Download DIAGNOSTIC PERFORMANCE OF LOW-DOSE ATTENUATION-CORRECTED REST/STRESS Tc-99m TETROFOSMIN

DIAGNOSTIC PERFORMANCE OF LOW-DOSE ATTENUATION-CORRECTED REST/STRESS Tc-99m TETROFOSMIN
MYOCARDIAL PERFUSION SPECT USING THE DNM530C CZT CAMERA (QUANTITATIVE VERSUS EXPERT VISUAL ANALYSIS)
FP Esteves, RD Folks, L Verdes, JR Galt, EV Garcia, Emory University School of Medicine, Atlanta, GA
Objective
1Esteves FP, Raggi P, Folks RD, Keidar Z, Askew JW, Rispler S, O'Connor MK, Verdes L,
Garcia EV. Novel solid-state-detector dedicated cardiac camera for fast myocardial perfusion
imaging: multicenter comparison with standard dual detector cameras. J Nucl Cardiol. 2009
Nov-Dec;16(6):927-34.
Discovery NM 530c System Design
Pharm
or
Exercise
Rest
imaging
0 min
Pinhole projections for stress (20
mCi, 6 minute acquisition) and rest
(6 mCi, 8 minute acquisition)
acquired on the DNM 530c SPECT
system.
5-7 min
Stress
imaging
1 hr
2 hr
All patients received a dose of 6 mCi at rest and 20 mCi
at stress. Rest images were obtained over 7, 8, or 9
minutes and stress over 5, 6, or 7 minutes depending on
chest size. On the same day, low dose CT (average 0.4
mSv) of the patient’s thorax was obtained on a standalone
CT scanner using a shallow free breathing protocol. CT
images were imported and manually registered with the
SPECT images prior to reconstruction (MLEM) with AC
using the manufacturer’s software.
Patient Characteristics
Low Likelihood
Pilot
Validation
Normalcy
Sample size
40
150
55
40
Male/Female
20/20
78/72
31/24
14/26
53 ± 9
Age ± % std. dev.
50 ± 11
63 ± 12
63 ± 11
BMI ± % std. dev.
29 ± 6
28 ± 5
29 ± 5
29 ± 5
Chest pain
0% (0)
75% (112)
82% (45)
0% (0)
Exercise stress
100% (40)
59% (88)
20% (11)
38% (15)
Stress
Rest
CT
Low dose CT scan acquired on a
stand-alone CT scanner on the
same day of the SPECT.
Multi pinhole technology
• CZT detectors
•
2.46 mm detectors
The normal database for DNM 530c rest/stress AC Tc-99m tetrofosmin
myocardial perfusion SPECT is shown above right in the form of a polar plot. A
single database is used for both sexes.
*
Stress
Manual software registration of the
Stress and Rest SPECT scans
(reconstructed without AC) with the
CT. Note that the same CT is used
for both Stress and Rest SPECT
scans.
Rest
Normal Database Development
ECTb software was used for processing and
interpretation. Forty patients with low (<5%) likelihood
(LLK) of coronary artery disease (CAD) were used to
define the normal count distribution. An expert reader
scored a pilot group of 150 patients using the 17-segment
LV model on a 5-point scale (0-4).
Optimum criteria for abnormality were determined by
ROC curve analysis for each of 7 myocardial segments.
Abnormality criteria were prospectively validated in 55
patients who had coronary angiography within 2 months of
the SPECT and in 40 patients with LLK of CAD.
Obstructive CAD was defined as a 70% luminal stenosis in
at least one major coronary artery.
AC Normal Database
Specificity
Stress
(20 mCi)
7-9 min
Combined Male/Female
Acquisition and processing of DNM 530c SPECT with AC is illustrated
below. Subject is a 60 year old female with history of myocardial
infarction in the inferolateral wall who weighed 77kg (170 lbs), was 163
cm tall (5’4”), with a BMI of 29.2 kg/m2 at the time of imaging.
Results - Overall
Background
The DNM 530c SPECT system has a unique design with 19 CZT
detectors, each equipped with pinhole collimation, arranged in three
rows surrounding the patient in a 180° RAO to LPO arc. During
acquisition the detectors are stationary with the central row of (9) in
line with the patient’s heart and rows above and below the heart (5
each) providing views in a more longitudinal direction. Compared to
standard SPECT systems, this multiple pinhole design provides very
high sensitivity and allows for shorter imaging times and lower
radiation doses.1
The unique detector also produces differences in attenuation
patterns compared to conventional SPECT. In a study that compared
DNM 530c SPECT to conventional SPECT, some patients with breast
attenuation noted in the anterior wall on conventional SPECT had no
loss of counts on D530c SPECT. Instead some of these patients had
a mild count loss in the inferior wall.1 Attenuation correction may be
performed on the DNM 530c by registering the SPECT scan with a
separately acquired CT scan, converting the CT Hounsfield units to
99mTc attenuation coefficients, and incorporating attenuation correction
into the MLEM reconstruction. A routine procedure in the clinic where
this study was performed, care is taken to position each patient on the
CT scanner in the same manner as they are positioned for SPECT.
Rest
(6 mCi)
Results
Example
Tc-99m Tetrofosmin SPECT Protocol
Sensitivity
To compare the diagnostic performance of quantitative analysis
(QA) against visual analysis (VA) on rest/stress attenuation corrected
(AC) Tc-99m tetrofosmin myocardial perfusion SPECT using the
DNM530c CZT camera.
Methods
* P = 0.03
All others P = NS
Visual and quantitative sensitivity and specificity by vascular territory and
overall are shown using a 70% stenosis as criteria for obstructive CAD.
Conclusions
The overall diagnostic performance of quantitative attenuation correction is
comparable to that of visual analysis in low-dose Tc-99m tetrofosmin myocardial
perfusion SPECT with the DNM530c CZT camera. Quantitative analysis may
increase physician’s confidence in image interpretation and can be particularly
useful to cardiac readers unfamiliar with the DNM530c camera.
Stress/Rest oblique slices and ECTb polar plots for non-AC (for comparison, on
left) and AC. Blackout regions of the polar plots are determined by comparison to
the normal database using the abnormality criteria.
DISCLOSURE
This work was funded by GE Healthcare. Two of the authors (RDF, EVG) and Emory University receive royalties
from the sale of the Emory Cardiac Toolbox(TM) utilized in this poster. The terms of this arrangement have been
reviewed and approved by Emory University in accordance with its conflict-of-interest practice.