High-Resolution Ultrasound in the Detection of Silicone Gel Breast Implant Shell Failure: Background, In Vitro Studies, and Early Clinical Results

Abstract

Magnetic resonance imaging (MRI) has historically been considered the "gold standard" for imaging silicone gel breast implants and is currently recommended by the US Food and Drug Administration for device surveillance. Recent studies, however, have questioned its accuracy as the best screening test for implant failure. In addition, the high cost of MRI is a significant deterrent to follow-up, especially among asymptomatic patients. Recent advancements in ultrasound technology have led to the development of high-resolution devices with the potential to accurately image breast implants and breast tissues. The authors evaluate the feasibility of portable, high-resolution ultrasound (HRUS) for imaging of silicone gel breast implants and perform preliminary comparisons of HRUS to MRI in the assessment of both intact and failed implants in a clinical setting by both radiologists and plastic surgeons. Phase 1 was composed of in vitro and ex vivo scanning model assessments in a variety of implant models utilizing multiple HRUS hardware platforms (GE LOGIQ-9, LOGIQ-e, LOGIQ-i, and Venue-40 devices) and transducer heads (range, 8-16 MHz, mainly GE12ML transducer). In Phase 2, these technologies were applied clinically to provide imaging experience in three patients previously diagnosed with unilateral implant failure. Phase 3 was a preliminary prospective evaluation of HRUS of 29 implants in 15 consecutive patients for whom MRI and independent surgeon-performed and radiologist-performed HRUS scans were compared to subsequent surgical findings. In Phase 1, all hardware models easily detected both intact and intentionally damaged shells in currently marketed fourth-generation responsive gel implants and in investigational, fifth-generation highly-cohesive gel devices. Although multiple transducers were able to detect shell failure, the 12-MHz head produced the best images at the normal clinical depth range. In Phase 2, confirmatory HRUS scans correctly identified the side of rupture and were consistent with MRI and surgical findings in all patients. In Phase 3, MRI, surgeon-performed HRUS, and radiologist-performed HRUS scans were all accurate in predicting implant shell integrity in 29 of 29 imaged breasts (100%) as confirmed at the time of surgery in both symptomatic and asymptomatic patients. Preliminary results with a variety of base and transducer systems demonstrated that HRUS provides excellent visualization of current fourth- and fifth-generation silicone gel implants both in the in vitro and ex vivo scanning models. In vivo surgeon-performed HRUS accurately identified implant status and correlated with radiologist-performed HRUS, MRI, and surgical findings. An ongoing Phase 4 prospective study is under way to help define the sensitivity and specificity of HRUS technologies in the evaluation of current implant designs. However, the relative affordability, accessibility, availability, and dynamic real-time visualization provided by HRUS represent significant potential advantages of HRUS over MRI in both the screening and future diagnosis of breast implant shell failure.