Asnc2008 poster session ii - comparison with other modalities

POSTER SESSION II
Friday, September 12, 2008, 9:30 a.m. – 11:00 a.m.
Comparison with Other Modalities
septum and the apex in normal eye analysis of grayscale. The interventric- BETA BLOCKER UTILIZATION AND SAFETY IN AN OUTPATIENT ular septum enhanced the most followed by the lateral wall, posterior wall, CARDIAC COMPUTED TOMOGRAPHIC ANGIOGRAPHY ENVIRONMENT inferior wall, anterior wall, and the apex (as a reference, the CT number for D Gopalakrishnan,1 D Abner,2 A Gopal,2 N Ahmadi,2 RS Pal,2 MJ Budoff2 a non-enhanced tissue: 40 HU). In 25 individuals, the HU of dark areas in 1The Heart Hospital Baylor Plano, Plano, TX, 2Los Angeles Biomedical the myocardium by eye analysis was very low (6.5 Ϯ 35.4 mean HU) Research Institute at Harbor UCLA Medical Center, Torrance, CA showing a wide range of hypoenhancement.
Background: An optimal heart rate control is mandatory with the current
Conclusion: “Normal” grayscale analysis by the human eye appeared to
64-slice scanners to prevent motion artifacts during cardiac computed correctly identify the normally enhanced myocardium with wide spectrum tomographic angiography (CTA). Depending on the practitioner’s comfort of contrast enhancement differences from apex to septum. Additionally, the level, beta blocker (BB) administration is done either exclusively by the dark areas based on eye analysis clearly reflected significantly lower HU aggressive rapid intravenous (IV) administration while on the table (IV metoprolol 5 mg every 1 minute up to a maximum dose of 50 mg) or by the
conservative oral outpatient 3-day preparation with metoprolol or atenolol.
Concerns regarding their safety have resulted in some practices considering
the purchase of more expensive scanners with increased temporal resolution
to eliminate the use of such heart rate slowing medications during CTA.
This study was done to evaluate the feasibility and safety of heart rate
slowing medications in an outpatient CTA setting.
Methods: Fifty-one consecutive patients (age 56 ϩ 13 years, 71% males) with
an intermediate likelihood of coronary artery disease who underwent CTA were
studied. The heart rate was initially assessed 3 days before the procedure and the
patients were started on BB, either metoprolol or atenolol once a day. On the
day of testing, the heart rate was measured again and if the HR was Ͼ 65, IV
metoprolol was used at doses of 5-10 mg every 3 minutes up to a dose of 40 mg.
As needed, IV verapamil up to 10 mg was also used to assist the BB.
Results: Seventy-one percent of the individuals received a daily outpatient
BB based on the initial heart rate assessment. Twenty-nine percent required
IV BB on the day of CTA and only one required additional IV verapamil.
Receiving outpatient BB reduced the use of IV BB on the day of CTA to
slow the heart rate by 69%. In those who already received outpatient BB,
only 31% required IV medications to further assist versus 27% with optimal
initial heart rate who did not receive prior oral BB (P Ͼ 0.9). Heart rate at
the time of acquisition was 59.7 ϩ 5.4 bpm. The images were acquired with
a 64-slice scanner and were of good quality without significant motion
artifacts. There were no adverse affects while the patients were in the lab or
after going home, and these medications were well tolerated.
Conclusion: Medications like BB and calcium channel blockers can be safely
used in an outpatient setting and significantly reduce the IV BB to slow heart
rate on the day of CTA. During CTA, contrast-related reactions are more likely PROSPECTIVE TRIGGERING WITH DUAL SOURCE CARDIAC CT: to happen than those caused by such heart rate slowing medications.
INITIAL CLINICAL EXPERIENCER Blankstein, AB Shah, RA Pale, S Abbara, W Mamuya, U Hoffmann, Massachusetts General Hostpital, Boston, MA COMPARISON TO THE ACTUAL CONTRAST ENHANCEMENT IN Background: Prospectively triggered (PT) cardiac CT, whereby radiation is
only administered at a predefined phase of the cardiac cycle, has been shown N Ahmadi, EP Young, RS Pal, K Nasir, V Gabrielian, SS Mao, to substantially reduce patient radiation dose. Recent preliminary study showed the feasibility of this technique when a single source scanner was Los Angeles Biomedical Research Institute at Harbor UCLA Medical used and all patients were treated with beta-blocker medications. The aim of our study was to assess the use of this technique in a clinical patient Background: Given the limitations of human eye for grayscales, only
population using the Dual Source Cardiac CT (DSCT; temporal resolution myocardial lesions that are infarcted or those areas replaced by calcium or 83msec) when beta blockers were not routinely administered.
fat will be promptly recognized from normal grayscale analysis in cardiac Methods: Among 200 consecutive patients (pts) referred for a clinical DSCT
computed tomographic angiography (CTA). On the other hand, subtle exam, prospective triggering was utilized in 20 pts. kV (range 100-140, average changes in the “normal” grayscale may be overlooked. We sought to 116) and mAs (average 119) were chosen for each patient based on physician analyze the pattern of contrast enhancement of the walls of the left assessment of body habitus. Beta blockers were not administered (except for 1 ventricular (LV) chamber in a normally appearing LV chamber.
pt). Effective radiation dose was calculated by multiplying the dose-length Methods: CTA images from 75 individuals (age 63 Ϯ 13 years, 73% males)
product (DLP) times a conversion factor (k ϭ 0.017 mSv/mGy/cm). All data were analyzed by an experienced reader using a GE Advantage Workstation related to scan acquisition parameters was collected prospectively. Per-patient (AW 4.3). The maximum, mean, and the SD (standard deviation or the analysis of non-evaluable segments was determined based on clinical readings.
noise) Hounsfield Units (HU) were measured using the ROI (region of Results: Of the 20 PT scans performed, (mean age 42, mean BMI 29.3, 70%
interest) tool in these LV chambers of relatively normal appearing gray- male), 13 pts were referred for coronary evaluation, 6 were referred for evaluation of coronary arteries and aortic disease (larger coverage), and 1 pt was Results: There were differences in the pattern of contrast enhancement of
referred for evaluation of a cardiac mass. The average heart rate was 66 bpm.
the LV chamber walls that was close to a 30% difference between the All pts were in normal sinus rhythm, and 20% had either PACs or partial- volume correction (PVC) during the acquisition. The average radiation dose for all 20 exams was 3.5 mSv (range 1.3 - 6.7) As expected, lower effective dose DETECTION OF CORONARY ARTERY DISEASE BY FREE-BREATHING, was achieved with lower kV and smaller coverage. When excluding the 6 WHOLE HEART CORONARY MAGNETIC RESONANCE ANGIOGRAPHY: studies which had a larger coverage, the average dose was 3.2 mSv. Of these, those which used 100kV resulted in an average dose of 1.5mSv. Twenty percent T Kunimasa, N Matsumoto, Y Nakano, Y Suzuki, S Yoda, T Kanai, (4/20) of pts had at least one non-evaluable segment. The reasons for non-evaluable parts included ectopy resulting in slab artifacts (2 pts), calcium (1 Nihon University School of Medicine, Tokyo, Japan
Background: Free-breathing, whole heart coronary magnetic resonance
pt) and presence of intra coronary stent (1pt).
angiography (MRA) has gained a great attention as a totally noninvasive Conclusions: Even in the absence of beta-blockers, use of prospective
diagnostic modality for the detection of coronary artery disease. We triggering DSCT is feasible and results in a significant reduction of radiation examined the accuracy of coronary MRA to identify the presence or absence dose. Use of beta-blockers to decrease ectopy may reduce the number of of coronary artery stenosis in comparison with conventional coronary angiography.
Methods: Free-breathing, whole heart coronary MRA with diaphragm drift
correction software was performed in 39 consecutive patients undergoing ATHEROSCLEROTIC PLAQUE MORPHOLOGY ASSESSMENT conventional coronary angiography. A total of 156 coronary arteries and ACCORDING TO BURDEN OF CORONARY ARTERY DISEASE 312 coronary artery segments were analyzed.
IN PATIENTS WITH TYPE 2 DIABETES MELLITUS BY NON- Results: In the vessel-based analysis, the sensitivity to detect coronary
stenosis Ն 50% was 80% and the specificity to define luminal narrowing Ͻ UN Ibebuogu,1 K Nasir,2 N Ahmadi,3 A Gopal,3 D Goodwin,3 MJ Budoff3 50% was 97%. The accuracy, positive predictive value, and the negative 1Medical College of Georgia, Augusta, GA, 2Massachusetts General predictive value were 92%, 93%, and 92%, respectively.
Hospital, Boston, MA, 3Los Angeles Biomedical Research Institute at Conclusions: Free-breathing, whole heart coronary MRA that provides
diaphragm drift correction software yields excellent diagnostic accuracy to Background: Patients with diabetes mellitus (DM) have a higher risk of
detect significant coronary artery disease and has potential to become the coronary artery disease (CAD) and are likely to have a higher underlying routine diagnostic modality for patients with suspected coronary artery atherosclerotic burden. However the atherosclerotic plaque composition in these patients is not studied. In this study we evaluated the plaque burden,morphology, and distribution in type 2 DM patients using multi-detector computed tomography angiography (MDCTA).
EVALUATION OF ISCHEMIC MYOCARDIAL STUNNING FOLLOWING Methods: The study population consisted of 40 symptomatic diabetic subjects
ADENOSINE VASODILATOR STRESS USING CARDIAC MAGNETIC (63 Ϯ 10 years, 55% men), who underwent contrast-enhanced MDCTA.
RESONANCE IMAGINGL Bakhos, TA Holly, E Wu, DC Lee Enrolled patients had an intermediate pre-test probability of obstructive CAD.
Results: A majority of individuals (n ϭ 33, 83%) had at least one segment with
Background: Several studies have reported myocardial stunning in up to
any plaque; 69% of diabetic had detectable coronary artery calcification (CAC) 30% of patients undergoing adenosine SPECT imaging. However, in and 36% had CAC Ն 400. Among individuals with any plaque the mean patients with significant perfusion defects, the edge-detection algorithm that number of segments involved were 5.7 Ϯ 3.0 segments; the respective mean gated SPECT relies on may fail to properly assess wall motion. Therefore, number of segments with exclusively non-calcified, calcified and mixed plaques the purpose of this study was to evaluate the incidence of myocardial were 1 Ϯ 1, 2.7 Ϯ 2.4, and 2 Ϯ 2 segments, respectively. Among those with stunning following adenosine stress by cardiac magnetic resonance imaging any plaque, the overall proportion of segments that had noncalcified, calcified, (CMR), a modality in which functional analysis is independent of perfusion.
and mixed atherosclerotic plaques were 25%, 44%, and 31%, respectively Methods: Thirty-three patients undergoing adenosine vasodilated CMR
In our study, 12 patients (30%) had at least one coronary segment perfusion imaging were retrospectively identified. Basal, mid, and apical with significant stenosis (luminal narrowing Ն 50%). Type 2 DM patients in short axis cines were acquired before and immediately after adenosine our study with significant stenosis were more likely to have plaque composition, infusion. Ejection fraction (EF) was measured blindly using CAAS-MRV that was mixed in nature (39% vs. 28%) and less likely to be exclusively
noncalcified plaque alone (17% vs. 26%) when compared to those without
significant stenosis. On the other hand, no difference was observed in the
respective proportion of exclusively calcified plaque (44% vs. 46%).
Conclusions: Our study demonstrates a high burden of CAD in patients
with diabetes. Majority of atherosclerotic plaque in these patients had
calcification, however among patients with significant CAD, mixed plaque
composition was more commonly observed. The prognostic value of these
different atherosclerotic plaque morphologies on MDCT in high risk DM
patients need to be assessed in larger prospective studies.

(PieMedical, Maastricht, The Netherlands). First-pass perfusion imaging mA, and 3 mm slice thickness. The parameter of MDCT was 140 kVp, 430 was performed at matched slice positions and blindly analyzed visually MA, 350 ms/per rotation and 2.5 mm in thickness. The calcium score were (0-normal to 3-severe defect) on a 16-segment model. Sum difference score measured with the Aquarius workstation (Terarecon, Inc). Each calcific foci (SDS) was the stress defect score minus rest defect score (maximum 48).
was measured. The Chi-square test (X2) was completed to test the difference Results: The mean EF prior to adenosine infusion was 44.8%. The mean EF
in sensitivity to detect small and low density calcium foci.
following adenosine infusion was 46.0%. As seen in Figure 1, pre- and Results: The foci number that can be detected and scored was 58/66 and
post-adenosine EF are highly correlated with a slope of 1.0 (95% confidence 40/66 in the in-vivo group and 46/57 and 35/57 in the in-vitro study using interval 0.90 - 1.05). Severe reversible perfusion defects (SDS Ն 5) were GE64 and EBT respectively (PϽ0.05).
seen in 15/32 (47%) of patients. In these patients, mean pre- and post- Conclusions: There was a significant increase in the sensitivity to detect
adenosine EF were 44.7% and 46.8% respectively.
small or low density calcium foci with 64-MDCT in both patients and Conclusion: Myocardial stunning was not seen in patients undergoing
adenosine vasodilated CMR perfusion imaging, even in the presence ofsignificant reversible perfusion defects. The apparent fall in EF reported in nuclear single-photon emission computed tomography studies may be due ROLE OF CORONARY FLOW RESERVE DURING HIGH RESOLUTION to suboptimal edge detection rather than myocardial stunning due to B-MODE ULTRASOUND IN THE FUNCTIONAL ASSESSMENT OF CORONARY ARTERY STENOSIS: COMPARISON WITH GATEDSPECT M Damiano, L Ferrara, P Gallo, C De Simone, A D’Errico, RELATIONSHIP BETWEEN CORONARY STENOSIS BY MULTI-SLICE CT ANGIOGRAPHY AND ISCHEMIA BY TC-99M PERFUSION SPECT T Chua,1 S Tan,1 J Huang,2 F Keng,1 R Tan,1 F Cheah2 Background: To estimate the value of coronary flow reserve (CFR), as
1National Heart Centre, Singapore, Singapore, 2Singapore General measured by transthoracic Doppler echocardiography, for functional assess- ment of stenosis severity in comparison with exercise gated single-photon Background: Multislice computed tomography (MSCT) shows great prom-
emission computed tomography (SPECT).
ise for the detection of coronary artery disease (CAD), but there is limited Methods: We studied 142 patients suspected of having coronary artery
data comparing its accuracy with established methods such as Tc-99m disease. The flow rate in the distal left anterior descending (LAD) coronary perfusion (single-photon emission computed tomography (SPECT).
artery was measured by echocardiography both at rest and during intrave- Methods: We compared the results of 16-Slice MSCT angiography using a
nous infusion of dipyridamole. CFR was calculated as the ratio of hyper- Siemens scanner with stress-rest Tc-99m Sestamibi Gated SPECT (dual emic to basal peak diastolic flow velocities. The CFR measurements by head gamma camera, Philips) in 59 patients who underwent both procedures ultrasonography were compared with the results of gated-SPECT. All within 6 months at our hospital with no intervening revascularization or patients underwent coronary angiography as control.
cardiac events. SPECT was performed with symptom-limited exercise in 52 Results: Complete Doppler ultrasound data were acquired for 136 of 142
patients and vasodilator stress in 7 patients for the diagnosis of CAD or study patients. Of these 136 patients, SPECT confirmed reversible perfusion detection of ischemia. Results were defined as normal or abnormal based on defects in the LAD territories in 82 patients (group A). Fifty-four patients the presence or absence of ischemia. CAD detected by MSCT was graded had normal perfusion in the LAD territories (group B). Peak CFR (mean as absent, mild (0-40% diameter stenosis), moderate (40-70%), severe value Ϯ SD) were 1.6 Ϯ 0.5 in group A and 2.4 Ϯ 0.7, respectively (p Ͻ 0.001). CFR Ͻ 2.0 predicted reversible perfusion defects, with a sensitivity Results: All patients with total occlusions had ischemic SPECT studies,
and specificity of 88% and 84%, respectively.
compared to 15/20 (75%) patients with severe or total occlusions, 17/28 Conclusions: Noninvasive measurement of CFR by transthoracic Doppler
(61%) patients with moderate or worse stenosis, 19/60 (32%) patients with ultrasonography provides a functional estimation of LAD stenosis severity any grade of CAD and 4/19 (21%) patients with no detectable CAD by MSCT (p ϭ 0.008). Of 36 patients with normal SPECT studies, 21 (58%)had evidence of atherosclerosis by CTA, though only 5/36 (14%) had severe disease. Of 23 patients with abnormal SPECT studies, 19/23 (83%) had SIGNIFICANT DIFFERENCES IN LEFT VENTRICULAR VOLUMES evidence of CAD, with 15/23 (65%) assessed as severe or occluded by AND EF BETWEEN GSPECT AND ECHOCARDIOGRAPHY AND CTA. One-quarter of patients with stenosis assessed as severe by CTA had A Hovland,1 H Bjørnstad,1 UH Staub,1 J Prytz,1 A Støylen,2 H Vik-Mo2 Conclusions: There is a stepwise relationship between SPECT findings and
1Nordland Hospital, Bodo, Norway, 2St Olav Hospital, Trondheim, MSCT, with an increasing prevalence of ischemia as the degree of stenosis increases. However, 25% of patients with stenosis assessed as severe by Background: Left ventricular ejection fraction (EF) is a powerful predictor
CTA had normal SPECT studies, suggesting the absence of ischemia despite of prognosis in coronary artery disease and is often estimated by echocar- diography or nuclear imaging. The purpose of the current study was tomeasure end diastolic and end systolic volumes and hence EF for echocar- diography compared with different gated single-photon emission computed COMPARISON OF EBT AND 64 MDCT TO DETECT SMALL OR tomography (GSPECT) software and to measure interrater differences using LOW-DENSITY CALCIUM FOCI: AN IN-VIVO AND IN-VITRO Methods: Eighty-four patients scheduled for nuclear imaging due to known
MJ Budoff,1 SS Mao,2 N Ahmadi,1 F Flores,1 H Wu,1 N Neal,1 N Petal1 or suspected coronary artery disease underwent GSPECT (patients with 1Los Angeles Biomedical Research Institute, Torrance, CA, 2Los Angeles atrial fibrillation or left bundle branch block were excluded). We used a Biomedical Research Institute, Carson, CA 1-day protocol GSPECT with 99mTc-tetrofosmin, using 8 frames/cardiac Background: The aim of this study was to estimate the sensitivity to detect
cycle during rest images. GSPECT images were processed by two raters small or low density calcium foci with Electron Beam Tomography (EBT) who estimated left ventricular volumes and EF using the Cedar-Sinai and 64 multidetector computed tomography (MDCT) (GE 64).
quantitative gated-SPECT (QGS), Emory Cardiac Toolbox (ECTB) and Methods: The in-vivo group included 66 single coronary arteries with solid
4D-MSPECT of the University of Michigan. Echocardiographic volumes calcium foci of Ͻ10 Agatston score (AS), selected from 93 patients who were measured by biplane Simpson‘s method. Measures are means Ϯ SD.
underwent coronary calcium scanning (CAC) with both EBT and MDCT on Differences were compared with t-tests. Interrater differences were calcu- the same day. All foci can be scored by either scanner or both EBT or lated as mean difference (95% limits of agreement).
MDCT. The in-vitro study included a cork chest phantom with dog heart, 57 Results: The 84 patients were 62.1 years Ϯ 8.9, 35% were women, and 26%
small (Ͻ6.6-18.9 mm3 with 200 mg/cc) or low density foci (14.8-157.8 previously had myocardial infarction. Echocardiographic end diastolic mm3, 150mg/cc) were inserted into a coronary artery and sealed with wax.
volume was 90.4 ml Ϯ 28.1, end systolic volume 32.1 ml Ϯ 18.7, and EF The technique parameters of EBT (GE Imatron) were 100 ms, 130kVp, 630 65.8% Ϯ 10.5. GSPECT volumes and EF are shown in Table 1.
Interrater mean difference in end diastolic volumes by QGS was 0.92 very good with no documented hard cardiac events to date. The findings of (-12.6-14.5), ECT -2.9 (-31.9-26.2), pϭ0.013 compared to QGS, 4D- severely obstructive CAD in 5 patients with CAC of 0 preclude the use of MSPECT -0.52 (-10.6-9.6), pϭ0.041. Interrater difference in end systolic CAC for excluding CAD, particularly in symptomatic younger patients.
volumes by QGS was 1.5 (-5.7-8.7), ECT -0.45 (-9.1-8.1) pϭ0.004,4D-MSPECT 0.32 (-7.5-8.2) pϭ0.046.The interrater mean difference in EF for QGS was 1.1( -5.3-7.5), mean difference for ECT was 0.05 and 95% ( -9.4.9.5) ,pϭ 0.062 compared to QGS and mean interrater difference was MYOCARDIAL PERFUSION IMAGING AND DOBUTAMINE 0.69 for 4D-MSPECT and 95% limits of agreement were -8.59-9.97 STRESS ECHOCARDIOGRAPHY IN THE DETECTION OF CORONARY ARTERY DISEASE IN HYPERTENSIVE PATIENTS Conclusions: Ejection fraction and left ventricular volumes calculated by
N Sultana,1 I Nisa,1 S Banarjee,2 AH Jehan3 different software programs in gated SPECT differ significantly from those 1Institute of Nuclear Medicine and Ultrasound, Dhaka, Bangladesh, obtained by biplane echocardiography. There are significant interrater 2BSM Medical university, Dhaka, Bangladesh, 3Center for Nuclear differences in end diastolic and end systolic volumes with different software. The discerning clinician should keep in mind that volumes and EF Background: Simultaneous dobutamine stress myocardial perfusion imag-
reported by a GSPECT study may differ between raters and from echocar- ing and dobutamine stress echocardiography for the evaluation of the presence and extent of coronary artery disease (CAD) are assessed for ahead-to-head comparison regarding the diagnostic accuracy of the two tests Table 1. Volumes and EF, *p Ͻ 0.05, **p Ͻ 0.001 compared to echocardiography
in hypertensive group of patients. The aim of this study is to compare the End diastolic
End systolic
sensitivity and specificity of these two imaging modalities, for the detection Software
volume, ml
volume, ml
Methods: This prospective observational study was performed at the
Institute of Nuclear Medicine & Ultrasound in collaboration with the cardiology department of BSM Medical University, Dhaka from January2006 to July 2007. A total of 80 patients (male 82% and female 18%, mean age 55.0 Ϯ 1.02 years, range 31-68 years) are included in this study and VALUE OF SAME-SESSION CORONARY ARTERY CALCIUM (CAC) informed about the necessity, risk, and benefit of these two tests. Informed SCORING IN THE SETTING OF STRESS MYOCARDIAL PERFUSION consent was obtained. All patients underwent simultaneous dobutamine IMAGING (MPI) USING CT-BASED ATTENUATION CORRECTION stress echocardiography (DSE) followed by dobutamine stress myocardial perfusion imaging (MPI) in single day stress-rest protocol with Tc 99m tetrofosmin. Coronary angiogram was performed in all patients within one month after imaging studies (significant stenosis was Ͼ 50%).
Background: Findings of a zero or very low coronary artery calcium score
Results: The overall sensitivity, specificity, and predictive accuracy of
(CAC) or a normal or low-risk myocardial perfusion single-photon emission dobutamine stress myocardial perfusion imaging for the detection of computed tomography (SPECT) study using computed tomography-based coronary artery disease were 97%, 62.5%, and 90%, respectively. Positive attenuation correction (CTAC), have excellent negative predictive value for predictive values were 91.2% and negative predictive value was 83.3%. The short- and intermediate-term hard and soft coronary events. CAC is overall sensitivity, specificity, and predictive accuracy of dobutamine stress sensitive for underlying coronary artery disease (CAD), but not specific for echocardiography for the detection of coronary artery disease were 92%, degree of luminal stenosis, whereas SPECT myocardial perfusion imaging 75%, and 88.7%, respectively. Positive predictive values were 93.7%, while (MPI) displays greater specificity for flow-limiting CAD and positive negative predictive value was 70.6%. Dobutamine stress MPI showed predictive value for short-term events. Data evaluating the potential syner- higher sensitivity than DSE (p Ͼ 0.05), but specificity lower than DSE (P Ͼ gism between these two techniques are limited. We examined the correla- 0.05), that does not rich the level significance.
tion between degree of CAC and MPI abnormalities using CTAC and Conclusions: In our study, both non-invasive methods for the detection of
assessed concordance in patients further evaluated with angiography.
CAD showed a good diagnostic accuracy. Nevertheless the dobutamine Methods: A total of 250 subjects with no history of CAD referred for stress
stress myocardial perfusion imaging showed higher sensitivity in compar- MPI were studied. Imaging was performed using Siemens SYMBIA-T6 ison with DSE in this specific group of patients.
SPECT-CT imaging systems (Siemens Medical Solutions, Hoffman Estates,IL) and a stress Tc-99 sestamibi protocol. Separate breathhold CT acquisi- tions were acquired for CTAC and CAC. SPECT images were reconstructed NON-INVASIVE STRESS TESTING OF MYOCARDIAL PERFUSION for attenuation correction (including scatter correction and resolution DEFECTS: HEAD-TO-HEAD COMPARISON OF THALLIUM-201 recovery) using manufacturers’ software without modification. CAC was tabulated by vessel using a 13-segment coronary artery model and corre- GM Vincenti,1 R Nkoulou,1 J Vallee,1 G Ambrosio,2 F Mach,1 O Ratib,1 lated with perfusion defects scored for severity, i.e. summed stress and difference scores (SSS, SDS), and extent in each coronary distribution using 1Geneva University Hospital, Geneva, Switzerland, 2Perugia University a standard 17-segment model. Results were analyzed for concordance and further correlated with subsequent angiography in those patients referred Background: To evaluate the diagnostic value of magnetic resonance
with a positive SPECT MPI study. Clinical outcome was based on review perfusion imaging in the assessment of hemodynamically obstructive coronary artery disease (CAD) in a head-to-head comparison with thallium- Results: Of 250 subjects (51% male, 35% diabetic, 65% symptomatic),
201 (201TI) single-photon emission compute tomography (SPECT) as CAC of 0, 1-10, 11-99, 100-399, Ն 400 were found in 100, 16, 52, 32 and 55 subjects, with CAC of Ͼ100 in 56% of normal MPI. Twenty-eight Methods: Eighteen patients (mean age 61 Ϯ 5 years, 14 men, 4 women)
percent of all cases had CAC of Ͻ10 with normal MPI. Patients’ age ranged with a history of angina pectoris were studied with 201TI-SPECT and fast from 36 to 84 years, with a mean age of 55 years. Forty-eight patients had gradient-echo (GRE) magnetic resonance imaging (MRI) of myocardial subsequent coronary angiography. Higher SSS and SDS correlated with perfusion during dipyridamole-induced vasomotor stress. Within 5 days of angiographic disease severity across all CAC score ranges. Discordant 201TI-SPECT, GRE-MRI perfusion imaging was performed with the same angiographic findings were present in 5 patients, 3 with typical angina protocol as 201TI-SPECT. Images for both scans were obtained in oblique symptoms, with CAC scores of 0, of whom 4 were under 50 years old. All horizontal, vertical long axis and short axis planes. Myocardial segments patients classified as normal/low risk were alive per CareWeb at 7 to 17 were assessed visually and myocardial perfusion was graded on a semi- quantitative 5-point scoring system (0 ϭ normal, 1 ϭ mildly reduced Conclusions: In patients with normal MPI, potentially valuable additional
perfusion, 2 ϭ moderately reduced perfusion, 3 ϭ severely reduced information regarding underlying subclinical CAD was found in 58% with perfusion, and 4 ϭ nearly absent perfusion) to derive the summed stress CAC imaging, however, the intermediate outcome in this group remains score (SSS), summed rest score (SRS) and summed difference score (SDS).
In all patients coronary angiography was performed to evaluate CAD clinical practice, its clinical impact on health care resource utilization needs to be better understood. We sought to determine the clinical impact of CTA Results: Seven out of 18 patients studied had a history of previous
on ICA referrals, CTA accuracy, and CTA normalcy rate.
myocardial infarction. 201TI-SPECT determined myocardial ischemia (ab- Methods: To determine the ‘impact of CTA’, 7,017 consecutive patients
normal SSS Ͼ 4) during dipyridamole stimulation was found in the regional undergoing ICA prior to (n ϭ 3,538) and after (n ϭ 3,479) implementing a myocardial territories supplied by the LAD (n ϭ 6), LCX (n ϭ 5), and RCA cardiac CT program were reviewed. For the CTA ‘accuracy’, we evaluated (n ϭ 7) with a SSS of 22 Ϯ 11, SRS of 14 Ϯ 10, and SDS of 8 Ϯ 4. By consecutive CTA patients who underwent ICA. For ‘normalcy rate’, we 201TI-SPECT, the sensitivity to detect hypoperfused segments was 61% identified 201 patients with a low pre-test probability for obstructive CAD (11/18) with the GRE-MRI analysis. The difference in the detection of stress-induced perfusion defects (n ϭ 7) between 201TI-SPECT and Results: With the implementation of a cardiac CT program, the frequency
GRE-MRI reached statistical significance (pϽ0.001 by chi2-test). The of ‘normal’ ICA decreased from 31.5% (1,114/3,538 patients) to 26.8% sensitivity, specifity, negative predictive and positive predictive value, and (932/3,479 patients) (p Ͻ 0.001). Analysis of the 148 CTA patients that accuracy for GRE-MRI analysis of myocardial perfusion in the detection of underwent ICA showed CTA had excellent “per patient” sensitivity (99%), flow-limiting epicardial coronary artery lesions, as defined as diameter positive predictive value (92%), and negative predictive value (95%) for stenosis Ͼ50%, were 72%, 100%, 54%, 100%, and 73%, respectively.
obstructive CAD. The positive likelihood and negative likelihood ratios Conclusions: These preliminary results indicate a moderate diagnostic
were 11.9 and 0.06, respectively. Using thresholds of Ն50% diameter accuracy of cardiac GRE-MRI perfusion imaging during dipyridamole- stenoses, the ‘normalcy rate’ of CTA was 94%.
induced coronary flow increases in the detection of flow-limiting epicardial Interpretation: The clinical implementation of CTA appears to positively
artery lesions. The concordance between 201TI-SPECT and cardiac GRE- impact ICA by reducing the rate of normal ICA. As well, the high diagnostic MRI imaging in the detection of stress-induced perfusion defects was accuracy of CTA supports its role as a clinically useful tool.
relatively low, and 39% of 201TI-SPECT-determined perfusion defectswere not identified by cardiac GRE-MRI, which deserves further investi- REPRODUCIBILITY OF EJECTION FRACTION ASSESSMENT BYCT-ANGIOGRAPHY F Alqaisi, C Poopat, C Nelson, M Al-Mallah COMPUTED TOMOGRAPHIC CORONARY ANGIOGRAPHY AND FRAMINGHAM RISK FACTOR SCORES: FURTHER SUPPORT FOR Introduction: Cardiac computed tomography (CTA) has emerged as a new
tool to assess left ventricular ejection fraction (LVEF). The accuracy of BJW Chow,1 RS Beanlands,1 TD Ruddy,1 GA Wells,1 L Chen,1 Y Yam,1 multiphase CTA in the measurement of LVEF has been demonstrated in comparison to other imaging modalities. However, the reproducibility of 1University of Ottawa Heart Institute, Ottawa, ON, Canada, 2Ottawa CTA assessment of LVEF has not well been demonstrated. The aim of this study is to assess the interobserver variability in the LVEF assessment by Background: Framingham risk scores (FRS) are routinely used to identify
individuals who may benefit from aggressive risk factor modification but Methods: Multiphase CTA (20 phases, 2.5 mm slice thickness) of 45
FRS does not account for genetic and environmental variables. Computed patients were reprocessed for LVEF assessment by three investigators: an tomographic coronary angiography (CTA) can non-invasively assess coro- experienced computed tomography technologist with 3D imaging training, nary atherosclerosis and is an ideal modality for the identification of an inexperienced cardiology trainee, and a cardiologist with level-3 training subclinical atherosclerosis. The objective of this study is to understand the in CTA. The investigators were blinded to each others interpretations. All relationship between FRS and coronary atherosclerosis as measured by processing was done using the Vitrea Workstation from Vital Images (Minneapolis, MN). Pearson’s correlation coefficients were calculated.
Methods: Consecutive patients who underwent CTA were prospectively
Results: The mean age was 57.6 Ϯ 10.4 years and 58% were males. There
enrolled and categorized according to FRS. Atherosclerotic calcific and was excellent inter-observer correlation between the LVEF measurement by the experienced technologist and the level 3 cardiologist. (r ϭ .92, p Ͻ Results: In this study, 447 patients without a history of MI, DM and not on
.001). The median difference in the LVEF assessment was 3% (range 1-11).
statin therapy were categorized according to FRS. Coronary atherosclerosis There was significant but less strong correlation between the LVEF was present in 60.6% of patients. Of the 258 patients with very low FRS assessment by the trainee and either the experienced technologist (rϭ0.78, (calculated 10-year risk Յ5%), atherosclerotic plaque was visually present pϽ.001) or the cardiologist (r ϭ 0.72, p Ͻ .001).
in 113 (43.8%). Furthermore, 9.1% of patients with high FRS had no Conclusion: Assessment of LVEF by CTA was highly reproducible by an
evidence of atherosclerotic plaque. Although mean atherosclerotic plaque experienced technologist and a level 3 cardiologist. Whether extensive burden increased with the 10-year Framingham risk the correlation between training improves the precision of LVEF by inexperienced trainee is yet to FRS and plaque was fair (r ϭ 0.50; p Ͻ 0.001).
Conclusions: Although FRS and other established clinical variables are
predictive of coronary artery disease events, CTA provides evidence of
calcific and non-calcific coronary atherosclerosis in many patients with low- THE IMPACT OF VOLUME AND READER EXPERIENCE ON THE to intermediate-risk Framingham scores. Furthermore, a small minority of DIAGNOSTIC ACCURACY OF CORONARY CT ANGIOGRAPHY patients with high FRS have no evidence of atherosclerosis. Prospective S Bhojraj, T Larson, K Abdul-Nour, MV Pantelic, M Al-Mallah studies are required to determine the value of identifying subclinical coronary atherosclerosis with CTA and modifying therapy based on these Background: Sixty-four-slice computed tomography angiography (CTA)
has emerged as a powerful non-invasive tool to rule out coronary arterydisease (CAD). Most of the diagnostic accuracy data comes from high- volume centers with highly experienced readers. The aim of this study is to DIAGNOSTIC ACCURACY AND IMPACT OF COMPUTED establish the diagnostic accuracy of CTA in a community setting, and to TOMOGRAPHIC CORONARY ANGIOGRAPHY ON UTILIZATION determine the impact of the reader experience on the positive and negative predictive value (PPV, NPV) of CTA changes with reader experience.
BJW Chow,1 A Abraham,1 GA Wells,1 L Chen,1 TD Ruddy,1 Y Yam,1 Methods: We included 41 consecutive patients without known CAD who
had coronary CTA and coronary angiography within 90 days. The CTA and 1University of Ottawa Heart Institute, Ottawa, ON, Canada, 2Ottawa angiograms were each evaluated using the standard American Heart Association 16-segment model. A “by segment” analysis was performed Background: Since computed tomographic coronary angiography (CTA)
using diagnostic coronary angiography as a gold standard.
has a high negative predictive value, it is a potential gatekeeper for invasive Results: A total of 655 segments were analyzed. The sensitivity, specificity,
coronary angiography (ICA). Before CTA can be further accepted into PPV, and NPV of CTA were 52%, 90%, 44% and 93% respectively.
Comparing the earlier cases to the later cases, there was an increase in the widely used load-independent measure of DF, in a population with low PPV (39% v. 50%) of CTA with no change in the NPV.
likelihood of cardiovascular disease.
Conclusions: CTA retains a high NPV when performed outside of high
Methods: This study evaluated 53 patients (55% male), mean age 57 years
volume centers, thus retaining its ability to significantly rule out coronary (Ϯ 14), who had both GSPECT and TDI within 1 month and no significant artery disease. Readers’ experience results in improving the specificity and change in their clinical status between the two studies. Patients with positive predictive value of the CTA.
coronary artery disease, hypertension, diabetes mellitus, atrial fibrillation,severe valvular disease, heart rate Ͼ100, or technically inadequate studies Experience
Sensitivity
Specificity
were excluded. All echocardiograms were required to have normal ejection fraction, normal wall motion, and no evidence of diastolic dysfunction by standard TDI criteria, defined as a lateral wall E’ Ͼ10 cm/s for patientsbetween 45 and 54 years, Ͼ 9 cm/s for patients between 55 and 65 years and Ͼ 8 cm/s for patients greater than 65 and an E/E’ ratio Ͻ 10. All NORMAL LIMITS OF LEFT VENTRICULAR DIASTOLIC FILLING GSPECT studies were required to have normal ejection fraction, normal RATE BY GATED MYOCARDIAL PERFUSION SPECT: VALIDATION wall motion, and no evidence of perfusion defects. LV PFR (end diastolic volumes/second) by GSPECT was assessed using QGS software and JA Gluck, D Desai, K Ngai, F Messineo, O Akinboboye adjusted for heart rate (calculated PFR (cPFR)).
New York Hospital Medical Center of Queens, Flushing, NY Results:
Background: Impaired diastolic function (DF) precedes systolic dysfunc-
tion in ischemic heart disease and is associated with systolic heart failure
Average E’’
(EDV/s) ؎ SD
Average E/E’’
and hypertrophic heart disease. Consequently, assessment of DF is animportant component of the evaluation of patients for cardiovascular disease. Gated single-photon emission computed tomography myocardial perfusion scintigraphy (GSPECT) can provide measurements of left ven- tricular systolic and diastolic function. However, normal diastolic filling parameters with this technique have not been fully validated. The objectiveof this study was to validate normal values for left ventricular peak filling Conclusion: This study establishes age-specific normal values for diastolic
rate (PFR) by GSPECT with traditional tissue Doppler imaging (TDI), a filling parameters by GSPECT validated with standard TDI parameters.

Source: https://www.asnc.org/imageuploads/ASNC2008%20Poster%20Session%20II.pdf