UTILITY OF A GENERAL PROGNOSTIC SCORE IN IDENTIFYING PATIENTS WITH POOR OUTCOMES AFTER AORTIC VALVE REPLACEMENT

S Shi,J Afilalo,J Popma,K Khabbaz,R Laham,K Guibone,L Lipsitz,D Kim
DOI: https://doi.org/10.1093/geroni/igy023.3183
2018-01-01
Innovation in Aging
Abstract:The 2017 American College of Cardiology guidelines recommend consideration of life expectancy during aortic valve replacement (AVR) evaluation; however, how to estimate prognosis remains uncertain. We evaluated whether a popular general prognostic score, the Lee index (JAMA 2006), could aid in identifying patients with limited life expectancy who may not benefit from AVR. We prospectively enrolled 246 older patients undergoing surgical or transcatheter AVR at an academic center and assessed their ability to perform activities of daily activity and physical tasks over 12 months. The Lee index (range: 0–41) was calculated before surgery. Poor outcome was defined as death, or New York Heart Association Class III or IV with functional decline over 12 months. Of 91 surgical and 137 transcatheter patients with available outcome data, the mean Lee index score was 9.2 in surgical patients (range: 3–17) and 13.4 in transcatheter patients (range: 7–23). In the combined cohort, the risk of poor outcome increased with higher risk score quartiles (6.8%, 17.9%, 20.0%, 34.0%; p-for-trend<0.001). A similar trend was observed in the surgical patients (2.1%, 4.0%, 15.4%, 20.0%; p-for-trend=0.05). In comparison, no such trend existed in transcatheter patients (27.3%, 29.0%, 31.3%, 35.4%; p-for-trend=0.42). Our results suggest that while a prognostic model developed from the general population may be useful for surgical AVR patients, it has limited utility in identifying patients undergoing transcatheter AVR, with a higher burden of comorbidity, frailty, and functional limitations. Thus, better prognostic tools are needed to guide decisions for this specific population.
What problem does this paper attempt to address?