Detemir vs neutral protamine Hagedorn insulin for diabetes mellitus in pregnancy: a comparative effectiveness,...

Insulin detemir, being used increasingly during pregnancy, may have pharmacologic benefits compared with neutral protamine Hagedorn.

We evaluated the probability that compared with treatment with neutral protamine Hagedorn, treatment with insulin detemir reduces the risk for adverse neonatal outcome among individuals with type 2 or overt type 2 diabetes mellitus (gestational diabetes mellitus diagnosed at <20 weeks' gestation).

We performed a multiclinic randomized controlled trial (September 2018 to January 2020), which included women with singleton gestation with type 2 or overt type 2 diabetes mellitus who sought obstetrical care at ≤21 weeks' gestation. Participants were randomized to receive either insulin detemir or neutral protamine Hagedorn by a clinic-stratified scheme. The primary outcome was a composite of adverse neonatal outcomes, including shoulder dystocia, large for gestational age, neonatal intensive care unit admission, respiratory distress (defined as the need of at least 4 hours of respiratory support with supplemental oxygen, continuous positive airway pressure or ventilation at the first 24 hours of life), or hypoglycemia. The secondary neonatal outcomes included gestational age at delivery, small for gestational age, 5-minute Apgar score of <7, lowest glucose level, need for intravenous glucose, respiratory distress syndrome, need for mechanical ventilation or continuous positive airway pressure, neonatal jaundice requiring therapy, brachial plexus injury, and hospital length of stay. The secondary maternal outcomes included hypoglycemic events, hospital admission for glucose control, hypertensive disorder of pregnancy, maternal weight gain, cesarean delivery, and postpartum complications. We used the Bayesian statistics to estimate a sample size of 108 to have >75% probability of any reduction in the primary outcome, assuming 80% power and a hypothesized effect of 33% reduction with insulin detemir. All analyses were intent to treat under a Bayesian framework with neutral priors (a priori assumed a 50:50 likelihood of either intervention being better; National Clinical Trial identifier 03620890).

There were 108 women randomized in this trial (57 in insulin detemir and 51 in neutral protamine Hagedorn), and 103 women were available for analysis of the primary outcome (n=5 for pregnancy loss before 24 weeks' gestation). Bayesian analysis indicated an 87% posterior probability of reduced primary outcome with insulin detemir compared with neutral protamine Hagedorn (posterior adjusted relative risk, 0.88; 95% credible interval, 0.61-1.12). Bayesian analyses for secondary outcomes showed consistent findings of lower adverse maternal outcomes with the use of insulin detemir vs neutral protamine Hagedorn: for example, maternal hypoglycemic events (97% probability of benefit; posterior adjusted relative risk, 0.59; 95% credible interval, 0.29-1.08) and hypertensive disorders (88% probability of benefit; posterior adjusted relative risk, 0.81; 95% credible interval, 0.54-1.16).

In our comparative effectiveness trial involving individuals with type 2 or overt type 2 diabetes mellitus, use of insulin detemir resulted in lower rates of adverse neonatal and maternal outcomes compared with neutral protamine Hagedorn.