The accuracy of haemoglobin A1c as a screening and diagnostic test for gestational diabetes: a systematic review and meta-analysis of test accuracy studies.

Abstract

Gestational diabetes mellitus (GDM) is associated with adverse pregnancy complications. Accurate screening and diagnosis of gestational diabetes are critical to treatment, and in a pandemic scenario like coronavirus disease 2019 needing a simple test that minimises prolonged hospital stay. We undertook a meta-analysis on the screening and diagnostic accuracy of the haemoglobin A1c (HbA1c) test in women with and without risk factors for gestational diabetes. Unlike the oral glucose tolerance test, the HbA1c test is simple, quick and more acceptable. There is a growing body of evidence on the accuracy of HbA1c as a screening and diagnostic test for GDM. We searched Medline, Embase and Cochrane Library and selected relevant studies. Accuracy data for different thresholds within the final 23 included studies (16 921 women) were pooled using a multiple thresholds model. Summary accuracy indices were estimated by selecting an optimal threshold that optimises either sensitivity or specificity according to different scenarios. HbA1c is more useful as a specific test at a cut-off of 5.7% (39 mmol/mol) with a false positive rate of 10%, but should be supplemented by a more sensitive test to detect women with GDM.