The Diagnostic Capacity of Pre-treatment 18F-FDG PET/CT for Predicting the Extranodular Spread of Lymph Node Metastases in Patients with Oral Squamous Cell Carcinoma.

The aim of this study was to evaluate the ability of pretreatment 90-min 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) to predict the extranodular spread of lymph node metastases in oral squamous cell carcinoma. We retrospectively reviewed the cases of 56 patients who underwent pretreatment 18F-FDG PET/CT and surgery with neck dissection. Maximum standardized uptake value, metabolic tumor volume, and total lesion glycolysis were measured for the 56 primary sites and maximum standardized uptake value was measured for 115 lymph node levels. Extranodular spread was present at 9 lymph node levels in 7 patients. Significant differences were found in metabolic tumor volume and total lesion glycolysis of the primary site, and in lymph node maximum standardized uptake value, between patients with and without extranodular spread (p<0.05). Combining primary site total lesion glycolysis and lymph node maximum standardized uptake volume at their respective optimal cutoffs, the sensitivity, specificity, and accuracy for predicting extranodular spread were 89%, 92%, and 92%, respectively. Pretreatment 18F-FDG PET/CT is useful for predicting extranodular spread in patients with oral squamous cell carcinoma. The combined use of primary site total lesion glycolysis and lymph node maximum standardized uptake value showed greater predictive value than either predictor singly.

Basic study of susceptibility-weighted imaging at 1.5T.

With the aim of sequence optimization in susceptibility-weighted imaging (SWI), 2 image acquisition parameters (slice thickness and matrix size) and 2 image processing conditions (number of slices per minimum intensity projection (MIP) and Sliding Window) were investigated using a 1.5-T magnetic resonance imaging (MRI) system. The subjects were 12 healthy volunteers and the target region for scanning was the whole brain. Informed consent was obtained from all subjects. First, susceptibility-weighted images were acquired with various slice thicknesses from 1mm to 5mm and various matrix sizes from 256 x 256 to 512 x 512, and the images were assessed in terms of the contrast-to-noise ratio (CNR) and were also visually evaluated by three radiologists. Then, the number of slices per MIP and the usefulness of the Sliding Window were investigated. In the study of the optimal slice thickness and matrix size, the results of visual evaluation suggested that a slice thickness of 3mm and a matrix size of 448 x 448 are optimal, while the results of evaluation based on CNR were not significant. As regards the image processing conditions, the results suggested that the number of slices per MIP should be set to a minimum value of 2 and that the use of Sliding Window is effective. The present study provides useful reference data for optimizing SWI sequences.