- Email: [email protected]
PIP breast implant removal: A study of 828 cases
Journal of Plastic, Reconstructive & Aesthetic Surgery (2014) 67, 302e307
PIP breast implant removal: A study of 828 cases S. Oulharj*, J. Pauchot, Y. Tropet Department of Orthopedics, Traumatology, Plastic and Reconstructive Surgery and Hand Support, Besancon University Hospital, Besancon, France Received 4 June 2013; accepted 17 December 2013
KEYWORDS PIP implants; Rupture; Silicone; Cancer
Summary In March, 2010, the French Health Products Safety Agency suspended the sale of prefilled silicone breast implants manufactured by Poly Implants Prosthe `se Prothese (PIP) because of a high failure rate and the use of an inappropriate silicone gel that did not comply with CE marking. These findings led to an international medical crisis. In France, 30,000 female patients had PIP implants. In our Department, 1150 PIP breast implants had been implanted in 630 patients since 2001. A retrospective study was conducted to define the rupture rate of these implants and the complications that arise. The women included in the study underwent implant removal from May 2010 to September 2012 for preventive or curative reasons. Data were collected from medical records that included: results of clinical examination, breast ultrasound before removal, rates of implant rupture, results of biopsy of periprosthetic capsule and pericapsule tissue and postoperative complications. A total of 828 PIP breast implants were removed in 455 patients. The rate of ruptured implants was 7.73% (64/828), corresponding to 11.6% of patients. A periprosthetic effusion was associated with rupture in 44% of cases. Breast ultrasound indicated a rupture for 87 implants; 32% were true positives and 3% were false negatives. Periprosthetic capsule biopsy demonstrated the presence of a foreign body, which seemed to be silicone, in 26% of cases and the presence of inflammation in 13% of cases. No siliconoma-type lesion was identified in the pericapsular tissue at biopsy. A total of 14 implants presented perspiration at removal. A statistically significant difference was found between the rates of rupture for texturised implants as compared to the smooth-surfaced implants. There were eight post-revisional-surgery complications (1%) and three cases of breast adenocarcinoma. The preventive explantation of PIP breast implants is justified given the high failure rate (7.73%) and given patients’ exposure to silicone gel that did not comply with CE standards in the absence of rupture, through the early perspiration of implants. ª 2013 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Hospital Jean Minjoz, Boulevard Fleming, 25000 Besancon, France. Tel.: þ33 3 81 66 78 18. E-mail address: [email protected] (S. Oulharj). 1748-6815/$ - see front matter ª 2013 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.bjps.2013.12.016
PIP breast implant removal On 29 March 2010, the French Health Products Safety Agency (AFSSAPS) suspended the sale of prefilled silicone breast implants manufactured by Poly Implants Prothe `ses (PIP), a company founded by Jean-Claude Mas in 1991, as well as the distribution, export and use of these implants. This decision followed an inspection by the AFSSAPS of the PIP premises in March 2010, which was itself due to an increasing number of incidents of ruptures of these breast implants in late 2009. The inspection revealed the use of silicone gel that did not comply with CE marking.1 A physicochemical analysis of these implants was conducted by judicial authorities and revealed a silicone gel of low quality that provoked irritation and significant heterogeneity from one prosthesis to another. The rupture test at elongation revealed fragility in the capsule of the implants. Cytotoxicity tests did not find an acute toxic effect or a genotoxic effect on tissue. Therefore, physicians were invited to monitor patients implanted with these PIP implants with a clinical examination and breast ultrasound every 6 months. Explantation of the implants was recommended in cases of suspected rupture. In December 2011, following the death of a patient with PIP implants from anaplastic large cell lymphoma (ALCL), an expert committee led by health agencies in connection with learned societies indicated that there was no significant increase in risk of cancer in women with PIP implants compared to other brands. However, the recommendations were reinforced and preventive removal of these implants was recommended for all patients, even in the absence of clinical signs of implant rupture. Several studies on PIP implants have been published, but the study populations were small, with fewer than 100 patients, and included only patients in whom implant failure was suspected at breast ultrasound. In our Department, PIP implants were placed between 2001 and March 2010 in 630 patients. Here, we conducted a retrospective study of patients who underwent implant removal from May 2010 to September 2012, in order to define the rupture rate of these implants and the complications presented.
Patients and methods This was a retrospective single-centre study. It included patients with PIP breast implanted in our Department from 2001 to March 2010 and underwent removal from May 2010 to September 2012. The removals were preventive or followed a diagnosis of implant failure. All patients were managed by the same surgeon for both placement and removal of the implants. Exclusion criteria were patients operated on by different surgeons. Initial indications for surgery were cosmetic, reconstructive or plastic. Patients were contacted by a letter informing them of the decision of AFSSAPS and the need to proceed to a control appointment with the referring surgeon. During this visit, signs of implant rupture or the presence of a periprosthetic capsule or locoregional lymph nodes were sought. A control breast ultrasound was systematically proposed to patients. In the case of ambiguous ultrasound results, an additional magnetic resonance imaging (MRI) was performed. Preventive removal of PIP implants was then routinely proposed to patients. The decision of removal or of a biannual clinical and ultrasound monitoring was made jointly with the
303 patient, according to the results of the clinical examination and further examination. We were interested in the initial placement indications for these PIP implants, the characteristics of the implants, clinical or ultrasound signs of implant failure, the time between placement and removal and the implant failure rate. During explantation, as recommended by the AFSSAPS, a biopsy of the periprosthetic capsule was routinely performed from April 2011 and a biopsy of the pericapsular tissue from January 2012. We used patient records to collect these data.
Results A total of 828 PIP implants were removed in 455 patients from May 2010 to September 2012; 398 patients had had a surgical procedure for cosmetic reasons, 29 for plastic surgery and 28 for breast reconstruction. The average age of patients at removal was 42.8 11.9 years. Patient follow-up ranged from 10 months to 10 years. All PIP implants were round in shape. The average volume was 290 cc 64.9. A total of 580 implants were high profile, 150 ultra-high profile, 81 medium profile and 17 low profile. Moreover, 711 implants were smooth surfaced and 117 were texturised. During the placement surgery, the incision was lower hemiareolar for 471 implants, axillary for 297, by mastectomy scar for 36 and sub-mammary for 24. A total of 387 implants were placed in retroglandular and 441 in retromuscular position. On physical examination, a periprosthetic capsule of Baker stage 2e3 was found in 115 cases. No periprosthetic capsule of Baker stage 4 was seen in all patients. A change in the type of breast to soft breast or swollen was found in 13 cases. Pre-placement breast ultrasound results were identified for 613 implant cases. Results indicated a suspected rupture for 87 implants and ambiguity for 31. Of these preoperative scans, 32% were true positives and 3% false negatives, with a sensitivity of ultrasound in this study of 65% and a specificity of 87%. MRI was performed in 26 cases where doubt about the rupture persisted at the ultrasound stage. The rupture was suspected in 14 cases, of which only five were true positives. The surgical removal of the implant was performed under general anaesthesia in 446 patients and under local anaesthesia for nine patients. The mean operative time was 30 14 min. During removal of PIP implants, periprosthetic effusion was found in 43 cases (5.19%), milky effusions in 16 cases and serous effusions in 27 cases. In 19 cases, the effusion was associated with implant rupture (44%). The cytology of the effusion fluid could be found in nine cases. It highlighted inflammatory fluid in two cases, and the presence of silicone particles in seven cases, of which four cases were not associated with implant rupture. Finally, 64 ruptured PIP implants were found, corresponding to an implant rupture rate of 7.73%. All ruptures were intracapsular (Figure 1). This corresponded to 11.6% of patients. In seven patients, the rupture was bilateral. For these implants, the interval between placement and removal was 6 2 years (Table 1). At removal, we noticed the presence of yellowed implants in 71 cases (8.57%) and a phenomenon of
304
S. Oulharj et al. lymphadenitis in three cases and silicone in three cases. For these three cases, lymphadenopathy was ipsilateral to intracapsular implant rupture. There were eight cases of postoperative complications (1% of patients); all were cases of breast haematoma requiring surgical evacuation. Three cases of breast adenocarcinoma were diagnosed. No periprosthetic infection was detected. We found a statistically significant difference between the rate of implant ruptures with a texturised envelope and those with a smooth-surfaced envelope (p < 0.0001). Rupture is more common with texturised (17%) than with smooth-surfaced implants (6.21%).
Discussion
Figure 1
Ruptured yellow PIP implant.
perspiration or gel bleed in 14 implants (1.69%) (Figure 2). A biopsy of periprosthetic capsule for anatomopathological examination was performed in 537 cases. It showed fibrosis in 329 cases (61%), inflammation in 67 cases (13%) of which 97% were not associated with rupture and a resorptive granuloma on foreign body which could be silicone in 141 cases (26%) of which 83% were not associated with rupture. A pericapsular tissue biopsy was performed in 269 cases and no siliconoma-type lesion was found. Seven axillary lymph nodes were biopsied and histological examination showed
Table 1
Characteristics of the 64 ruptured PIP implants.
Characteristics
Value
Mean volume of the implant Mean duration of placement Implant profile Medium High Ultra high Texture of envelope Microtexture Macrotexture Placement Low hemiareolar Axillar Submammary Implant position Retromuscular Retroglandular Capsule of Baker stage 2e3 Breast modification Pericapsular effusion Periprosthetic capsule biopsy Fibrosis Inflammation Silicone Axillary lymphadenopathy
300 cc 6 years 3 49 10 46 18 37 22 2 38 26 13 7 19 8 2 54 3
The PIP case is a medical scandal that has had a socioeconomic impact at an international level, affecting nearly 400,000 patients in 55 different countries. In France, more than 30,000 patients are carriers of these faulty breast implants. Until 2010, only two cases had been described in the literature. In 2005, Lahiri and Waters described the case of a patient who had a ruptured PIP implant after a period of 6 years, which was discovered after the onset of ipsilateral axillary lymph nodes.2 In 2007, Berry et al. described a similar case after a period of only 3 years.3 In our study, the rupture rate of PIP implants was 7.73%. The AFSSAPS, when announcing the withdrawal of PIP implants in March 2010, justified its decision with the results of a retrospective survey of six health facilities. It found rupture rates ranging from 0.37% to 11.1%, but concerned only cases of ruptured PIP implants reported before March 2010.4 Studies have since been carried out by different teams; they also identified high rates of rupture: 12% for Aktouf et al. and 10.6% for Blugermann et al.5,6 The failure rate of 7.73% observed in our series is thus in the low average range, according to the AFSSAPS. This can be explained by the choice made by our team to preventively remove these implants from May 2010 even in the absence of evidence of rupture, and before the recommendations of the AFSSAPS were made, as well as by the large number of patients included in our study. If the rupture rate is calculated using the duration of the period during which the breast implant was carried by the patient, the early rupture rate for a period of 5 years or less would be of 4.94% (Figure 3). However, this rate remains high compared to that in a series studying another brand of prostheses. A 1.1% failure rate over 6 years was found in a study of Mentor implants.7 Heden et al. found a rupture rate of 1.7% at 8 years in their study of 410 Allergan implants, and Spear et al. found a rate of 3.5% at 6 years with the same Allergan implants.8,9 The rupture rate of texturised implants in our study was found to be higher than that of smooth-surfaced implants; this difference was statistically significant (p < 0.0001) (Figure 4). This can be explained by the poor quality of the PIP implant envelope, which was demonstrated by the AFSSAPS.10 It is likely that these two different envelopes were made with a different material. Clinical examination is not very sensitive (30%) and the clinical profile ranges from complete absence of symptoms, to pain, or signs of local inflammation.11 Breast deformation may rather suggest the presence of a periprosthetic
PIP breast implant removal
305
Figure 2
Yellow PIP implant at removal.
Figure 3 Number of ruptured implants by implant year reference.
capsule of Baker stage 3e4. Further investigation is necessary and breast ultrasound remains the first-line examination, as it is cheaper, faster and accessible. An MRI is used in case of remaining doubt at ultrasound. This is the most sensitive test for the diagnosis of rupture of the prosthesis, with a sensitivity of 89% and a specificity of 97%, which increases together with other tests at the expense of sensitivity.12 In our series, ultrasound had a sensitivity of 65%, suggesting implant failure in 18% of cases, only 33% of which were true positives, with 3% of false negatives, corresponding to a normal breast ultrasound although the implant was found ruptured at removal. If we only consider
Figure 4
scans performed <3 months before removal, the sensitivity increases to 79% and the number of false negatives drops to 2.5%. MRI, which was performed in 24 cases because of ambiguous ultrasound, confirmed rupture in 14 cases, five of which were true positives. It suggests much less accuracy of MRI in the asymptomatic patient.13 However, there was no false negative at MRI. All these tests remain operator dependent and thus need to be performed in a specialised and trained radiological centre. Like any foreign body, breast implants are surrounded by a periprosthetic capsule, which is a constant and normal scarring reaction. However, in some cases the capsule undergoes abnormal constriction, whose aetiology is still unknown. This is termed a periprosthetic fibrous capsule. Periprosthetic capsules of Baker stage 2e3 were present in 13.9% of cases. However, we did not identify any link between the presence of a periprosthetic capsule and the occurrence of implant failure in our series (p Z 0.05). Silicone particles present in breast implants can migrate through the prosthetic wall after erosion or even through an intact wall; this is termed gel bleeding. This risk increases with the age of the implant, its retromuscular position or the presence of a periprosthetic capsule. Silicone particles can be transported to the lymph areas by macrophages of the reticuloendothelial system, causing a granulomatous reaction that forms lymphadenopathy that can be found in the axillar, clavicle or intra-breast areas.14 Gel bleeding was found in 2.15e11.1% of the cases and the majority was
High rupture rate of texturised implants, with a statistically significant difference (p < 0.0001).
306 reported in the 3 years following the date of placement of the PIP implant.10 The AFSSAPS has indicated in a statement that an internal PIP study showed that PIP gel at 8 weeks had a rate of transudation that was 30% greater than that of an implant filled with silicone gel, as described in the CE marking files.10 In our series, 14 implants showed intra-operative gel bleeding, but this figure is probably underestimated because it is difficult to quantify and this feature was not found in all surgical reports. At removal, 8.57% of implants had a yellow colour, but no similar cases have been reported in the literature. This colour can be explained by the poor quality of the envelope of PIP implants, which becomes permeable, and then allows the silicone gel to absorb fat-soluble components in the periprosthetic breast fatty tissue.15 This affinity of silicone with grease has already been studied by Nakamura et al.16 It could represent a pre-rupture stage of the implant: lipid infiltration of the silicone elastomer can result in structural degeneration and eventual failure.17 Seven axillary lymph nodes were biopsied. Silicone was found in three nodes with associated ipsilateral rupture. This is a known and described complication of breast augmentation. The presence of silicone in the lymph areas is frequently seen, although a significant link has been reported in the literature between the presence of silicone and the development of autoimmune diseases.18 A periprosthetic effusion was present in 43 cases, and in 19 cases it was associated with an implant rupture, followed by a milky effusion. Silicone particles were present in the cytological analysis in nine cases, four of which were not associated with a rupture. Histological examination of periprosthetic capsule biopsies showed inflammation in 13% of the cases, 97% without rupture, and silicone in 26% of the cases, 83% without rupture. Thus, silicone particles were present outside the implant even in the absence of rupture, which would cause the appearance of an effusion and periprosthetic inflammation. Patients are exposed to improper silicone gel even in the absence of rupture. The presence of silicone particles outside the implant and in the absence of rupture confirms the gel bleed, which is abnormally high for these defective implants. This is an early exposure source to PIP gel, and one that is hardly detectable by clinical or imaging. This gel diffusion through the defective envelope of the implant can weaken the texture and also explain the higher rate of rupture. The AFSSAPS has recommended preventive removal of PIP implants since December 2011 after evaluating the riskebenefit of the intervention, but our complication rate of 1% shows that the risk is very low.19 We had three cases of breast adenocarcinoma in patients aged 50, 51 and 55. In all three cases, the PIP implants had been placed for aesthetic reasons. The duration of placement was 1, 3 and 8 years, respectively. These patients had not presented with implant failure, periprosthetic effusion or locoregional lymph nodes. Periprosthetic capsule biopsies did not indicate silicone or inflammation. Two patients underwent removal of the implant with a mastectomy, and the third a change of implant with a tumorectomy. For one of these patients, the diagnosis of cancer was made during removal on a thickened capsule biopsy; the preoperative mammogram was normal. In December 2011, following the death of
S. Oulharj et al. a patient with a PIP implant from ALCL, the AFSSAPS decided to recommend the preventive removal of PIP implants.19 The Medicines and Healthcare Products Regulatory Agency (MHRA) in England has not made this decision because it could not identify a risk of cancer in these patients.20 De Jong et al. stated that there was a very rare correlation of one in a million between ALCL and breast implants, but no causal link.21 A literature review covering all articles published from 1991 to 2001 did not establish a link between breast implants and ALCL.22 It is therefore necessary to reassure patients of the low incidence of reported cases but to remain vigilant in cases of late periprosthetic effusion or swelling of the breast.
Conclusions Many anxious patients have demanded the removal of their PIP implants before ultrasound results and, in view of the lack of knowledge on the consequences of prolonged contact with improper PIP gel, our team decided on preventive removal before the recommendations of the AFSSAPS were made. This seems all the more justified in the context of a high 7.73% prosthetic rupture rate and the early and significant PIP gel bleeding found in our study. These complications lead to exposure of patients to bad gel, which is often asymptomatic, even in the absence of prosthetic failure. We must provide clear and honest information to our patients. Ultrasounds constitute the first-line examination for the diagnosis of prosthetic failure, but MRI, which did not detect any false negatives in our study, seems to be the most appropriate test for the annual monitoring of patients who do not wish to undergo immediate preventive removal of their PIP implants. Preventive removal and curative surgery in France has been supported by the National Health Insurance since December 2011. The placement of new implants is allowed in the same surgical act, but is not supported. This change in management facilitates the decision making of our patients. However, this system is not yet in place in many European countries.
Ethical approval Not required.
Funding None.
Conflicts of interest None declared.
References 1. GMED. Directive 2003/12/CE de la commission du 3 fe ´vrier 2003 concernant la reclassification des implants mammaires dans le cadre de la directive 93/42/CEE relative aux dispositifs me http://www.gmed.fr/pdf/directive_03_12.pdf ´dicaux. [Accessibility verified October 9, 2013].
PIP breast implant removal 2. Lahiri A, Waters R. Locoregional silicone spread after high cohesive gel silicone implant rupture. J Plast Reconstr Aesthet Surg 2006;59:885e6. 3. Berry RB. Rupture of PIP breast implants. J Plast Reconstr Aesthet Surg 2007;60:967e8. 4. AFSSAPS. Re ´sultats des tests sur les implants mammaires a ` base de gel de silicone de la socie ´te ´ Poly Implant Prothe `se. http:// ansm.sante.fr/content/download/27936/370699/version/2/ [Accessibility verified April 9, 2013]. 5. Aktouf A, Auquit-Auckbur I, Coquerel-Beghin D, Delpierre V, Milliez P-Y. Breast augmentation by poly implant prothe `ses silicone implants: retrospective study about 99 patients. Rupture analysis and management. Ann Chir Plast Esthe´tique 2012;57:558e66. 6. Blugerman G, Schavelzon D, Chomyzyn A, et al. Rupture of poly implant prothe `se (PIP) breast implants: experience with the removal of 884 implants. Eur J Plast Surg http://rd.springer. com/article/10.1007%2Fs00238-013-0813-8#page-2 [Accessibility verified April 11, 2013]. 7. Cunningham B, McCue J. Safety and effectiveness of mentor’s memorygel implants at 6 years. Aesthetic Plast Surg 2009;33: 440e4. 8. Hede ´n P, Bronz G, Elberg JJ, et al. Long-term safety and effectiveness of style 410 highly cohesive silicone breast implants. Aesthetic Plast Surg 2009;33:430e6. 9. Spear SL, Murphy DK, Slicton A, Walker PS. Inamed silicone breast implant core study results at 6 years. Plast Reconstr Surg 2007;120:8Se16S. 10. AFSSAPS. Synthe `se des donne ´es de vigilance sur les prothe `ses mammaires implantables PIP. http://ansm.sante.fr/content/ download/37705/495687/version/2/file/Syntheses-donneesvigilance-avril2011.pdf [Accessibility verified April 11, 2013]. 11. Ho ¨lmich 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: 583e9. 12. Ho ¨lmich LR, Vejborg I, Conrad C, Sletting S, McLaughlin JK. The diagnosis of breast implant rupture: MRI findings compared with findings at explantation. Eur J Radiol 2005;53:213e25.
307 13. Song JW, Kim HM, Bellfi LT, Chung KC. The effect of study design biases on the diagnostic accuracy of magnetic resonance imaging for detecting silicone breast implant ruptures: a meta-analysis. Plast Reconstr Surg 2011 Mar;127(3):1029e44. 14. Adams ST, Cox J, Rao GS. Axillary silicone lymphadenopathy presenting with a lump and altered sensation in the breast: a case report. J Med Case Reports 2009;3:6442. 15. Beretta G, Malacco M. Chemical and physicochemical properties of the high cohesive silicone gel from poly implant prothe `se (PIP) breast prostheses after explantation: a preliminary, comparative analytical investigation. J Pharm Biomed Anal 2013;78:75e82. 16. Nakamura K, Refojo MF, Crabtree DV, Pastor J, Leong FL. Ocular toxicity of low-molecular-weight components of silicone and fluorosilicone oils. Invest Ophthalmol Vis Sci 1991;32: 3007e20. 17. Adams Jr WP, Robinson Jr JB, Rohrich RJ. Lipid infiltration as a possible biologic cause of silicone gel breast implant aging. Plast Reconstr Surg 1998 Jan;101(1):64e8. 18. Dragoumis DM, Assimaki AS, Vrizas TI, Tsiftsoglou AP. Axillary silicone lymphadenopathy secondary to augmentation mammaplasty. Indian J Plast Surg Off Publ Assoc Plast Surg India 2010;43:206e9. 19. Institut National du Cancer. ropositions de conduite a ` tenir pour les femmes porteuses de prothe `ses mammaires PIP: avis d’experts. http://ansm.sante.fr/content/download/37708/ 495717/version/2/file/PIP_Conduite-a-tenir-Experts_Inca23122011.pdf [Accessibility verified April 9, 2013]. 20. Medicines and Healthcare products Regulatory Agency (MHRA). PIP breast implants e UK medical devices regulator says no evidence to support routine removal. http://www.mhra.gov. uk/NewsCentre/CON137935 [Accessibility verified April 9, 2013]. 21. De Jong D, Vasmel WLE, de Boer JP, et al. Anaplastic large-cell lymphoma in women with breast implants. JAMA e J Am Med Assoc 2008;300:2030e5. 22. Kim B, Roth C, Chung KC, et al. Anaplastic large cell lymphoma and breast implants: a systematic review. Plast Reconstr Surg 2011;127:2141e50.
PIP breast implant removal: A study of 828 cases S. Oulharj*, J. Pauchot, Y. Tropet Department of Orthopedics, Traumatology, Plastic and Reconstructive Surgery and Hand Support, Besancon University Hospital, Besancon, France Received 4 June 2013; accepted 17 December 2013
KEYWORDS PIP implants; Rupture; Silicone; Cancer
Summary In March, 2010, the French Health Products Safety Agency suspended the sale of prefilled silicone breast implants manufactured by Poly Implants Prosthe `se Prothese (PIP) because of a high failure rate and the use of an inappropriate silicone gel that did not comply with CE marking. These findings led to an international medical crisis. In France, 30,000 female patients had PIP implants. In our Department, 1150 PIP breast implants had been implanted in 630 patients since 2001. A retrospective study was conducted to define the rupture rate of these implants and the complications that arise. The women included in the study underwent implant removal from May 2010 to September 2012 for preventive or curative reasons. Data were collected from medical records that included: results of clinical examination, breast ultrasound before removal, rates of implant rupture, results of biopsy of periprosthetic capsule and pericapsule tissue and postoperative complications. A total of 828 PIP breast implants were removed in 455 patients. The rate of ruptured implants was 7.73% (64/828), corresponding to 11.6% of patients. A periprosthetic effusion was associated with rupture in 44% of cases. Breast ultrasound indicated a rupture for 87 implants; 32% were true positives and 3% were false negatives. Periprosthetic capsule biopsy demonstrated the presence of a foreign body, which seemed to be silicone, in 26% of cases and the presence of inflammation in 13% of cases. No siliconoma-type lesion was identified in the pericapsular tissue at biopsy. A total of 14 implants presented perspiration at removal. A statistically significant difference was found between the rates of rupture for texturised implants as compared to the smooth-surfaced implants. There were eight post-revisional-surgery complications (1%) and three cases of breast adenocarcinoma. The preventive explantation of PIP breast implants is justified given the high failure rate (7.73%) and given patients’ exposure to silicone gel that did not comply with CE standards in the absence of rupture, through the early perspiration of implants. ª 2013 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Hospital Jean Minjoz, Boulevard Fleming, 25000 Besancon, France. Tel.: þ33 3 81 66 78 18. E-mail address: [email protected] (S. Oulharj). 1748-6815/$ - see front matter ª 2013 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.bjps.2013.12.016
PIP breast implant removal On 29 March 2010, the French Health Products Safety Agency (AFSSAPS) suspended the sale of prefilled silicone breast implants manufactured by Poly Implants Prothe `ses (PIP), a company founded by Jean-Claude Mas in 1991, as well as the distribution, export and use of these implants. This decision followed an inspection by the AFSSAPS of the PIP premises in March 2010, which was itself due to an increasing number of incidents of ruptures of these breast implants in late 2009. The inspection revealed the use of silicone gel that did not comply with CE marking.1 A physicochemical analysis of these implants was conducted by judicial authorities and revealed a silicone gel of low quality that provoked irritation and significant heterogeneity from one prosthesis to another. The rupture test at elongation revealed fragility in the capsule of the implants. Cytotoxicity tests did not find an acute toxic effect or a genotoxic effect on tissue. Therefore, physicians were invited to monitor patients implanted with these PIP implants with a clinical examination and breast ultrasound every 6 months. Explantation of the implants was recommended in cases of suspected rupture. In December 2011, following the death of a patient with PIP implants from anaplastic large cell lymphoma (ALCL), an expert committee led by health agencies in connection with learned societies indicated that there was no significant increase in risk of cancer in women with PIP implants compared to other brands. However, the recommendations were reinforced and preventive removal of these implants was recommended for all patients, even in the absence of clinical signs of implant rupture. Several studies on PIP implants have been published, but the study populations were small, with fewer than 100 patients, and included only patients in whom implant failure was suspected at breast ultrasound. In our Department, PIP implants were placed between 2001 and March 2010 in 630 patients. Here, we conducted a retrospective study of patients who underwent implant removal from May 2010 to September 2012, in order to define the rupture rate of these implants and the complications presented.
Patients and methods This was a retrospective single-centre study. It included patients with PIP breast implanted in our Department from 2001 to March 2010 and underwent removal from May 2010 to September 2012. The removals were preventive or followed a diagnosis of implant failure. All patients were managed by the same surgeon for both placement and removal of the implants. Exclusion criteria were patients operated on by different surgeons. Initial indications for surgery were cosmetic, reconstructive or plastic. Patients were contacted by a letter informing them of the decision of AFSSAPS and the need to proceed to a control appointment with the referring surgeon. During this visit, signs of implant rupture or the presence of a periprosthetic capsule or locoregional lymph nodes were sought. A control breast ultrasound was systematically proposed to patients. In the case of ambiguous ultrasound results, an additional magnetic resonance imaging (MRI) was performed. Preventive removal of PIP implants was then routinely proposed to patients. The decision of removal or of a biannual clinical and ultrasound monitoring was made jointly with the
303 patient, according to the results of the clinical examination and further examination. We were interested in the initial placement indications for these PIP implants, the characteristics of the implants, clinical or ultrasound signs of implant failure, the time between placement and removal and the implant failure rate. During explantation, as recommended by the AFSSAPS, a biopsy of the periprosthetic capsule was routinely performed from April 2011 and a biopsy of the pericapsular tissue from January 2012. We used patient records to collect these data.
Results A total of 828 PIP implants were removed in 455 patients from May 2010 to September 2012; 398 patients had had a surgical procedure for cosmetic reasons, 29 for plastic surgery and 28 for breast reconstruction. The average age of patients at removal was 42.8 11.9 years. Patient follow-up ranged from 10 months to 10 years. All PIP implants were round in shape. The average volume was 290 cc 64.9. A total of 580 implants were high profile, 150 ultra-high profile, 81 medium profile and 17 low profile. Moreover, 711 implants were smooth surfaced and 117 were texturised. During the placement surgery, the incision was lower hemiareolar for 471 implants, axillary for 297, by mastectomy scar for 36 and sub-mammary for 24. A total of 387 implants were placed in retroglandular and 441 in retromuscular position. On physical examination, a periprosthetic capsule of Baker stage 2e3 was found in 115 cases. No periprosthetic capsule of Baker stage 4 was seen in all patients. A change in the type of breast to soft breast or swollen was found in 13 cases. Pre-placement breast ultrasound results were identified for 613 implant cases. Results indicated a suspected rupture for 87 implants and ambiguity for 31. Of these preoperative scans, 32% were true positives and 3% false negatives, with a sensitivity of ultrasound in this study of 65% and a specificity of 87%. MRI was performed in 26 cases where doubt about the rupture persisted at the ultrasound stage. The rupture was suspected in 14 cases, of which only five were true positives. The surgical removal of the implant was performed under general anaesthesia in 446 patients and under local anaesthesia for nine patients. The mean operative time was 30 14 min. During removal of PIP implants, periprosthetic effusion was found in 43 cases (5.19%), milky effusions in 16 cases and serous effusions in 27 cases. In 19 cases, the effusion was associated with implant rupture (44%). The cytology of the effusion fluid could be found in nine cases. It highlighted inflammatory fluid in two cases, and the presence of silicone particles in seven cases, of which four cases were not associated with implant rupture. Finally, 64 ruptured PIP implants were found, corresponding to an implant rupture rate of 7.73%. All ruptures were intracapsular (Figure 1). This corresponded to 11.6% of patients. In seven patients, the rupture was bilateral. For these implants, the interval between placement and removal was 6 2 years (Table 1). At removal, we noticed the presence of yellowed implants in 71 cases (8.57%) and a phenomenon of
304
S. Oulharj et al. lymphadenitis in three cases and silicone in three cases. For these three cases, lymphadenopathy was ipsilateral to intracapsular implant rupture. There were eight cases of postoperative complications (1% of patients); all were cases of breast haematoma requiring surgical evacuation. Three cases of breast adenocarcinoma were diagnosed. No periprosthetic infection was detected. We found a statistically significant difference between the rate of implant ruptures with a texturised envelope and those with a smooth-surfaced envelope (p < 0.0001). Rupture is more common with texturised (17%) than with smooth-surfaced implants (6.21%).
Discussion
Figure 1
Ruptured yellow PIP implant.
perspiration or gel bleed in 14 implants (1.69%) (Figure 2). A biopsy of periprosthetic capsule for anatomopathological examination was performed in 537 cases. It showed fibrosis in 329 cases (61%), inflammation in 67 cases (13%) of which 97% were not associated with rupture and a resorptive granuloma on foreign body which could be silicone in 141 cases (26%) of which 83% were not associated with rupture. A pericapsular tissue biopsy was performed in 269 cases and no siliconoma-type lesion was found. Seven axillary lymph nodes were biopsied and histological examination showed
Table 1
Characteristics of the 64 ruptured PIP implants.
Characteristics
Value
Mean volume of the implant Mean duration of placement Implant profile Medium High Ultra high Texture of envelope Microtexture Macrotexture Placement Low hemiareolar Axillar Submammary Implant position Retromuscular Retroglandular Capsule of Baker stage 2e3 Breast modification Pericapsular effusion Periprosthetic capsule biopsy Fibrosis Inflammation Silicone Axillary lymphadenopathy
300 cc 6 years 3 49 10 46 18 37 22 2 38 26 13 7 19 8 2 54 3
The PIP case is a medical scandal that has had a socioeconomic impact at an international level, affecting nearly 400,000 patients in 55 different countries. In France, more than 30,000 patients are carriers of these faulty breast implants. Until 2010, only two cases had been described in the literature. In 2005, Lahiri and Waters described the case of a patient who had a ruptured PIP implant after a period of 6 years, which was discovered after the onset of ipsilateral axillary lymph nodes.2 In 2007, Berry et al. described a similar case after a period of only 3 years.3 In our study, the rupture rate of PIP implants was 7.73%. The AFSSAPS, when announcing the withdrawal of PIP implants in March 2010, justified its decision with the results of a retrospective survey of six health facilities. It found rupture rates ranging from 0.37% to 11.1%, but concerned only cases of ruptured PIP implants reported before March 2010.4 Studies have since been carried out by different teams; they also identified high rates of rupture: 12% for Aktouf et al. and 10.6% for Blugermann et al.5,6 The failure rate of 7.73% observed in our series is thus in the low average range, according to the AFSSAPS. This can be explained by the choice made by our team to preventively remove these implants from May 2010 even in the absence of evidence of rupture, and before the recommendations of the AFSSAPS were made, as well as by the large number of patients included in our study. If the rupture rate is calculated using the duration of the period during which the breast implant was carried by the patient, the early rupture rate for a period of 5 years or less would be of 4.94% (Figure 3). However, this rate remains high compared to that in a series studying another brand of prostheses. A 1.1% failure rate over 6 years was found in a study of Mentor implants.7 Heden et al. found a rupture rate of 1.7% at 8 years in their study of 410 Allergan implants, and Spear et al. found a rate of 3.5% at 6 years with the same Allergan implants.8,9 The rupture rate of texturised implants in our study was found to be higher than that of smooth-surfaced implants; this difference was statistically significant (p < 0.0001) (Figure 4). This can be explained by the poor quality of the PIP implant envelope, which was demonstrated by the AFSSAPS.10 It is likely that these two different envelopes were made with a different material. Clinical examination is not very sensitive (30%) and the clinical profile ranges from complete absence of symptoms, to pain, or signs of local inflammation.11 Breast deformation may rather suggest the presence of a periprosthetic
PIP breast implant removal
305
Figure 2
Yellow PIP implant at removal.
Figure 3 Number of ruptured implants by implant year reference.
capsule of Baker stage 3e4. Further investigation is necessary and breast ultrasound remains the first-line examination, as it is cheaper, faster and accessible. An MRI is used in case of remaining doubt at ultrasound. This is the most sensitive test for the diagnosis of rupture of the prosthesis, with a sensitivity of 89% and a specificity of 97%, which increases together with other tests at the expense of sensitivity.12 In our series, ultrasound had a sensitivity of 65%, suggesting implant failure in 18% of cases, only 33% of which were true positives, with 3% of false negatives, corresponding to a normal breast ultrasound although the implant was found ruptured at removal. If we only consider
Figure 4
scans performed <3 months before removal, the sensitivity increases to 79% and the number of false negatives drops to 2.5%. MRI, which was performed in 24 cases because of ambiguous ultrasound, confirmed rupture in 14 cases, five of which were true positives. It suggests much less accuracy of MRI in the asymptomatic patient.13 However, there was no false negative at MRI. All these tests remain operator dependent and thus need to be performed in a specialised and trained radiological centre. Like any foreign body, breast implants are surrounded by a periprosthetic capsule, which is a constant and normal scarring reaction. However, in some cases the capsule undergoes abnormal constriction, whose aetiology is still unknown. This is termed a periprosthetic fibrous capsule. Periprosthetic capsules of Baker stage 2e3 were present in 13.9% of cases. However, we did not identify any link between the presence of a periprosthetic capsule and the occurrence of implant failure in our series (p Z 0.05). Silicone particles present in breast implants can migrate through the prosthetic wall after erosion or even through an intact wall; this is termed gel bleeding. This risk increases with the age of the implant, its retromuscular position or the presence of a periprosthetic capsule. Silicone particles can be transported to the lymph areas by macrophages of the reticuloendothelial system, causing a granulomatous reaction that forms lymphadenopathy that can be found in the axillar, clavicle or intra-breast areas.14 Gel bleeding was found in 2.15e11.1% of the cases and the majority was
High rupture rate of texturised implants, with a statistically significant difference (p < 0.0001).
306 reported in the 3 years following the date of placement of the PIP implant.10 The AFSSAPS has indicated in a statement that an internal PIP study showed that PIP gel at 8 weeks had a rate of transudation that was 30% greater than that of an implant filled with silicone gel, as described in the CE marking files.10 In our series, 14 implants showed intra-operative gel bleeding, but this figure is probably underestimated because it is difficult to quantify and this feature was not found in all surgical reports. At removal, 8.57% of implants had a yellow colour, but no similar cases have been reported in the literature. This colour can be explained by the poor quality of the envelope of PIP implants, which becomes permeable, and then allows the silicone gel to absorb fat-soluble components in the periprosthetic breast fatty tissue.15 This affinity of silicone with grease has already been studied by Nakamura et al.16 It could represent a pre-rupture stage of the implant: lipid infiltration of the silicone elastomer can result in structural degeneration and eventual failure.17 Seven axillary lymph nodes were biopsied. Silicone was found in three nodes with associated ipsilateral rupture. This is a known and described complication of breast augmentation. The presence of silicone in the lymph areas is frequently seen, although a significant link has been reported in the literature between the presence of silicone and the development of autoimmune diseases.18 A periprosthetic effusion was present in 43 cases, and in 19 cases it was associated with an implant rupture, followed by a milky effusion. Silicone particles were present in the cytological analysis in nine cases, four of which were not associated with a rupture. Histological examination of periprosthetic capsule biopsies showed inflammation in 13% of the cases, 97% without rupture, and silicone in 26% of the cases, 83% without rupture. Thus, silicone particles were present outside the implant even in the absence of rupture, which would cause the appearance of an effusion and periprosthetic inflammation. Patients are exposed to improper silicone gel even in the absence of rupture. The presence of silicone particles outside the implant and in the absence of rupture confirms the gel bleed, which is abnormally high for these defective implants. This is an early exposure source to PIP gel, and one that is hardly detectable by clinical or imaging. This gel diffusion through the defective envelope of the implant can weaken the texture and also explain the higher rate of rupture. The AFSSAPS has recommended preventive removal of PIP implants since December 2011 after evaluating the riskebenefit of the intervention, but our complication rate of 1% shows that the risk is very low.19 We had three cases of breast adenocarcinoma in patients aged 50, 51 and 55. In all three cases, the PIP implants had been placed for aesthetic reasons. The duration of placement was 1, 3 and 8 years, respectively. These patients had not presented with implant failure, periprosthetic effusion or locoregional lymph nodes. Periprosthetic capsule biopsies did not indicate silicone or inflammation. Two patients underwent removal of the implant with a mastectomy, and the third a change of implant with a tumorectomy. For one of these patients, the diagnosis of cancer was made during removal on a thickened capsule biopsy; the preoperative mammogram was normal. In December 2011, following the death of
S. Oulharj et al. a patient with a PIP implant from ALCL, the AFSSAPS decided to recommend the preventive removal of PIP implants.19 The Medicines and Healthcare Products Regulatory Agency (MHRA) in England has not made this decision because it could not identify a risk of cancer in these patients.20 De Jong et al. stated that there was a very rare correlation of one in a million between ALCL and breast implants, but no causal link.21 A literature review covering all articles published from 1991 to 2001 did not establish a link between breast implants and ALCL.22 It is therefore necessary to reassure patients of the low incidence of reported cases but to remain vigilant in cases of late periprosthetic effusion or swelling of the breast.
Conclusions Many anxious patients have demanded the removal of their PIP implants before ultrasound results and, in view of the lack of knowledge on the consequences of prolonged contact with improper PIP gel, our team decided on preventive removal before the recommendations of the AFSSAPS were made. This seems all the more justified in the context of a high 7.73% prosthetic rupture rate and the early and significant PIP gel bleeding found in our study. These complications lead to exposure of patients to bad gel, which is often asymptomatic, even in the absence of prosthetic failure. We must provide clear and honest information to our patients. Ultrasounds constitute the first-line examination for the diagnosis of prosthetic failure, but MRI, which did not detect any false negatives in our study, seems to be the most appropriate test for the annual monitoring of patients who do not wish to undergo immediate preventive removal of their PIP implants. Preventive removal and curative surgery in France has been supported by the National Health Insurance since December 2011. The placement of new implants is allowed in the same surgical act, but is not supported. This change in management facilitates the decision making of our patients. However, this system is not yet in place in many European countries.
Ethical approval Not required.
Funding None.
Conflicts of interest None declared.
References 1. GMED. Directive 2003/12/CE de la commission du 3 fe ´vrier 2003 concernant la reclassification des implants mammaires dans le cadre de la directive 93/42/CEE relative aux dispositifs me http://www.gmed.fr/pdf/directive_03_12.pdf ´dicaux. [Accessibility verified October 9, 2013].
PIP breast implant removal 2. Lahiri A, Waters R. Locoregional silicone spread after high cohesive gel silicone implant rupture. J Plast Reconstr Aesthet Surg 2006;59:885e6. 3. Berry RB. Rupture of PIP breast implants. J Plast Reconstr Aesthet Surg 2007;60:967e8. 4. AFSSAPS. Re ´sultats des tests sur les implants mammaires a ` base de gel de silicone de la socie ´te ´ Poly Implant Prothe `se. http:// ansm.sante.fr/content/download/27936/370699/version/2/ [Accessibility verified April 9, 2013]. 5. Aktouf A, Auquit-Auckbur I, Coquerel-Beghin D, Delpierre V, Milliez P-Y. Breast augmentation by poly implant prothe `ses silicone implants: retrospective study about 99 patients. Rupture analysis and management. Ann Chir Plast Esthe´tique 2012;57:558e66. 6. Blugerman G, Schavelzon D, Chomyzyn A, et al. Rupture of poly implant prothe `se (PIP) breast implants: experience with the removal of 884 implants. Eur J Plast Surg http://rd.springer. com/article/10.1007%2Fs00238-013-0813-8#page-2 [Accessibility verified April 11, 2013]. 7. Cunningham B, McCue J. Safety and effectiveness of mentor’s memorygel implants at 6 years. Aesthetic Plast Surg 2009;33: 440e4. 8. Hede ´n P, Bronz G, Elberg JJ, et al. Long-term safety and effectiveness of style 410 highly cohesive silicone breast implants. Aesthetic Plast Surg 2009;33:430e6. 9. Spear SL, Murphy DK, Slicton A, Walker PS. Inamed silicone breast implant core study results at 6 years. Plast Reconstr Surg 2007;120:8Se16S. 10. AFSSAPS. Synthe `se des donne ´es de vigilance sur les prothe `ses mammaires implantables PIP. http://ansm.sante.fr/content/ download/37705/495687/version/2/file/Syntheses-donneesvigilance-avril2011.pdf [Accessibility verified April 11, 2013]. 11. Ho ¨lmich 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: 583e9. 12. Ho ¨lmich LR, Vejborg I, Conrad C, Sletting S, McLaughlin JK. The diagnosis of breast implant rupture: MRI findings compared with findings at explantation. Eur J Radiol 2005;53:213e25.
307 13. Song JW, Kim HM, Bellfi LT, Chung KC. The effect of study design biases on the diagnostic accuracy of magnetic resonance imaging for detecting silicone breast implant ruptures: a meta-analysis. Plast Reconstr Surg 2011 Mar;127(3):1029e44. 14. Adams ST, Cox J, Rao GS. Axillary silicone lymphadenopathy presenting with a lump and altered sensation in the breast: a case report. J Med Case Reports 2009;3:6442. 15. Beretta G, Malacco M. Chemical and physicochemical properties of the high cohesive silicone gel from poly implant prothe `se (PIP) breast prostheses after explantation: a preliminary, comparative analytical investigation. J Pharm Biomed Anal 2013;78:75e82. 16. Nakamura K, Refojo MF, Crabtree DV, Pastor J, Leong FL. Ocular toxicity of low-molecular-weight components of silicone and fluorosilicone oils. Invest Ophthalmol Vis Sci 1991;32: 3007e20. 17. Adams Jr WP, Robinson Jr JB, Rohrich RJ. Lipid infiltration as a possible biologic cause of silicone gel breast implant aging. Plast Reconstr Surg 1998 Jan;101(1):64e8. 18. Dragoumis DM, Assimaki AS, Vrizas TI, Tsiftsoglou AP. Axillary silicone lymphadenopathy secondary to augmentation mammaplasty. Indian J Plast Surg Off Publ Assoc Plast Surg India 2010;43:206e9. 19. Institut National du Cancer. ropositions de conduite a ` tenir pour les femmes porteuses de prothe `ses mammaires PIP: avis d’experts. http://ansm.sante.fr/content/download/37708/ 495717/version/2/file/PIP_Conduite-a-tenir-Experts_Inca23122011.pdf [Accessibility verified April 9, 2013]. 20. Medicines and Healthcare products Regulatory Agency (MHRA). PIP breast implants e UK medical devices regulator says no evidence to support routine removal. http://www.mhra.gov. uk/NewsCentre/CON137935 [Accessibility verified April 9, 2013]. 21. De Jong D, Vasmel WLE, de Boer JP, et al. Anaplastic large-cell lymphoma in women with breast implants. JAMA e J Am Med Assoc 2008;300:2030e5. 22. Kim B, Roth C, Chung KC, et al. Anaplastic large cell lymphoma and breast implants: a systematic review. Plast Reconstr Surg 2011;127:2141e50.