Blood Serotonin Levels in Autism: A Systematic Review and Meta-Analysis

Background: Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder characterized by a broad heterogeneity in clinical symptoms and developmental trajectories. Given this complexity, there has been an intensive search to identify biological markers able to aid clinicians in early diagnosis, clinical prognosis and prediction of treatment response. Fifty years ago, elevated blood serotonin (5-hydroxytryptamine [5-HT])  levels, or hyperserotonemia, was identified as a biomarker for autism and it is still one of the most consistent quantitative traits associated with the disease.

Objectives: Many studies on hyperserotonemia have been reported in the literature, measuring differences between samples of autistics and controls. Given the different  methodologies used through the years to measure peripheral serotonin, we performed a systematic review and conducted a series of meta-analyses in order to provide an overall estimate of the effect size and significance of the association between hyperserotonemia and autism, as well as to verify whether and to what extent different methodologies have influences this effect size.

Methods: First we searched the Pubmed database using a combination of keywords related to 5-HT and ASD, and reviewed only articles reporting mean and standard deviation values for autistics and controls. We then pooled studies depending on biological substrate (Whole Blood [WB] [values in ng/mL or ng/10⁹PLT]; Platelet-Rich Plasma [PRP] [values in ng/10⁹ PLT] and Platelet-Poor Plasma [PPP] [values in ng/mL]) or methodological procedures (HPLC or fluorometric assays). Finally, data from each publication were meta-analysed to generate a pooled effect size using a fixed or random effects model, depending on between-study homogeneity or heterogeneity, respectively. 

Results: Sixteen studies evaluated 5-HT levels in WB. Meta-analysis on these studies shows significantly higher serotonin levels in autistics compared to controls. An higher effect size has been found by studies measuring 5-HT levels normalized by platelet count (O.R. = 6.7; P < 0.001), as compared to 5-HT values expressed in ng/mL (O.R.= 3.14 ; P < 0.001) even if a substantial between-study heterogeneity was found (P < 0.05). Furthermore, four studies measured 5-HT levels in PRP. No between-study heterogeneity was found (p-value = 0.11), and significantly higher 5-HT levels were found in autistics (O.R.=  2.6; P < 0.001). Only three studies were selected for serotonin levels in PPP. In this case, no significant group difference was observed (O.R. = 0.54; p-value = 0.36) Finally, meta-analyses on studies using HPLC versus fluorometric assays, found a similar effect size on 5-HT measures in autism.

Conclusions: These results confirm a strong overall association between higher levels of blood serotonin with autistic disorder, as repeatedly observed in autism research. In particular, more stringent association was found with ASD when 5-HT assay were performed in WB and PRP, normalized for platelet content. Meta-analyses clearly indicate both as excellent substrates for 5-HT assessment in autism. Moreover, both HPLC and fluorometric assays, result as valid measurement methodologies. However, our study reinforces the role of serotonin as a biomarker in autism, and it could provide indicative elements also for clinical use.