Hemodynamic Performance of the Medtronic Mosaic Porcine Bioprosthesis Up to Ten Years

Transcription

1 Hemodynamic Performance of the Medtronic Mosaic Porcine Bioprosthesis Up to Ten Years Friedrich-Christian Riess, MD, Ralf Bader, MD, Eva Cramer, MD, Lorenz Hansen, MD, Bèr Kleijnen, MS, Gunther Wahl, MD, Jürgen Wallrath, PhD, Stephan Winkel, MD, and Niels Bleese, MD Albertinen Heart Center, Hamburg, Germany, and Medtronic Bakken Research Center, Cardiac Surgery Clinical Research Department, Maastricht, the Netherlands Background. The Mosaic bioprosthesis (Medtronic, Minneapolis, MN) is a third-generation stented porcine bioprosthesis combining physiologic fixation and amino oleic acid antimineralization treatment to improve hemodynamic performance and durability. The findings of this single-center experience with this valve were evaluated to determine the clinical and hemodynamic performance. Methods. Between February 1994 and October 1999, we enrolled 255 patients with aortic valve replacement (AVR) with a mean age of 67 years (range, 23 to 82 years) and 47 patients with mitral valve replacement (MVR) with a mean age of 67 years (range, 41 to 84 years) in this post-united States Food and Drug Administration approval prospective and nonrandomized clinical trial. Patients were followed-up, including serial echocardiographic assessment, within 30 days, at 6 months, and annually thereafter. The cumulative follow-up was 1540 patient-years for AVR (mean, 6.1 years; maximum, 10 years) and 250 patient-years for MVR (mean, 5.4 years, maximum; 10 years). Results. Early mortality after AVR (<30 days) was 0.8%; late mortality per patient-year was 3.5%, including a valve-related/unexplained mortality of 1.1%. Early mortality after MVR (<30 days) was 0.0%; late mortality per patient-year was 2.8%, including a valve-related/unexplained mortality of 1.2%. Median postoperative gradient and effective orifice area for all valves after AVR were (early, n 252; 5 years, n 161; 9 years, n 43) 13.7, 12.3, and 11.7 mm Hg and 1.9, 1.8, and 1.8 cm 2 at early, 5 years, and 9 years, respectively. With MVR respective data were (early, n 46; 5 years, n 25; 7 years, n 13) 4.6, 4.1, and 3.9 mm Hg and 1.8, 2.2, and 2.3 cm 2. At 10 years, freedom from adverse events in the AVR group and MVR group was, respectively, thromboembolism, 86.6% 6.6% and 86.3% 9.8%; permanent neurologic event, 91.2% 6.8% and 90.9% 8.7%; valve thrombosis, 98.2% 0.8% and 100%; structural valve deterioration, 87.1% 6.7% and 100%. Conclusions. Our midterm results demonstrate clinical safety and good performance of the Mosaic bioprosthesis. Continued follow-up will determine if this new design will provide increased durability. (Ann Thorac Surg 2007;83:1310 8) 2007 by The Society of Thoracic Surgeons The main advantage of bioprosthetic cardiac valves compared with mechanical prostheses is the lower incidence of antithromboembolic-related hemorrhages. However, bioprostheses have limited durability due to progressive tissue degeneration and calcification resulting in structural valve deterioration (SVD) and suboptimal hemodynamic performance. The Medtronic Mosaic (Medtronic, Minneapolis, MN) bioprosthesis is a supraannular third-generation stented porcine bioprosthesis that was introduced It is built on the historical durability of the Hancock II (Medtronic) valve [1], and technical innovations were incorporated into the design in an attempt to improve hemodynamic performance and durability [2]. Tissue fixation with the Medtronic Physiologic Fixation process is performed with glutaraldehyde to minimize the consequences of antigenicity after porcine valve implantation [3]. The valve design also includes predilatation of the porcine aortic root and using zero net pressure across the leaflets [4]. This treatment generally preserves natural leaflet morphology. The tissue is mounted on a low-profile flexible polymer stent to minimize hemodynamic disturbance and to make it suitable for patients with small aortic root diameters. The bioprosthesis is treated with the long-chain fatty acid -amino oleic acid (AOA), which binds to the aldehyde fractions of the glutaraldehyde-preserved porcine tissue by forming Schiff-base covalent linkages with the aldehydes that remain after fixation with glutaraldehyde. The AOA process has been shown in several animal studies to reduce porcine valve mineralization of both leaflets and aortic wall and improve valve gradients [5 7]. Our hospital was a contributing center to the United States Food and Drug Administration (FDA) multi- Accepted for publication July 18, Address correspondence to Dr Riess, Albertinen Heart Center, Department of Cardiac Surgery, Suentelstrasse 11a, Hamburg, Germany; friedrich-christian.riess@albertinen.de. Mr Kleijnen and Dr Wallrath disclose that they have a financial relationship with Medtronic by The Society of Thoracic Surgeons /07/$32.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg RIESS ET AL 2007;83: YEAR MOSAIC RESULTS 1311 center, prospective nonrandomized trial that was completed actuarially in late This is a report of our data obtained from the ongoing post-fda approval long-term clinical evaluation study of the Medtronic Mosaic valve. Efficacy, safety, and clinical performance, including hemodynamic data from 302 patients are provided, collected by prospective serial standardized echocardiographic follow-up of up to 10 years. Patients and Methods Patient Population and Study Design Patients diagnosed with valvular heart disease and requiring isolated replacement of either the aortic or mitral valves were eligible to enter the study. Concomitant procedures in addition to valve replacement were permitted. Patients requiring a concomitant valve replacement or who had a preexisting prosthetic valve in another position were excluded. Between February 1994 and October 1999, 302 patients were enrolled into this Table 1. Characteristics of Patients Undergoing Aortic Valve Replacement or Mitral Valve Replacement With the Mosaic Bioprosthesis Variable AVR n 255 (%) MVR n 47 (%) Gender Male 150 (58.8) 14 (29.8) Female 105 (41.2) 33 (70.2) Concomitant procedures Coronary artery bypass 95 (37.3) 8 (17.0) Ascending aorta replacement/ 31 (12.2) 0 repair Aortic root enlargement 20 (7.8) 0 Myotomy/Myectomy 11 (4.3) 0 Age at Implant (years) (3.6) 1 (2.1) (27.5) 12 (19.4) (23.1) 15 (31.9) (31.0) 12 (25.5) (14.0) 7 (14.9) Cardiac rhythm Sinus rhythm 229 (89.8) 25 (53.2) Atrial fibrillation 13 (5.1) 20 (42.6) Heart block 6 (0.6) 0 Paced rhythm 7 (2.8) 2 (4.2) NYHA classification I 4 (1.6) 0 II 74 (29.0) 17 (36.2) III 150 (58.8) 28 (59.6) IV 27 (10.6) 2 (4.3) Valvular lesion Stenosis 43 (16.9) 2 (4.3) Insufficiency 38 (14.9) 30 (63.8) Mixed 174 (68.2) 15 (31.9) AVR aortic valve replacement; MVR mitral valve replacement; NYHA New York Heart Association. Fig 1. Supraannular valve implantation technique. prospective and nonrandomized study. Aortic valve replacement (AVR) was performed in 255 patients (mean age, years; range, 23 to 82 years) and mitral valve replacement (MVR) in 47 patients (mean age, years; range, 41 to 84 years). Coronary artery bypass surgery was performed as a concomitant procedure in 95 patients (37.3%) during AVR and in 8 patients (17%) during MVR. Further demographic data are summarized in Table 1. The study was approved by the respective institutional ethics committee, and all patients provided their informed consent to participate. Surgical Technique Patients were operated on using standard cardioplegia, cardiac arrest, and crystalloid or modified blood cardioplegia (Fig 1). Prosthetic valves were implanted in a supraannular technique with pledgeted single stitches. In case of MVR, posterior chordae were preserved, if possible. Postoperative Anticoagulation Postoperatively, all patients received unfractionated heparin intravenously and subcutaneously until complete mobilization. A total of 95 patients (37.1%) received phenprocoumon for 3 months postoperatively, with an international normalized ratio (INR) target range of 2.5 to 3 in the AVR group and 38 patients (80.9%) with an INR target range of 3 to 4 in MVR group. Indications for phenprocoumon treatment were chronic atrial fibrillation or severely impaired left ventricular ejection fraction ( 0.30). Follow-Up Clinical and hemodynamic follow-up data were collected at the early evaluation (before or 30 days after implantation), at the late evaluation (at 3 to 6 months after implant), at 1 year (11 to 14 months after implant), and annually thereafter. The examination included a patient

3 1312 RIESS ET AL Ann Thorac Surg 10-YEAR MOSAIC RESULTS 2007;83: Table 2. Echocardiographic Results After Aortic Valve Replacement Valve Size (mm) a Variable/Follow-up 19 (n) 21 (n) 23 (n) 25 (n) 27 (n) 29 (n) Mean SVG (mm Hg) 1 year (7) (59) (86) (72) (12) (2) 5 years (2) (39) (56) (52) (10) (2) 9 years (6) (16) (17) (3) 8.1 (1) 10 years 13.9 (1) (5) (3) (2) EOA (cm 2 ) 1 year (7) (59) (86) (72) (12) (2) 5 years (2) (39) (54) (52) (10) (2) 9 years (6) (17) (17) (3) 3.2 (1) 10 years 1.5 (1) (5) (3) (2) a Data are presented as means standard deviation. EOA effective orifice area; SVG systolic valve gradient. interview, the registration of valve-related adverse events, electrocardiogram, and a laboratory check for hemolysis. Assessment of hemodynamic performance and valve structure was performed using transthoracic echocardiography. The mean transvalvular gradient for the aortic bioprosthesis was calculated using the long form of the Bernoulli equation, and effective orifice area was calculated using the continuity equation. The valve-related complications, composites of complications, and deaths were classified and reported according to the guidelines of the Society of Thoracic Surgeons, of the American Association of Thoracic Surgery, and of the European Association of Cardio-Thoracic Surgery [8]. A total of 57 patients from the AVR group and 14 patients from the MVR group were lost to follow-up. This provided for a cumulative follow-up of 1540 patient- years for AVR (mean, 6.1; maximum, 10 years) and 250 patient-years for MVR (mean, 5.4; maximum, 10 years). Statistical Analysis Statistical analysis was performed using the SAS statistical software (SAS Institute, Cary, NC). Descriptive statistics were used to characterize the patient population data, operative, and follow-up clinical data. For continuous variables, the number of patients, mean SD, minimum, and maximum were provided. For categoric variables, the number and percentage of patients were provided. The early events rate was calculated as the number of patients having the event divided by the total number of patients, expressed as a percentage. Linearized rates (percentage per patient-year) were used to summarize late events and were calculated by dividing the number of late events by the sum of the late patientyear of experience, expressed as a percentage. Survival analyses using the actuarial Kaplan-Meier method were used to estimate survival and the freedom from valverelated adverse events. Peto s formula [9] was used to calculate standard errors of these estimates. Events that Table 3. Echocardiographic Results After Mitral Valve Replacement Valve Size (mm) a Variable/Follow-up 25 (n) 27 (n) 29 (n) 31 (n) Mean SVG (mm Hg) 1 year (7) (12) (16) (6) 5 years (2) (7) (11) (5) 9 years 6.2 (1) 4.8 (1) 10 years 3.9 (1) 3.0 (1) 4.0 (1) EOA (cm 2 ) 1 year (7) (12) (16) (6) 5 years (2) (6) (11) (5) 9 years 1.7 (1) 2.2 (1) 10 years 3.3 (1) 2.5 (1) 3.4 (1) a Data are presented as means standard deviation. EOA effective orifice area; SVG systolic valve gradient.

4 Ann Thorac Surg RIESS ET AL 2007;83: YEAR MOSAIC RESULTS 1313 Table 4. NYHA Classification of Patients After Mosaic Valve Replacement in the Aortic or Mitral Position AVR (%) MVR (%) NYHA Class 1 Year 5 Years 10 Years 1 Year 5 Years 10 Years I 159 (66.8) 73 (45.3) 1 (9.1) 10 (24.4) 2 (8.0) 0 (0.0) II 76 (31.9) 79 (49.1) 10 (90.9) 31 (75.6) 23 (92.0) 3 (100.0) III 2 (0.8) 3 (1.9) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) IV 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) UTA 1 (0.4) 6 (3.7) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) AVR aortic valve replacement; MVR mitral valve replacement; NYHA New York Heart Association; UTA unable to assess. occurred during the early and late postoperative periods were included in the analysis. Results Echocardiography Echocardiographically obtained mean systolic valve gradients (SVG) and effective orifice area 5 and 10 years after Mosaic bioprosthesis implantation in aortic position are presented in Table 2. Corresponding data for the mitral valve group are summarized in Table 3. Postoperative New York Heart Association (NYHA) class recorded after AVR or MVR is presented in Table 4, and Table 5 presents the number of patients with transvalvular regurgitation and degree of regurgitation. Cardiac Rhythm At 5 years after prosthetic AVR, 138 patients (85.7%) stayed in sinus rhythm, 15 (9.3%) in atrial fibrillation/ flutter, and 8 (5.0%) were paced. At the same time point, 97 patients (61.0%) received aspirin, 10 (5.7%) were taking phenprocoumon, and 52 (33.1%) had no anticoagulation therapy. In the MVR group, 14 (56.0%) stayed in sinus rhythm, 9 (36.0%) in atrial fibrillation/flutter, and 2 (8.0%) were paced. At the same time, 9 patients (36.0%) received phenprocoumon, 8 (32%) were taking aspirin, and 6 (24.0%) had no anticoagulation therapy. Ten years after prosthetic AVR, 9 patients (81.8%) stayed in sinus rhythm, 1 (9.1%) in atrial fibrillation/ flutter, and 1 (9.1%) was paced. At the same time, 5 patients (45.5%) received aspirin, 2 (18.2%) received phenprocoumon, 1 (9.1%) clopidogrel, and 3 (27.3%) had no anticoagulation therapy. Ten years after prosthetic MVR, 1 patient (33.3%) stayed in sinus rhythm and 2 (66.6%) in atrial fibrillation/flutter. At the same time, 2 patients (66.6%) received phenprocoumon and 1 patient (33.3%) received aspirin. Adverse Events Late valve-related adverse events and actuarial freedom from valve-related adverse events at 5, 9, and 10 years after AVR are summarized in Table 6. Corresponding data of the MVR group are presented in Table 7. Structural Valve Deterioration In the AVR group, four cases of SVD were observed during the 10-year follow-up, with combined aortic valve regurgitation and stenosis in 3 patients and isolated stenosis or regurgitation in 1 patient each. In the four SVD cases, the explanted valves were studied for calcium content microscopically and confirmed by radiologic and histologic investigations of all valve prostheses. No SVD at all was observed in the MVR group (Fig 2). Thrombosed Valves Four cases of aortic valve prosthesis thrombosis were observed. One 73-year-old woman showed valve thrombosis at postoperative day 87 with a mean SVG of 70 mm Hg. Phenprocoumon had been administered during the early postoperative course but was discontinued because of severe nasal angiodysplasia with recurrent hemorrhage. The valve prosthesis was replaced with a Hancock II prosthesis, which also developed thrombosis (110 mm Hg mean SVG) during the early postoperative course. Thrombophilic investigations revealed a congen- Table 5. Patients With Transvalvular Regurgitation After Mosaic Valve Replacement in the Aortic (AVR) or Mitral (MVR) Position AVR MVR Regurgitation 1 Year (n 238) 5 Years (n 161) 10 Years (n 11) 1 Year (n 41) 5 Years (n 25) 10 Years (n 3) I II III 2 1 IV AVR aortic valve replacement; MVR mitral valve position.

5 1314 RIESS ET AL Ann Thorac Surg 10-YEAR MOSAIC RESULTS 2007;83: Table 6. Frequency of Adverse Events and Actuarial Freedom From Valve-Related Adverse Events, Five, Nine and Ten Years After Aortic Valve Replacement (AVR) Late Events Actuarial Freedom From Event a Adverse Event N %/pt-yr 5 Years 9 Years 10 Years Thromboembolism Permanent neurologic event Transient neurologic event Acute myocardial infarction Valve thrombosis Peripheral embolic event Infarction of spleen Structural valve deterioration Endocarditis Paravalvular leak Mismatch Hemolysis Major hemorrhage Reoperation Explant a Data are presented as mean% standard deviation. N number of patients; %/pt-yr percent per patient year. ital antithrombin (AT) III deficiency (AT III activity, 20%). The thrombosed valve prosthesis was replaced with a mechanical valve. Analysis of the explanted valves showed bland thrombotic occlusion without infection or calcification. The second case was a 53-year-old woman receiving aspirin treatment with aortic valve prosthesis thrombosis at postoperative day 891. The mean SVG was 42 mm Hg, and the effective orifice area was 0.68 cm 2. The valve was replaced with a mechanical valve. Analysis of the explanted valve showed extensive thrombosis and no signs of cuspal degeneration, calcification, or infection. In the third case, aortic valve prosthesis thrombosis developed 1827 days after surgery in a 67-year-old man who was receiving aspirin treatment. In the early postoperative course, acquired AT III deficiency and acute heparin-induced thrombocytopenia type II developed in this patient and he had to be treated with AT III concen- Table 7. Frequency of Adverse Events and Actuarial Freedom From Valve-Related Adverse Events Five, Nine and Ten Years After Mitral Valve Replacement Late Events Actuarial Freedom From Event Adverse Event N %/pt-yr 5 Years 9 Years 10 Years Thromboembolism Permanent neurological event Transient neurological event Acute myocardial infarction Valve thrombosis Peripheral embolic event Structural valve deterioration Paravalvular leak Mismatch Endocarditis Paravalvular leak Hemolysis Major hemorrhage Reoperation Explant a Data are presented as mean% standard deviation. N Number of patients; %/pt-yr percent per patient year.

6 Ann Thorac Surg RIESS ET AL 2007;83: YEAR MOSAIC RESULTS 1315 hemorrhagic complications were observed after aortic valve replacement, including cerebral hematoma and gastrointestinal bleeding in 2 patients each, and hematoma of the thigh after an accident in 1. In the mitral group, a 75-year-old woman experienced diffuse oral bleeding that occurred under phenprocoumon treatment 8 days after surgery. Two cases of late hemorrhage complication occurred after prosthetic mitral valve implantation in patients receiving phenprocoumon treatment. The first was a 73-year-old man who had a massive intestinal hemorrhage from the colon (colitis and diverticulitis) 73 days after surgery that required transfusion. The second was a 67-year-old man who required surgical intervention at day 127 for a subdural hematoma. Fig 2. Freedom from structural valve deterioration (SVD) at 10 years after Mosaic valve replacement in the aortic (AVR) or mitral (MVR) position. trates and recombinant hirudin, respectively. The valve was replaced with a Mosaic valve. The fourth case was an 81-year-old man who had aortic stenosis with a mean SVG of 32 mm Hg. This led to reoperation on day 3304 and replacement with a Hancock II valve. Analysis of the explanted valve revealed brown thrombotic appearing material on two cusps. Thromboembolism Two early cases of thromboembolism occurred in the AVR group but in no one in the MVR group. In one 76-year-old woman with paroxysmal atrial fibrillation, a hemiparesis occurred 7 days after aortic valve implantation while she was receiving heparin treatment. The second patient, a 72-year-old woman with chronic atrial fibrillation, experienced a transient neurologic event with nausea, vertigo, and amnesia 13 days after aortic valve replacement. A total of 13 late thromboembolism events occurred after aortic valve replacement during the 10-year followup. Four of these patients sustained permanent neurologic dysfunction (stroke), and 1 patient had chronic atrial fibrillation and had no anticoagulation. In the MVR group, two late thromboembolic events occurred. One patient with sinus rhythm and no anticoagulation had a left renal infarction caused by renal artery thrombus revealed by computed tomography scan. The patient received heparin and was treated with phenprocoumon. Several weeks later, the renal artery was shown to be patent. At 2650 days after mitral valve implantation, a 64-year-old woman with atrial fibrillation and aspirin treatment sustained a stroke with accompanying dysphasia that resolved completely after several days. Bleeding At 24 days after aortic valve implantation, a 73-year-old woman who was receiving phenprocoumon treatment experienced severe epistaxis, and balloon tamponade was required to stop bleeding. In addition, five late Paravalvular Leak Two paravalvular leaks were observed on days 3 and 6 after AVR. The degree of aortic regurgitation was II to III in both patients, one of these had valve replacement with a Mosaic prostheses at postoperative day 102. Six late paravalvular leaks (all degree II) were observed with NYHA class I to II. No surgical intervention was performed in these patients. No paravalvular leaks were observed in the MVR group. Prosthesis Mismatch In 3 patients of the AVR group, late patient prosthesis mismatch between size of the bioprothesis and the body surface area was found, resulting in aortic stenosis. Prosthesis replacement was necessary in 2 patients owing to increased mean systolic valve gradient. Both valves were replaced with Mosaic prosthesis of larger size and additional patch enlargement of the ascending aorta. Endocarditis No early endocarditis was observed in the AVR and MVR groups; however, seven cases of late endocarditis were observed after aortic valve replacement (Table 6). Streptococcus was isolated in 4 patients and enterococcus in 1 patient. In 2 patients, no bacterium could be cultured. Six Fig 3. Survival at 10 years after Mosaic valve replacement in the aortic (AVR) or mitral (MVR) position.

7 1316 RIESS ET AL Ann Thorac Surg 10-YEAR MOSAIC RESULTS 2007;83: Fig 4. Freedom from valve-related or unexplained death after Mosaic valve replacement in the aortic (AVR) or mitral (MVR) position. patients were treated with valve replacement, and 1 patient was treated successfully with antibiotics. In the MVR group, two cases of late endocarditis occurred, one each induced by enterococcus or streptococcus. Both patients were treated with mitral prosthesis replacement. Explants One valve was explanted in the AVR group due to early prosthetic valve thrombosis, and no valve was explanted in the MVR group during the early ( 30 days) follow-up. Twenty aortic valve prostheses (1.3% per patient-year) were explanted by the 10-year follow-up. Reasons for explant were endocarditis in 7 patients, SVD in 4, valve thrombosis in 4, incidental replacement due to aneurysm of the ascending aorta in 2, mismatch in 2, and paravalvular leak in 1 patient. Two patients in the MVR group (0.8% per patient-year) were explanted because of prosthetic endocarditis. Mortality The early ( 30 days) mortality rate in the AVR group was 0.8% (2 patients). In one patient, pulmonary hypertension due to hypertrophic obstructive cardiomyopathy with left ventricular outlet obstruction occurred on postoperative day 3; and in the other, acute pericardial tamponade due to aortic dissection and perforation occurred on postoperative day 13. The linearized rate for late mortality was 3.5% per patient-year (Fig 3). Of 54 late deaths, 3 were valve-related, 11 were cardiac, 26 were noncardiac, and 14 were unexplained, for respective patient-year rates of 0.2%, 0.7%, 1.8%, and 1.0% (Fig 4). The three valve-related deaths occurred after AVR. One patient was a 66-year-old man receiving phenprocoumon treatment due to chronic atrial fibrillation (INR 2). He suffered from a cerebral hemorrhage requiring a relieving operation and developed septicemia. He died 2119 days after the primary valve implantation. The second patient was a 49-year-old man, who developed prosthetic valve endocarditis. This patient was reoperated and a new tissue valve was implanted. However, this patient died one day after re-operation, and 3098 days after primary valve implantation, from a lowoutput syndrome. The third patient was a 68-year-old man, who died from heart failure 983 days after valve implantation due to a stroke. No autopsy was performed in any of these patients. The 11 cardiac deaths by 10-year follow-up included myocardial infarction in 5 patients, heart failure in 5, and cor pulmonale in 1. No early deaths occurred in the MVR group. The linearized rate for late mortality (7 deaths) was 2.8% per patient-year (Fig 3). Of these seven deaths, one was cardiac (0.4% per patient-year), three were noncardiac (1.2% per patient-year), and three were unexplained (1.2% per patient-year) (Fig 4). Survival at 10 years was 65.5% 4.4% for AVR and 84.1% 6.0% for MVR. Freedom from valve-related or unexplained death at 10 years was 88.0% 3.2% (SE) for the AVR group and % 3.8% for the MVR group. There was no case of valve-related death during the total follow-up and only one case of cardiac death after MVR in a 75-year-old woman with cor pulmonale at postoperative day 1117 due to heart failure. Comment These are data from an ongoing post-fda approval trial investigating the hemodynamic performance and the clinical outcome of patients after implantation of the Medtronic Mosaic porcine prosthesis in aortic or mitral position. The design of an annual prospective serial standardized echocardiographic follow-up allowing assessment of functional and hemodynamic data was generally not performed in earlier generation bioprostheses. The serial echo follow-up here is unique to this study cohort. The intermediate 10-year results demonstrate excellent hemodynamic and clinical performance of this new third-generation porcine valve prosthesis in the aortic and mitral positions. Results are comparable with those of other frequently implanted tissue valves [10 13]. Thus, there is no statistically significant difference between the second-generation Carpentier-Edwards supraanular porcine valve (Irvine, CA), the Carpentier-Edwards Perimount pericardial prosthesis, and the third-generation Medtronic Mosaic porcine valve with respect to mean transvalvular gradients and effective orifice area [14]. The present echocardiographic data demonstrate stable transvalvular gradients and a transvalvular effective orifice area during the 10 years of follow-up. Data are comparable with data obtained in earlier Mosaic series that reported up to 7 years of echocardiography findings [2, 14, 15]. Relatively low gradients were found in small valve sizes. It is assumed that these low transvalvular gradients are a direct benefit of this new valve design and tissue processing that preserves the normal architecture of the collagen tissue [4]. In contrast with the previous Intact Medtronic valve prosthesis, where fixation in a nondilated root was associated with overcrowding of the leaflets, the Mosaic valve includes dilatation of the aortic root at the time of

8 Ann Thorac Surg RIESS ET AL 2007;83: YEAR MOSAIC RESULTS 1317 preservation, resulting in normal plane closure of leaflets without restriction. Nevertheless, freedom from SVD was low in the Intact valve at the 10-year follow-up [16, 17]. It was expected that improvement in tissue fixation and additional antimineralization treatment with -amino oleic acid [17 19] would provide longer durability than current bioprostheses, especially in a younger population [2]. In our study population, of 255 aortic implants, four patients (age at implant 31, 55, 55, 63) needed valve replacement for SVD by the 10-year follow-up, resulting in a freedom from SVD of 87.1% 6.7% (Fig 2). Because hazards are not constant, the presentation of linearized rates is not really quantitative. None of these documented cases of SVD in the AVR group occurred in a patient older than 65 years at implant. Thus, 10-year freedom from SVD in patients older than 65 years was 100%. Jamieson and colleagues [20] and Malligan and associates [21] found that although SVD occurred at all ages, the freedom was greater with advancing of age. These results suggest that the evolution in valve design allows for performance superior to that reported for earlier generation bioprostheses. Sarris reported Hancock standard 10-year SVD freedom rates of 59% 9% and 72% 2%, respectively, for aortic and mitral replacement [22]. Jamieson and colleagues [23] reported 10-year freedom rates of 79% and 72% with the Carpentier-Edwards aortic and mitral replacements, respectively. Jones and colleagues [24] documented 10-year freedom from SVD with standard prosthesis of 79% for AVR and 63% for MVR. Pelletier [25] reported 87% freedom from SVD at 10 years for the Carpentier-Edwards Perimount pericardial valve in aortic position. Most remarkably in our study, not a single case of SVD was observed in 47 patients who underwent MVR during the 10-year follow-up and no case of moderate or severe transvalvular regurgitation was observed. This observation contrasts other authors who reported tissue valves to be more durable in the aortic position than in the mitral position [1, 16, 24, 26, 27, 28]. It was suggested [23] that the difference in durability may be due to elevated closing pressures and, thus, increased hemodynamic stresses in the mitral position. This suggests that the Mosaic valve seems to be especially able to withstand the high stress in the mitral position. Valve thrombosis occurred in 4 patients of the AVR group and in none of the MVR group. In one patient with a prosthesis thrombosis, a congenital antithrombin deficiency with residual activity of only 20% was detected, which can be considered to be the reason for this adverse event. This theory is supported by the recurrence of thrombosis after implantation of a Hancock II prosthesis. In another patient, valve thrombosis of the Mosaic prosthesis developed 1827 days after implant. In this patient, heparin-induced thrombocytopenia type II as well as AT III deficiency developed during the early postoperative course and was treated with recombinant hirudin (Lepirudin, Pharmion, Windsor Berkshire, UK) as an alternative anticoagulation instead of heparin. The overlapping treatment with phenprocoumon had to be stopped early after initiation because of a gastrointestinal hemorrhage. Intraoperative findings and histologic investigation supported the theory of a heparin-induced thrombocytopenia, because a thin layer of white clot formation was found in all three cusps of the Mosaic valve resulting in prosthetic stenosis with a transvalvular peak gradient of 55 mm Hg. Endocarditis developed in 7 patients of the AVR group and in 2 patients of the MVR group during the postoperative course, resulting in freedom from endocarditis rates of 95.4% 4.6% for the AVR group and 95.0% 3.5% for the MVR group. Thus, the rates of infective endocarditis after AVR and MVR with Mosaic valve prosthesis are similar to those reported for other stented porcine and pericardial valves such as Hancock II [1], Carpentier-Edwards Perimount [26, 29, 30], Biocor (St. Jude Medical, St Paul, MN) [31, 32], Medtronic Intact [16] and the Carpentier-Edwards porcine prosthesis [33]. The risk of endocarditis was reported by David and colleagues [1] to be highest during the first year after the operation. In contrast to this, in our cohort, 7 cases of endocarditis in the AVR occurred 261, 615, 782, 833, 1886, 1909, and 3015 days after implant, and 2 patients with endocarditis in the MVR group were observed 133 and 1265 days after implant. A major benefit of a bioprosthesis compared with a mechanical valve is the low incidence of thromboembolic and major anti-thromboembolic-related hemorrhage. In the present trial, late thrombotic rates of 0.8% per patient-year were observed in the AVR and MVR group, respectively, which is favorable with other studies. In this context, it seems to be important that all patients without the need for phenprocoumon (eg, chronic atrial fibrillation) were treated with aspirin. A comparable low rate of major hemorrhages was observed for both groups, with only 0.3% per patientyear after AVR and 0.8% per patient-year after MVR. The rates for freedom from hemorrhage at 10 years were 96.8% 1.3% (AVR) and 93.5% 3.6% (MVR), respectively. The number of major hemorrhage was lower than in other series [34]. The reason might be the relatively low percentage of patients of our study who were receiving continuous anticoagulation therapy at 6 months after surgery compared with other studies (52.5% per patientyear) [34]. In conclusion, the long-term performance of the Mosaic valve is encouraging. This third-generation porcine bioprosthesis continues to provide excellent hemodynamic data, a small number of valve-related adverse events, and a low incidence of structural valve deterioration. Continued clinical follow-up and monitoring of this patient population should demonstrate if indeed this valve will provide patients with increased durability and low morbidity compared with bioprosthetic valves of an earlier design and generation. We thank Petra Schlizio for her skillful secretarial assistance.

9 1318 RIESS ET AL Ann Thorac Surg 10-YEAR MOSAIC RESULTS 2007;83: References 1. David TE, Ivanov J, Armstrong S, Feindel CM, Cohen G. Late results of heart valve replacement with the Hancock II bioprosthesis. J Thorac Cardiovasc Surg 2001;121: Fradet G, Bleese N, Busse E, et al. The Mosaic valve clinical performance at seven years: results from a multicenter prospective clinical trial. J Heart Valve Dis 2004;13: Carpentier A, Lemaigre G, Robert L, Carpentier S, Dubost C. Biological factors affecting long-term results of valvular heterografts. J Thorac Cardiovas Surg 1969;58: Vesley I. Analysis of the Medtronic Intact bioprosthesis valve. Effects of zero-pressure fixation. J Thorac Cardiovasc Surg 1991;101: Duarte IG, MacDonald MJ, Cooper WA, et al. In vivo hemodynamic, histologic, and antimineralization characteristics of the Mosaic bioprosthesis. Ann Thorac Surg 2001;71: Girardot MN, Girardot JM, Schoen FJ. Alpha amino oleic acid, a new compound, prevents calcification of bioprosthetic heart valves. Trans Soc Biomater 1991;14: Walther T, Falk V, Diegeler A, et al. Effectiveness of different anticalcification treatments for stentless aortic bioprosthesis. Thorac Cardiovasc Surg 1999;47: Edmunds LH, Clark RE, Cohn LH, Grunkemeier GL, Miller DC, Wiesel RD. Guidelines for reporting morbidity and mortality after cardiac valvular operations. Ann Thorac Surg 1996;62: Peto R, Pike MC, Armitage P, et al. Design and analysis of randomized clinical trials requiring prolonged observation of each patient. Br J Cancer 1977;35: Salomon NW, Okies JE, Krause AH, Page US, Bigelow JC, Colburn LQ. Serial follow-up of an experimental bovine pericardial aortic bioprosthesis. Circulation 1991;84: David TE, Armstrong S, Sun Z. Clinical and hemodynamic assessment of the Hancock II bioprosthesis. Ann Thorac Surg 1992;54: McDonald ML, Daly RC, Schaff HV, et al. Hemodynamic performance of small aortic valve bioprostheses: is there a difference? Ann Thorac Surg 1997;63: Jamieson WR, Janusz MT, Burr LH, Ling H, Miyagishima RT, Germann E. Carpentier-Edwards supraannular porcine bioprosthesis: second-generation prosthesis in aortic valve replacement. Ann Thorac Surg 2001(5 Suppl);71:S Jamieson WRE, Janusz MT, MacNab J, Henderson C. Hemodynamic comparison of second and third generation stented bioprostheses in aortic valve replacement. Ann Thorac Surg 2001(5 Suppl);71:S Wong SP, Legget ME, Greaves SC, Barratt-Boyes BG, Milsom FP, Raudkivi PJ. Early experience with the mosaic bioprosthesis: a new generation porcine valve. Ann Thorac Surg 2000;69: Barratt-Boyes BG, Jaffe WM, Whitlock RM. The Medtronic intact porcine valve: Ten-year clinical review. J Thorac Cardiovasc Surg 1998;116: Jones M, Eidbo EE, Hilbert SL, Ferrans VJ, Clark RE. Anticalcification treatments of bioprosthetic heart valves: in vivo studies in sheep. J Card Surg 1989;4: Gott JP, Pan-Chih, Dorsey LMA, et al. Calcification of porcine valves: a successful new method of antimineralization. Ann Thorac Surg 1992;53: Melina G, Rubens MB, Birks EJ, Bizzarri F, Khaghani A, Yacoub MH. A quantitative study of calcium deposition in the aortic wall following Medtronic freestyle compared with homograft aortic root replacement. A prospective randomized trial. J Heart Valve Dis 2000;9: Jamieson WRE, Janusz MT, Miyagishima RT, et al. Carpentier-Edwards standard porcine bioprostheses primary tissue failure (structural valve deterioration) by age groups. Ann Thorac Surg 1988;46: Magilligan DJ, Lewis JW, Stein P, Alam M. The porcine bioprosthetic heart valve: experiences at 15 years. Ann Thorac Surg 1989;48: Sarris GE, Robbins RC, Miller DC, et al. Randomized, prospective assessment of bioprosthetic valve durability. Hancock versus Carpentier-Edwards valves. Circulation 1993;88:II Jamieson WRE, Allen P, Miyagishima RT, et al. The Carpentier-Edwards standard porcine bioprosthesis: a first-generation tissue valve with excellent long-term clinical performance. J Thorac Cardiovasc Surg 1990;99: Jones EL, Weintraub WS, Craver JM, et al. Ten-year experience with the porcine bioprosthetic valve: interrelationship of valve survival and patient survival in 1,050 valve replacements. Ann Thorac Surg 1990;49: Pelletier LC, Carrier M, Leclerc Y, Dyrda I. The Carpentier- Edwards pericardial bioprosthesis: clinical experience with 600 patients. Ann Thorac Surg 1995;60(2 Suppl):S Poirier NC, Pelletier LC, Pellerin M, Carrier M. 15 year experience with the Carpentier-Edwards Pericardial bioprosthesis. Ann Thorac Surg 1998;66: Jamieson WR, Munro AI, Miyagishima RT, Allen P, Burr LH, Tyers GF. Carpentier-Edwards standard porcine bioprosthesis: clinical performance to seventeen years. Ann Thorac Surg 1995;60: Akins CW, Carroll DL, Buckley MJ, Daggett WM, Hilgenberg AD, Austen WG. Late results with Carpentier-Edwards porcine bioprosthesis. Circulation 1990;82(Suppl IV): Banbury MK, Cosgrove DM 3rd, Lytle BW, Smedira NG, Sabik JF, Saunders CR. Long-term results of the Carpentier- Edwards pericardial aortic valve: a 12-year follow-up. Ann Thorac Surg 1998;66:S73 S Frater RW, Furlong P, Cosgrove DM, et al. Long-term durability and patient functional status of the Carpentier- Edwards Perimount pericardial bioprosthesis in the aortic position. J Heart Valve Dis 1998;7: 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: Dellgren G, Eriksson MJ, Brodin LA, Radegran K. Eleven years experience with the Biocor stentless aortic bioprosthesis: clinical and hemodynamic follow-up with long-term relative survival rate. Eur J Cardiothorac Surg 2002;22: Bernal JM, Rabasa JM, Lopez R, Nistal JF, Muniz R, Revuelta JM. Durability of the Carpentier-Edwards porcine bioprosthesis: Role of age and valve position. Ann Thorac Surg 1995;60:S Gansera B, Botzenhardt F, Grünzinger R, Spiliopoulos K, Angelis I, Kemkes BM. The Mosaic Bioprosthesis in aortic position: seven years results. J Heart Valve Dis 2003;12: