Analysis of silicon in human tissues with special reference to silicone breast implants.

The increase, in the last two decades, in the application of silicones (polysiloxanes) and inorganic silicon compounds in medicine and the food industry, has exposed the human body to extensive contacts with these substances. Most silicone breast implants contain a gel consisting of a crosslinked silicone elastomer swollen by silicone oil (PDMS). Diffusion of PDMS through the silicone elastomer envelope and rupture of the envelope with release of the gel contents both occur clinically. The amount and distribution of silicone compounds in various tissues are key issues in the assessment of health problems connected with silicone implants. We have measured by GFAAS the Si content of tissues from normal and implant patients and the organic solvent extractable Si levels (assumed to be silicone), using careful control of sample collection and preparation. Whole blood levels were: implant patients mean 38.8 (SD 25.6) (microg/kg), controls mean 24.2 (SD 26.7) (microg/kg) in one study and subsequently 103.8 (SD 112.1) and 74.3 (SD 86.5) (microg/kg) in another study. Capsular tissue levels were: gel implants 25047 (SD 39313) (mg/kg of dry tissue), saline implants 20.0 (SD 27.3) (mg/kg of dry tissue) and controls 0.24 (SD 0.39) (mg/kg of dry tissue). Breast milk levels were: implant patients mean 58.7 (SD 33.8) (microg/kg), controls mean 51.1 (SD 31.0) (microg/kg); infant formula mean was 4.40 (mg/kg). Various precautions were undertaken to avoid Si contamination in this work, the most important being a) the use of a Class 100 laboratory for sample preparation and b) application of strict and elaborate washing procedure for specimen collection tools and laboratory plasticware. This data demonstrated that to properly interpret the importance of these numbers for human health, a larger study of "normal" levels of Si in human tissues should be undertaken and factors such as diet, water, race and geographical location should be considered.

Breast milk contamination and silicone implants: preliminary results using silicon as a proxy measurement for silicone.

In response to concerns about contamination of human breast milk from silicone gel-filled breast implants, and because silicon levels are assumed to be a proxy measurement for silicone, we compared silicon levels in milk from lactating women with and without implants. Two other sources of infant nutrition, cow's milk and infant formulas, were also analyzed for silicon. The survey took place at the Breast-feeding Clinic at Women's College Hospital in Toronto. A convenience sample of lactating women, 15 with bilateral silicone gel-filled implants and 34 with no implants, was selected. Women with foam-covered or saline implants or with medically related silicone exposures were ineligible. Collection of samples was scrupulously controlled to avoid contamination. Samples were prepared in a class 100 "ultraclean" laboratory and analyzed using graphite furnace atomic absorption spectrophotometry. Silicon levels were analyzed in breast milk, whole blood, cow's milk, and 26 brands of infant formulas. Comparing implanted women to controls, mean silicon levels were not significantly different in breast milk (55.45 +/- 35 and 51.05 +/- 31 ng/ml, respectively) or in blood (79.29 +/- 87 and 103.76 +/- 112 ng/ml, respectively). Mean silicon level measured in store-bought cow's milk was 708.94 ng/ml, and that for 26 brands of commercially available infant formula was 4402.5 ng/ml (ng/ml = parts per billion). We concluded that lactating women with silicone implants are similar to control women with respect to levels of silicon in their breast milk and blood. Silicon levels are 10 times higher in cow's milk and even higher in infant formulas.

Titanemia from total knee arthroplasty. A case resulting from a failed patellar component.

The subject of this case report is a patient with elevated serum levels of titanium (77 parts/billion [ppb]; normal, 3.3 ppb) and vanadium (0.38 ppb; normal, 0.17 ppb) resulting from excessive wear of a metal-backed patellar component in a total knee arthroplasty. The patellar component was worn through both its polyethylene and metal backing as a result of abnormal contact between the patellar and femoral components. Scanning electron microscopic examination of the ingrowth surface of the patellar component indicated that particle debonding occurred as a result of overloading of the sintered neck regions at the particle-substrate interface, suggesting a possible damage during initial insertion of the device, which may have predisposed it to loosening and abnormal contact with the femoral component. Wear particles resulted in staining of the tissues within the knee and an inflammatory and immune response in the synovium consisting of giant cells and T lymphocytes. The serum metal levels were reduced 22 weeks after replacing the patellar component; however, the titanium level was still slightly elevated (8 ppb).

Release of metal ions from dental implant materials in vivo: determination of Al, Co, Cr, Mo, Ni, V, and Ti in organ tissue.

Little is known of the release of trace elements in vivo from dental implant materials. Conflicting data have been reported in the literature as to the levels of trace elements released and their potential consequences, mainly because of sampling and analytical methodological errors. In this study methods for average concentration levels of Al, Cr, Co, Mo, Ni, Ti, and V in organs were developed using rabbit tissue from an in vivo implantation study. At least 50% of the brain, liver, lung, kidney, and spleen were taken minimizing contamination. The tissue was homogenized by cryogenic milling at LN2 temperature and then freeze-dried. Samples were digested in nitric acid and hydrogen peroxide using microwave energy. Standard reference materials were utilized for quality control. One liver sample was analyzed 10 times to assess the method precision. Absorbance values in blanks, standards, and test samples were measured using a Varian GTA 95 graphite furnace and 875 spectrophotometer. Very satisfactory method precision and quality control were recorded. Low or very low levels of the trace elements were found in the various organs.

Determination of chromium, cobalt and molybdenum in synovial fluid by GFAAS.

An artificial knee joint made of a cobalt alloy (Cr 27-38%, Mo 5-7%, others 9%, Co balance) failed in an orthopaedic patient with severe tissue reaction. Samples of synovial fluid and blood were analyzed for cobalt, chromium and molybdenum concentration using graphite furnace atomic absorption spectrophotometry (GFAAS). NIST SRM 909 Human Serum containing 91.3 micrograms/L chromium was used as the Cr Standard. To obtain reliable values for cobalt and molybdenum, three different independent analytical methods were employed to analyze each sample: 1) dilution of the samples with Triton TX-100 solution, followed by GFAAS; 2) microwave decomposition of the samples, followed by GFAAS; 3) classic nitric-perchloric acid decomposition of the samples, followed by GFAAS. The two decomposition methods were used to control the accuracy of the dilution method. The results showed that the non-decomposition method gave comparable results to the decomposition techniques for analysis of synovial fluid and blood. The data revealed that the patient had elevated levels of Co, Cr and Mo in both synovial fluid and blood with the values being 2-3 orders of magnitude higher in the synovial fluid from the affected knee.