Imaging of idle breast implants with ultrasound-strain elastography- A first experimental study to establish criteria for accurate imaging of idle implants via ultrasound-strain elastography

Abstract

AIM: To investigate whether there are fundamental sonographic and elastographic criteria to precisely assess different surfaces and fillings of idle breast implants and to determine their most distinctive parameters. This was a comparative study of different unused breast implant materials, neighter in animals nor in humans. This knowledge should be transferred in vivo to develop an objective measurement tool.

METHODS: Nine idle breast implants-silicone and polyurethane (PU)-were examined in an experimental study by using ultrasound B-mode with tissue harmonic imaging (THI), speckle reduction imaging (SRI, level 0–4), cross-beam (CB, low, medium, high), photopic and the colour coded ultrasound-strain elastography with a multifrequency probe (9–15 MHz).

Using a standardised protocol the implants’ centre as well as the edge were analysed by one experienced examiner. Two independent readers performed analysis and evaluation. For image interpretation a score was created (score 0:inadequate image, score 5:best image quality).

RESULTS: The highest score result for the centre was achieved by using ultrasound with B-mode in addition with CB level medium, SRI level 2, THI and photopic (mean:3.22±SD:1.56), but without any statistic significant difference (t-value = 0.71). With elastography the implants’ edge in general was represented without disruptive artefacts (3.89±0.60) with statistic significant difference (t-value = 5.29). Implants filled with inner cohesive silicone gel II° showed best imaging conditions for their centre via ultrasound (5±0) as well as for their edge via elastography (4.50±0.71).

CONCLUSION: Ultrasound-strain elastography and high resolution ultrasound represent a valuable measurement tool to evaluate different properties of idle breast implants. These modified ultrasound examinations could be an additional help for clinical investigations and be correlated with Baker’s Classification.

Keywords

Breast implants elastography Baker’s Classification

1 Introduction

So far breast augmentation with breast implants is one of the most popular and common surgeries in the field of plastic, reconstructive and aesthetic surgery. One of the first breast augmentations using breast implants was performed in 1963. Since then, implants have become increasingly popular with a growing popularity each year worldwide.

According to the International Society of Aesthetic Plastic Surgeons (ISAPS) 55.160 women in Germany were undergoing breast augmentation procedures in 2013, in the USA the amount was even higher with 313.703 procedures in 2013 [14]. That is why this surgery is one of the most popular procedures in the field of plastic and aesthetic reconstruction worldwide with a continuously increasing demand each year. Consequently, breast implants come under scrutiny, undergo steady developments and improvements and are part of continuous thorough investigations.

Surface characteristics have undergone steady changes. Besides silicone gel–filled and saline-filled implants, polyurethane-coated implants were developed to lower the occurrence of capsular fibrosis [10]. Nevertheless, national public health authorities regarding side effects as well as potential cancerogenic effects closely monitor all of these materials.

One of the major complications following breast augmentation with implants is the risk of developing a capsular contracture surrounding the implant [19]. Fibrocollagenous capsular fibrosis occurs at different degrees between smooth compared to textured surfaces. So far, the pathomechanisms are complex and the reasons are not completely identified assuming a multifactorial genesis and different contributory factors. Different studies discuss that seroma, infections, hematoma, bacterial colonisation and biofilms might have an impact on the development of capsular fibrosis [21].

A thin capsule of fibrous tissue (scar tissue) is formed around every implant inserted into a particular part of the human body as a foreign object frequently. In 1.3% to 30% instances, the fibrous capsule undergoes progressive thickening, hardness and shrinkage, which causes tenderness or pain occurring with a probability of 92% during the first 12 months [3, 4].

This major complication as well as other complications are often hard to detect during clinical examination due to insufficient imaging of common ultrasound and magnetic resonance imaging (MRI) [13]. Best clinical practice since 1978 for measuring capsular fibrosis remains the Baker score [25], a non-validated score, depending on examiner’s experience.

This inevitably leads to the fact, that Baker score is a subjective scale. Pain and pronounced fibrous capsule formation that alters the shape of the breast is a helpful predictor of capsular contracture or implants’ rupture, but absence of these symptoms does not exclude their presence. To obtain a more objective evaluation in breast examination, different imaging techniques have been investigated, including common ultrasound, computed tomography (CT), magnetic resonance imaging (MRI) and mammography.

MRI, ultrasound, CT and mammography are of considerable assistance in the diagnostic field, but they do not provide precise information about different stages of capsular contracture, the distinction of breast tissue and implant or follow-up of unrelated breast diseases such as mastitis or breast cancer [11]. Often, changes belonging to capsule fibrosis cannot be appreciated on imaging. Furthermore it is hard to detect smaller implant deformities. A specific and exact diagnostic in imaging is still missing.

The clinical utility of compressive colour coded-strain ultrasound and ultrasound-elastography has been well evaluated in the breast [6] and prostate [2], emphasising the potential clinical advantage for strain elastograms to differentiate benign from malignant masses based on the stiffness of soft tissues. Recent studies suggest ultrasound-strain elastography to be a potential additive for an objective evaluation of capsular contracture and could be an additional new method applicable for screening examinations of patients with breast implants [22]. Rjosk-Dendorfer et al. underline the importance of combination of different radiological examination methods –color Doppler sonography and sonoelastography –to enhance the reliability in identifying benign from malignant breast masses [23, 24].

Ultrasound-strain elastography is an available non-invasive clinical imaging technique, fast and radiation-free, that has been developed in the last decade [1 , 26]. Actually, there are guidelines about the new possibilities using ultrasound-strain elastography [27], still missing final evaluation for clinical examinations of different soft tissue changes.

Different tissue compression leads to different strain of tissue. Using a cross-correlation technique to determine the amount of displacement of each B-mode image pixel assesses information about the hardness of tissue. Elastograms are provided, that show pre- and post-compression frames to produce images of local strain, appearing as elasticity colour map images.

In this study we examined different idle breast implants with ultrasound elastography and determined their most discriminatory parameters for future transfer and exact diagnosis of breast implants and their surrounding tissue in vivo. Implants’ centre and edge was examined to further establish common parameters for ultrasound-elastography in evaluating breast implants and their surrounding tissue in vivo.

2 Materials and methods

In total, 108 observations (n = 108) were done of nine idle breast implants with different surface characteristics, silicone and polyurethane with either textured or smooth surfaces and different cohesivity of silicone gel. Two silicone implants with cohesivity I°, two implants with cohesivity II°, two implants with cohesivity III° and three breast implants with polyurethane. Cohesivity I° means that the inner cohesive silicone gel filling of the implant contains the softest gel, cohesive II° gel is a slightly firmer gel and III° means that the consistency is the firmest. All observations were conducted as a complete study on unused breast implants.

Using a standardised protocol the breast implants’ centre as well as the outer edge were analysed by one experienced radiologist specialized in breast imaging. Therefor the probe was connected directly to the implants surface with gel. The implants were located on a flat panel area, where they could not slip throughout the entire examination process. Two independent readers performed analysis and evaluation in consensus.

The assessment of ultrasound- and colour coded-strain elastography-images was based on a predetermined, subjectively chosen score, divided into six scores depending on the images’ qualities. Images’ quality, optimising parameters, level of purity, stability and reproducibility as well as the occurrence of artefacts were assessed, given score zero an inadequate, non-evaluable or lacking image with disruptive artefacts and score five indicated excellent, optimised and pure image quality. More complex structures with a tight density can lead to complex strain artefacts surrounding them. Interpretation via score four implied consistently well, but not optimal display, score three showed almost good quality with proportional artefacts compared to the implants’ inner structure. For score two significant artefacts compared to inner and outer structures were displayed. Score one also showed significant artefacts everywhere as well as additional interference patterns.

In accordance with the protocol all compressive elastography imaging modes were performed by one experienced ultrasound examiner with the colour coded dynamic compression-elastography mode [17] with a multifrequency linear probe (9–15 MHz, GE LOGIQ^® E9, GE Medical Systems, Milwaukee, WI, USA) (Fig. 1).

Different B-modes were applied to optimise the original image, such as speckle reduction imaging (SRI)-level 1 for low up to 4 for high-, spatial compounding (cross-beam [CB])- level 1 for low up to 3 for high-, tissue harmonic imaging (THI) (Fig. 2) and colour imaging with the help of photopic (Fig. 3). As THI is more useful in comparison to single B-mode ultrasound in distinguishing suspicious lesions in breast tissue [16], it was additionally used for precise interpretation of the original image.

In addition to that compressive colour coded-strain elastography was used. The elastography image is a semi-transparent colour map of stiffness overlaying the ultrasound image, with a colour spectrum from green to red (Fig. 4). The hardest tissues displayed as green and progressively softer tissues displayed as red, with colour nuances from blue, yellow and orange to display tissue stiffness in between. The default quantitative scale was 0–180 kPa. Gentle compression movements with the transducer were applied from a standardised distance on each implants’ centre and the outer edge. Different images, depending on the used mode, were created by computing relative tissue deformation globally and displaying the information within a user-defined region of interest.

We examined the implants’ centre as well as edge with regards to elasticity, thickness and signs of leakage. Necessary conditions for elastography images in relation with the pressure applied by the transducer were assessed on a five point scale using graphic reference standard ranging with five points for best image quality. The probe was held for five –ten seconds for the image to stabilize before recording the image for quantitative measurements.

For each implant, colour coded-strain elastography sequences were recorded for ten seconds and ten colour histogram frames were obtained. A quality marker was used to evaluate the best compression mode. Only cine sequences with the highest image quality with five green points were used for an appositional evaluation by a quantification mode (Q-analysis) integrated in the ultrasound machine workstation. User-defined regions of interest (ROIs) were placed on the implant. In total, 108 measurements were obtained. Five time measurements ensured that the elasticity index (EI) was not affected by compression variances caused by the examiner.

3 Statistics

Statistical analyses were performed using SPSS software for Windows (version 20.0; SPSS, Inc., Chicago, IL).

Continuous variables were described using mean±SD. A p-value of less than 0.05 was considered statistically significant. Paired comparisons t-test for dependent samples and a two-sample t-test assuming unequal variances were used.

4 Results

For the first time this study showed, that it is possible to evaluate different breast implant qualities with elastography and multifrequency ultrasound using different settings.

The implants’ different surfaces and fillings had significant influence on the occurance of artefacts or interferences.

High-resolution ultrasound with multifrequency probe is suitable for clear presentation of different breast implants’ centre. Whereas colour coded-strain elastography seems to be a useful instrument to evaluate the edge of different breast devices.

In recent validation studies, including comprehensive systematic reviews of studies that examined breast imaging via elastography, we determined that the amount of clear demarcation was significant higher in 45 kPa breast lesions [22].

In all investigations using a linear high-resolution multifrequency probe a coupling to the implants’ surface was achieved without significant interferences. An additional delay was not necessary. Whereas 15 MHz is the maximum frequency and limited to image surfaces’ changes, we used medium frequency of 11 MHz with level 2 of SRI and with medium level of cross-beam. In special cases contour irregularities were imaged more precisely using additional photopic. Depending on surface and filling of the breast implant the use of THI and medium levels of SRI and cross-beam with application of photopic were advantageous tools.

In general, the implants’ centre was presented more clearly than the edge in every modality of ultrasound. Here, the best imaging result for the centre was achieved by using ultrasound with B-mode in addition with CB, SRI, THI and photopic (3.22±1.56), see Fig. 5. However, t-value was 0.71, meaning that there is no statistic significant difference in ultrasound concerning the implants’ centre.

Whereas colour coded-strain elastography was a useful measurement tool to enhance the display of the implants’ outer edge generally without the occurance of disruptive artefacts (3.89±0.60), see Fig. 6. There was a statistic significant difference in the image of implant’s edges comparing to implant’s centre with elastography (t-value = 5.29). Strain elastography was a useful measurement tool to evaluate the implants’ outer edge. In all cases representative values could be obtained for good image quality up to score five.

We additionally analysed silicone implants depending on their grade of cohesivity. Breast implants with inner cohesive silicone gel filling II° showed best imaging conditions for their centre via ultrasound using all modalities (5±0) as well as for their edge via elastography (4.50±0.71), see Figs. 7 and 8. Furthermore, cohesivity II° showed best results in imaging and evaluation without any artefacts in ultrasound (4.25±0.96) and elastography (3.25±1.26) in general, but without any statistical significance (t-value = 1.26), see Figs. 7 and 4. All results are shown in (Table 1).

Regarding solid presentation of unused implants via ultrasound, we identified the higher the ultrasound image resolution the better the imaging of implants’ centre (Fig. 5). These ultrasound modalities permit an almost clear imaging with a minimum of artifacts. Vice versa strain elastography is a useful additional feature to show implants’ edge more detailed (Fig. 6).

5 Discussion

Plastic surgeons, gynaecologists as well as radiologists would benefit by becoming familiar with normal and abnormal findings in patients with breast implants. In particular, radiologists shall be able to recognize the normal appearance and distinguish between an abnormal or pathologic change of commonly used implants on various imaging techniques, particularly in the early stages. Therefore an objective measurement tool is essential to detect and prevent capsular contracture, even in its early stages. To combine thickness of the capsule and compressibility of the breast implant a deeper understanding of the imaging of surface and material of the breast implant itself is necessary.

This study is a first and so far unique study to depict criteria for ultrasound-strain elastography concerning imaging of different idle breast implants. Various studies have been published about breast imaging for MRI, mammography [20] or (contrast-enhanced) ultrasound [28] have been published, but clear and uniform diagnostic criteria for elastography [7, 27] as well as in general [15] are still missing. One of the reasons might be, that no one before tested image presentation and enhancement of surface properties of different pure and idle implants without any surrounding tissue.

Our scoring system was a useful tool to classify different categories concerning imaging of implants, which seems to be clinically promising. For further investigations this knowledge needs to be transformed into in-vivo diagnosis of breast implants and the anticipation of their possible complications. Capsular contracture is one of the most frequent complications following breast augmentation with implants [4, 5, 29]. In order to avoid such complications, surgeons as well as the industry continuously focus on improving implants’ textures, their shapes, surfaces and compositions [18, 29].

Our study might help to develop a classification for strain elastography screening concerning early detection of capsular fibrosis in breasts with implants. An objective and clear diagnostic system of capsular contracture is missing, especially for the early stages of this complication. Developing such a classification might help to prevent capsular contracture and protect against operation [9, 12] and to narrow the clinical diagnostic.

Imaging of a homogenous interface between implants’ surface and material via common ultrasound has always been challenging and ambitious. However, even this combination of maximal ultrasound modalities and elastography for high-resolution imaging did not always show best surface imaging without any artifacts. Constraints are the different surface and material properties of different breast implants, which interact in different ways with elastography imaging depending on their material. In comparison there are other implants’ materials that offer a high resolution with best imaging conditions.

Ultrasound with B-mode optimised by using SRI, CB, and THI and photopic can be useful to differentiate between implants, scanning the centre. Furthermore, strain elastography seems to be a useful device to examine the implants’ edge. Therefore strain elastography should be used as an additional measurement tool. We achieved precise and reproduceable results by using strain elastography, which is a first attempt to categorise imaging of idle implants via strain elastography.

Our investigation offers a pattern of frequency, which could be implemented for strain elastography in patients with breast implants. Furthermore it could facilitate to distinguish between vital tissues, malign or infected processes in vivo. Additional in vivo-studies are necessary for further evaluation and to establish a uniform standard for strain elastography and imaging of breast implants.

6 Conclusion

Aim of this study was to characterise sonographic and strain elastographic criteria for different idle breast implants. It can be challenging to examine breasts after augmentation with implants. In first place it seems to be important to get detailed information about the presentation of different idle implant materials and their surface properties in ultrasound and elastography. This knowledge could be useful for patients with augmented breasts for future investigation to ease clinical diagnosis, imaging and accurate interpretation in humans.

References

Adriaenssens

, Belsack

, Buyl

, Ruggiero

, Breucq

, De Mey

, et al. Ultrasound elastography as an objective diagnostic measurement tool for lymphoedema of the treated breast in breast cancer patients following breast conserving surgery and radiotherapy, Radiol Oncol 46(4) (2012)284–295.

Aigner

, Pallwein

, Junker

, Schfer

, Mikuz

, Pedross

, et al. Value of real-time elastography targeted biopsy for prostate cancer detection in men with prostate specific antigen 1.25ng/ml or greater and 4.00ng/ml or less, J Urol 184(3) (2010)913–917.

Araco

, Caruso

, Araco

, Overton

and Gravante

, Capsular contractures: A systematic review, Plast Reconstr Surg 124(6) (2009)1808–1819.

Araco

, Gravante

, Araco

and Delogu

D.C.V.

, Capsular contracture: Results of patients with aesthetic breast augmentation, Plast Reconstr Surg 118(6) (2006)1499–1500.

Burkhardt

B.R.

, Schnur

P.L.

and Tofield

J.J.D.P.

, Objective clinical assessment of fibrous capsular contracture, Plast Reconstr Surg 69(5) (1982)794–801.

Burnside

E.S.

, Hall

T.J.

, Sommer

A.M.

, Hesley

G.K.

, Sisney

G.A.

, Svensson

W.E.

et al. Differentiating benign from malignant solid breast masses with US strain imaging, Radiology 245(2) (2007)401–410.

Chang

S.C.

, Lee

Y.W.

, Lai

Y.C.

, Tiu

C.M.

, Wang

H.K.

, Chiou

H.J.

, Hsu

Y.W.

and Chou

Y.H.C.R.

, Automatic slice selection and diagnosis of breast strain elastography, Med Phys 41(10) (2014)102902.

Das

, Gupta

, Kaur

and Kalucha

, Elastography: The next step, J Oral Sci 53(2) (2011)137–141.

Gylbert

L.O.

, Applanation tonometry for the evaluation of breast compressibility, Scand J Plast Reconstr Surg Hand Surg 23(3) (1989)223–229.

10.

Handel

, Jensen

J.A.

, Black

, Waisman

J.R.

and Silverstein

M.J.

, The fate of breast implants: A critical analysis of complications and outcomes, Plast Reconstr Surg 96(7) (1995)1521–1533.

11.

Hauth

E.A.

, Hoffmann

, Zadow-Eulerich

, Kimmig

and Forsting

, Magnetic resonance tomography in the diagnostics of breast implants: A survey, Gynakol Geburtshilfliche Rundsch 48(2) (2008)76–83.

12.

Hoflehner

, Pierer

and Rehak

, “Mammacompliance”: An objective technique for measuring capsular fibrosis, Plast Reconstr Surg 92(6) (1993)1078–1084.

13.

Hölmich

L.R.

, Fryzek

J.P.

, Kjøller

, Breiting

V.B

Jørgensen

, and Krag

C.M.J.

, The diagnosis of silicone breast-implant rupture: Clinical findings compared with findings at magnetic resonance imaging, Ann Plast Surg 54(6) (2005)583–589.

14.

ISAPS International Society of Aesthetic Plastic Surgeons. International Survey on Aesthetic/Cosmetic Procedures 2013. Available at: http://http://www.isaps.org/Media/Default/global-statistics/2014%20ISAPS%20Global%20Stat%20Results.pdf. Accessed January 3, (2015).

15.

Joe

B.N.

and Sickles

E.A.

, The evolution of breast imaging: Past to present 1 evolution of breast imaging technique, Radiology 273(2) (2014), 23–44.

16.

Jung

E.M.

, Clevert

D.-A.

, Lutz

, Kett

and Rupp

, Preoperative wire localisation of breast lesions by tissue harmonic imaging (THI) sonography, Rofo 174 (2002)1121–1125.

17.

LOGIQ E9 Elastography. Available at: http://www3.gehealthcare.com/en/products/categories/ultrasound/logiq/new_logiq_e9. Accessed October 18, (2014).

18.

Marques

, Brown

S.A.

, Oliveira

, Cordeiro

M.N.D.S.

, Morales-Helguera

, Rodrigues

, et al., Long-term follow-up of breast capsule contracture rates in cosmetic and reconstructive cases, Plast Reconstr Surg 126(3) (2010)769–778.

19.

McGrath

M.H.

and Burkhardt

B.R.

, The safety and efficacy of breast implants for augmentation mammaplasty, Plast Reconstr Surg 74(4) (1984)550–560.

20.

Munn

, Special consent for mammography in women with breast implants?, Plast Reconstr Surg 125(4)167e–168e.

21.

Pinsolle

, Grinfeder

, Mathoulin-Pelissier

and Faucher

, Complications analysis of 266 immediate breast reconstructions, J Plast Reconstr Aesthetic Surg 59(10) (2006)1017–1024.

22.

Prantl

, Englbrecht

M.A.

, Kuehlmann

, Schoeneich

, Jung

E.M.

and Kubale

, Semiquantitative measurements of capsular contracture with elastography –First results in correlation to Baker Score, Clin Hemorheol Microcirc 58(4) (2014)521–528.

23.

Rjosk-Dendorfer

, Gürtler

V.M.

, Sommer

W.H.

, Reiser

and Clevert

D.A.

, Value of high resolution compression elastography and color doppler sonography in characterisation of breast lesions: Comparison of different high-frequency transducers, Clin Hemorheol Microcirc 57(2) (2014)129–135.

24.

Rjosk-Dendorfer

, Reu

, Deak

, Hetterich

, Kolben

, Reiser

, et al.,High resolution compression elastography and color doppler sonography in characterization of breast fibroadenoma, Clin Hemorheol Microcirc 58(1) (2014)115–125.

25.

Spear

S.L.

and Baker

J.L.

, Classification of capsular contracture after prosthetic breast reconstruction, Plast Reconstr Surg 96(5) (1995)1119–1123.

26.

Thomas

, Fischer

, Frey

, Ohlinger

, Grunwald

, Blohmer

J.U.

, et al.,Real-time elastography –An advanced method of ultrasound: First results in 108 patients with breast lesions, Ultrasound Obstet Gynecol 28(3) (2006)335–340.

27.

Wojcinski

, Degenhardt

, Peisker

and Beussel

S.H.M.

, Sonoelastography usage among german DEGUM- certified breast ultrasound specialists einsatz der sonoelastografie in der mammadiagnostik: Analyse unter DEGUM-zertifizierten untersuchern, Ultraschall Med 35(1) (2014)59–66.

28.

Xiao

, Ou

, Yang

, Wu

and Luo

, Breast contrast-enhanced ultrasound: Is a scoring system feasible? A preliminary study in China,e, PLoS One 9(8) (2014)105517–.

29.

Zahavi

, Sklair

M.L.

and Ad-El

D.D.

, Capsular contracture of the breast: Working towards a better classification using clinical and radiologic assessment, Ann Plast Surg 57(3) (2006)248–251.