Systemic Sclerosis: Environmental Factors

PATHOPHYSIOLOGY

Mechanisms responsible of the development of environmentally induced systemic autoimmunity are poorly understood. There are several hypotheses, but none of them are accurately sustained by direct evidence. Those mechanisms supposed to be involved in the beginning of the disease may be different to those worsening an established disease.

Environmental pathogenesis of SSc may be divided into 3 major mechanisms²⁸, as follows.

Immune Tolerance Interference

Loss of self-tolerance is crucial for the development of autoimmunity as a general concept. Many studies suggest that environmental exposure to certain agents affects tolerance at several points.

The haptene hypothesis proposes that a drug or its metabolites bind to certain proteins, changing their immunogenicity and triggering an immune response. The T cell recognition of such a complex is altered and results in immune tolerance interference. Human lymphocytes stimulated with silica express high levels of Fas receptor (CD95) and undergo apoptosis, with autoantigen alteration, concentration, and release, provoking an autoimmune attack²⁹ and post-translational protein changes²⁸.

Epigenetics comprise all inherited changes in gene expression that are not coded in the DNA sequence itself. Two major mechanisms mediate epigenetic changes: DNA methylation and histone modification^30–33. There is a direct causal relationship between methylation-dependent transcriptional repression and histone modifications. Some drugs may alter genomic activity by means of changes on DNA structure. Methylation of DNA suppresses genetic activity and regulates genes potentially harmful for cell functions. Hydralazyne and procainamyde are examples of DNA methylation inhibitors that enhance the expression of normally silenced genes and may lead to autoimmunity^33–37.

Scleroderma-associated cellular abnormalities persist in multiple generations of SSc fibroblasts in vitro and this profibrotic phenotype persists outside the disease microenvironment, suggesting the inheritance and transmission of in vivo imprinting of this disease phenotype among generations of fibroblasts. Increased levels of epigenetic mediators in SSc fibroblasts were noted in a study by Wang and colleagues³⁰. The augmented collagen synthesis by SSc fibroblasts was linked to epigenetic repression of the collagen suppressor gene FLI1. Heavy methylation of the CpG islands in the FLI1 promoter region was demonstrated in SSc fibroblasts and skin biopsy specimens.

Immune System Activation

It has been postulated that many cases of autoimmunity that occur after the therapeutic administration of cytokines and resolve after withdrawal may be the result of the direct immune activation by these cytokines. Other studies show that respirable silica particles are phagocytized by alveolar macrophages, driving cell activation and release of soluble mediators such as cytokines: tumor necrosis factor-α, interleukin 1-ß, and transforming growth factor-ß³⁸. T cell incubation with silica and silicates may cause lymphoid polyclonal activation in vitro²⁹.

The role of solvents in the initiation of SSc is not clear. Autoimmune disease-prone MRL +/+ mice exposed to trichloroethylene (TCE) and its metabolites induced autoantibody production³⁹. In a population exposed to domestic contamination of water with TCE, there were found high levels of CD4+ CD8+ T lymphocytes in peripheral blood, suggesting that this agent may alter normal immunity by modification of autoantigens⁴⁰. Solvents penetrate through the skin and the airways, initiating cellular and humoral autoimmunity and stimulating the production of fibrogenic proteins and growth factors⁴¹.

Vinyl chloride can enhance the immunogenicity of certain intracellular molecules, CD8+ activation⁴², and skin thickening related to collagen deposition⁴³.

Molecular Mimicry

This mechanism is characterized by an immune response to an environmental agent cross-reacting with the host antigens. The best epidemiologic evidence for this mechanism in human disease is beta-hemolytic streptococci infection and rheumatic fever⁴⁴. It is challenging to elucidate if this cross-reactivity is an epiphenomenon of such infection, or if autoimmunity appears after a loss of T cell ability to discriminate self- from non-self antigens through the shared epitope mechanism⁴⁵.

Maul, et al identified an 11-amino acid epitope on the C-terminal extreme of topoisomerase (topo)-1 sharing 6 of 11 sequential amino acids of the group-specific antigen p30gag from certain mammal retroviruses⁴⁶. Muryoi and colleagues described an epitope the recognizes the N-terminal extreme of topo-1 in patients with SSc and TSK mice. This epitope seems to have a cross-reactivity with other fragments of topo-1 and shares some degree of homology with UL70 protein of human cytomegalovirus (HCMV)⁴⁷.

There may be no single mechanism to explain environmental exposure triggering such an heterogeneous disease as SSc. It is more likely to exist a combination of all these mechanisms. Most of the alterations of the immune system may involve antigen recognition and processing, cell signaling, and cytokine production.

New approaches in biology have led to the implementation and combination of genomic, proteomic, and metabolomic studies, bringing us closer to identifying the molecular signature of some environmental exposures.

EPIDEMIOLOGY

Epidemiologic evidence for the association between environmental and occupational risk factors and SSc has been extensively analyzed. Because those exposures are frequently of long duration, there is growing interest in study of damage to health in the workplace, to establish a safe occupational environment for human health.

The majority of individuals are not aware of the specific agent they have been exposed to during their work, and without validated biomarkers, many of these studies are based on indirect evidence of exposure.

Environmental exposure may vary widely and it may be considered to include all non-genetic factors. Commonly, chemicals or drugs are considered environmental factors; stress events of daily life, other factors from lifestyle and incidental exposures like ultraviolet radiation could also be environmental risk factors for autoinmune diseases. With this in mind, environmental factors can be classificated into 3 categories: occupational, infectious, and non-occupational/non-infectious.

The association between environmental exposure and the beginning of SSc may be difficult to demonstrate. In this situation, the association is rare due to the fact that most of the patients with SSc did not have any exposure, just as the majority of the exposed subjects do not develop the disease, unless we work with a small group of subjects in which both situations occur simultaneously.

Case-control studies are of limited value because of the rarity of the exposure, as are cohort studies because of the rarity of the disease that could develop in an exposed group. Thus, many of these studies do not have enough statistical power to exclude the real effects or to detect moderate associations. The inadequate classification of the type of environmental exposure and other confounding variables may bias the estimated association between exposure and SSc.

Occupational Factors

Silica. The frequency of occupational exposure to crystaline silica dust, as a generator of work damage, has been underestimated, even though it is recognized as a risk factor for many systemic autoimmune diseases including rheumatoid arthritis, scleroderma, systemic lupus erythematosus, and small-vessel vasculitis with renal involvement (antineutrophil cytoplasmic antibody related vasculitis)^38,48–50.

Silicon is an ubiquitous mineral in the environment. It constitutes the second most abundant element in the earth’s crust and is part of air pollution particles; it may represent a significant proportion of environmental dust levels in some geographic areas^51,52. It primarily exists in the crystalline state as quartz, which is structurally and chemically different from amorphous silica (diatomaceous earth), silicates (starch and asbestos), and silicones (a polymer containing silicon: polydimethyl siloxane [(Si[CH₃]₂O)_n]). Crystalline silica may be harmful when it is inhalated as respirable dust (particles of less than 5 μm). The allowed limits of exposure for respirable silica stablished by the US Occupational Health and Safety Administration are close to 0.1 mg/m³ (estimated for an 8 h exposure for quartz), constituting double the recommended limit of exposure: 0.05 mg/m³, suggested by the US National Institute of Occupational Safety and Health, to minimize the risk of silicosis^53,54.

Professions classically related to occupational exposure to this agent are mining, sandblasting, and pottery. There are many other occupational sources not frequently associated with the induction of autoimmunity. Some examples are mechanical dentistry, agricultural tasks, asphalt work, cosmetology, use of abrasives, jewelry, etc.^50,54 (Table 2)⁵⁵.

Technological advances provided powerful tools to the mining industry, increasing the workers’ exposure to silica dust. Pneumoconiosis became more frequent and developed faster when steam machinery was installed in factories by the end of the 19th century. The intensive use of pneumatic hammers made this occupation more dangerous for human health⁵⁶.

In 1914, a physician from Chalmers Hospital and consultant to the Edinburgh Royal Infirmary named Bramwell described the incidence of scleroderma in 9 Scottish patients. Surprisingly, in his study, there was just 1 female patient⁵⁷.

Four decades later, Erasmus expressed his interest in the high prevalence of SSc in 40,013 gold miners from the Witwatersrand¹. Rodnan, et al reported a prevalence study about silica exposure in 60 men with SSc between 1955 and 1965. They showed evidence that SSc in patients with occupational exposure to silica is higher than in the general population, suggesting that silicosis could be a predisposing factor in the pathogenesis of the disease⁷.

Other reports documented SSc cases among patients working in mechanical dentistry⁵⁸; a case of conjugal SSc related to silica exposure was also described⁵⁹.

In the 1990s case reports, case-control studies, exposure prevalence and incidence studies in patients with Ssc (institutionally or population based) were published. Even though the number of exposed subjects may vary from different samples and the spacio-temporal cluster, the data obtained did not sustain an important role for silica in women, but in some cases evidenced the occupational background of silica exposure for SSc in men (published findings in Tables 3–6)^60–73.

In Argentina, a clinic-based case-control study assessed silica dust exposure in 20 patients with SSc between 1998 and 2002. A questionnaire designed by environmental hygienists was applied to cases and controls discriminating the type of task carried by the subjects: construction, janitorial, agricultural, and dental⁷². Although this study did not achieve adequate statistical power, the group with SSc showed 60% previous exposure to free silica sources [odds ratio (OR) 21; 95% confidence interval (CI) 4.7-101; p < 10⁻⁷]. Sixty-seven percent of patients with SSc and silica exposure were women (OR 24.6; 95% CI 3.6-216; p < 10⁻⁵). Time of latency between the beginning of the exposure and disease diagnosis was 21 years, on average. In an emerging country, the role of women in tasks classically assigned to men may be similar in middle and low income classes.

Vinyl chloride

This is an uncolored combustible gas (CH2 = CHCl), used in plastic manufacturing. In the mid-1960s a new syndrome affecting workers of vinyl chloride polymerization was described by Wilson and colleagues⁷⁴. These patients developed finger paresthesias, Raynaud’s phenomenon (RP), pseudo-acropachy, skin thickening, edema of hands and forearms, and chest radiographic changes⁷⁵. Risk for suffering these alterations is related to cumulative exposure over time and not necessarily with manipulation of the final product, polyvinyl chloride (PVC)⁷⁴. There seems to be an apparent increase of the prevalence of MHC DR3 and DR3/B8 haplotypes in patients with this disease⁷⁶. Skin changes in vinyl chloride disease may be similar to morphea, and the most striking hallmark of the syndrome is phalanx acroosteolysis⁷⁵. Several immune alterations have been described, such as polyclonal IgG increase, cryoglobulins, complement activation, and antinuclear antibodies (ANA) in low titers¹⁰. Clinical and symptomatic improvement after withdrawal of the exposure is characteristic⁷⁴.

Organic solvents

The determination of the type of solvent that a subject has been exposed to is crucial to the accuracy of epidemiological studies. It is frequently difficult to infer which agent is involved, not only because of limitations in the individual’s knowledge but also because frequent users of solvents do not look for a specific chemical but for the chemical properties of solvents or their blends. Examples of them are trichloroethylene (TCE), perchloroethylene (PCE), toluene, paint thinners, varnish, gasoline, and aromatic hydrocarbons⁷⁷.

Estimations of industrial exposure to solvents may widely vary depending on the type of use or industry. Domestic exposure and use of solvents in hobbies such as cleaning, and furniture and ship restoration are even more variable due to differences in ventilation, lack of training in solvent use, absence of legal regulations, and the idiosyncratic nature of working habits of people in their homes⁷⁷.

Exposure levels in Table 7 represent the upper limit of typical exposures for these agents in several situations, more than their whole rank. These levels could vary from high to trace (Table 7). Some high level exposures are from historical records and they should not happen within the present regulatory framework.

The American Conference of Governmental Industrial Hygienists recommended a maximum level of exposure for an 8 h exposure for each type of solvent. These recommendations (threshold limit value) for average exposure of 8 h are concentrations in the environment’s aerosol not believed to have serious consequences on adult health⁷⁷. Exceeding these limits is not prohibited by any regulatory authority, but they are recognized as reasonable limits in many countries.

The first description of TCE induced SSc reported a group of German workers in 1957⁷⁸. Years later, Walder referred to the association of exposure to solvents and the development of SSc¹⁴.

In 2007, Kettaneh, et al’s metaanalysis of case-control studies about occupational exposure to solvents and the sex-related risk of SSc (Figure 1)^79–86. They stated that occupational exposure to solvents confers an increased risk of SSc to men [OR 3.0 (1.9–4.6), p < 0.0001] compared to women [OR 1.8 (1.5–2.1), p < 0.0001].

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Metanalysis of Kettaneh and colleagues⁷⁹ on exposure to any organic solvent and risk of systemic sclerosis in the whole sample. Points and diamond represent the odds ratio (OR) with 95% confidence interval (CI). The size of each box is approximately proportional to the weight of the study in the metaanalysis.

Other agents

In 1977 Fessel published 2 cases of SSc in the context of an intense professional exposure to welding fumes⁸⁷. In 2002, Diot, et al stated that an existing exposure to these fumes could carry an OR of 3.74 (1.06–13.18) for SSc⁸⁴.

Exposure to vibrations is considered a risk factor for SSc due to case reports of patients using vibrating tools in their professional work^88,89. In 2001 Bovenzi, et al carried 2 case-control studies to verify this occupational hypothesis. The OR obtained in each study were not significant⁹⁰. The study of Maitre, et al in 2004 depicted an OR of 3.9 (0.8–19), statistically nonsignificant due to the limited number of exposed cases⁸⁶.

The process of industrial polymerization of epoxy resins has been associated in several reports with skin sclerosis, arthralgias, and myalgias, but not with visceral involvement nor with RP or autoantibodies¹⁶. Silman and Jones studied a population of 56 men with SSc and found an OR of 1.7 (0.4–7.3) on exposure to epoxy resins⁶⁹. The limited number of studies and the heterogeneity of the results do not permit assigning a role as a risk factor for these substances.

Infectious Factors

Bacterial and viral infections have long been considered as contributing factors to the development and progression of SSc. This hypothesis could be sustained by the fact that many sclerodermiform symptoms are transiently provoked by infectious agents in healthy subjects⁹¹.

Bacterial agents

There are 2 lines of evidence involving bacterial infection in the pathogenesis of SSc. One of them is anecdotal evidence that antibiotic therapy ameliorates SSc symptoms in some people. The other is evidence that graft versus host disease (GVHD), which shares many features with SSc, cannot be induced in “aseptic” animals^92,93 and its incidence is significantly reduced in childrent pretreated with antibiotic to eradicate their normal bacterial flora⁹⁴.

Helicobacter pylori (HP) is a gastric bacterium that has been involved in some vascular disease⁹⁵ and has been studied as a potential risk factor for SSc. In a study of patients with primary RP, eradication of HP infection with a triple antibiotic schedule was associated with disappearance of RP in 17% of treated patients and a decrease of symptoms in another 72% of subjects⁹⁶. Although this was not a double blind trial, it is interesting to observe the lack of improvement in those patients in which HP infection was not completely eradicated. Another study of comparable design reported similar results⁹⁷.

Attempts to associate HP infection and SSc have yielded conflicting results. One study identified high incidence rates of infection in patients with autoimmune disease, such as SSc⁹⁸. In contrast, 2 studies found no differences in HP infection rates between patients with RP and SSc and healthy controls^99,100. Although is true that the rates of HP infection do not correlate with SSc, this does not rule out its role in the disease¹⁰¹. There is evidence that HP infection is not different between patients with SSc and healthy controls, but it has been shown that 90% of patients with SSc were infected with the HP strain CagA, in comparison with 37% of infected controls¹⁰¹. For this reason, confounders like co-infections, differences between HP strains, and host factors should be observed and controlled to understand HP’s role in RP and SSc.

Viral agents

SSc is also associated with viral infection. The pathogenic role of 2 viruses, HCMV and parvovirus B19 (B19), has been recently proposed. Herpes viruses, in particular HCMV, have been involved in several vascular diseases including atherosclerosis, allograft rejection vasculopathy, vascular restenosis, and GVHD.

Neointimal formation is a common feature shared by SSc and all the above vascular diseases. When human cytomegalovirus (HCMV) infects vascular endothelium and this process is characterized by latency, reactivation, and spreading of viruses to distant tissues. In murine models it was shown that CMV infection leads to the development of intimal lesions^102,103. There is indirect evidence of a HCMV role in SSc, linking high levels of HCMV antibodies and the prevalence of specific SSc autoantibodies in patients with scleroderma^104–106. Recently, molecular mimicry mechanisms have shown a relationship between antibodies against the HCMV derived protein UL94 and SSc pathogenesis¹⁰⁶. UL94 epitope shares homology with NAG-2, a surface molecule highly expressed in endothelial cells. Antibodies against UL94 purified from sera of patients with SSc induce endothelial cell apoptosis by enhancing a NAG-2/integrin complex. NAG-2 is expressed in dermal fibroblasts and anti-UL94 bind to fibroblasts activating them, with subsequent fibrosis.

HCMV infection and its downstream effects on immune, vascular, and repairing effects serve as a trigger for SSc. The identification of genes controlling susceptibility to CMV in mice by genetic analysis may be a path to understanding human genetic susceptibility to HCMV infection.

Strain C57BL/6 mice, genetically resistant to CMV, do not develop vascular pathology in response to the infection, while susceptible mice (129 interferon R−/−) develop neointimal lesions that reproduce CMV infection in a dose dependent manner¹⁰⁷.

Microchimerism is another possible mechanism to explain sporadic development of SSc regarding its 8:1 rate of women/men affected by the disease. Microchimerism refers to prolonged allotypic lymphocyte survival (fetal T cells acquired during pregnancy or cells received by transfusion or organ transplant), usually in circulating blood. Fetal microchimeric T cells were isolated 27 years after delivery in a patient with SSc, and it has been shown that they are more common in patients with SSc than in age matched healthy controls¹⁰⁸. Pathogenic effects of these cells are unknown, but microchimeric cells were also found in cell infiltrates of scleroderma lesions¹⁰⁹. Engraftment and survival of these cells depends on the complex relationship between tissue antigens from the mother and the fetus (or host and donor), and so, they are highly variable¹¹⁰.

In vitro studies showing T cells exposed to allotypic endothelial cells activate and proliferate more if they are infected by HCMV, sustaining the idea that HCMV can induce proliferation of microchimeric cells¹¹¹.

Vascular endothelium is an allotypic stimulus for circulating microchimeric cells. If endothelium is infected with HCMV, cytokine production may be amplified, possibly triggering endothelial cell activation, vascular inflammation, and neointimal formation such as it happens with T cells transplanted in GVHD.

Genomic sequences of B19 have been found in 57% (12/21) of bone marrow specimens from patients with SSc and in none of the control group. It has been speculated that bone marrow could be a reservoir of the virus from which viruses disseminate to target tissues in SSc¹¹². There was also an increase in the frequency of anti NS1-B19 antibodies in sera of patients with SSc (33%) compared to controls (13%).

In a case series, ANA with a nucleolar pattern were detected in a patient with B19 arthritis¹¹³.

Altschuler¹¹⁴ highlighted that SSc is a relatively new disease, with the first case reported in 1753¹¹⁵, coincidental with the B19 virus reintroduction in Europe (in the XVI century).

Non-occupational/Non-infectious Factors

There are few epidemiologic studies regarding such factors and SSc, even though there is a wide list of them (Figure 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Epidemiologic studies regarding non-occupational/non-infectious factors and systemic sclerosis.

Drugs — estrogens

Beebe, et al¹¹⁶ studied the association between reproductive history, oral contraceptives, and estrogen replacement therapy in 472 women with SSc and 2227 controls. They did not find a relationship between risk of Ssc and contraceptive intake, early menarche, or multiple delivery. Hormone replacement therapy with estrogens showed an association with a mild but significant increase of SSc, with an adjusted OR of 1.40 (1.10–1.77). Mean age at start of therapy was not different between cases and controls, suggesting that early menopause did not cause the results.

Fraenkel and colleagues¹¹⁷ assessed the association between estrogen replacement therapy and RP in 497 postmenopausal women in the US Framingham Offspring Study. RP prevalence was 8.4% in women not receiving treatment, 19.1% of those receiving estrogens and 9.8% of women taking estrogens plus progesterone. Adjusted OR was 2.5 (1.2–5.3) for estrogens and 0.9 (0.3–2.6) for estrogens plus progesterone. This risk was associated with development of the disease after initiation of treatment, and does not necessarily imply a worse prognosis by their intake itself¹¹⁸.

Appetite suppressants

There are a few case reports documenting SSc after prolonged use of these drugs. Examples of them are dexamphetamine, diethylpropion, fenfluramine, fenproporex, mazindol, methaqualone, and phenmetrazine¹¹⁹. Duration of therapy, variety of drugs used, and lack of prescription criteria in the general population make it difficult to analyze if there is a chance of co-occurrence or if it is a real phenomenon.

Carbidopa and L-5 hydroxytryptophan (5-HTP)

Carbidopa inhibits conversion of 5-HTP to 5-hydroxytryptomen and the alternative pathway to tryptophan metabolism leads to an increased synthesis of quineurin. High levels of this molecule are associated with SSc, and this is not observed in patients continuously treated with carbidopa/L-5HTP.

A patient receiving carbidopa as treatment for myoclonus developed SSc after therapy¹²⁰.

Bleomycin

Bleomycin is an antitumoral antibiotic drug used for several types of cancer, primarily isolated from Streptomyces verticillus¹²¹. Lung fibrosis is a known side effect of this drug. Skin changes like hyperpigmentation, alopecia, gangrene, edema, RP, fibrosis, scleroderma, and others were described in relation with bleomycin¹²².

Bleomycin induced lung fibrosis is an established murine model, pathologically and biochemically similar to human lung fibrosis. Mice develop acute alveolitis followed by an intense interstitial inflammation when bleomycin is administered intratracheally. Late in its evolution, there is an increase of the extracellular matrix (collagen, fibronectin, hyaluronates, and small proteoglycans)^123,124.

In vitro, this drug upregulates collagen messenger RNA in the lungs and dermal fibroblasts^125,126. Mountz, et al reported that mice repeatedly injected with sublethal doses of bleomycin develop severe dermal fibrosis — similar to human SSc — in a 58-week period, with structural anomalies in collagen fibers¹²⁷.

Five cases of scleroderma-like disease following bleomycin use were described in 2 case reports^18,128. Bleomycin use enhanced an in vivo lymphoproliferative response in one of these patients¹⁸.

Gadolinium

Nephrogenic systemic fibrosis (NSF)¹²⁹ was described in 2000¹³⁰. This entity is characterized by extensive cutaneous fibrosis similar to SSc, sparing the face and neck in the absence of autoantibodies, also involving lungs, heart, liver and muscle¹³¹.

A link with contrast agents containing gadolinium was first established in 2006¹³², leading regulatory bureaus to set recommendations for its use. In 2007, the US Food and Drug Administration warned health professionals to avoid gadolinium use in patients at risk of developing NSF, to evaluate renal function, not to exceed the dose recommended on the product label of this substance, and eventually, to consider immediate hemodialysis after infusion of a gadolinium-containing agent¹³³.

Other drugs

Cases of cutaneous sclerosis proven histologically — even without clinical features of Ssc — have been described in relation to pentazocine abuse²¹.

A few articles have linked cocaine abuse with SSc development in young men^134,135. Cocaine causes vasoconstriction, and may constitute a first step in the pathogenesis of SSc in a susceptible host.

Mercury chloride is an immunomodulating agent that may cause immunocomplex mediated glomerulonephritis and autoantibody production in susceptible mice strains^136,137. These autoantibodies recognize the fibrillar protein fibrillarin¹³⁸. Arnett, et al suggested that mercury modifies fibrillarin, causing autoantigenicity¹³⁹. Anti-fibrillarin antibodies are found in 8% of patients with SSc and are associated with severe forms of the disease.

Pesticides

There are potential endocrine modulators under the form of pesticides that could also increase the risk of SSc. Some pesticides are known as endocrine disruptors, like polychlorinated biphenyls (PCB), DDT metabolites, and 2.3.7.8-tetrachlorinedibenzo p-dioxin (TCDD).

PCB exert an anti-estrogen effect by inhibition of estrogen-mediated transduction signals^140,141. Similarly, a DDT metabolite (p,p-DDE) may bind the aryl-hydrocarbon receptor, which plays a role in hormone signaling¹⁴². Acute and chronic exposure of mice to TCDD affects their bacterial pathogen innate immunity by supressing total hemolytic complement and C3¹⁴³.

A case of SSc was documented after exposure to a combination of herbicides: bromobuthyl-methyluracil, dichlorophenyl-dimethylurea, and aminotriazole¹⁴⁴. Aminotriazol may cause contact dermatitis¹⁴⁵, but none of these compounds was previously associated with autoimmune diseases. In a case-control study in women, exposure to pesticides and herbicides by self-reference showed an adjusted OR of 2.19 for SSc, without distinction between both exposures¹⁴⁶.

Silicones

These are synthetic polymers with organic groups bound to silicone atoms by carbon-silicon bonds; polydimethyl siloxane is the most commonly used silicone in medicine¹⁴⁷.

Silicone gel has been used since 1962 to fill breast implants¹¹⁸. Silicone elastomer containing chemically treated amorphous silica is added to the capsule of certain breast augmentation implants to increase its tensile strength, reducing the potential of implant rupture¹⁴⁸.

The first report of a connective tissue disease related to silicone was in a young woman who received silicone injections for breast augmentation¹⁴⁹. Other reports referred to this circumstance as “adjuvant induced disease”¹⁵⁰.

Association between silicone breast implants and the occurrence of SSc was not observed in several epidemiological studies^71,151–154. The risk for SSc in implanted medical devices was also evaluated⁷¹ but this study did not find an association.