Hyperglycemic conditions, particularly diabetes mellitus (DM), are strongly associated with the incidence of cognitive impairment, including both mild cognitive impairment and dementia (Biessels and Despa, 2018; van Sloten et al., 2020). Chronic peripheral hyperinsulinemia and insulin resistance are the main features of DM, but hyperglycemia is increasingly thought to be the cause of cognitive impairment in elderly patients with DM (Umegaki et al., 2017; Tahmi et al., 2021). Several studies have shown that patients with Alzheimer’s disease (AD) have desensitized insulin signals in their brains, even in the absence of DM (Jash et al., 2020). Extensive abnormalities in insulin and insulin-like growth factor type I and II (IGF-I and IGF-II) signaling pathways in the brains of patients with AD suggest that AD may partially share characteristics with a neuroendocrine disease similar to DM (Xu et al., 2015). Chronic peripheral hyperinsulinemia can cause brain insulin resistance and defective insulin receptor activity by impairing the blood–brain barrier and insulin transport to the brain (He et al., 2020; Milstein and Ferris, 2021). Therefore, impaired brain insulin signaling may be one of the mechanisms underlying neurodegenerative disease that causes progressive impairment of learning, memory, and cognitive functions.

A previous randomized controlled trial has reported that patients with diabetes have worse cognitive performance than patients without diabetes; however, whether the incidence of dementia or cognitive impairment in patients with DM could benefit from glycemic control remains controversial (Moore et al., 2013; Biessels et al., 2021). The aim of this meta-analysis was therefore to investigate whether glycemic control in patients with DM or hyperglycemia can delay the degree of decline according to cognitive function assessment scores.

The incidence of hyperglycemia or DM and cognitive impairment both increase progressively with age. In a 10-year population-based cohort study of individuals aged 65 years and older, a modest degree of hyperglycemia was proven to independently predispose to faster cognitive decline, and glucose and hemoglobin A1c (HbA1c) were proposed as more sensitive markers of glycemia (Ganguli et al., 2020). Other studies have shown that the risk of developing cognitive decline or dementia in patients with type 2 DM is 1.25 to 2 times higher than that in patients without diabetes (Gudala et al., 2013; Xue et al., 2019). Morris et al. (2016) used the hyperinsulinemic-euglycemic clamp technique to detect systemic insulin resistance in patients with mild cognitive impairment (MCI) and AD as compared to normal controls, observing increased insulin resistance in 15 patients with cognitive impairment. Even in children with newly diagnosed type 1 DM, a single DKA episode was found to be associated with cognitive decline, particularly in subtle memory function (Ghetti et al., 2020). Although severe hypoglycemia may also lead to poor global cognition in older adults (Lacy et al., 2020), mounting clinical evidence has shown that cognitive impairment is exacerbated by hyperglycemia or DM in large populations.

The pathophysiological process of cognitive decline in patients with hyperglycemia or DM is complex and may involve common features with the pathogenesis of AD and vascular dementia (Gerstein et al., 2020), although the molecular interactions between the two diseases are not fully understood. The physio-pathological mechanisms that characterize AD, including molecular, biochemical, and signaling abnormalities, are known to be similar to those of patients with diabetes. In addition, reduced insulin signaling in the brain due to insulin dysfunction may be the primary mechanism shared by both diseases (Duarte et al., 2012). The concept of “insulin-resistant brain state (IRBS)” has thus been proposed to better describe the nature of AD (de la Monte and Wands, 2008). Insulin resistance is associated with reduced cortical insulin receptor activation, impaired clearance of amyloid-β (Aβ) oligomers, increased cerebral abnormal neurotic plaque burden, and the cerebral microvascular dysfunction which is associated with memory loss or decline of cognition (Umegaki et al., 2017; van Sloten et al., 2020). Glucotoxicity from the accumulation of advanced glycation end products (AGEs) and their precursor methylglyoxal (MGO) could induce dopaminergic dysfunction, thereby playing a role in DM-associated cognitive impairment (Pignalosa et al., 2021).

In the present study, we reviewed and evaluated the potential protective effect of blood glucose control therapy on cognitive function in patients with DM, hyperglycemia, or insulin resistance. Thirteen trials with 19,134 participants were enrolled for preliminary outcome analysis. The MMSE, MoCA, AVLT, WMS-R, 3MS, and AVLT were used as primary cognitive function assessment methods. Overall analysis showed that glycemic control significantly attenuated cognitive decline. Several recent reviews and meta-analyzes have also investigated the relationship between antidiabetic therapy and cognitive status, with inconsistent primary findings (Areosa Sastre et al., 2017; Cao et al., 2018; McMillan et al., 2018). These inconsistencies are mainly due to differences in the focus and detailed design of the studies. For example, Areosa Sastre’s review only enrolled patients diagnosed with type 2 DM, while Cao’s study enrolled patients diagnosed with Alzheimer’s disease, but was not restricted to those with DM. In McMillan’s review, only the incidence of dementia was analyzed, and the change in cognitive score which may compromise the potential cerebral protection of blood glucose control therapy was not evaluated (McMillan et al., 2018).

The present review and analysis have several limitations which should be noted. First, the enrolled studies applied different cognitive function assessment methods, resulting in heterogeneity between groups. Second, the studies had various follow-up times, and longer follow-up periods would have allowed for more accurate detection of changes in cognitive function. Third, the optimal glycemic range for the prevention of cognitive decline could not be determined in this study, and thus further exploration through high-quality clinical trials is required.

In conclusion, the current study provides evidence that glycemic control could improve the cognitive impairment through cognitive function assessment scores in patients with DM, hyperglycemia or insulin resistance.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

YL, KW, and ZW designed and carried out the study. ZG and CM participated in analyzing and interpretation of the results. YL and KW wrote the manuscript with other authors’ inputs. ZW revised the manuscript. All authors contributed to the article and approved the submitted version.

This research was partially funded by grants from Tianjin “project + team” key cultivation program (No. XC202034) and Tianjin Key Medical Discipline (Specialty) Construction Project (No.TJYXZDXK-009A).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The handling editor XG declared a shared parent affiliation with the author KW at the time of review.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.