Continuous Glucose Monitoring Profiles Research Articles

Impact of Telemedicine Versus In-Person Pediatric Outpatient Type 1 Diabetes Visits on Immediate Glycemic Control: Retrospective Chart Review

Background Children and adolescents with type 1 diabetes require frequent outpatient evaluation to assess glucose trends, modify insulin doses, and screen for comorbidities. Continuous glucose monitoring (CGM) provides a detailed glycemic control assessment. Telemedicine has been increasingly used since the COVID-19 pandemic. Objective To investigate CGM profile parameter improvement immediately following pediatric outpatient diabetes visits and determine if visit modality impacted these metrics, completion of screening laboratory tests, or diabetic emergency occurrence. Methods A dual-center retrospective review of medical records assessed the CGM metrics time in range and glucose management indicator for pediatric outpatient diabetes visits during 2021. Baseline values were compared with those at 2 and 4 weeks post visit. Rates of completion of screening laboratory tests and diabetic emergencies following visits were determined. Results A total of 269 outpatient visits (41.2% telemedicine) were included. Mean time in range increased by 1.63% and 1.35% at 2 and 4 weeks post visit (P=.003 and .01, respectively). Mean glucose management indicator decreased by 0.07% and 0.06% at 2 and 4 weeks post visit (P=.003 and .02, respectively). These improvements in time in range and glucose management indicator were seen across both telemedicine visits and in-person visits without a significant difference. However, patients seen in person were 2.69 times more likely to complete screening laboratory tests (P=.03). Diabetic emergencies occurred too infrequently to analyze. Conclusions Our findings demonstrate an immediate improvement in CGM metrics following outpatient visits, regardless of modality. While statistically significant, the magnitude of these changes was small; hence, multiple visits over time would be required to achieve clinically relevant improvement. However, completion of screening laboratory tests was found to be more likely after visits occurring in person. Therefore, we suggest a hybrid approach that allows patient convenience with telemedicine but also incorporates periodic in-person assessment.

Effects of night-to-night variations in objectively measured sleep on blood glucose in healthy university students.

We examined associations between daily variations in objectively measured sleep and blood glucose in a sample of non-diabetic young adults to complement laboratory studies on how sleep affects blood glucose levels. 119 university students underwent sleep measurement using an Oura Ring 2 and continuous glucose monitoring (CGM) for up to 14 days. In 69 individuals who consumed a standardised diet across the study, multilevel models examined associations between sleep duration, timing and efficiency and daily CGM profiles. Separately, in 58 individuals, multilevel models were used to evaluate postprandial glycaemic responses to a test meal challenge on 7 days. Participants also underwent oral glucose tolerance testing once after a night of ad libitum sleep, and again following a night of sleep restriction by 1-2 hours relative to that individual's habitual sleep duration. Between-condition glucose and insulin excursions, HOMA-IR and Matsuda index were compared. Nocturnal sleep did not significantly influence following-day CGM profiles, postprandial glucose, or nocturnal mean glucose levels (all Ps>0.05). Longer sleep durations were associated with lower same-night glucose variability (all Ps<0.001). However, the range of variation of sugar levels was small and unlikely to be of functional significance. Considering naps in the analysis did not alter the findings. Sleep restriction by an average of 1.73 hours (SD=0.97) did not significantly impact excursions in glucose or insulin or insulin sensitivity the following morning (all Ps>0.05). Glucose handling in young, healthy adults may be more resilient to real-life fluctuations in sleep patterns than previously thought.

Estimating Breakfast Characteristics Using Continuous Glucose Monitoring and Machine Learning in Adults With or at Risk of Type 2 Diabetes

Background: Continuous glucose monitoring (CGM) systems allow detailed assessment of postprandial glucose responses (PPGR), offering new insights into food choices’ impact on dysglycemia. However, current approaches to analyze PPGR using a CGM require manual meal logging, limiting the scalability of CGM-driven applications like personalized nutrition and at-home diabetes risk assessment. Objective: We propose a machine learning (ML) framework to automatically identify and characterize breakfast-related PPGRs from CGM profiles in adults at risk of or living with noninsulin-treated type 2 diabetes (T2D). Methods: Our PPGR estimation framework uses a random forest ML algorithm trained on 15 adults without diabetes who wore a CGM for up to four weeks. The algorithm performance was evaluated on a held-out subset of the participants’ CGM data as well as on an external validation data set of 36 individuals at risk for or with noninsulin-treated T2D. Results: Our algorithm’s estimations of breakfast PPGRs displayed no statistically significant differences to annotated PPGRs, in terms of incremental area under the curve and glucose rise ( P &gt; .05 for both data sets), while a small difference in prebreakfast glucose was found in the nondiabetes data set ( P = .005) but not in the validation T2D data set ( P = .18). Conclusions: We designed an ML framework to automatically estimate the timing of meal events from CGM data in individuals without diabetes and in individuals at risk or with T2D. This could provide a more scalable approach for analyzing postprandial glycemia, increasing the feasibility of CGM-based precision nutrition and diabetes risk assessment applications.

Time in range prediction using the experimental mobile application in type 1 diabetes

BACKGROUND: Time in range (TIR) is a promising indicator of glycemic control used for evaluation of continuous glucose monitoring (CGM) for patients with diabetes mellitus (DM). The current problem is the assessment and prediction of TIR for patients who use self-monitoring of blood glucose (SМBG) corresponding low CGM availability for the majority of diabetic patients.AIM: To develop a predictive model of TIR for patients with T1DM based on data of the experimental mobile application.MATERIALS AND METHODS: An analysis of 1253 professional CGM profiles of patients with T1DM was performed. On the base of included records, TIR(CGM) was calculated and training models of 7-point SMBG profiles were generated. SMBG profiles’re loaded into the developed experimental mobile application that calculated standard glycemic control parameters. The dataset was divided into main and test samples (80 and 20%). For the main sample, the following methods’re used to develop predictive models: simple linear regression (SLR), multiple linear regression (MLR), artificial neural network (ANN). The effectiveness of the developed models was assessed on the test sample with the calculation of the mean absolute error (MAE), the root mean square error (RMSE).RESULTS: The 568 CGM profiles’re included in the study. TIR in the main group (n=454) — 45 [33; 65]%, in the test group (n=114) — 43 [33; 58]%. The most significant predictors of the regression models were the derived TIR (dTIR), p&lt;0,001; derived time below range level 1 (dTBR1), p&lt;0,001; standard deviation of blood glucose (SD), p=0,007. Determination coefficient for SLR (predictor: dTIR) — 0,844; for MLR (predictors: dTIR, dTBR1, SD) — 0,907. ANN multilayer perceptron models with two and one hidden layers’re developed, with the RMSE on the validation set 4,617 and 6,639%, respectively. The results of the forecast efficiency on the test sample were: dTIR: MAE — 6,82%, RMSE — 8,60%; SLR: MAE — 5,66%, RMSE — 7,34%; MLR: MAE — 4,18%, RMSE — 5,28%; ANN (2 layers): MAE — 4,14%, RMSE — 5,19%; ANN (1 layer): MAE — 4,44%, RMSE — 5,52%.CONCLUSION: ANN with two hidden layers and MLR demonstrated the best ability for TIR prediction. Further studies are required for clinical validation of developed prognostic models.

Continuous glucose monitoring profile in COVID-19 patients with and without diabetes receiving methylprednisolone.

Methylprednisolone is widely used during the COVID-19 epidemic. We aimed to evaluate the glucose profile of COVID-19 patients with and without diabetes receiving methylprednisolone. 36 patients with COVID-19 admitted to hospital were included: 17 with and 19 without diabetes. Methylprednisolone 40 mg was administered at about 9:00 a.m. Glucose levels were assessed by blinded intermittently scanned continuous glucose monitoring (isCGM) for an average of 6.8 ± 2.4 days. Excess hyperglycemia was defined as time above range (TAR) > 10.0 mmol/L (TAR>10.0) ≥ 25%, or TAR > 13.9 mmol/L (TAR>13.9) ≥ 10%. Glucose management indicator (GMI) was significantly higher than the admission glycated hemoglobin A1c (HbA1c) level in patients without diabetes [6.7 (6.1-7.0) % vs. 5.9 (5.9-6.1) %, P < 0.001], while no significant difference was found in patients with diabetes [9.0 (7.5-9.5) % vs. 8.9 (7.5-10.2) %, P > 0.05]. The difference between GMI and HbA1c (∆GMI-HbA1c) in patients without diabetes was significantly higher than in patients with diabetes [0.7 (0.2-1.0) % vs. -0.2 (-1.5-0.5) %, P = 0.005]. The circadian patterns of glucose were similar in the two groups. In patients without diabetes, excess hyperglycemia occurred in 31.6% (6/19) of participants, with 31.6% (6/19) having a TAR>10.0 ≥ 25%, while 21.1% (4/19) had a TAR>13.9 ≥ 10%. The impact of methylprednisolone on glycemia was more pronounced in COVID-19 patients without diabetes, compared to those with diabetes. A significant burden of methylprednisolone-induced hyperglycemia was observed in patients without diabetes.

Novel Detection and Progression Markers for Diabetes Based on Continuous Glucose Monitoring Data Dynamics.

Develop a diabetes diagnostic tool based on two markers of continuous glucose monitoring (CGM) dynamics: CGM entropy rate (ER) and Poincaré plot (PP) ellipse area (S). 5,754 daily CGM profiles from 843 individuals with type 1, type 2 diabetes, or healthy individuals with or without islet autoantibody status were used to compute two individual dynamic markers: ER (in bits per transition; BPT) of daily probability matrices describing CGM transitions between eight glycemic states, and the area S (mg2/dL2) of individual CGM PP ellipses using standard PP descriptors. The Youden's index was used to determine "optimal" cut-points for ER and S for health vs. diabetes (case 1); type 1 vs. type 2 (case 2); and low vs. high type 1 immunological risk (case 3). The markers' discriminative power was assessed through the area under the receiver operating characteristics curves (AUC). Optimal cut-off points were determined for ER and S for each of the three cases. ER and S discriminated case 1 with AUC = 0.98 (95% CI: 0.97-0.99) and AUC = 0.99 (95% CI: 0.99-1.00), respectively, (cut-offs ERcase1 = 0.76 BPT, Scase1 = 1993.91 mg2/dL2), case 2 with AUC = 0.81 (95% CI, 0.77-0.84) and AUC = 0.76 (95% CI, 0.72-0.81), respectively (ERcase2 = 1.00 BPT, Scase2 = 5112.98 mg2/dL2), and case 3 with AUC = 0.81 (95% CI, 0.77-0.84) and AUC = 0.76 (95% CI, 0.72-0.81), respectively (ERcase3 = 0.52 BPT, Scase3 = 923.65 mg2/dL2). CGM dynamics markers can be an alternative to fasting plasma glucose or glucose tolerance testing and identifying individuals at higher immunological risk of progressing to type 1 diabetes.

Ultra rapid lispro improves postprandial glucose control versus lispro in combination with basal insulin: a study based on CGM in type 2 diabetes in China.

To evaluate the efficacy and safety of URLi (ultra rapid lispro insulin) compared to insulin lispro as bolus insulin with basal insulin using CGM in the individuals with type 2 diabetes(T2D) in China. This was a double-blind, randomized, parallel, prospective, phase 3 study. Subjects with uncontrolled T2D were recruited and randomized 1:2 into the insulin lispro and URLi groups. Subjects received a consistent basal insulin regimen during the study and self-administered insulin lispro or URLi before each meal throughout the treatment period. Subjects underwent a 3-day continuous glucose monitoring (CGM) at the baseline and endpoint respectively, and then CGM data were analyzed. The primary endpoint was to compare the difference in postprandial glucose (PPG) control using CGM between the two groups. A total of 57 subjects with T2D completed the study. Our CGM data showed that postprandial glucose excursions after breakfast (BPPGE) in the URLi group was lower than that in the insulin lispro group (1.59 ± 1.57 mmol/L vs 2.51 ± 1.73 mmol/L, p = 0.046). 1-hour PPG was observed to decrease more in the URLi group than that in the insulin lispro group (-1.37 ± 3.28 mmol/L vs 0.24 ± 2.58 mmol/L, p = 0.047). 2-hour PPG was observed to decrease more in the URLi group than that in the insulin lispro group (-1.12 ± 4.00 mmol/L vs 1.22 ± 2.90 mmol/L, p = 0.021). The mean HbA1c level decreased by 1.1% in the URLi group and 0.99% in the insulin lispro group, with no treatment difference (p = 0.642). In the CGM profile, TBR was not significantly different between the two groups (p = 0.743). The weight gain also did not differ between the two groups (p = 0.303). URLi can control breakfast PPG better than insulin lispro in adults with T2D in China, while it is non-inferior in improving HbA1c. The incidence of hypoglycemic and weight gain were similar between the two groups.

Impact of Glucocorticoid Replacement Therapy on Nocturnal Hypoglycemia in Adrenal Insufficiency: An Analysis of Multiple Case Studies.

Adrenal insufficiency (AI) is a hormonal disorder characterized by insufficient glucocorticoid production. Nocturnal hypoglycemia (NH) occurs in patients with AI. However,the effect of glucocorticoid replacement therapy (GCRT) on AI and NH remains unclear. This study aimed to investigate the relationship between AI and NH by evaluating the impact of GCRT on NH in patients newly diagnosed with AI. The present study was conducted between October 2018 and December 2022 at the Department of Diabetes, Metabolism and Endocrinology of the Tokyo Rosai Hospital, Japan. In total, 15 patients aged ≥18 years with newly diagnosed AI or NH were included in this study. The NH frequency was measured using continuous glucose monitoring (CGM). The primary outcome was the change in NH frequency before and after the GCRT intervention. GCRT significantly decreased NH frequency. Severe NH frequency and minimum nocturnal glucose levels changed significantly while fasting blood glucose and glycated hemoglobin levels did not change significantly. GCRT intervention improved CGM profiles' time below range, time in range, and average daily risk range. The present study suggests that GCRT can help newly diagnosed patients with AI manage NH. These findings show that CGM can detect NH in patients with newly diagnosed AI, determine the optimal GCRT dosage, and hence prevent an impaired quality of life and even serious adverse effects in these patients. Further large multicenter studies should validate these findings and delve deeper into the mechanistic link between AI and NH.

Q-Score Complements the Time in Range in the Evaluation of Short-Term Glycemic Control.

The Q-Score is a single-number composite metric that is constructed based on the following components: central glycemic tendency, hyperglycemia, hypoglycemia, and intra- and interday variability. Herein, we refined the Q-Score for the screening and analysis of short-term glycemic control using continuous glucose monitoring (CGM) profiles. Continuous glucose monitoring profiles were obtained from noninterventional, retrospective cross-sectional studies. The upper limit of the Q-Score component hyperglycemia' that is, the time above target range (TAR), was adjusted from 8.9 to 10 mmol/L (n = 1562 three-day-sensor profiles). A total of 302 people with diabetes mellitus treated with intermittent CGM for ≥14 days were enrolled. The time to stability was determined via correlation-based analysis. There was a strong correlation between the Q-Scores of the two TARs, that is, 8.9 and 10 mmol/L (Q-ScoreTAR10 = -0.03 + 1.00 Q-ScoreTAR8.9, r = .997, p < .001). The times to stability of the Q-Score and TIR were 10 and 12 days, respectively. The Q-Score was correlated with fructosamine concentrations, the glucose management indicator (GMI), the time in range (TIR), and the glycemic risk index (GRI) (r = .698, .887, -.874, and .941), respectively. The number of Q-Score components above the target increased as the TIR decreased, from two (1.7 ± 0.9) in CGM profiles with a TIR between 70% and 80% to four (3.9 ± 0.5) in the majority of the CGM profiles with a TIR below 50%. A conversion matrix between the Q-Score and glycemic indices was developed. The Q-Score is a tool for assessing short-term glycemic control. The Q-Score can be translated into clinician opinion using the GRI.

Continuous Glucose Monitoring Profiles and Health Outcomes after Dapagliflozin Plus Saxagliptin vs Insulin Glargine.

Glycemic variability and hypoglycemia during diabetes treatment may impact therapeutic effectiveness and safety, even when glycated hemoglobin (HbA1c) reduction is comparable between therapies. We employed masked continuous glucose monitoring (CGM) during a randomized trial of dapagliflozin plus saxagliptin (DAPA+SAXA) vs insulin glargine (INS) to compare glucose variability and patient-reported outcomes (PROs). 24-week sub-study of a randomized, open-label, two-arm, parallel-group, phase 3b study. Multicenter study (112 centers in 11 countries). 283 adults with type 2 diabetes (T2D) inadequately controlled with metformin ± sulfonylurea. DAPA+SAXA vs INS. Changes in CGM profiles, HbA1c, and PROs. Changes from baseline in HbA1c with DAPA+SAXA were similar to those observed with INS, with mean difference [95% CI] between decreases of -0.12% [-0.37 to 0.12%], P = .33. CGM analytics were more favorable for DAPA+SAXA, including greater percent time in range (> 3.9 and ≤ 10 mmol/L; 34.3 ± 1.9 vs 28.5 ± 1.9%, P = .033), lower percent time with nocturnal hypoglycemia (area under the curve ≤ 3.9 mmol/L; 0.6 ± 0.5 vs 2.7 ± 0.5%, P = .007), and smaller mean amplitude of glycemic excursions (-0.7 ± 0.1 vs -0.3 ± 0.1 mmol/L, P = .017). Improvements in CGM were associated with greater satisfaction, better body weight image, less weight interference, and improved mental and emotional well-being. DAPA+SAXA and INS were equally effective in reducing HbA1c at 24 weeks, but people with T2D treated with DAPA+SAXA achieved greater time in range, greater reductions in glycemic excursions and variability, less time with hypoglycemia, and improved patient-reported health outcomes.

Clinically relevant stratification of patients with type 2 diabetes by using continuous glucose monitoring data.

The wealth of data generated by continuous glucose monitoring (CGM) provides new opportunities for revealing heterogeneities in patients with type 2 diabetes mellitus (T2DM). We aimed to develop a method using CGM data to discover T2DM subtypes and investigate their relationship with clinical phenotypes and microvascular complications. The data from 3119 patients with T2DM who wore blinded CGM at an academic medical centre was collected, and a glucose symbolic pattern (GSP) metric was created that combined knowledge-based temporal abstraction with numerical vectorization. The k-means clustering was applied to GSP to obtain subgroups of patients with T2DM. Clinical characteristics and the presence of diabetic retinopathy and albuminuria were compared among the subgroups. The findings were validated in an independent population comprising 773 patients with T2DM. By using GSP, four subgroups were identified with distinct features in CGM profiles and parameters. Moreover, the clustered subgroups differed significantly in clinical phenotypes, including indices of pancreatic β-cell function and insulin resistance (all p < .001). After adjusting for confounders, group C (the most insulin resistant) had a significantly higher risk of albuminuria (odds ratio = 1.24, 95% confidence interval: 1.03-1.39) relative to group D, which had the best glucose control. These findings were confirmed in the validation set. Subtyping patients with T2DM using CGM data may help identify high-risk patients for microvascular complications and provide insights into the underlying pathophysiology. This method may help refine clinically meaningful stratification of patients with T2DM and inform personalized diabetes care.

Association of glycemic variability and time in range with lipid profile in type 1 diabetes

PurposeHyperglycemia is associated with abnormalities of lipoproteins. The aim of this study was to analyze, in patients with Type 1 Diabetes (T1D), the association of glycemic control with lipid profile, focusing on glycemic variability and time in range obtained from Continuous Glucose Monitoring (CGM).MethodsWe performed a retrospective cohort in patients with T1D. We analyzed clinical parameters, HbA1c, CGM and lipid profile in two moments 6 to 18 months apart. We evaluated the association of HbA1c and CGM metrics with lipid profile in cross-sectional (n = 242) and longitudinal (n = 90) analyses.ResultsThe mean age of the study population was 36.6 ± 12.6 years, 51.7% were male, and the mean diabetes duration was 16.8 ± 10.3 years. In the cross-sectional analysis, higher HbA1c, higher glucose management indicator (GMI), higher time above range and lower time in range were associated with higher triglyceride levels. In the longitudinal analysis, an increase in time below range was associated with a decrease of HDL cholesterol. In both analyses, an increase in the coefficient of variability (CV) was associated with a significant decrease of HDL cholesterol. HbA1c and CGM were not associated with total cholesterol or LDL cholesterol.ConclusionsWe observed a negative association between CV and HDL cholesterol levels and a positive association between hyperglycemia metrics and triglyceride levels. These findings suggest that CGM parameters may be a helpful tool to guide the improvement of both glycemic control and lipid profile in T1D.

170-LB: Do Telemedicine vs. In-Person Outpatient Type 1 Diabetes Visits Immediately Improve Glycemic Control?

Background: Children with type 1 diabetes (T1D) require frequent outpatient evaluation to assess glucose trends, modify insulin doses, and stay up to date with screening. Continuous glucose monitoring (CGM) has become more prevalent in recent years and provides a detailed glycemic control assessment. Telemedicine is increasingly utilized in lieu of in-person visits, especially since the onset of the COVID-19 pandemic. Objective: To assess a difference in CGM profile parameters in the periods immediately before and after T1D follow-up visits and determine whether being seen by telemedicine versus in-person impacts this difference. Methods: A dual center retrospective chart review was conducted including outpatient T1D follow up visits for patients aged 5-21 years old utilizing Dexcom CGM during the 2021 calendar year. CGM profile parameters assessed were Time in Range (TIR) and Glucose Management Indicator (GMI) during the period 4-weeks leading up to each outpatient visit, compared to the 2-week and 4-week period immediately following. Results: A total of 269 outpatient visits (41.2% telemedicine) among 135 patients were included from both study sites. Mean TIR increased by 1.63 and 1.35% during the 2-week and 4-week period after each visit, respectively (p = 0.003, 0.013). Mean GMI also improved, decreasing by 0.07% and 0.06% during the 2-week and 4-week period after each visit, respectively (p = 0.003, 0.021). These changes in TIR and GMI were not statistically different between telemedicine versus in-person visits (p = 0.40, 0.29). Conclusions: This study detected a small, but statistically significant improvement in CGM profile parameters immediately after outpatient visits. Telemedicine versus in-person visits were comparable. These findings suggest that multiple visits over time are needed to obtain a sustainable, clinically relevant change in glycemic control, and telemedicine can be utilized just as effectively as traditional visits to achieve this goal. Disclosure C. J. Ferber: None. H. Wilhalme: None. S. Mittelman: None. T. Moin: None. R. A. Hicks: None.

Clinically-Similar Clusters of Daily CGM Profiles: Tracking the Progression of Glycemic Control Over Time.

The adoption of CGM results in vast amounts of data, but their interpretation is still more art than exact science. The International Consensus on Time in Range (TIR) proposed the widely accepted TIR system of metrics, which we now take forward by introducing a finite and fixed set of clinically-similar clusters (CSCs), such that the TIR metrics of the daily CGM profiles within a cluster are homogeneous. CSC definition and validation used 204,710 daily CGM profiles in health, type 1 and type 2 diabetes (T1D, T2D), on different treatments. The CSCs were defined using 23,916 daily CGM profiles (Training set), and the final fixed set of CSCs was obtained using another 37,758 profiles (Validation set). The Testing set (143,036 profiles) was used to establish the robustness and generalizability of the CSCs. The final set of CSCs contains 32 clusters. Any daily CGM profile was classifiable to a single CSC which approximated common glycemic metrics of the daily CGM profile, as evidenced by regression analyses with 0 intercept (R-squares≥0.83, e.g., correlation≥0.91), for all TIR and several other metrics. The CSCs distinguished CGM profiles in health, T2D, and T1D on different treatments, and allowed tracking of the daily changes in a person's glycemic control over time. Daily CGM profiles can be classified into one of 32 prefixed CSCs, which enables a host of applications, e.g. tabulated data interpretation and algorithmic approaches to treatment, database indexing, pattern recognition, and tracking disease progression.

A deep learning nomogram of continuous glucose monitoring data for the risk prediction of diabetic retinopathy in type 2 diabetes.

Continuous glucose monitoring (CGM) data analysis will provide a new perspective to analyze factors related to diabetic retinopathy (DR). However, the problem of visualizing CGM data and automatically predicting the incidence of DR from CGM is still controversial. Here, we explored the feasibility of using CGM profiles to predict DR in type 2 diabetes (T2D) by deep learning approach. This study fused deep learning with a regularized nomogram to construct a novel deep learning nomogram from CGM profiles to identify patients at high risk of DR. Specifically, a deep learning network was employed to mine the nonlinear relationship between CGM profiles and DR. Moreover, a novel nomogram combining CGM deep factors with basic information was established to score the patients' DR risk. This dataset consists of 788 patients belonging to two cohorts: 494 in the training cohort and 294 in the testing cohort. The area under the curve (AUC) values of our deep learning nomogram were 0.82 and 0.80 in the training cohort and testing cohort, respectively. By incorporating basic clinical factors, the deep learning nomogram achieved an AUC of 0.86 in the training cohort and 0.85 in the testing cohort. The calibration plot and decision curve showed that the deep learning nomogram had the potential for clinical application. This analysis method of CGM profiles can be extended to other diabetic complications by further investigation.

Glucose variability in maintenance hemodialysis patients with type 2 diabetes: Comparison of dialysis and nondialysis days.

Hemodialysis (HD) induces several physiological changes that can affect plasma glucose levels in patients with diabetes and in turn their glycemic control. Studies using continuous glucose monitoring (CGM) to assess glucose variations on dialysis days compared with nondialysis days report conflicting results. Here, we used CGM to examine glucose variations induced by HD in patients with type 2 diabetes. Patients with type 2 diabetes undergoing maintenance HD were included. CGM (Ipro2®, Medtronic) was performed at baseline and Week 4, 8, 12, and 16 for up to 7 days at each visit. CGM profiles on days where participants received HD were compared with days without HD using a linear mixed model. Twenty-seven patients were included. The median number of CGM days performed was 8 (interquartile range [IQR] 6-10) for dialysis days and 16 (IQR 12-17) for nondialysis days. The median sensor glucose was 9.4 (95% confidence interval [CI] 8.8-10.2) mmol/L on dialysis days compared with 9.5 (95% CI 8.9-10.2) mmol/L on nondialysis days (p=0.58). Nocturnal mean sensor glucose was higher on dialysis days compared with nondialysis days: 8.8 (95% CI 8.0-9.6) mmol/L versus 8.4 (95% CI 7.7-9.2) mmol/L (p=0.029). Similar median sensor glucose values were found for days on and off HD. Nocturnal glucose levels were modestly increased on dialysis days. Our findings indicate that antidiabetic treatment does not need to be differentiated on dialysis versus nondialysis days in patients with type 2 diabetes undergoing maintenance HD.

Back Cover

The outside back cover image is based on the Original Article Objective sleep characteristics and continuous glucose monitoring profiles of type 2 diabetes patients in real-life settings by Yan Zhao MD et al., https://doi.org/10.1111/dom.14930.

Prevaccination Glucose Time in Range Correlates With Antibody Response to SARS-CoV-2 Vaccine in Type 1 Diabetes.

Poor glucose control has been associated with increased mortality in COVID-19 patients with type 1 diabetes (T1D). This work aimed to assess the effect of prevaccination glucose control on antibody response to the SARS-CoV-2 vaccine BNT162b2 in T1D. We studied 26 patients with T1D scheduled to receive 2 doses, 21 days apart, of BNT162b2, followed prospectively for 6 months with regular evaluation of SARS-CoV-2 antibodies and glucose control. Immunoglobulin G (IgG) to spike glycoprotein were assessed by enzyme-linked immunosorbent assay, and serum neutralization by a live SARS-CoV-2 assay (Vero E6 cells system). Glycated hemoglobin A1c (HbA1c) and continuous glucose monitoring (CGM), including time in range (TIR) and above range (TAR), were collected. The primary exposure and outcome measures were prevaccination glucose control, and antibody response after vaccination, respectively. Prevaccination HbA1c was unrelated to postvaccine spike IgG (r = -0.33; P = .14). Of note, the CGM profile collected during the 2 weeks preceding BNT162b2 administration correlated with postvaccine IgG response (TIR: r = 0.75; P = .02; TAR: r = -0.81; P = .008). Patients meeting the recommended prevaccination glucose targets of TIR (≥ 70%) and TAR (≤ 25%) developed stronger neutralizing antibody titers (P < .0001 and P = .008, respectively), regardless of HbA1c. Glucose control along the study time frame was also associated with IgG response during follow-up (TIR: r = 0.93; P < .0001; TAR: r = -0.84; P < .0001). In T1D, glucose profile during the 2 weeks preceding vaccination is associated with stronger spike antibody binding and neutralization, highlighting a role for well-controlled blood glucose in vaccination efficacy.

Objective sleep characteristics and continuous glucose monitoring profiles of type 2 diabetes patients in real-life settings.

To investigate the association between objective sleep parameters and glycaemic variability determined by continous glucose monitoring (CGM) among patients with type 2 diabetes, given the significant role of sleep in glycaemic control. In this study, CGM was carried out in 28 patients with T2D (aged 62.3 ± 4.8 years, 57% women). Sleep characteristics were assessed by actigraphy within the CGM period. CGM-derived outcomes included glucose level, and percentages of time in range (TIR) and time above range (TAR) during the monitoring period. Associations between intraindividual night-to-night variations in sleep characteristics and overall CGM outcomes were analysed using linear regression. Associations between sleep characteristics during each night and time-matched CGM outcomes were analysed using linear mixed models. A total of 249 person-days of CGM, coupled with 221 nights of sleep characteristics, were documented. Greater standard deviation (SD) of objective sleep duration (minutes) between measurement nights was associated with higher glucose level (coefficient 0.018 mmol/L [95% confidence interval {CI} 0.004, 0.033], P=0.017), smaller proportion of TIR (% in observation period; coefficient -0.20% [95% CI -0.36, -0.03], P=0.023), and greater proportion of TAR (coefficient 0.22% [95% CI 0.06, 0.39], P=0.011). Later sleep midpoint (minutes from midnight) was associated with greater SD of glucose during the same sleep period (coefficient 0.002 minutes [95% CI 0.0001, 0.003], P=0.037), longer nocturnal sleep duration was associated with smaller coefficient of variation of glucose level in the upcoming day (-0.015% [95% CI -0.03, -0.001], P=0.041). Objectively determined sleep duration and sleep midpoint, as well as their daily variability, are associated with CGM-derived glucose profiles in T2D patients.

Difference on Glucose Profile From Continuous Glucose Monitoring in People With Prediabetes vs. Normoglycemic Individuals: A Matched-Pair Analysis.

Comprehensive characteristics of the glycemic profile for prediabetes derived by continuous glucose monitoring (CGM) are unknown. We evaluate the difference of CGM profiles between individuals with prediabetes and normoglycemic individuals, including the response to oral glucose tolerance test (OGTT). Individuals with prediabetes matched for age, sex, and BMI with normoglycemic individuals were instructed to use professional CGM for 1 week. OGTT was performed on the second day. The primary outcomes were percentages of glucose readings time below range (TBR): <54 or <70 mg/dL, time in range (TIR): 70 to 180 mg/dL, and time above range (TAR): >180 or >250 mg/dL. Area under the curve (AUC) was calculated following the OGTT. Glucose variability was depicted by coefficient of variation (CV), SD, and mean amplitude of glucose excursion (MAGE). Wilcoxon sign-ranked test, McNemar mid P-test and linear regression models were employed. In all, 36 participants (median age 51 years; median body mass index [BMI] = 26.4 kg/m2) formed 18 matched pairs. Statistically significant differences were observed for 24-hour time in range (TIR; median 98.5% vs. 99.9%, P = .013), time above range (TAR) >180 mg/dl (0.4% vs. 0%, P = .0062), and 24-hour mean interstitial glucose (113.8 vs. 108.8 mg/dL, P = .0038) between people with prediabetes compared to normoglycemic participants. Statistically significant differences favoring the normoglycemic group were found for glycemic variability indexes (median CV 15.2% vs. 11.9%, P = .0156; median MAGE 44.3 vs. 33.3 mg/dL, P = 0.0043). Following OGTT, the AUC was significantly lower in normoglycemic compared to the prediabetes group (median 18615.3 vs. 16370.0, P = .0347 for total and 4666.5 vs. 2792.7, P = .0429 for incremental 2-hour post OGTT). Individuals with prediabetes have different glucose profiles compared to normoglycemic individuals. CGM might be helpful in individuals with borderline glucose values for a more accurate reclassification.