Eleni Konstantatou1, Cheng Fang1, Odyssefs Romanos1, Lorenzo E Derchi2, Michele Bertolotto3, Massimo Valentino4, Christina Kalogeropoulou5, Paul S Sidhu1. 1. Department of Radiology, King's College Hospital, Denmark Hill, London, United Kingdom. 2. Department of Health Sciences (DISSAL), University of Genoa, AND Emergency Radiology, Ospedale Policlinico San Martino, Genoa, Italy. 3. Department of Radiology, University of Trieste, Trieste, Italy. 4. Emergency Department, S. Antonio Hospital, Tolmezzo, Italy. 5. Department of Radiology, University Hospital of Patras, Patras, Greece.
Abstract
OBJECTIVE: To investigate the role of strain elastography using calculated strain ratio and visual elastography score in differentiating nonneoplastic, benign, and malignant neoplastic intratesticular lesions. MATERIALS AND METHODS: The study was approved by the hospital review board as a retrospective review of 86 patients examined with gray scale, color Doppler ultrasonography and strain elastography (visual elastography score and strain ratio). Sensitivity, specificity, and positive and negative likelihood ratio of color Doppler and stain elastography were documented. Receiver operator characteristic curves assessed the diagnostic accuracy of strain elastography to discriminate nonneoplastic, benign, and malignant neoplasms. Histology or follow-up ultrasonography determined lesion character. RESULTS: Thirty-one of 86 (36.0%) intratesticular malignant neoplasms, 17 of 86 (19.8%) benign neoplasms, and 38 of 86 (44.2%) nonneoplastic lesions were confirmed with histology (n = 52) or follow-up sonography (n = 34); 89.5% of intratesticular lesions were heterogeneous or hypoechoic on gray scale, with no difference between benign and malignant. Sensitivity, specificity, positive and negative likelihood ratio for nonneoplasm versus neoplasm were documented: color Doppler: 68.8%, 97.4%, 26.5, 0.32; visual elastography score: 81.3%, 57.9%, 1.93, 0.32; strain ratio: 68.8%, 81.6%, 3.73, 0.38. Neoplastic lesions showed a higher strain ratio than nonneoplastic lesions (P < .001), with strong correlation between median strain ratio and visual elastography score (Spearman's coefficient, 0.693; P < .001). Strain ratio is a significantly better assessment than visual elastography score for malignant lesions (P = .025). Logistic regression analysis revealed significant associations between size (P = .001), hypervascularity (P < .001), and malignancy. CONCLUSION: Higher strain ratio and visual elastography score are associated with neoplastic lesions and offer an alternative to assess tissue characteristics but do not improve the diagnostic accuracy when compared with the color Doppler pattern.
OBJECTIVE: To investigate the role of strain elastography using calculated strain ratio and visual elastography score in differentiating nonneoplastic, benign, and malignant neoplastic intratesticular lesions. MATERIALS AND METHODS: The study was approved by the hospital review board as a retrospective review of 86 patients examined with gray scale, color Doppler ultrasonography and strain elastography (visual elastography score and strain ratio). Sensitivity, specificity, and positive and negative likelihood ratio of color Doppler and stain elastography were documented. Receiver operator characteristic curves assessed the diagnostic accuracy of strain elastography to discriminate nonneoplastic, benign, and malignant neoplasms. Histology or follow-up ultrasonography determined lesion character. RESULTS: Thirty-one of 86 (36.0%) intratesticular malignant neoplasms, 17 of 86 (19.8%) benign neoplasms, and 38 of 86 (44.2%) nonneoplastic lesions were confirmed with histology (n = 52) or follow-up sonography (n = 34); 89.5% of intratesticular lesions were heterogeneous or hypoechoic on gray scale, with no difference between benign and malignant. Sensitivity, specificity, positive and negative likelihood ratio for nonneoplasm versus neoplasm were documented: color Doppler: 68.8%, 97.4%, 26.5, 0.32; visual elastography score: 81.3%, 57.9%, 1.93, 0.32; strain ratio: 68.8%, 81.6%, 3.73, 0.38. Neoplastic lesions showed a higher strain ratio than nonneoplastic lesions (P < .001), with strong correlation between median strain ratio and visual elastography score (Spearman's coefficient, 0.693; P < .001). Strain ratio is a significantly better assessment than visual elastography score for malignant lesions (P = .025). Logistic regression analysis revealed significant associations between size (P = .001), hypervascularity (P < .001), and malignancy. CONCLUSION: Higher strain ratio and visual elastography score are associated with neoplastic lesions and offer an alternative to assess tissue characteristics but do not improve the diagnostic accuracy when compared with the color Doppler pattern.
Authors: Subramaniyan Ramanathan; Michele Bertolotto; Simon Freeman; Jane Belfield; Lorenzo E Derchi; Dean Y Huang; Francesco Lotti; Karolina Markiet; Olivera Nikolic; Parvati Ramchandani; Jonathan Richenberg; Laurence Rocher; Paul S Sidhu; Katarzyna Skrobisz; Athina Tsili; Pieter De Visschere; Irene Campo; Oliwia Kozak; Vikram Dogra Journal: Eur Radiol Date: 2021-01-15 Impact factor: 5.315