Tests commissioned by the IRG

Investigating group
Professor A.J. Freemont, University of Manchester
Mr J. Denton, University of Manchester

1. Introduction
The IRG determined that it was important to test a particular aspect of the hypothesis put forward by Professor Shanklin. This was that silica, which has known autoimmune properties, was present in tissues surrounding silicone gel breast implants.

The IRG requested Professor Shanklin to supply samples of tissues for analysis and arranged for testing to be carried out at the University of Manchester, Department of Pathological Science to establish the identity of the foreign material(s) present. Professor Shanklin supplied the Independent Review Group with glass mounted tissue from 8 tissue blocks from 7 patients.

Table 1 lists the Investigating Group's findings and the case numbers correspond to those on Professor Shanklin's original sheet. The Group's findings are as follows:-

2. Silica
The Investigating Group was unable to identify any silica in the tissues submitted using Professor Shanklin's criteria of bifringency (i.e. the ability of the material to refract light, splitting it into two beams). It should be noted that Professor Shanklin made it clear that many of his cases were outside referrals from which tissue sections were not available. It is possible that silica was present in those specimens alone.

3. Silicone
The Investigating Group had at first hoped to undertake electron probe analysis to identify silicon and oxygen. The only material that Professor Shanklin could provide was sections mounted on glass slides.

Electron probe analysis involves bombarding the sections with an electron beam in the vacuum of an electron microscope, and measuring the wavelength or energy of emitted x-rays. The electron beam burns through tissues leaving a visible track. Unfortunately it would have burned through the full thickness of the tissue section and vaporised some part of the underlying slide which, being glass, is rich in both silicon and oxygen. Fortunately, as electron probe analysis was not an option, the Group was able to identify a silicone solvent which, in test tissue sections, dissolved only silicone. This was applied to Professor Shanklin's sections.

Silicone was present in 6 of the 8 sections. The other 2 sections had no silicone. The Group recognised silicone in 3 different states within the tissue sections:

• there were fine globules in some macrophages giving appearances of "foamy macrophages"
• there were shards of solid silicone
• some of the silicone was in the form of "triangles" in the histological sections.

The latter 2 forms elicited a giant cell reaction. Interestingly, the triangular material seemed to be covered by congophilic proteinaceous material in some of the sections (i.e. protein material with an affinity for Congo red stain). The Group had some difficulty in understanding the three-dimensional structure of a material that, when cut in 2 dimensions, is always triangular. This seems, however, characteristic of these breast tissue sections.

4. Other material
Cottonoid was seen in most of the sections. It is introduced at most surgical procedures from swabs and patties. The appearance of this is characteristic with hollow tubes of birefringent material. It is unlikely that an experienced histopathologist would confuse this material, with it's characteristic appearance, with silicone. Formalin pigment, which is a bifringent fine granular material, was present in one slide.

5. Conclusions
The Investigating Group demonstrated the presence of silicone in most of the sections left by Professor Shanklin, but no silica. It is probable that the absence of silica is a reflection of the restricted nature of the material available, which is, in turn, due to the referral practice of Professor Shanklin.

The silicone solvent soluble material seems to have interesting physical properties which might represent a signature for breast implant material.

Table 1: Findings of the Investigating Group

Case Number	Silicone type*	Silica	Formalin pigment	Cottonoid	Other
55001B	T, L, S		YES	YES
55003A	L			YES
55004E	T
55005B	?L			YES
55006A	L, T			YES	Congo Red edges
55006B	L, T				Villus
55007B				YES
55008				YES

*
S= solid
L= liquid
T= triangles