In June of last year, I discussed the results of a large epidemiological study in women that showed that women with larger breasts have an increased risk of developing type-2 diabetes.
As soon as Travis and I read this study, we knew we had to do a follow-up study of our own to see if this finding was simply spurious or if there was actually something to large breasts that indicated health risk - beyond that explained by obesity per se.
The project that Travis and I began over a year ago has culminated in both a hot-off-the-press publication in the journal Obesity, as well as my presentation at this year's Obesity Society meeting in Washington D.C.
In the study, we used body composition data acquired through MRI on about 100 premenopausal women to directly quantify breast size. By using MRI data we significantly improved the methodology used by the authors of the original study on breast size and diabetes risk, who relied on over 20 year recall of cup size as their key measure.
First, we sought to examine if breast tissue volume was associated with any cardiometabolic risk factors, such as glucose tolerance (a known antecedent to type 2 diabetes) and various blood lipids. Since the original authors found an association between cup size and diabetes risk, we expected to find an association between breast volume and cardiometabolic risk factors.
What did we actually find?
Breast volume was not associated with any of the cardiometabolic risk
factors measured in any of the statistical models used. However, in
these analyses, visceral or intra-abdominal fat was a strong predictor
of numerous risk factors - a finding reported in numerous studies.
Next we decided to examine the associations between breast volume and
other body fat depots. Here is where the story got interesting. First
off, as one would predict, women with larger breasts had more
subcutaneous (under the skin) fat in their thighs, abdomen, as well as
more visceral and inter-muscular (or ectopic) fat. That is, bigger
breasted women tended to carry more fat everywhere.
However, once we controlled for their level of obesity (body mass
index and waist circumference) bigger breasted women were no more likely
to have more subcutaneous fat in either lower body or abdomen, but were
much more likely to have excess fat deposition in the dangerous
visceral and inter-muscular depots.
Specifically, our study found that given the same age, body mass
index (total obesity), waist circumference (abdominal obesity), women
with the highest breast volume had approximately 1.1 and 1.3 kg more
visceral fat and intermuscular fat in comparison to women with the
smallest breast volume.
In other words, large breasts appear to indicate a phenotype
characterized by the augmented deposition of fat in ectopic depots, such
as visceral and inter-muscular fat - each of which is independently
associated with increased cardiometabolic risk. The extrapolation of our
findings suggests that excess visceral or intermuscular fat may be the
conduit which explains the previously documented association between
breast size and type-2 diabetes.
While our findings are intriguing (someone at the Obesity Society
conference suggested I should get an award from the most "unique"
study), there exist a couple clinical examples which corroborate our
For example, approximately 40-50% of women undergoing liposuction of
subcutaneous adipose tissue from the hips, thighs, or abdomen present
with a paradoxical enlargement of breast size of at least one cup as
well as a relative increase in visceral fat post-surgery. Additionally,
highly active antiretroviral therapy among HIV-positive women is
associated with a peripheral loss of functional subcutaneous fat but a
compensatory increase in visceral and intermuscular fat (well documented
lipodystrophy), in association with a significant enlargement in breast
Now before women with large breasts head out to get breast
reductions, it is key to consider that our findings are quite
preliminary, and furthermore that breast size only appears to be a proxy
for other factors which are more likely to be causally related to
health risk. Thus, further research in this area is clearly warranted
before we begin screening disease risk by breast size. Nevertheless,
when explaining our study findings at the conference, I heard many sighs
of relief from women who perceived themselves to have smaller breasts.
P., Saunders, T., & Ross, R. (2009). Breast Volume is an
Independent Predictor of Visceral and Ectopic Fat in Premenopausal Women
Obesity DOI: 10.1038/oby.2009.336
This article was originally posted on Obesitypanacea.com on October 30, 2009, just after the article was published online ahead of print. It was published in paper form earlier this week.
Question: What are we to do with this information? If women do have large breasts, how should we act differently if further research substantiates this theory?
Excellent question. You could argue that this could be of clinical relevance - this suggests that for a given waist circumference, large breasts could be an indicator of increased visceral fat, and therefore increased metabolic risk. So it could potentially be a quick and dirty way of estimating metabolic risk. The waist-to-hip ratio is sometimes used in a similar way. But breast size is difficult to measure (not many people have access to MRI machines, and if you have access to one, you might as well directly measure visceral fat itself), so in the end I think these results are more useful in simply trying to understand why certain patterns of fat distribution are associated with a greater risk than other patterns.
Did you factor breast density into this too, or just volume?
Just volume - actually, we measured surface area on several of cross-sectional images, then used that to estimate volume. There could have been some compression since women were laying on their front, but unfortunately you can't account for that with MRI, and you'd hope that it would be fairly consistent from one woman to the next.
I started developing breasts when I was 8 years old, and at the age of 40, wear an H cup. My grandmother was similar in her breast size also. So it appears to be an inherited trait. If so, then how can it be a proxy for health risk? Or is it that people who inherit this trait are more likely to have visceral fat deposits? If so, I am sorry to hear this!
You hit the nail on the head - women who are predisposed to having larger breasts (which I would assume would have a lot to do with their genes) are more like to have higher levels of visceral fat, for a given waist circumference. Don't feel bad - my inherited trait (being male) probably accounts for far more risk than breast size ever will!
So, if after liposuction one grows larger breasts and visceral fat deposits, does it work vice versa - if I get breast reduction, will I grow even bigger pot belly?
Hmm... that's a great question, and off the top of my head I'm not really sure. It seems like fat accumulates in the breasts and visceral cavity once other "healthier" depots (e.g. the legs) are full. So my first guess (and it's a complete guess) is that it wouldn't do much. But that's a total guess.
Now before women with large breasts head out to get breast reduction
Presumably, on your interpretation, breast reduction should leave one (well, two) less places for external fat deposition, leading to even more visceral and intermuscular fat when the patient regains the lost weight (which she presumably will, absent changed dietary habits, breast tissue not being metabolically inert)?
Now, what we all really want to know is: do manboobs carry the same implication?
Does exercise help to reduce the amount of visceral fat and inter-muscle fat?
I've had large breasts since I was 13. I am now over 50 and have been moderately active for most of my life.
Great question - exercise typically reduces visceral fat even in the absence of weight loss. Meaning that all else being equal, the person who exercises more is likely to have less visceral fat.
Did you compare activity level in the two groups? I wouldn't be surprised if large breasted women tend to have a lower activity level. Breast movement during exercise can hurt, and it is harder for women to find adequately supportive bras if they have large breasts. Breast size could also contribute to concerns about harassment during exercise. If you didn't control for it then a difference in activity level, rather than a phenotype that combines large breasts and visceral fat, could explain your results.
Good question. All women in this particular sample were inactive (the scans are actually from the baseline data of a physical activity intervention). The issues you mention can definitely influence physical activity levels, but I don't think it could entirely explain the results of the present study. Since this is a pretty homogeneous group of women (all overweight and inactive), it would be nice to see if the relationship is similar in a more representative sample, controlling for other factors like physical activity.