60 60 3.577 2.575 2.970 2.267
67 1.70 20 4.952 3.830 4.167 3.432
67 1.70 40 4.560 3.414 3.731 2.998
67 1.70 60 4.168 2.998 3.295 2.564
71 1.80 20 5.544 4.254 4.493 3.729
71 1.80 40 5.152 3.838 4.057 3.295
71 1.80 60 4.760 3.422 3.621 2.861
75 1.90 20 6.135 4.677 4.818 4.026
75 1.90 40 5.743 4.261 4.382 3.592
75 1.90 60 5.351 3.845 3.946 3.158
79 2.00 20 6.727 5.100 5.144 4.323
79 2.00 40 6.335 4.684 4.708 3.889
79 2.00 60 5.943 4.268 4.272 3.455
FIGURE 3. Effect of lung size (as represented by vital capacity) on the exhalation profile. At a given exhaled volume (e.g., 1.5 l),
BrAC/AAC is inversely related to lung size. The model simulated a lung performing an IC inhalation (IC = 0.75 Æ VC) and a VC
exhalation at a rate of 200 ml s)1. The horizontal solid bars indicate the end-exhaled normalized BrAC at an average exhaled
volume. The relative average end-exhaled breath to alveolar concentration ratios are 0.767, 0.722 and 0.705 for subject vital
capacities of 2.0, 4.0, and 6.0 l, respectively.
Single-Exhalation Alcohol Breath Test
Radially, the airways are divided into six concentric
layers: (1) the airway lumen, (2) a thin mucous layer,
(3) connective tissue (epithelium and mucosal tissue),
(4) the bronchial circulation, (5) the adventitia, and (6)
the pulmonary circulation. Functionally, the upper
respiratory tract and cartilaginous airways (generation<
10) only have the first four layers. Within each
radial layer, concentration and temperature values are
bulk averages for the entire layer. Mass and energy are
transported between lumenal control volumes by bulk
convection and axial diffusion. Radial transport
between the gas phase and mucous layer is described
with heat and mass transfer coefficients. Radial
transport of water and soluble gas between concentric
layers occurs via filtration (from bronchial circulation
to mucus) and diffusion (Fick’s law). In the alveolar
unit, the concentration of soluble gas is allowed to vary
with time and depends on the pulmonary blood flow,
ventilation, blood solubility, and concentration of
soluble gas in the incoming blood as described by a
mass balance on the alveolar compartment.
Because airway volume increases with increasing
lung size, the lengths and diameter of the intraparenchymal
airways were scaled to ensure the ratio of the
airway volume to the VC was constant. Since the VC
of the Weibel lung model is 5000 ml, these dimensions
were scaled by the factor (VC/5000)1/3. None of
these airway dimensions changed dynamically during
the breathing cycle. The dimensions of the airway wall
compartments were calculated using data and a
method outlined previously.2
Mass and energy balances around a control volume
produce three partial-differential in time, t, and space,
z and nine ordinary differential equations. The equations
are solved simultaneously for the following 12
dependent variables: the mole fraction of soluble gas in
the air, mucous, connective tissue, bronchial bed, and
adventitial tissue layers; the temperature of the air,
mucous, connective tissue, bronchial bed, and adventitial
tissue layers; the mole fraction of water in the air;
and the mucous thickness. The 12 differential equations
are solved numerically using previously published
boundary conditions.2 The spatial derivatives are
approximated by upwind finite difference while the
time derivatives are solved using LSODE, an integrating
software package developed by Hindmarsh.7
Computer Simulations
Before an ABT was simulated, the model first must
reach breath-to-breath steady-state conditions. The
temperature, water concentrations, and ethanol concentrations
within the mathematical model were
brought to steady-state conditions by simulating tidal
breathing at FRC. A respiratory rate of 12 br min)1, a
sinusoidal flow waveform, and a tidal volume equal to
10% of VC were used for the case study (Table 2). For
the parameter study, tidal volume was varied between
200 and 600 ml in 100 ml increments. The inspired air
temperature and relative humidity were set to 23C
and 50%, respectively. The bronchial blood flow rate
was set to 1 ml s)1. The concentration of ethanol in the
pulmonary arterial blood was constant and equal to
0.10 g dl)1 of blood. Steady-state conditions were
reached when the end-exhaled water and ethanol
concentrations changed by less than 0.1% between
breaths. Then, the model simulated a single inhalation
of a volume equal to or a fraction of IC, the volume
from FRC to TLC, at a constant rate of 1500 ml s)1.
Inspiratory capacity was approximated to be 75% of
the VC.6 Then, the model simulated a prolonged
exhalation; the lung was emptied at a rate of
200 ml s)1 until the lung volume reached RV.
RESULTS
For highly soluble gas like ethyl alcohol, exhaled
concentration continues to increase with continued
exhalation due to airway gas exchange. An example of
an exhaled ethyl alcohol profile is shown in Fig. 1. In
this example, a male subject with a BAC  0.09 g/dl
inhaled quickly to TLC, exhaled at a constant flow
rate, and stopped exhalation at RV.4 Several different
expiratory profiles for the same subject are shown.
During exhalation at a constant exhaled flow rate, the
exhaled ethanol concentration rises continuously during
the final phase (phase III) of the ethanol profile.
When the subject stops exhalation (either due to
reaching RV or simply because the subject chooses to
stop), the alcohol concentration plotted against time
levels off because exhalation has stopped and no new
air enters the breath test machine.8 At this time, a
sample is taken and assumed to be ‘‘alveolar’’ in nature.
However, any breath sample is ‘‘always’’ lower in
alcohol concentration than AAC. The classical interpretation
assumes that the EEAC is related to the BAC
with an average BBR of 2100. This factor neglects the
exchange of alcohol with the airways of the lungs and
any variability in this ratio among individuals.
From the model’s predictions of exhaled ethanol
profiles from human subjects,4 we can describe the
mechanisms underlying ethanol exchange in the airways.
As fresh air is inhaled, it absorbs ethanol from
the mucous layer, thereby depleting the ethanol concentration
in the airway wall. Because of the small
bronchial blood flow (Qbr) and the significant diffusion
barrier between the bronchial circulation and mucous
layer, the mucus is not replenished with ethanol before
M.P. HLASTALA AND J.C. ANDERSON
exhalation begins. During exhalation, respired air
encounters a lower concentration of ethanol in the
mucus and, therefore, a large driving force for the
deposition of ethanol onto the mucus. This large airto-
mucus gradient promotes recovery of ethanol by the
mucous layer, decreases the ethanol concentration in
the air, and delays the rise in ethanol concentration at
the mouth. A large (small) air-to-mucus gradient causes
a slowly (rapidly) increasing phase III slope. These
absorption–desorption phenomena decrease the ethanol
concentration leaving the lung (relative to the
alveolar concentration) throughout exhalation and
are the major mechanisms of pulmonary ethanol
exchange.
The mathematical model simulated the effect of lung
size on the exhalation profile (Fig. 3). After a steadystate
was reached during tidal breathing (RR = 12 br
min)1 and VT = 400 ml), the model simulated a full
inhalation from FRC to TLC and then a constant
(200 ml s)1) exhalation to RV. These conditions were
simulated in five lung sizes as represented by the VC
that varied from 2 l to 6 l. The normalized BrAC after
a maximum exhalation (to RV) was 0.79 for all five
lung sizes and appears to be unaffected by lung size
(i.e., VC). However, many times subjects do not exhale
their entire VC and, in addition, most alcohol breathtesting
instruments only require a minimum exhaled
volume (e.g., 1.5 l) before a breath test is acceptable.
We examined the normalized BrAC in Fig. 3 after 1.5 l
of air had been exhaled from lungs of different sizes:
small (VC = 2 l), medium (VC = 4 l) and large
(VC = 6 l). The normalized BrAC was 0.74, 0.61, and
0.55, respectively. At this exhaled volume, the ratio of
change in normalized BrAC to change in lung size is
)0.048 l)1. Additionally, we examined how lung size
affected the normalized BrAC (Fig. 3) after an average
exhalation. We assumed that, on average, an individual
would exhale a volume that is the mean of the
minimum (1.5 l) and maximum (VC) volume. Thus,
for an individual with VC = 6 l, an average exhaled
volume (after an IC inhalation) is 3.75 l and results in a
normalized BrAC of 0.705. Subjects with smaller lung
size, 4 and 2 l, and providing an average exhalation
have normalized BrAC of 0.722 and 0.767, respectively.
For an average exhalation, individuals with
smaller lung size provide BrAC samples that are
greater than those with larger lung size because of the
minimum exhalation volume requirement in combination
with the mechanics of airway gas exchange. The
effect of lung size on this average BrAC is )0.015 l)1.
Thus, a one liter increase in VC decreases the normalized
BrAC at this average volume by 0.015.
The minimum, average, and maximum BrAC values
for subjects with different vital capacities are shown in
Fig. 4. Results are shown for vital capacities varying
between 2.0 and 7.0 l and for an inspiration of a full
IC. As lung VC increases, the average BrAC decreases.
For lungs with vital capacities less than 2.0 l, it is often
difficult for the subject to fulfill the mininum 1.5 l
minimum exhalation volume.
We simulated the effect of inspiratory volume on the
exhalation profile for a given lung size (Fig. 5). Once a
periodic steady-state was achieved (VT = 400 ml), the
model simulated an inhalation from FRC. The inhaled
volume depended on the simulation. For a maximum
IC inhalation, the inhaled volume was assumed to be
0.75ÆVC. Smaller inhaled volumes of 66%, 33%, and
10% of IC were simulated. After inhalation, a constant
(200 ml s)1) exhalation to RV was simulated. Figure 5
shows the effect of inhaled volume on normalized
BrAC from three lungs of varying size, VC = 2 l
(panel A), 4 l (panel B) and 6 l (panel C). For every VC
studied, a decrease in inhaled volume causes: (1) an
increase in normalized BrAC at a given exhaled volume;
(2) an increase in the normalized BrAC from a
minimum (1.5 l) and average exhalation; and (3) a
decrease in the normalized BrAC after a maximum
exhalation to RV. Specifically, a decrease in inspired
volume in a lung with VC = 4 l causes the normalized
BrAC after a minimum exhalation to increase by
0.048 l)1, the normalized BrAC after an average
exhalation to increase 0.004 l)1, and the normalized
BrAC after a maximum exhalation to decrease
0.022 l)1. These rates of change of normalized BrAC
per inspired volume are a function of VC. A two liter
increase (decrease) in VC causes these rates to decrease
(increase) by 15%. As compared with individuals with
small VC, subjects with large VC can choose from
more possible inspired volumes that will result in a
minimum exhaled volume and an acceptable breath
test. We examined the effect of tidal volume on BrAC
and found that a 100 ml increase in tidal volume
FIGURE 4. The relationship between normalized breath
alcohol concentration and lung size (based on vital capacity)
are shown for IC inhalations followed by different exhaled
volumes: maximum (VC), average and minimum (1.5 l). See
text for definitions.
Single-Exhalation Alcohol Breath Test
decreased all three measures (minimum, average, and
maximum exhalation) of normalized BrAC by 0.01.
The variation of lung volume among individuals of
differing gender, body height and age are shown in
Table 2. Typical values are presented in Table 2 for
normal Caucasian and African American male and
female adults. Lung volumes are greater in equally sized
and aged males compared with females, in Caucasians
FIGURE 5. Effect of inspiratory volume on the exhalation profile for a given lung size. At a given exhaled volume (e.g., 1.5 l),
BrAC/AAC is inversely related to volume of gas inhaled (VI). The model simulated a lung inhaling a volume, VI, from FRC and
exhaling to RV at a rate of 200 ml s)1. VC represents lung size. For each panel, VC is 2 l (panel A), 4 l (panel B), and 6 l (panel C).
M.P. HLASTALA AND J.C. ANDERSON
compared with African Americans and in younger
adults compared with older adults. Table 3 shows the
predicted BrAC normalized by AAC taken from Fig. 4.
The predictions of the mathematical model show a
greater BrAC (relative to AAC) in all cases comparing a
smaller lung volume with a larger lung volume.
DISCUSSION
Alcohol breath testing-instruments require a minimum
exhaled volume before a breath sample is taken
at the end of an exhalation. For a subject with a small
lung size, a greater fraction of the VC must be exhaled
before the sample criteria are fulfilled. Most breath test
instruments require a minimum exhalation pressure (or
flow) for a minimal duration of time (4–6 s and a
minimal exhalation volume (between 1.1 l and 1.5 l).
For our calculations, we chose 1.5 l as the minimum
exhaled volume. Once the minimum criteria are fulfilled,
a sample will be taken when the change in
exhaled alcohol partial pressure levels off (always
achievable when the exhaled flow is stopped). For a
subject with a VC of 6 l using a BAC Verifier Datamaster
(minimum volume is 1.5 l), a sample can be
obtained any where between 1.5 and 6.0 l of exhalation
because the subject may choose to stop exhalation any
where between 1.5 l and VC. For a subject with a VC
of 2 l, a sample can be obtained using a BAC Verifier
Datamaster anywhere between 1.5 and 2.0 l of exhalation.
A subject with a small lung size will proceed
further up the increasing BrAC exhaled profile before a
sample is taken (Fig. 3).
One of the fundamental assumptions of the ABT is
that during exhalation, the BrAC continues to increase
until alveolar air reaches the mouth. At this point the
BrAC levels off. This observation has been assumed to
indicate that EEAC is equal to AAC. However, breath
alcohol always increases during exhalation as air
moves out of the mouth,4 never reaching AAC. The
flatness of the slope of the exhaled alcohol profile
simply means that exhalation has stopped. It is not an
indication of alveolar air. Additional support of this
idea follows from two studies using isothermal rebreathing
in human subjects,10,13 which showed that
EEAC (with a single-exhalation maneuver) is always
less than AAC. The difference, on average, is
approximately 15%8 and consistent with the ideas
described in this paper. Individuals with smaller lung
size are predicted to have a smaller difference between
EEAC and AAC such that an individual with a smaller
lung size, would have an ABT that is greater than an
individual with a larger lung size.
The major thesis of this paper is that lung size and
breathing pattern influence the BrAC reading determined
with a breath-testing instrument. Figure 3
shows exhaled alcohol profiles for subjects taking a full
inspiration followed by a full expiration. For each lung
size (represented by VC), the end exhaled BrAC is the
same. In other words, if a subject takes a full inspiration
followed by a full exhalation, there would be no
size dependence. If these subjects were to exhale just to
the minimum volume requirement (1.5 l), the greatest
discrepancy is predicted between subjects with differing
lung size. Every thing else being equal (including
BAC), the subject with the smallest lung size would
TABLE 3. Relative BrAC comparisons.
Predicted VC (l) BrAC/AAC
Min Avg Max
55’’ vs. 75’’ – Male 40 Years
55’’ Male – 40 Years 2.786 0.681 0.747 0.794
75’’ Male – 40 Years 5.743 0.509 0.675 0.770
BrAC Ratio of small to large volume 1.34 1.11 1.03
67’’ Female vs. 67’’ Male – 40 Years
67’’ Female – 40 Years 3.414 0.629 0.723 0.785
67’’ Male – 40 Years 4.560 0.560 0.696 0.776
BrAC Ratio of small to large volume 1.12 1.04 1.01
67’’ AA Male vs. 67’’ Caucasian Male – 40 Years
67’’ AA Male – 40 Years 3.731 0.607 0.714 0.782
67’’ Caucasian Male – 40 Years 4.560 0.560 0.696 0.776
BrAC Ratio of small to large volume 1.08 1.03 1.01
75’’ Male – 60 Years vs. 20 Years
75’’ Male – 60 Years 5.544 0.516 0.678 0.771
75’’ Male – 20 Years 6.351 0.487 0.667 0.767
BrAC Ratio of small to large volume 1.06 1.02 1.00
Single-Exhalation Alcohol Breath Test
have the greatest BrAC. Table 3 summarizes this effect
by comparing the relative ratio of BrAC between two
hypothetical subjects that differ in height, gender, race,
or age. Comparing a male and female of the same
height, the female has a minimum exhalation BrAC
that is approximately 12% greater than the male.
Comparing a 55-inch tall male with a 75-inch tall male,
at minimum exhalation, the smaller male has a 34%
greater BrAC than the taller male. With a minimum
exhalation, the overestimate for the smaller lung individual
is substantial.
On the average, a subject with a valid breath test can
exhale to any point between the minimum volume and
the maximum exhalation. When the subject stops
exhaling, new breath is no longer being delivered for
analysis. Therefore, the BrAC levels off when plotted
against time. An average of the different exhalation
volumes would be approximately equal to the mean of
the volumes exhaled at 1.5 l and the maximum exhalation.
For hypothetical subjects that differ in either
their height, gender, race or age, the ratios of average
BrAC between matched subjects are shown in Column
4 of Table 3. Comparing a 67-inch tall 40-year-old
male and with a female of the same height and age, the
female has an average exhaled BrAC that is approximately
4% greater than the male. Comparing a 55-inch
tall 40-year-old male with a 75-inch tall 40-year-old
male, at average exhalation, the smaller male has an
11% greater BrAC than the taller male. Comparing a
67-inch tall 40-year-old African American male with a
67-inch tall 40-year-old Caucasian male, at average
exhalation, the African American male has a 3%
greater BrAC than the Caucasian male. Comparing a
75-inch tall 20-year-old Caucasian male with a 75-inch
tall 60-year-old Caucasian male, at average exhalation,
the African American male has a 2% greater BrAC
than the Caucasian male. With an average exhalation,
the bias for the smaller lung individual is less than the
bias predicted for the minimum exhalation. The largest
discrepancy is related to body height because of the
greatest difference in relative lung size.
The mechanism of airway gas exchange has been
described briefly above and used to explain how ethanol
exchanges in the lung.2–4 Based on this mechanism
of ethanol exchange, the effect of changes in
inspired volume on BrAC can be understood. A small
inhaled volume will reduce the ethanol concentration
in the airway mucus and tissue layers to a lesser extent
than a large inhaled volume. During exhalation, the
former case will have a smaller air-to-mucus gradient
than the latter case. A smaller gradient causes less
ethanol to be deposited to the airway surface and, as a
result, the BrAC rises more rapidly when the inhaled
volume is small than when it is large (Fig. 5). The
maximum BrAC/AAC depends on the ratio of inspiratory-
to-expiratory time, but because the flow rates
are prescribed, inhaled volumes are defined by percent
of VC and exhalation always proceeds to RV, the
maximum BrAC/AAC only depends on inhaled volume
(VI) as shown in Fig. 5.
The ability to fulfill the minimum exhalation criteria
for a breath test instrument is limited by individuals
with smaller lungs and less than full inhalations. Figure
5 illustrates the combined impact lung size and
inspiratory volume have on the ability to provide a
minimum sample volume. As the size of the individual’s
lungs decrease, it becomes more important to
inspire a greater volume before exhalation. This finding
is consistent with the observations of Jones and
Andersson12 showing the probability of failing to
provide a minimum sample is greater in females than
males. Both genders show an increased in the probability
of an insufficient sample with increasing age.
There are two recent studies that can be used to
compare with our model predictions. Ska˚ le et al.14 and
Jones and Andersson11 determined the blood–breath
ratio (or partition ratio) for several subjects (male and
female) with varying heights, ages and body weight.
Jones studied 9 male and 9 female subjects and found
average BBRs of 2553±576 for males and 2417±494
for females. Although not statistically significant, the
trend agrees with our predictions. The ratio of females
to males is 1.056. The smaller lung size females had a
5.6% greater BrAC than the males. Ska˚ le et al. studied
9 male and 15 female subjects and found that the
blood–breath ratio was dependent on body weight.
The average BBR for subjects with body weights of
50–70 kg was approximately 2250 while the BBR for
subjects with body weights of 90–100 kg was approximately
2476. The ratio is 1.10. The BrAC for the
smaller subjects was 10% greater than the larger
subjects. Neither of these two papers measured lung
VC as this was not part of their hypotheses. So we
cannot directly compare our data. However the trends
are consistent with the hypothesis put forward in this
paper that individuals with smaller lung size have
greater BrAC in comparison to the BAC3 .
The present hypothesis is consistent with published
data and with the mechanisms of pulmonary gas exchange.
We encourage future investigators to include
3 The Blood–Breath Ratio (BBR) is a commonly used term in
forensic science. Because alcohol is a very highly soluble gas, the
ratio of concentration in the blood normalized by that in the breath
is a very large number (typically around 2000). For a given Blood
Alcohol Concentration (BAC), the Breath Alcohol Concentration
(BrAC) is about 1/2000 x BAC. With smaller lung volumes, the
BrAC is greater, hence the BBR (= BAC/BrAC) is lesser. In one
case the BrAC is in the numerator (BrAC/AAC). In the other case,
the BrAC is in the denominator. So a greater BBR is the same as a
lesser BrAC/AAC.
M.P. HLASTALA AND J.C. ANDERSON
the measurement of lung VC with the measurements
of BBR in order to provide data to test our
hypothesis. Surely, if there is anatomically dependent
variation in the alcohol breath test, it is important to
make corrections for the bias of the test. Once these
data are obtained, several possible alternative solutions
can be used: appropriate corrections to the
BrAC values can be made; adjustable legal limits can
be used for individuals of differing lung size; or rebreathing
can be used to obtain a better sample of
AAC.
In conclusion, alcohol exchanges between the respired
air and the airway tissue during both inspiration
and expiration. This airway gas exchange causes the
exhaled alcohol concentration to always be less than
the AAC. A consequence of this airway exchange is
that BrAC depends on lung size and the amount of
effort provided by the subject.
ACKNOWLEDGMENTS
This work was supported, in part, by National
Institute for Biomedical Imaging and Bioengineering
Grant T32 EB001650 and by National Heart, Lung,
and Blood Institute Grants HL24163 and HL073598.
REFERENCES
1American Thoracic Society. Lung function testing: Selection
of reference values and interpretative strategies. Am.
Rev. Respir. Dis. 144:1202–1218, 1991.
2Anderson, J. C., A. L. Babb, and M. P. Hlastala. Modeling
soluble gas exchange in the airways and alveoli. Ann.
Biomed. Eng. 31:1402–1422, 2003.
3Anderson, J. C. and M. P. Hlastala. Breath tests and airway
gas exchange. Pulm. Pharmacol. Ther. in press, 2006.
4George, S. C., A. L. Babb, and M. P. Hlastala. Dynamics
of soluble gas exchange in the airways. III. Single-exhalation
breathing maneuver. J. Appl. Physiol. 75:2439–2449,
1993.
5Harding, P. Methods for breath analysis. In: Medical–Legal
Aspects of Alcohol (4th ed.), edited by Garriott J. C.
Tucson: Lawyers & Judges Publishing Co., 2003, pp. 185–
211.
6Hildebrandt, J. Structural and mechanical aspects of respiration.
In: Textbook of physiology, edited by Patton H.
D., Fuchs A. F., Hille B., Scher A. M., and Steiner R.
Philadelphia: W.B. Saunders Co., 1989, pp. 991–1011.
7Hindmarsh, A. LSODE (computer software). Laurence
Livermore Laboratory, Livermore, CA.
8Hlastala, M. P. The alcohol breath test – a review. J. Appl.
Physiol. 84:401–408, 1998.
9Hlastala, M. P. Invited editorial on ‘‘the alcohol breath
test’’. J. Appl. Physiol. 93:405–406, 2002.
10Jones, A. W. Role of rebreathing in determination of the
blood–breath ratio of expired ethanol. J. Appl. Physiol.
55:1237–1241, 1983.
11Jones, A. W. and L. Andersson. Comparison of ethanol
concentrations in venous blood and end-expired breath
during a controlled drinking study. Forensic Sci. Int.
132:18–25, 2003.
12Jones, A. W. and L. Andersson. Variability of the blood/
breath alcohol ratio in drinking drivers. J. Forensic. Sci.
41:916–921, 1996.
13Ohlsson, J., D. D. Ralph, M. A. Mandelkorn, A. L. Babb,
and M. P. Hlastala. Accurate measurement of blood alcohol
concentration with isothermal rebreathing. J. Stud.
Alcohol 51:6–13, 1990.
14Ska˚ le, A. G., L. Slørdal, G. Wethe, and J. Mørland. Blood/
breath ratio at low alcohol levels: A controlled study. Ann.
Toxicol. Analytique. XIV:41.
15Tsu, M. E., A. L. Babb, D. D. Ralph, and M. P. Hlastala.
Dynamics of heat, water, and soluble gas exchange in the
human airways: 1. A model study. Ann. Biomed. Eng.
16:547–571, 1988.

No comments: