Gestational Diabetes and Pre-Pregnancy Fat Intake

Gestational diabetes -- hyperglycemia during pregnancy -- effects a considerable number of women.  It is also a known risk factor for developing diabetes (T2) later in life.    Some degree of peripheral insulin resistance normally develops during pregnancy.  Why?  Because the fetus' growth and development takes priority in nutrient partitioning so glucose is conserved for the fetus much like it is conserved for the brain in the fasted/starved/glucose-deprived state.  The link of GD to T2 most likely reveals the genetic predisposition towards IR -- in other words, while diet can cause insulin resistance, there is a genetic component in most who eventually develop hyperglycemia as a result.  Most women will compensate for the mild IR state and maintain normal glycemia during pregnancy, while those who are perhaps even mildly IR to begin with will exceed their capacity to compensate during the pregnancy.  In most cases, deliver the baby and the diabetes goes away.  



This study used data from the Nurses Health Study II, which began tracking female nurses in the US in 1989 (age range in '89 22-44 y.o.a.).  This particular study included women who reported a single pregnancy that lasted at least 6 months (between 1991 and 2001) numbering 13,475.  There were 860 cases of GD (6.4%) reported.   They looked at dietary intake data collected (from periodic self reported intake*) and analyzed it for the total fat content as well as content of classes of fats (SFA, MUFA, PUFA) etc.
In multivariate models, total fat, fat subtypes, and the source of fats (animal compared with vegetable fat) were expressed as the nutrient density (percentage of calories from fat) and modeled as quintiles of intake. Quintiles were defined by the distribution of each nutrient at baseline.
Quintiles divide the sample into 5 groups of equal numbers of subjects.  The lowest quintile was used as the reference for determining risk ratios (RR).  Here's the baseline data -- you can click to enlarge


Risk Ratios for Total Fat, Animal Fat and Veggie Fat

Risk Ratios for Classes of Fats


Note:
RR1: Adjusted for Age (5yr increment) and BMIRR2: Additionally adjusted for parity, physical activity, energy intake, smoking & alcohol intake and diabetes historyRR3: Additionally adjusted for cereal fibers, glycemic load and other fats


Now, let's keep in mind the limitations of these types of studies. Also, as LCHF advocates are always quick to point out, this fat intake is in the context of the SAD.  Still, my reason for sharing this is that it is somewhat surprising given what we keep hearing about veggie oils and the glories of sat fats and fat consumption.  I doubt these can be predictive at all for the LCHF diet, but some of the numbers really stand out to me.  I'll summarize those below -- focusing on the number of cases per quintile and comparing lowest (Quint1) to highest quintile (Quint5)

  • SFA:  133 cases in Quint1 vs. 196 cases in Quint5
  • MUFA:  145 cases in Quint1 vs. 173 cases in Quint5
  • PUFA:   173 cases in Quint1 vs. 140 cases in Quint5
  • PUFA/SFA ratio:  216 cases in Quint1 vs. 122 cases in Quint5
That GD incidence increased significantly with SFA intake, and decreased significantly with PUFA intake is the sort of data that flies in the face of this notion that veggie oils are to blame for the increased incidence of diabetes in general, and tends to implicate SFA's.   That ratio of PUFA/SFA is rather striking and stood out to me.  
  • Animal Fat:  125 cases in Quint1 vs. 230 cases in Quint5
  • Veggie Fat:  200 cases in Quint1 vs. 144 cases in Quint5
I would also make note of the fact that the glycemic loads were actually lower in the higher fat quintiles.  The carb intake in the animal fat quintiles went from 58% of energy to 43% of energy from Quint1 to Quint5, and from 53% to 48% in the veggie fat group.  

I'm not going to be microanalyzing this data, certainly there are some things one can pull out to explain the data that counters any dearly held beliefs.  If anyone wants the full text of this for your own reference, shoot me an email (carbsane at gmail dot com).  I just came across this looking for something else, as often happens, but the results were surprising enough that I thought I'd share them with you.  

Comments

SamAbroad said…
Interesting study. Wouldn't be the first to discover that increased PUFA being associated with better health either.

But I think this study suffers from the same problems as those other studies. Namely the fact that epidemiology in industrial nations suffers from the 'self-fulfilling prophesy' effect. We tell people something is good for them and then the people who give a crap about their health implement that recommendation along with all the other tiny unmeasurable things that health-conscious people do to protect their health. That's why it's surprising to me that more cohort studies don't indict saturated fat in general.

Did the study control for energy intake in general?
CarbSane said…
Did the study control for energy intake in general? I don't think they needed to since intake varied from 1786 to 1850 kcal/day at most.

To take a devil's advocate approach to your well-stated problems with such studies: If the PUFA content of health-conscious people is increased, and there are other unmeasurable things associated with actually listening to the dreaded conventional wisdom on diet and health, perhaps the recommendations aren't sooooo far off base after all?!
bentleyj74 said…
I did 4 pregnancies with my usual "bananas and oatmeal" eating style and one higher fat/low[er] carb.

My weight gains were about the same for all but only one involved upward trending blood pressure, edema, beaus lines, failed OGTT, and ultimately an emergency c-section. Guess which one? All anecdotal of course.
SamAbroad said…
@Evelyn (reply link won't work for me?)

I agree, most conventional health advice is pretty good. Don't smoke (good!), avoid deep-fried foods and trans-fats (good!), eat oily fish (good!), eat fresh fruits and veggies (good!) avoid excess refined sugar (good!), avoid highly-processed food (good!).

I think people the problem arises when people conflate conventional wisdom with what food manufacturers 'interpret' as being healthy, which is mainly still processed crap, and also the national bodies that endorse that processed crap.

@bentley, was your pregnancy you had the trouble in your latest one? because there is some evidence in the literature (though conflicting in cases) that parity (number of previous children) or the time between having kids is a risk factor for GDM. I'd wager probabaly due to reduced nutrient status. I wanted to do a RCT for my masters degree to see if magnesium supplementation would improve GDM but the regulatory hoops I would have to jump through made it unfeasible.

I still think there is pretty compelling evidence that vitamin D and magnesium could help improve IR in GDM. I'm involved in a pilot trial that is testing vitamin D supplementation along with healthy eating advice, the D doses are paltry though so it probably won't make much of a difference.
Anonymous said…
Just a quick note. I use to look at these type of studies (involving RR and quintiles) to see if there is a dose-response effect. If you look at table 4, you can see that in both models, the RR for the 5th quintile is higher than the first, but those on the 4th quintile have a reduced RR (only marginally) compared with the 3rd.
Anonymous said…
Hmm. I didn't get above about 100g or so of carbs/day during my pregnancy, when I could eat. No GD or even ketosis. My first trimester was mostly VLC, although I did see my carb tolerance increase as the morning sickness lessened and i could hold food down better.

My child is in top of charts health and well ahead on all milestones. Breastfeeding has been and continues to be pretty simple (except for the whole high caloric need, but I have gotten used to it).

And I myself have much improved endurance and cardiovascular health.

I don't run from carbs, I just find that having to go mainly with tubers and dairy means living pretty low carb (<150g/day) even if i eat a fair amount of carbs in a given meal. and from the perspective of being a nursing mother, my child is far more physically robust than average and seems to be well adapted to eat what i eat (and her father eats) and breastmilk as her diet.
Dawn said…
Bentley, I wondered the same about your pregnancies. If it was your last one it's just as likely that the problems resulted from being older (ahem, none of us are getting any younger) and having had several previous pregnancies rather than your diet.

I've had three GD pregnancies, all of which I treated by reducing carbs (no drugs or insulin). Each time I had to be a bit more strict and the third time I ended up with a few weeks of high blood pressure following delivery (but not before!) and persistent "pre-diabetic" glucose levels. I never overindulged on fat, though, and always kept my carb intake up as much as I could without having glucose go too high. In my anecdotal case, at least, it would seem that my body was predisposed to diabetes no matter what and that GD was my "early warning."

Oh, and I was about 25 lb overweight with the first and was at my ideal weight with the second and third. I'd hoped that losing weight would make a difference, but it didn't (another indicator that I must have diabetes for some reason other than overnutrition/obesity).
Keenan said…
Instead of looking at epidemiological data which is often flawed why don't you look at the controlled studies?

O3 deteoriates glycemic response-
"Average blood glucose concentrations during the third week were significantly higher fasting (+ 15%, p < 0.01), and during the day at 1100 h (+18%, p < 0.001) and 1500 h (+ 17%, p=0.002) on PUFA than on the saturated fat diet."

http://onlinelibrary.wiley.com/doi/10.1111/j.1464-5491.1992.tb01748.x/abstract

Dietary supplementation with n-3 fatty acids may impair glucose homeostasis in patients with non-insulin-dependent diabetes mellitus.-
"The blood glucose concentration tended to increase during MaxEPA treatment, and to decrease during the placebo period, the changes under the two regimes being significantly different (P less than 0.01). In addition, the rate constant for glucose disappearance (k value) for the intravenous insulin-tolerance test, which reflected the peripheral insulin sensitivity, tended to decrease during MaxEPA treatment and increase during administration of the placebo, there being a significant difference (P less than 0.03) between the changes during the two treatments."
http://www.ncbi.nlm.nih.gov/pubmed/2394967

Fatty acid tissue composition and diabetes-
"The FA composition of serum phospholipids (S-PL) measured by gas liquid chromatography and insulin action during a 2-step hyperinsulinemic isoglycemic clamp (1 and 10 mU/kg. min) were determined in 21 newly diagnosed DM2 subjects (DMN), in groups of long-term DM2 patients treated with hypoglycemic agents (DMH; n = 21) or diet alone (DMD; n = 11), and in 24 healthy subjects (HS)....Increased contents of highly unsaturated n-6 family FA (P <.01), arachidonic acid in particular, were found in all groups of diabetics compared with HS."

http://www.ncbi.nlm.nih.gov/pubmed/11735096

LA diet-
Fasting blood glucose and insulin levels were significantly higher on the linoleic acid diet compared with the oleic acid diet (P < 0.01 and P < 0.002, respectively). Plasma cholesterol and LDL cholesterol levels were also significantly higher on the linoleic acid diet (P < 0.001). Likewise, fasting chylomicron apo B48 and apo B100 (P < 0.05) and postprandial chylomicron and VLDL apo B48 and B100 (P < 0.05) were also higher on the linoleic acid diet.

http://care.diabetesjournals.org/content/23/10/1472

Acceleration of experimental diabetic retinopathy in the rat by omega-3 fatty acids-
Despite these biochemical changes, diabetes-associated pericyte loss remained unaffected and the formation of acellular, occluded capillaries was increased by 75% in the fish oil treated diabetic group (115.1 +/- 26.8; untreated diabetic 65.2 +/- 15.0 acellular capillary segments/mm2 of retinal area). We conclude from this study that dietary fish oil supplementation may be harmful for the diabetic microvasculature in the retina.

http://www.ncbi.nlm.nih.gov/pubmed/8721768

Accute effects of fatty acids on insulin secretion-
Fatty acids (0.5 mM) acutely stimulated insulin release from rat islets of Langerhans in static incubations in a glucose-dependent manner. The greatest effect was seen at high glucose concentration (16.7 mM) and little or no response was elicited at 3.3 or 8.7 mM glucose. Long-chain fatty acids (palmitate and stearate) were more effective than medium-chain (octanoate). Saturated fatty acids (palmitate, stearate) were more effective than unsaturated (palmitoleate, linoleate, elaidate).

http://joe.endocrinology-journals.org/content/173/1/73


PUFA consumption has pretty much directly correlated with obesity and degenerative diseases unlike SFA.
bentleyj74 said…
We'll never know for certain but it was not the opinion of my OB that it had anything to do with age or serial pregnancy.

From an experiential standpoint high carb low fat yielded unremarkable easy pregnancies and good glucose sensitivity, LC did the opposite. Victim of my own perception as per usual :)
Dawn said…
When it comes to pregnancy there are so many variables it is really hard to know what causes certain problems. In my case the first delivery was difficult and long and required intervention (pitocin, epidural) but the next two were straightforward and accomplished without any drugs. My first baby was malpositioned, which explains the problems with that birth, although I've always wondered if my being overweight was a problem, too. Giving birth to my third at age 41 was much easier than the first at age 31, though, so I'd be the last to suggest that age was the leading factor in your issues. ;)
carbsane said…
I don't know for sure about what might be appropriate, but those levels are well within normal range. There's no basis for a goal of 5 (or sub-5) promoted in LC circles that I've seen in the literature.