T ventricle and the pulmonary artery has allowed repair. Extracardiac Valved Conduits in the Pulmonary Circuit

Transcription

1 Extracardiac Valved Conduits in the Pulmonary Circuit Shunji Sano, MD, PhD, Tom R. Karl, MS, MD, and Roger B. B. Mee, MB, ChB, FRACS Victorian Paediatric Cardiac Surgical Unit, Royal Children s Hospital, Melbourne, Australia Extracardiac valved conduits represent one of the weakest facets of reconstructive surgery for congenital heart disease in that they invariably need to be replaced because of growth of the patient or because of valve or conduit failure. Between 1979 and 1989, 141 patients had 169 valved conduits placed between the heart and the pulmonary artery circuit. There were 81 male and 60 female patients, aged 2 days to 35 years (mean age, 5.9 years), with 46 patients less than 1 year of age. We performed primary repair in 117 patients; in this group, there have been 28 conduit replacements in 27 patients. In 17 patients initial repair with a conduit was performed elsewhere and we replaced these conduits in 15 and removed them in 2. A further group of 9 patients were seen after repair of tetralogy of Fallot or double-outlet right ventricle, with severe pulmonary incompetence or right ventricular outflow tract aneurysm. All had valved conduits inserted as secondary procedures. The types of valved conduits used were xenograft (n = 126) and homograft (n = 43). There were six hospital deaths (3.6%; 70% confidence limits [CL], 2% to 6%) and seven late deaths (4.1%; CL, 2.5% to 6.5%) in a total of 169 conduit insertions. Forty-five conduits have been removed and 43 reinserted without early or late mortality (0%; CL, 0% to 4%). Actuarial survival after conduit insertion was 87% at 5 years (CL, 80% to 92%), including operative mortality. Actuarial freedom from conduit replacement was 37% at 5 years (CL, 20% to 56%). Conduit insertion in infants and small children ensures subsequent replacement, but this can be done at low risk. Our policy therefore has been to attempt to match the conduit to the size of the patient rather than to insert a large conduit. Our current approach is to use a Dacron conduit at initial repair to establish a prosthetic sewing margin adjacent to the left main and left anterior descending coronary arteries for safe subsequent replacement, preferably with a cryopreserved homograft. Replacement should be done before development of important right ventricular dysfunction. (Ann Tliorac Surg 1991;52:285-90) he use of an extracardiac conduit between the right T ventricle and the pulmonary artery has allowed repair of pulmonary atresia, truncus arteriosus, transposition of the great arteries with pulmonary stenosis, and other complex forms of congenital heart disease (1-41. Although conduit repair may provide dramatic improvement in clinical status with elimination of cyanosis and considerable increase in exercise tolerance, external valved conduits still represent one of the weak facets of reconstructive surgery for congenital heart disease in that they invariably need to be replaced because of growth of the patient or because of valve or conduit failure [58]. In this report, we review our 10 years experience with valved conduits in the pulmonary circuit. Patients and Methods Between January 1979 and September 1989, 141 patients had 169 valved conduits placed between the heart and the pulmonary artery circuit. There were 81 male and 60 female patients, aged 2 days to 35 years (mean age, 5.9 years), with 46 patients less than 1 year of age. Accepted for publication May 16, 1991 Address reprint requests to Mr Mee, Victorian Paediatric Cardiac Surgical Unit, Royal Children s Hospital, Flemington Rd, Parkville, 3052, Victoria, Australia. One hundred seventeen patients had their initial conduit operation in our own institution. Primary diagnoses for these patients are listed in Table 1. Within this group, 27 of 117 patients have subsequently required 28 conduit changes. The age distribution for this group is shown in Figure 1. Another 17 patients had initial conduit placement elsewhere and were referred to us subsequently. Fifteen had conduit replacement and 2 had right ventricle to pulmonary artery continuity reestablished without use of a conduit. An additional 9 patients had initial intracardiac repairs performed elsewhere without use of a conduit and were referred to us for secondary operation. All 9 initially had undergone transannular patch repair of tetralogy of Fallot or double-outlet right ventricle and had development of severe pulmonary insufficiency or patch-related aneurysms 51 to 176 months (mean, 108 months) later. The type of conduit used varied over the course of this study. Hancock composite Dacron-porcine valved conduits were used most frequently (n = 77). Cryopreserved aortic and pulmonary homografts, harvested and prepared by us according to the protocol of O Brien and associates [9], were next most frequent (n = 43). Other conduits used were Tascon (n = 41), Polystan (n = 4), Ionescu composite Dacron-bovine pericardial valved (n = 3), and nonvalved (n = 1) by The Society of Thoracic Surgeons /91/$3.50

2 ~ 286 SANOETAL Ann Thorac Surg 1991;52:28590 Table 1. Diagnosis and Early and Late Outcome of Conduit Insertion in 11 7 Patients Having Primary Repair at the Royal Children s Hospital Diagnosis No. of Hospital Late Replacement Patients Deaths Deaths of Conduits Truncus, hemitruncus arteriosus Pulmonary atresia VSD TGA + VSD + PS AV-VAD + PS Double-inlet ventricle Other Total AV-VAD = atrioventricular and ventriculoarterial discordance; PS = pulmonary stenosis; TGA = transposition of great arteries; VSD = ventricular septal defect. Follow-up was accomplished by direct contact with referring cardiologists, and included clinical examination, echocardiography, and cardiac catheterization. Follow-up data were obtained for all patients except 7 who reside outside Australia. Mean follow-up time for 131 patients was 41 months (range, 1 to 118 months). Actuarial data were calculated using the Kaplan-Meier formulas. Sample proportions are expressed with confidence limits (CL) corresponding to one standard deviation, with continuity corrections for upper and lower limits. Results Conduit Insertion There were six hospital deaths (3.68%; CL, 2% to 6%) in a total of 169 primary and secondary conduit insertions. All deaths occurred in the group of 117 patients undergoing conduit insertion as part of primary intracardiac repair in our own unit (see Table 1). Therefore, the early mortality within this subgroup was 5% (CL, 2.9% to 7.9%). Of these hospital deaths, one was related to pulmonary hypertensive crisis occurring in a patient with truncus arteriosus. A second resulted from unknown causes in a patient with the same diagnosis. The third was related to pulmonary hemorrhage after repair of pulmonary atresia, ventricular septal defect, and major aortopulmonary collateral arteries. Two patients had support withdrawn due to irreversible noncardiac congenital defects that became apparent after operation, and 1 died of bronchopneumonia 1 month after repair of pulmonary atresia and ventricular septal defect. Thus, we could not identify any technical problems relating to conduit insertion as directly contributory to operative mortality. Nine patients referred to us for conduit insertion subsequent to transannular right ventricle to pulmonary artery patch reconstruction experienced no hospital or late deaths (0%; CL, 0% to 19%). Non-conduit-related operative complications requiring reoperation included complete heart block (n = 3), left bronchial compression (n = 2), and left bronchial plus right pulmonary arterial compression (n = 1). There was also 1 patient who required early conversion to Fontan operation owing to hypoplastic right ventricle and tricuspid stenosis. There were seven late deaths, also occurring exclusively in the group of 117 patients undergoing primary conduit placement in our unit (6%; CL, 5.5% to 6.5%). Of these late deaths, only one was conduit related (Table 2). This occurred in a patient who had right atrium to pulmonary artery conduit placement as part of a Fontan operation and died 98 months later of conduit stenosis and right heart failure. This is the only Fontan operation included in this series, and the only patient in whom we have used an extracardiac conduit as part of a Fontan repair. The actuarial survival for patients requiring extracardiac valved conduits was 87% (CL, 83% to 91%) at 5 years, including operative mortality (Fig 2). Conduit Replacement Forty-three patients, including 28 patients initially operated on at Royal Children s Hospital, have undergone Fig 1. Age distribution of 11 7 patients undergoing conduit insertion as part of a primary intracardiac repair at the Royal Children s Hospital. 60 NUMBER =TOTAL WHD OLD 51 <lmo 1-3mo AGE 1-4yr 14yr

3 Ann Thorac Surg 1991;52:28590 SANOETAL 287 Table 2. Characteristics of Patients Who Died Late After Extracardiac Conduit Insertion Interval After Patient Age at Operation No. Diagnosis Operation Cause of Death (mo) 1 DIV, PS (Fontan) 11 y Obstructed conduit 98 2 Hemitruncus 7 mo Bronchopneumonia 2 3 Truncus 20 d Septicemia 9 4 PA, VSD, 4 mo Tracheobronchial 2 MAPCAs malacia 5 PA, VSD, 10 y Sudden death 25 MAPCAs (arrythmia?) 6 PA, VSD, 9 y Sudden death 27 MAPCAs (arrythmia?) 7 PA, VSD 6 y Congestive heart 18 failure MAPCAs = major aortopulmonary collateral arteries; PA = pulmonary atresia; I'S = pulmonary stenosis; VSD = ventricular septal defect. replacement of conduits 11 to 169 months (mean, 57 months) after initial operation (Table 3). Original diagnoses of the 28 patients initially operated on at the Royal Children's Hospital are summarized in Table 1. One patient with truncus arteriosus required a second reoperation owing to severe pulmonary incompetence from a nonvalved conduit, which was replaced with a 20-mm aortic homograft conduit. All 43 patients having conduit replacement were assessed by cardiac catheterization or echocardiography, or both, and the mean systolic right ventricle to pulmonary artery pressure gradient before reoperation was 52 mm Hg (range, 0 to 107 mm Hg) (Fig 3). Table 4 details the indications for conduit replacement and the sites of conduit obstruction. Four of the 42 patients had associated procedures at the time of conduit change, consisting of aortic valve replacement (n = Z), truncal valve replacement (n = l), and repair of residual ventricular septal Table 3. Conduit Size and Mean Intervals to Replacement for Patients Originally Operated on at the Royal Children's Hospital Xenograft Replacement Homograft Replacement Size Interval Interval (mm) Total N (mo) Total N (mo) (50%) (53%) (24%) (5%) (10%) Total (27%) (19%) defect (n = 1). Mortality for conduit removal, reinsertion, and repair of any associated defects was 0% (CL, 0% to 4%). Actuarial freedom from conduit replacement (or conduit-related late death in 1 patient) for long-term survivors was 37% (CL, 20% to 56%) at 5 years (Fig 4). The interval between insertion and replacement of valved conduits is shown in Table 3. Clinical Status Of the 129 survivors, 7 overseas patients were lost to follow-up. Six patients are mildly to moderately restricted because of elevated pulmonary vascular resistance (n = 2), neurological deficit (n = 2), angina pectoris (n = l), and ventricular dysfunction after aortic valve replacement (n = 1). The remaining 116 patients are alive and in New York Heart Association class I or 11. Comment In 1966, Ross and Somerville [l] reported the repair of tetralogy of Fallot with pulmonary atresia using an aortic PROBABlLl TY I UPPER CL LOWER CL \ ; ;,,, MONTHS AFTER INSERTION nary circuit (including operative mortality). Seventy percent confidence limits (CL) are expressed in broken curves.

4 288 SANOETAL Ann Thorac Surg 1991;52: Fig 3. Systolic pressure gradients (mm Hg) across extracardiac valued conduits in 43 patients undergoing conduit replacement n mm 41-60mm 61-80mm mm >100mm homograft as a conduit from the systemic venous ventricle to the pulmonary artery. In 1973 Bowman and coworkers [2] reported the first clinical use of a glutaraldehyde-preserved porcine valve and a Dacron tube graft. Numerous modifications of these techniques to establish systemic venous to pulmonary artery continuity have been described since, permitting repair of a wide variety of complex congenital heart defects. Currently, conduit placement with low operative mortality is possible even in small infants [lo]. A number of different types of valved conduits have been used, including those with aortic and pulmonary homograft valves, porcine valves, pericardial valves, and mechanical valves placed within Dacron tubes. Early calcific stenosis of irradiated and frozen aortic homograft valved conduits is a widely known late complication and has led to preferential usage in many institutions of heterograft extracardiac valved conduits [5, 6, 11, 121. Although several centers have reported satisfactory early results with the use of heterograft extracardiac conduits, a number of late follow-up studies of these patients have suggested that this conduit as well may Table 4. lndications for Late Reoperation in 43 Patients With Extracardiac Conduits Indication Conduit obstruction Proximal (kinking of the conduit) Valve Distal Valve + distal Valve + proximal Conduit obstruction + residual VSD Conduit obstruction + incompetence Conduit valve incompetence Total VSD = ventricular septal defect. No. of Patients have a significant incidence 1 relatively early obstruction [I, 7, 81. Bisset and colleagues [13] reported a 70% incidence of conduit failure within 6 years, and Jonas and associates [7] reported a 24% reoperation rate at a mean follow-up interval of 67 months. At 10 years, all of the conduits in the latter series had required replacement. The antibiotic-sterilized or cryopreserved homograft valved conduits have performed relatively well but are not readily available in many centers, especially in the smaller sizes [ 14-16]. Fontan and colleagues [ 141 reported that among 25 patients recatheterized at a mean interval of 12 years after homograft insertion for Fallot s tetralogy, only 1 had a gradient greater than 15 mm Hg. Similar excellent results have been reported by Kay and Ross [15]. They reviewed 51 patients with antibiotic-sterilized homografts used to repair pulmonary atresia and ventricular septal defect at a mean follow-up of 8 years. The actuarial probability of conduit obstruction requiring reoperation was 13% at 10 years, and actuarial survival after discharge was 70% at 10 years. However, their patients were older than ours, with a median age at operation of 11 years and a homograft size of 19 mm or greater, making survival comparison problematic. An extensive analysis of conduit operations in 249 children from Great Ormond Street, London, failed to reveal overall differences in performance of homograft and heterograft conduits in the long or short term [8]. Outcome appeared to correlate to the underlying pathological condition for which conduit replacement was required. Lacking an ideal conduit for use in small children and given the improved survival statistics for conduit placement in babies, reoperation for conduit obstruction is now required with increasing frequency. Ebert [ 171 reported that 55 of 86 survivors of truncus arteriosus repair required conduit replacement at a median interval of 4 years. Boyce and co-workers [18] reviewed 42 survivors of 12-mm porcine valved conduit insertion and reported that 28 required conduit replacement at a mean interval of 44 months. In the Mayo clinic experience, mean interval to

5 Ann Thorac Surg 1991;52: SANOETAL 289 PROBABlLl TY 1 2, Fig 4. Actuarial freedom from conduit replacement (or conduit-related death in 2 patient) in 217 patients. Seventy percent confidence limits (CL) are expressed in broken curves MONTHS AFTER IMPLANTATION conduit replacement was 5.4 years but not directly related to age at first operation [6]. Our own experience suggests that the probability of reoperation for conduit replacement will be greater than 60% at 5 years, noting that 40% of our patients were less than 1 year of age at initial conduit implantation. A number of studies have emphasized the importance of serial catheterization in patients with extracardiac conduits because symptoms may be absent despite moderate obstruction [7, 19, 201. Heck and associates [19] catheterized 13 patients 1 year after the Rastelli procedure and found a 57% incidence of serious conduit obstruction despite absence of symptoms in 88%. Stewart and coworkers [20] reported late clinical results for patients with Hancock conduits. Cardiac catheterization demonstrated a gradient of 50 mm Hg or greater across the conduit in 2 of 16 patients 1 year after operation and in 7 of 15 patients 6 years after operation. Nevertheless, all patients were essentially asymptomatic. Reoperation is indicated if the pressure in the pulmonary ventricle approaches systemic pressure, even in the absence of symptoms [6,21]. Clinical signs and symptoms of dysfunction of the pulmonary ventricle and serious atrioventricular valve insufficiency of the pulmonary ventricle may be indications for conduit replacement in the presence of a lesser degree of obstruction. Marked increase in pressure in the pulmonary ventricle concomitant with ventricular dilatation and loss of systolic function may occur rapidly, especially in small children. Ventricular dysfunction secondary to long-standing severe pressure overload may not be completely reversible, even if the outflow obstruction is completely relieved. We believe that reoperation should ideally be performed at the first sign of newly acquired pulmonary ventricular dysfunction or right-sided atrioventricular valve insufficiency. All patients undergoing conduit replacement will therefore not necessarily have large right ventricle to pulmonary artery gradients (see Fig 3). Replacement of the extracardiac conduit can be performed at a low risk. In the 43 patients undergoing replacement of extracardiac conduits in our own series, there have been no early or late deaths. All conduits were replaced with another conduit 18 mm or more in diameter. Similarly low mortality rates for conduit replacement have been reported previously by others [lo, 131. Long-term survival after a conduit change is expected to be good, although careful follow-up of such patients is imperative. At initial conduit placement, we currently favor the use of a conduit that is 5 to 8 mm larger in diameter than the size of a main pulmonary artery (normalized to the patient s weight and body surface area). We believe that insertion of too large a conduit at the initial operation can be a major source of reported problems such as coronary compression, bleeding, sternal erosion and distortion of pulmonary arteries. We generally prefer a porcine-valved Dacron (Hancock) conduit initially. This conduit is readily available and will usually function adequately until an adult-size aortic or pulmonary homograft is inserted at subsequent operation. Furthermore, the initial insertion of a Dacron tube improves the technical ease and safety of future conduit changes. In choosing to insert smaller conduits, we accept that reoperation (at low risk) will be necessary. Our technique for conduit replacement usually employs mild hypothermia (34 C nasopharyngeal) and cardiopulmonary bypass through a single right atrial cannula and an aortic cannula, as described by the Mayo Clinic group [6]. Dissection is limited to the cannulation sites and conduit (and pulmonary arteries if repair of stenosis is required). With the heart beating, the old conduit is excised, leaving a margin of prosthetic material along the left main and left anterior descending coronary arteries to provide an area for safe suture placement. The distal and then proximal anastomoses are performed. The pulmonary arteries can be enlarged with an autologous pericardial or homograft patch if necessary. Whenever possible, we avoid Dacron tube extension of the homograft [8]. In conclusion, low operative mortality with excellent medium-term survival can be achieved with operations involving conduit placement in the pulmonary circuit.

6 290 SANOETAL Ann Thorac Surg 1991;52:28590 Reoperation at low risk is currently accepted to match the initial conduit size to the patient s size. Subsequent conduit change is facilitated by the use of a porcine-valved Dacron tube in infants and small children, with later insertion of an adult-size cryopreserved aortic or pulmonary homograft. Replacement should be carried out before important right ventricular dysfunction develops. References 1. Ross DN, Somerville J. Correction of pulmonary atresia with a homograft aortic valve. Lancet 1966;2: Bowman FO Jr, Hancock WD, Malm JR. A valve-containing Dacron prosthesis: its use in restoring pulmonary arteryright ventricular continuity. Arch Surg 1973;107:72&8. 3. McGoon DC, Rastelli GC, Ongley PA. An operation for the correction of truncus arteriosus. JAMA 1968;205: Rastelli GC, McGoon DC, Wallace RB. Anatomic correction of transposition of the great arteries with ventricular septa1 defect and subpulmonary stenosis. J Thorac Cardiovasc Surg 1969;58: McGoon DC, Danielson GK, Puga FJ, Ritter DG, Mair DD, Ilstrup DM. Late results after extracardiac conduit repair for congenital cardiac defects. Am J Cardiol 1982;49: Schaff HV, DiDonato RM, Danielson GK, et al. Reoperation for obstructed pulmonary ventricle-pulmonary artery conduits: early and late results. J Thorac Cardiovasc Surg 1984; Jonas RA, Freed MD, Mayer JE Jr, Castaneda AR. Long term followup of patients with synthetic right heart conduits. Circulation 1985;72(Suppl 2): Bull C, Macartney FJ, Horvath P, et al. Evaluation of longterm results of homograft and heterograft valves in extracardiac conduits. J Thorac Cardiovasc Surg 1987;94: O Brien MF, Stafford EG, Gardner M, et al. The viable cryopreserved allograft aortic valve. J Cardiac Surg 1987; 2(Suppl 1):15> Ebert PA, Turley K, Stanger P, Hoffman JIE, Heyman MA, Rudolph AM. Surgical treatment of truncus arteriosus in the first month of life. Ann Surg 1984;200: Ciaravella JM Jr, McGoon DC, Danielson GK, Wallace RB, Mair DD, llstrup DM. Experience with the extra-cardiac conduit. J Thorac Cardiovasc Surg 1979:78: Bailey WW, Kirklin JW, Bargeron LM Jr, Pacifico AD, Kouchoukos NT. Late results with synthetic valved external conduits from venous ventricle to pulmonary arteries. Circulation 1977;56(Suppl 2):7> Bisset GS, Schwartz DC, Benzing G, Helmsworth JA, Schreiber JT, Kaplan S. Late results of reconstruction of right ventricular outflow tract with porcine xenografts in children. Ann Thorac Surg 1981;31: Fontan F, Choussat A, Deville C, Coutremepuich C, Coupilland J, Vosa C. Aortic valve homografts in the surgical treatment of complex cardiac malformations. J Thorac Cardiovasc Surg 1984;87: Kay PH, Ross DN. Fifteen years experience with the aortic homograft. The conduit of choice for right ventricular outflow tract reconstruction. Ann Thorac Surg 1985;40: Di Carlo D, de Leva1 MR, Stark J. Fresh antibiotic sterilized aortic homografts in extracardiac valved conduits: long-term results. Thorac Cardiovasc Surg 1984;32: Ebert PA. Current techniques and results in infancy. In: Moulton AL, ed. Congenital heart surgery. Current techniques and controversies. Pasadena, CA: Appleton Davies, 1984: Boyce SW, Turley K, Yee ES, Verrier ED, Ebert PA. The fate of the 12mm porcine valved conduit from the right ventricle to the pulmonary artery. A ten year experience. J Thorac Cardiovasc Surg 1988;95: Heck HA Jr, Schieken RM, Lauer RM, Doty DB. Conduit repair for complex congenital heart disease. Late followup. J Thorac Cardiovasc Surg 1978;75:80& Stewart S, Manning J, Alexson C, Harris P. The Hancock external valved conduit. A dichotomy between late clinical results and late cardiac catheterization findings. J Thorac Cardiovasc Surg 1983;86: Stark J, Pacifico AD. Reoperations in patients with extracardiac valved conduit. In: Stark J, Pacifico AD, eds. Reoperations in cardiac surgery. Berlin: Springer-Verlag, 1989: