T the ST segment during attacks caused by coronary

Transcription

1 Different Responses of Coronary Artery and Internal Mammary Artery Bypass Grafts to Ergonovine and Nitroglycerin in Variant Angina Soichiro Kitamura, MD, Ryuichi Morita, MD, Kanji Kawachi, MD, Sogo Iioka, MD, Toshio Seki, MD, Kiyoshi Inoue, MD, and Shigeki Taniguchi, MD Thoracic and Cardiovascular Surgery, Department of Surgery 111, Nara Medical College, Nara, Japan The dynamic responses of a coronary artery and an internal mammary artery (IMA) graft to pharmacological intervention were examined by arteriography in 5 patients with variant angina who had undergone coronary artery bypass grafting with an in situ IMA to the left anterior descending coronary artery. Preoperative electrocardiographic findings included elevated ST segments in chest leads during attacks of angina, and all patients had severe fixed lesions in addition to marked spasm of the left anterior descending coronary artery after the administration of ergonovine maleate. Postoperatively with ergonovine stimulation, complete occlusion or marked subtotal narrowing was again observed at the primary fixed lesion in the proximal portion of the left anterior descending coronary artery, but the IMA graft and the coronary artery distal to the anastomotic site maintained satisfactory patency with no further occurrence of anginal pain or ST segment elevation. By computer-assisted graphic analysis, which allows highly reproducible measurements of vascular internal diameters, the diameter of the IMA showed only small changes under ergonovine (p = not significant) or nitroglycerin (p < 0.05) stimulation in contrast to the marked vascular reactivity of the coronary artery (p < 0.05 and <0.01, respectively). These findings indicate that the IMA graft is unresponsive to ergonovine at least in the amount required to produce coronary artery spasm in patients with variant angina and fixed lesions. The IMA graft appears to function well from a clinical and pharmacological viewpoint in patients with variant angina. (Ann Thorac Surg 1989;47:756-60) he variant form of angina pectoris with elevation of T the ST segment during attacks caused by coronary artery spasm is frequently provoked by administration of ergonovine maleate [ We performed coronary artery bypass grafting by anastomosing the left internal mammary artery (IMA) to the left anterior descending coronary artery (LAD) in 5 patients with variant angina in whom major fixed lesions associated with coronary artery spasm were observed on arteriograms during ergonovine and nitroglycerin loading. Postoperatively, repeat coronary arteriography with pharmacological intervention demonstrated a difference in vascular reactivities to pharmacological stimulation between the IMA graft and the coronary artery. This report describes a favorable characteristic of IMA grafts used in patients with variant angina associated with structural fixed coronary stenosis. Material and Methods Five patients with variant angina associated with major fixed lesions underwent myocardial revascularization with an in situ IMA graft to the LAD. Postoperative angiography with pharmacological intervention was performed to evaluate the comparative vascular reactivities of the coronary artery and the IMA graft. Preoperative Accepted for publication Dec 1, 1988 Address reprint requests to Dr Kitamura, Department of Surgery 111, Nara Medical College, 840 Shijo-cho. Kashihara, Nara, 634, Japan. electrocardiographic findings included elevated ST segments in chest leads during spontaneous anginal attacks, a finding compatible with a diagnosis of variant angina. In all patients, preoperative stimulation with ergonovine yielded a positive response in the LAD (the occurrence of angina, coronary artery spasm, and ST segment elevation on the electrocardiogram). Patient profiles are shown in Table 1. There were 4 men and 1 woman ranging in age from 35 to 65 years (mean age, 51 f 11 years). In all patients, angina usually occurred soon after waking up in the morning and was accompanied by reversible ST segment elevation in chest leads. None of the patients had had a previous myocardial infarction. Preoperative coronary arteriography demonstrated that coronary artery spasms occurring after the administration of ergonovine produced total or subtotal occlusion of the LAD. In addition, the administration of nitroglycerin revealed a mean fixed stenosis in the LAD that was 94% f 5% of the luminal diameter. All patients had a structural LAD lesion, and 4 of the 5 patients also had other diseased vessels. The mean number of grafts was per patient. The left IMA was anastomosed to the LAD with the use of cold diltiazem potassium-blood cardioplegic solution [4] to prevent perioperative coronary artery spasm in all patients. Saphenous vein grafts were also employed in patients with multivessel disease by The Society of Thoracic Surgeons /89/$3.50

2 Ann Thorac Surg 1989: KITAMURA ET AL 757 PHARMACOLOGlCAL RESPONSES OF IMA GRAFTS Table I. Summary of Patient Data ST Peri- Post- Segment opera- opera- Post- Patient Age History Angina Elevation Coronary Artery tive tive operative No. (yr) Sex of MI at Rest on Attack Disease (27510) No. of Grafts MI Angina Follow-up 1 49 M No Yes Yes LAD 1 (LIMA-LAD) No No 2yr6mo 2 35 F No Yes Yes LAD, LMT 2 (LIMA-LAD, No No 2yr6mo 3 65 M No Yes Yes LAD, LCX, RCA 3(LIMA-LAD, No No 2yr4mo SVG-RCA, 4 51 M No Yes Yes LAD, RCA 2 (LIMA-LAD, No No 1yr6mo 5 55 M No Yes Yes LAD, LCX SVG-RCA) 2 (LIMA-LAD, NO No 1yr4mo Mean 51? 11 yr 2.0? 0.7/patient 2.0? 0.6 yr LAD = left anterior descending coronary artery; LCX = left circumflex artery; LIMA = left internal mammary artery; LMT = left main trunk; MI = myocardial infarction; RCA = right coronary artery; SVG = saphenous vein graft. Coronary and graft arteriography with pharmacological stimulation was repeated 1 to 3 months (mean period, 1.2 * 0.8 months) postoperatively. Informed consent was obtained from the patients. Medication was withdrawn 48 hours before the examination. Cineangiograms of the IMA graft were made by manual injection of the contrast material (5 to 6 ml) through an F7 or F8 catheter. We tried to control the pressure generated during injection as evenly as possible in each study. First, control angiography was performed (ie, without drug administration). Then the examination was repeated during intravenous injection of ergonovine (0.2 to 0.4 mg) over a period of nine to 12 minutes and similarly during administration of nitroglycerin (0.5 mg). Responses of the LAD and IMA diameters to pharmacological stimulation were compared by computer-assisted graphic analysis, details of which have been reported elsewhere [5, 61. Briefly, end-diastolic cineangiographic images were represented in a 256 x 256 pixel array and quantified at each pixel according to a maximum 64-grade luminescence scale with a model LA-4 image-analyzing system (Pias Co, Osaka, Japan). End-diastolic images were digitized, and edges of the vessels were determined according to the method described by Spears and associates [7], ie, by the first derivative of the densitometric gradient curve (count profile curve) along a line perpendicular to the vessels. The accuracy of this method has been examined in model studies [5, 61. A significant correlation was observed between known and measured values of the diameters of several different types of wire (Y = +0.99, p < 0.001). Using this method, the diameters of the IMA and LAD were determined at three sites each (in the IMA just proximal to the IMA-LAD anastomosis and in the LAD, just distal to it), delineated by 30-degree right anterior oblique projections. Values were corrected for magnification using images of a Judkins catheter of known diameter. The differences were examined by a paired or unpaired Wilcoxon test and single-factor analysis of variance. Changes were considered significant at a p value of less than Results Operative Results There were no operative or late deaths, and no myocardial infarctions occurred early or late postoperatively. Followup ranged from 1 year 4 months to 2 years 6 months (mean follow-up, years). Postoperative management of all patients included prophylactic medication with oral isosorbide dinitrate (15 mg/day). There has been no recurrence of angina, and all of the patients are in New York Heart Association functional class I. Graft Patency The IMA-LAD anastomosis proved to be patent in all patients, but two of five saphenous vein grafts were occluded at the time of follow-up; however, no critically ischemic regions remained. Arteriographic Findings With Ergonovine and Nitroglycerin Administration With the administration of ergonovine, the primary fixed lesion in the proximal portion of the LAD became completely occluded in 3 of the 5 patients, and showed marked narrowing (99%) in the remaining 2 patients. However, the IMA maintained adequate perfusion to the area supplied by the distal LAD without the occurrence of angina or ST segment elevation on the electrocardiograms. Figure 1 shows postoperative images of the IMA graft and the LAD in a patient (patient 4) with a 90% fixed stenosis in segment 6 (American Heart Association classification) of the LAD who exhibited complete occlusion during ergonovine loading preoperatively. In the control arteriogram, the IMA retrogradely supplied the proximal portion of the LAD. During ergonovine stimulation, however, the coronary arteries became slightly spastic, and

3 758 KITAMURA ET AL Ann Thorac Surg 1989;47:75-0

4 Ann Thotac Surg 1989;47:75&60 KITAMURA ET AL 759 Table 2. Measurements of Internal Vascular Diameters and Responsiveness to Pharmacological lnteruention Internal Diameter (mm) Percent of Control Diameter (76) Patient No. Artery Control Ergonovine Nitroglycerin Control Ergonovine Nitroglycerin 1 IMA LAD IMA LAD IMA LAD IMA LAD IMA LAD Mean" IMA 1.93 % f f f 0 98? gbsc 116 f lot LAD 1.90 f * f f llb*d 148 f 24d " Data are shown as the mean? the standard deviation. pharmacological intervention (ergonovine vs nitroglycerin); nitroglycerin) IMA = internal mammary artery; LAD = left anterior descending coronary artery. Significance: p < 0.05 versus other vessel; ' significance: y 0.05 versus other Significance: p < 0.01 versus other pharmacological intervention (ergonovine vs the LAD was completely occluded. In contrast, the IMA showed no notable changes in internal diameter and perfused the distal LAD region satisfactorily. With the administration of nitroglycerin, both the coronary artery and the IMA were dilated, and the IMA filled the entire LAD region as well as part of the region supplied by the circumflex artery. Measurements of Internal Vascular Diameters and Responsiveness to Pharmacological Intervention Mean internal diameters at control arteriography were as follows: IMA, 1.93 f 0.25 mm, and LAD, mm (Table 2). The ratio between the two diameters was 1.06 f The diameter of the IMA was almost the same as that of the LAD on the control angiograms. Changes in the internal diameter of the IMA and LAD during pharmacological stimulation were evaluated, and the results are shown in Table 2. Because the control diameter was different in each patient, comparison was made based on the percentage of the control angiographic value (100%). The LAD became spastic, and its diameter was reduced under ergonovine loading to 81% f 11% of the control value. In contrast, the diameter of the IMA remained at 98%? 9% of the control value, and its change was significantly smaller compared with that of the LAD (p < 0.05). With nitroglycerin, the LAD showed markedly significant dilation to 148% f 24% of the control value (p < 0.01), whereas the IMA was much less significantly dilated to only 116% f 10% of the control value ( p < 0.05). The sequential changes in vascular diameter are shown in Figure 2. Comment The assessment of spasm and spasmogenicity of coronary arteries has been successfully accomplished by coronary arteriography with pharmacological intervention, which is apparently the only method currently available for evaluating dynamic vascular reactivity in humans [l-3,8]. Manual injection of contrast material through a catheter can lead to variability in the filling of the arteries, thus resulting in somewhat unreliable vascular imaging. However, errors in the measurement of vascular diameters caused by angiographic variability in imaging appear to be smaller in comparative studies involving the same individual rather than different individuals. Therefore, percent changes in vessel diameter rather than actually measured calibers were used in Figure 2 to compare the U P<O 05 T U ns 1Control Ergonovine NTG Fig 2. Percent of control angiographic diameters of the internal mummary artery (IMA) graft and the left anterior descending coronary artery (LAD) at the time of pharmacological intervention (the control angiographic value was considered to be 100%). The vascular reactivity in response to ergonovine maleate and nitroglycerin (NTG) was more significant in the LAD than in the IA4A graft. (m+sd = mean f standard deviation; ns = not significant.)

5 760 KITAMURA ET AL Ann Thorac Surg 1989: diameters of different arteries in different individuals. Visual evaluation of internal vascular diameters in cineangiographic images demonstrated poor reproducibility because of interobserver variations [5, 71. With our method, which uses digital image processing of cinefilms (5, 61, we could accurately measure the internal vascular diameter. Our method also proved useful for evaluating changes in the vascular diameter during pharmacological stimulation. Singh and Sosa [9] reported dilation of the coronary artery and the IMA graft after nitroglycerin administration. We quantitatively examined the responsiveness of the IMA graft and the LAD to ergonovine and nitroglycerin in 5 patients with variant angina who underwent coronary artery bypass grafting. The diameter of the LAD was reduced during ergonovine stimulation to approximately 80% of the control value, but that of the IMA remained almost unchanged. Moreover, with nitroglycerin stimulation, the LAD was dilated approximately 1.5 times the control value, whereas the IMA was dilated only about 1.2 times, the degree of dilation being significantly smaller. These findings suggest that although the IMA graft is viable [6, 91, it shows resistance to both ergonovine and nitroglycerin stimulation compared with the coronary artery. According to Hackett and co-workers [8], ergonovine exerts a direct local effect on the coronary artery and causes coronary spasms by inducing localized arterial hyperreactivity. Yokoyama and colleagues [ 101 noted differences in the responsiveness to ergonovine between a systemic artery, such as the femoral artery, and the coronary artery in hyperlipidemic rabbits, and suggested that these differences were derived from differences in the receptor level. Although the difference in the responsiveness to ergonovine between the systemic and coronary arteries has not yet been sufficiently demonstrated in humans, our present study indicates at least a quantitative difference in response between the viable IMA graft and the coronary artery. Spontaneous IMA graft spasm after coronary artery bypass grafting has been reported [ll, 121. The vascular response of the IMA grafts to ergonovine was not assessed in these patients, and the mechanism of IMA graft spasm may differ from that of the pharmacologically provoked spasm. However, it would appear to be very difficult to differentiate between IMA graft spasm and coronary artery spasm [13] in catastrophic perioperative events. In the present study, major IMA spasm could not be induced with ergonovine administered in the amount required to produce major coronary artery spasm in patients with variant angina. Thus, the mechanisms of occurrence of IMA graft spasm and coronary artery spasm may differ, at least in terms of dose requirements of vasoconstrictive drugs. Although the number of patients is limited, our study has demonstrated an important physiological phenomenon in humans in regard to comparative vascular reactivities of the coronary artery and the IMA in response to ergonovine and nitroglycerin loading. We think the IMA graft can function well from a clinical and pharmacological viewpoint in patients with variant angina associated with fixed coronary lesions. References 1. Schroeder JS, Bolen JL, Quint RA, et al. Provocation of coronary spasm with ergonovine maleate. Am J Cardiol 1977;40: Heupler FA, Proudfit WL, Razavi M, Shirey FK, Greenstreet R Sheldon WC. Ergonovine maleate provocation test for coronary arterial spasm. Am J Cardiol 1978;41: Norell M, Balcon R. Utility of provocative tests for coronary artery spasm. Eur Heart J 1985;6(Suppl F): Kitamura S, Nishii T, Kawachi K, et al. Evaluation of effectiveness and safety of cold-diltiazem-potassium cardioplegia in coronary artery bypass surgery. Nippon Geka Gakkai Zasshi 1985;86: Seki T, Kitamura S, Morita R, et al. New methods for evaluating chronological changes of graft length and diameter in coronary artery bypass graft surgery for Kawasaki heart disease. Acta Cardiol Pediatr Jpn 1988;3: Kitamura S, Seki T, Kawachi K, et al. Excellent patency and growth potential of internal mammary artery grafts in pediatric coronary artery bypass surgery: new evidence for a "live" conduit. Circulation 1988;78(Suppl 1): Spears JR, Sandor T, Als AV, et al. Computerized image analysis for quantitative measurement of vessel diameter from cineangiograms. Circulation 1983;68: Hackett D, Larkin S, Chierchia S, Davies G, Kashi JC, Maseri A. Induction of coronary artery spasm by a direct local action of ergonovine. Circulation 1987;75: Singh RN, Sosa JA. Internal mammary artery: a "live" conduit for coronary bypass. J Thorac Cardiovasc Surg 1984;8793& Yokoyama M, Akita H, Mizutani T, Fukuzaki H, Watanabe Y. Hyperreactivity of coronary arterial smooth muscles in response to ergonovine from rabbits with hereditary hyperlipidemia. Circ Res 1983;53: Sarabu MR, McClung JA, Fass A, Reed GE. Early postoperative spasm in left internal mammary artery bypass grafts. Ann Thorac Surg 1987;44: Blanche C, Chaux A. Spasm in mammary artery grafts [Letter]. Ann Thorac Surg 1988;45: Lemmer JH Jr, Kirsh MM. Coronary artery spasm following coronary artery surgery. Ann Thorac Surg 1988;46: