Circulatory support for right ventricular dysfunction

J THoRAc

CARDIOVASC SURG

1987;94:95-103

Circulatory support for right ventricular dysfunction New modes of circulatory support for right ventricular dysfunction have recently been described. The present study compared the effectiveness of pulmonary artery balloon counterpulsation with a right ventricularassist device for support of surgicaUy induced right ventricular dysfunction. Right ventricular hypertrophy was created in 16 neonatal lambs by pulmonary artery banding. Right ventricular dysfunction was produced in aU animals by performing a right ventriculotomy and maintaining the pulmonary artery band. Four unassisted animals developed severe acute right heart failure and died. Six sheep had pulmonary artery balloon counterpulsation with a Dacron graft anastomosed to the proximal pulmonary artery as a reservoir for a 40 m1 intra-aortic balloon after the onset of heart failure. The remaining six sheep had a pneumaticaUy activated ventricular assist device inserted between the proximal pulmonary artery and the right ventricular apex. Periods of circulatory support with the balloon pump and the assist device on and off were compared. Decreases in right atrial pressure were observed with both balloon counterpulsation and right ventricular assistance: 14 ± 1 to 11 ± 1 mm Hg, p < 0.0001, versus 19 ± 2 to 12 ± 2 mm Hg, p < 0.0002, respectively. Cardiac output increased with both balloon COWlterpulsatio.n and ventricular assistance: 1.45 ± 0.16 to 2.03 ± 0.13 Ljmin, p < 0.0001, versus 0.72 ± 0.15 to 2.24 ± 0.23 Ljmin, p < 0.0002, respectively. Aortic systolic pressure increased in both support groups: 78 ± 7 to 99 ± 6 mm Hg, p < 0.0004, versus 53 ± 9 to 85 ± 9 mm Hg, p < 0.0001, respectively. Ventricular assistance produced greater changes in the right atrial pressure (39% ± 6% versus 17% ± 3%, P < 0.01), cardiac output (153% ± 39% versus 54% ± 11 %, p < 0.05), and aortic systolic pressure (85% ± 13% versus 39% ± 9%, p < 0.01). The insertion of a right ventricular assist device caused a significant increment in right ventricular dysfunction. These data, obtained with the devices in place but not operating, showed significantly increased right atrial and right ventricular end-diastolic pressures and approximately 50 % less cardiac output than with the pulmonary artery balloon counterpulsation system. The results demonstrate that both modes of circulatory support were etrective in reversing surgically induced right ventricular failure. Right ventricular assistance resulted in greater amelioration of right ventricular dysfunction than did pulmonary artery balloon counterpulsation, but was significantly detrimental to native right ventricular function.

G. Kimble Jett, M.D.,* Anthony L. Picone, M.D., and Richard E. Clark, M.D., Bethesda, Md.

Right ventricular (RV) dysfunction after intracardi-

ac repair of congenital heart defects is common.':' The

inability of the RV to support the circulation results from the ventriculotomy, with or without pulmonary valvulotomy, residual RV outflow tract obstruction, or from postoperative pulmonary hypertension. Medical management for RV dysfunction includes inotropic

From the Surgery Branch, National Heart, Lung, and Blood Institute, Bethesda, Md. Received for publication June 9, 1986. Accepted for publication July 16, 1986. Address for reprints: Richard E. Clark, M.D., Building 10, Room 2N244, National Institutes of Health, Bethesda, Md. 20892. *Current address: Department of Cardiothoracic Surgery, Emory University Hospital, Atlanta, Ga. 30322.

support and use of drugs to decrease pulmonary vascular resistance.v' Although most patients can be managed pharmacologically, occasionally mechanical circulatory support is needed. Although mechanical support of ischemic left ventricular dysfunction has been extensively investigated.r" only a few recent reports have described support of RV dysfunction.v':" Little attention, however, has been devoted to mechanical circulatory support in pediatric patients, in whom RV dysfunction is common and associated with increased impedance. Mechanical circulatory support devices have been shown to be capable of supporting the circulation after cardiac operations.>" Balloon counterpulsation is moderately effective in supporting the systemic circulation,":" although it cannot provide total ventricular support." 29 95

96

The Journal of Thoracic and Cardiovascular Surgery

Jett, Picone, Clark

A ___

8

~t-stenotic

dilatation

Graft

c------Post-stenotic dilatation

30 em in length

To pneumatic control

Fig. 1. The operative procedures to produce right ventricular failure and to provide circulatory support. Banding of the main pulmonary artery trunk (PT) was performed in newborn lambs (A). After they reached adulthood, a 6 to 8 em right ventriculotomy was made and thereafter sutured closed (B,C). A pulmonary artery balloon (PABCP) (B) or a right ventricular assist device (RVAD) (C) was then inserted to provide circulatory support after the development of right ventricular dysfunction. Ao, Aorta. RA, Right atrium. RV, Right ventricle. LV, Left ventricle.

The purpose of the present study was to compare the effectiveness of an RV assist device (RVAD) with a pulmonary artery (PA) balloon in a model of surgically induced severe R V failure that simulated the clinical setting of the pediatric cardiac surgical patient.

Materials and methods Sixteen lambs, 2 weeks of age and weighing 7 to 9 kg, underwent nonconstrictive banding of the main PA trunk with \4 inch wide Teflon tape (Fig. 1, A). The procedure was performed through a short left thoracotomy in the fourth intercostal space. The lambs were then maintained by routine husbandry techniques. All animals received humane care in compliance with the "Principles of Laboratory Animal Care" set forth by the National Society for Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH Publication No. 80-23 revised, 1978). As the animals grew, the band produced a gradually developing constriction of the PA.

When a mature weight of 50 to 60 kg was achieved after 12 months, the sheep were anesthetized with intravenous thiopental (30 mg/kg), and a cuffed endotracheal tube was inserted. Ventilation was provided with a volume respirator and anesthesia was maintained with alpha chloralose (60 rug/kg). The animal was positioned supine and the heart exposed through an anterior bilateral transverse thoracotomy in the fourth intercostal space. Aortic and left atrial pressures were measured through short, fluid-filled catheters inserted into the carotid artery and a pulmonary vein, respectively. RV pressure was measured with a similar catheter inserted in the RV apex. Right atrial and PA pressures were measured with a flow-directed catheter inserted through the jugular vein and positioned in the PA distal to the band. Continuous, instantaneous systemic blood flow (cardiac output minus coronary blood flow) was measured with a Statham S 17515 electromagnetic flow transducer (Statham Instruments, Inc., Oxnard, Calif.) placed around the ascending aorta just above the coronary arteries. The flow transducer was connectedto a Statham P2202 flowmeter. All pressure and blood

Volume 94 Number 1 July 1987

flow recordings were made on a Gould Brush 480 multichannel recorder (Gould Inc., Cleveland, Ohio). Baseline hemodynamic measurements were recorded when steady-state conditions were obtained before further interventions. Heparin (300 units/kg) was administered and cannulas were inserted in the femoral artery for systemic arterial perfusion and the superior and inferior venae cavae for venous drainage. Tapes were placed and secured around the venae cavae. Cardiopulmonary bypass was initiated, and the animal was cooled to a body temperature of 30° C. A vertical incision was made in the RV outflow tract originating just below the PA anulus and extending 6 to 8 em in length. Care was taken not to damage the anterior papillary muscle or the chordae tendineae of the tricuspid valve. The incision was then closed with a continuous braided suture buttressed with,Teflon felt strips along each side of the incision (Fig. 1, B and C). In six animals, a 20 mm diameter low-porosity woven Dacron tubular graft was anastomosed with a continuous monofilament suture to the PA proximal to the previously placed band (Fig. 1, B). The vascular graft functioned as a pumping reservoir for a 40 ml intraaortic balloon.The balloon was pneumatically inflated and deflated by a Datascope 3510 intra-aortic balloon pump unit (Datascope Corp., Paramus, N. J.). The animals were then warmed to 40° C and separated from cardiopulmonary bypass. After the development of hemodynamic evidence of RV dysfunction, the PA balloon counterpulsation (PABCP) system pump was activated. Counterpulsation timing was controlled by the PABCP unit by synchronization with the animal's electrocardiogram so that the balloon inflated during RV diastole and deflated during RV systole. Pressures and cardiac outputs were then measured at intervals with the PABCP system on and off. A ventricular assist device, designed and manufactured by ABIOMED (Applied Biomedical Corporation, Danvers, Mass.) was inserted in six animals in an end-to-side manner proximal to the previously placed band betweenthe main PA and the RV apex (Fig. 1, C). Acontinuous monofilament suture was used for the PA anastomosis, and interrupted, pledget-supported sutures were used to secure the stent of the RVAD to the RV apex. The valved blood pump was made from a continuous elastomer (Avcomat 610).30 Its stroke volume was 50 ml, and it incorporated trileaflet, polyurethane valves integral to the inflow and outflow conduits. 31 The blood pump was pneumatically driven by a Datascope 3510 intra-aortic balloon pump unit. The unit provided a pressure-limited assist with positive drive pressure continuously variable from 0 to 100 mm Hg

Circulatory support for RV dysfunction

97

and vacuum variable from 1 to 10 mm Hg. Assistance was controlled by a built-in minicomputer that provided an asynchronous fixed rate at 80 beats/min or a· synchronous electrocardiogram-triggered assist with pump ejection and filling occurring at operator-set percentages of the R-R intervals. These animals were then warmed and separated from cardiopulmonary bypass. As in the balloon-supported animals, the RVAD was activated after the development of hemodynamic evidence of R V dysfunction. The device filled during RV systole and emptied during RV diastole. Pressures and cardiac outputs were then measured with the device on and off at frequent intervals, as in the PABCPsupported animals. Four animals with a banded PA underwent right ventriculotomy only. Hemodynamic determinations were made at similar times as in the experimental groups. These animals demonstrated the effect of a vertical ventriculotomy and the course of RV dysfunction without mechanical circulatory support in sheep with high RV impedance. Systolic pressure-time index (in mm Hg sec/min) of the RV was calculated by multiplying the planimetric area under the systolic portion of the R V pressure contour during ejection by the systolic ejection period. RV stroke work index (in gm-rn/kg/beat) was calculated as (RVSM - RVEDP) X SVI X 0.0136, where RVSM is RV mean systolic pressure, RVEDP is RV end-diastolic pressure, and SVI is stroke volume index. Comparisons of the hemodynamic values obtained with the two devices in the on or off state were made. The percent changes in hemodynamic values with each device on and off were calculated and compared to each other. All comparisons were made by the Student's t test and considered statistically significant when p < 0.05. At the conclusion of each study, the heart was excised and separated into regions consisting of septum, R V and left ventricle, each of which was weighed.

Results Effect of neonatal PA banding. All 16 lambs survived PA banding. There was no clinical evidence of congestive heart failure, and there was no impairment of growth, edema, or ascites. Neonatal PA banding produced moderate RV hypertension and RV hypertrophy by adulthood. The RV/body weight ratio was 2.31 ± 0.31 gm/kg, whereas the same ratio for 15 normal sheep was 1.24 ± 0.08 gm/kg. The RV peak systolic pressure was elevated (65 ± 9 mm Hg) and a moderate peak systolic gradient was present across the PA band (38 ± 9 mm Hg). The physical characteristics of the animals are given in Table I. Right ventriculotomy without circulatory assis-

The Journal of

98

Jett, Picone, Clark

Thoracic and Cardiovascular Surge/Y

Table I. Physical characteristics PABCP (n = 6)

Control (n = 4) Age (rno) Body weight (kg) Heart weight (gm) RV weight (gm) RV thickness (mm) RV weight/body weight (gm/kg) LV weight (gm) LV thickness (mm) LV weight/body weight (gm/kg) Septal weight (gm) Septal thickness (mm) Septal weight/body weight (gm/kg)

10.1 ± 50 ± 400 ± 128 ± 11 ± 2.27 ± 88 ± 15 ± 1.50 ± 71 ± 13 ± 1.43 ±

0.4 5 30 18 2 0.30 5 2 0.10 4

12.8 ± 54 ± 431 ± 137 ± 15 ± 2.54 ± 93 ± 15 ± 1.72 ± 73 ± 16 ± 1.38 ±

2 0.15

RVAD (n = 6)

0.5 5 50 7 2 0.30

12.0 ± 0.4 58 ± 4 395 ± 28 121 ± 7 14 ± I 2.15 ± 0.21 90 ± 5 14 ± 1 1.52 ± 0.05 76 ± 4 15 ± I 1.35±0.13

5 2 0.20

4 2 0.15

Legend: PABCP. Pulmonary artery balloon counterpulsation. RVAD, Right ventricular assist device. RV, Right ventricular. LV, Left ventricular. Values are mean ± standard error of the mean.

Table Il, Hemodynamics with and without circulatory support Circulatory support off Parameter RAP (mm Hg) RVEDP (mm Hg) RVSP (mm Hg) Distal PAP (mm Hg) LAP (mm Hg) AoSP (mm Hg) AoDP (mm Hg) CO (Lj'min) CI (Lyrnin/kg) RVSPTI (mm Hg sec/min) RVSWI (grn-m/kg/beat)

PABCP 14 15 56 31 12 78 50 1.45 29.7 1140 0.081

± I

± I ± 5 ± 2 ± 1 ± 7 ± 5 ± 0.16 ± 4.5 ± 79 ± 0.011

I

RVAD 19 ± 2 19 ± 3 44 ± 8 27 ± 3 7± 1 53 ± 9 29 ± 4 0.72 ± 0.15 14.8 ± 5.2 1514 ± 232 0.042 ± 0.019

Circulatory support on p Value <0.01

NS NS NS

<0.01 <0.005 <0.01 <0.01 <0.05

NS NS

PABCP 11 ± 11 ± 41 ± 40 ± 13 ± 99 ± 59 ± 2.03 ± 41.3 ± 710 ± 0.121 ±

1 I 3

1 1

6 5 0.13 4.7 65 0.017

I

RVAD 12 12 43 52 11 85 40 2.23 30.0 969 0.075

± 2 ± I ± 6 ± 5 ± 1 ± 9 ± 5 ± 0.23 ± 5.9 ± 211 ± 0.026

p Value

NS NS NS <0.01

NS NS

<0.001

NS NS NS <0.05

Legend: Values are mean ± standard error of the mean. PABCP, Pulmonary artery balloon counterpulsation. RV AD, Right ventricular assist device. RAP, Right atrial mean pressure. RVEDP, Right ventricular end-diastolic pressure. RVSP, Right ventricular peak systolic pressure. Distal PAP, Pulmonary artery peak systolic pressure distal to the band. LAP, Left atrial mean pressure. AoSP, Aortic systolic pressure. AoDP, Aortic diastolic pressure. CO, Cardiac output. Cl, Cardiac index. RVSPTI. Right ventricular systolic pressure time index = RVEM X SEP, where RVEM = right ventricular ejection mean and SEP = systolic ejection period. RVSWI. Right ventricular stroke work index = (RVEM - RVEDP) X SVI X 0.0136, where SVI = stroke volume index. NS, not significant.

tance, The first four sheep studied underwent right ventriculotomy without mechanical assistance. RV dysfunction developed in all these animals, characterized by increasing right atrial and RVend-diastolic pressures and decreasing R V systolic pressure. This RV dysfunction was progressive and was associated with progressive circulatory collapse (hypotension and decreasing cardiac output). None of the animals survived longer than 40 minutes after separation from cardiopulmonary bypass. These data demonstrated that in the absence of effective mechanical support, this model of surgically induced R V dysfunction is rapidly fatal. The effectiveness of this model has been previously demonstrated." . Right ventriculotomy with PABCP. Six sheep underwent a right ventriculotomy with the insertion of a PABCP system. In these animals RV dysfunction developed in a manner identical to that in the control

sheep. Mechanical assistance with PABCP was initiated once systemic arterial hypotension developed, definedas an arterial systolic pressure zs 80 mm Hg. PABCP produced prompt and dramatic hemodynamic improvement. During each period with the balloon pump off, RV dysfunction and arterial hypotension recurred but were again rapidly reversed by reinstituting balloon pumping. In contrast to the unassisted sheep, those assisted with PABCP could be maintained for prolonged periods (I to 15 hours, average 7 ± 2 hours) before studies were electively terminated. The hemodynamic effects of PABCP are listed in Table II and illustrated in Figs. 2 to 5. RV afterload was decreased by counterpulsation, as evidenced by a 23% ± 4% reduction in RV peak systolic pressure (41 ± 3 versus 56 ± 5 mm Hg, p < 0.0001) and a 34% ± 7% decrease in RV systolic pressure-time index

Volume 94 Number 1

Circulatory support for RV dysfunction

July 1987

2.5

20

c

'E

0)

-lI

--

2.0

:::>

1.5

-l

E 15 E I-~

0:::

l-


I-

I::J

a..

10

:::> 0

c.?(f) -(f)

O:::UJ 0:::

a..

1.0

U


5

Off On

PABCP

(710 ± 65 versus 1,140 ± 79 nun

0.5

Off On

Off On

RVAD

Fig. 2. The effect of pulmonary artery balloon counterpulsation (PABCPj ora right ventricular assist device (RVADj on right atrial pressure is demonstrated.

p < 0.0001).

99

Hg

PABCP

Off On

RVAD

Fig. 3. Demonstrated is the effect on cardiac output of pulmonary artery balloon counterpulsation (PABCPj and a right ventricular assist device (RVADj.

sec/min,

Improved RV hemodynamics with balloon pumping was indicated by significant reductions in both right atrial pressure (11 ± 1 versus 14 ± 1 nun Hg, p < 0.0001) and RV end-diastolic pressure (11 ± 1 versus 15 ± 1 nun Hg, p < 0.0001). The PABCP provided significant augmentation of PA pressure distal to the band. The augmented PA peak systolic pressure increased by 34% ± 7% (40 ± 1 versus 31 ± 2 nun Hg, p < 0.0001) compared to the PA systolic pressure without PABCP. Improvement of RV function by PABCP produced significant improvement in systemic arterial hemodynamics. Aortic systolic pressure was increased by 35% ± 9% (99 ± 6 versus 78 ± 7 nun Hg, p < 0.0004) and aortic diastolic pressure was increased by 27% ± 12%(59 ± 5 versus 50 ± 5 nun Hg, p < 0.01). Systemic blood flow improved by 54% ± 11% with PABCP, from 1.45 ± 0.16 to 2.03 ± 0.13 L/min (p < 0.0001). Left atrial pressure was not significantly changed by PABCP. Improvement of RV function with PABCP was further demonstrated by the 66% ± 26% increase in the net RV stroke work index, which improved from 0.081 ± 0.011 without PABCP to 0.121 ± 0.017 gmrnjkgjbeat with PABCP (p < 0.01). Right ventriculotomy with RVAD. Six sheep had a right ventriculotomy with the insertion of an RVAD system. RV dysfunction developed in these sheep after the right ventriculotomy, as in the other animals. Mechanical assistance with RVAD was initiated once systemic arterial hypotension developed. RVAD pro-

W

0:

-40

W::::) (9(J)

Z~

«0:

ICL

U-J 1-«

Zo: WI-

-30 -20

U« 0: WI-

CLI

.<.:1

-10

0:

PABCP RVAD Fig. 4. Percent change in right atrial pressure caused by pulmonary artery balloon counterpulsation (PABCPj and a right ventricular assist device (RVADj.

duced prompt and dramatic hemodynamic improvement. These animals were also maintained for prolonged periods (1 to 10 hours, average 5 ± 2 hours) before the studies were electively terminated. The hemodynamic effects of R V assistance are listed in Table II and illustrated in Figs. 2 to 5. Although RV afterload in terms of RV peak systolic pressure was unchanged by the device, the RV systolic pressure-time index decreased 45% ± 14% (969 ± 211 versus 1,514 ± 232 nun Hg/sec/rnin, p < 0.01). Improved RV hemodynamics with the RVAD were

1 00

WI-

CD::>

za...

«lI::> uO

I-U

Z« W-

U Cl

a:: a:: W« a...U

The Journal of Thoracic and Cardiovascular

Jett, Picone, Clark

Surgery

m. Percent changes in hemodynamics with circulatory support

200 180 160 140 120 100 80 60 40 20

Table

Parameter RAP RVEDP RVSP Distal PAP LAP AoSP AoDP

CO CI RVSPTI RVSWI

PABCP RVAD Fig. 5. Percent change in cardiac output with pulmonary artery balloon counterpulsation (PARep) and right ventricular assistance (RVAD).

indicated by significant reductions in both right atrial pressure (12 ± 2 versus 19 ± 2 mm Hg, p < 0.01) and RV end-diastolic pressure (12 ± 1 versus 19 ± 3 mm Hg, p < 0.01). The RVAD provided significant augmentation of PA pressure distal to the band; systolic pressure increased by 114% ± 26% (52 ± 5 versus 27 ± 3 mm Hg, p < 0.01) as compared to the PA systolic pressure without RVAD. Improvement of RV function by RVAD produced significant improvement in systemic arterial hemodynamics. Aortic systolic pressure was increased by 85% ± 13% (85 ± 9 versus 53 ± 9 mm Hg, p < 0.01) and aortic diastolic pressure was increased by 46% ± 17% (40 ± 5 versus 29 ± 4 mm Hg, p < 0.01). Systemic blood flow improved by 153% ± 39% with RV AD, from 0.72 ± 0.15 to 2.23 ± 0.23 L/min (p < 0.01). Left atrial pressure was significantly increased 63% ± 15% (11 ± 1 versus 7 ± 1 mm Hg, p < 0.01). Improvement of R V function was further demonstrated by the 215% ± 87% increase in the net RV stroke work index, increasing from 0.042 ± 0.019 without R VAD to 0.075 .± 0.26 gm/kg/beat with RVAD (p < 0.01). PABCP versus RVAD. Comparison of the hemodynamic parameters preoperatively revealed no significant differences between the two groups. The hemodynamic effects produced by both forms of circulatory support were similar (Table 11). PABCP increased the aortic diastolic pressure and the R V stroke work index to a greater extent than RVAD. PA pressure distal to the band was augmented more by RVAD than by PABCP. The other hemodynamic parameters with circulatory

PABCP (%) -17 ± -27 ± -23 ± 34 ± 12 ± 35 ± 27 ± 54 ± 55 ± -34 ± 66 ±

3 3 4 7 5 9 12 II II 7

26

RVAD(%) -39 -35 -17 114

± 6 ± 7

± 9 ±

26

63 ± 85 ± 46 ± 153 ±

15

13 17 39 148 ± 53 -45 ± 14 215 ± 87

p Value

<0.01

NS NS

<0.01 <0.05 <0.01

NS <0.05 <0.05

NS <0.05

Legend: Values are mean ± standard error of the mean. PABCP, Pulmonary artery balloon counterpulsation. RV AD, Right ventricular assist device. RAP, Right atrial mean pressure. RVEDP, Right ventricular end-diastolic pressure. RVSP, Right ventricular peak systolic pressure. Distal PAP, Pulmonary artery peak systolic pressure distal to the band. LAP, Left atrial mean pressure. AoSP, Aortic systolic pressure. AoDP, Aortic diastolic pressure. CO, Cardiac output. CI, cardiac index. RVSPTI, Right ventricular systolic pressure time index = RVEM X SEP. where RVEM = right ventricular ejection mean and SEP = systolic ejection period. RVSWl, Right ventricular stroke work index = (RVEM - RVEDP) X SVI X 0.0136. where SVI = stroke volume index.

support were not significantly different when PABCP was compared to R VAD. The hemodynamics of each group with the mechanical assist devices turned off were markedly different (Table II). The RVAD-supported animals had a greater degree of R V dysfunction then the PABCP-supported animals, as evidenced by increased right atrial pressure (19 ± 1 versus 14 ± 1 mm hg, p < 0.01), but lower left atrial pressure (7 ± 1 versus 12 ± 1 mm Hg, p < 0.01), systemic arterial pressure (53/29 ± 9/4 versus 78/ 51 ± 7/5 mm Hg, p < 0.05/0.01), and systemic blood flow (0.72 ± 0.15 versus 1.45 ± 0.16 L/rnin, p < 0.01). The percent changes in hemodynamics produced by circulatory support were also significantly different when PABCP was compared to RVAD (Table III, Figs. 4 and 5). RVAD produced a greater decrease in the right atrial pressure (39% ± 6% versus 17% ± 3%, p < 0.01), provided more blood flow to the left ventricle (153% ± 39% versus 54% ± 11%, P < 0.05), and higher left atrial pressure (63% ± 15% versus 12% ± 5%, P < 0.05), PA peak systolic pressure (114% ± 26% versus 34% ± 7%, P < 0.01), and aortic systolic pressure (85% ± 13% versus 35% ± 9%, p < 0.01). RVAD also increased the R V stroke work index significantly more than did PABCP (215% ± 87% versus 66% ± 26%, P < 0.05). Discussion RV dysfunction is now frequently recognized. It is seen in adult patients with R V infarctionv" and in

Volume 94 Number 1 July 1987

those patients supported with left ventricular assist devices. 12. 15 In addition, RV dysfunction frequently follows intracardiac repair of congenital heart defects.!" Few modalities presently exist to support R V dysfunction. The primary treatment of R V dysfunction is medical management and includes inotropic support' and reduction of RV afterload.' Frequently, however, medical management is inadequate and mechanical support is needed. Mechanical assist devices have been shown to be capable of supporting the circulation during ischemic left ventricular dysfunction, both clinicallyr'" 19-21 and experimentally.v" Recent clinical reports have described mechanical support of RV dysfunction either after left ventricular dysfunction'v": 22. 24 or for postoperative low cardiac output after repair of congenital heart defects.9.35-37 Experimental evaluation and comparison of modalities for circulatory support for R V dysfunction, however, have been limited because of the absence of a satisfactory, clinically applicable model of severe RV dysfunction. We32 previously reported on an animal model of operatively induced R V dysfunction that simulates the frequently observed pediatric clinical situation. RV dysfunction was produced by a vertical ventriculotomy in a hypertrophied RV with moderate persistent RV outflow tract obstruction. All animals undergoing the ventriculotomy had progressive R V dysfunction associated with progressive circulatory collapse. This model of surgically induced RV dysfunction is more clinically relevant than one recently described by Fischer and associates" and those previously reported. 28.29,39 The RV dysfunction and progressive circulatory collapse produced by the ventriculotomy could be rapidly reversed by instituting mechanical circulatory assistance provided by the PABCP or RVAD. Overall cardiac function significantly improved with both PABCP and RVAD, as evidenced by increased cardiac output and aortic pressure and decreased R Vend-diastolic and right atrial pressures. The RV peak systolic pressure was decreased by PABCP but unchanged by RVAD. Both modalities decreased the systolic ejection period, which resulted in significantly decreased systolic pressure-time index. A decrease in systolic pressure-time index implies a reduced RV oxygen consumption with circulatory assistance, since a close correlation has been shown between systolic pressure-time index and oxygen consamption," RV stroke work index was also increased by both modalities of circulatory support, which suggests that both methods of circulatory assistance improve the decreased systolic and diastolic compliances associated with ventricular failure. The hemodynamics appear similar for both modali-

Circulatory support for RV dysfunction

I0I

ties when the absolute numbers are compared. Marked differences were seen, however, when the percent changes in hemodynamics produced by circulatory support were compared. The decrease in right atrial pressure was greater with RVAD, and RVAD was more effective at increasing blood flow to the left ventricle than PABCP. Thus RVAD was more effective than PABCP in improving R V function and providing circulatory support. Recent reports have shown that PABCP is not capable of total ventricular support. 19. 28 Spence and colleagues" found that PABCP was ineffective in providing circulatory support when ventricular fibrillation was present. Gaines and associates" compared four methods for circulatory support of profound RV failure (ventricular fibrillation) and showed that PABCP alone provided inadequate pulmonary circulatory support. RVAD was the recommended method of pulmonary circulatory support in profound RV failure. Our results also suggest that R VAD is needed for profound RV dysfunction, although either method is effective in providing circulatory support for less profound RV dysfunction. The insertion of an R VAD, however, resulted in significantly greater RV dysfunction postoperatively. Although RVAD was effective in reversing RV dysfunction, when the device was discontinued, the degree of R V dysfunction was much more marked than that with the PABCP off. The more marked degree of RV dysfunction after insertion of the assist device partially accounted for the greater changes in hemodynamics provided by RVAD as compared to PABCP. Comparison of the two groups of animals preoperatively revealed no significant differences in any of the hemodynamic parameters. This implies that the insertion of the RVAD itself caused greater R V dysfunction than did insertion of a balloon. The greater degree of RV dysfunction probably resulted from not only the longer length of the operation and cardiopulmonary bypass time, but also from the insertion of the RVAD into the RV apex. The R VAD involved not only the presence of a rigid stent in the RV apex, but also the removal of a core of R V myocardium. Although both devices were capable of effective circulatory support, the insertion of a PABCP induced less RV dysfunction than did the insertion of an RVAD. Perhaps insertion of the device from the atrium would have produced less RV dysfunction and yet have provided adequate circulatory support." 38 In summary, our results indicate that both PABCP and RVAD are effective in supporting the circulation after operatively induced RV dysfunction. RVAD produced more dramatic reversal of R V dysfunction than did PABCP. Although both devices are capable of

The Journal of

10 2

lett, Picone, Clark

Thoracic and Cardiovascular Surgery

effective circulatory support, the insertion of a PA balloon was associated with less RV dysfunction than was the insertion of an RVAD in the RV apex. REFERENCES I. Tatooles CJ, Ardekani RG, Miller RA, Seratto M. Results following physiological repair for tricuspid atresia. Ann Thorac Surg 1976;22:578-83. 2. Gale AW, Danielson GK, McGoon DC, Wallace RB, Mair DD. Fontan procedure for tricuspid atresia. Circulation 1980;62:91-6. 3. Laks H, Williams RG, Hellenbrand WE, et al. Results of atrial to right ventricular and right atrial to pulmonary artery conduits for complex congenital heart disease. Ann Surg 1980;192:382-9. 4. Lee TD, Roveti GC, Ross RS. The hemodynamic effects of isoproterenol on pulmonary hypertension in man. Am Heart J 1963;65:361-7. 5. Sade RM, Dearing JP. Augmentation of pulmonary blood flow after right ventricular bypass. J THoRAc CARDIOVASC SURG 1981;81:928-33. 6. Bernhard WF, LaFarge CG, Husain M, Yamamura N, Robinson TC. Physiologic observations during partial and total left heart bypass. J THoRAc CARDIOVASC SURG I 970;60:807-17. 7. DeBakey ME. Left ventricular bypass pump for cardiac assistance: clinical experience. Am J Cardiol 1971;27:3-

I\. 8. Litwak RS, Koffsky RM, Jurado RA, et al. Use of a left heart assist device after intracardiac surgery: technique and clinical experience. Ann Thorac Surg 1976;21: 191202. 9. Laks H, Marco JD, Farner TL, Standeven JW, Kaiser GC, Willman VL. A Servocontrolled atrial-aortic assist device: experimental findings and clinical experience. Ann Thorac Surg 1976;22:546-56. 10. Norman JC, Fuqua JM, Hibbs CW, Edmonds CH, Igo SR, Cooley DA. An intracorporeal (abdominal) left ventricular assist device: initial clinical trials. Arch Surg 1977; 112:1442-5\. II. Pierce WS, Donachy JH, Landis DI, et al. Prolonged mechanical support of the left ventricle. Circulation 1978;58(Pt 2):1133-46. 12. Turina MT, Bosio R, Senning A. Paracorporeal artificial heart in postoperative heart failure. Artif Organs 1979; 2:273-6. 13. Berger RL, Merin G, Carr J, Sussman HA, Bernhard WF. Successful use of a left ventricular assist device in cardiogenic shock from massive postoperative myocardial infarction. J THORAC CARDIOVASC SURG 1979;78:62632. 14. Olsen eK, Shaffer LJ, Pae WE, Parr GVS, Rosenberg G, Pierce WS. Biventricular mechanical assistance in the postcardiotomy patient. Trans Am Soc Artif Intern Organs 1980;26:29-33. 15. Pae WE, Rosenberg G, Donachy JH, et al. Mechanical

circulatory assistance for postoperative cardiogenic shock: a three year experience. Trans Am Soc Artif Intern Organs 1980;26:256-61. 16. Taguchi K, Murashita J, Nakagaki M, et al. Clinical application of biventricular bypass with six consecutive patients. Trans Am Soc Artif Intern Organs 1980;26:4283\. 17. Parr GVS, Pierce WS, Rosenberg G, Waldhausen lA. Right ventricular failure after repair of left ventricular aneurysm. J THORAC CARDIOVASC SURG 1980;80:79-84. 18. O'Neill MJ, Pierce WS, Wisman MD, Osbaken MD, Parr GV, Waldhausen JA. Successful management of right ventricular failure with the ventricular assist pump following aortic valve replacement and coronary bypass grafting. J THORAC CARDIOVASC SURG 1984;87:106-11. 19. Buckley MJ, Craver JM, Gold HK, Mundth ED, Daggett WM, Austen WG. Intra-aortic balloon pump assist for cardiogenic shock after cardiopulmonary bypass. Circulation 1973;47,48(Pt 2):11190-4. 20. Willerson JT, Curry GC, Watson JT, et al. Intra-aortic balloon counterpulsation on patients with cardiogenic shock, medically refractory left ventricular failure and/or recurrent ventricular tachycardia. Am J Med 1975;58:839\. 21. McEnany MT, Kay HR, Buckley MJ, et al. Clinical experience with intra-aortic balloon pump support in 728 patients. Circulation 1978;58(Pt 2):124-32. 22. Miller DC, Moreno-Cabral RJ, Stinson EB, Shinn lA, Shumway NE. Pulmonary artery balloon counterpulsation for acute right ventricular failure. J THORAC CARDI(). VASC SURG 1980;80:760-3. 23. Fledge JB Jr, Wright CB, Reisinger TJ. Successful balloon counterpulsation for right ventricular failure. Ann Thorac Surg 1984;37: 167-8. 24. Moran JM, Opravil M, Gorman AJ, Rastegar H, Meyers SN, Michaelis LL. Pulmonary artery balloon counterpul· sation for right ventricular failure. II. Clinical experience. Ann Thorac Surg 1984;38:254-9. 25. Symbas PN, McKeown PP, Santora AH, Vlasis SE. Pulmonary artery balloon counterpulsation for treatment of intraoperative right ventricular failure. Ann Thorac Surg 1985;39:437-40. 26. Kralios AC, Zwart HHJ, Moulopoulos SD, Callan R, Kwan-Gett CS, Kolff W J. Intrapulmonary artery balloon pumping assistance of the right ventricle. J THORAC CARDIOVASC SURG 1970;60:215-32. 27. Spotnitz HM, Berman MA, Reis RL, Epstein SE. The effects of synchronized counterpulsation of the pulmonary artery on right ventricular hemodynamics. J THORAC CARDIOVASC SURG 1971;61:167-74. 28. Gaines WE, Pierce WS, Prophet GA, Holtzman K. Pulmonary circulatory support: a quantitative comparison of four methods. J THoRAc CARDIOVASC SURG 1984; 88:958-64. 29. Spence PA, Weisel RD, Easdown J, Jabr KA, Salerno T A. Pulmonary artery balloon counterpulsation in the

Volume 94 Number 1 JUly 1987

30.

31.

32.

33.

34. 35.

management of right heart failure during left heart bypass. J THORAC CARDIOVASC SURG 1985;89:264-8. Singh P, Lederman D, Cumming R, et al. Development and evaluation of a seamless left heart assist blood pump. NHLBI Contract No. NOI-112-7-2937. Annual Report, March 1980 (available from NTIS, Springfield, Va.). Russell FB, Lederman DM, Singh PI, et al. Development of seamless tri-leaflet valves. Trans Am Soc Artif Intern Organs 1980;26:66-71. Jett GK, Applebaum RE, Clark RE. Right ventricular assistance for experimental right ventricular dysfunction. J THORAC CARDIOVASC SURG 1986;92:272-8. Isner JM, Roberts We. Right ventricular infarction secondary to coronary artery disease. Am J Cardiol 1978;42:885-94. Cohn IN. Right ventricular infarction revisited. Am J Cardiol 1979;43:666-8. Soeter JR, Mamiya RT, Sprague AY, McNamara 11. Prolonged extracorporeal oxygenation for cardiorespiratory failure after tetralogy correction. J THORAC CARDIO· VASC SURG 1973;66;214-8.

Circulatory support for RV dysfunction

10 3

36. Bartlett RH, Gazzaniga AB, Fong SW, Burns NE. Prolonged extracorporeal cardiopulmonary support in man. J THORAC CARDIOVASC SURG 1974;86:918-32. 37. Bartlett RH, Gazzaniga AB, Jefferies MR, Roohk HV, Haiduc N. Extracorporeal membrane oxygenator support for cardiopulmonary failure. J THORAC CARDIOVASC SURG 1977;73:375-86. 38. Fishcer EIC, Willshaw P, Armentano RL, Delbo MI, Pichel RH, Favaloro RG. Experimental acute right ventricular failure and right ventricular assist in the dog. J THORAC CARDIOVASC SURG 1985;90:580-5. 39. Opravil M, Gorman AJ, Krejcie TC, Michaelis LL, Moran JM. Pulmonary artery balloon counterpulsation for right ventricular failure. I. Experimental results. Ann Thorac Surg 1984;38:242-53. 40. Sarnoff SJ, Braunwald E, Welch GH. Hemodynamic determinants of oxygen consumption of the heart with special reference to the tension time index. Am J Physiol 1958;192:148-56.