Coevolution of white matter hyperintensities and cognition in the elderly

Abstract

Objective: To investigate the effects of baseline white matter hyperintensity (WMH) and rates of WMH extension and emergence on rate of change in cognition (episodic memory and executive function).

Methods: A total of 150 individuals including cognitively normal elderly individuals and those with Alzheimer disease and mild cognitive impairment completed serial episodic memory and executive function evaluations and serial MRI scans sufficient for longitudinal measurement of WMH (mean delay 4.0 years). Incident WMH voxels were categorized as extended (baseline WMH that grew larger) or emergent (newly formed WMH). We used a stepwise regression approach to investigate the effects of baseline WMH and rates of WMH extension and emergence on rate of change in cognition (episodic memory and executive function).

Results: WMH burden significantly increased over time, and approximately 80% of incident WMH voxels represented extensions of existing lesions. Each 1 mL/y increase in WMH extension was associated with an additional 0.70 SD/y of subsequent episodic memory decrease (p = 0.0053) and an additional 0.55 SD/y of subsequent executive function decrease (p = 0.022). Emergent WMHs were not found to be associated with a change in cognitive measures.

Conclusions: Aging-associated WMHs evolve significantly over a 4-year period. Most of this evolution represents worsening injury to the already compromised surround of existing lesions. Increasing WMH was also significantly associated with declining episodic memory and executive function. This finding supports the view that white matter disease is an insidious and continuously evolving process whose progression has clinically relevant cognitive consequences.

Figures

Figure 1. Regression lines relating changes in episodic memory (red) and executive function measures (black) as a function of baseline white matter hyperintensity (WMH) volumes (A) and rates of WMH extension (B) and emergence (C), adjusting (plain lines) or not (dashed lines) by age

Figure 2. Spatial frequency of (A) baseline white matter hyperintensities (WMHs), (B) WMH extension, and (C) WMH emergence