Content uploaded by Manishi L Narayan
Author content
All content in this area was uploaded by Manishi L Narayan on Oct 06, 2023
Content may be subject to copyright.
Eur J Nucl Med Mol Imaging (2023) 50 (Suppl 1): S1-S898
Aim/Introduction: Acute myocardial infraction (AMI) is one of the
major cause of mortality and morbidity in the developed world. If
a diagnostic algorithms of detection and treatment of this acute
condition is clear and developed enough, predictors of adverse
course of the disease have not been suciently identied.
Aim of this study to reveal of a complex diagnostic accuracy of
dynamic SPECT and cardiac MRI in predicting of adverse course
of the disease. Materials and Methods: Ninety patient with
rst identied AMI average age 62.5±10.5 years (61% male )
were included. Complete clinical and instrumental examination
according to ESC guide-lines was provided. Additionally, СMR
with gadolinium enhancement due to 2-7 days and dynamic
rest-stress SPECT within 7-10 day after admission were provided.
Rates of incident MACE following 12 months after acute coronary
event was assessed. Results: MACE were identied in 11 patients:
nonfatal stroke in 5 patients, 5 cases of cardiovascular death and
nonfatal MI in one patient. MACE group of patients characterized
signicant increase of myocardial mass, end systolic and diastolic
volumes of LV; end systolic volume of RV compare to group
without MACE: 285 (198;450.6)g vs 182 (157.9;225.5)g, 98 (65; 317)
ml vs 41.3 (32.7;55.5) ml; 160 (152;385.6)ml vs 120 (102;137.4)ml,
54 (47;85)ml vs 41 (32;56)ml, respectively. Ejection fraction of LV
and RV as well as myocardial salvage index were lower in patients
with MACE: 39.2 (16;55)ml vs 63.5 (54.9;70)ml and 35 (28;43)
ml vs 51 (46;56)ml; 11.62 (0.51;13.5) vs 30.95 (10.5;44.4) - against
non-MACE patients, respectively. Other CMR indices were not
signicant dierence. Compare of dynamic SPECT indices showed
that MACE group of patient against non-MACE characterized
increase of summed stress score and summed rest score 6.5 (6;30)
vs 5 (3;12) and 6 (4.5;26) vs 3 (0;7), respectively. Moreover, stress
and rest myocardial blood ow decreased in MACE patients:
0.86 (0.44;1.12)ml/min/g vs 1.09 (0.81;1.67) ml/min/g and 0.48
(0.28;0.48) ml/min/g vs 0.67 (0.40;1.04) ml/min/g. However
myocardial ow reserve did not dier between enrolled groups
of patients. Multivariate logistic regression analysis revealed that
ESV LV (1,04 [1,02-1,06]), EF RV (0,84 [0,76-0,98]) and rest MBF (0,85
[0,77-0,94]) were independed predictors of MACE in patients with
AMI (sensitivity 0.85, specicity 0.67, accuracy 0.79 and AUC 0.91).
Conclusion: The complex evaluation of cardiac MRI and dynamic
SPECT performed in subacute period of AMI is an informative
approach and can be used in the prognosis of MACE. These results
require further investigation.
EP-0449
Attenuation correction in SPECT-Myocardial perfusion
imaging:It can be corrected, but is it really useful?
S. Abhilash, S. Jeevan, P. Pradeep, S. Sarath, L. N. Manishi;
All India Institute of Medical Sciences, Rishikesh, INDIA.
Aim/Introduction: SPECT myocardial perfusion imaging (MPI)
is a non-invasive diagnostic method for assessing myocardial
perfusion in patients with coronary artery disease (CAD). Soft tissue
attenuation can aect its accuracy, but attenuation correction
(AC) compensates for this by correcting for photon absorption.
With the advent of hybrid SPECT/CT cameras, CT-based AC is now
widely used. This study evaluated the benets of AC in three major
vascular territories and compared it to coronary angiography
(CAG) and echocardiography (ECHO). Materials and Methods:
We retrospectively analysed 42 patients who were referred by the
cardiologist for assessment of myocardial perfusion (SPECT-MPI).
CAG done within 5 months of MPI along with recent ECHO
ndings were taken as the reference standard. A coronary artery
stenosis >75% and wall motion abnormality on ECHO were
considered the reference standards. For all patients, cardiac
SPECT with low-dose CT was acquired 1 hour after i.v. injection of
10-15 mCi of 99mTc Sestamibi at rest. Images were reconstructed
and CT-based attenuation correction was applied. Images were
processed using Emory Cardiac Toolbox software V:3.2 (ECT box;
Emory University, Atlanta, GA) on a Xeleris 4.0 Nuclear Medicine
workstation. Both AC and NAC images were compared. Results:
42 patients (39 males and 3 females) with a mean age of 57 years
were studied. Using a standard 17-segment polar map, a total of
126 major coronary artery territories (42 LAD, 42 RCA, and 42 LCX)
were analysed.On visual and semiquantitative analysis, 23/126
(RCA = 9, LCX = 1) vascular territories showed normalisation of
perfusion defects after AC. 1/126 new perfusion defects were
observed after AC in apparently normal territories (LAD = 1). On
comparing MPI ndings with reference standards, LAD territory
defects that were observed (34/126) on MPI remained unchanged
after AC and correlated well with reference standards. 13/21 RCA
and 11/17 LCX territory defects had signicant stenosis and
RWMA, which were persistent even after AC, were considered true
positives. However, 8/21 RCA and 2/17 LCX defects normalised
after AC, thus being considered false negatives. Conclusion:
In our study, though fallacious inferior wall defects involving
RCA in non-AC MPI images resolved with AC, few apparent
perfusion defects were underestimated with AC. It is important to
understand that CT-based AC can overcorrect perfusion defects
and introduce new artefacts. It should be used with caution; both
non-corrected and corrected image datasets should be reviewed
before integrating them into the nal report.
EP-26
e-Poster Area
B: Imaging Clinical Studies -> B4
Cardiovascular Imaging Clinical Study -> B42
Metabolism and Innervation
EP-0450
Automated absolute quantitation of cardiac sympathetic
activity using convolutional neural network and
123I-MIBG SPECT/CT
S. Saito1, K. Nakajima2, L. Edenbrandt3,4, M. Larsson5, O. Enqvist5,
J. Ulén5, S. Kinuya1;
1Department of Nuclear Medicine, Kanazawa University,
Kanazawa, JAPAN, 2Department of Functional Imaging
and Articial Intelligence, Kanazawa University,
Kanazawa, JAPAN, 3Sahlgrenska University Hospital,
Gothenburg, SWEDEN, 4University of Gothenburg,
Gothenburg, SWEDEN, 5Eigenvision, Malmö, SWEDEN.
Aim/Introduction: We have already created algorithm for
segmentation of cardiac region in 123I-MIBG SPECT images without
X-ray CT using convolutional neural networks (CNN). We have also
attempted to use 123I-MIBG SPECT/CT images and successfully
conducted automatic heart segmentation based on CNN. Since
CNN-based automatic absolute quantitation in 123I-MIBG SPECT/
CT has not been evaluated, the purpose of this study was to
quantify CNN-based absolute heart counts and standardized
uptake value (SUV) using 123I-MIBG SPECT combined with
low-dose CT for attenuation correction purposes and compare
these values with conventional planar image-based quantitation.
Materials and Methods: A total of 70 patients (46% men, mean
S608