摘要

IMPORTANCE Accumulating evidence suggests that brain-derived neurotrophic factor (BDNF) may be implicated in the developmental outcomes of children with autism spectrum disorder (ASD). OBJECTIVE To use meta-analysis to determine whether children with ASD have altered peripheral blood levels of BDNF. DATA SOURCE A systematic search of PubMed, PsycINFO, and Web of Science was performed for English-language literature through February 7, 2016. The search terms included brain-derived neurotrophic factor or BDNF in combination with autism, without year restriction. Two additional records were retrieved after a review of the reference lists of selected articles. STUDY SELECTION Studies were included if they provided data on peripheral blood levels of BDNF in children with ASD and healthy control children. Studies that included adults or with overlapping samples were excluded. DATA EXTRACTION AND SYNTHESIS Data were extracted by 2 independent observers from 19 included studies. Data were pooled using a random-effects model with Comprehensive Meta-analysis software. MAIN OUTCOMES AND MEASURES Blood levels of BDNF in children with ASD compared with healthy controls. Altered levels of BDNF were hypothesized to be related to ASD. RESULTS This meta-analysis included 19 studies with 2896 unique participants. Random-effects meta-analysis of all 19 studies showed that children with ASD had significantly increased peripheral blood levels of BDNF compared with healthy controls (Hedges g, 0.490; 95% CI, 0.185-0.794; P = .002). Subgroup analyses in 4 studies revealed that neonates diagnosed with ASD later in life had no association with blood levels of BDNF (Hedges g, 0.384; 95% CI, -0.244 to 1.011; P = .23), whereas children in the nonneonate ASD group (15 studies) demonstrated significantly increased BDNF levels compared with healthy controls (Hedges g, 0.524; 95% CI, 0.206 to 0.842; P = .001). Further analysis showed that children in the nonneonate ASD group had increased BDNF levels in serum (10 studies) (Hedges g, 0.564; 95% CI, 0.168 to 0.960; P = .005) but not in plasma (5 studies) (Hedges g, 0.436; 95% CI, -0.176 to 1.048; P = .16). Meta-regression analyses revealed that sample size had a moderating effect on the outcome of the meta-analysis in the nonneonate group. In addition, no publication bias was found in the meta-analysis. CONCLUSIONS AND RELEVANCE Children with ASD have increased peripheral blood levels of BDNF, strengthening the clinical evidence of an abnormal neurotrophic factor profile in this population.