Differential diagnosis of breast lesions using ultrasound elastography Ioana Andreea Gheonea, Zoia Stoica, and Simona Bondari Department of Radiology and Imaging, University of Medicine and Pharmacy Craiova, Romania Correspondence: Dr. Ioana Andreea Gheonea, Petru Rares Street, No 2, Craiova – 200349, Romania. E-mail: iagheonea@gmail Copyright : © Indian Journal of Radiology and Imaging This is an open-access article distributed under the terms of the Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract Context: The recent introduction of elastography has increased the specificity of USG and enabled early diagnosis of breast cancer. Quantitative elastography, especially with strain ratio (SR) index, improves diagnostic accuracy and decreased number of biopsies. Aims: The purpose of this study was to assess the role of USG elastography in the differential diagnosis of breast lesions. Settings and Design: This prospective study was conducted in the University of Medicine and Pharmacy Research Centre of Craiova. Materials and Methods: Fifty-eight patients diagnosed with breast lesions between January 2009 and January 2010 were included in this prospective study. All the patients were examined in the supine position, and the B-mode USG image was displayed alongside the elastography strain image. For obtaining the elastography images we used a EUS Hitachi EUB 8500 ultrasound system with a 6.5-MHz linear probe. The elastography strain images were scored according to the Tsukuba elasticity score. Statistical Analysis: We performed receiver operator characteristic (ROC) analysis for assessment of the role of USG elastography in the diagnosis of breast lesions. Results: We obtained a sensitivity of 86.7% and a specificity of 92.9% for elasticity score and a sensitivity of 93.3% and a specificity of 92.9% for SR (when a cutoff point of 3.67 was used). There was very good correlation between SR and elasticity score (Spearman coefficient of 0.911). Conclusions: Elastography is a fast, simple method that can complement conventional USG examination. This method has the lowest cost/efficiency ratio and it is also the most noninvasive and accessible imaging method, with an accuracy comparable to MRI. Keywords: Breast lesions, receiver operator characteristic analysis, ultrasound elastography Introduction The high incidence of breast cancer and its slow evolution before diagnosis have led to research on new diagnostic techniques.[1–3] The recent introduction of elastography has increased the specificity of USG and enabled earlier diagnosis of breast cancer. The use of quantitative elastography with strain ratio (SR) improves diagnostic accuracy in cases with equivocal Stavros criteria (stages 3 and 4 BI-RADS). USG elastography (SE) differentiates between benign and malignant lesions on the basis of their elasticity: benign lesions have an elasticity similar to the surrounding tissue, while malignant lesions are harder than adjacent tissue.[4] The purpose of this study was to assess the role of SE in the diagnosis of breast lesions. Malignant tumors have reduced elasticity and also display larger dimensions on elastography due to the accompanying desmoplastic reaction.[5,6] Benign lesions appear similar to the adjacent tissues and have a smaller diameter than on B-mode USG images.[7,8] Materials and Methods In this prospective study, consecutive patients presenting with palpable breast lesions were assessed with conventional B-mode USG. Those confirmed to have a breast lesion were then assessed with SE after informed consent was obtained. SE was performed by a single experienced physician who was not aware of the results of previous investigations. The operator was not blinded at conventional USG because the lesion was localized first with conventional B-mode USG and then SE was done. The patients were examined in the supine position with the arm placed behind the head. The USG probe, lubricated with gel, was placed on the breast and a radial, ductal exploration was made as follows: the transducer was placed perpendicularly to the skin and radially on the breast, with one end overlapping on the areola and the other end directed toward the periphery. The orientation of the transducer was such that the nipple appeared at the left-hand side of the image. The transducer was then rotated around the areola. When a duct was identified, the rotation of the transducer was halted and it was moved back and forth laterally for thorough evaluation of the duct and its branches and the lobules. The transducer was then rotated again until the next ducto-lobular complex was found. This procedure was repeated until all the ductal structures were evaluated. A second rotating sweep was performed over the upper outer peripheral part of each breast. The B-mode US image was displayed alongside the elastography strain image to ensure that the assessment was made in the area of interest. We included in the area of interest the lesion and also the subcutaneous layers and the pectoralis muscle, without the costal cartilages. A EUS Hitachi EUB 8500 US system (Hitachi Medical, Tokyo, Japan) with an elastography module and a 6.5 MHz linear probe was used to obtain the B-mode and elastography strain images. The images were acquired in a ductal, radial manner as described above and the elastography strain images were scored according to the Tsukuba elasticity score developed by Itoh and Ueno.[4] Ductal USG combined with SE is defined as full-breast elastography, a new concept initiated by Amy D[9] We used five scores for characterizing the lesions: score 1 for lesions with elasticity similar to the surrounding breast tissue, displayed in green color on elastography; score 2 for lesions with inhomogeneous elasticity, with green and blue elastography appearance; score 3 for lesions with an elastic green periphery and a stiff blue centre; score 4 for nodules that were entirely stiff, showing no deformation; score 5 for cases where the whole lesion and the adjacent tissue showed a blue appearance on the elasticity image. For all lesions we calculated the strain ratio (SR). The average strain of the lesion was determined by selecting a region of interest (ROI) from the lesion and a corresponding ROI of the adjacent adipose tissue. Using specific software, the SR value was displayed on a static image as the ratio of tumor-adjusted ROI and the ROI placed in the adjacent fatty tissue. We used fine needle aspiration cytology (FNAC) (n=12) or excision biopsy (n=18) for histopathological analysis of the malignant lesions. The benign lesions were diagnosed by a combination of FNAC (n=10), excision biopsy (n=7), and follow-up for 6 months (n=11). Results In this study we included 58 patients with breast lesions confirmed on US. The average age of the women was 45.3 years. There were 28 (48.27%) benign and 30 (51.73%) malignant lesions. Among the benign nodules the common lesions were fibroadenoma, cyst, and fibrocystic change. Among the malignant nodules, the most common lesion was infiltrative ductal carcinoma. Ductal carcinoma in situ was diagnosed in 10 cases [Table 1]. Fibroadenomas appeared either softer than or had the same elasticity as adjacent glandular tissue [Figures [Figures11 and and2].2]. Breast cysts had an elasticity score of 1 with a characteristic three-layered appearance: blue-green-red (BGR), blue being the superficial color and red the deep one, even in large dimension lesions [Figure 3]. Fibrocystic nodules had elasticity similar to surrounding parenchyma [Figure 4]. The mean elasticity score for benign lesions was 1.92±1.01. Breast carcinomas showed an average elasticity score of 4.23±0.89; they appeared larger on the elastography image because of better visualization of the surrounding desmoplastic reaction [Figures [Figures55 and and6;6; Table 2]. After FNAC and excision biopsy, four lesions (14.28%) with elasticity score 3, one lesion (3.57%) with elasticity score 4 [Figure 7], and one lesion (3.57%) with elasticity score 5 were found to be benign; also, one lesion (3.33%) with elasticity score 1 and three lesions (10.72%) with elasticity score 3 [Figure 8] turned out to be malignant lesions. The average SR for benign lesions was 2.08, which was significantly lower than that for malignant lesions (mean SR: 6.28). To calculate the sensitivity and specificity of elastography, lesions with elasticity scores 1–3 were classified as benign, while those with scores of 4 or 5 were classified as malignant. For assessment of the role of SE in the differential diagnosis of breast lesions, we performed a receiver operator characteristic (ROC) analysis. We obtained a sensitivity of 86.7% and a specificity of 92.9% [Figure 9A] for elasticity score (area under the ROC curve=0.928; 95% CI=0.829 to 0.979; P=0.0001) and a sensitivity of 93.3% and a specificity of 92.9% for SR, when a cutoff point of 3.67 was used (area under the ROC curve=0.965, 95% CI=0.880 to 0.995; P=0.0001) [Figure 9B]. Furthermore, the Spearman coefficient of rank correlation for SR values and elasticity score was 0.911 (95% CI 0.853 to 0.946; P
Posted on: Tue, 02 Jul 2013 09:32:28 +0000
Trending Topics
Recently Viewed Topics
© 2015