Contralesional thalamic surface atrophy and functional disconnection 3 months after ischemic stroke

Nawaf Yassi, Charles B. Malpas, Bruce C.V. Campbell, Bradford Moffat, Christopher Steward, Mark W. Parsons, Patricia M. Desmond, Geoffrey A. Donnan, Stephen M. Davis, Andrew Bivard

Research output: Contribution to journalArticleResearchpeer-review

25 Citations (Scopus)

Abstract

Background: Remote structural and functional changes have been previously described after stroke and may have an impact on clinical outcome. We aimed to use multimodal MRI to investigate contralesional subcortical structural and functional changes 3 months after anterior circulation ischemic stroke. Methods: Fifteen patients with acute ischemic stroke had multimodal MRI imaging (including high resolution structural T1-MPRAGE and resting state fMRI) within 1 week of onset and at 1 and 3 months. Seven healthy controls of similar age group were also imaged at a single time point. Contralesional subcortical structural volume was assessed using an automated segmentation algorithm in FMRIB's Integrated Registration and Segmentation Tool (FIRST). Functional connectivity changes were assessed using the intrinsic connectivity contrast (ICC), which was calculated using the functional connectivity toolbox for correlated and anticorrelated networks (Conn). Results: Contralesional thalamic volume in the stroke patients was significantly reduced at 3 months compared to baseline (median change -2.1%, interquartile range [IQR] -3.4-0.4, p = 0.047), with the predominant areas demonstrating atrophy geometrically appearing to be the superior and inferior surface. The difference in volume between the contralesional thalamus at baseline (mean 6.41 ml, standard deviation [SD] 0.6 ml) and the mean volume of the 2 thalami in controls (mean 7.22 ml, SD 1.1 ml) was not statistically significant. The degree of longitudinal thalamic atrophy in patients was correlated with baseline stroke severity with more severe strokes being associated with a greater degree of atrophy (Spearman's rho -0.54, p = 0.037). There was no significant difference between baseline contralesional thalamic ICC in patients and control thalamic ICC. However, in patients, there was a significant linear reduction in the mean ICC of the contralesional thalamus over the imaging time points (p = 0.041), indicating reduced connectivity to the remainder of the brain. Conclusions: These findings highlight the importance of remote brain areas, such as the contralesional thalamus, in stroke recovery. Similar methods have the potential to be used in the prediction of stroke outcome or as imaging biomarkers of stroke recovery.

Original languageEnglish
Pages (from-to)232-241
Number of pages10
JournalCerebrovascular Diseases
Volume39
Issue number3-4
DOIs
Publication statusPublished - 6 May 2015
Externally publishedYes

Keywords

  • fMRI
  • Ischemic stroke
  • Magnetic resonance imaging
  • Stroke
  • Stroke recovery
  • Thalamus

Cite this