End-stage ankle arthritis is often treated surgically by total ankle arthroplasty (TAA) due to its potential to improve gait through increased joint range of motion and reduce pain. However, TAA's effect on gait stability is not well understood. This study explores the impact of TAA on gait stability, measured by Margin of Stability (MoS), in 148 patients with end-stage ankle arthritis. Kinematic data were collected pre-operatively, at 1-year post-op, and at 2-years post-op and the MoS was determined at heel strike and midstance for the anteroposterior (MoSAP) and mediolateral (MoSML) directions. A linear mixed effects model including gait speed as a factor was used to assess the effects of limb, session, and their interaction on outcome measures. A significant interaction (p < 0.002) between limb (surgical, nonsurgical) and session (pre-op, 1-year post-op, 2-years post-op) was identified for each MoS variable of interest. Cumulatively, our results suggest that the nonsurgical limb, MoSAP at heel strike and MoSML at midstance improved (increased) as time from surgery increased. These results suggest patients developed a compensatory movement pattern to navigate surgical limb single support. TAA reduces this compensation improving side-to-side symmetry, while not fully restoring symmetry by 2-years post-op. These results indicate that TAA could improve gait stability in patients with end-stage ankle arthritis, but further work is needed to understand the impact of TAA on altering fall risk.