Multiparametric Ultrasound Evaluation of the Thyroid: Elastography as a Key Tool in the Risk Prediction of Undetermined Nodules (Bethesda III and IV)-Histopathological Correlation.

The purpose of this study was to select thyroid nodules most at risk for malignancy among those cytologically undetermined by fine-needle aspiration biopsy (FNAB), using B-mode, color duplex Doppler and 2-D shear wave shear wave elastography (2-D-SWE). This was a prospective diagnostic accuracy study with 62 Bethesda III/IV nodules according to FNAB (atypia/follicular lesion of undetermined significance or follicular neoplasia/Hürthle cell neoplasm). Ultrasonography (US) data were compared with resection histologic results, revealing 35 of 62 benign nodules (56.4%) and 27 of 62 carcinomas (43.6%). Conventional US was uses to evaluate nodule echogenicity, dimensions, contours, presence of halo and microcalcifications. Doppler US was used to assess the vascularization (exclusively or predominantly peripheral or central) and mean resistance index of three nodule arteries. Elastography was used to evaluate the nodule elastographic pattern; mean nodule deformation index; deformation ratio between nodule and adjacent thyroid parenchyma; and mean deformation ratio between nodule and pre-thyroid musculature (MDR). Statistical analysis included χ2, Fisher's exact, Student's t, Mann-Whitney tests and multivariable analysis by multiple logistic regression. Areas under the receiver operating characteristic curves (AUC-ROCs) were used for accuracy analysis. Fifty-eight participants (54.7 ± 14.0 y, 51 women) were studied. The parameters that were statistically significant to the univariate analysis were hypo-echogenicity, nodule diameter greater than width and all parameters analyzed from Doppler and elastography. Multivariate analysis revealed that the MDR (in kPa) was the best parameter for risk analysis of indeterminate nodules. Nodules with MDRs >1.53 exhibited a greater chance of malignancy (AUC-ROC = 0.98). We conclude that 2-D-SWE is able to select malignant nodules among those cytologically indeterminate, thus avoiding unnecessary surgery in these cytologic groups.

Suction versus slow-pull for endoscopic ultrasound-guided fine-needle aspiration of pancreatic tumors: a prospective randomized trial.

BACKGROUND: Suction (S) is commonly used to improve cell acquisition during endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA). Slow-pull (SP) sampling is another technique that might procure good quality specimens with less bloodiness. We aimed to determine if SP improves the diagnostic yield of EUS-FNA of pancreatic masses. METHODS: Patients with pancreatic solid masses were randomized to four needle passes with both techniques in an alternate fashion. Sensitivity, specificity, positive, and negative predictive values were calculated. Cellularity and bloodiness of cytological samples were assessed and compared according to the technique. RESULTS: Sensitivity, specificity, and accuracy of suction vs. SP were 95.2% vs. 92.3%; 100% vs. 100; 95.7% vs. 93%, respectively. As to the association of methods, they were 95.6, 100 and 96%, respectively. Positive predictive values for S and SP were 100%. There was no difference in diagnostic yield between S and SP (p = 0.344). Cellularity of samples obtained with SP and Suction were equivalent in both smear evaluation (p = 0.119) and cell-block (0.980). Bloodiness of SP and suction techniques were similar as well. CONCLUSIONS: S and SP techniques provide equivalent sensitivity, specificity, and accuracy. Association of methods seems to improve diagnostic yield. Suction does not increase the bloodiness of samples compared to slow-pull.