Article ID: CJ-21-0528
Background: This study retrospectively evaluated the long-term patient outcomes and durability of the Mosaic mitral porcine bioprosthesis in the Japan Mosaic valve long-term multicenter study.
Methods and Results: The medical records of 390 patients who underwent mitral valve replacement with the Mosaic bioprosthesis at 10 centers in Japan (1999–2014) were reviewed. Patient data were collected using the Research Electronic Data Capture software. Patient survival was determined using the Kaplan-Meier method. Freedom from structural valve deterioration (SVD) and valve-related reoperation and death were determined using actuarial methods. The median (interquartile range [IQR]) age of the cohort was 73 (69–77) years. The median (IQR) follow-up period was 4.83 (1.84–8.26) years. The longest follow-up period was 15.8 years. The 30-day mortality rate was 5.4%. The 12-year actuarial survival rate was 54.1±4.5%, and the freedom from valve-related death was 85.3±3.4%. The freedom from reoperation at 12 years was 74.3±5.7%. The freedom from SVD at 12 years was 81.4±6.6% for patients aged ≥65 years and 71.6±11.1% for those aged <65 years. The median (IQR) mean pressure gradient was 4.1 (3.0–6.0) and 5.6 (4.0–6.7) mmHg at 1 and 10 years, respectively. The median (IQR) effective orifice area was 1.7 (1.4–2.0) and 1.4 (1.2–1.6) cm2 at 1 and 10 years, respectively.
Conclusions: The Mosaic porcine bioprosthesis offered satisfactory long-term outcomes for up to 12 years.
Porcine bioprosthetic heart valves were introduced for cardiac valve replacement in the late 1960s.1,2 However, 1st-generation bioprosthetic valves have been recognized as having a high rate of early failure due to valve calcification and subsequent structural valve deterioration (SVD).3,4 In recent years, continued improvements in bioprosthesis design and tissue fixation techniques, in addition to the introduction of tissue antimineralization technology, have improved bioprosthesis performance and overall durability.
The Mosaic bioprosthesis (Medtronic plc, Minneapolis, MN, USA), a 3rd-generation stented porcine bioprosthesis, was originally released for investigational use in the USA in 1994 and later approved for clinical use by the US Food and Drug Administration in 2000. It is fixed with 0.2% glutaraldehyde solution at zero pressure to maintain the leaflet structure and minimize the consequences of antigenicity after implantation. Furthermore, this valve is treated with α-amino oleic acid (antimineralization) to improve tissue durability and hemodynamic performance. Several long-term studies have established its clinical safety and efficacy, especially in terms of hemodynamic performance and thromboembolic event rates.5–8 However, most of the reports have addressed patients undergoing Mosaic valve implantation in the aortic position. There is limited data available for the long-term performance of this bioprosthesis in the mitral position, and there are no reports on long-term outcomes in a large population of Japanese patients who underwent mitral valve replacement (MVR) with the Mosaic bioprosthesis.
The present large, multicenter, retrospective Japan Mosaic valve (J-MOVE) study examined the long-term patient outcomes and durability of the Mosaic bioprosthesis device over a period of 15 years.
This study retrospectively reviewed the records of 390 patients who underwent MVR with the Mosaic bioprosthesis (alone or in combination with coronary artery bypass grafting or other cardiac procedures) at 10 centers in Japan between November 1999 and December 2014. The follow-up survey was completed in September 2016. Data were collected retrospectively using the Research Electronic Data Capture (REDCap),9 which is a secure web-based software platform that supports data capture for research studies by providing (1) an intuitive interface for validated data capture, (2) audit trails for tracking data manipulation and export procedures, (3) automated export procedures for seamless data downloads to common statistical packages, and (4) procedures for data integration and interoperability with external sources. Patients who underwent aortic valve surgery prior to the implantation of a Mosaic valve in the mitral position were excluded. Mortality and valve-related morbidity were reported in accordance with the guidelines of the Society of Thoracic Surgeons, the American Association for Thoracic Surgery, and the European Association for Cardio-Thoracic Surgery.10 Early death was defined as occurring <30 days after surgery, regardless of the patient’s location (e.g., home or in a healthcare facility). Late death was defined as occurring >30 days after surgery. SVD was defined as dysfunction or deterioration of the prosthetic valve (excluding infection and thrombosis), such as calcification, tear, stiffness, or cusp perforation, as determined by echocardiography (at follow-up) and/or reoperation. Nonstructural valve deterioration (NSVD) was defined as any abnormality that was not intrinsic to the valve itself and resulted in stenosis or regurgitation of the prosthesis, such as pannus formation, perivalvular leakage, valve distortion, and valve dehiscence. Structural deterioration caused by endocarditis or thrombosis was not included. On strict echocardiographic assessment, the bioprosthesis was considered to have deteriorated whenever severe mitral stenosis (mean transvalvular gradient >10 mmHg, mitral valve area <1.0 cm2) or severe mitral regurgitation (effective regurgitant orifice >0.4 cm2, regurgitation volume >60 mL) was observed, even if the patient was asymptomatic. Thrombosis was defined as any thrombus not caused by infection, attached to or near an operated valve, and occluding part of the blood flow path, interfering with valve function, or large enough to warrant treatment. Embolism was defined as an embolic event occurring in the absence of infection in the immediate perioperative period. Embolism may manifest as a neurologic event or a non-cerebral embolic event. A bleeding event was defined as any episode of major internal or external bleeding that led to death, hospitalization, or permanent injury or necessitated transfusion. Valve-related death was defined as any death caused by SVD, NSVD, thrombosis, embolism, bleeding event, or operated valve endocarditis, death related to reintervention on the operated valve, or sudden unexplained death. Cardiac death included all deaths due to cardiac causes, which included valve-related deaths, sudden unexplained deaths, and deaths from non-valve-related cardiac causes (e.g., heart failure, acute myocardial infarction, or documented arrhythmias). A sudden unexplained death was defined as not having the cause of death determined by clinical investigation or autopsy findings, and an unknown relationship to the operated valve. This category included in valve-related death. Echocardiographic follow-up data were collected at the initial evaluation (before or ≤30 days after implant) and at 1 (≥6 months and <3 years), 5 (≥3 years and <8 years), and 10 years (≥8 years and <13 years) after implant.
The ethics review board of each participating center reviewed and approved the study. Individual patient consent was waived because the data were collected retrospectively and anonymized.
Statistical AnalysisContinuous variables are reported as median and interquartile range (IQR), and categorical variables are reported as frequencies and percentages. Categorical and continuous variables were analyzed using the chi-square test and Wilcoxon rank-sum test, respectively. Patient survival was determined by the Kaplan-Meier method. Valve-related reoperation, thrombosis, embolism or bleeding, valve-related death, and freedom from SVD were determined by actuarial methods. The analyses of age groups (≥65 vs. <65 years) were conducted in accordance with the American College of Cardiology (ACC) and American Heart Association (AHA) guidelines for prosthesis type.11 Comparisons between groups were performed using Cox regression models, with sex and dialysis as confounding factors. Hemodynamic performance was evaluated longitudinally using a random-intercept model. All tests were two-sided, and P values <0.05 were considered statistically significant. Statistical analysis was performed using the statistical software R (version 3.6.1, R Foundation for Statistical Computing, Vienna, Austria).
The total follow-up period for MVR was 2,089 patient-years, with a median (IQR) follow-up period of 4.83 (1.84–8.162) years (range: 0–15.8 years). The 1-, 5-, and 10-year follow-up rates were 93.5%, 80.2%, and 60.1%, respectively. The median (IQR) age of the cohort was 73 (69–77) years. Approximately 63% of patients were female. A total of 185 patients (47.9%) were classified as New York Heart Association class II. Degenerative mitral valve disease was the most common indication (64.1%) for valve replacement surgery, with mitral regurgitation being the most common diagnosis (56.9%). Mitral stenosis was observed in 22.1% of patients. The median (IQR) left ventricular ejection fraction was 64% (54–71%). The patient characteristics are summarized in Table 1A,B.
| A | |||
|---|---|---|---|
| Patient characteristics | n (%) | ||
| Sex | |||
| Female | 245 (62.8) | ||
| Male | 145 (37.2) | ||
| Age at implantation (years) | |||
| Median (IQR) | 73 (69–77) | ||
| <60 | 37 (9.5) | ||
| 60–64 | 21 (5.4) | ||
| 65–69 | 55 (14.1) | ||
| 70–74 | 137 (35.1) | ||
| 75–80 | 97 (24.9) | ||
| >80 | 43 (11.0) | ||
| Body surface area (m2) | 1.40 (1.30–1.54) | ||
| NYHA functional class | |||
| I | 35 (9.0) | ||
| II | 187 (47.9) | ||
| III | 107 (27.4) | ||
| IV | 50 (12.8) | ||
| Unknown | 11 (2.8) | ||
| Cardiac rhythm | |||
| Sinus | 151 (38.7) | ||
| Atrial fibrillation | 227 (58.2) | ||
| Other | 12 (3.1) | ||
| Mitral valve lesion | |||
| Stenosis | 86 (22.1) | ||
| Regurgitation | 222 (56.9) | ||
| Mixed | 61 (15.6) | ||
| Other | 21 (5.4) | ||
| Mitral pathology | |||
| Degenerated | 250 (64.1) | ||
| Infective endocarditis | 31 (7.9) | ||
| Barlow | 3 (0.8) | ||
| Redo | 90 (23.1) | ||
| Other | 57 (14.6) | ||
| Operative timing | |||
| Elective | 333 (85.4) | ||
| Urgent | 30 (7.7) | ||
| Emergency | 25 (6.4) | ||
| Salvage | 2 (0.5) | ||
| Ejection fraction (%) | 64 (54–71) | ||
| Comorbidities | |||
| Pulmonary hypertension | 150 (38.5) | ||
| Diabetes mellitus | 63 (16.2) | ||
| Renal dysfunction (creatinine >2.0 mg/dL) | 21 (5.4) | ||
| Cerebrovascular disease | 48 (12.3) | ||
| Chronic obstructive pulmonary disease | 20 (5.1) | ||
| Cardiogenic shock | 18 (4.6) | ||
| Extracardiac vascular disease | 15 (3.8) | ||
| Liver dysfunction | 26 (6.7) | ||
| B | Mitral stenosis | Mixed (mitral stenosis + regurgitation) |
Mitral regurgitation |
| LVDd (mm) | 45 (40–51) | 49 (43–51) | 54 (48–61) |
| LVDs (mm) | 28 (24–33) | 32 (29–37) | 34 (29–41) |
| EF (%) | 67 (59–73) | 61 (49–68) | 65 (53–72) |
| EOA/BSA (cm2/m2) | 0.69 (0.59–0.86) | 0.85 (0.65–1.11) | 1.39 (1.06–1.83) |
| MPG (mmHg) | 6.9 (5.0–9.0) | 5.6 (4–11) | 5.7 (3–9) |
| LV mass/BSA (g/m2) | 103 (84–120) | 105 (88–134) | 138 (111–168) |
Data are presented as number (percentage) or median (IQR). BSA, body surface area; EF, ejection fraction; EOA, effective orifice area; IQR, interquartile range; LV, left ventricular; LVDd, LV end-diastolic dimension; LVDs, LV end-systolic dimension; MPG, mean pressure gradient; NYHA, New York Heart Association.
The operative timing was elective in 85.4% of patients. Redo surgery was performed in 23.1% of the patients (n=90). The nominal diameter of the implanted valve was 25 mm in 23% of patients, 27 mm in 40%, 29 mm in 29%, 31 mm in 6%, and 33 mm in 2%. The median (IQR) cardiopulmonary bypass time was 161 (126–213) min, and the median (IQR) aortic cross-clamp time was 103 (79–135) min (data not shown).
Patient Survival and Valve DurabilityThe early mortality rate was 5.4% (n=21 in the operative period), and late mortality rate was 21.5% (n=84). Of the 105 deaths that occurred during the follow-up period, 39 (37.1%; including 4 sudden unexplained deaths) were cardiac, 51 (48.6%) were non-cardiac, and 15 (14.3%) were due to unknown causes. Of the 39 cardiac deaths, 6 (17.1%) were valve-related. The overall survival (OS) rate was 81.0±2.1% at 5 years and 54.1±4.5% at 12 years (Figure 1A). In patients aged ≥65 and <65 years, the 12-year OS rates were 51.4±5.4% and 61.5±8.5%, respectively (P=0.323; Figure 1B). The freedom from valve-related death was 85.3±3.4% at 12 years. The 12-year actuarial freedom from valve-related death for patients aged ≥65 and <65 years was 83.1±4.3% and 91.4±5.1%, respectively (P=0.454; Figure 2).
Kaplan-Meier curves for (A) overall survival and (B) overall survival, stratified by age group ≥65 years vs. <65 years. Log-rank test. MVR, mitral valve replacement.
Kaplan-Meier curves for freedom from valve-related death, stratified by age group ≥65 years vs. <65 years. Log-rank test. MVR, mitral valve replacement.
A total of 36 reoperations were performed over a period of 15 years. Of these, 23 were for the Mosaic bioprosthesis (patients aged ≥65 years, n=17; patients aged <65 years, n=6), the etiologic reasons for which are detailed in Table 2. SVD was the major reason for reoperation (Table 2); 10 reoperations for cardiac dysfunction, excluding Mosaic valve dysfunction, and 3 non-cardiac reoperations were performed. The 12-year actuarial freedom from reoperation was 74.3±5.7% (patients aged ≥65 years, 73.0±7.4%; patients aged <65 years, 74.8±8.5%; P=0.122; Figure 3).
| Reason for reoperation | All patients (%) (n=390) |
Age <65 years (%) (n=58) |
Age ≥65 years (%) (n=332) |
|---|---|---|---|
| Structural valve deterioration | 16 (4.10) | 6 (10.34) | 10 (3.01) |
| Perivalvular leakage | 3 (0.77) | 0 (0.00) | 3 (0.90) |
| Development of prosthetic valve regurgitation due to technical errors |
1 (0.26) | 0 (0.00) | 1 (0.30) |
| Endocarditis | 3 (0.77) | 1 (1.72) | 2 (0.60) |
| Total | 23 (5.90) | 7 (12.07) | 16 (4.82) |
Data are presented as number (percentage).
Kaplan-Meier curves for freedom from reoperation, stratified by age group ≥65 years vs. <65 years. Log-rank test. MVR, mitral valve replacement.
The actuarial freedom from SVD at 12 years was 77.9±5.8%; when stratified by age, it was 81.4±6.6% for patients aged ≥65 years and 71.6±11.1% for those aged <65 years (P=0.187; Figure 4). A total of 16 explants occurred due to SVD during the follow-up period. The median (IQR) age at implantation was 66.4 (64.0–70.8) years, and the median (IQR) duration since the implant was 9.8 (7.0–12.2) years. The predominant lesion was calcification in 9 patients (56.2%) and leaflet tear in 7 patients (43.8%).
Kaplan-Meier curves for freedom from reoperation for structural valve deterioration (SVD), stratified by age group ≥65 years vs. <65 years. Log-rank test. MVR, mitral valve replacement.
No cases of thrombosis were reported. A total of 9 embolic events (patients aged ≥65 years, n=8; patients aged <65 years, n=1) and 18 bleeding events (patients aged ≥65 years, n=16; patients aged <65 years, n=2) were reported. The 12-year actuarial freedom from a composite of thrombosis, embolism, and bleeding, stratified by age group, was 89.4±2.5% for patients aged ≥65 years and 91.9±4.9% for those aged <65 years (P=0.460; Figure 5).
Kaplan-Meier curves for freedom from a composite of thrombosis, embolism, and bleeding, stratified by age group ≥65 years vs. <65 years. Log-rank test. MVR, mitral valve replacement.
The hemodynamic performance is illustrated in Figure 6A,B. The median (IQR) mean pressure gradient was 4.1 (3.0–6.0) and 5.6 (4.0–6.7) mmHg at 1 and 10 years, respectively. The mitral mean (IQR) pressure gradient for all valve sizes was 4.0 (3.0–5.7) mmHg for patients aged ≥65 years and 5.1 (4.6–6.0) mmHg for those aged <65 years at 1 year, and 5.0 (4.0–6.4) mmHg for patients aged ≥65 years and 6.2 (4.3–7.1) mmHg for those aged <65 years at 10 years (P=0.206). The median (IQR) effective orifice area (EOA) was 1.7 (1.4–2.0) and 1.4 (1.2–1.6) cm2 at 1 and 10 years, respectively. The median (IQR) EOA index was 1.7 (1.4–2.0) cm2/m2 for patients aged ≥65 years and 1.6 (1.4–1.7) cm2/m2 for those aged <65 years at 1 year, and 1.5 (1.2–1.8) cm2/m2 for patients aged ≥65 years and 1.2 (1.2–1.3) cm2/m2 for those aged <65 years at 10 years (P=0.113).
(A) Mitral mean pressure gradient (PG) and (B) effective orifice area (EOA) index for all valve sizes.
J-MOVE has been the largest retrospective, multicenter clinical study to investigate the long-term results of the Mosaic bioprosthesis device in Japan, and it was a large series of long-term outcomes in patients undergoing MVR with a bioprosthetic valve. According to the ACC and AHA recommendations, it is rational to select a bioprosthesis in patients aged ≥65 years who require MVR and a mechanical prosthesis in those aged <65 years who have no contraindication to anticoagulation.12 There has been a considerable shift away from mechanical valves toward mitral bioprosthetic valves, possibly due to the risks of anticoagulation therapy associated with the use of mechanical valves in an aging population.13,14 However, no large-scale studies regarding the long-term outcomes of bioprosthetic vs. mechanical MVR have been conducted. There is a trade-off between the incremental risk of SVD and reoperation associated with bioprosthetic valves and the greater risk of stroke and major bleeding with mechanical valves. The treatment of younger patients with longer life expectancies has raised questions regarding valve durability because ideally it should be longer than the patient’s life expectancy. SVD leads to valve dysfunction, which reduces valve durability. Therefore, valve durability has emerged as a fundamental issue.
In the current study, the long-term outcomes and durability of the Mosaic bioprosthesis were evaluated in the mitral position, stratified by age group. The 5-year OS rate was 81.0%. The long-term OS rate observed in this study at 12 years (54.1%) with the Mosaic bioprosthesis was similar to that observed with the Carpentier-Edwards pericardial valve (50.9% at 10 years and 16.5% at 15 years).15
In this study, no significant difference was observed in the 12-year OS rate between patients aged ≥65 and <65 years. The 12-year OS rate in patients aged <65 years with the Mosaic valve (61.5%) was similar to that observed with the Carpentier-Edwards pericardial valve (56.1% at 10 years and 35.1% at 15 years).15
In the current study, 23 reoperations were performed over a period of 12 years. A retrospective long-term study in patients aged 50–69 years reported a 15-year cumulative incidence of reoperation of 5.0% with a mechanical prosthesis vs. 11.1% with a bioprosthesis, with no significant difference in long-term survival between the valve types,14 implying that bioprosthesis reoperation does not have an increased risk of death in the younger population when compared with mechanical prosthesis.
One of the major problems associated with bioprostheses is SVD, which reduces valve durability and necessitates reoperation. In this study, SVD was more common in patients aged <65 years than in those aged ≥65 years. This is consistent with the findings of other studies that have reported younger age at implant as a significant risk factor for SVD.15,16 In the current study, explants due to SVD were required in 16 patients during the follow-up period, with calcification being the predominant cause of SVD in 9 cases (56.2%) and leaflet tear in 7 cases (43.8%). A study of the Carpentier-Edwards pericardial valve reported calcification as the major cause of SVD (75%), followed by leaflet tear (23%).17 The difference in results could be due to the antimineralization of the Mosaic porcine valve with α-amino oleic acid, which reduces calcification.18
In this study, the freedom from SVD at 12 years for the Mosaic valve was 81.4% for patients aged ≥65 years and 71.6% for those aged <65 years. In contrast, with the Carpentier-Edwards pericardial valve, the freedom from SVD was 86.3% at 10 years and 62.8% at 15 years, and 62.5% for patients aged ≥65 years and 47.3% for those aged <65 years at 15 years,15 which was considerably higher than the freedom of SVD at 12 years, especially in the young population. The Mosaic bioprosthesis may be superior to pericardial tissue valves in patients aged <65 years with regard to long-term outcomes.
The freedom from a composite of thrombosis, embolism, and bleeding at 12 years in this study was 89.4% for patients aged ≥65 years and 91.9% for those aged <65 years. These results were similar to the rates observed with the Carpentier-Edwards pericardial valve (freedom from thromboembolic and bleeding events at 15 years was 91.5% and 91.6%, respectively).15
With mitral bioprostheses, the aim with respect to hemodynamic performance is to decrease the transprosthetic gradient and increase the EOA.19 In this study, the mean diastolic pressure gradient was 4.1 and 5.6 mmHg at 1 and 10 years, respectively, which was similar to that observed with the Carpentier-Edwards pericardial valve (5.6 mmHg at 10 years).20 Further, in the current study, the mean EOA was 1.7 and 1.4 cm2 at 1 and 10 years, respectively. Given no increase in the transprosthetic gradient and no decrease in the EOA with time, the hemodynamic performance of the Mosaic porcine valve was considered to be excellent over a period of 10 years.
In this era, when bioprostheses are increasingly being used,21 the associated low mortality risk of reoperation due to SVD with bioprostheses, when compared with the increased risk of thrombosis, embolism, and bleeding secondary to anticoagulation with mechanical prostheses, potentially makes the use of bioprosthetic valve an acceptable option in the younger population. Furthermore, transcatheter mitral valve replacement (TMVR) for degenerated bioprostheses has emerged as a new treatment option for the management of SVD in patients deemed at high surgical risk.22–24 As the long-term outcomes of TMVR become clear, the use of valve-in-valve TMVR after SVD may resolve the issues with reoperation, and the trend toward the use of bioprosthetic valves in the younger generation may increase. This 15-year study revealed satisfactory long-term clinical outcomes with MVR using the Mosaic bioprosthesis even in patients aged <65 years, which is encouraging for surgeons as well as patients who are faced with making decisions on valve selection.
Study LimitationsThis study has several limitations related to its retrospective, observational, and nonrandomized nature. The number of patients lost to follow-up was high, leading to fewer patients available for analysis at 15 years. The median follow-up period was only 4.84 years. The 10- and 15-year follow-up rates were 60.1% and 44.6%, respectively. The results may have significantly been affected by the small number of longer follow-up patients. In addition, no information was recorded on postoperative anticoagulation or antiplatelet management.
The Mosaic porcine bioprosthesis in the mitral position offered satisfactory long-term outcomes over a period of 12 years, including OS, freedom from valve-related death, reoperation for SVD, and a composite of thrombosis, embolism, and bleeding. The actuarial freedom from SVD and reoperation for SVD were acceptable even in patients younger than 65 years, suggesting that the Mosaic bioprosthesis is a reasonable alternative to mechanical prosthetic valves in this population for patients who wish to avoid lifelong anticoagulation.
Y.S. is a member of Circulation Journal’s Editorial Team. Other authors have no conflicts of interest related to the manuscript.
Ethical Review Board of Osaka University Hospital (reference no. 14283). The Ethics Committee of Kobe University Graduate School of Medicine (reference no. 1809). Ethical Review Board of Kyoto Prefectural University of Medicine (reference no. ERB-C-484). IRB of National Cerebral and Cardiovascular Center (reference no. M27-004). Saiseikai Kumamoto Hospital Ethics Committee (reference no. 447). Ethical Review Committee of Sakakibara Heart Institute (reference no. 15-018). IRB of Japanese Red Cross Nagoya Daiichi Hospital (reference no. 2015-001). New Tokyo Hospital Research Ethics Committee (reference no. 0086). Kobe City Medical Center General Hospital Research Ethics Committee (reference no. 15031). Ethical Committee of The Sakakibara Heart Institute of Okayama (reference no. 20150804).
