SIRG minutes - 14 December 2000

Silicone Gel Breast Implants: Independent Review Group Meeting
14 December 2000, Hannibal House, Elephant and Castle

Present:
     Professor R Sturrock (Chairman)
     Professor R Batchelor
     Mrs V Harpwood
     Professor D London
     Mr T Milward
     Professor R Walker

Secretariat:
     Dr A Austin
     Dr S Ludgate
     Mrs A Miller
     Mr J Tinkler

Apologies:
     Ms C Rayner
     Professor A Silman

In attendance:
     Dr D Jeffereys (items 1,2,8-11)

1. Introduction
The Chairman welcomed all members of the Group present. MDA's new Chief Executive, David Jeffereys was introduced to the Group.

2. Minutes of the last meeting
The minutes of the previous meeting held on the 4 December 2000 (SIRG(00)90) were confirmed after correction of spelling errors. The Annex (scientific review reports) are not yet complete and these would be circulated to the Group, for approval, before their publication on the IRG Website.

3. Issues arising from the Open day
Several scientific papers were sent to the Group prior to the Open Day by members of the public. It was agreed that those papers which had not been reviewed previously should be identified and reviewed as new evidence. Papers previously reviewed would not be reviewed again at this time.

(i) The secretariat had written to the Chief Executive of the Benefits Agency (BA) regarding the points raised about difficulties in getting Disability Living Allowance(DLA). A reply has been received which states that the "entitlement to DLA does not depend on the illness or disability a person has, but on the effects of the condition on their care and mobility needs". The BA indicated that it would be happy to arrange for the claims of those who felt they had been unfairly treated to be investigated to make sure that the correct procedures had been followed. The Group agreed that as they could not pass on personal details the information should be passed to the woman concerned.

(ii) A number of questions were raised at the Open Day (including a "dossier of clinical based responses" to the IRG report) for which it was necessary to obtain further information by written correspondence. The secretariat has written to several clinicians and scientists. It was agreed that responses received should be reviewed by the Group and a summary document compiled and placed on the website.

(iii) Platinum catalysts are used during the manufacture of some breast implants. The effect of platinum on health was raised at the Open day and by Dr Myhill (see point 10). It was agreed that the Group should investigate the toxicity of platinum in breast implants further and the secretariat was asked to investigate the scope of the work required.

(iv) The question of what is meant by "safety" was discussed. Professor London had reviewed the Department of Health's Patient Information Leaflet and suggested that the information on life expectancy of the implants and the need for further operations should be clarified. The comments on the Leaflet will be forwarded to the part of the Department of Health which had responsibility for the leaflet.

A summary of the Open day has been prepared and will be placed on the Website along with an update on the recommendations.

4. Membership of Group
Professor Sturrock informed the Group that Claire Rayner had resigned from the Group. She indicated that she felt that personally she had nothing further that she could usefully contribute. The need for a replacement patient's representative was discussed and it was agreed that the Patient's Association should be approached for a nomination.

It was agreed to increase the membership to include a Toxicologist and that the British Toxicology Society should be approached for a nomination

5. Scientific Papers: Discussion and Update (SIRG(00)93-107)
The Chair reminded members that they should submit their written comments on scientific papers to the Secretariat for compilation. The IRG's comments on these reviewed papers are attached as an annex to these minutes.

Members present agreed that these papers did not alter the IRG's original conclusions, but one or two papers had identified areas that required continued monitoring.

6. Breast implants and mammography (SIRG(00)108)
It was agreed at the last meeting that information on breast implants and mammography, including a link to advice from the NHS Breast Screening Programme, should be provided on the IRG website. The secretariat apologized that this document was not yet ready for circulation. It will be completed early in the new year and circulated to the Group for comment.

7. Hydrogel Breast Implants (SIRG(00)109)
Copies of the recent MDA Device Alerts regarding hydrogel filled breast implants were circulated to the Group for information. In response to comments made at the Open Day, MDA faxed/sent copies of the device alerts to the women's support groups at the same time as hospitals were informed.

8. Correspondence concerning the IRG relating to:
8.1 The Registry (SIRG(00)110)
Correspondence had been received relating to the delay in the proposed study using the NBIR. It was noted by the Group that the Research and Development Directorate was taking this forward and had advertised for proposals.

8.2 Remploy/Biosil/Nagor - (SIRG(00)111)
Further correspondence relating to Remploy/Biosil/Nagor has been received. The IRG noted that the position was unchanged.

8.3 Trilucent breast implants (SIRG(00)112)
Further correspondence relating to Trilucent breast implants has been received. The IRG noted this correspondence.

8.4 Supply of Implants (SIRG(00)113)
A letter had been received from a member of the public and her MP stressing that breast implants should continue to be made available.

9. Letters dated xx/xx/xx (SIRG(00)114) and xx/xx/xx (SIRG(00)115)
The Group noted the contents of these two letters (from the same author) and stressed the fact that the issues raised had all been reviewed previously by the Group or were the subject of further actions following the Open Day in October (see point 3). An acknowledgement should be sent to the author but a detailed reply was not required to individual points.

10. Letter from Dr Myhill xx/xx/xx
This letter raises the issue of the effects of the platinum in breast implants and asks whether monies are available from the IRG for further work in this area, Professor Sturrock has already replied, informing Dr Myhill that IRG did not hold any funds for research. The effects of platinum is one of the issues raised at the Open day which is being investigated further.

11. Frequency and dates of future meetings
It was agreed that the next meeting would take place on 24 May 2001. It was agreed that the Group should continue to meet annually thereafter to review any new scientific data, additional meetings would be arranged if necessary.

Annex 1 : Recently published literature : Summaries provided by IRG members

Blanks RG et al "Effect of NHS breast screening programme on mortality from breast cancer in England and Wales 1990-8 comparison of observed with predicted mortality"
British Medical Journal 2000:321:665-9 [SIRG (00) 95]
This paper uses national data to document the 21% decline in breast cancer mortality in women aged 55-69 compared to that predicted prior to the introduction of a national screening programme. As this was an analysis of mortality and not of incidence there are no matters of relevance within the paper for the IRG.

Brown SL et al, "Prevalence of rupture of silicone gel breast implants revealed on MR imaging in a population of women in Birmingham, Albama".
American Journal of Radiology 2000:175;1057-1064 [SIRG (00) 85]
This is an interesting study examining the prevalence of ruptured implants using sensitive magnetic resonance imaging techniques. The authors document the previous literature discussing how rupture rates based on clinical presentation may underestimate the true occurrence, as judged by the frequency of rupture found on explanation undertaken for other reasons. However, even that population may be selected towards those, for example, with a high incidence of rupture. There are also data available from mammography suggesting that about 5% of asymptomatic women with an implant have had a rupture; again this might not be a very sensitive technique. The authors base their study on the growing awareness that MRI could be a technique with much improved sensitivity. As the study involved interpretation of images the authors used three radiologists to attempt to achieve a consensus.

The cohort was heavily selected. The National Cancer Institute had identified 13,000 women with a breast implant from 18 plastic surgery centres. Women for this study were recruited from 2 of those centres if they had answered a questionnaire and still lived locally. This amounted to approximately 1,250 women. Of these approximately 75% responded to a telephone interview of which 2/3 reported that they still had an implant. From a combination of reasons including resources and willingness to participate, slightly over half of these women (359) were eventually studied. The authors imply that there was excellent agreement between radiologists in diagnosing rupture with weighted Kappa values of at least 0.88. These are indeed high levels of agreement.

The major result was that, based on a consensus of two or more radiologists, 55% of the 687 implants in these 359 women showed evidence of rupture. Alternatively 68% of all women had at least one rupture. Rupture did not necessarily indicate migration of silicone. This, however, was observed in 1 in 8 implants, equivalent to 21% of all women. There was, however, less agreement between radiologists in the classification of silicone migration.

The authors went on to look at a number of features that were associated with implants. The most important finding was that rupture rate at an implant age of over 10 years was approximately twice that of implants implanted for a shorter period. Submuscular implants were at least 50% more likely to rupture than subglandular implants. There are not enough numbers reliably to comment on differences between manufacturer and generation.

Clearly when concern is the potential for referral bias, it is possible that the women who agreed to participate did so on the basis of having some concerns, perhaps in pursuit of a legal claim. Thus the study might have concentrated those women with suspicious symptoms. Alternatively, the data are consistent with results from women who were ex-planted for "other" reasons in that the majority of implants do show evidence of some rupture after 10 years.

Brinton LA et al, "Breast cancer following augmentation mammoplasty":
Cancer Causes and Control 2000: 11;819-827 [SIRG (00) 97]
This study was undertaken to examine the hypothesis that the risk of breast cancer may be altered as a consequence of breast implantation undertaken for cosmetic reasons. Other concerns relate to the implant leading to a delay in diagnosis and treatment. The authors quote that most previous studies had actually shown a reduction in breast cancer incidence in women with implants, the magnitude of this reduction sometimes being substantial at approximately 50%. The authors did quote, however, the important need to have the appropriate control group given the influence of potential confounding factors.

In this study, therefore, the authors compared the occurrence of breast cancer in women receiving unrelated cosmetic surgery to those receiving a cosmetic breast implant. They tried to identify breast cancer in both living and dead subjects. The authors used a reasonable lag phase in that the cosmetic surgery happened between 1962 to 1988 and the authors considered breast cancer and breast cancer mortality up to the end of 1996. Given the long duration of the study, response rates were reasonable, at about 7% in both groups. The incidence of breast cancer in the implant recipients compared to expected population data was 0.89 (95% CI 0.8 - 1.1). This is not dissimilar from the incidence in those receiving other cosmetic procedures, incidence 0.96 (95% CI 0.7-1.2). There were, however, important differences in relation to year of follow-up, age of initial implant and calendar year of implant. Those women with more than 15 years of follow-up, who had undergone initial surgery prior to the age of 30, or who had had their surgery after 1984, all had approximately 40% of the breast cancer incidence of the group that had undergone other cosmetic procedures.

In interpreting results it is clearly more a concern to know whether all potential confounders have been adjusted. The main aspect of interest is whether the nature of the breast structure that causes a woman to seek a cosmetic implant is one that alters her risk of breast cancer, independent of any subsequent surgery. Nonetheless, the authors acknowledge, these data do not support clinical reports suggesting that implants are associated with an increased risk of breast cancer.

Collis N et al, "Ten-year review of a prospective randomised controlled trial of textured versus smooth subglandular silicone gel breast implants".
Plastic and Reconstructive Surgery 2000:106;786-791 [SIRG (00) 98]
In this study the authors report the result of a ten-year follow-up study of 53 women who in 1989 participated in a double-blind clinical trial comparing smooth with textured gel implants. The question at issue is whether there was a difference in the development of contractures in that ten-year period. This study is severely limited by small numbers. It is not easy to follow the data given in the paper but it does appear that women who had a textured implant had a substantially reduced likelihood of contracture, compared with those with a smooth implant. The relative proportions were 65% of contracture rates for smooth and 11% for textured. The authors also wonder whether the risk of contracture is greater in smokers, although their study probably did not have sufficient power to assess this.

This is a very important paper carrying out a further assessment of the 53 women put into a double-blind randomised trial to compare the difference between textured surface and smooth surface silicone gel breast implants with regard to capsule formation. The key statement in the summary is "At 10 years, the incidence of capsular contracture was 65 percent of patients with smooth implants (an increase of 6 percent on the 3-year results) and 11 percent for the textured implant patients (no change on the 3-year results)."

This appears to be the only prospective double-blind randomized trial to explore this very important matter. It shows that, at 10 years, the rate of capsular contracture with smooth implants shows a considerable increase on the 3-year results, whereas the results with textured implant are the same at 10 years as they were at 3 years.

Brawer AE et al, "Amelioration of systemic disease after removal of silicone gel-filled breast implants"
Journal of Nutritional & Environmental Medicine 2000:10;125-132 [SIRG (00) 100]
This is a study of 156 SBI patients with symptoms attributed to their implants and the subsequent effects of removal of the implant. It is, in essence, a clinical series attempting to document what happened to the symptoms of the 156 women who, as a consequence of their symptoms, decided to have breast explantation. The author examined all the patients himself before and after explantation. His conclusion is that many patients improved following removal of their implants. All the women surveyed apparently had systemic illness, although the criteria for such an illness is not defined in the paper. It is difficult, given the absence of such data, to evaluate what is achieved after explantation. Similarly the author classifies improvement as a lessening of the frequency and/or severity of at least 50% of the total "number of symptoms".

This study is anecdotal and there was no independent observer. The clinical criteria for assessing the rheumatological symptoms of the patients was not specified and the conclusions reached by the author are not wholly justified. This seems to be a very difficult and subjective assessment of the causes of a problem, in that, in the absence of a clinical trial one cannot know whether changes in symptoms happened as a consequence of explantation or might have happened anyway. The author divides his patients into a variety of groups and illustrates his findings with case histories. However, given (i) the absence of data showing a standardised approach to assessment of both pre- and post-explantation symptoms, (ii) any suggestion that the study was done blinded and, particularly, (iii) the absence of a comparison group, it is almost impossible to interpret the results. Anecdotally a number of these women did improve following explantation. The authors argue that results indicate that women who are symptomatic following an implant should be recommended to have an explantation, even if the implant is not ruptured. However, given there is no discussion in this paper on what constitutes "symptomatic" such recommendation is difficult to put into practice.

Shanklin DR et al "Perinatal aspects of reproductive casualty after silicone device implantation".
Journal of Investigative Medicine 2000:48'171 (Meeting Abstract) [SIRG (00) 107]
In this abstract, the authors review the obstetric aspects of implantation by comparing reproductive outcome of women who, apparently (although it is not clear in the abstract), had had pregnancies both prior to and after an implant. There are significantly increased frequencies of both miscarriage and neonatal death in the post implant period, indeed the neonatal death rate after implant was 1 in 12. No neonatal deaths were observed in 86 women who had had a pregnancy prior to an implant. The authors also observed an increase in the frequency of placenta previa although, given the small numbers, this is not statistically significant. It is, however, difficult to judge the scientific accuracy of what could be a very important study just based on an abstract. It will be important therefore to see a full report of this research when it is eventually published.

Vergnes J.S., Jung R., Thakur A.K., Barfknecht T.R. and Reynolds V.L.; "Genetic Toxicity Evaluation of Octamethylcylcylotetrasiloxane"
Environmental and Molecular Mutagenesis 2000; 36 13-21 SIRG[(00)103]
In this laboratory study, 4 test systems were used to assess genetic toxicity of an octomethylsiloxane (D4). The 4 tests applied were bacterial mutagenicity, chromosomal aberration in chinese hamster ovary cells (CHO), chromatid exchange in CHO cells and chromosomal aberrations in rat bone marrow cells. No evidence for genotoxicity of D4 could be found in any of these systems.

D4 was thus assessed in properly designed and conducted tests for point mutations in bacteria and chromosome effects in cultured mammalian cells. Negative results were obtained. In addition there were no chromosome effects in rat bone marrow in a screening test involving inhalation exposure. However, no in vitro test for mutations in mammalian cells was carried out.

The tests conducted were robust and, together with the negative findings of previous work by Isquith et al (1998) in a mammalian cell mutation test, allow the firm conclusion that D4 is not genotoxic. This conclusion counters the further finding by Isquith (1988) of chromosome effects at toxic doses.

Hildebrand and Bombard; "Gene expression analysis in Livers of rats treated with Octamethylcylcylotetrasiloxane (D4)", Meeting Abstract [SIRG(00)82]
This is a short report on the effects of a polysiloxane, D4, on rat liver at the transcriptional level. D4 was compared with Phenobarbital in this system.
A sensitive method was used to detect induction of hepatic enzymes involved in various toxic effects or repair mechanisms. The effect of orally dosed D4 was compared to that of phenobarbital (PB), which induces a wide range of hepatic enzymes.

A similar, dose dependent, degree of enzyme induction was seen in D4-treated rats as in PB-treated rats, for phase-1 and phase-2 enzymes (involved in foreign compound metabolism and detoxification). Although there were changes in the expression of genes regulating apoptosis (programmed cell death), no effect was detected in markers for cell damage.

Gene expression analysis using micro-arrays did not demonstrate the expression of any damaging genes leading to DNA defects or cell cycle abnormalities.

This is consistent with the low order of toxicity seen for silicones and if, as some have suggested, metabolism of low molecular weight silicones occurs, provides evidence for the induction of metabolic enzymes.

Albers J.W. and Berent S. "Controversies in neurotoxicology: Current Status"
Clinical Neurobehavioral Toxicology 2000;18/3: 741-763 [SIRG(00)96]
This is a review article on neurotoxicology in which there is a section on SBIs. The review points out that there is no evidence for an association between SBIs and any specific neurological disorder.

Alderman A.K., Wilkins E.G., Lowery J.C., Kim M. and Davis J.A.;"Determination of patients satisfaction in postmastectomy breast reconstruction"
Plastic and Reconstructive Surgery 2000; 769-776 SIRG(00)99
This paper is not relevant to the Silicone Breast Implant Independent Review Group as it is merely reporting on patient satisfaction with various methods of breast reconstruction and not related directly to the safety of silicone gel implants.

3 Meeting Abstracts: Immunology of Organ Transplantation, FASEB J 1999, 13 [SIRG(00)101]:
A)McDonald, A.H.; Silicone-containing machrophages prevent activation-induced cell death in murine lymphocytes,
B) Naim J., Yoshida H., Buehner A., Nusz., Kurtelawicz., Cramer S.F. and Reeves W.H.Induction of hypergammaglobulinemia by silicone gels and oils in SW Mice,
C) O'Hanlon T.P, Lawless O. and Miller F.W.; Histopathology and T cell receptor (TCR) gene expression in silicone breast implant (SBI) capsules and other inflammatory tissues.

This short report is an abstract that contains no primary data; it reaches a number of conclusions, the validity of which have to be taken on trust since there is no data. Briefly, Dr McDonald states that work from her laboratory suggests that silicone gel fails to induce autoimmune disease in BAI B/c mice, but can exacerbate the autoimmune disease that occurs spontaneously in NZB mice. She carried out experiments designed to investigate the role of several T cell surface molecules known to be involved in switching T cells into cell division. She used antibodies against Class II H-2 molecules, B7.1, B7.2, CTLA-4, ICAM-1, and CD40L. Her test systems were a) T cell proliferation in vitro, measured by uptake of 3H-thymidine, and b) co-culture of silicone containing macrophages and T cells, and determining whether T cell activation-induced cell death was affected. She found that antibodies against only CD40L and ICAM-1 inhibited the proliferative response of BALB/c T cells. Different results were observed with the NZB derived T cell cultures, antibodies against class II 11-2, B7.1, B7.2, and ICAM-1 decreased T cell proliferation, but antibody against CD40L did not. Dr McDonald suggests that the results indicate that the T cell subsets activated in BALB/c mice but not NZB mice are possibly controlled by regulatory T cells.

Unfortunately, it is not possible to assess the validity of conclusions of abstracts that contain no data. There seems to be a working hypothesis that there is a human counterpart of the genetic autoimmune disease seen in NZB strain mice, and that such humans if implanted with silicone gels would develop an exacerbated form of the autoimmune disease. If this is the author's hypothesis, much further evidence would be required in support before it would be generally accepted.

Krayenbuhl B.H and Panizzon R.G. "Silicone Granuloma"
Dermatology, 2000; 200: 360-362 [SIRG(00)93]
Case report describing typical giant cell reaction occurring at site of implant of silicone elastomer implant into leg following surgery 27 years previously. Presence of silicone confirmed by x-ray dispersive analysis. Nothing new was described; the paper emphasises that problems can occur after a long time.

Lugowski S.J., Smith., Bonek., Lugowski J., Peters W. and Semple J. "Analysis of silicon in human tissues with special reference to silicone breast implants" ,
J. Trace Elements Med.Biol. 2000; 14 31-42 [SIRG(00)102]
Concerned with analytical procedures for measuring silicon (Si) in tissues and body fluids, with contamination-free sample preparation. Detailed descriptions of how blood was taken, urine and milk collected, and connective tissue from capsules, preparation of plastic ware and cleaning procedures. The population groups studied were Caucasian women with bilateral silicone gel-filled implants and age-matched controls. The authors compared results with those reported in literature over past 20 years and noted generally lower values. Although there was a significant increase in whole blood Si in the implant group there was a large variation in values in the control group, with a similar large variation in urine levels, which was not significantly different between implant and control groups. Si concentrations in connective tissue adjacent to implants were higher than levels from reduction mammoplasties.

This study emphasises the need to ensure that any studies of Si use sensitive methods, with care to avoid contamination, but also that it is important that the levels of Si within large numbers of human tissues is determined to be able to interpret the significance of Si levels in women with implants.

Pfleiderer B., Moore A., Tokareva E., Ackerman J.L., Garridao L. "Biodegradation of polysiloxanes in lymph nodes of rats measured with 29Si NMR"
Biomaterials 1999; 20: 561-571 [SIRG(00)94]
The aim was to compare the reactivity in rats of polysiloxane compounds of known molecular weight and chemical functionality with polymeric materials extracted from silicone gel-filled prostheses. The authors developed a rat model to study uptake of silicone into the lymphatic system, using 29Si NMR spectroscopy. The injected aqueous emulsions of polysiloxanes were associated with presence of unchanged molecules and products of biodegradation. All polymers were biodegradable but higher molecular weight forms required longer to degrade, and degradation was dose- and duration-related. Functional groups containing silicone were always found. It is proposed that after phagocytosis by macrophages, silicone metabolites are released into lymphatics, transported to other organs and then excreted (but no renal/urine measurements are given).

Polysiloxanes injected into rat thigh musculature were studied for bio-degradation in draining lymph nodes using 29SI spectroscopy. The experiment demonstrated that biodegradation in the nodes does occur but at a rate dependent on the molecular weight of the polysiloxane. The amount of siloxane present in the tissues of the rats is an order of magnitude higher than that in the tissues as a result of SBI bleed and therefore the results cannot be readily extrapolated to the clinical situation in SBI patients.

Kane C; The Cold Theory - A Hypothesis as to the Cause of the "Silicone" Illness. Silicone Information Network Website. [SIRG(00)70]
The Hypothesis proposed is that symptoms described by women with implants are due to intolerance to heat and cold (Temperature Regulation Disorder), and that this could be due to the high resistance to heat, cold water and passage of electricity that silicone has. No scientific data are presented.

This review proposes that symptoms in SBI women can be due to temperature dysregulation. Silicone is described as a cold inert substance which when implanted in the body reduces the tissue temperature in its locality and consequently results in hypothalamic dysregulation and immune disturbance.

No evidence base is cited for this hypothesis and the paper must be judged as pure speculation.