Measurement of energy cost by the physiological cost index in walking after stroke

Objective: To compare the Physiological Cost Index (PCI) with direct measurement of oxygen consumption ((V) over dot O-2) as an estimate of energy cost in persons with stroke and healthy subjects. Design: Test-retest on separate days. A comparison of 2 methods of measurement. Measurements with and without an orthosis. Setting: A university hospital. Participants: A convenience sample of 20 persons with hemiparesis more than 6 months after stroke and 16 healthy subjects, ages 30 to 63 years. Interventions: Five minutes of treadmill walking at self-selected speeds while recording (V) over dot O-2 levels and heart rates. Additional data was recorded for 11 of the stroke subjects with and without an ankle-foot orthosis. Main Outcome Measures: (V) over dot O-2 and the PCI. Results: No significant differences were found in the PCI or (V) over dot O-2 between test and retest. Both PCI and (V) over dot O-2 per distance were higher for the stroke subjects compared with healthy subjects. PCI showed a larger dispersion than (V) over dot O-2 between test and retest. The regression analysis for PCI showed that the model including age, sex, group assignment, and (V) over dot O-2 could explain 53% of the variation. The PCI did not show a significant difference in walking with or without an orthosis, whereas (V) over dot O-2 differed significantly. Conclusions: The PCI showed limited reliability and validity as a measure of energy cost after stroke due to the extensive variability between test and retest.