Prediabetes in pregnancy, can early intervention improve outcomes? A feasibility study for a parallel randomised clinical trial

Objective Measurement of glycated haemoglobin (HbA1c) in early pregnancy is routine in New Zealand to identify women with diabetes and prediabetes. However, the benefit of early intervention in women with prediabetes is inconclusive. Our aim was to test the feasibility of a two arm parallel randomised controlled trial of standard care versus early intervention in pregnancies complicated by prediabetes. Setting Two tertiary referral centres in New Zealand. Participants Women <14 weeks gestation and HbAlc >= 5.9%-6.4% (41-46 mmol/mol) measured at booking, without pre-existing diabetes. Interventions Randomisation was done by remote web based allocation into one of two groups. Women in the early intervention group attended an antenatal diabetes clinic, commenced daily home blood glucose monitoring, and medication was prescribed if lifestyle measures failed to maintain target blood glucose levels. Controls received lifestyle education, continued standard care with their midwife and/or obstetrician, and were asked to perform a 75 g oral glucose tolerance test at 24 weeks' gestation with a referral to clinic if this test was positive. Both groups received lifestyle questionnaires at recruitment and in late pregnancy. Outcome measures Recruitment rate, adherence to protocol and validation of potential primary outcomes. Results Recruitment rates were lower than expected, especially in Maori and Pacific women. Non-adherence to allocated treatment protocol was significant, 42% (95% Cl 24% to 61%) in the early intervention group arid 30% (95% CI 16% to 51 %) in controls. Caesarean section arid pre-eclampsia were signalled as potential primary outcomes, due to both the high observed incidence in the control group and ease of measurement. Conclusions For a future definitive trial, extending the gestation of eligibility and stepped-wedge cluster randomisation may overcome the identified feasibility issues. Consistent with published observational data, pre-eclampsia and emergency caesarean section could be included as primary outcome measures, both of which have a significant impact on maternal arid neonatal morbidity arid ltealtltcare costs.