Supplementary Material for: Non-motor effects of low-frequency deep brain stimulation of the subthalamic nucleus in Parkinson’s disease: a systematic review

Introduction:
Deep brain stimulation of the subthalamic nucleus is an effective therapy for the motor symptoms of Parkinson’s disease. Typically, stimulation is applied at a high-frequency (>=100 Hz) to alleviate motor symptoms. However, the effects on non-motor symptoms can be variable. Low-frequency oscillations are increasingly recognized as playing an important role in the non-motor functions of the subthalamic nucleus. Therefore, it has been hypothesized that low-frequency stimulation of the subthalamic nucleus (<100Hz) may have a direct effect on these non-motor functions, thereby preferentially impacting non-motor symptoms of Parkinson’s disease. Despite important therapeutic implications, the literature on this topic has not been summarized.

Method:
To understand the current state of the field, we performed a comprehensive systematic review of the literature assessing the non-motor effects of low-frequency stimulation of the subthalamic nucleus in Parkinson’s disease. We performed a supplementary meta-analysis to assess the effects of low vs. high-frequency stimulation on verbal fluency outcomes.

Results:
Our search returned 7009 results of which we screened 4199. 145 studies were further assessed for eligibility, and a total of 21 studies met our inclusion criteria, representing 297 patients. These studies were a mix of case reports and control trials. The four clinical outcomes measured were sleep, sensory perception, cognition, and mood. A supplementary meta-analysis of six studies investigating the impact of low-frequency stimulation on verbal fluency did not find any significant results when pooling across subgroups.

Conclusion:
LFS of the STN may have benefits on a range of cognitive and affective symptoms in PD. However, current studies in this space are heterogeneous and the effect sizes are small. Factors that impact outcomes can be divided into stimulation and patient factors. Future work should consider the interactions between stimulation location and stimulation frequency as well as how these interact depending on the specific non-motor phenotype.