Multiple reoperations on the aortic valve: outcomes and implications for future potential valve-in-valve strategy

Multiple reoperations on the aortic valve: outcomes and implications for future potential... Abstract OBJECTIVES Surgical mortality and long-term outcomes are important considerations when determining strategies for multiple reoperations on the aortic valve (AV). With the rise of percutaneous valve-in-valve, we sought to evaluate the current outcomes of conventional surgery for AV reoperation, focusing first on the effect of the number of previous AV interventions with a subsequent analysis of the risk factors for adverse outcomes. METHODS From January 2007 to December 2016, 316 consecutive patients underwent an open redo operation (replacement) on their AV at a single centre. It was the first AV reintervention in 263 patients (Group 1), second in 42 patients (Group 2) and third or more in 11 patients (Group 3). RESULTS There were 230 men and 86 women, with a median age of 58 (Q1–Q3: 46–70) years. Structural valve deterioration (SVD) of the bioprosthesis (n = 136, 44%), endocarditis (n = 57, 18%) and prosthetic valve dehiscence (n = 41, 13%) were the most common reasons for reintervention. Overall, in-hospital mortality was 7.3%: 7.2% in Group 1, 4.76% in Group 2 and 18.2% in Group 3 (P = 0.233) and ranged from 3.7% for SVD to 14.0% when endocarditis was the reason for reintervention. Higher preoperative New York Heart Association (NYHA) class (III/IV) [odds ratio (OR) 15.9, P = 0.011], injury during re-entry (OR 16.9, P = 0.015), endocarditis (OR 3.7, P = 0.038) and concomitant mitral valve replacement (OR 5.6, P = 0.006) were independent risk factors for in-hospital mortality. Survival at 8 years was 79.0 ± 3.0% for the entire cohort and 88.4 ± 3.2% for re-replacement after SVD. CONCLUSIONS Multiple AV reoperations carry an acceptable risk of early postoperative mortality, particularly for isolated valve replacements of SVD. Aortic valve replacement, Redo aortic valve, Aortic valve, Bioprosthesis INTRODUCTION Redo-aortic valve replacement (AVR) following the first procedure can be challenging. A combination of increased life expectancy post-valvular surgery, an ageing population (particularly in Western societies) and an increasing preference for bioprosthetic valves has led to an increasing number of surgical patients presenting for redo valve replacements [1, 2]. To date, surgical open AVR remains the gold standard for managing aortic valve (AV) disease. With a 1% incidence in the USA, AV disease can include pathologies from the stenotic valve to the regurgitant valve and rheumatic heart disease. It represents a significant health burden, particularly in the elderly population. Currently, transcatheter valve-in-valve implantation is considered largely in patients who are at high operative risk; however, in this current paradigm of bioprosthetic structural valve deterioration (SVD), it is a management option that is gaining momentum. Indeed, a retrospective single-centre study by Erlebach et al. [3] suggested no significant difference in the 30-day mortality in patients who underwent transcatheter valve-in-valve implantation after previous open valve replacement compared with patients undergoing traditional operative redo-AVR. However, 1-year survival was significantly better in patients who underwent a redo open surgery. Moreover, when it comes to patients who are undergoing a second or a third or more AVR, it is not clear at this stage which valve substitute and which strategy would be best suited. We sought to evaluate the postoperative mortality and morbidity of patients undergoing redo-AVR for the first, second, third or more times in the last decade, with the intention of using the results to inform management decisions as patients increasingly present for redo procedures. MATERIALS AND METHODS This is a retrospective 10-year period study conducted between January 2007 and December 2016. We identified 316 consecutive patients at the Georges Pompidou European Hospital in Paris, France, who underwent open redo surgery on their AV. Patients were included in this study if they previously had an open surgery to their AV. Non-AV open-heart surgeries were not considered in the count of previous interventions to the AV. The study cohort was divided into 3 groups based on the number of current reinterventions to their AV: Group 1, patients with a first redo intervention (n = 263); Group 2, patients with a second redo intervention (n = 42); and Group 3, patients with a third redo intervention or beyond (n = 11). Once a patient was included in the study, he/she was not counted again if reintervention occurred during the follow-up period, thus keeping the samples independent for statistical analysis. The study was approved by the institutional review board. AVR in the study cohort was performed via the usual open surgical manner, reprising the median sternotomy scar from previous operations. In-hospital data were collected retrospectively from medical records. Follow-up was done through direct phone contact with the referring cardiologist and the patient. Mean follow-up period at the completion of the study was 60 ± 30 months (calculation of means made with censoring deceased patients at the time of death). Statistical analysis Descriptive statistics were expressed as median and 25–75% quartiles [median (Q1–Q3)] for continuous variables and numbers (proportions) for categorical variables. To identify the significant differences in preoperative and intraoperative categorical data and 30-day mortality between groups, a χ2 test was used, except when the expected contingent cell in cross-tabulation was <5, in which case a Fisher’s exact test was preferred. One-way analysis of variance or non-parametric Kruskal–Wallis test was also used for differences in continuous data, depending on the normality distribution of the variables within groups. Estimates for long-term survival and recurrence of any type of AV reintervention (including percutaneous procedures) were made using the Kaplan–Meier method. Considering the total study cohort, preoperative data of the patients were then entered into a stepwise binary logistic regression analysis to identify the independent predictors for in-hospital mortality. The appropriateness of the tested variables was first determined by univariable analysis using the χ2 or Fisher’s exact test for categorical data and the Student’s t-test for continuous data with a P-value <0.2. The log-rank test was used to determine univariable predictive factors of survival. Each variable with a P-value <0.2 was entered into a Cox proportional hazard regression multivariable analysis with a backward stepwise selection process to identify the independent multivariable predictive factors of long-term mortality. Statistical analyses were performed using IBM SPSS Statistics v22 software (IBM Corp, Armonk, NY, USA). RESULTS There were 230 men and 86 women with a median age of 58 (46–70) years. Baseline characteristics of the 3 groups based on the number of AV reinterventions are listed in Table 1. Compared with Group 3, patients in Group 1 had a lower mean systolic pulmonary arterial pressure and a higher mean left ventricular ejection fraction. Reasons for reintervention were SVD in 43.0% (n = 136), AV endocarditis in 18.0% (n = 57), prosthetic valve dehiscence (PVD) in 13.0% (n = 41), recurrent AV disease after previous valve repairs in 11.1% (n = 35), mechanical valve dysfunction in 8.9% (n = 28) and redo-Bentall procedure for aortic aneurysm in 6.0% (n = 19). Details of the reasons for reintervention in each group are shown in Fig. 1. Table 1: Baseline characteristics of the study population based on the number of previous AV interventions categorized by groups   No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Age (years), median (Q1–Q3)  58 (46–71)  58 (48–68)  50 (42–59)  Gender, n (%)   Female  74 (28.1)  9 (21.4)  3 (27.3)   Male  189 (71.9)  33 (78.6)  8 (72.7)  Serum creatinine (µmol/l), median (Q1–Q3)  88 (75–107)  102 (83–120)  93 (75–105)  COPD, n (%)  8 (3.0)  1 (2.4)  0 (0.0)  Diabetes, n (%)  18 (6.9)  3 (7.1)  0 (0.0)  NYHA class, n (%)   I  27 (10.3)  2 (4.8)  1 (9.1)   II  115 (43.7)  10 (23.8)  2 (18.2)   III  91 (34.6)  15 (35.7)  5 (45.5)   IV  30 (11.4)  8 (19.0)  2 (18.2)  LVEF (%), median (Q1–Q3)  60 (55–65)  58 (50–65)  50 (35–65)*  sPAP (mmHg), median (Q1–Q3)  35 (25–40)  35 (30–45)  50 (35–65)*  Urgency, n (%)   Elective  102 (38.8)  21 (50.0)  5 (45.5)   Urgent  121 (46.0)  15 (35.7)  4 (36.4)   Emergent  35 (13.3)  6 (14.3)  2 (18.2)   Salvage  5 (1.9)  0 (0.0)  0 (0.0)  EuroSCORE II, median (Q1–Q3)  7.12 (3.80–12.40)  7.56 (4.14–14.97)  5.99 (2.49–15.99)  STS mortality score, median (Q1–Q3)  1.80 (1.24–3.08)  2.22 (1.38–3.09)  2.18 (1.18–2.49)    No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Age (years), median (Q1–Q3)  58 (46–71)  58 (48–68)  50 (42–59)  Gender, n (%)   Female  74 (28.1)  9 (21.4)  3 (27.3)   Male  189 (71.9)  33 (78.6)  8 (72.7)  Serum creatinine (µmol/l), median (Q1–Q3)  88 (75–107)  102 (83–120)  93 (75–105)  COPD, n (%)  8 (3.0)  1 (2.4)  0 (0.0)  Diabetes, n (%)  18 (6.9)  3 (7.1)  0 (0.0)  NYHA class, n (%)   I  27 (10.3)  2 (4.8)  1 (9.1)   II  115 (43.7)  10 (23.8)  2 (18.2)   III  91 (34.6)  15 (35.7)  5 (45.5)   IV  30 (11.4)  8 (19.0)  2 (18.2)  LVEF (%), median (Q1–Q3)  60 (55–65)  58 (50–65)  50 (35–65)*  sPAP (mmHg), median (Q1–Q3)  35 (25–40)  35 (30–45)  50 (35–65)*  Urgency, n (%)   Elective  102 (38.8)  21 (50.0)  5 (45.5)   Urgent  121 (46.0)  15 (35.7)  4 (36.4)   Emergent  35 (13.3)  6 (14.3)  2 (18.2)   Salvage  5 (1.9)  0 (0.0)  0 (0.0)  EuroSCORE II, median (Q1–Q3)  7.12 (3.80–12.40)  7.56 (4.14–14.97)  5.99 (2.49–15.99)  STS mortality score, median (Q1–Q3)  1.80 (1.24–3.08)  2.22 (1.38–3.09)  2.18 (1.18–2.49)  * P = 0.01: non-parametric Kruskal–Wallis test (Group 3 vs Group 1). AV: aortic valve; COPD: chronic obstructive pulmonary disease; LVEF: left ventricular ejection fraction; NYHA: New York Heart Association; Q1: 25% quartile; Q3: 75% quartile; sPAP: systolic pulmonary arterial pressure; STS: Society of Thoracic Surgeons. Table 1: Baseline characteristics of the study population based on the number of previous AV interventions categorized by groups   No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Age (years), median (Q1–Q3)  58 (46–71)  58 (48–68)  50 (42–59)  Gender, n (%)   Female  74 (28.1)  9 (21.4)  3 (27.3)   Male  189 (71.9)  33 (78.6)  8 (72.7)  Serum creatinine (µmol/l), median (Q1–Q3)  88 (75–107)  102 (83–120)  93 (75–105)  COPD, n (%)  8 (3.0)  1 (2.4)  0 (0.0)  Diabetes, n (%)  18 (6.9)  3 (7.1)  0 (0.0)  NYHA class, n (%)   I  27 (10.3)  2 (4.8)  1 (9.1)   II  115 (43.7)  10 (23.8)  2 (18.2)   III  91 (34.6)  15 (35.7)  5 (45.5)   IV  30 (11.4)  8 (19.0)  2 (18.2)  LVEF (%), median (Q1–Q3)  60 (55–65)  58 (50–65)  50 (35–65)*  sPAP (mmHg), median (Q1–Q3)  35 (25–40)  35 (30–45)  50 (35–65)*  Urgency, n (%)   Elective  102 (38.8)  21 (50.0)  5 (45.5)   Urgent  121 (46.0)  15 (35.7)  4 (36.4)   Emergent  35 (13.3)  6 (14.3)  2 (18.2)   Salvage  5 (1.9)  0 (0.0)  0 (0.0)  EuroSCORE II, median (Q1–Q3)  7.12 (3.80–12.40)  7.56 (4.14–14.97)  5.99 (2.49–15.99)  STS mortality score, median (Q1–Q3)  1.80 (1.24–3.08)  2.22 (1.38–3.09)  2.18 (1.18–2.49)    No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Age (years), median (Q1–Q3)  58 (46–71)  58 (48–68)  50 (42–59)  Gender, n (%)   Female  74 (28.1)  9 (21.4)  3 (27.3)   Male  189 (71.9)  33 (78.6)  8 (72.7)  Serum creatinine (µmol/l), median (Q1–Q3)  88 (75–107)  102 (83–120)  93 (75–105)  COPD, n (%)  8 (3.0)  1 (2.4)  0 (0.0)  Diabetes, n (%)  18 (6.9)  3 (7.1)  0 (0.0)  NYHA class, n (%)   I  27 (10.3)  2 (4.8)  1 (9.1)   II  115 (43.7)  10 (23.8)  2 (18.2)   III  91 (34.6)  15 (35.7)  5 (45.5)   IV  30 (11.4)  8 (19.0)  2 (18.2)  LVEF (%), median (Q1–Q3)  60 (55–65)  58 (50–65)  50 (35–65)*  sPAP (mmHg), median (Q1–Q3)  35 (25–40)  35 (30–45)  50 (35–65)*  Urgency, n (%)   Elective  102 (38.8)  21 (50.0)  5 (45.5)   Urgent  121 (46.0)  15 (35.7)  4 (36.4)   Emergent  35 (13.3)  6 (14.3)  2 (18.2)   Salvage  5 (1.9)  0 (0.0)  0 (0.0)  EuroSCORE II, median (Q1–Q3)  7.12 (3.80–12.40)  7.56 (4.14–14.97)  5.99 (2.49–15.99)  STS mortality score, median (Q1–Q3)  1.80 (1.24–3.08)  2.22 (1.38–3.09)  2.18 (1.18–2.49)  * P = 0.01: non-parametric Kruskal–Wallis test (Group 3 vs Group 1). AV: aortic valve; COPD: chronic obstructive pulmonary disease; LVEF: left ventricular ejection fraction; NYHA: New York Heart Association; Q1: 25% quartile; Q3: 75% quartile; sPAP: systolic pulmonary arterial pressure; STS: Society of Thoracic Surgeons. Figure 1: View largeDownload slide Proportion of reasons for reintervention based on the number of previous aortic valve surgeries categorized by groups. AVRr: aortic valve replacement after previous AV repair; Bentall redo: redo Bentall procedure; MVD: mechanical valve dysfunction; PVD: prosthetic valve dehiscence; SVD: structural valve deterioration of a bioprosthesis. Figure 1: View largeDownload slide Proportion of reasons for reintervention based on the number of previous aortic valve surgeries categorized by groups. AVRr: aortic valve replacement after previous AV repair; Bentall redo: redo Bentall procedure; MVD: mechanical valve dysfunction; PVD: prosthetic valve dehiscence; SVD: structural valve deterioration of a bioprosthesis. Intraoperative characteristics Descriptive data regarding surgical procedures specific for each group are listed in Table 2. All patients underwent an AV replacement. A total of 192 (60.8%) patients underwent one or more concomitant procedures, with 29 (9.3%) patients undergoing tricuspid valve repair or replacement, 38 (12.3%) a mitral valve replacement (MVR), 38 (12.0%) a replacement of the ascending aorta and 49 (15.5%) a redo-Bentall procedure. Concomitant coronary artery bypass grafting was also performed in 18 (5.7%) patients. Table 2: Intraoperative characteristics of the study population based on the number of previous AV interventions categorized by groups   No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Cross-clamp time (min), median (Q1–Q3)  93 (67–127)  90 (72–150)  95 (85–117)  CPB time (min), median (Q1–Q3)  125 (93–179)  132 (97–212)  180 (150–211)  Circulatory arrest, n (%)  21 (8.0)  3 (7.1)  1 (9.1)  AVR prosthesis type, n (%)   Mechanical valve  108 (41.1)  19 (45.2)  6 (54.5)   Bioprosthesis  148 (56.3)  22 (52.4)  5 (45.5)   Homograft  7 (2.6)  1 (2.4)  0 (0.0)  CABG, n (%)  16 (6.1)  2 (4.8)  0 (0.0)  MVR, n (%)  28 (10.6)  9 (21.4)  1 (9.1)  MV repair, n (%)  9 (3.4)  1 (2.4)  1 (9.1)  PVR, n (%)  1 (0.4)  1 (2.4)  0 (0.0)  TVR, n (%)  4 (1.5)  3 (7.1)  1 (9.1)  TV repair, n (%)  18 (6.8)  2 (4.8)  1 (9.1)  Ascending aorta replacement, n (%)  32 (12.2)  6 (14.3)  0 (0.0)  Aortic root enlargement, n (%)  14 (5.3)  5 (11.9)  1 (9.1)  Bentall, n (%)  41 (16.0)  6 (14.3)  2 (18.2)    No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Cross-clamp time (min), median (Q1–Q3)  93 (67–127)  90 (72–150)  95 (85–117)  CPB time (min), median (Q1–Q3)  125 (93–179)  132 (97–212)  180 (150–211)  Circulatory arrest, n (%)  21 (8.0)  3 (7.1)  1 (9.1)  AVR prosthesis type, n (%)   Mechanical valve  108 (41.1)  19 (45.2)  6 (54.5)   Bioprosthesis  148 (56.3)  22 (52.4)  5 (45.5)   Homograft  7 (2.6)  1 (2.4)  0 (0.0)  CABG, n (%)  16 (6.1)  2 (4.8)  0 (0.0)  MVR, n (%)  28 (10.6)  9 (21.4)  1 (9.1)  MV repair, n (%)  9 (3.4)  1 (2.4)  1 (9.1)  PVR, n (%)  1 (0.4)  1 (2.4)  0 (0.0)  TVR, n (%)  4 (1.5)  3 (7.1)  1 (9.1)  TV repair, n (%)  18 (6.8)  2 (4.8)  1 (9.1)  Ascending aorta replacement, n (%)  32 (12.2)  6 (14.3)  0 (0.0)  Aortic root enlargement, n (%)  14 (5.3)  5 (11.9)  1 (9.1)  Bentall, n (%)  41 (16.0)  6 (14.3)  2 (18.2)  AV: aortic valve; CABG: coronary artery bypass grafting; CPB: cardiopulmonary bypass; MV: mitral valve; MVR: mitral valve replacement; PVR: pulmonary valve replacement; Q1: 25% quartile; Q3: 75% quartile; TV: tricuspid valve; TVR: tricuspid valve replacement. Table 2: Intraoperative characteristics of the study population based on the number of previous AV interventions categorized by groups   No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Cross-clamp time (min), median (Q1–Q3)  93 (67–127)  90 (72–150)  95 (85–117)  CPB time (min), median (Q1–Q3)  125 (93–179)  132 (97–212)  180 (150–211)  Circulatory arrest, n (%)  21 (8.0)  3 (7.1)  1 (9.1)  AVR prosthesis type, n (%)   Mechanical valve  108 (41.1)  19 (45.2)  6 (54.5)   Bioprosthesis  148 (56.3)  22 (52.4)  5 (45.5)   Homograft  7 (2.6)  1 (2.4)  0 (0.0)  CABG, n (%)  16 (6.1)  2 (4.8)  0 (0.0)  MVR, n (%)  28 (10.6)  9 (21.4)  1 (9.1)  MV repair, n (%)  9 (3.4)  1 (2.4)  1 (9.1)  PVR, n (%)  1 (0.4)  1 (2.4)  0 (0.0)  TVR, n (%)  4 (1.5)  3 (7.1)  1 (9.1)  TV repair, n (%)  18 (6.8)  2 (4.8)  1 (9.1)  Ascending aorta replacement, n (%)  32 (12.2)  6 (14.3)  0 (0.0)  Aortic root enlargement, n (%)  14 (5.3)  5 (11.9)  1 (9.1)  Bentall, n (%)  41 (16.0)  6 (14.3)  2 (18.2)    No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Cross-clamp time (min), median (Q1–Q3)  93 (67–127)  90 (72–150)  95 (85–117)  CPB time (min), median (Q1–Q3)  125 (93–179)  132 (97–212)  180 (150–211)  Circulatory arrest, n (%)  21 (8.0)  3 (7.1)  1 (9.1)  AVR prosthesis type, n (%)   Mechanical valve  108 (41.1)  19 (45.2)  6 (54.5)   Bioprosthesis  148 (56.3)  22 (52.4)  5 (45.5)   Homograft  7 (2.6)  1 (2.4)  0 (0.0)  CABG, n (%)  16 (6.1)  2 (4.8)  0 (0.0)  MVR, n (%)  28 (10.6)  9 (21.4)  1 (9.1)  MV repair, n (%)  9 (3.4)  1 (2.4)  1 (9.1)  PVR, n (%)  1 (0.4)  1 (2.4)  0 (0.0)  TVR, n (%)  4 (1.5)  3 (7.1)  1 (9.1)  TV repair, n (%)  18 (6.8)  2 (4.8)  1 (9.1)  Ascending aorta replacement, n (%)  32 (12.2)  6 (14.3)  0 (0.0)  Aortic root enlargement, n (%)  14 (5.3)  5 (11.9)  1 (9.1)  Bentall, n (%)  41 (16.0)  6 (14.3)  2 (18.2)  AV: aortic valve; CABG: coronary artery bypass grafting; CPB: cardiopulmonary bypass; MV: mitral valve; MVR: mitral valve replacement; PVR: pulmonary valve replacement; Q1: 25% quartile; Q3: 75% quartile; TV: tricuspid valve; TVR: tricuspid valve replacement. A total of 6 re-entry injuries (1.9%) occurred, involving the thoracic aorta (n = 4), the superior vena cava (n = 1) and the right atrium (n = 1). The latter 2 were fatal. Immediate postoperative outcomes Postoperative outcomes are shown in Fig. 2. Overall hospital mortality was 7.3% (7.2% in Group 1, 4.76% in Group 2 and 18.18% in Group 3; P = 0.233, Fisher’s exact test). Low cardiac output syndrome, defined as inotropic requirement over 48 h to maintain a systolic arterial pressure above 90 mmHg, differed between groups, being far more frequent in Group 3 compared with Groups 1 and 2 (P = 0.04, χ2 test). In addition to ionotropic support, 15 patients with low cardiac output syndrome were treated with intra-aortic balloon pump and 3 with extracorporeal membrane oxygenation. A total of 23 surgical revisions were performed for excessive bleeding (7.3%). Figure 2: View largeDownload slide In-hospital postoperative outcomes of the study population based on the number of previous AV surgeries categorized by groups. †P = 0.036: χ2 test. AV: aortic valve; LCOS: low cardiac output syndrome; MOF: multiorgan failure; PM: pacemaker. Figure 2: View largeDownload slide In-hospital postoperative outcomes of the study population based on the number of previous AV surgeries categorized by groups. †P = 0.036: χ2 test. AV: aortic valve; LCOS: low cardiac output syndrome; MOF: multiorgan failure; PM: pacemaker. Long-term follow-up Global survival rate at 8 years was 79.0 ± 3.0%. Based on the number of previous AV interventions, survival rates at 8 years were 79.6 ± 3.3% in Group 1, 80.3 ± 9.0% in Group 2 and 50.4% ± 10.1% in Group 3, respectively (Fig. 3). Figure 3: View largeDownload slide Overall survival curve using the Kaplan–Meier method, based on the number of previous aortic valve surgery categorized by groups. Figure 3: View largeDownload slide Overall survival curve using the Kaplan–Meier method, based on the number of previous aortic valve surgery categorized by groups. Recurrence of any new AV procedure at 8 years was 86.7 ± 3.4%, 89.8 ± 2.6% in Group 1 and 84.1 ± 6.7% in Group 2 (log-rank test, P = 0.198). None of the Group 3 patients were reoperated on AV during follow-up. Based on the initial reason for reintervention, recurrence of any new AV procedure at 8 years was 97.8 ± 1.8% for SVD, 80.9 ± 10.0 for redo-Bentall procedure, 81.8 ± 7.9 for endocarditis, 68.2 ± 14% for PVD, 86.9 ± 6.2% for AV disease after previous valve repair and 76.5 ± 17.4 for mechanical valve dysfunction (log-rank test, P = 0.032). Survival at 8 years after AV re-replacement for SVD of a bioprosthesis was 88.4% ± 3.2%. There was no statistical difference in long-term survival between patients receiving a bioprosthesis or a mechanical valve (survival rates were 76.6 ± 4.6% and 81.7 ± 4.0% at 8 years, respectively, log-rank test, P = 0.425). Multivariable predictors of in-hospital and long-term outcomes Based on the reason for current reintervention, in-hospital mortality was 3.7% for SVD, 5.3% for redo-Bentall’s procedure, 5.7% for AV disease after previous valve repair, 9.8% for PVD, 10.7% for mechanical valve dysfunction and 14.0% for endocarditis. Considering the whole cohort by multivariable analysis, risk factors associated with overall in-hospital mortality are presented in Table 3. The area under the curve for the receiver operating characteristics of EuroSCORE II and Society of Thoracic Surgeons (STS) mortality score as predictors of surgical mortality were 0.830 (0.757–0.904) and 0.826 (0.742–0.910), respectively (Fig. 4). Table 3: Univariable and multivariable analyses of in-hospital mortality   Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Endocarditis  2.65  1.07–6.62  0.036  3.66  1.07–12.50  0.038  Female gender  2.66  1.13–6.29  0.025  1.44  0.37–5.51  0.596  Serum creatinine level >120 µmol/l  2.46  0.98–6.17  0.054  1.03  0.28–3.85  0.970  NYHA Class III or IV  11.36  2.58–50.00  0.001  15.87  1.90–142.86  0.011  LVEF  1.03  0.99–1.07  0.128  1.02  0.96–1.07  0.559  Non-elective surgery  2.44  0.88–6.76  0.086  2.65  0.60–16.39  0.174  sPAP≥ 55 mmHg  3.57  1.26–10.10  0.017  1.58  0.42–5.99  0.501  Tricuspid annuloplasty  3.41  1.05–11.11  0.042  3.69  0.74–18.52  0.112  MVR  3.70  1.41–9.71  0.008  5.65  1.66–19.23  0.006  Injury during resternotomy  6.85  1.18–40.00  0.032  16.95  1.74–166.67  0.015    Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Endocarditis  2.65  1.07–6.62  0.036  3.66  1.07–12.50  0.038  Female gender  2.66  1.13–6.29  0.025  1.44  0.37–5.51  0.596  Serum creatinine level >120 µmol/l  2.46  0.98–6.17  0.054  1.03  0.28–3.85  0.970  NYHA Class III or IV  11.36  2.58–50.00  0.001  15.87  1.90–142.86  0.011  LVEF  1.03  0.99–1.07  0.128  1.02  0.96–1.07  0.559  Non-elective surgery  2.44  0.88–6.76  0.086  2.65  0.60–16.39  0.174  sPAP≥ 55 mmHg  3.57  1.26–10.10  0.017  1.58  0.42–5.99  0.501  Tricuspid annuloplasty  3.41  1.05–11.11  0.042  3.69  0.74–18.52  0.112  MVR  3.70  1.41–9.71  0.008  5.65  1.66–19.23  0.006  Injury during resternotomy  6.85  1.18–40.00  0.032  16.95  1.74–166.67  0.015  LVEF: left ventricular ejection fraction; MVR: mitral valve replacement; NYHA: New York Heart Association; OR: odds ratio; sPAP: systolic pulmonary artery pressure. Table 3: Univariable and multivariable analyses of in-hospital mortality   Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Endocarditis  2.65  1.07–6.62  0.036  3.66  1.07–12.50  0.038  Female gender  2.66  1.13–6.29  0.025  1.44  0.37–5.51  0.596  Serum creatinine level >120 µmol/l  2.46  0.98–6.17  0.054  1.03  0.28–3.85  0.970  NYHA Class III or IV  11.36  2.58–50.00  0.001  15.87  1.90–142.86  0.011  LVEF  1.03  0.99–1.07  0.128  1.02  0.96–1.07  0.559  Non-elective surgery  2.44  0.88–6.76  0.086  2.65  0.60–16.39  0.174  sPAP≥ 55 mmHg  3.57  1.26–10.10  0.017  1.58  0.42–5.99  0.501  Tricuspid annuloplasty  3.41  1.05–11.11  0.042  3.69  0.74–18.52  0.112  MVR  3.70  1.41–9.71  0.008  5.65  1.66–19.23  0.006  Injury during resternotomy  6.85  1.18–40.00  0.032  16.95  1.74–166.67  0.015    Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Endocarditis  2.65  1.07–6.62  0.036  3.66  1.07–12.50  0.038  Female gender  2.66  1.13–6.29  0.025  1.44  0.37–5.51  0.596  Serum creatinine level >120 µmol/l  2.46  0.98–6.17  0.054  1.03  0.28–3.85  0.970  NYHA Class III or IV  11.36  2.58–50.00  0.001  15.87  1.90–142.86  0.011  LVEF  1.03  0.99–1.07  0.128  1.02  0.96–1.07  0.559  Non-elective surgery  2.44  0.88–6.76  0.086  2.65  0.60–16.39  0.174  sPAP≥ 55 mmHg  3.57  1.26–10.10  0.017  1.58  0.42–5.99  0.501  Tricuspid annuloplasty  3.41  1.05–11.11  0.042  3.69  0.74–18.52  0.112  MVR  3.70  1.41–9.71  0.008  5.65  1.66–19.23  0.006  Injury during resternotomy  6.85  1.18–40.00  0.032  16.95  1.74–166.67  0.015  LVEF: left ventricular ejection fraction; MVR: mitral valve replacement; NYHA: New York Heart Association; OR: odds ratio; sPAP: systolic pulmonary artery pressure. Figure 4: View largeDownload slide ROC of EuroSCORE II and STS mortality score for in-hospital mortality. AUC: area under the curve; ROC: receiver operating characteristics; STS: Society of Thoracic Surgeons. Figure 4: View largeDownload slide ROC of EuroSCORE II and STS mortality score for in-hospital mortality. AUC: area under the curve; ROC: receiver operating characteristics; STS: Society of Thoracic Surgeons. In Cox multivariable regression analysis, preoperative New York Heart Association (NYHA) Class III/IV, concomitant MVR and PVD were independent factors affecting long-term survival (Table 4). Table 4: Univariable and multivariable analyses of long-term mortality   Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Serum creatinine level >120 µmol/l  2.29  1.28–4.12  0.005  1.52  0.76–3.06  0.237  Group 3  2.57  0.93–7.14  0.070  2.56  0.84–7.81  0.098  NYHA Class III/IV  3.19  1.65–6.19  0.001  2.31  1.10–4.83  0.027  sPAP≥55 mmHg  3.24  1.30–4.93  0.006  1.48  0.70–3.13  0.308  MVR  2.99  1.58–5.62  0.006  2.82  1.37–5.78  0.005  Endocarditis  1.73  0.92–3.25  0.090  2.00  0.91–4.23  0.086  PVD  2.21  1.16–4.22  0.016  2.19  1.01–4.74  0.047    Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Serum creatinine level >120 µmol/l  2.29  1.28–4.12  0.005  1.52  0.76–3.06  0.237  Group 3  2.57  0.93–7.14  0.070  2.56  0.84–7.81  0.098  NYHA Class III/IV  3.19  1.65–6.19  0.001  2.31  1.10–4.83  0.027  sPAP≥55 mmHg  3.24  1.30–4.93  0.006  1.48  0.70–3.13  0.308  MVR  2.99  1.58–5.62  0.006  2.82  1.37–5.78  0.005  Endocarditis  1.73  0.92–3.25  0.090  2.00  0.91–4.23  0.086  PVD  2.21  1.16–4.22  0.016  2.19  1.01–4.74  0.047  MVR: mitral valve replacement; NYHA: New York Heart Association; OR: odds ratio; PVD: prosthetic valve dehiscence; sPAP: systolic pulmonary artery pressure. Table 4: Univariable and multivariable analyses of long-term mortality   Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Serum creatinine level >120 µmol/l  2.29  1.28–4.12  0.005  1.52  0.76–3.06  0.237  Group 3  2.57  0.93–7.14  0.070  2.56  0.84–7.81  0.098  NYHA Class III/IV  3.19  1.65–6.19  0.001  2.31  1.10–4.83  0.027  sPAP≥55 mmHg  3.24  1.30–4.93  0.006  1.48  0.70–3.13  0.308  MVR  2.99  1.58–5.62  0.006  2.82  1.37–5.78  0.005  Endocarditis  1.73  0.92–3.25  0.090  2.00  0.91–4.23  0.086  PVD  2.21  1.16–4.22  0.016  2.19  1.01–4.74  0.047    Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Serum creatinine level >120 µmol/l  2.29  1.28–4.12  0.005  1.52  0.76–3.06  0.237  Group 3  2.57  0.93–7.14  0.070  2.56  0.84–7.81  0.098  NYHA Class III/IV  3.19  1.65–6.19  0.001  2.31  1.10–4.83  0.027  sPAP≥55 mmHg  3.24  1.30–4.93  0.006  1.48  0.70–3.13  0.308  MVR  2.99  1.58–5.62  0.006  2.82  1.37–5.78  0.005  Endocarditis  1.73  0.92–3.25  0.090  2.00  0.91–4.23  0.086  PVD  2.21  1.16–4.22  0.016  2.19  1.01–4.74  0.047  MVR: mitral valve replacement; NYHA: New York Heart Association; OR: odds ratio; PVD: prosthetic valve dehiscence; sPAP: systolic pulmonary artery pressure. DISCUSSION Currently, as the number of heart operations increases, multiple redo valve surgeries are becoming increasingly common. In a large series of patients with various reasons for reintervention, we demonstrated that outcomes of redo-AV surgery were more influenced by preoperative characteristics such as symptoms at presentation, aetiology and the need for concomitant MV surgery than the number of sternal re-entries. Our overall operative mortality results were in accordance with previous reports by Akins et al. [4] and Jones et al. [5]. Groups 1 and 2, representing patients undergoing redo- and re-redo-AV interventions, respectively, shared a similar aetiological pattern. They also had similar complication rates and outcomes. Therefore, a second redo procedure on the AV did not appear to increase morbidity compared with a single redo procedure with respect to the reason for reintervention. Similar findings have been reported by Tyers et al. [6] and recently by Davierwala et al. [7]. We also found that patients undergoing a fourth or more AV intervention (Group 3) seemed to have a higher all-complications rate but the limited number of patients in this group restricts any comparison with the other 2 groups. Moreover, patients in Group 3 were often referred late to surgery with more altered LV functions and higher pulmonary pressures at presentation. One of the major concerns regarding cardiac reinterventions is injury at resternotomy. Our incidence of re-entry injury was smaller than those reported by Roselli et al. [8] or Ellman et al. [9]. As recommended by Morishita et al. [10], a systematic preoperative investigation by a computed tomography scan in our patients who had previously undergone aortic surgery or coronary artery bypass grafting was certainly helpful in localizing the potential hazards during dissection and assessing the spatial relationship between the inner aspect of the sternum and the underlying mediastinal, cardiac and vascular structures. Nevertheless, as described previously by these authors, we found that the number of re-entry injuries was influenced by the number of re-sternotomies and had an impact on operative mortality. The presence of patent internal thoracic artery grafts may also compromise optimal myocardial protection if not controlled and has been assigned as an independent risk factor for injury by Park et al. [11] in an analysis of more than 2500 cardiac reinterventions. When the internal thoracic artery was considered impossible to be controlled, the so-called ‘open internal thoracic artery’ technique described by Vistarini et al. [12] has been used for years by some of the surgeons of our department with good results. There were no injuries to the coronary grafts reported in our study cohort. Furthermore, as already reported by Khaladj et al. [13], the presence of previous coronary artery bypass grafting was not related to adverse outcomes. Combined MVR, as opposed to mitral valve repair, has been known to negatively affect mortality and morbidity during AVR [14, 15]. Our results confirmed this established fact in redo-AV surgery. Furthermore, we found that concomitant MVR was also a risk factor for long-term mortality. In contrast, classical combined procedures such as aortic root replacement, coronary artery bypass grafting (described earlier) or tricuspid valve surgery were not associated with increased mortality in our study. The risk related to the latter procedure certainly may have benefitted from the large proportion of prophylactic tricuspid annuloplasties performed in our department, based more on the annular diameter rather than the dysfunction severity of the tricuspid valve. Although non-elective surgery was not associated with increased in-hospital mortality in our study, we found that urgency remained frequent in the population undergoing AV reinterventions. In addition, more than one-third of our patients presented with severe preoperative dyspnoea, which was an independent risk factor for operative mortality and affected long-term prognosis. This appeals for earlier timing, whenever possible, in addressing patients who require AV reintervention, as already advocated by Vogt et al. [16], almost 20 years ago. The various causes for reintervention in our analysis showcased very disparate postoperative course profiles. For endocarditis, which exclusively appeared on prosthetic valves, we reported an in-hospital mortality rate as high as 14% and an odds ratio of 3.66. This result was rationally lower than most of the other mortality rates reported in the literature [17, 18] but confirms that aortic prosthetic valve endocarditis is a serious disease, for which surgery remains challenging. Although we did not demonstrate a real statistical significance, it may also have had an influence on long-term prognosis. In our experience, PVD accounted for 12.6% of the cases for AV reintervention and the proportion appeared to increase with the number of reoperations. Although defined as a classical complication after AV replacement, occurrence of non-infectious PVD has been reduced to 2.2% in a recent report by Duncan et al. [19]. However, in terms of surgical mortality, PVD remains associated with a 10% mortality rate that confirmed the results recently published by Taramasso et al. [20]. Moreover, our data demonstrated that PVD was an independent risk factor for long-term all-cause mortality. Thus, we believe that alternative approaches such as transcatheter closure using vascular plugs should be considered in this setting, as early results are encouraging. In comparison, SVD, the leading cause for reintervention in this study, was associated with excellent short- and long-term outcomes. Our data confirmed the general trend towards a reduction of mortality in patients requiring AV reintervention for SVD [21]. This has mainly 2 implications: first, it could serve as a benchmark to compare with transcatheter valve-in valve therapy, which has recently been shown to have some drawbacks. Indeed, the recent retrospective analysis of the Global Valve-in-Valve Registry from Dvir et al. [22] has shown a 15.3% rate of valve malposition and a 3.5% rate of ostial obstruction complications that are rare in open redo-AVR. A single-centre Canadian study also demonstrated that patients receiving a valve-in-valve procedure following bioprosthetic SVD had a lower survival at 1 year [3]. Therefore, particularly in a younger population, transcatheter aortic valve implantation may not be a suitable option for SVD at the moment. Secondly, our results regarding redo-AVR for SVD may have an implication on the choice of prosthetic valve. Indeed, the main factors in making this choice include the risk of anticoagulation, thromboembolism and bleeding events that could occur with mechanical valves, whereas bioprosthetic valves are not as durable and thus have a risk of SVD leading to reintervention [23, 24]. In fact, a recent study by Glaser et al. [25] noted that patients between 50 and 69 years undergoing the first ever intervention to their AV and who opted for a bioprosthetic valve had a significantly higher risk of reintervention compared to those opting for a mechanical AV (hazard ratio 2.36). Thus, at present, mechanical valves are usually offered to patients younger than 60 years [26]. However, Ruel et al. [27] have demonstrated that selection of bioprosthetic valves in patients below 60 years of age had no impact on 30-year survival. Furthermore, our data suggested that the first and second AV reinterventions for SVD have a similar low operative risk, close to the initial AVR surgical procedure [28]. This is in line with Spampinato et al. [29] who suggested, 20 years ago, that the choice of a bioprosthetic valve for a redo surgery following SVD was not hazardous. Limitations The limitations of this study include its inherent retrospective design and analysis of observational data. Furthermore, the groups do not comprise a similar number of patients, particularly Group 3, with 11 patients over 10 years compared with 263 and 42 in Groups 1 and 2, respectively. However, this would be reflective of true clinical practice as it is not common to have patients presenting for a third or more redo-AVRs, and a much longer, and larger, study incorporating more than one facility is possibly required to obtain higher numbers. Finally, the Cardiac STS Score could not be utilized for those undergoing concomitant AVR and mitral valve procedures, and for this reason, the EuroSCORE II was also used; perhaps future models of prediction could incorporate this. CONCLUSIONS Multiple AV reinterventions carry an acceptable risk of in-hospital mortality, especially in the setting of isolated AVR for SVD of a bioprosthetic valve. These findings may favour the use of a biological valve substitute for AVR in young patients. Furthermore, they could serve as a benchmark for defining the role of valve-in-valve implantation in the setting of multiple AV reinterventions, which is still unknown. Given the inherently higher risks of AV reintervention associated with the transcatheter approaches, the latter may be considered more selectively in older patients with high preoperative risk well assessed by the EuroSCORE II and STS mortality scores. ACKNOWLEDGEMENTS We thank the surgeons and staff of the Cardiovascular Department at the Georges Pompidou European Hospital. Conflict of interest: none declared. REFERENCES 1 Leontyev S, Borger MA, Davierwala P, Walther T, Lehmann S, Kempfert J et al.   Redo aortic valve surgery: early and late outcomes. Ann Thorac Surg  2011; 91: 1120– 6. Google Scholar CrossRef Search ADS PubMed  2 Balsam LB, Grossi EA, Greenhouse DG, Ursomanno P, Deanda A, Ribakove GH et al.   Reoperative valve surgery in the elderly: predictors of risk and long-term survival. Ann Thorac Surg  2010; 90: 1195– 200. Google Scholar CrossRef Search ADS PubMed  3 Erlebach M, Wottke M, Deutsch M-A, Krane M, Piazza N, Lange R et al.   Redo aortic valve surgery versus transcatheter valve-in-valve implantation for failing surgical bioprosthetic valves: consecutive patients in a single-center setting. J Thorac Dis  2015; 7: 1494– 500. Google Scholar PubMed  4 Akins CW, Buckley MJ, Daggett WM, Hilgenberg AD, Vlahakes GJ, Torchiana DF et al.   Risk of reoperative valve replacement for failed mitral and aortic bioprostheses. Ann Thorac Surg  1998; 65: 1545– 51. Google Scholar CrossRef Search ADS PubMed  5 Jones JM, O’Kane H, Gladstone DJ, Sarsam MA, Campalani G, MacGowan SW et al.   Repeat heart valve surgery: risk factors for operative mortality. J Thorac Cardiovasc Surg  2001; 122: 913– 8. Google Scholar CrossRef Search ADS PubMed  6 Tyers GF, Jamieson WR, Munro AI, Germann E, Burr LH, Miyagishima RT et al.   Reoperation in biological and mechanical valve populations: fate of the reoperative patient. Ann Thorac Surg  1995; 60: S464– 8. Google Scholar CrossRef Search ADS PubMed  7 Davierwala PM, Borger MA, David TE, Rao V, Maganti M, Yau TM. Reoperation is not an independent predictor of mortality during aortic valve surgery. J Thorac Cardiovasc Surg  2006; 131: 329– 35. Google Scholar CrossRef Search ADS PubMed  8 Roselli EE, Pettersson GB, Blackstone EH, Brizzio ME, Houghtaling PL, Hauck R et al.   Adverse events during reoperative cardiac surgery: frequency, characterization, and rescue. J Thorac Cardiovasc Surg  2008; 135: 316– 23; 323.e1–6. Google Scholar CrossRef Search ADS PubMed  9 Ellman PI, Smith RL, Girotti ME, Thompson PW, Peeler BB, Kern JA et al.   Cardiac injury during resternotomy does not affect perioperative mortality. J Am Coll Surg  2008; 206: 993– 7. Google Scholar CrossRef Search ADS PubMed  10 Morishita K, Kawaharada N, Fukada J, Yamada A, Masaru T, Kuwaki K et al.   Three or more median sternotomies for patients with valve disease: role of computed tomography. Ann Thorac Surg  2003; 75: 1476– 80. Google Scholar CrossRef Search ADS PubMed  11 Park CB, Suri RM, Burkhart HM, Greason KL, Dearani JA, Schaff HV et al.   Identifying patients at particular risk of injury during repeat sternotomy: analysis of 2555 cardiac reoperations. J Thorac Cardiovasc Surg  2010; 140: 1028– 35. Google Scholar CrossRef Search ADS PubMed  12 Vistarini N, Aubert S, Leprince P, Pavie A. A simplified surgical approach for aortic valve replacement after previous coronary artery bypass grafting. Eur J Cardiothorac Surg  2009; 36: 404– 6. Google Scholar CrossRef Search ADS PubMed  13 Khaladj N, Shrestha M, Peterss S, Kutschka I, Strueber M, Hoy L et al.   Isolated surgical aortic valve replacement after previous coronary artery bypass grafting with patent grafts: is this old-fashioned technique obsolete? Eur J Cardiothorac Surg  2009; 35: 260– 4. Google Scholar CrossRef Search ADS PubMed  14 Rankin JS, Hammill BG, Ferguson TB, Glower DD, O’Brien SM, DeLong ER et al.   Determinants of operative mortality in valvular heart surgery. J Thorac Cardiovasc Surg  2006; 131: 547– 57. Google Scholar CrossRef Search ADS PubMed  15 Kim GS, Kim JB, Han S, Choo SJ, Chung CH, Lee JW et al.   Mitral valve repair versus replacement for moderate-to-severe mitral regurgitation in patients undergoing concomitant aortic valve replacement. Interact CardioVasc Thorac Surg  2014; 18: 73– 9. Google Scholar CrossRef Search ADS PubMed  16 Vogt PR, Brunner-LaRocca H, Sidler P, Zünd G, Truniger K, Lachat M et al.   Reoperative surgery for degenerated aortic bioprostheses: predictors for emergency surgery and reoperative mortality. Eur J Cardiothorac Surg  2000; 17: 134– 9. Google Scholar CrossRef Search ADS PubMed  17 Luciani N, Mossuto E, Ricci D, Luciani M, Russo M, Salsano A et al.   Prosthetic valve endocarditis: predictors of early outcome of surgical therapy. A multicentric study. Eur J Cardiothorac Surg  2017; 52: 768– 74. Google Scholar CrossRef Search ADS PubMed  18 Alonso-Valle H, Fariñas-Álvarez C, García-Palomo JD, Bernal JM, Martín-Durán R, Díez JFG et al.   Clinical course and predictors of death in prosthetic valve endocarditis over a 20-year period. J Thorac Cardiovasc Surg  2010; 139: 887– 93. Google Scholar CrossRef Search ADS PubMed  19 Duncan BF, McCarthy PM, Kruse J, Andrei A-C, Li Z, Russell HM et al.   Paravalvular regurgitation after conventional aortic and mitral valve replacement: a benchmark for alternative approaches. J Thorac Cardiovasc Surg  2015; 150: 860– 8.e1. Google Scholar CrossRef Search ADS PubMed  20 Taramasso M, Maisano F, Denti P, Guidotti A, Sticchi A, Pozzoli A et al.   Surgical treatment of paravalvular leak: long-term results in a single-center experience (up to 14 years). J Thorac Cardiovasc Surg  2015; 149: 1270– 5. Google Scholar CrossRef Search ADS PubMed  21 Jamieson WRE, Burr LH, Miyagishima RT, Janusz MT, Fradet GJ, Ling H et al.   Re-operation for bioprosthetic aortic structural failure-risk assessment. Eur J Cardiothorac Surg  2003; 24: 873– 8. Google Scholar CrossRef Search ADS PubMed  22 Dvir D, Webb J, Brecker S, Bleiziffer S, Hildick-Smith D, Colombo A et al.   Transcatheter aortic valve replacement for degenerative bioprosthetic surgical valves: results from the global valve-in-valve registry. Circulation  2012; 126: 2335– 44. Google Scholar CrossRef Search ADS PubMed  23 Head SJ, Çelik M, Kappetein AP. Mechanical versus bioprosthetic aortic valve replacement. Eur Heart J  2017; 38: 2183– 91. Google Scholar CrossRef Search ADS PubMed  24 Kulik A, Bédard P, Lam B-K, Rubens FD, Hendry PJ, Masters RG et al.   Mechanical versus bioprosthetic valve replacement in middle-aged patients. Eur J Cardiothorac Surg  2006; 30: 485– 91. Google Scholar CrossRef Search ADS PubMed  25 Glaser N, Jackson V, Holzmann MJ, Franco-Cereceda A, Sartipy U. Aortic valve replacement with mechanical vs. biological prostheses in patients aged 50-69 years. Eur Heart J  2016; 37: 2658– 67. Google Scholar CrossRef Search ADS PubMed  26 Sakamoto Y, Yoshitake M, Matsumura Y, Naruse H, Bando K, Hashimoto K. Choice of aortic valve prosthesis in a rapidly aging and long-living society. Ann Thorac Cardiovasc Surg  2016; 22: 333– 9. Google Scholar CrossRef Search ADS PubMed  27 Ruel M, Chan V, Bédard P, Kulik A, Ressler L, Lam BK et al.   Very long-term survival implications of heart valve replacement with tissue versus mechanical prostheses in adults <60 years of age. Circulation  2007; 116: I294– 300. Google Scholar CrossRef Search ADS PubMed  28 Potter DD, Sundt TM, Zehr KJ, Dearani JA, Daly RC, Mullany CJ et al.   Operative risk of reoperative aortic valve replacement. J Thorac Cardiovasc Surg  2005; 129: 94– 103. Google Scholar CrossRef Search ADS PubMed  29 Spampinato N, Gagliardi C, Pantaleo D, Fimiani L, Ascione R, De Robertis F et al.   Bioprosthetic replacement after bioprosthesis failure: a hazardous choice? Ann Thorac Surg  1998; 66: S68– 72. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. 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

Multiple reoperations on the aortic valve: outcomes and implications for future potential valve-in-valve strategy

Loading next page...
 
/lp/ou_press/multiple-reoperations-on-the-aortic-valve-outcomes-and-implications-hUB9E2fv8p
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. 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/ezx469
Publisher site
See Article on Publisher Site

Abstract

Abstract OBJECTIVES Surgical mortality and long-term outcomes are important considerations when determining strategies for multiple reoperations on the aortic valve (AV). With the rise of percutaneous valve-in-valve, we sought to evaluate the current outcomes of conventional surgery for AV reoperation, focusing first on the effect of the number of previous AV interventions with a subsequent analysis of the risk factors for adverse outcomes. METHODS From January 2007 to December 2016, 316 consecutive patients underwent an open redo operation (replacement) on their AV at a single centre. It was the first AV reintervention in 263 patients (Group 1), second in 42 patients (Group 2) and third or more in 11 patients (Group 3). RESULTS There were 230 men and 86 women, with a median age of 58 (Q1–Q3: 46–70) years. Structural valve deterioration (SVD) of the bioprosthesis (n = 136, 44%), endocarditis (n = 57, 18%) and prosthetic valve dehiscence (n = 41, 13%) were the most common reasons for reintervention. Overall, in-hospital mortality was 7.3%: 7.2% in Group 1, 4.76% in Group 2 and 18.2% in Group 3 (P = 0.233) and ranged from 3.7% for SVD to 14.0% when endocarditis was the reason for reintervention. Higher preoperative New York Heart Association (NYHA) class (III/IV) [odds ratio (OR) 15.9, P = 0.011], injury during re-entry (OR 16.9, P = 0.015), endocarditis (OR 3.7, P = 0.038) and concomitant mitral valve replacement (OR 5.6, P = 0.006) were independent risk factors for in-hospital mortality. Survival at 8 years was 79.0 ± 3.0% for the entire cohort and 88.4 ± 3.2% for re-replacement after SVD. CONCLUSIONS Multiple AV reoperations carry an acceptable risk of early postoperative mortality, particularly for isolated valve replacements of SVD. Aortic valve replacement, Redo aortic valve, Aortic valve, Bioprosthesis INTRODUCTION Redo-aortic valve replacement (AVR) following the first procedure can be challenging. A combination of increased life expectancy post-valvular surgery, an ageing population (particularly in Western societies) and an increasing preference for bioprosthetic valves has led to an increasing number of surgical patients presenting for redo valve replacements [1, 2]. To date, surgical open AVR remains the gold standard for managing aortic valve (AV) disease. With a 1% incidence in the USA, AV disease can include pathologies from the stenotic valve to the regurgitant valve and rheumatic heart disease. It represents a significant health burden, particularly in the elderly population. Currently, transcatheter valve-in-valve implantation is considered largely in patients who are at high operative risk; however, in this current paradigm of bioprosthetic structural valve deterioration (SVD), it is a management option that is gaining momentum. Indeed, a retrospective single-centre study by Erlebach et al. [3] suggested no significant difference in the 30-day mortality in patients who underwent transcatheter valve-in-valve implantation after previous open valve replacement compared with patients undergoing traditional operative redo-AVR. However, 1-year survival was significantly better in patients who underwent a redo open surgery. Moreover, when it comes to patients who are undergoing a second or a third or more AVR, it is not clear at this stage which valve substitute and which strategy would be best suited. We sought to evaluate the postoperative mortality and morbidity of patients undergoing redo-AVR for the first, second, third or more times in the last decade, with the intention of using the results to inform management decisions as patients increasingly present for redo procedures. MATERIALS AND METHODS This is a retrospective 10-year period study conducted between January 2007 and December 2016. We identified 316 consecutive patients at the Georges Pompidou European Hospital in Paris, France, who underwent open redo surgery on their AV. Patients were included in this study if they previously had an open surgery to their AV. Non-AV open-heart surgeries were not considered in the count of previous interventions to the AV. The study cohort was divided into 3 groups based on the number of current reinterventions to their AV: Group 1, patients with a first redo intervention (n = 263); Group 2, patients with a second redo intervention (n = 42); and Group 3, patients with a third redo intervention or beyond (n = 11). Once a patient was included in the study, he/she was not counted again if reintervention occurred during the follow-up period, thus keeping the samples independent for statistical analysis. The study was approved by the institutional review board. AVR in the study cohort was performed via the usual open surgical manner, reprising the median sternotomy scar from previous operations. In-hospital data were collected retrospectively from medical records. Follow-up was done through direct phone contact with the referring cardiologist and the patient. Mean follow-up period at the completion of the study was 60 ± 30 months (calculation of means made with censoring deceased patients at the time of death). Statistical analysis Descriptive statistics were expressed as median and 25–75% quartiles [median (Q1–Q3)] for continuous variables and numbers (proportions) for categorical variables. To identify the significant differences in preoperative and intraoperative categorical data and 30-day mortality between groups, a χ2 test was used, except when the expected contingent cell in cross-tabulation was <5, in which case a Fisher’s exact test was preferred. One-way analysis of variance or non-parametric Kruskal–Wallis test was also used for differences in continuous data, depending on the normality distribution of the variables within groups. Estimates for long-term survival and recurrence of any type of AV reintervention (including percutaneous procedures) were made using the Kaplan–Meier method. Considering the total study cohort, preoperative data of the patients were then entered into a stepwise binary logistic regression analysis to identify the independent predictors for in-hospital mortality. The appropriateness of the tested variables was first determined by univariable analysis using the χ2 or Fisher’s exact test for categorical data and the Student’s t-test for continuous data with a P-value <0.2. The log-rank test was used to determine univariable predictive factors of survival. Each variable with a P-value <0.2 was entered into a Cox proportional hazard regression multivariable analysis with a backward stepwise selection process to identify the independent multivariable predictive factors of long-term mortality. Statistical analyses were performed using IBM SPSS Statistics v22 software (IBM Corp, Armonk, NY, USA). RESULTS There were 230 men and 86 women with a median age of 58 (46–70) years. Baseline characteristics of the 3 groups based on the number of AV reinterventions are listed in Table 1. Compared with Group 3, patients in Group 1 had a lower mean systolic pulmonary arterial pressure and a higher mean left ventricular ejection fraction. Reasons for reintervention were SVD in 43.0% (n = 136), AV endocarditis in 18.0% (n = 57), prosthetic valve dehiscence (PVD) in 13.0% (n = 41), recurrent AV disease after previous valve repairs in 11.1% (n = 35), mechanical valve dysfunction in 8.9% (n = 28) and redo-Bentall procedure for aortic aneurysm in 6.0% (n = 19). Details of the reasons for reintervention in each group are shown in Fig. 1. Table 1: Baseline characteristics of the study population based on the number of previous AV interventions categorized by groups   No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Age (years), median (Q1–Q3)  58 (46–71)  58 (48–68)  50 (42–59)  Gender, n (%)   Female  74 (28.1)  9 (21.4)  3 (27.3)   Male  189 (71.9)  33 (78.6)  8 (72.7)  Serum creatinine (µmol/l), median (Q1–Q3)  88 (75–107)  102 (83–120)  93 (75–105)  COPD, n (%)  8 (3.0)  1 (2.4)  0 (0.0)  Diabetes, n (%)  18 (6.9)  3 (7.1)  0 (0.0)  NYHA class, n (%)   I  27 (10.3)  2 (4.8)  1 (9.1)   II  115 (43.7)  10 (23.8)  2 (18.2)   III  91 (34.6)  15 (35.7)  5 (45.5)   IV  30 (11.4)  8 (19.0)  2 (18.2)  LVEF (%), median (Q1–Q3)  60 (55–65)  58 (50–65)  50 (35–65)*  sPAP (mmHg), median (Q1–Q3)  35 (25–40)  35 (30–45)  50 (35–65)*  Urgency, n (%)   Elective  102 (38.8)  21 (50.0)  5 (45.5)   Urgent  121 (46.0)  15 (35.7)  4 (36.4)   Emergent  35 (13.3)  6 (14.3)  2 (18.2)   Salvage  5 (1.9)  0 (0.0)  0 (0.0)  EuroSCORE II, median (Q1–Q3)  7.12 (3.80–12.40)  7.56 (4.14–14.97)  5.99 (2.49–15.99)  STS mortality score, median (Q1–Q3)  1.80 (1.24–3.08)  2.22 (1.38–3.09)  2.18 (1.18–2.49)    No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Age (years), median (Q1–Q3)  58 (46–71)  58 (48–68)  50 (42–59)  Gender, n (%)   Female  74 (28.1)  9 (21.4)  3 (27.3)   Male  189 (71.9)  33 (78.6)  8 (72.7)  Serum creatinine (µmol/l), median (Q1–Q3)  88 (75–107)  102 (83–120)  93 (75–105)  COPD, n (%)  8 (3.0)  1 (2.4)  0 (0.0)  Diabetes, n (%)  18 (6.9)  3 (7.1)  0 (0.0)  NYHA class, n (%)   I  27 (10.3)  2 (4.8)  1 (9.1)   II  115 (43.7)  10 (23.8)  2 (18.2)   III  91 (34.6)  15 (35.7)  5 (45.5)   IV  30 (11.4)  8 (19.0)  2 (18.2)  LVEF (%), median (Q1–Q3)  60 (55–65)  58 (50–65)  50 (35–65)*  sPAP (mmHg), median (Q1–Q3)  35 (25–40)  35 (30–45)  50 (35–65)*  Urgency, n (%)   Elective  102 (38.8)  21 (50.0)  5 (45.5)   Urgent  121 (46.0)  15 (35.7)  4 (36.4)   Emergent  35 (13.3)  6 (14.3)  2 (18.2)   Salvage  5 (1.9)  0 (0.0)  0 (0.0)  EuroSCORE II, median (Q1–Q3)  7.12 (3.80–12.40)  7.56 (4.14–14.97)  5.99 (2.49–15.99)  STS mortality score, median (Q1–Q3)  1.80 (1.24–3.08)  2.22 (1.38–3.09)  2.18 (1.18–2.49)  * P = 0.01: non-parametric Kruskal–Wallis test (Group 3 vs Group 1). AV: aortic valve; COPD: chronic obstructive pulmonary disease; LVEF: left ventricular ejection fraction; NYHA: New York Heart Association; Q1: 25% quartile; Q3: 75% quartile; sPAP: systolic pulmonary arterial pressure; STS: Society of Thoracic Surgeons. Table 1: Baseline characteristics of the study population based on the number of previous AV interventions categorized by groups   No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Age (years), median (Q1–Q3)  58 (46–71)  58 (48–68)  50 (42–59)  Gender, n (%)   Female  74 (28.1)  9 (21.4)  3 (27.3)   Male  189 (71.9)  33 (78.6)  8 (72.7)  Serum creatinine (µmol/l), median (Q1–Q3)  88 (75–107)  102 (83–120)  93 (75–105)  COPD, n (%)  8 (3.0)  1 (2.4)  0 (0.0)  Diabetes, n (%)  18 (6.9)  3 (7.1)  0 (0.0)  NYHA class, n (%)   I  27 (10.3)  2 (4.8)  1 (9.1)   II  115 (43.7)  10 (23.8)  2 (18.2)   III  91 (34.6)  15 (35.7)  5 (45.5)   IV  30 (11.4)  8 (19.0)  2 (18.2)  LVEF (%), median (Q1–Q3)  60 (55–65)  58 (50–65)  50 (35–65)*  sPAP (mmHg), median (Q1–Q3)  35 (25–40)  35 (30–45)  50 (35–65)*  Urgency, n (%)   Elective  102 (38.8)  21 (50.0)  5 (45.5)   Urgent  121 (46.0)  15 (35.7)  4 (36.4)   Emergent  35 (13.3)  6 (14.3)  2 (18.2)   Salvage  5 (1.9)  0 (0.0)  0 (0.0)  EuroSCORE II, median (Q1–Q3)  7.12 (3.80–12.40)  7.56 (4.14–14.97)  5.99 (2.49–15.99)  STS mortality score, median (Q1–Q3)  1.80 (1.24–3.08)  2.22 (1.38–3.09)  2.18 (1.18–2.49)    No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Age (years), median (Q1–Q3)  58 (46–71)  58 (48–68)  50 (42–59)  Gender, n (%)   Female  74 (28.1)  9 (21.4)  3 (27.3)   Male  189 (71.9)  33 (78.6)  8 (72.7)  Serum creatinine (µmol/l), median (Q1–Q3)  88 (75–107)  102 (83–120)  93 (75–105)  COPD, n (%)  8 (3.0)  1 (2.4)  0 (0.0)  Diabetes, n (%)  18 (6.9)  3 (7.1)  0 (0.0)  NYHA class, n (%)   I  27 (10.3)  2 (4.8)  1 (9.1)   II  115 (43.7)  10 (23.8)  2 (18.2)   III  91 (34.6)  15 (35.7)  5 (45.5)   IV  30 (11.4)  8 (19.0)  2 (18.2)  LVEF (%), median (Q1–Q3)  60 (55–65)  58 (50–65)  50 (35–65)*  sPAP (mmHg), median (Q1–Q3)  35 (25–40)  35 (30–45)  50 (35–65)*  Urgency, n (%)   Elective  102 (38.8)  21 (50.0)  5 (45.5)   Urgent  121 (46.0)  15 (35.7)  4 (36.4)   Emergent  35 (13.3)  6 (14.3)  2 (18.2)   Salvage  5 (1.9)  0 (0.0)  0 (0.0)  EuroSCORE II, median (Q1–Q3)  7.12 (3.80–12.40)  7.56 (4.14–14.97)  5.99 (2.49–15.99)  STS mortality score, median (Q1–Q3)  1.80 (1.24–3.08)  2.22 (1.38–3.09)  2.18 (1.18–2.49)  * P = 0.01: non-parametric Kruskal–Wallis test (Group 3 vs Group 1). AV: aortic valve; COPD: chronic obstructive pulmonary disease; LVEF: left ventricular ejection fraction; NYHA: New York Heart Association; Q1: 25% quartile; Q3: 75% quartile; sPAP: systolic pulmonary arterial pressure; STS: Society of Thoracic Surgeons. Figure 1: View largeDownload slide Proportion of reasons for reintervention based on the number of previous aortic valve surgeries categorized by groups. AVRr: aortic valve replacement after previous AV repair; Bentall redo: redo Bentall procedure; MVD: mechanical valve dysfunction; PVD: prosthetic valve dehiscence; SVD: structural valve deterioration of a bioprosthesis. Figure 1: View largeDownload slide Proportion of reasons for reintervention based on the number of previous aortic valve surgeries categorized by groups. AVRr: aortic valve replacement after previous AV repair; Bentall redo: redo Bentall procedure; MVD: mechanical valve dysfunction; PVD: prosthetic valve dehiscence; SVD: structural valve deterioration of a bioprosthesis. Intraoperative characteristics Descriptive data regarding surgical procedures specific for each group are listed in Table 2. All patients underwent an AV replacement. A total of 192 (60.8%) patients underwent one or more concomitant procedures, with 29 (9.3%) patients undergoing tricuspid valve repair or replacement, 38 (12.3%) a mitral valve replacement (MVR), 38 (12.0%) a replacement of the ascending aorta and 49 (15.5%) a redo-Bentall procedure. Concomitant coronary artery bypass grafting was also performed in 18 (5.7%) patients. Table 2: Intraoperative characteristics of the study population based on the number of previous AV interventions categorized by groups   No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Cross-clamp time (min), median (Q1–Q3)  93 (67–127)  90 (72–150)  95 (85–117)  CPB time (min), median (Q1–Q3)  125 (93–179)  132 (97–212)  180 (150–211)  Circulatory arrest, n (%)  21 (8.0)  3 (7.1)  1 (9.1)  AVR prosthesis type, n (%)   Mechanical valve  108 (41.1)  19 (45.2)  6 (54.5)   Bioprosthesis  148 (56.3)  22 (52.4)  5 (45.5)   Homograft  7 (2.6)  1 (2.4)  0 (0.0)  CABG, n (%)  16 (6.1)  2 (4.8)  0 (0.0)  MVR, n (%)  28 (10.6)  9 (21.4)  1 (9.1)  MV repair, n (%)  9 (3.4)  1 (2.4)  1 (9.1)  PVR, n (%)  1 (0.4)  1 (2.4)  0 (0.0)  TVR, n (%)  4 (1.5)  3 (7.1)  1 (9.1)  TV repair, n (%)  18 (6.8)  2 (4.8)  1 (9.1)  Ascending aorta replacement, n (%)  32 (12.2)  6 (14.3)  0 (0.0)  Aortic root enlargement, n (%)  14 (5.3)  5 (11.9)  1 (9.1)  Bentall, n (%)  41 (16.0)  6 (14.3)  2 (18.2)    No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Cross-clamp time (min), median (Q1–Q3)  93 (67–127)  90 (72–150)  95 (85–117)  CPB time (min), median (Q1–Q3)  125 (93–179)  132 (97–212)  180 (150–211)  Circulatory arrest, n (%)  21 (8.0)  3 (7.1)  1 (9.1)  AVR prosthesis type, n (%)   Mechanical valve  108 (41.1)  19 (45.2)  6 (54.5)   Bioprosthesis  148 (56.3)  22 (52.4)  5 (45.5)   Homograft  7 (2.6)  1 (2.4)  0 (0.0)  CABG, n (%)  16 (6.1)  2 (4.8)  0 (0.0)  MVR, n (%)  28 (10.6)  9 (21.4)  1 (9.1)  MV repair, n (%)  9 (3.4)  1 (2.4)  1 (9.1)  PVR, n (%)  1 (0.4)  1 (2.4)  0 (0.0)  TVR, n (%)  4 (1.5)  3 (7.1)  1 (9.1)  TV repair, n (%)  18 (6.8)  2 (4.8)  1 (9.1)  Ascending aorta replacement, n (%)  32 (12.2)  6 (14.3)  0 (0.0)  Aortic root enlargement, n (%)  14 (5.3)  5 (11.9)  1 (9.1)  Bentall, n (%)  41 (16.0)  6 (14.3)  2 (18.2)  AV: aortic valve; CABG: coronary artery bypass grafting; CPB: cardiopulmonary bypass; MV: mitral valve; MVR: mitral valve replacement; PVR: pulmonary valve replacement; Q1: 25% quartile; Q3: 75% quartile; TV: tricuspid valve; TVR: tricuspid valve replacement. Table 2: Intraoperative characteristics of the study population based on the number of previous AV interventions categorized by groups   No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Cross-clamp time (min), median (Q1–Q3)  93 (67–127)  90 (72–150)  95 (85–117)  CPB time (min), median (Q1–Q3)  125 (93–179)  132 (97–212)  180 (150–211)  Circulatory arrest, n (%)  21 (8.0)  3 (7.1)  1 (9.1)  AVR prosthesis type, n (%)   Mechanical valve  108 (41.1)  19 (45.2)  6 (54.5)   Bioprosthesis  148 (56.3)  22 (52.4)  5 (45.5)   Homograft  7 (2.6)  1 (2.4)  0 (0.0)  CABG, n (%)  16 (6.1)  2 (4.8)  0 (0.0)  MVR, n (%)  28 (10.6)  9 (21.4)  1 (9.1)  MV repair, n (%)  9 (3.4)  1 (2.4)  1 (9.1)  PVR, n (%)  1 (0.4)  1 (2.4)  0 (0.0)  TVR, n (%)  4 (1.5)  3 (7.1)  1 (9.1)  TV repair, n (%)  18 (6.8)  2 (4.8)  1 (9.1)  Ascending aorta replacement, n (%)  32 (12.2)  6 (14.3)  0 (0.0)  Aortic root enlargement, n (%)  14 (5.3)  5 (11.9)  1 (9.1)  Bentall, n (%)  41 (16.0)  6 (14.3)  2 (18.2)    No. of prior AV surgeries   Group 1 (n = 263)  Group 2 (n = 42)  Group 3 (n = 11)  Cross-clamp time (min), median (Q1–Q3)  93 (67–127)  90 (72–150)  95 (85–117)  CPB time (min), median (Q1–Q3)  125 (93–179)  132 (97–212)  180 (150–211)  Circulatory arrest, n (%)  21 (8.0)  3 (7.1)  1 (9.1)  AVR prosthesis type, n (%)   Mechanical valve  108 (41.1)  19 (45.2)  6 (54.5)   Bioprosthesis  148 (56.3)  22 (52.4)  5 (45.5)   Homograft  7 (2.6)  1 (2.4)  0 (0.0)  CABG, n (%)  16 (6.1)  2 (4.8)  0 (0.0)  MVR, n (%)  28 (10.6)  9 (21.4)  1 (9.1)  MV repair, n (%)  9 (3.4)  1 (2.4)  1 (9.1)  PVR, n (%)  1 (0.4)  1 (2.4)  0 (0.0)  TVR, n (%)  4 (1.5)  3 (7.1)  1 (9.1)  TV repair, n (%)  18 (6.8)  2 (4.8)  1 (9.1)  Ascending aorta replacement, n (%)  32 (12.2)  6 (14.3)  0 (0.0)  Aortic root enlargement, n (%)  14 (5.3)  5 (11.9)  1 (9.1)  Bentall, n (%)  41 (16.0)  6 (14.3)  2 (18.2)  AV: aortic valve; CABG: coronary artery bypass grafting; CPB: cardiopulmonary bypass; MV: mitral valve; MVR: mitral valve replacement; PVR: pulmonary valve replacement; Q1: 25% quartile; Q3: 75% quartile; TV: tricuspid valve; TVR: tricuspid valve replacement. A total of 6 re-entry injuries (1.9%) occurred, involving the thoracic aorta (n = 4), the superior vena cava (n = 1) and the right atrium (n = 1). The latter 2 were fatal. Immediate postoperative outcomes Postoperative outcomes are shown in Fig. 2. Overall hospital mortality was 7.3% (7.2% in Group 1, 4.76% in Group 2 and 18.18% in Group 3; P = 0.233, Fisher’s exact test). Low cardiac output syndrome, defined as inotropic requirement over 48 h to maintain a systolic arterial pressure above 90 mmHg, differed between groups, being far more frequent in Group 3 compared with Groups 1 and 2 (P = 0.04, χ2 test). In addition to ionotropic support, 15 patients with low cardiac output syndrome were treated with intra-aortic balloon pump and 3 with extracorporeal membrane oxygenation. A total of 23 surgical revisions were performed for excessive bleeding (7.3%). Figure 2: View largeDownload slide In-hospital postoperative outcomes of the study population based on the number of previous AV surgeries categorized by groups. †P = 0.036: χ2 test. AV: aortic valve; LCOS: low cardiac output syndrome; MOF: multiorgan failure; PM: pacemaker. Figure 2: View largeDownload slide In-hospital postoperative outcomes of the study population based on the number of previous AV surgeries categorized by groups. †P = 0.036: χ2 test. AV: aortic valve; LCOS: low cardiac output syndrome; MOF: multiorgan failure; PM: pacemaker. Long-term follow-up Global survival rate at 8 years was 79.0 ± 3.0%. Based on the number of previous AV interventions, survival rates at 8 years were 79.6 ± 3.3% in Group 1, 80.3 ± 9.0% in Group 2 and 50.4% ± 10.1% in Group 3, respectively (Fig. 3). Figure 3: View largeDownload slide Overall survival curve using the Kaplan–Meier method, based on the number of previous aortic valve surgery categorized by groups. Figure 3: View largeDownload slide Overall survival curve using the Kaplan–Meier method, based on the number of previous aortic valve surgery categorized by groups. Recurrence of any new AV procedure at 8 years was 86.7 ± 3.4%, 89.8 ± 2.6% in Group 1 and 84.1 ± 6.7% in Group 2 (log-rank test, P = 0.198). None of the Group 3 patients were reoperated on AV during follow-up. Based on the initial reason for reintervention, recurrence of any new AV procedure at 8 years was 97.8 ± 1.8% for SVD, 80.9 ± 10.0 for redo-Bentall procedure, 81.8 ± 7.9 for endocarditis, 68.2 ± 14% for PVD, 86.9 ± 6.2% for AV disease after previous valve repair and 76.5 ± 17.4 for mechanical valve dysfunction (log-rank test, P = 0.032). Survival at 8 years after AV re-replacement for SVD of a bioprosthesis was 88.4% ± 3.2%. There was no statistical difference in long-term survival between patients receiving a bioprosthesis or a mechanical valve (survival rates were 76.6 ± 4.6% and 81.7 ± 4.0% at 8 years, respectively, log-rank test, P = 0.425). Multivariable predictors of in-hospital and long-term outcomes Based on the reason for current reintervention, in-hospital mortality was 3.7% for SVD, 5.3% for redo-Bentall’s procedure, 5.7% for AV disease after previous valve repair, 9.8% for PVD, 10.7% for mechanical valve dysfunction and 14.0% for endocarditis. Considering the whole cohort by multivariable analysis, risk factors associated with overall in-hospital mortality are presented in Table 3. The area under the curve for the receiver operating characteristics of EuroSCORE II and Society of Thoracic Surgeons (STS) mortality score as predictors of surgical mortality were 0.830 (0.757–0.904) and 0.826 (0.742–0.910), respectively (Fig. 4). Table 3: Univariable and multivariable analyses of in-hospital mortality   Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Endocarditis  2.65  1.07–6.62  0.036  3.66  1.07–12.50  0.038  Female gender  2.66  1.13–6.29  0.025  1.44  0.37–5.51  0.596  Serum creatinine level >120 µmol/l  2.46  0.98–6.17  0.054  1.03  0.28–3.85  0.970  NYHA Class III or IV  11.36  2.58–50.00  0.001  15.87  1.90–142.86  0.011  LVEF  1.03  0.99–1.07  0.128  1.02  0.96–1.07  0.559  Non-elective surgery  2.44  0.88–6.76  0.086  2.65  0.60–16.39  0.174  sPAP≥ 55 mmHg  3.57  1.26–10.10  0.017  1.58  0.42–5.99  0.501  Tricuspid annuloplasty  3.41  1.05–11.11  0.042  3.69  0.74–18.52  0.112  MVR  3.70  1.41–9.71  0.008  5.65  1.66–19.23  0.006  Injury during resternotomy  6.85  1.18–40.00  0.032  16.95  1.74–166.67  0.015    Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Endocarditis  2.65  1.07–6.62  0.036  3.66  1.07–12.50  0.038  Female gender  2.66  1.13–6.29  0.025  1.44  0.37–5.51  0.596  Serum creatinine level >120 µmol/l  2.46  0.98–6.17  0.054  1.03  0.28–3.85  0.970  NYHA Class III or IV  11.36  2.58–50.00  0.001  15.87  1.90–142.86  0.011  LVEF  1.03  0.99–1.07  0.128  1.02  0.96–1.07  0.559  Non-elective surgery  2.44  0.88–6.76  0.086  2.65  0.60–16.39  0.174  sPAP≥ 55 mmHg  3.57  1.26–10.10  0.017  1.58  0.42–5.99  0.501  Tricuspid annuloplasty  3.41  1.05–11.11  0.042  3.69  0.74–18.52  0.112  MVR  3.70  1.41–9.71  0.008  5.65  1.66–19.23  0.006  Injury during resternotomy  6.85  1.18–40.00  0.032  16.95  1.74–166.67  0.015  LVEF: left ventricular ejection fraction; MVR: mitral valve replacement; NYHA: New York Heart Association; OR: odds ratio; sPAP: systolic pulmonary artery pressure. Table 3: Univariable and multivariable analyses of in-hospital mortality   Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Endocarditis  2.65  1.07–6.62  0.036  3.66  1.07–12.50  0.038  Female gender  2.66  1.13–6.29  0.025  1.44  0.37–5.51  0.596  Serum creatinine level >120 µmol/l  2.46  0.98–6.17  0.054  1.03  0.28–3.85  0.970  NYHA Class III or IV  11.36  2.58–50.00  0.001  15.87  1.90–142.86  0.011  LVEF  1.03  0.99–1.07  0.128  1.02  0.96–1.07  0.559  Non-elective surgery  2.44  0.88–6.76  0.086  2.65  0.60–16.39  0.174  sPAP≥ 55 mmHg  3.57  1.26–10.10  0.017  1.58  0.42–5.99  0.501  Tricuspid annuloplasty  3.41  1.05–11.11  0.042  3.69  0.74–18.52  0.112  MVR  3.70  1.41–9.71  0.008  5.65  1.66–19.23  0.006  Injury during resternotomy  6.85  1.18–40.00  0.032  16.95  1.74–166.67  0.015    Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Endocarditis  2.65  1.07–6.62  0.036  3.66  1.07–12.50  0.038  Female gender  2.66  1.13–6.29  0.025  1.44  0.37–5.51  0.596  Serum creatinine level >120 µmol/l  2.46  0.98–6.17  0.054  1.03  0.28–3.85  0.970  NYHA Class III or IV  11.36  2.58–50.00  0.001  15.87  1.90–142.86  0.011  LVEF  1.03  0.99–1.07  0.128  1.02  0.96–1.07  0.559  Non-elective surgery  2.44  0.88–6.76  0.086  2.65  0.60–16.39  0.174  sPAP≥ 55 mmHg  3.57  1.26–10.10  0.017  1.58  0.42–5.99  0.501  Tricuspid annuloplasty  3.41  1.05–11.11  0.042  3.69  0.74–18.52  0.112  MVR  3.70  1.41–9.71  0.008  5.65  1.66–19.23  0.006  Injury during resternotomy  6.85  1.18–40.00  0.032  16.95  1.74–166.67  0.015  LVEF: left ventricular ejection fraction; MVR: mitral valve replacement; NYHA: New York Heart Association; OR: odds ratio; sPAP: systolic pulmonary artery pressure. Figure 4: View largeDownload slide ROC of EuroSCORE II and STS mortality score for in-hospital mortality. AUC: area under the curve; ROC: receiver operating characteristics; STS: Society of Thoracic Surgeons. Figure 4: View largeDownload slide ROC of EuroSCORE II and STS mortality score for in-hospital mortality. AUC: area under the curve; ROC: receiver operating characteristics; STS: Society of Thoracic Surgeons. In Cox multivariable regression analysis, preoperative New York Heart Association (NYHA) Class III/IV, concomitant MVR and PVD were independent factors affecting long-term survival (Table 4). Table 4: Univariable and multivariable analyses of long-term mortality   Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Serum creatinine level >120 µmol/l  2.29  1.28–4.12  0.005  1.52  0.76–3.06  0.237  Group 3  2.57  0.93–7.14  0.070  2.56  0.84–7.81  0.098  NYHA Class III/IV  3.19  1.65–6.19  0.001  2.31  1.10–4.83  0.027  sPAP≥55 mmHg  3.24  1.30–4.93  0.006  1.48  0.70–3.13  0.308  MVR  2.99  1.58–5.62  0.006  2.82  1.37–5.78  0.005  Endocarditis  1.73  0.92–3.25  0.090  2.00  0.91–4.23  0.086  PVD  2.21  1.16–4.22  0.016  2.19  1.01–4.74  0.047    Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Serum creatinine level >120 µmol/l  2.29  1.28–4.12  0.005  1.52  0.76–3.06  0.237  Group 3  2.57  0.93–7.14  0.070  2.56  0.84–7.81  0.098  NYHA Class III/IV  3.19  1.65–6.19  0.001  2.31  1.10–4.83  0.027  sPAP≥55 mmHg  3.24  1.30–4.93  0.006  1.48  0.70–3.13  0.308  MVR  2.99  1.58–5.62  0.006  2.82  1.37–5.78  0.005  Endocarditis  1.73  0.92–3.25  0.090  2.00  0.91–4.23  0.086  PVD  2.21  1.16–4.22  0.016  2.19  1.01–4.74  0.047  MVR: mitral valve replacement; NYHA: New York Heart Association; OR: odds ratio; PVD: prosthetic valve dehiscence; sPAP: systolic pulmonary artery pressure. Table 4: Univariable and multivariable analyses of long-term mortality   Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Serum creatinine level >120 µmol/l  2.29  1.28–4.12  0.005  1.52  0.76–3.06  0.237  Group 3  2.57  0.93–7.14  0.070  2.56  0.84–7.81  0.098  NYHA Class III/IV  3.19  1.65–6.19  0.001  2.31  1.10–4.83  0.027  sPAP≥55 mmHg  3.24  1.30–4.93  0.006  1.48  0.70–3.13  0.308  MVR  2.99  1.58–5.62  0.006  2.82  1.37–5.78  0.005  Endocarditis  1.73  0.92–3.25  0.090  2.00  0.91–4.23  0.086  PVD  2.21  1.16–4.22  0.016  2.19  1.01–4.74  0.047    Univariable analysis   Multivariable analysis   OR  95% confidence interval  P-value  OR  95% confidence interval  P-value  Serum creatinine level >120 µmol/l  2.29  1.28–4.12  0.005  1.52  0.76–3.06  0.237  Group 3  2.57  0.93–7.14  0.070  2.56  0.84–7.81  0.098  NYHA Class III/IV  3.19  1.65–6.19  0.001  2.31  1.10–4.83  0.027  sPAP≥55 mmHg  3.24  1.30–4.93  0.006  1.48  0.70–3.13  0.308  MVR  2.99  1.58–5.62  0.006  2.82  1.37–5.78  0.005  Endocarditis  1.73  0.92–3.25  0.090  2.00  0.91–4.23  0.086  PVD  2.21  1.16–4.22  0.016  2.19  1.01–4.74  0.047  MVR: mitral valve replacement; NYHA: New York Heart Association; OR: odds ratio; PVD: prosthetic valve dehiscence; sPAP: systolic pulmonary artery pressure. DISCUSSION Currently, as the number of heart operations increases, multiple redo valve surgeries are becoming increasingly common. In a large series of patients with various reasons for reintervention, we demonstrated that outcomes of redo-AV surgery were more influenced by preoperative characteristics such as symptoms at presentation, aetiology and the need for concomitant MV surgery than the number of sternal re-entries. Our overall operative mortality results were in accordance with previous reports by Akins et al. [4] and Jones et al. [5]. Groups 1 and 2, representing patients undergoing redo- and re-redo-AV interventions, respectively, shared a similar aetiological pattern. They also had similar complication rates and outcomes. Therefore, a second redo procedure on the AV did not appear to increase morbidity compared with a single redo procedure with respect to the reason for reintervention. Similar findings have been reported by Tyers et al. [6] and recently by Davierwala et al. [7]. We also found that patients undergoing a fourth or more AV intervention (Group 3) seemed to have a higher all-complications rate but the limited number of patients in this group restricts any comparison with the other 2 groups. Moreover, patients in Group 3 were often referred late to surgery with more altered LV functions and higher pulmonary pressures at presentation. One of the major concerns regarding cardiac reinterventions is injury at resternotomy. Our incidence of re-entry injury was smaller than those reported by Roselli et al. [8] or Ellman et al. [9]. As recommended by Morishita et al. [10], a systematic preoperative investigation by a computed tomography scan in our patients who had previously undergone aortic surgery or coronary artery bypass grafting was certainly helpful in localizing the potential hazards during dissection and assessing the spatial relationship between the inner aspect of the sternum and the underlying mediastinal, cardiac and vascular structures. Nevertheless, as described previously by these authors, we found that the number of re-entry injuries was influenced by the number of re-sternotomies and had an impact on operative mortality. The presence of patent internal thoracic artery grafts may also compromise optimal myocardial protection if not controlled and has been assigned as an independent risk factor for injury by Park et al. [11] in an analysis of more than 2500 cardiac reinterventions. When the internal thoracic artery was considered impossible to be controlled, the so-called ‘open internal thoracic artery’ technique described by Vistarini et al. [12] has been used for years by some of the surgeons of our department with good results. There were no injuries to the coronary grafts reported in our study cohort. Furthermore, as already reported by Khaladj et al. [13], the presence of previous coronary artery bypass grafting was not related to adverse outcomes. Combined MVR, as opposed to mitral valve repair, has been known to negatively affect mortality and morbidity during AVR [14, 15]. Our results confirmed this established fact in redo-AV surgery. Furthermore, we found that concomitant MVR was also a risk factor for long-term mortality. In contrast, classical combined procedures such as aortic root replacement, coronary artery bypass grafting (described earlier) or tricuspid valve surgery were not associated with increased mortality in our study. The risk related to the latter procedure certainly may have benefitted from the large proportion of prophylactic tricuspid annuloplasties performed in our department, based more on the annular diameter rather than the dysfunction severity of the tricuspid valve. Although non-elective surgery was not associated with increased in-hospital mortality in our study, we found that urgency remained frequent in the population undergoing AV reinterventions. In addition, more than one-third of our patients presented with severe preoperative dyspnoea, which was an independent risk factor for operative mortality and affected long-term prognosis. This appeals for earlier timing, whenever possible, in addressing patients who require AV reintervention, as already advocated by Vogt et al. [16], almost 20 years ago. The various causes for reintervention in our analysis showcased very disparate postoperative course profiles. For endocarditis, which exclusively appeared on prosthetic valves, we reported an in-hospital mortality rate as high as 14% and an odds ratio of 3.66. This result was rationally lower than most of the other mortality rates reported in the literature [17, 18] but confirms that aortic prosthetic valve endocarditis is a serious disease, for which surgery remains challenging. Although we did not demonstrate a real statistical significance, it may also have had an influence on long-term prognosis. In our experience, PVD accounted for 12.6% of the cases for AV reintervention and the proportion appeared to increase with the number of reoperations. Although defined as a classical complication after AV replacement, occurrence of non-infectious PVD has been reduced to 2.2% in a recent report by Duncan et al. [19]. However, in terms of surgical mortality, PVD remains associated with a 10% mortality rate that confirmed the results recently published by Taramasso et al. [20]. Moreover, our data demonstrated that PVD was an independent risk factor for long-term all-cause mortality. Thus, we believe that alternative approaches such as transcatheter closure using vascular plugs should be considered in this setting, as early results are encouraging. In comparison, SVD, the leading cause for reintervention in this study, was associated with excellent short- and long-term outcomes. Our data confirmed the general trend towards a reduction of mortality in patients requiring AV reintervention for SVD [21]. This has mainly 2 implications: first, it could serve as a benchmark to compare with transcatheter valve-in valve therapy, which has recently been shown to have some drawbacks. Indeed, the recent retrospective analysis of the Global Valve-in-Valve Registry from Dvir et al. [22] has shown a 15.3% rate of valve malposition and a 3.5% rate of ostial obstruction complications that are rare in open redo-AVR. A single-centre Canadian study also demonstrated that patients receiving a valve-in-valve procedure following bioprosthetic SVD had a lower survival at 1 year [3]. Therefore, particularly in a younger population, transcatheter aortic valve implantation may not be a suitable option for SVD at the moment. Secondly, our results regarding redo-AVR for SVD may have an implication on the choice of prosthetic valve. Indeed, the main factors in making this choice include the risk of anticoagulation, thromboembolism and bleeding events that could occur with mechanical valves, whereas bioprosthetic valves are not as durable and thus have a risk of SVD leading to reintervention [23, 24]. In fact, a recent study by Glaser et al. [25] noted that patients between 50 and 69 years undergoing the first ever intervention to their AV and who opted for a bioprosthetic valve had a significantly higher risk of reintervention compared to those opting for a mechanical AV (hazard ratio 2.36). Thus, at present, mechanical valves are usually offered to patients younger than 60 years [26]. However, Ruel et al. [27] have demonstrated that selection of bioprosthetic valves in patients below 60 years of age had no impact on 30-year survival. Furthermore, our data suggested that the first and second AV reinterventions for SVD have a similar low operative risk, close to the initial AVR surgical procedure [28]. This is in line with Spampinato et al. [29] who suggested, 20 years ago, that the choice of a bioprosthetic valve for a redo surgery following SVD was not hazardous. Limitations The limitations of this study include its inherent retrospective design and analysis of observational data. Furthermore, the groups do not comprise a similar number of patients, particularly Group 3, with 11 patients over 10 years compared with 263 and 42 in Groups 1 and 2, respectively. However, this would be reflective of true clinical practice as it is not common to have patients presenting for a third or more redo-AVRs, and a much longer, and larger, study incorporating more than one facility is possibly required to obtain higher numbers. Finally, the Cardiac STS Score could not be utilized for those undergoing concomitant AVR and mitral valve procedures, and for this reason, the EuroSCORE II was also used; perhaps future models of prediction could incorporate this. CONCLUSIONS Multiple AV reinterventions carry an acceptable risk of in-hospital mortality, especially in the setting of isolated AVR for SVD of a bioprosthetic valve. These findings may favour the use of a biological valve substitute for AVR in young patients. Furthermore, they could serve as a benchmark for defining the role of valve-in-valve implantation in the setting of multiple AV reinterventions, which is still unknown. Given the inherently higher risks of AV reintervention associated with the transcatheter approaches, the latter may be considered more selectively in older patients with high preoperative risk well assessed by the EuroSCORE II and STS mortality scores. ACKNOWLEDGEMENTS We thank the surgeons and staff of the Cardiovascular Department at the Georges Pompidou European Hospital. Conflict of interest: none declared. REFERENCES 1 Leontyev S, Borger MA, Davierwala P, Walther T, Lehmann S, Kempfert J et al.   Redo aortic valve surgery: early and late outcomes. Ann Thorac Surg  2011; 91: 1120– 6. Google Scholar CrossRef Search ADS PubMed  2 Balsam LB, Grossi EA, Greenhouse DG, Ursomanno P, Deanda A, Ribakove GH et al.   Reoperative valve surgery in the elderly: predictors of risk and long-term survival. Ann Thorac Surg  2010; 90: 1195– 200. Google Scholar CrossRef Search ADS PubMed  3 Erlebach M, Wottke M, Deutsch M-A, Krane M, Piazza N, Lange R et al.   Redo aortic valve surgery versus transcatheter valve-in-valve implantation for failing surgical bioprosthetic valves: consecutive patients in a single-center setting. J Thorac Dis  2015; 7: 1494– 500. Google Scholar PubMed  4 Akins CW, Buckley MJ, Daggett WM, Hilgenberg AD, Vlahakes GJ, Torchiana DF et al.   Risk of reoperative valve replacement for failed mitral and aortic bioprostheses. Ann Thorac Surg  1998; 65: 1545– 51. Google Scholar CrossRef Search ADS PubMed  5 Jones JM, O’Kane H, Gladstone DJ, Sarsam MA, Campalani G, MacGowan SW et al.   Repeat heart valve surgery: risk factors for operative mortality. J Thorac Cardiovasc Surg  2001; 122: 913– 8. Google Scholar CrossRef Search ADS PubMed  6 Tyers GF, Jamieson WR, Munro AI, Germann E, Burr LH, Miyagishima RT et al.   Reoperation in biological and mechanical valve populations: fate of the reoperative patient. Ann Thorac Surg  1995; 60: S464– 8. Google Scholar CrossRef Search ADS PubMed  7 Davierwala PM, Borger MA, David TE, Rao V, Maganti M, Yau TM. Reoperation is not an independent predictor of mortality during aortic valve surgery. J Thorac Cardiovasc Surg  2006; 131: 329– 35. Google Scholar CrossRef Search ADS PubMed  8 Roselli EE, Pettersson GB, Blackstone EH, Brizzio ME, Houghtaling PL, Hauck R et al.   Adverse events during reoperative cardiac surgery: frequency, characterization, and rescue. J Thorac Cardiovasc Surg  2008; 135: 316– 23; 323.e1–6. Google Scholar CrossRef Search ADS PubMed  9 Ellman PI, Smith RL, Girotti ME, Thompson PW, Peeler BB, Kern JA et al.   Cardiac injury during resternotomy does not affect perioperative mortality. J Am Coll Surg  2008; 206: 993– 7. Google Scholar CrossRef Search ADS PubMed  10 Morishita K, Kawaharada N, Fukada J, Yamada A, Masaru T, Kuwaki K et al.   Three or more median sternotomies for patients with valve disease: role of computed tomography. Ann Thorac Surg  2003; 75: 1476– 80. Google Scholar CrossRef Search ADS PubMed  11 Park CB, Suri RM, Burkhart HM, Greason KL, Dearani JA, Schaff HV et al.   Identifying patients at particular risk of injury during repeat sternotomy: analysis of 2555 cardiac reoperations. J Thorac Cardiovasc Surg  2010; 140: 1028– 35. Google Scholar CrossRef Search ADS PubMed  12 Vistarini N, Aubert S, Leprince P, Pavie A. A simplified surgical approach for aortic valve replacement after previous coronary artery bypass grafting. Eur J Cardiothorac Surg  2009; 36: 404– 6. Google Scholar CrossRef Search ADS PubMed  13 Khaladj N, Shrestha M, Peterss S, Kutschka I, Strueber M, Hoy L et al.   Isolated surgical aortic valve replacement after previous coronary artery bypass grafting with patent grafts: is this old-fashioned technique obsolete? Eur J Cardiothorac Surg  2009; 35: 260– 4. Google Scholar CrossRef Search ADS PubMed  14 Rankin JS, Hammill BG, Ferguson TB, Glower DD, O’Brien SM, DeLong ER et al.   Determinants of operative mortality in valvular heart surgery. J Thorac Cardiovasc Surg  2006; 131: 547– 57. Google Scholar CrossRef Search ADS PubMed  15 Kim GS, Kim JB, Han S, Choo SJ, Chung CH, Lee JW et al.   Mitral valve repair versus replacement for moderate-to-severe mitral regurgitation in patients undergoing concomitant aortic valve replacement. Interact CardioVasc Thorac Surg  2014; 18: 73– 9. Google Scholar CrossRef Search ADS PubMed  16 Vogt PR, Brunner-LaRocca H, Sidler P, Zünd G, Truniger K, Lachat M et al.   Reoperative surgery for degenerated aortic bioprostheses: predictors for emergency surgery and reoperative mortality. Eur J Cardiothorac Surg  2000; 17: 134– 9. Google Scholar CrossRef Search ADS PubMed  17 Luciani N, Mossuto E, Ricci D, Luciani M, Russo M, Salsano A et al.   Prosthetic valve endocarditis: predictors of early outcome of surgical therapy. A multicentric study. Eur J Cardiothorac Surg  2017; 52: 768– 74. Google Scholar CrossRef Search ADS PubMed  18 Alonso-Valle H, Fariñas-Álvarez C, García-Palomo JD, Bernal JM, Martín-Durán R, Díez JFG et al.   Clinical course and predictors of death in prosthetic valve endocarditis over a 20-year period. J Thorac Cardiovasc Surg  2010; 139: 887– 93. Google Scholar CrossRef Search ADS PubMed  19 Duncan BF, McCarthy PM, Kruse J, Andrei A-C, Li Z, Russell HM et al.   Paravalvular regurgitation after conventional aortic and mitral valve replacement: a benchmark for alternative approaches. J Thorac Cardiovasc Surg  2015; 150: 860– 8.e1. Google Scholar CrossRef Search ADS PubMed  20 Taramasso M, Maisano F, Denti P, Guidotti A, Sticchi A, Pozzoli A et al.   Surgical treatment of paravalvular leak: long-term results in a single-center experience (up to 14 years). J Thorac Cardiovasc Surg  2015; 149: 1270– 5. Google Scholar CrossRef Search ADS PubMed  21 Jamieson WRE, Burr LH, Miyagishima RT, Janusz MT, Fradet GJ, Ling H et al.   Re-operation for bioprosthetic aortic structural failure-risk assessment. Eur J Cardiothorac Surg  2003; 24: 873– 8. Google Scholar CrossRef Search ADS PubMed  22 Dvir D, Webb J, Brecker S, Bleiziffer S, Hildick-Smith D, Colombo A et al.   Transcatheter aortic valve replacement for degenerative bioprosthetic surgical valves: results from the global valve-in-valve registry. Circulation  2012; 126: 2335– 44. Google Scholar CrossRef Search ADS PubMed  23 Head SJ, Çelik M, Kappetein AP. Mechanical versus bioprosthetic aortic valve replacement. Eur Heart J  2017; 38: 2183– 91. Google Scholar CrossRef Search ADS PubMed  24 Kulik A, Bédard P, Lam B-K, Rubens FD, Hendry PJ, Masters RG et al.   Mechanical versus bioprosthetic valve replacement in middle-aged patients. Eur J Cardiothorac Surg  2006; 30: 485– 91. Google Scholar CrossRef Search ADS PubMed  25 Glaser N, Jackson V, Holzmann MJ, Franco-Cereceda A, Sartipy U. Aortic valve replacement with mechanical vs. biological prostheses in patients aged 50-69 years. Eur Heart J  2016; 37: 2658– 67. Google Scholar CrossRef Search ADS PubMed  26 Sakamoto Y, Yoshitake M, Matsumura Y, Naruse H, Bando K, Hashimoto K. Choice of aortic valve prosthesis in a rapidly aging and long-living society. Ann Thorac Cardiovasc Surg  2016; 22: 333– 9. Google Scholar CrossRef Search ADS PubMed  27 Ruel M, Chan V, Bédard P, Kulik A, Ressler L, Lam BK et al.   Very long-term survival implications of heart valve replacement with tissue versus mechanical prostheses in adults <60 years of age. Circulation  2007; 116: I294– 300. Google Scholar CrossRef Search ADS PubMed  28 Potter DD, Sundt TM, Zehr KJ, Dearani JA, Daly RC, Mullany CJ et al.   Operative risk of reoperative aortic valve replacement. J Thorac Cardiovasc Surg  2005; 129: 94– 103. Google Scholar CrossRef Search ADS PubMed  29 Spampinato N, Gagliardi C, Pantaleo D, Fimiani L, Ascione R, De Robertis F et al.   Bioprosthetic replacement after bioprosthesis failure: a hazardous choice? Ann Thorac Surg  1998; 66: S68– 72. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. 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: Dec 26, 2017

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