Teaching Case

38-year-old woman, DCIS grade 1 breast cancer, mastectomy in 2009. CT chest, abdomen, pelvis done for a subsequently benign complex left ovarian cyst first seen on US showed right supraclavicular and left internal mammary nodes which were FDG-avid on 18F-FDG PET/CT. Right Breast US, mammogram showed no mass. Neck Ultrasound, Breast MRI acquired.

The 18F-FDG PET/CT demonstrated a metabolically active right supraclavicular node (SUV max 4.1) and physiological left thyroid activity (Figure 1). In addition the 18F-FDG PET/CT showed a metabolically active left internal mammary node (SUV max 3) (Figure 2). The same metabolically active left internal mammary node is also seen on the non-hybrid PET image (Figure 3). Both metabolically active nodes were thought to represent nodal metastases. An ultrasound of the right supraclavicular node showed areas of "cloudy" high echogenicity with marked posterior enhancement ("snowstorm sign")(Figure 4) whilst a Silicone-specific MRI breast scan showed high signal intensity in the left internal mammary node (Figure 5). Intact left breast implant on MRI Breast (Figure 6).

The ultrasound of the right supraclavicular node showed areas of "cloudy" echogenicity with marked posterior enhancement ("snowstorm sign"). This is the classic appearance of a siliconoma [1] whilst a silicone-specific MRI of the breast showed high signal intensity in the left mammary node. This is also a feature of Siliconoma.

Several studies have demonstrated that metastasis to the internal mammary lymph nodes in breast carcinoma patients portends a worse prognosis [2]. Breast reconstruction/silicone breast implants are common procedures following mastectomy. Siliconoma is thought to occur following the transit of silicone droplets from breast implants to lymph nodes by macrophages and should always be considered as a differential diagnosis in patients in whom silicone prostheses are present. The risk of implant rupture/leak increases with implant age. A minimum of 15% of modern implants can be expected to rupture/leak between the third and tenth year after implantation [3]. Silicone breast injection has numerous adverse effects including skin reaction, pain, infection and formation of granulomas (i.e. siliconoma). Silicone usually migrates to the axillary lymph nodes but may migrate to other lymph nodes as well as extra nodal sites [4, 5]. In our case, given the intact breast implant, it is possible that there was silicone leak through the microscopic pores of the implant shell, without any discernible tear via a mechanism known as silicone gel bleeding. It is also possible that the silicone leak may have occurred through the medial aspect of the breast implant and migrated through the medial lymphatic channels to the supraclavicular and internal mammary nodes bypassing the axillary nodes (Figure 7). It is important to be able to tell the difference between locoregional nodal metastasis and a siliconoma to avoid unnecessary treatment.

18F-FDG PET/CT is a useful tool in assessing metastatic disease in breast cancer patients. However, as in this case occasionally false positive features can be encountered. It is well known that inflammatory granulomatous response following silicone injections can cause increased FDG uptake on PET-CT [6]. FDG uptake in the breast after a silicone implant may mimic a new primary breast tumour or recurrent breast tumour in patients with previous disease or positive nodes [7]. This concept is of utmost importance to the radiologist in image interpretation.
Ultrasound and MRI breast silicone sequences should be considered whenever there is doubt about activity within the nodes on 18F-FDG PET/CT particularly in low risk patients and also in cases where biopsy is considered difficult to obtain (i.e. internal mammary nodes).