Article Text

Original research
Trends in transcatheter aortic valve implantation practice and clinical outcomes at an Irish tertiary referral centre
  1. Richard Tanner1,
  2. Daniele Giacoppo2,
  3. Hassan Saber2,
  4. David Barton2,
  5. Declan Sugrue2,
  6. Andrew Roy2,
  7. Gavin Blake1,2,
  8. Mark S Spence2,
  9. Ronan Margey2 and
  10. Ivan P Casserly1,2
  1. 1Department of Cardiology, Mater Misericordiae University Hospital, Dublin, Ireland
  2. 2Department of Cardiology, Mater Private Hospital, Dublin, Ireland
  1. Correspondence to Dr Ivan P Casserly; icasserly{at}


Objective A paucity of data exists on how transcatheter aortic valve implantation (TAVI) practice has evolved in Ireland. This study sought to analyse temporal trends in patient demographics, procedural characteristics, and clinical outcomes at an Irish tertiary referral centre.

Methods The prospective Mater TAVI database was divided into time tertiles based on when TAVI was performed: Group A, November 2008–April 2013; Group B, April 2013–September 2017; and Group C, September 2017–February 2022. Patient and procedural characteristics and clinical outcomes were compared across groups.

Results A total of 1063 (Group A, 59; Group B, 268; and Group C:, 736) patients were treated with TAVI during the study period (mean age 81.1±7.4, mean Society of Thoracic Surgeons score 5.9±5.1).

Conscious sedation (Group A, 0%; Group B, 59.9%; and Group C, 90.2%, p<0.001) and femoral artery access (Group A, 76.3%; Group B, 90.7%; and Group C, 96.6%, p<0.001) were used more frequently over time. The median length of hospital stay reduced from 9 days (IQR 7, 18) in Group A to 2 days (IQR 2, 3) in Group C. In-hospital death was numerically higher in Group A compared with Group C (6.8% vs 1.9%, p=0.078). At 1-year follow-up, the rate of death and/or stroke was similar in Group A and Group C (20.3% vs 12.0%, adjusted HR 1.49, 95% CI (0.59 to 3.74)).

Conclusion There was exponential growth in TAVI procedural volume during the study period. A minimalist approach to TAVI emerged, and this was associated with significantly shorter procedure duration and hospital stay. Clinical outcomes at 1-year follow-up did not change significantly over time.

  • Aortic Valve Stenosis
  • Transcatheter Aortic Valve Replacement

Data availability statement

Data are available upon reasonable request. Reasonable requests for data will be considered by the senior author, Professor Ivan Casserly.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Transcatheter aortic valve implantation (TAVI) has become the mainstay of treatment for severe aortic stenosis in older patients over the last decade.


  • TAVI has experienced exponential growth in Ireland since the first procedure in 2008.

  • A move to a minimalist TAVI procedure over the study period proved feasible and was associated with low complication rates, shorter procedure times, and hospital stays.

  • In keeping with TAVI practice in France, patients’ age and operative risk profile at our centre did not change significantly over time, suggesting potential differences in TAVI practice between Europe and North America.


  • This study supports a minimalist approach to TAVI in experienced centres. In addition, the need to appropriately resource hospitals to meet the growing demand for TAVI is underlined.


Transcatheter aortic valve implantation (TAVI) is now the default treatment for older patients with severe symptomatic aortic stenosis (AS).1–3 The transition from surgical aortic valve replacement (SAVR) to TAVI as the treatment of choice for AS in this patient cohort has been underpinned by cardiology societal guidelines and multiple randomised control trials showing non-inferiority or superiority of the transcatheter approach (figure 1).4–6

Figure 1

Pivotal clinical trials, guidelines and device iterations that have underpinned the evolution of TAVI in clinical practice. ACC, American College of Cardiology; AHA, American Heart Association; EACTS, European Association for Cardio-Thoracic Surgery; ESC, European Society of Cardiology; PARTNER, Placement of Aortic Transcatheter Valves trials; SAVR, surgical aortic valve replacement; SURTAVI, surgical replacement and transcatheter aortic valve implantation; TF, transfemoral; TAVI, transcatheter aortic valve implantation; VHD, valvular heart disease. *First recapturable system.

Within the Irish setting, limited published data are available on the use of TAVI, and a dearth of information exists on the evolution of TAVI practice since the procedure was first introduced.7–9 Given the exponential growth in TAVI across North America and other European countries and the move to treating younger patients, an incumbent need exists to understand the evolving characteristics of the population being treated in national TAVI programmes.1 2 10 Furthermore, clinical outcomes need to be scrutinised and benchmarked against international registries to ensure optimal patient care is being delivered and identify areas needing improvement. Additionally, knowledge of trends in TAVI practice for specific programmes enables future planning and resourcing at a local and national level.

Against this background, we sought to describe the clinical experience and temporal change in patient demographics, procedural characteristics, and clinical outcomes at a tertiary referral centre in Ireland.


Data collection

The prospective Mater TAVI database was analysed. This comprises all patients treated with TAVI for severe AS at two co-located hospitals, namely, the Mater Private Hospital (MPH) and Mater Misericordiae University Hospital (MMUH). Dendrite Clinical Systems designed the database and collects information under variable headings standard to other registry databases such as the STS/American College of Cardiology Transcatheter Valve Therapy (STS/ACC TVT) registry. Clinical and procedural data are entered prospectively. Dedicated structural clinical nurse specialists in MPH and MMUH with physician oversight are responsible for inputting data and maintaining the database. This approach facilitated efficient workflow and mitigated physician bias in reporting.

Patient selection

All patients treated with TAVI for severe AS since the inception of the TAVI programme in November 2008 through February 2022 were eligible for inclusion in this analysis. The extracted patient cohort was divided into time tertiles (Groups A, B and C) based on the procedure’s date. Groups A, B and C comprised patients treated with TAVI between November 2008–April 2013, April 2013–September 2017 and September 2017–February 2022, respectively.

Patient and public involvement

There was no patient or public involvement in this study.

Definitions and outcomes measured

Operative mortality risk was defined using the STS and the European System for Cardiac Operative Risk Evaluation (EuroSCORE) II risk assessment tools. High-risk, intermediate-risk and low-risk patients were defined based on an STS score of >8%, 4%–8% and <4%, respectively. Definitions for background medical conditions and clinical outcomes align with the Valve Academic Research Consortium-2 criteria.11 Procedural success was defined as the correct positioning of the valve in the absence of major cardiovascular and cerebrovascular events (death, stroke, conversion to open heart surgery and myocardial infarction). A neurologist or stroke physician reviewed all neurological events and adjudicated on stroke diagnosis. Baseline clinical and procedural characteristics and long-term clinical outcomes were compared between each time tertile (Groups A, B and C). In-hospital complications and major complication rates at 1-year follow-up within each group (A, B and C) were compared with examination for changes in clinical outcomes over the duration of the study.

Statistical analysis

Continuous variables were summarised as mean and SD or median and IQR and compared them through the analysis of variance or Kruskal-Wallis test according to the results of the Shapiro-Wilk test. The outcomes of interest were assessed at 1 year. The Kaplan-Meier method computed the incidence of events across tertiles and compared them using the log-rank test. The cumulative distribution over time and the number at risk at the beginning of each time interval were displayed. For each outcome, the risk between tertiles was estimated by Cox proportional hazards regression, and the results were reported as HRs and corresponding 95% CIs. The proportional hazards assumption was assessed by the Grambsch-Therneau test and scaled Schoenfeld residuals. Considering the potential influence of several confounding effects on crude risk estimates, any result was adjusted by multivariable Cox proportional hazards regression, including clinical and procedural variables.


A total of 1063 patients (Group A, 59; Group B, 268; and Group C, 736) underwent TAVI for severe AS during the study period and were included for analysis. For the overall patient cohort, the mean age was 81.1±7.4 years, 55.6% were male and the mean STS score was 5.9±5.1. The number of TAVIs performed per year significantly increased during the study period, from a low of 5 in 2010 to a high of 227 in 2021 (figure 2).

Figure 2

Number of TAVI procedures per year. TAVI, transcatheter aortic valve implantation. *First TAVI performed in November 2008.

Patient characteristics

During the study period, there was no significant change between groups in the mean age of patients treated with TAVI (p=0.439). The frequency of patients with chronic lung disease (p<0.001) or a background of coronary artery bypass graft (p<0.001) reduced significantly over time. However, the proportion of patients with a low STS score (<4%) did not change significantly. Computed tomography angiography (CTA) replaced intraprocedure transoesophageal echocardiography (TOE) as the modality of choice for assessing aortic valve annular dimensions in Groups B and C. Transthoracic echo (TTE) aortic valve measurements were consistent with severe AS in all groups, with a mean transvalvular gradient of 45.7±13.6 mm Hg, mean peak systolic velocity of 4.3±0.6 metres/second and mean aortic valve area of 0.7±0.2 cm2 for the overall cohort. Severe left ventricular dysfunction was present in 12.7% (n=135) of all patients, with no significant differences between groups (table 1).

Table 1

Baseline clinical characteristics

Procedural characteristics

Conscious sedation replaced general anaesthesia (GA) as the sedation modality over time. Femoral artery access was the preferred access site in all groups. Group A and Group C had the lowest (76.3%, n=45) and highest (96.6%, n=712) proportion of patients being treated via the femoral artery. A total of 29 transapical cases were performed, all in Group A (23.7%) or Group B (5.6%). Although the range of alternate access options expanded in Groups B and C, the proportion of patients treated with TAVI via non-transapical alternative access remained small (<3.5%). Balloon-expandable (BE) valves were chosen in >80% of all patients, with the Edwards Lifesciences Sapien BE family of valves being used in all cases. Meanwhile, the Medtronic Evolut family of valves was the most commonly used self-expanding valve (13.4%, n=142). During the study period, procedure duration reduced significantly, with Groups A, B and C having a median duration of 120 min (IQR 105, 150), 75 min (IQR 60, 95) and 65 min (IQR 50, 80), respectively, p<0.001 (table 2).

Table 2

Procedural characteristics

In-hospital complications

Procedural success was achieved in 95.3% of the overall patient cohogrt: Group A, 84.7%; Group B, 96.3%; and Group C, 96.0%. In-hospital all-cause mortality reduced numerically over the study period: Group A, 6.8%; Group B, 2.2%; and Group C, 1.9%; p=0.078 (table 1, figure 3). The rate of in-hospital stroke was highest in Group A (3.4%, n=2), and a similar rate of stroke was recorded in Group B (0.7%, n=2) and Group C (0.9%, n=7) (p=0.211). Major vascular complications were reduced to <1% in Groups B and C. Five cases (0.5%) of annular rupture were recorded for the overall cohort, with all these patients undergoing TAVI with a BE valve. Four patients underwent emergency open heart surgery during the study period, with the indication being annular rupture in three cases and ventricular perforation in one case. Ultimately, three of these four patients survived to hospital discharge. Overall, 4.3% of the study cohort required an in-hospital permanent pacemaker (PPM). A significant stepwise reduction in the duration of inpatient stay was seen from Group A through to Group C: 9 days (IQR 7, 18), 3 days (IQR 2, 6) and 2 days (IQR 2, 3) for Groups A, B and C, respectively. For the overall patient cohort, survival to hospital discharge was 97.8% (n=1039) with a non-significant trend towards increased survival over the study period: for Group A, 93.1%; Group B, 97.7%; and Group C, 98.1%, p=0.056. Concerning antiplatelet therapy, Group A was predominately discharged on dual antiplatelet therapy (83.6%), whereas the most common discharge antiplatelet therapy for Group C was aspirin monotherapy (48.2%) (online supplemental table 1) (tables 3 and 4).

Figure 3

Major in-hospital adverse events. Group A, November 2008–April 2013; Group B, April 2013–September 2017; and Group C, September 2017–February 2022. PPM, permanent pacemaker.

Table 3

In-hospital complications

Table 4

Clinical outcomes at 1-year follow-up with Kaplan-Meier event incidence

Clinical outcomes at 1-year follow-up

At 1-year follow-up, the Kaplan-Meier estimate of stroke and/or death was numerically lower in Group B (11.1%) and Group C (12.0%) compared with Group A (20.3%). The adjusted (Adj) risk of stroke or death at 1 year after TAVI was not significantly different in Group B compared with Group A (HRAdj 1.19, 95% CI (0.50 to 2.82)) or Group C compared with Group A (HRAdj 1.49, 95% CI (0.59 to 3.75)). After adjustment, the rate of death in Group B was not significantly different from Group A (HRAdj 1.51 95%CI (0.58 to 3.93)) or Group C compared with Group A (HRAdj 1.82, 95% CI (0.66 to 5.07)) (table 4, figure 4).

Figure 4

All-cause death or stroke for each group. Group A (1st time tertile), November 2008–April 2013; Group B (2nd time tertile), April 2013–September 2017; and Group C (3rd time tertile), September 2017–February 2022.


This study reports the largest analysis of TAVI practice in Ireland to date. An exponential growth in the number of TAVI procedures performed per year was observed over the 13-year study period. Although there were changes in some patient characteristics, there was no significant change in patient age or estimated surgical risk of the cohort over time. Key changes in TAVI practice included an increase in the proportion of patients being treated under conscious section and using TF access, a reduction in the use of balloon dilation before TAVI and abandonment of the transapical access route in favour of other alternate access sites in non-TF TAVI cases. There was a significant shortening in procedural duration and hospital length of stay over time. While there was a non-significant trend towards improved survival to hospital discharge over the study period, there was no significant change in the 1-year estimate of death, stroke or the combined endpoint of stroke and/or death.

Patient population and characteristics

During the initial years of this study, the number of TAVIs performed per year was limited by severe financial constraints within the public health system and significant restrictions in coverage by private health insurers in Ireland. However, since 2015, the increase in cases performed per year has been exponential, with a greater than fourfold increase in cases performed in 2021 (n=227) relative to 2015 (n=50). Furthermore, growth in the latter 2 years of the study progressed despite the implementation of extensive COVID-19 restrictions. Interestingly, despite the publication of intermediate-risk and low-risk TAVI trials, which included younger patients during the study period, the mean age of patients (81.1±7.4 years) being treated at this centre did not change significantly.3 12 This finding contrasts with a significant trend in treating younger patients in the USA.10 However, it is consistent with evidence from the FRANCE 2 TAVI registry that reported an unchanged median age (83 years (IQR 79, 87)) of patients treated between 2010 and 2021.1 13 14 In addition, we found no significant changes in the operative risk (STS score and EuroSCORE II) of patients undergoing TAVI over time. Hence, the growth in TAVR volume did not reflect a move to treating younger and lower-risk patients. It is plausible that the increase in volume was due to physicians gaining confidence in referring patients for TAVI, given the publication of multiple trials showing TAVI to be effective in treating severe AS during the study period. Furthermore, patient confidence in TAVI likely increased as the procedure transitioned to the mainstay of treatment for AS in the elderly population. For now, practice in Ireland has been to reserve TAVI for older patients with multiple comorbidities, while younger patients are still treated with SAVR. However, this is an evolving field, and younger patients will likely be treated going forward, given recent guideline changes (figure 1).15 16

Move to a ‘minimalist’ TAVI procedure

The move towards a more streamlined and simplified TAVI procedure occurred throughout the study. An essential element in this transition was the use of conscious sedation rather than GA for TAVI procedures. This approach is endorsed by studies showing comparable safety to GA and additional benefits such as reduced cost and length of hospital stay.17 18 Furthermore, a significant number of patients in this study had their sedation managed by the cardiac team in the absence of an anaesthetist (25%). This approach is supported by research from our centre and modest-sized studies from Israel (n=174) and Germany (n=292), which found no adverse impact on in in-hospital or 30-day outcomes using this strategy to management of sedation.9 19 20 Importantly, these findings are from experienced TAVI centres where timely input from anaesthesiology is available if needed and hence may not be generalisable to all centres. Routine use of preprocedure CTA to assess annular dimension rather than intraprocedure TOE has also played a vital role in the move to a minimalist approach. Finally, the use of balloon predilation significantly reduced over time. This finding reflects evidence that emerged during the study period that found a lack of clinical benefit to routine predilation.21–23 These changes in procedural characteristics were facilitated by growth in the cardiac team’s experience over time and coincided with a significant reduction in procedure duration and hospital length of stay after TAVI. Notably, the minimalist approach to TAVI and the efficiencies achieved are critical to the delivery and growth of a TAVI programme, particularly in a public health system that is frequently limited by resource limitations.

Vascular access sites for TAVI

TF access was used in 96.6% of patients in the most contemporary tertile of this cohort. Repeated studies have shown TF-TAVI to be associated with lower in-hospital mortality and stroke.16 24 Hence, the TF route is strongly favoured at our centre, even when this might require surgical cut-down of the common femoral artery or adjunctive iliac angioplasty. Despite new iterations of valve delivery systems, this study demonstrates that 3% and 4% of patients remain unsuitable for TF-TAVI in contemporary practice. Subclavian artery access (via surgical cut-down) and transcaval access have replaced transapical access as the favoured non-TF access sites for TAVI in this cohort. Subclavian access was the leading alternate access route between 2016 and 2019 in the STS/ACC TVT registry, followed by transaxillary access, accounting for 1.5% and 0.97% of all TAVI cases, respectively, in 2019. However, recent evidence has suggested that subclavian/axillary access is associated with a higher rate of stroke compared with TF access at 30 days (6.5% vs 3.5%).25 Moreover, certain anatomical features (vessel tortuosity, extreme root angulation, and patent internal thoracic artery graft) can preclude this access option. These limitations, the advantage of a fully percutaneous procedure and the favourable ergonomics of operating from the femoral position have led to the adoption of transcaval access (n=12) for non-TF TAVI at our centre. A recent study (n=238 transcaval, n=106 axillary and n=132 TF) found transcaval access TAVI to have a significantly lower risk of stroke and/or TIA compared with axillary access (2.5% vs 13.2%, p=0.01) and a comparable frequency of major bleeding.24

Alternate access TAVI will likely continue to evolve and provide a treatment option for the small percentage of patients unsuitable for TF-TAVI. Given the low volume of patients requiring alternate access to TAVI and the unique challenges associated with each approach, it makes sense that these procedures are performed at a referral specialist centre to achieve optimal clinical outcomes.

Procedure-related complications

The rate of in-hospital complications recorded in this study compares favourably with outcome data from large European and US registries, accepting the limitation of comparing heterogeneous patient cohorts from different time periods (online supplemental table 2).1 26 27 Although not reaching statistical significance, the reduction in in-hospital all-cause mortality during the study period from 6.8% in Group A to 1.9% in Group C is encouraging and is similar to the rate recorded in the most recent STS/ACC TVT registry publication.1 Furthermore, given that the mean age and overall risk profile (STS, EuroSCORE II) of patients did not change significantly between Groups A, B and C, our finding of no significant change in in-hospital mortality is not unexpected.

Furthermore, an in-hospital stroke rate of <1% in Groups B and C compares favourably with available registry data and is lower than reported in the control arm of the recently published cerebral embolic protection PROTECTED TAVR trial (2.9%).1 26 28 An outlier in our data when compared with other registries is the lower requirement for PPM insertion before hospital discharge. This may be attributed to the high proportion of BE valves implanted (82.4%) and the use of three-dimensional rotational angiography to accurately delineate the aortic valve annular plane during the early years of the TAVI programme.29 The rate of conversion to open heart surgery was reassuringly low (0.4%) during the study period. Although infrequent, the need for urgent cardiothoracic support reaffirms the need for TAVI to be performed at a centre with on-site cardiothoracic surgery. Furthermore, while acknowledging selection bias, the finding that three out of the four cases in this series that required emergent conversion to open heart surgery survived to hospital discharge underlines how these emergency cases can be salvaged when the appropriate support is available. Finally, operator experience improved during the study period, likely contributing to the low in-hospital complication rate, particularly in Groups B and C.

Clinical outcomes at 1 year

Although there was no statistically significant difference in the frequency of all-cause death or stroke across groups at 1-year follow-up, the numerical drop in the Kaplan-Meier estimate from 20.3% in Group A to 12% in Group C is a reassuring finding. In addition, these outcomes are similar to the STS/ACC TVT registry, which reported the frequency of all-cause death at 1 year of 24.3% before 2013 and 12.5% in 2018.1 Finally, beyond discharge, the need for vigilance and regular review is underlined by the number of TAVI cases that undergo PPM insertion between hospital discharge and 1 year (3.9% and 9.8%, respectively).


This study has some notable limitations. The data presented in this study are non-randomised and should be interpreted within the inherent limitations of a registry-based study. In addition, the distribution of cases between time tertiles is not equal due to the exponential growth in the number of TAVIs performed per year. Moreover, the findings of this study are mainly applicable to patients treated with BE valves, as these were used in >80% of patients. Furthermore, the low rate of PPM insertion may have been influenced by a high threshold of clinicians for pacing, which is impossible to quantify.


This study showed an exponential growth in TAVI procedural volume over the study period between 2008 and 2022. A minimalist approach to TAVI, including increased use of conscious sedation and femoral arterial access, was associated with significantly shorter procedure times and hospital stay. In-hospital complication rates and clinical outcomes at 1 year were reassuringly similar to those recorded in large European and North American TAVI registries.

Data availability statement

Data are available upon reasonable request. Reasonable requests for data will be considered by the senior author, Professor Ivan Casserly.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants, and the ethical approval for this study was obtained from the MMUH and MPH Institutional Review Board (Reference: 1/378/2150 TMR). However, patients provided consent for their information to be stored in our database for the purpose of research and publication of outcome data.


We acknowledge our structural clinical nurse specialists (Barbara Moran, Jamie Byrne, Susan Groarke and Jacqueline Geraghty) for their dedication to patient care and contributions to maintaining the Mater TAVI database.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Twitter @RichCardioHST

  • Contributors I confirm the contribution of all the coauthors attested to the following: (1) conception and design or analysis and interpretation of data or both, (2) drafting of the manuscript or revising it critically for important intellectual content and (3) final approval of the manuscript submitted. IC is the guarantor for this study.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.