By Jeffrey Zimmet, MD, PhD

Associate Professor of Medicine, University of California, San Francisco; Director, Cardiac Catheterization Laboratory, San Francisco VA Medical Center

Dr. Zimmet reports no financial relationships relevant to this field of study.

SYNOPSIS: Assessment of frailty adds important prognostic information about risk of death and disability following both surgical aortic valve replacement and transcatheter aortic valve replacement. Among the available instruments for assessing frailty, a scale known as the Essential Frailty Toolkit demonstrated the best correlation with outcomes.

SOURCE: Afilalo J, Lauck S, Kim DH, et al. Frailty in older adults undergoing aortic valve replacement: The FRAILTY-AVR study. J Am Coll Cardiol 2017;70:689-700.

As the options for treating aortic stenosis by
surgical and transcatheter procedures have increased, assessment of patients’ suitability for the procedure and subsequent prognosis has become more complex. Tools such as the Society of Thoracic Surgeons (STS) risk score and EuroSCORE are highly useful in predicting short-term risk of surgical aortic valve replacement (SAVR), but these instruments neglect many important details and are not validated for predicting outcomes for transcatheter aortic valve replacement (TAVR). Among the unmeasured variables is frailty, which carries great intuitive appeal in this realm, but clinicians have used it on a limited basis. Reasons for this restrained uptake include the time and effort involved in performing the various tests and, more importantly, a lack of consensus on which tools should be used to measure frailty. Although gait speed as a single measure has been the most commonly used test, multidomain frailty scales are preferred to achieve higher degrees of specificity for clinical outcomes. Most of these scales have been validated in individual studies, but head-to-head comparisons are lacking.

Accordingly, Afilalo et al presented the results of the FRAILTY-AVR study, which sought to prospectively evaluate the value of seven different frailty assessment tools in predicting outcomes in patients undergoing SAVR and TAVR. For the trial, patients > 70 years of age anticipating SAVR or TAVR were enrolled at 14 centers in Canada, the United States, and the Netherlands. Data were collected, and frailty was assessed by trained individuals using the Fried, Fried+, Rockwood, Short Physical Performance Battery, Bern, Columbia, and the Essential Frailty Toolset (EFT). The primary outcome measure was all-cause death at one year, with secondary outcomes of death at 30 days and a composite of death and increased disability at 12 months. Over the five-year study period, 1,020 older adults were enrolled, of whom 646 underwent TAVR and 374 underwent SAVR. The median age was 82 years, and the average STS score (predicted risk of mortality) was 4.3% (5.4% in the TAVR group, and 2.7% in the SAVR group). Notably, although frailty was assessed by these tools on a scale from least to most frail, it was reported as a dichotomous variable; patients either were judged to be frail or not. Frailty was approximately two-fold higher among TAVR patients compared to SAVR patients.

As expected, frailty was predictive of hard outcomes, with substantial variability among the different scales. The results showed that the EFT frailty assessment outperformed the other scales and was most strongly associated with one-year mortality, with an odds ratio of 3.72. The EFT also was the strongest predictor of death at 30 days and of worsening disability at one year. Further, it added incremental value to prediction models using the STS predicted risk of mortality score and procedure type in terms of predicting these hard outcomes.

The authors concluded that frailty is a strong predictor of mortality and disability following both SAVR and TAVR. Among available tools, the EFT demonstrated the most robust performance characteristics regarding predicting poorer outcomes following AVR.


Frailty as a concept in assessing patients for outcomes after cardiovascular interventions is intuitive and appealing, but in practice its measurement has been challenging to operationalize. The large number of measurement tools, some of which are challenging and time-consuming to administer, has led to confusion over the very definition of frailty, and has hampered its uptake as a clinical tool. In FRAILTY-AVR, the prevalence of frailty varied between 26% and 68%, depending on the particular tool used. This is a striking amount of variability, which highlights the need for this study. The tool that outperformed the others, EFT, is a relatively simple four-element, 5-point scale. Patients are scored for time to stand five times from a seated position (1 point if ≥ 15 seconds, 2 points if unable to complete), cognition (1 point for Folstein Mini-Mental State Examination [MMSE] score < 24), hemoglobin (1 point if < 13 g/dL in men or < 12 g/dL in women), and serum albumin (1 point if < 3.5 g/dL). Patients with ≥ 3 points are deemed frail, while 5/5 points defines severe frailty. The tests can be conducted easily and fairly rapidly (the most time-consuming part of the exam is the MMSE) in the office environment, and inter-observer variability is relatively low.

Assessment of older patients with severe aortic stenosis increasingly involves not just the choice of treatment modality (SAVR vs. TAVR), but also the determination in some patients about whether to treat. In this study, although procedural success was very high and short-term outcomes were good, the incidence of death or marked disability at one year was more than one-third for the whole group of patients. For those deemed frail by the EFT, the number was > 50%, while for those marked as severely frail (5 out of 5 points), 80% were dead or disabled at one year. These are sobering numbers.

With a relatively straightforward and validated tool, the assessment of frailty takes its rightful place as a central component in the evaluation of older adults with severe AS. Along with defining which patients are likely to benefit from AVR procedures, frailty assessment can assist in determining which patients are less likely to receive full benefit, either because they are unlikely to survive past one year, or because they will experience increased disability or worsened quality of life. Ultimately, patients and their families will benefit from this receiving this information as part of a shared decision-making process, as well as elements of informed consent.