BACKGROUND: This study investigated the clinical efficacy of asymmetric frequency stimulation (AFS), which involved the unilateral low-frequency and contralateral high-frequency stimulation, in the treatment of PD. METHODS: In a randomized controlled design, we compared the efficacy of Bi-LFS and Bi-HFS (n = 33) and explored the efficacy of AFS (n = 20) in the acute setting in one cohort of postural instability gait difficulty (PIGD) motor subtype of patients with PD. Motor evaluations included MDS UPDRS-III, Berg balance scale (BBS), and gait kinematic analysis, and 6 to 12 months of telephone follow-up. RESULTS: Bi-LFS was superior to Bi-HFS in improving axial (p = 0.044) and bradykinesia (p = 0.049) sub-score of the MDS UPDRS-III and BBS (p = 0.030). AFS significantly reduced the total score of the MDS-UPDRS-III compared to Bi-HFS (p = 0.015) and Bi-LFS (p = 0.010). AFS was comparable to Bi-LFS in axial, bradykinesia and balance symptoms but outperformed Bi-HFS (p = 0.022, 0.013, 0.040), while AFS was comparable to Bi-HFS but superior to Bi-LFS (p = 0.027) regarding the non-axial symptoms. Seventeen out of 20 participants (85.0%) who received AFS in the randomized phase favored AFS stimulation for the subsequently open-label phase, while half of those not tested with AFS switched back to Bi-HFS (46.2%). CONCLUSION: AFS may offer benefits for both axial and appendicular symptoms of PD and could be an alternative programming strategy in the management of parkinsonian axial symptoms refractory to conventional high-frequency STN DBS.