糖尿病流行病学研究中应用OGTT资料评估胰岛β细胞功能的可能性——468例非糖尿病Pima印第安人葡萄糖钳研究资料分析

Possibility of evaluating islet β-cell function b y oral glucose tolerance test in epidemiological study on diabetes mellitus——a nalysis of euglycemic clamp study in 468 non-diabetic Pima Indians

目的 探讨以口服葡萄糖耐量试验 (OGTT)的简单参数在糖尿病流行病研究中评估胰岛 β细胞功能的可能性。方法 美国Pima印第安人糖耐量正常者 332例、糖耐量低减者 1 36例参与本试验。各例均做静脉葡萄糖耐量试验 (IVGTT) ,OGTT及正葡萄糖钳试验。计算Homa IR ,Homa β ,第一时相胰岛素分泌量 (AIR) ,胰岛素曲线下面积 ,ΔI30 /ΔG30 ,胰岛素介导的葡萄糖代谢率 (M ,mg·kg- 1 ·min- 1 )。从基本病理生理学概念推导新 β细胞功能指数MBCI=(FINS×FPG) / (PG2h +PG1h- 2×FPG) (FINS :空腹胰岛素 ;FPG :空腹血糖 ;PG1h和PG2h分别为OGTT 1h和 2h的血糖 )。以线性回归分析正葡萄糖钳技术测定的胰岛素敏感性与不同的β细胞功能指数的组合对血糖水平的贡献 ,探讨以OGTT的简单参数评估β细胞功能的可能性。 结果 (1 )调整M后 ,IGT组AIR、ΔI30 /ΔG30 、Homa β和MBCI与OGTT 2h血糖水平的偏相关系数分别为 - 0 .30、- 0 .30、- 0 .2 9及 - 0 .37(P均 <0 .0 0 1 ) ,但在NGT组则分别为 - 0 .0 6(P >0 .0 5)、0 .0 1 (P >0 .0 5)、- 0 .30 (P <0 .0 0 1 )及 - 0 .43(P <0 .0 0 1 )。 (2 )以OGTT 2h血糖为因变量 ,分别以M +AIR ,M +ΔI30 /ΔG30 ,M +Homa β,M +MBCI为自变量做线性回归分析 ,显示M

Objective To in tr oduce a modified β-cell function index MBCI, 〔MBCI=(FPG×FINS)/(PG2h+PG1h-2× FPG)〕, based on oral glucose tolerance test (OGTT) and to investigate if contri bution of this new index to the changes of plasma glucose level will surpass tha t of some other commonly used indices. Methods T hree hundred and thirty-two subjects with normal glucose tolerance (NGT) and 13 6 subjects with impaired glucose tolerance (IGT) in Pima Indians were included. Intravenous glucose tolerance test (IVGTT), OGTT and euglycemic clamp test were completed and Homa-IR, Homa-β, acute insulin response (AIR), ΔI 30 /ΔG 30 and insulin area under curve were calculated. Insulin-mediated glucose disposal rate (M, mg·kg -1 ·min -1 ) was evaluated by euglycemic clam p technique. Multivariable regression analysises were performed with SAS softwar e to find the relationship among post-load glucose levels, MBCI and other indic es used in literature, such as Homa-β, AIR, and ΔI 30 /ΔG 30 . Contr ibutions of combinations of insulin resistance (IR) and β-cell function indice s to increment of glucose levels after glucose load were evaluated. Results (1) Among AIR, ΔI 30 /ΔG 30 , Homa-β an d MBCI, MBCI was more significantly correlated with PG2h in IGT group after the adjustment of M (r=-0.30, -0.30, -0.29 and -0.37 respectivel y, all P<0.001). In NGT group, only Homa-β and MBCI were significantly correlated with PG2h (r=-0.30, P<0.001 and r=-0.43, P<0.001 res pectively). (2) After controlling for the influence of M, β-cell function eval uated by AIR, ΔI 30 /ΔG 30 , ΔI 60 /ΔG 60 and MBCI in IGT g roup were lower than those of the NGT group (70%, 75%, 65% and 51% vs 100%, P<0.001), however Homa-β still failed to make this difference (89% vs 100%, P>0 .05). General linear regression showed M+AIR, M+ΔI 30 /ΔG 30 , M+Homa -β and M+MBCI could explain 21%, 18%, 21% and 36% change of PG2h. By replacing M with Homa-IR, among these pairs of IR and β-cell function indices, MBCI st ill could explain most of the changes of PG2h (21%, 20%, 27% vs 55%). Conclusion (1) It is possible to eval uate β-cell function based on simple data of OGTT. MBCI is a better index than AIR, Homa-β or ΔI 30 /ΔG 30 . (2)In NGT subjects, evaluating β-ce ll function by AIR or ΔI 30 /ΔG 30 should be taken with great caution because these indices are not significantly correlated with PG2h.