What does this research mean for the field?

Mean Hemoglobin Glycation Index (mHGI) calculated using the Glucose Management Indicator (GMI) is a stable patient phenotype independent of mean blood glucose and therapy, and can reliably classify patients into complication risk categories using four or more measurements. Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

This research aims to assess the stability of the mean Hemoglobin Glycation Index (mHGI) in youth with type 1 diabetes before and after starting automated insulin delivery.

June 7, 2026

2945-LB: The Glucose Management Indicator (GMI) Enables Calculation of the Hemoglobin Glycation Index (HGI) and Reliable Patient Classification into High and Low Risk Phenotypes for Complications

Key Points

This research aims to assess the stability of the mean Hemoglobin Glycation Index (mHGI) in youth with type 1 diabetes before and after starting automated insulin delivery.
Conducted repeated measures assessment with HbA1c and GMI data from 145 youth with type 1 diabetes over 90 days.
Calculated HGI as the difference between HbA1c and GMI; classified patients into tertiles based on mHGI.
Utilized mixed-effects models to analyze variability and reliability of mHGI classification.
40% of HGI variation was linked to individual factors; mean mHGI was -0.14±0.38.
Both HbA1c and mean blood glucose decreased after AID, while mHGI remained stable across the study.
Reliability for mHGI classification was high, increasing to 0.81 with six repeated measures.

Abstract

Introduction and Objective: HGI quantifies individual variation in HbA1c level not due to mean blood glucose (MBG) and is strongly associated with complications risk. Continuous Glucose Monitoring provides a predicted HbA1c as the GMI, enabling easy HGI calculation. Using repeated measures assessment, we evaluated whether mean HGI (mHGI) is a stable patient phenotype both before and after start of automated insulin delivery (AID) as well as the effect of ethnicity, MBG and sex. Methods: HbA1c and GMI data (90 days) were obtained before (1-2 measures) and after (2-3 measures) the start of AID in 145 youth with type 1 diabetes (age 12.4±3.6 y; duration 5.9±3.8 y). HGI was calculated as HbA1c-GMI. Patients were ranked by their mHGI from lowest to highest and divided into tertiles. Mixed-effects models accounted for missing data, quantified sources of HGI variability, and estimated reliability of mHGI tertile classification. Results: Based on 435 measures, 40 % of HGI variation was attributable to individual factors. Overall, mean mHGI was -0.14±0.38 (range -1.12 to 0.85). Both HbA1c and MBG declined after AID initiation, while mHGI remained stable. mHGI was higher in African heritage compared with White and South Asian groups (p0.001), and slightly decreased with age (p=0.0114). mHGI slightly decreased after AID in African heritage (p = 0.0472), but not in White or South Asian groups. mHGI was not associated with MBG or sex. Reliability for classification into Low or High mHGI phenotypes increased with repeated measures: 0.74 with 4, 0.78 with 5, and 0.81 with 6 repeats. Conclusion: GMI enables easy calculation of HGI. Over time mHGI represents a stable phenotype independent of MBG and therapy. Four or more HGI measurements have high reliability for classification of individuals into low or high mHGI risk phenotypes, This data supports the use of GMI derived mHGI as an individualized characteristic for diabetes assessment and management. Disclosure J. Pemberton: Consultant; Current; Roche Diabetes Care, Abbott Diabetes, Dexcom, Inc., Insulet Corporation, Tandem Diabetes Care, Inc. R.E. Krone: None. S. Uday: None. Z. Fang: None. S. Chalew: None.

Bookmark

Cite This Study

PEMBERTON et al. (Fri,) studied this question.

synapsesocial.com/papers/6a250ae37def13d035e1ae69 https://doi.org/https://doi.org/10.2337/db26-2945-lb

Bookmark