OGC P23 Modified Frailty Index is not a good predictive tool of postoperative mortality following minimally invasive oesophagectomy

Abstract

Abstract Background Evaluating frailty in the elderly population is crucial in preoperative assessment to identify high-risk patients. The modified frailty index (m-FI) is known to predict postoperative mortality in various surgical populations, but its predictive value in oesophagectomy patients' long-term mortality is unclear. No studies have specifically quantified frailty and its association with long-term postoperative mortality in oesophagectomy patients. This study investigates the association between m-FI score and long-term mortality in patients undergoing minimally invasive oesophagectomy (MIO). Methods The m-FI is a scoring tool of 11 parameters quantifies frailty in surgical patients. Retrospective analysis (1/1/2014-31/12/2023) of patients aged ≥75 years undergoing minimally invasive oesophagectomy procedures for cancer, at a single h[ka1] igh volume tertiary referral centre was carried out. Primary outcome was defined as 1-year all-cause mortality with overall length of survival as the secondary outcome. After sensitivity analyses m-fi score of less than 1 or greater than 2 were most predictive of mortality. Results Ninety-one patients (mean age 78 years, age range 75-88, male to female ratio of 4:1) were included, with a median m-FI score of 1 (range: 0-5). Patients with an m-FI score of ≥2 had a higher 1-year mortality rate (30% vs. 19%, p=0.195) and shorter overall median survival time (34 months vs. 47 months, p=0.399) compared to those with a score of ≤1. After adjusting for covariates, including preoperative chemotherapy, age, sex, and cancer staging, an m-FI score of ≥2 had a hazard ratio of 1.65 (95% CI: 0.68 to 3.98, p=0.267) for 1-year mortality compared to ≤1. Conclusions The m-FI tool can predict post operative mortality in a range of surgical specialities. The results of our study would suggest that higher m-FI score would correlate with worse long term mortality outcomes, however at this stage the m-FI tool cannot be applied in a clinical setting to accurately define elderly patient's probable outcome following MIO. Additional multicentre data is needed to establish its true predictive value in this patient cohort.