Background: Gestational diabetes mellitus (GDM) is associated with adverse perinatal outcomes, if not treated. International guidelines recommend screening "all or high-risk women" at the initial prenatal visit, when a fasting plasma glucose (FPG) between 92 and 126 mg/dL is diagnostic for GDM. However, glucose testing may be affected by a great pre-analytical variability (usually overlooked), due to, for example, kind of sample (serum/plasma), temperature of storage, time between blood draw and centrifugation (in-tube glycolysis), and use of a glycolysis inhibitor. So GDM may be easily missed. We aimed to evaluate the potential characteristics of this important issue.
Subjects and methods: FPG was tested by both "routine" and "gold standard" protocols in 60 women at the first trimester of gestation, presenting for GDM screening. "Routine" blood plasma was collected in a tube with sodium fluoride, kept at room temperature, centrifuged, and tested 30-45 min after blood draw. "Gold standard" was a specimen from the same blood sample that was centrifuged within 5 min and tested together with the "routine" specimen.
Results: In the "routine" protocol, 10 mg/dL on average was lost for each determination. Thirteen cases of GDM and two of overt diabetes (FPG >126 mg/dL) were missed in this preliminary series.
Conclusions: The risk for GDM underdiagnosis in the first half of pregnancy appears to be actual and wide. A closer collaboration between clinicians and pathologists is critical, allowing a stricter adherence to the laboratory guidelines to be ensured.