FDG-PET/CT Imaging Method
The patients were instructed to fast for six hours before the imaging. Blood glucose level was measured, and if it was <10 mmol/L, the study was performed. After an intravenous injection of approximately 350–450 MBq (5 MBq/kg) of 18F-FDG, the patients rested for 60 minutes in a quiet, dimly lit room. Images were acquired with a Gemini PET/CT scanner (Philips, USA).
After conducting a CT survey (120 kV, 30 mAs), the imaging was started applying a head and neck protocol. Body images were obtained first, followed by head and neck images. The body imaging was conducted from clavicle to mid-thigh, with arms raised above shoulder level. CT (120–140 kV, 50–60 mAs, a section width of 4 mm) was applied first followed by PET (8 cm bed position, 1.5 minutes per frame).
During the head and neck images, the patients’ arms were positioned down, and the head was stabilized with a head rack; CT (120 KeV, 50 mAs, a section width of 3 mm) was again applied first, followed by PET (8 cm bed position, 2.5 minutes per frame).
Focal uptake distinguishable from the background, which could not be considered physiologic, reactive, or inflammatory, was interpreted to be pathological uptake. No predetermined standardized uptake value (SUV) threshold was used in the analysis. A nuclear physician and a radiologist examined the PET/CT images together, and the nuclear physician made the final interpretation.
In this study, the results are based on the original interpretations. In the analysis, a scan with no pathological focal FDG-accumulation was considered negative even in the presence of residual anatomical changes in the CT portion of the imaging.
The patient characteristics of the 88 eligible patients are presented in Table 1. FDG-PET/CT was performed at a median of 13 weeks post (C)RT (range 10–18 weeks). The median follow-up time of patients with negative FDG-PET/CT and no proven recurrence was 26 months (range 5–55 months).
TABLE 1. Patient characteristics.
|Oral cavity||1 (1)|
FDG-PET/CT was negative for residual disease in 67 (76%) patients. In nine (13%) out of these 67 patients, residual disease or recurrent disease was confirmed during further follow-up. FDG-PET/CT was positive indicating residual disease in 21 (24%) patients. In 17 (81%) out of these 21 patients, residual disease was confirmed (Table 2).
Accordingly, the overall NPV, PPV, specificity, sensitivity, and accuracy of FDG-PET/CT in detecting residual disease were 87%, 81%, 94%, 65%, and 85%, respectively (Table 3). The corresponding figures specifically for the primary site and the neck nodes are presented in Tables 2 and 3.
TABLE 2. True and false positive and negative findings.
(n = 88)
(n = 88)
(n = 54)*
|Note: *Only patients diagnosed with N+ disease.|
TABLE 3. Ability of FDG-PET/CT to detect residual disease.
|OVERALL (%)||PRIMARY (%)||NECK (%)|
|Abbreviations: NPV, negative predictive value; PPV, positive predictive value.|
Distant metastases were detected in three (3%) patients who all had persistent disease also locoregionally.