can flow in the smaller artery (fig. 1). In the present

Transcription

1 Cross-sectional Area of the Proximal Portions of the Three Major Epicardial Coronary Arteries in 98 Necropsy Patients with Different Coronary Events Relationship to Heart Weight, Age and Sex CHARLES S. ROBERTS AND WILLIAM C. ROBERTS, M.D. SUMMARY The cross-sectional area (the portion enclosed by the internal elastic membrane) of histologic sections from the first 5-mm long segments of the right, left anterior descending and left circumflex coronary arteries was determined by videoplanimetry in 98 necropsy patients with coronary heart disease and in 46 control subjects who did not have significant coronary narrowing. Significant (p < 0.001) differences were observed in the mean cross-sectional area of each of the three major coronary arteries in the subgroups of coronary patients and among and between the control subjects. These differences resulted primarily from differences in heart weight and, to a slight extent, in age. Difference in sex was not significant. The 20 patients with angina pectoris had the smallest coronary arteries (mean cross-sectional area of each of the 60 arteries 6.0 mm2) and the smallest hearts (mean weight 386 g). The 18 patients with healed myocardial infarcts and intractable congestive heart failure had the largest coronary arteries (mean area 8.6 mm') and the largest hearts (mean weight 588 g). The 23 patients with acute transmural myocardial infarcts and the 19 with sudden coronary death had similar-sized coronary arteries (mean area 7.6 mm') and similar-sized hearts (mean weight 471 g). The 18 patients with healed myocardial infarcts, subsequently asymptomatic courses and noncardiac deaths had slightly enlarged arteries (mean area 6.9 mm') and hearts (mean weight 430 g). The 31 control subjects with cancer and normal or near-normal-sized hearts (mean weight 309 g) had the smallest coronary arteries (mean area 5.0 mm'). The 16 controls with aortic valve disease had the largest hearts (mean weight 730 g) and the largest coronary arteries (mean area 9.6 mm'). When heart weights were equalized (450 g), older patients had larger coronary arteries than younger patients (mean area < 40 years 6.5 mm', years 6.8 mm' and > 60 years 7.6 mm'). DEGREES of coronary arterial luminal narrowing in patients with symptomatic or fatal coronary heart disease have been studied extensively in recent years. The degrees of coronary narrowing in necropsy patients with fatal coronary heart disease are usually recorded in terms of cross-sectional area. Studies from this laboratory and others indicate that patients with fatal coronary heart disease at necropsy usually have narrowing by atherosclerotic plaques of more than 75% in cross-sectional area of at least two of the three major (right, left anterior descending and right) epicardial coronary arteries, and that over 30% of the entire lengths of these three major arteries are narrowed to this extent.1-7 Although degrees of crosssectional area narrowing at necropsy of the major epicardial coronary arteries have demonstrated certain differences among subsets of patients with fatal coronary heart disease,3-7 cross sectional area does not provide complete anatomic information. A large artery and a small artery, for example, can be similarly narrowed in cross-sectional area and yet the area through which blood can flow in the large artery obviously is greater than the area through which blood From the Pathology Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland. Address for correspondence: William C. Roberts, M.D., Building IOA, Room 3E-30, National Institutes of Health, Bethesda, Maryland Received November 5, 1979; revision accepted March 31, Circulation 62, No. 5, can flow in the smaller artery (fig. 1). In the present study, therefore, we describe the sizes of the three major coronary arteries in necropsy patients with clinical evidence of coronary heart disease and in control subjects and examine whether the sizes of these arteries are similar or different in various subgroups of coronary patients and, if different, why. Patients and Methods Ninety-eight necropsy patients were included in this study. Twenty patients had clinically isolated angina pectoris4 (table 1). Each died within 3 days of an aortocoronary bypass procedure, and during life their only evidence of myocardial ischemia was angina pectoris; none had had clinical evidence (historical and electrocardiographic) of acute myocardial infarction or chronic congestive heart failure. Twenty-three patients had fatal transmural (involving greater than the inner half of the left ventricularwall) acute myocardial infarcts' which by history and by histologic examination were 24 hours to 30 days old. Nineteen patients died suddenly and unexpectedly, and this subgroup hereafter will be referred to as sudden coronary death.3 Each died within 6 hours after the onset of chest pain, which, if present, began outside the hospital. None ever had evidence of congestive heart failure. At necropsy, at least one of the three major epicardial coronary arteries was greater than 75% narrowed in cross-sectional area by atherosclerotic plaques. None of these 19 patients at necropsy had ventricular wall myocardial coagulation 953

2 954 CIRCULATION VOL 62, No 5, NOVEMBER 1980 Internal diameter Internal diameter 2 cm _ 75 _...v >75% X-sectional narrowing >75% X-sectional narrowing Residual lumen (cm2) \ 3 rv.75 FIGURE 1. Diagram of transverse sections of two coronary arteries, both of which are narrowed greater than 75% in cross-sectional area. Although both are similarly narrowed in cross-sectional area, one can accommodate a much larger blood flow because it is much larger. necrosis. At necropsy, 36 patients had healed transmural myocardial infarcts. These 36 patients were divided into two groups: 18 with nonfatal healed myocardial infarcts,7 i.e., patients who had had acute myocardial infarcts that healed and who subsequently died of noncardiac conditions, and 18 with fatal healed myocardial infarcts,6 i.e., patients who had had transmural acute myocardial infarcts that healed and either immediately or later developed chronic congestive cardiac failure that became intractable and TABLE 1. Mean Cross-sectional Area of the Right, Left Anterior Descending, and Left Circumflex Coronary Arteries, Heart Weight, Age and Sex in the Five Groups of Coronary Patients and in the Two Groups of Control Subjects Mean XSA Heart weight (mm2) of R, Sex Age (years), (g), LAD, LCCA, Group Pts M F 1-Angina pectoris (49) (386) (6.0) 2 Healed MI (noncardiac death) (63) (430) (6.9) 3-Acute MI (58) (482) (7.6) 4-Sudden coronary death (54) (459) (7.7) 5-Healed MI (cardiac death) (58) (588) (8.6) 1 Cancer (51) (309) (5.0) 2 Aortic valve (56) (730) (9.6) Abbreviations: XSA = cross-sectional area; R = right coronary artery; LAD = left anterior descending coronary artery; LCCA = left circumflex coronary artery; MI = myocardial infaretion.

3 CORONARY ARTERIAL SIZE/Roberts and Roberts 955 FIGURE 2. Drawing of projection microscope (left) and video planimetry (VP) console (right) used to quantify the degree of coronary arterial cross-sectional area narrowing. The histologic section of coronary artery, mounted on a glass slide, is placed on the stage of the projection microscope; the magnified (X 650) image is traced onto opaque paper (lower left). The tracing is then placed under the focused camera of the VP console. The reflected light is adjusted so that the camera sees either the original lumen (or as done in some other studies2 7) the area of the atherosclerotic plaque (lower right). fatal. Patients with associated valvular, congenital or pericardial heart diseases or hypertrophic cardiomyopathy or myocardial diseases not secondary to coronary disease were excluded from these coronary subgroups. The hearts in all 98 patients were studied in similar fashion. The hearts were cleaned of postmortem clots, all portions of parietal pericardium were excised, the aorta and pulmonary trunk were excised about 2 cm cephalad to their sinotubular junctions and the hearts were carefully weighed. After fixation in 10% buffered formalin for at least 24 hours, the major epicardial coronary arteries were excised intact from the hearts and fixed again for at least another 24 hours. They were then x-rayed and if calcific deposits were present the arteries that contained calcium were decalcified. The right, left main, left anterior descending, and left circumflex coronary arteries were then cut into 5-mmlong segments at right angles to the longitudinal axes of the arteries and each segment was numbered chronologically beginning at its origin from the aorta (right and left main coronary arteries), or in the case of the left anterior descending and left circumflex arteries, from their origin from the left main. Each 5- mm segment was processed and dehydrated in alcohols and xylenes, embedded in paraffin, cut, and stained by Movat's pentachrome method. The latter was used because it clearly delineates the artery's internal elastic membrane. We studied three histologic sections of coronary artery from each patient, for a total of 294 sections. From each patient, one Movat-stained section of the right, left anterior descending and left circumflex coronary arteries was examined. Each section was prepared from the first 5-mm long segment - the artery's most proximal portion - of the respective artery. The cross-sectional areas of the coronary arteries were determined by planimetry (KE-compensating polar planimeter) (fig. 2). Each histologic section was positioned on the stage of a projection-light microscope that magnified the image 650 times (fig. 2). The circumference of the internal elastic membrane was then traced on white paper. Although all sections examined had atherosclerotic plaques in the lumens, this study concerned the total area enclosed by the artery's internal elastic membrane irrespective of the presence or absence of superimposed atherosclerotic plaque. Where the internal elastic membrane was artifactually indented, the lumen was extrapolated to a circular form. Each tracing then was placed under a focused camera and the resultant video signal was passed through an electronic integrator and displayed on a television monitor (fig. 2).8 The threshold level of the television monitor was adjusted so that the area of opacification estimated by video planimeters activated by the video signal corresponded to the original lumen (area enclosed by the internal elastic membrane). The analog output voltages of the video planimeters, calibrated on a grid system, provided on-line measurements of the crosssectional area. The area of each artery enclosed by the internal elastic membrane provided by videoplanimetry was converted into actual area in the following manner. A slide containing a 2-mm scale was projected onto tracing paper and recorded via the projection microscope in the same manner in which histologic sections of coronary arteries were projected and traced. The traced scale was converted into a square box, which was placed under the focused camera of the video planimeter console. The scale of the video-planimeter was then set to read 4 mm2 because the original length of one side of the box had been projected and traced from a 2-mm long scale. The actual cross-sectional areas of the right, left anterior descending and left circumflex coronary arteries were added together to obtain the sum of the cross-sectional area of each

4 956 ClIRCULATION VOL 62, No 5, NOVEMBER 1980 patient. This number was divided by 3 to obtain the mean cross-sectional area for each patient. For comparison with the 98 coronary patients, 46 subjects served as controls. During life none of them had clinical evidence of coronary heart disease and at necropsy none had any of their major coronary arteries narrowed greater than 75% in cross-sectional area by atherosclerotic plaques. The control subjects consisted of two major groups: 31 patients died of various cancers and 15 died of complications of aortic valve disease. Of the 31 control subjects with cancer, 26 had hearts that weighed 400 g or less and the other five had hearts that weighed g. Of the 15 patients with aortic valve disease, 11 had stenosis with peak systolic pressure gradients between left ventricle and a systemic artery of mm Hg (average 74 mm Hg), and the remaining four patients had pure aortic regurgitation of severe degree. The reason for selecting these two groups of control subjects was to obtain both normal and near-normal-sized hearts and also very large hearts. The coronary arteries in the 46 control subjects were examined in the same manner as in the 98 study patients. Results The findings in this study are summarized in tables 1-4. Except for similar values in the acute myocardial infarct and sudden coronary death subgroups, the mean values of the mean cross-sectional areas of each of the three major coronary arteries in each of the subgroups of coronary patients differed significantly (p < ), and also, these values differed significantly (p < ) from those of the control subjects. The 20 patients with angina pectoris had the smallest coronary arteries (mean cross-sectional area of each of the 60 arteries 6.0 mm2) and the 18 patients with healed myocardial infarcts with progressive and eventually fatal congestive heart failure had the "largest" coronary arteries (mean cross-sectional area of each of the 54 arteries 8.6 mm2). The 23 patients with acute myocardial infarcts and the 19 with sudden cardiac death had similar-sized coronary arteries (mean cross-sectional area of each of the 126 arteries 7.6 mm2) and the 18 patients with healed myocardial infarcts and subsequently asymptomatic courses and noncardiac causes of death had relatively small coronary arteries (mean cross-sectional area of each of the 54 arteries 6.9 mm2). The 31 control subjects with cancer had the smallest coronary arteries of any group (mean cross-sectional area of each of the 93 arteries 5.0 mm2) and the 15 patients with aortic valve disease had the largest coronary arteries (mean area of each of the 45 arteries 9.6 mm2) (p < ). The differences in cross-sectional area of the coronary arteries among the subsets of 98 coronary patients and between and among the 46 control subjects is attributable primarily to differences in heart weight (tables 1-3). The coronary subgroup with angina pectoris had the smallest hearts (average 386 g); the subgroups, with healed myocardial infarction with chronic congestive cardiac failure, had the largest hearts (average 588 g). The other three groups were intermediate. The cancer control subjects, who had the smallest coronary arteries, had the smallest hearts TABLE 2. Relation of.m1ean Cross-sectional Area of the Right, Left Anterior Descending and Left Circumflex Coronary A rteries to the Heart Weight in the 98 Coronary Patients and in the 46 Control Subjects Heart weight (gs) Parameter < > 700 Totals 1 No. patients Age (years), (49) (532) (58) (59) (61) (61) (56) 3-Male:female ratio 0:2 23:13 24:2 21:1 6:1 5:0 79:19 4 Heart weight (g), (270) (360) (465) (;546) (635) (754) (468) 5-Mean XSA (mm2) of R, LAD and LCCA (4.4) (5.9) (7.2) (8.9) (9.7) (10.3) (7.4) 1-No. subjects O Age (years), (47) (52) (64) (61) (56) (54) (53) 3-Male:female ratio 7:9 6:4 4:1 3:1 3:0 7:1 30:16 4-Heart weight (g), (243) (352) (436) (568) (660) (838) (447) 5-Mean XSA (mm2) of R, LAD and LCCA, (3.8) (5.9) (7.0) (8.4) (8.2) (10.8) (6.5) Abbreviations: XSA = cross-sectional area: LCCA = left circumflex coronary arterv. P = right coroniary artery; LAD left anterior descending coronary artery;

5 CORONARY ARTERIAL SIZE/Roberts and Roberts 957 TABLE 3. Relation of Mean Cross-sectional Area of the Right, Left Anterior Descending and Left Circumflex Coronary Arteries to Sex with Actual and Equalized Heart Weights in the 98 Coronary Patients and in the 46 Control Subjects Mean XSA (mm2) Heart weight of R, LAD, and Mean XSA (mm2) No. (g), LCCA, Equalized mean for R, LAD and patients heart weight LCCA Men (490) (7.7) Women (376) (5.9) Men (509) (7.7) Women (329) (4.4) Abbreviations: XSA = cross-sectional area; R = right coronary artery; LAD = left anterior descending coronary artery; LCCA = left circumflex coronary artery. (average 309 g) and the aortic valve control subjects, who had the largest coronary arteries, had the largest hearts (average 730 g). When the heart weights for each of the seven groups (five subgroups of coronary patients and two control groups) were equalized by using the same regression coefficient, no significant differences (p = 0.63) were found among the mean values of cross-sectional area of the coronary arteries (table 3). The mean values of cross-sectional areas of the coronary arteries in the women were significantly (p < ) different from the values in the men (mean cross-sectional area of each of the 19 women's 57 coronary arteries was 5.9 mm2 [range mm2] and of the 79 men's 237 arteries, 7.7 mm2 [ mm2]) (table 3). These mean differences, however, resulted from differences in heart weight. The heart weights of the 19 women ranged from g (mean 376 g) and those of the 79 men, g (mean 490 g). When the heart weights for the women and men with coronary heart disease were equalized by using the same regression equation, no significance differences in mean areas of the-coronary arteries between the sexes was apparent (table 3). Similar findings regarding sex were observed in the control subjects (16 women and 30 men). Of the 31 cancer control subjects, 14 were women (mean coronary cross-sectional area 3.9 mm2 and mean heart weight 280 g) and 17 were men (mean coronary crosssectional area 6.0 mm2 and mean heart weight 334 g); of the 15 aortic valve disease control subjects, two were women (mean coronary cross-sectional area 7.9 mm2 and mean heart weight 675 g) and 13 were men (mean coronary cross-sectional area 9.9 mm2 and mean heart weight 738 g). Age had a small independent effect on the mean cross-sectional area of the coronary arteries in the study patients. The younger patients had smaller coronary arteries than the older patients. The mean crosssectional area of the 33 coronary arteries in the 11 patients ages 40 years or younger was 5.7 mm2; in the 150 arteries in the 50 patients ages years, the mean area was 6.8 mm2, and in the 111 arteries of the 37 patients over 60 years of age, the mean area was 8.5 mm2 (p < ) (table 4). Most of the difference in cross-sectional area among the age groups resulted from differences in heart weights. The younger patients had smaller hearts, on the average, than the older patients. The mean heart weight in the 11 patients ages 40 years and younger was 395 g; in the 50 patients ages years, 456 g, and in the 37 patients older than 60 years, 506 g (p < ). When heart weight was equalized among the 98 coronary patients after dividing them into five age groups (< 40, 41-50, 51-60, and > 70 years), the mean crosssectional areas were significantly (p > 0.001) different except in the age groups years (table 4). Thus, aging itself among the coronary patients increased the cross-sectional areas of the three major coronary arteries. The nature of selecting the control subjects, i.e., those with normal- or near-normal-sized hearts (cancer victims) and those with huge hearts (aortic valve disease victims), prevented adequate numbers of patients in each of the five age-group divisions to determine if age per se caused the coronary arteries of the control subjects to enlarge. Discussion Our findings indicate that there are significant differences in the mean cross-sectional areas of the three major coronary arteries in patients with various coronary events, but that these differences are nearly all accounted for by differences in heart weight and, to a slight extent, age. Thus, the size of the coronary artery (the area enclosed by the internal elastic membrane, irrespective of the degree of luminal narrowing by atherosclerotic plaque) is unimportant in determining whether a person develops evidence of myocardial ischemia. Although the area through which blood

6 958 CIRCULATION VOL 62, No 5, NOVEMBER 1980 TABLE 4. Relation of Mean Cross-sectional Area of the Right, Left Anterior Descending and Left Circutnflex Coronary Arteries to Age in the 98 Coronary Patients and in the 46 Control Subjects Age (years) Parameter < > 70 Totals 1 No. patients Age (years), (35) (47) (55) (65) (78) (56) 3 Heart weight (g) (392) (480) (433) (503) (510) (468) 4-Mean XSA of R, LAD and LCCA, (3.7) (7.1) (6.6) (8.1) (9.2) (7.4) 1-No. patients Age (years), (32) (46) (.56) (65) (78) (53) 3-Heart weight (g), (306) (405) 2, (633) (401) (t590) (447) 4-Mean XSA of R, LAD and LCCA, (4.5) (.5.8) (8.3) (6.6) (9.2) (6.5) Abbreviations: XSA cross-sectional area; R = right corolnary artery; LAI) = left anterior descending coronary artery; LCCA = left circumflex coronary artery. might flow in a large artery narrowed greater than 75% in cross-sectional area by atherosclerotic plaque is greater than that of a small artery with a similar degree of cross-sectional area narrowing, the larger artery is large because the myocardial mass that it must perfuse is large and vice versa. Consequently, a 75% cross-sectional area narrowing in a large coronary artery has the same effect on myocardial perfusion as a similar degree of narrowing in a small coronary artery. The patients with angina pectoris had the smallest coronary arteries because they had the smallest hearts. The patients with healed myocardial infarcts who are left with chronic congestive heart failure that becomes intractable and fatal had the largest coronary arteries because they had the largest hearts. However, patients with healed myocardial infarcts who recovered completely, never had further evidence of myocardial ischemia, and died from noncardiac causes tended to have relatively small hearts (just larger, on the average, than those of the pure angina pectoris group) and, therefore, relatively small coronary arteries. The patients with sudden coronary death and those with acute myocardial infarcts had similar-sized hearts, and, therefore, similar-sized coronary arteries. Previous findings revealed similar degrees of cross-sectional narrowing throughout the entire lengths of the coronary arteries in patients with sudden coronary death and acute myocardial infarction.3 In both groups, 35% of the four major coronary arteries were narrowed greater than 75% in cross-sectional area. Also of note is the previous observation4 that the angina patients had the most severe degrees of cross-sectional area narrowing (48% of their four major coronary arteries were narrowed more than 75% in cross-sectional area) and the patients with healed myocardial infarcts irrespective of whether or not they died from cardiac or noncardiac causes had the least degree of severe crosssectional area narrowing (30% of their major coronary arteries were greater than 75% narrowed in crosssectional area by atherosclerotic plaque).', 7 This study also demonstrated that sex did not have an independent effect on the size of the coronary arteries. Women, on the average, however, had smaller coronary arteries than men, but this difference is entirely accountable for by differences in heart weight. This sex difference in heart size might explain why the early mortality after aortocoronary bypass operations, as reported by Hall and associates,9 is higher in women than in men. Women have smaller hearts, on the average, than men, and most of the bypass operations are performed for angina, which in itself is associated on the average with the smallest hearts of any of the various coronary events. The combination of pure angina and womanhood, in general, makes for relatively small hearts and, therefore, relatively small coronary arteries. The smaller the coronary artery, the greater the difficulty in inserting a conduit and the greater the likelihood after the coronary anastomosis that the runoff via the conduit into the native coronary artery will not be good. When coronary bypass fails to result in an adequate increase in myocardial oxygenation, the greater the likelihood of early death. 101 Age affected the cross-sectional area of the coronary arteries. This observation may suggest that it takes less cross-sectional area narrowing in younger persons to produce myocardial ischemia than it does

7 CORONARY ARTERIAL SIZE/Roberts and Roberts 959 in older persons, heart weight and other factors being equal. Elderly persons are known to have larger and more tortuous coronary arteries than younger adults. This "senile dilatation" comes about through both transverse widening and longitudinal lengthening. The present study indicates that this so-called senile dilatation, which occurs as a more-or-less normal event in insignificantly narrowed coronary arteries, also occurs, but probably to a lesser extent, in coronary arteries that are significantly narrowed by atherosclerotic plaques. Because arteries tend to be larger in older adults than in younger adults, coronary bypass anastomoses may be more readily accomplished in the old than in the young. From a technical standpoint, the young woman with pure angina pectoris may be at a greater risk of early mortality or lesser improvement than the elderly woman with similar symptoms and a similar-sized heart. Acknowledgment We thank Margaret C. Wu, Ph.D., Biometrics Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, for providing the statistical analyses of the data presented. References 1. Roberts WC, Buja LM: The frequency and significance of coropary arterial thrombi and other observations in fatal myocardial infarction. A study of 107 necropsy patients. Am J Med 52: 425, Roberts WC: The coronary arteries and left ventricle in clinically isolated angina pectoris. Circulation 54: 388, Roberts WC, Jones AA: Quantitation of coronary arterial narrowing at necropsy in sudden coronary death. Analysis of 31 patients and comparison with 25 control subjects. Am J Cardiol 44: 39, Roberts WC, Virmani R: Quantitation of coronary arterial narrowing in clinically isolated unstable angina pectoris. An analysis of 22 necropsy patients. Am J Med 67: 792, Roberts WC, Jones AA: Quantification of coronary arterial narrowing at necropsy in acute transmural myocardial infarction. Analysis and comparison of findings in 27 patients and 22 controls. Circulation 61: 786, Virmani R, Roberts WC: Quantification of coronary arterial narrowing and of left ventricular myocardial scarring in healed myocardial infarction with chronic eventually fatal, congestive cardiac failure. Am J Med 68: 831, Virmani R, Roberts WC: Non-fatal healed transmural myocardial infarction and fatal non-cardiac disease. Qualification and quantification of coronary arterial narrowing and of left ventricular scarring in 18 necropsy patients. Br Heart J. In press 8. Dvorak JA, Schuette WH, Whitehouse WC: A simple video method for the quantification of microscopic objects. J Microscopy 102: 71, Hall RJ, Garcia E, Wukasch DC, Hallman GL, Cooley DA: Aortocoronary bypass (CAB): long-term follow-up. (abstr) Circulation 52 (suppl I1): 11-90, Spray TL, Roberts WC: Status of the grafts and the native coronary arteries proximal and distal to coronary anastomotic sites of aortocoronary bypass grafts. Circulation 55: 741, 1977 I l. Spray TL, Roberts WC: Changes in saphenous veins used as aortocoronary bypass grafts. Am Heart J 94: 500, 1977