The relative contribution of diurnal and nocturnal glucose exposures to HbA1c in type 1 diabetes males: a pooled analysis

Abstract

Purpose: The exact contribution of daily glucose exposure to HbA1c in people with type 1 diabetes (T1D) remains controversial. We examined the contribution of pre- and postprandial glycaemia, nocturnal and early-morning glycaemia, and glycaemic variability to HbA1c levels in T1D. In this analysis, we used clinical data, namely age, BMI and HbA1c, as well as glycaemic metrics (24-h glycaemia, postprandial, nocturnal, early-morning glycaemia, wake-up glucose, and glycaemic variability) obtained over a four-week period of continuous glucose monitoring (CGM) wear in thirty-two males with T1D.

Methods: The trapezoid method was used estimate the incremental area under the glucose curve (iAUC) for 24-h, postprandial (3-h period following breakfast, lunch, and dinner, respectively), nocturnal (between 24:00-04:00 AM), and early-morning (2-h period 2-h prior to wake-up) glycaemia. Linear regression analysis was employed whereby CGM-derived glycaemic metrics were explanatory variables and HbA1c was the outcome.

Results: Thirty-two T1D males (mean ± SD: age 29 ± 4 years; HbA1c 7.3 ± 0.9% [56 ± 13 mmol/mol]; BMI 25.80 ± 5.01 kg/m²) were included in this analysis. In linear models adjusted for age and BMI, HbA1c was associated with 24-h mean glucose (r ² = 0.735, p < 0.001), SD (r ² = 0.643, p = 0.039), and dinner iAUC (r ² = 0.711, p = 0.001). CGM-derived metrics and non-glycaemic factors explained 77% of the variance in HbA1c, in which postprandial glucose accounted for 32% of the variance explained. The single greatest contributor to HbA1c was dinner iAUC resulting in 0.6%-point (~7 mmol/mol) increase in HbA1c per SD increase in dinner iAUC.

Conclusions: Using comprehensive CGM profiling, we show that postprandial glucose, specifically evening-time postprandial glucose, is the single largest contributing factor to HbA1c in T1D.

Trial registration number: NCT02204839 (July 30th 2014); NCT02595658 (November 3rd 2015).

Keywords: CGM; Glucose variability; HbA1c; Postprandial glucose; Type 1 diabetes.

Fig. 1

A-J Participant CGM-derived metrics. Breakfast, lunch, and dinner iAUC was calculated for 3-h after each meal. Early morning iAUC was calculated for a 2-h period prior to wake-up time. Nocturnal iAUC was calculated between 02:00–04:00 for all participants

Fig. 2

Pairwise scatter plots of the interrelationships between CGM-derived metrics with mean and 95%CI

Fig. 3

A-J Pairwise scatter plots of the interrelationships between CGM-derived metrics vs. HbA1c, with mean and 95%CI

Fig. 4

A-C (A) Mean (95%CI) difference in HbA1c (%) by 1-SD difference in CGM-derived glycaemic metrics. Estimated differences are unadjusted (grey) and adjusted (black) for age and BMI. (B) Proportion of variance explained in HbA1c by postprandial factors (black bar; breakfast iAUC, lunch iAUC, dinner iAUC), preprandial factors (grey bar; nocturnal iAUC, early-morning iAUC, wake-up glucose), glycaemic variability (hashed bar; SD, CV%), and non-glycaemic factors (white bar; age, BMI). (C) Sensitivity analysis of proportion of variance explained in HbA1c by postprandial factors (black bar; dinner iAUC), preprandial factors (grey bar; nocturnal iAUC), glycaemic variability (hashed bar; SD), and non-glycaemic factors (white bar; age, BMI)