P120 Liver frailty index predicts transplant listing outcomes better than UKELD and MELD: a prospective cohort study

<h3>Background</h3> Frailty is increasingly recognised as a critical factor in the assessment and management of patients with chronic liver disease. The Liver Frailty Index (LFI) provides an objective measure of frailty, which is crucial for intervention and predicting patient outcomes. Our objective was to evaluate the association between LFI and transplant listing. <h3>Methods</h3> We conducted a prospective cohort study of all adults (18 years or older) evaluated electively for liver transplantation at King’s College Hospital between September 2023 and January 2024. Data collected included age, sex, transplant indications, liver disease aetiology, categorical LFI, UKELD, MELD, and listing outcomes (listed versus not listed). Descriptive statistics were used to summarise baseline characteristics. Chi-squared and t-tests assessed associations between LFI and listing outcomes. Crude and adjusted logistic regression identified factors independently associated with listing. We calculated the area under the curve (AUROC) to explore separately how LFI, UKELD, and MELD perform at predicting listing outcomes. <h3>Results</h3> The study included 72 patients with median age of 59 years (Q1 50, Q3 63). Fifteen (20.8%) were frail, 46 (63.9%) pre-frail, and 11 (15.3%) robust. Median UKELD was 52.5 (Q1 49, Q3 56.5) and MELD 13.5 (IQR Q1 9, Q3 20). The most common indication for listing was portal hypertensive complications (33.4%), followed by synthetic dysfunction (25.4%) and HCC (25.4%). There was a significant difference in listing outcomes for LFI (p=0.01) with 45.5% of those not listed and 10% of those listed categorized as frail, and 36% of those not listed and 76% of those listed categorized as pre-frail. There was no significant difference in listing outcomes by UKELD (p=0.66) or MELD (p=0.084). Pre-frailty as compared to frailty increased the odds of listing after confounder adjustment (OR 12.0; 95% CI 2.9–48.9). The AUROC for was 0.725 for the LFI model, 0.590 for the MELD model, and 0.556 for the UKELD model. A subgroup analyses excluding variant and hepatocellular carcinoma indications showed a trend towards robustness decreasing the odds of listing, but this did not achieve statistical significance due to small sample size. <h3>Conclusion</h3> The LFI is significantly associated with transplant listing outcomes and better predicts transplant listing outcomes as compared to UKELD and MELD. This suggests frailty assessment should be integrated in transplant evaluation protocols. Future studies will investigate how LFI changes over time on the transplant waiting list and its correlation with post-transplant outcomes.