AATS: Tuesday Afternoon, April 20,1999

Tuesday Afternoon, April 20,1999

Back to Annual Meeting Program

TUESDAY AFTERNOON, APRIL 20, 1999

1:45 p.m. SIMULTANEOUS SCIENTIFIC SESSION A-2 -

ADULT CARDIAC SURGERY

Ballroom, Ernest N. Morial Convention Center

Moderators: Bruce W. Lytle, M.D.

D. Glenn Pennington, M.D.

37. The Impact of the Maze Procedure on the Incidence of Stroke Due to Atrial Fibrillation

James L. Cox, Niv Ad*, Washington, DC

Discussant: Hartzell V. Schaff, M.D., Rochester, MN

OBJECTIVE: The incidence of stroke in non-anticoagulated patients with atrial fibrillation is 1-12 % per year, depending on associated risk factors. Warfarin therapy dramatically decreases the incidence of stroke but despite its effectiveness, the number of strokes due to atrial fibrillation in the U.S.A. remains at approximately 75,000 per year. This study evaluates the effect of abolishing atrial fibrillation with the Maze procedure on the subsequent incidence of stroke in both low-risk and high-risk patients.

METHODS: Between September, 1987 and September, 1998, we performed the Maze procedure on 275 patients with intermittent or chronic atrial fibrillation. Fifty-two (19 %) of those patients (Group I) had suffered at least one previous stroke, TLA, or systemic embolic episode directly attributable to atrial fibrillation prior to surgery. Two hundred twenty-three patients (Group II) (81%) had not experienced any evidence of thromboembolism preoperatively. These groups were compared to the expected rate of stroke in anticoagulated patients with atrial fibrillation.

RESULTS: During the follow-up period of up to 11 years (mean: 50.1+34.1 months), statistics would predict that even with appropriate warfarin therapy in all patients, 24 patients would have had a thromboembolic stroke. However, there were no thromboembolic strokes in either Group I or Group II patients undergoing the Maze procedure during the follow-up period despite the lack of anticoagulation postoperatively.

CONCLUSIONS: The Maze procedure abolishes the risk of stroke associated with atrial fibrillation.

*By Invitation

38. T-Grafts With Bilateral ITA Versus Left ITA And Radial Artery: Flow Dynamics In The ITA Mainstem

Olaf Wendler*, Benno Hennen*, Torsten Markwirth*, Dietmar Tscholl*, Qi Huang*, Erfane Shahangi* and Hans-Joachim Schsfers*, Homburg Saar, Germany

Sponsored by: Hans G. Borst, M.D., Munich, Germany

Discussant: Alfred J. Tector, M.D., Milwaukee, WI

OBJECTIVE: Complete arterial coronary artery bypass grafting (CABG) with two grafts may be attained in triple vessel disease when a T-configuration is employed. There is still scepticism wether the coronary flow reserve (CFR) in the ITA-mainstem is sufficient to supply more than one anastomosed coronary vessel.

METHODS: Between 3/96 and 9/98 118 patients (pts) (102 male; mean age 59 years) with triple vessel disease received complete arterial CABG with T-grafts. In 57 bilateral skeletonized ITAs (group I) and in 61 pts left skeletonized ITA and RA (group II) were used as conduits. A mean of 4.04 (3-6) (I) versus 4.33 (3-6) (II) coronary vessels were anastomosed per patient. One week postop-eratively resting flow in 32 (I) and 30 pts (II) was measured in the proximal ITA using a doppler guide wire (Cardiometrics FloWire; 0.014 inches). Maximum flow was determined after injection of adenosin (30 mg/kg). Six months later the investigation was repeated in 12 (I) and 9 pts (II).

RESULTS: The in-hospital mortalities were 3.5%(I) versus 0% (II). No case of bleeding, sternal wound infection or dehiscence occurred. On angiography 94.6% (I) versus 96.4% (II) of vessels were patent. ITA-mainstem flow after stimulation with adenosin increased significantly in all pts (p≤0,0001). There was no significant difference between base-line flow, maximum flow and CFR in the ITA-mainstem between the two groups. 1 week: (I) 78.3±33.8 / 136.0+54.8 / 1.81±0.30 (II) 64.0±30.3 / 115.1±49.3 / 1.87±0.34 6 months: (I) 42.9±15.2 / 118.9±37.3 / 2.89±0.82 (II) 67 .6±40.5 / 142.9±80.3 / 2.49±0.51 (flow data in ml/min±sd) CFR, however, increased significantly in both groups after the first six postoperative months (p≤0,0001).

CONCLUSIONS: Bilateral ITA or left ITA and RA as T-grafts produce complete arterial revascularization with good perioperative results. Using the T-graft configuration the CFR of ITA-mainstem is identical with puplished results after single CABG with ITA. Therefore revascularization with T-grafts using two arteries (ITA or RA) results in adequate coronary perfusion and portends good long-time prognosis of arterial revascularization.

*By Invitation

39. Isolated LITA to LAD: Late Consequences of Incomplete Revascularization

Roslyn Scott*, Eugene H. Blackstone, Patrick M. McCarthy, Bruce W. Lytle, Floyd D. Loop, Jennifer White* and Delos M. Cosgrove, Cleveland, Ohio

Discussant: Antonio Maria Calafiore, M.D., Cheiti, Italy

OBJECTIVE: Early and long-term outcome of isolated left internal thoracic artery (LITA) to the left anterior descending coronary artery (LAD) is the benchmark for minimally invasive and percutaneous intervention. However, changing indications for revascularization according to extent of disease over the years presents the opportunity to assess the impact of existing residual non-LAD system stenoses at the time of LITA-LAD.

METHODS: 2072 pts underwent primary isolated LITA-LAD 1971-1997. Using a 50% stenosis criterion, 26% and 13% would nowadays be considered 2-or 3-system disease, respectively. Incomplete revascularization fell from 60% in the early 1970s to 20-25% from the 1980s onward. Multivariable analysis of detailed coronary stenosis variables was conducted in the hazard-function domain for the 24,300 patient-years of follow-up (mean 12±6.0).

RESULTS: Survival was 99%, 86%, and 57% at 1,10, and 20 yrs. In addition to expected risk factors, presence of a 70% or greater lesion in the circumflex (LCx), but not in the right coronary (RCA) systems, was a late-phase risk factor (70% and 38% survival at 10 and 20 yrs with LCx disease vs. 89% and 62% without, P<.0001). Any left main disease unfavorably affected survival (P=.003). In contrast, residual stenoses, particularly in the RCA (P=.001), were unrelated to survival and only weakly predictive of later reintervention (60% free at 20 yrs).

CONCLUSIONS: Residual stenoses in non-LAD systems after LITA-LAD are poorly predictive of late reintervention, but residual LCx or left main disease importantly reduces late survival.

*By Invitation

40. Prospective Randomized Comparative Study of Brain Protection in Total Aortic Arch Replacement: Deep Hypothermic Circulatory Arrest with Retrograde Cerebral Perfusion or Selective Antegrade Cerebral Perfusion

Yutaka Okita*, Kenji Minatoya*, Osamu Tagusari*, Motomi Ando* Kazuyuki Nagatsuka* and Soichiro Kitamura, Osaka, Japan

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

OBJECTIVE: Comparing results of total aortic arch replacement with two different methods of brain protection, particularly in view of neurological outcomes.

METHODS: From June 1997 to September 1998, 47 consecutive patients who underwent total arch replacement through a midsternotomy were randomly allocated based upon the two methods of brain protection: deep hypothermic circulatory arrest with retrograde cerebral perfusion (RCP: 22 patients) and selective antegrade cerebral perfusion (SCP: 25 patients). Patients who had acute aortic dissection or ruptured aneurysm were excluded. Mean ages were 69±8 years. Pre- and postoperative (3 weeks) brain CT scan, neurological examination, and cognitive function tests were performed. Serum 100b protein was assayed before and after cardiopulmonary bypass, 24 hours, and 48 hours after the operation. The whole bypass time was shorter in the RCP (176±53 vs 218±89 min. p=0.05). Duration of the RCP was 46±15 min. and SCP was 116+42 min. No differences existed in patients' demography and other details of surgery.

RESULTS: Hospital mortality was noted in 2 patients only in the SCP (8.0%, p=0.18). New strokes occurred in 1(4.5%) in the RCP and in 1 (4.0%) in the SCP (p=0.93). The incidence of transient brain dysfunction was significantly higher in the RCP (8,36.3% vs 2,8.0%, p=0.02). Excluding the patients with strokes, S-100b values were identical in two groups, (RCP:SCP, pre-bypass 0.01±0.03: 0.002±0.01, post-bypass 1.90±0.97:1.96±1.00, 24 hours 0.67±0.58: 0.59+0.26, 48 hours 0.39±0.30: 0.31±0.13 mg/L, respectively. p=0.69). Postoperative wake-up time (RCP 6.1±3.3; SCP 5.2±2.4 hours, p=0.32) and extubation time (RCP 13.4±7.5; SCP 12.3±5.7 hours, p=0.60) were equal. Declined scores of the memory (RCP 0.74±0.99; SCP 0.55±1.19, p=0.59), orientation (RCP 1.11±1.29; SCP 0.50+.0.76, p=0.08), and intellectual function (RCP 1.21±1.27; SCP 1.05±1.15, p=0.68) showed no difference. Postoperative CT showed no abnormalities in either group. Stay in the ICU (RCP 3±1; SCP 4±3 days, p=0.15) and in the hospital (RCP 34±21; SCP 29±10 days, p=0.36) was equivalent.

CONCLUSIONS: Both methods of brain protection for patients undergoing total arch replacement provided acceptable mortality and morbidity. However, the incidence of transient brain dysfunction was significantly higher in patients with the RCP.

3:05 p.m. INTERMISSION

*By Invitation

3:50 p.m. SIMULTANEOUS SCIENTIFIC SESSION A-2 - ADULT CARDIAC SURGERY

Ballroom, Ernest N. Morial Convention Center

Moderators: Bruce W. Lytle, M.D.

D. Glenn Pennington, M.D.

41. Competing Risks after Bypass Surgery: The Influence of Death on Reintervention

Eugene H. Blackstone and Bruce W. Lytle, Cleveland, Ohio

Discussant: Robert H. Jones, M.D., Durham, NC

OBJECTIVE: Extensive arterial grafting lowers the incidence of reintervention (REINT), but is being performed in older, higher risk patients. Is reduced REINT real or simply the passive result of dying before needing REINT?

METHODS: Multivariable competing risks analysis was performed of 2001 patients undergoing CABG using bilateral internal thoracic artery (ITA) conduits (BITA) and 8123 receiving single ITA conduits (SITA) for the events death before REINT, REINT by angioplasty (PTCA), and redo CABG. Mean follow-up was 9.7±3.0 yrs and 10.8±5.2 yrs for the BITA and SITA groups, respectively.

RESULTS: BITA provided better survival (difference of 5% at 10 yrs, P<.0001) and fewer REINT (difference of 5% in redo CABG, P<.0001, but no difference in PTCA), while older age was associated with poorer survival (P<.0001) and fewer REINT (P<.0001). Death impacted estimates of REINT prevalence more often in SITA than BITA because of the simultaneous effects of decreased mortality and REINT.

% in Categories at 12 Years

	Age 35		Age 70
Category	SITA	BITA	SITA	BITA
Alive, no REINT	23	65	37	56
Dead before REINT	8	8	35	28
PTCA	43	24	20	12
Redo CABG	26	3	8	3

CONCLUSIONS: Across all ages, after accounting for death's confounding influence, more extensive arterial grafting was associated with fewer REINTs. However, at older ages, its influence on redo CABG narrows considerably.

*By Invitation

42. Clinical Benefits of Endoscopic Vein Harvesting in Coronary Artery Bypass Patients With Risk Factors for Saphenectomy Wound Infections.

Phillip A. Carpino*, Kamal R. Khabbaz*, Robert M. Bojar*, Hassan Rastegar*, Kenneth G. Warner*, Richard E. Murphy* and Douglas D. Payne*, Boston, Massachusetts

Sponsored By: Benedict D. T. Daly, Boston, Massachusetts

Discussant: Robert J. March, M.D., Chicago, IL

OBJECTIVE: The impact of the use of endoscopic techniques on the incidence of complications in the saphenectomy incision after coronary artey bypass surgery (CABG) is not defined for patients with higher risks for developing wound infections (WI).

METHODS: In 1473 CABG patients who had the saphenous vein harvested by either a continuous incision or skip incisions leaving intact skin bridges, the incidence of WI was 9.6%. The following variables were entered into a logistical regression analysis to identify the significant risk factors that are predictors of WI: Diabetes (DM), peripheral vascular disease, obesity, renal failure, steroid use, age, gender, and type of closure. We then randomized prospectively 132 patients found to be at high risk of WI to either endoscopic vein harvesting approach (ENDO) or continuous open incision (OPEN).

RESULTS: A univariate analysis showed female gender (p=0.0314), DM (p=0.002), and obesity (p=0.002) to be predictors of WI. In a mulrivariate model, only DM (p=0.022) and obesity 9p=0.025) were independent predictors. The incidence of wound infection in the high-risk group was 4.5% for the ENDO group vs. 20% for the OPEN group (p=0.002). However, the vein procurment time was longer in the ENDO group (65 min vs. 32 min.) and so was the number of vein repairs required (2.5 vs. 0.8).

CONCLUSIONS: The use of endoscopic vein harvesting decreases the incidence of post operative saphenectomy infections in patients with DM and or Obesity, independent predictors of that problem. Wether this translates into an economic benifit that justifies the additional cost of theat technology requires further complex analysis.

*By Invitation

43. Systematic Off-Pump Coronary Artery Revascularization in Multi-Vessel Disease: Experience of 230 Cases

Raymond Carrier*, Stacey Brann*, Francois Dagenais*, Raymond Martineau* and Yves Leclerc*, Montreal, Quebec, Canada

Sponsored By: Michel Carrier, Montreal, Quebec, Canada

Discussant: Stephen E. Colvin, M.D., New York, NY

OBJECTIVE: We report our recent experience with systematic off-pump coronary artery revascularization for multi-vessel disease.

METHODS: Between September 1996 and June 1998, 230 off-pump revascularization representing 80% of all revascularizations done during this time frame and 95% since January 1998 were performed by a single surgeon (RC) at the Montreal Heart Institute. There were 184 men and 46 women averaging 63.2±10.9 years old. Sixteen (7%) procedures were reoperative surgery. Main indication was unstable angina (61%). An average of 2.88±0.6 (1-5) grafts/ patient was performed, majority (70%) being triple and quadruple bypasses.

RESULTS: Complete revascularization was achieved in 91 % of the patient and single or double internal thoracic artery, saphenous vein, radial, and gastro-epiploic arteries were employed in respectively 95, 84, 10 and 1% of the patients. Coronary artery mechanical stabilization (Coroneo Corvasc. patent pending) and heart ventricalizing technique were used to reach circumflex area. Average total ischemic time was 29.8±0.9 (8-65) min. 68% of the patient did not require transfusion. Four percent of the patients were reexplored for bleeding, 2.5% experienced transmural myocardial infarction and only one required postoperative aortic counterpulsation assistance. There were two operative death, one due to multi-organ failure non-cardiac related and one from malignant arrhythmia already present prior to the surgery. Three patients experimented early recurrent angina and two among them had negative investigation. Early angiograms performed on the first 12 patients confirmed 100% patency with excellent run-off (95%).

CONCLUSIONS: With adequate surgical technique, complete coronary revascularization can be achieved without extracorporeal circulation in a majority of patients with excellent result and low morbidity and mortality.

4:50 p.m. EXECUTIVE SESSION (MEMBERS ONLY) - Ballroom

6:30 p.m. MAGIC OF MARDI GRAS RECEPTION

MARDI GRAS WORLD

*By Invitation

1:45 p.m. SIMULTANEOUS SCIENTIFIC SESSION B-2 - GENERAL THORACIC SURGERY

Room 211-213, Ernest N. Morial Convention Center

Moderators: David J. Sugarbaker, M.D.

Carolyn E. Reed, M.D.

44. Induction Chemotherapy Prior to Surgery for Early Stage Lung Cancer - A Novel Approach

Katherine M. W. Pisters*, Robert J. Ginsberg and Paul A. Bunn*, Houston, Texas, New York, New York and Denver, Colorado

Discussant: Mark K. Ferguson, M.D., Chicago, IL

OBJECTIVE: Patients (Pts) with clinical stage Ib, II and T3 N1 non-small cell lung cancer(NSCLC) have a poor survival following with surgery-less than 50% are cured. Adjuvant treatment has had little impact. Induction chemotherapy(CT) for N2 disease improves long-term survival. Newer CT has proven tolerable, user-friendly, and effective in advanced NSCLC. A phase II trial assessed the feasibility-measured by response rate, toxicity, resectabil-ity rate, and surgical morbidity and mortality-of perioperative paclitaxel and carboplatin in pts with early stage NSCLC-proven by mediastinoscopy and imaging studies.

METHODS: Pts with T1 N0 or superior sulcus tumors were excluded. All pts required adequate medical parameters to undergo induction CT and surgery. CT consisted of paclitaxel:225mgm/M²3hr infusion, and carboplatin:AUC=6 every 21 days for 2 cycles prior to Surgery. Three further cycles of CT were given to RO pts. Rl,2 pts were deemed off-study.

RESULTS: Between 06/97 and 07/98, 94 pts were entered, M/F = 65/29, median age 64yrs. All pts have completed therapy. Major (CR&PR) responses occurred in 50 of 92 pts (54%) eligible for surgery. 9 pts were deemed off-study prior to surgery because of : progression(3), CT reaction(2), death (1), MI(1), lost to follow-up(1), unresectable(1). Of 83pts (90%) explored, 75(82%) were completely resected. Two postop deaths have occurred. Four(4%)pathologic CRs have been observed. There was no increased or unexpected toxicity or surgical morbidity.

CONCLUSIONS: Induction CT is feasible and paclitaxel/carboplatin has a high response rate without morbidity in early stage NSCLC. These results have stimulated development of a randomized intergroup trial comparing induction CT and surgery to surgery alone in early stage lung cancer.

* For the Bimodality Lung Oncology Group (BLOT)

*By Invitation

45. Pulmonary Hemodynamics Contribute to Indicate Priority for Lung Transplantation in Patients with Cystic Fibrosis

Federico Venuta*, Erino A Rendina*, Giorgio Delia Rocca*, Tiziano De Giacomo*, Francesco Pugliese*, Anna Maria Ciccone* and Giorgio F. Coloni*, Rome, Italy

Sponsored by: G. Alexander Patterson, M.D., St. Louis, Missouri

Discussant: Robert Duane Davis, M.D., Durham, NC

OBJECTIVE: Lung transplantation is a viable therapeutic option for patients with cystic fibrosis (CF). Timing of referral and priority for transplantation are crucial to improve results and minimize mortality on the waiting list. The current strategy, based on pulmonary function tests and deterioration of quality of life, results in a high waiting list mortality. We reviewed the CF population accepted for lung transplantation in our program to ascertain if pulmonary hemo-dynamics could contribute to enhance referral and priority in the waiting list.

METHODS: Forty - two CF patients were accepted: 22 were transplanted, 10 died in the waiting list and 10 are still waiting. At the time of evaluation we recorded FEV₁, FVC, FEF_25-75, supplemental O₂, PaO₂/FiO₂, PaCO₂, 6-minute walking test, Right Ventricular Ejection Fraction (RVEF) and pulmonary hemodynamics with and without inhaled nitric oxide. We also recoreded the age at the time of diagnosis, gender, body weight and Schachman score. We compared the data from patients dying on the waiting list (Group I) with patients undergoing lung transplantation (Group II). A comparison was also made within Group II between the data collected at the time of evaluation and at the time of transplantation to quanitfy the functional deterioration during the waiting time.

RESULTS: Mean waiting time for Groups I and II was respectively 121 (1-281) and 112 (28-238) days. Age at time of diagnosis, gender, weight, Schachman score, spirometry, 6-minute walking test, RVEF and response to inhaled nitric oxide did not differ between Group I and II. A statistically significant was found for PaO₂/FiO₂ (191±54 mmHg in Group I vs 274±63 mmHg in Group II), PaCO₂ (64±23 vs 45±5, mmHg), mean pulmonary artery pressure (35±12 vs 23±6, mmHg) cardiac index (4.6±1 vs 3.5±0.6, LΣminΣm²), pulmonary wedge pressure (6.6±2.4 vs 3±2, mmHg) and intrapulmonary shunt (31±7 vs 23±3m %). The comparison within Group II showed a significative deterioration of pulmonary hemodynamics during the waiting time.

CONCLUSIONS: We conclude that pulmonary hemodynamics should contribute to indicate priority for lung transplantation in patients with cystic fibrosis.

*By Invitation

46. A Prospective Randomized Trial Comparing Suction to Water Seal for Air Leaks

Robert J. Cerfolio*, Ramu P. Tummala*, William L. Holman, George L. Zorn*, Charles R. Katholi* and Albert D. Pacifico, Birmingham, Alabama

Discussant: Claude Deschamps, M.D., Rochester, MN

OBJECTIVE: To compare whether suction or water seal for chest tubes is better at stopping air leaks

METHODS: One hundred forty consecutive pt who underwent elective pulmonary resection, were randomized to receive suction or water seal to their chest tubes after postoperative day (POD) #2. On the morning of POD #3, they were randomized to suction or seal. Chest tubes were checked daily for air leaks and were scored from 1 (least) to 7 (greatest) by a leak meter. Air leaks were also classified as forced expiratory, expiratory, inspiratory or continuous. Pt with air leaks that continued after POD #4 who had been randomized to suction were then placed on seal. Pt who had been randomized to seal who developed a pneumothorax were placed to -10 cm of suction.

RESULTS: There were 140 pt (96 men). On POD #1, 35 pt had an air leak. It was a forced expiratory leak in 21 pt (60%) and expiratory in 14 (40%). On POD #2, 33 pt had an air leak. It was a forced expiratory leak in 19 pt and expiratory in 14. On POD #3, 33 pt had air leaks, 18 pt were randomized to seal and 15 to suction. Of the 18 pt randomized to seal, air leaks resolved in 12 (66%) by the next morning. Four of the 6 other pt had air leaks greater than 3/7. In the suction group, only 1 pt's air leak resolved. The remaining 14 pt were placed to seal on POD #4 and 13 pt's leaks resolved after 24 hours. Eight pt who were placed on seal had a pneumothorax and 6 had leaks of 3/7 or greater.

CONCLUSIONS: Placing chest tubes on water seal is superior to suction for sealing air leaks after pulmonary resection (p=0.001). Water seal does not stop expiratory leaks that are greater than 4/7. Pneumothorax, although rare, can occur after placing chest tubes to water seal, especially with leaks greater than 4/7.

2:45 p.m. INTERMISSION - Visit Exhibits

*By Invitation

3:30 p.m. SIMULTANEOUS SCIENTIFIC SESSION B-2 - GENERAL THORACIC SURGERY

Room 211-213, Ernest N. Morial Convention Center

Moderators: David J. Sugarbaker, AID.

Carolyn E. Reed, M.D.

47. Surveillance Transbronchial Lung Biopsies: Implication for Survival after Transplantation

Scott J. Swanson*, John R. Reilly*, Steven J. Mentzer, Malcolm M. Decamp*, Edward P. Ingenito*, Raphael Bueno*, Lester Kobzik*, Jeanne M. Lukanich*, Michael T. Jaklitsch* and David J. Sugarbaker, Boston, Massachusetts

Discussant: Thomas M. Egan, M.D., Chapel Hill, NC

OBJECTIVE: Does early rejection(rej) after LTX by TBBX predict survival.

METHODS: 113 pts had LTX from 1990-1998. We have minimum 1-yr follow-up and results of first 3 TBBX on 89 consecutive pts. Survival was tabulated using Kaplan-Meier lifetable and statistical analysis done by Log-Rank Test. Surveillance TBBX was done in 1^st mo then at 3mo and 6mo. Standard immunosuppression was induction therapy with either Minnesota Antilymphocyte Globulin or Antithymocyte Gammaglobulin and methylprednisone and triple drug maintenance: prednisone,CyA,azathioprine. Acute rej was treated with methylprednisolone Igm/dx 3d, persistent acute rej (>2 consecutive) with total lymphoid irradiation and maintenance change to tacrolimus and mycophenolate in 5/9 pts. Blinded grading was done retrospectively using ISHLT classification.

RESULTS: l-yr survival for 89 is 79%,51% at 3yr. Survival was not significantly different in subset with rej the 1^st(n=36), 1^st and 2^nd(n=16), or l^st, 2^nd, 3^rd(n=9) or no rej on l^st, 2^nd, 3^rd(n=20) post-LTX TBBX. 61 positive biopsies were graded, 11/36 pts showing > 1 moderate/severe episodes. Survival for this group was not statistically different(p=0.10).

Rejection #	1yr%	2yr%	3yr%	5yr%	p
1 n=36	78 n=29	62 n=19	49 n=10	49 n=2	0.89
1,2 n=16	93 n=15	76 n=9	61 n=5	61 n=3	0.30
1,2,3 n=9	100 n=9	76 n=7	57 n=4	57 n=2	0.87
No 1,2,3 n=20	85 n=18	69 n=14	58 n=10	45 n=6	0.84

CONCLUSIONS: Surveillance and aggressive treatment of persistent early acute rej leads to survival comparable to pts who do not exhibit early acute rej.

*By Invitation

48. Does Pneumonectomy for Lung Cancer Adversely Influence Long-Term Survival?

Mark K. Ferguson and Theodore Karrison*, Chicago, Illinois

Discussant: Leslie J. Kohman, M.D., Syracuse, NY (not confirmed)

OBJECTIVE: The increased operative mortality associated with pneumonectomy has stimulated the use of lung sparing operations such as sleeve lobectomy. Whether pneumonectomy adversely affects long-term outcome after resection for lung cancer is unknown.

METHODS: We reviewed lobectomy/bilobectomy and pneumonectomy performed for stages I-III non-small cell lung cancer from 1980-97. Kaplan-Meier survival curves were compared using the log-rank test. Covariates were determined for operative mortality and survival using logistic regression analysis and Cox proportional hazards estimation.

RESULTS: 258 men and 179 women with a mean age of 62 yrs underwent lobectomy/bilobectomy (334) or pneumonectomy (103). 209 were stage I, 99 were stage II, and 129 were stage III. Operative mortality was 36 (8.2%) overall, 22 (6.6%) for lobectomy/bilobectomy and 14 (13.6%) for pneumonectomy. Mean follow-up was 41 mos (range 0 - 222). Median survival was worse for pneumonectomy (stage II: 17.9 vs 36.3 mos, p=0.05; stage III 11.4 vs 21.1 mos, p=0.07), an effect that was not significant excluding operative deaths (stage II: 21.7 vs 37.8 mos, p=0.14; stage III 14.4 vs 22.0 mos, p=0.17). Covariates for operative mortality were pneumonectomy (relative risk 2.7; 95% C.I. 1.3-5.6) and performance status (2.6; 1.5-4.7). Covariates for survival (operative deaths included, stratified by stage) were age (1.3; 1.1-1.4), performance status (1.4; 1.1-1.8), and postoperative predicted FEV₁% (1.2; 1.1-1.3). Pneumonectomy did not achieve statistical significance as a covariate for survival whether operative mortality was included (1.2; 0.8-1.8) or excluded (1.4; 0.9-2.1).

CONCLUSIONS: The adverse effect of pneumonectomy on survival relates primarily to its immediate operative risk. We demonstrated no significant long-term adverse influence of pneumonectomy on survival.

1998-99 AATS Graham Fellow

*By Invitation

49. Surgical Resection of Unilateral Lung Metastases: Unilateral or Bilateral Thoracotomy?

Riad N. Younes*, Jefferson L. Gross* and Daniel Deheinzelin*, Sao Paulo, Brazil

Sponsored by: Adib D. Jatene, M.D., Sao Paulo, Brazil

Discussant: Joseph S. Friedberg, M.D., Philadelphia, PA

OBJECTIVE: To evaluate the need for bilateral thoracotomy in patients diagnosed with unilateral lung metastases.

METHODS: A retrospective evaluation of a prospective data base from a single institution(1990-1997) of all consecutive patients (n=267) diagnosed on admission with unilateral (n=179) or bilateral(n=88) lung nodules. Ipsilateral thoracotomy was performed on all patients with unilateral disease; contralateral lung was only explored if de novo nodules were detected. Bilateral thoracotomy was performed on all patients with bilateral lung metastases. Histology: adenocarcinoma(25%), osteosarcoma(23%), squamous cell carcinoma(18%), soft tissue sarcoma(18%). Median follow-up was 17 months. Contralateral-disease free survival and overall survival were determined. Univariate and multivariate analyses were performed to determine prognostic factors for overall and contralateral-disease free survival. The 2 groups of patients with confirmed bilateral metastases (synchronous or metachronous) were compared.

RESULTS: Acruarial overall 5-year survival was 34.9%. Contralateral-recurrence free 6 months, 12 months, and 5 year survival were 95%, 89%, and 78%, respectively. Patients who recurred in contralateral lung within 3, 6 and 12 months had an overall 5-year survival of 24%, 30%, and 37%, respectively. When patients who recurred in contralateral lung were compared to patients with bilateral metastases on admission, there was no significant difference in overall survival. Only histology and the number of pathologically-proven metastases significantly (p<0.05) predicted recurrence in contralateral lung.

CONCLUSIONS: Bilateral exploration for unilateral lung metastases is not warranted. Most patients will only have unilateral disease, and delaying contralateral thoracotomy until radiologically detected disease does not affect outcome.

*By Invitation

50. Long-term Results of Cricopharyngeal Myotomy for Muscular Disease

Talat S. Chughtai*, Long-qi Chen*, Dimitrios Nastos*, Raymond Taillefer*, Pasquale Ferraro*, and Andre C. Duranceau, Montreal, Quebec, Canada

Discussant: Antoon E. M. R. Lerut, M.D., Leuven, Belgium

OBJECTIVE: Muscular disease may cause progressive oropharyngeal dysphagia and tracheobronchial aspiration. When these symptoms are present, short-term improvement has been consistently documented following Cricopharyngeal myotomy. Our aim is to analyze the long-term effects of this operation in patients where muscular dystrophy is responsible for the dysphagia.

METHODS: 13 dystrophic patients having undergone Cricopharyngeal myotomy for more than 10 years were retrospectively assessed. The effects of myotomy were measured clinically, using a symptom score (0: no symptom to 3: severe or frequent). Radiologic, manometric and radionuclide pharyngeal emptying studies were obtained. All parameters were measured for both short-term (<6 years) and long-term (>6 years) results.

RESULTS:

	Pre-op	p	Short-term	p	Long-term
Symptom (0 to 3)
Dysphagia to solids	2.92	0.0001	0.46	0.003	1.77
Regurgitation	1.46	0.0005	0	0.002	1.23
Aspiration with meals	1.15	0.03	0.15	0.007	1.31
Voice change	0.08	NS	0.15	0.006	0.85
Limb weakness	0.08	NS	0.15	0.02	0.85
Esophago-gram (%)
Abnormal c-p relaxation	69.2	0.0003	0	NS	0
Stasis	30.8	NS	23.1	0.02	85.7
Aspiration	30.8	NS	15.4	NS	57.1
Manometry (mm Hg)
UES Resting Pressure	35	0.01	20.89	NS	25
UES Closing Pressure	51.33	0.03	31.67	NS	45.2
Emptying Scintiscan
% Stasis at 2 min.	10	NS	14.38	NS	18.57

CONCLUSIONS: When dystrophy causes debilitating dysphagia, Cricopharyngeal myotomy results in significant and consistent early improvement. Physiologic alterations include a decrease in resting and closing pressures at the pharyngoesophageal junction. Late assessment reveals: 1) the unchanged physiologic effects of the operation 2) reappearance of oropharyngeal symptoms 3) manifestation of dystrophy in previously intact peripheral muscle groups 4) increasing hypopharyngeal stasis with time. Cricopharyngeal myotomy is a palliation in the natural evolution of the dystrophic process.

4:50 p.m. EXECUTIVE SESSION (MEMBERS ONLY) - Ballroom

6:30 p.m. MAGIC OF MARDIGRAS RECEPTION

MARDI GRAS WORLD

*By Invitation

1:45 p.m. SIMULTANEOUS SCIENTIFIC SESSION C-2 - CONGENITAL HEART DISEASE

Room 208-210, Ernest N. Morial Convention Center

Moderators: Frank L. Hartley, M.D. John E. Mayer, M.D.

51. Surgical Reintervention of Neopulmonary Arteries after Complete Unifocalization in Patients with Ventricular Septal Defect, Pulmonary Atresia, and Major Aorto Pulmonary Collaterals

Vadiyala Mohan Reddy*, Zahid Amin*, Phillip More*, David F. Teitel* and Frank L. Hanley, San Francisco, California

Discussant: Roger B. B. Mee, M.D., Cleveland, OH

OBJECTIVE: With early and complete one stage unifocalization of major aortopulmonary collaterals and the use of native collaterals, there is concern about the growth of the native collateral tissue and the need for subsequent reintervention. The purpose of this report is to examine the patterns of surgical reintervention after unifocalization and the outcomes.

METHODS: Between July 1992 and September 1998,81 patients (median age 7 months, range 10d to 37 yrs)have undergone complete one stage unifocalization with (groupI n=54) or without (groupII n=27) ventricular septal defect closure. All group I patients were evaluated by lung perfusion scans, echocardiography and when indicatedcardiac catheterization. All group II patients underwent elective cardiac catheterization 3 to 6 months after complete unifocalization.

RESULTS: Seventy early survivors were followed. There were 4 nonsurgical late deaths. Among the 66 survivors (group I n=46; group II n= 20). In group 1,6 patients (6/46; 13%)required balloon angioplasty and five of these patients also required surgical neopulmonary augmentation. In group II, 15 patients have undergone balloon angioplasty and 12 patients have undergone surgical neopulmonary artery augmentation with successful closure of the VSD 13 patients. The stenosis were primarily at anastamotic sites,in the central neopulmonary arteries or the native distal stenotic segments of the collateral vessels which could not be surgically addresses (in group II) patients.

CONCLUSIONS: Up to 6 year followup shows that incidence of neopulmonary artery reintervention in completely repaired patients is very low. In patients with complicated anatomy neopulmonary rehabilitation is successful in the majority with good hemodynamic outcome

*By Invitation

52. Twenty-five Year Experience with Rastelli Repair for Transposition of the Great Arteries.

Christian Kreutzer*, Julia De Vivie*, Kimberley Gauvreau*, Guido Oppido*, Jaqueline Kreutzer*, Michael Freed*, John E. Mayer, Richard A. Jonas and Pedro J. Del Nido, Boston, Massachusetts

Discussant: Gordon K. Danielson, M.D., Rochester, MN

OBJECTIVE: To evaluate the outcome of the Rastelli repair for transposition of the great arteries (TGA).

METHODS: From 3/73 to 4/98, 101 pts with d-TGA and ventricular septal defect (VSD) underwent a Rastelli repair. Pts with double outlet right ventricle and bilateral conus were excluded. The mean age at operation was 4.9 ± 5.7 yrs (ld-27 yrs) and the mean weight 16.8 ± 13.3 kg (3.3-71 kg). Pulmonary stenosis was present in 73 and pulmonary atresia in 18; 4 pts had multiple VSD's.

RESULTS: Right ventricle to pulmonary artery continuity was achieved by the use of 39 aortic homografts, 18 pulmonary homografts, 18 Hancock conduits, 11 Carpentier Edwards, 6 Dacron, 4 Tascon and 5 right ventricle-pulmonary artery direct anastomosis. The VSD was enlarged in 48 pts. There were 7 early deaths(6.9%) with none in the last 7 yrs. Risk factors for early death by univariate analysis included complete heart block (p=0.02), straddling tricuspid valve (p=0.04), longer cardiopulmonary bypass (p=0.02)and cross clamp times (p=0.04). At a mean follow-up of 6.6 ± 5.7 yrs (1 m to 22 yrs), there were 14 late deaths (5 sudden deaths), and 1 heart transplant. Late arrhythmias developed in 9 pts. Reoperations for conduit stenosis were performed in 45, for left ventricular outflow tract obstruction (LVOTO) in 10, and 31 had catheter interventions for conduit obstruction. Overall freedom from death or transplant (Kaplan-Meier) was 82.5%, 79.1% and 62.7 % at 5, 10 and 15 yrs respectively. Freedom from death or reintervention (transcatheter or surgical) was 43.6%, 16.0% and 8.0% at 5,10 and 15 years of follow up.

CONCLUSIONS: Rastelli repair for TGA and LVOTO can be performed with low early mortality. However, there is significant late morbidity and mortality associated with conduit stenosis, LVOTO and arrhythmias.

1998-99 AATS Graham Fellow

*By Invitation

53. Results of Norwood Stage-one Operation: Comparison of Hypoplastic Left Heart Syndrome with Other Malformations

Sabine H. Daebritz*, Georg D. A. Nollert*, Philipe N. Khalil*, John E. Mayer, Pedro J. Del Nido and Richard A. Jonas, Boston, Massachusetts

Discussant: William I. Norwood, M.D., Wilmington, DE

OBJECTIVE: Norwood stage-one operation is performed in hypoplastic left heart syndrome (HLHS) and other complex malformations with ductus dependent systemic circulation. We investigated the outcome in these two groups of patients.

METHODS: Between 1990 and 1998, eight surgeons performed the Norwood stage-one procedure in 194 patients at a median age of 5 days (weight 3.5+/-2.5kg; 32.3% female). Malformations in 131 patients were classified as HLHS in the presence of aortic and mitral atresia or severe stenosis with normal seg-mental anatomy (SDS), intact ventricular septum and hypoplasia of the left ventricle; 63 had other lesions with aortic outflow obstruction: hypoplastic left ventricle and VSD (n=18), unbalanced complete AV-canal (n=9), complex double outlet right ventricle (n=14), single LV or double inlet LV (n=11), tri-cuspid atresia with transposition of the great arteries (n=6), and others (n=5) including heterotaxia.

RESULTS: Operative and one year survival was significantly lower for patients with HLHS compared to those with other lesions (63.4% versus 81%, p=0.013 and 55.7% versus 73%, p=0.027, respectively). The presence of a non-hypoplastic left ventricle (n=27) was associated with significantly higher operative survival in uni- and multivariate analysis (96.3% versus 64.7%, p=0.001). Other echocardiographic measurements of anatomical structures such as size of the ascending aorta were not found to have an impact on operative survival. Prematurity was the only additional patient related risk factor (p=0.022).

CONCLUSIONS: The outcome of patients with malformations other than HLHS after Norwood stage-one procedure is better than of those with HLHS. The presence of an anatomically left ventricle is the single most important predictor of survival.

*By Invitation

54. Repair of the Hypoplastic Left Heart: Survival, Quality-of-Life, and Cost.

Deborah L. Williams*, Judy H. Ng*, Emily Crawford", Annetine C. Gelijns*, Alan J. Moskowitz*, Constance J. Hayes*, Mark E. Galantowicz* and Jan M. Quaegebeur, New York, New York

Discussant: John L. Myers, M.D., Hershey, PA

OBJECTIVE: The debate about the hypoplastic left heart syndrome (HLHS) is moving from whether to how to treat patients with this defect. Beyond survival, little is known about the QoL and costs of alternative treatment approaches. This paper analyzes these endpoints for the Norwood staged repair.

METHODS: Between 1993-98, 62 patients underwent staged repair for HLHS (Stage 1:62; Stage 2:25; Stage 3:7; 2 patients required conversion to cardiac transplantation). Survival was analyzed by the Kaplan Meier method, QoL was measured by the Infant/Toddler Child Health Questionnaire (I/T CHQ), developmental status measured by the Ages and Stages Questionnaire (ASQ). Inpatient costs were calculated with the ratio-of-cost-to-charges approach, outpatient costs were calculated using payments.

RESULTS: Overall survival at 4.7 years was 56%; survival beyond stage 2 and 3 was 96% and 100%, respectively. QoL ratings (mean 1.9 years, 0-100 scale) were as follows: global health (89.1±18.7); physical abilities (82.9±21.1); soc. interactions (68.6±9.9);and health percept.(53.9±21.5). However, 47% scored below established norms on the overall ASQ. The mean inpatient cost for stage 1, 2, and 3 repairs was $59,280 (±114,605), $26,700 (±13,215), $38,925 (±26,013), respectively. Total outpatient costs were less than 2% of total costs.

CONCLUSIONS: Despite progress, survival after stage 1 remains uncertain and needs improvement. QoL is surprisingly high by parents' standards but developmental status lags behind peers at this early stage. The majority of costs are inpatient costs (which are comparable to cardiac transplantation), while outpatient costs, by contrast, are low.

2:45 p.m. INTERMISSION - VISIT EXHIBITS

*By Invitation

3:25 p.m. SIMULTANEOUS SCIENTIFIC SESSION C-2 - CONGENITAL HEART DISEASE

Room 208-210, Ernest N. Morial Convention Center

Moderators: Frank L. Hartley, M.D.

John E. Mayer, M.D.

55. Biventricular Repair for Aortic Atresia or Hypoplasia and Ventricular Septal Defect

Richard G. Ohye*, Koji Kagisaki*, Lisa Lee*, Ralph S. Mosca*, Caren Goldberg* and Edward L. Bove, Ann Arbor, Michigan, Osaka, Japan

Discussant: Richard A. Jonas, M.D., Boston, MA

OBJECTIVE: Aortic valve atresia or hypoplasia can present with a VSD and a normal mitral valve and left ventricle. These patients may be suitable for a biventricular repair (BVR). The optimal management of aortic atresia/hypoplasia with VSD remains uncertain.

METHODS: From 1991-1998, 17 patients with aortic atresia/hypoplasia and VSD underwent BVR. Aortic atresia was present in 5 patients and 12 had aortic valve hypoplasia. Among the group with aortic hypoplasia, Z scores for the diameter of the aortic annulus ranged from -8.8 to -2.7. Associated anomalies were common and included interrupted aortic arch (10), coarctation (5), AP window (1), and heterotaxy (1). Eight patients were staged with an initial Norwood procedure followed by BVR, while 9 were corrected with a single procedure.

RESULTS: Among the 8 patients undergoing staged repair, there were no deaths after the Norwood procedure and one death after BVR due to low cardiac output and sepsis. For the 9 patients having a primary BVR, there was one early death due to low cardiac output, and two late deaths from non-cardiac causes. Follow-up ranged from 1 to 85 months (mean, 28 months). Actuarial survival for the entire group was 76 ± 12% at 5 years and was not significantly different for the staged repair group (88%) when compared to the patients undergoing primary BVR (67%). There was no significant morbidity among late survivors.

CONCLUSIONS: Both primary and staged BVR for patients with aortic atresia or hypoplasia and VSD may be performed with good late survival. Although the superiority of either approach was not clearly established in this series, patients with diminished pulmonary function, who would tolerate shunt dependent pulmonary blood flow poorly, should be considered for primary repair. Morbidity and mortality is largely related to associated anomalies.

*By Invitation

56. Does the Degree of Cyanosis Affect Myocardial Bioenergetics and Function?

Hani K. Najm*, Jack Wallen*, Michael P. Belanger*, John G. Coles, Glen S. Van Arsdell*, Michael D. Black*, William G. Williams, and Carin Wittnich*, Toronto, Ontario, Canada

Discussant: Bradley S. Allen, M.D., Oak Lawn, IL

OBJECTIVE: Animal studies indicated detrimental effects of exposure to chronic hypoxia on myocardial metabolism and function. Whether the presence or the degree of cyanosis adversely affects myocardial bioenergetics, ventricular function and clinical outcome in children.

METHODS: 48 children undergoing repair of tetralogy of Fallot were divided according to their preoperative saturation: group I; ≥ 90% (n=14), group II; ≥80 - 89% (n=16) and group III; 65-79% (n=18). ATP was measured from RV biopsies taken (a) before ischemia, (b) 15 minute of ischemia, (c) end ischemia and (d) 15 minute reperfusion. Ventricular function was assessed by echocardiography in the pre, intra and early postoperative period.

RESULTS: Group III had lower ATP levels at baseline (15.1 vs 19.1 vs 21.4 µmol/g/dry wt, group III, II, I respectively, P <0.01 ) and at 15 minutes of ischemia (11.2 vs 14.77 vs 17.6 μmol/g/dry wt, group III, II, I respectively, P <0.01). With reperfusion both cyanotic groups lost further ATP from end is-chemic level compared to an actual recovery in the acaynotic group (-22% vs 20% vs 18%, group III, II, I respectively, P <0.01). Cyanotic children also had lower preoperative ejection fraction (59 vs 66 vs 65%, group III, II, I respectively, P <0.01). Clinical outcome of children in group III was complicated as evidenced by longer ventilatory support (85 vs 31 vs 40 hours, group III, II, I respectively, P 0.07), inotropic support ( 86 vs 38 vs 36 hours, group III, II, I respectively, P <0.01) and intensive care unit stay (160 vs 60 vs 82 hours, group III, II, I respectively, P 0.02).

CONCLUSIONS: Cyanotic children undergoing cardiac surgery are at a precarious metabolic and functional status, and these children should be identified to be at a potentially higher risk of complications.

*By Invitation

57. Selective Cerebral Perfusion in Infants/Neonates Undergoing Complex Aortic Arch Reconstruction

Michael D. Black*, Bruno Bissonette* and Vivek Rao*, Toronto, Ontario, Canada

Discussant: Robert H. Jones, M.D., Durham, NC

Category

Sponsored By: Bruce A. Reitz, Stanford, California