Glycaemic patterns in breastfeeding women with type 1 diabetes is poorly characterised. We aimed to 1) describe glycaemia and risk of hypoglycaemia in breastfeeding and artificially feeding women and 2) describe the acute changes in glycaemia induced by suckling.
METHODS: We conducted a prospective study using blinded continuous glucose monitoring 2-4 months postpartum in eight breastfeeding (BF) and eight artificially feeding (AF) women. Patient demographics, medical and pregnancy history were recorded. Women kept a diary and ate a standardised breakfast during monitoring. A third group (clinic controls, CC) of age, BMI and HbA1c matched women were retrospectively identified.
RESULTS: There was a trend towards higher daily carbohydrate intake in BF compared to AF women (p=0.090) despite similar insulin requirements. Compared to the BF group, the high blood glucose index (HBGI) and standard deviation of glucose (SD) were higher in the AF (p=0.031) and CC (p=0.022) groups. There was no difference in the low blood glucose index (LBGI), number, duration or severity of hypoglycaemic episodes between AF and BF groups. The LBGI significantly increased after suckling compared to prior (p=0.003) and during (p <0.001) suckling. The average glucose declined at 120-150mins (p=0.022) and 150-180mins (p=0.020) after the onset of suckling. Hypoglycaemia occurred within three hours of suckling in 14.49% of episodes. The significant predictors of hypoglycaemia were the time from the last oral intake (p=0.037) and time from last rapid-acting insulin (p=0.032). After a standardised breakfast, the net change in glucose was positive for AF women. In BF women, the AUC was positive if suckling did not occur within an hour of eating but negative if suckling occurred within 30 minutes of eating (p=0.025).
CONCLUSION: Breastfeeding women with type 1 diabetes had a comparable incidence of hypoglycaemia but lower glucose variability than women who artificially fed. The onset of suckling is associated with a subsequent reduction in maternal glucose but the magnitude was insufficient to cause hypoglycaemia in 85% of episodes.