Translating HbA1c measurements into estimated average glucose values in pregnant women with diabetes

• DOI: 10.1007/s00125-017-4205-7
• Publication Date: 2017-01-19T19:15:32Z
• Authors: Graham R. Law; Mark S. Gilthorpe; Anna L. Secher; Rosemary Temple; Rudolf Bilous; Elisabeth R. Math...; Helen R. Murphy; Eleanor M. Scott
• Keywords: Coefficient of variation

This study aimed to examine the relationship between average glucose levels, assessed by continuous monitoring (CGM), and HbA1c levels in pregnant women with diabetes determine whether calculations of standard estimated (eAG) from measurements are applicable diabetes. CGM data 117 (89 type 1 diabetes; 28 2 diabetes) were analysed. Average calculated 5–7 day profiles (mean 1275 values per profile) paired a corresponding (±1 week) measure. In total, 688 glucose–HbA1c pairs obtained across pregnancy six participant). level was used as dependent variable regression model. Covariates gestational week, centre HbA1c. There strong association (coefficient 0.67 [95% CI 0.57, 0.78]), i.e. 1% (11 mmol/mol) difference corresponded mmol/l glucose. The random effects model that included week curvilinear (quadratic) covariate fitted best, allowing calculation pregnancy-specific eAG (PeAG). showed an 8.0% (64 gave PeAG 7.4–7.7 (depending on week), compared 10.2 mmol/l. associated maintaining 6.0% (42 during 6.4 6.7 mmol/l, depending week. HbA1c–average is altered pregnancy. Routinely generated do not account for this non-pregnant individuals and, thus, should be Instead, deduced current recommended antenatal clinical care.