Very Low-Carbohydrate Diet Beneficial in Type 1 Diabetes

Miriam E. Tucker

May 07, 2018

Following a very low-carbohydrate diet (VLCD) can produce "exceptional" glycemic control for both adults and children with type 1 diabetes, new research finds. 

The results, from more than 300 patients with type 1 diabetes, were published online May 7 in Pediatrics by Belinda S. Lennerz, MD, PhD, of Boston Children's Hospital and Harvard Medical School, Massachusetts, and colleagues.

Study participants all followed the VLCD regimen described in the book Dr Bernstein’s Diabetes Solution, whose author, Richard K Bernstein, MD, of New York Diabetes Center in Mamaroneck, and who has type 1 diabetes himself, was a study investigator. 

Both the adult and pediatric study participants were able to achieve near-normal HbA1c levels without significant increases in some of the potential adverse effects that have been raised as concerns about such an approach, including hypoglycemia, dyslipidemia, or growth impairment among children. 

"We suggest that a VLCD may allow for exceptional control of type 1 diabetes without increased risk of adverse events. This possibility is mechanistically plausible because of the dominant effects of dietary carbohydrates on postprandial glycemia and the lower insulin doses required with a VLCD," Lennerz and colleagues write.

However, they caution, "In light of study limitations, these findings by themselves should not be interpreted as sufficient to justify a change in diabetes management."

In an accompanying editorial, Carly Runge, BS, and Joyce M. Lee, MD, MPH, both of the University of Michigan, Ann Arbor, call the study "an important contribution to the literature," given the dearth of information about optimal dietary strategies for type 1 diabetes.

Indeed, Runge and Lee note, current guidance given to patients with type 1 diabetes — as opposed to type 2 diabetes — focuses almost entirely on insulin adjustment around carbohydrates, despite the fact that the timing of action of even short-acting analogs lags behind carbohydrate absorption, resulting in post-meal glucose spikes.

Indeed, a father of one of Lee's patients told her in response to her asking for his opinion about the advice provided to the family at the time the child was diagnosed, "You should never have told us that our son can eat anything he wants. It's just not true."

But the editorialists also point out that the generalizability of the findings and long-term safety of carbohydrate restriction remain unknown. Moreover, "even if the therapy is determined to be efficacious, the uptake by patients and providers may be a barrier."

"Exceptional" Blood Glucose Control   

The study included 316 individuals, of whom 54% were adult patients and 42% were parents of children with type 1 diabetes. All were part of an international Facebook group for people with type 1 diabetes who follow Bernstein's recommended VLCD diet.

The approach includes a weight-based carbohydrate prescription of no more than 30 grams/day, derived from fibrous vegetables and nuts with a low glycemic index. High-protein foods are substituted for carbohydrates, and insulin doses adjusted empirically by postprandial and fasting glucose levels.  

Most of the participants were from the United States, Canada, Europe, or Australia, the majority (88%) were white, and 84% were college graduates. Mean age at diabetes diagnosis was 16 years, diabetes duration was 11 years, and time following the VLCD diet was 2.2 years.

Confirmatory data were obtained from diabetes care providers and medical records, including a multi-tiered investigation to ensure all participants had type 1 diabetes — not type 2 diabetes or a genetic variant. However, not all data points were available for all respondents.  

Participants reported consuming an average of 36 grams/day of carbohydrate. The mean reported HbA1c was 5.67%, a drop of 1.45 percentage points following adoption of the VLCD (P < .001). Average blood glucose level was 104 mg/dL in the 137 patients who had continuous glucose monitoring data.

In a regression analysis, carbohydrate intake goal was the only significant predictor of variation in HbA1c (P = .001), with an increase in HbA1c of 0.1% per 10 g of carbohydrate consumed. The mean daily insulin dose was 0.40 U/kg/day.

Short-Term Adverse Events Not Increased

Just seven participants (2%) reported 14 hospitalizations within the past year, including four hospitalizations for diabetic ketoacidosis and nine for other reasons. Slightly more than two thirds (69%, n = 205) of participants reported having symptomatic hypoglycemia in the past month, but just seven (2%) reported severe hypoglycemia with seizure or coma, and 11 (4%) required glucagon.

Fasting lipid profiles were mixed. Triglyceride levels were low (mean 74 mg/dL), and HDL cholesterol high (74 mg/dL), but total and LDL cholesterol levels were high (234 mg/dL and 147 mg/dL, respectively).   

HbA1c and other parameters were similar for children and adults.

Among the children, participant- and provider-reported standard deviation scores (SDS) for height were 0.26 and 0.25, respectively, and there was no correlation of height SDS with carbohydrate intake goal (P = .20) or diet duration (P = .16).

Patient-Provider Communication Lacking

Of concern, 27% of respondents said they don't discuss their VLCD adherence with their diabetes care provider. And of those who do discuss it, fewer than half (49%) agreed or strongly agreed that those providers were supportive.

Reasons given for not discussing the VLCD with providers included disagreement about treatment goals and approach, perceived provider disinterest or unfamiliarity with VLCD, desire to avoid conflict with the provider, and, among parents, fear of being accused of child abuse. 

In contrast, 82% of participating providers perceived the therapeutic relationship as very good or excellent, and were significantly more likely than the patients/parents to say that they were, in fact, supportive of the VLCD (P = .007). 

Not Ready for Universal Recommendation

Lennerz and colleagues write, "Additional research is needed to determine the degree of carbohydrate restriction (and other dietary aspects) necessary to achieve these benefits, optimal insulin regimen to accompany a VLCD specifically, with regard to avoiding severe hypoglycemia, safety, and efficacy (in randomized controlled trials)."

And, they add that once those are established, trials should be conducted to evaluate the effectiveness of VLCD in preventing long-term diabetes complications.

Addressing the patient-provider communication issue, Runge and Lee say, "This finding reveals the need for improved communication and shared decision-making between the patient, caregiver, and provider regarding the overall management of type 1 diabetes and the need for greater dialogue within the type 1 diabetes community regarding dietary standards of care."

The study was supported by the National Institute of Diabetes and Digestive and Kidney Diseases. Bernstein receives royalties for books on diabetes management. Runge and Lee have reported no relevant financial relationships.

Pediatrics. Published online May 7, 2018. Full text, Editorial

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.