283. MODIFIED FRAILTY INDEX AND CLINICAL FRAILTY SCALE TOOLS ARE NOT GOOD PREDICTORS OF SURVIVAL AFTER MINIMALLY INVASIVE ESOPHAGECTOMY

Abstract Background Assessing frailty in elderly patients is pivotal for preoperative risk stratification. Whilst the modified Frailty-Index (m-FI) and Clinical Frailty Scale (CFS) have been used to predict postoperative mortality and morbidity in various surgical contexts, their prognostic value for complications and overall survival following esophagectomy remains uncertain. Several studies have assessed pre-operative application of frailty indices and whether they have prognostic value for long-term postoperative outcomes, however, none have compared m-FI and CFS in patients undergoing esophagectomy. This study examines and compares m-FI and CFS with overall survival and post-operative complications in patients undergoing Minimally Invasive Esophagectomy (MIO). Methods We retrospectively analysed data from a prospectively maintained database, including patients aged ≥65 years undergoing MIO for esophageal cancer from January 2014 to December 2023, at a high-volume esophago-gastric tertiary centre. We compared the m-FI 11 and m-FI 5, CFS, all measuring frailty parameters in surgical patients. Frailty was defined as m-FI 11 ≥3, m-FI 5 ≥2, or CFS ≥4, based on the current literature evidence for application of these frailty assessment tools. Primary outcomes were 2- and 5-year all-cause mortality, with anastomotic leak, serious complications (Clavien-Dindo grade ≥IIIa) and average length of stay (LOS) as secondary outcomes. Results Among 174 patients (median age 72 years, range 65-85, male-to-female ratio 4:1), no significant associations were found between age, sex, TNM stage, and pre/post-op chemoradiotherapy within each scoring category. Neither tool showed significant associations for 2-, 5-year, and overall mortality (Table 1), nor for average length of stay. However, an m-FI 5 score ≥2 exhibited an odds ratio (OR) of 2.41 (95% CI 1.00–5.78, p = 0.049) for anastomotic leak, and a CFS score ≥4 had an OR of 2.46 (95% CI 1.02–5.90, p = 0.044) for serious complications. Conclusion In frail patients undergoing MIO, both the m-FI and CFS tools demonstrated no significant prognostic value for predicting overall survival. However, m-FI 5 ≥2 correlated significantly with anastomotic leaks, while CFS ≥4 correlated with serious complications. Relying on scoring tools for frailty assessment may inadvertently trivialise what is intended to be a multidimensional evaluation of health domains. Nevertheless, their utility in predicting complications should not be overlooked. We propose a more strategic use of frailty assessment tools to triage patients for comprehensive geriatric evaluation. This approach identifies patients at highest risk of complications and facilitates targeted prehabilitation for optimising outcomes.