TUESDAY AFTERNOON, April 27, 1993

1:45 p.m. SIMULTANEOUS SCIENTIFIC SESSION A CARDIAC SURGERY

Grand Ballroom B

Moderators: D. Glenn Pennington, M.D.

Irving L. Kron, M.D.

17. Results of Non-Guided Subtotal Endocardectomy Associated With Left Ventricular Reconstruction in 106 Patients With Ischemic Ventricular Arrhythmias

VINCENT DOR, M.D., MICHEL SABATIER, M.D.*,

FRANCOISE MONTIGLIO, M.D.*, PHILIPPE ROSSI, M.D.*,

MARISA DI DONA TO, M.D.* and

ANNA TOSO, M.D.*

Monte-Carlo, Monaco and Firenze, Italy

From June 1987 to September 1992, 284 pts underwent left ventricular (LV) reconstruction with endoventricular circular patch plasty and septal exclusion for postinfarction LV aneurysm or severe LV wall motion abnormalities. 106 out of these pts, presenting spontaneous and/or inducible ventricular tachycardia (VT), represent the study group (mean age 58 ± 8 years). There were 97 anterior akinetic or dyskinetic and 9 posterior aneurysms. 69 pts were in NYHA class III/IV; indication for surgery was angina in 35% of pts, intractable VA in 11% and a combination of angina, congestive failure and VA in the remaining pts. 18 pts were operated in emergency. In these 106 pts before LV patch reconstruction, subtotal endocardiectomy of fibrous in-traventricular scar was performed; cryotherapy at the border of the lesion was associated in 67 pts and coronary revascularization, including infarcted area, was performed in 93% of pts. All pts underwent complete hemo-dynamic study including programmed ventricular stimulation (PVS), when not contraindicated, before and early after surgery (10-15 days). Clinical follow up is available in all pts (min 2 max 62 months, mean follow-up 19,8 m). At present 34 pts have complete hemodynamic control, including PVS after 1 year.

Preoperative data: 49 pts had documented episodes of spontaneous VT; 57 without spontaneous VTs had inducible VT at PVS and 20 had spontaneous and inducible VTs. PVS was contraindicated in 23 pts. Mean EF was 33 ± 11%; contractile EF was 41 ± 10%; EDVI was 124 ±63 ml/m2; CI was 2.8 ± 6.5 l/min/m2; mean pulmonary artery pressure (PAP) was 20 ± 8 mmHg. Postoperative data: perioperative mortality rate was 7.5% (8 pts). EF improved significantly (46 ± 10% p<.001) as well as PAP decreased (17 ± 7 mmHg p<.001); CI did not change significantly (2.7 ± 5.5 l/min/m2). Spontaneous VTs was recorded in 2 pts who had spontaneous VTs pre-operatively; VT was inducible in 8 pts (Wilcoxon test p<.0001). Four of these were contraindicated and 4 were inducible before surgery. At late hemodynamic control EF was still significantly increased (32 ±10 basal; 47 ± 11 early control and 46 ± 11 % after 1 year p< .01, n = 34). In this group there were 11 spontaneous VTs and 23 inducible VTs in basal conditions. After 1 year VT was induced in 4/34 pts and no spontaneous VTs occurred. Three of the 4 pts with inducible VT were under amiodarone and they were inducible also at early control. Follow-up: two sudden deaths occurred among 7 late deaths; no spontaneous VTs were clinically recorded in the surviving pts, all controlled.

In conclusion, non guided subtotal endocardiectomy +/- cryotherapy can be safely performed during surgery for LV aneurysm in pts with severely depressed LV function and it drastically and significantly reduces the occurrence of spontaneous and inducible VT early after surgery. In pts controlled after 1 year the beneficial effect on ventricular arrhythmias is maintained as well as the improvement of LV geometry and function.

*By invitation

18. The Effects of Myocardial Revascularization on the Incidence of Implanted Defibrillator Discharge in Patients With Cardiac Dysfunction

HOOSHANG BOLOOKI, M.D.,

MICHAEL D. HOROWITZ, M.D.*,

GEORGE M. PALATIANOS, M.D.*,

ALBERTO INTERIAN, JR., M.D.*, MICHAEL BARRON, M.D.*

and RICHARD A. FERRYMAN, M.D.*

Miami, Florida

We studied 115 patients (pts) who had survived sudden death and had received an automatic implantable cardioverter defibrillator (AICD) either alone (Group A, n = 70) or after coronary bypass operation (CABG + AICD = Group B, n=45). All patients had inducible ventricular tachycardia/ fibrillation after previous myocardial infarction associated with coronary disease and left ventricular (LV) dysfunction. Pts with LV aneurysm and ablative procedures were not included. Mean age (63 vs 64 yrs), N.Y.H.A. Class IV (95% vs 93%), LV ejection fraction (27% vs 27%), LV diastolic pressure (21 vs 20mmHg), operative mortality 2.9% vs 2%, mean follow-up 30 vs 42 months (range 18-90 months), and sudden death rate 3.5% vs 2.4%/pt year were similar in both groups. However, the incidence of AICD discharge was significantly higher in Group A than Group B pts (75% vs 59%, P = 0.04). Furthermore, all Group A pts who had defibrillator discharge experienced the first shock in the first 24 months while Group B pts experienced their first shock within the first 12 months after implantation of the device. The overall survival rate at 6.5 years was 58% and 85% (P = n.s.) and the defibrillator shock-free survival rate was 24% and 47% (p = 0.04) for Groups A and B pts respectively. Patients with malignant ventricular arrhythmia and LV dysfunction who receive myocardial revascularization and an AICD implantation experience fewer defibrillator shocks than pts who receive AICD alone.

*By Invitation

19. Nonthoracotomy Lead System for Implantable Defibrillator

BRADFORD P. BLAKEMAN, M.D.*,

HENRY J. SULLIVAN, M.D., ALVARO MONTOYA, M.D.*,

DAVID WILBER, M.D.*, BRIAN OLSHANSKY, M.D.*,

JEFFREY BAERMAN, M.D*, JOHN KALL, M.D.*

and ROQUE PIFARRE, M.D.

Maywood, Illinois

Automatic implantable defibrillators have become a standard therapy for ventricular arrhythmias. A new lead system consisting of one (CPI) or two (Medtronic) endocardial leads and a subcutaneous patch not requiring a thoracotomy are currently under investigation at our institution. Eighty-five insertions for the nonthoracotomy lead system (NTL) have been attempted and sixty-four were successful (75 percent). Sixty-five of the total patients were male and mean age was 57 years. Forty-nine patients (56 percent) had previous open heart surgery. Left ventricular ejection fraction for the entire group demonstrated a mean of 28.4 percent, a range of 12 to 74 percent. Operative data noted defibrillation thresholds (DFT) for the sixty-four successful patients to be a mean of 18.9 joules (range 3-25). The number of defibrillations necessary for successful NTL implants was a mean of 10.3. The reasons for unsuccessful implants were insufficient DFT's - 19 patients and inability to position endocardial lead - 2 patients. Conventional lead systems were implanted in NTL failure patients by the following: lateral thoracotomy 11, sternotomy 9 and subxyphoid 1. Ten of the twenty-one NTL failure patients required three or more conventional patches to attain adequate DFT's. Length of procedure for a successful NTL system was a mean of 123.7 minutes (range 30-270 minutes). Success of implant could not be linked to previous heart surgery, size of chest wall or ejection fraction. No inappropriate or unsuccessful defibrillations have occurred with implanted systems to date. Complications directly related to the device requiring further surgery included lead migration - 5 patients, hematoma - 3 patients and infection - 1 patient. The nonthoracotomy lead system demonstrates reasonable promise in this population.

2:45 p.m. INTERMISSION - VISIT EXHIBITS

*By Invitation

3:15 p.m. SIMULTANEOUS SCIENTIFIC SESSION A CARDIAC SURGERY

Grand Ballroom B

20. Aprotinin for Coronary Bypass Surgery: Efficacy, Safety and Influence on Early Vein Graft Patency. Results of a Multi-center, Randomized, Double-Blind, Placebo-Controlled Study

JOHN H. LEMMER, JR., M.D.*, WILLIAM STANFORD, M.D.,

SHARON L. BONNEY, M.D.*, JEROME F. BREEN, M.D.*,

EVA V. CHOMKA, M.D.*, W. JAY ELDREDGE, M.D.*,

WILLIAM W. HOLT, M.D.*, ROBERT B. KARP, M.D.,

GLENN W. LAUB, M.D.*, MARTIN J. LIPTON, M.D.*,

HARTZELL V. SCHAFF, M.D., CONSTANTINE J. TATOOLES, M.D.

and JOHN A. RUMBERGER, Ph.D., M.D.*

Iowa City, Iowa

Two hundred sixteen patients (pts) undergoing primary (151) and repeat (65) coronary bypass surgery (CABG) procedures at 5 hospitals were randomized to receive high-dose aprotinin or placebo during surgery. Saphenous vein graft (SVG) patency (330 total grafts) was evaluated by cine-computed tomography 7 to 45 days after operation in 164 (76%) pts. Assessment of patency was determined by group consensus without knowledge of the patients' randomized status. Patency was analyzed on both a per-patient and per-graft basis using Chi Square and Fishers Exact Test methods. Significance was at a p≤0.05 level.

Results:
	Aprotinin	Placebo	p-Value
Primary CABG pts requiring RBCs	28/74 (38%)	35/67 (52%)	0.052
Repeat CABG pts requiring RBCs	7/23 (30%)	23/32 (72%)	0.001
Primary pts-RBC vol transfused (n = 141)	362ml	606ml	0.023
Repeat pts-RBC vol transfused (n = 54)	164ml	931 ml	0.005
Pts with ≥ 1 closed SVG	13/83 (16%)	7/81 (9%)	0.170
Closed SVGs	14/176 (8%)	8/163 (5%)	0.248
Myocardial infarcts	10/108 (9%)	7/108 (6%)	0.448
Deaths	6/108 (6%)	4/108 (4%)	0.517

Conclusions: Prophylactic aprotinin decreases transfusion requirements in CABG pts, particularly in repeat procedures. In this study, there was a trend toward a higher rate of early SVG closure in patients who received aprotinin as compared to those who received placebo, although this result did not reach statistical significance. While this trend did not translate into a difference in perioperative myocardial infarctions or patient deaths, further investigations regarding the safety of routine aprotinin use appear indicated.

*By Invitation

21. Arterial Revascularization in 300 Patients With the Right Gastroepiploic Artery and Internal Mammary Arteries

JAN G. GRANDJEAN, M.D.*, PIET W. BOONSTRA, M.D., Ph.D.*,

PETER DEN HEIJER, M.D.* and TJARK EBELS, M.D., Ph.D.*

Groningen, Ho/land

Sponsored by: John W. Kirklin, M.D., Birmingham, Alabama

From September 1989 to September 1992, the right gastroepiploic artery (GEA) in combination with the internal mammary artery (IMA) was used in 300 patients who underwent coronary artery bypass grafting. The GEA was the primary choice in preference to a saphenous vein. There were 263 men and 37 women, ranging in age from 31 to 77 years (mean age 58,2 years). Thirty-nine patients (13%) underwent previous bypass procedures with vein grafts. In 150 patients (50%) we used the left IMA in conjunction with the GEA (in two patients combined with a vein graft), and in 133 patients (44.3%) both IMA's were used with the GEA in one operation. In 17 patients (5.7%) we used the GEA as a single graft. Revascularization in 9 patients (3%) was combined with another cardiac procedure; three times an aortic valve replacement, two mitral valve repairs, and four resections of an aneurysm of the left ventricle. Ten patients died in hospital (3.3%; 70% CL 2.3-4.8%); two cases were directly related to the GEA. There was no late mortality. Four patients had to be operated again; one had a new stenosis in a previously not stenosed coronary vessel, one patient due to mitral and tricuspid valve endocarditis, one patient with an open GEA had an occlusion of the left IMA, and one with a closure of a single GEA was reoperated with a right IMA. Eighty patients were re-catheterized 1 to 18 months postoperatively (mean 9 months). Graft patency in GEA increased from 77% in the first semester of the program to 94% in the fourth semester. Patency of the internal mammary grafts was 96%. We conclude that graft failure of the GEA was related to a "learning curve". Furthermore, the GEA may well be the graft of choice in conjunction with the internal mammary arteries.

*By Invitation

22. Donor Shortage in Heart Transplantation: Is Extension of Donor Age Limits Justified?

UGOLINO LIVI, M.D.*, UBERTO B. BORTOLOTTI, M.D.*,

GIOVANBATTISTA LUCIANI, M.D.*,

GIOVANNI BOFFA, M.D.*, GAETANO THIENE, M.D*

and DINO CASAROTTO, M.D.*

Padova, Italy

Sponsored by: Norman E. Shumway, M.D., Stanford, California

Chronic shortage of donor organs for HTx led us to extend donor age limits. To verify the effectiveness of such policy we have compared the results of HTx in 40 patients (pts) using donors >40 years (yrs) (Group 1) with 69 pts >50 yrs of age who had HTx using donors <40 yrs (Group 2) from November 1985 to September 1992. The 2 groups were comparable in terms of mean recipient age, recipient and donor sex and indication for HTx. Mean donor age was 46 ± 4 yrs in Group 1 (range, 40 to 59) and 23 ± 7 yrs in Group 2 (range, 8 to 38) (p<0.01). In Group 1 cerebrovascular accidents were more common as cause of donor death (60% vs 16%, p<0.01), while no difference was found in ischemic time (139 ± 41 vs 151 ± 52 m¹, p = ns). There were 4 early (<30 days) deaths in Group 1 (10%) and 10 in Group 2 (14%) (p = ns); 2 pts (5%) died late post-HTx in Group 1 and 3 (4%) in Group 2 (p = ns). Acute graft failure leading to death or re-HTx was more frequent in Group 1 (10% vs 6%, p<0.01). Mean follow-up is 29 ±20 months (range, 1 to 72) in Group 1 and 30 ± 20 months (range, 2 to 74) in Group 2 pts (p = ns). Actuarial survival is 86 ± 6% vs 83 ± 7% and 84 ± 7% vs 80 ± 8% (p = ns) at 1 and 4 yrs in Group 1 vs Group 2, respectively. Angiographic control has shown a similar left ventricular ejection fraction at 1 (59 ± 14% vs 63 ± 10%) and 4 yrs (66 ± 14% vs 62 ± 10%) (p = ns). However, Group 2 pts had a higher freedom from coronary artery disease (CAD) of any degree at 4 yrs (84 ± 7% vs 75 ± 8%, p<0.01).

Donors >40 yrs of age can be used for HTx with mid-term results comparable to that of younger donors. A higher incidence of CAD and acute graft failure seems not to affect survival after HTx with donors >40 years, but the impact of CAD on the performance of older grafts must be assessed at longer follow-up.

*By Invitation

23. Seven Years Experience With Bridging to Cardiac Transplantation

D. GLENN PENNINGTON, M.D.,

LAWRENCE R. McBRIDE, M.D.*, PAMELA S. PEIGH, M.D.*,

LESLIE MILLER, M.D. * and

MARC T. SWARTZ, B.A.*

St. Louis, Missouri

Although bridging to cardiac transplantation has become a therapeutic option for transplant candidates who deteriorate while awaiting a donor heart, short term efficacy has not been proven and long term survival has not been reported. We retrospectively reviewed 42 patients (pts) who had circulatory assist devices placed as a bridge to transplant between May 1985 and July 1992. The 33 men and 9 women ranged in age from 12-65 years (mean 43 years). Thirty pts were supported with Thoratec (17 left ventricular, 13 biventricular), 10 with Novacor and 2 with Jarvik J-7-70 devices. The duration of device support was from 4 hours - 440 days (mean 47 days). Fourteen pts were not transplanted because of infection (10 pts), renal failure (5 pts), bleeding (9 pts), cerebrovascular accident (3 pts) and died. Two pts were weaned from support and survived without transplantation. Twenty-six pts were transplanted, with 25 survivors (96%). Overall survival is 64% (27/42). Duration of survival has ranged from 3-90 months (mean 35.3 months). Of the 27 survivors, there were 3 late deaths (all transplants) at 4, 6 and 14 months. Post-transplant actuarial survival at 1, 5 and 7 years is 86%, 81% and 81%. Twenty-three of the 24 pts presently alive are NYHA functional status I.

These data demonstrate the short and long term efficacy of bridging to transplantation with circulatory support devices. The excellent survival rates and full functional recovery of transplanted patients ensures that donor organs are not being "wasted" on the sickest patients.

4:40 p.m. EXECUTIVE SESSION (Members Only)

7:00 p.m. ANNUAL DINNER/DANCE Black Tie (Optional)

Regency Ballroom

*By Invitation

TUESDAY AFTERNOON, April 27, 1993

1:45 p.m. SIMULTANEOUS SCIENTIFIC SESSION B GENERAL THORACIC SURGERY

Grand Ballroom D

Moderators: L. Penfield Faber, M.D.

J. Kent Trinkle, M.D.

24. Determinants of Outcome for Lung Transplantation in Cystic Fibrosis

HANI SHENNIB, M.D.*, GILBERT MASSARD, M.D.*,

MICHAEL P. KAYE, M.D., JOHN WALLWORK, M.D.*,

MICHEL NOIRCLERC, M.D. *, DA VID MULDER, M.D.

and BRUCE REITZ, M.D. Montreal, Canada

Cystic Fibrosis (CF) is currently the commonest indication for bilateral lung replacement. The choice of patient, procedure and post-operative management is controversial. We analyzed data on 227 CF transplant recipients reported to the International Heart and Lung Transplant Registry as of December 1991. There were 156 heart/lung (HLT) and 71 bilateral lung transplants (BLT) [29 En bloc (EB) and 42 sequential single (SL) transplants]. Recipient's age, sex, CMV and functional status, type of procedure and site of transplantation (North America/United Kingdom), donor age, CMV match, ischemia time and whether induction steroids and cytolytic therapy were used or not, were examined as univarite and multivarite determinants of outcome. Overall, one year survival was 68.5% for HLT and 60% for BLT. One year survival of HLT in U.K. (74%) was significantly better than in North America (NA) (45%) (p < 0.01) where less HLT were done. Patient characteristics were similar, however ischemia time was longer in U.K. (204 ± 60 min) than in NA (173 ± 72 min) (p < 0.025). Bilateral lung transplants, on the other hand, did much better at 1 year in NA (63%) than in U.K. (33%). However, much smaller numbers were done in the U.K. to allow valid statistical testing. On further analysis of North American data, 1 year survival of SL (71%) were noted to be superior to EB transplants (51%) and patients with NYH-III or less functional status (72%) did better than NYH-IV patients (29%) (p < 0.01). Steroid administration was associated with significant (p < 0.05) improvement in survival (steroid 64°7o, no steroids 37%) while cytolytic therapy had no effect. Results also appeared to improve with time (p < 0.03). As most mortality occurred in the early post-operative period [EB (41%), HLT (22%), SL (17%)], we conclude that first year survival in CF transplant recipients is determined primarily by 1) The type of surgical procedure and where it is performed i.e. center's experience, 2) Pre-operative functional status of recipient and, 3) Early post-operative administration of steroids. In North America, the best results were achieved with bilateral sequential lung transplantation of patients with NYH-III or less functional status and with early post-operative use of steroids.

*By Invitation

25. Anastomotic Pitfalls in Lung Transplantation

BARTLEY P. GRIFFITH, M.D., MITCHELL J. MAGEE, M.D.*,

REMI HOUEL, M.D.*, IVAN F. GONZALES, M.D.*,

JOHN M. ARMITAGE, M.D. * and

ROBERT J. KEENAN, M.D.*

Pittsburgh, Pennsylvania

While airway, arterial, and venous connections required for lung transplantation appear simple, in practice we have encountered morbid early stenosis and obstructions which are now avoided by technical modifications gradually made since 1985 in 184 cases (60 SL and 74 DL).

Our eight initial DL procedures were performed with tracheal anastomoses and omental wraps, but ischemic disruption, a 75% (6/8) rate of complications, resulted in our subsequent use of bi-bronchial connections (Table). 192 bronchial anastomoses (BA) have been reviewed (60 SL, 66 DL). While all have been constructed between the donor trimmed to 1-2 rings above the upper lobe origin and host divided at its emergence from the mediatinum, the suture technique has evolved. 9/28 (32%) early BA with end-to-end suture and intercostal muscle wrap developed ischemic or stenotic complications, but the telescoping technique without wrap in 164 BA has reduced the problem to 12%(17/164). Twelve anastomoses required temporary intraluminal stenting. Because the telescopic method using horizontal mattress suture has rarely been associated with an obstructing flange of invaginated cartilage, it has been replaced by a telescoping suture figure-of-eight which holds the invaginated rings closely to the outer wall.

AIRWAY ANASTOMOSES
	N	Ischemia	Dehiscence	Stenosis	TOTAL
TRACHEAL
End-to-End	8	47% (3)	47% (3)	-0-	75% (6)
BRONCHIAL
End-to-End	28	13% (4)	13% (4)	4% (1)	32% (9)
Telescoping	164	7% (11)	-0-	5% (8)	12% (19)

Vascular anastomotic obstructions have occurred in 6 arterial (excessive length 4, restrictive suture/clot 2) and 3 venous (excessive length 1, restrictive/clot 2) connections. Suspicion of arterial obstruction has been prompted by persisting pulmonary hypertension and reduced flow to the allograft measured by postoperative nuclear scan and widened A-a gradient. Venous abstractions are suggested by persisting radiographic and clinical pulmonary; dema. Currently, donor and recipient arteries are shortened to within 2-3 cm 3f the hilum and mediastinum respectively, and the donor venous atrial cuffs and trimmed to a muscle border of only 5 mm. The arteries are divided to maximize circumference of the suture line which is now interrupted.

Modifications of earlier techniques have improved our early success in lung transplantation and might be considered by others entering this demanding field.

*By Invitation

26. Results of Single and Bilateral Lung Transplantation in 130 Consecutive Recipients

JOEL D. COOPER, M.D.,

G. ALEXANDER PA TTERSON, M.D.

and ELBERTP. TRULOCK, M.D.*

St. Louis, Missouri

From its inception in July, 1988 until July 31, 1992, 142 lung transplants were performed in 138 patients by our lung transplant group. Eight en bloc double lung transplants were performed in the initial year with 75% mortality, after which this procedure was abandoned in favor of the bilateral sequential technique. Experience with the remaining 134 single (SLT) or bilateral (BLT) transplants performed in 130 recipients forms the basis for this report. Seventy-three patients underwent SLT and 57 patients underwent BLT for the following indications: emphysema-68; cystic fibrosis (C.F.)-20; primary pulmonary hypertension (P.P.H.)/Eisnemenger's-20; idiopathic pulmonary fibrosis (I.P.F.)-14 and other diagnoses-8. Hospital mortality occurred in 11 patients (8%) and late mortality in an additional 12 (9%). One hundred and seven patients (82%) are currently alive with a mean survival time of 17 months. By the actuarial method one-year survival is 81% and two-year survival is 80%. Survival, and results by diagnostic groups are as follows:

SURVIVAL
Diagnosis	n	% alive		mean followup (months)		1 year actuarial survival
emphysema	68	82%		16		84%
C.F.	20	85%		12		81%
P.P.H./Eisenmenger's	20	85%		16		81%
I.P.F	14	71%		15		68%
RESULTS
Diagnosis	6 minute walk (meters)				Room air PO2 (mmHg)
	preop		1 year		preop		1 year
emphysema	266		595		56		89
C.F.	348		711		48		98
P.P.H./Eisenmenger's	194		596		60		82
I.P.F	261		550		53		76

The 11 hospital deaths resulted from cardiac failure (2); aspergillus infection (2); sepsis (3); ARDS (2), arrhythmia (1), and airway dehiscence (1). Bronchiolitis obliterans currently affects approximately 25% of survivors and was the direct or indirect cause of 6 out of the 12 late deaths.

Lung transplantation can now achieve early results similar to those obtained with the more established types of organ transplants. The major unsolved problems relate to morbidity and mortality associated with chronic immunosuppression and chronic rejection, and the acute shortage of donor organs.

2:45 p.m. INTERMISSION - VISIT EXHIBITS

*By Invitation

3:15 p.m. SIMULTANEOUS SCIENTIFIC SESSION B GENERAL THORACIC SURGERY

Grand Ballroom D

27. Surgical Management of Non Small Cell Carcinoma With N2 Disease

GOPI C. MANNAM, FRCS*, PETER COLOSTRA W, FRCS*,

DA VID K. KAPLAN, FRCS* and

PANOS MECHAIL, M.D.*

London, England and Rhodes, Greece

Sponsored by: Thomas W. Shields, M.D., Chicago, Illinois

Between 1979 and 1989, 876 patients with Non Small Cell Carcinoma (NSCLC) were referred to our unit. One hundred and forty-six patients were judged not suitable for surgical treatment on clinical, radiological or bronchoscopic findings. Cervical mediastinoscopy and/or anterior mediastinotomy showed that 151 patients had metastases into the superior mediastinal lymph nodes (N2 disease) and were therefore deemed inoperable. Except for one patient who had single nodal station positive at mediastinoscopy, all other patients proceeding to thoracotomy, 578, were thought on the basis of Ct scan (89) and/or mediastinal exploration (59), not to have N2 disease. Despite our efforts to avoid surgery in patients with N2 disease, routine mediastinal node dissection showed that 149 patients had previously unsuspected N2 disease. Resection was possible in 130 (87.3%) by pneumonectomy (72), bilobectomy (7), lobectomy (49), or lesser resection (2). In three patients the resection was incomplete (2.3%), but in 127 cases a complete resection was performed (85%). The histology of tumors in these 149 patients showed, 72 were squamous cell carcinoma, 54 adenocarcinoma, 14 large cell carcinoma and 9 of mixed type.

Five patients died in hospital following thoracotomy. Complete follow-up was obtained in 109 patients and the mean follow-up period was 27.25 months (1-116). The actuarial 5 year survival for those having complete resection was 19.4%. Neither cell type nor the nodal station of the metastatic node influenced long-term survival. There was however, a statistically significant difference favoring long-term survival in those patients with only one nodal station involved compared to those with more than one (p<0.033).

Despite rigorous preoperative investigations it is possible to encounter mediastinal node metastasis first time at thoracotomy by routine mediastinal node dissection. We consider resection is justified in these patients who have already necessarily incurred the morbidity and mortality of thoracotomy, as long as complete resection is possible.

*By Invitation

28. Comparison of Survival and Lung Function Following Sleeve Lobectomy and Pneumonectomy for Lung Cancer

HENNING A. GAISSERT, M.D.*,

DOUGLAS J. MATHISEN, M.D., HERMES C. GRILLO, M.D.,

ASHBY C. MONCURE, M.D., JOHN C. WAIN, M.D.*

and ALAN D. HILGENBERG, M.D.

Boston, Massachusetts

Sleeve lobectomy is a widely accepted bronchoplastic procedure for patients with normal and compromised lung function. Scarce information is available regarding postoperative lung function and comparative survival after pneumonectomy. We have performed 71 sleeve lobectomies for lung cancer. Pulmonary or cardiac function was compromised in 38 patients and normal in 33. Histology was squamous cell in 49, adenocarcinoma in 18, large cell in 3, and adenosquamous in 1. Resection involved the upper lobe in 47 patients (right 37, left 10), lower and middle lobe in 10, and bilobectomies in 14. Postsurgical stage was I in 28 patients (39 percent), II in 33 patients (47 percent), and III in 10 patients (14 percent).

Bronchoplastic procedures were compared to 53 patients undergoing pneumonectomy for lung cancer. Operative mortality for sleeve lobectomy was 2.8 percent compared to 7.5 percent for pneumonectomy. Cumulative 5-year survival following pneumonectomy was 42 percent vs. 40 percent for all sleeve lobectomies. In the bronchoplastic group, 5-year survival in patients with normal lung function was 57 percent vs. 32 percent for compromised function. Survival in NO disease was 52 percent vs. 36 percent in Nl disease. Survival following upper sleeve lobectomy was 44 percent and 40 percent for lower, middle, and bilobectomies.

Mean postoperative reduction in FEV1 was 8.6 percent in patients with compromised function and 10 percent in patients with normal function. Ventilation perfusion scans confirmed preservation of function in remaining pulmonary parenchyma.

Sleeve lobectomy is the procedure of choice for anatomically suitable lesions for patients with both normal and compromised lung function. Survival is comparable to pneumonectomy and superior for patients with normal lung function. Preservation of lung function in the remaining lobes is confirmed by our studies.

*By Invitation

29. Prospective Evaluation of Unilateral Adrenal Masses in Patients With Operable Non-Small Cell Lung Cancer: Impact of Magnetic Resonance Imaging

MICHAEL E. BURT, M.D., Ph.D., ROBERT HEELAN, M.D.*

DANIEL COIT, M.D.*,

PATRICIA M. McCORMACK, M.D.

and ROBERT J. GINSBERG, M.D.

New York, New York

With computed tomography (CT) in the staging of NSCLC, 4 percent of otherwise operable patients have been found to have a unilateral adrenal mass. Previous studies have suggested that MRI has the ability to differentiate between benign (adenoma or hyperplasia) and malignant adrenal masses. Since this differentiation is critical to treatment planning, we designed a prospective study to evaluate the efficacy of MRI in diffentiating a benign from a malignant adrenal mass in patients with otherwise operable NSCLC.

Methods: All patients with potentially operable NSCLC were prospectively staged by history, physical examination, pulmonary function testing, cardiac evaluation, CT scan of the chest and upper abdomen (including the adrenals), CT head scan, and bone scan. All operable patients with a unilateral adrenal mass underwent respiratory and cardiac gated thin section MRI of the adrenals (1.5 Tesla GE signa system). One radiologist interpreted the MRI blinded and based on the T1 and T2 weighted images judged whether the adrenal mass was benign or malignant. The patients then underwent a percutaneous needle biopsy of the adrenal mass, if technically feasible. If the result of the needle biopsy was non-diagnostic, or if the biopsy was not feasible, an adrenalectomy through a posterior approach was performed. Data expressed as mean ± SD. Differences were determined by Fisher exact test or Student's unpaired t-test. Signficance defined as p<0.05.

Results: Twenty-seven patients with a unilateral adrenal mass entered the study; there were 11 men and 16 women whose ages ranged from 42-75 yrs (median 58). Four patients had epidermoid and 23 adenocarcinoma. The locoregional stage was I in 8, II in 4, IIIA in 12, and IIIB in 3. Twenty-five completed the MRI (2 did not, due to claustrophobia). Five adrenal masses (19%) were metastatic NSCLC (adeno CA = 4, epidermoid = 1); 22 masses (81%) were benign (adenoma = 20, hyperplasia = 2). There were no significant differences in age, sex, histology, or locoregional stage between those with a benign versus a malignant mass. However, the malignant masses were significantly larger (3.8 ± 1.9 cm; range 2.5 - 7.1; median 3.1) than the benign (2.0 ± 0.4 cm, range 1.2 - 2.8; median 2.0) (p = 0.002). Of those having an MRI (n = 25), MRI correctly predicted a malignant mass in the four patients with a histologically confirmed metastasis from NSCLC. However, of the 21 histologically benign masses, the MRI was interpreted as benign in only 4 and malignant in 25. Therefore, although the false negative rate was 0%, the false positive rate was 81%.

Conclusion: Most adrenal masses in otherwise operable patients with NSCLC are benign. If during staging an adrenal mass is found, histologic diagnosis must be obtained. MRI is not useful in the differentiation of benign and malignant adrenal masses in patients with NSCLC.

*By Invitation

30. Treatment and Prognosis in Bronchial Carcinoids Involving Regional Lymph Nodes

NAEL MARTINI, M.D., MUHAMMAD ZAMAN, M.D.*,

MANJITS. BAINS, M.D., MICHAEL E. BURT, M.D., Ph.D.,

PATRICIA M. McCORMACK, M.D.,

VALERIE W. RUSCH, M.D. and

ROBERT J. GINSBERG, M.D.

New York, New York

Bronchial carcinoids constitute less than 5% of all lung tumors and 10-15% of these have regional lymph node metastases at diagnosis. Over a 40 year period (1953-92), 25 patients were surgically treated by us for bronchial carcinoids with metastases to regional lymph nodes (Nl or N2). The tumors were located centrally, involving main or lobar bronchi in 12 patients and were peripheral in 13. None had a carcinoid syndrome of Ml disease. Histologically, the carcinoids were classified as typical in 12 and atypical (neuroendocrine carcinoma) in 13. The median age of patients with typical carcinoids was 42 years and for atypical carcinoids 62.

Pneumonectomy was performed in 11 patients, sleeve lobectomy in 1, lobectomy in 7 and bilobectomy in 6. All had a mediastinal lymph node dissection. At final staging, 10 had N1 disease and 15 had N2. The number of N1 or N2 patients was equally divided between the 2 types of carcinoids.

No adjuvant treatment was given to the 10 patients with N1 disease. External radiation therapy was given postoperatively to 9 of 15 N2 patients, and oral cyclophosphamide to 1. There was only 1 local recurrence (in a patient with Nl disease) and 7 distant metastases in liver, bone or brain (6 in patients with N2 and 1 in a patient with Nl).

The overall 5-year survival (Kaplan-Meier) was 83% (median follow-up: 62 months). There was no difference in disease free survival between patients with N1 or N2 disease. However survival and recurrence rates differed between typical and atypical carcinoids. In typical carcinoids, the 5-year survival was 92% and the 5-year disease free survival 100% (the one recurrence occurred at 8!/2 years). In atypical carcinoids, the 5-year survival was 73% (p = .06) and the 5-year disease free survival 57% (p = .025).

We conclude that complete resection is effective for bronchial carcinoids, despite the presence of regional lymph node metastases, and results in long-term survival. In this group of patients, recurrence appears more dependent on histologic subset than nodal status. There is no evidence that postoperative radiation therapy is beneficial and we are unable to assess the merit of systemic adjuvants since none had effective systemic treatment.

4:40 p.m. EXECUTIVE SESSION (Members Only)

7:00 p.m. ANNUAL DINNER/DANCE Black Tie (Optional)

Regency Ballroom

*By Invitation

TUESDAY AFTERNOON, April 27, 1993

1:45 p.m. SIMULTANEOUS SCIENTIFIC SESSION C CONGENITAL HEART DISEASE

Grand Ballroom F

Moderators: Edward L. Bove, M.D.

Thomas L. Spray, M.D.

31. Cardiopulmonary Bypass Significantly Impairs Surfactant Activity in Children

FRANCIS X. McGOWAN, M.D.*, PEDRO J. DEL NIDO, M.D.*,

GEOFFREY KURLAND, M.D.*, MACHIKO IKEGAMI, Ph.D.*,

ETSURO K. MOTOYAMA, M.D.* and RALPH D. SIEWERS, M.D.*

Pittsburgh, Pennsylvania and Los Angeles, California

Sponsored by: Bartley P. Griffith, M.D., Pittsburgh, Pennsylvania

Pulmonary dysfunction following cardiopulmonary bypass (CPB) in children remains a primary source of morbidity and mortality, particularly in infants. The effects of cardiopulmonary bypass on surfactant activity and lung mechanics in children has not been directly evaluated. In experimental animal models of lung injury (endotoxin), surfactant activity has been shown to significantly decrease with loss of the surface active, high density large surfactant aggregates (LA), and an associated rise in the less active, low density small aggregates (SA), producing a decrease in the LA/SA ratio. To determine the effects of CPB on lung surfactant and lung mechanics we studied 12 children, ages 0.6 to 12 years, undergoing elective cardiac surgery for congenital heart disease. Pulmonary function testing, with deflation flow volume curves, was done to measure forced vital capacity (FVC) and maximum air flow at 25% lung volume (V_max25%). Bronchoalveolar lavage was then performed to assay surfactant aggregates and lavage cell counts pre-CPB, 1 hr and 6 hrs post-CPB. CPB duration was 112 + /-28 min. The results are shown below:

	LA/SA ratio	FVC (1/sec)	Vmax25%(%)	Cell Count (PMN's/ml)
PRE-CPB	6.8 + / - 1.0	1.4 + / - .3	70 + / - 9	7 + / - 5
1 hr POST	3.6 + / - 0.5*	1.0 + / - .2*	47 + / - 7*	35 + / - 11*
6 hrs POST#	1.5 + / - 0.3*	0.9 + / - .1*	52 + / - 7*	55 + / - 4*

(all values are mean + / - SE, * - p<0.05 vs pre-CPB, n = 12 except for # where n = 3)

Along with the increase in PMN's there was a significant decline in lung monocytes in the lavage fluid post-CPB at both time points. Lung compliance was also significantly decreased by 6 hrs post-CPB.

We conclude that in children, cardiopulmonary bypass of even moderate duration exerts a deleterious effect on surfactant activity with an associated decline in lung mechanics. The effects of CPB on surfactant activity may be of even more importance in neonates undergoing open heart surgery due to their limited ability to produce surfactant.

*By Invitation

32. Evaluation of Cerebral Metabolism and Quantitative EEC Following Hypothermic Circulatory Arrest and Low Flow Cardioplmonary Bypass

CRAIG K. MEZROW, M.S.*, PETER K. MIDULLA, M.D.*,

ALIM. SADEGHI, M.D.*, ALEJANDRO GANDSAS, M.D.*,

ROSARIO ZAPPULLA, M.D. *

and RANDALL B. GRIEPP, M.D.

New York, New York

Although widely used for repair of complex cardiovascular pathology, long intervals of hypothermic circulatory arrest (HCA) and low flow cardiopulmonary bypass (LFCPB) may both result in cerebral injury (CI). This study examines cerebral hemodynamics, metabolism, and electrical activity in order to evaluate the relative risks of CI after 60 min of HCA at 8 C, 13 C and 18 C, compared with 60 min LFCPB at 18 C.

Twenty-four puppies were randomly assigned to one of 4 experimental groups, and centrally cooled to the appropriate temperature. Serial evaluations of quantitative EEG (QEEG), radioactive microsphere determinations of cerebral blood flow (CBF), calculations of cerebral oxygen consumption (CMRO2), cerebral glucose consumption (CMRglu), cerebral vascular resistance (CVR), cerebral oxygen extraction, systemic oxygen metabolism and systemic vascular resistance (SVR) were carried out. Measurements were obtained at baseline (37 C), at the end of cooling, at 30 C after rewarming, and at 2, 4, and 8 hrs after HCA or LFCPB. A p<0.05 as determined by ANOVA was accepted as statistically significant.

At the end of cooling, CVR remained at baseline levels in all groups, but SVR was almost triple at 18 C, almost twice normal at 13 C, and 1 Vi baseline at 8 C. CMRO2 became progressively lower as temperature was reduced: it was only 5% of baseline at 8 C; 20% at 13 C; and 34% and 39% at 18 C. QEEG was silent in the 8 and 13 C groups, but significant slow wave activity was present at 18 C.

SVR and CMRO2 returned to baseline values in all groups by 2 hrs after HCA or LFCPB, but CVR remained elevated at 2 and 4 hrs, not returning to baseline until 8 hrs after HCA or LFCPB.

All of the long-term survivors (20/24) appeared neurologically normal, with one exception: after HCA at 8 C, one animal had an unsteady gait. Comparison of QEEG preoperatively and 6 days postoperatively showed a significant increase in slow wave activity and decrease in fast wave activity in all animal groups. These changes were more pronounced after HCA and LFCPB at 18 C.

Although undetected postoperatively by simple behavioral and neurological assessment, significant differences in cerebral metabolism, vasomotor responses and QEEG do exist during and following HCA and LFCPB at various temperatures, and may be implicated in the occurrence of CI. The data from this study suggest that for an interval of 60 minutes, HCA at 8 or 13 C may provide cerebral protection superior to HCA or LFCPB at 18 C.

*By Invitation

33. Energy Expenditure in Children With Congenital Heart Disease Before and After Cardiac Surgery

IAN M. MITCHELL, M.D.*, PETER S. W. DAVIES, Ph.D.*,

JANICE M.E. DAY*, JAMES C.S. POLLOCK, FRCS*

and MORGAN P.O. JAMIESON, FRCS*

Glasgow, Scotland and Cambridge, England

Sponsored by: Professor D.J. Wheatley, Glasgow, Scotland

Poor growth and failure to thrive are common features of children with congenital heart disease; some degree of nutritional impairment being evident, even in seemingly asymptomatic patients. Whether this is the result of a poor intake, or whether it is due to an abnormally high basal metabolic rate, is unknown, yet the state of nutrition has a profound effect on the metabolic response to injury and strongly influences the outcome from surgery. In nutritionally compromised children it is clearly important to recommence feeding at an early stage, but the exact energy requirements in the postoperative period are also unknown. The aim of this study was therefore to measure the pre- and postoperative energy requirements in children with congenital heart disease, to determine not only why growth should be poor, but also the calorie cost of cardiac surgery.

Seventeen children undergoing cardiac surgery were studied aged (mean age 15.8 months, range 4 - 33 months), cardiopulmonary bypass being required in 14. Each child was given two oral doses of doubly labelled water (H₂¹⁸O and ²H₂O), the first, one week prior to operation and the second, 6 hours after the end of surgery. By measuring the relative concentrations of each isotope in daily urine samples, carbon dioxide production and hence energy expenditure could be calculated. Preoperative results demonstrate that energy expenditure was essentially normal in 5 children, elevated in 8 and low in 4, suggesting that an elevated basal metabolic rate is an important factor in the observed failure to thrive. In the week following surgery however, total body water fell by approximately 5% and energy requirements by 36% (range 4% to 73%), reaching values below normal for healthy (non-operated) children, irrespective of whether or not cardiopulmonary bypass had been required. These results suggest that the stress of surgery leads to smaller energy requirements than have previously been thought.

2:45 p.m. INTERMISSION - VISIT EXHIBITS

*By Invitation

3:15 p.m. SIMULTANEOUS SCIENTIFIC SESSION C CONGENITAL HEART DISEASE

Grand Ballroom F

34. The Effect of the Hypoplastic Left Heart Class on Outcomes in Interrupted Aortic Arch

RICHARD A. JONAS, M.D., JAN M. QUAEGEBEUR, M.D.*,

GEORGE R. DAICOFF, M.D.,

EUGENE H. BLACKSTONE, M.D.

and JOHN W. KIRKLIN, M.D.

Boston, Massachusetts, New York, New York, St. Petersburg,

Florida and Birmingham, Alabama

Among 232 neonates with interrupted aortic arch (IAA) entering 30 institutions for treatment (1987-1992), 167 had coexisting VSD. Considering aortic atresia to be hypoplastic left heart class IV (HLH IV), 10 of the 167 pre-repair had HLH III (two additional important left heart obstructions, such as supravalvar, valvar, or subvalvar mitral narrowing, left ventricular hypoplasia, subvalvar, valvar, or annular aortic valve narrowing, or ascending aortic hypoplasia) and 37 had HLH II (one additional important left heart obstruction). The HLH assignment strongly (negatively) correlated with the pre-repair Z-values (echocardiography) of the diameter of the subvalvor area, the anulus, and the ascending aorta (for the 50th percentile, -6.3, -4.4, and -4.1, respectively). In some patients the subvalvar and annular Z-values decreased soon after repair.

Twenty-two different initial procedures were performed! One, 12, 24, and 36 month survival after the initial procedure was 75%, 66%, 65% and 64%. Risk factors (hazard function domain were identified (Table); the strength of some variables is illustrated in the nomograms. If Z-value of the aortic anulus replaces "HLH class" in the equation, its strength is shown.

Thus, when HLH II or III coexists with IAA and VSD, or when the Z-value of the anulus is ≤ about -4, an initial classical repair is reasonable, but important LV-aortic gradients may require later Konno type repair. But, when instead a DKS repair is performed initially, early survival is lessened even though the procedure may not be necessary (HLH I). The resolution of this dilemma requires continuation of the study and longterm follow-up, and re-analysis in 3-5 years.

*By Invitation

35. Unifocalization in Pulmonary Atresia With Ventricular Septal Defect and Major Aorto-Pulmonary Collateral Arteries

TOSHIKATSU YAGIHARA, M.D.*, FUMIO YAMANOTO, M.D.*, KYOUICHINISHIGAKI, M.D.*, OSAMUMATSUKI, M.D.*, HIDEKI UEMURA, M.D.* and YASUNARU KA WASHIMA, M.D.

Osaka, Japan

For the purpose of extending the indication of corrective surgery for patients with pulmonary atresia, ventricular septal defect and major aorto-pulmonary collateral arteries (MAPCA), we have applied the surgical procedures to unifocalize the pulmonary blood supply. Since December 1985, 48 patients underwent unifocalization at the age from 3 months to 26 years with an average of 5.6 years.

Eighty staged unifocalizations were performed in 47 patients, while one patient received one stage repair. Procedures included anastomosis between central pulmonary artery (CPA) and in-trapulmonary artery (IPA) supplied by MAPCA, directly or with interposition of graft (n = 32), pulmonary angioplasty (n = 7), creation of CPA with heterogenous pericardial roll (n = 33), and others (n = 8). In the patients group without CPA or with vestiginal small CPA and needed heterogenous pericardial roll, one end of the pericardial roll was anastomosed to IPA and another end was fixed at mediastinal pleura anteriorly with a shunt from the subclavian artery. Anastomosis was made mainly inside the lung dividing in-terlobular fissure. In addition to these procedures, central palliation which aids a small CPA to grow was performed in 3 patients. There were 5 operative and 5 late deaths out of these 83 procedures.

Twenty-five patients underwent intracardiac repair after the staged unifocalization. In 15 patients, confluency of surgically created CPA's were achieved as well as the reconstruction of right ventricular outflow tract using external conduit. Six patients are ready for correction. Right to left ventricular systolic pressure ratio immediately after intracardiac repair in 26 patients including one who underwent one stage repair ranged 0.24 to 0.91 with an average of 0.56 ± 0.18. There were 1 operative and 4 late deaths 2 months to 2 years after surgery, which related to bronchial hemorrhage or respiratory infection.

Surgical unifocalization is possibly a beneficial procedure as a part of correction for the patients with pulmonary atresia, ventricular septal defect and MAPCA, even for patients without CPA.

*By Invitation

36. Extended Aortic Valvuloplasty for Recurrent Aortic Valvar Stenosis and Regurgitation in Children

MICHEL N. ILBAWI, M.D., JOSEPH CASPI, M.D.*,

DAVID A. ROBERSON, M.D.*, R. ABDULLAH, M.D.*,

WILLIAM PICCIONE, M.D.* and

HASSAN NAJAFI, M.D.

Oak Lawn, Illinois

Recurrent significant aortic valvar stenosis and/or regurgitation (AVSR) following balloon or open valvotomy in pediatric patients, often requires aortic valve replacement (AYR). In an attempt to preserve the aortic valve, extended aortic valvuloplasty repair (EAV) was performed in 13 children with recurrent A VSR from 1/89 to 5/92. Previous related procedures were one or more open aortic valvotomy (n = 9), balloon valvotomy (n = 3), and repair of iatrogenic valve tear (n = 1). Mean age at the time of the EAV was 4 ± 1.4 yrs. Mean pressure gradient (MPG) across the aortic valve was 53 ± 12 torr. Regurgitation was moderate (Grade 2 to 3) in 7, and severe (Grade 4) in 6 pts. EAV techniques consisted of: thinning of valve leaflets (n = 7), augmentation of scarred retracted or torn leaflets using autologous pericardium (n = 5), release of rudimentary commissure from aortic wall (n = 2), resuspension of valve commissure (n = 8), extension of the valvotomy incision into the aortic wall on both sides of the commissure (n = 12), and patch repair of sinus of valsalva perforation (n= 1). There was no operative death or morbidity. Postoperative MPG assessed by most recent Doppler echocar-diography or cardiac catheterization at a follow-up of 24 ± 9 mos was 16±5mmHg (p<0.01 vs preoperative). Aortic regurgitation was absent in 9, mild in 2, and moderate to severe requiring subsequent AVR in 2. This short term experience indicates that EAV is a safe and effective surgical approach that minimizes the need for AVR in children with significant recurrent AVSR.

*By Invitation

37. Late Outcome Following Repair of Supravalvar Aortic Stenosis

JACQUES A. VAN SON, M.D.*,

GORDONK. DANIELSON, M.D., DWIGHT C. McGOON, M.D.,

HARTZELL V. SCHAFF, M.D., AMITA RASTOGI, M.D.*

and FRANCISCO J. PUG A, M.D.

Rochester, Minnesota

To determine the long-term outcome we reviewed 79 patients who had repair of localized (group A) (n = 67) or diffuse (group B) (n = 12) supravalvar aortic stenosis (SAS) from 1956 to 1992, including 30 patients with the Williams-Beuren syndrome. In group A the aortic root was enlarged with a diamond-shaped (dS) patch (n = 61) or a pantaloon-shaped (PS) patch (n = 6). In group B patch enlargement of the aorta was confined to the root (n = 4) or extended into the ascending aorta or aortic arch (n = 7); 1 patient had a left ventricular-aortic conduit. Two patients in group B in whom the patch enlargement was confined to the aortic root died intraoperatively (2.5%). During follow-up extending to 33 years there were 5 late deaths (2 related to coronary pathology) in group A and 1 in group B. There was no significant difference between patients with a DS or PS patch in terms of late gradient (mean ± SEM: 18 ± 2 mm Hg vs 9 ± 4 mm Hg, respectively) and aortic insufficiency (AI). By Cox multivariate model independent predictors of late death were concomitant aortic valvotomy (p = 0.04) and presence of preoperative AI (p= 0.006). In group B, survival was better in patients who received an extended patch vs aortic root patch (p=0.02). Risk factors for the development of late AI by univariate analysis were: bicuspid aortic valve (p = 0.02), valvar aortic stenosis (p = 0.01), absence of Williams-Beuren syndrome (p= 0.007), and concomitant aortic valvotomy (p = 0.006). On multivariate analysis the latter 2 factors were independent predictors of late AI (both p<0.01). Conclusions: 1. Both the DS and PS patch techniques provide excellent long-term relief of localized SAS; 2. In diffuse SAS aortic enlargement should be extended into the ascending aorta or beyond if necessary; 3. Concomitant aortic valvotomy may be associated with development of late AI as well as increased late death; 4. Early surgical intervention prior to development of significant AI or coronary artery disease may improve the long-term survival.

4:40 p.m. EXECUTIVE SESSION (Members Only)

7:00 p.m. ANNUAL DINNER/DANCE Black Tie (Optional)

Regency Ballroom

*By Invitation