Imaging in Breast Implant Rupture 

In the early 20th century, many cosmetic breast prosthetic techniques and materials were used for implants; these included polyvinyl alcohol prostheses, paraffin injections, and injections of free silicone. All of these techniques and materials had notable drawbacks and for the most part were abandoned.

Silicone gel breast implants (SGBIs) have been marketed in the United States since 1962. (An intact SGBI is seen in the image below.) The US Food and Drug Administration (FDA) began regulating SGBIs in 1976 with the passage of the Medical Device Amendments. In 1992, after months of public and private meetings, the FDA restricted the use of SGBIs to specific instances of medical necessity, such as those involving patients who had undergone mastectomy. Saline breast implants have replaced SGBIs as the common breast prosthesis. Details of the FDA rulings and current status updates are available at the Breast Implants information section of the FDA Web site.^{[1, 2, 3, 4, 5]}

The ability to reliably evaluate SGBIs with imaging is important because the findings at clinical examination often are nonspecific. The incidence of implant rupture increases with time, and the long-term systemic effects of SGBIs, if any, remain unclear. The diagnosis of SGBI rupture is useful to clinicians and patients; it aids in surgical decision making and helps the patient gain peace of mind.

This article does not address imaging of the rupture of single-lumen saline implants (which usually is clinically obvious because extravasated saline is rapidly absorbed and breast volume quickly decreases at examination).

Preferred examination

The imaging examinations for SGBI rupture are the following:

• MRI^{[6, 7]}
• Ultrasonography^{[6, 7]}
• Computed tomography (CT) scanning
• Mammography^{[6, 7]}

Patients should undergo mammography per the American Cancer Society recommendations for breast cancer screening. When implant rupture and extravasation is detected at mammography, further imaging studies may not be required. However, screening mammography alone is insensitive and its findings often are nonspecific in the detection of SGBI rupture.^[8]

Limitations of techniques

MRI is the most accurate imaging examination for the evaluation of SGBI rupture.^[9] MRI's drawbacks include its cost and possible unavailability. Ultrasonography is fairly accurate and more available than MRI but is highly operator dependent and has a steep learning curve. CT scanning can show findings similar to those obtained with MRI, but the modality involves ionizing radiation, and it has not been systematically studied to the extent that MRI has been. Mammography is inexpensive. Its findings can be specific if free silicone is present, but it has very low sensitivity.^[10]

Findings at mammography include the following:

• Rupture with deflation and silicone extravasation (shown in the images below)Craniocaudal mammogram shows free extracapsular silicone around a ruptured silicone gel breast implant. Mediolateral oblique mammogram shows extracapsular silicone. Note the opaque axillary lymph node, which is consistent with nodal uptake of silicone.
• Measurable periprosthetic opaque band or rim of tissue
• Periprosthetic calcification
• Asymmetry in the size and/or shape of the implant
• Focal implant herniation

Silicone injections and SGBIs can limit the sensitivity of mammography. Implant-displaced views should be obtained, if possible. Comparison with prior images is strongly advised.

Degree of confidence

When rupture and extravasation are present, confidence is high. However, these findings are uncommon, and other signs are both insensitive and relatively nonspecific. However, when screening mammography performed for other reasons reveals incidental signs of rupture, the sensitivity is high.

With intracapsular ruptures, CT scans show findings analogous to the linguine sign at MRI, and CT can demonstrate the extracapsular spread of silicone. To our knowledge, no findings from large studies examining the effectiveness of CT have been published. Because CT scanning involves ionizing radiation and its multiplanar capabilities are limited, MRI typically is the examination of choice.

MRI can be used to exploit differences in silicone, water, and fat resonance frequencies to deliver high-resolution images of SGBIs.^{[9, 11, 12]} The approximate resonance frequency of silicone is 100 Hz lower than that of fat and 320 Hz lower than that of water. Because the resonance of silicone is similar to that of fat, silicone appears similar to fat on nonselective chemical fat- or water-suppressed images.^{[13, 14, 15]}

On short–inversion time inversion-recovery (STIR) images, the fat signal is suppressed and water and silicone are bright. Water-suppressed STIR sequences produce a silicone-only image. Chemical-shift imaging with the modified 3-point Dixon protocol provides both a silicone-only and a water-and-fat image with a single sequence. The relative appearances of silicone, fat, and water on various types of MRIs are summarized in the following table:

Table. Relative Appearances of Silicone, Fat, and Water on MRI of SGBI Rupture^* (Open Table in a new window)

While exact protocols differ among institutions, in general, orthogonal-plane imaging with several pulse sequences is performed. Image quality is maximized with the use of a dedicated phased-array breast coil and a high–field-strength magnet; however, in our experience, an adequate examination can also be performed with a low–field-strength, open-sided magnet.

Intact silicone breast implants are shown in the MRI scans below.

MRI findings in intact single-lumen implants

These include the following:

• Smooth, low–signal-intensity silicone membrane shell
• Low–signal-intensity radial folds in the shell, shown in the image below (may be complex; always abut the implant at its periphery and span the gel substance only at periphery) Magnetic resonance image shows a normal radial fold, which appears as a low–signal-intensity line through the implant.
• A few internal water droplet signals (common; not a reliable indication of rupture)
• Reactive fluid around textured implants (common; not indicative of rupture)
• Fibrous capsule (dark, ringlike structure around the implant)

MRI findings in double-lumen implants

Gradual deflation of the saline chamber over time results in complex fold patterns and occasional, nonspecific findings of fluid around the outer capsule of the implant.

MRI findings in intracapsular rupture

The linguine sign refers to a collapsed and folded elastomer shell that is floating in gel. This is the most reliable sign of intracapsular rupture.

The keyhole (ie, teardrop, inverted teardrop, noose) sign refers to the presence of silicone both inside and outside a radial fold. These signs are shown in the images below.

MRI findings in extracapsular rupture

Macroscopic extrusion of silicone through the fibrous capsule into the surrounding parenchyma, pectoralis muscle, or lymph nodes is present.^[7] Findings in intracapsular rupture should be expected. Extracapsular silicone is seen in the image below.

MRI findings in internal rupture of double-lumen implants

Failure of the inner shell may be depicted as saline droplets that are floating in the silicone gel; this is considered a form of intracapsular rupture. The presence of some saline droplets is a normal finding in single-lumen implants.

MRI findings in capsular contracture

Asymmetrical, serrated, focal folding of the fibrous capsule that changes the normal ovoid appearance of the implant may be present.

A transverse diameter of less than twice the anteroposterior depth corresponds well to clinically evident contracture.

Degree of confidence

MRI has a reported a sensitivity of 76-95% and a specificity of 93-97% in the detection of rupture.

False positives/negatives

Extensive gel bleeding can have the same findings as those of an intracapsular rupture, in which the keyhole sign is present and the linguine sign is absent. Some authors believe that a tiny tear is present and that it can be found with diligent examination at surgery. Other authors maintain that microscopic gel bleeding alone is the cause. Radial folds are normal invaginations of the silastic membrane, and they should not be confused with the keyhole sign. Silicone should not be present both inside and outside a radial fold.

Typically, a high-frequency (eg, 10-MHz) transducer is used.

Ultrasonographic findings in an intact implant

These include the following:

• Smooth, thin, linear membrane
• Sonolucent anechoic interior (most reliable but insensitive finding)
• Reverberation artifact from the proximal membrane wall
• Radial folds possible
• Linear internal echoes

With regard to radial folds, parts of the implant envelope may be connected by scanning adjacent portions of the implant.

Linear internal echoes do not indicate rupture. The abundance of the linear echoes has been evaluated, and their effect in the prediction of rupture is not statistically significant.^[17]

Ultrasonographic findings in intracapsular rupture

These include the following:

• Stepladder sign
• Echogenic implant lumen
• Focal, prominent, irregular bulge in the implant
• Ill-defined or poorly visualized implant margin

The stepladder sign is identified as multiple, discontinuous, parallel, linear echoes in the lumen. It is the most reliable ultrasonographic finding in intracapsular rupture and is analogous to the linguine sign at MRI.

Ultrasonographic findings in extracapsular rupture

These include the following^[7] :

• Snowstorm, or hyperechoic, noise (shown in the images below)Longitudinal ultrasonogram shows the snowstorm, or hyperechoic, noise associated with the capsule in proven extracapsular silicone gel breast implant rupture. Usually, the interior of an intact implant is sonolucent. Longitudinal ultrasonogram shows the snowstorm appearance of the soft tissues of the breast, which is consistent with extracapsular silicone gel breast implant rupture and free parenchymal silicone.
• Silicone masses (granulomas) - The same hyperechoic noise pattern for snowstorm noise (described below); possibly hyperechoic or hypoechoic and similar to other breast lesions (including carcinoma) or nearly sonolucent and similar to breast cysts (hyperechoic noise is often near or around cystic-appearing masses)
• Snowstorm pattern in involved axillary lymph node

Snowstorm noise refers to intense, homogeneous echogenicity with loss of the posterior detail of free silicone. (Silicone can replace portions of the implant envelope, be extruded into the breast parenchyma, forming nodules, and migrate to lymph nodes.)

Indeterminate findings

These include the following:

• Coarse echogenic aggregates
• Cobwebs
• Smooth focal bulge in the implant
• Peri-implant fluid collections

In one study, coarse echogenic aggregates occurred in 41% of surgically confirmed implant ruptures.^[18] They have also been found to occur in approximately the same percentage of implants that have been confirmed as intact at surgery.

Cobwebs are delicate, short, linear echoes that are diffusely scattered throughout the implant. (Most authors consider this finding to be a normal variant; they are also called "commas" in some reports.)

Degree of confidence

Sensitivity ranges from 47-74%; specificity, from 55-96%. An anechoic interior, although rare (4 of 64 implants in one series), is a strong indicator of an intact implant.

False positives/negatives

Ultrasonography is limited in the evaluation of the posterior implant wall, because of marked beam attenuation caused by silicone. Use of a low-frequency transducer (eg, 5 MHz) can help improve visualization of the posterior wall of the implant and the adjacent soft tissues. Reverberation artifacts from the near-side interface may limit visibility of the superficial implant margins. Prior silicone injections limit visualization of the implant.