The Postoperative Breast

General Postoperative Assessment Tips

Scars can have an alarming appearance on tomosynthesis—presenting as a spiculated mass or architectural distortion often extending for large distances in the breast—and either not seen at all or not as well on 2D images. These findings can mimic malignancy and will catch the eye when viewing the tomosynthesis images ( Fig. 12.1 ).

Any known breast scar or area of surgical procedure, no matter how remote, should be brought to the attention of the radiologist by the technologist performing the study. Although scar markers can cause distraction on both 2D and three-dimensional (3D) images and are not routinely necessary, the interpreting radiologist must nonetheless be made aware of the presence and location of all previous breast biopsies. A patient intake form or electronic medical record data entry system is typically used to note the location of each scar. In some cases in which concern persists, imaging can be repeated with dermal scar markers, particularly those composed of nonmetallic material designed specifically for tomosynthesis, on the appropriate area to permit correlation. Review of prior preoperative or needle localization images is the most direct method of correlating findings with prior surgical sites ( Fig. 12.2 ). This is particularly important for periareolar incisions, in which the surface scar may be barely perceptible and not provide information as to the actual surgical site within the deeper breast tissues. Sometimes getting an accurate history of remote surgery can be difficult, especially if the patient is elderly and the scar has faded. It is important in cases of a suspected remote scar to spend time trying to elicit the correct biopsy history, from the patient or even her physician, to avoid unnecessary work-up of a patient for a very remote benign biopsy.

Scars typically present as architectural distortion or a focal asymmetry, varying from mostly fatty to very dense ( Fig. 12.3 ). The radiating spicules can extend great distances in the breast, reaching from nipple to chest wall, depending on the type of surgery. Scars are often planar and more prominent on one mammographic projection than the other. When identified on one set of tomosynthesis projections and localized to a specific depth in the breast, a scar can usually be identified on the corresponding projection. Focal skin thickening and/or peripheral contour deformity may be present.

Postoperative changes are usually most pronounced on the first imaging exam performed generally 6 months to 1 year following most procedures. Some findings are common to many operative procedures, such as seromas and hematomas, which are found frequently in the early postoperative period. These are usually round or oval and circumscribed and are sometimes palpable. They resolve slowly over time in most patients, although some will become encapsulated and remain virtually unchanged for years ( Fig. 12.4 ). With resolution, the circumscribed mass decreases and is replaced by scar, which is typically more spiculated. These findings rarely present interpretation difficulties.

Fat necrosis is a common postoperative finding after any type of breast surgery because there is an abundant amount of fat within breast tissue. Fat necrosis often presents as rounded lucencies or oil cysts ( Fig. 12.5 ). These changes are usually solely noted on imaging but sometimes present as a palpable finding, particularly causing alarm in patients with a history of malignancy, in which case the differentiation from recurrent malignancy is paramount. Characteristic findings of fat necrosis include visualization of fat content within a mass. As discussed elsewhere, observation of fat within a lesion does not itself deem the finding benign. Many malignancies are shown by tomosynthesis to have fatty areas within them. Fat necrosis should ideally be encapsulated, fat-containing areas, often with thin, well-circumscribed margins, that may calcify over time. When such imaging findings are noted, the area can be confidently characterized as fat necrosis. However, in some instances, fat necrosis involving a postoperative scar can appear as an ill-defined or spiculated mass and mimic cancer. Biopsy of such lesions may be necessary ( Fig. 12.6 ).

Calcifications associated with scarring may develop over time. Dystrophic calcifications are common in women who have undergone postoperative radiation therapy and typically present as coarse, heterogeneous calcifications in the surgical bed. Although these findings are often typically benign, in the early phase they may be difficult to differentiate from malignant findings. Tomosynthesis is particularly useful because viewing the tissue in thin slices permits better assessment of calcifications, which can often be located to small focal areas of fat within a larger scar ( Fig. 12.7 ).

Infection is an early postoperative complication. Such cases are often managed clinically, but imaging may be requested to exclude abscess formation and/or monitor response to antibiotic treatment or drainage. These patients present with palpable, tender masses, and as in 2D imaging such masses have ill-defined borders, especially when within a surgical site. Ultrasound will be the imaging modality most appropriate for further assessment.

Retained foreign bodies are rare, but imaging can help to determine their presence and location. Likewise a missed lesion or clip following needle localization and excisional biopsy is also uncommon, but when these problems do occur, mammography is vital in documenting residual lesion or breast markers left behind at surgery ( Fig. 12.8 ).

Metallic markers, or clips, are routinely used to mark percutaneous biopsy sites, and those placed in benign lesions not requiring excision will be seen repeatedly on subsequent imaging. Rarely, the collagen or gel pellet that houses the metal component may not get fully resorbed and will also be seen. Likewise, surgical clips placed during lumpectomy or axillary lymph node excision will also be visualized. Because metal can produce a black streak artifact on tomosynthesis, imaging-processing algorithms should be used to reduce artifact, which can obscure or detract from fully assessing the underlying tissue ( Fig. 12.9 ).

Plastic surgery techniques, including reduction mammoplasty and mastopexy and different surgical techniques for lumpectomy and other oncoplastic surgeries, rearrange breast tissue to varying degrees. Concern has been raised over whether these surgeries may complicate subsequent mammographic interpretation and detection of malignancy. Several studies of 2D mammography have been performed that show the rates of imaging work-ups and biopsy recommendations in women who have undergone a variety of plastic surgery procedures, particularly reduction mammoplasty and oncoplastic surgery, are actually similar compared with those who have not undergone such surgery. Because tomosynthesis is more sensitive to postoperative findings, it is possible that an increase in architectural distortions due to scars may lead to an increase in imaging follow-ups or even biopsies. However, when used carefully and by recognizing the common appearances of scars, the information gained from tomosynthesis will hopefully improve diagnostic performance beyond 2D imaging alone.

Needle Biopsy

Core needle biopsies have been used for diagnosis of breast lesions for more than 25 years. There are very few sequelae after a benign core needle biopsy. Unless the biopsy was complicated by bleeding and hematoma formation, which may leave a residual finding for many years, most uncomplicated biopsies leave no imaging finding. Occasionally, very subtle areas of architectural distortion may remain at a prior needle biopsy site noted on tomosynthesis and not seen on 2D imaging. Correlation with prior imaging should be performed to exclude a new suspicious finding at or near a prior benign biopsy site, and in some cases, short-term follow-up may be necessary ( Fig. 12.10 ).

Surgical Biopsy

Surgical biopsies for benign or high-risk lesions may be performed as either excisional (removing the entire lesion) or incisional (representative sampling of a portion of the lesion) biopsies. In either case the goal is to remove adequate tissue while maintaining cosmesis. Because these nonmalignant cases do not require radiation therapy, the postsurgical effects seen on mammography are usually minimal. With 2D mammography, benign surgical scars are often not evident, and indeed it has been taught that one should regard a mammographically visible scar from remote benign biopsy with suspicion. This is certainly not the case with tomosynthesis imaging! The majority of postoperative scars will actually be seen even though they were previously occult on 2D mammography ( Fig. 12.11 ). This is one of the challenges of tomosynthesis because old scars can present as subtle architectural distortion and, because they cannot be seen on prior 2D imaging, assessing for stability when comparing with 2D imaging alone may be difficult. As discussed previously, information about prior surgery and location of scars is important when interpreting tomosynthesis images. Most cases will not present a problem in interpretation and will be confidently regarded as benign. Scars should stabilize or regress over time. If suspicious changes are noted on tomosynthesis in an area of a prior scar, further investigation including ultrasound will be necessary.