The measurement of glycated haemoglobin (HbAI) is now well established as a marker of medium term glycaemic control in diabetic patients. Furthermore, it is widely accepted that optimal glycaemic control in women prior to conception and during pregnancy reduces fetal morbidity and mortality.' For HbAI measurements to be useful during pregnancy if is necessary to establish appropriate pregnancy specific reference ranges. Whilst obtaining HbAI values from non-diabetic pregnant women for reference range purposes we observed very high results in three patients which could not be attributed to impaired glucose tolerance. We therefore investigated the cause.
METHODSBlood samples (5 mL) were collected into disodium EDTA from 25 pregnant women at sixteen, 26 and 34 weeks gestation. Ethical Committee approval was obtained for the collection of samples. All of the pregnant women had a fasting plasma glucose concentration of less than 5'5mmol/L by our laboratory glucose oxidase method. None of the women had either a family history of diabetes or previous gestational diabetes. Samples were stored at + 20°C and were analysed within 24 h of collection.HbAI was measured by agar gel electroendosmosis using the Ciba Corning glycosylated haemoglobin system." Between run imprecision (% CV) was 6·0% at an HbAI level of 13·2%. Haemoglobin F (HbF) was measured by Singer's alkali denaturation method.' The adult reference range for percentage HbF using this technique is less than 0·5%.
RESULTSWith three exceptions all the HbAI results in the non-diabetic pregnant women fell within the adult non-pregnant reference range of 5·0 to 8·0% (Table I). Furthermore, there was no significant difference in HbAI values between the three gestational ages (Mann-Whitney Utest: P