Retrospective clinical analysis of stented vs. stentless porcine aortic bioprostheses q

Transcription

1 European Journal of Cardio-thoracic Surgery 18 (2000) 46±53 Retrospective clinical analysis of stented vs. stentless porcine aortic bioprostheses q Mario Vrandecic*, Fernando Antonio Fantini, Bayard Gontijo Filho, Ozanan CeÂsar de Oliveira, Idail Martins da Costa JuÂnior, Erika Vrandecic Biocor Institute, Avenida Alameda da Serra, 217, Bairro Vila da Serra, Belo Horizonte, Minas Gerais , Brazil Received 6 September 1999; received in revised form 15 February 2000; accepted 21 February 2000 Abstract Objective: The study was designed to compare hemodynamic performance, structural failure and survival of patients undergoing aortic valve replacement (AVR) with a composite aortic stented or stentless porcine bioprosthesis. Methods: From January 1990 to June 1999, the clinical data of 725 patients undergoing AVR with stented porcine aortic bioprosthesis were reviewed. We de ned two groups of patients with similar clinical characteristics: 202 patients receiving aortic stented and 205 patients stentless valves. The two patients groups were similar in age, sex, valve lesion, valve size, preoperative New York Heart Association (NYHA) class status and follow-up. Results: The number of patients available for follow-up, excluding hospital and late mortality, reoperations and patients lost to follow-up, was 157 for the stented and 175 for the stentless group. There was a higher incidence of rheumatic heart disease in the stented (59%) vs. stentless group (44%), (P ˆ 0:003). Fewer patients had prior aortic bioprosthetic dysfunction in the stented (7.6%) compared to the stentless group (25%) (P, 0:001). The mean intensive care unit stay, hospital mortality and late mortality were similar (P, NS). The total complication rate was higher in the stented (12%) than the stentless (3.4%)(P ˆ 0:005). Valve related death was higher in the stented (2.5%) than the stentless (0%) (P ˆ 0:049). Postoperatively, the aortic effective ori ce area (AEOA) was larger (P, 0:001) and the transvalvular peak and mean gradients were lower in the stentless group (P, 0:001). The lea et tissue degeneration analysis was 8.0% in patients at risk for stented and 0.6% for stentless (P ˆ 0:001). Actuarial analysis disclosed no statistical difference in patient survival between groups (P ˆ 0:18). Reoperations were less frequent in the stentless group (P ˆ 0:010). Conclusions: Hemodynamic bene ts in the stentless group were evident and expressed by larger AEOA, lower gradients, better left ventricular remodeling with signi cant decrease of the left ventricular mass. Lower complication rates, lower reoperation rates, less lea et tissue degeneration, and lower valve related mortality rates were seen in the stentless group. A controlled clinical comparison trial with longer follow-up will be required to con rm these clinical and hemodynamic bene ts. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Porcine aortic stented bioprosthesis; Stentless aortic heart valve; Heterologous tissue heart valve; Tissue heart valve substitutes 1. Introduction After the successful experimental work of Morsch, Duran and Gunning in 1964, Jean Paul Binet in Paris implanted the rst porcine bioprosthesis in man [1]. Since then the use of aortic porcine bioprosthesis has provided patients with an excellent quality of life with low risk of major thromboembolic events and freedom from anticoagulant-related hemorrhage. An important drawbacks to the use of bioprosthesis have been the level or degree of structural tissue degeneration (STD) which has primarily been related to patient's age [2±4]. The clinical results with the use of porcine stented q Presented at the 13th Annual Meeting of the European Association for Cardio-thoracic Surgery, Glasgow, Scotland, UK, September 5±8, * Corresponding author. Tel.: ; fax: address: biocor@biocor.com.br (M. Vrandecic). bioprosthesis have shown satisfactory results as far as hemodynamics performance and limited durability [5]. A comparison with mechanical valves has also shown favorable results with respect to freedom from all valves-related adverse effects at 10 years in patients 60 years of age and older [6]. Efforts to produce a longer-lasting tissue valve, for use in young patients, has spurred research to identify the causes of STD, such as the use of aldehyde [7,8] and, more recently, collagen tissue fatigue as an important cause of STD pointed out by Vyavhare [9]. Since tissue stress is an important determinant of valve failure, the absence of the valve stent may improve heterologous valve durability. This concept that gave origin to the stentless design late in 1988, with major clinical application beginning in 1990 for the Biocor± SJM aortic porcine stentless valve /00/$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S (00)

2 M. Vrandecic et al. / European Journal of Cardio-thoracic Surgery 18 (2000) 46±53 47 The present study was designed as to determine and compare the hemodynamic performance, structural tissue degeneration and survival in patients undergoing AVR over the same time period, with the use of stented or stentless valves at the same institution. 2. Patients and methods 2.1. Tissue heart valves The heterologous valves used in this study were manufactured by Biocor InduÂstria e Pesquisas Ltda, since 1979, in Belo Horizonte, Brazil. In 1996 Biocor was acquired by St. Jude Medical, Inc., St. Paul, MN. The company continues to provide tissue heart valve substitutes with the following characteristics: fresh tissue harvesting, tissue selection and individual cuspal tanning by balanced salt and aldehyde solution and processing without pressure during xation. Composite mounting of selected porcine cusps into an anatomical exible stent, (provided with bovine pericardial bias tape at the out ow edge to avoid tissue contact with the polyester stent covering), and individual valve testing with pulse duplicator. The stentless model is composed of three selected aortic porcine cusps, attached to the inner aspect of a scalloped shape bovine pericardial tube, with anatomical similarity to the native aortic valve. This design provides improved effective ori ce area (AEOA), since the sewing ring and stent (part of the stented) no longer exists in the stentless model (Fig. 1) Patients From January 1990 to June 1999, 205 patients received porcine aortic stentless bioprosthesis implantations at our institution. During the same time period, 725 patients underwent AVR with porcine aortic stented bioprosthesis. Since these groups were not identical in relation to age and other clinical variables, a computer model was design to select, from these 725 patients, those with ages ranging between 30 and 79 years, which was the age range of the stentless group. The next step that the computer model performed was to select those with similar implant dates, accepting those operated within 3 months of the stentless patients. At the Fig. 1. Biocor±SJM tissue valves used were of composite design. end of this analysis 202 patients were selected in the stented group, which were similar to the 205 stentless patients. An analysis of these two groups disclosed no statistical difference of the chosen preoperative variables such as sex, valve lesion, NYHA functional class, and bioprosthesis diameter as shown in Table 1. The mean age was 48.1 ^ 17.4 years for the stented and 46.1 ^ 16.4 years for the stentless (P ˆ 0:23). There were 129 patients (64%) of male gender for the stented group and 148 patients (72%) for the stentless (P ˆ 0:09). The comparison between preoperative functional NYHA class in both groups showed similar data (P ˆ 0:80). The type of valve lesion and the bioprosthesis diameter were also equivalent in both groups (respectively, P ˆ 0:20 and P ˆ 0:48). The most frequently used valve sizes were 25 mm or smaller in both groups (71% for the stented and 73% for the stentless). Sizes: 19±21 mm represented 8.6% for the stented vs. 12% in the stentless (P ˆ 0:27). The follow-up ranged from 3 to 114 months with a mean of 53 months (4.4 ^ 1.8 years) for the stented patients and 55 months (4.6 ^ 1.5 years) for the stentless patients (P ˆ 0:22). The total patient follow-up was 693 patients-years for the stented group and 801 patient-years for the stentless group. Table 2 shows the differences related to etiology and associated procedures between both groups. The assessment of the preoperative data disclosed a higher incidence of rheumatic heart disease (RHD) (59%) in the stented compared to the stentless group (44%)(P ˆ 0:003). Fewer Table 1 Preoperative comparative patient pro le of both groups Stented (n ˆ 202) mean ^ SD n (%) Stentless (n ˆ 205) mean ^ SD n (%) P-value Patient pro le Age (30±79 years) 48.1 ^ ^ Gender Male 129 (64) 148 (72) 0.09 Female 73 (36) 57 (28) Valve lesion Insuf ciency 77 (38) 96 (47) Stenosis 87 (43) 75 (36) 0.20 Mixed lesion 38 (19) 34 (17) NYHA preoperative functional class I±II 50 (25) 45 (22) III 97 (48) 102 (50) 0.80 IV 55 (27) 58 (28) Bioprosthesis diameter 19 0 (0.0) 2 (1.0) (8.6) 23 (11) (28) 64 (31) (35) 60 (29) (23) 49 (24) (5.0) 7 (3.5)

3 48 M. Vrandecic et al. / European Journal of Cardio-thoracic Surgery 18 (2000) 46±53 Table 2 Etiology and associated procedures Stented (n ˆ 202) n (%) Stentless (n ˆ 205) n (%) P-value Etiology Rheumatic HD 119 (59) 91 (44) Senile aortic degeneration 34 (17) 12 (6.0) Endocarditis 7 (3.4) 11 (5.5) 0.49 Congenital 14 (6.8) 26 (12) 0.07 Aortic bioprosthesis dysfunction 15 (7.6) 51 (25), Other 13 (6.7) 14 (6.8) 0.97 Associated procedures n ˆ 88 (43.5%) n ˆ 40 (19.5%) Mitral valve replacement 37 (18) 23 (11) 0.06 Myocardial revascularization 30 (15) 9 (4.4) Mitral valvuloplasty 10 (5.0) 7 (3.4) 0.60 Other 11 (5.4) 1 (0.5) None 114 (56) 165 (80), patients with aortic bioprosthetic dysfunction were operated on in the stented (7.6%) than in the stentless group (25%) (P, 0:001), because of our preference to use a stentless valve in patients requiring second tissue valve. Higher incidence of associated procedures seen in the stented group may be explained because of the need for longer cardiopulmonary bypass time for the stentless group. The following parameters were comparable: endocarditis (3.4%) on the stented group and (5.5%) for the stentless (P ˆ 0:49). Preoperatively, 19 (9.2%) of the stented patients and 17 (8.2%) of the stentless were in atrial brillation (P ˆ 0:82). Data collected included cardiopulmonary bypass, intensive care unit stay and hospital stay. Total and partial cardiopulmonary bypass time was compared between patient groups with and without associated procedures. Analysis of the echocardiographic data compared the following parameters: Left ventricular function: ejection fraction (LVEF), end systolic volume (LVESV), interventricular septal thickness (IST), posterior wall left ventricular thickness (PWLVT), LV mass (LVM), aortic effective ori ce area (AEOA), transvalvular peak and mean gradients (TPG± TMG), aortic competence (AC) and aortic lea et tissue degeneration (ALTD) De nitions We followed the guidelines of Edmunds [10] for the de nitions of morbidity and mortality in heart valve replacement. The incidence of thromboembolism was assessed according to the standard `proposal for reporting thrombosis, embolism and bleeding after heart valve replacement' by Bodnar and collaborators Echocardiographic assessment The echocardiographic analysis was done before surgery, 1 month after hospital discharge and yearly thereafter. Two senior sonographers were responsible for the nal reading of the echocardiography data. The revision and analysis of the echo ndings was performed periodically and at the end of the study closing, no major differences were found. The ATL±HDI 3000 with a 2±3-MHz transducer was used. M- mode and 2-dimensional echocardiography. Both M-mode recordings and of ine measurements were guided by the 2- dimensional image. M-mode measurements were made according to the recommendations of the American Society of Echocardiography. The ejection fraction was calculated according to Teichholz and associates and the LV mass by means of the cube formula [11]. Doppler measurements. Blood ow velocity in the LV out ow tract was estimated by pulsed wave Doppler from an apical 4-chamber view. Peak velocities during systole and diastole were measured. Continuous wave Doppler signals were recorded from multiple windows by a 2±3-MHz probe. Pressure gradients were calculated according to the simpli ed Bernoulli equation (Doppler pressure gradient ˆ 4 [peak velocity 2 ]). The mean gradients were calculated from off-line planimetry of the continuous wave Doppler recordings. The effective ori ce area was calculated according to the integral method [12]. Tissue degeneration analysis assessed by echocardiography followed standard parameters using M-mode, 2-dimensional and color Doppler and were graded as mild, moderate and severe. The overall valve function was assessed including effective ori ce area, lea et morbidity, valve gradient and valve competency. Lea et thickness, extent of brous degeneration, lea et refringency and calci cation (acoustic shadow) were all taken into consideration for the analysis. Mild degeneration was considerate when lea et thickness measured between 3 and 4 mm. Fibrosis involved one third of the lea et with less functional activity. Refringency compared the `white' color to the gray echo color scale. Moderate and severe structural failure was considerate when higher degree of alterations were encountered Statistical analysis Statistical analysis was done on extracted data from les formatted for the statistical analysis system with the Epi

4 M. Vrandecic et al. / European Journal of Cardio-thoracic Surgery 18 (2000) 46±53 49 Info Software version 6.04b ± Center for Disease Control, Atlanta, GA [13]. Signi cance between means was determined using the unpaired t-test. The comparison between proportions was employed the chi square test and, when necessary, Fisher's exact test. The Kaplan±Meier life table technique was used to provide actuarial estimates [14]. The log rank test for independent groups was used to test the signi cance of differences. A P-value of,0.05 was considered signi cant. 3. Results Fig. 2. Hospital mortality. The mean intensive care unit stay was 2.5 days for the stented group and 2.6 days for the stentless group (P ˆ 0:50). The total hospital stay for the stented group was 9.2 days and 11.0 days for the stentless group (P ˆ 0:004). The stentless group with an isolated valve procedure had a longer hospital stay compared to stented group (11.2±9.9 days, P, 0:001). When other surgical procedures were performed, the same tendency was noted (11.0±8.7 days, P ˆ 0:022). The hospital mortality was 11 patient deaths for the stented (5.4%) and nine for the stentless (4.4%), with P ˆ 0:79 (Fig. 2). There was no statistical difference in hospital mortality between groups, with or without associated procedures. Forty-four percent of the stented patients and 20% of the stentless group had associated surgical procedures (P, 0:001). Table 3 shows the cardiopulmonary bypass times: a mean of 86.3 min for the stented group and 99.9 min for the stentless group, (P, 0:001). The aortic cross clamping time was 65.0 min for the stented and 80.9 min for the stentless group (P, 0:001). Although a longer extra corporeal time was needed for the stentless patients this did not affect the nal outcome. The follow-up ranged from three months to 114 months with a mean of 53 months (4.4 ^ 1.8 years) for the stented patients and 55 months (4.6 ^ 1.5 years) for the stentless patients (P ˆ 0:22). The total patient follow-up was 693 patient-years for the stented group and 801 patient-years for the stentless group. Late mortality was represented by 16 (8.4%) patient deaths for the stented and 14 (7.1%) for the stentless. There was no statistical difference between both groups when considering cardiac and no-cardiac death (P ˆ 1:00) (Fig. 3). Among 16 late death in the stented group, ve were of cardiac origin (two patients with acute myocardial infarction and three with progressive heart failure). The remaining deaths were not cardiac-related. Among 14 late deaths in the stentless group, four were of cardiac origin (one due to ventricular arrhythmia, one of acute myocardial infarction, Table 3 Cardiopulmonary bypass time ± hospital stay Stented mean ^ SD Stentless mean ^ SD P-value CPB-time (min) Without associated procedure 69.5 ^ ^ 26.4, With associated procedure ^ ^ All patients 86.3 ^ ^ 28.8, CCT (min) Without associated procedure 51.5 ^ ^ 23.4, With associated procedure 81.7 ^ ^ All patients 65.0 ^ ^ 25.2, Intensive care unit stay (days) Without associated procedure 2.4 ^ ^ With associated procedure 2.6 ^ ^ All patients 2.5 ^ ^ Total hospital stay (days) Without associated procedure 8.7 ^ ^ 7.5, With associated procedure 9.9 ^ ^ All patients 9.2 ^ ^

5 50 M. Vrandecic et al. / European Journal of Cardio-thoracic Surgery 18 (2000) 46±53 Fig. 3. Late mortality in both groups broken down into cardiac and noncardiac causes. one of AV block and one with progressive heart failure). The other ten late deaths were not cardiac related. The actuarial analysis (excluded hospital mortality and reoperations) disclosed no statistical difference in patient survival between both groups with P ˆ 0:18, as shown in Fig. 4. The number of patients available for follow-up, excluding hospital and late mortality, reoperations and patients lost to follow-up, was 157 patients in the stented and 175 patients for the stentless group. Although there were three reoperations on the stentless group, it is worth noting the fact that two reoperations were due to paravalvular leaks in the stentless group, with only one replaced. One other valve with partial inferior suture dehiscence was repaired. There were eight reoperations in stented group. The actuarial analysis regarding freedom from all reoperations shows that they were less frequent on the stentless group (P ˆ 0:010) (Fig. 5). The following adverse effects were found: structural failure 1.3% for the stented and 0.0% for the stentless group (P ˆ 0:22), paravalvular leak in 1.3% for the stented and 1.1% for the stentless group (P ˆ 1:00), endocarditis 2.5% for the stented and 0.6% for the stentless group (P ˆ 0:19), thromboembolism 1.9% for the stented and 0.0% for the stentless group (P ˆ 0:10). All adverse effects were 12% for the stented and 3.4% for the stentless group (P ˆ 0:005). Valve related mortality was 2.5% for the stented and 0.0% for the stentless group (P ˆ 0:049) (Table 4). There were fewer patients in atrial brillation in both groups at the end of this follow-up, when compared with initial data: preoperatively, 19 (9.2%) of the stented patients were in atrial brillation and 17 (8.2%) of the stentless patients (P ˆ 0:82). At the end of this follow-up, there were eight patients (5.0%) for stented and ve patients (3.0%) for stentless (P ˆ 0:44). The comparison of the echocardiographic data showed statistical equivalence in the following variables: LVEF, LVESV, IST, PWLVT, aortic regurgitation and stenosis. The regression between preoperative and postoperative LV mass index had the P, 0:001, in the stented and stentless groups, independently. The AEOA was larger (P, 0:001) and TPG, TMG were lower in the stentless group (P, 0:001) (Table 5). The lea et tissue degeneration analysis disclosed 13 patients were at risk (8.0%) in the stented and one patient (0.6%) in the stentless (P ˆ 0:001) (Fig. 6). Both groups were equivalent when compared postoperative functional NYHA class, (P ˆ 0:39) (Table 6). 4. Discussion Ever since Ross [15] rst implanted a human homograft in the aortic position, similar efforts have continued to the present with favorable clinical results [16,17]. Higher demand and limited supplies triggered the current availability of aortic homografts in a stentless design. Over the last decade there has been consistently reports of hemodynamics bene ts with stentless valves, such as lower gradients and effective postoperative LV mass index reduction [18,19]. The Biocor±SJM porcine aortic stentless valve is a composite (three selected porcine aortic cusps) with conical design allowing a good t for most aortic pathologies. The use of stentless valves (when homograft is not available) is generally accepted even in patients with aortic valve endocarditis Fig. 4. Actuarial survival between stented and stentless groups.

6 M. Vrandecic et al. / European Journal of Cardio-thoracic Surgery 18 (2000) 46±53 51 Fig. 5. Actuarial freedom from reoperations between the stented and stentless groups. [20]. The aortic stentless valve usage may be contraindicated when the surgeon is unfamiliar with the technique, when concomitant ascending aortic diseases exists or when the duration of aortic cross clamping is of concern. The present study compares the clinical outcomes of two patient groups undergoing AVR during the same time period and with similar relevant preoperative clinical characteristics. The fact that both valves were manufactured by the same rm, made with composite porcine aortic valves, and processed without pressure xation, allows a good comparison of clinical outcomes with these two different valve designs. The nature of this retrospective study raises questions about differences between the groups. Among these is the fact that there were relatively more preoperative patients with rheumatic heart disease in the stented group. This re ects our ongoing preference for stentless valves in patients with a failed stented bioprosthesis. This criteria is also re ected by the increased number of reoperations required because of bioprosthetic dysfunction in the stentless group. It should be emphasized that at a referral center such as ours, patients with degenerated bioprosthesis had valves made by several manufacturers. It is well known that durability between manufacturers is not uniform, and this may explain the higher number of cases with prosthetic dysfunction encountered in this study. Rheumatic heart disease recurrence is rarely seen in patient with aortic valve disease at 50 years of age. In mitral valve and in a Table 4 Long-term complications Complications Stented (n ˆ 157) n (%) Stentless (n ˆ 175) n (%) P-value Thromboembolism 3 (1.9) 0 (0) 0.10 Endocarditis 4 (2.5) 1 (0.6) 0.19 Paravalvular leak 2 (1.3) 2 (1.1) 1.00 Structural failure 2 (1.3) 0 (0.0) 0.22 All complications 19 (12) 6 (3.4) Valve-related death 4 (2.5) 0 (0) younger age group, recurrence represents major problem. The same may be applicable to atrial brillation, which is more frequent in young patients with mitral valve involvement. There were a higher number of associated procedures, including mitral valve replacement and coronary revascularization, in the stented group. The longer extra-corporeal circulation time needed for the stentless valve placement limited the use of stentless valves in these patients. The hospital mortality was similar in both groups, including those with or without associated procedures. The slightly lower mortality encountered in the stentless group was acceptable with no speci c reason found on the analysis. Shorter hospital stay (2 days less) related to fewer hours of respiratory assistance, longer cardiopulmonary by-pass time, the same duration of ICU stay. This study has shown favorable valve hemodynamics and LV remodeling with a trend towards less tissue degeneration in the stentless group. Tissue calci cation reported in stentless valves is, in general, more prevalent in the aortic wall and less in the cusps. Since Biocor±SJM includes minimal amount of aortic wall in its design this may favor durability. The choice of an aortic bioprosthesis versus a mechanical valve should be individualized according to age, prospective survival, and the basis of patient's expected compliance with the anticoagulation regime [21,22]. Results obtained with the stentless Biocor±SJM valves are comparable or better to their stented counterparts at 10 and 13 years [6]. Both stented and stentless valves in this study showed good clinical performance. This retrospective comparative analysis looked at the similarities in patients' ages, valve lesions, and preoperative clinical functional class in both groups. The end point of this study was the comparison of the hemodynamic performance and structural tissue degeneration over approximately 10 years of follow-up. Several studies have shown excellent hemodynamics performance with this aortic stentless when compared to the stented valve similar to the results seen with homografts [23±25]. In this study, relevant data obtained by the postoperative echodoppler examination during follow-up included a larger effective ori ce area, lower transvalvular

7 52 M. Vrandecic et al. / European Journal of Cardio-thoracic Surgery 18 (2000) 46±53 Table 5 Echodoppler data Echodoppler results Stented (n ˆ 157) mean ^ SD Stentless (n ˆ 175) mean ^ SD P-value Aortic effective ori ce area 1.4 ^ ^ 0.4, Transvalvular peak gradient 27.2 ^ ^ 10.6, Mean transvalvular gradient 14.8 ^ ^ 6.1, Preoperative left ventricular mass (g-m 2 ) 153 ^ ^ 24, Postoperative left ventricular mass (g-m 2 ) 129 ^ ^ gradients with the mean of 9 mmhg. It was noteworthy that more than 70% of this population required stented and stentless valves sizes 25 mm or smaller. Favorable postoperative LV mass regression was demonstrated in both groups with 30% lower LV mass index in the stentless patients. Aortic regurgitation was rare, trivial and non-progressive. Comparison of the second parameter, lea et tissue degeneration, was favorable in the stentless group with only one patient with mild lea et change versus 13 patients with mild or moderate degeneration in the stented group. Hemodynamic performance, long term LV remodeling and structural tissue failure were signi cantly better in the stentless patients. 5. Study limitations The present study was not prospectively randomized. This was a retrospective comparative analysis between two patient groups similar in speci cally chosen parameters, known to be determinants of valve durability (age, sex, valve lesion, valve size, preoperative NYHA class and follow-up). There were more patients with rheumatic heart diseases in the stented group and a higher incidence of bioprosthesis dysfunction indicating valve replacement in the stentless group. Although these limitations exist, both groups were comparable in the most clinically important parameters. The patient's relatively young age, long follow-up and use of a tissue valves with identical xation Table 6 Postoperative NYHA functional class NYHA Stented (n ˆ 165) n (%) Stentless (n ˆ 174) n (%) I±II 142 (86) 158 (91) III 13 (8.1) 9 (5.1) IV 10 (5.9) 7 (3.9) methods provides a relevant clinical comparison between the use of stented and stentless valves in the aortic position. 6. Conclusions Both the aortic porcine stented bioprosthesis and the aortic porcine stentless valves performed well during up to 10 years follow-up (mean of 4.4 years). Hemodynamic bene ts in the stentless group were evidenced by a larger aortic effective ori ce area, lower gradients, and better LV remodeling with signi cant decreased of the LV mass index. There was no structural tissue degeneration requiring reoperation in the stentless group. Lower complication rates, lower reoperation rates, and lower valve-related mortality rates were seen in the stentless group. Longer follow-up studies will be required to con rm continuing clinical and hemodynamic bene ts, continuing lack of tissue degeneration, and better quality of life with the use of the stentless bioprosthesis in the aortic position. Acknowledgements The authors gratefully acknowledge the statistical assistance of EugeÃnio Marcos Andrade Goulart, MD, who is responsible for the medical statistical of this Institute. References Fig. 6. Tissue lea et degeneration detected during echocardiographic examination. [1] Binet JP, Duran CG, Carpentier A, Langlois J. Heterologous aortic valve transplantation. Lancet 1965;2:1275. [2] Jamieson WRE, Munro AI, Miyagishima RT, Allen P, Burr LH, Tyers GFO. Carpentier-Edwards standard porcine bioprosthesis-clinical performance to seventeen years. Ann Thorac Surg 1995;60:999± [3] Fann JI, Miller DC, Moore KA, Mitchell RS, Oyer PE, Stinson EB, Robbins RC, Reitz BA, Shumway NE. Twenty-year clinical experi-

8 M. Vrandecic et al. / European Journal of Cardio-thoracic Surgery 18 (2000) 46±53 53 ence with porcine bioprosthesis. Ann Thorac Surg 1996;62:1301± [4] Burr LH, Jamieson WRE, Munro AI, Miyagishima RT, German E. Porcine bioprosthesis in the elderly: clinical performance by age groups and valve positions. Ann Thorac Surg 1995;60:S264±S269. [5] Vrandecic M, Gontijo Filho B, Paula e Silva JA, Fantini FA, Barbosa JT, SaÄo Jose MC, Dos Santos Pinto CA, Vieira GL, Oliveira HG, Rabelo S. Clinical results with the Biocor porcine bioprosthesis. J Cardiovasc Surg (Torino) 1991;32(6):807±813. [6] Myken PS, Caidahl K, Larsson S, Berggren HE. 10-year experience with the Biocor porcine bioprosthesis in the aortic position. J Heart Valve Dis 1994;3(6):648±656. [7] Hilbert SL, Barrick MK, Ferrans VJ. Porcine aortic valve bioprosthesis: A morphologic comparison of the effects of xation pressure. J Biomed Mater Res 1990;24:773±787. [8] Flomenbaum MA, Schoen FJ. Effects of xation back pressure and antimineralization treatment on the morphology of porcine aortic bioprosthetic valves. J Thorac Cardiovasc Surg 1993;105:154±164. [9] Vyavahare N, Ogle M, Schoen FJ, Zand R, Gloeckner DC, Sacks M, Levy RJ. Mechanisms of bioprosthetic heart valve failure: fatigue causes collagen structural denaturation and glycosaminoglycan loss. J Biomed Mater Res 1999;46:44±50. [10] Edmunds LHJ, Clark RE, Cohn LH, Miller DC, Weisel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. J Thorac Cardiovasc Surg 1998;96(3):351±353. [11] Teichholz LE, Kreulen T, Herman MV, Gorlin R. Problems in echocardiographic volume determinations: echocardiographic-angiographic correlations in the presence or absence of asynergy. Am J Cardiol 1976;37:7±11. [12] Skjaerpe T, Hegranaes L, Hatle L. Noninvasive estimation of valve area in patients with aortic stenosis by Doppler ultrasound and twodimensional echocardiography. Circulation 1985;72:810. [13] Dean AG, Dean JA, Coulombier D, Brendel KA, Smith DC, Burton AH, Dicker RC, Sullivan K, Fagan RF, Arner TG. Epi Info, Version 6: Word Processing, Database and Statistics Program for Epidemiology on microcomputers. Atlanta, GA: Centers of Disease Control and Preservation, [14] Campos Filho N, Franco EL. KMSURV ± A microcomputer program for univariate survival data analysis. Comp Meth Prog Biomed 1998;27:223±228. [15] Ross DN. Homograft replacement of the aortic valve. Lancet 1962;2:487±488. [16] Yacoub M, Rasmi NRH, Sundt TM, Lund O, Boyland E, Radley- Smith R, Khaghani A, Mitchell A. Fourteen-year experience with homovital homografts for aortic valve replacement. J Thorac Cardiovasc Surg 1995;110:186±194. [17] Kirklin JK, Smith D, Novick W, Naftel DC, Kirklin JW, Paci co AD, Nanda NC, Helmcke FR, Bourge RC. Long-term function of cryopreserved aortic homografts: a ten-year study. J Thorac Cardiovasc Surg 1993;106:154±166. [18] Christakis GT, Joyner CD, Morgan CD, Fremes SE, Buth KJ, Sever JY, Rao V, Panagiotopoulos KP, Murphy PM, Goldman BS. Left ventricular mass regression early after aortic valve replacement. Ann Thorac Surg 1996;62:1084±1089. [19] Vrandecic MP, Gontijo BF. Biocor stentless aortic valve at 7 years. In: Huysmans HA, David TE, editors. Stentless bioprostheses, Isis Medical Media, pp. 53±58. [20] Gontijo FB, Vrandecic M, Fantini FA, Oliveira MH, Avelar SO. Porcine stentless aortic valve in replacements and acute aortic endocarditis. J Heart Valve Dis 1995;4:171±175. [21] Peterseim SD, Cen YY, Cheruvu S, Landolfo K, Bashore TM, Lowe JE, Wolfe WG, Glower DD. Long-term outcome after biologic versus mechanical aortic valve replacement in 841 patients. J Thorac Cardiovasc Surg 1999;117:890±897. [22] Beth-Hanssen O, Wallentin I, Larsson S, Caidahl K. Reference Doppler echocardiographic values for St. Jude Medical, Omnicarbon and Biocor prosthetic valves in the aortic position. J Am Soc Echocardiogr 1998;11:466±477. [23] Dellgren G, Erikson M, Brodin LA, Radegran K. The extended Biocor stentless aortic bioprosthesis. Early clinical experience. Scand Cardiovasc J 1999;33(5):259±264. [24] Jin XY, Gibson DG, Yacoub MH, Pepper RJ. Perioperative assessment of aortic homograft. Toronto stentless valve, and stented valve in the aortic position. Ann Thorac Surg 1995;60:S395±S401. [25] Doty DB, Flores JH, Millar RC. Aortic valve replacement with the Medtronic Freestyle bioprosthesis. Stentless Bioprosthesis 1999:93±101.