TUESDAY AFTERNOON, APRIL 19, 1988

1:45 p.m. Scientific Session - Ballroom

22. Survivorship After the Arterial Switch Repair for Transposition

WILLIAM I. NORWOOD, ANTHONY R. DOBELL,

JAMES E. LOCK*, JOHN W. KIRKLIN,

EUGENE H. BLACKSTONE and THE CONGENITAL

HEART SURGEONS SOCIETY

From 20 cooperating institutions

Four hundred sixty-six neonates, less than 15 days of age and with transposition of the great arteries (TGA) of all types, have been entered into an ongoing 20 institution study. Among these, 212 with simple TGA or TGA and ventricular septal defect (VSD) have undergone an arterial switch repair and are the basis of this report. 187 (67%) had the usual coronary anatomy, 34 (18%) had origin of the circumflex as well as the right coronary artery from sinus 2 (posterior sinus), and 16 patients (9%) had all coronary arteries arising from a single sinus. Survivorship did not correlate with the coronary anatomy.

Survivorship of the patients varied among the 16 institutions performing the arterial switch repair (4 others did none). Among the 6 "low risk" institutions, the one week, one month, one year, and 2½ year survivorships of patients with simple TGA (n = 114) were 96%, 94%, 91%, and 90% respectively. Among the total of 155 patients undergoing an arterial switch repair in the "low risk" institutions, only older age at repair in the case of simple TGA and TGA with ventricular septal defect were risk factors for death after repair. The survivorship for one month predicted from the multivariate equation for simple TGA was 94% when the repair was done at 6 days of age and 71% when done at 20 days of age.

Overall, among the 212 patients the one week, one month, one year, and 2!/2 year survivorships were 82%, 80%, 79%, and 78% respectively. Birth weight (smaller), associated cardiac or noncardiac anomalies, age at repair (older) in the case of simple TGA, and total elapsed repair time (longer) were risk factors for death after repair. The prevalence of the risk factors was no different in the "low risk" institutions compared with all others (all P-values >.5).

The inferences are that in simple TGA the arterial switch repair should be performed as soon after birth as is possible; that the arterial switch repair is as safe for TGA with VSD in the first month of life as later; and that, in "low risk" institutions, protocols leading to the arterial switch repair provide early and intermediate term survivorships which are better than those of protocols leading to an atrial switch repair.

*By Invitation

23. The Infant Mustard Procedure (100 Days): 10 Year Follow-Up

KEVIN TURLEY, FRANK L. HANLEY*,

EDWARD D. VERRIER*, SCOTMERRICK*

and PAUL A. EBERT

San Francisco, California

In 1978, at the American Association for Thoracic Surgery, a unique group of infants were presented with D-Transposition of the Great Arteries and Intact Ventricular Septum (dTGA - IVS) operated < 100 days of age, using the Mustard procedure. As true infant repairs, such patients represented the benchmark against which subsequent techniques applied to infants must be compared.

During the period 1975 -1981, 36 infants, < 100 days of age with dTGA -IVS underwent Mustard repair at the University of California, San Francisco. Ages ranged from 4-98 days (mean 50 days) and weights from 2.3 - 6.6 kg (mean 3.6 kg). There were no early deaths (operative 1 year). Follow-up 7 -13 years (mean 10 years) postoperatively was performed by direct contact of the cardiologists and echocardiographic evaluation in all 36.

Late survival is 94%, mean 10 years. There were two late deaths, one accidental drowning, and one procedure-related due to ventricular fibrillation. No neurologic complications have occurred. Rhythm is normal sinus in 22, with abnormalities in 12, minor atrial 7, major atrial 1, pacemaker 2, and premature ventricular contractions in 2. Reoperations were performed in 3 for superior vena caval obstruction, 2 of whom required pacemaker while no inferior vena caval or pulmonary venous obstructions have occurred. Finally, serial echocardiographic evaluation has failed to reveal deterioration of right ventricular function.

The Mustard procedure performed in the first 100 days of life results in a high-early - 100%, and 94% - late survival at mean 10 years (one procedure-related death), and a surprisingly low incidence of late complications, against which other techniques of infant repair bear comparison.

*By Invitation

24. Infant Orthotopic Heart Transplantation

CONSTANTINE MAVROUDIS, JON B. KLINE*,

HAROLD L. HARRISON*, LAMANA. GRAY, JR.,

BRIAN L. GANZEL* and SAMUEL R. WELLHAUSEN*

Louisville, Kentucky

Infant OHT has been recently applied to various forms of congenital heart disease with encouraging short term results. From June 1986 to September 1987 we evaluated 16 infants for OHT. Fourteen had hypoplastic left heart syndrome (HLHS), one had endocardial fibroelastosis with aortic atresia and one had anomalous pulmonary artery origin of the left main coronary (ACA). Eight families accepted the treatment program and 8 families refused (2 because of infant associated anomalies and 6 on philosophical grounds). Of the 8 families who accepted, one died 6 hours after diagnosis, one was allowed to die after 60 days due to acquired neurologic complications and the other had congenital CMV. The remaining 5 patients (4 HLHS, 1 AGA) had OHT.

OHT was performed with PDS absorbable suture using deep hypothermia and circulatory arrest in 4 neonates for HLHS (average time 47 min.) and bicaval cannulation and continuous bypass in one eleven month infant for ACA. In-house retrieval was used in all. One neonate died from complications due to pre-transplant donor heart dysfunction and size discrepancy while the remaining 3 neonates and one infant survived and are home 15 months (HLHS), 10 months (ACA), 5 months (HLHS) and 1 month (HLHS) postoperatively. Triple drug immunosupression included cyclosporine, azathioprine and prednisone.

Rejection was diagnosed by clinical evaluation of child activity and monocyte/cell cycle analysis from peripheral blood samples without myocardial biopsies. Routine echocardiography, EKG, and chest x-ray were not helpful. Six episodes of rejection were successfully treated in 3 patients. Twelve month post-op cath in one patient (HLHS) showed appropriate graft growth, no aortic or PA anastomotic strictures, normal RV and LV function, and no coronary artery disease. All four children are normally achieving their growth and personality milestones.

We conclude that while infant OHT is still an investigational procedure, our experience suggests that it can be applied to severe forms of congenital heart disease with excellent short term results.

2:45 p.m. Intermission - Visit Exhibits

*By Invitation

3:30 p.m. Scientific Sessions - Ballroom

25. Seven Years' Experience with the Pierce-Donachy Ventricular Assist Device

D. GLENNPENNINGTON, KIRKR. KANTER*,

LA WRENCE R. McBRIDE*, GEORGE C. KAISER,

HENDRICK B. BARNER, LESLIE W. MILLER *

KEITH R. NAUNHEIM* and VALLEE L. WILLMAN

St. Louis, Missouri

Of the current methods for circulatory support, the Pierce-Donachy external pneumatic ventricular assist device (PD-VAD) has proved to be one of the most versatile and effective. In order to define the current role of the PD-VAD, we reviewed the records of 41 patients (pts)aged 15-71 yrs (mean 53 yrs) with shock refractory to drugs and intraaortic balloon (IABP) who were supported with PD-VADs from 1981-1987. Of 26 pts who received a PD-VAD for postcardiotomy shock, 8 had biventricular (BVAD), 12 had left ventricular (LVAD) and 6 had right ventricular (RVAD) and IABP support. Flows of 2.3-4.8 liters/min (mean 3.6) were maintained for 3-312 hrs (mean 83). Eight pts died in the operating room, 14 were weaned and 10 (38.5%) are long-term survivors. Of 19 pts with biventricular failure, 0 of 5 with univen-tricular support survived while 6 of 14 with biventricular support survived. Perioperative myocardial infarction (MI) occurred in 12 of 16 nonsurvivors and in only one of 10 survivors (p<.01). Other causes for death were bleeding, renal failure and infection.

Nine pts were supported with PD-VAD as "bridge to cardiac transplant (TX)," five pts with BVAD and four with LVAD. Six pts were supported 8-120 hrs (mean 65) and three pts for 22, 24 and 75 days. Five pts were TX, all of whom are long-term survivors; four pts were denied TX because of complications. Four other pts were supported for 12-144 hrs (mean 81) with PD-VADs for cardiogenic shock after TX. In spite of biventricular support (two BVAD and two RVAD & IABP), none survived. Two pts with acute myocardial infarction (MI) shock were supported for 3.5-8.5 days with BVAD and LVAD. One pt was weaned, but neither survived.

Bleeding occurred in 23/41 pts but was decreased by improved surgical technics and sternal closure. Thrombi found in three PD-VADs resulted in emboli in only one pt and were always due to interruption of PD-VAD flow. Anticoagulants were used only during weaning and prolonged support (>10 days). One pt had a stroke due to LV thrombus. Renal failure requiring dialysis (11 pts) was uniformally fatal. Factors predictive of death were myocardial infarction, renal failure, severe infection and the lack of biventricular support if needed. PD-VAD is a safe, effective device which has provided salvage of pts with postcardiotomy shock and pts requiring bridge to TX. Its effectiveness in pts with shock after MI or failed TX are not yet proven.

*By Invitation

26. Dynamic Cardiomyoplasty. Early Clinical Experience and Preliminary Conclusions

ALAIN F. CARPENTIER,JUAN-CARLOS CHACHQUES*,

PIERRE GRANDJEAN*, PATRICK PERIER*,

SYLVAIN CHAUVAUD* and

SERBAN MIHAILEANU*

Paris, France

Since January 24, 1985, the date of the first clinical case of dynamic cardiomyoplasty (cardiac partial reinforcement or substitution with a stimulated striated muscle), 12 patients with acquired myocardial diseases have been operated on utilizing this technique. There were 11 males and 1 female, aged 18 to 67 years. The causes of the myocardial damage were: ischemia (8), tumor (2), cardiomyopathy (1), valvular (1).

All patients but 2 were in class functional IV and were thought to be contraindications to cardiac transplantation for the following reasons: multi-organ failure (4), age (3), diabetes (3), renal insufficiency (2), extensive cardiac tumor (2), pulmonary hypertension (1), social consideration (1). The two other patients had a giant left ventricular aneurysm for which a cardiomyoplasty was judged preferable to transplantation. Ejection fraction ranged from 9 to 24% (mean 14%). Mitral valve incompetence and tricuspid valve incompetence were present in 2 patients.

A dynamic Cardiomyoplasty was carried out using the pediculed left latissimus dorsi transposed into the thorax through a window made in the chest by removing the second left rib. The muscle was used either to reinforce the whole heart (9) or as a myocardial substitute to repair a large defect after removal of an extensive tumor (1) or aneurysm (2). Associated procedures were: subtotal resection and plastic repair of the diaphragme (1), prosthetic ring mitral valve repair (1), prosthetic ring tricuspid valve repair (1), coronary artery bypass (1). 6 patients required circulatory assistance, either post-operatively (4) or both pre and post-operatively (2), using intraaortic balloon conterpulsation (6) and a Pierce-Thoratec artificial left ventricle (1). The programmed sequential stimulation (AC) was begun 15 days after operation, to allow for complete fusion of the muscle to the heart, and was completed in an average of 8 weeks.

Five patients died in the post-operative period, the causes of death were: cardiac failure (3), ventrivular tachycardia (1), renal failure (1). All patients but one had multiorgan failure prior to the operation. The average age was 61 years (42-67).

The 7 patients who survived the operation are in functional class I (1), or II (6). The first patient, operated upon 2 years and 8 months ago in desperate condition is living a normal family life as a mother of two children with no medication. Post-operative Echo-Doppler and 99 MTC heart Scan performed in all patients showed an average ejection fraction of 13.6% without stimulation and 20.6%with stimulation (P<0.01).

This early experience calls for confirmation by a larger number of cases and a longer follow-up. It nevertheless demonstrates that dynamic car-diomyoplasty was effective in all surviving patients after the electrical stimulation has been instituted. This technique may prove to be helpful in the future whenever cardiac transplantation is contraindicated or as an alternative to cardiac transplantation in carefully selected cases.

*By Invitation

27. Current Status of Surgery for Bullous Emphysema

JOHN E. CONNOLLY and ARCHIE F. WILSON*

Irvine, California

Bullous emphysema is the term used to describe patients with acquired emphysema and bullae of the upper lung fields. These bullae are usually bilateral, appear as cotton candy-like multiloculated cystic sacs, and commonly progressively compress relatively normal lower lobes, resulting in severe pulmonary impairment.

Careful selection of patients whose upper and middle lobe bullae severely compress uninvolved lower-lobe tissue permits excision of the bullous tissue with re-expansion of the compressed lower lobe tissue.

Over the past 20 years the authors have had experience with 19 such carefully selected patients. All were severely symptomatic. The majority of these patients required staged bilateral operations with intervals between operations ranging from one month to 11 years. Evidence of compression of relatively normal lower lobe tissue was sine qua non for operation. Selection was initially based on preoperative bronchograms and pulmonary angiography. However, currently tomography and radioisotope lung scanning have been found to be adequate to identify and quantitate compression.

Conservation of all uninvolved lung with excision of only destroyed tissue was carefully adhered to. In no case was lobectomy employed. The use of the stapler, multiple chest tubes, and careful ventilatory support postoperative resulted in no mortality. While prolonged postoperative ventilatory support was often required, all patients ultimately benefited from surgery, in some case dramatically so, with sustained improvement as long as over 11 years postoperatively. Objective pulmonary function data will be presented to document the effectiveness of operation in carefully selected patients. Although these cases represent only a minority of the emphysematous population, patients with bullous emphysema and compression can significantly be improved by appropriate surgical resection.

*By Invitation

28. Diffusing Capacity Predicts Morbidity and Mortality Following Pulmonary Resection

MARK K. FERGUSON*†, LOUISE R. LITTLE*,

DEBORAH MANJONEY*, ALAN LEFF*,

KEITH J. POPOVICH* and ALEX G. LITTLE

Chicago, Illinois

Spirometry and cardiovascular status are routinely used to exclude high risk patients from pulmonary resection. The majority of operative deaths in acceptable candidates are said to be random and unpredictable. We retrospectively reviewed 176 consecutive patients who underwent lobectomy or pneumonectomy to evaluate other pulmonary function parameters as predictors of operative morbidity or mortality. Data on 40 preoperative variables were collected and statistical analyses were performed to determine their relationship to 21 operative and postoperative events. There were 97 males and 78 females with a mean age of 60.2 years. 153 had lung cancer (58 Stage I, 23 Stage II and 72 Stage IIIA) while 22 had benign disease. 59 had pneumonectomy and 116 had lobectomy or bilobectomy. Overall hospital mortality was 8.5%. The incidence of pulmonary complications (PC; identified as ven-tilatory support > 24 hours, need for bronchodilators or reintubation, pneumonia, lobar collapse) was unrelated to spirometric values. PC were significantly influenced by diffusing capacity of the lung for carbon monoxide (DLCO, hemoglobin and volume corrected, expressed as ml/min/torr). Pulmonary complications occurred in 31.4% of patients with DLCO < 20 and 18.1% with DLCO ‰¥ 20 (p < 0.025; 101 patients evaluable). Operative mortality was also importantly influenced by DLCO. Death occurred in 20% of patients with DLCO < 20 and 0% with DLCO ‰¥ 20 (p < 0.025; n = 101). The incidence of pulmonary complications and death was inversely related to DLCO:

DLCO is a measure of pulmonary capillary hemoglobin concentration and is a sensitive indicator of the anatomic integrity of the pulmonary alveolus. As such, it can reveal subtle but important changes in lung structure undetected by spirometry. These data show that DLCO independently predicts the risk of pulmonary complications and death following major pulmonary resection and should be routinely used in the preoperative evaluation of such patients.

AN HISTORICAL VIGNETTE

Lyman A. Brewer, III, Pasadena, California

4:50 p.m. Executive Session (Members Only)

7:00 p.m. President's Reception (Tickets Required)

*By Invitation

†Edward D. Churchill Research Scholarship