MONDAY AFTERNOON, May 7, 1984

2:00 p.m. Forum Session - Grand Ballroom - Third Floor

7. Thoracic Disease in Patients with Acquired Immune Deficiency Syndrome

HARVEY I. PASS*, DOROTHY A POTTER*,

JACK A. ROTH*, JAMES H. SHELHAMMER*,

HENRY MASUR*, CHERYL M. REICHERT*

and ABE M. MACHER*

Bethesda, Maryland

Sponsored by: RICHARD E. CLARK

Bethesda, Maryland

The acquired immune deficiency syndrome (AIDS) is characterized by the development of multiple recurrent opportunistic infections and unusual neoplasms in individuals with no prior history of immune suppression. This report summarizes the role of diagnostic modalities currently used in the evaluation of thoracic disease in AIDS patients.Efficacy of treatment was then determined by correlation with post mortem findings in all patients. The mean age of 15 AIDS patients with thoracic disease was 39 years, all were male, and risk factors included homosexuality (10 of 15), bisexuality (3 of 15), intravenous drug abuse (1 of 15) and Haitian nationality (1 of 15).

Pulmonary disease (hypoxia, pulmonary infiltrates) was present in all tients necessitating 23 transbronchial biopsies (TBB) in 11 patients. 65% were diagnostic, with Pneumocystic Carinii Pneumonia (PCP) and cytomegalovirus (CMV) pneumonia being the most common findings. Nine open lung biopsies (OLB) in 8 patients documented either PCP or Kaposi's sarcoma (KS). Esophageal disease, characterized by dysphagia, was found in 4 patients, and endoscopic evaluation demonstrated Candida esophagitis (2), esophageal KS (1), and CMV esophagitis (1).

Mean time to death from diagnosis of AIDS was 7.7 months, with post mortem examination performed in all patients. Although Pneumocystic carinii was the most common antemortem cause of pneumonia in our AIDS patients, it was rarely found at autopsy as PCP was successfully eradicated in 70% of the patients treated. At autopsy, four additional cases of unsuspected pulmonary KS were discovered as well as nine cases of CMV pneumonia (the most common cause of death, 10 of 15, 67%). Esophageal disease documented at autopsy included: CMV esophagitis (3), KS (2), and Candida (1).

Our data reveal that PCP and Candida esophagitis can be successfully treated in AIDS patients if appropriately diagnosed. The major cause of mortality is pulmonary insufficiency, often due to severe CMV infection. Thoracic surgeons must continue to play an aggressive and important role in the early diagnosis and management of potentially treatable pulmonary and esophageal disease in these patients.

*By Invitation

8. The Effect of Mid-Sternal Thoracotomy on Pulmonary Function in Patients Undergoing Coronary Revascularization

NADIV SHAPIRA *, DA VID M.F. MURPHY*,

SALVATORE M. ZABBATINO*,

DANIEL B. SULLIVAN*, NABIL KHALIL*

and GERALD M. LEMOLE

Browns Mills, New Jersey

It is generally assumed that non-complicated cardiac operation performed through a midsternal incision is associated with only minimal impairment of pulmonary function (PF). In order to verify this assumption, PF was studied in 30 consecutive male patients (age range 42 - 71 yr) undergoing surgery for coronary artery disease. Lung volumes, flow rates, diffusion capacity, and arterial blood gases were determined prior to surgery, at discharge (avg 8.7 ± 1.9 d), and at 3 months. In addition, peak flow was obtained immediately after extubation. Seventeen patients had smoking history, but none had history of pulmonary dysfunction. All underwent non-complicated coronary revascularization (4.4 ± 1.8 grafts/patient) performed by one surgeon. The left internal mammary artery (LIMA) was utilized in 24 patients. The left pleura was inadvertently opened in 14 and a left chest tube was left in in 6. Aortic cross-clamp time was 68 ± 27 min and pump time was 119 ± 43 min. The changes in each parameter were analyzed by paired t test and expressed as mean ± S.D. (*p.05).

Pleural invasion or LIMA dissection did not affect the results. None had abnormal PF preoperatively. These data indicate that a significant impairment in lung volumes, flow rates, and oxygenation - but not in diffusion capacity - do occur in association with a midsternal thoracotomy in patients undergoing coronary revascularization. Return to preoperative values was observed in 3 months.

*By Invitation

9. Use of T-lymphocyte Analysis in Early Diagnosis of Adult Respiratory Distress Syndrome

MARTIN L. DALTON and CARL S. RIGBY*

Jackson, Mississippi

Adult respiratory distress syndrome (ARDS) afflicts an estimated 150,000 Americans annually with a mortality rate in excess of 50%. Although frequently associated with trauma, sepsis and shock, the precise etiology remains obscure. Certainly the role of the immune system in the development of ARDS needs further definition. Cell-mediated immunity via T-lymphocytes has been shown to be severly depressed following trauma, major surgical procedures and shock. These conditions, plus sepsis, are the usual precursors of ARDS.

We have studied by monoclonal antibody staining and flourescence-activated cell sorting (FACS) twelve massively traumatized patients of the type particularly prone to develop ARDS. In each instance, total T-lymphocytes and subsets of helper T-lymphocytes and suppressor T-lymphocytes were precisely measured at the time of admission, following surgery and on alternate days postoperatively.

All twelve patients have survived and a definite pattern of T-lymphocyte aberration has emerged. Nine patients progressed satisfactorily, with findings of increased helper T-lymphocyte and increased helper/suppressor ratio. Two patients developed delayed infection with no sustained increase in helper T-lymphocyte and little change in the helper/suppressor ratio. One patient developed ARDS preceeded by a marked increase in suppressor T-lymphocytes and a decrease in the helper/suppressor ratio. No other patient in the series has developed ARDS and no other patient has developed this pattern of T-lymphocyte response. Recovery of the patients was accompanied by an increase in helper cells and an increase in helper/suppressor ratio.

We conclude that T-lymphocyte analysis offers a promising means of evaluation of patients considered highly susceptible to ARDS. Certainly, early detection of ARDS would facilitate management and increase survival. It is conceivable that pharmacologic modification of the immune response could become a vital part of the treatment of ARDS.

*By Invitation

10. Perioperative Blood Transfusion Adversely Affects Prognosis after Resection of Stage I (NO) Non-Oat Cell Lung Cancer

PAUL I. TARTTER*, LEWIS BURROWS*

and PAUL A. KIRSCHNER

New York, New York

Recent studies suggest that pretransplant blood transfusions prolong kidney graft survival by non-specific immune suppression. Because immune suppression in cancer patients may be associated with earlier tumor recurrence and shorter survival, we studied the relationship of perioperative blood transfusion to prognosis in 165 consecutive resections of Stage I (NO) non-oat cell lung cancer over the 15 year period 1966 - 1980.

Using life table and Cox proportional hazards analysis two statistically significant prognostic factors emerge: 1) extent of resection (p = 0.0056) and 2) perioperative transfusion (p = 0.0282).

Other factors such as age, sex, tumor size (T1 or T2), histopathology, admission and discharge hematocrit, estimated blood loss, duration of operation and anesthetic agents were not statistically significant.

All were operated by one surgeon to minimize surgeon-related technical variables which may be important determinants of the need for perioperative transfusion.

Inasmuch as only 15 of the 165 patients underwent pneumonectomy, they were eliminated from the final analysis leaving a more homogenous group of 150 lobectomies and lesser procedures. Survival advantage was noted in non-transfused patients (77% versus 68% disease-free at five years, p = 0.0783).

These results indicate that perioperative transfusion during pulmonary resection for lung cancer adversely affects prognosis and accelerates the appearance of recurrent or metastatic disease. This supports the finding of previous studies that perioperative transfusion adversely affects the prognosis for breast and colon cancer patients.

3:00 p.m. Intermission - Visit Exhibits - Second Floor

Complimentary Coffee

*By Invitation

3:45 p.m. Scientific Session - Grand Ballroom - Third Floor

11. The Fate of Arm Veins Used for Coronary Artery Bypass Grafts

WILLIAM S. STONEY, WILLIAM C. ALFORD,

GEORGE R. BURRUS*, DAVID M. GLASSFORD*,

MICHAEL R. PETRACEK* and CLARENCE S. THOMAS

Nashville, Tennessee

Arm veins have been a common second choice as a graft conduit for those patients who have insufficient saphenous veins for coronary bypass operations. To define the durability and late patency of arm vein grafts, we reviewed our patients with one or more arm vein grafts used for coronary revascularization between 1974 and early 1983. A total of 58 patients required at least one arm vein graft and 50 patients are presently alive. Arm veins were used because of prior bilateral saphenous vein stripping in 39 (67%) patients, sclerotic or thrombosed vein in 7 (12%), prior use of saphenous veins in 6 (10%), varicosities in 5 (9%), and unknown in 1 (2%).Postoperative arteriograms were obtained in 25 of the 50 patients at an average of 21 months after the operative procedure. These 25 patients had received a total of 51 arm vein grafts.a nd restudy showed that 24 (47%) grafts were patent with no abnormality, 22 (43%) were completely occluded, and 5 (10%) were patent but with a significant area of stenosis in the graft. Seventeen internal mammary artery grafts were also used in this same group of patients. Fifteen of these were investigated and 14 (93%) were patent, one with poor flow. For comparison, 28 additional patients with saphenous vein grafts were reviewed who had had a recent restudy arteriogram for symptoms. The time interval between operation and restudy for this group was 46 months. A total of 57 saphenous vein grafts were used and 42 (74%) were patent with 15 (26%) occluded. From this study, we conclude that arm vein grafts have a high failure rate of 43% (22/51) and are not as dependable as saphenous vein grafts or internal mammary artery grafts.

*By Invitation

12. Late Surgical Results for Ischemic Mitral Regurgitation: Role of Wall Motion Scores and Severity of Regurgitation

C. WRIGHT PINSON*, ADNAN COBANOGLU*,

MARK T. METZDORFF*, GARY L. GRUNKEMEIER*,

PHILIP KAY* and ALBERT STARR

Portland, Oregon

The indication for concomitant valve surgery for ischemic mitral regurgita-tion (IMR) is examined in 120 consecutive patients with IMR who had coronary bypass surgery (CBS) since 1970. IMR was mild in 67%, moderate in 21%, and severe in 32%. IMR patients compared with 3334 CBS patients without MR, had significantly more cardiomegaly (31% vs. 5%), left heart failure (LHF) (42% vs. 6%) and abnormal CASS wall motion scores (WMS) (71% vs. 42%).

Eighty-six IMR patients (72%) had CBS alone and 34 (28% had concomitant valve surgery. No patient with mild IMR as compared to 24% with moderate and 91% with severe IMR had valve surgery.

Operative mortality for mild, moderate, and severe IMR was 5%, 10%, and 38%; 5-year survival was 82%, 60%, and 47%, respectively. Other significant determinants of survival were cardiomegaly, LHF, type of operation, and WMS. Patients with mild IMR and WMS of 5-10 (n = 40) had 5-year survival of 93%, comparing favorably with CBS patients without MR. For patients with either moderate/severe IMR (n = 27) or WMS of 11-20 (n = 25), 5-year survival was 70%, while in patients with both high wall motion scores and moderate/severe IMR (n = 20), it was 45%, demonstrating an additive detrimental effect.

In conclusion, IMR is a major additive risk factor to WMS in CBS. Mild IMR is best managed by CBS alone. CBS alone in patients with moderate IMR and with high WMS yielded poor results, suggesting this subset of patients should have their valves explored. Severe IMR requires concomitant mitral valve surgery.

*By Invitation

13. Primary Myocardial Revascularization: Trends in Surgical Mortality

DELOS M. COSGROVE, FLOYD D. LOOP,

BRUCE W. LYTLE*, CARL C. GILL*,

LEONARD R. GOLDING*, PAUL C. TAYLOR*

and MARLENE GOORMASTIC*

Cleveland, Ohio

From 1970-1982, 24,672 patients underwent primary isolated myocardial revascularization: Group I, 4,517 patients operated on from 1970-1973; Group II, 6,181 patients 1974-1976; Group III, 6,869 patients 1977-1979; Group IV, 7,105 patients 1980-1982. The operative mortality rate was 1.2% for the entire experience and 1.2%, 1.4%, 1.6% and 0.8% for Groups I-IV respectively. The mortality rate for Group IV was significantly lower, p<0.001.

Clinical, angiographic and operative variables were analyzed for operative risk factors using univariate analysis. All univariately significant factors were then analyzed using a multivariate logistic regression analysis for the entire experience.

Group I-IV were independently analyzed for variables which increase operative mortality.

Cardiac causes accounted for 203 (66.2%) patients. This has gradually decreased from 75.3% in Group II to 58.5% in deaths of Group IV. Neurologic deficit was the second most frequent cause of death, 29 (9.6%), reaching a high in Group IV (18.9%). The median interval from surgery to hospital death has increased from two days in Group I to eight days in Group IV.

We conclude 1) there has been a significant decrease in operative mortality in the 1980's. 2) Emergency surgery has been the principle risk factor. 3) Left main disease has been neutralized as a risk factor. 4) Death ascribed to cardiac causes is decreasing in incidence.

*By Invitation

14. Sequential Internal Mammary Artery Grafts: Expanded Utilization of an Ideal Conduit

M. LAXMAN KAMATH*, LINDA S. MATYSIK*

and DONALD H. SCHMIDT*

Milwaukee, Wisconsin

Sponsored by: EDWARD F. PARKER

Charleston, South Carolina

Several long-term patency studies have documented the superiority of internal mammary artery grafts (IMAG) over vein grafts (VG) as conduits for the revascularization of the ischemic myocardium. Thus far, each IMAG has been used for anastomosis at one site on the coronary artery limiting the area of revascularization. In an effort to expand the utility of IMAG we have performed sequential anastomoses in 82 patients (pts) over the past 3 years and our experience is presented in this report. Based on the nature of anastomoses the pts are divided into three groups. Pts in group I received single sequential IMAG, group II received one sequential IMAG and one end-to-side IMAG, while pts in group III received two sequential IMAG. In addition pts in each group received VG as needed. Graft patency was evaluated by coronary angiography in 30 pts (36.5%). The results are presented in the table.

*One IMAG could not be injected, patency of two grafts-unknown.

**One Gortex graft, patent.

Functional adequacy of myocardial revascularization was evaluated by exercise stress test in 77 pts (94%) including every pt who had angiography. Among pts who were evaluated for graft patency, stress test was negative in 26 (86.6%) pts, positive in 3 (10%) pts and equivocal in 1 (3%) pts. In the total study population, stress test was negative in 64 (83%) pts, positive in 5 (6.4%) pts and equivocal in 8 (10%) pts. At the time of this reporting 96% of the pts were alive and well.

The results show that patency of sequential IMAG is comparable to the patency rate reported for single end-to-side IMAG. It is also evident that expansion of the area of revascularization using sequential IMAG is technically feasible and the results of stress tests suggest that such anastomoses provide adequate nutrition to the revascularized myocardium.

*By Invitation

15. Coronary Artery Bypass for Unsuccessful Percutaneous Transluminal Coronary Angioplasty

GEORGE J. REUL, DENTON A. COOLEY,

DAVID A. OTT*, J. MICHAEL DUNCAN*,

JAMES J. LIVESAY* and O. H. FRAZIER*

Houston, Texas

Of the 518 consecutive patients undergoing percutaneous transluminal coronary angioplasty (PTCA) for 571 coronary lesions, 184 eventually underwent aortocoronary artery bypass (ACB) because of failure of the procedure. Immediate failure of PTCA resulted in ACB in 157 patients, all operated upon the day of PTCA attempt. Late failure manifested by recurrent symptoms 1 week to 2 years post-PTCA resulted in ACB in 27 patients. Age range was 34 to 79 years (mean, 56.5 years). There were 130 men and 54 women. Complicating factors were previous ACB (16 patients); previous PTCA (12 patients), and pre-PTCA acute myocardial infarction (3 patients). Delayed ACB (another hospital admission) was done in 27 patients (no deaths). Immediate ACB (0-24 hours) because of failure to dilate the vessel or because of complications of the dilatation attempt was done in 87 patients (2 deaths). Emergency ACB (less than 3 hours) was required in 63 patients who were unstable because of unrelieved angina, persistently ischemic EKG, or hypotension (1 death). In the remaining 7 patients, cardiac arrest occurred during PTCA and required immediate cardiopulmonary bypass and emergency ACB (3 deaths). Overall mortality was 3.3% (6/184). More than one graft was required in 49% (90/184), indicating multiple vessel involvement. Despite emergency ACB, evolution to transmural infarction occurred in 26 of 70 (38%) who were either unstable with acute ischema or had cardiac arrest during PTCA. Early mortality of ACB following unsuccessful PTCA (6/184) compared with early mortality following elective ACB during this same period (42/3500) was significantly different (p<0.05). Perioperative infarction following failed PTCA (26/184) was significantly different (p<0.001) when compared to infarction following elective ACB (88/3500). Thus, during PTCA immediately available surgical back-up is imperative, and proper patient selection is essential since there is a high incidence of perioperative infarction and operative mortality following unsuccessful PTCA.

*By Invitation