Can the Tilburg Frailty Indicator predict post-operative quality of recovery in patients with gynecologic cancer? A prospective cohort study

ObjectiveFrailty is a marker of physiologic decline within multiple organ systems. The Tilburg Frailty Indicator (TFI) is an instrument for assessing frailty. We evaluated the ability of the TFI to predict the quality of post-operative recovery in patients with gynecologic cancer and explored the associations between frailty, post-operative complications, and length of stay. MethodsWe conducted a prospective cohort study of patients scheduled for radical gynecologic cancer surgery between May 2021 and January 2022, and defined a TFI score >= 5 as 'frailty'. Our primary outcome was the post-operative quality of recovery based on the Quality of Recovery-15 (QoR-15), and the secondary outcomes were post-operative complications and length of stay. Multiple logistic regression was used to examine the relationship between frailty and outcomes. We developed receiver operating characteristics (ROCs) and assessed areas under the ROC curves (AUCs) to explore the ability of frailty to predict the quality of post-operative recovery. ResultsA total of 169 patients were included. The prevalence of frailty using the TFI was 47.9% in this cohort. In the multivariate regression analysis, frailty emerged as a significant predictor of the 3-day QoR-15 score (aOR 11.69, 95% CI 4.26 to 32.08; p<0.001) and complications (aOR 10.05, 95% CI 1.66 to 60.72; p=0.012). Frailty was not associated with length of stay (aOR 2.12, 95% CI 0.87 to 5.16; p=0.099). The combination of the TFI, American Society of Anesthesiologists (ASA) classification, and types of cancer resulted in an increase in the AUC compared with the TFI alone (AUC 0.796, 95% CI 0.727 to 0.865; p<0.05). ConclusionsThe use of the TFI may assist surgeons in estimating the risk with respect to post-operative quality of recovery and complications in patients with gynecologic cancer. Combining the TFI with ASA classification and cancer type is expected to improve the predictive ability of poor quality of recovery.