Twice daily versus four times daily insulin dose regimens for diabetes in pregnancy: randomised controlled trial.

OBJECTIVE To compare perinatal outcome and glycaemic control in two groups of pregnant diabetic patients receiving two insulin regimens. DESIGN Randomised controlled open label study. SETTING University affiliated hospital, Israel. PARTICIPANTS 138 patients with gestational diabetes mellitus and 58 patients with pregestational diabetes mellitus received insulin four times daily, and 136 patients with gestational diabetes and 60 patients with pregestational diabetes received insulin twice daily. INTERVENTION Three doses of regular insulin before meals and an intermediate insulin dose before bedtime (four times daily regimen), and a combination of regular and intermediate insulin in the morning and evening (twice daily regimen). MAIN OUTCOME MEASURES Maternal glycaemic control and perinatal outcome. RESULTS Mean daily insulin concentration before birth was higher in the women receiving insulin four times daily compared with twice daily: by 22 units (95% confidence interval 12 to 32) in patients with gestational diabetes and by 28 units (15 to 41) in patients with pregestational diabetes. Glycaemic control was better with the four times daily regimen than with the twice daily regimen: in patients with gestational diabetes mean blood glucose concentrations decreased by 0.19 mmol/l (0.13 to 0.25), HbA(1c) by 0.3% (0.2% to 0.4%), and fructosamine by 41 micromol/l (37 to 45), and adequate glycaemic control (mean blood glucose concentration <5.8 mmol/l) was achieved in 17% (8% to 26%) more women; in patients with pregestational diabetes mean blood glucose concentration decreased by 0.44 mmol/l (0.28 to 0.60), HbA(1c) by 0.5% (0.2% to 0.8%), and fructosamine by 51 micromol/l (45 to 57), and adequate glycaemic control was achieved in 31% (15% to 47%) more women. Maternal severe hypoglycaemic events, caesarean section, preterm birth, macrosomia, and low Apgar scores were similar in both dose groups. In women with gestational diabetes the four times daily regimen resulted in a lower rate of overall neonatal morbidity than the twice daily regimen (relative risk 0.59, 0.38 to 0.92), and the relative risk for hyperbilirubinaemia and hypoglycaemia was lower (0.51, 0.29 to 0.91 and 0.12, 0.02 to 0.97 respectively). The relative risk of hypoglycaemia in newborn infants to mothers with pregestational diabetes was 0.17 (0.04 to 0.74). CONCLUSIONS Giving insulin four times rather than twice daily in pregnancy improved glycaemic control and perinatal outcome without further risking the mother.