Interactive CardioVascular and Thoracic Surgery Advance Access published March 8, 2015
Cite this article as: Tourmousoglou C, Rao V, Lalos S, Dougenis D. What is the best approach in a patient with a failed aortic bioprosthetic valve: transcatheter aortic valve replacement or redo aortic valve replacement? Interact CardioVasc Thorac Surg 2015; doi:10.1093/icvts/ivv037.
What is the best approach in a patient with a failed aortic bioprosthetic valve: transcatheter aortic valve replacement or redo aortic valve replacement? Christos Tourmousogloua,*, Vivek Raob, Spiros Lalosc and Dimitrios Dougenisd a b c d
University of Ioannina Medical School, Ioannina, Greece Division of Cardiovascular Surgery, Peter Munk Cardiac Center, Toronto General Hospital, University of Toronto, Toronto, Canada Department of Cardiothoracic Surgery, Hippokratio General Hospital, Athens, Greece Department of Cardiothoracic Surgery, University Hospital of Patra, Patra, Greece
* Corresponding author. University of Ioannina Medical School, 13 Poutetsi St, 45332 Ioannina, Greece. Tel: +30-6977748313; e-mail: [email protected] (C.E. Tourmousoglou). Received 3 April 2014; received in revised form 29 January 2015; accepted 12 February 2015
Abstract A best evidence topic in cardiac surgery was written according to a structured protocol. The question addressed was whether transcatheter aortic valve-in-valve replacement (viv-TAVR) or redo aortic valve replacement (rAVR) is the best strategy in a patient with a degenerative bioprosthetic aortic valve. Altogether, 162 papers were found using the reported search, of which 12 represented the best evidence to answer the question. The authors, journal, date, country of publication, patient group, study type, outcomes and results of papers are tabulated. The results of the studies provided interesting results. All the studies are retrospective. Four papers reported the results of redo aortic valve replacement in patients with failed aortic bioprosthetic valve, six papers demonstrated their results with transcatheter aortic valve-in-valve replacement for the same indication and two papers reported their propensity-matched analysis of outcomes between viv-TAVR and rAVR in patients with previous cardiac surgery. Thirty-day mortality for rAVR was 2.3–15.5% and 0–17% for viv-TAVR. For rAVR, survival rate at 30 days was 83.6%, 76.1% at 1 year, 70.8% at 3 years, at 51.3–66% at 5 years, 61% at 8 years and 61.5% at 10 years. For viv-TAVR, the overall Kaplan–Meier survival rate at 1 year was 83.2%. After viv-TAVR at 1 year, 86.2% of surviving patients were at New York Heart Association (NYHA) class I/II. The complications after rAVR were stroke (4.6–5.8%), reoperation for bleeding (6.9–9.7%), low-cardiac output syndrome (9.9%) whereas complications after viv-TAVR at 30 days were major stroke (1.7%), aortic regurgitation of at least moderate degree (25%), new permanent pacemaker implantation rate (0–11%), ostial coronary obstruction (2%), need for implantation of a second device (5.7%) and major vascular complications (9.2%). It is noteworthy to mention that there is a valve-in-valve application that provides information to surgeons for choosing the correct size of the TAVR valve. Transcatheter aortic valve-in-valve procedures are clinically effective, at least in the short term, and could be an acceptable approach in selected high-risk patients with degenerative bioprosthetic valves. Redo AVR achieves acceptable medium and long-term results. Both techniques could be seen as complementary approaches for high-risk patients. Keywords: Aortic valve • Failed • Bioprosthesis • Implantation
INTRODUCTION A best evidence topic was constructed according to a structured protocol. This is fully described in ICVTS [1].
The patient has become increasingly dyspnoeic during the last year. Transthoracic echocardiography and transoesophageal echocardiography revealed severe aortic regurgitation (three-fourth). The valve has to be replaced. There are two options for the patient: transcatheter aortic valve-in-valve replacement or redo AVR.
THREE- PART QUESTION In [a patient with a degenerative bioprosthetic aortic valve] is [transcatheter aortic valve-in-valve replacement or redo aortic valve replacement] the best strategy in order to optimize [event-free survival]?
CLINICAL SCENARIO A 78-year old patient, with a history of previous aortic valve replacement with a bioprosthetic valve 20 years ago, comes to your office.
SEARCH STRATEGY A review of the English language literature was performed on Medline using the Ovid interface from 1980 to July 2014 [transcatheter aortic valve implantation OR transcatheter aortic valve replacement OR transcatheter aortic valve-in-valve procedure OR re-replacement of aortic valve OR redo aortic valve replacement) AND ( previous aortic valve replacement OR degenerative bioprostheses)].
© The Author 2015. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
BEST EVIDENCE TOPIC
BEST EVIDENCE TOPIC – ADULT CARDIAC
Interactive CardioVascular and Thoracic Surgery (2015) 1–7 doi:10.1093/icvts/ivv037
C. Tourmousoglou et al. / Interactive CardioVascular and Thoracic Surgery
2
Table 1: Best evidence papers Author, date and country Study type (level of evidence)
Patient group
Key outcomes
Results
Eitz et al. (2006), Ann Thorac Surg, Germany [2]
71 patients (1991–2004), mean duration of follow-up 3.3 ± 2.8 years
‘AVR-type’ Bioprosthesis Mechanical
69/71 patients 2/71 patients
‘Age group’ 80–84 years ≥85 years Median time since first AVR Operative mortality rate Mortality during follow-up Survival at 30 days, 1 year, 3 years and 5 years in patients with reoperation
Cohort study (level 2b)
Estimated median survival of patients with reoperations Davierwala et al. (2006), J Thorac Cardiovasc Surg, Canada [3] Cohort study (level 2b)
216 patients (1990–2002) underwent aortic valve rereplacements from a total group of 2673 patients that underwent aortic valve surgery
‘Redo-AVR type’ No. of operations Second operation Third operation Fourth operation Mechanical valve Bioprostheses Median duration since first AVR Hospital mortality ‘Postoperative outcomes’ Low-cardiac output syndrome Postoperative MI Postoperative stroke Postoperative renal failure Reoperation rate for bleeding Pacemaker insertion
Kempfert et al. (2010), Ann Thorac Surg, Germany [4]
11 patients (2007–2009), follow-up 330 ± 291 days
Cohort study (level 2b)
Mean age since first AVR Mean logistic EuroSCORE Mean STS score Successful valve-in-valve implantation Off-pump treatment ‘Postoperative outcomes’ 30-day mortality rate Stroke rate New pacemaker implantation rate Trans/paravalvular leak first degree Maximal AV gradient, mmHg Mean AV gradient, mmHg
Leontyev et al. (2011), Ann Thorac Surg, Germany [5] Cohort study (level 2b)
155 patients (1994–2008), mean duration of the follow-up 2.7 ± 2.8 years, 86 isolated redo AVR, 69 redo aortic root replacement
Mean age Mean logistic EuroSCORE Mean interval since first AVR Early mortality Reoperation for bleeding Stroke Low-cardiac output syndrome
90.2% 9.8% 8 years 15.5% 40.8% 83.6%, 76.1%, 70.8% and 51.3%, respectively
Comments
Low-cardiac output syndrome and intestinal failure were independent risk factors of 30-day survival in the cohort of reoperated patients Early and mid-term survival was predominantly influenced by unexpected postoperative complications and not by preoperative risk factors, with the exception of diabetes mellitus
5.6 years
191 patients 23 patients 2 patients 52% 48% 10.3 years 4.6%
4.6%
Patients in the redo AVR group had predominantly AR Hospital mortality was independently predicted by peripheral vascular disease, active endocarditis, worsening NYHA class and need for annular enlargement Reoperation itself did not predict hospital mortality
0.9% 4.6% 4.2% 6.9% 14% 7.5 ± 3.4 years 31.7 ± 14.5% 7.2 ± 2.6% 100%
None of the patients had new onset postoperative AV block
100%
0% 0% 0% 18.2% 21.4 ± 7.5 11 ± 4.2 58.1 ± 16.3 years 27 ± 23% 6.7 ± 7.9 years 4.5% 9.7% 5.8% 9%
Only one-quarter of patients who received isolated rAVS in this study were potentially suitable for a valve-in-valve transcatheter procedure
Continued
C. Tourmousoglou et al. / Interactive CardioVascular and Thoracic Surgery
BEST EVIDENCE TOPIC
3
Table 1: (Continued) Author, date and country Study type (level of evidence)
Eggebrecht et al. (2011), JACC Cardiovasc Interv, Germany [6]
Patient group
47 patients (2005–2010), trans-femoral (n = 25) or transapical (n = 22) valve-in-valve TAVR
Cohort study (level 2b)
Key outcomes
Results
Survival at 5 years and 8 years Freedom from reoperation at 5 years and 8 years
66 ± 5%, 61 ± 6%, respectively 87.1 ± 4%, 82.3 ± 6%, respectively
Mean age Time since first AVR Mean logistic EuroSCORE, % Mean STS score
79.8 ± 7.1 years 113 ± 65 months 35 ± 18.5%
‘Type of first bioprosthesis’ Stented valve Stentless aortic root valve
96% 4%
‘Prosthesis used for viv-TAVI’ Edwards-Sapien valve Medtronic CoreValve Procedural success Mean gradient after TAVI
75% 25% 98% 17 ± 10 (3–35) mmHg
‘Regurgitation after TAVI’ None Mild Moderate Procedural complications Vascular access complications Bail-out implantation of second TAVI prosthesis New permanent pacemaker implantation All-cause 30-day mortality rate Pasic et al. (2011), J Thorac Cardiovasc Surg, Germany [7]
14 patients (2008–2010), follow-up 20 months, transapical Edwards-Sapien valve
Cohort study (level 2b)
Mean age Mean time since first AVR Mean STS score Mean logistic EuroSCORE Procedural success Conversion to open surgery ‘Mean aortic transvalvular gradient’ preoperatively postoperatively New pacemaker implantation rate New clinical neurological deficit Reoperation rate during follow-up Aortic Regurgitation after TAVI (end of follow-up) Death rate during the follow-up
Bapat et al. (2012), J Thorac Cardiovasc Surg, UK [8] Cohort study (level 2b)
23 patients (2009–2011), mean follow-up 8.1 months, valve-in-valve (transapical 21/23)
11.6 ± 8.5%
Mean age ‘Type of AV used during initial operation’ Stented Stentless Mean time since first AVR for stented vs stentless valves
Comments
A trend towards lower gradients for Medtronic CoreValve viv-TAVI procedures in small, 21-mm bioprostheses compared with Edwards-Sapien valves was seen
55% 43% 2% 19% 13% 4% 11% 17% (1 procedural death and 7 in-hospital deaths) 73.3 ± 13.1 years 8.9 ± 4.7 years 21.9 ± 10.9% 45.3 ± 22.2% 100% 0%
There were three limitations in that study: (i) there was no control group with conventional AVR, (ii) the number of patients was small and (iii) the follow-up was limited
37.1 ± 25.7 mmHg 13.1 ± 6.4 mmHg 0% 0% 7% 0% 14% 79.6 ± 14.4 years
15/23 patients 8/23 patients 9.4 ± 3.8 years vs 14 ± 6.2 years
Large case series with short follow-up Concluded that patients with small bioprosthetic valves might be treated with VIV if the risk of conventional surgery is prohibitive
Continued
C. Tourmousoglou et al. / Interactive CardioVascular and Thoracic Surgery
4
Table 1: (Continued) Author, date and country Study type (level of evidence)
Linke et al. (2012), Circ Cardiovasc Interv, Germany [9]
Patient group
27 patients, mean follow-up 421 ± 198 days, TAVR with Medtronic CoreValve (trans-femoral)
Cohort study (level 2b)
Pechlivanidis et al. (2013), J Thorac Cardiovasc Surg, Italy [10] Cohort study (level 2b)
131 patients (2002–2010), mean follow-up 55.1 ± 32.4 months, underwent redo aortic valve replacement
Key outcomes
Results
Mean logistic EuroSCORE Mean STS score Procedural success rate Mean aortic valve gradient preprocedural vs post-procedural AR paravalvular postimplantation mild Permanent pacemaker implantation rate MyocardiaI Infarction rate In-hospital mortality 30-day mortality Late death rate during the follow-up
31.6 ± 15.3% 6.3 ± 3.8% 100% 31.2 ± 17.06 mmHg vs 9.13 ± 4.9 mmHg
Mean age Mean logistic EuroSCORE, Mean STS score Time since first AVR Conversion to open surgery MI intraprocedurally Death rate intraprocedurally Mean AV gradient reduction All-cause mortality at 30 days Major stroke Life-threatening bleeding Major access site complications New permanent pacemaker rate Survival rate at 6 months and at 1 year
74.8 ± 8 years 31.3 ± 16.5% 3–30.9% 67 ± 46 months 0%
Mean age
65.6 ± 12.9 years
‘Previous aortic valve prosthesis’ Biological Mechanical
80.9% 19.1%
‘Type of intervention’ Redo AVR Redo AVR plus CABG
75.5% 24.5%
‘Type of prosthesis used’ Biological Mechanical Homograft
65.6% 29% 5.3%
Hospital mortality Reoperation rate for bleeding Low output cardiac syndrome Stroke 10-year survival 10-year freedom from acute heart failure 10-year freedom from reoperations
Comments
22% 0% 0% 0% 0% 13%
0% 0% 42 ± 15 to 10 ± 8 mmHg
Trans-femoral implantation of Medtronic CoreValve into a wide range of failed bioprosthetic aortic valves was feasible, safe and resulted in marked improvement in haemodynamics in selected high-risk patients
7.4% 3.7% 7.4% 11.1% 3.7% 92%, 88% respectively
2.3% 9.2% 9.9% 3.1% 61.5 ± 8.6% 62.9 ± 6.9%
Survival during follow-up was lower, but not statistically significant, for older patients (patients >75 years vs patients with 2) in r-TAVI/r- AVR 30-day mortality in r-TAVI/r-AVR Actuarial survival rate at 4 years in r-TAVI/r-AVR Actuarial major stroke rate at 4 years in r-TAVI/ r-AVR Mean transvalvular gradient in r-TAVI/r-AVR
81 ± 4 years vs 80 ± 3 years
Paravalvular leaks in r-TAVI/r-AVR Grade I/II Grade III/IV
24 ± 6% vs 19 ± 6% 11.1 ± 2.8% vs 10.4 ± 3% 7 ± 5 vs 8 ± 6 years 3 vs 17% 0 vs 8% 0 vs 13% (P = 0.04) 8 vs 16% (P = 0.06)
Valve-in-valve outcomes were worse in patients with small surgical valves (label size ≤21 mm) and in patients with stenosis as mechanism of failure The registry showed that there was an elevated rate of residual leak in patients with baseline regurgitation
There was a trend towards a decreased 30-day mortality following r-TAVI The incidence of major adverse events including death and permanent neurological complications within 30 days postoperatively reached statistical significance between the two groups (there was a higher rate in the open surgical group)
75 ± 3% vs 73 ± 4% (P = 0.56) 25 vs 43% (P = 0.01)
9.8 ± 3 mmHg vs 10.1 ± 2 mmHg
33 vs 0% 0 vs 0%
NYHA: New York Heart Association; EuroSCORE: European System for Cardiac Operative Risk Evaluation; STS: society of thoracic surgeons.
SEARCH OUTCOME Using the reported search, 162 papers were identified of which 12 papers provided the best evidence to answer the question. These papers are summarized in Table 1.
RESULTS Eitz et al. [2] showed that perioperative mortality rate was 15.5% in reoperations of the aortic valve in a group of 71 octogenarians. Indications for surgery were structural valve deterioration (76.1%),
paravalvular leak (4.2%) and prosthetic valve endocarditis (11.3%). Survival rates at 30 days, 1 year, 3 years and 5 years were 83.6, 76.1, 70.8 and 51.3%, respectively. Estimated median survival was 5.6 years for patients with reoperations. Mid-term survival was similar when compared with a group of patients matched for age, gender and date of operation, having their first AVR. Davierwala et al. [3] showed that mortality rate was 4.6% in rAVR in 216 patients. The proportion of patients that received mechanical valves was 52% in the rAVR group, significantly higher than that in the first AVR operation in the same group of patients (17%). Reoperative aortic valve surgery was not a significant predictor of hospital mortality.
Kempfert et al. [4] showed that 30-day mortality was 0% in viv-TAVR in 11 patients. Stroke rate and new pacemaker implantation rate was 0%. Postoperatively, trans/paravalvular leak of first degree was 18.2% and mean AV gradient was 11 ± 4 mmHg. Leontyev et al. [5] demonstrated that early mortality was 4.5% for all rAVR in 155 patients. Indications for redo AV surgery were endocarditis (27.1%), paravalvular leak (18.1%), bioprosthetic structural valve deterioration (16.1%), endocarditis with root abscess (14.2%) and bioprosthetic degeneration (7.7%). Five- and 8-year survival was 66 ± 5% and 61 ± 6%, respectively. Eggebrecht et al. [6] showed that 30-day mortality was 17% in 47 patients that underwent viv-TAVR. Indications for surgery were isolated stenosis (47%), regurgitation (32%) and mixed stenosis and regurgitation (21%). Increased transvalvular gradients ≥20 mm were observed in 44% of patients, especially in those with bioprostheses of small diameters. Vascular access complications occurred in 13% of patients. New pacemaker implantation rate was 11%. Pasic et al. [7] showed that operative mortality was 0% in 14 patients that were treated with viv-TAVR. Mean aortic transvalvular gradient decreased from 37.1 ± 25.7 mmHg to 13.1 ± 6.4 mmHg, postoperatively. There was no postoperative valve insufficiency. Up to 20 months postoperatively, 86% of patients were in New York Heart Association (NYHA) class I or II. Bapat et al. [8] found that 30-day mortality was 0% in 14 patients that were treated with viv-TAVR. Indication for surgery was predominantly aortic valve stenosis (56%). Mean AV gradient decreased from 31.2 ± 17.06 mmHg to 9.13 ± 4.9 mmHg. Late death rate during the follow-up was 13%. Linke et al. [9] showed that all-cause mortality at 30 days was 7.4% in a group of 27 patients that were treated with viv-TAVR. Major stroke rate was 3.7% and major access site complications were 11.1%. Survival rate at 6 months and 1 year was 92 and 88%, respectively. Mean AV gradient decreased from 42 ± 15 to 10 ± 8 mmHg. Pechlivanidis et al. [10] demonstrated that hospital mortality was 2.3% in reoperations of the aortic valve in 131 patients. Ten-year survival was 61.5 ± 8.6%. Ten-year freedom from acute heart failure, reoperations, stroke and thromboembolisms were 62.9 ± 6.9%, 97.8 ± 1.5%, 93.2 ± 3% and 91.2 ± 3.2%, respectively. Redo AVRs achieved good results for elective cases, cases without endocarditis and NYHA class I/II cases. Wilbring et al. [11] showed that 30-day mortality was 9.4% in a group of 53 patients who were treated with viv-TAVR versus 5.7% in a propensity-matched group of 53 patients who were treated with conventional redo AVR. Survival at 6 months was 83% in the viv-TAVR group vs 86.8% in the rAVR group. Mean grade of paravalvular regurgitation was 0.8 ± 0.2 in the viv-TAVI group vs 0% in the rAVR group. Dvir et al. [12] showed that 30-day mortality was 7.6% in viv-TAVR in their international registry of 459 patients. Modes of aortic bioprosthesis failure were stenosis (39.4%), regurgitation (30.3%) and combined (30.3%). Major stroke rate was 1.7% and permanent pacemaker implantation rate was 8.3%. Overall 1-year Kaplan–Meier survival rate was 83.2%. Patients in the stenosis group had worse 1-year survival (76.6%) in comparison with the regurgitation group (91.2%) and the combined group (83.9%). Patients with small valves had also worse 1-year survival (74.8%) versus patients with intermediate-sized valves (81.8%) and patients with large valves (93.3%).
7
Papadopoulos et al. [13] showed that 30-day mortality was 8% in the viv-TAVR group versus 16% in the r-AVR group by using propensity analysis of matched subgroups of 40 patients. Actuarial survival at 4 years was 75 ± 3% in the viv-TAVR group vs 73 ± 4% in the rAVR group. Kaplan–Meier estimates of major adverse events (death and permanent neurological deficits) at 4 years were 25% for the viv-TAVR group vs 43% for the rAVR group.
CLINICAL BOTTOM LINE Both viv-TAVR and rAVR provided good clinical results in high-risk patients with previous cardiac surgery. Both techniques could be seen as complementary approaches. An interdisciplinary heart team is necessary to discuss every patient and choose the best individualized approach. Conflict of interest: none declared.
REFERENCES [1] Dunning J, Prendergast B, Mackway-Jones K. Towards evidence-based medicine in cardiothoracic surgery: best BETS. Interact CardioVasc Thorac Surg 2003;2:405–9. [2] Eitz T, Fritzsche D, Kleikamp G, Zittermann A, Horstkotte D, Korfer R. Reoperation of the aortic valve in octogenarians. Ann Thorac Surg 2006; 82:1385–91. [3] Davierwala P, Borger M, David T, Rao V, Maganti M, Yau T. Reoperation is not an independent predictor of mortality during aortic valve surgery. J Thorac Cardiovasc Surg 2006;131:329–35. [4] Kempfert J, van Linden A, Linke A, Borger M, Rastan A, Mukherjee C et al. Transapical off-pump valve in valve implantation in patients with degenerated aortic xenografts. Ann Thorac Surg 2010;89:1934–41. [5] Leontyev S, Borger M, Davierwala P, Walter T, Lehmann S, Kempfert J et al. Redo aortic valve surgery: early and late outcomes. Ann Thorac Surg 2011; 91:1120–6. [6] Eggebrecht H, Schafer U, Treede H, Boekstegers P, Babin-Ebell J, Ferrari M et al. Valve-in valve transcatheter aortic valve implantation for degenerated bioprosthetic heart valves. JACC Cardiovasc Interv 2011;4:1218–27. [7] Pasic M, Unbehaum A, Dreysse S, Buz S, Drews T, Kukucka M et al. Transapical aortic valve implantation after previous aortic valve replacement: clinical proof of the “valve-in-valve” concept. J Thorac Cardiovasc Surg 2011;142:270–7. [8] Bapat V, Attia R, Redwood S, Hancock J, Wilson K, Young C et al. Use of transcatheter heart valves for a valve-in-valve implantation in patients with degenerated aortic bioprosthesis: technical considerations and results. J Thorac Cardiovasc Surg 2012;144:1372–80. [9] Linke A, Woitek F, Merx M, Schiefer C, Mobius-Winkler S, Holzhey D et al. Valve-in-valve implantation of Medtronic CoreValve with failing bioprosthetic aortic valves. Circ Cardiovasc Intev 2012;5:689–97. [10] Pechlivanidis K, Onorati F, Petrilli G, Santini F, Milano A, Torre S et al. In which patients is transcatheter aortic valve replacement potentially better indicated than surgery for redo aortic valve disease? Long-term results of a 10-year surgical experience. J Thorac Cardiovasc Surg 2014; 148:500–8.e1. [11] Wilbring M, Tugtekin SM, Alexiou K, Simonis G, Matschke K, Kappert U. Transapical transcatheter aortic valve implantation vs conventional aortic valve replacement in high-risk patients with previous cardiac surgery: a propensity-score analysis. Eur J Cardiothorac Surg 2013;44:42–7. [12] Dvir D, Webb J, Bleiziffer S, Pasic M, Waksman R, Kodali S et al. For the Valve-in-Valve International Data Registry Investigators. Transcatheter aortic valve implantation in failed bioprosthetic surgical valves. JAMA 2014;312:162–70. [13] Papadopoulos N, Schiller N, Fichtlscherer S, Lehmann R, Weber C, Moritz A et al. Propensity matched analysis of longterm outcomes following transcatheter based aortic valve implantation versus classic aortic valve replacement in previous cardiac surgery. J Cardiothorac Surg 2014;9:99.
BEST EVIDENCE TOPIC
C. Tourmousoglou et al. / Interactive CardioVascular and Thoracic Surgery
Cite this article as: Tourmousoglou C, Rao V, Lalos S, Dougenis D. What is the best approach in a patient with a failed aortic bioprosthetic valve: transcatheter aortic valve replacement or redo aortic valve replacement? Interact CardioVasc Thorac Surg 2015; doi:10.1093/icvts/ivv037.
What is the best approach in a patient with a failed aortic bioprosthetic valve: transcatheter aortic valve replacement or redo aortic valve replacement? Christos Tourmousogloua,*, Vivek Raob, Spiros Lalosc and Dimitrios Dougenisd a b c d
University of Ioannina Medical School, Ioannina, Greece Division of Cardiovascular Surgery, Peter Munk Cardiac Center, Toronto General Hospital, University of Toronto, Toronto, Canada Department of Cardiothoracic Surgery, Hippokratio General Hospital, Athens, Greece Department of Cardiothoracic Surgery, University Hospital of Patra, Patra, Greece
* Corresponding author. University of Ioannina Medical School, 13 Poutetsi St, 45332 Ioannina, Greece. Tel: +30-6977748313; e-mail: [email protected] (C.E. Tourmousoglou). Received 3 April 2014; received in revised form 29 January 2015; accepted 12 February 2015
Abstract A best evidence topic in cardiac surgery was written according to a structured protocol. The question addressed was whether transcatheter aortic valve-in-valve replacement (viv-TAVR) or redo aortic valve replacement (rAVR) is the best strategy in a patient with a degenerative bioprosthetic aortic valve. Altogether, 162 papers were found using the reported search, of which 12 represented the best evidence to answer the question. The authors, journal, date, country of publication, patient group, study type, outcomes and results of papers are tabulated. The results of the studies provided interesting results. All the studies are retrospective. Four papers reported the results of redo aortic valve replacement in patients with failed aortic bioprosthetic valve, six papers demonstrated their results with transcatheter aortic valve-in-valve replacement for the same indication and two papers reported their propensity-matched analysis of outcomes between viv-TAVR and rAVR in patients with previous cardiac surgery. Thirty-day mortality for rAVR was 2.3–15.5% and 0–17% for viv-TAVR. For rAVR, survival rate at 30 days was 83.6%, 76.1% at 1 year, 70.8% at 3 years, at 51.3–66% at 5 years, 61% at 8 years and 61.5% at 10 years. For viv-TAVR, the overall Kaplan–Meier survival rate at 1 year was 83.2%. After viv-TAVR at 1 year, 86.2% of surviving patients were at New York Heart Association (NYHA) class I/II. The complications after rAVR were stroke (4.6–5.8%), reoperation for bleeding (6.9–9.7%), low-cardiac output syndrome (9.9%) whereas complications after viv-TAVR at 30 days were major stroke (1.7%), aortic regurgitation of at least moderate degree (25%), new permanent pacemaker implantation rate (0–11%), ostial coronary obstruction (2%), need for implantation of a second device (5.7%) and major vascular complications (9.2%). It is noteworthy to mention that there is a valve-in-valve application that provides information to surgeons for choosing the correct size of the TAVR valve. Transcatheter aortic valve-in-valve procedures are clinically effective, at least in the short term, and could be an acceptable approach in selected high-risk patients with degenerative bioprosthetic valves. Redo AVR achieves acceptable medium and long-term results. Both techniques could be seen as complementary approaches for high-risk patients. Keywords: Aortic valve • Failed • Bioprosthesis • Implantation
INTRODUCTION A best evidence topic was constructed according to a structured protocol. This is fully described in ICVTS [1].
The patient has become increasingly dyspnoeic during the last year. Transthoracic echocardiography and transoesophageal echocardiography revealed severe aortic regurgitation (three-fourth). The valve has to be replaced. There are two options for the patient: transcatheter aortic valve-in-valve replacement or redo AVR.
THREE- PART QUESTION In [a patient with a degenerative bioprosthetic aortic valve] is [transcatheter aortic valve-in-valve replacement or redo aortic valve replacement] the best strategy in order to optimize [event-free survival]?
CLINICAL SCENARIO A 78-year old patient, with a history of previous aortic valve replacement with a bioprosthetic valve 20 years ago, comes to your office.
SEARCH STRATEGY A review of the English language literature was performed on Medline using the Ovid interface from 1980 to July 2014 [transcatheter aortic valve implantation OR transcatheter aortic valve replacement OR transcatheter aortic valve-in-valve procedure OR re-replacement of aortic valve OR redo aortic valve replacement) AND ( previous aortic valve replacement OR degenerative bioprostheses)].
© The Author 2015. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
BEST EVIDENCE TOPIC
BEST EVIDENCE TOPIC – ADULT CARDIAC
Interactive CardioVascular and Thoracic Surgery (2015) 1–7 doi:10.1093/icvts/ivv037
C. Tourmousoglou et al. / Interactive CardioVascular and Thoracic Surgery
2
Table 1: Best evidence papers Author, date and country Study type (level of evidence)
Patient group
Key outcomes
Results
Eitz et al. (2006), Ann Thorac Surg, Germany [2]
71 patients (1991–2004), mean duration of follow-up 3.3 ± 2.8 years
‘AVR-type’ Bioprosthesis Mechanical
69/71 patients 2/71 patients
‘Age group’ 80–84 years ≥85 years Median time since first AVR Operative mortality rate Mortality during follow-up Survival at 30 days, 1 year, 3 years and 5 years in patients with reoperation
Cohort study (level 2b)
Estimated median survival of patients with reoperations Davierwala et al. (2006), J Thorac Cardiovasc Surg, Canada [3] Cohort study (level 2b)
216 patients (1990–2002) underwent aortic valve rereplacements from a total group of 2673 patients that underwent aortic valve surgery
‘Redo-AVR type’ No. of operations Second operation Third operation Fourth operation Mechanical valve Bioprostheses Median duration since first AVR Hospital mortality ‘Postoperative outcomes’ Low-cardiac output syndrome Postoperative MI Postoperative stroke Postoperative renal failure Reoperation rate for bleeding Pacemaker insertion
Kempfert et al. (2010), Ann Thorac Surg, Germany [4]
11 patients (2007–2009), follow-up 330 ± 291 days
Cohort study (level 2b)
Mean age since first AVR Mean logistic EuroSCORE Mean STS score Successful valve-in-valve implantation Off-pump treatment ‘Postoperative outcomes’ 30-day mortality rate Stroke rate New pacemaker implantation rate Trans/paravalvular leak first degree Maximal AV gradient, mmHg Mean AV gradient, mmHg
Leontyev et al. (2011), Ann Thorac Surg, Germany [5] Cohort study (level 2b)
155 patients (1994–2008), mean duration of the follow-up 2.7 ± 2.8 years, 86 isolated redo AVR, 69 redo aortic root replacement
Mean age Mean logistic EuroSCORE Mean interval since first AVR Early mortality Reoperation for bleeding Stroke Low-cardiac output syndrome
90.2% 9.8% 8 years 15.5% 40.8% 83.6%, 76.1%, 70.8% and 51.3%, respectively
Comments
Low-cardiac output syndrome and intestinal failure were independent risk factors of 30-day survival in the cohort of reoperated patients Early and mid-term survival was predominantly influenced by unexpected postoperative complications and not by preoperative risk factors, with the exception of diabetes mellitus
5.6 years
191 patients 23 patients 2 patients 52% 48% 10.3 years 4.6%
4.6%
Patients in the redo AVR group had predominantly AR Hospital mortality was independently predicted by peripheral vascular disease, active endocarditis, worsening NYHA class and need for annular enlargement Reoperation itself did not predict hospital mortality
0.9% 4.6% 4.2% 6.9% 14% 7.5 ± 3.4 years 31.7 ± 14.5% 7.2 ± 2.6% 100%
None of the patients had new onset postoperative AV block
100%
0% 0% 0% 18.2% 21.4 ± 7.5 11 ± 4.2 58.1 ± 16.3 years 27 ± 23% 6.7 ± 7.9 years 4.5% 9.7% 5.8% 9%
Only one-quarter of patients who received isolated rAVS in this study were potentially suitable for a valve-in-valve transcatheter procedure
Continued
C. Tourmousoglou et al. / Interactive CardioVascular and Thoracic Surgery
BEST EVIDENCE TOPIC
3
Table 1: (Continued) Author, date and country Study type (level of evidence)
Eggebrecht et al. (2011), JACC Cardiovasc Interv, Germany [6]
Patient group
47 patients (2005–2010), trans-femoral (n = 25) or transapical (n = 22) valve-in-valve TAVR
Cohort study (level 2b)
Key outcomes
Results
Survival at 5 years and 8 years Freedom from reoperation at 5 years and 8 years
66 ± 5%, 61 ± 6%, respectively 87.1 ± 4%, 82.3 ± 6%, respectively
Mean age Time since first AVR Mean logistic EuroSCORE, % Mean STS score
79.8 ± 7.1 years 113 ± 65 months 35 ± 18.5%
‘Type of first bioprosthesis’ Stented valve Stentless aortic root valve
96% 4%
‘Prosthesis used for viv-TAVI’ Edwards-Sapien valve Medtronic CoreValve Procedural success Mean gradient after TAVI
75% 25% 98% 17 ± 10 (3–35) mmHg
‘Regurgitation after TAVI’ None Mild Moderate Procedural complications Vascular access complications Bail-out implantation of second TAVI prosthesis New permanent pacemaker implantation All-cause 30-day mortality rate Pasic et al. (2011), J Thorac Cardiovasc Surg, Germany [7]
14 patients (2008–2010), follow-up 20 months, transapical Edwards-Sapien valve
Cohort study (level 2b)
Mean age Mean time since first AVR Mean STS score Mean logistic EuroSCORE Procedural success Conversion to open surgery ‘Mean aortic transvalvular gradient’ preoperatively postoperatively New pacemaker implantation rate New clinical neurological deficit Reoperation rate during follow-up Aortic Regurgitation after TAVI (end of follow-up) Death rate during the follow-up
Bapat et al. (2012), J Thorac Cardiovasc Surg, UK [8] Cohort study (level 2b)
23 patients (2009–2011), mean follow-up 8.1 months, valve-in-valve (transapical 21/23)
11.6 ± 8.5%
Mean age ‘Type of AV used during initial operation’ Stented Stentless Mean time since first AVR for stented vs stentless valves
Comments
A trend towards lower gradients for Medtronic CoreValve viv-TAVI procedures in small, 21-mm bioprostheses compared with Edwards-Sapien valves was seen
55% 43% 2% 19% 13% 4% 11% 17% (1 procedural death and 7 in-hospital deaths) 73.3 ± 13.1 years 8.9 ± 4.7 years 21.9 ± 10.9% 45.3 ± 22.2% 100% 0%
There were three limitations in that study: (i) there was no control group with conventional AVR, (ii) the number of patients was small and (iii) the follow-up was limited
37.1 ± 25.7 mmHg 13.1 ± 6.4 mmHg 0% 0% 7% 0% 14% 79.6 ± 14.4 years
15/23 patients 8/23 patients 9.4 ± 3.8 years vs 14 ± 6.2 years
Large case series with short follow-up Concluded that patients with small bioprosthetic valves might be treated with VIV if the risk of conventional surgery is prohibitive
Continued
C. Tourmousoglou et al. / Interactive CardioVascular and Thoracic Surgery
4
Table 1: (Continued) Author, date and country Study type (level of evidence)
Linke et al. (2012), Circ Cardiovasc Interv, Germany [9]
Patient group
27 patients, mean follow-up 421 ± 198 days, TAVR with Medtronic CoreValve (trans-femoral)
Cohort study (level 2b)
Pechlivanidis et al. (2013), J Thorac Cardiovasc Surg, Italy [10] Cohort study (level 2b)
131 patients (2002–2010), mean follow-up 55.1 ± 32.4 months, underwent redo aortic valve replacement
Key outcomes
Results
Mean logistic EuroSCORE Mean STS score Procedural success rate Mean aortic valve gradient preprocedural vs post-procedural AR paravalvular postimplantation mild Permanent pacemaker implantation rate MyocardiaI Infarction rate In-hospital mortality 30-day mortality Late death rate during the follow-up
31.6 ± 15.3% 6.3 ± 3.8% 100% 31.2 ± 17.06 mmHg vs 9.13 ± 4.9 mmHg
Mean age Mean logistic EuroSCORE, Mean STS score Time since first AVR Conversion to open surgery MI intraprocedurally Death rate intraprocedurally Mean AV gradient reduction All-cause mortality at 30 days Major stroke Life-threatening bleeding Major access site complications New permanent pacemaker rate Survival rate at 6 months and at 1 year
74.8 ± 8 years 31.3 ± 16.5% 3–30.9% 67 ± 46 months 0%
Mean age
65.6 ± 12.9 years
‘Previous aortic valve prosthesis’ Biological Mechanical
80.9% 19.1%
‘Type of intervention’ Redo AVR Redo AVR plus CABG
75.5% 24.5%
‘Type of prosthesis used’ Biological Mechanical Homograft
65.6% 29% 5.3%
Hospital mortality Reoperation rate for bleeding Low output cardiac syndrome Stroke 10-year survival 10-year freedom from acute heart failure 10-year freedom from reoperations
Comments
22% 0% 0% 0% 0% 13%
0% 0% 42 ± 15 to 10 ± 8 mmHg
Trans-femoral implantation of Medtronic CoreValve into a wide range of failed bioprosthetic aortic valves was feasible, safe and resulted in marked improvement in haemodynamics in selected high-risk patients
7.4% 3.7% 7.4% 11.1% 3.7% 92%, 88% respectively
2.3% 9.2% 9.9% 3.1% 61.5 ± 8.6% 62.9 ± 6.9%
Survival during follow-up was lower, but not statistically significant, for older patients (patients >75 years vs patients with 2) in r-TAVI/r- AVR 30-day mortality in r-TAVI/r-AVR Actuarial survival rate at 4 years in r-TAVI/r-AVR Actuarial major stroke rate at 4 years in r-TAVI/ r-AVR Mean transvalvular gradient in r-TAVI/r-AVR
81 ± 4 years vs 80 ± 3 years
Paravalvular leaks in r-TAVI/r-AVR Grade I/II Grade III/IV
24 ± 6% vs 19 ± 6% 11.1 ± 2.8% vs 10.4 ± 3% 7 ± 5 vs 8 ± 6 years 3 vs 17% 0 vs 8% 0 vs 13% (P = 0.04) 8 vs 16% (P = 0.06)
Valve-in-valve outcomes were worse in patients with small surgical valves (label size ≤21 mm) and in patients with stenosis as mechanism of failure The registry showed that there was an elevated rate of residual leak in patients with baseline regurgitation
There was a trend towards a decreased 30-day mortality following r-TAVI The incidence of major adverse events including death and permanent neurological complications within 30 days postoperatively reached statistical significance between the two groups (there was a higher rate in the open surgical group)
75 ± 3% vs 73 ± 4% (P = 0.56) 25 vs 43% (P = 0.01)
9.8 ± 3 mmHg vs 10.1 ± 2 mmHg
33 vs 0% 0 vs 0%
NYHA: New York Heart Association; EuroSCORE: European System for Cardiac Operative Risk Evaluation; STS: society of thoracic surgeons.
SEARCH OUTCOME Using the reported search, 162 papers were identified of which 12 papers provided the best evidence to answer the question. These papers are summarized in Table 1.
RESULTS Eitz et al. [2] showed that perioperative mortality rate was 15.5% in reoperations of the aortic valve in a group of 71 octogenarians. Indications for surgery were structural valve deterioration (76.1%),
paravalvular leak (4.2%) and prosthetic valve endocarditis (11.3%). Survival rates at 30 days, 1 year, 3 years and 5 years were 83.6, 76.1, 70.8 and 51.3%, respectively. Estimated median survival was 5.6 years for patients with reoperations. Mid-term survival was similar when compared with a group of patients matched for age, gender and date of operation, having their first AVR. Davierwala et al. [3] showed that mortality rate was 4.6% in rAVR in 216 patients. The proportion of patients that received mechanical valves was 52% in the rAVR group, significantly higher than that in the first AVR operation in the same group of patients (17%). Reoperative aortic valve surgery was not a significant predictor of hospital mortality.
Kempfert et al. [4] showed that 30-day mortality was 0% in viv-TAVR in 11 patients. Stroke rate and new pacemaker implantation rate was 0%. Postoperatively, trans/paravalvular leak of first degree was 18.2% and mean AV gradient was 11 ± 4 mmHg. Leontyev et al. [5] demonstrated that early mortality was 4.5% for all rAVR in 155 patients. Indications for redo AV surgery were endocarditis (27.1%), paravalvular leak (18.1%), bioprosthetic structural valve deterioration (16.1%), endocarditis with root abscess (14.2%) and bioprosthetic degeneration (7.7%). Five- and 8-year survival was 66 ± 5% and 61 ± 6%, respectively. Eggebrecht et al. [6] showed that 30-day mortality was 17% in 47 patients that underwent viv-TAVR. Indications for surgery were isolated stenosis (47%), regurgitation (32%) and mixed stenosis and regurgitation (21%). Increased transvalvular gradients ≥20 mm were observed in 44% of patients, especially in those with bioprostheses of small diameters. Vascular access complications occurred in 13% of patients. New pacemaker implantation rate was 11%. Pasic et al. [7] showed that operative mortality was 0% in 14 patients that were treated with viv-TAVR. Mean aortic transvalvular gradient decreased from 37.1 ± 25.7 mmHg to 13.1 ± 6.4 mmHg, postoperatively. There was no postoperative valve insufficiency. Up to 20 months postoperatively, 86% of patients were in New York Heart Association (NYHA) class I or II. Bapat et al. [8] found that 30-day mortality was 0% in 14 patients that were treated with viv-TAVR. Indication for surgery was predominantly aortic valve stenosis (56%). Mean AV gradient decreased from 31.2 ± 17.06 mmHg to 9.13 ± 4.9 mmHg. Late death rate during the follow-up was 13%. Linke et al. [9] showed that all-cause mortality at 30 days was 7.4% in a group of 27 patients that were treated with viv-TAVR. Major stroke rate was 3.7% and major access site complications were 11.1%. Survival rate at 6 months and 1 year was 92 and 88%, respectively. Mean AV gradient decreased from 42 ± 15 to 10 ± 8 mmHg. Pechlivanidis et al. [10] demonstrated that hospital mortality was 2.3% in reoperations of the aortic valve in 131 patients. Ten-year survival was 61.5 ± 8.6%. Ten-year freedom from acute heart failure, reoperations, stroke and thromboembolisms were 62.9 ± 6.9%, 97.8 ± 1.5%, 93.2 ± 3% and 91.2 ± 3.2%, respectively. Redo AVRs achieved good results for elective cases, cases without endocarditis and NYHA class I/II cases. Wilbring et al. [11] showed that 30-day mortality was 9.4% in a group of 53 patients who were treated with viv-TAVR versus 5.7% in a propensity-matched group of 53 patients who were treated with conventional redo AVR. Survival at 6 months was 83% in the viv-TAVR group vs 86.8% in the rAVR group. Mean grade of paravalvular regurgitation was 0.8 ± 0.2 in the viv-TAVI group vs 0% in the rAVR group. Dvir et al. [12] showed that 30-day mortality was 7.6% in viv-TAVR in their international registry of 459 patients. Modes of aortic bioprosthesis failure were stenosis (39.4%), regurgitation (30.3%) and combined (30.3%). Major stroke rate was 1.7% and permanent pacemaker implantation rate was 8.3%. Overall 1-year Kaplan–Meier survival rate was 83.2%. Patients in the stenosis group had worse 1-year survival (76.6%) in comparison with the regurgitation group (91.2%) and the combined group (83.9%). Patients with small valves had also worse 1-year survival (74.8%) versus patients with intermediate-sized valves (81.8%) and patients with large valves (93.3%).
7
Papadopoulos et al. [13] showed that 30-day mortality was 8% in the viv-TAVR group versus 16% in the r-AVR group by using propensity analysis of matched subgroups of 40 patients. Actuarial survival at 4 years was 75 ± 3% in the viv-TAVR group vs 73 ± 4% in the rAVR group. Kaplan–Meier estimates of major adverse events (death and permanent neurological deficits) at 4 years were 25% for the viv-TAVR group vs 43% for the rAVR group.
CLINICAL BOTTOM LINE Both viv-TAVR and rAVR provided good clinical results in high-risk patients with previous cardiac surgery. Both techniques could be seen as complementary approaches. An interdisciplinary heart team is necessary to discuss every patient and choose the best individualized approach. Conflict of interest: none declared.
REFERENCES [1] Dunning J, Prendergast B, Mackway-Jones K. Towards evidence-based medicine in cardiothoracic surgery: best BETS. Interact CardioVasc Thorac Surg 2003;2:405–9. [2] Eitz T, Fritzsche D, Kleikamp G, Zittermann A, Horstkotte D, Korfer R. Reoperation of the aortic valve in octogenarians. Ann Thorac Surg 2006; 82:1385–91. [3] Davierwala P, Borger M, David T, Rao V, Maganti M, Yau T. Reoperation is not an independent predictor of mortality during aortic valve surgery. J Thorac Cardiovasc Surg 2006;131:329–35. [4] Kempfert J, van Linden A, Linke A, Borger M, Rastan A, Mukherjee C et al. Transapical off-pump valve in valve implantation in patients with degenerated aortic xenografts. Ann Thorac Surg 2010;89:1934–41. [5] Leontyev S, Borger M, Davierwala P, Walter T, Lehmann S, Kempfert J et al. Redo aortic valve surgery: early and late outcomes. Ann Thorac Surg 2011; 91:1120–6. [6] Eggebrecht H, Schafer U, Treede H, Boekstegers P, Babin-Ebell J, Ferrari M et al. Valve-in valve transcatheter aortic valve implantation for degenerated bioprosthetic heart valves. JACC Cardiovasc Interv 2011;4:1218–27. [7] Pasic M, Unbehaum A, Dreysse S, Buz S, Drews T, Kukucka M et al. Transapical aortic valve implantation after previous aortic valve replacement: clinical proof of the “valve-in-valve” concept. J Thorac Cardiovasc Surg 2011;142:270–7. [8] Bapat V, Attia R, Redwood S, Hancock J, Wilson K, Young C et al. Use of transcatheter heart valves for a valve-in-valve implantation in patients with degenerated aortic bioprosthesis: technical considerations and results. J Thorac Cardiovasc Surg 2012;144:1372–80. [9] Linke A, Woitek F, Merx M, Schiefer C, Mobius-Winkler S, Holzhey D et al. Valve-in-valve implantation of Medtronic CoreValve with failing bioprosthetic aortic valves. Circ Cardiovasc Intev 2012;5:689–97. [10] Pechlivanidis K, Onorati F, Petrilli G, Santini F, Milano A, Torre S et al. In which patients is transcatheter aortic valve replacement potentially better indicated than surgery for redo aortic valve disease? Long-term results of a 10-year surgical experience. J Thorac Cardiovasc Surg 2014; 148:500–8.e1. [11] Wilbring M, Tugtekin SM, Alexiou K, Simonis G, Matschke K, Kappert U. Transapical transcatheter aortic valve implantation vs conventional aortic valve replacement in high-risk patients with previous cardiac surgery: a propensity-score analysis. Eur J Cardiothorac Surg 2013;44:42–7. [12] Dvir D, Webb J, Bleiziffer S, Pasic M, Waksman R, Kodali S et al. For the Valve-in-Valve International Data Registry Investigators. Transcatheter aortic valve implantation in failed bioprosthetic surgical valves. JAMA 2014;312:162–70. [13] Papadopoulos N, Schiller N, Fichtlscherer S, Lehmann R, Weber C, Moritz A et al. Propensity matched analysis of longterm outcomes following transcatheter based aortic valve implantation versus classic aortic valve replacement in previous cardiac surgery. J Cardiothorac Surg 2014;9:99.
BEST EVIDENCE TOPIC
C. Tourmousoglou et al. / Interactive CardioVascular and Thoracic Surgery
