Study at a glance

In an n-of-1 study conceived and led by the participant himself, a 54-year-old man with Parkinson’s disease tested three deep brain stimulation frequencies in a single day, finding that frequency changes had a clinically meaningful impact on his gait, speech, and cognitive-motor performance.

Summary

Disclaimer: The following summary details independent academic research. Theia3D is intended for research, sports performance and educational use only. It is not cleared as a medical device and is not intended to be used for the diagnosis, treatment, mitigation or prevention of any disease, injury or medical condition.

Editor’s note: Theia3D version: not specified in the paper. The study cites Kanko et al. (2021) for the underlying markerless gait methodology but does not state a specific Theia3D software version.

Study Overview

Design

This was an n-of-1 (single-subject) study using a randomized, blinded, within-day crossover design. The study was proposed and led by its sole participant, who worked with a research team to design the DBS frequency comparisons based on his personal notes from prior DBS programming sessions; the proposed programs were reviewed by his treating neurologist, deemed safe, and uploaded to his patient controller, and a separate neurologist with DBS expertise assessed the participant at study commencement. Five 40-minute trial sessions, each using a different DBS program, were conducted across a single day, with the participant and project team blinded to session order until initial analyses were complete. Each session assessed gait, speech, cognitive-motor performance, and cortical activity (the latter reported separately in a different publication).

Participants

The study involved a single participant: a 54-year-old right-handed man diagnosed with Parkinson’s disease (right-side-dominant symptoms) at age 40. At age 50, following worsening symptoms, he underwent bilateral subthalamic nucleus DBS implantation (Vercise Gevia system with Cartesia directional leads, Boston Scientific). His medication regimen (pramipexole extended-release 3 mg daily and amantadine 100 mg twice daily) remained unchanged throughout the study, and he had not taken levodopa for more than three years prior. The study was approved by Queen’s University’s Health Sciences Research Ethics Board (approval no. 6041027), and informed consent was obtained on the day of the study.

Data Collection

The baseline session (Session #1) used the participant’s existing clinical DBS settings (179 Hz, -3.1/-2.9 mA, 60 µs on the most-ventral electrode contacts) and was not blinded. The four subsequent sessions tested two alternative frequencies (104 Hz and 149 Hz) at two current strengths each, in randomized order, with 15- and 60-minute rest periods held between current and frequency changes respectively to minimize carryover effects. During each session, gait was assessed for 5 minutes using a markerless motion capture system (Theia3D, Theia Markerless Inc., Kingston, ON, Canada) while the participant walked on a treadmill at 1.7 m/s, with arm swing (wrist-to-pelvis distance in the anterior-posterior plane) used as an indicator of rigidity and gait quality, and torso lean (thorax angle relative to vertical) used as an indicator of postural instability. Speech was recorded for 5 minutes per session using an iPhone, with the participant responding to prompts designed to elicit extemporaneous speech, and later analyzed by a speech-language pathologist experienced in Parkinson’s disease using a five-point impairment scale. The participant also completed four Kinarm robotic tasks (visually-guided reaching, reverse visually-guided reaching, spatial span, and object hit), performed with each arm where applicable, assessing motor, cognitive-motor, memory, and bimanual coordination function. Functional near-infrared spectroscopy (fNIRS) was also used to measure cortical activity during gait and speech tasks, with those results reported in a separate publication.

Analysis

Markerless gait data were analyzed per stride for arm swing range of motion and torso lean angle, with results presented by session as interquartile ranges. The proportion of time the participant held the treadmill’s support handrails during each session was also recorded as an indicator of gait difficulty. Speech recordings were rated by a speech-language pathologist on a five-point impairment scale covering voice quality, resonance, articulatory precision, and rhythm. Kinarm task performance was converted to impairment zeta scores, standardized against an age-, sex-, and dominant-hand-matched normative database, with scores above 1.96 indicating statistically significant impairment relative to that normative population.

Key Findings

  • Gait difficulty was substantially worse under the 104 Hz frequency sessions. The participant held the treadmill support handrails for 51% and 45% of the time during Sessions #2 and #3 (both 104 Hz), compared with 0% during the baseline session and both 149 Hz sessions (#4 and #5), indicating markedly reduced walking stability at the lower frequency.
  • Arm swing asymmetry and variability increased at the higher frequency relative to baseline. During Sessions #4 and #5 (149 Hz), the participant showed greater left-right arm swing asymmetry and greater stride-to-stride variation in arm swing compared with baseline, suggesting more subtle gait difficulty even without the handrail-holding seen at 104 Hz.
  • Rigidity was consistently greater in the right arm than the left across all five sessions, measured as reduced arm swing range of motion, consistent with the participant’s right-side-dominant Parkinson’s disease presentation.
  • Cognitive-motor performance on Kinarm tasks was significantly impaired specifically under the 104 Hz sessions. Reverse visually-guided reaching and other reaching tasks showed significant impairment relative to a matched normative database during Sessions #2 and #3, while the baseline session was the least impaired across five of the six Kinarm tests. Spatial span, a measure of short-term memory, showed no impairment under any session.
  • Speech impairment varied by session and did not track neatly with gait difficulty. Ratings ranged from mild during Session #2 to moderate-severe during Session #5, with the participant reporting that moderate-severe speech impairment, if it became his consistent baseline, would require him to take early retirement.
  • Torso lean, a marker of postural instability, did not vary substantially across the five DBS settings tested, in contrast to the more pronounced changes observed in arm swing, handrail use, and cognitive-motor scores.

What This Means for Movement Disorder Researchers

For researchers studying movement disorder interventions, this study’s central message is that stimulation frequency, a DBS parameter often adjusted less deliberately than current strength or electrode contact selection, can meaningfully shift gait stability, speech clarity, and cognitive-motor performance within the same patient, sometimes within the same day. The finding that lower frequency (104 Hz) was associated with substantially worse gait stability and cognitive-motor impairment in this individual cannot be generalized to other patients, given the single-subject design, but it illustrates a research approach: pairing objective, multi-domain outcome measures, including markerless gait analysis, with patient-reported experience to study how DBS parameter changes affect function. The authors position this as a starting point for future, larger clinical trials, not a basis for current treatment decisions.

The Role of Theia3D in Movement Analysis

In this study, Theia3D was used to assess gait quality objectively during each DBS testing session, measuring arm swing kinematics from treadmill walking as an indicator of rigidity and motor symptom severity, and torso lean as an indicator of postural instability. This markerless approach allowed the research team to capture repeated 5-minute gait assessments across five separate sessions within a single day without marker setup time between conditions, generating the stride-by-stride arm swing data that, together with the participant’s support-rail use, provided some of the clearest evidence in the study for how DBS frequency affected his functional mobility.

Study Limitations and Future Directions

The authors are direct about the central limitation of this work: as an n-of-1 study, its findings cannot be generalized to other individuals with Parkinson’s disease or DBS systems, and the authors explicitly describe the study as providing “proof of concept for future clinical trials” rather than definitive clinical guidance. The study tested only a narrow range of frequencies and currents on one day, and whether these acute, single-day findings would persist, or whether responses would differ, under sustained longer-term use of each frequency setting was not assessed. The fNIRS cortical activity data collected alongside the gait, speech, and cognitive-motor measures were not reported in this paper and are intended for separate publication, meaning the full multi-modal picture of the participant’s response is not yet complete in the available literature. The study design, in which the participant proposed the specific DBS parameters tested based on his own clinical notes, and served as the sole study subject, is also acknowledged by the authors as central to the study’s novelty but is inherently a departure from conventional investigator-led, blinded clinical trial design.

Citation: McAuley, M. D., Dean, J. M., Heming, E., Ionova, A. A., Onabajo, C. A., Solie, M., Stephenson, S. A., & Boehnke, S. E. (2025). Parkinson’s disease participant-led research on the impact of deep brain stimulation frequency. Journal of Parkinson’s Disease, 15(3), 668–671.

Link: https://doi.org/10.1177/1877718X251331904

Studying How DBS Parameters Affect Movement and Function?

This case illustrates how much can shift, in gait stability, speech, and cognitive-motor function, when a single DBS parameter changes, and how difficult that kind of change can be to capture without objective, repeatable measurement. If your research team is exploring how markerless motion capture could add objective gait data to movement disorder research protocols, talk to our team about what that kind of measurement could look like in your study design.

Recent Posts