1312-P: Admission Plasma Glucose and HbA_1c in Emergency Hospital Admissions by Ethnicity

Introduction: As HbA_1c is now widely used for the diagnosis of diabetes with WHO cut-off ≥ 48 mmol/mol/6.5%, it is debatable whether the value should vary by ethnicity.

Aim: This study compares plasma glucose and HbA_1c in white European WE, South Asian SA, unknown/other U and Afro-Caribbean AC people admitted as an emergency.

Methods: Electronic patient records were interrogated over 1 year (2014/5) for a clinical audit, CARMS-12031R. Glucose was measured on wards with meters/gas machines or the laboratory along with HbA_1c on Tosoh G8 HPLC analysers if haemoglobin normal.

Results: There were 30,664 people of whom 22,045 (72%) had glucose recorded and 2,083 (7%) HbA_1c. 1,904 (6%) had both glycaemic markers and were aged 64 (50-78) years, median (IQ range); 839 (44%) female; 560 (29%) prior diabetes diagnosis; 1,405 (74%) WE, 296 (16%) SA, 112 (6%) U and 91 (5%) AC. Glucose was 7.4 (5.8-10.6), 8.6 (6.2-12.3), 8.1 (5.9-10.7), 7.0 (5.9-11.6) mmol/L and HbA_1c 42 (37-55), 50 (41-70), 42 (38-60), 46 (39-70) mmol/mol/6.0 (5.5-7.2), 6.7 (5.9-8.6), 6.0 (5.6-7.6), 6.4 (5.7-8.6)%, respectively. HbA_1c was higher relative to glucose in SA and AC compared to WE and U, p<0.001.

Conclusions: Haematological differences related to red blood cell turnover may account for the variance; this should be recognised by those involved in producing protocols/guidelines for diagnosis. More research is required on using HbA_1c to confirm undiagnosed diabetes in admissions.

View largeDownload slide

View largeDownload slide

Close modal