Gestational diabetes mellitus (GDM) frequently provides the first presentation of glucose intolerance in women with mutations in the glucokinase gene (GCK). Identifying such women during, or immediately after, their period of antenatal care is desirable since a molecular diagnosis of GCK deficiency provides valuable prognostic information on the expected future patterns of glucose homeostasis of both mother and fetus [1, 2]. However, molecular screening for GCK mutations is labour-intensive and currently impractical for universal use in GDM, and, as a result, preselection remains a necessary component of any molecular screening strategy.

Ellard et al. recently described a set of clinical criteria, which when applied to a large cohort of GDM women, successfully preselected 15 European GDM women yielding a high prevalence (80%) of GCK mutations [2]. We have applied similar (but, in our case, purely biochemical) criteria to our local multiethnic antenatal population, namely:(i) mild fasting hyperglycaemia (5.5±8.0 mmol/l) in pregnancy;(ii) persisting mild fasting hyperglycaemia outside pregnancy (5.5±8.0 mmol/l); and (iii) an increment between the fasting and 2-h plasma glucose concentrations< 3.5 mmol/l during the 75-g OGTT postpartum. These criteria were met by 17 women of multiethnic origin (Table 1) with a median (interquartile range) age of 37.5 (34.0±40.3) years and a median parity of 2. Singlestranded conformational polymorphism analysis followed by direct sequencing was used to screen these women for