View
2
Download
0
Category
Preview:
Citation preview
University of Groningen
Quantification and data optimisation of heart and brain studies in conventional nuclearmedicineDobbeleir, André Alfons
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.
Document VersionPublisher's PDF, also known as Version of record
Publication date:2006
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):Dobbeleir, A. A. (2006). Quantification and data optimisation of heart and brain studies in conventionalnuclear medicine. s.n.
CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.
Download date: 11-10-2020
RIJKSUNIVERSITEIT GRONINGEN
Quantification and data optimisation of heart and brain studies in
conventional nuclear medicine
Proefschrift
ter verkrijging van het doctoraat in de
Medische Wetenschappen
aan de Rijksuniversiteit Groningen
op gezag van de
Rector Magnificus, dr. F. Zwarts,
in het openbaar te verdedigen op
woensdag 25 januari 2006
om 14.45 uur
door
André Alfons Dobbeleir
geboren op 14 september 1949
te Kruibeke, België
2
PROMOTORES :
Prof. dr. R.A.J.O. Dierckx
Prof. dr. H.R. Ham
Prof. dr. A.M.J. Paans
Prof. dr. A-S. E. Hambÿe
BEOORDELINGSCOMMISSIE :
Prof. dr. W. Vaalburg
Prof. dr. J.H.A. De Keyser
Prof. dr. Ph. Franken
ISBN-10 : 9090198687
ISBN-13 : 9789090198682
3
Paranimfen : Ann Vervaet
Gerda Dobbeleir
The publication of this thesis was supported by a grant from Amersham-Health (part of GE
Healthcare) Belgium.
4
Stellingen
Behorende bij het proefschrift :
Quantification and data optimisation of heart and brain studies in conventional nuclear medicine.
1. De analyse van gated SPECT vervolg studies bij een individuele patient dienen gedaan te
worden met dezelfde kwantitatieve gated SPECT software.
2. Snelle, herhaalde en reproduceerbare hoge telsnelheid linker ventriculaire functie studies
kunnen verkregen worden met het kortlevende radionuclide 191m
Ir.
3. Zonder scatter correctie wordt myocardiale mismatching tussen BMIPP en MIBI gedeeltelijk
verborgen.
4. De SPECT kwantitatieve bevindingen met HMPAO wijzen erop dat in DAT patienten de
perfusie van het cerebellum onaangetast is waardoor deze regio kan gebruikt worden als
referentie.
5. Indien berekend met een resolutie onafhankelijke methode kunnen dopamine transporters
resultaten met 123
I-FP-CIT afkomstig van verschillende centra en systemen met elkaar
vergeleken worden.
6. Zelfs na geschikte scatter en attenuatiecorrectie, blijven de problemen verbonden aan de
beperkte resolutie van het SPECT systeem onopgelost voor accurate kwantificatie. P. Jarritt
and K. Kouris. New trends in nuclear neurology and psychiatry 1993: p57.
7. Wij zijn allen voor vereenvoudiging en standaardisering maar de geschiedenis van nucleaire
geneeskunde leert ons dat dit zelden leidt naar absolute kwantificatie. A. Gottschalk.
Yearbook of Nuclear Medicine 1995: p275.
8. Statistiek is als een lantaarnpaal voor een dronken man, meer ter ondersteuning dan ter
verlichting.
9. Zoals dit proefschrift is creativiteit een langzaam proces.
10. 2005 is het jaar van de natuurkunde; daarom is Fietsica dit jaar uitgeroepen tot een belangrijke
activiteit in Groningen.
11. Alleen wie tegen de stroom inzwemt komt aan de bron. (Volgens A. Vervaet en G. Dobbeleir
ook van toepassing op A. Dobbeleir)
5
Contents
1. Introduction and outline of the thesis
2. Determination of left ventricular ejection fraction by first pass and gated SPECT
studies.
2.1 Performance of a single crystal digital gamma camera for first pass cardiac studies.
2.2 Variability of left ventricular ejection fraction and volumes by quantitative gated
SPET : influence of algorithm, pixel size and reconstruction parameters in normal
and small-sized hearts.
2.3 Clinical applications of first pass studies (abstracts of articles)
-Clinical usefulness of ultrashort-lived Iridium-191m from a carbon-based
generator system for the evaluation of the left ventricular function.
-Comparison between exercise myocardial perfusion and wall motion using 201
Tl
and 191m
Ir simultaneously.
3. Myocardial perfusion and viability.
3.1 SPET generated colour-coded polar maps to quantify the uptake of 99mTc-
sestaMIBI and 123I-BMIPP in chronically dysfunctional myocardium: comparison
with coronary anatomy and wall motion.
3.2 Influence of high-energy photons on the spectrum of iodine-123 with low-and
medium-energy collimators: consequences for imaging with 123I- labelled
compounds in clinical practice.
3.3 Influence of methodology on the presence and extent of mismatching between
99mTc-MIBI and 123I-BMIPP in myocardial viability studies.
3.4 Clinical applications ( abstracts of articles)
-BMIPP imaging to improve the value of sestamibi scintigraphy for predicting
functional outcome in severe chronic ischemic left ventricular dysfunction.
-Quantification of 99mTc-sestaMIBI and 123I-BMIPP uptake for predicting
functional outcome in chronically ischaemic dysfunctional myocardium
-Prediction of functional outcome by quantification of sestamibi and BMIPP after
acute myocardial infarction.
4. Perfusion of the brain.
4.1 Quantification in SPECT using non-invasive methods.
4.2 Quantification of technetium-99m hexamethylpropylene amine oxime brain uptake
in routine clinical practice using calibrated point sources as an external standard:
phantom and human studies.
4.3 Clinical applications ( abstracts of articles)
6
-Parameters influencing SPET regional brain uptake of technetium-99m
hexamethylpropylene amine oxime measured by calibrated point sources as an
external standard.
-Validation of the cerebellum as a reference region for SPECT quantification in
patients suffering from dementia of the Alzheimer type.
5. Dopamine transporter imaging in the human brain.
Quantification of Iodine-123-FPCIT SPECT with a resolution independent technique.
6. Summary and future directions
Samenvatting en toekomstperspectieven
7. List of publications
8. Dankwoord
7
1. Introduction and outline of the thesis
Introduction
After the introduction of planar nuclear medicine images obtained with the Anger gamma camera
(1957) and connection to a computer system (early 1970s), attempts have been made to enhance the
quality of the scintigrams by filtering and noise subtraction(1,2). Using manual or automatic regions
of interest on static or dynamic images, quantitative analysis of the radiopharmaceutical distribution
let to the development of nuclear medicine procedures for many organs. Quantification reduces the
inter- and intra-observer variability and improves the sensitivity and specificity of the nuclear
medicine procedures (3). Using planar imaging, the superimposition of activity in front and behind
the organ results in decreased image contrast and prevents accurate quantitative measurements. At
the end of the 1970s three-dimensional information of the radionuclide distribution in humans was
obtained by means of single photon emission computed tomography (SPECT) (4-6).
The ultimate goal was to quantify the absolute distribution of radioactivity. To achieve this goal
many obstacles need to be overcome, some inherent to the gamma camera and planar acquisition,
others to the tomographic reconstruction (7-11). In SPECT, photon absorption and scatter,
particularly in the chest, produce regional inhomogeneities. Poor attenuation maps and
misalignment between transmission and emission data also influence quantitative measurements
(12). Step by step, solutions appear in literature and the new generation gamma cameras permit
correction for photon absorption and scatter (13). In spite of all that, some physical properties
hamper accurate quantification.
In the next paragraph physical parameters of the gamma camera are summarised. Then their impact
on different quantification methods is highlighted. Finally in the aim and outline of the thesis, the
application of correction methods or alternative approaches necessary for quantification of heart and
brain studies is mentioned.
Physical properties
Calibration
The first requirement of an imaging system is that the image of an object is independent of its
position in the field of view. Originally this is not the case due to impurities in the crystal and
variations in the response of the photomultipliers, affecting both the energy estimate and event
localisation. Variations of energy, non-linearity and uniformity can be corrected by calibration
measurements. Tomographic reconstruction needs additional calibration for the centre of rotation of
the detectors (14-16).
Scatter
Due to a limited energy resolution, usually a 20% energy resolution is used. Therefore, scattered
events can amount to 20% in a typical brain study and even to 40 % of the total counts in a cardiac
study. The nature of scatter is thus study dependant in a complex manner both on the composition
of the patient, the distribution of the tracer and the collimator and detector characteristics. Scatter is
nonstationary. In SPECT, these scattered events must be removed before attenuation correction or
in some cases a reduced linear attenuation coefficient can be used (17-19).
Dead-time
The sensitivity of an imaging system is defined by the number of counts per unit time detected by
the device for a unit activity in the source. Only a fraction of the photons passes through the
collimator and is absorbed by the crystal. Some of these events are rejected depending of the setting
8
of the photopeak window. Some of the events are lost because the system is still busy processing a
previous event. The probability of this situation increases with higher activities. As a result of the
dead time of the system, the detected counts do not rise linearly with the activity. With further
increase of activity, a saturated stage and even drop in the detected count rate may be observed
(20,21).
Partial volume effect
Planar and SPECT images have a characteristic resolution. Images of objects larger than 2x the full
width at half maximum of the point spread function will reflect both the size and radioactive
concentration of the object. However for smaller objects, the signal is blurred; so that the total
counts is preserved although the activity per pixel is decreased. This effect, known as partial
volume effect, is particularly severe for SPECT images of the brain. Textures are below the
resolving power of the methodology and accurate determination of radioactive concentrations is
impossible (22-24).
Acquisition Parameters
The acquisition matrix must be chosen in function of the detector size, and the obtained pixel size
defines the spatial sampling. Optimally, the pixel size should be less than the FWHM / 3 e.g. about
3-4 mm for a SPECT system characterized by a 10-12 mm FWHM resolution. The angular sample
is defined by the number of projections in 360°. In order to ensure similar spatial and angular
sampling for the reconstruction region, the angular interval should be such that the arch length is
equal to the spatial sampling interval. For the circumference of the brain, optimal angular sampling
interval should be about 3°. In cardiac studies, due to larger detector distance the FWHM is higher
and the heart is more in the centre of the reconstructed volume, so angular sampling between 4° and
6° is used (25-27).
Reconstruction parameters
Reconstructing the angular images by filtered backprojection needs filtering by a Ramp filter to
correct for image blurring. This enhances however the high spatial frequency noise in the
reconstructed image. To suppress this noise, filters with specific parameters defining the degree of
smoothing of the image are used. This reduces the information in the reconstructed image (27-29).
Quantification
Quantification of radioactive tracer concentrations depends and is limited by previous mentioned
factors. Quantification analysis can be subdivided in three subclasses: the measurement of size and
volume of features within the image, the relative activity concentrations within regions and the
absolute tracer concentration in units of MBq/ml.
Size and volume
Size measurements always require some kind of contour definitions. Accurate size measurements
are limited by the finite resolution of the system and the statistical errors in the reconstruction. The
accuracy of the edge detection algorithms depends on the signal-to-noise ratio and the contrast
range within the image. The partial volume effect defined by the finite resolution of the imaging
process requires a different threshold for different sized structures. System and user defined
parameters influence size measurements. The user can effect the final resolution of the image by the
choice of the collimator, the acquisition pixel size and the reconstruction filter.
9
Relative concentration
Quantification comparing total activity or concentrations within several regions is the most
common method in nuclear medicine procedures. The amount of activity can also be expressed
relative to the maximum activity or mean activity in the image or the study. Reference anatomic
regions or control subjects are used. The choice of region of interest size and placement, the
adequate definition of anatomical regions and the identification of the reference site are critical.
Taking care, data obtained within one centre or between centres with identical equipment for
acquisition and processing might be compared. For centres using different detector systems with
different resolutions, different acquisitions protocols and reconstruction methods, results will
almost certainly be different.
Comparing different activities supposes a uniform sensitivity of the imaging system. This is
achieved by calibration. At high count rate activity linearity during dynamic acquisitions is no
longer present and dead time corrections are obliged.
In myocardial perfusion studies profiles normalised to the maximum activity or the most normal
region can be compared with those of normal subjects. The localisation, extent and severity of a
defect can be calculated and compared between rest and stress studies.
In myocardial viability studies, the profile of perfusion with Tc99m-sestaMIBI can compared to the
profile obtained with I123-BMIPP. Even acquired separately a supplementary problem rises. In the
photo-peak Iodine-123 considerably higher scatter is measured from high-energy photons.
Moreover, the distribution of the tracers are different. The nature of scatter depends in a complex
manner both on the distribution of the tracer and the collimator and detector characteristics. Scatter
is non-stationary and adds background activity in the myocardial profile. Scatter correction must be
applied for quantitative comparison of these different isotopes. Absolute quantification
The elusive but ultimate goal of quantification in nuclear medicine is the measurement of absolute
tracer concentration in units of MBq/ml or in % of the injected dose. This would take in account all
centre specific problems and data become non-centre specific. Absolute quantification performed in
a specific centre can also be used to prove that an anatomic region remains stable within different
patients, groups or treatment and thus can be used further on as reference region in relative
quantitative measurements.
Although not totally accurate, absolute quantification might also be a supplementary tool for studies
where small organs are involved and relative quantification is hampered by a huge partial volume
effect.
Aim of the thesis
The aim of this work was to obtain accurate quantitative measurements useful in clinical practice in
some heart and brain nuclear medicine procedures, applied in our department.
In the development of each method several of the following steps have to be covered:
- investigate the appropriate physical characteristics
- optimise and / or correct for physical characteristics
- figure out a practical method useful for clinical practice
- determine the accuracy of the method
- apply the method in patient studies
10
Outline of the thesis.
Chapter 2. Determination of left ventricular ejection fraction by first pass and gated SPECT
studies.
2.1 At high count rate activity linearity during dynamic acquisitions is no longer present and
dead time corrections are required. The performance of a single crystal digital gamma camera was
studied for the evaluation of the left ventricular function.
Ultrashort-lived Iridium-191m permitted rapid, repeat first pass studies.
2.2 System and user defined parameters influence size measurements. The variability of left
ventricular ejection fraction and volumes calculated by quantitative gated SPECT modifying the
acquisition pixel size and the reconstruction filter was measured. The impact on normal and small-
sized hearts calculated by different algorithms on several processing stations was studied.
2.3 First pass studies were applied in patients at increasing levels of exercise. Exercise
myocardial perfusion and wall motion using 201
Tl and 191m
Ir simultaneously was studied.
Chapter 3. Myocardial perfusion and viability of the heart
3.1.1 In myocardial studies profiles are normalised to the maximum activity or the most normal.
We generated colour-coded polar maps to quantify the uptake of 99mTc-sestaMIBI and 123I-
BMIPP in chronically dysfunctional myocardium. The difference in extent and severity of a defect
was compared with coronary anatomy and wall motion.
3.2.1 In the photo-peak of Iodine-123 a considerably higher scatter portion is measured than with
Tc99m-sestaMIBI. The influence of high-energy photons on the spectrum of iodine-123 with low-
and medium-energy collimators is studied and the consequences for imaging with 123I-labelled
compounds in clinical practice discussed.
3.2.2 The influence of methodology on the presence and extent of mismatching between perfusion
using 99mTc-MIBI and metabolism using 123I-BMIPP in myocardial viability studies was
investigated.
3.3 Several clinical applications were published taking into account the previous mentioned
spectral analysis. Comparative quantification of 99mTc-MIBI and 123I-BMIPP tomography
predicted functional outcome in chronically ischaemic dysfunctional myocardium and after acute
myocardial infarction. BMIPP imaging improved the value of sestamibi scintigraphy for predicting
functional outcome in severe chronic ischaemic left ventricular dysfunction.
Chapter 4. Perfusion of the brain
4.1 A review of quantification of brain perfusion and cerebral blood flow was published in a
textbook presenting an up-to-date and systematic approach of SPECT in the major neurological and
psychiatric disorders.
4.2 We calculated the absolute technetium-99m hexamethylpropylene amine oxime (HMPAO)
brain uptake and proved that the cerebellum remains stable within different patients, groups or
treatment and can be used as reference region in relative quantitative measurements.
11
4.3 Parameters influencing the SPECT regional brain uptake of technetium-99m HMPAO were
studied in volunteers and patients. The cerebellum was validated as a reference region for SPECT
quantification in patients suffering from dementia of the Alzheimer type.
Chapter 5. Dopamine transporter imaging in the human brain
In a small organ like the striatum relative quantification is hampered by a huge partial volume
effect. We developed a region of interest independent method. Using gamma camera calibration
factors for the radio-ligand Iodine-123-FPCIT we transformed the striatal uptake in absolute
quantification. Although not totally accurate, absolute quantification might also be a supplementary
tool for inter-centre comparison.
References:
1. Goris ML. Nontarget activities: can we correct for them ? J Nucl Med 1979; 20: 1294,1300.
2. King MA, Schwinger RB, Doherty PW, Penney BC. Two-dimensional filtering of SPECT
images using the Metz and Wiener filters. J Nucl Med 1984; 25: 1234-1240.
3. Berger BC, Watson DD, Taylor GJ et al. Quantitative thallium-201 exercise scintigraphy for
the detection of coronary artery disease. J Nucl Med 1981; 22: 585-593.
4. Jaszczak RJ, Murphy PH, Huard D, Burdine J. Radionuclide emmision computed
tomography of the head with 99m-Tc and a scintillation camera. J Nucl Med 1977; 18: 373-
380.
5. Keyes JW, Orlandea N, Heetderks WJ et al. The Humongotron – a scintillation camera
transaxial tomograph. J Nucl Med 1977; 18: 381-387.
6. Larsson SA. Gamma camera emmision tomography. Acta Radiol 1980; 363: 5-75.
7. Goulding P, Burjan A, Smith R et al. Semi-automatic quantification of regional cerebral
perfusion in primary degenerative dementia using technetium-99m hexamethylpropylene
amine oxime and single photon emmision computer tomography. Eur J Nucl Med 1990; 17:
77-82.
8. Hooper HR, McEwan AJ, Lentle BC et al. Interactive three-dimensional region of intrest
analysis of HMPAO SPECT studies. J Nucl Med 1990; 31: 2046-2051.
9. Eisner RL, Tamas MJ, Cloninger K et al. Norma SPECT thallium-201 bull’s-eye display:
gender differences. J Nucl Med 1988; 29: 1901-1909.
10. Nuyts J, Mortelmans L, Suetens P et al. Model-based quantification of myocardial perfusion
images from SPECT. J Nucl Med 1989; 30: 1992-2001.
11. Garcia EV, Van Train K, Maddahi J et al. Quantification of rotational thallium-201
myocardial tomography. J Nucl Med 1985; 26: 17-26.
12. Ficaro E, Wackers F. Should SPET attenuation correction be
more widely employed in routine clinical practice? Eur J Nucl Med 2002;29: 409-415.
13. Fricke H, Fricke E, Weise R et al. A method to remove artifacts in attenuation-corrected
myocardial perfusion SPECT Introduced by misalignment between emission scan and CT-
derived attenuation maps. J Nucl Med. 2004;45:1619-25.
14. Fahey FH, Harkness BA, Keyes JW et al. Sensitivity, resolution and image quality with a
multi-head SPECT camera. J Nucl Med 1992; 33: 1859-1863.
15. Lim CB, Walker R, Pinkstaff C et al. Triangular SPECT system for 3-D organ volume
imaging: performance results and dynamic imaging capability. IEEE Trans Nucl Sci 1986;
33: 501-504.
16. Cyrill Burger and Gustav K. von Schultness. (1998) Gamma rays: nuclear medicine. In:
Gustav K. von Schultness and Jurgen Hennig (eds) Functional Imaging. Lippincott-Raven
Publishers. (157-216)
12
17. Ljundberg M, Strand SE. Scatter and attenuation correction in SPECT using density maps
and Monte Carlo simulated scatter functions. J Nucl Med 1990; 31: 1560-1567.
18. Ogawa K, Harata Y, Ichihara T et al. A new method for scatter correction in SPECT. Med
Phys 1990; 17: 518.
19. Jaszczak RJ, Floyd CE, Coleman RE. Scatter compensation techniques for SPECT. IEEE
Trans Nucl Sci 1985; 32: 786-793.
20. Ullman V, Husak V, Dubroka L. Deadtime correction in dynamic radionuclides studies by
computer. Eur J Nucl Med 1978; 3: 197-202.
21. Johnston AS, Arnold JE, Pinsky SM. Anger camera deadtime: marker source correction and
two parameter model. J Nucl Med 1975; 16: 539.
22. King MA, Long DT, Brill BA. SPECT volume quantification: influence of spatial
resolution, source size and shape, and voxel size. Med Phys 1991; 18: 1016-1024.
23. Kim HJ, Zeeberg BR, Fahey FB et al. Three-dimensional SPECT simulations of a complex
three-dimensional mathematical brain model and measurements of the three-dimensional
brain phantom. J Nucl Med 1991; 32: 1923-30.
24. Kim HJ, Zeeberg BR, Reba RC. Compensation for three-dimensional detector response,
attenuation and scatter in gray matter imaging using an iterative reconstruction algorithm
which incorporates a high resolution anatomical image. J Nucl Med 1992; 33: 1225-1234.
25. Muehllehner G. Effect of resolution improvement on required count density in ECT
imaging: a computer simulation. Phys Med Biol 1985; 30: 163-173.
26. Mueller SP, Pollak JF, Kijewski MF, Holman BL. Collimator selection for SPECT brain
imaging: the advantage of high resolution. J Nucl Med 1986; 27: 1729-1738.
27. Jarritt P.H. and Kouris K. (1993) Instrumentation for brain SPET: guidelines and
quantification. In: D.C. Costa, G.F. Morgan, N.A. Lassen (eds) New trends in neurology and
psychiatry. John Libbey & Company. (39-62)
28. Lee KH, Liu H, Chen D et al. Volume calculation by means of SPECT: analysis of imaging
acquisition and processing factors. Radiology 1988; 167: 259-262.
29. Blokland K, Reiber H and Pauwels E. Quantitative analysis in single photon emission
tomography (SPET). Eur J Nucl Med 1992;19:47-61.
13
2. Determination of left ventricular ejection fraction by first pass
and gated SPECT studies.
2.1.Performance of a single crystal digital gamma camera for first pass cardiac
studies.
A. Dobbeleir
1, P.R. Franken
1, H.R. Ham
2, C. Brihaye
3, M. Guillaume
3, F.F. Knapp
4 and
J. Vandevivere1
1Division of Nuclear Medicine, Middelheim General Hospital Antwerpen, 2020 Belgium,
2Division
of Nuclear Medicine, St Peter’s Hospital, 1000 Bruxelles, Belgium, 3Cyclotron Research Center,
University of Liège, Belgium, 4Nuclear Medicine Group, Health and Safety Research Division, Oak
Ridge National Laboratory (ORNL), Oak Ridge, TN 37831-0622, USA
Nuclear Medicine Communications 1991; 12: 27-34.
Summary
First pass radionuclide angiocardiography (FPRNA) has gained increasing interest because of the
development of new 99Tc
m-labelled perfusion agents and of new
191Os/
191Irm generator systems. The
aim of the study was to evaluate the performance capacities of a small field of view crystal digital
gamma camera for 99Tc
m and
191Irm at high count rates. The camera dead time for
99Tc
m (window
30%) was well corrected up to 300 kcps in fast acquisition mode using the relative decrease of a
small shielded reference source. Using the decaying activity method for 191
Irm the non-linearity
response of the gamma camera was corrected by an 191
Os reference source up to 210 kcps at 70
keV, 75 kcps at 129 keV and 320 kcps including both peaks. Saturation count rates were
respectively 270 kcps, 150 kcps and 420 kcps and high count rate resolution (FWHM) 9.0, 7.3 and
10.3 mm. Since the accuracy of the first pass measurements is more sensitive to count rate than to
spatial resolution the 50-150 keV window was chosen for clinical studies. In data obtained from 32
ECG gated FPRNA patient studies, the whole field of view count rate during the left ventricular
phase ranged from 100 to 250 kcps with 80 to 120 mCi (2960-4400 MBq) of 191
Irm and 100 to 180
kcps with 20 to 25 mCi (750-925 MBq) of 99Tc
m red blood cells permitting for both tracers accurate
non-linearity correction.
Introduction
First pass radionuclide angiocardiography (FPRNA) has recently gained increasing interest for
measuring left ventricular function. Firstly, because of the availability of new 99Tc
m-labelled
myocardial perfusion agents allowing simultaneous assessment of myocardial perfusion and
function [1, 2]. Secondly, because of the development of new high performance 191
Os/191
Irm
generator systems [3-5] offering the opportunity to conduct rapid, repeat, multiple first pass studies
of the cardiovascular system with the ultrashort half-lived 191
Irm [6-8]. The aim of this study was to
evaluate the performance capacities and the limitations of a single crystal digital gamma camera
(SCDGC) with respect to the high count rates needed for accurate measurements of ventricular
function with the FPRNA method.
14
Materials and methods
191
Irm
191Irm is the daughter of
191Os (
191Os:β-
emission ; T1/2 = 15,4 days) and decays with a half-life of
4,96 s to stable iridium emitting a gamma-ray at 129 keV and three X-rays at 63 keV, 16%; 65 keV,
28%; and 74 keV, 12%. The X-rays cannot be resolved with the Na crystal and appear thus as one
single peak at about 69 keV (Fig 1.). In this study, 191
Irm was produced by elution of a carbon-based
191Os/
191Irm generator system with pH 2, 0,9% NaCl solution containing potassium iodide and
subsequently neutralized with a TRIS buffer. Details concerning the preparation and use of this
generator system have been published elsewhere [3, 8, 9].
Data acquisition
Data were acquired in the ‘normal’ acquisition mode or in the ‘fast’ acquisition mode with a small
field of view (20 cm) SCDGC (APEX 215M, Elscint) equipped with a very high sensitivity, low-
energy, parallel hole collimator. In the ‘fast’ mode, a higher number of counts can be acquired using
a different electronic circuit integrating only the first 400 ns of the scintilation.
Fig. 1. Spectrum of
191Ir
m measured with a gamma camera.
Camera resolution
The resolution of the camera for different energies was tested with 99Tc
m, 201
Tl and 191Os point
sources in the ‘normal’ and in the ‘fast’ acquisition modes [10]. 191
Os decays to 191
Irm by β-
emission without emitting photons. The FWHM, FW20M and FW10M were calculated in a 30%
window centered over the 140 keV 99Tc
m photopeak, in a 40% window centered over the 70 keV
201Tl photopeak and in the 50-100, 100-150 and 50-150 keV windows of the
191Irm spectrum.
Count-rate linearity
The linearity of the gamma camera for 99Tc
m was measured by placing an increasing number of
small vials (1 cm diameter) on the collimator. The activities were measured using a dose calibrator
15
and no scattering material was used. Data were acquired in the ‘fast’ mode with the energy of the
pulse height analyser set at 140 keV with a 30% window. Increasing 99Tc
m activities were measured
together with a 10 kcps activity shielded reference source of 99Tc
m placed in the field of view of the
camera. The measured activity was corrected for the dead time of the camera by a software
correction program based on the detected counts before the pulse height analyser (provided by the
manufacturer) and by the relative decrease of activity of a reference source [11, 12]. The linearity of
the gamma camera for 191
Irm was tested using the decaying source method. For that purpose a bolus
of 100 mCi (3700 MBq) of 191
Irm was extracted from the generator system using 15 ml of normal
saline solution, collected in an extension tube and directly divided through a three-way stopcock
into two 100 ml beakers placed on the collimator. The beakers contained a small quantity of water
in order to obtain a distributed source. A 10 kcps activity shielded 191
Os source was placed on the
camera as reference. Data were acquired in dynamic mode (25 frames per second) for 30 s. This
procedure was repeated three times in the 50-100, 100-150 and 50-150 keV windows. Time-activity
curves were then generated from regions of interest (ROI) drawn over the reference source, over the
two beakers and over a region between the two beakers ROIs, the latter being used to estimate the
relative amount of misplaced pile-up events [13, 14]. The activity curves of the two beakers were
added together and corrected for camera dead time by the relative decrease of the reference source
activity. The linearity response of the gamma camera was then established by comparing the dead
time corrected activity curve to the theoretical decaying curve of 191
Irm. For this purpose, a linear fit
with a slope = -0.140 corresponding to the decay constant of 191
Irm was applied on the low values of
the corrected activity curve expressed in the natural log (Fig. 2). Accepting a 1% deviation as
criteria, the limits of the linearity response of the system was determined for the above-mentioned
windows of the 191
Irm spectrum.
Fig. 2. Linear fit with a slope of –0.140 is applied to the dead time corrected decay curve of
191Ir
m in order to
determine the limits of linearity response of the system.
Patient studies
First pass radionuclide angiocardiographic studies were obtained at rest in 32 patients with 80-120
mCi (2960-4400 MBq) of 191
Irm and a few minutes later with 20-25 mCi (750-925 MBq) of
99Tc
m
red blood cells. Pulse height analyser windows were set over the 50-150 keV windows for 191
Irm
16
studies and over the 119-161 keV windows for the 99Tc
m studies. Data were collected in the ‘fast’
mode in a 32 x 32 x 8 matrix (25 frames per second) for 30 s.
Results
Camera resolution
The FWHM, FW20M and FW10M with 99Tc
m, 201Tl and
191Os point sources are given in table 1.
Table 1. Spatial resolution on point sources of
99Tc
m, 201Tl and
191Ir
m (50-100, 100-150 and 50-150 keV
windows) in the ‘normal’ (N) and ‘fast’ (F) acquisition modes.
191Ir
m
99
Tcm
201Tl 50-100 keV 100-150 keV 50-150 keV
N F N F N F N F N F
FWHM 6.0 7.3 6.5 8.6 6.9 9.0 6.4 7.3 7.7 10.3
FW20M 9.8 11.2 10.3 12.9 10.3 14.2 9.9 12.0 11.2 15.0
FW10M 12.4 14.2 13.3 16.3 12.5 16.3 12.0 14.2 13.3 18.1
As expected, the spatial resolution of the system was better for 99Tc
m than for
201Tl in the ‘normal’
as well as in the ‘fast’ acquisition modes. In the latter, a small but consistent degradation of the
resolution was observed with both isotopes. The resolution of the system for 191
Irm depends
obviously on the window selection. The resolution was similar to that of 201Tl for the 50-100 keV
window (FW20M 10.3 mm versus 10.3 mm) and similar to that of 99Tc
m for the 100-150 keV
window (FW20M 9.9 mm versus 9.8 mm) while the largest window (50-150 keV) gave the largest
FW20M (11.2 mm). Again the ‘fast’ acquisition mode induced a degradation of the spatial
resolution for all energy windows. This influence of window selection on the spatial resolution of
the gamma camera was further observed in clinical studies comparing 191
Irm FPRNA to
99Tc
m. The
left ventricular ROI area averaged 164 ± 29 pixels in 99Tcm studies and 192 ± 28 pixels in the 191Irm studies (P<0.0001).
Camera linearity
Using the relative decrease of the shielded point source activity as a reference, the camera dead time
for 99Tc
m (with a 30% window) was corrected up to 80 kcps in the ‘normal’ acquisition mode and
up to 300 kcps in the ‘fast’ acquisition mode, corresponding to true count rates of about 160 and
650 kcps, respectively. The software correction resulted in systematic undercorrection of the
linearity response of the camera. The saturation count rate for 191
Irm was 270 kcps in the 50-100
keV window, 150 kcps in the 100-150 keV window and 420 kcps in the 50-150 keV window. Using
the decaying source method, the camera dead time was corrected with an error of less than 1% up to
210 kcps in the 50-100 keV window, 320 kcps in the 50-150 keV window, but only up to 75 kcps in
the 100-150 keV window. The number of pile-up events was estimated from the ROI drawn
between the two beakers. At bolus arrival up to 9% of the total measured activity in the camera field
of view was related to misplaced events in the 100-150 keV window compared to 2.5% in the 50-
100 keV window and 5.5% in the 50-150 keV window (Fig. 3). A 1% or less misplaced events were
observed in the 50-100 keV window at the maximal count rate capacity of the camera (270 kcps), in
the 50-150 keV window at 70% (300 kcps) of the maximal capacity, but in the 100-150 keV at only
40% (60 kcps) of the maximal capacity of the camera system.
Patient studies
The highest count rates in the WFOV and in the right and left ventricular ROIs during the first
transit of 191
Irm (50-150 keV window) observed in 3 of the 32 patients are given in table 2. In
Patient 1, although left ventricular count rate was rather low the WFOV count rate during the right
17
phase of the transit was just below the maximal limit of accurate dead time correction of the system.
In Patient 2, and certainly in Patient 3, the count rates during the right transit phase were out of the
limits of dead time correction: the maximum WFOV count rate was reached at 0.3 and 2.6 s,
respectively, after the passage of the bolus in the right ventricle, indicating saturation of the gamma
camera and precluding simultaneous assessment of right and left ventricular function studies during
a single injection of 191
Irm.
Table 2. Maximal count rates during
191Ir
m (50-150 keV window) FPRNA studies in three patients.
FPRNA RV phase LV phase
WFOV RV WFOV LV WFOV
Patient (kcps) (kcps) (kcps) (kcps) (kcps)
Delay
max FPRNA-
max RV phase ( s)
1 306 201 306 51 132 0.0
2 400 278 398 99 212 0.3
3 441 187 380 195 331 2.6
The maximal count rate in the 20 cm field of view of the camera observed in most patients during
the left ventricular transit of the 191
Irm bolus ranged between 100 and 250 kcps. The maximal count
rate observed in those patients during the left ventricular transit of the 99Tc
m bolus ranged between
100 and 180 kcps. Left ventricular counts in the 40 ms end-diastolic image of the ECG-gated left
ventricular representative cycle averaged 14.8 kcounts (range 5.3-30.3) with 191
Irm and 10.2 kcounts
(range 3.6-22.1) with 99Tc
m.
Discussion
The count rates observed during left ventricular FPRNA studies using 99Tc
m and
191Irm were within
the limits of accurate dead time correction for this gamma camera system. Left ventricular counts
were sufficiently high to measure left ventricular function accurately [15].
Window selection on the 191
Irm spectrum with the pulse height analyser is of major importance for
both camera resolution and linearity when performing studies with this tracer. Although the 100-
150 keV window is associated with the best camera resolution, this selection is the worst with
respect to count rate capacities and dead time correction because the relative low contribution of
those photons to the total number of photons reaching the crystal and because of the pile-up events.
Accurate dead time corrections with the reference activity source were obtained, in the 50-100 keV
as well as in the 50-150 keV window, for count rates higher than those observed in patients during
left ventricular first pass studies. Although the spatial resolution of the 50-100 keV was somewhat
better than the 50-150 keV window, this latter was chosen for clinical studies because the accuracy
of first pass measurements is known to be more sensitive to count rate than to spatial reslution [15].
Using this window, the number of counts in the left ventricular cavity with 191
Irm were at least equal
to those obtained with 99Tc
m in all patients.
During the right ventricular transit phase of the bolus, the total count rate was obviously over the
count rate capacities of the camera in most patients precluding simultaneous studies of the right en
left ventricles during a single injection of 191
Irm.
In our patient population, the mean total activity in the 20 cm field of view of the camera was about
1.15 times higher during diastole than during systole, introducing a relative dead time correction of
1.03 to 1.04. On the other hand, for 191
Irm, the relative decay correction between diastolic and
systolic frames ranged between 1.04 and 1.06. As the linearity correction factor and the decay
correction factor work in the opposite direction, an error of maximum 3% would be made on the
ejection fraction not applying any correction. For large field of view gamma cameras one can
expect a smaller relative dead time correction between diastolic and systolic frames due to a less
varying total activity in the large field of view.
18
Fig. 3. Time-activity curves of a decaying
191Ir
m source (initial activity 100 mCi) in the 50-100, 100-150 and
50-150 keV windows, respectively. For display purposes, the activity of the 191Os reference source and of the
misplaced events recorded simultaneously, were multiplied by a factor of 5.
References
1. Sporn V, Perez Balino N, Holman BL et al. Simultaneous measurement of ventricular
function and myocardial perfusion using the technetium-99m isonitriles. Clin Nucl Med
1988; 13: 77-81.
2. Baillet GY, Mena IG, Kuperus JH et al. Simultaneous technetium-99m MIBI angiography
and myocardial perfusion imaging. J Nucl Med 1989;30: 38-44.
3. Brihaye C, Butler TA, Knapp FF Jr et al. A new osmium-191/iridium-191m radionuclide
generator system using activated carbon. J Nucl Med 1986; 27: 380-7.
19
4. Packard AB, Treves ST, O’Brien GM, Lim KS. An osmium-191/iridium-191m radionuclide
generator using an oxalato osmate parent complex. J Nucl Med 1987; 28: 1571-6.
5. Issachar D, Abrashkins S, Weinigerr J et al. Osmium-191/iridium-191m generator based on
silica gel imregnated with tridodecylmethylammonium chloride. J Nucl Med 1989; 30: 538-
41.
6. Heller GV, Treves ST, Parker JA et al. Comparison of ultrashort-lived iridium-191m with
technetium-99m for first pass radionuclide angiocardiographic evaluation of right and left
ventricular function in adults. J Am Coll Cardiol 1986; 7: 1295-302.
7. Hellman C, Zafrir N, Shimoni A et al. Evaluation of ventricular function with first pass
iridium-191m radionuclide angiography. J Nucl Med 1989; 30: 450-7.
8. Franken PR, Dobbeleir A, Ham HR et al. Clinical usefulness of ultrashort-lived iridium-
191m from carbon-based generator system for the evaluation of the left ventricular function.
J Nucl Med 1989; 30: 1025-31.
9. Brihaye C, Dewez S, Guillaume M et al. Reactor production and purification of osmium-
191 for use in a new OS-191/Ir-191m radionuclide generator system. Appl Radiat Isot 1989;
40: 183-9.
10. Performance standards of scintillation cameras, Standards Publication/No. NU 1-1986.
National Electrical Manufacturers Association.
11. Ullman V, Husak V, Dubroka L. Deadtime correction in dynamic radionuclides studies by
computer. Eur J Nucl Med 1978; 3: 197-202.
12. Johnston AS, Arnold JE, Pinsky SM. Anger camera deadtime: marker source correction and
two parameter model. J Nucl Med 1975; 16: 539.
13. Lange D, Hermann HJ, Wetzel E, Schenck P. Critical parameters to estimate the use of a
scintillation camera in high dose dynamic studies. Medical Radionuclide Imaging (Proc.
Symp. Los Angeles) 1. Vienna: IAEA 1977; 85-100.
14. Johnston AS, Gergans GA, Kim I et al. Deadtime of computers coupled with anger cameras:
counting losses and false counts. Single photon emission computed tomography and other
selected computer topics (Proc. Symp. Miami 1980). Sorenson, ed. New York: Society of
Nuclear Medicine.
15. Dymond DS, Elliot A, Stone D et al. Factors that affect the reproducibility of measurements
of left ventricular function from first pass radionculide ventriculograms. Circulation 1982;
65: 311-22.
20
2.2 Variability of left ventricular ejection fraction and volumes by quantitative gated
SPET : influence of algorithm, pixel size and reconstruction parameters in normal
and small-sized hearts.
Anne-Sophie Hambye1, Ann Vervaet
2, André Dobbeleir
2,3
1Nuclear Medicine, CHU-Tivoli, La Louvière, Belgium
2Nuclear Medicine, Middelheim Hospital, Antwerp, Belgium
3Nuclear Medicine, University Hospital Ghent, Ghent, Belgium
Eur J Nucl Med Mol Imaging 2004; 31: 1606-1613.
Abstract
Several software are commercially available for quantification of left ventricle ejection fraction and
volumes from myocardial gated SPET, all with a high reproducibility. However, their accuracy has
been questioned in patients with a small-sized heart. This study aimed at evaluating the
performances of different software and the influence of modifications in acquisition or
reconstruction parameters on ejection fraction and volumes measurements, depending on the heart
size. Methods: Sixty-four2 and 1282 matrix size acquisitions were consecutively obtained in 31
patients referred for gated SPET. After reconstruction by filtered backprojection (Butterworth, 0.4,
0.5 or 0.6 cyc/cm cutoff, order 6), LVEF and volumes were computed with different software (3
versions of Quantitative Gated SPECT (QGS), Emory Cardiac Toolbox (ECT) and the Stanford
University (SU) Medical School algorithm), and processing workstations. Depending upon their
end-systolic volume (ESV), patients were classified into 2 groups: Group I (ESV>30ml, n=14) and
Group II (ESV <30ml, n=17). Agreement between the different software, and the influence of
matrix size and sharpness of the filter on LVEF and volumes were evaluated in both groups.
Results: In Group I, the correlation coefficients between the different methods ranged from 0.82 to
0.94 except for SU (r=0.77), and were slightly lower for volumes than ejection fraction. Mean
differences between the methods were not significant, except for ECT which LVEF values were
systematically higher by more than 10%. Changes in matrix size had no significant influence on
LVEF or volumes. On the other hand, a sharper filter was associated with significantly larger
volume values though this did usually not result in significant LVEF changes. In Group II, many
patients had a LVEF at the higher range. The correlations coefficients between the different
methods ranged between 0.80 and 0.96 except for SU (r=0.49), and were slightly worse for volumes
than LVEF values. Contrary to Group I, a majority of mean differences between LVEF
measurements was significant. LVEF was systematically the highest by ECT and the lowest by SU.
With QGS, changes in matrix size from 642 to 1282 were associated with significantly larger
volumes as well as lower LVEF values. Increasing the filter cutoff frequency had the same effect.
With SU-Segami, a larger matrix was associated with larger end-diastolic and smaller end-systolic
volumes, resulting in a highly significant increase in LVEF. Increasing the filter sharpness on the
other hand had no influence on LVEF though the measured volumes were significantly larger.
Conclusion: In patients with a normal-sized heart, LVEF and volume estimates computed from
different commercially available software for quantitative gated SPET are well correlated. LVEF
and volumes are little sensitive to changes in matrix size. Smoothing on the other hand was
associated with significant changes in volumes but usually not in LVEF values. However, owing to
21
the specific characteristics of each algorithm, software should not be interchanged for follow-up in
an individual patient.
In small-sized hearts on the other hand, both the used software and the matrix size or smoothing
significantly influence the results of quantitative gated SPET. LVEF at the higher range are
frequently observed with all the studied software except for SU-Segami. A larger matrix or a
sharper filter could be suggested to enhance the accuracy of most commercial software, more
particularly in patients with a small heart.
Keywords: quantitative gated SPET – LVEF – small heart – inter-software comparison
Introduction
Gated myocardial SPET has become the state-of-the-art for myocardial perfusion imaging, offering
the simultaneous evaluation of left ventricular perfusion and function with a single test. Different
methods to quantify left ventricular ejection fraction (LVEF) and volumes have been described [1-
6], all with a high reproducibility and a good agreement with various non nuclear or nuclear
techniques [6-10].
However, owing to the specific characteristics of each algorithm, software interchangeability for
repeated examinations in an individual patient should not be recommended [9,11] despite the good
correlations reported between different software computing the same gated SPET data [8,9,11].
Moreover, experimental data have revealed the sensitivity of gated SPET measured LVEF to
particular acquisition conditions such as time of imaging, background activity or injected dose [12],
filtering and zooming [13-15], and larger discrepancies between the methods have been described
for LV volumes [8], particularly at both ends of the scope of volume values.
Another problem in using quantitative gated SPET for LVEF calculation is encountered in patients
with a small heart such as children or some small women. Indeed, due to the limited spatial
resolution of the gamma cameras, the opposite endocardial edges of the left ventricle overlap, so
that the ventricular cavity may become almost virtual especially at end-systole. This results in an
underestimation of volumes, hence overestimation of LVEF [13-17], particularly using algorithms
based upon edge detection.
The purpose of our study was to compare LVEF and volumes computed from the same gated SPET
data by different versions of the QGS-package [1], the Emory Cardiac Toolbox [4,5] and the
Standford-University algorithm [6], and to evaluate the influence of filter and matrix size on the
measurements.
Material and methods
Patients and acquisition
During a 3-month period, QGS-analysis [1] was systematically performed in all patients undergoing
a stress test as a part of a two-day stress-rest gated myocardial SPET. Depending upon their end-
systolic volume (ESV) calculated on a GE-Elscint Expert system, the patients were classified into a
group with a normal or large-sized heart (ESV >30 ml, Group I) and a group with a small-sized
heart (ESV <30 ml, Group II). This value of 30ml-ESV was chosen based upon data from Ford et
al, reporting that the difference between measured and true LVEF in a cardiac phantom becomes
pronounced when the end-diastolic volume is <70 ml and the true LVEF is >40% [14]. Clinical
characteristics of the both patients groups are reported in Table 1.
Among those who required a comparative rest test, 31 underwent two consecutive gated SPET at
rest: 14 of Group I and 17 of Group II. Decision to perform this double rest study was based solely
upon the availability of free time-slots on the gamma-camera. The first acquisition in matrix 642,
zoom 1.28 (6.9 mm-pixel size) started about 1 hour after injection of 740-1000 MBq 99mTc-
sestamibi and was immediately followed by a second acquisition in a 1282 matrix, zoom 1.28 (3.45
mm-pixel size). Both SPET acquisitions lasted 25-30 minutes and were performed with a GE-
22
Elscint VariCam dual-head gamma camera equipped with VPC-35 collimators (system resolution of
9.0 mm FWHM at 10 cm distance; 290 cpm/µCi), using an eight-bin gated protocol (90 projections
(45/head) of 35 second/each; 360°-rotation; automatic body-contouring).
Gated SPET Analysis
Acquisition data sets were transferred from the GE-Elscint Expert system to a Sun Ultra10 Link
Medical system (Link Medical, Hamshire, UK), a PC-Windows NT GE system (GE Medical
Systems, Milwaukee, USA) and a PC-Windows NT Segami system (Segami, Columbia, USA)
using DicomP10, and to a Nuclear Diagnostic Hermes system (Nuclear Diagnostic, Stockholm,
Sweden) using modified interfile. The rough 642 and 1282 matrix gated SPET acquisitions were
reconstructed with Butterworth filters of 0.4, 0.5 or 0.6 cyc/cm cutoff (order 6) on different
workstations. LVEF and volumes were automatically quantified from the gated coronal slices using
commercially available software routinely used by the nuclear medicine community (three versions
of Quantitative Gated SPECT (QGS), Cedars-Sinai Medical Center, Los Angeles, CA; Emory
Cardiac Toolbox (ECT), Emory University, Atlanta, GA; Stanford University (SU) Medical School
algorithm). These six different processings will be further referred to as QGS-Link, QGS-GE, QGS-
Hermes, QGS-eNTEGRA, ECT-eNTEGRA and SU-Segami respectively. All have been described
in detail elsewehere [1,3-5] and widely validated.
Table 1. Clinical characteristics of the patient population (p=NS if >0.05).
Group I: ESV > 30ml; Group II: ESV <30 ml; CRF: cardiovascular risk factors; MI: myocardial infarction;
bicycle: upright bicycle stress test, 25W increment/2 min up to maximum heart rate; adenosine:
140µg/kg.min during 6 minutes; dobutamine: 10 to 40µg/kg.min with 3-min increments, + atropine if
required.
Group I (n=14) Group II (n=17) P value
Age (years); mean±SD 55.3±14.6 65.1±12.1 0.048
Gender (M/F) 7 / 7 2 / 15 0.044
CRF 7 10 NS
Prior MI 5 1 NS
Prior revascularization 5 4 NS
Referral reason
Chest pain
Abnormal stress EKG
Other
9
2
3
16
1
0
NS
NS
NS
Kind of stress test
(bicycle/adenosine/dobutamine)
9 / 4 / 1
8 / 9 / 0
NS
Evidence of stress ischemia on SPECT 5 6 NS
23
Statistical analysis
Results are expressed in absolute EF units for LVEF and in ml for the volumes.
All statistical analyses were performed using the SPSS statistical program package (SPSS Inc,
Chicago, USA). Inter-method variability was expressed as mean difference +/- SD. The significance
of the difference between two groups of data was assessed by the paired or unpaired Student’s t-test
and chi squared test or Fisher’s exact test, when appropriate. Paired data among three or more
groups were compared using repeated measurements ANOVA. A p value of 0.05 or less was
considered significant.
To identify the differences for multiple testing, a Bonferroni correction was applied for comparing
each pair of methods. With this correction, a p value of less than 0.0033 was considered significant.
Pearson correlation coefficients were calculated, and Bland-Altman plots [18] were generated to
search for trends by plotting the differences versus averages of paired values. For this part of the
analysis, QGS-Link was arbitrarily chosen as a reference against which the other methods were
plotted, as it constituted the last version of the most widely spread quantification method.
Fig 1. Bland-Altman plots showing the agreement for ejection fraction between the reference method (QGS-
Link) and the other packages. (ESV>30ml in open circles; ESV<30ml in solid circles).
LVEF: left ventricular ejection fraction; ESV: end-systolic volume.
-25
-20
-15
-10
-5
0
5
10
15
20
25
30
35
0,0 20,0 40,0 60,0 80,0 100,0
mean
QGS GE - QGS Link
-25
-20
-15
-10
-5
0
5
10
15
20
25
30
35
0,0 20,0 40,0 60,0 80,0 100,0
mean
QGS Hermes-QGS Link
-25
-20
-15
-10
-5
0
5
10
15
20
25
30
35
0,0 20,0 40,0 60,0 80,0 100,0
mean
QGS eNTEGRA-QGS Link
-25
-20
-15
-10
-5
0
5
10
15
20
25
30
35
0,0 20,0 40,0 60,0 80,0 100,0
mean
SU Segami-QGS Link
-25
-20
-15
-10
-5
0
5
10
15
20
25
30
35
0,0 20,0 40,0 60,0 80,0 100,0
mean
ECT eNTEGRA-QGS Link
24
Results
Using repeated measures analysis of variance between the six processings and the two groups of
patients, a highly significant interaction was found, indicating that the impact of the software
differed for the two patients groups. In addition, a significant overall difference was found between
the six methods and also between the two patients groups (both p<0.001).
Influence of the processing algorithm on ejection fraction and volume values.
Group I (ESV> 30 ml).
The different methods were fairly correlated with QGS-Link, with r values ranging between 0.82
and 0.94, except for SU-Segami (r=0.77).
Mean LVEF and volume values were quite similar for the different methods, except for ECT-
eNTEGRA which resulted in higher LVEF (Table 2). By Bland-Altman analysis, no significant
trend toward higher or lower LVEF was found across the whole range of values for any method but
ECT-eNTEGRA compared to QGS-Link (Figure 1, open circles). By paired Student’s t-test, highly
significant differences (0.0001<p<0.0033) were noted between ECT-eNTEGRA and the other
methods for LVEF and end-systolic but not for end-diastolic volume values (Table 3). Significant
differences were also found for volume values between the QGS versions.
Group II (ESV<30 ml)
In keeping with Group I, all methods except SU-Segami correlated well with QGS-Link (r values
between 0.80 and 0.96; r=0.49 for SU-Segami).
Mean LVEF values were above 70% for all programs except for SU-Segami (mean LVEF: 60.4%,
Table 2), and was highest by ECT-eNTEGRA. Opposite to Group I however, inter-method
variability was quite large and most mean LVEF differences were significant (Table 3). Significant
disparities in volume estimates were more frequent for end-systolic than end-diastolic volumes, and
were particularly large for SU-Segami (between 10ml and 20ml, all p values <0.0001, Table 3).
Compared to QGS-Link, LVEF was systematically higher by ECT-eNTEGRA (8.1±5.46%) and
lower by SU-Segami (-13.7±8.01%) as shown on the Bland-Altman plots (Figure 1, solid circles).
More surprisingly, a small but systematic difference in LVEF was also found by Bland-Altman
analysis between the three versions of the QGS software, reaching the level of statistical
significance for QGS-Hermes (Table 3).
Table 2. Mean±SD ejection fraction and volumes for the different software
(Group I: ESV > 30ml; Group II: ESV <30 ml; LVEF: left ventricular ejection fraction;
EDV: end-diastolic volume; ESV: end-systolic volume).
QGS Link QGS GE QGS Hermes QGS
eNTEGRA
ECT
eNTEGRA
SU Segami
LVEF (%) 45.1±12.98 47.4±12.43 49.4±12.48 47.5±13.53 62.0±14.13 50.1±12.67
EDV (ml) 119.5±66.24 122.4±63.85 112.6±63.85 119.3±65.76 108.2±51.22 119.3±49.19
Group I
ESV (ml) 72.3±59.96 69.9±58.32 62.7±52.47 69.9±57.89 46.1±41.14 64.3±45.45
LVEF (%) 74.5±9.06 70.1±7.35 78.1±8.49 73.1±7.80 82.4±8.24 60.4±5.43
EDV (ml) 53.6±17.23 57.8±15.68 51.6±18.21 55.6±17.31 54.9±17.25 70.9±15.25
Group
II
ESV (ml) 14.8±9.02 17.1±6.86 12.8±8.38 15.9±8.41 10.1±5.76 28.5±7.65
25
Influence of filtering on left ventricular ejection fraction and volume values.
For this part of the study, 642 matrix size images were used. Due to technical limitations of some
programs at our disposal at the time of the study, only QGS-GE, QGS-Hermes and SU-Segami
were compared.
Group I (ESV> 30 ml).
Increasing the cutoff frequency of the Butterworth filter from 0.4 to 0.6 cyc/cm (order 6) resulted in
significantly larger volumes for both QGS versions, and smaller volumes
for SU-Segami. However, the subsequent changes in LVEF were significant only by QGS-GE
(Table 4). The effect of filtering was more striking for the end-systolic volumes in relative values,
although the absolute changes were usually higher for the end-diastolic (Table 4).
Group II (ESV<30 ml)
Sharper filtering resulted in significantly larger volumes for QGS, and particularly the end-diastolic,
and in smaller volumes for SU-Segami. The ensuing LVEF change was however significant only
for the QGS-versions, the SU-Segami LVEF remaining remarkably stable (Table 4).
Table 3. Mean±SD difference in ejection fraction and volumes value according to the processing method
used.
(Group I: ESV > 30ml; Group II: ESV <30 ml; LVEF: left ventricular ejection fraction (%); EDV: end-
diastolic volume (ml); ESV: end-systolic volume (ml)).
P values are calculated using the Student’s paired t-test after Bonferroni correction. *: 0.0001<p<0.0033; **:
p<0.0001. All p values >0.0033 are considered as not significant.
Group I Group II
LVEF EDV ESV LVEF EDV ESV
QGS Link-QGS GE -2.2+/-7.61 -2.9+/-12.26 2.4+/-11.52 3.3+/-4.39 -3.8+/-6.81 -1.6+/-3.07
QGS Link-QGS Hermes -4.3+/-4.60 6.9+/-9.33 9.6+/-8.36 * -2.9+/-2.50 * 2.2+/-8.27 1.7+/-2.95
QGS Link-QGS eNTEG -2.3+/-7.57 1.8+/-14.59 3.8+/-13.58 1.1+/-2.47 -0.9+/-2.76 -0.7+/-1.88
QGS Link-ECT eNTEG -16.9+/-8.16 ** 11.3+/-21.95 26.1+/-22.52 * -8.1+/-5.46 ** 0.6+/-8.79 5.1+/-4.50 *
QGS Link-SU Segami -4.9+/-8.69 0.2+/-23.9 8.0+/-19.48 13.7+/-8.01 ** -17.5+/-6.09 ** -13.5+/-4.29 **
QGS GE-QGS Hermes -2.1+/-5.94 9.8+/-8.46 * 7.1+/-10.60 -6.2+/-3.42 ** 5.6+/-11.9 3.1+/-3.29
QGS GE-QGS eNTEG -1.0+/-3.74 5.4+/-10.53 2.5+/-8.56 -2.4+/-3.05 2.8+/-6.88 1.1+/-2.82
QGS GE-ECT eNTEG -14.6+/-4.8 ** 14.2+/-17.68 23.7+/-18.94 * -11.8+/-6.35 ** 3.5+/-11.91 6.9+/-2.53 **
QGS GE-SU Segami -2.7+/-4.91 3.1+/-18.03 5.6+/-15.16 9.7+/-7.28 ** -13.2+/-7.61 ** -11.4+/-4.65 **
QGS eNT-QGS Herm -2.0+/-5.87 5.0+/-12.35 5.9+/-11.71 -4.1+/-2.25 ** 2.9+/-7.65 2.4+/-1.91 *
QGS eNT-ECT eNTEG -14.4+/-4.87 ** 10.9+/-16.8 23.0+/-16.84 * -9.4+/-5.32 ** 0.7+/-8.20 5.8+/-3.87 **
QGS eNTEG-SU Seg -2.2+/-7.06 -0.7+/-18.5 4.5+/-15.42 12.3+/-7.45 ** -16.4+/-4.92 ** -12.8+/-3.69 **
QGS Herm-ECT eNT -12.6+/-7.25 ** 4.4+/-15.69 16.6+/-15.66 * -5.2+/-5.54 -1.4+/-8.01 3.5+/-4.16
QGS Herm-SU Segami -0.6+/-6.77 -6.6+/-16.29 -1.6+/-11.98 17.1+/-8.43 ** -19.1+/-6.76 ** -15.1+/-3.92 **
SU Sega-ECT eNTEG -11.9+/-6.63 ** 11.1+/-12.45 18.1+/-9.4 ** -21.7+/-9.02 ** 17.1+/-8.27 ** 18.6+/-4.27 **
26
Table 4: Influence of the filter cutoff frequency (order 6) on mean values for ejection fraction, end-diastolic
and end-systolic volumes. The p values are calculated by overall repeated measurements ANOVA (p=NS if
>0.05)
(Group I: ESV > 30ml; Group II: ESV <30 ml; BW: Butterworth filter; LVEF: left ventricular ejection
fraction; EDV: end-diastolic volume; ESV: end-systolic volume).
Group I
Group II
BW 0.4 BW 0.5 BW 0.6 P value BW 0.4 BW 0.5 BW 0.6 P value
EF (%) 48.1 47.4 45.1 0.008 72.6 70.1 69.8 0.003
EDV (ml) 111.1 122.4 122.3 <0.001 50.7 57.8 57.5 <0.0001
QGS GE
ESV (ml) 62.0 69.9 72.7 <0.0001 13.7 17.1 17.1 <0.0001
EF (%) 49.6 49.5 48.7 NS 82.5 78.1 77.1 0.007
EDV (ml) 100.9 112.6 118.9 <0.0001 47.8 51.6 56.3 <0.0001
QGS
Hermes
ESV (ml) 55.7 62.7 67.7 0.002 10.0 12.8 13.9 0.004
EF (%) 49.5 50.1 49.0 NS 60.3 60.4 60.5 NS
EDV (ml) 128.6 119.3 112.9 <0.0001 74.3 70.9 67.9 0.002
SU
Segami
ESV (ml) 69.8 64.3 62.1 <0.001 29.8 28.5 26.9 0.003
Influence of matrix size on left ventricular ejection fraction and volume values.
For this part of the study, the 0.5 cyc/cm Butterworth filter images (order 6) were processed by
QGS-GE, QGS-Hermes and SU-Segami.
Group I (ESV> 30 ml).
In this group, modifying the matrix size did not significantly influence mean LVEF and volume
values except for the end-diastolic volumes by SU-Segami (Table 5). Mean± SD differences (matrix
642 – 128
2) for LVEF, EDV and ESV were respectively 0.7±5.88%, 3.6±13.3 ml and 1.9±12.05 ml
for QGS-GE, 0.6±6.15%, –1.7±8.97 ml and -1.3±8.96 ml for QGS-Hermes, and -3.9±7.18%, -
9.1±10.51 ml and -1.1±9.36 ml for SU-Segami.
Group II (ESV<30 ml)
Decreasing the pixel size from 6.9 to 3.45 mm significantly modified the LVEF and volume values
regardless of the used processing (Table 5).
Using QGS, a smaller pixel size was associated with lower LVEF and larger volumes. Mean± SD
differences (matrix 642 – 128
2) for LVEF, EDV and ESV were respectively 2.8±4.36% (p=0.021), -
6.8±8.28 ml (p=0.004) and -4.1±4.24 ml (p=0.001) for QGS-GE, and 5.1±4.66% (p=0.001), –
6.7±7.02 ml (p=0.003) and -3.8±3.07 ml (p<0.0001) for QGS-Hermes. The effect of a smaller pixel
size seemed particularly marked for end-diastolic volumes of 60ml and below.
Using SU-Segami, results were divergent for end-diastolic and end-systolic volumes, the former
increasing from a mean value of 70.9 ml to 76.1ml for the 1282 matrix (p=0.014), and the latter
decreasing from 28.5 ml to 20.5 ml (p<0.001). As a consequence, LVEF increased by 12.9±5.74%
on average (p<0.0001).
27
Table 5: Influence of the acquisition matrix (642 or 128
2) on mean values for ejection fraction, end-diastolic
and end-systolic volumes. The p values are calculated by paired Student’s t-test (p=NS if >0.05).
(Group I: ESV > 30ml; Group II: ESV <30 ml; LVEF: left ventricular ejection fraction; EDV: end-diastolic
volume; ESV: end-systolic volume).
Group I
Group II
Matrix size 642 128
2 P value 64
2 128
2 P value
LVEF (%) 47.4 46.6 NS 70.1 67.4 0.021
EDV (ml) 122.4 118.9 NS 57.8 64.6 0.004
QGS GE
ESV(ml) 69.9 68.0 NS 17.1 21.2 <0.001
LVEF (%) 49.4 48.9 NS 78.1 71.5 0.001
EDV (ml) 112.6 114.4 NS 51.6 58.5 0.003
QGS
Hermes
ESV(ml) 62.7 64.0 NS 12.8 17.4 <0.0001
LVEF (%) 50.1 53.9 NS 60.5 73.4 <0.0001
EDV(ml) 119.3 128.4 0.006 70.9 76.1 0.014
SU
Segami
ESV(ml) 64.3 65.4 NS 28.5 20.5 <0.0001
Discussion
Using gated myocardial SPET, several algorithms have been developed for the calculation of LVEF
and volumes, each owing its specific assumptions for left ventricle modeling. Among the various
commercial programs, Cedars-Sinai Quantitative Gated SPECT (QGS, 1) is currently the most
widely used in the clinical setting. Its reliability and reproducibility are excellent and have been
validated against a whole range of methods. Nevertheless, with increased routine use, some
limitations have appeared, such as a falsely elevated LVEF in patients with a small-sized heart like
children or some women [14,16].
In patients with a normal- or large-sized heart, our study confirms the good agreement for LVEF
between different processing methods [7-9] and the absence of significant bias through the whole
range of LVEF values. Indeed, except for ECT-eNTEGRA that systematically overestimated LVEF
by more than 10%, no significant method-related mean differences in LVEF were noted. This
overestimation of LVEF by ECT has also been reported by others, including the authors of the
program themselves [9,19], and might be due to specificities in time sampling or shape used for LV
modeling [19]. Despite this good agreement, interchanging algorithms, or even consecutive
versions of the same algorithm for follow-up studies in an individual patient should not be
recommended because of the rather large standard deviation of the differences between the
methods. For the volume values, and more particularly the end-systolic, we found a larger
variability than for LVEF, with significant differences not only between ECT-eNTEGRA and the
other programs, but also between the three versions of QGS, maybe due to (minor) modifications of
its algorithm. In this patient population, increasing the matrix size had no significant influence on
volume or LVEF values. Increasing the filter cutoff frequency on the other hand significantly
modified the volume measurements, though this resulted in significant changes in LVEF only with
QGS-GE.
In patients with a small-sized heart, most mean differences in LVEF were significant despite a good
agreement between the different methods except for SU-Segami. Moreover, a systematic bias was
noted not only for ECT-eNTEGRA but also for SU-Segami which volumes were systematically
28
markedly larger, probably due to differences in the location of the ventricular wall which
corresponds to the average position for the latter and to the endocardial surface for the former [7].
Also changes in matrix of filter cutoff significantly influenced volume and LVEF values in small
hearts. This influence of the pixel size has already been reported by Nakajima et al in a cardiac
phantom study [13]. They found a decrease from 49% to 3% in the overestimation of a 37-ml
chamber volume by increasing the zoom from none to 2x during the acquisition, and confirmed
their findings in a pediatric population, but only in children younger than 7 years [13]. However, the
1.28x-zooming applied in the present study is the maximum magnifying factor that can be used for
a 60cm-field of view gamma camera without mechanical device keeping the heart in the center of
rotation, so that we were compelled to increase the matrix from 642 to 128
2 to reduce the pixel size
from 6.9 to 3.45 mm and so improve the delineation of the left ventricle endocardial border. Using
QGS, this modification resulted in significantly larger volumes and lower LVEF, particularly for
end-diastolic volumes of 60ml and below. By SU-Segami on the other hand, the combination of
larger end-diastolic and smaller end-systolic volumes for a 1282 matrix resulted in a highly
significant increase in LVEF, probably because of an insufficient count density and thus enhanced
statistical fluctuations.
By increasing the cutoff frequency of the Butterworth filter from a smooth 0.4 to a sharper 0.6
cyc/cm, larger volumes and a significant decrease in LVEF was obtained by QGS. By SU-Segami
on the contrary, LVEF remained stable despite significantly smaller volumes with a sharper filter,
probably because of parallel changes in end-diastolic and end-systolic volumes. The influence of
smoothing on LVEF and volumes could be due to the fact that, because of the limited spatial
resolution of a gamma-camera, the proportion of LV volume contained in an individual pixel is
larger in small than in large-sized hearts. In this way, changes in count density of the (especially
endocardial) pixels related to the cardiac motion are probably more abrupt for higher cutoff
frequency filtering. With a smooth filter, the systolo-diastolic transition in count density might be
softer, hence volume estimates smaller and LVEF higher. The lesser filter-dependence observed
with SU-Segami could be explained by the fact that its algorithm relies on the average ventricular
wall position instead of the endocardial surface.
This study compared different processing methods for quantitative estimates of LVEF and volumes
using gated myocardial perfusion SPET. Despite good correlations with regard to the calculated
values, clear differences were found between the algorithms, and more particularly between SU-
Segami and the other methods, especially in patients with a small heart. No single external standard
was available in our patients to determine the “true” values, so that the most recent version of the
most widely used program was arbitrarily chosen as a reference. Therefore, the calculated results
might be only a rough estimation of the patients’ real LVEF and volumes. However, since we
aimed at correlating different processing methods computing the same gated SPET data, the use of
an external standard does not seem an absolute prerequisite to validate the results. Another
limitation consists in the use of low-energy general-purpose collimators (system resolution: 9.0 mm
FWHM at 10 cm distance) for the gated SPET acquisition. Indeed, a high-resolution collimator
should be preferred from a theoretical point of view since resolution recovery is expected to affect
small volumes more than large. With a high resolution collimator, a 1.5 mm gain in resolution could
be anticipated, but at the expense of a 40 %-count reduction which would require a smoother filter,
hence loss of resolution, for acceptable image quality. The choice of general-purpose collimators
constitutes thus a compromise between resolution and noise, especially using an automatic body-
contouring to reduce the patient-collimator distance. A last limitation concerns the small number of
patients included. Despite this small sampling, highly significant results could be found so that this
should not considered a major drawback, all the more as our purpose was to compare the different
software currently available and not to identify the best of them.
29
Conclusion
In patients with a normal-sized heart, quantitative estimates of left ventricular functional data
computed from gated myocardial perfusion SPET by different commercially available software
show excellent correlation. Inter-software, or even inter-version variability for an individual
software is however present, especially regarding the volume values. Technical parameters such as
matrix size or filter cutoff frequency have little influence on LVEF measurements but a sharper
filter significantly modify the calculated volumes. Consequently, definition of specific normal
limits should be advised for each algorithm, and software permutation should be avoided for
follow-up studies in an individual patient.
In small-sized hearts on the other hand, ejection fraction value in the (very) high range, most
probably overestimated, is observed in a significant number of cases, so that the accuracy of gated
SPET measured LVEF and volumes in these patients might be questioned. However, increasing the
matrix size or the filter cutoff frequency results in significantly lower, probably more realistic
LVEF with all the tested software except the SU-Segami. Although further confirmation of our
results and validation of the correctness of the measurements is required, a smaller pixel size and/or
a sharper filter might be suggested for quantitative gated SPET in patients with a small-sized heart.
Acknowledgments
The authors whish to thank H. Ham, MD, PhD, for his friendly comments and criticisms in the
review of this manuscript. None of the authors has a financial interest in any software package. This
study did not receive any vendor support.
References
1. Germano G, Kiat H, Kavanagh PB, Moriel M, Mazzanti M, Su H, et al. Automatic
quantification of ejection fraction from gated myocardial perfusion SPECT. J Nucl Med 1995;
36:2138-47.
2. Germano G, Kavanagh PB, Kavanagh JT, Wishner SH, Berman DS, Kavanagh GJ.
Repeatability of automatic left ventricular cavity volume measurements from myocardial perfusion
SPECT. J Nucl Cardiol 1998; 5:477-483.
3. Goris ML,Thompson C, Malone L, PR Franken. Modeling the integration of myocardial
regional perfusion and function. Nucl Med Commun 1994; 15: 9-20.
4. Faber TL, Akers MS, Peshock RM, Corbett JR. Three dimensional motion and perfusion
quantification in gated single-photon emission computed tomograms. J. Nucl Med 1991; 32:2311-
2317.
5. Faber Tl, Cooke CD, Folks RD et al. Left ventricular function and perfusion from gated
perfusion images : an integrated method. J Nucl Med 1999; 40: 650-659.
6. Nichols K, DePuey RG, Rozanski A. Automation of gated tomography left ventricular
ejection fraction. J Nucl Cardiol 1996; 3: 475-482
7. Everaert H, Bossuyt A, Franken P. Left ventricular ejection fraction and volumes from gated
single photon emission tomographic myocardial perfusion images: Comparison between two
algorithms working in three-dimensional space. J Nuclear Cardiology 1997;4:472-476.
8. Nichols K, Lefkowitz D, Faber T, et al. Echocardiographic validation of gated SPECT
ventricular function measurements. J Nucl Med 2000;41:1308-14.
30
9. Nakajima J, Higuchi T, Taki J, Kawano M, Tonami N. Accuracy of ventricular volume and
ejection fraction measured by gated myocardial SPECT: Comparison of 4 software Programs. J
Nucl Med 2001; 42: 1571-78.
10. Vourvouri E, Poldermans D, Bax JJ, et al. Evaluation of left ventricular function and
volumes in patients with ischaemic cardiomyopathy : gated single-photon emission computed
tomography versus two-dimentional echocardiography. Eur J Nucl Med 2001;28:1610-15.
11. Lum DP, Coel MN. Comparison of automatic quantification software for the measurement
of ventricular volume and ejection fraction in gated myocardial perfusion SPECT. Nucl Med Comm
2003; 24: 259-266
12. Vallejo E, Dione DP, Bruni WL, et al. Reproducibility and accuracy of gated SPECT for
determination of left ventricular volume and ejection fraction: experimental validation using MRI. J
Nucl Med 2000; 41:874-882.
13. Nakajima K, Taki J, Higuchi T, Kawano M, Taniguchi M, Maruhashi K, Sakazume S,
Tonami N. Gated SPET quantification of small hearts: mathematical simulation and clinical
application. Eur J Nucl Med 2000;27:1372-79.
14. Ford P, Chatziioannou S, Moore H, Dhekne R. Overestimation of the LVEF by quantitative
gated SPECT in simulated left ventricles. J Nucl Med 2001;42:454-459.
15. Manrique A, Hitzel a, Gardin I, Dacher JN, Vera P. Influence of Wiener filter in
determining the left ventricle volume and ejection fraction using thallium-201 gated SPECT. Nucl
Med Comm 2003; 24: 907-914
16. De Bondt P, Van de Wiele C, De Sutter J, De Winter F, De Backer G, Dierckx RA. Age-
and gender-specific differences in left ventricular cardiac function and volumes determined by
gated SPET. Eur J Nucl Med 2001;28:620-24.
17. Achtert AD, King MA, Darlberg ST, et al. An investigation of the estimation of ejection
fractions and cardiac volumes by a quantitative gated SPECT software package in simulated gated
SPECT images. J Nucl Cardiol 1998;5:144-152.
18. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of
clinical measurements. Lancet 1986; 1: 307-310.
19. Nichols K, Santana CA, Folks R et al. Comparison between ECT and QGS for assessment
of left ventricular function from gated myocardial perfusion SPECT. J Nucl Cardiol 2002; 9:285-93
31
2.3 Clinical applications.
Clinical usefulness of ultrashort-lived Iridium-191m from a carbon-based generator system for the evaluation of the left ventricular function.
P.R. FRANKEN, A. DOBBELEIR, H.R. HAM, C. BRIHAYE, M. GUILLAUME, F.F. KNAPP and J. VANDEVIVERE. Nuclear Medicine, Middelheim Hospital, Antwerp and St-Peters Hospital , Brussels, Belgium. Cycloton Research Center, University of Liege, Belgium and Nuclear Medicine Group, Oak Ridge National Laboratory Tennessee, USA. Journal of Nuclear Medicine, 1989; 30: 1025-1031. Abstract
Ultrashort-lived
191mIr (4.96 sec; 63-74 and 129 keV photons) is potentially advantageous for first-pass radionuclide
angiocardiography, offering the opportunity to perform repeat studies with very low absorbed radiation dose to the
patient. Left ventricular (LV) first-pass studies were performed in 72 patients with 191m
Ir from a new bedside 1.3 Ci (48.1
GBq) 191Os/
191mIr generator system using an activated carbon support that offers high
191mIr yields (15-18%) and
consistent low 191Os breakthrough (2-4 x 10
-4 %/bolus). Using a single crystal digital gamma camera, uncorrected end-
diastolic counts in the left ventricular representative cycle ranged from 10 up to 30 k counts. The reproducibility of
repeated LV ejection fraction (LVEF) determination at 2-min intervals in 50 patients was r = 0.97, mean diff. = 2.08 ± 1.55
EF units. Comparison between 191m
Ir (80-120 mCi; 2960-4400 MBq) and 99mTc (20-25 mCi; 750-925 MBq) LV count rates
indicates a 3 wk useful shelf life of this new generator system for cardiac studies. Iridium-191m determined LVEF
correlated closely with 99mTc determined LVEF in 32 patients (r = 0.96, mean diff. = 1.87 ± 1.23 EF units). Parametric
images for LV wall motion analysis were comparable with both isotopes. We conclude that rapid, repeat, and
reproducible high count rate first-pass left ventricular studies can be obtained with 191m
Ir from this new 191Os/
191mIr
generator system using a single crystal gamma camera.
32
Comparison between exercise myocardial perfusion and wall motion
using 201Tl and 191mIr simultaneously.
P.R. FRANKEN, A. DOBBELEIR, H.R. HAM, R. RANQUIN, S. LIEBER, F. VAN DEN BRANDEN, P. VAN DEN HEUVEL, C. BRIHAYE, M. GUILLAUME, F.F. KNAPP and J. VANDEVIVERE. Nuclear Medicine and Cardiology, Middelheim Hospital, Antwerp, Belgium Nuclear Medicine St-Peters Hospital , Brussels, Belgium. Cycloton Research Center, University of Liege, Belgium and Nuclear Medicine Group, Oak Ridge National Laboratory Tennessee, USA.
Nuclear Medicine Communications, 1991; 12: 473-484. Summary
By exploiting the ultrashort halflive
191mIr as tracer for left ventricular first-pass angiocardiography and
201Tl as myocardial
perfusion agent, direct comparison between myocardial perfusion and regional wall motion was obtained during the
same exercise stress test in patients with non-significant coronary artery disease, in patients with recent myocardial
infarction, and in patients six weeks after successful percutaneous transluminal coronary angioplasty (PTCA). A good
agreement between regional myocardial perfusion and regional wall motion was observed in patients with non-significant
coronary artery disease and in most patients with recent myocardial infarction. In contrast, discrepancies occurred at
maximal exercise in patients studied six weeks after successful PTCA: only 38% of the patients with no evidence of
restenosis and with a completely normal myocardial perfusion scintigraphy had a normal regional wall motion at maximal
exercise stress. According to these results, a normal uptake of 201Tl six weeks after PTCA would mean that the
circulation has been successfully re-established but without predicting the functional capacities of the myocardial cells
which remain altered at least six weeks after the revascularization procedure in about two-thirds of the patients. We
conclude that 191m
Ir in combination with 201Tl offers the opportunity of performing myocardial perfusion and wall motion
studies simultaneously both at rest and during exercise.
201Tl myocardial perfusion score
2 1 0
Patients with <5% likelihood of CAD (group I)
2 28 2 0
1 0 0 0
191Irm left ventricular RNA
Wall motion
Score
0 0 0 0
Patients with recent
myocardial infarction (group II)
2 30 9 1
1 6 20 4
191Irm left ventricular RNA
Wall motion Score
0 1 5 14
Patients with successful PTCA (group III)
2 30 0 0
1 13 1 0
Comparison between regional myocardial perfusion score and regional wall motion
score at maximal exercise stress in the
anterior projection.
191Irm left ventricular RNA
Wall motion Score
0 1 0 0
Myocardial perfusion score: 2=normal, 1=moderate hypoperfusion,0=severe hypoperfusion. Regional wall motion score: 2=normal, 1=hypokinesis, 0=akinesis or dyskinesis.
33
3. Myocardial perfusion and viability.
3.1 SPET generated colour-coded polar maps to quantify the uptake of 99m
Tc-
sestaMIBI and 123I-BMIPP in chronically dysfunctional myocardium: comparison
with coronary anatomy and wall motion.
A.S. Hambÿe1, A. Dobbeleir
1, P.R. Franken
2.
1Middelheim General Hospital, Antwerp and
2University Hospital, Free University of Brussels,
Brussels, Belgium.
Nuclear Medicine Communications 1997;18: 1135-1147.
Abstract
The combined use of 123
I-BMIPP and 99m
Tc-sestaMIBI SPET imaging has been proposed as an
alternative to PET for the noninvasive detection of jeopardized myocardium after a myocardial
infarction, a mismatching quite accurately indicating jeopardized but still viable tissue. In this paper
a new quantitative approach is described, expressing the presence and degree of mismatching in %
of the left ventricular surface globally as well as for each major epicardial artery by means of
clearly identified colour-coded polar maps. With this method, the relative proportion of normal and
scar tissue, each characterized by a specific colour, is measured using thresholds of sestaMIBI
uptake of respectively 60% and 30% of the expected mean normal value, whereas the presence and
extent of mismatching between BMIPP and sestaMIBI are calculated only between these two
thresholds, typically corresponding to a flow decrease associated with a possible but uncertain post-
revascularization recovery. Applied to 15 patients with severely impaired left ventricular function
post myocardial infarction, small intra- and interobserver differences were noted in the assessment
of the relative % of normal, mismatched and scar tissue. More specifically analyzing the variability
in the calculated % mismatching, a good reproducibility was observed, with intra- and interobserver
correlation coefficients of 0.96 and 0.94 respectively, mean intraobserver difference of 0.25 % (SD:
2.0%) for the left ventricle globally, 1.65% (SD: 2.9 %) for the LAD, -1.56 % (SD: 3.6 %) for the
LCX and -1.24% (SD: 2.8%) for the RCA territories and mean interobserver variability of 0.91%
(SD: 2.4%) for the left ventricle globally, -1.51% (SD: 3.0%) for the LAD, -0.53% (SD: 2.9%) for
the LCX and -0.34% (SD: 3.9%) for the RCA territories. Using the second standard deviation of the
interobserver difference as criterion of significancy, significant mismatching between BMIPP and
sestaMIBI was noted in 13 arterial territories, corresponding to significant stenoses on coronary
angiogram and/or wall motion abnormalities in all cases. These results suggest that this new
quantitative method, showing a good reproducibility, may constitute a reliable and interesting tool
for the noninvasive evaluation of myocardial viability with SPET.
Key words: myocardial viability - single photon emission tomography - iodinated fatty acids
analogues - perfusion tracer - quantification.
34
Introduction
Despite its high cost and limited availability, positron emission tomography (PET) with 18F-
fluorodeoxyglucose (18F-FDG) remains the golden standard for the noninvasive assessment of
viability after a myocardial infarction (MI). However, attempts are made to replace it by less
expensive techniques, such as stress echocardiography with low-dose dobutamine[1], planar or
tomographic rest/redistribution imaging with 201
Tl [2], or quantification 99m Tc-sestaMIBI uptake at
rest[3]. Regarding this last method, disagreement exists concerning the value of sestaMIBI uptake
separating normal, jeopardized and scar myocardium. Usually, the lower limit of uptake
corresponding to viable tissue is set at 60 % of the maximum, and the upper limit of definitely
necrotic tissue at 30% to 40 %[3,4]. Myocardial regions with perfusion in between constitute a grey
zone, typically considered to contain an admixture of necrotic, viable and normal cells in various
amounts in which potential improvement post-revascularization is not clearly defined.
Recently, several studies have reported an improvement in the accuracy of sestaMIBI perfusion
imaging to assess viability by coupling it to the evaluation of cardiac metabolism using 123
I-labelled
modified fatty acids[5-8]. Indeed, due to an early metabolic switch from beta-oxydation to
glycolysis to preserve the ATP production in hypoxic myocardium, the uptake of fatty acids is
expected to be more reduced than flow in jeopardized but still viable myocardial segments.
Therefore, a mismatching between flow and metabolism, mirroring that obtained with 18F-FDG
PET, is supposed to represent a hallmark of myocardial viability [9]. We have developed a new method to quantify the degree and extent of mismatching between
metabolism (assessed by iodine 123-labeled beta-methyl iodophenyl pentadecanoic acid [123
I-
BMIPP]) and perfusion in terms of % of the left ventricle surface by means of colour-coded polar
maps for various values of 99m
Tc-sestaMIBI uptake.
The current paper describes this new quantitative analysis, and the preliminary results obtained in
15 patients with severe left ventricular dysfunction post myocardial infarction, regarding the intra-
and inter-observer reproducibility and the concordance with the coronary anatomy and the wall
motion assessed by contrast ventriculography.
Material and methods
Patient population
Fifteen patients (13 men; 2 women, mean age +/- SD: 66.1 +/-6.7 years, range: 57-78 years) were
included, all suffering from at least one myocardial infarction and presenting left ventricular
dysfunction (Table 1).The diagnosis of infarction was based upon documented CPK increase at the
time of the acute event, significant electrocardiographic (ECG) modifications and angiographically
proven wall motion abnormalities. Overall, on ECG there were 3 anterior, 7 anteroseptal, 5 inferior,
1 anterolateral and 2 lateral infarctions, with a mean age +/- SD (for the most recent) of 27.1 +/-
58.4 weeks (range: 2-230 weeks). Thrombolysis was the first-line treatment in 9 patients, balloon
angioplasty in 1 and intravenous heparin in 4, the last one having received other drugs.
In all patients, coronary angiogram and left ventricular ventriculogram, sestaMIBI and BMIPP
single photon emission tomography (SPET), and equilibrium radionuclide angiography were
performed within 10 days. Mean ejection fraction value at the time of the tests was 36.1% (SD:
9.6%, range: 19.9% to 51.6%°).
35
Table 1. Clinical characteristics of the patient population (MI: myocardial infarctio; Thr.: thrombolysis;
Hep.: heparin; PTCA: percutaneous transluminal coronary angioplasty; Misc.: other drugs; CABG: coronary
artery bypass grafting; LAD: left anterior descending artery; LCX: left circumflex artery; RCA: right
coronary artery; R. Interm: ramus intermedius; D1: first diagonal branch; L1: first lateral branch.)
Pat number Localization MI
(on ECG)
Age MI
(weeks)
Treated by Previous
revasculari-
zation
Coronary
anatomy (%
stenosis)
LVEF (%)
1 Anterolat. 4 Thr. No LAD:90
D1: 99
32.9
2 Anterosept. 10 Thr. No LAD: 75
LCX: 80
22.4
3 Lateral 3 Hep. No LAD: 70
LCX: 99
32.0
4 Anterosept. 4 Thr. No LAD: 100 41.9
5 Anterior 14 Thr. No LAD: 100
RCA: 50
38.8
6 Anterosept. 2 Thr. No LAD: 99
R. interm: 90
47.0
7 Anteroapic 40 Thr. PTCA LAD: 70
R. interm: 99
51.6
8 Anterosept. 22 Thr. No LAD: 100
LCX: 80
39.7
9 Inferior
Lateral
5
148
Thr. No LAD:70
D1: 99
LCX: 80
26.8
10 Anterosept.
Inferior
3
150
PTCA PTCA LAD: 100
LCX:70
RCA: 90
33.3
11 Anterosept. 3 Misc. No LAD: 99
RCA: 60
25.5
12 Inferior 230 Hep. No LAD: 70
LCX: 60
RCA: 100
19.9
13 Inferior 2 Hep. CABG LCX: 100
RCA: 100
47.6
14 Anterior
Inferior
6
350
Thr. CABG LAD: 100
D1: 70
LCX: 70
43.0
15 Anterosept. 58 Hep. CABG LAD: 100
LCX: 50
L1: 70
39.3
Coronary angiography and contrast ventriculography
Cardiac catheterization was performed using the Judkins technique and recorded on videotape. The
degree of stenosis was determined quantitatively on at least two orthogonal views using a
computer-assisted approach that compares the stenotic segment defined by the observer with a
"normal" segment in the same vessel and expresses the result as a percent narrowing. Significant
stenosis was defined as a reduction of at least 50% of the luminal diameter of one of the major
36
epicardial arteries and/or of their main side branches. Patients with significant narrowing in the
ramus intermedius were considered as suffering from a two-vessel disease (left anterior descending
[LAD] and left circumflex artery [LCX]). Using this criterion, there were no main left stenosis, 2
one-vessel disease, 11 two-vessel disease and 2 triple-vessel disease.
Regional left ventricular wall motion was visually assessed by well-trained invasive cardiologists
from the contrast left ventriculogram, using left and right anterior oblique projections. Seven
myocardial segments (anterobasal, anterolateral, septal, diaphragmatic, posterobasal, posterolateral
and apical) were defined and reported to the three main arterial territories (the first three segments
were assumed to represent the territory of the LAD, the posterobasal and diaphragmatic segments
the territory of the right coronary artery [RCA] and the posterolateral segment that of the LCX, the
attribution of the apical abnormalities if present depending on the location of coexistent
abnormalities in contiguous regions). Segmental contractility was described as normal, hypokinetic,
akinetic or dyskinetic.
Scintigraphic data
Radioiodination of BMIPP was realized at the Free University of Brussels using 123I (p, 5n) and
the Cu(I)-assisted isotopic exchange reaction developed by Mertens [10]. Rest sestaMIBI and BMIPP SPET images were acquired within three days, starting 90 min
postinjection of 925 MBq for 99m
Tc -sestaMIBI, and 30 min postinjection of 166.5 MBq for 123
I -
BMIPP. Both studies were performed after a fasting period of at least 6h and without
discontinuation of any of the patients’ medications. Potassium perchlorate was administered to the
patients 15 min before the injection of BMIPP to block thyroidal uptake of free iodine.
The tomographic images were acquired using a triple-head gamma camera (Triad 88, Trionix Lab.,
Twinsburg, Ohio, USA) equipped with low-energy all-purpose collimators, by using a 360° step-
and-shoot protocol in which each head rotates over 120° and acquires 30 frames of 40 sec for
sestaMIBI and of 60 sec for BMIPP.
Prior to reconstruction, the projection images were corrected for scatter, using a scatter image
acquired in a second window just under the photopeak. For this purpose, the photopeaks were set
respectively at 140 keV for 99m
Tc (window: 126-154 keV) and at 160 keV for 123
I (window: 143-
175 keV), the scatterpeaks being acquired between 100 and 124 keV for 99m
Tc and between 114
and 141 keV for 123
I.
During the processing, the images were compensated for scatter using a substraction method with k
values for the compensation of 0.7 for 99mTc and 1.0 for 123I. These values had been preliminary
established by measurements obtained in phantom studies in our laboratory[11]. Afterwards, reconstruction and reorientation were performed as usual using a Butterworth prefilter (cutoff
frequency 0.75 for sestaMIBI and 0.6 for BMIPP, and an order of 5) and a Ramp-backprojection
filter.
Image interpretation
SestaMIBI and BMIPP series were independently normalized to their own maximum,
corresponding to the region with the highest activity. The distribution of sestaMIBI and BMIPP
uptake was visually analyzed in the three standard orthogonal tomograms and on polar maps using a
10% colour-coded scale and a side-by-side display. The anterior, septal and apical walls were
assumed to represent the territory of the LAD, the lateral wall the territory of the LCX and the
inferior wall the territory of the RCA.
37
Fig. 1. Quantitative analysis of the perfusion and metabolism in a patient with a two-vessel disease, a 5-week
old inferior and 148-week old lateral infarction (patient nr 9).
SestaMIBI (D1) and BMIPP (D2) bull’s eyes are displayed in column D, next to the nine viability polar
maps obtained from the comparison between BMIPP and sestaMIBI uptake in between different thresholds
of perfusion corresponding to normal tissue on one hand (rows 1, 2, 3), and to scar on the other hand
(columns A, B, C). Row 1: normal perfusion if > 70% of the maximum, row 2: normal perfusion if > 60% of
the maximum, row 3: normal perfusion if > 50% of the maximum. Column A: scar if < 40% of the
maximum, column B: scar if < 30% of the maximum, column C: scar if < 20% of the maximum. Normal
tissue is represented by the red colour, scar by the blue colour, mismatching > 20% by the orange colour,
mismatching between 10% and 20% by the yellow colour and matching (equal activity) between BMIPP and
sestaMIBI by the green colour.
Quantitative analysis: the viability polar maps
Starting from the bull’ eyes generated from the short axis sestaMIBI and BMIPP data, nine polar
maps, called the viability polar maps and representing the accumulation of BMIPP for different
levels of sestaMIBI uptake, were generated as follows.
In a first step, since no attenuation correction was applied, the patient’s sestaMIBI and BMIPP polar
maps were divided by a polar map of normal subjects and multiplied by 100 in order to set the
perfusion in all normal myocardial bull’s eye pixels to the same maximum (100%), assuming that
both tracers have the same distribution in normals. The polar map of normal subjects had been
generated from data obtained from a previously described population of 20 healthy volunteers with
a low likelihood (<5%) for coronary artery disease [12].
38
Fig 2. Quantification of the amount of normal, jeopardized and scar myocardium for thresholds of sestaMIBI
uptake corresponding a to > 60% for normal tissue and < 30% for scar (viability polar map displayed on
upper left corner). The second upper polar map shows the surface correction applied to take into account the
relative contribution of each pixel and of each slice of the bull’s eye to the total myocardial surface. The 5
following bull’s eyes (upper row right and mid row left) depict graphically the proportion of the left
ventricular surface corresponding to each specific category (normal, mismatching >20%, mismatching 10%-
20%, equal activity (matching) and scar), proportions which are expressed in % of the left ventricular surface
globally and for the main coronary arteries in the lower row.
Second, since there is no clear consensus in the literature concerning both the lower limit of
sestaMIBI uptake corresponding to normal tissue and the higher value of perfusion definitely
considered as scar, nine different colour-coded bull’s eyes were created that represented the nine
possible combinations obtained by varying the lowest threshold of normal uptake between 50, 60
and 70 % of the expected mean normal value on one hand, and of highest uptake corresponding to
scar between 20, 30 and 40 % on the other hand. The lower limit of normal tissue, defined as a
minimal uptake of sestaMIBI of respectively 50, 60 or 70 %, was represented in red. Scar tissue,
defined as an uptake of sestaMIBI of less than respectively 20, 30 or 40 %, was represented in blue.
In between the defined lower “normal” and upper “scar” limits, the presence and degree of
mismatching between sestaMIBI and BMIPP (representing the amount of jeopardized myocardium)
was quantified and expressed by colour codes. Green was used for matched regions, where the
difference between the uptake of both tracers was less than 10 %. Yellow indicated regions with a
BMIPP uptake between 10 and 20 % lesser than that of sestaMIBI, and orange corresponded to
39
segments where the mismatching was beyond 20 % (Figure 1). Using the polar map display, the
amount of normal, jeopardized and scar tissue as well as the degree and extent of mismatching were
easily identified by means of the different colours.
Lastly, a quantitative analysis of the relative proportion of normal, jeopardized and scar tissue was
performed globally and for each arterial territory after surface correction (Figure 2) and expressed
in % of the left ventricular surface. This surface correction was made necessary to take into account
the surface deformation due to the bull’s eye display. For this purpose, a surface correction bull’s
eye was created. This bull’s eye, generated from data obtained in 14 normal subjects, represents the
relative contribution of each pixel of the bull’s eye to the total myocardial surface as well as of each
bull’s eye slice to the total myocardial surface.
Intra- and interobserver reproducibility
To evaluate the reproducibility of the quantification, intra- and interobserver variability were
assessed respectively by one observer processing the data twice with a 3-weeks interval without
knowledge of his first results, and by two observers unaware of each other reorientation or number
of slices used to create the bull’s eyes.
In both cases, BMIPP and sestaMIBI data were completely reprocessed in all patients starting from
the transaxial tomograms, using a standard software dual-tomo program to reorient the slices. In
this way, the same rotation was applied to all the transaxial images for each individual processing,
and the same number of short axis slices was used to create both sestaMIBI and BMIPP bull’s eyes.
However, the rotation angle as well as the number of slices could differ from one processing to the
other.
The variability of the obtained measurements was evaluated for three categories of tissue: normal,
jeopardized and scar. For the present study, normal tissue was defined as showing a sestaMIBI
uptake of >60% of the mean normal expected value (red colour on the polar map display),
jeopardized tissue as showing a significant mismatching and a sestaMIBI uptake between 30 and
60% (corresponding to the addition of the yellow (10-20% mismatching) and orange (>20%
mismatching) colours of the viability polar map), and scar as the sum of regions with an uptake of
sestaMIBI below 30% and those with matched sestaMIBI and BMIPP uptake (blue + green
colours), since it has been suggested that matching reflected nonviable tissue[5].
Statistics
Mean differences and their standard deviations (SD) were calculated to compare the measurements.
The mean difference constitutes an estimate of the average bias of one result relative to the other,
while the SD represents the likelihood of agreement between both values for an individual patient.
The 95% limits of agreement are given by the mean difference +/- 2 SD. Scatter diagrams of the
data were generated to compare the linear regression equation to the line of equality. The more
informative Bland-Altman analysis was used to evaluate the eventual systematic error in the
measurement of the left ventricular surface[13]. Briefly, this method plots the mean of the paired
observations on the abcissa against the difference between their value on the ordinate. In this way, it
depicts the degree of agreement between the two measurements and the degree of bias of one result
related to the other, and checks visually whether or not the differences are related to the size of the
sampling, using the average as the best estimate of the unknown value. Using this method, a r value
nearing 0, corresponding to an absence of bias between the two observations, must be achieved.
Wilcoxon signed rank test and linear regression analysis were used to evaluate the differences. A p
value of 0.05 or less was considered significant.
40
Fig 3. Scatter plot diagram of the correlation between the percentage jeopardized tissue measured by the two
observers for the global left ventricular surface, quantifying the degree and extent of mismatching for values
of sestaMIBI uptake between 30% and 60%, the values below 30% being considered as myocardial scar and
beyond 60% as normal tissue. The two estimates are highly correlated, with a r value of 0.94.
Results
Intra- and interobserver reproducibility
Using the 30-60% cutoff of sestaMIBI uptake, the amount of mismatching in the left ventricle
ranged between 0.3% and 24.6% of its total surface.
The mean intraobserver variability in assessment of the percentage mismatching in the 15 patients
was 0.25 % (SD: 2.0%) for the left ventricle globally (p=0.68), and 1.65% (SD: 2.9%, p=0.02), -
1.56 % (SD: 3.6 %, p=0.18) and -1.24% (SD: 2.8%, p=0.05) for the LAD, LCX and RCA territories
respectively.
For the assessment of the interobserver reproducibility, the average intraobserver measurement was
compared to the % mismatching obtained by the second observer, resulting in a mean difference of
0.91% (SD: 2.4%) for the total left ventricle surface (p=0.23), -1.51% (SD: 3.0%) for LAD
(p=0.15), -0.53% (SD: 2.9%) for LCX (p=0.68) and -0.34% (SD: 3.9%) for RCA territories
(p=0.91). The correlation coefficients for the left ventricle globally were 0.94 when the two
measurements were performed by two different observers, and 0.96 when the processing was
repeated by the same observer, with standard error of estimate of 2.55% and 2.08% respectively,
and a linear regression analysis showing a slope nearing the unit and a quite small intercept in both
cases as depicted in Figure 3 for the global interobserver variability. This good reproducibility was
confirmed by the Bland-Altman analysis showing small inter- and intraobserver variabilities for the
total left ventricle as well as for each arterial territory (with mean intra- and interobserver
differences +/- SD of 0.25 +/- 2.01% and -0.79 +/- 2.46% respectively), and very similar differences
regardless of the total amount of mismatching for both intra- and interobserver measurements
(interobserver variability: y = -0.04x - 0.51, r = 0.12, Figure 4).
Evaluating the intra- and interobserver reproducibility for the normal and scar tissue, the same
range of mean differences was found, without any statistically significant difference between the
measurements (Table 2).
41
Fig 4. Bland-Altman representation of the interobserver difference in assessment of the amount mismatching
for the left ventricular surface globally, showing a quite small mean +/- SD difference and almost no
relationship between the % mismatching and the total amount of jeopardized tissue (r=0.12).
Quantitative assessment of the presence and degree of mismatching
For the quantitative assessment of the mismatching, in keeping with the 2 standard deviation cutoff
criterion of positivity classically used to define the presence of ischemia when comparing stress and
rest imaging [14], it was decided that the percentage of jeopardized tissue should reach at least the value of the second standard deviation of the interobserver variability to be considered significant
(4.8 % of the left ventricular surface globally, 6.0% for the LAD, 5.8 % for the LCX and 7.8 % for
the RCA). Using these values, significant mismatching was noted in at least one arterial territory in
8 patients (LAD territory in 5 cases, LCX and RCA in 4 cases each). In all cases, the regions with
significant mismatching corresponded to significant stenosis and/or wall motion abnormalities on
coronary angiogram (Table 3). Notheworthy, the age of the myocardial infarction was not
necessarily related to the presence and extent of mismatching, since large amounts of jeopardized
tissue were observed in the LCX region (57.0%) and in the RCA territory (27.6%) in 2 patients
(patients 9 and 10) with 3-year old lateral and inferior MI respectively.
Correlation with the coronary anatomy
A good concordance was noted between the regions with decreased BMIPP and/or sestaMIBI
uptake (regardless of the presence of significant mismatching) on one hand, and the presence of
significant stenosis and/or wall motion abnormalities on coronary angiography on the other hand
even in the RCA territory (which could be expected to be less sensitive due to interference of liver
activity and its possible influence on the test accuracy).
42
Table 2. Intra- and interobserver variability for the assessment of normal, jeopardized or scar tissue, normal
tissue corresponding to a sestaMIBI uptake of more than 60% of the expected mean normal value,
jeopardized to a mismatching BMIPP-sestaMIBI of at least 10% in the sestaMIBI uptake range 30-60%, and
scar to a perfusion of less than 30% or a matching BMIPP-sestaMIBI. The results represent the mean
(standard deviation) surface (in %) corresponding to each category of tissue for the left ventricle globally as
well as for each major arterial territory. (LV: left ventricle; LAD: left anterior descending artery; LCX: left
circumflex artery; RCA: right coronary artery.)
Intraobserver variability (%) Interobserver variability (%)
Normal Mismatching Scar Normal Mismatching Scar
Total LV 0.39 (3.6) 0.25 (2.0) -0.63 (3.3) 0.71 (3.2) 0.91 (2.4) 0.21 (3.9)
LAD -0.33 (5.4) 1.65 (2.9) -0.69 (5.5) 2.03 (4.3) -1.51 (3.0) -0.72 (5.0)
LCX 0.53 (1.8) -1.56 (3.6) 1.08 (4.2) 0.04 (2.7) -0.53 (2.9) 0.49 (3.5)
RCA 0.89 (4.9) -1.24 (2.8) 0.33 (6.1) -2.01 (5.1) -0.34 (3.9) 2.4 (6.1)
Indeed, significant stenoses were present in the LAD territory in 14 patients, all showing wall
motion abnormalities (12 akinetic, 2 hypokinetic), in the LCX territory in 11 patients, only 4 of
them showing an impaired motion (hypokinesis in all cases), and in the RCA territory in 5 patients,
with only 3 showing motion disturbances (1 akinetic and 2 hypokinetic) while 2 other patients were
classified as having inferior akinesis despite the absence of significant stenosis after thrombolysis
(diffuse atheromatous vessel in both cases).
Using sestaMIBI, a decreased uptake (≤60% of the maximum after correction for the normal
distribution) was observed in the LAD territory in 12 of the 14 patients (all but one with akinesia)
and in the LCX territory in 4 of the 11 patients. In the RCA territory, a decreased sestaMIBI uptake
was noted in 4 of the 5 patients with significant stenosis and one of the two patients with akinesis
without significant narrowing.
Using BMIPP, a very high concordance was found with the sestaMIBI findings, although the
abnormalities observed with BMIPP were significantly more severe in 8 patients, suggesting the
presence of jeopardized but viable myocardium. A decreased uptake (≤60% of the maximum uptake
after correction for the normal distribution) was noted in all cases of impaired sestaMIBI uptake in
the LCX and RCA territories, while one more patient with LAD stenosis was correctly identified,
showing an impaired BMIPP distribution despite low normal sestaMIBI uptake. In this particular
patient (patient 12) with triple vessel disease (distal RCA 100%, proximal LAD 70% and distal
LCX 60%), severely impaired left ventricular function and a 4.5 years old inferior infarction,
besides a significant mismatching (16.4%) in the region of the infarction, a severe and extended
mismatching was found in the whole LAD territory (Figure 5) on visual analysis, corresponding to
an akinetic anteroseptal wall. This feature, which was almost not detected by the quantitative
because of a low normal sestaMIBI uptake (about 60% to 70% of the maximum), presumably
represented an area of stunned rather than hibernating myocardium as confirmed by the clinical
evolution of this patient who died immediately prior to the planned cardiac bypass operation from
an extended anterior infarction with sustained ventricular fibrillation.
43
Table 3. Concordance between coronary anatomy, wall motion abnormalities and sestaMIBI uptake in the
regions showing a significant mismatching on quantitative analysis. (LAD: left anterior descending artery;
LCX: left circumflex artery; RCA: right coronary artery.) Pat. number Localization
mismatching
% mismatching % uptake
sestaMIBI
% coronary
stenosis
Wall motion
2 LAD 16.7 40 75 Akinesis
3 LCX 9.8 40 99 Hypokinesis
6 LAD 25.4 50 99 Akinesis
9 LCX
RCA
57.0
32.3
30
50
80
Atheromatous
Hypokinesis
Akinesis
10 LAD
LCX
RCA
7.5
5.8
27.6
40
100
40
100
70
90
Akinesis
Normal
Normal
12 LAD
RCA
7.6
16.4
70
20
70
100
Akinesis
Hypokinesis
13 LCX
RCA
27.8
9.5
30
40
100
100
Hypokinesis
Hypokinesis
15 LAD 10.3 50 100 Akinesis
Discussion
Noninvasive assessment of residual viability in patients with poor left ventricular function post
myocardial infarction has become a question of major importance due to the proved influence of
revascularization on those patients’outcome[15,16]. However, since a severe ventricular dysfunction is associated with a higher peri-treatment morbidity and mortality, especially when
performing coronary bypass surgery, an accurate classification of the patients between “high” and
“low” probability of benefiting from the revascularization is warranted before referring them to the
surgeon or the invasive cardiologist. Up to now, this preselection is based mainly upon the use of
PET when available, on dobutamine stress echocardiography or on different scintigraphic protocols
using 201
Tl[17]. 99mTc-sestaMIBI, despite its recognized properties of being a marker of cellular
viability[18] has not yet gained a well defined place in the clinical diagnosis of viability, apparently
underestimating its extent compared to the other techniques[3,19]. However, a relative agreement
exists regarding the relationship between severe sestaMIBI defects (<30% of the peak activity) and
myocardial scar on one hand[4], and between a sestaMIBI uptake of >60% to 70% of the maximum
and viable tissue on the other hand[3]. The regions with an uptake situated between these two thresholds constitute a grey zone in which more or less 50% of the segments will probably benefit
from revascularization therapy.
44
Fig 5. Short axis tomograms (upper pannel), vertical (left) and horizontal (right) long axis tomograms (mid
pannel) and bull’s eye representations (lower pannel) of sestaMIBI (upper rows) and BMIPP (lower rows) of
a patient with triple-vessel disease and a 4.5 years old inferior myocardial infarction (10%-colour scale).
Visually, an extended area of mismatching is noted in the whole LAD territory (despite a low-normal
sestaMIBI uptake thereby probably representing myocardial stunning), while a small amount of mismatching
is observed in the inferior and inferoseptal wall, corresponding to hibernating tissue within the infarcted
region (quantitatively measured: 16.4% of the surface of the RCA territory).
Radioiodinated fatty acids analogues such as 123
I-BMIPP have recently been proposed as an
interesting approach for metabolic imaging of the myocardium, improving the diagnostic accuracy
of the conventional scintigraphic studies to predict the presence of inotropic reserve during low-
dose dobutamine infusion[5] and the improvement in wall motion and/or left ventricular ejection
fraction[6,7] in patients with recent myocardial infarction.
It has been suggested that BMIPP could reflect both regional myocardial perfusion and free fatty
acid metabolism. Indeed, a >90%-agreement has been reported between the uptake of BMIPP and
flow tracers such as 201
Tl or the early uptake of 11C-palmitate [20,21], assuming that BMIPP could
be used as a flow tracer besides its property of metabolic marker. In our small group of patients, this
assumption is confirmed by the very high concordance observed between BMIPP and sestaMIBI
uptake.
However, BMIPP is primarily used as a metabolic tracer, since it has been postulated that regional
mismatches could be related to alterations of the usage of fatty acids as energy substrates for the
production of high energy phosphate in myocardial regions with an low oxygen supply[22]. The combined approach of perfusion (with either
201Tl or a
99mTc-flow tracer) and metabolism with
SPET has demonstrated a good concordance with 18F-FDG PET (despite a trend toward
45
underestimation of viability[9]) and has been found a valuable predictor of long term cardiac
events[23]. It seems to correlate better with the presence of jeopardized tissue in patients with
myocardial infarction than either low-dose dobutamine echocardiography or the analysis of the
uptake of sestaMIBI alone, showing a high positive and negative acurracy[7]. One of the possible factors encoutering for the lower diagnostic accuracy of SPET viability studies
compared to PET could be the use of qualitative or semiquantitative methods in many studies to
evaluate the tracer uptake or the degree of mismatching, and to relate it to the presence of
jeopardized tissue. Indeed, using a quantitative approach, a preliminary study has reported
encouraging results regarding the predictive value of a mismatching between 123
I-BMIPP and 201
Tl
and the wall motion recovery post cardiac bypass surgery[24]. In the current study, we describe an original approach to quantify the extent of normal, jeopardized
and scar myocardial tissue in terms of % of the total left ventricular surface and of each arterial
territory, using colour-coded polar maps and sestaMIBI and BMIPP as perfusion and metabolism
tracers. In keeping with recent data from the literature[3,4], the presence and degree of mismatching
between both tracers were analyzed in regions with a sestaMIBI uptake between 30% and 60% of
the expected normal value, and both the reproducibility of the method and the relationship between
significant mismatching and the coronary anatomy were evaluated in 15 patients with severe left
ventricular dysfunction due to old myocardial infarctions. This patient population was chosen to
determine if a quantitative approach of the mismatching could more accurately identify preserved
viability in regions with severe and chronic hypoperfusion than the visual analysis, because of the
higher frequency of discordant BMIPP/201
Tl uptake reported in recent (<4 weeks old) than older
infarction[8] suggesting that this method better identifies myocardial stunning than hibernation.
Using the 30-60% perfusion threshold, a good intra- and interobserver reproducibility was
demonstrated for the assessment of normal, jeopardized and scar tissue, with quite small standard
deviations and a better interobserver agreement than reported with visual analysis[25]. Regarding the presence and extent of mismatching in the regions representing jeopardized
myocardium, a high correlation between the measurements was noted for the left ventricle surface
globally as well as for each individual arterial territory, without significant relationship between the
amount jeopardized tissue and the measured differences. Using the second standard deviation value
of the interobserver difference as threshold of significancy for the presence of clinically significant
jeopardized tissue, a very good agreement with the presence of wall motion abnormalities and/or
coronary stenosis was found, confirming the results reported by Schad et al[26]. Interestingly,
unlike Tamaki et al[8], we were unable to identify a clear relationship between the age of the infarction, the administrated therapy (either thrombolysis, angioplasty or intravenous heparin) and
the presence and degree of mismatching, since we observed extensive mismatching even in very old
myocardial infarctions (> 3 years old), hence indicating that quantitative SPET imaging with
BMIPP and sestaMIBI can clearly detect myocardium at risk even years after the acute event.
The new quantitative method described here seems a potentially interesting approach to improve the
accuracy of the fatty acid/perfusion SPET imaging to identify patients with severe left ventricular
dysfunction who are most susceptible to benefit from a revascularization procedure. However, the
exact amount of mismatching (in terms of percentage of the left ventricular surface) necessary to
predict functional recovery post revascularization requires additional follow-up studies.
References
1. La Canna G, Alfieri O, Giubbini R, Gargano M, Ferrari R, Visiolo O. Echocardiography during
infusion of dobutamine for identification of reversible dysfunction in patients with chronic coronary
artery disease. J Am Coll Cardiol 1994; 23: 617-626.
2. Ragosta M, Beller GA, Watson DD, Kaul S, Gimple LW. Quantitative planar rest/redistribution
201-Tl imaging in detection of myocardial viability and prediction of improvement of left
46
ventricular function after coronary bypass surgery in patients with severely depressed left
ventricular function. Circulation 1993; 87: 1630-1641.
3. Sawada SG, Allman KC, Muzik O, Beanlands RSB, Wolfe ER, Gross M, Fig L, Schwaiger M.
Positron emission tomography detects evidence of viability in rest technetium-99m sestamibi
defects. J Am Coll Cardiol 1994; 23: 92-98.
4. Altehoefer C, vom Dahl J, Biedermann M, Uebis R, Beilin I, Sheehan F, Hanrath P, Buell U.
Significance of defect severity in Technetium-99m-MIBI SPECT at rest to assess myocardial
viability: comparison with fluorine-18-FDG PET. Eur J Nucl Med 1994; 35: 569-574.
5. Franken PR, De Geeter F, Dendale P, Demoor D, Block P, Bossuyt A. Abnormal free fatty acid
uptake in subacute myocardial infarction after coronary thrombolysis: correlation with wall motion
and inotropic reserve. J Nucl Med 1994; 35: 1758-1765.
6. Ito T, Tanouchi J, Kato J, Morioka T, Nishino M, Iwai K, Tanahashi H, Yamada Y, Hori M,
Kamada T. Recovery of impaired left ventricular function in patients with acute myocardial
infarction is predicted by the discordance in defect size on I123-BMIPP and 201Tl SPET images.
Eur J Nucl Med 1996; 23: 917-923.
7. Franken PR, Dendale P, De Geeter F, Demoor D, Bossuyt A, Block P. Prediction of functional
outcome after myocardial infarction using BMIPP and sestamibi scintigraphy. J Nucl Med 1996;
37: 718-722.
8. Tamaki N, Kawamoto M, Yonekura Y, Fujibayashi Y, Takahashi N, Konishi J, Nohara R,
Kambara H, Kawai C, Ikekubo K, Kato H. Regional metabolic abnormality in relation to perfusion
and wall motion in patients with myocardial infarction: assessment with emission tomography using
a iodinated branched fatty acid analog. J Nucl Med 1992; 33: 659-667.
9. Kawamoto M, Tamaki N, Yonekura Y, Tadamura E, Fujibayashi Y, Magata Y, Nohara R,
Sasayama S, Ikekubo K, Kato H, Konishi J. Combined study with I-123 fatty acid and thallium-201
to assess ischemic myocardium: comparison with thallium redistribution and glucose metabolism.
Ann Nucl Med 1994; 8: 47-54.
10. Mertens J, Eersels J, Van Ryckeghem W. New high yield Cu(I) assisted I-123 radioiodination of
15(p-I-phenyl)-9-methyl pentadecanoic acid, a potential myocardial tracer. Eur J Nucl Med 1987;
13: 159-160.
11. Van Steelandt E, Dobbeleir A, Vanregemorter J. Compton scatter correction for scintigraphic
imaging. Proceedings of the 11th annual symposium of the Belgian Association of Hospital
Physicists. 1995; p23.
12. A.S. Hambye, A.Dobbeleir, E. Stulens, A. Vervaet, J. Vandevivere, P.R. Franken. 240°: why
not ? Nucl Med Commun 1996;17: 583-590.
13. Altman DG. Some common problems in medical research, method comparison studies. In:
Altman DG, ed. Practical statistics for medical research. 1st edition. London: Chapman & Hall;
1991: 396-403.
14. Iskandrian AS, Verani MS. Exercise perfusion imaging in coronary artery disease: physiology
aand diagnosis. In: Nuclear cardiac imaging: principles and aplications. 2d edition. Philadelphia:
FA Davis Company,1996: 97.
15. Gioia G, Powers J, Heo J, Iskandrian AS. Prognostic value of rest-redistribution tomographic
thallium-201 imaging in ischemic cardiomyopathy. Am J Cardiol 1995; 75: 759-762.
16. DiCarli MF, Davidson M, Little R, Khanna S, Mody FV, Brunken RC, Czernin J, Rokhsar S,
Stevenson LW, Laks H, Hawkins R, Schelbert HR, Phelps ME, Maddahi J. Value of metabolic
imaging with positron emission tomography for evaluating prognosis in patients with coronary
artery disease and left ventricular dysfunction. Am J Cardiol 1994; 73: 527-533.
17. Gimple LW, Beller GA. Myocardial viability. Assessment by cardiac scintigraphy. Cardiology
Clinics, 1994; 12: 317-332.
18. Piwnica-Worms D, Kronauge JF, Chiu ML. Uptake and retention of hexakis (2-
methoxyisobutyl isonitrile) Technetium (I) in cultured chick myocardial cells. Mitochondrial and
plasma membrane potential dependence. Circulation 1990; 82: 1826-1838.
47
19. Marzullo P, Sambuceti G, Parodi O, Gimelli A, Picano E, Giorgetti A, L’Abbate A. Regional
concordance and discordance between thallium-201 ans sestamibi imaging for assessing tissue
viability: comparison with postrevascularization functional recovery. J Nucl Cardiol 1995; 2: 309-
316.
20. Chouraqui P, Maddahi J, Henkin R, Karesh SM, Galie E, Berman DS. Comparison of
myocardial imaging with iodine-123-iodophenyl-9-methyl pentadecanoic acid and thallium-201-
chloride for assessment of patients with exercise-induced myocardial ischemia. J Nucl Med 1991;
32: 447-452
21. Kawamoto M, Tamaki N, Yonekura Y, Magata Y, Tadamura E, Nohara R, Matsumori A,
Sasayama S, Konishi J. Significance of myocardial uptake of iodine 123-labeled beta-methyl
iodophenyl pentadecanoic acid: comparison with kinetics of carbon 11-labeled palmitate in positron
emission tomography. J Nucl Cardiol 1994; 1: 522-528.
22. Takeda K, Saito K, Makino K, et al. Iodine-123-BMIPP myocardial washout during exercise in
normal and ischemic hearts. J Nucl Med 1997; 38: 559-563.
23. Tamaki N, Tadamura E, Kudoh T, Hattori N, Yonekura Y, Nohara R, Sasayama S, Ikekobu K,
Kato H, Konishi J. Prognostic value of iodine-123 labelled BMIPP fatty acid analogue in patients
with myocardial infarction. Eur J Nucl Med 1996; 23: 272-279.
24. Inubushi M, Kudoh T, Hattori N, Magata Y, Ohno N, Nishimura K, Tamaki N, Konishi T. I123-
BMIPP/Tl-201 mismatching predicts wall motion recovery after CABG: assessment with
quantitative analysis. Ann Nucl Med 1996; 10: S79.
25. Taillefer R, Iskandrian A, Verani M, Orlandi C, Davies G, Borer JA, Pippin J. Observer
variability in myocardial viability assessment with I123-IPPA: results from the multicenter IPPA
study. J Nucl Med 1996; 37: 184P.
26. Schad N, Wagner RK, Hallermeier J, Daus HJ, Vattimo A, Bertelli P. Regional rates of
myocardial fatty acid metabolism: comparison with coronary angiography and contrast
ventriculography. Eur J Nucl Med 1990; 16: 205-212.
48
3.2 Influence of high-energy photons on the spectrum of iodine-123 with low and
medium-energy collimators: consequences for imaging with 123I labelled compounds
in clinical practice.
André A. Dobbeleir1 , Anne-Sophie E. Hambÿe
1 and Philippe R. Franken
2 .
1Nuclear Medicine, Middelheim Hospital, Antwerp, and
2Nuclear Medicine, Free University of
Brussels (VUB), Belgium.
European Journal of Nuclear Medicine 1999; 26: 655-658.
Abstract. Using iodine-123 labelled radiotracers, the presence of 2.5% high energy photons causes
image deterioration due to increased scatter. To investigate the influence of these photons on image
quality, we measured the spectrum of 123
I with a medium energy (ME), a low energy all purpose
(LEAP) and a high resolution (LEHR) collimator. Even in air, using low energy collimators, a high
baseline activity was observed over the total energy detection range of the gamma camera. The 159
keV photopeak to scatter activity ratio dropped from 5.9 for ME to 3.6 and 2.9 for LE collimators.
Acquiring images with LEHR collimators with energy windows set at 159 keV and 500 keV
demonstrated that the 159 keV LEHR image is a combination of ME image of the object and of the
LEHR 500 keV image. Because of their important septal penetration and greater geometric
detection efficiency compared to the 159 keV photons of 123
I, the contribution of high energy
photons is dependent on the source-detector distance. For a small source placed in air, the scatter to
photopeak activities varied from 17.4% at 80 cm to 37.8% at 5 cm distance from a LEHR
collimator. Considering only the scatter problem, medium energy collimators are the best choice for 123
I studies. Using low energy collimators for high resolution tomography with 123
I labelled
compounds, scatter contribution from high energy photons has to be corrected for quantitative
analysis or when dual isotope studies are performed, whether or not simultaneously acquired.
Key words: iodine-123, high energy photons, spectral analysis, collimator.
49
Introduction
Iodine has become important as an isotopic label because the chemistry of iodination is versatile
and well understood. However, besides the 159 keV photons, high purity 123
I emits high energy
photons: 2.4% between 440 and 625 keV, and 0.15% between 625 and 784 keV. These high energy
photons cause septal penetration and scatter detected in the energy window of the 159 keV
photopeak. Therefore, some authors recommended to use medium-energy (ME) collimators,
especially when quantitation is required [1,2]. Analyzing the recent literature published in this
journal using 123
I labelled dopamine receptors, fatty acids or antibodies, we noted 15 studies over a
1 year period. A large variety of collimators were used, from medium energy (ME) and fanbeam
collimators to low energy high resolution (LEHR) and all purpose (LEAP) collimators. When semi-
quantitative or quantitative analysis has to be applied, ME collimators are preferred. Alternatively
when high spatial resolution is important for accurate quantification , as for neuroreceptor SPET,
LEHR collimators supported by scatter correction are used. However Messa et al. still correct for
collimator-septa penetration while using a specialised annular geometry CERASPET system and
ME collimators for quantitative analysis of dopaminergic receptors [123
I]Beta-CIT [3].
In myocardial viability studies, iodine-123 beta-methyl-p-iodophenyl-pentadecanoic acid (BMIPP),
a free fatty acid analogue, is the most frequently used tracer for metabolic SPET studies, a
discordant uptake with less BMIPP than perfusion being well correlated with viable tissue. LEAP
[4-8] or LEHR collimators are used [9-14] although higher noise level in the 123
I image can disguise
the reduction in fatty acid uptake related to perfusion. Special care must be paid to the methodology
when quantitative measurements are intended, in order to avoid image misinterpretation due to
differences in physical characteristics [15]. One solution consists in using medium energy [16]
instead of low energy collimators, despite a certain loss of resolution [1]. Alternatively, low energy
collimators can be used, if a scatter correction is applied [17,18].
In a previously published series of 10 patients with myocardial infarction, we measured after
sequential studies 12.1% higher background activity for 123
I-BMIPP than for 99m
Tc-sestamibi using
a LEAP collimator. Performing dual window scatter correction , mean difference in activity
between technetium and 123-iodine in the infarcted area increased from 0.4% without background
correction to 6.4% after scatter subtraction, increasing dramatically the number of viable territories
[19]. In an attempt to clarify the image degradation with iodine-123 and low energy collimators, we
performed a few phantom studies.
Materials and methods
All data were acquired with a Trionix Triad 88 triple headed gamma camera using LEHR, LEAP
and ME collimators. Table 1 presents the specifications of these collimators for 99m
Tc according to
the manufacturer, as well as resolution measurements for 123
I determined in our department. The
Trionix collimators has been fabricated from lead foil by Precision (Tennessee, USA).
In a first time, the spectrum of 123
I in air was measured sequentially for each collimator with the
source at 10 cm from the same detector. A 2-ml volume source containing 74 MBq (2.0 mCi) of 123
I
was used. For each acquisition, the photopeak window was placed between 143 and 175 keV and
the scatter window just below, between 116 and 142 keV. As a comparison, the same measurements
were repeated in air for 99m
Tc, with the photopeak window between 126 and 154 keV and the
scatter window between 100 and 125 keV. The ratio photopeak to scatterpeak activity was
calculated in order to compare the signal to noise ratio for each collimator. To eliminate the
eventual effect of intrinsic detector performance, this study was repeated for the three detectors.
In a second time, the 2-ml volume 123
I source was placed in air at the center of the three detector
gamma camera, at an equal distance of 20 cm from each detector, and three simultaneous planar
images were obtained. The first detector was equipped with a LEHR collimator and the photopeak
was set at 159 keV. The second detector was equipped with a ME collimator with the same energy
50
window as the first, whereas for the third detector, also fitted with a LEHR collimator, the 20%
photopeak window was placed at 500 keV.
Finally the influence of distance on septal penetration for the 3 collimators was investigated. The
spectra of 123
I in air was measured for the 2-ml volume source placed at variable distances (5, 10,
20, 40 and 80 cm) of each of the 3 collimators. This was repeated for each detector. Scatter to
photopeak ratio was calculated as an indication of background observed in clinical studies.
Table 1. Collimator specifications for
99mTc according to the manufacturer
and measured data for 123I between brackets.
System resolution
in mm @ 10 cm
FWHM FWTM
Sensitivity
cpm / µCi
Septal
Penetration
LEAP 9.5 (9.9) 17.1 (18.3) 281 0.4 %
LEHR
7.1 (7.4) 13.0 (14.5)
132
0.2 %
MEAP
9.7 (10.1) 16.5 (17.1)
165
1.5% (67Ga)
Results
As expected, the spectra for 99m
Tc were very similar with the 3 types of collimators. The photopeak
to scatter ratio was close to 14 to 1 (14:1) in air regardless of the used collimator. The mean values
for the three detector/collimator combinations are presented in Table 2.
The spectra of iodine-123 acquired in air without scatter medium for each kind of collimator are
shown in Figure 1. The position and width of the photopeak (143-175 keV) and scatter window
(116-142 keV) are superimposed on the spectra. A small influence of the high energetic photons is
visible on the spectrum using medium energy collimator. For the low energy collimators , the high
energetic peak becomes more important in comparison to the 159 keV photopeak, and even in air a
large amount of scatter is present over the total energy range. The mean photopeak to scatter ratio is
5.9:1 with ME collimator and drops to 3.6:1 with LEAP collimator and to 2.9:1 with LEHR
collimator (Table 2).
The influence of high energy photons on the image quality is illustrated in Figure 2, showing an
image of the 123
I source in air acquired with ME and LEHR collimators; the photopeak was set at
159 keV. The third image is obtained with the LEHR collimator and a 20% energy window set at
500 keV. The 159 keV photopeak images are overexposed to attract the attention to the differences
in septal penetration between ME and LEHR collimators. The difference in scatter activity observed
at 159 keV between the two collimators resemble the scatter distribution observed at 500 keV. The
counts measured in a large region drawn over the source acquired using the LEHR collimator
amounted to 71 kcounts in the 500 keV and 1.49 Mcounts in the 159 keV image ( ratio 500 keV /
159 keV = 4.8% ), while the activity over the total 40-20 cm field of view raise to 1.00 Mcounts for
500 keV and 2.15 Mcounts in the 159 keV photopeak image (ratio 500 keV / 159 keV = 46.5%). In
the 159 keV image with the ME collimator 1.70 Mcounts was measured in the region over the
source and 1.81 Mcounts for the total field of view, an increase of only 6.5% compared to 44% with
the LEHR collimator. The cross-shaped artefact seen in both images with the LEHR collimator
represents penetration of the high energy photons through the thinner septa.
51
In Figure 3, the influence of distance on septal penetration by the high energy photons is depicted
for a 123
I point source in air using LEHR collimator. This figure clearly demonstrates that scatter
from the high energetic photons is distance-dependent, the closer to the collimator the more
important the scatter to photopeak ratio, ranging from 17.4% for 80 cm to 37.8% for 5 cm distance.
In Table 3, the scatter fraction for each distance and each collimator is presented.
Table 2. Activity ratio of photopeak on scatter window for
ME, LEHR and LEAP collimators for 123I in comparison
to a 99m
Tc source. Mean value for the three detectors
ME LEHR LEAP
99mTc in air
13.8 : 1
14.1 : 1
14.0 : 1
123I in air
5.87 : 1
2.88 : 1
3.60 : 1
Fig. 1. The spectra of iodine-123 in air with medium energy all purpose (ME) and high
resolution (HR) low energy collimators. The position and width of the photopeak
(143-175 keV) and scatter window (116-142 keV) are also shown.
52
Fig 2. The difference in scatter at 159 keV between LE and ME collimators is shown to be related to the
high-energy 123I photons.
Fig 3. Influence of source-collimator distance on septal penetration by high-energy photons, for a
123I source
in air using LEHR collimator. The distances considered are 5, 10, 20, 40 and 80 cm.
2 78 153 229 304 380 456 keV
Spectra159 keV
High energy photons
5 cm
80 cm distance
Discussion
We demonstrated that with 123
I and low energy collimators a larger contribution of scatter is
observed from sources closer to the detector. On the images of Figure 2, it can be seen that the
photons detected around 159 keV are the sum of true 159 keV photons and scattered photons from
the high energies. For the true 159 keV photons, those passing through the holes of the collimator
are only the ones travelling perpendicular to the detector. Changing the distance will only have a
53
Table 3. Ratio scatter on photopeak 123I activity (in %) for
3 collimators and variable distance in air between source
and gamma camera. Mean value for the three detectors.
Colli-
mator
5 cm 10 cm 20 cm 35 cm 80
cm
LEHR
37.8
34.7
29.0
22.5
17.4
LEAP
30.0
27.7
22.7
17.9
15.3
ME
18.2
17.1
16.8
15.4
15.1
small effect on the number reaching the crystal. On the contrary, due to septal penetration, high
energetic photons behave themselves as if the gamma camera was without a collimator and the
number of photons detected decrease by the inverse of the square of the distance. Despite the small
proportion of high energetic photons (about 2.5%), their greater geometric detection efficiency
compared to 159 keV photons make their relative contribution in the detected photons very
important if the distance between the camera and the object is small. This can explain why De
Geeter et al. [1] measured a relative sensitivity for a LEHR compared to a ME collimator of 0.74 for 99m
Tc whereas for 123
I a value of 1.53 was obtained. They used a plastic dish placed directly on the
collimator surface. Another-distance dependent effect is the changing effective path length through
the septa of the collimator depending on the angle of incidence of the photon and thus modifying
attenuation of high energy photons in the lead of the collimator. From the point of view of scatter,
using a low energy high resolution collimator for 123
I imaging is the worst solution because of a
poor photopeak to scatter ratio (2.9:1). Due to a higher photopeak sensitivity and almost equal
septal penetration, the low energy all purpose collimator has a slightly better ratio (3.6:1). Using
the medium energy collimator, the septal penetration becomes very low and the ratio raises to 5.9:1.
Since the observed results are collimator-dependent, slight differences could be expected with
collimators from other manufactures, mainly depending on the septal length and thickness.
Furthermore, changing the position and width of the scatter window will provide different
photopeak to scatter ratios. Although, this will not change the general findings observed when using
low-energy collimators for imaging with 123
I radioiodinated tracers.
In summary, the influence of high energy photons on the image depends on collimator specificities
and geometric factors (organ-detector distance). Considering only the scatter problem, medium
energy collimators are the best choice for 123
I studies. However for neuroreceptor studies, where
high spatial resolution is important to avoid partial volume artefacts for accurate quantification,
LEHR parallel or fanbeam collimators are prefered. For quantitative analysis or when comparing a 123
I-labelled compound to another isotope scatter correction is important when low energy
collimators are used.
References
1. De Geeter F, Franken PR, Defrise M, Andries H, Saelens E, Bossuyt A. Optimal collimator
choice for sequential iodine-123 and technetium-99m imaging. Eur J Nucl Med 1996; 23: 768-774.
2. Macey DJ, DeNardo GL, Denardo S, Hines HH. Comparison of low- and medium-energy
collimators for SPECT imaging with iodine-123-labelled antibodies. J Nucl Med 1986; 27: 1467-
1474.
54
3. Messa C, Volonte M, Fazio F, Zito F, Carpinelli A, d'Amico A, Rizzo G, Moresco R, Paulesu E,
Franceschi M, Lucignani G. Differential distribution of striatal [123
I]Beta-CIT in Parkinson's disease
and progressive supranuclear palsy, evaluated with single-photon emission tomography. Eur J Nucl
Med 1998; 25: 1270-1276.
4. Ito T, Tanouchi J, Kato J, et al. Recovery of impaired left ventricular function in patients with
acute myocardial infarction is predicted by the discordance in defect size on 123
I -BMIPP and 201
Tl-
SPET images. Eur J Nucl Med 1996; 23: 917-923.
5. Tamaki N, Tadamura E, Kudoh T, et al. Prognostic value of iodine-123-labelled BMIPP fatty
acid analogue in patients with myocardial infarction. Eur J Nucl Med 1996; 23: 272-279.
6. Tadamura E, Kudoh T, Hattori N, Inubushi M, Magata Y, Konishi J, Matsumori A, Nohara R,
Sasayama S, Yoshibayashi M, Tamaki N. Impairment of BMIPP uptake precedes abnormalities in
oxygen and glucose metabolism in hypertrophic cardiomyopathy. J Nucl Med 1998; 39: 390-396.
7. Kawai Y, Tsukamoto E, Nozaki Y, Kishino K, Kohya T, Tamaki N. Use of 123
I -BMIPP single-
photon emission tomography to estimate areas at risk following successful revascularization in
patients with acute myocardial infarction. Eur J Nucl Med 1998; 25: 1390-1395.
8. Yoshida S, Ito M, Mitsunami K, Kinoshita M. Improved myocardial fatty acid metabolism after
coronary angioplasty in chronic coronary artery disease. J Nucl Med 1998; 39: 933-938.
9. Vanzetto G, Janier M, Fagret D, Cinotti L, Andre-Fouet X, Comet M, Machecourt J. Metabolic
myocardial assessment with iodine 123-16-iodo-3-methyl hexadecanoic acid in recent myocardial
infarction: comparison with thallium-201 and fluorine-18 fluorodeoxyglucose. Eur J Nucl Med
1997; 24: 170-178.
10. Takeishi Y, Chiba J, Abe S, Tonooka I, Komatani A, Tomoike H. Heterogeneous myocardial
distribution of iodine-123 15-(p-iodophenyl)-3-R,S-methylpentadecanoic acid (BMIPP) in patients
with hypertrophic cardiomyopathy. Eur J Nucl Med 1992; 19: 775-782.
11. Nakata T, Hashimoto A, Kobayashi H, Miyamoto K, Tsuchihashi K, Miura T, Shimamoto K.
Outcome significance of thallium-201 and iodine-123-BMIPP perfusion-metabolism mismatch in
preinfarction angina. J Nucl Med 1998; 39: 1492-1499.
12. Kobayashi H, Kusakabe K, Momose M, Okawa T, Inoue S, Iguchi N, Hosoda S.
Evaluation of myocardial perfusion and fatty acid uptake using a single injection of iodine-123-
BMIPP in patients with acute coronary syndromes. J Nucl Med 1998; 39: 1117-1122.
13. Taki J, Nakajima K, Matsunari I, Bunko H, Takata S, Kawasuji M, Tonami N. Assessment of
improvement of myocardial fatty acid uptake and function after revascularization using iodine-123-
BMIPP. J Nucl Med 1997; 38: 1503-1510.
14. Takeishi Y, Fujiwara S, Atsumi H, Takahashi K, Sukekawa H, Tomoike H. Iodine-123-BMIPP
imaging in unstable angina: a guide for interventional strategy. J Nucl Med 1997; 38: 1407-1411.
15. Tamaki N, Kawamoto M, Yonekura Y, et al. Regional metabolic abnormality in relation to
perfusion and wall motion in patients with myocardial infarction: assessment with emission
tomography using a iodinated branched fatty acid analog. J Nucl Med 1992; 33: 659-667.
16. Franken PR, Dendale P, De Geeter F, Demoor D, Bossuyt A, Block P. Prediction of functional
outcome after myocardial infarction using BMIPP and sestamibi scintigraphy. J Nucl Med 1996;
37: 718-722.
17. Gilland DR, Jaszczak RJ, Turkington TG, Greer KL, Coleman RE. Volume and activity
quantification with iodine-123 SPECT. J Nucl Med 1994; 35: 1707-1713.
18. Takeda K, Saito K, Makino K, Saito Y, Aoki S, Koji T, Matsumura K, Nomura Y, Kitano T,
Nakagawa T. Iodine-123-BMIPP myocardial washout and cardiac work during exercise in normal
and ischemic hearts. J Nucl Med 1997; 38: 559-563.
19. Dobbeleir A, Hambye AS, Franken PR. Influence of the methodology on the presence and
extent of mismatching between 99m
Tc-MIBI and 123
I -BMIPP in myocardial viability studies. J Nucl
Med 1999; 40: (in press).
55
3.3 Influence of the methodology on the presence and extent of mismatching between 99m
Tc-MIBI and 123I-BMIPP in myocardial viability studies.
André A. Dobbeleir1 MSc, Anne-Sophie E. Hambye
1 MD and Philippe R. Franken
2 MD, PhD.
1Nuclear Medicine, Middelheim Hospital, Antwerp, and
2Nuclear Medicine, Free University of
Brussels, Brussels, Belgium.
Journal of Nuclear Medicine 1999; 40: 707-714.
Abstract
Discordant uptake (mismatching) of 123
I-BMIPP less than 99m
Tc-MIBI is a good predictor of
myocardial viability. However, methodological factors can influence the assessment of the presence
of mismatching because of differences in background activity between the tracers. In the present
study, we investigated the influence of methodological parameters on the mismatching between
BMIPP and MIBI in patients with chronic ischemic heart disease.
Methods. Polar maps were created to quantify the extent of mismatched tissue measured in 10
patients with myocardial infarction according to different methods for data processing: no
correction, substraction of the background activity measured in the left ventricle cavity, and dual
window scatter correction. Mismatching was expressed in % of the surface of the left ventricle
globally as well as for each arterial territory using a BMIPP uptake of at least 10% less than MIBI
as threshold. The results of dobutamine stress echocardiography and the evolution of the regional
contractility at 6 months follow-up were used as references.
Results. Mean background activity in the ventricle cavity was 9.3% of the maximum activity for
MIBI and 21.4% for BMIPP before, and 2.8% and 8.3% after scatter correction. Fourteen arterial
vascular territories demonstrated baseline wall motion abnormalities, of which 9 with contractile
reserve under dobutamine. Significant mismatching was found in 5/14 regions without correction,
9/14 after scatter correction and 13/14 after background substraction. Compared to the evolution of
resting regional contractility at follow-up, optimal results were found when using the scatter
corrected data. Without correction, mismatching between BMIPP and MIBI was partially disguised
because of the higher noise level in the iodine images. On the contrary, substraction of background
measured by means of a single ROI overestimated the magnitude of mismatching due to the
heterogeneous background distribution in the ventricular cavity.
Conclusion. Quantifying the presence and extent of mismatching between MIBI and BMIPP in
chronic ischemic heart disease, significant differences in the detection of viability are noted
according to the acquisition and processing methods used. Scatter correction of the acquisition data
is the most accurate and liable method for identifying viable myocardium.
Key words: iodine-123-BMIPP, chronic ischemic myocardium, myocardial viability,
quantification.
56
Using 99m
Tc-MIBI for evaluation of myocardial viability, a normal uptake is a good predictor of
functional recovery but a decreased uptake clearly underestimates tissue viability (1). Furthermore,
the lower limit of normal with this tracer is not precisely defined and fluctuates between 50% and
60% (2,3).
Due to an early metabolic switch from beta-oxydation to glycolysis to preserve the
production of high energy phosphates in hypoxic myocardium, single-photon emission computed
tomography (SPECT) with radioiodinated free fatty acid analogues has been proposed as an
alternative to 18F-Fluorodeoxyglucose for imaging cardiac metabolism (4), a mismatching with fatty
acid uptake more severely reduced than the flow reliably identifying jeopardized but viable
myocardium (4-7).
However, comparing the tissue uptake of radiopharmaceuticals labeled with different
isotopes requires a special attention for the acquisition parameters. Particularly, the use of 123
I-
labeled compounds can lead to an underestimation of the defect contrast, due to the emission of
2.5% high energy photons (440-625 keV). When using low-energy collimators, these photons cause
septal penetration and scatter, partly detected in the 159 keV photopeak window. This phenomenon
results in an increased background level in the 123
I image, that can disguise partly or completely the
reduction in fatty acid uptake related to the perfusion tracer, or even result in an increased fatty
acid/perfusion uptake as recently reported by Sloof et al in chronic ischemic heart disease (8). To
overcome this problem when acquiring 123
I-imaging, one solution consists in using medium-energy
instead of low-energy collimators, despite a certain loss of resolution (9). Alternatively, low-energy
high-resolution collimators can be used, even for quantitative 123
I SPECT imaging, if a scatter
correction is applied (10).
The present study aimed at clarifying the influence of methodology on the quantitative
assessment of mismatching between iodine-123-beta-methyl iodophenyl pentadecanoic acid
(BMIPP) and 99m
Tc-MIBI. For this purpose, we quantified the presence and degree of mismatching
in ten patients with prior transmural myocardial infarction, using different methodological
approaches and a previously described quantitative method (11), and compared the results to the
findings of dobutamine stress echocardiography. We used low-energy all-purpose instead of high
resolution collimators because the amount of septal penetration is similar with both kinds of low-
energy collimators when 123
I-labeled compounds are used, but the photopeak sensitivity is higher
with the all purpose, resulting in a significant improvement of the signal/noise ratio.
Materials and methods
Study design
Ten patients with old transmural myocardial infarction (median: 3 months, range: 1 month-
10years) were included. Within a week, coronary angiography, dobutamine stress
echocardiography, radionuclide angiography, BMIPP and MIBI studies were obtained (the last two
with a 3-day interval).
Using echocardiography, the left ventricle was divided into 8 segments. The anterobasal,
anterolateral and anteroseptal segments were ascribed to the left anterior descending artery (LAD),
the posterolateral to the left circumflex artery (LCX), and the diaphragmatic, posteroseptal and
posterobasal to the right coronary artery (RCA), while the apex was attributed to the LAD unless a
dominant RCA or LCX was reported. In each arterial territory, wall motion was analyzed at rest and
during intravenous infusion of 5 and 10 µg/kg.min dobutamine and graded as normal, slightly,
mildly or severely hypokinetic, or a/dyskinetic. Viability was defined as an increase in wall motion
of at least one grade in a segment with resting abnormalities.
After completion of the tests, revascularization was performed within the month in 6
patients in whom it was technically feasible, the remaining being conservatively treated. Follow-up
data was obtained 6 to 7.5 months later.
57
All patients received written information about the study and gave informed consent. The
study protocol had been approved by the Commission of Medical Ethics of the Hospitals of
Antwerp.
Scintigraphic imaging protocol
Radioiodination of BMIPP was realized at the Free University of Brussels using 123
I (p,5n) and the
Cu(I)-assisted isotopic exchange reaction developed by Mertens (12).
123
I-BMIPP was intravenously injected in resting condition at a mean dose of 159 MBq (4.3
mCi) after at least 6h fasting. Potassium perchlorate was administered to the patients 15 min before
injection to block thyroidal uptake of free iodine. SPECT started 30 min postinjection using a triple-
head gamma camera (Triad, Trionix Lab, Twinsburg, USA), detector size 40*20 cm, equipped with
all-purpose low-energy collimators. Ninety projections (30/head) of 60 sec duration were acquired
over a 360° non-circular body contour orbit, using a 128*64 matrix. A scatter image was obtained
in a second window just under the photopeak according to the method of Jaszczak (13). The
photopeak image was set at 159 keV with a window between 143 and 175 keV, and the scatter
image acquired between 116 and 142 keV.
Resting 99m
Tc-MIBI SPECT was started at a mean time of 80 min postinjection of 925 MBq
(25 mCi) using a similar protocol as for the BMIPP, but with 40 sec acquisition time per projection
and different photo- and scatterpeak characteristics (photopeak set at 140 keV with a window
between 126 and 154 keV; scatter window between 100 and 125keV).
Determination of the k fraction for dual window scatter correction
In SPECT, scatter compensation consists in substracting a fraction k of the compton image C(x,y)
from the photopeak image P(x,y) to obtain a scatter compensated image I(x,y) = P(x,y) - k*C(x,y).
The value of this k factor depends on the acquisition geometry, the energy resolution of the camera,
the energy settings and the size of the object (14).
Using the previously described collimators and energy window settings, the projection
version of the dual window scatter correction was implemented by performing weighted
substractions with an experimentally determined k value directly on the projection images, before
reconstruction (15).
Experimental determination of this k value was performed as follows. Using the same
window settings and collimators that would be used for the patients studies, the scatter substraction
fraction was calculated for 99m
Tc and 123
I and different source geometries. Scatter correction was
performed on planar images since SPECT data consist of a set of planar images and quantitative
distortions will propagate in the tomographic studies.
Four different source geometries were tested for both isotopes: a small point source, a 5 cm
diameter-1 cm thick cylinder, a 10 cm diameter-2 cm thick cylinder and a large 10 cm diameter-10
cm thick cylinder. Images were acquired in air and with increasing depths of attenuating medium: 4,
8, 12, 16 and 20 cm of water.
When the logarithm of the counts are plotted against depth, the resulting straight line has a slope
equal to the attenuation coefficient for broad beam geometry, and after accurate elimination of
scatter, for narrow beam geometry (16). Therefore, k was iteratively changed for each source and
the subsequent scatter corrected counts were fitted to the single monoexponential function up to the
attenuation coefficient corresponded to the values for water for the two isotopes (0.15 cm-1 for
99mTc and 0.146 cm
-1 for
123I).
Processing and analysis of the scintigraphic data
The extent of viable tissue was evaluated for 4 different types of processing: 1°) Without correction.
2°) After background substraction performed as follows: a 1-cm2 region of interest (ROI) was
drawn at the basal part of a 1 cm thick midventricular long axis slice on both MIBI and BMIPP
images and the activity measured within these ROI’s was subtracted from the data before bull's eyes
were created. 3°) Using a dual window scatter correction with k values of 0.7 for 99m
Tc and 1.0 for
58
123I. 4°) With the same scatter correction method but a substraction constant k=1.3 instead of 1.0 for
123I. These 2 values of k for
123I, corresponding respectively to the point source and the to 10 cm
diameter-10 cm height phantom, had been found the minimum and maximum correction fractions
in the experimental measurements.
The three standard orthogonal tomograms were obtained after filtered backprojection and
appropriate reorientation of the images using a Butterworth prefilter (cutoff frequency 0.75 cyc/cm
for MIBI and 0.6 cyc/cm for BMIPP, order 5) and a Ramp-backprojection filter, and polar maps
were created.
Quantification of the mismatching
Color-coded polar maps allowing a quantitative analysis of the relative proportion of normal,viable
and scar tissue for the left ventricle globally and for each arterial territory were created according to
a previously described method (11). By quantifying the magnitude of viable tissue measured by the
four processing methods, the influence of methodology on the quantitative assessment of
mismatching between 123
I-BMIPP and 99m
Tc-MIBI could be evaluated.
These polar maps, comparing the MIBI and BMIPP images, are obtained as follows. The
lower limit of normal MIBI uptake is defined as 60% of the normal local value (2) and represented
in red, while scar tissue, defined as a MIBI uptake of <30%, is represented in blue. In between these
two limits, MIBI is compared to BMIPP to evaluate the presence and degree of mismatching. Green
is used for matched regions, meaning regions were the difference between BMIPP and MIBI uptake
is <10%, yellow for regions with a mismatching of BMIPP 10-20 % less than MIBI, and orange
corresponds to segments with a >20% mismatching, both considered as jeopardized but viable
tissue. The extent of mismatched tissue is considered significant if it amounts to at least 4.8 % of
the global left ventricular surface, 6.0 % of the LAD region, 5.8 % of the LCX and 7.8 % of the
RCA surface (11).
At baseline, the presence of viability by scintigraphy was compared to the findings of the
dobutamine stress echocardiography for the four processing methods. At follow-up, it was related to
the evolution of resting regional contractility assessed by echocardiography.
Results
Determination of the k fraction for scatter correction
The optimum k values for scatter correction measured for the four different sources geometries are
reported in Table 1. As expected according to the literature, smaller k values were found for larger
objects (15).
Table 1. Substraction coefficients for the dual energy window scatter correction method
Phantom 99m
Tc k fraction 123I k fraction
Point source 0.80 1.3
Cylinder D= 5 cm
H= 1 cm
0.75 1.2
Cylinder D= 10 cm
H= 2 cm
0.70 1.1
Cylinder D= 10 cm
H= 10 cm
0.70 1.0
Measurement of intraventricular cavity and infarcted area activities
Mean intraventricular activity was measured on both MIBI and BMIPP vertical midventricular long
axis slices. For MIBI, mean activity for the 10 patients was 9.3% of the maximum activity,
59
compared to 21.4% for BMIPP. After scatter correction, mean activity dropped to 2.8% for MIBI
and 8.3% for BMIPP with a k value of 1.0, indicating a slight undercorrection for the latter.
Mean tissue activity within the infarcted area for the 10 patients was measured by placing a
1-cm2 ROI over the central part of the infarcted region. The differences in mean measured activities
according to the correction method used are reported in Table 2 and are particularly prominent with
BMIPP, the mean difference in uptake between MIBI and BMIPP amounting to 0.4% without
correction, 6.4% after scatter correction and up to 12.8% after background substraction. Using this
last method, negative values of BMIPP uptake of -15.4% and -11.5% were found in two patients,
both with an extended anterior infarction. In these patients, a high basal to apical background
gradient was observed in the left ventricular cavity, responsible for this overcorrection.
Figure 1 shows the short axis, vertical long axis and bull's eyes MIBI and BMIPP images of a patient with an
history of severe heart failure (ejection fraction: 15%) due to a at least 1-year old silent anterior infarction. In
this region, MIBI uptake was almost normal (59% of the maximum activity and BMIPP uptake only slightly
reduced (55% of the maximum activity without correction). Applying scatter correction, the mean BMIPP
activity dropped to 43%, versus 39% with background substraction (respectively 16% and 20% less than
perfusion), hence making the mismatching more obvious. Six months after bypass surgery, the patient was
symptom-free, his ejection fraction had raised to 31%, and regional improvement was noted by
echocardiography.
60
Table 2: Mean activity in the infarcted area according to the correction method (in %).
MIBI BMIPP Difference
No correction 28.7 28.3 0.4
Background substraction 21.7 8.9 12.8
Scatter correction 22.1 15.7 6.4
Figure 2 shows the short axis, vertical long axis and bull's eyes MIBI and BMIPP images of a patient with a
14-week old anterior infarction and an ejection fraction of 39% at baseline. Without correction, BMIPP
uptake was >10% higher than MIBI (53.2% versus 40.9%) in the anterior and apical regions. After scatter
substraction, the activities measured in the infarcted area were almost equal for both tracers: 37.2% for MIBI
and 38.7% for BMIPP. At 6 months follow-up, the ejection fraction decreased by 2.5% and regional wall
motion remained unchanged.
Quantification of the mismatching according to the correction method
Analysis was focused on the 14 arterial territories showing resting wall motion abnormalities by
echocardiography. Using MIBI alone and a 60%-uptake as threshold for viable tissue (2), these
regions should have been considered nonviable because all of them had less than 60% MIBI uptake.
Comparing the BMIPP to the MIBI uptake in these regions, significant mismatching was observed
in 5/14 territories without correction, in 9 with both k=1.0 and 1.3 scatter correction factors and in
13 with background subtraction (Table 3). Interestingly, none of the regions supplied by nonstenotic
arteries showed significant mismatching. According to previous results based upon inter-observer
reproducibility, mismatching was considered significant if it involved at least 6.0 %, 5.8 % and 7.8
% of the surface perfused by the LAD, LCX and RCA arteries respectively (11). The extent of
61
mismatching measured by the different processing methods is represented in Figure 3 for the 14
arterial territories. This graph clearly illustates the influence of methodology on the magnitude of
viable tissue.
Table 3: Extent of mismatching in each arterial territory (in % of the surface) for the different
correction methods. Significant surface is represented in bold characters.
LAD: left anterior descending artery. LCX: left circumflex artery. RCA: right coronary artery.
No: no correction. Bg: background substraction. Scat 1.0:scatter correction with k value of 1.0 for BMIPP.
Scat 1.3: scatter correction with k value of 1.3 for BMIPP.
Arterial territories
LAD LCX RCA
No Bg Scat
1.0
Scat
1.3
No Bg Scat
1.0
Scat
1.3
No Bg Scat
1.0
Scat
1.3
Pat 1 0.5 14.1 8.4 10.8 0 0 0 0 0 0 0 0
Pat 2 0 0 0.6 3.1 0 23.8 12.4 9.2 0 2 1.7 0
Pat 3 1.7 14.1 3.4 5.5 0 0 0 0 0.2 3.9 1 1.6
Pat 4 21 36.6 35.2 39.3 0 0 0 0 0 1.9 1.9 2.2
Pat 5 1 4.6 0.8 5 14.3 67.2 15.2 42.1 4 37.4 27.7 30.8
Pat 6 0 9.1 0.7 0.8 0 0 0 0 0 2.3 0 0
Pat 7 0 7.3 3.2 4.5 0 1.1 1.3 0 0 0 0 0
Pat 8 15.8 37.5 30.2 32.7 9.1 22.3 15.3 17.2 14.2 31.9 25.3 28.4
Pat 9 1.8 27.9 14.6 22.3 0 0 0 0 0 0 0 1.3
Pat
10
0 1.5 0 0 0.3 11.2 0.6 1.8 0.4 7.2 5.2 5.4
When the two k values for scatter correction with 123
I were compared, a small difference in
the % mismatched surface was noted. However this did not modify the classification of the
concerned territories as either significantly mismatched or not.
As a comparison, dobutamine stress echocardiography demonstrated evidence of contractile
reserve in 9 territories.
At follow-up, echocardiographic improvement was noted in all revascularized regions,
versus in 1 nonrevascularized (Table 4). Accuracy of the different processing methods to predict the
evolution of regional contractility at follow-up was 64% when no correction was applied, 93% for
scatter correction and 79% for the background correction (versus 93% for dobutamine stress
echocardiography).
Discussion
When the uptake of 99m
Tc-labeled sestamibi and 123
I-BMIPP are quantitatively compared in the
setting of myocardial viability assessment, special attention must be paid to the acquisition
parameters. Indeed, comparing the data without correction underestimates the magnitude of viable
myocardium, and appliance of a background correction overestimates it. Dual window scatter
correction constitutes the optimal approach, predicting functional outcome with the same accuracy
as dobutamine stress echocardiography in patients with chronic ischemic heart disease and
myocardial infarction.
62
Figure 3: Extent of the mismatched surface for the 14 arterial territories (in %) according to the different
correction methods. No: no correction. Bg: background substraction. Scat 1.0: scatter correction with k value
of 1.0 for BMIPP. Scat 1.3: scatter correction with k value of 1.3 for BMIPP
Free fatty acid metabolism and its relationship to myocardial viability studies
In myocardial viability assessment, a normal uptake of sestamibi, usually defined as at least 60% of
the peak activity (2) identifies residual viable tissue with a high accuracy. However, a decreased
uptake is a rather poor predictor of myocardial scar, and additional metabolic studies are required to
differentiate poorly perfused but still viable from fibrotic tissue. Radioiodinated free fatty acids,
suitable for SPECT imaging, are a useful tool for metabolic imaging. Their uptake, depending
primarily upon regional blood flow (17) and regulated by a membrane fatty acid-binding protein, is
followed by the ATP-dependent initial steps of the native fatty acids enzymatic activation to acyl-
coenzyme A (18), before they are esterified to triglyceride and incorporated in the endogenous lipid
pool.
Using BMIPP, prolonged myocardial retention time is obtained thanks to the presence of a
methyl-group precluding direct beta-oxidation. However, a significant proportion of the BMIPP-
CoA is beta-oxided after an intermediate alpha-oxidation step (19), and only a small amount of the
extracted BMIPP is backdiffused. In pathological conditions with impaired myocardial oxygen
supply, alteration of the usage of fatty acids as energy substrate for the production of high energy
phosphate results in a increased backdiffusion of BMIPP, decreased tissular concentration of
BMIPP and alpha-oxidized metabolites, and hence mismatching with the flow tracers.
In patients, this pattern of uptake with BMIPP less than perfusion has been found a good predictor
of myocardial viability, while a matched decreased uptake of both tracers reliably identified
myocardial scar (4-7). Reverse mismatching (BMIPP higher than perfusion) is less frequently
reported and its significance unclear. It could be associated with unstable conditions (20).
0
10
20
30
40
50Bg Scat 1.0 Scat 1.3No
Extent (%)
Method
m=24.8
m=5.9
m=14.1m=17.9
63
Table 4: Echocardiographic and scintigraphic findings in the 14 arterial territories with resting wall motion
abnormalities.
LAD: left anterior descending artery. LCX: left circumflex artery. RCA: right coronary artery.
H: hypokinesis. A: akinesis. D: dyskinesis. Dobutamine stress test: (+) = contractile response under
dobutamine; (-) = unchanged wall motion. No: no correction. Bg: background substraction.
Scat: scatter correction. MIBI/BMIPP: (+) = mismatching; (-) = matched decreased uptake.
CABG: coronary artery bypass grafting. PTCA: percutaneous transluminal coronary angioplasty.
Involved
artery
Baseline
wall
motion
Dobuta
mine
stress
MIBI/BMIPP Treatment Follow-up
wall motion
Scat No Bg
Patient 1 LAD H (+) (+) (-) (+) PTCA Normalized
Patient 2 LCX A (-) (+) (-) (+) CABG Improved
Patient 3 LAD A (-) (-) (-) (+) Medical Unchanged
Patient 4 LAD A (+) (+) (+) (+) PTCA Improved
Patient 5 RCA
LCX
A
A
(+)
(+)
(+)
(+)
(-)
(+)
(+)
(+)
CABG Improved
Improved
Patient 6 LAD severe H (+) (-) (-) (+) Medical Improved
Patient 7 LAD D (-) (-) (-) (+) Medical Unchanged
Patient 8 LAD
RCA
LCX
severe H
severe H
mild H
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
(+)
CABG Improved
Improved
Improved
Patient 9 LAD A (+) (+) (-) (+) CABG Improved
Patient 10 RCA
LCX
D
A
(-)
(-)
(-)
(-)
(-)
(-)
(-)
(+)
Medical Unchanged
Unchanged
Influence of the methodology on the pattern of distribution of BMIPP related to perfusion:
mismatching versus reverse mismatching.
Using low-energy collimators for cardiac imaging with 123
I-radioiodinated free fatty acids, the
magnitude of septal penetration varies from one study to the other depending on the surrounding
activity, even when high-purity 123
I is used. This must be kept in mind especially when the uptake
of BMIPP must be quantitatively compared to that of a perfusion tracer. Indeed, the heterogeneous
distribution of the resultant background might mask the presence of mismatching or even result in a
"reverse mismatching" with BMIPP uptake higher than the perfusion.
This feature, recently reported by Sloof et al in a significant number of patients with chronic
ischemic heart disease, incited these authors to suggest that the mismatching with BMIPP less than
perfusion might be more typical for the (sub)acute phase of a myocardial infarction while a reverse
mismatching, mirroring the findings observed with 18F-fluorodeoxyglucose, should be more
frequently associated with viability in the chronic phase of the disease (8). However, in their paper,
the authors observed a spread of BMIPP uptake from 20% higher to 10% lower than thallium.
Furthermore, they considered a 7% to 8% difference in uptake as threshold of significancy, in order
to take into account the differences in attenuation between the 2 isotopes as recommended by
Tamaki (21). Since we found 12% more background activity in the left ventricle cavity with BMIPP
due to septal penetration of high energy photons, it is likely that this might at least partially explain
the higher BMIPP than thallium uptake they noted.
Indeed, although reverse mismatching has been reported in a small number of patients, it
seems related to unstable conditions (20) rather than to the delay between the acute event and the
64
tests since Tamaki found no reverse mismatching but a decreased BMIPP uptake relative to 201
Tl in
29/50 consecutive patients with chronic infarction (22).
Influence of the methodology on the presence and extent of viable myocardium
In our patients, the mean tissue activity within the infarcted area for the 10 patients was very similar
for both tracers when no correction for the higher noise level in the BMIPP images was applied
(mean BMIPP uptake 0.4% lower than MIBI). After scatter correction, the difference increased to
6.4%.
Our results demonstrate that methodological factors can influence the image interpretation
when comparing data obtained by using isotopes with different physical characteristics. In
myocardial viability assessment, comparing 123
I-BMIPP and 99m
Tc-sestamibi without correcting for
the differences in signal-to-noise ratio, scatter and attenuation results in an underestimation of the
magnitude of mismatching, partially disguised by the higher background in the iodine images,
especially in regions with low uptake.
Conversely, measuring the background activity by means of a single region of interest located at the
basal part of the myocardium to apply a background subtraction overcorrects the BMIPP images,
particularly in patients with extended apical defects, hence worsening the specificity. Indeed,
because of the heterogeneous distribution of background activity within the ventricle, the activity in
low count regions is influenced by the surrounding tissue activity when applying filtered
backprojection (23), and one single ROI placed on the base of the ventricle is not representative for
the real distribution of background activity within the whole ventricular cavity, as shown by the
negative values of BMIPP uptake found in two of our patients.
Scatter correction is thus the method of choice when the uptake of BMIPP must be quantitatively
compared to a perfusion tracer. In our small group of patients, it demonstrated the same accuracy as
dobutamine stress echocardiography.
Regarding the perfusion tracer of choice, the value of 99m
Tc-labeled compounds for
assessing myocardial viability remains controversial compared to 201
Tl (24). However, we should
recommend 99m
Tc-labeled agents for quantitative comparison with 123
I-BMIPP, because of the small
difference in attenuation between 123
I and 99m
Tc.
Using the dual window method, the value of the substraction factor k is difficult to
determine from phantom studies because it depends on the distribution of activity, and different
approaches result in slightly different values (13). Based upon phantom studies showing that scatter
substraction factors of 1.0 and 1.3 were the lowest and highest possible values for 123
I, we applied
these two values to 123
I-BMIPP imaging in patients. Using a scatter factor k=1.0 for BMIPP, the
mean left ventricular background activity remained higher than in the MIBI images corrected for
scatter with a factor k=0.7, indicating a slight undercorrection. However, when quantifying the
mismatching between MIBI and BMIPP with k=1.0 and 1.3 for 123
I, the difference in the %
mismatched surface was small and had no influence on the classification of a region as either viable
or not. Triple window scatter correction avoids the scatter factor determination (25).
In the present study, BMIPP and MIBI were acquired sequentially with an interval of 3
days. Acquiring two different energies simultanously would likely cause errors in this quantitative
analysis because of cross-talk from one energy window into the other (6). Using small asymmetric
energy windows and correcting for cross-talk, Madsen et al (26) concluded in a phantom study that
the true difference in regional tracer uptake should exceed 10 % to get reliable results. Since a
difference of ±10% uptake between both tracers is defined as mismatching, hence viable tissue,
simultanous acquisitions are not appropriate.
Low-energy all-purpose instead of high-resolution collimators were used in this work, to
improve the signal-to-noise ratio after scatter correction. Further optimization of the method could
be achieved by using medium-energy collimators for 123
I-BMIPP and low-energy collimators for 99m
Tc-MIBI with the same resolution.
65
Conclusion
When quantifying the mismatching between 123
I-BMIPP and 99m
Tc-MIBI in myocardial viability
assessment, special attention must be paid to the acquisition parameters. Indeed, direct comparison
between the two tracers without any correction results in an underestimation of the presence of
mismatching while background substraction frequently overestimates the amount of viable tissue.
Compared to dobutamine stress echocardiography in a small group of patients, quantification of
scatter corrected images was the most optimal method for identifying the presence of viable
myocardium.
References
1. Altehoefer C, vom Dahl J, Biedermann M et al. Significance of defect severity in technetium-
99m-MIBI SPECT at rest to assess myocardial viability: comparison with fluorine-18-FDG PET. J
Nucl Med 1994; 35: 569-574.
2. Udelson JE, Coleman PS, Metherall J et al. Predicting recovery of severe regional ventricular
dysfunction: comparison between resting scintigraphy with 201
Tl and 99m
Tc-sestamibi.Circulation
1994; 89: 2552-2561
3. Maes AF, Borgers M, Flameng W, et al. Assessment of myocardial viability in chronic coronary
artery disease using technetium-99m sestamibi SPECT. J Am Coll Cardiol 1997; 29: 62-68
4. Tamaki N, Tadamura E, Kawamoto M, et al. Decreased uptake of iodonated branched fatty acid
analog indicates metabolic alterations in ischemic myocardium. J Nucl Med 1995; 36: 1974-1980
5. Franken PR, De Geeter F, Dendale P, Demoor D, Block P, Bossuyt A. Abnormal free fatty acid
uptake in subacute myocardial infarction after coronary thrombolysis: correlation with wall motion
and inotropic reserve. J Nucl Med 1994; 35: 1758-1765.
6. Ito T, Tanouchi J, Kato J, et al. Recovery of impaired left ventricular function in patients with
acute myocardial infarction is predicted by the discordance in defect size on 123
I-BMIPP and 201
Tl-
SPET images. Eur J Nucl Med 1996; 23: 917-923.
7. Franken PR, Dendale P, De Geeter F, Demoor D, Bossuyt A, Block P. Prediction of functional
outcome after myocardial infarction using BMIPP and sestamibi scintigraphy. J Nucl Med 1996;
37: 718-722.
8. Sloof G, Visser F, Bax J, et al. Increased uptake of Iodine-123-BMIPP in chronic ischemic heart
disease: comparison with Fluorine-18-FDG SPECT. J Nucl Med 1998 ; 39: 255-260.
9. De Geeter F, Franken PR, Defrise M, Andries H, Saelens E, Bossuyt A. Optimal collimator
choice for sequential iodine-123 and technetium-99m imaging. Eur J Nucl Med 1996; 23: 768-774.
10. Gilland DR, Jaszczak RJ, Turkington TG, Greer KL, Coleman RE. Volume and activity
quantification with iodine-123 SPECT. J Nucl Med 1994; 35: 1707-1713.
11. Hambye AS, Dobbeleir A, Franken PR. SPET generated colour-coded polar maps to quantify
the uptake of 99mTc-sestaMIBI and 123I-BMIPP in chronically dysfunctional myocardium:
comparison with coronary anatomy and wall motion. N Med Commun 1997; 18: 1135-1147.
12. Mertens J, Eersels J, Van Ryckeghem W. New high yield Cu(I) assisted I-123 radioiodination of
15(p-I-phenyl)-9-methyl pentadecanoic acid, a potential myocardial tracer. Eur J Nucl Med 1987;
13: 159-160.
13. Jaszczak RJ, Greer KL, Floyd CE, Harris CC, Coleman RE. Improved SPECT quantification
using compensation for scatter photons. J Nucl Med 1984; 25:403-408.
14. Buvat I, Benali H, Todd-Pokropek A, Di Paola R. Scatter correction in scintigraphy: the state of
the art. Eur J Nucl Med 1994; 21: 675-694.
15. Gilardi MC, Bettinardi V, Todd-Pokropek A, Milanesi L, Fazio F. Assessment and comparison
of three scatter correction techniques in single photon emission computed tomography. J Nucl Med
1988; 29: 1971-1979.
66
16. Pretorius PH, van Rensburg AJ, van Aswegen A, Lötter MG, Serfontein DE, Herbst CP. The
channel ratio method of scatter correction for radionuclide image quantitation. J Nucl Med 1993;
34: 330-335
17. Fujibayashi Y, Nohara R, Hosokawa R, et al. Metabolism and kinetics of iodine-123-BMIPP in
canine myocardium. J Nucl Med 1996; 37: 757-761
18. Fujibayashi Y, Yonekura Y, Takemura Y, et al. Myocardial accumulation of iodinated beta-
methyl branched fatty acid analogue, iodine-125-15-(p-iodophenyl)-3-(R,S) methyl pentadecanoic
acid (BMIPP) in relation to ATP concentration. J Nucl Med 1990; 31: 1818-1822
19. Yamamichi Y, Kusuoka H, Morishita K, et al. Metabolism of iodine-123-BMIPP in perfused
rat hearts. J Nucl Med 1995; 36: 1043-1050
20.. Saito T, Yasuda T, Gold HK et al. Differentation of regional perfusion and fatty acid uptake in
zones of myocardial injury. N Med Commun 1991; 12: 663-675
21. Tamaki N, Kawamoto M, Yonekura Y, et al. Regional metabolic abnormality in relation to
perfusion and wall motion in patients with myocardial infarction: assessment with emission
tomography using a iodinated branched fatty acid analog. J Nucl Med 1992; 33: 659-667
22. Tamaki N, Tadamura E, Kudoh T, et al. Prognostic value of iodine-123-labelled BMIPP fatty
acid analogue in patients with myocardial infarction. Eur J Nucl Med 1996; 23: 272-279
23. Nuyts J, Dupont P, Van den Maegdenbergh V, Vleugels S, Suetens P, Mortelmans L. A study of
the liver-heart artifact in emission tomography. J Nucl Med 1995; 36: 133-139.
24. Marcassa C, Galli M, Cuocolo A, Scappellato F, Maurea S, Salvatore M. Rest-redistribution
thallium-201 and rest Technetium-99m sestamibi SPECT in patients with stable coronary artery
disease and ventricular dysfunction. J Nucl Med 1997; 38: 419-424
25. Ichihara T, Ogawa K, Motomura N, Kubo A, Hashimoto S. Compton scatter compensation
using the triple-energy window method for single- and dual-isotope SPECT. J Nucl Med 1993; 34:
2216-2221.
26. Madsen MT, O’Leary DS, Andreasen NC, Kirchner PT. Dual isotope brain SPECT imaging for
monitoring cognitive activation: physical considerations. N Med Commun 1993; 14: 391-396.
67
3.4 Clinical applications.
BMIPP Imaging to Improve the Value of Sestamibi Scintigraphy for Predicting Functional Outcome in Severe Chronic Ischemic Left Ventricular Dysfunction. Anne-Sophie E. Hambye, André A. Dobbeleir, A. Vervaet, Paul A. Van den Heuvel and Philippe R. Franken. Nuclear Medicine and Cardiology, Middelheim Hospital, Antwerp, and Nuclear Medicine, Free University of Brussels (VUB), Brussels, Belgium.
Journal of Nuclear Medicine 1999; 40: 1468-1476.
Mismatching between BMIPP and perfusion accurately predicts functional outcome after acute myocardial infarction. The
current investigation aimed at evaluating the value of this method to predict the evolution of global function according to
the applied treatment in patients with chronic ischemic heart disease. Methods: Twenty patients with infarction and
chronic left ventricular dysfunction were studied (median infarction age: 12 weeks, range: 2 weeks-15 years).
Radionuclide angiography, 2D-echocardiography and BMIPP and gated sestamibi scintigraphy were obtained at rest
before and >6 months after treatment (revascularization in 13 and conservative therapy in 7). In 7 patients, radionuclide
angiography was repeated after 1 year. Results: On a patient basis, mismatching with BMIPP less than sestamibi was
noted in 15 patients at baseline. Eleven of these 15 patients demonstrated significant functional improvement at follow-
up, versus only one of the 5 with a matched decreased uptake. Hence, the combined sestamibi/BMIPP was 73% positive
and 80% negative predictive of functional outcome, with a global accuracy of 75%. On a segmental basis, using an
optimal threshold of uptake defined by ROC curve analysis, sestamibi was only 63% accurate to predict regional
outcome. Adding BMIPP improved the accuracy to 80% (p=0.001).
At follow-up, significant mismatching was still noted in 7 patients in the revascularized group, and 1 in the medically
treated. It was associated with a further increase in ejection fraction at one year follow-up only in the revascularized
group. Conclusion: In patients with chronic left ventricular dysfunction post-infarction, a mismatching with BMIPP less
than sestamibi reliably identifies jeopardized but viable myocardium, and predicts functional recovery with an accuracy
similar to that reported in the (sub)acute phase of the infarction.
68
Mid-ventricular short axis (SA), vertical long axis (VLA), and bull’s eye (BE) images of a 45-year old male patient with a 37 week-old anteroseptal infarction associated with extended akinesis (9 segments). Baseline EF amounted to 37% and 8/9 segments were mismatched. Six months after CABG, EF did not significantly change (39%) although the patient was clinically symptom free. On scintigraphy, 6/9 segments with baseline dysfunction were still mismatched despite an increased sestamibi and BMIPP uptake. At one year, EF amounted to 44%.
69
Quantification of Sestamibi and BMIPP Uptake for Predicting Functional Outcome in Chronically Ischemic Dysfunctional Myocardium. A.S. HAMBYE, A. VERVAET, A. DOBBELEIR Nuclear Medicine, Middelheim Hospital, Antwerp, Belgium.
Nuclear Medicine Communications, 1999; 20: 737-745. Summary
In chronic ischemic heart disease, little is known about the usefulness of free fatty acid scintigraphy for assessing
viability. To investigate this, we quantified the uptake of 99Tc
m-sestamibi and
123I- BMIPP at rest twice with 6 months
interval in 20 patients with chronic ischemic left ventricular dysfunction and infarction. Depending on the relative
distribution of both tracers, 4 patterns of uptake were observed and named normal, mismatched, matched and scar. The
proportion of the left ventricle surface corresponding to each pattern was expressed in % of the total surface using a
polar map. In the meantime between the 2 series of tests, patients were either addressed for revascularization or
conservatively treated.
The quantitative results were compared to those of dobutamine stress echocardiography (DSE) in arterial territories with
resting contractile dysfunction and correlated to the evolution of regional and global function at follow-up.
At baseline, 25 arterial territories were analyzed. Using sestamibi, on average 1/3 of their surface was considered as
normally perfused. No clear association was found between the % normally perfused surface and the DSE findings.
Adding BMIPP and using a value of >7% of the arterial surface with BMIPP lower than sestamibi (mismatch) as cutoff of
significance for viability, 14/18 mismatched regions were considered viable by DSE, and 6/7 with <7% mismatched
surface or matching were not.
On a patient basis, 15 patients were viable, of whom 13 were revascularized (16 territories). At follow-up, global function
improved in 11 of the 15 viable patients, all in the revascularized group. Regional improvement was noted in 11/16
revascularized territories, and was associated with a significant increase in sestamibi and BMIPP uptake and in the %
normally perfused myocardial surface. In the 5 patients without significant viability, no functional deterioration or changes
in the quantitative parameters were observed under medical treatment.
These results suggest that quantitative analysis of sestamibi and BMIPP uptake is a reliable method to objectivate the
presence of myocardial viability in chronic ischemic heart disease and to predict functional improvement after
revascularization.
Evolution with time of the percent uptake of sestamibi and BMIPP in the regions with wall motion abnormalities by echocardiography relative to the treatment used.
Sestamibi uptake (%) BMIPP uptake (%)
Baseline
Follow-up
Mean
difference
P-value
Baseline
Follow-up
Mean
difference
P-value
Revascularized territories (n=16) 47.2±11.4 53.0±12.0 5.8±5.7 0.004 38.5±13.2 47.8±14.7 9.3±6.9 <0.001
Conservatively treated territories (n=9) 41.0±10.9 40.6±13.4 -0.4±8.8 0.86 40.1±9.6 39.2±12.7 -0.8±6.8 0.77
*Values are expressed as the mean ± standard deviation.
70
Predicting Functional Outcome by quantification of sestamibi and BMIPP after acute myocardial infarction. Anne-Sophie E. Hambye, Ann Vervaet, André A. Dobbeleir, Paul Dendale and Philippe R. Franken. Nuclear Medicine, Middelheim Hospital, Antwerp, and Nuclear Medicine and Cardiology, Free University of Brussels (VUB), Brussels, Belgium.
European Journal of Nuclear Medicine 2000; 27: 1494-1500.
Iodine-123 15-(p-iodophenyl)-3-R,S-methyl-pentadecanoic acid (BMIPP) can be used to image myocardial fatty acid
regional distribution and utilization with single-photon emission tomography (SPET). By visual analysis, a mismatching
with regional uptake of BMIPP less than that of a perfusion tracer has been show to predict myocardial viability and
functional improvement after restoration of flow in patients with myocardial infarction. The current study aimed to
evaluate a newly developed quantitative method of analysis of sestamibi and BMIPP uptake for the prediction of
functional recovery after revascularization in patients with acute infarction. BMIPP and gated sestamibi SPECT studies at
rest were obtained before and >3 months after revascularization in 18 patients with recent infarction. A color coded polar
map was generated from the comparison of sestamibi and BMIPP uptake. Depending on the relative distribution of the
two tracers, different patterns of uptake were identified and their extent expressed as percentages of the surface of the
whole left ventricle and of the three main coronary artery territories. At follow-up, recovery was defined as ≥5% increase
in ejection fraction compared to baseline. Receiver-operating characteristic curve analysis was performed to analyse the
data. At baseline, significant correlations were found between ejection fraction and the % surface with decreased
sestamibi or BMIPP uptake (r=-0.68, p=0.001, and r=-0.72, p<0.0001 respectively). When combining both tracers,
ejection fraction was significantly associated with extent of myocardium showing decreased sestamibi uptake with lower
BMIPP uptake (mismatching; r=-0.68, p=0.001). At follow-up, significant functional recovery was found in 13/18 patients.
By ROC curve analysis, the optimal pattern of distribution predicting recovery was a mismatching with uptake of
sestamibi <70% and uptake of BMIPP at least 10% lower. For this parameter, optimal cut-off of extent was 10% of the
whole left ventricle surface (sensitivity 69%, specificity 80%, accuracy 72%) and 25% of the infarct related arterial
territory (sensitivity 77%, specificity 80%, accuracy 78%). The areas under the curve were 79% for the left ventricle
surface and 72% for the individual arterial territories. These results suggest that in patients with acute infarction,
quantitative analysis of sestamibi and BMIPP could offer an objective and reproducible method for estimating the severity
of cardiac dysfunction and predicting the evolution of ejection fraction after revascularization.
Optimal
cut-off
(% surface)
Sens./
Spec. pair
(%)
AUC
(SEE)
(%)
Normal sestamibi uptake
80
54/40
32 (14)
Normal BMIPP uptake 55 69/40 52 (14)
NPDFA 10 46/40 32 (16)
DPDFA-mismatch 10 69/80 79 (12)
NPDFA+DPDFA-mismatch 25 62/60 46 (19)
DPDFA-match
5 62/60 65 (16)
Optimal % left ventricle surface predicting ≥5%
increase in ejection fraction at follow-up for each
distribution pattern of sestamibi and BMIPP, and their respective sensitivity/specificity pairs and
area under the curve (AUC) with the corresponding
standard error of the estimate (SEE) calculated by ROC curve analysis. The best parameter is
indicated in italics.
NPDFA: normal perfusion, decreased fatty acid uptake
DPDFA: decreased perfusion, decreased fatty acid uptake
71
4. Perfusion of the brain.
4.1 Quantification in SPECT using non-invasive methods.
A. Dobbeleir, R.A. Dierckx, J. Vandevivere, P.P. De Deyn.
Departments of Nuclear Medicine and Neurology, Middelheim Hospital Antwerp and Nuclear
Medicine, University Hospital Ghent, Belgium.
In SPECT in Neurology and Psychiatry. Editors De Deyn, Dierckx, Alavi and Pickut. 1997 John
Libbey & Company Ltd.
Summary
Development of dedicated tomographic equipment, in the early eighties, led to the regional
quantification of cerebral blood flow. Using the inert radioactieve gas Xenon-133, Lassen and co-
workers measured the cortical blood flow in ml/min. After the introduction of I-123 IMP and Tc-99m
HMPAO, during the late eighties, kinetic models were developed in several centres using acquired data
from conventional nuclear medicine equipment. However, due to the necessity of arterial sampling for
quantitation, visual interpretation of the tomographic images is still the method of choice in clinical
practise.
Using Tc-labeled tracers and high resolution tomographic gamma cameras, tracer uptake of the main
brain structures can be analyzed. Regions with reduced tracer uptake are visually detected by
comparing to the cerebellar or contralateral uptake. Semiquantitative analysis is performed by
comparing activity ratios of different regions. Many different methods for defining the regions of
interest have been applied, like fixed size regions, functional cortical regions and circumferential
profiles. 3D translation and rotation techniques are used to combine neuroanatomic and functional
images for delineation of the brain structures. Visual interpretation and semiquantification are widely
used.
An hypothetical volume calculation method was presented by Mountz. Assuming that the contralateral
brain region remains uninvolved, this solid method takes into account the severity of hypoperfusion
and extent of the lesion. We used this method in order to express acute ischaemic infarction as
millilitre of zero perfusion.
A non-invasive method for quantitative assessment of brain perfusion was presented by Matsuda. Our
group presented a simple method, using calibrated point sources as a scaling factor, whereby the
tomographic slices are displayed as regional Tc-99m HMPAO brain uptake per cm3 brain tissue in
proportion of the injected lipophilic dose. This method was used to test the influence of several
parameters such as heart rate, weight, height, brain volume, age and drug activation on the HMPAO
brain uptake. Hence, evaluation of global alterations in neurological disorders may be considered.
Introduction
Many neurological disorders are associated with a decrease of perfusion. Since several decenia
attempts at quantification of brain function have been made. In 1945, Kety and Schmidt 1 measured
absolute cerebral blood flow, using inhalation of nitrous oxide in low concentrations. As shown by the
Fick 2 equation, arterial and venous blood samples are needed.
QBRAIN(T) = FLOW * (AC - VC) dt , in which
QBRAIN = amount of substance delivered to the brain, AC = arterial concentration and VC = venous
concentration
72
Several models, especially for PET studies, have been derived from this equation. For absolute
quantification, several presumptions are made in the kinetic blood flow models. In SPECT, absolute
quantification is mainly used to validate semi-quantification methods for specific patient groups.
Cerebral blood flow
Due to the introduction of I123-amphetamine and certainly Tc99m-HMPAO at the end of the eighties
and the use of high resolution tomographic scanners the interest in the regional distribution of the brain
perfusion increased tremendously. The first pass extraction of HMPAO was explained by Nickel 3 in a
simple circulation model.
ABRAIN = Ainject * (1 - Rlung) * (CBF/CO) * Rbrain , in which
ABRAIN = activity trapped in the brain, Ainject = total injected activity, Rlung = retention factor of the lung,
CBF = cerebral blood flow, CO = cardiac output and RBrain = retention factor of the brain
After intravenous injection of a known activity of Tc99m-HMPAO the tracer flows through the right
heart, the lung and the left heart chambers. Within the lung a small part of the activity is trapped due to
diffusion to the lung tissue. This fraction is indicated as the retention factor of the lung. Afterwards, in
the aorta the blood flow splits into the brain circulation and the systemic circulation. Therefore, only
the fraction of activity given by the ratio of cerebral blood flow on cardiac output reaches the brain.
The amount of activity, the brain uptake, retained in the brain depends on the extraction and retention
fraction of the brain.
Again to quantitate the retention fraction in the brain arterial and venous blood samples are needed. A
technique avoiding these blood samples is necessary for clinical applications. Generally for practical
reasons it is also considered that the retention fraction is constant over the brain and that the HMPAO
SPECT image represents mainly a distribution of blood flow.
Absolute quantification
The first non-invasive absolute quantification method was
proposed by Lassen, Kanno, Celsis and co-workers 4,5.
Their method is based on the inhalation and expiration of the inert gas Xe133. Four 1 minute
sequential tomographic images of the cross-section of the brain were obtained using a dedicated
tomograph. The shape of the arterial input function is recorded using a separate detector placed over
the right lung. This was the only non-invasive method until recently. The technical disadvantages
are the need for a dedicated expensive fast rotating single photon tomograph and the poor resolution
of the low energetic Xenon isotope.
Besides arterial sampling methods, compartimental models have been developed by predicting the
blood flow from a linear relationship with a calculated index like the absolute quantification method of
Matsuda 6. This method, most applied worldwide at this moment, uses the brain input and aortic arch
curves to calculate a brain perfusion index which is correlated to blood flow. Detailed explanation is
given in another chapter.
Visual and semi-quantitative analysis
Quantification remains difficult in routine practice. Therefore most physicians evaluate the distribution
of blood flow in the brain by visual interpretation. In an attempt to be more sensitive, reducing the
variation coefficient by refering to a stable region, many semi-quantitative methods using
Tc99m-HMPAO have been described over the past few years 7. Operator defined or automatic left to
right ratios have been used. The activity in functional regions has been studied. In order to delineate
73
more precisely the anatomic structures, SPECT images have been matched to CT or NMR images. The
volume of the defect has been calculated comparing to the contralateral healthy structures.
Figure 1. Hofmann brain phantom. First and third row represent planar images of slices through the phantom.
Second and fourth row are tomographic reconstructions.
Significant improvement in image resolution has been achieved by using multidetector systems and
high resolution fanbeam collimators 8. When comparing images made with a single detector to a multi-
detector system, physicians had to adapt their visual interpretation due to a dramatic improvement of
reconstructed resolution, from about 15 mm to about 8 mm providing functional detailes never
observed before. The images became close to the ones observed with PET systems.
The group of Matsuda at the Kanazawa University in Japan presented the first high resolution
functional brain atlas 9. Major topographical reference points could be defined and a comparison was
made with the MRI data 10.
Meanwhile progres was also made in computerspeed permitting reconstruction or reprojection methods
which were too time consuming a few years ago 11. Three-dimensional display utilizing volume
rendering facilitates understanding of spatial relationships and anatomo-functional correlation 12.
In smaller structures, a relative hypoperfusion is observed. This is not necessary a physiological
phenomenon, but is associated to the tomographical resolution of the gamma camera and is well
74
known as partial volume effect. For this reason, visually or quantitatively structures of the left
hemisphere are compared to the same structures in the right hemisphere.
In figure 1 a Hofmann brain phantom is presented. The first and third row represents planar images of
slices through the phantom. The second and fourth row are tomograpic reconstructions of the phantom.
Due to a worse resolution, the images of the tomographic reconstruction are less sharp and a drop of
activity is observed in smaller structures 13.
In figure 2 a circumferential profile of the phantom in a basal ganglia slice is presented. Although the
same radioactivity is present everywhere, different activities of the cortex are observed. Frontal cortical
activity is higher then in the other cortical regio's. This corresponds to the regio in the phantom where
the cortex has the largest diameter. This means also that normal distribution of the tracer in the brain
depends of the resolution of the acquisition system. A mental picture of normality and pictures of
different disease states are therefore mandatory. Gender differences in brain uptake were described by
Podreka in a large group of normal subjects 14.
Figure 2. A circumferential maximum value profile of a Hofmann phantom in a basal ganglia slice
normalized to the mean activity. 90° corresponds to frontal activity.
That's why some centers try to establish a normal database of regional cerebral uptake 15. Sectorial
regions of interest were drawn on brain slices of normal volunteers in order to obtain strict reference
values to be used in interindividual and intergroup clinical comparisons. The activity in each region is
generally compared the the contralateral activity and to the cerebellar activity. A mean and SD value
can be obtained for each region. Most of the cortical normal values are situated at a level of about 80 to
85 % of the cerebellar activity. The spread was in general around 5 % exept in the fronto-orbital
region where it was about 10 %. In all similar studies, for example in Alzheimer patients, the
assumption has to be made on clinical grounds that the cerebellar activity in patients is normal since
the cerebellum is needed as reference region. In the same way the mean activity in the whole slice or
even in the whole brain is used as reference. Automatic transformation of 3D images into a stereotaxic
Talairach atlas for establishing a normal database and patient comparison has been achieved 16.
Effective volume determination
Defect volumes can be calculated. ROIs have to be drawn over the lesion on all slices involved. These
ROIs are mirrored to the contralateral homologous region. The total functional volume loss, expressed
as an imaginary volume of zero perfusion can then be calculated. This method, first used by Mountz 17,
75
takes into account the pixel volumes, number of slices and the activity difference at each slice between
the normal and in our case ischaemic hemisphere.
We used this method in patients suffering from unilateral acute ischemic infarction 18. By expressing
the infarction as millilitre zeroperfusion, this semiquantification method simultaneously takes into
account the extent and degree of hypoperfusion. SPECT defects may be measured in an accurate and
reproducible way. Moreover this study showed a significant correlation between the volume
zeroperfusion and the neurologic deficit on admittance as scored on the Orgogozo scale.
Brain uptake quantification
We have developed an absolute quantification method 19, not to quantify blood flow, but to quantify
exactly what is seen on the images: the regional brain uptake of Tc99m-HMPAO being the product of
blood flow and retention of the brain. We studied the influence of individual body surface, brain
volume and heart rate fluctuation on the brain uptake 20. The influence of these factors were
investigated in 33 healthy volunteers. Afterwards, they were applied to 13 patients suffering from
probable dementia of Alsheimer's type, in order to correct for interindividual variation 21. An equation
has been tested to calculate from known point source activities a factor K. Multiplying SPECT slices
on pixel by pixel base by this factor provides image values expressed as 10-6 of the injected lipophilic
dose per ml brain tissue.
This allows intra or interindividual comparison between different Tc99m HMPAO studies. In healthy
volunteers with similar heart rate at injection in the two studies, a mean deviation of 7.2% in cerebellar
uptake was observed between repeated acquisitions 19. On the contrary testing the reproducibility of the
method on healthy volunteers, we noted a significant influence of the heart rate at injection time. This
could be expected since the brain uptake is inversely proportional to the cardiac output.
Table 1 shows the cerebellar uptake of the total group of 33 healthy volunteers, 18 men, 15 women and
13 Alzheimer patients. The mean uptake at basal heart rate indicated in the first column varies from
43.1 for men to 75.2 for the Alzheimer patients. The second column shows the brain uptake corrected
for the body surface. The value 49.9 for men becomes closer to the uptake value of 66.2 for the
Alzheimer patients, who had a much smaller body weight. The third column shows the results after the
correction for body surface and brain volume index. In each of these groups, the cerebellar uptake
becomes very close to about 52 10-6 of the injected dose per ml brain tissue, with a similar standard
deviation of about 7.
Table 1. Cerebellar uptake (mean and standard deviation) in healthy volunteers and DAT patients at rest
heart rate, corrected for a body surface factor of BS/1.73 and corrected for body surface and a brain volume
factor of BV/1350.
76
Figure 3. Regional brain uptake rBU in healthy volunteers and DAT patients, after cumulative corrections for
heart rate, body surface and brain volume, plotted versus age. The decline of cerebellar uptake with age
becomes negligible rBU = 54.7 – 0.04 x age.
In figure 3 the cerebellar uptake of 33 healthy volunteers and 13 Alzheimer patients, after cumulative
corrections for heart rate, body surface and brain volume are plotted versus age. As seen on this graph
the decline of the cerebellar uptake with age becomes negligible.
Normalized brain uptake values allow extension of the application of the quantification method from
longitudinal studies to group comparisons. This study proved that the cerebellum is a practical and
reliable region to choose as a reference in patients with Alzheimer's dementia.
Discussion
The aim of quantification is to provide a reliable and reproducible numerical measure of brain
perfusion in different regions and time windows. Since absolute quantification is difficult and time
consuming in daily practise, absolute quantification is certainly important for identifying stable
regions at different disease state, regions which afterwards can be used in semi-quantitation as
control or reference regions. In those cases semi-quantitation is advisable since possible
measurement errors are smaller. However in cases, where a overall change of blood flow can be
expected like carotic obstruction or in stimulation studies, the more difficult and susceptive to errors
absolute quantification method is mandatory.
References
1. Kety SS, Schmidt CF.(1945): The determination of cerebral blood flow in man by use of nitrous
oxide in low concentrations. Am J Physiol. 143,53-66.
2. Fick A. (1870): Uber die Messung des Blutquantums in den Herzventrikeln. Verhandl Phys Med
Ges Wurzburg. (Translated Hoff HE, Scott HJ, N Engl J Med 1948. 239, 476-483).
3. Nickel O, Nagele-Wohrle B, Ulrich P, Eisner D, Roesler A, Grimm W and Hahn K.(1989):
RCBF-quantification with 99mTc
HMPAO-SPECT: Theory and first results. Eur J Nucl Med. 15,1-8.
77
4. Kanno I, Lassen NA.(1979): Two methods for calculating regional cerebral blood flow from
emission computed tomography of inert gas concentrations. J Comput Assist Tomogr. 3, 1, 71-76.
5. Celsis P, Goldman T, Henriksen L and Lassen NA.(1981): A method for calculating regional
cerebral blood flow from emission computed tomography of inert gas concentrations. J Comput Assist
Tomogr. 5, 5, 641-645.
6. Matsuda H, Tsuji S, Shuke N, Sumiya H, Tonami N, Hisada K.(1992): A quantitative approach to
technetium-99m hexamethylpropylene amine oxime. Eur J Nucl Med. 19, 195-200.
7. Mountz J.(1991): Quantification of the SPECT Brain Scan. Nuclear Medicine Annual, Raven Press,
New York. 67-98.
8. Dobbeleir A, Dierckx R, Vandevivere J. (1991): High spatial resolution SPECT using a three-head
rotating gamma camera and super fine lead fanbeam collimators. Eur J Nucl Med.18, 8, 600.(abstract)
9. Matsuda H, Oskoie SD, Kinuya K, Tsuji S, Sumiya H, Tonami N, Hisada K. (1990): Tc-99m
HMPAO brain perfusion tomography atlas using high resolution SPECT system. Clin Nucl Med. 15,
428.
10. Dierckx R, Dobbeleir A, Martin JJ, De Deyn PP. (1993): Tc-99m HMPAO tomography using a
three-headed SPECT system equipped with lead fan-beam collimators. Clin Nucl Med. 18, 532-
534.
11. Wallis J, Miller T.(1990): Volume rendering in three-dimensional display of SPECT images. J
Nucl Med. 31, 1421-1430.
12. Dierckx R, Dobbeleir A, Borggreve F, De Deyn PP, Vandevivere J. (1991): Three-dimension
reconstruction of brain perfusion. Eur J Nucl Med. 18, 597.(abstract)
13. Kim H, Zeeberg B, Fahey F, Bice A, Hoffman E and Reba R.(1991): Three-dimensional SPECT
simulations of a complex three-dimensional mathematical brain model and measurements of the
three-dimensional physical brain phantom. J Nucl Med. 32, 1923-1930.
14. Podreka I, Goldenberg G, Baumgartner C, Lang W, Steiner M, Schmidbauer M, Suess E, Bruecke
T, Asenbaum S, Deecke L.(1989): HMPA0 brain uptake in young normal subjects: gender differences
and hemispheric asymmetries. J Cereb Blood Flow Metab. 9 Suppl 1, 202. (abstract)
15. De Sadeleer C, de Metz K, Somers G, Bossuyt A.(1994): Relative quantification of Tc99m
HMPAO SPECT based on sectorial regions of interest in normal volunteers. Eur J Nucl Med. 21, 8,
781.(abstract)
16. Vanhove C, Monte C, Colaert H, Demonceau G.(1994): Standardised atlas for brain studies. Eur J
Nucl Med. 21, 8, 773.(abstract)
17. Mountz J.(1989): A method of analysis of SPECT blood flow image data for comparison with
computed tomography. Clin Nucl Med. 14, 192-196.
18. Dierckx R, Dobbeleir A, Pickut B, Timmermans L, Dierckx I, Vervaet A, Vandevivere J, Deberdt
W, De Deyn PP.(1995): Technetium-99m HMPAO SPET in acute supratentorial ischaemic infarction,
expressing deficits as millilitre of zero perfusion. Eur J Nucl Med. 22, 5, 427-433.
19. Dobbeleir A, Dierckx R.(1993): Quantification of Tc-99m HMPAO brain uptake in routine clinical
practice using calibrated point sources as an external standard: phantom and human studies. Eur J Nucl
Med. 20, 684-689.
20. Dierckx R, Dobbeleir A, Maes M, Pickut B, Vervaet A, De Deyn PP.(1994): Parameters
influencing SPET regional brain uptake of Tc-99m HMPAO measured by calibrated point sources as
an external standard. Eur J Nucl Med. 21, 514-520.
21. Dobbeleir A, De Deyn PP, Dierckx R.(1994): The cerebellum as reference region: comparison
of absolute cerebellar uptake of Tc-99m HMPAO in 33 healthy volunteers and 13 DAT patients.
Eur J Nucl Med. 21, 10, S5.(abstract)
78
4.2 Quantification of technetium-99m hexamethylpropylene amine oxime brain
uptake in routine clinical practice using calibrated point sources as an external
standard: phantom and human studies
A. Dobbeleir, R. Dierckx
Nuclear Medicine Department, Middelheim Hospital, Lindendreef 1, B-2020 Antwerp, Belgium
European Journal of Nuclear Medicine 1993; 20: 684-689.
Abstract. Quantitative methods for calculation of regional cerebral blood flow with technetium-
99m hexamethylpropylene amine oxime (99m
Tc-HMPAO) have been proposed. These methods are
very labour intensive and therefore are not useful in routine practice. We describe a simple
alternative method, using calibrated point sources as a scaling factor, whereby the tomographix
slices are displayed as regional 99m
Tc-HMPAO brain uptake per cm3 brain tissue in 10
-6 of the
injected lipophilic dose. The method was validated on Jaszczak and Hoffman phantoms using a
three-detector system with HR parallel and HR fan-beam collimators. Under the optimal conditions
described in this paper, the measured to real activity ratio was 1.00 (SD = 0.06). The reproducibility
of the cerebellar uptake in a group of ten normal volunteers and five patients was studied. Intra-
individually a mean deviation of 12.6% was observed for the total group. For those persons with a
heart rate difference of less than 5 units between the two studies, a mean deviation of 7.2% was
obtained. Quantitative 99m
Tc-HMPAO brain uptake images can be useful for longitudinal studies,
especially for follow-up, activation and pharmacological studies.
Key words: Technetium-99m hexamethylpropylene amine oxime – Quantification – Phantom
studies – Brain – Human studies
Introduction
Several authors [1-3] have shown that technetium-99m hexamethylpropylene amine oxime (99m
Tc-
HMPAO) distribution reflects regional cerebral blood flow (rCBF). In recent years, various
methods have been proposed for the quantification of rCBF on the basis of 99m
Tc-HMPAO uptake
[4-7]. All these methods are very elaborate and not applicable on a routine basis. A non-invasive
method for the quantitative assessment of brain perfusion using 99m
Tc-HMPAO was presented by
Matsuda et al. [8]. Although brain uptake is influenced by various parameters such as cardiac output
[5], retention fraction [9], and lipophilic and hydrophilic fraction [10], a simple quantification of the
regional brain uptake (rBU) of 99m
Tc-HMPAO may be useful for follow-up, pharmacological and
activation studies. A simple quantification method to convert brain counts into brain uptake is
presented, using calibrated point sources as an external standard. This method, which avoids the
need for regular phantom calibration of the single-photon emission tomography (SPET) system,
was tested on phantoms, and human reproducibility studies are also described.
Materials and methods
Reference sources.
Four point sources of 5 µCi (185 kBq) 99m
Tc were made by pipetting 5 µl of a 1mCi/ml (37
MBq/ml) solution, using a capillary precision pipette, onto scarcely absorbent paper and dried. The
point sources with a diameter of 3 mm and an activity precision of 2% were prepared in less than 10
min. The exact activity of the 1 ml solution was measured in an ionisation chamber. The four point
sources were fixed along the orbitomeatal axis of each patient before performing the 99m
Tc-
79
HMPAO SPET study. We used a three-detector system (Trionix, Triad) with a detector size of 40 x
22 cm and an energy window of 20%. For phantom studies 5-µl point sources of varying activities
were fixed on the Jaszczak of Hoffman phantom.
Phantom studies.
The spheres (internal diameter 33.5, 28, 22, 16, 13.5 and 10.5 mm) of the Jaszczak Deluxe phantom
were filled with a known concentration of about 3 µCi/ml (111 kBq/ml). Point sources of 1, 2, 4 and
6 µCi (37, 74, 148 and 222 kBq) were fixed on the Jaszczak phantom. The phantom was filled with
a low activity, about 0.1 µCi/ml (3.7 kBq/ml), in order to permit visualisation of the physical
boundary of the phantom. A total of 12 tomographic acquisitions, each of 120 angular views, were
performed on the three-detector system. Acquisition arrays were 128 x 128 and 64 x 64 for the
ultra-high-resolution parallel (UHRP) collimators and 256 x 128 and 128 x 64 for the super-fine
fan-beam (SFFB) collimators. When using the parallel collimators a zoom factor of 1.6 was applied
because of the rectangular shape of the detector. Reconstruction was performed in a 128 x 128 or 64
x 64 array depending on the acquisition array, resulting in a pixel size of 2.2 and 4.4 mm for the
UHRP collimators and 1.8 and 3.6 mm for the SFFB collimators. The acquisitions were repeated
for a total acquisition time of 10, 20, or 30 min. The projection data were reconstructed by filtered
backprojection using a Butterworth filter with a cutoff frequency of 0.7 cyc/cm and roll-off 5.
Chang’s first-order attenuation correction was applied using an attentuation coefficient of 0.12 cm-1.
Circular ROIs, 21-29 mm in diameter depending on the collimator and acquisition array, were
drawn over the point sources. The ROIs were applied on a number of slices with a total axial
distance equal to the ROI diameter. The adjacent slices were used for background substraction. The
diameter necessary to include the total point source activity had been determined previously (Table
1). Total point source activity was calculated applying the ROI on multiple slices and substracting
the ROI activity (background) of the first and last slices of the cylinder, corrected for the volume
ratio, from the total cylinder activity. Background substraction using adjacent slices was done to
eliminate blurred activity from the phantom or brain tissue into the point source region. The mean
activity in a 5 x 5 mm ROI in the centre of the spheres was measured and the activity/cm3
calculated, taking into account the pixel size and the measured and real activity of the point sources.
The same method was used for a three-dimensional Hoffman phantom [11] except that point
sources of 2, 4, 8 and 12 µCi (74, 148, 296 and 444 kBq) were used. The phantom was scanned at
least 6 h after filling and all activities are given at scan time. A 20-min acquisition (40 images of 30
s each, three detectors) was performed using the UHRP ans SFFB collimators. The activities in
cerebellum and frontal cortex (highest activity in the phantom) were measured.
Human studies: reproducibility of 99m
Tc-HMPAO brain uptake.
Ten normal volunteers (age range 24-52 years) and five patients (age range 54-72 years) were
studied. Two acquisitions were performed for each individual within a period of 1 week. The total
study period was 2 months. The protocol was approved by the local Ethics Committee and informed
consent was obtained. Using a previously fixed butterfly needle, 20 mCi (740 MBq) 99m
Tc-HMPAO
was administered to the patient sitting in a quiet and dimmed room, with eyes open and ears
unplugged. To ensure proper formulation of HMPAO, the manufacturer’s package insert was
followed, which implies one vial per patient. The injection was given less than 10 min after
preparation. In these conditions the injected lipophilic fraction can be estimated as 90% of the
injected dose [12]. Heart rate at the time of injection was noted. 5µCi (185 kBq) point sources were
fixed along the orbitomeatal axis and 120 images of 60 s were acquired using the three-detector
system and SFFB collimators starting 20 min after injection. After reorientating the slices parallel to
the orbitomeatal axis by means of the point sources, ROIs were drawn over the point sources and
the total point source counts were calculated as previously described. The mean of the four point
source counts was calculated. Knowing the pixel size (Z) in cm, the mean total point source counts
(CS), the activity of the point sources (AS), the injected liphophilic fraction (ID) and the delay
between the measurement in the ionisation chambre of the source activity and the injected dose (T),
80
and adding a multiplication factor to obtain whole numbers, a factor K is calculated: K=AS.e-
0.00193.T.10
6/CS.ID.Z
3. Multiplying SPET slices on a pixel-by-pixel basis by factor K gives image
values expressed in 10-6 of the injected dose per cm
3 brain tissue. The mean cerebellar uptake in a
11-mm-square ROI centered on the peak pixel value was compared in the two acquisitions. The
cerebellum was chosen since it is often used as a reference region in semi-quantitative analyses.
Results
Reference sources
Typical point source activities and the influence of ROI diameter in patient acquisitions are shown
in Table 1. The data show a greater influence of blurred activity of brain tissue in the ROI at the
meatus than at the canthus region. This difference in blurred activity is almost completely
eliminated after substraction of the background activity in the adjacent slices. For individual
sources a deviation of less than 5% of the mean value was observed in all patients. From Table 1 we
considered an ROI with a diameter of 21 mm large enough to include all the activity from the point
source when using the fan-beam collimator, a 128 x 128 reconstruction array and a Butterworth
filter. A similar approach showed the necessary ROI diameter to be 29 mm when using the parallel
collimator and a 64 x 64 reconstruction array.
Table 1. Influence of ROI diameter on measured point source activities (kcounts) for a fan-beam collimator
and 128 x 128 array patient study. Data are presented for the cylindrical volume, the adjacent slices and the
background substracted activity (bg substr.) ROI
diameter
(mm)
6 x 3.6 mm
slices
2 x 3.6 mm
slices
bg
subtr.
6 x 3.6 mm
slices
2 x 3.6 mm
slices
bg
subtr.
Canthus, right Canthus, left
14.6 146.7 2.7 138.5 145.0 2.1 139.0
16.7 165.4 2.9 156.5 160.0 2.6 152.3
18.5 177.1 3.5 166.9 170.2 3.0 161.0
20.6 184.8 4.0 172.8 176.2 3.5 165.6
22.4 190.0 5.3 174.4 183.3 3.7 172.3
24.5 192.6 5.3 176.3 184.8 4.9 170.0
Meatus, right Meatus, left
14.6 170.5 6.7 150.4 157.4 3.6 146.4
16.7 188.8 10.5 157.2 174.8 5.2 159.4
18.5 204.8 12.6 166.9 189.0 6.5 169.4
20.6 220.2 13.1 180.9 200.5 8.4 175.1
22.4 223.8 14.7 179.9 208.6 12.0 172.6
24.5 237.4 18.5 182.0 213.7 12.7 175.7
Phantom studies
In Table 2 the ratio of measured to real concentration in the centre of the largest sphere (chosen to
exclude any partial volume effect) is shown as a function of point source activity for the two
collimators, with 64 x 64 and 128 x 128 reconstruction arrays and 10-30 min acquisition time.
When the parallel collimator was used, the point source activity had only a minor influence except
in the 128 x 128 array at 10-min acquisition time for the lowest source activities. Excluding these
low pixel count acquisitions, the mean ratio of measured to real activity was 1.00 (SD=0.06). Using
the fan-beam collimator, the concentration in the spheres of the Jaszczak phatom was overestimated
at low pixel density due to underestimation of the point source activity. For the high pixel density
acquisitions a mean ratio of 1.00 (SD=0.04) was measured.
81
Table 2. Ratio of measured to real concentration, for different point source activities, in the centre of the
33.5-mm hollow sphere of the Jaszczak phantom filled with 3 µCi/ml (111 kBq/ml) 99m
Tc
Point source activity
37 kBq 74 kBq 148 kBq 222 kBq
Parallel 64x64 array
10 min acq. 1.04 1.10 1.00 1.02
20 min acq. 1.02 0.99 0.98 0.95
30 min acq. 0.95 0.98 0.98 0.95
Parallel 128 x 128 array
10 min acq. 1.19 1.15 0.99 1.07
20 min acq. 1.08 1.06 0.95 1.01
30 min acq. 1.02 1.08 0.97 1.05
Fan-beam 64 x 64 array
10 min acq. 1.08 1.12 1.00 1.03
20 min acq. 1.13 1.14 1.01 1.01
30 min acq. 0.98 1.00 0.93 0.98
Fan-beam 128 x 128 array
10 min acq. 2.52 1.83 1.24 1.30
20 min acq. 1.22 1.17 0.98 1.04
30 min acq. 1.19 1.12 1.00 1.06
Table 3. Ratio of measured to real concentration in the centre of the hollow spheres of the Jaszczak phantom
filled with 3 µCi/ml (111 kBq/ml) 99m
Tc, demonstrating the partial volume effect
Sphere diameter (mm)
33.5 28.0 22.0 16.0 13.5 10.5
Parallel 64 x 64 array 0.98 0.95 0.88 0.62 0.49 0.30
Parallel 128 x 128 array 1.03 0.99 0.94 0.60 0.48 0.27
Fan-beam 64 x 64 array 1.03 0.96 0.91 0.70 0.49 0.33
Fan-beam 128 x 128 array 1.02 1.01 1.00 0.83 0.66 0.42
The influence of sphere diameter on the ratio of measured to real concentration in the centre of the
hollow spheres (mean value of three acquisitions) is shown in Table 3. The partial volume effect is
larger for the parallel collimator than for the fan-beam collimator since the FWHM (reconstructed
in air, radius 12.5 cm) is 10.1 mm for the former compared to 7.3 mm for the latter.
Table 4 shows the results of the measured to real concentration for the cerebellar region of the
Hoffman phantom. Again, high point source activity is necessary for the fan-beam acquisitions,
whereas at high source activity compared to organ activity, the point source is slightly
overestimated after reconstruction for the parallel collimator. In this respect, Galt et al. [13]
described activity overestimation due to the cutoff frequency when comparing high- and low-
activity sources. We presumed that for the cerebellar region of the Hoffman phantom the partial
volume effect was negligiable. Lower activities due to a partial volume effect are observed in
different regions of the cortex, as shown by the circumferential maximum value profile in Fig. 1.
Human studies: reproducibility of 99m
Tc-HMPAO brian uptake
The reproducibility of the point source activity in repeated human studies is shown in Table 5. A
mean deviation of 2.4% is observed intra-individually. The interval between the two studies in each
volunteer was restricted to less than 1 week in an attempt to avoid changes in the sensitivity of the
gamma camera. Due to the reproducibility of the point source method, this technique of data
calibration with sources can be applied for SPET studies, correcting for the changes in sensitivity of
the gamma camera over time. The individual cerebellar uptake values are presented in Table 6. A
mean deviation of 12.6% (SD=10.3%) in cerebellar uptake is observed intra-individually. The
82
correlation coefficient between the first and second acquisition was 0.80 (Fig. 2). Six volunteers and
four patients had a less than 5-unit difference in heart rate between the two studies (Table 6). The
mean deviation in cerebellar uptake for those individuals was 7.2% (SD=4.1%), and the correlation
coefficient was 0.94 (Fig. 3). Probably due to stress, four volunteers and one patient had a
significant higher heart rate (mean 28%) and a lower cerebellar uptake (mean 21%) in the first
investigation than in the second. Unlike in the case of rCBF, interindividual comparison with regard
to percentage of brain uptake is not possible, mainly due to differences in cardiac output or body
surface area
Table 4. Ratio of measured to real concentration in the Hoffman phantom filled with 3 µCi/ml (111 kBq/ml) 99m
Tc for different point source activities and a total acquisition time of 20 min. The concentration was
measured in a 5 x 5 mm ROI corresponding to the area of highest activity in the cerebellar region of the
phantom (largest hot region in phantom, where no partial volume effect is expected)
Point source activity
74 kBq 148 kBq 296 kBq 444 kBq
Parallel 64 x 64 array 1.02 0.97 0.97 0.93
Parallel 128 x 128 array 1.00 0.97 0.94 0.90
Fan-beam 64 x 64 array 1.16 1.14 1.03 0.97
Fan-beam 128 x 128 array 1.40 1.21 1.07 1.00
Fig. 1. Circumferential maximum value profile of a Hoffman phantom in a basal ganglia slice normalised to
the mean activity. 90° corresponds to frontal activity.
Table 5. Reproducibility of point source activity, in counts/µCi, for repeated HMPAO brain studies in ten
volunteers.
Volunteer Acquisition 1 Acquisition 2
1 29199 27536
2 31132 30674
3 30990 31177
4 31476 30885
5 28940 29230
6 33910 32751
7 31709 32065
8 32498 33512
9 33067 31833
10 31850 31970
83
Fig. 2. Reproducibility of HMPAO uptake in the cerebellum (10-6/cm
3.ID) in ten volunteers and five patients.
A correlation coefficient of 0.80 was obtained (uptake 2=1.10 x uptake 1)
Table 6. Reproducibility of cerebellar uptake in 10
-6 of injected dose per cm
3 brain tissue for repeated
HMPAO brain studies in ten normal volunteers and five patients
Uptake 1 Uptake 2 % change Heart rate 1 Heart rate 2
Volunteer 1 72.2 65.0 -10.5 74 74
Volunteer 2 42.0 39.7 -5.7 60 60
Volunteer 3 38.6 44.1 13.4 68 68
Volunteer 4 51.0 53.3 4.5 76 72
Volunteer 5 66.6 68.6 3.0 76 80
Volunteer 6 33.3 31.9 -4.4 58 63
Volunteer 7 49.4 60.2 17.6 84 70
Volunteer 8 37.3 55.2 38.7 80 68
Volunteer 9 47.9 55.0 13.8 74 56
Volunteer 10 41.6 56.2 30.0 108 72
Patient 1 36.2 37.1 2.4 77 72
Patient 2 53.3 51.2 -4.0 82 86
Patient 3 48.2 55.0 13.1 88 88
Patient 4 49.7 54.7 9.6 104 105
Patient 5 10.1 46.9 15.6 96 80
Discussion
Point sources were used as an external standard to convert counts from the camera into
concentrations of activity in the brain. In order to use the optimal point source activity, similar
once-only phantom work is necessary when employing a single detector system or different
collimators. Since the correct calibration is strictly dependent on ROI extension and background
substraction, the ROI diameter and axial distance (number of slices) should be adapted to the
resolution of the system. The use of point sources for calibration introduces a supplementary
measurement error compared to the cylindrical uniform phantom technique. On the other hand,
tomographic sensitivity variations of the gamma camera, even during scan, do not influence the
point source method because of the use of the target organ on point source ratio. Regular, possible
daily, calibration using the cylindrical uniform phantom is necessary. Since a relatively high
number of counts have to be acquired at a relatively low count rate to avoid any influence of the
84
non-linearity response of the gamma camera, implementing the cylindrical uniform phantom
method in routine practice might be more cumbersome.
Fig.3. Reproducibility of HMPAO uptake in the cerebellum (10
-6/cm
3.ID) in six volunteers and four patients
in whom no significant heart rate difference was observed between the two investigations. A correlation
coefficient of 0.94 was obtained (uptake 1=1.02 x uptake 2)
Despite the greater point source activity dependence for correct estimation of the phantom
concentrations, in patient studies we preferred the SFFB and 128 x 128 reconstruction array because
of the minimal partial volume effect. An acquisition time of 40 min was then needed [14]. From the
results obtained with the Hoffman and Jaszczak phantoms we deduced that point sources of
approximately 5µCi (185 kBq) were optimal. For a parallel collimator the point source activity is
less critical. The filter cutoff frequency was not the cause of overestimation of the fan-beam data at
low source activity. For these data a negligible difference was observed when a Butterworth filter
was applied, varying the cutoff frequency from 0.5 to 1.7 cyc/cm. Overestimation of fan-beam data
at low point source activity with a short acquisition time is probably due to the rebinning process,
the transformation from fan-beam to parallel data, before backprojection. The phantom studies seem
to indicate that there is a minimum count density below which the rebinning algorithm cannot be
used. This should be verified by comparing different computer systems. Reprojecting the fan-beam
data without rebinning eliminated the underestimation of the point sources. However, this method is
not preferred because of high statistical noise in the reconstructed slices. The same effect might
explain higher contrast between grey and white matter in fan-beam reconstructed images.
Kojima et al. described the effect of spatial resolution on quantification [15]. Due to a partial
volume effect, we could only recover real activity in spheres larger than 20 mm using SFFB and
larger than 25 mm using UHRP collimators. Similar results were presented by Szabo et al. [16] for
high-resolution SPET with a three-detector system. This partial volume effect explains the
underestimation in the circumferential profile of the Hoffman phantom, as previously described by
Kim [17]. This can be a major problem in comparing different patients because of different
physiology in respect of not only rBu but also rCBF. Nevertheless, providing tomographic slices in
rBU values might be useful, intra-individually, for activation studies, follow-up and
pharmacological studies. Since no internal reference region is needed, clinically the scope of
indications may be extended and methodological discussions on the choice of reference region are
unneccessary [18]. In order to reduce the cost, by splitting the vial, the lipophilic fraction should be
investigated for absolute quantification. Good intra-individual reproducibility was obtained in cases
of constant heart rate, although large interindividual differences were obtained, as previously
described by Prodreka et al. [19]. The influence of heart rate on rBU is currently being studied in a
larger patient group. It has been shown that the brain extraction of HMPAO is lower at higher flow
85
[20]. Measurement of the absolute uptake might be useful in order to determine the retention factor
for the linearisation correction method of Lassen et al. [20].
To conclude: Although the results are based on a three-detector system with high-resolution
collimators and a relatively small rotation radius, the quantification method should be applicable to
most SPET systems after careful phantom work. Different results can be expected for small brain
structures using a one-detector system with lower resolution collimators and a larger rotation radius.
The major advantage of the described method is its simplicity for routine clinical use, the same
sources being used daily for all patient acquisitions and the final result being represented in terms of
regional brain uptake.
Acknowledgements.
The authors wish to thank Dr. Vandevivere, Head of the Nuclear Medicine Departement, for
providing the opportunity to develop this method and Prof. De Deyn, Head of the Departement of
Neurology, for moral support and addvice. They are also grateful to the technicians of the
department for their collaboration in performing the patient studies.
References
1. Sharp PF, Smith FW, Gemmell HG, et al. Technetium-99m HMPAO stereoisomers as
potential agents for imaging regional cerebral blood flow: human volunteer studies. J Nucl
Med 1986;27:171-177.
2. Costa DC, Ell PJ, Cullum ID, Jarritt PH. The in vivo distribution of 99Tc
m HMPAO in
normal man. Nucl Med Commun 1986:7:647-658.
3. Nowotnik DP, Canning LR, Cumming SA, et al. Development of a TC-99m-labelled
radiophamaceutical for cerebral blood flow imaging. Nucl Med Commun 1985;6:499-506.
4. Matsuda H, Oba H, Seki H, et al. Determination of flow and rate constants in a kinetic
model of 99m
Tc-hexamethyl-propylene amine oxime in the human brain. J Cereb Blood
Flow Metab 1988;8:61-68.
5. Nickel O, Nägele-Wöhrle B, Ulrich P, et al. rCBF-quantification with 99m
Tc-HMPAO-
SPECT: theory and first results. Eur J Nucl Med 1989;15:1-8.
6. Pupi A, De Cristofaro M, Bacciotttini L et al. An analysis of the arterial input curve for
technetium-99m-HMPAO: quantification of rCBF using single-photon emission computed
tomography. J Nucl Med 1991;31:1501-1506.
7. Murase K, Tanada S, Fujita J, et al. Kinetic behavior of technetium-99m-HMPAO in the
human brain and quantification of cerebral blood flow using dynamic SPECT. J Nucl Med
1992;33:135-143.
8. Matsuda H, Tsuji S, Shuke N, et al. A quantitative approach to technetium-99m
hexamethylpropylene amine oxime. Eur J Nucl Med 1992;19:195-200.
9. Szabo Z, Monsein LH, Maruki et al. Quantitative imaging of CBF with Tc-99m HMPAO
[abstract]. Eur J Nucl Med 1991;18:667.
10. Nakamura K, Tukatani Y, Kubo A, et al. The behavior of 99m
Tc-hexamethyl
propyleneamineoxime (99m
Tc-HMPAO) in blood and brain. Eur J Nucl Med 1989;15:100-
107.
11. Hoffman EJ, Cutler PD, Digby WM, Maziotta JC. 3-D phantom to simulate cerebral blood
flow and metabolic images for PET. IEEE Trans Nucl Sci 1990;37:616-620.
12. Hung JC, Corlija M, Volkert WA, Holmes RA. Kinetic analysis of technetium-99m d,I-
HMPAO decomposition in aqueous media. J Nucl Med 1988;29:1568-1576.
86
13. Galt JR, Grant SF, Alazraki NP. Effect of system resolution on quantitative measurements
of the cerebral cortex and cerebellum and Spect brain images [abstract]. J Nucl Med
1991;32:728.
14. Dobbeleir A, Dierickx R, Vandevivere J. High spatial resolution Spect using a three-head
rotating gamma camera and super fine lead fan-beam collimators [abstract]. Eur J Nucl Med
1991;18:600.
15. Kojima A, Matsumoto M, Takahashi M, et al. Effect of spatial resolution on Spect
quantification values. J Nucl Med 1989;30:508-514.
16. Szabo Z, Seki C, Rhine J, et al. Effect of spatial resolution on absolute quantification with
high resolution Spect [abstract]. Eur J Nucl Med 1991;18:604.
17. Kim HJ, Zeeberg B, Fahey F, et al. Three-dimensional Spect simulations of a complex
three-dimensional mathematical brain model and measurements of the three-dimensional
physical brain phantom. J Nucl Med 1991; 32:1923-1930.
18. Syed G. Eagger S, Toone D, et al. Quantification of regional cerebral blood flow (rCBF)
using 99Tc
mHMPAO and SPECT: choice of the reference region. Nucl Med Commun
1992;13:811-816.
19. Podreka I, Goldenberg G, Baumgartner C, et al. HMPAO brain uptake in young normal
subjects: gender differences and hemispheric asymmetries. J Cereb Blood Flow Metab
1989;9 Suppl 1:202.
20. Lassen N, Andersen A, Friberg L, Paulson O, The retention of 99Tc
m-d,I-HMPAO in the
human brain after intracarotid bolus injection: a kinetic analysis. J Cereb Blood Flow Metab
1988;8(Suppl):S13-S22.
87
4.3 Clinical applications.
Parameters influencing SPET regional brain uptake of technetium-99m hexamethylpropylene amine oxime measured by calibrated point sources as an external standard. R.A. DIERCKX, A. DOBBELEIR, M. MAES, B.A. PICKUT, A. VERVAET, P.P. DE DEYN. Nuclear Medicine and Neurology , Middelheim Hospital and Born-Bunge Foundation Antwerp, Belgium. Psychiatry, University Hospitals of Cleveland, USA.
European Journal of Nuclear Medicine, 1994; 21: 514-520. Abstract
Using calibrated point sources as external standard to convert single-photon emision tomography (SPET) brain counts
into absolute values of regional brain uptake (rBU) of technetium-99m hexamethylpropylene amine oxime (HMPAO), the
relative contribution of different parameters to inter-individual variability of cerebellar rBU was examined in 33 healthy
volunteers. Stepwise regression analysis identified body surface as the most important factor underlying inter-individual
variability (P<0.001), when compared with brain volume. In the normal volunteer population presented, age decrement of
rBU corrected for body surface and brain volume equalled
60.05-0.20xage. Based on the data of eight normal volunteers, including four test-retest studies with heart rate (HR)
differences greater than 5 units and four test-retest studies with doubling of heart rate after bicycle exercise, influence of
heart rate may be expressed by the equation ∆rBU=0.35 ∆HR. Clinically, estimation of the relative influence of different
factors allows normalization and extension of the applicability of the rBU quantification method used from longitudinal
studies to group comparisons. Interestingly, results of the Daily Stress Inventory Scale and a subjective rating scale
suggest the absence of a significant influence of minor stress on rBU. When using one vial per patient, chromatography
may be omitted in clinical routine practice and lipophilicity may be estimated as 90% of the injected dose, if administered
within 10 min after preparation. Finally, sensitivity of the quantification method was tested in eight volunteers using
acetazolamide brain activation and showed a mean increase in cerebellar rBU of 30.2%, varying between 14.1% and
75.9%.
Group rBU at rest heart rate rBU BS corrected rBU BS+BV corrected
Global (n=33)
48.4±9.7
52.1±7.75
52.0±7.27
Men (n=18) 43.1±7.56 49.9±8.4 51.6±8.68
Women (n=15)
54.8±8.12 54.7±6.18 52.5±5.45
Summary of rBU (mean±stan- dard deviation) in men and women separ- ately and in the whole group, rBU at rest heart
rate, rBU after correction of body surface
and rBU after cumulative correc- tions for body surface and brain volume
rBU (mean±standard deviation) in 10-6 of the injected lipophilic dose per cm3 brain tissue; n,
number of volunteers; BS, body surface (index per 1.73 m2); BV, mean brain volume of 1350 ml
88
Validation of the cerebellum as a reference region for SPECT quantification in patients suffering from dementia of the Alzheimer type.
B.A. PICKUT, R.A. DIERCKX, A. DOBBELEIR, K. AUDENAERT, K. VAN LAERE, A. VERVAET, P.P. DE DEYN. Neurology and Nuclear Medicine, Middelheim Hospital and Born-Bunge Foundation Antwerp, Belgium. Nuclear Medicine and Psychiatry, University Hospital Gent, Belgium. Psychiatry Research: Neuroimaging Section 1999; 90: 103-112. Abstract
In longitudinal brain studies of dementia of the Alzheimer type (DAT), the cerebellum is often used as a reference region
for single photon emission computed tomography (SPECT) quantification, which assumes no significant regional
influence of physiological fluctuations or pathology. With the use of absolute quantification in DAT patients, reproducibility
of cerebellar uptake of technetium-99m-d,l- hexamethylpropyleneamine oxime (HMPAO) was tested and compared with
the mean absolute cerebellar tracer uptake in DAT patients and healthy control subjects. In 13 DAT patients SPECT
studies were repeated within 2 weeks to assess reproducibility of cerebellar regional brain uptake (rBU). With calibrated
point sources as scaling factors, cerebellar activity was expressed as rBU of HMPAO per cm3 brain tissue in percent of
the injected lipophilic dose of 740 MBq (20 mCi). Also, mean cerebellar rBU in patients suffering from DAT was
calculated and compared with a previously established database obtained in healthy volunteers. Repeated SPECT
studies within a 2-week interval in clinically stable patients resulted in a mean rBU increase of 6.8±10.3% in the second
study as compared with the first. A similar shift was previously reported in healthy volunteers. Mean cortical cerebellar
rBU values in DAT patients and in the healthy reference population concurred, after cumulative corrections for body
surface and for a mean brain volume of 1350 ml (obtained in healthy control subjects), showing respective mean values
of 53.9±7.4 and of 52.0±7.3 x 10 –6 of the injected lipophilic dose 740 MBq (20 mCi) of HMPAO per cm
3 brain tissue . A
unidirectional shift in mean absolute cerebellar uptake values occurs between repeat examinations in DAT patients
similar to previous findings in a group of healthy volunteers. The origin of this phenomenon remains elusive but deserves
further study with regard to SPECT (semi)quantification in DAT patients. Most interestingly, the presented findings
suggest that the use of HMPAO SPECT in DAT patients the cerebellum remains scintigraphically uninvolved.
Demographic data, MMSE scores, and individual cerebellar regional brain uptake in 13 patients with dementia of the Alzheimer type (DAT), respectively, at resting heart rate and after cumulative corrections for body surface and for a mean brain volume of
1350 ml (determined in healthy volunteers)
Patient number
Gender/age MMSE score
rBU1 rBU2 BS index
rBU2- corr1
BV (in cm3)
rBU2- corr2
1 F/68 11 83.8 85.4 0.75 63.9 1150 54.4
2 F/76 12 62.5 73.9 0.89 65.8 1220 59.5
3 F/83 16 70.1 72.9 0.84 61.2 1060 48.1 4 F/67 22 98.3 124.7 0.72 90.4 920 61.6
5 F/72 16 72.6 72.8 1.00 72.8 1037 55.9
6 F/74 19 72.6 82.9 0.89 73.8 1125 61.5 7 F/73 7 81.8 78.7 0.92 72.4 1194 64.0
8 F/78 5 58.9 71.1 0.77 54.7 1053 42.7
9 F/84 13 60.4 61.3 0.90 55.2 1060 43.3 10 F/83 15 62.2 64.4 0.87 56.0 1105 45.8
11 M/78 25 46.1 43.0 1.18 50.7 1274 47.8
12 M/78 19 78.2 75.7 0.95 71.9 1092 58.1 13 M/82 17 57.7 70.8 1.02 72.2 1086 58.1
Notes. rBU, regional (cerebellar) brain uptake, 10-6 of the injected lipophilic dose 740 MBq (20 mCi) of Tc-99m HMPAO per
cm3 brain tissue; rBU1, rBU for first examination; rBU2, rBU for repeat study; BS, body surface, index per 1.73 m2; BV, brain volume in cm3; rBU2 corr1, regional brain uptake for repeat study corrected for BS; rBU2 corr2, regional brain uptake for repeat
study corrected for BS and BV of 1350 ml.
89
5. Dopamine transporter imaging in the human brain.
Quantification of Iodine-123-FPCIT SPECT with a resolution independent technique.
Andre A. Dobbeleir
1,3, Anne-Sophie E. Hambye
2, Ann M. Vervaet
1 and Hamphrey R. Ham
3.
1Nuclear Medicine, Middelheim Hospital, Antwerp, Belgium
2Nuclear Medicine, CHU-Tivoli, La Louvière, Belgium
3Nuclear Medicine, University Hospital Ghent, Ghent, Belgium
Submitted.
Abstract Accurate quantification of small-sized objects by SPECT is hampered by the partial volume effect.
The present work evaluates the magnitude of this phenomenon with iodine-123 in phantom studies,
and presents a resolution-independent method to quantify striatal 123
I-FP-CIT uptake in patients.
Methods: First, five syringes with internal diameters varying between 9 and 29mm and an
anthropomorphic striatal phantom were filled with known concentrations of iodine-123 and imaged
by SPECT using different collimators and radii of rotation. Data were processed with and without
scatter correction. From the measured activities, calibration factors were calculated for each specific
collimator.
Second, a resolution-independent method for FP-CIT quantification using large regions of interest
was developed and validated in 34 human studies (controls and patients) acquired in 2 different
hospitals, by comparing its results to those obtained by a semi-quantitative striatal-to-occipital
analysis. Taking the injected activity and decay into account, the measured counts/volume could be
converted into absolute tracer concentrations.
Results: For the fan-beam, high resolution and medium energy collimators, the measured
maximum activity in comparison to the 29 mm-diameter syringe was respectively 38%, 16% and
9% for the 9 mm-diameter syringe and 82%, 80% and 30% for the 16 mm syringe, and not
significantly modified after scatter correction. For the anthropomorphic phantom, the error in
measurement in % of the true concentration ranged between 0.3-9.5% and was collimator
dependent. Medium energy collimators yielded the most homogeneous results.
In the human studies, inter-observer variability was 11.4% for the striatal-to-occipital ratio and
3.1% for the resolution-independent method, with correlation coefficients >0.8 between both. The
resolution-independent method was 89%-sensitive and 100%-specific to separate the patients
without and with abnormal FP-CIT uptake (accuracy: 94%). Also the quantification in % of the
injected activity was well correlated with both the striatal-to-occipital and the resolution-
independent ratios.
Conclusion: Partial volume effect severely affects the quantification of FP-CIT uptake and results
in a rather large inter-observer variability. Measurement of striatal and non-specific activity in large
regions of interest circumvents this problem and provides stable and reproducible resolution-
independent results, allowing a direct comparison between data acquired with different imaging
systems or even in different hospitals.
Keywords: dopamine transporters, Parkinson’s disease, 123
I-FP-CIT, partial volume effect, SPECT,
quantification, anthropomorphic striatal phantom.
90
Introduction
Parkinson’s disease is a degenerative process of the dopaminergic neurons in the caudate and
putamen nuclei of the substantia nigra, resulting in progressive cellular loss (1-3).
The cocaine-like radioligand, iodine-123-FP-CIT ([123
I]N-ω-fluoropropyl-2β-carbomethoxy-3β-(4-iodophenyl)nortropane) developed by Neumeyer et al binds with high affinity to the striatal
dopamine transporters (DAT) in humans (4), and several studies have demonstrated the value of
FP-CIT brain SPECT imaging in the work up of patients with movement disorders (5-16).
Though the primary evaluation of a FP-CIT scan relies on the visual analysis, many authors
recommend a quantitative approach besides the visual to increase the objectivity of interpretation,
especially in doubtful cases (17-19).
Usually, this quantification consists in measuring the ratio of striatal (specific) to occipital (non
specific) FP-CIT uptake using manually drawn and/or positioned regions of interest (ROIs).
Due to the rather small size of these regions however, quite large variations in the measured ratios
have been reported, questioning the robustness and reliability of the technique. Even in normal
subjects, Seibyl et al found values for the putamen and caudate ratios between 1.9 and 6.3 (17).
Using SPECT, the low spatial resolution of the gamma camera and its consequent partial volume
effect hampers an accurate measurement of the count density in small-sized objects.
In FP-CIT studies, the small size of the human striatum has the effect that quantification is
influenced by the ROIs size and positioning, and therefore observer-dependent (17-19). In a recent
publication, Fleming and al have demonstrated that the influence of partial volume effect, and the
resulting variability in quantifying the radioactive content of small objects, can however be largely
reduced by using comfortable ROIs to measure the total uptake in an object rather than its activity
concentration (20).
FP-CIT imaging is gaining increasing importance in the differential diagnosis of patients with
movement disorders, and the availability of a reliable, reproducible and simple index to quantify
FP-CIT uptake should probably be of value to set up multi-centric studies.
Therefore, we designed the present work first to evaluate the magnitude of the partial volume effect
and its potential negative consequences in iodine-123 SPECT imaging. For this purpose, phantoms
of various sizes were filled with known concentration of radioactivity and imaged by SPECT with
different collimators and radii of rotation to quantify the influence of the system resolution on the
results. Processing was done without or with scatter subtraction to analyze the possible effects of
scatter.
Second, we developed a resolution-independent method for striatal FP-CIT quantification, based
upon the use of large ROIs to circumvent the partial volume effect. This method was validated in a
retrospective study using human data acquired in two different hospitals, by comparing its results to
those obtained by a semi-quantitative striatal-to-occipital analysis.
Materials and methods
Phantom studies
Syringes:
Five syringes of 2, 5, 10, 20 and 50 ml (internal diameter: 9, 13, 16, 20 and 29 mm respectively)
were filled with the same concentration of iodine-123 (1.30 MBq/ml - 35 µCi/ml). In total, 45
SPECT acquisitions were performed in air at a radius of rotation of 13.5, 15.0 and 16.5 cm with a
Triad triple-head gamma camera (Trionix, Ohio, USA) equipped with either fan-beam, high
resolution or medium energy collimators. Hundred and twenty images (3x40) of 4 seconds with a
pixel size of 3.56 mm were acquired using the triple energy window method described by Ichihara,
with the photopeak set at 143-175 kev (159 ± 10%) and the scatterpeaks at 125-141 keV and 180-
196 keV (21). The triple energy window method attempts to correct for both scatter and septal
penetration generated by the high energy photons of iodine-123 (22).
91
Each photopeak and scatter corrected acquisition was prefiltered with a Butterworth filter (order 6,
cut-off 1.0 cyc/cm) and 3.56 mm-thick transaxial slices were reconstructed. A linear profile of the
activity through the transaxial slices was measured to determine the maximum. All the maximum
activities were extrapolated to the initial time of filling to correct for radioactive decay.
With its 29 mm internal diameter, the 50 ml syringe is more than twice the SPECT resolution.
Therefore, it was considered as unaffected by the partial volume effect and used as a reference to
assess the magnitude of partial volume effect in the smaller syringes.
Additionally, from the known activity in each syringe and the total counts measured in the
transaxial slices, the gamma camera sensitivity factor for each collimator could be calculated with
and without scatter correction, allowing to transform the measured counts into MBq.
Anthropomorphic phantom:
An anthropomorphic striatal phantom was used to investigate the magnitude of the partial volume
effect and the reliability of the measured striatal activity. The investigations were performed in
cascades.
First, only the left and right striatum (both 11 ml) were filled with 555 and 407 kBq (15 and 11 µCi)
iodine-123. SPECT acquisition was performed using the same triple energy window, with the 3
previously described collimators but with 40 sec/image at a radius of 15 cm. Second, the remaining
brain structures were filled with 7.40 MBq (200 µCi) iodine-123 in 1300 ml and the phantom was
rescanned. The true contrast ratio between striatum and brain was measured afterwards in a well-
counter. Finally, the filled brain was incorporated in the skull and a 50 cm long catheter filled with
740 kBq (20 µCi) 99m
Tc was fixed around the skull at the canthomeatal level before repeating the
acquisition.
For the processing, each photopeak and scatter corrected acquisition was prefiltered using a
Butterworth filter (order 6, cut-off 1.0 cyc/cm) and 3.56 mm-thick transaxial slices were
reconstructed. Linear attenuation correction was performed on the second and the third sets of data
within an ellipse drawn around the brain or on the catheter fixed on the skull. Attenuation
coefficients of respectively 0.11 cm-1 for the brain, 0.12 cm
-1 for the brain + skull, and 0.15 cm
-1
after scatter correction were applied.
Human studies
Acquisition:
Thirty-four FP-CIT studies acquired in two different hospitals were retrospectively analyzed,
including 6 healthy volunteers (medical staff) and 28 patients referred to the nuclear medicine
department because of clinically unclear differential diagnosis between essential tremor and
Parkinson’s disease. By visual analysis of the FP-CIT study, 10 patients were considered as normal
and 18 patients as having significant abnormalities of the dopaminergic system. To validate the
resolution-independent quantification, the 10 patients with a visually normal FP-CIT scan and the 6
healthy volunteers were considered as the control group (mean age 62.9 y; range 39-90) while the
18 with a pathological test constituted the patients group (mean age 64.7 y; range 31-81).
Each subject received an intravenous injection of 148-222 MBq (4-6 mCi) [I123
]FP-CIT (Amersham
Cygne, Eindhoven, The Netherlands). Prior to the injection, potassium iodide or natrium
perchlorate was orally administered to block the thyroid uptake of free iodine. SPECT acquisition
was performed on a the Triad triple-head gamma camera equipped with fan-beam collimators or on
a MultiSPECT3 triple-head gamma camera (Siemens, Illinois, USA) equipped with low energy
high resolution collimators. Fourty views/head of 45 second each were acquired with a 20%
symmetric energy window centred around 159 keV (total scan duration: 30 minutes). Before
imaging, four skin markers containing 148 kBq (4 µCi) iodine-123 were taped at the level of the
canthomeatal line, two on each side of the subject’s head. They were used for posthoc reorientation
of transaxial images, and delineation of the ellipse around the brain for attenuation correction. In
92
some patients, these markers were replaced by a 50 cm long catheter filled with 740 kBq (20 µCi) 99m
Tc.
Raw data (pixel size: 3.56 mm) were transferred to a SunBlade 1500 Link Medical computer
system (Link Medical, Hampshire, UK) and reconstructed with a Butterworth prefilter (order 6, cut-
off 1.0 cyc/cm) and a ramp reconstruction filter. After reorientation parallel to the canthomeatal
plane, transaxial slices were corrected for attenuation using Chang’s algorithm on an ellipse fitted
on the catheter or through the skin markers. A linear attenuation factor of 0.12 cm-1 was used (17).
No scatter correction was applied.
Data analysis:
First, the 3 axial slices with the highest striatal activity were manually selected and summed
(classical semi-quantification). ROIs were manually drawn around the nucleus caudate, the
putamen, and on the occipital cortex as the non-specific binding area, and a semi-quantitative
striatal-to-occipital ratio (V”3) was calculated as:
V”3 = (caudate or putamen activity - occipital activity) / occipital activity.
Second, larger ROIs were drawn to obtain a resolution-independent semi-quantification. For this
purpose, 13 slices enclosing the whole striatum were summed (total thickness 4.6 cm), on which
two large ROIs were drawn: a 200 cm3 “striatal” ROI over the striatum (normal volume 20 cm
3),
and an elliptical ROI enclosing almost the whole brain (brain ROI).
The following parameters were then calculated.
(1) A resolution-independent striatal uptake, similar to the specific uptake size index recently
presented by Fleming (20): by subtracting the striatal ROI activity from the brain ROI, an operator-
insensitive non specific brain activity was obtained. After normalisation for the ROI size, the non
specific brain activity was subtracted from the striatal ROI activity to obtain the total striatal
activity. The resolution independent striatal uptake ratio was obtained by dividing the total striatal
activity by the mean brain activity/cm3.
Bg = Brain _ striatum
ROIind = Striatumcts – ( Bg
cts x Striatum
cm3/Bg
cm3 )
Bgcts / Bg
cm3
(2) The total striatal uptake, expressed as a percentage of the injected activity: using the gamma
camera sensitivity factor (specific for the collimator considered and total scan time), the total
striatal activity was transformed to real activity expressed in MBq. The striatal uptake in % of the
injected activity was then calculated, left and right striatum uptake obtained from the total striatal
uptake, and left and right activity measured in large mirrored ROIs on the background corrected
striatal image.
For display purposes, a density image of the striatum was generated from the total uptake values
expressed in % of the injected activity per cm3 striatal tissue. A colour-coded scale was used, each
colour corresponding to a specific range of percentage of the injected activity.
Reproducibility study:
To calculate the reproducibility of striatal uptake measurement for both the striatal-to-occipital ratio
and the resolution-independent method, the 34 data sets were processed by 2 observers
independently, starting from the raw acquisition. The inter-observer variability was expressed as the
mean ± S.D. of the absolute differences in % between the two observers.
93
Results
Phantom studies
The magnitude of the partial volume effect is reported in Table 1 for the different types of
collimators. Expressed as a fraction of the maximum activity of the 50ml-syringe, the measured
activity varied from 9%-40% for the 2 ml-syringe up to 50-100% for the 20ml-syringe and was
clearly collimator dependent, the best results being measured with the fan-beam collimators and the
worse with the medium energy. Scatter subtraction did not significantly improve the accuracy of the
measurements, as can be expected when imaging a small source with little scatter material. On the
other hand, increasing the scan radius of rotation from 13.5 to 16.5 cm resulted in an additional
mean maximum activity loss of respectively 6.0%, 4.9% and 3.9% with the medium energy, high
resolution and fan-beam collimators.
Table 1. Influence of the partial volume effect for the different collimators, without and with scatter
subtraction. The maximum activity of the linear profile through the transaxial slices of the 2, 5, 10 and 20 ml
syringes is expressed in % of the maximum activity of the 50 ml-syringe used as the reference.
Fan-beam High resolution
Medium energy
Syringe
volume
Internal
diameter Max
counts
Max cnts
- scatter
Max
counts
Max cnts
- scatter
Max
counts
Max cnts
- scatter
2 ml 9 mm 38 40 16 17 9 9
5 ml 13 mm 63 67 33 36 17 17
10 ml 16mm 82 88 80 84 30 31
20 ml 20 mm 98 100 81 84 50 51
50 ml 29 mm 100 100 100 100 100 100
From the ratio between the measured total counts in the syringe and the true activity for a scan
duration of 30 minutes, gamma camera sensitivity factors could be calculated without and with
scatter correction for the each type of collimator. The values of the different sensitivity factors are
given in Table 2.
Table 2: Gamma-camera sensitivity factors (in counts/µCi.scan) for the different collimators, without and
with scatter correction (scan duration: 30 minutes).
Fan-beam High resolution Medium energy
Non
corrected
Scatter
corrected
Non
corrected
Scatter
corrected
Non
corrected
Scatter
corrected
Sensitivity
6770
6176
7511
6760
9127
8411
94
Using the anthropomorphic phantom, the most homogeneous results were obtained with the
medium energy collimators, all ratios between measured and true striatal activity remaining within a
3 % error regardless of whether the striatum was imaged alone, with brain activity or with skull
attenuation (Table 3). With the other sets of collimators, the error amounted to 3.6% at the very
most when scanning the striatum alone, but increased up to a maximum of 9.5 % after addition of
activity in the other brain structures (Table 3).
Table 3: Ratio (in %) between the measured and the true activity in the anthropomorphic striatal phantom for
the different collimators, without and with scatter correction.
Fan-beam High resolution Medium energy
Non
corrected
Scatter
corrected
Non
corrected
Scatter
corrected
Non
corrected
Scatter
corrected
Striatum alone 103.6 102.8 102.9 100.3 98.2 97.9
Striatum + brain 107.4 96.0 109.5 100.5 97.2 98.1
Striatum + brain
+ skull
103.9 101.5 99.1 106.8 98.5 102.5
In a well counter, the true contrast ratio between left striatum and brain was 7.7. Measured on the
images, its value was 3.5 for the fan-beam and 3.0 for the high resolution collimator in the centre of
the left striatum, and dropped to 2.5 and 2.0 respectively within a 1 cm diameter ROI drawn on the
central slice. A partial volume factor of a magnitude of 3 seems therefore a good approximation for
the count loss.
Human studies
Using the striatal-to-occipital ratio, mean±SD absolute inter-observer variability was 11.4±16.9 %
(3.6±3.1 % in the control group and 21.4±22.4 % in the patients group due to the much lower
uptake), versus only 3.1±1.8 % with the resolution-independent striatal-to-brain method.
The results of the three different methods for striatal uptake measurement are presented in Table 4.
Globally, as well the striatal-to-occipital ratio as both resolution-independent methods (striatal-to-
brain activity/cm3 and in % of the injected activity) allowed a good separation between controls and
patients. Two of the 18 patients (11%) had values within the normal range with the striatal-to-brain
methods (Figure 1), resulting in a sensitivity of 89%, a specificity of 100% and an accuracy of 94%.
With the method expressing the uptake as a percentage of the injected activity, only one patient was
not correctly classified but 3 controls had values at the lower range (sensitivity: 94%, specificity:
81%, accuracy: 88%) (Figure 2). The best separation between controls and patients was obtained
after normalisation for the body weight (sensitivity: 94%, specificity: 100%, accuracy: 97%)
(Figure 2).
95
Figure 1. Striatal activity divided by the mean brain activity/cm3 in the summed 13 slices using the resolution
independent method. The total striatal activity is shown on the left graph, and the individual left and right
activity at the right. For each graph, the normal subjects are at the left and the Parkinson patients at the right.
total striatum/mean brain activity
0,0
50,0
100,0
150,0
200,0
250,0
normal subjects patients
striatum / mean brain activity
0,0
20,0
40,0
60,0
80,0
100,0
120,0
left striatum
right striatum
patientsnormal subjects
Figure 2. Plots of the total, left and right striatal uptake expressed in % of injected dose, without (top graphs)
and with normalization (bottom graphs) for the patient’s weight.
For each graph, the normal subjects are at the left and the Parkinson patients at the right.
total uptake-raw data
0,000
0,200
0,400
0,600
0,800
1,000
1,200
normal subjects patients
left - right uptake raw data
0,000
0,100
0,200
0,300
0,400
0,500
0,600
left uptake
right uptake
normal subjects patients
total uptake weight normalised
0,000
0,200
0,400
0,600
0,800
1,000
1,200
normal subjects patients
left - right weight normalised
0,000
0,100
0,200
0,300
0,400
0,500
0,600
left uptake
right uptake
normal subjects patients
By regression analysis, the three methods were nicely correlated with each other (Figure 3). For the
right and the left striatum respectively, the correlation coefficients were 0.86 and 0.82 between the
striatal-to-occipital ratio and the striatal-to-brain method, 0.77 and 0.71 between the striatal-to-
occipital ratio and the striatal uptake expressed in % of the injected activity, and 0.94 and 0.90
between the two resolution-independent methods.
A display of the results measured by the three methods in a healthy volunteer is shown in Figure 4.
96
Table 4. Striatal activity measured in normal subjects and Parkinson patients with the different methods. The
mean specific striatal to non specific uptake is given in the first two columns, the resolution independent
method (left and right striatal activity divided by the mean brain activity/cm3) in the two median columns and
the uptake in % of injected dose, normalised for the patient weight, in the last two columns.
Name V”3 left V”3 right ROI ind left ROI ind right uptake left uptake right
Subject 1 2,56 2,48 95,8 89,3 0,464 0,433
Subject 2 2,31 2,15 75,2 80,8 0,358 0,385
Subject 3 1,84 1,80 77,6 64,5 0,385 0,320
Subject 4 2,88 2,90 94,3 91,2 0,379 0,366
Subject 5 2,04 2,20 85,1 78,3 0,373 0,343
Subject 6 2,03 1,85 90,4 78,1 0,498 0,432
Subject 7 2,24 1,93 74,2 80,3 0,332 0,359
Subject 8 2,33 2,24 74,0 76,2 0,292 0,301
Subject 9 2,43 2,40 108,2 99,9 0,526 0,486
Subject 10 2,58 2,55 78,4 73,4 0,309 0,289
Subject 11 2,13 2,18 81,6 80,4 0,370 0,365
Subject 12 1,68 1,52 86,0 71,8 0,442 0,369
Subject 13 2,31 2,28 75,4 70,1 0,358 0,333
Subject 14 2,06 1,93 79,4 75,8 0,359 0,343
Subject 15 1,96 2,13 70,1 74,2 0,316 0,334
Subject 16 2,80 3,04 76,8 98,8 0,281 0,361
mean 2.26 2.22 82.7 80.2 0.378 0.364
S.D. 0.33 0.39 10.1 9.9 0.071 0.051
range 1.68–2.88 1.52–3.04 70.1–108.2 64.5–99.9 0.281–0.526 0.289–0.486
Patient 1 0,88 1,08 47,8 61,2 0,133 0,170
Patient 2 0,69 0,61 40,9 45,4 0,143 0,159
Patient 3 0,31 1,12 11,2 60,5 0,053 0,290
Patient 4 0,44 0,74 35,4 47,3 0,172 0,230
Patient 5 0,38 0,39 37,8 39,2 0,160 0,166
Patient 6 1,27 1,31 52,9 58,9 0,227 0,252
Patient 7 1,87 1,76 64,7 54,0 0,297 0,248
Patient 8 0,27 0,37 38,5 42,1 0,147 0,161
Patient 9 0,77 0,96 54,2 65,0 0,235 0,282
Patient 10 0,83 0,88 57,5 46,7 0,253 0,205
Patient 11 0,60 0,71 44,0 50,7 0,156 0,180
Patient 12 1,94 2,08 82,3 82,7 0,271 0,273
Patient 13 0,68 0,67 64,8 60,5 0,299 0,279
Patient 14 0,85 1,03 50,6 55,0 0,234 0,254
Patient 15 0,76 1,07 32,2 56,7 0,150 0,264
Patient 16 1,64 1,86 83,0 88,1 0,394 0,419
Patient 17 0,63 0,77 37,7 50,2 0,202 0,269
Patient 18 0,87 1,38 72,2 16,6 0,300 0,069
mean 0.87 1.04 50.4 54.5 0.213 0.232
S.D. 0.50 0.48 18.4 15.8 0.082 0.075
range 0.27–1.94 0.37–2.08 11.2–83.0 16.6–88.1 0.053–0.394 0.069–0.419
97
Figure 3. Regression between the traditional method with manually drawn ROIs and the resolution
independent method at the top. Regression between the traditional method and the striatal uptake in % of
injected dose at the bottom (X-axis = V”3 traditional method)
V'3 - ROI independent left striatum
R = 0.86
0,0
20,0
40,0
60,0
80,0
100,0
120,0
0 1 2 3 4
V'3 - ROI independent right striatum
R = 0.82
0,0
20,0
40,0
60,0
80,0
100,0
120,0
0 1 2 3 4
V'3 - % uptake I.D. left striatum
R = 0.77
0,000
0,100
0,200
0,300
0,400
0,500
0,600
0 1 2 3 4
V'3 - % uptake I.D. right striatum
R = 0.71
0,000
0,100
0,200
0,300
0,400
0,500
0,600
0 1 2 3 4
Discussion
In quantitative SPECT studies of small objects, accurate measurement of the radioactive content is
significantly influenced by the partial volume effect, and the degree of reduction of the measured
compared to true activity is highly dependent upon the resolution of the gamma camera-collimator
system.
When imaging syringes of various diameters by SPECT with different types of collimators, we
found that the underestimation of the measured activity for iodine-123 was lowest with the fan-
beam collimators, and that scatter correction did not significantly modify the influence of partial
volume effect. For this part of the work, we used a 50ml syringe as a reference, assuming that with
its internal diameter of more than twice the SPECT resolution, it should not be significantly
affected by partial volume effect. For lower resolution collimators however, this assumption could
be partially incorrect so that our results might not be completely reproduced using another system.
When imaging the human striatum with 123
I-FP-CIT, its small size (about 1 cm) makes an accurate
quantification of the true activity by SPECT particularly sensitive to partial volume effect and
dependent upon the system and collimator resolution. The highest resolution collimator is therefore
the most appropriate choice. However, even with fan-beam collimators, we estimated the magnitude
of underestimation of activity for the striatum to be about 3 in an anthropomorphic brain phantom.
The influence of partial volume effect probably at least partly explains the fluctuations in count
density observed at different anatomical levels of the striatum in the same individual, as the real
diameter of a human striatum decreases from 12 mm at the head of the caudate nucleus to 6 mm at
the tail of the putamen, and also the differences in normal striatal-to-occipital values between
98
SPECT and PET (1.9 to 3.8 for 123
I-FP-CIT (17-19) versus >7.8 for the PET dopamine transporter
ligand 18F-FECNT (23)).
Since the importance of partial volume effect depends on the system resolution, multi-centric
studies with FP-CIT using quantitative or even semi-quantitative indices are difficult to carry out,
and reference normal values from one centre cannot be used in another without caution.
To circumvent the influence of partial volume on the measurement of activity in small organs,
Fleming et al have recently shown that the determination of the total uptake in the whole organ with
large regions of interest instead of small ROIs drawn around the target alone constitutes an
interesting approach (20).
The method presented in this paper uses two very large regions of interest. Knowing that the
volume of the human striatum is about 20 cm3 and the real concentration of dopaminergic
radioligands 8 to 9 times higher in the striatum than in the remaining brain tissue, the total striatal to
the mean brain activity/cm3 values between 150 and 200 obtained in our normal subjects seem
consistent.
This method offers several advantages compared to the classical striatal-to-occipital ratio usually
found in the literature. Being significantly larger than the SPECT resolution, the ROIs are not
subjected to partial volume effect so that the striatal-to-brain ratio is gamma-camera, acquisition
and reconstruction independent, allowing a direct comparison between results from different
centres, without the need for any special soft- or hardware adjustment.
Thanks to their large size, the placement of the ROIs becomes less critical, resulting in an important
reduction of the inter-observer variability compared to the traditional striatal-to-occipital method
while keeping its performance in separating normal subjects from Parkinson’s patients.
Using adequate calibration factors, the total count rate can be expressed in MBq, hence in
percentage of the injected activity. In our study, the striatal uptake expressed in percentage of the
injected activity was well correlated with both the classical striatal-to-occipital ratio and the
resolution-independent striatal-to-brain method, but slightly less accurate in differentiating normal
subjects from Parkinson’s patients. Normalisation for the body weight allowed a better separation
between both groups.
However, besides calibration errors, physiological differences between patients might interfere in
the calculation of the absolute uptake, and this absolute quantification is more complex than the
intermediate results of the resolution-independent method presented here, which could be easily
implemented in most centres.
In the anthromorphic striatal phantom, we were able to measure the activity with an error of less
than 10 % compared to the true value, and even smaller after scatter correction.
In the patients studies however, data were not corrected for scatter or septal penetration. Applying
these corrections might change the ratio between the striatum and brain background. This needs
further investigation.
Conclusion
In small objects, partial volume results in a dramatic underestimation of the true radioactive content,
making an accurate quantification of the activity concentration a little hazardous. In a quantitative
anthropomorphic phantom study, the magnitude of partial volume effect was of the order of 3 for
FP-CIT striatal uptake, and was collimator and processing-dependent. The classically used striatal-
to-occipital ratio can therefore hardly be applied for multi-centric studies or even to define widely
usable normalcy rates.
Our resolution-independent method uses very large regions of interest and is thus not influenced by
the partial volume effect. Well correlated with the classical ratio but providing a better inter-
observer variability, and easy to implement in a nuclear medicine department, it offers a
reproducible, reliable and simple way to separate the patients with a normal or pathological FP-CIT
scan, and should be helpful to compare quantitatively the results between different imaging systems
and even different centres.
99
Figure 4. Normal subject: At the top left a density image of the striatum is expressed in % of the
injected dose per ml with an appropriate colour scale. At the bottom the traditional (striatum –
occipital) / occipital image is represented. Values for striatal activity divided by the mean brain
activity/cm3, % of the injected dose and traditional ratio’s are shown.
Acknowledgments
The authors would like to express their gratitude to Amersham-Health Belgium (part of GE
Healthcare) for providing the anthropomorphic striatal phantom, and to Philippe Delsarte and Rudi
Vandermeiren for the technical assistance.
The authors have indicated they have no conflict of interest.
References
1. Hornykiewicz O. Dopamine and brain function. Pharmacol Rev. 1966; 925-964.
2. Bernheimer H, Birkmayer W, Hornykiewicz O, Jellinger K, Seitelberger F. Brain dopamine
and the syndromes of Parkinson and Huntington. J. Neurol Sci. 1973; 20: 415-455.
3. Marsden CD. Parkinson’s disease. Lancet. 1990;335:948-952.
100
4. Neumeyer JL, Wang S, Gao Y, Milius RA, Kula NS, Campbell A, Baldessarini RJ, Zea-
Ponce Y, Baldwin RM, Innis RB. N-ω-fluoroalkyl analogs of (IR)-2β-carbomethoxy-3β-(4-
iodophenyl)-tropane (β-CIT): Radiotracers for Positron Emission Tomography and Single
Photon Emmision Computed Tomography Imaging of Dopamine Transporters. J Med Chem
1994; 37:1558-1561.
5. Lorberboym M, Djaldetti R, Melamed E, et al. 123I-FP-CIT SPECT imaging of dopamine
transporters in patients with cerebrovascular disease and clinical diagnosis of vascular
parkinsonism. J Nucl Med. 2004;45:1688-93.
6. Schwartz M, Groshar D, Inzelberg R, Hocherman S. Dopamine-transporter imaging and
visuo-motor testing in essential tremor, practical possibilities for detection of early stage
Parkinson's disease. Parkinsonism Relat Disord. 2004;10:385-9.
7. Ceravolo R, Volterrani D, Gambaccini G, et al. Presynaptic nigro-striatal function in a group
of Alzheimer's disease patients with parkinsonism: evidence from a dopamine transporter
imaging study. J Neural Transm. 2004;111:1065-73.
8. Catafau AM, Tolosa E. Impact of dopamine transporter SPECT using 123I-Ioflupane on
diagnosis and management of patients with clinically uncertain Parkinsonian syndromes.
Mov Disord. 2004;19:1175-82.
9. Walker Z, Costa DC, Walker RW et al. Striatal dopamine transporter in dementia with Lewy
bodies and Parkinson disease: a comparison. Neurology. 2004;62:1568-72.
10. Winogrodzka A, Bergmans P, Booij J et al. [123I]FP-CIT SPECT is a useful method to
monitor the rate of dopaminergic degeneration in early-stage Parkinson's disease. J Neural
Transm. 2001;108:1011-9.
11. Booij J, Speelman JD, Horstink MW, Wolters EC. The clinical benefit of imaging striatal
dopamine transporters with [123I]FP-CIT SPET in differentiating patients with presynaptic
parkinsonism from those with other forms of parkinsonism.
Eur J Nucl Med. 2001;28:266-72.
12. Booij J, Bergmans P, Winogrodzka A, et al. Imaging of dopamine transporters with
[123I]FP-CIT SPECT does not suggest a significant effect of age on the symptomatic
threshold of disease in Parkinson's disease. Synapse. 2001;39:101-8.
13. Benamer HT, Patterson J, Wyper DJ et all. Correlation of Parkinson's disease severity and
duration with 123I-FP-CIT SPECT striatal uptake. Mov Disord. 2000;15:692-8.
14. Benamer TS, Patterson J, Grosset DG, et al. Accurate differentiation of parkinsonism and
essential tremor using visual assessment of [123I]-FP-CIT SPECT imaging: the [123I]-FP-
CIT study group. Mov Disord. 2000;15:503-10.
15. Tissingh G, Booij J, Bergmans P, et al. Iodine-123-N-omega-fluoropropyl-2beta-
carbomethoxy-3beta-(4-iod ophenyl)tropane SPECT in healthy controls and early-stage,
drug-naive Parkinson's disease. J Nucl Med. 1998;39:1143-8.
16. Booij J, Tissingh G, Boer GJ, et al. [123I]FP-CIT SPECT shows a pronounced decline of
striatal dopamine transporter labelling in early and advanced Parkinson's disease. J Neurol
Neurosurg Psychiatry. 1997;62:133-40.
17. Seibyl JP, Marek K, Sheff K, et al. Iodine-123-β-CIT and iodine-123-FPCIT SPECT measurement op dopamine transporters in healthy subjects and Parkinson’s patients. J Nucl
Med 1998; 39:1500-1508.
18. Booij J, Habraken J, Bergmans P et al. Imaging of dopamine transporters with iodine-123-
FPCIT SPECT in healthy controls and patients with Parkinson’s disease. J Nucl Med 1998;
39:1879-1884.
19. Habraken J, Booij J, Slomka P, et al. Quantification and visualization of defects of the
functional dopaminergic system using an automatic algorithm. J Nucl Med 1999; 40:1091-
1097.
20. Fleming J, Bolt L, Stratford J, Kemp P. The specific uptake size index for quantifying
radiopharmaceutical uptake. Phys Med Biol 2004; 49:227-234.
101
21. Ichihara T, Ogawa K, Motomura N, et al. Compton scatter compensation using the triple-
method for single and dual-isotope SPECT. J Nucl Med 1993; 34:2216-2221.
22. Dobbeleir A, Hambye A-S, Franken P. Influence of high-energy photons on the spectrum of
iodine-123 with low- and medium-energy collimators: consequences for imaging with 123
I-
labelled compounds in clinical practice. Eur J Nucl Med 1999; 26:655-658.
23. Davis M, Votaw J, Bremner D et al. Initial human PET imaging studies with the
dopamine transporter ligand 18F-FECNT. J Nucl Med 2003; 44:855-861.
102
6. Summary and future directions.
Part I deals with cardiac studies. In chapter 2, methods aimed at evaluating left ventricular ejection fraction are described.
In the first part, the performance capacities and limitations of a single crystal digital gamma camera are evaluated with respect to the high count rate required for an accurate measurement of ejection fraction by first pass radionuclide angiography. Using the ultrashort half-lived 191Irm, the high yield of the generator (120 mCi - 4400 MBq) provided more than 1 million real counts per second whereas the measured camera saturation was 420 kcps. Compared to a large field of view detector system, a small field of view (20 cm) has the advantage to have less activity in the field of view reducing the non-linearity problem. Using an 191Os reference source, we were able to correct for the non-linearity up to 320kcps. Applied to patient studies, a maximum count rate of 250 kcps was measured during the left ventricular phase, with a system resolution loss of 2-3 mm in fast mode. Repeated LVEF determination at 2 min-intervals in 50 patients was highly reproducible with a mean
difference of 2.08±1.55 EF units (r=0.97). Furthermore, the simultaneous use of 191Irm and of 201Tl permitted a combined evaluation of myocardial perfusion and function both at rest and during exercise.
In the second part, the performances of different software for left ventricular ejection fraction (LVEF) and volume measurements by gated myocardial tomography are studied, and the influence of modifying acquisition and reconstruction parameters are evaluated, especially in patients with small hearts. In patients with a normal-sized heart, the different commercially available software for quantitative gated SPET was well correlated. Changes in matrix size had little influence on LVEF and volumes whereas smoothing significantly modified the volume measurements. In small-sized hearts on the other hand, LVEF at the higher range were frequently observed. The results of quantitative gated SPET were software, matrix size and smoothing dependent. Probably more realistic, significantly lower LVEF and larger volumes were found by increasing the matrix size or sharpening the filter.
Chapter 3 deals with the quantification of perfusion and metabolism in the specific context of myocardial viability assessment.
The first part reports the development and clinical validation of a quantification whereby the activity of the myocardial perfusion tracer 99mTc -sestamibi and the free fatty acid metabolism tracer 123I-BMIPP was quantified on a pixel-to-pixel basis. Based upon the difference in uptake between sestamibi and BMIPP, the presence and extent of normal, viable and scar myocardium was expressed in % of the surface of the left ventricle as a whole and of the three main coronary arteries separately and visually displayed using colour-coded polar maps. This analysis was applied to patient studies. Inter-observer difference in the % viable myocardial surface was rather small, amounting to 1.5% at most. Moreover, a good concordance was found between the presence of decreased sestamibi and BMIPP uptake and a significant stenosis on coronary angiogram. In the second part the influence of high-energy emitting photons on the spectrum of iodine-123 was quantified for low- and medium-energy collimators in phantom studies.
103
In the third part this scatter influence was shown in patient studies. The newly developed quantification with colour-coded polar maps was applied to calculate the extent of viable tissue, defined as a mismatched uptake with BMIPP uptake lower than sestamibi. Since the contribution of scatter in the iodine images is not negligible, its potential influence on the calculated amount of viable tissue was measured by quantifying sestamibi and BMIPP uptake without correction, after background subtraction, and with scatter correction. Echocardiographically assessed changes in segmental wall motion at six months after treatment was used as the gold standard. The evolution of contractile function was correctly predicted in 64% of the segments without correction, 79% after background subtraction and 93% after scatter correction. The fourth part contains the abstracts of the articles that we have published about the clinical applications of myocardial perfusion/metabolism imaging in patients with chronic ischemic heart disease post-infarction. From these clinical studies, it seems that 99mTc -sestamibi alone is a suboptimal tracer to identify myocardial viability in patients with chronic ischemic heart disease post-infarction, even when a quantitative analysis is applied. Adding a metabolic tracer such as 123I-BMIPP significantly improves the diagnostic accuracy, and the combination of sestamibi and BMIPP imaging is able to identify myocardial viability in chronic ischemic heart disease with an accuracy similar to that reported in the literature for 18F-FDG PET, or for the combined BMIPP/sestamibi study in the acute or subacute phase of a myocardial infarction. However, due to the influence of high energy photons in the iodine-123 imaging, scatter correction is recommended and special attention must be paid to the used collimator.
Part 2 deals with brain studies. In Chapter 4, an absolute quantification method of the brain perfusion is described. Methodologically, because of the clinical need for an absolute SPECT parameter, a simple approach was developed using calibrated point sources as scaling factor, to display tomographic images as regional 99mTc HMPAO brain uptake (rBU) per cm3 brain tissue in percent of the injected lipophilic dose. The method was validated on Jaszczak and Hoffman phantoms using a three detector SPECT system with parallel and fan-beam collimators. A mean reproducibility of 7.2 % was obtained in human studies. Application of the method in 33 healthy volunteers pointed to body surface as the most important factor explaining interindividual variability when compared to brain volume. The same study stressed the need in longitudinal studies for normalization of rBU to the rest heart rate and suggested the absence of significant influence of minor stress on regional 99mTc HMPAO brain uptake. The former evaluation of cerebellar rBU in a healthy population was extended to patients suffering from dementia of the Alzheimer type (DAT). rBU values in operator-defined cerebellar regions of interest may be considered highly symmetrical, reproducible and stable in time in healthy volunteers. Moreover, after cumulative corrections for body surface and brain volume a similar and reproducible, absolute cerebellar 99mTc HMPAO uptake value was found for the group of DAT patients and the group of healthy volunteers. The presented findings suggested that the cerebellum may be a good choice as reference region in SPECT analysis of DAT patients.
104
In Chapter 5, dopamine transporters in the striatum are quantified. Parkinson’s disease is characterised by a severe degeneration of dopaminergic neurons in the substantia nigra, resulting in a loss of dopamine transporters in the caudate and putamen nuclei visualized with 123I-FP-CIT. Using SPECT, it is well know that semi-quantitative analysis of small organs like the striatum is hampered by the partial volume effect due to the low spatial resolution of the gamma camera. For source diameters < 12 mm, the count density was reduced by a factor 3 to 5. We devised a resolution independent method by calculating the total striatum activity divided by the mean brain activity per ml using two very large region of interests. Knowing that the volume of the human striatum is about 20 ml and the real concentration of dopaminergic system radioligands is 8 to 9 times higher in the striatum than in the remaining brain tissue, the total striatal to the mean brain activity/cm3 values between 150 and 200 obtained in our normal subjects seem consistent. Moreover using classical striatal and occipital ROIs, we obtained an inter-observer variability of 11.4 % compared to 3.1% using the resolution independent method. Additionally the total striatal uptake was expressed as percentage of injected dose using a gamma camera calibration factor. Globally, good separation was obtained between normal and Parkinsons using both, the conventional and our method. When corrected for the patient’s weight, striatal uptake expressed in percentage of the injected dose allowed a better separation between normal subjects and Parkinson’s patients compared to the conventional method. For more accurate anatomic localization of defects, we created two images for visual interpretation: a striatum/brain ratio image and an uptake image expressing the % of the injected dose per ml striatal tissue. The resolution independent method is gamma-camera, acquisition and reconstruction independent. Using this method, results from different centres can directly be compared without the need of any special soft- or hardware adjustment.
Future directions. The current trends in conventional nuclear instrumentation and data analysis can be divided in three main topics: image fusion, software development and new devices. The fusion of modalities becomes standard in clinical practice. Hybrid imaging systems SPECT equipment with X-ray tubes (CT) are on the market. Initial attempts to co-register functional and anatomical images acquired on two different machines failed to disclose the proper alignment and are too cumbersome on a routine basis (1). SPECT/CT improves the diagnostic accuracy of SPECT in various clinical situations (2) although misalignment artefacts between emission and transmission can still be present, especially in cardiac studies (3). A debate is still going on pro and contra attenuation correction for cardiac studies (4). However a CT based attenuation map improves semi-quantitative cardiac studies (5,6) and precise absolute quantification becomes realistic when including also scatter and collimator depth corrections in the iterative reconstruction methods (7,8). Attempts are also made to correct partial volume effect in small textures by anatomical information (9). Improvement of reconstruction algorithm’s remained a major topic on the IEEE meeting in Rome 2004. New software starts entering clinical practice. Three dimensional models of coronary artery tree created by biplane angiograms are aligned with 3D perfusion SPECT images (10). Motion-frozen gated images can be created using phase-to-phase motion vectors (11). Statistic parametric mapping (SPM), frequently used for brain research becomes a standard procedure (12).
105
The major improvement in nuclear medicine is expected from new devices and material. The goal is to reduce the intrinsic spatial resolution and energy resolution (13,14). New detection materials with better physical characteristics than NaI concerning stopping power, energy resolution, light output, fragility and density will most likely replace NaI crystals (15). Position sensitive photo multiplier tubes (PSPMT) has become available which are coupled to pixellated NaI(Tl) crystals or new scintillation material like CsI(Tl) (16,17). It is expected that new imaging devices with several thousands of tiny crystals or semiconductor array detectors will improve the sensitivity and specificity of clinical studies (15,18). Increased detector sensitivity will permit dynamic tomographic studies and more precise quantitative data for compartimental analysis already performed for planar studies. Small surgical probes based on CZT semiconductors or PSPMT tubes become popular in surgery tracing regional metastases (19). It remains however questionable whether these small devices, now a day used for animal studies or as surgical probes, can be developed as large detectors for human studies at a reasonable price.
1. Keidar Z, Isreal O, Krausz Y. SPECT/CT in tumor imaging: technical aspects and clinical applications. Semin Nucl Med 2003; 33:205-18.
2. Schillaci O, Danieli R, Manni C, Simonetti G. Is SPECT/CT with a hybrid camera useful to improve scintigraphic imaging interpretation? Nucl Med Commun. 2004;25:705-10.
3. Fricke H, Fricke E, Weise R et al. A method to remove artifacts in attenuation-corrected myocardial perfusion SPECT Introduced by misalignment between emission scan and CT-derived attenuation maps. J Nucl Med. 2004;45:1619-25.
4. Figaro E, Wackers F. Should SPET attenuation correction be more widely employed in routine clinical practice? Eur J Nucl Med 2002; 29: 409-415.
5. Grossman G, Garcia E, Bateman T et al. Quantitative Tc99m sestamibi attenuation-corrected SPECT development and multicenter trial validation of myocardial perfusion stress gender-independent normal database in an obese population. J Nucl Cardiol 2004; 11: 239-241.
6. Dondi M, Fagioli G, Salgarello M et al. Myocardial SPECT: what do we gain from attenuation correction (and when)? Q J Nucl Med Mol Imaging 2004; 48:181-7.
7. El Fakhri G, Buvat I, Benali H et al. Relative impact of scatter, collimator response, attenuation and finite spatial resolution corrections in cardiac SPECT. J Nucl Med 2000; 41: 1400-8.
8. Links J, Becker L, Rigo P et al. Combined corrections for attenuation, depth dependent blur and motion in cardiac SPECT: a multicenter trial. J Nucl Cardiol 2000; 7: 414-25.
9. Matsuda H, Ohnishi T, Asada T et al. Correction for partial volume effects on brain perfusion SPECT in healthy men. J Nucl Med 2003; 44: 1243-52.
10. Faber T, Santana C, Garcia E et al. Three dimensional fusion of coronary arteries with myocardial perfusion distributions: clinical validation. J Nucl Med 2004; 45: 745-53.
11. Slomka P, Nishina H, Berman D et al. “Motion-Frozen” display and quantification of myocardial perfusion. J Nucl Med 2004; 45: 1128-34.
12. Friston K, Ashburner J, Holmes A and Poline J-B. SPM: Statistical parametric mapping, software for functional neuroimaging. Welcome department of Cognitive Neurology, University College London.
13. Williams M, Goode A, Galbis-Reig V et al. Performance of a PSPMT based detector for scintimammography. Phys Med Biol 2000; 45: 781-800.
14. Loudos G, Nikita K, Uzunoglu N et al. Improving spatial resolution in SPECT with the combination of PSPMT based detector and iterative reconstruction algoritms. Comput Med Imaging Graph 2003; 27: 307-13.
106
15. Fidler V. Current trends in nuclear instrumentation in diagnostic nuclear medicine. Radiol Oncol 2000; 34: 381-5.
16. Weisenberger A, Kross B, Majewski S et al. Dual low profile detector heads for a restraint free small animal SPECT imaging system. IEEE conference Rome 2004: p136.
17. Pani R, Pellegrini R, Cinti M et al. New devices for imaging in nuclear medicine. Cancer Biother Radiopharm. 2004;19:121-8.
18. Wieczorek H, Goedicke A, Shao L et al. Analytical model for pixellated SPECT detector concepts. IEEE conference Rome 2004: p142.
19. Blevis L, Reznik A. Intra-operative imaging probe using CZT. IEEE conference Rome 2004: p197.
107
Samenvatting en toekomstperspectieven.
Deel 1 betreft hartstudies. In hoofdstuk 2 worden methoden ter bepaling van de linker ventriculaire ejectiefraktie beschreven.
In het eerste deel, wordt de prestatie en beperkingen van een kleinveld digitale gamma camera met betrekking tot hoge telcapaciteit voor accurate bepaling van de linker ventriculaire functie bij de eerste doorstroming voor radio-nucleaire angiocardiography beschreven. Het kort levend 191Irm (5sec) van een nieuwe hoge opbrengst 191Os/191Irm generator leverde meer dan 1 miljoen werkelijke slagen per seconde daar waar de gemeten verzadiging van de gamma camera 420kcps was. De kleine gezichtsveld (20 cm) gamma camera heeft het voordeel ten opzichte van een grootveld dat er minder activity gemeten wordt en dus ook het activiteits niet-lineaire probleem vermindert. Op een nauwkeurige manier verbeterden wij deze niet-lineaire respons van de gamma camera tot 320kcps door middel van een 191Os referentie bron. In patient studies was de maximum telsnelheid gedurende de linker ventriculaire faze 250 kcps. Het resolutieverlies van het systeem in snelle telmode was 2-3 mm. De reproduceerbaarheid van herhaalde LVEF bepalingen in 2 min-intervallen in 50 patienten
was r=0.97 and het gemiddelde verschil=2.08±1.55 EF eenheden. Verder, maakte het gebruik van 191Irm als merkstof voor linker ventriculaire angiography en 201Tl voor myocard perfusie en wandbeweging gelijktijdige bepalingen mogelijk, zowel in rust als gedurende inspanning. In het tweede deel wordt de prestatie van verschillende software programma’s bestudeerd en tevens de invloed van opnamemethode en reconstructieparameters op de ejectiefractie en hartvolumes, vooral bij patienten met een klein hart. In patienten met een hart van normaal volume, LVEF en volume, bepaald door middel van verschillende commerciële software voor kwantitatieve gated SPECT zijn vergelijkbaar. LVEF en volume zijn weinig gevoelig aan wijzigingen in de opnamematrix. Door smoothing (afvlakken) werd het volume aanmerkelijk gewijzigd maar niet de LVEF waarde. Bij kleine harten daarentegen, beïnvloeden zowel het gebruikte programma, de opname matrix als smoothing in belangrijke mate de resultaten van kwantitatieve gated SPECT. Hoge ejectiefraktie waarden worden dikwijls waargenomen. Een grotere matrix en scherpere reconstructie filter worden gesuggereerd om de accuraatheid van de commerciële software te verbeteren, voornamelijk bij patiënten met een klein hart. Hoofdstuk 3 handelt over de kwantificatie van perfusie en metabolisme in het specifieke kader van leefbaar hartweefsel. In het eerste deel beschreven we de ontwikkeling en klinische validering van een methode om de aktiveit van de perfusietracer 99mTc-sestamibi en de vetzuur metabolisme tracer 123I-BMIPP pixel per pixel te kwantificeren, aan de hand van kleur-gecodeerde polaire voorstellingen. Met als basis het opnameverschil tussen sestamibi en BMIPP, wordt de aanwezigheid en uitgebreidheid van normaal, leefbaar en littekenweefsel uitgedrukt in % van het linker ventrikel als geheel en voor de vaatgebieden van de 3 hoofdkransslagaders afzonderlijk berekend en in beeld gebracht. Deze analyse werd toegepast op patiëntengegevens. De inter-observer verschillen in het percentage mismatched oppervlak waren eerder klein, ten hoogste 1.5%. Bij vergelijking
108
met de coronaire anatomie werd een goede overeenkomst vastgesteld tussen een verminderde traceropname en een significante vernauwing van de arterie.
In het tweede deel van dit hoofdstuk wordt de invloed van hoog energetische fotonen afkomstig van 123I op het gemeten spectrum gekwantificeerd met collimatoren voor lage energie en middelhoge energie in een fantoom studie.
In het derde deel wordt deze scatter invloed aangetoond bij patiënten. Voor deze studie maakten we gebruik van de nieuwe polaire kleur-gecodeerde voorstelling om de uitgebreidheid van leefbaar myocard te kwantificeren, gedefinieerd als mismatching met BMIPP opname lager dan sestamibi. Gezien de bijdrage van scatter in de jodium beelden en de mogelijke invloed op de berekende hoeveelheid leefbaar weefsel, werd de sestamibi en BMIPP opname berekend zonder correktie, met background en met scatter correktie. Echocardiografische veranderingen in wandbeweging 6 maanden na behandeling werd gebruikt als referentie. De accuraatheid om de evolutie van de regionale contractiliteit bij vervolgonderzoek te voorspellen, was als volgt: 64% indien geen correctie werd toegepast, 79% na correctie voor background en 93% na scattercorrectie.
Het vierde deel bevat de abstracten van de artikelen gepubliceerd betreffende klinische toepassingen van de gecombineerde studie van perfusie en metabolisme bij patiënten met chronische linker ventrikeldysfunctie na infarct. Uit deze studies blijkt dat de voorspellende waarde van scintigrafie met sestamibi alleen sub-optimaal is om leefbaar weefsel op te sporen, zelfs bij kwantitatieve analyse. Toevoeging van de resultaten van de 123I-BMIPP scintigrafie verbeterde de accuraatheid aanzienlijk en de combinatie van sestamibi en BMIPP beeldvorming gaf de mogelijkheid om leefbaar weefsel bij patiënten met chronische linker ventrikeldysfunctie op te sporen met dezelfde accuraatheid als met 18F-FDG vermeld in de literatuur. Tevens werden identieke resultaten bekomen als bij patiënten in de acute of subacute fase na een hartinfarct met dezelfde gecombineerde sestamibi/BMIPP methode. Wegens de invloed van hoog energetische fotonen in beelden van een jodium-123 gemerkt produkt zoals BMIPP is scatter correctie aanbevolen en dient aandacht besteed te worden aan de gebruikte collimator.
Deel 2 behandelt hersenstudies In hoofdstuk 4 wordt een absolute kwantifikatiemethode voor hersenperfusie beschreven. Methodologisch werd wegens de klinische vraag naar een absolute SPECT parameter, een eenvoudige benadering ontwikkeld, gebruikmakend van gekalibreerde puntbronnen als schalingsfactor, om tomografische beelden voor te stellen als regionale 99mTc HMPAO hersenopname per cm3 hersenweefsel in percent van de geïnjecteerde lipofiele dosis. De methode werd gevalideerd op Jaszczak en Hoffman fantomen, gebruikmakend van een 3-detector SPECT systeem met parallelle en fan-beam collimators. Een gemiddelde reproduceerbaarheid van 7.2% werd bekomen in een referentiepopulatie. Toepassing bij 33 gezonde vrijwilligers wees op lichaamsoppervlakte als belangrijkste factor in vergelijking met hersenvolume ter verklaring van de interindividuele variabiliteit. Dezelfde studie benadrukte tevens de noodzaak in longitudinale studies een normalisatie uit te voeren voor het hartritme in rust en suggereerde de afwezigheid van een significante invloed van lichte stress op de regionale 99mTc HMPAO hersenopname. De voorgaande evaluatie van cerebellaire rBU in een gezonde populatie werd uitgebreid tot patiënten lijdend aan demensie van het Alzheimer type (DAT). rBU waarden van cerebellaire operator-bepaalde regios van interesse kunnen worden beschouwd als symmetrisch, reproduceerbaar en stabiel in de tijd in gezonde vrijwillers. Bovendien werd
109
na cumulatieve correcties voor lichaamsoppervlakte en hersenvolume een vergelijkbare en reproduceerbare, absoluut cerebellaire 99mTc HMPAO opname gevonden in de groep DAT patiënten en de groep gezonde vrijwilligers. De voorgestelde bevindingen suggereerden dat het cerebellum een goede keuze is als referentie regio voor de SPECT analyse van patiënten lijdend aan dementie van het Alzheimer type. Hoofdstuk 5: dopamine transporters in het striatum worden quantitatief bepaald. Parkinson’s ziekte wordt gekenmerkt door een hevige degeneratie van dopaminegevoelige neuronen in de grijze stof (dopaminergic neurons in the substantia nigra), met gevolg een verlies van dopamine transporters in nucleus caudata en putamen, in beeld gebracht met 123I-FP-CIT. Met SPECT, is het goed gekend dat semi-kwantitative analyse van kleine organen zoals het striatum belemmerd wordt door het partiëel volume effekt veroorzaakt door de slechte resolutie van de gamma camera. Voor bronnen met diameters < 12 mm werd een aktiviteit met een faktor 3 tot 5 te laag gemeten. Wij hebben een resolutie onafhankelijke methode ontwikkeld die erin bestaat de totale activiteit van het striatum te bepalen en te delen door de gemiddelde hersenaktiviteit, gebruik makend van twee grote interessezones. Wetende dat het volume van het menselijk striatum ongeveer 20 ml is en de werkelijke concentratie van de radioactieve tracer 8 tot 9 keer hoger is in het striatum dan in de rest van het hersenweefsel, is de gemeten verhouding van totale striatale op gemiddelde hersenactiviteit per ml van 150 tot 200 hiermee in overeenstemming. Met de klassieke striatale and occipitale ROIs, bekwamen we een inter-observator variabiliteit van 11.4 % in vergelijking met 3.1% met de resolutie onafhankelijke methode. Aanvullend hebben wij de totale striatale opname uitgedrukt in procent van de ingespoten dosis, gebruik makend van een gamma camera ijkingsfactor. Een goede scheiding werd bekomen tussen normale en Parkinson patiënten met de conventionele en onze methode. Met de methode waarbij de opname in het striatum uitgedrukt werd in procent van de dosis en na correctie door het gewicht van de patiënt werd een betere scheiding waargenomen dan met de conventionele methode. Voor meer accurate anatomische lokalisatie van defecten hebben wij twee bijkomende beelden ontworpen voor visuele interpretatie : een striatum/brain ratio beeld een een opnamebeeld uitgedrukt in % van de ingespoten dosis per ml striataal weefsel. De resolutie onafhankelijke methode is niet gamma camera, beeldopname of reconstructie afhankelijk. Met deze methode kunnen dan ook resultaten van verschillende centra onmiddelijk met elkaar vergeleken worden zonder soft of hardware aanpassingen.
Toekomstsperspectieven.
De huidige stroming in conventionele nucleaire instrumentatie en data analyse kan opgesplitst worden in drie belangrijke richtingen: beeldfusie, software ontwikkeling en nieuwe toestellen. De fusie van verschillende beeldvormende technieken wordt de norm in de klinische praktijk. Hybride SPECT camera’s uitgerust met X-straalbuizen (CT) worden verkocht. Initiële pogingen om functionele en anatomische beelden van twee verschillende toestellen te co-registreren onthullen de moeilijkheid om de strukturen in overeenstemming te brengen en zijn te arbeidsintensief voor routinematig gebruik (1). SPECT/CT verhoogt de diagnostieke accuraatheid van SPECT bij verschillende klinische onderzoeken (2). Nietemin blijven co-registratie artefacten tussen emmissie en transmissie beelden mogelijk, vooral bij hartstudies (3). Er is tevens nog discussie pro en contra attenuatiecorrectie bij hartstudies. (4) Daartegenover staat dat semi-quantitative gegevens verbeterd worden door middel van een CT attenuatie map (5,6) en preciese absolute
110
kwantificatie wordt realistisch wanneer tevens scatter en collimator afhankelijke diepte correcties toegevoegd wordt aan de iteratieve reconstructie methode (7,8). Pogingen worden tevens ondernomen om het partiëel volume effect bij kleine strukturen te corrigeren met behulp van anatomische informatie (9). Verbetering van reconstructiealgoritmes was nog steeds een belangrijk onderwerp op de IEEE meeting in Rome 2004. Nieuwe software gaat deel uitmaken van de kliniek. Drie-dimentionele modellen van de kransslagaders verkregen door twee-dimentionele angiografie en 3D perfusie SPECT beelden worden op elkaar gepast (10). Bewegings-bevroren gated beelden kunnen verkregen worden met behulp van fase tot fase bewegingsvectoren (11). Statistische parametrische mapping (SPM), frequent gebruikt in hersenonderzoek wordt een standaard procedure (12). De belangrijkste vooruitgang in nucleaire geneeskunde dient te komen van nieuwe toestellen en materiaal. Het doel is de intrinsieke ruimtelijke resolutie en energieresolutie te verbeteren (13,14). Nieuw detectormateriaal met betere fysische karakteristieken dan NaI, wat betreft stralingsabsorptie, energieresolutie, licht opbrengst, breekbaarheid en densiteit, zal hoogstwaarschijnlijk NaI kristallen gaan vervangen (15). Positie gevoelige fotomultiplicatoren (PSPMT) worden gebruikt samen met pixelgrootte NaI(Tl) kristalen of nieuw scintillatie materiaal zoals CsI(Tl) (16,17). Men rekent erop dat nieuwe detectoren, bestaande uit duizenden kleine kristallen of een semi-conductoren matrix, de sensitiviteit and specificiteit van klinische studies zal verhogen (15,18). Verhoogde detector sensitiviteit maakt ook dynamische tomografische studies mogelijk en genereert meer nauwkeurige kwantitatieve gegevens als input voor compartimentele analyse reeds uitgevoerd met planaire beeldopnames. Kleine heelkundige sondes met CZT semi-conductoren of PSPMT worden gebruikt tijdens de operatie om lokale metastases op te sporen (19). Het blijft echter twijfelachtig, of deze kleine toestellen gebruikt voor dierproeven of als heelkundige sondes kunnen ontwikkeld worden tot grote detectoren, voor klinisch gebruik, tegen een aanvaardbare prijs.
1. Keidar Z, Isreal O, Krausz Y. SPECT/CT in tumor imaging: technical aspects and clinical applications. Semin Nucl Med 2003; 33:205-18.
2. Schillaci O, Danieli R, Manni C, Simonetti G. Is SPECT/CT with a hybrid camera useful to improve scintigraphic imaging interpretation? Nucl Med Commun. 2004;25:705-10.
3. Fricke H, Fricke E, Weise R et al. A method to remove artifacts in attenuation-corrected myocardial perfusion SPECT Introduced by misalignment between emission scan and CT-derived attenuation maps. J Nucl Med. 2004;45:1619-25.
4. Figaro E, Wackers F. Should SPET attenuation correction be more widely employed in routine clinical practice? Eur J Nucl Med 2002; 29: 409-415.
5. Grossman G, Garcia E, Bateman T et al. Quantitative Tc99m sestamibi attenuation-corrected SPECT development and multicenter trial validation of myocardial perfusion stress gender-independent normal database in an obese population. J Nucl Cardiol 2004; 11: 239-241.
6. Dondi M, Fagioli G, Salgarello M et al. Myocardial SPECT: what do we gain from attenuation correction (and when)? Q J Nucl Med Mol Imaging 2004; 48:181-7.
7. El Fakhri G, Buvat I, Benali H et al. Relative impact of scatter, collimator response, attenuation and finite spatial resolution corrections in cardiac SPECT. J Nucl Med 2000; 41: 1400-8.
8. Links J, Becker L, Rigo P et al. Combined corrections for attenuation, depth dependent blur and motion in cardiac SPECT: a multicenter trial. J Nucl Cardiol 2000; 7: 414-25.
9. Matsuda H, Ohnishi T, Asada T et al. Correction for partial volume effects on brain perfusion SPECT in healthy men. J Nucl Med 2003; 44: 1243-52.
111
10. Faber T, Santana C, Garcia E et al. Three dimensional fusion of coronary arteries with myocardial perfusion distributions: clinical validation. J Nucl Med 2004; 45: 745-53.
11. Slomka P, Nishina H, Berman D et al. “Motion-Frozen” display and quantification of myocardial perfusion. J Nucl Med 2004; 45: 1128-34.
12. Friston K, Ashburner J, Holmes A and Poline J-B. SPM: Statistical parametric mapping, software for functional neuroimaging. Welcome department of Cognitive Neurology, University College London.
13. Williams M, Goode A, Galbis-Reig V et al. Performance of a PSPMT based detector for scintimammography. Phys Med Biol 2000; 45: 781-800.
14. Loudos G, Nikita K, Uzunoglu N et al. Improving spatial resolution in SPECT with the combination of PSPMT based detector and iterative reconstruction algoritms. Comput Med Imaging Graph 2003; 27: 307-13.
15. Fidler V. Current trends in nuclear instrumentation in diagnostic nuclear medicine. Radiol Oncol 2000; 34: 381-5.
16. Weisenberger A, Kross B, Majewski S et al. Dual low profile detector heads for a restraint free small animal SPECT imaging system. IEEE conference Rome 2004: p136.
17. Pani R, Pellegrini R, Cinti M et al. New devices for imaging in nuclear medicine. Cancer Biother Radiopharm. 2004;19:121-8.
18. Wieczorek H, Goedicke A, Shao L et al. Analytical model for pixellated SPECT detector concepts. IEEE conference Rome 2004: p142.
19. Blevis L, Reznik A. Intra-operative imaging probe using CZT. IEEE conferencence Rome 2004:p197.
112
7. List of publications.
1)Thyroid Research, 590-593 (December 1976)
Early thyroidal iodide and pertechnetate kinetics: new approach with a scintillation camera and a
computer.
Decostre P, Brooke P, Dobbeleir A, Erbsmann F.
2) European Journal of Nuclear medicine 2,173-177 (1977).
Measurement of separate kidney clearance by means of 99m-Tc-DTPA complex and a
scintillation camera.
A.Piepsz, A.Dobbeleir, F.Erbsmann.
3) Journal of Nuclear medicine (1977) Volume 18, Number 10.
Comments on Tc-99m DTPA scintillation camera renography.
A.Piepsz, H.R.Ham, A.Dobbeleir, F.Erbsmann.
4) Journal of Pediatrics. 1978 ; 5 : 769-74.
A simple method for measuring separate glomerular filtration rate using a single injection
of Tc-DTPA and the scintillation camera.
A.Piepsz, R.Denis, H.R.Ham, A.Dobbeleir, C.Schulman, F.Erbsmann.
5) Journal of Nuclear medicine (1981),Volume 22,Number 8.
Radionuclide quantitation of left-to-right cardiac shunts using deconvolution analysis.
H.R.Ham, A.Dobbeleir, P.Viart, A.Piepsz, A.Lenaers.
6) Radionuclides in Nephrology, 269-273, Grune and Stratton London 1982.
How to exclude renal obstruction in children ? Comparison of intrarenal transit times,
cortical times and the furosemide test.
A.Piepsz,H.R.Ham,A.Dobbeleir,M.Hall,F.Collier.
7) Nuclear medicine Communications 4 (1983); 276-281.
Background determination for 99m-Tc DTPA renal studies. A comparison between interpolative
background subtraction and surface ratio method.
A.Piepsz, A.Dobbeleir , H.R.Ham.
8) Nuclear medicine Communications 6 (1985); 477-483.
Influence of statistical noise on the determination of the single kidney 99m-Tc DTPA clearance.
A.Piepsz, A.Dobbeleir, H.R.Ham.
9) European Journal of Nuclear medicine. (1985) 11 : 17-21.
Evaluation of methods for qualitative and quantitative assessment oesophagal transit of liquid.
H.R.Ham, B.Georges, M.Guillaume, F.Erbsmann, A.Dobbeleir.
10) Nucl. Med. Comm. 8 (1987) 365-373 .
Measurement of right ventricular volumes from ECG gated steady state krypton-81m
angiocardiography.
P.R. Franken, P. Mols, E. Delcourt, A. Dobbeleir, B. Georges, H.R. Ham .
113
11) Nuclear Medicine Communications 9, 603-612 (1988).
Assessment of vertebral mineral content by means of a single crystal scintillation camera.
J. Vandevivere, A. Dobbeleir, H.R. Ham, L. Williame.
12) J. Nuclear Medicine. Volume 30, 6 1989. (1025-1031)
Clinical Usefulness of Ultrashort-lived Iridium-191m from a Carbon-Based Generator System for
the Evaluation of the Left Ventricular Function.
P.R. Franken, A. Dobbeleir, H.R. Ham, C. Brihaye, M. Guillaume, F.F. Knapp, J. Vandevivere.
13) Clinical Nuclear Medicine. Volume 14, 12 1989.
Continuous anterior acquisitions in gastric emptying: comparison with the mean.
J. Roland, A. Dobbeleir, H.R. Ham, J. Vandevivere.
14) Nuclear Med. Comm. 10,1989.
Renal background.
A. Piepsz, A. Dobbeleir, H.R. Ham.
15) Nuclear Med. Comm. 11, 1990.
Effect of mild mental stress on solid phase gastric emptying in healthy subjects.
J. Roland, A. Dobbeleir, J. Vandevivere and H.R. Ham.
16) Eur. J. Nucl. Med. Volume 17, 130-133 1990.
Evaluation of reproducibility of solid-phase gastric emptying in healthy subjets.
J. Roland, A. Dobbeleir, J. Vandevivere and H.R. Ham.
17) J. Nuclear Medicine. Volume 31, 4 1990.
Effect of background correction on separate Tc99m-DTPA renal clearance.
A. Piepsz, A. Dobbeleir, H.R. Ham.
18) Eur. J. Nucl. Med. Volume 18, 83-86, 1991.
Technetium 99m mercaptoacetyltriglycine (Mag3) gamma camera clearance calculations:
methodological problems.
M. Tondeur, A. Piepsz, A. Dobbeleir, H.R. Ham.
19) Nuclear Med. Comm. 12, 27-34, 1991.
Performance of a single crystal digital gamma camera for first pass cardiac studies.
A. Dobbeleir, P.R. Franken, H.R. Ham, C. Brihaye, M. Guillaume, F.F. Knapp and J. Vandevivere.
20) Nuclear Med. Comm. 12, 473-484, 1991.
Comparison between exercise myocardial perfusion and wall motion using Tl201 and Ir191m
simultaneously.
P.R. Franken, A. Dobbeleir, H.R. Ham, R.Ranquin, S. Lieber, F. Van den Branden, P. Van den
Heuvel, C. Brihaye, M. Guillaume, F.F. Knapp and J.Vandevivere.
21) Clin Nucl Med 1992, 17: 378-389
Visualisation of brainstem perfusion using a high spatial resolution SPECT system.
R. Dierckx, A. Dobbeleir, J. Vandevivere, H. Abts, P. DeDeyn.
22) Clin Nucl Med 1993,18:532-534
Tc-99m HMPAO tomography using a three headed SPECT system equipped with lead fanbeam
collimators.
R. Dierckx, A. Dobbeleir, J. Martin, P. De Deyn.
114
23) Clin Nucl Med 1993, 18 : 83-84.
High spatial resolution Tc-99m HMPAO brain SPECT in cerebellar emboligenic infarct.
R. Dierckx, L. Fidlers, A. Dobbeleir, P. De Deyn, J. Vandevivere.
24) Epilepsy Res 1992, 12 : 131-139.
Single photon emission computed tomography (SPECT) in seizure disorders using perfusion
tracers.
R. Dierckx, J. Vandevivere, L. Dom, K. Melis, G. Janssens, A. Dobbeleir, P. De Deyn.
25) European Heart Journal 1992, 13 : 1189-1194
Improvement in the efficacy of exercise first-pass radionuclide angiocardiography in
detecting coronary artery disease and the effect of patient age.
P. Franken, A. Vervaet, R. Ranquin, S. Lieber, P. Van Den Heuvel, F. Van Den Branden,
A. Dobbeleir and J. Vandevivere.
26) Eur. J. Nucl. Med. 1993, 20 : 684-689.
Quantification of technetium-99m hexamethylpropylene amine oxime brain uptake in routine
clinical practice using calibrated point sources as an external standard: phantom and human studies.
A. Dobbeleir, R. Dierckx.
27) Nucl. Med. Com. 1993, 14 : 792-797.
Sensitivity and specificity of Tc-99m HMPAO single headed SPECT in dementia.
R. Dierckx, M. Vandewoude, J. Saerens, T. Hartoko, P. Mariën, I. Capiau, A. Vervaet,
A. Dobbeleir and P. De Deyn.
28) Eur. J. Nucl. Med. 1994, 21 :514-520.
Parameters influencing SPET regional brain uptake of technetium-99m hexamethylpropylene amine
oxime measured by calibrated point sources as an external standard.
R. Dierckx, A. Dobbeleir, M. Maes, B. Pickut, A. Vervaet, P. De Deyn.
29) Eur. J. Nucl. Med. 1994, 21 :621-633.
Sensitivity and specificity of thallium-201 single-photon emission tomography in the
functional detection and differential diagnosis of brain tumours.
R. Dierckx, J. Martin, A. Dobbeleir, R. Crols, I. Neetens, P. De Deyn
30) Eur. J. Nucl. Med. 1995, 22 :427-433.
Technetium-99m HMPAO SPET in acute supratentorial ischaemic infarction, expressing deficits
as millilitre of zero perfusion.
R. Dierckx, A. Dobbeleir, B. Pickut, L. Timmermans, I. Dierckx, A. Vervaet, J. Vandevivere, W.
Deberdt, P. De Deyn.
31) Yearbook of Nuclear Medicine 1995: 273-275.
Quantification of technetium-99m hexamethylpropylene amine oxime brain uptake
in routine clinical practice using calibrated point sources as an external standard:
phantom and human studies.
A. Dobbeleir, R. Dierckx.
32) Nucl. Med. Com. 1996, 17 : 583-590.
240 Degrees: Why not ?
A.S. Hambÿe, A. Dobbeleir, E. Stulens, A. Vervaet, J. Vandevivere, P.R. Franken.
115
33) Nucl. Med. Com. 1997, 18 : 751-760.
Can we rely on Tc99m -sestamibi gated tomographic myocardial perfusion imaging to quantify left
ventricular function? A comparative study with classical isotopic techniques
for the measurement of ejection fraction.
A.S. Hambÿe, A. Dobbeleir, A. Vervaet, H. Chi-Chou.
34) Clin Nucl Med 1997, 22 : 172-175.
Determination of systolic thickening index with gated Tc99m Sestamibi SPECT:
A new parameter of myocardial viability ?
A.S. Hambÿe, A. Dobbeleir, M. Derveaux, J. Vandevivere, P. Van Den Heuvel.
35) Nucl. Med. Com. 1997, 18 : 1135-1147
SPET generated colour-coded polar maps to quantify the uptake of 99mTc-sestaMIBI and 123I-
BMIPP in chronically dysfunctional myocardium: comparison with coronary anatomy and wall
motion.
A.S. Hambÿe, A. Dobbeleir, P.R. Franken.
36) Nuklearmedizin 1998, 37: S1-S6
BMIPP imaging to identify residual myocardial viability in patients with acute and
chronic left ventricular dysfunction.
P.R. Franken, A.S. Hambÿe, A. Dobbeleir, F. De Geeter, P. Dendale.
37) J. Nuclear Medicine 1998. Volume 39: 1845-1850
Abnormal BMIPP uptake in chronically dysfunctional myocardial segments: correlation
with contractile response to low-dose dobutamine.
A.S. Hambÿe, M. Vaerenberg, A. Dobbeleir, P. Van den Heuvel, P.R. Franken.
38) Psychiatry Research: Neuroimaging Section 90, 1999 : 103-112.
Validation of the cerebellum as a reference region for SPECT quantification in patients suffering
from dementia of the Alzheimer type.
B. Pickut, R. Dierckx, A. Dobbeleir, K. Audemaert, K. Van Laere, A. Vervaet, P De Deyn.
39) Nucl. Med. Com. 1999, 20 : 737-745
Quantification of 99mTc-sestaMIBI and 123I-BMIPP uptake for predicting functional outcome in
chronically ischaemic dysfunctional myocardium.
A.S. Hambÿe, A.Vervaet, A. Dobbeleir.
40) Eur. J. Nucl. Med. 1999, 26 :655-658.
Influence of high-energy photons on the spectrum of iodine-123 with low- and medium-energy
collimators: consequences for imaging with 123I-labelled compounds in clinical practice.
A. Dobbeleir, A.S. Hambÿe, P.R. Franken.
41) J. Nuclear Medicine 1999. Volume 40: 707-714
Influence of methodology on the presence and extent of mismatching between 99mTc-MIBI and
123I-BMIPP in myocardial viability studies.
A. Dobbeleir , A.S. Hambÿe, P.R. Franken.
42) J. Nuclear Medicine 1999. Volume 40: 1468-1476
BMIPP imaging to improve the value of sestamibi scintigraphy for predicting functional outcome in
severe chronic ischemic left ventricular dysfunction.
A.S. Hambÿe, A. Dobbeleir, A. Vervaet, P. Van den Heuvel, P.R. Franken.
116
43) Nucl. Med. Com. 1999, 20 : 1031-1040
Quantification of 99mTc-HMPAO brain SPET in two series of healthy volunteers using different
triple-headed SPET configurations: Normal databases and methodological considerations.
K Van Laere, C. De Sadeleer, A. Dobbeleir, A. Bossuyt, P. De Deyn, R. Dierckx.
44) J. Nuclear Medicine 2000. Volume 41: 213-214
Effect of methodology on mismatching between 99mTc-MIBI and 123I-BMIPP.
A. Dobbeleir , A.S. Hambÿe, P.R. Franken.
45) Yearbook of Nuclear Medicine 2000: 279-280.
Influence of high-energy photons on the spectrum of iodine-123 with low- and medium-energy
collimators: consequences for imaging with 123I-labelled compounds in clinical practice.
A. Dobbeleir, A.S. Hambÿe, P.R. Franken.
46) Eur. J. Nucl. Med. 2000, 27 :1494-1500.
Prediction of functional outcome by quantification of sestamibi and BMIPP after acute myocardial
infarction.
A.S. Hambÿe, A.Vervaet, A. Dobbeleir. P. Dendale, P.R. Franken.
47) World Journal of Nuclear Medicine, Volume 1, Number 1, 10-2002
Image quality with Rhenium-188 and Technetium-99m: Comparative Planar and Spect Evaluation
in a Phantom Study and implications for dosimetry.
A.S. Hambÿe, A. Dobbeleir, A. Vervaet, FF (Russ) Knapp.
48) Nucl. Med. Com. 2004, 25, 347-353
Quantitative gated sestamibi SPECT for diagnosing coronary artery disease: a comparative study of
non corrected and scatter-corrected summed and end-diastolic images.
A.S. Hambÿe, A.Vervaet, A. Dobbeleir
49) Eur. J. Nucl. Med and Molecular Imaging 2004; 31: 1606-1613
Variability of left ventricular ejection fraction and volumes by quantitative gated SPET : influence
of algorithm, pixel size and reconstruction parameters in normal and small-sized hearts. A.S. Hambÿe, A.Vervaet, A. Dobbeleir
50) Submitted.
Quantification of Iodine-123-FPCIT SPECT with a resolution independent technique.
A. Dobbeleir, A.S. Hambÿe, A.Vervaet, H. Ham.
117
8. Dankwoord.
Dit proefschrift kan moeilijk als klassiek beschouwd worden daar het data bevat die verzameld
werden gespreid over meerdere decennia. De noodzakelijke methodologie gaat zelfs terug tot het
begin van mijn loopbaan en daarom ben ik ook dank verschuldigd aan personen die niet voorkomen
in dit werk maar die mij wetenschappelijk hebben gevormd. Mijn dankbetuigingen zijn dan ook min
of meer in chronologische volgorde weergegeven.
Aan Dr Philippe Decostre die mij meer dan 30 jaar geleden, ondanks nog af te leggen militaire
dienst, uitkoos om onder zijn leiding te werken in het St-Pieter ziekenhuis (Universiteit Brussel).
Hij heeft mij de basis van nucleaire geneeskunde bijgebracht.
Aan François Erbsmann, fysicus en wetenschapper in hart en nieren, die mij wetenschappelijk heeft
gevormd. Onder zijn inspirerende leiding werd de afdeling fysica uitgebouwd en de eerste
wetenschappelijke werken gepubliceerd.
Aan Prof Ami Piepsz, die mij het belang van nauwkeurig werk heeft bijgebracht en toonde wat
doorzettingsvermogen is. Onder zijn impuls heeft nucleaire een belangrijke impact gekregen op de
nierfunktie.
Aan Prof Hamphrey Ham, die elk probleem toch nog op een andere manier zag en met zijn inzicht,
vernieuwende ideeën te voorschijn toverde.
Nooit zal ik de rit in mijn auto vergeten naar het congres in Lausanne in 1976 samen met François,
Ami en Hamphrey waar wij met jong enthousiasme de gescheiden nierklaring en whole body scan
voorstelden.
Half de jaren tachtig overtuigde Dr Vandevivere mij om in Middelheim te Antwerpen te werken
waar een tweede periode in mijn loopbaan aanbrak. Aldaar heb ik samengewerkt met Dr Roland en
Rudi Vandermeiren die veel routine op zijn schouders nam zodat ik de tijd kreeg om nieuwe zaken
te ontwikkelen. Tevens werd er de basis gelegd voor de aangename post-Middelheim
samenwerking met Dr Koen Melis.
Aan Prof Philippe Franken, die mij met zijn grote kennis van nucleaire en klinische cardiologie
stimuleerde om de resultaten te kwantificiëren.
Aan Prof Rudi Dierckx, die van oordeel is dat nucleaire geneeskunde best kan functioneren onder
multi-disciplinaire samenwerking. Onder zijn impuls werden de hersenen uitgebreid bestudeerd.
Aan Ann Vervaet, statisticus, die vele resultaten controleerde en deze daarna ook in een
publiceerbare vorm voorstelde.
Aan Prof Anne-Sophie Hambÿe, tevens mijn echtgenote, die de voornaamste rol heeft in veel van
de voorgestelde werken. Niet alleen haar kennis maar ook haar litteraire kwaliteiten dragen veel bij
tot dit werk. Zonder haar zouden vele gegevens begraven zijn in een schuif van mijn bureau.
Aan Prof Anne Paans voor zijn hartelijk onthaal en begeleiding in Groningen.
Heel belangrijk is dat dit werk uitgevoerd is in een vriendschappelijke samenwerking en dat die
vriendschap gebleven is door de jaren heen.
Tot slot bedank ik de leden van de jury en tevens mijn kinderen, zus en schoonbroer.
Recommended