Drugs in Pregnancy and Lactation: Tenth Edition






Approximately 2 million women in the United States alone have received silicone implants (1), but no estimation of the number of pregnant women exposed to these devices has been located. The passage of polydimethylsiloxane (PDMS) to the fetus should not occur because of the high molecular weight of this polymer, and no evidence has been published that any of its multiple breakdown products cross the placenta. However, silicon, the second most abundant element in the earth’s crust and a major component of biologic systems including the human skeleton (2), crosses to the fetus with concentrations in the amniotic fluid ranging from 34 to 800 ng/mL (mean 154.7 ng/mL) at 16–19 weeks’ gestation (3). Transplacental passage of maternal immunoglobulin antibodies (IgG, IgA, or IgM) from silicone-induced immune disease is potentially possible, but this was not found in the study cited below. Moreover, some data presented in the Breastfeeding Summary argue against the clinical significance of this occurrence. Furthermore, a meta-analysis found no evidence that silicone implants are associated with connective-tissue or autoimmune diseases.


Silicone breast implants are composed of a shell of high-molecular-weight PDMS (dimethicone) gum (i.e., elastomer or rubber) containing either saline, a silicone “oil” composed of unlinked polymers, or a lower-viscosity PDMS gel (2,4). The difference in the viscosity of the organosiloxane shell and filler is dependent on the average molecular weight and molecular number distribution of the polymer (4). Leakage (bleeding) of the filler onto the surface of the shell may occur by simple diffusion and is associated with contracture of fibrous tissue around the implant, a foreign body response, or both. Gel bleeding may decrease the tensile strength of the shell and, by implication, may increase the incidence of implant rupture. The prevalence of implant rupture has been estimated to be 4%–6% (4).

Silicone breast implants have been generally available since the early 1980s, although they were first used experimentally in the 1940s (2). Several serious health concerns have been raised in relation to these implants, including silicone gel implant bleed, contracture (moderate degrees of contracture may be beneficial), implant rupture, carcinogenesis, immune disorders, and impaired breast cancer detection (1,2,4). However, a causal relationship between breast implants and connective-tissue and autoimmune disease seems unlikely based on the results of a meta-analysis described below.

A 2000 study conducted a meta-analysis of 20 studies (9 cohort, 9 case–control, and 2 cross-sectional) to determine if breast implants were associated with an increased risk of connective-tissue and autoimmune diseases (5). The study was part of a report prepared by a scientific panel formed to advise the federal judiciary on silicone breast implants (6). No evidence of an increase in individual connective-tissue diseases (rheumatoid arthritis, systemic lupus erythematosus, scleroderma or systemic sclerosis, and Sjogren’s syndrome), all connective-tissue diseases combined, or other autoimmune or rheumatic conditions was found. Similarly, no evidence of a significantly increased risk was found specifically for silicone-gel-filled implants (5).

A study published in 1996 found no significant difference in the prevalence of autoantibodies between children (N = 80) born to mothers with silicone breast implants and control children (N = 42) born to mothers without implants (7). Moreover, no association between the clinical symptoms in the children and the presence of autoantibodies was found. Control children had been referred to the authors because of irritable bowel syndrome or lactose intolerance (N = 21) or fibromyalgia (N = 21), whereas the children in the study group had been referred because of concerns about adverse effects from the mother’s implants. The authors concluded that determination of the antibodies was of limited clinical value in this patient population (7).

In a three-part study, investigators studied whether the immunogenicity of silicone, which appears to have been confirmed in humans and in at least one animal model, could be transferred from the mother to her offspring (8). In part 1 of the study, using silicon dioxide (silica), T lymphocyte cell-mediated immune responses were elicited in 21 of 24 children from 15 women with silicone breast implants. Subjects in part 2 of the study were the offspring of three women who gave birth to four children before they received their implants and to seven children after the implants. Five of the postimplant offspring were found to be positive to T-cell memory for silica compared with none of the preimplant offspring. Part 3 was a blinded study that evaluated 30 children of mothers with silicone implants compared with 10 control children of mothers without implants. A significant increase in T-cell stimulation was measured in the exposed children in comparison to the controls. Because not all of the above offspring were breastfed, the investigators concluded that the results indicated either the transplacental passage of immunogens from silicone or the transfer by maternal–fetal cellular exchange (8).

A 1998 epidemiologic cohort study conducted in Denmark examined the occurrence of esophageal disorders, connective tissue diseases, and congenital malformations in children (born during 1977–1992) of mothers with breast implants (9). A total of 939 children born from mothers with breast implants for cosmetic reasons (660 before the implant surgery and 279 after) were compared with 3906 children born from mothers who had undergone breast reduction surgery (1739 before surgery and 2167 after). The mean times from surgery to delivery for the two groups were 5.0 and 5.5 years, respectively. In the implant group, the observed/expected ratios (95% confidence intervals in parentheses) for various offspring outcomes before and after implants were as follows: esophageal disorders 2.7 (1.4–4.7) and 2.9 (0.8–7.4); rheumatic diseases 0.7 (0.0–4.1) and 0.0 (0.0–12.9); all types of congenital malformations 1.2 (0.9–1.5) and 1.3 (0.8–2.0); and defects of the digestive organs 1.4 (0.5–3.0) and 1.3 (0.2–4.6). Among the four children with an esophageal disorder who were born after the implants, three were hospitalized because of mild regurgitation that resolved without treatment. The fourth case involved a child with microcephaly, which was diagnosed as part of a genetic syndrome, and cerebral palsy. The results provided no evidence that silicone breast implants were associated with an increased risk of connective tissue diseases or malformations. Similar, nonsignificant differences for each of the categories were also found in the breast reduction group. The investigators concluded that women in both groups were more likely to seek professional medical care for problems normally solved outside the hospital (9).


Studies concerning the excretion of PDMS (see above) or the breakdown products of this macromolecule into milk have not been located, but two reports discussed below have described unusual disease symptoms in breastfed infants of mothers with silicone breast implants (10,11). For the present, any association between the symptoms and the organosiloxane components of the implants is speculative and further studies are needed to establish a causal relationship.

A 1994 report described 67 (56 breastfed, 11 bottle-fed) children born to mothers with silicone breast implants, who had been self-referred because of concerns relating to implant-induced toxicity (10). Forty-three (35 breastfed, 8 bottle-fed) of the children had complaints of recurrent abdominal pain, and 26 (20 breastfed, 6 bottle-fed) of this group had additional symptoms, such as recurrent vomiting, dysphagia, decreased weight:height ratio, or a sibling with these complaints. Of these latter 26 children, 11 (8 breastfed, 3 bottle-fed; mean age 6.0 years, range 1.5–13 years; 6 boys and 5 girls) agreed to undergo further evaluation. A group composed of 17 subjects (mean age 10.7 years; range 2–18 years; 11 boys and 6 girls) with abdominal pain who had not been exposed to silicone breast implants served as controls. No significant differences were found between the implant-exposed breastfed and bottle-fed subjects or between the total exposed group and controls (7 of the 17 were tested) in autoantibodies to eight antigens (nuclear, Sci-70, centromere, ribonucleoprotein, Sm, Ro, La, and phospholipid). Endoscopy was performed on all subjects, and no gross visual abnormalities were observed. Chronic esophagitis was discovered on biopsy specimens in 8 exposed (6 breastfed, 2 bottle-fed) children (all graded as mild) and in 13 of 16 controls (1 not tested) (mild-to-moderate in 7, severe in 6). No granulomas or crystals were identified in the biopsy specimens. The histology of the specimens did not differ between the exposed children or between the exposed children and controls. When esophageal manometry was used to test the esophageal motility of the subjects, six of the eight breastfed exposed children were found to have significantly abnormal motility with nearly absent peristalsis in the distal two-thirds of the esophagus. Esophageal sphincter pressure, esophageal wave propagation, and wave amplitude were measured and compared in the breastfed exposed, bottle-fed exposed, and control groups by an investigator blinded to the clinical status of the children. The following results were obtained in the three groups: 13.1 (ns compared with controls), 22.7, and 24.8 mmHg, respectively; 14.7% (ns compared with controls), 64.3%, and 53.0%, respectively; and 42.3, 60.3, and 50.6 mmHg, respectively. No improvement in the motility abnormalities was found in three of the breastfed exposed subjects who were retested 10 months later after long-term ranitidine therapy had reduced the episodes of abdominal pain (10).

The symptoms present in the breastfed exposed children were considered to be characteristic of systemic sclerosis, although the children did not meet the clinical criteria for the disease (10). Moreover, the investigators excluded the possibility that the abnormal motility was a consequence of chronic esophagitis. The blinded manometric findings suggested that the esophageal disorder might have been related to exposure to substances in breast milk because the bottle-fed exposed children had values similar to those of controls. The nature of these substances, if any, could not be determined by this study, but the investigators considered the possibilities to include silicone and other breakdown products of the implants that could be transferred across the immature intestinal barrier of the nursing infant and eventually lead to immunologically mediated damage (10).

In an accompanying editorial to the above report, several possible mechanisms for silicone-induced toxicity were explored (12). These included mother-to-child transmission of silicone products or maternal autoantibodies across the placenta or through breast milk. However, an argument against the latter mechanism is the usually short-term effects of passively acquired antibodies compared with the prolonged nature of the disorders in the affected children (12).

The second study, also published in 1994, described two female children, ages 2.67 and 9 years of age, who had long-standing, unusual, diffuse myalgias and arthralgias, not consistent with juvenile arthritis, and positive antinuclear antibodies (1:80 and 1:160, respectively; both speckled pattern) (11). The 9-year-old girl had a markedly elevated titer of antibodies against denatured human type II collagen. Both girls had been breastfed, the youngest for 3 months and the other for 6 months, by mothers with silicone breast implants. The right implant in the mother of the youngest girl had ruptured during pregnancy, and a recent breast ultrasound of the other mother was suggestive for implant rupture, but the timing was unknown (11).

A number of comments were published in response to the above two studies (1325). In one of the comments, the authors described the results of a study in which no silicone was detected (detection level 0.5 mcg/mL) in two women with silicone breast implants (17). Two of the references stated that without more evidence, breastfeeding by women with silicone breast implants should not be contraindicated (22) or should be recommended (23).

Macrophage activation was suggested from the results of a case–control study of 38 breastfed children from mothers with silicone breast implants compared with 30 controls (healthy children N = 10, children with gastrointestinal symptoms similar to study patients N = 10, children with benign urinary abnormalities N = 7, and children with joint symptoms similar to study patients N = 3) (26). Researchers measured the urinary excretion of stable nitric oxide (NO) metabolites (NO3 plus NO2) and neopterin, inflammatory mediators released by phagocytosis of foreign material by macrophages. Mean levels of NO metabolites in the study patients were higher than those in controls, but significantly higher only when compared with levels in the subgroup of healthy children. Mean neopterin excretion in study patients was higher than each of the four control subgroups, but significantly so only in comparison to the healthy, gastrointestinal, and joint symptom subgroups. The investigators speculated that macrophage activation by silicone results in the release of NO and other substances with subsequent inhibition of esophageal peristalsis (26).

At follow-up (mean 2.1 years) of 11 children with esophageal dysmotility who had been breastfed by mothers with silicone breast implants, 7 had subjective clinical improvement in their symptoms (27). Esophageal sphincter pressures (both lower and upper) and percent of wave propagation into the distal esophagus following swallowing were statistically similar to the values obtained at initial manometric testing. Wave amplitude in the distal esophagus, however, did increase significantly. Urinary neopterin decreased significantly, whereas urinary nitrates (NO3 plus NO2) decreased but not significantly. The data suggested that the dysmotility had become a chronic condition in this group of children (27).

In a 1998 study, silicon concentrations in milk and blood of 15 women with bilateral silicone gel-filled implants were compared with similar samples in 34 women with no implants, store-bought cow’s milk, and 26 brands of commercially available infant formula (28). Silicon was used as a “proxy” measure for silicone. Mean milk and blood silicon levels were statistically similar in the implant group compared with women without implants: 55.45 and 79.29 ng/mL vs. 51.05 and 103.76 ng/mL, respectively. Much higher mean silicon concentrations were found in cow’s milk (708.94 ng/mL) and commercial infant formula (4402.5 ng/mL) (28).

Two studies have described unusual signs and symptoms in children who had been breastfed by mothers with silicone breast implants. Follow-up studies by the same investigators suggested that these effects may have been caused by the transfer of maternal mutagenicity to silicone to the fetuses or infants during pregnancy or breastfeeding and that the pathogenesis in the offspring might be caused by macrophage activation. However, these conclusions are controversial (2931) and have not been confirmed. Many experts recommend that women with silicone breast implants should be encouraged to breastfeed because the benefits of breastfeeding appear to far outweigh the potential, if any, risk to the nursing infant. Mothers with silicone breast implants should be fully informed of the current state of knowledge so that they are actively involved in the decision whether to breastfeed. The American Academy of Pediatrics concluded that the available evidence does not justify classifying silicone implants as a contraindication to breastfeeding (32).


1.Kessler DA, Merkatz RB, Schapiro R. A call for higher standards for breast implants. JAMA 1993;270:2607–8.

2.Yoshida SH, Chang CC, Teuber SS, Gershwin ME. Silicon and silicone: theoretical and clinical implications of breast implants. Regul Toxicol Pharmacol 1993;17:3–18.

3.Hall GS, Carr MJ, Cummings E, Lee M-L. Aluminum, barium, silicon, and strontium in amniotic fluid by emission spectrometry. Clin Chem 1983;29:1318.

4.Council on Scientific Affairs, American Medical Association. Silicone gel breast implants. JAMA 1993;270:2602–6.

5.Janowsky EC, Kupper LL, Hulka BS. Meta-analyses of the relation between silicone breast implants and the risk of connective-tissue diseases. N Engl J Med 2000;342:781–90.

6.Hulka BS, Kerkvliet NL, Tugwell P. Experience of a scientific panel formed to advise the federal judiciary on silicone breast implants. N Engl J Med 2000;342:812–5.

7.Levine JJ, Lin H-C, Rowley M, Cook A, Teuber SS, Ilowite NT. Lack of autoantibody expression in children born to mothers with silicone breast implants. Pediatrics 1996;97:243–5.

8.Smalley DL, Levine JJ, Shanklin DR, Hall MF, Stevens MV. Lymphocyte response to silica among offspring of silicone breast implant recipients. Immunobiol 1996/1997;196:567–74.

9.Kjoller K, McLaughlin JK, Friis S, Blot WJ, Mellemkjaer L, Hogsted C, Winther JF, Olsen JH. Health outcomes in offspring of mothers with breast implants. Pediatrics 1998;102:1112–5.

10.Levine JJ, Ilowite NT. Sclerodermalike esophageal disease in children breast-fed by mothers with silicone breast implants. JAMA 1994; 271:213–6.

11.Teuber SS, Gershwin ME. Autoantibodies and clinical rheumatic complaints in two children of women with silicone gel breast implants. Int Arch Allergy Immunol 1994;103:105–8.

12.Flick JA. Silicone implants and esophageal dysmotility. Are breast-fed infants at risk? JAMA 1994;271:240–1.

13.Bartel DR. Sclerodermalike esophageal disease in children of mothers with silicone breast implants. JAMA 1994;272:767.

14.Cook RR. Sclerodermalike esophageal disease in children of mothers with silicone breast implants. JAMA 1994;272:767–8.

15.Epstein WA. Sclerodermalike esophageal disease in children of mothers with silicone breast implants. JAMA 1994;272:768.

16.Placik OJ. Sclerodermalike esophageal disease in children of mothers with silicone breast implants. JAMA 1994;272:768–9.

17.Liau M, Ito S, Koren G. Sclerodermalike esophageal disease in children of mothers with silicone breast implants. JAMA 1994;272:769.

18.Levine JJ, Ilowite NT. Sclerodermalike esophageal disease in children of mothers with silicone breast implants. In reply. JAMA 1994;272:769–70.

19.Brody GS. Sclerodermalike esophageal disease in children of mothers with silicone breast implants. JAMA 1994;272:770.

20.Flick JA. Sclerodermalike esophageal disease in children of mothers with silicone breast implants. In reply. JAMA 1994;272:770.

21.Williams AF. Silicone breast implants, breastfeeding, and scleroderma. Lancet 1994;343:1043–4.

22.Berlin CM Jr. Silicone breast implants and breast-feeding. Pediatrics 1994;94:547–9.

23.Jordan ME, Blum RWM. Should breast-feeding by women with silicone implants be recommended? Arch Pediatr Adolesc Med 1996;150:880–1.

24.Epstein WA. Silicone breast implants and sclerodermalike esophageal disease in breast-fed infants. JAMA 1996;275:184.

25.Levine JJ, Ilowite NT. Silicone breast implants and sclerodermalike esophageal disease in breast-fed infants. JAMA 1996;275:184–5.

26.Levine JJ, Ilowite NT, Pettei MJ, Trachtman H. Increased urinary NO3 + NO2 and neopterin excretion in children breast fed by mothers with silicone breast implants: evidence for macrophage activation. J Rheumatol 1996;23:1083–7.

27.Levine JJ, Trachtman H, Gold DM, Pettei MJ. Esophageal dysmotility in children breast-fed by mothers with silicone breast implants. Long-term follow-up and response to treatment. Dig Dis Sci 1996;41:1600–3.

28.Semple JL, Lugowski SJ, Baines CJ, Smith DC, McHugh A. Breast milk contamination and silicone implants: preliminary results using silicon as a proxy measurement for silicone. Plast Reconstr Surg 1998; 102:528–33.

29.Epstein WA. Silicone breast implants and breast feeding. J Rheumatol 1997;24:1013.

30.Hoshaw SJ, Klykken PC, Abbott JP. Silicone breast implants and breast feeding. J Rheumatol 1997;24:1014.

31.Levine JL, Ilowite NT, Pettei MJ, Trachtman H. Silicone breast implants and breast feeding. J Rheumatol 1997;24:1014–5.

32.Committee on Drugs, American Academy of Pediatrics. The transfer of drugs and other chemicals into human milk. Pediatrics 2001; 108:776–89.