User accommodation to an active microprocessor-controlled knee in individuals with unilateral transfemoral amputation: a 5-week non-randomized trial.

Only seven participants were studied, so the results are very small and cannot show how most people would respond. The trial was not randomized for receiving the knee; only the order of testing devices was randomized. Participants used the powered knee for only one hour per week for five weeks, and the authors state five one-hour sessions might be insufficient for accommodation. Table 1 shows missing data for many outcomes from session 2 onward, commonly 28.6%, and some heart rate values were missing at 42.9%. Participants were already rehabilitated, had comfortable sockets, and were aged 25 to 75 years with Medicare Functional Classification level K3-4, so findings may not apply to people who have not completed rehabilitation, have poor fit, pain, wounds, neurological conditions, or lower functional levels. The current prosthesis varied between participants and included passive and passive microprocessor-controlled knees, so the comparison was not against one standard device. The powered knee was used with a Pro-Flex XC foot and the person's own socket, so the effect of the knee alone cannot be separated from the foot, socket fit, alignment, or added weight. The powered knee added weight compared with the current knee, with differences between 690 and 1650 g, which may have affected speed and comfort. Tests did not include stair climbing or sit-to-stand, although the paper mentions these as important daily activities. The study used several primary outcomes and uncorrected p-values, and the correlation analysis was exploratory without p-values.

Do not conclude that a powered microprocessor knee is inferior for everyday walking. Seven people used it only one hour each week for five weeks, and the authors state this may be insufficient for accommodation.