Background:
There is some debate within the plastic surgical community as to whether the implant shapes the breast, or the breast, via the healing scar envelope, shapes the implant. Examination of breast implants explanted with their capsules intact has shown a variety of bizarre shapes and contour deformities departing widely from the original shape of the implant. It is difficult to examine for implant contour deformities in vivo due to the overlying breast tissue's ability to camouflage most implant deformities. However, through the use of three dimensional reconstruction of breast MR images, we have been able to recreate computerized models of implants as they sit in the breast. Surprisingly, most exhibit some degree of deformity.
Methods:
A retrospective review was conducted of consecutive breast MRI reports until 50 patients with implants were identified. Images of these patients were obtained, and those with short TI inversion recovery (STIR) images were included in the study. All patients were scanned using the same 1.5 T dedicated breast coil in the prone position with their breasts suspended uncompressed. Computerized three-dimensional reconstructions of the implant were created using Osirix software (Geneva, Switzerland). T2 image slices were also used for confirmation and clarity. These images were then reviewed for contour deformities by four different reviewers: the primary author, a senior plastic surgeon, and two radiologists trained in breast imaging. The implants were categorized as non-deformed, mildly deformed, or significantly deformed. Implants were grouped according to material and placement, and the rate of distortion was calculated for each group.
Results:
23 patients with a total of 43 implants were included in the study. There was some degree of distortion present in 31 implants (72.1%). Twelve implants were mildly distorted (27.9%), and 19 were significantly distorted (44.2%). Only 3 implants had distortions that were evident in the overlying skin (6.9%). Of the 24 saline implants, 5 were mildly distorted (20.1%), and 8 were significantly distorted (33.3%). Of the 19 silicone implants, 3 were mildly distorted (15.8%) and 11 were significantly distorted (58.9%). The overall distortion rate was 54.2% of the saline implants and 73.7% of the silicone. Of the 15 subpectoral implants, 6 were mildly distorted (40%) and 3 were significantly distorted (20%). Of the 28 subglandular implants, 7 were mildly distorted (25%) and 15 were significantly distorted (53.6%). The overall distortion rate was 60% for subpectoral implants and 78.6% for subglandular implants.
Conclusions:
A significant majority of implants included in this study expressed some degree of deformity. Implant distortions may be due to extrinsic factors such as surgically inappropriate pocket, irregular contraction of the capsule, or intrinsic factors such as implant folds. There was a higher overall rate of distortion in silicone implants and subglandular implants. Of all the deformed implants in this study, only three had any visible effect on the overlying skin, making deformed implants difficult to detect through physical diagnosis or routine mammography. With adequate soft-tissue coverage, deformed implants usually will have no effect on the external appearance of the breast. This study sheds some light upon the debate as to how round implants can perform like shaped devices in mimicking an axillary tail.
Figure 1:
Significantly deformed right-sided silicone gel implant with inferior
flattening and medial angular distortion.
The external breast appeared normal with no contour deformities.
Figure 2:
Significantly deformed right-sided silicone gel implant with
anteromedial depression, and medial extension of the implant across the
sternum. The anteromedial
depression was reflected in the external breast.