Outcomes after living-donor liver transplantation (LDLT) at high Model for End-stage Liver Disease (MELD) scores are not well characterized in the United States. This was a retrospective cohort study using Organ Procurement and Transplantation Network data in adults listed for their first liver transplant alone between 2002 and 2021. Cox proportional hazards models evaluated the association of MELD score (<20, 20-24, 25-29, and ≥30) and patient/graft survival after LDLT and the association of donor type (living versus deceased) on outcomes stratified by MELD. There were 4495 LDLTs included with 5.9% at MELD 25-29 and 1.9% at MELD ≥30. LDLTs at MELD 25-29 and ≥30 LDLT have substantially increased since 2010 and 2015, respectively. Patient survival at MELD ≥30 was not different versus MELD <20: adjusted hazard ratio 1.67 (95% confidence interval, 0.96-2.88). However, graft survival was worse: adjusted hazard ratio (aHR) 1.69 (95% confidence interval, 1.07-2.68). Compared with deceased-donor liver transplant, LDLT led to superior patient survival at MELD <20 (aHR 0.92; P = 0.024) and 20-24 (aHR 0.70; P < 0.001), equivalent patient survival at MELD 25-29 (aHR 0.97; P = 0.843), but worse graft survival at MELD ≥30 (aHR 1.68, P = 0.009). Although patient survival remains acceptable, the benefits of LDLT may be lost at MELD ≥30.