Original Preoperative Localization Technique of Parathyroid Adenomas by 3-Dimensional Virtual Neck Exploration

Abstract

Objective. Preoperative imaging in primary hyperparathyroidism (PHPT) is essential for planning of parathyroidectomy-particularly for selection of a minimally invasive approach. The objective of this cohort study was to evaluate the diagnostic precision of 3D virtual neck exploration (3D-VNE), to evaluate its impact on choice of surgical approach, and to document the correlation with long-term outcomes. Methods. 235 consecutive patients with PHPT were studied (January 2014 to December 2018), with 6-month follow-up. 220 patients had a preoperative computed tomography (CT), 172 of these had a 3D-VNE based on the CT, and 226 patients had a Tc-99m sestamibi scan. Results. Sensitivity of exact, per gland, adenoma localization was 57.09% (95% CI: 50.85-63.10%) for nonspecialized radiologist interpretation of CT scan, 58.17% (95% CI: 51.99-64.10%) for Tc-99m sestamibi scan, and 90.21% (95% CI: 85.21-93.64%) for 3D-VNE, and thereby favoring 3D-VNE compared to CT scan alone (OR 34.5, 95% CI: 9.19-290.56%, P < 2.2 × 10^-16) and to Tc-99m sestamibi scan (OR 16.25, 95% CI: 6.05-61.42%, P = 3.1 × 10^-15). Specificity was 87.38% for CT scan, 86.36% for 3D-VNE, and 90% for Tc-99m sestamibi scan (P > .05). The cure rate was 100%. The long-term recurrence rate (RR) was 2.978%. The RR was 1.324% in the video-assisted parathyroidectomy group of 151 patients and 5.952% in the group of 84 patients with cervicotomy (P = .0459). Conclusion. CT-based 3D-VNE proved to be the most accurate localizing study in PHPT and aided in selecting patients for targeted minimally invasive parathyroidectomy, which was associated with the lower recurrence rate. 3D-VNE could be proposed as a first-line imaging study in patients with PHPT.

Keywords: 3D virtual neck exploration; cohort study; diagnostic accuracy study; minimally invasive parathyroidectomy; primary hyperparathyroidism; technetium Tc-99m sestamibi; three-dimensional imaging.