Frailty Index Enhances Mortality Risk Prediction in TAVR

Patrice Wendling

March 08, 2018

A prospective study has shown for the first time that adding frailty assessment to conventional risk scores improves the prediction of 1-year mortality for patients undergoing transcatheter aortic valve replacement (TAVR).

Not only did mortality increase with increasing frailty, but Swiss researchers also showed that their frailty index accounted for 58.2% of the predictive information when combined with the logistic EuroSCORE and for 77.6% when combined with the Society of Thoracic Surgeons (STS) risk score.

"Our frailty index is just one proposition, and it was surely not the intention of our paper to say that this specific frailty index should be used," Andreas Schoenenberger, MD, a geriatrician at Bern University Hospital, Switzerland, told | Medscape Cardiology in an email. "Our paper rather wants to prove a concept, namely that frailty, howsoever it is measured, adds to the risk prediction."

The paper was published online in JACC: Cardiovascular Interventions.

"Frailty should be considered by every heart team when assessing the risk for the individual TAVR candidate," Lars Søndergaard, MD, DMSc, Copenhagen University Hospital, Denmark, and colleagues write in a related editorial. Guidelines also recommend using an objective approach rather than eyeballing frailty, but no gold standard exists.

For the study, 330 consecutive patients at least 70 years of age (median age, 83.6 years; 56.4% female) who underwent TAVR for symptomatic severe aortic stenosis received a baseline examination that included the previously validated Mini-Mental State Examination (MMSE), Timed Up and Go (TUG) test, Mini Nutritional Assessment (MNA), basic activities of daily living (BADL), and instrumental activities of daily living (IADL).

Using their previously developed frailty index, the researchers calculated a score ranging from 0 to 7 from the baseline components:  2 points if MMSE score was less than 21 points and 1 point for each of the following: MMSE score at least 21 and less than 27 points, TUG score at least 20 seconds, MNA score less than 12 points, BADL at least one limited activity, IADL at least one limited activity; and a preclinical mobility disability, defined as decreased frequency of walking 200 meters or of climbing stairs during the previous 6 months.

At baseline, the median frailty index score was 3.0 points, median logistic EuroSCORE was 19.2%, and median STS score was 6%. On the basis of the frailty index, one third of the patients had cognitive impairment and almost one third had mobility impairment.

The overall 1-year mortality rate was 0.18 (95% CI, 0.13 - 0.23). The mortality risk for patients with a frailty index score of 0 was 0.04 (95% CI, 0.01 - 0.27), steadily increasing to 0.78 for those with a score of 6 (95% CI, 0.37 - 1.63) and 3.94 for those with a score of 7 (95% CI, 1.27 - 12.2).

The EuroSCORE, STS score, and frailty index were all predictive of 1-year mortality in Cox regression models, although C-statistics were only 0.67, 0.64, and 0.66, respectively.

Sensitivity analyses, however, showed the frailty index improved risk prediction for survival, even when known risk factors, such as anemia, atrial fibrillation, and left ventricular ejection fraction less than 35%, were added to the conventional risk scores.

Notably, IADL and preclinical mobility disability did not predict survival, while TUG was best at predicting 1-year mortality (hazard ratio, 3.41; 95% CI, 1.95 - 5.97; likelihood ratio chi-square test, 19.84).

Asked why physicians shouldn't just add TUG to conventional risk algorithms, Schoenenberger responded that several studies have shown that mobility is probably the most important single component for risk prediction, but that how best and most efficiently to assess functional aspects is still an open research question.

"In my opinion, there is a need for a multicentric prospective study evaluating a variety of easy functional tests and examining which of these tests adds most to the conventional risk scores," he said. "The functional tests should be performance-based and not rely on patient self-report or the observer's subjective rating."

Schoenenberger said the frailty index takes about 20 minutes to administer and that it makes sense to measure frailty systematically in all patients after age 70 or 75. However, it would not go amiss to measure frailty in younger patients if they make a frail impression.

The researchers previously showed that frailty may be the consequence of severe aortic stenosis and reversed by TAVR, but  these results suggest "there might be a level of frailty which is irreversible."

They note that 1-year mortality exceeded 50% in the two highest-risk groups and "might be considered unacceptably high for performing TAVR. Therefore, our frailty index might have the potential to guide triage in patients being evaluated for TAVR and to recommend palliative treatment in these highest-risk patients."

Søndergaard and his editorial colleagues agree that incorporating frailty into a risk algorithm for TAVR patients "would be a much needed tool that could aid clinical recommendations and also provide information on potential reversible risk factors that could guide in-hospital stay and post-procedural management to a better outcome for the patient."

They caution, however, that the results were derived from a single center, most patients (93.6%) underwent transfemoral TAVR, and advances in operator experience and knowledge and device design have resulted in a lower procedural mortality rate, even in high- and extreme-risk patients.

Further, the logistic EuroSCORE is no longer recommended in a clinical setting, and another study has shown that poor mobility as defined in the EuroSCORE II was the best predictor of long-term mortality after TAVR.

"Finding the optimal frailty scores and their TAVR-specific scale warrants investigation in larger trials," Søndergaard and colleagues write.

The study was supported by research grants from Bern University Hospital and the Swiss National Science Foundation. Schoenenberger and the editorialists report no relevant financial relationships.

Follow Patrice Wendling on Twitter: @pwendl. For more from, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.