Tailored deep brain stimulation improves gait in Parkinson's disease

In patients with Parkinson's disease, changes in walking ability can be very pronounced. The so-called "Parkinsonian gait" can include changes in stride length and asymmetry between the legs. These gait disturbances reduce a person's mobility, increase the risk of falls, and significantly affect the patient's quality of life.

Although high-frequency deep brain stimulation (DBS) is highly effective in reducing symptoms of tremor, rigidity, and bradykinesia (slowness of movement), its effects on gait are more variable and less predictable in patients with severe gait disorders. The main challenges in improving the results of DBS for the treatment of gait disorders remain the lack of a standardized gait metric for clinicians to use when adjusting stimulation parameters, as well as a lack of understanding of the effects of different stimulation factors on gait.

In a recent study, researchers at the University of California, San Francisco (UCSF) developed a systematic way to quantify key aspects of gait characteristic of Parkinson’s disease and used machine learning techniques to select optimal DBS settings for each patient. These personalized settings led to significant improvements in walking, such as faster and more stable strides, without worsening other symptoms.

The results of their study are published in npj Parkinson's Disease.

"We approached the task of optimizing the DBS settings as an engineering problem, with the goal of modeling the relationship between stimulation parameters, brain activity, and gait performance," said first author Hamid Fekri Azghomi, PhD, a postdoctoral fellow in the UCSF Wang Lab.

How to Optimize Gait Performance

In the study, patients with Parkinson's disease were implanted with DBS devices that not only stimulated the brain but also recorded neural activity as they walked. During clinic visits, DBS parameters were varied within safe ranges to study their effects on gait function. In response to each set of settings, patients walked a closed circuit of approximately six meters while neural data and gait kinematics were continuously recorded.

The researchers developed the Walking Performance Index (WPI), which assessed gait metrics such as stride length, stride speed, arm swing amplitude, and gait coherence. By combining these metrics, the WPI provided a comprehensive gait assessment, covering multiple dimensions of motor function affected by Parkinson's disease.

“Our results confirmed that changes in DBS settings were effectively captured by the WPI and were consistent with patient and clinician assessments at each visit,” Azgomi said. “This validation confirms that the WPI is an effective metric for assessing and targeting gait improvement in people with Parkinson’s disease. Using these methods, we were able to predict and identify personalized DBS settings that improved the WPI.”

The researchers also identified patterns of brain activity associated with improved walking. Using multivariate models, the authors identified unique neural dynamics that differentiate optimal gait from less efficient patterns. Improved gait was associated with decreased beta-band brainwave activity during specific phases of the gait cycle in the globus pallidus, a brain region associated with muscle loss in people with Parkinson’s disease.

These data, along with the identified individual neural biomarkers, highlight the importance of personalised, data-driven interventions to improve gait in people with Parkinson's disease.

"This work not only deepens our understanding of how DBS influences movement, but also demonstrates the potential for personalized neuromodulation for Parkinson's disease and other neurological disorders, bringing us closer to smarter, more effective therapies," said senior study author Doris Wang, MD, PhD, a neurosurgeon and associate professor of neurosurgery at UCSF.