Monday Morning, May 4, 1998

The American Association

for Thoracic Surgery

78TH ANNUAL MEETING

May 3-6, 1998

Hynes Convention Center

Boston, Massachusetts

MONDAY, MAY 4, 1998

8:00 a.m. BUSINESS SESSION (Limited to Members)

Ballroom, Hynes Convention Center

8:15 a.m. PLENARY SCIENTIFIC SESSION

Ballroom, Hynes Convention Center

Moderators: Floyd D. Loop, M.D.

James L. Cox, M.D.

1. THIRTY YEARS OF CARDIAC TRANSPLANTATION AT ONE INSTITUTION.

Robert C. Robbins, M.D.*, Clifford Barlow, M.D.*, Philip F. Oyer, M.D., Sharon A. Hunt, M.D.*, Bruce A. Reitz, M.D., Edward B. Stinson, M.D. and Norman E. Shumway, M.D.

Stanford, California

Discussant: Bartley P. Griffith, M.D., Pittsburgh, Pennsylvania

We reviewed the records of all 865 patients (pts) who have received a cardiac transplant (Tx) at our center. Patients were divided into 3 groups (Gps) based on immunosuppression (IMS) received: Gpl = no cyclosporin (CSA) (n = 200, 1/68-11/80), Gpl = CSA (n = 269, 12/80 - 5/86), Gp3 = CSA + OKT3 (n = 396, 6/86-10/97). Important data relating to the groups are as follows (mean ± SD):

	Gp1	Gp2	Gp3
Recipient age (years) (range)	38 ± 12 (12-55)	39 ± 14 (0.1-63)	42 ± 19 (0.1-70)
Donor age (years) (range)	22 ±6 (13-41)	23 ± 7 (1-53)	27 ± 12 (0.5-54)
Wait list time (days)	44 ± 35	49 ± 69	157 ± 195
Length hospital stay (days)	69 ± 36	36 ± 23	21 ± 34
Time to first rejection (days)	36 ± 55	63 ± 175	95 ± 259

Actuarial analyses of short and long-term outcomes for the Gps are as follows:

3 months

1 year

5 years

10 years

15 years

p value (ANOVA)

Survival (%)

Gp1

Gp2

Gp3

p < 0.05

Death from rejection (%)

Gp1

Gp2

Gp3

p < 0.05

Death from infection (%)

Gp1

Gp2

Gp3

p < 0.705

Death from lymphoid malignancy (%)

Gp1

Gp2

Gp3

Death from coronary artery disease (%)

Gp1

Gp2

Gp3

Retrospective HLA typing did not affect outcome. There are 141 pts who have survived > 10 years with the longest survivor being 23 years post Tx. There have been 65 re-Tx in 63 pts with 1, 5, and 10 year actuarial survivals of 51, 25, and 18 % respectively. Conclusions: The evolution of three decades of experience with cardiac Tx has resulted in an improved overall survival of pts who are currently at both extremes of age, wait longer for Tx, and have more complex medical conditions. The incidence of death from infection and rejection has decreased with time as a result of improved IMS and treatment. Continued advances in peri-operative management and IMS could further refine this initial Tx experience and improve the survival and quality of life of pts with end-stage heart failure.

*By invitation

2. RESULTS OF FIRST 100 CASES OF COMPLETE REPAIR OFCONGENITAL HEART DEFECTS IN PATIENTS WEIGHING700 TO 2500 GRAMS.

V. Mohan Reddy, M.D.*, Doff B. McElhinney, M.S.*, Theresa Sagrado, B.A.*, Andrew J. Parry, M.D.*, David F. Teitel, M.D.*, Norman J. Silverman, M.D.* and Frank L. Hanley, M.D.

San Francisco, California

Discussant: Roger B.B. Mee, M.B., Ch.B., Cleveland, Ohio

Infants born with congenital heart disease are more likely than normal children to be small for gestational age at birth. Published data suggest that low birth weight is a risk factor for corrective surgery for many cardiac defects. Congenital heart defects in this patient population are typically managed with supportive therapy or palliative surgery, and definitive repair is delayed. However, the morbidity of such an approach has been shown to be high. Since 1990, complete repair of congenital heart defects (other than isolated patent ductus arteriosus) has been performed in 102 infants ≤ 2500 g (median 2100 g, range 700-2500 g), including 16 who weighed ≤ 1500 g. Defects included ventricular septal defect with other left to right shunt lesions (23), tetralogy of Fallot (19), transposition complexes (13), coarctation of the aorta (12), interrupted aortic arch (9), truncus arteriosus (8), atrioventricular septal defect (6), total anomalous pulmonary venous return (4), and other defects (8). Preoperative morbidity was more common in patients who were referred late for corrective surgery. Standard techniques of neonatal cardiopulmonary bypass were used, including circulatory arrest in 17 pts. There were 10 early deaths (10%), due to cardiac failure (4), arrhythmia (1), sepsis (1), idiopathic coronary artery intimal necrosis (1), foot gangrene (1), pulmonary hemorrhage (1), and technical error (1). No patient had evidence of post-bypass intracerebral hemorrhage. At a median follow-up of 36 months, there had been 6 late deaths and 7 patients had undergone surgical and/or catheter reintervention for subaortic stenosis (2), conduit obstruction (2), pulmonary artery stenosis (2), or recurrent coarctation (1). Among the first 50 patents, followed for > 3 yrs, there was no evidence of neurological sequelae. Median weight for age was at the 20th percentile, with a direct correlation between weight for age and birth weight (p = 0.01). In most cases, delay in repair of congenital heart defects in low and very low birth weight infants does not confer any benefit and is associated with a higher incidence of preoperative morbidity. Complete repair of both simple and complex congenital heart lesions can be performed successfully in low and very low birth weight infants with good early and medium term results. Postoperative growth is accelerated following repair and approximates the normal growth curve for low birth weight infants without congenital heart disease. It is recommended that such infants, especially when symptomatic, undergo early surgical correction rather than prolonged medical management or other forms of palliation.

*By invitation

3. CLINICAL EXPERIENCE WITH CARINAL RESECTION.

John D. Mitchell, M.D.*, Douglas J. Mathisen, M.D. Dean M. Donahue, M.D.*, Ashby C. Moncure, M.D. John C. Wain, M.D.*, Cameron D. Wright, M.D. and Hermes C. Grillo, M.D.

San Diego, California and Boston, Massachusetts

Discussant: Jean DesLauriers, M.D., Sainte-Foy, Quebec, Canada

Pathologic processes that involve the carina pose a tremendous challenge to thoracic surgeons. Techniques have been developed to allow primary resection and reconstruction. The procedures are demanding, management of patients complex, and the potential for complications high. Few institutions have accumulated sufficient experience to allow meaningful conclusions about indications and contraindications for surgery, morbidity and mortality rates. Since 1962 135 patients have undergone 143 carinal resections (134 primary resection, 9 re-resection) at our institution. Indications for carinal resection included bronchogenic cancer (58), other airway neoplasms (60), benign or inflammatory strictures (16). Thirty-seven patients had prior lung or airway surgery not involving the carina. Carinal resection without pulmonary resection was accomplished in 52 patients; 57 patients had carinal pneumonectomy (44 right, 13 left); 14 patients had carinal plus lobar resection; and 11 patients had carinal resection following prior pneumonectomy (9 left, 2 right). There were 15 different combinations of reconstruction. Techniques were employed to reduce anastomotic tension. The overall mortality in the 134 patients for primary carinal resection was 12.7% (17/134). Adult respiratory distress syndrome was responsible for 9 early deaths. Significant multivariate predictors of postoperative death included length of resected airway, development of anastomotic complications, and postoperative mechanical ventilation. Complications occurred in 35 patients (26%) including atrial arrhythmias (17), anastomotic complications (9) end pneumonia (7). Mortality by procedure and indication for surgery was as follows.

	Right	Left	Carina	Carina	Prior	Carinal	Lung	Other	Benign
	Carinal	Carinal	alone	& Lobe	Pneu-	Re-resect.	Cancer	Neoplasms	Strictures
	Pneumon.	Pneumon.			Monect.
N	44(7)	13(4)	52(4)	14(1)	11(1)	9(1)	58(9)	60(5)	16(3)
Mortality	15.9%	30.8%	7.7%	7.1%	9.1%	11.1%	15.5%	8.3%	18.7%

Left carinal pneumonectomy is associated with high operative mortality and consideration of the underlying pathologic process and chance for long-term survival must be carefully considered before recommending such a procedure. Carinal resection with preservation of lung and for low grade neoplasms is associated with acceptable mortality rates. Chance for long-term survival must be carefully balanced against operative mortality in recommending carinal resection for lung cancer.

*By invitation

4. PRIMARY AORTIC VALVE REPLACEMENT WITH HOMOGRAFTS OVER 25 YEARS: VALVE AND PROCEDURE RELATED DETERMINANTS OF OUTCOME.(106)

Ole Lund, M.D., Ph.D.*, Magdi Yacoub, F.R.C.S., DSc., V. Chandrasekaran, F.R.C.S.*, Richard Grocott-Mason, M.R.C.P., M.D.*, Hassan Elwidaa, M.R.C.P.* and Rashid Mazhar, F.R.C.S.*

London, England

Discussant: Mark F. O'Brien, M.D., Brisbane, Australia

Homografts offer many advantages over prosthetic valves, however, homograft durability varies considerably.

From 1969 through 1993, 618 patients aged 15-84 (mean 51) years underwent their first AYR with an aortic homograft. Concomitant surgery included root tailoring (N = 64), replacement of the ascending aorta (N = 56), coronary bypass grafting (N = 87). Homograft implantation was done using a freehand' subcoronary technique (N = 551) or total root replacement (N = 67). The homografts were antibiotic sterilized (N = 479), cryo-preserved (N=12), or homovitals' (nutrient medium, inserted within 72 h; N = 127). Maximum follow-up was 27.1 (mean 10.1) years.

Thirty-day mortality was 5.0% and 10- and 20-year crude survivals ± standard error (SE) 67 ± 2% and 35 ± 3%, respectively. Ten- and 20-year complication freedoms ± SE included: endocarditis 93 ± 1% and 89 ± 2%, respectively; primary tissue failure, 62 ± 3% and 18 ±3%, respectively; and re-do AYR, 81 ± 2% and 35 ± 4%, respectively. Multivariate COX analyses identified several valve and procedure related determinants which included: rising homograft donor age and antibiotic sterilized/cryopreserved homograft for mortality; donor > 10 years older than patient for endocarditis; rising donor minus patient age, rising implantation (from harvest to AYR) time, and donor age > 65 years for tissue failure; and rising donor minus patients age, young patient age, rising implantation time, subcoronary implantation, and aortic root tailoring for re-do AYR. Estimated 10- and 20-year freedoms from tissue failure for a 60-year-old patient with a homovital' and no other risk factors were 88% and 52%, respectively, for a donor age of 30 years and 72% and 21%, respectively, for a donor of 60 years; the freedoms for a 30-year-old patient were 78% and 31%, respectively, for a 30-year-old donor and 56% and 7%, respectively, for a 60-year-old donor. Beneficial influences of a homovital' were largely covered by short harvest time (0 for homografts from brain dead organ donors or heart transplant recipients) and short implantation time.

It is concluded that primary homograft AYR can give acceptable results for up to 25 years and that the late results can be improved by the use of a homovital' valve, by matching patient and donor age, and by more liberal use of a free root replacement with re-implantation of the coronary arteries rather than tailoring the root to accomodate a subcoronary implantation.

*By invitation

9:40 a.m. EVARTS A. GRAHAM MEMORIAL TRAVELING FELLOW PRESENTATION

Jun Wang, M.D. Beijing, People's Republic of China

9:45 a.m. INTERMISSION - VISIT EXHIBITS

MONDAY, MAY 4, 1998

10:30 a.m. PLENARY SCIENTIFIC SESSION

Ballroom, Hynes Convention Center

Moderators: Lawrence H. Cohn, M.D.

James L. Cox, M.D.

5. THE FIRST GENERATION OF ENDOVASCULAR STENT-GRAFTS FOR DESCENDING THORACIC ANEURYSMS.

D. Craig Miller, M.D., R. Scott Mitchell, M.D.*, James I. Fann, M.D.* and Michael D. Dake, M.D.*

Stanford, California

Discussant: Randall B. Griepp, M.D, New York, New York

From July 1992 to October 1997, 109 patients (mean age 68 years [range 34-89 years]) underwent endovascular stent-grafting of descending thoracic aortic disorders using a custom fabricated, self-expanding device. Follow-up was 100% complete, and averaged 21 months. Eighty-five (78%) patients had atherosclerotic aneurysms, 8 were due to aortic dissection, 8 post-traumatic, 5 anastomotic, 2 mycotic (healed), and 1 was degenerative. Twelve patients presented with aortic rupture. Fourteen (13%) patients had undergone one or more previous thoracic aortic surgical procedures. Over one-half of patients had been judged not to be reasonable candidates for "open" conventional surgical repair. Access was from the femoral artery in 60% of patients, abdominal aorta in 31%, iliac artery in 7%, and the aortic arch in 2%. Twenty-four patients underwent simultaneous abdominal aortic aneurysm repair. Complete thrombosis of the aneurysm was achieved in 104 (95%) patients (14 required post-procedure stent/graft extensions or coil embolization) prior to discharge. One patient died intraoperatively of aneursym perforation. Early mortality rate was 8 ± 3% (± 70% CL). The only multivariate independent risk factor for early death was smaller aortic diameter. Early complications included paraplegia in 4%, stroke in 3%, iliac artery avulsion in 2%, proximal aortic dissection in 2%, myocardial infarction in 1%, colon ischemia in 1%, and renal failure in 1%. Acturial survival estimates at 6 months, 1 year, and 5 years were 88%, 84%, and 77 ± 5% (± 1 SE), respectively. According to the intent to treat principle, "Treatment Failure" (conservatively defined as all late sudden, unexplained deaths, deaths due to aneurysm rupture, need for subsequent stent-grafting or operation, and endoleak) occurred in 18 patients (or 16%). Actuarial estimates of freedom from Treatment Failure at 1 and 5 years were 84% and 71 ± 8%, respectively. Emergency stent-grafting was the only independent risk factor for Treatment Failure. Three (3%) patients subsequently required definitive (open) surgical repair after 4, 43, and 61 months.

Conclusions: This 5 year clinical trial employing a relatively primitive stent/graft device indicates that endovascular descending thoracic aortic stent-grafting is feasible with acceptable medium-term results, which are comparable to the outcome after open operation; however, additional follow-up is still required to assess fully the long-term efficacy of this technique. The more refined, commercially-developed devices used today offer less traumatic and more precise stent/graft deployment; these major technical advantages, coupled with the important lessons we have learned over time, should be associated with even more salutary long-term clinical results in the future.

*By invitation

6. TWO INTERNAL THORACIC ARTERY GRAFTS ARE BETTER THAN ONE.

Bruce W. Lytle, M.D., John H. Arnold, M.D.*, Floyd D. Loop, M.D., Penny Whiteman, M.S.*, Robert W. Stewart, M.D., Patrick M. McCarthy, M.D. and Delos M. Cosgrove, M.D.

Cleveland, Ohio

Discussant: Hendrick B. Barnes, M.D., St. Louis, Missouri

Do bilateral ITA (BITA) grafts produce better outcomes than a single ITA graft (SITA)? To examine this issue for patients undergoing primary coronary artery bypass grafting, we reviewed the surgical strategies and outcomes for 2015 consecutive patients receiving BITA grafting with or without additional vein grafts and, for comparison, 8059 patients receiving SITA grafts and at least one vein graft. To achieve long follow-up intervals, patients were selected for review who underwent operation from 1971 through 1989. Treatment selection was not randomized and the BITA group had a lower mean age, included more men, fewer patients with diabetes, and more patients with triple vessel disease (all p < 0.01). The in-hospital mortality rate was 0.7% for both groups. Late follow-up (mean postoperative interval 10.3 years after BITA, 304 BITA patients with > 12 year follow-up) documented 51 reoperations and 177 percutaneous interventions (PTCA) after BITA grafting. Survival for the BITA group was 94%, 84%, and 67% and for the SITA group 91%, 79% and 64% at 5, 10 and 15 postoperative years, respectively (p < 0.001). Cox regression analyses were used to test patient and treatment related variables in multi-variate models for their associations with late outcomes and to establish the increases in relative risks associated with those variables. The table shows tests of significance and the adjusted risk ratios (p value [relative risk ratio]) for the patients in the SITA group relative to the BITA patients according to age group and outcomes. For all age groups, the risks of death and reoperation were higher for patients in the SITA group, except for the risk of death for patients < 50 years old.

	Late Death	Reoperation	Late Death or Reoperation	Late Death, Reoperation, or PTCA
All patients	<0.001 (1.31)	<0.001 (3.74)	<0.001 (1.61)	<0.001 (1.51)
< 50 years	N/S (1.05)	<0.001 (6.78)	<0.001 (1.77)	<0.001 (1.55)
50-60 years	0.002 (1.41)	<0.001 (3.58)	<0.001 (1.77)	<0.001 (1.56)
> 60 years	<0.001 (1.36)	0.001 (2.33)	<0.001 (1.44)	<0.001 (1.44)

Neither total arterial revascularization nor the specific vessels grafted with BITA improved results when compared to the general strategy of BITA grafting.

These data suggest that BITA grafting improved the survival rate over that for

SITA grafting in all age groups except for those < 50 years old. Futhermore, they provide strong evidence that BITA grafting substantially decreases the risk of reoperation for all age groups.

11:15 a.m. PRESIDENTIAL ADDRESS

"...The First Living and the Last Dying"

Floyd D. Loop, M.D., Cleveland, Ohio

12:00 p.m. ADJOURN FOR LUNCH - VISIT EXHIBITS

*By invitation