Beyond HbA1c : using continuous glucose monitoring metrics to enhance interpretation of treatment effect and improve clinical decision-making.

Assessment of glycaemic outcomes in the management of Type 1 and Type 2 diabetes has been revolutionized in the past decade with the increasing availability of accurate, user-friendly continuous glucose monitoring (CGM). This advancement has brought a need for new techniques to appropriately analyse and understand the voluminous and complex CGM data for application in research-related goals and clinical guidance for individuals. Traditionally, HbA1c was established using the Diabetes Control and Complications Trial (DCCT) and other trials as the ultimate measure of glycaemic control in terms of efficacy and, by default, risk of microvascular complications of diabetes. However, it is acknowledged that HbA1c alone is inadequate at describing an individual's daily glycaemic variation and risks for hypo- and hyperglycaemia, and it does not provide the guidance needed to decrease those risks. CGM data provide means by which to characterize an individual's daily glycaemic excursions on a different time scale measured in minutes rather than months. As a consequence, clinical reports, such as the ambulatory glucose profile, increasingly include summary statistics related to averages (mean glucose, time in range) as well as markers related to glycaemic variability (coefficient of variation, standard deviation). However, there is a need to translate those metrics into specific risks that can be addressed in an actionable plan by individuals with diabetes and providers. This review presents several clinical scenarios of glycaemic outcomes from CGM data that can be analysed to describe glycaemic variability and its attendant risks of hyperglycaemia and hypoglycaemia, moving towards relevant interpretation of the complex CGM data streams.

Modular closed-loop control of diabetes.

Modularity plays a key role in many engineering systems, allowing for plug-and-play integration of components, enhancing flexibility and adaptability, and facilitating standardization. In the control of diabetes, i.e., the so-called "artificial pancreas," modularity allows for the step-wise introduction of (and regulatory approval for) algorithmic components, starting with subsystems for assured patient safety and followed by higher layer components that serve to modify the patient's basal rate in real time. In this paper, we introduce a three-layer modular architecture for the control of diabetes, consisting in a sensor/pump interface module (IM), a continuous safety module (CSM), and a real-time control module (RTCM), which separates the functions of insulin recommendation (postmeal insulin for mitigating hyperglycemia) and safety (prevention of hypoglycemia). In addition, we provide details of instances of all three layers of the architecture: the APS© serving as the IM, the safety supervision module (SSM) serving as the CSM, and the range correction module (RCM) serving as the RTCM. We evaluate the performance of the integrated system via in silico preclinical trials, demonstrating 1) the ability of the SSM to reduce the incidence of hypoglycemia under nonideal operating conditions and 2) the ability of the RCM to reduce glycemic variability.

Muscle microvascular recruitment predicts insulin sensitivity in middle-aged patients with type 1 diabetes mellitus.

AIMS/HYPOTHESIS: Insulin delivery to muscle is rate-limiting for insulin's metabolic action and is regulated by insulin's own action to increase skeletal muscle blood flow and to recruit microvasculature. Microvascular dysfunction has been observed in insulin resistant states. We investigated the relation between insulin's action to recruit microvasculature and its metabolic action in type 1 diabetes. METHODS: Near euglycaemia was obtained by an overnight insulin infusion during 17 inpatient admissions of participants with type 1 diabetes. This was followed by a 2 h 1 mU kg⁻¹ min⁻¹ euglycaemic-hyperinsulinaemic clamp. Microvascular blood volume (MBV) was assessed using contrast-enhanced ultrasound 10 min before and 30 min after starting the clamp. RESULTS: We observed that, after overnight modest hyperinsulinaemia (average ≈ 286 pmol/l), MBV was positively related to the steady-state insulin sensitivity measured during the subsequent clamp (r = 0.62, p = 0.008). The more marked hyperinsulinaemia during the clamp (average steady-state insulin ≈ 900 pmol/l) increased MBV in the more insulin resistant participants within 30 min but not in the insulin sensitive participants. The change in MBV during the clamp was negatively correlated to the insulin sensitivity (r = -0.55, p = 0.022). As a result, MBV after 30 min of marked hyperinsulinaemia was comparable between the insulin sensitive and resistant participants. CONCLUSIONS/INTERPRETATION: We conclude that moderate overnight hyperinsulinaemia recruited microvasculature in the more sensitive participants, while higher levels of plasma insulin were needed for more insulin resistant participants. This suggests that microvascular responsiveness to insulin is one determinant of metabolic insulin sensitivity in type 1 diabetes.

Anticipating the next meal using meal behavioral profiles: a hybrid model-based stochastic predictive control algorithm for T1DM.

Automatic control of Type 1 Diabetes Mellitus (T1DM) with subcutaneous (SC) measurement of glucose concentration and subcutaneous (SC) insulin infusion is of great interest within the diabetes technology research community. The main challenge with the so-called "SC-SC" route to control is sensing and actuation delay, which tends to either destabilize the system or inhibit the aggressiveness of the controller in responding to meals and exercise. Model predictive control (MPC) is one strategy for mitigating delay, where optimal insulin infusions can be given in anticipation of future meal disturbances. Unfortunately, exact prior knowledge of meals can only be assured in a clinical environment and uncertainty about when and if meals will arrive could lead to catastrophic outcomes. As a follow-on to our recent paper in the IFAC symposium on Biological and Medical Systems (MCBMS 2009), we develop a control law that can anticipate meals given a probabilistic description of the patient's eating behavior in the form of a random meal (behavioral) profile. Preclinical in silico trials using the oral glucose meal model of Dalla Man et al. show that the control strategy provides a convenient means of accounting for uncertain prior knowledge of meals without compromising patient safety, even in the event that anticipated meals are skipped.

Self-treatment of hypoglycemia while driving.

OBJECTIVE: While it is clear that progressive diabetic hypoglycemia leads to neuroglycopenia, which impairs driving, it is not clear what contributes to patients' detection and subsequent self-correction of hypoglycemia/driving impairments. Drivers with Type 1 Diabetes Mellitus (T1DM) who did and did not engage in self-treatment during experimental hypoglycemia driving are compared physiologically and psychologically. METHOD: 38 drivers with T1DM drove a sophisticated driving simulator during euglycemia and progressive hypoglycemia. Subjects were continually monitored for driving performance, EEG activity and whether they self-treated with a glucose drink. Every 5 min measures were taken of blood glucose (BG) and epinephrine levels, perceived neurogenic and neuroglycopenic symptoms and driving ability. For the four weeks prior to this hospital study, subjects participated in a field study. Using a hand-held computer just prior to routine self-measurements of BG, subjects rated neurogenic and neuroglycopenic symptoms and made judgements about BG level and ability to drive as they did in the hospital. RESULTS: Drivers who did and did not self-treat did not differ in terms of their pre-hospital exposure to hypoglycemia, their depth and rate of BG fall during experimental testing, or their epinephrine response to hypoglycemia. Subjects who self-treated detected more neurogenic and neuroglycopenic symptoms than those who did not self-treat. They also experienced less EEG defined neuroglycopenia during the progressive hypoglycemic drive as compared to those who did not self-treat. Perceived need to self-treat and EEG parameters correctly classified 88% of those who did treat from those who did not self-treat. Further, subjects who self-treated were more aware of hypoglycemia and when not to drive while hypoglycemic in the field study. CONCLUSION: There is a narrow window between a patient's detection of hypoglycemic symptoms and the need to self-treat, and neuroglycopenia, which impairs self-treatment. Consequently, drivers with T1DM should be vigilant for signs of hypoglycemia and driving impairment (e.g. trembling, uncoordination, visual difficulties) and encouraged to treat themselves immediately when they suspect hypoglycemia while driving.

Episodes of severe hypoglycemia in type 1 diabetes are preceded and followed within 48 hours by measurable disturbances in blood glucose.

This study quantifies blood glucose (BG) disturbances occurring before and after episodes of severe hypoglycemia (SH). For 6-8 months, 85 individuals with type 1 diabetes and a history of SH (age, 44+/-10 yr; 41 women and 44 men; duration of diabetes, 26+/-11 yr; hemoglobin A1c, 7.7+/-1.1%) used Lifescan One Touch BG meters for self-monitoring three to five times daily and recorded the date and time of SH episodes in diaries. For each subject, the timing of SH episodes was located in the temporal stream of SMBG readings recorded by the meter, and characteristics, including the Low BG index (LBGI), were computed in 24-h increments. In the 24-h period before the SH episode LBGI rose (P < 0.001), average BG was lower (P = 0.001), and BG variance increased (P = 0.001). In the 24 h after SH, LBGI and BGvariance remained elevated (P < 0.001), but average BG returned to baseline. These disturbances disappeared in 48 h. On the basis of LBGI we identified subjects at low, moderate, and high risk of SH, who reported, on the average, 1.7, 3.4, and 7.4 SH episodes (P < 0.005) during the study. In addition, we designed an algorithm that predicted 50% of all SH episodes that occurred in this subject group. We conclude that episodes of SH are preceded and followed by quantifiable BG disturbances, which could be used to devise warnings of imminent SH.

Progressive hypoglycemia's impact on driving simulation performance. Occurrence, awareness and correction.

OBJECTIVE: Progressive hypoglycemia leads to cognitive-motor and driving impairments. This study evaluated the blood glucose (BG) levels at which driving was impaired, impairment was detected, and corrective action was taken by subjects, along with the mechanisms underlying these three issues. RESEARCH DESIGN AND METHODS: There were 37 adults with type 1 diabetes who drove a simulator during continuous euglycemia and progressive hypoglycemia. During testing, driving performance, EEG, and corrective behaviors (drinking a soda or discontinuing driving) were continually monitored, and BG, symptom perception, and judgement concerning impairment were assessed every 5 min. Mean +/- SD euglycemia performance was used to quantify z scores for performance in three hypoglycemic ranges (4.0-3.4, 3.3-2.8, and <2.8 mmol/l). RESULTS: During all three hypoglycemic BG ranges, driving was significantly impaired, and subjects were aware of their impaired driving. However, corrective actions did not occur until BG was <2.8 mmol/l. Driving impairment was related to increased neurogenic symptoms and increased theta-wave activity. Awareness of impaired driving was associated with neuroglycopenic symptoms. increased beta-wave activity, and awareness of hypoglycemia. High beta and low theta activity and awareness of both hypoglycemia and the need to treat low BG influenced corrective behavior. CONCLUSIONS: Driving performance is significantly disrupted at relatively mild hypoglycemia, yet subjects demonstrated a hesitation to take corrective action. The longer treatment is delayed, the greater the neuroglycopenia (increased theta), which precludes corrective behaviors. Patients should treat themselves while driving as soon as low BG and/or impaired driving is suspected and should not begin driving when their BG is in the 5.0-4.0 mmol/l range without prophylactic treatment.

Biopsychobehavioral model of severe hypoglycemia. II. Understanding the risk of severe hypoglycemia.

OBJECTIVE: To evaluate the clinical/research utility of the biopsycho-behavioral model of severe hypoglycemia in differentiating patients with and without a history of severe hypoglycemia and in predicting occurrence of future severe hypoglycemia. RESEARCH DESIGN AND METHODS: A total of 93 adults with type 1 diabetes (mean age 35.8 years, duration of diabetes 16 +/- 10 years, HbA1 8.6 +/- 1.8%), 42 of whom had a recent history of recurrent severe hypoglycemia (SH) and 51 who did not (NoSH), used a handheld computer for 70 trials during 1 month recording cognitive-motor functioning, symptoms, blood glucose (BG) estimates, judgments concerning self-treatment of BG, actual BG readings, and actual treatment of low BG. For the next 6 months, patients recorded occurrence of severe hypoglycemia. RESULTS: SH patients demonstrated significantly more frequent and extreme low BG readings (low BG index), greater cognitive-motor impairments during hypoglycemia, fewer perceived symptoms of hypoglycemia, and poorer detection of hypoglycemia. SH patients were also less likely to treat their hypoglycemia with glucose and more likely to treat with general foods. Low BG index, magnitude of hypoglycemia-impaired ability to do mental subtraction, and awareness of neuroglycopenia, neurogenic symptoms, and hypoglycemia correlated separately with number of SH episodes in the subsequent 6 months. However, only low BG index, hypoglycemia-impaired ability to do mental subtraction, and awareness of hypoglycemia entered into a regression model predicting future severe hypoglycemia (R2 = 0.25, P < 0.001). CONCLUSIONS: Patients with a history of severe hypoglycemia differed on five of the seven steps of the biopsychobehavioral model of severe hypoglycemia. Helping patients with a recent history of severe hypoglycemia to reduce the frequency of their low-BG events, become more sensitive to early signs of neuroglycopenia and neurogenic symptoms, better recognize occurrence of low BG, and use fast-acting glucose more frequently in the treatment of low BG, may reduce occurrence of future severe hypoglycemia.

The effects of age and alcohol intoxication on simulated driving performance, awareness and self-restraint.

AIMS: To investigate whether, compared with middle-aged men (aged 30-50), older men (age > or =60) (i) perform more poorly on a driving simulator and (ii) are more sensitive to the effects of ethanol in terms of blood alcohol concentration (BAC) and driving performance, but more aware of their driving difficulties, and therefore exercise better driving judgement. METHODS: 14 Healthy middle-aged men (mean age 36 years) were compared with 14 healthy older men (mean age 69 years) on an interactive driving simulator, while sober and while legally intoxicated (BAC >80 mg/dl). RESULTS: Older age was associated with poorer driving performance on the simulator. While sober, older men exhibited more improper braking, slower driving, greater speed variability, fewer appropriate full stops and more crashes, and spent more time executing left turns (across oncoming traffic); all values < or =0.02. BACs > or =80 mg/dl were associated with impaired driving, with more inappropriate braking, fewer appropriate full stops and more time executing left turns (all values > or =0.02) and trends towards more speed variability, more low speed collisions and more wrong turns (values <0.1). However, similar ethanol consumption did not produce higher peak BAC or more driving impairments in older drivers. While there were no differences between age groups in terms of awareness of intoxication or driving difficulties, older men were unwilling to drive while legally intoxicated because of fear of physical injury, whereas middle-aged men were more likely to avoid driving when intoxicated due to fear of legal ramifications. CONCLUSION: While both age and legal intoxication affected driving performance, older men were no more sensitive to ethanol in terms of peak BACs, driving performance or awareness/judgement than middle-aged men.

Assessment of risk for severe hypoglycemia among adults with IDDM: validation of the low blood glucose index.

OBJECTIVE: To evaluate the clinical/research utility of the low blood glucose index (LBGI), a measure of the risk of severe hypoglycemia (SH), based on self-monitoring of blood glucose (SMBG). RESEARCH DESIGN AND METHODS: There were 96 adults with IDDM (mean age 35+/-8 years, duration of diabetes 16+/-10 years, HbA1 8.6+/-1.8%), 43 of whom had a recent history of SH (53 did not), who used memory meters for 135+/-53 SMBG readings over a month, and then for the next 6 months recorded occurrence of SH. The SMBG data were mathematically transformed, and an LBGI was computed for each patient. RESULTS: The two patient groups did not differ with respect to HbA1, insulin units per day, average blood glucose (BG) and BG variability. Patients with history of SH demonstrated a higher LBGI (P < 0.0005) and a trend to be older with longer diabetes duration. Analysis of odds for future SH classified patients into low- (LBGI <2.5), moderate- (LBGI 2.5-5), and high- (LBGI >5) risk groups. Over the following 6 months low-, moderate-, and high-risk patients reported 0.4, 2.3, and 5.2 SH episodes, respectively (P = 0.001). The frequency of future SH was predicted by the LBGI and history of SH (R2 = 40%), while HbA1, age, duration of diabetes, and BG variability were not significant predictors. CONCLUSIONS: LBGI provides an accurate assessment of risk of SH. In the traditional relationship history of SH-to-future SH, LBGI may be the missing link that reflects present risk. Because it is based on SMBG records automatically stored by many reflectance meters, the LBGI is an effective and clinically useful on-line indicator for SH risk.

Evaluating driving performance of outpatients with Alzheimer disease.

BACKGROUND: Alzheimer disease (AD) is a progressive disease, with multiple physiologic, psychologic, and social implications. A critical issue in its management is when to recommend restrictions on autonomous functioning, such as driving an automobile. This study evaluates driving performance of patients with AD and its relation to patient scores on the Mini-Mental State Exam (MMSE). METHODS: This study compared 29 outpatients with probable AD with 21 age-matched control participants on an interactive driving simulator to determine how the two groups differed and how such differences related to mental status. RESULTS: Patients with AD (1) were less likely to comprehend and operate the simulator cognitively, (2) drove off the road more often, (3) spent more time driving considerably slower than the posted speed limit, (4) spent less time driving faster than the speed limit, (5) applied less brake pressure in stop zones, (6) spent more time negotiating left turns, and (7) drove more poorly overall. There were no observed differences between AD patients and the control group in terms of crossing the midline and driving speed variability. Among the AD patients, those who could not drive the simulator because of confusion and disorientation (n = 10) had lower MMSE scores and drove fewer miles annually. Those AD patients who had stopped driving also scored lower on their MMSE but did not perform more poorly on the driving simulator. Factor analysis revealed five driving factors associated with AD, explaining 93 percent of the variance. These five factors correctly classified 27 (85 percent) of 32 AD patients compared with the control group. Of the 15 percent who were improperly classified, there were three false positives (control participants misclassified as AD patients) and two false negatives (AD patients misclassified as control participants). The computed total driving score correlated significantly with MMSE scores (r = -.403, P = 0.011). CONCLUSION: Driving simulators can provide an objective means of assessing driving safety.

Electroencephalographic and psychometric differences between boys with and without attention-deficit/Hyperactivity disorder (ADHD): a pilot study.

Attention-Deficit/Hyperactivity Disorder (ADHD) is reported to have an incidence of 3-5%, and is associated with a variety of interpersonal, academic, and social problem behaviors. There is controversy as to whether ADHD is a learned behavioral or brain dysfunction. Research has explored a variety of measures to assess behavioral and brain dysfunctions in this population, with no consistent and clearly diagnostic results. We investigated whether a new psychometric and a new electroencephalographic procedure would clearly differentiate ADHD. The psychometric was based on DSM-IV criteria and the EEG measure was based on the assumption that ADHD interferes with cognitive transition from one discrete task to another. Parents of four ADHD boys (ages 8-12) and four age- and interest-matched non-ADHD boys completed the ADHD Symptom Inventory, while their sons' EEG was monitored during viewing of a video and reading of a book. For the ADHD boys, this was repeated a second time, 3 months later, to assess test-retest reliability. Both the psychometric and the EEG measures clearly differentiated the two samples (p's < .01) with no overlap in scores, were reliable over 3 months (r = .87), and were significantly correlated with one another (r = .85). While a small sample size, these robust, related and reliable findings suggest that both the psychometric and the psychophysiological EEG measures deserve further replication and exploration.

Symmetrization of the blood glucose measurement scale and its applications.

OBJECTIVE: To introduce a data transformation that enhances the power of blood glucose data analyses. RESEARCH DESIGN AND METHODS: In the standard blood glucose scale, hypoglycemia (blood glucose, < 3.9 mmol/l) and hyperglycemia (blood glucose, > 10 mmol/l) have very different ranges, and euglycemia is not central in the entire blood glucose range (1.1-33.3 mmol/l). Consequently, the scale is not symmetric and its clinical center (blood glucose, 6-7 mmol/l) is distant from its numerical center (blood glucose, 17 mmol/l). As a result, when blood glucose readings are analyzed, the assumptions of many parametric statistics are routinely violated. We propose a logarithmic data transformation that matches the clinical and numerical center of the blood glucose scale, thus making the transformed data symmetric. RESULTS: The transformation normalized 203 out of 205 data samples containing 13,584 blood glucose readings of 127 type 1 diabetic individuals. An example illustrates that the mean and standard deviation based on transformed, rather than on raw, data better described subject's blood glucose distribution. Based on transformed data: 1) the low blood glucose index predicted the occurrence of severe hypoglycemia, while the raw blood glucose data (and glycosylated hemoglobin levels) did not; 2) the high blood glucose index correlated with the subjects' glycosylated hemoglobin (r = 0.63, P < 0.001); and 3) the low plus high blood glucose index was more sensitive than the raw data to a treatment (blood glucose awareness training) designed to reduce the range of blood glucose fluctuations. CONCLUSIONS: Using symmetrized, instead of raw, blood glucose data strengthens the existing data analysis procedures and allows for the development of new statistical techniques. It is proposed that raw blood glucose data should be routinely transformed to a symmetric distribution before using parametric statistics.

Frequency of severe hypoglycemia in insulin-dependent diabetes mellitus can be predicted from self-monitoring blood glucose data.

Severe hypoglycemia is associated with insulin-dependent diabetes mellitus and may occur more frequently as metabolic control approaches normal. The goal of this study was to determine whether the frequency of severe hypoglycemia could be predicted by the following predictor variables: 1) frequency and degree of low blood glucose (BG) readings, 2) degree of BG variability during routine self-monitoring blood glucose (SMBG) readings, and 3) level of glycemic control measured by glycosylated hemoglobin-A1 (HbA1). Seventy-eight insulin-dependent diabetes mellitus subjects from 3 different sites had their glycosylated HbA1 assayed and then performed 50 SMBG recordings during the next 2-3 weeks. Over the following 6 months, subjects recorded their severe hypoglycemic episodes (stupor or unconsciousness). There was no difference in the number of severe hypoglycemic episodes between subjects in good vs. poor metabolic control. A higher frequency of severe hypoglycemia during the subsequent 6 months was predicted by frequent and extreme low SMBG readings and variability in day to day SMBG readings. Regression analysis indicated that 44% of the variance in severe hypoglycemic episodes could be accounted for by initial measures of BG variance and the extent of low BG readings. Patients who recorded variable and frequent very low BG readings during routine SMBG were at higher risk for subsequent severe hypoglycemia. Individuals who had lower glycosylated Hb levels were not at higher risk of severe hypoglycemic episodes.