Author + information
- Received April 5, 2005
- Revision received August 30, 2005
- Accepted September 8, 2005
- Published online January 17, 2006.
- Benjamin J.W. Chow, MD, FRCPC, FACC⁎,†,⁎ (, )
- Rob S. Beanlands, MD, FRCPC, FACC⁎,†,a,
- Andrea Lee⁎,
- Jean N. DaSilva, PhD⁎,
- Robert A. deKemp, PhD⁎,
- Abdulkareem Alkahtani, MD⁎ and
- Terrence D. Ruddy, MD, FRCPC, FACC⁎,†
- ↵⁎Reprint requests and correspondence:
Dr. Benjamin J. W. Chow, University of Ottawa Heart Institute, 40 Ruskin Street, Ottawa, Ontario, Canada, K1Y 4W7.
Objectives The aim of this study was to compare treadmill exercise (TEX) and dipyridamole stress on the uptake and retention of N-13 ammonia.
Background Size and severity of stress-induced myocardial perfusion defects are clinically important. Because ammonia uptake and retention seems to be related to perfusion, viability, and metabolism, exercise stress might induce larger perfusion defects than dipyridamole stress.
Methods Twenty-six patients underwent TEX and dipyridamole stress N-13 ammonia positron emission tomography (PET). Images were assessed with a 17-segment model and a five-point score. Summed stress score (SSS), summed rest score (SRS), and summed difference score (SDS) were calculated. Left ventricular (LV) defect sizes were measured quantitatively with a 70% threshold for abnormal perfusion.
Results Compared with dipyridamole stress, TEX yielded larger SSS (9.1 ± 5.7 vs. 6.9 ± 5.9; p < 0.01), SDS (5.8 ± 4.7 vs. 3.7 ± 4.6; p < 0.02), and percentage of LV stress defect (19.3 ± 11.5% vs. 13.8 ± 13.6%; p < 0.02).
Conclusions In patients achieving adequate exercise, TEX N-13 ammonia PET myocardial perfusion imaging (MPI) yields larger stress perfusion defects than dipyridamole stress and might reflect the true myocardial ischemic burden. Treadmill exercise might be the preferred method of stress for routine N-13 ammonia PET MPI.
Size and severity of stress-induced myocardial perfusion defects are clinically important. Differences in perfusion defects with various stressors might have prognostic implications (1,2).
Positron emission tomography (PET) is traditionally performed with vasodilator stress, and the evaluation of treadmill exercise (TEX) PET myocardial perfusion imaging (MPI) is limited. A previous study comparing TEX versus dipyridamole stress rubidium-82 PET showed that defect size and severity were similar (3). Because ammonia uptake and retention seems to be related to perfusion, viability, and metabolism (glutamine pathway), exercise stress might induce larger perfusion defects than with dipyridamole stress (4–7). The objective of this study was to compare TEX and dipyridamole stress on the uptake and retention of N-13 ammonia in patients able to perform adequate exercise stress.
Between August 2003 and September 2004, 31 patients, able to perform adequate exercise stress, were prospectively referred to this single-center, randomized, single-blinded study. All patients had documented coronary artery disease (CAD) or an intermediate-to-high pre-test probability for CAD (3). Patients lacking informed consent, age <18 years, unable to exercise, or with a contraindication to dipyridamole or radiation were excluded. This study was approved by the University of Ottawa Heart Institute Human Research Ethics Board.
The order of stress was randomized (dipyridamole vs. TEX). Dipyridamole and TEX stress N-13 ammonia PET were performed on different days, and each was preceded by rest imaging. Patients abstained from caffeine, xanthine derivatives, and atrioventricular nodal blocking drugs ≥12 h and fasted (except for medications) ≥6 h before each study.
The PET images were acquired with an ECAT ART scanner (Siemens/CTI, Knoxville, Tennessee). A 4-min Cesium-137 singles transmission scan was acquired to confirm proper patient positioning and for attenuation correction (8). Ten millicuries (370 MBq) of N-13 ammonia was injected at rest and with stress. Static images were created by summing 17 min of emission data. An 8-min transmission scan was acquired post-stress for attenuation correction (Fig. 1).
The dipyridamole stress protocol has been previously described (3); N-13 ammonia was administered over 30 s, and static uptake images were created by summing 17 min of emission data.
Symptom-limited Bruce protocol was performed. During the last 1.5 min of peak exercise, N-13 ammonia was administered. Patients were immediately repositioned in the PET camera, and a 17-min emission scan was acquired and used to create static images.
Electrocardiography and image analysis
Electrocardiography analysis has been previously described (3). The PET images were assessed with a 17-segment model and a five-point grading system by two expert observers blinded to stress (2). Summed stress score (SSS), summed rest score (SRS), and summed difference score (SDS) were calculated. Images were also analyzed by a sectored analysis approach with a 70% threshold cutoff for abnormal perfusion (3). Image quality was assessed visually and quantitatively on the basis of myocardial count density (Bq/cc) and target/background ratios (3).
With SPSS version 11.5 (Chicago, Illinois), paired and independent samples of continuous variables were evaluated with the paired or unpaired ttest and non-continuous variables with the McNemar test. The concordance of the observer grading of the exercise and dipyridamole stress scans was evaluated with Kappa scores. Perfusion defects (summed scores and percentage of left ventricular [LV] defect) were correlated with a Pearson correlation coefficient and Bland-Altman plot analyses.
Thirty-one patients were referred to our study. Four patients were excluded because of left bundle branch block (n = 1), uncontrolled atrial tachyarrhythmia (n = 2), and insufficient dose of N-13 ammonia (n = 1). One patient was withdrawn from the study because of extensive severe ischemia with TEX N-13 ammonia PET. The remaining 26 patients underwent both TEX and dipyridamole stress N-13 ammonia PET MPI (Table 1).
The mean time interval between TEX and dipyridamole stress PET was 6.0 ± 5.7 days (Table 2).Peak stress heart rate and blood pressure were significantly higher with exercise than dipyridamole stress. Chest pain was more common with dipyridamole, and ST-segment depression was more common with TEX. All five patients with dipyridamole-induced ST-segment depression had abnormal N-13 PET MPI. Mean time interval from onset of N-13 ammonia infusion to initiation of emission data acquisition was 3.5 ± 0.5 min.
There was good correlation between dipyridamole and TEX stress N-13 PET (Fig. 2).Stress defect sizes (percentage of LV) with a 70% threshold for abnormal had fair correlation between stressors (Table 3).Rest defect sizes (percentage of LV) before exercise and dipyridamole stress also had fair correlation. Treadmill exercise stress produced larger and more severe induced stress (SSS) and ischemic perfusion defects (SDS) compared with dipyridamole stress. Bland Altman analysis demonstrated a bias toward larger and more severe SSS and SDS with TEX than with dipyridamole stress (Fig. 3).
The 17 patients with exercise-induced ischemic ST-segment depression had larger perfusion defects (SSS) with exercise than with dipyridamole (10.5 ± 6.3 and 8.2 ± 7.0, respectively; p = 0.03) with good correlation (r = 0.81). In the remaining nine patients without exercise-induced ST-segment depression, there was a trend toward greater exercise SSS than dipyridamole SSS (6.4 ± 3.6 and 4.3 ± 2.7, respectively; p = 0.08). Furthermore, a similar trend was noted when patients with and without exercise-induced ST-segment depression were compared (10.5 ± 6.3 and 6.4 ± 3.6, respectively; p = 0.09).
The interobserver variability of SSS showed good agreement with Kappa scores of 0.71 for exercise SSS (normal [SSS <4] or abnormal [SSS ≥4]) and 0.74 for dipyridamole SSS.
Patients were categorized as having disease in the left anterior descending artery, left circumflex, and/or right coronary artery territories. Kappa analysis demonstrated fair correlation between exercise and dipyridamole stress in each territory (left anterior descending artery = 0.57, left circumflex artery = 0.69, and right coronary artery = 0.54; p < 0.01). Patients were also categorized as normal or single-vessel versus multivessel disease. Four of the 12 (33%) patients initially categorized as having normal or single-vessel disease with dipyridamole stress had multivessel disease with TEX N-13 ammonia PET. Furthermore, patients were categorized into normal to mildly abnormal SSS (≤8) and moderate to severely abnormal SSS (>8). Five of the 18 (28%) patients with dipyridamole-induced normal or mildly abnormal SSS had moderate to severely abnormal SSS with TEX.
Most exercise stress PET MPI studies have focused on supine bicycle (9–11). Though TEX has been successfully performed with rubidium-82 and FDG PET MPI (3,12), this is the first study to combine TEX with N-13 ammonia PET.
In patients who achieve adequate exercise, TEX N-13 ammonia PET seems to induce larger and more severe stress and ischemic perfusion defects than dipyridamole stress. Retention of N-13 ammonia is adenosine triphosphate-dependant and requires the metabolism of ammonia to glutamine (13). Thus, uptake and retention might be altered by changes in metabolic state (14,15). Exercise-induced ischemia might cause metabolic stunning, reduce the metabolism of ammonia to glutamine, and attenuate N-13 ammonia retention in areas of ischemia.
Exercise MPI has been compared with various vasodilator stressors (4,16–18) Comparing exercise, dipyridamole, and adenosine stress Tc-99m single-photon emission tomography (SPECT), exercise resulted in greater defect extent, severity, and reversibility than dipyridamole but no differences in defect size between exercise and adenosine (17). Treadmill exercise SPECT stress defects were larger than with dipyridamole in 45% of patients (16). Larger exercise stress defects were typically observed in patients with higher heart rates (16). Nishimura et al. (19) showed that adenosine stress induced larger defect sizes than exercise Tl-201 SPECT. Abe et al. (4) compared adenosine and TEX Tl-201 MPI and found no difference in patients who were able to perform adequate exercise.
Though our study demonstrated that TEX provoked larger and more severe perfusion defects than dipyridamole stress, this was more apparent in patients with mild to moderate SSS and SDS with dipyridamole stress. This finding might be explained by: the high level of exercise achieved in our patients, which might have resulted in more ischemia; the superior accuracy of N-13 ammonia PET that might have enabled the detection of differences between exercise and dipyridamole; and the rapid return of flow to normal after exercise compared with a dipyridamole-aminophylline protocol, which might delay N-13 ammonia washout from normal tissue compared with ischemic tissue, thus amplifying the defect size.
Five patients having normal to mildly abnormal SSS with dipyridamole were reclassified as having moderate to severely abnormal SSS with TEX. Similarly, four patients classified as normal or single-vessel disease with dipyridamole stress had evidence of multivessel disease with TEX N-13 ammonia PET. These differences in exercise and dipyridamole stress SSS, SDS, and defect size might be clinically relevant and might influence patient management.
Although TEX does not permit the benefits of dynamic acquisition for the quantification of blood flow, it does provide information about exercise tolerance and is preferred by most patients (3).
Resting myocardial/lung ratio was significantly different before exercise and dipyridamole stress. Because resting target/background ratios should not differ in the same patient, we believe that this finding likely occurred by chance.
The successful performance of the TEX stress N-13 ammonia PET required the coordination of the cyclotron laboratory and the imaging team. The completion of N-13 ammonia synthesis and its delivery occurred simultaneously with the initiation of TEX. A surplus of N-13 ammonia was manufactured (2.5 times the injected dose) to ensure that an adequate quantity of radiotracer was available at peak exercise.
Our patient population was predominantly male, and our small sample size is prone to patient-related systematic errors and was not powered to detect significant differences in the subgroup analysis. Coronary angiography was not routinely performed to validate the accuracy of TEX versus dipyridamole N-13 ammonia PET and warrants further investigation.
In patients achieving adequate exercise, TEX N-13 ammonia PET MPI yields larger and more severe defects than dipyridamole stress. Images acquired after TEX might reflect the true ischemic burden. Treadmill exercise might be the preferred method of stress for routine N-13 ammonia PET MPI.
The authors extend their gratitude to the staff at the National Cardiac PET Centre for their technical expertise.
↵a Dr. Beanlands was supported by the Canadian Institute of Health Research and the Ontario Premier’s Research Excellence Award.
This research was supported in part by the Ontario Research and Development Challenge Fund (#00-May-0710) for the Ontario Consortium of Cardiac Imaging.
- Abbreviations and Acronyms
- coronary artery disease
- left ventricle/ventricular
- myocardial perfusion imaging
- positron emission tomography
- summed difference score
- single-photon emission tomography
- summed rest score
- summed stress score
- treadmill exercise
- Received April 5, 2005.
- Revision received August 30, 2005.
- Accepted September 8, 2005.
- American College of Cardiology Foundation
- Hachamovitch R.,
- Berman D.S.,
- Shaw L.J.,
- et al.
- Chow B.J.W.,
- Ananthasubramaniam K.,
- deKemp R.A.,
- Dalipaj M.M.,
- Beanlands R.S.B.,
- Ruddy T.D.
- Schelbert H.R.,
- Phelps M.E.,
- Huang S.C.,
- et al.
- Rosenspire K.C.,
- Schwaiger M.,
- Mangner T.J.,
- Hutchins G.D.,
- Sutorik A.,
- Kuhl D.E.
- Krivokapich J.,
- Smith G.T.,
- Huang S.C.,
- et al.
- Camici P.,
- Araujo L.I.,
- Spinks T.,
- et al.
- Rauch B.,
- Helus F.,
- Grunze M.,
- et al.
- Bergmann S.R.,
- Hack S.,
- Tewson T.,
- Welch M.J.,
- Sobel B.E.
- Nishimura S.,
- Mahmarian J.J.,
- Boyce T.M.,
- Verani M.S.