Carbohydrate Beta Cell Function and Glucose Control in Children With Diabetes

Effect of Dietary Carbohydrate on Diabetes Control and Beta Cell Function in Children With Newly Diagnosed Diabetes

The goal of this clinical trial is to test the effects of a ketogenic diet on the progression and control of type 1 diabetes in children with newly diagnosed diabetes. The main questions to answer are:

• Does a ketogenic diet prolong the honeymoon period of type 1 diabetes?
• Does a ketogenic diet improve diabetes control?
• Is a ketogenic diet safe, acceptable and sustainable in children with newly diagnosed diabetes?
• What are the microbiome, inflammatory and metabolic changes linking diet to β-cell function?

Participants will receive a combination of free meals, groceries, micronutrient supplements, and intensive diet and diabetes education for 9 months.

• Diabetes care devices will be connected for cloud-based data collection.
• Bi-weekly data downloads and remote check-ins will assess dietary intake, satisfaction with diet and study procedures, and possible safety concerns.
• During four study visits held at at baseline, 1, 5, and 9 months, an intravenous catheter (IV) will be placed for collection of 5 blood samples before and up to 2 hours after a liquid test meal (protein shake) to assess insulin response. A stool sample will also be collected to assess microbiome changes.
• Children and their caregivers may be invited to participate in a semi-structured interview, and online questionnaires to assess their experience with the diet and diabetes care, general well-being and quality of life.
• Children and their caregivers may be invited to participate in a follow-up visit to evaluate long-term effects after 24 months.

Comparison will be made between a ketogenic vs standard diet.

Study Overview

• Status: Recruiting
• Conditions: Diabetes Mellitus, Type 1
• Intervention / Treatment: Other: Ketogenic diet, food delivery and education; Other: Standard diet, food delivery and education

Detailed Description

Type I diabetes is caused by an autoimmune destruction of insulin producing β-cells in the pancreas, resulting in absolute insulin deficiency. In the first months after diagnosis, a small number of β-cells typically remain and, by producing insulin, significantly improve diabetes control and reduce disease burden.

Preliminary data suggest that this early disease stage entitled the "honeymoon period" might be extended by a ketogenic diet, which would provide a major therapeutic advantage and may reduce chronic disease burden.

To test the hypothesis that a ketogenic vs. standard diet will extend the honeymoon period and improve diabetes control in children, the researchers are conducting a study employing education and food deliveries of a ketogenic or standard diet to children and their families. Fifty-two children aged 5 to 12 years with newly diagnosed diabetes will participate. Children will be assigned by chance (randomized) to receive either a ketogenic or a standard diet for 9 months. Chances to be assigned to either diet are 50:50 like a coin flip, and 26 children will participate in each diet arm.

Participants will receive a combination of free meals, groceries, micronutrient supplements, and intensive diet and diabetes education throughout the 9 months. Continuous glucose monitoring (CGM) and diaries will be used for cloud-based data collection. Bi-weekly data downloads and remote check-ins will be performed to assess dietary intake, satisfaction with diet and study procedures, and possible safety concerns. Participants are instructed to measure blood ketone levels with their home ketone meter anytime blood glucose levels exceed a safety threshold and to call the study physician for persistent low glucose levels or ketones above diet specific safety thresholds.

Study visits are held at at baseline, 1, 5, and 9 months to collect height, weight, stool and blood samples for hormones, metabolites and inflammatory biomarkers. At each visit, an intravenous catheter (IV) will be placed to collect fasting blood samples, followed by a liquid test meal (protein shake) and collection of four additional blood samples from the IV over the course of two hours. Prior to each visit, participants will collect stool samples at home using provided kits. In addition, participants and their families may be invited to participate in a semi-structured interview, and online questionnaires to asses their food intake, experience with the diet, diabetes care burden and complications, and general well-being and quality of life. They may also be invited to participate in a follow-up visit to evaluate long-term effects after 24 months.

• Study Type: Interventional
• Enrollment (Estimated): 52
• Phase: Not Applicable

Contacts and Locations

• Study Contact: Name: Belinda Lennerz, MD PhD; Phone Number: 857-218-3896; Email: belinda.lennerz@childrens.harvard.edu

Study Locations

• United States
  • Massachusetts
    • Boston, Massachusetts, United States, 02115
      • Recruiting
      • Boston Children's Hospital
      • Contact: Belinda Lennerz

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Children aged 5 to 12 years.
• Within 3 month of diabetes diagnosis.
• Insulin adjusted HbA1c ≤9 if enrolled ≥ 2 months pat diagnosis.
• Type 1 diabetes confirmed by immediate insulin requirement and any 2 of the following criteria: autoimmunity marker [glutamate decarboxylase-65, islet-antigen-2, zinc transporter-8, insulin [prior to first insulin dose]; age under 10 years, BMI <95th percentile.
• Family committed and able to participate in study education and implement dietary intervention.

Exclusion Criteria:

• Dietary needs or habits incompatible with the study meal plans, (e.g., vegan, major food intolerances/allergies, ketogenic).
• Eating disorders as assessed by Chede-Q8.
• Major medical illness or use of medications other than insulin that could interfere with metabolic or glycemic variables.
• Major psychiatric illness.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Treatment
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Single

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: ketogenic diet; The diet will be high in protein and healthy fats and comprise meat, fish, fibrous vegetables, nuts, dairy, and berries. Macronutrient composition will be ~ 5% carbohydrate, 20% protein, 70% fat. Participants will receive a daily multi-vitamin, magnesium supplement, and supplemental salt (bouillon cubes) to ascertain micronutrient sufficiency and help with transition to the diet.	Other: Ketogenic diet, food delivery and education; Meals and groceries will be delivered and participants will receive education on nutrition, meal preparation, and diabetes care strategies. Participants will consume study-prescribed foods exclusively.
Active Comparator: standard diet; The diet will be consistent with prevailing dietary guidelines and recommendations and contain meat, fish, grains, vegetables, fruit and dairy. At least 50% of grain-based products will be whole grains. Meats will be primarily lean, and dairy products will be fat-free or low-fat. Macronutrient composition will be ~50% carbohydrate (<10% added sugars), 20% protein, 30% fat. Participants will receive a daily multi-vitamin supplement to ascertain micronutrient sufficiency.	Other: Standard diet, food delivery and education; Meals and groceries will be delivered and participants will receive education on nutrition, meal preparation, and diabetes care strategies. Participants will consume study-prescribed foods exclusively.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Decline in Beta-cell Function	Change in C-peptide 2-h area under the curve after a mixed-meal tolerance test (ΔCP).	Change over 1, 5, and 9 months, corrected for baseline

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Time in Range (TIR) 70-180 mg/dl	From continuous glucose monitoring (CGM) - percent time spent in the specified glycemic target range will be computed throughout study participation in 2-week increments.	Over 9 months and optional at 24 months
Duration of Clinical Diabetes Remission	Calculated based on percent children with insulin dose corrected HbA1c (IDAA1c) <9.	1, 5, 9, and optional at 24 months
Time in low Range (TIR) <70 mg/dl	From CGM - percent time spent in the specified glycemic target range will be computed throughout study participation in 2-week increments.	Over 9 months and optional at 24 months
Time in very low Range (TIR) <55 mg/dl	From CGM - percent time spent in the specified glycemic target range will be computed throughout study participation in 2-week increments.	Over 9 months and optional at 24 months
Time in high Range (TIR) >180 mg/dl	From CGM - percent time spent in the specified glycemic target range will be computed throughout study participation 2-week increments.	Over 9 months and optional at 24 months
Time in very high Range (TIR) >250 mg/dl	From CGM - percent time spent in the specified glycemic target range will be computed throughout study participation in 2-week increments.	Over 9 months and optional at 24 months
Average Blood Glucose	From CGM - will be computed throughout study participation in 2-week increments.	Over 9 months and optional at 24 months
Coefficient of Glycemic Variation (CV)	From CGM - will be computed by dividing glucose standard deviation by glucose average throughout study participation in 2-week increments.	Over 9 months and optional at 24 months
Mean Amplitude of Glycemic Excursions (MAGE)	From CGM - will be computed using published formula throughout study participation in 2-week increments.	Over 9 months and optional at 24 months
Total Daily Insulin Dose	From insulin administration device uploads - will be computed in units per kg throughout study participation in 2-week increments.	Over 9 months and optional at 24 months
HOMA-IR (Homeostatic Model Assessment for Insulin Resistance)	Calculated from fasting blood draw [fasting insulin (µU/ml) × fasting plasma glucose (mg/dl)]/405.	1, 5, 9, and optional at 24 months
BMI	Weight divided by height squared.	1, 5, 9, and optional at 24 months
Lipid panel	Fasting blood - total, LDL and HDL cholesterol, and triglycerides.	1, 5, 9, and optional at 24 months
HDL to Triglyceride Ratio	Fasting blood	1, 5, 9, and optional at 24 months
Lipoprotein Subfractions	Fasting blood	1, 5, 9, and optional at 24 months
Inflammasome, targeted	Interleukins 1β, 17, 23, 6, 10; high sensitivity c-reactive protein; tumor necrosis factor α, interferon gamma	1, 5, 9, and optional at 24 months
Microbiome, targeted and untargeted	Extraction and sequencing will be performed by Qiagen PowerSoil DNA extraction using Qiagen's DNeasy 96 PowerSoil Pro QIAcube HT Kit (480), followed by whole genome sequencing (WGS) using a miniaturized version of the NEBNext Ultra FS II method.	1, 5, 9, and optional at 24 months
Metabolome, targeted and untargeted	Blood samples will be processed using liquid chromatography-mass spectrometry (LC-MS) and nuclear magnetic resonance (NMR). The LC-MS analyses will be carried out on a Sciex triple quadrupole mass spectrometer couple to an Exion ultra-performance LC system. The targeted analysis will utilize the Biocrates Q500 targeted metabolomics assay which quantifies more than 500 metabolites over 26 chemical classes (Biocrates Inc., Innsbruck, Austria). Data processing to yield metabolite concentrations in micromolar units will utilize the Biocrates MetIDQ software. The NMR data will be acquired on a Bruker Avance NEO 700 MHz NMR equipped with a TCI cryoprobe and a SampleXPress automatic sample changer. The data will be processed using the Chenomx NMR Processor and Profiler packages (Chenomx, Edmonton, CA) to yield quantitative data in millimolar units.	1, 5, 9, and optional at 24 months
Problem Areas in Diabetes (PAID-Ped) - child	Validated questionnaire, scored according to published standards. Scores range 0-100, higher scores indicate greater burden.	1, 5, 9, and optional at 24 months
Problem Areas in Diabetes (PAID-PR) - parent	Validated questionnaire, scored according to published standards. Scores range 0-100, higher scores indicate greater burden.	1, 5, 9, and optional at 24 months
Pediatric Quality of Life (PEDSQL) General Module - parent	Validated questionnaire, scored according to published standards. Scores range 0-100, higher scores indicate better quality of life.	1, 5, 9, and optional at 24 months
Pediatric Quality of Life (PEDSQL) General Module - child	Validated questionnaire, scored according to published standards. Scores range 0-100, higher scores indicate better quality of life.	1, 5, 9, and optional at 24 months
Pediatric Quality of Life (PEDSQL) Diabetes Module - parent	Validated questionnaire, scored according to published standards. Scores range 0-100, higher scores indicate less problems.	1, 5, 9, and optional at 24 months
Pediatric Quality of Life (PEDSQL) Diabetes Module - child	Validated questionnaire, scored according to published standards. Scores range 0-100, higher scores indicate less problems.	1, 5, 9, and optional at 24 months
Child Eating Disorder Examination Questionnaire (ChEDE-Q8)	Validated questionnaire, scored according to published standards. Scores range 0-42, higher scores are worse.	1, 5, 9, and optional at 24 months
Perceptions on Diet Management of Diabetes	Questionnaire to assess participants' and caregivers' perceptions of the influence of the diet on their diabetes management.	1, 5, 9, and optional at 24 months
Qualitative patient perspectives, interview - parent	Interviews will be held with children and caregivers separately after implementation and completion of the intervention.	optional at 9 months
Qualitative patient perspectives, interview - child	Interviews will be held with children and caregivers separately after implementation and completion of the intervention.	optional at 9 months

Other Outcome Measures

Outcome Measure	Measure Description	Time Frame
Study termination for growth deceleration	Safety Measure - Undesired weight loss or significant deceleration in longitudinal growth may warrant termination of study participation. Total number of events will be computed.	Over 9 months
Study termination for dyslipidemia	Safety Measure - LDL >200 mg/dl will trigger review of additional risk factors and may prompt diet modification to lower intake of saturated fats. If persistent, study participation may be terminated. Total number of events will be computed.	Over 9 months
Study termination for disordered eating	Safety Measure - ChEDE-Q8 diagnostic score with clinical confirmation. Total number of events.	Over 9 months
Time in tight Range (TIR) 70-140 mg/dl	From CGM - percent time spent in the specified glycemic target range will be computed throughout study participation in 2-week increments.	Over 9 months and optional at 24 months
Time above tight Range (TIR) >140 mg/dl	From CGM - percent time spent in the specified glycemic target range will be computed throughout study participation in 2-week increments.	Over 9 months and optional at 24 months
BOHB (beta-hydroxybutyrate), fasting blood concentration	Obtained at daily increasing to weekly intervals as effect modifier of beta-cell function.	Over 9 months and optional at 24 months
Growth	Safety Measure - Height standard deviation score will be calculated from serial height measurements obtained during study visits using CDC age and sex specific references.	1, 5, 9, and optional at 24 months
Growth velocity	Safety Measure - Growth velocity will be calculated from serial height measurements obtained during study visits.	1, 5, 9, and optional at 24 months
Weigh-gain	Safety Measure - Weight SDS and gain will be calculated from serial weight measures obtained during study visits with calibrated scale.	1, 5, 9, and optional at 24 months
Confirmed Ketoacidosis	Safety Measure - Defined by elevated BOHB, blood pH <7.3 and serum bicarbonate <15. Rates will be computed as total number of events divided by total patient years of follow-up.	Over 9 months and optional at 24 months
Severe Hypoglycemia	Safety Measure - Defined as blood glucose < 55 mg/dl and requiring glucagon or resulting in seizure or coma. Rates will be computed as total number of events divided by total patient years of follow-up.	Over 9 months and optional at 24 months
Diabetes Related Emergency Visits	Safety Measure - Rates will be computed as total number of events divided by total patient years of follow-up.	Over 9 months and optional at 24 months
Diabetes Related Hospitalizations	Safety Measure - Rates will be computed as total number of events divided by total patient years of follow-up.	Over 9 months and optional at 24 months

Collaborators and Investigators

• Sponsor: Boston Children's Hospital
• Collaborators: University of South Florida; Indiana University
• Investigators: Principal Investigator: Belinda Lennerz, Boston Children's Hospital

Publications and helpful links

General Publications

• Lennerz BS, Barton A, Bernstein RK, Dikeman RD, Diulus C, Hallberg S, Rhodes ET, Ebbeling CB, Westman EC, Yancy WS Jr, Ludwig DS. Management of Type 1 Diabetes With a Very Low-Carbohydrate Diet. Pediatrics. 2018 Jun;141(6):e20173349. doi: 10.1542/peds.2017-3349. Epub 2018 May 7.
• Lennerz BS, Koutnik AP, Azova S, Wolfsdorf JI, Ludwig DS. Carbohydrate restriction for diabetes: rediscovering centuries-old wisdom. J Clin Invest. 2021 Jan 4;131(1):e142246. doi: 10.1172/JCI142246.

Study record dates

Study Major Dates

• Study Start (Actual): May 31, 2024
• Primary Completion (Estimated): April 30, 2028
• Study Completion (Estimated): April 30, 2029

Study Registration Dates

• First Submitted: February 1, 2023
• First Submitted That Met QC Criteria: June 1, 2023
• First Posted (Actual): June 12, 2023

Study Record Updates

• Last Update Posted (Actual): April 24, 2025
• Last Update Submitted That Met QC Criteria: April 19, 2025
• Last Verified: April 1, 2025

More Information

Terms related to this study

• Keywords: diabetes; keto; low-carb; carbohydrate reduced
• Additional Relevant MeSH Terms: Endocrine System Diseases; Metabolic Diseases; Autoimmune Diseases; Immune System Diseases; Glucose Metabolism Disorders; Diabetes Mellitus; Diabetes Mellitus, Type 1
• Other Study ID Numbers: IRB-P00044330

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Plan Description: Upon publication, de-identified raw data for each original article will be uploaded to the appropriate an NIH maintained repository.

IPD Sharing Time Frame

Study protocol and statistical analysis plan will be shared prior to enrollment of the first participant.

De-identified data will be shared upon manuscript publication.

• IPD Sharing Access Criteria: Data will be downloadable.
• IPD Sharing Supporting Information Type: STUDY_PROTOCOL; SAP

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No