Home Aerobic Exercise Beats Balance Training for Cerebellar Ataxia in 12-Month RCT

Cerebellar ataxias affect roughly 150,000 Americans, causing progressive loss of coordination, balance, and mobility, with mean annual healthcare costs exceeding $18,000 per person. While balance training has been the clinical standard, it is generally thought to help patients compensate for deficits rather than alter disease course. This trial asked whether high-intensity aerobic exercise — which has shown neuroprotective effects in animal models and benefits in other neurodegenerative conditions like Parkinson's and Alzheimer's disease — could outperform balance training in this population.

The study enrolled 62 adults with cerebellar ataxia (mean age 54.4 years, 47% female, mean SARA score 12.1) from Columbia University Medical Center. Participants were randomized 1:1 to either home stationary cycling or home balance exercises, both prescribed at 30 minutes per session, five days per week. The aerobic group progressively increased intensity up to 85% of age-predicted maximum heart rate; the balance group performed structured exercises at progressively harder difficulty levels. Both groups received biweekly phone support for the first six months only, after which all structured support was withdrawn. Assessments occurred at baseline, 6, 9, and 12 months, and training adherence was objectively tracked via Fitbit or Apple Watch.

The primary outcome was the Scale for the Assessment and Rating of Ataxia (SARA, range 0–40, minimal clinically important difference ~1.0 point). Using linear mixed-effects models on an intent-to-treat basis, the aerobic group showed significantly greater SARA improvement than the balance group at 6 months (β = −1.53, 95% CI −2.44 to −0.61, p=0.001), 9 months (β = −1.34, p=0.006), and 12 months (β = −1.40, p=0.005). These differences exceeded the minimal clinically important difference at every time point. Secondary outcomes reinforced these findings: the aerobic group improved VO2max by 4.26 mL/kg/min more than the balance group (95% CI 2.1–6.4, p<0.001) and reduced fatigue scores (FSS) by 9.38 points more (95% CI −15.1 to −3.7, p=0.001).

A particularly compelling finding emerged from the long-term follow-up. Among aerobic group participants who continued training independently after study support ended, SARA scores improved by a mean of −3.81 points from baseline (95% CI −2.2 to −5.4) at 12 months. In contrast, those who reduced or stopped training saw their scores trend back toward baseline (change of +0.4, 95% CI −0.4 to 1.2). This dose-response relationship suggests a genuine exercise-dependent neurobiological effect rather than a learned compensation. Training adherence was over 70% in both groups during the supported first six months, dropping to 38.5% (aerobic) and 24% (balance) at 12 months after support was withdrawn.

The study carried a low dropout rate of 17.7% at one year, and adverse events were minimal and not significantly different between groups. Participants included a wide range of ataxia subtypes, with most having genetically confirmed spinocerebellar ataxia. Structural MRI data were collected and will be reported separately, potentially shedding light on whether aerobic exercise induces measurable cerebellar volumetric changes. These results build a compelling case that high-intensity aerobic exercise should be integrated into ataxia care guidelines, not merely as a complement to balance therapy but as a primary intervention with disease-modifying potential.