Home / Resources / Articles / Very Low Carbohydrate Diet Equals Better Diabetes Control

Very Low Carbohydrate Diet Equals Better Diabetes Control

May 19, 2018

Very low carbohydrate diet (VLCD) equals lower blood sugars and control, with less hypoglycemia, but how low do diabetes patients need to go?

Endocrinologists have traditionally focused on the adjustment of insulin rather than diet as a primary means for controlling glucose levels; for example, the 2018 Standards of Medical Care in Diabetes for children and adolescents does not even address dietary management. This gap is partly due to the lack of evidence regarding outcomes of optimal dietary strategies for improving glycemia in T1D. This is why the recent study by Belinda Lennerz MD et al, in which they looked at glycemic outcomes of individuals with T1D on a very low–carbohydrate diet (VLCD), is an important contribution to the literature.

We know from experience that the more insulin a patient has to take, the less control they will have, as the use of insulin is not an exact science and depends on many factors. Plus we know that the more carbohydrates one eats, the more insulin is required to control blood sugars.

The authors enrolled 316 participants from a closed Facebook community of adults with T1D and parents of children with T1D who follow recommendations outlined in the book Dr Bernstein’s Diabetes Solution. Only participants who self-reported following the recommendation for a VLCD (≤30 g per day from mostly fibrous vegetables and nuts) for at least 3 months before the study were enrolled. Data collected from patient surveys were validated with medical records and provider surveys, and the authors took a rigorous approach to ensure confidence in a true diabetes diagnosis based on age at diagnosis, BMI, positive results for diabetes antibodies, and insulin requirements immediately after diagnosis. Adult participants reported following the VLCD for a mean of 2.7 ± 3.6 years, with a mean daily carbohydrate intake of 36 ± 16 g. In the pediatric group, the mean VLCD duration was 1.4 ± 1.2 years, and the mean daily carbohydrate intake was 36 ± 14 g.

Before the initiation of the VLCD, the mean hemoglobin A1c (HbA1c) for all participants was 7.15% ± 1.15%. At the time of data collection, the mean reported HbA1c for all participants was 5.67% ± 0.66%, and it was 5.71% ± 0.58% for the 131 pediatric participants, which is an exquisite level of control. Also there was an almost 0.1% increase in A1C for each additional 10 grams of carbohydrates consumed.

Among a subset of patients who used continuous glucose monitoring, the mean reported blood glucose value was 102 ± 17 mg/dL with an SD of 26 ± 12 mg/dL for adults (n = 64) and 107 ± 15 mg/dL with an SD of 29 ± 12 mg/dL for kids (n = 51). Authors of previous studies of the VLCD have found comparable decreases in HbA1c, albeit not as dramatic. Authors of another study of individuals who were on a carbohydrate-restricted diet of 30g per day for 8 months to 61 months reported a mean change in HbA1c from 6.8% ± 1.1% to 5.5% ± 0.8%, but this was a small study of just 10 participants with T1D, and was performed only in adults.

Reassuringly, the authors found relatively low rates of adverse events before and after the initiation of the VLCD for diabetes-related hospitalizations (8% before and 2% after), emergency encounters (8% and 3%), hypoglycemia requiring help from others (13% and 7%), and hypoglycemia requiring glucagon (7% and 4%); however, the rates may be susceptible to recall bias. Nonetheless, this is the first report of the frequency of adverse events and hospitalizations related to diabetes before and after the initiation of a VLCD because other studies have been limited by sample size or a lack of reporting.

However, limitations of the study, including the fact that it was performed in a subgroup of highly motivated patients who had well-controlled diabetes even before starting the diet (mean HbA1c of 7.15%), and the fact that detailed dietary records were not collected, making the carbohydrate sources unknown. Furthermore, the VLCD may have been accompanied by changes in other aspects of diabetes management, such as maintaining stricter adherence to the insulin regimen, that could not be controlled for due to the study design.

Even if the therapy is determined to be efficacious, the uptake by patients and providers may be a barrier. The authors assessed the participants’ satisfaction with their diabetes management and care team with a focus on the VLCD. Interestingly, 27% of participants reported not discussing their VLCD with their provider, and only approximately one-half of those who did discuss it agreed that their providers were supportive. This finding reveals the need for improved communication and shared decision-making between the patient, caregiver, and provider regarding the overall management of T1D, and the need for greater dialogue within the T1D community regarding dietary standards of care.

The researchers reported that the average blood glucose levels among the group of patients who reported these values were 104 ± 16 mg/dL, as they published in Pediatrics.

One of the benefits of those who followed this VLCD was also that they had very few adverse events, with only 2% of the total respondents reporting a diabetes-related hospitalization within the past year — 1% for ketoacidosis and 1% for hypoglycemia.

Following a switch from a regular diet to the VLCD, participants reported an average change in A1c of -1.45% ± 1.04% (P<0.001). There was also a roughly 0.1% increase in HbA1c for every 10 grams of additional carbohydrates consumed.

The study participants noticed that the source and amount of carbohydrates consumed affect postprandial hyperglycemia and glycemic variability more than any other dietary factor such as protein and fats, providing a conceptual basis for interest in carbohydrate-modified diets for T1DM.

The participants were confirmed to have a positive diabetes antibody test result, a diagnosis of diabetes prior to the age of 20, and a body mass index (BMI) under 30 for adults or lower than the 95% percentile for children. The international cohort included individuals from the U.S., Canada, Europe, and Australia. The findings held similar for both pediatric and adult patients.

In this new published study in Pediatrics, they concluded that for type 1 patients who adhered to a very low-carbohydrate diet, they showed very good glycemic control, 97% of whom achieved the American Diabetes Association’s recommended glycemic targets.

The world’s largest study of how a low carbohydrate diet can impact and help control type 1 diabetes is set to start this autumn. The Dietary Science Foundation in Sweden, which funds scientific studies focused on diet and how it can affect health, has been raising funds to carry out the work.

Practice Pearls:

  • More studies are needed to assess not only the effect on HbA1c, but also the frequency of diabetic ketoacidosis, hypoglycemia, BMI, linear growth outcomes, cholesterol, and percentage of time in range.
  • It was not that long ago that the American Diabetes Association included in the standards of care that low carbohydrate diets can be useful for some as we individualize treatments.
  • Now with continuous glucose monitors, it is even safer to reduce glucose levels into the normal ranges without hyperglycemia.
  • Low carbohydrate diets for those with diabetes could become the norm.

Pediatrics, May 2018; From the American Academy of Pediatrics