Can low carb or keto help with gestational diabetes in pregnancy?
When Natalie Thompson Cooper was diagnosed with gestational diabetes in her first pregnancy, at age 28, she was very concerned. The condition, which affects at least one in seven pregnancies to as many as one in five, causes blood sugars to rise abnormally high, called hyperglycemia.1
Natalie knew hyperglycemia exposed her body tissues and her unborn daughter in glucose, putting both her and her fetus at risk for a wide range of potential complications, including miscarriage, birth defects, macrosomia (very large size), high blood pressure, pre-eclampsia, seizures, birth trauma, and higher rates of C-section and even stillbirth.2
Moreover, gestational diabetes (GD) — also called ‘carbohydrate intolerance of pregnancy’ — greatly increases the risk that the mother and her offspring will both face future health problems, such as much higher rates of eventual type 2 diabetes, metabolic conditions, and cardiovascular disease.3
Gestational diabetes is one of the most common and significant complications of pregnancy. Prenatal guidelines the world over recommend the routine screening of all pregnant women and then, if positive, strict management, starting with dietary therapy, then if that does not work, insulin injections.4
However, to this day, what constitutes the best “dietary therapy” is hotly debated, with some researchers proposing a diet high in complex carbohydrates (60% carbs) and others lower carbohydrates (40% carbs).5
However, the recommended “lower carb” GD diet is still far higher than the under 20 g per day of the strict low-carb high-fat or ketogenic diet. In fact, many guidelines for GD recommend women, on an ostensibly “lower-carb” diet, eat a minimum 175 g of carbohydrate daily, a level at which many women see their blood sugar rise out of control.6
Natalie Cooper Thompson is a case in point.8
Natalie, an elementary school teacher in Buford, Georgia, was told after the diagnosis of GD in her first pregnancy to eat carbohydrates at every meal and at three snacks a day, getting a daily carbohydrate requirement of at least 175 g — enough so that no ketones would be found in her urine.
She ate what they advised — bananas, whole grain toast, fruit, rice, steel cut oats — and was checking her blood sugar with a home monitor seven times a day. The results were always too high. Soon she was on four insulin injections a day, but still the GD was uncontrolled. She developed preeclampsia (high blood pressure and protein in her urine). Her daughter was born by C-section at 37 weeks, weighing 8 lbs (4 kg). “The diet did not help me AT ALL!”
Finding the ketogenic diet
Six years later — after failed fertility treatments for her PCOS, a miscarriage, and a successful adoption — Natalie discovered ketogenic eating. She lost 40 lbs (18 kg) in three months, which brought her weight down to 210 lbs (95 kg), helping her infertility and her PCOS symptoms. Much to her surprise, after six years of infertility, she was suddenly pregnant.
In the second pregnancy she developed GD, again, at 18 weeks. Her doctors and dietitian gave her exactly the same advice — carbs at every meal and at three snacks a day to a minimum daily requirement of 175 g. This time Natalie just smiled and nodded but kept her carbs below 50 g and monitored her blood sugar regularly with a home meter. She eventually needed insulin, but just one injection at bedtime. “It wasn’t perfect, but I was much more successful at keeping my blood sugars in check. There was no way I could eat the amount of carbs they recommended. If I did, my blood sugars would be way too high.”
Now, age 36, she is happily pregnant again — 24 weeks9 — with her third pregnancy. She has continued keto eating these past two years, which may have helped resolve her infertility, resulted in a total weight loss of 95 lbs (43 kg) and a healthy pre-pregnancy weight of 155 lbs (70 kg). In June, at 18 weeks, she had a glucose tolerance test again to screen for GD. She drank a sickly sweet drink with 50 g of sugar, and had her blood glucose checked one hour later. This time she passed with flying colours: her blood sugar was 85 mg/dl (4.7 mmol/l). “That is about as perfect as you can get!”
Will she still develop GD later in this pregnancy? She doesn’t know, but she will do her utmost not to up her carb consumption and will monitor her blood sugars regularly. She has set the goal of under 50 g of carbs a day, and hopes to get to 25 g if possible. “I cannot believe that 175 g of carbs is still being recommended for GD! It is insane.”10
Is the current advice based on solid science?
Dr. Jovanovic and U.S. dietitian Lily Nichols both agree. Jovanovic has been consulting in the last few years around the US, telling endocrinologists and ob/gyn doctors that women can safely cut carbs to the level that keeps their blood glucose tightly controlled (see below) even if that means no carbs at all. She says to doctors looking after women with GD: “Repeat after me: pregnant women do not need to eat carbohydrates!”11
Nichols has specialized in gestational diabetes for most of her career, particularly with a California GD program called “Sweet Success”, which also recommended the minimum 175 g of carbs daily for GD. “It was always really disappointing to see how many of my prenatal clients would end up, as we say, ‘failing diet therapy’ and would have to go on insulin. It didn’t make sense to me why somebody with gestational diabetes, which translates as carbohydrate intolerance, would need such a high level of carbs,” said Nichols.
That experience set Nichols off digging into the historical research literature to find out why 175 g a day had become the arbitrary measure for lower-carb GD dietary management. “It turns out that this number is a best guess and not based in solid science. There is no evidence that eating fewer than 175 g of carbohydrates is harmful. The only reason medical professionals continually push a higher-carb diet is due to unfounded fears around ketosis.”12
As detailed in an earlier post: Is low carb safe in pregnancy? good scientific studies around low-carb ketogenic eating in pregnancy are sorely lacking. However, ketones in the urine of pregnant women with GD “freak doctors out,” says Nichols, because they fear the life-threatening conditions of diabetic ketoacidosis in pregnancy or starvation ketosis. “These are not the same as nutritional ketosis at all,” says Nichols whose book has an entire chapter on the misconceptions surrounding ketosis in pregnancy. Dr. Jovanovic agrees: “Put the ketone stix away. Ketones in urine do not mean anything.”13
Both women, as well as reproductive expert Dr. Michael Fox, note that nutritional ketosis in pregnancy is a natural state, which is normal and safe, and likely beneficial to general fetal growth as well as fetal nerve and brain growth.14
Dr. Fox notes that in obstetrics training, doctors are taught that, most commonly, ketones in the urine would represent starvation ketosis. This is often seen in hyperemesis patients (severe nausea and vomiting) and in patients who are not eating enough calories during pregnancy.
As long as pregnant women are eating frequently (at least every 3 hours) and consuming plenty of calories (about 2500 calories or more per day), ketones in the urine or blood is not an abnormal state and it is expected with a ketogenic approach.15
While pregnant women might get push-back from doctors not yet familiar with nutritional ketosis, Dr. Fox suggests one way to educate the medical team would be to say “I am following a ketogenic nutritional approach and ketones are expected in the urine. Unless you believe that I am sick, I will continue my diet.” Then recommend that the physician visit Diet Doctor.
In short, all three experts say, pregnant women with GD can reduce their carbohydrates and not fear ketones in their urine as long as their blood sugars remain in normal ranges and they are eating enough calories from nutrient dense foods for themselves and their babies. (Note: High blood sugar combined with very high blood ketones is ketoacidosis and a medical emergency. Diabetic ketoacidosis cannot be diagnosed with urine ketones; it is only found in blood ketones and generally found among women with type 1 diabetes.)
In extensive interviews for this post, these experts offered four practical tips for managing the risk of GD for all pregnant women, especially those at higher risk of GD, such as women with polycystic ovarian syndrome (PCOS), pre-diabetes, obesity, and previous GD in pregnancy.
Tip #1 Have early blood sugar tests to assess your risk
Women who go into pregnancy with pre-existing insulin resistance, or undiagnosed pre-diabetes, are much more likely to develop GD.16 “A lot of women unknowingly have blood sugar issues, even before pregnancy. And the physiology of pregnancy naturally increases insulin resistance in all women. By the 10th week of pregnancy, insulin levels are up to 3 times higher than they were pre-pregnancy.”
An HbA1c test — which measures the average blood sugar levels in the previous three months — taken prior to pregnancy or early in the first trimester can help identify women at risk in advance. Values greater than 5.7% indicate prediabetes and many doctors treat this the same as gestational diabetes. The higher the number, the higher your average blood sugar. A level of 5.9% or higher accurately predicates gestational diabetes 98.4% of the time.17
A fasting blood glucose prior to pregnancy or early in the first trimester can also help flag potential problems, if values are higher than 5.6 mmol/l (100 mg/dl), which is suggestive of prediabetes. Both women note that, depending on your result, you may need to make blood sugar control a bigger priority. “Tackling it up front can prevent blood sugar issues in the pregnancy and future complications,” Nichols said. Jovanovic agrees: “Waiting until 24 to 28 weeks to diagnosis GD means the fetus may have many weeks earlier in the pregnancy exposed to high blood sugars.18 You want to be on top of it early.” Note: HbA1C tests later in the pregnancy are less reliable because of rapid blood cell turnover and low iron stores.19
Tip #2 Self monitor with a home glucose monitor
Women at higher risk of GD, with a confirmed GD diagnosis, or who are interested in knowing about how their pregnancy and their diet is impacting their blood sugar, should consider getting a home glucose monitor. Times of the day to test are: a fasting blood glucose first thing in the morning and one to two hours after every meal or snack.
Both note that if you have a couple of weeks showing normal blood sugar, you can keep eating the way you are eating. Just be aware that if your readings looked great in early pregnancy, you need to check again for a few weeks in the 24-28 week window since insulin resistance goes up in later pregnancy.21 Some people like Natalie Thompson Cooper have chosen to self-monitor their blood sugar regularly through their entire pregnancy.
Many types of home glucose monitors are available from your local pharmacy. Follow the instructions that come for your particular meter. In general, wash your hands with soap and water and dry well before testing. Do not use alcohol wipes as this may alter the result. Also so not touch any food before testing, as this can also raise the readings.
- Fasting blood glucose in the upper range limit of 90-99 mg/dl (5.0 to 5.5 mmol/l); under 90 mg/dl is optimum.
- 1 hour post meal (postprandial) blood glucose less than 140 mg/dl (7.8 mmol/l); under 100 mg/dl is optimum.
- 2 hour post meal blood glucose less than 120 mg/dl (6.7 mmol/l); back down to 90 mg/dl is optimum.
- Avoid maintaining blood glucose levels too low (below 4.5 mg/dl) as there is an increase likelihood of small for gestational age infants; you may not be giving your baby enough nutrition (although it is unclear how ketosis would affect this recommendation in pregnancy).
Keep track of your daily blood glucose readings, the food you ate, and the exercise you did, either in a notebook or through one of the online trackers. The free SugarStats tracker is very easy to use.
Tip #3 Understand the options for screening and diagnosis
Screening for GD typically happens in the first trimester for those at high risk (women with previous GD or PCOS) and at 24-28 weeks for all other pregnant women. The first test is typically the glucose challenge test: chugging a syrupy sweet drink with 50 mg of sugar and having blood glucose tested one hour later.
Nichols blogged about her own experience with the test during her pregnancy in 2015 and how she failed it by 1 point, reading 141 mg/dl (7.9 mmol/l) rather than the cut off of 140 mg/dl (7.8 mmol/l). Her failure was likely because she was eating a low-carbohydrate diet prior to the test. Her home glucose monitoring revealed perfectly normal blood sugar levels and ruled out a gestational diabetes diagnosis.
False positives in the glucose challenge test are common, particularly among women who eat a low-glycemic, low-carb or ketogenic diet, because their bodies are not accustomed to handling that amount of sugar at one time.23 Many women find drinking the sugary liquid very difficult, causing nausea, vomiting, dizziness, and headache. Studies have found eating 50 g of sugar in 28 jelly beans or 10 sticks of a very popular (in North America) strawberry-flavoured licorice candy, called “Twizzlers” may be easier to tolerate, with fewer side effects, but still provide the same results.24
Women who fail the glucose challenge screening typically must then have the three-hour Oral Glucose Tolerance Test (OGTT) to confirm GD diagnosis. That entails drinking twice the amount of sugar (100 g) and having blood tested hourly over the next three hours.25
Nichols notes in her blog about failing the test that the two-step screening method for gestational diabetes is outdated because a fairly high percentage of healthy women “fail” the first test while some women with excessive insulin production “pass” it and are never formally diagnosed.
“That’s why the International Association of Diabetes and Pregnancy Study Group (IADPSG), the World Health Organization (WHO), and nearly all countries aside from the United States recommend the more reliable and specific 2-hour, 75 g glucose tolerance test, which is done fasting and includes more rigid cut-offs for diagnosis,” she says. However, the 75 g fasted test is not yet the standard test in all countries, and it may lead more women to be diagnosed with GD (likely because it catches women with “mild” gestational diabetes).26
Nichols and Jovanovic note that women can forego the tests if they instead choose to track their blood sugar at home. If you show your doctor the results of your readings, you can usually convince them to forego other tests. “You can always decline the tests. You have that right. But you should be tracking your own sugars,” says Nichols. Jovanovic advises: “Just show them your home blood sugars. That is all that matters.”27
Some women eating low-carb/ high-fat or ketogenic diets decide to refuse the standard GD tests altogether. Katie Bravie of Hobart Tasmania did just that. Now 32 weeks pregnant, Bravie, 26, lost 23 kg (52 lbs) over the previous two years by following the ketogenic diet. She is loathe to add any sugar, even in a test, back into her body. Her midwife demanded she take the screening test and told her there were no other options.
“I went to my family doctor for backup, who knew my journey with my weight loss and keto. She said drinking that amount of sugar would make me sick after two years of no sugar and it would result in a false positive.”
Instead the doctor did an HbA1c test as well as a test called a fructosamine blood test, which is another way to test for evidence of high blood sugar. Her results were fine. She is staying the course on her keto eating during pregnancy. “I love having a keto pregnancy,” says Katie, who says her midwife continues to be “freaked out” by the presence of ketones in her urine. “But I just secretly laugh it off. I know I am fine.”28
Tip #4 Eat a nutrient-dense “whole-food” diet
The cover of Nichols book features two fried eggs with sliced avocado and cherry tomatoes — a low-carb high-fat meal. “It is the perfect breakfast, or any meal, for gestational diabetes,” she says. Her advice: eat nutrient-dense foods over empty calories.
“Some low-carb people are just living off fat bombs and bulletproof coffee. You need to have good, common sense with variety in your diet.” She recommends having protein like meat, fish and eggs, snacks like nuts and seeds, lots of vegetables with healthy fat like olive oil or butter. Try to eat liver once or twice a week, to help ensure a good natural intake of vitamins, minerals and antioxidants in your diet.
She further recommends avoiding all highly processed foods, added sugar, refined carbs, as well as high-glycemic fruit and fruit juices. If your blood sugar is still high, consider cutting out all fruit except berries. “You do not need to add in bread and cereals to have a healthy pregnancy,” says Nichols. Pay attention to how you feel, and monitor your blood sugars. “Some pregnant women may feel better eating slightly more carbs, but I think the majority of women with GD, and all pregnant women in general, do better with at least one-half to one-quarter the carbohydrates recommended in conventional nutrition guidelines (which is upwards of 65% of calories from carbohydrates).29
Are gestational diabetes programs around the globe now registering that message and advising lower carb? Alas, not yet.
“What do you think would happen if I followed their advice?” said Sarah incredulously, emailing the advice she received to Andreas and the Diet Doctor Team. She believes she has been likely insulin resistant and prediabetic for years.
She found the low-carb high-fat diet, and Diet Doctor, five years ago. During this pregnancy, her third, she has been eating a more liberal low-carb diet, consuming more fruit than normal.
But now with the GD diagnosis, she is sticking to a lower-carb diet and monitoring her blood sugars regularly by a home meter, and doing well. “Home monitoring is brilliant. I love doing it. I can see the impact of the foods I eat immediately,” says the math teacher. “I know if I followed the NHS dietary advice my blood sugar would get much, much worse.”
Medical comment from Dr. Michael D. Fox30
In obstetrics training, we are taught that ketones are pathologic (abnormal) in pregnancy. Most commonly ketones in the urine would represent starvation ketosis. This is often seen in hyperemesis patients (severe nausea vomiting) and in patients who are not eating enough calories during pregnancy.
Most training for OB/GYN’s happens in inner city hospitals, where there is a high prevalence of drug users in pregnancy. This group is notorious for starvation and subsequent complications in the child including early delivery, growth retardation and low birth weight. Doctors fear these complications and are going to react to this positive test.
In people with diabetes, we have been trained to have an exaggerated fear of diabetic ketoacidosis, which would also generate a positive test on the urine dipstick done at most of the ob appointments.
One of the biggest impediments in educating doctors and healthcare workers about the ketogenic approach relates to the confusion between nutritional ketosis (ketogenic diet) and diabetic keto acidosis, a completely different and life-threatening pathologic condition.
In medical school, doctors are not well trained in nutrition and definitely not made aware of nutritional ketosis. Until I became interested in metabolic medicine, I would never have been able to explain and didn’t understand nutritional ketosis. Therefore, it is not a deficiency in the obstetric doctors knowledge base because they were never trained in this process.
I have reassured patients (without type 1 diabetes) that as long as they are eating frequently (at least every 3 hours) and consuming plenty of calories (about 2,500 calories or more per day), that ketosis is not an abnormal state and it is expected with a ketogenic approach.
One way to educate the medical team would be to say “I am following a ketogenic nutritional approach and ketones are expected in the urine. Unless you believe that I am sick, I will just continue my diet.” Then recommend that the physician visit Diet Doctor.
Did you enjoy this guide?
We hope so. We want to take this opportunity to mention that Diet Doctor takes no money from ads, industry or product sales. Our revenues come solely from members who want to support our purpose of empowering people everywhere to dramatically improve their health.
Will you consider joining us as a member as we pursue our mission to make low carb simple?
- International Diabetes Federation: Gestational Diabetes [overview article; ungraded]
- International Journal of Women’s Health 2010: Gestational Diabetes: Risks, Management, and Treatment Options [overview article; ungraded]
- Clinical & Investigative Medicine 1996: Carbohydrate Intolerance in Pregnancy: Incidence and Neonatal Outcomes [retrospective observational study; weak evidence]
- Seminars in Perinatology 2015: Pregnancy As a Window to Future Health: Excessive Gestational Weight Gain and Obesity [overview article; ungraded]
- American Diabetes Association 2007: Summary and Recommendations of the Fifth International Workshop – Conference on Gestational Diabetes Mellitus [overview article; ungraded]
- The Journal of the American College of Nutrition 1990: Dietary Manipulation As a Primary Treatment Strategy for Pregnancies Complicated by Diabetes [non-randomized study; weak evidence]
- The American Diabetes Association 2016: Carbohydrate Content in the GDM Diet: Two Views: View 1: Nutrition Therapy in Gestational Diabetes: The Case for Complex Carbohydrates [overview article; ungraded] (60% carbs)
- Diabetes Spectrum 2016: Carbohydrate Content in the GDM Diet: Two Views: View 2: Low-Carbohydrate Diets Should Remain the Initial Therapy for Gestational Diabetes [overview article; ungraded](40% carbs)
We are not aware of any clinical trials in pregnant women that examined a very low-carb or ketogenic diet compared to other dietary patterns. Therefore, in the absence of clinical data, much of the information presented in this guide is based on clinical experience and extrapolation of data from non-pregnant women. ↩
As of Sunday July 23 2017 ↩
This is based on Drs. Jovanovic’s and Nichol’s review of the literature which was unable to find a study substantiating the recommendation for 175grams of carbohydrates. ↩
This is based on Drs. Jovanovic’s and Nichol’s review of the literature which was unable to find a study substantiating the recommendation for 175grams of carbohydrates. ↩
- Obstetrics & Gynecology 2014: Diabetic Ketoacidosis in Pregnancy [overview article; ungraded]
- European Journal of Obstetrics and Gynecology: Starvation Ketoacidosis in Pregnancy 2013 [overview article; ungraded]
This is based on the clinical experience of these particular practitioners. [anecdotal report; very weak evidence]
It is also substantiated by normal physiology seen in non-pregnant individuals in whom nutritional ketosis is safe and distinctly different than ketoacidosis. You can read more in our guide on ketosis.
The American Diabetes Association 2014: An Early Pregnancy HbA1c ≥5.9% (41 mmol/mol) Is Optimal for Detecting Diabetes and Identifies Women at Increased Risk of Adverse Pregnancy Outcomes [non-controlled study; weak evidence] ↩
This is based mostly on her opinion and clinical experience although some studies seem to support her hypothesis.
- An International Journal of Obstetrics & Gynaecology 2012: Glucose Challenge Test for Detecting Gestational Diabetes Mellitus: A Systematic Review [systematic review of randomized trials; strong evidence]
- American Journal of Obstetrics & Gynecology 1999: Jelly Beans As an Alternative to a Fifty-Gram Glucose Beverage for Gestational Diabetes Screening [randomized trial; moderate evidence]
- American Journal of Obstetrics & Gynecology 2015: Candy Twists As an Alternative to the Glucola Beverage in Gestational Diabetes Mellitus Screening [randomized trial; moderate evidence]
American Journal of Obstetrics & Gynecology 2016: Single Abnormal Value on 3-Hour Oral Glucose Tolerance Test During Pregnancy Is Associated with Adverse Maternal and Neonatal Outcomes: A Systematic Review and Metaanalysis [review of observational studies; weak evidence] ↩
American Journal of Obstetrics & Gynecology 2015: The Impact of Adoption of the International Association of Diabetes in Pregnancy Study Group Criteria for the Screening and Diagnosis of Gestational Diabetes [non-controlled study; weak evidence] ↩
This is not officially endorsed by medical societies but represents the opinion of the experts interviewed. ↩
This is based on her clinical experience ↩
The following is based on his opinion and clinical experience ↩