Noninvasive detection of subcritical coronary arterial narrowings with a coronary vasodilator and myocardial perfusion imaging

EXPERIMENTAL STUDIES

Noninvasive Detection of Subcritical Coronary Arterial Narrowings With a Coronary Vasodilator and Myocardial Perfusion Imaging

H. WILLIAM STRAUSS, MD BERTRAM PITT, MD

Baltimore, Maryland

From the Divisions of Nuclear Medicine and Cardiology, The Johns Hopkins Medical Institutions, Baltimore, Maryland. This work was supported in part by U. S. Public Health Service Grant GM10548 and Grant P50HL17655-02 from the National Institutes of Health, Bethesda, Maryland and the Department of Health, Education, and Welfare, Washington, D. C. Manuscript received July 6, 1976; revised manuscript received September 23, 1976, accepted September 29, 1976. Address for reprints: Bertram Pitt, MD, 601 North Broadway, Baltimore, Maryland 21205.

Myocardial perfusion imaging after administration of the potent coronary vasodilator ethyl adenosine-5'-carboxylate, which increases flow to normal areas in excess of that to areas supplied by subcritically stenosed vessels, was investigated as a nonischemia-producin9 stimulus for detecUng subcriUcal coronary stenosis. Preliminary studies in 10 dogs with reactive hyperemia were performed with thallium-201 and potassium-43 to determine which tracer was a better indicator of increased flow. Neither agent was a linear indicator of increased flow caused by reactive hyperemia but thallium-201, because of its imaging characteristics, was selected as a flow indicator after administration of ethyl adenosine. Five dogs were studied after placement of a subcriUcal stenosis on the left circumflex coronary artery. Strontium-86 microspheres were injected into the left atrium after placement of the stenosis to verify that changes in resting blood flow were only minimal. Thereafter, intravenous administration of ethyl adenosine was followed by injection of chromlum-51labeled microspheres into the left atrium and intravenous administration of thallium-201. The mean ratio of left circumflex to left anterior descending coronary arterial flow was 0.96 4- 0.16 for the control experiment after subcriUcal stenosis; after administration of the vasodilator the ratio of activity levels in the two arteries was 0.43 4- 0.09 with the chromium-51 microspheres and 0.56 -I- 0.07 with thallium-201. Imaging performed in three additional dogs after injection of microspheres in the presence of subcritical stenosis revealed a normal pattern, whereas imaging after administration of the vasodilator and thallium-201 revealed a perfusion deficit. In two additional dogs without subcritical stenosis, thallium was administered after injection of ethyl adenosine to determine that the drug alone did not cause perfusion deficits. The perfusion scans in these two dogs were normal. These studies suggest that a coronary vasodilator and thallium-201 myocardial imaging can be used to detect subcriUcal coronary stenosis.

The noninvasive methods currently used to detect subcritical stenosis of a coronary artery require the development of myocardial ischemia either to induce S-T segment abnormalities in the electrocardiogram 1 or to cause regional changes in tracer concentrations in myocardial perfusion images. 2 This level of stress may be difficult or impossible to achieve in patients with peripheral vascular or chronic pulmonary disease and in some patients treated with beta adrenergic blocking drugs. We undertook this study to develop and validate a noninvasive method of detecting subcritical coronary stenosis that does not require establishment of an ischemic state. To achieve this goal, a new coronary vasodilator of the adenosine type, ethyl-adenosine-5'-carboxylic acid, 3'4 was used to induce differential perfusion between normal coronary beds and

March 1977

The American Journal of CARDIOLOGY Volume 39

403

NONINVASIVE DETECTION OF CORONARY NARROWING--STRAUSS AND PITT

those distal to a subcritical stenosis, and myocardial perfusion imaging was used to detect the resultant difference in regional perfusion. Materials and Methods

Thallium-201 was supplied as a sterile pyrogen-free ionic radiochemical in 0.9 percent sodium chloride solution (Philips-Duphar, Petten, Holland). Potassium-43 was supplied as •a sterile pyrogen-free radiochemical in a 0.9 percent sodium chloride solution (Oak Ridge National Laboratories ERDA, Oak Ridge, Tennessee). Carbonized microspheres, 7 to 10 in diameter, labeled with either strontium-85 or chromium-51 at a specific activity level of 50 mCi/mg were used to detect regional myocardial blood flow (3M Company, St. Paul, Minnesota) as previously described. '~-7 Ethyl adenosine-5'carboxylate hydrochloride was supplied as a sterile powder (Abbott Company), and dissolved immediately before use in saline solution. Two series of experiments were performed: one to determine which radiopharmaceutical agent was the most suitable noninvasive indicator for detecting regional changes of increased perfusion, and the second to determine whether the agent chosen could detect a subcritical stenosis after administration of the vasodilator. Experiment 1: This series of experiments utilized reactive hyperemia to increased blood flow to a segment of myocardium and compared the regional distribution of microspheres with that of ionic potassium-43 or thallium-201 activity. Ten dogs weighing 20 to 30 kg were anesthetized with sodium pentobarbital (30 mg/kg body weight), intubated and placed on a Harvard respirator. A lateral thoracotomy was performed in the left fifth intercostal space. After the pericardium was sectioned, a 19 gauge polyethylene catheter was placed in the left atrium for subsequent administration of microspheres. A small segment of the left circumflex coronary artery was exposed and a snare placed around it. The artery was occluded for 30 seconds, and 15 seconds after release of the occlusion, potassium-43 (50 #Ci) was given intravenously and strontium-85 labeled microspheres (50 ttCi, 500,000 particles) were administered through the left atrial catheter. In the remaining five dogs, the same experimental model was used, but thallium-201 (100 ttCi) rather than potassium-43 was administered intravenously. All animals were killed with an overdose of pentobarbital 5 minutes after administration of the tracer. The heart was removed and the left ventricle divided by gross dissection into areas supplied by the left circumflex coronary artery and by the left anterior descending coronary artery. Three to four full thickness sections, 2 to 3 g, were taken from the center of the region supplied by each ventricle, weighed and counted in a 5 inch (12.7 cm) diameter sodium iodide well type scintillation counter. Thallium-201 was counted at a window of 70 to 100 key to include the mercury X-ray. Potassium-43 was counted from 600-650 key and strontium from 470 to 500 key. Corrections were made for crossover into each of the windows. After crossover corrections, errors due to counting were less than 2 percent. Experiment 2: Eight adult mongrel dogs weighing 20 to 30 kg were prepared as in experiment 1. In addition to the polyethylene snare, an electromagnetic flow transducer and screw clamp were also positioned around the left circumflex artery. A 19 gauge polyethylene catheter was passed through the carotid artery to the central thoracic aorta and connected to a pressure transducer. A multilead electrocardiogram was monitored throughout the experiment. The left circumflex coronary artery was transiently occluded by the snare for 30

404

March 1977

seconds and the occlusion then released. Peak blood flow after release of the snare (peak reactive hyperemic flow) was calculated from the electromagnetic flow recording. The screw clamp distal to the left circumflex arterial flow transducer was then carefully tightened until the reactive hyperemia response was just eliminated. After the subcritical stenosis was established, 100 uCi of strontium-85-1abeled microspheres was administered into the left atrium to document the regional distribution of coronary blood flow. Five minutes later, the vasodilator ethyl-adenosine-5'-carboxylic acid was administered intravenously in a bolus dose of 1 mg/kg; 30 seconds thereafter, at the peak increase of blood flow, 3 100 #Ci of thallium-201 was administered intravenously and 100 #Ci of chromium-51-1abeled microspheres injected into the left atrium. Five minutes later, the dogs were killed with an overdose of sodium pentothal, the hearts removed and divided into sections as described in experiment 1. The activity of each tracer was determined by differential gamma spectrometry using a window of 470 to 550 key for strontium-85, 270 to 350 key for chromium-51 and 70 to 90 key for thallium-201. In three animals two scans were obtained using a Picker 500 Magnascanner with a high energy collimator (2112). The first scan was performed after establishing the subcritical stenosis and 100 #Ci of strontium-85 microspheres was administered into the left atrium to determine the control perfusion level. A window of 460 to 550 key with 20 percent contrast enhancement and a count density of 1,000 counts/cm 2 was used. The second scan was performed after administration of ethyl adenosine and 1 mCi of thallium-201 was given intravenously to determine the change in regional perfusion induced by the vasodilator (a window of 70 to 90 kev with 20 percent contrast enhancement at 1,000 counts/cm2). In two additional animals, 1 mCi of thallium-201 was administered after intr.avenous administration of ethyl adenosine, and scans were performed as before to determine if the drug alone caused changes in the perfusion scan. All data was analyzed using the Student's t test. Results Series 1

Potassium-43: After release of a 30 second left circumflex coronary arterial occlusion, during reactive hyperemia, the ratio of strontium-85 microsphere activity in the left circumflex arterial distribution compared with t h a t in the left anterior descending distribution was 4.2 + 0.7 (standard deviation), whereas the ratio of ionic potassium-43 activity in the same segments was 1.7 4- 0.3. This value was significantly lower than t h a t of the microspheres (P <0.05). Thallium-201: T h e ratio of left circumflex to left anterior descending coronary arterial s t r o n t i u m - 8 5 microsphere activity during reactive hyperemia was 3.9 4- 1.1, whereas the ionic thallium-201 ratio for the same segments was 2.3 4- 0.5; this value was significantly less than t h a t for strontium-85 (P <0.05). T h e difference between the s t r o n t i u m - 8 5 microsphere and potassium-43 distribution at hyperemia was 55 percent c o m p a r e d with the 41 p e r c e n t difference between the s t r o n t i u m microsphere and thallium-201 distribution; the difference between the thallium-201 and potassium-43 distribution was not significant (P <0.1). Because thallium-201 a p p e a r e d to reflect the increased flow better than potassium-43, it was selected as the ionic tracer for the second series of experiments.

The American Journal of CARDIOLOGY Volume 39

NONINVASIVE DETECTION OF CORONARY NARROWING--STRAUSS AND PITT

Series 2

TABLE I

After creation of a subcritical coronary arterial stenosis, the resting heart rate was 164 + 15 beats/min and the mean aortic pressure was 126 + 10 mm Hg. The ratio of left circumflex to left anterior descending coronary arterial flow was 0.96 ± 0.16 as determined by strontium-85 microspheres (Table I). Myocardial imaging performed at the strontium-85 energy setting failed to reveal any perfusion deficit in the areas supplied by the subcritically narrowed vessel (Fig. 1). After administration of ethyl adenosine, the heart rate was not significantly changed from the control value of 162 4- 18 beats/min; the mean aortic pressure decreased slightly but significantly to 118 ± 10 mm Hg (P <0.05) and returned to control values 90 seconds after administration of the vasodilator. The regional distribution of perfusion was changed significantly by ethyl adenosine as evidenced by the ratio of left circumflex to left anterior descending arterial chromium-51 microsphere activity of 0.43 ± 0.09, compared with the control value of 0.96 + 0.16 obtained with the strontium spheres (P <0.02). The thallium-201 activity ratio of 0.56 ± 0.16 also differed significantly from the control ratio obtained with the strontium-85 microspheres and the ratio obtained with the chromium-51 microspheres (P <0.05). Myocardial perfusion images performed at the thallium-201 energy setting revealed a marked decrease in tracer concentration in the zone of the left ventricle supplied by the left circumflex coronary artery (Fig. 1). Scans in the two dogs without occlusion were normal. Electrocardiographic recordings remained unchanged after administration of ethyl adenosine.

Ratio of Left Circumflex to Left Anterior Descending Coronary Arterial Flow After Vasodilatation

Discussion

D e t e c t i o n of subcritical coronary stenosis with a vasodilator: This technique relies upon the fact that flow in the normal coronary bed can increase severalfold compared with resting flow whereas a coronary bed supplied by a subcritical stenosis cannot increase flow to a similar degree. 7,s Ischemia is not produced with the coronary vasodilator ethyl adenosine4; flow actually increases to the ischemic area whereas oxygen demands increase, as reflected by the small but significant re-

Control Study After Subcritical Stenosis

Mean

Effect of Ethyl Adenosine

(Strontium-85 Microspheres)

Chromium-51 Microspheres

Thallium-201

0.91 0.98 0.71 1.08 1.10 0.96 -+0.16

0.56 0.37 0.35 0.49 0.39 0.43 +0.09 * t

0.56 0.46 0.55 0.67 0.55 0.56 _+0.07*t

*Significantly different from control (P <0.02). t r = 0.51, thallium-201 significantly different from chromium-51 microspheres (P <0.05).

duction in mean arterial pressure with an essentially unchanged heart rate. Because ischemia is not produced, methods other than the electrocardiogram must be used to detect the subcritical stenosis. In our study myocardial perfusion imaging with thallium-201 was used to detect the change in coronary blood flow between the normal and critically narrowed vessel induced by ethyl adenosine. Previous studies revealed a linear relation of potassium-43 and thallium-2019J° to the distribution of myocardial blood flow at rest and during myocardial ischemia and infarction. However, during reactive hyperemia, uptake of neither potassium-43 nor thallium-201 paralleled the myocardial flow values indicated by the strontium microspheres. There are two possible reasons for this finding: (1) The increases in flow examined were transient and the microspheres cleared from the blood in one pass through the tissue whereas the ionic tracers cleared in multiple passes. Therefore, the microspheres reflected only one point in time whereas thallium reflected the mean flow during the time of tracer clearance from the blood. (2) It is possible that the extraction fraction for the ionic tracers is significantly altered by increasing flow above demands. Thallium-201 was selected as the tracer for

A.

FIGURE 1. Thallium-201 and radioactive microsphere images in the left lateral position in a dog with subcritical stenosis of the left circumflex coronary artery. The microsphere image (right) was performed before administration of the vasodilator. The thallium-201 image (left) was obtained with radionuclide injection performed 30 seconds after administration of vasodilator and imaging began 5 minutes later. Note the appearance of a perfusion defect in the thallium-201 image after administration of the vasodilator compared with the uniform perfusion in the microsphere image. The arrows point to the ischemic zone in the region supplied by the left circumflex coronary artery.

j,L

March 1977

The American Journal of CARDIOLOGY

Volume 39

,r i, ,

405

NONINVASIVE DETECTION OF CORONARY NARROWINC'---STRAUSS AND PITT

comparison with microspheres in the study with ethyl adenosine because of its better imaging properties and because of its slightly, though statistically insignificant, improvement over potassium in reactive hyperemia. Ethyl adenosine was selected as the vasodilator for these studies because it causes an intense coronary vasodilatation almost equal to peak reactive hyperemia with a relatively small change in systemic arterial pressure.3, 4 Nitroglycerin would be less effective in causing a difference in flow between normal and abnormal coronary arteries because its peripheral effects exceed its action on the coronary vascular bed. Other vasodilators such as dipyridamole could be used to produce the degree of coronary vasodilatation necessary to detect subcritical stenosis, but the decrease in systemic pressure associated with these agents may result in a "myocardial steal syndrome" and myocardial ischemia. 7 The ability of the technique used in our study to detect changes in regional myocardial perfusion depends upon two factors, the intensity of the coronary vasodilatation in the nonstenosed beds and the duration of the vasodilator response. Vasodilatation must be maintained long enough to permit clearance of the tracer from the vasculature. One minute after administration of thallium-201 about 14 percent of the dose remains in the blood; 86 percent is cleared by the tissues. 11 After intravenous administration of ethyl adenosine, coronary vasodilatation occurs at 20 to 30 seconds and flow remains one and a half to two times normal for 30 to 60 seconds. The time and intensity of this response are sufficient to permit uptake of thallium-201 and therefore detection of the subcritical coronary arterial lesions. A similar technique has been used to detect subcritical lesions at the time of arteriography. 12 The inves•tigators used angiographic contrast material as the vasodilator and radioactive microspheres injected directly into the coronary artery as the flow indicator. 12 Two

microsphere labels were used: one administered before the vasodilator and the second at the time of maximal coronary vasodilatation. Images of the distribution of regional myocardial perfusion with the two tracer labels allowed detection of subcritical coronary arterial lesions in patients. Clinical applications: Our study used a noninvasive method to produce a difference in flow between normal and subcritically narrowed coronary arteries. In clinical practice, patients would initially undergo thallium-201 myocardial perfusion imaging after administration of a vasodilator. If a perfusion defect were evident in the initial perfusion image a second study without pretreatment with the vasodilator would be performed several days later• Fixed zones of decreased myocardial perfusion in these images would represent areas of myocardial scarring whereas perfusion defects appearing only after administration of the vasodilator would suggest subcritical coronary arterial lesions. This approach to the diagnosis of subcritical coronary arterial lesions has two potential advantages over current methods: First, it requires less active patient participation than exercise stress testing. It is thus similar to right atrial pacing but has the advantage of being noninvasive. Second, because it does not produce ischemia, the risk of the procedure should be low. Although the principle of maximal arterial vasodilatation for the detection of subcritical coronary arterial lesions appears to be sound, the sensitivity and specificity of the technique remain to be determined under clinical conditions.

Acknowledgment We thank Dr. Kenneth Poggenburg of the Oak Ridge National Laboratories for supplying the potassium-43 and Dr. Rolf DeJong of the Philips-Duphar Company, Petten, Holland for supplying the thallium-201. In addition, we are grateful for the encouragement and support we received from Dr. Henry N. Wagner, Jr. and Dr. Richard S. Ross during these investigations.

References 1. Borer JS, Brensike JF, Redwood DR, et ah Limitations of the electrocardiographic response to exercise in predicting coronary artery disease. N Engl J Med 293:367-371, 1975 2. Zaret BL, Strauss HW, Martin ND, et al: Noninvasive regional myocardial perfusion with radioactive potassium: study of patients at rest, with exercise, and during angina pectoris. N Engl J Med 288:809-812, 1973 3. Somani P: Coronary vasodilator prQperties of ethyl adenosine5'-carboxylate HCI. In, Recent Advances in Studies on Cardiac Structure and Metabolism, Vol 7 (Harris P, Bing RJ, Fleckenstein A. eds). Baltimore, University Park Press, 1976, p 413-420 4. Helmann DB, Pitt B: Effect of Abbott 40557 (ethyl adenosine-5'carboxylate HCI) on regional myocardial blood flow following coronary artery occlusion. Am J Physiol, in press 5. Rudolph AM, Heymann MA: The circulation in the fetus in utero. Circ Res 21:163-184, 1967 6. Kaihara S, Van Heerden PD, Migita T, et ah Measurement of the distribution of cardiac output. J Appl Physiol 25:696-700, 1968 7. Becker L, Fortuin NJ, Pitt B: Effect of ischemia and antianginal

406

March 1977

The American Journal of CARDIOLOGY

8.

9. 10. 11. 12.

Volume 39

drugs on the distribution of microspheres in the canine left ventricle. Circ Res 28:263-269, 1971 Gould KL, Hamilton GW, Lipscomb K, et al: Method for assessing stress induced regional malperfusion during coronary arteriography. Experimental validation and clinical application. Am J Cardiol 34:557-564, 1974 Prokop EK, Strauss HW, Shaw J, et al: Comparison of regional myocardial perfusion determined by ionic potassium-43 to that determined by microspheres. Circulation 50:978-984, 1974 Strauss HW, Harrison K, Langan JK, et ah Thallium-201 for myocardial perfusion imaging: relation of thallium-201 to regional myocardial perfusion. Circulation 51:641-645, 1975 Bradley-Moore PR, Lebowitz E, Green MW, et ah Thallium-201 for medical use. II. Biological behavior. J Nucl Med 16:156-160, 1975 Rltchie JL, Hamilton GW, Gould KL: Myocardial imaging with indium- 113m and technetium-99m labeled macroaggregatedalbumin. Am J Cardiol 35:380-389,1975