WEDNESDAY MORNING, APRIL 20, 1988

6:45 a.m. SIMULTANEOUS BREAKFAST SESSIONS**

(See page 5 for further information)

C) CARDIAC ASSIST DEVICES AND STRATEGIES

Larry W. Stephenson, M.D., Philadelphia, Pennsylvania

D) PRE-OPERATIVE CARDIOPULMONARY EVALUATION

Joel D. Cooper, M.D., Toronto, Ontario, Canada

8:30 a.m. SCIENTIFIC SESSION - Ballroom

29. Experience with the Gore-Tex Surgical Membrane for Pericardia! Closure in Congenital Heart Surgery

JOSEPH J. AMATO, JOSEPH V. COTRONEO*,

RALPH J. GALDIERI*, JOSE R. ANTILLON*,

R. LEE VOGEL* and THOMAS M. CONNOR*

Newark, New Jersey

From 1984 through September 1987, pericardial closure in 88 patients (pts.) who underwent repair of congenital heart defects, was completed with placement of the Gore-Tex (polytetrafluoroethylene) surgical membrane. Review of these cases shows increased use of the membrane over the 3½ year (yrs) period, with 6 cases in 1984, 19 in 1985, 27 in 1986 and 34 cases in the first 9 months of 1987. The average age was 3.17 yrs. with a range of 3.0 to 44.6 kg. The pts. were divided into groups; Group I - 32 pts. with simple repairs (ventricular septal defects, atrial septal defects, partial canal defects, pulmonary stenosis, and uncomplicated tetralogy of Fallot); Group II - 16 pts., complex repairs with conduits; Group III - 40 pts. with complex repairs (Mustard, Jatene, Fontan procedures, total anomalous pulmonary venous drainage, aortic stenosis).

Early results show that there were no infections in any group and also no complications secondary to the Gore-Tex membrane. No increase in pericardial drainage was found in any group when compared to similar groups without the membrane. There were a total of 16 deaths (17.7%). Group I 1/32 (3%), Group II 6/16 (37.5%, and Group III 9/40 (22.5%). None of these deaths could be attributed to the placement of the Gore-Tex membrane.

There were two late reoperations which tested the usefulness of the membrane and in both, the findings were the same. There were no adhesions between the chest wall and the membrane which had basically become transparent. It was loosely attached to epicardium and easily removed except at the periphery where sutures had been placed. On the epicardium was a thin layer of collagenous fibrous tissue which did not interfere with the visualization of the heart anatomy including the coronary vessels.

In the postoperative follow-up, there was no difficulty in the M - mode or 2 - D echocardiographic examination of the heart. When the Gore-Tex membrane was placed externally on the chest wall and then the echo probe placed on top of the membrane, there was moderate blurring of the finer details of the cardiac architecture. Clinically this was evaluated by reviewing the 2 - D echocardiogram of all the post operative Mustard pts. with the Gore-Tex membrane and comparing these with Mustard pts. with no membrane. There was no difference in the clarity of the studies between the groups.

Conclusions are:

1. The Gore-Tex surgical membrane can be safely used for pericardial closure.

2. Reoperation shows no adhesions to the chest wall or epicardium and visualization of the cardiac surface, especially the coronary vessels, is not obscured.

3. No difficulty was encountered in the echocardiographic examination of the pts. with the Gore-Tex membrane.

*By Invitation

**Admission will be by ticket only and will be limited. Tickets must be obtained in the Registration Area of the Century Plaza Hotel prior to 2:00 p.m. on Monday, April 18. There are no provisions for pre-registration. Breakfast will be served until 7:00 a.m. only.

30. Prevention of Reperfusion Injury in the Neonatal Heart Using Leukocyte-Depleted Blood

MICHAEL A. BREDA *, DAVIS C. DRINKWA TER*.

HILLEL LAKS, ANTONIO F. CORNO*,

HAKOB G. DAVTYAN* and PAUL CHANG*

Los Angeles, California

Activated leukocytes release oxygen free radicals and cause microvascular occlusion. This experiment tests the hypothesis that reperfusion with leukocyte-depleted blood reduces injury following extended ischemic preservation.

An in vitro model consisting of an isolated, working neonatal piglet heart and an adult support pig was used. Three groups were compared. Control hearts (n = 45), perfused with whole blood without a period of preservation, developed a left ventricular stroke work index (SWI) at a left atrial pressure of 9 mmHg of 1.29 ± 0.39 x 10⁴ erg/g (Mean ± S.D.). In Groups 1 and 2, hearts were given crystalloid cardioplegia, excised, and stored in 4°C saline for 12 hours. Group 1 (n = 8), reperfused with whole blood, had a SWI of 0.031 ± 0.088 x 10⁴ erg/g (2.4% of control). Group 2 (n = 6), reperfused with blood rendered leukocyte-depleted by a polyester filter, had a SWI of 1.16 ± 0.25 x 10⁴ erg/g (89.9% of control). This difference was highly significant (p<0.0001). Group 2 had normal ultrastructure on electron microscopy. Group 1 exhibited severe injury with myofibrillar necrosis, mitochondria! disruption, nuclear chromatin clumping, and moderate interstitial edema.

We conclude that reperfusion with leukocyte-depleted blood prevents reperfusion injury and results in excellent myocardial function after long-term heart preservation.

*By Invitation

31. End-To-Side and End-To-End Vascular Anastomoses Using CO₂ Laser

SEISUKE NAKA TA *, CHARLES D. CAMPBELL,

RUTH PICK* and ROBERT L. REPLOGLE

Chicago, Illinois

There are several theoretical advantages to laser anastomoses. This study was designed to compare laser with conventional suture anastomoses. 120 end-to-end (E-to-E) and 40 end-to-side (E-to-S) laser anastomoses were performed on rabbit carotid arteries (2.0-3.0 mm). In each of 80 rabbits the divided left carotid artery was anastomosed using continuous suture technique, and the right carotid was anastomosed using a CO₂ laser. In each of other 40 rabbits both E-to-E laser and E-to-S laser anastomosis were performed on a carotid artery. The laser technique involved the placement of 3 stay sutures (E-to-E) or 4 stay sutures (E-to-S) of 7-0 polyplopylene and everting laser seal (65 mW, Ricoh Co. TC 3-C 100). The stay sutures were placed slightly further from the vessel edges than the standard suture technique. Examination of patency by angiography, tissue bonding strength and scanning electron microscopy as well as standard histology of laser anastomoses was carried out sequentially from one hour to one year.

The overall patency rate was 98% (78/80) in E-to-E laser, 79% (63/80) in suture and 95% (38/40) in combined E-to-E and E-to-S laser. No aneurysmal formation was observed in any group. Microscopic findings in laser anastomoses demonstrated degeneration of collagen and protein in the adventitia and media, but much less intimal injury as compared with suture anastomoses. Reendothelialization began on day 3-7 and was completed by day 14 in the laser groups in contrast to day 14 and 30, respectively, in the suture group. The tissue bonding strength in laser was weaker at hour 1 as compared with suture, but withstood the challenge of intraluminal pressure loads of 350 mmHg.

Results indicate CO₂ laser vascular anastomosis results in substantially lesser intimal injury, allowing for rapid early reendothelialization and excellent patency rate. Slight alterations of laser technique produces consistent success of anastomosing this size of vessels without aneurysmal formation. We will demonstrate the laser technique for the anastomoses by videotape. This technique may improve healing and patency in clinical aorto-coronary artery bypass operation.

*By Invitation

32. Metabolic Support of Remote Myocardium After Acute Coronary Occlusion

FRIEDHELM BEYERSDORF*, GERALD D. BUCK BERG,

and CHRISTOPHER ACAR*

Los Angeles, California

HYPOTHESIS: Metabolic support of remote "non-ischemic" myocardium during acute infarction will reverse the trend towards cardiogenic shock.

METHODS: Of 34 dogs undergoing LAD ligation and 50% stenosis of the circumflex artery (Cx) (allowing 70 - 100% reactive hyperemia), 20 developed irreversible ventricular fibrillation. The 14 survivors were followed for 6 hours measuring global and regional left ventricular function (CI, SWI, ultrasonic crystals) and regional blood flow (radioactive microspheres). After 2 hours 8 dogs received an intravenous infusion of glutamate/aspartate, glucose, insulin, potassium and Coenzyme Q₁₀ for 4 hours. Data are expressed as mean ± SEM; differences were considered significant at the p < 0.05 level.

RESULTS: The substrate infusion for 4 hours caused no change in regional or global cardiac function or coronary blood flow in 5 control dogs. The 6 untreated dogs developed cardiogenic shock (40% decrease in CI, 50% decrease in SWI)* due to persistent dyskinesis in the LAD region (- 40% systolic shortening [SS])* and hypocontractility in the Cx region (48% SS)* despite normal transmural blood flow in the posterior wall of the left ventricle (76 ml/100 g/min). In contrast, treated dogs recovered hypercontractility in the circumflex segment (138% SS)* and raised SWI to control levels (91%) without changing regional blood flow.

CONCLUSIONS: Cardiogenic shock after myocardial infarction is due to impaired ability of "non-ischemic" myocardium to maintain hypercontractility. This limitation can be prevented by metabolic support of viable muscle with intravenous substrate infusion before the definite treatment (e.g. CABG) is started. *p < 0.05)

9:10 a.m. Basic Science Lecturer

PHYSIOLOGY AND PATHOPHYSIOLOGY OF ESOPHAGEAL PERISTALSIS

Raj K. Goyal, M.D., Boston, Massachusetts

9:55 a.m. Intermission - Visit Exhibits

*By Invitation

10:25 a.m. Scientific Session - Ballroom

33. Coronary Angioplasty vs Coronary Bypass: Three Year Follow-up of a Matched Series of 250 Patients

MARK S. HOCHBERG, ISAAC GIELCHINSKY,

VICTOR PARSONNET, SYED M. HUSSAIN*,

ERIC MIRSKY* and DANIEL A. FISCH*

Newark, New Jersey

Two hundred fifty consecutive patients treated for one or two vessel coronary artery disease with either balloon angioplasty or surgical bypass were followed for three years to determine the comparative long term effectiveness of each treatment. The 125 angioplasty patients were matched with the 125 bypass patients so that both groups had a similar number of patients with single or double vessel disease. The two groups did not significantly differ in age, male:female ratio, NYHA Class, risk factors or ejection fraction. The LVEDP was 11 ± 7 mmHg in the angioplasty group and 14 ± 9 mmHg in the surgical patients (p = 0.0046).

Angioplasty was deemed initially successful in 88% (110/125), it failed in 10% (12/125) and in 2% (3/125) the lesion could not be crossed. Emergency bypass was performed in 10% (12/125). Four of these 125 angioplasty patients (3%) died within 30 days. Coronary artery bypass grafting was successfully performed on the matched set of surgical patients with 99% (124/125) discharged well. There was one (1%, 1/125) surgical death. The average hospital stay per patient was 4.8 ± 3.1 days for angioplasty and 12.1 ± 4.2 days for bypass.

Three year post procedure follow-up was obtained on 96% (236) of the 245 patients discharged alive. A second angioplasty was required in 18% and 11 angioplasty patients subsequently required surgical bypass. Overall, 19% (23/121) of the angioplasty patients ultimately required bypass. Four late deaths occurred in the angioplasty group, bringing the early and late mortality to 7% (8/121). There were two late surgical deaths bringing the combined surgical mortality to 2.5% (3/120), p = 0.1263.

The evaluation of symptoms revealed that 62% (75/121) of the angioplasty patients are alive and in NYHA Class I or II three years following one or two angioplasty procedures. This compares to 92% (110/120) of surgical patients alive and in the same two NYHA Classes, (p = 0.0000).

*By Invitation

34. Late Results After Intracoronary Thrombolysis and Early Bypass Surgery for Acute Myocardial Infarction

BRUNO J. MESSMER, RAINER UEBIS*,

CHRISTOPH RIEGER*, CARMINE MINALE*,

FERDINAND HOFSTADTER* and SVEN EFFERT*

Aachen, Federal Republic of Germany

Treatment of acute myocardial infarct has undergone major changes ever since thrombolytic therapy has proved to reduce significantly early mortality in comparison to conventional therapy. Thrombolysis results in reperfusion but does not alter the underlying arteriosclerotic stenosis. Additional and more definite treatment is necessary. Between April 1980 and the end of 1985 we performed early (1-10 days) bypass surgery in 70 patients. Hospital mortality was 1.4%. During a follow-up period from two to seven years (average 56 months) two cardiac and four non-cardiac deaths occurred. Actuarial survival was 90% at seven years. Reoperation was necessary in one patient. Postoperative angiocardiography was done in 50 patients (72%) to assess left ventricular wall motion in the formerly ischemic area. Normal or near normal wall motion was more often (68%) present when ischemia had been below three hours. In patients with reperfusion after three hours more persistent damage was recorded but complete recovery was still found in 45%. Correlation between ischemic time interval and late wall motion score was only 0.35.

In 24 patients transmural needle biopsies for electron microscopy studies had been taken from the formerly ischemic area at the time of surgery. Similar to late wall motion the extent of necrosis did not correlate with the ischemic time interval (r = 0.17) nor did the peak enzyme level or total enzyme activity. A clear correlation was, however, present between enzyme levels and the amount of necrosis within the biopsies (r = 0.76) as well as between the amount of necrosis and late wall motion (r = 0.69) in those patients who had intraoperative biopsy and late angiography.

It is concluded that early surgery after successful thrombolysis yields excellent long-term results but more reliable criteria for optimal aptient selection are mandatory.

*By Invitation

35. Six Months Postoperative Clinical and Angiographic Assessment of 163 Consecutive Sequential Mammary Grafts

ROBERT DION*, ROBERT VERHELST*,

MICHEL ROUSSEAU*, MARTIN GOENEN*,

ROBERT PONLOT* and CHARLES CHALANT*

Brussels, Belgium

Sponsored by: MARK BRAIMBRIDGE, London, England

Between October 1985 and March 1987, 163 consecutive sequential mammary grafts (SMG) were performed in 155 patients. Age averaged 61 years; 78% had a triple vessel disease; 69% had a history of myocardial infarction (MI); ejection fraction was below 0.4 in 25%; 33% had unstable angina. There were 16 (9.7%) reoperations. The length of the internal mammary artery (IMA) pedicle was the only limitation imponed on its use, and complex IMA grafting was confronted to the whole spectrum of grafting circumstances. No account was taken of the IMA free cut end flow. There were 8 free SMG, 16 triple SMG. Of the 340 sequential mammary anastomoses, 319 were built with the left IMA; 83 (24.4%) were diamond-shaped anastomoses. The right IMA was brought 27 times through the transverse sinus, 7 times for sequential grafting of circumflex branches. Taking in account the adjunctive venous anastomoses, and the single IMA anastomoses, there were 4.5 distal anastomoses per patient, 2.4 being mammary anastomoses, 2.2 being sequential mammary anastomoses. Extensive coronary endarterectomy could not be avoided in 22 cases (14%): in 17 cases, the left IMA was implanted into a venous patch primarily used to close the long arteriotomy. Additional procedures were valvular replacements or repairs (2 aortic, 3 mitral), right coronary ostium patch plastics (2), left main trunk plasty (1) and carotid endarterectomies (2). Hospital mortality rate was 1.3%. Perioperative significant MI rate was 3.3%. Follow-up averaged 15 months. 4 patients (2.5%) still experienced moderate angina pectoris and 4 other patients complained of residual dyspnea. 93 patients (60%) underwent a maximal stress test combined to a thallium scintigraphy at an average of 6 months post-operatively, with abnormal results in respectively 6/3% and 5.8% of the cases. 108 patients (70%) consented to a repeat catheterization at an average of 6 months after operation: 323 (95%) sequential mammary anastomoses were still patent. The patency rate of the diamond-shaped mammary anastomoses was 97%, that of the anastomoses corresponding to the right IMA's brought through the transverse sinus was 95%. As the attrition rate of the IMA grafts beyond 6 months postoperatively has proved to be minimal, gratifying long-term results with systematic use of SMG can be anticipated.

*By Invitation

36. Prospective Study of Adjuvant Surgery After Chemotherapy for Limited Small Cell Lung Cancer

ROBERT J. GINSBERG, FRANCES A. SHEPHERD*,

G. ALEXANDER PATTERSON*, RONALD FELD* and

MARTIN E. BLACKSTEIN*

Toronto, Ontario, Canada

Seventy-two patients with central L-SCLC were identified as candidates for adjuvant surgical resection (ASR) after remission induction with chemothrapy (CT). There were 47 male and 25 female patients, median age 61 yrs (range 39-77). Pre-treatment clinical staging revealed 21 stage I, (T1-2 NO), 16 stage II (T1-2N1), and 35 stage III (MO). Pre-operative CT included cyclophosphamide, adriamycin, and vincristine (CAV) for 62, CAV and etoposide for 7, and cisplatin and etoposide for 3 patients. Twenty-six patients achieved complete remission, 29 partial remission, 15 less than partial remission, and there were 2 early deaths. Thirty-four patients did not undergo ASR for the following reasons: randomized to radiation in another study protocol (10), patient refusal (9), inadequate response to CT (4), medically unfit for thoracotomy (3), early death (2). The remaining 38 patients underwent thoracotomy (8 pneumonectomy, 25 lobectomy). Five patients were not resected at thoracotomy (4 unresectable, 1 no tumor identified). Post-operative pathological stage for the 38 ASR patients revealed 7 stage I, 10 stage II, and 21 stage III. N1 disease was not identified pre-operatively for 2 patients, N2 disease for 6, and M1 (pleural) for 1. Postoperative pathology showed no tumor for 3, SCLC only for 29, non-SCLC for 4, and mixed histology for 2. Post-operative treatment included mediastinal irradiation (27) and cranial irradiation (24), but no further CT. Twenty-eight of 72 patients were alive without disease (18 of 38 ASR and 10 of 34 non-ASR). Seven patients died without recurrent SCLC and 37 died in relapse. Median survival for the ASR group is 91 wks versus 47 wks for the non-ASR group (p =0.001). Within the ASR group, patients with stage I (NO) disease had significantly (p = 0.037) longer survival (median not reached) than stage II or III patients (median 69 and 52 wks, respectively). Their survival was also significantly longer than that of the 10 stage I non-ASR patients (p = 0.001). No survival advantage was seen for stage II and III ASR patients (p = 0.086). The median survival (51 wks) of the 19 patients who were eligible but did not undergo ASR (10 randomized, 9 refused) was significantly less than that of the ASR patients (p = 0.049), and only 1 of 6 stage I patients is alive in this group.

In summary, ASR for responding patients after chemotherapy results in long-term survival and perhaps cure for a significant number of stage I patients. Because a survival advantage cannot be demonstrated for patients with stages II and III, intensive pre-treatment and pre-operative investigation including mediastinoscopy is essential to exclude patients who would not benefit from such combined modality therapy. At this time, surgery cannot be recommended outside the clinical trial setting.

*By Invitation

37. The Effect of Vertebral Column Invasion on the Surgical Treatment of Lung Cancer

CLAUDIO BOTTI*. TOM R. DeMEESTER

and PETER J. DAWSON*

Omaha, Nebraska

The New International TNM Staging System, as proposed by Mountain, characterizes all tumors invading the vertebral column as having a sufficiently poor prognosis to discourage resection. This is contrary to our experience with 12 patients treated between 1976 and 1987 whose tumors clinically extended into the vertebral column. All received 3000 rads of preoperative radiation followed by en bloc resection of the lung and costovertebral angle including a portion of the vertebral body. A complete mediastinal lymphadenectomy was also performed. In two patients the lowest trunk of the brachial plexus was resected, including the subclavian vessels in one. Preoperatively, three patients had a true Pancoast syndrome and in eight the tumor was located in the apex of the lung. All patients were free of mediastinal node involvement as assessed by pre-radiation mediastinoscopy. Resectability was based on the absence of tumor invasion as assessed by tomograms or CT scans into the triangle formed by the rib and the transverse process and pedicle of the vertebra (costo-transverse foramen) (Figure 1). The first two structures were transected posteriorly and pushed inward, allowing the vertebral body to be resected with a tangentially oriented osteotome (Figure 2). There were eight adenocarcinoma, three squamous carcinoma and one large cell carcinoma. All were T3NO. Follow-up ranged from five months to 11 years with an average of 36 months. Five patients died within 12 months from recurrent disease (one local, two distant and two local and distant). One patient died at five months disease free. Six patients are alive at five months to 11 years (three beyond five years) without evidence of tumor recurrence and arthritic pain. The overall five and ten year survival rate (Kaplan-Meier method) was 42 percent, which is similar to other resected T3NO tumors. Vertebral involvement does not always preclude resection. In proper patient selection, the removal of the involved portion of the vertebral body as described is associated with long term survival without sequelae.

*By Invitation

38. Primary Isolated Aortic Valve Replacement: Early and Late Results

BRUCE W. LYTLE, DELOS M. COSGROVE,

PAUL C. TAYLOR*, FLOYD D. LOOP,

LEONARD A.R. GOLDING*, ROBERT W. STEWART*,

CARL C. GILL* and MARLENE GOORMASTIC*

Cleveland, Ohio

One thousand three hundred eighty-two consecutive patients underwent primary isolated aortic valve replacement (AYR) from 1972 through 1983. There were 52 (3.8%) in-hospital deaths. Multivariate analysis identified advanced age (p<0.0001), preoperative shock or cardiac arrest (p = 0.0002) and preoperative BUN >25 mg/100 ml (p = 0.006) as independent variables increasing in-hospital mortality, and the use of cardioplegia for myocardial protection (p = 0.006) as a factor decreasing mortality.

Follow-up (mean postoperative interval 103 months, range 39-187 months for late survivors, total of 10,125 patient years of follow-up) documented survival of 85% and 66% and event-free survival of 71% and 43% at 5 and 10 postoperative years, respectively. Mutivariate models were used to examine the impact of variables over a 10-year follow-up and advanced age, moderate or severe impairment of left ventricular function, coronary artery disease and preoperative BUN of >25 mg/100 ml were identified as factors decreasing late survival and event-free survival (all p<0.05). Patients with bioprostheses had better survival (p = 0.0001) and event-free survival (p = 0.0007) than patients with mechanical valves. Patients with bioprostheses had superior results only if not taking warfarin, and experienced more reoperations and endocarditis; those with mechanical prostheses suffered more strokes, myocardial infarctions, bleeding complications and thromboembolic events. Survival and event-free survival curves for patients grouped according to prosthesis type converged late in the follow-up. At 5 postoperative years univariate comparisons showed a significant advantage for bioprostheses while at 10 postoperative years they did not.

Analysis of patients grouped according to age at operation showed that bioprostheses were associated with improved survival and event-free survival for patients >60 years, improved survival but not improved event-free survival for patients in the 40-59 year range and did not influence either survival or event-free survival for patients <40 years.

We conclude the 10-year results after isolated AYR are influenced by both patient-related and management-related factors and the impact of those factors is different for patients of different ages.

12:30 p.m. ADJOURN

*By Invitation