Sensitivity and reproducibility of accelerometry and heart rate in physical strain assessment during prosthetic gait.

Accelerometry and heart rate (HR) are frequently used indicators of physical strain during normal daily life. The present study focused on the sensitivity and reproducibility of accelerometry (body motility, the intensity of body movement measured with accelerometry) and HR (percentage maximal heart rate reserve, %HRR(max)) in the assessment of physical strain during walking in persons with a lower leg amputation, using persons without an amputation as reference. Ten patients with an amputation of the leg and ten comparison subjects performed, at an interval of 1 month, the same walking protocol three times. Subjects walked at a preferred speed and at fixed speeds. At their preferred walking speed, speed (0.63 vs 1.31 m s(-1), P=0.001), body motility [0.53 vs 0.91 (arbitrary unit), P=0.001] and %HRR(max) (42.5 vs 27.6, P=0.02) differed between the amputation group and the comparison group. At fixed walking speeds, only %HRR(max) differed between groups ( P</=0.002) and showed a session effect ( P=0.02). The relationship between body motility and %HRR(max) in the patient group was significantly different from that in the comparison group. It can be concluded that accelerometry is strongly related with walking speed, but not sensitive to differences and changes in economy, contrary to %HRR(max). The use of accelerometry as an indicator of the level of physical strain in persons with an amputation is not feasible. The added value of a calibration procedure has to be studied.