OUP user menu

Mid-term follow-up after transcatheter aortic valve implantation

John G. Webb
DOI: http://dx.doi.org/10.1093/eurheartj/ehs018 947-948 First published online: 6 February 2012

This editorial refers to ‘Transcatheter aortic valve implantation: 3-year outcomes of self-expanding CoreValve prosthesis’, by G.P. Ussia et al., on page 969

The prognosis for patients with symptomatic aortic stenosis is poor. Although open surgical aortic valve replacement (SAVR) is the gold standard for the management of aortic stenosis, there are no randomized trials or formal evaluations of the sort that would currently be the norm to gain approval of such a costly and invasive treatment. However, like the use of parachutes when jumping from airplanes, randomized trials were presumably considered unnecessary. There is no doubt that SAVR can relieve symptoms and prolong life.

With surgery already offering a tried and true option for patients with aortic stenosis, it is reasonable to expect more rigorous evaluation of transcatheter aortic valve implantation (TAVI).1 Ussia et al. now report 3-year outcomes following CoreValve implantation from a large and well compiled Italian registry.2 In light of this and other recent studies,35 it may be helpful to review what we now know about some of the early concerns about TAVI that are gradually being addressed. Concerns such as: (i) aortic stenosis might not be adequately relieved by transcatheter valves; (ii) paravalvular leaks might have serious consequences; (iii) durability might be poor; and (iv) the TAVI procedure might be inherently more dangerous than SAVR.

Is transcatheter valve function comparable with that of surgical valves? Yes. Standardized in vitro testing routinely documents that haemodynamic performance is comparable with that seen with surgical valves. Multiple lines of evidence suggest that, for a given annulus size as determined by echocardiography, better orifice areas may be achieved with transcatheter than with surgical valves.6 Most convincingly, the randomized PARTNER study documented significantly better valve areas and lower gradients with transcatheter as compared with surgical valves.7 The study of Ussia et al. documents similarly exceptional echocardiographic valve gradients and orifice areas for the CoreValve device. The result is an early and durable improvement in functional class and quality of life with both widely available transcatheter valves.2,5,8,9

Are paravalvular leaks a problem with transcatheter valves? Yes. The majority of patients undergoing TAVI have paravalvular leaks, if looked for.24 Fortunately, the great majority of these are mild or moderate. On average, most patients have less paravalvular aortic regurgitation than they had valvular regurgitation prior to their procedure.7 Admittedly we may not fully appreciate the long-term effect of moderate regurgitation. Although this is a minor consideration in elderly patients, this may become an issue if TAVI is applied in younger and lower risk patients. More importantly, paravalvular leaks sometimes can be severe. In the study of Ussia et al., persistent severe leaks were not observed, although it is worth noting that almost one in five patients required some sort of intervention on the implanted valve (dilation, repositioning, or a second valve).2 Such interventions may not be without consequences.10

Are transcatheter heart valves (THVs) durable? Yes, so far. All prosthetic cardiac valves are required to undergo prolonged accelerated wear testing and failure mode analysis. However, there is no substitute for clinical experience. Prior experience with surgical valves demonstrates that the 10–20 year durability anticipated with some valves is often not achieved with other valves. We have reasonable clinical documentation of 3 year durability with the SAPIEN valve2 and now with the CoreValve bioprostheses.2,11 Concerns about early leaflet deterioration, frame fracture, and instability have been largely allayed. However, late follow-up beyond 5 years remains relatively sparse and is not yet adequate to support use in patients who are good candidates for surgery.

Is the TAVI procedure inherently more dangerous than medical management? Initially yes, but this early procedural risk is soon overwhelmed by the later survival and functional benefits. In the PARTNER 1B trial of patients considered too high risk for surgery, there was an absolute 20% reduction in mortality at 1 year as compared with medical management, with survival curves continuing to diverge out beyond 2 years.12,13

Is the TAVI procedure more dangerous than SAVR? In the PARTNER 1A high risk trial, TAVI (transarterial and transapical groups combined) was ‘non-inferior’ to SAVR. However, if one looks more closely at the transfemoral procedure, TAVI actually bested SAVR, with a mortality that was less than half that with surgery (3.7 vs. 8.2% 30-day mortality as treated, P = 0.05).7

TAVI outcomes were particularly favourable in the randomized PARTNER trials.7,12 This was particularly so given a very early TAVI experience (new operators using early generation, 22- to 24-French diameter THV delivery systems) as compared with a very mature surgical procedure in the hands of very experienced surgeons. In the Italian registry,2 the Society of Thoracic Surgeons (STS) predicted risk of surgical mortality score was almost identical to that seen in the contemporaneous PARTNER trials; however, transarterial TAVI procedural mortality was higher (11.2%, vs. 5% and 3.7% in the Italian registry, PARTNER 1A and 1B, respectively). Presumably procedural outcomes continued to improve in this experience as the larger Italian CoreValve registry reported a lower mortality of 5.4%.

It must be remembered that the PARTNER 1 trials only examined TAVI in the highest risk 10% of patients (Figure 1). There are no randomized studies comparing TAVI and surgery for the other 90%, meaning that the high rates of TAVI in some countries are based on limited evidence. The ongoing PARTNER 2 and SURTAVI intermediate risk trials will only evaluate TAVI in the highest risk 25% of patients, meaning that extending the role of TAVI into the great bulk of lower risk patient will remain controversial for a long time.

Figure 1

The predicted risk of surgical mortality in patients undergoing isolated aortic valve replacement. High and extreme risk patients (predicted risk of surgical mortality at 30 days of >8%) were studied in the PARTNER 1 trials, representing the highest risk 10% of patients undergoing aortic valve surgery. Intermediate risk patients (predicted mortality >4%) are currently being studied in the PARTNER 2 and SURTAVI studies, representing the highest risk 25% of patients. Based on data from the Society of Thoracic Surgery (STS) database for isolated aortic valve replacement surgery from the year 2009 from >24 000 patients.

Clearly TAVI is a moving target. The procedure continues to improve due to increasing experience, patient selection, and technical improvements.14 So too will clinical outcomes. Valve function and durability seem more than adequate for most patients currently considered to be at high risk with surgery. Whether clinical outcomes and durability will be adequate to compete with surgery in younger and healthier patients is unknown.

Conflict of interest: Edwards Lifesciences.


  • The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.


View Abstract