Breast Surgery
Preliminary Report
Diagnosis of Ruptured Breast Implants Through High-Resolution Ultrasound Combined With Real-Time Elastography
Aesthetic Surgery Journal 2015, Vol 35(4) 410–418 © 2015 The American Society for Aesthetic Plastic Surgery, Inc. Reprints and permission: [email protected] DOI: 10.1093/asj/sju057 www.aestheticsurgeryjournal.com
Angrit Stachs, MD; Max Dieterich, MD; Steffi Hartmann, MD; Johannes Stubert, MD; Toralf Reimer, PhD; and Bernd Gerber, PhD
Abstract Background: Implant rupture as a late complication of breast implant surgery is often a silent phenomenon that is difficult to diagnose. Sonoelastography is a new ultrasound-based technique that allows assessment of tissue elasticity. Objectives: This study was undertaken to evaluate elastographic findings in normal and ruptured breast implants. Methods: This prospective study included 28 implants in 16 patients, all of whom underwent high-resolution ultrasound and real-time elastography. The diagnosis of implant rupture was confirmed by surgery. Results: Implant rupture was diagnosed in 5 out of 28 implants (17.9%). In those patients with ruptured implants, 3 had no symptoms, 1 presented with pain, and 1 complained of ipsilateral axillary lymph node swelling. Implants with a homogenous anechoic interior were considered to be intact. Ultrasound findings indicating implant rupture included multiple parallel echogenic lines in the implant interior in 2 cases and a mix of hyperechoic and hypoechoic masses in 3 cases. The feasibility of real-time elastography of implants was demonstrated in all cases. Elastograms of intact implants revealed a typical blue-green-red pattern familiar from cystic lesions. In all 5 ruptured implants, elastography revealed yellow-green figures without typical layering. Conclusions: To the authors’ knowledge this is the first series to combine high-resolution ultrasound with real-time elastography for the diagnosis of implant rupture. Since there are distinct differences between elastograms of intact and ruptured implants, addition of real-time elastography to conventional ultrasound may improve implant surveillance and obviate the need for magnetic resonance imaging. Level of Evidence: 3
Accepted for publication September 5, 2014; online publish-ahead-of-print March 24, 2015.
Over the past 25 years, the number of silicone breast implants for cosmetic or reconstructive surgery has increased substantially.1 In 2006, the US Food and Drug Administration (FDA) recommended routine magnetic resonance imaging (MRI) evaluations for postoperative implant surveillance.2 A recent meta-analysis of studies on the diagnostic accuracy of MRI for detecting silicone breast implant rupture revealed sensitivity of 87.0% and specificity of 89.9%.3 However, because diagnostic accuracy was 14-fold higher in symptomatic compared with asymptomatic patients, it was concluded that most of the studies considered for the meta-analysis were methodologically flawed because only patients with
Therapeutic
symptoms were included.3 In terms of expense and specific contraindications (eg, metallic implants, claustrophobia), MRI cannot be generally recommended as a screening method for implant rupture in asymptomatic women.
From the Breast Unit, Department of Gynecology and Obstetrics, University of Rostock, Rostock, Germany. Corresponding Author: Angrit Stachs MD, Department of Gynecology and Obstetrics, University of Rostock, Südring 81, 18059, Rostock, Germany. E-mail: [email protected]
Stachs et al
There is also no evidence to indicate whether or not patients with silent implant rupture should proceed to surgical explantation. Nevertheless, there exists a need for noninvasive assessment of breast implants under various conditions. Ultrasound (US) is the standard imaging technique in young women presenting with breast symptoms. Although the application of US for evaluation of breast implants is widespread, its diagnostic performance falls short of the accuracy achieved with MRI. The results of US examination depend mainly on examiner experience and the technical equipment used. A retrospective study by Cilotti et al4 in which symptomatic patients were examined using high-frequency transducers (≥10 megahertz [MHz]) revealed a sensitivity of 92% and specificity of 100% for the diagnosis of implant complications. Another study evaluating the usefulness of high-resolution ultrasound (HRUS) in the detection of implant failure demonstrated 100% concordance between MRI, US, and surgical findings in 18 asymptomatic and 11 symptomatic breast implants.5 In terms of costs, applicability, time, and patient convenience, US represents a valuable tool for the diagnostic detection of implant complications. Sonographic findings in breast implant rupture have also been summarized in a recently published pictorial review by Lake et al.6 Sonoelastography is a new imaging technique that estimates the relative strain on tissues in response to external compression.7 Earlier studies have demonstrated the value of sonoelastography for differentiating benign from malignant breast lesions.8-10 The objective of the present study was to evaluate the use of real-time elastography (Philips iU22, Bothwell, WA, USA) in addition to ultrasound for the diagnosis of implant failure.
METHODS In order to evaluate the feasibility of sonoelastography for the detection of silicone implant failure, a 2-phase preliminary study was undertaken from May 2013 until December 2013. All 16 patients recruited provided written informed consent. The study was approved by the Institutional Review Board of the University of Rostock and was conducted in accordance with the principles of the Declaration of Helsinki. Phase 1 was a pilot study to assess the use of elastography in breast implants and included 2 patients who had undergone bilateral augmentation for cosmetic reasons and had been scheduled for follow-up in the wake of the Poly Implant Prosthesis (PIP) scandal. In Phase 2, we started a prospective trial including 14 consecutive patients who had received breast implants for cosmetic and reconstructive reasons and who had been referred to our hospital with cosmetic complaints or for breast cancer follow-up.
411
All patients were examined using US and additional sonoelastography carried out by an experienced breast surgeon (AS). A Philips iU22 with a high-frequency linear transducer (12 MHz) and commercially available real-time elastography function was used for US and elastography. The sonoelastographic function of the Philips software is based on tissue compression during thoracic respiratory excursions. Sonoelastography comprised at least 4 images per implant (1 image in each quadrant). Implants were considered to be intact where US findings revealed well-defined implant edges without interruption, homogeneous hypoechoic implants, and no external echoes. US signs of inhomogeneity of the implant lumen as well as the stepladder sign (multiple, discontinuous linear echoes in the lumen) were regarded as typical indicators of intracapsular rupture. Extracapsular rupture was diagnosed in cases of increased echogenicity of breast tissue and loss of normal parenchymal interfaces (snowstorm pattern), extracapsular hypoechoic nodules representing extracapsular silicone, and snowstorm pattern of axillary lymph nodes. Patients with abnormal sonomorphological and clinical findings underwent additional MRI examination. Surgical procedures were carried out in cases with confirmed implant rupture or other complications, eg, capsular contraction, infection.
RESULTS Eleven out of a total of 16 female patients had received bilateral breast implants for cosmetic reasons. Of the remaining 5 patients, 4 had unilateral implants and 1 had bilateral implants for reconstruction after mastectomy. The patients’ mean age was 39 years, with a range from 23 to 65 years. The average follow-up period was 6 years (range: 1-14 years). Patients presented with symptoms (pain or a palpable mass) in 11 out of 16 cases (68.7%). Twelve out of 28 (42.9%) individual silicone implants were associated with symptoms. Clinical examination revealed no signs of implant failure in any case. All patients underwent HRUS followed by real-time elastography (Table 1). Elastograms of sufficiently high quality were obtained within 5 minutes. Altogether, 5 implant ruptures (17.9%) were detected, including 2 extracapsular ruptures (Patients 1 and 11) and 3 intracapsular ruptures (Patients 2, 6, and 15).
Phase 1 Ultrasound revealed signs of unilateral implant failure in both patients, and the elastography series confirmed the presence of 2 normal and 2 ruptured implants. In the 2 implants with normal ultrasound findings, elastography revealed a typical blue-green-red pattern consistent with the bull’s eye artifact11 that is familiar from cystic lesions (Figure 1A,B). Abnormal US findings in the left implant
Aesthetic Surgery Journal 35(4)
412
Table 1. Patient Characteristics, Imaging Findings, and Surgical Findings (n = 16) Study Phase
1
Patient Side
Age (yr)
Symptoms
Follow-up (yr)
Ultrasound
Elastogram
MRI
Surgical Findings
pt 1 L
30
None
8
Ruptured
Ruptured
Ruptured
Ruptured
None
8
Intact
Intact
Intact
Intact
None
14
Ruptured
Ruptured
Ruptured
Ruptured
None
14
Intact
Intact
Intact
Intact
pt 1 R pt 2 L
43
pt 2 R 2
pt 3 R
46
Pain
1
Intact
Intact
n.d.
n.d.
pt 4 R
23
None
1
Intact
Intact
n.d.
n.d.
None
1
Intact
Intact
n.d.
n.d.
Mass
2
Intact
Intact
n.d.
n.d.
None
3
Intact
Intact
n.d.
n.d.
Axillary mass
7
Ruptured
Ruptured
Ruptured
Ruptured
None
7
Intact
Intact
Intact
Intact
Pain
4
Intact
Intact
n.d.
n.d.
None
4
Intact
Intact
n.d.
n.d.
pt 4 L pt 5 L
51
pt 5 R pt 6 L
32
pt 6 R pt 7 L
24
pt 7 R pt 8 L
62
None
1
Intact
Intact
n.d.
n.d.
pt 9 L
40
Pain
1
Intact
Intact
n.d.
n.d.
None
1
Intact
Intact
n.d.
n.d.
None
8
Intact
Intact
n.d.
n.d.
Skin necrosis
8
Intact
Intact
n.d.
Intact
Swelling
10
Ruptured
Ruptured
n.d.
Ruptured
None
10
Intact
Intact
n.d.
Intact
Pain
5
Intact
Intact
n.d.
Intact
Pain (fibrosis)
5
Intact
Intact
n.d.
Intact
None
1
Intact
Intact
n.d.
n.d.
Mass
1
Intact
Intact
n.d.
n.d.
pt 9 R pt 10 L
38
pt 10 R pt 11 L
35
pt 11 R pt 12 L
34
pt 12 R pt 13 L
33
pt 13 R pt 14 L
65
Axillary mass
9
Intact
Intact
n.d.
n.d.
pt 15 L
34
None
9
Intact
Intact
n.d.
Intact
None
9
Ruptured
Ruptured
n.d.
Ruptured
Pain
3
Intact
Intact
n.d.
n.d.
pt 15 R pt 16 R
42
L, left breast; MRI, magnetic resonance imaging; n.d., not done; R, right breast.
from patient 1 included marginal irregularity and snowstorm patterning (Figure 2A). Elastography in this patient revealed a mixed pattern consisting of green and blue regions without typical layering (Figure 2B). Patient 2 presented with US abnormalities consistent with inhomogeneity of the left implant lumen (Figure 3A). The corresponding elastogram revealed an identical pattern to that seen for
patient 1 (Figure 3B). For both patients, unilateral implant rupture was confirmed by MRI and surgical findings.
Phase 2 Three unilateral implant ruptures were diagnosed in 14 patients with 24 silicone implants. Sonomorphological signs of
Stachs et al
413
Figure 1. A 24-year-old woman (Patient 7) 4 years after bilateral breast augmentation with cohesive silicone gel implants. Ultrasound (US) and elastography images of an intact breast implant in split screen mode. (A) US: noninterrupted shell and a normal reverberation pattern. (B) Elastography: blue-green-red pattern artifact.
implant rupture in Patient 6 included multiple discontinuous linear echoes in the lumen of the medial part of the left implant (Figure 4A). Real-time elastography of this region demonstrated inhomogeneity of the red-green-yellow pattern in the lower part of the prosthesis, whereas the upper part was almost blue (Figure 4B). Ultrasound findings and elastograms for Patients 11 and 15 revealed features similar to those described for Patient 6. All patients with sonographic findings indicative of intracapsular implant rupture underwent MRI (Figure 3C). Implant rupture was confirmed by surgery in all cases (Figure 4C).
DISCUSSION Since the Poly Implant Prosthesis (PIP) scandal, women who have undergone breast augmentation are now more frequently concerned about possible implant rupture. The FDA recommendation regarding repeated breast MRI for implant surveillance in asymptomatic women has not met with widespread acceptance. Although numerous studies have evaluated MRI and ultrasound in the detection of implant rupture in the past, the results need to be interpreted with caution.12-20 In a recent systematic meta-analysis conducted in 2011, Song et al identified 1175 publications related to this topic, of which only 21 were eligible for inclusion. The pooled sensitivity and specificity as calculated by logistic regression modeling were 87.0% and 89.9%,
respectively, for MRI and 60.8% and 76.3%, respectively, for US.3 However, the meta-analysis identified methodological biases in many of the studies reviewed, with those conducted in symptomatic patients leading to an overestimation of the diagnostic accuracy of MRI. Moreover, MRI is also specifically contraindicated for patients with metallic implants or suffering from claustrophobia, and the expense of the technique constitutes a further disadvantage. HRUS, carried out by an experienced examiner, is able to detect implant failure in a high percentage of cases. In 2012 Bengtson and Eaves reported 100% concordance between MRI, high-resolution US, and surgical findings in a prospective trial including 29 implants in 15 patients, half of them without symptoms.5 The conclusions drawn are limited by the small sample size and the fact that in real-time US examination it is impossible to blind the examiner to patients’ symptoms. On the other hand, realtime, dynamic US examination offers the advantage that compression or movement can be used to further explore regions of suspected rupture. The disadvantages associated with US include poorer visualization of the deep parts of the implant, and it is more examiner-dependent than MRI. Our study confirms the potential of HRUS for diagnosis of implant failure. Moreover, we have presented initial findings concerning the additional application of real-time sonoelastography for implant surveillance. The principle of sonoelastography has been described elsewhere.8,21 Currently, 2 different
414
Aesthetic Surgery Journal 35(4)
Figure 2. Extracapsular rupture of left breast implant in this 30-year-old asymptomatic woman (Patient 1) presenting for follow-up 8 years after implantation of Poly Implant Prosthesis implants. (A) Ultrasound: marginal irregularity and inhomogeneity (snowstorm pattern). At left (arrowed) a hypoechoic nodule representing extracapsular silicone. (B) Elastography: a mixed pattern without typical layering.
techniques are available: strain elasticity imaging in which the relative strain or elasticity of tissue is determined in a qualitative or semiquantitative manner, and shear wave elasticity imaging in which the velocity of a shear wave induced by an initial ultrasound pulse is measured. Because the velocity of the shear wave is proportional to stiffness, the strain modulus can be estimated (in kilopascals). Both techniques have demonstrated high sensitivity and specificity for characterizing breast lesions as benign or malignant.10,22-25 Furthermore, potential applications of elasticity imaging in distinguishing cystic lesions from solid masses have been suggested, since cystic breast lesions show a blue-green-red pattern artifact on elastography.26,27 This artifact can be explained in terms of the physical principles of elastography, as reported in detail by Cho et al.26 In brief, soft tissues with an extremely low internal echo have a very small gradient of displacement. Therefore, a linear interpolation process occurs in between the background and the soft tissue, leading to a reversed gradient of displacement in the superficial layer (coded as blue) and an overestimation in the deep layer of the cyst (coded as red). To our knowledge, our group is the first to describe this artifact in intact silicone breast implants. Additionally, our study demonstrates that this artifact cannot be reproduced in ruptured implants. On the contrary, depending on the degree of destruction present, elastography revealed a
mixed pattern without any layering. In contrast to the study by Bengtson and Eaves, which revealed differences in the pattern of rupture depending on implant durometer (gel vs form-stable implants), we did not notice any differences.5 This may be explained by the fact that only 1 patient (Patient 1) had a noncohesive gel implant whereas the other 4 patients with implant rupture had form-stable implants. Elastography was straightforward to perform, and evaluation did not require a long learning curve. The combination of high-resolution US and real-time elastography has the advantage that equivocal US findings, such as minimal inhomogeneity, can be immediately verified by real-time elastography. Conversely, in cases with a definite US diagnosis of implant failure, the need for expensive MRI examinations is obviated. Our first Phase 1 study patient underwent MRI (although extracapsular rupture had been diagnosed by HRUS) because she was the first patient in our institution to be affected by the PIP scandal. In Phase 2, for patients with certain ultrasonographic signs of extracapsular implant rupture, MRI was omitted. In our opinion, expensive MRI should be skipped in cases where the results do not affect patient management. In our study, the elastographic findings merely corroborate the ultrasonographic diagnosis. However, ultrasonographic findings in implant rupture are highly varied, and diagnostic accuracy is dependent on investigator
Stachs et al
415
Figure 3. Intracapsular rupture of left breast implant in this 43-year-old asymptomatic woman (Patient 2) 14 years after bilateral augmentation with silicone implants. (A) Ultrasound: implant showing inhomogeneity and echoic pattern. (B) Elastography: absence of typical blue-green-red layering. (C) Magnetic resonance imagine scan of Patient 2, horizontal view. The collapsed implant shell (“linguine sign”) and intracapsular gel are clearly visualized on the left side.
experience. Therefore, elastography has the potential to serve as an additional method for patients with equivocal findings or for examiners with a low level of experience.
Our study has several limitations. First, we have presented preliminary results based on a small number of patients. Second, in our study population (n = 16), the current gold
416
Aesthetic Surgery Journal 35(4)
Figure 4. Intracapsular rupture in this 32-year-old woman (Patient 6) suffering from axillary lymphadenopathy 7 years after bilateral implantation of cohesive silicone prostheses. (A) Ultrasound: echoic longitudinal stripes (stepladder sign) in the medial part of the implant. (B) Elastography: blue-colored upper part of the implant, but inhomogeneity in the lower part of the prosthesis, showing a mixed pattern of red and green areas. (C) Ruptured implant after explantation (Patient 6): the fractured shell fragment is lifted to demonstrate fragmentation of the cohesive fill material.
standard of MRI as well as surgery has not been applied to all patients, but only to those in whom implant rupture was suspected, in line with our local clinical practice.
Without performing MRI examinations on all patients, we are unable to provide data regarding the sensitivity and specificity of HRUS and elastography. From an ethical
Stachs et al
standpoint, in asymptomatic patients without US findings of implant failure, there is no indication for surgery, and expensive diagnostic modalities such as MRI with potential side effects and specific contraindications, eg, claustrophobia, metallic implants, should be avoided. Elastography did not reveal distinct differences between intra- and extracapsular rupture, but this may be explained by the preliminary nature of our study and the limited number of cases. Finally, in terms of visualizing the deep implant surface, high-frequency US as well as real-time elastography may be of limited utility for patients with large-volume implants. We do not, in fact, believe that elastography can replace HRUS, but it might be a very helpful addition for diagnosing implant rupture. To evaluate the potential of the combination of HRUS and elastography in comparison with the gold standard MRI, a larger study of patients scheduled for implant exchange (and including power analysis) is planned. However, in view of the fact that implant rupture is (fortunately) a rare complication, the recruitment period may extend over several years.
CONCLUSIONS Our results demonstrate for the first time that the combination of HRUS with real-time elastography is a valuable tool for the diagnosis of implant rupture. Sonoelastography is easy to perform and is not associated with any specific side effects. Since elastographic findings in intact and ruptured implants are distinctively different, elastography may well prove useful as an additional method for patients with equivocal US findings or for examiners with a low level of experience.
417
4.
5.
6.
7. 8.
9.
10.
11.
12.
13.
14.
Disclosures The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.
15.
Funding
16.
The authors received no financial support for the research, authorship, and publication of this article.
REFERENCES 1. 2.
3.
Cosmetic Surgery National Data Bank: Statistics 2013. Aesthet Surg J. 2014;34 (1 suppl):1S-22S. US Food and Drug Administration. Regulatory history of implants in the US. Available at: http://www.fda.gov/ MedicalDevices/ProductsandMedicalProcedures/Implants andProsthetics/BreastImplants/ucm064461.htm. Accessed May 13, 2014. Song JW, Kim HM, Bellfi LT, Chung KA. The effect of study design biases on the diagnostic accuracy of
17.
18.
19.
magnetic resonance imaging to detect silicone breast implant ruptures: a meta-analysis. Plast Reconstr Surg. 2011;127:1029-1044. Cilotti A, Marini C, Iacconi C, et al. Ultrasonographic appearance of breast implant complications. Ann Plast Surg. 2006;56:243-247. Bengtson BP, Eaves FF III. High-resolution ultrasound in the detection of silicone gel breast implant shell failure: background, in vitro studies, and early clinical results. Aesthetic Surg J. 2012;32:157-174. Lake E, Ahmad S, Dobrashian R. The sonographic appearances of breast implant rupture. Clin Radiol. 2013;68: 851-858. Garra BS. Imaging and estimation of tissue elasticity by ultrasound. Ultrasound Q. 2007;23:255-268. Itoh A, Ueno E, Tohno E, et al. Breast disease: clinical application of US elastography for diagnosis. Radiology. 2006;239:341-350. Sadigh G, Carlos RC, Neal CH, Dwamena BA. Ultrasonic differentiation of malignant from benign breast lesions: a meta-analytic comparison of elasticity and BIRADS scoring. Breast Cancer Res Treat. 2012;133:23-35. Stachs A, Hartmann S, Stubert J, et al. Differentiating between malignant and benign breast masses: factors limiting sonoelastographic strain ratio. Ultraschall Med. 2013;34:131-136. Barr RG, Lackey AE. The utility of the "bull’s-eye" artifact on breast elasticity imaging in reducing breast lesion biopsy rate. Ultrasound Q. 2011;27:151-155. Di Benedetto G, Cecchini S, Grassetti L, et al. Comparative study of breast implant rupture using mammography, sonography, and magnetic resonance imaging: correlation with surgical findings. Breast J. 2008;14:532-537. Chung KC, Wilkins EG, Beil RJ Jr, et al. Diagnosis of silicone gel breast implant rupture by ultrasonography. Plast Reconstr Surg. 1996;97:104-109. Hölmich LR, Fryzek JP, Kjøller K, et al. The diagnosis of silicone breast-implant rupture: clinical findings compared with findings at magnetic resonance imaging. Ann Plast Surg. 2005;54:583-589. Collis N, Litherland J, Enion D, Sharpe DT. Magnetic resonance imaging and explantation investigation of longterm silicone gel implant integrity. Plast Reconstr Surg. 2007;120:1401-1406. Paetau AA, McLaughlin SA, McNeil RB, et al. Capsular contracture and possible implant rupture: is magnetic resonance imaging useful? Plast Reconstr Surg. 2010;125: 830-835. Berg WA, Caskey CI, Hamper UM, et al. Single- and doublelumen silicone breast implant integrity: prospective evaluation of MR and US criteria. Radiology. 1995;197:45-52. Everson LI, Parantainen H, Detlie T, et al. Diagnosis of breast implant rupture: imaging findings and relative efficacies of imaging techniques. AJR Am J Roentgenol. 1994;163:57-60. Quinn SF, Neubauer NM, Sheley RC, et al. MR imaging of silicone breast implants: evaluation of prospective and retrospective interpretations and interobserver agreement. J Magn Reson Imaging. 1996;6:213-218.
Aesthetic Surgery Journal 35(4)
418
20.
21. 22.
23.
Maijers MC, Niessen FB, Veldhuizen JF, et al. MRI screening for silicone breast implant rupture: accuracy, interand intraobserver variability using explantation results as reference standard. Eur Radiol. 2014;24:1167-1175. Barr RG. Sonographic breast elastography – a primer. J Ultrasound Med. 2012;31:773-783. Wojcinski S, Farrokh A, Weber S, et al. Multicenter study of ultrasound real-time tissue elastography in 779 cases for the assessment of breast lesions: improved diagnostic performance by combining the BI-RADS®-US classification system with sonoelastography. Ultraschall Med. 2010;31:484-491. Cho N, Moon WK, Kim HY, et al. Sonoelastographic strain index for differentiation of benign and malignant nonpalpable breast masses. J Ultrasound Med. 2010;29:1-7.
24. Yi A, Cho N, Chang JM, et al. Sonoelastography for 1,786 non-palpable breast masses: diagnostic value in the decision to biopsy. Eur Radiol. 2012;22:1033-1040. 25. Berg WA, Cosgrove DO, Doré CJ, et al. BE1 Investigators: Shear-wave elastography improves the specificity of breast US: the BE1 multinational study of 939 masses. Radiology. 2012;262:435-449. 26. Cho N, Moon WK, Chang JM, et al. Aliasing artifact depicted on ultrasound (US)-elastography for breast cystic lesions mimicking solid masses. Acta Radiol. 2011;52: 3-7. 27. Booi RC, Carson PL, O’Donnell M, et al. Characterization of cysts using differential correlation coefficient values from two dimensional breast elastography: preliminary study. Ultrasound Med Biol. 2008;34:12-21.
Preliminary Report
Diagnosis of Ruptured Breast Implants Through High-Resolution Ultrasound Combined With Real-Time Elastography
Aesthetic Surgery Journal 2015, Vol 35(4) 410–418 © 2015 The American Society for Aesthetic Plastic Surgery, Inc. Reprints and permission: [email protected] DOI: 10.1093/asj/sju057 www.aestheticsurgeryjournal.com
Angrit Stachs, MD; Max Dieterich, MD; Steffi Hartmann, MD; Johannes Stubert, MD; Toralf Reimer, PhD; and Bernd Gerber, PhD
Abstract Background: Implant rupture as a late complication of breast implant surgery is often a silent phenomenon that is difficult to diagnose. Sonoelastography is a new ultrasound-based technique that allows assessment of tissue elasticity. Objectives: This study was undertaken to evaluate elastographic findings in normal and ruptured breast implants. Methods: This prospective study included 28 implants in 16 patients, all of whom underwent high-resolution ultrasound and real-time elastography. The diagnosis of implant rupture was confirmed by surgery. Results: Implant rupture was diagnosed in 5 out of 28 implants (17.9%). In those patients with ruptured implants, 3 had no symptoms, 1 presented with pain, and 1 complained of ipsilateral axillary lymph node swelling. Implants with a homogenous anechoic interior were considered to be intact. Ultrasound findings indicating implant rupture included multiple parallel echogenic lines in the implant interior in 2 cases and a mix of hyperechoic and hypoechoic masses in 3 cases. The feasibility of real-time elastography of implants was demonstrated in all cases. Elastograms of intact implants revealed a typical blue-green-red pattern familiar from cystic lesions. In all 5 ruptured implants, elastography revealed yellow-green figures without typical layering. Conclusions: To the authors’ knowledge this is the first series to combine high-resolution ultrasound with real-time elastography for the diagnosis of implant rupture. Since there are distinct differences between elastograms of intact and ruptured implants, addition of real-time elastography to conventional ultrasound may improve implant surveillance and obviate the need for magnetic resonance imaging. Level of Evidence: 3
Accepted for publication September 5, 2014; online publish-ahead-of-print March 24, 2015.
Over the past 25 years, the number of silicone breast implants for cosmetic or reconstructive surgery has increased substantially.1 In 2006, the US Food and Drug Administration (FDA) recommended routine magnetic resonance imaging (MRI) evaluations for postoperative implant surveillance.2 A recent meta-analysis of studies on the diagnostic accuracy of MRI for detecting silicone breast implant rupture revealed sensitivity of 87.0% and specificity of 89.9%.3 However, because diagnostic accuracy was 14-fold higher in symptomatic compared with asymptomatic patients, it was concluded that most of the studies considered for the meta-analysis were methodologically flawed because only patients with
Therapeutic
symptoms were included.3 In terms of expense and specific contraindications (eg, metallic implants, claustrophobia), MRI cannot be generally recommended as a screening method for implant rupture in asymptomatic women.
From the Breast Unit, Department of Gynecology and Obstetrics, University of Rostock, Rostock, Germany. Corresponding Author: Angrit Stachs MD, Department of Gynecology and Obstetrics, University of Rostock, Südring 81, 18059, Rostock, Germany. E-mail: [email protected]
Stachs et al
There is also no evidence to indicate whether or not patients with silent implant rupture should proceed to surgical explantation. Nevertheless, there exists a need for noninvasive assessment of breast implants under various conditions. Ultrasound (US) is the standard imaging technique in young women presenting with breast symptoms. Although the application of US for evaluation of breast implants is widespread, its diagnostic performance falls short of the accuracy achieved with MRI. The results of US examination depend mainly on examiner experience and the technical equipment used. A retrospective study by Cilotti et al4 in which symptomatic patients were examined using high-frequency transducers (≥10 megahertz [MHz]) revealed a sensitivity of 92% and specificity of 100% for the diagnosis of implant complications. Another study evaluating the usefulness of high-resolution ultrasound (HRUS) in the detection of implant failure demonstrated 100% concordance between MRI, US, and surgical findings in 18 asymptomatic and 11 symptomatic breast implants.5 In terms of costs, applicability, time, and patient convenience, US represents a valuable tool for the diagnostic detection of implant complications. Sonographic findings in breast implant rupture have also been summarized in a recently published pictorial review by Lake et al.6 Sonoelastography is a new imaging technique that estimates the relative strain on tissues in response to external compression.7 Earlier studies have demonstrated the value of sonoelastography for differentiating benign from malignant breast lesions.8-10 The objective of the present study was to evaluate the use of real-time elastography (Philips iU22, Bothwell, WA, USA) in addition to ultrasound for the diagnosis of implant failure.
METHODS In order to evaluate the feasibility of sonoelastography for the detection of silicone implant failure, a 2-phase preliminary study was undertaken from May 2013 until December 2013. All 16 patients recruited provided written informed consent. The study was approved by the Institutional Review Board of the University of Rostock and was conducted in accordance with the principles of the Declaration of Helsinki. Phase 1 was a pilot study to assess the use of elastography in breast implants and included 2 patients who had undergone bilateral augmentation for cosmetic reasons and had been scheduled for follow-up in the wake of the Poly Implant Prosthesis (PIP) scandal. In Phase 2, we started a prospective trial including 14 consecutive patients who had received breast implants for cosmetic and reconstructive reasons and who had been referred to our hospital with cosmetic complaints or for breast cancer follow-up.
411
All patients were examined using US and additional sonoelastography carried out by an experienced breast surgeon (AS). A Philips iU22 with a high-frequency linear transducer (12 MHz) and commercially available real-time elastography function was used for US and elastography. The sonoelastographic function of the Philips software is based on tissue compression during thoracic respiratory excursions. Sonoelastography comprised at least 4 images per implant (1 image in each quadrant). Implants were considered to be intact where US findings revealed well-defined implant edges without interruption, homogeneous hypoechoic implants, and no external echoes. US signs of inhomogeneity of the implant lumen as well as the stepladder sign (multiple, discontinuous linear echoes in the lumen) were regarded as typical indicators of intracapsular rupture. Extracapsular rupture was diagnosed in cases of increased echogenicity of breast tissue and loss of normal parenchymal interfaces (snowstorm pattern), extracapsular hypoechoic nodules representing extracapsular silicone, and snowstorm pattern of axillary lymph nodes. Patients with abnormal sonomorphological and clinical findings underwent additional MRI examination. Surgical procedures were carried out in cases with confirmed implant rupture or other complications, eg, capsular contraction, infection.
RESULTS Eleven out of a total of 16 female patients had received bilateral breast implants for cosmetic reasons. Of the remaining 5 patients, 4 had unilateral implants and 1 had bilateral implants for reconstruction after mastectomy. The patients’ mean age was 39 years, with a range from 23 to 65 years. The average follow-up period was 6 years (range: 1-14 years). Patients presented with symptoms (pain or a palpable mass) in 11 out of 16 cases (68.7%). Twelve out of 28 (42.9%) individual silicone implants were associated with symptoms. Clinical examination revealed no signs of implant failure in any case. All patients underwent HRUS followed by real-time elastography (Table 1). Elastograms of sufficiently high quality were obtained within 5 minutes. Altogether, 5 implant ruptures (17.9%) were detected, including 2 extracapsular ruptures (Patients 1 and 11) and 3 intracapsular ruptures (Patients 2, 6, and 15).
Phase 1 Ultrasound revealed signs of unilateral implant failure in both patients, and the elastography series confirmed the presence of 2 normal and 2 ruptured implants. In the 2 implants with normal ultrasound findings, elastography revealed a typical blue-green-red pattern consistent with the bull’s eye artifact11 that is familiar from cystic lesions (Figure 1A,B). Abnormal US findings in the left implant
Aesthetic Surgery Journal 35(4)
412
Table 1. Patient Characteristics, Imaging Findings, and Surgical Findings (n = 16) Study Phase
1
Patient Side
Age (yr)
Symptoms
Follow-up (yr)
Ultrasound
Elastogram
MRI
Surgical Findings
pt 1 L
30
None
8
Ruptured
Ruptured
Ruptured
Ruptured
None
8
Intact
Intact
Intact
Intact
None
14
Ruptured
Ruptured
Ruptured
Ruptured
None
14
Intact
Intact
Intact
Intact
pt 1 R pt 2 L
43
pt 2 R 2
pt 3 R
46
Pain
1
Intact
Intact
n.d.
n.d.
pt 4 R
23
None
1
Intact
Intact
n.d.
n.d.
None
1
Intact
Intact
n.d.
n.d.
Mass
2
Intact
Intact
n.d.
n.d.
None
3
Intact
Intact
n.d.
n.d.
Axillary mass
7
Ruptured
Ruptured
Ruptured
Ruptured
None
7
Intact
Intact
Intact
Intact
Pain
4
Intact
Intact
n.d.
n.d.
None
4
Intact
Intact
n.d.
n.d.
pt 4 L pt 5 L
51
pt 5 R pt 6 L
32
pt 6 R pt 7 L
24
pt 7 R pt 8 L
62
None
1
Intact
Intact
n.d.
n.d.
pt 9 L
40
Pain
1
Intact
Intact
n.d.
n.d.
None
1
Intact
Intact
n.d.
n.d.
None
8
Intact
Intact
n.d.
n.d.
Skin necrosis
8
Intact
Intact
n.d.
Intact
Swelling
10
Ruptured
Ruptured
n.d.
Ruptured
None
10
Intact
Intact
n.d.
Intact
Pain
5
Intact
Intact
n.d.
Intact
Pain (fibrosis)
5
Intact
Intact
n.d.
Intact
None
1
Intact
Intact
n.d.
n.d.
Mass
1
Intact
Intact
n.d.
n.d.
pt 9 R pt 10 L
38
pt 10 R pt 11 L
35
pt 11 R pt 12 L
34
pt 12 R pt 13 L
33
pt 13 R pt 14 L
65
Axillary mass
9
Intact
Intact
n.d.
n.d.
pt 15 L
34
None
9
Intact
Intact
n.d.
Intact
None
9
Ruptured
Ruptured
n.d.
Ruptured
Pain
3
Intact
Intact
n.d.
n.d.
pt 15 R pt 16 R
42
L, left breast; MRI, magnetic resonance imaging; n.d., not done; R, right breast.
from patient 1 included marginal irregularity and snowstorm patterning (Figure 2A). Elastography in this patient revealed a mixed pattern consisting of green and blue regions without typical layering (Figure 2B). Patient 2 presented with US abnormalities consistent with inhomogeneity of the left implant lumen (Figure 3A). The corresponding elastogram revealed an identical pattern to that seen for
patient 1 (Figure 3B). For both patients, unilateral implant rupture was confirmed by MRI and surgical findings.
Phase 2 Three unilateral implant ruptures were diagnosed in 14 patients with 24 silicone implants. Sonomorphological signs of
Stachs et al
413
Figure 1. A 24-year-old woman (Patient 7) 4 years after bilateral breast augmentation with cohesive silicone gel implants. Ultrasound (US) and elastography images of an intact breast implant in split screen mode. (A) US: noninterrupted shell and a normal reverberation pattern. (B) Elastography: blue-green-red pattern artifact.
implant rupture in Patient 6 included multiple discontinuous linear echoes in the lumen of the medial part of the left implant (Figure 4A). Real-time elastography of this region demonstrated inhomogeneity of the red-green-yellow pattern in the lower part of the prosthesis, whereas the upper part was almost blue (Figure 4B). Ultrasound findings and elastograms for Patients 11 and 15 revealed features similar to those described for Patient 6. All patients with sonographic findings indicative of intracapsular implant rupture underwent MRI (Figure 3C). Implant rupture was confirmed by surgery in all cases (Figure 4C).
DISCUSSION Since the Poly Implant Prosthesis (PIP) scandal, women who have undergone breast augmentation are now more frequently concerned about possible implant rupture. The FDA recommendation regarding repeated breast MRI for implant surveillance in asymptomatic women has not met with widespread acceptance. Although numerous studies have evaluated MRI and ultrasound in the detection of implant rupture in the past, the results need to be interpreted with caution.12-20 In a recent systematic meta-analysis conducted in 2011, Song et al identified 1175 publications related to this topic, of which only 21 were eligible for inclusion. The pooled sensitivity and specificity as calculated by logistic regression modeling were 87.0% and 89.9%,
respectively, for MRI and 60.8% and 76.3%, respectively, for US.3 However, the meta-analysis identified methodological biases in many of the studies reviewed, with those conducted in symptomatic patients leading to an overestimation of the diagnostic accuracy of MRI. Moreover, MRI is also specifically contraindicated for patients with metallic implants or suffering from claustrophobia, and the expense of the technique constitutes a further disadvantage. HRUS, carried out by an experienced examiner, is able to detect implant failure in a high percentage of cases. In 2012 Bengtson and Eaves reported 100% concordance between MRI, high-resolution US, and surgical findings in a prospective trial including 29 implants in 15 patients, half of them without symptoms.5 The conclusions drawn are limited by the small sample size and the fact that in real-time US examination it is impossible to blind the examiner to patients’ symptoms. On the other hand, realtime, dynamic US examination offers the advantage that compression or movement can be used to further explore regions of suspected rupture. The disadvantages associated with US include poorer visualization of the deep parts of the implant, and it is more examiner-dependent than MRI. Our study confirms the potential of HRUS for diagnosis of implant failure. Moreover, we have presented initial findings concerning the additional application of real-time sonoelastography for implant surveillance. The principle of sonoelastography has been described elsewhere.8,21 Currently, 2 different
414
Aesthetic Surgery Journal 35(4)
Figure 2. Extracapsular rupture of left breast implant in this 30-year-old asymptomatic woman (Patient 1) presenting for follow-up 8 years after implantation of Poly Implant Prosthesis implants. (A) Ultrasound: marginal irregularity and inhomogeneity (snowstorm pattern). At left (arrowed) a hypoechoic nodule representing extracapsular silicone. (B) Elastography: a mixed pattern without typical layering.
techniques are available: strain elasticity imaging in which the relative strain or elasticity of tissue is determined in a qualitative or semiquantitative manner, and shear wave elasticity imaging in which the velocity of a shear wave induced by an initial ultrasound pulse is measured. Because the velocity of the shear wave is proportional to stiffness, the strain modulus can be estimated (in kilopascals). Both techniques have demonstrated high sensitivity and specificity for characterizing breast lesions as benign or malignant.10,22-25 Furthermore, potential applications of elasticity imaging in distinguishing cystic lesions from solid masses have been suggested, since cystic breast lesions show a blue-green-red pattern artifact on elastography.26,27 This artifact can be explained in terms of the physical principles of elastography, as reported in detail by Cho et al.26 In brief, soft tissues with an extremely low internal echo have a very small gradient of displacement. Therefore, a linear interpolation process occurs in between the background and the soft tissue, leading to a reversed gradient of displacement in the superficial layer (coded as blue) and an overestimation in the deep layer of the cyst (coded as red). To our knowledge, our group is the first to describe this artifact in intact silicone breast implants. Additionally, our study demonstrates that this artifact cannot be reproduced in ruptured implants. On the contrary, depending on the degree of destruction present, elastography revealed a
mixed pattern without any layering. In contrast to the study by Bengtson and Eaves, which revealed differences in the pattern of rupture depending on implant durometer (gel vs form-stable implants), we did not notice any differences.5 This may be explained by the fact that only 1 patient (Patient 1) had a noncohesive gel implant whereas the other 4 patients with implant rupture had form-stable implants. Elastography was straightforward to perform, and evaluation did not require a long learning curve. The combination of high-resolution US and real-time elastography has the advantage that equivocal US findings, such as minimal inhomogeneity, can be immediately verified by real-time elastography. Conversely, in cases with a definite US diagnosis of implant failure, the need for expensive MRI examinations is obviated. Our first Phase 1 study patient underwent MRI (although extracapsular rupture had been diagnosed by HRUS) because she was the first patient in our institution to be affected by the PIP scandal. In Phase 2, for patients with certain ultrasonographic signs of extracapsular implant rupture, MRI was omitted. In our opinion, expensive MRI should be skipped in cases where the results do not affect patient management. In our study, the elastographic findings merely corroborate the ultrasonographic diagnosis. However, ultrasonographic findings in implant rupture are highly varied, and diagnostic accuracy is dependent on investigator
Stachs et al
415
Figure 3. Intracapsular rupture of left breast implant in this 43-year-old asymptomatic woman (Patient 2) 14 years after bilateral augmentation with silicone implants. (A) Ultrasound: implant showing inhomogeneity and echoic pattern. (B) Elastography: absence of typical blue-green-red layering. (C) Magnetic resonance imagine scan of Patient 2, horizontal view. The collapsed implant shell (“linguine sign”) and intracapsular gel are clearly visualized on the left side.
experience. Therefore, elastography has the potential to serve as an additional method for patients with equivocal findings or for examiners with a low level of experience.
Our study has several limitations. First, we have presented preliminary results based on a small number of patients. Second, in our study population (n = 16), the current gold
416
Aesthetic Surgery Journal 35(4)
Figure 4. Intracapsular rupture in this 32-year-old woman (Patient 6) suffering from axillary lymphadenopathy 7 years after bilateral implantation of cohesive silicone prostheses. (A) Ultrasound: echoic longitudinal stripes (stepladder sign) in the medial part of the implant. (B) Elastography: blue-colored upper part of the implant, but inhomogeneity in the lower part of the prosthesis, showing a mixed pattern of red and green areas. (C) Ruptured implant after explantation (Patient 6): the fractured shell fragment is lifted to demonstrate fragmentation of the cohesive fill material.
standard of MRI as well as surgery has not been applied to all patients, but only to those in whom implant rupture was suspected, in line with our local clinical practice.
Without performing MRI examinations on all patients, we are unable to provide data regarding the sensitivity and specificity of HRUS and elastography. From an ethical
Stachs et al
standpoint, in asymptomatic patients without US findings of implant failure, there is no indication for surgery, and expensive diagnostic modalities such as MRI with potential side effects and specific contraindications, eg, claustrophobia, metallic implants, should be avoided. Elastography did not reveal distinct differences between intra- and extracapsular rupture, but this may be explained by the preliminary nature of our study and the limited number of cases. Finally, in terms of visualizing the deep implant surface, high-frequency US as well as real-time elastography may be of limited utility for patients with large-volume implants. We do not, in fact, believe that elastography can replace HRUS, but it might be a very helpful addition for diagnosing implant rupture. To evaluate the potential of the combination of HRUS and elastography in comparison with the gold standard MRI, a larger study of patients scheduled for implant exchange (and including power analysis) is planned. However, in view of the fact that implant rupture is (fortunately) a rare complication, the recruitment period may extend over several years.
CONCLUSIONS Our results demonstrate for the first time that the combination of HRUS with real-time elastography is a valuable tool for the diagnosis of implant rupture. Sonoelastography is easy to perform and is not associated with any specific side effects. Since elastographic findings in intact and ruptured implants are distinctively different, elastography may well prove useful as an additional method for patients with equivocal US findings or for examiners with a low level of experience.
417
4.
5.
6.
7. 8.
9.
10.
11.
12.
13.
14.
Disclosures The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.
15.
Funding
16.
The authors received no financial support for the research, authorship, and publication of this article.
REFERENCES 1. 2.
3.
Cosmetic Surgery National Data Bank: Statistics 2013. Aesthet Surg J. 2014;34 (1 suppl):1S-22S. US Food and Drug Administration. Regulatory history of implants in the US. Available at: http://www.fda.gov/ MedicalDevices/ProductsandMedicalProcedures/Implants andProsthetics/BreastImplants/ucm064461.htm. Accessed May 13, 2014. Song JW, Kim HM, Bellfi LT, Chung KA. The effect of study design biases on the diagnostic accuracy of
17.
18.
19.
magnetic resonance imaging to detect silicone breast implant ruptures: a meta-analysis. Plast Reconstr Surg. 2011;127:1029-1044. Cilotti A, Marini C, Iacconi C, et al. Ultrasonographic appearance of breast implant complications. Ann Plast Surg. 2006;56:243-247. Bengtson BP, Eaves FF III. High-resolution ultrasound in the detection of silicone gel breast implant shell failure: background, in vitro studies, and early clinical results. Aesthetic Surg J. 2012;32:157-174. Lake E, Ahmad S, Dobrashian R. The sonographic appearances of breast implant rupture. Clin Radiol. 2013;68: 851-858. Garra BS. Imaging and estimation of tissue elasticity by ultrasound. Ultrasound Q. 2007;23:255-268. Itoh A, Ueno E, Tohno E, et al. Breast disease: clinical application of US elastography for diagnosis. Radiology. 2006;239:341-350. Sadigh G, Carlos RC, Neal CH, Dwamena BA. Ultrasonic differentiation of malignant from benign breast lesions: a meta-analytic comparison of elasticity and BIRADS scoring. Breast Cancer Res Treat. 2012;133:23-35. Stachs A, Hartmann S, Stubert J, et al. Differentiating between malignant and benign breast masses: factors limiting sonoelastographic strain ratio. Ultraschall Med. 2013;34:131-136. Barr RG, Lackey AE. The utility of the "bull’s-eye" artifact on breast elasticity imaging in reducing breast lesion biopsy rate. Ultrasound Q. 2011;27:151-155. Di Benedetto G, Cecchini S, Grassetti L, et al. Comparative study of breast implant rupture using mammography, sonography, and magnetic resonance imaging: correlation with surgical findings. Breast J. 2008;14:532-537. Chung KC, Wilkins EG, Beil RJ Jr, et al. Diagnosis of silicone gel breast implant rupture by ultrasonography. Plast Reconstr Surg. 1996;97:104-109. Hölmich LR, Fryzek JP, Kjøller K, et al. The diagnosis of silicone breast-implant rupture: clinical findings compared with findings at magnetic resonance imaging. Ann Plast Surg. 2005;54:583-589. Collis N, Litherland J, Enion D, Sharpe DT. Magnetic resonance imaging and explantation investigation of longterm silicone gel implant integrity. Plast Reconstr Surg. 2007;120:1401-1406. Paetau AA, McLaughlin SA, McNeil RB, et al. Capsular contracture and possible implant rupture: is magnetic resonance imaging useful? Plast Reconstr Surg. 2010;125: 830-835. Berg WA, Caskey CI, Hamper UM, et al. Single- and doublelumen silicone breast implant integrity: prospective evaluation of MR and US criteria. Radiology. 1995;197:45-52. Everson LI, Parantainen H, Detlie T, et al. Diagnosis of breast implant rupture: imaging findings and relative efficacies of imaging techniques. AJR Am J Roentgenol. 1994;163:57-60. Quinn SF, Neubauer NM, Sheley RC, et al. MR imaging of silicone breast implants: evaluation of prospective and retrospective interpretations and interobserver agreement. J Magn Reson Imaging. 1996;6:213-218.
Aesthetic Surgery Journal 35(4)
418
20.
21. 22.
23.
Maijers MC, Niessen FB, Veldhuizen JF, et al. MRI screening for silicone breast implant rupture: accuracy, interand intraobserver variability using explantation results as reference standard. Eur Radiol. 2014;24:1167-1175. Barr RG. Sonographic breast elastography – a primer. J Ultrasound Med. 2012;31:773-783. Wojcinski S, Farrokh A, Weber S, et al. Multicenter study of ultrasound real-time tissue elastography in 779 cases for the assessment of breast lesions: improved diagnostic performance by combining the BI-RADS®-US classification system with sonoelastography. Ultraschall Med. 2010;31:484-491. Cho N, Moon WK, Kim HY, et al. Sonoelastographic strain index for differentiation of benign and malignant nonpalpable breast masses. J Ultrasound Med. 2010;29:1-7.
24. Yi A, Cho N, Chang JM, et al. Sonoelastography for 1,786 non-palpable breast masses: diagnostic value in the decision to biopsy. Eur Radiol. 2012;22:1033-1040. 25. Berg WA, Cosgrove DO, Doré CJ, et al. BE1 Investigators: Shear-wave elastography improves the specificity of breast US: the BE1 multinational study of 939 masses. Radiology. 2012;262:435-449. 26. Cho N, Moon WK, Chang JM, et al. Aliasing artifact depicted on ultrasound (US)-elastography for breast cystic lesions mimicking solid masses. Acta Radiol. 2011;52: 3-7. 27. Booi RC, Carson PL, O’Donnell M, et al. Characterization of cysts using differential correlation coefficient values from two dimensional breast elastography: preliminary study. Ultrasound Med Biol. 2008;34:12-21.
