ABSTRACT
AIM: Poor glucose control is associated with adverse outcomes in pregnancies with pre-existing diabetes. However, strict glucose control increases the risk of severe hypoglycaemia, particularly in the first trimester. Therefore, we aimed to investigate whether less tight glucose control in the first trimester determines adverse outcomes or can be compensated for by good control in late pregnancy. METHODS: Retrospective data were collected from 517 singleton pregnancies complicated by pre-existing diabetes delivering between 2010 and 2017. Three hundred and thirty-six pregnancies fulfilled the inclusion criteria of having available HbA1c values either pre-conception or in the first trimester (65% type 1 diabetes, 35% type 2 diabetes). RESULTS: Higher HbA1c values in the first trimester were associated with increasing rates of large for gestational age (LGA) neonates, preterm delivery or neonatal intensive care unit admissions. Multiple regression analysis demonstrated third trimester HbA1c , type 1 diabetes, multiparity and excess weight gain, but not first trimester HbA1c , to be independently predictive for LGA. Pre-eclampsia and third trimester HbA1c increased the risk for preterm delivery. If HbA1c was ≤ 42 mmol/mol (6.0%) in the third trimester, rates of adverse outcomes were not significantly higher even if HbA1c targets of ≤ 48 mmol/mol (6.5%) had not been met in the first trimester. Good first trimester glucose control did not modify the rates of adverse outcomes if HbA1c was > 42 mmol/mol (6.0%) in the third trimester. CONCLUSIONS: Less tight glycaemic control, for example due to high frequency of severe hypoglycaemia in the first trimester, does not lead to increased adverse neonatal events if followed by tight control in the third trimester. Besides glycaemic control, excess weight gain is a modifiable predictor of adverse outcome.
Subject(s)
Diabetes Mellitus, Type 1/drug therapy , Diabetes Mellitus, Type 2/drug therapy , Glycemic Control/methods , Hypoglycemia/chemically induced , Hypoglycemic Agents/therapeutic use , Pregnancy in Diabetics/drug therapy , Adult , Cohort Studies , Diabetes Mellitus, Type 1/metabolism , Diabetes Mellitus, Type 2/metabolism , Embryonic Development , Female , Fetal Macrosomia/epidemiology , Gestational Weight Gain , Glycated Hemoglobin/metabolism , Humans , Intensive Care Units, Neonatal/statistics & numerical data , Parity , Pre-Eclampsia/epidemiology , Pregnancy , Pregnancy Trimester, First/metabolism , Pregnancy Trimester, Third/metabolism , Pregnancy in Diabetics/metabolism , Premature Birth/epidemiology , Retrospective StudiesABSTRACT
AIMS: To compare glycaemic control, maternal and neonatal outcomes in pregnancies with Type 1 diabetes, managed either by continuous subcutaneous insulin infusion, multiple daily insulin injection or switch from multiple daily insulin injection (MDI) to continuous subcutaneous insulin infusion (CSII) in early pregnancy. RESEARCH DESIGN AND METHODS: Data from 339 singleton pregnancies were retrospectively reviewed. HbA1c values were measured preconception and in each trimester. In a secondary analysis, use of CSII pre-pregnancy was compared with initiation of CSII during pregnancy. RESULTS: MDI was used in 140 pregnancies (41.3%) and CSII was used in 199 (58.7%), including 34 pregnancies (10.0%) during which the women switched to CSII. In pregnancies during which CSII was used duration of diabetes [median (interquartile range) 16.0 (8.0-23.0) years vs 11.0 (5.5-17.5) years; P<0.001] was longer, and the Institute of Medicine recommendations for appropriate weight gain were exceeded more often (64.8% vs. 50.8%; P=0.01). CSII use and pre-pregnancy BMI were independent predictors of excess weight gain. There was no difference in glucose control, but CSII was associated with higher birth weight [median (interquartile range) 3720 (3365-4100) g vs 3360 (3365-4100) g; P<0.001] and higher large-for-gestational-age (LGA) rate (44.7% vs. 33.6%; P=0.04) than MDI. HbA1c concentration in the third trimester and excess weight gain were predictive of LGA infants [odds ratio 2.33 (95% CI 1.54-3.51); P<0.001 and 1.89 (95% CI 1.02-3.51); P=0.04]. In pregnancies where CSII therapy was initiated in the first trimester and in those with pre-pregnancy use, similar glucose control and outcome was achieved. CONCLUSIONS: There was no advantage of CSII with respect to glycaemic control and neonatal outcomes. The rate of LGA neonates was higher in the CSII group, possibly mediated by excess maternal weight gain, which was more frequent than in women treated with MDI.