Objective To observe day-to-day variability in glucose levels in pregnant women with Type 1 diabetes using the Continuous Glucose Monitoring System (CGMS) and to assess the usefulness of continuous glucose measurements for adjustment of insulin treatment.
Design A prospective observational study.
Setting The obstetrical outpatient clinic of the University Medical Centre Utrecht.
Population Pregnant women with Type 1 diabetes mellitus.
Methods Thirty-one pregnant women with Type 1 diabetes used the CGMS for two consecutive days. Patients were classified in two groups (high vs low day-to-day variability) based on visual inspection of the glucose excursions. The relationship between day-to-day variability and the variables HbA1c, maternal age and body mass index (BMI), duration of diabetes, number of self-monitored blood glucose levels, number of insulin injections, gestational age, nutrition, physical activity, White classification, living with children and method of insulin administration was determined. The two days of the first 20 CGMS measurements were separated and four physicians were asked to give recommendations on treatment adjustment for each separate day.
Main outcome measures Mean absolute difference (MAD) was calculated for each patient as measure of day-to-day variability.
Results Seventeen patients (55%) were classified as having low (MAD 0.92–2.33 mmol/L) and 14 (45%) as having high day-to-day variability (MAD 2.41–6.12 mmol/L). Of the variables measured, only the relation between MAD and HbA1c was significant (r= 0.58, P= 0.001). The difference in recommendation on treatment adjustment between the days of the CGMS measurement ranged from 29% to 48%. This percentage was significantly higher in the high day-to-day variability group (48 vs 33%, P= 0.01).
Conclusion Day-to-day glucose variability is high and the treatment of pregnant women with Type 1 diabetes is a problem. Fine-tuning of insulin regimens based on two-day measurements with the CGMS is not advisable.