Graft-Specific Neuromuscular Responses and EMG Activities During Running Before and After Running-Induced Fatigue in...

Fatigue during intense activity challenges the neuromuscular system and can reveal deficits not evident during lower-demand functional tasks. In individuals with anterior cruciate ligament reconstruction (ACLR), differences in graft source may be associated with graft-specific neuromuscular adaptations. Although fatigue-related neuromuscular deficits after ACLR have been reported, few studies have examined whether these responses differ according to graft type during running. This study investigated the effects of running-induced fatigue on lower-limb muscle activity during the stance phase in ACLR patients with different graft types, compared with healthy controls.

Fifty-three active males were allocated into four groups: single-bundle hamstring tendon (SB-HT), double-bundle hamstring tendon (DB-HT), allograft, and healthy controls. Participants completed a treadmill-based fatigue protocol that was terminated when they reached a Borg score > 17, which served as the primary criterion for confirming a high level of perceived fatigue. Heart rate ≥ 85% of estimated maximum heart rate was recorded as a complementary physiological indicator of high-intensity exertion. Electromyography (EMG) amplitudes of eight lower-limb muscles were recorded during the early and late stance phases of running, before and after fatigue. A two-way mixed-model ANOVA evaluated the effects of graft type and fatigue condition.

Fatigue reduced semitendinosus (p = 0.042, ηp² = 0.091) and gluteus medius (p < 0.001, ηp² = 0.442) activation during early stance regardless of group. Late-stance group effects were significant for the tibialis anterior, gastrocnemius, rectus femoris, biceps femoris, semitendinosus, and gluteus medius (p ≤ 0.05). Across stance phases, the SB-HT group showed greater hamstring and gluteus medius activation than healthy controls (p = 0.003-0.041), while the allograft group showed greater tibialis anterior, gastrocnemius, and rectus femoris activation during late stance (p = 0.005-0.043). The only significant group × fatigue interaction was observed for vastus medialis (p = 0.002, ηp² = 0.279), reflecting a fatigue-related reduction in the DB-HT group.

Graft type was associated with distinct lower-limb EMG activation patterns during running after ACLR. Fatigue revealed a DB-HT-specific reduction in vastus medialis activation, while the allograft and SB-HT groups showed different compensatory activation profiles. These findings suggest that graft-specific neuromuscular adaptations during fatigue should be considered when developing rehabilitation protocols to improve functional stability and future joint health.