We conducted a randomized placebo-controlled trial in the Child & Adolescent Psychiatric Clinic at two hospitals in Penang, Malaysia.
Eligibility and enrolment
Patient records were screened from February till October 2012. All patients born between 31.12.2000 and 1.1.2006 with the diagnosis of ADHD were invited to participate either directly at the clinic or via a telephone call. The diagnosis of ADHD was made in accordance to the Diagnostic and Statistical Manual of Mental Disorder- IV (DSM-IV) criteria by physicians working in these hospitals using diagnostic interviews . If the patient was on ADHD medication (e.g. methylphenidate or atomoxetine), the dose of the medication had to remain unchanged for at least 3 months before enrolment. Patients with syndromes, inborn errors of metabolism, structural brain lesions, co-existing chronic liver disease and those on concurrent anticoagulants or antiplatelet drugs were excluded. Children who were unable to swallow the capsule were also excluded.
Randomization and allocation concealment
Computer-generated sequence generation and randomization were performed by an independent person not involved in the trial. Permutated block randomization of unknown size to investigators was used. Allocation concealment was by sequentially numbered sealed opaque envelopes.
Intervention and placebo
Children were randomized to receive 200 mg of TRF or placebo daily. These were given as 2 softgel capsules daily containing either 100 mg TRF per capsule or placebo (soya bean based cooking oil).
Each 100 mg capsule of TRF (Tocovid Suprabio) contained the following: alpha tocotrienol 30.76 mg, gamma tocotrienol 56.40 mg, delta tocotrienol 12.84 mg, alpha tocopherol 45.80 IU, plant squalene 25.64 mg, phytosterol complex 10.24 mg, phyto-carotenoid complex 180.00 mcg.
Blinding of the study personnel, participants and outcome assessors were achieved by identical packaging of both TRF and placebo manufactured by Hovid Bhd. They were similar in size, shape, colour, texture and outer-taste. At the end of the study, the effectiveness of blinding was assessed by asking parents if they could guess what their child received.
The investigators purchased both the TRF and placebo capsules from the manufacturer.
Conduct of study
All of the children underwent a 1 month run-in with placebo. The children were allowed to continue with their existing ADHD medication. However, those who were taking any over-the-counter nutritional supplements were instructed to discontinue these at the point of entry to the study and throughout the study. They were not informed of the run-in period with placebo. After the run-in period, children were randomized to receive either placebo or TRF. The capsules were dispensed as monthly supplies and they had to return the previous month’s bottle. Adherence was assessed by performing pill counts.
At the point of randomization (month 0), the accompanying parent or guardian would complete the NICHQ Vanderbilt ADHD Parent Rating Scale (VAPRS). The NICHQ Vanderbilt ADHD Teacher Rating Scale (VATRS) was given to the children to be handed over to their class teacher. The teachers’ scores were collected back using an enclosed addressed and stamped envelope.
The children were then seen at 3 months and 6 months with a repeat VAPRS and VATRS scoring. Their medical records were checked for alterations made to their medications. All adverse events reported during these visits were recorded. The children also had their blood taken for tocotrienol levels at recruitment and after 3 and 6 months of the study. (See Fig. 1)
This study was approved by the National Medical Research Ethics Committee, Ministry of Health Malaysia (NMRR No. 6767). Written consent was obtained from the parent or guardian. Each child was told the nature of his or her participation. This study was also registered with ClinicalTrials.gov (NCT01855984).
The primary outcome is the mean ADHD symptom score as measured using the NICHQ Vanderbilt ADHD Parent and Teacher Rating Scales (VAPRS & VATRS). The secondary outcomes were adverse events, reported changes in existing medication and correlation of plasma tocotrienol levels with the ADHD symptom score.
NICHQ Vanderbilt ADHD Parent and Teacher Rating Scale (VAPRS & VATRS)
The VAPRS and VATRS were taken from the American Academy of Paediatric ADHD Toolkit 1st Edition 2002 . The psychometric properties of the VAPRS and VATRS have been evaluated and found to be reliable and valid to be used in practice and research [20, 21]. Both these scales were translated into the local language and the translation had been previously tested in a test population showing good internal reliability .
These scales are made up of two main components—the ‘Inattention’ and ‘Hyperactivity/Impulsivity’ scores. The sum of these is known as the ‘Total Symptom Score’. The maximum score for the Total Symptom Score is 54 (maximum 27 points for each of the two components). Higher scores signify more problematic behaviour.
Tocotrienol levels were taken for all children and collected in EDTA bottles. Each sample was centrifuged immediately and the plasma frozen at −20 °C. All samples were transported in frozen state for laboratory analysis not more than 3 months from the date of collection. The plasma concentrations of alpha, gamma and delta-tocotrienols were measured using a validated high-performance liquid chromatographic (HPLC) method with fluorescence detection .
Sample size calculation
There had been no previous studies using TRF for ADHD. Sample size was calculated based on the study “Effect of Supplementation with Polyunsaturated Fatty Acids and micronutrients on learning and behaviour problems associated with child ADHD” using the Total DSM-IV scores from the Conner’s Parent Rating Scale . The scale is very similar to the VAPRS. By using the Power and Sample Size Program for T test Version 2.1 with 80 % power and 5 % level of significance, the sample size required was 63 participants for the treatment group and 22 for placebo group . However, we aimed to recruit 80 children in each group in order to have a balanced ratio and to account for 20 % drop-out rate.
All analysis was done based on intention-to-treat (ITT) principle. We used ANOVA repeated measure analysis to compare the mean VAPRS and VAPRS at baseline, 3 months and 6 months. Missing data was handled by interpolating values (for VAPRS) and iterative Markov chain Monte Carlo (MCMC) method (for VATRS). Categorical data were analysed using Chi square test. Correlation between tocotrienol levels and VAPRS Total Symptom Scores were done using Spearman Rho’s. Statistical analyses were done using SPSS Version 22  and Stata Version 12 .