Most of the studies of antenatal steroids for lung maturation included mothers at less than 34 weeks. Most babies at 34 to 36 weeks do fine, although there is a growing realization of their increased long term risks. Even the proportion who develop respiratory disease usually have relatively short-lived disease, and rarely get very sick, although that can happen. Therefore, to show the efficacy and safety of steroids at later preterm gestational ages you would need to perform a very large study.
So here it is…
Gyamfi-Bannerman C, et al. Antenatal Betamethasone for Women at Risk for Late Preterm Delivery. The New England journal of medicine. 2016; 2800 mothers at risk for preterm delivery at 34 weeks 0 days up to 36 weeks and 5 days were randomized to IM betamethasone or IM placebo. Mothers were not eligible if they were expected to deliver in less than 24 hours, in order to try and get enough time for the steroids to have optimal effect. In fact the median time from starting the intervention until they did indeed deliver was about 30 hours, with interquartile range from 14 to 110 hours.
The primary outcome was a composite made up of signs of respiratory failure, including needing CPAP (or high flow cannulae) for more than 2 hours, or needing more than 30% oxygen for over 4 hours, or being ventilated. That outcome was decreased from 14.4% (placebo) to 11.6% (betamethasone). Among other secondary outcomes, there was also a reduction in the “need for resuscitation at birth” which was defined as : any intervention in the first 30 minutes other than blow-by oxygen, This dropped from 18.7% to 4.5%
Which all sounds great.
The incidence of hypoglycemia (blood sugar under 2.2 mmol/L, that’s 40 mg/dl for the Americans) was greatly increased, from 15% to 24% in the steroid group. There were no other safety issues identified, maternal adverse effects were similar in the 2 groups, and there was no increase in infection (maternal or neonatal). I can’t tell from the data presented how many of the babies needed invasive treatment for their hypoglycemia, or how many had more severe hypoglycemia.
If that is a real effect of the antenatal steroids, I am not sure what the mechanism would be, I wouldn’t have thought that 30 hours of antenatal steroids would be enough to induce hyperinsulinism in the babies, but I do think this is a serious potential hazard of the steroids in this group. Almost all of the control babies recovered from their respiratory distress without a significant complication, so if the hypoglycemia had long-term consequences in some babies then the balance of benefits and adverse effects might not be positive.
For now this seems hopeful that antenatal betamethasone at 34 to 36 weeks could lead to an absolute risk reduction of 3% in the development of respiratory failure, a 0.7% decrease in the use of assisted ventilation. But unfortunately a 9% increase in hypoglycemia, most of which was presumably mild and of short duration. I am not sure that this should become routine therapy unless we can get some long term follow up. I for one hope the MFM units network of the NICHD will fund such a follow up to this study, at least to 3 years of age.