Phenotypic Frailty Score Predicts Perioperative Outcomes for Geriatric Total Joint Arthroplasty

Various assessment tools are often used to predict perioperative morbidity among patients older than 75 years who undergo total joint arthroplasty. Yet, few studies describe the use of phenotypic frailty as a predictor for outcomes. The goal of this study was to assess phenotypic frailty with the Sinai Abbreviated Geriatric Evaluation (SAGE) and compare its utility with established assessment tools used in practice. We specifically asked: (1) Can SAGE predict 30-day outcomes, including postoperative delirium? (2) Can SAGE determine the risk of prolonged hospital length of stay? (3) Is SAGE predictive for 30-day readmissions? (4) Can SAGE determine the risk of discharge to a specialized facility? Patients undergoing total hip arthroplasty and total knee arthroplasty were evaluated with the American Association of Anesthesiologists Physical Status (ASA), Charlson Comorbidity Index (CCI), 5-point Modified Frailty Score (5-FS), and SAGE. Assessment scores were determined for each patient, and every incremental change in score was used to predict the likelihood of perioperative complications. A receiver operating characteristic analysis was also performed to calculate testing sensitivity for each assessment tool. The SAGE scores were more likely to predict 30-day complications (odds ratio [95 CI], 2.21 [1.32-3.70]), postoperative delirium (6.40 [1.78-23.03]), and length of stay greater than 2 days (3.90 [1.00-15.7]) compared with ASA, CCI, and 5-FS values. The SAGE scores were not predictive of readmission (1.77 [0.66-4.72]) or discharge to a specialized facility (1.48 [0.80-2.75]). The SAGE score was a more sensitive predictor (area under the curve, 0.700) for perioperative morbidity compared with ASA (0.638), CCI (0.662), and 5-FS (0.644) values. Therefore, SAGE scores can reliably assess risk of perioperative morbidity and may have better clinical utility than ASA, CCI, and 5-FS values for patients undergoing total joint arthroplasty. [Orthopedics. 2022;45(6):e315-e320.].