As the elderly population has grown, so symptomatic aortic stenosis has become an increasing problem with a poor prognosis if untreated.1 2 Trans-catheter aortic valve implantation (TAVI) has been designed to treat patients who would be at high risk during standard cardiac surgery. TAVI allows aortic valve implantation without the need for a sternotomy or cardiopulmonary bypass. The procedure may be performed from the femoral or trans-apical approach depending on the calibre of the peripheral vascular vessels and the device being used. The first human procedure was performed in 2002 and recently two devices have received a CE mark. The Edwards Sapien valve and the Corevalve are now available for commercial use in the United Kingdom. The Edwards valve is a balloon-expandable valve delivered either from the transfemoral or transapical (left 5th/6th intercostals space) approach, while the Corevalve is a self-expanding device which currently can only be delivered by the transfemoral or transaxillary approach. The procedure is usually (but not always) carried out under a general anaesthetic and during the transfemoral approach the femoral artery is usually (but not always) closed by an open surgical technique.

Surgical aortic valve replacement (AVR) is undoubtedly the “gold standard” procedure for symptomatic aortic stenosis with excellent short-term and long-term results, even in high-risk patients.2 3 Despite this, surveys have shown that many patients are excluded from this procedure, often because of co-morbidities.4 The prognosis in these patients is poor. The need for a less invasive treatment for this condition has therefore become increasingly clear.

The current CE-marked indications for TAVI include a logistic Euroscore of >20, a Society of Thoracic Surgeons (STS) score of >10 or the patient being turned down for surgery (ideally by two surgeons). Although these risk assessments have been criticised, the probable risk of death or significant morbidity in this group of patient is >20%. Outcome data for TAVI are somewhat difficult to interpret because much of the literature contains the learning curve of many units and therefore risk may be overestimated. In experienced units the 30-day mortality of TAVI may be in the region of 8-10% with a stroke rate of 3–4%.5 For those patients who survive, haemodynamic and symptomatic improvement appears excellent and there has been no late failure of any device thus far. The longest survivor of TAVI is now approaching five years. Importantly there has so far been no structural valve failure of either of the two commonly used devices.

No randomised trial comparing TAVI to high-risk surgery has yet been reported but at least one is currently under way. This trial (“Partner US”) is comparing TAVI with both medical therapy (in surgical turndowns) and standard surgery (in high-risk surgery) in a randomised fashion.6 Despite the lack of randomised data many units in the United Kingdom have enthusiastically embraced this technology.

In June 2008 the National Institute for Health and Clinical Excellence (NICE) produced interventional procedural guidance for TAVI.7 This is an assessment of safety and efficacy of the procedure and not of cost-effectiveness. The conclusions of NICE can be summarised as follows:

1. The evidence is limited to small numbers of patients and therefore centres performing the procedures should have special arrangements with regard to clinical governance, consent, audit and research.

2. Patient selection and procedure performance should be by a multidisciplinary team (MDT) including interventional cardiologists, cardiac surgeons, imaging specialists and cardiac anaesthetists.

3. Data on all patients undergoing the procedure should be entered onto the national database (UK Central Cardiac Audit Database).

This guidance stresses the need for an informed and MDT approach to this procedure. Patients should be told that there is no randomised or long-term data to support the procedure. This is usually somewhat irrelevant because most patients have been referred by a surgeon who is reluctant to perform an AVR. The patient is frequently delighted that some form of treatment might be possible.

The MDT should consist of interventional cardiologists, cardiac surgeons, cardiac anaesthetists and imaging specialists. In my opinion the surgeons should be the “gatekeepers” to TAVI. If they think a standard AVR carries a high or prohibitive risk then the patient should be considered for TAVI. Screening procedures for patient selection include transthoracic echocardiography, angiography and computed tomography (CT). Echocardiography is performed to measure aortic annulus size (current technology limits the maximum size of annulus to 25 mm) and to demonstrate severe calcific aortic stenosis. Coronary angiography is required to identify concomitant coronary artery disease. During the same screening angiography quantitative iliac and femoral angiography is necessary to establish whether a transfemoral approach is possible for the procedure. Current technology is such that 18F–24F sheaths are necessary to perform the procedure requiring peripheral vessels with a minimum calibre of 7 mm. Finally, aortic CT scanning is important for the assessment of aortic, iliac and femoral calcification and tortuosity. Peripheral vascular complications remain the one of the highest complications of the procedure, with an incidence of 10–15%. Adequate use of the angiography and CT scanning data can help to limit this complication.

Governance issues are vital for the successful introduction and assessment of this new technique. In the recent assessment, NICE recommended that all data from TAVI procedures in the United Kingdom are reported on the UK Central Cardiac Audit Database. The minimum dataset required for this is currently in development. Another area of interest for the United Kingdom will be which centres will perform the procedures and who will select them? This process has varied in countries throughout the world. Currently in the United Kingdom it has been the commercial companies who have decided. The potential for commercial pressures rather than true expertise and governance arrangements to influence this process makes the current UK arrangements suboptimal. A recent position statement from the European Society of Cardiology has suggested that centres should be limited to “a limited number of high-volume centres which have both cardiology and cardiac surgery departments”.8

The future of this technology remains unclear. Assuming the results are robust in the medium term the risk profile of the patients will gradually be reduced. This should take place within the framework of an MDT and in consultation with our surgical colleagues. One important and potential future use of this technique may well be the concept of “valve within valve” for a failing aortic bioprosthesis.9 If this becomes an established and robust technique then it would signal the end of the metallic aortic prosthesis even during AVR in “young” patients, with the concept that TAVI could be performed at least once in later life for a failing bioprosthesis.

CONCLUSIONS

This new technique has the potential to revolutionise the treatment of aortic valve stenosis. However, it is very early in the development of the technique and the robustness of the valve will dictate how quickly it is introduced into lower risk patients. The success of TAVI will depend on hospitals and individuals taking a careful MDT approach to its development. It is vital that this is done with the cooperation of the surgical community. With this approach then TAVI will undoubtedly be a breakthrough technology. Without it, disaster could be just around the corner and this is certainly not a technique for a lone, maverick interventional cardiologist.