The Early and Midterm Function of Decellularized Aortic Valve Allografts

The Early and Midterm Function of Decellularized Aortic Valve Allografts Francisco D. A. da Costa, MD, Ana Claudia B. A. Costa, Roberta Prestes, Ana Carolina Domanski, MD, Eduardo Mendel Balbi, MD, Andreia D. A. Ferreira, MD, and Sergio Veiga Lopes, MD Department of Cardiac Surgery, Santa Casa de Curitiba, Pontificia Universidade Catolica do Parana, and Institute of Neurology and Cardiology of Curitiba, Curitiba, Paraná, Brazil Background. This study evaluates the early and midterm results of decellularized aortic valve allografts (DAVA) as an aortic valve replacement. Methods. Between October 2005 and February 2010, 41 patients, 28 of whom were male, with a median age of 34 years (range, 0.1 to 71), had aortic valve replacement with DAVA. Decellularization was obtained with a 0.1% sodium dodecyl sulfate solution. Postoperative evaluation was performed with serial echocardiograms, magnetic resonance imaging, and multislice computed tomography studies to evaluate valve hemodynamics, allograft conduit dimensions, and calcification scores. Results. There were 3 early deaths and 1 late death, with a mean follow-up of 19 months (range, 1 to 53). There was 1 reoperation due to a failed mitral valve repair. By echocardiography in all patients, the median immediate postoperative peak gradient was 7 mm Hg (range, 1 to 26 mm Hg), and at last follow-up it was 4 mm Hg (range, 1 to 16 mm Hg); valvular regurgitation was graded as none or trivial in all but 1 patient, who had a regurgitation graded as mild to moderate. By magnetic resonance imaging (n 4), mean root dimensions were stable at the annulus (24 mm), sinus of Valsalva (33 mm), and sinotubular junction (28 mm). By computed tomography (n 22), there was only discrete conduit calcification (median calcium score 63 Hounsfield units [HU]; range, 0 to 894 HU) to 3 years of follow-up. Conduit biopsy in the patient who underwent reoperation demonstrated well-preserved wall structure, absence of calcification, and limited in vivo host repopulation. Conclusions. The early and midterm results with DAVA demonstrated stable structural integrity, low rate of calcification, and adequate hemodynamics. Although longer periods of observation are necessary, DAVA appears to be a promising alternative for aortic valve replacement in selected patients. (Ann Thorac Surg 2010;90:1854 61) 2010 by The Society of Thoracic Surgeons Aortic valve allografts (AVA) have been used for more than 50 years with satisfactory clinical outcomes. Owing to the superior hemodynamic performance compared with stented bioprosthesis, low incidence of thromboembolic events, and resistance to infection, they still should be considered an excellent alternative for selected patients needing aortic valve replacement (AVR) [1, 2]. Although cryopreservation is widely accepted as the gold standard method for processing and storage of AVA, preserved cellular viability is associated with a low-grade immune response from the host, which partially explains its limited durability, especially in children and young adults [2 4]. More recently, decellularization has been proposed as a promising alternative for processing biological tissues. In theory, the complete removal of cells and cellular debris results in a nonantigenic extracellular matrix, which is capable of being repopulated in vivo after implantation, and hence with the potential for enhanced durability [5 8]. Accepted for publication Aug 12, 2010. Address correspondence to Dr da Costa, Rua Henrique Coelho Neto 55, Curitiba, Paraná, 82200-120, Brazil; e-mail: fcosta13@mac.com. At the Pontificia Universidade Catolica do Parana (PUCPR), we have developed a decellularization technique that is based in a 0.1% sodium dodecyl sulfate (SDS) solution. After extensive in vitro and in vivo testing in the sheep model, we started our clinical experience with decellularized pulmonary valve allografts for right ventricular outflow tract reconstruction during the Ross operation. Results have demonstrated that decellularized pulmonary valve allografts have good biological behavior in humans, with low gradients and competent valves, with no reoperations for allograft dysfunction for as long as 5 years of follow-up [9 11]. Others have reported similar results, with different decellularization techniques [12 14]. Although decellularized allografts perform well in the right side of the heart, there are concerns that, owing to altered physical properties of the decellularized tissue, complications such as progressive aneurysmatic conduit dilation or valvular cusp prolapse might occur when subjected to the higher pressures on the systemic circulation [15]. In the medical literature, there is only one report on the use of decellularized aortic valve allograft (DAVA) for AVR, showing favorable short-term results [16]. The purpose of this study is to demonstrate our short- 2010 by The Society of Thoracic Surgeons 0003-4975/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2010.08.022

Ann Thorac Surg DA COSTA ET AL 2010;90:1854 61 DECELLULARIZED AORTIC VALVE ALLOGRAFTS 1855 term and midterm results of AVR with DAVA used as a root replacement in a selected high-risk subset of patients. Patients and Methods This study was conducted at Santa Casa de Misericórdia de Curitiba-PUCPR and the Institute of Neurology and Cardiology of Curitiba and was approved by the Ethical Committee (CEP 5300). All patients gave an informed consent to have this new valve. Aortic Valve Allografts All AVA were processed at the Human Heart Valve Bank of Santa Casa de Curitiba according to protocols previously described [17]. In the beginning of the experience, cryopreserved allografts were thawed and subsequently descellularized before implantation. More recently, allografts were sterilized in Russel Park Memorial Institute (RPMI) nutrient medium 1640 containing antibiotics, and decellularized still fresh without cryopreservation. Decellularization was accomplished in a patented solution based in SDS 0.1% for 24 hours under continuous stirring in shakers, followed by consecutive washing in Ringer Lactate solution for 10 days to assure complete removal of the detergents. Histologic control to verify tissue acellularity and adequate preservation of the extracellular matrix was done with a segment of the distal aortic wall with hematoxylin and eosin (HE) and Movat pentachromic staining (Fig 1). Patients Between October 2005 and February 2010, 41 patients underwent AVR using DAVA as a root replacement. Twenty-eight were male, and ages varied between 0.1 and 71 years (median 34). The majority of patients had at least one associated pathology or risk factor, including multivalvular disease (29%), reoperations (36%), active bacterial endocarditis (17%), ascending aorta/arch aneurysm (12%), coronary insufficiency, (7%) or associated congenital cardiac anomalies (4%). Only 7 patients (17%) were elective isolated primary AVR. Patient demographics are listed in Table 1. Operative Technique Operations were performed with extracorporeal circulation with moderate hypothermia at 32 C, and myocardial Table 1. Clinical Data on 41 Patients Undergoing Aortic Valve Replacement With Decellularized Aortic Valve Allografts Clinical Data Age, years 41 (0.1 71) Sex Male 28 Female 13 Valvular lesion Aortic stenosis 12 Aortic insufficiency 21 Mixed lesion 8 Etiology Rheumatic 8 Congenital 12 Degenerative 7 Prosthetic valve dysfunction 7 Endocarditis 7 Associated pathologies Mitral valve disease 12 Coronary insufficiency 4 Ascending aorta/arch aneurysm 5 Pulmonary stenosis 1 Persistent ductus arteriosus 1 Idiopathic hypertrophic 1 subaortic stenosis protection was performed with intermittent antegrade cold blood cardioplegia. Median aortic cross-clamp time was 88 minutes (minimum 55, maximum 166), and extracorporeal circulation time was 128 minutes (minimum 75, maximum 279). The DAVA were implanted with the root replacement technique in all cases, as previously described [18]. Briefly, the aorta was transected just above the sinotubular junction, the aortic valve excised and debridement of all calcium at the annulus performed when necessary. The proximal anastomosis was done with interrupted 4-0 multifilament sutures, putting the allograft in an intraannular position. Whenever possible, the proximal suture line was reinforced with a second running suture of polypropylene 4-0 approximating the remnants of the native aortic wall to the adventitia of the DAVA. However, in cases with destruction of the aortic annulus by n Fig 1. Histologic control of the sodium dodecyl sulfate decellularization process. (A) Fresh aortic wall (hematoxylin and eosin; original magnification 100). (B) Decellularized aortic wall (hematoxylin and eosin; original magnification 100).

1856 DA COSTA ET AL Ann Thorac Surg DECELLULARIZED AORTIC VALVE ALLOGRAFTS 2010;90:1854 61 infection or in complex reoperations, the first suture line was supported with a Teflon (Impra Inc, Tempe, AZ ) strip. The distal anastomosis was accomplished with running 4-0 polypropylene sutures, and the coronary buttons were anastomosed side to side with continuous 5-0 or 6-0 polypropylene sutures. Cases with important mismatch ( 3 to 4 mm) were adjusted with annular reduction or enlargement as indicated. Associated procedures were performed according to well-established techniques. Details of the operations are listed in Table 2. Clinical Evaluation and Follow-Up All patients were evaluated preoperatively with clinical examination, electrocardiogram, chest roentgenogram, and bidimensional Doppler echocardiography. By Doppler velocities, calculations were done to estimate mean and maximum instantaneous gradients. Aortic valve insufficiency was graded as none, trivial, mild, moderate, or severe [19]. During hospital stay, all complications and causes of death were recorded. All patients had a control echocardiogram before hospital discharge. Patients were oriented to return at 6 and 12 months postoperatively and annually thereafter. For patients not returning to our outpatient clinic, data were collected with the referring physician or directly by phone contact. During 2009, patients were actively located to update their clinical status and asked to undergo a chest computed tomography (CT) scan or magnetic resonance imaging (MRI) study, or both. The CT scans were performed using a GE Healthcare (Buckinghamshire, UK) VCT/XT 64-slice CT model. The calcium score method represented the total volume of the calcium (mm 3 )according to the Agatson index [20]. Table 2. Operative Data on 41 Patients Undergoing Aortic Valve Replacement With Decellularized Aortic Valve Allografts Operative Data Associated procedures Mitral 12 Repair 6 Comissurotomy 5 Replacement 1 Pulmonary homograft 3 Myocardial revascularization 4 Congenital cardiac repair 1 Subaortic membrane resection 1 Septal myomectomy 1 Additional aortic procedures Annular reduction 3 Annular enlargement 1 Aortoplasty 1 Dacron graft 3 Allograft diameter, mm (range) 22 (6 28) Cross-clamp time, minutes (range) 88 (55 166) Extracorporeal circulation time, minutes (range) 128 (75 279) n The MRI measurements were performed using a 1.5T HDX GE system and gradients of 30 mt/m strength. Image acquisitions were processed to determine the aortic annulus, aortic sinus, and sinotubular junction diameter in telesystolic and telediastolic phase. Allograft distensibility was determined as [(dmax dmin)/dmax] 100 [21]. In 1 patient, who underwent reoperation at 18 months due to mitral stenosis, a fragment of the DAVA wall was obtained for histologic analysis with HE and Movat staining. Statistical Analysis Data were expressed as median values with corresponding ranges. Estimation of actuarial survival and freedom from valve-related events were performed with Kaplan-Meier curves and confidence limits of 95%. Comparison of the echocardiography measurements as well as the CT calcium scores at the early and midterm postoperative follow-up were made with the nonparametric matched pairs Wilcoxon test. A linear regression analysis with calcium score as the dependent variable and time as the independent variable was performed to test the null hypothesis that time does not influence calcification of the DAVA. Results Early hospital mortality was 7% (3 of 41 cases). Two patients with active bacterial endocarditis died of sepsis and multiorgan failure despite satisfactory surgical repairs, and the third patient died of low cardiac output syndrome. Clinical information was possible for 35 of the 38 survivors (92% completeness of follow-up), after a mean of 19 months (range, 1 to 53). There was 1 late death, of a newborn operated on within 1 month of age. Despite a normal functioning allograft by echocardiography, the child died suddenly 6 months after the operation. Kaplan-Meier late survival was 90% at 3 years (confidence limits: 86% to 94%). Postoperatively, there was marked functional improvement, and all patients are in New York Heart Association class I and II. There were no cases of thromboembolic or hemorrhagic events, and only 5 patients (13%) with concomitant atrial fibrillation or mechanical mitral valve are under systemic anticoagulation. There were also no new or recurrent cases of bacterial endocarditis. One rheumatic patient needed a reoperation for recurrent mitral stenosis at 18 months of follow-up, but none required reoperation owing to failure of the DAVA. Freedom from reoperation on the DAVA was 100%, and freedom from all causes of reoperations was 96% (confidence limits: 92% to 100%) at 3 years. By echocardiography, early and midterm hemodynamic performance was adequate, with consistently low mean gradient ( Pmed) and maximum instantaneous gradient ( Pmax) across the DAVA (Fig 2, Table 3). With the exception of 1 patient, who presented with an eccentric jet judged as mild to moderate aortic insufficiency at the

Ann Thorac Surg DA COSTA ET AL 2010;90:1854 61 DECELLULARIZED AORTIC VALVE ALLOGRAFTS 1857 Fig 2. Maximum instantaneous gradients after decellularized aortic valve allograft (DAVA) implantation. The regression equation that best defines the gradient trend over time is given by: maximum instantaneous gradient ( Pmax) 8.14 to 0.261 months, suggesting stable or even lower gradients during follow-up (p 0.031). first month postoperatively, all others had none or trivial regurgitation during the follow-up period. Accordingly, there was significant left ventricular mass regression and improvement in left ventricular ejection fraction (Table 3). We did not observe progressive aortic root dilation. Four patients (2 with sequential studies) had late MRI demonstrating normal geometry and root dimensions, with no pathologic dilation of the DAVA. Mean distensibility at the level of the aortic root was 10%, but only 3% at the sinus of Valsalva and sinotubular junction level. Calcium score was measured in 22 patients, 10 of whom had sequential measurements. Overall, mean calcium score was 63 Hounsfield units (HU [minimum 0, Table 3. Early and Late Echocardiographic Data After Decellularized Aortic Valve Allograft Implantation Data Preoperative Early a (n 34) Late b (n 31) Pmax, mm Hg 7 (1 26) 4 (1 16) Pmed, mm Hg 5 (1 17) 2 (1 11) LV septum thickness, 12 (8 16) 13 (10 14) 12 (9 15) mm LV posterior wall 11 (8 16) 12 (10 14) 10 (8 12) thickness, mm LVDD, mm 57 (36 73) 51 (46 66) 50 (37 67) LVSD, mm 36 (19 49) 37 (29 46) 30 (22 50) LVMI, g/m 2 141 (92 252) 137 (108 197) 112 (74 191) Aortic root 33 (26 36) 34 (24 43) dimension, mm Ejection fraction, % 63 (37 73) 58 (49 70) 68 (47 77) a Early echocardiography refers to examinations done during the first 30 days after the operation (n 34). b Late echocardiography is the latest examination done by each individual patient. Mean late echocardiography follow-up time was 18 months (minimum 6, maximum 34). Results are given as median values and ranges (minimum and maximum). Pmax maximum instantaneous gradient; Pmed mean gradient; LV left ventricular; LVDD left ventricular diastolic dimension; LVMI left ventricular mass index; LVSD left ventricular systolic dimension. Fig 3. Computed tomography scan analysis and calcium score determination of decellularized aortic valve allograft (DAVA). The conduit wall of the DAVA is free of calcium whereas some calcification is visible in the native coronary ostia buttons. maximum 894]). Logistic regression analysis demonstrated a small but statistically significant increase in calcium content over time (p 0.036). However, this difference was not observed when comparison was made between the 10 patients with sequential measurements: the median initial calcium score was 50 HU (minimum 0, maximum 894) after a mean follow-up of 7 months, and in the second measurement, it was 75 HU (minimum 0, maximum 402) after a mean follow-up of 18 months (p 0,845; Figs 3 and 4). Histologic analysis of the conduit wall in the 1 patient who underwent reoperation at 18 months of follow-up Fig 4. Total calcium scores for the decellularized aortic valve allografts (DAVA) over time. The regression equation that best defines this line is given by: calcium 23.7 2.3 months, indicating a slight and possibly clinical not important increase in calcium content during follow-up (p 0.036). (HU Hounsfield units.)

1858 DA COSTA ET AL Ann Thorac Surg DECELLULARIZED AORTIC VALVE ALLOGRAFTS 2010;90:1854 61 demonstrated well-preserved elastic and collagen fibers in the media and intimal hyperplasia of moderate intensity. A few cells with fibroblast appearance were present in the media coming from both the luminal and adventicial sides. There was a periadventicial inflammatory reaction, but no lymphocytes were present, suggesting that there was no important immune reaction to the DAVA (Fig 5). Comment In our current practice, we still recommend the use of AVA for various subsets of patients, including bacterial endocarditis patients, patients who do not wish to receive or have contraindications to anticoagulants, young rheumatic multivalvular patients who are not good candidates for the Ross operation, and patients with small aortic roots. Even with the advances in modern biological and mechanical valves, patient-prosthesis mismatch is still not an uncommon event and probably has a significant impact on long-term survival, especially for young patients and in the presence of left ventricular dysfunction [22, 23]. Although cryopreserved AVA have excellent longterm results in older patients, durability is limited in the younger patient, especially those less than 20 years of age [2]. The mechanisms responsible for calcification and structural tissue degeneration are multiple and not completely understood. It is widely accepted that AVA causes an immune reaction, but the correlation between the intensity of this reaction and late graft function is not well defined [3, 4]. Conversely, it is postulated that different methods of sterilization, processing, and storage may induce different degrees of adverse inflammatory reactions and thus may have a significant influence on graft survival [24]. The use of acellular matrices, with low antigenicity and capable of being repopulated in vivo after implantation may, at least in theory, enhance longterm AVA durability [5 8]. During the operations, there were no technical difficulties with DAVA implantation. Decellularized tissue had excellent handling properties and appropriate resistance at the suture lines, comparable to those of the cryopreserved counterparts. During this experience, we had no immediate complication related to the DAVA itself. Some reports, including those from our own institution, have documented favorable short- and mediumterm results with the use of decellularized pulmonary valve allografts in the right ventricular outflow tract, with good hemodynamic performance and significant reduction in the immune reactivity [9 14]. However, the experimental and clinical results of DAVA for AVR are scarce. Baraki and colleagues [25] have demonstrated, experimentally, that DAVA performed better than cryopreserved allografts for as long as 9 months in the sheep model. In their experience, DAVA were partially repopulated in vivo, did not calcify, and had no evidence of pathologic dilation. Our experimental experience (to be published) with DAVA implanted in the subcoronary position have shown that they were intact and normally functioning for as long as 6 months in juvenile sheep. Zehr and colleagues [16] reported their experience with Synergraft DAVA for AVR using the root replacement technique. Although the small number of patients limits their results, they demonstrated that there were no major complications related to the use of DAVA after a mean follow-up of 30 months. The allografts had stable hemodynamic function, and there were no cases of progressive dilation. Furthermore, DAVA had significantly lower PRA levels when compared with conventional cryopreserved AVA. Our results confirm these observations, with the DAVA demonstrating adequate hemodynamics and no progressive valvular regurgitation in 38 patients after a mean follow-up of 19 months. The single case with mild to moderate insufficiency had an eccentric jet, possibly due to some distortion of the DAVA during implantation in a patient with important annular destruction due to endocarditis. Our findings with CT scans and MRI, although preliminary, may give some important information relating to the biological behavior of the DAVA. The MRI images Fig 5. Histologic analysis of a segment of the decellularized aortic valve allograft (DAVA) wall obtained during a reoperation 18 months after implantation. (A) There is intimal hyperplasia and a few cells, with fibroblast appearance that could migrate superficially in the aortic media (arrows [hematoxylin and eosin; original magnification 100]). B) Extracellular matrix of the DAVA, with normally aligned elastic fibers and well-preserved glicosaminoglicans (Movat stain; original magnification 100).

Ann Thorac Surg DA COSTA ET AL 2010;90:1854 61 DECELLULARIZED AORTIC VALVE ALLOGRAFTS 1859 were concordant with the nominal size of the implanted grafts, and the anatomy of the neoaortic root was normal at the all levels. We also demonstrated that the DAVA had some degree of distensibility during the cardiac cycle, especially at the annulus level. Although distensibility index was smaller than predicted for the normal native aortic root, it was similar to those reported by Reid and colleagues [22] for the pulmonary autograft after the Ross operation. These data should be interpreted with caution, as all MRI were performed during the first postoperative year, and the long-term fate of these grafts is still unknown. One very important aspect of this study is that SDS DAVA appear to be less prone to calcification than cryopreserved allografts. El-Hamamsy and colleagues [26] studied, comparatively, 166 patients with a mean age of 66 years receiving either a fresh/cryopreserved AVA or porcine Free-Style porcine valves. The conduit wall of the AVA had premature calcium deposits, and calcification was progressive in both groups up to 8 years. Mean calcium scores for the cryopreserved AVA were 573 and 977 at 1 and 2 years respectively. Intuitively, this progressive calcification will probably be related to long-term valve structural dysfunction. Although a direct comparison is not possible, our patients were younger and had a lower median calcium score of only 63 HU after a mean follow-up of 18 months. In 2 patients, in whom these scores were above 200 HU, this was possibly related to residual calcium at the annular level or around the coronary ostia in older patients. Although linear regression analysis demonstrated a slight and possibly clinical not significant increase in calcium content over time, sequential analysis in 10 patients showed the opposite. Our impression is that DAVA has a slower rate of calcification with some patients showing calcium score of zero with follow-up as long as 22 months. Of course longer follow-up is necessary to draw more definitive conclusions, but if these results are confirmed, this technology may represent a major advance, especially for children and young adults with aortic valve disease. Several experimental studies in sheep and pigs have demonstrated that decellularized allografts are fully repopulated in vivo after implantation [7, 8, 16]. However, histologic analysis of a few clinical explants have shown that in humans, repopulation may be much less complete and dependent on the site of implantation. The thinner and less dense tissue wall of pulmonary valve allografts may be more easily repopulated than the thicker and compact elastic wall of the AVA [10, 11, 27]. In our patient, conduit wall biopsy demonstrated that the extracellular matrix was relatively intact with good preservation of the elastic fibers; however, repopulation was only focal. There were few cells with fibroblast appearance coming both from the luminal side and from the periadventicial inflammatory reaction, but they were sparse and did not penetrate into the deep layers of the media. These findings are very similar to those reported by Miller and colleagues [28], who studied one Synergraft DAVA explanted 2 years after the operation, and suggest that refinements in tissue decellularization techniques or other tissue engineering approaches will be necessary to obtain grafts that are fully repopulated with regenerative and growth potential. This study has obvious limitations owing to the relative small number of patients and short follow-up. Furthermore, it is a retrospective study, and imaging evaluation including echocardiography, CT, and MRI were not performed systematically. Despite these limitations, our results have shown that the early and midterm results with DAVA were safe and satisfactory, with adequate hemodynamic performance for as many as 4 years of follow-up. The DAVA appears to be more resistant to calcification and did not show any major structural and morphologic alteration. If these results are confirmed with longer follow-up periods, DAVA may be a promising alternative for AVR for a selected group of patients. Francisco Diniz Affonso da Costa is chief of cardiovascular surgery and also medical director of the Heart Valve Bank of Santa Casa de Curitiba. He has a national patent for tissue decellularization, which was the technique employed in this study. However, owing to government regulation of human tissue banks in Brazil, the author has no direct financial interest with this valve. References 1. Ross DN. Homograft replacement of the aortic valve. Lancet 1962;2:487 9. 2. O Brien MF, Harrocks S, Stafford EG, et al. The homograft aortic valve: a 29-year, 99.3% follow up of 1,022 valve replacements. J Heart Valve Dis 2001;10:334 44. 3. Dignan R, O=Brien M, Hogan P, et al. Aortic valve allograft structural deterioration is associated with a subset of antibodies to human leukocyte. J Heart Valve Dis 2003;12:382 90. 4. Baskett RJ, Nanton MA, Warren AE, et al. Human leukocyte antigen-dr and ABO mismatch are associated with accelerated homograft valve failure in children: implications for therapeutic interventions. J Thorac Cardiovasc Surg 2003; 126:232 9. 5. O=Brien MF, Goldstein S, Walsh S, et al. The SynerGraft valve: a new acellular (nonglutaraldehyde-fixed) tissue heart valve for autologous recellularization first experimental studies before clinical implantation. Semin Thorac Cardiovasc Surg 1999;11(Suppl 1):194 200. 6. Booth C, Korossis SA, Wilcox HE, et al. Tissue engineering of cardiac valve prostheses I: development and histological characterization of an acellular porcine scaffold. J Heart Valve Dis 2002;11:457 62. 7. Dohmen PM, da Costa F, Yoshi S, et al. Histological evaluation of tissue-engineered heart valves implanted in the juvenile sheep model: is there a need for in-vitro seeding? J Heart Valve Dis 2006;15:823 9. 8. Affonso da Costa FD, Dohmen PM, Lopes SV, et al. Comparison of cryopreserved homografts and decellularized porcine heterografts implanted in sheep. Artif Organs 2004; 28:366 70. 9. da Costa FD, Dohmen PM, Duarte D, et al. Immunological and echocardiographic evaluation of decellularized versus cryopreserved allografts during the Ross operation. Eur J Cardiothorac Surg 2005;27:572 8. 10. Costa F, Dohmen P, Vieira E, et al. Ross Operation with decellularized pulmonary allografts: medium-term results. Rev Bras Cir Cardiovasc 2007;22:454 62.

1860 DA COSTA ET AL Ann Thorac Surg DECELLULARIZED AORTIC VALVE ALLOGRAFTS 2010;90:1854 61 11. da Costa FD, Santos LR, Collatusso C, et al. Thirteen years experience with the Ross Operation. J Heart Valve Dis 2009;18:84 94. 12. Dohmen PM, Lembcke A, Holinski S, et al. Midterm clinical results using a tissue-engineered pulmonary valve to reconstruct the right ventricular outflow tract during the Ross procedure. Ann Thorac Surg 2007;84:729 36. 13. Brown JW, Elkins RC, Clarke DR, et al. Performance of the CryoValve SG human decellularized pulmonary valve in 342 patients relative to the conventional CryoValve at a mean follow-up of four years. J Thorac Cardiovasc Surg 2010;139: 339 48. 14. Bechtel JF, Stierle U, Sievers HH. Fifty-two months mean follow up of decellularized SynerGraft-treated pulmonary valve allografts. J Heart Valve Dis 2008;17:98 104. 15. Korossis SA, Booth C, Wilcox HE, et al. Tissue engineering of cardiac valve prostheses II: biomechanical characterization of decellularized porcine aortic heart valves. J Heart Valve Dis 2002;11:463 71. 16. Zehr KJ, Yagubyan M, Connolly HM, et al. Aortic root replacement with a novel decellularized cryopreserved aortic homograft: postoperative immunoreactivity and early results. J Thorac Cardiovasc Surg 2005;130:1010 5. 17. da Costa FD, da Costa MB, da Costa IA, Poffo, et al. Clinical experience with heart valve homografts in Brazil. Artif Organs 2001;25:895 900. 18. Costa F, Fornazari DF, Matsuda CM, et al. [Dez anos de experiência com a substituição da valva aórtica com homoenxertos valvares aórticos implantados pela técnica da substituição total da raiz]. Rev Bras Cir Cardiovasc 2006;21:155 64. 19. Perry GJ, Helmcke F, Nanda NC, et al. Evaluation of aortic insufficiency by Doppler color flow mapping. J Am Coll Cardiol 1987;9:952 9. 20. Agatston AS, Janowitz WR, Hildner FJ, et al. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol 1990;15:827 32. 21. Reid SA, Walker PG, Fisher J, et al. Quantification of pulmonary autograft characteristics using magnetic resonance imaging. J Heart Valve Dis 2004;13:78 85. 22. Ruzicka DJ, Hettich I, Hutter A, et al. The complete supraannular concept: in vivo hemodynamics of bovine and porcine aortic bioprostheses. Circulation 2009; 120(Suppl):139 45. 23. Smedira NG, Blackstone EH, Roselli EE, et al. Are allografts the biologic valve of choice for aortic valve replacement in nonelderly patients? Comparison of explantation for structural valve deterioration of allograft and pericardial prostheses. J Thorac Cardiovasc Surg 2006;131:558 64. 24. Bechtel JF, Marquardt A, Müller-Steinhardt M, et al. Anti- HLA antibodies and pulmonary valve allograft function after the Ross procedure. J Heart Valve Dis 2009;18:673 80. 25. Baraki H, Tudorache I, Braun M, et al. Orthotopic replacement of the aortic valve with decellularized allograft in a sheep model. Biomaterials 2009;30:6240 6. 26. El-Hamamsy I, Zaki M, Stevens LM, et al. Rate of progression and functional significance of aortic root calcification after homograft versus freestyle aortic root replacement. Circulation 2009;120(Suppl):269 75. 27. Sayk F, Bos I, Schubert U, et al. Histopathologic findings in a novel decellularized pulmonary homograft: an autopsy study. Ann Thorac Surg 2005;79:1755 8. 28. Miller DV, Edwards WD, Zehr KJ. Endothelial and smooth muscle cell populations in a decellularized cryopreserved aortic homograft (SynerGraft) 2 years after implantation. J Thorac Cardiovasc Surg 2006;132:175 6. INVITED COMMENTARY The surgical treatment of aortic valve disease today, particularly in young patients, is not optimal yet. During the last decade, increased interest was focused on valvesparing surgery for a variety of pathologies and patients of different ages due to the absence of ideal valve prostheses. However, repair techniques are demanding and do not always allow growth potential. Either interventional or surgical aortic valvuloplasty in neonatal aortic stenosis is predestined for reinterventions due to residual and recurrent stenosis or iatrogenic regurgitation, which has an impact on the long-term outcome. However, aortic valve regurgitation with dilatation of the aortic root commonly has concomitant cusp pathology that also requires treatment. Therefore, nonviable or gluteraldehyde-fixed materials are used to support the aortic root or increase leaflet cooptation, which can also deteriorate and show no ability to grow. Ideal valve prostheses are required, which shows excellent hemodynamic behavior, freedom of structural deterioration, low incidence of thromboembolic events, resistance to endocarditis, optimal left ventricular regression, and growth potential in childhood. Neither mechanical nor allogenic or xenogenic biologic prostheses fulfill this requirement. Cryopreserved allografts show excellent hemodynamic behavior, but immune response limited durability and ability to grow with the child is not given. Several experimental and clinical studies were performed with tissue-engineered heart valves to reconstruct the right ventricular outflow tract, which provides promising data. The present study by da Costa and colleagues [1] shows early-term and midterm clinical results of decellularized aortic valve allografts. They need to be congratulated because this is the first article showing clinical mid-term results of decellularized valves implanted into the systemic circulation as a full root replacement. The data presented in this article do not only show excellent hemodynamic behavior, means low-pressure gradients, and absence of central regurgitation of decellularized heart valves, but also absence of calcification. These promising results are also important for adult patients, such females who are likely to give birth, patients on dialysis, or other young patients who may not need a heart valve with growth potential. Furthermore, the host recellularization of such a decellularized heart valve allows the patient to have a natural defense again endocarditis, as the endothelium prevents the development of a biofilm. By creating a tissueengineered heart valve, based on a decellularized scaffold, the patient will be his own bioreactor and no complex laboratory procedures are needed. If ingrowth of valvular interstitial cells occurs, regeneration and remodeling potential of the extracellular matrix is given, which could furthermore lead to growth potential for the patient in childhood. Therefore, long- 2010 by The Society of Thoracic Surgeons 0003-4975/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2010.09.035