Emerging Trends in Breast Ultrasound: Q&A

October 31, 2007

The live audience for Diagnostic Imaging’s Netseminar “Emerging Trends in Breast Ultrasound” had a number of questions for the faculty. Below, Dr. William Svensson responds to their queries:

The live audience for Diagnostic Imaging's Netseminar "Emerging Trends in Breast Ultrasound" had a number of questions for the faculty. Below, Dr. William Svensson responds to their queries:

Q: You claim that there is an improved diagnostic accuracy with strain imaging. What is the sensitivity and specificity of the procedure?

Dr. Svensson: Using an elasticity B-mode ratio of 0.75 as an indicator for cancer and intermedial lesions (lesions that have a malignant potential such as radial scars, phyloides tumours, papillomas, etc.) gives a sensitivity of 96% and specificity of 53% [Svensson WE, Amiras DG, Shousha S, et al. Elasticity imaging of 234 breast lesions shows that it could halve biopsy rates of benign lesions. Eur Radiol 2006;16 supp 1:213-214].

With updated software, the sensitivity is over 97%, with little change in specificity. This is based on the elasticity imaging alone. All the false negatives had mammographic or B-mode appearances, which had a high index of suspicion with a BI-RADS grading of 4 or 5. The results of our work to date suggest that all lesions with a BI-RADS 2 grading on ultrasound, which have a elasticity to B-mode ratio of less than 0.75, are benign and would not need biopsy. Larger numbers are needed before we can determine whether this can become a standard of practice.

Q: What is your comment on the reproducibility between operators in strain imaging?

We have shown there is good reproducibility between observers in strain imaging [Burnside ES, Hall TJ, Sommer AM, et al. Using ultrasound strain imaging to improve the decision to biopsy solid breast masses. Radiology 2007;245:401-410].

Our experience is that reproducibility between operators using the technique is similar to that obtained in other areas of ultrasound once the operators have gained sufficient skill in the technique. The overall published results in strain imaging are remarkably similar, which would support the hypothesis that reproducibility between operators is similar to other area of ultrasound.

Q: Are these techniques effective for all types of breast cancer?

Elasticity imaging is effective for all types of breast cancer. Even so-called soft cancers, such as mucinous cancers and ductal carcinoma in situ, exhibit increased stiffness compared with the surrounding tissues. The key to good interpretation of elasticity imaging is a good understanding of the physics behind it and the algorithms used in producing the elasticity images.

Differences in results between different methods of obtaining elasticity or strain images probably reflect the different algorithms used in the programming to produce the images.

Q: Does calcification cause any problem with interpretation of breast elastogram?

Microcalcifications do not cause problems with interpretation of breast elastograms. Macrocalcifications can result in interesting patterns and artifact, which are recognizable because of the very obvious B-mode changes associated with macrocalcifications.

Q: How does fat necrosis show up?

Fat necrosis usually shows more extensive B-mode abnormality than elasticity or strain abnormality. When cancers have an appearance that is similar to fat necrosis, our initial results suggest that this is reversed and the strain appearances of such cancers is larger than the B-mode abnormality.

Q: Does this finding make obsolete the need for cyst aspiration?

The very classical pattern of elasticity imaging seen with cysts may well make the need for cyst aspiration obsolete. Like all new imaging techniques, we need greater numbers to be certain that it can safely remove the need for aspiration in the case of very complex echogenic cysts.

Q: When ductal carcinoma is the diagnosis, is the biopsy/aspiration procedure different?

Solid lesions require core biopsy. Cysts call for fine-needle aspiration.

Q: Do you report the results of breast ultrasound examinations according to American College of Radiology BI-RADS Atlas standards?

In the U.K., breast ultrasound examinations are reported in a manner similar to that of the ACR BI-RADS Atlas standards. There are minor differences between BI-RADS II and BI-RADS III compared with ultrasound grade II and ultrasound grade III. BI-RADS advises ultrasound follow-up in cases that we would biopsy.

In the U.K., ultrasound grade III is biopsied at the time of first examination almost without exception. Also in the U.K., BI-RADS II lesions, which are solid, are nearly always core biopsied at first presentation and not followed up. The difference is not so much in the reporting as in the management.