PLoS ONE (Jan 2016)

Incremental Value of a Dedicated Head and Neck Acquisition during 18F-FDG PET/CT in Patients with Differentiated Thyroid Cancer.

  • Renaud Ciappuccini,
  • Nicolas Aide,
  • David Blanchard,
  • Jean-Pierre Rame,
  • Dominique de Raucourt,
  • Jean-Jacques Michels,
  • Emmanuel Babin,
  • Stéphane Bardet

DOI
https://doi.org/10.1371/journal.pone.0162482
Journal volume & issue
Vol. 11, no. 9
p. e0162482

Abstract

Read online

OBJECTIVES:18F-FDG-PET/CT is a useful tool used to evidence persistent/recurrent disease (PRD) in patients with differentiated thyroid cancer and iodine-refractory lesions. The aim of this study was to compare the diagnostic value at the cervical level of the routine whole-body (WB) acquisition and that of a complementary head and neck (HN) acquisition, performed successively during the same PET/CT study. METHODS:PET/CT studies combining WB and HN acquisitions performed in 85 consecutive patients were retrospectively reviewed by two nuclear medicine physicians. 18F-FDG uptake in cervical lymph nodes (LN) or in the thyroid bed was assessed. Among the 85 patients, the PET/CT results of the 26 who subsequently underwent neck surgery were compared with surgical and pathological reports. The size of each largest nodal metastasis was assessed by a pathologist. RESULTS:In the 85 patients, inter-observer agreement was excellent for both WB and HN PET/CT interpretation. Of the 26 patients who underwent surgery, 25 had pathology proven PRD in the neck. Of these 25 patients, 15 displayed FDG uptake on either WB or HN PET. In these 15 patients, HN PET detected more malignant lesions than WB PET did (21/27 = 78% vs. 12/27 = 44%, P = 0.006). Node/background ratios were significantly higher on HN than on WB PET (P<0.0001). Three false-negative studies (20%) on WB PET were upstaged as true-positive on HN PET. The mean size of the largest LN metastasis was 3 mm for the LN detected neither on WB nor on HN PET, 7 mm for the metastasis detected on HN but not on WB PET, and 13 mm for those detected on both acquisitions (P = 0.0004). Receiver-Operating Characteristic analysis showed that area under the curve was higher for HN PET than for WB PET (0.97 [95%CI, 0.90-0.99] vs 0.88 [95%CI, 0.78-0.95], P = 0.009). CONCLUSIONS:HN acquisition improves the ability to detect PRD in the neck compared with WB acquisition alone. We recommend systematically adding an HN acquisition when PET/CT is performed to detect PRD in the neck.