ISPAD Clinical Practice Consensus Guidelines 2018: Assessment and management of hypoglycemia in children and adolescents with diabetes

A multi-center study on the adaptability of a shared foundation model for electronic health records

Foundation models are transforming artificial intelligence (AI) in healthcare by providing modular components adaptable for various downstream tasks, making AI development more scalable and cost-effective. Foundation models for structured electronic health records (EHR), trained on coded medical records from millions of patients, demonstrated benefits including increased performance with fewer training labels, and improved robustness to distribution shifts. However, questions remain on the feasibility of sharing these models across hospitals and their performance in local tasks. This multi-center study examined the adaptability of a publicly accessible structured EHR foundation model (FMSM), trained on 2.57 M patient records from Stanford Medicine. Experiments used EHR data from The Hospital for Sick Children (SickKids) and Medical Information Mart for Intensive Care (MIMIC-IV). We assessed both adaptability via continued pretraining on local data, and task adaptability compared to baselines of locally training models from scratch, including a local foundation model. Evaluations on 8 clinical prediction tasks showed that adapting the off-the-shelf FMSM matched the performance of gradient boosting machines (GBM) locally trained on all data while providing a 13% improvement in settings with few task-specific training labels. Continued pretraining on local data showed FMSM required fewer than 1% of training examples to match the fully trained GBM's performance, and was 60 to 90% more sample-efficient than training local foundation models from scratch. Our findings demonstrate that adapting EHR foundation models across hospitals provides improved prediction performance at less cost, underscoring the utility of base foundation models as modular components to streamline the development of healthcare AI.

Recommendations for recognizing, risk stratifying, treating, and managing children and adolescents with hypoglycemia

There has been continuous progress in diabetes management over the last few decades, not least due to the widespread dissemination of continuous glucose monitoring (CGM) and automated insulin delivery systems. These technological advances have radically changed the daily lives of people living with diabetes, improving the quality of life of both children and their families. Despite this, hypoglycemia remains the primary side-effect of insulin therapy. Based on a systematic review of the available scientific evidence, this paper aims to provide evidence-based recommendations for recognizing, risk stratifying, treating, and managing patients with hypoglycemia. The objective of these recommendations is to unify the behavior of pediatric diabetologists with respect to the timely recognition and prevention of hypoglycemic episodes and the correct treatment of hypoglycemia, especially in patients using CGM or advanced hybrid closed-loop systems. All authors have long experience in the specialty and are members of the Italian Society of Pediatric Endocrinology and Diabetology. The goal of treating hypoglycemia is to raise blood glucose above 70 mg/dL (3.9 mmol/L) and to prevent further decreases. Oral glucose at a dose of 0.3 g/kg (0.1 g/kg for children using "smart pumps" or hybrid closed loop systems in automated mode) is the preferred treatment for the conscious individual with blood glucose <70 mg/dL (3.9 mmol/L), although any form of carbohydrate (e.g., sucrose, which consists of glucose and fructose, or honey, sugary soft drinks, or fruit juice) containing glucose may be used. Using automatic insulin delivery systems, the oral glucose dose can be decreased to 0.1 g/kg. Practical flow charts are included to aid clinical decision-making. Although representing the official position of the Italian Society of Pediatric Endocrinology and Diabetology (ISPED), these guidelines are applicable to the global audience and are especially pertinent in the era of CGM and other advanced technologies.

Demographic, Clinical, Management, and Outcome Characteristics of 8,004 Young Children With Type 1 Diabetes

OBJECTIVE

To compare demographic, clinical, and therapeutic characteristics of children with type 1 diabetes age <6 years across three international registries: Diabetes Prospective Follow-Up Registry (DPV; Europe), T1D Exchange Quality Improvement Network (T1DX-QI; U.S.), and Australasian Diabetes Data Network (ADDN; Australasia).

RESEARCH DESIGN AND METHODS

An analysis was conducted comparing 2019-2021 prospective registry data from 8,004 children.

RESULTS

Mean ± SD ages at diabetes diagnosis were 3.2 ± 1.4 (DPV and ADDN) and 3.7 ± 1.8 years (T1DX-QI). Mean ± SD diabetes durations were 1.4 ± 1.3 (DPV), 1.4 ± 1.6 (T1DX-QI), and 1.5 ± 1.3 years (ADDN). BMI z scores were in the overweight range in 36.2% (DPV), 41.8% (T1DX-QI), and 50.0% (ADDN) of participants. Mean ± SD HbA1c varied among registries: DPV 7.3 ± 0.9% (56 ± 10 mmol/mol), T1DX-QI 8.0 ± 1.4% (64 ± 16 mmol/mol), and ADDN 7.7 ± 1.2% (61 ± 13 mmol/mol). Overall, 37.5% of children achieved the target HbA1c of <7.0% (53 mmol/mol): 43.6% in DPV, 25.5% in T1DX-QI, and 27.5% in ADDN. Use of diabetes technologies such as insulin pump (DPV 86.6%, T1DX 46.6%, and ADDN 39.2%) and continuous glucose monitoring (CGM; DPV 85.1%, T1DX-QI 57.6%, and ADDN 70.5%) varied among registries. Use of hybrid closed-loop (HCL) systems was uncommon (from 0.5% [ADDN] to 6.9% [DPV]).

CONCLUSIONS

Across three major registries, more than half of children age <6 years did not achieve the target HbA1c of <7.0% (53 mmol/mol). CGM was used by most participants, whereas insulin pump use varied across registries, and HCL system use was rare. The differences seen in glycemia and use of diabetes technologies among registries require further investigation to determine potential contributing factors and areas to target to improve the care of this vulnerable group.

Prevalence and associations of impaired awareness of hypoglycemia in a pediatric type 1 diabetes population - The Norwegian Childhood Diabetes Registry

AIMS

To determine the prevalence and associations of impaired awareness of hypoglycemia (IAH) in pediatric type 1 diabetes.

METHODS

Nationwide, population-based cross-sectional study with 51 % participation. Participants (n = 1329; 53 % males) aged 2-19 years (median 13.3) with type 1 diabetes ≥ 6 months (median 4.6 years) self-assessed hypoglycemia awareness with a validated questionnaire ('Clarke'). Parents responded for children aged < 9 years (n = 235). We estimated associations between IAH and clinical data in the Norwegian Childhood Diabetes Registry.

RESULTS

The overall prevalence of IAH was 22 %, but gradually decreased from 53 % in preschoolers to 12 % in adolescents aged ≥ 16 years. IAH was associated (adjusted OR; 95 %CI) with episodes of severe hypoglycemia (6.0; 3.04, 11.8) and diabetic ketoacidosis (3.45; 1.37, 8.68) the preceding year, increased fear of hypoglycemia (highest quartile vs. lowest: 2.27; 1.51, 3.40), female sex (1.41; 1.05, 1.90), and HbA1c ≥ 8.5 % (69 mmol/mol) vs. 7.5-8.4 % (58-68 mmol/mol) (1.48; 1.01, 2.18), but not with disease duration, use of insulin pump or continuous glucose monitoring, or HbA1c < 7.5 % (58 mmol/mol).

CONCLUSIONS

IAH is prevalent in pediatric diabetes and more likely reported in young children. IAH is associated with severe hypoglycemia and fear of hypoglycemia, but good metabolic control seems achievable without increased risk of IAH.

Characterizing the limitations of using diagnosis codes in the context of machine learning for healthcare

BACKGROUND

Diagnostic codes are commonly used as inputs for clinical prediction models, to create labels for prediction tasks, and to identify cohorts for multicenter network studies. However, the coverage rates of diagnostic codes and their variability across institutions are underexplored. The primary objective was to describe lab- and diagnosis-based labels for 7 selected outcomes at three institutions. Secondary objectives were to describe agreement, sensitivity, and specificity of diagnosis-based labels against lab-based labels.

METHODS

This study included three cohorts: SickKids from The Hospital for Sick Children, and StanfordPeds and StanfordAdults from Stanford Medicine. We included seven clinical outcomes with lab-based definitions: acute kidney injury, hyperkalemia, hypoglycemia, hyponatremia, anemia, neutropenia and thrombocytopenia. For each outcome, we created four lab-based labels (abnormal, mild, moderate and severe) based on test result and one diagnosis-based label. Proportion of admissions with a positive label were presented for each outcome stratified by cohort. Using lab-based labels as the gold standard, agreement using Cohen's Kappa, sensitivity and specificity were calculated for each lab-based severity level.

RESULTS

The number of admissions included were: SickKids (n = 59,298), StanfordPeds (n = 24,639) and StanfordAdults (n = 159,985). The proportion of admissions with a positive diagnosis-based label was significantly higher for StanfordPeds compared to SickKids across all outcomes, with odds ratio (99.9% confidence interval) for abnormal diagnosis-based label ranging from 2.2 (1.7-2.7) for neutropenia to 18.4 (10.1-33.4) for hyperkalemia. Lab-based labels were more similar by institution. When using lab-based labels as the gold standard, Cohen's Kappa and sensitivity were lower at SickKids for all severity levels compared to StanfordPeds.

CONCLUSIONS

Across multiple outcomes, diagnosis codes were consistently different between the two pediatric institutions. This difference was not explained by differences in test results. These results may have implications for machine learning model development and deployment.

Family structures and parents' occupational models: its impact on children's diabetes

AIMS

This study examines how family-related factors influence the management of children and adolescents with type 1 diabetes (T1DM). We investigate the relationship between family patterns, parental work schedules and metabolic control.

MATERIALS AND METHODS

We analysed data from a nationwide diabetes survey (DPV) focusing on HbA1c, severe hypoglycaemia, diabetic ketoacidosis, hospital admissions and inpatient treatment duration. We used linear regression and negative binomial regression models. Our study includes 15,340 children under the age of 18 with data on family structure and parental division of labour.

RESULTS

Children from two-parent households have better HbA1c outcomes than children from single-parent, blended or no-parent households (p < .0001). Higher HbA1C levels are associated with children living with an unemployed father, as opposed to those with full-time working parents or with a full-time working father and a part-time working mother (p < .001).

CONCLUSIONS

These findings emphasise the importance of carefully considering family structure and working time models in the management of paediatric T1DM. Our results highlight risk factors within the family environment and emphasise the need for family-focused counselling of high-risk patients or severe cases in clinical practice.

Evaluation of changes in glycemic control and diabetic complications over time and factors associated with the progression of diabetic complications in Japanese patients with juvenile-onset type 1 diabetes mellitus

BACKGROUND

This study aimed to evaluate the changes in glycemic control and diabetic complications over time in Japanese patients with juvenile-onset type 1 diabetes mellitus and to clarify the factors associated with the progression of diabetic complications.

METHODS

We tracked 129 patients with type 1 diabetes mellitus (21.8 ± 4.1 years old [mean ± SD] with a diabetes duration of 12.6 ± 5.7 years) for up to 19 years and analyzed data on glycated hemoglobin (HbA1c) and indicators related to the severity of diabetic complications (estimated glomerular filtration rate [eGFR], urinary albumin excretion rate [UAE], carotid intima-media thickness [CIMT], and brachial-ankle pulse wave velocity [baPWV]) using linear mixed model and decision tree analysis.

RESULTS

Although the HbA1c and UAE levels improved over time, the eGFR, CIMT, and baPWV worsened. Decision tree analysis showed that HbA1c and the glycoalbumin/HbA1c ratio for eGFR; HbA1c and systolic blood pressure for UAE; low-density lipoprotein cholesterol/high-density lipoprotein cholesterol ratio, glycoalbumin/HbA1c ratio, and body mass index (BMI) for CIMT; and HbA1c for baPWV were associated factors.

CONCLUSIONS

In this retrospective observational study, glycemic control and albuminuria improved; however, renal function and arteriosclerosis worsened over time. HbA1c levels, glycemic excursion, and blood pressure are associated with nephropathy progression. HbA1c levels, glycemic excursion, lipid levels, and BMI are associated with the progression of atherosclerosis.

The Relationship Between Parent Fear of Hypoglycemia and Youth Glycemic Control Across the Recent-Onset Period in Families of Youth with Type 1 Diabetes

BACKGROUND

This study aims to examine the relationship between parents' fear of hypoglycemia (FH) over a 1-year period and child glucose metrics in 126 families of youth recently diagnosed with type 1 diabetes (T1D).

METHODS

Parents completed the Hypoglycemia Fear Survey for Parents (HFS-P) and uploaded 14 days of glucose data at a baseline, 6-month, and 12-month assessment.

RESULTS

Parents' HFS-P total and worry scores increased to a clinically meaningful degree from baseline to 6-month assessment, while multilevel models revealed within- and between-person variability in parents' HFS-P worry and behavior scores over time associated with child glycemia. Specifically, a significant negative relationship for within-person worry scores suggested that when parents reported higher than their average worry scores, their children recorded fewer glucose values in the target range, while within-person behavior scores suggested that when parents reported lower than their average behavior scores, their children recorded more values above the target range. There was also a negative relationship for between-person behavior scores with child glycated hemoglobin and a positive relationship for between-person behavior scores with child glucose values in the target range.

CONCLUSIONS

In the recent-onset period of T1D, parental FH worry and behavior associated with child glycemia possibly due to changes in parents' perceptions of their child's hypoglycemia risk. The clinically meaningful increases in parent FH in the recent-onset period and the negative association for between-person behavior scores with child glycated hemoglobin suggest that clinics should consider screening parents for FH, especially among parents of children with lower glycemic levels.

Effects of therapeutic patient education program on glycemic control and quality of life among children and adolescents with type 1 diabetes mellitus in Fez city, Morocco

Context

Type 1 diabetes mellitus (T1DM) is a chronic disease, mainly observed in children or youth, with a significantly increased incidence in young children. Structured therapeutic patient education (TPE) is a must to help them manage their disease effectively and lead a healthy lifestyle.

Aims

This study aimed to assess the effects of a structured TPE program on glycemic markers and quality of life (QOL) of T1DM children and adolescents in Fez city, Morocco.

Settings and Design

It is a quasi-experimental study.

Materials and Methods

One hundred T1DM children and adolescents, aged 8-18, participated in a TPE intervention at the pediatric department in a hospital center in Fez, Morocco. Glycemic markers were measured and QOL was assessed by a validated questionnaire.

Statistical Analysis Used

Parametric and nonparametric tests were used and statistical significance determined by P < 0.05.

Results

At 3 months' follow-up, both global and dimensional QOL mean scores improved significantly (P ≤ 0.0001), whereas glycosylated hemoglobin (HbA1c) decreased (10.28% vs. 10.62%), tough with no statistical significance (P = 0.160). Furthermore, a significant improvement was observed in the maximum preprandial (2, 11 g/L [1.51-2.58] vs. 2, 37 g/L [1.81-3.21], P = 0.001) and postprandial blood glucose levels (2, 50 g/L [1.90-3.27] vs. 2, 95 g/L [2.07-3.99], P = 0.001) after 3 months; with no significant change in their minimum.

Conclusion

Although this TPE intervention was more effective in improving patients' QOL than their HbA1c, it is worth striving to implement regular TPE programs for T1DM pediatric patients and adjust them to achieve a better patients' glycemic markers levels.

A comparative study using insulin pump therapy and continuous glucose monitoring in newly diagnosed very young children with type 1 diabetes: it is possible to bend the curve of HbA1c

AIMS

The target of metabolic control (HbA1c < 7% or 53 mmol/mol) recommended by the ADA and ISPAD is attained by 30% of children with Type 1 Diabetes (T1D). Advances in technologies for T1D aim to improve metabolic outcomes and reduce complications. This observational study assesses the long-term outcomes of advanced technologies for treatment of T1D compared to conventional approach started at onset in a group of very young children with T1D.

METHODS

54 patients with less 4 years old at onset of T1D were enrolled and followed for up to 9 years after diagnosis. 24 subjects started continuous subcutaneous insulin (CSII) treatment and 30 subjects received MDI therapy from onset. Auxological data, HbA1c and total daily insulin dose (TDD/kg) have been collected at admission and every 4 months. HbA1cAUC>6%, rates of acute complications, glycemic variability indices and glucometrics were also recorded.

RESULTS

Patients with CSII therapy had significantly lower mean HbA1c values compared to subjects receiving MDI treatment. CSII approach also recorded lower mean HbA1cAUC>6% and TDD/kg than MDI therapy. At the last download data, the time in range (TIR) was higher in patients with CSII and hyperglycemia events were lower. Better glycemic variability indices have been described during CSII therapy, including mean glycemia, standard deviation, coefficient of variation (CV), glycemia risk index (GRI) and high blood glucose index (HBGI). There was no statistically significant difference between frequency of severe hypoglycemia and ketoacidosis episodes between groups.

CONCLUSIONS

Early initiation of diabetes technologies is safe and able to determine a better long term glycemic control in young children with T1D. It also allows to flatten the trajectory of HbA1c, probably reducing microvascular, macrovascular and neurological complications of diabetes in this very peculiar age group.

Profile of plasma free amino acids, carnitine and acylcarnitines, and JAK2v617f mutation as potential metabolic markers in children with type 1 diabetic nephropathy

Fifty diabetic nephropathy (DN) children with type 1 diabetes mellitus (T1DM) and 50 healthy matched controls were included. Chromatographic assays of 14 amino acids, free carnitine and 27 carnitine esters using high-performance liquid chromatography/electrospray ionization-mass spectroscopy, and genetic testing for JAK2v617f mutation using real-time PCR were performed. Patients had significantly lower levels of tyrosine, branched-chain amino acids (BCAAs), and BCAA/AAA (aromatic chain amino acids) ratios, glycine, arginine, ornithine, free carnitine and some carnitine esters (C5, 6, 12 and 16) and higher phenylalanine, phenylalanine/tyrosine ratio and C18 compared with the controls and in the macro-albuminuria vs. the microalbuminuria group (p < 0.05 for all) except for free carnitine. Plasma carnitine was negatively correlated with eGFR (r = -0.488, p = 0.000). There were significant positive correlations between tyrosine with UACR ratio (r = 0.296, p = 0.037). The plasma BCAA/AAA ratio showed significant negative correlations with UACR (r = -0.484, p = 0.000). There was a significantly higher frequency of the JAK2V617F gene mutation in diabetic nephropathy patients compared with the control group and in macro-albuminuria than the microalbuminuria group (p = 0.000) for both. When monitoring children with T1DM, plasma free amino acids and acylcarnitine profiles should be considered, especially if they have tested positive for JAK2V617F for the early diagnosis of DN.

The effect of sodium-glucose cotransporter 2 inhibitors as an adjunct to insulin in patients with type 1 diabetes assessed by continuous glucose monitoring: A systematic review and meta-analysis

AIMS

Patients undergoing insulin-based therapy for type 1 diabetes often experience poor glycemic control characterized by significant fluctuations. This study was undertaken to analyze the effect of sodium-glucose cotransporter 2 inhibitors (SGLT2Is), as an adjunct to insulin, on time in range (TIR) and glycemic variability in patients with type 1 diabetes, using continuous glucose monitoring (CGM). In addition, we examined which type of SGLT2I yielded a superior effect compared to others.

METHODS

We conducted a comprehensive search of PubMed, EMBASE, the Cochrane Library, Web of Science, and clinical trial registry websites, retrieving all eligible randomized clinical trials (RCTs) published up until February 2023. We analyzed the mean TIR, mean amplitude of glucose excursions (MAGE), mean daily glucose (MDG), diabetic ketoacidosis (DKA), standard deviation (SD), total insulin dose, and severe hypoglycemia to evaluate the efficacy and safety of SGLT2Is. A random-effects model was also employed.

RESULTS

This study encompassed 15 RCTs. The meta-analysis revealed that the use of SGLT2Is as an adjuvant therapy to insulin led to a significant increase in TIR (MD = 10.78, 95%CI = 9.33-12.23, I2 = 42 %, P < 0.00001) and a decrease in SD (MD = -0.38, 95%CI = -0.50 to -0.26, I2 = 0 %, P < 0.00001), MAGE (MD = -0.92, 95%CI = -1.17 to -0.67, I2 = 19 %, P < 0.00001), MDG(MD = -1.01, 95%CI = -1.32 to -0.70, I2 = 48 %, P < 0.00001), and total insulin dose (MD = -5.81, 95%CI = -7.81 to -3.82, I2 = 32 %, P < 0.00001). No significant increase was observed in the rate of severe hypoglycemia (RR = 1.04, 95 % CI = 0.76-1.43, P = 0.80). However, SGLT2I therapy was associated with increased DKA occurrence (RR = 2.79, 95 % CI = 1.42-5.48; P = 0.003, I2 = 16 %). In addition, the subgroup analyses based on the type of SGLT2Is revealed that dapagliflozin might exhibit greater efficacy compared to other SGLT2Is across most outcomes.

CONCLUSIONS

SGLT2Is exhibited a positive effect on improving blood glucose level fluctuations. Subgroup analysis showed that dapagliflozin appeared to have more advantages. However, giving due consideration to preventing adverse effects, particularly DKA, is paramount.

REGISTRATION

Prospero CRD42023408276.

Post-Hypoglycemic hyperglycemia are highly relevant markers for stratification of glycemic variability and partial remission status of pediatric patients with new-onset type 1 diabetes

AIMS

To evaluate whether parameters of post-hypoglycemic hyperglycemia (PHH) correlated with glucose homeostasis during the first year after type 1 diabetes onset and helped to distinguish pediatric patients undergoing partial remission or not.

METHODS

In the GLUREDIA (GLUcagon Response to hypoglycemia in children and adolescents with new-onset type 1 DIAbetes) study, longitudinal values of clinical parameters, continuous glucose monitoring metrics and residual β-cell secretion from children with new-onset type 1 diabetes were analyzed during the first year after disease onset. PHH parameters were calculated using an in-house algorithm. Correlations between PHH parameters (i.e., PHH frequency, PHH duration, PHH area under the curve [PHHAUC]) and glycemic homeostasis markers were studied using adjusted mixed-effects models.

RESULTS

PHH parameters were strong markers to differentiate remitters from non-remitters with PHH/Hyperglycemia duration ratio being the most sensitive (ratio<0.02; sensitivity = 86% and specificity = 68%). PHHAUC moderately correlated with parameters of glucose homeostasis including TIR (R2 = 0.35, p-value < 0.05), coefficient of variation (R2 = 0.22, p-value < 0.05) and Insulin-Dose Adjusted A1c (IDAA1C) (R2 = 0.32, p-value < 0.05) and with residual β-cell secretion (R2 = 0.17, p-value < 0.05). Classification of patients into four previously described glucotypes independently validated PHH parameters as reliable markers of glucose homeostasis and improved the segregation of patients with intermediate values of IDAA1C and estimated C-peptide (CPEPEST). Finally, a combination of PHH parameters identified groups of patients with specific patterns of hypoglycemia.

CONCLUSION

PHH parameters are new minimal-invasive markers to discriminate remitters from non-remitters and evaluate glycemic homeostasis during the first year of type 1 diabetes. PHH parameters may also allow patient-targeted therapeutic management of hypoglycemic episodes.

Self-supervised machine learning using adult inpatient data produces effective models for pediatric clinical prediction tasks

OBJECTIVE

Development of electronic health records (EHR)-based machine learning models for pediatric inpatients is challenged by limited training data. Self-supervised learning using adult data may be a promising approach to creating robust pediatric prediction models. The primary objective was to determine whether a self-supervised model trained in adult inpatients was noninferior to logistic regression models trained in pediatric inpatients, for pediatric inpatient clinical prediction tasks.

MATERIALS AND METHODS

This retrospective cohort study used EHR data and included patients with at least one admission to an inpatient unit. One admission per patient was randomly selected. Adult inpatients were 18 years or older while pediatric inpatients were more than 28 days and less than 18 years. Admissions were temporally split into training (January 1, 2008 to December 31, 2019), validation (January 1, 2020 to December 31, 2020), and test (January 1, 2021 to August 1, 2022) sets. Primary comparison was a self-supervised model trained in adult inpatients versus count-based logistic regression models trained in pediatric inpatients. Primary outcome was mean area-under-the-receiver-operating-characteristic-curve (AUROC) for 11 distinct clinical outcomes. Models were evaluated in pediatric inpatients.

RESULTS

When evaluated in pediatric inpatients, mean AUROC of self-supervised model trained in adult inpatients (0.902) was noninferior to count-based logistic regression models trained in pediatric inpatients (0.868) (mean difference = 0.034, 95% CI=0.014-0.057; P < .001 for noninferiority and P = .006 for superiority).

CONCLUSIONS

Self-supervised learning in adult inpatients was noninferior to logistic regression models trained in pediatric inpatients. This finding suggests transferability of self-supervised models trained in adult patients to pediatric patients, without requiring costly model retraining.

Assessing Parental Competence and Self-Ratings in Management of Pediatric Type 1 Diabetes and Emergency Glucagon Administration-An Exploratory Observational Study

BACKGROUND

Parents of pediatric patients with type I diabetes require competence in hypoglycemia management and skills in glucagon administration to deal with potentially life-threatening severe hypoglycemia. We aimed to compare parents' subjective self-ratings to an objective expert assessment of competences and skills in dealing with severe hypoglycemia.

METHODS

We interviewed 140 participants to assess their subjective self-ratings. The objective expert assessments used a standardized clinical case scenario of severe hypoglycemia and a practical demonstration of glucagon administration.

RESULTS

The participants self-rated their competence in hypoglycemia management as good (5) or very good (6), and their skills in administering glucagon as acceptable (3) [Scale: very poor (1) to very good (6)]. In the standardized clinical case scenario, 1.4% (2/140) of participants named all relevant steps of severe hypoglycemia management. In the practical demonstration of glucagon administration, 92.9% (130/140) of participants committed at least one drug handling error; 52.1% (73/140) committed at least one drug handling error rated with high clinical risk.

CONCLUSIONS

We found discrepancies regarding participants' subjective self-ratings compared to their performance in the respective objective expert assessments. These discrepancies indicate a lack of error awareness and the need for intervention studies to improve competence in hypoglycemia management and glucagon administration.

Autoantibody test for type 1 diabetes in children: are there reasons to implement a screening program in the general population? A statement endorsed by the Italian Society for Paediatric Endocrinology and Diabetes (SIEDP-ISPED) and the Italian Society of Paediatrics (SIP)

In recent years screening of type 1 diabetes (T1D) in both at risk children and general population has been widely discussed with the aim of increasing awareness of the importance to early detect (and possibly treat) at-risk children in early stages of the chronic autoimmune progression to T1D.In fact, it is well known that first-degree relatives have the highest risk of T1D, but up to 90% of children who develop T1D do not have a family history and belong to the general population.The reasons for screening children well before the clinical onset of T1D include prevention of DKA (still up to 60% children are diagnosed with DKA) and related morbidities and mortality, reducing the need for hospitalisation, time to provide emotional support and education to ensure a smooth transition to insulin treatment, and opportunities for new treatments to prevent or delay progression.There are pros and cons of screening of T1D in children, but recent evidences suggest that it is now time to consider possible a screening for T1D in children.Recently, the European Society for Paediatric Endocrinology (ESPE) has endorsed a Position Statement, discussing the pros and cons of screening for T1D in the general population; ESPE supports national Societies for Paediatric Endocrinology to spread information on this important issue in various countries throughout Europe; the Italian Society for Paediatric Endocrinology and Diabetes (SIEDP-ISPED) and the Italian Society of Paediatrics (SIP) endorse this document with the specific aim of increasing awareness on screening for paediatric T1D in the general population.

Hypoglycemia risk with physical activity in type 1 diabetes: a data-driven approach

Physical activity (PA) provides numerous health benefits for individuals with type 1 diabetes (T1D). However, the threat of exercise-induced hypoglycemia may impede the desire for regular PA. Therefore, we aimed to study the association between three common types of PA (walking, running, and cycling) and hypoglycemia risk in 50 individuals with T1D. Real-world data, including PA duration and intensity, continuous glucose monitor (CGM) values, and insulin doses, were available from the Tidepool Big Data Donation Project. Participants' mean (SD) age was 38.0 (13.1) years with a mean (SD) diabetes duration of 21.4 (12.9) years and an average of 26.2 weeks of CGM data available. We developed a linear regression model for each of the three PA types to predict the average glucose deviation from 70 mg/dl for the 2 h after the start of PA. This is essentially a measure of hypoglycemia risk, for which we used the following predictors: PA duration (mins) and intensity (calories burned), 2-hour pre-exercise area under the glucose curve (adjusted AUC), the glucose value at the beginning of PA, and total bolus insulin (units) within 2 h before PA. Our models indicated that glucose value at the start of exercise and pre-exercise glucose adjusted AUC (p < 0.001 for all three activities) were the most significant predictors of hypoglycemia. In addition, the duration and intensity of PA and 2-hour bolus insulin were weakly associated with hypoglycemia for walking, running, and cycling. These findings may provide individuals with T1D with a data-driven approach to preparing for PA that minimizes hypoglycemia risk.

Continuous glucose monitoring versus blood glucose monitoring for risk of severe hypoglycaemia and diabetic ketoacidosis in children, adolescents, and young adults with type 1 diabetes: a population-based study

BACKGROUND

The effect of continuous glucose monitoring on the risk of severe hypoglycaemia and ketoacidosis in patients with diabetes is unclear. We investigated whether rates of acute diabetes complications are lower with continuous glucose monitoring, compared with blood glucose monitoring, and which metrics predict its risk in young patients with type 1 diabetes.

METHODS

In this population-based cohort study, patients were identified from 511 diabetes centres across Austria, Germany, Luxembourg, and Switzerland participating in the Diabetes Prospective Follow-up initiative. We included people with type 1 diabetes aged 1·5-25·0 years, with a diabetes duration of more than 1 year, who had been treated between Jan 1, 2014, and June 30, 2021, and had an observation time of longer than 120 days in the most recent treatment year. Severe hypoglycaemia and ketoacidosis rates during the most recent treatment year were examined in people using continuous glucose monitoring and in those using blood glucose monitoring. Adjustments of statistical models included age, sex, diabetes duration, migration background, insulin therapy (pump or injections), and treatment period. Rates of severe hypoglycaemia and diabetic ketoacidosis were evaluated by several continuous glucose monitoring metrics, including percentage of time below target glucose range (<3·9 mmol/L), glycaemic variability (measured as the coefficient of variation), and mean sensor glucose.

FINDINGS

Of 32 117 people with type 1 diabetes (median age 16·8 years [IQR 13·3-18·1], 17 056 [53·1%] males), 10 883 used continuous glucose monitoring (median 289 days per year), and 21 234 used blood glucose monitoring. People using continuous glucose monitoring had lower rates of severe hypoglycaemia than those using blood glucose monitoring (6·74 [95% CI 5·90-7·69] per 100 patient-years vs 8·84 [8·09-9·66] per 100 patient-years; incidence rate ratio 0·76 [95% CI 0·64-0·91]; p=0·0017) and diabetic ketoacidosis (3·72 [3·32-4·18] per 100 patient-years vs 7·29 [6·83-7·78] per 100 patient-years; 0·51 [0·44-0·59]; p<0·0001). Severe hypoglycaemia rates increased with percentage of time below target glucose range (incidence rate ratio 1·69 [95% CI 1·18-2·43]; p=0·0024, for 4·0-7·9% vs <4·0% and 2·38 [1·51-3·76]; p<0·0001, for ≥8·0% vs <4·0%) and glycaemic variability (coefficient of variation ≥36% vs <36%; incidence rate ratio 1·52 [95% CI 1·06-2·17]; p=0·022). Diabetic ketoacidosis rates increased with mean sensor glucose (incidence rate ratio 1·77 [95% CI 0·89-3·51], p=0·13, for 8·3-9·9 mmol/L vs <8·3 mmol/L; 3·56 [1·83-6·93], p<0·0001, for 10·0-11·6 mmol/L vs <8·3 mmol/L; and 8·66 [4·48-16·75], p<0·0001, for ≥11·7 mmol/L vs <8·3 mmol/L).

INTERPRETATION

These findings provide evidence that continuous glucose monitoring can reduce severe hypoglycaemia and ketoacidosis risk in young people with type 1 diabetes on insulin therapy. Continuous glucose monitoring metrics might help to identify those at risk for acute diabetes complications.

FUNDING

German Center for Diabetes Research, German Federal Ministry of Education and Research, German Diabetes Association, and Robert Koch Institute.

Severe Hypoglycemia and the Use of Glucagon Rescue Agents: An Observational Survey in Adults With Type 1 Diabetes

Severe hypoglycemia (SH) is the most frequent and potentially serious complication affecting individuals with type 1 diabetes and can have major clinical and psychosocial consequences. Glucagon is the only approved treatment for SH that can be administered by non-health care professionals (HCPs); however, reports on the experiences and emotions of people with type 1 diabetes associated with SH and glucagon rescue use are limited. This survey study demonstrated that an increasing number of individuals with type 1 diabetes have current and filled prescriptions for glucagon and have been educated about glucagon rescue use by an HCP. Despite this positive trend, challenges with SH remain, including a high level of health care resource utilization, considerable out-of-pocket expenses for glucagon kits, a high prevalence of hypoglycemia unawareness, and a negative emotional impact on individuals with diabetes. Nocturnal and exercise-related hypoglycemia were concerns for most survey participants.

Metabolic control during the first two years of the COVID-19 pandemic in pediatric patients with type 1 diabetes: results from the German DPV initiative

AIM

To assess effects of the SARS-CoV2 pandemic on metabolic control in youth with type 1 diabetes (T1D) in Germany in a population-based analysis.

METHODS

Data from 33,372 pediatric T1D patients from the Diabetes Prospective Follow-up (DPV) registry, with face-to-face visits or telemedicine contacts in the years 2019-2021, were available. Datasets from eight time periods between March 15, 2020, and December 31, 2021, according to SARS-CoV2 incidence waves, were compared to those from five control time periods. Parameters of metabolic control were assessed with adjustment for sex, age, diabetes duration, and repeated measurements. Laboratory-measured HbA1c values and those estimated from CGM were aggregated into a combined glucose indicator (CGI).

RESULTS

There was no clinically relevant difference in metabolic control between pandemic and control time periods with adjusted CGI values ranging from 7.61% [7.60-7.63] (mean [95% confidence interval (CI)]) in the third quarter of 2019 to 7.83% [7.82-7.85] in the time period from January 1 to March 15 2020, in the other control periods, and during the pandemic, CGI values lay between these values. BMI-SDS rose during the pandemic from 0.29 [0.28-0.30] (mean [95% CI]) in the third quarter of 2019 to 0.40 [0.39-0.41] during the fourth wave. Adjusted insulin dose rose during the pandemic. Event rates for hypoglycemic coma and diabetic ketoacidosis remained unchanged.

CONCLUSIONS

We found no clinically relevant change of glycemic control or incidence of acute diabetes complications during the pandemic. The observed BMI increase may represent an important health risk for youth with T1D.

The efficacy of insulin degludec and insulin glargine over NPH insulin among toddlers and preschoolers with type 1 diabetes using glycemic variability and time in range

Optimizing glycemic control without risking hypoglycemia is crucial in toddlers and preschoolers with type 1 diabetes (T1D) to avoid cognitive impairment later in life. Hence, this study aims to compare glycemic parameters among toddlers and preschoolers with T1D in relation to different basal insulins. Sixty toddlers and preschoolers with T1D with mean age of 3.53 ± 1.17 years (range, 2-6) and mean diabetes duration of 9.37 ± 1.85 months were randomly assigned into three equal groups; group A received insulin degludec, group B received insulin glargine, and group C were on NPH. At baseline, the three groups were matched regarding clinical and laboratory parameters (p > 0.05). They were followed up at 3 and 6 months for insulin daily dose (IDD), hypoglycemia and severe-hypoglycemia frequency, and glycated hemoglobin (HbA1c). At the study endpoint, continuous glucose monitoring (CGM) was assessed in a random sample of 10 patients from each group. The mean time in range (TIR) of the studied cohort was 55.07 ± 24.05%, and their mean coefficient of variation (CV) was 42.82 ± 11.69%. The TIR was significantly higher in the degludec group (69.36 ± 18.54) and the glargine group (55.43 ± 26.51) than the NPH group (32.56 ± 9.11), p < 0.001. Meanwhile, the CV was significantly lower in the degludec group (35.12 ± 6.47) than the gargine (44.1 ± 13.13) and the NPH (53.8 ± 7.54) groups, p < 0.001. The insulin degludec and glargine groups had significantly lower HbA1c (p = 0.002), hypoglycemia (p = 0.006), severe hypoglycemia (p = 0.029), and IDD (p = 0.015) than the NPH group.

CONCLUSION

Insulin degludec and glargine resulted in better HbA1c and TIR with reduced hypoglycemia and IDD than NPH among toddlers and preschoolers with T1D. Moreover, CV was lowest in the insulin degludec group.

WHAT IS KNOWN

• Insulin therapy is the mainstay of T1D management. • Optimal insulin therapy for young children with T1D should provide effective glycemic.

WHAT IS NEW

• Insulin degludec and insulin glargine have better efficacy than NPH insulin among toddlers and preschoolers with T1D in the term of significantly lower coefficient of variation, HbA1c and IDD and significantly higher time in range. • Insulin degludec and insulin glargine have better safety in the term of less hypoglycemia and severe hypoglycemia episodes than NPH insulin among toddlers and preschoolers with T1D.

More hypoglycemia not associated with increasing estimated adiposity in youth with type 1 diabetes

BACKGROUND

Despite the widespread clinical perception that hypoglycemia may drive weight gain in youth with type 1 diabetes (T1D), there is an absence of published evidence supporting this hypothesis.

METHODS

We estimated the body fat percentage (eBFP) of 211 youth (HbA1c 8.0-13.0%, age 13-16) at baseline, 6, and 18 months of the Flexible Lifestyles Empowering Change trial using validated equations. Group-based trajectory modeling assigned adolescents to sex-specific eBFP groups. Using baseline 7-day blinded continuous glucose monitoring data, "more" vs. "less" percent time spent in hypoglycemia was defined by cut-points using sample median split and clinical guidelines. Adjusted logistic regression estimated the odds of membership in an increasing eBFP group comparing youth with more vs. less baseline hypoglycemia.

RESULTS

More time spent in clinical hypoglycemia (defined by median split) was associated with 0.29 the odds of increasing eBFP in females (95% CI: 0.12, 0.69; p = 0.005), and 0.33 the odds of stable/increasing eBFP in males (95% CI: 0.14, 0.78; p = 0.01).

CONCLUSIONS

Hypoglycemia may not be a major driver of weight gain in US youth with T1D and HbA1c ≥8.0. Further studies in different sub-groups are needed to clarify for whom hypoglycemia may drive weight gain and focus future etiological studies and interventions.

IMPACT

We contribute epidemiological evidence that hypoglycemia may not be a major driver of weight gain in US youth with type 1 diabetes and HbA1c ≥8.0% and highlight the need for studies to prospectively test this hypothesis rooted in clinical perception. Future research should examine the relationship between hypoglycemia and adiposity together with psychosocial, behavioral, and other clinical factors among sub-groups of youth with type 1 diabetes (i.e., who meet glycemic targets or experience a frequency/severity of hypoglycemia above a threshold) to further clarify for whom hypoglycemia may drive weight gain and progress etiological understanding of and interventions for healthy weight maintenance.

Integrated safety and efficacy analysis of dasiglucagon for the treatment of severe hypoglycaemia in individuals with type 1 diabetes

AIMS

To perform an integrated analysis of the safety and efficacy of dasiglucagon, a glucagon analogue available in a ready-to-use aqueous formulation, to treat severe hypoglycaemia (SH) in type 1 diabetes (T1D).

MATERIALS AND METHODS

An integrated analysis of dasiglucagon safety was conducted on data from two placebo-controlled trials (placebo-controlled pool) and two placebo-controlled and four non-placebo-controlled trials (broad pool) in adults with T1D. An integrated analysis of dasiglucagon efficacy was conducted of pooled data and within demographic subgroups from the two placebo-controlled and two non-placebo-controlled trials in adults with T1D.

RESULTS

Dasiglucagon had a similar safety and tolerability profile to that of reconstituted glucagon. In the placebo-controlled datasets, no serious adverse events (AEs), AEs leading to withdrawal from the trial, or deaths were reported. The most common causally related AEs were nausea (56.5%) and vomiting (24.6%). The broad pool safety analysis showed similar results. Dasiglucagon efficacy in time to plasma glucose recovery from insulin-induced SH was similar to that of reconstituted glucagon (median 10.0 and 12.0 minutes, respectively) and superior to placebo (median 40.0 minutes; P < 0.0001). The median recovery time was consistent across all placebo-controlled trial subgroups.

CONCLUSIONS

Dasiglucagon was well tolerated and effective as a rapid rescue agent for insulin-induced SH in people with T1D.

Artificial intelligence for non-invasive glycaemic-events detection via ECG in a paediatric population: study protocol

Purpose

Paediatric Type 1 Diabetes (T1D) patients are at greater risk for developing severe hypo and hyperglycaemic events due to poor glycaemic control. To reduce the risk of adverse events, patients need to achieve the best possible glycaemic management through frequent blood glucose monitoring with finger prick or Continuous Glucose Monitoring (CGM) systems. However, several non-invasive techniques have been proposed aiming at exploiting changes in physiological parameters based on glucose levels. The overall objective of this study is to validate an artificial intelligence (AI) based algorithm to detect glycaemic events using ECG signals collected through non-invasive device.

Methods

This study will enrol T1D paediatric participants who already use CGM. Participants will wear an additional non-invasive wearable device for recording physiological data and respiratory rate. Glycaemic measurements driven through ECG variables are the main outcomes. Data collected will be used to design, develop and validate the personalised and generalized classifiers based on a deep learning (DL) AI algorithm, able to automatically detect hypoglycaemic events by using few ECG heartbeats recorded with wearable devices.

Results

Data collection is expected to be completed approximately by June 2023. It is expected that sufficient data will be collected to develop and validate the AI algorithm.

Conclusion

This is a validation study that will perform additional tests on a larger diabetes sample population to validate the previous pilot results that were based on four healthy adults, providing evidence on the reliability of the AI algorithm in detecting glycaemic events in paediatric diabetic patients in free-living conditions.

Trial registration

ClinicalTrials.gov identifier: NCT03936634. Registered on 11 March 2022, retrospectively registered, https://www.clinicaltrials.gov/ct2/show/NCT05278143?titles=AI+for+Glycemic+Events+Detection+Via+ECG+in+a+Pediatric+Population&draw=2&rank=1.

Supplementary information

The online version contains supplementary material available at 10.1007/s12553-022-00719-x.

Influence of the occurrence and duration of partial remission on short-term metabolic control in type 1 diabetes: the DIABHONEY pediatric study

OBJECTIVE

To evaluate the residual effect of partial remission (PR) on immediate post-PR glycemic control according to its occurrence and duration in a cohort of children with type 1 diabetes mellitus (T1DM).

PATIENTS AND METHODS

Values of glycemic control parameters [i.e. HbA_1C, insulin dose-adjusted hemoglobin A_1C (IDAA_1C), glycemic target-adjusted HbA_1C (GTAA_1C)] and data from glucose monitoring devices from 189 pediatric patients with new-onset type 1 diabetes were collected retrospectively from 24 months. Patients were characterized according to their remission status (PR⁺ and PR^-). PR⁺ patients were subdivided into three subgroups regarding PR duration [i.e. short (⩾3-⩽6 months), intermediate (>6-⩽12 months), and long PR (>12-⩽14 months)]. We compared glycemic control data from each PR⁺ subgroup at +6 and +12 months post-PR with PR^- patients at the same postdiagnosis time. Second, PR⁺ subgroups were compared with each other.

RESULTS

PR⁺ patients showed improved glycemic control (i.e. HbA_1C, IDAA_1C, and GTAA_1C) at + 6 months post-PR when compared with nonremitters (PR^-), independently of the PR duration subgroups (p < 0.05). Interestingly, patients in long PR⁺ subgroup exhibited higher positive residual effect than short PR⁺ subgroup with lower GTAA_1C scores (p = 0.02), better time in range (TIR) (p = 0.003), less time in hypoglycemia (10.45 versus 16.13%, p = 0.03) and less glycemic variability (83.1 mg/dl versus 98.84 mg/dl, p = 0.03). No significant differences were found for glucose control between PR⁺ and PR^- patients at +12 months post-PR.

CONCLUSION

This study supports the positive impact of PR occurrence and duration on short-term metabolic control (better HbA_1C levels, IDAA_1C and GTAA_1C scores, TIR, and less glycemic variability) with the residual effect increasing according to PR duration.

Efficacy of Low-dose Dapagliflozin in Young People with Type 1 Diabetes

Objective Young people with type 1 diabetes are likely to gain body weight and not achieve optimal glycemic control with only high doses of insulin. This study examined the efficacy of the sodium-glucose cotransporter 2 (SGLT2) inhibitor dapagliflozin as an adjunct-to-insulin therapy in young Japanese subjects with type 1 diabetes who had been diagnosed before 15 years old, were overweight, and had inadequate control despitereceiving intensive insulin therapy. Methods Twenty-two patients with type 1 diabetes (12 boys and 10 girls 16.0-33.9 years old) were involved in the study. All patients had a body mass index (BMI) >25 kg/m², glycated hemoglobin (HbA1c) level >7.0%, and daily insulin dose >0.5 units/kg. They were treated with a low dose of dapagliflozin (5.0 mg/day) as an adjunctive therapy to insulin. Fourteen patients were treated with multiple daily injections of insulin, while eight used an insulin pump. Results The body weights and BMIs were significantly reduced during the 12-month study period (change of -4.4 kg and -1.7 kg/m², p<0.001, respectively). Their insulin dose was significantly decreased (-0.17 units/kg, P <0.001), and glycemic control was significantly improved (fasting plasma glucose: -18.7 mg/dL, HbA1c: -0.62%, p<0.001) during the study period. There was one episode of diabetic ketoacidosis, with no other problematic adverse events, including severe hypoglycemia, observed. Conclusion The use of low-dose dapagliflozin as an adjunct therapy may be beneficial in overweight young people with poorly controlled type 1 diabetes.

Prevalence of hypoglycemic conditions in adolescents with type 1 diabetes mellitus in real clinical practice

BACKGROUND. Hypoglycemia and fear of hypoglycemia remain critical problems in the treatment of adolescents with type 1 diabetes mellitus (DM1) and are factors limiting proper control of glycemia and preventing the achievement of metabolic compensation of the disease. The use of pump insulin therapy involves the prevention of hypoglycemic conditions.

AIM. To analyze the frequency and duration of hypoglycemia episodes, their effect on the metabolic compensation of the disease in adolescents with type 1 diabetes mellitus (DM1) in real clinical practice, depending on the mode/method of insulin administration.

MATERIALS AND METHODS. The study involved 117 adolescents with DM1 aged 12 to 19 years (average age 15.5 years). 37 adolescents received therapy by continuous subcutaneous insulin infusion (CSII); 80 adolescents received therapy by multiple insulin injections (MII). The level of glycated hemoglobin (HbA1c) was determined for all adolescents, and its main indicators were evaluated using a 6 days continuous glucose monitoring (CGM) by the «blind» method of a professional system with an iPro 2 sensor (Medtronic MiniMed, USA).

RESULTS. Episodes of a decrease in glucose levels <3,9 mmol/l were recorded in 87% of patients (n=102), 63% (n=74) showed a decrease in glucose levels <3,0 mmol/l. Episodes decrease in glucose levels <3,9 mmol/l at night were recorded in 68% of patients (n=80), and with glucose levels <3,9 mmol/l in 46% (n=54). The frequency of episodes of glucose lowering <3,9 mmol/l had no statistically significant differences depending on the methods of insulin administration (by continuous subcutaneous insulin infusion or multiple insulin injections), however, they are more common in adolescents with HbA1c <7,0% (p=0,03). The median time spent by patients in the range of <3,9 mmol/l was 5% per day, and a longer time in this range was observed in patients with HbA1c <7,0% (p=0,006). The median time in the range of <3,0 mmol/l was 1% per day and had no significant differences depending on the level of HbA1c (p=0,559). There were also no significant differences depending on the groups using CSII and MII (p=0,640 and p=0,250).

CONCLUSION. Episodes of glucose reduction in the range of <3,9 mmol/l according to CGM data are more common in adolescents with HbA1c target values, regardless of the method of insulin administration. Significantly more time in range of <3,9 mmol/l is spent by adolescents with target values of HbA1c i.е. <7,0% compared with HbA1c ≥7,0%, however, in both groups, a large number of patients had time in the range below the target level was higher than recommended values.

Liraglutide for Weight Management in Children and Adolescents With Prader-Willi Syndrome and Obesity

CONTEXT

Prader-Willi syndrome (PWS) is characterized by lack of appetite control and hyperphagia, leading to obesity. Pharmacological options for weight management are needed.

OBJECTIVE

To determine whether liraglutide treatment for weight management is superior to placebo/no treatment in pediatric individuals with PWS.

METHODS

This was a multicenter, 52-week, placebo-controlled trial with a 16-week double-blinded period. Adolescents (n = 31, aged 12-17 years; Tanner stage 2-5) and children (n = 24, aged 6-11 years; Tanner stage <2) with PWS and obesity were included. Patients were randomized 2:1 to liraglutide 3.0 mg (or maximum-tolerated dose) or placebo for 16 weeks, after which placebo was stopped. Liraglutide was continued for 52 weeks. All patients followed a structured diet and exercise program throughout the trial. The coprimary endpoints were change in body mass index (BMI) standard deviation score (SDS) from baseline to 16 and 52 weeks. Secondary endpoints included other weight-related parameters, hyperphagia, and safety.

RESULTS

Change in BMI SDS from baseline to weeks 16 and 52 was not significantly different between treatments in adolescents (estimated treatment difference: -0.07 at week 16 and -0.14 at week 52) and children (-0.06 and -0.07, respectively). Changes in other weight-related parameters between treatments were not significant. At week 52, hyperphagia total and drive scores were lower in adolescents treated with liraglutide vs no treatment. The most common adverse events with liraglutide were gastrointestinal disorders.

CONCLUSION

Although the coprimary endpoints were not met, changes in hyperphagia total and drive scores in adolescents warrant further studies on liraglutide in this population.

The Development and Validation of a Mobile Application for Guidance for Management of Severe Diabetic Ketoacidosis

OBJECTIVE

To develop and validate a mobile application-based tool for the management guidance of children and adolescents with diabetic ketoacidosis (DKA).

METHODS

The study involved the development of a mobile application-based tool for DKA management in accordance with the International Society of Pediatric and Adolescent Diabetes (ISAPD) guidelines, 2018. The impact of the mobile application in preventing protocol deviation and resultant complications was assessed. Case records of 70 children and adolescents [39 boys, 8.9 (4.1) y of age] with severe DKA managed in the authors' intensive care unit were examined. The application guidance and real-time management were compared to the standard protocol.

RESULTS

Protocol deviations were observed in 58 (82.9%), with two or more errors in 28 (40%). These included lack of initial fluid bolus (4, 5.7%), excessive fluid supplementation (8, 11.4%), inadequate initial fluid (25, 35.7%) and potassium supplementation (13, 18.6%), delayed response to fall in potassium (15, 21.4%) and glucose levels (24, 34.3%), and erroneous insulin administration (19, 27.1%). These errors contributed to 42.1% of severe hypokalemia and 56% of significant hypoglycemia episodes. The mobile application guidance was in accordance with the protocol in all the case scenario.

CONCLUSION

Deviation from the management protocol is common in DKA and associated with adverse outcomes. Mobile application guidance is expected to reduce the protocol deviation with a potential of improving outcomes.

An effective and cost-saving structured education program teaching dynamic glucose management strategies to a socio-economically deprived cohort with type 1 diabetes in a VIRTUAL setting

OBJECTIVES

Compare the clinical and cost-effectiveness of an established face to face (F2F) structured education program to a new remote (VIRTUAL) program teaching dynamic glucose management (DynamicGM) to children and young people with type 1 diabetes (CYPD) using continuous glucose monitoring (CGM). To ascertain the most effective DynamicGM strategies predicting time in range (TIR) (3.9-10.0 mmol/L) and incorporating these into a user-friendly teaching aid.

DESIGN AND METHODS

Effectiveness of the F2F and VIRTUAL programs were ascertained by comparing the mean change (Δ) from baseline to 6 months in HbA1c, TIR and severe hypoglycemia. Delivery cost for the two programs were evaluated. Factors predicting TIR in the combined cohort were determined and incorporated into a user-friendly infographic.

RESULTS

First 50 graduates per group were evaluated. The mean difference in Δ HbA1c, Δ TIR and Δ episodes of severe hypoglycemia between VIRTUAL and F2F groups were 1.16 (p = 0.47), 0.76 (p = 0.78) and -0.06 (p = 0.61) respectively. Delivery cost per 50 CYPD for VIRTUAL and F2F were $5752 and $7020, respectively. The strongest predictors of TIR (n = 100) were short bursts of exercise (10-40 min) to lower hyperglycemia (p < 0.001), using trend arrow adjustment tools (p < 0.001) and adjusting pre-meal bolus timing based on trend arrows (p < 0.01). These strategies were translated into a GAME (Stop highs), SET (Stay in target), MATCH (Prevent lows) mnemonic.

CONCLUSION

Teaching DynamicGM VIRTUALLY is just as effective as F2F delivery and cost saving. Short bursts of exercise and using CGM trend arrows to adjust insulin dose and timing improves TIR.

Nationally Subsidized Continuous Glucose Monitoring: A Cost-effectiveness Analysis

OBJECTIVE

The Continuous Glucose Monitoring (CGM) Initiative recently introduced universal subsidized CGM funding for people with type 1 diabetes under 21 years of age in Australia. We thus aimed to evaluate the cost-effectiveness of this CGM Initiative based on national implementation data and project the economic impact of extending the subsidy to all age-groups.

RESEARCH DESIGN AND METHODS

We used a patient-level Markov model to simulate disease progression for young people with type 1 diabetes and compared government-subsidized access to CGM with the previous user-funded system. Three years of real-world clinical input data were sourced from analysis of the Australasian Diabetes Data Network and National Diabetes Services Scheme registries. Costs were considered from the Australian health care system's perspective. An annual discount rate of 5% was applied to future costs and outcomes. Uncertainty was evaluated with probabilistic and deterministic sensitivity analyses.

RESULTS

Government-subsidized CGM funding for young people with type 1 diabetes compared with a completely user-funded model resulted in an incremental cost-effectiveness ratio (ICER) of AUD 39,518 per quality-adjusted life-year (QALY) gained. Most simulations (85%) were below the commonly accepted willingness-to-pay threshold of AUD 50,000 per QALY gained in Australia. Sensitivity analyses indicated that base-case results were robust, though strongly impacted by the cost of CGM devices. Extending the CGM Initiative throughout adulthood resulted in an ICER of AUD 34,890 per QALY gained.

CONCLUSIONS

Providing subsidized access to CGM for people with type 1 diabetes was found to be cost-effective compared with a completely user-funded model in Australia.

Not only for caregivers: intranasal glucagon for severe hypoglycaemia in a simulation study

AIMS

To evaluate: (i) the propensity of paediatrics and emergency medicine residents to select different therapeutic options and (ii) the speed and administration success in a high-fidelity simulation of severe hypoglycaemia in a child with type 1 diabetes (T1DM).

METHODS

In this single-centre high-fidelity simulation study, 51 paediatrics or emergency medicine residents were exposed to a scenario of severe hypoglycaemia in a T1DM child attending an ambulatory setting, before and after a training on the preparation and administration of both injectable and IN glucagon. Time for drug delivery and its effectiveness were collected.

RESULTS

Before training, 45.1% of participants chose to administer injectable glucagon, 43.1% intravenous glucose solution, 5.9% intranasal (IN) glucagon, and 5.9% took no action. Administration was successful in 74% of injectable glucagon, 33.3% intravenous glucose solution, and 22.7% IN glucagon. After training, 58.8% of participants chose IN and 41.2% injectable glucagon, with 100% of successful administrations for IN glucagon and 90.5% for injectable glucagon. Time to successful administration was shorter for IN than injectable glucagon (23 ± 10 vs. 38 ± 7 s, p < 0.0001).

CONCLUSIONS

IN glucagon is an easy and effective option for severe hypoglycaemia treatment, with an almost zero possibility of failure provided that adequate training is imparted.

Characteristics of Type 2 Diabetes in Female and Male Youth

The incidence of type 2 diabetes in children is rising and carries a worse prognosis than in adults. The influence of sex on pediatric type 2 diabetes outcomes has not been well investigated. We studied 715 youth with type 2 diabetes diagnosed at a median age of 13.7 years and compared sex differences in demographic, clinical, and laboratory characteristics within the first year of diagnosis. Females diagnosed with type 2 diabetes were younger and at a higher stage of pubertal development than males, yet presented with lower A1Cs, a lower prevalence of diabetic ketoacidosis, and higher HDL cholesterol levels.

Impact on variables of severe hypoglycaemia and healthcare costs of the use of the FreeStyle system in paediatric population with type 1 diabetes mellitus

INTRODUCTION

Analysis of the impact on severe hypoglycaemia and direct costs of the introduction of the FreeStyle Libre sensor in paediatric population with type 1 Diabetes Mellitus.

MATERIAL AND METHODS

Ambispective single-centre study to assess the impact on severe hypoglycaemia and direct costs, focusing on consumption of materials, in paediatric population with type 1 Diabetes Mellitus before and after introduction of the FreeStyle Libre 1 sensor.

RESULTS

A significant decrease was found in episodes of severe hypoglycaemia, with 4.2 episodes of severe hypoglycaemia per 100 patients under follow-up versus 0.25 episodes per 100 patients a year after introduction of the system. This represents a cost difference for severe hypoglycaemia, estimated at €6559.52 before introduction and €409.97 after introduction of the FreeStyle Libre sensor. We found a decrease in the daily consumption of capillary blood glucose strips, which translates as a decrease in the cost of materials and helps mitigate the cost of the sensor. The cost in materials for the patient with FreeStyle Libre was €185.13 per patient and year higher than conventional control with capillary blood glucose strips.

Significance of the CGM metric of time in range in children and adolescents with type 1 diabetes

Continuous glucose monitoring (CGM) has been widely used in children and adolescents as well as adults with type 1 diabetes. CGM metrics include three key measurements of target glucose: time in range (TIR: 70-180 mg/dL), time below range (TBR: <70 mg/dL), and time above range (TAR: >180 mg/dL). The primary goal of optimal glycemic control is to increase TIR to more than 70%, while simultaneously reducing TBR to less than 4%, while minimizing severe hypoglycemia to less than 1%, as proposed by the Advanced Technologies and Treatments for Diabetes (ATTD) panel. However, several studies have indicated that the TIR goal is quite difficult to achieve in pediatric patients who have remarkable interindividual and day-to-day glycemic variation due to their irregular lifestyles. Previous studies have demonstrated that patients without an automated insulin delivery system are unlikely to attain the recommended glycemic goals. On the other hand, reduction of hypoglycemia, particularly minimizing severe hypoglycemia, is a critical issue in the effective management of children with type 1 diabetes. Frequent episodes of severe hypoglycemia and hypoglycemia can cause lasting neurological damage. Accordingly, we propose reducing the TBR to less than 5%, rather than just targeting the TIR to more than 70%. In CGM metrics this should be the cardinal glycemic goal for pediatric patients who are either being treated with multiple daily injections of insulin or a conventional insulin pump, but who are not using an automated insulin delivery system.

Pharmacodynamics, pharmacokinetics, safety, and tolerability of a ready-to-use, room temperature, liquid stable glucagon administered via an autoinjector pen to youth with type 1 diabetes

BACKGROUND

Prompt and reliable management of hypoglycemia in youth with diabetes is important to prevent serious medical complications.

OBJECTIVES

To determine efficacy, pharmacodynamics (PD), pharmacokinetics (PK), safety, and tolerability of a ready-to-use, liquid stable glucagon formulation administered subcutaneously via an autoinjector pen to youth with type 1 diabetes (T1D).

METHODS

After plasma glucose concentration was < 80 mg/dL (< 4.4 mmol/L) after insulin, participants aged 2 to < 12 years with T1D were administered 0.5 mg of glucagon; participants aged 12 to < 18 years instead received 1 mg of glucagon. Then, adolescents were challenged with 0.5 mg after a 7- to 28-day washout period. Primary endpoint was mean plasma glucose concentration at 30 min after glucagon.

RESULTS

Plasma glucose concentrations significantly (p < 0.001) increased from baseline to 30 min after glucagon, with mean change in plasma glucose concentration between baseline and 30 min for each age cohort as follows: 2 to < 6 years (n = 7; 81.4 mg/dL [4.5 mmol/L]); 6 to < 12 years (13; 84.2 mg/dL [4.7 mmol/L]); 12 to < 18 years (11; dose, 1 mg; 54.0 mg/dL [3.0 mmol/L]); and 12 to < 18 years (11; 0.5 mg; 52.4 mg/dL [2.9 mmol/L]). Among age cohorts, no clinically relevant differences were observed for PD and PK parameters. Common adverse events were nausea, vomiting, and hypoglycemia.

CONCLUSION

Age-appropriate dosing of this glucagon formulation was effective at 30 min in reversing plasma glucose concentrations from < 80 mg/dL in youth with T1D.

Scientific advisory on nocturnal hypoglycemia in insulin-treated patients with diabetes: Recommendations from Indian experts

INTRODUCTION

Insulin is one of the commonly prescribed glucose lowering agents in diabetes. Hypoglycemia is the most common complication, and severe hypoglycemia is the most serious complication of insulin therapy. Almost half of all severe hypoglycemia episodes (HEs) occur at night. However, patients are often unaware of their nocturnal hypoglycaemia (NH) risk. Additionally, both healthcare professionals and patients find it difficult to manage NH. The purpose of this expert group meeting is to improve NH awareness and provide guidance for the physicians to recognize and manage NH.

METHOD

The panel of experts in an e-board deliberated extensively upon the available literature and guidelines on hypoglycemia and NH discussed the consensus on definition, detection, reporting, monitoring, treatment, and optimization of therapy in NH.

RESULT

& Conclusion: Though there are many guidelines on the management of HEs in patients with diabetes, very few touch the topic of NH. This scientific advisory on management of NH in insulin treated patients with diabetes is formulated to address this gap in understanding regarding management of NH. The experts provide recommendations for the nocturnal window, defining NH based on blood glucose values, recognition, prevention and management of NH.

Comparison Between Continuous Versus Flash Glucose Monitoring in Children, Adolescents, and Young Adults with Type 1 Diabetes: An 8-Week Prospective Randomized Trial

INTRODUCTION

To assess the impact of real-time continuous glucose monitoring (RT-CGM) instead of first-generation flash glucose monitoring (FGM) on hypoglycaemia in children and adolescents with type 1 diabetes.

METHODS

In this randomized controlled interventional study, young individuals with type 1 diabetes used RT-CGM or FGM for 8 weeks. We evaluated changes in time below range (TBR), severe hypoglycaemia (SH), HbA1c, glycaemic variability, and impaired awareness of hypoglycaemia with RT-CGM (intervention group) in comparison with FGM.

RESULTS

We randomly assigned 37 participants to either the intervention group (n = 19) or the control group (n = 18). At 8 weeks, we did not find a decrease in TBR in either group, but there was a significant reduction in SH in the intervention group. For participants with TBR ≥ 5% at baseline, we observed significant reductions in 24-h TBR, wake TBR, sleep TBR, and glucose variability at 8 weeks in the intervention group.

CONCLUSIONS

The use of RT-CGM versus FGM decreased SH in young individuals with type 1 diabetes, and TBR and glucose variability in patients with a higher TBR at baseline. The patient's history should be taken into account when advising on the method of blood glucose monitoring, as RT-CGM could be more effective in younger patients at high risk for SH.

TRIAL REGISTRATION

ClinicalTrials.gov NCT04249102.

The Cost of Control: Cost-effectiveness Analysis of Hybrid Closed-Loop Therapy in Youth

OBJECTIVE

Hybrid closed-loop (HCL) therapy is an efficacious management strategy for young people with type 1 diabetes. However, high costs prevent equitable access. We thus sought to evaluate the cost-effectiveness of HCL therapy compared with current care among young people with type 1 diabetes in Australia.

RESEARCH DESIGN AND METHODS

A patient-level Markov model was constructed to simulate disease progression for young people with type 1 diabetes using HCL therapy versus current care, with follow-up from 12 until 25 years of age. Downstream health and economic consequences were compared via decision analysis. Treatment effects and proportions using different technologies to define "current care" were based primarily on data from an Australian pediatric randomized controlled trial. Transition probabilities and utilities for health states were sourced from published studies. Costs were considered from the Australian health care system's perspective. An annual discount rate of 5% was applied to future costs and outcomes. Uncertainty was evaluated with probabilistic and deterministic sensitivity analyses.

RESULTS

Use of HCL therapy resulted in an incremental cost-effectiveness ratio of Australian dollars (AUD) $32,789 per quality-adjusted life year (QALY) gained. The majority of simulations (93.3%) were below the commonly accepted willingness-to-pay threshold of AUD $50,000 per QALY gained in Australia. Sensitivity analyses indicated that the base-case results were robust.

CONCLUSIONS

In this first cost-effectiveness analysis of HCL technologies for the management of young people with type 1 diabetes, HCL therapy was found to be cost-effective compared with current care in Australia.

Predictors of associated and multiple autoimmunity in children and adolescents with type 1 diabetes mellitus

PURPOSE

Type 1 diabetes mellitus (T1DM) is an autoimmune condition characterised by the presence of antipancreatic antibodies. The autoimmune process is also directed against other organs, most frequently against the thyroid gland, intestinal mucosa, and gastric parietal cells.

METHODS

Our investigation included 121 children with T1DM with a mean age±standard deviation of 11.99±4.63 years (range, 2.0-20.0 years). We explored the frequency of associated autoimmunity; the presence of predictive factors such as current age, sex, and severity at diabetes diagnosis; T1DM duration; and family history of autoimmunity.

RESULTS

Associated autoimmunity was present in 28.9% of T1DM patients. Children with associated autoimmunity were older at diabetes diagnosis (p=0.009) and had a longer diabetes duration compared to children without associated autoimmunity (p=0.044). Adolescents aged 12-20 years had a statistically significant higher chance of developing thyroid autoimmunity compared to children aged 1-5 years (p=0.019). Multiple autoimmunity (MA), T1DM, and 2 or more autoimmune diseases were present in 5.8% of the study population. All children with MA presented with ketoacidosis at diabetes diagnosis and had a higher percentage of familial autoimmunity (p=0.042). The familial autoimmunity of these patients most frequently affected ≥3 relatives (p=0.026) and was more frequently diagnosed before 5 years of age (p=not significant).

CONCLUSION

Associated autoimmunity was present in almost one-third of T1DM patients. Significant associations with associated autoimmunity were longer diabetes duration, female sex, older age at diabetes diagnosis, and glutamic acid decarboxylase positivity. Predictors of MA were age <5 years at T1DM diagnosis, the presence of diabetic ketoacidosis at diagnosis, and a significant family history of autoimmunity.

Breaking Barriers With Basal Insulin Biosimilars in Type 2 Diabetes

Despite increases in the availability and effectiveness of other therapies, insulin remains an essential treatment for approximately 30 million people with type 2 diabetes worldwide. The development of biosimilars has created the potential for significant health care cost savings and may lead to greater access to basal insulin for vast populations. In this review, we discuss evidence demonstrating equipoise between basal insulin biosimilars and the patented analogs they may replace.

Drug Induced Diabetes Mellitus in Pediatric Acute Lymphoblastic Leukemia: Approach to Diagnosis and Management

Corticosteroids and l -asparaginase used in the treatment of pediatric acute lymphoblastic leukemia (ALL) can cause drug-induced diabetes mellitus (DIDM). DIDM can lead to dyselectrolytemia, a higher risk of infections including cellulitis, bacteremia, fungemia, and a higher incidence of febrile neutropenia and may have an impact on the outcome of ALL. Literature on the management of DIDM among children with ALL is sparse and the diagnostic criteria for pediatric diabetes should be carefully applied considering the acute and transient nature of DIDM during ALL therapy. Insulin remains the standard of care for DIDM management and the choice of Insulin regimen (stand-alone Neutral Protamine Hagedorn or basal bolus) should be based on the type and dose of steroids used for ALL and the pattern of hyperglycemia. A modest glycemic control (postmeal 140 to 180 mg/dL, premeal <140 mg/dL) to prevent complications of hyperglycemia, as well as hypoglycemia, would be the general approach. This review is intended to suggest evidence-based practical guidance in the diagnosis and management of DIDM during pediatric ALL therapy.

Childhood diabetes and sleep

Sleep modulates glucose metabolism, both in healthy states and in disease. Alterations in sleep duration (insufficient and excessive) and obstructive sleep apnea may have reciprocal ties with obesity, insulin resistance and Type 2 diabetes, as demonstrated by emerging evidence in children and adolescents. Type 1 diabetes is also associated with sleep disturbances due to the influence of wide glycemic fluctuations upon sleep architecture, the need to treat nocturnal hypoglycemia, and the need for glucose monitoring and insulin delivery technologies. In this article, we provide an extensive and critical review on published pediatric literature regarding these topics, reviewing both epidemiologic and qualitative data, and provide an overview of the pathophysiology linking sleep with disorders of glucose homeostasis.

In-home use of a hybrid closed loop achieves time-in-range targets in preschoolers and school children: Results from a randomized, controlled, crossover trial

AIM

To obtain additional information on the incremental differences between using a sensor-augmented pump (SAP) without automated insulin delivery (AID), using it with predictive low-glucose management (PLGM) or as hybrid closed loop (HCL), in preschool and school children.

METHODS

We conducted a monocentric, randomized, controlled, two-phase crossover study in 38 children aged 2-6 and 7-14 years. The primary endpoint was the percentage of time in range (TIR) of 70-180 mg/dl. Other continuous glucose sensor metrics, HbA1c, patient-related outcomes (DISABKIDS questionnaire, Fear of Hypoglycaemia Survey) and safety events were also assessed. Results from 2 weeks of SAP, 8 weeks of PLGM and 8 weeks of HCL were compared using a paired t-test or Wilcoxon signed-rank test.

RESULTS

Overall, we found a high rate of TIR target (>70%) achievement with HCL in preschool (88%) and school children (50%), with average times in Auto Mode of 93% and 87%, respectively. Preschool children achieved a mean TIR of 73% ± 6% (+8% vs. SAP, +6% vs. PLGM) and school children 69% ± 8% (+15% vs. SAP and + 14% vs. PLGM). Overall, HbA1c improved from 7.4% ± 0.9% to 6.9% ± 0.5% (P = .0002). Diabetes burden and worries and fear of hypoglycaemia remained at low levels, without significant changes versus PLGM. No events of severe hypoglycaemia or diabetic ketoacidosis occurred.

CONCLUSIONS

Preschool children profit from AID at least as much as those aged 7 years and older. To ensure safe use and prescribing modalities, regulatory approval is also required for young children.

Early vs late histological confirmation of coeliac disease in children with new-onset type 1 diabetes

AIM

Screening for coeliac disease in asymptomatic children with new-onset type 1 diabetes is controversial. The aim of this study was to analyse whether the confirmation of coeliac disease in children with new-onset type 1 diabetes and positive screening results can be postponed.

METHODS

This was a multicentre population-based cohort study based on the German/Austrian/Swiss/Luxembourgian Prospective Diabetes Follow-up Registry (Diabetes Patienten Verlaufsdokumentation [DPV]). Participants aged ≤18 years diagnosed with type 1 diabetes between 1995 and June 2021 and with elevated IgA tissue transglutaminase antibodies (anti-tTGA) at diabetes onset on screening for coeliac disease were included. We compared outcomes of participants with a diabetes duration of more than 1 year between those in whom coeliac disease was confirmed histologically within the first 6 months and those in whom coeliac disease was confirmed between 6 and 36 months after diabetes diagnosis.

RESULTS

Of 92,278 children and adolescents with a diagnosis of type 1 diabetes, 26,952 (29.2%) had documented anti-tTGA data at diabetes onset. Of these, 2340 (8.7%) had an elevated anti-tTGA level. Individuals who screened positive were younger (median age 9.0 vs 9.8 years, p<0.001) and more often female (53.1% vs 44.4%, p<0.001). A total of 533 participants (22.8% of those who screened positive) had a documented biopsy, of whom 444 had documented histological confirmation of coeliac disease. Of 411 participants with biopsy-proven coeliac disease within the first 36 months of diabetes and follow-up data, histological confirmation was performed in 264 (64.2%) within the first 6 months and in 147 (35.8%) between 6 and 36 months after diabetes onset. At follow-up (median diabetes duration 5.3 years and 5.1 years, respectively), estimated median HbA1c levels (62.8 mmol/mol vs 62.2 mmol/mol [7.9% vs 7.8%]), cardiovascular risk markers (lipids, rate of microalbuminuria, blood pressure), rates of acute diabetes complications (diabetic ketoacidosis, severe hypoglycaemia) and the proportions of participants reaching anti-tTGA levels within the normal range did not differ between groups. Participants with delayed histological confirmation of coeliac disease showed no negative effects on growth or weight gain during the observation period.

CONCLUSIONS

Our study suggests that the histological confirmation of coeliac disease in asymptomatic individuals with new-onset type 1 diabetes could be postponed.

Multi-Timepoint Metabolic Fingerprinting of a Post-Episode Period of Hypoglycemia and Ketoacidosis Among Children With Type 1 Diabetes

Background: Acute complications of type 1 diabetes mellitus such as diabetes ketoacidosis (DKA) and hypoglycemia (HG) are detrimental in a short- and long-term perspective. Restoration of normoglycemia and correction of pH do not mean that all metabolic disturbances caused by HG or DKA are immediately reversed.

Aim: This study aimed to identify serum metabolic changes caused by an episode of DKA and HG that may indicate the mechanisms contributing to long-term consequences of DKA/HG.

Materials and methods: Four groups of children with type 1 diabetes were recruited. The first two study groups included patients after an episode of DKA or HG, respectively. Additionally, two comparative groups were recruited—children with established type 1 diabetes (EDM) and patients with newly diagnosed diabetes without diabetes ketoacidosis (NDM). Serum samples were collected in three group-specific time points (since the hospital admission): HG 0h-12h–48h; DKA or NDM 0h-24h–72 h; and one random fasting sample from patients with EDM. Two batches of 100 samples each were created: for DKA batch 20 × 3 DKA patients, 10 × 3 NDM and 10 EDM; for HG batch: 10 × 3 HG patients, 25 EDM and 15 × 3 NDM. All patients within the batches were age and sex matched. Metabolic fingerprinting was performed with LC-QTOF-MS.

Results: Four metabolites were associated with a DKA episode occurring in the preceding 72 h: three were found higher after the DKA episode versus comparative groups: lysophosphatidylcholine (LPC) (18:1), sphingomyelins (SM) (34:0 and d18:0/15:0), and one was found lower: LPC (18:0). Similarly, four metabolites were identified for the HG episode in the last 48 h: three were found higher after the HG episode versus comparative groups: two lysophosphatidylethanolamines (LPE) (18:2 and 20:3) and one LPC (18:2); and one was found lower after the HG episode: oxy-phosphatidylocholine (PC O-34:4).

Conclusions: We found eight metabolites whose levels may be traced in the serum, indicating the DKA or HG episode for up to 72 h and 48 h, respectively. Acute complications of diabetes may cause persistent metabolic disturbances long after pH and glucose level normalization.