Current dietary advice for women with gestational diabetes mellitus is to avoid diets that result in elevated ketone levels. This guidance stems from a concern that maternal ketones are associated with poor fetal and childhood outcomes, including reduced childhood intelligence quota. The evidence behind these guidelines is conflicting and inconsistent. Given that dietary counseling is the initial treatment strategy for women with diabetes in pregnancy, it is important that clinicians understand the concern regarding maternal ketones. This review examines the physiology of ketogenesis in pregnancy, the prevalence of elevated maternal ketone levels, and the relationship between maternal ketones and fetal and childhood outcomes.

In 2002 and 2008, the American Diabetes Association (ADA) published evidence-based nutrition principles for the treatment and prevention of diabetes (1,2). Both articles recommended that women with gestational diabetes mellitus (GDM) consume a diet with adequate energy levels for appropriate gestational weight gain and absence of ketones. ADA dietary guidelines currently recommend a minimum of 175 g carbohydrate per day in women with diabetes in pregnancy (DIP) (3). The stated concern with low-carbohydrate diets in pregnancy is the potential for increased maternal ketone levels. Ketones are synthesized in the liver from the breakdown of lipids as the body switches from glucose to lipid metabolism to ensure adequate energy production. Ketone synthesis is accelerated in pregnancy, particularly in the third trimester. This is due to increasing maternal insulin resistance and rising fetal demand for energy, which leads to increased maternal lipolysis (46). Elevated maternal ketone levels have been associated with adverse fetal outcomes, including low childhood intelligence quota (IQ) (7), oligohydramnios, fetal heart decelerations, and nonreactive nonstress tests (8). However, studies do not consistently find associations with these outcomes (9).

Carbohydrate restriction has been shown to be an effective method for reducing blood glucose levels in patients with diabetes outside pregnancy (10,11). It has also been evaluated as a treatment option for women with GDM, although with conflicting results (1214). The popularity of low-carbohydrate diets in the general population has increased in recent years, with the rise of approaches such as the Atkins and Paleolithic diets. Given that dietary counseling is the key initial strategy for management of GDM (15), diet trends in the broader community are likely to influence pregnant women. The exact prevalence of carbohydrate restriction among women with DIP is unknown. A study in Denmark of 204 women found that 52% of those with type 2 diabetes and 40% with type 1 diabetes consumed <175 g of carbohydrate per day in early pregnancy (16). This suggests that carbohydrate restriction may be fairly common in women with DIP. It is therefore important that clinicians understand whether there are adverse outcomes from a low-carbohydrate maternal diet that has the potential to lead to increased ketone levels. This article reviews the physiology of ketones in pregnancy, the range and prevalence of elevated maternal ketones, and the relationship between ketones and pregnancy outcomes. Where the term DIP is used, this refers to the combination of pregestational and gestational diabetes.

Ketones are organic compounds produced from the breakdown of fatty acids in the liver. Ketone production increases during periods of low glucose availability to maintain an adequate supply of energy in the form of transportable acetyl-CoA that can be converted into ATP in the brain and other organs. Ketone production requires two separate metabolic processes, lipolysis and ketogenesis. Lipolysis is where triglycerides are broken down into fatty acids and glycerol, a process inhibited by insulin and stimulated by glucagon. Fatty acids are released into the circulation, taken up by the liver where they are transported into the mitochondria, and converted into ketones via ketogenesis (17) (see Fig. 1). Both processes are upregulated in low glucose and low insulin states. The first ketone produced is acetoacetate, which can be converted into β-hydroxybutyrate (BHB) and acetone. BHB is produced by the reduction of acetoacetate in the mitochondrial matrix using NADH as the electron donor. The normal ratio of acetoacetate to BHB is 1:1. Acetone is a breakdown product of acetoacetate and cannot be used as an energy source. In healthy, nonpregnant individuals, overnight fasting levels of serum BHB range from 0.1 to 0.4 mmol/L (18). “Normal” serum BHB levels are therefore considered to be <0.5 mmol/L (19). Serum BHB levels are considered elevated if the level is >1.0 mmol/L, and ketoacidosis is likely to occur at levels >3.0 mmol/L (19,20).

Figure 1

Ketogenesis pathway in the liver. HMG, Hydroxymethylglutaryl. Enzymes appear in italics (17).

Figure 1

Ketogenesis pathway in the liver. HMG, Hydroxymethylglutaryl. Enzymes appear in italics (17).

Close modal

Ketones and Dietary Carbohydrate

Low-carbohydrate diets result in elevated ketone levels due to the reduction in available glucose. In a crossover meal study of nonpregnant individuals, the average BHB level 5 h after a 10% carbohydrate meal was 0.2 mmol/L, compared with 0.08 mmol/L after a 60% carbohydrate meal (P < 0.007) (21). It is not known exactly what level of carbohydrate restriction leads to increased ketone levels because this varies between individuals and metabolic demand. In the nonpregnant population, it has been generally reported that diets with <50 g carbohydrate per day will induce ketosis (22,23). In pregnancy, it is also unknown exactly what level of carbohydrate restriction leads to increased ketone levels. A recent study of women with GDM found that a modestly low-carbohydrate diet of 165 g of carbohydrate per day does not result in increased fasting BHB levels (24). Another study measured total serum ketone levels in seven women with GDM consuming a diet restricted to 150 g carbohydrate per day. Total ketone levels on the restricted diet were compared with levels from four of these women following their usual diet at a different time, and three other women with GDM following their usual diet. Total ketone levels were higher when women followed the carbohydrate-restricted diet; however, the difference was not statistically significant, potentially due to the low numbers in the study (blood total ketone body levels 0.26 ± 0.15 mmol/L vs. 0.19 ± 0.05 mmol/L) (25).

Ketone Physiology in Pregnancy

Nutrients from the maternal circulation cross the placenta to the developing fetus. The fetus is capable of using multiple energy substrates, including glucose, amino acids, free fatty acids, and ketones (26). Maternal ketogenesis is accelerated in pregnancy, particularly in the third trimester (4,5). In women admitted to the hospital who were fasted, levels of serum BHB and acetoacetate rose significantly faster and were up to three times higher in pregnant women in the third trimester compared with nonpregnant control subjects (4). For this reason, pregnancy has been termed a state of accelerated starvation.

The increase in maternal ketogenesis is secondary to multiple metabolic changes. Maternal insulin resistance develops in the second trimester largely due to secretion of placental growth hormone. This hormone functions as an insulin antagonist, impairing maternal glucose utilization and stimulating lipolysis, which leads to increased ketogenesis (6). In addition, maternal fasting glucose levels fall in pregnancy (2729), resulting in increased maternal lipolysis and ketogenesis (30).

The Placenta and Its Role

Ketones cross the placenta by passive diffusion from the mother to the fetus (26). The concentration of maternal serum ketones is twice that of the fetus, creating a concentration gradient that drives this process (5,26,31). Measurement of the arteriovenous differences across the placenta show that energy transferred from the mother to the fetus from ketones is ∼6% of that from glucose (2.5 cal/kg in 24 h vs. 39 cal/kg in 24 h) (32,33). Enzymes have been found in the placenta capable of converting acetyl-CoA into BHB, meaning the placenta itself has the ability to undertake ketogenesis (34). There is also in vitro evidence in rats that the placenta is capable of using ketones as a fuel and may do so preferentially over glucose (35). When rat placental tissue is incubated with labeled fuel mixtures containing a high concentration of BHB, the CO2 derived from ketones is higher than that derived from glucose and lactate.

Ketone Physiology and the Fetus

The role of ketones in the developing central nervous system is not well understood. The fetal brain, kidney, and liver contain enzymes capable of catabolizing ketones, with the brain expressing these enzymes already at 8–10 weeks’ gestation (36). The activity of these enzymes is increased in conditions that lead to elevated maternal ketones such as starvation and low glucose states (5). Ketones are used by the fetal brain for energy and are also incorporated into cerebral lipids and proteins, suggesting that ketones are a building block for the developing brain (36). The molar consumption of labeled BHB by isolated fetal brain obtained by therapeutic abortion between 12 and 21 weeks’ gestation is 1.47 times that of glucose, demonstrating that ketones potentially provide a large proportion of the fuel to the fetal brain (37). Evidence therefore suggests that ketones are a fuel for the fetal brain and potentially a necessary component of brain development. However, the evidence is limited, making it difficult to draw definite conclusions.

Two animal studies from a single research group in the early 1980s showed that the ketone bodies acetoacetate and BHB independently inhibit de novo biosynthesis of pyrimidines and purines in fetal rat brain in vitro (38,39). The authors speculated that this may be a causal link between maternal ketones and lower childhood IQ (38,39). They state that “under conditions of maternal starvation, ketone bodies could function as a maternally derived ‘regulatory hormone’ by slowing rates of new cell growth in the presence of limited availability of nutrients from the mother.” The first study examined the de novo pathway for purine synthesis by incubating fetal rat brains with BHB and acetoacetate for 3 h at 37.5°C (38). The second study examined the de novo pathway for pyrimidine synthesis by incubating fetal rat brains with BHB and acetoacetate for 2 h at 37.5°C (39). Both studies included 10 mL of Krebs-improved Ringer II buffer solution, which contains 5.5 mmol/L of glucose. BHB and acetoacetate both independently inhibited the synthesis of purines and pyrimidines, and increasing concentrations of BHB led to progressive inhibition. Furthermore, this inhibition occurred at concentrations of BHB as low as 1.4 mmol/L. Pyrimidines and purines are building blocks for the synthesis of both RNA and DNA.

These studies have not been replicated, and the physiological response to ketones in human fetal brains may differ from rodents. There may be no inhibition of de novo synthesis of purines and pyrimidines due to differing enzymes and enzymatic pathways, the abundance of available amino acids, which also can be converted into nucleic acids, or differences in the environmental pH. In these studies, brain cells were incubated for 2–3 h in a solution with ketone bodies ranging from 1.4 to 43.2 mmol/L. Whether this mimics what happens in vivo in humans is unknown and raises the following questions.

  • Do ketone levels stay consistently elevated for such a time period in humans?

  • Do human fetal brain cells “incubate” in ketones or do the ketone levels that the cells are exposed to fluctuate?

  • Is the human fetal brain exposed to high ketone levels in the presence of adequate glucose?

Given these uncertainties, it is difficult to draw conclusions about the effect of elevated ketone levels in fetal human brains.

Studies in mice have shown differences in brain development in offspring of mothers consuming ketogenic diets (KDs). Offspring of dams fed a KD consisting of 0.6% carbohydrate and 67.4% fat had significant alterations in brain structure compared with offspring from dams consuming a standard diet (SD) with 76.1% carbohydrate and 5% fat (40). Analysis of brain structure at 21.5 days postpartum showed that offspring from KD dams had statistically significant reductions in the volume of the cortex, corpus callosum, fimbria, lateral ventricles, and hippocampus compared with offspring from SD dams. In addition the hypothalamus and medulla were relatively enlarged. The same authors repeated the study with different mice and undertook neurobehavioral tests in the offspring at 8 weeks of age (41), including the open-field, forced-swim, and exercise wheel tests. Offspring from KD dams displayed increased physical activity and reduced susceptibility to anxiety and depression compared with offspring from SD dams.

Ketones can be measured in serum, urine, and expired breath. Historically, urinary ketone measurements were the mainstay of clinical care and research. Urinary testing kits detect acetoacetate, and the results are semiquantitative rather than an exact level (19,20). The test kits have a significant rate of false-positive and false-negative results (42). Urine ketone testing has been surpassed in both clinical care and research by serum BHB levels because this is the main ketone responsible for diabetic ketoacidosis (DKA) (20). Capillary blood testing for BHB became available in the late 1990s and is reliable and precise, enabling rapid and easy-to-perform point-of-care testing (20).

Given that many studies evaluating the safety of maternal ketones have tested urinary ketones, it is important to understand the correlation between urine and serum ketones. Few studies have directly evaluated this correlation in pregnancy; however, those that have, have found a poor correlation. Fasting serum BHB levels were tested simultaneously with urinary ketone levels in 180 Greek women with GDM in the third trimester. The prevalence of urinary ketones was significantly higher than the prevalence of elevated serum ketones (36% vs. 5%, P < 0.001). This indicates there was a poor association between fasting urine and serum ketone levels because women with moderate and high amounts of urinary ketones often had normal serum ketone levels (43). The poor correlation may be due to several factors. The renal excretion of ketones is not always linear because ketones are both reabsorbed and excreted by the kidney to varying degrees dependent on the serum ketone level (44,45). Urine is also stored in the bladder for hours before it is excreted, meaning that a urinary ketone level likely reflects the serum ketone level some hours prior. Urine ketone measurements are also affected by dehydration, as the more concentrated the urine, the higher the measured ketone level.

Studies reporting the prevalence of ketones in pregnant women consuming their usual diet are lacking, particularly for serum ketones. A study of 180 women with GDM found that 5% of women had moderate levels of serum ketones at routine clinic visits, and 1.6% of women had high levels (43). More information exists regarding the prevalence of maternal urinary ketones, with most of the studies being performed in women with DIP. Studies report a wide variation in prevalence, ranging from 9 to 89% of women (12,46). Studies vary in their methodology, and the level of urinary ketones considered to be positive differs between studies. One study measuring urinary ketone levels once at term found a prevalence of 9% (8), and another measuring urinary ketones every 3–4 days from the end of the first trimester found a prevalence of 89% over the course of pregnancy (46). This suggests that urinary ketones may vary from day to day and that single time point measurements are likely to underestimate the prevalence. A summary of studies can be found in Table 1. While it is generally accepted that ketogenesis is accelerated in pregnancy, the exact prevalence of detectable urinary ketones or elevated BHB levels in not known. Furthermore, it is not clear what level of ketones in pregnancy is considered harmful and, therefore, what the relevance of an “unacceptably high ketone levels is.”

Table 1

Articles reporting on the prevalence of urinary ketones in pregnant women following their usual diet

ArticleSample population methodologyLevel of ketones considered to be positive on urine dipstickPrevalence of elevated urinary ketone levels
Churchill et al. (7) 1969 N = 107. Diabetes of any type. Urine tested anytime in T3 except for the 24 h before delivery. +1 (small) 58% 
Churchill et al. (51) 1969 N = 213. Diabetes of any type. Urine tested anytime in the 24 h before delivery. +1 (small) 57% 
Stehbens et al. (53) 1977 N = 78. Diabetes of any type. Urine tested every 6 h during any hospitalization in the pregnancy. Not stated 51% 
Persson and Gentz (58) 1984 N = 85. Type 1 diabetes (n = 64), GDM (n = 21). 1–2 weekly 12-h urine collection until 32 weeks’ gestation, then daily 12-h urine collection until delivery. Not stated Type 1 diabetes: 26%; GDM: 50% 
Chez and Curcio (46) 1987 N = 9. No diabetes. 24-h urine collection every 3–4 days from the end of T1 until delivery. 0.5 mmol/L (trace) 89% 
Hamdi et al. (8) 2006 N = 360. No diabetes. Single urine test at >40 weeks’ gestation. ≥0.1 mmol/L (small) 9% 
Spanou et al. (43) 2015 N = 180. GDM in T3. Urine tests at routine clinic reviews. +/++ (moderate) 35% 
Robinson et al. (62) 2018 N = 187. Overweight or obese women. Urine tests once per trimester. 0.5 mmol/L (trace) 22% 
ArticleSample population methodologyLevel of ketones considered to be positive on urine dipstickPrevalence of elevated urinary ketone levels
Churchill et al. (7) 1969 N = 107. Diabetes of any type. Urine tested anytime in T3 except for the 24 h before delivery. +1 (small) 58% 
Churchill et al. (51) 1969 N = 213. Diabetes of any type. Urine tested anytime in the 24 h before delivery. +1 (small) 57% 
Stehbens et al. (53) 1977 N = 78. Diabetes of any type. Urine tested every 6 h during any hospitalization in the pregnancy. Not stated 51% 
Persson and Gentz (58) 1984 N = 85. Type 1 diabetes (n = 64), GDM (n = 21). 1–2 weekly 12-h urine collection until 32 weeks’ gestation, then daily 12-h urine collection until delivery. Not stated Type 1 diabetes: 26%; GDM: 50% 
Chez and Curcio (46) 1987 N = 9. No diabetes. 24-h urine collection every 3–4 days from the end of T1 until delivery. 0.5 mmol/L (trace) 89% 
Hamdi et al. (8) 2006 N = 360. No diabetes. Single urine test at >40 weeks’ gestation. ≥0.1 mmol/L (small) 9% 
Spanou et al. (43) 2015 N = 180. GDM in T3. Urine tests at routine clinic reviews. +/++ (moderate) 35% 
Robinson et al. (62) 2018 N = 187. Overweight or obese women. Urine tests once per trimester. 0.5 mmol/L (trace) 22% 

T1, first trimester; T3, third trimester.

DKA

Maternal DKA is associated with adverse outcomes in the fetus, with high rates of perinatal mortality and morbidity (47). The serum ketone concentration in DKA is extremely high: acetoacetate and BHB can be 100–200 times higher than normal, reaching levels of 10–20 mmol/L (48). Cardiotocography performed during maternal DKA suggest fetal distress, and in one study, late fetal decelerations improved with correction of the DKA (49,50). However, it is not clear whether it is the elevated ketones themselves that cause harm to the fetus or whether it is due to other confounding factors such as the acidosis, dehydration, or the underlying cause of the DKA event.

Maternal Ketones and Childhood IQ in Women With DIP

The two early influential studies examining ketones in pregnancy were retrospective studies performed by an overlapping set of authors in the United States in 1969. They reported an association between maternal urinary ketones in women with DIP and low childhood Stanford-Binet IQ score (7,51). In both studies, urine tests were performed at irregular intervals for unspecified reasons. Women were separated into two groups. Case subjects consisted of women with DIP, either pregestational or gestational as per White’s classification, which at the time would have reflected severe disturbance of glucose metabolism (52). If a woman had a urine test at any time that showed a ketone level of +1 or more, she was classified as a “ketone-positive” case. “Ketone-negative” case subjects were women with no ketones in the urine or only trace amounts. Control subjects consisted of women without DIP who had no urinary ketones or only trace amounts. Control subjects were matched for hospital of birth, race, socioeconomic status, maternal age, and birth order. Mean childhood IQ from the offspring of case subjects and control subjects was tested at 4 years of age. The IQs of offspring from control subjects were firstly compared with the IQs of offspring from case subjects with ketones and secondly with the IQs of offspring from case subjects without ketones.

In the first study, urine was tested at any time in the third trimester except the 24 h before delivery (7). There were 122 ketone-positive case subjects and 91 ketone-negative case subjects, with 213 matched control subjects. Gestation at delivery was significantly shorter in case subjects compared with control subjects (38.1 vs. 39.6 weeks, P < 0.001), and mean birth weight was significantly higher (3.41 vs. 3.26 kg, P < 0.05). The IQs of children born to ketone-negative women with diabetes were not significantly different compared with control subjects (101 vs. 101, P = NS). However, children born to ketone-positive women with diabetes had significantly lower IQs than children born to control subjects, thus linking the presence of urinary ketones with low childhood IQ (93 vs. 102, P < 0.001). The effect on IQ remained the same when women with GDM and pregestational diabetes were examined separately. In addition, women without DIP were analyzed separately. Children of 111 ketone-positive women had significantly lower IQs than children of matched ketone-negative women (89 vs. 98, P < 0.001).

In the second study, urine was tested at any time in the 24 h before delivery (51). Again, the gestation at delivery was significantly shorter in case subjects compared with control subjects (38.2 vs. 39.7 weeks, P < 0.001). However, there was no significant difference in the mean birth weight. The IQs of children born to 45 ketone-negative women with diabetes were not significantly different compared with control subjects (100 vs. 101, P = NS). Children born to 62 ketone-positive women with diabetes had significantly lower IQs than children born to control subjects, again linking the presence of urinary ketones with low childhood IQ (93 vs. 102, P < 0.001).

Criticisms of the two studies by Churchill et al. (7,51) include the fact that ketone testing was not systematic. Women could have a negative result at the time of the test but have ketones at other times. No adjustments were made for maternal illness, reason for urine testing, or gestation at urine testing. In addition, case subjects consisted of women with DIP and control subjects consisted of women without DIP. The studies also did not control for nonnutritional factors known to affect IQ. As a result of these criticisms, the same data were reanalyzed by a different author using only women without DIP (9). After adjustment for nonnutritional factors known to affect childhood IQ, no association with maternal urinary ketones and childhood IQ was found. Since the above studies by Churchill et al. linking maternal ketones to low childhood IQ in women with DIP were conducted, multiple studies have evaluated the same association, and results continue to be conflicting.

Three studies have found an association between maternal ketones and low childhood IQ (5355). In a study of 76 women with pregestational and gestational diabetes (53), urinary ketones were measured every 6 h while women were hospitalized at any gestation. The reason for hospital admission was not clear and could have affected ketone production. Childhood IQ at 3 years was not significantly different between women with and without urinary ketones (92 vs. 99, P = NS). However, childhood IQ at 5 years of age was significantly lower if women had any urinary ketones in the pregnancy (92 vs. 103, P < 0.05). This difference remained after adjustment for birth weight. Other factors were not adjusted for.

The second study tested fasting serum BHB levels every 1–2 weeks in the second and third trimester of women with pregestational and gestational diabetes (54). Scores of IQ in the 139 offspring at ages 3–5 and 7–11 were inversely correlated with mean maternal serum BHB levels in the third trimester (age 3–5, r = −0.24, P < 0.01; age 7–11, r = −0.24, P < 0.01) Results were corrected for the mother’s IQ, parental education, socioeconomic status, and ethnicity.

The third study assessed 223 women, 89 with pregestational diabetes, 99 with GDM, and 35 with normal carbohydrate tolerance (55). Fasting serum BHB levels and urinary ketones were measured at each outpatient visit and twice weekly during hospitalizations in the pregnancy. Mean serum BHB levels in the third trimester of pregnancy were inversely correlated with the average of three childhood IQ scores performed at ages 3, 4, and 5 years (r = −0.2, P < 0.02), again with adjustment for the mother’s IQ, parental education, socioeconomic status, and ethnicity. Maternal urinary ketones were not found to be correlated with mental development or IQ scores. It should be noted that the strength of the relationship in the serum ketone studies was weak, with correlation coefficients of only −0.2.

Two other studies have found no association between maternal ketones and childhood IQ. The first investigated the association between ketoacidosis in pregnancy in women with pregestational or gestational diabetes and childhood IQ (56). The IQs of offspring from 37 mothers with ketoacidosis were compared with the IQs of offspring from 498 mothers without ketoacidosis. An initial association was found between ketoacidosis and low childhood IQ at 4 years of age; however, this association disappeared once evidence of amniotic fluid infection was accounted for. Amniotic fluid infection, or chorioamnionitis, has been associated with increased odds of cognitive impairment and neurodevelopmental delay in offspring (57). No other study of maternal ketones and childhood IQ has accounted for amniotic fluid infection.

The second study included 85 women, 64 with type 1 diabetes and 21 with GDM (58). Twelve-hour urine collections were performed every 1–2 weeks at outpatient visits and daily during hospitalization from 32 weeks’ gestation. The percentage of urine samples that were positive for ketones was not correlated with childhood IQ at 5 years of age (value not stated). This study did not account for factors known to affect IQ.

It is difficult to draw conclusions regarding the association between maternal ketones in women with DIP and childhood IQ based on existing studies. The studies vary greatly in their methodology. Some studies account for factors known to affect IQ, whereas others do not. In addition, it is likely that the exact timing of the exposure of the fetal brain to elevated ketone levels is important in terms of effect on IQ. Most of the studies do not contain detailed information regarding the exact gestation at which the exposure occurred. If Churchill’s studies are disregarded, given that no association was found with childhood IQ when the data were reanalyzed accounting for factors known to affect IQ, that leaves three studies showing an association and two studies showing no association. Of these studies, more weight should be given to those analyzing serum ketones, given that the biological link between ketones and IQ must occur through serum ketones. However, again, the studies of serum ketones found conflicting results, with two finding an association and one finding no association. The two reporting an association were systematic in the testing of ketones, testing them at multiple times throughout the pregnancy. In contrast, the one serum study that showed no association was of DKA in pregnancy, and it did not comment on the frequency of DKA. One could argue that more weight should be given to the two studies where the testing was systematic and examined a longitudinal presence of maternal ketones through pregnancy. It could also be argued that DKA is the situation where we would expect to see an impact on IQ if one existed given the severity of the event. On the other hand, a single episode of DKA in a pregnancy may not impact on childhood IQ, but elevated ketone levels over a sustained period might.

Given that ketones are potentially a necessary building block of the fetal brain, it is difficult to understand how elevated ketone levels could lead to reduced childhood IQ. Perhaps there is a continuum with a small amount of ketones being beneficial for fetal development but progressively increasing concentrations is harmful, leading to reduced IQ in the first instance and fetal mortality in the extreme case of DKA.

A summary of studies evaluating maternal ketones and childhood IQ is shown in Table 2.

Table 2

Articles assessing the relationship between maternal ketones in women with DIP and childhood IQ

Study geographical locationMethodologyKetones: serum vs. urinary. Urinary level considered positiveOutcomeAdjustment of factors known to affect IQ
Churchill et al. (7) 1969 Multicenter study, U.S. Retrospective study (N = 107). Women with DIP compared with women without DIP and no ketonuria. Urine tested in T3 except for 24 h predelivery. IQ (Binet) tested at age 4. Urine dipstick +1 (small) Offspring IQ in mothers with DIP and ketones < offspring IQ in mothers without DIP and ketones (controls). No difference in offspring IQ in DIP mothers without ketones and controls. Offspring IQ in mothers without DIP with ketones < controls. Nil 
Churchill et al. (51) 1969 Multicenter study, U.S. Retrospective study (N = 213). Women with DIP compared with women without DIP and no ketonuria. Urine tested anytime in the 24 h before delivery. IQ (Binet) tested at age 4. Urine dipstick +1 (small) Offspring IQ in mothers with DIP and ketones < offspring IQ in mothers without DIP and ketones (controls). No difference in offspring IQ in DIP mothers without ketones and controls. Nil 
Stehbens et al. (53) 1977 Iowa, U.S. Prospective study (N = 78). Women with DIP. Urine tested every 6 h during any hospitalization. IQ (Stanford-Binet) tested at age 3 and 5. Urine dipstick. Level not stated IQ at age 5 was lower in offspring of mothers with ketones. No difference in IQ at age 3. Birth weight 
Naeye et al. (56) 1979 Multicenter study, U.S. Prospective study (N = 53,518). Episodes of DKA in women with DIP. IQ (Stanford-Binet) tested at age 4. Multiple patient exclusions* Serum ketones No association between maternal ketoacidosis and childhood IQ at age 4. Amniotic fluid infection, SES, ethnicity, gestation at delivery 
Persson and Gentz (58) 1984 Stockholm, Sweden Prospective study (N = 85). Women with DIP. 12-h urine collection: 1–2 weekly until week 32 then daily. IQ (Terman-Merril) tested at age 5. Urine dipstick. Level not stated Percentage urine samples with ketones for each woman not correlated with childhood IQ at age 5. Nil 
Rizzo et al. (55) 1991 Chicago, U.S. Prospective study (N = 223). Women with and without DIP. Serum BHB at each outpatient visit and 2× weekly during hospitalizations in T2 and T3. Urine ketones in women with DIP 4×/day as outpatients and on each inpatient urine sample. IQ (Stanford-Binet) tested at ages 3, 4, and 5. Serum BHB and urinary dipstick (moderate) Mean T3 BHB levels inversely correlated with average childhood IQ at age 3, 4 and 5. No correlation between % days with positive urine ketones and IQ. Mother’s IQ, parental education, SES, ethnicity 
Silverman et al. (54) 1998 Chicago, U.S. Prospective study (N = 139). Women with DIP. Overnight fasting BHB 1–2 weekly in T2 and T3. IQ (Stanford-Binet) tested at age 3–5 and Wechsler Intelligence scale for children-revised at ages 7–11. Serum BHB Mean third trimester BHB levels were inversely correlated with IQ at age 3–5 and IQ at age 7–11. Mother’s IQ, parental education, SES, ethnicity 
Study geographical locationMethodologyKetones: serum vs. urinary. Urinary level considered positiveOutcomeAdjustment of factors known to affect IQ
Churchill et al. (7) 1969 Multicenter study, U.S. Retrospective study (N = 107). Women with DIP compared with women without DIP and no ketonuria. Urine tested in T3 except for 24 h predelivery. IQ (Binet) tested at age 4. Urine dipstick +1 (small) Offspring IQ in mothers with DIP and ketones < offspring IQ in mothers without DIP and ketones (controls). No difference in offspring IQ in DIP mothers without ketones and controls. Offspring IQ in mothers without DIP with ketones < controls. Nil 
Churchill et al. (51) 1969 Multicenter study, U.S. Retrospective study (N = 213). Women with DIP compared with women without DIP and no ketonuria. Urine tested anytime in the 24 h before delivery. IQ (Binet) tested at age 4. Urine dipstick +1 (small) Offspring IQ in mothers with DIP and ketones < offspring IQ in mothers without DIP and ketones (controls). No difference in offspring IQ in DIP mothers without ketones and controls. Nil 
Stehbens et al. (53) 1977 Iowa, U.S. Prospective study (N = 78). Women with DIP. Urine tested every 6 h during any hospitalization. IQ (Stanford-Binet) tested at age 3 and 5. Urine dipstick. Level not stated IQ at age 5 was lower in offspring of mothers with ketones. No difference in IQ at age 3. Birth weight 
Naeye et al. (56) 1979 Multicenter study, U.S. Prospective study (N = 53,518). Episodes of DKA in women with DIP. IQ (Stanford-Binet) tested at age 4. Multiple patient exclusions* Serum ketones No association between maternal ketoacidosis and childhood IQ at age 4. Amniotic fluid infection, SES, ethnicity, gestation at delivery 
Persson and Gentz (58) 1984 Stockholm, Sweden Prospective study (N = 85). Women with DIP. 12-h urine collection: 1–2 weekly until week 32 then daily. IQ (Terman-Merril) tested at age 5. Urine dipstick. Level not stated Percentage urine samples with ketones for each woman not correlated with childhood IQ at age 5. Nil 
Rizzo et al. (55) 1991 Chicago, U.S. Prospective study (N = 223). Women with and without DIP. Serum BHB at each outpatient visit and 2× weekly during hospitalizations in T2 and T3. Urine ketones in women with DIP 4×/day as outpatients and on each inpatient urine sample. IQ (Stanford-Binet) tested at ages 3, 4, and 5. Serum BHB and urinary dipstick (moderate) Mean T3 BHB levels inversely correlated with average childhood IQ at age 3, 4 and 5. No correlation between % days with positive urine ketones and IQ. Mother’s IQ, parental education, SES, ethnicity 
Silverman et al. (54) 1998 Chicago, U.S. Prospective study (N = 139). Women with DIP. Overnight fasting BHB 1–2 weekly in T2 and T3. IQ (Stanford-Binet) tested at age 3–5 and Wechsler Intelligence scale for children-revised at ages 7–11. Serum BHB Mean third trimester BHB levels were inversely correlated with IQ at age 3–5 and IQ at age 7–11. Mother’s IQ, parental education, SES, ethnicity 

SES, socioeconomic status; T1, first trimester; T2, second trimester; T3, third trimester.

*

Excluded children with Down syndrome, hydrocephalus, major congenital malformations, hypothyroidism, lead intoxication, fractured skull, inborn error of metabolism, and central nervous system infection. Excluded children of parents with intellectual disability and of mothers with alcohol or drug addictions or hypovolemic shock in labor.

No association has been reported between maternal serum ketone levels and fetal malformations. Serum BHB levels were tested in the first trimester of patients with type 1 diabetes and women without diabetes, and the rate of fetal malformations was not higher in women with higher BHB levels (59).

Two studies have found an association between urinary ketones at term and adverse in utero fetal outcomes. One study measured urinary ketones one time in 360 women without DIP at >40 weeks’ gestation (8). Women were excluded if they had vomiting, diarrhea, multiple pregnancy, history of renal disease, hypertension, or a high-risk pregnancy. An association was found between the presence of urinary ketones and oligohydramnios (17.6% vs. 2.5%, P = 0.01), fetal heart rate decelerations (29.4% vs. 5.8%, P < 0.001), and nonreactive nonstress tests (51.9% vs. 11% P < 0.01). Urine specific gravity was used as a marker of dehydration and did not differ between those with and those without ketones. The second study assessed maternal urinary ketones in 1,895 patients at gestations >41 weeks and found a significant correlation between the presence of urinary ketones and oligohydramnios (24% vs. 9.3%, P < 0.001), nonreactive nonstress tests (6.2% vs. 2.15%, P < 0.0001), and spontaneous fetal heart rate decelerations (14% vs. 9.2%, P = 0.0039). Women were excluded if they had diabetes, hypertension, renal disease, or another high-risk complication (60). Again, there was no significant difference in urine specific gravity in women with and without ketones. In contrast, an earlier study found no association between maternal ketones and adverse fetal outcomes in 59 women, although different outcomes were measured. Urinary ketones in the second and third trimester were not associated with fetal distress, asphyxia neonatorum, defined as the failure to start regular respirations within a minute of birth, or low 5-min Apgar score (48).

Making comparisons between these three studies is difficult because the outcome measures differed. The two studies that found an association involved markedly more women, so arguably, more weight should be given to these results. Both of these studies found an association between maternal ketonuria at term and specific measures of adverse in utero outcomes. Both studies attempted to account for maternal dehydration by testing urine specific gravity. However, studies in athletes have found that using urine specific gravity as a maker of dehydration is problematic (61). Further studies should be done evaluating the association between adverse in utero fetal outcomes and maternal serum ketone levels because these are not affected by dehydration.

Although multiple studies have shown an association between maternal ketones and adverse fetal and childhood outcomes, there is no clear evidence that maternal ketones directly cause these adverse outcomes. Ketones may instead be a marker of maternal pathology that causes both an adverse fetal environment and elevated ketone levels. Any pathology resulting in a decrease in glucose availability will lead to an increase in ketone levels. In addition, any pathology that leads to maternal dehydration will result in increased urinary ketone levels although not serum levels. Such pathologies could include infection, reduced oral intake, or vomiting and diarrhea. It is therefore possible that maternal ketones are a bystander rather than the cause of adverse fetal and childhood outcomes.

Conundrum Caused by Insufficient Evidence Regarding the Safety of Maternal Ketones

The current research on ketones in pregnancy leaves many unanswered questions that need to be addressed if we are to confidently advise our patients with DIP about safe carbohydrate intake. The current recommendation of 175 g carbohydrate per day does not appear to have a strong evidence base, and recent evidence suggests that a diet with 165 g carbohydrate per day does not lead to increased fasting ketone levels. None of the studies evaluating ketones in women with DIP and reduced childhood IQ commented on maternal carbohydrate intake. It therefore is difficult to determine not only whether elevated maternal ketone levels are harmful but also exactly what amount of carbohydrate intake in pregnancy is sufficient to prevent elevated ketone levels.

While it is difficult to draw strong conclusions on the safety of ketones and carbohydrate restriction in pregnancy, we believe that there is insufficient evidence to support current recommendations on necessary carbohydrate intake and avoidance of ketones. We propose that current recommendations for pregnant women to consume a minimum of 175 g carbohydrate per day and to consume a diet that does not result in ketones be reviewed. There is insufficient evidence to support either of these recommendations, and they have the potential to limit the ability of women with DIP to restrict their carbohydrate intake in an effort to control blood glucose levels. In addition, given the high prevalence of maternal ketones, these recommendations have the potential to cause unnecessary anxiety among pregnant women.

To further clarify the safety of carbohydrate restriction in DIP and of elevated ketones in pregnancy, many questions need to be answered. A table of questions has been generated in Table 3.

Table 3

Unanswered questions and proposed research

Unanswered questionProposed research
What amount of carbohydrate intake in pregnancy is required to prevent elevated fasting ketone levels? Interventional studies in human pregnancies measuring fasting BHB in women consuming diets varying in carbohydrate content. 
What is the prevalence of elevated serum ketones in pregnancy, both in women with and without DIP? Observational studies in humans with systematic collection of serum BHB levels throughout pregnancy. 
Is there an association between maternal serum ketones at term and in utero markers of fetal distress? Observational study in human pregnancies evaluating for an association between serum BHB and markers of fetal distress 
Is there an association between maternal serum ketones and childhood IQ? Further large-scale prospective studies in human pregnancies evaluating for association between serum BHB collected systematically throughout all trimesters and childhood IQ at multiple ages. 
What is the correlation between maternal urinary and serum ketones. If it is poor, should we cease measurement of urinary ketones as a marker of ketogenesis? Observational study of human pregnancy assessing for correlation between fasting serum BHB and urinary ketone levels. 
Does BHB inhibit de novo synthesis of pyrimidines and purines in human fetal brains? Replication of existing work using a primate model. 
Are ketones a necessary fuel for the fetal brain and a necessary component of brain development? Consideration of using a primate model to investigate the role of ketones in fetal and newborn brain development further. 
Unanswered questionProposed research
What amount of carbohydrate intake in pregnancy is required to prevent elevated fasting ketone levels? Interventional studies in human pregnancies measuring fasting BHB in women consuming diets varying in carbohydrate content. 
What is the prevalence of elevated serum ketones in pregnancy, both in women with and without DIP? Observational studies in humans with systematic collection of serum BHB levels throughout pregnancy. 
Is there an association between maternal serum ketones at term and in utero markers of fetal distress? Observational study in human pregnancies evaluating for an association between serum BHB and markers of fetal distress 
Is there an association between maternal serum ketones and childhood IQ? Further large-scale prospective studies in human pregnancies evaluating for association between serum BHB collected systematically throughout all trimesters and childhood IQ at multiple ages. 
What is the correlation between maternal urinary and serum ketones. If it is poor, should we cease measurement of urinary ketones as a marker of ketogenesis? Observational study of human pregnancy assessing for correlation between fasting serum BHB and urinary ketone levels. 
Does BHB inhibit de novo synthesis of pyrimidines and purines in human fetal brains? Replication of existing work using a primate model. 
Are ketones a necessary fuel for the fetal brain and a necessary component of brain development? Consideration of using a primate model to investigate the role of ketones in fetal and newborn brain development further. 

Summary

Elevated ketones are a key concern with low carbohydrate diets and are cited as a reason to avoid such diets during pregnancy. Given that dietary intervention is the primary treatment for GDM and that there are also potentially large numbers of women with DIP following a low-carbohydrate diet, it is important to carefully examine what is actually known about the impact of increased maternal ketone levels. It is well known that ketogenesis is more pronounced in pregnancy due to changes in maternal metabolism. However, data on the prevalence of urinary and serum ketones in pregnancy, including in women with DIP, are lacking. Therefore, at present, there is a limited understanding of “normal” ketone levels in pregnancy, either affected by diabetes or not.

There is evidence of fetal harm in DKA, which may be related to high levels of ketones but may also be related to the causes and consequences of this highly pathological state. Outside of DKA, studies in women with DIP evaluating the association between elevated maternal ketone levels and childhood IQ are conflicting. Several studies show a correlation between maternal ketones and reduced childhood IQ; however, many of these studies do not account for factors known to affect childhood IQ, such as socioeconomic status, parental education, parental IQ, and neurological and metabolic abnormalities in the children. In addition, most of these studies have based their findings on measured urinary ketones. Given that the correlation between urinary and serum maternal ketone levels in women with GDM has been shown to be poor, it is difficult to interpret the validity of this correlation.

Further studies are urgently required to explore the impact of maternal ketones on pregnancy and fetal outcomes. Areas that need to be studied include a prospective examination of serum ketone levels throughout pregnancy, both in women with and without DIP. Long-term, methodologically rigorous prospective observational studies are required to examine the relationship between maternal serum ketone levels, adverse fetal outcomes, and childhood IQ. This work is essential so the public health implications of low-carbohydrate diets in women with and without DIP can be better understood and accurate advice can be provided. Women need definitive evidence-based advice regarding the safety of low-carbohydrate diets in pregnancy. And finally, clinicians need high quality research to answer the question: “Do ketones really matter?”

Funding. M.D.N. receives a salary from the University of Queensland for research and teaching. L.C. receives a salary from the Royal Brisbane and Women’s Hospital in her position as the Director of Research for Women’s and Newborn Services. H.B. receives a research scholarship from the National Health and Medical Research Council of Australia and is also supported by the Mater Research Foundation.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. H.L.T. wrote the manuscript. M.D.N. reviewed and edited the manuscript. L.K.C. reviewed and edited the manuscript. H.L.B. reviewed and edited the manuscript.

1.
Franz
MJ
,
Bantle
JP
,
Beebe
CA
, et al
.
Evidence-based nutrition principles and recommendations for the treatment and prevention of diabetes and related complications
.
Diabetes Care
2002
;
25
:
148
198
2.
American Diabetes Association
;
Bantle
JP
,
Wylie-Rosett
J
,
Albright
AL
, et al
.
Nutrition recommendations and interventions for diabetes: a position statement of the American Diabetes Association
[published correction appears in Diabetes Care 2010;33:1911]
.
Diabetes Care
2008
;
31
(
Suppl. 1
):
S61
S78
3.
American Diabetes Association
.
13. Management of diabetes in pregnancy: Standards of Medical Care in Diabetes—2018
.
Diabetes Care
2018
;
41
(
Suppl. 1
):
S137
S143
4.
Metzger
BE
,
Ravnikar
V
,
Vileisis
RA
,
Freinkel
N
.
“Accelerated starvation” and the skipped breakfast in late normal pregnancy
.
Lancet
1982
;
1
:
588
592
5.
Herrera
E
.
Lipid metabolism in pregnancy and its consequences in the fetus and newborn
.
Endocrine
2002
;
19
:
43
55
6.
Newbern
D
,
Freemark
M
.
Placental hormones and the control of maternal metabolism and fetal growth
.
Curr Opin Endocrinol Diabetes Obes
2011
;
18
:
409
416
7.
Churchill
JA
,
Berendes
HW
.
Intelligence of children whose mothers had acetonuria during pregnancy
. In
Perinatal Factors Affecting Human Development, proceedings of the special session held during the Eighth Meeting of the PAHO Advisory Committee on Medical Research, 10 June 1969, Scientific Publication 185, p. 30–35
8.
Hamdi
K
,
Bastani
P
,
Gafarieh
R
,
Mozafari
H
,
Hashemi
SH
,
Ghotbi
MH
.
The influence of maternal ketonuria on fetal well-being tests in postterm pregnancy
.
Arch Iran Med
2006
;
9
:
144
147
9.
Naeye
RL
,
Chez
RA
.
Effects of maternal acetonuria and low pregnancy weight gain on children’s psychomotor development
.
Am J Obstet Gynecol
1981
;
139
:
189
193
10.
Gannon
MC
,
Nuttall
FQ
.
Effect of a high-protein, low-carbohydrate diet on blood glucose control in people with type 2 diabetes
.
Diabetes
2004
;
53
:
2375
2382
11.
Kirk
JK
,
Graves
DE
,
Craven
TE
,
Lipkin
EW
,
Austin
M
,
Margolis
KL
.
Restricted-carbohydrate diets in patients with type 2 diabetes: a meta-analysis
.
J Am Diet Assoc
2008
;
108
:
91
100
12.
Major
CA
,
Henry
MJ
,
De Veciana
M
,
Morgan
MA
.
The effects of carbohydrate restriction in patients with diet-controlled gestational diabetes
.
Obstet Gynecol
1998
;
91
:
600
604
13.
Moreno-Castilla
C
,
Hernandez
M
,
Bergua
M
, et al
.
Low-carbohydrate diet for the treatment of gestational diabetes mellitus: a randomized controlled trial
.
Diabetes Care
2013
;
36
:
2233
2238
14.
Farabi
SS
,
Hernandez
TL
.
Low-carbohydrate diets for gestational diabetes
.
Nutrients
2019
;
11
:
1737
15.
Han
S
,
Crowther
CA
,
Middleton
P
,
Heatley
E
.
Different types of dietary advice for women with gestational diabetes mellitus
.
Cochrane Database Syst Rev
2013
:
CD009275
16.
Ásbjörnsdóttir
B
,
Ronneby
H
,
Vestgaard
M
, et al
.
Lower daily carbohydrate consumption than recommended by the Institute of Medicine is common among women with type 2 diabetes in early pregnancy in Denmark
.
Diabetes Res Clin Pract
2019
;
152
:
88
95
17.
Frayn
KN
,
Arner
P
,
Yki-Järvinen
H
.
Fatty acid metabolism in adipose tissue, muscle and liver in health and disease
.
Essays Biochem
2006
;
42
:
89
103
18.
Balasse
EO
,
Féry
F
.
Ketone body production and disposal: effects of fasting, diabetes, and exercise
.
Diabetes Metab Rev
1989
;
5
:
247
270
19.
Laffel
L
.
Ketone bodies: a review of physiology, pathophysiology and application of monitoring to diabetes
.
Diabetes Metab Res Rev
1999
;
15
:
412
426
20.
Taboulet
P
,
Deconinck
N
,
Thurel
A
, et al
.
Correlation between urine ketones (acetoacetate) and capillary blood ketones (3-beta-hydroxybutyrate) in hyperglycaemic patients
.
Diabetes Metab
2007
;
33
:
135
139
21.
Walsh
CO
,
Ebbeling
CB
,
Swain
JF
,
Markowitz
RL
,
Feldman
HA
,
Ludwig
DS
.
Effects of diet composition on postprandial energy availability during weight loss maintenance
.
PLoS One
2013
;
8
:
e58172
22.
Adam-Perrot
A
,
Clifton
P
,
Brouns
F
.
Low-carbohydrate diets: nutritional and physiological aspects
.
Obes Rev
2006
;
7
:
49
58
23.
VanItallie
TB
,
Nufert
TH
.
Ketones: metabolism’s ugly duckling
.
Nutr Rev
2003
;
61
:
327
341
24.
Mijatovic
J
,
Louie
JCY
,
Buso
MEC
, et al
.
Effects of a modestly lower carbohydrate diet in gestational diabetes: a randomized controlled trial
.
Am J Clin Nutr
2020
;
112
:
284
292
25.
Potter
JM
,
Reckless
JP
,
Cullen
DR
.
Diurnal variations in blood intermediary metabolites in mild gestational diabetic patients and the effect of a carbohydrate-restricted diet
.
Diabetologia
1982
;
22
:
68
72
26.
Rudolf
MC
,
Sherwin
RS
.
Maternal ketosis and its effects on the fetus
.
Clin Endocrinol Metab
1983
;
12
:
413
428
27.
Silverstone
FA
,
Solomons
E
,
Rubricius
J
.
The rapid intravenous glucose tolerance test in pregnancy
.
J Clin Invest
1961
;
40
:
2180
2189
28.
Victor
A
.
Normal blood sugar variation during pregnancy
.
Acta Obstet Gynecol Scand
1974
;
53
:
37
40
29.
Riskin-Mashiah
S
,
Damti
A
,
Younes
G
,
Auslander
R
.
Normal fasting plasma glucose levels during pregnancy: a hospital-based study
.
J Perinat Med
2011
;
39
:
209
211
30.
Phelps
RL
,
Metzger
BE
,
Freinkel
N
.
Carbohydrate metabolism in pregnancy. XVII. Diurnal profiles of plasma glucose, insulin, free fatty acids, triglycerides, cholesterol, and individual amino acids in late normal pregnancy
.
Am J Obstet Gynecol
1981
;
140
:
730
736
31.
Kalhan
SC
,
D’Angelo
LJ
,
Savin
SM
,
Adam
PA
.
Glucose production in pregnant women at term gestation. Sources of glucose for human fetus
.
J Clin Invest
1979
;
63
:
388
394
32.
Robinson
AM
,
Williamson
DH
.
Physiological roles of ketone bodies as substrates and signals in mammalian tissues
.
Physiol Rev
1980
;
60
:
143
187
33.
Sabata
V
,
Wolf
H
,
Lausmann
S
.
The role of free fatty acids, glycerol, ketone bodies and glucose in the energy metabolism of the mother and fetus during delivery
.
Biol Neonat
1968
;
13
:
7
17
34.
Hosoya
N
,
Hagerman
D
,
Villee
C
.
Stimulation of fatty acid synthesis by oestradiol in vitro
.
Biochem J
1960
;
76
:
297
301
35.
Shambaugh
GE
 III
,
Koehler
RA
,
Freinkel
N
.
Fetal fuels II: contributions of selected carbon fuels to oxidative metabolism in rat conceptus
.
Am J Physiol
1977
;
233
:
E457
E461
36.
Patel
MS
,
Johnson
CA
,
Rajan
R
,
Owen
OE
.
The metabolism of ketone bodies in developing human brain: development of ketone-body-utilizing enzymes and ketone bodies as precursors for lipid synthesis
.
J Neurochem
1975
;
25
:
905
908
37.
Adam
PA
,
Räihä
N
,
Rahiala
EL
,
Kekomäki
M
.
Oxidation of glucose and D-B-OH-butyrate by the early human fetal brain
.
Acta Paediatr Scand
1975
;
64
:
17
24
38.
Bhasin
S
,
Shambaugh
GE
 III
.
Fetal fuels. V. Ketone bodies inhibit pyrimidine biosynthesis in fetal rat brain
.
Am J Physiol
1982
;
243
:
E234
E239
39.
Shambaugh
GE
 III
,
Angulo
MC
,
Koehler
RR
.
Fetal fuels. VII. Ketone bodies inhibit synthesis of purines in fetal rat brain
.
Am J Physiol
1984
;
247
:
E111
E117
40.
Sussman
D
,
Ellegood
J
,
Henkelman
M
.
A gestational ketogenic diet alters maternal metabolic status as well as offspring physiological growth and brain structure in the neonatal mouse
.
BMC Pregnancy Childbirth
2013
;
13
:
198
41.
Sussman
D
,
Germann
J
,
Henkelman
M
.
Gestational ketogenic diet programs brain structure and susceptibility to depression & anxiety in the adult mouse offspring
.
Brain Behav
2015
;
5
:
e00300
42.
Meas
T
,
Taboulet
P
,
Sobngwi
E
,
Gautier
JF
.
Is capillary ketone determination useful in clinical practice? In which circumstances
?
Diabetes Metab
2005
;
31
:
299
303
43.
Spanou
L
,
Dalakleidi
K
,
Zarkogianni
K
, et al
.
Ketonemia and ketonuria in gestational diabetes mellitus
.
Hormones (Athens)
2015
;
14
:
644
650
44.
Wildenhoff
KE
.
The renal excretion of acetoacetate and 3-hydroxybutyrate during absolute fasting
.
Acta Med Scand
1972
;
192
:
475
479
45.
Sapir
DG
,
Owen
OE
.
Renal conservation of ketone bodies during starvation
.
Metabolism
1975
;
24
:
23
33
46.
Chez
RA
,
Curcio
FD
 III
.
Ketonuria in normal pregnancy
.
Obstet Gynecol
1987
;
69
:
272
274
47.
Sibai
BM
,
Viteri
OA
.
Diabetic ketoacidosis in pregnancy
.
Obstet Gynecol
2014
;
123
:
167
178
48.
Coetzee
EJ
,
Jackson
WP
,
Berman
PA
.
Ketonuria in pregnancy--with special reference to calorie-restricted food intake in obese diabetics
.
Diabetes
1980
;
29
:
177
181
49.
Schneider
MB
,
Umpierrez
GE
,
Ramsey
RD
,
Mabie
WC
,
Bennett
KA
.
Pregnancy complicated by diabetic ketoacidosis: maternal and fetal outcomes
.
Diabetes Care
2003
;
26
:
958
959
50.
Hughes
AB
.
Fetal heart rate changes during diabetic ketosis
.
Acta Obstet Gynecol Scand
1987
;
66
:
71
73
51.
Churchill
JA
,
Berendes
HW
,
Nemore
J
.
Neuropsychological deficits in children of diabetic mothers. A report from the Collaborative Study of Cerebral Palsy
.
Am J Obstet Gynecol
1969
;
105
:
257
268
52.
White
P
,
Gillespie
L
,
Sexton
L
.
Use of female sex hormone therapy in pregnant diabetic patients
.
Am J Obstet Gynecol
1956
;
71
:
57
69
53.
Stehbens
JA
,
Baker
GL
,
Kitchell
M
.
Outcome at ages 1, 3, and 5 years of children born to diabetic women
.
Am J Obstet Gynecol
1977
;
127
:
408
413
54.
Silverman
BL
,
Rizzo
TA
,
Cho
NH
,
Metzger
BE
;
The Northwestern University Diabetes in Pregnancy Center
.
Long-term effects of the intrauterine environment
.
Diabetes Care
1998
;
21
(
Suppl. 2
):
B142
B149
55.
Rizzo
T
,
Metzger
BE
,
Burns
WJ
,
Burns
K
.
Correlations between antepartum maternal metabolism and intelligence of offspring
.
N Engl J Med
1991
;
325
:
911
916
56.
Naeye
R
.
Pregnancy Metabolism, Diabetes and the Fetus
.
The Netherlands
,
Excerpta Medica
,
1979
57.
Pappas
A
,
Kendrick
DE
,
Shankaran
S
, et al.;
Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network
.
Chorioamnionitis and early childhood outcomes among extremely low-gestational-age neonates
.
JAMA Pediatr
2014
;
168
:
137
147
58.
Persson
B
,
Gentz
J
.
Follow-up of children of insulin-dependent and gestational diabetic mothers. Neuropsychological outcome
.
Acta Paediatr Scand
1984
;
73
:
349
358
59.
Jovanovic
L
,
Metzger
BE
,
Knopp
RH
, et al
.
The Diabetes in Early Pregnancy Study: beta-hydroxybutyrate levels in type 1 diabetic pregnancy compared with normal pregnancy. NICHD-Diabetes in Early Pregnancy Study Group (DIEP). National Institute of Child Health and Development
.
Diabetes Care
1998
;
21
:
1978
1984
60.
Onyeije
CI
,
Divon
MY
.
The impact of maternal ketonuria on fetal test results in the setting of postterm pregnancy
.
Am J Obstet Gynecol
2001
;
184
:
713
718
61.
Zubac
D
,
Reale
R
,
Karnincic
H
,
Sivric
A
,
Jelaska
I
.
Urine specific gravity as an indicator of dehydration in Olympic combat sport athletes; considerations for research and practice
.
Eur J Sport Sci
2018
;
18
:
920
929
62.
Robinson
HL
,
Barrett
HL
,
Foxcroft
K
,
Callaway
LK
,
Dekker Nitert
M
.
Prevalence of maternal urinary ketones in pregnancy in overweight and obese women
.
Obstet Med
2018
;
11
:
79
82
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at https://www.diabetesjournals.org/content/license.