Relationship Between Estimated Glycosylated Hemoglobin Using Flash Glucose Monitoring and Actual Measured Glycosylated Hemoglobin in a Chinese Population

Evaluation of HbA1c from CGM traces in an Indian population

Introduction

The development of continuous glucose monitoring (CGM) over the last decade has provided access to many consecutive glucose concentration measurements from patients. A standard method for estimating glycated hemoglobin (HbA1c), already established in the literature, is based on its relationship with the average blood glucose concentration (aBG). We showed that the estimates obtained using the standard method were not sufficiently reliable for an Indian population and suggested two new methods for estimating HbA1c.

Methods

Two datasets providing a total of 128 CGM and their corresponding HbA1c levels were received from two centers: Health Centre, Savitribai Phule Pune University, Pune and Joshi Hospital, Pune, from patients already diagnosed with diabetes, non-diabetes, and pre-diabetes. We filtered 112 data-sufficient CGM traces, of which 80 traces were used to construct two models using linear regression. The first model estimates HbA1c directly from the average interstitial fluid glucose concentration (aISF) of the CGM trace and the second model proceeds in two steps: first, aISF is scaled to aBG, and then aBG is converted to HbA1c via the Nathan model. Our models were tested on the remaining 32 data- sufficient traces. We also provided 95% confidence and prediction intervals for HbA1c estimates.

Results

The direct model (first model) for estimating HbA1c was HbA1cmmol/mol = 0.319 × aISFmg/dL + 16.73 and the adapted Nathan model (second model) for estimating HbA1c is HbA1cmmol/dL = 0.38 × (1.17 × ISFmg/dL) - 5.60.

Discussion

Our results show that the new equations are likely to provide better estimates of HbA1c levels than the standard model at the population level, which is especially suited for clinical epidemiology in Indian populations.

Development and economic assessment of machine learning models to predict glycosylated hemoglobin in type 2 diabetes

Background: Glycosylated hemoglobin (HbA1c) is recommended for diagnosing and monitoring type 2 diabetes. However, the monitoring frequency in real-world applications has not yet reached the recommended frequency in the guidelines. Developing machine learning models to screen patients with poor glycemic control in patients with T2D could optimize management and decrease medical service costs. Methods: This study was carried out on patients with T2D who were examined for HbA1c at the Sichuan Provincial People's Hospital from April 2018 to December 2019. Characteristics were extracted from interviews and electronic medical records. The data (excluded FBG or included FBG) were randomly divided into a training dataset and a test dataset with a radio of 8:2 after data pre-processing. Four imputing methods, four screening methods, and six machine learning algorithms were used to optimize data and develop models. Models were compared on the basis of predictive performance metrics, especially on the model benefit (MB, a confusion matrix combined with economic burden associated with therapeutic inertia). The contributions of features were interpreted using SHapley Additive exPlanation (SHAP). Finally, we validated the sample size on the best model. Results: The study included 980 patients with T2D, of whom 513 (52.3%) were defined as positive (need to perform the HbA1c test). The results indicated that the model trained in the data (included FBG) presented better forecast performance than the models that excluded the FBG value. The best model used modified random forest as the imputation method, ElasticNet as the feature screening method, and the LightGBM algorithms and had the best performance. The MB, AUC, and AUPRC of the best model, among a total of 192 trained models, were 43475.750 (¥), 0.972, 0.944, and 0.974, respectively. The FBG values, previous HbA1c values, having a rational and reasonable diet, health status scores, type of manufacturers of metformin, interval of measurement, EQ-5D scores, occupational status, and age were the most significant contributors to the prediction model. Conclusion: We found that MB could be an indicator to evaluate the model prediction performance. The proposed model performed well in identifying patients with T2D who need to undergo the HbA1c test and could help improve individualized T2D management.

Correlation Between Hemoglobin Glycation Index Measured by Continuous Glucose Monitoring With Complications in Type 1 Diabetes

OBJECTIVE

HbA1C is the "gold standard" parameter to evaluate glycemic control in diabetes; however, its correlation with mean glucose is not always perfect. The objective of this study was to correlate continuous glucose monitoring (CGM)-derived hemoglobin glycation index (HGI) with microvascular complications.

METHODS

We conducted a cross-sectional study including permanent users of CGM with type 1 diabetes mellitus or latent autoimmune diabetes of the adult. HGI was estimated, and presence of microvascular complications was compared in subgroups with high or low HGI. A logistic regression analysis to assess the contribution of high HGI to chronic kidney disease (CKD) was performed.

RESULTS

In total, 52 participants who were aged 39.7 ± 14.7 years, with 73.1% women and 15.5 years (IQR, 7.5-29 years) since diagnosis, were included; 32.7% recorded diabetic retinopathy, 25% CKD, and 19.2% neuropathy. The median HbA1C was 7.6% (60 mmol/mol) and glucose management indicator (GMI) 7.0% (53 mmol/mol). The average HGI was 0.55% ± 0.66%. The measured HbA1C was higher in the group with high HGI (8.1% [65 mmol/mol] vs 6.9% [52 mmol/mol]; P < .001), whereas GMI (7.0% [53 mmol/mol] vs 7.0% [53 mmol/mol]; P = .495) and mean glucose were similar in both groups (153 mg/dL vs 153 mg/dL; P = .564). In the high HGI group, higher occurrence of CKD (P = .016) and neuropathy were observed (P = .025). High HGI was associated with increased risk of CKD (odds ratio [OR]: 5.05; 95% CI: 1.02-24.8; P = .04) after adjusting for time since diagnosis (OR: 1.09; 95% CI: 1.02-1.16; P = .008).

CONCLUSION

High HGI measured by CGM may be a useful marker for increased risk of microvascular diabetic complications.

The impact of the COVID-19 pandemic on glycaemic control in people with diabetes: A systematic review and meta-analysis

AIM

To identify, appraise and synthesize the available evidence on the impact of the coronavirus disease 2019 (COVID-19) pandemic and lockdown (LD) on glycaemic control in people with diabetes.

MATERIALS AND METHODS

We searched multiple databases up to 2 February 2021 for studies reporting HbA1c, time in range (TIR), average or fasting glucose, severe hypoglycaemia and diabetic ketoacidosis. Data were pooled using random effects meta-analysis and are presented as mean difference (MD) with 95% confidence intervals (CI). This review was preregistered on PROSPERO (CRD42020179319).

RESULTS

We include 59 studies; 44 (n = 15 464) were included in quantitative syntheses and 15 were narratively synthesized. Pooled data were grouped by diabetes type. Results from 28 studies (n = 5048 type 1 diabetes [T1D] and combined diabetes participants) showed that TIR increased during LD compared with before LD (MD 2.74%, 95% CI 1.80% to 3.69%). Data from 10 studies (n = 1294 T1D participants) showed that TIR increased after LD compared with before LD (MD 5.14%, 95% CI 3.12% to 7.16%). Pooled results from 12 studies (n = 4810 T1D and type 2 diabetes participants) resulted in average glucose decreasing after LD compared with before LD (MD -6.86 mg/dl, 95% CI -8.54 to -5.18). Results for other outcomes, including HbA1c, were not statistically significantly different.

CONCLUSIONS

The COVID-19 pandemic was associated with small improvements across multiple outcomes of glycaemic control, although there was insufficient evidence to suggest that this led to changes in HbA1c. Most evidence came from people with access to diabetes technologies in high-income countries; more research is needed in less advantaged populations.

Correlation between glucose measurement parameters of continuous flash monitoring and HbA1c. Real life experience in Asturias

INTRODUCTION

Despite continuous glucose monitoring having been proven useful in patients with type 1 diabetes mellitus, A1C remains the gold standard for assessing disease management.

MATERIAL AND METHODS

Descriptive, retrospective study which included 252 patients, 40.5% male, mean age 44.91±14.57 years, mean duration of diabetes 22.21±13.12 years, 88.1% on basal-bolus insulin therapy and 11.9% users of continuous subcutaneous insulin infusion. Glucose measurement, analytical and anthropometric data were obtained.

RESULTS

The mean time in range was 60.18±15.60% and was associated with A1C after adjusting for age, gender, duration of diabetes, BMI, insulin regimen, %CV and time below range (ß: -0.548; p<0.01). The glucose management indicator (GMI) was 7.19±0.69% and was also associated with A1C (ß: 0.957; p<0.01) regardless of age, gender, duration of diabetes, BMI, insulin treatment, %CV and time in range. The average difference between A1C and GMI was 0.17±0.65% (-2.70-3.40%), being higher as A1C increased, in a linear and significant manner, without being influenced by the duration of diabetes or CV.

CONCLUSIONS

Although we found a positive correlation between continuous glucose monitoring glucose measurement parameters and A1C, there is still not enough evidence to replace one parameter with another.

Understanding the clinical implications of differences between glucose management indicator and glycated haemoglobin

Laboratory measured glycated haemoglobin (HbA1c) is the gold standard for assessing glycaemic control in people with diabetes and correlates with their risk of long-term complications. The emergence of continuous glucose monitoring (CGM) has highlighted limitations of HbA1c testing. HbA1c can only be reviewed infrequently and can mask the risk of hypoglycaemia or extreme glucose fluctuations. While CGM provides insights in to the risk of hypoglycaemia as well as daily fluctuations of glucose, it can also be used to calculate an estimated HbA1c that has been used as a substitute for laboratory HbA1c. However, it is evident that estimated HbA1c and HbA1c values can differ widely. The glucose management indicator (GMI), calculated exclusively from CGM data, has been proposed. It uses the same scale (% or mmol/mol) as HbA1c, but is based on short-term average glucose values, rather than long-term glucose exposure. HbA1c and GMI values differ in up to 81% of individuals by more than ±0.1% and by more than ±0.3% in 51% of cases. Here, we review the factors that define these differences, such as the time period being assessed, the variation in glycation rates and factors such as anaemia and haemoglobinopathies. Recognizing and understanding the factors that cause differences between HbA1c and GMI is an important clinical skill. In circumstances when HbA1c is elevated above GMI, further attempts at intensification of therapy based solely on the HbA1c value may increase the risk of hypoglycaemia. The observed difference between GMI and HbA1c also informs the important question about the predictive ability of GMI regarding long-term complications.

Real-Time Flash Glucose Monitoring Had Better Effects on Daily Glycemic Control Compared With Retrospective Flash Glucose Monitoring in Patients With Type 2 Diabetes on Premix Insulin Therapy

BACKGROUND AND AIMS

To compare the effects of real-time and retrospective flash glucose monitoring (FGM) on daily glycemic control and lifestyle in patients with type 2 diabetes on premix insulin therapy.

METHODS AND RESULTS

A total of 172 patients using premix insulin, with HbA1c ≥ 7.0% (56 mmol/mol), or the time below the target (TBR) ≥ 4%, or the coefficient of variation (CV) ≥36% during the screening period, were randomly assigned to retrospective FGM (n = 89) or real-time FGM group (n = 83). Another two retrospective or real-time 14-day FGMs were performed respectively, 1 month apart. Both groups received educations and medication adjustment after each FGM. Time in range (3.9~10.0 mmol/l, TIR) increased significantly after 3 months in the real-time FGM group (6.5%) compared with the retrospective FGM group (-1.1%) (p = 0.014). HbA1c decreased in both groups (both p < 0.01). Real-time FGMs increased daily exercise time compared with the retrospective group (p = 0.002).

CONCLUSIONS

Real-time FGM with visible blood glucose improves daily glycemic control and diabetes self-care behaviors better than retrospective FGM in patients with type 2 diabetes on premix insulin therapy.

CLINICAL TRIAL REGISTRATION

https://clinicaltrials.gov/NCT04847219.

Utilizing the New Glucometrics: A Practical Guide to Ambulatory Glucose Profile Interpretation

Traditional continuous glucose monitoring and flash glucose monitoring systems are proven to lower glycated haemoglobin levels, decrease the time and impact of hypoglycaemia or hyperglycaemia and, consequently, improve the quality of life for children and adults with type 1 diabetes mellitus (T1DM) and adults with type 2 diabetes mellitus (T2DM). These glucose-sensing devices can generate large amounts of glucose data that can be used to define a detailed glycaemic profile for each user, which can be compared with targets for glucose control set by an International Consensus Panel of diabetes experts. Targets have been agreed upon for adults, children and adolescents with T1DM and adults with T2DM; separate targets have been agreed upon for older adults with diabetes, who are at higher risk of hypoglycaemia, and women with pregestational T1DM during pregnancy. Along with the objective measures and targets identified by the International Consensus Panel, the dense glucose data delivered by traditional continuous glucose monitoring and flash glucose monitoring systems is used to generate an ambulatory glucose profile, which summarizes the data in a visually impactful format that can be used to identify patterns and trends in daily glucose control, including those that raise clinical concerns. In this article, we provide a practical guide on how to interpret these new glucometrics using a straightforward algorithm, and clear visual examples that demystify the process of reviewing the glycaemic health of people with T1DM or T2DM such that forward-looking goals for diabetes management can be agreed.

Cost-Effectiveness of Flash Glucose Monitoring for the Management of Patients with Type 1 and Patients with Type 2 Diabetes in China

INTRODUCTION

To compare the cost-effectiveness of flash glucose monitoring versus self-monitoring of blood glucose/point of care testing (SMBG/POCT) in both patients with type 1 and patients with type 2 diabetes (T1D/T2D) receiving insulin therapy.

METHODS

The IQVIA CORE Diabetes Model (version 9.5) was used to project the lifetime costs and health outcomes of flash glucose monitoring and SMBG/POCT from a Chinese societal perspective. We considered both hospital and individual version flash glucose monitoring to reflect the clinical practice in China. The clinical inputs leveraged the outcomes from both clinical trials and real-world studies. Cohort characteristics, intervention costs, treatment-related disutility and mortality were extracted from the literature. We also conducted scenario analyses and probabilistic sensitivity analyses to test the robustness of results.

RESULTS

Compared with SMBG/POCT using efficacy results from clinical trial, flash glucose monitoring brought the incremental costs of Chinese yuan (CNY) 58,021 and CNY 90,997 and additional quality-adjusted life years (QALYs) of 1.22 and 0.65 for patients with T1D and patients with T2D, respectively. According to the "WHO-CHOICE threshold" of three times the gross domestic product per capita in China (CNY 217,341 in 2020) as cost-effectiveness threshold, flash glucose monitoring was cost-effective for both patients with T1D and patients with T2D with incremental cost-effectiveness ratios (ICER) of CNY 47,636 and CNY 140,297 per QALY gained, respectively. According to the real-world effectiveness data, flash glucose monitoring was dominant for patients with T1D (lower costs and better effectiveness) and cost-effective for patients with T2D with an ICER of CNY 124,169 per QALY gained compared with SMBG/POCT. Scenario analyses and probabilistic sensitivity analyses confirmed the robustness of the results.

CONCLUSION

Flash glucose monitoring is likely to be considered as a cost-effective strategy compared to SMBG/POCT for Chinese patients with T1D and patients with T2D receiving insulin therapy.

Correlation between short- and mid-term hemoglobin A1c and glycemic control determined by continuous glucose monitoring

BACKGROUND

Glucose monitoring is vital for glycemic control in patients with diabetes mellitus (DM). Continuous glucose monitoring (CGM) measures whole-day glucose levels. Hemoglobin A1c (HbA1c) is a vital outcome predictor in patients with DM.

METHODS

This study investigated the relationship between HbA1c and CGM, which remained unclear hitherto. Data of patients with DM (n = 91) who received CGM and HbA1c testing (1-3 months before and after CGM) were retrospectively analyzed. Diurnal and nocturnal glucose, highest CGM data (10%, 25%, and 50%), mean amplitude of glycemic excursions (MAGE), percent coefficient of variation (%CV), and continuous overlapping net glycemic action were compared with HbA1c values before and after CGM.

RESULTS

The CGM results were significantly correlated with HbA1c values measured 1 (r = 0.69) and 2 (r = 0.39) months after CGM and 1 month (r = 0.35) before CGM. However, glucose levels recorded in CGM did not correlate with the HbA1c values 3 months after and 2-3 months before CGM. MAGE and %CV were strongly correlated with HbA1c values 1 and 2 months after CGM, respectively. Diurnal blood glucose levels were significantly correlated with HbA1c values 1-2 months before and 1 month after CGM. The nocturnal blood glucose levels were significantly correlated with HbA1c values 1-3 months before and 1-2 months after CGM.

CONCLUSIONS

CGM can predict HbA1c values within 1 month after CGM in patients with DM.

A novel telemedicine protocol improved outcomes for high-risk patients with type 1 diabetes: A 3-month quality improvement project during the COVID-19 pandemic

Our endocrinology practice needed to protect its highest-risk patients with type 1 diabetes (T1D) during the COVID-19 pandemic. To do so, we needed to identify these patients and develop a protocol to keep them out of the hospital (to limit risk of infection and conserve medical resources), and do so without in-person visits. So we used our peer-reviewed software, Diabetes Reporting, to identify 87 patients whose glucose management indicator (GMI) scores were over 9%. The GMI is a method for estimating the laboratory A1C using the patient's actual blood glucose measurements over the past 90 days. A GMI (or A1C) over 9% indicates a heightened risk of diabetic ketoacidosis (DKA) and, possibly, a slightly higher risk of severe hypoglycemia (SH), the two most common acute complications leading patients with T1D to be hospitalized. We contacted these 87 at-risk patients and enrolled them in a quality improvement project. This project consisted of additional online meetings with their doctors as well as weekly reports generated by Diabetes Reporting for three months, between March 28, 2020 and June 28, 2020. We hypothesized that this heightened communication would reduce the incidence of DKA and SH among the participants by reducing their GMI. As a comparison group, we used data from the T1D Exchange, which showed that, among patients with an A1C over 9%, 6.7% were hospitalized for DKA and 7% experienced SH leading to loss of consciousness in a three-month period. This led us to predict 6 incidences of DKA and 6 incidences of SH among our 87 participants during the three-month period. Instead, we saw 2 incidences of DKA and 1 incidence of SH. Moreover, the mean GMI of our participants dropped from 9.91% to 9.25%, a clinically-significant 0.66% improvement, which supports the conclusion that our protocol helped avoid acute complications among a cohort of at-risk patients with T1D by improving glycemic control during a time when we were limited to largely online care. This telemedicine protocol merits further research for its potential to improve and lower costs of care for patients with T1D, particularly for those at higher risk for acute complications.

Postchallenge glucose increment was associated with hemoglobin glycation index in subjects with no history of diabetes

We investigated the association between postchallenge glucose increment and hemoglobin glycation index (HGI), the difference between observed and predicted glycated hemoglobin (HbA1c), in subjects with no history of diabetes. We enrolled 1381 subjects who attended our outpatient clinic for an oral glucose tolerance test (OGTT) to screen for diabetes. HGI was defined as observed HbA1c minus predicted HbA1c. The predicted HbA1c was calculated by entering fasting plasma glucose (FPG) level into an equation [HbA1c(%)=FPG(mg/dL)*0.029+2.9686] determined from an HbA1c versus FPG regression analysis using data from an independent cohort of 2734 subjects with no history of diabetes. The association between 2-hour glucose increment and HGI was analyzed using linear regression analyses with adjustment of relevant parameters. Overall, the proportions of subjects with normal glucose tolerance, pre-diabetes, and newly diagnosed diabetes were 42.3%, 41.3%, and 16.4%, respectively. Compared with subjects who had an HGI≤0, subjects with an HGI>0 had a lower FPG (95.0±13.3 vs 98.5±15.3 mg/dL, p<0.001) but a higher 2-hour plasma glucose (151.1±52.8 vs 144.6±51.4 mg/dL, p=0.027) and 2-hour glucose increment (56.1±46.1 vs 46.1±45.0 mg/dL, p<0.001). The 2-hour glucose increment after an OGTT was independently associated with HGI (β coefficient 0.003, 95% CI 0.002 to 0.003, p<0.001). Our findings suggested that postchallenge glucose increment was independently associated with HGI in subjects with no history of diabetes.