Star J, Canick J A, Palomaki G E, Carpenter M W, Saller D N, Sung C J, Tumber M B, Coustan D R
Brown University School of Medicine, Women and Infants Hospital, Department of Obstetrics and Gynecology, Providence, RI 02905, USA.
Prenat Diagn. 1997 Feb;17(2):149-54.
Associations between elevated amniotic fluid glucose and insulin levels in the second trimester and the subsequent development of gestational diabetes have been reported. We conducted a case-control study to determine which analyte best predicts future maternal glucose intolerance. Thirty-nine women diagnosed with gestational diabetes (criteria of Carpenter and Coustan, Am. J. Obstet. Gynecol., 144, 768, 1982) who had undergone genetic amniocentesis for advanced maternal age were matched with euglycaemic controls. Glucose and insulin concentrations were determined by analysis of stored amniotic fluid samples. No significant difference was detected between cases and controls for amniotic fluid glucose concentrations. Amniotic fluid insulin concentrations were significantly higher in cases (mean rank 4.44, P < 0.01, using matched rank analysis of variance, where 1 is the lowest and 6 is the highest rank). After conversion to multiples of the median, 20 per cent of women with subsequent gestational diabetes were found to have amniotic fluid glucose levels at or above the 90th centile, while 35 per cent of cases had similarly elevated amniotic fluid insulin levels. We conclude that second-trimester amniotic fluid insulin is a more sensitive predictor of impending glucose intolerance than amniotic fluid glucose, although neither is sufficiently powerful to use alone as a screening test.
已有报道称,孕中期羊水葡萄糖和胰岛素水平升高与随后发生的妊娠期糖尿病之间存在关联。我们进行了一项病例对照研究,以确定哪种分析物最能预测未来母亲的糖耐量异常。39名因母亲年龄较大而接受基因羊膜穿刺术并被诊断为妊娠期糖尿病的女性(采用Carpenter和Coustan的标准,《美国妇产科杂志》,144, 768, 1982)与血糖正常的对照组进行了匹配。通过分析储存的羊水样本测定葡萄糖和胰岛素浓度。病例组和对照组的羊水葡萄糖浓度未检测到显著差异。病例组的羊水胰岛素浓度显著更高(平均秩次4.44,P < 0.01,采用配对秩次方差分析,其中1为最低秩次,6为最高秩次)。转换为中位数倍数后,发现随后发生妊娠期糖尿病的女性中有20%的羊水葡萄糖水平处于或高于第90百分位数,而35%的病例有同样升高的羊水胰岛素水平。我们得出结论,孕中期羊水胰岛素比羊水葡萄糖更能敏感地预测即将发生的糖耐量异常,尽管两者都不足以单独用作筛查试验。