Changes of migraine-related white matter hyperintensities after 3 years: a longitudinal MRI study

Objective/background: The aim of this longitudinal study was to investigate changes of migraine-related brain white matter hyperintensities 3 years after an initial study. Baseline quantitative magnetic resonance imaging (MRI) studies of migraine patients with hemispheric white matter hyperintensities performed in 2009 demonstrated signs of tissue damage within the hyperintensities. The hyperintensities appeared most frequently in the deep white matter of the frontal lobe with a similar average hyperintensity size in all hemispheric lobes. Since in this patient group the repeated migraine attacks were the only known risk factors for the development of white matter hyperintensities, the remeasurements of migraineurs after a 3-year long follow-up may show changes in the status of these structural abnormalities as the effects of the repeated headaches.

Methods: The same patient group was reinvestigated in 2012 using the same MRI scanner and acquisition protocol. MR measurements were performed on a 3.0-Tesla clinical MRI scanner. Beyond the routine T1-, T2-weighted, and fluid-attenuated inversion recovery imaging, diffusion and perfusion-weighted imaging, proton magnetic resonance spectroscopy, and T1 and T2 relaxation time measurements were also performed. Findings of the baseline and follow-up studies were compared with each other.

Results: The follow-up proton magnetic resonance spectroscopy studies of white matter hyperintensities showed significantly decreased N-acetyl-aspartate (median values 8.133 vs 7.153 mmol/L, P=.009) and creatine/phosphocreatine (median values 4.970 vs 4.641 mmol/L, P=.015) concentrations compared to the baseline, indicating a more severe axonal loss and glial hypocellularity with decreased intracellular energy production. The diffusion values, the T1 and T2 relaxation times, and the cerebral blood flow and volume measurements presented only mild changes between the studies. The number (median values 21 vs 25, P<.001) and volume (median values 0.896 vs 1.140 mL, P<.001) of hyperintensities were significantly higher in the follow-up study. No changes were found in the hemispheric and lobar distribution of hyperintensities. An increase in the hyperintensity size of preexisting lesions was much more common than a decrease (median values 14 vs 5, P=.004). A higher number of newly developed hyperintensities were detected than disappeared ones (130 vs 22), and most of them were small (<.034 mL). Small white matter hyperintensities in patients with a low migraine attack frequency had a higher chance to disappear than large white matter hyperintensities or white matter hyperintensities in patients with a high attack frequency (coefficient: -0.517, P=.034).

Conclusions: This longitudinal MRI study found clinically silent brain white matter hyperintensities to be predominantly progressive in nature. The absence of a control group precludes definitive conclusions about the nature of these changes or if their degree is beyond normal aging.