SIRG minutes - 6 December 2001

Silicone Gel Breast Implants: Independent Review Group Meeting
6 December 2001, Hannibal House, Elephant and Castle

Present:
     Professor R Sturrock (Chairman)
     Mrs V Harpwood
     Professor I Kimber
     Mr T Milward
     Professor A Silman
     Professor R Walker

Apologies:
     Professor D London
     Ms K Murphy
     Ms N Ratcliff

Secretariat:
     Mr G A Crosbie
     Mr N Grocia
     Mr J Tinkler

1. Welcome
The Chairman welcomed members of the Group and introduced Professor Kimber, who gave the Group a brief overview of his background and experience in immunology and toxicology.

2. Minutes of the last meeting
The draft minutes (SIRG(01)61-62) of the previous meeting, held on the 14 May 2001 were reviewed. A number of typographic errors were identified, and the secretariat agreed to correct these before the final version was published.

3. Matters arising
With regard to the IRG website (agenda item 6 from the May 2001 meeting) it was noted that the new website provider would take over in April 2002. Major changes to the site could not be undertaken until after that date.

Prof Silman asked how many "hits" the website was currently receiving. The secretariat agreed to obtain that information for the IRG.

There were no other matters arising that were not already on the agenda.

4. Membership of the IRG
Immunologist - The Chairman reported that the Group was still searching for a replacement for Prof Batchelor. It was agreed that the Chairman would contact the British Society of Immunology seeking a suitable nominee.

Toxicologist - Prof Kimber had joined the IRG in this capacity.

The Group noted its disappointment that neither of the consumer representatives could be present due to conflicting commitments. The Chairman pointed out that Ms Ratcliff had written to the Group (SIRG(01)76) with regard to membership and declarations of interest. The secretariat tabled copies of the "Code of Practice for Board Members of Advisory Non-Departmental Public Bodies" (SIRG(01)78). This was discussed briefly and it was agreed that the secretariat would also send members copies of the relevant declaration form for them to complete and return.

5. Update on progress on IRG recommendations
Recommendation 1 - The secretariat reported that DH officials had meet with Baroness Emerton on 28 November 2001 to discuss revision of the patient information leaflet (originally published in October 2000). A draft revision would be produced by DH and circulated to members of the original working group for their consideration. Mrs Harpwood asked to be kept updated on progress.

Recommendation 2 - The IRG noted that Care Standards Act was now in place. Mrs Harpwood commented that the Act should have considerable impact on the advertising and promotion of private cosmetic surgery procedures.

Recommendation 3 - The IRG noted that the working party convened by the Royal College of Surgeons (under the chairmanship of Mr Brough) continued work on the drafting of clinical guidelines for surgeons carrying out breast implant surgery. The Chairman and Mr Milward agreed to write to Mr Brough seeking an update on progress.

Recommendation 4 - It was noted that the issue of consent to surgery was now being considered as a general issue across the whole of the NHS. Mr Milward commented that in his view there was still an argument for there being a specific form for breast implant procedures.

Recommendation 5 - The IRG noted that the Care Standards Act specifically required there to be a "cooling-off" period prior to surgery.

Recommendation 6 - The secretariat reported that the National Breast Implant Registry (NBIR) had now finalized a new consent from for registration. This will be introduced to use in late 2001 or early 2002. Prof Silman commented that the introduction of the form might limit the comprehensiveness of the registry, since highlighting the voluntary nature of registration might introduce a selection bias and affect the quality of the results of any studies carried out based upon NBIR registrants. The possibility of the use of annonymised data was raised and it was agreed that this needed clarification.

Recommendation 7 - Reporting of adverse incidents - no further developments.

Recommendation 8 - The secretariat reported that the pilot study to look at the feasibility of using data held by the NBIR to analyse early complications of breast implant surgery was now proceeding. With regard to studies of chronic low grade infection, the secretariat reported that DH Research and Development Division had agreed in principle to fund such work - they were awaiting a written proposal(s). The IRG asked the secretariat to keep them updated on developments on these projects.

Recommendation 9 - The group continues to monitor new scientific papers.

6. Breast implants and mammography
The proposed text of information for the IRG website had been circulated as (SIRG(01)64). The group agreed that the wording was acceptable, with the proviso that some changes should be made to the opening paragraph. The secretariat agreed to place the amended document on the IRG website.

7. Scientific papers
Members present summarized their findings from the papers they had reviewed, which included recently published or submitted scientific papers. Details of the reviews of published papers can be found in the annex to these minutes.
It was agreed that the papers by Briton et al (SIRG(01)20 & (01)40) and Brown et al (SIRG(01)62) should be considered in more detail and discussed again at the next meeting.

8. IRG Open Day - Review of actions taken
A compilation of correspondence which had taken place since the October 2000 Open day (SIRG(01)45) was discussed. The Group agreed that the summary of this work (included in the document as a draft) should be placed on the IRG website.

9. Platinum toxicity review
The secretariat reported that work on this matter was progressing. They had contacted manufacturers seeking further information on what platinum compounds were present in the silicones used in the manufacture of breast implants. The secretariat were in the process of obtaining this information.

10. Health Select Committee hearing on breast implants
The Government's response to the report of the Health Select Committee hearing on breast implants (SIRG(01)66) had been circulated to Group members. The Group noted the document.

11. Update on Europe
The Group noted that the European Commission's "Communication" regarding breast implants had been published in November 2001 (SIRG(01)74). The secretariat tabled a summary of the recommendations contained in the document and a summary of relevant actions already taken in the UK (SIRG(01)75). The Chairman commented that most of the recommendations had already been implemented in the UK in response to the IRG report.

12. Correspondence
The Group noted items of correspondence circulated (SIRG(01)67-9).

13. Any other business
None

14. Date of next meeting
It was agreed that the next meeting should take place on 23 May 2002 (full day).

DECEMBER 2001: Annex 1 : Recently published literature : Summaries provided by IRG members
Kjoller K, Friis S, Mellemkjaer L, et al, Connective tissue disease and other rheumatic conditions following cosmetic breast implantation in Denmark Archives of Internal Medicine 2001; 161: 973-979 [SIRG(01)43]
Objective: The objective of this study was to examine the occurrence of CTDs and other rheumatic conditions in Danish women with cosmetic silicone breast implants, in comparison to an appropriate control group.

Design, Subjects and Methods: 2761 with implants, 8807 controls (who had cosmetic surgery other than breast implant or who had consultation, without surgery, only) - from private and public plastic surgery clinics/departments. All were followed from Jan 1, 1977 to Dec 31 1996. The implant group was matched with controls by age (±3yr) and year of procedure/consultation (±18months) - although matching by age was impossible at one of the clinics. Women were followed-up from date of hospital discharge/outpatient visit for breast/cosmetic surgery until death, emigration, or Dec 31 1996. National in-patient registers were used to identify admissions with any diagnosis of a connective tissue, arthritic or auto-immune disorder. Since 1995 outpatient diagnoses were also available. Ratio of observed to expected cases (O/E 95% CI) of CTDs and ill-defined and other rheumatic conditions were calculated assuming a Poisson distribution.

Results:
Age distribution at time of surgery was similar for the two implant cohorts (private vs. public) and the reduction cohort (median 31yr) while the privately treated comparison group had a slightly higher age (median 33yr).

Ratios for definite CTDs and other rheumatic conditions
As the results for the outcomes under study were similar in both private, public cohorts and for both control groups they were combined for comparison

The only outcome that was statistically significant increased was "unspecified rheumatism". This was increased in both the implant group, 85 cases observed, 45 expected O/E = 1.9 (95% CI 1.5-2.3) and in the comparison group 238 cases observed, 152 expected (O/E=1.6, 95% CI 1.4-1.8).

There was no increase in any of the specific diagnoses in the implant group.

Direct comparison of implant with the control cohorts showed no differences for any definite CTDs (OR=1.0, 95% CI 0.4-1.7) or for ill-defined and other rheumatic conditions

Media coverage did not affect diagnosis (discharge rates for the outcomes were studied and no pattern was observed) - although the data for this is not shown.

There were 2 cases of scleroderma in the implant group compared to only 0.5 expected. The numbers are too small for useful comment.

Limitations of Study: The study had a well-defined population and strong follow-up due to use of nationwide hospital and population registers. Diagnoses were based on recorded diagnoses, rather than information from the medical records.

Implications of the Study:
No excess of definite CTDs were found in 2761 women with breast implants compared to either national rates of other women contemplating plastic surgery, which is consistent with other studies.
Those women undergoing/or seeking to undergo any cosmetic plastic surgery have excess unspecified rheumatic disorders compared with the general population and this should be taken into account when identifying control groups for studies of CTD in women with breast implants.

Englert H, Joyner E, McGill N et al., Women's Health after plastic surgery, Internal Medicine Journal 2001; 31: 77-89. [SIRG(01)44]
Objective: To compare prevalence/incidence of autoimmune and connective tissue disorders (CTDs) in a population-cohort of augmentation mammoplasty patients in Sydney.

Design, Subjects and Methods: Population-based retrospective cohort study comparing females who had cosmetic augmentation mammoplasty with patients who had non-silicone plastic surgery between the years 1979-1983. Groups were matched for age (± 5 yr), year of surgery (± 2 yr), plastic surgeon, anaesthetist and mode of anaesthesia. Excluded from the study were those with plastic surgery due to malignancy, those exposed to other silicone-containing prostheses or those diagnosed with CTDs prior to plastic surgery. Subjects were obtained from 16 (out of 28) plastic surgeons practising in the Sydney area between 1979 and 1983. Non-participation was due to medical records being destroyed (8), death of surgeon (2) and unwilling to participate (2). Four methods were used to study the cohort: womens health after surgery questionnaire; clinical examination; nailfold capillaroscopy (on fully participating subjects); antinuclear antibody (ANA) status. Various outcomes were measured: Rheumatoid arthritis (RA); systemic lupus erythematosus (SLE); scleroderma (SSc); sicca symptoms; CTD overlap syndrome; digital vasospasm; multiple sclerosis (MS); thyroid disease; abnormal nailfold capillaroscopy; elevated antinuclear titres. Univariate analysis examining the risk of each adverse outcome was expressed in terms of relative risk (RR) or odds ratio (OR).

Results: Prevalence of autoimmune and non-autoimmune diseases:
A positive ANA (>1/160) occurred more commonly in the exposed cohort: OR=1.3 (95% CI 1.0-1.6).

There were no significant differences in the frequency of high titre ANA (>1/320), SSc, SLE, sicca symptoms, RA, digital vasospasm, MS, thyroid disease, abnormal nailfolds and livedo reticularisin between the cohorts.

Restricting analysis to disease onsets post-surgery gave an overall RR for any CTD of 1.1 (95% CI 0.8-1.5).

Axillary adenopathy correlations: Axillary adenopathy (lymph gland enlargement) was increased in the mammoplasty group (OR=3.5, 95% CI 2.1-5.8).

The exposed cohort with axillary adenopathy had an increased risk of self-reported digital vasospasm after index surgery (RR=3.2, 95% CI 1.5-7.0) and an increased rate of capsular contracture (RR=2.4, 95% CI 1.2-4.6).

Limitations of the study:
This study is one of a number of similar cohort studies. It does however have a long follow up, validation of self-reported symptoms and high subject tracing and examination. Response rates in both groups were modest but there was no obvious evidence of a non-participation bias. Adequate physician blinding was demonstrated - by ascertaining the frequency of correctly diagnosing cohort status in the first 100 patients and the last 100 patients assessed.

Implications of the study:
Higher prevalances of axillary lymphadenopathy (16.7%) and low titre-positive ANA (11.2%) were detected more frequently in the exposed cohort.

There was no association between augmentation mammoplasty and various CTDs.

The clinical significance of this ANA result in the absence of an association with CTD is unclear.

D M Dush Breast implants and illness: a model of psychological factors, Ann Rheum Dis 2001 60, 653-657 [SIRG(01)48]
This paper by an American Psychologist under a section of Rheumatic Diseases entitled "hypothesis" postulates that there is a great similarity in the spectrum of symptoms suffered by worried patients with breast implants and those symptoms complained of in fibromyalgia, irritable bowel syndrome and chronic fatigue syndrome. He stresses that the symptoms must be taken seriously, but medical rehabilitation and behavioural therapy may be the way forward.

This review from a psychologist conjectures that the symptoms reported by women having a silicone breast implant are by their nature similar to those reported in synottomes attributed to somatisation - defined as the physical presentation of psychological distress. In relation to silicone breast implants, the author further conjectures that publication and reporting of possible adverse health consequences, could raise fears and increase the likelihood of such a process referred to as mass somatisation.

This is an interesting hypothesis, which is untested and possibly impossible to study directly.

S J Hoshaw et al, Breast implants and cancer: Causation, delayed detection and survival, Plastic and Reconstructive Surgery 2001 107, 1393-1408 [SIRG(01)49]
Concerns about implants relate to development of cancer (breast and others), delayed detection and decreased survival. This paper a) summarises the literature to date that has looked at any of these factors and b) reports a meta-analysis of the risk for breast cancer or other cancers.

For inclusion in the meta-analysis, each study had to provide a relative risk (or data needed to calculate it) and be in English. This resulted in 10 studies being available, published between 1995 and 2000. There was no evidence to support a causal association between breast implants and any type of cancer and there was no additional risk for breast cancer (possibly a protective effect). The implant type was unknown for many of the publications. The findings are similar to the recent publications of Brinton et al.

This competent review was carried out by employees of Dow Corning It is a detailed review of the evidence suggesting a link between silicone implants and cancers, both of the breast and elsewhere. In addition, data were analysed from studies investigating whether the presence of an implant delayed detection of breast cancer. Neither of these hypotheses has been clearly supported by previously published studies and it is thus not surprising that this review, which included a meta-analysis, excluded any important effect. Indeed the weight of evidence suggests a lower risk of breast cancer in women with breast implants. Confounding factors might be an explanation for this apparent negative effect.

JF Winther et al, Neurological disease among women with silicone breast implants in Denmark, Acta Neurologica Scandinavica 2001, 103, 93-6 [SIRG(01)56]

This was a Danish study based on linkage to hospital discharge registers of 1653 women with an implant and 1736 women with other plastic surgery procedures from the same private clinics. The particular outcome was a discharge diagnosis with a neurological disease. In general there was no trend or suggestion of any increased hospitalisation for all neurological disorders combined. Analysis for specific diagnoses is constrained by the very small numbers. Further hospitalisations may not be an appropriate outcome for many neurological disorders, which are predominantly treated in an outpatient setting.

L B Signorello et al, Offspring health risk after cosmetic breast implantation in Sweden, Annals of Plastic Surgery 2001 46, 279-86 [SIRG(01)57]
Objective: The objective of this study was to determine whether there was an increased risk of adverse health outcomes among children born to women who had undergone cosmetic breast implant surgery.

Design, subjects and methods: A retrospective cohort study in Sweden of 5,874 children born to 2,910 women with a prior cosmetic implants and 13,274 children born to women who had breast reduction surgery. A second comparison group were 4,285 children born to the implant women prior to surgery. Various health outcomes were studied including: hospitalisation rates for rheumatic and oesophageal disorders, cancer incidence, prevalence rates for congenital malformation and perinatal death. The data were derived from linking four national registers to identify breast surgery in the mothers and hospitalisations in their offspring: inpatient register, birth register, cancer register and death register. Women with surgery due to breast cancer were excluded from the study.

Results:
Characteristics of children/mothers
'Implant children' were born 5.6 ± 3.8 years after the implant surgery and followed for 8.9 ± 6.2 years whereas reduction children were born 5.9 ± 3.7 years after surgery and followed for 6.5 ± 5.0 years. 27% of the children were followed for more than 12 years. The average age of the mother at time of birth was higher in the implant mothers (30.9 years) compared to 28.5 years for reduction mothers (P <0.001).

Relative risks (RR) for various health outcomes (implant vs. reduction)
There was no increased risk for the children of implanted mothers for: esophageal disorders (RR=1.0; 95% CI 0.7-1.6), rheumatic diseases (RR=1.1; 95% CI 0.2-5.3), cancer incidence (RR=0.3; 95% CI 0-2.5), congenital malformations (total) (RR=1.0; 95% CI 0.6-1.5), malformations involving digestive organs (RR=0.5; 95% CI 0.2-1.3) or perinatal death (RR=0.9; 95% CI 0.5-1.8).

Relative risks for various health outcomes (children born before implant vs. children born after implant)
1,589 children born to women with implants after surgery were compared to the 4,285 children born before the surgery took place. There was no increased risk for esophageal disorders (RR =1.0; 95% CI 0.6-1.7). For those children born after the implant surgery: there was a significant deficit in congenital malformations (RR=0.6; 95% CI 0.4-0.9). Non-significant deficits for digestive malformations (RR=0.3; 95% CI 0.1-1.1), cancer incidence (RR=0.6; 95% CI 0.1-5.1), and rheumatic/autoimmune disease (RR=0.8; 95% CI 0.2-4.1) were reported. A marginally significant excess in perinatal death was reported (RR=2.1; 95% CI 1.0-4.8).

Limitations of the study:
Numbers of certain outcomes were small e.g. only 8 congenital malformations of the gastro-intestinal tract. Data on potential confounders of maternal lifestyle behaviour linked to adverse foetal/infant health (alcohol, cigarettes) was not included. However, large sample sizes were used, long observation period and two different comparison groups were used.

Implications of the study:
No evidence that important adverse health outcomes are more likely in children born to mothers who have cosmetic breast implants. The results support the findings of a previous Danish epidemiological study (Kjoller, 1998).

Holmich et al, Prevalence of silicone breast implant rupture among Danish women Plastic and Reconstructive Surgery 2001 vol 108 p848-863 [SIRG(01)72]
This is an analysis of a group of Danish women selected without bias, with the instance of implant rupture detected by MRI scans, with multiple review of the images and identification of implants by generation of manufacture and implant position and effect of close capsulotomy on rupture prevalence.

As Leroy Young and Maria Watson state in the discussion that follows the article, the weaknesses of the study include the small sample size, failure to sample the entire cohort and lack of surgical conformation of imaging diagnosis to establish the frequency of positives and negatives.

The study establishes the prevalence of rupture, i.e. "how many implants are intact at a moment in time". It is not able to show the time of rupture or the actual incidence rate at which the rupture occurs. The reviewer's point out that, to determine the incidence, one or more additional imaging screening separated by one or more years for the same cohort of patients are necessary. They look forward to follow-up imaging studies of this cohort which Holmich et al. are planning.

The reviewers put in a plea for the next step to be carried out, now that the cohort of patients is established and the methodology validated. They suggest that a base line magnetic imaging examination should be carried out to determine the prevalence of rupture, followed by repeat imaging examinations to determine the incidence of rupture. This article and the discussion that follows it deserve reading.

This was a detailed and carefully conducted study of implant rupture, using MR imaging, of 533 implants in 271 women. 436 had been selected to attend with sampling of women with the longest duration of follow-up. Analysis for possible non-participation bias did not reveal any serious concerns. Extensive training and standardisation between techniques and the observers at the 3 centres where MRI was undertaken was carried out. Ruptures were divided into definite and possible.
Overall, 26% of all implants had ruptured with a further 6% showing a possible rupture. Rupture was more frequent, the younger the woman was at the time of implantation (age 17-29: 32%, age 50-64: 16%) and were substantially more common in second generation implants (45%). Overall 22% of implants were extracapsular.
The study was limited, because of small numbers, in its ability to make comparisons in subgroups such as manufacturer.

Laing TJ et al, Potential risk factors for undifferentiated connective tissue disease among women: implanted medical devices American Journal of Epidemiology 2001 vol 154 p610-7 [SIRG(01)73]
This study examined the role of exposure to silicone containing and non-silicone containing medical devices by telephone interview of 205 women with so-called undifferentiated connective tissue disease (UCTD), 2095 controls and 660 patients with other scleroderma.

Compared to controls the UCTD cases reported about twice as high previous exposure to both silicone containing and non-silicone containing devices. The relative risk following breast implants was 2.2 (95% CI 0.7-7.6), which was lower than that following artificial (non-silicone containing) joint replacement (risk 5.0, 1.6-15.7). The latter was also higher than in-patients with scleroderma. There are a number of methodological issues raised appropriately by the authors on factors such as patient selection and recall. The former was of particular concern given the moderate participation rates. The authors were reluctant to give an overall conclusion as to the implication of their findings.

Lori Brown et al Infections Related to breast implants reported to the Food and Drug Administration 1977-1997 Journal of Long Term Effects of Medical Implants 2001 11, 1-12 [SIRG(01)47]
This paper records the occurrence of implantation-associated infections reported to the FDA in the time period of the study. The passive methodology used in this study means that the figures given are no more than the bare minimum of likely events. Because there are no data on the number of subjects at risk, no inference can be drawn on incidence. Two observations are of interest, the first relates to the pattern of reports when the highest numbers occurred during the years of maximal publicity and litigation, the second is the disparity in timing between the infections after saline implantation compared with the silicone gel device, with the latter being a more delayed event.

The very fact that late infection can be a problem might bring us back to a question asked during our original review, namely whether such events could be a causal link with the chronic ill health of which some implantees complain.

AW Eller et al, Migration of silicone oil into the brain: A complication of intraocular silicone oil for retinal tamponade American Journal of Ophthalmology 2000, 129 685-688 [SIRG(01)50]
Case report describing migration of silicone oil along optic nerve and into lateral ventricles of brain following repair of retinal detachment. Not relevant to silicone breast implants.

DB Sarwer et al, Cosmetic breast augmentation surgery: a critical overview, Journal of women's health and gender-based medicine 2000, 9(8) 843-56 [SIRG(01)54]
This is essentially a review of the breast augmentation story from an historical perspective with a focus on the psychological aspects involved in this form of breast surgery. It emphasises that it is important for health professionals dealing with breast augmentation patients to be well versed in the physiological and psychological implications of this type of surgery.

American Academy of Paediatrics, Committee on Drugs, The Transfer of Drugs and other Chemicals into Human Milk Peadiatrics 2001 Vol 108 p776-789 [SIRG(01)70]
This is a statement from the Committee on Drugs of the American Academy of Paediatrics on the risks of breast feeding by mothers' taking a wide variety of drugs. There is a small section on silicone breast implants which concludes that there is no evidence that silicone implants are a contraindication to breast feeding.

AD Malyon et al, Expanding silicone granuloma Br J Plastic Surgery 2001 54, 257-259 [SIRG(01)51]
This is a case of report of a woman with bilateral silicone implants for 30 years who underwent capsulectomy and replacement and then presented 8 months later with a 2 month history of marked swelling in prepectoral region. Histologically this proved to be an extensive silicone granuloma, despite concern that it might be neoplastic, due to its rapid growth following traumatic rupture of the breast implant almost thirty years after insertion. This is the first report of a silicone granuloma growing so quickly. There is no explanation provided as to why the rupture of this particular implant caused such an aggressive silicone granuloma, but radical surgery was required to remove the granuloma.

M O'Toole & C Caskey, Imaging spectrum of breast implants complications: mammography, ultrasound and magnetic resonance imaging, Seminars in Ultrasound CT and MRI, 2000, 21, 351-361 [SIRG(01)53]
This is a review article in a series entitled "Seminars in Ultrasound CT and MRI". It compares various imaging techniques for investigating breast implant complications.

M Kamel et al The peri-implant breast capsule: an immunophentypic study of capsule taken at explantation surgery Journal of Biomedical Materials Research 2001, 58,1, 88-96 SIRG(01)55
Study of 63 patients involving histological reviews of explanted capsules and immunohistochemical analysis for a range of lymphoid and macrophage markers.

Results were related to clinical features. A low grade non-specific reactive/inflammatory response was identified, which did not relate to the duration of implantation, presence or absence of calcification, arthritis or arthralgia, or fibromyalgia. As contracture increased, the inflammatory response decreased. An association was found between the incidence of pain and the presence of CD45 RO.

The findings support the view that there is no specific pattern of lymphocytic or histiocytic response that would suggest an immune or sensitivity response to silicone implants.

Main findings: no significant correlation between implant integrity and the number of lymphocytes or histiocytes present. Some link between capsular contracture and CD3/CD68 and between pain and CD45RO.

Conclusion: no evidence for a specific immune response.

Comments: (a) only symptomatic implantees were examined; (b) the duration of implant was 4 - 25 years (mean 14 years) - there was apparently no effect of duration on phenotype; (c) a mixture of bilateral and unilateral implants were examined; (d) no polymorphonuclear cells or B lymphocytes were identified.

EW Karlson et al, Monoclonal gammopathy of undetermined significance and exposure to breast implants. Archives of international medicine 2001,161, 864-7 [SIRG(01)58]
Monoclonal gammopathy is a benign condition that is a precursor to multiple myeloma or related diseases. Previous animal studies and cases of multiple myeloma developing at younger ages in women with silicone breast implants have suggested a possible association between exposure to silicone gel and multiple myeloma.

Plasma samples from 288 women with silicone breast implants and 288 randomly selected age-matched women were analysed (in a blinded manner) by immunofixation for the presence of monoclonal gammopathy. No differences were found between the two groups. The 5 women with implants who had a monoclonal gammopathy were older than the 4 in the control group. Due to the infrequency of such gammopathies the study may have limited power to detect anything but large increases in relative risk.

Background: Supposed increase in young age multiple myeloma among women with silicone gel breast implants. Evidence that mineral oil/ silicone gel can cause plasmacytomas in susceptible strains of mice.

Approach: A retrospective cohort study - within a larger women's prospective health study. 288 cases exposed to breast implants (only those with implants for cosmetic or prophylactic purposes considered, those with previous gammopathies or related diseases excluded) 235 of 288 had silicone gel implants; 53 of 288 had other implants. Mean age of cases 52.0 (and of controls 51,7).

Results:
5 of 288 (1.7% of those with implants) were positive for gammopathy
3 of 235 (1.3% with silicone gel implants) were positive for gammopathy
4 of 288 (1.4% of controls) were positive for gammopathy.

No differences in isotype of gammopathy; no multiple myelomas.

Breast implant exposed gammopathies in cases were in older women than those in controls.

Comment: Although it could be argued that the study lacks power, there is no evidence here for an association between breast impants and either monoclonal gammopathy or multiple myeloma.

CM Contant et al, First evaluation study of Dutch working party on silicone breast implants (SBI) and silicone-related symptom complex (SRSC), Clinical Rheumatology 2000, 19, 456-63 [SIRG(01)59]
Overview: A Prospective (longitudinal) cohort study that examined a total of 63 women in Rotterdam that had had mastectomy (usually for breast cancer) followed by immediate breast reconstruction with silicone breast implant. The aim was to assess the prevalence of medical complaints which have been related to the silicone-related symptom complex (SRSC) and the prevalence of anti nuclear antibodies (ANA).

Main findings:
There were 63 women with a total of 73 silicone implants, with a mean age of 46 years at operation. 10 (15.8%) were shown to be ANA positive. They were screened 2 months to 6.3 years after implantation, and positivity was found 2-29 months after implantation. A range of symptoms was reported but there was no difference between the ANA +ve and -ve groups.

There was no knowledge of the ANA status prior to surgery. All that can be concluded is that there was no relationship between SRSC symptoms (as judged by questionnaire) and ANA status.

Conclusion: ANA status was unrelated to symptoms of SRSC.

Comments:

1. No indication of ANA prevalence in age matched control women with or without surgery or other therapeutic interventions (including anaesthesia)
2. No indication of how many of the 63 cases examined had ANA prior to surgery
3. Is a questionnaire a satisfactory method to measure symptoms of SRSC?
4. Prevalence of ANA is lower than reported in other similar studies. Could this be due to the effect of treatment and/or median period from surgery to analysis - in this case 2 years - in other studies 4-10 years?

S Kasamaki et al, A case of inflammatory breast cancer following augmentation mammoplasty with silicone gel, Breast Cancer 2000 7(1) 71-4 SIRG(01)60
Describes a case presenting with features of an inflammatory breast cancer, diagnosed by MRI. The patient had had an implant 30 years previously. Both a carcinoma and silicone gel granulomas were found histologically but the manuscript does not contribute to any causal relationship.

CM Ward, The breast-implant controversy: a medico-moral critique Br J Plastic Surgery 2001 54, 352-357 [SIRG(01)52]
This article provides an excellent overview of the silicone gel controversy and its repercussions. In its analysis of the ethical and moral dimensions of the subject, it reaffirms the value of real science and reinforces the conclusions of the Silicone Gel Breast Implant Independent Review Group. The book "Science on trial" by Marcia Angell(1) provides the main source of material for the discussion, taking as the starting point Marcia Angell's statement:

"As the breast implant story unfolded, three professions - law, medicine and journalism - failed in their responsibilities to hold the controversy within the bounds of reason."
(1) Angell, M "Science on trial". London: W.W. Norton & Co, 1997

The article beings with a valuable historical summary of the now familiar events leading up to and surrounding the silicone gel breast implant controversy and litigation in America. The impossible position of the FDA is explained, and the respective roles of the media, lawyers and medical expert witnesses are set out very graphically. This summary provides a useful account for those coming for the first time to the subject.

The author continues by examining the silicone gel controversy from the ethical perspectives of paternalism, utilitarianism and justice.

Paternalism
The author discusses practical and ethical questions surrounding the doctrine of informed consent in the context medical paternalism, and emphasises the need to respect the autonomy of patients during the consent process. However, the author points out that unconstrained autonomy is not necessarily in the interest of patients, and that simply because a patient demands an unsuitable and risky surgical procedure there is not always a need for a doctor to agree to perform it. Occasionally a doctor is required to undertake a paternalistic role on ethical grounds or because the patient demands this. While recognising that paternalism can sometimes be a force for the good, the author argues that when the doctor allows his or her personal values to conflict with established medical practice or the patient's reasonable needs, paternalism can become a harmful influence. This is a strong moral criticism of some of the doctors involved in the silicone gel controversy, either as expert witnesses or as surgeons performing breast augmentation surgery.

The FDA is criticised in the article for taking a paternalistic stance in endorsing a restrictive policy relating to the use of silicone gel breast implants, and in what can only be interpreted as a sexist attitude, in the moratorium on silicone gel breast implants, towards women's right to decide for themselves about breast augmentation.

Utilitarianism
On a utilitarian analysis of the FDA decision, the author concludes that far from producing a health benefit to society as a whole, there was widespread psychological damage caused to women who had previously been happy with their silicone gel implants. It is pointed out that the silicone gel scare also led to widespread misinformation and to the introduction of alternative types of implants that have subsequently been proved dangerous. The result was that the action taken by the FDA seriously impaired public confidence in the doctor-patient relationship.

Against that, the only benefits were that a few women benefited enormously financially from the litigation, and that the FDA decision has prompted more rigorous scientific controls. However, these benefits have been overshadowed by the overwhelming harm caused by the decision. Thus as a utilitarian decision, looking at consequences, the stance taken by the FDA was disastrous.

Justice
In terms of distributive justice and resource allocation, the author makes the point that the litigation drove some of the main manufacturers out of business and lined the pockets of lawyers at the expense of research and development. In terms of rights-based justice, the argument is that individuals have the right to accurate information based on properly conducted scientific research, but this has been denied to many of the women who have had implants. Legal justice admittedly does not always correspond with what might be regarded as morally just or fair, and there is an obvious discrepancy between law and moral justice in the decisions made in the name of the law in the silicone gel litigation. On a social level what happened is unfair on individuals and it also brings the law into disrepute.

The article concludes with the notion that "the only good that might emerge is that if a similar eruption were to occur then a different course of action might be taken". It is the author's opinion that in such an outcome ethics could play a valuable practical part by informing the professions concerned of the correct path to choose. At a more abstract level, the author concludes that the wider issues raised by the silicone gel controversy include the problems involved in the inescapable conflict between the forces of utilitarianism and the rights of the individual. The final issue raised is the recognition of vulnerability and the limited role that ethics can play in that context, as ethics are easily discarded by professional people in the face of human greed.

The Perspective of the Independent Review Group
This article provides a valuable and insightful overview of the main ethical issues raised by the silicone gel breast implant problem, and reassuringly endorses the approach taken by the IRG, particularly in respect of their concern for scientific truth and the individual's right to informed consent on the basis of accurate information. Without complacency it must be concluded that the views of the IRG relating to on-going research, the value of a register, the proper regulation of independent clinics and the need to obtain informed consent to surgery on the basis of a sound information leaflet are evidence of consistency with the moral stance taken by the author.