Eur J Cardiothorac Surg 2000;17:624-630
© 2000 Elsevier Science NL
Reconstruction of the RVOT with valved biological conduits: 25 years experience with allografts and xenografts
Michael Homann,
John Cristoph Haehnel,
Nikolaus Mendler,
Sung Un Paek,
Klaus Holper,
Hans Meisner,
Rüdiger Lange
German Heart Center Munich, Department of Cardiac Surgery, Lazarettstrasse 36, D-80636 München, Germany
Corresponding author. Tel.: +49-89-1218-4111; fax: +49-89-1218-4113
e-mail: homann{at}dhm.mhn.de
 |
Abstract
|
|---|
Objective: The reconstruction of the RVOT in congenital heart disease often requires the implantation of a valved conduit. Although allografts are considered the conduit of choice their availability is limited and therefore xenografts are implanted as well. We compared the long-term durability of both grafts in the RVOT over a 25-year period. Methods: Between January 1974 and August 1999, 505 patients (median age 4.0 years, range 2 days–31 years; median weight 14.5 kg, range 2.2–76.6 kg; median body length 103 cm, range 48–183 cm) with congenital malformations (PA 25.3%, TOF 14.5%, TOF+PA 2.4%, DORV 4.2%, TGA+PS 8.7%, TAC 24.8%, and other 20.2%) received their first valved conduit (174 xenografts: median diameter 14 mm, range 8–27 mm; 331 allografts: median diameter 19 mm, range 8–30 mm). Results: Follow-up is 3017 patient-years. The 10-year survival-probability for all patients. was 66% with a mean reoperation-free interval for conduit-exchange of 13.3 years (mean reoperation-free interval for allografts, 16.0 years; mean reoperation-free interval for xenograft, 10.3 years). One hundred and thirteen patients underwent a conduit-exchange, mostly due to conduit stenosis. Fourteen patients had a second exchange and three patients a third exchange. For patients with conduit diameters <18 mm (n=235: allograft n=116, xenograft n=119; median age 9 months, range 0–27.3 years), the mean reoperation-free interval was 11.2 years (mean interval allograft, 13.1 years; mean interval xenograft, 8.6 years, P=0.03). For conduit diameters 
18 mm (n=270: allograft n=215, xenograft n=55, median age 7.4 years, range 0–34.3 years) the mean interval from freedom of conduit exchange was 15.1 years (for allografts 14.1 years, for xenografts 12.5 years, P<0.01). Comparing xenografts to allografts, we found no difference in patient survival probability (P=0.62). There was no significant difference between antibiotic (n=198) preserved vs. cryopreserved (n=133) allografts (P=0.06). Blood group compatibility of allografts to recipients had no significant influence on allograft function (P=0.42). The donors allograft origin, whether aortic or pulmonary valve, had also no significant influence on allograft long-term function (P=0.15). Conclusion: For the reconstruction of the right ventricular outflow tract (RVOT) allografts show significantly better long-term durability than xenografts regardless of the age at implantation and the diameter.
Key Words: Biological heart valves • Xenograft • Allograft
|
1. Introduction
|
|---|
Biological valved conduits have been used to correct congenital malformations in neonates and infants [1]. Nevertheless the question remains open whether allografts are superior to xenografts. We therefore analyzed 505 patients with congenital malformations who received a valved conduit as a first implant between January 1974 and August 1999.
The first clinical use of allograft tissue in cardiovascular surgery was in 1948, when Gross used cadaver arterial grafts to construct systemic to pulmonary artery shunts in patients with tetralogy of Fallot, and to repair coarctation of the aorta [2]. Allograft valves have been implanted in humans for nearly four decades, since their first implantation by Donald Ross in London, UK [3] and Barrat-Boyes in Auckland, New Zealand [4] in 1962. Their use has been restricted by their limited availability and by their different long-term durability, possibly due to the multitude of preservation techniques [5]. These problems led to the development of standardized processing techniques for bioprothesis with long-storage capabilities. Basically these improvements in storing techniques increased the availability of allografts for correction of congenital heart diseases. However, at the same time the glutaraldehyde-preserved xenograft porcine valves (Alain Carpentier, Paris in 1969 [6] in a woven Dacron tube also became available [7,8]. In this retrospective study the differences between right ventricular outflow tract (RVOT) implanted allografts and xenografts were determined, analyzing the factors, which have an influence on the durability of valved biological conduits. For this purpose we analyzed the survival and long-term function of xenografts and allografts in two groups of patients. To address the question whether the durability of either allo- or xenograft is dependent from the graft-size, we stratified for diameter size. Specifically for allografts, we investigated the influence of donors graft origin, blood group incompatibility and method of conversation on durability.
|
2. Patients and methods
|
|---|
From January 1974 until August 1999 a total of 635 biological valved conduits were placed in 505 patients to establish right ventricular to pulmonary artery continuity at the German Heart Center, Munich, Germany. These 505 patients received as their first valved conduit 174 porcine xenografts and 331 allografts. The median diameter of all conduits was 18 mm (range 8–30 mm). This diameter size was the cut-point for stratification. The hospital records of these patients were reviewed retrospectively and the follow up was complete in 98% (n=495). The ages of the patients (Fig. 1) at the time of operation ranged from 2 days to 31 years (median 4.0 years). To evaluate retrospectively the long-term function of the conduits we analyzed the interval of freedom from reoperation after xeno-, allografts implantation for two groups, with diameter <18 mm and >18 mm, respectively.
A further subgroup of patients was analyzed who had received conduits <15 mm. This was done to differentiate between conduits that had to be explanted because of structural deterioration and those, which were explanted because the child had outgrown the conduit size. Xenografts were used only in those cases where no allograft of comparative size was available. No patient underwent dilatation of a stenosed conduit.
Furthermore, we evaluated three factors determining the time-interval for allograft-replacement. These factors were the preservation techniques, the origin of the allograft and the blood group compatibility.
Xenograft patients were slightly younger (mean age 5.4 years, SEM 0.5, range 0 days–27.3 years; for allografts, mean age 6.7 years, SEM 0.4, range 2 days–34.3 years).
The diagnosis at the time of operation is shown in Table 1. The most common diagnosis was pulmonary atresia and truncus arteriosus. There is no significant difference in the distribution of xenografts and allografts (P=0.72) for different diagnosis.
A separate analysis was conducted for patients less than 1 year, which revealed no difference in xenograft use and the rate of conduit-replacement (Table 2).
All procedures were performed using standard techniques such as hypothermic cardiopulmonary bypass, bicaval venous cannulation and aortic cannulation with periods of low-flow or hypothermic circulatory arrest. In case of additional intracardiac repairs, such as atrial or ventricular septal defect closure, infundibular muscle resection, or distal pulmonary artery reconstruction, the aorta was cross-clamped and cold cardioplegic solution was infused into the aortic root.
The right ventricle was opened by a vertical incision, thereby exposing the ventricular septal defect if necessary. The proximal muscular part at the base of the human valve was bordered as short as possible, while the distal part was tailored obliquely for optimal adaption to the pulmonary vessels [9]. The distance from the pulmonary connection can be variably tailored to calculate the anatomic available space, before the length of the patch determines an orthograde, nearly free position to the circumferential tissue. Optimal valve function was achieved by performing a proximal patch-roof that extends the gap for closure. The patch material was Gore-Tex® (W.L. Gore & Associates, AZ), or double velour woven Dacron. Using an aortic allograft, an attempt was made to form the roof with the dissected rest of the anterior mitral leaflet.
In cases of interposition a xenograft conduit (Hancock, Medtronic Inc., MN) the ventricular end of the Dacron tube was cut obliquely in an orthograde related low angle. A Dacron-roof was formed with an oval posterior orifice, which covers anteriorly the full length of the ventriculotomy. Distal anastomosis between the Dacron-tube and the pulmonary trunk was end to end.
Aortic and pulmonary valve allografts were procured from donors free of cardiac disease within 48 h after death. One hundred and eighty nine of the allografts were antibiotically preserved. Since January 1991, a new cryopreservation procedure was employed for standardized uniform cooling in 133 allografts.
Briefly, the valves were dissected in our tissue-bank, examined macroscopically and their internal diameters measured using graduated obturators. The excess tissue was dissected away, specimens were taken for bacteriological and fungal culture. After sizing, the allograft was stored for 18–20 h in a nutrient medium containing antibiotics (cefoxitin sodium, clindamycin, polymyxin B sulfate, vancomycin hydrochloride, nystatin) at 4°C for sterilization. If the microbiological cultures were sterile allografts were stored and implanted within four weeks.
Recently our institution started the cryopreservation period. For sterilization a solution containing cephalotin, piperacillin, polymyxin-B sulfate, neomycinsulfat, nystatin and human serum albumin was used. Without changing the above-described temperature and time procedure, the allografts were put in a nutrient solution with 10% dimethyl sulfoxide [10] in a doubly sealed package. The package was placed in a freezing chamber where the temperature was reduced by 1°C/min until it reached -60°C [11] using a specially developed heat sink. With this heat sink technique a faster initial lowering of the temperature is achieved. Therefore the increase of temperature caused by the crystallization process is better compensated and the crystallization period is shortened. After 1 h the allografts are then transferred to their permanent storage over the liquid nitrogen at -196°C.
Conduit exchange was performed when the pressure gradient between the right ventricle and the pulmonary artery exceeded 40–50 mmHg. Our policy is to replace the conduit before clinical signs of right ventricular failure appear.
Data are presented as median and range. All data related to survival and reoperation for biological valve dysfunction or outgrow parameters were analyzed by the method of Kaplan and Meier, differences were evaluated with the log-rank test.
|
3. Results
|
|---|
The Kaplan–Meier function shows the patient survival probability after first implantation for all patients being 66% after 10 years. Thirty-day mortality was 9.2% following xenograft implantation (174/16) and 10.0% following allograft interposition (331/33). Thirty-day mortality following the first graft exchange was 0.9% (113/1). The 10-year survival probability after allograft implantation was 58.3% and after xenograft-implantation 67.0% (Fig. 2), this difference not being significant (P=0.62).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 2. Kaplan–Meier function for overall freedom from biological ROVT conduit replacement: xenograft vs. allograft; # at risk indicates number of patients at risk. Patients at risk are the number of patients who are exposed at any particular time.
|
|
For the 174 xenografts the median diameter was 14 mm, range 8–27 mm. The 331 allograft-diameters ranged from 8–30 mm with median diameter of 19 mm. A multivariate Cox regression was conducted to evaluate the simultaneous effects of age at initial implantation, initial conduit diameter and conduit type xeno- vs. allograft on durability. Table 3 shows, that the conduit type remains a significant risk factor for conduit durability independent from age and conduit diameter.
View this table:
[in this window]
[in a new window]
|
Table 3. The conduit type remains a significant risk factor for conduit durability independent from age and conduit diameter
|
|
The mean reoperation-free interval for all valved biological conduits was 13.3 years.
Comparing xenografts (n=174) with allografts (n=331) we found a significant difference (Fig. 3) in favor of allografts (P<0.01). The mean reoperation-free intervals for allografts were 16.0 years and for xenografts 10.3 years. After a period of 10 years, 30% of the children initially interposed with allografts, and 70% of the patients with xenograft had undergone replacement of their conduits.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 3. Implanted valved biological conduits at the moment of first replacement; # indicates number of patients at risk.
|
|
For conduits >18 mm the mean freedom from conduit exchange was 15.1 years (Fig. 4).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 4. Freedom of valved biological conduit replacement with implanted diameters 18 mm (P=0.01); # indicates number of patients at risk.
|
|
After a period of 10 years, 28% of the children interposed with allografts, and 62% of the patients with xenografts had undergone replacement of their conduits. The mean reoperation-free interval for allografts >18 mm was 14.1 years, for xenografts >18 mm 12.5 years.
For conduits <18 mm the mean reoperation-free interval was 11.2 years (Fig. 5).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 5. Freedom of valved biological conduit replacement with implanted diameters <18 mm (P=0.03); # indicates number of patients at risk.
|
|
After a period of 10 years, 47% of the children initially interposed with allografts, and 76% of the patients with xenografts had undergone replacement of their valved biological conduits.
The mean reoperation-free interval for allografts <18 mm was 13.1 years and for xenografts <18 mm 8.6 years.
There was no significant difference (Fig. 6) between antibiotic (n=198) preserved vs. cryopreserved (n=133) allografts on graft survival (P=0.06).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 6. Durability after first implantation with different conservation techniques: Antibiotic vs. cryopreserved allografts.
|
|
Blood group compatibility had no significant influence on allograft survival (P=0.15).
In 34 patients with ABO mismatch the reoperation-free interval was not different from that in 77 patients in whom donor blood group matched recipient blood group (Fig. 7).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 7. Durability after first implantation according to ABO compatibility. When donor blood group and recipient blood group were discordant, mismatch was assumed.
|
|
The donors allograft origin, whether aortic or pulmonary valve, had no significant influence on allograft function in RVOT reconstruction (P=0.15), (Fig. 8).
Concerning the reoperation-free interval there was no difference between 102 patients, who received an aortic allograft and 216 patients who were implanted with a pulmonary allograft.
For conduits <15 mm the mean freedom from conduit exchange was 6 years (P=0.24), (Fig. 9). After a period of 10 years, 51% of the children interposed with allografts and 76% of the patients with xenografts had undergone replacement of their conduits. There was no difference in the mean reoperation-free interval for the allografts in comparison with xenografts due to the outgrow situation in this diameter group.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 9. Freedom of valved biological conduit replacement with implanted diameters <15 mm. # indicates number of patients at risk.
|
|
From this diameter group the allografts and xenografts were inspected after explantation and exhibited regular function in the majority of the cases. So the time for exchange is determined by the outgrow of the small diameters.
Therefore, retrospectively the parameters indicating pending outgrow were analyzed. The ratio of the diameter of the actually implanted conduit to the size of the pulmonary artery was computed and compared to the ratio in normal children with the same body length, according to van Meurs van Woezik et al. [12].
From our results we recommend close monitoring of right ventricular function parameters, when the implanted diameter drops to less than 83% of the normal pulmonary valve size that would be expected at any given body length (Fig. 10).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 10. Relation between expected and actual pulmonary valve size to body length and age of reoperated group.
|
|
|
4. Discussion
|
|---|
Which is the preferable valve for the reconstruction of the RVOT, porcine conduit or allograft? This question has accompanied the implantation of valved biological grafts for 25 years. In our study there was no difference in patients' survival after implantation of an allograft or xenograft over a 25-year period. The Kaplan–Meier function shows the survival probability after first allograft implantation for all patients being 66% after 10 years. However, freedom from graft replacement was significantly better in the allograft group for both, smaller and larger diameters. In the majority of these cases the indication for replacement of the conduit was graft stenosis. As to this date, 22.4% of all primary implanted conduits have already been replaced.
An early indicator of graft failure is the development of tricuspid incompetence and these patients require short-term control. This includes echocardiography examination to detect early signs of reduction of right ventricular function. The reoperation is indicated when reduction of right ventricular function is evident or when the gradient over the right ventricular outflow tract reaches more than 50 mmHg. Clinically, most of these patients are asymptomatic.
Comparing the long-term function of xenografts with allografts, we found a significant difference in favor of allografts. After a period of 10 years, 30% of the children initially corrected with allografts and 70% of the patients with xenografts had undergone replacement of their conduits. The shortest interval free of reoperation is observed in patients being corrected as infants with conduits smaller than 18 and 15 mm, respectively. After correction, these children grow out the implanted conduit [13].. In the majority of the cases these explanted valves show no signs of degeneration, however, a relative stenosis had occurred while the children were growing. When diameters smaller or equal 15 mm were used, the children came for replacement nearly 6 years after the first implantation had been performed, at a time when they were babies or infants. Our results showed, that there is no difference in the durability of xeno- and allografts when only the first 5–6 years after the operation are observed. Since by that time most conduits had already been replaced because of the child's outgrowth of the conduit, a xenograft conduit may very well be chosen in the newborn period [14,15].
In contrary, after 10 years, xenograft conduits had already been replaced in 62% compared to only 28% of the allograft conduits. These conduits then show obvious signs of degeneration, such as sclerosis and calcification. When the first implantation was done before school age with an allograft conduit sized 15–18 mm, the patient was free of reoperation for 13.1 years. The type of conservation, blood group compatibility, and the anatomic origin of the donor graft seemed to have no influence on the durability of the human valve. We found no difference between the implanted antibiotically preserved (+4°C) or the cryoconserved allografts in respect to long-term durability. Similarly, Bodnar et al. analyzed results from a 20-year follow-up of cryopreserved (-196°C) viable and fridge-preserved (+4°C) non-viable grafts and found no significant differences between both groups [16]. In contrast, Mair et al. reported about longer freedom from reoperation for porcine-valved grafts than for allografts [17]. However, the allografts of the Mayo Clinic were irradiated while our allografts were antibiotically sterilized and cryoconserved.
A difference in respect to long-term function between the pulmonary and aortic allografts has been described by Bando et al. [18]. These results indicate that both aortic and pulmonary allografts provide excellent intermediate-term patient survival after RVOT-reconstruction, but that pulmonary allografts are more durable than aortic allografts with less calcification and obstruction. Also Clarke et al. [19] reported about a significantly higher percentage of aortic valve allografts being explanted, which corroborates published data and advocates the use of pulmonary valve allografts to reconstruct the right ventricular outflow tract. Such a difference was not observed by Stark et al. [20] and could not be confirmed by our results either.
In some patients the recipients’ immunologic response could be a borderline rejection, sub acute over years, which will reduce the durability of the allografts [5].. Our valve donors and recipients retrospectively were not matched for blood group or human leukocyte antigens (HLA) and immunosuppressive therapy was not given to valve recipients at any time. Our study showed no evident influence of blood group compability upon the durability of allografts. Theoretically, a preoperative genetic monitoring could be useful to select a specific HLA allograft recipient borderline group for elective immunosuppressive treatment to reduce the immunogenicity of the allograft valve and to improve the long-term results after first allograft implantation and replacement.
Long-term durability of all biological conduits is still far from being satisfying.
Endothelialization with autologous cells might become a major breakthrough [21]. A denatured, genetically inert allograft will receive an autologous, vital ‘de novo’ endothelial cell layer. Theoretically autologous endothelium may serve as a ‘genetic-coat’ against immunologic and biochemical stress and thus improve long-term durability. Until long-term results for these new approaches are available, allografts remain the conduit of choice in reconstruction of the right ventricular outflow tract.
|
Footnotes
|
|---|
Presented at the 13th Annual Meeting of the European Association for Cardio-thoracic Surgery, Glasgow, Scotland, UK, September 5–8, 1999.
|
Appendix A Conference discussion
|
|---|
Dr D. DiCarlo (Rome, Italy): I would like to ask you to be more precise with your indication for conduit replacement, as your slide indicated a pressure gradient of 50 mmHg and in your presentation you quoted 80 mmHg pressure gradient.
Besides, I think that it should be kept in mind that recent data suggest that, at secondary implantation homograft duration tends to be much shorter than after first implantation. I therefore would agree with what you suggested; it is really not wise to use homografts in small children, if one considers that your second homograft may last much less than your first one.
Dr Homann: I shall answer your second question first. At the moment, we have only small numbers about the fate after second replacement of the allografts. When we have the first results after the second implantation and there will be a shorter durability in comparison with the first implantation period, we shall present these results at the EACTS meeting.
Corresponding to your first question we always consider the reoperation with our cardiologists, and we are looking forward when the right ventricular dysfunction will not be increased. Normally we indicate the reoperation, when the invasive evaluated gradient reaches between 40 and 50 mmHg.
|
References
|
|---|
-
Hopkins R.A. Rationale for use of cryopreserved allograft tissues for cardiac reconstruction. In: Hopkins R.A., ed. Cardiac reconstruction with allograft valves. Berlin: Springer Verlag, 1989:15-19.
-
McElhinney D.B., Reddy V.M., Hanley F.L. Homografts in congenital heart disease: current applications and future directions. Isr J Med Sci 1996;32(10):880-885.[Medline]
-
Ross D.N. Homograft replacement of the aortic valve. Lancet 1962;2:487.[Medline]
-
Barrat-Boyes B.G. Homograft aortic valve replacement in aortic incompetence and stenosis. Thorax 1964;19:131-150.
-
O'Brien H.F., Staford E.G., Gardner A.H., Polner P.G., McGrifin D.C. A comparison of aortic valve replacement with viable cryoconserved and fresh allograft valves, with a note on chromosomal studies. J Thorac Cardiovasc Surg 1987;94:812-823.[Abstract]
-
Carpentier A., Lemaigre G., Ladislas R., Dubost C. Biological factors affecting long term results of valvular heterografts. J Thorac Cardiovasc Surg 1969;58:467.[Medline]
-
Planche C., Binet J.P., Langlois J., Conso J.F. Reconstruction de la voie d'ejection du ventricule droit a l‘aide de tubes valves. Nouv Presse Med 1972:541.
-
Bowman F.O., Hancock W.D., Malm J.R. A valve containing Dacron prothesis. Arch Surg 1973;107:724.[Medline]
-
Clarke D.R., Bishop D.A. Allograft reconstruction of the right ventricular outflow tract: technique and results. In: Yankah A.C., Yacoub M.H., Hetzer R., eds. Cardiac valve allografts. Science and practice. Darmstadt: Steinkopff. New York: Springer, 1997:263-274.
-
Lovelock J.E., Bishop M.W.H. Prevention of freezing damage to living cells by dimethylsulphoxide. Nature (Cond.) 1959;183:1394.
-
Mendler N., Ettner U., Meisner H. Heat sink technique for allograft heart valve freezing. 6th International Conference on Tissue Banking, Edingburgh. 1997:60.
-
Van Meurs van Woezik H., Debets T., Klein H.W. Growth of the internal diameters in the pulmonary arterial tree in infants and children. J Anat 1987;151:107-115.[Medline]
-
Borrow K.M., Green L.H., Castaneda A.R., Keane J.F. Left ventricular function after repair of tetralogy of fallot and it's relationship to age of surgery. Circulation 1980;61:1150.[Abstract/Free Full Text]
-
Homann M., Haehnel J.C., Wottke M., Meisner H., Mendler N., Sebening F. Allograft vs. xenograft – 20 years experience with valved biologic conduits. Thorac Cardiovasc Surgeon 1995;43(Suppl I):60 Abstract 40.[Medline]
-
Weipert J., Meisner H., Mendler N., Haehnel J.C., Homann M., Paek S.-U., Sebening F. Allograft implantation in pediatric cardiac surgery: Surgical experience from to 1994. Ann Thorac Surg 1995;52:285-290.[Abstract]
-
Bodnar E., Matsuki O., Parker R., Ross D.N. Viable and nonviable aortic homografts in the subcoronary position: a comparative study. Ann Thorac Surg 1989;47:799-805.[Abstract]
-
Mair D.D., Danielson G.K. Long-term results of ventricle to pulmonary artery conduits. The Second World Congress of Pediatric Cardiology and Cardiac Surgery, Honolulu, Hawaii 1996;1997:293 PL.
-
Bando K., Danielson G.K., Schaff H.V., Mair D.D., Julsrud P.R., Puga F.J. Outcome of pulmonary and aortic homografts for right ventricular outflow tract reconstruction. J Thorac Cardiovasc Surg 1995;109(3):509-517 Discussion 517–518.[Abstract/Free Full Text]
-
Clarke D.R., Bishop D.A. Allograft reconstruction of the right ventricular outflow tract: technique and results. In: Yankah A.C., Yacoub M.H., Hetzer R., eds. Cardiac Valve allografts. science and practice. Darmstadt: Steinkopff. New York: Springer, 1997:263-274.
-
Stark J., Bull C., Stajevic M., Jothi M., Elliott M., de Leval M. Fate of subpulmonary homograft conduits: determinants of late homograft failure. J Thorac Cardiovasc Surg 1998;115(3):506-514 Discussion 514–516.[Abstract/Free Full Text]
-
Shinoka T., Shum-Tim D., Ma P.X., Tanel R.E., Isogai N., Langer R., Vacanti J.P., Mayer J.E., Jr Creation of viable pulmonary artery autografts through tissue engineering. J Thorac Cardiovasc Surg 1998;115(3):536-545 Discussion 545–546.[Abstract/Free Full Text]
Received September 6, 1999;
received in revised form January 10, 2000;
accepted February 21, 2000.
Top
Abstract
1. Introduction
