18F fluorocholine PET/MR imaging in patients with primary hyperparathyroidism and inconclusive conventional imaging: A prospective pilot study

Purpose: To investigate the performance of flourine 18 (18F) fluorocholine (FCH) positron emission tomography (PET)/magnetic resonance (MR) imaging in patients with hyperparathyroidism and nonlocalized disease who have negative or inconclusive results at ultrasonography (US) and technetium 99m (99mTc) sestamibi scintigraphy. Materials and Methods: This study was approved by the institutional review board. Between May and December 2015, 10 patients (mean age, 70.4 years; range, 58-82 years) with biochemical primary hyperparathyroidism and inconclusive results at US and 99mTc sestamibi scintigraphy were prospectively enrolled. All patients gave informed consent. Directly after administration of 3 MBq/kg of FCH, PET imaging was performed, followed by T1-and T2-weighted MR imaging before and after gadolinium enhancement. Intraoperative localization and histologic results were the reference standard for calculating sensitivity and positive predictive value. The Wilcoxon rank test was used to calculate the mean difference in maximum standardized uptake value (SUVmax) between abnormal parathyroid uptake and physiologic thyroid uptake. The Wilcoxon rank-sum test was performed. Results: MR imaging alone showed true-positive lesions in five patients and a false-positive lesion in one patient. FCH PET/MR imaging allowed correct localization of nine of 10 adenomas (90% sensitivity), without any false-positive results (100% positive predictive value). One patient had four-gland hyperplasia, of which three hyperplastic glands were not localized. The median SUVmax of the nine preoperatively identified adenomas was 4.9 (interquartile range, 2.45-7.35), which was significantly higher than the SUV, 2.7 (interquartile range, 1.6-3.8), of the thyroid (P = .008). Conclusion: FCH PET/MR imaging allowed localization of adenomas with high accuracy when conventional imaging results were inconclusive and provided detailed anatomic information. More patients must be examined to confirm our initial results, and the accuracy of FCH PET/MR imaging for localization of glands in patients with four-gland hyperplasia remains to be investigated.