| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Meakins-Christie Laboratories
and Department of Biomedical Engineering, Adler, Andy, Norihiro Shinozuka, Yves Berthiaume, Robert
Guardo, and Jason H. T. Bates. Electrical impedance tomography can
monitor dynamic hyperinflation in dogs. J. Appl.
Physiol. 84(2): 726-732, 1998.
positive end-expiratory pressure; intensive care unit; imaging
VENTILATED PATIENTS suffering from high expiratory
airway resistance may become dynamically hyperinflated if there is
insufficient time to expire before the onset of the next inspiration.
In the intensive care unit (ICU), various monitoring strategies are
used to estimate the degree of dynamic hyperinflation (DH) so as to effectively manage patients. It has been shown that the information obtained from cardiopulmonary monitoring procedures affects decisions concerning treatment (17, 23) and can decrease mortality rates by
50-80% in certain patients (9). Presently, measuring DH in ICU
patients is problematic and requires either an esophageal balloon or
end-expiratory airway occlusion (16). Neither of these techniques is
entirely satisfactory, being either somewhat invasive or requiring a
maneuver that interferes with normal breathing. Furthermore, neither
technique gives any information about the site of obstruction in the
lungs.
Electrical impedance tomography (EIT) can serve as a technique for
monitoring lung volumes that potentially overcomes many of these
disadvantages. EIT calculates a cross-sectional image of the change in
conductivity distribution in a body from electrical measurements made
at a series of electrodes placed around it. Because air is
significantly less conductive than the other tissues in the thorax,
changes in lung volume ( To assess EIT for pulmonary monitoring in the ICU, it is necessary to
determine the precision with which it can measure DH. Therefore, in
this study we determined the accuracy with which EIT measured changes
in lung inflation in anesthetized, paralyzed dogs, in which independent
determinations of Experimental procedures.
All procedures were reviewed and approved by the Animal Ethics
Committee of McGill University. Experiments were performed on eight
mongrel dogs (7 of which weighed 22 ± 2 kg, and 1 weighed 6.8 kg).
The dogs were anesthetized with an intravenous bolus of pentobarbital
sodium (25-30 mg/kg) and maintained by additional bolus injections
of 130 mg hourly. A rigid cannula (20 mm inner diameter) was inserted
into the trachea and connected to a Harvard volume ventilator (model
618, Harvard Apparatus, South Natick, MA). The expiratory line of the
ventilator was connected to an adjustable water trap that applied a set
level of PEEP. Pao was measured via a side tap at the tracheal cannula,
and Pes was measured with an esophageal balloon catheter. Pressure
measurements were made with a piezo-resistive pressure transducer
(Fujikura FPM-02PG, Servoflo, Lexington, MA). The dogs were ventilated
while in a supine position.
ABSTRACT
Top
Abstract
Introduction
Methods
Results
Discussion
References
We assessed in
eight dogs the accuracy with which electrical impedance tomography
(EIT) can monitor changes in lung volume by comparing the changes in
mean lung conductivity obtained with EIT to changes in esophageal
pressure (Pes) and to airway opening pressure (Pao) measured after
airway occlusion. The average volume measurement errors were
determined: 26 ml for EIT; 35 ml for Pao; and 54 ml for Pes.
Subsequently, lung inflation due to applied positive end-expiratory
pressure was measured by EIT
(
VEIT) and Pao
(VPao) under both inflation
and deflation conditions. Whereas
VPao was equal under both
conditions, VEIT was 28 ml
greater during deflation than inflation, indicating that EIT is
sensitive to lung volume history. The average inflation VEIT was 67.7 ± 78 ml
greater than VPao, for an
average volume increase of 418 ml. Lung inflation due to external
expiratory resistance was measured during ventilation by EIT
(VEIT,vent) and Pes
(VPes,vent) and at occlusion
by EIT (VEIT,occl), Pes, and
Pao. The average differences between EIT estimates and
VEIT,occl were 5.8 ± 44 ml
for VEIT,vent and 49.5 ± 34 ml for VEIT,occl. The average
volume increase for all dogs was 442.2 ml. These results show that EIT
can provide usefully accurate estimates of changes in lung volume over
an extended time period and that the technique has promise as a means
of conveniently and noninvasively monitoring lung hyperinflation.
INTRODUCTION
Top
Abstract
Introduction
Methods
Results
Discussion
References
VL)
induce marked changes in the conductivity distribution of the thorax
and can be imaged by EIT. Additionally, EIT is noninvasive, uses
current levels 1/10 of the threshold for cutaneous perception (8), and
is minimally cumbersome because the use of thoracic electrodes allows the airway to remain unobstructed and so does not restrict access to
the lungs. EIT also produces an image of the lungs from which regional
inhomogeneities can be determined and is able to monitor continuously,
with sampling frequencies up to 25 images/s (10, 24).
VL could be
made with the greatest precision. We produced volume changes by
applying positive end-expiratory pressure (PEEP) and by using an
external expiratory resistance. Lung volume estimates by EIT were
compared with those calculated from airway opening pressure (Pao)
during airway occlusion and from measurements of esophageal pressure (Pes).
METHODS
Top
Abstract
Introduction
Methods
Results
Discussion
References
View larger version (29K):
[in a new window]
Fig. 1.
Block diagram of electrical impedance tomography (EIT) system used. A
current of 300 µA at 10 kHz is injected across a pair of electrodes,
while voltage differences produced are measured at all other electrode
pairs. Process is then repeated for all patterns of current injection
and measurement to form 1 EIT data set, which is then sent to the
computer, where image of change in conductivity distribution between
the taking of 2 data sets is calculated.
VL were calculated from the
image of the conductivity change between a data set of interest and a
reference data set. A reference data set was taken by stopping the
ventilation, removing any applied PEEP, allowing the animal to
passively expire for 5 s, and then continuously acquiring EIT data for
10 s and ensemble averaging the measured data sets. The EIT images of
VL with respect to the
reference data set can thus be considered to represent the volume level
above functional residual capacity (FRC). Three different reference
data set protocols were used: 1)
sets taken immediately after the experimental protocol,
2) sets taken 3 min previously, and
3) sets taken between 30 and 60 min
previously. A reference level of Pes was acquired at the same time as
the EIT reference.
A region of interest (ROI) surrounding the lung was selected, and image
pixels in the region were summed. The ROI encompassed ~75% of the
image. Exclusion of image pixels in the heart region and those near the
body surface reduces the contribution of cardiac activity to volume
measurement and that from artifacts due to electrode movement and
postural change (3). Data sets acquired during an interval of interest
were ensemble averaged to eliminate the effect of cardiac activity and
to reduce the measurement noise.
The EIT images and both Pao and Pes were calibrated to lung volume
above FRC. First, the lungs were twice inflated to 3 kPa, after which
the ventilation was stopped, the dog was allowed to expire to FRC, and
a reference EIT data set was acquired. Then, three different gas volume
levels were introduced into the lungs in a stepwise fashion by using a
large syringe, during which period EIT data were continuously acquired.
The average value of the pixel sum in the ROI of the EIT image and Pao
and Pes were determined at each lung volume. This protocol was repeated
for lung volumes of 50, 100, 150, 200, 300, 400, 500, 600, 700, 800, 900, and 1,000 ml (the order of volume levels was randomized). Finally,
a calibration factor between the imposed
VL and the changes in Pao and
Pes, and the pixel sum of the EIT image ROI were calculated by linear regression. Subsequently, these calibration factors were used to
calculate the volume estimates:
VPao was calculated from Pao, VPes was calculated from Pes,
and VEIT was calculated from
the pixel sum in the EIT image ROI. Calibration factors were calculated for each animal.
Inflation response to applied PEEP.
VEIT and
VPao, as a function of the
applied PEEP, were measured in eight dogs. Figure
2 illustrates the protocol used and shows
an airway pressure trace from one animal. To study the effect of
differences in lung pressure-volume (PV) history on the volume estimates by EIT and Pao, measurements were made on both the inflation and deflation paths. Initially, volume history was normalized by two
inflations to 3 kPa, after which the dog was allowed to expire to the
applied PEEP level while being ventilated for 60 s. Ventilation was
then stopped, and the dog allowed to expire against the applied PEEP
for 15 s while EIT data were continually acquired. The volume estimates
for the deflation path,
VEIT,defl and
VPao,defl, were calculated from
the ensemble averages of the acquired data during this period.
Subsequently, the dog was ventilated without PEEP for 60 s, after which
the same level of PEEP was reapplied. Again, ventilation was stopped
and the dog was allowed to expire against PEEP for 15 s while data were
acquired, from which the volume estimates for the inflation path,
VEIT,infl and
VPao,infl, were calculated.
This protocol was repeated for PEEP levels of 0.3, 0.5, 0.7, 1.0, and
1.2 kPa.
|
Inflation response to expiratory resistance.
VEIT and
VPao due to increased
expiratory resistance were measured in four dogs. Figure
3 illustrates the protocol used and shows
an airway pressure trace from one animal. Volume history was normalized
by two inflations to 3 kPa, only one of which is shown in Fig. 3, and
the animal was ventilated normally for ~60 s, and then a
linear-resistive element was fixed to the expiratory tube for 140 s.
EIT data were acquired continuously during the last 30 s of this
period. The airway was then occluded, ventilation was stopped, and EIT
data were acquired for 15 s. This protocol was repeated four to six
times with different resistances.
|
VEIT,occl), Pao (VPao,occl), and Pes
(VPes,occl) were calculated
from the ensemble average of data acquired at occlusion.
2) The volume estimates during
ventilation measured by EIT
(VEIT,vent) and Pes
(VPes,vent) were calculated
by acquiring data for 30 s, generating a volume estimate for each data
set and averaging the lowest 10% of the values. The figure of 10% was
chosen to keep enough data sets to provide some immunity to measurement
noise while excluding data sets that are far from end-expiratory lung
volume.
Finally, in one dog, to assess how EIT images give information about
the distribution of ventilation between the lungs, one lung was blocked
by inserting a plug into the left main stem bronchus, and the images
were compared with those obtained before blockage.
| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
Figure 4A shows an image in a cross section of the thorax of the conductivity change due to a volume increase of 500 ml from FRC. The grey-scale level corresponds to the magnitude of conductivity change. Dark areas indicate decreasing conductivity, light areas indicate increase, and neutral grey indicates no conductivity change. The two areas of decreasing conductivity correspond to the lungs. It was possible to see two clearly defined lung regions in all dogs.
|
The location of the maximum conductivity change was identified by
selecting all image pixels greater than one-half of the maximum image
pixel value and calculating the center of gravity (CG) with respect to
the center of the thorax. CG values in the lateral and
anterior-posterior (AP) dimensions were calculated as a percentage,
where zero indicates the center of the thorax, 100% indicates the
front or right surface, and 
100% indicates the back or left.
The average values for all dogs for a
VL of 800 ml were calculated:
the EIT image CG in the lateral dimension was 6.2 ± 8%,
which is roughly the center and indicates that the conductivity changes
in each lung were equal, whereas the CG in the AP dimension was 29.7 ± 4%, indicating a significant displacement of maximum
conductivity change toward the front of the animal. The AP dimension of
the CG decreased with increasing VL by 1.8 ± 0.5% for
each 100-ml change.
Figure 4B shows the conductivity change image due to a 400-ml volume increase in the animal with the blocked left lung. The image clearly shows most of the conductivity change on the right side. The CG in the lateral dimension averaged over tidal volumes from 50 to 600 ml was 18.9 ± 14%.
The volume estimates VEIT,
VPao, and
VPes as a function of time
during a calibration protocol in one dog are shown in Fig. 5. At time = 0 s, the ventilation is
stopped and the dog is allowed to passively expire to FRC. Volume steps
to 50, 500, and 900 ml above FRC are then introduced, after which the
dog is again allowed to expire to FRC.
|
Figure 6 shows the mean and standard
deviation (SD) values of VEIT,
VPao, and
VPes as a function of
VL for all dogs measured during the calibration protocol. This linear relationship between the
measurements and VL was seen
in all animals. The average correlation coefficient between
VL and both
VEIT and
VPao was 0.996 ± 0.002, whereas for VPes it was
0.986 ± 0.007.
|
We calculated the measurement error as the SD of the difference between
the various volume estimates and the syringe volume. By using
calibration data for all animals, we found that the average measurement
error was 26.2 ml for VEIT,
34.7 ml for VPao, and 53.6 ml
for VPes.
Inflation response to applied PEEP.
The relationship between PEEP and the change in volume measured by each
technique is essentially linear and can be described by the compliance,
C, for each dog for both EIT and Pao measurements, under inflation and
deflation conditions. Thus, for example, the deflation compliance by
EIT (CEIT,defl) is calculated by
finding the linear regression fit between
VEIT,defl and the applied PEEP. The following values were calculated (in units of ml/kPa ± SD): CEIT,defl: 683 ± 199; CEIT,infl: 647 ± 193; CPao,defl: 563 ± 157; CPao,infl: 559 ± 157;
CPes,defl: 621 ± 147; and
CPes,infl: 598 ± 139. The two
CPao values for each dog were very
similar: the average ratio between
CPao,defl and
CPao,infl was 1.003 ± 0.013.
VEIT as a function of
VPao for all dogs, for both the
inflation and deflation measurements. On average,
VEIT,defl overestimated
VPao by 95.8 ± 84 ml, and
VEIT,infl overestimated VPao by 67.7 ± 78 ml. The average value of VPao
for all dogs was 416.1 ml.
|
VEIT,infl by using two
different reference data sets: 1) those taken 3 min
previously
(
)
and 2) those taken at least 30 min previously
(
).
The average to-
ratio calculated for all dogs was 1.02 ± 0.03, and the
-to- ratio calculated in all dogs except two was 1.02 ± 0.10. In the remaining dogs, the ratios were very different (0.47 and 0.62, respectively), possibly because the dogs had been moved during the
protocol and the electrode positions had shifted. This could have
resulted in both a different electrode spacing and a change in the
electrode-tissue impedance, both of which can affect EIT images.
Inflation response to expiratory resistance.
The inflation response to expiratory resistance was calculated during
ventilation and occlusion for several expiratory resistances. For each
dog, constants (k) were calculated
by linear regression to describe the best-fit linear relationships
between VEIT,occl, VEIT,vent,
VPao,vent, and
VPes,vent compared with
Vpao,occl for all resistance
values. For example,
kEIT,occl is the
ratio of VEIT,occl to
VPao,occl. The following values
were calculated (±SD):
kEIT,occl: 1.135 ± 0.063;
kEIT,vent: 1.018 ± 0.072;
kPes,occl: 1.169 ± 0.190; and
kPes,vent: 0.955 ± 0.112.
VEIT,occl and
VEIT,vent as functions of
VPao,occl. The average
differences between the following quantities and
VPao,occl were calculated:
VEIT,vent: 5.8 ± 44 ml; VEIT,occl: 49.5 ± 34 ml; VPes,vent: 22.7 ± 84 ml; and VPes,occl:
46.0 ± 65 ml. The average
VPao,occl for all dogs was
442.2 ml.
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
One of the goals of monitoring ventilated patients in ICU environments is to estimate the degree of DH and its progression over time in response to therapy and to changes in the ventilatory regime of the patient. Ideally, a monitoring technique would noninvasively provide a continuous reading of lung volume without requiring maneuvers that interfere with normal breathing and would not be cumbersome to the patient or for clinical staff. Additionally, it should provide information about regional inhomogeneities in the distribution of ventilation. Because EIT potentially fulfills these criteria, we have studied the precision with which EIT can measure DH in dogs. Changes in lung inflation due to applied PEEP and to external expiratory resistance were determined by EIT and compared with the values calculated from Pao and Pes.
Step changes in lung volume were shown to be measurable by EIT with a
precision similar to that from Pao values, with the advantage that EIT
did not show the overshoot response to volume steps characteristic of
Pao measurement (Fig. 5). EIT estimates of
VL to applied PEEP were
sufficiently close to Pao values to be within the 100-ml error value
considered acceptable for spirometers by the American Thoracic Society
(7). One contribution to the difference in estimates was the
sensitivity of EIT to lung PV hysteresis, which is not taken into
account by measurements of airway pressure.
VEIT,occl values after
application of external expiratory resistance were close to Pao and Pes
estimates. Measurements of
VEIT,vent were close to both
VPes,vent and
VPao,occl. In addition to measuring VL
during ventilation with a precision comparable to that of occlusion
pressure measurement, EIT was able to provide a cross-sectional image
from which the regional inhomogeneities in the distribution of
ventilation could be determined. EIT was also capable of accurately
monitoring hyperinflation volume changes with respect to reference data
sets acquired 30 min previously.
The EIT, Pao, and Pes values were calibrated with respect to known
volume changes to calculate the change in volume above FRC from
experimental values of these quantities. Additionally, the volume
measurement errors for each parameter were calculated from the
calibration data. The errors for
VEIT and
VPao were similar, whereas the
error for VPes was
approximately twice as large. Because these estimates assumed the
syringe volume measurements to be accurate, they tended to overestimate
the actual errors by an amount related to the inaccuracy in the syringe
volume measurements. The shape of the
VEIT curve as a function of
time did not show any overshoot after the volume steps, as seen in the
VPao and VPes curves. Pressure
overshoots to volume steps are due to the flow across the airway
resistance and to the stress relaxation in the lung tissue (6). Because
EIT is not sensitive to either phenomenon, it is able to directly track
the lung volume.
The lung inflation response to applied PEEP was measured on both the
inflation and deflation paths. The two Pao estimates, VPao,defl and
VPao,infl, were very similar,
whereas VEIT,defl was, on
average, 6% higher than
VEIT,infl. These results agree with the fact that hysteresis in the lung PV curve results in a higher
volume at the same pressure with deflation from TLC than with inflation
from FRC. According to the analysis of Bachofen et al. (4), the average
difference between inflation and deflation volume over a PV loop is
0.14 times the loop volume for several species, including dogs. This
analysis would predict an average difference
(VEIT,defl VEIT,infl) of 58 ml (0.14 times
the average VPao), which is
reasonably close to the measured difference of 28 ml.
VEIT was systematically larger
than VPao by 67.7 ml on the
deflation path and by 95.8 ml on inflation. The higher EIT estimates
may be due to the PV hysteresis, because calibration data were taken a
maximum of 30 s after inflation to TLC, whereas experimental
measurements were made 60-240 s after the last TLC inflation. Thus
the lung may have had time to settle into a state similar to that of
the inflation path of the PV curve, even for VEIT,defl estimates.
The inflation response to expiratory resistance was measured during
ventilation and at airway occlusion, whereas Pao measurements are taken
at airway occlusion. VEIT,occl
was, on average, 50 ml higher than
VPao,occl, which is similar to
the difference between EIT and Pao for applied PEEP on the deflation
path. During ventilation, however, the EIT volume estimate was roughly
equal to VPao,occl. The volume
estimates by use of Pes were similar to those by use of EIT:
VPes,occl was almost identical
to VEIT,occl, whereas
VPes,vent was ~30 ml less
than VEIT,vent. This indicates that EIT enables measurement of DH during ventilation without the need
for an occlusion maneuver. Although DH measurement during ventilation
by Pes was only slightly less accurate than EIT measurement, placement
of esophageal balloons in patients is somewhat invasive and suffers
from errors due to expiratory muscle activity and cardiogenic
oscillations on the pressure signals (22).
The SD of the difference between
VEIT and
VPao for all measurements of
volume increase to both applied PEEP and expiratory resistance was 68 ml. This is close to the sum of the individual volume measurement
errors of EIT and Pao; this suggests that additional sources of error
during the experimental protocol compared with the calibration do not
significantly increase the volume-measurement error.
EIT imaging with respect to a reference data set taken sometime previously is more representative of the proposed clinical applications of EIT, in which it would be important to follow changes in the ventilatory regime over periods of minutes to hours. Our results indicate that the EIT system used for these measurements is stable for periods of 30 min. In most animals, the reference data set taken 30 min previously was almost as good as that taken immediately after the measurement. However, in two animals, the previous reference was significantly different. This difference may be due to movement of the electrodes during the experimental protocol. Small movements of electrodes between data sets are known to introduce large artifacts into the conductivity change images (3, 5). This experimental protocol, using subcutaneous needle electrodes, is prone to electrode movement whenever the animal is bumped or moved. Because of this sensitivity to electrode movement, in awake patients it will be important to use electrodes that are as mechanically stable as possible.
The conductivity change images produced by EIT, while of low resolution
compared with standard medical imaging modalities, clearly show useful
anatomic information. The images separate the two lung regions and also
indicate that the zone of maximum conductivity change occurs in the
ventral portion of the images. This contrasts with our previous results
(1), in which dogs were ventilated while in a prone position. In that
study, the average CG in the AP dimension was 2.3 ± 7% and
was 0.6 ± 2% in the lateral dimension, for a
VL of 700 ml, in contrast to an average AP dimension CG of 29.7% in this study. Thus the maximum conductivity change due to ventilation was significantly more ventral
in dogs ventilated while supine than in those ventilated while prone.
Additionally, the image regions corresponding to the two lungs tended
to merge and were less easily distinguishable in the prone dogs.
Although the dependent portion of the lungs receives a greater portion
of the volume in spontaneously breathing individuals (14), the reverse
is the case for paralyzed and mechanically ventilated normal subjects
(12). Thus our results, showing movement of the maximum conductivity
change region into the upper portion of the lungs, agree with the
expected change for ventilated subjects. The movement of the maximum
conductivity change region to the dependent portion of the lung for
spontaneously breathing normal subjects has also been shown in a recent
study of the gravity dependance of EIT images (11). EIT was also
clearly able to detect the blockage of one lung, by showing a
conductivity change only on one side of the thorax. The ability of EIT
to detect unilateral ventilation also has been shown by Hahn et al.
(13) in pigs, by Morice et al. (18) in patients with pneumothorax, and
by Newell et al. (19) in dogs.
One limitation of this study is the comparison of EIT measurement to
volume measurements calculated from Pao and Pes. Volume measurements
based on pressure are subject to various errors and cannot be
considered a "gold standard" against which the precision of EIT
can be compared. One such problem is that pressure-based volume
measurement is insensitive to PV hysteresis. However, by calibrating
the measurement errors of the techniques used, it was possible to
determine the significance of the differences between the estimates.
Additionally, DH measurement made by using the occlusion technique is
relatively well understood. For example, Rossi et al. (20) compared the
intrinsic PEEP at end expiration obtained from airway occlusion to the
Pao at the onset of inspiratory flow in patients with DH. From their
data, we calculate that the Pao measurement error for occlusion is
0.080 kPa. This is very similar to the measurement error for
VPao in this study, divided by
the average compliance, 0.062 kPa. Another limitation is that our EIT
system only takes measurements in a single electrode plane and thus
provides only a two-dimensional conductivity change image. Because much
of the increase in lung volume in the dog occurs by descent of the
diaphragm, it is encouraging that, with this configuration, we were
still able to assess VL in
terms of changes in thoracic cross section. Furthermore, it was
possible to discriminate the regional differences in the
cross-sectional plane imaged. To assess heterogeneity of ventilation
within the lungs as a whole, a three-dimensional EIT system would be
required. The technical feasibility of such a system has recently been
shown by Metherall et al. (15).
In summary, this study has shown that
VL due to applied PEEP and
expiratory resistance can be measured by EIT in dogs with an accuracy
acceptable for monitoring in an ICU environment. The results suggest
that EIT offers several advantages for the monitoring of DH. In
addition to its precision being comparable or better than that of
airway occlusion- or Pes-based techniques, EIT is not affected by
flow-resistance and stress-relaxation effects. Significantly, EIT is
able to noninvasively and continuously monitor the lung volume over
periods of at least 30 min without interfering with normal breathing or
being cumbersome to the patient or staff. EIT images also provide
useful information about the distribution of ventilation. Thus EIT
shows promise as a technique for routine monitoring of ICU patients,
which could measure the progression of the level of DH in response to
therapy or to changes in a patient's ventilatory regime.
| |
ACKNOWLEDGEMENTS |
|---|
This work was supported by the Medical Research Council of Canada, Inspiraplex, the Association Pulmonaire du Québec, and the J. T. Costello Memorial Research Fund. A. Adler is the recipient of a fellowship award from the Natural Sciences and Engineering Research Council. Y. Berthiaume is a Chercheur-Boursier Clinicien of the Fonds de la Recherche en Santé du Québec (FRSQ), and J. H. T. Bates is a Chercheur-Boursier of the FRSQ.
| |
FOOTNOTES |
|---|
Address for reprint requests: J. H. T. Bates, Meakins-Christie Laboratories, McGill Univ., 3626 St-Urbain St., Montreal, Canada H2X 2P2 (E-mail: jason{at}Meakins.lan.McGill.ca).
Received 28 April 1997; accepted in final form 3 October 1997.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
| 1. |
Adler, A.,
R. Amyot,
R. Guardo,
J. H. T. Bates,
and
Y. Berthiaume.
Monitoring changes in lung air and liquid volumes with electrical impedance tomography.
J. Appl. Phys.
83:
1762-1767,
1997 |
| 2. | Adler, A., and R. Guardo. Electrical impedance tomography: regularized imaging and contrast detection. IEEE Trans. Med. Imaging 15: 170-179, 1996. [Medline] |
| 3. | Adler, A., R. Guardo, and Y. Berthiaume. Impedance imaging of lung ventilation: do we need to account for chest expansion? IEEE Trans. Biomed. Eng. 43: 414-421, 1996[Medline]. |
| 4. |
Bachofen, H.,
and
J. Hildebrandt.
Area analysis of pressure-volume hysteresis in mammalian lungs.
J. Appl. Physiol.
30:
493-497,
1971 |
| 5. | Barber, D. C., and D. H. Brown. Errors in reconstruction of resistivity images using a linear reconstruction technique. Clin. Phys. Physiol. Meas. 9A: 101-104, 1988. |
| 6. |
Bates, J. H. T.,
P. Baconnier,
and
J. Milic-Emili.
A theoretical analysis of interrupter technique for measuring respiratory mechanics.
J. Appl. Physiol.
64:
2204-2214,
1988 |
| 7. | Crapo, R. O., J. L. Hankinson, C. Irvin, N. R. Macintyre, K. Z. Voter, and R. A. Wise. Standardization of spirometry. 1994 update. Am. J. Respir. Crit. Care Med. 152: 1107-1136, 1995[Medline]. |
| 8. | Dalziel, C. F. Effect of electric shock on man. IRE Trans. Med. Electron. 5: 44-62, 1956. |
| 9. |
Del Guercio, L. R.,
and
J. D. Cohn.
Monitoring operative risk in the elderly.
JAMA
243:
1350-1355,
1980 |
| 10. | Edic, P. M., G. J. Saulnier, J. C. Newell, and D. Isaacson. A real-time electrical impedance tomograph. IEEE Trans. Biomed. Eng. 42: 849-859, 1995[Medline]. |
| 11. | Frerichs, I., G. Hahn, and G. Hellige. Gravity-dependent phenomena in lung ventilation determined by functional EIT. Physiol. Meas. 17: A149-A157, 1996. |
| 12. | Froese, A. B., and A. C. Bryan. Effects of anesthesia and paralysis on diaphragmatic mechanics in man. Anesthesiology 41: 242-255, 1974[Medline]. |
| 13. | Hahn, G., I. Sipinková, F. Baisch, and G. Hellige. Changes in the thoracic impedance distribution under different ventilatory conditions. Physiol. Meas. 16: A161-A173, 1995[Medline]. |
| 14. |
Kaneko, K.,
J. Milic-Emili,
M. B. Dolovich,
A. Dawson,
and
D. V. Bates.
Regional distribution of ventilation and perfusion as a function of body position.
J. Appl. Physiol.
21:
767-777,
1966 |
| 15. | Metherall, P., D. C. Barber, R. H. Smallwood, and B. H. Brown. Three dimensional electrical impedance tomography. Nature 380: 509-512, 1996[Medline]. |
| 16. |
Milic-Emili, J. F. M.,
F. M. Robatto,
and
J. H. T. Bates.
Respiratory mechanics in anaesthesia.
Br. J. Anaesth.
65:
4-12,
1990 |
| 17. | Mimoz, O., A. Rauss, N. Rekik, C. Brun-Buisson, F. Lemaire, and L. Brochard. Pulmonary artery catherization in critically ill patients: a prospective analysis of outcome changes associated with catheter-prompted changes in therapy. Crit. Care Med. 22: 573-579, 1994[Medline]. |
| 18. | Morice, A. H., N. Harris, J. Campbell, F. Zhang, and B. Brown. EIT in the investigation of chest disease. In: Clinical and Physiological Applications of Electrical Impedance Tomography, edited by D. Holder. London: UCL, 1993, p. 236-241. |
| 19. | Newell, J. C., D. Isaacson, and R. S. Blue. Regional ventilation and perfusion assessed by electrical impedance imaging. Proc. IEEE Eng. Med. Biol. Soc. Conf., Amsterdam, Oct. 31-Nov. 3, 1996. (Paper no. 601) |
| 20. | Rossi, A., B. Gottfried, L. Zocchi, B. D. Higgs, S. Lennox, P. M. Calverley, P. Begin, A. Grassino, and J. Milic-Emili. Measurement of static compliance of the total respiratory system in patients with acute respiratory failure during mechanical ventilation. Am. Rev. Respir. Dis. 131: 672-677, 1985[Medline]. |
| 21. | Savoie, G. Applications de la tomographie d'impédance électrique à l'étude de phénomènes physiologiques (Master's thesis). Montréal, Canada: École Polytechnique de Montréal, 1994. |
| 22. | Schuessler, T. F., S. B. Gottfried, and J. H. T. Bates. A model of the spontaneously breathing patient: applications to intrinstic PEEP and work of breathing. J. Appl. Physiol. 82: 1695-1703, 1997. |
| 23. |
Weidermann, H. P.,
M. A. Matthay,
and
R. A. Matthay.
Cardiovascular-pulmonary monitoring in the intensive care unit (Part 2).
Chest
85:
656-668,
1984 |
| 24. | Zhu, Q. S., C. N. McLeod, F. J. Lidgey, and W. R. B. Lionheart. Development of a realtime adaptive current tomograph. Physiol. Meas. 15: A37-A43, 1994. |
This article has been cited by other articles:
|
|
 |
|
  J. A. Victorino, J. B. Borges, V. N. Okamoto, G. F. J. Matos, M. R. Tucci, M. P. R. Caramez, H. Tanaka, F. S. Sipmann, D. C. B. Santos, C. S. V. Barbas, et al. Imbalances in Regional Lung Ventilation: A Validation Study on Electrical Impedance Tomography Am. J. Respir. Crit. Care Med., April 1, 2004; 169(7): 791 - 800. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hinz, P. Neumann, T. Dudykevych, L. G. Andersson, H. Wrigge, H. Burchardi, and G. Hedenstierna Regional Ventilation by Electrical Impedance Tomography: A Comparison With Ventilation Scintigraphy in Pigs Chest, July 1, 2003; 124(1): 314 - 322. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
