Benefits of hamstring lengthening surgery on the sagittal plane in children with cerebral palsy have been previously demonstrated, but there is limited information on its effects on the transverse plane. This study compared the effects of medial hamstring lengthening (MHL) with those of medial and lateral hamstring lengthening (MLHL) procedures in the transverse plane. Children with gross motor function classification system (GMFCS) levels I-III who had MHL or MLHL were included. Baseline, short-(1-2 years), and long-term (3+ years) postoperative threedimensional gait analysis outcomes were compared using analysis of variance. Children were excluded if they had concurrent osteotomies or tendon transfers. One hundred fifty children (235 limbs) were included, with 110 limbs in the MHL group (age 8.5 ± 4.1 years, GMFCS I-27%, II-52%, and III-21%) and 125 limbs in the MLHL group (age 10.0 ± 4.0 years, GMFCS I-23%, II-41%, and III-37%). Time between surgery and short-and long-term follow-up gait analysis was 1.5 ± 0.6 years and 6.6 ± 2.9 years, respectively. Transmalleolar axis became more external after MHL at both short and long terms (P < 0.05), whereas there were only significant differences at long term in MLHL (P < 0.05). Although hamstring lengthening has a positive impact on stance phase knee extension in children with cerebral palsy, intact lateral hamstrings after MHL likely contribute to increased tibial external rotation after surgery. Significant increases in external rotation at the knee in the long term are likely related to a trend present with growth in children with cerebral palsy rather than a direct result of surgical intervention.