People with diabetes may wish to fast for dietary, medical, or religious reasons. Advance planning with a health care professional may reduce complications.
When Martin M. Grajower, MD, encountered patients with diabetes through his clinical practice who were committed to fasting for religious and other reasons, he was inspired to support them in finding ways to fast safely. Here, he discusses strategies that can help people with diabetes avoid health complications while fasting.
Q: What is a fast?
A: A fast is any defined period of time in which someone chooses not to eat. That could be six hours, eight hours, or 24 hours or more. A fast can mean no intake of food, or it can mean nothing at all by mouth, including no liquids.
Q: Why do people with diabetes fast?
A: People with diabetes fast for a variety of reasons, just like people without diabetes do. Most commonly today, people fast for dietary reasons. They want to lose weight, and so they may engage in calorie restriction by means of intermittent fasting.
People also fast for spiritual or religious reasons. For example, various religious practices call for fasting over a 24-hour period, over a certain number of hours per day for an entire month, or for one day a month throughout the year, or on some other schedule.
People may also fast because they have been told to do so before going into surgery. For example, sometimes people are told to eat nothing after midnight, but their surgery may not be until 3 oclock in the afternoon the next day, so thats a 15-hour fast. Some people going in for procedures, such as a bowel procedure, may be told to consume only liquids for 24 hours, which is also a form of fasting.
Q: What are the risks from fasting for people with diabetes?
A: Hypoglycemia is the number one risk. Hyperglycemia can also become an issue because people will cut back on their diabetes medicine, especially if theyre on insulin, to avoid hypoglycemia. But if they cut back too much, then their blood glucose will go too high and theyre at risk of hyperglycemia, and even potentially diabetic ketoacidosis.
Dehydration is another risk and depends on the nature of the fast. If its a complete fast where the person fasting cant take in liquids or has been instructed not to take anything orally to prepare for surgery, then dehydration becomes a major risk. The dehydration could be due to not drinking or because some diabetes medicines, such as the SGLT-2 inhibitors and diuretic medicines, induce diuresis. Furthermore, hyperglycemia induces diuresis.
Q: How do the different types of diabetes affect fasting risks?
A: People with type 1 diabetes are at greater risk when fasting compared to people with type 2 diabetes, because they are on insulin. The amount of insulin they take when fasting needs to be adjusted. If they dont cut back enough, they risk hypoglycemia, but if it is cut back too much, they can develop hyperglycemia. People with type 1 diabetes also face increased risk of dehydration and diabetic ketoacidosis, compared to people with type 2 diabetes. However, there are more people with type 2 diabetes, so at the population level there are more people at risk when fasting.
For women with gestational diabetes, if theyre not on medicine, especially if theyre not on insulin, fasting is the best insulin sensitizer that we have. So, I never hesitate to let women with gestational diabetes restrict calories for a short period. Pregnant women do have to factor in the risk of dehydration, taking into consideration their blood pressure and any edema.For women with gestational diabetes who wish to partake in the longer fasting regimen of certain religious practices, thats a separate discussion.
Q: What challenges does intermittent fasting pose for people with diabetes?
A: There are two types of intermittent fasting. Theres the type that you do every dayfor example, every day you eat for only eight hours during the day. I don't necessarily recommend it, but I have no problem with a person with diabetes following this eating pattern, because they're doing the same thing every day and you can just adjust their medications accordingly. The intermittent fasting where you fast two days a week or every other day is more problematic because it can become very complicated to adjust the medication. It can be done, but it requires the time and the expertise of the physician and it requires the compliance of the patient. So, I don't recommend patients do it on their own, but it can be done safely under medical supervision.
Q: What is your approach with patients who wish to fast for religious or spiritual reasons?
A: I became interested in this subject because of a couple of patients. A member of my religious community went to the rabbi and said, Last year my doctor said anyone with diabetes shouldnt fast, but I did anyway, on my own, and my sugar dropped low. So, what do I do this year? The rabbi called me up and asked what I should tell him. I found out that the only medicine this person was taking was a sulfonylurea, so I told him to stop taking his medicine 36 hours before Yom Kippur, and he did fine.
I also had an elderly Orthodox Jewish woman as my patient. I said to her, I dont think you should fast on Yom Kippur because youre elderly, youre on heart medicines, and youre on a complicated insulin regimen of three shots a day. She looked me straight in the eye and said, Doctor, I fasted on Yom Kippur in the concentration camps, so dont tell me not to fast now. Im going to fast with or without your help, but Im going to fast.
This was a powerful lesson. The determination to fast is found not just in the Jewish religion, but also in the Muslim faith. People hold Ramadan to be a very holy time, and theyre going to fast either with or without their doctors help. People of other faiths or who adhere to other spiritual or meditative traditions fast as well. Thats why I've become a very big proponent of allowing people with diabetes to fast, but under medical supervision.
Its our obligation as health care professionals to adapt diabetes to our patients religious beliefs. I did my fellowship under Dr. Harold Rifkin, who co-wrote the first textbook on diabetes. He taught me that you need to adjust the management of diabetes to the patients lifestyle, not the other way around.
I really think nurse practitioners and nutritionists could take the lead on this, because doctors unfortunately dont always have the time. If you have patients who are Jewish, Muslim, or a member of the Church of the Latter-day Saints, you can ask, Do you fast for religious reasons? And if they do, talk with them about how theyre going to manage it. Because if you dont ask the question, patients will do it on their own, and thats when theyre going to run into problems.
Q: What are the concerns when people must fast prior to surgery?
A: The major concerns are hypoglycemia and dehydration, both of which can be avoided by adjustment of medication and scheduling the surgery, for early in the day when possible.
Q: How do you help manage the patient who wants to fast?
A: Health care professionals need to consider the pharmacodynamics (mechanism of action) and pharmacokinetics (the onset and duration of action) of the diabetes medicine a patient is on. How long does the medicine work? How long does it stay in the system? Does the medicine increase hypoglycemia risk or is its action glucose dependent?
Sulfonylureas, the short-acting meglitinides, and insulin, are associated with hypoglycemia. The sulfonylureas have a 24- or 36-hour duration of action, so those need to be stopped at least 24 and preferably 36 hours before the patient is going to fast. Meglitinide and Nateglinide generally are taken three times a day before each meal because it has a duration of action of only 4-6 hours. Patients should not take a glinide medicine if theyre not eating or if theyre not going to eat carbohydrates.
Insulin requires a major adjustment, so the health care professional should understand the duration of action for the kind of insulin that the patient is on. For example, certain long-acting insulins are taken every day and have a duration of action of 36 to 42 hours. If a patient takes insulin on Monday, the effect is going to last until Wednesday. If I have a patient with this kind of insulin going in for a medical procedure on Tuesday, I advise him or her to reduce their dose of insulin on Sunday, two days prior, as well as on Monday, one day prior. I provide detailed instructions on how much to reduce the dosage, as described in the article on medication adjustment referenced below.
The older NPH (isophane) insulin has a duration of action of about 12 to 16 hours, and other forms of long-acting insulins have a duration of between 16 and 24 hours. For these medicines, you would have to help the patient adjust dosages mostly the day before the procedure.
Metformin, pioglitazone, and DPP-4 inhibitors rarely cause hypoglycemia, so health care professionals dont have to adjust them. But the patient should not take it on the day of fasting if its a 24-hour fast. With patients doing intermittent fasting, where they are eating during 8 hours of the day and going on a 16-hour fast, I dont tell them to stop taking the medicine, because they rarely cause hypoglycemia, and the medicine should be in their system for those 6 or 8 hours while they are eating to prevent hyperglycemia.
Q: What about dehydration concerns?
A: As far as dehydration goes, it really depends on the kind of fast. With intermittent fasting, fluid intake is never restricted; just calories are restricted. So, people with diabetes can drink water, diet soda, tea, or black coffee without hesitating, and dehydration should generally not be an issue. However, patients who normally get a lot of their liquids from foods like soups, shakes, jello, and yogurt may not realize that three-quarters of their fluid intake is really coming through food. Even if they drink as much while fasting as they do at other times, they will not be consuming enough liquid and they could run into a problem with dehydration.
Health care professionals also need to keep in mind that the SGLT-2s, besides lowering blood sugar, have a diuretic effect. Both aspects of the medicine must be considered when adjusting the dosages. I generally will stop the SGLT-2 two days before a patient begins a fast because of the dehydration aspect.
Health care professionals should also consider other medicines the patient is on, especially diuretics. These may also require adjusting. We also keep in mind the patients other medical conditions. A patient who has had a heart attack or a stroke within the last three months is at increased risk from dehydration and the resultant drop in blood pressure. If the patient becomes hypotensive from dehydration, this could lead to another heart attack or another stroke.
If A1C is not controlled, the patient is also at increased risk for dehydration, because glucose in the urine acts as a diuretic. If a patient has an A1C of 9 or greater, I will strongly discourage fasting due to the risk of dehydration from the high blood sugar or, if the patient has type 1 diabetes, the risk of going into diabetic ketoacidosis.
A patient who is running any fever in the last week or so should not be fasting, again because of the risk of dehydration resulting from fluid loss due to sweating. Health care professionals need to be conscious of these other issues before going ahead and giving a blanket recommendation regarding fasting.
Q: How do you advise patients regarding glucose monitoring during a fast?
A: Patients who are on insulin and fasting should do even more frequent glucose monitoring than usual until they get a sense of the safety of their revised insulin regimen. For example, the patient on a long-acting insulin who decides to intermittently fast two days a week, with the help of a health care professional, should adjust the insulin the day before the start of the fast. Then over the first two or three fasting periods, the patient should check glucose levels even more frequently than normal, until it can be established that the lower dose of insulin is correct. Subsequently, the normal frequency of testing can be resumed.
Someone whos not on a sulfonylurea or insulin doesnt have to test any more frequently than normal because the risk of hypoglycemia is extremely low.
Note: For detailed guidelines on medication adjustment and other considerations while fasting with diabetes, see the articles listed at the end of this interview.
Q: Do you have any other tips for helping patients with diabetes manage fasting?
A: At the time of a patients pre-fast visit, I write down all my instructions. I hand the patient a copy (to avoid misunderstandings), and I keep a copy in the patients chart. In the instructions, I put down medication adjustments, how often to check blood glucose readings, and what to do if the blood glucose reading goes above or below a certain specific number (individualized for the patient depending on age, the presence of hypoglycemia unawareness, and comorbid conditions). Soon after the fast, either at the next visit or via a follow-up telephone call, I ask the patient how he or she did. I make a note of that in the chart. The next time the fasting observance comes around, I look back at my previous note in the chart. If the patient did well, I simply make a photocopy and say, Here are your instructions. And for me, instead of spending 10 minutes, now it takes only 30 seconds.
Also, on the occasion when Ive told patients that I dont think they should be fasting, I ask permission to discuss it with their clergy. Youd be surprised how often a patient will let me do that. And then when their religious advisor tells them not to fast, they feel much more comfortable about it.
Guidance on fasting with diabetes by Dr. Grajower and others:
How do you address the subject of fasting with your patients who have diabetes?
