Objective:It has been shown thatβ-cell function reduced by 50%at the time of T2D diagnosis.Traditional parameters could only reflect limited aspects of the complex process ofβ-cell dysfunction.β-cell glucose sensitivities,calculated from C-peptide,estimate the early stage impairment ofβ-cells.However,in the clinical setting,C-peptide was not alwanys available and a routine detecting parameter.Thus,in the present study our aim was to explore the feasibility of using an alternative parameter,the changes of insulin release rate and its response to glucose during oral glucose tolerance test(OGTT),for traditional C-peptide-based model in evluationβ-cell glucose sensitivity(βCGS),and investigateβCGS across glucose tolerance category in Chinese individuals,and determine the role of impairedβCGS in the development of type 2 diabetes.Furthermore,we also aimed to investigate whetherβ-cell function especially IRRG levels have an impact on glycemic control after treated with common oral hypoglycemic agents,or influenced by the drugs.Methods:Participants were enrolled from the outpatient department at Shanghai Jiao Tong University affiliated Sixth People’s Hospital between February 2006 and November 2012.All subjects underwent complete physical examinations and routine biochemical analyses of blood.Glucose tolerance status were defined according to OGTT.1599 subjects with measurement of insulin and C-peptide levels were selected to investigated the clinical possibility of utilize insulin instead of C-peptide to evaluatedβ-cell glucose sensitivity.In the follow-up study,2321 subjects were selected to evaluateβ-cell function in Chinese population with different glucose tolerance and the role ofβ-cell glucose sensitivity in the development of T2D.Furthermore,totally 694 subjects with T2D were randomly treated by Acarbose or Metformin for 1 year and were selected to study the impact of IRRG on hypoglycemic agents.Subjects with HAb1c<7%after treatment were considered as effective glycemic control otherwise were non-effective.All participants received OGTT or the standard meal test(SMT)with measurement of glucose,insulin and/or C-peptide levels before test and at 30 and/or 60,120 and 180 minutes.β-cell function were evaluated from OGTT or SMT.EmpiricalβCGS were calculated by the standard C-peptide deconvolution approach.We computed insulin release rate(IRR)from insulin concentration during OGTT or SMT,and IRR response to glucose(IRRG).The comparison between IRRG andβCGS were performed,multiple logistic regression were conducted to analyze the predictive role of IRRG andβCGS in DI1st or T2D.Results:1)In the first part,there were totally 450 subjects with normal glucose tolerance(NGT),432 with impaired glucose tolerance(IGT),and 717 with newly diagnosed T2D.No matter in NGT,IGT and T2D,the insulin secretion pattern was similar to the C-peptide whereas C-peptide had a longer half-life than insulin.IRRG calculated from insulin decreased progressively from NGT to IGT to T2D,as well asβCGS derived from C-peptide.IRRG andβCGS were highly correlated(NGT:r=0.2139,P=0.001,IGT:r=0.3125,P<0.001,T2D:r=0.4783,P<0.001).In the logistic regression analysis,IRRG andβCGS methods were consistent in predicting disposition index of first-phase insulin secretion(DI1st)independently regardless of age,sex,blood pressure and diabetic family history in NGT and T2D.2)In the second part,there were 702 NGT subjects,630 IGT,and 989 T2D patients respectively.Pancreatic IRRG decreased progressively spanning the range from NGT to IGT to T2D(P<0.01).By design,IRRG decreased across NGT tertiles at baseline.Stumvoll first and second-phase insulin secretion declined progressively throughout IRRG tertiles(P<0.01).After adjustment for insulin resistance,despite the difference among disposition index of second-phase insulin secretion(DI2nd)were disappeared,statistical difference among DI1stst were still significant(P<0.01).At 4.74±1.73 years,32(5.41%)NGT subjects had developed T2D.In the multiple logistic regression model,IRRG was an independent predictor of T2D development[Odds ratio(OR):0.942,95%Confidence interval(CI)0.904-0.982;P<0.01](Model 1).This predictive role remained after adjustment for sex,age,blood pressure,insulin resistance(HOMA-IR),insulin sensitivity(ISIM),DI1st and diabetic family history[OR:0.941(95%CI 0.903-0.980);P<0.01](Model 4).3)In the third part,baseline IRRG levels were significantly higher in subjects with improved glycaemic control compared to poor glycaemic control after treatment by Acarbose(P<0.01),however,no significant difference was found between baseline IRRG levels according to Metformin glyccemic control effective.No differences were found in both Acarbose and Metformin group between IRRG levels before and after treatment.Logistic regression analysis showed that IRRG was an independent predictor in improved glycemic control after Acarbose treatment,whereas OGIS was an independent predictor in improved glycemic control after Metformin treatment,all regardless of age,sex,blood pressure,diabetic duration and DI1st.Conclusions:The insulin based IRRG correlated significantly withβCGS derived from the C-peptide approach in subjects with different glucose metabolism state.Moreover,IRRG was closely correlated with DI1st.Impaired IRRG levels had already existed in NGT subjects and associated with deterioration ofβ-cell function and develpemnt of T2D,suggesting that IRRG could be used as an alternative index in subjects lacking C-peptide to estimateβ-cell function.The baseline IRRG was associetd with the effect of Acarbose,but not with that Metformin,on glycemic control.No changes could be viewed in IRRG after one-year treatment of both Acarbose and Metformin. |