Longitudinal Structural and Functional Differences Between Proportional and Poor Motor Recovery After Stroke.

This is an observational imaging study, not a treatment trial, so it cannot show that any rehabilitation method caused the differences. All patients received the same standard inpatient and outpatient physical and occupational therapy program, so the study cannot separate the contribution of any single therapy component. The sample came from one inpatient rehabilitation unit and excluded patients who recovered almost completely within the first two weeks. It was more severe and less representative than real-world settings. Assessments were at 2 to 4 weeks and 3 months, so some recovery may have occurred before the first measurement or after the follow-up. Fifty-five patients had baseline diffusion tensor imaging and 46 had follow-up imaging, so not every participant had complete structural data at both time points. Prediction results were evaluated mainly within the same dataset with leave-one-out cross-validation. External validation in other stroke populations is needed.

The authors declared no commercial interests relevant to the manuscript. The supplied text does not state a funding source.

Do not conclude that a specific rehabilitation treatment can convert poor recovery into proportional recovery. The study observed associations in patients who all received standard physical and occupational therapy, and no treatment was tested. The prediction results also came from a selected inpatient sample, so they should not be applied as a routine clinical test without external validation.