1198 MINAKATA K et al. Circ J 2017; 81: ORIGINAL ARTICLE doi: /circj.CJ

Transcription

1 1198 MINAKATA K et al. Circ J 2017; 81: ORIGINAL ARTICLE doi: /circj.CJ Valvular Heart Disease Comparison of the Long-Term Outcomes of Mechanical and Bioprosthetic Aortic Valves A Propensity Score Analysis Kenji Minakata, MD, PhD; Shiro Tanaka, PhD; Nobushige Tamura, MD, PhD; Shigeki Yanagi, MD, PhD; Yohei Ohkawa, MD, PhD; Shuichi Okonogi, MD; Tatsuo Kaneko, MD, PhD; Akihiko Usui, MD, PhD; Tomonobu Abe, MD, PhD; Mitsuomi Shimamoto, MD, PhD; Yoshiharu Takahara, MD, PhD; Kazuo Yamanaka, MD, PhD; Hitoshi Yaku, MD, PhD; Ryuzo Sakata, MD, PhD Background: The aim of this study was to assess the long-term outcomes of aortic valve replacement (AVR) with either mechanical or bioprosthetic valves according to age at operation. Methods and Results: A total of 1,002 patients (527 mechanical valves and 475 bioprosthetic valves) undergoing first-time AVR were categorized according to age at operation: group Y, age <60 years; group M, age years; and group O, age 70 years). Outcomes were compared on propensity score analysis (adjusted for 28 variables). Hazard ratio (HR) was calculated using the Cox regression model with adjustment for propensity score with bioprosthetic valve as a reference (HR=1). There were no significant differences in overall mortality between mechanical and bioprosthetic valves for all age groups. Valve-related mortality was significantly higher for mechanical valves in group O (HR, 2.53; P=0.02). Reoperation rate was significantly lower for mechanical valves in group Y (HR, 0.16; P<0.01) and group M (no events for mechanical valves). Although the rate of thromboembolic events was higher in mechanical valves in group Y (no events for tissue valves) and group M (HR, 9.05; P=0.03), there were no significant differences in bleeding events between all age groups. Conclusions: The type of prosthetic valve used in AVR does not significantly influence overall mortality. Key Words: Bioprosthetic valve; Complication; Mechanical valve; Reoperation; Survival The American College of Cardiology/American Heart Association (ACC/AHA) recently published their revised guidelines on the selection of prosthetic valves. 1 According to their class I recommendation, patient values and preferences must be taken into account, based on full disclosure and understanding of the indications for anticoagulant therapy and the potential need for and risk of reoperation. Besides patient preference, age has been one of the most important factors influencing prosthetic valve choice: mechanical valves are recommended for patients <60 years of age, and bioprosthetic valves for those >70 years of age. For those between 60 and 70 years of age, either type of valve prosthesis can be chosen. The aim of this study was to assess the mortality and valverelated complications and compare the long-term outcomes of mechanical vs. bioprosthetic valves in Japanese patients undergoing aortic valve replacement (AVR) stratified into 3 age groups (<60 years; years, and 70 years). Methods Patients This study was a pooled analysis of 2 large retrospective Received February 13, 2017; accepted March 3, 2017; released online April 12, 2017 Time for primary review: 18 days Department of Cardiovascular Surgery, Kyoto University Graduate School of Medicine, Kyoto (K.M., R.S.); Department of Pharmacoepidemiology, Kyoto University Graduate School of Medicine and Public Health, Kyoto (S.T.); Division of Cardiovascular Surgery, Kumamoto Central Hospital, Kumamoto (N.T., S.Y.); Division of Cardiovascular Surgery, Cardiovascular Center Hokkaido Ohno Hospital, Sapporo (Y.O.); Division of Cardiovascular Surgery, Gunma Prefectural Cardiovascular Center, Maebashi (S.O., T.K.); Department of Cardiac Surgery, Nagoya University Graduate School of Medicine, Nagoya (A.U., T.A.); Division of Cardiovascular Surgery, Shizuoka City Shizuoka Hospital, Shizuoka (M.S.); Department of Cardiovascular Surgery, Funabashi Municipal Medical Center, Funabashi (Y.T.); Department of Cardiovascular Surgery, Tenri Hospital, Tenri (K.Y.); and Department of Cardiovascular Surgery, Kyoto Prefectural University of Medicine, Kyoto (H.Y.), Japan This abstract was presented at the American Heart Association Scientific Session 2015, Orlando, FL, USA. Mailing address: Kenji Minakata, MD, PhD, Division of Cardiovascular Surgery, Temple University School of Medicine, 3401 North Broad Street, Suite C-301, Philadelphia, PA 19140, USA. k_minakata@hotmail.com ISSN All rights are reserved to the Japanese Circulation Society. For permissions, please cj@j-circ.or.jp

2 Prosthetic Valve Choice in AVR 1199 Figure 1. Flowchart of patient selection and follow-up. AVR, aortic valve replacement; CEP, Carpentier-Edwards pericardial; MVR, mitral valve replacement; SJM, St. Jude Medical; yrs, years. cohort studies with similar study designs that assessed the very long-term outcomes of either bioprosthetic valves (Carpentier-Edwards pericardial [CEP] Perimount valve; Edwards Life Sciences, Irvine, CA, USA) or mechanical valves (St. Jude Medical [SJM] bileaflet mechanical valve; St. Jude Medical, Minneapolis, MN, USA) in Japanese patients undergoing AVR. 2,3 In the bioprosthetic valve study cohort, there were 591 patients who underwent AVR with CEP from 1985 to 2000 in 9 Japanese cardiac centers. Of these, 475 patients were enrolled after excluding those who underwent redo AVR, aortic root replacement, and concomitant mitral valve replacement. In the mechanical valve cohort, there were 816 patients who underwent AVR with SJM mechanical valves from 1991 to 2001 in 5 Japanese cardiac centers. Of these, 527 patients were enrolled after excluding patients using the same criteria. A total of 1,002 patients from these 2 sets of the patients formed the present cohort (Figure 1). Of note, all 5 cardiac centers that participated in the SJM study were included in the 9 centers that participated in the CEP study. All surviving patients had follow-up surveys with a 10-year followup rate of 83.2% in the CEP study and 83.7% in the SJM study. We integrated the 2 sets of data and analyzed longterm outcomes according to age at operation to verify the current ACC/AHA guidelines in Japanese patients. Definitions of Events Definitions of structural valve deterioration (SVD), and other valve-related events such as bleeding, thromboembolism, reoperation, and valve-related death were based on the Society of Thoracic Surgeons (STS) guidelines for reporting mortality and morbidity after cardiac valve interventions, published in All clinical events were evaluated at the participating hospitals, and then assessed by an independent clinical events evaluation committee if necessary. Of note, valve-related death was defined as any death caused by structural or non-structural valve dysfunc- tion, valve thrombosis, embolism, bleeding event, prosthetic valve endocarditis, death related to reintervention on the operated valve or sudden, unexplained death. Death caused by heart failure in patients with advanced myocardial disease and satisfactorily functioning cardiac valves was not counted. Statistical Analysis Postoperative overall survival, freedom from valve-related death, reoperation due to SVD or non-structural dysfunction, thromboembolic events, and major bleeding events were estimated using the Kaplan-Meier method. Between the 2 cohorts there were significant differences in patient background characteristics, such as age and other risk factors. Therefore, we conducted a propensity score analysis to adjust for 28 variables: age; gender; preoperative cardiac/kidney/respiratory functions; body mass index; body surface area; New York Heart Association (NYHA) functional class; diabetes mellitus; dyslipidemia; hypertension; coronary artery disease (CAD); history of myocardial infarction (MI) or percutaneous coronary artery intervention; stroke; malignancy; atrial fibrillation (AF); peripheral artery disease (PAD); pathology of aortic valve; concomitant procedures; and year of operation. Hazard ratios (HR) were calculated using the Cox regression model with adjustment for propensity score with a bioprosthetic valve as a reference, that is, HR >1 indicates that bioprosthetic valve (CEP) is superior, while HR <1 indicates that mechanical valve (SJM) is superior. All continuous variables are expressed as mean ± SD. All P-values were 2-sided, and P<0.05 was considered statistically significant. All statistical analysis was carried out by an academic statistician (S.T.) using SAS version 9.4 (SAS Institute, Cary, NC, USA).

3 1200 MINAKATA K et al. Table 1. Baseline Subject Characteristics Characteristics CEP group SJM group (n=475) (n=527) P value Age (years) 72.2± ±10.6 <0.01 <60 years 18 (3.8) 215 (40.8) years 92 (19.4) 220 (41.7) 70 years 365 (76.8) 92 (17.5) Female 228 (48.0) 185 (35.1) <0.01 BSA (m 2 ) 1.50± ±0.17 <0.01 LVEF (%) 62.3± ± Comorbidity Hypertension Dyslipidemia Diabetes mellitus Atrial fibrillation COPD Peripheral artery disease <0.01 CKD (egfr <60) <0.01 Hemodialysis <0.01 Malignancy <0.01 CAD <0.01 Old MI <0.01 LVEF 40% Prior cerebral infarction Prior intracranial hemorrhage NYHA functional class <0.01 I/II 16.9/ /57.3 III/IV 31.6/ /2.6 Aortic valve pathology Bicuspid Rheumatic <0.01 Infective endocarditis Aortic valve lesion <0.01 Aortic stenosis Aortic regurgitation Mixed lesion Data given as mean ± SD, n (%) or %. BSA, body surface area; CAD, coronary artery disease; CEP, Carpentier-Edwards pericardial; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; egfr, estimated glomerular filtration rate; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NYHA, New York Heart Association; SJM, St. Jude Medical. Results The baseline characteristics of the entire cohort are listed in Table 1. Mean age at operation was 72.2±8.2 years in the CEP group, and 60.0±10.6 years in the SJM group (P<0.01). In addition, there were significant differences in gender, prevalence of PAD, chronic kidney disease, CAD, AF, history of MI and malignancy between the 2 groups. Furthermore, preoperative NYHA functional class was more advanced in the CEP group than in the SJM group. In terms of aortic valve pathology, bicuspid disease, rheumatic degeneration, and infective endocarditis were more common in the SJM group than in the CEP group. Accordingly, pure aortic regurgitation was more common in the SJM group, while pure aortic stenosis was more common in the CEP group. In terms of the operative variables, including concomitant procedures and ischemic time (Table 2), there were no significant differences between the groups. With regard to late results, the causes of late death are summarized and the detailed indications of aortic valve reoperation given in Table 3. Unadjusted survival estimates for both the CEP and SJM groups are given using Kaplan-Meier curves according to age at operation with propensity score-adjusted HR in Figures 2 4. There were no significant differences in overall survival between SJM and CEP valves for all age groups (Figure 2). In terms of valve-related mortality, HR was not estimated in group Y, because there was no valverelated death in the CEP group. Valve-related mortality was significantly higher for SJM valves in group O compared with CEP valves (HR, 2.36; P=0.03; Figure 3). In terms of freedom from reoperation in the aortic position (Figure 3), the risk of reoperation rate was significantly lower for SJM valves in group Y (HR, 0.16; P<0.01). Because there were no patients requiring reoperation in SJM group M, HR was not estimated. Of note, there were a certain number of patients in this age group who required reoperation for SVD. Although there was no difference in

4 Prosthetic Valve Choice in AVR 1201 Table 2. Operative Variables Characteristics CEP group (n=475) SJM group (n=527) P value Isolated aortic valve replacement 301 (63.4) 354 (67.2) 0.21 Concomitant procedures 174 (36.6) 173 (32.8) CABG 99 (20.8) 86 (16.3) Mitral valve repair 40 (8.4) 47 (8.9) Tricuspid valve repair 14 (2.9) 20 (3.8) Annular enlargement 15 (3.2) 9 (1.7) Aortic aneurysm repair 36 (7.6) 26 (4.9) Mean aortic cross-clamp time (min) 95±40 95± Mean CPB time (min) 133±55 141± Data given as n (%) or mean ± SD. CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass. Other abbreviations as in Table 1. Table 3. Postoperative and Follow-up Data Characteristics CEP group (n=475) SJM group (n=527) Late death 233 (49.1) 198 (37.6) Cause of late death Malignancy 41 (8.6) 32 (6.1) Respiratory 31 (6.5) 27 (5.1) Cardiac 29 (6.1) 46 (8.7) Cerebrovascular 12 (2.5) 31 (5.9) Renal 8 (1.7) 3 (0.6) Infection 7 (1.5) 14 (2.7) Hemorrhage 6 (1.3) 19 (3.6) Multiple organ 3 (0.6) 2 (0.4) Gastrointestinal 2 (0.4) 2 (0.4) Liver 2 (0.4) 2 (0.4) Sudden 6 (1.3) 14 (2.7) Unknown/others 82 (17.3) 31 (5.9) Valve-related 23 (4.8) 50 (9.5) Aortic valve reoperation 19 (4.0) 13 (2.5) Structural valve deterioration 14 (2.9) 0 Non-structural obstruction/pannus 0 5 (1.0) Paravalvular leak (hemolysis) 1 (0.2) 5 (1.0) Prosthetic valve endocarditis 4 (0.8) 3 (0.6) Thrombosis 0 0 Data given as n (%). Abbreviations as in Table 1. group O, the SJM valve was much more favorable for age <70 years in terms of freedom from reoperation. With regard to the thromboembolic events (Figure 4), there were no events in CEP group Y, thus HR was not estimated. In group M, however, the event rate was significantly higher in SJM valves. There was no difference between bioprosthetic and mechanical valves in group O. In terms of major bleeding (Figure 4), because there was no event in CEP group Y, HR was not estimated. In addition, there were no significant differences in major bleeding events in the other age groups. Figure 5 summarizes the aforementioned HR of overall mortality, valve-related survival, reoperation, thromboembolism, and major bleeding according to age group. Discussion To date, this is the largest comparative study to assess the long-term outcomes in Japanese patients undergoing AVR with either mechanical or bioprosthetic valves. 5 In particular, this study is unique because we compared these 2 types of valves using propensity score analysis by stratifying a large cohort according to the age cut-offs 60 and 70 years, as noted in the most recent ACC/AHA treatment guidelines. 1 We, as clinicians in daily practice, must insure that patients and their families are fully informed about the risks and benefits of both types of prosthetic valves, and we must allow them to decide based on this information. In order to do so, we must have reliable data. Unfortunately, such data have been lacking in the literature. Therefore, the data described in this manuscript will likely be useful information for all care providers.

5 1202 MINAKATA K et al. Figure 2. Freedom from all-cause death according to age at operation. Abbreviations as in Figure 1. There have been only a few prospective randomized clinical trials comparing biological and mechanical prostheses. In the Unites States Veterans Administration (USVA) Cooperative Study on Valvular Heart Disease, Hammermeister et al randomized either mechanical valves (Bjork-Shiley monocusp tilting valve) or bioprosthetic valves (Hancock porcine valve) for patients undergoing AVR. 6 In their study, the majority of the patients were in their 50 s (40.9%) or 60 s (39.6%) at the time of operation. They concluded that patients who underwent AVR with mechanical valves had better survival at 15 years than those who received bioprosthetic valves (34% vs. 21%; P=0.02). In addition, primary valve failure requiring reoperation was more common with bioprosthetic valves in those aged <65 years, and not different in those aged 65 years. The thromboembolic event rate was similar between the 2 valves, but the bleeding event rate was higher with mechanical valves. Similar to the USVA study, in the Edinburgh trial, Oxenham et al conducted a prospective randomized study in which they used either mechanical valves (Bjork-Shiley monocusp tilting valve) or bioprosthetic valves (Hancock or Carpentier-Edwards porcine valve). 7 The age distribution in that trial consisted of <50 years (33.6%), years (54.6%) and >65 years (11.8%). They found no differences in the 20-year survival rate between mechanical and bioprosthetic valves. Also, when they combined death and reoperation as endpoints, they confirmed that mechanical valve patients had improved survival years after surgery. In addition, major bleeding was more common in mechanical valves, but there were no significant differences in thromboembolic events and endocarditis. Another prospective randomized trial in patients aged years, which was conducted more recently in Italy, noted no significant differences in survival, freedom from thromboembolic events, and bleeding events between mechanical valves (SJM or CarboMedics bileaflet) and bioprosthetic valves (CEP or CE porcine valve), but patients with bioprosthetic valves faced a significant higher risk of structural failure and reoperation. 8 Although both mechanical and biological prostheses used in the former 2 prospective randomized trials were first-generation prostheses, which are no longer used in current practice, the essential findings of these 2 studies must be taken seriously. The core messages from all 3 of these prospective randomized trials are that: (1) there was no significant difference in overall survival between the 2 valve types, or slightly better with mechanical valves; (2) the risk of reoperation was higher for biological valves, especially in those who received the valves at a younger age; and (3) the risk of bleeding was likely higher in mechanical valves, but there were no differences in thromboembolic events. In terms of long-term overall survival, many retrospective non-randomized studies have been carried out. Chiang et al compared survival and long-term outcome in patients aged years undergoing AVR on propensity-matched analysis derived from a large New York State registry. 9 They reported that there was no significant difference in the 15-year survival or stroke rate, and patients with bioprosthetic valves had a greater likelihood of reoperation but a lower likelihood of major bleeding. Also, Brennan et al reported that long-term mortality rates were similar for those who received bioprosthetic vs. mechanical valves on propensity score analysis using the STS database, 10 which was consistent with the present findings. In general, preoperative basic characteristics are different in those receiving

6 Prosthetic Valve Choice in AVR 1203 Figure 3. Freedom from valve-related death and reoperation according to age at operation. Abbreviations as in Figure 1. younger patients with longer life expectancy), which was not captured by the database and comparative analysis; (2) intrinsic prosthesis valve performance (maintaining normal function in mechanical valves for years), and (3) benefit of long-term anticoagulation (protective effect against thromboembolic events). 14 In a previous study, we demonstrated that in patients aged <65 years who underwent AVR with a CEP bioprosthesis, reoperation due to SVD started to become necessary after 8 years. 15 In patients requiring reoperation, mean duration from initial operation to reoperation was approximately 12 years. Also, >50% of the patients required reoperation due to SVD within 15 years postoperatively. The perioperative risk of reoperation has dramatically improved in current practice. Potter et al reported that the risk of reoperative AVR is similar to that for primary AVR after excluding patients with prosthetic valve endocarditis, and the data support the extended use of bioprosthetic valves in younger patients. 16 Based on the Vancouver database, van Geldorp et al reported that the mortality due to bleeding was much higher than that due to reoperation (22% vs. 7.3%). 17 They reported that the lower risk of bleeding events offered by a bioprosthesis outweighs the increased risk associated with SVD. Therefore, they concluded that event-free life expectancy is better with a biobioprosthetic and mechanical valves. Undoubtedly, age is one of the most important factors. McClure et al compared late outcomes of patients aged <65 years with bioprosthetic and mechanical valves in the aortic position using propensity-matched analysis. 11 They found that despite an increase in the rate of reoperation with bioprosthetic valves and an increase in major bleeding events with mechanical valves, there was no significant difference in mortality at late follow-up. In contrast, a few reports showed that mechanical valves provide better long-term survival, especially in younger patients. Weber et al found that mechanical valves (SJM or ATS bi-leaflet mechanical valves) in the aortic position were associated with better 10-year survival compared with bioprosthetic (CEP) valves in patients aged <60 years. 12 They reported that the survival advantage seemed related to the better hemodynamic performance of the mechanical valves and/or the protective effect of long-term anticoagulation. Furthermore, Brown et al reported improved long-term outcomes with mechanical valves (SJM) compared with bioprosthetic valves (CEP) in patients aged years on propensitymatched analysis from the Mayo Clinic database. 13 In addition, Suri and Schaff argued that the possible explanations of the survival benefit of mechanical valves include: (1) selection bias (mechanical valves tend to be chosen in

7 1204 MINAKATA K et al. Figure 4. Freedom from thromboembolic event and major bleeding event according to the age at operation. Abbreviations as in Figure 1. prosthesis even for patients aged <60 years. According to their calculations, the risk of bleeding with mechanical valves and the risk of reoperation with bioprosthetic valves become equal at age 55 years for AVR. Thus, a bioprosthesis may be a suitable option in much younger patients than previously thought. Due to the improved durability of bioprosthetic valves and the desire of patients to avoid long-term anticoagulation, there is a significant trend towards the use of bioprosthetic valves in younger patients It is also true, however, that a certain number of patients who need reoperation for SVD are not considered to be good surgical candidates because of their higher risk. Recently, transcatheter aortic valve replacement (TAVR) has become available for patients not only with severe native valve aortic stenosis 21 but also with failed bioprosthetic valves. 22 Both in the USA and Europe, there are several TAVR devices approved for such patients with failed bioprosthetic valves. Also, a relatively large database of TAVR usage in failed bioprosthetic valve cases was recently published. 23 Although preoperative conditions such as the diameter of the aortic annulus, vascular access and comorbidities need to be evaluated in each patient, some of the patients with failed bioprostheses can certainly be excellent candidates for valve-in-valve TAVR. It is also expected that newer devices with more advanced technology for a wider range of prostheses and a more flexible delivery system can resolve many of the current technical limitations in the near future. It remains unknown, however, whether this valve-in-valve technology will change the indication for bioprostheses in AVR because there have been no longterm follow-up studies on TAVR. Study Limitations There were several limitations in this study. First, this study cohort was formed by combining the data from 2 different retrospective, non-randomized studies. The study centers were not identical for the 2 studies. Although we used propensity score analysis to adjust for preoperative background characteristics and operative variables that may have affected the results, it was virtually impossible to exclude all selection biases: for example, the indications for bioprostheses were not the same as they are now, particularly in the 1980 s and early 1990 s. Likewise, the 10-year follow-up completeness was relatively low, at 83.2% in the CEP study and at 83.7% in the SJM study. The total number of patients was relatively large, but when these were classified into 3 subsets according to age, this was still not

8 Prosthetic Valve Choice in AVR 1205 Figure 5. Hazard ratio of overall survival, valve-related survival, reoperation, thromboembolic event, and major bleeding event according to age at operation. y, years. Other abbreviations as in Figure 1. sufficiently large to enable comparisons, especially in younger patients. These factors may have influenced the results significantly. Conclusions Although there were no differences in overall survival between bioprosthetic and mechanical valves in all age groups, bioprosthetic valves may provide better freedom from valve-related mortality and thromboembolic events in elderly patients. Mechanical valves should be the choice in patients <70 years of age who wish to avoid reoperation. Acknowledgments The original work was supported by educational grants from the Research Institute for Production Development (Kyoto, Japan). We are indebted to the participating hospitals, investigators, and clinical coordinators for their great contributions to data collection. The clinical coordinators were Ms. Yui Kinoshita, Ms. Asuka Takahashi, and Ms. Miya Hanazawa. We also thank Mr. Christian Rowthorn for his excellent English proofreading. Disclosures The authors declare no conflict of interest. References 1. Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin JP 3rd, Guyton RA, et al AHA/ACC Guideline for the Management of Patients With Valvular Heart Disease: Executive summary: A report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines. Circulation 2014; 129: Minakata K, Tanaka S, Okawa Y, Shimamoto M, Kaneko T, Takahara Y, et al. Long-term outcomes of the Carpentier-Edwards pericardial valve in the aortic position in Japanese patients. Circ J 2014; 78: Minakata K, Tanaka S, Okawa Y, Kaneko T, Okonogi S, Usui A, et al. Twenty-year outcomes of aortic valve replacement with St. Jude Medical mechanical valves in Japanese patients. Circ J 2015; 79: Akins CW, Miller DC, Turina MI, Kouchoukos NT, Blackstone EH, Grunkemeier GL, et al. Guidelines for reporting mortality and morbidity after cardiac valve interventions. Ann Thorac Surg 2008; 85: Nishida T, Sonoda H, Oishi Y, Tatewaki H, Tanoue Y, Shiokawa Y, et al. Long-term results of aortic valve replacement with mechanical prosthesis or Carpentier-Edwards Perimount bioprosthesis in Japanese patients according to age. Circ J 2014; 78: Hammermeister K, Sethi GK, Henderson WG, Grover FL, Oprian C, Rahimtoola S. Outcomes 15 years after valve replacement with a mechanical versus a bioprosthetic valve: Final report of the Veterans Affairs randomized trial. J Am Coll Cardiol 2000; 36: Oxenham H, Bloomfield P, Wheatley DJ, Lee RJ, Cunningham J, Prescott RJ, et al. Twenty year comparison of a Bjork-Shiley mechanical heart valve with porcine bioprostheses. Heart 2003; 89: Stassano P, Di Tommaso L, Monaco M, Iorio F, Pepino P, Spampinato N, et al. Aortic valve replacement: A prospective randomized evaluation of mechanical versus biological valves in patients ages 55 to 70 years. J Am Coll Cardiol 2009; 54: Chiang YP, Chikwe J, Moskowitz AJ, Itagaki S, Adams DH, Egorova NN. Survival and long-term outcomes following bioprosthetic vs mechanical aortic valve replacement in patients aged 50 to 69 years. JAMA 2014; 312: Brennan JM, Edwards FH, Zhao Y, O Brien SM, Douglas PS, Peterson ED; Developing Evidence to Inform Decisions About Effectiveness Aortic Valve Replacement (DEcIDE AVR) Research Team. Long-term survival after aortic valve replacement among high-risk elderly patients in the United States: Insights from the Society of Thoracic Surgeons Adult Cardiac Surgery Database, 1991 to Circulation 2012; 126: McClure RS, McGurk S, Cevasco M, Maloney A, Gosev I,

9 1206 MINAKATA K et al. Wiegerinck EM, et al. Late outcomes comparison of nonelderly patients with stented bioprosthetic and mechanical valves in the aortic position: A propensity-matched analysis. J Thorac Cardiovasc Surg 2014; 148: Weber A, Noureddine H, Englberger L, Dick F, Gahl B, Aymard T, et al. Ten-year comparison of pericardial tissue valves versus mechanical prostheses for aortic valve replacement in patients younger than 60 years of age. J Thorac Cardiovasc Surg 2012; 144: Brown ML, Schaff HV, Lahr BD, Mullany CJ, Sundt TM, Dearani JA, et al. Aortic valve replacement in patients aged 50 to 70 years: Improved outcome with mechanical versus biologic prostheses. J Thorac Cardiovasc Surg 2008; 135: Suri RM, Schaff HV. Is tissue valve the preferred option for patients aged 60 years and older? Circulation 2013; 128: Minakata K, Tanaka S, Takahara Y, Kaneko T, Usui A, Shimamoto M, et al. Long-term durability of pericardial valves in the aortic position in younger patients: When does reoperation become necessary? J Card Surg 2015; 30: Potter DD, Sundt TM 3rd, Zehr KJ, Dearani JA, Daly RC, Mullany CJ, et al. Operative risk of reoperative aortic valve replacement. J Thorac Cardiovasc Surg 2005; 129: van Geldorp MW, Eric Jamieson WR, Kappetein AP, Ye J, Fradet GJ, Eijkemans MJ, et al. Patient outcome after aortic valve replacement with a mechanical or biological prosthesis: Weighing lifetime anticoagulant-related event risk against reoperation risk. J Thorac Cardiovasc Surg 2009; 137: Brown JM, O Brien SM, Wu C, Sikora JA, Griffith BP, Gammie JS. Isolated aortic valve replacement in North America comprising 108,687 patients in 10 years: Changes in risks, valve types, and outcomes in the Society of Thoracic Surgeons National Database. J Thorac Cardiovasc Surg 2009; 137: Committee for Scientific Affairs, Japanese Association for Thoracic Surgery, Masuda M, Okumura M, Doki Y, Endo S, Hirata Y, Kobayashi J, et al. Thoracic and cardiovascular surgery in Japan during 2014: Annual report by The Japanese Association for Thoracic Surgery. Gen Thorac Cardiovasc Surg 2016; 64: Kaneko T, Cohn LH, Aranki SF. Tissue valve is the preferred option for patients aged 60 and older. Circulation 2013; 128: Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med 2010; 363: Hamid NB, Khalique OK, Monaghan MJ, Kodali SK, Dvir D, Bapat VN, et al. Transcatheter valve implantation in failed surgically inserted bioprosthesis: Review and practical guide to echocardiographic imaging in valve-in-valve procedures. JACC Cardiovasc Imaging 2015; 8: Dvir D, Webb JG, Bleiziffer S, Pasic M, Waksman R, Kodali S, et al. Transcatheter aortic valve implantation in failed bioprosthetic surgical valves. JAMA 2014; 312: