OUP user menu

The art of risk stratification in TAVI

Nicolas M. Van Mieghem, Patrick W. Serruys
DOI: http://dx.doi.org/10.1093/eurheartj/eht137 1859-1861 First published online: 13 May 2013

This editorial refers to ‘Clinical outcomes of patients with estimated low or intermediate surgical risk undergoing transcatheter aortic valve implantation’, by P. Wenaweser et al., on page 1894

Over the last decade, the technology of transcatheter aortic valve implantation (TAVI) has revolutionized treatment of patients with symptomatic severe aortic valve stenosis (AS) and strengthened the ties between interventional cardiology and cardiac surgery. In the recently updated European Society of Cardiology (ESC)/European Association for Cardio-Thoracic Surgery (EACTS) guidelines on valvular heart disease, TAVI has emerged as a class I recommendation for inoperable patients with severe AS and a viable treatment alternative to surgical aortic valve replacement (SAVR) in those patients with a high operative risk provided a multidisciplinary heart team has confirmed the TAVI indication and there is a sufficient life expectancy.1

Operative risk assessment is thus a key element in patient selection for TAVI. Both the Society of Thoracic Surgeons (STS) Predicted Risk of Mortality (PROM) model and the logistic EuroSCORE were validated in standard surgical patients, but they lose accuracy and precision at the higher spectrum of operative risk.2 These surgical risk models consider baseline demographic and medical variables but do not assess relevant anatomical factors (chest radiation and deformity, prior coronary artery bypass grafting with use of the internal mammary artery or other grafts crossing the midline or adherent to the sternum precluding a safe re-sternotomy, or porcelain aorta) or the individual's biological age.3 Frailty is a phenotype characterized by impaired physiological reserve and decreased resistance to stressors, and is the resultant of physical status (muscle wasting, malnutrition, weakness, slow gait speed) and neurocognitive function and mental health.4 Overall, up to one-third of those aged 80 and older are considered to be frail.5 In general practice, experienced surgeons and cardiologists pragmatically and informally estimate individual frailty based on the practical eyeball test. Yet uniform, reproducible, and more objective frailty assessment tools are necessary. Given the overall pivotal role of risk stratification in patient selection for TAVI, one may advocate—in analogy with independent clinical event adjudication—independent risk adjudication for reliable and accurate patient risk assessment and subsequent inclusion in randomized trials on TAVI. Wenaweser et al.6 present the clinical outcome after TAVI defined according to the Valve Academic Research Consortium (VARC) definitions in patients with estimated low, intermediate, and high surgical risk according to the STS risk score.7 Two questions arise: (i) Is there a clinical need to perform TAVI in patients at the lower end of the operative risk spectrum? (ii) How accurate is the operative risk estimation?

Is there a need for transcatheter aortic valve implantation in lower risk patient populations?

Since its introduction in the early 1960s, SAVR has set a solid track record in patients with severe AS. Contemporary insights from the STS Adult Cardiac Surgery Database demonstrated an operative mortality of 2.5–3.2% in patients with an STS PROM of <5% as compared with 10–0.9% in patients with an STS PROM ≥ 5.8 The EURO Heart Survey unmasked a specific patient population that was denied valve replacement because of advanced age, left ventricular (LV) dysfunction, and the presence of co-morbidities.9 The role of TAVI in inoperable patients is undisputed since the randomized Placement of Aortic Transcatheter Valves (PARTNER) trial–Cohort B pointed to an impressive absolute 20% reduction in 1-year mortality.10 The randomized PARTNER Cohort A trial demonstrated comparable mortality outcome in SAVR and TAVI up to 2 years.11 Interestingly a pre-specified quality of life analysis of PARTNER Cohort A suggested favourable quality of life and economic perspectives in the (particularly) transfemoral TAVI cohort in comparison with SAVR.12 Arguably patient and physician's preference for a less invasive strategy may be a justifiable incentive to explore safety and efficacy in lower risk patient populations.

How accurate was the Society of Thoracic Surgeons-based risk assessment?

Wenaweser et al. calculated the STS score and subcategorized 389 consecutive patients treated with TAVI in the tertiary care hospital of Bern.6 A total of 41 patients were at low risk (defined by an STS score ≤3), 254 at intermediate risk (with an STS score of >3 to ≤8), and 94 at high risk (with an STS score of >8). The cut-off STS scores may be arbitrary at best; more importantly, the authors themselves nicely illustrate the principle limitation of using exclusively the STS risk model. The Bern heart team deemed 39% of the STS low-risk and half of STS medium-risk patients inoperable or at excessive operative risk. The majority of the studied patients were thus at (very) high operative risk. No additional information was provided as to why patients were considered higher risk than estimated by the STS score, but it is conceivable that many patients were judged to be at higher risk based on frailty. Still, 44% of patients apparently had an intermediate operative risk profile. From the study itself, it is not clear why this considerable number of operable patients (intermediate risk) actually went for TAVI and not SAVR. It definitely illustrates the explosive growth of TAVI practice prone to shift to lower risk patient populations. Ethical considerations may cast gloom on the horizon, as the scientific basis on which to justify TAVI in these patients is rather scarce, especially if one considers the study time window of between August 2007 and October 2011. Furthermore, the first attempt at a randomized trial comparing transapical TAVI with SAVR in all-comers—and thus including lower risk patients—was a fiasco and was prematurely halted because of an excess of adverse events in the TAVI cohort.13 That being said, the 30-day mortality (2.4%) and major stroke rate (3.9% and 3.2%, respectively) in low and intermediate risk patients match what can be expected with SAVR. Compared with the high-risk cohort, the lower risk patients had improved 30-day and 1-year survival and suffered less renal failure and fewer vascular access site complications, with significant reductions in the combined safety endpoint. Of further encouragement is the fact that with growing experience (see the Supplementary material online of Wenaweser et al.6), TAVI became a more precise and safer procedure, with significant reductions in procedure time and contrast load, and fewer bleeding complications. This echoes what was recently demonstrated in a larger multicentre registry where mounting experience resulted in significant reductions in major vascular and bleeding complications and improved 1-year survival.14 TAVI has historically been associated with more paravalvular aortic regurgitation (AR) and (device-related) conduction disturbances and the need for permanent pacemakers as compared with SAVR.15 These issues deserve special attention if the TAVI technology moves to truly lower risk patients with a longer life expectancy. The 2% incidence of paravalvular AR post-TAVI in the latest quartile of the Bern study is impressive, yet data on newly developed conduction disturbances are not provided, and the need for permanent pacemakers did not drop with experience. Longer follow-up is definitely warranted to see what the impact of these events will be. Ultimately, only randomized trials will demonstrate whether TAVI can compare with SAVR in lower risk patient populations. The PARTNER 2 and the SURgical replacement and Transcatheter Aortic Valve Implantation (SURTAVI) trials are responding to this urgency and are currently enrolling intermediate risk patients based on a combination of STS risk calculation and heart team assessment. Wenaweser et al. should be congratulated for sharing preliminary valuable and reassuring insights from their TAVI practice and clearly illustrating the crucial role of the multidisciplinary heart team in global risk assessment.

In conclusion, appropriate risk stratification is quintessential to select the optimal treatment strategy for elderly patients with symptomatic severe AS (Figure 1). In the absence of accurately validated and calibrated risk models for TAVI, the institution of a multidisciplinary heart team seems a conditio sine qua non to guarantee that the right treatment is provided to the right patient. To paraphrase Hippocrates: ‘It's more important to know what sort of person has a disease than to know what sort of disease a person has’.

Figure 1

aSociety of Thoracic Surgeons or (Logistic) EuroSCORE. bFor example, frailty or pending urgent major non-cardiac surgery. Lower panel: subcategorization based on the operative risk as estimated by the heart team and resultant treatment strategy. CABG, coronary artery bypass grafting; PCI, percutaneous coronary intervention; SAVR, surgical aortic valve replacement; TAVI, transcatheter aortic valve implantation.

Conflict of interest: none declared.


  • 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.

  • doi:10.1093/eurheartj/eht086.