Efficacy of Low Glycemic Index Diet vs Modified Atkins Diet and Gluten Free Casein Free Diet as add-on Therapy With Standard of Care in Children With Autism Spectrum Disorder

Children aged 3 to 8 years presenting to pediatric neurology OPD and Autism clinic in the department of Pediatrics, AIIMS, New Delhi, will be screened for eligibility. Those children fulfilling the inclusion and exclusion criteria will be enrolled.

Eligible children will be randomised into three groups. Group I would receive add-on Low Glycemic Index diet (LGIT) with standard of care, Group II would receive add-on Modified Atkins diet (MAD) with standard of care and group III would receive add-on Gluten free casein free (GFCF) diet with standard of care. The process of randomisation will be done by a person who has no role in collecting the baseline variables, applying intervention, data collection or analysis of outcomes. Block randomisation would be performed in 1:1:1 ratio with variable block size using computer generated random numbers. These will be generated by a person not involved in the study.

Allocation concealment: Serially numbered, sealed, identical envelope containing the random allocation will be kept at study site as per allocation sequence.

Blinding : The study will be an open label randomised controlled trial. The individual performing behavioural and neuro-cognitive assessments will be blinded to the group to which the patient is randomised.

Baseline demographic details will be collected and following evaluation will be done at baseline, 3 months and 6 months post intervention:

• Autism severity assessment using CARS -2, ABC
• Neurocognitive assessment using (DQ/IQ/SQ) -VSMS
• Behavioral assessment using CBCL
• Sleep related issues using CSHQ-SF

The above copyrighted scales would be administered using authorized materials obtained through appropriate licensing arrangements.

Standard of care: All three groups will receive standard of care intervention plan, which will be devised individually for each subject, including the following:

• Behavioural therapy:

• Behavioural modification techniques, Applied Behavioural Analysis
• Psycho education
• Cognitive behavioural therapy
• Activity based interventions like eye contact exercises, attention enhancement exercises, self-help skills
• Parental training
• Speech therapy
• Occupational therapy
• Sensory integration therapy
• Pharmacotherapy - Anti-psychotic medications (Risperidone/ Aripiprazole) as required in case of aggression/ irritability/ self-mutilation.

The parents of the children will be called telephonically every week to check for any issues and reinforcement provided to ensure that appropriate therapies as advised are being provided.

Dietary therapy:

During the initial 2 weeks, the following workup will be completed: Baseline nutritional assessment in terms of anthropometry, dietary history with 24-hour dietary recall, estimation of required calorie and protein intake along with understanding of individual food preferences.

A pre-diet workup will be done in the form of ECG, RFT, LFT, CBC, fasting lipid profile, Urine calcium creatinine ratio and USG KUB for nephrocalcinosis. Ferritin and 25-OH Vitamin D3 will also be tested to screen for pre-existing nutritional deficiencies and if so, will be supplemented accordingly.

Group I- LGIT with standard of care arm :

• If pre-diet work up is normal, then LGIT diet will be initiated. All medications will be changed to carbohydrate-free formulations and sugar-free toothpaste will be advised.
• Medium and high glycaemic index carbohydrates (GI>50) will be eliminated from diet. A dietary chart with low glycaemic foods (GI<50) as well as foods to be avoided will be shared with the parents and they would be taught to read food labels for the same. Total carbohydrate intake permitted will be up to 40-60 g/day (or 10% of calorie intake, whichever is lesser). The diet will be introduced gradually over a period of 2 weeks.
• A diet plan will be provided accordingly meeting recommended calorie and protein intake for age - Estimated Average Requirement (EAR), for ideal body weight as per ICMR NIN guidelines, keeping in mind the child's and family's food preferences. There will be no restriction on fluid intake. Carbohydrate-free Vitamin D3, Calcium and multivitamin supplements for B complex vitamins meeting RDA will be given, with treatment dose for the former in case of pre-existing deficiency.
• Urine ketone measurement will not be done, as it is not a standardised requirement in most studies

Group II-MAD with standard of care arm :

• If pre diet work up is normal, then MAD diet will be initiated in the ketogenic ratio of 1:1. All medications will be changed to carbohydrate-free formulations and sugar free toothpaste will be advised. Carbohydrate intake will be restricted to 10 g/day. A diet plan meeting recommended calorie and protein intake (EAR) for age (based on ideal body weight) will be devised keeping in mind individual food preferences, with no restriction on fluid intake. The diet will be introduced gradually over a period of 2 weeks. Carbohydrate-free Vitamin D3, Calcium and multivitamin supplements for B complex vitamins meeting RDA will be given, with treatment dose for the former in case of preexisting deficiency.
• Urine ketones will be checked daily using ketone dipsticks, with target of moderate-heavy ketosis indicating adequacy of the diet.

Group III -GFCF diet with standard of care arm

• Parents will be instructed on elimination of gluten containing foods (wheat, oats, rye, barley) and casein containing foods (milk and milk products) and taught to check food labels, given a list of food items to be avoided.
• Alternative food options will be provided as per dietary preferences, with a diet plan satisfying calorie and protein requirement for age as per EAR (considering ideal body weight) as per the child's and family's food preferences, with no restriction on fluid intake. The diet will be introduced gradually over a period of 2 weeks
• Vitamin D3, Calcium and Multivitamin (for B complex vitamins) supplementation will be added satisfying RDA, with treatment doses of the former in case of pre existing deficiency.

Follow up protocol

• Patients would be telephonically contacted weekly to ensure compliance and check for adverse effects
• Physical follow up visits will be scheduled at 4, 12, 20 and 24 weeks. During follow up visits, adverse effects, review of compliance via dietary records of the last 3 days prior to the visit will be done with estimation of carbohydrate intake.
• Blood investigations: CBC, RFT, LFT, fasting lipid profile, and random blood glucose will be performed at 1 month, 3 months and 6 months post diet initiation. Measurement of weight, height and BMI will be done at each physical visit.
• Follow up assessments (Autism severity, behavioural, neurocognitive and sleep assessment) will be done at 12 weeks (3 months) & 24 weeks (6 months) post diet initiation.
• Compliance rate and adverse events will be assessed from patient diary/log at each follow up visit
• Those children who are unable to tolerate LGIT/MAD/GFCF diet requiring discontinuation of diet therapy will be considered as deviates.

Data collection and Analysis: The data required for the study will be first collected in the predesigned proforma. The collected data will then be transferred to Microsoft excel and analysed using STATA. Data will be checked for normality using Shapiro Wilk. For continuous variables that are normally distributed, the unpaired-t test will be applied and if the data is non- parametric then Mann Whitney U test. For comparison within a group between the paired sample, paired-t test will be applied if the data is parametric and Wilcoxon sign rank test will be applied if the data is non parametric. For assessing the correlation between variables, Pearson correlation coefficient will be used if the data is parametric, else spearman correlation coefficient will be used. Pearson's chi-square test will be used to evaluate differences between independent groups for categorized variables provided all the values in the expected table is >5, else Fisher exact test will be applied if table is 2*2, else Fisher test with 10000 simulations will be applied. ANOVA will be used for group comparisons between the three groups.

Inclusion criteria

Children aged 3 to 8 years diagnosed with ASD according to DSM -V criteria

Exclusion criteria

• ASD in remission : DSM V criteria fulfilled only historically and/or current CARS-2 score <30
• Genetic/acquired metabolic disorders including fatty acid oxidation defects, pyruvate carboxylase deficiency, porphyria, diabetes mellitus, dyslipidemia
• Presence of kidney stones
• Children with failure to thrive/underweight/stunting, poor oral intake
• Gastroesophageal reflux
• Chronic systemic illness: renal/cardiac/hepatic/lung disease, chronic metabolic acidosis, malabsorption
• Changes in antipsychotic medications over the last 1 month
• Associated epilepsy
• Pre-existing food fads
• Food allergies: to gluten and/or casein, celiac disease
• Those who have been on Ketogenic diet/ Modified Atkins diet/ Low Glycemic Index Diet / Gluten free and/or Casein free diet in the last 6 months
• Refusal of consent