Universal two-step screening strategy for gestational diabetes has weak relevance in French Mediterranean women: Should we simplify the screening strategy for gestational diabetes in France?

• Published in: Diabetes & metabolism 2011-11, Vol.37 (5), p.419-425
• Main Authors: Chevalier, N, Fénichel, P, Giaume, V, Loizeau, S, Bongain, A, Daideri, G, Brucker-Davis, F, Hiéronimus, S
• Format: Article
• Language: English
• Subjects: Adult; Biological and medical sciences; Body Mass Index; Diabetes, Gestational - diagnosis; Diabetes, Gestational - epidemiology; Diabetes. Impaired glucose tolerance; Diabète gestationnel; Dépistage; Endocrine pancreas. Apud cells (diseases); Endocrinology & Metabolism; Endocrinopathies; Facteur de risque; Female; France - epidemiology; Gestational diabetes; Glucose Intolerance - diagnosis; Glucose Intolerance - epidemiology; Glucose Tolerance Test - methods; Glucose Tolerance Test - statistics & numerical data; HGPO; Humans; Internal Medicine; Mass Screening - methods; Mass Screening - statistics & numerical data; Medical sciences; Mediterranean Region - epidemiology; O'Sullivan test; Obesity - diagnosis; Obesity - epidemiology; OGTT; Pregnancy; Prevalence; Prospective Studies; Risk factor; Risk Factors; Screening strategy; Sensitivity and Specificity; Test de O'Sullivan; Two-step strategy
• ISSN: 1262-3636; 1878-1780
• DOI: 10.1016/j.diabet.2011.01.004

Abstract Aim Currently, there is no international consensus for gestational diabetes mellitus (GDM) diagnosis. This is a report of our experience of GDM screening according to the 1996 French guidelines. Methods For 5 years, all pregnant women followed at our hospital ( n = 11,545) were prospectively screened for GDM between weeks 24 and 28 of pregnancy with a two-step strategy: the O'Sullivan test (OS) with a threshold at 130 mg/dL, followed by a 100-g OGTT if positive. GDM was diagnosed according to Carpenter and Coustan criteria. Results Prevalence of GDM was 4.26% [344/1451 of patients with an OS of 130–199 mg/dL (12.1%); and 148 patients with an OS greater than 200 mg/dL]. The false-positive rate for the OS was 76.8%. Compared with 140 mg/dL, a threshold of 130 mg/dL caused 401 additional negative OGTTs in 90% of cases. In 80.7% GDM patients, fasting glucose was less than 95 mg/dL. The time lag between OS and OGTT was 3 weeks (1–84 days). Risk factors associated with GDM were maternal age, preconception overweight and obesity, parity, personal history of GDM or macrosomia, and familial history of obesity ( P < 0.05), but not diabetes. Also, 20% of GDM patients had no risk factors, whereas they were present in 75% of patients without GDM. Conclusion In our population, a two-step screening strategy for GDM was neither relevant nor efficient. It could be simplified with a single-step definitive screening strategy using a 75-g OGTT, as used in the HAPO study, and as recommended by the IADPSG and the recent French Expert Consensus. At present, there are still no evidence-based arguments to help in deciding between selective or universal screening for GDM.