Breast implants may make cancer harder to spot

Wednesday May 1 2013

“Breast implants may harm breast cancer survival chances,” The Guardian warns, along with other media sources reporting on the same subject.

It is important to stress that the research the media has reported does not suggest that breast implants cause breast cancer.

Instead, the research suggests that breast implants may cause a delay in diagnosis in women who have breast cancer, which may increase their risk of dying from the condition.

The researchers suggest that the implants could hide cancerous tissue that would otherwise be detected during screening.

To test this, the researchers reviewed several small studies looking at whether having cosmetic breast implants was associated with (no causal association) a delay in diagnosis, and whether women with breast cancer who had implants were at increased risk of dying from the disease.

They found some evidence of an association. Women with breast implants had a 26% increased risk of being diagnosed at a later stage of breast cancer than those without implants. Women with implants also had a 38% greater risk of dying from breast cancer than women without implants.

However, as the authors rightly point out, the results of these analyses should be viewed with caution. This is because they couldn’t find previous research of high enough quality to draw firm conclusions, and better quality studies would be needed to confirm the association.

Where did the story come from?

The study was carried out by researchers from the Université Laval, the Public Health Agency of Canada, the University of Toronto, Health Canada and the University of Ottawa. It was funded by grants from a number of public institutions in Canada.

The study was published in the peer-reviewed British Medical Journal.

Coverage of this study in the UK media was very good. All of the papers who covered the stories carried comments about its limitations from both the authors and independent experts. It was refreshing to see that there was no attempt to ‘spin’ this research into a scare story.

What kind of research was this?

This was a systematic review of observational studies that examined whether the stage at which breast cancer is diagnosed is different between women who have previously had cosmetic implants and those who have not.

It also looked at whether mortality from breast cancer differed between the two groups. The review included two meta-analyses.

The authors point out that cosmetic implants to increase breast size have become increasingly popular, although there is controversy about the long-term health effects. Concerns have been raised that implants may impair the ability to diagnose breast cancer at an early stage by mammography. This is because cosmetic breast implants are ‘radio-opaque’, which means they may obscure breast tissue on an X-ray.

Previous studies of whether implants impair the ability to detect breast cancer early (when survival is more favourable) have had conflicting results.

What did the research involve?

The researchers carried out a systematic search for all eligible studies, using a number of established electronic databases. They also contacted international experts to request any relevant published or unpublished papers. All studies included had to be original, peer reviewed and published. The search was limited to articles in French and English.

Eligible publications were those that included women diagnosed as having breast cancer, who had previously had cosmetic breast implants and that included a comparison group of women who had breast cancer but who had not had implants. These latter groups were drawn from the general female population, or from women who had had other common cosmetic surgeries. To be included, the studies had to provide information on the stage at which the women from each group were diagnosed with breast cancer, or on whether their cancer had spread beyond the breast.

Publications eligible for inclusion in the analysis of breast implants and survival following diagnosis of breast cancer needed to provide figures comparing the mortality rate due to breast cancer after diagnosis, between women who had implants and those who did not.

The researchers evaluated all the studies to see if they met their criteria, and those that did underwent a more detailed assessment. Only invasive breast cancers (cases where the cancer has the ability to spread out of the breast and into other parts of the body) were included in this analysis. Due to the different classification systems for diagnosis used in different countries, the researchers divided cases into two broad groups for their first analysis: 

  • localised cancer (less advanced, early stage cancers)
  • non-localised cancer (more advanced or later stage cancers)

What were the basic results?

The first meta-analysis included 12 observational studies, mainly from the US, Canada and Northern Europe. It found that among women diagnosed with breast cancer, those who had cosmetic breast implants had a 26% higher chance of being diagnosed at a non-localised (later) stage of the disease than those without implants (odds ratio (OR) 1.26, 95% confidence interval (CI) 0.99 to 1.60).

The second meta-analysis was based on five observational studies. It found that among women diagnosed with breast cancer, those with implants had a 38% greater risk of dying from the disease than those without (hazard ratio (HR) 1.38, 95% CI 1.08–1.75).

How did the researchers interpret the results?

The researchers say that their analyses suggest that women with cosmetic breast implants have later stage tumours at diagnosis of breast cancer than those without. This may mean that cosmetic breast augmentation adversely affects the survival of women who are subsequently diagnosed with breast cancer.

They suggest that this might be explained by the implant material (silicone or saline) obscuring breast tissue examined by mammography, as well as other factors related to implants. They also say that although techniques to solve this problem are widely used in mammography, breast tissue may still not be sufficiently well examined.

However, they add that their results should be interpreted with caution. Further research is warranted into the long-term effects of cosmetic breast implants on the detection and survival rates of breast cancer, they conclude.

Conclusion

As the authors point out, their review had several limitations. This was not the fault of the current researchers, as the quality of some of the evidence they had to rely on was impaired by poor study methodology. For example, certain studies included cases of ‘in situ’ (non-invasive) breast cancer, which could have resulted in some misclassification.

Further, the second meta-analysis, which showed a higher risk of breast cancer mortality among women with breast cancer who had implants, included only five studies, three of which did not adjust their results for the women’s age at diagnosis. This is an important potential confounding factor that means the results should be viewed with some caution.

All five studies were unadjusted for the women’s BMI (another potential confounder), while one study included mortality from all causes rather than just from breast cancer (which means that some of the women may have died from other causes).

Further research is needed into the important issue of breast cancer detection and diagnosis in women with cosmetic implants.

Analysis by Bazian
Edited by NHS Choices