Abstract

Background: The use of a glucose challenge test as the universal screening for gestational diabetes is common in many countries. This test represents significant costs for laboratories and inconveniences for the patients, who have to wait for one hour and, very often, feel discomfort and nausea. In this work we propose the use of fasting glycemia, in a population with low prevalence of gestational diabetes as a pre-screening test that would avoid the oral glucose overload in those pregnant women with low risk of gestational diabetes.
Methods: The study was done with the fasting glucose levels of 6,573 pregnant women who underwent a two steps strategy to screen for gestational diabetes, a first step consisting of a 50 g glucose challenge test, followed when glycemia ≥ 140 mg/dL by a 100 g Oral Glucose Tolerance Test, based on recommendations made by National Diabetes Data Group.
Results: The ROC curve for fasting glucose was calculated, and we obtained an AUC = 0.633 (0.569 - 0.696). The sensitivity, specificity, and predictive values were established for different thresholds.
Conclusions: We proposed that women with fasting glycemia ≤ 62 mg/dL, (S = 91.3%, NPV = 98.79% and LR- = 0.87) are in low risk of suffering gestational diabetes, which means that 10% of our population would not undergo the glucose challenge test.

DOI: 10.7754/Clin.Lab.2017.170920