A Nitkunan, RA Charlton, TR Barrick, DJO McIntyre, FA Howe, HS Markus
BACKGROUND: Cerebral small vessel disease (SVD) is an important cause of cognitive impairment, but the pathophysiological mechanisms remain unclear. We used (1)H MRS to investigate brain metabolic differences between patients with SVD and controls and correlated this with cognition. METHODS: 35 patients with SVD (lacunar stroke and radiological evidence of confluent leukoaraiosis) and 35 controls underwent multi-voxel spectroscopic imaging of white matter to obtain absolute metabolite concentrations of N-acetylaspartate (NAA), total creatines, total cholines, myo-inositol, and lactate. A range of cognitive tests was performed on patients with SVD, and composite scores were calculated. RESULTS: Scans of sufficient quality for data analysis were available in 29 cases and 35 controls. NAA was significantly reduced in patients compared with controls (lower by 7.27%, P = 0.004). However, when lesion load within each individual voxel (mean 22% in SVD vs 5% in controls, P < 0.001) was added as a covariate, these differences were no longer significant, suggesting that the metabolite differences arose primarily from differences in lesioned tissue. In patients with SVD, there was no correlation between cognitive scores and any brain metabolite. No lactate, an indicator of anaerobic metabolism, was detected. CONCLUSIONS: The most consistent change in SVD is a reduction in NAA, a marker of neuronal integrity. The lack of correlation with cognition does not support the use of MRS as a surrogate disease marker.