The CORONARY AR’lERY DlSEASE American NOVEMBER 1, 1994, VOL. 74, NO. 9

/ Journal

Effects of Nitroglycerin by TechnetiunMSm Sestamibi Tomoscintigraphy on Resting Regional

Myocardial Hypoperfusion in Stable Patients with Healed Myocardial Infarction

Michele Galli, MD, Claudia Marcassa, MD, Alessandro Imparato, MD, Riccardo Campini, MD, Pedro Silva Orrego, MD, and Pantaleo Giannuzzi, MD

My&Sal sestamibi uptake reflects regional flow distribution and cellular i~grity; however, some wgments showhtg reduced tracer uptake at rest may consist of viable, atthoum hypoper fused, myocardhrm. lt is speculated that the administration of nitroglycerin (NTG) before the sestamibi injection wouki improve the tracer uptake in restia hypopeHused regions. Thirty- six stable patients with previous myocardial infarction (56 * 2 years; mean ejection fraction 42 + 2%), in whom perfusion defects coukl be seen at restiN sestamibi -Y, repeated the sdntiik study 2 to 6 days iater, receiv- ing N7G (0.3 to 0.6 mg sublingually) before the tracer injection. lhe size of the tracer uptake defect was quantified from circumferential pro- files in 3 sho&axis f&es by integrating the area below the lower normal limit (mean - 2 SD). After NTG, the mean perfusion defect significant- ly decreased (from 6,324 f 619 to 6,365 + 516, p 40.01). The defect was reduced beyond the reproducibility limits in 20 patients (56%, group 1) and was unchanged or iB in 16 (440/o, group 2). The resting sestamibi defect size was comparable between the 2 @oups. The average percent reduction of the perfusion defect after NT0 was 29 + 4% (range 7 to 74). The perfusion defects that improved after N7G were associated with a less severe 2dimensional e&ocardio- graphic regional wall motion score (2.1 + 0.1 vs 2.6 f 0.1 in segments showhtg fixed defect, p *O.OOl), a lower rate of patency (37% vs 63%, p <O.OS), and a worse grade of the sestamibi defectmlated vessel flow according to the

From the Division of Cardiology and Nuclear Medicine, Clinica de1 Lavoro Foundation IRCCS, Medical Center of Rehabilitation of Veruno (No), Italy. This study was supported in part by grant Ricerca Corrente fPP2/93 from the Minister0 della Sanita, Rome, Italy. Manuscript received February 22, 1994; revised manuscript received and accepted May 2.5, 1994.

Address for reprints: Michele Galli, MD, Fondazione Clinica de1 Lavoro, via Revislate 13, Veruno 28010 (No), Italy.

Thrombolysis in Myocardial lnfaretion trial (1.2 + 0.3 vs 2.4 + 0.3, b ~0.05). After MG, changes in regional myocardial flow distrJbutii may occur in a sizable number of stable patlents with healed myocardial infarction, as reflected by their improved sestamibi uptake in areas show- ingrestingperfusiondefectsintheabsenceof symptoms of myocardlal ischemia.

(Am J Cardiol l-74:843-848)

T hrough its arterial and venous vasodilatory actions, nitroglycerin (NTG) reduces ventricular energy requirements and relieves myocardial ischemia;

NTG has also been shown to be beneficial in increasing regional myocardial blood flow in patients with coronary artery disease. lv2 Technetium-99m sestamibi scintigra- phy has been introduced into clinical practice for the evaluation of myocardial &hernia. Tracer uptake and retention are sensitive and proportional to regional myo- cardial blood flow and viability.3d In some patients with stable coronary disease, however, resting sestamibi uptake has been found to be reduced in segments with normal wall motion, as well as in asynergic regions still containing viable tissue. 7-10 This lends support to the claim that regional hypoperfusion may be present at rest in the absence of other ischemic symptoms.U*12 We spec- ulated that by improving the regional myocardial blood flow at the time of tracer injection, resting perfusion pat- tern would improve in some patients with viable but hypoperfused myocardium. The present study assessed whether regional sestamibi uptake defects would be sig- nificantly reduced at a second scan if the tracer injection were preceded by sublingual administration of NTG. Variation of the defect size was quantified and correlat- ed with the corresponding regional wall motion pattern.

MEWODS Study cohort and ~~MBcoI: We studied 36 consec-

utive patients (aged 56 f 2 years) with Q-wave myocar- dial infarction and a significant sestamibi defect at the resting tomography. All patients had been referred to our

NITROGLYCERIN AND SESTAMIBI SCAN 843

nuclear laboratory for functional evaluation of ischemic heart disease. Patients with recent myocardial infarction (within 1 month), unstable angina, or overt heart failure were excluded. Sestamibi tomographic studies were per- formed both at rest and after administration of sublin- gual NTG (0.3 to 0.6 mg) within 2 to 6 days. All patients were stable between the 2 sestamibi studies; cardioac- tive therapy remained unchanged and was not withheld before the 2 studies. In the NTG study, heart rate and blood pressure were measured before and every 5 min- utes after NTG administration; the tracer was injected 15 minutes after NTG, or sooner if a decrease in sys- tolic blood pressure or an increase in heart rate values

210% from baseline were observed. Global and region- al ventricular function were assessed at rest by 2-dimen- sional echocardiography within 1 week of the NTG scan. Standard coronary angiography was also performed in 31 of 36 stable patients (86%) within 8 weeks. The study protocol was approved by the local ethics committee for human research. All patients gave informed consent.

Sestamibi imaging and analysis: Scintigraphic images were obtained 90 to 150 minutes after the trac- er injection (30 mCi/70 kg) with a circular head-rotat- ing camera (Apex 409, Elscint, Israel). The imaging was performed over a 180” semicircular orbit, at 3” incre- ments of 25 seconds each. in a step-and-shoot mode.

RGURE 1. Representative example of rest-nitroglycerin repeat imaging and quantification of perfusion defect size in a woman with &vessel dis- ease, an occluded proximal circum- flex artery, and history of heart fail- ure after an inferolaterai infarction. The 3 short-axis slices from images at rest (I&j and after administration of nitroglycerin (right) and their corre sponding plots of quantitative region al sestamibi activity by circumferen- tial profiles are reported. The yellow line indicates the count distribution profile of the patient’s images; the gPeen line is the lower limit of normal sestamibi distribution; the white area below the normal profile indicates the abnormal hypoperfused area. The resting perfusion defect in the pori- terolateral wall was reduced after administration of nitroglycerin, from 3,396 to 2,077 integral units (-36.6%).

FlGURE 2. Scintigraphic findings at rest and after nitroglycerin in a man with a recent anterior infarction, severe left ventricular dysfunction, and a critically stenosed proximal left anterior descending artery. Qualita tive and quantitative representations are as in Figure 1. There were exterr sive abnormalities in septal, anterior, and anterolateral perfusion on resting imaging that improved after nitroglye erin (from 11,609 to 6,939 integral units, 49.7%).

844 THE AMERICAN JOURNAL OF CARDIOLOGYa VOLUME 74 NOVEMBER 1, 1994

Data were collected in a 64 X 64 array with a pixel size of 4.5 mm. Particular care was taken to avoid major arti- facts, such as patient motion during acquisition. Transax- ial slices were reconstructed using a back-projection algorithm with a modified Wiener filter with a dumping factor of 0.8, without attenuation or scatter correction; flood correction was applied during reconstruction. The spatial resolution in the transaxial plane was 8 mm. From the transaxial slices, sections in the horizontal and ver- tical long- and short-axis views were reconstructed.

For the quantitation of sestamibi uptake defects, 3 representative slices (apical, midventricular, and basal) from the short-axis view were considered. The apical slice was chosen by selecting the distal slice where the left ventricular cavity was tirst visible; the basal slice was chosen by selecting the slice in which a decrease in the septal activity was tirst visualized and by moving 1 to 2 slices further toward the apex. The midventricular slice was chosen halfway between the apical and basal slices. From each slice, the sestamibi uptake was dis- played as a circumferential profile (Figures 1 and 2) and normalized to the peak profile value. Each profile was then compared with the corresponding normal profile obtained in a group of 20 subjects with a ~5% proba- bility of coronary disease: a tracer uptake defect was considered to be present when the patient’s profile was <2 SD from the mean normal values. The defect size was quantified by integrating the area between the low- er normal limit and the patient’s profile. The integral val- ue is expressed by arbitrary units, reflecting both the extent and the severity of the perfusion defect. For each patient, the total defect size was obtained by adding the integral values obtained in the 3 short-axis slices. All studies were processed by a single experienced operator (CM). In a randomly selected group of 11 patients, the quantitative analysis was repeated after 2 weeks and the method reproducibility was assessed by linear regres- sion analysis (regression line: Y = 0.97x + 309; r = 0.98; p <O.OOl). After NTG, changes in sestamibi defect size exceeding the 95% confidence limits of the method vari- ability were considered significant.

Sestamibi images were also reviewed in side-by-side pairs (resting/NTG) by 2 experienced observers (MG, CM) who were unaware of other patients’ data. From the 3 short-axis slices, 16 myocardial segments (proxi- mal, mid-, and apical segments of the anterior and infe- rior walls; proximal and midanterolateral, posterolater- al, anteroseptal, and inferoseptal segments; and distal, septal, and lateral segments) were distinguished. Seg- ments with abnormal tracer uptake at quantitative analy- sis were identified. In cases of disagreement, a consen- sus was reached.

Twodimensional echocardiography: Thirty patients (83%) underwent a complete study in multiple views with a commercially available imaging system (Hewlett- Packard 77020). The 3 apical views (4 and 2 chambers and long axis) were analyzed. The left ventricle was divided into 16 segments in order to match the same left ventricular segmentation used at sestamibi tomography. Systolic wall thickening and inward wall motion were assessed visually off-line by 2 experienced operators (AI and PG) who were unaware of other information. Each

segment was graded on a 4-point scoring system (1 = normal; 2 = hypokinetic; 3 = akinetic; and 4 = dyski- netic).13 In cases of discrepancies, a consensus was reached. Biplane measurements of left ventricular vol- umes and ejection fraction were also obtained as previ- ously reported.14

Data interpretation: Myocardial sestamibi uptake and regional wall motion were compared in corre- sponding segments. Vascular attribution of the sestamibi defect to the distribution of the major coronary arteries was performed according to the Cedars-Sinai Laborato- ry attribution. l5 Coronary angiograms were interpreted by consensus of 2 experienced observers (MG and PS) who had no knowledge of the sestamibi and echocar- diographic results. The arbitrary cutoff point of >50% lumen diameter narrowing (by caliper measurements) was used for definition of significant stenosis. The Thrombolysis in Myocardial Infarction (TIMI) trial grade of the sestamibi defect-related vessel flow and the presence of collateral vessels were also assessed. A TIM1 grade of 0 to 1 defined an occluded artery and a TIM1 grade of 2 to 3 defined a patent artery.

Statistical analysis: Resting versus NTG hemody- namic and sestamibi data were compared using the paired Student’s t test and the Wilcoxon signed rank test for paired nonparametric data, respectively. Compar- isons between groups were obtained using the unpaired Student’s t test or the Mann-Whitney rank-sum test when appropriate. The chi-square test was used to determine the significance of differences in rates of occurrence. A p value co.05 (2-tailed) was considered significant. Data are reported as mean + SEM.

RESULTS The electrocardiographic site of Q-wave infarction

was anterior in 27 (75%), inferior in 6 (17%), and lat- eral in 3 (8%) patients. Wall motion abnormalities were documented in all (including the 30 patients with good echo images and 6 other patients with regional dys- function at contrast ventriculography). Mean ejection fraction was 42 f 2%. Multivessel coronary disease was documented in 17 of 31 patients (55%). In all patients, the resting sestamibi defect was located in asynergic (severely hypokinetic, akinetic, or dyskinetic) regions with corresponding abnormal Q waves and subtended by stenotic vessels.

After administration of NTG, blood pressure decreased significantly (systolic values: from 126 f 2 to 114 f 2 mm Hg, p ~0.01; diastolic values: from 75 * 1 to 72 + 2 mm Hg, p ~0.05) and heart rate was unchanged (from 72 + 3 to 76 -I 3 beats/min, p = NS). For the group as a whole, a significant reduction in the mean sestamibi defect size was observed after NTG (from 6,324 f 619 to 5,365 f 516, p ~0.01). The defect size after NTG administration was reduced beyond the reproducibility limits in 20 of 36 patients (56%) (group 1: from 7,370 + 852 to 5,370 + 734, p <O.OOl) (Figures 1 and 2) and was unchanged or increased in 16 of 36 patients (44%) (group 2: 5,016 + 810 vs 5,359 f 739, p = NS). Figure 3 shows individual sestamibi defect size values at rest and after NTG in the 2 groups. The resting perfusion defect size was comparable between the 2 groups. In

NITROGLYCERIN AND SESTAMIBI SCAN 845

TABLE I Clinical and Angiographic Findings in Patients With (group 1) or Without (group 2) Improvement in Regional Ses- tamibi Uptake Defect After Nitroglycerin

Group 1 Group 2 p No. (%) No. (%) Value

Number of patients 20 Age (years)’ 56 f 2 Male 18 (90) Nitrate medication at resting study 6 (30) Left ventricular ejection fraction (%)* 40 f 2 Thrombolysis in the acute phase 15 (75) Coronary angiography 19 (95)

1 -vessel disease 10 (53) 3-vessel disease 9 (47)

Sestamibi defect-related vessel Stenosis 51-75% 0 Stenosis 75-90% 6 (32) Stenosis >90% 13 (68) Mean stenosis (%)* 94 f 2

TIMI grade 0 9 (47) Grade 1 3 (16) Grade 2 1 (5) Grade 3 6 (32) Average TIMI grade* 1.2 It 0.3

Patent vessel+ 7 (37) Collaterals 9 (47)

16 56 f 2

16 (100) 7 (44) 45 f 3

13 (81) 12 (75)

4 (33) 8 (67)

2 (16) 5 (42) 5 (42) 85 f 5

1 03) 1 (8)

2 (17) 8 (67)

2.4 i 0.3 10 (83)

2 (17)

NS

NS NS NS

NS

co.05

co.05 co.05

NS

*Mean f SE. +lncludes Thrombolysis in Myocardial infarction trial grades 2 and 3. TIMI = Thrombolysis in Myocardial Infarction trial.

group 1, the average percent reduction of the perfusion defect after NTG was 29 f 4% (range 7 to 74). Perti- nent data of patients with or without NTG-reversible sestamibi defects are reported in Table I. The 2 groups were comparable in terms of age, ventricular function, and extent of coronary disease. In contrast, a more severe sestamibi defect-related vessel stenosis was found in group 1 patients, and the mean TIM1 grade was lower because of a greater prevalence of nonpatent vessels. There were no differences between groups regarding nitrate medication at the resting sestamibi study.

In the 30 patients with good echo images, different regional wall motion scores were found in normal and abnormal sestamibi regions (1.2 + 0.03 vs 2.6 f 0.07, respectively; p <O.OOl). The degree of contraction abnor- mality was lower in hypoperfused segments showing tracer uptake reversibility after NTG (2.1 f 0.1 vs 2.8 f. 0.1 in segments with fixed defect, p ~0.001) (Figure 4).

DISCUSSION Atherosclerotic coronary arteries show decreased en-

dotbelium-dependent relaxation and enhanced respons- es to vasoconstrictors. Vascular dysfunction improves after administration of L-arginine, the substrate of nitric oxide synthases. l6 Nitrates exert beneficial effects on both myocardial oxygen supply and demand. They

16000

0

Group 1 Group 2 (n= 20) (n= 16)

,-----p<o.o01-----,

Rest Nitroglycerin Rest Nitroglycerin

*=p<o.o01

FlGURE 3. Individual sestamibi defect size values at rest and after nitroglycerin in patients with (group 1, left) or without (group 2, rigbf) significant postnitroglycerin decrease in resting perfusion defect. Large open circles wifh bar imlicate average values.

FIGURE 4. A, average regional wall motion score in corresponding seg- ments with normal (open bar) or reduced (solid bar) sestamibi uptake at rest. 6, average regional wall motion score in corresponding segments with nitroglycerirr reversible (dashed bar) or fixed (hatched bar) sestamibi defects. The degree of the wall motion abnormality was less severe in segments showing defect reversibility after nitroglycerin.

846 THE AMERICAN JOURNAL OF CARDIOLOGY@ VOLUME 74 NOVEMBER 1, 1994

decrease myocardial work and oxygen requirement by reducing both ventricular preload and afterload.’ The mechanisms through which NTG increases coronary blood flow include the increase in large coronary arter- ies diameter, vessel spasm relaxation, ventricular end- diastolic pressure reduction, and diastolic compression decrease.’ Furthermore, NTG augments blood flow through the severely obstructed coronary artery by enlarging the vessel stenosis, reversing the artery con- striction, and improving the collateral circulation to the jeopardized area. lv2,17

In experimental studies, myocardial sestamibi uptake and retention have been related to both coronary blood flow and myocardial viability.3a Our findings of im- proved tracer distribution in asynergic regions after NTG suggest some improvement of regional myocardial blood flow to hypoperfused, dysfunctioning, but still viable regions. The presence of hypoperfused viable myocardi- urn in the absence of other signs or symptoms of myo- cardial ischemia has been documented distal to severe coronary stenoses.1’~‘2~18*‘9 In some cases, a residual vasodilator reserve can still be elicited in the face of decreased resting blood flo~/.‘~,~~ Recurrent myocardial stunning in critically supplied segments may also result in reduced resting blood flow, and a prolonged postis- chemic impairment of coronary vasodilation (“microvas- cular stunning”) has been documented in animals.21,22 Although our patients were stable and symptom-free at rest, we cannot exclude that they continued to exhibit recurrent, clinically silent episodes of brief myocardial ischemia.

Most of our patients showing improved sestamibi dis- tribution after NTG had severe stenosis and a worse TIM1 grade of the sestamibi defect-related vessel. It is possible that NTG induced changes in the luminal nar- rowing of the culprit vessel, increasing the regional blood flow and the tracer delivery. One additional expla- nation is that collateral vessels to the jeopardized area were recruited by NTG. In patients with proximal occlu- sion of the left anterior descending artery, NTG reversed the asynergy of the myocardium supplied by well-devel- oped collateral vessels. 23 In our study, collateral vessels to the sestamibi defect-related vessel were more fre- quently seen in patients whose tracer defect was reduced after NTG, although this difference was not significant because of the small sample size. However, the limita- tions of coronary angiography in the assessment of col- lateral vessels are well known.24,25

In the interpretation of post-NTG “reperfusion” of the infarcted area, another explanation deserves mention. It has been shown that “apparent” tracer defects can occur due to abnormal segmental contraction, despite preserved regional myocardial blood flow and tracer dis- tribution (the “partial volume effect”).26,27 The possibil- ity that segmental wall motion improved after NTG and attenuated the sestamibi defect seems logical. On the contrary, the explanation of post-NTG perfusion defect reduction due to changes in ventricular size seems un- likely. The hemodynamic effects of sublingual NTG are brief, and ventricular diameters and volumes return to baseline 45 to 60 minutes later.28 We acquired sestamibi

scans 90 to 150 minutes after NTG, when its hemody- namic effects had presumably ceased.

In our study, few patients had worsened regional ses- tamibi uptake after NTG. Although it is likely that NTG has only minor effects on coronary vessels <lOO km in diameter,29 which limits the compound’s effect on coro- nary flow and prevents coronary steal, the occurrence of the latter after NTG cannot be excluded in some of our patients.

Study limitations: The evaluation of NTG effects on regional sestamibi distribution was nonrandomized and uncontrolled; only patients with resting sestamibi defects were considered, and an additional sestamibi scan (i.e., after placebo) was not allowed for dosimetric reasons and patients’ compliance. Although closely related, resting wall motion and sestamibi distribution were not assessed simultaneously. However, all patients were stable and asymptomatic at rest and their therapy did not change between the 2 studies. Another limita- tion stems from the fact that coronary angiography was not available in all patients.

Clinicel implications: In our study, the degree of wall motion abnormalities was less severe in the corre- sponding regions showing perfusion defects that im- proved after NTG; this finding suggests that a greater amount of viable tissue was present in the hypoperfused area. The reversibility of the sestamibi defect after NTG casts doubt on the ability of a conventional resting ses- tamibi study (i.e., not preceded by NTG) to reliably quantify the real extent of the scarred area. Although in experimental models sestamibi uptake was not signifi- cantly affected by either myocardial stunning5 or short- term hibemation,“O the issue of whether resting sestamibi myocardial uptake reflects viability beyond blood flow distribution is now debated.s-lo Because sestamibi tracks with blood flow but does not redistribute appreciably,3 this tracer may not differentiate fibrotic from viable myocardium in the presence of a flow-limiting stenosis inducing persistent myocardial dysfunction (hibernating myocardium). Our results suggest that the accuracy of sestamibi scans for the detection of dysfunctioning but viable myocardium could be improved by NTG admin- istration. Apart from subjects with nonpatent tributary vessels who more frequently had a reduced perfusion defect after NTG, we are still unable to predict which patients with ventricular dysfunction will most benefit from NTG administration before the tracer injection. Further studies are required to better elucidate in whom the administration of NTG substantially improves the detection of residual viability by sestamibi scan.

Acknowledgments: We are grateful to Fabio Fringuelli, MD, and Orazio Zoccarato, PhD, for their participation in the scintigraphic data collection, to Elisa Gavinelli and Giacomina Romolo for their technical advice, and to Gillian Jarvis for her invaluable editorial assistance.

1. Abrams J. Mechanisms of action of the organic nitrates in the treatment of myo- cardial ischemia. Am .I Cardiol 1992;70:30B42B. 2. Brown BG, B&on EL. Peterson RB, Pierce CD, Dodge AT. The mechanism

NITROGLYCERIN AND SESTAMIBI SCAN 847

of nitroglycerin action. Stenosis vasodilation as a major component of drug re- sponse. Circulation 1981;65: 1089-1097. 3. Okada R, Glover D, Gaffney T, Williams S. Myocardial kinetics of 99mTc-hexa- kis-2.methoxy-2.methylpropyl-isonitrile. Circulation 1988;77:491-%98. 4. Verani MS, Jeroudi MO, Mahmarian JJ, Boyce TM, Barges-Neto S, Pate1 E, Bol- Ii R. Quantification of myocardial infarction during coronary occlusion and myo cardial salvage after reperfusion using cardiac imaging with technetium-99m hexa- kis-2-methoxy-isobutyl isonittile imaging. J Am Coil Cardiol 1988; 12:1573-15X1. 5. Sinusas AJ, Watson DD, Cannon JM, Belier GA. Effect of ischemia and postis- chemic dysfunction on myocardial uptake of technetium-99m-labeled methoxy- isobutyl-isonitrile and thallium-201. J Am Co/l Cardiol 1989;14: 1785-1793. 6. Beanlands RS, Dawood F, Wen W, McLaughin PR, Butany J, D’Amati G, Liu PP. Are the kinetics of technetium99m methoxyisobutyl isonitrile affected by cell metabolism and viability? Circulation 1990:X2: 1802-1814. 7. Rocco TP, Dilsizian V, Strauss WH, Boucher CA. Technetium-99m isonitrile myocardial uptake at rest. 2. Relation to clinical markers of potential viability. J Am Co/l Cardiol 1989;14:1678-1684. 8. Marzullo P, Parodi 0, Reisenhofer B, Sambuceti G, Picano E, Distante A, Giel- li A, L’Abbate A. Value of rest thallium-20l/technetium-99m sestamibi scans and dobutamine echocardiography for detecting myocardial viability. Am J Cardioll993; 71:16&172. 9. Dilsizian V, Bonow RO. Current diagnostic techniques of assessing myocardi- al viability in patients with hibernating and stunned myocardium. Circulation 1993; 87: l-20. 10. Sawada S, Allman K, Muzik 0, Beanlands R, Wolfe E, Gross M, Fig L, Schwaiger M. Positron emission tomography detects evidence of viability in rest technetium99m sestamibi defects. J Am Coil Cardiol 1994;23:92-98. ll. Berger BC, Watson D, Butwell LR, Crosby IK, Wellons HA, Testes CD, Belier GA. Redistribution of thallium at rest in patients with stable and unstable angina and the effect of coronary artery bypass surgery. Circulation 1979;60:111~1125. 12. Rahimtoola S. The hibernating myocardium. Am Heart J 1989;117:21 l-221. l3. Schiller NB, Shah PM, Crawford M, DeMaria A, Devereux R, Feigenbaum H, Gutgesell H, Reichek N, Sahn D, Schnittger I, Silverman NH, Tajik AI. Recom- mendations for quantitation of the left ventricle by 2-D echocardiography. .I Am Sot Echocardiogr 1989;2:358-367. l4. Giannuzzi P, De Vito F, Imparato A, Tavazzi L. Influence of left ventricular cavity dimension on electrocardiographic estimation of the extent of wall motion abnormalities. C/in Cardiol 1987;10:521-527. 1% Van Train K, Maddahi J, Berman DS, Kiat H, Areeda J, Prigent F, Friedman J. Quantitative analysis of tomographic stress thallium-201 myocardial scintigrams: a multicenter trial. J Nucl Med 1990;31:1168-1179. 19. Drexler H, Zeiher AM, Meinzer K, Just H. Correction of endothelial dysfunc-

tion in coronary microcirculation of hypercholesterolaemic patients by L-arginine. Lancer 1991;338:15461550. 17. Jugdutt BJ, Becker LC, Utchings GM, Bulkley BH, Reid PR, Kallman CH. Effect of intravenous nitroglycerin on collateral blood flow and infarct size in the conscious dog. Circulation 1981;63:17-28. 18. Nichols AB, Brown C, Han J, Nickoloff EL, Esser PD. Effect of coronary stenot- ic lesions on regional myocardial blood flow at rest. Circukztion 1986;74:74&757. 19. Parodi 0, Sambuceti G, Roghi A, Testa R, Inglese E, Pirelli S, Spinelli F, Cam- polo L, L’Abbate A. Residual coronary reserve despite decreased resting blood flow in patients with critical coronary lesions. A study by technetium99m human albu- min microsphere myocardial scintigraphy. Ciradation 1993;87:330-344. 20. Gould KL, Lipscomb K, Calvert C. Compensatory changes of the distal vascw lar bed during progressive coronary constriction. Circulation 1975;5 1: 1085-1094. 22. Bolli R. Myocardial stunning in man. Circukzrion 1992;86:1671-1691. 22. Bolli R, Triana F, Jeroudi M. Prolonged impairment of coronary vasodilation after reversible ischemia. Evidence for microvascular stunning. Circ Res 1990;67: 332-343. 22. Fujita M, Yamanishi K, Hirai T, Miwa K, Ejiri M, Asanoi H, Sasayama S. Sig- nificance of collateral circulation in reversible left ventricular asynergy by nitro- glycerin in patients with relatively recent myocardial infarction. Am Heart J 1990, 120:521-528. 24. Cohen MV. Coronary Collaterals: Clinical and Experimental Observations. Mount Kisco, NY: Futura, 1985:1-91. 25. Sabia PI, Powers ER, Ragosta M, Saremboc IL, Boxwell L, Kaul S. An asso- ciation between collateral blood flow and myocardial viability in patients with recent myocardial infarction. N Engl J Med 1992;327:1825-1831. 26. Parodi 0, Schelbert HR. Schwaiger M, Hansen H, Selin C, Hoffman El. Car- diac emission tomography: underestimation of regional tracer concentration due to wall motion abnormalities. J Camp Assist Tomogr 1984;8:1083-1092. 27. Sinusas AJ, Shi Q, Vitols PJ, Fetterman RC, Maniawski P, Zaret BL, Wack- en FIT. Impact of regional ventricular function, geometry, and dobutamine stress on quantitative -Tc-sestamibi defect size. Circulation 1993;88:2224-2234. 28. Abrams J. Pharmacology of nitroglycerin and long-acting nitrates and their use- fulness in the treatment of congestive heart failure. In: Gould KL, Reddy CVR, eds. Vasodilator Therapy for Cardiac Disorden. Mount Kisco, NY: Futura, 1979: 129-167. 29. Sellke FW, Myers PR, Bates IN, Harrison DG. Influence of vessel size on the sensitivity of porcine microvessels to nitroglycerin. Am J Physiol 1990;258: H5 1%H520. a0. Beller GA, Glover DK, Edwards NC, Ruiz M, Simanis JP, Watson DD.*Tc- sestamibi uptake and retention during myocardial ischemia and reperfusion. Cir- culation 1993;87:2033-2042.

848 THE AMERICAN JOURNAL OF CARDIOLOGYa VOLUME 74 NOVEMBER 1, 1994