Mean Motor Disconnectivity Predicts Hand Function Impairment After Stroke: Beyond Motor Cortex Output.

Abstract

Persistent compromised hand function is one of the most common long-term deficits after stroke. It is related to dysfunction of the primary motor cortex (M1) and corticospinal tract (CST) as assessed by magnetic resonance imaging-derived estimates of CST lesion load or by transcranial magnetic stimulation-derived measures, such as motor evoked potential (MEP) status. However, substantial interindividual variability remains with these measures. We tested whether a novel measure, mean motor disconnectivity (MMD), explains additional variation in the hand function of subacute stroke patients. Thirty-two participants (15 M/17 F; age, 58.6±9.65 years) after unilateral ischemic stroke involving the CST and related upper extremity weakness were studied within 4 weeks of stroke in a cross-sectional study design at Emory University between 2015 and 2021. The primary outcome measure was skilled hand function assessed using the Jebsen-Taylor test. Gross hand function and proximal UE function were secondary outcome measures. Transcranial magnetic stimulation was used to measure MEP status. Lesion masks were hand-drawn on each participant's structural magnetic resonance imaging image and were used to calculate lesion volume, CST lesion load, and MMD across a 20-region bilateral motor network consisting of the precentral gyrus, postcentral gyrus, thalamus, caudate, putamen, and brainstem regions. MMD is the mean of the disconnection values across the 20×20 nodes of this network of the adult human connectome; disconnection values quantify how severely a participant's lesion disrupts structural connectivity between region pairs. Univariable analysis indicated MEP status (MEP+/-), CST lesion load, and MMD were significantly correlated with preserved hand function; lesion volume was not. Adding MMD to a model containing MEP status significantly improved the fit of a multivariable model. MMD explains additional variability in skilled hand function after accounting for MEP status, suggesting that disconnectivity measures capture stroke impact on motor areas beyond M1. URL: https://www.clinicaltrials.gov; Unique identifier: NCT02544503.