A Single-Institution Experience With the Ross Operation Over 11 Years ADULT CARDIAC

A Single-Institution Experience With the Ross Operation Over 11 Years Jürgen O. Böhm, MD, Wolfgang Hemmer, MD, PhD, Joachim-Gerd Rein, MD, PhD, Alexander Horke, MD, Detlef Roser, MD, Gunnar Blumenstock, MD, and Cornelius A. Botha, FCS(SA) Sana Cardiac Surgical Clinic, Stuttgart, Department of Medical Biometry, Eberhard-Karls-University, Tuebingen, Germany; and Cardiac Clinic Bodensee, Konstanz, Germany Background. Although the Ross operation requires double-valve replacement for aortic valve pathology, it is the only autologous, aortic valve replacement available. We report a single-unit s 11-year experience. Methods. Before August 2006, 467 patients (mean age, 41 15 years; 358 males) underwent a Ross operation. The right ventricular outflow tract was repaired with a cryopreserved pulmonary homograft. Follow-up was 94.4% complete. Results. The 30-day mortality was 0.6%. The Kaplan- Meier survival estimate at 120 months was 94.4% 2.9% (standard error [SE], 0.0146). Reoperation was due to autograft failure in 15 patients (7 repairs, 8 replacements), with a Kaplan-Meier freedom from autograft failure measured as reoperation or regurgitation exceeding grade II at 120 months of 94.2% 2.8% (SE, 0.0142). Homograft replacement, mostly due to stenosis, occurred in 11 patients. Freedom from homograft dysfunction, defined as homograft reoperation or peak homograft gradient of 30 mm Hg or more, at 120 months was 79.3% 7.3% (SE, 0.0372). Freedom from all autograft- and homograftrelated reoperations at 120 months was 85.9% 6.3% (SE, 0.0321). Autograft or homograft endocarditis occurred in 8 patients, and 1 patient had simultaneous endocarditis of both valves. Conclusions. Patient survival and freedom from prostheses-related events over 11 years still compares favorably with conventional heart valve prostheses. Mortality and morbidity remain low. Reoperation for autograft or homograft failure is higher than our previous reports, and endocarditis is also evident, 1.9% (9 of 467). Homograft dysfunction is higher in younger recipients. (Ann Thorac Surg 2009;87:514 20) 2009 by The Society of Thoracic Surgeons The pulmonary autograft operation is a complex double-valve procedure for aortic valve malfunction [1]. The benefits are an autologous valve replacement free from anticoagulation and possessing physiologic function, but with the potential dysfunction of 2 valves that have undergone operation. Our experience with the Ross operation dates from 1995 and includes children and adults who required combined procedures. We report the experience of the Sana Cardiac Surgical Clinic, Stuttgart, following earlier publications [2, 3]. Patients and Methods Accepted for publication Oct 30, 2008. Address correspondence to Dr Böhm, Mühlrain 74 b, Stuttgart, D-70180, Germany; e-mail: joboehm@z.zgs.de. This study was approved by the institutional ethics committee. We obtained informed consent from patients or their guardians. From February 1995 to August 2006, 467 patients with a mean age of 41 15 years (range, 1 to 67 years) underwent Ross operations. Patient characteristics and the valve pathology are listed in Table 1. The pulmonary autograft was implanted in 456 patients as a freestanding root [3], and 11 patients with ideal anatomic and geometric relation of autograft to aortic root received a subcoronary implant. The pulmonary valve was always replaced by a cryopreserved homograft. Homografts were sourced from Continental Europe, the United Kingdom, North America, or South Africa. Patients with coronary artery disease were accepted only when amenable to complete arterial grafting. Repairable mitral valve pathology was also accepted, as were other correctable conditions (Table 2). Two physicians undertook the follow-up. This included clinical details and transthoracic echocardiography (2-dimensional, M-mode, color flow, and Doppler echocardiography). Autograft regurgitation was graded according to Perry and colleagues [4], and pulmonary homograft regurgitation was estimated whenever possible with the method of Chan and colleagues [5]. Systolic autograft and homograft gradients were calculated with the modified Bernoulli equation. Diameters of the aortic annulus, sinus of Valsalva, and the sinotubular junction were measured from the parasternal long-axis view. Data Analysis Data are presented as mean standard deviation, unless indicated. Patient survival and time-to-event analyses for homograft or autograft dysfunction were performed using Kaplan-Meier methods. Estimates of survival and 2009 by The Society of Thoracic Surgeons 0003-4975/09/$36.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2008.10.093

Ann Thorac Surg BÖHM ET AL 2009;87:514 20 ROSS OPERATION OVER 11 YEARS 515 Table 1. Preoperative Patient Data Characteristic Mean SD or No. (%) Patients, No. 467 Gender, No. Male 358 Female 109 Age, years 41 15 Weight, kg 75.1 38.3 Height, cm 170.6 15 Body surface area, m 2 1.87 0.38 Aortic stenosis 176 (37.7) Aortic regurgitation 123 (26.3) Combined lesion 144 (30.8) Prosthesis replacement 9 (1.9) Active endocarditis 15 (3.2) Previous thoracic operation 47 (10.1) Ascending aorta aneurysm 83 (17.8) Coronary artery disease 26 (5.6) Unicuspid/bicuspid aortic valve 251 (53.7) NYHA III/IV pre-op 208 (44.5) SD stan- NYHA New York Heart Association functional class; dard deviation. freedom of events are given with 95% confidence intervals. Data analyses were done with SAS 9.1.2 statistical software (SAS Institute, Cary, NC). Results Early Complications and Mortality There were no surgical deaths. Early complications are listed in Table 3. The 30-day mortality was 0.6% (3 of 467), Table 2. Operative Data and Combined Procedures Operative Data Mean SD, or No. (%) Bypass time, min 158 31 Cross-clamp time, min 128 25 Autograft implantation technique Full root replacement 456 (97.6) Subcoronary implant 11 (2.4) Combined procedures 211 (45.2) Ascending aorta remodeling 29 (6.2) Ascending aorta replacement 113 (24.2) Mitral valve repair 8 (1.7) Maze procedure 4 (0.9) CABG 35 (7.5) Ostioplasty 5 (1.1) LVOT enlargement 6 (1.3) ASD closure 5 (1.1) VSD closure 4 (0.9) PAPVR correction 1 (0.2) Pacemaker replacement 1 (0.2) ASD atrial septal defect; CABG coronary artery bypass grafting; LVOT left ventricular outflow tract; PAPVR partial anomalous pulmonary veins return; SD standard deviation; VSD ventricular septal defect. Table 3. Early and Late Complications Complications No. (%) Early 30-day mortality 3 (0.6) Reentry for bleeding, second look, etc 10 (2.1) Acute autograft rupture 1 (0.2) Deep sternal wound infection 3 (0.6) Minor neurologic deficit 10 (2.1) Major neurologic deficit 2 (0.4) Elevated myocardial enzymes 3 (0.6) Pacemaker insertion 4 (0.9) CABG (not planned) 9 (1.9) Late Late mortality 14 (3.0) Hemorrhagic events 1 (0.2) Thromboembolic events 4 (0.9) Pacemaker insertion 2 (0.4) Cardioverter implantation 1 (0.2) Constrictive pericarditis 1 (0.2) Myocarditis 1 (0.2) Aneurysm of ascending aorta 2 (0.4) CABG coronary artery bypass grafting. all in the later part of the series. The first death occurred after an emergency operation in a 55-year-old woman with pulmonary edema and severe aortic stenosis. She was weaned from cardiopulmonary bypass with a rightheart assist device. Multiorgan failure ensued, and she died 10 days later. The second death was in a 54-year-old woman. Four days after an unremarkable postoperative course, acute myocardial infarction with shock occurred. An emergency operation and revascularization were unsuccessful. Autopsy revealed a blood clot compressing the left mainstem. The third death was a 38-year-old man with atrial fibrillation. Cardiac arrest occurred 9 days after discharge, and a medication-induced bradycardia was suspected. A fatal outcome was prevented in a further patient after acute autograft rupture with exsanguination after extubation. The patient survived without complications after autotransfusion and autograft replacement with a stentless valve conduit. Late Morbidity and Mortality Follow-up of the 464 survivors was 94.4% complete at a mean of 66 4 months, equalling 2397 patient-years. All major and minor complications are listed in Table 3. The study duration was 134 months, during which 14 late deaths occurred. The first was due to sternal infection and mediastinitis in a 46-year-old man. Despite surgical revision, he was readmitted 3 months later with autograft and homograft endocarditis. Both valves were replaced with homografts, but the patient died within days. Six additional patients had a demonstrable cardiac cause of death. Insufficient clinical information was available to determine the cause in a further 4 patients with sudden death after 6 to 64 months. The remaining 3 patients died of documented noncardiac causes. Kaplan-Meier esti-

516 BÖHM ET AL Ann Thorac Surg ROSS OPERATION OVER 11 YEARS 2009;87:514 20 Fig 1. Kaplan-Meier estimates of survival by age group. mate of survival for the series at 120 months is 94.4% 2.9% (standard error, 0.0146). Estimates of survival respective to age are presented in Figure 1. Endocarditis We have not previously documented homograft or autograft endocarditis [2, 3]. We currently report 9 patients with endocarditis: 1 double-valve infection, 4 with autograft endocarditis, and 4 infected homografts. The patients who contracted endocarditis were men with a Fig 3. Kaplan-Meier estimates of homograft reoperation and a gradient of 30 mm Hg or more. mean age of 42.2 11.1 years (range, 22 to 60 years). Autograft function was normal previously. Seven of the endocarditis infections presented within 3 months, and two after 25 and 53 months. The bacteriemia was cultureconfirmed in 5 patients. The remaining 3 patients remained culture-negative, and their diagnoses were made from the clinical course and vegetations on echocardiography. One additional postmortem diagnosis of homograft endocarditis was made in a patient who was presumed to have died of pneumonia and hemoptysis. A medical cure without loss of valve competence was achieved in 5 of the 9 patients with endocarditis. Three required an operation. One of these patients had doublevalve endocarditis and died after double-valve repeat replacement, as described previously. Non-Valve-Related Reoperations Three cardiac non-valve-related reoperations occurred. Two were due to ascending aortic aneurysms, with bicuspid native aortic valves. Autograft function was salvaged despite the redo operation. Symptomatic constrictive pericarditis and moderate tricuspid incompetence developed in a third patient after 19 months. Pericardectomy and tricuspid annuloplasty was required. The pathology was in keeping with the sequelae of Hodgkin lymphoma and subsequent radiotherapy 27 years earlier. No late coronary artery or ostial reinterventions were observed. Fig 2. Kaplan-Meier estimates of autograft reoperation and autograft regurgitation exceeding grade 2 (II ). Reoperation for Autograft Dysfunction No stenosis or autograft calcification was observed. Autograft-related reoperations occurred in 18 patients. Apart from the 3 patients mentioned who required redo operations for autograft endocarditis and the patient

Ann Thorac Surg BÖHM ET AL 2009;87:514 20 ROSS OPERATION OVER 11 YEARS 517 Fig 4. Kaplan-Meier estimates of all autograft- and homograft-related reoperations. treated for acute autograft rupture, another 14 patients required late revision due to autograft regurgitation. Of these, 9 patients received their Ross operation during the first 5 years of this series, which may reflect a learning curve after which we described our stabilizing techniques for the aortic annulus and sinotubular junction in geometric mismatch [6]. In only 3 patients, progressive annular dilatation was the sole cause of autograft regurgitation. It was combined in 5 further patients with prolapse of 1 or more valvecusps. Iatrogenic autograft dysfunction caused by an external stitch through a cusp was discovered in 1 patient at the redo operation. In addition, 2 patients had cusp perforations and 1 a commissural tear as the cause of autograft incompetence, discovered at the redo procedure. The cause of these perforations was uncertain, either due to intraoperative injury, undiagnosed endocarditis, or congenital fenestrations that enlarged. The two remaining cases of autograft reoperation were annular dehiscence and pseudoaneurysm formation below the left sinus. In summary, progressive autograft regurgitation occurred predominately in men (12:2), with a mean age of 48.4 9.8 years (range, 29 to 62 years). Reoperation of failed autografts occurred after a mean of 39.3 20.8 months (range, 9 to 84). Seven of the 18 reoperated autografts could be repaired. Kaplan-Meier estimates of freedom from autograft failure, defined as autograft-related reoperation or autograft regurgitation exceeding grade II at 120 months is 94.2% 2.8% (standard error, 0.0142). Estimates with respect to different age are provided in Figure 2. Reoperation for Homograft Failure Apart from the single patient mentioned, who required a redo operation for double-valve endocarditis, another 11 required an operation for homograft failure, almost exclusively due to stenosis (n 10). Usually, the homograft cusps were less affected by the fibrosis than the wall. The exception to this was one incompetent homograft, where only remnants of cusp tissue remained at the redo operation. As described previously [6], most homograft replacements occur in younger patients. Homograft stenosis has also predominated in male patients (9:1), with a mean age of 25.7 20.1 years and with 7 aged younger than 20 years. Reoperation due to dysfunctional homografts occurred after a mean of 50.8 21.0 months (range, 22 to 90 months). In all patients, the implanted homograft at insertion had a diameter of at least 23 mm, for which guidelines were previously published [3]. For this study, estimate of freedom from homograft failure, defined as homograft-related reoperation or a peak gradient of 30 mm Hg or higher across the homograft at 120 months, is 79.3% 7.3% (standard error, 0.0372). Additional Kaplan-Meier estimates regarding age are illustrated in Figure 3. Reoperations Related to Total Valve Failure A total of 29 valve-related reoperations were required, comprising both autograft and homograft, and resulted Table 4. Echocardiographic Data of 384 Patients During Follow-up Variable Mean SD or % Autograft p max, mm Hg 5.8 2.9 p mean, mm Hg 3.2 1.5 Aortic insufficiency None 17.2 Trace 55.4 Grade I 23.9 Grade I II 3.0 Grade II 0.5 Sinus diameter, mm 36.9 5.0 Annulus diameter, mm 24.8 3.5 EOA, cm 2 3.73 1.52 Indexed EOA, cm 2 /m 2 1.93 0.70 Homograft p max, mm Hg 16.8 11.1 p mean, mm Hg 9.4 6.8 p 30 mm Hg 7.3 Pulmonary insufficiency None 25.7 Trace 29.8 Grade 1 33.9 Grade I II 10.3 Grade II 0.3 EOA effective orifice area; SD standard deviation.

518 BÖHM ET AL Ann Thorac Surg ROSS OPERATION OVER 11 YEARS 2009;87:514 20 in a Kaplan-Maier freedom from all valve-related reoperations at 120 months of 85.9% 6.3% (standard error, 0.0321). Kaplan-Meier estimates representing different ages are illustrated in Figure 4. Echocardiography Echocardiographic follow-up was available in 384 patients. Physiologic gradients were documented for all autografts (mean gradient, 3.2 1.5 mm Hg; effective orifice area, 3.73 1.52 cm 2 ). Apart from those patients reoperated on for autograft failure, 2 asymptomatic patients with a stable autograft regurgitation of grade II are under observation. Another 11 patients with autograft insufficiency of grade I to II at their latest follow-up remain asymptomatic without cardiac dilatation. Stenosis is the dominant cause of homograft dysfunction for the pulmonary homograft. The median peak gradient of 15.1 9.1 mm Hg for the pulmonary homograft is abnormally high. In 27 patients, the pulmonary homograft gradients exceed 30 mm Hg. Clinically relevant homograft regurgitation is absent, however. Echocardiographic data are summarized in Table 4. Comment Early morbidity and mortality increased slightly compared with previous publications from our group [3]. Early morbidity is now 9% (42 of 467), and early mortality has increased to 0.6% (3 of 467). We had previously reported 404 patients [7] without a death. Nevertheless, the current report still compares favorably compared with groups with an early mortality of 2.6% [8], and also when compared with contemporary reports of 4% to 5.3% after mechanical and biologic aortic valve replacement [9 12]. We described a patient who had early autograft dehiscence and left the hospital with a stentless aortic valve prostheses and pulmonary homograft. A further 8 patients required eventual autograft replacement so that in 9 patients, a healthy native pulmonary valve was sacrificed unnecessarily. Another common criticism of the Ross operation is the likelihood of coronary artery complications [13]. In this series, 9 patients (1.9%) received unplanned coronary grafting, mostly for right ventricular ischemia and probably due to reimplant malposition of the right coronary artery. This complication was reported more often in early operations [14]. Obstruction of either coronary ostium is also described with all heart valve prostheses. Some units favor the subcoronary implantation of the pulmonary autograft [15], which does avoid malrotation of the coronary artery buttons. But compromise of either coronary ostium is still possible. Particularly in patients with bicuspid aortic valves, the openings of the coronary arteries are often displaced within the aorta, which adds to the difficulty of the subcoronary insertion. Myocardial infarction some months after surgery ended fatally in an 11-year-old girl, despite a saphenous vein bypass to a small right coronary for right heart failure on weaning from cardiopulmonary bypass. At autopsy, however, the right coronary arterial system was patent, and an anterior myocardial infarction of an unclear etiology was found [7]. Favorable outcome for the Ross operation in active endocarditis has been reported by our group and others [16, 17]. Some have suggested that the viable autologous tissue of the autograft is resistant to infection. Our observations are now ambivalent. Although 15 patients in this series underwent the operation for active endocarditis, without recurrence, endocarditis did subsequently occur contrary to expectations in some patients without prior endocarditis, as has been described by other authors [18]. The rate of autograft and homograft infection in this series of 1.9% (9 of 467) is still lower than that for manufactured prostheses [9 12, 19], but the expectation of autograft and homograft immunity to infection has not been confirmed, which another group also observed [20]. The most valid concern regarding the Ross operation is autograft longevity within the systemic circulation. Few series [21] other than the original Ross series [22] are longer than 20 years. In earlier reports we speculated that autograft regurgitation was mainly due to annular dilatation. Residual coarctation and upper-body hypertension, we concluded, was a contraindication to the Ross operation. Carr-White and colleagues [23] independently expressed concerns about hypertension and thin autograft tissue. The combination of coarctation of the aorta with bicuspid valve disease is well documented, and the relevance of the latter to the Ross operation has received attention [24 26]. The current autograft failure of 3.2% (15 of 464) is low. We introduced various innovations in an attempt to stabilize the autograft dimensions and function, which may explain the variance of these results to others [8], but the follow-up is not long enough to resolve the debate comparing subcoronary insertion with root replacement techniques. Our group initially favored the root replacement technique, but a small number of patients in whom the sinus dimensions continued to enlarge over years [27, 28], despite buttressing the sinotubular and annulus diameters, led one of the participating surgeons to introduce a few subcoronary insertions later in the series, when the optimal ideal geometric match was present. A comparison of the durability resulting from the two techniques of the Ross operation will only become possible once more units that use different methods for the operation publish their long-term results. The increased number of patients in this series in whom repair of autograft dysfunction was possible and in whom replacement could be avoided deserves mention. This is rarely an option in redo operations for dysfunctional manufactured prostheses. Failure of the pulmonary homograft occurred predominately in 11 younger patients in this series. Fibrosis and shrinkage of the entire conduit was observed in all of these patients and was most pronounced at the muscular portion of the homograft. The cause is speculative [29]. Our series was too small to allow conclusions. Where histology of the explanted homografts

Ann Thorac Surg BÖHM ET AL 2009;87:514 20 ROSS OPERATION OVER 11 YEARS 519 was available, the results were nonspecific, and neither rejection nor infection was documented. Calcification was absent from the specimens. In children and adolescents, we routinely implant the largest homograft possible because homograft size seems to be the most important homograft-related factor for stenosis [30, 31], and in all children it was possible to implant an adult-sized homograft. With one exception in the younger patients, most of the adults received a homograft of a compatible blood group. At the reoperations for degenerated homografts, we used decellularized homografts (Cryolife Inc, Kennesaw, GA) in 2 patients, with disappointing results. Gradual increase of the postoperative gradients still was evident; as a consequence, the remaining 8 such patients underwent antihuman leukocyte antigen (HLA) antibody quantification. An appropriate HLA match was sought according to accepted cardiac transplantation guidelines [32]. It is too soon to determine the success of the latter strategy, but early results are encouraging. In these patients we also remove as much muscle from the replacement homograft proximally as possible and insert a short polyester conduit between the homograft and right ventricle similar to the proposal of using autologous pericardium [33]. Older patients in this study have the advantage over younger patients regarding long-term pulmonary homograft function and are likely to avoid further interventions on the aortic or pulmonary valve for the duration of their lives. This observation is in concurrence with our previous reports [6]. All patients in this series are free of systemic anticoagulation and have physiologic autograft valve function. Several studies [34 36] have documented exercise capacities and physiologic hemodynamic performance of the autograft similar to that of a comparative population without heart valve pathology. Despite no systemic anticoagulation, thromboembolism and hemorrhagic events are rare in this study, with a frequency of 0.9% (4 of 464) and 0.2% (1 of 464) respectively. This concurs with Sievers and colleagues [15], who presented an annual stroke incidence in their Ross patients that equalled a community-based study in Germany, namely 1.82 events/year per 1000 patients. In contrast, the freedom from thromboembolism after 10 years for a mechanical prosthesis varies from 67% to 91%. [37]. In conclusion, the Ross operation can be performed in all age groups with an early morbidity and mortality that compares well with conventional techniques of aortic valve replacement. The total rate of valve failure in this series remains acceptably low. When prostheses failure occurs, for the autograft it is always due to incompetence, and for the homograft, almost exclusively due to stenosis. Homograft dysfunction occurs particularly in younger recipients. This study of 11 years of experience and 467 recipients of the Ross operation describes the causes of autograft and homograft failure. Both patient survival with the pulmonary autograft and freedom from prostheses-related events over 11 years still compare very favorably with conventional heart valve prostheses. Reoperation rates, however, which include both autograft and homograft, are higher than we previously reported. Early on in our experience, we were led to believe that the autograft would stabilize after the first year, and dysfunction would cease to occur. This is not the case. The current endocarditis rate of 1.9 % (9 of 467) is also higher than previously documented. Any supposed immunity that the Ross operation may offer from bacterial endocarditis could not be confirmed. Older recipients of the Ross operation are more likely to avoid reoperations. References 1. Ross DN. Replacement of aortic and mitral valves with a pulmonary autograft. Lancet 1967;2:956 8. 2. Böhm JO, Botha CA, Hemmer W, et al. The Ross operation as a combined procedure and in complicated cases is there an increased risk? Thorac Cardiovasc Surg 2001;49: 300 5. 3. Böhm JO, Botha CA, Rein J-G, Roser D. Technical evolution of the Ross operation: midterm results in 186 patients. Ann Thorac Surg 2001;71:S340 3. 4. Perry GJ, Helmcke F, Nanda NC, Byard C, Soto B. Evaluation of aortic insufficiency by Doppler colour flow mapping. J Am Coll Cardiol 1987;9:952 9. 5. Chan KC, Fyfe DA, McKay CA, Sade RM, Crawford FA. Right ventricular outflow reconstruction with cryopreserved homografts in paediatric patients: intermediate-term follow-up with serial echocardiographic assessment. J Am Coll Cardiol 1994;24:483 9. 6. Böhm JO, Botha CA, Hemmer W, et al. Older patients fare better with the Ross operation. Ann Thorac Surg 2003;75: 796 802. 7. Böhm JO, Botha CA, Horke A, et al. Is the Ross operation still an acceptable option in children and adolescents? Ann Thorac Surg 2006;82:940 7. 8. Takkenberg JJM, Dossche KME, Hazekamp MG, et al. Report of the Dutch experience with the Ross procedure in 343 patients. Eur J Cardiothoracic Surg 2002;22:70 7. 9. Alexiou C, McDonald A, Langley SM, et al. Aortic valve replacement in children: are mechanical prostheses a good option? Eur J Cardiothorac Surg 2000;17:125 33. 10. David TE, Ivanov J, Armstrong S, et al. Late results of heart valve replacement with the Hancock II bioprosthesis. J Thorac Cardiovasc Surg 2001;121:268 78. 11. Butchart EG, Li HH, Payne N, et al. Twenty years experience with the Medtronic Hall valve. J Thorac Cardiovasc Surg 2001;121:1090 100. 12. Emery RW, Krogh C, Arom KV, et al. The St. Jude Medical cardiac valve prosthesis: a 25-year experience with single valve replacement. Ann Thorac Surg 2005;79:776 83. 13. David T. Invited commentary. Ann Thorac Surg 2003;75:802. 14. Botha CA, Böhm JO, Hemmer W. Coronary artery complications and the Ross operation. Ann Thorac Surg 2004;78: 384. 15. Sievers HH, Hanke T, Stierle U, et al. A critical reappraisal of the Ross operation. Circulation 2006;114:504 11. 16. Oswald JD. Pulmonary autograft: Valve of choice for aortic infective endocarditis? In: Yankah AC, Yacoub MH, Hetzer R, eds. Cardiac valve allografts. New York: Springer; 1997: 179 85. 17. Joyce F, Tingleff J, Pettersson G. Treatment of aortic valve endocarditis with the Ross operation. In: Yankah AC, Yacoub MH, Hetzer R, eds. Cardiac valve allografts. New York: Springer; 1997:187 99. 18. Prat A, Saez de Ibarra JI, Vincentelli A, et al. Ross operation for active culture-positive aortic valve endocarditis with

520 BÖHM ET AL Ann Thorac Surg ROSS OPERATION OVER 11 YEARS 2009;87:514 20 extensive paravalvular involvement. Ann Thorac Surg 2001;72:1492 6. 19. Oxenham H, Bloomfield P, Wheatley DJ, et al. A twenty-year comparison of a mechanical heart valve with a porcine bioprosthesis. Cardiac Surg Today 2003;1:52 9. 20. Ashrafian H, Griselli M, Rubens MB, et al. Pulmonary homograft endocarditis 19 years after the Ross procedure. Thorac Cardiovasc Surg 2007;55:55 6. 21. Elkins RC, Elkins CC, Lane MM, et al. Ross operation Sixteen year experience. Presented at: the 85 th AATS meeting, San Francisco, CA, Apr 10 13, 2005. 22. Chambers JC, Somerville J, Stone S, Ross DN. Pulmonary autograft procedure for aortic valve disease. Circulation 1997;96:2206 14. 23. Carr-White GS, Kilner PJ, Hon JKF, et al. Incidence, location, pathology and significance of pulmonary homograft stenosis after the Ross operation. Circulation 2001;104(suppl I):16 20. 24. Kouchoukos NT, Masetti P, Nickerson NJ, et al. The Ross procedure: long-term clinical and echocardiographic followup. Ann Thorac Surg 2004;78:773 81. 25. De Sa M, Moshkovitz Y, Butany J, David TE. Histologic abnormalities of the ascending aorta and pulmonary trunk in patients with bicuspid aortic valve disease: clinical relevance to the Ross procedure. J Thorac Cardiovasc Surg 1999;118:588 96. 26. Botha CA, Rupp W, Böhm JO, Roser D, Rein J-G. The pulmonary autograft (Ross operation) as aortic valve replacement. S Afr Med J 1999;89:C202 8. 27. Hokken RB, Bogers AJJC, Taams MA, et al. Does the pulmonary autograft in the aortic position in adults increase in diameter? An echocardiographic study. J Thorac Cardiovasc Surg 1997;113:667 74. 28. Simon-Kupilik N, Bialy J, Moidl R, et al. Dilatation of the autograft root after the Ross procedure. Eur J Cardiothorac Surg 2002;21:470 3. 29. Xie G-Y, Bhakta D, Smith MD. Echocardiographic follow-up study of the Ross procedure in older versus younger patients. Am Heart J 2001;142:331 5. 30. Freier H, Collart F, Ghez O, et al. Risk factors, dynamics and cutoff values for homograft stenosis after the Ross procedure. Ann Thorac Surg 2005;79:1669 75. 31. Meyns B, Jashari R, Gewillig M, et al. Factors influencing the survival of cryopreserved homografts. Eur J Cardiothoracic Surg 2005;28:211 6. 32. Kouchoukos NT, Blackstone EH, Doty DB, Hanley F, Karp R, eds. Determination of histocompatibility. Kirklin/Barratt- Boyes cardiac surgery. 3rd ed. Philadelphia: Churchill Livingstone; 2003:1733. 33. Schmidtke C, Dahmen G, Stierle, U, Sievers HH. The influence of autologous pericardium used for homograft muscle replacement on valve function after the Ross procedure. Thorac Cardiov Surg 2005;53(suppl1):S23. 34. Oury JH, Doty DB, Oswalt JD, et al. Cardiopulmonary response to maximal exercise in young athletes following the Ross procedure. Ann Thorac Surg 1998;66:S153 4. 35. Sievers HH, Schmidtke C, Graf B. Hemodynamic of semilunar valves at rest and exercise at an average of more than two years after the Ross procedure. J Heart Valv Dis 2001; 10:166 70. 36. Wang A, Jaggers J, Ungerleider RM, et al. Exercise echocardiographic comparison of pulmonary autograft and aortic homograft replacement for aortic valve disease in adults. J Heart Valve Dis 2003;12:202 8. 37. Akins CW. Results with mechanical cardiac valvular prostheses. Ann Thorac Surg 1995;60:1836 44. Southern Thoracic Surgical Association: 2009 Submission of Abstracts Abstracts for papers to be presented for the STSA 2009 Annual Meeting can now be submitted. Abstracts must be submitted online using the CTSNet Online Abstract Submission System, which can be accessed via the STSA Web Site at http://www.stsa.org. Abstracts must be submitted by April 13, 2009, by midnight Central Daylight Time. Paper abstracts cannot be submitted. 2009 by The Society of Thoracic Surgeons Ann Thorac Surg 2009;87:520 0003-4975/09/$36.00 Published by Elsevier Inc