Mechanical correlates of the third heart sound.

In seven chronically instrumented conscious dogs, micromanometers measured left ventricular pressure, and ultrasonic dimension transducers measured left ventricular minor-axis diameter; the latter recording was filtered to examine data between 20 and 100 Hz. Acceptable external heart sounds were recorded with a phonocardiographic microphone in four of the seven dogs. With each dog sedated, intubated and mechanically ventilated, data were obtained during hemodynamic alterations produced by volume loading, phenylephrine, calcium infusion and vena caval occlusion. Damped oscillations were noted consistently in the left ventricular diameter waveform toward the end of rapid ventricular filling. These wall vibrations, assessed by the filtered diameter, correlated well with the third heart sound (S3) on the phonocardiogram. The peak frequency of the wall vibrations increased with increased diastolic pressure (p = 0.004), probably reflecting an increase in myocardial wall stiffness. In contrast, the amplitude of the vibrations varied directly with left ventricular filling rate (p = 0.0001). Thus, S3 seemed to be related specifically to ventricular wall vibrations during rapid filling, and the spectra of the amplitude-frequency relation shifted toward the audible range with increases in diastolic pressure, wall stiffness or filling rate. Spectral analysis of S3 may be useful in assessing pathologic changes in myocardial wall properties.

Characteristics of chronic left ventricular hypertrophy induced by subcoronary valvular aortic stenosis. I. Myocardial blood flow and metabolism.

Using a canine model of subcoronary valvular aortic stenosis, we determined myocardial blood flow, high-energy phosphate content, and mitochondrial function in eight hearts with chronic left ventricular hypertrophy. Fourteen normal hearts were used for control data. Myocardial blood flow was determined by injection of tracer microspheres. During cardiopulmonary bypass, left ventricular transmural biopsy specimens were taken for metabolic analyses. Subepicardial and subendocardial content of adenosine triphosphate (ATP) and creatine phosphate (CP) were assayed. Respiratory control indices for isolated mitochondria were measured by use of NAD-linked and FAD-linked substrates. Endocardial blood flow, subendocardial high-energy phosphate content, and respiratory control indices for NAD-linked substrate in the hearts with chronic left ventricular hypertrophy were significantly lower than the normal values. These data provide insight into the metabolic and myocardial blood flow abnormalities occurring in cardiac hypertrophy and provide a framework for understanding the altered response of hypertrophied hearts to ischemia.

Characteristics of chronic left ventricular hypertrophy induced by subcoronary valvular aortic stenosis. II. Response to ischemia.

The increased susceptibility of hearts with chronic left ventricular hypertrophy (CLVH) to damage during ischemia has been suggested but not documented. The purpose of this study was to isolate ischemic events in hearts with CLVH from reperfusion events. Using physiological and biochemical parameters, we compared the rate and extent of myocardial injury during ischemic contracture between eight canine hearts with CLVH induced by subcoronary valvular aortic stenosis and 14 normal canine hearts. Preischemic myocardial blood flow was determined by injection of tracer microspheres. During cardiopulmonary bypass, each heart was instrumented with a left ventricular balloon and made globally ischemic. At control, contracture initiation, and contracture completion left ventricular transmural biopsy specimens were assayed for subepicardial and subendocardial adenosine triphosphate (ATP) and creatine phosphate (CP). Mitochondrial respiratory control indices for NAD-linked and FAD-linked substrates were measured. Preischemic endocardial blood flow in hearts with CLVH was significantly lower than in normal hearts. At control, subendocardial ATP and CP and the respiratory control index for NAD-linked substrate were significantly lower in hearts with CLVH than in normal hearts. Hearts with CLVH reached contracture initiation significantly sooner than normal hearts. All hearts demonstrated significant decreases in high-energy phosphate content and mitochondrial function during ischemia. Reperfusion injury notwithstanding, we concluded that hearts wih CLVH are more susceptible to ischemic injury than are normal hearts, perhaps related to lower endocardial blood flow, lower subendocardial high-energy phosphate stores, and depressed mitochondrial function prior to ischemia.

Response of hypertrophied myocardium to ischemia: correlation with biochemical and physiological parameters.

The increased susceptibility of hypertrophied hearts to ischemic injury during cardiac operations has long been recognized. Although the imbalances in oxygen supply and demand which may occur with hypertrophy during hypotension, ventricular fibrillation, or reperfusion have been extensively studied, the biochemical response of hypertrophied myocardium to ischemia has not been fully elucidated. In the present investigation, rat hearts in which hypertrophy was induced by chronic pressure overload were used to examine the relationship of the physiological parameter, ischemic contracture, to high-energy phosphate content and mitochondrial function during global ischemia. Hypertrophied hearts developed ischemic contracture after significantly shorter duration of ischemia than did normal hearts (5.8 +/- 0.3 minutes versus 10.1 +/- 0.7 minutes). High-energy phosphate content was lower in hypertrophied hearts at control and at ischemic contracture initiation and completion than in normal hearts, whereas mitochondrial function was consistently greater in the hypertrophy group. This investigation demonstrates that the hypertrophied myocardium, independent of flow-related events, is more vulnerable to ischemic injury than normal myocardium and suggests that the increased susceptibility may result from lower high-energy phosphate stores present at the onset of ischemia. The results emphasize the need for rapid cardiac arrest with the induction of ischemia in hypertrophied myocardium and suggest the potential for increasing myocardial high-energy phosphate content in the hypertrophied ventricle by interventions such as arrested perfusion with substrate containing oxygenated cardioplegic solutions prior to the onset of planned ischemia.

Metabolic deterioration during global ischemia as a function of time in the intact normal dog heart.

High-energy phosphate content and mitochondrial function were analyzed at the initiation and completion of ischemic contracture in dog hearts exposed to normothermic ischemia while on cardiopulmonary bypass. Contracture initiation and completion were detected by a balloon catheter placed within the left ventricle. In seven dogs, inner and outer layers of the myocardium were assayed for adenosine triphosphate (ATP) and creatine phosphate (CP). ATP and CP content in these two layers were compared prior to ischemia and at contracture initiation and completion. Inner layer ATP levels were 23.88 +/- 0.73 (mean +/- SM) mu moles/gm dry weight prior to ischemia, 5.14 +/- 0.49 at initiation, and 0.73 +/- 0.2 at completion. Inner layer CP content was 41.29 +/- 0.87 prior to ischemia, 3.49 +/- 0.34 at initiation, and 4.06 +/- 0.48 at completion. Mitochondrial respiratory control indices (RCI) were assayed in a second group of seven dogs prior to ischemia, at contracture initiation, and at contracture completion and were, respectively, 11.5 +/- 1.18, 3.1 +/- 0.43 and 1.76 +/- 0.29 (alpha ketoglutarate as substrate). Despite the specific degrees of metabolic deterioration associated with the events of contracture, ischemic time required to develop contracture initiation and completion was variable, ranging from 29.5 to 72 minutes for initiation and 60.25 to 101 minutes for completion. A third group of five dogs had biopsy specimens taken for ATP at fixed ischemic time intervals, and at 45 minutes of ischemia they were found to have greater ranges in ATP values than the ranges associated with contracture initiation. In contrast to ischemic time, the physiological events of ischemic contracture are reliable predictors of the degree of metabolic injury in the intact dog heart exposed to normothermic ischemic arrest during cardiopulmonary bypass.

Beneficial effects of endotracheal extubation on ventricular performance. Implications for early extubation after cardiac operations.

Early endotracheal extubation has been shown to be a safe postoperative management option in patients having cardiac operations. However, few objective data exist on the response of ventricular performance to early termination of controlled ventilation. Seven patients undergoing routine elective coronary artery bypass grafting or adult repair of atrial septal defect were studied after intraoperative placement of left ventricular micromanometers, left ventricular minor axis dimension crystals, and left atrial and intrapleural pressure catheters. Physiologic data were recorded intraoperatively, during controlled mandatory ventilation in the intensive care unit, and during spontaneous respiration immediately after extubation. Extubation to spontaneous breathing was associated with a significant decline in intrapleural pressure and significant increases in left ventricular end-diastolic diameter, ejection diameter shortening, stroke work, and cardiac output. The augmented left ventricular diastolic filling seemed to result from the fall in intrapleural pressure and perhaps from normalization of right ventricular afterload. The preload recruitable stroke work relationship showed that myocardial contractility remained constant after extubation, and ventricular function improved primarily because of increased preload associated with shifting of the capacitance blood volume toward the chest. Thus endotracheal extubation enhances cardiac performance after uncomplicated cardiac surgical procedures, and by this mechanism early extubation may be clinically beneficial as a routine adjunct to postoperative care.

Physiologic determinants of coronary blood flow during external cardiac massage.

Adequate coronary blood flow is a major determinant for successful resuscitation from cardiopulmonary arrest. To develop compression techniques that optimize coronary blood flow, we implanted in eight dogs electromagnetic flow probes that measured circumflex coronary blood flow and ascending aortic blood flow. Micromanometers measured left ventricular and aortic pressures. Each dog was anesthetized and intubated, and the heart was fibrillated electrically. High-impulse manual chest compressions were performed with the dog in the supine position, and compression rate was varied from 60/min to 150/min. Antegrade coronary blood flow occurred primarily during artificial diastole, and there was a brief period of retrograde coronary blood flow with compression during artificial systole. Cardiac output and diastolic aortic pressure increased with compression rate, significantly augmenting peak coronary blood flow velocity. However, diastolic perfusion time decreased linearly with compression rate and limited coronary perfusion at rates greater than 120/min. As a result, net coronary blood flow during high-impulse manual chest compression was determined primarily by diastolic aortic pressure and diastolic perfusion time. Coronary blood flow was optimized in this model at a compression rate of 120/min.

A physiologic comparison of external cardiac massage techniques.

On the basis of recent investigation, controversy has arisen regarding which of several cardiopulmonary resuscitation methods optimizes hemodynamics. The present study was designed to compare five recently described chest compression techniques: high-impulse manual chest compression at 150/min, mechanical compression at 60/min with simultaneous ventilation, mechanical compression at 60/min with simultaneous ventilation and either systolic or diastolic abdominal compression, and pneumatic vest compression at 60/min. Eight dogs were chronically instrumented with electromagnetic flow probes in the ascending and descending aorta while matched micromanometers measured aortic, left ventricular, and pleural pressures. At study, each dog was anesthetized with morphine, intubated, and the heart was fibrillated by rapid ventricular pacing. The five cardiopulmonary resuscitation methods were performed randomly in each preparation within 7 to 10 minutes of arrest. In four dogs, brachiocephalic blood flow was computed as total cardiac output minus descending aortic blood flow, and in all dogs coronary perfusion pressure was calculated as mean diastolic aortic pressure minus mean diastolic left ventricular pressure. Average cardiac output for seven studies was 662 +/- 61 ml/min with high-impulse manual compression, 340 +/- 46 ml/min with mechanical compression and simultaneous ventilation, 336 +/- 45 ml/min with mechanical compression and simultaneous ventilation with systolic abdominal compression, 366 +/- 52 ml/min with mechanical compression and simultaneous ventilation with diastolic abdominal compression, and 196 +/- 29 ml/min with vest resuscitation (high-impulse manual compression significantly greater than other techniques by multivariate analysis, p less than 0.05). Brachiocephalic blood flow generally followed cardiac output and was statistically the greatest with high-impulse manual compression at 273 +/- 47 ml/min (p less than 0.05). Finally, high-impulse manual compression provided the highest coronary perfusion pressure of 31 +/- 4 mm Hg (p less than 0.05) compared to 23 +/- 2 mm Hg for mechanical compression and simultaneous ventilation, 23 +/- 2 mm Hg for mechanical compression and simultaneous ventilation with systolic abdominal compression, 23 +/- 3 mm Hg for mechanical compression and simultaneous ventilation with diastolic abdominal compression, and 11 +/- 2 mm Hg for vest resuscitation. These data demonstrate that high-impulse manual compression generated physiologically and statistically superior hemodynamics when compared with other methods in this model of cardiopulmonary resuscitation.

Experimental mitral regurgitation. Physiological effects of correction on left ventricular dynamics.

It has been suggested that mitral valve replacement for mitral regurgitation can precipitate acute myocardial failure by increasing left ventricular afterload. However, most studies of this problem have involved anesthesia, acute surgical trauma, or ischemic cardioplegia, each of which can influence myocardial function. The pure hemodynamic consequences of mitral valve replacement were investigated by surgically instrumenting eight dogs with ultrasonic transducers to measure left ventricular diameter, electromagnetic flow probes to measure ascending aortic blood flow, and micromanometers to measure left ventricular and pleural pressures. At the time of implantation, an 8 mm stainless steel shunt was inserted through the left ventricular myocardium at the base of the anterior wall and sutured to the left atrial appendage, producing simulated mitral regurgitation of 20% to 40% of total ventricular output. Balloon occluders were placed around the left atrial shunt and both venae cavae. One to 7 days after implantation, each dog was studied in the conscious state, and data were recorded during acute occlusion of the shunt. After shunt occlusion, left ventricular mean ejection pressure increased significantly in all studies. Systolic wall tension also increased by an average of 8%, diameter shortening decreased by 21%, and forward cardiac output increased by 17%. Thus the higher afterload associated with elimination of mitral regurgitation produced an acute fall in stroke shortening and total left ventricular output. However, forward cardiac output increased in all studies, implying improved pump efficiency and overall cardiac performance. Thus the improvement in pump efficiency associated with restoration of mitral valve competence uniformly increases forward cardiac output despite an increased ventricular afterload and a decreased total stroke volume. Although there may be differences between this relatively acute model and chronic forms of mitral regurgitation encountered clinically, these data suggest that forward cardiac output should increase with correction of mitral regurgitation and that the associated augmentation in afterload is probably not a major factor causing low cardiac output after correction.

Management of an infected mucocele occurring in a bypassed excluded esophageal segment.

An infected mucocele of an excluded retained esophageal segment is an infrequently reported complication of esophageal bypass surgery. The subtle symptoms of this entity and its management are discussed.

Review of coronary-subclavian steal following internal mammary artery-coronary artery bypass surgery.

The syndrome of coronary-subclavian steal through an internal mammary artery graft following coronary artery bypass grafting is rare. We are aware of only eight cases reported in the world literature. The cases of these 8 patients are reviewed, and the case of the ninth patient is described. All patients but 1 have been successfully managed by subclavian-carotid artery bypass.

Dynamic geometry of the intact left ventricle.

Knowledge of left ventricular chamber dynamics is central to our understanding of cardiac physiology. The complicated changes in left ventricular geometry observed in the dog during various phases of the cardiac cycle can be represented as distinct linear relationships between chamber eccentricity and intracavitary volume during diastole and ejection, and probably represent structural properties of the ventricular wall. Chamber geometry of the left ventricle is a major determinant of overall myocardial function. The slope of the radius of curvature (r) to wall thickness (h) relationship is a geometric constant that determines the mural force at any given transmural pressure. Chronic pressure and volume overload produce changes in this geometric relationship as a result of increased mural force resisting ejection. The adaptive mechanism of ventricular hypertrophy in this setting alters the r/h ratio and returns systolic mural force toward normal. Coronary occlusion induces acute changes in regional geometry characterized by holosystolic wall bulging and systolic wall thinning, which shift the r/h relationship upward and to the left. The geometric alteration during ischemia probably increases systolic mural force and could adversely affect myocardial function. Recent studies with patients have shown the r/h ratio to be of value in distinguishing between reversible and irreversible impairment of myocardial performance. Because most myocardial diseases produce major alterations in the structure of the ventricular wall, analysis of dynamic chamber geometry may prove of prognostic value in assessing patients with cardiac disorders.

Effect of acutely increased right ventricular afterload on work output from the left ventricle in conscious dogs. Systolic ventricular interaction.

In seven conscious, chronically instrumented dogs, left ventricular volume was calculated with an ellipsoidal model from the anteroposterior, septal-free wall, and base-to-apex left ventricular dimensions, measured by implanted ultrasonic transducers. Matched micromanometers measured left and right ventricular transmural and transseptal pressures. Ventricular pressures and volumes were varied by inflation of implanted vena caval and pulmonary arterial occluders. When compared with vena caval occlusion at matched left ventricular end-diastolic volumes, graded pulmonary arterial occlusions were associated with higher right ventricular systolic pressures, reduced left-to-right transseptal systolic pressure gradients, and leftward systolic septal displacement, with increased septal-free wall segment shortening (all p less than 0.05). Graded pulmonary arterial occlusions, like vena caval occlusions, reduced left ventricular end-diastolic volume, but left ventricular stroke work at a given end-diastolic volume was greater during pulmonary arterial occlusions (2,674 +/- 380 10(-3) erg) than during vena caval occlusion (1,886 +/- 450 10(-3) erg, p less than 0.05). These data indicate that, while transient pulmonary arterial occlusion reduces left ventricular preload, the concomitant increase in right ventricular systolic pressure, which is the pressure external to the interventricular septal segment of the left ventricle, augments septal shortening and assists left ventricular pump function at a given preload through direct systolic ventricular interaction.

Diastolic anisotropic properties of the left ventricle in the conscious dog.

The role of myocardial anisotropy in determining change in left ventricular shape during diastolic filling has not yet been demonstrated. Therefore, 11 conscious dogs were instrumented with global ultrasonic dimension transducers to measure left ventricular major and minor axis diameters and equatorial wall thickness. Myocardial geometry was represented as a three-dimensional ellipsoidal shell. Left ventricular transmural pressure was measured with micromanometers, and ventricular volume was varied by inflation of vena caval occluders. Left ventricular wall strains and stresses calculated from the ellipsoidal shell model agreed closely with those measured directly by myocardial force and dimension transducers. Unequal normalized diastolic stress-strain relations were observed in the latitudinal, longitudinal, and wall thickness directions, reflecting anisotropic mechanical properties of the myocardium. Although a greater wall stress in the latitudinal versus longitudinal axis was predicted adequately from left ventricular geometry alone, the observed latitudinal strain exceeded longitudinal strain by an amount greater than was predicted by geometric considerations alone, suggesting that myocardial anisotropy contributes significantly to changes in ventricular shape during diastolic filling.

Dynamic ventricular interaction in the conscious dog.

In nine conscious, chronically instrumented dogs, ultrasonic dimension transducers measured left ventricular anterior-posterior and septal-free wall minor axis and major axis diameters. Matched micromanometers measured right and left ventricular transmural and transeptal pressures. Ventricular pressures and volumes were varied by inflation of implanted vena caval and pulmonary artery occluders, and the functional significance of the two types of ventricular interaction, i.e., direct and series, was determined. The left ventricle was represented by a modified ellipsoidal geometry. Left ventricular stroke volume calculated from the dimension data correlated well with that measured directly from ascending aortic electromagnetic flow probes during all interventions (r greater than or equal to 0.96). Partial pulmonary artery occlusion significantly increased right ventricular diastolic and systolic pressures as compared to values obtained during control and venal caval occlusion. During pulmonary artery occlusion, latitudinal septal eccentricity was increased throughout the cardiac cycle compared to control and vena caval occlusion (P less than 0.05), indicating leftward interventricular septal shifting and significant alteration of left ventricular shape. The normalized diastolic pressure-volume curve was shifted to the left with pulmonary artery occlusion compared to control and indicated a decrease in left ventricular chamber compliance. However, when selected cardiac cycles with similar end-diastolic volumes from vena caval and pulmonary artery occlusions were compared, parameters of left ventricular systolic function (stroke volume, percent systolic shortening, peak aortic blood flow, peak left ventricular pressure, and its first derivative) remained relatively constant. These data suggest that mean myocardial fiber length is the major preload determinant of left ventricular systolic function independent of chamber pressure and shape, and that direct ventricular interaction mediated by interventricular septal shifting has minimal effects on systolic myocardial performance in this model. Thus, series ventricular interaction during acute imbalances in biventricular loading, where the output of the right ventricle determines the input of the left, seems to be far more important than direct interaction to the regulation of overall cardiac function.

Diminished stroke volume during inspiration: a reverse thoracic pump.

In 12 conscious dogs, a three-dimensional array of pulse-transit ultrasonic transducers was used to measure left ventricular anterior-posterior minor, septal-free wall minor, and basal-apical major diameters. Matched micromanometers measured left ventricular, right ventricular, and intrapleural pressures. Electromagnetic ascending aortic blood flow and right ventricular transverse diameter were measured in five of the dogs. A major cause of the inspiratory decline in stroke volume in this preparation appeared to be reflex tachycardia and autonomic changes associated with inspiration. However, when heart rate was controlled by atrial pacing or pharmacologic autonomic attenuation (propranolol and atropine), stroke volume still decreased around 10%, with an inspiratory decrease in pleural pressure of 10 mm Hg. Based on the measurements of ventricular dimension, venous return to the right ventricle appeared to be augmented significantly during inspiration and the transient increase in right ventricular volume was associated with leftward interventricular septal shifting and altered diastolic left ventricular geometry. However, left ventricular end-diastolic volume was changed minimally, implying that alterations in preload were not important. Moreover, transmural left ventricular ejection pressure, calculated as intracavitary minus pleural pressure, was not significantly changed, and it seemed that neither pressure nor geometric components of afterload were altered significantly by inspiration. The inspiratory fall in left ventricular stroke volume correlated best with the decline in intracavitary left ventricular ejection pressure referenced to atmospheric pressure. It is hypothesized that during ejection, left ventricular pressure referenced to atmospheric pressure is the hydraulic force effecting stroke volume and that the decline in this effective left ventricular ejection pressure is responsible for the inspiratory fall in stroke volume through a reverse thoracic pump mechanism.

An improved transducer for measurement of cardiac dimensions with sonomicrometry.

Although pulse-transit sonomicrometry is an accurate and sensitive technique for measuring cardiac dimensions, transducer directionality has been a significant problem in certain applications. Construction of an improved ultrasonic dimension transducer is described for the measurement of global cardiac diameters. The transducer consists of a hemispheric piezoelectric crystal with a quarter-wavelength impedance matched faceplate. The in vitro acoustic characteristics of this transducer reveal a substantially improved beam pattern compared to the flat piezoceramic transducer commonly used. This improvement in transducer design greatly facilitates sonic alignment, enhances tracking throughout the cardiac cycle, and allows the measurement of more complex cardiac dimensions.

The deformational characteristics of the left ventricle in the conscious dog.

We studied left ventricular minor and major axis diameters and equatorial wall thickness in eleven conscious dogs with chronically implanted pulse-transit ultrasonic dimension transducers. Left ventricular transmural pressure was measured with micromanometers. Left ventricular volume was varied by inflation of implanted vena caval or aortic occluders. The geometry of the left ventricle was represented as a three-dimensioal ellipsoidal shell. Left ventricular eccentricity was found to be a linear function of ventricular volume during both diastole and ejection. However, the relationship was not the same for diastole and ejection, and during diastole the left ventricle was mre spherical at large volumes and more elliptical at small volumes than during ejection. The rearrangements in geometry observed during isovolumic contraction appeared to be transitional stages from the diastolic to the ejection-phase relationship. Thus, during isovolumic contraction, the left ventricle became more elliptical at large volumes and more spherical at small volumes. These relationships were not altered significantly by increased afterload or inotropic interventions. We also observed that the diastolic deformation of the ventricular chamber occurred in a set and predictable manner that seemed to be determined by the three-dimensional mechanical properties of the myocardium. The geometric inter-relationships of the ventricular wall determined the relationship between diastolic transmural pressure and mural stress. These findings probably reflect basic structural characteristics of the myocardium and provide a convenient method for quantitatively representing the dynamic geometry of the left ventricle.