American Association for Thoracic Surgery

67TH ANNUAL MEETING

Scientific Program

MONDAY MORNING, April 6, 1987

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

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

1. New Approaches to the Management of Acute Ischemic Mitral Regurgitation

J. SCOTT RANKIN*, MICHAEL P. FENELEY*,

RICHARD D. FLOYD*, J. G. REYES*

and ROBERT M. CALIFF*

Durham, North Carolina

Sponsored by: DAVID C. SABISTON, JR.

Durham, North Carolina

Severe mitral regurgitation (MR) secondary to acute myocardial infarction has been a particularly difficult management problem with disappointing clinical results. Over a 20 month period, 15 patients were admitted with acute myocardial infarction and severe MR associated with: acute pulmonary edema in 14, low cardiac output requiring balloon pump placement in 10, and acute renal dysfunction in 7. The infarcts were posterior in 14 patients and anterior in 1; 3 had ruptured papillary muscles. Six patients underwent thrombolysis or angioplasty reperfusion within 6 hours of coronary occlusion, and the degree of MR then was monitored serially with Doppler echo-cardiography. In 5 patients (Group I), reperfusion was associated with resolution of MR to 1 + levels or less over 48-72 hours. Four of these patients underwent conventional coronary revascularization, and 1 received percutaneous angioplasty, with 4 long-term survivors (>6 months). In 10 patients (Group II), the MR persisted at severe levels, and all 10 underwent complete coronary revascularization, transventricular mitral valvuloplasty through the infarcted area, and infarct exclusion/plication. The valve repairs included posterior commissural annuloplasty in 8, papillary muscle shortening procedures in 7, reimplantation of ruptured papillary muscles in 3, and chorda! transfer in 1. Valve function was monitored intraoperatively with transesophageal echocardiography. One acute failure of repair necessitated transventricular mitral valve replacement, and 8 Group II patients have survived long-term. All survivors are Class I for angina and heart failure. Late postoperative color-flow Doppler studies of both groups revealed 0-1 + MR in 9 and 2+ MR in 3. Thus, in severe acute ischemic MR, a policy of early reperfusion when feasible, followed by 1) serial Doppler assessment of valve competence, 2) subsequent coronary revascularization, and 3) a uniform application of transventricular mitral valve repair when necessary resulted in an 80% long-term survival. Late symptomatic status and valve function appear satisfactory, and this approach offers the potential for improving therapeutic results in this difficult clinical subgroup.

*By Invitation

2. Aortic Valve Replacement with Cryopreserved Homograft Valves and with Antibiotic 4°C Stored Valves: A Comparative Follow-Up Study

MARK F. O'BRIEN*, E.G. STAFFORD*,

P.G. POHLNER*, M.A.H. GARDNER*

and D.D. McGIFFIN*

Chermsibe, Brisbane, Australia

Sponsored by: JOHN W. KIRKLIN

Birmingham, Alabama

Aortic valve replacement, with or without associated procedures, was performed in 304 patients with allograft (homograft) aortic valves; the allografts were Cryopreserved between 1975 and August 1986 (n = 184) and antibiotic sterilized and stored at 4°C between 1969 and 1975 (n = 120). All patients have follow-up, with date of inquiry of August 1986. The 10 year actuarial survival, including hospital deaths, of patients with the Cryopreserved valves and with the 4°C stored valves was 70% (± 5% SD) and 67% (± 5% SD) respectively. With the 4°C stored allografts, the 14 year survival was 51%.

Ten year actuarial freedom from homograft valve reoperation in surviving patients with Cryopreserved allografts and 4°C stored valves was 92% (± 3% SD) and 85% (± 4% SD) respectively. The 14 year actuarial freedom in the case of 4°C stored allografts was 62%. The longest follow-up in patients with Cryopreserved allografts was 12 years and in those with 4°C stored allografts was 16 years. Within these periods, no patients with Cryopreserved allografts (0% of 30 day survivors) underwent reoperation for cusp perforation while 15 patients with 4°C stored allografts (14% of 30 day survivors) underwent reoperation for cusp perforation (p for difference <.0001).

Multivariate analysis demonstrated the dominant effect of the homograft valve preservation technique on the prevalence of reoperation for cusp rupture. Earlier studies showed fibroblast viability in Cryopreserved allografts and lack of fibroblast viability in antibiotic sterilized 4 °C stored allografts. These considerations, plus the low prevalence of thromboembolism (2% of 30 day survivors) and valve endocarditis (3% of 30 day survivors) in the patients receiving Cryopreserved allograft aortic valves, indicate the superiority of this method of storage.

*By Invitation

3. The Carpentier-Edwards Supra-Annular Porcine Bioprosthesis - A New Generation Tissue Valve with Excellent Clinical Performance

W.R. ERIC JAMIESON, ALFRED N. GEREIN*,

A. IAN MUNROE*, ROBERT T. MIYAGISHIMA*,

MICHAEL T. JANUSZ* and G. FRANK O. TYERS

Vancouver, British Columbia

This investigational valve was implanted in 1176 patients (1185 operations, 1270 valves) November 1981-December 1985 (age 13-85 years, mean 61 years). Early mortality was 7.0% (with concomitant procedures 10.6%, without 4.7%); (previous operation 10.2%, without 6.4%). Late mortality was 4.5%/patient year (AYR 4.0%, MVR 4.8%, MR 0.4%). Total cumulative follow-up was 2288 years. Thromboembolism (TE) 2.6%/patient year (fatal 0.3%/patient year), (major 1.4%, minor 1.2%). Anticoagulant hemorrhage (ACH) 0.7% (fatal 0.04%). Hemolyses 0.04%. Prosthetic valve endocarditis (PVE) 0.3% (fatal 0.1%) Periprosthetic leak (PPL) 0.3% (fatal 0.04%). Structural failure (SF) (primary tissue failure/structural failure) 0.09%/patient year. Hemodynamic valve dysfunction (HVD) 1£ (0.7). Re-operation 0.6%/patient year (TE 0.09%, hemolysis 0.04%, PVE 0.04%, PPL 0.3%, SF 0.09%). TE incidence reduced after 2 years, ACH throughout observation, PVE early, PPL throughout, SF late. Survival 79.2 ± 3.3% (5 years). Freedom from (5 years) TE 91.8 ± 2.5% ACH 96.5 ± 2.4% (4 years), PVE 95.7 ± 6.1%, PPL 88.5 ± 8.2%, SF 98.8 ± 1.6%, reoperation 97.6 ± 1.8%. Freedom from all complications (5 years) 85.5 ± 3.7%, complication mortality 98.8 ± 0.8% and valve failure (mortality and reoperation) 96.7 ± 1.6%. This investigational Carpentier-Edwards, supra-annular porcine bioprosthetic valve has provided excellent clinical performance and remains our overall prosthesis of choice.

*By Invitation

4. Do Heart Valve Bioprostheses "Degenerate" Metabolically? Or Just "Wear and Tear" (Forum)

SHLOMO GABBAY*, PANKESH KADAM*,

STEPHEN FACTOR* and TAK KE CHEUNG*

Newark and Piscataway, New Jersey and Bronx, New York

Sponsored by: WILLIAM E. NEVILLE

Newark, New Jersey

The ideal substitute for diseased cardiac valves has yet to be found, dissatisfaction exist with currently available valvular prostheses, due to the still high rate of thromboembolic complications of mechanical valves, and the limited durability of bioprostheses. Failure of bioprostheses is defined as "degeneration," which is a broad term, and considered as metabolic consequences, while most of the valves just' ‘wear and tear.'' (Except the calcification in children). We have tested a large number of heart valves, porcine and pericardial bioprostheses, using a Fatigue Test System with a high degree of reproducible test conditions. The results have allowed us to define causes of "wear and tear," previously insufficiently stressed, in each type of valve tested. There is a clear difference in factors influencing tissue disruption between porcine and pericardia! valves. We have compared these in vitro results with in vivo clinical findings. The main conclusions are: 1) Bioprostheses rupture and fail in the same fashion in vitro as in vivo, 2) Mechanical and design factors are involved in tissue failure, 3) The ratio in vivo/vitro in durability is not 1:1. The ratio depends on the test conditions, 4) Pericardial valves fail by damage at time of closure, while porcine valves wear and tear in both opening and closing (mostly opening sequence) because of design features, 5) There is a wide variability in durability of similar valve types. Design improvement can minimize the variability. (Valves can fail as early as 2-3 million cycles or more than 100 million cycles). Similarly bioprostheses can be explanted in few months or last 10-13 years in patients, 6) "Domino effect" exist in tissue disruption: once one cusp fail and prolapse, the other cusps will fail in an accelerated fashion, 7) Most valves that last less than 40 to 50 x 10⁶ cycles under our test conditions, probably will not last in patients for a period of 10 years. Durability of 100 x 10⁶ cycles can be considered excellent. Very few bioprostheses reach this record, 8) Fatigue Testing is an excellent and valuable tool in developing heart valves. This study definitely shows that design improvement can prolong the durability of bioprostheses.

9:55 a.m. Intermission - Visit Exhibits - Wacker Hall

Complimentary coffee available

*By Invitation

10:40 a.m. Scientific Session - Grand Ballroom

5. Cerebral Microembolism During Cardiopulmonary Bypass: Retinal Microvascular Studies In Vivo Using Fluorescein Angiography

CHRISTOPHER I. BLAUTH*, JOHN V. ARNOLD*

EDWARD W. SCHULENBURG*,

ALISON C. MCCARTNEY*,

and KENNETH M. TAYLOR* London, England

Sponsored by: FLOYD D. LOOP

Cleveland, Ohio

We have recently reported the use of retinal fluorescein angiography to demonstrate the occurrence of occlusions in the cerebrovascular micro-circulation during cardiopulmonary bypass (CPB). In the present study, 21 patients undergoing elective coronary artery surgery had retinal fluorescein angiography 5 minutes before the end of CPB. Patients with diabetes or known cerebro vascular disease were excluded. Anaesthetic and perfusion protocols were standardised in all cases, with non-pulsatile flow and a bubble oxygenator (Harvey H1700). After 31-167 minutes of CPB, 21/21 (100%) of patients had retinal microvascular occlusions consistent with microembolism. Control fluorescein angiograms obtained in 5 patients immediately before CPB but after heparinisation and aortic cannulation showed normal retinal perfusion. During CPB there were a mean of 3·5 (range 1-7) blocked arteries of less than 50mm caliber, and a mean of 6·3 (range 1-10) focal areas of capillary non-perfusion per 30° field of retina centered on the macula per patient. The microembolic count did not correlate with bypass time (r = 0-14). Repeat studies 30 minutes after discontinuation of CPB showed partial resolution with occlusions in 4/5 (80%) of patients. In later studies at a median of 8 days (range 5-11) postoperatively, only 2/16 (12·5%) patients had persistently occluded retinal vessels, in both cases single areas of capillary non-perfusion. Psychometric testing with 4 standardised tests was performed in 18 patients. The mean total microembolic count in patients with psychometric deficit was 11*1 occlusions per field (n = 7), compared to 8·6 occlusions per field in patients with no deficit (n = 11). In a parallel study of 15 dogs, 8/9 (88·9%) had retinal microvascular occlusions during CPB after 10-90 minutes of perfusion, and retinal histology revealed platelet-fibrin microemboli 20-70 mm diameter in 7 of the 8 dogs (87·5%) with angiographic occlusions. 6 dogs undergoing sham CPB with heparinisation and aortic cannulation had normal retinal perfusion. This study demonstrates a very high incidence of microvascular occlusions in the territory of the internal carotid artery during CPB consistent with microembolism.

*By Invitation

6. Reconstruction of Left Ventricular Wall with Autologous Pericardium

TIRONE E. DAVID, CHRISTOPHER M. FEINDEL*

and GLORIANNE ROPCHAN*

Toronto, Ontario

Autologous pericardium has been used to reconstruct different areas of the left ventricle in 24 patients (pts). Fifteen pts had active infective endocarditis with abscess formation in an infarcted myocardium (3 pts), in the aortic root (5 pts), in the central fibrous body and mitral annulus (3 pts), in the central fibrous body and tricuspid annulus (1 pt) and in the mitral annulus alone (3 pts). The remaining patients had acute ventricular septal defect (5 pts), acute rupture of the posterior wall of the left ventricle following mitral valve replacement (1 pt) or excessive calcification of the mitral annulus to properly secure a prosthesis (3 pts).

The diseased portion of the myocardium and or fibrous skeleton of the heart was excised when indicated, and reconstruction was accomplished by suturing an appropriately tailored autologous pericardial patch directly to the endocardium. A double patch was used in pts with acute ventricular septal defect (one on each side of the septum). All other pts had a single patch sutured directly to the endocardium with running polypropelene suture. In cases of valvular disease, a prosthetic valve was secured to the patch in those areas where the pericardium replaced the fibrous skeleton.

There were 3 hospital deaths (one pt had acute VSD and 2 pts had annular abscesses) but none were related to the technique or patch material. All survivors have been followed from 3 to 61 months, mean of 32. There has been no late death and every patient has had at least one non-invasive assessment test of the pericardial patch. No patient has had patch or valve dehiscence or pericardial patch aneurysm.

Autologous pericardium can be safely used to reconstruct left ventricular wall. It handles very well and problems with suture line bleeding is practically non-existent. Autologous pericardial patch is particularly useful in pts with myocardial and or annular abscess secondary to infective endocarditis.

*By Invitation

7. Identification of Free Radicals Produced During Myocardial Ischemia and Reperfusion Using Electron Paramagnetic Resonance Spectroscopy and High Precision Liquid Chromatography (Forum)

JOHN M. LUBER*, PARINAM S. RAO*

and MARK S. CROWDER*

New Hyde Park, New York and Danbury, Connecicut

Sponsored by: DENIS H. TYRAS

New Hyde Park, New York

The presence of free radicals (FR) were assessed in left ventricular myocardium (LV) of 1 month old swine (N = 5) during cardiopulmonary bypass pre crossclamp (C), with 30 minutes of normothermic ischemia (I) and after 10 minutes of reperfusion (R). LV biopsies were quick frozen in liquid N₂ and examined by EPR at 10°K and by high precision liquid chromatography (HPLC) using electrochemical (EC) and UV detectors. The EPR difference spectra between C and I revealed a 3 line spectrum (g = 2.015, 2.005, 1.988) with splitting between lines of 2.25 m Tesla. Power saturation studies revealed g = 2 signal saturation at lower microwave powers than the 3 line EPR signal. This suggests a nitrogen triplet or an organic biradical. EPR signals with I were irreversibly lost at 200°K. With R the 3 line spectrum was present but diminished.

HPLC verification of FR was accomplished by incubating the LV homogenate (50 mg) with a spin trap (5.5-dimethyl-l-pyrroline-n-oxide [DMPO]); 50 ul, 10mM at pH = 7.8 in 50 mM phosphate buffer using a 5 u resolve C₁₈ column and EC. A DMPO adduct signal was noted. This could not be modified by superoxide dismutase (SOD), or dimethylthiourea indicating that it is not an oxy derived species. Lipid peroxidation measured by tissue malondialdehyde (MDA) with UV absorbance at 254 nm and catechol levels using EC revealed the following:

These data indicate that (1) with 30 minutes of normothermic ischemia there are directly measurable FR by EPR and HPLC (2) these FR do not seem to be oxygen derived species and, therefore, cannot be effected by SOD or catalase (3) lipid peroxidation and catechol oxidation may play a role in cellular damage during I but not R. (4) The FR concentration does not increase with R and, therefore, efforts at free radical scavenging emphasizing the reperfusion period may be in error.

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

"Some Thoughts From The Other Side of The Table, or The Last Presidential Address"

Norman E. Shumway, M.D., Stanford, California

12:15 p.m. Adjourn for Lunch

Luncheon Service available in Exhibit Area - Wacker Hall

*By Invitation