GDM and the Ketone Boogeyman

This was posted yesterday to a Facebook group to which I belong:

Gestational Diabetes. Sister in law is 32 weeks pregnant. Doc is telling her she needs 150-250 carbs per day. Yet her glucose is still high pretty much all day and especially in the morning. She is doing finger pricks like 4 times a day and if she doesn't get her levels below 130, consistently, she will have to have other interventions. I know this can be helped with food...

Being of personal and professional interest to me, I went searching for resources and got busy with some synthesis. The trouble is, ancestral health (AH) sciences have not had a chance to penetrate into the corners of human health conditions yet, meaning the vast majority of relevant published literature focuses on AH issues in adult men, or CW methods of addressing obstetric conditions. There's also a long-standing tradition in medicine to view the adult male as the default physical model, leaving women - and heaven forbid pregnant women - as de facto pathology cases. Combining that conceptual substrata with the diabetic disease state sets off any number of ringing bells and blinking lights that promptly summon the heavy pharmaceutical artillery.

But if we took a step back and changed the underlying assumption to one more in line with AH, and accepted that the female also functions, and that pregnancy is a normal aspect of female physiology, what would we find?

First we have to accept that most westerners are not living la vida low carb. They start the reproductive cycle from that basis.

Second, diabetes is a disease state. Sure, it is rather well studied and we have some solid understanding of how it works and thus how to control it, but it is by definition pathological and so its complexity should not be underestimated. 

Third, pregnancy metabolism is a tricky beast.

Taken together, AH responses to GDM are going to be hard to pin down. Additionally considering the request that prompted this post is an urgent one, I must demur that my answer will be incomplete.

Why is diabetes (DM) a problem, and what do we do about it?

Diabetes mellitus isn't fun. There are two types (autoimmune, or type 1, and induced, or type 2.) Both are malfunctions of the body's insulin metabolism; in T1DM the body kills off its own insulin-producing cells, and in T2DM the body stops responding to the insulin that's present so the pancreas produces an ever-increasing amount until it burns out. The end result is uncontrolled blood glucose, which is highly damaging to many bodily systems, and can be fatal. If insulin is the police, sugar is the unruly mob. Diabetes is gang warfare.

Generally speaking, and the American Diabetes Association's recommendations aside, we know that carbohydrate intake is the primary determinant of blood glucose values, and keeping blood glucose low and steady keeps the hordes under control and improves the health and lifespan of diabetics, so a low-carb diet is advised. There's loads on this and I'm going to assume you're aware of it and not delve into it any further. If you want resources beyond LMGTFY, please comment below.

What is gestational diabetes (GDM) and why is it a problem?

Insulin resistance is a normal phenomenon emerging in the second half of pregnancy, which is thought to secure glucose supply to the growing fetus. In women with GDM, normal insulin resistance spins out of control. Wikipedia suggests that its causes may be "similar to those in type 2 diabetes: autoimmunity, single gene mutations, obesity, and other mechanisms" and warns that

Because glucose travels across the placenta (through diffusion facilitated by GLUT3 carriers), the fetus is exposed to higher glucose levels. This leads to increased fetal levels of insulin (insulin itself cannot cross the placenta). The growth-stimulating effects of insulin can lead to excessive growth and a large body (macrosomia). After birth, the high glucose environment disappears, leaving these newborns with ongoing high insulin production and susceptibility to low blood glucose levels (hypoglycemia).

Babies of GDM pregnancies are at increased lifetime risk of  metabolic disorders as a result of this epigenetic starvation signalling.

Pregnancy in women with pre-existing diabetes is associated with a tenfold increased risk for sudden late-stage intrauterine deaths, either of an unexplained nature or due to placental immaturity and insufficient blood transference, and/or impaired oxygen diffusion resulting in acidosis. The risk of adverse outcome is related to the degree of maternal glycaemic control. Regular monitoring of glycosylated haemoglobin (HbA1c) levels are of limited value in indicating glucose control since a physiologic decrease occurs during pregnancy.

Why is the G a problem in GDM management?

Pregnancy is a state of tension between the mother and fetus for nutrients, with the battle front policed by the placenta. Initially the needs of the fetus are low enough that the mother experiences an anabolic state: her body stocks the larder by means of increased insulin sensitivity, lipogenic and glycerologenic activity. The production of insulin-antagonistic hormones progressively rises during gestation as fetal demands increase, while rising progesterone slows digestion, increases absorption of carbohydrates, and thereby promotes hyperglycaemia. Despite the increased physiological need for insulin action, insulin sensitivity may fall by as much as 56% during gestation. This is a boon for the fetus, which starts mainlining glucose to feed brain growth and fat deposits. That draw-down on the maternal system is powerful enough to induce maternal hypogycemia in the midst of this candy store, a switch to insulin-resistance and a net catabolic state, and the breakdown of the fat deposits laid down earlier.

One study specifically investigating carb metabolism in normal and GDM pregnancies gave women glucose tolerance tests prior to conception and at 12-14 weeks, and 34-36 wk of gestation. In all cases insulin sensitivity decreased during gestation as expected, but in GDM it was excessively reduced between points 1 and 2. (The authors go on to note that "basal endogenous glucose production increased during gestation, but there was resistance to suppression by insulin infusion in late gestation", suggesting that the standard treatment is in fact ineffective) That suggests that GDM puts women behind in the larder-stocking stage, inducing jagged blood sugars and prompting "rescue eating".

A common response to hypoglycemia is to have regular snacks providing fast sugars. The trouble with this method is that

• it's terribly difficult to ensure that you're getting just enough sugar, and very easy to get far too much. 
• insulin response is slowed during pregnancy, so that sugar overdose hangs around far too long
• excess sugar becomes toxic to the body and "gums up" circulation
• clogged pathways convey insulin signalling even slower than before
• insulin resistance is increased

So what happens if/when we don't have access to a cookie every hour? The metabolic changes that accompany pregnancy create a predisposition to ketosis. While doctors are often dire about the effects of ketone bodies on the fetus,

In human fetuses at mid-gestation, ketones are not just an alternative fuel but appear to be an essential fuel because they supply as much as 30% of the energy requirement of the brain at that age (Adam et al., 1975, from here)

Latter-stage gestational release of adipose tissue, stemming from fetal resource-hogging leading to decreased maternal fasting glucose which is perceived by the body as relative starvation, stimulates liver synthesis of triglycerides and their later release into circulation. The liver then breaks these free fatty acids (FFAs) down into ketone bodies when fasting (such as during sleep), protecting the fetus from maternal metabolic insult. This state of

exaggerated maternal hypertriglyceridemia ... besides being a floating metabolic reserve for emergency conditions such as starvation, constitutes an essential substrate for milk synthesis around parturition in preparation for lactation. [source]

Not bad at all! So why the scary stories?

Often ketosis and ketoacidosis are conflated, and the appropriate warnings about the risks of the latter are extended to the former. After combing through quite a few hits, I concluded that PaleoHacks was the most succinct source of an explanation for the mechanisms behind and the differences between ketosis and ketoacidosis. For brevity I have smooshed a few replies together. To the best of my knowledge this has not altered the meaning at all.

Ketosis occurs when free fatty acids are being broken down into ketones and then burned in cells, instead of the cells burning glucose. Usually, these ketones don't build up because they are produced to be used up. Ketoacidosis is caused when there is an extreme build-up of ketones (by something like a factor of 10 compared to ketosis).  In a healthy person, ketone production will be kept in check by insulin, but in a state of absolute insulin deficiency or a non-functional isoform thereof, there is a

lack of suppression of lipolysis; uncontrolled increase in free fatty acid levels which causes excessive ketone bodies to be produced.

lack of suppression of gluconeogenesis; uncontrolled increase in blood glucose levels.

As a result of these simultaneous processes, ketone bodies accumulate to dangerous levels. The body works to maintain blood pH, but it eventually becomes overwhelmed by this pathological production of ketone bodies, and metabolic acidosis results. 

Diabetic ketoacidosis presents a serious danger to mother and baby, and aspects of pregnancy increase the risk of its occurrence compared to non-pregnant diabetics. At the risk of putting all my eggs in one basket, this article addresses the major issues:

the incidence of diabetic ketoacidosis in pregnancy is 1%–3%, with fetal loss rate of 9%... The exact mechanism by which maternal diabetic ketoacidosis affects the fetus is unknown. Ketoacids as well as glucose readily cross the placenta. Whether it is the maternal acidosis, hyperglycaemia, severe volume depletion, or electrolyte imbalance that has the most detrimental effect on the fetus is unclear.[15]

The development of diabetic ketoacidosis in pregnancy can be rapid and can also occur at lower blood glucose levels compared to nonpregnant women. The use of reagent strips to detect ketones in urine (Ketostix) when blood glucose levels are high, or if symptoms of intercurrent illness appear, may be one way of early identification of this complication. However, the presence of minor ketonuria in normal pregnancy, especially in the presence of significant emesis, should be borne in mind during evaluation of such patients. The use of reagent strips to detect ketones in blood may help in the differentiation of these two conditions, although this needs validation for its use in routine clinical practice.

What issues might not pose a problem with GDM?

Even Wikipedia notes up front that the 3-10% prevalence of high blood glucose levels during pregnancy may indicate it is in fact "a natural phenomenon". This would certainly not be the first time that CW had set a tighter bell curve than Mother Nature!

There has been some concern that maternal hyperglycemia causes fetal cardiac distress. Using continuous fetal heart rate and maternal blood glucose monitors, a Dutch study observed the correlation of diabetic mothers' blood sugar and their babies' heart rate:

FHR [fetal heart rate] and STV [its short term variability] were within limits of normality during all episodes [of dysglycaemia] demonstrating that dysglycaemic excursions were well tolerated by the fetuses in our study population. Multilevel analysis of the hyperglycaemic events showed a significant effect of glucose on FHR, but not on STV... This effect probably represents an expression of physiology rather than pathology.

They do indicate that the results are not so benign when ketoacidosis is involved, though. But otherwise both parties seem to handle a moderate sugar rush reasonably well.

AH-friendly treatment approaches to GDM

Despite the normal changes in insulin response throughout pregnancy, the wonky wheel that is diabetes cannot be left to fix itself.  Murphy et al. (cited here) reported an observation of increased glycemic control with advancing T2 diabetic pregnancy. In contrast, "the degree of glycaemic variability during the first trimester does not predict variability in later pregnancy" in T1D pregnant women on insulin. This may be due to the relative effectiveness of personal blood glucose monitors in inducing behaviour change in each of the two DM types. Fortunately then, GDM behaves more like T2D than T1D.

And that means, as ever, going low carb. Even when it is defined at levels the paleo community would swoon at, like the 42-50% found to be beneficial in this GDM study, reducing the fuel for the hot and fast fire and asking the body to use low and slow burning fats and proteins instead has been found to make a significant difference in blood glucose values.

Before anyone goes zero carb as a result of this post (!) let me point out that there are a couple of problems with sudden drops in carb intake during pregnancy. As noted, the fetus' preferred fuel is glucose. They've got a lot to do in not very long, and quick and easy suits their nutritional plan just fine, thankyouverymuch. Going cold turkey on grains, fruits, processed sugars (and in some cases, dairy) will set off a starvation response in both parties. In the adjustment period, the maternal body may not be able to compensate sufficiently for the loss of its usual energy source, and the nutrient supply to the fetus may be reduced. The mother will get carb flu which she can manage and get through, but the baby's developing endocrine system will get a memo about food scarcity that it will never forget and which may predispose that child to metabolic disorders throughout their life. More immediately, the intricacy of gestation is akin to the laugher of god, and I wouldn't want to guess at when a safe week would be to impair that.

Which is not to say anyone is doomed to Ritz crackers, injections, and large-headed babies. Buying or having prescribed a standard blood glucose monitor and using it to get to know your own body's need for and response to food is wise in any case.

• Eat mindfully and track intake while testing blood sugar before meals and at hourly intervals afterwards for a week before making any changes. Then start lowering the carb component, tracking carefully to keep sugar levels from spiking. 
• Since insulin resistance is highest in mornings, carbs need to be restricted at breakfast; identify how many eggs you need to get through to lunch without a crash, and when you need to have an afternoon snack to prevent panic-munching while preparing dinner. If morning readings continue to be high it might be a sign of cortisol response as a reaction to hypogycemia over night; employ Ketostix before breakfast and try adding a small, high-fat snack before bed (half a cup of plain yogurt or a few macadamias would do).
• Make sure you are not reducing overall caloric intake unnecessarily; there is no need to gain large amounts of weight during pregnancy, but nor is it wise to attempt to use your baby as an endogenous tape worm for weight loss. I believe that much of the medical establishment's indulgence of women's eating habits while pregnant comes down to the hope that quantity of food will ensure sufficient nutrient intake. Focussing on nutrient density therefore relieves the need to gorge, with its concomitant ill effects.

Exercise is the best remedy for high trigs, and will thereby - in concert with a reduced-carb diet - address the potential for overload of ketone bodies. Go for a walk! Aim for a 2-minute full squat 10 times a day; that'll get your biggest muscle groups working as well as give you an advantage for birthing. Go to yoga in the evening while you still can. The other benefit of this arm of the treatment is its stress-reduction aspects.

And we all know that stress is terrible for pregnant women!

___________________________________________________________

For further reading, please avail yourself of the embedded links or wander over to Keto-adaptation: what it is and how to get through it