Total and cause specific mortality among Swedish women with cosmetic breast implants: prospective studyBMJ 2003; 326 doi: http://dx.doi.org/10.1136/bmj.326.7388.527 (Published 08 March 2003) Cite this as: BMJ 2003;326:527
- V C M Koot, clinician ()a,
- P H M Peeters, associate professora,
- F Granath, associate professorb,
- D E Grobbee, professora,
- O Nyren, professorb
- a Julius Centre for Health Sciences and Primary Care, University Medical Centre, Box 85500, 3508 GA Utrecht, Netherlands
- b Department of Medical Epidemiology, Karolinska Institute, Box 281, S-171 77 Stockholm, Sweden
- Correspondence to: V C M Kuck-Koot, Comprehensive Cancer Centre Middle Netherlands, PO Box 19079, 3501 DB Utrecht, Netherlands
- Accepted 17 January 2003
The potential health hazards of breast implants have been heavily debated for the past decade, yet only one study has reported on long term mortality among women with such implants, and around one fifth of the participants were lost to follow up. 1 2 We assessed total and cause specific mortality among Swedish women who underwent augmentation mammoplasty between 1965 and 1993. As a desire for cosmetic surgery represents underlying psychopathology in some patients, we hypothesised that deaths due to suicide may be over-represented.3
Subjects and methods
Details about accrual of the cohort have been given elsewhere.4 We obtained records from the Swedish Inpatient Register of all 15–69 year old women who had had breast implants (n=7585) in 1965-93. We identified records with erroneous registration numbers or where emigration or death occurred before surgery through linkages with registers held by Statistics Sweden, using the unique national registration numbers. We excluded such records and records where surgery occurred at hospitals without surgical services (n=138). We also excluded women who had received an implant after surgery for breast cancer (n=3926), identified through the cancer register. The final study cohort comprised 3521 women, with a mean age of 31.6 (SD 8.6) years.
Follow up started on the day of first implantation surgery and stopped at date of emigration, death, or end of follow up (31 December 1994), whichever occurred first. The cohort members were followed for an average of 11.3 (range 0.3-29.9) years, corresponding to 39 735 person years at risk. We compared the observed number of deaths with the expected number of deaths, the ratio of these two numbers giving the standardised mortality ratio. We obtained the expected number of deaths by multiplying the observed number of person years at risk in the cohort, divided into 5 year age strata and 1 calendar year strata, by the stratum specific mortality rates, derived from official Swedish death statistics. The standardised mortality ratio can therefore be viewed as a measure of relative risk, with the Swedish female population matched for age and calendar year serving as reference. We calculated 95% confidence intervals, assuming that the number of observed events followed a Poisson distribution. We coded underlying causes of death according to the international classification of diseases (7th, 8th, and 9th revisions) into suicide, unintentional injury, cardiovascular diseases, malignancies, and other causes.
Although 58.7 deaths were expected, 85 women died (standardised mortality ratio 1.5, 1.2 to 1.8; table). Fifteen women committed suicide, compared with 5.2 expected deaths (2.9, 1.6 to 4.8). Excess deaths were also due to malignant disease (1.4, 1.0 to 1.9), mainly lung cancer. The number of deaths for all other causes was close to expected.
Women who undergo cosmetic surgery for breast augmentation are more likely to commit suicide than women from the general population. The 50% excess mortality found by us in our prospective study of 3000 Swedish women contrasts with the decreased mortality reported from the United States.2 This may reflect different reasons for self selection for plastic surgery or may be an effect of losses to follow up in the American study. Both the American study and our study did, however, show an increased risk for suicide in women opting for breast augmentation. Our excess mortality was explained by the excess of suicides and deaths from malignant disease. Deaths due to malignancy were mainly linked to smoking, previously shown as common in our cohort.5 Given the well documented link between psychiatric disorders and a desire for cosmetic surgery, the increased risk for death from suicide may reflect a greater prevalence of psychopathology rather than a causal association between implant surgery and suicide.3 Surgeons evaluating candidates for breast implant surgery need to be vigilant for subtle signs of psychiatric problems.
Contributors: VCMK, the principal investigator, discussed the core ideas, performed the record linkages, outlined and performed analyses, and wrote most of the paper. PHMP discussed the core ideas and participated in data interpretation and writing of the paper. FG participated in discussions about the core ideas, made suggestions about analyses, and helped VCMK with the practical analysis. DEG discussed the core ideas, the design of the study, the interpretation of the data, and writing of the paper. ON initiated the research, discussed the core ideas, formulated the primary study hypothesis, made suggestions about analyses and interpretation of the data, and supervised the writing. DEG and ON will act as guarantors for the paper.
Funding This work was supported by a contract from the International Epidemiology Institute, Rockville, MA. VCMK was financially supported by the Netherlands Organization for Scientific Research.
Competing interests None declared.