Ann Thorac Surg 1997;63:1044-1049
© 1997 The Society of Thoracic Surgeons
Original Article: Cardiovascular
Right Ventricular Performance and Left Ventricular Assist Device Filling
William A. Mandarino, MSME,
Stephen Winowich, BSChE,
John Gorcsan, III, MD,
Thomas A. Gasior, MD,
Si M. Pham, MD,
Bartley P. Griffith, MD,
Robert L. Kormos, MD
Divisions of Cardiothoracic Surgery, Cardiology, and Anesthesiology, University of Pittsburgh and VA Medical Centers, Pittsburgh, Pennsylvania
Accepted for publication October 29, 1996.
 |
Abstract
|
|---|
Background. Right ventricular (RV) function is believed to be an important determinant of left ventricular assist device (LVAD) filling. This study was designed to demonstrate this relation in patients.
Methods. To demonstrate the interaction between RV ejection and LVAD filling, 10 patients (mean age, 49 ± 13 years) supported with an LVAD were studied. Right ventricular pressure–area loops from cross-sectional area using transesophageal echocardiographic automated border detection and high-fidelity RV pressure were recorded simultaneously with LVAD volume during intraoperative inferior vena cava occlusion. Beat-by-beat RV ejection phase indices were calculated: stroke area, peak ejection rate, and stroke work. The LVAD filling rate was calculated as the first derivative of the volume, and the peak filling rate and the mean filling rate during RV systole were determined for each cardiac cycle.
Results. Right ventricular stroke area, peak ejection rate, and stroke work were closely correlated with LVAD peak filling rate (r = 0.87 ± 0.09, r = 0.83 ± 0.09, and r = 0.85 ± 0.10, respectively). Also, baseline LVAD mean filling rate correlated with RV stroke work (r = 0.77) and LVAD peak filling rate with RV peak ejection rate for the group (r = 0.75).
Conclusions. These correlations demonstrate predictable associations of RV ejection with LVAD filling.
|
Introduction
|
|---|
The mechanical left ventricular assist device (LVAD) may be a lifesaving therapy for patients with end-stage cardiomyopathy awaiting cardiac transplantation and for patients in cardiogenic shock after a cardiac operation [1–3]. Right ventricular (RV) function is important for proper LVAD filling, although clinical assessment of RV performance has been difficult. Pressure–volume relations have been established as an important means to assess ventricular function because of their relative load insensitivity, but they have only rarely been used to assess RV function in clinical settings because on-line volume acquisition is not readily available [4–9].
We [10] have recently validated a method to assess RV performance in an animal model using cross-sectional area by echocardiographic automated border detection as a surrogate for RV volume. We have demonstrated that there is a linear relationship between RV area and volume at different left ventricular volumes and that RV pressure–area relations can be used to assess RV performance in a manner similar to pressure–volume relations. This method provides an opportunity to evaluate RV performance in patients on an LVAD. The objective of this study was to determine the relationship between RV ejection and LVAD filling at different preloads by using RV pressure–area relationships.
|
Material and Methods
|
|---|
Fifty-two patients with end-stage heart failure received a Novacor LVAD (Novacor left ventricular assist system; Baxter Healthcare Corporation, Oakland, CA) as a bridge to transplantation between July 1, 1987, and April 18, 1996, at the University of Pittsburgh Medical Center. This study was carried out on a subset of 10 of these patients (mean age, 49 ± 13 years). The study was approved by the Institutional Review Board of the University of Pittsburgh, and written informed consent was obtained from all patients. Five patients had idiopathic dilated cardiomyopathy, and 5 had ischemic cardiomyopathy.
|
Instrumentation
|
|---|
A rapid-response thermodilution pulmonary artery catheter (Edwards Scientific, Santa Ana, CA) was inserted through an internal jugular vein before induction. Patients also had instrumentation with fluid-filled femoral artery catheters connected to strain-gauge pressure transducers (Baxter Summit, Irvine, CA). After insertion of an endotracheal tube, a 64-element, single-plane 5-MHz transesophageal transducer was placed. This was connected to a prototype ultrasound system (SONOS OR model 77035A; Hewlett-Packard, Andover, MA) with automated border detection capabilities. A 4F high-fidelity fiberoptic pressure catheter (model 110-4; Camino Laboratories, San Diego, CA) was advanced into the right ventricle through a pursestring suture in the right atrium. An umbilical tape was passed around the inferior vena cava (IVC) to intermittently reduce venous return during the study protocol.
|
Automated Border Detection
|
|---|
Automated measures of RV cross-sectional area from the body of the right ventricle were acquired as an index of volume using echocardiographic automated border detection. This technique allows continuous measurement of RV cavity area on-line by differentiating between the acoustic backscatter characteristics of blood and myocardial tissue within a user-defined region of interest [11]. The threshold for discriminating blood from tissue backscatter characteristics was directly influenced by adjusting manual gain settings as previously described [12–15]. A region of interest was then manually drawn beyond the RV endocardial border so that the RV cavity area remained within this region during transient vena caval occlusion maneuvers. The short-axis RV image at the mid–papillary muscle level of the left ventricle was used, and if this image was not available because of dropout, the RV four-chamber midventricular view was used. When this view was used, time gain compensation controls were increased in the right atrial cavity adjacent to the tricuspid valve plane to maximize automated tracking of the RV blood area throughout the cardiac cycle [16].
|
Data Acquisition and Analysis
|
|---|
All studies were performed after the patient was weaned from cardiopulmonary bypass approximately 1 hour after LVAD implantation before chest closure. Simultaneous pressures, areas, and LVAD volume signals were recorded immediately before (baseline) and during transient IVC occlusion maneuvers with suspended ventilation to minimize cardiopulmonary interactions.
The analog signal of the pressure transducers, the electrocardiogram, and the volume waveform from the LVAD were continuously digitized at a sampling rate of 150 Hz by a computer workstation (Apollo Computer Inc, model A1421; Hewlett-Packard, Chelmsford, MA) with a customized hardware and software interface [12–16]. The electrocardiogram, femoral artery pressure, central venous pressure, pulmonary artery pressure, Novacor volume, RV pressure, and RV area were continuously recorded during the protocol. Right ventricular pressure and area signals were plotted to display pressure–area loops in real time. Maximal area values were aligned with the point preceding the onset of isovolumic contraction from the RV pressure for each run to allow for the variable delay in the area signal output. Physiologic waveform data acquired during the IVC occlusion protocol from 1 patient are shown in Figure 1
.
View larger version (60K):
[in this window]
[in a new window]
|
Fig 1. . Example of waveform data during inferior vena cava (IVC) occlusion from 1 patient with a left ventricular assist device (LVAD). The electrocardiogram, right ventricular (RV) pressure, RV area, first derivative of RV area (dA/dt), LVAD volume, and first derivative of LVAD volume (dV/dt) (ie, LVAD filling) are shown.
|
|
Data were separated into cardiac cycles by using the R wave of the electrocardiogram. Cycles with ectopic beats were eliminated from subsequent analysis. The following clinical variables were measured for each cardiac cycle: heart rate (HR), systolic and mean arterial pressures, mean pulmonary artery pressure, and central venous pressure. The following two measurements were obtained from the Novacor LVAD: stroke volume (SV) and pump output (SV x HR). The first derivative of the volume signal was used to calculate the flow rate of the LVAD. From this, the peak filling rates (PFRs) of the LVAD were calculated. A plot showing the change in LVAD flow rate during the IVC occlusion protocol from 1 patient is shown in Figure 2.
View larger version (44K):
[in this window]
[in a new window]
|
Fig 2. . Superimposed left ventricular assist device (LVAD) filling profiles from several cardiac cycles from 1 patient during inferior vena cava occlusion. (dV/dt = first derivative of LVAD volume, ie, LVAD filling.)
|
|
The filling profile was divided into an initial filling phase and a final filling phase. The initial filling phase was taken from the onset of native heart systole and continued until the end of LVAD filling. This filling phase occurred during RV systole. The final filling phase was taken as the filling that occurred during late RV diastole. Although the exact mechanism for this final-phase filling is unknown, it may be due to the leftward shift of the interventricular septum during RV diastole, which may result in some passive filling of the LVAD. The PFR and the mean filling rate occurring during each phase were then calculated.
|
Pressure–Area Indices
|
|---|
A plot showing the RV pressure–area loops resulting from the IVC occlusion protocol from 1 patient is shown in Figure 3. The following indices of RV performance were calculated from each individual pressure–area loop generated during the IVC occlusion protocol: end-diastolic area (EDA), end-systolic area (ESA) or minimum area, stroke area (EDA - ESA), peak ejection rate (PER = peak negative first derivative of the RV area), and stroke work (SW = 
P dA).
View larger version (34K):
[in this window]
[in a new window]
|
Fig 3. . Right ventricular (RV) pressure–area loops generated during inferior vena cava occlusion from a representative patient.
|
|
|
Statistical Analysis
|
|---|
Linear regression analysis using the method of least squares was used to correlate the PFR that occurred during RV systole with RV stroke area, PER, and SW on a beat-to-beat basis during the IVC occlusion protocol for each patient. The LVAD PFR from baseline prior to IVC occlusion was also correlated with RV PER, and LVAD mean filling rate was correlated with RV SW for all patients. Fisher's exact test was used in a post hoc fashion to evaluate the association of RV SW and LVAD mean filling rate with the clinical need of right ventricular assist device support.
|
Results
|
|---|
Technically adequate baseline data were available from all 10 patients. Inferior vena caval occlusion data were available from 8 of the 10 patients. Group mean hemodynamics and LVAD variables are shown in Table 1. Group mean baseline RV pressure–area data are shown in Table 2. The correlation coefficients for the relationship between LVAD mean filling rate and RV stroke area, PER, and SW from the 8 patients with IVC occlusion data are shown in Table 3. A linear relation existed with a mean r value of 0.87 ± 0.09 for stroke area, 0.83 ± 0.09 for PER, and 0.85 ± 0.10 for SW. An example of the regressions from 1 patient is shown in Figure 4. A plot showing the correlation between baseline LVAD PFR and RV PER and between LVAD mean filling rate and RV SW for the 10 patients is shown in Figure 5. Of the 10 patients studied, 2 required additional RV mechanical circulatory support (RV) assist device. These were patients 2 and 4 in Tables 1 and 2. There was a trend toward an association between lower RV SW and RV assist device requirement (p = 0.067), but it did not reach significance with this small patient group.
View larger version (28K):
[in this window]
[in a new window]
|
Fig 4. . Linear regression plots from 1 patient showing left ventricular assist device (LVAD) filling rate versus (A) right ventricular (RV) stroke area, (B) RV peak ejection rate, and (C) RV stroke work.
|
|
View larger version (26K):
[in this window]
[in a new window]
|
Fig 5. . Correlation between (A) baseline left ventricular assist device (LVAD) peak filling rate and right ventricular (RV) peak ejection rate and (B) LVAD mean filling rate and RV stroke work for 10 patients.
|
|
|
Comment
|
|---|
This study demonstrates that on-line measures of RV pressure–area relations using transesophageal echocardiographic automated border detection can be used to rapidly examine RV performance in patients being supported by an LVAD. Filling of the LVAD was closely coupled with RV ejection phase indices of stroke area, PER, and SW on a beat-by-beat basis. Further, LVAD mean filling rate from individual patients correlated with RV SW, and LVAD PFR correlated with RV PER. These data support the importance of RV function for proper LVAD filling in patients receiving mechanical circulatory assistance.
Although this study focuses on the effects of RV function on LVAD filling, effects of an LVAD on RV mechanical function have been examined previously [17–19]. Farrar and colleagues [17–19] proposed that the mechanisms by which the left and right ventricles interact during left ventricular mechanical support are a combination of hemodynamic and mechanical interactions. Hemodynamic interactions result from the output of the right ventricle becoming the input of the left ventricle. Mechanical interactions are due to the coupling of the two ventricles by the interventricular septum. The authors suggested that it may be alterations in these mechanisms that account for RV failure in patients supported with an LVAD. The end-systolic pressure–dimension relation of an RV free wall–septal dimension and a free wall circumferential segment length during alterations in preload in pigs was used to investigate the effects of left ventricular unloading on RV function [19]. Left ventricular unloading did not alter the slope of the free wall–septal end-systolic pressure–dimension relation but only increased the axis intercept, and there were no changes in the free-wall segment length. These studies suggest that left ventricular unloading affects primarily the septum and not the RV free wall.
These findings were similar to a study by our group [20] in which humans implanted with an LVAD showed a reduction in septal fractional area change as measured by transesophageal echocardiography. This reduction was more pronounced in a subset of these patients who required RV mechanical assistance. Our group [21] also used RV pressure–area relations to study the effects of an LVAD on RV function in humans. Right ventricular cross-sectional area was hand-digitized off-line and combined with RV pressure measured with a high-fidelity micromanometer to develop the end-systolic pressure–area relation. By evaluating ventricular and arterial correlation, it was shown that afterload reduction is probably the mechanism by which RV ejection improves after LVAD implantation. In another study of RV function after LVAD implantation, we [1] found that an increase in RV end-diastolic length measured by ultrasonic crystals placed in the RV free wall after LVAD placement occurred in a series of patients who required mechanical RV circulatory support. The present study expands the importance of RV function in patients with an LVAD by demonstrating the close correlation between RV ejection and LVAD filling.
The advance of automated border detection in the operating room setting has allowed real-time assessment of RV function in these same groups of patients. We [16, 22] recently demonstrated the potential utility of high-quality automated border detection–measured RV area to allow more sophisticated load-independent measures of RV performance, such as end-systolic elastance and preload-recruitable stroke work in patients with severe heart failure. These automated pressure–area relations have also been used to assess RV function during decreasing RV mechanical assistance to help predict the success of withdrawing mechanical support [23].
Two limitations of this study are that two-dimensional cross-sectional area was used to reflect three-dimensional volume of the anatomically complex right ventricle and that the alterations in RV geometry induced by the LVAD may affect the RV area–volume relationship. However, the linearity of the relationship of relative changes in RV area with changes in RV volume has been demonstrated in an animal model [10]. Further, alterations in left ventricular volume have been shown to produce parallel shifts in the RV area–volume relationship. Although the use of on-line cross-sectional area may represent a limitation, this method may be considered an advance from one-dimensional techniques previously used because it provides more on-line data [19]. Another limitation of the present study is that a relatively small group of LVAD patients was studied, and these observations should be extended to include a larger group. Also, the study was performed while the patient's chest was open, and differences may occur in these relationships after chest closure.
Despite these limitations, RV pressure–area relations may have potential clinical utility to assess RV function in patients with an LVAD and may be important to help demonstrate the importance of RV performance on LVAD filling. Although this study did not use the load-independent measures of end-systolic elastance and preload-recruitable stroke work, it has shown a correlation exists between LVAD filling and RV performance as demonstrated by SW. Right ventricular pressure–area relations from echocardiographic automated border detection appear to be a promising method to further evaluate the association of RV performance with LVAD filling in patients requiring mechanical assistance. In addition, they may possibly be useful to help predict the need of additional RV mechanical support after LVAD implantation, although they do not directly prevent or treat RV dysfunction in a patient with an LVAD.
|
Footnotes
|
|---|
Address reprint requests to Dr Kormos, C-700 Presbyterian University Hospital, 200 Lothrop St, Pittsburgh, PA 15213.
|
References
|
|---|
- Portner PM, Oyer PE, Pennington DG, et al. Implantable electrical left ventricular assist system: bridge to transplantation and the future. Ann Thorac Surg 1989;47:142–50.[Abstract]
- Kormos RL, Borovetz HS, Gasior T, et al. Experience with univentricular support in mortally ill cardiac transplant candidates. Ann Thorac Surg 1990;49:261–72.[Abstract]
- Pennington DG, McBride LR, Swartz MT, et al. Use of the Pierce-Donachy ventricular assist device in patients with cardiogenic shock after cardiac operations. Ann Thorac Surg 1989;47:130–5.[Abstract]
- Suga H, Sagawa K. Instantaneous pressure–volume relationships and their ratio in the excised, supported canine left ventricle. Circ Res 1974;35:117–26.[Abstract/Free Full Text]
- Redington AN, Gray HH, Hodson ME, Rigby ML, Oldershaw PJ. Characterization of the normal right ventricular pressure–volume relation by biplane angiography and simultaneous micromanometer pressure measurements. Br Heart J 1988;59:23–30.[Abstract/Free Full Text]
- Schwiep F, Cassidy SS, Ramanathan M, Johnson RL. Rapid in vivo determination of instantaneous right ventricular pressure and volume in dogs. Am J Physiol 1988;254:H622–30.[Abstract/Free Full Text]
- Maughan WL, Shoukas AA, Sagawa K, Weisfeldt ML. Instantaneous pressure–volume relationship of the canine right ventricle. Circ Res 1979;44:309–15.[Free Full Text]
- Brown KA, Ditchey RV. Human right ventricular end-systolic pressure–volume relations defined by maximal elastance. Circulation 1988;78:81–91.[Abstract/Free Full Text]
- Karunanithi MK, Michniewicz J, Copeland SE, Feneley MP. Right ventricular preload recruitable stroke work, end-systolic pressure–volume, and dp/dtmax–end-diastolic volume relations compared as indexes of right ventricular contractile performance in conscious dogs. Circ Res1992;70:1169–79.[Abstract/Free Full Text]
- Oe M, Gorcsan J, Mandarino WA, Kawai A, Griffith BP, Kormos RL. Automated echocardiographic measures of right ventricular area as an index of volume and end-systolic pressure–area relations to assess right ventricular function. Circulation 1995;92:1026–33.[Abstract/Free Full Text]
- Pérez JE, Waggoner AD, Barzilai B, Melton HE, Miller JG, Sobel BE. On-line assessment ventricular function by automatic boundary detection and ultrasonic backscatter imaging. J Am Coll Cardiol 1992;19:313–20.[Abstract]
- Gorcsan J, Romand JA, Mandarino WA, Deneault LG, Pinsky MR. Assessment of left ventricular performance by on-line pressure–area relations using echocardiographic automated border detection. J Am Coll Cardiol 1994;23:242–52.[Abstract]
- Gorcsan J, Gasior TA, Mandarino WA, Deneault LG, Hattler BG, Pinsky MR. Assessment of the immediate effects of cardiopulmonary bypass on left ventricular performance by on-line pressure–area relations. Circulation 1994;89:180–90.[Abstract/Free Full Text]
- Gorcsan J, Morita S, Mandarino WA, et al. Two-dimensional echocardiographic automated border detection accurately reflects changes in left ventricular volume. J Am Soc Echocardiogr 1993;6:482–9.[Medline]
- Gorcsan J, Gasior TA, Mandarino WA, Deneault LG, Hattler BG, Pinsky MR. On-line estimation of changes in left ventricular stroke volume by transesophageal echocardiographic automated border detection in patients undergoing coronary artery bypass grafting. Am J Cardiol 1993;72:721–7.[Medline]
- Gorcsan J, Murali S, Counihan PJ, Mandarino WA, Kormos RL. Right ventricular performance and contractile reserve in patients with severe heart failure: assessment by pressure–area relations and association with outcome. Circulation 1996;94:3190–7.[Abstract/Free Full Text]
- Farrar DJ, Compton PG, Hershon JJ, Fonger JD, Hill JD. Right heart interaction with the mechanically assisted left heart. World J Surg 1985;9:89–102.[Medline]
- Farrar DJ, Compton PG, Hershon JJ, Hill JD. Right ventricular function in an operating room model of mechanical left ventricular assistance and its effects in patients with depressed left ventricular function. Circulation 1985;72:1279–85.[Abstract/Free Full Text]
- Farrar DJ, Compton PG, Verderber A, Hill JD. Right ventricular end-systolic pressure–dimension relationship during left ventricular bypass in anesthetized pigs. ASAIO J 1986;32:278–81.
- Kawai A, Kormos RL, Mandarino WA, et al. Differential regional function of the right ventricle during the use of a left ventricular assist device. ASAIO J 1992;38:M676–8.[Medline]
- Morita S, Kormos RL, Mandarino WA, et al. Right ventricular/arterial coupling in the patient with left ventricular assistance. Circulation 1992;86(Suppl 2):316–25.
- Kormos RL, Mandarino WA, Gasior TA, Griffith BP, Gorcsan J. Assessment of the immediate effects of mechanical left ventricular assistance on right ventricular performance by automated pressure–area relations [Abstract]. J Am Coll Cardiol 1995;25:150A.
- Mandarino WA, Gorcsan J, Armitage JM, Griffith BP, Kormos RL. Assessment of the response of right ventricular performance to decreasing levels of mechanical assistance by on-line pressure–area relations. ASAIO J 1994;40:1032–5.[Medline]
This article has been cited by other articles:
|
|
|
|
 
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]
|
|
|
|
|
|
|
C. H. Van Meter Jr
Right heart failure: best treated by avoidance
Ann. Thorac. Surg.,
March 1, 2001;
71
(2007):
S220 - S222.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
