Energy expenditure and accelerometer cut-points for sedentary behavior and physical activity in spinal cord injury : implication for guiding and prevention

Abstract

Background: A motor-complete spinal cord injury (SCI) alters the prerequisites for physical activity (PA) and subsequently energy expenditure. Persons with SCI above thoracic level six have a compromised physiological response, which further compromises energy expenditure, during exercise. Weight gain and lower levels of PA increase the risk of lifestyle-related diseases and their risk factors. Yet little is known about energy expenditure and the intensity of different activities from rest to maximal effort, nor about clinically useful cut-points for accelerometer in motor-complete SCI.

Aim: The aim of the work reported in this dissertation was to extend knowledge about energy expenditure, oxygen consumption and heart rate during rest, standardized activities and peak capacity in people with motor-complete SCI. A further aim was to evaluate how clinically accessible methods can be used to measure and describe activity patterns and intensity levels.

Methods: Participants were 64 persons with motor-complete SCI. Seventeen were women. Twenty-six had tetraplegia (C5-C8) and 38 (T7-T12) had paraplegia. Studies I and II are based on data from indirect calorimetry during rest and during standardized activities. In study III data from peak capacity (VO2peak), peak heart rate (HRpeak) and Borg rating of perceived exertion (RPE) were used for categorizing standardized activities into different levels of intensity. In study IV dominant-wrist-worn accelerometer (ActiGraph GT3X+) cutpoints were created by using receiver operating characteristic (ROC) curves, and the relative intensities established in Study III.

Results: Studies I and II showed that mean resting oxygen consumption for the whole group, no gender differences, was 2.52 ml·kg-1 ·min-1 and the variable that best explained the variance for energy expenditure during rest (24 hours) was bodyweight r 2= 0.37 for the total cohort. During non-exercise activities (wheeling indoors/outdoors Borg RPE 10-11 and setting table), the activity energy expenditure (total energy expenditure minus resting energy expenditure) for tetraplegia increased between two and four times compared to sedentary, and between three and five times during exercise activities. Motor-complete paraplegia could increase energy expenditure between three and six times during non-exercise activities and between 6 and 14 times during exercise. In study III absolute VO2peak was 0.76 L∙min-1 in tetraplegia and 1.36 L∙min-1 in paraplegia, differing significantly between men and women for both tetraplegia and paraplegia (p≤0.001). The significant difference disappeared for both groups when the VO2 was related to body weight (p=0.43). Further, in study III all activities were categorized into sedentary, light, moderate and vigorous levels of intensity, based on percentage of VO2peak, heart rate and Borg. Thus, many of the non-exercise physical activities (NEPA) were categorized as moderate or vigorous for persons with tetraplegia. Study IV showed a high correlation of 0.8-0.9 between percentage of VO2peak, absolute VO2 (MET) and accelerometer vector magnitude counts (VMC). The ROC curve analysis showed an area under the curve (AUC) of 0.8, which resulted in cut-points for different intensity levels such as, moderate-to-vigorous intensity of 4887 VMC (tetraplegia) and 9515 VMC (paraplegia).

Conclusion: Given the large inter-individual differences, person-specific information regarding RMR is crucial. The VO2peak was lower for person with tetraplegi, affecting the relative intensity level for activities of daily living. Activity energy expenditure, especially during daily activities, may increase total daily energy expenditure since daily activities are easily accessible and can be performed for long periods. Specific accelerometer cut-points for motor-complete tetraplegia and paraplegia from ROC curve analysis may be used in rehabilitation and research to capture activity patterns objectively.