Cohort study: The ability of 5 frailty tests and a comprehensive geriatric assessment to predict outcomes after aortic valve replacement was assessed.
Hosler QP, Maltagliati AJ, Shi SM, et al. A Practical Two-Stage Frailty Assessment for Older Adults Undergoing Aortic Valve Replacement. J Am Geriatr Soc. 2019 Jun 18. doi: 10.1111/jgs.16036.

OBJECTIVES: Despite evidence, frailty is not routinely assessed before cardiac surgery. We compared five brief frailty tests for predicting poor outcomes after aortic valve replacement and evaluated a strategy of performing comprehensive geriatric assessment (CGA) in screen-positive patients.

DESIGN: Prospective cohort study.

SETTING: A single academic center.

PARTICIPANTS: Patients undergoing surgical aortic valve replacement (SAVR) (n = 91; mean age = 77.8 y) or transcatheter aortic valve replacement (TAVR) (n = 137; mean age = 84.5 y) from February 2014 to June 2017.

MEASUREMENTS: Brief frailty tests (Fatigue, Resistance, Ambulation, Illness, and Loss of weight [FRAIL] scale; Clinical Frailty Scale; grip strength; gait speed; and chair rise) and a deficit-accumulation frailty index based on CGA (CGA-FI) were measured at baseline. A composite of death or functional decline and severe symptoms at 6 months was assessed.

RESULTS: The outcome occurred in 8.8% (n = 8) after SAVR and 24.8% (n = 34) after TAVR. The chair rise test showed the highest discrimination in the SAVR (C statistic = .76) and TAVR cohorts (C statistic = .63). When the chair rise test was chosen as a screening test (=17 s for SAVR and =23 s for TAVR), the incidence of outcome for screen-negative patients, screen-positive patients with CGA-FI of .34 or lower, and screen-positive patients with CGA-FI higher than .34 were 1.9% (n = 1/54), 5.3% (n = 1/19), and 33.3% (n = 6/18) after SAVR, respectively, and 15.0% (n = 9/60), 14.3% (n = 3/21), and 38.3% (n = 22/56) after TAVR, respectively. Compared with routinely performing CGA, targeting CGA to screen-positive patients would result in 54 fewer CGAs, without compromising sensitivity (routine vs targeted: .75 vs .75; P = 1.00) and specificity (.84 vs .86; P = 1.00) in the SAVR cohort; and 60 fewer CGAs with lower sensitivity (.82 vs.65; P = .03) and higher specificity (.50 vs .67; P < .01) in the TAVR cohort.

CONCLUSIONS: The chair rise test with targeted CGA may be a practical strategy to identify older patients at high risk for mortality and poor recovery after SAVR and TAVR in whom individualized care management should be considered.

Specialty Area Score
Hospital Doctor/Hospitalists
Internal Medicine
Comments from MORE raters


The great difference in outcome between surgical (8.8%) and transcatheter (24.8%) valve replacement is suggesting a great difference in the selection strategy, which should be taken into account also for a different frailty analysis in the two groups.

Anthony Maltagliati on 2019-07-12
Yes, the prospective nature of the patient selection in the study placed more frail individuals into the less invasive TAVR cohort (after they were deemed too high risk for SAVR) and is almost certainly the reason for the poor outcome disparity between groups. The data analysis was then done after observing what happened to individuals over time (i.e., retrospectively to determine which tests and moreover posits a practical two-step strategy that feeds those who fail rapid screening to a more robust secondary screen, the Comprehensive Geriatric Assessment) that would have had the greatest ability to detect poor outcomes in both SAVR/TAVR cohorts. The subsequent CGA for individuals who fail the chair-rise screen serves as a secondary more lengthy screen to confirm/deny frailty along with the degree of frailty, which helps provide realistic outcome expectations and offers a nuanced view into what specific domains of an individual's life are contributing to frailty. The hope is that these findings could possibly shape future frailty assessment and peri-operative planning for older adults undergoing aortic valve replacement.
0 0
Comments from JournalWise subscribers
Dr. Mark Mcconnell on 2019-07-06
Why shouldn’t ‘individualized care management‘ be done in all cases?
0 0
Anthony Maltagliati on 2019-07-22
If I recall, the data show that even the mandated gait speed in the ACC TAVR registry is only done ~50% of the time. This suggests a disconnect between how important frailty is for patient outcomes and the weight of importance it evidently currently carries to be worked up in the perioperative setting (broadly speaking, at least for determining gait speed in the context of AVR). This study is trying to show a feasible strategy to get the foot in the door and show that meaningful workups using an interdisciplinary model can theoretically improve outcomes and be efficient. Also, a CGA on every older adult undergoing AVR might not be totally necessary (many still have good outcomes) and might not be feasible or the most cost-effective strategy for many hospitals for a number of reasons (resources, staffing, etc).
0 0