Population Pharmacokinetic Model of N-Acetylcysteine During Periods of Recurrent Hypoglycemia in Healthy Volunteers.

Recurrent hypoglycemia leads to impaired awareness of hypoglycemia where the blood glucose threshold that elicits the counterregulatory response is lowered. Hypoglycemia-induced oxidative stress is hypothesized to contribute to impaired awareness of hypoglycemia development and hypoglycemia-associated autonomic failure. Our group conducted a randomized, double-blinded, placebo-controlled, crossover study in healthy individuals undergoing experimentally induced recurrent hypoglycemia to evaluate the impact of intravenous N-acetylcysteine (NAC) during experimental hypoglycemia to preserve the counterregulatory response to subsequent hypoglycemia. The work presented herein aimed to characterize the NAC pharmacokinetics and its effects on oxidative stress. Whole blood and plasma samples were collected at specified time points during separate NAC and placebo infusions from 10 healthy volunteers. Samples were analyzed for NAC, cysteine, and glutathione (GSH) concentrations. A 2-compartment population NAC pharmacokinetic model was developed. Estimates for central compartment clearance and volume of distribution were 19.8 L/h, and 12.2 L, respectively, for a 70-kg person. Peripheral compartment clearance and volume of distribution estimates were 34.9 L/h and 13.1 L, respectively, for a 70-kg person. The PK parameters estimated here were different from those reported in the literature, suggesting a higher NAC clearance during hypoglycemic episodes. NAC leads to a significant increase in circulating cysteine concentration in a NAC concentration-dependent manner, suggesting rapid biotransformation. A transient decrease in plasma GSH was observed, supporting the hypothesis that NAC can act as a reducing agent displacing glutathione from the disulfide bond allowing for increased clearance and/or distribution of GSH.

Development and Validation of an Electronic Health Record-Based Risk Assessment Tool for Hypoglycemia in Patients With Type 2 Diabetes Mellitus.

BACKGROUND: The recent availability of high-quality data from clinical trials, together with machine learning (ML) techniques, presents exciting opportunities for developing prediction models for clinical outcomes. METHODS: As a proof-of-concept, we translated a hypoglycemia risk model derived from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study into the HypoHazardScore, a risk assessment tool applicable to electronic health record (EHR) data. To assess its performance, we conducted a 16-week clinical study at the University of Minnesota where participants (N = 40) with type 2 diabetes mellitus (T2DM) had hypoglycemia assessed prospectively by continuous glucose monitoring (CGM). RESULTS: The HypoHazardScore combines 16 risk factors commonly found within the EHR. The HypoHazardScore successfully predicted (area under the curve [AUC] = 0.723) whether participants experienced least one CGM-assessed hypoglycemic event (glucose <54 mg/dL for ≥15 minutes from two CGMs) while significantly correlating with frequency of CGM-assessed hypoglycemic events (r = 0.38) and percent time experiencing CGM-assessed hypoglycemia (r = 0.39). Compared to participants with a low HypoHazardScore (N = 19, score <4, median score of 4), participants with a high HypoHazardScore (N = 21, score ≥4) had more frequent CGM-assessed hypoglycemic events (high: 1.6 ± 2.2 events/week; low: 0.3 ± 0.5 events/week) and experienced more CGM-assessed hypoglycemia (high: 1.4% ± 2.0%; low: 0.2% ± 0.4% time) during the 16-week follow-up. CONCLUSIONS: We demonstrated that a hypoglycemia risk model can be successfully adapted from the ACCORD data to the EHR, with validation by a prospective study using CGM-assessed hypoglycemia. The HypoHazardScore represents a significant advancement toward implementing an EHR-based decision support system that can help reduce hypoglycemia in patients with T2DM.

BMI, insulin sensitivity, and cognition in early type 2 diabetes: The Glycemia Reduction Approaches in Diabetes: A Comparative Effectiveness Study.

OBJECTIVE: This study explored the association of BMI and insulin sensitivity with cognitive performance in type 2 diabetes. METHODS: A cross-sectional analysis of data from the baseline assessment of the Glycemia Reduction Approaches in Diabetes: a Comparative Effectiveness Study (GRADE) was conducted. BMI was used as a surrogate of adiposity and the Matsuda index as the measure of insulin sensitivity. Cognitive tests included the Spanish English Verbal Learning Test, the Digit Symbol Substitution Test, and the letter and animal fluency tests. RESULTS: Cognitive assessments were completed by 5018 (99.4%) of 5047 participants aged 56.7 ± 10.0 years, of whom 36.4% were female. Higher BMI and lower insulin sensitivity were related to better performance on memory and verbal fluency tests. In models including BMI and insulin sensitivity simultaneously, only higher BMI was related to better cognitive performance. CONCLUSIONS: In this study, higher BMI and lower insulin sensitivity in type 2 diabetes were cross-sectionally associated with better cognitive performance. However, only higher BMI was related to cognitive performance when both BMI and insulin sensitivity were considered simultaneously. The causality and mechanisms for this association need to be determined in future studies.

Translational aspects of glucagon: current use and future prospects.

Glucagon is secreted by the pancreatic alpha cell and has long been known to oppose insulin action. A lyophilized form of the hormone has been available to treat episodes of insulin-induced hypoglycemia in insulin-treated people with diabetes for decades, but the difficulty of use was a barrier to widespread utilization. Newer formulations of glucagon are stable at room temperature in single-use devices that many caregivers find are easier to use than the original glucagon emergency kit. In this review , we will review what is known about the role of glucagon in normal physiology and diabetes and then discuss how the research in this area has been translated into treatment for metabolic conditions.

Hypoglycaemia in type 1 diabetes mellitus: risks and practical prevention strategies.

Hypoglycaemia, which occurs when blood levels of glucose fall below what is considered a normal range, is a well-known complication of insulin therapy in individuals with type 1 diabetes mellitus. Despite advances in diabetes mellitus management, hypoglycaemia has continued to affect the majority of these individuals, leading to suboptimal care and decreased quality of life. Multiple epidemiological studies have demonstrated the risks associated with hypoglycaemic events. With this understanding, various advances have been made in therapeutics for diabetes mellitus management. Diabetes mellitus education continues to form the foundation for management and prevention of hypoglycaemia. The advent of newer diabetes mellitus technologies and newer insulins herald improvements in management strategies and hypoglycaemia prevention. Improved understanding of these newer approaches is needed to ensure delivery of safe and effective care to individuals with type 1 diabetes mellitus, leading to reductions in both the short-term and long-term morbidity and mortality associated with hypoglycaemic events.

Impaired Awareness of Hypoglycemia in Type 1 Diabetes: A Report of An NIDDK Workshop in October 2021.

Hypoglycemia remains a limiting factor in the optimal treatment of type 1 diabetes. Repeated episodes of hypoglycemia result in impaired awareness of subsequent hypoglycemic events, inducing a vicious feed-forward cycle and increasing the risk of morbidity and mortality. Why this occurs and how to manage the problem in clinical practice remain uncertain. To address the obstacles and barriers that have hindered progress in this clinically important area, the National Institute of Diabetes and Digestive and Kidney Diseases convened a workshop on 14-15 October 2021. This perspective offers a summary of this outstanding meeting, which brought clinical and basic scientists from the fields of diabetes, neuroscience, psychology, psychiatry, and imaging together, on how to best advance the field of impaired awareness of hypoglycemia and hypoglycemia in general in patients with diabetes.

Relationship Between Hypoglycemia Awareness Status on Clarke/Gold Methods and Counterregulatory Response to Hypoglycemia.

Context: Impaired awareness of hypoglycemia (IAH) is characterized by the diminished ability to perceive symptoms of hypoglycemia. Gold and Clark questionnaires are commonly used to identify patients with IAH. The relationship between IAH status on questionnaires and a person's symptom and epinephrine responses to hypoglycemia are not well understood. Objective: We aimed to examine the relationship between hypoglycemia awareness status on Clarke and Gold questionnaires with both hormonal and symptomatic responses to experimental hypoglycemia. Methods: In this university medical center study, we examined data from 78 subjects with type 1 diabetes (T1D) who completed both questionnaires and underwent a hyperinsulinemic hypoglycemic clamp (target glucose 50 mg/dL). Results: Clarke and Gold scores were highly correlated with one another (r = 0.82) and each had a moderate negative relationship with epinephrine (Clarke: r = -0.51, Gold: r = -0.50) and total symptom response (Clarke: r = -0.59, Gold: r = -0.57). However, 32% of the subjects were classified inconsistently by Clark vs Gold. A clustering analysis was done to examine how disagreement between the 2 questionnaires on IAH classification relates to epinephrine and symptoms responses during hypoglycemia. Subjects who had partial loss of symptoms or of epinephrine response were more likely to be classified inconsistently. Conclusion: Our results show that IAH classification may be discordant between Clark and Gold questionnaires and that hypoglycemia awareness status on Clarke and Gold questionnaires poorly predicts hormonal and symptomatic responses to hypoglycemia in subjects with T1D and moderate blunting of symptoms or epinephrine.

Heterogeneity in Epinephrine Response to Experimental Hypoglycemia in Type 1 Diabetes and Controls.

Context: The epinephrine response (Epi) to a first episode of hypoglycemia (HG) has been proposed to be predictive of Epi in subsequent HG and to provide insight into the risk for developing HG-associated autonomic failure (HAAF) in healthy controls (HCs). Objective: To determine if Epi and symptom response (SR) to the first episode of HG predicts who will develop HAAF after exposure to recurrent HG in volunteers with type 1 diabetes (T1D) and in HCs. Design: Review of data collected between 2013 and 2019. Setting: Academic clinical research unit. Patients or Participants: Volunteers with T1D and HCs. Interventions: Subjects participated in a preinduction protocol where they were exposed to three 2-hour episodes of clamped HG over 2 days. Data collected during clamp 1 were compared with data collected during clamp 3. Main outcome measure: Difference in Epi and SR. Results: Using the standard definition of HAAF in which HG-induced Epi during clamp 3 is at least 20% lower than during clamp 1, 21/28 HCs and 13/19 volunteers with T1D developed HAAF. Epi during clamp 1 was significantly higher in those subjects who developed HAAF than in those who did not in both groups (P = 0.02). If HAAF is defined as achieving a 20% reduction in HG-induced SR measured during clamp 3 compared with clamp 1, 10/27 HCs and 10/19 volunteers with T1D developed SR-based HAAF. Conclusion: There was heterogeneity in the response to the preinduction protocol. Epi during clamp 1 was higher than in clamp 3 in HCs and in those with T1D who developed HAAF.

Different FreeSurfer versions might generate different statistical outcomes in case-control comparison studies.

PURPOSE: Neuroimaging pipelines have long been known to generate mildly differing results depending on various factors, including software version. While considered generally acceptable and within the margin of reasonable error, little is known about their effect in common research scenarios such as inter-group comparisons between healthy controls and various pathological conditions. The aim of the presented study was to explore the differences in the inferences and statistical significances in a model situation comparing volumetric parameters between healthy controls and type 1 diabetes patients using various FreeSurfer versions. METHODS: T1- and T2-weighted structural scans of healthy controls and type 1 diabetes patients were processed with FreeSurfer 5.3, FreeSurfer 5.3 HCP, FreeSurfer 6.0 and FreeSurfer 7.1, followed by inter-group statistical comparison using outputs of individual FreeSurfer versions. RESULTS: Worryingly, FreeSurfer 5.3 detected both cortical and subcortical volume differences out of the preselected regions of interest, but newer versions such as FreeSurfer 5.3 HCP and FreeSurfer 6.0 reported only subcortical differences of lower magnitude and FreeSurfer 7.1 failed to find any statistically significant inter-group differences. CONCLUSION: Since group averages of individual FreeSurfer versions closely matched, in keeping with previous literature, the main origin of this disparity seemed to lie in substantially higher within-group variability in the model pathological condition. Ergo, until validation in common research scenarios as case-control comparison studies is included into the development process of new software suites, confirmatory analyses utilising a similar software based on analogous, but not fully equivalent principles, might be considered as supplement to careful quality control.

Monitoring the Neurotransmitter Response to Glycemic Changes Using an Advanced Magnetic Resonance Spectroscopy Protocol at 7T.

The primary excitatory and inhibitory neurotransmitters glutamate (Glu) and gamma-aminobutyric acid (GABA) are thought to be involved in the response of the brain to changes in glycemia. Therefore, their reliable measurement is critical for understanding the dynamics of these responses. The concentrations of Glu and GABA, as well as glucose (Glc) in brain tissue, can be measured in vivo using proton (1H) magnetic resonance spectroscopy (MRS). Advanced MRS methodology at ultrahigh field allows reliable monitoring of these metabolites under changing metabolic states. However, the long acquisition times needed for these experiments while maintaining blood Glc levels at predetermined targets present many challenges. We present an advanced MRS acquisition protocol that combines commercial 7T hardware (Siemens Scanner and Nova Medical head coil), BaTiO3 dielectric padding, optical motion tracking, and dynamic frequency and B0 shim updates to ensure the acquisition of reproducibly high-quality data. Data were acquired with a semi-LASER sequence [repetition time/echo time (TR/TE) = 5,000/26 ms] from volumes of interest (VOIs) in the prefrontal cortex (PFC) and hypothalamus (HTL). Five healthy volunteers were scanned to evaluate the effect of the BaTiO3 pads on B 1 + distribution. Use of BaTiO3 padding resulted in a 60% gain in signal-to-noise ratio in the PFC VOI over the acquisition without the pad. The protocol was tested in six patients with type 1 diabetes during a clamp study where euglycemic (~100 mg/dL) and hypoglycemic (~50 mg/dL) blood Glc levels were maintained in the scanner. The new protocol allowed retention of all HTL data compared with our prior experience of having to exclude approximately half of the HTL data in similar clamp experiments in the 7T scanner due to subject motion. The advanced MRS protocol showed excellent data quality (reliable quantification of 11-12 metabolites) and stability (p > 0.05 for both signal-to-noise ratio and water linewidths) between euglycemia and hypoglycemia. Decreased brain Glc levels under hypoglycemia were reliably detected in both VOIs. In addition, mean Glu level trended lower at hypoglycemia than euglycemia for both VOIs, consistent with prior observations in the occipital cortex. This protocol will allow robust mechanistic investigations of the primary neurotransmitters, Glu and GABA, under changing glycemic conditions.

Structural Alterations in Deep Brain Structures in Type 1 Diabetes.

Even though well known in type 2 diabetes, the existence of brain changes in type 1 diabetes (T1D) and both their neuroanatomical and clinical features are less well characterized. To fill the void in the current understanding of this disease, we sought to determine the possible neural correlate in long-duration T1D at several levels, including macrostructural, microstructural cerebral damage, and blood flow alterations. In this cross-sectional study, we compared a cohort of 61 patients with T1D with an average disease duration of 21 years with 54 well-matched control subjects without diabetes in a multimodal MRI protocol providing macrostructural metrics (cortical thickness and structural volumes), microstructural measures (T1-weighted/T2-weighted [T1w/T2w] ratio as a marker of myelin content, inflammation, and edema), and cerebral blood flow. Patients with T1D had higher T1w/T2w ratios in the right parahippocampal gyrus, the executive part of both putamina, both thalami, and the cerebellum. These alterations were reflected in lower putaminal and thalamic volume bilaterally. No cerebral blood flow differences between groups were found in any of these structures, suggesting nonvascular etiologies of these changes. Our findings implicate a marked nonvascular disruption in T1D of several essential neural nodes engaged in both cognitive and motor processing.

Infusion of N-acetyl cysteine during hypoglycaemia in humans does not preserve the counterregulatory response to subsequent hypoglycaemia.

AIM: Administration of N-acetyl cysteine (NAC) during hypoglycaemia will preserve the counterregulatory response to subsequent hypoglycaemia in healthy humans. METHODS: This was a randomized double-blind cross over study where humans were given either a 60-minute infusion of NAC (150 mg/kg) followed by a 4-hour infusion of NAC (50 mg/kg) or saline starting 30 minutes before the initiation of a 2-hour hypoglycaemic (HG) clamp at 8 am. After rest at euglycaemia for ~2 hours, subjects were exposed to a 2nd HG clamp at 2 pm and discharged home in euglycaemia. They returned the following day for a 3rd HG clamp at 8 am. RESULTS: Twenty-two subjects were enrolled. Eighteen subjects completed the entire protocol. The epinephrine response during clamp 3 (171 ± 247 pg/mL) following clamp 1 NAC infusion was lower than the response during the clamp 1 NAC infusion (538 ± 392 pg/mL) (clamp 3 to clamp 1 NAC: P = .0013). The symptom response during clamp 3 (7 ± 5) following clamp 1 NAC infusion was lower than the response during the clamp 1 NAC infusion (16 ± 10) (clamp 3 to clamp 1 NAC: P = .0003). Nine subjects experienced rash, pruritus or nausea during NAC infusion. CONCLUSION: We found no difference in the hormone and symptom response to experimental hypoglycaemia measured in subjects who were administered NAC as opposed to saline the day before. This observation suggests that further development of NAC as a therapy for impaired awareness of hypoglycaemia in patients with diabetes may be unwarranted.

Prevalence of microvascular and macrovascular disease in the Glycemia Reduction Approaches in Diabetes - A Comparative Effectiveness (GRADE) Study cohort.

AIMS: The Glycemia Reduction Approaches in Diabetes - A Comparative Effectiveness (GRADE) trial is a randomized clinical trial comparing glycemic effects of four diabetes medications added to metformin in type 2 diabetes (T2D). Microvascular and macrovascular diseases are secondary outcomes. We evaluated the prevalence and risk factor relationships for microvascular and macrovascular complications in the GRADE cohort at study entry. METHODS: Complication prevalence and risk factors were analyzed based on data from screening in all consenting participants meeting GRADE eligibility. Logistic regression and Z-statistics were used to assess risk factor relationships with complications. RESULTS: We enrolled 5047 T2D participants [mean age 57 years; 36% female; mean known T2D duration 4 years (all < 10 years); mean HbA1c 8.0% (∼64 mmol/mol) at screening]. Urinary albumin/creatinine ratio (ACR) ≥ 30 mg/gram was present in 15.9% participants; peripheral neuropathy (by Michigan Neuropathy Screening Instrument) in 21.5%; cardiovascular autonomic neuropathy by electrocardiography-derived indices in 9.7%; self-reported retinopathy in 1.0%. Myocardial infarction ascertained by self-report or electrocardiogram was present in 7.3%, and self-reported history of stroke in 2.0%. CONCLUSIONS: In the GRADE cohort with < 10 years of T2D and a mean HbA1c of 8.0%, diabetes complications were present in a substantial fraction of participants, more so than might otherwise have been expected.

Multiple predictively equivalent risk models for handling missing data at time of prediction: With an application in severe hypoglycemia risk prediction for type 2 diabetes.

The presence of missing data at the time of prediction limits the application of risk models in clinical and research settings. Common ways of handling missing data at the time of prediction include measuring the missing value and employing statistical methods. Measuring missing value incurs additional cost, whereas previously reported statistical methods results in reduced performance compared to when all variables are measured. To tackle these challenges, we introduce a new strategy, the MMTOP algorithm (Multiple models for Missing values at Time Of Prediction), which does not require measuring additional data elements or data imputation. Specifically, at model construction time, the MMTOP constructs multiple predictively equivalent risk models utilizing different risk factor sets. The collection of models are stored and to be queried at prediction time. To predict an individual's risk in the presence of incomplete data, the MMTOP selects the risk model based on measurement availability for that individual from the collection of predictively equivalent models and makes the risk prediction with the selected model. We illustrate the MMTOP with severe hypoglycemia (SH) risk prediction based on data from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study. We identified 77 predictively equivalent models for SH with cross-validated c-index of 0.77 ± 0.03. These models are based on 77 distinct risk factor sets containing 12-17 risk factors. In terms of handling missing data at the time of prediction, the MMTOP outperforms all four tested competitor methods and maintains consistent performance as the number of missing variables increase.

State-Dependent Changes in Brain Glycogen Metabolism.

A fundamental understanding of glycogen structure, concentration, polydispersity and turnover is critical to qualify the role of glycogen in the brain. These molecular and metabolic features are under the control of neuronal activity through the interdependent action of neuromodulatory tone, ionic homeostasis and availability of metabolic substrates, all variables that concur to define the state of the system. In this chapter, we briefly describe how glycogen responds to selected behavioral, nutritional, environmental, hormonal, developmental and pathological conditions. We argue that interpreting glycogen metabolism through the lens of brain state is an effective approach to establish the relevance of energetics in connecting molecular and cellular neurophysiology to behavior.

Hypoglycemia in diabetes: The dark side of diabetes treatment. A patient-centered review.

Hypoglycemia is a frequent occurrence in patients with diabetes who are treated with insulin and insulin secretagogues. Hypoglycemia is the limiting factor that prevents patients from achieving the glycemic control known to reduce the microvascular complications of diabetes. Recurrent episodes of hypoglycemia can lead to impaired awareness of hypoglycemia where the first symptom of a low blood sugar is unconsciousness. The fear of hypoglycemia has a significant effect on the quality of life of patients and their families. In the acute setting, hypoglycemia can kill, and clinical trials have demonstrated that a single episode of severe hypoglycemia increases the risk of subsequent mortality and cardiovascular events. Clinicians must make efforts to recognize and prevent hypoglycemia in order to prevent the adverse events associated with this event. Patient education is central to these efforts. Recent developments in glucose monitoring and drug development have provided more approaches that can be used to reduce the risk of hypoglycemia in patients with diabetes.

Central Mechanisms of Glucose Sensing and Counterregulation in Defense of Hypoglycemia.

Glucose homeostasis requires an organism to rapidly respond to changes in plasma glucose concentrations. Iatrogenic hypoglycemia as a result of treatment with insulin or sulfonylureas is the most common cause of hypoglycemia in humans and is generally only seen in patients with diabetes who take these medications. The first response to a fall in glucose is the detection of impending hypoglycemia by hypoglycemia-detecting sensors, including glucose-sensing neurons in the hypothalamus and other regions. This detection is then linked to a series of neural and hormonal responses that serve to prevent the fall in blood glucose and restore euglycemia. In this review, we discuss the current state of knowledge about central glucose sensing and how detection of a fall in glucose leads to the stimulation of counterregulatory hormone and behavior responses. We also review how diabetes and recurrent hypoglycemia impact glucose sensing and counterregulation, leading to development of impaired awareness of hypoglycemia in diabetes.

Development of a model to predict 5-year risk of severe hypoglycemia in patients with type 2 diabetes.

OBJECTIVE: We constructed a predictive model of long-term risk for severe hypoglycemia (SH: hypoglycemia requiring assistance) in patients with type 2 diabetes (T2DM). RESEARCH DESIGN AND METHODS: Data from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study (original n=10 251, n=5135 used in the current analysis), a randomized, multicenter, double 2×2 factorial design study examining the effect of glycemic, blood pressure, and lipid control on cardiovascular outcomes in patients with diagnosed T2DM, were used. Over the follow-up (3.76±1.12 years), the ACCORD participants experienced 607 incident SH events. Cox regression was used to identify the SH risk prediction model. RESULTS: We identified 17 predictors-glycemic management, age, race, education, waist circumference, medications (insulin, antihypertensive, HMG-CoA reductase inhibitors, sulfonylurea, biguanide and meglitinide), years since diabetes diagnosis, history of hypoglycemia in the last week, systolic blood pressure, diastolic blood pressure, serum creatinine, and urinary albumin creatinine ratio-to construct a prediction model for SH (c-statistic=0.782). Using this information, we derived point scores to estimate the 5-year risk for SH in individual patients with T2DM. After adjusting for other variables in the model, the three strongest predictors for SH over 5 years were intensive glycemic management (HR=2.37, 95% CI 1.99 to 2.83), insulin use (HR=2.14, 95% CI 1.77 to 2.59), and antihypertensive medication use (HR=1.90, 95% CI 1.26 to 2.86). CONCLUSION: Using the ACCORD data, we identified attributes to predict 5-year risk of SH in patients with T2DM, which warrant evaluation in broader populations to determine applicability.

Beyond the brain: do peripheral mechanisms develop impaired awareness of hypoglycemia?

The mechanisms responsible for the development of the impaired awareness of hypoglycemia often seen in insulin-treated patients with diabetes remain uncertain, but cerebral adaptations to recurrent hypoglycemia are frequently hypothesized. In this issue of the JCI, Ma et al. demonstrate that neuropeptide Y (NPY) secretion from adrenal chromaffin cells persists during exposure to recurrent hypoglycemia and activation of the sympathetic nerves at the same time that epinephrine secretion is reduced. This results in the inhibition of tyrosine hydroxylase, the rate-limiting enzyme for catecholamine synthesis. These observations suggest that a peripheral mechanism downstream from the brain contributes to the development of impaired awareness of hypoglycemia.