Abstract / Summary
Abstract Background Subthalamic nucleus–deep brain stimulation (STN–DBS) reliably improves cardinal motor symptoms in Parkinson's disease (PD). However, gait disorders and non-motor symptoms (NMS) may persist or emerge following surgery, diminishing quality of life and complicating clinical management. Objectives To characterize the dominant axes of clinical covariance across motor and non-motor domains before and after STN–DBS, and to determine whether non-motor symptom severity moderates post-DBS gait and mobility change. Methods 100 PD patients undergoing STN–DBS (mean age 60.2 ± 7.6 years; 67% male) were studied. Principal component analysis (PCA) was performed on thirteen motor and non-motor sub-scores at pre-DBS, 1-year follow-up (1-y FU), and a subset ( n = 33) at 3–5-year follow-up. Moderation was tested using analysis of covariance (ANCOVA): each 1-year motor outcome was regressed on its pre-DBS baseline, a non-motor moderator, and their interaction, adjusting for age and disease duration, with a significant interaction indicating that baseline non-motor severity was associated with the degree of subsequent motor recovery. Results PCA identified a core motor triad: gait, mobility, and freezing of gait as the primary source of clinical variance at all timepoints (PC1: 35% pre-DBS, declining to 28% at 3–5-y FU), PC1 was the only component to exceed a permutation null ( p < 0.001), with memory-related impairment increased at longer-term follow-up. Moderation analysis showed that higher pre-operative mood burden specifically for psychomotor and psychosomatic symptoms were associated with poorer recovery of mobility (Psychomotor β = 1.16, p < 0.05). At 1-y FU anxiety and psychosomatic symptoms showed concurrent associations with mobility and overall motor burden (Anxiety β = 2.30, p < 0.05). Conclusions STN–DBS outcomes extend beyond global motor improvement and may be shaped by associations spanning the motor, cognitive, and mood domains. Higher NMS baseline burden was associated with poorer mobility recovery, and mood symptoms at follow-up were associated with concurrent motor burden. These exploratory findings support considering multi-domain pre-operative assessment and post-operative mood monitoring, as potential targets for optimizing motor outcomes.