Full text of DOI:10.1007/s12009-001-0056-3

A B S T R A C T
The purpose of this study was to assess whether intraocular silicone
oil causes systemic antipolymer antibody (APA) production at a
greater incidence than in individuals without exposure to intraocu-
lar silicone oil. The cross-sectional, randomized sample was
obtained from 3 university-based hospitals (21 patients with com-
plex vitreoretinal disorders) and treated with intraocular silicone
oil. Compared with the general population, patients treated with sil-
icone oil did not have an increased incidence of APA production.
O R I G I N A L A R T I C L E
Serum Antibody Testing in Patients
Treated With Intraocular Silicone Oil
Cameron G. Javid, MD, Donald P. Maxwell,
Jr., MD, James E. Diamond, MD, & Robert F.
Garry, PhD
n the past several years, silicone gel-filled breast
I
implants have been a controversial and highly pub-
licized issue. The association between silicone and sys-
temic connective tissue disease remains controversial.
A previously published study showed a statistically
significant correlation between various symptoms of
these inflammatory diseases and the presence of
serum antipolymer antibodies (APA) to silicone oil in
patients exposed to silicone breast implants (SBI).1
Antipolymer antibodies have been shown to be cross-
reactive with antisilicone antibodies (unpublished
data, Y. Cao, S. A. Tenenbaum, R. F. Garry, PhD, 1997).
Cross-reactivity between antipolymer antibodies and
silicone antibodies.
Silicone oil has been used in ophthalmology for
several years as a treatment for complicated retinal
detachment, proliferative vitreoretinopathy, severe
diabetic retinopathy, and giant retinal tears.2 Silicone
oil has been left in the eye for extended time peri-
ods.2 In a literature search, we found no published
data on the association between intraocular silicone
oil and a serum antibody response in humans. This
pilot study is believed to be the first to attempt to
address this topic.
Reprints:
Cameron G. Javid, MD, Retina Associates, 6561 E. Carondelet Drive, Tucson, AZ 85710.
Drs. Javid, Maxwell, and Diamond are from the Department of Ophthalmology and Dr.
Garry is from the Department of Microbiology, Tulane University School of Medicine,
New Orleans, La.
This work was presented as a scientific poster, 1997 Vitreous Society Annual Meeting,
New Orleans, La., September 18-21, 1997.
Acknowledgment
Dr. Garry acknowledges financial support from Autoimmune Technologies, LLC, New
Orleans, La. The other authors hold no proprietary interest in any of the materials or
methods described herein.
ANN OPHTHALMOL. 2001;33(1):53–55
53

Patients & Methods
T A B L E 1
Between May and June 1997, twenty-one patients with
complex vitreoretinal disease underwent surgical
repair using intraocular silicone oil. Following
informed consent, a randomized postoperative sample
of peripheral blood was obtained from all patients. The
time between retinal surgery and assay ranged from 2
weeks to 22 months. The average time of silicone oil
exposure was 5.92 months. Patients were operated on
and followed up at the Tulane University Medical Cen-
ter, the Medical Center of Louisiana (Charity Hospital),
or the Veterans Administration Hospital, all in New
Orleans, La. All samples were processed by the Tulane
University Medical Center’s Department of Microbiol-
ogy using the APA test to silicone, which has been pre-
viously described.1 To summarize briefly, this test is a
blood assay to an antigen of a high-molecular-weight
complex of synthetic polymers. It has been shown that
serum antibody production is based on the presence of
these polymers, not to a specific protein.1,3
Silicone Antipolymer Antibody Assay Results
Silicone Oil
Exposure (months)
Antibody
Patient
Response*
1
2
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
3
7
4
12
8
5
6
3
7
2
8
6
9
7
10
11
12
13
14
15
16
17
18
19
20
21
7
4
4
Aliquots of the polymeric complex were applied to
nitrocellulose membranes and allowed to air dry. The
samples were diluted in distilled water tenfold, 100-
fold, and 1000-fold. Sample strips, 4-mm wide, of the
nitrocellulose membranes were placed in 20-well trays
(MedTec Inc, New Bern, NC) The strips were then
rinsed with washing buffer (Tris-buffered saline con-
taining 0.3% polyoxyethylene sorbitan monolaurate
and 0.005% thimerosal, pH 7.4) for 45 minutes before
the addition of 5 µL of patient serum (1:400 dilution).
These samples were incubated in 2 mL of blocking
buffer (Tris-buffered saline containing 5% powdered
instant milk, 4% goat serum, and 0.008% thimerosal,
pH 7.4). Serum samples and the strips were combined
and incubated for an additional 90 minutes. These
steps were performed on a rocking platform at room
temperature. Following these incubations and wash-
ings, the blocking buffer was removed. The strips
were washed 3 times, for 5 minutes each time, in
washing buffer. After washing, 2 mL of another block-
ing buffer, containing biotin conjugated goat anti-
human IgG (Kirkengaard & Perry Laboratories,
Gaithersburg, Md), diluted 1:1000, was added to the
nitrocellulose trips. After 60 minutes of incubation,
the strips were washed as described earlier. Blocking
buffer containing avidin conjugated horseradish per-
oxidase (Jackson ImmunoResearch Laboratories Inc,
West Grove, Pa) was then added (2 mL, diluted 1:500).
The strips were incubated for another 60 minutes and
washed again. Then, 2 mL of detection buffer (phos-
phate-buffered saline containing 30% methanol; 0.6
mg/mL of 4-chloro-1-napthol; and 0.03% hydrogen per-
oxide, pH 7.4) was added. Rinsing the strips in dis-
tilled water stopped the reaction after 15 minutes.
After the assay, visual scoring and image analysis
for positive reactions were performed.
6
10
8
9
0.5
0.5
3
1.5
22
*On the scale developed by Robert F. Garry, PhD, 0
indicates absolute negative in response to the con-
trol; 1+, minimal response; 2+, weak response;
3+, low positive response; and 4+, strong positive
response.
previous report,4 we observed that 3 of 21 patients
gave a weak or minimal positive result. A different
configuration of the APA assay was used in this study.
The original samples were reassayed using the cur-
rent protocol. The original configuration of the APA
gives a higher nonspecific background reactivity (ie,
increased false-positives). There were no “strong pos-
itive” (4+) findings with either the original or modi-
fied APA assay configurations when matched with
individual controls. These results were read by 1
microbiologist (R.F.G.) and were objectively verified
by densitometry.
Discussion
The APA assay is an evaluation of the binding of
serum immunoglobulin (IgG) to a complex of partially
polymerized acrylamides (PPA), immobilized on nitro-
cellulose. Silicone can, in a concentration-dependent
manner, partially block APA reactivity in a subset of
APA-positive serum samples. Conversely, PPA, the
antigen used in the APA assay, blocks antisilicone
Results
All 21 patients treated with intraocular silicone oil had
negative responses to the APA assay (Table 1). In our
54
ANN OPHTHALMOL. 2001;33(1)

antibody reactivity in a subset of positive serum sam-
ples. These results indicate that APA are cross-reac-
tive with antisilicone antibodies and suggest why this
laboratory marker correlated with the severity of mus-
culoskeletal and other signs and symptoms in recipi-
ents of SBI.1 After silicone exposure, certain
individuals produce antibodies to silicone, which rec-
ognize a cross-reactive epitope (chemical moiety) pre-
sent on PPA. However, not all patients exposed to
silicone produce the particular subset of antibodies
capable of cross-reacting with PPA. These persons,
who have positive reactions on the APA assay, appear
significantly more likely to develop musculoskeletal
signs and symptoms and other disease complications
after exposure to silicone. Positive results of the APA
assay have been observed in more than 50% of the per-
sons reporting complications after SBI.1 Serum sam-
ples obtained randomly from 100 healthy blood donors
at the Tulane University Hospital blood bank showed
only a 9% positivity. Serum samples from patients
with intraocular silicone oil showed a 0% positivity
rate (Table 2). In patients with rheumatoid arthritis,
APA positivity occurred in less than 10% of the cases
(data not shown), so it appears that APA positivity is
not generally a marker for rheumatic diseases.
T A B L E 2
APA Positivity in Patients With Intraocular Silicone Oil
Exposure and in Silicone Breast Implant Recipients
No. of Positive
Donor Group/
Diagnosis
Results/
No. Tested
Percent
Positive
P
Healthy blood
donors
9/100
363/667
0/21
9.0
54.4
0
*
Silicone
breast
implant
recipients
<.0005†
Patients with
intraocular
silicone oil
exposure
NS
APA indicates antipolymer antibody; NS, not
significant.
*Data previously reported by Tenenbaum et al.1
†Odds ratio=17.0; 95% confidence interval
(CI)=7.55 – 46.7.
Silicone polymers are widely used in medicine and
commerce. It would be important to resolve whether
SBI and other silicone products induce musculoskele-
tal and other signs and symptoms in a subset of
exposed individuals. Lack of an accepted medical def-
inition, frequent use of nonscientific legal terminolo-
gy, and reliance on anecdotal studies have tended to
hamper studies of potential complications following
exposure to SBI and other silicone products. Recipi-
ents of SBI with moderate to advanced signs and
symptoms of systemic disease are significantly more
likely to have a positive result on the APA assay than
are SBI recipients with limited or mild signs and
symptoms (P<.0005).1 This represents yet another fac-
tor that has served to hamper the interpretation of
data in previous reports. As shown in Table 2, the inci-
dence of a positive response to APA in the general
population is 8% to 9%, probably due to environmen-
tal silicone exposure (unpublished observation, R.
Garry PhD, June 1997). Patients receiving SBI, how-
ever, have a reported 54.4% incidence of positivity.
The APA silicone antibody is a highly sensitive and
specific marker for serum antibody production to sili-
cone.1 In the present study, 21 patients were randomly
sampled and showed no statistically significant
increase in the incidence of positive serum antibody
responses, compared with the control group (patients
having no known medical silicone exposure). The
blood-retinal barrier probably plays an important role
in protecting the silicone oil from systemic immuno-
logic exposure. In contrast, patients receiving SBI
may be at higher risk due to the exposure of silicone
oil in a highly vascular nonimmunologic privileged
site when breast tissue is compared with retinal tis-
sue. This would add credence to why the eye, with a
blood-retinal barrier, would not have the increased
rate of APA production compared with SBI recipients.
Even though none of the samples in our small pilot
study was positive, one should not infer that the true
incidence of serum positivity to APA is lower in
patients treated with intraocular silicone oil injection
than that in the general population (ie, less than 8% to
9%). This study suggests that patients do not have an
increased systemic APA production when exposed to
intraocular silicone oil, compared with the general
population. This study will be expanded to include a
larger sample of patient serum samples tested at spe-
cific intervals (before, during, and after intraocular sil-
icone oil injection).
References
1. Tenenbaum SA, Rice JC, Espinoza LR, et al. Use of antipolmer
antibody assay in recipients of silicone breast implants. Lancet.
1997;349:449–454.
2. The Silicone Study Group. Vitrectomy with silicone oil or sulfur
hexafluoride gas in eyes with severe proliferative vitreoretinopa-
thy: results of a randomized clinical trial. Arch Ophthalmol. 1992;
110:770–779.
3. Tenenbaum SA, Cuellar ML, Garry RF, Espinoza LR. Production
of antibodies to partially cross-linked polymers in silicone breast
recipients. Arthritis Rheum. 1993;S118:A123.
4. Javid CG, Maxwell DP Jr, Diamond JE, Garry RF. Poster present-
ed at: Annual Meeting of the Vitreous Society; September 18–21,
1997; New Orleans, La
ANN OPHTHALMOL. 2001;33(1)
55