The effects of microprocessor prosthetic knee use in early rehabilitation: A pilot randomized controlled trial.

Only 15 people completed the study. Enrollment stopped well short of the planned 24, partly because COVID-19 restrictions shut down in-person study activity and partly because few people with a recent above-knee amputation in one city are prosthesis candidates at all, so the trial had very little chance of detecting a real difference. People could not be blinded to which knee they were given, and the outcomes that did improve were the ones they reported themselves, so knowing which knee they had could have shaped those answers. Twenty outcomes were compared without any adjustment for multiple testing, and the authors say every result should be treated as exploratory. Three significant questionnaire results out of that many comparisons could be chance. There was no measure of how well people functioned before their amputation and no record of limb length, strength or range of motion at the start, so the two groups could have differed in ways the study could not see. Follow-up was three months, which the authors say is too short, and falls too uncommon, to judge anything about falls. The sample size estimates for future trials come from pilot data, and the authors themselves warn these are rough estimates rather than dependable numbers.

Funded by the US Office of the Assistant Secretary of Defense for Health Affairs through the Orthotics and Prosthetics Outcomes Research Program, award W81XWH 17-1-0617, with the note that the opinions are the authors' and not necessarily endorsed by the Department of Defense. The prosthetic knees and feet used in the study were provided by Otto Bock Healthcare LP (Austin, TX); the paper states Otto Bock had no influence on the design or conduct of the study, on data collection or analysis, or on the decision to publish. One author, Dr Friedly, is editor in chief of the journal and sits on other boards and committees.

Do not read the significant questionnaire results as proof that a microprocessor knee works better in early rehabilitation. The performance-based walking and mobility tests, which were the trial's main comparison, showed no significant differences, and with only 15 people finishing and 20 outcomes compared without adjustment for multiple testing, the three significant questionnaire findings could be chance.