Background Local complications (encapsulation, rashes, rupture, and le
akage) can occur after placement of silicone-gel-containing breast imp
lants (SBI). Whether SBI exposure results in systemic manifestations i
n some recipients is controversial. We have carried out a blinded stud
y to assess whether there is any difference between SBI recipients and
non-exposed controls in the proportions positive for serum antibodies
directed against polymeric substances. Methods We recruited female SB
I recipients (including those without symptoms) who presented to a sin
gle rheumatology clinic, A physician global assessment was used to cla
ssify SBI recipients who did not meet criteria for specific autoimmune
diseases according to the severity of local and systemic signs and sy
mptoms. Controls were recruited from among clinic staff and their acqu
aintances. Results of the antipolymer antibody (APA) assay were compar
ed with those of an assay for antinuclear antibodies (ANA) and with th
e severity of the signs and symptoms. Findings Positive APA results we
re found in one (3%) of 34 SBI recipients with limited symptoms, two (
8%) of 26 with mild symptoms, seven (44%) of 16 with moderate symptoms
, and 13 (68%) of 19 with advanced symptoms. Four (17%) of 23 healthy
non-SBI-exposed controls and two (10%) of 20 non-exposed women with cl
assic autoimmune diseases were positive for APA. Thus, women with mode
rate or advanced symptoms were significantly more likely than those wi
th limited or mild symptoms, or nonexposed controls to have APA (p<0.0
01). The proportion with positive ANA results was higher for women wit
h classic autoimmune diseases 14 (70%) of 20 than for any SBI-exposed
subgroup (0-33%). Interpretation The APA assay can objectively contrib
ute to distinguishing between SBI recipients with limited or mild sign
s and symptoms, SBI recipients with more severe manifestations, and pa
tients with specific autoimmune diseases. Further studies will be need
ed to define the signs and symptoms associated with exposure to SBI.