Clinical factors associated with motor imagery ability in individuals with subacute stroke.
Yuki Fukumoto, Tsubasa Kawasaki, Takanao Shirai and 7 others
PMID 42605918WHAT IT FOUND
In 40 subacute stroke patients, motor recovery, proprioception, and working memory together were associated with motor imagery ability, but no single factor independently predicted it.
Cross-sectional design; associations only, not causation.
Key findings
01Motor imagery ability was significantly correlated with motor recovery stage (r = −0.316, p = 0.047) and proprioceptive function (r = −0.422, p = 0.007), but not with working memory (p = 0.282) or time since stroke onset (p = 0.983).
02When all three variables were entered simultaneously in a multiple regression, the overall model was significant (R² = 0.217, p = 0.030), but no single variable independently predicted motor imagery ability (BRS-H p = 0.402; SIAS-PS p = 0.060; BDS p = 0.278).
03In an exploratory classification analysis, combining motor recovery, proprioception, and working memory distinguished high from low motor imagery ability above chance (AUC = 0.716; permutation test p = 0.032).
STILL TO COME
How it was doneWhat they foundWhat it means for PTsWhat it means for OTs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
Cross-sectional design: associations only, no causal direction can be inferred between motor recovery, proprioception, working memory, and motor imagery ability. Patients with severe motor impairment (Brunnstrom ≤ 3) were excluded, so findings do not apply to the most affected patients. 34 of 74 enrolled patients were excluded (missing data, Brunnstrom ≤ 3, left-handedness), leaving only 40 for analysis; the authors note potential selection bias from missing-data exclusions. Exploratory subgroup analyses used z-score dichotomization, producing small and imbalanced groups (e.g., 11 vs 29 for SIAS-PS), limiting the stability of effect-size estimates. The SVM classifier showed substantial variability across folds (AUC SD = 0.297) and has not been externally validated. Motor recovery was assessed only by the Brunnstrom hand stage; comprehensive upper-extremity measures (Fugl-Meyer, ARAT) were not included. Sensory function and working memory were each captured by a single clinical item, which may not reflect the full multidimensional nature of these functions. F-wave analysis was limited to 4 participants and one facility; results are preliminary and hypothesis-generating only.
Declared interests
No conflicts of interest declared. Funded by a Grant-in-Aid for Scientific Research (B) from the Japan Society for the Promotion of Science (JSPS KAKENHI; grant number JP25K02985).
The easy way to misread this
Do not read the significant correlations with motor recovery and proprioception as meaning either one determines a patient's motor imagery capacity. When all three factors were modelled together, none independently predicted imagery ability, and the cross-sectional design means no causal direction can be inferred from these associations.