All patients with severe symptomatic degenerative AS rejected for surgical replacement were assessed for TAVI by a femoral approach over a period of two years. Our center's assessment protocol has been described previously.9 Severe AS was defined as an aortic valve area calculated by planimetry or the continuity equation of <1cm2 or <0.6cm2/m2 indexed to body surface area by echocardiography, as recommended in the ESC guidelines.4 Patients were assessed by a team of clinical and interventional cardiologists, cardiac surgeons and anesthesiologists who evaluated surgical risk, with the help of standard risk scores: EuroSCORE and the Society of Thoracic Surgeons (STS) Predicted Risk of Mortality.10,11 Patients rejected for surgery (those with logistic EuroSCORE >20% and/or STS >10%) were assessed for transfemoral TAVI, all undergoing coronary catheterization and femoral angiography, transesophageal echocardiography to measure aortic valve annulus diameter, and high-resolution CT angiography of the iliofemoral axis.

Patients with bicuspid aortic valve, valve annulus <18mm or >25mm, left ventricular ejection fraction (LVEF) <20%, myocardial infarction in the previous 30 days, or stroke or transient ischemic attack in the previous 6 months, were excluded. Those with significant coronary disease were scheduled for percutaneous coronary intervention (PCI) and TAVI was scheduled for 30 days after PCI. Also excluded were patients with excessive tortuosity or calcification, and those with a minimum iliofemoral axis diameter of <7mm with the first-generation RetroFlex® catheter, and from May 2010 onwards, <6mm with the second-generation NovaFlex® catheter. Although severe septal hypertrophy has been considered an exclusion criterion, due to the risk of prosthesis embolization during implantation, in our center the implantation procedure is slightly modified in these patients.12

In the 30-month study period (April 2008–October 2010), a total of 76 patients were evaluated, of whom 44 (57.9%) were rejected for TAVI, 25 (32.9%) were accepted, 5 (6.6%) died during the evaluation period, and 2 (2.6%) are currently on the waiting list for the first transapical implantation in our center. Of the 44 patients rejected, the reason was poor iliofemoral access in 17 (38.6%), comorbidities that significantly limited the patient's life expectancy or quality of life in 8 (18.2%), non-severe or asymptomatic AS in 7 (15.9%), refusal by the patient or the patient's family in 5 (11.4%), and severe ventricular dysfunction (LVEF<20%) in 4 (9.1%); four patients (9.1%) were referred again for surgical valve replacement after re-evaluation showed the surgical risk to be acceptable. Of the 25 patients accepted for TAVI, implantation proved impossible in 2 (8%) due to vascular access complications. The remaining 23 patients (92%) constitute the study population.

The procedure was performed under general anesthesia in a sterile environment in the hemodynamic laboratory, with patients being extubated in the same room in most cases. The Edwards SAPIEN aortic prosthesis consists of a trileaflet bovine pericardial prosthesis mounted on a balloon-expandable metal stent (Fig. 1). Two sizes are currently available: 23mm (for a valve annulus of 18–21.5mm) and 26mm (for a valve annulus of 21.5–25mm). A temporary pacemaker lead is positioned via the femoral vein in the right ventricle, and balloon aortic valvuloplasty is performed under rapid ventricular pacing (∼200bpm) following the standard technique. A retrograde approach is used as described previously.13 The sheath is introduced by sequential dilation of the access site from the femoral artery to the infrarenal aorta, and through this the balloon-mounted prosthesis is inserted. The system is advanced through the native valve and the balloon is inflated, again under rapid pacing, to expand the stent and implant it in the native valve annulus, compressing the valve leaflets (Fig. 2). The procedure is monitored by angiography and three-dimensional transesophageal echocardiography. The antiplatelet therapy prescribed for most patients was aspirin 100mg/day and clopidogrel 75mg/day for three months.

Figure 1.

Edwards SAPIEN prosthetic valve. A and B: the unfurled prosthesis as it would appear following expansion over the native valve. C: the prosthesis mounted on the catheter and balloon prior to transfemoral introduction.

(0.15MB).

Figure 2.

Transcatheter aortic valve implantation. A: aortography showing the stent with the balloon-mounted prosthesis in the left ventricular outflow tract. B: balloon completely inflated, with the prosthesis expanded in its final position.

(0.12MB).

The variables analyzed in this study are those recommended by consensus documents and assessed in other TAVI registries.14,15 The data were stored in a specially designed database and analyzed using SPSS 15.0 (Chicago, Illinois, USA). Categorical variables are expressed as percentages and continuous variables as means±standard deviation. Mortality during follow-up is shown by Kaplan–Meier survival curve analysis.

TAVI was performed in a total of 23 patients, whose baseline characteristics are summarized in Table 1. Mean age was 8.15±6 years (68–90). The most common comorbidity was ischemic heart disease, in 13 patients (56.5%), 10 of whom had undergone revascularization prior to aortic valve implantation (four by coronary artery bypass grafting and six by PCI). Ten (43.5%) had chronic renal failure, of whom two (8.7%) were undergoing dialysis.

The etiology of AS was degenerative in all patients except one (rheumatic valve disease). Of those with coexisting valve disease, 3 (13%) had moderate mitral stenosis, 7 (30%) had moderate mitral regurgitation and 7 (30%) had moderate or severe tricuspid regurgitation. Severe mitral regurgitation was considered an exclusion criterion, but one such patient first underwent percutaneous mitral valvuloplasty followed by TAVI.

Table 2 shows the procedural results. The valve was successfully implanted in all cases. The hemodynamics of the newly implanted valve was excellent, a valve area of over 2cm2 being achieved in all cases. Around half of patients (n=12; 52.2%) were implanted using the first-generation system (RetroFlex®, 22 and 24 F), and the rest (n=11; 47.8%), using the second-generation system (NovaFlex®, 18 and 19 F). The introduction of the NovaFlex® system in May 2010 led to a significant reduction in the number of patients rejected for TAVI because of insufficient iliofemoral diameter, as well as fewer vascular access complications: four out of 12 (33%) patients with the RetroFlex® vs. none out of 11 with the NovaFlex® (chi-square, p=0.015).

Complications during and after the procedure are shown in Table 3. There were no deaths in the hemodynamic laboratory (intraprocedural mortality 0%). One patient died on the third day after implantation due to cardiac tamponade; the autopsy revealed perforation of the right ventricle by the pacemaker lead.16 Mortality at 30 days was thus 4.4%.

The following major complications occurred: laceration of the femoral artery requiring immediate surgical repair in one case, need for transfusion in five cases (22%), and need for permanent pacemaker implantation in two cases (9%). The latter were due to damage to the atrioventricular node and His bundle by prosthesis expansion.16 There were no cases of significant prosthesis dysfunction.

Clinical follow-up was achieved in all patients. Mean follow-up was 12.9±11 months. The first patient to undergo TAVI in our center has now been followed for 30 months and is alive, with good prosthesis function. Between the 30th day and the end of follow-up two more patients died (8.7%), both between the first and second month after the intervention and neither due to cardiac cause (one from pneumonia and the other following general deterioration and multiple organ failure). Accumulated mortality at the end of follow-up was 13%. As shown by Kaplan–Meier survival analysis, all deaths occurred in the first three months, the survival curve being flat thereafter (Fig. 3). With regard to major adverse cardiovascular events (rehospitalization for prosthesis dysfunction, heart failure, stroke or myocardial infarction), two patients were rehospitalized, one for heart failure at two months (with normal valve function) and the other at three months for stroke. Survival free of major events (death or cardiovascular rehospitalization) was 91.3% at one month and 78.3% at one year. No cases of prosthesis dysfunction were recorded during follow-up.

Figure 3.

Kaplan–Meier survival curve. Survival at 30 days is 95.8% and 87% at one year. All deaths occurred in the first three months following implantation.

(0.05MB).

Degenerative AS is the most common valve disease, and its prevalence is projected to increase in the coming years due to aging populations.17 This has implications not only for those specializing in cardiovascular disease; AS is increasingly diagnosed and treated by other medical specialties, including internal medicine, geriatrics and intensive care.

Valve replacement surgery is an effective and safe treatment that changes the natural history of AS, but in certain circumstances, especially at advanced ages, it can carry an unacceptably high surgical risk.3,4 The recent development of transcatheter aortic valve implantation has provided an alternative treatment for such patients that is being used in an increasing number of centers. In this article we present the results of this technique in the first two years since its introduction in our center.

In our series, 30-day mortality was 4.4%, an excellent result considering that the surgical mortality predicted by the EuroSCORE in this population was 22%. These results are comparable to those of other published series; in the SAPIEN Aortic Bioprosthesis European Outcome (SOURCE) Registry, with 463 patients treated by a transfemoral approach, initial implantation success was 95.2%, and 30-day mortality was 6.3%.15 In the Placement of Aortic Transcatheter Valve (PARTNER) trial, 358 patients with severe symptomatic AS refused for surgery were randomized to TAVI or standard therapy (medical and/or balloon aortic valvuloplasty). One-year mortality was 30.7% in the TAVI group and 50.7% in the standard therapy group (p<0.001), an absolute reduction of 20% and a relative reduction of 40%.13

The rate of complete atrioventricular block requiring permanent pacemaker implantation in our series (9%) was comparable to that of other series and well below that of studies on the other currently available transcatheter-implantable aortic valve prosthesis (CoreValve, Medtronic, USA).18

Long-term follow-up in our population has shown excellent results, with one-year survival of 87% and no deaths from cardiac cause or complications associated with the prosthetic valve such as prosthesis dysfunction and endocarditis. One-year survival in the SOURCE registry was 81.1%.19

Although TAVI is actually performed by interventional cardiologists, a variety of other medical personnel are involved throughout the process, from diagnosis to discharge and follow-up. Degenerative AS is diagnosed and managed not only in cardiology departments but also often by those working in primary care, internal medicine, geriatrics, intensive care or other environments, who work with cardiologists, cardiac surgeons and anesthesiologists, in consultation with patients and their families to decide whether to proceed with valve replacement surgery, based on the patient's clinical condition and surgical risk. As our study demonstrates, TAVI is a new therapeutic option for patients rejected for surgery due to high surgical risk that provides effective treatment for those who previously had no possibility of valve replacement.

The technical aspects of the procedure require not only interventional cardiologists, but also many other medical personnel specifically trained in the technique, including cardiologists specializing in transesophageal echocardiography, vascular surgeons, interventional radiologists and cardiovascular anesthesiologists. This range of specialties reflects the increasingly multidisciplinary nature of modern medicine.

One important aspect is the need for a rapid selection process, since short-term mortality in these patients is significant.1 In our series, five patients (6.6% of the total) died during evaluation or while on the waiting list, less than three months after first clinical contact with our group. These figures reveal the terminal condition of many AS patients.20

Further technical improvements in valve prostheses, and accumulated experience, can be expected to reduce mortality and risk of procedure-related complications in the future. New valve technology is at various stages of development, including changes in valve design and release mechanisms, which will lead to the technique becoming more widespread. In this context, the results of the A cohort of the PARTNER trial are eagerly awaited: patients accepted for surgical valve replacement but with relatively high risk are randomized to either surgery or TAVI, the first time transcatheter replacement has been compared directly with the standard treatment. The results of this trial may well lead to the indication for TAVI being extended to patients accepted for surgery but with relatively high surgical risk. It is expected that a risk score for TAVI will soon be developed that will predict mortality and complications, so as to improve the information available to patients and their families and to facilitate comparison with surgical replacement. Studies will also need to be performed on the cost-effectiveness of TAVI compared to standard therapy and surgery, since at the moment the technique is costly.