Fw: Silicone Gel Breast Implant Rupture, Extracapsular Silicone, and Health Status in a Population of Women ~ New Journal of Rheumatology

[Guest guest]

I know this relates to silicone gel for the most part, but the point must be

taken that statistically significant results have been obtained to show

causation between implants and disease...

Patty

----- Original Message -----

From: " ilena rose " <ilena@...>

<Recipient List Suppressed:>

Sent: Monday, April 30, 2001 11:08 AM

Subject: Silicone Gel Breast Implant Rupture, Extracapsular Silicone, and

Health Status in a Population of Women ~ New Journal of Rheumatology

VOLUME 28: NO. 5 MAY 2001

http://jrheum.com/subscribers/01/05/996.html

Silicone Gel Breast Implant Rupture, Extracapsular Silicone, and Health

Status in a Population of Women

S. LORI BROWN, GENE PENNELLO, WENDIE A. BERG, MARY SCOTT SOO, and MICHAEL

S. MIDDLETON

ABSTRACT.

Objective. To assess whether breast implant rupture or extracapsular

silicone are associated with selected symptoms of self-reported

physician-diagnosed connective tissue disease (CTD).

Methods. Women with silicone gel breast implants responded to a

questionnaire that included questions on health status, satisfaction with

implants, symptoms of CTD, and physician-diagnosed disease. These women

then had magnetic resonance imaging (MRI) of their breasts to determine the

status of the implants with respect to rupture and extracapsular silicone.

Results. Women with breast implant rupture diagnosed by MRI were no more

likely to report a diagnosis of selected CTD than those with intact

implants or those with implants of indeterminate status. Women with

extracapsular silicone (silicone gel outside of the fibrous scar that forms

around breast implants) were more likely to report having fibromyalgia (FM,

p = 0.004) or other CTD, which included dermatomyositis, polymyositis,

Hashimoto's thyroiditis, mixed CTD, pulmonary fibrosis, eosinophilic

fasciitis, and polymyalgia (p = 0.008) than other women in the study. The

association with FM remained statistically significant when adjusted for

multiple comparisons (7 diagnoses) and implant age, implant location, or

implant manufacturer (p < 0.05 in all cases), but became of borderline

statistical significance when adjusted for multiple comparisons and

self-perceived health status (p = 0.094) or self-perceived rupture status

(p = 0.051). The association with other CTD remained statistically

significant when adjusted for multiple comparisons and implant location or

implant manufacturer, but became borderline or insignificant when adjusted

for multiple comparisons and for implant age (p = 0.051), self-perceived

health status (p = 0.434), or self-perceived rupture status (p = 0.145).

Logistic regression was used to compute odds ratios of self-reported

diagnoses comparing women with and without extracapsular silicone. The odds

ratios were 2.8 (95% CI 1.2 to 6.3) for FM, and 2.6 (95% CI 0.8 to 8.5) for

other CTD after adjustment for implant age, implant location, implant

manufacturer, implant type, self-perceived health, self-perceived rupture

status, and site of surgery practice.

Conclusion. These data suggest an association between extracapsular

silicone from ruptured silicone breast implants and FM. If this association

persists in other studies, women with silicone gel breast implants should

be informed of the potential risk of developing fibromyalgia if their

breast implants rupture and the silicone gel escapes the fibrous scar

capsule. (J Rheumatol 2001;28:996-1003)

Key Indexing Terms:

BREAST IMPLANTS

BREAST PROSTHESES

SILICONE

FIBROMYALGIA

ADVERSE EVENTS

------------------------------------------------------------------------

Silicone gel breast implants have been available in the United States since

1963. In 1976, with the passage of the Medical Device Amendments, implants

became regulated products. After manufacturers of silicone gel breast

implants failed to provide adequate scientific evidence of the safety and

effectiveness of their products in 1991 and because of emerging health

concerns, the US Food and Drug Administration (FDA) imposed a short

voluntary moratorium on their sale and implantation in 1992. After a second

panel meeting in 1992, the FDA agreed with the General and Plastic Surgery

Devices Advisory Panel that there was a public health need for silicone gel

breast implants to remain on the market for use only by women requiring

implants for reconstruction after breast cancer or for other medical

indications while manufacturers performed safety and efficacy studies1.

Women desiring breast implants for reconstruction were required to receive

silicone gel implants in an adjunct study to ensure that they gave informed

consent prior to implantation. Unanswered questions about the prevalence of

breast implant rupture and the potential link, if any, between implants and

immune-related disorders were cited as important factors in the FDA's

decision2.

Recent studies have ruled out a large increased risk of connective tissue

disease (CTD) overall for women with breast implants3-5, but they have not

ruled out small increases in specific rare CTDs. Neither have these studies

ruled out the possibility that women with breast implants may have a

complex of symptoms or a syndrome that is not typical of diagnosed CTD. A

genetic predisposition for developing symptoms has been investigated and

may be a factor in the development of symptoms in women with breast

implants6. Another factor that has not been studied is the potential role

of exposure to silicone gel from ruptured implants.

In our study, a population of women with silicone breast implants underwent

MR imaging examinations to determine the status of their silicone gel

breast implants. These women were invited to participate in the MR imaging

examination without regard to health status or self-reported symptoms or

diagnoses. Our study reports on the health status of women with ruptured

implants.

MATERIALS AND METHODS

Patients. Participating women were from 2 plastic surgery practices in

Birmingham, Alabama, included in a National Cancer Institute (NCI) study to

examine whether implants were associated with cancer or connective tissue

disease. A cohort of women having their first augmentation mammoplasty

prior to 19887 were identified in the medical records from the implanting

plastic surgeon and information was abstracted from the record on the

surgery and the implant. Women were subsequently located, contacted by

mail, and asked to complete a questionnaire. The protocol for the current

study was reviewed and approved by 5 institutional review boards and a

Certificate of Confidentiality was obtained from the Department of Health

and Human Services. Women who had responded to the questionnaire in the NCI

study and still lived in Alabama at the time of that questionnaire were

eligible for the current cohort.

Of 1247 eligible women, 907 responded to a computer assisted telephone

interview that focused on past surgeries in which implants were removed or

replaced (the results of this study have been submitted elsewhere). Women

were also asked questions regarding their current health status including

whether they had certain symptoms or disease diagnosed by their doctors. Of

these 907 women, 837 reported still having implants and of those, 654

reported having either single or double lumen silicone gel breast implants.

Women were invited to undergo an MRI to determine the status of their

implants after completing the telephone interview providing there were no

contraindications for undergoing an MRI (metal implant or battery activated

stimulator, pregnancy, tattoos, body weight exceeding 300 pounds, or a

history of metal fragments in the eye). The order in which women were

called and invited to participate in the study was random in that the order

in which women were interviewed was random. The study had funding for up to

400 MRI examinations and was also constrained by the contract period the

FDA had with the MRI facility. Women were invited to participate in the

examination until all possible appointments were filled. Initially, only

women within a 50 mile radius of the clinic were invited to participate. As

the study progressed, women outside of the area were called back and

invited, as were subsequent contacts. Of the 445 women invited to

participate in the study, 359 (80.7%) accepted during the allotted time

period. Fourteen with saline breast implants and one woman with no implants

in either breast were excluded. The remaining 344 women with silicone gel

breast implants comprised our study population. A comparison between the

women accepting and undergoing the examination and other women has been

described8. Women gave informed consent prior to receiving an MRI. The

telephone interviews were completed between December 1, 1997 and May 20,

1998 and the MRI were completed between January 6 and May 26, 1998.

Radiological assessment. Three radiologists independently assessed the MR

images for evidence of breast implant rupture. Implants were rated as

having no evidence of rupture (intact), indeterminate (suspicious for

rupture, but not certain), or ruptured. The agreement between radiologists

was almost perfect9 as has been reported8. Radiologists also assessed the

images of the breasts for the presence of extracapsular silicone that had

migrated out of the fibrous scar capsule that forms around the implant. The

agreement between radiologists for extracapsular silicone was moderate to

substantial8,9.

Analyses. Women with ruptured implants were compared to all others (intact

or indeterminate) and women with extracapsular silicone were compared to

all others. Rupture and extracapsular silicone were examined for

association with increases in self-reported symptoms or physician-diagnosed

diseases. The associations were measured with exact p values based on the

Fisher-Freeman-Halton test for 2 by C tables, which generalizes Fisher's

exact test for 2 by 2 tables. Significant associations were reassessed by

stratifying the data on a number of covariates. The possible covariates

were implant satisfaction, participant age, implant age (from time of

mammoplasty), current problems with implants, perceived rupture, implant

manufacturer (Surgitek, -Uphoff International, Dow Corning, Mentor or

Heyer-Schulte, and McGhan or 3M/McGhan), implant type (single or double

lumen), and implant location (subglandular or submuscular). After

stratifying on a covariate, homogeneity of stratum-specific odds ratios

(OR) was tested with Zelen's test. If this hypothesis was not rejected,

then a common OR across strata was assumed and was tested against a value

of one by the uniformly most powerful (UMP) test based on the conditional

distribution given by Gart. The p value of the test was computed by the

horizontal line method of summing all probabilities of data sets with

probability less than or equal to the probability of the data set observed.

An exact maximum likelihood estimate of the common OR was computed with

respect to the conditional distribution. An approximate 95% CI was computed

by inverting the UMP test by the method of and using the mid p value

adjustment to compute upper and lower bounds. All of these calculations

were made in StatExact10.

Logistic regression was used to compute OR simultaneously adjusted for

multiple covariates. In addition to the covariates mentioned above, the

logistic models included site of surgery practice and interactions between

covariates.

P values were adjusted for multiple comparisons by computing an upper bound

on the smallest familywise significance level at which the comparison would

still be significant, where familywise significance level is the

probability of a falsely significant result among any of the comparisons in

the family. The upper bound was based on Sidek's inequality and is 1 - (1 -

p)k, where p is the unadjusted p value and k is the number of comparisons

in the family11.

RESULTS

The mean age for the 344 women at the time of their first mammoplasty was

34.1 ± 7.9 years and their mean age at the time of the MRI was 51.4 ± 8.4

years. Women received their first implant between 1970 and 1988 and the

mean implant age was 16.5 ± 3.4 years. As reported8, the radiologists found

that 236 (68.6%) women had at least one ruptured implant. While a majority

of women had a ruptured implant, 73 (21.2%) had extracapsular silicone gel

that had migrated outside of the fibrous scar capsule that forms around the

implant. Of the 265 women with either a ruptured or indeterminate implant,

72 (27.2%) had extracapsular silicone gel. One woman had extracapsular

silicone but no evidence of implant rupture or self-reported explantation

of a previous implant. Overall 378 (55.0%) of 687 implants had ruptured and

85 (12.4%) had extracapsular silicone gel.

Table 1 shows the distribution of women with ruptured implants or with

extracapsular silicone according to satisfaction with implants,

self-perceived health, participant age, and report of current local

complication with implants, including self-perceived rupture of

implants(s). Women with ruptured implants did not differ from other women

(with intact or indeterminate implants) with respect to satisfaction with

their implant(s), self-perceived health, age, or report of current implant

problems. Women with ruptured implants were also no more likely than other

women to report that they suspected that their implants were ruptured.

Women with extracapsular silicone did not differ from other women with

respect to satisfaction with implants or report of current problems with

implants, but were more likely to report poor health status. Women with

extracapsular silicone depicted by MRI were also more likely to suspect

that their implant(s) were ruptured.

Table 1. Satisfaction with implant, self-perceived health, age group,

current problem with implant, and self-perceived rupture by implant status

among 344 women with silicone gel breast implants (column percents).

Table 2 shows the prevalence of women with ruptured implants or with

extracapsular silicone according to implant type, implant location, implant

age, and implant manufacturer. The prevalence of rupture differed

significantly among manufacturers, implant age groups, and locations of the

implant (submuscular or subglandular). The prevalence of extracapsular

silicone did not differ significantly among the levels of any variable.

Table 2. Implant location, implant type, and implant manufacturer by

implant status.

During the interview, women were asked whether they currently had any of 5

symptoms. Neither rupture nor extracapsular silicone was associated with an

increase in the report of any symptom by women (Table 3) nor were they

associated with the number of symptoms reported (latter results not shown).

Women were also asked whether they currently had each of 7

physician-diagnosed conditions (Table 4). While rupture was not associated

with an increase in any of these diagnoses, extracapsular silicone was

significantly associated with fibromyalgia (FM, p = 0.004) and other

connective tissue disease (p = 0.008), which included dermatomyositis,

polymyositis, Hashimoto's thyroiditis, mixed connective tissue disease,

pulmonary fibrosis, eosinophilic fasciitis, and polymyalgia. Extracapsular

silicone was also borderline significantly associated with Raynaud's

phenomenon (p = 0.086). Adjustments for multiple comparisons had little

impact on the conclusions. Multiplicity adjusted p values based on applying

Sidek's inequality to the family of 7 diagnoses tested for association with

extracapsular silicone were 0.028 for FM, 0.055 for other connective tissue

disease, and 0.467 for Raynaud's (Table 4). A logistic regression was used

to calculate the odds ratio that at least one of a woman's implants had

extracapsular silicone for each of the symptoms and diagnoses in Tables 3

and 4. The OR was 2.7 (95% CI = 1.4 to 5.2) for FM, indicating that the

odds of self-reported FM were 2.7 times greater for women with

extracapsular silicone than women without extracapsular silicone. The OR

was 2.9 (95% CI = 1.0 to 8.7) for Raynaud's, and 3.7 (95% CI = 1.4 to 9.4)

for other CTD. Statistically significant or borderline associations of

extracapsular silicone with diagnosis were reassessed by stratifying by the

participant's age group at time of MRI, self-perceived implant status, and

self-perceived health, (Table 5) and by implant age group, implant

location, and implant manufacturer (Table 6). Stratification variables were

considered one variable at a time. For each diagnosis, stratum-specific OR

were not significantly different for each variable, according to Zelen's

exact test, the smallest p value being 0.135 for FM stratified on implant

age. We therefore assumed a common OR across strata and tested the

hypothesis that the common OR is one. Under this assumption, associations

with extracapsular silicone that were borderline before stratification

(Raynaud's) remained borderline or became statistically significant.

Associations with extracapsular silicone that were statistically

significant (FM and other CTD) continued to be statistically significant

except in the case of other CTD stratified on self-perceived health, which

was borderline (p = 0.078). P values were adjusted for multiple comparisons

within the family of 7 diagnoses considered by applying Sidek's inequality.

Borderline associations (Raynaud's) became insignificant and statistically

significant associations (FM and other CTD) remained significant except for

other CTD stratified on implant age, which just missed being significant (p

= 0.051).

Table 3. Persistent symptoms reported by implant status.

Table 4. Self-reported physician diagnosis by implant status.

Table 5. Raynaud's phenomenon, FM, and other CTD by extracapsular silicone

stratified on participant age, self-perceived health and self-perceived

implant rupture.

Table 6. Raynaud's phenomenon, FM, and other CTD by extracapsular silicone

stratified on implant location, implant age, or implant manufacturer.

Logistic regression was used to compute OR of diagnoses, comparing women

with and without extracapsular silicone, that were adjusted simultaneously

for multiple covariates. The logistic models included the variables listed

in Tables 5 and 6, except for patient age, which was not a significant

predictor of extracapsular silicone. The models also included implant type,

site of surgery practice, implant manufacturer by implant age interaction,

and implant manufacturer by site interaction. These extra 4 variables were

significant predictors of women with ruptured implants or ruptured or

indeterminate implants as explained elsewhere8. Implant age was modeled as

a continuous variable with a linear effect on the log scale rather than as

a categorical variable as in Tables 2 and 6. The OR were 4.2 (95% CI

1.1-16.0, p = 0.037) for Raynaud's, 2.8 (95% CI 1.2-6.3, p = 0.013) for FM,

and 2.7 (95% CI = 0.8-8.5, p = 0.102) for other CTD.

DISCUSSION

Extracapsular silicone was associated with an increase in self-reported

physician-diagnosed fibromyalgia and other connective tissue disease in

women with silicone gel breast implants. These associations remained

statistically significant after separately controlling for the woman's age,

implant age, location, and manufacturer. After an adjustment for multiple

comparisons, the association with FM remained significant. Breast implant

rupture alone was not associated with self-reported physician-diagnosed FM

or other CTD. The natural history of breast implant rupture is not

completely understood but it is believed in some cases that rupture with

intracapsular containment of silicone gel may progress to extracapsular

silicone gel.

In a report on 29 patients with ruptured silicone gel implants diagnosed

during open capsulotomy, 17.2% tested positive for antinuclear antibody

(ANA), which was not significantly different than control subjects12.

Whether the implant rupture was extracapsular was not stated in this study.

A positive test for ANA may be seen in undifferentiated connective tissue

disease but it is not a hallmark of FM. Another study reported the presence

of positive ANA in women with breast implants referred by rheumatologists

to be 18/24 (75%) of patients, but there was no comparison or control

group13. The authors of this study also reported that a history of trauma

to the breast, which was presumed to cause implant rupture, accelerated the

onset of CTD. However, no evidence of the rupture status, either by imaging

or explantation, was provided. Epidemiologic studies have ruled out a large

increase in CTD associated with breast implants3-5,14,15 but the status of

implants with respect to rupture or extracapsular silicone in these studies

was unknown.

American College of Rheumatology criteria for FM include a history of

widespread pain and tenderness at specified tender points16 and it has been

estimated that FM occurs with a prevalence of 2% in the US and 3.4% for

women17. Symptoms reported by women with silicone gel breast implants are

common in patients diagnosed with FM, suggesting that the atypical syndrome

in women with breast implants reported by rheumatologists18-21 may be FM22.

In one study, the most common diagnosis for women with breast implants

referred for evaluation was FM (27/70, 38.6%) but the status of their

implants with respect to rupture or extracapsular silicone was not known23.

In another study of patients with breast implants referred to a

rheumatology practice, 124/300 (41.3%) met the criteria for a diagnosis of

FM24. In a study of women having their implants removed, 10/100 (10%)

patients had FM25. While it was reported that 57% of the implants in that

study had ruptured, any association between ruptured implants or

extracapsular silicone and FM was not reported. s, et al did report

that although there was an initial improvement after removal of implants in

women with FM, symptoms of FM recurred within the next 6 to 12 months25.

Not all clinicians regard the diagnosis of FM as definitive, and some

consider FM and other syndromes whose diagnoses rely on subjective symptoms

as functional somatic syndromes, implying that the pathogenesis is

psychosomatic or related to litigation26. Others describe the

pathophysiology of this syndrome in terms of pain amplification due to

biochemical imbalances in the nervous or immune system27. Women in our

study were more likely to think that their implants were ruptured when

extracapsular silicone was present (Table 1). However, after adjusting for

self-perceived rupture status, the association between self-reported

physician-diagnosed FM and extracapsular silicone remained significant.

Women in our study were also asked whether a physician had diagnosed them

with other CTD such as dermatomyositis, polymyositis, Hashimoto's

thyroiditis, mixed connective tissue disease, pulmonary fibrosis,

eosinophilic fasciitis, or polymyalgia. Women with extracapsular silicone

gel were more likely to report that they had other CTD than were other

women in the study. It is not possible to determine whether any of these

disorders predominated. Our category of other CTD is artificial and

therefore those results are difficult to interpret.

Raynaud's phenomenon may occur in isolation or be present in association

with CTD such as scleroderma. We found the association between

extracapsular silicone and Raynaud's syndrome had borderline statistical

significance while there was no significant association observed for

scleroderma and extracapsular silicone. The 3 cases of scleroderma reported

to us were by women whose implants had ruptured but who did not have

extracapsular silicone.

The principle limitation of our study was its small size It would not be

possible to rule out rare diseases (such as scleroderma) in association

with ruptured implants, ruptured or indeterminate implants, or

extracapsular silicone. Our study is cross-sectional in that we only have

information on current implant status and current self-reported symptoms

and diseases. We were unable to determine whether symptoms or disease

occurred before or after the development of extracapsular silicone or

whether it was before or after the implant was surgically implanted. Also,

the results from this study were based on self-report of

physician-diagnosed disease and therefore have the weaknesses attributed to

self-reported disease. However, since the women did not know the status of

their implants with respect to rupture, any error in disease

misclassification would be expected to dilute the association of disease

with rupture. We cannot know what effect the media or litigation had on

participation in this study: anecdotal evidence suggests that some

litigants were encouraged to participate in the NCI study and others were

discouraged by their attorney(s) or others. Despite these limitations, the

results of this study are valuable because it is the first study in which

the implant status has been evaluated for all women. The women for the MRI

study were invited to participate from a larger population of women with

breast implants from Alabama without regard to their health status or

knowledge of their implant status. While not definitive, our study does

raise the question of whether implant status, especially implant rupture

with extracapsular spread of silicone gel, may have a bearing on the health

status of implanted women.

It should be noted that we did not compare women with implants to those

without. It is possible that there may be an increased prevalence of

disease or symptoms in women with breast implant rupture compared to women

without implants but we could not make this comparison.

Our results indicate that women with extracapsular silicone gel may be at

an increased risk for FM. If future studies confirm our findings,

consideration as to whether women with silicone gel breast implants should

be screened for implant rupture should be considered. Clearly, if the risk

of rupture and subsequent extracapsular silicone increases the risk of

fibromyalgia, women considering augmentation or reconstruction mammoplasty

with silicone gel breast implants should be informed.

ACKNOWLEDGMENT.

We acknowledge the contributions of the women who participated in this

study and appreciate the contributions of Drs. Louise Brinton and Jay Lubin

from the National Cancer Institute who graciously provided access to the

NCI cohort. Cay Burich, Vargish, Marilyn Sawyer, and Jon

Schmalz from Abt Associates, Inc. in Chicago are to be commended for their

work on this study. We thank Audrey Sheppard and Dr. Margaret of the

FDA Office of Women's Health for their continued support for this work and

thank the many reviewers of this study including Drs. al and

Larry Kessler.

From the Office of Surveillance and Biometrics, Center for Devices and

Radiological Health, Food and Drug Administration Rockville; Division of

Breast Imaging, Department of Radiology and Greenebaum Cancer Center,

University of land School of Medicine, Baltimore, MD; Department of

Radiology, Duke South Hospital, Duke University Medical Center, Durham, NC;

Department of Radiology, University of California at San Diego School of

Medicine, San Diego, CA, USA.

Supported in part by the Food and Drug Administration, National Institutes

of Health, and the US Department of Health and Human Services.

S.L. Brown, PhD, MPH, Senior Research Scientist Officer; G. Pennello, PhD,

Mathematical Statistician, Office of Surveillance and Biometrics, Center

for Devices and Radiological Health, Food and Drug Administration; W.A.

Berg, MD, PhD, Director of Breast Imaging, Division of Breast Imaging,

Department of Radiology and Greenebaum Cancer Center, University of

land School of Medicine; M. Soo, MD, Assistant Professor of

Radiology, Department of Radiology, Duke South Hospital, Duke University

Medical Center; M.S. Middleton, MD, PhD, Radiologist, Department of

Radiology, University of California at San Diego School of Medicine.

The opinions or assertions presented herein are the private views of the

authors and are not to be construed as conveying either an official

endorsement or criticism by the US Department of Health and Human Services,

the Public Health Service, or the Food and Drug Administration.

Address reprint requests to Dr. S.L. Brown, Office of Surveillance and

Biometrics, Center for Devices and Radiological Health, Food and Drug

Administration, 1350 Piccard Drive, HFZ-541, Rockville, MD 20850, USA.

Submitted March 15, 2000 revision accepted November 10, 2000.

REFERENCES

1. Controversy over the silicone gel breast implant: Current status

and clinical implications. Texas Med 1993;89:52-8.

2. Kessler DA. The basis of the FDA's decision on breast implants. N

Engl J Med 1992;326:1713-5.

3. SE, O'Fallon WM, Kurland LT, Beard CM, Woods JE, Melton LJ.

Risk of connective-tissue diseases and other disorders after breast

implantation. N Eng J Med 1994;330:1697-702.

4. Hennekens CH, Lee I-M, Cook NR, et al. Self-reported breast

implants and connective-tissue diseases in female health professionals: a

retrospective cohort study. JAMA 1996; 275:616-21.

5. -Guerrero J, Colditz GA, Karlson EW, Hunter DJ, Speizer FE,

Liang MH. Silicone breast implants and the risk of connective-tissue

diseases and symptoms. N Eng J Med 1995;332:1666-70.

6. Young VL, Nemecek JR, Schwartz BD, Phelan DL, Schorr MW. HLA typing

in women with breast implants. Plast Reconstr Surg 1995;96:1497-519.

7. Brinton LA, Toniolo P, Pasternack BS. Epidemiologic follow-up

studies of breast augmentation patients. J Clin Epidemiol 1995;48:557-63.

8. Brown SL, Middleton MS, Berg WA, Soo MS, Pennello G. Silicone gel

breast implant rupture prevalence in a population of women in Birmingham,

Alabama. AJR 2000;175:1057-64.

9. Shoukri MM, Edge VL. Statistical methods for health sciences. Boca

Raton, Florida: CRC Press; 1996.

10. Mehta C, Patel N. StatXact-Turbo User Manual. Cambridge, MA: CYTEL

Software Corporation; 1989.

11. Westfall PH, Young SS. Resampling-based multiple testing: Examples

and methods for p value adjustment. New York; Chichester: Wiley &

Sons; 1993.

12. s W, Keystone E, Snow K, Rubin L, D. Is there a

relationship between autoantibodies and silicone-gel implants? Ann Plast

Surg 1994;32:1-7.

13. Press RI, Peebles CL, Kumagai Y, Ochs RL, Tan EM. Antinuclear

autoantibodies in women with silicone breast implants. Lancet

1992;340:1304-7.

14. Silverman BG, Brown SL, Bright RA, Kaczmarek RG, Arrowsmith-Lowe

JB, Kessler DA. Reported complications of silicone gel breast implants: an

epidemiologic review. Ann Intern Med 1996;124:744-56.

15. Brown SL, Langone JJ, Brinton LA. Silicone breast implants and

autoimmune disease. J Am Med Wom Assoc 1998;53:21-4.

16. Wolfe F, Smythe HA, Yunus MB, et al. The American College of

Rheumatology 1990 criteria for the classification of fibromyalgia: report

of the multicenter criteria committee. Arthritis Rheum 1990;33:160-72.

17. Goldenberg DL. Fibromyalgia, chronic fatigue syndrome, and

myofascial pain. Curr Opin Rheumatol 1996;8:113-23.

18. Freundlich B, Altman C, Sandorfi N, Greenberg M, Tomaszewski J. A

profile of symptomatic patients with silicone breast implants: a

Sjögrens-like syndrome. Semin Arthritis Rheum 1994; 24 Suppl 1:44-53.

19. Borenstein D. Siliconosis. A spectrum of illness. Semin Arthritis

Rheum 1994;24 Suppl 1:1-7.

20. G. A clinical and laboratory profile of symptomatic women with

silicone breast implants. Semin Arthritis Rheum 1994; 24 Suppl 1:29-37.

21. Vasey FB, Havice DL, Bocanegra TS, et al. Clinical findings in

symptomatic women with silicone breast implants. Semin Arthritis Rheum

1994;24 Suppl 1:22-8.

22. Wolfe F. Silicone related symptoms are common in patients with

fibromyalgia: no evidence for a new disease. J Rheumatol 1999;26:1172-5.

23. Blackburn Jr WD, Grotting JC, Everson MP. Lack of evidence of

systemic inflammatory rheumatic disorders in symptomatic women with breast

implants. Plast Reconstr Surg 1997;99:1054-60.

24. Cuellar ML, Gluck O, Molina JF, Gutierrez S, C, Espinoza R.

Silicone breast implant-associated musculoskeletal manifestations. Clin

Rhematol 1995 14:667-72.

25. s W, D, Fornasier V, Lugowski S, Ibanez D. An outcome

analysis of 100 women after explantation of silicone gel breast implants.

Ann Plast Surg 1997;39:9-19.

26. Barsky AJ, Borus JF. Functional somatic syndromes. Ann Intern Med

1999;130:910-21.

27. Wallace DJ. The fibromyalgia syndrome. Ann Med 1997;29:9-21.