American Association for

Thoracic Surgery

64TH ANNUAL MEETING

Scientific Program

MONDAY MORNING, May 7, 1984

8:30 a.m. Business Session (Limited to Members)

Grand Ballroom - Third Floor

8:45 a.m. Scientific Session - Grand Ballroom - Third Floor

1. Computed Tomography: An Effective Technique for Mediastinal Staging in Lung Cancer

BENEDICT D.T. DALY*, L. JACK FAILING*,

ROBERTD. PUGATCH*, Y. JUNG-LEGG*,

M. ELON GALE*, GUNARS BITE*

and GORDON L. SNIDER*

Boston, Massachusetts

Sponsored by: HAROLD F. RHEINLANDER

Boston, Massachusetts

Computed tomographic (CT) scans of the chest were utilized to stage the mediastinum in 148 instances of bronchogenic carcinoma considered for resection in 146 patients. Nodes > 1.5 cm in size were interpreted as abnormal. All nodes positive by CT were evaluated by mediastinoscopy (31), anterior mediastinotomy (7), or thoracotomy (7) depending on their anatomic location. All patients with negative findings underwent thoracotomy without prior surgical staging. Patients undergoing thoracotomy for staging or resection were divided into two Groups. In Group I (first 51 instances), routine mediastinal exploration was not carried out; in Group II Oast 97 instances) the mediastinum was explored in every patient and nodes were submitted for histopathology. This report compares the CT and pathologic findings on the mediastinal lymph nodes.

# Patients	True Positive	True Negative	False Positive	False Negative	Sensitivity	Specificity	Accuracy
Group I	15	32	2	2	88%	94%	92%
Group 11	18	65	8	6	75%	89%	86%
Central Tumor	26	38	9	4	87%	81%	83%
Peripheral Tumor	7	59	1	4	64%	98%	93%

Ten patients had false-positive scans, two with extensive granulomatous disease and seven with postobstructive pneumonitis; nine of the ten had central lesions and seven were located in the right upper lobe. Eight patients had false negative scans, six had either paraaortic or subaortic nodes (4) or posterior subcarinal nodes (2). These nodes would not have been accessible to mediastinoscopy. One patient had positive low paratracheal and subcarinal nodes but the metastases were focal and intranodal. One patient had small paratracheal nodes replaced by tumor and is the only patient with false negative nodes in whom routine mediastinoscopy would have prevented thoracotomy and resection. We believe CT staging of the mediastinum is indicated for all patients with lung cancer in whom surgery is contemplated. CT directs the most appropriate staging procedure for patients with positive findings and obviates mediastinoscopy for patients with negative findings.

*By Invitation

2. Adverse Prognostic Effect of N2 in Treated Small Cell Carcinoma of The Lung

JOHN A. MEYER, JOHN J. GULLO*,

PHILLIP M. IKINS*, ROBERT L. COMIS*,

WILLIAM A. BURKE* and FREDERICK B. PARKER, JR.

Syracuse, New York

Patients treated nonsurgically for "limited" small cell carcinoma of the lung relapse most frequently within the chest. We have sought to control this mode of treatment failure by adjunctive surgical resection, in addition to chemotherapy. Since 1975, we have treated patients with disease in clinical Stages I and II by initial resection and the full course of chemotherapy thereafter. Beginning in 1979, suitable patients with Stage III-MO disease (T3 and/or N2, MO) have been treated initially with two cycles of chemotherapy, followed by resection of the primary tumor, dissection of mediastinal nodes, prophylactic cranial irradiation, and continuing chemotherapy. Diagnosis of N2 in this study was made by positive biopsy at mediastinoscopy (nine cases), or in one case by CT demonstration of enlarged subcarinal nodes, plus later histologic finding of tumor in the excised nodes. Patients classified here as NO or N1 had negative mediastinoscopy, and no radiographic evidence of mediastinal node enlargement.

Many reviews find that continuing complete remission at 30 months after the start of treatment may be considered equivalent to disease eradication. Thirteen of our patients without the finding of N2 began treatment more than 30 months ago (three T3 N1; four T2 N1; four T2 NO; two T1 NO); one has died of a surgical complication and two (T3 N1, T2 N1) have suffered relapse at 14 and 17 months. The other ten patients (77%) have remained disease free.

Conversely, all of the first ten patients with an initial finding of N2 have developed known relapse within this period (nine patients), or in one instance, died at home with suspected recurrent disease. Known recurrences have appeared in the liver in four cases; in liver plus bone in two; in meninges, in opposite main bronchus, and in lung, one case apiece. Two relapses appeared very late, at 27 and 30 months after the start of treatment.

These findings suggest that disease control is possible in the majority of MO patients without demonstrable mediastinal node involvement. The initial finding of N2 may be indicative of dissemination sufficient in most cases to interfere with the patient's chances for long survival.

*By Invitation

3. Survival Following Sequential Resections for Second or Third Primary Lung Cancers

DOUGLAS J. MATHISEN*, ROBERT J. JENSIK,

L. PENFIELD FABER and C. FREDERICK KITTLE

Chicago, Illinois

The performance of sequential resections and the consideration of new lesions as second or third primary lung cancers remain controversial issues. Criteria to define these as new primary lesions depend upon a difference in histologic types, a prolonged interval between initial and second or third resections, and location in the contralateral lung or a different ipsilateral lobe. Ninety patients have undergone multiple resections for bronchogenic carcinoma from 1962 to November 1983. There were 10 examples of synchronous lesions and the remaining 78 were metachronous with the longest interval between resections being 17 years and four months.

The initial surgical procedures were pneumonectomy-11; lobectomy-42; sleeve lobectomy-9; segmentectomy-27; carinal resection-1. At the second operation, the procedures were: Segmentectomy-55; lobectomy-11; completion lobectomy-6; completion pneumonectomy-15. Two patients had sternotomy with bilateral resections and one patient had a tracheal resection. The third procedures were: Segmentectomy-6; completion segmentectomy-1; completion lobectomy-2; completion pneumonectomy-2.

In 18 patients undergoing the second procedure and 2 undergoing a third resection, a different cell type was identified. The peri-operative mortality following the second operation was 7 of 90 patients (8%), and there were no deaths in those patients undergoing three resections.

Cumulative survival following second resection in 78 patients with metachronous tumors was 33% at 5 years and 20% at 10 years.

*By Invitation

4. Experience With Primary Neoplasms of The Trachea and Carina

F. GRIFFITH PEARSON, THOMAS T.R. TODD*

and JOEL D. COOPER

Toronto, Ontario

Between 1963 and 1983, forty-three patients were seen with primary malignant tracheal neoplasms and managed on our surgical service. The final pathology was adenoid cystic carcinoma - 28 cases, squamous cell carcinoma - 8 cases, leiomyosarcoma - 4 cases, and 3 miscellaneous tumours.

Thirty-two patients were managed by resection and primary anastomosis: trachea only -12, trachea and carina -12, trachea and cricoid - 4, trachea and larynx - 4. There were two operative deaths.

Six patients had defects replaced with a heavy duty Mrlex prosthesis, and all but one of these was operated on during the 1960's. Three of six died post-operatively due to innominate artery erosion.

Three patients with non-resectable tumours were treated with an indwelling silicone stent: one Montgomery T-tube, two silicone TY-tubes. All three had worthwhile palliation. Two patients with extensive, but non-obstructive adenoid cystic carcinomas were managed by primary irradiation. Good clinical remission has been maintained in each at 30 and 36 months respectively.

The following observations are noted:

1) Adjunctive radiotherapy was used (either pre or post-operatively) in patients with adenoid cystic carcinoma. No patient having a complete resection has suffered a recognizable clinical recurrence. Furthermore, even patients with incomplete resections achieve good long-term palliation in most cases. This includes four patients with slowly progressive, asymptomatic pulmonary metastases.

2) Resection of cricoid with sparing of the larynx is possible in occasional patients with primary malignancies of the proximal airway.

3) With experience it is apparent that median sternotomy provides the best exposure for selected cases requiring carinal resection.

10:00 a.m. Intermission - Visit Exhibits - Second Floor

Complimentary Coffee

*By Invitation

10:45 a.m. Scientific Session - Grand Ballroom - Third Floor

5. Pulmonary Artery Banding

ROBERT A. ALDUS*, GEORGE A. TRUSLER,

TERUO IZUKAWA* and WILLIAM G. WILLIAMS*

Toronto, Ontario

From January 1972 through December 1982, 209 patients underwent pulmonary artery banding at The Hospital for Sick Children, Toronto. A method for estimating the required band circumference was used in all children without significant pulmonary vascular disease. Children were divided into four categories according to diagnosis. Infants in Group I had ventricular septal defect (VSD), atrioventricular septal defect (AVD) or double outlet right ventricle with a subaortic VSD (DORVA). We attempted to band these infants tightly at a band circumference of 20mm + 1mm for each kilogram of infant body weight. Overall mortality was low (9%), response was moderate or good (89%) and pulmonary hypertension was well controlled (normal pulmonary artery pressure at subsequent repair in 88%). The highest mortality occurred in infants with VSD and coarctation of the aorta (16%). Group II infants had complete transposition of the great arteries (TGA) with VSD, double outlet right ventricle with a subpulmonic VSD, single ventricle or tricuspid atresia. These infants were banded more loosely at 24mm + 1mm for each kilogram of infant body weight. Overall mortality was good (13.5%), response to banding was moderate or good in 86% and subsequent pulmonary artery pressure was normal in 79%. Group III consisted of infants with mitral atresia, mitral stenosis and a variety of other complex lesions. Control of congestive cardiac failure was only slight or moderate, and early and late mortality was high.

Pulmonary artery banding by this method allows for safe and effective control of congestive cardiac failure and pulmonary hypertension in infants with congenital heart disease and excessive pulmonary blood flow.

*By Invitation

6. Valvotomy for Pulmonary Atresia with Intact Ventricular Septum: A Disciplined Approach to Achieve a Functioning Right Ventricle

ADNAN COBANOGLU*, MARK T. METZDORFF*,

C. WRIGHT PINSON*, GARY L. GRUNKEMEIER*

and ALBERT STARR

Portland, Oregon

During the past 20 years, a consistent policy in applying early valvotomy has resulted in a unique opportunity to appraise the long-term results of this approach in pulmonary atresia with intact ventricular septum (PA:IVS). Since 1964, 27 of 35 patients with PA:IVS had Type 1 or 2 right ventricle (RV); 25 of these had early valvotomy, 7 with and 18 without concomitant shunt. The remaining 2 patients with Type 2 RV and the 8 patients with Type 3 RV were treated with shunt alone. The overall operative mortality was 34%; those patients weighing more than 3 kilograms and those operated since 1977 had operative mortalities of 18% and 16% respectively. There were 17 survivors of early valvotomy: 12 had valvotomy alone and 5 valvotomy with shunt; 12 had Type 1 RV and 5 Type 2 RV. Survival (± standard error) for these 17 patients was 85(± 10)% and 68(± 17)% at 5 and 10 years respectively. The probability of reoperation was 100% by 6 years of age, with outflow patch reconstruction successfully employed in all cases where reoperation has been performed. Aggressive follow-up and early recatheterization were essential features of management. Delayed reconstruction after shunt alone was unsuccessful in 3 patients, 2 with Type 3 RV and 1 with Type 2 RV.

Primary direct valvotomy without shunt is the operation of choice for patients with PA:IVS and Type 1 RV. Concomitant shunt may be required for some Type 1 and most Type 2 RV patients, selected preoperatively by angiography or postoperatively by clinical necessity. Delayed right ventricle reconstruction after shunt alone is not an acceptable approach when an outflow tract is present.

11:30 a.m. Presidential Address - Grand Ballroom - Third Floor

A Laboratory For Progress

DWIGHT C. McGOON

12:15 p.m. Adjourn for Lunch - Visit Exhibits

*By Invitation