WEDNESDAY MORNING, April 30, 1986

8:30 a.m. SCIENTIFIC SESSION - Grand Ballroom

32. Fontan Type Operation for Complex Lesions: Surgical Considerations to Improve Survival

SERAFIN Y. DELEON*, MICHEL N. ILBAWI*,

FAROUK S. IDRISS, ALEXANDER J. MUSTER*,

SAMUEL S. GIDDING*, TERESA E. BERRY*

and MILTON H. PAUL*

Chicago, Illinois

Twenty-two of 44 patients who underwent the Fontan type operation (FO) had complex lesions other than tricuspid atresia with ventriculo-arterial concordance. Their age ranged from 2 to 20 years. Nine had single ventricle, 5 had tricuspid atresia with transposition, and 8 had various other lesions. Two groups were identified. Group I had pulmonic stenosis with (N = 6) and without N = 9) previous systemic-pulmonary artery shunt. Group II (N = 7) had pulmonary artery banding. Three patients had significant subaortic stenosis. Twenty-six palliative operations, including 6 preliminary Glenn shunts, were performed prior to the FO. The mean pulmonary artery pressure and vascular resistance ranged from 10 to 26 torr and .5 to 2.9 u.M² respectively. Direct RA-to-PA anastomosis was carried out in 18 patients (7 with valve) and valved conduits were used in 4 patients. Eleven patients had right atrioventricular (A-V) valve patch closure (3 running, 8 interrupted suture technique).

In Group I, 40% developed significant pleural and/or pericardial effusion compared with 100% in Group II (p<.02) and the average length of hospital stay was 20 days and 45 days respectively. Significant patch disruption occurred in 3 patients, 2 of whom had the running suture technique. Two patients developed late conduit occlusion in whom the Glenn shunt was lifesaving. Four patients (3 from Group II and all with subaortic stenosis) required Fontan takedown (FT) for persistent low cardiac output (PLCO), 2 of whom died (2/22, 9%). None of the patients with established Glenn shunt required FT. Twenty patients including the 2 patients who survived FT were followed over 3.5 years (range 2 months to 6 years) and did well except for 2 late deaths (10%, 4 months, and 4 years later). We believe that early FT in patients with PLCO, interrupted suture technique for A-V valve closure, and a preliminary Glenn shunt in patients with pulmonary artery banding and/or subaortic stenosis can further improve the results with the Fontan operation for complex lesions.

*By Invitation

8:40 a.m.

33. Comparison of Atriopulmonary Versus Atrioventricular Connections for Repair of Tricuspid Atresia

CHUEN N. LEE*, HARTZELL V. SCHAFF*,

GORDON K. DANIELSON, FRANCISCO J. PUGA

and DAVID J. DRISCOLL*

Rochester, Minnesota

Physiologic repair as described by Kreutzer and Fontan has provided excellent palliation for many children with tricuspid valve atresia. This study compares the early and late results of two variations of physiologic repair of tricuspid atresia: (1) direct atriopulmonary (AP) connection without interposed valves and (2) atrioventricular (AV) connection leaving the native pulmonary valve in place and utilizing the pumping capacity of the residual right ventricle. From January 1979 through March 1984, 50 pts had AP connection and 25 pts had AV connection during physiologic repair. Preoperative characteristics including age, severity of disability, and presence of cardiomegaly were similar. Pts with AP connection had fewer Glenn shunts (14% vs 44%, p<0.004) and a higher incidence of previous Waterston shunts (42% vs 16%, p<0.024) than pts with AV connection. Operative results in the two pt groups were similar, including cardio-pulmonary bypass time, aortic cross-clamp time, postoperative morbidity, and operative risk (30-day mortality 6% for AP connections vs 8% for AV connections). The only significant difference postoperatively was higher mean right atrial filling pressure in the AP group (18.4 mm Hg vs 16.3 mm Hg, p=0.006) and the more frequent need for reoperation in the AV group (28% vs 6%, p=0.013). Overall survival through 107 pt-yrs was 86% after AP connections versus 80% after AV connections. We conclude that there are no clinically important differences in the early results of physiologic repair of tricuspid atresia using AP versus AV connection. The choice of either technique should be dictated by the anatomy of the great arteries at the time of repair.

*By Invitation

8:50 a.m.

34. Extending the Limits for Modified Fontan Procedures

JOHN E. MAYER, JR., RICHARD A. JONAS*,

ALDO R. CASTANEDA, HRODMAR HELGASON*

and PETER LANG*

Boston, Massachusetts

A variety of physiologic and anatomic limits have been proposed by Fontan's group as predictors of the outcome after atrio-pulmonary anastamotic (APA) procedures, but how far these limits can be extended with successful oucome remains unclear. Since a large number of complex cyanotic defects are physiologically correctable, only by APA procedures, we have extended these limits in the areas of age, systemic venous or pulmonary venous anomalies, pulmonary artery anatomy, and pulmonary hemodynamics. One or more of these limits was exceeded in 89 of a total of 148 patients undergoing APA procedures for tricuspid atresia (59) or more complex defects (89). The age limit of 4-15 years was exceeded in 65 patients, 41 under 4 (range 4-47 months) and 24 over 15 (range 16-30 years). Mortality was 34% under 4 years, 8.3% over 15 years, and 20% for the 4-15 year group. In the <4 group where other anatomic and physiologic limits were not exceeded, mortality was reduced to 22%. If indexed pulmonary arteriolar resistance (PAR) was ≤1.5 Woods units, 10 of 11 under 4 years survived, but if PAR was >1.5 units, only 8 of 19 survived. Abnormal pulmonary or systemic venous connections occurred in 11, and 9 survived APA procedures combined with re-arrangement of venous drainage. Pulmonary artery distortion requiring reconstruction occurred in 21 patients, and 9 died (42%). Five of these deaths occurred in patients with PAR >2 units. The pulmonary artery pressure limit of 15 mmHg was exceeded in 25 patients, and 17 survived (68%). Of these, 19 had PAR <2 units and 17 survived (89%). All 6 patients with PAP >15 mmHg and PAR >2 units died. Late follow-up of the 64 survivors exceeding one or more criteria shows 40 in Class I, 11 in Class II, 3 late deaths, and 10 lost to follow-up. We conclude that age, anomalies of pulmonary or systemic venous drainage, pulmonary artery distortion, or elevated pulmonary artery pressure do not individually preclude satisfactory outcome after APA procedures. Criteria for selection of patients under 4 years should be more stringent than older patients, particularly regarding calculated pulmonary arteriolar resistance.

9:00 a.m. Discussion of the Fontan Procedure - Francis M. Fontan, Bordeaux, France

9:45 a.m. Intermission - Visit Exhibits - Exhibit Hall

*By Invitation

10:30 a.m. Scientific Session - Grand Ballroom

35. A New Technique for Repair of Aortic Coarctation: Subclavian Flap Aortoplasty with Preservation of Arterial Blood Flow to the Left Arm

MILTON A. MEIER, FERNANDO A. LUCCHESE*,

WALDIR JAZBIK*, IVO A. NESRRALLA* and

JOSE TELES MENDONCA*

Rio de Janeiro, Porto Alegre and Aracaju, Brazil

Sponsored by: JOHN W. KIRKLIN, Birmingham, Alabama

Between 1967 and 1976, 106 children survived Mustard procedure for transposition of great arteries (TGA). Fifty-five had simple and 51 complex TGA. Their age ranged between 2 weeks and 117 months (median = 12). Ninety-five patients (pts) had 121 previous palliative procedures. Follow-up data up to 18 years (mean = 10.9 ± 3.3) is available on 98 (92.5%) pts. Late death occurred in 8 pts (1 simple, 7 complex TGA) between 2 months to 13 years (mean = 4.6). Sudden death occurred in 4 pts., 2 died of congestive heart failure, one of pulmonary vascular disease and one of tricuspid re-gurgitation. Late Sinus node dysfunction occurred in 31 pts, 6 of these required permanent pacemakers. Of these 31 pts, 17 had unroofing of coronary sinus, while in 14, it was left intact. Postoperative cardiac catheterization showed mild superior vena cava obstruction in 4, mild pulmonary venous obstruction in 3 and baffle leak in 4 pts. Of these, one patient underwent reoperation elsewhere for a baffle leak. Five pts. developed new subpulmonic obstruction, two of which required resection and myotomy. Tricuspid regurgitation occurred in 2, one underwent reoperation. The actuarial survival for 10 years is 93 ± 2.4% and for 18 years is 91 ± 2.3% (95% confidence interval). The probability of event-free survival at 18 years is 82 ± 3.8% (number at risk = 69, events being death, reoperation and insertion of permanent pacemaker). Eighty-seven pts. are in NYHA class I and 3 in class II.

Conclusions: (1) Long-term and event-free survival has been satisfactory. (2) Late death was signficiantly higher in complex TGA (P=.027). (3) Sinus node dysfunction was common, but was not related to the unroofing of coronary sinus (p=NS). Of these, only 6 pts. required permanent pacemakers. (4) The incidence of late complications and reoperation was low.

10:40 a.m. Discussion

*By Invitation

10:50 a.m.

36. The Arterial Switch Operation: An Eight-Year Experience

JAN M. QUAEGEBEUR *, JOHN ROHMER *,

JAAP OTTENKAMP*, TIJIK BUIS*,

JOHN W. KIRKLIN, EUGENE H. BLACKSTONE

and A. GERARD BROM

Leiden, The Netherlands and Birmingham, Alabama

Sixty-six patients (23 neonates with transposition (TGA) and intact ventricular septum (VS), 33 infants and children with TGA and large ventricular septal defect (VSD), and 10 with the Taussig-Bing heart) have received an arterial switch operation as a routine since 1977. One with TGA and intact VS, 6 with TGA and VSD, and 1 with the Taussig-Bing heart died in the hospital. Including hospital deaths, 11 month overall actuarial survival was 81%, and no deaths occurred subsequently among the 33 patients traced thereafter for as long as 8 years. The hazard function for death had only a single early phase, and its confidence limits overlapped the hazard function of a matched general population within 12 months of operation. Incremental risk factors for death included low birth weight (but not weight or age at operation), TGA with large VSD, the Taussig-Bing heart, and presence of a patent ductus arteriosus (PDA). Coronary artery morphology and position of the great arteries were not risk factors. Earlier date of operation was a risk factor (P=.004), and among 27 patients operated upon since April 1984 there have been no early or late deaths; 1 year survival, including hospital deaths, after operation in 1985, predicted from the multivariate equation, is 99.9% (CL 98%-100%) for neonates with TGA and intact VS, and 99.6% (99%-99.9%) for those with TGA and large VSD or the Taussig-Bing heart. The late functional results were excellent, and the rhythm was sinus in 96% of the 55 surviving patients.

A formal comparison was made of these results with those of the atrial switch and the intraventricular tunnel repair, which indicates that the arterial switch repair is superior.

11:00 a.m. Discussion

*By Invitation

11:10 a.m.

37. Long-Term Evaluation of Homograft and Heterograft Valves in Extracardiac Valved Conduits

JAROSLAV STARK*, PAVEL HORVATH*,

RUI ALMEIDA *, WALTER MERRILL*,

JAMES F.N. TAYLOR *, CATHERINE BULL*,

FERGUS J. MACARTNEY* and MARC R. DE LEVAL*

London, England

Sponsored by: JAMES R. MALM New York, New York

Long-term results of 249 children, who received extracardiac conduits between 1971 and December 1982 were evaluated. Follow-up ranged from between 3 and 13 years. Major diagnostic groups were Truncus Arteriosus (TA) 70, Transposition of the Great Arteries (TGA)/Double Outlet Right Ventricle (DORV) complex 64, Fontan type operation 37, Pulmonary Atresia (PA) and Ventricular Septal Defect (VSD) 29, Congenitally Corrected TGA (CTGA) 22 and miscellaneous 27. Aortic homografts (AH) preserved in nutrient and antibiotic solution or deep frozen were used in 108, porcine heterografts (PH) in 107, pericardial valves in 30 and valveless conduits in 6. Hospital mortality related to the age of the patient, complexity of the lesion and severity of the pre-operative condition. Hospital mortality was 77% in the first 3 months of life, 33% between 1-4 years and 23% over the age of 4 years. Status of survivors at the last follow-up was good in 92.5% and fair in 7.5%. Late results were evaluated actuarially. Probability of survival at 5 and 10 years (hospital deaths excluded) for the major diagnostic group was: TGA/DORV 82%/50%, PA + VSD 90%/78%, TA 78%/70%. Probability of survial with the original conduit for the same diagnostic group at 5 and 10 years was TGA/DORV 70%/45%, PA + VSD 85%/25%, TA 62%/20%. At 5 years there was no difference in "conduit survival" between AH and PH. At 10 years 50% of AH patients were alive with their original conduit but only 30% of PH patients. Only 2 AH calcified and became stenotic. Cause of obstruction in AH were related to peel formation in dacron extension and anastomotic strictures. We conclude that AH conduits used without dacron extension are preferable to PH in children. Collagen or fibrinogen presealed dacron tubes may reduce peel formation, but our experience with those techniques is too short for meaningful evaluation.

11:20 a.m. Discussion

*By Invitation

11:30 a.m.

38. Management of Pulmonary Atresia with Intact Ventricular Septum

JOHN E. FOKER, ELIZABETH A. BRAUNLIN*,

JOHN A. ST. CYR*, DAVID HUNTER*, JAMES H. MOLLER*,

W. STEVES RING* and J. ERNESTO MOLINA

Minneapolis, Minnesota

Newborns with pulmonary atresia with intact ventricular septum may also have suprasystemic right ventricular (RV) pressures, tricuspid valve insufficiency, and hypoplastic RVs and pulmonary arteries (PAs). The complexity of these lesions is reflected in the variety of surgical approaches taken and the high mortality rates reported. The purpose of this report is to document the advantages of early RV decompression by a RV outflow tract patch (RVOP) and a variable approach to the provision of adequate pulmonary blood flow (PBF). Fifteen neonates, 1-3 days of age, with this diagnosis and suprasystemic RV pressures were placed on prostaglandin E, (PGE₁) prior to surgery to improve PBF. Only 2/15 could be predicted to have satisfactory postoperative RV mechanics and had an RVOP and ductus ligation. For 9/15 a RVOP was placed but diminutive RV size and/or tricuspid incompetency precluded ductus ligation. PGE, was infused postoperatively (average 6.8 days) to provide PBF in these patients until RV function became adequate. In 3 infants with severely hypoplastic RVs, early RV function was not anticipated, therefore, an aortopulmonary shunt was placed in addition to decompression by a RVOP. One infant, early in series, had only a shunt. Postoperatively, adequate PBF was present in all and 11/15 (73%) survived. Three deaths (avg. 2.8 days) following RVOP and PGE₁ infusion were due to continued low cardiac output, not primary hypoxia. One neonate with a RVOP and shunt died from myocardial ischemia because of coronary artery steal through RV sinusoids. One late death occurred in the child with only a shunt, presumably due to an arrhythmia. The remaining survivors (10/15, 61%)are alive and completely repaired. These patients have revealed that adequate RV size requires more liberal definition. Four patients with a residual atrial septal defect (ASD) and a right to left shunt at age 1-3 years, were found by angiography to have a tricuspid valve diameter from 2.2-4.1 SE below the expected mean and RV volumes below the fifth percentile. Nevertheless, balloon occlusion of the ASD revealed the RVs were functionally adequate and the ASDs (in two, shunts were also closed) could be repaired. In summary, all survivors are corrected in contrast to reported studies in which many are functioning with shunts. Certain unifying treatment principles have emerged: (1) RVOP can be performed in the neonatal period with good results and maximizes the potential for RV growth by relief of RV hypertension; (2) In most patients, PBF can be assured for at least 1-2 weeks with PGE, infusion. RV function usually improves and only rarely is shunt placement required. (3) In cases of apparent inadequate RV size in follow-up, RV function can be assessed preoperatively by balloon occlusion of the ASD and, if tolerated, will be satisfactory after ASD closure.

11:40 a.m. Discussion

*By Invitation