Impact of late pregnancy haemoglobin A at 29-30 weeks' gestation on adverse pregnancy outcomes among women with pre-existing diabetes: a retrospective analysis.

This study was to assess the relationship between late pregnancy haemoglobin A (HbA) at 29-30 weeks of gestation and adverse pregnancy outcomes (APOs) in 272 pregnant women with pre-existing diabetes. HbA ≥6.1% was associated with significantly increased risk of preterm delivery, Caesarean section, large for gestational age (LGA), neonatal respiratory distress, neonatal hypoglycaemia, and composite adverse neonatal outcome (p < .05). The risk of pre-eclampsia increased significantly at the lower HbA cut-off of > 5.6% (p = .039). Reduction of HbA cut-off from 6.1% to 5.6% improved the sensitivity but reduced the specificity for prediction of APOs. Overall, the receiver operating characteristic (ROC) curves demonstrated the moderate predictive value of late pregnancy HbA for APOs. In conclusion, elevated late pregnancy HbA levels at 29-30 gestational weeks had a negative impact on APOs in pregnant women with pre-existing diabetes. However, HbA cut-off levels of neither ≥6.1% nor >5.6% were ideal for predicting APOs. Impact statement What is already known on this subject: Poorly controlled diabetes is associated with adverse pregnancy outcomes (APOs). Periconceptual haemoglobin A (HbA) correlates well with the risk of foetal anomaly but is not predictive of APOs at time of delivery. New evidence suggested that late pregnancy HbA is predictive of APOs but the definitions of a late pregnancy gestational week and target HbA cutpoint remain in doubt. What the results of this study add: This study investigated the relationship between late pregnancy HbA levels at 29-30 weeks of gestation and the APOs among pregnant women with pre-existing diabetes. Late pregnancy HbA ≥ 6.1% correlated with the risk of APOs but the increased risk of pre-eclampsia only became significant at the lower cut-off of >5.6%. Reducing HbA cut-off from 6.1% to 5.6% improved the sensitivity but reduced the specificity for prediction of APOs. Overall, late pregnancy HbA had a moderate predictive value for APOs. What the implications are of these findings for clinical practice and/or further research: HbA cut-off levels of neither ≥6.1% nor >5.6% were ideal in predicting APOs among pregnant women with pre-existing diabetes. As HbA levels tend to drop in pregnancy, caution should be taken when interpreting HbA in pregnancy. More multi-centred studies are required to explore the respective glycaemic target for each APO and to determine the ideal timing for late pregnancy HbA measurement.