Gait predicts cognitive decline in early Parkinson’s

PORTLAND, ORE. – Gait characteristics, rather than cognition, in early Parkinson’s disease predicted cognitive decline 3 years later in a small longitudinal study, suggesting that gait measures may provide a low cost clinical biomarker when combined with other predictors of cognitive decline in Parkinson’s disease.

Cognitive decline and dementia take a significant toll on patients’ personal, social, and economic affairs. Being able to identify at-risk individuals may allow for early behavioral and possibly pharmacological interventions. Gait has been shown to be a sensitive measure of cognitive decline in older adults, and the investigators here tested it as a predictor of cognitive decline in Parkinson’s disease (PD).

Led by Rosie Morris, PhD, of Newcastle University, Newcastle upon Tyne, England, researchers assessed gait characteristics and cognition in 119 patients within 4-6 months of a diagnosis of idiopathic PD and then 18 and 36 months later. The patients were recruited from the Incidence of Cognitive Impairment in Cohorts with Longitudinal Evaluation – Parkinson’s Disease (ICICLE-PD) study.

In this ICICLE-Gait substudy, participants walked for 2 minutes around a circuit that included a 7-meter instrumented walkway (GaitRite) that recorded 16 gait characteristics in the five independent domains of pace, variability, rhythm, asymmetry, and posture control. The investigators used the Montreal Cognitive Assessment (MoCA) battery of tests to assess global cognition, attention, fluctuating attention, executive function, working memory, visual memory, and visuospatial ability. They adjusted the results for covariates of age, education, gender, depression, and levodopa equivalent daily doses.

Dr. Rosie Morris

“We found that seven characteristics of gait were able to predict changes in measures of attention,” Dr. Morris said during a poster session at the World Parkinson Congress. The characteristics of gait were “mainly pace and variability and at diagnosis were sensitive to decline in tests of attention.”

With the multiple measures used in this study, a P value of less than or equal to .01 was considered statistically significant.

There were significant cognitive declines in the areas of attention, fluctuating attention, executive function, and visual memory. Fluctuating attention was predicted by measures of gait pace (step velocity, step length, and step swing: all P less than or equal to .007), gait variability (step stance and step length: both P less than or equal to .01), and postural control (P = .004). The pace measure of step length also predicted a decline in visual memory (P = .01).

Baseline MoCA predicted a decline in attention (P less than .001) but not in fluctuating attention or in visual memory.

The authors said that this study is the first “to demonstrate that discrete gait characteristics can predict measures of cognitive decline in early PD.” Asked why gait may predict cognitive decline, Dr. Morris postulated that certain characteristics of gait, such as speed and postural control, are related to cognition more so than other characteristics such as asymmetry.

These techniques may have clinical utility and not be merely of research interest. The parameters of gait may be measurable in the community using wearable sensors such as accelerometers “to measure discrete characteristics of gait ... So, ideally, it will be a possible future biomarker, and then we can measure them in the clinic and then identify risks from there,” she said, adding that she is planning research to explore the association of gait and cognition. Confounding factors yet to be examined are the effects of being on or off different medications and different medication cycles when patients are free-living in the community.

If researchers can validate that gait is a sensitive predictor of dementia, then the question becomes what to do about it. Dr. Morris suggested that physical therapy should incorporate more cognitive tasks.

Jori Fleisher, MD, a movement disorders neurologist at New York University Langone Medical Center, said that the study is “really interesting because right now there’s a huge search underway for biomarkers” to reliably predict progression of Parkinson’s disease or predict who is at risk for the disease.

“These investigators looked at whether something as seemingly simple as gait – so the way that someone walks, the way that their balance is, the pace of their walking – could predict decline in cognition. And that’s something of great interest to clinicians and to researchers: How do we figure out which of our patients are going to experience cognitive decline and in what different domains?” she said.

Although the researchers in this study used sophisticated and sensitive measures of several aspects of gait not generally available to clinicians, Dr. Fleisher predicted that it may be possible to extrapolate the findings to look for specific aspects of gait variability (possibly through the use of wearable devices) so the technique may become widely useful.