Although synaptic loss is thought to be core to the pathophysiology of
schizophrenia, the nature, consistency and magnitude of synaptic
protein and
mRNA changes has not been systematically appraised. Our objective was thus to systematically review and meta-analyse findings. The entire PubMed database was searched for studies from inception date to the 1st of July 2017. We selected case-control postmortem studies in
schizophrenia quantifying synaptic
protein or
mRNA levels in brain tissue. The difference in
protein and
mRNA levels between cases and controls was extracted and meta-analysis conducted. Among the results, we found a significant reduction in
synaptophysin in
schizophrenia in the hippocampus (effect size: -0.65, p < 0.01), frontal (effect size: -0.36, p = 0.04), and cingulate cortices (effect size: -0.54, p = 0.02), but no significant changes for
synaptophysin in occipital and temporal cortices, and no changes for SNAP-25, PSD-95, VAMP, and
syntaxin in frontal cortex. There were insufficient studies for meta-analysis of complexins,
synapsins, rab3A and
synaptotagmin and
mRNA measures. Findings are summarised for these, which generally show reductions in SNAP-25, PSD-95,
synapsin and
rab3A protein levels in the hippocampus but inconsistency in other regions. Our findings of moderate-large reductions in
synaptophysin in hippocampus and frontal cortical regions, and a tendency for reductions in other pre- and postsynaptic
proteins in the hippocampus are consistent with models that implicate synaptic loss in
schizophrenia. However, they also identify potential differences between regions and
proteins, suggesting synaptic loss is not uniform in nature or extent.