International Journal of Cardiology 189 (2015) 282–288

Contents lists available at ScienceDirect

International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard

Transcatheter aortic valve implantation in patients with bicuspid aortic valve: A patient level multi-center analysis Altayyeb Yousef a,1, Trevor Simard a,1, John Webb b, Josep Rodés-Cabau c, Charis Costopoulos d,e,f, Janusz Kochman g, José M. Hernández-Garcia h, Paul T.L. Chiam i, Robert C. Welsh j, Harindra C. Wijeysundera k,l,m, Eulogio García n, Henrique B. Ribeiro c, Azeem Latib d,e, Zenon Huczek g, Miriam Shanks j, Luca Testa o, Michael E. Farkouh p,q, Danny Dvir b, James L. Velianou r, Buu-Khanh Lam a, Ali Pourdjabbar a, Christopher Glover a, Benjamin Hibbert a, Marino Labinaz a,⁎ a

Division of Cardiology, University of Ottawa Heart Institute, Ottawa, Ontario, Canada Division of Cardiology, St. Paul's Hospital, University of British Columbia, Vancouver, British Columbia, Canada Quebec Heart and Lung Institute, Quebec City, Quebec, Canada d Interventional Cardiology Unit, San Raffaele Scientific Institute, Milan, Italy e Interventional Cardiology Unit, EMO-GVM Centro Cuore Columbus, Milan, Italy f Interventional Cardiology Unit, Imperial College London, London, United Kingdom g Ist Department of Cardiology, The Medical University of Warsaw, Poland h Hospital Universitario Virgen de la Victoria, Málaga, Spain i Department of Cardiology, National Heart Centre Singapore, The Heart And Vascular Centre, Mount Elizabeth Medical Centre, Singapore j Department of Cardiology, Mazankowski Alberta Heart Institute, University of Alberta, Alberta, Canada k Schulich Heart Center, Department of Medicine, Sunnybrok Health Sciences Center, University of Toronto, Toronto, Ontario, Canada l Institute for Clinical Evaluative Sciences (ICES), Toronto, Canada m Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, Canada n Hospital Universitario Clínico San Carlos, Madrid, Spain o Department of Cardiology, IRCCS Pol. San Donato, San Donato Milanese, Milan, Italy p Peter Munk Center, University Health Network, Toronto, Ontario, Canada q Richard Lewar Centre in Cardiovascular Research, University of Toronto, Toronto, Ontario, Canada r Hamilton Health Sciences, McMaster University, Hamilton, Ontario, Canada b c

a r t i c l e

i n f o

Article history: Received 19 February 2015 Accepted 9 April 2015 Available online 11 April 2015 Keywords: Transcatheter aortic valve implantation (TAVI) Bicuspid aortic valve Aortic valve stenosis Congenital aortic valve stenosis

a b s t r a c t Objective: We sought to evaluate the safety and efficacy of transcatheter aortic valve implantation (TAVI) in patients with bicuspid aortic valve (BiAV). Background: BiAV remains a relative contraindication to TAVI resulting in exclusion from TAVI trials and thus limiting data on the clinical performance of transcatheter valves in these patients. Methodology: We conducted an international patient level multicenter analysis on outcomes in patients with BiAV undergoing TAVI. The primary outcome of the study was the combined early safety endpoint — a composite of 30 day mortality, stroke, life-threatening bleeding, acute kidney injury, coronary artery obstruction, major vascular complication and valve related dysfunction. Secondary endpoints included the individual components of the primary endpoint as well as post-TAVI paravalvular leak (PVL), rehospitalization, new pacemaker insertion and device success rates at 30 days and 1 year. Results: A total of 108 patients with BiAV were identified in 21 centers in Canada, Spain, Italy, Poland and Singapore who underwent TAVI between January 2005 and March 2014. The composite primary outcome occurred in one quarter of patients (26.9%) — mainly driven by re-intervention for valve malposition (9.3%). The 30-day and 1 year mortality rates were 8.3% and 16.9% respectively with AR ≥ 3+ occurring in 9.6% of patients. Device success was achieved in 85.2% of cases with pacemaker insertion in 19.4%. While PVL was not associated with an increased risk of 30 day or 1 year mortality — Type I BiAV anatomy with left and right cusp fusion had significantly better outcomes than other valve variants.

⁎ Corresponding author at: University of Ottawa Heart Institute, 40 Ruskin Street Room 1264, Ottawa, Ontario K1Y 4W7, Canada. E-mail address: [email protected] (M. Labinaz). 1 Equally contributing authors.

http://dx.doi.org/10.1016/j.ijcard.2015.04.066 0167-5273/© 2015 Elsevier Ireland Ltd. All rights reserved.

A. Yousef et al. / International Journal of Cardiology 189 (2015) 282–288

283

Conclusion: In selected patients with BiAV and severe aortic stenosis, TAVI appears both safe and feasible with acceptable clinical outcomes. Clinical studies of TAVI in this patient population are warranted. © 2015 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

2.2. Outcomes and definitions

Bicuspid aortic valve (BiAV) is the most common congenital cardiac abnormality and is the leading cause of aortic stenosis (AS) in patients under the age of 65 [1,2]. Notably, a recent study has suggested that up to 20% of stenotic aortic valves in patients over 80 years of age are congenitally bicuspid [3]. As a result, a large cohort of elderly patients with symptomatic severe AS and elevated operative risk will require intervention and present a unique set of management challenges for the heart teams managing them. Transcatheter aortic valve implantation (TAVI) is the standard of care for managing elderly patients with high operative risk suffering from symptomatic severe aortic valve stenosis [4,5]. While the benefits of TAVI are clear, recent guidelines have omitted patients with BiAVs from their recommendations [6,7]. This stems largely from the theoretical risk of para-valvular leak (PVL), device expansion failure and due to the exclusion of patients with BiAV from major clinical trials evaluating TAVI for aortic stenosis. We recently performed a systematic review of TAVI in patients with BiAVs reporting similar short and long-term survival rates to those reported for patients with TAVIs [8]. However, limited conclusions may be drawn from this study given the relatively small number of case series and case reports included in the analysis. Accordingly, to ascertain the clinical performance of TAVI in patients with BiAV we performed this international, multi-center, patient-level analysis of consecutive cases.

Based on the Valve Academic Research Consortium-2 consensus document (VARC-2) definitions, we employed a primary composite endpoint of combined early safety at 30 days, which is described in details in VARC-2 article [10]. In short, this includes all-cause mortality, all stroke, life-threatening bleeding, acute kidney injury — stage 2 or 3, coronary artery obstruction requiring intervention, major vascular complication, and valve related dysfunction requiring intervention. Our secondary endpoints included the individual components of the primary composite endpoint, device success at both 30 day and 1 year outcomes. When available, centers reported 1 year mortality, valvular related rehospitalization, aortic regurgitation (AR), need for pacemaker, and prosthetic valve dysfunction. The latter is defined as mean aortic valve pressure gradient ≥ 20 mm Hg, AVA ≤ 1.1 or moderate to severe AR (AR ≥ 3+) [10]. Aortic regurgitation was categorized as paravalvular, transvalvular or mixed and was graded as none/trace (0), mild (1+), moderate (2+), moderate–severe (3+) or severe (4+). The ellipticity ratio was defined as longest/shortest axis of the aortic valve annulus. The aortic valve annulus was measured by MSCT or TEE according to the protocol of the individual participating institute.

2. Methods 2.1. Study population We included all consecutive patients with BiAV undergoing TAVI from each participating center. Inclusion required clear documentation of the presence of a bicuspid aortic valve by each participating center. This was achieved by short axis images of aortic valve obtained either by transthoracic echocardiogram/transesophageal echocardiogram (TTE or TEE) or Multiscliced computed tomography (MSCT) imaging. BiAV was defined as abnormal aortic valve morphology consisting of 2 anatomical or functional cusps with less than three zones of parallel apposition between cusps [9]. All participating centers reviewed and thereafter confirmed the diagnosis and (where available) classification of BiAV retrospectively. The BiAV anatomical classification definitions were adopted from Sievers and Schmidtke classification system, which classifies bicuspid valves based on the number of raphe and fused leaflets seen on imaging [9]. When both TEE and MSCT were performed, cases were excluded if the diagnosis of BAV was not consistent or remained speculative. Patients were included if they had symptomatic severe aortic stenosis, suitable aortic and vascular anatomy for TAVI, and high surgical risk as determined by the centers' heart team. Patients were excluded if they had functional but not anatomical bicuspid aortic valves, if there was a lack of 30-day follow-up, or if the TAVI was aborted due to identification of a BiAV intra-procedurally. Each center's human research and ethics committees approved their respective registries. Anonymized data was transferred to the University of Ottawa Heart Institute for Central Core-Laboratory analysis.

2.3. Statistical analysis All continuous variables were described as mean ± standard deviation for normally distributed variables and medians for others. Categorical variables described as number (%). For composite endpoints all components are reported individually [11]. For unadjusted analysis of the primary and secondary outcomes comparisons were made using Chi-square test or Fisher test as appropriate. All analyses were performed using SAS software version 9.2. 3. Results 3.1. Patient cohort From January 2005 to April 2014, 5828 TAVI procedures were performed in 21 centers Fig. 1. In total, patients were from five countries — Canada (5 centers), Spain (10 centers), Italy (2 centers), Poland (3 centers) and Singapore (1 center). In these centers, only 110 (1.9%) TAVIs were performed in patients with stenotic BiAVs. Based on our exclusion criteria, 2 patients were excluded (1 patient due to inability to deliver the transcatheter across aortic arch and the other patient was lost to 30 day post-TAVI followup). Baseline characteristics are described in Table 1. Briefly, the mean age of the cohort is 75.5 years with 63.9% of the patients being male. Dyspnea was the most common presenting symptom in 94.4% — with the majority of patients presenting with NYHA III–IV symptoms. Despite having a mean Logistic Euroscore of 17.2, patients were denied SAVR based on the treating heart team assessment due to increased frailty, hostile chest, porcelain aorta or other factors precluding them from undergoing surgery. As expected, pre-TAVI patients had low aortic valve areas (AVA) with a median of 0.7 cm2 and a mean aortic valve pressure gradient of 48.4 mm Hg (Table 2) that fell to 10.5 mm Hg post-intervention and the median AVA increased from 0.7 to 1.6 cm2 post-TAVI and was maintained out to 1 year. Notably, the diagnosis of BiAV was missed in 18.7% of patients on echocardiogram and was diagnosed incidentally by MSCT. Overall, 56.5% of our cohort received the balloon expandable

284

A. Yousef et al. / International Journal of Cardiology 189 (2015) 282–288

Total TAVI Procedures – 5,828 Total number of centers – 21 (January 2005-April 2014) 110 TAVIs in patients with BAV 2 Patients excluded: 1 – Device undeliverable due to aortic anatomy 1 – Mortality at 30 days is not available

108 Patients analyzed Canada – 5 centers (43 patients) Italy – 2 centers (21 patients) Singapore –1 center (4 patients) Spain – 10 centers (23 patients) Poland – 3 centers (17 patients)

89 1-year follow-up available Fig. 1. Flow diagram of study centers and patients.

valves and transfemoral access was the preferred access site for the procedures (83.3%). 3.2. Primary outcomes The combined composite early safety endpoint at 30 day occurred in 29 patients (26.9%, Table 3) — primarily being driven by 10 cases requiring re-interventions (second TAVI or conversion to surgery) for valve malposition (9.3%). There was total of 9 deaths (8.3%), with 6 cases of cardiovascular mortalities (Table 4). Furthermore, 7 patients (6.5%) that had life threatening bleeding, major vascular complication and acute kidney injury were reported. Similar event rates were observed between balloon expandable and self-expanding prostheses (Supplemental Table 1). 3.3. Secondary outcomes 3.3.1. 30 day outcomes In the total cohort, 92 patients (85.2%) met the VARC-2 definition of device success — a result mainly driven by ≥3+ AR occurring in 10 patients (9.6%) and 10 patients with valve malposition from first deployment (Table 4). There was one intra-procedure death due to annulus rupture in a patient with balloon expandable transcatheter valve. New pacemaker insertion was required in 21 patients (19.4%) of the total cohort, No patients suffered from NYHA class III/IV dyspnea at 30 day postTAVI.

3.3.1. 1 year outcomes. One year follow-up data was available in 89 patients with only one patient lost to followup. The one year mortality rate was 16.9% (15 patients) with 7 deaths occurring after 30 days. Of these 7 patients, 3 were due to cardiovascular mortality (Table 4). Despite the need for rehospitalization of 6 patients within the first year post-TAVI, only 2 patients (4.3%) remained symptomatic with NYHA class III or IV. 3.3.2. Paravalvular leak The major concern regarding TAVI in patients with BiAV is the theoretical risk of increased rates of significant paravalvular AR. Indeed, 30.8% and 27.7% had ≥2+ AR at 30 day and 1 year respectively. One patient had mixed aortic regurgitation, one patient had transvalvular aortic regurgitation and the remainder were attributed to paravalvular leak. We analyzed the impact of ≥2+ AR on outcomes (Supplemental Table 2). Of the total cohort, 104 patients had echocardiographic follow-up of which 32 (30.8%) had ≥2+ AR. The combined early safety endpoint occurred more frequently with AR ≥2+ (40.6% vs. 18.1%, p = 0.01, Fig. 2) driven primarily by more frequent valve malposition requiring additional interventions (18.8% vs. 4.2%, p = 0.04). Importantly, neither 30-day mortality (9.4% vs. 5.6%, p = 0.76) nor 1 year mortality (19.2% vs. 18.2%, p = 0.75) was higher in patients with ≥2+ AR. 3.3.3. BiAV anatomy and procedural outcomes Data was available on the bicuspid valve anatomy in 76 patients. As expected, Type I with left and right fusion was the most common variant and was identified in 44 patients (56.4%, Supplemental Table 3).

A. Yousef et al. / International Journal of Cardiology 189 (2015) 282–288

285

Table 1 Baseline characteristics. Total cohort (n = 108)

Edward Sapien (n = 61)

Patients demographics Age, year (SD) Male — n (%) BMI — kg/m2 (SD) Smoking history — n (%)

75.5 (14.4) 69 (63.9%) 28.0 (7.1) 43 (39.8%)

74.4 (11.7) 44 (72.1%) 27.7 (6.5%) 24 (39.3%)

77.0 (8.0) 25 (53.2%) 28.4 (5.5) 19 (42.2%)

Presenting symptoms Dyspnea — n (%) NYHA grades III–IV Angina — n (%) Angina (CCS classes III–IV) — n (%) Syncope — n (%)

102 (94.4%) 74 (72.5%) 26 (24.1%) 11 (10.2%) 11 (10.5%)

60 (98.4%) 41 (67.2%) 15 (24.6%) 8 (16.0%) 8 (13.6%)

42 (89.4%) 33 (70.2%) 11 (23.4%) 3 (6.7%) 3 (6.7%)

54 (50.0%) 1 (0.9%) 53 (49.1%)

34 (55.7%) 0 (0.0%) 27 (44.3%)

20 (42.6%) 1 (2.1%) 26 (55.3%)

77 (72.0%) 40 (37.4%) 59 (55.7%) 26 (24.5%) 29 (27.1%) 20 (18.7%) 20 (18.7%) 23 (21.7%) 4 (4.3%) 0 (0.0%) 1 (0.9%) 24 (22.6%) 7 (6.6%) 30 (31.9%) 30 (31.9%) 6 (6.4%) 94.4 (80–120) 53.8 (40–69) 56 (51.9%) 1 (0.9%) 119 (34.8) 17.2 (12.2)

42 (68.9%) 22 (36.1%) 38 (63.3%) 13 (21.7%) 16 (26.7%) 14 (23.3%) 9 (15.3%) 11 (18.3%) 0 (0.0%) 0 (0.0%) 1 (1.7%) 12 (20.0%) 6 (10.0%) 17 (32.7%) 15 (28.3%) 5 (9.6%) 93.0 (74–121) 51.7 (39.8–80.5) 28 (45.9%) 1 (1.7%) 122.9 (16.0) 16.1 (13.8)

TAVI indications AS alone — n (%) AR alone — n (%) AS + AR — n (%) Past medical history Hypertension — n (%) Diabetes mellitus — n (%) Dyslipidemia — n (%) Prior myocardial infarction — n (%) Prior percutaneous coronary intervention — n (%) Prior coronary artery bypass graft — n (%) Prior stroke — n (%) Peripheral vascular disease — n (%) Prior carotid surgery — n (%) Prior valve surgery — n (%) Aortic coarctation intervention — n (%) Atrial fibrillation/flutter — n (%) Defibrillator or biventricular pace — n (%) Pulmonary hypertension — n (%) Chronic obstructive pulmonary disease — n (%) Cancer — n (%) Creatinine — mmol/L (IQR) eGFR — mL/min/1.73 m2 (IQR) eGFR b 60 mL/min/1.73 m2 — n (%) Dialysis — n (%) Hemoglobin g/dL (SD) Logistic EuroScore (SD)

Core valve (n = 47)

35 (76.1%) 18 (39.1%) 21 (45.7%) 13 (28.3%) 13 (27.7%) 6 (12.8%) 11 (23.4%) 12 (26.1%) 4 (9.8%) 0 (0.0%) 0 (0.0%) 12 (26.1%) 1 (2.2%) 13 (29.5%) 15 (34.9%) 1 (2.4%) 96.8 (85.5–115.9) 54.0 (42.1–61.1) 28 (60.0%) 0 (0.0%) 107.6 (41.0) 19.4 (11.9)

BMI: body mass index; NYHA: New York Heart Association; CCS: Canadian Cardiovascular score; AS: aortic stenosis; AR: aortic regurgitation; TAVI: transcutaneous aortic valve implantation; eGFR: estimated Glomerular Filtration Rate.

When compared to other types of BiAV, Type I (L–R) was associated with lower 30 day and 1 year mortality (0.0% vs. 25.0%, p b 0.001 & 3.0% vs. 36.7%, p b 0.01, Fig. 3). Type I (L–R) valve had significantly lower early combined safety endpoint events (20.5% vs. 43.8%, p = 0.05), higher incidence of device success 90.9% vs 78.1%, p = 0.21), valve malposition requiring intervention (9.1% vs. 15.6%, p = 0.61) or AR ≥ 2+ (34.1% vs 41.4%, p = 0.70) were observed. 4. Discussion Patients with BiAV represent a substantial portion of elderly high risk patients with symptomatic severe aortic stenosis and may benefit from TAVI. The current report demonstrates in a large cohort of consecutive patients that transcatheter techniques are feasible with reasonable short term outcomes in patients with BiAV who are deemed not to be surgical candidates. Specifically, we noted 30 day and 1 year mortality rates of 8.3% and 16.9% respectively and 85.2% device success rate. These are similar findings to those reported for patients with tricuspid aortic valve (TrAV). Moreover, we have noted that type I BiAV valve anatomy, with left and right cusp fusion, may be associated with improved clinical outcomes. These findings should help heart teams understand the clinical performance of TAVI in patients with BiAV when considering it as part of a treatment plan.

4.1. Safety of TAVI in BiAV Most recently Khatri et al. conducted a meta-analysis of 16,000 patients undergoing TAVI looking at mortality and morbidity [12]. Notably, the 30 day and 1 year mortality rates reported in this meta-analysis of TrAV patients were 9.1% and 21% respectively — similar to the 8.3% and 16.9% mortality rates seen in our cohort with BiAV. Furthermore, our data are in line with the recently published cohort by Mylotte et al. where 139 patients with BiAV had 5.0% and 17.5% mortality rates at 30 day and 1 year respectively [13], as well as two recent reviews of published cases which noted 30 day and 1 year mortality rates of 8.6% and 15% [8,14]. Thus, the current data suggests that mortality with TAVI in highly selected patients with BiAV is at least comparable to published outcomes in patients with TrAV and is consistent. With regard to procedure related morbidities, we chose to select the VARC-2 composite early safety endpoint as our primary outcome [10]. We observed reasonable early safety with events in only 26.9% of patients experiencing an event meeting the VARC-2 definition. Comparatively, Genereux and colleagues reported a composite safety event rate of 32.7% in their meta-analysis of 16 studies of TAVI in TrAV [15]. Furthermore, Mylotte et al. reported 29.9% safety endpoint events. Thus, our findings with regard to short-term mortality and morbidity, in this highly selected cohort of patients with BiAV appear to perform comparably to patients with TrAV.

286

A. Yousef et al. / International Journal of Cardiology 189 (2015) 282–288

Table 2 Baseline echocardiography and prosthesis features.

Table 3 Primary and secondary outcomes.

Total cohort

Edwards Sapien

Core valves

Total n (%a)

Total n (%a)

Total n (%a)

BiAV diagnostic modality (n = 108) Echocardiogram 90 (83.3%) MSCT 18 (18.7%)

51 (83.6%) 10 (16.4%)

39 (83.0%) 8 (17.0%)

BiAV anatomy (n = 108) Type 0 Type 1 (L–R) Type 1 (R–N) Type 1 (L–N) Type 2 (R–L) Type 2 (R–N) Undetermined/unavailable

40 (65.0%) 5 (12.5%) 25 (62.5%) 6 (15.0%) 2 (5.0%) 1 (2.5%) 1 (2.5%) 21 (35.0%)

38 (80.9%) 8 (21.1%) 19 (50.0%) 4 (10.5%) 1 (2.6%) 5 (13.2%) 1 (2.6%) 9 (19.1%)

27 (25.2%) 50 (15.6) 41 (40.2%) 44.0 (35–52)

14 (23.3%) 50.7 (14.7) 23 (40.4%) 44.0 (33–50)

13 (27.7%) 49.3 (13.8) 18 (40.0%) 45 (40–54)

48.4 (17.0)

47.0 (16.4)

50.2 (15.1)

Pre-TAVI EF b 40% — n (%) Mean EF (%) (SD) Pulmonary hypertension Median PAPS mm Hg (IQR — mm Hg) Mean aortic valve pressure gradient (SD) Median AVA — cm2 (IQR — cm2) AR at baseline AR ≥ 2

78 (72.2.%) 13 (16.7%) 44 (56.4%) 10 (12.8%) 3 (3.8%) 6 (7.7%) 2 (2.6%) 30 (27.8%)

0.7 (0.5–0.8)

0.7 (0.5–0.8)

0.7 (0.6–0.8)

81 (75.7%) 27 (25.2%)

48 (80.0%) 16 (26.7%)

33 (70.2%) 11 (23.4%)

84 (95.6%) 22.58 (9.8)

47 (100%) 21.9 (9.8)

37 (90.2%) 23.6 (9.6)

28 (30.0%) 24.9 (12.2)

10 (16.7%) 23.0 (10.8)

18 (58.1%) 26.7 (13.8)

Valve size Total number Size 23 — valve Size 26 — valve Size 29 — valve Size 31 — valve

16 (14.8%) 42 (38.9%) 45 (41.7%) 5 (4.6%)

16 (26.2%) 30 (49.2%) 15 (24.6%) 0 (0.0%)

0 (0.0%) 12 (25.5%) 30 (63.8%) 5 (10.6%)

Access used Femoral Subclavian Transaortic Transapical

90 (83.3%) 5 (5.6%) 5 (5.6%) 8 (8.7%)

49 (80.3%) 0 (0.0%) 4 (6.7%) 8 (13.3%)

41 (87.2%) 5 (10.6%) 1 (2.1%) 0 (0.0%)

TAVI parameters Balloon predilation Predilation balloon size — mm (SD) Balloon postdilation Postdilation balloon size — mm (SD)

BiAV: bicuspid aortic valve; MSCT: multi-slice computed tomography; EF: ejection fraction; PAPS: pulmonary artery pressure; AVA: aortic valve area; AR: aortic regurgitation; MS: mitral stenosis; MR: mitral regurgitation; TR: tricuspid regurgitation. a %: Percent of the available data.

Number of events

%

Primary endpoint (30 days) Early combined safety endpoint (30 days) 30-day total mortality CVS mortality Stroke Life threatening bleeding Major vascular complication Valve-related dysfunction requiring repeat procedure AKI stages 2–3 Coronary artery obstruction requiring intervention

29 9 7 3 7 7 10 7 5

26.9% 8.3% 7.6% 2.8% 6.5% 6.5% 9.3% 6.5% 4.6%

Secondary endpoint (30 days) Re-hospitalization related to valve dysfunction NYHA III–IV Mean aortic valve pressure gradient ≥ 20 mm Hg AR ≥ 2+ AR ≥ 3+ New pacemaker Device success

2 1 2 32 10 21 92

2.3% 1.7% 2.3% 30.8% 9.6% 19.4% 85.2%

Long-term outcomes at 1 year (n = 89) Total mortality CVS mortality NYHA stages III–IV Re-hospitalization related to valve dysfunction AR ≥ 2+ Mean aortic valve pressure gradient ≥ 20 mm Hg AVA ≤ 1.1 cm2

15 9 2 6 13 4 3

16.9% 10.1% 4.3% 13.3% 27.7% 6.7% 8.1%

CVS: cardiovascular; AKI: acute kidney injury; NYHA: New York Heart Association; AR: aortic regurgitation; AVA: aortic valve area.

PVL in our study could be related to the inherent aortopathy associated with BiAV, but also can be attributed to lower rates of MSCT performed (62.1%) pre-TAVI for sizing of the valve. While the presence of PVL (≥2+) was associated with an increased incidence of the composite primary safety outcome, it was reassuringly not associated with increased 30 day and 1 year mortality — though the cohort is admittedly underpowered to assess this endpoint. While significant debate exists on the relationship between PVL and mortality — some studies, including the PARTNER trial, have noted an adverse impact on long term outcomes in patients with TrAV [18]. One important difference may be the higher rates of AR prior to TAVI in patients with BiAV conditioning the ventricle in the event of post-procedure AR. Nonetheless, the current data does not suggest an impact of PVL on outcomes following TAVI but future studies should still focus on identifying mechanisms to minimize PVL in patients with BiAV. 4.3. Outcomes of TAVI in BiAV

4.2. Feasibility of TAVI in BiAV One of the major hesitations in performing TAVI in patients with BiAV is its feasibility. In this cohort, device success was achieved in 85.2%. This is comparable to that achieved in patients with TrAV (88%) and is consistent with other reports of BiAV patients (84%) in a recent systematic review [14]. The major cause for this was attributed to significant paravalvular leak (9.6%). By far, PVL is the major concern for TAVI in patients with BiAV, owing to incomplete prosthesis apposition resultant from increased calcium or annular eccentricity. Most recently, a comprehensive review of the impact of PVL after TAVI noted a wide range of incidence owing to differences in timing and technique of assessment following implantation, and the data of our study, is not an exception [16,17]. Nonetheless, in patients with TrAV it is estimated that moderate to severe PVL (AR ≥ 3+) may occur in as many as 24% of patients [16] — with the PARTNER trial reporting an incidence of 12% when assessed by a core laboratory [4]. Comparatively we noted a 9.6% incidence of ≥3+ AR with incidence rising as high as 30.8% when including all patients with ≥2+ AR. Such higher rates of

Our study makes three important observations with regard to patients with BiAV and their management with TAVI. First, the outcome of TAVI in this highly selected cohort resulted in an acceptable improvement in symptoms as well as comparable safety outcomes to TrAV with the exception of a higher rate of PVL — a finding which did not appear to impact hard outcomes. Second, the presence of a BiAV was missed in up to a fifth of patients undergoing TAVI workup with transthoracic echocardiogram and only subsequently identified by CT. While others have demonstrated the value of CT for appropriate sizing of the prosthesis [19,20], in our cohort it had the added benefit of increasing the sensitivity for BiAV in patients undergoing evaluation. Thus, in early series using transthoracic echocardiogram alone BiAV patients may have inadvertently been included. Finally, we evaluated the bicuspid valve anatomy based on the Sievers and Schmidtke classification system [9] and noted less PVL and improved clinical outcomes in patients with fusion of the left and right cusps. While the reason type I BiAV leads to improved prosthesis performance is unclear — it is well documented that differences in aortic architecture and outcomes following aortic valve repair

A. Yousef et al. / International Journal of Cardiology 189 (2015) 282–288 Table 4 Details of procedural outcomes in TAVI for patients with bicuspid aortic valve stenosis. 12 4 8 0

Valve malpositioned requiring another intervention Second TAVI (TAVI-in-TAVI) Third TAVI (TAVI-in-TAVI-in-TAVI) SAVR for Valve malposition Balloon Valvuloplasty

11 6 0 4 2

Device failure Procedural mortality Incorrect positioning from first valve deployment Moderate AR 3+ Severe AR 4+ Post-TAVI MPG N 20 Valve embolization Conversion to SAVR Annulus rupture Annular dissection

1 10 6 2 2 4 4 1 0

1a 2 2 2 2

Beyond 30 day mortality CV deaths Myocardial infarction Pulmonary hypertension Hemothorax after pacemaker insertion Non-CV death Hepatic failure Septic shock Pneumonia

1 1 1 1 2 1

TAVI = transcatheter aortic valve implantation, SAVR = surgical aortic valve replacement, AR = aortic regurgitation, MPG= Mean pressure gradient across aortic valve, CV = cardiovascular. a The only procedural death.

differ depending on bicuspid type [21,22]. Future studies of patients with BiAV should pay careful attention to BiAV anatomic variants to both confirm the current study's findings and to potentially risk stratify patients referred for TAVI with BiAV.

50 AR ≥2+ AR < 2+

40

Percentage of cohort (%)

p = 0.05

40

p < 0.01

30 p

Transcatheter aortic valve implantation in patients with bicuspid aortic valve: A patient level multi-center analysis.

We sought to evaluate the safety and efficacy of transcatheter aortic valve implantation (TAVI) in patients with bicuspid aortic valve (BiAV)...
328KB Sizes 0 Downloads 10 Views