Manuela Götzberger1, Susanne Krueger2, Roland Gärtner3, Martin Reincke3, Matthias Pichler3, Gerald Assmann4, Thomas Mussack5, Michael Zimmermann6, Veit Gülberg3. 1. Medizinische Klinik und Poliklinik IV, Klinikum der Universität München, Munchen, Germany. goetzbergerm@klinikum-freising.de. 2. Klinikum Bogenhausen, Klinikum der Universität München, Munchen, Germany. 3. Medizinische Klinik und Poliklinik IV, Klinikum der Universität München, Munchen, Germany. 4. Pathologisches Institut, Klinikum der Universität München, Munchen, Germany. 5. Klinik für Viszeral- und Endokrine Chirurgie - Campus Innenstadt, Klinikum der Universität München, Munchen, Germany. 6. Maria-Theresia-Klinik, Munich, Germany.
Abstract
OBJECTIVE: Strain-elastography provides a new ultrasound-based method that can offer information about the stiffness of thyroid nodules as an indicator of malignancy. The aim of our study was to compare the utility of color-Doppler and strain-elastography in differentiating between benign and malignant nodules. DESIGN AND METHODS: 77 thyroid nodules (70 benign and 7 malignant) from 70 unselected patients (48 female/22 male, mean age 49.7±14.3 years) were evaluated with color-Doppler and elastography based on a five-scale elastogram score for qualitative elastography and strain ratio for quantitative elastography. As reference tissue we chose normal thyroid tissue [strain ratio a (SR a)] and cervical muscles [strain ratio b (SR b)]. The cytological or histological results were used as a reference standard. Diagnostic performances of qualitative and quantitative elastography were compared using ROC curves. RESULTS: Vascularization score 3 or 4 was associated with malignancy (p=0.024) as well as elastogram score 4 or 5 (p=0.070, n.s.s.). SR a was indicatively higher and SR b lower in the group of malignant nodules (p=0.065 and p=0.246, n.s.s.). The best cut-off points predicting malignancy were 3.32 for SR a (66.7% sensitivity, 83.3% specificity) and 0.10 for SR b (71.4% sensitivity, 67.1% specificity). CONCLUSION: In our study, the accuracy of elastography did not surpass other sonographic parameters in differentiating thyroid nodules. The technique can play a role as a supplementary parameter in assessment of malignancy to improve diagnostic efficacy. The best parameter is SR a, but SR b can serve as an alternative if SR a is not assessable.
OBJECTIVE: Strain-elastography provides a new ultrasound-based method that can offer information about the stiffness of thyroid nodules as an indicator of malignancy. The aim of our study was to compare the utility of color-Doppler and strain-elastography in differentiating between benign and malignant nodules. DESIGN AND METHODS: 77 thyroid nodules (70 benign and 7 malignant) from 70 unselected patients (48 female/22 male, mean age 49.7±14.3 years) were evaluated with color-Doppler and elastography based on a five-scale elastogram score for qualitative elastography and strain ratio for quantitative elastography. As reference tissue we chose normal thyroid tissue [strain ratio a (SR a)] and cervical muscles [strain ratio b (SR b)]. The cytological or histological results were used as a reference standard. Diagnostic performances of qualitative and quantitative elastography were compared using ROC curves. RESULTS: Vascularization score 3 or 4 was associated with malignancy (p=0.024) as well as elastogram score 4 or 5 (p=0.070, n.s.s.). SR a was indicatively higher and SR b lower in the group of malignant nodules (p=0.065 and p=0.246, n.s.s.). The best cut-off points predicting malignancy were 3.32 for SR a (66.7% sensitivity, 83.3% specificity) and 0.10 for SR b (71.4% sensitivity, 67.1% specificity). CONCLUSION: In our study, the accuracy of elastography did not surpass other sonographic parameters in differentiating thyroid nodules. The technique can play a role as a supplementary parameter in assessment of malignancy to improve diagnostic efficacy. The best parameter is SR a, but SR b can serve as an alternative if SR a is not assessable.
Authors: Thomas Studeny; Wolfgang Kratzer; Julian Schmidberger; Tilmann Graeter; Thomas F E Barth; Andreas Hillenbrand Journal: BMC Med Imaging Date: 2021-10-30 Impact factor: 1.930