American Association for

Thoracic Surgery

63RD ANNUAL MEETING

Scientific Program

MONDAY MORNING, April 25, 1983

8:30 a.m. Business Session (Limited to Members)

Grand Ballroom

8:45 a.m. Forum Session - Grand Ballroom

1. Is a Left Ventricular Vent Necessary for Coronary Artery Bypass Procedures Performed with Cardioplegic Arrest?

ROBERT H. BREYER*, J. WAYNE MEREDITH*,

STEPHEN A. MILLS*, MICHAEL BARRINGER*,

ZAK K. SHIHABI*, ALBERTO TRILLO* and

A. ROBERT CORDELL

Winston Salem, North Carolina

Before the widespread adoption of hypothermic cardioplegic arrest (HCA), the physiologic benefits of left ventricular venting (LVV) were convincingly shown. Now, more than 90% of cardiac surgeons utilize HCA during the performance of coronary artery bypass graft procedures (CABG) and many no longer use a vent. Equivalent myocardial cooling and ventricular performance, as assessed by radionuclide angiography, were recently reported in a series of vented (V) and non-vented (NV) patients undergoing CABG. However, animal studies by other investigators have demonstrated depressed left ventricular function (LVF) when left ventricular end diastolic pressure (LVEDP) was elevated to 20 mmHg during reperfusion, even though ventricular distension during the arrest period per se was not harmful. In clinical practice, LVEDP rarely exceeds 10 mmHg during reperfusion and radionuclide angiography may not be sufficiently sensitive to detect subtle deficiencies in myocardial protection. Consequently, the role of LVV in CABG remains unclear.

We evaluated the need for LVV using a canine model (35 dogs) that closely simulated the conditions during routine CABG. The dogs were placed on car-diopulmonary bypass, subjected to 60 min. of HCA followed by 30 min. of reperfusion. The left ventricle was vented in 18 of the 35 dogs, not vented in 17 dogs. Myocardial temperature (Temp.) and left atrial pressure (LAP) were recorded continuously. Before (pre) and 30 min. after (post) HCA: (1) LVF curves (6 V, 6 NV dogs) were generated at constant arterial pressure (MAP) and the pre and post values for left ventricular stroke work index (LVSWI) fitted to second order polynomials of LAP. (2) Maximum dP/dt was determined at constant LAP and MAP. (3) The left ventricle was biopsied for determinations of ATP (12V, 11 NV dogs) and semiquantitative grading of ultrastructure (6V, 6 NV).

LAP in NV dogs was 6.9 ± 3.9 mmHg during HCA and 5.0 ± 3.0 mmHg during reperfusion. Temp, during HCA was 10.0 ± 4.6 C in V dogs and 8.8 ± 2.2 C in NV dogs (p = 0.5). Comparison of fractional change in LVSWI (LAP = 8) demonstrated no differences in LVF curves between V and NV dogs (p = 0.47) and the V and NV dogs could not be separated by linear discriminant analysis (p= .40). Equivalent recovery of maximum dP/dt was seen in both groups, 86.2% in V dogs and 97.4% in NV dogs (p>0.1). Post ATP was 96.7% of control (4.30/µmol/gm) in V dogs and 94.6% (4.37 µmol/gm) in NV dogs (p = 0.7). Mitochondrial ultrastructure was equally preserved in both groups. The absence of LVV did not lead to left ventricular distension or to more rapid rewarming. The equivalent myocardial protection (preservation of LVF, ATP and ultrastructure) seen in all dogs is contradictory to recent reports that suggest LVV is necessary during reperfusion after HCA.

*By invitation

2. Autogenous Pericardial Patch Enlargement of the Aortic Root or Annulus During Aortic Valve Replacement: Long-Term Follow-Up

JEFFREY M. PIEHLER*, GORDON K. DANIELSON,

JAMES R. PLUTH, THOMAS A. ORSZULAK*,

FRANCISCO J. PUGA*, HARTZELL V. SCHAFF*,

WILLIAM D. EDWARDS* and CLARENCE SHUB*

Rochester, Minnesota

Patch enlargement of the aortic root or annulus is a widely accepted technique when restrictive anatomy is encountered during aortic valve replacement. Patches made of prosthetic material have been almost exclusively utilized, and patches of autogenous pericardium have not received wide acceptance. Although pericardium is advantageous because of its low cost, ready availability, non-porosity, and ease of handling, its long-term durability has not been fully established. From 1965-1981, 96 patients had autogenous pericardial patch enlargement of the aorta during aortic valvular replacement. There were 44 males and 52 females with a mean age of 46.5 years. In 81 patients, the patch was placed solely in a supravalvular location to facilitate aortic closure around the valvular prosthesis, while in 15 patients the patch was placed in a transannular fashion, and thus in both subvalvular and supravalvular locations, allowing for annular expansion and insertion of a larger prosthesis. The four operative deaths were from low cardiac output and were unrelated to the use of the patch. In a mean follow-up of 5.4 years (range 6 months to 15 years), no patient has had clinical evidence of sudden patch failure and none has had patch aneurysms detected by routine chest roentgenography. One patient required reoperation for peri valvular leak where the prosthesis had been sutured to the patch. Objective data concerning the late postoperative status of the patch were available in 48 patients. Twenty-four underwent a subsequent cardiac operation where the patch was visualized; of the remaining 24, 16 underwent two-dimensional echocardiography, 2 aortic root angiography, and 6 postmortem examination. In each patient, the patches were well incorporated into the adjacent tissues and patch aneurysms were universally absent. Postmortem histologic examination of the patches revealed a thick fibro-elastic neo-intimal layer to the pericardium with a smooth transition to adjacent aortic intima. This proven durability suggests that autogenous pericardium is an excellent patch material when required during aortic valve replacement.

*By Invitation

3. Experimental Mitral Regurgitation: Physiologic Effects of Correction on Left Ventricular Dynamics

JOHN A. SPRATT*, CRAIG O. OLSEN*,

GEORGE J. TYSON*, DONALD D. GLOWER, JR.*

and J. SCOTT RANKIN*

Durham, North Carolina

Sponsored by: DAVID C. SABISTON, JR.,

Durham, North Carolina

It has been suggested that mitral valve replacement for mitral regurgitation (MR) can precipitate acute myocardial failure by increasing left ventricular (LV) afterload. To investigate this hypothesis, 9 dogs were surgically instrumented with ultrasonic transducers to measure LV diameter (D), electromagnetic flow probes to measure ascending aortic blood flor (Q), and micromanometers to measure LV and pleural pressures (P). At the time of implantation, an 8 mm stainless steel shunt was inserted through the LV myocardium at the base of the anterior wall and sutured to the left atrial (LA) appendage, producing simulated MR of 30-50% of total LV output. A balloon occluder was placed around the LA appendage. One to 7 days after implantation, each dog was studied in the conscious state, and data were recorded during acute occlusion of the shunt. Heart rate was maintained constant by atrial pacing. Data were analyzed digitally, mean ejection wall tension (T) was calculated as ejection LVP x D, and Q was integrated to compute forward cardiac output (FCO). Systolic D shortening (ΔD) was calculated as the change in D from beginning to end ejection. After shunt occlusion, mean LV ejection pressure (EP) increased* in all studies. LVT increased* by an average of 10%, ΔD decreased* by 24%, and FCO increased* by 13%. Thus, the higher afterload after elimination of MR produced an acute fall in stroke shortening and total LV output. However, FCO increased in all studies, implying improved pump efficiency and overall cardiac function. These data suggest that FCO should increase with correction of MR and that the associated augmentation in afterload is probably not a major factor causing low cardiac output after correction.

Shunt	EP	LVT	EDO	ΔD	FCO
Status	(mmHg)	(cm mmHg)	(mm)	(mm)	(ml/min)
Open	92.4 ± 19.0	3070 ± 443	35.1 ± 3.5	18.4 ± 4.7	1756 ± 741
Occluded	101. 3 ± 19.7*	3385 ± 404*	34.9 ± 3.7	13.9 ± 6.2*	1953 ± 572*
* = significant difference using Student's t test for paired data (p<0.05).

*By Invitation

4. A Comparison of Valved with Nonvalved Extra Cardiac Conduits: An Experimental Study

ANDREW C. FIORE*, PAMELA S. PEIGH*,

ROBERT J. ROBISON*, MICHAEL D. GLANT*,

HAROLD KING and JOHN W. BROWN*

Indianapolis, Indiana

Extra cardiac conduits (ECC) are essential in operations for congenital discontinuity between the right ventricle (RV) and the pulmonary artery (PA). The disturbing degree of obstruction reported in clinical series of ECC containing porcine valves has been largely attributed to stenosis at the RV anastomosis, early valve deterioration and the development of a thick neointimal lining within the Dacron graft. The purpose of this study is to compare transconduit resistance and thickness of the neointimal lining in right ventricular ECC with and without a porcine valve.

Sixteen millimeter woven Dacron conduits were implanted in 15 adult mongrel dogs followed by proximal PA occlusion with Dacron tape. In 6 dogs, the ECC contained a porcine valve, while in the other 9, it did not. Cardiac output (C), transconduit gradient (G) and resistance (G/CO) were measured at operation, 6 and 12 months postoperatively. After one year, the ECC was removed, the Dacron graft opened longitudinally and the cross-sectional thickness of the neointimal lining (excluding suture lines) was measured microscopically. Data were collated by time and group. Groups were compared statistically using a T test and two way analysis of variance.

		Operation	6 Months	12 Months
Valved	G/CO	5.30 ± 1.25	6.26 ± 1.11	7.32 ± 1.42
(N = 6)	NT			1370 ± 313.1
Nonvalved	G/CO	8.84 ± 1.78	7.90 ± 1.27	9.56 ± 1.14
(N = 9)	NT			367.7 ± 28.07
CO in liters/minute; G in mmHg; NT is neointimal thickness in microns.
Each value expressed as a mean ± SEM.

Cardiac output and resistance were not significantly different between the two groups. The thickness of the neointimal peel was nearly four-fold greater in valved conduits (P<.005). Nonvalved conduits had uniform neointimae, while conduits containing valves had fenestrated intimal linings and varying degrees of valve cusp thrombosis and calcification. The neointimal thickening seen in valved conduits was greatest distal to the valve.

This study demonstrated: (1) the presence of a porcine valve in an extracardiac conduit is associated with a thicker and fenestrated neointima than when the valve is absent, and (2) the absence of a valve in the conduit did not adversely affect the cardiac output or resistance in our model at one year. These data suggest that right ventricular extra cardiac conduits with a different valve design or without valves warrant further investigation.

*By Invitation

5. Internal Mammary Artery Versus Saphenous Vein Graft: Comparative Performance in Patients with Combined Revascularization

RAM N. SINGH*, JULIO A. SOSA* and

GEORGE E. GREEN

Pittsburgh, Pennsylvania; Albany and New York, New York

Thirty-three patients with coronary artery disease undergoing combined myocardial revascularization with internal mammary artery (IMA) and saphenous vein grafts (SVGs) underwent angiographic studies up to ten years after surgery. Each patient had one IMA graft and one or more SVGs. The studies were analyzed to assess the state of preservation of the grafts as well as patency, and findings were correlated with symptoms. Eleven asymptomatic patients, studied one month to five years (mean 1.9 years) after surgery, had intact IMA grafts and SVGs in a good state of preservation. Of the six patients developing symptoms in the first year of surgery, three had evidence of poor flow in the IMA graft due to large side branches and the other three had stenosis or occlusion of the SVGs. Sixteen patients became symptomatic after several years of symptom-free status and were studied three to ten years (mean 6 years) after surgery. In this group, one IMA graft was occluded and the remaining 15 were in excellent condition. Of the 23 SVGs in this group, 17 (74%) were either occluded or severely stenosed and only six (26%) were in good condition. SVG failure is the predominant cause of late development of symptoms in patients with combined revascularization. Our findings suggest that late SVG failure is a result of an intimal proliferative process from which the IMA grafts tend to remain free. The long term performance of the IMA grafts is far superior to the SVGs.

*By Invitation

6. Pulmonary Artery Balloon Counterpulsation for Right Ventricular Failure

LELAND G. SIWEK*, G. KIMBLE JETT*,

ANTHONY L. PICONE*, ROBERT E. APPLEBAUM*

and MICHAEL JONES*

Newton and Boston, Massachusetts; Bethesda, Maryland

Sponsored by: W. GERALD AUSTEN, Boston, Massachusetts

Right ventricular (RV) failure frequently occurs in patients undergoing correction of congenital cardiac defects, as well as in other settings. RV hypertrophy (RVH) was created in sheep by pulmonary artery banding. RV failure was then produced by performing a right ventriculotomy. Unlike normal hearts, those with RVH and moderate RV outflow obstruction all developed intractable RV failure after ventriculotomy. All six unassisted controls died. Six experimental animals received mechanical assistance by pulmonary artery balloon Counterpulsation (PABCP). A Dacron graft anastomosed end-to-side to the proximal PA served as a reservoir for a 40cc Intra-Aortic Balloon Pump. PABCP effectively reversed RV failure, low cardiac output, and systemic hypotension. Measurements were made in experimental animals during consecutive alternating periods with or without PABCP. Each animal served as its own control. The data from experimental periods were compared by the paired t-test. PABCP increased cardiac output from 1.45 ± 0.16 to 2.03 ± 0.13 L/min (p<0.0001), and increased aortic systolic pressure from 78 ± 7 to 99 ± 6 mmHg (p<0.001). PABCP produced significant reduction in RV afterload, decreasing RV systolic pressure from 56 ± 5 to 41 ± 3 mmHg (p<0.0001) and the RV systolic pressure time index from 1140 ± 79 to 710 ± 65 mmHg·sec·min^-1 (p<0.0001). While RV systolic pressure decreased, PA systolic pressure distal to the band increased from 31 ± 1.5 to 40 ± 1.1 mmHg (p<0.0001). RA pressure decreased from 13.9 ± 0.7 to 11.4 ± 0.6 mmHg (p<0.0001) with PABCP and RVEDP similarly fell from 14.9 ± 0.6 to 10.7 ± 0.5 mmHg (p<0.0001). PABCP produces dramatic hemodynamic improvement and should prove clinically useful in managing otherwise refractory RV failure.

10:15 a.m. Intermission - Visit Exhibits - Lower Level

(Galleria) - Complimentary Coffee

*By Invitation

11:00 a.m. Scientific Session - Grand Ballroom

7. Late Functional and Hemodynamic Status of Surviving Patients Following Insertion of a Left Heart Assist Device

DANIEL M. ROSE*, STEVEN M. COLVIN*,

ALFRED TO CULLIFORD*, O. W. ISOM,

JOSEPH N. CUNNINGHAM, JR. and FRANK C. SPENCER

New York, New York

Since 1978 we have inserted a left atrial to ascending aorta left heart assist device in 35 patients with 17 acute and 13 long-term survivors. Thirty-three patients could not be successfully weaned from cardiopulmonary bypass and required insertion of the assist device. Two patients had refractory ventricular arrhythmias and profound cardiac failure in the early postoperative period and required insertion of the assist device. We employ a left heart assist device modified from the assist device initially described by Litwak in 1976. With this type of device, flow rates of 3.5 to 4.0 L/min can be attained. Once the patient's own cardiac function recovers satisfactorily, the patient can be weaned from the assist device. Surviving patients were maintained on the assist device for 16 - 92 hours. There were no significant complications related to the use of the assist device.

Seventeen patients were successfully weaned from the assist device. Four early deaths occurred 60-120 days following removal of the assist device (one from cardiac causes, and three from sepsis). Of the 13 long-term survivors (8 months - 42 months), 4 patients have mild to moderate recurrence of their angina (NYHA Classification II-III) and 9 patients are completely asymptomatic (NYHA Classification I) and 7 are working full time. Three of 13 patients have significantly decreased ejection fractions from their preoperative level, while 10 of 13 have either maintained or increased their ejection fraction from their preoperative level.

We conclude that use of a left heart assist device not only can improve acute survival in patients with profound cardiac failure but can also help preserve long-term ventricular function.

*By Invitation

8. The Implantation of the Total Artificial Heart for the Treatment of Endstage Cardiomyopathy

WILLIAM C. DeVRIES* and L. D. JOYCE*

Salt Lake City, Utah

Sponsored by: DAVID C. SABISTON,

Durham, North Carolina

On December 2, 1982, a 61 year old man with endstage cardiomyopathy (Class IV-B) was taken to the operating room during bouts of symptomatic ventricular tachycardia, at which time his heart was removed and an orthotopic pneumatic driven total artificial heart was implanted. Preoperatively, the patient's myocardium had progressively deteriorated clinically and histologically (endomyocardial biopsy) despite therapeutic trials of digoxin, furosemide, hydralazine hydrochloride, amiodarone, azathioprine, and prednisone. After extensive medical, social, psychiatric and financial screening, the patient was approved for surgery. The postoperative period was complicated by intermittent renal failure and pulmonary insufficiency, seizures, valve breakage, compensated disseminated intravascular coagulation and anticoagulation difficulties. The events leading to the surgery, the surgical procedure and the complicated postoperative management will be discussed. The effect of the device upon the formed blood elements, as well as the physiological and pathological compensations in this patient will also be presented.

11:30 a.m. Presidential Address - Grand Ballroom

INTELLECTUAL CREATIVITY IN THORACIC

SURGEONS

FRANK C. SPENCER

12:15 p.m. Adjourn for Lunch - Visit Exhibits

*By Invitation