Relative contributions of basal vs postprandial hyperglycaemia to overall glycaemic control in newly diagnosed Chinese adults with type 2 diabetes before and after 3 months' intensive glucose‐lowering therapy

ABSTRACT

Understanding the relative contributions of basal and postprandial hyperglycaemia (BH and PPH) to hyperglycaemia is pivotal to optimising the therapeutic strategies for type 2 diabetes (T2D). We used continuous glucose monitoring (CGM) to quantify longitudinal changes in the relative contributions of BH and PPH to overall hyperglycaemia in 22 treatment‐naïve Han Chinese adults with T2D before and after 3 months of intensive glucose‐lowering therapy. At baseline (HbA1c 10.0 ± 0.3%), BH predominated, contributing ~73–77% of overall hyperglycaemia. Following treatment (HbA1c 7.1 ± 0.2%), the contribution of BH decreased markedly (~35–38%), while PPH became the dominant contributor of hyperglycaemia (~62–66%). The improvement in glycaemic control was also accompanied by enhanced insulin secretion and sensitivity, without changes in gastric emptying. These findings provide longitudinal, CGM‐based evidence of a dynamic shift in glycaemic determinants with therapy. Therapeutic strategies should be adapted accordingly, with initial focus on BH followed by increasing emphasis on PPH as glycaemic control improves.

In newly‐diagnosed Chinese with T2D,

• Contribution of basal hyperglycaemia to hyperglycaemic exposure decreased substantially and the relative importance of postprandial hyperglycaemia increased markedly after 3 months glucose‐lowering therapy.

• Therapy should be re‐evaluated after initial treatment, and that a greater focus on limiting postprandial glucose excursions may then be warranted.

INTRODUCTION

Achieving tight glycaemic control (i.e. HbA1c ≤7% or even 6.5%) is a key therapeutic goal in type 2 diabetes (T2D) ¹ . Understanding the relative contributions of basal and postprandial hyperglycaemia (BH and PPH) to hyperglycaemia is pivotal to the therapeutic approach ² . Cross‐sectional analysis of self‐reported data from people with T2D indicate that BH is the dominant contributor when HbA1c is >8.0%, while PPH predominates when HbA1c is <7.5% ³ . However, the dynamic changes in their contributions within individuals remain poorly characterised due to limitations of self‐reported ⁴ , ⁵ or brief daytime glucose profiles ⁶ .

We have quantified, for the first time using continuous glucose monitoring (CGM), the within‐individual longitudinal changes in the relative contributions of BH and PPH to overall hyperglycaemia before and after 3 months of intensive glucose‐lowering therapy in treatment‐naïve Han Chinese with newly diagnosed T2D. This approach overcomes the limitations of prior cross‐sectional studies and intermittently sampled glucose profiles, providing precise insights into the evolving burden of hyperglycaemia. This population is also distinctive from Caucasians with T2D, typically presenting with marked hyperglycaemia, substantially impaired beta‐cell function and insulin sensitivity and accelerated gastric emptying ⁷ . Given the relevance of these parameters to the regulation of basal and/or postprandial glycaemia ⁸ , ⁹ , ¹⁰ , ¹¹ , a 75 g oral glucose tolerance test (OGTT) was also performed before and following glucose‐lowering therapy.

METHODS

We recruited 22 treatment‐naïve Han Chinese adults with newly diagnosed T2D (HbA1c 10.0 ± 0.3%) from Nanjing First Hospital, China. Ethics approval (ID: KY 20220214–08) and clinical registration (NCT05284344) were obtained. All participants provided written informed consent and were hospitalised for evaluation.

On Day 1, a CGM device (Medtronic Incorporated, USA) was fitted. On Day 2, participants consumed standardised meals and underwent 24 h CGM (7 am–7 am) as reported previously ⁷ . On Day 3, they ingested a glucose drink containing 75 g glucose and 150 mg ¹³C‐acetate. Venous blood was collected over 180 min for measurements of plasma glucose and serum insulin and C‐peptide (Modular Analytics E170, Roche, Germany), and breath samples for measurement of the gastric half‐emptying time (T50) ¹² . Insulin sensitivity was assessed by the Matsuda index ¹³ and HOMA‐IR ¹⁴ . Beta‐cell mass index was calculated as C‐peptide AUC_0‐30min/glucose AUC_0‐30min ¹⁵ . Between Days 4 and 7, participants commenced insulin pump therapy (insulin aspart) ¹⁶ for ‘rescue’ of hyperglycaemia and received structured diabetes education. At discharge, they were prescribed: metformin alone (n = 12), metformin + a DPP‐4 inhibitor (n = 2), metformin + a SGLT‐2 inhibitor (n = 4) and metformin + a GLP‐1RA (n = 4). The heterogeneity of post‐discharge pharmacotherapy reflects real‐world clinical practice; accordingly, the primary objective of the study was to assess overall changes in glycaemic patterns rather than drug‐specific effects. After 3 months and 48 h ‘washout’, reassessments were performed.

BH and PPH were calculated from CGM using 6.1 mmol/L as the normal fasting glucose cutoff (the World Health Organization criterion) ¹⁷ . Hyperglycaemia was defined as the total area under the glucose curve above 6.1 mmol/L (AUC_T), discounting negative values. PPH was estimated as AUC above fasting glucose (Model A: AUC_P‐A) ³ or premeal glucose levels over 4 h (Model B: AUC_P‐B) ¹⁸ . Relative contributions of BH and PPH were calculated as (AUC_T‐AUC_P)/AUC_T*100% and AUC_P/AUC_T*100%, respectively ³ . A sensitivity analysis used 5.6 mmol/L (the American Diabetes Association criterion) as the cutoff.

Power analysis indicated that 22 participants yielded >80% power at α = 0.05 to detect a 30% change in BH contribution ³ . Data are expressed as means ± SEM or medians (interquartile range) and analysed using paired Student's t‐tests or the Wilcoxon signed‐rank tests (GraphPad, 188 La Jolla, CA, USA). P < 0.05 was considered statistically significant.

RESULTS

All participants tolerated therapy without hypoglycaemia or weight change. After 3 months, HbA1c decreased from 10.0 ± 0.3% to 7.1 ± 0.2% (Table 1), with an improvement in glycaemic excursions (Figure S3) and plasma insulin and C‐peptide response to oral glucose (Figure S4). Although gastric emptying was not different before and after treatment (Figure S5), before treatment, BH was the major contributor to hyperglycaemia: 72.5 ± 2.2% (Model A) and 77.3 ± 1.5% (Model B). After treatment, BH contribution decreased to 38.1 ± 10.7% or 34.5 ± 8.8%, respectively, while PPH increased to 61.9 ± 10.7% or 65.5 ± 8.8% (Figure 1). Sensitivity analyses (cutoff 5.6 mmol/L) showed similar patterns (Figure S1). Glucose‐lowering therapy improved insulin and C‐peptide responses to oral glucose, beta‐cell mass index and insulin sensitivity markers, but did not alter gastric emptying (Table 1). In the small subgroup receiving GLP‐1 receptor agonists (n = 4), no significant change in the relative contribution of PPH was observed after treatment (P = 0.52); however, this analysis was exploratory and markedly underpowered (Figure S2A). In the participants receiving other glucose‐lowering treatments, the relative contribution of PPH was 22.4 ± 1.8% at baseline and increased to 68.4 ± 9.9% after treatment (P = 0.002) (Figure S2B).

Table 1.

Demographics of Chinese participants with newly diagnosed type 2 diabetes (T2D)

	Baseline (n = 22)	Post‐treatment (n = 22)	P
Sex (male/female)	11/11	11/11	–
BMI (kg/m2)	25.5 ± 0.7	25.2 ± 0.7	0.25
HbA1c (%)	10.0 ± 0.3	7.1 ± 0.2	0.003
HbA1c (mmol/mol)	85.8 ± 3.1	54.4 ± 2.5	0.003
Gastric half‐emptying time (T50 min)	85.6 ± 5.9	93.8 ± 11.3	0.29
24‐h interstitial glucose recorded by CGM
Mean glucose (mmol/L)	13.7 ± 0.4	7.5 ± 0.3	0.001
SD (mmol/L)	2.9 ± 0.2	1.6 ± 0.2	0.002
CV (%)	21.3 (16.6, 24.9)	19.9 (10.1, 25.9)	0.44
MAGE (mmol/L)	6.5 ± 0.4	4.2 ± 0.5	0.01
TIR (%)	7.5 (0.0, 26.9)	94.8 (88.5, 98.6)	0.001
Insulin sensitivity and beta‐cell function derived from 75 g oral glucose tolerance test
Insulin AUC0–120 min (uU/mL*min)	2289.0 ± 415.5	3932.0 ± 647.6	0.003
C‐peptide AUC0–120 min (ng/mL*min)	458.9 ± 41.9	726.8 ± 97.8	0.002
Matsuda index	3.5 ± 0.3	5.9 ± 0.9	0.005
HOMA‐IR	4.7 ± 0.7	2.1 ± 0.3	0.001
Beta‐cell mass index (%)	1.0 ± 0.1	1.8 ± 0.2	0.001
Overall and postprandial hyperglycaemia recorded by 24 h‐CGM
AUCT (mmol/L*min)	11161.0 ± 587.7	2336.0 ± 351.4	0.001
AUCP‐A (mmol/L*min)	3084.0 ± 263.5	1219.0 ± 204.1	0.001
AUCP‐B (mmol/L*min)	2487.0 ± 159.8	1231.0 ± 142.7	0.001

Normally distributed data are expressed as means ± SEM, while data that were not normally distributed are expressed as medians (interquartile range). Differences between baseline and post‐treatment were compared by paired Student's t‐tests or Wilcoxon signed‐rank tests, where appropriate.

AUC_P‐A, postprandial hyperglycaemia (Model A); AUC_P‐B, postprandial hyperglycaemia (Model B); AUC_T, hyperglycaemia over 24 h; BMI, body mass index; CV, glucose coefficient of variation; HOMA‐IR, Homeostasis Model Assessment of insulin resistance; MAGE, mean amplitude of glycaemic excursion; SD, standard deviation of 24‐h mean interstitial glucose; TIR, time in range between 3.9 and 10.0 mmol/L.

Figure 1.

Relative contributions of basal and postprandial hyperglycaemia (BH and PPH) to overall hyperglycaemia in participants with newly diagnosed type 2 diabetes (T2D) before and after 3 months of glucose‐lowering therapy (n = 22). Overall hyperglycaemia was calculated as the total area under the glucose curve above 6.1 mmol/L (AUC_T). PPH was calculated as the AUC above the fasting glucose level (Model A: AUC_P‐A) and the AUC above premeal glucose levels over 4 h (Model B: AUC_P‐B). The relative contributions of BH and PPH to overall hyperglycaemia were calculated as (AUC_T‐AUC_P)/AUC_T*100% and AUC_P/AUC_T*100%, respectively. Data are means ± SEM.

DISCUSSION

This study showed that (i) in newly diagnosed Han Chinese with T2D with marked hyperglycaemia, BH was the primary contributor before treatment, (ii) intensive pharmacotherapy improved glycaemic control, beta‐cell function and insulin sensitivity, without affecting gastric emptying, (iii) the relative contribution of BH fell markedly with treatment, while PPH became the dominant determinant of hyperglycaemia and (iv) in a small number of patients receiving a GLP‐1RA, the improvement in overall glycaemic control was associated with numerically (but not statistically) delayed gastric emptying and a minimal increase in the contribution of PPH to hyperglycaemia.

While the concept that BH predominates at higher HbA1c and PPH at lower HbA1c is well established, prior evidence has been derived largely from cross‐sectional analyses or limited glucose sampling. Our study extends this framework by providing longitudinal, CGM‐based evidence within individuals, demonstrating the dynamic transition in glycaemic contributions with therapy in a treatment‐naïve cohort. These observations suggest that glucose‐lowering therapy in people with newly diagnosed, poorly controlled T2D should initially target BH. However, as glycaemic control improves, PPH assumes greater importance and appears to become a key limiting factor in achieving target HbA1c. This is consistent with findings in trials in which normalisation of fasting glucose with insulin glargine failed to achieve HbA1c targets due to residual PPH ¹⁹ .

We recently reported that Han Chinese with newly diagnosed T2D often exhibit abnormally rapid gastric emptying, which contributes to postprandial hyperglycaemia ⁷ . Moreover, 4 weeks of insulin pump therapy lowered fasting and PPH markedly and slowed gastric emptying in those patients, which contrasts with the effect of 3 months of pharmacotherapy on gastric emptying in the current study. This discrepancy suggests that gastric emptying is not substantially modulated by chronic hyperglycaemia, but rather by specific interventions. In the small subgroup of participants receiving long‐acting GLP‐1 receptor agonists, there was a non‐significant tendency towards delayed gastric emptying, accompanied by minimal change in the relative contribution of PPH. However, given the very small sample size and the limited washout period, these findings should be interpreted with caution and regarded as exploratory. Although metformin is known to slow gastric emptying ²⁰ , it was discontinued (>7 half‐lives) prior to reassessments.

Several limitations should be acknowledged. The sample size was modest, and subgroup analyses, particularly in those receiving GLP‐1 receptor agonists, were exploratory and underpowered. Post‐discharge pharmacotherapy was heterogeneous, reflecting real‐world practice, but may introduce confounding and preclude attribution of effects to specific agents. Furthermore, despite a 48‐h washout, residual effects of longer‐acting therapies cannot be excluded. Finally, the cohort comprised newly diagnosed Han Chinese individuals with marked hyperglycaemia, which may limit generalisability to other populations, disease stages or treatment settings.

Our findings highlight the need to re‐evaluate therapy after initial treatment, with increasing emphasis on limiting postprandial glucose excursions to achieve optimal glycaemic targets.

FUNDING

Y. Sun is supported by an Adelaide University Research Scholarship, C. Xie by a Royal Adelaide Hospital Florey Fellowship, C. K. Rayner by the Michell Bequest Foundation and T. Wu by the Australian Medical Research Future Fund (MRFCDDM000009).

AUTHOR CONTRIBUTIONS

Y. Sun was involved in clinical data collection, statistical analysis and writing of the manuscript; Y. Luo contributed to the recruitment and clinical management of participants and data collection. C. Xie, K. L. Jones, M. Horowitz and C. K. Rayner were involved in data interpretation and reviewing of the manuscript. J. Ma and T. Wu were involved in the conception and design of the study, data interpretation, statistical analysis and drafting of the manuscript, and were guarantors of this work.

DISCLOSURE

T. Wu has received research funding from DSM‐Firmenich. K. L. Jones has received research funding from Diabetology UK. The remaining authors have no personal or financial conflict of interest to declare.

Approval of the research protocol: N/A.

Informed consent: N/A.

Registry and the registration no. of the study/trial: This study was approved by the Human Ethics Committee of Nanjing First Hospital (approval number: KY 20220214‐08) and prospectively registered at ClinicalTrials.Gov (NCT05284344). All participants provided written informed consent.

Animal studies: N/A.

Supporting information

Figure S1. Relative contributions of basal and postprandial hyperglycaemia (BH and PPH) to overall hyperglycaemia in participants with newly diagnosed type 2 diabetes (T2D) before and after 3 months of glucose‐lowering therapy. Overall hyperglycaemia was calculated as the total area under the glucose curve above 5.6 mmol/L (AUC_T). PPH was calculated as the AUC above the fasting glucose level (Model A: AUC_P‐A) (A) and the AUC above premeal glucose levels over 4 h (Model B: AUC_P‐B) (B). The relative contributions of BH and PPH to overall hyperglycaemia were calculated as (AUC_T‐AUC_P)/AUC_T*100% and AUC_P/AUC_T*100%, respectively. Data are means ± SEM.

Figure S2. Relative contributions of basal and postprandial hyperglycaemia (BH and PPH) to overall hyperglycaemia in participants with newly diagnosed type 2 diabetes (T2D) before and after 3 months of glucose‐lowering therapy, stratified according to the use of GLP‐1 RAs. BH was calculated as the total area under the glucose curve above 6.1 mmol/L (AUC_T). PPH was calculated as the AUC above the fasting glucose level (Model A: AUC_P‐A) (A) and the AUC above premeal glucose levels over 4 h (Model B: AUC_P‐B) (B). The relative contributions of BH and PPH to overall hyperglycaemia were calculated as (AUC_T‐AUC_P)/AUC_T*100% and AUC_P/AUC_T*100%, respectively. Data are means ± SEM.

Figure S3. The 24‐h interstitial glucose profile (Day 2: 7 am ‐ Day 3: 7 am) recorded by a continuous glucose monitoring (CGM) device in participants with newly diagnosed type 2 diabetes (T2D) at baseline and after 3 months of treatment (n = 22). Data are means ± SEM.

Figure S4. Fasting and post‐OGTT plasma glucose (A), serum C‐peptide (B) and insulin (C) concentrations in participants with newly diagnosed type 2 diabetes (T2D) before and after 3 months of glucose‐lowering therapy (n = 22). Data are means ± SEM.

Figure S5. Gastric half‐emptying time (T50) in newly diagnosed type 2 diabetes (T2D) before and after 3 months of glucose‐lowering therapy (n = 22). Data are means ± SEM.

DATA AVAILABILITY STATEMENT

The datasets generated during and/or analysed during the current study are not publicly available but can be provided by the corresponding authors upon reasonable request.