Objectives: To validate a diagnostic test for gestational diabetes which predicts the risk of macrosomia.
Methods: A prospective study was carried out among 354 women at risk for gestational diabetes to compare two glucose tests diagnosing pregnancies at risk of macrosomia. The "practical" test consisted in glucose measurement in the fasting state and two hours after an usual breakfast and the "reference" test was the test proposed in France (O'Sullivan test with or without a 100 g oral glucose tolerance test). Both tests were made between the 24(th) and 28(th) week of gestation. Women at high risk for macrosomia were treated. The first assessment criterion was macrosomia (babies large for gestational age). Because of the presence of women treated for gestational diabetes in our sample, the sensitivity and specificity of the tests in diagnosing pregnancies at risk of macrosomia were calculated using either the incidence of macrosomia observed in our population, or the incidence of macrosomia observed theoretically in the absence of treatment (22% in literature).
Results: Macrosomia was diagnosed in 49 neonates (14%). The "practical" test was significantly more sensitive than the reference test (respectively 46.9% versus 16.3%, p=0.0001 in the first case, and 54.3% versus 20.1%, p=0.0001 in the second case). The "reference" test was significantly more specific than the "practical" test (respectively 80% versus 68.2%, p=0.0001 in the first case, and 80.6% versus 70%, p=0.0001 in the second case).
Conclusion: Our study shows that the simplified "practical" test is more sensitive than the "reference" test currently used in France in screening women at risk of macrosomia.