Despite substantial advances in insulin delivery and glucose monitoring technologies, optimisation of glycaemic control in type 1 diabetes (T1D) remains challenging, particularly postprandially [1]. This likely reflects imperfect coordination between the timing of prandial insulin action and intestinal absorption of carbohydrates [2]. Gastric emptying, which governs the rate at which nutrients enter the small intestine for absorption, is a key determinant of postprandial glycaemia in both health and diabetes [3]. In health, it occurs at a relatively constant caloric rate in the range of 1–4 kcal/min [3]. In T1D, gastric emptying exhibits substantially greater variability due to a high prevalence of both accelerated and delayed emptying [3]. That individuals with delayed emptying (gastroparesis) require less insulin early postprandially (0–2 h), but more later (3–4 h) to maintain euglycaemia [4], suggests that measurement of gastric emptying may inform more personalised and effective use of prandial insulin. The majority of studies examining gastric emptying in T1D have been conducted in Western populations and information relating to Asian cohorts, particularly Han Chinese individuals, is limited, despite evidence suggesting that Han Chinese may have more rapid gastric emptying [5]. This study aimed to characterise gastric emptying in Chinese individuals with T1D, determine whether it differs from that of healthy individuals, and examine its relationship with the glycaemic response to an oral glucose load and prandial insulin requirements. Twenty-eight Chinese participants with T1D (Table 1) were recruited from the Department of Endocrinology and Metabolism at Nanjing First Hospital, China, during hospitalisation for optimisation of insulin therapy. Participants were diagnosed according to the 2020 Chinese Diabetes Society criteria, were aged < 75 years, and provided informed consent. Review of their medical records indicated that 12 had peripheral neuropathy and 5 had diabetic retinopathy. None had significant gastrointestinal symptoms, a history of gastrointestinal surgery, or the use of medications affecting gastrointestinal function. The study was approved by the Human Ethics Committee of Nanjing First Hospital (ID: KY20230915-11) and registered prospectively (NCT06173934). Data from 19 healthy participants of comparable age, BMI and sex distribution were derived from a published study [6] and used as a reference for interpretation. Following hospital admission, T1D participants underwent clinical biochemistry assessments. Screening was conducted on Day 1 to confirm eligibility. From Day 2, participants were provided with standardised meals (at ~07:00, ~11:00 and ~17:00 h; total 1512 kcal; carbohydrate 46%, fat 23% and protein 31%). Basal and bolus insulin doses were optimised over 3–4 days guided by seven-point capillary glucose monitoring until relatively acceptable glycaemic control was achieved (Figure S1). Prandial insulin dose relative to basal or total insulin was calculated to estimate prandial insulin requirement while accounting for inter-individual differences in insulin sensitivity. Immediately following the completion of insulin titration, participants ingested a 300 mL drink containing 75 g glucose and 150 mg 13C-acetate, within 5 min (t = 0–5 min) after an overnight fast. Breath samples were collected every 15 min over 3 h to assess gastric emptying. Venous blood samples were obtained every 30 min to measure plasma glucose. As reported previously [6], healthy participants underwent an identical procedure for measurements of GE and plasma glucose concentrations. 13CO2 in breath samples was measured by an isotope ratio mass spectrometer (HCBT-01; Headway, Shenzhen, China). Gastric half-emptying time (T50) was calculated using the Wagner-Nelson method, as previously described [7, 8]. HbA1c was measured by high-performance liquid chromatography (Bio-Rad, CA, USA), and plasma glucose by the glucose oxidase method (Modular Analytics E170; Roche, Mannheim, Germany). Data distribution was assessed using the Shapiro–Wilk test. Normally distributed data are presented as mean ± SD, while non-normally distributed data are presented as median (interquartile range). The primary endpoint was the relationship between T50 and postprandial glycaemia. Between-group comparisons were performed using unpaired Student's t-tests, Wilcoxon signed-rank test, analysis of Covariance (ANCOVA), or Fisher's exact test, where appropriate. Relationships between the incremental area under the glucose curves (iAUCs, calculated as the total AUCs subtracting the baseline values) or prandial insulin dose and T50 in T1D participants were assessed using Pearson's correlation analysis. Plasma glucose responses between the two groups were analysed using two-way repeated-measures ANOVA. All analyses were conducted using GraphPad Prism (Version 9.0, La Jolla, CA, USA), with the exception of the comparison of T50 between T1D and healthy participants, which was performed using SPSS Statistics (Version 24; IBM, NY, USA) with adjustment for sex and BMI. p < 0.05 was considered statistically significant. T50 varied from 48.7 to 135.2 min (mean ± SD: 83.71 ± 27.36 min) in healthy participants and 52.1–192.8 min [median (interquartile range): 93.35 (74.18, 123.88) min] in T1D participants, without significant difference between groups both before and after adjustment for sex and BMI (p = 0.123 and 0.054 respectively) (Figure 1A). Plasma glucose concentrations before and after the glucose drink were higher in T1D than healthy participants (Figure 1B). Plasma glucose iAUCs between 0–60, 0–120 and 0–180 min following the glucose drink were all related inversely to T50 in T1D participants (r = −0.542, p = 0.003; r = −0.638, p < 0.001 and r = −0.548, p = 0.003, respectively) (Figure 1C–E). Furthermore, prandial insulin dose following stabilisation of blood glucose was related inversely to T50, both before and after adjustment for basal insulin dose or total daily insulin dose (r = −0.462, p = 0.013; r = −0.425, p = 0.024 and r = −0.382, p = 0.045, respectively) (Figure 1F–H). This study demonstrates that in Chinese with T1D, both the glycaemic response following an oral glucose load and the prandial insulin requirement for mixed meals are related directly to the rate of gastric emptying (inversely to T50), supporting the concept that gastric emptying should be incorporated into algorithms determining prandial insulin delivery. To our knowledge, this is the first study to characterise gastric emptying in a Chinese population with T1D. Compared with healthy controls, mean gastric emptying of the T1D participants appeared to be numerically (but insignificantly) slower, which may reflect the prevalence of autonomic neuropathy in a subset of individuals, given that a modest prevalence of microvascular complications was observed in this cohort. Consistent with previous reports in Caucasians with T1D [9, 10], the inter-individual variation in gastric emptying was also more marked in Chinese with T1D. The variability in gastric emptying is likely to contribute to the heterogeneity in postprandial glycaemic responses observed in the management of T1D. Consistent with findings in previous studies [6, 9], both early and overall glycaemic excursions following the glucose drink were related inversely to T50 in Chinese individuals with T1D, highlight the importance of gastric emptying in determining the glycaemic response to carbohydrate intake. The inverse relationship between prandial insulin dose and T50 further underscores the clinical relevance of gastric emptying in guiding more individualised prandial insulin dosing strategies in T1D. Importantly, gastric emptying can be assessed using a 13C-breath test, which entails no radiation exposure and is convenient [11], making it highly feasible for routine clinical implementation. Several limitations should be noted while interpreting our findings. First, the sample size was modest, particularly for the healthy control group derived from a previous study. Second, gastric emptying was assessed using a glucose drink rather than a solid meal. Gastric emptying of solids is characterised by an initial lag phase, reflecting trituration, followed by a more linear emptying pattern. In contrast, non-nutrient liquids empty in an approximately exponential manner without a lag phase, although increasing nutrient content slows gastric emptying and shifts it towards a more linear, caloric-dependent process. While nutrient-containing liquids empty at a caloric rate comparable to that of digestible solids in health [12], their relationship is relatively weak in individuals with longstanding diabetes [13]. Accordingly, caution is required when extrapolating these findings to solid meals. However, the T50 of a glucose drink has been shown to predict the glycaemic response to mixed meals in T2D [6] and to correlate with prandial insulin dose titrated based on glycaemic responses to mixed meals in the current study, supporting its physiological relevance. Finally, gastric emptying was assessed on a single occasion; however, its intra-individual variability is known to be small and not influenced by spontaneous fluctuations in blood glucose in T1D [14]. In conclusion, gastric emptying is a key determinant of postprandial glycaemic responses and prandial insulin requirements in Chinese individuals with T1D, which supports the concept that routine assessment of gastric emptying may facilitate more personalised insulin therapy in T1D. Y.L. was involved in the conception and design of this study, recruitment and clinical management of study participants, data collection and analysis and drafting of the manuscript. Z.J. and Y.Z. were involved in data collection, statistical analysis and drafting of the manuscript. Y.L. was involved in data collection and statistical analysis. C.X. and Y.S. were involved in data analysis and reviewing of the manuscript. M.H., K.L.J. and C.K.R. were involved in the conception of this study and reviewing of the manuscript. J.M. and T.W. were involved in the conception and design of the study, data interpretation, statistical analysis and revisions of the manuscript and were guarantors of this work. Open access publishing facilitated by Adelaide University, as part of the Wiley - Adelaide University agreement via the Council of Australasian University Librarians. The authors have nothing to report. The authors declare no conflicts of interest. 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. The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/dom.70811. Figure S1: The seven-point blood glucose profiles for type 1 diabetes participants. Data are means ± SEM. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
No takes yet. Share an insight, caveat, or question.
Luo et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: