JACC Vol. 20. No.5 November I. 1992:1073-9

1073

Effect of Progression of Left Ventricular Hypertrophy on Coronary Artery Dimensions in Aortic Valve Disease

BRUNO VILLARI, MD,* OTTO M. HESS, MD, DEBORAH MOCCETTI, BA, GIUSEPPE VASSALLI, MD, HANS P. KRAYENBUEHL, MD

Zurich. Switzerland

Objectives. The effect of progression of left ventricular hyper. trophy on coronary artery dimensions was studied in patients with aortic valve disease.

Methods. Cross·sectional area of the left and right coronary arteries was determined by quantitative coronary arteriography in 12 control subjects and in 10 patients with aortic valve disease at baseline and after a follow·up period of 66 months.

Results. The cross·sectional area of the left coronary artery was larger in patients with aortic valve disease than in control subjects (left anterior descending artery 13 vs. 8 mm1

, p < 0.001; left circumflex artery 13 vs. 6 mm2

, p < 0.001). At the follow·up examination, cross·sectional area of the left coronary artery increased (left anterior descending artery 17 mm2

, p < 0.01 vs. baseline; left circumflex artery 15 mm1, p < 0.01 vs. baseline). The cross· sectional area of the right coronary artery was not different in patients with aortic valve disease from that in control subjects. Left ventricular muscle mass was larger in patients with aortic valve disease both at baseline (269 g, p < 0.001) and after follow·up examination (339 g, p < 0.001) than in control subjects

Several observations suggest that in the hypertrophied left ventricle there is increased vulnerability to ischemia (1,2). Although in the overloaded left ventricle coronary blood flow at rest is within normal limits (3), relative subendocar· dial hypoperfusion may develop during exercise or rapid cardiac pacing (1,2,4-6). It has been reported (7) that in patients with aortic valve disease and massive hypertrophy, left coronary artery size was enlarged preoperatively but decreased after valve replacement at an equal rate with the regression of left ventricular muscle mass. The purpose of the present study was to evaluate changes in coronary artery dimensions after progression of left ventricular hypertrophy

From the Department of Internal Medicine, Cardiology, University Hos· pital. Zurich. Switzerland. This study was supported in part by the Swiss National Science Foundation. Berne. Switzerland. Dr. Villari is a recipient of a fellowship of the Italian Society of Cardiology. Rome. Italy.

'Present address: Division of Cardiology, Federico II University of Naples, Via S. Pansini 7.80131. Naples. Italy.

Manuscript received December 12. 1991; revised manuscript received April 2, 1992, accepted May 8. 1992.

Address for correspondence: Otto M. Hess. MD. Department of Internal Medicine, Division of Cardiology. University Hospital, Raemistrasse 100. 8091 Zurich. Switzerland.

©1992 by the American College of Cardiology

--(136 g). The appropriateness of coronary artery size with respect to muscle mass was evaluated by normalizing cross·sectional area of the left coronary artery (left anterior descending plus left circumflex artery) per 100 g of left ventricular muscle mass (mm1/l00 g). This index was 10.9 mm2/100 g in control subjects, and decreased in subjects with aortic valve disease from 10.3 mml/l00 g at baseline to 8.6 mm2/100 g at the follow·up measurement (p < 0.05 vs. control values).

Conclusions. In patients with aortic valve disease, the progreso sion of left l'entricular hypertrophy is associated with an increase in left anterior descending and left circumflex coronary artery dimensions, whereas the size of the right coronary artery remains unchanged. Despite the enlargement of the left coronary artery, the cross·sectional area of the left coronary artery per 100 g of left ventricular muscle mass decreased. Hence, the increase in coronary artery size appears to be inadequate when the severity of left ventricular hypertrophy increases.

(J Am CoU CardioI1992j20:1073-9)

in patients with aortic valve disease who were not yet surgical candidates at the baseline cardiac catheterization.

Methods Ten patients (mean age 51 ± 9 years, range 35 to 79) with

aortic valve disease and normal coronary arteries were studied at baseline and after a follow.up period of 66 ± 36 months (range 24 to 122) (mean age 56 ± 11 years, range 46 to 85). Three patients had a bicuspid aortic valve. Twelve normal subjects (mean age 52 ± 8 years, range 38 to 65) evaluated for atypical chest pain served as control subjects. No control subject had a bicuspid valve. Body surface area did not differ between control subjects (1.85 ± 0.22 m2

) and patients with aortic valve disease either at baseline (1.81 ± 0.18 m2

) or at follow·up study (1.82 ± 0.19 m2). Physical

working capacity was determined by upright bicycle exer· cise testing at baseline and after the follow·up period. The maximal work load, which was achieved under steady state conditions (3·min steps), was expressed in percent of the agee, gender· and height·corrected normal value (8).

Cardiac catheterization. Informed consent was obtained from all patients. The study protocol required selective

0735·10971921$5.00

1074 . VILLARI ET AL VENTRICULAR HYPERTROPHY AND CORONARY DIMENSIONS

JACe Vol. 20, No.5 November 1, 1992:1073-9

coronary arteriograms of high quality. These were obtained during diagnostic arteriography without any further exami­nation for scientific purposes. Premedication consisted of 10 mg of chlordiazepoxide given orally 1 h before cardiac catheterization. Vasoactive substances were withheld for 24 h before catheterization. Left ventricular pressure was measured transseptaily with use of an 8.5F Brockenbrough catheter, whereas aortic pressure was determined through an 8F pigtail catheter introduced retrogradely from the right femoral artery. Pulmonary artery pressure was measured with a 7F Cournand catheter. Mean coronary perfusion pressure was calculated as mean aortic pressure minus mean right atrial pressure. Mean systolic pressure gradient and aortic valve area were calculated according to standard formula. Aortic regurgitation was assessed by thermodilu­tion technique.

Left ventricular angiograms were recorded simulta­neously in the right and left anterior oblique projections at a filming rate of 50 frames/so Left ventricular volumes and ejection fraction were calculated with the area-length method (9). Left ventricular muscle mass was determined according to the method of Rackley et al. (10).

Selective left and right coronary arteriography was car­ried out from the right femoral artery (Judkins technique, 8F catheters) with multiple views for optimal visualization of the coronary arteries. Only patients with normal coronary arteries were included in the present analysis.

Quantitative coronary arteriography. Quantitative evalu­ation of coronary angiograms was performed with a semi­automatic computer system (7,1 ,12). The system is based on a 35-mm film projector (Tagarno 35 eX), a slow-scan CCD-camera for image digitation and a computer work station (Apollo DN 3000) for image storage and processing. Contour detection was carried out with use of a geometric­densitometric edge detection algorithm (12-14) (Fig. 1). The method of computerized analysis of coronary angiograms has been described elsewhere (7,11-17).

The proximal cross-sectional area of the three major coronary vessels Cieft anterior descending, left circumflex and right coronary arteries) was measured from one to three end-diastolic cine frames. The proximal cross-sectional area of the left anterior descending and left circumflex arteries was defined as the vessel segment immediately beyond the bifurcation of the left main coronary artery over a length of approximately 1 cm. The computer traced this segment automatically and calculated the mean area over this seg­ment. A circular lumen was assumed because only patients vvith normal coronary arteries were included (Fig. The proximal cross-sectional area of the right coronary artery was defined as the vessel segment 1 to 2 em distal to the coronary ostium. A vessel segment over a length of approx­imately 1 em was analyzed, and the mean cross-sectional area was calculated as for the left coronary artery. For each vessel segment, three measurements in different projections were obtained and averaged to correct for biologic variations in coronary artery dimensions (1 ,17,18). Calibration was

Figure 1. Representative coronary arteriogram of a patient with aortic stenosis at baseline (upper panel) and at the follow-up examination (lower panel). The contour of the left anterior descend­ing coronary artery is automatically detected by the computer (dotted rectangle). At the follow-up examination an increase in left coronary artery cross-sectional area is evident (from 14.5 mm2 to 16.6 mm2

).

performed automatically by using the proximal part ofthe 8F Judkins catheter as a scaling device (7,19). Left or right coronary artery dominance was evaluated according to standard criteria (20). No correlation between left domi­nance and bicuspid valve was observed. As an index of the enlargement of the coronary arteries with respect to muscle mass, the cross-sectional area of the left coronary artery (left anterior descending plus left circumflex artery) per 100 g of left ventricular angiographic mass was calculated (7,21).

Statistics. Statistical comparisons of hemodynamic and angiographic data among control subjects and baseline and follow-up measurements in patients with aortic valve disease were carried out by a one-way analysis of variance. When the analysis was significant, the Scheffe procedure was applied. The paired Student t test was used to compare baseline and follow-up data in patients aortic valve disease. The chi-square test was used to compare the type of coronary dominance between control subjects and patients with aortic valve disease.

VILLARI ET AL. 1075 JACC Vol. 20, No.5 November I, 1992:1073-9 VENTRICULAR HYPERTROPHY AND CORONARY DIMENSIONS

Table 1. Hemodynamic and Angiographic Data.

Control Patients With AVD (n = 10) SUbjects (n = 12) Baseline Follow-Up

HR (beats/min) 70 ± 13 73 ± 8 80 ± 10 LVSP (mm Hg) 114 ± 20 163 ± 2U 174 ± 22* L VEDP (mm Hg) 9±3 10 ± 3§ 15 ± 6t MPAP(mm Hg) 15 ± 3 19 ± 10 23 ± 10* MCPP(mmHg) 86 ± 8 102 ± 33 92 ± 15 EF(%) 65 ± 4 63 ± 8 52 ± 1U EDV (ml) 137 ± 28 223 ± 1St 283 ± 115:j:,§ LMM (g) 136 ± 30 269 ± 60~ 339 ± 73t.!1

*p < 0.05, tp < 0.01, ~p < 0.001 versus control values. §p < 0.01, lip < 0.001 versus baseline measurements in patients with aortic valve disease. AVD = aortic valve disease; EDV = end-diastolic volume; EF = ejection fraction; HR = heart rate; LMM = left ventricular mass; L VEDP = left ventricular end-diastolic pressure; L VSP = left ventricular systolic pressure; MCPP = mean coronary perfusion pressure; MPAP = mean pulmonary artery pressure.

Results Five patients had pure or predominant aortic stenosis

(mean systolic pressure gradient 41 mm Hg [range 25 to 61]; aortic valve area 1.08 cm2 [range 0.9 to 1.5]) and five had pure or predominant aortic regurgitation (regurgitant frac­tion 29%, range 21% to 37%). In patients with aortic steno­sis, mean systolic pressure gradient was 69 mm Hg (range 30 to 90) (p < 0.01 vs. baseline) and aortic valve area was 0.6 cm2 (range 0.45 to 0.98) (p < 0.01 vs. baseline) at the follow-up examination, whereas regurgitant fraction was minimal. In patients with aortic regurgitation, mean regur­gitant fraction was 54% (35% to 65%) (p < 0.01 vs. baseline) at the follow-up examination. Functional classification ac­cording to the New York Heart Association was significantly higher at the follow-up evaluation than at baseline (2.43 vs. 1.25, respectively, p < 0.(01) in both aortic stenosis and regurgitation. Physical working capacity was comparable to control values (97 ± 15%) in patients with aortic valve disease at baseline (82 ± 18%) but was decreased (p < 0.01) at the follow-up evaluation (56 ± 22%).

Hemodynamic and angiographic data (Table 1). All pa­tients were in sinus rhythm. Heart rate was similar in all three groups. Left ventricular peak systolic pressure was significantly higher in patients with aortic valve disease than in control subjects both at baseline and at follow-up exami­nation. Left ventricular end-diastolic pressure was signifi­cantly higher in patients with aortic valve disease at follow-up than at baseline or than in control subjects, whereas mean pulmonary artery pressure was higher in patients with aortic valve disease at follow-up than in control subjects. Mean coronary perfusion pressure was similar in control subjects and in patients with aortic valve disease both at baseline and follow-up. Ejection fraction was com­parable in control subjects and in patients with aortic valve disease at baseline; however, it decreased significantly at the follow-up examination when compared with both baseline

Table 2. Quantitative Coronary Arteriographic Data

Control Patients With AVD (n = 10) Subjects (n = 12) Baseline Follow-Up

CSALAD (mm2) 7.8 ± 3.6 13.3 ± 2.5~ 16.9 ± 4.5§ CSALc• (mm2) 6.3 ± 3.4 12.9 ± 5.3~ 14.9 ± 3.3§ CSARCA (mm2) 9.0 ± 4.7 10.4 ± 5.3 11.2 ± 4.0 f(mm2/IOO g) 10.9 ± 3.2 10.3 ± 2.3 8.6 ± 4.0*

*p < 0.05, tp < 0.01, +p < 0.001 versus control values. §p < 0.01 versus baseline measurements in patients with aortic valve disease. A VD = aortic valve disease; CSALAD = cross-sectional area of left anterior descending coronary artery; CSALc• = cross-sectional area of left circumflex artery; CSARCA = cross-sectional area of right coronary artery; f = (CSALAD + CSALc.)I100 g of left ventricular muscle mass.

evaluation or with control values. Left ventricular mass was significantly increased in patients with aortic valve disease with respect to control values; a further increase in left ventricular mass occurred at the follow-up examination.

Coronary artery dimensions (Table 2, Fig. 2). In seven subjects in the control group, the right coronary artery was dominant, whereas the left coronary artery was dominant in three and a balanced type of distribution was present in two. Of the patients with aortic valve disease, right coronary dominance was present in four patients, left dominance in another four and a balanced distribution in two. These types of coronary dominance did not differ between the control group and patients with aortic valve disease.

At baseline the proximal cross-sectional area of the left anterior descending and the left circumflex artery was sig­nificantly larger in patients with aortic valve disease than in control subjects; however, after the follow-up period the dimensions of the left anterior descending and circumflex

Figure 2. Coronary artery dimensions in control subjects and in patients with aortic valve disease (AVO) at baseline (B) and follow-up (F-U) evaluation. The cross-sectional area of the left anterior descending (LAD) and the left circumflex (LCX) coronary artery of the patients was increased significantly at baseline when compared with values in control subjects; both dimensions in­creased further at follow-up evaluation. The size of the right coronary artery (RCA) did not change.

24 (mm 2

)

RI

;16 iii c: .2 u 011 CII 8 en III 0 (3

0

o LAD A LCX • RCA

ftl Controls AVO (8)

AA ...

AVO (F·U)

•• = p<O.01 , ... = p<O.OO1 vs. Controls. ""= p<O.01 vs. AVO (8)

1076 VILLARI ET AL. VENTRICULAR HYPERTROPHY AND CORONARY DIMENSIONS

JACC Vol. 20, No.5 November I, 1992:1073-9

Table 3. Correlations Between Coronary Artery Dimensions and Hemodynamic Variables

Left Coronary Right Coronary

Cross-Sectional Artery (n = 32) Artery (n = 32)

Area (mm2) p* r p*

vs. LVSP (mm Hg) 0.58 <0.001 0.09 NS vs. LVEDP (mm Hg) 0.47 <0.01 0.19 NS vs. MPAP (mm Hg) 0.43 <0.05 0.47 <0.01 vs. EF (%) -0.48 <0.01 -0.03 NS vs. LVEDV (mI) 0.66 <0.001 0.27 NS vs. LMM (g) 0.86 <0.001 0.28 NS vs. MCPP (mm Hg) 0.09 NS 0.25 NS vs. PWC (%) 0.08 NS 0.18 NS

*By linear regression analysis. LVEDV = left ventricular end-diastolic volume; PWC = physical working capacity; other abbreviations as in Table I.

arteries increased significantly with respect to baseline. The proximal cross-sectional area of the right coronary artery did not differ from control values at baseline or at follow-up.

The index of the appropriateness of left coronary artery size per 100 g of left ventricular muscle mass was compara­ble in patients with aortic valve disease at baseline and in control subjects (10.3 vs. 10.9 mm21100 g) but decreased significantly at the follow-up examination in the patients with valve disease (8.6 mm2/100 g).

Correlations (Table 3). Correlations were calculated from all available data (n == 32) from control subjects as well as those from patients with aortic valve disease at baseline and at follow-up examination. There was a significant correlation between cross-sectional area of the left coronary artery and left ventricular peak systolic and end-diastolic pressure, mean pulmonary artery pressure, left ventricular end­diastolic volume and left ventricular muscle mass (Fig. 3). Left coronary cross-sectional area was inversely related to ejection fraction (r == -0.48, p < 0.01) but did not correlate with mean coronary perfusion pressure or maximal work load in percent of the age-, gender- and height-corrected normal value. Cross-sectional area of the right coronary

Figure 3. Correlation between left coronary artery cross-sectional area and left ventricular muscle mass (n = 32). The dashed lines represent the 95% confidence limits. Abbreviations as in Figure 2.

n=32 r = 0.86, pcO.OO1

o 50 100 150 200 250 300 350 400 450 500

Left ventricular muscle mass (g)

(mll'f 1100 g)

14 • 12

III 10 l!! III~

8 'iii~ C....I .2D1

6 ug I .... u,C£j 4 eo. n =10 0 2 r = -0.65, p<0.05

0 0 20 40 60 80 100 120 140

Follow-up months

Figure 4. Correlation between left coronary cross-sectional area per 100 g of left ventricular muscle mass (LMM) and time period of follow-up evaluation in the patients with aortic valve disease (n = 10).

artery was correlated significantly only with mean pulmo­nary artery pressure (r = 0.47, p < 0.01). A close inverse relation was present between the cross-sectional area of the left coronary artery per 100 g of left ventricular mass and the duration of the follow-up period (r == -0.65, p < 0.05, Fig. 4).

Discussion Coronary artery enlargement consequent to left ven­

tricular hypertrophy has been shown to represent an adap­tive mechanism to maintain blood flow and shear stress constant (22,23). Because coronary blood flow velocity is approximately the same in normal and hypertrophied ven­tricles (22), the relation between coronary artery dimension and left ventricular muscle mass should also be constant. The data of the present study suggest that in the presence of moderate left ventricular hypertrophy the increase in coro­nary artery dimensions is proportionate to the increase in left ventricular muscle mass because the index of the appropri­ateness of left coronary cross-sectional area/l00 g of left ventricular muscle mass is comparable to that in control subjects (7,21,24,25). A further increase in left ventricular mass is accompanied by a further enlargement of coronary artery size; however, this increase is not proportionate to the increase in mass.

Progression of left ventricular hypertrophy and coronary artery size. Marcus et al. (26) have demonstrated in dogs that the duration of left ventricular hypertrophy does not influence the severity of the decrease in coronary vasodilator reserve. However, in the same experimental study left ventricular mass was comparable at baseline and follow-up evaluation. In the present study the severity of aortic valve disease was mild to moderate, and ejection fraction was preserved at the time of the baseline evaluation. Ng et al.

VILLARI ET AL. 1077 JACC Vol. 20, No.5 November I, 1992:1073-9 VENTRICULAR HYPERTROPHY AND CORONARY DIMENSIONS

(27) have demonstrated that, among patients with aortic valve stenosis, those with absent or mild left ventricular functional impairment are more likely to have progression of valve disease. In the present study all patients with aortic stenosis had a reduction in aortic valve area of >20% and an increase in the mean systolic pressure gradient of >40%. In patients with aortic regurgitation, the regurgitant fraction increased by >55% of the baseline value at follow-up eval­uation. Thus, the severity of aortic valve disease increased in all patients in our study. The persistence and the increased severity of the overload condition led to a significant in­crease in left ventricular mass and a decrease in systolic ejection performance. The inverse relation between these two variables has been previously described by Murakami et al. (28).

In preoperative patients with aortic valve disease and left ventricular hypertrophy, it was shown that the dimensions of the left anterior descending and left circumflex arteries, but not of the right coronary artery, are increased. The present study confirms these findings in patients with mild to mod­erate aortic valve disease who were not yet candidates for surgery at baseline examination. After a follow-up period of approximately 6 years, a significant increase in the severity of aortic valve disease was accompanied by a significant increase in left ventricular mass and a further increase in left coronary artery dimensions, whereas those of the right coronary artery did not change. Nevertheless, at follow-up evaluation the enlargement of the left coronary artery was not adequate to match the increased left ventricular mass; thus, the cross-sectional area per 100 g of left ventricular mass decreased significantly. This inadequate growth of the cross-sectional area was mainly due to the lower percent increase in left coronary artery dimensions with respect to the percent increase in left ventricular mass (21% vs. 27%, respectively, Fig. 5).

Murray and Vatner (29) have demonstrated in dogs with severe right ventricular hypertrophy that the increase in right ventricular mass was not accompanied by a proportion­ate increase of the total cross-sectional area of the coronary vasculature supplying the hypertrophied right ventricle. Sev­eral other studies (7,21,24,25) reported a reduction of cross­sectional area per 100 g of left ventricular mass in the hypertrophied left ventricle. Our findings suggest that the main determinant for the increase in coronary size is left ventricular hypertrophy. A previous report from oUf labora­tory (7) has shown that in preoperative patients with aortic valve disease (360 g) the cross-sectional area of the left coronary artery per 100 g of left ventricular mass was reduced. After valve replacement and regression of left ventricular hypertrophy, the residually hypertrophied myo­cardium (250 g) was supplied by a coronary artery that was enlarged but adequately sized with respect to left ventricular muscle mass (7). In the present study we observed the same pattern of variations; that is, coronary cross-sectional area per 100 g was normal in the presence of moderate left ventricular hypertrophy (left ventricular mass 269 g),

(%) 50 NS

40 r-- p<0.05 --.

r-- p<0.05 -,

30

20

10

0 LC RCA LMM Cross·sectlonal area

Figure 5. Percent changes in left ventricular mass (LMM) and cross-sectional area of the left (LC) and right (RCA) coronary artery at the follow·up evaluation (mean percent change and standard error). The percent change in left ventricular muscle mass was larger than the increase in cross·sectional area of the coronary arteries, although this difference reached statistical significance only for the right and not for the left coronary artery.

whereas the increase in coronary artery dimensions was inappropriate when left ventricular hypertrophy was severe (left ventricular mass 339 g).

Determinants of coronary artery size. The stimulating factor for increasing coronary cross-sectional area is not known. Several mechanisms are possible, such as coronary flow, myocardial hypertrophy or changes in microvascula­ture.

Coronary blood./low. Certainly coronary blood flow plays a major role. The continuous release of vasoactive sub­stances, such as the endothelium-derived relaxing factor, increases with an increase in coronary blood flow and thus might stimulate the growth of coronary artery size ... is unknown if flow at rest or maximal flow is the stimulating factor for the increase in coronary artery size; flow at rest is increased in patients with myocardial hypertrophy, whereas maximal flow is normal and thus coronary flow reserve (maximal flow divided by resting flow) is decreased (3).

If ./low at rest is the stimulating factor for the increase in coronary artery size, the growth of coronary arteries in patients with myocardial hypertrophy could be easily ex­plained by the increase in flow at rest. However, the inadequate growth of the coronary arteries in patients with severe hypertrophy (decrease in coronary cross-sectional area per 100 g of muscle mass, Table 2) is difficult to explain but could be due to the relative decrease in coronary flow per 100 g mass that has been observed in patients with severe aortic valve disease (30,31).

If maximal flow is the stimulating factor for the increase in coronary artery size, then the reduction in coronary flow reserve in patients with severe myocardial hypertrophy

'lO78 VILLARI ET AL. VENTRICULAR HYPERTROPHY AND CORONARY DIMENSIONS

JACC Vol. 20, No.5 November I, 1992:1073-9

could explain the inadequate growth of the coronary arteries because maximal blood flow does not increase in proportion to left ventricular muscle mass (3) but remains more or less unchanged. The occurrence of myocardial ischemia with redistribution of coronary blood flow from the subendocar­dium to the subepicardium (32-34) also could influence the growth of the coronary arteries by decreasing maximal flow to the subendocardium during exercise (32,33).

Myocardial hypertrophy. Left ventricular hypertrophy plays an important role in the regulation of coronary blood flow and thus coronary artery size. The increase in cross­sectional area of the coronary arteries in patients with severe aortic valve disease can be explained by the increase in coronary blood flow associated with myocardial hypertro­phy. However, humoral (growth factors) or mechanical (stretch) stimuli could be responsible for the growth of the vascular smooth muscle; in other words, the growth factors causing left ventricular hypertrophy may also cause the increase in coronary artery size. An increase in the size of the epicardial coronary arteries was observed in postmortem hearts of patients with left ventricular hypertrophy (35,36). The relation between total heart weight and the diameters of the coronary arteries was found to be linear only up to a heart weight of 500 g (35).

Microvasculature. Changes in the microvasculature (5,37) could lead to a decrease in coronary flow reserve and thus could be associated with an inadequate growth of the epicardial coronary arteries. However, it has been shown in patients with aortic valve disease (3) that coronary flow reserve tends to normalize after successful valve replacement, suggesting that the microvasculature is not altered but that hypertrophy is not associated with an appropriate increase in the cross-sectional area of the micro­vascular bed.

Angiogenesis has been observed in dogs (38) during long-term left ventricular hypertrophy that is sufficient to increase the cross-sectional area of the resistance vessels in proportion to the increase in left ventricular mass. However, in contrast to experimental studies in animals, no evidence exists that significant angiogenesis occurs in humans with left ventricular hypertrophy (39).

Limitations of the study. It is important to consider other determinants of coronary artery size, such as physical work­ing capacity, vessel dominance and coronary vasomotor tone.

Physical working capacity has been reported to have a direct influence on coronary artery size (7). However, at the follow-up measurements there was an increase in cross­sectional area of the left coronary artery and a decrease in physical working capacity. Thus, it is unlikely that the enlargement in coronary artery size was mediated by the physical working capacity.

It is well known that a close relation exists between myocardial territory size and proximal coronary artery di­ameter (40,41). However, in the present study coronary dominance was the same in control subjects and patients

with aortic valve disease. Moreover, as in a previous study (7), after sUbgrouping the patients according to left or right coronary artery dominance, the same patterns of variations in the left and right coronary artery dimensions were ob­served.

Coronary vasomotor tone is another important factor that influences coronary artery dimensions. We did not evaluate coronary vasomotor tone, but it can be assumed that "over­all" autonomic activity was similar because heart rate was the same in control subjects and patients with aortic valve disease (Table 1). However, we recognize that both humoral and endothelium-derived vasoactive factors may have varied among our patients.

Conclusions. In patients with aortic valve disease, the progression of left ventricular hypertrophy is associated with an increase in left anterior descending and left circumflex dimensions, whereas the size of the right coro­nary artery remains unchanged. Despite the enlargement of the left coronary artery dimension, the cross-sectional area of the left coronary artery per 100 g ofleft ventricular muscle mass decreases with progression of left ventricular hyper­trophy. Hence, the increase in left coronary artery size appears to be inadequate to match the severity of left ventricular hypertrophy.

References I. Salcedo EE, Marwick TH, Korzick DH, Goormastic M, Go RT. Left

ventricular hypertrophy sensitizes the myocardium to the development of ischemia. Eur Heart J 1990;11(suppl G):72-8.

2. Bache RJ, Arentzen CE, Simon AB, Vrobel TR. Abnormalities in myocardial perfusion during tachycardia in dogs with left ventricular hypertrophy: metabolic evidence for myocardial ischemia. Circulation 1984;69:409-17.

3. Eberli FR, RitterM, Schwitter J, Bortone A, Hess OM, Krayenbuehl HP. Coronary reserve in patients with aortic valve disease before and after successful aortic valve replacement. Eur Heart J 1991;12:127-38.

4. Holtz J, Restort! W, Bard P, Bassenge E. Transmural distribution of myocardial blood flow and of coronary reserve in canine left ventricular hypertrophy. Basic Res CardioI1977;72:286-91.

5. Mueller TM, Marcus ML, Kerber RE, Young JA, Barnes RW, Abboud FM. Effect of renal hypertension and left ventricular hypertrophy on the coronary circulation in dogs. Circ Res 1978;42:543-9.

6. Bache RJ, Vrobel TR, Arentzen CE, Ring WS. Effect of maximal coronary vasodilation on transmural myocardial perfusion during tachy­cardia in dogs with left ventricular hypertrophy. Circ Res 1981;49:742-9.

7. Villari B, Hess OM, Meier Ch, et aI. Regression of coronary artery dimensions after successful aortic valve replacement. Circulation 1992; 85:972-8.

8. Rutishauser W, Krayenbuehl HP. Herz. In: Siegenthaler W, ed. Klinische Pathophysiologie. 6th ed. Stuttgart: Georg Thieme Verlag, 1987:576-641.

9. Dodge HT, Sandler H, Baxley WA, Hawley RR. Usefulness and limita­tions of radiographic methods for determining left ventricular volume. Am J Cardioll966;18:10-24.

10. Rackley CE, Dodge HT, Coble YO, Hay RE. A method for determining left ventricular mass in man. Circulation 1964;29:666-71.

11. Buechi M, Hess OM, Kirkeeide RL, et aI. Validation ofa new automatic system for biplane quantitative coronary arteriography. Int J Card Imag­ing 1990;5:93-103.

12. Suter TM, Buechi M, Hess OM, Haemmerli-Saner C, Gaglione A, Krayenbuehl HP. Normalization of coronary vasomotion after percuta­neous transluminal coronary angioplasty. Circulation 1992; 85:86-92.

13. Kirkeeide RL, Gould KL, Parsel L. Assessment of coronary stenoses by

VILLARI ET AL. 1079 lACC Vol. 20, No.5 November I, 1992:1073-9 VENTRICULAR HYPERTROPHY AND CORONARY DIMENSIONS

myocardial perfusion imaging during pharmacologic coronary vasodila­tion. VII. Validation of coronary flow reserve as a single integrated functional measure of stenosis severity reflecting all its geometric dimen­sions.] Am Coli CardioI1988;7:103-13.

14. Gould KL. Identifying and measuring severity of coronary artery steno­sis: quantitative coronary arteriography and positron tomography. Circu­lation 1988;78:237-45.

15. Gould KL, Kirkeeide RL, Buchi M. Coronary flow reserve as a physio­logic measure of stenosis severity. Part I. Relative and absolute coronary flow reserve during changing aortic pressure and cardiac work load. Part II. Determination from arteriographic stenosis dimensions under stan­dardized conditions. ] Am Coli Cardioll990;15:459-74.

16. Gould KL. Coronary Artery Stenosis. New York: Elsevier, 1990:93-107.

17. Gage ]E, Hess OM, Murakami T, Ritter M, Grimm ], Krayenbuehl HP. Vasoconstriction of stenotic coronary arteries during exercise in patients with classic angina pectoris: reversibility by nitroglycerin. Circulation 1986;73:865-76.

18. Bortone AS, Hess OM, Eberli FR, et al. Abnormal coronary vasomotion during exercise in patients with normal coronary arteries and reduced coronary flow reserve. Circulation 1989;79:516-27.

19. Brown BG, Bolson E, Frimer M, Dodge HT. Quantitative coronary arteriography: estimation of dimensions, hemodynamic resistance, and atheroma mass of coronary artery lesions using arteriogram and digital computation. Circulation 1977;55:329-37.

20. Guidelines for coronary arteriography. A report of the American College of Cardiology, American Heart Association Task Force. Assessment of diagnostic and therapeutic cardiovascular procedures. Circulation 1987; 76{suppl A):963A-77A.

21. Stack RS, Rembert ]C, Schirmer B, Greenfield ]C Jr. Relation of left ventricular mass to geometry of the proximal coronary arteries in the dog. Am] CardioI1983;51:1728-31.

22. O'Keefe]H, Owen RM, Bove AA. Influence of left ventricular mass on coronary artery cross-sectional area. Am] CardioI1987;59:1395-7.

23. Strauer BE. Myocardial oxygen consumption in chronic heart disease: role of wall stress, hypertrophy, and coronary reserve. Am ] Cardiol 1979;44:730-40.

24. Roberts CS, Roberts WC. 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. Circulation 1980;62:953-9.

25. Wilens SL, Plair CM, Henderson D. Size of the major epicardial coronary arteries at necroscopy. ]AMA 1966;198:1325-9.

26. Marcus ML, Mueller TM, Easthman CL. Effects of short- and long-term

left ventricular hypertrophy on coronary circulation. Am ] Physiol 1981;241:H358-H62.

27. Ng ASH, Holmes DR, Smith HC, et al. Hemodynamic progression of adult valvular aortic stenosis. Cathet Cardiovasc Diagn 1986;12:145-50.

28. Murakami T, Hess OM, Gage ]E, Grimm ], Krayenbuehl HP. Diastolic filling dynamics in patients with aortic valve disease. Circulation 1986;73: 1162-74.

29. Murray PA, Vatner SF. Reduction of maximal coronary vasodilator capacity in conscious dogs with severe right ventricular hypertrophy. Cire Res 1981;48:27-33.

30. Johnson LL, Sciacca RR, Ellis K, Weiss MB, Cannon Pl. Reduced left ventricular myocardial blood flow per unit mass in aortic stenosis. Circulation 1978;57:582-90.

31. Schwitter], Eberli FR, Ritter M, Turina M, Krayenbuehl HP. Myocardial oxygen consumption in aortic valve disease with and without left ventric­ular dysfunction. Br Heart] 1992;67:161-9.

32. Bache RI, Dai X-Z, Alyono D, Vrobel Th, Homans DC. Myocardial blood flow during exercise in dogs with left ventricular hypertrophy produced by aortic banding and perinephritic hypertension. Circulation 1987;76:835-42.

33. Bache RI, Wright L, Laxson DD, Dai X-Z. Effect of coronary stenosis on myocardial blood flow during exercise in the chronically pressure­overloaded hypertrophied left ventricle. Circulation 1990;81:1967-73.

34. Vatner SF, Shannon R, Hittinger L. Reduced subendocardial coronary reserve: a potential mechanism for impaired diastolic function in the hypertrophied and failing heart. Circulation 1990;8I(suppl I1I):III-8-14.

35. Schoenmackers ]. Die Herzkranzschlagadem bei der arterio-kardialen Hypertrophie. Z Kreislaufforsch 1949;38:321-36.

36. Vogelberg K. Die Lichtungsweite der Koronarostien an normalen und hypertrophen Herzen. Z Kreislaufforsch 1957;46:101-15.

37. Marcus ML, Harrison 00, Chilian WM, et al. Alterations in the coronary circulation in hypertrophied ventricles. Circulation 1987;75(suppll):I-19-25.

38. Tomanek RI, Schalk KA, Marcus ML, Harrison 00. Coronary angio­genesis during long·term hypertension and left ventricular hypertrophy in dogs. Circ Res 1989;65:352-9.

39. Marcus ML. The Coronary Circulation in Health and Disease. New York: McGraw-Hili, 1983:296-99.

40. Vieweg WVR, Alpert ]S, Hagan AD. Caliber and distribution of normal coronary arterial anatomy. Cathet Cardiovasc Diagn 1976;2:269-80.

41. Koiwa Y, Bahn RC, Ritman EL. Regional myocardial volume perfused by the coronary artery branch: estimation in vivo. Circulation 1986;74: 157-63.