Postsurgical Breast Imaging

The accurate interpretation of images of the postsurgical breast depends on the availability of high-quality pictures and pertinent medical and surgical breast history. Many findings can be mistaken for cancer. Skin thickening, architectural distortion, and other indicators of malignancy can be seen in both malignant and benign conditions. Prior breast surgery, trauma, and breast conservation treatment (BCT), or lumpectomy, can result in scarring and distortion as seen on imaging studies. These findings can be mistakenly interpreted as suspicious for malignancy. Thus, understanding the expected postsurgical imaging findings is important to ensuring an accurate interpretation and recommendation. ^{[1, 2, 3, 4]}

The majority of recurring cancers in reconstructed breasts after mastectomy are in the skin and subcutaneous tissues, with recurrence rates being about 1-2% annually and 2-15% overall. ^[2]

(See the image below.)

The pathophysiology of postsurgical changes, as observed on mammograms, is associated with the type of surgical intervention and the time elapsed since the procedure. The 4 most common breast interventional procedures are percutaneous biopsy; excisional breast biopsy; breast conservation treatment (BCT); and breast reduction, augmentation, or reconstruction. Postsurgical mammographic findings are also related to the time sequence from the procedure and can be classified into 2 general categories: acute changes and chronic changes.

(See the image below.)

Acute mammographic changes refer to the immediate postoperative period extending for the first several weeks and months. Acute changes include hematoma, seroma, and edema. Chronic changes refer to findings identified after the acute period, usually several months to years after surgery. These include scar formation, retraction, development of dystrophic calcifications, tissue asymmetry (from tissue removal), fat necrosis, and architectural distortion.

(See the images below.)

Architectural distortion is the disturbance of the normal-appearing curvilinear crescentic planes of the breast. The mammogram may demonstrate a pulling in of the Cooper ligaments to form a spiculated appearance. Architectural distortion may be the only mammographic indication of cancer. However, this finding is also seen after surgery, and it may be observed as a result of the superimposition of normal structures. Therefore, a thorough mammographic evaluation is needed to evaluate any areas of architectural distortion and to correlate the finding with the clinical history.

Percutaneous Breast Biopsy

Percutaneous breast biopsies commonly are performed on masses and calcifications and include fine-needle aspiration, core-needle biopsy, and cyst aspiration. ^[5]  These procedures involve the introduction of a needle into the suspicious lesion, usually under ultrasonographic or stereotactic guidance. A sample of tissue is removed and analyzed by the pathologist.

(See the image below.)

Mammographic findings immediately after percutaneous biopsy are usually related to bleeding and the local anesthesia injected into the biopsy area. These findings include increased density in the area, formation of a mass (hematoma), and trabecular thickening from edema. A clip may be left intentionally in the biopsy area to document the sampled lesion. In most women, the biopsy area heals with little or no residual evidence on the mammogram other than reduced or absent calcifications or the presence of a marking clip.

Excisional Breast Biopsy

Excisional breast biopsy is usually performed by the surgeon and involves a skin incision and removal of breast tissue. The amount of breast tissue removed and the degree of tissue interruption are variable and are dependent on the surgical technique and subsequent treatment, such as radiation (see the images below). Acute postsurgical changes are most prominent in the immediate postoperative period (as long as 1-2 wk) and are related to hematoma, edema, and tissue disruption. Mammographic images may show an ill-defined mass, area of increased density, skin thickening, and/or distortion.

Gradually, as healing occurs, the surgical site matures. Fibrosis may occur, leading to scarring. The mammogram may show a spiculated mass, area of architectural distortion, or development of dystrophic calcifications. Some patients may heal with little or no mammographic findings. Seldom, extensive scarring can occur, resulting in keloid formation (see the image below). Careful documentation on the history form and the placement of scar markers can help prevent misinterpretation of these lesions. In general, a baseline postsurgical mammogram is performed 6 months after biopsy. The findings on this mammogram are regarded as the new baseline.

Breast Conservation Treatment

BCT involves the removal of the breast cancer with a surrounding rim of noncancerous tissue. Some patients undergo complete lymph node dissection, whereas other patients undergo only a sentinel lymph node biopsy. Most patients undergo adjuvant radiation therapy to the breast to eradicate any possible residual occult cancer.

The expected changes on mammography after breast conservation surgery include skin thickening or edema, parenchymal edema, postoperative fluid collection, scarring, fat necrosis, and dystrophic calcifications, which are more marked up to 6 months after therapy. Recurrence on mammographic imaging may be observed as a mass or microcalcifications, increased skin thickening, increased breast density, scar enlargement, axillary nodal recurrence, or Paget disease. ^[6]

Acute mammographic changes observed after BCT are usually related to the extent of surgery and the time elapsed since radiation therapy. Mammograms performed during the acute surgical period usually demonstrate findings related to the surgery, such as skin and/or trabecular edema, seromas, architectural distortion, and surgical clips placed within the tumor bed. Patients may have early postoperative mammograms to document possible residual calcifications and nodules or masses.

Interpreting early mammogram findings may be confusing because masses and architectural distortion may be misinterpreted as residual cancer. Therefore, correlating mammographic findings with the surgical approach and the pathology report is important.

Additional imaging such as compression and/or magnification views and ultrasonography may be useful. A sonogram of a solid mass within the tumor bed would be worrisome for residual cancer in place of the expected finding of a complex fluid collection indicating a seroma or hematoma (see the image below). Differentiating a complex fluid collection from a solid mass is sometimes difficult sonographically. ^[7]

Chronic mammographic findings after BCT are related to the volume of tissue excised and to radiation therapy. Architectural distortion, a spiculated or ill-defined mass, and a change in an appearance of the breast are common findings identified after a lumpectomy. Follow-up mammographic evaluation may involve a 6-month series of mammographic examinations to assess the treated breast. However, some institutions may have different protocols.

Mediolateral oblique, craniocaudal, and mediolateral mammographic images are frequently supplemented by magnification and compression views. In general, radiation and postsurgical changes are noted to be most pronounced immediately after surgery and radiation, with maximum radiation changes at 6-12 months. On subsequent imaging, the areas of distortion and tissue edema should regress or remain stable. Therefore, any developing density or mass or calcifications need to be regarded as suspicious for recurrent cancer and thoroughly evaluated. Tissue sampling of any suspicious or indeterminate finding should be performed.

Recurrences may present at clinical examination or may be detected only on mammography as suspicious microcalcifications or masses. The rate of local recurrence after breast cancer surgery is 1-2% per year. Stability is defined as no interval change on 2 successive mammographic studies and is generally observed at 2-3 years after the completion of radiation therapy. Any retrograde change in imaging findings such as a new mass, microcalcifications, architectural distortion, or an area of increased density at the scar site after stability has been established should raise suspicion for tumor recurrence. ^[6]

(See the images below.)

The development of calcifications after BCT is problematic because one third to one half of irradiated breasts develop calcifications. Most of the calcifications can be attributed to fat necrosis resulting from the surgery and/or radiation. Suture calcifications may also be seen, especially if catgut sutures were used; however, these types of sutures are no longer used. Biopsy may be indicated when the calcifications appear suspicious or indeterminate.

Breast Reduction, Augmentation, or Reconstruction

Mammographic appearances of postsurgical changes after breast reduction, breast reconstruction, and breast augmentation commonly are encountered. A variety of surgical techniques are used in breast reduction surgery. One of the most common is the keyhole incision technique. In this procedure, an incision is made around the areola and extended vertically in the 6-o'clock position to the inferior mammary fold. Typical mammographic findings may include alteration of the parenchymal architecture, cranial displacement of the nipple, patchy densities due to tissue removal and scarring, and the development of fat necrosis. Approximately 6 months after surgery, a new baseline mammogram should be obtained. Any new findings from the baseline examination, such as a developing density, mass, or calcifications, require a thorough evaluation, including possible tissue sampling.

Breast reconstruction may be performed after a mastectomy by means of reconstruction with autogenous tissue transfer and/or implants. The most common autogenous tissue transfer site is from the panniculus or from a free myocutaneous flap. The most frequent location of the donor tissue is from a flap harvested from the latissimus dorsi muscle or the transverse rectus abdominis muscle (TRAM) flap.

Mammographic imaging of the reconstructed breast may be requested for the evaluation of a clinically suspicious finding, such as a palpable mass. Standard mammographic views are performed with additional views (compression, magnification, tangential) and ultrasonography if needed. In general, most of the mammographic and clinical findings are related to the development of dystrophic changes within the donor tissue, such as oil cysts and fat necrosis. Typically, dystrophic changes can be recognized easily on the mammogram as benign. However, fat necrosis, dystrophic microcalcifications, and scarring also can mimic cancer, thus prompting biopsy. Oil cysts are well-defined, are round or oval in shape, and contain fat. Oil cysts are identifed in about 20% of patients and can resolve or get smaller. ^[1]

(See the images below.)

The postsurgical mammographic observations identified after breast augmentation are related to the technical placement of the implant and the type of the implant. Standard and implant-displaced views are recommended. Assessment of the implants includes the location (subglandular or subpectoral), type (silicone, saline, mixed), contour (evaluation for possible rupture or weakening), and evaluation for possible complications (rupture, capsular formation). The evaluation of the native breast tissue may be obscured by the implant, thus hampering breast cancer detection. Rarely, breast tissue may be augmented by using native tissue harvested from the muscle or pedunculus. This produces an unusual mammographic appearance.

Reduction mammoplasty iremoves a significant portion of the breast, and changes to the breast parenchyma are often seen on breast images. Any radiologist must be able to recognize these changes to avoid unnecessary recall from screening or breast biopsy. ^[1]

In a retrospective review of 64 patients who underwent partial mastectomy with immediate oncoplastic reduction mammoplasty reconstruction, although substantial tissue rearrangement was performed, there were low rates of abnormal postoperative mammograms and subsequent biopsies during the first 2 years following the procedure. The most common cause of nonmalignant architectural distortion is postsurgical scarring. ^[8]

Implants in transgender women

Implant integrity in transgender women can be evaluated by mammography, ultrasound, and MRI, with imaging features being the same as those in cisgender women with implants. Silicone or dual-lumen implants are best evaluated by MRI or ultrasonography. ^[9]  

The sensitivity is very low for mammography and ultrasonography for breast cancer screening in transgender women who have had silicone injections. In such cases, contrast-enhanced MRI with silicone suppression should be employed as the first-line imaging modality. ^[9] ​ 

Imaging Modalities

The workup of postsurgical breast changes includes a thorough mammographic evaluation and correlation with the clinical history and pathologic findings. Scar markers are helpful to document the surgical approach. Additional mammographic projections (tangential, compression, and/or magnification views), comparison with prior mammograms, clinical breast examination, and ultrasonography are helpful in the workup. Tissue sampling is recommended for any suspicious or indeterminate findings. Breast MRI may also be helpful in the evaluation of postsurgical changes. ^{[10, 11, 12]}

The development of pleomorphic microcalcifications, especially in a branching pattern, is regarded as highly suggestive of new or recurrent cancer. Biopsy should be performed promptly. The development of oil cysts or fat necrosis is common, and routine follow-up can be performed. Architectural distortion is a common mammographic finding after breast conservation surgery and should be closely monitored after an appropriate workup is performed (with magnification and/or compression views). Any developing or changing area of architectural distortion should be viewed as worrisome. Skin and trabecular thickening are frequently observed after radiation therapy.

Mammographic and ultrasonographic findings of malignant and benign lesions overlap. A thorough workup is needed for any new or changing mass, developing calcifications, and developing areas of architectural distortion or density. Biopsy should be considered for any indeterminate or suspicious findings.

Scar markers are helpful to document the surgical approach. Additional mammographic projections (tangential, compression, and/or magnification views), comparison with prior mammograms, clinical breast examination, and ultrasonography are helpful in the workup. Tissue sampling is recommended for any suspicious or indeterminate findings. ^{[13, 10]}  Scintimammography is best used in clinical scenarios where mammography and ultrasound are inconclusive. ^[14]

Digital breast tomosynthesis (DBT) entails imaging of the breast tissue in multiple sections (at varied angles) instead of a 2-dimensional image, as with conventional mammography. DBT helps in triangulation of a lesion and can reduce the requirement for additional views. ^[6]

ACR recommendations

The American College of Radiology (ACR) has made the following recommendations after mastectomy and breast reconstruction ^[2] :

• Mammography or digital breast tomosynthesis for breast cancer screening may be appropriate for females with a history of cancer and autologous reconstruction on breasts with or without implants.
• Imaging for breast cancer screening is usually not appropriate for females with a history of cancer and no reconstruction on breasts that underwent mastectomy.
• Ultrasound of the breast as initial imaging is usually appropriate for females with a palpable lump or clinically significant pain on the side of the mastectomy without reconstruction or with reconstruction (autologous or nonautologous).

The imaging appearance of postsurgical changes can be confusing because masses, calcifications, and architectural distortion can mimic cancer. In addition, cancers can develop in areas of prior surgery. Therefore, a thorough clinical history, including information regarding the type of surgical intervention, and pathologic correlation are needed for adequate assessment. In general, indeterminate or suspicious findings may require tissue sampling to exclude malignancy.

Acute mammographic changes refer to the immediate postoperative period extending for the first several weeks and months. Acute changes include hematoma, seroma, and edema. Chronic changes refer to findings identified after the acute period, usually several months to years after surgery. These include scar formation, retraction, development of dystrophic calcifications, tissue asymmetry (from tissue removal), fat necrosis, and architectural distortion.

The pathophysiology of postsurgical changes, as observed on mammograms, is associated with the type of surgical intervention and the time elapsed since the procedure. The 4 most common breast interventional procedures are percutaneous biopsy; excisional breast biopsy; breast conservation treatment (BCT); and breast reduction, augmentation, or reconstruction. Postsurgical mammographic findings are also related to the time sequence from the procedure and can be classified into 2 general categories: acute changes and chronic changes.

Noroozian et al examined how often screening mammography after prophylactic mastectomy depicts clinically occult malignancy following breast reconstruction with autologous myocutaneous flaps, and they found that the cancer detection rate is comparable to that of just one native breast of an age-matched woman. They therefore determined that such screening mammography adds little value after prophylactic mastectomy. ^[3]

(See the mammographic images below.)

 

Early postoperative MRI is affected by the strong enhancement of resection margins in response to inflammatory postoperative reactions. Therefore, MRI is unable to exclude possible residual tumor until at least 12-18 months after completion of BCT. However, MRI is often used as part of the routine posttreatment follow-up, because it can discriminate between postsurgical tissue modifications and tumor relapse with a high negative predictive value and a sensitivity of 90-100% and a specificity of 89-92%. ^[13]

Acute changes seen on MRIs include the following ^[15] :

• Hemorrhage can have variable signal intensity on nonenhanced T1-weighted images.
• Early or intense enhancement, or both, may occur.
• Hematomas and seromas manifest as a cavity with variable signal intensity (low to high), depending on whether it is acute or chronic.
• Generally, hematomas and seromas have a thinner, smoother capsule, whereas residual or recurrent tumor is more nodular and irregular.
• Granulation tissue usually enhances moderately, and enhancement is delayed; however, fast enhancement may lead to false-positive diagnosis.

Chronic and later changes seen on MRIs include the following ^[16] :

• Between 9 and18 months, enhancement may or may not be present at a lumpectomy site. ^[17] Therefore, residual or recurrent disease is excluded if no enhancement is seen.

• A false-positive diagnosis may be made if enhancement is seen.

• The morphology of the enhancement (ie, smooth and thin walled vs irregular and thick walled) may help differentiate between the 2 conditions.

• Scar tissue should not be enhancing after 18 months. Therefore, tissue that enhances must be considered suspicious, and biopsy should be considered.

On MRI, fat necrosis (FN) has different presentations depending on the stage of the process. The most common MRI finding is a round or oval mass showing high signal intensity on T1-weighted non–fat-saturated images; the mass appears hyperintense on T2-weighted non–fat-saturated images and hypointense on fat-saturated images. T1-weighted fat-suppressed sequences are helpful to differentiate fat from blood, which also shows high signal intensity on T1-weighted images. Because fat-containing lesions are extremely uncommon in malignant conditions, the presence of fat is extremely useful in differentiating FN from a malignant lesion. ^[18]  The T1 signal from fat necrosis is typically isointense compared to other fats in the breast fat necrosis may, but fat necrosis can occasionally have a slightly darker T1 signal because of hemosiderin deposition or chronic inflammation. ^[1]

Overlap exists between benign and malignant disease. Suspicious findings may warrant tissue sampling. A false-positive diagnosis may be made if enhancement is seen.

Ultrasonography of the breast, in conjunction with mammography and clinical breast examination, can be extremely helpful in evaluating postsurgical changes in the breast. Ultrasonography is usually performed with a high-frequency linear-array transducer (10 MHz or greater). The examination is directed toward palpable or mammographic lesions. Information regarding the internal architecture, shape, and margins of a lesion are used to determine if a lesion is cystic or solid.

Sonographic findings subsequent to surgery may include skin thickening (scarring), architectural distortion of the parenchyma, and development of a mass and/or calcifications (scarring, fibrosis, fat necrosis). These findings can appear on the sonogram as hypoechoic or hyperechoic structures and/or masses with acoustic shadowing. They can mimic cancer, thus prompting a thorough workup to determine if tissue sampling is indicated to exclude malignancy. In the acute stage, hematomas may be ill defined or may present as a mass containing complex echoes with distal acoustic enhancement or shadowing. In time, the hematoma organizes and becomes more masslike, and the fluid becomes more hypoechoic and may even appear cystic.

(See the ultrasound images of the postsurgical breast below.)

Simple Cysts

Simple cysts have 4 criteria:

• They are usually well defined and round or ovoid.

• They have thin smooth walls.

• They contain no internal echoes.

• They have posterior acoustic enhancement (see the image below).

Diagnosing a simple cyst means that the lesion is benign and that no further testing is needed. However, if a lesion does not meet all 4 criteria for a simple cyst, it is classified as either a complex lesion or a solid lesion.

Complex Cyst

A complex cyst is a lesion that lacks one or more of the 4 criteria for a simple cyst. Complex cysts may contain internal echoes, limited posterior acoustic enhancement, wall irregularity, and/or an irregular shape. Further evaluation with cyst aspiration usually is recommended to exclude malignancy.

Solid Lesion

Solid lesions can display several characteristics on sonograms. Assessing if a lesion is a complex cyst or solid lesion may be difficult. Certain characteristics are helpful in assessing a solid lesion for malignant potential. Worrisome lesions may have irregular walls, contain internal echoes, be taller than they are wide, and/or exhibit broad acoustic shadowing. Further evaluation with tissue sampling is needed to exclude malignancy.

Acute Postoperative Period

In the acute postoperative period, sonographic findings include diffuse hemorrhage and/or edema, which is seen on sonograms as a region of increased or decreased echogenicity and/or an area of architectural distortion.

Scar Formation

Chronic or later (months to years) postsurgical ultrasonographic findings are usually related to scar formation (ie, retraction, dystrophic calcifications, and fat necrosis [lipophagic granuloma, oil cyst]). Scar formation can be seen as a discrete area of architectural distortion with hypoechoic structures, acoustic shadowing, and interruption of the normal parenchyma. Frequently, these findings originate at the scar and extend into the breast parenchyma. This appearance can mimic that of cancer.

Fat Necrosis and Dystrophic Calcifications

Fat necrosis and dystrophic calcifications can be seen on the sonogram as a hypoechoic or hyperechoic irregular mass with acoustic shadowing. Correlation with mammographic images, surgical history, pathology findings, and clinical breast examination are important for accurate assessment. Serial ultrasounds and/or mammograms or biopsy (for suspicious lesions) may be needed.

Nonspecific Findings

Ultrasonographic findings can be nonspecific, unless the lesion has the characteristics of a simple cyst. For example, benign pathology such as fat necrosis and scarring can mimic cancer on sonograms. Both can be hypoechoic or hyperechoic, and they can appear as irregular, hypoechoic masses with acoustic shadowing. Thus, because of the overlap between benign and malignant disease, a thorough workup of the abnormalities is needed. Indeterminate or suspicious findings may warrant tissue sampling.

Scintimammography is performed by using technetium-99m sestamibi (MIBI). ^[19] Research demonstrates that sestamibi is a lipophilic complex that can penetrate the cell membrane, preferentially in tumor cells. ^[20]  Scintimammography employs a wide range of instrumentation applications. Conventional planar scintimammography has been enhanced by single-photon emission computed tomography (SPECT) and hybrid SPECT/CT. Hybrid SPECT/CT adds clinical value by co-registering physiologic with anatomic data to assist nonpalpable lesion biopsies, radiotherapy planning, and treatment follow-up. ^[14]

Single-photon emission computed tomography improves resolution of planar-prone images. Initial data showed 83-97% sensitivity for palpable lesions and less than 50% sensitivity for lesions smaller than 1 cm. Sestamibi imaging is limited because tracer uptake is dependent on tumor size and tumor histology; other limitations are due to detector resolution.

(See the images below.)

Dedicated small field of view (FOV) breast-specific gamma imaging (BSGI) devices have also emerged as an alternative to contrast-enhanced MRI because of better depiction of location and size of lesions. A meta-analysis comparing diagnostic performance of BSGI to MRI reported the pooled sensitivities of BSGI and MRI to be 0.84 (95% CI, 0.79-0.88) and 0.89 (95% CI, 0.84-0.92), respectively, and the pooled specificities of BSGI and MRI were 0.82 (95% CI, 0.74-0.88) and 0.39 (95% CI, 0.30-0.49), respectively. ^[21]

Molecular breast imaging may also have a role in imaging-guided percutaneous biopsy. In one study, the sensitivity of molecular breast imaging was shown not to be impacted by high breast density, the presence of implants, or radiation scars from previous treatment. ^[22]