Very-Low-Carbohydrate Diet vs. Oral Diabetes Medication

A recent study by Kaiser Permanente and Yale Medical Center found that severe hypoglycemia is quite common in people with Type 2 diabetes who take insulin-stimulating medications. This was true regardless of level of control, meaning those with elevated blood sugar and A1c levels experienced low blood sugar as well as those who were within or below their A1c goal. Severe hypoglycemia is defined as a blood glucose level less than 50 mg/dL and is associated with increased risk for heart attack, stroke, loss of consciousness, and death, particularly when it occurs during sleep. Its symptoms can be frightening and include sweating, shaking, dizziness, unsteadiness, and heart palpitations.

Although having tight control (A1c <6%)  is considered dangerous by many diabetes specialists because it suggests frequent low blood glucose levels, it's evident that anyone taking medications which cause the pancreas to secrete insulin is at risk for hypoglycemia as well.  (For the record, I think having a lower A1c is good, provided it's not due to  frequent lows). According to Dr. Kasia Lipska, an endocrinologist at Yale Medical Center, "It's important to note that it's not the HbA1c that directly causes hypoglycemia; it's the therapies we use to lower it."  Exactly.

Insulin-stimulating medications cause unpredictable blood glucose response in several ways. Typically prescribed to be taken twice a day at meals in fixed dosages, they are unable to make the pancreas produce the precise amount of insulin needed to cover the carbohydrate ingested at a meal, nor do they start working at exactly the right time to match the digestion of carbohydrate. In the poorly controlled overweight person with diabetes, taking this type of medication practically guarantees at least occasional episodes of low blood sugar leading to overtreating with juice, soda, or candy, resulting in hyperglycemia and weight gain. It also places a burden on the beta cells of the pancreas by causing them to secrete large amounts of insulin, thereby increasing progression of the disease. Precisely the problems diabetes management is supposed to avoid.

I feel that carbohydrate restriction should be offered as an alternative to taking these types of medications, and I outline the basis for why this way of eating is ideal for diabetes management in my article.  I have heard about or spoken with many people -- including  Type 2 bloggers Dan Brown, Steve Cooksey, and Eddie Mitchell,  as well as Dr. Jay Wortman -- who have been able to stop their insulin-stimulating diabetes meds and improve their blood glucose control by following a very-low-carbohydrate ketogenic diet (VLCKD). In most cases, those who adopt a VLCKD require only metformin, an insulin sensitizer that does not place a person at risk for hypoglycemic events.

The message given by many of my fellow dietitians and CDEs is, "You can eat the same foods everyone else does as long as you take your diabetes medication."  I don't feel comfortable naming names, but the vast majority of articles by RDs and CDEs that I've read advise individuals with diabetes to eat low-fat, high-carb meals and snacks and take whatever meds are needed to keep blood glucose in check. I realize many people may not want to change their eating habits, and that is of course their choice. But I think they should be told about the risks of these medications, including the strong likelihood that they will periodically experience low blood sugar when taking them. Some will want to assume the risk, but others may be interested in an alternative way of eating that involves less medication and no risk of low blood sugar.  Every patient I talk to who has ever experienced severe hypoglycemia would prefer to avoid it all costs.

I'm not saying that there isn't a need for diabetes medication in some people. People with Type 1 diabetes obviously require long-acting and mealtime insulin, although considerably less of the latter when on a carbohydrate-restricted diet (Hypoglycemia is also minimized with this approach). Those with Type 2 who adopt a VLCKD may only need metformin and possibly a long-acting insulin, depending on how much beta cell function they have remaining. Dr. William Yancy and Dr. Eric Westman have demonstrated that insulin and oral diabetes medications can  be reduced and in some cases eliminated in people following a VLCKD, and that blood sugar control improves across the board with this method. Why not encourage and support those who are interested in trying it?


1. Kaiser Permanente. Severe low blood sugar occurs often in patients with Type 2 diabetes. Science Daily. July 30, 2013.
2.  Westman EC, et al.A low-carbohydrate, ketogenic diet to treat type 2 diabetes. Nutr Metab 2:34, 2005
Yancy WS,  et al. The effect of a low-carbohydrate, ketogenic diet versus a low-glycemic index diet on glycemic control in type 2 diabetes mellitus. Nutr Metab 5:36, 2008. 

  • Janknitz

    Most individuals with Type II diabetes have hyperinsulemia until they are in the latter stages of the disease. Why are there meds that INCREASE insulin???

- Enter Your Location -
- or -