This test was designed to evaluate the effect of lower-limb dominance and non-dominance shuttle runs under load carriage during different exercise load at the same exercise intensity on the balance responses. Ten healthy young males were joined in this experiment, they were (20.80±2.04) years old and (173.99±2.87) cm tall. In a randomized cross-over design, they performed four times shuttle runs under unilateral load carriage:20 m×5 at dominant side, 20 m×5 at non-dominant side, 20 m×10 at dominant side, 20 m×10 at non-dominant side respectively. Balance abilities were evaluated immediately and 20 minute post-exercise respectively, and R-R interval was recorded. The HR, EPOC and TRIMP for all exercise load were increased significantly after shuttle runs compared to rest (P<0.01), the EPOC and TRIMP for 20 m×10 shuttle runs were significantly higher than those for 20 m×5 shuttle runs at the dominant and non-dominant side (P<0.01). Compared to the rest, the instability indices (general, anteroposterior and mediolateral movement) for 20 m×5 and 20 m×10 shuttle runs immediately post-exercise at the dominant and non-dominant side were increased significantly (P<0.05), and the magnitude of the mediolateral movement for 20 m×10 shuttle runs was higher than that of 20 m×5 shuttle runs (P<0.05). While there were no obvious difference between the dominant and the non-dominant side (P>0.05), which showed symmetrical change. In addition, during 20 minute recovery, the balance ability for all exercise load was returned to the rest value (P>0.05). The shuttle runs could impair the trunk control ability immediately post-exercise, the magnitude of mediolateral movement was increased as the exercise load increased. The changes of balance responses were similar between the dominant and the non-dominant side, the dominant and the non-dominant side might show cross-effects.