SIRG minutes - 22 November 2002

Silicone Gel Breast Implants: Independent Review Group
Meeting of 22nd November 2002 Hannibal House, Elephant and Castle, London

Present:
     Professor R Sturrock (Chairman) (RS)
     Professor V Harpwood (VH)
     Professor A Silman (AS)
     Mr T Milward (TM)
     Ms N Ratcliffe (NR)
     Dr I Kimber (IK)

Apologies:
     Professor D London
     Professor A Cooke
     Ms K Murphy
     Professor R Walker

Secretariat:
     Dr A Austin
     Dr S Ludgate
     Mr J Tinkler
     Miss E Taiwo
     Dr S Bourne

MDA
     Dr S Costigan (item 7 only)

1. Welcome
The Chairman welcomed members of the Group and congratulated Professor Harpwood on her elevation to Chair at Cardiff University. He also welcomed the return of Mr Milward following his recovery from a car accident.

2. Minutes of the last meeting
The draft minutes [SIRG(02)058] of the previous meeting, held on the 23rd May 2002 were agreed as a correct record of the meeting.

3. Matters arising
a) General/update on progress
The Group was informed that the Medical Devices Agency's (MDA's) breast implant update is now available on the MDA's website.

b) Group conclusion from Lori Brown Teleconference
The Chair reminded the Group that Lori Brown's article (2001) had concluded that fibromyalgia was more common in women with extracapsular rupture of their implants, The Group agreed that it was valuable study as it compared women with and without rupture. It was agreed that some reassurance was gained as the incidence of self-reported symptoms was similar in both groups (those with unruptured and ruptured breast implants). It was noted that although the self-reported symptoms were similar in both groups the reports of fibromyalgia was higher in the group with extracapsular rupture, which may reflect the difficulties in diagnosing the condition.

4. Declaration of Interest
The Groups reviewed the declarations of interest document [SIRG(02) 59]. This document would be finalised and placed on the website. TM requested that the last sentence of his entrey should be removed as he is no longer carrying out any surgery. NR has some amendments to her entry, which she will forward to the Secretariat. She also stated that "none" should be entered under her conflicts of interest.

It was noted that membership of the IRG would not preclude individuals from writing a response to published articles where they had the appropriate expertise and would do this as part of their normal professional activity.

5. Update on the progress of the recommendations from the Independent Review Group report [SIRG(02)60]
Recommendation 1 - The revised DH information booklet contains a section which indicates that the risk of breast cancer is decreased in women with silicone breast implants. The Group had received some feedback on this. This statement was based on a large study that had found that cancer risk was reduced in women with implants. This and other comments on the booklet would be consedered further when the next revision takes place.

Recommendation 2 - It was noted that this recommendation has been progressed as far as posible and that it any future action would be taken by the Advertising Standards Authority.

Recommendation 3 - It was noted that the Royal College of Surgeons of England's clinical practice guidelines for breast implants was in the final stages of preparation and that individual members of the IRG had commented on the draft. The final document will be circulated to the Group as and when it became available.

Recommendation 4 - The IRG noted that the issue of consent to surgery was now being considered as a general issue across the whole of the NHS.

Recommendation 5 - The IRG noted that this was the responsibility of the Care Standards Agency, which has now been become part of CHI (Commission for Health Improvement).

Recommendation 6 - The possibility of obtaining a Section 60 exemption (which would allow data to be recorded without consent) was raised. This would allow compulsory registration of breast implants with the NBIR. However it was noted that this was only possible for the cancer registry due to the overwhelming public health interest. The Breast Implant Registry does not fulfil the necessary criteria.

Recommendation 7 - The Secretariat informed the group that a new poster (advising clinicians of reportable events in relation to breast implants) and a dedicated adverse incident form were now available. MDA had promoted these documents at the British Association of Plastic Surgeons Annual Conference (November 2002).

Recommendation 8 - SIRG(02)61 - Dr Ridgway (UCL) had submitted a proposal for research into microbiological investigation of subclinical infection. It was noted that it would be difficult to carry out this work without invasive procedures and that (due to the large study population required) ethics approval would be difficult to obtain. The IRG agreed that this project is not feasible at this point in time and taht options would be reviewed again in the future. The Chair reported that the pilot study using NBIR data was underway and that their report is expected in late 2003.

Recommendation 9 - The Secretariat informed the group that a paper on breast implants and mammography had been added to the website.

6. Scientific papers [SIRG (02)63 to SIRG (02)75 and SIRG (02)82]
Members presented summaries of their findings from the papers they had reviewed, which included recently published or submitted scientific papers. Details of the reviews and the published papers can be found in the annex to these minutes.

7. Platinum toxicity review - update
IK updated the Group on the progress of the report and informed members that relevant issues had been comprehensively addressed, although there were some points that need to be resolved in further detail. These included the issue of platinum sensitization, as one study in guinea pigs had been positive, contrary to previous indications. It was noted that this is an old test and may not be the most appropriate by modern standards, however the raw data needs to be looked at and a conclusion drawn before final conclusions can be reached. The Group noted it is unlikely that platinum is contribution to problems, but that this conclusion cannot be certain, as ti is difficult to prove negative. However, the lack of positive data could also form an important part of a risk assessment.

8. Adverse Incident Guidance [SIRG (02)77]
The poster relating to this item had already been discussed under item 5, recommendation 7.

9. Correspondence from interested parties
The Group noted the correspondence that had been received [SIRG (02)78-81] and the Chair thanked the Secretariat for responding on behalf of the IRG.

NR had reviewed the way MDA handles correspondence relating to breast implants. She concluded that in general the current system is thorough in responding to members of the public, she suggested that the website could be revamped to make it more accessible and had made some suggestions for this, including adding a section for frequently asked questions (FAQs). NR's detailed report would be consedered by MDA.

10. AOB
The Group discussed the future of the IRG.

11. Date of the next meeting
14th November 2003 (to be confirmed).

November 2002: Annex:
Recently published literature : Summaries provided by IRG members

A Prospective Study On Silicone Breast Implants and the Silicone-Related Symptoms Complex. Contant CME et al, Clinical Rheumatology (2002) 21:215-219 [SIRG(02)63]
This was a prospective study of a relatively small number of women (57) who were examined before and after silicone implant to determine the new occurrence of a range of clinical symptoms, the development of anti-nuclear antibodies and the development of rupture.

There were no new cases of anti-nuclear antibodies in this small number of women. There were changes in a number of symptomatic complaints pre-operatively to post-operatively, although for a substantial number of the complaints reporting did not alter. One of the major interests was in relation to symptoms suggestive of Sjogren's syndrome, that is of dry eyes and dry mouth. There were more symptoms of dry eyes and dry mouth post operatively. The significance of this is difficult to judge. The prevalence of these symptoms was fairly high pre-operatively and certainly much higher than the prevalence of Sjogren's syndrome in the population. Symptoms alone also, in the absence of specific autoantibodies, are not very predictive of the presence of immune damage caused by conditions such as Sjogren's syndrome. The data on implant rupture was uninformative.

Replacement Surgery and Silicone Gel Breast Implant Rupture: Self-Report By Women After Mammoplasty. Brown Lori et al. Journal of Women's Health & Gender-Based Medicine (2002) 11(3):255-264. [SIRG(02)64]
This study was a prospective interview survey of 907 women who had had a silicone gel implant who were asked about their self-report of rupture and other complications following the implant. The women were also asked about possible reasons for either implant removal or other replacement surgery.

One of the problems is that of the 2,994 women in the original population, only 907 responded. Some of these were because of non-response whereas subjects who were not living in the local area were not contacted. Of the 907 respondents, 303 had had their implants removed. Of these, 71% allowed the retrieval of their medical records but these were only obtained for 165. Thus this study provides some data on the self-reported, and partially validated, risks of replacement surgery and implant removal on a group of women following silicone gel implants.

An Association of Silicone-Gel Breast Implant Rupture and Fibromyalgia. Brown Lori S et al. Current Rheumatology Reports (2002) 4:293-298. [SIRG(02)65]
This is a review article by the authors investigating possible association between silicone gel breast implant rupture and fibromyalgia. In this review the authors mainly focus on their own work which is a true comparative epidemiological study, but they do also discuss other less robust epidemiological work investigating this association. Thus their main comments and conclusions are based on their own study which has been previously reviewed [SIRG(01)39)].

Metastatic Silicone Granuloma: Lupus Miliaris Disseminatus Faciei-like Facial Nodules and Sicca Complex in a Silicone Breast Implant Recipient. Suzuki K et al. American Medical Associations (2002) 138: 537-538. [SIRG (02) 66]
This is a short report describing a 56 year old woman who presented with features of a connective tissue disorder (Sjogren's syndrome) but who had no serological abnormalities. She then developed lupus miliaris disseminatus faciei-like lesions around the eyelids and a biopsy showed non-caseating granulomas. Further investigations as part of a sarcoidosis survey revealed fibrous implant capsules. The patient admitted to having implants in her 20's and 8 months prior to the symptoms had undergone mammography, which had been painful. The skin biopsy was examined by electron microscopy, which revealed silicone. Facial oedema went following subcutaneous mastectomy and axillary surgery although the skin nodules remained.
The authors suggest that contents of the implants might have spread following the pressure of mammography.

Breast Implant -Related Silicone Granulomas: the Literature and the Litigation. Austad ED. Plastic and Reconstructive Surgery (2002) 109(5):1724-1732. Comment: 1731-1732. [SIRG(02)67]
The granuloma is a response of tissues often to the presence-of fine material. It has often been postulated that if silicone leaks out of an implant it can cause a granuloma, although granulomas themselves are not necessarily harmful as opposed to just representing a normal tissue response. This article is a review of the literature that covers definitions of granuloma, cause of silicone granulomas (unclear, infection may be contributory but may represent response to foreign bodies), incidence and prevalence, location and clinical relevance. It is stated that "there is no scientific evidence to support the hypothesis that silicone granulomas play a part in so-called immune responses to silicone". The review of the literature reveals silicone granulomas to be a particular type of foreign body response. The author notes that granulomas are a very rare consequence of breast implants and concludes that granulomas do not play an important role in any of the presumed implant-related diseases. The Comment by RJ Rohrich (p1731-1732), supports these conclusions.

Silicone Gel Breast Implant Rupture Rates: Research Issues. Cook RR et al. Annals of Plastic Surgery (2002) 48(1). [SIRG(02)68]
This discussion of the problems of documenting the rate of rupture of silicone breast implants is essential reading for anyone planning to do a research project on this subject. It recommends that future work in this area should be carried out by multi-disciplinary teams, including plastic surgeons, radiologists, material scientists, biomaterial engineers, epidemiologists and biostatisticians. It also recommends that multi-disciplinary advisory groups could be set up to provide meaningful input into study protocols, data collection, analysis and interpretation to endeavour to produce future research papers which are really meaningful and stand a good chance of circumventing the many pitfalls of inaccurate assessment of rupture rates.
This paper is partly funded by Dow Corning.

Expression of Cyclooxygenase-2 in the Periprosthetic Capsule Surrounding a Silicone Shell Implant in the Rat. McLean AL et al. Annals of Plastic Surgery (2002) 48: 292-297. [SIRG (02) 69]
The aim of the study was to characterise the inflammatory response to silicone breast implants using an animal model, the hypothesis being that variations in the capsular inflammatory reaction may promote different intensities of contracture. Cyclooxygenase-2 (COX-2) is an inducible enzyme expressed in response to inflammatory stimuli. It is inhibited by glucocorticoids plus specific COX-2 inhibitors.

Silicone implants were inserted into male Spraque-Dawley rats. 59 animals were sacrificed at days 4, 7, 14, 28 and 80. De novo capsular tissue was dissected away and formalin fixed and processed. COX-2 was detected by immunohistochemistry; the specificity of the primary antibody was not given, but peptide blocking was undertaken. Endothelial cells, macrophages and fibroblasts were identified and COX-2 staining assessed but no semi-quantitation was applied. COX-2 was detected in the three cell types at all time points.

The findings are not surprising, are inconclusive and do not contribute to the literature.

Dynamics of Wound Healing After Silicone Device Implantation. Shanklin DR and SmalleyDL. Experimental and Molecular Pathology (1999) 67:26-39 [SIRG(02)72]
and
T-lymphocyte Memory in non-mammary silicone encounters. Shanklin DR et al. Journal of Nutritional Environmental Medicine (2002) 12:101-106. [SIRG(02)74]
These two papers are reviewed together as they are from the same research group and are somewhat related. They are discussed from the immunological perspective.

This group presents data showing that T cell responses to silicone products or derivatives can be detected in patient groups at levels higher than controls. As they have collected together data over a long time they have used stimulation index as a measure of T cell responses. This way of portraying data is not generally liked as it can cover a multitude of problems. However on face value and without having access to the primary data sets it looks as though there is convincing evidence of a higher T cell response in patients with implants. It furthermore looks from SIRG(02)72 as though the response diminishes substantially after excision of the implant only in the group with a saline filled implant. This could suggest that the agent causing proliferation (silica or its derivatives) disappears more readily from surrounding tissues with saline filled implants than with others. There is some confusing discussion on page 36 (1st para; SIRG(02)72) where the term hapten is used in a loose way and the term adjuvant is also thrown in.

It is however interesting to note that it may be that some silicone derivatives could act as an adjuvant (no hard data for this in these papers) and that Si might modify some self antigens or MHC molecules as is seen with Ni, Be and Au and this could lead to T cell hypersensitivity. Also iodinated peptides can be recognised by T cells. More work would need to be done to clarify this. Note in the context of Au or Ni there may be MHC associations so it might be expected that this might also occur with Si. It might have been interesting to have further data on this point in the context of Si.

It would appear that mechanical constraints affect wound healing and angiogenesis but a pathologist would be more able to talk to this point.

Clinicopathological review of pulmonary silicone embolism with special emphasis on the resultant histologic diversity in the lung. A review of five cases. Chung KY et al. Medical Journal (2002) 43: 152-159. [SIRG (02) 75]
This describes five cases from Korea, where there has been concern about illicit silicone injection. All cases were adult women with no previous ill health and all had received an injection of an unknown volume of silicone in the vaginal wall. One case received additional injections into the breast and right shoulder. At autopsy of this case, who died 15 hours after injection, lipid droplets were found in pulmonary alveolar capillaries. Case 2 died 10 hours after the last silicone injection and similar changes were found in the lungs. Case 3 lost consciousness during the second silicone injection, and died 5 days later. Both lungs were extensively oedematous and had globular vacuoles in capillaries. Case 4 developed dyspnoea after 2 silicone injections; an open lung biopsy showed globular material in capillaries with an associated inflammatory reaction (acute pneumonitis). She recovered after intensive treatment. Case 5 developed dyspnoea 2 days after the silicone injection and was treated for a pulmonary embolus. She died one month later. Autopsy showed diffuse alveolar damage and reactivated tuberculosis (due to corticosteroid treatment). Electron microscopy demonstrated globular deposits in pulmonary alveolar capillaries.
Subcutaneously injected silicone can gain access to the pulmonary vascular tree. The extent of the damage found in the different cases could be due to other factors. The study highlights a problem of injecting material but it is difficult to extrapolate this to silicone implants since the method of administration and dose is different from what could occur with leakage of silicone from an implant. Factors such as dose and purity cannot be assessed.

The Impact of Breast Augmentation on the Breast-screening Programme. Colville RJI and McLean NR. British Journal of Plastic Surgery (2002) 55:588-589. [SIRG(02)82]
This letter is considering the number of women screened in Newcastle/Northumberland from 1990-1999 plus number with implants. It documents the increase in patients with breast implants coming to the screening programme from 61 in 1990 - 1993 to 135 in 1996 - 1999 and anticipates a further increase in the future. In view of the fact that screening a patient with a breast implant for breast cancer is more time consuming than screening a patient without a breast implant. No screen detected or interval cancers were found in the implant group, This letter is making a plea to put more forward into breast screening programmes to cope with this increase in work.

Mortality among augmentation mammoplasty patients. Brinton LA et al Epidemiology (2001) 12(3):321-326 [SIRG(02)83]
and
Cancer risk at sites other than the breast following augmentation mammoplastic. Brinton LA et al. Annals of Epidemiology (2001); 11(4):248-256 [SIRG(02)84]
These two publications are based on a retrospective study of a cohort of women from 18 plastic surgery practices in six geographic areas. All women who had bilateral augmentation mammoplastic between 1960 and 1988 were included, but women with breast cancer were excluded. A total of 13,488 subjects were matched with 3,936 women who had other forms of plastic surgery. After 12 years of follow-up questionnaires were given to subjects who were located and alive, and attempts were made to obtain death certificates and medical verification for all reported cancers. Expected numbers of cancers were derived using general population incidence rates. The average age of the implant group w2as 34.0 years compared to 40.9 years for the control.

One of the cancers that was found to be present in excess in the augmentation group as compared to controls and the general population was brain cancer, and it was concluded that the reason for this were not apparent.

Brain cancers (glioblastoma multiforme) are uncommon tumours, but when they occur they are more frequent in the 30-40 year age group. The mean age of the augmentation group was 34.0 years, so the age distribution may have resulted in an apparent (but not significant) excess. A review of the literature has failed to identify any link between exposure to silicone in any form and the development of brain cancer, and this has been confirmed by neuropathologists. The various factors that have been linked to the development of brain cancer are methylene chloride (in men exposed as part of occupation) Am J Ind Med 1996, 30:504-9; herbicide exposure J Neurosci Nurs 1992; 24:260-4, but other studies have not confirmed this (J Neurooncol 1990; 8:55-60) and have suggested a link with decreased susceptibility to allergies.

In conclusion, there does appear to be a relative, non-significant excess of brain cancers in women with implants, but the numbers are small and the reasons are not apparent. There is no evidence to suggest a link between exposure to silicone and development of brain cancer.