J Appl Physiol 96: 1127–1136, 2004.
First published November 14, 2003; 10.1152/japplphysiol.00092.2003.
Physiological evaluation of a new quantitative SPECT method measuring
regional ventilation and perfusion
Johan Petersson,1 Alejandro Sánchez-Crespo,2,3 Malin Rohdin,4 Stéphanie Montmerle,4
Sven Nyrén,5 Hans Jacobsson,2,5 Stig A. Larsson,2,3 Sten G. E. Lindahl,1
Dag Linnarsson,4 Robb W. Glenny,1,6 and Margareta Mure1
1
Department of Anesthesiology and Intensive Care, 2Section of Nuclear Medicine, Department of Hospital Physics,
and 5Department of Radiology, Karolinska Hospital, 171 76 Stockholm; 3Medical Radiation Physics, Department
of Oncology-Pathology, Stockholm University and Karolinska Institutet and 4Section of Environmental Physiology,
Department of Physiology and Pharmacology, Karolinska Institutet, 171 77 Stockholm, Sweden; and 6Departments
of Medicine and Physiology and Biophysics, University of Washington, Seattle, Washington 98195
Submitted 29 January 2003; accepted in final form 6 November 2003
single-photon-emission computed tomography; multiple inert-gas
elimination technique; gas exchange
ventilation and regional lung perfusion in the whole lung. The
method has previously been evaluated by using phantom studies (47). In the present study, the multiple inert-gas elimination
technique (MIGET) (54, 55) and respiratory gas exchange
were used as physiological evaluations of the new SPECT
method. We evaluated not only the ability of the method to
depict the distributions of radioactivity in the subject, but also
how well the method, including the methods of administering
the isotopes, describes the actual distributions of regional
ventilation and regional perfusion in human subjects.
METHODS
Subjects
Ten healthy volunteers (5 men and 5 women), aged 20–40 yr, were
studied. All subjects were of normal weight (range 57–70 kg), height
(range 160–175 cm), and body mass index (range 20–26). We recruited subjects ⬍180 cm in height to ensure that all lung regions
would fit into the scanning field of the SPECT camera. None of the
subjects had a history of pulmonary disease, all were nonsmokers, and
all had normal spirometry and lung volumes. The subjects received
written information about the procedure, and informed verbal consent
was obtained. Under sterile conditions and after infiltration of local
anesthetics, an arterial line was inserted into each subject’s radial
artery at the wrist. An intravenous catheter was inserted into the
antecubital vein of the same arm. Results from one of the subjects had
to be excluded from the study because of technical problems with the
transmission source causing artifacts in the SPECT results. The study
was approved by the local ethical committee and the local radiation
safety committee.
EFFICIENT GAS EXCHANGE in the lungs is the result of intimate
matching of regional alveolar ventilation and perfusion. Regional distributions of ventilation and perfusion in human
subjects have previously been analyzed by use of a number of
techniques, many of which have used external detection of
radioactive markers for ventilation and perfusion (3, 15, 21, 29,
32, 36, 39, 42, 57, 58). Measurement of regional distributions
of ventilation and perfusion provides insights into how different treatment regimes (e.g., positioning) influences gas exchange. To explain such mechanisms, a method is needed that
provides simultaneous quantitative measures of both regional
ventilation and perfusion for the whole lung. To meet these
needs, we developed a new dual-isotope quantitative singlephoton-emission computed tomography (SPECT) method that
simultaneously measures the distributions of regional alveolar
Radiopharmaceuticals. Regional distribution of ventilation was
marked by using inhaled Technegas, microscopic graphite particles
labeled with radioactive technetium (99mTc) (11). Regional distribution of perfusion was marked by use of macroaggregates of albumin
(LyoMAA, Mallinckrodt Medical, Petten, The Netherlands) labeled
with radioactive indium (113mIn). Both radiotracers were administered
to subjects in the sitting upright posture. The principal photon energy
of 99mTc is 140 keV and for 113mIn 392 keV. Labeling of the
macroaggregates was performed according to Watanabe et al. (56)
with a few modifications. 99mTc was obtained from a 99Mo-99mTc
generator (Nycomed/Amersham), whereas 113mIn was obtained from
Address for reprint requests and other correspondence: J. Petersson, Dept. of
Anesthesiology and Intensive Care, Karolinska Hospital, 171 76 Stockholm,
Sweden (E-mail: [email protected]).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
http://www.jap.org
SPECT Imaging
8750-7587/04 $5.00 Copyright © 2004 the American Physiological Society
1127
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 14, 2017
Petersson, Johan, Alejandro Sánchez-Crespo, Malin Rohdin,
Stéphanie Montmerle, Sven Nyrén, Hans Jacobsson, Stig A. Larsson, Sten G. E. Lindahl, Dag Linnarsson, Robb W. Glenny, and
Margareta Mure. Physiological evaluation of a new quantitative
SPECT method measuring regional ventilation and perfusion. J Appl
Physiol 96: 1127–1136, 2004. First published November 14, 2003;
10.1152/japplphysiol.00092.2003.—We have developed a new quantitative single-photon-emission computed tomography (SPECT)
method that uses 113mIn-labeled albumin macroaggregates and Technegas (99mTc) to estimate the distributions of regional ventilation and
perfusion for the whole lung. The multiple inert-gas elimination
technique (MIGET) and whole lung respiratory gas exchange were
used as physiological evaluations of the SPECT method. Regional
ventilation and perfusion were estimated by SPECT in nine healthy
volunteers during awake, spontaneous breathing. Radiotracers were
administered with subjects sitting upright, and SPECT images were
acquired with subjects supine. Whole lung gas exchange of MIGET
gases and arterial PO2 and PCO2 gases was predicted from estimates of
regional ventilation and perfusion. We found a good agreement
between measured and SPECT-predicted exchange of MIGET and
respiratory gases. Correlations (r2) between SPECT-predicted and
measured inert-gas excretions and retentions were 0.99. The method
offers a new tool for measuring regional ventilation and perfusion in
humans.
1128
REGIONAL VENTILATION AND PERFUSION BY QUANTITATIVE SPECT
Assessment of Global Ventilation and Perfusion
Quantitative SPECT data represent regional ventilation and perfusion only in relative terms. Prediction of gas exchange from SPECT
requires separate measurements of global lung ventilation and perfusion.
Global alveolar ventilation. Whereas excretion and retention of
MIGET gases depends on total minute ventilation, Technegas distribution only describes alveolar ventilation. Estimation of gas exchange
from SPECT results therefore requires measurement of alveolar ven-
tilation. Anatomic dead space was measured with the Fowler method
(13), and alveolar ventilation was estimated from the total minute
ventilation minus the anatomic dead space ventilation. CO2 concentration and flow at the mouth were recorded during several slow
expirations by using a calibrated in-line infrared capnometer (model
14360 A, Hewlett-Packard, Palo Alto, CA) and a volume-calibrated
pneumotachograph (Fleisch no. 2, Siemens-Elema, Solna, Sweden,
linear for flows up to 3 l/s). Data from the capnometer and the
pneumotachograph were digitally recorded with a sampling frequency
of 50 Hz by using the software program Workbench (Strawberry Tree,
Sunnyvale, CA). Measured flow was corrected for the difference
between gas temperature at the point of measurement and body
temperature. The data were converted to text files and analyzed with
the use of Microsoft Excel. The same equipment was used for
registration of respiratory rate and minute ventilation during MIGET
sampling and during the administration of the radiopharmaceuticals
(during these recordings, the sampling frequency was reduced to 20
Hz). Alveolar ventilation was calculated by using the minute ventilation, the measured anatomic dead space, the apparatus dead space
(different for each situation, measured with water), and the respiratory
rate. Minute ventilation, alveolar ventilation, and tidal volumes were
adjusted to BTPS (body temperature and pressure, water vapor saturated gas).
Cardiac output. Cardiac output was measured with a total rebreathing technique using Freon 22 (CHClF2) (6, 45, 46). Gas flow was
measured with a pneumotachometer (type 3813, Hans Rudolph, Kansas City, MO) and gas concentrations with a quadrupole mass spectrometer (AMIS 2000, Innovision A/S, Odense, Denmark). Before
and after each experiment, the mass spectrometer was calibrated
against gases of known composition, and the flowmeter was calibrated
with a 3-liter syringe within the experimental flow range. The response latency of the mass spectrometer was determined from a
sudden, simultaneous change of gas composition and flow direction at
the inlet of the sampling capillary. During the rebreathing maneuver,
pulmonary capillary blood flow was estimated from the uptake of
Freon 22, and cardiac output was considered to equal pulmonary
capillary blood flow. Each subject performed five rebreathing maneuvers, and the mean value of the three most consistent measurements
was used to estimate cardiac output.
Regional Alveolar Ventilation and Perfusion
Coordinates and number of events per voxel within the total
delineated lung region were extracted from original reconstructed
transverse SPECT 113mIn-LyoMAA and 99mTc-Technegas image
data. Because of a limitation in the number of voxel data that could be
handled by the software used in the subsequent calculations of gas
exchange, the original set of 128 ⫻ 128 ⫻ 128 voxels was converted
into a smaller number (32 ⫻ 32 ⫻ 32) of larger voxels (1.4 ⫻ 1.4 ⫻
Fig. 1. Transverse lung section from 1 subject. Size of 1 ⫻ 1
(3.56 ⫻ 3.56 mm2) and 4 ⫻ 4 (1.4 ⫻ 1.4 cm2) pixels in relation
to lung size are shown. Coloring is according to a relative scale
for each image. Perfusion 1 ⫻ 1, range 0.04–0.09 ml䡠min⫺1 䡠
voxel⫺1; perfusion 4 ⫻ 4, range 2.6–5.6 ml䡠min⫺1 䡠voxel⫺1;
ventilation 1 ⫻ 1, range 0.05–0.14 ml䡠min⫺1 䡠voxel⫺1; ventilation 4 ⫻ 4, range 3.3–7.6 ml䡠min⫺1 䡠voxel⫺1. Max, maximum;
Min, minimum.
J Appl Physiol • VOL
96 • MARCH 2004 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 14, 2017
a 113Sn-113mIn generator (Radioisotope Center, Polatom, Otwock
Swierk, Poland). Technegas was inhaled during quiet tidal breathing
from a Technegas generator (Tetley Manufacturing, Sydney, Australia). This was done through a box that mixed Technegas (initially
100% argon) with air. Pulse oximetry was monitored during the
Technegas inhalation to demonstrate normal hemoglobin oxygen
saturation. During the Technegas inhalation, a radiation protection
monitor (Proportional chamber, Bertold, Bad Widbad, Germany) was
used to register the counts over the lung regions. Inhalation was
terminated when the counts in the lower dorsal fields reached a level
of 200 counts/min. Immediately after inhalation of Technegas, 100
MBq of 113mIn-LyoMAA was injected via the peripheral venous
catheter followed by a flush of normal saline. The subjects were
estimated to receive a total effective dose of ⬃5 mSv.
SPECT. The dual-isotope SPECT technique used in this work has
previously been presented in detail (47). In brief, SPECT images are
obtained with a three-headed gamma camera (TRIAD XLT 20,
Trionix Research Laboratory, Twinsburg, OH) equipped with medium
energy general-purpose parallel-hole collimators. SPECT scans are
performed in 72 projections, 62 s per projection, by use of a fourenergy window acquisition protocol. Thus four images (128 ⫻ 128
pixels with a pixel size of 3.56 ⫻ 3.56 mm2) are obtained at each
data-acquisition angle. An additional transmission tomography with a
moving 99mTc line source is performed to obtain data for the attenuation correction routine. The two sets of projected images, one for
each principal photon energy (140 and 392 keV), are corrected for
photon scattering and attenuation as well as for the contribution of
high-energy photons in the lower photon-energy window and the
radioactive decay before image reconstruction. The lung areas are
delineated in the reconstructed transverse transmission images by
using a previously described edge-detection algorithm (47). To verify
that only lung tissue was included in the images, they were reviewed
by a radiologist. In a few of the images, a small number of peripheral
pixels were considered nonlung tissue and therefore removed. All
images were obtained with subjects in the supine posture while
breathing against an expiratory pressure of 2.5 cmH2O to compensate
for the reduction in functional residual capacity (FRC) when changing
from the sitting to the supine position.
REGIONAL VENTILATION AND PERFUSION BY QUANTITATIVE SPECT
1129
Fig. 2. Timeline describing the experimental protocol. CO, cardiac output; V̇O2, oxygen consumption;
V̇CO2, carbon dioxide elimination; MIGET, multiple
inert-gas elimination technique; SPECT, single-photon-emission computed tomography; iv, intravenous.
each subject. A value 0.04 below the arterial pH was used as the
mixed venous pH.
Oxygen consumption and carbon dioxide production. Expired
gases were collected in a Douglas bag while the number of breaths
during 10 min were counted. After thorough mixing, the content of the
bag was analyzed by use of the same mass spectrometer as used for
the cardiac output measurements. Expired volume was measured by
use of a standard gas meter (AB Nordgas, Stockholm, Sweden).
Oxygen consumption and carbon dioxide elimination were calculated,
and oxygen consumption was corrected for the difference between
inspired and expired volume by Haldane transformation of the expired
volume (35).
Comparison of SPECT and MIGET
MIGET
A continuous intravenous infusion of a standard solution of six
inert gases (sulfur hexafluoride, ethane, cyclopropane, enflurane, diethyl ether, and acetone) was administered to the subjects and allowed
to equilibrate for 60 min. Inert-gas concentrations in arterial and
mixed exhaled gas were measured with a gas chromatograph (Star
3400 CX GC, Varian Chromatography Systems, used with a Varian
4400 Integrator) by using previously described techniques (54). Mean
values from duplicate samples of both expired gases and arterial blood
were used for the calculations. Retentions and excretions for the inert
gases were calculated without measured mixed venous concentrations
of the inert gases according to Gale et al. (14). Excretions were
corrected for anatomic dead space.
Mixed Venous Oxygen and Carbon Dioxide Content
Experimental Protocol
Prediction of respiratory gas exchange requires estimates of mixed
venous oxygen and carbon dioxide content. Because a pulmonary
catheter was not used, contents were calculated from arterial blood
content, oxygen consumption, and carbon dioxide elimination. Arterial blood oxygen content was calculated from blood hemoglobin
concentration and the percentage of oxyhemoglobin in arterial blood
by use of a standard equation.1 Mixed venous oxygen content was
estimated from arterial oxygen content, cardiac output, and oxygen
consumption. Arterial carbon dioxide content was estimated from the
arterial PCO2 (PaCO2), according to Douglas et al. (12), and mixed
venous carbon dioxide contents were subsequently estimated from the
arterial contents, cardiac output, and carbon dioxide elimination for
1
CaO2 ⫽ [Hb] ⫻ SO2% ⫻ 1.34. CaO2, arterial blood oxygen content [ml
(STPD)/100 ml]; [Hb], blood hemoglobin concentration (g/dl), SO2% (% of
hemoglobin saturated with oxygen); 1.34, the amount of oxygen bound to 1 g
of fully saturated hemoglobin.
J Appl Physiol • VOL
By using the method of Altemeier et al. (1), regional measurements
of ventilation and perfusion obtained by SPECT were used to estimate
arterial PO2 (PaO2), PaCO2, and retentions and excretions of the MIGET
gases. These calculations were performed with a Microsoft Excel
spreadsheet with macros written with Visual Basic for Applications;
input data consist of ventilation and perfusion (in ml/min per voxel),
barometric pressure, blood hemoglobin concentration, cardiac output,
mixed venous pH, PO2 and PCO2, and the inert gas blood solubilities.
Finally, SPECT-derived estimates of PaO2, PaCO2, and retentions and
excretions of the MIGET gases were compared with measured values.
Parameters derived from the 50-compartment model of the MIGET
method were compared with the ventilation and perfusion distributions estimated by the SPECT method.
Figure 2 provides an overview and timeline for the experiments.
The subjects sat in the upright posture until administration of the
radiopharmaceuticals was completed. Before each data collection, the
subjects rested undisturbed for a minimum of 5 min to attain as stable
conditions as possible throughout the experiments. Heart rate and
invasive blood pressure were monitored from the start of the physiological measurements until the administration of the radioactive
isotopes was completed. Pulse oximetry was monitored during rebreathing maneuvers and during Technegas administration by use of
an Ultima S device (Datex, Helsinki, Finland). Respiratory rate and
tidal volume were monitored during the MIGET sampling and during
the administration of the radiopharmaceuticals. After administration
of the radiotracers, subjects were transported to the gamma camera in
a wheelchair. Subjects were told to refrain from any movements
during image acquisition. Physiological measurements and isotope
administration lasted 2–3 h, and image acquisitions in the SPECT
camera required another 1–2 h.
96 • MARCH 2004 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 14, 2017
1.4 cm3), by adding 4 ⫻ 4 ⫻ 4 original voxels together. Figure 1
illustrates the increment in voxel size in relation to lung size. Indium
and technetium counts per voxel were converted to ventilation and
perfusion (both ml/min) per voxel by multiplying the fraction of the
total number of counts for each isotope in that voxel by the total
cardiac output and total alveolar ventilation, respectively.
Arterial blood gases. Arterial blood samples were obtained at the
start and finish of both the collection of expired gases and the MIGET
sampling. The mean values of these two samples were used as the
arterial blood-gas values during each situation. One arterial blood
sample was obtained immediately before the administration of the
radioactive isotopes. The samples were analyzed with a AVL OMNI
1-9 blood-gas analyzer (Roche Diagnostics, Graz, Austria).
1130
REGIONAL VENTILATION AND PERFUSION BY QUANTITATIVE SPECT
Table 1. Circulatory and respiratory parameters
CO, l/min
HR, beats/min
V̇O2, ml/min
V̇CO2, ml/min
SBP, mmHg
V̇E, l/min
V̇A, l/min
VT, ml
RR, breaths/min
PaO2, Torr
PaCO2, Torr
4.3⫾0.6
76⫾10
280⫾30
240⫾40
130⫾17
7.4⫾1.7
5.7⫾1.4
690⫾140
11⫾2
98⫾3
38⫾5
Values are means ⫾ SD for all subjects at the first measurement of each
parameter. CO, cardiac output; HR, heart rate; V̇O2, oxygen consumption;
V̇CO2, carbon dioxide elimination (V̇O2 and V̇CO2 are ml STPD); SBP, systolic
blood pressure; V̇E, expired minute volume; V̇A, alveolar ventilation; VT, mean
tidal volume during the registration (all ventilatory volumes as BTPS); RR,
mean respiratory rate during the registration; PaO2, measured arterial PO2;
PaCO2, measured arterial PCO2.
All data, unless otherwise stated, are presented as means ⫾ SD.
The agreements between measured retentions and excretions of the
MIGET gases and arterial blood gases were compared with those
estimated from the SPECT data by using plots of linear correlation
and by the method of Bland and Altman (5).
RESULTS
Physiological Measurements
General physiological descriptors of the subjects are summarized in Table 1. Cardiac output was 4.3 ⫾ 0.6 l/min,
oxygen consumption 280 ⫾ 30 ml/min, and carbon dioxide
elimination 240 ⫾ 40 ml/min (volumes in STPD, standard
temperature and pressure, dry gas). At the first measurements
of the protocol, the average minute ventilation was 7.4 ⫾ 1.7
l/min, alveolar ventilation was 5.7 ⫾ 1.4 l/min, and tidal
volume was 690 ⫾ 140 ml (all volumes BTPS). The stability of
physiological measurements during the course of the study
protocol is summarized in Table 2. The variability, expressed
as the mean of all intraindividual coefficients of variation, was
less than 15% for all parameters. The lowest pulse oximetry
reading recorded during Technegas administration was 95%.
SPECT
Subjects received 86–105 MBq (mean 96 MBq) of 113mIn,
corresponding to 270,000–360,000 LyoMAA particles. TechTable 2. Stability of physiological parameters
during experiments
Parameter
Mean Value
Mean CV, %
HR, beats/min
SBP, mmHg
V̇E, l/min
V̇A, l/min
VT, ml
RR, breaths/min
PaO2, Torr
PaCO2, Torr
74
130
7.7
5.9
710
11
99
38
8.6
4.4
10.3
9.9
10.6
13.2
2.4
2.7
The parameters are those that were measured repeatedly. The mean value is
the average of all the intraindividual means. Likewise, the coefficient of
variation (CV) is the mean of the intraindividual coefficients. All ventilatory
volumes are in BTPS.
J Appl Physiol • VOL
Fig. 3. Comparison of measured and SPECT-estimated arterial PCO2 (PaCO2)
and PO2 (PaO2), plot according to Bland and Altman (5). X-axis: mean of
measured and estimated arterial blood gases. Y-axis: difference between
measured and estimated arterial blood gases. Solid lines mark mean differences
between measured and estimated gases; dashed line corresponds to a perfect
agreement between measured and estimated arterial blood gases.
96 • MARCH 2004 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 14, 2017
Statistics
negas breathing required 120–387 s (mean 208 s). There was a
good agreement between the volume of SPECT lung masks
(mean 3,184 ml, range 2,215–4,237 ml) and measured FRC in
the sitting posture (mean 3,219 ml, range 2,150–4,180 ml).
Unlike the FRC, the SPECT lung mask does not include the
anatomic dead space. The volumes also differ in that the
SPECT lung mask represents not only the gas volume in the
lungs but also the total volume of the lungs, including tissue
and the intravascular blood volume. The number of voxels
used for the gas-exchange calculations ranged from 751 to
1,476 voxels/subject. Average radioactivity counts per voxel
varied from 1,383 to 6,148 for indium and for technetium from
5,835 to 26,954.
Physiological SPECT evaluation. The precision of the arterial blood gases predicted from regional ventilation and perfusion as measured with the SPECT method is shown in Fig. 3.
Mean SPECT-estimated PaO2 was 107 Torr, and mean SPECTestimated PaCO2 was 37 Torr. Corresponding measured values
for PaO2 and PaCO2 were 100 and 38 Torr, respectively. The
agreements between measured and SPECT-derived inert gas
retentions and excretions are shown in Figs. 4 and 5. The
correlation between measured and calculated excretions for all
subjects and all inert gases grouped together was 0.99 (Fig. 4).
The correlation between measured and calculated retentions of
the inert gases was equally high, 0.99 (Fig. 5). In none of the
subjects did SPECT data demonstrate blood flow to regions
without ventilation (shunt) nor ventilation of regions without
blood flow (alveolar dead space). Analysis of the measured
MIGET data demonstrated no or insignificant shunt in eight
subjects. In one subject, the estimated shunt corresponded to
7% of cardiac output. Compared with the SPECT measurements, the MIGET method demonstrated a larger fraction
of the total ventilation going to regions with very high ventilation-to-perfusion ratio (V̇A/Q̇), i.e., dead space. The difference between the SPECT and MIGET methods in this regard
was expected because the SPECT method only identifies alveolar dead space, whereas the MIGET method identifies total
dead space. The heterogeneity of SPECT-estimated regional
ventilation and perfusion was less than the heterogeneity of the
MIGET V̇A/Q̇ distributions (Table 3). An example of measured
MIGET result and MIGET results derived from the SPECTestimated V̇A/Q̇ distribution from one subject is presented in
Fig. 6.
REGIONAL VENTILATION AND PERFUSION BY QUANTITATIVE SPECT
1131
Fig. 4. Comparison of measured and SPECTestimated retentions of MIGET gases. A: linear
correlation. Solid line represents the identity
line. B: Bland-Altman plot (5). Solid line marks
mean difference between measured and SPECTestimated retentions.
DISCUSSION
Limitations of the Study Design
In this study, our objective was to develop and evaluate a
method to measure regional ventilation and perfusion in humans. Validating or proving measurements of regional ventilation and perfusion require gold standards for comparison.
Lacking an applicable gold standard, we have employed a
strategy comparing multiple indirect assessments that are each
“necessary but insufficient” to validate our method. A further
limitation of the present study is that only healthy subjects
were studied. In normal lungs, the heterogeneity of regional
ventilation and perfusion distributions is small, and the correlation between these distributions is high. Gas exchange estimated from these distributions is, therefore, mostly dependent
on the measurement of global alveolar ventilation and global
lung perfusion, with regional ventilation and perfusion data
Fig. 5. Comparison of measured and SPECTestimated excretions of MIGET gases. Measured
excretions corrected for the effect of anatomic
dead space. A: linear correlation. Solid line represents the identity line. B: Bland-Altman plot
(5). Solid line marks mean difference between
measured and SPECT-estimated excretions.
J Appl Physiol • VOL
96 • MARCH 2004 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 14, 2017
The principal finding of this study is the good agreement
between measured gas exchange and gas exchange predicted
from SPECT-derived estimations of regional ventilation and
perfusion. This supports the use of the SPECT method to
measure regional perfusion and ventilation in small volume
units of lung in healthy human volunteers. Several methodological aspects of our study deserve comments. The following
discussion details the limitations of the study design, limitations of the SPECT method, and differences between the
SPECT and MIGET methods.
making little contribution to the results of the calculations. In
fact, gas exchange calculated by using a single-compartment
model, i.e., using only global measurements of cardiac output
and minute ventilation, predicts arterial blood gases and inertgas retentions and excretions very similar to those predicted by
the SPECT method. Although the ability to predict whole lung
gas exchange supports the accuracy of our measurements, it
cannot prove that they are correct. However, if we could not
predict gas exchange, it would prove that our measurements
are not valid. Accurate prediction of arterial blood gases from
SPECT studies of patients with abnormal ventilation and perfusion distributions, producing gas-exchange abnormalities,
would provide stronger evidence that our method accurately
estimates regional ventilation and perfusion. Despite that, we
considered it reasonable to include only healthy volunteers in
this first study of a new method. Our results include multiple
measures of ventilation and perfusion distributions and of
whole lung gas exchange. Neither alone can validate our
methods, but the fact that all are in agreement is evidence that
our method may have merit. Prior validations of new imaging
methods for regional ventilation and perfusion have solely
compared their distributions with previous results with similar
techniques. Other than Tokics et al. (51), we believe this is the
only other study to attempt validation of an imaging technique
with independent methodologies.
Limitations of the study protocol. Ideally, physiological
measurements and MIGET should have been performed simul-
1132
REGIONAL VENTILATION AND PERFUSION BY QUANTITATIVE SPECT
Table 3. Characteristics of the V̇/Q̇ distributions
Mean V̇A/Q̇ of Q̇
log SDQ̇
Qs/Qt
Mean V̇A/Q̇ of V̇A
log SDv
DISP R-E*
Predicted PaO2
Predicted PaCO2
MIGET
SPECT
1.44⫾0.33
0.62⫾0.41
0.01⫾0.02
1.88⫾0.37
0.37⫾0.12
4.07⫾3.78
89⫾12.6
37⫾3.0
1.38⫾0.34
0.23⫾0.05
0⫾0
1.46⫾0.38
0.27⫾0.07
0.89⫾0.39
107⫾7.0
37⫾2.9
Fig. 6. Results from 1 subject. A: inert-gas retention and excretion estimated from SPECT
data. B: ventilation-to-perfusion ratio (V̇/Q̇) distribution of SPECT-measured regional ventilation and perfusion. C: measured retention and
excretion of inert gases. Measured inert-gas excretion was corrected for anatomic dead space.
D: V̇/Q̇ distribution of regional ventilation and
perfusion derived from measured inert-gas exchange. Dead space estimated by MIGET includes anatomic and apparatus dead space; in
contrast, the SPECT method only estimates alveolar dead space, hence the difference in the
dead space in the V̇/Q̇ distributions.
J Appl Physiol • VOL
96 • MARCH 2004 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 14, 2017
Values are means ⫾ SD. MIGET, parameters derived from the 50-compartment model of the multiple inert-gas elimination technique method. SPECT,
actual ventilation and perfusion distributions as estimated by the singlephoton-emission computed tomography method. Mean V̇A/Q̇ of Q̇, mean of
the perfusion (Q̇) distribution with reference to log V̇A/Q̇; log SDQ̇, standard
deviation of the Q̇ distribution with reference to log V̇A/Q̇; Qs/Qt, shunt blood
flow as fraction of total blood flow (cardiac output); mean V̇A/Q̇ of V̇A, mean
of the V̇A distribution with reference to log V̇A/Q̇; log SDv, standard deviation
of the V̇A distribution with reference to log V̇A/Q̇. DISP R-E*, MIGET
dispersion index, corrected for apparatus and anatomic dead space. The
dispersion index is defined as the root square differences between the retention
and excretion (R-E) for the 6 inert gases. Correction for apparatus and
anatomic dead space is necessary to make it comparable with dispersion index
calculated from SPECT-derived Q̇A/Q̇ distributions. The differences in mean
V̇A/Q̇ are caused by different minute ventilation during MIGET sampling and
Technegas inhalation and by minute ventilation during MIGET, including
anatomic and apparatus dead space ventilation. The dead space fraction of total
ventilation is omitted from the table because the fractions obtained with the
MIGET and SPECT methods are not comparable.
taneously with the administration of the radiotracers. Additionally, a pulmonary catheter providing mixed venous blood
samples would have eliminated the need to measure oxygen
uptake and carbon dioxide elimination. Also, cardiac output
could have been measured simultaneously with MIGET sampling and administration of radiopharmaceuticals. Any error in
these measurements and calculations may contribute to the
differences between measured and calculated values. Because
it is used several times in the calculations, the estimates of
cardiac output have an especially great influence on estimates
of respiratory and inert-gas exchange, and also on measured
MIGET retentions and excretions. Right heart catheterization
was not considered justified in the present study of healthy
volunteers because alternative techniques were available. The
data analysis assumes that cardiac output, oxygen uptake, and
carbon dioxide elimination remained unchanged during the
study. Any deviations from this assumption may reduce the
agreement between measured and SPECT-estimated arterial
blood gases and MIGET indexes. To maintain as stable conditions as possible, the subjects remained seated all through the
first part of the protocol, and each measurement was preceded
by a 5-min period of inactivity. Although not completely
stable, the variability of the physiological measurements during the course of the experiment was relatively low. In animal
studies, regional ventilation and perfusion in the same posture
have been demonstrated to be stable over time (17, 18, 43).
REGIONAL VENTILATION AND PERFUSION BY QUANTITATIVE SPECT
Changes in the distributions of ventilation and perfusion between MIGET measurements and administration of radiotracers are, therefore, unlikely to have influenced the agreement
between measured and estimated values. For the same reason,
we believe that the sequential, rather than simultaneous, administration of the radiotracers did not affect the results.
Limitations of the SPECT Method
J Appl Physiol • VOL
ton (81mKr) in both healthy subjects and patients and found that
the two tracers produced somewhat different ventilation distributions. In contrast, Amis et al. (2), also using planar imaging,
compared the distribution of Technegas with the distribution of
radioactive xenon (133Xe) and found no differences. Hartmann
et al. (22) compared 81mKr and Technegas as the marker of
regional ventilation for SPECT diagnosing pulmonary embolism: in 15 of 92 patients, the 81mKr and Technegas examinations produced different diagnostic conclusions. We are not
aware of any study quantitatively comparing the distribution of
Technegas with that of a true gas at the level of spatial
resolution used in this study. Pellegrino et al. (41) used Technegas and SPECT to study regional airflow limitation during
methacholine-induced bronchoconstriction. In this study,
Technegas deposition in larger airways increased with increasing bronchoconstriction. It is thus possible that in diseased lung
there is a larger discrepancy between the distribution of Technegas and the distribution of regional ventilation. Obviously,
the estimates of regional ventilation obtained with the SPECT
method will suffer from any failure of the deposition of the
Technegas particles to mimic the true distribution of alveolar
ventilation. The duration of the image acquisition also requires
the radiotracer distributions in the lung to be stable over an
extended period of time. The hydrophobic nature of the Technegas particles is claimed to be important to lung imaging,
preventing deposition in the airways and thereby preventing
the distribution to be influenced by mucociliary clearance.
Amis et al. (2) found no change in the distribution of Tc
activity in the lung after Technegas inhalation when imaging
was repeated after 20 min. Again, this evaluation used planar
imaging and comparing lung regions much larger than the
voxels used in the present study. Xu et al. (59) observed an
unintentional contamination of Technegas with Pertechnegas,
which is produced by transformation of Technegas in the
presence of moisture. This is an important observation because
Pertechnegas is rapidly cleared into the blood with a half-life of
⬃10 min. Thus, in quantitative studies using Technegas, the
absence of Pertechnegas needs to be verified. These results
were published after the completion of the present studies. We
have not been able to exclude the possibility that the Technegas
used in this study was contaminated in this manner, but,
because the minimum time from Technegas administration to
image acquisition was 30 min, we believe that any contamination will not have influenced our results.
Differences Between the SPECT and MIGET Methods
Although they are fundamentally different, the MIGET and
SPECT methods share the ability to produce information on
ventilation and perfusion distributions. MIGET uses measured
global gas exchange to estimate the distributions of perfusion
and ventilation to 50 compartments with different V̇A/Q̇. The
method produces no data on regional ventilation and perfusion
because it does not include any spatial information; there is no
defined anatomic counterpart to the 50 V̇A/Q̇ compartments.
MIGET, therefore, cannot be used to explore the factors that
determine the spatial distribution of regional ventilation and
perfusion. In contrast, the SPECT method results in estimates
of regional spatial distributions of ventilation and blood flow.
These data can be transformed to V̇A/Q̇ distributions similar to
those obtained with MIGET. Furthermore, the SPECT method
96 • MARCH 2004 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 14, 2017
Correction for scatter and attenuation. The SPECT method
uses a gamma camera to externally detect photons emitted
from radiotracers within the body. Because of interaction with
matter along its pathway through the body, only part of the
radiation emitted from a certain volume unit of the lung will be
detected by the gamma camera. When two different tracers are
present simultaneously, scattered photons originating from the
isotope with the higher energy adds to the counts detected in
the energy window of the other isotope (known as downscatter). These phenomena must be taken into account when
SPECT is used for quantitative physiological measurements.
The SPECT method used in this study incorporates new methods of correcting for attenuation, scatter, and downscatter. In
phantom studies, these methods have been shown to allow an
accurate retrieval of the true regional radioactivity, with mean
differences between SPECT data from a low-density phantom
and corrected value in a thorax phantom of ⫺0.9% for 113mIn
and ⫺1.9% for 99mTc (47).
Spatial resolution. The resolution of imaging equipment is
characterized by the full-width half-maximum (FWHM). The
FWHM for our SPECT system is 16 mm. This measure means
that 50% of the counts of one lung voxel will be counted in a
voxel of another lung region 16 mm from the first one.
Consequently, the radioactivity measured within one voxel is
not independent of the activity in the surrounding voxels. This
is the partial volume effect. The movements of the lung during
breathing and movements caused by the pulsations of the heart
and the large vessels further reduce the spatial resolution. The
partial volume effect also causes a gradual attenuation of the
radioactivity measured at the lung edges.
Macroaggregates as markers of regional perfusion. With
SPECT, imaging of regional blood flow is accomplished by
microembolization of radionuclide-labeled particles in the arterial pulmonary circulation. It is based on the principle that the
number of particles trapped in a particular lung volume is
proportional to regional blood flow. Both macroaggregates of
albumin and 15-␮m microspheres have been shown to faithfully measure regional pulmonary blood flow compared with
other measurement methods (4, 19, 38).
Technegas as marker of regional ventilation. Technegas is a
dispersion of ultrafine graphite particles labeled with 99mTc
(11). Although determination of the particle size with different
techniques has yielded somewhat different results (10, 11, 26,
27, 33, 34, 48, 49), most authors seem to agree that the
diameter of the majority of particles is ⬍200 nm, a particle size
that is associated with predominating alveolar deposition (7,
50). Isawa et al. (24) reported that 84.9% of Technegas retained in the lung was deposited in alveoli. Several studies
using both planar imaging and SPECT have found minor
discrepancies between the distributions of true radioactive
gases and Technegas. Isawa et al. (25), using planar imaging,
compared the distributions of Technegas and radioactive kryp-
1133
1134
REGIONAL VENTILATION AND PERFUSION BY QUANTITATIVE SPECT
J Appl Physiol • VOL
lung. In summary, because of the differences between the
MIGET and SPECT methods, a perfect agreement between
measured and estimated inert-gas exchange is not to be expected. The agreement will tend to decrease with increasing
mismatch between regional ventilation and perfusion. Compared with MIGET, the SPECT method will underestimate
shunt blood flow, alveolar dead space ventilation, as well as the
width of the V̇A/Q̇ distributions. In theory, this reasoning
applies to all imaging methods and also to the microsphere
method. With increasing spatial resolution, the differences
compared with MIGET will be less.
Practical Considerations
Approximately 3–5 h were required to complete the experimental protocol. The SPECT study in itself required 1–2 h
(most studies were completed in 1 h). Thus in future studies
1–2 h should be adequate for each experiment, if relative
quantitative measures of regional ventilation and perfusion are
considered sufficient. Absolute values of regional ventilation
and perfusion (ml/min) require measurement of anatomic dead
space, which in this study took ⬃10 min to perform, and
measurements of global ventilation and cardiac output. The
time required for the latter measurements, of course, depends
on the chosen methodologies. Our subjects found it somewhat
difficult but still possible to avoid movements during the image
acquisition.
Repeated studies. In this study, the subjects were estimated
to receive an effective dose of 5 mSv. The limits for acceptable
radiation exposure of healthy individuals vary worldwide. The
radiation dose used in this study is of the same magnitude as
the effective annual dose from radon and background radiation
for the average Swedish individual and is accepted by the local
Radiation Safety Committee for studies of volunteers and
patients; 5 mSv is about one-half of what is obtained from a
clinical computed tomography (CT) examination to diagnose
pulmonary embolism. In the present examinations, the signalto-noise ratio was adequate, indicating that a reduction of the
administered radioactivities is possible. Reducing the doses by
50% would probably not distort the information too much, but
allow for a repeated examination to study some clinical or
physiological variable without increasing the radiation dose.
However, the second examination should not be performed
until after at least 2 days because of the physical half-life of 6 h
for 99mTc.
Comparison With Other Imaging Techniques
Jones et al. (28) used electron-beam CT to image regional
perfusion. Xenon-enhanced CT has been used to assess regional ventilation (20, 37), and lately Kreck et al. (31) used
sequential CT imaging during xenon washin to estimate both
regional ventilation and perfusion. Methods using CT have a
spatial and temporal resolution that is superior to that of
SPECT. Positron emission tomography (PET) has been used in
several studies of regional ventilation and perfusion (8, 9, 40,
52). PET has a spatial resolution that is two to three times finer
than the SPECT method. Finally, functional lung imaging
utilizing magnetic resonance imaging (MRI) is presently being
developed (30, 53). MRI combines the advantages of a spatial
and temporal resolution similar to CT without using ionizing
radiation. The primary advantage afforded by our new SPECT
96 • MARCH 2004 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 14, 2017
can be used to identify anatomic lung regions with V̇A/Q̇
mismatch. We believe this spatial information to be crucial if
the goal is not only to describe but also explain the cause of
gas-exchange impairment or differences in gas exchange in
different situations. For example, localizing blood flow through
low V̇A/Q̇ regions might provide insights into mechanisms
causing difficulties in oxygenation.
The SPECT method did not detect any intrapulmonary shunt
in any of the subjects, whereas the MIGET results indicated
that there was a small shunt in several of the subjects. These
discrepancies could be explained by the fundamental differences between the methods. The shunt flow measured by the
MIGET method corresponds to the sum of all blood flow
through nonventilated lung regions. This blood flow might be
localized to a single lung region, or it could be dispersed to any
number of different regions. The SPECT method in contrast
would only detect shunt blood flow if it corresponds to a lung
region of a certain size. This size depends on the spatial
resolution of the method; for our system, it would correspond
to a region larger than the voxel size used in this study. Shunt
flow through regions of a smaller size would be detected as low
V̇A/Q̇ regions by the SPECT method, whereas the MIGET
method would still produce an estimate of the total blood flow
through nonventilated lung regions. In fact, from this study, we
have no proof that the SPECT method is able to identify
intrapulmonary shunt. Thus the MIGET method has the greater
sensitivity for detecting intrapulmonary shunt. The same principle applies to estimates of alveolar dead space obtained with
the two methods. SPECT-derived estimates of shunt blood
flow and alveolar dead space ventilation will, therefore, be
smaller than MIGET-derived estimates. Because of the finite
spatial resolution of the SPECT method, also V̇A/Q̇ heterogeneity is underestimated, and gas exchange efficiency of the
lung is overestimated. This is demonstrated by our results with
MIGET V̇A/Q̇ distributions being comparable to previous studies of normal subjects (23, 44), whereas SPECT V̇A/Q̇ distributions were narrower (Table 3, Fig. 6). Compared with the
MIGET method, the SPECT method only accounted for 14 and
53%, respectively, of the variance of the perfusion and ventilation V̇A/Q̇ distributions. This suggests that for perfusion
(86%) and for ventilation (47%) of the heterogeneity occurred
at a spatial scale below the resolution of the SPECT method.
The difference in variance is consistent with previous studies
that have demonstrated that the heterogeneity of perfusion is
dependent on the scale of resolution used to measure regional
blood flow (16). Thus whole lung gas exchange predicted from
SPECT V̇A/Q̇ distributions is expected to result in arterial
oxygenation greater than predicted from the MIGET distributions and greater than observed, which is again verified by our
results (Table 3, Fig. 3). The edge effect causes the SPECT
method to underestimate regional ventilation and perfusion in
peripheral parts of the lung, and, as a consequence, values for
central lung regions are systematically overestimated. Even if
this does not affect estimates of regional V̇A/Q̇, it will have an
effect on estimates of regional ventilation and blood flow. The
edge effect will, therefore, affect the arterial blood gases and
inert-gas exchange calculated from SPECT-estimated regional
ventilation and perfusion. Another consequence of the edge
effect is that it imposes a correlation between SPECT-estimated regional ventilation and perfusion because both are
attenuated to similar degree in the most peripheral parts of the
REGIONAL VENTILATION AND PERFUSION BY QUANTITATIVE SPECT
ACKNOWLEDGMENTS
We gratefully acknowledge the dedication of our subjects and the excellent
technical support from Ann-Marie Danielsson, Annette Ebberyd, Marie Finnbogason, Gunilla Fyhr, Ingeborg Gottlieb-Inacio, Bo Tedner, and Jan-Olov
Thorell. We are also most grateful to William Altemeier for providing the
software used for calculating gas exchange from regional ventilation and
perfusion data.
GRANTS
This research was supported by Swedish Medical Research Council Grant
K2003-74X-10401-11A, the Swedish Heart and Lung Foundation Grant
200141470, and by Linde Healthcare AGA.
REFERENCES
1. Altemeier WA, Robertson HT, and Glenny RW. Pulmonary gasexchange analysis by using simultaneous deposition of aerolized and
injected microspheres. J Appl Physiol 85: 2344–2351, 1998.
2. Amis TC, Crawford AB, Davison A, and Engel LA. Distribution of
inhaled 99mtechnetium labelled ultrafine carbon particle aerosol (Technegas) in human lungs. Eur Respir J 3: 679–685, 1990.
3. Anthonisen NR, Dolovich MB, and Bates DV. Steady state measurement
of regional ventilation to perfusion ratios in normal man. J Clin Invest 45:
1349–1356, 1966.
4. Beck KC. Regional trapping of microspheres in the lung compares well
with regional blood flow. J Appl Physiol 63: 883–889, 1987.
5. Bland JM and Altman DG. Statistical methods for assessing agreement
between two methods of clinical measurement. Lancet 1: 307–310, 1986.
6. Bonde-Petersen F, Norsk P, and Suzuki Y. A comparison between freon
and acetylene rebreathing for measuring cardiac output. Aviat Space
Environ Med 51: 1214–1221, 1980.
7. Brain JD and Valberg PA. Deposition of aerosol in the respiratory tract.
Am Rev Respir Dis 120: 1325–1373, 1979.
J Appl Physiol • VOL
8. Brudin LH, Rhodes CG, Valind SO, Jones T, and Hughes JM.
Interrelationships between regional blood flow, blood volume, and ventilation in supine humans. J Appl Physiol 76: 1205–1210, 1994.
9. Brudin LH, Rhodes CG, Valind SO, Jones T, Jonson B, and Hughes
JM. Relationships between regional ventilation and vascular and extravascular volume in supine humans. J Appl Physiol 76: 1195–1204, 1994.
10. Burch WM, Boyd MM, and Crellin DE. Technegas: particle size and
distribution. Eur J Nucl Med 21: 365–367, 1994.
11. Burch WM, Sullivan PJ, and McLaren CJ. Technegas—a new ventilation agent for lung scanning. Nucl Med Commun 7: 865–871, 1986.
12. Douglas AR, Jones NL, and Reed JW. Calculation of whole blood CO2
content. J Appl Physiol 65: 473–477, 1988.
13. Fowler WS. Lung function studies. II. The respiratory dead space. Am J
Physiol 154: 405–416, 1948.
14. Gale GE, Torre-Bueno JR, Moon RE, Saltzman HA, and Wagner PD.
Ventilation-perfusion inequality in normal humans during exercise at sea
level and simulated altitude. J Appl Physiol 58: 978–988, 1985.
15. Glazier JB and DeNardo GL. Pulmonary function studied with the
xenon-133 scanning technique. Normal values and a postural study. Am
Rev Respir Dis 94: 188–194, 1966.
16. Glenny RW, Bernard SL, and Robertson HT. Pulmonary blood flow
remains fractal down to the level of gas exchange. J Appl Physiol 89:
742–748, 2000.
17. Glenny RW, McKinney S, and Robertson HT. Spatial pattern of
pulmonary blood flow distribution is stable over days. J Appl Physiol 82:
902–907, 1997.
18. Glenny RW, Polissar NL, McKinney S, and Robertson HT. Temporal
heterogeneity of regional pulmonary perfusion is spatially clustered.
J Appl Physiol 79: 986–1001, 1995.
19. Glenny RW and Robertson HT. Fractal properties of pulmonary blood
flow: characterization of spatial heterogeneity. J Appl Physiol 69: 532–
545, 1990.
20. Gur D, Drayer BP, Borovetz HS, Griffith BP, Hardesty RL, and
Wolfson SK. Dynamic computed tomography of the lung: regional
ventilation measurements. J Comput Assist Tomogr 3: 749–753, 1979.
21. Harf A, Pratt T, and Hughes JM. Regional distribution of V̇A/Q̇ in man
at rest and with exercise measured with krypton-81m. J Appl Physiol 44:
115–123, 1978.
22. Hartmann IJ, Hagen PJ, Stokkel MP, Hoekstra OS, and Prins MH.
Technegas versus 81mKr ventilation-perfusion scintigraphy: a comparative
study in patients with suspected acute pulmonary embolism. J Nucl Med
42: 393–400, 2001.
23. Hedenstierna G. Ventilation-perfusion relationships during anaesthesia.
Thorax 50: 85–91, 1995.
24. Isawa T, Teshima T, Anazawa Y, Miki M, and Motomiya M. Technegas for inhalation lung imaging. Nucl Med Commun 12: 47–55, 1991.
25. Isawa T, Teshima T, Anazawa Y, Miki M, and Soni PS. Technegas
versus krypton-81m gas as an inhalation agent. Comparison of pulmonary
distribution at total lung capacity. Clin Nucl Med 19: 1085–1090, 1994.
26. Jackson P, Baker RJ, McCulloch DG, Mackey DW, van der Wall H,
and Willett GD. A study of Technegas employing X-ray photoelectron
spectroscopy, scanning transmission electron microscopy and wet-chemical methods. Nucl Med Commun 17: 504–513, 1996.
27. Jackson P, Mackey D, and Van der Wall H. Physical and chemical
nature of technegas. J Nucl Med 39: 1646–1649, 1998.
28. Jones AT, Hansell DM, and Evans TW. Pulmonary perfusion in supine
and prone positions: an electron-beam computed tomography study.
J Appl Physiol 90: 1342–1348, 2001.
29. 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.
30. Kauczor HU, Hanke A, and Van Beek EJ. Assessment of lung ventilation by MR imaging: current status and future perspectives. Eur Radiol
12: 1962–1970, 2002.
31. Kreck TC, Krueger MA, Altemeier WA, Sinclair SE, Robertson HT,
Shade ED, Hildebrandt J, Lamm WJ, Frazer DA, Polissar NL, and
Hlastala MP. Determination of regional ventilation and perfusion in the
lung using xenon and computed tomography. J Appl Physiol 91: 1741–
1749, 2001.
32. Lavender JP, Al-Nahhas AM, and Myers MJ. Ventilation perfusion
ratios of the normal supine lung using emission tomography. Br J Radiol
57: 141–146, 1984.
96 • MARCH 2004 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 14, 2017
method over prior methods is the ability to analyze simultaneous distributions of ventilation and perfusion for whole lung.
Another advantage is that the radiopharmaceuticals can be
administered outside the camera, which means that the distribution of ventilation and perfusion during conditions different
from those during imaging (e.g., a different posture) can be
studied. In this situation, the interpretation of regional distributions is complicated by any changes in the distribution of
lung tissue within the thorax. It is also an advantage that
SPECT imaging does not require any special respiratory maneuver, which might affect both the distribution of both regional ventilation and perfusion. The disadvantages of the
SPECT method are mostly related to the spatial resolution,
which is lower than for some of the other methods, and the
partial volume effect. Time required for image acquisition is
longer for SPECT than for CT- and PET-based methods. The
radiation dose is an obvious disadvantage, which limits the
number of examinations that can be performed in the same
individual. Although some PET techniques use a radiation dose
much lower than the doses used in this study (40), dose
restriction applies to a similar or even greater extent to other
CT and PET techniques.
In this study, we found a good agreement between measured
gas exchange and estimates of regional and whole lung gas
exchange from SPECT measurements. Despite the limitations
of our study design and the SPECT method, we remain optimistic about the potential for this new method to provide new
insights into the mechanisms responsible for the distributions
of regional ventilation and perfusion in humans. In addition,
this method offers a tool for measuring regional ventilation and
perfusion under physiological conditions not amenable for CT,
PET, and MRI methods.
1135
1136
REGIONAL VENTILATION AND PERFUSION BY QUANTITATIVE SPECT
J Appl Physiol • VOL
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
pulmonary tissue volume, and cardiac output measured by a rebreathing
technique. Am Rev Respir Dis 111: 157–165, 1975.
Sánchez-Crespo A, Petersson J, Nyren S, Mure M, Glenny RW,
Thorell JO, Jacobsson H, Lindahl SG, and Larsson SA. A novel
quantitative dual-isotope method for simultaneous ventilation and perfusion lung SPET. Eur J Nucl Med Mol Imaging 29: 863–875, 2002.
Senden TJ, Moock KH, Gerald JF, Burch WM, Browitt RJ, Ling CD,
and Heath GA. The physical and chemical nature of technegas. J Nucl
Med 38: 1327–1333, 1997.
Strong JC and Agnew JE. The particle size distribution of technegas and
its influence on regional lung deposition. Nucl Med Commun 10: 425–430,
1989.
Stuart BO. Deposition of inhaled aerosols. Arch Intern Med 131: 60–73,
1973.
Tokics L, Hedenstierna G, Svensson L, Brismar B, Cederlund T,
Lundquist H, and Strandberg A. V̇/Q̇ distribution and correlation to
atelectasis in anesthetized paralyzed humans. J Appl Physiol 81: 1822–
1833, 1996.
Treppo S, Mijailovich SM, and Venegas JG. Contributions of pulmonary perfusion and ventilation to heterogeneity in V̇A/Q̇ measured by PET.
J Appl Physiol 82: 1163–1176, 1997.
Uematsu H, Levin DL, and Hatabu H. Quantification of pulmonary
perfusion with MR imaging: recent advances. Eur J Radiol 37: 155–163,
2001.
Wagner PD, Naumann PF, and Laravuso RB. Simultaneous measurement of eight foreign gases in blood by gas chromatography. J Appl
Physiol 36: 600–605, 1974.
Wagner PD, Saltzman HA, and West JB. Measurement of continuous
distributions of ventilation-perfusion ratios: theory. J Appl Physiol 36:
588–599, 1974.
Watanabe N, Shirakami Y, Tomiyoshi K, Oriuchi N, Hirano T,
Higuchi T, Inoue T, and Endo K. Direct labeling of macroaggregated
albumin with indium-111-chloride using acetate buffer. J Nucl Med 38:
1590–1592, 1997.
West JB. Regional differences in gas exchange in the lung of erect man.
J Appl Physiol 17: 893–898, 1962.
West JB and Dollery CT. Distribution of blood flow and ventilationperfusion ratio in the lung, measured with radioactive CO2. J Appl Physiol
15: 405–410, 1960.
Xu JH, Moonen M, Johansson A, and Bake B. Dynamics of ‘Technegas’ deposited in the lung. Nucl Med Commun 22: 383–387, 2001.
96 • MARCH 2004 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 14, 2017
33. Lemb M, Oei TH, Eifert H, and Gunther B. Technegas: a study of
particle structure, size and distribution. Eur J Nucl Med 20: 576–579,
1993.
34. Lloyd JJ, Shields RA, Taylor CJ, Lawson RS, James JM, and Testra
HJ. Technegas and Pertechnegas particle size distribution. Eur J Nucl Med
22: 473–476, 1995.
35. Lumb AB. Nunn’s Applied Respiratory Physiology. Oxford, UK: Butterworth-Heinemann, 2000.
36. Maeda H, Itoh H, Ishii Y, Todo G, Mukai T, Fujita M, Kambara H,
Kawai C, and Torizuka K. Pulmonary blood flow distribution measured
by radionuclide-computed tomography. J Appl Physiol 54: 225–233, 1983.
37. Marcucci C, Nyhan D, and Simon BA. Distribution of pulmonary
ventilation using Xe-enhanced computed tomography in prone and supine
dogs. J Appl Physiol 90: 421–430, 2001.
38. Melsom MN, Flatebo T, Kramer-Johansen J, Aulie A, Sjaastad OV,
Iversen PO, and Nicolaysen G. Both gravity and non-gravity dependent
factors determine regional blood flow within the goat lung. Acta Physiol
Scand 153: 343–353, 1995.
39. Milic-Emili J, Henderson JAM, Dolovich MB, Trop D, and Kaneko K.
Regional distribution of inspired gas in the lung. J Appl Physiol 21:
749–759, 1966.
40. Musch G, Layfield JD, Harris RS, 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.
41. Pellegrino R, Biggi A, Papaleo A, Camuzzini G, Rodarte JR, and
Brusasco V. Regional expiratory flow limitation studied with Technegas
in asthma. J Appl Physiol 91: 2190–2198, 2001.
42. Rhodes CG, Valind SO, Brudin LH, Wollmer PE, Jones T, Buckingham PD, and Hughes JM. Quantification of regional V̇/Q̇ ratios in
humans by use of PET. II. Procedure and normal values. J Appl Physiol
66: 1905–1913, 1989.
43. Robertson HT, Glenny RW, Stanford D, McInnes LM, Luchtel DL,
and Covert D. High-resolution maps of regional ventilation utilizing
inhaled fluorescent microspheres. J Appl Physiol 82: 943–953, 1997.
44. Roca J and Wagner PD. Principles and information content of the
multiple inert gas elimination technique. Thorax 49: 815–824, 1993.
45. Rohdin M and Linnarsson D. Differential changes of lung diffusing
capacity and tissue volume in hypergravity. J Appl Physiol 93: 931–935,
2002.
46. Sackner MA, Greeneltch D, Heiman MS, Epstein S, and Atkins N.
Diffusing capacity, membrane diffusing capacity, capillary blood volume,