jats:pBACKGROUND: A transfemoral bone-anchored prosthesis (TF-BAP) can be fitted with non-microprocessor-controlled knees (N-MPKs), or with microprocessor-controlled knees, which can be passive (P-MPKs) or active (A-MPKs). The next generation of A-MPKs, including powered knees, is emerging. The understanding of the loading applied on TF-BAP fitted with these A-MPKs is limited. OBJECTIVE: This cross-sectional study aimed to characterize the load applied on instrumented TF-BAP fitted with an A-MPK (Power Knee, Össur, Iceland) during standardized daily activities. Furthermore, some load characteristics applied during walking were compared with TF-BAP fitted with N-MPK and P-MPK reported in the literature using similar approach. METHODOLOGY: Thirteen males fitted with a transfemoral press-fit osseointegrated implant participated in this study between 2021 and 2022. Forces and moments applied on the instrumented TF-BAP, fitted with a Power Knee (PKA01) and Pro-Flex (LP, XC) or Balance S feet (ÖSSUR, Iceland), were measured wirelessly using an iPecsLab (RTC Electronics, USA) during walking, ascending and descending ramp and stairs. We followed a 28-step process to characterize the loading pattern considering spatiotemporal gaits variables as well as loading boundaries and extrema. FINDINGS: Overall, 1,327 steps were analyzed. The cadence ranged between 34 ± 6 and 49 ± 13 strides/min. The maximum forces and moments recorded on the long, anteroposterior and mediolateral axes of the transducer were 1,258 N, 331 N and 234 N as well as 19 Nm, 74 Nm and 91 Nm, respectively. CONCLUSION: The Power Knee, combined with Pro-Flex or Balance S feet, may improve participants’ capacity to ambulate. Comparations with reference values indicated that transitions from N-MPKs or P-MPKs to the Power Knee are considered safe and likely to improve efficiency. This study contributed to evidence-based recommendations of TF-BAP fitted with powered knees. Hopefully, this work will advance clinical practice guidelines for the growing population choosing bionic solutions. Layman's Abstract Advanced prosthetic knees play a key role in achieving rehabilitation success when a bionic limb is attached directly to the remaining bone in people with lower limb loss. The next generation of bionic knees, called powered knees, is emerging. Indeed, they become more frequently prescribed for individuals fitted with implant inserted into their bone. However, more scientific evidence is required to confirm the safety and efficacy of these knees before they can be accepted as the standard of care. Consequently, this study examined the forces applied between a powered knee and the implant in the residual bone, measured directly using a wireless device embedded in the prosthesis. Thirteen individuals with transfemoral amputations participated in the study between 2021 and 2022. The mechanical constraint was directly measured using a wireless device embedded into the prosthesis. The outcomes were benchmark against reference data provided in the literature. The outcomes of this study confirmed that the newly developed component, Power Knee, is both safe and potentially effective in restoring gait for users of bionic limbs. This work also improves standard methods for evaluating how prosthetic components mechanically affect the implant-bone interface in direct skeletal attachment of bionic limbs. Overall, this work will, hopefully, contribute to the development of clinical practice guidelines for the rehabilitation of the growing population choosing bionic solutions. Article PDF Link: https://jps.library.utoronto.ca/index.php/cpoj/article/view/45790/34428 How To Cite: Frossard L, Laux S, Geada M, Tronicke L, Fridriksson T, Lechler K. Evidence-based recommendation of a powered knee for transfemoral bone-anchored prostheses: A cross-sectional study. Canadian Prosthetics & Orthotics Journal. 2025; Volume 8, Issue 2, No. 4. Https://doi.org/10.33137/cpoj.v8i2.45790 Corresponding Author: Professor Laurent Frossard, (PhD)Affiliation: 1) YourResearchProject Pty Ltd, Brisbane, Australia; 2) Griffith University, Southport, Australia; 3) Queensland University of Technology, Brisbane, Australia; 4) University of the Sunshine Coast, Sippy Downs, Australia.E-Mail: laurentfrossard@outlook.comORCID ID: https://orcid.org/0000-0002-0248-9589</jats:p>