Association between measures of cognitive function on physical function in novice users of a lower limb prosthesis

Background: Cognitive impairment is prevalent in people with lower limb amputations (PLLA) and is associated with adverse outcomes, such as falls and worse rehabilitation outcomes. Physical function tests are essential to examine abilities; however, no research in PLLA has clarified the magnitude of cognitive demands amongst available tests in users novice at walking with a prosthesis.

Research question: Is there an association between cognitive and physical function for PLLA novice at walking with a prosthesis?

Methods: People from inpatient prosthetic rehabilitation were recruited. Inclusion criteria were: age ≥ 50 years, unilateral transtibial amputation and able to walk independently. Gait velocity and the L Test under single-task (usual) and dual-task (walking while counting backwards) conditions assessed functional mobility. The Four Square Step Test (FSST) examined dynamic balance. The Montreal Cognitive Assessment (MoCA) and the Trail Making Test (TMT-B) assessed global cognitive status and executive function, respectively. Multivariable linear regressions evaluated the association of cognition on physical function.

Results: Twenty-two people participated (age: 62.3 ± 8.9 years, male: 68.18%). The mean MoCA score was 26.23 ± 2.90. A 1-point MoCA increase was independently associated with faster gait velocity (cm/s) [single-task: 5.45 (95%CI: 2.35-8.54, AdjR² =0.46), dual-task: 5.04 (95%CI: 1.33-8.75, AdjR² =0.20) and a quicker L Test (s) [single-task: - 4.75 (95%CI: 7.22-2.28, AdjR² =0.45), dual-task: - 5.27 (95%CI: 8.74-1.80, AdjR² =0.38)]. A 1-second TMT-B increase was also independently associated with worse L Test performance [single-task: 0.21 s (95%CI: 0.03-0.39, AdjR² =0.20), dual-task: 0.29 s (95%CI: 0.06-0.51, AdjR² =0.30)]. No association was observed between MoCA or TMT-B on the FSST (p > 0.13).

Significance: Better global cognitive function and executive function were independently associated with faster gait velocity and improved functional mobility, but not dynamic balance. The present study demonstrates a unique relationship between cognition and physical function that warrants further research on the cognitive demands among clinical tests of physical function in PLLA.