Background: The objective of this study was to determine whether doubly labeled water (DLW) and a multi-sensor armband (SWA) could detect the variation in energy expenditure incurred by a period of increased exercise (EXE) versus a period of high sedentary activity (SED), in individuals with spinal cord injury (SCI).
Methods: Eight participants with SCI were submitted to 2 testing periods of energy expenditure assessment: 1) a 14-day phase during which sedentary living conditions were imposed and 2) a 14-day phase during which an exercise training intervention was employed. For each phase, total daily energy expenditure (TDEE) and physical activity energy expenditure (PAEE) were measured by DLW and SWA.
Results: Mean TDEE assessed by DLW, was significantly higher during EXE versus SED (11,605 ± 2151 kJ·day⁻¹ and 10,069 ± 2310 kJ·day⁻¹). PAEE predicted by DLW was also significantly higher during EXE versus SED (5422 ± 2240 kJ·day⁻¹ and 3855 ± 2496 kJ·day⁻¹). SWA-predicted PAEE significantly underestimated PAEE measured by the DLW during SED and EXE.
Conclusion: DLW is sensitive to detect variation in within-individual energy expenditure during voluntary increase in physical activity in individuals with SCI. SWA failed to detect statistically significant variations in energy expenditure between periods of high versus low activity in SCI.