Continuous glucose monitoring improves outcomes in pregnant women with type 1 diabetes (CONCEPTT)

Référence

Feig DS, Donovan LE, Corcoy R, et al, for the CONCEPTT Collaborative Group. Continuous glucose monitoring in pregnant women with type 1 diabetes (CONCEPTT): a multicentre international randomised controlled trial. Lancet 2017;390(10110):2347-2359.

Sommaire

These authors recruited women with type 1 diabetes into 2 parallel trials—one trial included the women who were pregnant, and the other included the women who were planning to become pregnant. All the women had diabetes for at least 1 year, used multiple daily injections, or used an insulin pump. The pregnant women had a singleton pregnancy of less than 14 weeks' gestation and a baseline glycohemoglobin (A1C) level of between 6.5% and 10.0% (48 - 86 mmol/mol) while the baseline A1C level for those planning to become pregnant was between 7.0% and 10.0% (53 - 86 mmol/mol). All women participated in a run-in phase of at least 6 days in which all the women wore CGM masks. At the end of the run-in period, the researchers randomized the highly adherent women to capillary testing with CGM (n = 161) or capillary testing without CGM (n = 64). All women, regardless of assignment, were to self-test at least 7 times daily and were given written instructions for insulin dose adjustments. The researchers followed up with the women multiple times throughout pregnancy, and for 24 weeks with the women planning pregnancy. If members of the latter group became pregnant, they were subsequently assessed as the pregnant women. For such a small number of women the authors assessed an enormous amount of outcomes, including A1C levels and various maternal and fetal outcomes, but made no statistical adjustments to mitigate the potential for finding random associations. One-third of the women who were planning a pregnancy became pregnant during the study period. The researchers found small but statistically significant decreases in the mean A1C levels in the pregnant women who used CGM. The nonpregnant women had no statistically significant decrease in their A1C. Additionally, the pregnant women using CGM were slightly more likely than those using capillary testing alone to have achieved an A1C level of less than 6.5% (48 mmol/mol; 68% vs 61%). Just under 20% of women in each group had severe hypoglycemic episodes. Although the study likely lacked power to detect small differences, the pregnant women in each group had comparable rates of hypertensive disorders, preeclampsia, gestational age at delivery, cesarean deliveries, and pregnancy loss. Women who used CGM had fewer infants with birth weights over the 90th percentile compared with women in the control group (53% vs 69%; number needed to treat [NNT] = 7; 95% CI 4 - 77); approximately 25% of the infants in each group were macrosomic. Although it is common practice to intensively monitor infants born to women with type 1 diabetes, in this study the infants delivered in the CGM group were less likely than infants delivered in the control group to require more than 24 hours in intensive care (27% vs 43%; NNT = 7; 4 - 36) and less likely to require intravenous dextrose (15% vs 28%; NNT = 8; 5 - 64). The infants born to women wearing a CGM mask had an average length of hospital stay that was 1 day shorter than the control patients. The study was too small to see any effect on infant mortality. Just less than half the women using CGM experienced skin rashes compared with fewer than 10% of the control patients. More than 80% of the women using CGM voiced frustrations.