Vol. 90, Issue 4, 1415-1423, April 2001
Single-breath washouts in a rotating stretcher
M. J.
Rodríguez-Nieto1,
G.
Peces-Barba1,
N.
González
Mangado1,
S.
Verbanck2, and
M.
Paiva3
1 Laboratorio de Fisiopatología Respiratoria,
Fundación Jiménez Díaz, Madrid, Spain;
2 Respiratory Division, Academic Hospital, Vrije Universiteit
Brussel, 1090 Brussels; and 3 Laboratoire de Physique
Biomédicale, Université Libre de Bruxelles, 1070 Brussels,
Belgium
 |
ABSTRACT |
Vital capacity
single-breath washouts using 90% O2-5% He-5%
SF6 as a test gas mixture were performed with subjects
sitting on a stool (upright) or recumbent on a stretcher (prone,
supine, lateral left, lateral right, with or without rotation at end of inhalation). On the basis of the combinations of supine and prone maneuvers, gravity-dependent contributions to N2 phase III
slope and N2 phase IV height in the supine posture were
estimated at 18% and 68%, respectively. Whereas both He and
SF6 slope decreased from supine to prone, the
SF6-He slope difference actually increased (P = 0.015). N2 phase III slopes, phase IV
heights, and cardiogenic oscillations were smallest in the prone
posture, and we observed similarities between the modifications of He
and SF6 slopes from upright to prone and from upright to
short-term microgravity. These results suggest that phase III slope is
partially due to emptying patterns of small units with different
ventilation-to-volume ratios, corresponding to acini or groups of
acini. Of all body postures under study, the prone position most
reduces the inhomogeneities of ventilation during a vital capacity
maneuver at both inter- and intraregional levels.
ventilation inhomogeneity; posture; phase III slope; phase IV
height; cardiogenic oscillations
| |
INTRODUCTION |
THE SINGLE-BREATH
WASHOUT (SBW) has been extensively used as a simple and
noninvasive test of ventilation inhomogeneity. It presents the
following features: after a vital capacity (VC) inspiration of
O2, the N2 concentration measured at the mouth
during the next expiration presents first a zero concentration (phase
I), a rapid rise (phase II), a quasi-linear sloping alveolar plateau
(phase III), and a final rise due to airway closure (phase IV). In some cases, N2 concentration decreases at the very end of
expiration (phase V). In recent years, experiments performed in
sustained microgravity (µG) have contributed to a better
understanding of the SBW. For instance, the 22% reductions in
N2 phase III slope of the VC SBW in sustained µG
suggested considerable gravity-independent ventilation inhomogeneity
(9). This observation is consistent with intra-acinar
diffusion-convection-dependent inhomogeneity (DCDI) (14)
and convection-dependent inhomogeneity (CDI) between lung units larger
than acini (7). A more elaborate phase III slope analysis
of multiple-breath washout tests, which enables the separation between
DCDI and CDI, indeed demonstrated considerable contribution from both
these components during tidal breathing in sustained µG and suggested
that the nongravitational CDI originates in part from lung units as
small as groups of acini (15).
The inhomogeneity of ventilation in the periphery of the lung
has been studied with gases of different diffusivities: SBW with He and
SF6 as tracers take advantage of a sixfold ratio of molecular diffusion coefficients: SF6-derived indexes are
sensitive to the structure of more peripheral lung regions than is He,
because the SF6 diffusion front penetrates more in the
lung. In normal subjects on the ground, the phase III slope of
SF6 is always steeper than that of He, as predicted by the
DCDI mechanism (14). If we accept that most of the slope
is due to DCDI occurring within the acini or between close acini, it
was expected that both the SF6 and He slopes would be
reduced by the same amount (same absolute change) in µG,
corresponding to the elimination of gravitational CDI. This was not the
case for He and SF6 phase III slope decreases seen in
sustained µG, neither in the VC SBW (16) nor in the first breath of the multiple-breath washout test (15). In
fact, for the VC SBW tests, the SF6-He slope difference was
no longer significantly different from zero in sustained µG, because
of a larger phase III slope decrease for SF6 than for He
(16). However, when the same tests were performed during
short periods of µG (parabolic flights), the results were quite
different, with the He slope decreasing more than the SF6
slope so that the SF6-He slope difference actually
increased (11). No clear explanation yet had been found
for these observations.
Before these experiments became possible in µG, several groups
studied the gravity-dependent effect on the VC SBW by modifying body
posture with respect to the gravity field (2-4, 16,
17). SBW tests were performed in recumbent postures (2, 3,
17) or in a head-down tilt (4, 16) with or without
changes in body posture between inhalation and exhalation. The
paradoxical He and SF6 results obtained in µG gave new
impetus to such an approach. The main question we address here is how
to compute, on ground, the gravity-dependent contribution to
ventilation inhomogeneity and how to evaluate the role of the lung
periphery. We have used the SBW, which is the traditional test of
ventilation inhomogeneity and, in particular, its phase III and IV. We
take advantage of the very different diffusivities of He and
SF6 to identify the role of the lung periphery. Because
previous studies modifying body posture suffer from several drawbacks,
we have designed a dedicated rotating stretcher system for a systematic
He and SF6 SBW study in different recumbent postures, with
the possibility of a rotation between the different postures while the
subjects are performing the test. A complete analysis of phase III and phase IV in the different postures is applied to N2, He,
and SF6 gases.
| |
MATERIAL AND METHODS |
Experimental system.
A new experimental system was developed to allow gas concentration and
volume measurement during breathing experiments performed in different
recumbent body postures around the cephalocaudal axis. It consisted of
a cylindrical cage (Fig. 1) accommodating a stretcher that could rotate around its longitudinal axis. The driving
system consisted of an electric motor (Verye Matricera, Madrid, Spain)
producing adjustable clockwise and counterclockwise rotating velocities
up to 60 rotations/min. The stretcher was equipped with a leather
holding garment covering only the anterior rib cage like a cuirass and
leather straps over the waist and ankles to keep the subject in a
steady posture for any radial position of the stretcher. The garment
exerted no pressure on the abdominal region and kept the subject
comfortably suspended (for the body postures other than supine).
A mouthpiece that connected the subject to the breathing assembly and
monitoring system was positioned in a flexible manner above the head
side of the stretcher. The mouthpiece connected to a breathing tube
from which the mass spectrometer (MGA1100, Marquette, Milwaukee, WI)
capillary continuously sampled gas away at ~1 ml/s. Beyond the
mouthpiece, an electronically operated three-way sliding valve (Model
8560, Hans Rudolph, Kansas City, MO) and a two-way nonrebreathing valve
separating inspiratory and expiratory breathing paths (Model 1410, Hans
Rudolph) amounted to a total dead space of 42 ml. All breathing paths
were incorporated in a bag-in-box system, containing inspiratory and
expiratory 40-liter meteorological balloons, a pneumotachograph (series
3700, Hans Rudolph) in its wall, and a pressure transducer
(Poch-Millas, I+D, Madrid, Spain). All of these, as well as a small
display in the subject's field of view (for flow and volume feedback), were fixed to the rotating part of the system. The connections to the
external mass spectrometer, valve control, and data acquisition system
were led through an opening at the head side of the cage, ensuring that
the wires and tubings were not subject to torsion. Data acquisition of
gas and volume data at 1,000 Hz was handled by Testpoint software
(Billerica, MA) using a 12-bit analog-to-digital card (Keithly
Metrabyte, Taunton, MA).
Protocol.
The 10 subjects participating in this study (Table
1) performed SBW tests using a VC
maneuver with 90% O2-5% He-5% SF6 as a test
gas mixture. The SBW tests were first performed as five static
maneuvers, i.e., with the subjects in the same body posture throughout
inhalation and exhalation. These postures were either sitting on a
stool (upright) or recumbent on the stretcher (prone, supine, lateral
left, lateral right). In the five static SBW maneuvers, a 3-s
end-inspiratory breath-hold was used by default. Five subjects performed an additional set of static SBW tests without end-inspiratory breath-hold (ST, SP, and PR in Table 1), and seven subjects performed an additional set of static SBW tests with a 5-s end-inspiratory breath-hold (SP and PR in Table 1).
The SBW tests were also performed as semistatic maneuvers whereby the
stretcher was rotated 1) by 180° from one posture to another between inhalation and exhalation (prone-supine, supine-prone, lateral right-lateral left, lateral left-lateral right) or
2) by 360° from one posture to the same one between
inhalation and exhalation (supine-supine, prone-prone). In the
semistatic SBW maneuvers, subjects were instructed to breath-hold at
the end of inhalation, during which rotation could be accomplished
using a 15-rpm rotation speed. For the 180 and 360° rotation
maneuvers, the subjects used a 3- and 5-s breath-hold, respectively,
and were readily positioned in the new posture before starting to exhale.
Table 1 summarizes all combinations of static and semistatic SBW
maneuvers performed by each subject. For any given maneuver, three
tests were obtained on each subject. Inspiratory and expiratory flows
were targeted around 400 ml/s.
Data analysis.
The SBW data were analyzed after all 1,000-Hz gas concentration and
volume signals had been run through a software median filter and gas
concentration had been plotted as a function of exhaled volume at a
subsampling rate of 100 Hz. The resulting plots in the
different postures are exemplified by Fig.
2, showing typical expiratory
N2 traces obtained in subject 1. The two
top panels are upright SBW, without and with the 3-s
end-inspiratory breath-hold. The remaining panels represent SBW tests
with 3-s end-inspiratory breath-hold in the four static and four
semistatic maneuvers. Of all maneuvers depicted in Fig. 2, only the SBW
tests performed upright, prone, supine, prone-supine, and supine-prone were analyzed in terms of phase III and phase IV.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 2.
Typical N2 concentration traces as a function
of expired volume (in %vital capacity) obtained in subject
1 in the different static and semistatic SBW maneuvers. In 2 top panels, the upright maneuver is represented without
end-inspiratory breath-hold (left) and with a 3-s
breath-hold (right), whereas all other maneuvers in the
panels below correspond to SBW tests with a 3-s breath-hold.
|
|
All gases were normalized and rescaled to represent 100% resident
concentration before inhalation. In this way, phase III slopes or phase
IV heights for N2, He, and SF6 could be
directly compared. Phase III slopes were determined by linear
regression between ~30 and 80% expired VC, deviating slightly from
these limits only to compensate for the possible influence of
cardiogenic oscillations. This was achieved by considering volume
limits corresponding to midamplitude of cardiogenic oscillations. After
N2, He, and SF6 phase III slopes were
determined, these were normalized by mean expired N2, He,
or SF6 concentration. Phase IV height was computed as the
concentration difference between the phase IV maximum and the phase III
slope extrapolated values for the corresponding expired volumes. Phase
IV volume was considered as the volumetric difference between the end
of expiration and the intercept of phase III and phase IV regression
lines. The cardiogenic oscillation amplitude was computed as the
average height of the oscillation maxima with respect to the phase III
regression line, considering only the cardiogenic oscillations in the
phase III portion of the N2 curves.
Unless stated otherwise, all results are expressed as means ± SE.
All pairwise comparisons are intrasubject comparisons for which we used
Wilcoxon rank tests, accepting significance at P < 0.05.
| |
RESULTS |
Table 2 shows expired VC and phase
IV volume obtained on all subjects in the upright, supine, prone,
supine-prone, and prone-supine SBW maneuvers and residual volumes
measured by body plethysmography (in the upright posture). A
significant VC decrease on the order of 250 ml was observed between
upright and recumbent postures, whereas phase IV volume remained
unaffected (P > 0.1). There was no significant
difference in VC between the four recumbent postures (P > 0.1 for all pairwise comparisons).
Figure 3 shows the comparison of the
N2 slope and SF6-He slope difference, with and
without the 3-s end-inspiratory breath-hold in upright, supine, and
prone postures, obtained on subjects 1-5. In subsequent
figures, in which static and semistatic SBW maneuvers are compared for
all 10 subjects, only SBW maneuvers including a 3-s breath-hold are
represented (in the semistatic SBW, the breath-hold is necessary to
allow the 180° rotation between inhalation and exhalation). Figure
4 summarizes the behavior of
N2 phase III slope (A) and SF6-He
slope difference (B) of the SBW tests obtained in upright,
supine, and prone, as well as in the two semistatic maneuvers,
supine-prone and prone-supine (with 3-s end-inspiratory breath-hold in
all maneuvers). Between upright and supine maneuvers, there were no
significant changes in N2 phase III slope nor in
SF6-He slope difference. By contrast, the prone maneuver
showed a significantly decreased N2 phase III slope (P = 0.009) and a significantly increased
SF6-He slope difference (P = 0.015) with
respect to supine. The semistatic prone-supine and supine-prone
maneuvers yielded N2 and SF6-He results that were not significantly different from each other (P > 0.1).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 3.
Mean ± SE values of N2 phase III slope and
SF6-He phase III slope difference in single-breath washout
(SBW) tests with and without a 3-s end-inspiratory breath-hold obtained
in subjects 1-5 (Table 1) in 3 static maneuvers
(upright, supine, and prone).
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 4.
Mean ± SE and individual values of N2 phase III slope
(A) and SF6-He phase III slope difference
(B) in SBW tests with 3-s end-inspiratory breath-hold
obtained on all subjects in 3 static (upright, supine, and prone) and 2 semistatic (supine-prone and prone-supine) maneuvers.
|
|
Figure 5 compares SF6-He
slope difference increase between upright and prone (left)
with that between ground (1 G) and µG during parabolic flight
(right) as obtained by Lauzon et al. (11). For
this particular comparison, upright and prone SBW data are considered
without end-inspiratory breath-hold, which corresponds to
subjects 1-5 (Table 1). Note the difference in left and
right y-axis scale, because in addition to the 100%
inspired gas rescaling, also used in Lauzon et al. (11),
we normalized He and SF6 slopes by the mean expired He or
SF6 concentration. In our subjects, the increase in
SF6-He slope difference between upright and prone results
from a He slope decrease from 0.042 ± 0.014 liter
1
(upright) to 0.021 ± 0.008 liter1 (prone) and a
corresponding SF6 slope decrease from 0.057 ± 0.020 liter1 (upright) to 0.046 ± 0.012 liter1 (prone).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 5.
Comparison of SF6-He slope difference obtained in the
present study (upright vs. prone) and the parabolic flight study [1 G
upright vs. microgravity (µG)] by Lauzon et al. (11).
Respective slope normalizations were only different due to the fact
that, in addition to the normalizations in Lauzon et al.
(11) (slopes in %/liter; right axis), our
slopes were divided by mean expired concentrations (slopes in
liter1; left axis).
|
|
Figure 6 shows N2 phase IV
height, which was not significantly different from phase IV height
measured on He and SF6 curves in any maneuver (not
represented in Fig. 6). Upright and supine maneuvers produced similar
phase IV heights, which were in turn significantly larger than in prone
(P = 0.02 for supine vs. prone). In the semistatic
maneuvers, phase IV was attenuated even more and also changed sign, yet
phase IV height was not significantly different between supine-prone
and prone-supine (P = 0.06). Figure 7 shows that cardiogenic oscillation
amplitude was similar in upright and supine maneuvers and significantly
larger in supine than prone (P = 0.04). Cardiogenic
oscillation amplitude was also smaller in prone-supine than in
supine-prone (P = 0.02).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 6.
Mean ± SE and individual values of N2 phase IV height
in SBW tests with 3-s end-inspiratory breath-hold obtained on all
subjects in 3 static (upright, supine, and prone) and 2 semistatic
(supine-prone and prone-supine) maneuvers.
|
|
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 7.
Mean ± SE values of cardiogenic oscillation amplitude on phase III
in SBW tests with 3-s end-inspiratory breath-hold obtained on all
subjects in 3 static (upright, supine, and prone) and 2 semistatic
(supine-prone and prone-supine) maneuvers. O1 and
O2 refer to measurements in the first and second halves of
the phase III portion of the SBW, respectively.
|
|
Table 3 summarizes the effect of
end-inspiratory rotation on phase III slope, phase IV height, and
cardiogenic oscillation amplitude by comparing supine and prone SBW
maneuvers with or without 360° rotation between in- and exhalation
during a fixed end-inspiratory breath-hold period of 5 s
(performed by seven subjects; Table 1). In the supine SBW maneuvers,
the 360° rotation induced a significant N2 phase III
slope decrease but did not affect SF6-He slope difference.
In the prone SBW maneuvers, the 360° rotation did not significantly
affect N2 phase III slope nor SF6-He slope
difference. Table 3 shows a significant decrease of N2
phase III slope and a significant increase of SF6-He slope difference between supine and prone, extending the findings of Fig. 4,
A and B (using a 3-s end-inspiratory breath-hold)
to a 5-s breath-hold in these static postures. Also, phase III slope behavior between supine and prone in this subgroup of seven subjects was mimicked by their respective behavior between supine-supine and
prone-prone. Phase IV height and cardiogenic oscillation amplitude did
not reveal any significant effect of the 360° rotation.
| |
DISCUSSION |
Typical expiratory curves obtained on one subject who performed
the 10 SBW maneuvers (Fig. 2) show that each curve is specific to a
given maneuver, including the amplitude and shape of the cardiogenic
oscillations. For the lateral decubitus maneuvers, the shape of phase
III is due to very large patterns of sequential emptying between the
dependent and nondependent lung, as was previously demonstrated
(5, 8). Frazier et al. (8) obtained SBW curves from individual lungs in awake volunteers in lateral decubitus posture, showing extreme sequencing between lungs, which makes the
interpretation of phase III and IV very difficult. Not only does the
contribution of each lung to phase III change dramatically during
expiration, but the onset of phase IV is almost always impossible to
measure. For these reasons, we decided not to measure the slope of
phase III or phase IV height in these lateral postures nor to speculate
on the amplitude and shape of the cardiogenic oscillations. The
modifications in shape and amplitude of the oscillations between supine
and prone and between lateral left and lateral right show that the role
of the heart is posture dependent. The SBW curves of Fig. 2 also
indicate that gravity plays a primary role in the shape of the curves.
Despite that, the mean slope of phase III in maneuvers with 3-s
breath-holding are remarkably insensitive to body posture, with or
without end-inspiratory rotation (Fig. 4).
The standard model for the interpretation of the gravity-dependent
slope of the alveolar plateau in any body posture is the so-called
"onion-skin" diagram described by Milic-Emili et al. (13) and the demonstration by Anthonisen et al.
(2) of gravity-dependent sequential emptying of lung
regions throughout the VC. The model of Milic-Emili et al. assumes a
fixed gradient of pleural pressure and uniform intrinsic elastic
properties of the lung. Because of the curvilinear pressure-volume (PV)
relationship, nondependent lung regions (higher up on their PV curve)
are more expanded and inflate less during a breath than dependent
regions situated on the steeper part of the PV curve. Because the
pleural pressure gradient is attributed to the weight of the lung,
gravity influences the gradient and hence the vertical difference in
expansion. Therefore, this model predicts that, in µG, vertical
distribution of inspired gas is more uniform and phase III slope is
reduced. However, only minor reductions in phase III slope were
observed during transient (12) and sustained
(9) µG. These results suggest other sources of
ventilation inhomogeneity. On the basis of experiments performed in
dogs, Engel et al. (7) have shown that small lung regions subtended by peripheral airways also manifest an inhomogeneity of gas
concentrations among units emptying sequentially (CDI), a possible
mechanism being that of differing mechanical properties of relatively
small lung units. The other mechanism is DCDI in the lung periphery
which, according to theoretical studies (14), also leads
to a sloping phase III.
Posture-dependence of phase III: Role of gravity.
The marked N2 phase III slope decreases between supine and
prone postures with or without breath-hold (Figs. 3 and 4) indicate that, at least in the supine posture, phase III slope is partially due
to emptying patterns of lung units larger than acini with different
volume-to-ventilation ratios (i.e., CDI).
Insofar as the onion-skin models obtained in supine and prone
postures are similar (10), it would be unlikely that the
difference in N2 phase III slope between static supine and
prone SBW maneuvers are due to different gravity-dependent inspired
ventilation distribution. If, on the other hand, we extrapolate to
humans the result obtained by Wiener-Kronish et al. (18)
in dogs, the vertical gradient of pleural pressure in the prone posture
is smaller than in the supine posture, and this can explain, at least
in part, the observation. In any case, the CDI contribution to
N2 phase III slope appears to be smallest in the prone posture.
The gravity-dependent contribution to ventilation inhomogeneity in the
prone and supine postures can be assessed by comparison of the static
(supine or prone) and semistatic (supine-prone or prone-supine) SBW
maneuvers. To validate this approach, we have also evaluated the
possible transient effects of the end-inspiratory rotation in the
supine-prone and prone-supine measurements. This was done by comparing
prone and supine SBW maneuvers with or without an end-inspiratory
360° rotation (Table 3). We decided to keep the same angular velocity
(15 rpm) and extended end-inspiratory breath-holding from 3 to 5 s. In fact, N2 phase III slope and SF6-He slope
difference obtained with a 5-s breath-hold (SP vs. PR in Table 3)
showed similar behavior to that obtained with a 3-s breath-hold (SP vs.
PR in Fig. 4).
The fact that a significant decrease in N2 phase III slope
was observed in the supine posture with a 360° rotation compared with
the equivalent static SBW suggests a small dynamic effect. The fact
that the SF6-He slope difference remained unaffected by the
360° rotation indicates that this transient effect is CDI in origin.
In this respect, the absence of any transient effect in the prone
experiments, even for the N2 phase III slope, is consistent
with minimal CDI inhomogeneities in this posture. Finally, because the
5-s breath-hold increases the relative importance of CDI (i.e.,
ventilation inhomogeneities between units larger than those in which 2 extra seconds allow homogenization by diffusion), the transient dynamic
effect in the supine-prone and prone-supine measurements (with 3-s
breath-hold) is expected to be even smaller. Therefore, we will neglect
it in the computation of the gravity-dependent ventilation inhomogeneities.
On the basis of the comparison between static and semistatic maneuvers
combining supine and prone postures (Fig. 4), we can now evaluate the
role of gravity on phase III slope. Indeed, if gravity were the only
factor generating phase III, we would expect almost perfect inversion
of the slopes when comparing maneuvers with or without 180° rotation
between inspiration and expiration (provided that all other conditions
such as flow, VC, and end-inspiratory breath-hold are similar in static
and semistatic maneuvers). Let us consider the mean
N2 phase III slope values depicted in Fig. 4:
0.055 liter1 for supine, 0.041 liter1 for prone, and 0.035 liter1 for
supine-prone and prone-supine (the latter value is taken as the average
of both semistatic maneuvers, because they are not significantly
different from each other). If gravity were the only mechanism present,
e.g., in the supine posture, one would expect a mean slope of 
0.055
liter1 for the supine-prone SBW maneuver, i.e., a 0.11 liter1 decrease. The actual 0.020 liter1 decrease between supine and supine-prone suggests
that the average gravity dependence of the slope in the supine posture
is only 18% (0.020/0.110). The same reasoning for prone vs.
prone-supine yields an even smaller gravity dependence in the prone
posture of 7% (0.006/0.082). The estimated 18% gravity-dependence of
the supine N2 slope is less than the 32% N2
slope decrease observed by Guy et al. (9), when the supine
condition was compared with sustained µG. Unfortunately, no
N2 phase III slope data exist in the same individuals in
the supine posture and in short-term µG.
Posture dependence of phase III: Previous SBW studies.
The role of gravity on the SBW was also previously studied on ground by
tilting the subjects. Demedts et al. (4) performed a
systematic study of regional distributions of lung volumes on six
healthy young male subjects. The authors have shown that the regional
volume distribution in the head-down posture is indeed inverted but not
in a perfect mirror image of the upright posture. Although the obtained
results could be interpreted from the action of gravity on the lung
tissue as predicted from the model of Milic-Emili et al.
(13), one must consider that, in these two body postures, the lung is asymmetric in respect to the configuration of the surrounding structures in the apicodiaphragmatic direction. Also, when
the subjects are tilted from upright to head-down posture, lung blood
volume increases and VC decreases by 13% (4).
A complementary work (17) overcame some of these problems
by studying the subjects in the four recumbent postures, with or
without body inversions, between inspiration and expiration at total
lung capacity. However, the breath-hold time at end-inspiration was
~5 s (without body inversion) and 10 s (with body inversion), so
that a significant part of the inhomogeneities generating phase III
slope may have disappeared. Indeed, the inhomogeneities in the lung
periphery, which are less gravity dependent, may have partially
disappeared during the breath-hold period, leading to an overestimation
of the role of gravity on the slope. To what extent the N2
phase III slopes in that study may have been subject to transient
effects of body inversion remains uncertain.
The fact that in Verhamme et al. (17) flows were about
half those used here (target flows ~200 ml/s instead of ~400 ml/s here) may in part explain why these authors obtained, contrary to us,
identical N2 phase III slopes in supine and prone postures. Cortese et al. (3) who had 10 subjects perform VC SBW
maneuvers in upright, supine, and prone postures, with 5-s breath-hold
but using flows between 200 and 400 ml/s, did find a significantly smaller phase III slope prone than in supine. This result is consistent with ours, despite the fact that phase III slopes in Cortese et al.
(3) were measured from 70% VC onward (as opposed to 30% VC used here). The main conclusion of previous works on the
interpretation of phase III (3, 4, 17) and our
N2 phase III data (Fig. 3) is that phase III slope is
partially due to emptying patterns of small lung units with different
volume-to-ventilation ratios, which extends previous observations in
dogs to humans (7).
Posture dependence of phase III: Role of the lung periphery.
The significant change in SF6-He phase III slope difference
between supine and prone with a 3-s breath-hold (Fig. 4) or without breath-hold (Fig. 5, left) indicates a posture-dependent
effect on the lung periphery as well. We have observed a surprising
analogy between the increase in SF6-He slope difference
from upright to prone (Fig. 5, left) and from 1 G (upright)
to short-term µG (Fig. 5, right). In both cases, this
results from a larger decrease of the He slope, whereas VC decreases by
comparable amounts: 8% in our experiments (Table 2) and 17% in
short-term µG (A. M. Lauzon, personal communication). The larger
He phase III slope decrease suggests that, in this posture, the
ventilation inhomogeneities between parallel units located around the
acinar entrance are minimal.
Previous results in sustained µG (16) had shown that the
SF6-He slope difference actually disappears, due to a
larger SF6 decrease in sustained µG. This emphasizes the
fact that the observed SF6-He behavior in the prone posture
appears to mimic SF6-He behavior in short-term rather than
sustained µG, the latter probably being related to the modifications
in the vascular compartment. In an attempt to explain the results
obtained in sustained µG, Prisk et al. (16) performed
head-down tilt SBW experiments also including tilting between
inspiration and expiration. Whereas He and SF6 slope
changes were obtained in the different maneuvers, the
SF6-He slope difference remained invariant.
Posture dependence of phase IV and cardiogenic oscillations: Role
of gravity.
From Fig. 6, we can also evaluate the role of gravity on phase IV
height in the supine posture, considering phase IV height values: 3.9%
for supine, 2.1% for prone, and 1.4% for supine-prone (the latter
value is taken as the average of both semistatic maneuvers, because
they are not significantly different from each other). Following the
same calculations as used for the phase III slopes, this yields a 68%
contribution of gravity, which compares well with the 75% decrease
observed by Guy et al. (9) in sustained µG in respect to
supine posture. Also, the absence of phase IV volume changes in the
different recumbent postures (Table 2) is consistent with the fact that
µG did not affect phase IV volume (9).
Cardiogenic oscillations are also markedly affected by gravity. Guy et
al. (9) observed a 44% reduction in the amplitude of
selected oscillations on phase III when comparing µG to 1 G standing.
This finding was consistent with the conclusion of Engel (6) in a review on the subject that the oscillations on
phase III are related to inhomogeneities of concentration between large lung units. In fact, Engel and colleagues (7) had also
shown that the oscillations present in the lung periphery and,
measured directly in dogs, are smoothed during the expiration in
the successive branch points of the airways and largely cancel out.
Therefore, the oscillations seen at the mouth reflect primarily
concentration differences between large lung units. In this respect,
the fact that the lowest cardiogenic oscillation amplitude is observed in the prone condition (Fig. 7) gives further support to minimal CDI
contribution to ventilation inhomogeneity in this posture.
Prone vs. short-term µG.
On the basis of the similarity between the SF6-He increase
between upright and prone postures, and between 1 G and short-term µG
(Fig. 5), it is tempting to speculate that prone posture is the best
ground analog for µG. The lowest N2 phase III slope, lowest phase IV height, and lowest cardiogenic oscillation amplitude in
the prone posture further support this. Also pertinent to the analogy
between prone and µG is the minimal weight of the heart on the lungs
in the prone posture, as observed by Albert and Hubmayr (1). Maybe in the postures other than prone, gravity
distorts the human lung in such a way that it induces both ventilatory inhomogeneities between parallel units with branch points situated before (CDI) and at the level of (DCDI) the diffusion front.
Anatomically, this corresponds to inhomogeneities between acini or
small clusters of acini.
In summary, we have studied the slope of phase III, height of phase IV,
and cardiogenic oscillation amplitude of SBW in several postures in an
attempt to better understand the role of gravity in these curves.
Although SBW curves show marked dependence on body posture with or
without end-inspiratory rotation, the slope of phase III is remarkably
insensitive to the gravitational field. Of all postures under study,
the least ventilation inhomogeneity in terms of phase III, phase IV,
and cardiogenic oscillations was observed in the prone posture. In
addition, the prone posture showed virtually no contribution from
gravity to the ventilation inhomogeneity reflected in its phase III
slope. More surprisingly, the SF6-He phase III slope
difference was largest in the prone position, due to a smaller He
slope, suggesting that in this posture the ventilation inhomogeneities
between parallel units corresponding to acini or groups of acini are minimal.
| |
ACKNOWLEDGEMENTS |
This study was funded by Dirección General para la
Investigación Científica y Técnica PB/941277, the
Fund for Scientific Research-Flanders (Actie Levenslijn), and the
Federal Office for Scientific Affairs (program PRODEX).
| |
FOOTNOTES |
Address for reprint requests and other correspondence: M. Paiva, Laboratoire de Physique Biomédicale, ULB
Campus Erasme, Route de Lennik 808, 1070 Brussels, Belgium (E-mail:
mpaiva{at}ulb.ac.be).
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. Section 1734 solely to indicate this fact.
Received 9 April 1999; accepted in final form 20 October 2000.
| |
REFERENCES |
1.
Albert, RK,
and
Hubmayr RD.
The prone position eliminates compression of the lungs by the heart.
Am J Respir Crit Care Med
161:
1660-1665,
2000[Abstract/Free Full Text].
2.
Anthonisen, NR,
Robertson PC,
and
Ross WRD
Gravity-dependent sequential emptying of lung regions.
J Appl Physiol
28:
589-595,
1970[Free Full Text].
3.
Cortese, DA,
Rodarte JR,
Rehder K,
and
Hyatt RE.
Effect of posture on the single-breath oxygen test in normal subjects.
J Appl Physiol
41:
474-479,
1976[Abstract/Free Full Text].
4.
Demedts, M,
Clarysse I,
Verhamme M,
Marcq M,
and
De Roo M.
Regional gas distributions and single-breath washout curves in head-down position.
J Appl Physiol
54:
361-365,
1983[Abstract/Free Full Text].
5.
Don, HF,
Ingram RH,
and
Green M.
Relationship of phase IV to closing volume in lateral body positions.
J Appl Physiol
39:
390-394,
1975[Abstract/Free Full Text].
6.
Engel, LA.
Dynamic distribution of gas flow.
In: Handbook of Physiology. The Respiratory System. Mechanics of Breathing. Bethesda, MD: Am. Physiol. Soc, 1986, sect. 3, vol. III, pt. 2, chapt. 32, p. 575-593.
7.
Engel, LA,
Utz G,
Wood LDH,
and
Macklem PT.
Ventilation distribution in anatomical lung units.
J Appl Physiol
37:
194-200,
1974[Free Full Text].
8.
Frazier, AR,
Rehder K,
Sessler AD,
Rodarte JR,
and
Hyatt RE.
Single-breath oxygen tests for individual lungs in awake man.
J Appl Physiol
40:
305-311,
1976[Abstract/Free Full Text].
9.
Guy, HJB,
Prisk GK,
Elliott AR,
Deutschman RA, III,
and
West JB.
Inhomogeneity of pulmonary ventilation during sustained microgravity as determined by single-breath washouts.
J Appl Physiol
76:
1719-1729,
1994[Abstract/Free Full Text].
10.
Kaneko, K,
Milic-Emili J,
Dolovich MB,
Dowson A,
and
Bates DV.
Regional distribution of ventilation and perfusion as a function of body position.
J Appl Physiol
21:
767-777,
1966[Free Full Text].
11.
Lauzon, AM,
Prisk GK,
Elliott AR,
Verbanck S,
Paiva M,
and
West JB.
Paradoxical helium and sulfurhexafluoride single-breath washouts in short-term versus sustained microgravity.
J Appl Physiol
82:
859-865,
1997[Abstract/Free Full Text].
12.
Michels, DB,
and
West J.
Distribution of pulmonary ventilation and perfusion during short periods of weightlessness.
J Appl Physiol
45:
987-998,
1978[Abstract/Free Full Text].
13.
Milic-Emili, J,
Henderson JAM,
Dolovich MB,
Trop D,
and
Kaneko K.
Regional distribution of inspired gas in the lung.
J Appl Physiol
21:
749-759,
1966[Free Full Text].
14.
Paiva, M,
and
Engel LA.
Gas mixing in the lung periphery.
In: Respiratory Physiology: An Analytical Approach. Edited by Chang HK and Paiva M. New York: Dekker, 1989, p. 245-276.
15.
Prisk, GK,
Elliott AR,
Guy HJB,
Verbanck S,
Paiva M,
and
West JB.
Multiple breath washin of helium and sulfur hexafluoride in sustained microgravity.
J Appl Physiol
84:
244-252,
1998[Abstract/Free Full Text].
16.
Prisk, GK,
Lauzon AM,
Verbanck S,
Elliott A,
Guy HJ,
Paiva M,
and
West JB.
Anomalous behavior of helium and sulfur hexafluoride during single breath tests in sustained microgravity.
J Appl Physiol
80:
1126-1132,
1996[Abstract/Free Full Text].
17.
Verhamme, M,
Roelandts J,
De Roo M,
and
Demedts M.
Gravity-dependence of phases III, IV, and V in single-breath washout curves.
J Appl Physiol
54:
887-895,
1983[Abstract/Free Full Text].
18.
Wiener-Kronish, JP,
Gropper MA,
and
Lai-Fook SJ.
Pleural liquid pressure in dogs measured using a rib capsule.
J Appl Physiol
59:
597-602,
1985[Abstract/Free Full Text].
J APPL PHYSIOL 90(4):1415-1423
8750-7587/01 $5.00
Copyright © 2001 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
