1:30 p.m. PLENARY SCIENTIFIC SESSION

Sheraton Ballroom

Moderators: Mortimer J. Buckley, M.D.

Andrew S. Wechsler, M.D.

7. PEDIATRIC AND ADULT LUNG TRANSPLANTATION FOR CYSTIC FIBROSIS.

Eric N. Mendeloff, M.D.*, Charles B. Huddleston, M.D.*, George B. Mallory, M.D.*, Alan Cohen, M.D.*, Stewart Sweet, M.D.*, Burt P. Trulock, M.D.*, Sudhir Sundaresan, M.D.*, Joel D. Cooper, M.D. and G. Alec Patterson, M.D.

St. Louis, Missouri

Discussant: Frank C. Detterbeck, M.D.

It has been suggested that lung transplantation for cystic fibrosis is fraught with increased morbidity and mortality. Since 1989, we have performed 103 bilateral sequential lung transplants in patients with cystic fibrosis (46 pediatric, 48 adult) including 9 redo transplants. Mean age at transplantation for the entire population was 21 ± 10 years (13 ± 3 years in the pediatric population and 29 ± 8 years in the adult group) and mean weight was 43 ± 15 Kg (32 ± 10 Kg in the pediatric and 52 ± 12 in the adult group). Average waiting periods from time of listing to time of transplantation were 231 ± 175 days and 362 ± 210 days in the pediatric and adult populations, respectively. All transplants except one in the pediatric age group were performed using cardiopulmonary bypass while this modality was employed selectively in the adults (23%). Hospital mortality for the entire population was 2.1%, with both early deaths occurring in the adult age group. Bronchial anastomotic complications requiring dilation, stent placement or surgery took place with equal frequency in the pediatric and the adult population occurring in 15 of 206 (7.3%) anastomoses at risk. At an average length of follow-up of 2.1 ± 1.6 years. The one and three year actuarial survival for the entire group were 80% and 59%, with no significant difference between the pediatric and adult age groups. Mean forced expiratory volume in 1 second at the time of listing for transplant was 25 ± 9% for the entire population, while average values at 1 month and 1 year post-transplantation improved to 54 ± 17% and 79 ± 35%, respectively. There was an average of 1 episode of acute rejection per patient-year, with the majority occurring in the first 6 months post-transplant. Actuarial freedom from bronchiolitis obliterans (biopsy proven or by clinical criteria) was 63% at 3 years. The subset who underwent redo transplantation consisted entirely of patients in the pediatric age range (<18 years old), and the mean time from the initial transplant to re-transplantation was 56 ± 44 weeks. The combined early and late mortality in this group was 44%. Eight living related donor transplants have been performed (4 as primary transplants and 4 as redo transplants) with an early survival of 87.5%. Lung transplantation in patients with end stage cystic fibrosis can be performed with low peri-operative mortality and with complication rates similar to those seen in pulmonary transplantation for other disease entities.

*By invitation

8. MID-TERM RESULTS AFTER MINIMALLY INVASIVE CORONARY SURGERY (LAST OPERATION).

Antonio M. Calafiore, M.D.*, Giovanni Teodori, M.D.*, Gabriele Di Giammarco, M.D.*, Giuseppe Vitolla, M.D.*, Angela Iacó, M.D.*, Teresa lovino, M.D.* and Sergio Cirmeni, M.D.*

Chieti, Italy

Sponsored by: Tomas Antonio Salerno, M.D., Buffalo, New York

Discussant: Steven R. Gundry, M.D.

Background. Left internal mammary artery (LIMA) to left anterior descending (LAD) artery via a left anterior small thoracotomy (LAST) is a recently proposed procedure specially designed to be effective, reproducible and to increase patients' comfort. We reviewed our experience to evaluate if these goals could be considered achieved.

Methods. From November 1994 to October 1996 366 patients (pts) underwent LIMA to LAD grafting via a LAST. One hundred eighty-two pts (49.8%) had a single LAD disease, in 184 LAD lesion was part of multiple vessel disease. High risk factors for, cardiopulmonary bypass were present in 51 pts (13.9%). Intravenous diltiazem was infused during the operation. LIMA was harvested for the length enough to reach the LAD.

Results. One hundred forty-seven pts (40.1%) were extubated in the OR or in the 1st hour. Mean ICU stay was 4.2 h; mean postoperative in hospital stay was 53 h; 30 day mortality was 0.8% (3 pts); late mortality was 1.1% (4 pts). All pts who died but one had a patent anastomosis. Eighteen pts were reoperated on early (< 30 d) and 7 late (> 30 d) due to conduit or anastomotic malfunction; 4 were reoperated on with patent anastomosis for progression of disease (3) or pericarditis (1). Four pts had angina, 3 due to anastomotic stenosis (spontaneously reversed in 2) and 1 due to progression of disease. A 5th pt had dispnea; LIMA was patent but apical dyskinesia, was present. A patent and well functioning anastomosis, checked by angiography or stress doppler flow assessment, was obtained in 340 pts (92.9%). Twenty-three months after surgery, actuarial survival was 98.0% (100% in 1-v disease and 96.4% in 2/3-v disease, p = ns); event free was 87.9% (90.9% in 1-v disease and 85.8 in 2/3-v disease, p = 0.006). In the last 130 pts (from April 1, 1996), with increased experience and better instruments, a patent well functioning anastomosis was obtained in 128 pts (98.5%); 7 month survival was 99.1% (100% in 1-v disease and 97.9% in 2/3-v disease, p = ns) and event free survival was 93.5% (95.1% in 1-v disease and 91.5% in 2/3-v disease, p = ns).

Comment. LAST Operation is a safe operation that gives good midterm results. The great majority of events happened in the first four months, and are, at our opinion, due to technical factors or selection of the patients. However for single LAD lesion our experience compares favourably with stent PTCA procedures on LAD.

*By invitation

9. SURVIVAL AFTER PHOTODYNAMIC THERAPY FOR ENDOBRONCHIAL MALIGNANCY: A 14 YEAR STUDY.

James S. McCaughan, Jr., M.D.* and Thomas E. Williams, M.D. Columbus, Ohio

Discussant: Douglas E. Wood, M.D.

Background: After being injected intravenously, the photosensitizer dihematoporphyrin ether is selectively retained in the tumor cells. The photosensitizer absorbs 630 nanometer wavelength light (red) energy delivered from a laser and produces a singlet oxygen which destroys the tumor. A limiting factor in the effectiveness of PDT is the fact that the light only penetrates 5 to 10 mm. The bronchi, however, have a maximum diameter of 9 mm and therefore photodynamic therapy is ideally suited to relieving obstruction due to endobronchial tumors. Photodynamic therapy (PDT) was performed using 630 nm light generated by an argon dye laser system delivered through cylinder diffusing tip quartz fibers passed through the biopsy channel of a flexible endoscope.

Objectives: Determine factors affecting survival rates, benefits and complications of patients with endobronchial cancer treated with photodynamic therapy.

Methods: All patients had already received, refused, or were ineligible for other modalities; none was refused PDT because of a low performance status; and some were on a respirator when first treated. All signed informed consents approved by the Institutional Review Board. From 1982 to May, 1996 photo-dynamic therapy was performed on 175 patients with endobronchial and endo-tracheal tumors (158 squamous, 17 adeno). All were clinically staged at the time of PDT. Sixteen were Stage I, 9 Stage II, 42 Stage IIIA, 64 Stage IIIB, and 44 Stage IV. All patients were followed until death or November, 1996.

Results: Multivariate analysis of survival using a model of the effects of age, sex, race, histology, Karnofsky Performance Status (KPS) and clinical stage showed the clinical stage (p<.0001) was the only statistically significant factor. Sixteen Stage I patients had a 93% five-year disease related estimated survival. Median (months) survivals after PDT were: Stage I = not reached; Stage II = 22.5; Stage IIIA = 5.7; Stage IIIB = 5.5; Stage IV = 5.0. KPS does become significant when it reaches 50 but is not significant for Stages I or II. Within Stages III and IV a Karnofsky Performance Status (KPS) ≥ 50 had a significant effect. For Stage IIIA the median survival was 8.2 months when the KPS was ≥ 50 and 2.0 for a KPS < 50 . For Stage IIIB the median survival was 7.2 months when the KPS was ≥ 50 and 4.0 for a KPS < 50. For Stage IV survival was 6.5 months when the KPS was ≥ 50 and 2.6 for a KPS < 50.

Conclusions: Photodynamic therapy may be considered as an alternative treatment for patients under consideration for surgery for Stage I carcinoma who are high surgical risks. The length of palliation for "non-curative" patients was equal to or better than that reported historically for most other treatment regimens.

2:30 p.m. BASIC SCIENCE LECTURE

Implications for Gene Therapy in Treating Coronary Artery Disease and Lung Cancer.

Ronald G. Crystal, M.D., New York, New York

3:15 p.m. INTERMISSION - VISIT EXHIBITS

*By invitation

4:00 p.m. PLENARY SCIENTIFIC SESSION

Sheraton Ballroom

Moderators: Mortimer J. Buckley, M.D.

Andrew S. Wechsler, M.D.

10. LONG-TERM RESULTS, OVER 10 YEARS, OF CONSERVATIVE SURGERY OF CONGENITAL MITRAL VALVE INSUFFICIENCY.

Sylvain Chauvaud, M.D.*, Jean-Francois Fuzellier, M.D.*, Remi Houel, M.D.*, Alain Berrebi, M.D.*, Serban Mihaileanu, M.D.* and Alain F. Carpentier, M.D., Ph.D.

Paris, France

Discussant: Richard A. Hopkins, M.D.

Background: Previous publications from various authors have stressed the benefits of mitral valve repair over mitral valve replacement in children. Very few communications have reported the long-term results and none with follow-up over 10 years. This paper reports our results in a series of 141 patients (pts) operated on for congenital mitral valve insufficiency (MVI) using the same technique (Carpentier technique) in the same center.

Patients and Methods: Between 1970 and 1995, 141 patients (pts) younger than 12 years underwent surgery for congenital MVI. Mean age was 5.8 ± 3.1 years ranging from 0.5 to 12 years. According to Carpentier classification, 30 pts mitral dysfunction were classified type I (normal leaflet motion), 77 classified type II (leaflet prolapse), 34 classified type III (restricted leaflet motion), 14 with normal papillary muscle and 20 with abnormal papillary muscle (hammock or parachute valve). Associated lesions were present in 38 pts (27%). Conservative surgery was possible in 134 pts (95%). Among them, 66 pts required a prosthetic annuloplasty and 10 valve extension with patch. Valve replacement was necessary in 7 pts (5%).

Results: In-hospital mortality was 5.6% (8 pts). No early death was observed in the group of pts who underwent valvular replacement. In-hospital mortality in type I was 10%, in type II, 3.9%, in type III, 5.9% (p : NS). Only 2 of these pts had an associated lesion (ventricular septal defect). Early reoperation was required in 3 pts for recurrent MVI.

Mean follow up was 7.2 ± 6 years (0.4 to 25 years), available in 129 pts (97%). There were 5 late deaths. Actuarial survival was 89.2% ± 6.4% at 15 years and respectively 90.8% and 68.6% in pts who underwent mitral valve repair and in pts who underwent mitral replacement. Late reoperation was required in 6% (8 pts) : 5.5% (7 pts) in pts who had undergone mitral repair and 14.3% (1 pt) in pts with valve replacement. Causes of reoperation were recurrent MVI (6 pts), mitral stenosis (1 pt) and bioprosthesis degenerescence (1 pt). Actuarial freedom from reoperation was 85.9 ± 11.4% at 15 years and a linearized rate of pts exposed to reoperation was 0.9% pts-year. No thromboembolic event was observed.

Conclusion: Congenital MVI can be repaired in infancy with low mortality. Conservative surgery using Carpentier techniques is feasible in the majority of cases of congenital MVI. This technique offers stable long-terms results with a low rate of reoperation.

*By invitation

11. EFFICACY OF ENDOVENTRICULAR PATCH PLASTY REPAIR IN LARGE POST-INFARCTION AKINETIC SCAR AND SEVERE LV DYSFUNCTION. COMPARISON WITH A SERIES OF LARGE DYSKINETIC SCAR.

Vincent Dor, M.D., Marisa Di Donate, M.D.*, Michel Sabatier, M.D.*, Anna Toso, M.D.*, Fran9oise Montiglio, M.D.* and Mauro Maioli, M.D.*

Monte Carlo, Monaco and Florence, Italy

Discussant: Michael K. Pasque, M.D.

In previous studies we have demonstrated that endoventricular circular patch plasty repair (EVCCP) for post-infarction anterior LV aneurysm improves early and late clinical and haemodynamic status. Since 1984 more than 700 pts with different degree and type of post-infarction asynergies have been proposed in our Center for EVCPP, associated coronary grafting and cryotherapy (when indicated for ventricular arrhythmias). Large akinetic scars, associated with severely depressed pump function, are more difficult to treat by surgery as the limit between scar and sound tissue is not as clear as in pure dyskinetic aneurysm. Therefore the technique of the patch anchorage slightly differs, the site of the patch inside the left ventricle depending on the size of the chamber that is « worthy » to leave. The present report concerns 49 pts (60 ± 8 yrs) with large akinetic scar and compares to 40 pts (61 ± 9 yrs) with large dyskinetic scar proposed for EVCPP and coronary grafting. Pts were selected if EF ≤ 30% and A% (A is the extent of left ventricular perimeter involved by the asynergy) ≥ 60%. Regional wall motion was quantitatively evaluated with the centerline method before and after surgery. All pts have had an anterior myocardial infarction (Ml), groups were comparable for symptoms, indication for surgery, delay from MI and other clinical variables. Heart failure was the major indication for surgery in both groups, 72% of pts were in NYHA class III/IV, 13 pts were operated on emergency. Ventricular tachycardia (VT) was inducible in 43% (group 1) and 48% (group 2).

The mortality rate was 10, 2% (5/49) in akinetic and 17% in dyskinetic (7/40) -ns-. Associated procedures were mitral repair or replacement in 10 pts (5 in group 1 and 5 in group 2) and cryotherapy in 22/49 (group 1) and 20/40 (group 2). Coronary grafting was performed in 98% of pts (LIMA in 89%); the mean number of bypass was 1, 9 ± 0, 4. VT was still inducible in 7/33 pts of group 1; no pt in group 2 had inducible VT after surgery. Results show that the surgical outcome of EVCPP does not depend on the presence or absence of dyskinesia pts with large akinetic scar have even worse pre-operative hemodynamics nevertheless, they benefit from a surgical technique previously reserved only for dyskinetic aneurysms. The reduction of wall tension and oxygen demand due to the marked decrease in volume and the increase in oxygen supply due to complete myocardial revascularization play the major role in improving pump function in either akinetic or dyskinetic post-infarction scars with severely depressed pump function. Therefore, EVCPP can be considered as an alternative to heart transplantation in pts with end stage ischemic cardiomyopathy and predominant akinesia.

*By invitation

12. FIRST RESULTS WITH VIDEO-ASSISTED MINIMALLY INVASIVE MITRAL VALVE REPAIR USING THE PORT-ACCESS-SYSTEM.

Friedrich W. Mohr, M.D., Ph.D.*, Volkmar Falk, M.D.*, Anno Diegeler, M.D.*, Thomas Walther, Ph.D.*, Jaques A.M. van Son, M.D., Ph.D.* and Rudiger Autschbach, M.D., Ph.D.*

Leipzip, Germany

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

Discussant: W. Randolph Chitwood, Jr., M.D.

Background: This study was performed to evaluate the application of the Port-Access-System (Heartport, Redwood, CA) for video assisted minimally invasive mitral valve repair. As yet this is the largest series using this technique worldwide.

Patients and Methods: After approval by the local ethical committee 18 consecutive patients (mean age 65.1 ± 9.4 years, 16 female, 2 male, LVEF 58 ± 12%) were included in the study. Mitral insufficiency (MI) grade III-IV° was present in 14 patients while 4 had predominant mitral stenosis (MS). Patients (pts) were placed on femoro-femoral bypass and an aortic endoclamp (Heartport) was inflated in the ascending aorta under flouroscopy and TEE control. A minithoracotomy was performed in the 5th intercostal space (length of incision 3.8 to 6.5 cm). Cardiac arrest was induced by antegrade crystalloid cardioplegia via the distal lumen of the endoclamp. The left atrium was opened and a stereoscope inserted through a separate port. The heart was vented through a transvenously placed endopulmonary vent (Heartport).

Results: With videoscopic assistance quadrangular resection and ring implantation was performed in 7 pts. In 4 pts. commissurotomy alone (n = 2) or in combination with a ring annuloplasty (n = 2) and additional chordal replacement (n = 2) was performed. In 5 pts. partial or complete ring annuloplasty was performed. One pt. had persistent MI 11° after quadrangular resection, chordal replacement, and ring implantation and consequently underwent mitral valve replacement using the same approach. Mean duration of operation, cardiopulmonary bypass, and crossclamp time were 180 ± 33, 114 ± 21, and 71 ± 16 min, respectively. Intubation time was 25.6 hours (range 5 to 123 hours). Postoperative pain index (day 2) was low averaging 1.1 ± 0.8 on a 0-10 scale. Duration of ICU treatment and hospital stay were 2 days (1-11 days) and 12 days (10-20 days), respectively. There was one non cardiac related postoperative death. Two pts. required reexploration for bleeding (intercostal artery n = 1, port site n = 1). In 2 pts. transient psychosis most likely due to incomplete deairing was noted. At a mean follow up of 10 ± 6 weeks all patients are in a NYHA class I or II. Echocardiography revealed excellent results of mitral valve repair in all patients with only trivial regurgigation (equal or less MI I°) in 5 pts.

Conclusion: Using the Port-Access-System even complex mitral valve repairs can be performed minimally invasively with good results. The stereoscope allows visualization of all valvular structures in great detail and facilitates repair. With gaining experience operation time decreased to less than 2.5 hours making this new approach a valuable alternative to conventional mitral valve repair.

*By invitation

§13. PARTIAL LIQUID VENTILATION MINIMIZES PULMONARY PARENCHYMAL AND VASCULAR INJURY AND IMPROVES CARDIAC OUTPUT IN A NEONATAL SWINE MODEL OF CARDIOPULMONARY BYPASS.

Ira M. Chaifetz, M.D.*, Michael L. Cannon, M.D.*, Damian M. Craig, B.S.*, George Quick*, Ross M. Ungerleider, M.D., Peter K. Smith, M.D. and Jon N. Meliones, M.D.*

Durham, North Carolina

Discussant: John E. Mayer, M.D.

During cardiopulmonary bypass (CPB) organ protective strategies are traditionally directed at the myocardium and brain. Without a strategy for lung protection, the pulmonary parenchyma and vasculature may suffer severe injury after CPB, especially in neonates. CPB may result in a hypoxic/ ischemic injury, reperfusion injury, surfactant dysfunction, and immune system/complement activation. These processes often result in decreased pulmonary compliance, increased pulmonary vascular resistance, and, potentially, decreased cardiac output (CO). Partial liquid ventilation (PLV) has been shown to be beneficial in both clinical and animal evaluations of acute lung injury. The beneficial effects of PLV result from its oxygen carrying capability, surfactant function, alveolar distending properties, and anti-inflammatory properties. Thus, we hypothesized that PLV might minimize the pulmonary parenchymal and vascular injuries seen in neonates after CPB.

Methods: Twenty neonatal swine (2.0-3.4 kg) were randomized to receive CPB with (n = 9) or without (n = 11) PLV. In the liquid ventilated group, a single dose of perflubron (LiquiVent, Alliance Pharmaceutical Corp.) was administered to functional residual capacity prior to CPB. The control group (CTL) was ventilated conventionally. Each animal was placed on non-pulsatile CPB at 125 mL/kg/min and cooled to a nasopharyngeal temperature of 18°C over 20 minutes. Low-flow CPB (35 mL/kg/min) was then performed for 60 minutes. The flow rate was returned to 125 mL/kg/min, and the animals warmed to 37°C. The animals were removed from CPB, and data were obtained at 30, 60, and 90 minutes after CPB.

Results: The pre-CPB data (mean + sem) for each group were compared to the data 30 min. after CPB by paired t-tests (# p<0.05 vs. pre-CPB). The post-CPB data (30, 60, and 90 min.) were compared between the 2 groups using a linear regression model of analysis of variance with repeated measures (*p<0.05vs. CTL).

Conclusions: The lung protection strategy of partial liquid ventilation minimized the pulmonary parenchymal and vascular injuries associated with neonatal CPB while increasing cardiac output. PLV may become an important technique for protecting the lungs from the deleterious effects of CPB. The morbidity associated with CPB as well as the cost of post-operative care may be significantly reduced if the pulmonary sequelae of CPB can be diminished.

§Authors have a relationship with Alliance Pharmaceutical Corp.

*By invitation