Effects of the Anti-inflammatory Flavonoid Luteolin on Behavior in Children With Autism Spectrum Disorders

Background. Increasing evidence indicates that brain inflammation is important in the pathogenesis of neuropsychiatric disorders, including at least a significant proportion of subjects with Autism spectrum disorder (ASD). Natural flavonoids, such as luteolin and quercetin, exhibit potent antioxidant and anti-inflammatory activities, inhibit the release of inflammatory mediators from human mast cells, and reduce maternal interleukin 6-induced autism-like behavioral deficits related to social interactions in mice. In a case series of 37 children with ASD who took a dietary supplement containing luteolin and quercetin for 4 months reported gains in eye contact, attention and social interaction according to parental reports.

Aim. The purpose of this study was to assess the effectiveness and tolerability in white children with ASD of a dietary supplement containing 2 flavonoids, luteolin and quercetin, and the quercetin glycoside rutin.

Methods. Fifty children (42 boys and 8 girls) divided into 2 equal age groups (4-6 years old, and 7-10 years) with ASD were enrolled in a 26-week, prospective, open-label trial at the 2nd University Department of Psychiatry at "Attikon" General Hospital, Athens, Greece, the Ethics Committee of which approved the study. The parents of all subjects were informed of the study's aims, including risks versus benefits of participating and not participating as well as the inclusion and exclusion criteria, and they written consent for participation in the study.

Participants had already been diagnosed with ASD based on clinical assessments, and this diagnosis was corroborated at the 'Attikon' clinic by meeting the cutoff scores on both the DSM-IV-TR, symptom list and the ADOS algorithm. All children were medication naive. Apart from the diagnostic evaluation, the assessment also included a thorough medical evaluation comprising a physical examination and health history (including a review of allergic and gastrointestinal symptoms, as well as any food allergies or food intolerance). All concurrent interventions were thoroughly noted (type and hours), and the same was done at all visits. After meeting screening criteria, subjects were evaluated at the baseline visit, mid-trial visit at 18 weeks, and final visit at 26 weeks.

Children were administered a dietary formulation containing 2 flavonoids, luteolin (100 mg/capsule) and quercetin (70 mg/capsule), and the quercetin glycoside rutin (30 mg/capsule). The dose used was 1 softgel capsule per 10 kg (22 lb) weight per day with food for 26 weeks.

The primary outcomes were the age-equivalent scores in the 3 domains of the Vineland Adaptive Behavior Scales (VABS), communication, daily living skills, and socialization. The VABS was chosen because the impact of an agent on adaptive functioning in real life is even more important for obtaining a better quality of life than just alleviation of some symptoms. The raw scores from the interview can be also expressed as an age-equivalent score and a standard score compared with those of the subject's peers. There are also supplementary special norms for individuals with autism. Although standard scores could be more useful in subject characterization, their use as an outcome measure has been proven to be less sensitive due to floor effects and reduced variability, especially in short time periods, and thus these scores underestimate change. Conversely, scores of special norms tend to overestimate change, as a small increase in a raw score can produce a big improvement in special norm percentile rank. Thus, raw scores and age-equivalent scores seem to be the most appropriate for use as outcome measures, with the latter being more easily interpreted as change over time.

Secondary outcomes included the Aberrant Behavior Checklist (ABC), the Autism Treatment Evaluation Checklist (ATEC), and the Clinical Global Impression-Improvement score (CGI-I). To explore other possible effects of the formulation not captured from the aforementioned instruments, we chose to record any other benefits observed and reported by the parents during its use. For this, the primary clinician (K.F.) conducted telephone or in-person interviews of the parents, independently of the assessing clinician (A.T.), to discuss the possible gains of the child. CGI-I was also independently coded by the primary clinician with personal assessments as well as with information gathered by parents and, in the majority of cases, by the subjects' trainers.

Compliance was monitored by softgel capsule count and the parents' assurance that the capsules had actually been taken at each visit; in case of a capsule count <85% of the prescribed dosage at midterm and at the end of the study, the subject was excluded from the final analysis.

Adverse events were systematically recorded on an adverse event form by using scales indicating severity, relationship to the study procedures, action taken, and any therapy required.