Impact of Pre-Transplant Frailty on Early Outcomes Following Liver Transplantation: A Propensity-Matched Multicenter Cohort Study

Background/Objectives: Frailty is a validated predictor of waitlist mortality and perioperative risk in liver transplant candidates, but its association with early post-transplant outcomes in large real-world cohorts remains incompletely characterized. This study evaluated the association between administratively defined pre-transplant frailty and early clinical outcomes following liver transplantation. Methods: We conducted a retrospective cohort study using the TriNetX US Collaborative Research Network. Adults undergoing first-time isolated liver transplantation through February 2026 were included. Frailty was identified using ICD-10-CM codes for frailty, sarcopenia, cachexia, weakness, abnormal gait/mobility, or reduced mobility documented within 12 months before transplantation; patients coded only for nonspecific weakness were excluded from the frailty cohort. Patients underwent 1:1 propensity score matching using 18 baseline covariates, including demographics, comorbidities, laboratory values, albumin, and MELD-Na. The primary outcome was all-cause mortality at 7, 30, and 90 days. Secondary outcomes included acute kidney injury, prolonged mechanical ventilation, vasopressor requirement/hemodynamic instability, renal replacement therapy, ICU and hospital length of stay, and 90-day readmission. Sensitivity analyses used a restrictive ≥ 2-code frailty definition and substituted MELD 3.0 for MELD-Na in the propensity model. Results: Among 4860 eligible recipients, 742 had administratively defined frailty and 4118 did not. After matching, 730 patients remained in each group with well-balanced covariates. Administratively defined frailty was associated with higher mortality at 7, 30, and 90 days, with numerically smaller relative risks at later time points. It was also associated with higher risks of acute kidney injury, prolonged mechanical ventilation, vasopressor requirement/hemodynamic instability, renal replacement therapy, longer ICU and hospital stays, and 90-day readmission. Findings were directionally consistent in both sensitivity analyses. Etiology-stratified analyses were exploratory and showed no statistically significant heterogeneity across liver disease etiologies. Conclusions: In this large propensity-matched multicenter cohort, administratively defined pre-transplant frailty was associated with worse early outcomes after liver transplantation. Because frailty and several outcomes were identified using structured EHR and administrative data, findings should be interpreted as associative and hypothesis-generating. Prospective studies using validated frailty instruments and granular donor, intraoperative, and center-level variables are needed to confirm these findings.