Glycemic control refers to the maintenance of blood glucose levels (Perlmuter et al., 2008) and is closely associated with healthy cardiovascular and psychological functioning (Thayer et al., 2010), as well as optimal brain health (Garfield et al., 2021, Wennberg et al., 2016). Long-term glycemic control (over the preceding 3 months) can be indexed by glycated hemoglobin (Hb1Ac) (Canada, 2018). Studies of patients with diabetes also show that both glycemic control and brain function are linked to heart-rate variability (HRV), a surrogate of autonomic nervous (ANS) function (Napadow et al., 2008, Singh et al., 2000), suggesting that the autonomic nervous system (ANS), as indexed by HRV, plays a role in the association between glycemic control and brain function. While impaired glycemic control is typically associated with diabetes, individuals without a clinical diagnosis of diabetes also show a wide range of capacities for glycemic control. Certain individuals within the healthy adult population may be considered prediabetic (HbA1c between 5.7 and 6.4% (Cohen et al., 2010)), characterized by elevated blood glucose levels that do not meet criteria for a clinical diagnosis of diabetes (Public Health Agency of Canada, 2011). This is particularly the case among older adults. Despite the increasing importance of diabetes in modulating aging brain health (Strotmeyer, 2014), it remains unclear whether the findings in diabetes can generalize to healthy, non-clinical samples.

As HRV is strongly influenced by respiratory sinus arrhythmia, it is considered an excellent marker of parasympathetic and, to some extent, sympathetic ANS activity (Shaffer and Ginsberg, 2017). HRV indexes the variations between consecutive heartbeats (Shaffer and Ginsberg, 2017). A variety of HRV metrics exist. Root-mean-square of successive differences (Public Health Agency of Canada, 2011) (RMSSD) is strongly correlated with parasympathetic activity, making it a good ANS function (Otzenberger et al., 1998). HRV can be further specified as high-frequency (HF-HRV), an excellent marker of parasympathetic activity (Thayer et al., 2010). In studies of individuals with diabetes or prediabetes, serum glucose levels and HRV are inversely correlated, a relationship attributed to the development of cardiac autonomic neuropathy when levels of blood glucose are persistently elevated in diabetes (Almeida-Santos et al., 2016, Coopmans et al., 2020, Meyer et al., 2016, Rothberg et al., 2016, Singh et al., 2000, Ziegler et al., 2015). Cardiac autonomic neuropathy first damages the parasympathetic nervous system through the vagus nerve, and later affects the sympathetic system (Yu and Lee, 2021). In healthy individuals, the relationship between blood glucose and HRV less clear. While some studies of healthy adults report an inverse association between blood glucose and HRV, similar to findings in diabetes (Meyer et al., 2016, Takahashi et al., 2020), others found no association between blood glucose and HRV in healthy individuals (Cherkas et al., 2015, Hansen et al., 2019, Lutfi and Elhakeem, 2016, Stein et al., 2007). The discrepancy in findings across the literature may be attributable to the effects of age and sex on these relationships, which has not been previously examined in healthy individuals or clinical samples.

Moreover, advanced age is associated with higher blood glucose levels irrespective of sex (Ko et al., 2006, Yang et al., 1997). Aging is also associated with a reduction of parasympathetic activity of the ANS up to the age of 70, as indexed by lower root-mean-squared standard deviation (RMSSD) HRV (Almeida-Santos et al., 2016). With respect to sex differences, females have better long-term glycemic control compared to males until they reach menopause (Yang et al., 1997). Over the lifespan, females generally show lower HRV compared to males (Geovanini et al., 2020). HRV has also been suggested to reflect menopausal status, supporting an interaction between age and sex in modulating HRV (Ramesh et al., 2022).

Consistent with HRV as a measure of ANS activity, neuroimaging studies demonstrate correlations between fluctuations in HRV and activity within brain regions and networks that support ANS function. HRV is correlated with functional connectivity of various nodes, particularly the insula and amygdala, of the central-autonomic network (CAN) in both healthy participants and individuals with post-traumatic stress disorder (Napadow et al., 2008, Rabellino et al., 2017, Thome et al., 2017). HF-HRV also correlates with resting-state functional MRI (rs-fMRI) functional connectivity (FC) between the perigenual anterior cingulate cortex and medial prefrontal cortex, which are a part of the salience network (SN) (Jennings et al., 2016). Thus, glycemic control appears to be linked with ANS activity, as indexed by HRV, which in turn is associated with brain activity that is associated with ANS function. Despite the current literature linking HRV to brain function, whether there is a direct association between long-term glycemic control and FC of networks that support the ANS remains unknown. The roles of age and sex on these links are also unknown.

To address the knowledge gap mentioned earlier, the goal of this study was to examine the relationships among long-term glycemic control, ANS function and FC within the S-CAN and SN in healthy individuals, with a special focus on age and sex differences in these relationships. We hypothesized that poorer glycemic control, as indexed by higher HbA1c, would be associated with reduced parasympathetic activity, as indexed by RMSSD and HF-HRV (Hypothesis 1) and associated with reduced S-CAN and SN FC (Hypothesis 2). We further hypothesize that parasympathetic activity and FC would be correlated with one another (Hypothesis 3), and that the relationships among HbA1c, HRV, and FC would vary by age and sex (Hypothesis 4).