George B. Haasler

Intraoperative 2-dimensional echocardiography: Ejection of microbubbles from the left ventricle after cardiac surgery

Although the presence of entrapped air in the left ventricle at surgery has long been a concern, its prevalence and clinical significance are unknown. Two-dimensional echocardiography provides a unique means for observing intracardiac microbubbles intraoperatively. Accordingly, intraoperative 2-dimensional echocardiograms obtained in 79 patients immediately after cardiopulmonary bypass were examined for the presence of microbubbles. Left ventricular microbubbles were detected in 14% of coronary artery bypass operations (n = 22), 50% of multiple valve replacements (n = 8), 67% of single valve replacements (n = 58), and 100% of atrial septal defect closures (n = 4), left atrial myxoma resections (n = 2), and mitral commissurotomies (n = 2). Microbubbles were not observed in patients undergoing coronary artery bypass with valve replacement (n = 4) or with removal of left ventricular aneurysm (n = 1). Continuous ejection of microbubbles from the left ventricle was frequently noted for extended periods of time and, although no patient in this study showed evidence of gross neurologic dysfunction, the possibility of subclinical organ damage exists. We conclude that 2-dimensional echocardiography is useful for detection of intracardiac microbubbles and therefore may be instrumental in their elimination.

Compliance, mass, and shape of the canine left ventricle after global ischemia analyzed with two-dimensional echocardiography☆

End-diastolic pressure (EDP)-volume (EDV) curves were constructed for the left ventricle (LV) in 10 dogs on cardiopulmonary bypass before and after 45 min of normothermic ischemic arrest. LVEDV and mass were calculated from three perpendicular two-dimensional echocardiographic sections using a Simpsons rule algorithm. The echocardiographic method employed was specially developed in earlier work for physiologic studies in dogs. Postmortem pressure-volume curves were measured for comparison by direct saline infusion into sealed ventricles. LVEDV was plotted against LVEDP (0-20 mm Hg) and data were analyzed by two-way analysis of variance and also by exponential curve fitting to the relation EDP = alpha e beta EDV, where beta is the derived exponential constant reflecting slope and thereby ventricular stiffness. Postischemic EDP-EDV curves shifted to the left and the beta constant increased (beta = 0.02 +/- 0.005 (SE) vs 0.05 +/- 0.01, P less than 0.05). LV eccentricity, the ratio between long- and short-axis measurements, was unchanged (1.43 +/- 0.05 vs 1.46 +/- 0.06 postischemia, NS), but LV mass increased (105 +/- 7 vs 128 +/- 9 g postischemia, P less than 0.001). Postischemic LV mass by echo was not significantly different from measured postmortem LV mass (r = 0.87). The nonparallel alteration in the EDP-EDV curves, lack of change in LV shape (eccentricity), and increase in LV mass in the postischemic ventricle suggest that increased ventricular stiffness rather than direct encroachment by the edematous endocardium on chamber volume is responsible for postischemic changes in LV diastolic properties.

Canine left ventricular volume and mass during thoracotomy by two-dimensional echocardiography: Increased ventricular mass after ischemia and reperfusion☆

Left ventricular end-diastolic volume (Vc) and mass (Mc) were studied in 10 dogs by two-dimensional echocardiography after thoracotomy and construction of a pericardial well filled with sound-conducting gel. Five selected LV sections were obtained in vivo by two-dimensional ultrasound. After X-ray angiography and thermal dilution cardiac outputs, 9 of the 10 hearts were fixed by glutaraldehyde coronary perfusion while filling pressures were maintained equal to LVEDP and RVEDP in vivo. Silastic casts of the left ventricle were used to measure cavity volume (Vc) by displacement, and LV mass was determined by weighing (Vpm). Study of postmortem casts with a Simpsons rule algorithm revealed that volume calculations with two of six possible pairs of orthogonal sections were congruent with the line of identity vs cast displacement volume. Optimum results in vivo were obtained from averaged data based on these section pairs (from three echo sections) as Vc = 0.94Vc + 0.2 (r = 0.88). LV mass was given by Mc = 0.95Mpm + 6.6 (r = 0.90). Systolic volume, stroke volume, and ejection fraction were less satisfactory, reflecting difficulties with long axis sections during systole in the open chest. To examine the suitability of these methods for physiologic studies, LV mass was measured before and after 30 min of normothermic ischemic arrest and reperfusion in six dogs. LV mass increased from 109 ± 7 (SE) g before ischemia to 153 ± 13 g after ischemia (P < 0.01). Two-dimensional echocardiography thus demonstrates unique potential for experimental cardiovascular physiology.

Heart-rate effects on canine left-ventricular end-diastolic compliance measured by two-dimensional ultrasound

Despite recent interest in left ventricular (LV) compliance during open-heart surgery, the effect of heart rate (HR) remains unclear due to technical limitations of previous studies. In the presence of Formalin-induced complete atrioventricular (AV) block, AV sequentially paced heart rate was varied between 100, 125, and 150 beats/min in eight dogs on cardiopulmonary bypass. The heart was cradled in an open pericardial well filled with ultrasound-conducting gel. LV end-diastolic pressure (LVEDP) was varied over 0-20 mm Hg by flow regulation in the bypass circuit. Left ventricular end-diastolic volume (LVEDV) was calculated from three orthogonal two-dimensional (2-D) echocardiographic sections using Simpsons rule. Exponential curves derived from the relation LVEDP = ke(bxLVEDV) demonstrated a small, but significant increase in the mean exponential constants (b = 0.034, 0.037, and 0.049 at HR 100, 125, and 150, respectively) reflecting progressive stiffening with increasing heart rate. However, mean overall pressure-volume relations were not significantly altered when analyzed in four separate intervals of LVEDP. It is concluded that although LV volume determinations with 2-D echocardiography demonstrate a steady trend toward decreased end-diastolic LV compliance with increasing rate, this change is so small as to make heart rate a minor determinant of observed intraoperative compliance changes.

Effects of Simulated Clinical Cardiopulmonary Bypass and Cardioplegia on Mass of the Canine Left Ventricle

Myocardial edema is a well-documented response to ischemia and reperfusion in dogs and can be detected as an increase in left ventricular (LV) mass measured echocardiographically. Investigation in human beings has failed to demonstrate similar increases in LV mass after routine cardiac operations. However, direct comparison of these observations is not possible, because dogs have not been studied under conditions of cardiopulmonary bypass (CPB) and global ischemia that rigorously reproduce those of a clinical operation. Accordingly, clinical CPB and global ischemia were simulated in 8 adult dogs. Multisection two-dimensional echocardiograms and a computerized light pen were used to calculate LV volume and mass. The data were analyzed during four periods: before CPB; CPB before cross-clamping; CPB after cross-clamping, and after CPB. Echocardiographic LV mass increased significantly from 89.5 +/- 7.4 gm (before CPB) to 94.6 +/- 8.5 gm (CPB before cross-clamping) (p less than 0.05) and from 94.6 +/- 8.5 gm to 100.4 +/- 8.1 gm (CPB after cross-clamping) (p less than 0.05). Overall, mass increased 13.9% (p less than 0.05). Mean wall thickness increased 0.08 cm (p less than 0.05) overall and correlated with the increase in mass (r = 0.79). Postmortem mass correlated well with late echocardiographic mass (r = 0.95). Interobserver correlation of 98 separate mass determinations was high (r = 0.84) with a coefficient of variation of 10.5%. We conclude that canine LV mass increases significantly with CPB and global ischemia maintained within clinical standards and that human beings and dogs may differ in extent of edema during CPB.

Controlled Reperfusion Following Regional Ischemia

The ability to reverse acute coronary occlusion with fibrinolytic agents and percutaneous transluminal angioplasty has increased interest in the revascularization of ischemic myocardium. This study defines changes in global ventricular function, mass, and compliance during acute coronary occlusion and following reperfusion with blood in the beating and arrested heart. In 17 dogs on cardiopulmonary bypass, the proximal left anterior descending coronary artery was occluded for 45 minutes. In 12 dogs, flow was reestablished by releasing the coronary snare in the beating heart. In the other 5 dogs, the snare was released during a continuous 10-minute infusion of blood potassium cardioplegia in the arrested heart. Coronary occlusion resulted in significant decreases in stroke work index and left ventricular (LV) mass, but compliance was unchanged. Reperfusion in the beating heart increased LV mass compared with the values measured before ischemia (104 +/- 5 versus 95 +/- 5 gm; p less than 0.05) and decreased LV compliance (39 +/- 4 versus 53 +/- 4 ml at LV end-diastolic pressure of 8 mm Hg; p less than 0.05). In contrast, with blood cardioplegia-based reperfusion in the arrested heart, LV mass and LV compliance remained unchanged from control values. We conclude that revascularization of acutely ischemic myocardium in the beating heart further impairs LV function by increasing LV mass and decreasing compliance. This damage can be avoided by reperfusion with blood cardioplegia in the arrested heart.