Peripheral blood glutamate, glutamine, and GABA in Autism Spectrum disorder: A systematic review and meta-analysis of group differences with an exploratory assessment of diagnostic accuracy.

OBJECTIVE: To compare the levels of glutamate (Glu), glutamine (Gln), and γ-aminobutyric acid (GABA) in the blood, as well as the Glu/GABA ratio, between individuals with autism spectrum disorder (ASD) and the general population; and to conduct an exploratory assessment of the discriminatory ability of peripheral blood Glu, GABA, Gln, and related ratios in distinguishing between diagnosed ASD cases and healthy or typically developing controls. METHODS: A systematic search of PubMed, Web of Science, and Embase was performed. Cohort and case-control studies were eligible, and reporting followed PRISMA guidelines. Group differences were summarized as standardized mean differences (SMDs) with 95% confidence intervals. Diagnostic data were analyzed separately by biomarker. Because hierarchical bivariate models did not provide stable estimates, sensitivity and specificity for GABA and Glu were pooled separately using univariate random-effects meta-analyses and interpreted as exploratory marginal pooled estimates. Biomarkers represented by fewer than three datasets were reported descriptively. QUADAS-2 was used to assess the risk of bias in diagnostic-accuracy studies, and GRADE was used to assess the certainty of evidence. RESULTS: A total of 21 publications were included. 20 publications contributed 29 biomarker-specific group-comparison datasets, and 5 publications contributed 10 biomarker-specific diagnostic-accuracy datasets; four publications contributed to both analyses. The reported sample totals comprised 804 participants with ASD and 721 comparison participants. Compared with the control group, the ASD group had higher peripheral blood Glu levels (SMD = 1.04, 95% CI: 0.62-1.45, P < 0.001;95% PI: -2.109-4.439); no statistically significant difference was observed in peripheral blood GABA levels (SMD = 0.90, 95% CI: -1.03-2.83, P = 0.36; 95% PI: -8.471-10.272). Furthermore, there was extremely high heterogeneity among the studies, and the direction of effect was inconsistent across different studies. Peripheral blood Gln levels were lower in the ASD group (SMD = -0.78, 95% CI: -1.19 ~ -0.37, P < 0.001;95% PI: -2.439-0.881). Two studies suggested that the Glu/GABA ratio in peripheral blood may be reduced in the ASD group; however, the current evidence is insufficient to perform a reliable quantitative meta-analysis or draw definitive conclusions. Five publications contributed 10 biomarker-specific diagnostic-accuracy datasets, including 3 for GABA, 4 for Glu, and 1 each for Gln, the Glu/GABA ratio, and the Glu/Gln ratio. For GABA, exploratory marginal pooled sensitivity and specificity were 0.90 (95% CI: 0.77-0.99) and 0.89 (95% CI: 0.70-1.00), respectively. The corresponding estimates for Glu were 0.68 (95% CI: 0.28-0.97) and 0.78 (95% CI, 0.61-0.91). For Gln, the Glu/GABA ratio, and the Glu/Gln ratio, only a single study was available, so no quantitative pooling was performed. The certainty of the GRADE evidence for all indicators was very low. CONCLUSION: The diagnostic analyses yielded exploratory marginal pooled estimates of sensitivity and specificity for GABA and Glu and study-specific estimates for the remaining biomarkers, without establishing overall diagnostic accuracy. However, due to the limited number of included studies, potential selection and spectrum biases inherent in case-control designs, inconsistencies in threshold values across studies, and very low certainty of the evidence, there is currently insufficient evidence to support their use in general-population screening or clinical diagnosis.