Background: Although peripheral blood adipokines and insulin levels have
been considered to be biomarkers of Alzheimer’s disease (AD), previous researches
about levels of adipokines and insulin in blood are no conclusive. We designed this meta-analysis to validate whether peripheral
adipokines and insulin can be used as a candidate biomarker in AD diagnosis. Methods: We carried out a replication
study in serum by ourselves and further conducted a meta-analysis to estimate the different levels of peripheral blood
adipokines and insulin between AD patients and controls. In the section of meta-analysis, the pooled weighted mean difference
(WMD) and 95% confidence interval (CI) was used to compare the levels of adipokines and insulin in different
groups. Results: According to our replication study, there is statistically significant lower in the levels of leptin, but conspicuous
higher in the levels of adiponectin and insulin in the blood of AD patients than controls. We finally identified
four studies for leptin, four studies for adiponectin and eleven studies for insulin. From the random-effect model, the
pooled WMD of the levels of leptin, adiponectin and insulin of AD subjects compared with the controls was -3.90 ng/ml
(95% CI: [-5.68, -2.13]), 9.42 µg/mL (95% CI: [4.21, 14.62]), and 2.86 µIU/ml (95% CI: [1.21, 4.50]), respectively.
Conclusion: Our replication study and meta-analysis support lower levels of leptin and higher levels of adiponectin and
insulin in AD patients with respect to controls, and indicate their potential values as important risk factors for AD. Further
researches that using standardized assay for leptin, adiponectin, and insulin measurement are still needed to reveal the potential
change of peripheral blood leptin, adiponectin, and insulin levels in AD participants.