RESUMEN
CONTEXT: Some women living with type 1 diabetes complain of changes in glucose values according to the different phases of menstruation. OBJECTIVE: To evaluate this variability through continuous glucose monitoring (CGM) data in type 1 diabetes patients. DESIGN: Observational study. SETTING: Ambulatory data, recruitment in 2 centers in the Paris region. PATIENTS: Twenty-four women with type 1 diabetes having spontaneous menstrual cycles. INTERVENTION: Collection of CGM data for 62 spontaneous menstrual cycles, with evaluation of five 3-day phases during each cycle: (1) early follicular (menstruations), (2) mid-follicular, (3) peri-ovulatory, (4) mid-luteal, and (5) late luteal. MAIN OUTCOME MEASURE: Time in range (TIR, prespecified). RESULTS: TIR decreased for each consecutive phase (61â ±â 18%; 59â ±â 18%; 59â ±â 20%; 57â ±â 18%; and 55â ±â 20%, Pâ =â 0.02). The linear mixed model highlighted a decrease in TIR in the mid-luteal (Pâ =â 0.03) and late luteal (Pâ <â 0.001) phases compared with the early follicular phase. Time above range was significantly higher during the late luteal phase than the early follicular phase (Pâ =â 0.003). Time below range was significantly higher during the mid-follicular phase than in the early follicular phase. CONCLUSION: In most of the study population, glucose levels rose linearly throughout the menstrual cycle, reaching a maximum in the late luteal phase. A sharp decrease was seen for most participants at the beginning of menstrual bleeding. This should be taken into consideration in daily care of type 1 diabetes patients to avoid hypoglycemia.