Optimizing the Approach to Calculating CST Lesion Load in Ischemic Stroke Patients for Understanding Motor Outcomes

Abstract

Introduction

Corticospinal tract lesion load (CST-LL) is a biomarker used for studying motor outcomes after stroke. However, the optimal method for calculating this metric is unknown.

Methods

This is a cross-sectional study of a large ischemic stroke cohort from the ENIGMA Stroke Recovery Consortium (n = 221; mean age = 59.8 years, 56% male) to compare 4 lesion load metrics across 3 CST templates. We then validate these findings in another large, independent stroke cohort (n = 125; mean age = 64.6 years, 54% male).

Results

Results indicate that variance in behavioral outcome was best explained using the maximum weighted cross-sectional overlap between lesion and the CST (Max-WLL), and when using an age-appropriate normative CST template (generated from the HCP Aging study). This was true both when the outcome was motor impairment, measured using the Fugl-Meyer Upper Extremity scale (FMUE, relative explained variance (REV) = 58.9%), and when it was global function, measured using the Barthel Index (BI, REV = 60%). In the validation cohort, FMUE results were replicated (REV = 47.6%).

Conclusion

The findings indicate that Max-WLL, which represents the proportion of transected CST fibers, most accurately captures CST injury as it relates to motor and functional outcomes after stroke. Additionally, results suggest the importance of an age-appropriate template, a key consideration given that stroke is largely a disease of the elderly. Together, these findings provide an independently validated tool to optimize future research examining CST injury after stroke.