RESUMO
BACKGROUND: Assessment of metabolic energy expenditure in post-stroke patients using accelerometers is clinically important. However, understanding of the best placement of accelerometers on the body and methods for calculating activity counts are limited. METHODS: Thirty hemiparetic post-stroke patients participated in this cross-sectional study. Four triaxial accelerometers were attached to the hemiplegic and contralateral sides of the waist and ankles during various activities: lying, sitting, standing, stepping in place, and walking on a treadmill (1-5 kmh-1). Activity counts and metabolic energy expenditure of the patients were recorded simultaneously. Simple linear regression analyses were performed between the activity counts and energy expenditure. Activities were classified according to their intensity, using the definition of energetic sedentary behavior of post-stroke patients and a low fitness level group. RESULTS: The best estimate of energy expenditure was obtained when the accelerometer was worn on the contralateral ankle and the activity counts was calculated using the vertical and anteroposterior axes (R2=0.812). Six classes of activity intensity (sedentary: ≤1.5 METs, very light: 1.51-1.79, light: 1.80-2.59, moderate: 2.60-3.39, hard: 3.40-4.39, and very hard: ≥4.40) and corresponding activity counts cut-off points are presented. CONCLUSION: A triaxial accelerometer worn on the contralateral ankle and a method of calculating activity counts that includes at least the vertical and anteroposterior axes are recommended for estimating metabolic energy expenditure in post-stroke patients. The new activity counts cut-off points provide a significant advance in the interpretation of post-stroke monitoring in patients outside the hospital or rehabilitation center.
