Volumetric mapping of intra‐ and extracellular pH in the human brain using 31P MRSI at 7T. Issue 4 (1st April 2020)

• Record Type: Journal Article
• Authors: Korzowski, Andreas; Weinfurtner, Nina; Mueller, Sebastian; Breitling, Johannes; Goerke, Steffen; Schlemmer, Heinz‐Peter; Ladd, Mark E.; Paech, Daniel; Bachert, Peter
• Is Part Of: Magnetic resonance in medicine. Volume 84:Issue 4(2020)
• Page Start: 1707
• Page End: 1723
• Publication Date: 2020-04-01
• Subjects: 7T -- 31P MRSI -- extracellular pH -- high‐grade glioma -- intracellular pH -- pH imaging; Nuclear magnetic resonance -- Periodicals; Electron paramagnetic resonance -- Periodicals; 616.07548
• DOI: 10.1002/mrm.28255
• Languages: English
• ISSNs: 0740-3194

Abstract

Abstract : Purpose: In vivo 31 P MRSI enables noninvasive mapping of absolute pH values via the pH‐dependent chemical shifts of inorganic phosphates (Pi ). A particular challenge is the quantification of extracellular Pi with low SNR in vivo. The purpose of this study was to demonstrate feasibility of assessing both intra‐ and extracellular pH across the whole human brain via volumetric 31 P MRSI at 7T. Methods: 3D 31 P MRSI data sets of the brain were acquired from three healthy volunteers and three glioma patients. Low‐rank denoising was applied to enhance the SNR of 31 P MRSI data sets that enables detection of extracellular Pi at high spatial resolutions. A robust two‐compartment quantification model for intra‐ and extracellular Pi signals was implemented. Results: In particular low‐rank denoising enabled volumetric mapping of intra‐ and extracellular pH in the human brain with voxel sizes of 5.7 mL. The average intra‐ and extracellular pH measured in white matter of healthy volunteers were 7.00 ± 0.00 and 7.33 ± 0.03, respectively. In tumor tissue of glioma patients, both the average intra‐ and extracellular pH increased to 7.12 ± 0.01 and 7.44 ± 0.01, respectively, compared to normal appearing tissue. Conclusion: Mapping of pH values via 31 P MRSI at 7T using the proposed two‐compartment quantification model improves reliability of pH values obtained in vivo, and has the potential to provide novel insights into the pH heterogeneity of various tissues.