The CORONARY AR’fERY DISEASE NOVEMBER 1, 1993, VOL. 72,NO.14

Assessment of Left Ventricular Wall Motion in Angina Pectoris by Two-Dimensional

Echocardiography and Myocardial Perfusion by Technetium-99m Sestamibi Tomography During Adenosine-Induced Coronary Vasodilation and

Comparison with Coronary Angiography Aman M. Amanullah, MD, Sture Bevegird, MD, PhD,

Kaj Lindvall, MD, PhD, and Mikael Aasa, MD

Myocardial perfusion and regional wall motion dulc ing adenosine-induced coronary vasodilation were assessed in 40 patients with angina pectoris by technetiumSSm sestamibi singlephoton emission computed tomography (SPECT) and simultaneous 2dimensional echocardiography. Adenosine was infused intravenously at a dose of 240 ug/kg body weight per minute for 6 minutes, and tech netiunr65m sestamibi was injected at 3 minutes. Adenosine caused a significant decrease in sys tolic and diastolic blood pressure and a significant increase in heart rate and the heart rat+blood pressure product. Adverse effects were mild and transient and no patient required aminophylline. Completely or partially reversible defects on SPECT were present in 28 patients, a f’iied defect was seen in 4 patients, and no defect was seen in 6 patients. Two-dimensional echocardiography re- vealed a new or worsening wall motion abnormali- ty in 21 patients, a fixed abnormality in 4 patients and no abnormality in 15. Transient perfusion de fects were associated with transient wall motion abnormalities in 71% of cases. The overall sensi- tivity, specificity and predictive accuracy of adenosine echocardiography in detecting signifi- cant coronary artery disease (>50% diameter stenosis) were 74,100 and 76%, respectively, whereas those of adenosine SPECT were 64,100 and 95%, respectively (p 40.05, NS, and 40.05, re- spectively). Thus adenosine technetium99m ses-

From the Department of Cardiology and the Department of Clinical Physiology, Karolinska Institute at South Hospital, and the Department of Cardiology, Karolinska Hospital, Stockholm, Sweden. Manuscript received April 20,1993; revised manuscript received and accepted June 14, 1993.

Address for reprints: Aman M. Amanullah, MD, Department of Cardiology, South Hospital, S-l 18 83 Stockholm, Sweden.

tamibi SPECT has a higher sensitivity and predie tive accuracy than adenosine echocardiography, suggesti~ that adenosine-induced perfusion de- fects are not always associated with wall motion abnormality. Although the principal underlying mechanism of myocardial ischemia during aden* sine infusion is a “coronary steal” phenomenon, the ischemia is also due to an enhanced myocap dial oxygen demand, as indicated by an increased rat~essure product.

(Am J Cardiol1653;72:963-669)

P harmacologic stress testing in conjunction with 2- dimensional echocardiography and myocardial perfusion scintigraphy is now an accepted alter-

native to the traditional dynamic exercise test to detect myocardial ischemia, especially in patients who are un- able to perform adequate exercise. Previous studies have demonstrated the clinical usefulness of dipyridamole thallium scintigraphy and dipyridamole 2-dimensional echocardiography in the diagnosis of coronary artery disease (CAJI).lJ The coronary vasodilator effect of di- pyridamole is the indirect result of increased interstitial adenosine due to decreased cellular adenosine uptake. Adenosine is a potent coronary vasodilator, and intrave- nous adenosine given at a rate of 140 pg/kg/min pro- duces nearly maximal coronary vasodilation at approxi- mately 120 seconds after the start of infusion.5 The po- tential advantage of adenosme over dipyridamole is its extremely short half-life (~10 seconds) and a rapid on- set of action. Technetium-99m sestamibi has been wide- ly used in the diagnosis of CAD in patients with stable angina pectoris.6,7 Imaging using single-photon emission computed tomography (SPECT) has considerably im- proved the diagnostic yield.*s9 Technetium-99m sesta- mibi is distributed proportionally to the blood flow throughout the myocardium and, in contrast to thallium-

ADENOSINE-INDUCED CORONARY VASODILATION 983

TABLE I Demographic Characteristics of the Study Patients (n = 40)

Age (years) Gender (m/f) Cigarette smoker Diabetes mellitus Systemic hypertension (history) Healed myocardial infarction (history)

Q-wave infarction Stable angina pectoris Recent episode of unstable angina Medication

p blocker Long-acting nitrate Calcium antagonist Aspirin Digitalis

61 rt 10 (range 44-80) 3218

9 (23%) 5 (13%)

12 (30%) 12 (30%)

5 (13%) 16 (40%) 24 (60%)

31 (78%) 27 (68%) 20 (50%) 31 (78%)

0 (0%)

201, there is minimal redistribution.1o Stress SPECT with technetium-99m sestamibi can therefore be per- formed at a time long after the injection of the isotope. These properties make it useful even in patients with acute ischemia, e.g., those with acute myocardiaJ ischemia and in patients with spontaneous chest pain.11-13 Con- flicting findings regarding the diagnostic value of adeno- sine echocardiography have recently been reported.lL16 Also, there are only a few published studies that involve direct comparison of adenosine-induced myocardial per- fusion defects and wall motion abnormalities.14,i6 The present study was undertaken to assess the diagnostic accuracy of adenosine echocardiography and adenosine technetium-99m sestamibi SPECT for detecting CAD in patients with angina pectoris and to directly correlate the segmental perfusion defects detected by SPECT with wall motion abnormalities by echocardiography.

METHODS Subjects (Table I): The study group consisted of 40

patients who had been referred for coronary angiogra- phy. Twenty-four had a recent episode of unstable angi- na which was stabilized with medical treatment and 16 had a history of stable, exertional angina pectoris. Un- stable angina was de&-red as (1) new-onset angina of cl- month duration at low exertion, (2) crescendo angina (increased frequency and duration, more severe or pre- cipitated at lower work loads than previously in patients

2 -DE 2 - DE, ECG SPECT SPECT ECG f-HR, BP:

I I ! I I I

3 min g min

FIGURE 1. Study protocol for the ecquisition of adenosine echocerdiographic and myocardial perfusion images. BP q

blood pressure; 20E q 2-dimensional echocardiography; ECG q echocardiography; HR = heart rate; SPECT q single- photon emission computed tomography; T c+estamibi=teclk netium-BBm sestamibi.

with prior stable angina), and (3) angina at rest. None of the patients had asthma or chronic obstructive pul- monary disease and none were taking theophyllamine- containing medication. Patients with a second- or third- degree atrioventricular block and left bundle branch block were not included in the study. Myocardial perfusion imaging (SPECT) and simultaneous 2dimensional echocar- diography during adenosine infusion were performed in all patients within 6 & 15 days of the coronary angiog- raphy. No patient had had any cardiac event during this interval. Thirty-nine patients had also undergone symp- tom-limited exercise testing. The study was approved by the local ethical committee. The informed consent of all patients was obtained before their participation.

Coronary angiography: All coronary angiograms were recorded according to the Judkins technique and were analyzed by 2 observers who were unaware of the clinical, echocardiographic or scintigraphic data. Signif- icant CAD was detined as 250% diameter narrowing of a major coronary artery or one of its major branches. Left main coronary artery stenosis was regarded as 2- vessel CAD.

Exercise test: Patients performed a symptom-limit- ed exercise test sitting on a bicycle ergometer. A 1Zlead electrocardiogram and blood pressure were obtained at baseline and at each minute of exercise. A positive exer- cise electrocardiogram was defined as 21 mm of hori- zontal or downsloping ST-segment depression 80 ms af- ter the J point.

Adenosine infusion protocol (Figure 1): On the morning of the test, none of the patients consumed cof- fee, tea or caffeine-containing products; medications containing P-blocking agents and calcium antagonists were withheld. A baseline 1Zlead electrocardiogram, heart rate, and blood pressure were recorded. Adenosine was then infused at a rate of 140 ~.@k&nin through an antecubital vein with an infusion pump and was con- tinued for 6 minutes. At the end of the third minute of infusion, 500 MBq of technetium-99m sestamibi (Car- diolite8) was injected intravenously through a second antecubital v&n. A 12-lead electrocardiogram was mon- itored continuously and heart rate, blood pressure and electrocardiograms were recorded every minute during the adenosine infusion.

Adenosine echocardiography: Two-dimensional echocardiography at standard projections (parasternal long- and short-axis and apical 4- and 2-chamber views) were recorded at baseline and during the adeno- sine infusion and about 15 minutes after infusion using a commercially available system (Acuson 128XP/lO). Imaging was performed alternating between apical 4- and 2-chamber views and parastemal long- and short- axis views. The images were stored on videotapes and selected images at baseline and after 23 minutes of adenosine infusion were digitized in an off-line analysis station and displayed on a 4-quadrant screen using a continuous cineloop format (Cine’view Freelands Sys- tem). For purposes of analysis, the left ventricle was di- vided into 5 segments: anterior, lateral, inferior, septum and apex. The wall motion of each segment was as- sessed and scored, and a wall motion score index was derived as previously described.” The results of the wall

984 THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72 NOVEMBER 1,1993

motion analysis were classified as follows: normal = motion analysis for rest and stress images was assessed baseline normal wall motion that remains unchanged or in 32 randomly selected studies and was small (r = 0.91, increases during the adenosine infusion; ischemic = p <O.OOOl and r = 0.93, p <O.OOOl, respectively). Data worsening or new-onset wall motion abnormality during concerning echocardiography reported in this study are adenosine infusion; scar = baseline wall motion abnor- those of the first observer. All the images were inter- mality that does not change during the adenosine infu- preted without awareness of the clinical data and coro- sion. The inter- and intraobserver variability in the wall nary angiography or SPECT imaging results.

FIBURE 2. Bull’aeye (top) and midcaviby tomo@e@ic slii (bottom) obtained with adenosine singbphoton emission computed tomogrephic imaging (stress and delayed) in a patient with severe left circumffex artery stenssis. Reversible per- timIon defects are present in the lateral segments.

ADENOSINE-INDUCED CORONARY VASODILATION 985

TABLE II Hemodynamic Changes During Adenosine Infusion TABLE IV Comparison of Adenosine and Exercise Testing

Peak Baseline Adenosine p Value

Heart rate (HR) (beats/min) 61 f 9 81 f 15 <O.OOl Systolic BP (mm Hg) 142 f 22 125 z! 19 <O.OOl Diastolic BP (mm Hg) 85 t 13 77 f 14 <O.OOl Rate-pressure product 86 +- 18 101 k 25 < 0.001

(HR x systolic BP x 10-2)

BP = blood pressure.

Adenosine Exercise (n = 40) (n = 39) p Value

Peak HR (beatslmin) 81 f 15 122 -+ 26 <0.0001 Changes in HR* 20 + 10 57 2 25 <0.0001

,Peak systolic BP 125 t 19 176 or 26 <0.0001 Rate-pressure product 101 2 25 217 -+ 68 < 0.0001

(Peak HR x BP x 10-Z) Chest or throat pain 12 (30%) 23 (59%) <O.Ol Positive ECG response 18 (45%) 27 (69%) < 0.05 *Changes from baselrne to peak value. BP = blood pressure; ECG = electrocardiographic: HR = heart rate.

TABLE Ill Cardiac and Noncardiac Adverse Effects During Adenosine Infusion in 40 Patients

Symptoms

Dyspnea Flushing Chest or throat pain Headache Nausea No symptoms ST depression > 1 mm First-degree AV block Second- or third-degree AV block

Av = atrioventricular.

Number of Patients (%)

17 (43) 16 140) 12 (30)

6 E15) 1 (3) 7 (18)

18 (45) 1 (3) 0 (0)

Adenosine technetium99m sestamibi tomom phy: SPECT images were obtained 1 to 2 hours after injection of technetium-99m sestamibi. The injection of the isotope (700 MBq) was repeated 24 hours after the initial study and image acquisition was repeated. Using a large field-of-view gamma camera (General Electric 400 AC), tomographic image acquisition was effectuat- ed over a 180” arc, from the 45” right anterior oblique to the 45” left posterior oblique position. Oblique or- thogonal tomographic slices, each 6 mm thick, were generated parallel to the vertical and horizontal long and short axes of the left ventricle. From the short-axis slices, extending from the base to the apex, a bull’s-eye polar coordinate map was reconstructed.*8J9

Qualitative analysis of the tomographic images: The SPECT images were analyzed by 2 experienced ob- servers without knowledge of the coronary angiography, echocardiographic studies, or clinical history of the pa- tients. Any difference was mediated by consensus. The left ventricle was divided into 5 segments correspond- ing to the anterior wall, lateral wall, inferior wall, sep- tum and apex. Technetium-99m sestamibi uptake in each segment was classified as normal and mildly, moderate- ly or severely reduced or absent. Tomograms were also analyzed for the presence of complete, partial or no re- versibility.

Computer quantification of tomographic images (Figure 2): The polar maps for each patient were com- pared statistically with a data bank of 25 normal sub- jects. Myocardial pixels with counts below 2.5 SDS from the mean values of the corresponding pixel in the nor- mal data bank were considered abnormal. Focal defects involving 24% of the total left ventricular pixels were considered abnormal.

The site of wall motion abnormality by echocardiog- raphy and myocardial perfusion abnormality by SPECT and their relation to coronary arteriography were defmed as follows: septal and anterior segments corresponded to the left anterior descending coronary artery, lateral seg- ment to the left circumtlex artery, and the inferior seg- ment to the right coronary artery. The vascular distribu- tion of the apical segment was considered separately in each case, depending on defects in other territories.

Statistical analysis: Data are presented as mean rt SD. Comparisons between groups of continuous vari- ables for paired and unpaired data were made using Wilcoxon’s signed-rank test and the Mann-Whitney test. Frequency comparisons were made by &i-square anal- ysis. Fisher’s exact tests were used when appropriate.

RESULTS Hemodynamic effects of intravenous adenosine

(Table II): Adenosine caused a significant increase in heart rate and a significant decrease in systolic and dia- stolic blood pressures. The rate-pressure product, reflect- ing myocardial oxygen demand, increased signiticantly during adenosine infusion.

Adverse effects (Table Ill): Although most patients tolerated the adenosine infusion well, many of the pa- tients experienced adverse effects, most of which were mild and transient in nature. No patient required treat- ment with aminophylline and in only 1 patient was adenosine infusion terminated 90 seconds prematurely because of chest pain and throat and neck tightness. One patient developed tirst-degree atrioventricular block. In 4 patients with baseline tirst-degree atrioventricular block, no further increase in the PR interval occurred.

Comparison OF herodynsmi chim@s duting aderr osine and exercise testing (Table IV): Heart rate, blood pressure and rate-pressure product were significantly higher during exercise than during adenosine infusion. Also, many more patients had chest pain and ST de- pression during the exercise test.

Adenoslne echwwdiographyandadenceineteck netill--blsPEcTandcompadsonwitb

mglo&@y . The baseline echocardiogram (either in an apical or a parastemal projection) was tech- nically adequate for regional wall motion analysis in 196 segments (98%). In 36 patients, all the segments were analyzable, whereas in the remaining 4 patients, 4 of 5 segments in each patient were visualized. During adeno- sine infusion, 192 segments (96%) were technically suit-

986 THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72 OCTOBER 15,1993

able for analysis; in 32 patients, all the segments, and in the remaining 8 patients 4 of 5 segments in each patient were adequate for wall motion analysis. Twen- ty-one of 40 patients had an ischemic response during adenosine infusion, 4 had a fixed wall motion abnormal- ity, and 15 had a normal response. Significant angio- graphically verified CAD was present in 34 of 40 pa- tients; 20 had 3-vessel disease, 10 had 2-vessel disease, and 4 had l-vessel disease. Left main CAD was present in 9 patients. Abnormal perfusion scans were present in 32 of 40 patients. Completely reversible defects were present in 12 patients, partially reversible in 16, and 4 patients had iixed defects. The overall sensitivity, speci- ficity and predictive accuracy of adenosine echocardiog- raphy in detecting significant CAD were 74, 100 and 78%, respectively, whereas those of adenosine SPECT were 94, 100 and 95%, respectively (p ~0.05, NS, and ~0.05, respectively) (Table V). When patients with fixed wall motion abnormalities or tixed perfusion defects were excluded, the sensitivity of adenosine echo- cardiography and adenosine SPECT decreased to 62 and 82%, respectively. Adenosine echocardiography had a sensitivity of 48, 48, and 60%, respectively, for identi- fying stenoses of the left anterior descending, left cir- cumflex and right coronary arteries, while the corre- sponding results were 41, 67 and 76% (all p = NS) for adenosine SPECT.

Comparison of adenosine technetiunMSm sestz+ mibi SPEW with atienosine echocardiography (Fig- ure 3): Among 28 patients with complete or partial reversible perfusion defects during adenosine infusion, 20 (71%) had either new-onset wall motion abnormality or further deterioration of the preexisting abnormality. In most instances, the wall motion abnormalities and per- fusion defects in these 20 patients involved the same myocardial region. All patients with normal SPECT also had normal 2-dimensional echocardiography. Among 12 patients with a history of prior myocardial infarction, per- fusion defects on SPECT (fixed or partially reversible defects) were present in all patients. Abnormal wall motion on the echocardiogram during adenosine infusion (tixed abnormality worsening, or new-onset abnormality) occurred in 11 of these patients. There was a complete or partial correlation of the location of the wall motion abnormality on the echocardiogram with the site of the perfusion defects at SPECT in all of these 11 patients.

DISCUSSION The usefulness of exercise echocardiography and ex-

ercise myocardial perfusion scintigraphy is now well es- tablished.9J7q20 In patients who are unable to perform an adequate exercise test, an alternative form of stress test- ing is required. The present study is concerned with the evaluation of the diagnostic accuracy of adenosine echocar- diography and adenosine technetium-99m sestamibi scintigraphy in detecting significant CAD and to deter- mine whether myocardial ischemia, as demonstrated by wall motion abnormality detected on 2dimensional echocardiography, occurs in association with perfusion defects induced by standard adenosine infusion.

Dwim-deor#rary =W-byph-- IO@ coronary vasadilation: Pharmacologic agents ad-

TABLE V Comparative Sensitivity, Specificity, Predictive Value and Accuracy of Adenosine Echocardiography and Adenosine Technetium-Sestamibi SPECT in Detecting Significant Coronary Artery Disease

Adenosine Adenosine Tc-Sestamibi

Echocardiography SPECT p Value

Sensitivity 74%(25/34) 94% (32/34) <0.05 Specificity 100% (6/6) 100% (6/6) NS Positive predictive value 100% (25/25) 100% (32/32) NS Negative predictive value 40% (6/E) 75% (6/8) NS Predictive accuracy 78% (31/40) 95% (38/40) 10.05

SPECT = single-photon emission computed tomography; Tc = technetium.

ministered in conjunction with an imaging modality in stress testing include sympathomimetic drugs such as dobutamine, and coronary vasodilators such as dipyrid- amole and adenosine. The principal underlying mecha- nism of myocardial perfusion defects during adenosine- induced coronary vasodilation in patients with CAD is the disparity in regional myocardial perfusion between vascular territories supplied by normal coronary arteries and those supplied by stenotic vessels. Adenosine in- duces coronary arteriolar vasodilation which produces a three- to fivefold increase in blood flow in normal ves- sels with little or no increase in stenotic vessels.5J4 True myocardial ischemia, as suggested by angina, ST-seg- ment depression on electrocardiogram, and wall motion abnormality on echocardiography, therefore does not occur in all patients with adenosine-induced perfusion defects as supported by our present study and a previ- ous study.14

Adenosine technetiunr99m sestamibi SPECT: The present study is in agreement with previous reports of adenosine-induced coronary vasodilation combined with myocardial perfusion imaging,14j21 and shows that this is a highly sensitive, specific and accurate technique for detecting angiographicahy verified significant CAD. In these studies adenosine-induced perfusion defects were detected by thallium-201 imaging, which, unlike tech-

2 - D Echo

I I+S S N

FIGURE 3. Comparison of adenosine tschnetium99m ses tamibi (Tc-ssstamibi) singlephoton emission computsd tomography (SPECT) with adenosine 2dimensioaal echocalc dio&aphy (2-D Echo). I q ischemia; N = normal; S = scar.

ADENOSINE-INDUCED CORONARY VASODILATION 987

netium-99m sestamibi, has a potential for early redistri- bution and imaging must be performed soon after injec- tion of the radiotracer. Technetium-99m sestamibi un- dergoes minimal redistribution, and tomographic image acquisition performed several hours after injection of the isotope does not lead to underestimation of the presence and extent of the CAD. Although the primary mecha- nism of ischemia during adenosine infusion is a “coro- nary steal,” an increased myocardial oxygen demand, as indicated by the increase in the rate-pressure product observed in our study, further enhances the development of true myocardial ischemia in some patients with severe CAD.

Adenosine echocardiography: The sensitivity of adenosine echocardiography reported in this series was lower than the sensitivity of exercise echocardiography reported in previous studies,17,20 demonstrating that adenosine-induced myocardial ischemia occurs less fre- quently than exercise-induced ischemia. The lower sen- sitivity of adenosine echocardiography compared with adenosine SPECT also demonstrates that wall motion abnormality, a marker of myocardial ischemia, does not necessarily occur in all patients with perfusion defects caused by flow maldistribution in different myocardial regions. However, our lindings with adenosine echocar- diography are comparable to the sensitivity of high-dose dipyridamole echocardiography reported by Picano et alF2 The dose of adenosine used in the present study was 140 pg/k&nin. At this dose, maximal coronary vasodilation occurs in most (85%) but not all patients,5 and at a higher dose of adenosine, sensitivity in detect- ing CAD may be further enhanced. However, this must be balanced against the risk of an increased incidence of adverse effects. Jain et al4 demonstrated that partial- ly or completely reversible defects, as detected by thalli- um-201 SPECT after dipyridamole, were followed by a transient wall motion abnormality in 87.5% of patients. In the present study, adenosine-induced partially or to- tally reversible perfusion defects were associated in only 71% of cases with a transient wall motion abnormality. Nguyen et al,14 in a study of the diagnostic accuracy of adenosine thallium-201 SPECT in a group of 60 pa- tients, performed echocardiography in a subgroup of 25 patients, and found a lower sensitivity of adenosine echocardiography than reported in the present series. In their study, 20% of the myocardial segments were ex- cluded because of inadequate image quality, echocar- diographic images at baseline and during adenosine were not compared side by side for accurate interpreta- tion, and patients with a wall motion abnormality at baseline that did not change during adenosine infusion were considered normal. Moceover, there was only a 13% incidence of adenosine-mduced ST-segment de- pression in patients studied by Nguyen et al, compared with 45% in our series, indicating that the patients in their series may have had less severe CAD. Zoghbi et alI5 reported an overall sensitivity of 85% in detecting significant CAD as shown by a scar or ischemia, a fig- ure that is higher than in our study. A high percentage (37%) of their patients had fixed wall motion abnormal- ities compared with 10% in the present study. Also, sig- nificant CAD was defined as >75% stenosis compared

with ~50% stenosis in the present study. Marwick et ali6 reported a study in which they compared the results of 2dimensional echocardiography and technetium-99m sestamibi SPECT using both adenosine and dobutamine stress in a group of patients without previous myocar- dial infarction. The sensitivity of adenosine echocardi- ography was 58%, and was signiticantly less than that of adenosine SPECT, which was 86%. These results are comparable with our findings (62 and 82%, respective- ly) if we exclude patients with fixed wall motion abnor- malities or lixed defects.

Study limitations: The patient group studied had a high prevalence of CAD, thus enhancing the sensitivity. The results may have been different if a cohort of pa- tients with less severe CAD and more normal subjects had been included. There were only 4 patients with l- vessel CAD. The major advantage of stress perfusion imaging over stress echocardiography is detection of l- vessel disease. With more patients included with l-ves- sel CAD, the advantage of adenosine SPECT over adenosine echocardiography would have been more ap- parent. The sensitivity in detecting individual coronary artery stenoses was relatively low. This may be partly due to the extensive variability in the myocardial regions supplied by a coronary artery and the development of collateral vessels in the area of a stenotic or occluded vessel. The degree of coronary artery stenosis was based on visual estimations. Although this is still the method that is routinely used, it has several limitations.23 Quan- titative coronary angiography may have reflected the relationship of individual diseased vessels with the loca- tion, extent and severity of perfusion defects and wall motion abnormalities better. The high specificity in our study was based on only 6 patients with normal coro- nary angiograms. Also, only 8 women were included in the patient cohort. Attenuation artefacts from overlying breast tissue decrease the specificity of sestamibi. It is likely that in a larger group of patients including more women without CAD, the specificity of adenosine SPECT would be lower.

Acknowledgment: We are grateful to Ann-Christine Johansson, BS, Lena Schager, BS, and Astrid Femstedt for expert technical assistance.

1. Iskandrian AS, Heo J, Askenase A, Segal BL, Auerbach R. Dipyridamole car- diac imaging. Am Heart J 1988;115:432-443. 2. Rossen JD, Simonetti I, Mamns ML, Wiiford MD. Coronary dilation with standard dose dipyridamole and dipyridamole combined with handgrip. Circulation 1989;79:566-572. 3. DePuey EG, Guatler-Krawczynska E, D’Amato PH, Patterson RE. Thallium- 201 single photon emission computed tomography with intravenous dipyridamole to diagnose coronary artery disease. Coronary Artery Disease 1990;1:75-82. 4. Jain A, Suaw J, Mahmtian JJ, Zoghbi WA, Quinones MA, Verani MS. Func- tional significance of myocardial perfusion defects induced by dipyiidamole using dxdlium-201 single-photon emission computed tomography and two-dimensional echocardiography. Am J Cardiol 1990;66:802-806. 5. Wilson RF, Wyche K, Christensen BV, Zimmer S, Laxson DD. Effects of adeno- sine on human coronary arterial circulation. Circulation 1990;82:1595-1606. 6. Iskandrian AS, Heo J, Kong B, Lyons E, Marsh S. Use of technetium-99m isoni- bile (RP-30A) in assessing left ventricular perfusion and function at rest and dur- ing exercise in coronary artery disease, and comparison with coronary ateriography and exercise thallium-201 SPECT imaging. Am J Cardiol 1989;64:270-275. 7. Kiat H, Maddahi J, Roy LT, Van Train K, Friedman J, Rear K, Berman DS. Comparison of technetium99m methoxy isobutyl isonitrile and thallium-201 for

988 THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72 OCTOBER 15,1993

evaluation of coronary artery disease by planar and tomographic methods. Am Heart J 1989;117:1-11. 6. Fmtel DJ, Links JM, Brinker JA, Frank TL: Parker M, Becker LC. Improved diagnostic performance of exercise thallium-201 single photon emission computed tomography over planar imaging in the diagnosis of ischemic heart disease: a receiv- er operating characteristic analysis. J Am CoU Cardiol 1989;13:600-612. 9. Amanullah A, Beve&d S, Lindvall K, Herlitz E. Early exercise thallium-201 single photon emission computed tomography in unstable angina: a prospective study. Clin Physiol 1992; 12:607617. 10. Okada RD, Glover D, Gaffney T. Myocardial kinetics of technetium-99m-hex- skis-2-methoxy-2-methylpropyl-isoninile. Circulation 1988;77:491-498. 11. Gibbons RJ, Verani MS, Behrenbeck T, Pellia PA, O’Connor MK, Mah- marian JJ, Chesebro JH, Wackers FJ. Feasibility of tomographic 99mTc-hex&is-2- methoxy-2.methylpropyl-isonihile imaging for the assessment of myocaxlial area at risk and the effect of treatment in acute myocardial infarction. Circulation 1989;80:1277-1286. 12. Bilodeau L, Th&oux I’, Grkgoire J, Gagnon D, Arsenault A. Technetium-99m sestamibi tomography in patients with spontaneous chest pain: correlations with clinical, electrocaxliographic and angiograpbic findings. JAm Coil Car&l 1991;18: 1684-1691. 13. Grkgoire J, Th&oux P. Detection and assessment of unstable angina using myocardial perfusion imaging: comparison between technetium-99m sestamibi SPECT and 12-lead electrocardiogram. Am J Cardiol 1990;66:42E-46E. 14. Nguyen T, Heo J, Ogilby JD, Iskandrian AS. Single photon emission comput- ed tomography with thallium-201 during adenosine-induced coronary hyperemia: correlation with coronary tieriography, exercise thallium imaging and two-di- mensional echocardiography. J Am CoN Cardiol 1990;16:1375-1383. 15. Zoghbi WA, Cheirif J, Kleiman NS, Verani MS, Trakhtenbroit A. Diagnosis

of ischemic heart disease with adenosine echocardiography. J Am Cull Cardiol 1991;18:1271-1279. 16. Marwick T, Wilemart B, D’Hondt A, Baudhuin T, Wijns W, De&y J, Melin J. Selection of the optimal nonexercise stress for the evaluation of ischemic region- al myocardial dysfunction and malperfusion. Comparison of dobutamine and adeno- sine using echocardiography and 99*Tc-MJBI single photon emission computed tomography. Circulation 1993;87:345-354. 17. Amanullah AM, Lindvall K. Predischarge exercise echocardiography in patients with unstable angina who respond to medical treatment. C/in Cardiol 199215: 417-423. 18. Garcia EV, Van Train K, Maddahi J, Prigent F, Friedman J, Areeda J, Wax- man A, Berman DS. Quantification of rotational thallium-201 myocudial tomog- raphy. J Nucl Med 1985;26:17-26. 19. DePasquale EE, Nody AC, DePuey EG, Garcia EV, Pilcher G, Bredlau G, Gober A, Gruentzig A, Berger HJ. Quantitative rotational thallium-201 tomography for identifying and localizing coronary artery disease. Circulation 1988;77:316-327. 20. Marwick TH, Nemec JJ, Pastiow FJ, Stewart WJ, Salcedo EE. Accuracy and limitations of exercise echocardiography in a routine clinical setting. J Am Coil Cardiol 1992;19:7&81. 21. Gupta NC, Es&brooks DJ, Hil leman DE, Mohiuddin SM. Comparison of adenosine and exercise thallium-201 single-photon emission computed tomography (SPECT) myocardial petfusion imaging. J Am Co/l Cardiol 1992;19:248-257, 22. Picano E, Lattanzi F, Ma&i M, Distante A, L’Abbate A. High dose dipyrid- amole echocardiography test in effort angina pectoris. J Am Coil Car-dial 1986;8: 848-854. 23. White CW, Wright CB, Doty DB, Hiiatza LF, Eastham CL; Harris DE, Mar- cus ML Does visual interpretation of the coronary aaeriogram predict the physio- logic importance of a coronary stenosis? N Engl J Med 1984;310:81%824.

ADENOSINE-INDUCED CORONARY VASODILATION 989