Functional magnetic resonance imaging of pain-related brainstem nuclei in single subjects at 7 Tesla. Issue 3 (March 2016)

• Record Type: Journal Article
• Authors: Beissner, F.; Polimeni, J.; Kim, J.; Bianciardi, M.; Renvall, V.; Eichner, C.; Wald, L.; Napadow, V.
• Is Part Of: Clinical neurophysiology. Volume 127:Issue 3(2016:Mar.)
• Page Start: e3
• Publication Date: 2016-03
• Subjects: Neurophysiology -- Periodicals; Electroencephalography -- Periodicals; Electromyography -- Periodicals; Neurology -- Periodicals; 612.8
• DOI: 10.1016/j.clinph.2015.10.009
• Languages: English
• ISSNs: 1388-2457

Abstract

Abstract : Background: The function or dysfunction of pain-related nuclei in the brainstem has only been scarcely studied in humans due to the lack of non-invasive measurement methods. The performance of functional magnetic resonance imaging (fMRI), a standard non-invasive technique, is hampered by the close vicinity of the brainstem to large arteries and ventricles as well as its propensity to spatial distortions caused by the oral cavity. Furthermore, the small average size of brainstem nuclei necessitates higher spatial resolution and accuracy than in studies of the cortex. Here, we present a new approach based on ultra-high field fMRI acquisition at 7 Tesla and a brainstem-optimized analysis method (mICA), which we apply to study pain-related activity and connectivity of brainstem nuclei in single subjects. Methods: Following a multi-modal imaging approach based entirely on echo-planar imaging, we acquired distortion matched functional, T1-weighted structural as well as diffusion-weighted images at a resolution of 1.2 mm isotropic. Five 6 min runs of resting-state fMRI were acquired, during two of which subjects received a continuous pressure pain stimulus applied to their lower leg by an inflatable cuff. Results: Pain intensity was percept matched at 50/100 points on a visual analogue scale. Applying our recently developed mICA approach (masked independent component analysis), we were able to detect reproducible resting-state activity for specific brainstem nuclei, like the cuneiform nuclei, periaqueductal grey and as well as brainstem-cortex functional connectivity at the single-subject level. We identified a number of pain-related nuclei that showed distinctive activity changes during pain stimulation. Conclusions: Identification of nuclei was greatly aided by fractional anisotropy (FA) maps created from the diffusion data. Finally, assessing activity and functional connectivity of brainstem nuclei on the single-subject level may soon give us a deeper understanding of disease subtypes, individual differences in pain processing, as well as other functions localized in the brainstem.