| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1996;61:350-356
© 1996 The Society of Thoracic Surgeons
Division of Thoracic and Cardiovascular Surgery, University of Louisville, Louisville, Kentucky
Abstract
Background. In patients with postcardiotomy low cardiac output syndromes, right ventricular (RV) failure develops in approximately 25% of patients receiving left ventricular (LV) assist device support. Depressed RV function have been attributed to abnormalities of the RV myocardium, excessive load imposed on the RV during systole or diastole, or obstruction to RV inflow. However, recent studies also suggest that LV function may significantly affect RV function through ventricular interdependence.
Methods. We reviewed the data showing the importance of systolic ventricular interaction. We then related these observations to the RV response during LV assist device support, and present our ideas regarding the mechanisms responsible for this RV failure.
Results. Using an electrically isolated right heart preparation, Damiano observed double-peaked waveforms for RV pressure, and pulmonary artery blood flow occurred over a wide range (0 to 300 ms) of pacing intervals between the LV and RV. Numeric analysis indicated that RV systolic pressure and pulmonary artery blood flow were composed of both RV and LV components, with the LV component dominating (63.5% versus 36.5%).
Conclusions. The experimental studies indicate a very consistent RV response during LV assist device support: a decrease in RV afterload, increased compliance, and decreased contractility. In normal hearts, the net effect is an increase or no change in cardiac output. With a preexisting pathologic condition, the RV responses is qualitatively the same, but anatomic ventricular interaction is accentuated, leading to a greater decrease in RV contractility. The net effect is a decrease in cardiac output, which may require inotropic or RV mechanical support.
The major cause of death after coronary revascularization is postoperative low cardiac output, usually resulting from acute, often reversible, perioperative myocardial injury [1]. Between 4% and 7% of patients who undergo cardiac procedures with cardiopulmonary bypass require hemodynamic support with intraaortic balloon counterpulsation; up to 2% of patients require other forms of mechanical circulatory support [1, 2]. A survival rate of up to 50% has been reported with the use of a left ventricular assist device (LVAD) in patients with postcardiotomy low cardiac output syndromes. However, right ventricular failure will develop in approximately 25% of patients receiving LVAD support [1–3]. The major determinant of survival after LVAD placement is the ability of the right ventricle to provide sufficient output to fill the LVAD itself.
If severe right ventricular failure develops, cardiovascular collapse will occur and the patient will not survive without biventricular support. For example, Pennington and associates reported that in 26 patients with biventricular failure, 16 received only LVAD support and none survived [2]. In the 10 patients who received biventricular support, 5 were weaned and 3 survived. Thus, although patients with isolated left ventricular failure do well with an LVAD only, those with biventricular failure require biventricular mechanical support for survival. Further, because delay in the decision to use an assist device increases morbidity and mortality, predicting which patients are susceptible to right ventricular failure would improve clinical outcome.
Abnormalities of right ventricular function have been classically attributed to primary abnormalities of the right ventricular myocardium, excessive load imposed on the right ventricular during systole or diastole, or obstruction to right ventricular inflow. However, recent studies have also suggested that, in addition to the above mechanisms, left ventricular function may significantly affect right ventricular function. This so-called ventricular interdependence is defined herein as the forces that are transmitted from one ventricle to the other ventricle through the myocardium and pericardium, independent of neural, humoral, or circulatory effects [4]. These ventricular independent effects are immediate as compared with circulatory changes, which require several beats. Ventricular interdependence is a consequence of the close anatomic association between the ventricles: the ventricles are encircled by common muscle fibers, share a septal wall, and are enclosed within the pericardium.
In this article, we will first briefly review the mechanisms for diastolic ventricular interaction, followed by a more in-depth review of recent data showing the importance of systolic ventricular interaction. We will then relate these observations to the right ventricular response during LVAD support, and present our ideas regarding the mechanisms responsible for this right ventricular failure.
Diastolic Ventricular Interaction
The evidence for diastolic ventricular interaction is indisputable, and has been the source of in-depth reviews [4, 5]. Briefly, the volume or pressure in one ventricle can directly influence the volume and pressure in the other ventricle. This phenomenon, ventricular interdependence, was probably first observed by Henderson and Prince in 1914 and later verified in postmortem [6] and in isolated heart preparations [7, 8]. Increased distention of either ventricle during diastole alters the compliance and geometry of the opposite ventricle. Taylor and associates [6] in postmortem hearts and Santamore and colleagues [8] and Elzinga and co-workers [7] in isolated beating hearts observed acute changes in ventricular distensibility caused by changing the volume of the opposite ventricle. As left (or right) ventricular volume and pressure increased, the right (or left) ventricular pressure-volume curve shifted to the left and became steeper. This diastolic interaction occurred even with the pericardium open, although the coupling was stronger with it closed [9].
Diastolic ventricular interdependence is present on a moment-to-moment, beat-to-beat basis; ie, part of the measured diastolic ventricular pressure is caused by the opposite ventricle. Although always present, ventricular interdependence is most apparent with sudden changes in ventricular volume. For example, during spontaneous inspiration, right ventricular dimensions and volume increase [10]. Concomitant with these changes, left atrial transmural pressure increases and the septum moves toward the left ventricle in diastole [10]. Left ventricular end-diastolic volume either remains unaltered or decreases. This increase in filling pressure with a decrease in volume is consistent with a change in left ventricular distensibility. Thus, diastolic interaction is always present and the interactions are large enough to be of physiologic and pathophysiologic importance.
Systolic Ventricular Interaction
In recent years, the evidence for systolic ventricular interaction is becoming indisputable. In the sections that follow, we review the studies demonstrating the existence, magnitude, and mechanisms of systolic ventricular interdependence.
Existence
Figure 1
shows a very simple way to demonstrate systolic ventricular interdependence. Left and right ventricular pressures are recorded, while suddenly in diastole a partial constriction of the aorta is released (Fig 1A, B). On the subsequent systole, not only does right ventricular systolic pressure decrease, but left ventricular systolic pressure also decreases slightly. Because preload was not altered, only systolic ventricular interdependence can explain this decrease in left ventricular pressure. Figure 1 also shows the complementary study for the right ventricle. An aortic constriction is released in diastole, leading on the subsequent systolic contraction to a decrease in left ventricular systolic pressure and also, through ventricular interdependence, to a decrease in right ventricular systolic pressure.
|
Experimental studies have also shown that this systolic interaction is an immediate effect: rapid withdrawal or injections into the left ventricle caused immediate changes in right ventricular pressure and volume outflow [4, 13]. In whole animal studies, Woodard and associates [13] used a ventricular assist device to rapidly withdraw blood from the left ventricle. This withdrawal occurred in systole during a single cardiac cycle. Figure 2 shows the typical response. The withdrawal of blood from the left ventricle caused a rapid decrease in left ventricular pressure. Right ventricular pressure and flow outflow also decreased, resulting in a large change in developed pressure and outflow.
|
Magnitude
Although these studies proved the existence of systolic ventricular interdependence, they did not quantify the magnitude of systolic interdependence on right ventricular function. Are these observations just interesting phenomenon or are they physiologically important with significant clinical implications? Two studies addressed this question and quantified the magnitude of this left ventricular assistance, using an electrically isolated right heart preparation [14, 15]. The electrically isolated right heart preparation allowed for wide variations in the timing interval between right and left ventricular contractions. Double-peaked waveforms for right ventricular pressure and pulmonary artery blood flow occurred over a wide range (0 to 300 ms) of pacing intervals between the left and right ventricles (Fig 3). Numeric analysis indicated that these pressures and volume waveforms were due to two components [14]. One component could be directly related to right ventricular contraction, whereas the other component was directly related to left ventricular contraction. Right ventricular systolic pressure and pulmonary artery blood flow were composed of both right ventricular and left ventricular components, with the left ventricular component dominating. For right ventricular pressure, the left ventricular component was significantly greater than the right ventricular component (63.5% versus 36.5% peak-to-peak value; 65.2% versus 34.8% root-mean-square value). Similarly, for pulmonary artery blood flow, the left ventricular component was significantly greater than the right ventricular component (67.5% versus 32.5% peak-to-peak value; 68.3% versus 31.8% root-mean-square value). This study shows that left ventricular contraction is very important and may be the primary source for right ventricular developed pressure and volume outflow [14].
|
Many investigators have attributed ventricular interdependence to transseptal pressure gradients, septal motion, ventricular wall characteristics, and other factors [7, 10, 16–18]. However, ventricular pressures and volumes are really a consequence of the stress in the ventricular walls. Thus, we believe that ventricular interdependence should be viewed as the balance of forces at the interventricular sulcus [18] (Fig 4). At this junction, the force in the left ventricular free wall is balanced by forces in the right ventricular free wall and interventricular septum. This view implies that changing one wall directly affects all three walls; ie, a direct transfer of forces between the left and right ventricular free walls and septum occurs. Thus, changing left ventricular volume not only alters the left ventricular free wall and septum, but also alters the dimensions of the right ventricular free wall [19].
|
Left Ventricular Assist Devices
The experimental studies indicate a very consistent right ventricular response to left ventricular unloading by cardiac assist devices (LVADs): a decrease in right ventricular afterload, increased compliance, and decreased contractility [20–24]. In normal hearts, the net effect is an increase or no change in cardiac output. In an eloquent canine study, Moon and colleagues [20] implanted tantalum markers into the left and right ventricular walls for computation of biventricular volumes and geometry. With full LVAD support, left ventricular end-diastolic volume decreased and pulmonary artery input impedance decreased. Right ventricular end-diastolic volume increased, while right ventricular end-diastolic pressure decreased. A significant leftward septal shift occurred. The right ventricular pressure, volume, and dimension changes showed an increase in right ventricular compliance. Global right ventricular end-systolic elastance and preload recruitable stroke work decreased; however, right ventricular power output did not change significantly due to simultaneous changes in right ventricular preload and afterload.
Other experimental studies have shown similar changes in right ventricular afterload, compliance, and contractility. In normal dogs, Farrar and associates [21] observed significant reductions in right ventricular peak systolic pressure and mean pulmonary arterial pressure indicating a reduced afterload. No significant changes in cardiac output or right ventricular stroke volume occurred during left ventricular bypass. However, end-diastolic volume, stroke volume, and ejection fraction were unchanged in response to the reduced afterload. Further, the maximum and minimum rates of change in right ventricular pressure and the maximum rate of change in pulmonary arterial flow were reduced, suggesting reductions in right ventricular contractility.
In dogs, Fukamachi and co-workers [22] observed that left heart bypass decreased pulmonary arterial resistance (-15%), increased right ventricular chamber compliance (38%), and decreased the slope of the right ventricular end-diastolic pressure versus cardiac output relationship. The net effect of the left heart bypass was an increase in cardiac output (20%) for any given right ventricular end-diastolic pressure. Elbeery and associates [23] showed that during full support, left ventricular end-diastolic volume decreased by 31%, left ventricular septal-free wall diameter decreased by 7%, and the rate of rise of right ventricular pressure declined by 13%. Analysis of right ventricular end-diastolic pressure-volume relationships suggested improved right ventricular chamber compliance. In dogs, Miyamoto and colleagues [24] observed that incremental increases in the level of left ventricular unloading resulted in incremental decreases in the rate of right ventricular pressure change.
Therefore, during LVAD support, global right ventricular contractility is impaired with leftward septal shifting, but right ventricular myocardial efficiency and power output are maintained through a decrease in right ventricular afterload and an increase in right ventricular preload. However, not all the studies have observed a decrease in right ventricular contractility. In pigs, Chow and associates [25] calculated right ventricular maximal systolic elastance as the slope between an isovolumic beat and a series of transiently occluded beats. During LVAD support, there were no significant changes in cardiac output, mean systemic arterial pressure, or in maximum rate of change of right ventricular pressure. Most important, right ventricular maximal systolic elastance during LVAD unloading was unchanged from control.
With a preexisting pathologic condition, such as regional myocardial ischemia, the right ventricular response is qualitatively the same: a decrease in right ventricular afterload, increased compliance, and decreased contractility. However, because of the preexisting condition, anatomic ventricular interaction is accentuated, leading to a greater decrease in right ventricular contractility. The net effect is either no change or a decrease in cardiac output. For example, in pigs, LVAD support resulted in a leftward septal and a 13% increase in right ventricular septal-to-free wall dimension, with no changes in right ventricular cardiac output or stroke work [26]. In contrast, right coronary artery occlusion alone produced right heart failure with a 50% reduction in right ventricular global stroke work and 26% and 27% reductions in cardiac output and right ventricular peak systolic pressure, respectively. During LVAD support, right heart failure persisted, which resulted in further leftward septal shifting and unchanged but still depressed stroke work and flow output.
Chow and Farrar [27] produced congestive heart failure in pigs by rapid ventricular pacing. With congestive heart failure, left ventricular unloading by assist device led to a further impairment in cardiac output (-14%), mean arterial pressure (-15%), and the slope of the right ventricular global stroke-work curve (-50%). Based on analysis of the right ventricular pressure and free wall-to-septum dimension data, left ventricular pressure unloading resulted in a 48% reduction in the slope and 20% increase in the dimension intercept of the preload recruitable stroke-work relationship. There was also a 45% reduction in the slope and 16% increase in the dimension intercept of the end-systolic pressure-dimension relation. These slope changes plus reductions in cardiac output and global stroke work, which are indicative of impaired right ventricular function during left ventricular pressure unloading in the congestive heart failure pigs, were not seen in normal hearts.
In calves, when ventricular function was normal, establishment of left heart bypass did not significantly affect right ventricular function [28]. Septal ischemia during support caused a rapid and remarkable increased in right ventricular short-axis dimension in every experiment. Right ventricular end-diastolic area increased during septal ischemia by 275% and right ventricular fractional area change decreased from 82% to 28%. Right ventricular developed pressure decreased and a further decrease in intraventricular septal wall thickening occurred. The interventricular septum became thin with flattening of its normal contour. Septal reperfusion resulted in right ventricular recovery with significant improvement in all factors.
Nishigaki and associates [29] compared a control group with groups with ischemic injuries to the right ventricular free wall and interventricular septum. With LVAD support, septal ischemia shifted the right ventricular pressure-volume points to the right, indicating significant depression of function.
The response in human hearts is similar to the experimental studies: decreased right ventricular afterload, increased right ventricular filling, and an increase in cardiac output without regional ischemia. Farrar and co-workers [30] simulated an LVAD in the operating room by bypassing and unloading the left ventricle with the heart-lung machine before a routine open heart operation. During bypass, in patients with normal left ventricular function, mean pulmonary arterial pressure decreased from 17 to 10 mm Hg, which resulted in a decrease in right ventricular end-diastolic pressure (from 8 to 6 mm Hg) with no change in cardiac output. However, in patients with poor left ventricular function, during bypass, pulmonary arterial pressure decreased from 27 to 12 mm Hg and right ventricular end-diastolic pressure decreased from 13 to 8 mm Hg. This caused an increase in cardiac output during left ventricular bypass from 4.5 to 5.3 L/min.
Mechanism for Right Ventricular Failure During Left Ventricular Assist Device Support
The series circulatory connections and ventricular interdependence help to explain the observed right ventricular responses to LVAD support. Via the circulatory connections, decreasing left ventricular preload will decrease pulmonary artery pressure. This passive decrease in pulmonary pressures is caused by the reductions in left ventricular filling pressure during left ventricular bypass. With higher initial left ventricular end-diastolic pressures, greater reductions in left ventricular filling pressures will occur. This will result in greater decreases in pulmonary artery pressure.
Right ventricular filling is increased by the increased left ventricular output with LVAD support (the series circulatory component). Right ventricular filling is further assisted by the decrease in left ventricular diastolic volume (the ventricular interdependence component). The decrease in left ventricular diastolic volume alters right ventricular compliance, making it easier to fill the right ventricle. The observed changes in right ventricular shape and septal position and the observed increases in right ventricular volume with right ventricular filling pressure decreases reflect this altered compliance.
Without ventricular interdependence, the reduced pulmonary artery pressure and increased right ventricular filling would increase right ventricular stroke volume. Kinoshita and colleagues [31] examined this question in calves with artificial hearts. The total artificial heart eliminated ventricular interdependence because the ventricle did not have anatomic or mechanical interactions. Biventricular failure was simulated by reducing both ventricular drive pressures of the total artificial heart. In the simulated condition, LVAD support decreased pulmonary artery pressure and increased cardiac output.
These positive effects have to compensate for the reduced left ventricular systolic assistance, which leads to a decrease in right ventricular systolic function. Via ventricular interdependence, the decrease in left ventricular systolic function decreases left ventricular assistance to right ventricular function. This sets up a positive feedback mechanism: the decreased left ventricular assistance decreases right ventricular systolic pressure and stroke volume, which decrease left ventricular filling. Again, this depressed right ventricular function via ventricular interdependence is reflected in the observed changes in right ventricular septal position that persist throughout systole and the decreases in right ventricular elastance and maximal rate of change of right ventricular pressure.
As mentioned before, ventricular interdependence is best viewed as the balance of forces at the intraventricular sulcus. What is not obvious is that the stress in the left ventricular wall is not zero, even when left ventricular cavity pressure is zero: even with no left ventricular cavity pressure, stress is still present in left ventricular walls. For example, Kresh and associates [32] measured left ventricular intramyocardial pressure. They demonstrated that despite total left ventricular decompression tension-time index less than zero the systolic pressure-time integral estimated using intramyocardial pressure was only slightly reduced from 2,340 in a beating ejecting heart to 2,080 in a beating empty left ventricular bypass. Further, the systolic pressure-time integral could be increased by epinephrine infusion despite the absence of left ventricular tension-time changes. In a nonworking cat heart preparation, Mihailescu and Abel [33] measured significant intramyocardial pressure although left ventricular cavity pressure was zero. Although the heart is not performing pressure or volume work, it continues to generate intramyocardial stresses.
Figure 5 is a cartoon that summarizes the possible mechanisms for right ventricular failure with LVAD support. The right ventricle ejects blood through a uniform reduction in its free wall area and a decrease in the septal-to-free wall distance [34]. With its large surface area-to-volume ratio, small decreases in septal-to-free wall distance cause large volume displacements. This bellows effect was first proposed by Rushmer and associates [35], who considered the right ventricular free wall and septum to be the hands holding the bellows. Based on ventricular interdependence, we propose that one hand is the right ventricular free wall, whereas the other hand is the whole left ventricle.
|
), both contribute to right ventricular ejection. With LVAD support (see Fig 5B), right ventricular afterload is decreased. The right ventricular free wall moves more and contributes more (decreased afterload), while the left ventricle contributes less (due to the decrease in ventricular interdependence). With septal ischemia, the left ventricle moves away from the right ventricle, a negative contribution. The right ventricular free wall cannot compensate, and right ventricular failure occurs with a decrease in cardiac output. We would speculate that this movement of the septum away from the right ventricular cavity and toward the left ventricle probably occurs with right ventricular failure. During systole, the change in the interventricular septum from its normal shape to a flattened or inverted shape would result in a net motion of the septum away from the right ventricular cavity. Again, given the bellows-like contraction of the right ventricle, this septal motion will greatly decrease right ventricular output, and may explain the observed right ventricular failure. What separates mild from severe right ventricular depression with LVAD support? First, underlying right ventricular dysfunction appears necessary. Second, the degree of right ventricular afterload reduction is important. Third, regional ischemia is probably worse than global dysfunction, especially of the septum. Kawai and associates [36] evaluated the differential performance of the right ventricular free wall and septum to identify the need for additional right ventricular support. Cross-sectional views of the right ventricle were obtained immediately before and 1 hour after Novacor LVAD (Novacor Division, Baxter Healthcare Corp, Irvine, CA) implantation in 12 consecutive patients using transesophageal echocardiography. Percentage change between preimplantation and postimplantation right ventricular ejection fraction, fractional area change for the right ventricular free wall and the septum, and the fractional shortening of the free wall-to-septum distance were reported for patients based on the need for right ventricular support: minimal or maximal. With LVAD support, patients showed a reduction in septal fractional area change. However, this reduction was most pronounced in patients requiring maximal right ventricular support. The degree of septal impairment may explain the mechanism of right ventricular failure in patients on an LVAD.
Acknowledgments
This study was supported in part by a grant from the Rudd Heart and Lung Institute of Jewish Hospital.
Footnotes
Presented at The Third International Conference on Circulatory Support Devices for Severe Cardiac Failure, Pittsburgh, PA, Oct 28-30, 1994.
Address reprint requests to Dr Santamore, Division of Thoracic and Cardiovascular Surgery, University of Louisville, Louisville, KY 40292.
References
This article has been cited by other articles:
|
|
 |
|
  S. Aggarwal, F. Cheema, M. C. Oz, and Y. Naka Long-Term Mechanical Circulatory Support Card. Surg. Adult, January 1, 2008; 3(2008): 1609 - 1628. [Full Text]
|
|
|
|
 |
|
 M. E. Stone Current Status of Mechanical Circulatory Assistance Seminars in Cardiothoracic and Vascular Anesthesia, September 1, 2007; 11(3): 185 - 204. [Abstract] [PDF]
|
|
|
|
 |
|
 S. Chumnanvej, M. J. Wood, T. E. MacGillivray, and M. F. V. Melo Perioperative Echocardiographic Examination for Ventricular Assist Device Implantation Anesth. Analg., September 1, 2007; 105(3): 583 - 601. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Santambrogio, T. Bianchi, M. Fuardo, F. Gazzoli, R. Veronesi, A. Braschi, and M. Maurelli Right ventricular failure after left ventricular assist device insertion: preoperative risk factors Interactive CardioVascular and Thoracic Surgery, August 1, 2006; 5(4): 379 - 382. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O.H. Frazier, T. J. Myers, and I. Gregoric Biventricular assistance with the Jarvik FlowMaker: A case report J. Thorac. Cardiovasc. Surg., October 1, 2004; 128(4): 625 - 626. [Full Text] [PDF]
|
|
|
|
 |
|
 D. Nordhaug, T. Steensrud, S. Muller, K. V. Husnes, and T. Myrmel Intraaortic Balloon Pumping Improves Hemodynamics and Right Ventricular Efficiency in Acute Ischemic Right Ventricular Failure Ann. Thorac. Surg., October 1, 2004; 78(4): 1426 - 1432. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. D. Aaronson, H. Patel, and F. D. Pagani Patient selection for left ventricular assist device therapy Ann. Thorac. Surg., June 1, 2003; 75(90060): S29 - 35. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Stone, W. Soong, M. Krol, and D. L. Reich The Anesthetic Considerations in Patients with Ventricular Assist Devices Presenting for Noncardiac Surgery: A Review of Eight Cases Anesth. Analg., July 1, 2002; 95(1): 42 - 49. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Y. Schure, P. C. Laussen, and F. X. McGowan JR Mechanical Cardiopulmonary Support of Infants and Children With Congenital Heart Disease Seminars in Cardiothoracic and Vascular Anesthesia, March 1, 2001; 5(1): 46 - 54. [Abstract] [PDF]
|
|
|
|
|
|
E. Chinchoy, C. L. Soule, A. J. Houlton, W. J. Gallagher, M. A. Hjelle, T. G. Laske, J. Morissette, and P. A. Iaizzo Isolated four-chamber working swine heart model Ann. Thorac. Surg., November 1, 2000; 70(5): 1607 - 1614. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. Trumble, C. S. Park, and J. A. Magovern Copulsation Balloon for Right Ventricular Assistance : Preliminary Trials Circulation, June 1, 1999; 99(21): 2815 - 2818. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
