Objective To compare the clinical effects of palpation, ultrasound and electromyography (EMG) guided injection of botuli-num toxin type A (BTX-A) on spastic equinovarus in stroke patients. Methods Stroke patients with equinovarus were randomly assigned to palpation-guided group (n=19), ultrasound-guided group (n=21) and EMG-guided group (n=19). All of the patients received injection of BTX-A 300 U in gastrocnemius, soleus and tibialis posterior, guided in their own ways. They were measured with modified Ashworth Scale (MAS), Physician Rating Scale (PRS), speed of gait, passive range of motion (PROM) of ankle dorsiflexion and eversion before and 2 weeks, 4 weeks, 8 weeks, and 12 weeks after injection. Results Compared with the baseline, all the indexes significantly improved after in-jection in all the groups (P<0.05). Compared with the EMG-guided group, the ultrasound-guided group improved significantly MAS at 4 weeks, PROM of ankle eversion at 4 and 12 weeks (P=0.046, P=0.008, P=0.038). Compared with the palpation-guided group, the ultra-sound-guided group improved significantly in MAS (P=0.015), PRS (P=0.01), PROM of ankle dorsiflexion and eversion (P=0.003, P=0.011) at 4 weeks, in MAS (P=0.002) and PROM of ankle dorsiflexion (P=0.022) at 8 weeks, in MAS (P=0.011) and PROM of ankle ever-sion (P=0.018) at 12 weeks. Compared with the palpation-guided group, the EMG-guided group improved significantly in PRS at 4weeks, PROM of ankle dorsiflexion at 4 and 8 weeks (P=0.048, P=0.08, P=0.026). There was no significantly difference in the variations with the time among 3 groups. Conclusion The correction of equinovarus in stroke patients can be obtained by BTX-A injection with any type of guidance technique. Ultrasound-guided technique is considered a valid alternative, which has a slight advantage over EMG-guided tech-nique. Both the EMG-and ultrasound-guided techniques obtained superior results to palpation-guided technique.