Decision making in AS

Patients with AS are frequently elderly and frail with attendant comorbidity, and selection of the optimal treatment option is frequently challenging as a consequence. The development of a robust ‘Heart Team’ to evaluate, discuss and treat these patients is a fundamental principle of a successful TAVI service, and this multidisciplinary approach is strongly endorsed by the European Society of Cardiology and American College of Cardiology/American Heart Association Guidelines on Valvular Heart Disease. Guiding a patient and their family in the choice between AVR and TAVI (or no intervention in some patients where high-risk intervention would be futile) and comparing the risks and benefits of either approach requires careful explanation of what is known about the procedures and their immediate and long-term outcomes. Despite the rapid emergence of TAVI in the past five years, there is a paucity of outcome data beyond 3 years and nothing approaching the extensive data concerning the long-term performance of surgically implanted bioprosthetic valves whose expected durability, mode of failure, complication rates and risks of redo surgery are well described. Although only a small proportion of patients within the original TAVI recipient cohorts are likely to require repeat intervention (due to advanced age at the time of their index procedure), there is little experience to date on how, when and why percutaneously implanted valves degenerate and no defined options for repeat valve intervention.

In their Heart paper, Bouleti et al4 present the outcome of 141 patients who underwent TAVI at a single Parisian centre between October 2006 and December 2009 and then followed up for a median duration of 3.6 years (maximum 6 years). Six-year survival rate was 31%±5% with a reasonably constant attrition rate attributed to the influence of comorbidities—6-year cardiovascular survival rate was 66%±5% and >50% of deaths were related to non-cardiac causes adjudicated by Valve Academic Research Consortium-2 criteria. Multivariate analysis identified three factors predictive of late death after TAVI—symptomatic lower limb vascular disease, a high Charlson comorbidity index (≥5) and postprocedural paravalvular aortic regurgitation (≥grade 2/4). The patients treated were widely dispersed and data concerning the long-term performance of the percutaneously implanted valves or the mode of cardiac death (beyond cardiac failure) were unavailable due to a lack of standardised echocardiographic follow-up. However, functional status of the patients remaining alive at most recent follow-up was reassuring with 33/45 patients reporting New York Heart Association Class I/II symptoms.

How do these data compare with other series?

Analysis of medium-term survival data in the landmark Placement of Aortic Transcatheter Valve B study demonstrated a 3-year mortality of ∼50%5 with 41% of deaths related to a cardiovascular cause and maintained function of the TAVI prosthesis in all survivors. Similarly, Toggweiler et al6 recently reported the 5-year outcome of 88 patients (mean age 83 years) treated in the first TAVI cohort in Vancouver, Canada, with overall survival of 53% and 35% at 3 and 5 years, respectively, and a median survival (excluding procedural mortality) of 3.4 years after TAVI. Data concerning cause of death were not reported, but mortality was predicted by the presence of moderate or worse postprocedural paravalvular regurgitation and major comorbidities (notably pulmonary disease). Recent unpublished data from the UK TAVI registry demonstrate a 5-year survival of 48%, with long-term mortality related to comorbidities including diabetes, atrial fibrillation, pulmonary disease and renal impairment. Overall, these data suggest that 30–50% of the current cohort of high-risk patients undergoing TAVI can expect to be alive 5 years following intervention.

How do these data compare with the results of surgical AVR?

Surgical AVR is a procedure that has been refined and perfected over more than half a century with excellent results and procedural safety that are reproducible in all cardiac surgical centres. Latest UK data report an in-hospital mortality following isolated AVR of <2%, and this is the gold standard against which TAVI must ultimately be compared after adjustment for patient factors and predicted risk. Analysis of almost 150 000 patients undergoing AVR in North America7 reported a median survival of 6 years in octogenarians—a figure identical to the median survival of the age-matched general US population. These impressive results confirm that patient selection is key to ensuring optimal long-term outcomes. Indeed, the ‘art’ of risk stratification in high-risk patients with AS has been the focus of several recent publications, accompanied by long overdue research on the accurate quantification of frailty.8 Selection of patients who are at excessive risk for surgical AVR but likely to benefit significantly from TAVI and survive to 5 years and beyond is a fundamental role of the ‘Heart Team’ and often requires input from other medical disciplines including geriatricians, respiratory physicians and anaesthetists.

What else do we need to know?

Many questions relating to the long-term consequences and cost effectiveness of TAVI remain unanswered. Although TAVI-related infective endocarditis is rare and confined to case reports, it is frequently fatal—mechanisms have yet to be clarified, but nosocomial infection at the time of implantation in the catheter laboratory environment may well be relevant. Given the intrinsic similarity of current TAVI devices to bioprosthetic valves implanted surgically, progressive leaflet degeneration and valve failure seems inevitable. However, the likely timing of onset of valve dysfunction and the mode and acuity of valve failure have yet to be described. Treatment options in this setting remain limited, and the design of percutaneous valves that will permit future ‘TAVI in TAVI’ procedures will be a challenging consideration for engineers and device manufacturers over the next decade. Moreover, proven durability of percutaneously implanted valves will provide yet more challenging clinical scenarios—for example, progressive or acute manifestations of coronary artery disease may require percutaneous intervention and ensuring unobstructed access to the coronary ostia may favour certain TAVI concepts and designs.

Implications for future patient selection

The emergence of long-term outcome data following TAVI as highlighted by the present Parisian data begins to unravel the key remaining unknowns in the decision-making matrix for elderly patients with severe symptomatic AS. One fundamental question is whether the predicted durability of the implanted prosthesis will exceed the life expectancy of the patient—while the study of Bouleti et al does not provide specific data on percutaneously implanted valve performance at 5 years, the high rate of non-cardiac-related deaths reinforces the key role of careful assessment and screening for important comorbidities such as significant peripheral vascular or respiratory disease, and optimal procedural technique avoiding any significant remaining paravalvular regurgitation.