Articles in PresS. J Appl Physiol (October 7, 2010). doi:10.1152/japplphysiol.00220.2010
Vertical distribution of specific
ventilation in normal supine humans
measured using oxygen-enhanced
proton MRI
1
2
3
4
5
6
7
8
9
10
11
12
13
Pulmonary Imaging Laboratory,
14
Departments of 1 Medicine and 2 Radiology,
15
University of California, San Diego, La Jolla, California.
16
*
17
18
19
20
21
22
- these authors contributed equally to this work.
Running Head: Specific Ventilation Imaging in supine humans
Correspondence address: Rui Carlos Sá
23
Department of Medicine
24
University of California, San Diego
25
9500 Gilman Dr., La Jolla, CA 92093-0852, USA
26
Tel: + 1 858 822-5081, Fax: + 1 858 534-6046
27
E-mail: [email protected]
28
29
Copyright © 2010 by the American Physiological Society.
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
Rui Carlos Sá1, Matthew V. Cronin2,*, A. Cortney
Henderson1,*, Sebastiaan Holverda1, Rebecca J. Theilmann2,
Tatsuya J. Arai1, David J. Dubowitz2, Susan R. Hopkins1,2,
Richard B. Buxton2, and G. Kim Prisk1,2
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
30
Abstract
32
Specific ventilation (SV) is the ratio of fresh gas entering a lung region divided by its
33
end-expiratory volume. To quantify the vertical (gravitationally dependent) gradient of
34
SV in 8 healthy supine subjects we implemented a novel proton magnetic resonance
35
imaging (MRI) method. Oxygen is used as a contrast agent, which in solution changes the
36
T1 relaxation time in lung tissue. Thus, alterations in the MR signal resulting from the
37
regional rise in O2-concentration following a sudden change in inspired O2 reflect SV -
38
lung units with higher SV reach a new equilibrium faster than those with lower SV. We
39
acquired T1-weighted inversion recovery images of a sagittal slice of the supine right
40
lung using a 1.5 Tesla MRI system. Images were voluntarily respiratory gated at
41
functional residual capacity; 20 images were acquired breathing air, 20 breathing 100%
42
O2, and this cycle repeated 5 times. Expired tidal volume was measured simultaneously.
43
The specific ventilation maps presented an average spatial fractal dimension of 1.13 ±
44
0.03. There was a vertical gradient in SV of 0.029 ± 0.012 /cm with SV being highest in
45
the dependent lung. Dividing the lung vertically into thirds showed a statistically
46
significant difference in SV with specific ventilation of 0.42±0.14 (mean±SD), 0.29±0.10
47
and 0.24±0.08 in the dependent, intermediate and non-dependent region respectively (all
48
differences, P<0.05). This vertical gradient in SV is consistent with the known
49
gravitationally-induced deformation of the lung resulting in greater lung expansion in the
50
dependent lung with inspiration. This specific ventilation imaging technique can be used
51
to quantify regional specific ventilation in the lung using proton MRI.
52
53
54
55
Keywords: functional magnetic resonance imaging, respiration, specific ventilation,
56
gravity, oxygen-enhanced MRI.
57
2
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
31
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
58
9/16/2010
Introduction
59
60
Specific ventilation (SV) is the ratio of the volume of fresh gas (ΔV) moving into a
61
region of the lung to the end-expiratory volume (V0) of that region, SV = ΔV / V0.
62
Specific ventilation is thus a dimensionless quantity that provides a measure of how
63
efficiently a given lung region is ventilated. Regional specific ventilation is an important
64
metric from a physiological standpoint (14, 16, 23).
65
Three approaches have been used in the past to quantify the distribution of specific
67
ventilation in humans: 1) inhaling radioactive
68
information on the spatial distribution of ventilation; however the radiation dose limits
69
the applicability of this method in repeated measurements studies (14, 22); 2) using
70
multiple-breath nitrogen washouts, Lewis et al (16) described a method to estimate the
71
distribution of specific ventilation; however this method does not yield spatial
72
information; 3) more recently, magnetic resonance imaging using hyperpolarized gases
73
(129Xe and more commonly, 3He) have been used to quantify ventilation (1, 17, 21). MRI
74
measurements of hyperpolarized gas provide good spatial resolution and do not require
75
ionizing radiation, but have several disadvantages: 3He is a rare gas; 129Xe, although more
76
readily available, is soluble in blood and has anesthetic properties; and hyperpolarization
77
requires dedicated hardware and a specific 3He (or
78
different from those used for standard proton (1H) MR imaging. All of these factors limit
79
its generalization to routine use.
133
Xe gas (or other tracers) yields
129
Xe) dedicated MR imaging coil,
80
81
Using proton MRI, it has been shown that inhaled oxygen (O2) can be used as a contrast
82
agent to verify the presence or absence of ventilation (5). Oxygen is weakly
83
paramagnetic, and when in solution in lung tissues produces a measurable decrease in the
84
longitudinal relaxation time T1 of tissues, increasing the signal intensity of an
85
appropriately timed inversion recovery proton MR image. A small body of work has been
86
published on oxygen-enhanced ventilation, mainly focusing on optimizing MRI
3
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
66
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
acquisition and post processing of image data (3, 18-20, 27) and on detecting ventilatory
88
defects (25, 28, 29), reviewed recently in (26). These publications demonstrate the use of
89
O2 as a contrast agent in the lung, and have shown the ability to detect ventilation defects.
90
However, O2-enhanced MRI signal contains more information than just the presence or
91
absence of ventilation. The change in O2 concentration in lung tissues (determining the
92
local change in T1) depends on the rate of change of the O2 alveolar concentration, which
93
is a function of the regional specific ventilation (16). Thus, measurement of the regional
94
rate of change of the MRI signal allows the quantification of specific ventilation:
95
Following a change in inspired fractional oxygen content (FIO2), units that have higher
96
specific ventilation reach the new equilibrium faster than units that have a lower specific
97
ventilation. Thus the rapidity of the change in the MRI signal in a particular voxel or
98
region following a change in inspired FIO2 is a quantitative measure of the specific
99
ventilation of that portion of the lung.
100
101
Previous studies (14, 23) have shown a vertical gradient in specific ventilation, i.e.
102
specific ventilation is gravity dependent, with the dependent lung being better ventilated
103
(higher specific ventilation) than the non-dependent lung. We hypothesized that we could
104
measure specific ventilation using proton MRI by using O2 as a contrast agent. Based on
105
this approach we have mapped the distribution of specific ventilation in the lung, and as a
106
test case, its gravitationally dependent gradient.
107
Methods
108
Quantification of specific ventilation – Basis of the measurement
109
Following a sudden change in inspired fraction of O2 (FIO2), the rate of change of the
110
alveolar O2 concentration is a function of the local specific ventilation: lung units with
111
higher specific ventilation reach a new equilibrium faster than units with a lower specific
112
ventilation. Therefore, the time it takes to reach a new equilibrium is a measure of the
113
local specific ventilation. This time is reflected in the MR images. For a series of
114
inversion recovery images acquired with appropriate scanning parameters, the signal
115
intensity change observed following a change in inspired gas is determined by the change
4
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
87
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
116
in relaxation time T1, a change driven by the local amount of oxygen in solution, which is
117
a function of the local O2 partial pressure. Therefore, the time course of the regional MR
118
signal intensity resulting from a change in FIO2 reflects local specific ventilation. More
119
specifically, the rise time of the MRI signal intensity following the onset of O2
120
inhalation—the time required for the signal to change to a new equilibrium level—is
121
directly related to specific ventilation (Figure 1).
122
As an empirical measure of the rise time we used the time delay that maximizes the
124
cross-correlation of the measured signal time course from an image voxel with a square
125
wave representing the onset of increased inspired O2 (dashed lines, Figure 2A), and refer
126
to this as the correlation delay time. This approach is illustrated in Figure 2A.
127
Mathematically, the cross correlation is maximized when the response is shifted by about
128
half the rise time, so that different rise times effectively translate to different delays in the
129
cross-correlation analysis, allowing us to use the correlation delay time as an empirical
130
index of the rise time. This calculation is done for the entire time series corresponding to
131
each image voxel. (Note that dead space in the plumbing leading from the gas containers
132
to the subject introduces a true global delay, but this delay is calculated from the
133
geometry and flow rate of the delivery system, and eliminated before the cross-
134
correlation analysis.) In order to relate the correlation delay time to the specific
135
ventilation, we modeled the experiment with a simple lung unit, and simulated the signal
136
changes during the series of consecutive breaths following a sudden change in inspired
137
FIO2. The simulated data for units with different specific ventilations was then analyzed
138
with the identical cross-correlation method used for the experimental data to derive a
139
calibration curve of specific ventilation versus correlation delay time. This calibration
140
curve was used to convert the measured correlation delay times into a quantitative
141
estimate of specific ventilation for each image voxel (Figure 2B). The details of these
142
steps are described below.
143
144
145
5
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
123
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
146
Model and simulated data
147
In order to translate the MRI signal rise time into a quantification of specific ventilation,
148
we constructed a model of a lung unit, corresponding to the voxel size measured during
149
the MRI imaging (see below). The only variable of interest in the model is the rate of
150
equilibration - the rise time - and not the equilibrium values. Therefore, dead space, water
151
vapor and end tidal PCO2, factors that change the steady state equilibrium PAO2, but
152
remain largely constant during the experiment are ignored in the following description.
153
The model describes solely the temporal behavior of the O2 concentration from the air
154
breathing equilibrium value to the new 100% O2 steady state.
155
Different V / Q ratios result in different steady state oxygen concentrations, for both
157
inspired air (~range 50 – 140 mmHg, for V / Q ratios between 0.001 and 10) and 100%
158
inspired O2 (~ 550 to 600 mmHg, for the same range of V / Q ratios), (37). These steady
159
state equilibrium conditions provide adequate (and essentially similar) contrast for all
160
reasonable V / Q ratios. Further, our model is independent of the equilibrium or steady
161
state conditions - it depends solely on the rate of equilibration. Two lung units presenting
162
identical V / Q ratios and different specific ventilations are still distinguishable from each
163
other by the different rate of equilibration, with the one presenting a higher SV having a
164
faster turn-over, and thus equilibrating faster.
165
166
The two assumptions in this model are that:
167
1) Voxel dimension (1.6x1.6mm in the imaging plane, for a 15 mm thick slice,
168
volume ~ 40 mm3) is small enough such that concentrations inside the unit at end
169
expiration can be considered uniform, and therefore each voxel can be treated as a
170
single ventilatory unit;
171
2) The time interval between two consecutive acquired images (> 5 sec) is long
172
enough so that equilibrium between oxygen in the gas-phase and in solution in
173
blood and tissues is attained by end expiration;
174
6
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
156
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
175
The first of these assumptions is supported by simulation work by Paiva (30), Davidson
176
(4) and Engel (7). Using either continuous (trumpet model) or discrete (asymmetrical
177
branch point) models of the acinus, these approaches have shown that at the scale of our
178
resolution, oxygen concentration would be expected to be constant within a single
179
respiratory pathway. Simulations in asymmetrical branch point models suggest
180
differences in O2 concentration can persist among parallel respiratory units, yet within
181
each unit the O2 concentration gradient is essentially abolished, matching our assumption.
182
Subjects self gate their breathing to the 5 s interval between consecutive image
184
acquisitions. Oxygen dissolves in tissues very rapidly compared to this 5 s interval
185
between images. For instance, for a normal, room air inspired breath, it takes ~0.25 s for
186
oxygen to diffuse through a 0.5µm thick capillary wall and reach its equilibrium with
187
hemoglobin in the pulmonary capillary (36), a process that includes dissolving in tissue
188
as well as other processes. The process of simple dissolution in tissue thus is assumed to
189
be complete within this same time frame, as in our second assumption.
190
191
Each voxel is simulated as a single ventilatory unit, with end expiratory volume V0, to
192
which inspiration transiently adds ΔV during each breath (Figure 1). The initial
193
concentration in the unit at end expiration is denoted C0, and Cn (n=1, 2, 3, …) denotes
194
the concentration in the unit at end expiration after breath n.
195
196
n
denote the concentration of oxygen in the inspired gas for breath n; during the
Let C insp
197
first twenty breaths the subjects are breathing air, (inspired FIO2 = 0.21), and are then
198
changed to a gas mixture enriched in oxygen (in our experiment, FIO2=1.0).
199
200
At the end of the first breath following the switch in FIO2 the concentration in the unit is:
201
C1 =
V0 .C0 + ΔV .C1insp
V0 + ΔV
(1)
202
The same reasoning can be used to establish a recursive formula for concentration after n
203
breaths:
7
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
183
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
V0 .Cn−1 + ΔV .C n insp
V0 + ΔV
9/16/2010
204
Cn =
205
which in turn can be re-written as a function of the unit’s specific ventilation,
206
with SV =
(2)
ΔV
, as:
V0
207
208
Cn =
SV
1
C n insp
Cn−1 +
1+ SV
1+ SV
(3)
209
This equation is used to model oxygen concentration. A simulated response for an
211
individual voxel is obtained, for a given set of inspired O2 fraction breaths (dotted line in
212
figure 1). Figure 1 presents the time series of two units, one with low specific ventilation
213
(SV = 0.2) and a second with high specific ventilation (SV = 0.8).
214
215
A linear relationship between R1 = 1/T1 and FIO2 has been consistently reported (13, 33).
216
The slope and zero crossing values reported for the R1-FIO2 function (13) result in a
217
almost linear relationship between T1 and FIO2 in the range used in the present
218
experiment (FIO2 = 0.21 or 1.0).
219
Based on numerical simulations, for the inversion time of our experiment
220
(TI = 1000 msec) and the variation of T1 with inspired oxygen concentration reported by
221
Jakob et al (13), the error involved in assuming that the MR signal varies linearly with
222
oxygen concentration is < 3% for these studies. Our calibration curve is based on this
223
assumption.
224
225
Time-shifted cross correlation between the driving function and simulated
226
units
227
To determine the correlation delay time we computed the cross correlation between each
228
simulated ventilatory unit response (continuous and dashed lines, lower panel of Figure
229
1) and the inspired oxygen fraction (dotted line in the lower panel of Figure 1). The
230
inspired FIO2 curve was shifted breath-by-breath and the time shift that maximizes the
231
time-shifted crossed correlation is a measure of how fast the unit equilibrates –
8
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
210
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
232
correlation delay time (computed similarly to (32)). This is illustrated, for two simulated
233
units (SV=0.12 and SV=0.71), in the upper panel of Figure; the continuous line depicts
234
the unit response, the dashed line depicts the time-shifted inspired FIO2 curve that
235
maximizes cross correlation.
236
We simulated units with specific ventilation ranging from 0.05 to 1 in steps of 0.05, and
238
the simulated signal was analyzed using the time-shifted cross correlation. The outcome
239
of these simulations is shown in Figure 2B. This figure is a conversion tool, allowing the
240
translation of correlation delay time measured using the proton-MRI acquired time series
241
into quantitative values of ventilation, on a voxel-per-voxel basis. The figure also shows
242
that when specific ventilation is above 0.5 this analysis is incapable of determining
243
differences in specific ventilation and lumps all ventilations above that threshold. This is
244
because equilibration for these high specific ventilation units happens sufficiently fast
245
that the correlation delay time is ≤1 breath, and thus not resolvable. However, based on
246
previous studies, relatively few lung units have SV higher than 0.5 (16), and the presence
247
of lung disease is typically associated with the development of regions of reduced SV
248
(11), as opposed to increased SV, making this limitation of relatively minor significance.
249
250
Figure 2B also shows that, as expected, for a given measured correlation delay time (y
251
axis), the corresponding specific ventilation depends not only on the specific ventilation,
252
but also on an extrinsic delay resulting from plumbing volume in the inspiratory line (see
253
section specific ventilation imaging below). In essence, any volume that exists within the
254
experimental configuration between the subject’s mouth and the valve used to change
255
between gases of different FIO2 introduces a delay that will appear to artificially reduce
256
SV unless properly accounted for. This emphasizes the need to determine this delay by
257
measuring respiratory flow and inspiratory plumbing volume.
258
9
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
237
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
259
Magnetic Resonance Imaging data collection
260
Subjects
261
We studied 8 healthy subjects (3 female, 5 male subjects) in the supine posture. Table 1
262
presents subject characteristics (gender, age, height and weight) and pulmonary function
263
data (FEV1, FVC, and FEV1/FVC). The Human Subjects Research Protection program of
264
the University of California, San Diego, approved this study and subjects participated
265
after giving written informed consent.
266
MRI
268
Data was collected using a 1.5 Tesla Signa HDx TwinSpeed MRI system (General
269
Electric Medical Systems, Milwaukee, WI, USA). A single sagittal slice was selected in
270
the right lung, in order to avoid physiological noise arising from cardiac movements.
271
Slice selection was aimed at selecting the slice within the lung presenting the largest
272
anterior-posterior dimension, while avoiding major hilar vessels.
273
Specific Ventilation Imaging protocol
274
Two-dimensional T1 weighted images were acquired using an inversion recovery
275
(TI=1000 msec) single shot fast spin echo (SSFSE) sequence, with images being acquired
276
with a half Fourier acquisition (HASTE), with a 40x40cm field of view, echo time of
277
approximately 30ms, and a 15mm image slice thickness. A MRI homodyne
278
reconstruction algorithm rescaled the data to a 256x256 matrix, with each voxel thus
279
corresponding to ~ 1.6x1.6x15mm (~ 40 mm3). An inversion time (TI) of 1000 msec,
280
approximating T1 of lung tissue, ensured maximal sensitivity to changes in O2
281
concentration (3). The long repetition time used in this single shot fast spin echo
282
sequence (5 s, compared to the T1 of blood, 1.4 s) renders the O2 induced contrast
283
independent of lung density, however lung density does alter the local signal-to-noise
284
ratio. HASTE acquisition was utilized to keep the echo time short to minimize signal loss
285
due to the very short T2* observed in the lung (9, 34). Images were voluntarily respiratory
286
gated. Subjects were instructed to take a normal breath in following the noise made by
287
the MR image acquisition, and relax back to Functional Residual Capacity (FRC), at a
10
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
267
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
288
comfortable expiratory flow rate. All our subjects were comfortable with the default 5 s
289
inter-breath intervals (12 breaths per minute). Images were acquired during a short
290
(~hundreds of milliseconds), post-expiratory breath hold at FRC resulting in a relatively
291
natural respiratory maneuver, only constrained by a constant breathing rate.
292
The addition of O2 as a contrast agent followed a block design functional Magnetic
294
Resonance Imaging (fMRI) approach, typically utilized for neuroimaging studies. A lung
295
image was acquired every 5 s, with 20 images acquired with the subject inspiring air
296
(21% oxygen), and 20 images while inspiring 100% O2. This cycle was repeated 5 times,
297
and an additional 20 breaths of 100% oxygen were added at the end of the last cycle
298
making a total of 220 images (total imaging time was 18 min 20 sec). Blocks of 20
299
breaths of air and 100% oxygen were chosen to ensure an approach to full equilibration,
300
for the ranges of specific ventilation and V / Q ratios observed in normal healthy subjects
301
(16, 35), in accordance with what has been observed in prior multiple breath washout
302
studies (31). Five cycles of air-oxygen provided an acceptable signal to noise ratio, and
303
were implemented as a compromise between keeping the total acquisition time below
304
20min and an improved signal to noise ratio that would result from a longer sequence.
305
306
A face mask (Hans Rudolph, KS, USA, deadspace 73 to 113 ml, depending on mask size)
307
equipped with a non-rebreathing T-valve (deadspace 27.9 ml) was fitted to the subject.
308
One end of the T-valve was connected to the inlet, where a remote controlled 3 way
309
pneumatic sliding valve (Hans Rudolph, model 8500) allowed rapid switching between
310
room air and O2 contained in a 170 liter Douglas bag (Hans Rudolph type 6170). The
311
inspiratory path resistance on the room air and O2 circuits were matched to eliminate
312
changes in FRC following changes in inspiratory path. The outlet of the T-valve was
313
connected to a ~6 m long large bore low resistance expiratory line, leading out of the
314
scanner room, where expired tidal volume was simultaneously measured using a
315
ParvoMedics Metabolic Measurement System (ParvoMedics, Sandy, UT, USA).
316
317
The experimental configuration for controlling the inspired gas (room air, 100% oxygen)
318
introduced an extrinsic plumbing delay in the inspired signal that is determined by the
11
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
293
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
319
volume of tubing that connects the remote control valve to the subject. This volume was
320
made as small as possible (0.6 l) under constraints imposed by working in an MRI
321
environment. This delay can be computed from each subject’s tidal volume and the
322
tubing volume, by dividing the tubing volume by the tidal volume. In practice, this
323
extrinsic delay was taken into account by computing individual inspired fractional
324
n
in eq 3), taking
oxygen concentration (FIO2) time series on a breath-by-breath basis ( Cinsp
325
the tubing volume as a delay chamber of fixed volume, through which the inspired gas
326
must pass. Once this extrinsic plumbing delay is corrected for, what is left corresponds to
327
the signal intensity change over time, allowing the computation of specific ventilation as
328
described in the next section.
330
Specific Ventilation Imaging (SVI) data analysis
331
A time series of the 220 images corresponding to the SVI sequence was constructed.
332
Quality control of the acquired image was performed at this stage, and images where the
333
subject was not at FRC based on diaphragm position compared to adjacent images were
334
removed from the series and replaced by an interpolated image, constructed from the
335
preceding and following images. A region of interest was manually drawn, and the
336
subsequent analyses were restricted to the voxels inside the lung. The region of interest
337
encompassed the entire lung, yet avoided partial volume effects from regions close to the
338
chest wall and the diaphragm. Data analysis was performed on a voxel-by-voxel basis,
339
the time course of each voxel – MRI signal intensity versus time – was used to compute
340
the regional correlation delay time. All data analysis was performed using Matlab
341
(Mathworks, Natick, MA).
342
343
We computed the correlation delay time using a shifted cross correlation between the
344
inspired FIO2 and the acquired signal intensity time series. The time shifted cross
345
correlation is a fast and simple approach to implement, fitting only one parameter, the
346
correlation delay time, and is independent of the asymptotic signal intensity. By
347
removing this extra degree of freedom, the model can remain simple while capturing only
348
the time course of the transition, leaving out the physiological dead space, water vapor,
12
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
329
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
. .
CO2 concentration, and any effect of varying V/Q ratios; factors that alter the steady state
350
concentration, but not the time course of the transition. Moreover, the shifted correlation
351
delay approach acts as a smoothing low pass filter, eliminating some of the high
352
frequency noise present in our data that did not allow a direct fit of the model (eq 3). In
353
practice, the inspired FIO2 (driving function) is correlated with the time course of each
354
voxel, and this process is repeated for delayed versions of the driving function (delayed
355
by an integer number of breaths). The integer delay that maximizes the cross correlation
356
between the time shifted driving function and the actual voxel response is the correlation
357
delay time. The correlation delay time computed in this way will only retain voxels
358
whose correlation with the optimal shifted inspired fractional oxygen concentration was
359
significant (P<0.05), and the null hypothesis of no correlation rejected in these cases. In
360
voxels in which the null hypothesis was accepted, the corresponding lung voxel was
361
attributed no specific ventilation value and treated hereafter as missing data.
362
363
Figure 3 shows the time course of signal intensity (filled circles, continuous lines) for one
364
such voxel. The arbitrarily chosen voxel plotted was located in the dependent portion of
365
the lung. The corresponding driving function representing FIO2 is also shown (dotted
366
line). The delay between the driving function and the measured signal intensity is a
367
measure of the correlation delay time for that voxel.
368
369
370
Statistics
371
Each series of 220 breaths was considered as a single measure of specific ventilation,
372
computed as described above, creating for each subject one map of specific ventilation.
373
All the voxels presenting a statistically significant (p<0.05) correlation with the optimally
374
shifted drive function were included in the analysis. The remaining voxels were treated as
375
missing data. Two different analyses were implemented: a comparison of specific
376
ventilation by lung thirds, and a finer analysis, on a cm-per-cm basis moving up the
377
supine lung.
13
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
349
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
378
For the lung thirds analysis, specific ventilation was partitioned into three gravitational
379
regions, corresponding to thirds of the lung based on equal vertical extent: the dependent
380
portion, the intermediate region and the non-dependent region. The data was reduced to a
381
subject-by-subject average specific ventilation per lung gravitational region. One-way
382
repeated measures ANOVA was used to compare the gravitational gradient in specific
383
ventilation across the lung regions (3 levels: dependent, intermediate, non-dependent).
384
Where overall significance was present, post-hoc testing was conducted using Student’s
385
t-test, to determine where this significance occurred.
386
In a second analysis, linear regression was used to evaluate the linear relationship
388
between the vertical height of the lung as a function of specific ventilation, by dividing
389
the data into isogravitational slices of 1 cm thickness (~ 6 vertical voxels). As subjects
390
were studied in the supine position, the vertical height of the lung (isogravitational level)
391
was measured along the anterior-posterior axis, with zero height corresponding to the
392
most dependent isogravitational voxels. The linear relationships were evaluated
393
individually for each subject. The inter-subject averaged specific ventilation versus lung
394
height values correspond to averages over vertical 1cm regions for the 8 subjects.
395
Different subjects have different anterio-posterior lung dimensions; therefore, results
396
were reported only for lung heights that included data from all 8 subjects. The slope of
397
the individual relationships between height and specific ventilation were compared to a
398
zero slope using a one-group t-test.
399
400
All data are presented as mean ± SD. When data for the 8 subjects were averaged, SD
401
corresponds to the inter-subject variability (SD). When calculating relative dispersion,
402
spatial SD refers to the inter-voxel variability within a specific ventilation map. Spatial
403
fractal dimension, a scale independent index of spatial heterogeneity, was also computed
404
for each specific ventilation map (8). The null hypothesis (no effect) was rejected when
405
P<0.05, two tailed, except where otherwise indicated. All statistical analyses were
406
performed using Prism (GraphPad, San Diego, CA).
407
14
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
387
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
Results
409
General data
410
Subject descriptive data and pulmonary function measurements are presented in Table 1.
411
The subjects studied had normal spirometry, as indicated by an average FEV1=100±10
412
%-predicted, FVC=99±9 %-predicted and FEV1/FVC=101±5 %-predicted (Table 1).
413
Subjects maintained a relatively constant expired tidal volume throughout the
414
experiment, averaging 0.87±0.18 l. Heart rate during SVI data acquisition averaged 65±9
415
beats/min (no difference between inspiring air and 100% oxygen) and arterial oxygen
416
saturation measured by pulse oximetry was 97.4±1.7% while breathing room air, and
417
increased to 97.8±1.5% during inspiration of 100% oxygen. The selected breathing rate
418
was followed by all subjects and tidal volume was relatively stable throughout the 220
419
breaths (the average CV for the eight subjects was 16%). Tidal volume did not change
420
when comparing breaths taken while inspiring air and oxygen (average over the 8
421
subjects, 0.86 ± 0.15 Liters on air, 0.87 ± 0.17 Liters, 2-tailed t-test paired comparison,
422
p=0.82).
423
424
Specific Ventilation Imaging (SVI)
425
Figure 3 presents the time series of a voxel (continuous line) together with the inspired
426
FIO2 (dotted line) for a voxel in the dependent lung region. Five repetitions of a block of
427
20 air and 20 100%-oxygen breaths were performed. When the inspired line was changed
428
from air to 100% oxygen, signal intensity increased, and decreased for the opposite
429
change, from oxygen back to room air. After the 5 cycles air - 100% oxygen, 20
430
additional breaths on 100% oxygen were added. Noise, expressed as the coefficient of
431
variation of the steady state signal (first 20 breaths on air, and last 20 breaths inspiring
432
O2) averaged over the entire lung of the 8 subjects studies, was 22.8 ± 1.5 % for air, and
433
18.4 ± 1.8 % for oxygen. In order to test the robustness of the algorithm we have run the
434
model with similar levels of random noise. The average specific ventilation estimation
435
error introduced by 20% noise levels is extremely small (average error in estimating SV
15
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
408
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
436
~ 0.003); the maximal SV estimation error was ~0.013, and was observed for very low
437
specific ventilation units (SV < 0.02), below the normal healthy range (16).
438
The convergence of the data processing algorithm to a solution was tested using
440
simulated data with 25% added noise, (consistent with the experimental data). The data
441
processing algorithm estimated SV using an increasing number of air-oxygen cycles (first
442
40, 80, 120, 160 breaths, corresponding to 1, 2, 3 and 4 cycles respectively). These
443
estimations were compared to the best estimation of SV, obtained using the entire 220
444
breaths simulated time series. When the analysis was restricted to the first 40 breaths
445
51.6 % of simulated units had converged to their best estimation. As expected, as the
446
number of cycles increased so did the number of units having converged: 90.9% for 2
447
cycles, 97.7% for 3 cycles and 98.8% when the first 160 breaths were used for the
448
analysis (4 cycles). The number of units presenting a significant correlation with the
449
driving function increased as the analysis was extended to an increasing number of
450
cycles.
451
452
From each individual voxel response, a map of the correlation delay time for all voxels
453
within the lung was computed and is presented for a representative subject in Figure 4A.
454
The arrow indicates the direction of gravity; the head is located to the right of the image,
455
and the diaphragm to the left. In Figure 4A, warmer colors represent portions of the lung
456
with a longer correlation delay time. The extrinsic delay resulting from the inspiratory
457
plumbing was accounted for prior to the analysis. Using the results presented in Figure
458
2B, this correlation delay time was then translated into a quantitative measure of specific
459
ventilation (Figure 4B). In this figure, warmer colors represent regions of the lung with
460
higher specific ventilation. Average specific ventilation in a slice of the right lung,
461
averaged over all subjects, was 0.33 ± 0.11. Overall specific ventilation heterogeneity, as
462
measured by the relative dispersion averaged 0.63 ± 0.11 (individual range 0.50 to 0.79),
463
comparable to that previously reported for perfusion (12). The average spatial fractal
464
dimension of the specific ventilation maps was 1.13 ± 0.03 (individual range 1.09 to 1.16,
465
table 1), similar to that derived for perfusion maps (15).
466
16
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
439
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
467
On average (over all subjects), 3.7% of the lung failed to pass the null hypothesis test in
468
the shifted correlation (individual range 0.1 to 17.8%, only a single subject had >5% of
469
voxels), showing no statistically significant changes with changes in inspired FIO2. These
470
voxels were excluded from the subsequent analysis.
471
In Figure 4B a clear vertical (gravitational) gradient of specific ventilation is present.
473
Units in the dependent portions of the lung have higher specific ventilation than those in
474
the non-dependent portions of the lung. This relationship is quantified in Figure 5, where
475
results for the regional dependence of specific ventilation over the 8 subjects brings out
476
such a relationship: a statistically significant difference in specific ventilation was
477
observed, with the most dependent third of the lung having a specific ventilation of
478
0.42±0.14, the intermediate third 0.29±0.10 and the non-dependent third 0.24±0.08 (all
479
differences, P<0.05).
480
481
An approximately linear decrease in specific ventilation with height was observed, as
482
evidenced in Figure 6, where specific ventilation is plotted against height, for one subject
483
(top panel), for the 8 individual subjects (center panel). The average over all subjects is
484
presented in the bottom panel. The inverse of the slope of this line is a measure of the
485
decrease in specific ventilation per cm. Specific ventilation was found to decrease
486
0.029 ± 0.012 /cm (average slope over the 8 individual subjects). Subject-by-subject
487
fitted slopes, reported in Table 1, showed a decrease in specific ventilation with
488
increasing height ranging from 0.015 to 0.046/cm. The slope of the height – specific
489
ventilation relationship was significantly different from a zero slope (P=0.0002). An
490
exception to the linearity is present in the most dependent portions of the lung
491
(height<1.0cm). This most dependent portion of the lung had, on average, lower specific
492
ventilation than the layers immediately above them, however this difference failed to
493
reach statistical significance.
494
495
Two independent specific ventilation maps of subject S1 were obtained, 20 days apart. A
496
Bland-Altman analysis on the cm per cm specific ventilation data was performed; the
17
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
472
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
average difference between the two runs was found to be 3.0 %, with a 95 % confidence
498
interval of ± 9.5%.
499
Discussion
500
501
This study shows that using O2-enhanced proton MR imaging, regional quantification of
502
specific ventilation is possible, extracting more information from the method than the
503
mere absence or presence of ventilation. From a physiological standpoint, specific
504
ventilation imaging shows a clear gravitational gradient with higher values seen in
505
dependent lung as expected. The spring like self-deformation of the lung (23), the Slinky
506
effect (12, 23), is likely the main determinant of the gravitational gradient observed in
507
specific ventilation. However, in the most dependent third of the supine lung there are
508
deviations from this overall behavior suggesting important influences of other factors
509
such as the non-linearity of the pressure-volume curve of the lung, the large vessels in the
510
lung, and dependent airways closure.
511
512
Vertical gradient in specific ventilation
513
The measurements show a clear, and for the most part, linear vertical gradient in specific
514
ventilation (Figures 4, 6). As expected, dependent portions of the lung have a higher
515
specific ventilation that the non-dependent portions (Figure 5). Because the dependent
516
portion of the lung partially supports the weight of the upper portions, it is more
517
“compressed”, and therefore its end-expiratory volume is smaller. Further, it is subject to
518
a higher (less negative) pleural pressure than an equivalent non-dependent region,
519
therefore placing it on a steeper portion of the pressure-volume curve. Therefore, a given
520
change in pleural pressure resulting from an inspiratory effort will result in a greater
521
increase in volume. Both the lower initial (end-expiratory) volume and the increased
522
change in volume would be expected to contribute to the higher specific ventilation
523
observed in the dependent regions of the lung, when compared to the non-dependent (36).
524
18
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
497
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
525
Comparison with published specific ventilation distributions
526
Kaneko et al (14), using a
527
ventilation for 3 supine subjects (normal healthy subject of similar age to the group
528
studied here) of: 0.020/cm, 0.022/cm and 0.026/cm, in close accordance with the
529
0.029/cm measured using our Specific Ventilation Imaging (SVI) technique.
133
Xe technique, reported a vertical gradient of specific
530
Using two different krypton isotopes, Amis et al (2) reported a vertical gradient of
532
specific ventilation in 3 supine subjects, and found a ~45% decrease from the most
533
dependent part of the lung to the most non-dependent portion, in close accordance with
534
the ~53% decrease found in our data. Using PET, Musch et al (24) reported a specific
535
ventilation gradient relative to the mean, in the supine posture, of -4.5 ± 2.5 %, slightly
536
lower than the -8.4 ± 1.9 % (range 5.3 to 10.2) we observed in our group. The breathing
537
pattern used for image acquisition in this particular study is not directly comparable to the
538
one presented here, for it starts with a 40s breath hold, followed by ~160s of free
539
breathing.
540
541
Prisk et al (31) reported an average SV of 0.364 ± 0.028, in a population of 4 supine
542
subjects somewhat older than those studied here (average age 43 compared to 33 in this
543
study), and with tidal volume constrained to ~650 ml. This compares with an average of
544
0.33±0.11 in our subject group (individual range 0.19 to 0.47) and suggests a close
545
agreement between the techniques. Comparison with upright subjects is inappropriate
546
since there are large changes in FRC with posture (6), which serves to alter SV.
547
548
Possible effects of airways closure
549
The approximately linear increase in specific ventilation with decreasing lung height is
550
not always seen in the most dependent portion of the lung (Figure 6C). Airway closure
551
occurs when small airways, thought to be respiratory bronchioles, collapse, resulting in
552
trapped gas in the alveoli distal to that point. Trapped gas means higher initial volumes,
553
and also lower volumes of inspired air, both effects contributing to a decrease in specific
554
ventilation (36). We observed specific ventilation patterns consistent with airway closure
19
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
531
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
555
in 4 of the 8 subjects imaged (subjects 3, 4, 7 and to a lesser extent 5). For these subjects,
556
specific ventilation in the most dependent part of the lung (height <1cm) was lower than
557
in the subsequent 1-2cm-height slice.
558
Assumptions and limitations
560
The technique is currently limited to a single lung imaging slice. For each subject, we
561
selected a mid-lung, sagittal slice in the right lung that presented a low percentage of
562
large vessels. Large pulmonary veins transport to the imaging plane changes in O2
563
concentration that occurred in distant regions of the lung. In different lung slices or
564
imaging planes, this can be a confounding effect, addressable by masking of large vessels
565
based on voxel intensity (10).
566
The current voxel size corresponds to rod-like shaped structures, with 1.6x1.6mm in
567
plane surface, but 15mm in the perpendicular direction. An individual voxel may contain
568
multiple adjacent acini, and as such the method measures the average response. The true
569
resolution obtained in specific ventilation imaging is somewhat lower than the nominal
570
resolution, due to small registration errors than inevitably occur from image to image,
571
resulting in some averaging of responses of neighboring in-plane voxels.
572
573
In our current implementation, we have assumed that the subject reaches a constant and
574
reproducible FRC for each acquired image and that each breath is identical (a tidal
575
breath). Based on the measured tidal volume data this was true in this subject population,
576
however this might prove not to be the case for studies in some patient populations.
577
Patients might not be able to comply with the 220 breaths sequence, and be less proficient
578
in attaining a reproducible FRC. In such circumstances, subject training, respiratory MR
579
image acquisition triggering, image registration, or an individualized approach to
580
translate correlation delay time to specific ventilation in the face of changing total
581
ventilation might be required. Misregistration resulting from patient movement or failure
582
to attain FRC at the time of imaging introduces an additional error in the measurement of
583
SV. Image registration of deformable structures such as the lung is not an easy task, yet it
584
will likely be required for the extension of the proposed method to patients suffering from
585
severe pulmonary disorders. We anticipate that the signal to noise ratio associated with
20
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
559
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
586
our measurement will be worse in patients with lung disease such as COPD. Low signal
587
to noise ratio might only impact specific regions of the lung, namely regions with lower
588
specific ventilation than that observed in normal healthy subjects; low specific ventilation
589
regions are known to be present in patient population (16). In order to capture regions of
590
lower specific ventilation, longer blocks of air – oxygen would be required; longer blocks
591
improve contrast, especially in low specific ventilation units, as they allow more time for
592
equilibration. Depending on the specific signal to noise conditions of the population and
593
their ability to cope with longer periods in the scanner, this can be achieved either by
594
reducing the number of cycles, or extending the total acquisition time.
595
Breathing 100% oxygen will result in a small increase of hemoglobin-bound oxygen (by
597
~1% in this study). The main effect of hemoglobin bound O2 is a T2* change, and a much
598
smaller effect on the T1 relaxation time, (5). Adapting the data reported by Silvennoinen
599
et al for a 1.5T field (33), a 1% arterial oxygen saturation (SaO2) increase would be
600
expected to result in a smaller than 0.5% change in the T1 of blood, a negligible effect
601
compared to the observed 12.5% change in T1 between breathing 21% O2 and 100% O2
602
(3, 13). Thus the measurement is dominated by the increase in the amount of oxygen in
603
solution in blood and tissues and not by the increase in hemoglobin-bound oxygen.
604
Conclusion
605
In conclusion, quantification of specific ventilation in the human lung is practical using
606
proton-MRI Specific Ventilation Imaging (SVI) which uses O2 as a contrast agent. As
607
expected, there was a clearly defined vertical gradient in specific ventilation in the supine
608
human lung, and this was most likely a result of the self-deformation of the lung (the
609
Slinky effect). However, in the most dependent third of the lung, other factors such as
610
dependent airways closure, the non-linear nature of the lung pressure-volume curve, and
611
the large pulmonary vessels, likely play important roles in determining regional specific
612
ventilation over and above Slinky behavior.
613
21
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
596
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
614
Grants
615
This work was supported by NIH grants HL080203 and HL081171.
616
Rui Carlos Sá’s work was supported by the National Space Biomedical Research Institute
617
through NASA NCC 9-58. A. Cortney Henderson was supported by NIH grant
618
1K99HL093064 and David Dubowitz by NIH grant NS053934.
619
620
621
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
22
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
Bibliography
1.
Albert MS, Cates GD, Driehuys B, Happer W, Saam B, Springer CSJ, and
Wishnia A. Biological magnetic resonance imaging using laser-polarized 129Xe.
Nature 370: 199-201, 1994.
2.
Amis TC, Jones HA, and Hughes JM. Effect of posture on inter-regional
distribution of pulmonary ventilation in man. Respir Physiol 56: 145-167, 1984.
3.
Chen Q, Jakob PM, Griswold MA, Levin DL, Hatabu H, and Edelman RR.
Oxygen enhanced MR ventilation imaging of the lung. MAGMA 7: 153-161, 1998.
4.
Davidson MR. Further considerations in a theoretical description of gastransport in lung airways. Bull Math Biol 43: 517-548, 1981.
5.
Edelman RR, Hatabu H, Tadamura E, Li W, and Prasad PV. Noninvasive
assessment of regional ventilation in the human lung using oxygen-enhanced
magnetic resonance imaging. Nat Med 2: 1236-1239, 1996.
6.
Elliott AR, Prisk GK, Guy HJ, and West JB. Lung volumes during sustained
microgravity on Spacelab SLS-1. J Appl Physiol 77: 2005-2014, 1994.
7.
Engel LA. Gas mixing within the acinus of the lung. J Appl Physiol 54: 609618, 1983.
8.
Glenny RW, and Robertson HT. Fractal properties of pulmonary blood flow:
characterization of spatial heterogeneity. J Appl Physiol 69: 532-545, 1990.
9.
Hatabu H, Alsop DC, Listerud J, Bonnet M, and Gefter WB. T2* and proton
density measurement of normal human lung parenchyma using submillisecond echo
time gradient echo magnetic resonance imaging. Eur J Radiol 29: 245-252, 1999.
10.
Henderson AC, Prisk GK, Levin DL, Hopkins SR, and Buxton RB.
Characterizing pulmonary blood flow distribution measured using arterial spin
labeling. NMR Biomed 22: 1025-1035, 2009.
11.
Hickam JB, Blair E, and Frayser R. An open-circuit helium method for
measuring functional residual capacity and defective intrapulmonary gas mixing. J
Clin Invest 33: 1277-1286, 1954.
12.
Hopkins SR, Henderson AC, Levin DL, Yamada K, Arai TJ, Buxton RB, and
Prisk GK. Vertical gradients in regional lung density and perfusion in the supine
human lung: the Slinky effect. J Appl Physiol 103: 240-248, 2007.
13.
Jakob PM, Wang T, Schultz G, Hebestreit H, Hebestreit A, and Hahn D.
Assessment of human pulmonary function using oxygen-enhanced T(1) imaging in
patients with cystic fibrosis. Magn Reson Med 51: 1009-1016, 2004.
14.
Kaneko K, Milic-Emili J, Dolovich MB, Dawson A, and Bates DV. Regional
distribution of ventilation and perfusion as a function of body position. J Appl Physiol
21: 767-777, 1966.
15.
Levin DL, Buxton RB, Spiess JP, Arai T, Balouch J, and Hopkins SR. Effects
of age on pulmonary perfusion heterogeneity measured by magnetic resonance
imaging. J Appl Physiol 102: 2064-2070, 2007.
16.
Lewis SM, Evans JW, and Jalowayski AA. Continuous distributions of
specific ventilation recovered from inert-gas washout. J Appl Physiol 44: 416-423,
1978.
17.
MacFall JR, Charles HC, Black RD, Middleton H, Swartz JC, Saam B,
Driehuys B, Erickson C, Happer W, Cates GD, Johnson GA, and Ravin CE. Human
23
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
9/16/2010
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
lung air spaces: potential for MR imaging with hyperpolarized He-3. Radiology 200:
553-558, 1996.
18.
Mai VM, Bankier AA, Prasad PV, Li W, Storey P, Edelman RR, and Chen Q.
MR ventilation-perfusion imaging of human lung using oxygen-enhanced and
arterial spin labeling techniques. J Magn Reson Imaging 14: 574-579, 2001.
19.
Mai VM, Chen Q, Bankier AA, and Edelman RR. Multiple inversion recovery
MR subtraction imaging of human ventilation from inhalation of room air and pure
oxygen. Magn Reson Med 43: 913-916, 2000.
20.
Mai VM, Tutton S, Prasad PV, Chen Q, Li W, Chen C, Liu B, Polzin J,
Kurucay S, and Edelman RR. Computing oxygen-enhanced ventilation maps using
correlation analysis. Magn Reson Med 49: 591-594, 2003.
21.
Middleton H, Black RD, Saam B, Cates GD, Cofer GP, Guenther R, Happer
W, Hedlund LW, Johnson GA, and Juvan K. MR imaging with hyperpolarized 3He
gas. Magn Reson Med 33: 271-275, 1995.
22.
Milic-Emili J. Radioactive xenon in the evaluation of regional lung function.
Semin Nucl Med 1: 246-262, 1971.
23.
Milic-Emili J, Henderson JA, Dolovich MB, Trop D, and Kaneko K.
Regional distribution of inspired gas in the lung. J Appl Physiol 21: 749-759, 1966.
24.
Musch G, Layfield JD, Harris RS, Vidal Melo MF, Winkler T, Callahan RJ,
Fischman AJ, and Venegas JG. Topographical distribution of pulmonary perfusion
and ventilation, assessed by PET in supine and prone humans. J Appl Physiol 93:
1841-1851, 2002.
25.
Naish JH, Parker GJ, Beatty PC, Jackson A, Young SS, Waterton JC, and
Taylor CJ. Improved quantitative dynamic regional oxygen-enhanced pulmonary
imaging using image registration. Magn Reson Med 54: 464-469, 2005.
Ohno Y, and Hatabu H. Basics concepts and clinical applications of oxygen26.
enhanced MR imaging. Eur J Radiol 64: 320-328, 2007.
27.
Ohno Y, Hatabu H, Higashino T, Kawamitsu H, Watanabe H, Takenaka D,
van Cauteren M, and Sugimura K. Centrically reordered inversion recovery halfFourier single-shot turbo spin-echo sequence: improvement of the image quality of
oxygen-enhanced MRI. Eur J Radiol 52: 200-205, 2004.
28.
Ohno Y, Hatabu H, Takenaka D, Adachi S, Van Cauteren M, and Sugimura
K. Oxygen-Enhanced MR Ventilation Imaging of the Lung: Preliminary Clinical
Experience in 25 Subjects. Am J Roentgenol 177: 185-194, 2001.
29.
Ohno Y, Hatabu H, Takenaka D, Uematsu H, Ohbayashi C, Higashino T,
Nogami M, Yoshimura M, Fujii M, and Sugimura K. Dynamic MR imaging: value of
differentiating subtypes of peripheral small adenocarcinoma of the lung. Eur J Radiol
52: 144-150, 2004.
30.
Paiva M. Gas transport in the human lung. J Appl Physiol 35: 401-410, 1973.
31.
Prisk GK, Guy HJ, Elliott AR, Paiva M, and West JB. Ventilatory
inhomogeneity determined from multiple-breath washouts during sustained
microgravity on Spacelab SLS-1. J Appl Physiol 78: 597-607, 1995.
32.
Prisk GK, Hammer J, and Newth CJ. Techniques for measurement of
thoracoabdominal asynchrony. Pediatr Pulmonol 34: 462-472, 2002.
24
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
9/16/2010
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
33.
Silvennoinen MJ, Kettunen MI, and Kauppinen RA. Effects of hematocrit
and oxygen saturation level on blood spin-lattice relaxation. Magn Reson Med 49:
568-571, 2003.
34.
Theilmann RJ, Arai TJ, Samiee A, Dubowitz DJ, Hopkins SR, Buxton RB,
and Prisk GK. Quantitative MRI measurement of lung density must account for the
change in T(2) (*) with lung inflation. J Magn Reson Imaging 30: 527-534, 2009.
35.
Wagner PD, Laravuso RB, Uhl RR, and West JB. Continuous distributions
of ventilation-perfusion ratios in normal subjects breathing air and 100 per cent O2.
J Clin Invest 54: 54-68, 1974.
36.
West JB. Respiratory Physiology - The essentials. Lippincott Williams &
Wilkins, 2008.
37.
West JB. State of the art: ventilation-perfusion relationships. Am Rev Respir
Dis 116: 919-943, 1977.
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
9/16/2010
25
727
Figure captions
728
Figure 1 – Specific ventilation of a simple unit: V0 is the end-expiratory volume of the
729
unit, and ΔV the volume increase that occurs during inspiration. C0 is the initial
730
concentration at end expiration. Specific ventilation (SV) is defined as the ratio of ΔV/V0
731
.
732
higher specific ventilation. The lower panel presents the simulated response of both units
733
to a sudden change in inspired FIO2 (dotted line). The high specific ventilation unit
734
equilibrates faster (continuous line, SV = 0.8) than the lower specific ventilation one
735
(dashed line, SV = 0.2).
The top left schematic view depicts a low specific ventilation unit; top right a unit with
737
Figure 2 – Top (A): Correlation delay time, for two simulated units with different specific
738
ventilation (SV = 0.72 and SV= 0.12, continuous lines). The vertical dotted lines (breaths
739
20 and 40) represent changes in inspired gas. The shifted driving function that maximizes
740
the correlation with each unit is represented (square wave, dashed line). The integer
741
amount of breath by with the driving function is shifted so as to maximize correlation is
742
the correlation delay time (2 breaths, for SV = 0.72 and 6 breaths for SV = 0.12). Only
743
the first 60 breaths are represented here, yet the entire series (220 breaths) was used in the
744
computation of the correlation delay time.
745
Bottom (B): translation function from measured correlation delay time (Y axis) to
746
specific ventilation, for two different values of delay due to the plumbing – zero delay
747
(continuous line) and 1 breath delay (dashed line). The addition of a plumbing delay
748
resulted, as expected, in a parallel vertical displacement of the curve.
749
750
Figure 3 – Time series of signal intensity for a single voxel. When the subject changed
751
from breathing air to oxygen, signal intensity increased. The inspired O2 fraction (FIO2) is
752
represented by the dashed line. Once the extrinsic plumbing delay has been accounted
753
for, the correlation delay time (expressed in number of breaths) between the inspired FIO2
754
driving function and each voxel’s time series is a quantitative measure of ventilation.
755
Units with higher specific ventilation equilibrate faster, thus signal intensity increases
756
faster, and correlation delay time is shorter, than for units with lower specific ventilation.
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
736
Specific Ventilation Imaging (SVI)-JAPPL-00220-2010–R3
9/16/2010
757
Applying the modeling and signal-processing algorithm described in the methods section,
758
different correlation delay times were converted into a physiologically meaningful
759
measure of specific ventilation. The correlation delay time for this voxel was 2 breaths,
760
corresponding to a specific ventilation of 0.35 (P<0.0001).
761
Figure 4 – Top (A): Correlation delay time map (in number of breaths) in a sagittal slice
763
of the right lung of a typical subject. Bottom (B): Corresponding specific ventilation map,
764
using the ‘translation’ shown in figure 2B. In this plane, the head is located to the right,
765
the diaphragm to the left. The vector g indicates the direction of gravity. The subject was
766
supine. Note the shorter correlation delay time in the dependent regions (A) which
767
indicates a greater SV in these regions that in the non-dependent lung (B).
768
769
Figure 5 – Specific ventilation per third of the lung, grouped by the vertical distance from
770
the most dependent portion. ANOVA for repeated measures comparing all 3 regions
771
showed a highly significant difference (F=26.8, p<0.0001). Post-hoc testing (paired t-
772
test) p values are reported in the figure (7 degrees of freedom, t values of 5.5, 2.4 and 5.8
773
for the dependent-intermediate, intermediate-nondependent and dependent-nondependent
774
regions paired comparisons, respectively).
775
776
Figure 6 – Specific ventilation averaged along the isogravitational lines, in 1cm bins,
777
plotted versus the vertical distance from the most dependent portion of the lung for the
778
right lung slice. Height increases up along the Y-axis. As different subjects have different
779
lung heights, data was only plotted when all the subjects contributed to the average.
780
(A): specific ventilation plotted against height of the lung for a single subject (S8), with
781
horizontal error bars corresponding to the variability (SD) present in each isogravitational
782
bin.
783
(B): individual height profiles of mean specific ventilation for the 8 subjects studied, each
784
determined as in panel A.
785
(C): average height profiles of specific ventilation, with error bars corresponding to
786
inter-subject variability. The dashed line corresponds to the average slope determined by
787
averaging the individual slopes of the 8 individual height–SV profiles (panel B).
27
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
762
Δ
Δ
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
!
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
Table 1 – Subject characteristics, pulmonary function data and slope of the lung height – Specific ventilation relationship for the 8
subjects.
* Group average significantly different from zero, p=0.0002.
Gender
Age
(years)
Height
(m)
Weight
(kg)
S1
M
33
1.81
81
S2
M
26
1.85
93
S3
F
26
1.73
68
S4
F
39
1.62
66
S5
F
24
1.76
92
S6
M
52
1.86
113
S7
M
34
1.78
83
S8
M
28
1.70
60
33 ± 9
1.76 ± 0.08
82 ± 17
Mean ±
SD
FEV1
(liters)
(% predicted)
FVC
(liters)
(% predicted)
(% predicted)
3.99
88 %
4.57
92 %
4.13
115 %
3.03
101 %
3.08
102 %
4.50
105 %
3.84
89 %
4.43
107 %
3.85 ± 0.61
100 ± 10 %
5.17
93 %
5.28
87 %
5.04
118 %
3.65
99 %
3.65
99 %
5.36
96 %
4.98
93 %
5.16
103 %
4.79 ± 0.71
99 ± 9 %
0.77
95 %
0.87
105 %
0.82
96 %
0.83
101 %
0.84
102 %
0.84
109 %
0.77
95 %
0.86
104 %
0.82 ± 0.04
101 ± 5 %
FEV1/FVC
-1/slope
(cm-1)
Spatial
fractal
dimension
0.020
1.15
0.015
1.15
0.031
1.15
0.038
1.09
0.043
1.11
0.046
1.10
0.021
1.16
0.019
1.13
0.029 * ± 0.012
1.13 ± 0.03
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on July 31, 2017
Subject