SLPOtherNeurorehabilitation and neural repair2024

Predicting Language Function Post-Stroke: A Model-Based Structural Connectivity Approach.

Franziska E Hildesheim, Anja Ophey, Anna Zumbansen and 5 others

PMID 38602161

WHAT IT FOUND

After stroke, models using routine brain scans, non-verbal cognition, initial language score and language-network connectivity predicted naming and comprehension; stimulation type did not predict language outcomes.

Key findings

01Non-verbal cognitive ability was the most predictive factor for language function at baseline and second-highest at follow-up.

02A higher initial language test score was the most critical predictor of better follow-up outcomes for TT, BNT and sVF.

03Connectivity disruption of language-relevant regions decreased prediction error by up to 12.5% at baseline and 12.8% at follow-up.

STILL TO COME

How it was doneWhat they foundWhat it means for SLPs

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.

Already have one?

What it does not show

The sample was small for machine learning, with 76 patients at baseline and 67 at follow-up. The authors did not analyse subacute and chronic patients separately because the sample was not large enough. Time since stroke varied from 2 days to 25 years, so recovery phases were mixed together. The brain atlas used has limited detail in some temporal regions. The non-verbal cognitive subscore may still depend on language for understanding instructions. Model results may be affected by image normalization and hand-drawn lesion masks. Validation was internal, using a held-out third of the same dataset.

Declared interests

The authors declared no potential conflicts of interest. Funding was provided by the Canadian Institutes of Health Research.

The easy way to misread this

Do not conclude that rTMS or tDCS does not help aphasia recovery. This was a secondary prediction analysis, not a test of treatment effect, and all patients also received 45 minutes of individualized speech therapy, so the contribution of any single component cannot be separated.

Read it on PubMed →