Effect of Aortic Stenosis on Oxygen Balance in Partially Ischemic Myocardium

Transcription

1 Effect of Aortic Stenosis on Oxygen Balance in Partially Ischemic Myocardium Gary J. Grover, Ph.D., Peter M. Scholz, M.D., James W. Mackenzie, M.D., and'harvey R. Weiss, Ph.D. ABSTRACT Valvar aortic stenosis can result in myocardial underperfusion with or without coronary obstruction. The purpose of this study was to determine how hearts with valvar aortic stenosis without hypertrophy can maintain their oxygen supplylconsumption balance with partial left anterior descending coronary artery (LAD) occlusion. Open-chested, anesthetized dogs (n = 9) were subjected to mild valvar aortic stenosis and then to a reduction of LAD flow to 50% of baseline, while controls (n = 9) received partial LAD occlusion without aortic stenosis. Blood flows were determined before and after aortic stenosis and after LAD occlusion using radioactive microspheres. The hearts were then removed for microspectrophotometric analysis of regional venous and arterial oxygen saturation. Aortic stenosis resulted in a pressure gradient of approximately 50 mm Hg, representing mild aortic stenosis. Only a slight increase in myocardial blood flow was seen with aortic stenosis. Ischemia resulted in a significant drop in blood flow in control (40%) and aortic stenosis (55%) animals compared with their own preocclusion values. These ischemic region flows were not different from each other. Aortic stenosis itself did not alter oxygen extraction, although partial occlusion similarly increased extraction for all groups in the ischemic zone. The LAD occlusion resulted in a decreased oxygen consumption in the occluded region of all groups, with no differences noted between control and aortic stenosed animals. Thus, mild, acute aortic stenosis without hypertrophy does not appear to significantly increase the severity of an ischemic episode precipitated by partial LAD occlusion. Patients with aortic stenosis often exhibit findings of myocardial ischemia even in the absence of coronary artery disease [l, 21. It is well recognized that, because of chronic pressure overload, the hypertrophied left ventricle exhibits an increased vulnerability to surgically induced global ischemia [3-51. Much evidence suggests that subendocardium is relatively underperfused under From the Department of Surgery and Department of Physiology and Biophysics, Heart and Brain Circulation Laboratory, University of Medicine and Dentistry of New Jersey, Rutgers Medical School, Piscataway. NJ. Accepted for publication Apr 15, Address reprint requests to Dr. Weiss, Heart and Brain Circulation Laboratory, Department of Physiology and Biophysics, UMDNJ-Rutgers Medical School, P.O. Box 101, Piscataway, NJ stress, resulting in a mismatch between increased oxygen demand and decreased supply reserve [6, 71. Previous studies, from our laboratory as well as others, showed that acute valvar aortic stenosis, without accompanying ventricular hypertrophy, results in a relative underperfusion of the subendocardium even though total coronary flow increases with severe degrees of aortic stenosis [7, 81. The problem with aortic stenosis in myocardial ischemia has been studied previously but with often conflicting results depending on the model chosen to create left ventricular hypertrophy. Some studies indicated that changes in transmural myocardial blood flow decrease flow reserve under stress, whereas others have not [ In normal hearts, myocardium possesses both the flow and oxygen extraction reserves that can compensate for partially ischemic regions (7, 121. It is presently unknown how partial coronary artery occlusion affects the regional oxygen supply and consumption balance in aortic stenosis. It was our hypothesis that in the presence of mild to moderate aortic valvar stenosis, partial coronary artery occlusion would worsen subendocardial oxygen supply/ demand balance. We chose a model of acute aortic stenosis without left ventricular hypertrophy to allow evaluation of the effect of alterations in aortic and left ventricular pressure that occur with aortic stenosis separate from changes related to left ventricular hypertrophy. The purpose of the present study was to determine how the oxygen supplykonsumption balance is maintained in myocardium subjected to 50% flow reduction and mild to moderate acute valvar aortic stenosis. Materials and Methods Eighteen mongrel dogs of either sex (19-30 kg) were anesthetized using sodium pentobarbital (30 rng/kg IV). All animals were handled in accordance with the principles outlined by the National Research Council. The trachea was intubated, and each animal was supported with a volume ventilator so as to maintain eucapnia. Both femoral arteries and one vein were catheterized: the vein for drug injection, one artery for recording blood pressure, and the other for blood sampling. One artery was cannulated using a catheter-tip rnicromanometer (model MPL. 500; Millar). Recordings were made on a Gould ES1000 recorder. A left-sided thoracotomy was performed at the fifth interspace, and a left atrial catheter was introduced. A Millar transducer-tipped catheter was introduced into the left ventricle through the apex for measurement of ventricular pressure and determination of rate of change 270 Ann Thorac Surg 43: , Mar 1987

2 271 Grover, Scholz, Mackenzie, Weiss: Aortic Stenosis, O2 Balance, and Ischemia of pressure (dp/dt). Approximately 1 cm of the left anterior descending coronary artery (LAD) was isolated proximal to its first major bifurcation. A screw clamp was placed around the distal side of the isolated artery, and an electromagnetic flow probe (Biotronex Laboratories) was placed around the proximal end. A right-sided thoracotomy was then performed through the fourth interspace in preparation for the induction of aortic valvar stenosis. After a period of stabilization, simultaneous measurements of LAD flow, using the flow probe, arterial and ventricular pressures, and heart rate were made. Arterial blood samples were withdrawn and analyzed for ph, carbon dioxide tension, and oxygen tension using a Corning (model 165-2) blood gas analyzer and for hemoglobin on a CO oximeter (Instrumentation Laboratories). At this time regional myocardial blood flow was determined by injection of a dose of 1 to 2 x lo6 carbonized '41Ce-labeled microspheres ( pm in diameter; 3M Co.) followed by a saline flush. A reference sample method was used to obtain myocardial blood flows [13]. The sample was obtained from the femoral artery catheter with a peristaltic pump set at 5 ml/min. In the animals undergoing aortic stenosis, the right atrial appendage was retracted caudally through the right thoracic incision and the base of the right atrium was dissected from the right side of the aorta. The noncoronary sinus then was made visible. Two doublearmed vascular sutures with pledgets were used to plicate the noncoronary sinus and valve cusp. The needle entered the aorta at the left posterior extreme of the sinus. The second needle was put in the aorta in a more anterior position. A second pledget was threaded over the suture, which was then tied down tightly. This technique was adapted from the method of Sink and colleagues [14]. A mild degree of stenosis was sought for this study (left ventricular-aortic systolic pressure gradient of approximately 50 mm Hg). Control animals were treated similarly, including dissection of the aortic root, but they did not undergo aortic stenosis. After stabilization (15-20 minutes), aortic stenosed animals were reinjected with microspheres (85Sr) and blood flow was determined. Hemodynamic and blood gas parameters also were determined in these animals. Control and aortic stenosed animals then had their LADS partially occluded such that LAD flow, as measured by the electromagnetic flow probe, was 50% of preocclusion values. After 10 minutes, blood flows, as measured by microspheres (46Sc), and hernodynamic parameters were again determined in all animals. After these measurements, all the hearts were rapidly excised at the atrioventricular ring with shears, then quickly frozen and stored in liquid nitrogen. Cutting the heart this way ensured that freezing began simultaneously on both sides of the ventricular wall. It has been shown that the time required for freezing is not sufficient for alteration of oxygen saturation of small blood vessels [ 151. For flow determinations, the left ventricles were di- vided into four sections: subepicardium and subendocardium, in the occluded and nonoccluded regions. The ischemic regions in these frozen hearts could easily be discerned because the myocardium was discolored. The transmural pieces of myocardium were cut approximately half to obtain subepicardial and subendocardial pieces. The thin, external layers of the subepicardium and subendocardium were removed. The reference arterial blood samples and heart muscle samples were weighed and placed in counting tubes. The activity of the microspheres in these regions and blood samples was then determined on a Hewlett-Packard Autogamma Spectrometer. Blood flow was expressed as milliliters per minute per 100 g of tissue. To measure oxygen saturation in frozen arterial and venous blood vessels, a three-wavelength microspectrophotometric method was used as described previously (15, 161. Hearts were cut on a bandsaw at - 20 C, and plugs were obtained from the occluded (LAD) and nonoccluded (circumflex) regions of the left ventricular free wall adjacent to the plugs used for flow determinations. After the plugs were cut to a convenient size, they were mounted with an embedding medium (OCT compound; Lab-Tek Products, Naperville, IL). Then 25-pm sections were cut on a rotary microtome at -20 C in a nitrogen atmosphere, transferred to precooled glass slides, and covered with degassed silicone oil and a coverglass. These slides were placed on a Zeiss microspectrophotometer fitted with an N2-flushed cold stage to obtain readings of optical density at 560, 523, and 506 nm. The slit width was set at a 5-nm band pass, and the size of the measuring spot was 12 pm. Readings were obtained to determine oxygen saturation in the first seven arteries (Sa02) and seven veins (Sv02) that were 20 to 100 pm in diameter found in each of the four left ventricular regions: subepicardium and subendocardium, for the occluded (LAD) and nonoccluded (circumflex) regions. The oxygen content of the blood was obtained by multiplying the percentage of oxygen saturation by the hemoglobin concentration times By use of the Fick principle, the paired product of oxygen extraction and blood flow was obtained to determine oxygen consumption on a regional basis within the heart. The SaO2 in the nonischemic region was used in the ischemic region, and the actual ischemic region Sv02 was used for the oxygen extraction calculation. The use of SaO2 in the nonischemic region for determination of the ischemic regional oxygen extraction has been performed previously in our laboratory and is done to account for extraction of oxygen that takes place in the arterioles in ischemic myocardial tissue [ 171. A factorial analysis of variance with repeated measures was used to determine whether differences existed in hemodynamic, blood gas, or flow parameters before and after occlusion or aortic stenosis and between treatments. This analysis was also used to determine differences between treatments and regions for all oxygen supply/consumption parameters. A p value of less than

3 272 The Annals of Thoracic Surgery Vol 43 No 3 March was accepted as significant. All values are expressed asmean? SD. Results Hemodynamic and blood gas values are shown in Table 1. Creation of acute aortic stenosis resulted in significantly reduced systemic pressures when compared with their own pre-aortic stenosis values. When the LAD was partially occluded without aortic stenosis, blood pressure was unchanged in control animals compared with their own baseline values. The LAD occlusion did not significantly reduce pressure in aortic stenosed or control animals. During aortic stenosis alone, left ventricular systolic blood pressure increased from 121? 17 to 157? 38 mm Hg while ventricular end-diastolic pressure stayed relatively constant at 5? 3 mm Hg from a starting pressure of 4? 4 mm Hg. During LAD occlusion plus aortic stenosis, left ventricular systolic blood pressure decreased to 139? 34 mm Hg, which was still higher than the pre-aortic stenosis pressure, while left ventricular end-diastolic pressure increased slightly to 7 * 5 mm Hg. The maximal dp/dt increased significantly from 1,941? 644 to 3,074? 1701 mm Hg/sec with aortic stenosis alone. During LAD occlusion plus aortic stenosis, maximal dp/dt decreased to nearly pre-aortic stenosis values (2,179? 784 mm Hglsec). No differences in blood gas or ph values were noted for any group. Myocardial regional blood flows are given in Table 2. After partial LAD occlusion in controls, flows decreased significantly in the occluded LAD region (approximately 40% decrease overall). Flow remained constant in the nonoccluded region. In animals undergoing aortic stenosis, pre-aortic stenosis flow values were similar in all regions and were also similar to preocclusion flows in control animals. With induction of aortic stenosis, flows increased slightly in all regions, but only significantly in the subendocardium of the circumflex region. Occlusion of the LAD during aortic stenosis resulted in a lower Table 1. Hernodynamic and Arterial Blood Gas Values for Control and Aortic Stenosed Animals Before and After Left Anterior DescendinR Coronary Artery Occlusion Stenosis Control Preocclusion Postocclusion prestenosis (n = 9) (n = 9) (n = 9) Poststenosis (n = 9) Systolic blood pressure (mm Hg) 143 t Diastolic blood pressure (mm Hg) t Heart rate (beatslmin) 156 t Po2 (mm Hg) 71 t Pco2 (mm Hg) t PH 7.35 t z ? 0.13 After Aortic Stenosis and LAD Occlusion (n = 9) All values are mean? SD. Po2 = oxygen tension; Pcoz = carbon dioxide tension; LAD = left anterior descending coronary artery p <.05, compared with values before aortic stenosis. Table 2. Myocardial Regional Blood Flows in Control and Aortic Stenosed Animals Before and After Partial Left Anterior Descending Coronary Artery Occlusion Aortic Stenosis Control After Aortic Stenosis Preocclusion Postocclusion Prestenosis Posts tenosis and LAD Occlusion Variable (n = 9) (n = 9) (n = 9) (n = 9) (n = 9) Nonoccluded (circumflex) region (mvmid100 g) Epicardial : Endocardial t ?: 32b b Occluded (LAD) region (ml/min/100 g) Epicardial ? , Endocardial 69 z z a.b.c AU values are mean 2 SD. LAD = left anterior descending coronary artery <.05, compared with respective nonoccluded region value p <.05, compared with respective prestenosis value. p <.05, compared with respective poststenosis value.

4 273 Grover, Scholz, Mackenzie, Weiss: Aortic Stenosis, O2 Balance, and Ischemia I - T EPN I ENN Arterial (S,02) and venous (S,02) oxygen saturations in the occluded region of the subepicurdium (EPO) and subendocardium (ENO) and the nonoccluded region of the subepicardium (EPN) and subendocardium (ENN) in control and aortic stenosed animals. Asterisk indicates significant difference from its respective nonoccluded region (p C.05). flow in the occluded LAD region compared with the nonoccluded region, and these flows were similar to that seen in controls. These flows were also significantly lower (55%) compared with their own preocclusion values during aortic stenosis. No subepicardial-subendocardial differences were seen for any group. Mean SaOz was 91 t 3% and 89 * 4% in the nonoccluded circumflex regions of control and aortic stenosed animals, respectively, and these values were not significantly different (Figure). Mean SaOz was 78 * 8% and 79 * 6% in the occluded region of control and aortic stenosed animals, respectively, and these values were significantly lower than those in the nonoccluded region. Mean SvOz was 41 * 6% and 43 * 5% for control and aortic stenosed animals, respectively, in the nonoccluded circumflex region (see Figure). Occluded region Aortic Stenosis I EPO EN0 EPN ENN SvOz was 26 * 6% and % for control and aortic stenosed groups, respectively, and these values were significantly lower compared with their respective nonoccluded region values but were similar to each other. Subendocardial SvOz was significantly lower compared with subepicardial SvOz values in all regions of control animals and in the nonoccluded region of aortic stenosed animals. The oxygen extraction data are shown in Table 3. After LAD occlusion, oxygen extraction increased significantly, to the same degree in the occluded LAD region compared with the nonoccluded circumflex region of control and aortic stenosed animals. In all regions, subendocardial oxygen extractions were significantly higher than the values for the subepicardium in control ani- mals. No subepicardial-subendocardial differences were seen in aortic stenosed dogs. The oxygen consumption values are shown in Table 3. In control animals, myocardial oxygen consumption was reduced 31 % in the occluded LAD region compared with the nonoccluded circumflex region. With aortic stenosis, occlusion of the LAD resulted in a 48% reduction in oxygen consumption compared with the nonoccluded Table 3. Myocardial Oxygen Extraction and Consumption in Control and Aortic Stenosed Animals With and Without Left Anterior Descending Coronary Artery Occlusion Variable Control Aortic Stenosis No n o c c 1 u d e d Nonoccluded (circumflex) Occluded (LAD) (circumflex) Occluded (LAD) Region (n = 9) Region (n = 9) Region (n = 9) Region (n = 9) Oxygen extraction (ml O,/dl) Epicardial ? V Endocardial l.mb ,b 7.99? Oxygen consumption (ml 02/min/100 g) Epicardial 7.16? Endocardial All values are means? SD. LAD = left anterior descending coronary artery a p <.05, compared with respective nonoccluded region value. p <.05, compared with respective subepicardial region value.

5 274 The Annals of Thoracic Surgery Vol 43 No 3 March 1987 region, which was not different from the values seen in control animals. The greatest drop in occluded region oxygen consumption was in the subendocardial region of both control and aortic stenosed animals. Comment The most important finding of the present study was that mild to moderate acute aortic valvar stenosis did not worsen the severity of subendocardial underperfusion precipitated by partial coronary artery occlusion in the nonhypertrophied animal models used. Clinical findings in aortic stenosis with compensatory cardiac hypertrophy suggest increased vulnerability of the myocardium to ischemic injury. Despite normal coronary angiograms, patients with left ventricular hypertrophy may be seen with chest pain and electrocardiographic repolarization abnormalities suggestive of subendocardial ischemia. The increased minimum coronary vascular resistance and decreased flow reserves, demonstrated in long-term animal models of aortic stenosis with concomitant ventricular hypertrophy, may be due to several factors. This impedance to flow, especially of the subendocardium, may result from the increased ventricular mass or the increased extravascular ventricular compression caused by valvar aortic stenosis or to both factors [7, 18). Previous studies have shown conflicting responses to aortic stenosis depending on the method used to produce left ventricular hypertrophy. Previous studies of severe acute valvar stenosis showed a decrease in the subendocardial to subepicardial flow ratio in the absence of left ventricular hypertrophy [7, 81. In these studies, left ventricular outflow obstruction was produced by placement of a subvalvar balloon, which differs from obstruction caused by valvar plication. The purpose of the present study was to directly measure the effects of acute aortic stenosis on subepicardial and subendocardial oxygen supply and consumption in the absence of ventricular hypertrophy to separate the effects of increased ventricular mass and extravascular compression on subendocardial perfusion abnormalities. These studies were performed with and without partial coronary occlusion to determine how partially ischemic myocardium may compensate for mild to moderate valvar aortic stenosis. Although rare, acute aortic stenosis can occur because of aortic prosthetic valve thrombosis, following aortic valve replacement, and in patients with previous aortic insufficiency in whom the prosthetic valve is too small and represents an acute partial obstruction. The relative contribution of coexisting arteriosclerotic coronary artery disease, often present in these patients, to the decreased reserve flow encountered with aortic stenosis is unknown. In the present study, mild valvar aortic stenosis without LAD occlusion resulted in a slight decrease in the endocardial to epicardial flow ratio in the region to be occluded (0.94). This degree of redistribution of myocardial blood flow was similar to that seen with moderate aortic stenosis in studies by Vinten-Johansen and Weiss [7]. Falsetti and colleagues (91, however, found scant redistribution with a model of aortic stenosis similar in severity to our own. In another study, Feldman and associates [8] found a much more marked redistribution of flow away from the subendocardium with a moderate valvar stenosis. In the present study, partial occlusion of the LAD resulted in an endocardial to epicardial ratio that was reduced to 0.84 in the occluded region. This value was similar when compared with the ratio seen with LAD occlusion in nonstenosed animals. Feldman and co-workers [8] also found a further drop in this ratio with coronary narrowing in the face of valvar aortic stenosis. Previous studies from our laboratory indicated that a moderate to severe valvar aortic stenosis can result in an increased myocardial oxygen extraction and consumption [7]. This increased oxygen consumption is probably due to increased pressures generated by the left ventricle or increased wall stress resulting from higher ventricular pressures and dilation. In the present study of mild valvar aortic stenosis, neither oxygen extraction nor consumption was increased in the nonoccluded circumflex region. The partial occlusion of the LAD resulted in a significant drop in the left ventricular-aortic pressure gradient as well as in maximal dpldt. In the occluded LAD region, oxygen extraction was significantly higher and oxygen consumption lower in both control and aortic stenosed animals compared with their preocclusion values, particularly in the subendocardium. This effect of ischemia on regional oxygen extraction and consumption has been reported previously [ 171. Interestingly, aortic stenosis did not result in a significant change in oxygen consumption in the occluded region compared with the same region in controls. Acute mild aortic valvar aortic stenosis leads to a modest increase in coronary flow. With ischemia caused by a partial obstruction of a coronary artery, the flow decrement is partially compensated for by an increased oxygen extraction. This helps to maintain oxygen consumption, which falls significantly only in the subendocardium. Similar changes were seen in the stenotic dogs. Thus, ischemia is not worse than controls in our model of acute mild aortic stenosis. The clinical findings of a greater risk of ischemia in patients with long-term aortic stenosis, therefore, may be related to the accompanying hypertrophy or to the seventy of the stenosis. Supported in part by United States Public Health Service grant HL-26919, a grant from the Educational Foundation of America, and a grant-in-aid from the American Heart Association, New Jersey Affiliate. Dr. Grover was supported by Individual National Research Service Award HL References 1. Moller JH, Nakeb A, Edwards JE: Infarction of the papillary muscle and mitral insufficiency associated with congenital aortic stenosis. Circulation 34:87, Fallen EL, Elliot WC, Gorlin R: Mechanisms of angina in aortic stenosis. Circulation 34:87, 1966

6 275 Grover, Scholz, Mackenzie, Weiss: Aortic Stenosis, O2 Balance, and Ischemia 3. Levitsky S, Sloane RE, Mullin EM, et al: Normothermic myocardial anoxia: effects on the canine heart with left ventricular outflow obstruction. Ann Thorac Surg 11:229, Katz AM, Tada M: The "stone heart": a challenge to the biochemist. Am J Cardiol 29:578, Schaper J, Schwarz F, Flameng W, Hehrlein F: Tolerance to ischemia of hypertrophied human hearts during valve replacement. Basic Res Cardiol 73:171, Hoffman JIE, Buckberg GD: Transmural variation in myocardial perfusion. Prog Cardiol 5:37, Vinten-Johansen J, Weiss HR: Oxygen consumption in subepicardial and subendocardial regions of the canine left ventricle: the effect of experimental acute valvular aortic stenosis. Circ Res 46:139, Feldman RL, Nichols WW, Edgerton JR, et al: Influence of aortic stenosis on the hemodynamic importance of coronary artery narrowing in dogs without left ventricular hypertrophy. Am J Cardiol 51:865, Falsetti HL, Verani MS, Cramer JA, et al: Total, phasic, and regional myocardial blood flow in aortic stenosis. Am Heart J 98:331, Attarian DE, Jones RN, Currie LND, et al: Characteristics of chronic left ventricular hypertrophy induced by subcoronary valvular aortic stenosis: 1. Myocardial blood flow and metabolism. J Thorac Cardiovasc Surg 198:382, Alyono D, Anderson RW, Parrish DG, et al: Alterations of myocardial blood flow associated with experimental canine left ventricular hypertrophy secondary to valvular aortic stenosis. Circ Res 58:47, Lipp HH, Weiss HR: Blood flow and relative tissue oxygenation of normal and partially ischaemic myocardium: effect of COz. Clin Exp Pharmacol Physiol 5567, Dole WP, Jackson DL, Rosenblatt JI, et al: Relative error and variability in blood flow measurements with radiolabeled microspheres. Am J Physiol 243:H371, Sink JD, Attarian DE, Chitwood WR, et al: An improved technique for producing ventricular hypertrophy with a subcoronary valvular aortic stenosis model. Ann Thorac Surg 30:285, Weiss HR, Sinha AK: Regional O2 saturation of small arteries and veins in the canine myocardium. Circ Res 42:110, Sinha AK, Neubauer JA, Lipp JA, et al: Blood oxygen saturation determination in frozen tissue. Microvasc Res 13:133, Weiss HR: Effect of coronary artery occlusion on regional arterial and venous O2 saturation, O2 extraction, blood flow, and O2 consumption in the dog heart. Circ Res 47400, Blumenthal MR, Wang H, Wang SC: Effect of acute experimental aortic stenosis on coronary circulation. Circ Res 11:727, 1962