Abstract: Background. Transcatheter aortic valve implantation (TAVI) is the treatment of choice for high-risk patients presenting with severe symptomatic aortic stenosis. The aim of this study was to investigate the impact of second-generation (2G) devices in comparison to first-generation (1G) devices with regard to procedural and short-term clinical outcomes. Methods. Between November 2007 and May 2015, a total of 449 patients treated with 1G TAVI devices (Edwards Sapien XT, Medtronic CoreValve) were propensity matched (1:1) to 179 patients treated with 2G TAVI devices (Edwards Sapien 3, Medtronic Evolut R, Boston Scientific Lotus, Direct Flow Medical). The primary endpoint was 30-day safety according to the Valve Academic Research Consortium 2 (VARC-2) definition. Results. Patients treated with 1G devices suffered more adverse events at 30-day follow-up (freedom of adverse events, 75.3% vs 88.8%; hazard ratio, 2.4; 95% confidence interval (CI), 1.4-4.0; P=.01) and a significantly greater number of minor vascular complications (31.8% vs 10.4%; P<.001) and major vascular complications (3.2% vs 0.6%; P<.001) compared with patients treated with 2G devices. The presence of residual aortic regurgitation ≥2 was also greater in the 1G group (17.5% vs 5.8%; odds ratio, 0.30; 95% CI, 0.13-0.69; P<.001). There were no differences between groups with regard to 30-day all-cause mortality (5.2% vs 3.2%; odds ratio, 0.61; 95% CI, 0.20-1.92; P=.40). Conclusion. TAVI with contemporary 2G devices was associated with a significant safety benefit at 30 days and reduction of residual moderate or severe paravalvular leak. Longer-term follow-up in more patients is required to determine if these short-term benefits translate into improvements in long-term clinical outcomes.
J INVASIVE CARDIOL 2016;28(5):210-216
Key words: transcatheter aortic valve implantation, TAVI, TAVR, first generation, second generation
Transcatheter aortic valve implantation (TAVI) has now become the treatment of choice for patients presenting with symptomatic severe aortic stenosis (AS) who are deemed inoperable or of high surgical risk.1,2 In this select patient group, TAVI has been demonstrated to be non-inferior to surgical aortic valve replacement (SAVR) in a number of studies3-6 and associated with a mortality advantage in one recent study.7
In spite of these promising initial results and greater operator experience, TAVI has a number of limitations in comparison with SAVR. The most important of these is the presence of residual paravalvular leak (PVL), which has been found to be an independent predictor for long-term adverse outcomes and mortality.8,9 Other important considerations include vascular complications, conduction disturbances, and the occurrence of stroke.10,11 Addressing and overcoming these limitations is particularly important if TAVI is to be expanded to the treatment of younger, lower-risk individuals as an alternative to the current “gold standard” of SAVR.
In recent years, first-generation (1G) TAVI devices have been extensively modified (and newer, novel devices have been developed) in an attempt to address some of these limitations. Prostheses have been specifically designed to reduce the occurrence of residual PVL and many devices now also have the option to be repositioned or completely retrieved to aid in implantation and optimize procedural outcomes.12 In addition to the prostheses themselves, delivery systems have been significantly improved and contemporary sheaths are now lower profile and more flexible, with the dual aims of facilitating device delivery and reducing vascular and bleeding complications.13
The aim of this study was therefore to investigate the impact of the use of newer second-generation (2G) devices in comparison to 1G devices on procedural and short-term clinical outcomes following TAVI for the treatment of patients presenting with severe symptomatic AS.
Study population. All patients treated with TAVI between November 2007 and May 2015 at the San Raffaele Scientific Institute in Milan, Italy were included in this retrospective analysis. Patients treated with TAVI devices for the management of aortic, mitral, tricuspid, and pulmonary bioprosthesis dysfunction, pure aortic regurgitation, and via any non-transfemoral (TF) vascular access route were excluded. A dedicated multidisciplinary heart team consisting of interventional cardiologists, imaging cardiologists, cardiac surgeons, cardiac anesthetists, and general physicians discussed the management of all patients. All patients were deemed to be inoperable or of high surgical risk using a combination of accepted surgical risk scores (EuroScore; Society of Thoracic Surgeon [STS] risk scores) and clinical judgment to encompass patient factors (eg, frailty) that are not accounted for by the above objective assessments in keeping with current guidelines.1 Patients were allocated to the 1G TAVI group if they were treated with a Sapien XT valve (Edwards Lifesciences) or CoreValve (Medtronic) and were predominantly treated with these devices between 2008-2014. Patients were allocated to the 2G TAVI group if they were treated with a Direct Flow Medical (DFM) valve (Direct Flow Medical), Lotus valve (Boston Scientific), Evolut R valve (Medtronic), or Sapien 3 valve (Edwards Lifesciences) and were treated with these devices between 2014-2015. Patients treated with other devices during this period were excluded. All patients provided full informed consent for the procedure, subsequent follow-up, data collection, and subsequent analysis.
TAVI procedure. Prior to TAVI, all patients underwent multislice computed tomography (MSCT) or coronary angiography to exclude coronary artery disease and invasive or MSCT assessment of the peripheral vasculature. Prosthesis type and sizing were at the operator’s discretion based upon the MSCT and/or echocardiographic findings. All procedures were carried out under general anesthesia or conscious sedation provided by a cardiac anesthetist. Periprocedural and postprocedural medical therapy did not change over time, with the administration of periprocedure unfractionated heparin aiming for an activated clotting time (ACT) of 250-300 seconds. The TF route was the default access site, with percutaneous puncture sites closed with suture-based closure devices (Prostar and Proglide; Abbott Laboratories). Other access sites including transapical, transaortic, and transaxillary were considered if the TF route was contraindicated. Following the procedure, patients were administered dual-antiplatelet therapy (aspirin 100 mg daily and clopidogrel 75 mg daily) for a period of 3 months.
First-Generation TAVI Devices
Edwards XT valve. The Edwards Sapien XT valve is a balloon-expandable valve where the cobalt-chromium stent and the bovine valve leaflets are crimped onto a balloon that is expanded to deploy the valve with the aid of rapid ventricular pacing to induce ventricular standstill and facilitate optimal valve positioning and implantation. The valve requires the insertion of a 16 Fr E-sheath for 23 mm and 26 mm valves and an 18 Fr sheath for 29 mm valves when implanted via the TF route.
Medtronic CoreValve. The Medtronic CoreValve is a self-expanding valve that consists of a trileaflet porcine pericardial valve mounted on a self-expanding nitinol frame and requires the insertion of an 18 Fr TF sheath for delivery.
Second-Generation TAVI Devices
Edwards Sapien 3 valve. The Edwards Sapien 3 valve, like the Sapien XT valve, is a balloon-expandable valve but has been modified with the introduction of a lower “skirting” to reduce the occurrence of PVL following implantation and is implanted via a 14 Fr (23 mm, 26 mm valves) or 16 Fr (29 mm valve) expandable sheath.
Medtronic Evolut R valve. The Medtronic Evolut R is a modification of the CoreValve that is repositionable, resheathable, and retrievable in up to 80% of implantations, and is delivered via a 14 Fr in-line sheath.
Direct Flow Medical valve. The DFM valve consists of three bovine pericardial tissue leaflets sutured onto two inflatable Dacron rings (aortic and ventricular). It is entirely non-metallic, repositionable, and fully retrievable, and requires the insertion of an 18 Fr TF sheath for delivery.
Boston Scientific Lotus valve. The Boston Scientific Lotus valve is positioned using a mechanical deployment device and consists of bovine pericardial tissue mounted onto a nitinol stent. The device requires the insertion of a 20-22 Fr TF sheath and is fully repositionable and retrievable. It can be implanted via the TF or the direct aortic route.
Procedural outcomes and patient follow-up. Procedural and 30-day outcomes were prospectively collected in a dedicated TAVI database, and follow-up was conducted either by clinic visits or telephone consultations. All definitions of the clinical endpoints used were in accordance with the Valve Academic Research Consortium-2 (VARC-2) definitions14 that were independently adjudicated by at least two interventional cardiologists. Other information collected included the severity of aortic regurgitation (AR) following TAVI (by aortography or echocardiography) and the requirement for permanent pacemaker (PPM) implantation.
Study endpoints. The primary endpoint was freedom from the VARC-2 30-day safety combined endpoint following TAVI (composite of all-cause mortality, all stroke [disabling and non-disabling], life-threatening bleeding, stage 2 or 3 acute kidney injury [AKI], coronary artery obstruction requiring intervention, major vascular complication, and valve-related dysfunction requiring repeat procedure).
Secondary endpoints included the individual components of the primary outcome measure in addition to the requirement for PPM implantation and the occurrence of moderate/severe (grade ≥2) AR following TAVI.
Statistical analysis. Continuous variables are presented as mean ± standard deviation. Normality of each continuous variable was tested with the Kolmogorov-Smirnov test. Differences in continuous variables between groups were compared using a paired Student t-test or Mann-Whitney U-test for parametric and non-parametric variables, respectively. Categorical variables are presented as numerical values and percentages and differences between groups were investigated using the McNemar test. Logistic regression analysis was utilized to estimate in-hospital outcomes, and reported as odds ratios (ORs) and 95% confidence intervals (CIs). The cumulative incidences of clinical events at 30 days were estimated using Kaplan-Meier analysis, and reported as hazard ratios (HRs) and 95% CIs using Cox regression analysis. All reported P-values are 2-sided and P-values <.05 were regarded as statistically significant.
To minimize the effect of treatment-selection bias and potential confounders in this study, propensity-score matching (1:1) was performed. A propensity score was calculated with the aid of multivariable logistic regression analysis, which included the following variables: age, sex, weight, height, diabetes mellitus, hypercholesterolemia, hypertension, smoking, baseline renal failure, baseline dialysis, pulmonary hypertension, previous cerebrovascular disease, previous percutaneous coronary intervention, previous myocardial infarction, previous coronary artery bypass grafting, peripheral artery disease, chronic obstructive pulmonary disease, coronary artery disease, cardiac decompensation at time of TAVI, baseline ejection fraction, history of syncope, access site, type of anesthesia, logistic EuroScore, and STS mortality score. The C-statistic was 0.71, and the Hosmer-Lemeshow P-value was .36, indicating good discrimination and calibration of the propensity-matching model, respectively. In-hospital and 30-day outcomes in the matched population were calculated with logistic regression and Cox proportional hazards regression stratified to matched pairs, respectively. Analyses were performed with SPSS version 21.0 (SPSS, Inc) and GraphPad Prism version 5.0 (GraphPad).
Patient population. During the study period, a total of 829 patients underwent TAVI at our institution, with 449 patients treated via the TF route with devices in the predefined 1G group (224 patients with Sapien XT, 225 patients with CoreValve) and 179 patients treated with devices in the predefined 2G group (79 patients with DFM, 18 patients with Evolut R, 47 patients with Sapien 3, and 35 patients with Lotus valve). Propensity-matching analysis identified 154 matched pairs (1G group: 76 patients CoreValve, 76 patients Sapien XT; and 2G group: 60 patients DFM, 45 patients Sapien 3, 31 patients Lotus, 18 patients Evolut R). Baseline characteristics of both unmatched and propensity-matched groups are illustrated in Table 1. Propensity-matched groups demonstrated no significant differences specifically with regard to age (82.7 ± 6.7 years vs 81.7 ± 7.9 years; P=.23), male sex (40.3% vs 39.0%; P=.91), diabetes mellitus (33.8% vs 26.6%; P=.29), baseline renal dysfunction (41.6% vs 39.0%; P=.73), and ejection fraction ≤35% (13.0% vs 16.2%; P=.51) with comparable surgical risk scores (logistic EuroScore, 21.3 ± 16.4 vs 20.9 ± 17.9 [P=.78] and STS score, 6.5 ± 6.5 vs 6.6 ± 5.6 [P=.78]) between the 1G and 2G groups, respectively.
Procedural characteristics and outcomes. A minority of procedures were performed under general anesthesia (9.1% vs 8.4%; P>.99). Aortic valve predilation rates were similar between the two groups (64.9% vs 55.8%; P=.14); however, postdilation was carried out in significantly more patients following implantation of 1G devices in comparison with 2G devices (27.9% vs 7.1%; P<.001).
There was 1 incidence of coronary obstruction in the 1G group that was successfully treated with emergency percutaneous coronary intervention with no further sequelae. All patients successfully received a TAVI device in each group, with 13 patients (8.4%) requiring the implantation of a second device in the 1G group and 2 patients (1.3%) requiring a second device in the 2G group (P<.001). There was a significantly greater number of minor (31.8% vs 10.4%; P<.001) and major vascular complications (3.2% vs 0.6%; P<.001) in the 1G group when compared with the 2G group. Patients treated with 1G devices also had significantly longer fluoroscopic times (81.7 ± 50.2 minutes vs 33.1 ± 22.1 minutes; P<.001) and greater volumes of contrast administered (136.1 ± 95.1 mL vs 81.9 ± 57.7 mL; P<.001). There were no intraprocedural deaths in either group. An overview of the procedural characteristics and outcomes are summarized in Table 2.
30-day outcomes. Procedural and 30-day outcomes are summarized in Table 3. With regard to the primary endpoint of 30-day safety according to the VARC-2 definition following TAVI, patients treated with 1G devices suffered more adverse events compared with 2G devices (freedom of adverse events, 75.3% vs 88.8%; HR, 2.4; 95% CI, 1.4-4.0; P=.01) (Figure 1). This difference between groups was driven by a significant difference in the number of major vascular complications (21.4% vs 7.4%; OR, 0.28; 95% CI, 0.13-0.57; P<.001) in the 1G group vs the 2G group. Additionally, there were numerically more life-threatening bleeds in the 1G group (1.9% vs 0.6%; OR, 0.28; 95% CI, 0.03-2.7; P=.27) and a non-significant trend toward more patients suffering stage 2 or 3 AKI in the 1G group (14.9% vs 8.4%; OR, 0.53; 95% CI, 0.26-1.08; P=.08).
There were no differences between groups with regard to the requirement for PPM implantation (6.5% vs 7.8%; P=.46) and myocardial infarction (0.6% vs 0%). We did, however, observe a non-significant but numerically higher incidence of stroke in the 2G group (0.6% vs 2.6%; P=.10).
In spite of the greater proportion of patients who underwent postdilation in the 1G group, there were significantly more patients with residual ≥2 AR in this group when compared with the 2G group (17.5% vs 5.8%; OR, 0.30; 95% CI, 0.13-0.69; P<.001). There were no differences between groups with regard to 30-day all-cause mortality (5.2% vs 3.2%; OR, 0.61; 95% CI, 0.20-1.92; P=.40).
The aim of this study was to investigate the impact of 2G TAVI devices on procedural and short-term outcomes compared with the use of 1G TAVI devices. The principal findings of this study are: (1) 2G devices were associated with a significant benefit with regard to 30-day safety when compared with 1G devices; (2) this benefit was predominantly driven by a reduction in vascular complications and bleeding; (3) 2G devices were associated with a significantly lower incidence of residual AR ≥2 following TAVI; and (4) there were no significant differences at 30 days with regard to all-cause mortality between groups.
In this single-center, retrospective, propensity-matched study, device success was 100% in both 1G and 2G groups. The use of 2G devices was associated with a significantly greater freedom from adverse events (89.3% vs 78.5%; P=.01) when compared with 1G devices with regard to the primary 30-day safety endpoint as defined by the VARC-2 criteria. This was predominantly driven by a reduction in major vascular complications and major bleeding events. The development of smaller and more flexible delivery sheaths has enabled more patients to have safe TAVI via the more preferable TF route, with a lower risk of vascular complications, major bleeding, and resultant blood transfusion. This is particularly important because of the observation that vascular complications resulting in major or life-threatening bleeding requiring blood transfusion are associated with both short-term15 and medium-term16 adverse events. The impact of these improvements in delivery system design was reflected in our study, with patients treated with 1G devices suffering a significantly greater number of major vascular complications in conjunction with major bleeds. This translated into a significantly greater number of patients in this group receiving blood transfusions.
One of the most important limitations of 1GTAVI devices was the high rate of residual AR in comparison with “gold-standard” treatment of SAVR,6 with the reported incidence of ≥2 AR between 12%-40%.17 This is of particular relevance because ≥2 AR has been identified as an independent predictor for acute and long-term mortality.11,18,19 In the randomized CHOICE (comparison of balloon-expandable vs self-expandable valves in patients undergoing transcatheter aortic valve replacement) trial20 that compared the 1G Medtronic CoreValve with the Edwards Sapien XT valve, the frequency of residual moderate or severe AR was 42.5% vs 12.4%, respectively, immediately after valve implantation, which fell to 18.3% vs 4.1% after postdilation. Newer devices have therefore been designed with the aim of reducing AR with the introduction of “lower skirtings” (eg, the Lotus and Sapien 3 valves) or with the ability of the valve to conform more readily to the underlying annular anatomy (eg, the DFM valve) and reported rates of ≥2 AR for these 2G devices have been <5%.21-23 In our study, final rates of ≥2 AR were significantly greater in the 1G group when compared with the 2G group (19.5% vs 6.0%; P<.001), broadly supporting the published reports. The postdilation rates were also significantly greater in the 1G group (28.9% vs 6.7%; P<.001), again reflecting the greater incidence of residual AR following initial deployment of 1G TAVI devices.
There have been some concerns with regard to the higher incidence of conduction abnormalities following implantation of certain 2G TAVI devices;21,22 however, in our propensity-matched study, we did not find a significant difference between groups. Similarly we did not detect a significant difference in myocardial infarction at 30 days between groups.
Of note, we did observe a numerically (but not significant) higher incidence of stroke in the 2G group. Initial experiences with newer-generation valves have reported a lower incidence of stroke.21-23 Possible explanations for our findings include the potentially higher risk of stroke with repositionable valves due to greater instrumentation of the aorta, or the existence of other coexisting comorbidities of patients not accounted for when matching baseline characteristics. Larger studies with new-generation valves are required to further investigate this discrepancy.
Fluoroscopy times and contrast volumes administered were significantly lower in the propensity-matched 2G group, possibly suggesting the greater ease of implantation of newer devices, although the impact of greater operator and institutional expertise cannot be quantified. This may have accounted for the trend toward the lower incidence of stage 2 or 3 AKI in the 2G group, again contributing to the greater safety associated with newer devices.
Finally, in spite of a significant advantage associated with 2G devices when compared to 1G devices with regard to the primary safety endpoint at 30 days, this did not translate into a mortality benefit. Possible explanations for this may include the relatively small number of patients in each group and the lack of long-term follow-up currently available with 2G devices.
Study limitations. The principal limitations of this study are the relatively small sample size and the retrospective, single-center, non-randomized design. While we used propensity matching in a bid to overcome differences in baseline characteristics, it cannot eliminate all bias; for example, the impact of operator learning curve with 1G devices compared to 2G devices, which may well have had an impact upon procedural outcomes. We were unable to demonstrate a potential mortality difference between groups, and this may also be related to the relatively small number of patients in each group and the absence of long-term follow-up that is required to show the negative impact of bleeding, transfusion, and PVL. Finally, the sample sizes were relatively small and so we were unable to perform meaningful analyses of each of the individual valves that were included in this study.
In this retrospective, single-center, propensity-score matched analysis, TAVI with contemporary 2G devices was associated with a significant safety benefit at 30 days and a significant reduction in the incidence of residual moderate or severe PVL. Longer-term follow-up in greater numbers of patients is required to determine valve durability with 2G devices and to ascertain if these short-term benefits translate into improvements in long-term clinical outcomes.
1. Joint Task Force on the Management of Valvular Heart Disease of the European Society of C, European Association for Cardio-Thoracic S, Vahanian A, et al. Guidelines on the management of valvular heart disease (version 2012). Eur Heart J. 2012;33:2451-2496.
2. Holmes DR Jr, Mack MJ, Kaul S, et al. 2012 ACCF/AATS/SCAI/STS expert consensus document on transcatheter aortic valve replacement. J Am Coll Cardiol. 2012;59:1200-1254.
3. Kapadia SR, Leon MB, Makkar RR, et al. 5-year outcomes of transcatheter aortic valve replacement compared with standard treatment for patients with inoperable aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet. 2015;385:2485-2491.
4. Mack MJ, Leon MB, Smith CR, et al. 5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet. 2015;385:2477-2484.
5. Leon MB, Smith CR, Mack M, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010;363:1597-607.
6. Smith CR, Leon MB, Mack MJ, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011;364:2187-2198.
7. Adams DH, Popma JJ, Reardon MJ, et al. Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med. 2014;370:1790-1798.
8. Athappan G, Patvardhan E, Tuzcu EM, et al. Incidence, predictors, and outcomes of aortic regurgitation after transcatheter aortic valve replacement: meta-analysis and systematic review of literature. J Am Coll Cardiol. 2013;61:1585-1595.
9. Kodali S, Pibarot P, Douglas PS, et al. Paravalvular regurgitation after transcatheter aortic valve replacement with the Edwards Sapien valve in the PARTNER trial: characterizing patients and impact on outcomes. Eur Heart J. 2015;36:449-456.
10. Genereux P, Head SJ, Van Mieghem NM, et al. Clinical outcomes after transcatheter aortic valve replacement using Valve Academic Research Consortium definitions: a weighted meta-analysis of 3,519 patients from 16 studies. J Am Coll Cardiol. 2012;59:2317-2326.
11. Moat NE, Ludman P, de Belder MA, et al. Long-term outcomes after transcatheter aortic valve implantation in high-risk patients with severe aortic stenosis: the U.K. TAVI (United Kingdom Transcatheter Aortic Valve Implantation) Registry. J Am Coll Cardiol. 2011;58:2130-2138.
12. Taramasso M, Pozzoli A, Latib A, et al. New devices for TAVI: technologies and initial clinical experiences. Nature Reviews Cardiol. 2014;11:157-167.
13. Webb JG, Binder RK. Transcatheter aortic valve implantation: the evolution of prostheses, delivery systems and approaches. Arch Cardiovasc Dis. 2012;105:153-159.
14. Kappetein AP, Head SJ, Genereux P, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. J Am Coll Cardiol. 2012;60:1438-1454.
15. Tchetche D, Van der Boon RM, Dumonteil N, et al. Adverse impact of bleeding and transfusion on the outcome post-transcatheter aortic valve implantation: insights from the Pooled-RotterdAm-Milano-Toulouse In Collaboration Plus (PRAGMATIC Plus) initiative. Am Heart J. 2012;164:402-409.
16. Seiffert M, Conradi L, Terstesse AC, et al. Blood transfusion is associated with impaired outcome after transcatheter aortic valve implantation. Catheter Cardiovasc Interv. 2015;85:460-467.
17. Genereux P, Head SJ, Hahn R, et al. Paravalvular leak after transcatheter aortic valve replacement: the new Achilles’ heel? A comprehensive review of the literature. J Am Coll Cardiol. 2013;61:1125-1136.
18. Abdel-Wahab M, Zahn R, Horack M, et al. Aortic regurgitation after transcatheter aortic valve implantation: incidence and early outcome. Results from the German transcatheter aortic valve interventions registry. Heart. 2011;97:899-906.
19. Kodali SK, Williams MR, Smith CR, et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med. 2012;366:1686-1695.
20. Abdel-Wahab M, Mehilli J, Frerker C, et al. Comparison of balloon-expandable vs self-expandable valves in patients undergoing transcatheter aortic valve replacement: the CHOICE randomized clinical trial. JAMA. 2014;311:1503-1514.
21. Meredith Am IT, Walters DL, Dumonteil N, et al. Transcatheter aortic valve replacement for severe symptomatic aortic stenosis using a repositionable valve system: 30-day primary endpoint results from the REPRISE II study. J Am Coll Cardiol. 2014;64:1339-1348.
22. Webb J, Gerosa G, Lefèvre T, et al. Multicenter evaluation of a next-generation balloon-expandable transcatheter aortic valve. J Am Coll Cardiol. 2014;64:2235-2243.
23. Schofer J, Colombo A, Klugmann S, et al. Prospective multicenter evaluation of the direct flow medical transcatheter aortic valve. J Am Coll Cardiol. 2014;63:763-768.
From the 1San Raffaele Scientific Institute, Milan, Italy; 2EMO-GVM Centro Cuore Columbus, Milan, Italy; and 3Imperial College, London, United Kingdom.
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Colombo is a minor shareholder in Direct Flow Medical. Dr Latib is a consultant for Medtronic and Direct Flow Medical. The remaining authors report no conflicts of interest regarding the content herein.
Manuscript submitted December 15, 2015, provisional acceptance given January 18, 2016, final version accepted March 1, 2016.
Address for correspondence: Dr Azeem Latib, EMO-GVM Centro Cuore Columbus, Via Buonarroti 48, 20145, Milan, Italy. Email: firstname.lastname@example.org