There are sporadic references in the literature to single cases or small series of cases where an association between silicone gel implants and autoimmune diseases was claimed.

The hypothesis that silicone induces connective-tissue diseases was generated from such descriptive uncontrolled studies.

About 300 patients have been reported worldwide with rheumatic symptoms after receiving gel-filled breast implants. The most commonly reported connective-tissue disease was scleroderma, but several other connective-tissue disorders and vague musculoskeletal symptoms have also been described, particularly vague aches and pains (fibromyalgia), sleep disorders and fatigue - symptoms which are extraordinarily common in the community.
First definitive studies to examine the relationship between silicone implants and autoimmune disease appeared in 1994. Englert identified 556 cases of scleroderma in women who had resided in Sydney, of whom 270 were living and 213 were deceased (73 were "living status unknown").
They reported that the rates of augmentation mammoplasty were similar between the 251 women with scleroderma that they interviewed and 289 matched controls (the study carried a 90% chance of detecting a relative risk of 4.5).
These results were validated when the authors found no association between silicone breast implants and scleroderma (odds ratio [OR], 1.33; 95% confidence interval [CI], 0.26-6.71, and OR, 1.00; 95% CI, 0.16-6.16, after adjustment for confounders age, socioeconomic status and ethnicity).
Validation of augmentation mammoplasty status was possible in 532 of the original 556 cases of scleroderma.
Larger epidemiological studies were conducted in the United States. 

Gabriel performed a population-based case-control study of 749 women who had received a breast implant in Olmsted County, Minnesota.
Each subject was matched with two women of the same age who had not had a breast implant, but had had a medical evaluation within two years of the date of the subjects implant.
The implant group were followed for a mean of 7.8 years and the control group for a mean of 8.3 years.
The authors sought evidence from the medical records for a diagnosis of any connective-tissue diseases, autoimmune diseases or non-breast cancer, as well as for related symptoms. Only morning stiffness was significantly more common in the implant group. Five women with implants and 10 women without implants were diagnosed with one of the specified connective-tissue diseases. The authors found no statistically significant elevation in the relative risk of any of the specified connective-tissue diseases or other disorders in women with silicone breast implants.

Sanchez-Guerrero from Harvard Medical School reported on a cohort of over 120 000 registered nurses aged between 30 and 55 who had been followed up.
The mean follow-up period after surgery for the 1183 women with silicone breast implants was 9.9 years.The age-adjusted relative risk of definite connective-tissue disease in women with implants was 0.3 (95% CI, 0-1.9). The relative risk of self-reported signs or symptoms of connective-tissue disease for women with implants was 1.5 (95% CI, 0.9-2.4), and the risk of having any one of 41 signs, symptoms or laboratory features of connective-tissue disease was 0.7 (95% CI, 0.3-1.6). The authors concluded that there was no association between silicone breast implants and connective-tissue diseases, or signs or symptoms of these diseases. 

The American College of Rheumatology released a statement declaring that these studies provided compelling evidence that silicone implants expose patients to no demonstrable additional risk for connective-tissue or rheumatic disease.
The College affirmed that anecdotal evidence, while of importance in drawing attention to a potential problem, should no longer be used to support this relationship in the courts or by the Food and Drug Administration (FDA). They recognised that many women who have received silicone breast implants have musculoskeletal complaints that are also very common in the general population. They had stated in 1994 the importance and great need for scientific analysis of this question, and called upon the FDA and other regulatory agencies to allow professional societies to foster epidemiological studies. 

The only study suggesting a possible link was published by
Hennekens. He retrospectively reviewed a large cohort of 395 543 health professionals and found 10 830 women who reported breast implants and 11 805 who reported connective-tissue diseases. The relative risk of any connective-tissue disease among those reporting implants was 1.24 (95% CI, 1.08-1.41; P = 0.0015).

This indicated a small, but significant, increase in the risk of connective-tissue disease, but provided reassuring evidence against silicone implants being a large-scale hazard.

The results fell within the 95% confidence limits of the other two major studies.

The authors stress that biases from self-reporting of symptoms (questionnaires were sent to the women after the publicity surrounding the FDA ban) or a higher participation rate in women with such diseases must be considered as alternative explanations for their results.