Long-term results after concomitant mitral and aortic valve surgery: repair or replacement?

Long-term results after concomitant mitral and aortic valve surgery: repair or replacement? Abstract OBJECTIVES The reported superiority of mitral valve (MV) repair for isolated MV regurgitation has not been confirmed in mitroaortic valve surgery. Our goals were to evaluate the feasibility of repair in patients undergoing mitral and aortic valve surgery and to identify factors predisposing to MV replacement, to compare long-term outcomes (survival and MV reoperation) of repair and replacement and to perform a subgroup analysis in patients with rheumatic MV disease. METHODS From January 1992 through December 2016, 1122 consecutive patients were submitted to concomitant aortic and MV surgery in 2 different centres (Coimbra and Santiago). Of these, 837 patients underwent MV repair (74.6%) and 285 patients had MV replacement (25.4%). Rheumatic aetiology was predominant (666 patients; 59.4%). Cumulative follow-up was 9522.6 patient-years (25th–75th percentile 2.6–13.2 years) and was complete for 95.6% of patients. Propensity score matching (1:1) was performed in 232 patients for comparing each treatment option (MV repair and MV replacement). RESULTS Previous MV intervention, rheumatic aetiology, chronic obstructive pulmonary disease, higher degrees of tricuspid and mitral regurgitation and pulmonary hypertension were independently correlated with MV replacement. The 30-day mortality rate was higher in patients with MV replacement (4.2% vs 1.8%, P = 0.021) and was confirmed in the propensity score matching (4.7% vs 1.7%, P = 0.06). Late survival was lower in the MV replacement group (53.3 ± 4.5% vs 61.7 ± 2.0% at 12 years; P = 0.026) and was confirmed in the propensity score matching (54.6 ± 4.9% vs 63.2 ± 3.8%, P = 0.062) and rheumatic subgroup (57.9 ± 4.8% vs 68.0 ± 2.5%, P = 0.018). Freedom from MV reoperation at 12 years was higher in the MV repair group (94.7 ± 1.1% vs 89.0 ± 3.1%, P = 0.004) but similar in patients with rheumatic MV disease. CONCLUSIONS MV repair can be performed in most patients undergoing aortic valve replacement. It should be the procedure of choice whenever feasible, because it is associated with lower early and late mortality rates and with freedom from reoperation in non-rheumatic patients. View largeDownload slide View largeDownload slide Aortic valve replacement, Mitral valve repair, Mitral valve replacement, Long-term survival, Freedom from reoperation INTRODUCTION Aortic valve stenosis and mitral valve regurgitation (MR) have become the most frequent valvular diseases in Western countries in recent decades [1]. Increasing life expectancy allied with improved surgical outcomes has contributed to a growing need for concomitant aortic and mitral valve (MV) surgery [2, 3]. However, the potential benefit of MV repair over replacement in patients having double valve surgery remains uncertain. Theoretically, double valve replacement would be the likely procedure, because there would be no advantage in repairing a valve when another required replacement. Nevertheless, the excellent results of isolated MV repair in the correction of severe degenerative MR with regard to hospital mortality, long-term survival, preservation of left ventricular function and freedom from valve-related complications (thromboembolism, anticoagulant-related bleeding and endocarditis) [4–6] have encouraged its application in cases of aortic valve replacement (AVR). The lack of convincing data on the use of MV repair in combined or multiple-valve disease precludes one from making evidence-based recommendations. The major international guidelines for the management of valvular heart disease (European Society of Cardiology/European Association for Cardio-Thoracic Surgery and American College of Cardiology/American Heart Association) [7, 8] are vague in this regard, partially due to the absence of randomized controlled trials and the scarcity of observational studies. Although the current trend is to favour repair in patients with non-rheumatic MV disease [9], there is considerable controversy surrounding MV repair in patients with rheumatic disease. Leaflet fibrosis, subvalvular thickening and retraction and commissural fusion all contribute to lower probability and durability of the repair. Poorer outcomes have driven surgeons away from attempting repair in this context, which poses additional difficulties in interpreting surgical results, as there are even fewer studies addressing this subject [10–12]. Therefore, our goals were to (i) evaluate the feasibility of repair in patients undergoing mitral and aortic valve surgery and to identify factors predisposing to replacement, (ii) compare long-term outcomes (survival and MV reoperation) of repair and replacement in this context and (iii) perform a subgroup analysis in patients with rheumatic MV disease. MATERIALS AND METHODS Patient population From January 1992 through December 2016, 1122 patients were consecutively submitted to concomitant aortic (replacement only) and MV surgery (repair or replacement) in 2 different centres [university hospitals of Coimbra, Portugal (n = 1039) and Santiago de Compostela, Spain (n = 83)]. Of these, 837 patients underwent MV repair (74.6%) and 285 patients had MV replacement (25.4%). Patients with additional procedures, such as coronary artery bypass graft (CABG), tricuspid valve surgery and ascending aorta surgery, were admitted to the study. Patients with previous MV interventions other than MV replacement (percutaneous mitral balloon valvuloplasty, open or closed commissurotomy and other MV valvuloplasty) were also included. The mean age of the population was 63.0 ± 12.8 years (25th–75th percentile 53–70). Patients who had MV replacement were significantly older (63.6 ± 10.7 vs 59.7 ± 13.4 years of age, P < 0.001), more symptomatic, had more comorbidities and a greater prevalence of tricuspid valve disease. Patient characteristics are listed in Table 1. Rheumatic disease was the most frequent aetiology of MV disease, present in 666 patients (59.4%), followed by degenerative disease in 326 (29%) and secondary (functional) disease in 59 (5.3%). All patients from both centres had moderate to severe or severe MV disease, either as stenosis, regurgitation or mixed lesion. Table 1: Demographic and echocardiographic data Variables Overall (n = 1122) Mitral valve repair (n = 837) Mitral valve replacement (n = 285) P-value Age (years), mean ± SD 60.6 ± 12.8 59.7 ± 13.4 63.3 ± 10.7 <0.001 Male sex, n (%) 558 (49.7) 428 (51.1) 130 (45.6) 0.10 Body surface area, mean ± SD 1.66 ± 0.18 1.66 ± 0.17 1.67 ± 0.19 0.46 Aetiology of MV disease, n (%)  Degenerative 326 (29.0) 272 (32.5) 54 (18.9) <0.001  Rheumatic 666 (59.4) 461 (55.1) 205 (71.9) <0.001  Secondary 59 (5.3) 58 (6.9) 1 (0.4) <0.001  Others 71 (6.3) 46 (5.5) 25 (8.8) 0.069 MV pathology, n (%)  Stenosis 310 (27.6) 240 (28.7) 70 (24.6) 0.18  Regurgitation 547 (48.8) 451 (53.9) 96 (33.7) <0.001  Mixed lesion 265 (23.6) 146 (17.4) 119 (41.8) <0.001 Previous MV surgery, n (%) 127 (11.3) 49 (5.9) 78 (27.4) <0.001 Tricuspid valve disease, n (%) 233 (20.8) 145 (17.3) 88 (30.9) <0.001 Coronary artery disease, n (%) 122 (10.9) 106 (12.7) 16 (5.6) 0.001 NYHA III–IV, n (%) 731 (65.2) 525 (62.7) 206 (72.3) 0.003 Stroke, n (%) 64 (5.7) 45 (5.4) 19 (6.7) 0.41 Hypertension, n (%) 321 (28.6) 216 (25.8) 105 (36.8) <0.001 Diabetes, n (%) 101 (9.0) 59 (7.1) 42 (14.8) <0.001 Chronic kidney disease, n (%) 31 (2.8) 23 (2.8) 8 (2.8) 0.96 COPD, n (%) 85 (7.6) 49 (5.9) 36 (12.6) <0.001 Atrial fibrillation/flutter, n (%) 441 (39.3) 313 (37.4) 128 (44.9) 0.025 Ejection fraction (%), mean ± SD 61.2 ± 10.6 61.0 ± 10.7 61.6 ± 10.0 0.43 LV end-systolic dimension (mm), mean ± SD 41.8 ± 10.7 43.3 ± 10.8 36.8 ± 8.3 <0.001 Left atrial diameter (mm), mean ± SD 54.5 ± 10.5 53.3 ± 10.6 57.9 ± 9.6 <0.001 Mean aortic gradient (mmHg), mean ± SD 48.3 ± 22.2 49.1 ± 22.4 46.2 ± 21.5 0.22 sPAP, mean ± SD 53.8 ± 17.3 51.7 ± 16.9 55.9 ± 18.2 0.002 Variables Overall (n = 1122) Mitral valve repair (n = 837) Mitral valve replacement (n = 285) P-value Age (years), mean ± SD 60.6 ± 12.8 59.7 ± 13.4 63.3 ± 10.7 <0.001 Male sex, n (%) 558 (49.7) 428 (51.1) 130 (45.6) 0.10 Body surface area, mean ± SD 1.66 ± 0.18 1.66 ± 0.17 1.67 ± 0.19 0.46 Aetiology of MV disease, n (%)  Degenerative 326 (29.0) 272 (32.5) 54 (18.9) <0.001  Rheumatic 666 (59.4) 461 (55.1) 205 (71.9) <0.001  Secondary 59 (5.3) 58 (6.9) 1 (0.4) <0.001  Others 71 (6.3) 46 (5.5) 25 (8.8) 0.069 MV pathology, n (%)  Stenosis 310 (27.6) 240 (28.7) 70 (24.6) 0.18  Regurgitation 547 (48.8) 451 (53.9) 96 (33.7) <0.001  Mixed lesion 265 (23.6) 146 (17.4) 119 (41.8) <0.001 Previous MV surgery, n (%) 127 (11.3) 49 (5.9) 78 (27.4) <0.001 Tricuspid valve disease, n (%) 233 (20.8) 145 (17.3) 88 (30.9) <0.001 Coronary artery disease, n (%) 122 (10.9) 106 (12.7) 16 (5.6) 0.001 NYHA III–IV, n (%) 731 (65.2) 525 (62.7) 206 (72.3) 0.003 Stroke, n (%) 64 (5.7) 45 (5.4) 19 (6.7) 0.41 Hypertension, n (%) 321 (28.6) 216 (25.8) 105 (36.8) <0.001 Diabetes, n (%) 101 (9.0) 59 (7.1) 42 (14.8) <0.001 Chronic kidney disease, n (%) 31 (2.8) 23 (2.8) 8 (2.8) 0.96 COPD, n (%) 85 (7.6) 49 (5.9) 36 (12.6) <0.001 Atrial fibrillation/flutter, n (%) 441 (39.3) 313 (37.4) 128 (44.9) 0.025 Ejection fraction (%), mean ± SD 61.2 ± 10.6 61.0 ± 10.7 61.6 ± 10.0 0.43 LV end-systolic dimension (mm), mean ± SD 41.8 ± 10.7 43.3 ± 10.8 36.8 ± 8.3 <0.001 Left atrial diameter (mm), mean ± SD 54.5 ± 10.5 53.3 ± 10.6 57.9 ± 9.6 <0.001 Mean aortic gradient (mmHg), mean ± SD 48.3 ± 22.2 49.1 ± 22.4 46.2 ± 21.5 0.22 sPAP, mean ± SD 53.8 ± 17.3 51.7 ± 16.9 55.9 ± 18.2 0.002 COPD: chronic obstructive pulmonary disease; LV: left ventricle; MV: mitral valve; NYHA: New York Heart Association; SD: standard deviation; sPAP: systolic pulmonary artery pressure. Statistically significant P-values (<0.05) are in bold. Table 1: Demographic and echocardiographic data Variables Overall (n = 1122) Mitral valve repair (n = 837) Mitral valve replacement (n = 285) P-value Age (years), mean ± SD 60.6 ± 12.8 59.7 ± 13.4 63.3 ± 10.7 <0.001 Male sex, n (%) 558 (49.7) 428 (51.1) 130 (45.6) 0.10 Body surface area, mean ± SD 1.66 ± 0.18 1.66 ± 0.17 1.67 ± 0.19 0.46 Aetiology of MV disease, n (%)  Degenerative 326 (29.0) 272 (32.5) 54 (18.9) <0.001  Rheumatic 666 (59.4) 461 (55.1) 205 (71.9) <0.001  Secondary 59 (5.3) 58 (6.9) 1 (0.4) <0.001  Others 71 (6.3) 46 (5.5) 25 (8.8) 0.069 MV pathology, n (%)  Stenosis 310 (27.6) 240 (28.7) 70 (24.6) 0.18  Regurgitation 547 (48.8) 451 (53.9) 96 (33.7) <0.001  Mixed lesion 265 (23.6) 146 (17.4) 119 (41.8) <0.001 Previous MV surgery, n (%) 127 (11.3) 49 (5.9) 78 (27.4) <0.001 Tricuspid valve disease, n (%) 233 (20.8) 145 (17.3) 88 (30.9) <0.001 Coronary artery disease, n (%) 122 (10.9) 106 (12.7) 16 (5.6) 0.001 NYHA III–IV, n (%) 731 (65.2) 525 (62.7) 206 (72.3) 0.003 Stroke, n (%) 64 (5.7) 45 (5.4) 19 (6.7) 0.41 Hypertension, n (%) 321 (28.6) 216 (25.8) 105 (36.8) <0.001 Diabetes, n (%) 101 (9.0) 59 (7.1) 42 (14.8) <0.001 Chronic kidney disease, n (%) 31 (2.8) 23 (2.8) 8 (2.8) 0.96 COPD, n (%) 85 (7.6) 49 (5.9) 36 (12.6) <0.001 Atrial fibrillation/flutter, n (%) 441 (39.3) 313 (37.4) 128 (44.9) 0.025 Ejection fraction (%), mean ± SD 61.2 ± 10.6 61.0 ± 10.7 61.6 ± 10.0 0.43 LV end-systolic dimension (mm), mean ± SD 41.8 ± 10.7 43.3 ± 10.8 36.8 ± 8.3 <0.001 Left atrial diameter (mm), mean ± SD 54.5 ± 10.5 53.3 ± 10.6 57.9 ± 9.6 <0.001 Mean aortic gradient (mmHg), mean ± SD 48.3 ± 22.2 49.1 ± 22.4 46.2 ± 21.5 0.22 sPAP, mean ± SD 53.8 ± 17.3 51.7 ± 16.9 55.9 ± 18.2 0.002 Variables Overall (n = 1122) Mitral valve repair (n = 837) Mitral valve replacement (n = 285) P-value Age (years), mean ± SD 60.6 ± 12.8 59.7 ± 13.4 63.3 ± 10.7 <0.001 Male sex, n (%) 558 (49.7) 428 (51.1) 130 (45.6) 0.10 Body surface area, mean ± SD 1.66 ± 0.18 1.66 ± 0.17 1.67 ± 0.19 0.46 Aetiology of MV disease, n (%)  Degenerative 326 (29.0) 272 (32.5) 54 (18.9) <0.001  Rheumatic 666 (59.4) 461 (55.1) 205 (71.9) <0.001  Secondary 59 (5.3) 58 (6.9) 1 (0.4) <0.001  Others 71 (6.3) 46 (5.5) 25 (8.8) 0.069 MV pathology, n (%)  Stenosis 310 (27.6) 240 (28.7) 70 (24.6) 0.18  Regurgitation 547 (48.8) 451 (53.9) 96 (33.7) <0.001  Mixed lesion 265 (23.6) 146 (17.4) 119 (41.8) <0.001 Previous MV surgery, n (%) 127 (11.3) 49 (5.9) 78 (27.4) <0.001 Tricuspid valve disease, n (%) 233 (20.8) 145 (17.3) 88 (30.9) <0.001 Coronary artery disease, n (%) 122 (10.9) 106 (12.7) 16 (5.6) 0.001 NYHA III–IV, n (%) 731 (65.2) 525 (62.7) 206 (72.3) 0.003 Stroke, n (%) 64 (5.7) 45 (5.4) 19 (6.7) 0.41 Hypertension, n (%) 321 (28.6) 216 (25.8) 105 (36.8) <0.001 Diabetes, n (%) 101 (9.0) 59 (7.1) 42 (14.8) <0.001 Chronic kidney disease, n (%) 31 (2.8) 23 (2.8) 8 (2.8) 0.96 COPD, n (%) 85 (7.6) 49 (5.9) 36 (12.6) <0.001 Atrial fibrillation/flutter, n (%) 441 (39.3) 313 (37.4) 128 (44.9) 0.025 Ejection fraction (%), mean ± SD 61.2 ± 10.6 61.0 ± 10.7 61.6 ± 10.0 0.43 LV end-systolic dimension (mm), mean ± SD 41.8 ± 10.7 43.3 ± 10.8 36.8 ± 8.3 <0.001 Left atrial diameter (mm), mean ± SD 54.5 ± 10.5 53.3 ± 10.6 57.9 ± 9.6 <0.001 Mean aortic gradient (mmHg), mean ± SD 48.3 ± 22.2 49.1 ± 22.4 46.2 ± 21.5 0.22 sPAP, mean ± SD 53.8 ± 17.3 51.7 ± 16.9 55.9 ± 18.2 0.002 COPD: chronic obstructive pulmonary disease; LV: left ventricle; MV: mitral valve; NYHA: New York Heart Association; SD: standard deviation; sPAP: systolic pulmonary artery pressure. Statistically significant P-values (<0.05) are in bold. Definitions Mortality and morbidity were reported according to the latest ‘guidelines for reporting mortality and morbidity after cardiac valve interventions’ [13]. Early mortality was defined as death in hospital or within 30 days, and late mortality was defined as death occurring beyond this period. MV aetiology was classified following a thorough analysis of the clinical information, echocardiograms and operative reports. For the purpose of this work and attending to the most frequent aetiologies found, we divided MV disease into 4 major categories: degenerative, rheumatic, secondary (functional) and others (only 6.3% of the entire study population). Secondary MV disease was defined as dysfunction without structural abnormalities of the MV apparatus such as valve prolapse, significant calcification of leaflets or annulus, ruptured chordae (degenerative) and concomitant mitral stenosis (rheumatic) [14]. All patients had preoperative detailed echocardiographic and Doppler examinations. The severity of MR and other relevant measures were determined according to the accepted recommendations [15]. Intraoperative transoesophageal echocardiography, both pre- and post-repair, was routinely used from the beginning of the study, and no patient left the operating room after repair with greater than mild MR, as described previously [16]. Our anticoagulation protocol included a target international normalized ratio of 2.0–3.0 for patients undergoing isolated AVR (mechanical prosthesis) and 2.5–3.5 for those having double valve replacement (mechanical). Patients with a bioprosthesis, either aortic or mitral, were given anti-aggregants only, if there were no other indications for anticoagulation. Data collection Data were retrieved from dedicated databases and included relevant preoperative demographic, clinical and echocardiographic variables, surgical information and postoperative records. Follow-up information, complete for 95.6% of patients, was obtained through a mailed questionnaire or by telephone interview with surviving patients, family members or the patients’ personal physicians, and included vital status and the need for MV reoperation. The cumulative follow-up for the entire cohort was 9522.6 patient-years (mean 8.5 ± 6.7 years; 25th–75th percentile 2.6–13.2 years). All patients gave informed consent for surgery and granted us permission to use their medical records for research purposes, using forms approved by the ethic committee of the respective institutions. Operative procedures and data Mitral and aortic valve exposure was similar in both centres and included a left atriotomy posterior to Waterston’s groove and a ‘hockey-stick’ incision in the ascending aorta, respectively. However, both centres differed significantly regarding the management of MV disease. In Coimbra, repair was attempted in the majority of patients (79.1% repair rate). In Santiago, MV repair was performed less frequently (18.1%). The operative procedures are detailed in Table 2. Table 2: Operative data and mortality Procedures Mitral valve repair Mitral valve replacement P-value Mechanical aortic valve, n (%) 559 (66.8) 199 (69.8) 0.34 Mechanical mitral valve, n (%) 194 (45.6) Tricuspid valve surgery, n (%) 103 (12.3) 74 (41.8) <0.001 CABG, n (%) 72 (8.6) 12 (4.2) 0.023 MV repair techniques  Commissurotomy, n (%) 419 (50.1)  Papillotomy, n (%) 163 (19.5)  Annuloplasty (complete/posterior), n (%) 732 (87.5)  Mean ring size (mm), mean ± SD 32 ± 1.6  Posterior leaflet resection, n (%) 43 (5.1)  Neochordae, n (%) 74 (8.8)  Chordae/transposition shortening, n (%) 14 (1.7) ECC time (min), mean ± SD 87.4 ± 15.9 120.4 ± 51.9 <0.001 Aortic clamping time (min), mean ± SD 59.1 ± 17.7 90.3 ± 44.9 <0.001 30-Day mortality rate, n (%) 15 (1.8) 12 (4.2) 0.021 Procedures Mitral valve repair Mitral valve replacement P-value Mechanical aortic valve, n (%) 559 (66.8) 199 (69.8) 0.34 Mechanical mitral valve, n (%) 194 (45.6) Tricuspid valve surgery, n (%) 103 (12.3) 74 (41.8) <0.001 CABG, n (%) 72 (8.6) 12 (4.2) 0.023 MV repair techniques  Commissurotomy, n (%) 419 (50.1)  Papillotomy, n (%) 163 (19.5)  Annuloplasty (complete/posterior), n (%) 732 (87.5)  Mean ring size (mm), mean ± SD 32 ± 1.6  Posterior leaflet resection, n (%) 43 (5.1)  Neochordae, n (%) 74 (8.8)  Chordae/transposition shortening, n (%) 14 (1.7) ECC time (min), mean ± SD 87.4 ± 15.9 120.4 ± 51.9 <0.001 Aortic clamping time (min), mean ± SD 59.1 ± 17.7 90.3 ± 44.9 <0.001 30-Day mortality rate, n (%) 15 (1.8) 12 (4.2) 0.021 Statistically significant P-values (<0.05) are in bold. AL: anterior leaflet; ECC: extracorporeal circulation; PL: posterior leaflet; SD: standard deviation. Table 2: Operative data and mortality Procedures Mitral valve repair Mitral valve replacement P-value Mechanical aortic valve, n (%) 559 (66.8) 199 (69.8) 0.34 Mechanical mitral valve, n (%) 194 (45.6) Tricuspid valve surgery, n (%) 103 (12.3) 74 (41.8) <0.001 CABG, n (%) 72 (8.6) 12 (4.2) 0.023 MV repair techniques  Commissurotomy, n (%) 419 (50.1)  Papillotomy, n (%) 163 (19.5)  Annuloplasty (complete/posterior), n (%) 732 (87.5)  Mean ring size (mm), mean ± SD 32 ± 1.6  Posterior leaflet resection, n (%) 43 (5.1)  Neochordae, n (%) 74 (8.8)  Chordae/transposition shortening, n (%) 14 (1.7) ECC time (min), mean ± SD 87.4 ± 15.9 120.4 ± 51.9 <0.001 Aortic clamping time (min), mean ± SD 59.1 ± 17.7 90.3 ± 44.9 <0.001 30-Day mortality rate, n (%) 15 (1.8) 12 (4.2) 0.021 Procedures Mitral valve repair Mitral valve replacement P-value Mechanical aortic valve, n (%) 559 (66.8) 199 (69.8) 0.34 Mechanical mitral valve, n (%) 194 (45.6) Tricuspid valve surgery, n (%) 103 (12.3) 74 (41.8) <0.001 CABG, n (%) 72 (8.6) 12 (4.2) 0.023 MV repair techniques  Commissurotomy, n (%) 419 (50.1)  Papillotomy, n (%) 163 (19.5)  Annuloplasty (complete/posterior), n (%) 732 (87.5)  Mean ring size (mm), mean ± SD 32 ± 1.6  Posterior leaflet resection, n (%) 43 (5.1)  Neochordae, n (%) 74 (8.8)  Chordae/transposition shortening, n (%) 14 (1.7) ECC time (min), mean ± SD 87.4 ± 15.9 120.4 ± 51.9 <0.001 Aortic clamping time (min), mean ± SD 59.1 ± 17.7 90.3 ± 44.9 <0.001 30-Day mortality rate, n (%) 15 (1.8) 12 (4.2) 0.021 Statistically significant P-values (<0.05) are in bold. AL: anterior leaflet; ECC: extracorporeal circulation; PL: posterior leaflet; SD: standard deviation. Anticoagulant management Anticoagulation was initiated with warfarin on the first or second postoperative day. A target international normalized ratio of 2.0–3.0 was adopted for isolated mechanical AVR and 2.5–3.5 for double mechanical valve replacement. Patients with a bioprosthesis were maintained with aspirin after a 3-month period (after surgery) under anticoagulation, if there was no formal indication for anticoagulation. Statistical analysis The statistical analysis was performed according to the statistical and data reporting guidelines of the European Journal of Cardio-Thoracic Surgery and Interactive CardioVascular and Thoracic Surgery [17]. Continuous variables are reported as mean ± standard deviation and the median with the 25th–75th percentile and compared using the independent Student’s t-test if normally distributed and using the Mann–Whitney U-test for variables with non-normal distribution. Univariable analysis of categorical data was performed using the χ2 (when no cell was expected to count <5 in contingency tables) or the Fisher’s exact test, and the results are presented as percentages. A multiple logistic regression was performed to identify factors predisposing to replacement rather than repair. Overall survival and freedom from reoperation were plotted using the Kaplan–Meier method, and comparison was made by the log-rank test. Multivariable analyses to identify risk factors for late mortality or reoperation were performed using Cox regression models and calculating hazard ratios and 95% confidence intervals (CIs). All variables with a P-value <0.2 in the univariable analysis were entered in the multivariable analysis (logistic regression or Cox models); in the final model, only those variables with a P-value <0.05 or clinically relevant were retained. To estimate the probability that a patient would have the MV replaced rather than repaired, a multivariable logistic regression model was designed by incorporating demographic information, clinical status and relevant cardiac and non-cardiac comorbidities (Supplementary Material, Table S1), and a propensity score was determined for each patient. Patients were matched according to the propensity score previously calculated by the ‘nearest neighbour matching’ technique, using a calliper of 0.1. Each patient was matched to a single patient (no-replacement). After matching, 232 patients for each group (MV repair vs MV replacement) were obtained for comparison (survival and freedom from reoperation). Subgroup analysis was performed to evaluate the impact of repairing versus replacing the MV in patients with rheumatic MV disease undergoing AVR. This analysis was prespecified. Statistical significance was defined as a 2-tailed P-value <0.05. The data were analysed using the statistical package program SPSS (version 20.0. IBM Corp., Armonk, NY, USA). RESULTS Assessment of the feasibility of mitral valve repair MV repair was performed in 74.6% of the whole population and in 79.2% of patients who had primary MV surgery. In the higher volume centre (Coimbra), the rate of MV repair (primary MV intervention) was even higher, increasing to 94.5% in patients with degenerative MR (flail leaflet) and to 77.6% in those with rheumatic MV disease. Several factors were significantly associated with the probability of having MV replacement in the univariable analysis: MV aetiology; coronary artery disease (CAD); degree of tricuspid valve regurgitation; hypertension; diabetes; atrial fibrillation; chronic obstructive pulmonary disease; degree of MR; higher left atrium and ventricular (systolic) dimensions; higher transaortic gradients; and higher systolic pulmonary artery pressure, surgeon (most experienced) and surgical centre. In the multivariable analysis, only previous MV intervention, surgeon, surgical centre, rheumatic involvement, chronic obstructive pulmonary disease and higher degrees of tricuspid valve regurgitation, MR and systolic pulmonary artery pressure were correlated with MV replacement (Table 3). Table 3: Risk factors for mitral valve replacement (logistic regression) Risk factors OR 95% CI P-value Previous MV surgery 4.776 2.942–0.7753 <0.001 COPD 2.316 1.258–4.266 0.007 sPAP 1.019 1.042–1.381 <0.001 Rheumatic aetiology 5.378 3.211–9.006 <0.001 MV regurgitation degree 1.561 1.290–1.888 0.001 TV regurgitation degree 1.200 1.042–1.381 0.012 Surgeona 0.523 0.358–0.764 0.001 Surgical centreb 0.005 0.002–0.130 <0.001 Risk factors OR 95% CI P-value Previous MV surgery 4.776 2.942–0.7753 <0.001 COPD 2.316 1.258–4.266 0.007 sPAP 1.019 1.042–1.381 <0.001 Rheumatic aetiology 5.378 3.211–9.006 <0.001 MV regurgitation degree 1.561 1.290–1.888 0.001 TV regurgitation degree 1.200 1.042–1.381 0.012 Surgeona 0.523 0.358–0.764 0.001 Surgical centreb 0.005 0.002–0.130 <0.001 The Hosmer–Lemeshow goodness-of-fit χ2 value for this model was 8.0 (p = 0.307). C-statistic: 0.787 (95% CI 0.747–0.827). a The variable surgeon corresponded to the most experienced surgeon (in the number of surgeries performed and the rate of MV repair) in opposition to the remaining. b Coimbra centre. CI: confidence interval; COPD: chronic obstructive pulmonary disease; MV: mitral valve; OR: odds ratio; sPAP: systolic pulmonary artery pressure; TV: tricuspid valve. Table 3: Risk factors for mitral valve replacement (logistic regression) Risk factors OR 95% CI P-value Previous MV surgery 4.776 2.942–0.7753 <0.001 COPD 2.316 1.258–4.266 0.007 sPAP 1.019 1.042–1.381 <0.001 Rheumatic aetiology 5.378 3.211–9.006 <0.001 MV regurgitation degree 1.561 1.290–1.888 0.001 TV regurgitation degree 1.200 1.042–1.381 0.012 Surgeona 0.523 0.358–0.764 0.001 Surgical centreb 0.005 0.002–0.130 <0.001 Risk factors OR 95% CI P-value Previous MV surgery 4.776 2.942–0.7753 <0.001 COPD 2.316 1.258–4.266 0.007 sPAP 1.019 1.042–1.381 <0.001 Rheumatic aetiology 5.378 3.211–9.006 <0.001 MV regurgitation degree 1.561 1.290–1.888 0.001 TV regurgitation degree 1.200 1.042–1.381 0.012 Surgeona 0.523 0.358–0.764 0.001 Surgical centreb 0.005 0.002–0.130 <0.001 The Hosmer–Lemeshow goodness-of-fit χ2 value for this model was 8.0 (p = 0.307). C-statistic: 0.787 (95% CI 0.747–0.827). a The variable surgeon corresponded to the most experienced surgeon (in the number of surgeries performed and the rate of MV repair) in opposition to the remaining. b Coimbra centre. CI: confidence interval; COPD: chronic obstructive pulmonary disease; MV: mitral valve; OR: odds ratio; sPAP: systolic pulmonary artery pressure; TV: tricuspid valve. Early mortality and late survival The 30-day mortality rate for the entire study population was 2.4% (27 patients), significantly higher in the MV replacement group [4.2% vs 1.8%; odds ratio (OR) 2.4, 95% CI 1.1–5.2; P = 0.021] and confirmed in the propensity-matched group (4.7% vs 1.7%; OR 2.8, 95% CI 0.9–9.9; P = 0.06). Interestingly, in patients with rheumatic MV disease, the 30-day mortality rate was not significantly different (2.4% vs 1.5%; OR 1.6, 95% CI 0.5–5.1; P = 0.41). During the follow-up period, 388 patients died (35.4% of patients who survived surgery). The unadjusted survival rate at 4, 8 and 12 years was superior in the MV repair group (85.0 ± 1.3%, 74.1 ± 1.7% and 61.7 ± 2.0%) compared with that in the MV replacement group (81.7 ± 2.5%, 71.9 ± 3.2% and 53.3 ± 4.5%; P = 0.026) (Fig. 1A). This effect persisted in the propensity-matched population (Fig. 1B), where survival at 12 years was also better in the repair group, almost reaching statistical significance (63.3 ± 3.8% vs 54.6 ± 4.9%, P = 0.062). In the subgroup of patients with rheumatic MV disease (Fig. 2), the 12-year survival rate was also significantly better in patients undergoing repair (68.0 ± 2.5% vs 57.9 ± 4.8%; P = 0.018). These results were replicated in both centres when each was evaluated individually. Figure 1: View largeDownload slide Comparison of late survival of patients who had mitral valve repair versus replacement. (A) Non-adjusted survival (overall population) and (B) propensity-matched population. Figure 1: View largeDownload slide Comparison of late survival of patients who had mitral valve repair versus replacement. (A) Non-adjusted survival (overall population) and (B) propensity-matched population. Figure 2: View largeDownload slide Comparison of late survival of patients with rheumatic mitral valve disease who had mitral valve repair versus replacement. Figure 2: View largeDownload slide Comparison of late survival of patients with rheumatic mitral valve disease who had mitral valve repair versus replacement. Numerous factors were identified as risk factors for late mortality in the univariable analysis (P < 0.05): MV replacement, mechanical mitral and aortic prostheses, higher New York Heart Association (NYHA) class, CAD, hypertension, obesity, atrial fibrillation, CABG, age, low ejection fraction, higher degrees of tricuspid valve regurgitation, left atrium and ventricle (systolic) dilation and higher systolic pulmonary artery pressure. However, in the multivariable Cox analysis, only age, MV replacement, higher NYHA class, CAD and mechanical aortic prosthesis were considered independent risk factors for mortality (Table 4). Table 4: Risk factors for late mortality (Cox regression) Risk factors HR 95% CI P-value Mitral valve replacement 1.35 1.04–1.75 0.022 Age (per years) 1.03 1.02–1.04 <0.001 NYHA class 1.26 1.04–1.52 0.018 Aortic bioprosthesis 1.32 1.01–1.72 0.041 Coronary artery disease 1.97 1.47–2.64 <0.001 Risk factors HR 95% CI P-value Mitral valve replacement 1.35 1.04–1.75 0.022 Age (per years) 1.03 1.02–1.04 <0.001 NYHA class 1.26 1.04–1.52 0.018 Aortic bioprosthesis 1.32 1.01–1.72 0.041 Coronary artery disease 1.97 1.47–2.64 <0.001 CI: confidence interval; HR: hazard ratio; NYHA: New York Heart Association. Table 4: Risk factors for late mortality (Cox regression) Risk factors HR 95% CI P-value Mitral valve replacement 1.35 1.04–1.75 0.022 Age (per years) 1.03 1.02–1.04 <0.001 NYHA class 1.26 1.04–1.52 0.018 Aortic bioprosthesis 1.32 1.01–1.72 0.041 Coronary artery disease 1.97 1.47–2.64 <0.001 Risk factors HR 95% CI P-value Mitral valve replacement 1.35 1.04–1.75 0.022 Age (per years) 1.03 1.02–1.04 <0.001 NYHA class 1.26 1.04–1.52 0.018 Aortic bioprosthesis 1.32 1.01–1.72 0.041 Coronary artery disease 1.97 1.47–2.64 <0.001 CI: confidence interval; HR: hazard ratio; NYHA: New York Heart Association. Freedom from mitral valve reoperation There were 58 reoperations to the MV during the study period for a linearized incidence of 0.64%/patient-year. The mean interval to reoperation was 8.1 ± 6.5 years (median 6.9 years; 25th–75th percentile 2.4–12.6 years). There were 17 reoperations in the MV replacement group (6.0%) and 41 in the MV repair group (4.9%, P = 0.482). The causes of reoperation are listed in Table 5, progression of the MV disease and periprosthetic leakage being the most frequent causes in MV repair and MV replacement, respectively. Freedom from MV reoperation at 6 and 12 years was superior in the MV repair group (98.5 ± 0.5% and 94.7 ± 1.1% vs 94.1 ± 1.8% and 89.0 ± 3.1%; P = 0.004) (Fig. 3A and B); this advantage was also observed in the propensity-matched population (P = 0.018). Table 5: Causes of reoperation Causes of reoperation* n (%) MV repair  Technical failure 3 (7.4)   Ring dehiscence 2 (4.9)   Suture dehiscence 1 (2.4)  Disease progression 38 (92.6)   Fibrosis/leaflet retraction 32 (30.5)   New prolapses (rupture of native chordae) 6 (39.1)  Total 41 MV replacement  Leak (non-infectious) 10 (58.8)  Leak (infectious) 3 (17.6)  Pannus 4 (23.5)  Total 17 Causes of reoperation* n (%) MV repair  Technical failure 3 (7.4)   Ring dehiscence 2 (4.9)   Suture dehiscence 1 (2.4)  Disease progression 38 (92.6)   Fibrosis/leaflet retraction 32 (30.5)   New prolapses (rupture of native chordae) 6 (39.1)  Total 41 MV replacement  Leak (non-infectious) 10 (58.8)  Leak (infectious) 3 (17.6)  Pannus 4 (23.5)  Total 17 *Patients could have more than one cause of reoperation. MV: mitral valve. Table 5: Causes of reoperation Causes of reoperation* n (%) MV repair  Technical failure 3 (7.4)   Ring dehiscence 2 (4.9)   Suture dehiscence 1 (2.4)  Disease progression 38 (92.6)   Fibrosis/leaflet retraction 32 (30.5)   New prolapses (rupture of native chordae) 6 (39.1)  Total 41 MV replacement  Leak (non-infectious) 10 (58.8)  Leak (infectious) 3 (17.6)  Pannus 4 (23.5)  Total 17 Causes of reoperation* n (%) MV repair  Technical failure 3 (7.4)   Ring dehiscence 2 (4.9)   Suture dehiscence 1 (2.4)  Disease progression 38 (92.6)   Fibrosis/leaflet retraction 32 (30.5)   New prolapses (rupture of native chordae) 6 (39.1)  Total 41 MV replacement  Leak (non-infectious) 10 (58.8)  Leak (infectious) 3 (17.6)  Pannus 4 (23.5)  Total 17 *Patients could have more than one cause of reoperation. MV: mitral valve. Figure 3: View largeDownload slide Freedom from mitral valve reoperation and comparison of patients who had mitral valve repair versus replacement. (A) Overall population, (B) propensity-matched population and (C) rheumatic mitral valve disease subgroup. Figure 3: View largeDownload slide Freedom from mitral valve reoperation and comparison of patients who had mitral valve repair versus replacement. (A) Overall population, (B) propensity-matched population and (C) rheumatic mitral valve disease subgroup. Patients with degenerative MV disease had the most satisfactory results in this aspect: freedom from MV reoperation at 12 years was 97.7 ± 1.3% in contrast with those who had rheumatic disease (92.7 ± 1.3%, P = 0.038). Noteworthy, freedom from reoperation was similar between groups in rheumatic patients (91.4 ± 2.7% in MV repair vs 93.5 ± 1.5% in MV replacement at 12 years; P = 0.12) (Fig. 3C). In 6 cases of reoperation (5 in the MV replacement group and 1 in MV repair group), MV surgery was not the primary indication for surgery. Those cases included pannus obstruction, periprosthetic leakage and mitral restenosis. After exclusion of these patients, freedom from MV reoperation remained higher in the MV repair group, almost reaching statistical significance (P = 0.06). Moreover, in the propensity-matched group and in the rheumatic MV subgroup, the results remained the same. DISCUSSION The decision to repair or replace the MV in patients undergoing AVR is a challenge yet to be resolved. We believe that our data shed some light on this subject. Firstly, mitroaortic valve surgery, with or without concomitant procedures and including redo cases, can be performed with low mortality (2.4%). Secondly, it is possible to perform MV repair in most cases, even in those with rheumatic disease. Thirdly, there appears to be a survival advantage in repairing the MV in these circumstances, and this was also demonstrated in a propensity-matched population and validated in patients with rheumatic MV disease. Finally, there was no difference in the reoperation rates after repair or replacement of the MV. Which mitral valves can be repaired? MV repair for the treatment of isolated severe MR has been established as the gold standard of surgical care in patients with degenerative MV disease, particularly in cases of MV prolapse. This has been demonstrated even in complex cases, such as bileaflet prolapse and Barlow’s disease [16]. In contrast, the reported rate of MV repair during AVR is widely variable, ranging from 12% to 80% [10, 11, 14, 18, 19]. In a recent report from the Society of Thoracic Surgeons Adult Cardiac Surgery Database (STS-ACSD), which includes 23 404 patients operated on from 1993 to 2007, MV repair was performed in 46% of the patients and replacement in 54% of the patients during AVR [20]. The authors found that MV repair was associated with an independent lower risk of operative mortality (OR 0.61), which improved over time. Despite not having evaluated other important outcomes, such as survival, event-free survival and MV reoperation or MR recurrence, they claimed that during mitroaortic surgery, MV repair should remain the preferred treatment whenever feasible. As expected, rheumatic MV disease was identified as the most powerful independent predictor of MV replacement (5-fold risk) in our population, although we were able to repair 77.6% of rheumatic valves (excluding patients with previous MV surgery). Of note, the high rate of repair achieved in patients with degenerative MV disease (94.5%) in one of our two centres will be a probable scenario in the coming years, because patients with rheumatic valves are slowly disappearing in Western countries. Naturally, not all valves are amenable to repair, particularly those with heavily calcified leaflets or annulus, or with extensive rheumatic involvement (severely thickened, immobile and retracted leaflets and chordae). Thirty-day mortality and long-term survival Concomitant aortic and MV surgery carries an additional risk compared with isolated valve surgery. Data from the STS-ACSD revealed an unadjusted operative mortality rate of 8.2% when the MV was repaired and 11.6% when it was replaced (P < 0.0001). Furthermore, a report from the Northern New England Cardiovascular Disease Study Group showed an even higher in-hospital mortality rate of 15.5% (11.0% for patients <70 years, 18.0% for those 70–79 years and 24% for those >80 years) [21]. However, we and others have clearly demonstrated that, with experience, it is possible to have a significantly lower 30-day mortality rate (1–3%) even when other procedures are performed concomitantly [11, 14]. In all experiences, double valve replacement is associated with a greater risk of early death, probably related to perioperative complications, such as rupture of the atrioventricular sulcus, obstruction of the left outflow tract and low cardiac output. In our study, MV replacement was also associated with a significantly higher 30-day mortality rate (4.2% vs 1.8%, P = 0.021), and this was also observed in the propensity-matched population (4.7% vs 1.7%), almost reaching statistical significance (P = 0.06). This finding was also described in a recent meta-analysis of observational studies, evaluating the outcomes of MV repair compared with MV replacement in this setting [9]. Several risk factors have been identified, such as age, associated CABG and renal failure with dialysis, and they should also be taken into account when deciding which MV procedure to use. Beyond the immediate results, the impact of the surgical option on the long-term outcomes, particularly on late survival, is the focus of intense debate. There are several reasons for the heterogeneity of the results reported [11, 12, 22]. First, the populations among the studies differ with regard to age, patient characteristics, inclusion (or not) of other procedures and the aetiology of the MV disease. Second, almost all series published have significantly more patients in the replacement group than in the repair group. This difference is especially worrisome when evaluating the subgroup of patients with rheumatic MV disease, where the ‘repair arm’ is scarce, which raises concerns about the interpretation of the results. We believe that our study is singular because the majority of patients had MV repair, even the patients with rheumatic disease. Finally, the length of follow-up varied with different studies. In our series, late survival was significantly better after MV repair in the overall population, in the propensity-matched patients and in the subgroup of patients with rheumatic MV disease. Conversely, MV replacement was independently associated with long-term mortality—a fact also demonstrated in the meta-analysis performed by Saurav et al. [9]. CAD, age, mechanical aortic prosthesis and higher NYHA class were identified as predictors of late mortality. Turina et al. [23] found that lower ejection fraction and tricuspid valve surgery were also risk factors for impaired survival in this context. The latter factors and the presence of symptoms are characteristic of a more advanced stage of the disease, which may indicate that patients should be sent to surgery earlier. The majority of these factors was more prevalent in the MV replacement group, which can further worsen their prognosis. Mitral valve reoperation Surprisingly, freedom from reoperation to the MV was higher in the repair group (overall and propensity score matching population), which is in contradiction with the findings of the majority of studies [11, 18, 19]. Additionally, we did not find any difference in the rheumatic population, which is known to be associated with lesser durability of repair. Nonetheless, some caution should be exerted when drawing conclusions because in six patients submitted to redo MV surgery the primary indication for surgery was not the MV. After excluding these patients, freedom from MV reoperation remained higher in the MV repair group, although not significantly so. However, this finding did not change the results in the propensity score matching and rheumatic patients. Ho et al. [22] evaluated 609 patients with rheumatic heart disease who underwent AVR with either MV repair (n = 201) or MV replacement (n = 408). They also did not find any difference with regard to MV reoperation. Our results may be attributed to our extensive experience in repairing rheumatic MVs (over a thousand patients), which may not be reproducible by low-volume centres. In our experience, patients usually return to surgery on average 12–17 years after the initial operation, which means that these results would surely be different after a 20-year follow-up period. On the contrary, in degenerative disease this difference probably will not dissipate over time [24]. Limitations The lack of randomization is always prone to the appearance of unmeasured confounding factors and selection bias. We have tried to minimize these factors by performing the propensity score matching (232 patients per group) and multivariable analyses. The inclusion of 2 centres with distinct approaches to MV disease, 1 more ‘aggressive’ in repairing all kinds of MV aetiologies and the other keener on replacement, can lead to some heterogeneity of the data. However, we believe that the evaluation of the impact of the surgical choice (repair versus replacement) is not impaired by these 2 distinct strategies. Repairing rheumatic MV is more demanding and surgeon-dependent, and so the issue of the generalizability of the results should also be raised in this setting. Nevertheless, these patients also benefited from having their valves repaired, and so surgeons should strive to repair those valves, and these results should be regarded as an incentive. CONCLUSIONS This study confirms that, whenever feasible, MV repair is the procedure of choice in patients undergoing simultaneous AVR because it is associated with lower early and late mortality rates when compared with replacement and with similar rates of reoperation. We have also demonstrated that MV repair can be performed in the majority of cases. Therefore, strategies to improve MV repair rates in this setting should be pursued in the future. SUPPLEMENTARY MATERIAL Supplementary material is available at EJCTS online. Conflict of interest: none declared. REFERENCES 1 Iung B , Baron G , Butchart E , Delahaye F , Gohlke-Bärwolf C , W , Levang O et al. A prospective survey of patients with valvular heart disease in Europe: the Euro Heart Survey on Valvular Heart Disease . Eur Heart J 2003 ; 24 : 1231 – 43 . Google Scholar CrossRef Search ADS PubMed 2 Nowicki ER , Birkmeyer NJ , Weintraub RW , Leavitt BJ , Sanders JH , Dacey LJ et al. Multivariable prediction of in-hospital mortality associated with aortic and mitral valve surgery in Northern New England . Ann Thorac Surg 2004 ; 77 : 1966 – 77 . Google Scholar CrossRef Search ADS PubMed 3 Birkmeyer NJ , Marrin CA , Morton JR , Leavitt BJ , Lahey SJ , Charlesworth DC et al. Decreasing mortality for aortic and mitral valve surgery in Northern New England. Northern New England cardiovascular Disease Study Group . Ann Thorac Surg 2000 ; 70 : 432 – 7 . Google Scholar CrossRef Search ADS PubMed 4 Tribouilloy C , Enriquez-Sarano M , Schaff H , Orszulak TA , Bailey KR , Tajik AJ et al. Impact of preoperative symptoms on survival after surgical correction of organic mitral regurgitation: rationale for optimizing surgical indications . Circulation 1999 ; 99 : 400 – 5 . Google Scholar CrossRef Search ADS PubMed 5 Suri RM , Schaff H , Dearani JA , Sundt TM , Daly RC , Mullany CJ et al. Survival advantage and improved durability of mitral repair for leaflet prolapse subsets in the current era . Ann Thorac Surg 2006 ; 82 : 819 – 26 . Google Scholar CrossRef Search ADS PubMed 6 Gillinov AM , Blackstone EH , Nowicki ER , Slisatkorn W , Al-Dossari G , Johnston DR et al. Valve repair versus valve replacement for degenerative mitral valve disease . J Thorac Cardiovasc Surg 2008 ; 135 : 885 – 93 . Google Scholar CrossRef Search ADS PubMed 7 Falk V , Baumgartner H , Bax JJ , De Bonis M , Hamm C , Holm PJ et al. 2017 ESC/EACTS guidelines for the management of valvular heart disease: the Task Force for the Management of Valvular Heart Disease of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS) . Eur J Cardiothorac Surg 2017 ; 52 : 616 – 64 . Google Scholar CrossRef Search ADS PubMed 8 Nishimura RA , Otto CM , Bonow RO , Carabello BA , Erwin JP , Guyton R et al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines . J Am Coll Cardiol 2014 ; 63 : e57 – 185 . Google Scholar CrossRef Search ADS PubMed 9 Saurav A , Alla VM , Kaushik M , Hunter CC , Mooss AV. Outcomes of mitral valve repair compared with replacement in patients undergoing concomitant aortic valve surgery: a meta-analysis of observational studies . Eur J Cardiothorac Surg 2015 ; 48 : 347 – 53 . Google Scholar CrossRef Search ADS PubMed 10 Urban M , Pirk J , Szarszoi O , Skalsky I , Maly J , Netuka I. Mitral valve repair versus replacement in simultaneous aortic and mitral valve surgery . Expert Clin Cardiol 2013 ; 18 : 22 – 6 . 11 Hamamoto M , Bando K , Kobayashi J , Satoh T , Sasako Y , Niwaya K et al. Durability and outcome of aortic valve replacement with mitral valve repair versus double valve replacement . Ann Thorac Surg 2003 ; 75 : 28 – 33 . Google Scholar CrossRef Search ADS PubMed 12 Talwar S , Mathur A , Choudhary SK , Singh R , Kumar AS. Aortic valve replacement with mitral valve repair compared with combined aortic and mitral valve replacement . Ann Thorac Surg 2007 ; 84 : 1219 – 25 . Google Scholar CrossRef Search ADS PubMed 13 Akins CW , Miller DC , Turina MI , Kouchoukos NT , Blackstone EH , Grunkemeier GL et al. Guidelines for reporting mortality and morbidity after cardiac valve interventions . Eur J Cardiothorac Surg 2008 ; 33 : 523 – 8 . Google Scholar CrossRef Search ADS PubMed 14 Coutinho GF , Correia PM , Antunes MJ. Concomitant aortic and mitral surgery: to replace or repair the mitral valve? J Thorac Cardiovasc Surg 2014 ; 148 : 1386 – 92 . Google Scholar CrossRef Search ADS PubMed 15 Lang RM , Bierig M , Devereux RB , Flachskampf FA , Foster E , Pellikka PA et al. Recommendations for chamber quantification . Eur J Echocardiogr 2006 ; 7 : 79 – 108 . Google Scholar CrossRef Search ADS PubMed 16 Coutinho GF , Correia PM , Branco C , Antunes MJ. Long-term results of mitral valve surgery for degenerative anterior leaflet or bileaflet prolapse: analysis of negative factors for repair, early and late failures, and survival . Eur J Cardiothorac Surg 2016 ; 50 : 66 – 74 . Google Scholar CrossRef Search ADS PubMed 17 Hickey GL , Dunning J , Seifert B , Sodeck G , Carr MJ , Burger HU et al. Statistical and data reporting guidelines for the European Journal of Cardio-Thoracic Surgery and the Interactive CardioVascular and Thoracic Surgery . Eur J Cardiothorac Surg 2015 ; 48 : 180 – 93 . Google Scholar CrossRef Search ADS PubMed 18 Gillinov AM , Blackstone EH , Cosgrove DM , White J , Kerr P , Marullo A et al. Mitral valve repair with aortic valve replacement is superior to double valve replacement . J Thorac Cardiovasc Surg 2003 ; 125 : 1372 – 85 . Google Scholar CrossRef Search ADS PubMed 19 Kuwaki K , Kawaharada N , Morishita K , Koyanagi T , Osawa H , Maeda T et al. Mitral valve repair versus replacement in simultaneous mitral and aortic valve surgery for rheumatic disease . Ann Thorac Surg 2007 ; 83 : 558 – 63 . Google Scholar CrossRef Search ADS PubMed 20 Thourani VH , Suri RM , Rankin JS , He X , O'Brien SM , Badhwar V et al. Does mitral valve repair offer an advantage over replacement in patients undergoing aortic valve replacement? Ann Thorac Surg 2014 ; 98 : 598 – 7 . Google Scholar CrossRef Search ADS PubMed 21 Leavitt BJ , Baribeau YR , DiScipio AW , Ross CS , Quinn RD , Olmstead EM et al. ; Northern New England Cardiovascular Disease Study Group . Outcomes of patients undergoing concomitant aortic and mitral valve surgery in northern New England . Circulation 2009 ; 120 : S155 – 62 . Google Scholar CrossRef Search ADS PubMed 22 Ho H-QT , Nguyen V-P , Phan K-P , Pham N-V. Mitral valve repair with aortic valve replacement in rheumatic heart disease . Asian Cardiovasc Thorac Ann 2004 ; 12 : 341 – 5 . Google Scholar CrossRef Search ADS PubMed 23 Turina J , Stark T , Seifert B , Turina M. Predictors of the long-term outcome after combined aortic and mitral valve surgery . Circulation 1999 ; 100 : II48 – 53 . Google Scholar CrossRef Search ADS PubMed 24 Coutinho GF , Antunes MJ. Mitral valve repair for degenerative mitral valve disease: surgical approach, patient selection and long-term outcomes . Heart 2017 ; 103 : 1663 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Journal of Cardio-Thoracic Surgery Oxford University Press

Long-term results after concomitant mitral and aortic valve surgery: repair or replacement?

Loading next page...
 
/lp/ou_press/long-term-results-after-concomitant-mitral-and-aortic-valve-surgery-XHKla9uHu0
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
ISSN
1010-7940
eISSN
1873-734X
D.O.I.
10.1093/ejcts/ezy205
Publisher site
See Article on Publisher Site

Abstract

Abstract OBJECTIVES The reported superiority of mitral valve (MV) repair for isolated MV regurgitation has not been confirmed in mitroaortic valve surgery. Our goals were to evaluate the feasibility of repair in patients undergoing mitral and aortic valve surgery and to identify factors predisposing to MV replacement, to compare long-term outcomes (survival and MV reoperation) of repair and replacement and to perform a subgroup analysis in patients with rheumatic MV disease. METHODS From January 1992 through December 2016, 1122 consecutive patients were submitted to concomitant aortic and MV surgery in 2 different centres (Coimbra and Santiago). Of these, 837 patients underwent MV repair (74.6%) and 285 patients had MV replacement (25.4%). Rheumatic aetiology was predominant (666 patients; 59.4%). Cumulative follow-up was 9522.6 patient-years (25th–75th percentile 2.6–13.2 years) and was complete for 95.6% of patients. Propensity score matching (1:1) was performed in 232 patients for comparing each treatment option (MV repair and MV replacement). RESULTS Previous MV intervention, rheumatic aetiology, chronic obstructive pulmonary disease, higher degrees of tricuspid and mitral regurgitation and pulmonary hypertension were independently correlated with MV replacement. The 30-day mortality rate was higher in patients with MV replacement (4.2% vs 1.8%, P = 0.021) and was confirmed in the propensity score matching (4.7% vs 1.7%, P = 0.06). Late survival was lower in the MV replacement group (53.3 ± 4.5% vs 61.7 ± 2.0% at 12 years; P = 0.026) and was confirmed in the propensity score matching (54.6 ± 4.9% vs 63.2 ± 3.8%, P = 0.062) and rheumatic subgroup (57.9 ± 4.8% vs 68.0 ± 2.5%, P = 0.018). Freedom from MV reoperation at 12 years was higher in the MV repair group (94.7 ± 1.1% vs 89.0 ± 3.1%, P = 0.004) but similar in patients with rheumatic MV disease. CONCLUSIONS MV repair can be performed in most patients undergoing aortic valve replacement. It should be the procedure of choice whenever feasible, because it is associated with lower early and late mortality rates and with freedom from reoperation in non-rheumatic patients. View largeDownload slide View largeDownload slide Aortic valve replacement, Mitral valve repair, Mitral valve replacement, Long-term survival, Freedom from reoperation INTRODUCTION Aortic valve stenosis and mitral valve regurgitation (MR) have become the most frequent valvular diseases in Western countries in recent decades [1]. Increasing life expectancy allied with improved surgical outcomes has contributed to a growing need for concomitant aortic and mitral valve (MV) surgery [2, 3]. However, the potential benefit of MV repair over replacement in patients having double valve surgery remains uncertain. Theoretically, double valve replacement would be the likely procedure, because there would be no advantage in repairing a valve when another required replacement. Nevertheless, the excellent results of isolated MV repair in the correction of severe degenerative MR with regard to hospital mortality, long-term survival, preservation of left ventricular function and freedom from valve-related complications (thromboembolism, anticoagulant-related bleeding and endocarditis) [4–6] have encouraged its application in cases of aortic valve replacement (AVR). The lack of convincing data on the use of MV repair in combined or multiple-valve disease precludes one from making evidence-based recommendations. The major international guidelines for the management of valvular heart disease (European Society of Cardiology/European Association for Cardio-Thoracic Surgery and American College of Cardiology/American Heart Association) [7, 8] are vague in this regard, partially due to the absence of randomized controlled trials and the scarcity of observational studies. Although the current trend is to favour repair in patients with non-rheumatic MV disease [9], there is considerable controversy surrounding MV repair in patients with rheumatic disease. Leaflet fibrosis, subvalvular thickening and retraction and commissural fusion all contribute to lower probability and durability of the repair. Poorer outcomes have driven surgeons away from attempting repair in this context, which poses additional difficulties in interpreting surgical results, as there are even fewer studies addressing this subject [10–12]. Therefore, our goals were to (i) evaluate the feasibility of repair in patients undergoing mitral and aortic valve surgery and to identify factors predisposing to replacement, (ii) compare long-term outcomes (survival and MV reoperation) of repair and replacement in this context and (iii) perform a subgroup analysis in patients with rheumatic MV disease. MATERIALS AND METHODS Patient population From January 1992 through December 2016, 1122 patients were consecutively submitted to concomitant aortic (replacement only) and MV surgery (repair or replacement) in 2 different centres [university hospitals of Coimbra, Portugal (n = 1039) and Santiago de Compostela, Spain (n = 83)]. Of these, 837 patients underwent MV repair (74.6%) and 285 patients had MV replacement (25.4%). Patients with additional procedures, such as coronary artery bypass graft (CABG), tricuspid valve surgery and ascending aorta surgery, were admitted to the study. Patients with previous MV interventions other than MV replacement (percutaneous mitral balloon valvuloplasty, open or closed commissurotomy and other MV valvuloplasty) were also included. The mean age of the population was 63.0 ± 12.8 years (25th–75th percentile 53–70). Patients who had MV replacement were significantly older (63.6 ± 10.7 vs 59.7 ± 13.4 years of age, P < 0.001), more symptomatic, had more comorbidities and a greater prevalence of tricuspid valve disease. Patient characteristics are listed in Table 1. Rheumatic disease was the most frequent aetiology of MV disease, present in 666 patients (59.4%), followed by degenerative disease in 326 (29%) and secondary (functional) disease in 59 (5.3%). All patients from both centres had moderate to severe or severe MV disease, either as stenosis, regurgitation or mixed lesion. Table 1: Demographic and echocardiographic data Variables Overall (n = 1122) Mitral valve repair (n = 837) Mitral valve replacement (n = 285) P-value Age (years), mean ± SD 60.6 ± 12.8 59.7 ± 13.4 63.3 ± 10.7 <0.001 Male sex, n (%) 558 (49.7) 428 (51.1) 130 (45.6) 0.10 Body surface area, mean ± SD 1.66 ± 0.18 1.66 ± 0.17 1.67 ± 0.19 0.46 Aetiology of MV disease, n (%)  Degenerative 326 (29.0) 272 (32.5) 54 (18.9) <0.001  Rheumatic 666 (59.4) 461 (55.1) 205 (71.9) <0.001  Secondary 59 (5.3) 58 (6.9) 1 (0.4) <0.001  Others 71 (6.3) 46 (5.5) 25 (8.8) 0.069 MV pathology, n (%)  Stenosis 310 (27.6) 240 (28.7) 70 (24.6) 0.18  Regurgitation 547 (48.8) 451 (53.9) 96 (33.7) <0.001  Mixed lesion 265 (23.6) 146 (17.4) 119 (41.8) <0.001 Previous MV surgery, n (%) 127 (11.3) 49 (5.9) 78 (27.4) <0.001 Tricuspid valve disease, n (%) 233 (20.8) 145 (17.3) 88 (30.9) <0.001 Coronary artery disease, n (%) 122 (10.9) 106 (12.7) 16 (5.6) 0.001 NYHA III–IV, n (%) 731 (65.2) 525 (62.7) 206 (72.3) 0.003 Stroke, n (%) 64 (5.7) 45 (5.4) 19 (6.7) 0.41 Hypertension, n (%) 321 (28.6) 216 (25.8) 105 (36.8) <0.001 Diabetes, n (%) 101 (9.0) 59 (7.1) 42 (14.8) <0.001 Chronic kidney disease, n (%) 31 (2.8) 23 (2.8) 8 (2.8) 0.96 COPD, n (%) 85 (7.6) 49 (5.9) 36 (12.6) <0.001 Atrial fibrillation/flutter, n (%) 441 (39.3) 313 (37.4) 128 (44.9) 0.025 Ejection fraction (%), mean ± SD 61.2 ± 10.6 61.0 ± 10.7 61.6 ± 10.0 0.43 LV end-systolic dimension (mm), mean ± SD 41.8 ± 10.7 43.3 ± 10.8 36.8 ± 8.3 <0.001 Left atrial diameter (mm), mean ± SD 54.5 ± 10.5 53.3 ± 10.6 57.9 ± 9.6 <0.001 Mean aortic gradient (mmHg), mean ± SD 48.3 ± 22.2 49.1 ± 22.4 46.2 ± 21.5 0.22 sPAP, mean ± SD 53.8 ± 17.3 51.7 ± 16.9 55.9 ± 18.2 0.002 Variables Overall (n = 1122) Mitral valve repair (n = 837) Mitral valve replacement (n = 285) P-value Age (years), mean ± SD 60.6 ± 12.8 59.7 ± 13.4 63.3 ± 10.7 <0.001 Male sex, n (%) 558 (49.7) 428 (51.1) 130 (45.6) 0.10 Body surface area, mean ± SD 1.66 ± 0.18 1.66 ± 0.17 1.67 ± 0.19 0.46 Aetiology of MV disease, n (%)  Degenerative 326 (29.0) 272 (32.5) 54 (18.9) <0.001  Rheumatic 666 (59.4) 461 (55.1) 205 (71.9) <0.001  Secondary 59 (5.3) 58 (6.9) 1 (0.4) <0.001  Others 71 (6.3) 46 (5.5) 25 (8.8) 0.069 MV pathology, n (%)  Stenosis 310 (27.6) 240 (28.7) 70 (24.6) 0.18  Regurgitation 547 (48.8) 451 (53.9) 96 (33.7) <0.001  Mixed lesion 265 (23.6) 146 (17.4) 119 (41.8) <0.001 Previous MV surgery, n (%) 127 (11.3) 49 (5.9) 78 (27.4) <0.001 Tricuspid valve disease, n (%) 233 (20.8) 145 (17.3) 88 (30.9) <0.001 Coronary artery disease, n (%) 122 (10.9) 106 (12.7) 16 (5.6) 0.001 NYHA III–IV, n (%) 731 (65.2) 525 (62.7) 206 (72.3) 0.003 Stroke, n (%) 64 (5.7) 45 (5.4) 19 (6.7) 0.41 Hypertension, n (%) 321 (28.6) 216 (25.8) 105 (36.8) <0.001 Diabetes, n (%) 101 (9.0) 59 (7.1) 42 (14.8) <0.001 Chronic kidney disease, n (%) 31 (2.8) 23 (2.8) 8 (2.8) 0.96 COPD, n (%) 85 (7.6) 49 (5.9) 36 (12.6) <0.001 Atrial fibrillation/flutter, n (%) 441 (39.3) 313 (37.4) 128 (44.9) 0.025 Ejection fraction (%), mean ± SD 61.2 ± 10.6 61.0 ± 10.7 61.6 ± 10.0 0.43 LV end-systolic dimension (mm), mean ± SD 41.8 ± 10.7 43.3 ± 10.8 36.8 ± 8.3 <0.001 Left atrial diameter (mm), mean ± SD 54.5 ± 10.5 53.3 ± 10.6 57.9 ± 9.6 <0.001 Mean aortic gradient (mmHg), mean ± SD 48.3 ± 22.2 49.1 ± 22.4 46.2 ± 21.5 0.22 sPAP, mean ± SD 53.8 ± 17.3 51.7 ± 16.9 55.9 ± 18.2 0.002 COPD: chronic obstructive pulmonary disease; LV: left ventricle; MV: mitral valve; NYHA: New York Heart Association; SD: standard deviation; sPAP: systolic pulmonary artery pressure. Statistically significant P-values (<0.05) are in bold. Table 1: Demographic and echocardiographic data Variables Overall (n = 1122) Mitral valve repair (n = 837) Mitral valve replacement (n = 285) P-value Age (years), mean ± SD 60.6 ± 12.8 59.7 ± 13.4 63.3 ± 10.7 <0.001 Male sex, n (%) 558 (49.7) 428 (51.1) 130 (45.6) 0.10 Body surface area, mean ± SD 1.66 ± 0.18 1.66 ± 0.17 1.67 ± 0.19 0.46 Aetiology of MV disease, n (%)  Degenerative 326 (29.0) 272 (32.5) 54 (18.9) <0.001  Rheumatic 666 (59.4) 461 (55.1) 205 (71.9) <0.001  Secondary 59 (5.3) 58 (6.9) 1 (0.4) <0.001  Others 71 (6.3) 46 (5.5) 25 (8.8) 0.069 MV pathology, n (%)  Stenosis 310 (27.6) 240 (28.7) 70 (24.6) 0.18  Regurgitation 547 (48.8) 451 (53.9) 96 (33.7) <0.001  Mixed lesion 265 (23.6) 146 (17.4) 119 (41.8) <0.001 Previous MV surgery, n (%) 127 (11.3) 49 (5.9) 78 (27.4) <0.001 Tricuspid valve disease, n (%) 233 (20.8) 145 (17.3) 88 (30.9) <0.001 Coronary artery disease, n (%) 122 (10.9) 106 (12.7) 16 (5.6) 0.001 NYHA III–IV, n (%) 731 (65.2) 525 (62.7) 206 (72.3) 0.003 Stroke, n (%) 64 (5.7) 45 (5.4) 19 (6.7) 0.41 Hypertension, n (%) 321 (28.6) 216 (25.8) 105 (36.8) <0.001 Diabetes, n (%) 101 (9.0) 59 (7.1) 42 (14.8) <0.001 Chronic kidney disease, n (%) 31 (2.8) 23 (2.8) 8 (2.8) 0.96 COPD, n (%) 85 (7.6) 49 (5.9) 36 (12.6) <0.001 Atrial fibrillation/flutter, n (%) 441 (39.3) 313 (37.4) 128 (44.9) 0.025 Ejection fraction (%), mean ± SD 61.2 ± 10.6 61.0 ± 10.7 61.6 ± 10.0 0.43 LV end-systolic dimension (mm), mean ± SD 41.8 ± 10.7 43.3 ± 10.8 36.8 ± 8.3 <0.001 Left atrial diameter (mm), mean ± SD 54.5 ± 10.5 53.3 ± 10.6 57.9 ± 9.6 <0.001 Mean aortic gradient (mmHg), mean ± SD 48.3 ± 22.2 49.1 ± 22.4 46.2 ± 21.5 0.22 sPAP, mean ± SD 53.8 ± 17.3 51.7 ± 16.9 55.9 ± 18.2 0.002 Variables Overall (n = 1122) Mitral valve repair (n = 837) Mitral valve replacement (n = 285) P-value Age (years), mean ± SD 60.6 ± 12.8 59.7 ± 13.4 63.3 ± 10.7 <0.001 Male sex, n (%) 558 (49.7) 428 (51.1) 130 (45.6) 0.10 Body surface area, mean ± SD 1.66 ± 0.18 1.66 ± 0.17 1.67 ± 0.19 0.46 Aetiology of MV disease, n (%)  Degenerative 326 (29.0) 272 (32.5) 54 (18.9) <0.001  Rheumatic 666 (59.4) 461 (55.1) 205 (71.9) <0.001  Secondary 59 (5.3) 58 (6.9) 1 (0.4) <0.001  Others 71 (6.3) 46 (5.5) 25 (8.8) 0.069 MV pathology, n (%)  Stenosis 310 (27.6) 240 (28.7) 70 (24.6) 0.18  Regurgitation 547 (48.8) 451 (53.9) 96 (33.7) <0.001  Mixed lesion 265 (23.6) 146 (17.4) 119 (41.8) <0.001 Previous MV surgery, n (%) 127 (11.3) 49 (5.9) 78 (27.4) <0.001 Tricuspid valve disease, n (%) 233 (20.8) 145 (17.3) 88 (30.9) <0.001 Coronary artery disease, n (%) 122 (10.9) 106 (12.7) 16 (5.6) 0.001 NYHA III–IV, n (%) 731 (65.2) 525 (62.7) 206 (72.3) 0.003 Stroke, n (%) 64 (5.7) 45 (5.4) 19 (6.7) 0.41 Hypertension, n (%) 321 (28.6) 216 (25.8) 105 (36.8) <0.001 Diabetes, n (%) 101 (9.0) 59 (7.1) 42 (14.8) <0.001 Chronic kidney disease, n (%) 31 (2.8) 23 (2.8) 8 (2.8) 0.96 COPD, n (%) 85 (7.6) 49 (5.9) 36 (12.6) <0.001 Atrial fibrillation/flutter, n (%) 441 (39.3) 313 (37.4) 128 (44.9) 0.025 Ejection fraction (%), mean ± SD 61.2 ± 10.6 61.0 ± 10.7 61.6 ± 10.0 0.43 LV end-systolic dimension (mm), mean ± SD 41.8 ± 10.7 43.3 ± 10.8 36.8 ± 8.3 <0.001 Left atrial diameter (mm), mean ± SD 54.5 ± 10.5 53.3 ± 10.6 57.9 ± 9.6 <0.001 Mean aortic gradient (mmHg), mean ± SD 48.3 ± 22.2 49.1 ± 22.4 46.2 ± 21.5 0.22 sPAP, mean ± SD 53.8 ± 17.3 51.7 ± 16.9 55.9 ± 18.2 0.002 COPD: chronic obstructive pulmonary disease; LV: left ventricle; MV: mitral valve; NYHA: New York Heart Association; SD: standard deviation; sPAP: systolic pulmonary artery pressure. Statistically significant P-values (<0.05) are in bold. Definitions Mortality and morbidity were reported according to the latest ‘guidelines for reporting mortality and morbidity after cardiac valve interventions’ [13]. Early mortality was defined as death in hospital or within 30 days, and late mortality was defined as death occurring beyond this period. MV aetiology was classified following a thorough analysis of the clinical information, echocardiograms and operative reports. For the purpose of this work and attending to the most frequent aetiologies found, we divided MV disease into 4 major categories: degenerative, rheumatic, secondary (functional) and others (only 6.3% of the entire study population). Secondary MV disease was defined as dysfunction without structural abnormalities of the MV apparatus such as valve prolapse, significant calcification of leaflets or annulus, ruptured chordae (degenerative) and concomitant mitral stenosis (rheumatic) [14]. All patients had preoperative detailed echocardiographic and Doppler examinations. The severity of MR and other relevant measures were determined according to the accepted recommendations [15]. Intraoperative transoesophageal echocardiography, both pre- and post-repair, was routinely used from the beginning of the study, and no patient left the operating room after repair with greater than mild MR, as described previously [16]. Our anticoagulation protocol included a target international normalized ratio of 2.0–3.0 for patients undergoing isolated AVR (mechanical prosthesis) and 2.5–3.5 for those having double valve replacement (mechanical). Patients with a bioprosthesis, either aortic or mitral, were given anti-aggregants only, if there were no other indications for anticoagulation. Data collection Data were retrieved from dedicated databases and included relevant preoperative demographic, clinical and echocardiographic variables, surgical information and postoperative records. Follow-up information, complete for 95.6% of patients, was obtained through a mailed questionnaire or by telephone interview with surviving patients, family members or the patients’ personal physicians, and included vital status and the need for MV reoperation. The cumulative follow-up for the entire cohort was 9522.6 patient-years (mean 8.5 ± 6.7 years; 25th–75th percentile 2.6–13.2 years). All patients gave informed consent for surgery and granted us permission to use their medical records for research purposes, using forms approved by the ethic committee of the respective institutions. Operative procedures and data Mitral and aortic valve exposure was similar in both centres and included a left atriotomy posterior to Waterston’s groove and a ‘hockey-stick’ incision in the ascending aorta, respectively. However, both centres differed significantly regarding the management of MV disease. In Coimbra, repair was attempted in the majority of patients (79.1% repair rate). In Santiago, MV repair was performed less frequently (18.1%). The operative procedures are detailed in Table 2. Table 2: Operative data and mortality Procedures Mitral valve repair Mitral valve replacement P-value Mechanical aortic valve, n (%) 559 (66.8) 199 (69.8) 0.34 Mechanical mitral valve, n (%) 194 (45.6) Tricuspid valve surgery, n (%) 103 (12.3) 74 (41.8) <0.001 CABG, n (%) 72 (8.6) 12 (4.2) 0.023 MV repair techniques  Commissurotomy, n (%) 419 (50.1)  Papillotomy, n (%) 163 (19.5)  Annuloplasty (complete/posterior), n (%) 732 (87.5)  Mean ring size (mm), mean ± SD 32 ± 1.6  Posterior leaflet resection, n (%) 43 (5.1)  Neochordae, n (%) 74 (8.8)  Chordae/transposition shortening, n (%) 14 (1.7) ECC time (min), mean ± SD 87.4 ± 15.9 120.4 ± 51.9 <0.001 Aortic clamping time (min), mean ± SD 59.1 ± 17.7 90.3 ± 44.9 <0.001 30-Day mortality rate, n (%) 15 (1.8) 12 (4.2) 0.021 Procedures Mitral valve repair Mitral valve replacement P-value Mechanical aortic valve, n (%) 559 (66.8) 199 (69.8) 0.34 Mechanical mitral valve, n (%) 194 (45.6) Tricuspid valve surgery, n (%) 103 (12.3) 74 (41.8) <0.001 CABG, n (%) 72 (8.6) 12 (4.2) 0.023 MV repair techniques  Commissurotomy, n (%) 419 (50.1)  Papillotomy, n (%) 163 (19.5)  Annuloplasty (complete/posterior), n (%) 732 (87.5)  Mean ring size (mm), mean ± SD 32 ± 1.6  Posterior leaflet resection, n (%) 43 (5.1)  Neochordae, n (%) 74 (8.8)  Chordae/transposition shortening, n (%) 14 (1.7) ECC time (min), mean ± SD 87.4 ± 15.9 120.4 ± 51.9 <0.001 Aortic clamping time (min), mean ± SD 59.1 ± 17.7 90.3 ± 44.9 <0.001 30-Day mortality rate, n (%) 15 (1.8) 12 (4.2) 0.021 Statistically significant P-values (<0.05) are in bold. AL: anterior leaflet; ECC: extracorporeal circulation; PL: posterior leaflet; SD: standard deviation. Table 2: Operative data and mortality Procedures Mitral valve repair Mitral valve replacement P-value Mechanical aortic valve, n (%) 559 (66.8) 199 (69.8) 0.34 Mechanical mitral valve, n (%) 194 (45.6) Tricuspid valve surgery, n (%) 103 (12.3) 74 (41.8) <0.001 CABG, n (%) 72 (8.6) 12 (4.2) 0.023 MV repair techniques  Commissurotomy, n (%) 419 (50.1)  Papillotomy, n (%) 163 (19.5)  Annuloplasty (complete/posterior), n (%) 732 (87.5)  Mean ring size (mm), mean ± SD 32 ± 1.6  Posterior leaflet resection, n (%) 43 (5.1)  Neochordae, n (%) 74 (8.8)  Chordae/transposition shortening, n (%) 14 (1.7) ECC time (min), mean ± SD 87.4 ± 15.9 120.4 ± 51.9 <0.001 Aortic clamping time (min), mean ± SD 59.1 ± 17.7 90.3 ± 44.9 <0.001 30-Day mortality rate, n (%) 15 (1.8) 12 (4.2) 0.021 Procedures Mitral valve repair Mitral valve replacement P-value Mechanical aortic valve, n (%) 559 (66.8) 199 (69.8) 0.34 Mechanical mitral valve, n (%) 194 (45.6) Tricuspid valve surgery, n (%) 103 (12.3) 74 (41.8) <0.001 CABG, n (%) 72 (8.6) 12 (4.2) 0.023 MV repair techniques  Commissurotomy, n (%) 419 (50.1)  Papillotomy, n (%) 163 (19.5)  Annuloplasty (complete/posterior), n (%) 732 (87.5)  Mean ring size (mm), mean ± SD 32 ± 1.6  Posterior leaflet resection, n (%) 43 (5.1)  Neochordae, n (%) 74 (8.8)  Chordae/transposition shortening, n (%) 14 (1.7) ECC time (min), mean ± SD 87.4 ± 15.9 120.4 ± 51.9 <0.001 Aortic clamping time (min), mean ± SD 59.1 ± 17.7 90.3 ± 44.9 <0.001 30-Day mortality rate, n (%) 15 (1.8) 12 (4.2) 0.021 Statistically significant P-values (<0.05) are in bold. AL: anterior leaflet; ECC: extracorporeal circulation; PL: posterior leaflet; SD: standard deviation. Anticoagulant management Anticoagulation was initiated with warfarin on the first or second postoperative day. A target international normalized ratio of 2.0–3.0 was adopted for isolated mechanical AVR and 2.5–3.5 for double mechanical valve replacement. Patients with a bioprosthesis were maintained with aspirin after a 3-month period (after surgery) under anticoagulation, if there was no formal indication for anticoagulation. Statistical analysis The statistical analysis was performed according to the statistical and data reporting guidelines of the European Journal of Cardio-Thoracic Surgery and Interactive CardioVascular and Thoracic Surgery [17]. Continuous variables are reported as mean ± standard deviation and the median with the 25th–75th percentile and compared using the independent Student’s t-test if normally distributed and using the Mann–Whitney U-test for variables with non-normal distribution. Univariable analysis of categorical data was performed using the χ2 (when no cell was expected to count <5 in contingency tables) or the Fisher’s exact test, and the results are presented as percentages. A multiple logistic regression was performed to identify factors predisposing to replacement rather than repair. Overall survival and freedom from reoperation were plotted using the Kaplan–Meier method, and comparison was made by the log-rank test. Multivariable analyses to identify risk factors for late mortality or reoperation were performed using Cox regression models and calculating hazard ratios and 95% confidence intervals (CIs). All variables with a P-value <0.2 in the univariable analysis were entered in the multivariable analysis (logistic regression or Cox models); in the final model, only those variables with a P-value <0.05 or clinically relevant were retained. To estimate the probability that a patient would have the MV replaced rather than repaired, a multivariable logistic regression model was designed by incorporating demographic information, clinical status and relevant cardiac and non-cardiac comorbidities (Supplementary Material, Table S1), and a propensity score was determined for each patient. Patients were matched according to the propensity score previously calculated by the ‘nearest neighbour matching’ technique, using a calliper of 0.1. Each patient was matched to a single patient (no-replacement). After matching, 232 patients for each group (MV repair vs MV replacement) were obtained for comparison (survival and freedom from reoperation). Subgroup analysis was performed to evaluate the impact of repairing versus replacing the MV in patients with rheumatic MV disease undergoing AVR. This analysis was prespecified. Statistical significance was defined as a 2-tailed P-value <0.05. The data were analysed using the statistical package program SPSS (version 20.0. IBM Corp., Armonk, NY, USA). RESULTS Assessment of the feasibility of mitral valve repair MV repair was performed in 74.6% of the whole population and in 79.2% of patients who had primary MV surgery. In the higher volume centre (Coimbra), the rate of MV repair (primary MV intervention) was even higher, increasing to 94.5% in patients with degenerative MR (flail leaflet) and to 77.6% in those with rheumatic MV disease. Several factors were significantly associated with the probability of having MV replacement in the univariable analysis: MV aetiology; coronary artery disease (CAD); degree of tricuspid valve regurgitation; hypertension; diabetes; atrial fibrillation; chronic obstructive pulmonary disease; degree of MR; higher left atrium and ventricular (systolic) dimensions; higher transaortic gradients; and higher systolic pulmonary artery pressure, surgeon (most experienced) and surgical centre. In the multivariable analysis, only previous MV intervention, surgeon, surgical centre, rheumatic involvement, chronic obstructive pulmonary disease and higher degrees of tricuspid valve regurgitation, MR and systolic pulmonary artery pressure were correlated with MV replacement (Table 3). Table 3: Risk factors for mitral valve replacement (logistic regression) Risk factors OR 95% CI P-value Previous MV surgery 4.776 2.942–0.7753 <0.001 COPD 2.316 1.258–4.266 0.007 sPAP 1.019 1.042–1.381 <0.001 Rheumatic aetiology 5.378 3.211–9.006 <0.001 MV regurgitation degree 1.561 1.290–1.888 0.001 TV regurgitation degree 1.200 1.042–1.381 0.012 Surgeona 0.523 0.358–0.764 0.001 Surgical centreb 0.005 0.002–0.130 <0.001 Risk factors OR 95% CI P-value Previous MV surgery 4.776 2.942–0.7753 <0.001 COPD 2.316 1.258–4.266 0.007 sPAP 1.019 1.042–1.381 <0.001 Rheumatic aetiology 5.378 3.211–9.006 <0.001 MV regurgitation degree 1.561 1.290–1.888 0.001 TV regurgitation degree 1.200 1.042–1.381 0.012 Surgeona 0.523 0.358–0.764 0.001 Surgical centreb 0.005 0.002–0.130 <0.001 The Hosmer–Lemeshow goodness-of-fit χ2 value for this model was 8.0 (p = 0.307). C-statistic: 0.787 (95% CI 0.747–0.827). a The variable surgeon corresponded to the most experienced surgeon (in the number of surgeries performed and the rate of MV repair) in opposition to the remaining. b Coimbra centre. CI: confidence interval; COPD: chronic obstructive pulmonary disease; MV: mitral valve; OR: odds ratio; sPAP: systolic pulmonary artery pressure; TV: tricuspid valve. Table 3: Risk factors for mitral valve replacement (logistic regression) Risk factors OR 95% CI P-value Previous MV surgery 4.776 2.942–0.7753 <0.001 COPD 2.316 1.258–4.266 0.007 sPAP 1.019 1.042–1.381 <0.001 Rheumatic aetiology 5.378 3.211–9.006 <0.001 MV regurgitation degree 1.561 1.290–1.888 0.001 TV regurgitation degree 1.200 1.042–1.381 0.012 Surgeona 0.523 0.358–0.764 0.001 Surgical centreb 0.005 0.002–0.130 <0.001 Risk factors OR 95% CI P-value Previous MV surgery 4.776 2.942–0.7753 <0.001 COPD 2.316 1.258–4.266 0.007 sPAP 1.019 1.042–1.381 <0.001 Rheumatic aetiology 5.378 3.211–9.006 <0.001 MV regurgitation degree 1.561 1.290–1.888 0.001 TV regurgitation degree 1.200 1.042–1.381 0.012 Surgeona 0.523 0.358–0.764 0.001 Surgical centreb 0.005 0.002–0.130 <0.001 The Hosmer–Lemeshow goodness-of-fit χ2 value for this model was 8.0 (p = 0.307). C-statistic: 0.787 (95% CI 0.747–0.827). a The variable surgeon corresponded to the most experienced surgeon (in the number of surgeries performed and the rate of MV repair) in opposition to the remaining. b Coimbra centre. CI: confidence interval; COPD: chronic obstructive pulmonary disease; MV: mitral valve; OR: odds ratio; sPAP: systolic pulmonary artery pressure; TV: tricuspid valve. Early mortality and late survival The 30-day mortality rate for the entire study population was 2.4% (27 patients), significantly higher in the MV replacement group [4.2% vs 1.8%; odds ratio (OR) 2.4, 95% CI 1.1–5.2; P = 0.021] and confirmed in the propensity-matched group (4.7% vs 1.7%; OR 2.8, 95% CI 0.9–9.9; P = 0.06). Interestingly, in patients with rheumatic MV disease, the 30-day mortality rate was not significantly different (2.4% vs 1.5%; OR 1.6, 95% CI 0.5–5.1; P = 0.41). During the follow-up period, 388 patients died (35.4% of patients who survived surgery). The unadjusted survival rate at 4, 8 and 12 years was superior in the MV repair group (85.0 ± 1.3%, 74.1 ± 1.7% and 61.7 ± 2.0%) compared with that in the MV replacement group (81.7 ± 2.5%, 71.9 ± 3.2% and 53.3 ± 4.5%; P = 0.026) (Fig. 1A). This effect persisted in the propensity-matched population (Fig. 1B), where survival at 12 years was also better in the repair group, almost reaching statistical significance (63.3 ± 3.8% vs 54.6 ± 4.9%, P = 0.062). In the subgroup of patients with rheumatic MV disease (Fig. 2), the 12-year survival rate was also significantly better in patients undergoing repair (68.0 ± 2.5% vs 57.9 ± 4.8%; P = 0.018). These results were replicated in both centres when each was evaluated individually. Figure 1: View largeDownload slide Comparison of late survival of patients who had mitral valve repair versus replacement. (A) Non-adjusted survival (overall population) and (B) propensity-matched population. Figure 1: View largeDownload slide Comparison of late survival of patients who had mitral valve repair versus replacement. (A) Non-adjusted survival (overall population) and (B) propensity-matched population. Figure 2: View largeDownload slide Comparison of late survival of patients with rheumatic mitral valve disease who had mitral valve repair versus replacement. Figure 2: View largeDownload slide Comparison of late survival of patients with rheumatic mitral valve disease who had mitral valve repair versus replacement. Numerous factors were identified as risk factors for late mortality in the univariable analysis (P < 0.05): MV replacement, mechanical mitral and aortic prostheses, higher New York Heart Association (NYHA) class, CAD, hypertension, obesity, atrial fibrillation, CABG, age, low ejection fraction, higher degrees of tricuspid valve regurgitation, left atrium and ventricle (systolic) dilation and higher systolic pulmonary artery pressure. However, in the multivariable Cox analysis, only age, MV replacement, higher NYHA class, CAD and mechanical aortic prosthesis were considered independent risk factors for mortality (Table 4). Table 4: Risk factors for late mortality (Cox regression) Risk factors HR 95% CI P-value Mitral valve replacement 1.35 1.04–1.75 0.022 Age (per years) 1.03 1.02–1.04 <0.001 NYHA class 1.26 1.04–1.52 0.018 Aortic bioprosthesis 1.32 1.01–1.72 0.041 Coronary artery disease 1.97 1.47–2.64 <0.001 Risk factors HR 95% CI P-value Mitral valve replacement 1.35 1.04–1.75 0.022 Age (per years) 1.03 1.02–1.04 <0.001 NYHA class 1.26 1.04–1.52 0.018 Aortic bioprosthesis 1.32 1.01–1.72 0.041 Coronary artery disease 1.97 1.47–2.64 <0.001 CI: confidence interval; HR: hazard ratio; NYHA: New York Heart Association. Table 4: Risk factors for late mortality (Cox regression) Risk factors HR 95% CI P-value Mitral valve replacement 1.35 1.04–1.75 0.022 Age (per years) 1.03 1.02–1.04 <0.001 NYHA class 1.26 1.04–1.52 0.018 Aortic bioprosthesis 1.32 1.01–1.72 0.041 Coronary artery disease 1.97 1.47–2.64 <0.001 Risk factors HR 95% CI P-value Mitral valve replacement 1.35 1.04–1.75 0.022 Age (per years) 1.03 1.02–1.04 <0.001 NYHA class 1.26 1.04–1.52 0.018 Aortic bioprosthesis 1.32 1.01–1.72 0.041 Coronary artery disease 1.97 1.47–2.64 <0.001 CI: confidence interval; HR: hazard ratio; NYHA: New York Heart Association. Freedom from mitral valve reoperation There were 58 reoperations to the MV during the study period for a linearized incidence of 0.64%/patient-year. The mean interval to reoperation was 8.1 ± 6.5 years (median 6.9 years; 25th–75th percentile 2.4–12.6 years). There were 17 reoperations in the MV replacement group (6.0%) and 41 in the MV repair group (4.9%, P = 0.482). The causes of reoperation are listed in Table 5, progression of the MV disease and periprosthetic leakage being the most frequent causes in MV repair and MV replacement, respectively. Freedom from MV reoperation at 6 and 12 years was superior in the MV repair group (98.5 ± 0.5% and 94.7 ± 1.1% vs 94.1 ± 1.8% and 89.0 ± 3.1%; P = 0.004) (Fig. 3A and B); this advantage was also observed in the propensity-matched population (P = 0.018). Table 5: Causes of reoperation Causes of reoperation* n (%) MV repair  Technical failure 3 (7.4)   Ring dehiscence 2 (4.9)   Suture dehiscence 1 (2.4)  Disease progression 38 (92.6)   Fibrosis/leaflet retraction 32 (30.5)   New prolapses (rupture of native chordae) 6 (39.1)  Total 41 MV replacement  Leak (non-infectious) 10 (58.8)  Leak (infectious) 3 (17.6)  Pannus 4 (23.5)  Total 17 Causes of reoperation* n (%) MV repair  Technical failure 3 (7.4)   Ring dehiscence 2 (4.9)   Suture dehiscence 1 (2.4)  Disease progression 38 (92.6)   Fibrosis/leaflet retraction 32 (30.5)   New prolapses (rupture of native chordae) 6 (39.1)  Total 41 MV replacement  Leak (non-infectious) 10 (58.8)  Leak (infectious) 3 (17.6)  Pannus 4 (23.5)  Total 17 *Patients could have more than one cause of reoperation. MV: mitral valve. Table 5: Causes of reoperation Causes of reoperation* n (%) MV repair  Technical failure 3 (7.4)   Ring dehiscence 2 (4.9)   Suture dehiscence 1 (2.4)  Disease progression 38 (92.6)   Fibrosis/leaflet retraction 32 (30.5)   New prolapses (rupture of native chordae) 6 (39.1)  Total 41 MV replacement  Leak (non-infectious) 10 (58.8)  Leak (infectious) 3 (17.6)  Pannus 4 (23.5)  Total 17 Causes of reoperation* n (%) MV repair  Technical failure 3 (7.4)   Ring dehiscence 2 (4.9)   Suture dehiscence 1 (2.4)  Disease progression 38 (92.6)   Fibrosis/leaflet retraction 32 (30.5)   New prolapses (rupture of native chordae) 6 (39.1)  Total 41 MV replacement  Leak (non-infectious) 10 (58.8)  Leak (infectious) 3 (17.6)  Pannus 4 (23.5)  Total 17 *Patients could have more than one cause of reoperation. MV: mitral valve. Figure 3: View largeDownload slide Freedom from mitral valve reoperation and comparison of patients who had mitral valve repair versus replacement. (A) Overall population, (B) propensity-matched population and (C) rheumatic mitral valve disease subgroup. Figure 3: View largeDownload slide Freedom from mitral valve reoperation and comparison of patients who had mitral valve repair versus replacement. (A) Overall population, (B) propensity-matched population and (C) rheumatic mitral valve disease subgroup. Patients with degenerative MV disease had the most satisfactory results in this aspect: freedom from MV reoperation at 12 years was 97.7 ± 1.3% in contrast with those who had rheumatic disease (92.7 ± 1.3%, P = 0.038). Noteworthy, freedom from reoperation was similar between groups in rheumatic patients (91.4 ± 2.7% in MV repair vs 93.5 ± 1.5% in MV replacement at 12 years; P = 0.12) (Fig. 3C). In 6 cases of reoperation (5 in the MV replacement group and 1 in MV repair group), MV surgery was not the primary indication for surgery. Those cases included pannus obstruction, periprosthetic leakage and mitral restenosis. After exclusion of these patients, freedom from MV reoperation remained higher in the MV repair group, almost reaching statistical significance (P = 0.06). Moreover, in the propensity-matched group and in the rheumatic MV subgroup, the results remained the same. DISCUSSION The decision to repair or replace the MV in patients undergoing AVR is a challenge yet to be resolved. We believe that our data shed some light on this subject. Firstly, mitroaortic valve surgery, with or without concomitant procedures and including redo cases, can be performed with low mortality (2.4%). Secondly, it is possible to perform MV repair in most cases, even in those with rheumatic disease. Thirdly, there appears to be a survival advantage in repairing the MV in these circumstances, and this was also demonstrated in a propensity-matched population and validated in patients with rheumatic MV disease. Finally, there was no difference in the reoperation rates after repair or replacement of the MV. Which mitral valves can be repaired? MV repair for the treatment of isolated severe MR has been established as the gold standard of surgical care in patients with degenerative MV disease, particularly in cases of MV prolapse. This has been demonstrated even in complex cases, such as bileaflet prolapse and Barlow’s disease [16]. In contrast, the reported rate of MV repair during AVR is widely variable, ranging from 12% to 80% [10, 11, 14, 18, 19]. In a recent report from the Society of Thoracic Surgeons Adult Cardiac Surgery Database (STS-ACSD), which includes 23 404 patients operated on from 1993 to 2007, MV repair was performed in 46% of the patients and replacement in 54% of the patients during AVR [20]. The authors found that MV repair was associated with an independent lower risk of operative mortality (OR 0.61), which improved over time. Despite not having evaluated other important outcomes, such as survival, event-free survival and MV reoperation or MR recurrence, they claimed that during mitroaortic surgery, MV repair should remain the preferred treatment whenever feasible. As expected, rheumatic MV disease was identified as the most powerful independent predictor of MV replacement (5-fold risk) in our population, although we were able to repair 77.6% of rheumatic valves (excluding patients with previous MV surgery). Of note, the high rate of repair achieved in patients with degenerative MV disease (94.5%) in one of our two centres will be a probable scenario in the coming years, because patients with rheumatic valves are slowly disappearing in Western countries. Naturally, not all valves are amenable to repair, particularly those with heavily calcified leaflets or annulus, or with extensive rheumatic involvement (severely thickened, immobile and retracted leaflets and chordae). Thirty-day mortality and long-term survival Concomitant aortic and MV surgery carries an additional risk compared with isolated valve surgery. Data from the STS-ACSD revealed an unadjusted operative mortality rate of 8.2% when the MV was repaired and 11.6% when it was replaced (P < 0.0001). Furthermore, a report from the Northern New England Cardiovascular Disease Study Group showed an even higher in-hospital mortality rate of 15.5% (11.0% for patients <70 years, 18.0% for those 70–79 years and 24% for those >80 years) [21]. However, we and others have clearly demonstrated that, with experience, it is possible to have a significantly lower 30-day mortality rate (1–3%) even when other procedures are performed concomitantly [11, 14]. In all experiences, double valve replacement is associated with a greater risk of early death, probably related to perioperative complications, such as rupture of the atrioventricular sulcus, obstruction of the left outflow tract and low cardiac output. In our study, MV replacement was also associated with a significantly higher 30-day mortality rate (4.2% vs 1.8%, P = 0.021), and this was also observed in the propensity-matched population (4.7% vs 1.7%), almost reaching statistical significance (P = 0.06). This finding was also described in a recent meta-analysis of observational studies, evaluating the outcomes of MV repair compared with MV replacement in this setting [9]. Several risk factors have been identified, such as age, associated CABG and renal failure with dialysis, and they should also be taken into account when deciding which MV procedure to use. Beyond the immediate results, the impact of the surgical option on the long-term outcomes, particularly on late survival, is the focus of intense debate. There are several reasons for the heterogeneity of the results reported [11, 12, 22]. First, the populations among the studies differ with regard to age, patient characteristics, inclusion (or not) of other procedures and the aetiology of the MV disease. Second, almost all series published have significantly more patients in the replacement group than in the repair group. This difference is especially worrisome when evaluating the subgroup of patients with rheumatic MV disease, where the ‘repair arm’ is scarce, which raises concerns about the interpretation of the results. We believe that our study is singular because the majority of patients had MV repair, even the patients with rheumatic disease. Finally, the length of follow-up varied with different studies. In our series, late survival was significantly better after MV repair in the overall population, in the propensity-matched patients and in the subgroup of patients with rheumatic MV disease. Conversely, MV replacement was independently associated with long-term mortality—a fact also demonstrated in the meta-analysis performed by Saurav et al. [9]. CAD, age, mechanical aortic prosthesis and higher NYHA class were identified as predictors of late mortality. Turina et al. [23] found that lower ejection fraction and tricuspid valve surgery were also risk factors for impaired survival in this context. The latter factors and the presence of symptoms are characteristic of a more advanced stage of the disease, which may indicate that patients should be sent to surgery earlier. The majority of these factors was more prevalent in the MV replacement group, which can further worsen their prognosis. Mitral valve reoperation Surprisingly, freedom from reoperation to the MV was higher in the repair group (overall and propensity score matching population), which is in contradiction with the findings of the majority of studies [11, 18, 19]. Additionally, we did not find any difference in the rheumatic population, which is known to be associated with lesser durability of repair. Nonetheless, some caution should be exerted when drawing conclusions because in six patients submitted to redo MV surgery the primary indication for surgery was not the MV. After excluding these patients, freedom from MV reoperation remained higher in the MV repair group, although not significantly so. However, this finding did not change the results in the propensity score matching and rheumatic patients. Ho et al. [22] evaluated 609 patients with rheumatic heart disease who underwent AVR with either MV repair (n = 201) or MV replacement (n = 408). They also did not find any difference with regard to MV reoperation. Our results may be attributed to our extensive experience in repairing rheumatic MVs (over a thousand patients), which may not be reproducible by low-volume centres. In our experience, patients usually return to surgery on average 12–17 years after the initial operation, which means that these results would surely be different after a 20-year follow-up period. On the contrary, in degenerative disease this difference probably will not dissipate over time [24]. Limitations The lack of randomization is always prone to the appearance of unmeasured confounding factors and selection bias. We have tried to minimize these factors by performing the propensity score matching (232 patients per group) and multivariable analyses. The inclusion of 2 centres with distinct approaches to MV disease, 1 more ‘aggressive’ in repairing all kinds of MV aetiologies and the other keener on replacement, can lead to some heterogeneity of the data. However, we believe that the evaluation of the impact of the surgical choice (repair versus replacement) is not impaired by these 2 distinct strategies. Repairing rheumatic MV is more demanding and surgeon-dependent, and so the issue of the generalizability of the results should also be raised in this setting. Nevertheless, these patients also benefited from having their valves repaired, and so surgeons should strive to repair those valves, and these results should be regarded as an incentive. CONCLUSIONS This study confirms that, whenever feasible, MV repair is the procedure of choice in patients undergoing simultaneous AVR because it is associated with lower early and late mortality rates when compared with replacement and with similar rates of reoperation. We have also demonstrated that MV repair can be performed in the majority of cases. Therefore, strategies to improve MV repair rates in this setting should be pursued in the future. SUPPLEMENTARY MATERIAL Supplementary material is available at EJCTS online. Conflict of interest: none declared. REFERENCES 1 Iung B , Baron G , Butchart E , Delahaye F , Gohlke-Bärwolf C , W , Levang O et al. A prospective survey of patients with valvular heart disease in Europe: the Euro Heart Survey on Valvular Heart Disease . Eur Heart J 2003 ; 24 : 1231 – 43 . Google Scholar CrossRef Search ADS PubMed 2 Nowicki ER , Birkmeyer NJ , Weintraub RW , Leavitt BJ , Sanders JH , Dacey LJ et al. Multivariable prediction of in-hospital mortality associated with aortic and mitral valve surgery in Northern New England . Ann Thorac Surg 2004 ; 77 : 1966 – 77 . Google Scholar CrossRef Search ADS PubMed 3 Birkmeyer NJ , Marrin CA , Morton JR , Leavitt BJ , Lahey SJ , Charlesworth DC et al. Decreasing mortality for aortic and mitral valve surgery in Northern New England. Northern New England cardiovascular Disease Study Group . Ann Thorac Surg 2000 ; 70 : 432 – 7 . Google Scholar CrossRef Search ADS PubMed 4 Tribouilloy C , Enriquez-Sarano M , Schaff H , Orszulak TA , Bailey KR , Tajik AJ et al. Impact of preoperative symptoms on survival after surgical correction of organic mitral regurgitation: rationale for optimizing surgical indications . Circulation 1999 ; 99 : 400 – 5 . Google Scholar CrossRef Search ADS PubMed 5 Suri RM , Schaff H , Dearani JA , Sundt TM , Daly RC , Mullany CJ et al. Survival advantage and improved durability of mitral repair for leaflet prolapse subsets in the current era . Ann Thorac Surg 2006 ; 82 : 819 – 26 . Google Scholar CrossRef Search ADS PubMed 6 Gillinov AM , Blackstone EH , Nowicki ER , Slisatkorn W , Al-Dossari G , Johnston DR et al. Valve repair versus valve replacement for degenerative mitral valve disease . J Thorac Cardiovasc Surg 2008 ; 135 : 885 – 93 . Google Scholar CrossRef Search ADS PubMed 7 Falk V , Baumgartner H , Bax JJ , De Bonis M , Hamm C , Holm PJ et al. 2017 ESC/EACTS guidelines for the management of valvular heart disease: the Task Force for the Management of Valvular Heart Disease of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS) . Eur J Cardiothorac Surg 2017 ; 52 : 616 – 64 . Google Scholar CrossRef Search ADS PubMed 8 Nishimura RA , Otto CM , Bonow RO , Carabello BA , Erwin JP , Guyton R et al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines . J Am Coll Cardiol 2014 ; 63 : e57 – 185 . Google Scholar CrossRef Search ADS PubMed 9 Saurav A , Alla VM , Kaushik M , Hunter CC , Mooss AV. Outcomes of mitral valve repair compared with replacement in patients undergoing concomitant aortic valve surgery: a meta-analysis of observational studies . Eur J Cardiothorac Surg 2015 ; 48 : 347 – 53 . Google Scholar CrossRef Search ADS PubMed 10 Urban M , Pirk J , Szarszoi O , Skalsky I , Maly J , Netuka I. Mitral valve repair versus replacement in simultaneous aortic and mitral valve surgery . Expert Clin Cardiol 2013 ; 18 : 22 – 6 . 11 Hamamoto M , Bando K , Kobayashi J , Satoh T , Sasako Y , Niwaya K et al. Durability and outcome of aortic valve replacement with mitral valve repair versus double valve replacement . Ann Thorac Surg 2003 ; 75 : 28 – 33 . Google Scholar CrossRef Search ADS PubMed 12 Talwar S , Mathur A , Choudhary SK , Singh R , Kumar AS. Aortic valve replacement with mitral valve repair compared with combined aortic and mitral valve replacement . Ann Thorac Surg 2007 ; 84 : 1219 – 25 . Google Scholar CrossRef Search ADS PubMed 13 Akins CW , Miller DC , Turina MI , Kouchoukos NT , Blackstone EH , Grunkemeier GL et al. Guidelines for reporting mortality and morbidity after cardiac valve interventions . Eur J Cardiothorac Surg 2008 ; 33 : 523 – 8 . Google Scholar CrossRef Search ADS PubMed 14 Coutinho GF , Correia PM , Antunes MJ. Concomitant aortic and mitral surgery: to replace or repair the mitral valve? J Thorac Cardiovasc Surg 2014 ; 148 : 1386 – 92 . Google Scholar CrossRef Search ADS PubMed 15 Lang RM , Bierig M , Devereux RB , Flachskampf FA , Foster E , Pellikka PA et al. Recommendations for chamber quantification . Eur J Echocardiogr 2006 ; 7 : 79 – 108 . Google Scholar CrossRef Search ADS PubMed 16 Coutinho GF , Correia PM , Branco C , Antunes MJ. Long-term results of mitral valve surgery for degenerative anterior leaflet or bileaflet prolapse: analysis of negative factors for repair, early and late failures, and survival . Eur J Cardiothorac Surg 2016 ; 50 : 66 – 74 . Google Scholar CrossRef Search ADS PubMed 17 Hickey GL , Dunning J , Seifert B , Sodeck G , Carr MJ , Burger HU et al. Statistical and data reporting guidelines for the European Journal of Cardio-Thoracic Surgery and the Interactive CardioVascular and Thoracic Surgery . Eur J Cardiothorac Surg 2015 ; 48 : 180 – 93 . Google Scholar CrossRef Search ADS PubMed 18 Gillinov AM , Blackstone EH , Cosgrove DM , White J , Kerr P , Marullo A et al. Mitral valve repair with aortic valve replacement is superior to double valve replacement . J Thorac Cardiovasc Surg 2003 ; 125 : 1372 – 85 . Google Scholar CrossRef Search ADS PubMed 19 Kuwaki K , Kawaharada N , Morishita K , Koyanagi T , Osawa H , Maeda T et al. Mitral valve repair versus replacement in simultaneous mitral and aortic valve surgery for rheumatic disease . Ann Thorac Surg 2007 ; 83 : 558 – 63 . Google Scholar CrossRef Search ADS PubMed 20 Thourani VH , Suri RM , Rankin JS , He X , O'Brien SM , Badhwar V et al. Does mitral valve repair offer an advantage over replacement in patients undergoing aortic valve replacement? Ann Thorac Surg 2014 ; 98 : 598 – 7 . Google Scholar CrossRef Search ADS PubMed 21 Leavitt BJ , Baribeau YR , DiScipio AW , Ross CS , Quinn RD , Olmstead EM et al. ; Northern New England Cardiovascular Disease Study Group . Outcomes of patients undergoing concomitant aortic and mitral valve surgery in northern New England . Circulation 2009 ; 120 : S155 – 62 . Google Scholar CrossRef Search ADS PubMed 22 Ho H-QT , Nguyen V-P , Phan K-P , Pham N-V. Mitral valve repair with aortic valve replacement in rheumatic heart disease . Asian Cardiovasc Thorac Ann 2004 ; 12 : 341 – 5 . Google Scholar CrossRef Search ADS PubMed 23 Turina J , Stark T , Seifert B , Turina M. Predictors of the long-term outcome after combined aortic and mitral valve surgery . Circulation 1999 ; 100 : II48 – 53 . Google Scholar CrossRef Search ADS PubMed 24 Coutinho GF , Antunes MJ. Mitral valve repair for degenerative mitral valve disease: surgical approach, patient selection and long-term outcomes . Heart 2017 ; 103 : 1663 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

European Journal of Cardio-Thoracic SurgeryOxford University Press

Published: May 25, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off