Background

Transcatheter aortic valve implantation (TAVI) was introduced by Alain Cribier 10 years ago and is now an accepted treatment for high-risk patients with severe aortic stenosis. Since the approval of both the Edwards SAPIEN and the Medtronic CoreValve in Europe in 2007, TAVI has been performed in more than 50,000 patients worldwide . The current results of the technique are described in the other articles of this issue.

Today, TAVI has become the standard of care for inoperable patients with aortic stenosis and acceptable life expectancy. TAVI is also an alternative to aortic valve replacement in selected high-risk operable patients . The aim of this review is to forecast what may happen in the future, starting with patient selection and then examining technical aspects.

Patient selection

First of all, it should be stressed that today and in the future, the TAVI Heart Team approach is and will remain, essential for the management of patients with severe aortic stenosis and TAVI should be restricted to high-risk patients . This will apply at each step of the procedure: patient selection, performance of the procedure, post-procedural care and evaluation of the results.

The Heart Team is comprised of clinical cardiologists, interventionists, surgeons, anaesthetists and imaging specialists, all with expertise in the treatment of valve disease. The participation of other specialists, such as geriatricians, will be increasingly sought.

It is essential to assess both the risk of surgery and the risk of TAVI. Firstly, we need better scores to assess the risk of surgery. The current scoring systems, EuroSCORE or Society of Thoracic Surgeons Predicted Risk Of Mortality (STS PROM) , are limited in their prediction of outcomes in high-risk patients. New scoring systems should be based on a limited number of variables; aimed at the specific evaluation of valvular patients; elaborated from a broad spectrum of operative risk; externally validated in high- and low-volume centres; and updated on a regular basis. Besides evaluating cardiac and extracardiac factors, it is mandatory to include indices of functional and/or cognitive capacity and frailty. We need better definition and further evaluation of this last parameter .

Secondly, we need scoring systems that predict the outcome of TAVI, both in the immediate and the long term. Even with refined scoring systems it is likely that it will never be ‘magic numbers’. Assessment by the Heart Team, based primarily on clinical judgment, will remain critical, but will be supported by a certain degree of quantification using better scores.

The use of TAVI is limited to patients at high risk or with contraindication(s) to surgery. In this category, a number of subgroups require more-precise evaluation:

• •
  

  associated coronary artery disease: only data from retrospective studies involving a limited number of patients exist; we therefore lack the solid evidence necessary to guide our strategy. It is likely that randomized studies will be the best way to decide when percutaneous coronary intervention (PCI) should be performed and the timing of the procedure ;

  
  
• •
  

  bicuspid valves are a classic contraindication for TAVI. Here again, we have very limited data to guide us, but this feature will but this feature will become increasingly important with the consideration of lower-risk patients (i.e. younger patients). It may be that a specific valve design is needed in patients with bicuspid valves ;

  
  
• •
  

  the strategy in patients with severe left ventricular dysfunction is also a matter of debate. We should identify those patients unlikely to benefit from TAVI because of their low likelihood of a good outcome and in those we are willing to treat, we should compare a strategy of balloon aortic valvuloplasty as a bridge versus TAVI as a first intervention ;

  
  
• •
  

  transcatheter ‘valve in a valve’ is an attractive alternative in bioprosthesis failure. Preliminary results have been reported from observational studies, but more data are needed to better define: which valve prostheses are suitable for valve-in-valve implantation; sizing and the theoretical risk of thrombosis due to dual prosthetic implantation; and durability. It is possible that a new valve design will be needed .

With improved identification of the predictors of outcome after TAVI and thanks to the involvement of other specialists in patient selection, it is likely that we will be able to better define which patients should not be treated, either by surgery or by TAVI because their life expectancy and expected quality of life are too limited .

The most important issue for the future is to know when to treat lower risk patients. This has already happened in practice, as illustrated in a German registry, where as many as 16% of patients were treated as a result of patient choice while being at only intermediate risk for surgery . However, before moving towards lower risk patients we should accumulate evidence on the long-term results of the procedure. We currently have only anecdotal cases of valve dysfunction, but the available follow-up is seldom longer than 3–4 years and we need longer-term data to better define the timing and the mode of failure . Due to the similarity of the valves implanted surgically and using the transcatheter, it is likely that both are similar with regard to long-term outcomes, but this has yet to be proven. In addition, when structural dysfunction occurs, it is likely that in most cases it will be possible to put a ‘valve in a valve’ and then further delay conventional operation. The best way to assess whether to move towards lower risk patients is to perform randomized studies, such as the SURTAVI and Placement of Aortic Transcatheter Valves (PARTNER) II trials, which will include patients with a STS predicted mortality rate of 3–8%.

Finally, it is unlikely that TAVI will be used in a large number of patients with aortic regurgitation; most of these individuals have dystrophic disease with dilatation of the ascending aorta, which should be cured at the same time as aortic stenosis .

Sizing of the valve

In the future we should be able to solve the challenge of the proper sizing of the valve . Measurement of the aortic annulus was initially done by echocardiography alone. Three-dimensional echo or computed tomography (CT) provides more accurate measurements, but it remains to be established whether a strategy based on CT is more effective and safer than an echo-based strategy ; this should be proven by appropriately designed studies. It is likely that in the future, CT or magnetic resonance imaging (MRI) will be the best methods because they can be reproduced more easily and are also better suited to the assessment of the aorta and the peripheral vessels. The role of other factors such as distribution of calcium and the size of the sinus of Valsalva should also be assessed .