Initially untreated fasting hyperglycaemia in early pregnancy: prognosis according to occurrence of gestational diabetes mellitus after 22 weeks' gestation: a case-control study.

Aims To evaluate the percentage of women with untreated fasting hyperglycaemia in early pregnancy who develop gestational diabetes mellitus after 22 weeks' gestation, the determinants of gestational diabetes development in such women and the prognosis of early fasting hyperglycaemia according to whether the women go on to develop gestational diabetes. Methods From a large cohort of women who delivered in our hospital between 2012 and 2016, we retrospectively selected all those who had untreated early fasting hyperglycaemia and separated them into a 'gestational diabetes' and a 'no-gestational diabetes' group according to oral glucose tolerance test results after 22 weeks' gestation. We compared the incidence of a predefined composite outcome (preeclampsia or large-for-gestational-age infant or shoulder dystocia or neonatal hypoglycaemia) in both groups. Results A total of 268 women (mean fasting plasma glucose 5.3 +/- 0.3 mmol/l at a mean +/- sdof 10.2 +/- 4.2 weeks' gestation) were included. Gestational diabetes developed in 134 women and was independently associated with early fasting plasma glucose >= 5.5 mmol/l [odds ratio 3.16 (95%CI1.57, 6.33)], age >= 30 years [odds ratio 2.78 (95%CI1.46, 5.31)], preconception obesity [odds ratio 2.12 (95%CI1.11, 4.02)], family history of diabetes [odds ratio 1.87 (95%CI1.00, 3.50)] and current employment [odds ratio 0.46 (95%CI0.26, 0.83)]. Despite treatment, gestational diabetes induced a significant increase in the composite outcome as compared to no gestational diabetes (odds ratio 2.16 [95%CI1.08, 4.34]). The association disappeared after adjustment for risk factors. Conclusions Only half of the women with early fasting hyperglycaemia and no specific care subsequently developed gestational diabetes, and these women had a poor prognosis despite gestational diabetes treatment. Poor prognosis was mostly attributable to risk factors. Our results suggest that only women with certain risk factors should be screened for early fasting hyperglycaemia.