P211 COMPARISON OF PROGNOSTIC VALUE FOR G8 SCREENING TOOLS WITH FTRST IN OLDER PATIENTS OF ESOPHAGEAL CANCER
Abstract Aim We investigated two prognostic value of geriatric screening tools, G8 and Flemish version of the Triage Risk Screening Tool (fTRST), for overall survival and postoperative outcome in older patients undergoing esophagectomy for esophageal cancer (EC). Backgrounds&Method The elderly cancer population is a very heterogeneous group due to differences in comorbidities and functional status. G8 and fTRST are short and easy tools to administer in clinical settings and reported to be useful for identifying patients with geriatric risks. However, the prognostic value of G8 and fTRST has not been compared in EC patients after surgery. Patients aged ≥ 70 years old with EC were retrospectively included who received an operation at National Center for Global Health and Medicine from April 2014 to December 2017. G8 and fTRST were administered to all patients. The impaired were defined as a G8 ≤ 14 and fTRST ≥ 2. We evaluated overall survival (OS) and 30-day postoperative complications classified into Clavien-Dindo (CD) severity grade. Results 63 patients were included. Median age was 76 years (range, 70 to 89 years), and 84% of patients were men. 44 patients (69%) were G8 impaired, 22 patients (34%) were fTRST impaired, 23 patients (36%) were impaired on both screening tools. G8 was an independent predictor of overall survival (hazard ratio 9.9; 95% confidence interval 1.24-79.4, p=0.02), while fTRST was not. The CD≥3 postoperative complications occurred in 33 patients (52%). In univariable logistic regressions, ECOG-PS, G8 and fTRST were associated with CD≥3 complications. G8 alone was not independently predictive for CD≥3 complications, however combined with fTRST, the predictive value of G8 was increased (p=0.001). Conclusions G8 is useful for prognostic value of OS and prior to fTRST in EC. G8 combined with Ftrst has the strongest predictive value for postoperative CD≥3 complications. Further studies are needed to design interventions to improve outcomes for those frail patients.