Jung Min Chang MD, Presenter: Nothing to Disclose

Woo Kyung Moon MD, Abstract Co-Author: Nothing to Disclose

Nariya Cho MD, Abstract Co-Author: Nothing to Disclose

Sanghee Park, Abstract Co-Author: Nothing to Disclose

Se Yeong Chung MD, Abstract Co-Author: Nothing to Disclose

To investigate the factors influencing the quality of sonoelastography for the evaluation of suspicious breast lesions.

Between January 2009 and February 2009, real-time sonoelastography of 312 breast masses (245 benign, 67 malignant) were obtained in 279 patients (mean age 46, range 16-74) with a EUB-8500 scanner (Hitachi Medical) prior to US-guided core biopsy. Five breast radiologists who performed the examinations prospectively assessed the quality of elasticity images on a 3 point scale (0=failure to obtain strain information, 1=valid strain information but high noise level, 2=valid strain information with low noise level), based on noise level and consistency of strain information. Diagnostic score of elasticity was also assigned on a 3 point scale (0=high stain, 1 =low strain, 2= no strain), based on the degree of strain in the hypoechoic lesion without histology information. The clinical and imaging findings of patients such as age, body mass index (BMI), mammographic density, breast thickness, lesion size, and lesion depth on US were analyzed for their association with the image quality using the χ2test, student t-test and Spearman rank correlation method. Sensitivities, specificities, PPV, and NPV for the classification of benign and malignant masses were calculated with a cutoff point between the diagnostic scores of 1 and 2 , and compared between the groups of different quality scores using the χ2test test.  

The quality of elasticity images was assessed as score 0 in 21 cases (6.7%), score 1 in 134, and score 2 in 157. Quality score of elasticity images was negatively correlated with BMI (r =-.145, p=.010), breast thickness (r=-.266, p<.001), lesion size (r =-.175, p=.002), and lesion depth (r=-.145, p=0.011). However, the age and mammographic density was not associated with  image quality. There were significant differences in sensitivities (87.0% vs. 59.5%, p=.041) and NPV (97.3% vs. 84.7%, p=.002) between the groups of quality score 1 and 2 for classification of benign and malignant masses.

Breast thickness, BMI, lesion depth, and lesion size was correlated with image quality of sonoelastography. Sensitivity and NPV for classification of benign and malignant masses improved with higher quality scores.

Sonoelastography seems to be favorable for the evaluation of small superficially-located lesions in women with smaller breasts and lower BMI.

Chang, J, Moon, W, Cho, N, Park, S, Chung, S, Factors Influencing the Quality of Sonoelastography for the Evaluation of Suspicious Breast Lesions.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8010349.html