Clinical Results of Cryopreserved Valved Conduits in the Pulmonary

Ventricle-to-Pulmonary Artery Position Sanjeev Sharma, uo, Adnan Cobanoglu, MO, Jeri Dobbs, as, Mary Rice, MO, Portland. Oregon

Aortic valved homograft conduits (AVHC) have be- come valuable in the pulmonary ventricle (PV)-to- main pulmonary artery (MPA) reconstruction in congenital heart defects. Since 1985, 45 patients, ranging in age from 12 days to 32 years, under- went PV-to-MPA reconstruction utilizing cryopre- served AVHC. Operative deaths included seven pa- tients (16%), six of whom died as a result of the complexity of their underlying heart defects. One late death (2%) occurred as a result of infective endocarditis 48 months after conduit placement. The 38 patients who survived the operation re- mained in the intensive care unit for a mean of 5.7 4- 1.0 days (median: 4 days; range: 2 to 37 days). The mean hospital stay was 13.0 4- 1.8 days (me- dian: 9 days; range: 6 to 63 days). The mean fol- low-up was 40.0 4- 3.6 months (median: 40 months; range: 10 months to 7.1 years).

Only two patients (5%) required reoperation for conduit stenosis with systolic pressure gradients of 60 to 80 inm Hg at 10 and 14 months, respec- tively, after operation, and both reoperations were successful. During outpatient visits, 16 patients are totally asymptomatic, and 21 patients have minimal symptoms (New York Heart Association class II). Only 10 patients (26%) require digoxin, and 2 pa- tients (5%) need diuretics as part of their medical regimen. Recent echocardiographic examinations show insignificant pressure gradients in all 37 cur- rently surviving patients. Thus, barring operative mortality, which is almost always associated with the nature of the underlying heart defect, the use of cryopreserved AVHC is a safe and effective alterna- tive for PV-to-MPA reconstruction.

From the Departments of Cardiopulmonary Surgery (SS, AC, JD) and Pediatric Cardiology (MR), Oregon Health Sciences University, Port- land. Oregon.

Requests for reprints should be addressed to Adnan Cobanoglu, MD, L353, Oregon Health Sciences University, 3181 Southwest Sam Jackson Park Road. Portland, Oregon 97201-3098.

Presented at the 79th Annual Meeting of the North Pacific Surgi- cal Association, Tacoma. Washington, November 13-14. 1992.

C urrently, the use of aortic valved homograft conduits has become an important alternative for the recon-

struction of the fight ventricular outflow tract. In 1966, Donald Ross [1 ] pioneered the use of aortic valved homo- grafts by placing a homograft conduit in an 8-year-old boy with pulmonary atresia. At that time, the scarcity of fresh homograft conduits and early calcific obstruction of irradiated conduits led to the development of a porcine- valved Dacron prosthesis [2,3]. Agarwal et al [4,5] dem- onstrated, however, that late conduit obstruction can oc- cur in these heterograft conduits by two pathologic processes: the valve may become stenotic due to calcific degeneration, and the conduit may be obstructed either proximally or distally with a thick fibrous peel.

Presently, with the advent of cryopreservation and an increase in pediatric organ procurement, aortic and pul- monary valved homografts are more readily available. Cryopreserved, valved homograft conduits used by Ore- gon Health Sciences University (OHSU) were available from two centralized sources that supply these tissues nationwide. Since 1985, one of these centers has pre- served and provided for implantation 4,058 aortic valves, pulmonary valves, and aortic valved homograft conduits for right ventricular outflow reconstruction in U.S. cen- ters (written communication, CryoLife Cardiovascular Inc., Marietta, GA, September 1992). To date, only a few published series have documented the clinical outcome with regard to cryopreserved aortic valved homograft conduits in the reconstruction of the right ventricular outflow tract [6-8]. Accordingly, we examine the OHSU experience of homograft conduit performance by review- ing the clinical outcomes and echocardiographic results through the intermediate-term follow-up.

PATIENTS AND METHODS A retrospective review of the charts of 45 patients with

complex, congenital heart defects who received cryopre- served aortic valved homograft conduits from September 1985 to October 1991 was undertaken. The most com- mon underlying congenital malformation requiring a conduit placement or replacement was tetralogy of Fal- lot, which was present in 20 patients. Thirteen patients had truncus arteriosus, and 9 patients were treated for d- transposition of the great arteries with subpulmonic ste- nosis. The remaining three patients had a double-outlet right ventricle with pulmonary stenosis. The group con- sisted of 21 males and 24 females. Their mean age was 9.1 4- 1.1 years (median: 6.5 years; range: 12 days to 32 years) (Figure 1). Further, their mean weight was 28.8 4- 3.3 kg (median: 21.5 kg; range: 2.8 to 91.0 kg).

Thirty-nine patients (87%) had undergone previous

THE AMERICAN JOURNAL OF SURGERY VOLUME 165 MAY 1993 587

SHARMA ET AL

0-1 1-3 3-8 8-12 12-20 21-30 > 30 Age (Years)

Figure 1. Age distribution in 45 patients who underwent pulmo- nary ventr icle-to-pulmonary artery reconstruction with the use of a cryopreserved aort ic valved homograft conduit.

T A B L E I Previous Palliative Procedures

Procedure No.

Blalock-Taussig shunt 17 Waterston shunt 6 Pulmonary artery banding 6 Blalock-Hanlon septostomy 4 Potts shunt 2 Total 35

T A B L E II Previous Corrective Procedures

Procedure No.

Heterograft right ventricle to pulmonary artery conduit 11 Fallot correction 9 Homograft right ventricle to pulmonary artery conduit 3 VSD closure 3 ASD closure 2 Aortic valve replacement 2 Pulmonary valvotomy 2 Mustar(~ procedure 1 Valveless right ventricle to pulmonary artery conduit 1 Repair of PAPVR 1 Total 35

VSD = vent r icu lar septal defect: ASD = atrial seotal defect: PAPVR = partial

anomalous pulmonary venous return.

T A B L E II1 NYHA Classification

NYHA Class Preoperative Postoperative

I 2 16 II 4 21 III 28 IV 11 Total 45 37

NYHA = New York Heart Association

cardiac operations. Of this group, 25 patients (56%) had undergone prior palliative procedures, and 14 patients (31%) had prior corrective procedures before having ei- ther primary conduit placement as a part of total correc- tion or conduit replacement for graft dysfunction (Tables I and H). Indications for homograft conduit placement were either as a complement to total correction in 30 patients (primary group) or to replace a dysfunctional conduit, previously placed, in 15 patients (replacement group). Furthermore, the latter group consisted of six patients who had undergone prior median sternotomies once, six patients twice, and three patients three times. This replacement group was symptomatic with dyspnea, cyanosis, exercise intolerance, and an increasing trans- conduit, peak systolic gradient with ventricular hypertro- phy. Prior to surgery, the replacement group had a mean transconduit, peak systolic gradient of 67.8 4- 5.5 mm Hg (range: 30.0 to 105.0 mm Hg). Preoperative New York Heart Association (NYHA) classification demonstrated significant right ventricular failure in the majority of patients (Table III). With regard to the two patients classified as NYHA class I, one patient had significant pulmonary valve insufficiency following correction of the tetralogy of Fallot malformation, whereas the other pa- tient played high school varsity basketball with a right ventricle-to-pulmonary artery gradient of 60 mm Hg and increasing right ventricular hypertrophy.

The cryopreserved aortic valved homograft conduits were obtained from two centralized distribution centers (CryoLife Cardiovascular Inc., and Oregon Tissue Bank, Portland, OR) Cryopreservation, thawing, and prepara- tion have been described elsewhere [9]. The aortic valved homograft conduit, procured from either pediatric or adult donors, consisted of the septal leaflet of the mitral valve, interventricular septum, aortic valve, aortic root, ascending aorta, and aortic arch (Figure 2). For best results, freezing techniques allow the harvested conduit a storage life of 4 years. Homograft conduit size ranged from 10 to 25 mm and correlated with the preoperative weight of the patient (Figure 3). Conventional cardiopul- monary bypass with moderate hypothermia (26~ naso- pharyngeal) was utilized in 27 patients; deep hypother- mia (18~ nasopharyngeal) and total circulatory arrest were used in 12 patients; and 6 patients underwent deep hypothermia and low-flow bypass. Myocardial preserva- tion was accomplished by the use of cold, crystalloid cardioplegia solution to keep the septal temperature at 10~ to 12~

Conduit placement was usually the last step in the total correction and followed the repair of intracardiac defects. In conduit replacement operations, the previously placed dysfunctional heterograft, homograft, or valveless conduit was removed in its entirety. The largest possible homograft conduit size, while avoiding sternal compres- sion, was chosen. Distal anastomoses to the main pulmo- nary artery were made in an end-to-end or an end-to-side fashion with continuous polytetrafluoroethylene (Gortex, W.L. Gore, Flagstaff, AZ) suture. In addition, individual left-to-right pulmonary artery anastomoses or patch en- largement of the pulmonary arteries was performed, if

5 8 8 THE A M E R I C A N J O U R N A L OF SU RG ERY VOLUME 165 MAY 1993

CRYOPRESERVED VALVED CONDUITS

26 7 . �9

2 ' t . -

22 t �9 �9

201 . . . . �9 .

"1 " > , " " 16-L.. " . " . r=OTO 14- ' r �9 �9 �9 p< .OOl

12-~ ,~

1 0 1 =- . . . . . , . ~ . . . . ] . . 0 20 4=0 60 80

Weight (kg)

' i 100

Figure 3. The correlation between conduit size versus preopera- tive weight in 45 patients who underwent pulmonary ventricle-to- pulmonary artery reconstruction with the use of a cryopreserved aortic valved homograft conduit.

Figure 2. An aortic valved homograft conduit prior to implantation.

necessary. A vertical ventriculotomy was made, and the proximal end of the conduit was anastomosed to the ven- tricle with a continuous polytetrafluoroethylene (Gortex) suture. Ventricular anastomoses were augmented with a pericardial patch in order to prevent valvular distortion.

Follow-up was managed by the Child Development Rehabilitation Center at OHSU. Following discharge, clinic visits were at 2 weeks, 2 months, 6 months, and then at yearly intervals. Each patient was interviewed and examined by a pediatric cardiologist and underwent two- dimensional and Doppler echocardiography. During Doppler echocardiography, objective results were calcu- lated using the modified Bernoulli formula to assess the peak systolic gradient across any area of stenosis [10]. All patients were seen in the 3 months preceding the conclu- sion of this study (October 1991). Statistical analyses were performed with the Student's t-test and the Fisher's exact test. Also, all data for actuarial survival and reoper- ation were evaluated by the Kaplan-Meier method [11].

RESULTS In our group of 45 patients, there were 7 early deaths

(16%) and 1 late death (2%). Clinical and autopsy data revealed low cardiac output and biventricular heart fail- ure in six patients and a subarachnoid hemorrhage and cerebral edema in one patient as the causes of operative

TABLE IV Operat ive Results

Group 1" Group 2t (n = 30) (n = 15) p Value

Age (y) 7.5 12,0 <0.06 Ventilatory support (h) 82 28 < 0.01 Inotropic support (h) 76 22 < 0.004 Blood transfusion (U) 3 2 < 0.45 Intensive care unit stay (d) 5 3 < 0.17 Hospital stay (d) 12 7 < 0.04 Early mortality 7 0 < 0.08 Late mortality 0 1 - -

*Group 1 (primary group) includes those individuals who underwent homograf t

condui t p lacement for the first t ime,

tGroup 2 (replacement group) includes those individuals who had a homograf f

placed for a previously placed dysfunctional conduit .

~ 8O

i g �9 60

i E 4o o J=

o E

20

(38) (26) (20) (14) (s)

T i m e ( y e a r s )

Figure 4. Actuarial freedom from conduit reoperation for homo- graft conduit dysfunction in 45 patients in whom cryopreserved aortic valved homograft conduits were used to restore pulmonary ventricle-to-pulmonary artery continuity. Numbers in parentheses represent the number of patients at this year of follow-up.

THE AMERICAN JOURNAL OF SURGERY VOLUME 165 MAY 1993 589

SHARMA ET AL

mortality. The only late death occurred in a patient who died 48 months after conduit placement secondary to infective endocarditis. Tables III and IV and Figure 4 summarize the results.

To determine whether a primary conduit placement differs from a replacement procedure, the cohort of 45 patients was divided into 2 groups with respect to prior conduit placement. Group 1 (n -- 30) consisted of pa- tients for whom homograft conduit placement was a part of the total correction, whereas the patients in group 2 (n -- 15) had a dysfunctional conduit replaced. The opera- tive results of these groups are outlined in Table IV. All operative deaths occurred in those patients who had con- duits placed for the first time and as a part of the total correction of their heart defects (p <0.08). The replace- ment group did well with no early deaths, required signifi- cantly fewer hours of ventilatory and inotropic support, and spent fewer days in the hospital. They required a similar number of blood transfusions when compared with the group of patients who were undergoing a pri- mary conduit placement as a part of their total correction, even though this group consisted of patients who had undergone either at least one prior sternotomy (one in six patients, two in six patients, and three in three patients).

The mean follow-up for all 38 operative survivors is 40.0 4- 3.6 months (median: 40 months; range: 10 months to 7.1 years). Thirty-seven patients are currently alive. The actuarial survival is 84% 4- 5% at 1 year after conduit placement, and it is 78% 4- 8% 4 years after conduit placement. During follow-up clinical examination, 16 pa- tients were completely asymptomatic (NYHA class I), whereas 21 patients had mild cardiac symptoms (NYHA class II) (Table III). Currently, only 10 patients (26%) require digoxin, and 2 patients (5%) are administered diuretics as part of their medical management.

To date, two patients (5%) have required reoperation for conduit stenosis. One patient had subvalvular stenosis with an 80 mm Hg gradient at the proximal anastomosis and required reoperation 14 months later. Another pa- tient had a 60 mm Hg valvular gradient that required reoperation 10 months after the original conduit place- ment. Both homograft conduit reoperations were success- ful. The actuarial rates of freedom from reoperation for conduit dysfunction were 97% 4- 4% and 94% 4- 4% at 1 year and 4 years after conduit placement, respectively (Figure 4). Currently, 30 patients (81%) have peak sys- tolic gradients less than 30 mm Hg, and the remaining 7 patients have peak systolic gradients in the 30 to 40 mm Hg range, when examined by Doppler echocardiography. The latter group consists of patients in their growth and development phase following primary total correction.

COMMENTS During the last 5 to 6 years, there has been a renewed

interest and an increase in the use of cryopreserved, valved homograft conduits. Currently, in the United States, with the advent of cryopreservation and pediatric organ procurement, these homograft valves and valved conduits are available through centralized distribution centers. Furthermore, the porcine-valved conduits have

lost favor since Agarwal et al [4,5] demonstrated that their valves undergo calcific degeneration and that the Dacron sleeve may form a thick fibrous peel through neointimal hyperplasia, causing conduit obstruction. Also, prior unsatisfactory experiences with fresh antibiot- ic-sterilized or irradiated homografts have been reported [7-11]. To date, only a few published series have demon- strated results with the use of cryopreserved aortic and pulmonary valved homograft conduits [6-8]. We have retrospectively examined the OHSU experience with cryopreserved aortic valved homograft conduits in the repair of complex cardiac malformations.

In the past, several reports have examined the clinical outcomes with the use of cryopreserved homograft con- duits in the repair of the right ventricular outflow tract. Turley and Ebert [6] reported their early experience with cryopreserved aortic valved homograft conduits. All con- duits were used as a part of total correction. Operative mortality was 13%; however, follow-up was short, and late death and conduit replacement were not examined. In 1987, Kirklin et al [7] reported 6% as the reoperation rate after 3.5 years for conduit problems in a cohort of 147 patients. Recently, Hawkins et al [8] discussed 89 patients undergoing cryopreserved conduit placement. The investigators found no difference in aortic versus pulmonary valved homograft conduits with regard to postoperative conduit dysfunction. Operative mortality was 8% and involved patients who underwent primary conduit placement. There were nine late deaths, none of which were conduit related, and eight reoperations were performed. Freedom from reoperation was only 80% at 4 years after operation.

Our operative mortality is comparable with that found in these series, with an early death rate of 16%, with almost all operative deaths due to very complex primary repairs and the presence of low cardiac output syndrome: Late mortality is 2%. Our reoperation rate is better, with 94% 4- 4% freedom from reoperation 4 years after the operation. We believe that this is because of our tendency to use the largest possible conduit size, while avoiding sternal compression, and our augmentation of the anastomosis at the ventriculotomy site with a large pericardial patch to prevent valvular distortion. Further- more, our results with homograft conduits are superior with regard to the past experiences with fresh antibiotic- sterilized or irradiated homografts. In those series, opera- tive mortality rates ranged from 24% to 47% and late mortality rates from 10% to 22%, and reoperations for conduit obstruction were performed at much higher rates [12-16].

Also, our analysis assessed the differences among the 45 patients with regard to the cohort of patients who had primary conduit placement versus those patients who had replacement procedures. The placement of a cryopre- served aortic valved homograft conduit was either a pri- mary procedure (n -- 30), mostly as part of total correc- tion, or a replacement of a previously placed, dysfunctional conduit (n = 15). As expected, of the 38 survivors, patients in the primary conduit placement group received significantly more hours of ventilatory

5 9 0 THE AMERICAN JOURNAL OF SURGERY VOLUME 165 MAY 1993

CRYOPRESERVED VALVED CONDUITS

and inotropic support and were hospitalized longer than those in the replacement group. The patients in the for- mer group were mostly infants and younger patients (Ta- ble IV). Furthermore, all early deaths occurred in the group of patients who underwent primary conduit place- ment (p <0.08). Despite the previous multiple median sternotomies among the conduit replacement group, their blood transfusion requirements did not differ from those patients undergoing primary conduit placement as a part of total correction and was kept at 2 U per patient. We believe that extensive blood transfusions are unnecessary because of the better apposition of allograft tissues against native tissues and because the need for a stiff prosthetic material, such as Dacron or heterograft valves, is obviated. Homografts have simplified suturing tech- niques and have helped avoid tension at the anastomosis.

With regard to the freedom from reoperation rate, our cohort had an extremely low rate of conduit dysfunction. We believe that this low rate of homograft conduit dys- function is attainable by following certain guidelines. Technical errors occur if the homograft conduit is ob- structed by extrinsic sternal compression, if there is dis- tortion of the aortic valve annulus due to inappropriate suturing techniques, or if the conduit undergoes desicca- tion during implantation. To avoid these errors, the size of the conduit and its length must be tailored to avoid sternal compression and the anastomosis must be augmented at the ventriculotomy site by the use of a pericardial patch on the proximal end of the homograft conduit, thereby maintaining the circular nature of the aortic valve annu- lus. Also, saline should be used to keep the conduit moist during implantation.

Our follow-up needs to be extended in the future, but this review demonstrates that our experience with cryo- preserved aortic valved homograft conduits are superior to previous experiences with fresh antibiotic-sterilized and irradiated homograft conduits. Furthermore, we re- affirm the recent good results that other investigators have demonstrated with cryopreserved homograft con- duits by avoiding the valvular degeneration of the por- cine-valved conduits and the obstruction created by the fibrinous peel of the Dacron sleeve [4 ,5 ,14 ,17] . We real- ize that during longer-term follow-up, those patients who had small conduits as neonates or infants will need re- operation, simply because they outgrow the size of their homograft conduit. Based on our excellent experience in conduit replacements, we believe that these patients can be offered reoperations with minimal morbidity and mor- tality when they reach the appropriate age and size and require replacement.

The 37 current survivors uniformly demonstrate post- operative clinical improvement and a low attrition and reoperation rate. Thus, the use of a cryopreserved aortic valved homograft conduit is a safe and effective alterna- tive for right ventricular outflow tract reconstruction.

Although there is an increasing interest and demand for utilization, the supply remains limited and is associated with an availability trend similar to that of hearts for transplantation. In the future, increased availability will depend on improving the awareness of the public and health care professionals to the tremendous value and need of these tissues in cardiac reconstruction.

R E F E R E N C E S 1. Ross DN, Somerville J. Correction of pulmonary atresia with a homograft aortic valve. Lancet 1966; 2: 1446-7. 2. Merin G, McGoon DC. Reoperation after insertion of aortic homograft as a right ventricular outflow tract. Ann Thorac Surg 1973; 16: 122-6. 3. Bowman RO, Hancock WD, Malin JR. Aortic valve containing Dacron prosthesis: its use in restoring pulmonary artery-right ven- tricular continuity. Ann Surg 1973; 107: 724-32. 4. Agarwal KC, Edwards WD, Feldt RH, Danielson GK, Puga F J, McGoon DC. Clinicopathological correlates of obstructed right- sided porcine-valved extracardiac conduits. J Thorac Cardiovase Surg 1981; 81: 591-601. 5. Agarwal KC, Edwards WD, Feldt RH, Danielson GK, Puga F J, McGoon DC. Pathogenesis of nonobstructive fibrous peels in right- sided porcine-valved extracardiac conduits. J Thorac Cardiovasc Surg 1982; 83: 584-9. 6. Turley K, Ebert PA. Aortic allograft reconstruction of right ventricular-pulmonary artery continuity. Ann Thorac Surg 1989; 47: 278-81. 7. Kirklin JW, Blackstone EH, Maehara T, et al. Intermediate term fate of cryopreserved allograft and xenograft valved conduits. Ann Thorac Surg 1987; 44: 598-606. 8. Hawkins JA, Bailey WW, Dillon T, Schwartz DC. Midterm results with cryopreserved allograft valved conduits from the pul- monary ventricle to the pulmonary arteries. J Thorac Cardiovasc Surg 1992; 104: 910-6. 9. Brockbank KG, Carpenter JF, Dawson PE. Effects of storage temperature of variable bioprosthetic heart valves. Cryobiology 1992; 29: 537-42. 10. Wiles HB. Imaging congenital heart disease. Pediatr Clin North Am 1990; 37: 115-36. 11. Daniel W. Applied nonparametric statistics. 1st ed. Boston: Houghton-Mifflin Company, 1978: 82. 12. 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: 360-4. 13. Fontan F, Choussat A, Deville C, Doutremepuich C, Coupil- laud J, Vosa C. Aortic valve homografts in the surgical treatment of complex cardiac malformations. J Thorac Cardiovasc Surg 1984; 87: 649-57. 14. Schaff HV, DiDonato RM, Danielson GK, et al. Reoperation for obstructed pulmonary ventricle-pulmonary artery conduits. J Thorac Cardiovasc Surg 1984; 88: 334-43. 15. Bull C, Macartney F J, Horvath P, et al. Evaluation of long- term results of homograft and heterograft valves in extracardiac conduits. J Thorac Cardiovasc Surg 1987; 94: 12-9. 16. Burczynski PL, Mckay R, Arnold R, Mitchell DR, Sabino GP. Homograft replacement or the pulmonary artery bifurcation. J Thorac Cardiovasc Surg 1989; 98: 623-31. 17. Stark J. Do we really correct congenital heart defects? J Thorac Cardiovasc Surg 1989; 97: 1-9.

THE AMERICAN JOURNAL OFSURGERY VOLUME 165 MAY 1993 591