543
Effect of Renal Hypertension and Left Ventricular Hypertrophy on the Coronary Circulation in Dogs THOMAS M. MUELLER, MELVIN L. MARCUS, RICHARD E. KERBER, JOHN A. YOUNG, ROBERT W. BARNES, AND FRANCOIS M. ABBOUD
SUMMARY The purpose of this study was to investigate the effects of pressure-induced left ventricular hypertrophy on the coronary circulation. Hypertrophy was induced by single-kidney renal vascular hypertension in 12 dogs. Ventricular mass in the dogs with hypertrophy was about 50% greater than in 11 controls. A third group of six dogs, with a similar amount of left ventricular hypertrophy but normal blood pressure after repair of the renal artery stenosis, also was studied. Total and regional myocardial blood flow was measured with radioactive microspheres at rest, during pacing at a rate of 200 (in the control and hypertensive dogs), and during maximal vasodilation induced with adenosine (4.7 juM/kg x min). Results were as follows. (1) Regional distribution of coronary flow was normal at rest in dogs with hypertension and left ventricular hypertrophy and in dogs with hypertrophy alone. (2) Pacing caused at 16% decrease in the endocardial-epicardial perfusion ratio only in the hypertrophied ventricles of the hypertensive dogs. (3) During maximal coronary vasodilation, the coronary vascular resistance of the entire left ventricle was no different among the three groups: the controls, the hypertensive dogs with left ventricular hypertrophy, and the normotensive dogs with left ventricular hypertrophy (0.14 ± 0.02 SEM, 0.16 ± 0.02, and 0.14 ± 0.02 mm Hg/ml x min, respectively). The use of the minimal coronary vascular resistance measured during maximal vasodilation as an index of the functional cross-sectional area of the coronary bed suggests that the cross-sectional area does not increase with hypertrophy. This failure of the cross-sectional area of the coronary bed to increase commensurate with the degree of hypertrophy is due to an anatomical or architectural alteration of the relationship between the coronary bed and the cardiac muscle and is not due to a functional alteration caused by hypertension alone. Thus, the hypertrophied ventricle may be at greater risk of ischemic injury.
CLINICAL SUSPICION of inadequate myocardial perfusion due to pressure-induced hypertrophy of the left ventricle has existed for many years.'"4 Although several investigators have found that blood flow per unit mass of the myocardium of hypertrophied ventricles is normal,5"8 the decreases in capillary density described in hypertrophied myocardium9''" and preliminary reports suggesting relative endocardial hypoperfusion"'12 imply that the regional distribution of coronary flow may be abnormal in hypertrophied ventricles and that such ventricles might be at increased risk for ischemic injury. To study possible physiological limitations of the coronary circulation in pressure-induced hypertrophy, we produced single-kidney renal vascular hypertension in mongrel dogs. After left ventricular hypertrophy developed, we assessed regional myocardial perfusion at rest, during pacing, and during maximal vasodilation produced by an iv adenosine infusion. We also studied another group of dogs at rest and during adenosine infusion which had ventricular hypertrophy induced in the same manner but were rendered normotensive by repair of the renal artery stenosis. In this manner we were able to isolate the effects of hypertrophy From the Department of Internal Medicine, Division of Cardiovascular Disease, Department of Surgery, Division of Cardiovascular and Thoracic Surgery and the Cardiovascular Center, University of Iowa and Iowa City Veterans Administration Hospital, Iowa City, Iowa. This work was partially supported by Program Project HL 14388 and Veterans Administration Grant MRIS 584 5459. Dr. Marcus is the recipient of a Research Career Development Award, HL 00328, from the National Heart, Lung, and Blood Institute. Address for reprints: Thomas M. Mueller, M.D., Department of Internal Medicine, University of Iowa Hospitals, Iowa City, Iowa 52242. Received March 24, 1977; accepted for publication October 25, 1977.
from those of hypertension with respect to myocardial perfusion. Methods Preparation of the Dogs Twenty-four male mongrel dogs weighing 18-31 kg were made hypertensive. This was accomplished by an operation during which the animals were anesthetized with sodium pentobarbital (30 mg/kg, iv). Ventilation was maintained with a Harvard respirator. Bilateral flank incisions were made and, on one side, a nephrectomy was done while, on the contralateral side, a clamp such as that described by Ferrario,13 was placed around the renal artery. While renal blood flow was measured with an electromagnetic flow transducer, the clamp was tightened to a point at which resting renal artery flow was marginally compromised and then was released slightly to allow the control level of blood flow, but no more. Five to 6 weeks after this operation, the dogs underwent a second surgical procedure under sodium pentobarbital anesthesia (30 mg/kg, iv) in which cannulas were placed in the aorta through the internal thoracic artery and the left atrium through its appendage. Pacing leads were attached to the left ventricle in eight of the dogs. The cannulas and wires were tunneled subcutaneously and exteriorized dorsally in a skin button. The 12 control dogs had the thoracic operation but not the renal surgery. Six hypertensive dogs had a third operation within 1 week of the thoracic operation. Prior to this operation, their blood pressures were measured to make certain that they were
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
CIRCULATION RESEARCH
544
hypertensive. They were again anesthetized with sodium pentobarbital and a flank incision was made on the side of their remaining kidney. The renal artery was isolated, the clamp was removed, and the stenotic section of the renal artery was excised. The vessel was repaired with an endto-end anastomosis. Measurement of Myocardial Perfusion
Myocardial perfusion was measured with carbonized microspheres (7-10 /urn in diameter) and labeled with 4li Sc, 85Sr, 95Nb, or m C e . Because of the availability of appropriately labeled microspheres, in eight of the hypertensive and seven of the control dogs, the injection during resting conditions was made with 15 /am microspheres labeled with 125I. For each flow determination, between 3.5 x 10" and 5.5 x 106 microspheres suspended in less than 2 ml of saline were injected over a 10-second period into the left atrium, and the cannula was immediately flushed with 5 ml of saline. Starting 1 minute before injection and continuing until 2 minutes after injection, blood was withdrawn at a rate of 2.06 ml/min with a Harvard pump from the arterial cannula. Prior to injection, the vial containing the microspheres and 1 drop of Tween-80 was vigorously agitated mechanically for at least 4 minutes. Microscopic examination of spheres dispersed in this manner showed that in excess of 98% of the spheres were completely dispersed. After completion of this study, the heart was excised and the free wall of the right ventricle, the left atrium, the great vessels, valves, surface vessels, and epicardial fat were removed. We used the posterior descending artery as a starting point and divided the left ventricle into four levels of eight segments each. Each segment then was divided into three layers —epicardium, midwall, and endocardium. The thicknesses of these layers were approximately equal. To determine whether there was heterogeneity of flow within the endocardial third of the ventricular wall, these segments were, in turn, divided into outer, middle, and inner thirds in 7 control dogs and 10 hypertensive dogs with hypertrophy. The paillary muscles were sectioned into their proximal and distal halves. The relative geometric position of each segment was constant from dog to dog.14 The myocardial segments were weighed to the nearest milligram, placed in plastic tubes, and counted for 4 minutes in a 3-inch well-type sodium iodide y scintillation counter. Reference blood samples were divided so that their counting geometry was similar to that of the myocardial samples. Standard techniques15 were used for isotope separation. Myocardial perfusion was calculated, using the formula: MP =
Q, x 100 x BFr
in which MP = myocardial perfusion (ml/min x 100 g), Qn = counts/g of myocardium, BFr = reference blood flow (the rate of withdrawal from the reference artery), and Cr = the total counts in the reference blood.14 Total left ventricular perfusion was calculated by multiplying the normalized blood flow by the mass in grams of the left ventricle and dividing by 100. Coronary vascular resistance was calculated in two ways. The normalized coro-
VOL. 42, No. 4, APRIL
1978
nary vascular resistance was calculated by dividing the mean aortic pressure by the coronary flow/100 g (ml/min x 100 g), and the resistance of the whole left ventricle was calculated by dividing the mean aortic pressure by the total left ventricular coronary flow (ml/min). Protocol and Hemodynamic Measurements One week to 10 days after the thoracic operation, the dogs were returned to the laboratory for study. They had recovered from their previous surgical procedures, were not anemic, and were free of infection. To avoid possible artifacts induced by anesthesia and acute surgical trauma, the dogs were studied in the unanesthetized state. They were given 6-15 mg of morphine, iv while being readied for the experiment. The morphine was given 60-90 minutes before the first measurement, of regional blood flow. During the experiment, they stood quietly in a harness. Aortic and left ventricular pressures were monitored with Statham p23AA and p23BB pressure transducers placed at midchest level. The electrocardiogram was monitored continuously. The pressure tracings and the electrocardiogram were recorded on a direct-writing oscillographic recorder. Venous blood was taken for determination of hemoglobin and creatinine levels. When the dogs appeared to be accustomed to the laboratory setting, baseline hemodynamic parameters were measured. To measure regional myocardial perfusion, an injection of radioactively labeled microspheres was made in the left atrium. In the eight hypertensive and eight control dogs in which they were implanted, the pacing electrodes were attached to a Grass stimulator. The hearts were paced a rate of about 200 beats/min. Pacing at higher rates usually caused ventricular alternans. When the hemodynamic parameters and the electrocardiogram had stabilized for at least 3 minutes of pacing, a second batch of radioactively labeled microspheres was injected into the left atrium to measure regional myocardial perfusion during pacing. Pacing was continued for 4 minutes after the injection of microspheres. After pacing, the hemodynamic parameters were allowed to return to baseline levels. In 9 of the control dogs, 11 of the hypertensive dogs, and in all 6 of the normotensive dogs with left ventricular hypertrophy, an infusion of adenosine was begun. The adenosine was administered, iv, at a dose rate of 4.7 /*M/kg x min with a Harvard infusion pump. This dose was chosen after preliminary studies indicated that higher doses produced no further coronary vasodilation.* At the onset of the adenosine infusion, the blood pressure decreased and the heart rate increased. When these hemodynamic parameters had stabilized at new levels, a batch of microspheres with a third radioactive label was injected through the left atrial cannula. The adenosine infusion was continued for 4 minutes after the microspheres were injected. The dogs undergoing renal artery repair were studied 3-6 days after reanastomosis of the renal artery. Blood * Myocardial perfusion (ml/min X 100 g) for six control dogs was 545 ± 78 SEM at an adenosine dose of 4.7 /xM/kg x min and 476 ± 20 at 6.0 /iM/kg x min. The perfusion in five hypertensive dogs with hypertrophy was 512 ± 67 and 519 ± 59 for the lower and higher doses, respectively.
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
CORONARY FLOW IN HYPERTROPHY/Mueller et al. pressure was measured daily from the 3rd day, and when it had fallen such that the mean arterial pressure was less than 105 mm Hg, the myocardial perfusion was measured at rest and during adenosine infusion, as outlined above. All studies were carried out within 5 days of renal artery repair. At the conclusion of the experiment, the dogs were anesthetized with sodium pentobarbital (30 mg/kg, iv) and killed with potassium chloride. The chest was opened and the heart was removed. The heart was considered to be hypertrophied if the mass of the trimmed left ventricle was greater than 5.0 g/kg of the body mass of the dogs prior to any operative procedure. Six hypertensive dogs with lesser degrees of left ventricular hypertrophy were excluded from the study. Statistical Analysis
Student's /-test for paired and unpaired data or analysis of variance was used where appropriate to assess the statistical significance of the observed differences. Results are expressed as the mean ± 1 SEM. Results Baseline Characteristics
The left ventricular mass/kg body mass was 4.1 ± 0.1 g/ kg in the controls, 6.1 ± 0.2 in the hypertensive dogs with left ventricular hypertrophy, and 5.6 ± 0.2 in the normotensive dogs with left ventricular hypertrophy. At the 95% probability level, there was a statistical difference between the control group and the two groups with left ventricular hypertrophy. The two groups with hypertrophy also were statistically different from each other (P < 0.05). The dogs were not anemic. The hemoglobin concentration in g/dl was 13.8 ± 0.7 for the control group, 12.6 ± 0.5 for the hypertensive dogs with left ventricular hypertrophy, and 13.3 ± 1.8 for the normotensive dogs with hypertrophy. Renal function was mildly impaired in the hypertensive dogs. Creatinine levels in mg/dl were 0.8 ± 0.5 for the control group, 1.2 ± 0.08 for the hypertensive dogs with left ventricular hypertrophy, and 0.9 ± 0.1 for the normotensive group with left ventricular hypertrophy. The highest creatinine in any of the dogs was 1.8 mg/ dl. The hemodynamic measurements are presented in Table 1A. The resting heart rates for all three groups were less than 100 beats/min, and the left atrial pressures were normal. Mean aortic pressures in the hypertensive group were increased by about 50% compared to controls. The mean aortic pressures of the dogs with the renal artery repaired were significantly lower than pressures of the hypertensive dogs and were not statistically different at the 95% probability level from those of the control dogs. The mean aortic pressure in these dogs before renal artery repair was 138 ± 5 mm Hg. There was no significant difference among the three groups in left ventricular perfusion expressed as ml/min x 100 g under resting conditions. Total myocardial perfusion, due to the greater ventricular mass, was higher in the two groups with left ventricular hypertrophy, but this difference did not reach statistical significance at the 95%
545
probability level. This was probably due to the variability in the data occasioned by the variability of the left ventricular mass of the dogs which were of different sizes. The endocardial-epicardial perfusion ratios were similar in the three groups. When perfusion distribution was examined in small regions of the ventricle in terms of percent of the total left ventricular blood flow, there was no difference among the three groups in any of these regions (Fig. 1). In the most critical of these regional comparisons, the relative proportion of flow to the endocardium did not diminish with increasing ventricular mass (Fig. 2). Because the endocardial third of the hypertrophied ventricles was relatively thicker than that of the controls, we divided the subendocardial third of the left ventricular wall in turn into inner, middle, and outer thirds and the papillary muscles into proximal and distal halves in 7 control and 10 hypertensive dogs. The subendocardial third of both groups was perfused homogenously (Table 2), and therefore, considering the relatively thicker subendocardial third of the hypertrophied ventricles as a unit did not mask a potential relative hypoperfusion of a smaller segment of the subendocardium. The coronary vascular resistance/100 g of left ventricle was higher in the hypertensive dogs but was approximately the same in the controls and the normotensive dogs with left ventricular LEVELS
150,
• Control, n = l2 EH LVH +Hypertension, n=12
HJ LVH,n=6
100 50
A
C
LAYERS
50 00
E*5
50 o Epi
Mid
Endo
WALLS
150 100 50 0
Post
Sept
Ant
Lot
FIGURE 1 This figure shows the blood flow to various regions of the left ventricle at rest expressed as a percentage of the normalized blood flow to the entire ventricle and compares normal ventricles and hypertrophic ventricles in dogs that were hypertensive and dogs with normal blood pressure. The upper panel shows the layers from base (A) to apex (D). The middle panel shows the three layers of the ventricular wall: epicardium, midwall, and endocardium. The bottom panel shows the flow to the posterior, septal, anterior, and lateral walls. There is no significant difference among the three groups for any of the regions.
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
CIRCULATION RESEARCH
546
VOL. 42, No. 4, APRIL
1978
TABLE 1 Hemodynamics and Myocardial Perfusion Rest
Pacing
Adenosine
8 8 0
9 11 6
A. Hemodynamics Number of dogs Control HT + LVH LVH Heart rate (beats/min) Control HT + LVH LVH Mean left atrial pressure (mm Hg) Control HT + LVH LVH Mean blood pressure (mm Hg) Control HT + LVH LVH
12 12 6
89 + 5 91 ± 8 91 ± 8
198 ± 6* 182 ± 12*
119 + 8 125 ± 5 137 ± 5
5.2 + 0.5
6.9 ± 1.3 5.6 ± 0.8
4.5 ± 0.5 3.3 ± 0.3
5.2 ± 0.2 1.0 ± O.ltt 97 ± 3 144 ± 3t 105 ± 5t
1.0 ± O.ltt
105 ± 2 145 ± 8t
72 ± 6 108 ± 4t 80 ± 4t
B. Perfusion Measurements MP/100g(ml/min x 100 g) Control HT + LVH LVH
96 + 9 88 + 8 83 ± 3
182 ± 16* 160 ± 19*
557 ± 86 485 ± 51 453 ± 57
101 + 12 141 + 14 116 ± 8
193 ± 18* 253 ± 32'
551 ± 6 9 786 ± 93 606 ± 46
Endocardial/epicardial perfusion Control HT + LVH LVH
1.32 + 0.04 1.40 + 0.05 1.38 + 0.07
1.29 + 0.13 1.17 ± 0.11*
0.95 ± 0.07 0.94 + 0.03 1.01 ± 0.06
CVR/lOOg [(mmHg/mlmin) x lOOg] Control HT + LVH LVH
1.08 + 0.09 1.70 + 0.15t 1.18 + 0.1 It
0.70 ± 0.09* 1.03 ± 0.10*
0.15 ± 0.02 0.26 ± 0.03t 0.19 ± 0.02
CVR/total LV (mm Hg/mlmin) Control HT + LVH LVH
1.06 + 0.10 1.07 ± 0.10 0.95 + 0.11
0.60 ± 0.11*
0.14 + 0.01 0.16 ± 0.02 0.14 ± 0.01
MP/total LV (ml/min) Control HT + LVH LVH
0.64 + 0.11*
Values are mean ± SEM. HT = hypertension; LVH = left ventricular hypertrophy; LV = left ventricle; MP myocardial perfusion; CVR = coronary vascular resistance * Significant difference in paired comparison between control and pacing (P < 0.05). t Significantly different from control (P < 0.05). t Significant difference between HT + LVH and LVH (P < 0.05).
hypertrophy. The coronary vascular resistance of the total left ventricle, however, was not different among the three groups at rest. Measurements during Pacing (Table 1)
Rapid ventricular pacing did not significantly change mean aortic pressure or left atrial pressure in the control or hypertensive dogs (Table 1A). Pacing did, however, increase myocardial perfusion but there was no significant difference in the perfusion/100 g of myocardium between the ventricles of the hypertensive and control dogs. The endocardial-epicardial perfusion ratio was not significantly
different between the two experimental groups although it was lower in the hypertensive dogs. A paired comparison of the endocardial-epicardial ratio (rest vs. pacing) indicated that in the hypertensive dogs there was a small but statistically significant decrease (P < 0.05) in the proportion of flow to the endocardium, whereas in the controls the endocardial-epicardial perfusion ratio was unchanged. Coronary vascular resistance/100 g of left ventricle remained elevated in the hypertensive dogs compared to the controls during pacing, but the resistance calculated for the total coronary bed was not significantly different between the two groups. In the control dogs, the endocar-
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
CORONARY FLOW IN HYPERTROPHY/Mueller et al.
comparisons found in Figure 1 and Table 2), there were no differences among the three groups. During adenosine infusion, the coronary vascular resistance/100 g dropped considerably. However, the resistance/100 g in the hypertensive dogs was still about 75% above the controls and the normotensive dogs with left ventricular hypertrophy. The coronary vascular resistance for the total left ventricular coronary bed was not different among the three groups during maximal vasodilation.
2.0 r
•H
1.5
A
1.0 Q.
UJ
•z
0.5
c Ui
0
• Control A LVH + Hypertension • LVH
3.0
4.0
5.0
6.0
7.0
8.0
Left Ventricular Mass ( q ) / B o d y Mass (kg)
FIGURE 2 77i« /igwre .s/iovra f/ie relationship between the resting endocardial-epicardial flow ratio and the left ventricular mass normalized for body mass for the three groups: controls, hypertensive dogs with hypertrophy, and normotensive dogs with hypertrophy. It demonstrates that the endocardial-epicardial ratio does not decrease with increasing hypertrophy.
dial coronary vascular resistance/100 g was 0.87 ± 0.08 at rest and 0.57 ± 0.07 during pacing, and the epicardial resistance was 1.03 ± 0.12 at rest and 0.70 ± 0 . 1 2 with pacing. The resistance at rest and during pacing was 1.48 ± 0 . 1 5 and 0.96 ± 0 . 1 0 for the endocardial layer of the hypertensive dogs with left ventricular hypertrophy. For the epicardial layer in the hypertensive dogs, the resistance values were 2.10 ± 0.23 and 1.13 ± 0.13 at rest and during pacing, respectively. Measurements during Adenosine Infusion (Table 1)
Mean arterial pressure compared to resting levels dropped significantly in all three groups of dogs, although mean blood pressure remained higher in the hypertensive group. The heart rate was significantly increased during adenosine infusion in all three groups, but there was no important change in the left atrial pressure. Blood flow to the ventricles of all three groups increased five to six times the resting level. There were no significant differences among the three groups in the endocardial-epicardial perfusion ratios; in each group the ratio was approximately one during adenosine infusion. When we examined perfusion to small areas of the ventricle (similar to the TABLE 2 Distribution of Flow to Small Segments of the Left Ventricular Wall at Rest
Epicardium Midwall Endocardium Outer Middle Inner Papillary muscles Proximal Distal
547
Control
Left ventricular hypertrophy + hypertension (n = 10)
87 + 2 100 + 3
84 ± 2 97 ± 2
112 ± 3 112 + 2 111+2
110± 2 111 ± 2 113 ± 2
123 + 9 137+ 12
140 ± 8 151+11
Results are expressed as mean ± SEM. Values are percentages of flow/100 g of the whole ventricle.
Discussion The major contributions of this study are these findings: (1) at rest, the regional distribution of the myocardial blood flow in the pressure-hypertrophied ventricles that we studied is normal; (2) during pacing, there is a small decrease in the relative proportion of blood flow to the endocardium in the hypertrophied ventricles of the hypertensive dogs; and (3) the functional cross-sectional area of the coronary bed, as estimated by the coronary vascular resistance during maximal coronary vasodilation, does not increase proportionately with the degree of hypertrophy. The advantages in the design of this study are several. The study was conducted in unanesthetized dogs to avoid the effects of anesthesia and acute surgical trauma. Renal hypertension was used to induce hypertrophy because hypertension is the most common cause of clinical left ventricular hypertrophy.16 The degree of hypertrophy produced over the experimental period was considerable; ventricular mass was increased by about 50% on the average, and up to 100% in an occasional dog. Both total and regional coronary perfusion were measured at rest, during the stress of pacing, and during maximal coronary vasodilation. The effects of hypertrophy and of hypertension were separated by the repair of the stenotic renal artery in six dogs and their subsequent return to the normotensive state. There are several potential disadvantages of the experimental design. First, morphine was given in doses of 0.250.30 mg/kg to calm the dogs. The morphine probably did not significantly affect coronary vascular resistance because the dose was relatively small and its effect was largely dissipated before the measurements of blood flow were made.17 Second, the criterion used to define left ventricular hypertrophy was a ratio of left ventricular weight to body weight (>5.0 g/kg). Because the hypertensive dogs had one more surgical procedure than the control dogs, greater weight loss in this group (1-2 kg) could have caused a systematic error in the estimation of the degree of left ventricular hypertrophy. To avoid this, the weight of the dog prior to any operative procedure was used to normalize left ventricular mass. Third, we used microspheres of two sizes (15 /xm and 7-10 /am). Utley et al.18 studied the effects of microspheres of different sizes on regional blood flow measurements and found that 15-/u.m microspheres overestimated relative endocardial flow by 2% when compared to 7- to 10-/xm microspheres. In our hypertensive group, comparing endocardial-epicardial flow ratios, the relative endocardial flow at rest was 16% greater than the endocardial flow during pacing. Since this difference is comparatively large and was not observed in the control dogs, even though
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
548
CIRCULATION RESEARCH
they also were studied in the same way with spheres of two different sizes, it is unlikely that the use of spheres of different sizes accounts for the relative decrease in endocardial flow with pacing in the group with left ventricular hypertrophy and hypertension. Fourth, the interpretation of some of the results of this experiment is dependent upon the ability of the dose of adenosine we used to dilate the coronary bed maximally. We tested this in preliminary experiments in both control dogs and dogs with left ventricular hypertrophy and found that larger doses of adenosine did not cause a further decrease in coronary flow or a further decrease in coronary vascular resistance. In addition, Cobb et al.19 also found that a similar dose of adenosine (4.0 ^.M/kg x min, iv) caused maximal coronary vasodilation. Fifth, there was a significant difference in the weight of the left ventricles normalized for body mass in the hypertensive group compared to the normotensive dogs with hypertrophy. It is possible that some regression of the hypertrophy took place in the 3-5 days between the repair of the renal artery stenosis and study of the dogs. We doubt whether this regression was significant in degree. There was no abrupt drop of blood pressure; the pressures measured on the days after arterial repair but before study were intermediate between those reported for the hypertensive and the post-repair states. Furthermore, regression of hypertrophy has been shown to be a relatively slow process.20-21 Thus, in the aggregate, these considerations regarding our experimental methods and design probably did not alter the results of our study significantly. The minimal coronary vascular resistance of the entire left ventricle can be considered an index of the functional cross-sectional area of the left ventricular coronary vascular bed. If the functional cross-sectional area of the coronary bed had kept pace with the amount of hypertrophy that occurred in the left ventricle —an increase of approximately 50% in mass —we would have expected the minimal coronary vascular resistance to be approximately 30% less in the ventricles of the hypertensive dogs with left ventricular hypertrophy. There was, however, no decrease in the minimal coronary vascular resistance in this group. This failure of the minimal coronary vascular resistance to decrease commensurate with the degree of hypertrophy could occur for several reasons (Fig. 3). There could be an element of myogenic constriction,22 vasoconstriction secondary to a humoral substance,23 or an increase in intramural stress.24 These factors are functional and presumably would be reversed by the normalization of blood pressure. The failure of the functional cross-sectional area to increase with hypertrophy also could be explained by anatomic or architectural alterations in the relationship between the coronary microvascuiature and the cardiac muscle. The altered architectural relationships could be of two types. First, during hypertrophy there might be no new growth of coronary vessels, simply the addition of greater muscle mass around an unchanged coronary bed. There is some evidence, however, that new vessels may form during the process of hypertrophy. Bishop and Melsen25 found evidence of vascular endothelial growth in ventricles subjected to a pressure load. If one assumes that the vascular growth observed was at least, in part,
VOL. 42, No. 4, APRIL
1978
o oo o oo o oo
T
oo oo oo
o o o o O 0
Architectural
o oo o oo o oo
^*
o o o O
0
O
0
O
O
Functiona
FIGURE 3 This figure schematically illustrates the changes in the relationship between the coronary microvascuiature at the level of the critical resistance and the cardiac muscle that could occur due to hypertrophy. The small square represents a piece of normal myocardium. The circles represent the cross-sectional area of the coronary bed. The larger squares represent hypertrophied ventricles. If the cross-sectional area of the coronary bed kept pace with hypertrophy, the relationship would be as shown at the top of the figure. However, we found no change in the cross-sectional area, as shown in the three large squares below. This could occur, as on the lower left, if there were an addition of muscle tissue around an unchanged coronary bed, or the number of parallel vascular channels could increase, as shown in the lower middle drawing, but because of hypertrophy of the vessel wall, the lumen of each channel was smaller. These two changes are anatomical or architectural in nature. The cross-sectional area also could be limited if microvascular growth had taken place and if each of the parallel channels were decreased in caliber by the functional changes of increased systolic compression, a circulating vasoconstrictor substance, or increased myogenic tone, all due to hypertension. Our findings support an anatomical alteration.
new parallel vascular channels, as opposed to remodeling of existing vessels, then some neovascularization could occur in hypertrophy. Therefore, a second explanation would be that the increase in cross-sectional area caused by growth of new parallel vascular channels could be offset by the hypertrophy of vessel walls narrowing the lumen of the critical resistance vessels. Folkow et al.26-27 showed that such hypertrophy of the vessel wall does occur with hypertension in skeletal muscles; a similar phenomenon could occur in coronary vessels. Because the minimal coronary vascular resistance of the normotensive dogs with left ventricular hypertrophy was no different from the control or hypertensive dogs with hypertrophy, this suggests that there is an architectural explanation for the failure of the functional cross-sectional area of the coronary bed to increase commensurate with the degree of hypertrophy. The minimal coronary vascular resistance of the hypertensive group was, however, greater than that of either of the two normotensive groups, although not significantly so at the 95% probability level. It is possible that in this group there was a small functional increase in minimal coronary resistance due to hypertension (increased systolic
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
CORONARY FLOW IN HYPERTROPHY IM ueller et al. compressive forces or a humoral vasoconstrictor) in addition to the larger structural or architectural component discussed above. There is considerable evidence that, in normal ventricles, a greater proportion of the total vasodilatory capacity of the coronary bed is used in providing adequate flow to the subendocardial muscle at rest than to the subepicardial muscle, and that the subendocardial muscle is the portion of the left ventricle particularly at risk for ischemic injury.2* The subendocardial muscle of the hypertrophied left ventricle may be at even greater risk. We have shown the maximal vasodilatory capacity of the coronary bed does not increase in pressure-induced hypertrophy. Since, at rest, total and regional coronary blood flow is normal; a greater portion of the total vasodilatory reserve of the hypertrophied ventricle apparently is used to maintain adequate resting flow. With response to a given amount of stress, the vasodilatory capacity of the coronary bed, and particularly that of the endocardial muscle, would be fully expended in the hypertrophied ventricle at a point at which further vasodilatory capacity remains in the normal subendocardium. Thus, additional demands could lead to ischemia in the subendocardium of the pressure hypertrophied ventricle. Acknowledgments We gratefully acknowledge the technical assistance of Paul McKay, Dennis Lura, Charles Eastham, and Mary Kay Snell, as well as the aid in preparation of the manuscript provided by Jeanne Coffin and Ruth Bonar. We also appreciate the statistical advice provided by Dr. Judy Bean Kondo and Richard Lin and the critical review of this manuscript by Dr. Allyn Mark.
References 1. Buchner F: Qualitative morphology of heart failure. Light and electron microscopic characteristics of acute and chronic failure. Methods Achiev Exp Pathol 5: 60-120, 1971 2. Zoll PM, Wessler S, Schlesinger MJ: Interarterial coronary anastomoses in the human heart, with particular reference to anemic and relative cardiac anoxia. Circulation 4: 797-815, 1951 3. Harris CN, Aronow WS, Parker DP, Kaplan MA: Treadmill stress in left ventricular hypertrophy. Chest 63: 353-357, 1973 4. Hurst JW, Logue RB. Schlant RB, Wenger NK: The Heart, Arteries and Veins, ed 3. New York, McGraw-Hill, 1975, pp 1340-1341 5. Bing RJ, Hammond MM, Handelsman JC, Powers SR, Spencer FC, Eckenhoff JE, Goodale WT, Hafkenschiel JF, Kety SS: The measurement of coronary blood flow, oxygen consumption, and efficiency of the left ventricle in man. Am Heart J 38: 1-24, 1949 6. Brink AJ, Lewis CM: Coronary blood flow energetics, and myocardial metabolism in idiopathic mural endomyocardiopathy (14 pa-
549
tients). Am Heart J 73: 339-349, 1967 7. Rowe GG, Castillo CA, Maxwell GM, Crumpton CW: A hemodynamic study of hypertension including observations on coronary blood flow. Ann Intern Med 54: 405-412, 1961 8. Olson RE, Piatnek DA: Conservation of energy in cardiac muscle. Ann NY Acad Sci 72: 466-478, 1959 9. Wearn JT: The extent of the capillary bed of the heart. J Exp Med 47: 273-292, 1928 10. Rakusan K: Quantitative morphology of capillaries of the heart. Number of capillaries in animal and human hearts under normal and pathological conditions. Methods Achiev Exp Pathol 5: 272-286. 1971 11. Einzig S, Leonard JJ, Tripp MR, Burchell HB, Fox IJ: Regional myocardial blood flow (RMF) in closed-chested dogs with left ventricular hypertrophy (LVH) (abstr). Physiologist 18: 285, 1975 12. O'Keefe DD, Hoffman JIE: Regional coronary blood flow and vascular resistance in experimental left ventricular hypertrophy (LVH) in dogs (abstr). Circulation 54 (suppl II): 43, 1976 13. Ferrario C, Blumel C, Nadzam G, McCubbin J: An externally adjustable renal artery clamp. J Appl Physiol 31: 635-637, 1971 14. Marcus ML, Kerber RE, Ehrhardt J, Abboud FM: Three dimensional geometry of acutely ischemic myocardium. Circulation 52: 254-263, 1975 15. Heymann MA, Payne BD, Hoffman JI, Rudolph AM: Blood flow measurements with radionuclide-labeled particles. Prog Cardiovasc Dis 20: 55-79, 1977 16. Kannel WB, Gordon T, Offutt D: Left ventricular hypertrophy by electrocardiogram. Prevalence, incidence and mortality in the Framingham study. Ann Int Med 71: 89-101, 1969 17. Vatner SF, Braunwald E: Cardiovascular control mechanisms in the conscious state. N Eng J Med 293: 970-976, 1975 18. Utley J, Carlson EL, Hoffman JIE, Martinez HM, Buckberg GD: Total and regional myocardial blood flow measurements with 25jt, I5fi, 9fx, and filtered l-lOp diameter microspheres and antipyrine in dogs and sheep. Circ Res 34: 391-405, 1974 19. Cobb FR, Bache RJ, Greenfield JC: Regional myocardial blood flow in awake dogs. J Clin Invest 53: 1618-1625, 1974 20. Sasayama S, Ross J, Franklin D, Bloor CM, Bishop S, Dilley RB: Adaptations of the left ventricle to chronic pressure overload. Circ Res 38:172-178,1976 21. Marcus ML, Eckberg DL, Braxmeier J, Abboud FM: Ventricular hypertrophy: Relative rates of progression and regression. Clin Res 24:229A, 1976 22. Folkow B: Description of the myogenic hypothesis. Circ Res 14/15 (suppl I): 1279-1283, 1964 23. Green HD, Kepchar JH: Control of peripheral resistance in major systemic vascular beds. Physiol Rev 39: 617-686, 1959 24. Downey JM, Kirk ES: The transmural distribution of coronary blood flow during maximal vasodilation. Proc Soc Exp Biol Med 150: 189193,1975 25. Bishop SP, Melsen LR: Myocardial necrosis, fibrosis, and DNA synthesis in experimental cardiac hypertrophy induced by sudden pressure overload. Circ Res 39: 238-245, 1976 26. Folkow B, Hallback M, Lundgren Y, Weiss L: Background of increased flow resistance and vascular reactivity in spontaneously hypertensive rats. Acta Physiol Scand 80: 93-106, 1970 27. Folkow B, Hallback M, Lundgren Y, Weiss L: Structurally based increase of flow resistance in spontaneously hypertensive rats. Acta Physiol Scand 79: 373-379, 1970 28. Hoffman, JIE, Buckberg GD: Transmural variations in myocardial perfusion. Prog Cardiol 5: 37-39, 1976
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
Effect of renal hypertension and left ventricular hypertrophy on the coronary circulation in dogs. T M Mueller, M L Marcus, R E Kerber, J A Young, R W Barnes and F M Abboud Circ Res. 1978;42:543-549 doi: 10.1161/01.RES.42.4.543 Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 1978 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circres.ahajournals.org/content/42/4/543.citation
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation Research is online at: http://circres.ahajournals.org//subscriptions/
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
Effect of Renal Hypertension and Left Ventricular Hypertrophy on the Coronary Circulation in Dogs THOMAS M. MUELLER, MELVIN L. MARCUS, RICHARD E. KERBER, JOHN A. YOUNG, ROBERT W. BARNES, AND FRANCOIS M. ABBOUD
SUMMARY The purpose of this study was to investigate the effects of pressure-induced left ventricular hypertrophy on the coronary circulation. Hypertrophy was induced by single-kidney renal vascular hypertension in 12 dogs. Ventricular mass in the dogs with hypertrophy was about 50% greater than in 11 controls. A third group of six dogs, with a similar amount of left ventricular hypertrophy but normal blood pressure after repair of the renal artery stenosis, also was studied. Total and regional myocardial blood flow was measured with radioactive microspheres at rest, during pacing at a rate of 200 (in the control and hypertensive dogs), and during maximal vasodilation induced with adenosine (4.7 juM/kg x min). Results were as follows. (1) Regional distribution of coronary flow was normal at rest in dogs with hypertension and left ventricular hypertrophy and in dogs with hypertrophy alone. (2) Pacing caused at 16% decrease in the endocardial-epicardial perfusion ratio only in the hypertrophied ventricles of the hypertensive dogs. (3) During maximal coronary vasodilation, the coronary vascular resistance of the entire left ventricle was no different among the three groups: the controls, the hypertensive dogs with left ventricular hypertrophy, and the normotensive dogs with left ventricular hypertrophy (0.14 ± 0.02 SEM, 0.16 ± 0.02, and 0.14 ± 0.02 mm Hg/ml x min, respectively). The use of the minimal coronary vascular resistance measured during maximal vasodilation as an index of the functional cross-sectional area of the coronary bed suggests that the cross-sectional area does not increase with hypertrophy. This failure of the cross-sectional area of the coronary bed to increase commensurate with the degree of hypertrophy is due to an anatomical or architectural alteration of the relationship between the coronary bed and the cardiac muscle and is not due to a functional alteration caused by hypertension alone. Thus, the hypertrophied ventricle may be at greater risk of ischemic injury.
CLINICAL SUSPICION of inadequate myocardial perfusion due to pressure-induced hypertrophy of the left ventricle has existed for many years.'"4 Although several investigators have found that blood flow per unit mass of the myocardium of hypertrophied ventricles is normal,5"8 the decreases in capillary density described in hypertrophied myocardium9''" and preliminary reports suggesting relative endocardial hypoperfusion"'12 imply that the regional distribution of coronary flow may be abnormal in hypertrophied ventricles and that such ventricles might be at increased risk for ischemic injury. To study possible physiological limitations of the coronary circulation in pressure-induced hypertrophy, we produced single-kidney renal vascular hypertension in mongrel dogs. After left ventricular hypertrophy developed, we assessed regional myocardial perfusion at rest, during pacing, and during maximal vasodilation produced by an iv adenosine infusion. We also studied another group of dogs at rest and during adenosine infusion which had ventricular hypertrophy induced in the same manner but were rendered normotensive by repair of the renal artery stenosis. In this manner we were able to isolate the effects of hypertrophy From the Department of Internal Medicine, Division of Cardiovascular Disease, Department of Surgery, Division of Cardiovascular and Thoracic Surgery and the Cardiovascular Center, University of Iowa and Iowa City Veterans Administration Hospital, Iowa City, Iowa. This work was partially supported by Program Project HL 14388 and Veterans Administration Grant MRIS 584 5459. Dr. Marcus is the recipient of a Research Career Development Award, HL 00328, from the National Heart, Lung, and Blood Institute. Address for reprints: Thomas M. Mueller, M.D., Department of Internal Medicine, University of Iowa Hospitals, Iowa City, Iowa 52242. Received March 24, 1977; accepted for publication October 25, 1977.
from those of hypertension with respect to myocardial perfusion. Methods Preparation of the Dogs Twenty-four male mongrel dogs weighing 18-31 kg were made hypertensive. This was accomplished by an operation during which the animals were anesthetized with sodium pentobarbital (30 mg/kg, iv). Ventilation was maintained with a Harvard respirator. Bilateral flank incisions were made and, on one side, a nephrectomy was done while, on the contralateral side, a clamp such as that described by Ferrario,13 was placed around the renal artery. While renal blood flow was measured with an electromagnetic flow transducer, the clamp was tightened to a point at which resting renal artery flow was marginally compromised and then was released slightly to allow the control level of blood flow, but no more. Five to 6 weeks after this operation, the dogs underwent a second surgical procedure under sodium pentobarbital anesthesia (30 mg/kg, iv) in which cannulas were placed in the aorta through the internal thoracic artery and the left atrium through its appendage. Pacing leads were attached to the left ventricle in eight of the dogs. The cannulas and wires were tunneled subcutaneously and exteriorized dorsally in a skin button. The 12 control dogs had the thoracic operation but not the renal surgery. Six hypertensive dogs had a third operation within 1 week of the thoracic operation. Prior to this operation, their blood pressures were measured to make certain that they were
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
CIRCULATION RESEARCH
544
hypertensive. They were again anesthetized with sodium pentobarbital and a flank incision was made on the side of their remaining kidney. The renal artery was isolated, the clamp was removed, and the stenotic section of the renal artery was excised. The vessel was repaired with an endto-end anastomosis. Measurement of Myocardial Perfusion
Myocardial perfusion was measured with carbonized microspheres (7-10 /urn in diameter) and labeled with 4li Sc, 85Sr, 95Nb, or m C e . Because of the availability of appropriately labeled microspheres, in eight of the hypertensive and seven of the control dogs, the injection during resting conditions was made with 15 /am microspheres labeled with 125I. For each flow determination, between 3.5 x 10" and 5.5 x 106 microspheres suspended in less than 2 ml of saline were injected over a 10-second period into the left atrium, and the cannula was immediately flushed with 5 ml of saline. Starting 1 minute before injection and continuing until 2 minutes after injection, blood was withdrawn at a rate of 2.06 ml/min with a Harvard pump from the arterial cannula. Prior to injection, the vial containing the microspheres and 1 drop of Tween-80 was vigorously agitated mechanically for at least 4 minutes. Microscopic examination of spheres dispersed in this manner showed that in excess of 98% of the spheres were completely dispersed. After completion of this study, the heart was excised and the free wall of the right ventricle, the left atrium, the great vessels, valves, surface vessels, and epicardial fat were removed. We used the posterior descending artery as a starting point and divided the left ventricle into four levels of eight segments each. Each segment then was divided into three layers —epicardium, midwall, and endocardium. The thicknesses of these layers were approximately equal. To determine whether there was heterogeneity of flow within the endocardial third of the ventricular wall, these segments were, in turn, divided into outer, middle, and inner thirds in 7 control dogs and 10 hypertensive dogs with hypertrophy. The paillary muscles were sectioned into their proximal and distal halves. The relative geometric position of each segment was constant from dog to dog.14 The myocardial segments were weighed to the nearest milligram, placed in plastic tubes, and counted for 4 minutes in a 3-inch well-type sodium iodide y scintillation counter. Reference blood samples were divided so that their counting geometry was similar to that of the myocardial samples. Standard techniques15 were used for isotope separation. Myocardial perfusion was calculated, using the formula: MP =
Q, x 100 x BFr
in which MP = myocardial perfusion (ml/min x 100 g), Qn = counts/g of myocardium, BFr = reference blood flow (the rate of withdrawal from the reference artery), and Cr = the total counts in the reference blood.14 Total left ventricular perfusion was calculated by multiplying the normalized blood flow by the mass in grams of the left ventricle and dividing by 100. Coronary vascular resistance was calculated in two ways. The normalized coro-
VOL. 42, No. 4, APRIL
1978
nary vascular resistance was calculated by dividing the mean aortic pressure by the coronary flow/100 g (ml/min x 100 g), and the resistance of the whole left ventricle was calculated by dividing the mean aortic pressure by the total left ventricular coronary flow (ml/min). Protocol and Hemodynamic Measurements One week to 10 days after the thoracic operation, the dogs were returned to the laboratory for study. They had recovered from their previous surgical procedures, were not anemic, and were free of infection. To avoid possible artifacts induced by anesthesia and acute surgical trauma, the dogs were studied in the unanesthetized state. They were given 6-15 mg of morphine, iv while being readied for the experiment. The morphine was given 60-90 minutes before the first measurement, of regional blood flow. During the experiment, they stood quietly in a harness. Aortic and left ventricular pressures were monitored with Statham p23AA and p23BB pressure transducers placed at midchest level. The electrocardiogram was monitored continuously. The pressure tracings and the electrocardiogram were recorded on a direct-writing oscillographic recorder. Venous blood was taken for determination of hemoglobin and creatinine levels. When the dogs appeared to be accustomed to the laboratory setting, baseline hemodynamic parameters were measured. To measure regional myocardial perfusion, an injection of radioactively labeled microspheres was made in the left atrium. In the eight hypertensive and eight control dogs in which they were implanted, the pacing electrodes were attached to a Grass stimulator. The hearts were paced a rate of about 200 beats/min. Pacing at higher rates usually caused ventricular alternans. When the hemodynamic parameters and the electrocardiogram had stabilized for at least 3 minutes of pacing, a second batch of radioactively labeled microspheres was injected into the left atrium to measure regional myocardial perfusion during pacing. Pacing was continued for 4 minutes after the injection of microspheres. After pacing, the hemodynamic parameters were allowed to return to baseline levels. In 9 of the control dogs, 11 of the hypertensive dogs, and in all 6 of the normotensive dogs with left ventricular hypertrophy, an infusion of adenosine was begun. The adenosine was administered, iv, at a dose rate of 4.7 /*M/kg x min with a Harvard infusion pump. This dose was chosen after preliminary studies indicated that higher doses produced no further coronary vasodilation.* At the onset of the adenosine infusion, the blood pressure decreased and the heart rate increased. When these hemodynamic parameters had stabilized at new levels, a batch of microspheres with a third radioactive label was injected through the left atrial cannula. The adenosine infusion was continued for 4 minutes after the microspheres were injected. The dogs undergoing renal artery repair were studied 3-6 days after reanastomosis of the renal artery. Blood * Myocardial perfusion (ml/min X 100 g) for six control dogs was 545 ± 78 SEM at an adenosine dose of 4.7 /xM/kg x min and 476 ± 20 at 6.0 /iM/kg x min. The perfusion in five hypertensive dogs with hypertrophy was 512 ± 67 and 519 ± 59 for the lower and higher doses, respectively.
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
CORONARY FLOW IN HYPERTROPHY/Mueller et al. pressure was measured daily from the 3rd day, and when it had fallen such that the mean arterial pressure was less than 105 mm Hg, the myocardial perfusion was measured at rest and during adenosine infusion, as outlined above. All studies were carried out within 5 days of renal artery repair. At the conclusion of the experiment, the dogs were anesthetized with sodium pentobarbital (30 mg/kg, iv) and killed with potassium chloride. The chest was opened and the heart was removed. The heart was considered to be hypertrophied if the mass of the trimmed left ventricle was greater than 5.0 g/kg of the body mass of the dogs prior to any operative procedure. Six hypertensive dogs with lesser degrees of left ventricular hypertrophy were excluded from the study. Statistical Analysis
Student's /-test for paired and unpaired data or analysis of variance was used where appropriate to assess the statistical significance of the observed differences. Results are expressed as the mean ± 1 SEM. Results Baseline Characteristics
The left ventricular mass/kg body mass was 4.1 ± 0.1 g/ kg in the controls, 6.1 ± 0.2 in the hypertensive dogs with left ventricular hypertrophy, and 5.6 ± 0.2 in the normotensive dogs with left ventricular hypertrophy. At the 95% probability level, there was a statistical difference between the control group and the two groups with left ventricular hypertrophy. The two groups with hypertrophy also were statistically different from each other (P < 0.05). The dogs were not anemic. The hemoglobin concentration in g/dl was 13.8 ± 0.7 for the control group, 12.6 ± 0.5 for the hypertensive dogs with left ventricular hypertrophy, and 13.3 ± 1.8 for the normotensive dogs with hypertrophy. Renal function was mildly impaired in the hypertensive dogs. Creatinine levels in mg/dl were 0.8 ± 0.5 for the control group, 1.2 ± 0.08 for the hypertensive dogs with left ventricular hypertrophy, and 0.9 ± 0.1 for the normotensive group with left ventricular hypertrophy. The highest creatinine in any of the dogs was 1.8 mg/ dl. The hemodynamic measurements are presented in Table 1A. The resting heart rates for all three groups were less than 100 beats/min, and the left atrial pressures were normal. Mean aortic pressures in the hypertensive group were increased by about 50% compared to controls. The mean aortic pressures of the dogs with the renal artery repaired were significantly lower than pressures of the hypertensive dogs and were not statistically different at the 95% probability level from those of the control dogs. The mean aortic pressure in these dogs before renal artery repair was 138 ± 5 mm Hg. There was no significant difference among the three groups in left ventricular perfusion expressed as ml/min x 100 g under resting conditions. Total myocardial perfusion, due to the greater ventricular mass, was higher in the two groups with left ventricular hypertrophy, but this difference did not reach statistical significance at the 95%
545
probability level. This was probably due to the variability in the data occasioned by the variability of the left ventricular mass of the dogs which were of different sizes. The endocardial-epicardial perfusion ratios were similar in the three groups. When perfusion distribution was examined in small regions of the ventricle in terms of percent of the total left ventricular blood flow, there was no difference among the three groups in any of these regions (Fig. 1). In the most critical of these regional comparisons, the relative proportion of flow to the endocardium did not diminish with increasing ventricular mass (Fig. 2). Because the endocardial third of the hypertrophied ventricles was relatively thicker than that of the controls, we divided the subendocardial third of the left ventricular wall in turn into inner, middle, and outer thirds and the papillary muscles into proximal and distal halves in 7 control and 10 hypertensive dogs. The subendocardial third of both groups was perfused homogenously (Table 2), and therefore, considering the relatively thicker subendocardial third of the hypertrophied ventricles as a unit did not mask a potential relative hypoperfusion of a smaller segment of the subendocardium. The coronary vascular resistance/100 g of left ventricle was higher in the hypertensive dogs but was approximately the same in the controls and the normotensive dogs with left ventricular LEVELS
150,
• Control, n = l2 EH LVH +Hypertension, n=12
HJ LVH,n=6
100 50
A
C
LAYERS
50 00
E*5
50 o Epi
Mid
Endo
WALLS
150 100 50 0
Post
Sept
Ant
Lot
FIGURE 1 This figure shows the blood flow to various regions of the left ventricle at rest expressed as a percentage of the normalized blood flow to the entire ventricle and compares normal ventricles and hypertrophic ventricles in dogs that were hypertensive and dogs with normal blood pressure. The upper panel shows the layers from base (A) to apex (D). The middle panel shows the three layers of the ventricular wall: epicardium, midwall, and endocardium. The bottom panel shows the flow to the posterior, septal, anterior, and lateral walls. There is no significant difference among the three groups for any of the regions.
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
CIRCULATION RESEARCH
546
VOL. 42, No. 4, APRIL
1978
TABLE 1 Hemodynamics and Myocardial Perfusion Rest
Pacing
Adenosine
8 8 0
9 11 6
A. Hemodynamics Number of dogs Control HT + LVH LVH Heart rate (beats/min) Control HT + LVH LVH Mean left atrial pressure (mm Hg) Control HT + LVH LVH Mean blood pressure (mm Hg) Control HT + LVH LVH
12 12 6
89 + 5 91 ± 8 91 ± 8
198 ± 6* 182 ± 12*
119 + 8 125 ± 5 137 ± 5
5.2 + 0.5
6.9 ± 1.3 5.6 ± 0.8
4.5 ± 0.5 3.3 ± 0.3
5.2 ± 0.2 1.0 ± O.ltt 97 ± 3 144 ± 3t 105 ± 5t
1.0 ± O.ltt
105 ± 2 145 ± 8t
72 ± 6 108 ± 4t 80 ± 4t
B. Perfusion Measurements MP/100g(ml/min x 100 g) Control HT + LVH LVH
96 + 9 88 + 8 83 ± 3
182 ± 16* 160 ± 19*
557 ± 86 485 ± 51 453 ± 57
101 + 12 141 + 14 116 ± 8
193 ± 18* 253 ± 32'
551 ± 6 9 786 ± 93 606 ± 46
Endocardial/epicardial perfusion Control HT + LVH LVH
1.32 + 0.04 1.40 + 0.05 1.38 + 0.07
1.29 + 0.13 1.17 ± 0.11*
0.95 ± 0.07 0.94 + 0.03 1.01 ± 0.06
CVR/lOOg [(mmHg/mlmin) x lOOg] Control HT + LVH LVH
1.08 + 0.09 1.70 + 0.15t 1.18 + 0.1 It
0.70 ± 0.09* 1.03 ± 0.10*
0.15 ± 0.02 0.26 ± 0.03t 0.19 ± 0.02
CVR/total LV (mm Hg/mlmin) Control HT + LVH LVH
1.06 + 0.10 1.07 ± 0.10 0.95 + 0.11
0.60 ± 0.11*
0.14 + 0.01 0.16 ± 0.02 0.14 ± 0.01
MP/total LV (ml/min) Control HT + LVH LVH
0.64 + 0.11*
Values are mean ± SEM. HT = hypertension; LVH = left ventricular hypertrophy; LV = left ventricle; MP myocardial perfusion; CVR = coronary vascular resistance * Significant difference in paired comparison between control and pacing (P < 0.05). t Significantly different from control (P < 0.05). t Significant difference between HT + LVH and LVH (P < 0.05).
hypertrophy. The coronary vascular resistance of the total left ventricle, however, was not different among the three groups at rest. Measurements during Pacing (Table 1)
Rapid ventricular pacing did not significantly change mean aortic pressure or left atrial pressure in the control or hypertensive dogs (Table 1A). Pacing did, however, increase myocardial perfusion but there was no significant difference in the perfusion/100 g of myocardium between the ventricles of the hypertensive and control dogs. The endocardial-epicardial perfusion ratio was not significantly
different between the two experimental groups although it was lower in the hypertensive dogs. A paired comparison of the endocardial-epicardial ratio (rest vs. pacing) indicated that in the hypertensive dogs there was a small but statistically significant decrease (P < 0.05) in the proportion of flow to the endocardium, whereas in the controls the endocardial-epicardial perfusion ratio was unchanged. Coronary vascular resistance/100 g of left ventricle remained elevated in the hypertensive dogs compared to the controls during pacing, but the resistance calculated for the total coronary bed was not significantly different between the two groups. In the control dogs, the endocar-
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
CORONARY FLOW IN HYPERTROPHY/Mueller et al.
comparisons found in Figure 1 and Table 2), there were no differences among the three groups. During adenosine infusion, the coronary vascular resistance/100 g dropped considerably. However, the resistance/100 g in the hypertensive dogs was still about 75% above the controls and the normotensive dogs with left ventricular hypertrophy. The coronary vascular resistance for the total left ventricular coronary bed was not different among the three groups during maximal vasodilation.
2.0 r
•H
1.5
A
1.0 Q.
UJ
•z
0.5
c Ui
0
• Control A LVH + Hypertension • LVH
3.0
4.0
5.0
6.0
7.0
8.0
Left Ventricular Mass ( q ) / B o d y Mass (kg)
FIGURE 2 77i« /igwre .s/iovra f/ie relationship between the resting endocardial-epicardial flow ratio and the left ventricular mass normalized for body mass for the three groups: controls, hypertensive dogs with hypertrophy, and normotensive dogs with hypertrophy. It demonstrates that the endocardial-epicardial ratio does not decrease with increasing hypertrophy.
dial coronary vascular resistance/100 g was 0.87 ± 0.08 at rest and 0.57 ± 0.07 during pacing, and the epicardial resistance was 1.03 ± 0.12 at rest and 0.70 ± 0 . 1 2 with pacing. The resistance at rest and during pacing was 1.48 ± 0 . 1 5 and 0.96 ± 0 . 1 0 for the endocardial layer of the hypertensive dogs with left ventricular hypertrophy. For the epicardial layer in the hypertensive dogs, the resistance values were 2.10 ± 0.23 and 1.13 ± 0.13 at rest and during pacing, respectively. Measurements during Adenosine Infusion (Table 1)
Mean arterial pressure compared to resting levels dropped significantly in all three groups of dogs, although mean blood pressure remained higher in the hypertensive group. The heart rate was significantly increased during adenosine infusion in all three groups, but there was no important change in the left atrial pressure. Blood flow to the ventricles of all three groups increased five to six times the resting level. There were no significant differences among the three groups in the endocardial-epicardial perfusion ratios; in each group the ratio was approximately one during adenosine infusion. When we examined perfusion to small areas of the ventricle (similar to the TABLE 2 Distribution of Flow to Small Segments of the Left Ventricular Wall at Rest
Epicardium Midwall Endocardium Outer Middle Inner Papillary muscles Proximal Distal
547
Control
Left ventricular hypertrophy + hypertension (n = 10)
87 + 2 100 + 3
84 ± 2 97 ± 2
112 ± 3 112 + 2 111+2
110± 2 111 ± 2 113 ± 2
123 + 9 137+ 12
140 ± 8 151+11
Results are expressed as mean ± SEM. Values are percentages of flow/100 g of the whole ventricle.
Discussion The major contributions of this study are these findings: (1) at rest, the regional distribution of the myocardial blood flow in the pressure-hypertrophied ventricles that we studied is normal; (2) during pacing, there is a small decrease in the relative proportion of blood flow to the endocardium in the hypertrophied ventricles of the hypertensive dogs; and (3) the functional cross-sectional area of the coronary bed, as estimated by the coronary vascular resistance during maximal coronary vasodilation, does not increase proportionately with the degree of hypertrophy. The advantages in the design of this study are several. The study was conducted in unanesthetized dogs to avoid the effects of anesthesia and acute surgical trauma. Renal hypertension was used to induce hypertrophy because hypertension is the most common cause of clinical left ventricular hypertrophy.16 The degree of hypertrophy produced over the experimental period was considerable; ventricular mass was increased by about 50% on the average, and up to 100% in an occasional dog. Both total and regional coronary perfusion were measured at rest, during the stress of pacing, and during maximal coronary vasodilation. The effects of hypertrophy and of hypertension were separated by the repair of the stenotic renal artery in six dogs and their subsequent return to the normotensive state. There are several potential disadvantages of the experimental design. First, morphine was given in doses of 0.250.30 mg/kg to calm the dogs. The morphine probably did not significantly affect coronary vascular resistance because the dose was relatively small and its effect was largely dissipated before the measurements of blood flow were made.17 Second, the criterion used to define left ventricular hypertrophy was a ratio of left ventricular weight to body weight (>5.0 g/kg). Because the hypertensive dogs had one more surgical procedure than the control dogs, greater weight loss in this group (1-2 kg) could have caused a systematic error in the estimation of the degree of left ventricular hypertrophy. To avoid this, the weight of the dog prior to any operative procedure was used to normalize left ventricular mass. Third, we used microspheres of two sizes (15 /xm and 7-10 /am). Utley et al.18 studied the effects of microspheres of different sizes on regional blood flow measurements and found that 15-/u.m microspheres overestimated relative endocardial flow by 2% when compared to 7- to 10-/xm microspheres. In our hypertensive group, comparing endocardial-epicardial flow ratios, the relative endocardial flow at rest was 16% greater than the endocardial flow during pacing. Since this difference is comparatively large and was not observed in the control dogs, even though
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
548
CIRCULATION RESEARCH
they also were studied in the same way with spheres of two different sizes, it is unlikely that the use of spheres of different sizes accounts for the relative decrease in endocardial flow with pacing in the group with left ventricular hypertrophy and hypertension. Fourth, the interpretation of some of the results of this experiment is dependent upon the ability of the dose of adenosine we used to dilate the coronary bed maximally. We tested this in preliminary experiments in both control dogs and dogs with left ventricular hypertrophy and found that larger doses of adenosine did not cause a further decrease in coronary flow or a further decrease in coronary vascular resistance. In addition, Cobb et al.19 also found that a similar dose of adenosine (4.0 ^.M/kg x min, iv) caused maximal coronary vasodilation. Fifth, there was a significant difference in the weight of the left ventricles normalized for body mass in the hypertensive group compared to the normotensive dogs with hypertrophy. It is possible that some regression of the hypertrophy took place in the 3-5 days between the repair of the renal artery stenosis and study of the dogs. We doubt whether this regression was significant in degree. There was no abrupt drop of blood pressure; the pressures measured on the days after arterial repair but before study were intermediate between those reported for the hypertensive and the post-repair states. Furthermore, regression of hypertrophy has been shown to be a relatively slow process.20-21 Thus, in the aggregate, these considerations regarding our experimental methods and design probably did not alter the results of our study significantly. The minimal coronary vascular resistance of the entire left ventricle can be considered an index of the functional cross-sectional area of the left ventricular coronary vascular bed. If the functional cross-sectional area of the coronary bed had kept pace with the amount of hypertrophy that occurred in the left ventricle —an increase of approximately 50% in mass —we would have expected the minimal coronary vascular resistance to be approximately 30% less in the ventricles of the hypertensive dogs with left ventricular hypertrophy. There was, however, no decrease in the minimal coronary vascular resistance in this group. This failure of the minimal coronary vascular resistance to decrease commensurate with the degree of hypertrophy could occur for several reasons (Fig. 3). There could be an element of myogenic constriction,22 vasoconstriction secondary to a humoral substance,23 or an increase in intramural stress.24 These factors are functional and presumably would be reversed by the normalization of blood pressure. The failure of the functional cross-sectional area to increase with hypertrophy also could be explained by anatomic or architectural alterations in the relationship between the coronary microvascuiature and the cardiac muscle. The altered architectural relationships could be of two types. First, during hypertrophy there might be no new growth of coronary vessels, simply the addition of greater muscle mass around an unchanged coronary bed. There is some evidence, however, that new vessels may form during the process of hypertrophy. Bishop and Melsen25 found evidence of vascular endothelial growth in ventricles subjected to a pressure load. If one assumes that the vascular growth observed was at least, in part,
VOL. 42, No. 4, APRIL
1978
o oo o oo o oo
T
oo oo oo
o o o o O 0
Architectural
o oo o oo o oo
^*
o o o O
0
O
0
O
O
Functiona
FIGURE 3 This figure schematically illustrates the changes in the relationship between the coronary microvascuiature at the level of the critical resistance and the cardiac muscle that could occur due to hypertrophy. The small square represents a piece of normal myocardium. The circles represent the cross-sectional area of the coronary bed. The larger squares represent hypertrophied ventricles. If the cross-sectional area of the coronary bed kept pace with hypertrophy, the relationship would be as shown at the top of the figure. However, we found no change in the cross-sectional area, as shown in the three large squares below. This could occur, as on the lower left, if there were an addition of muscle tissue around an unchanged coronary bed, or the number of parallel vascular channels could increase, as shown in the lower middle drawing, but because of hypertrophy of the vessel wall, the lumen of each channel was smaller. These two changes are anatomical or architectural in nature. The cross-sectional area also could be limited if microvascular growth had taken place and if each of the parallel channels were decreased in caliber by the functional changes of increased systolic compression, a circulating vasoconstrictor substance, or increased myogenic tone, all due to hypertension. Our findings support an anatomical alteration.
new parallel vascular channels, as opposed to remodeling of existing vessels, then some neovascularization could occur in hypertrophy. Therefore, a second explanation would be that the increase in cross-sectional area caused by growth of new parallel vascular channels could be offset by the hypertrophy of vessel walls narrowing the lumen of the critical resistance vessels. Folkow et al.26-27 showed that such hypertrophy of the vessel wall does occur with hypertension in skeletal muscles; a similar phenomenon could occur in coronary vessels. Because the minimal coronary vascular resistance of the normotensive dogs with left ventricular hypertrophy was no different from the control or hypertensive dogs with hypertrophy, this suggests that there is an architectural explanation for the failure of the functional cross-sectional area of the coronary bed to increase commensurate with the degree of hypertrophy. The minimal coronary vascular resistance of the hypertensive group was, however, greater than that of either of the two normotensive groups, although not significantly so at the 95% probability level. It is possible that in this group there was a small functional increase in minimal coronary resistance due to hypertension (increased systolic
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
CORONARY FLOW IN HYPERTROPHY IM ueller et al. compressive forces or a humoral vasoconstrictor) in addition to the larger structural or architectural component discussed above. There is considerable evidence that, in normal ventricles, a greater proportion of the total vasodilatory capacity of the coronary bed is used in providing adequate flow to the subendocardial muscle at rest than to the subepicardial muscle, and that the subendocardial muscle is the portion of the left ventricle particularly at risk for ischemic injury.2* The subendocardial muscle of the hypertrophied left ventricle may be at even greater risk. We have shown the maximal vasodilatory capacity of the coronary bed does not increase in pressure-induced hypertrophy. Since, at rest, total and regional coronary blood flow is normal; a greater portion of the total vasodilatory reserve of the hypertrophied ventricle apparently is used to maintain adequate resting flow. With response to a given amount of stress, the vasodilatory capacity of the coronary bed, and particularly that of the endocardial muscle, would be fully expended in the hypertrophied ventricle at a point at which further vasodilatory capacity remains in the normal subendocardium. Thus, additional demands could lead to ischemia in the subendocardium of the pressure hypertrophied ventricle. Acknowledgments We gratefully acknowledge the technical assistance of Paul McKay, Dennis Lura, Charles Eastham, and Mary Kay Snell, as well as the aid in preparation of the manuscript provided by Jeanne Coffin and Ruth Bonar. We also appreciate the statistical advice provided by Dr. Judy Bean Kondo and Richard Lin and the critical review of this manuscript by Dr. Allyn Mark.
References 1. Buchner F: Qualitative morphology of heart failure. Light and electron microscopic characteristics of acute and chronic failure. Methods Achiev Exp Pathol 5: 60-120, 1971 2. Zoll PM, Wessler S, Schlesinger MJ: Interarterial coronary anastomoses in the human heart, with particular reference to anemic and relative cardiac anoxia. Circulation 4: 797-815, 1951 3. Harris CN, Aronow WS, Parker DP, Kaplan MA: Treadmill stress in left ventricular hypertrophy. Chest 63: 353-357, 1973 4. Hurst JW, Logue RB. Schlant RB, Wenger NK: The Heart, Arteries and Veins, ed 3. New York, McGraw-Hill, 1975, pp 1340-1341 5. Bing RJ, Hammond MM, Handelsman JC, Powers SR, Spencer FC, Eckenhoff JE, Goodale WT, Hafkenschiel JF, Kety SS: The measurement of coronary blood flow, oxygen consumption, and efficiency of the left ventricle in man. Am Heart J 38: 1-24, 1949 6. Brink AJ, Lewis CM: Coronary blood flow energetics, and myocardial metabolism in idiopathic mural endomyocardiopathy (14 pa-
549
tients). Am Heart J 73: 339-349, 1967 7. Rowe GG, Castillo CA, Maxwell GM, Crumpton CW: A hemodynamic study of hypertension including observations on coronary blood flow. Ann Intern Med 54: 405-412, 1961 8. Olson RE, Piatnek DA: Conservation of energy in cardiac muscle. Ann NY Acad Sci 72: 466-478, 1959 9. Wearn JT: The extent of the capillary bed of the heart. J Exp Med 47: 273-292, 1928 10. Rakusan K: Quantitative morphology of capillaries of the heart. Number of capillaries in animal and human hearts under normal and pathological conditions. Methods Achiev Exp Pathol 5: 272-286. 1971 11. Einzig S, Leonard JJ, Tripp MR, Burchell HB, Fox IJ: Regional myocardial blood flow (RMF) in closed-chested dogs with left ventricular hypertrophy (LVH) (abstr). Physiologist 18: 285, 1975 12. O'Keefe DD, Hoffman JIE: Regional coronary blood flow and vascular resistance in experimental left ventricular hypertrophy (LVH) in dogs (abstr). Circulation 54 (suppl II): 43, 1976 13. Ferrario C, Blumel C, Nadzam G, McCubbin J: An externally adjustable renal artery clamp. J Appl Physiol 31: 635-637, 1971 14. Marcus ML, Kerber RE, Ehrhardt J, Abboud FM: Three dimensional geometry of acutely ischemic myocardium. Circulation 52: 254-263, 1975 15. Heymann MA, Payne BD, Hoffman JI, Rudolph AM: Blood flow measurements with radionuclide-labeled particles. Prog Cardiovasc Dis 20: 55-79, 1977 16. Kannel WB, Gordon T, Offutt D: Left ventricular hypertrophy by electrocardiogram. Prevalence, incidence and mortality in the Framingham study. Ann Int Med 71: 89-101, 1969 17. Vatner SF, Braunwald E: Cardiovascular control mechanisms in the conscious state. N Eng J Med 293: 970-976, 1975 18. Utley J, Carlson EL, Hoffman JIE, Martinez HM, Buckberg GD: Total and regional myocardial blood flow measurements with 25jt, I5fi, 9fx, and filtered l-lOp diameter microspheres and antipyrine in dogs and sheep. Circ Res 34: 391-405, 1974 19. Cobb FR, Bache RJ, Greenfield JC: Regional myocardial blood flow in awake dogs. J Clin Invest 53: 1618-1625, 1974 20. Sasayama S, Ross J, Franklin D, Bloor CM, Bishop S, Dilley RB: Adaptations of the left ventricle to chronic pressure overload. Circ Res 38:172-178,1976 21. Marcus ML, Eckberg DL, Braxmeier J, Abboud FM: Ventricular hypertrophy: Relative rates of progression and regression. Clin Res 24:229A, 1976 22. Folkow B: Description of the myogenic hypothesis. Circ Res 14/15 (suppl I): 1279-1283, 1964 23. Green HD, Kepchar JH: Control of peripheral resistance in major systemic vascular beds. Physiol Rev 39: 617-686, 1959 24. Downey JM, Kirk ES: The transmural distribution of coronary blood flow during maximal vasodilation. Proc Soc Exp Biol Med 150: 189193,1975 25. Bishop SP, Melsen LR: Myocardial necrosis, fibrosis, and DNA synthesis in experimental cardiac hypertrophy induced by sudden pressure overload. Circ Res 39: 238-245, 1976 26. Folkow B, Hallback M, Lundgren Y, Weiss L: Background of increased flow resistance and vascular reactivity in spontaneously hypertensive rats. Acta Physiol Scand 80: 93-106, 1970 27. Folkow B, Hallback M, Lundgren Y, Weiss L: Structurally based increase of flow resistance in spontaneously hypertensive rats. Acta Physiol Scand 79: 373-379, 1970 28. Hoffman, JIE, Buckberg GD: Transmural variations in myocardial perfusion. Prog Cardiol 5: 37-39, 1976
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
Effect of renal hypertension and left ventricular hypertrophy on the coronary circulation in dogs. T M Mueller, M L Marcus, R E Kerber, J A Young, R W Barnes and F M Abboud Circ Res. 1978;42:543-549 doi: 10.1161/01.RES.42.4.543 Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 1978 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circres.ahajournals.org/content/42/4/543.citation
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation Research is online at: http://circres.ahajournals.org//subscriptions/
Downloaded from http://circres.ahajournals.org/ at University of Michigan--Ann Arbor on April 18, 2015
