Synopsis

Previous studies about resting-state functional activity reported that altered brain activity was demonstrated with type 2 diabetes and their findings were inconsistent ,which have not been quantitatively reviewed. Therefore, we conducted a quantitative meta-analysis including resting-state functional activity studies of patients with type 2 diabetes. Meta-analysis of pooled and subgroup meta-analyses found that type 2 diabetes patients showed abnormal brain activity mainly in default mode network(DMN) and visual-related regions. Our finding supports that type 2 diabetes could lead to diabetic brain activity alterations, which may correlate with cognitive impairment of type 2 diabetes patients.

Target audience: Researchers interested in functional MRI.

Purpose: The aim of this study was to conduct a quantitative meta-analysis of studies that included the amplitude of low-frequency fluctuation (ALFF) and regional homogeneity (ReHo) in patients with type 2 diabetes.

Introduction: Recent longitudinal studies have shown that individuals with insulin resistance have an associated increased risk of developing Alzheimer disease (AD) compared to healthy individuals [1]. However, the mechanism by which the insulin resistance of type 2 diabetes patients influences brain function through acting on specific brain regions has not been fully elucidated. Previous studies on resting-state functional activity have found that patients with type 2 diabetes demonstrate altered brain activity, but these findings are inconsistent and have not been quantitatively reviewed.

Methods: We selected relevant studies published before July 2016 in PubMed, Web of Knowledge, and Ovid. Six ALFF data sets and 5 ReHo data sets were included. The seed-based mapping (SDM) method was applied to quantitatively estimate regional neural activity abnormalities in type 2 diabetes patients. Seed-based mapping is a relatively new voxel-based meta-analytic tool that has been shown to be better than previous coordinate-based meta-analytical methods, such as the activation likelihood estimation (ALE) [2]. A systematic voxel-based jackknife sensitivity analysis at the whole-brain level was conducted to test the replicability of the results. And SDM has not been used in a meta-analysis of studies exploring functional activity alterations in patients with type 2 diabetes compared to healthy controls. Type 2 diabetes was diagnosed according to the latest criteria of diabetes in all included studies.

Results: A meta-analysis of 6 ALFF data sets (including 136 patients and 142 healthy controls) and 5 ReHo datasets (including 116 patients and 98 healthy controls) were conducted. Our study found that ALFF was higher in the left inferior temporal gyrus (ITG) and middle temporal gyrus (MTG) , right anterior cingulate cortex(ACC) and cerebellum of patients with type 2 diabetes compared to healthy controls (Figure 1), whereas ReHo was higher in the bilateral medial superior frontal gyrus(mSFG) and left middle frontal gyrus (MFG) (Figure 2) .

Discussion: The pooled meta-analyses revealed higher ALFF in the left MTG and ITG, right ACC and cerebellum in patients with type 2 diabetes relative to that in controls. The MTG, ITG and ACC all belong to the default mode network (DMN), which contributes significantly to normal cognitive function [3]. In addition to the regions within the DMN, increasing evidence shows that the cerebellum not only plays a pivotal role in motor functions but is also involved in cognitive, emotional, and sensory processing[4, 5]. On the other hand, the pooled meta-analyses revealed higher ReHo in the bilateral mSFG and left MFG of type 2 diabetes patients compared to that of healthy controls, which belong to the default mode network(DMN) or anatomically connected to DMN area. The altered ReHo and ALFF in these brain re­gions may be related to impairments in cognitive and visual function. However, this present meta-analysis study had limitations. First, there were only a small number of studies that met our inclusion criteria, which may reduce the ability to detect more neural activity changes. Second, the accuracy of our voxel-wise meta-analyses may be limited, as they should be based on raw statistical brain maps instead of the coordinates from published studies; however, all of the voxel-wise meta-analyses were performed previously as described here.

Conclusion: Despite the limitations mentioned above, our voxel-wise meta-analyses found higher resting-state ALFF in the left ITG, left MTG, right ACC and right cerebellum and higher ReHo in the bilateral mSFG and left MFG in patients with type 2 diabetes compared with healthy controls using the SDM method. The altered ReHo and ALFF in these brain re­gions may be related to impairments in cognitive and visual function. Furthermore, the subgroup meta-analyses suggest that SVD and type 2 diabetes have a complex interaction in the decline of cognitive and visual functions, which still needs to be explored further in a large sample with a more accurate diagnosis of the stage of type 2 diabetes.

Acknowledgements

This study was funded by National Natural Science Foundation of China (No. 31470047, 81271565).

References

1. Schrijvers, E.M., et al., Insulin metabolism and the risk of Alzheimer disease: the Rotterdam Study. Neurology, 2010. 75(22): p. 1982-7.

2. Radua, J., et al., A new meta-analytic method for neuroimaging studies that combines reported peak coordinates and statistical parametric maps. Eur Psychiatry, 2012. 27(8): p. 605-11.

3. Buckner, R.L., J.R. Andrews-Hanna, and D.L. Schacter, The brain's default network: anatomy, function, and relevance to disease. Ann N Y Acad Sci, 2008: p. 011.

4. Schmahmann, J.D., J.B. Weilburg, and J.C. Sherman, The neuropsychiatry of the cerebellum - insights from the clinic. Cerebellum, 2007. 6(3): p. 254-67.

5. Stoodley, C.J. and J.D. Schmahmann, Functional topography in the human cerebellum: a meta-analysis of neuroimaging studies. Neuroimage, 2009. 44(2): p. 489-501.