Tatsumi Moriya, Madoka Matsubara and Masafumi Koga
Background: Pregnant women are frequently complicated with iron deficiency. Because Hemoglobin A1c (HbA1c) shows an apparently high value in patients with iron-deficiency, HbA1c was reported to be elevated due to iron deficiency in pregnant women with or without diabetes. In the present study, we investigated whether HbA1c or Glycated Albumin (GA) accurately reflects glycemic control state estimated from the Mean Blood Glucose (MBG) obtained by Continuous Glucose Monitoring (CGM) in pregnant women with diabetes. Methods: We studied 20 pregnant women with diabetes whose glycemic control was stable and CGM was performed to determine MBG. The estimated HbA1c and GA were calculated from the MBG, and the measured values were compared with the estimated values. Correlations between Mean Corpuscular Hemoglobin (MCH), an index of iron-deficiency, and continuous variables were examined. Results: HbA1c was not significantly correlated with GA. The measured GA was not significantly different from the estimated GA (16.0 ± 2.7% vs. 16.0 ± 1.8%, P=0.982). On the other hand, the measured HbA1c was significantly higher than the estimated HbA1c [6.4 ± 0.9% (46.3 ± 10.3 mmol/mol) vs. 5.6 ± 0.5% (38.2 ± 5.4 mmol/mol), P<0.001]. The measured HbA1c (R= –0.598, P=0.005) and the measured HbA1c/estimated HbA1c ratio (R= –0.566, P=0.009) showed significant inverse correlations with MCH. Conclusions: HbA1c overestimates glycemic control due to iron deficiency in pregnant women with diabetes, whereas GA accurately reflected their glycemic control. Therefore, GA is a better index of glycemic control than HbA1c in pregnant women with diabetes.