Metabolomics Perturbations In Patients With Varying Glycaemic Variability And The Effect Of Meal Replacement On Glycaemic Variability

Objective: There is emerging evidence that glycaemic variability(GV)plays an important role in the development of diabetic complications.However,there is no study investigating the metabolic perturbations in patients with different characteristics of glycemia.In addtion,how to effectively alleviate GV remains to be elucidated.The current study aimed to compare the metabolic profiles of postload hyperglycemia and fasting hyperglycemia,and evaluate the effects of meal replacement(MR)on GV.Methods: In the metabolomics study,patients with newly diagnosed diabetes were stratified into fasting hyperglycemia(FH)group or postload hyperglycemia(PH)group.A total of 46 patients(FH group,n = 23;PH group,n = 23)were included as the exploratory set,and another 40 patients(FH group,n = 20;PH group,n = 20)were recruited as the validation set.In the meal replacement study,a total of 123 patients with newly diagnosed and untreated type 2 diabetes were randomized to receive either lifestyle intervention(LI)together with breakfast meal replacement(LI+MR)(n = 62)or LI alone(n = 61)for 4 weeks and completed the study.Each participant underwent 72-hour continuous glucose monitoring(CGM)both before and after intervention.GV assessed by CGM was compared.Results:(1)The metabolomics study showed that:(1)In the exploratory set,10 metabolites were selected as differential metabolites of PH(vs.FH).Of them,mannose and 5-aminoimidazole-4-carboxamide ribonucleotide(AICAR)were confirmed in the validation set to be significantly higher in FH than in PH(P < 0.05);(2)In the pooled dataset,mannose was independently associated with fasting plasma glucose(? = 0.151,P = 0.035)and HOMA-IR(? = 0.160,P = 0.026),respectively;(2)And the MR study part demonstrated that:(1)LI+MR caused significantly greater improvements in standard deviation of blood glucose values(LI+MR: ?=-0.6 ± 0.8 mmol/L,LI: ?=-0.2 ± 0.7 mmol/L,P = 0.005),glucose coefficient of variation(LI+MR: ?=-4.7 ± 8.1%,LI: ?=-0.2 ± 7.6%,P = 0.002),the mean amplitude of glycaemic excursions(LI+MR: ?=-1.6 ± 2.3 mmol/L,LI: ?=-0.6 ± 2.1 mmol/L mm Hg,P = 0.016)and incremental area under the curve of postprandial blood glucose(LI+MR: ?=-287.8 ± 368.7 min*mmol/L,LI: ?=-31.2 ± 301.6 min*mmol/L,P < 0.001)than did LI;(2)Multivariate linear regression showed that LI+MR(vs.LI)was independently associated with improvements in GV after adjustment of covariates(all P < 0.05).Conclusions: FH and PH exhibited distinct metabolic profiles.The perturbation of mannose may be involved in the pathophysiologic disturbances in diabetes.LI+MR led to significantly greater improvements in GV than LI.