A clinical study of the relationship between silicone breast implants and connective tissue disease.

OBJECTIVE: This study was a blinded, concurrent assessment of a historical cohort derived from a provincial registry (1978 to 1986) of breast implant recipients (cosmetic, not reconstructive) and controls (other cosmetic surgery) to test the hypothesis that connective tissue disease (CTD) is increased in breast implant recipients. METHODS: Women who underwent breast implant or other cosmetic surgery during the interval from 1978 to 1986 were contacted confidentially by Alberta Health and asked to participate in the study. Those willing to participate completed an extensive questionnaire and supplied a blood sample, subsequent to which all surgical records were reviewed to confirm implant type(s) or cosmetic surgery(ies). All participants with any suggestion of rheumatic disease were assessed blindly by a rheumatologist for CTD. RESULTS: One thousand five hundred seventy-six breast implant recipients were recruited, including 1112 who had received silicone gel-filled implants (> 13,500 person yrs exposure). Seven hundred twenty-six controls were recruited. Prevalence rates adjusted for sex and age for rheumatoid arthritis, systemic lupus erythematosus, scleroderma, and Sjögren's syndrome (the principal targeted conditions) were consistent with published reports for Caucasian women. While breast implant recipients self-reported significantly greater rates of symptoms than controls, post-surgical diagnoses of the principal targeted conditions did not indicate an increased incidence of typical or atypical CTD. CONCLUSION: The results of the study do not support the hypothesis that silicone gel-filled implants induce or promote CTD.

Gene synthesis and functional expression of a protein exhibiting monodomain IgG Fc binding.

A gene encoding a bacterial IgG Fc binding domain was designed and synthesized. The synthetic DNA fragment was cloned 3' to an inducible trpE promoter such that expression of the gene in Escherichia coli produced abundant Fc binding protein fused to the first seven amino acids of the trpE protein. The recombinant protein contained a single Fc binding domain and demonstrated efficient binding to human IgG in Western blot analysis. This protein degraded rapidly following cell lysis in the absence of protease inhibitors, but could be effectively protected by the addition of protease inhibitor. After purification of the protein by IgG affinity chromatography, IgG Fc binding ability was retained for at least 24 h at either 23 or 37 degrees C and on heating for 15 min at temperatures up to 65 degrees C. No immunoprecipitation was observed in interactions between the monodomain Fc binding protein and IgG molecules. Unlike staphylococcal protein A, no detectable binding of the monodomain IgG Fc binding protein was observed to either IgM or IgA. Truncated proteins, expressed from a series of 3' deletions of the synthetic gene, were used to estimate the minimum portion of a monodomain Fc binding protein that retained Fc binding ability.

A single Fc binding domain--alkaline phosphatase gene fusion expresses a protein with both IgG binding ability and alkaline phosphatase enzymatic activity.

A recombinant gene fusion was created and cloned using a previously constructed gene encoding a monodomain IgG Fc binding protein and the gene coding for bacterial alkaline phosphatase. The construct was able to express and secrete a fusion protein that exhibited both IgG binding and alkaline phosphatase enzymatic activities. Greater than 60% of the protein demonstrating both biological activities was detected from periplasmic space preparations. Nanogram concentrations of the Fc binding--alkaline phosphatase fusion protein allowed primary IgG antibody detection without the use of conjugated secondary antibodies. Removal of the domain coding for alkaline phosphatase resulted in decreased resistance of the protein to proteolytic degradation and the loss of IgG Fc binding ability. Using affinity-purified fusion protein, the specificity of binding to IgG, IgM and IgA was examined; binding was strong to IgG and barely detectable against IgM or IgA. Affinity for binding of the fusion protein to IgG (Kd = 6.7 x 10(-8) M) was determined to be equal to or greater than previously reported for protein A.

Breast cancer diagnosis and survival in women with and without breast implants.

The stage at diagnosis and the survival experience of 41 women who developed breast cancer after cosmetic breast augmentation were compared with those of all other patients with breast cancer (n = 13,246) diagnosed in Alberta from 1973 to 1990 (inclusive). The tumors in women with breast implants were smaller (65.9 percent < or = 2 cm) as compared with the tumors in women without implants (34.1 percent < or = 2 cm), but lymph node and distant metastases were equally frequent in the two groups. The distribution of tumor histologic types did not differ significantly between women with or without implants. Women who had an implant were younger at diagnosis of breast cancer compared with women with breast cancer and no breast implants. The relative 5- and 10-year survival rates did not differ significantly between the two groups, and the Kaplan-Meier survival estimate also was similar. It is concluded that women with breast implants in whom breast cancer develops are not diagnosed in a later stage and do not experience an impaired survival as compared with breast cancer patients without implants.

Breast augmentation: a risk factor for breast cancer?

BACKGROUND: A relation between breast augmentation and the subsequent risk of breast cancer has been postulated. Since an estimated 2 million women in the United States alone have received breast implants, even a small increase in the risk of breast cancer could have considerable public health consequences. METHODS: We performed a population-based nonconcurrent cohort-linkage study. All women in Alberta, Canada, who underwent cosmetic breast augmentation from 1973 through 1986 were included in the implant cohort (n = 11,676). This cohort was compared with the cohort of all women in Alberta in whom a first primary breast cancer was diagnosed (n = 13,557). The expected number of breast-cancer cases in the implant cohort was estimated by applying age-specific and calendar year--specific incidence rates of breast cancer (obtained from the Alberta Cancer Registry) to the implant cohort. Standardized incidence ratios were calculated by dividing the observed by the expected number of breast-cancer cases in the implant cohort. RESULTS: Forty-one patients with implants were subsequently found to have breast cancer. The expected number was 86.2. The standardized incidence ratio was thus 47.6 percent, significantly lower than expected (P less than 0.01). The average length of follow-up in the implant cohort was 10.2 years, and the average length of time from breast augmentation to the diagnosis of breast cancer was 7.5 years. CONCLUSIONS: Women who undergo breast augmentation with silicone implants have a lower risk of breast cancer than the general population. This finding suggests that these women are drawn from a population already at low risk and that the implants do not substantially increase the risk.

Urinary excretion of free toluenediamines in a patient with polyurethane-covered breast implants.

We report the detection of free 2,4-toluenediamine in urine of a patient implanted with polyurethane-covered breast implants. Samples were collected on several dates, ranging from 21 days to seven months after the insertion of the implants, and these samples all showed the presence of free 2,4-toluenediamine at a concentration of about 1 micrograms/L. The chemical was not found in a urine sample collected before implantation. This finding is important for risk assessment of cancer in patients with this type of breast implant because the chemical is a suspected carcinogen. Free 2,6-toluenediamine, an isomer, was also found in all samples from this patient.

Detection of toluenediamines in the urine of a patient with polyurethane-covered breast implants.

Breast prostheses are implanted for augmentation or during reconstructive surgery. One of the more commonly used prostheses is the polyurethane-sponge-covered silicone gel implant. Some clinicians are concerned about the safety of this product because the polyurethane foam disintegrates in vivo, and its subsequent fate is not known. Polyurethane is a polymer formed by reacting diisocyanates and polyols. This study indicates that the polymer sponge breaks down into its reactive monomers, 2,4- and 2,6-toluenediisocyanate, which are converted into their corresponding diamines. We present evidence of the excretion of the diamine metabolites in the urine of a patient implanted with polyurethane-covered prostheses.

Cervical spine injury in patients with facial skeletal trauma.

The records of 2,555 patients admitted to hospital with facial fractures were reviewed for concomitant neck injury. Cervical spine injury occurred in 1.3% of all patients with facial fractures, the incidence rising to 5.5% with facial fractures resulting only from MVA's. Neck injuries occurred most frequently in the setting of MVA's (85%) and multisystem trauma (70%) and tended to involve two main levels of the cervical spine: C2 (31%) and the two lower cervical vertebrae C6 and C7 (50%). Examination of the pattern of cervical spine injury with respect to the pattern of facial skeletal injury does not support the premise of a direct causal relation between them.

Surface localization of sialic acid on Actinomyces viscosus.

This study reports the presence of sialic acid in Actinomyces viscosus strains T14V and T14AV. Mild acid hydrolysis of whole organisms released a compound which reacted positively in the periodate-thiobarbituric acid, direct Ehrlich's and resorcinol assays, and which co-chromatographed on paper with authentic N-acetylneuraminic acid. Strain T14V contained 10-fold greater concentrations of sialic acid than did strain T14AV. Sialic acid content was dependent upon the stage of growth of the culture, reaching a maximum in early stationary phase. Epifluorescence microscopy of fluorescein isothiocyanate (FITC)-conjugated Limulus polyphemus agglutinin (LPA), a lectin specific for sialic acid, revealed a uniform distribution of bound lectin on the surfaces of strains T14V and T14AV. Additional evidence for surface localization was obtained by demonstration of whole-cell agglutination of both strains with LPA. All LPA interactions with A. viscosus were inhibited by the presence of 0.1 M-N-acetylneuraminic acid. Neuraminidases from Clostridium perfringens, Arthrobacter ureafaciens and Vibrio cholerae did not release detectable amounts of sialic acid, but the extracellular enzyme from A. viscosus cleaved amounts equivalent to those obtained by acid hydrolysis. Other laboratory strains (W1053, M100, W859, 5-5S, RC45, ATCC 19246, and 'binder') as well as recent clinical isolates of A. viscosus were agglutinated by LPA and released sialic acid upon mild acid hydrolysis. Surface-available sialic acid has been implicated in the inhibition of alternative complement pathway activation and subsequent opsonophagocytosis. Thus the occurrence of surface sialic acid in A. viscosus may represent a mechanism of pathogenesis for this oral bacterium.

Langmuir and scatchard parameters do not describe the binding of Actinomyces viscosus to saliva-treated hydroxyapatite.

The binding of Actinomyces viscosus T14V to saliva-treated spheroidal hydroxyapatite (SHA) beads was studied. The association constant (K) and the total number of binding sites (N) obtained from the Langmuir plots were in good agreement with those reported by other workers (approx. 3 X 10(-8) and 3 X 10(8), respectively). The values for N obtained from Scatchard plots differed from those obtained from Langmuir plots by factors of 10(6) or more. These results suggest that either these equations are inappropriate to describe binding or certain assumptions regarding this system are not being met. The use of these models requires, among other constraints, that the process be reversible and that measurements be taken at equilibrium. A method was developed which allowed a close examination of the equilibrium dynamics without perturbation of the system. The results suggest that the adsorption process is only poorly reversible. Adsorption to SHA was not at equilibrium after 1.5 h. Even when bacteria were allowed to adsorb for longer periods, and the system appeared to approach equilibrium, the increased time of adherence did not significantly alter the derived K or N values. Our results suggest that the use of Scatchard and Langmuir plots is inappropriate to describe binding of A. viscosus to SHA.

Resolution of prepubertal male gynecomastia following removal of a giant pigmented nevus.

A male infant is reported in whom substantial gynecomastia resolved following removal of a giant pigmented nevus. Endocrinological studies were normal. It is postulated that the nevus contributed to the development of gynecomastia. Surgeons should be aware of a possible relationship between these two seemingly unrelated conditions when making evaluation and management decisions.

The mitogenicity for murine splenocytes of specific surface components of the oral periodontopathic bacterium, Actinomyces viscosus.

Many characterized fractions obtained from A. viscosus were examined to identify the macromolecules responsible for mitogenicity for lymphocytes. Spleen-cell suspensions of CBA/J mice were cultured with 50-200 micrograms dry weight of A. viscosus strains T14V and T14AV cellular components. Lipopolysaccharide (LPS) (Escherichia coli) was used as a positive control. Mechanical disruption with a French pressure cell or sonication produced preparations with a stimulation of 69,082 and 45,183 counts above background (CAB), respectively. Mitogenic activity was also present in the culture supernatant (38,000 CAB). Other poorly mitogenic fractions included the peptidoglycan, cell-wall fractions, muramidase digests of cell walls, and the microcapsule extracted from whole cells with 0.5 M MgNO3. The results suggest that mitogenic activity is not associated with the isolated cell-wall structure. The activity was released from the cell surface by physical shearing forces, as well as released into the medium growth.

Role of macrophages in the lymphocyte response to Actinomyces viscosus.

Macrophage cooperation has been considered necessary for lymphocytes to express a variety of their differentiated functions. We attempted to characterize macrophage-lymphocyte cooperation in response to a known periodontal pathogen. Actinomyces viscosus T14V. Using adherent murine cells and subpopulations of splenocyte cultures, we assessed the effect of A. viscosus T14V fractions on lymphocyte proliferation. We determined that (i) various fractions of A. viscosus induced different proliferative responses; (ii) the physical state of the A. viscosus component determined the degree of dependence on adherent cells for proliferation; (iii) the proliferative response to supernatants of sonicated A. viscosus involved interaction among adherent cells and T and B cells; and (iv) the effects of adherent cells on the proliferative response were due to cell-to-cell interactions.

Morphological, chemical and antigenic characterization of M-1 N-acetylmuramidase-digested cell walls of the oral pathogenic bacterium Actinomyces viscosus T14V and T14AV.

The cell walls were enzymically solubilized with M-1 N-acetylmuramidase. The minimum amount of enzyme required for maximum response was 70 micrograms/mg of wall. The action of the enzyme seemed localized, producing holes in the wall structure. Chemistry and morphology suggested that all of the wall was solubilized. Antigenically, anti-T14AV formed precipitates with 11-14 antigens from the solubilized walls, of which 3 are unique to strain T14AV. Many of these antigens have not been observed previously due to the physical and/or chemical degradation associated with the extraction procedures. Antisera prepared against strain T14V whole cells formed precipitates with only a few antigens common to both strains. The results suggest that immunological processing and/or surface localization of these antigens are different.

"Staying power"--absorbable vs. nonabsorbable.

An experiment was designed to compare the "holding power" or "staying power" of absorbable (polyglycolic acid and polyglactin 910) and nonabsorbable (nylon) suture. The aim of this experiment was to determine what provides the lasting strength of the bond between soft tissues that are approximated or plicated. When correcting the rectus diastasis during abdominoplasty, we used nylon sutures in 15 patients and absorbable synthetic sutures in 15 other patients. We then marked the closed folds of the rectus sheath with small metal vascular clips. Two days later and approximately 6 months after operation an upright anteroposterior abdominal x-ray was taken and the position of the metal clips was compared in the test groups. Although there was usually slight separation of the clips after 6 months, no significant difference between the two groups was noted, thereby indicating that holding power is not related to type of suture material but more likely to fibroplasia.

Role of surface fimbriae (fibrils) in the adsorption of Actinomyces species to saliva-treated hydroxyapatite surfaces.

We studied the adsorption, morphological, and serological characteristics of selected Actinomyces and related species. Evaluation of uranyl acetate-stained cells by electron microscopy revealed wide variations among strains in the frequency of surface fimbriae. These variations did not always correlate with the percent adsorption to saliva-treated hydroxyapatite of the various Actinomyces strains. However, two strains of Rothia dentocariosa possessing no surface fimbriae and five strains of A. israelii possessing very few surface fimbriae exhibited feeble adsorption to saliva-treated hydroxyapatite. Although the calculated number of adsorption sites on saliva-treated hydroxypatite did not vary widely among the strains tested, significant differences were observed in the affinities calculated for some species or serotypes. The mean affinities for strains of A. viscosus serotype 2 and A. naeslundii serotype 3 were similar, and these strains adsorbed well to saliva-treated hydroxyapatite. The mean adsorption and affinity for the A. naeslundii strain serotype 1 and all strains of A. israelii tested were significantly less than those determined for the A. viscosus serotype 2 or A. naeslundii serotype 3 strains. Adsorption inhibition activity of antiserum to strain T14V, previously shown to be solely related to antibodies in immune serum directed against the VA1 fimbria (fibril) antigen, was removed by preadsorption of the antiserum with most A. viscosus and A. naelundii strains, but not with A. israelii strains. This suggests some cross-reactivity among strains of A. viscosus and A. naeslundii but not A. israelii. Adsorption to saliva-treated hydroxyapatite of all A. viscosus and A. naeslundii strains tested was strongly inhibited by fimbriae isolated from A. viscosus strain T14V. Collectively, these data suggest that the adsorption of certain A. viscosus and A. naeslundii strains is mediated by surface fimbriae, many of which appear serologically cross-reactive with strain T14V fimbriae.