Differentiation Between Malignant And Benign Thyroid Nodules And Stratification Of Papillary Thyroid Cancer With Aggressive Histological Features Using The Reduced Field Of View DWI

Part I Clinical evaluation of reduced field-of-view diffusion imaging in thyroid nodules patients at 3TPurpose:To evaluate the diagnostic performance of reduced field-of-view (r-FOV) thyroid DWI in a clinical population qualitatively and quantitatively, and compare it with conventional full FOV (f-FOV) DWI.Materials and methods:This retrospective study was approved by the institutional review board. Seventy patients with 82 pathologically confirmed thyroid nodules were finally included. All subjects underwent r-FOV and conventional f-FOV DWI of thyroid at 3T MRI. Image quality and lesion identification were evaluated both with a 4-point scale and quantification of SNR, CNR and geometric distortion rate. The ADC value of thyroid nodules and normal thyroid tissues were calculated. The differences between methods and different tissue groups were compared, and receiver operating characteristic curve (ROC) analysis was used to evaluate the diagnostic performance in discriminating benign and malignant nodules.Results:The thyroid r-FOV DWI achieved significantly better image quality (p<0.001),superior lesion identification (p<0.001) than f-FOV DWI with good agreement between readers. Although r-FOV DWI image significantly sacrificed its SNR, it achieved significantly lower geometric distortion rate (15.1±12.6 vs. 26.7±14.2,p<0.001) than f-FOV DWI. Mean ADC values derived from r-FOV DWI were significantly lower than f-FOV DWI, while the area under the curve (AUC) for differentiating malignant and benign thyroid nodules based on ADC was comparable for both DWI methods.Conclusion:The r-FOV DWI was superior to conventional f-FOV DWI in providing better image quality, superior lesion identification and comparable diagnostic performance, which is a promising method for detecting and stratifying malignant lesions in the workup of thyroid nodules.Part Ⅱ Differentiation between malignant and benign thyroid nodules and stratification of papillary thyroid cancer with aggressive histological features: Whole-lesion DWI histogram analysisPurpose:To explore the usefulness of whole-lesion histogram analysis of apparent diffusion coefficient (ADC) derived from reduced field-of-view (rFOV) diffusion weighted imaging (DWI) in differentiating malignant and benign thyroid nodules and stratifying papillary thyroid cancer (PTC) with aggressive histological features.Materials and methods:This institutional review board approved, retrospective study, included 93 patients with 101 pathological proven thyroid nodules. All patients underwent preoperative rFOV DWI at 3T. The whole-lesion ADC assessments were performed for each patient. Histogram-derived ADC parameters between different subgroups (pathologic type, extrathyroidal extension, lymph node metastasis) were compared. Receiver operating characteristic curve analysis was used to determinate optimal histogram parameters in differentiating benign and malignant nodules and predicting aggressiveness of PTC.Results:Mean ADC, median ADC,5th percentile ADC,25th percentile ADC,75th percentile ADC,95th percentile ADC(all p<0.001) and kurtosis (p=0.001) were significantly lower in malignant thyroid nodules, and mean ADC achieved the highest AUC (0.919) with cutoff value of 1842.78×10-6 mm2/s in differentiating malignant and benign nodules. Compared to the PTCs without extrathyroidal extension (ETE), PTCs with extrathyroidal extension showed significantly lower median ADC,5th percentile ADC, and 25th percentile ADC. And the 5th percentile ADC achieved the highest AUC (0.757) with cutoff value of 911.5×10-6 mm2/s for differentiating between PTCs with ETE and without extrathyroidal extension. None of ADC histogram metrics achieved significant difference for lymph node metastasis.Conclusion:Whole-lesion ADC histogram analysis might help to differentiate malignant nodules from benign ones and show the PTCs with ETE.