Background: During the past decade, 18F-fluorocholine (FCH) PET/CT has been continuously performed at Tenon Hospital (Paris, France) for the detection of hyperfunctioning parathyroid glands (PT).
Methods: A cohort of 401 patients, deliberately referred for HPT since September 2012, has been analyzed. The aim of this real-life retrospective study was to determine the diagnostic utility of FCH in this setting, overall and in subgroups according to the type of hyperparathyroidism (HPT), the context of FCH in the imaging work-up and in the patient's history: initial imaging or persistence or recurrence after previous parathyroidectomy (PTX). The influence of the histologic type of resected PTs, hyperplasia or adenoma, on the preoperatory detection on FCH PET/CT has been studied as well.
Results: Four hundred one FCH PET/CTs were included in the cohort, performed in 323 patients with primary HPT (pHPT), including 18 with familial HPT (fHPT), and in 78 patients with secondary renal HPT (rHPT). The overall positivity rate in the 401 FCH PET/CTs was 73%. The PTX rate was twice greater in patients whose FCH PET/CT was positive than negative (73% vs. 35%). Abnormal PT(s) were pathology proven in 214 patients: only hyperplastic gland(s) in 75 cases and at least one adenoma in 136 cases; FCH PET/CT sensitivity was 89% and 92%, respectively. Similarly, there was no significant difference in patient-based sensitivity whether FCH PET/CT was performed as 1st line or later in the imaging work-up, or indicated for initial imaging or for suspicion of persistent or recurrent HPT. Gland-based sensitivity was significantly lower for hyperplasia than for adenoma (72% and 86%, respectively). The lowest gland-based sensitivity value was 65%, observed in case of hyperplasia and when FCH was performed late in the imaging work-up. FCH PET/CT correctly showed multiglandular HPT (MGD) in 36/61 proven cases, 59%. Results of ultrasonography (US) and 99mTc-sestaMIBI (MIBI) imaging were available in 346 and 178 patients, respectively. For both modalities, the corresponding sensitivity values were significantly less than those of FCH PET/CT (e.g., overall gland-based sensitivity 78% for FCH, 45% for US, 30% for MIBI) and MGD was detected in 32% of cases by US and 15% by MIBI.
Conclusions: Although FCH PET/CT has been performed since 2017 as 1st line imaging for HPT at Tenon Hospital (Paris, France), a large majority of patients underwent prior US and/or MIBI in their preoperative work-up. Therefore, a selection bias is very likely, as most patients referred to FCH PET/CT had non-conclusive or discordant results of US and MIBI, explaining the low performance of those modalities in the present cohort compared to published results. Nevertheless, the superiority of FCH PET/CT over US and MIBI in detecting abnormal PTs reported in various comparative studies is definitely confirmed in this larger real-life cohort. The detection with FCH PET/CT of hyperplastic PTs was somewhat lower than that of adenomas but was better than using US or MIBI. The present results lead to recommend FCH PET/CT as the first line imaging modality in HPT when it is widely available or, if less available, at least in HPT with predominance of hyperplasia and/or MGD.