Cerebral Ketones Detected by 3T MR Spectroscopy in Patients with High-Grade Glioma on an Atkins-Based Diet.

BACKGROUND AND PURPOSE: Ketogenic diets are being explored as a possible treatment for several neurological diseases, but the physiologic impact on the brain is unknown. The objective of this study was to evaluate the feasibility of 3T MR spectroscopy to monitor brain ketone levels in patients with high-grade gliomas who were on a ketogenic diet (a modified Atkins diet) for 8 weeks. MATERIALS AND METHODS: Paired pre- and post-ketogenic diet MR spectroscopy data from both the lesion and contralateral hemisphere were analyzed using LCModel software in 10 patients. RESULTS: At baseline, the ketone bodies acetone and ß-hydroxybutyrate were nearly undetectable, but by week 8, they increased in the lesion for both acetone (0.06 ± 0.03 ≥ 0.27 ± 0.06 IU, P = .005) and ß-hydroxybutyrate (0.07 ± 0.07 ≥ 0.79 ± 0.32 IU, P = .046). In the contralateral brain, acetone was also significantly increased (0.041 ± 0.01 ≥ 0.16 ± 0.04 IU, P = .004), but not ß-hydroxybutyrate. Acetone was detected in 9/10 patients at week 8, and ß-hydroxybutyrate, in 5/10. Acetone concentrations in the contralateral brain correlated strongly with higher urine ketones (r = 0.87, P = .001) and lower fasting glucose (r = -0.67, P = .03). Acetoacetate was largely undetectable. Small-but-statistically significant decreases in NAA were also observed in the contralateral hemisphere at 8 weeks. CONCLUSIONS: This study suggests that 3T MR spectroscopy is feasible for detecting small cerebral metabolic changes associated with a ketogenic diet, provided that appropriate methodology is used.

Characterization of peripheral nerve sheath tumors with 3T proton MR spectroscopy.

BACKGROUND AND PURPOSE: The characterization of peripheral nerve sheath tumors is challenging. The purpose here was to investigate the diagnostic value of quantitative proton MR spectroscopy at 3T for the characterization of peripheral nerve sheath tumors as benign or malignant, compared with PET. MATERIALS AND METHODS: Twenty participants with 24 peripheral nerve sheath tumors underwent MR spectroscopy by use of a point-resolved sequence (TE, 135 ms). Six voxels were placed in 4 histologically proven malignant peripheral nerve sheath tumors and 22 voxels in 20 benign peripheral nerve sheath tumors (9 histologically proven, 11 with documented stability). The presence or absence of a trimethylamine signal was evaluated, the trimethylamine concentration estimated by use of phantom replacement methodology, and the trimethylamine fraction relative to Cr measured. MR spectroscopy results for benign and malignant peripheral nerve sheath tumors were compared by use of a Mann-Whitney test, and concordance or discordance with PET findings was recorded. RESULTS: In all malignant tumors and in 9 of 18 benign peripheral nerve sheath tumors, a trimethylamine peak was detected, offering the presence of trimethylamine as a sensitive (100%), but not specific (50%), marker of malignant disease. Trimethylamine concentrations (2.2 ± 2.8 vs 6.6 ± 5.8 institutional units; P < .049) and the trimethylamine fraction (27 ± 42 vs 88 ± 22%; P < .012) were lower in benign than malignant peripheral nerve sheath tumors. A trimethylamine fraction threshold of 50% resulted in 100% sensitivity (95% CI, 58.0%-100%) and 72.2% (95% CI, 59.5%-75%) specificity for distinguishing benign from malignant disease. MR spectroscopy and PET results were concordant in 12 of 16 cases, (2 false-positive results for MR spectroscopy and PET each). CONCLUSIONS: Quantitative measurement of trimethylamine concentration by use of MR spectroscopy is feasible in peripheral nerve sheath tumors and shows promise as a method for the differentiation of benign and malignant lesions. Trimethylamine presence within a peripheral nerve sheath tumor is a sensitive marker of malignant disease, but quantitative measurement of trimethylamine content is required to improve specificity.