Short post today on a new study suggesting that restricting carbohydrates in women with gestational diabetes mellitus (GDM) does not reduce their chances of needing insulin to control blood sugar during pregnancy. Spanish researchers randomly assigned 152 women with GDM to a control diet or a “low carbohydrate” diet. Each of the diets contained at least 1800 calories. The control diet contained 55% of calories from carbohydrate. And the “low carbohydrate” diet? It specified 40% of calories as carbs, or at least 180 grams. Definitely not low-carb, and significantly more than most of these women need, given the increased insulin resistance that is the hallmark of GDM. In addition, compliance was measured with three-day food records kept by the participants in their own homes rather than in a metabolic ward or other controlled setting where the amount of carbs actually consumed could be accurately assessed. The women’s food records indicated they consumed between 155-191 grams of carbs daily; however, it’s generally accepted that most people tend to underreport the amount of food they eat. The researchers concluded that a low-carbohydrate diet does not reduce insulin needs, but this is misleading given that this diet contained considerably more carbs than the women should have been advised to eat.
I’ve read other studies looking at similar “low carbohydrate” diets that actually weren’t low enough to be therapeutic. Part of the problem is that there’s no universally accepted definition of that term, but in my opinion, up to 190 grams of carbohydrate daily really seems like a stretch. Had these women been prescribed a diet containing less than 100 grams instead, I’m fairly confident the number requiring insulin would have been much lower than those following the control diet. I suspect they would have also had an easier time maintaining a healthy rate of weight gain during their pregnancies.
Unfortunately, given that the media often only report the “conclusions” in research abstracts, the take-home message will be that carb restriction doesn’t decrease the need for insulin in GDM, so there’s no reason to try it. And that’s a shame, because we know that decreasing carbohydrates does lower insulin requirements, provided that the amount consumed is low enough to keep blood sugar at a healthy level. 40% of caloric intake clearly exceeds that limit.
References
1. Morena-Castilla C, et al. Low-carbohydrate diet for the treatment of gestational diabetes: a randomized controlled trial. Diabetes Care. e-published ahead of print April 5, 2013 do1:30.2337/dc12-2714
Whew! Your headline (theirs, really) misled me. I thought you were endorsing this line of baloney. In reality, you were as skeptical — dismissive, actually, of this report as I am. And, of course, I also agree with your conclusion wrt the press take on this study. It will “feed” the notion that (real) low-carb does not benefit the GDM patient. I believe Howard Feinman and/or Richard Bernstein made an attempt to define LC and VLC and VLCKD a few years ago in a paper they published, perhaps in Diabetes Today or some other journal that Bernstain at least is on the Advisory Board of. I will try to find it for you.
Thanks, Dan! Sorry if the title was confusing. Glad you kept reading and discovered that I wholeheartedly endorse carb restriction for women with GDM, as well as all other people with diabetes. I’m pretty sure II have the paper that Dr. Feinman and Dr. Bernstein published with the various criteria for very-low-carb, low-carb, and moderate carb, but thanks so much for the offer. The thing is, those definitions still aren’t agreed upon by everyone, which is why the researchers were able to characterize 180 grams as “low carbohydrate.”
Great post, Franziska. Yeah, 40% carbs is more like Barry Sears’ Zone diet; low carb proper would probably be about half of that. Someone should really try to define this! (like the BMI cutoffs.) Maybe calorie percentages aren’t the most important/useful indicator of diet quality, but cutoffs would be useful for the media and even researchers so that we’re all on the same page!
Thanks so much, Bill! I did find an article by Dr. Richard Feinman proposing the following definitions for various levels of carb restriction. I like these and hope they’ll soon be adopted by all:
Low carbohydrate diet: less than 130 g/d or less
than 26 % of a 2000 kcal/d diet
Very low-carbohydrate ketogenic diet (VLCKD):
less than 20-50 g/d or less than approximately 10 %
of a 2000 kcal/d diet.
Moderate Carbohydrate Diet. 26- 45% of total caloric intake (130-225 g/d for 2000 kcal/d diet)
Hard to believe 180-190 carbs would qualify for any definition of LC. Makes me question if this study was intentionally designed to prove the outcome. Indeed, the headline will send the message, the wrong message, like the ACCORD study.
Thanks, Gerri. Great points.
I must be an anomaly because I managed to avoid insulin use with a low carb diet while I had GDM. So I guess it only worked for me!
Hi Anna,
It worked for you because it was truly low carb! 🙂
Hi Franziska,
Always a great read! I was just reading the Journal from The Academy (June 2013) and they say that GDM patients need a minimum of 175 g of CHO/day and no less than 45% of their calories should be coming from CHO because of the CHO needs of the fetus. What do you think about this? I always like to hear your thoughts since you are so up on the research. Thanks!
Thanks so much for your comments, Gina!
I disagree entirely that a woman with GDM should be consuming 175 grams of carbs at a minimum daily. She will likely need insulin at that carb level, and if her BG is not well controlled her baby runs the risk of developing macrosomia (abnormally large size), as well as other complications.The mother will provide her baby with enough glucose to meet its needs on far less than 175 grams of carbs a day.
There is some debate among low-carb proponents regarding whether being in ketosis during pregnancy poses any problems (none have ever been documented, as far as I know), but consuming 50-100 grams of carbs a day will keep a woman out of ketosis and still allow her to experience the benefits of improved glycemic control and weight management throughout her pregnancy.
Hi Franziska, Diabetes Dietitian from Australia here again. There were actually a couple of concerning studies in the early 1990s correlating 3rd trimester ketosis in DM and non-DM with lower intellegence and motor proficiency scores in the offspring at ages 2-9 years (Rizzo et al N Engl J Med 1991; 325:911-916 and Am J Obstet Gynecol. 1995 Dec;173(6):1753-8). There’s been little research since then, but they are the papers behind the “avoid ketosis” statements in multiple papers&position statements on GDM nutrition since.
Ketosis is much more prevalent in pregnancy metabolism (“accelerated starvation”) compared to non-pregnancy; and far more so in diabetes than non-diabetes. A spanish study in 2011 showed 41% of GDM ladies having significant ketonuria at one or more time points through the day (J Obstet & Gyn Res, 37(7):722-728). According to the IOM weight gain in pregnancy background paper, a chronic increase in ketonemia occurs in 47% of GDM vs only 12% in non-GDM.
But whether this is due to, or worsened by CHO restriction is unclear, although commonly assumed. Even so, I am not convinced that “50-100 grams of carbs will keep a woman out of ketosis” as you claim.
In the late 1990s we had all our GDM ladies testing urinary fasting ketones, and we would often restrict carbs down and not worry about it unless they presented with ketosis. But as it turns out, its not that simple. Ketone testing methodology and evidence is lacking – when to test, if to test, blood or urine, does it even relate to diet, does doing it lead to improved pregnancy outcomes etc etc So many unknowns!
Practically, in our large GDM clinic, we come across a few other low carb concerns. Ladies frequently restrict carb because, yes, it results in a lower BG 2 hours later when they test. But most end up getting excessively hungry, leading to binging at a later meal and high BGs in response. In other cases, where they do maintain low carb, they often end up losing weight or gaining suboptimal weight, thereby increasing their risk of an SGA baby. All this could likely be alleviated by more fat/protein, but that doesn’t always occur to them, and by the time we pick up what is happening, its already been going on for some time.
Another phenomenon that I haven’t been able to find any literature on, but is confirmed by multiple clinics across our country, including ours (at least a few patients per month) . . is the avoidance of an evening carbohydrate snack and/or low carb dinner resulting in a higher fasting BG the next morning. Starting these ladies on ~30-45 g carb at dinner and cup of milk before bed almost always lowers the fasting BG. No idea what is going on there metabolically – but can only surmise lower carb is somehow increasing overnight insulin resistance.
What does have some research evidence is low carb causing higher post prandial BGs in response to a carb load the next day. Specifically, studies into the oral glucose tolerance test showing low carb, particularly the night before the OGTT, giving a higher 2 hour BG during the test. Hence the common recommendation of at least 150g carb per day for the 3 days before OGTT – to reduce the risk of a false positive diagnosis of diabetes. What one recent OGTT study found is that there is a decrease in 1st page insulin secretion following low carb eating, that may contribute to this rebound BG effect.
So to me that’s a concern for my GDM ladies. If they restrict carb today, but then indulge tomorrow, are their BGs going to go even higher in response that 2nd day?
So even though low carb might be fine in non-pregnancy, I’m still feeling the need to be extra cautious in pregnancy, until there is better research to rule out harm. We find GDM can be well-controlled in most cases on ~175 g carb/day – particularly if most of that carb is low glycaemic index – with or without insulin, the safety of which in pregnancy is well documented.
Thank you for your input R. Smith, which is similar to my experience. Having been diagnosed with gestational diabetes and being on a low carb diet for around 1 month, I can testify that I have indeed lost weight resulting in tiredness and concentration loss, and feeling hungry all the time, and what’s more the baby has appeared smallish on the 32 weeks scan… Also I have noticed that low carb meals at night do result in higher glucose levels in the morning.
I have met quite a number of other pregnant women in hospital who are anguished because they are not putting on weight, some having not put more than 3 kg after 6 months !
I think there is something very wrong about this low carb fad, so I am using the glucose tests as a guideline to what is good for me to eat or not, and not weighing my portions as this is just insane. In my case it is the type of carb that matters, white rice and pulses are OK, milk based products are not very good and bread is not good at all….
This sort of malarkey is an old Navaho Indian trick (no disrespect to Navaho native Americans) A so called lowcarb diet is often anything but. The same game is played with fats. I have lost count of negative fat studies and trials, but when checking the small print, trans fats have been included in the diet used. Trans fats have played a major part in metabolic disease and obesity in my opinion. As have refined carbs and so many foods now loaded with sugar and HFCS. A lowcarb diet done properly harms no one, indeed it has been the salvation of many, especially diabetics. For me a lowcarb diet is around 50 grams of carb per day. More than enough to stay fit and healthy.
Keep up your great work, it is much appreciated.
Eddie
Thanks so much for commenting. You’re so right on all points!
Thanks also for your continued support. Your great work is appreciated as well!
“There is some debate among low-carb proponents regarding whether being in ketosis during pregnancy poses any problems (none have ever been documented, as far as I know), but consuming 50-100 grams of carbs a day will keep a woman out of ketosis and still allow her to experience the benefits of improved glycemic control and weight management throughout her pregnancy”
My friend Dr. Jay Wortman a type two diabetic and lowcarb expert second child was carried with his wife on a lowcarb diet. I think the end product speaks for itself. I know people who have met Jays daughter and she is perfect in every way.
Eddie
Jays daughter http://www.drjaywortman.com/blog/wordpress/2012/11/29/low-carb-cruise-2013/
Yes, I’ve read about Dr. Wortman’s wife and also Dr. Andreas Eenfeldt’s wife following LC and giving birth to very healthy and happy daughters. The pic of Jay’s daughter is absolutely adorable!
@ R. Smith,
Thanks very much for your comments and expertise in this area. Admittedly, gestational diabetes is not an issue I’ve encountered very much in my work at the Veterans hospital. I stand corrected in my response to Gina about no studies indicating problems with maternal ketosis.
You bring up some excellent points, and I agree with many of them, particularly the inability to accurately measure ketones and higher BG levels causing problems. And definitely share your opinion that women who alternately eat low carb and then binge will have worse glycemic control and weight control than those who eat a more moderate carb diet.
I just read the Rizzo 1991 study, and unfortunately they don’t specify what the women whose children had lower IQ scores and other problems were eating. In other words, were they in starvation ketosis from suboptimal calories rather than carbohydrates? Or was something else entirely going on that resulted in these issues? The carb level at which one enters ketosis is highly individualized, and perhaps for some pregnant women 50-100 grams may not be enough.
I’m surprised further research has not been done in this area. There are populations — ie, the Inuit — who subsist on extremely low carbohydrate intake (virtually none) for much of the year, and as far as I know there isn’t any evidence that this affects intelligence or causes other problems.
As always, I think a low-carb diet should be an option for people who are interested, but women who are pregnant — particularly those with diabetes — need to be monitored closely by their doctors to ensure a safe pregnancy, and carb intake should be adjusted on an individualized basis.
Thanks again for your comments and polite debate on this topic.
Hi Lilian, yes . . this tendency to undereat in response to GDM diagnosis is widespread. I see multiple patients every week doing this. But paradoxically, there are clear links between small for gestational age babies and insulin resistance/obesity/cardiovascular disease in later life (If interested, see BBC documentary on utube: “The 9 months that made you”). So undereating may well be just as problematic as overeating in the life long effects on that baby.
Its understandable though. Ladies are naturally worried about their baby and GDM diagnosis. Many are very fearful of going on to insulin injections. So they over-restrict their diets, particularly carbs and endure misery, severe hunger and frustration, thinking they are helping their baby. Some exercise fanatically after every meal, just to keep their glucose readings low.
In my GDM experience (15 years), the most successful approach seems to be low glyceamic index carbs spread over 2-3 hourly meals and snacks (eg 3 small meals and 3-4 snacks). Research supports this approach (see references at end), whereas there is no research on very low carb diets in GDM.
And that’s not because of some “cover-up-the-truth conspiracy” on the part of researchers, as implied by some comments above. Its simply because this is now an ethical catch 22, due to the ketosis articles I mentioned in my previous post. The authors of this study may well have wanted to study lower carb levels, but would not have been permitted by their research ethics committee, due to the risk of ketosis and association with neurological development shown in those studies from the 90s. So we’re stuck on that front.
But getting back to practicalities in managing GDM . . . I’ve seen thousands of glucose readings over the years, and the top 4 food reasons for elevated readings seen in our clinic are:
1. Rice
Streets ahead of all other carb sources in causing problematic BGLs, (however we do have high numbers of patients from rice-based cuisines). Basmati is much better than jasmine rice, but amounts still need to be quite controlled (“fist” size often works).
2. White bread. Most wholemeal and multigrain breads are also high GI, so potentially problematic once you go beyond about 2 slices at a meal (on average). Low glycaemic index breads are dramatically better, but are very brand-specific. We are fortunate Sydney University does a lot of GI testing, so we know our brands here.
3. Takeaway foods
4. Foods/beverages with high amounts of refined sugar – cakes, desserts, softdrinks etc
In contrast, carbs which I hardly ever see resulting in high BGs are:
1. milk and yoghurt (glycaemic index is extremely low, and most people even tolerate sugar-sweetened varieties with little difficulty). Great for the pre-bed snack in assisting with fasting BGs the next morning.
2. legumes/pulses/nuts
3. Many fruits – especially if taken as snacks in an amount the size of the fist.
So quite obviously, all carbs are not the same metabolically. I think that’s the one big weakness of the ‘low carb’ movement. It is far too primitive to assume all carbs act the same way; and that cutting out the whole group will be universally beneficial. Wholegrains for example, contain thousands of phytonutrients in addition to carbohydrate with health promoting effects. Do you really want to throw out the baby with the bathwater? We all like a simple approach, but the human body is just not that simple.
Moses R.G et al. The effect of a low glycemic index diet during pregnancy on obstetric outcomes. Am J Clin Nutr. 2006; 84(4) 807-812.
Moses R.G et al. Can a low glycemic index diet reduce the need for insulin in gestational diabetes mellitus? Diabetes Care. 2009;32(6) 996-1000.
Hi RS Smith,
Thanks again for your comment; however, I respectfully disagree that using moderate amounts of low GI foods will result in adequate control. The amount of total carbohydrate has at least as much impact on postprandial BG as the type.Of course, carb tolerance varies from person among PWDs and women with GDM, so some may do well at higher amounts. I also dispute the assertion that whole grains are a necessary part of the diet; fruits and vegetables contain more phytochemicals in their richly colored pigments.
Let’s just agree to disagree on this issue. I appreciate the time you took to respond. Have a nice week 🙂
Franziska
Third pregnancy, first time with GD here: Read all the comments and appreciate the multiple view points. I tried low carb dieting in fhe past and ended up with reduced seratonin production resulting in deprressive episodes. I’ve also found the tendency to undereat to be a burden. Instead of clinging to some extreme formula of ‘’correct’’ eating, I’m going to try a more balanced diet and keep track of my readings. One thing that’s clear is that different people need different diets at times.
Hello, Joy,
Thanks for your comments, and congratulations on your third pregnancy! I wish you the very best of luck going forward.
Franziska