Evaluating hyperglycaemia in the hospitalised patient: towards an improved system for classification and treatment

Background Hyperglycaemia is common amongst hospitalised patients. Some of this is due to pre-existing diabetes (either previously diagnosed or not), but a proportion is due to stress hyperglycaemia, a transient state of high blood sugars related to the underlying illness. Aims We aimed to estimate the prevalence of hyperglycaemia in an Irish hospital setting, including an assessment of what contribution is made by cases of stress hyperglycaemia. Methods Over a 9-day period, all bedside glucometer-measured point of care blood glucoses performed in medical and surgical wards in University Hospital Galway were examined. Medical case notes and our diabetes computerised database were analysed to identify individuals with pre-existing diabetes. Glycosylated haemoglobin (HbA 1c ) was measured on the remainder of patients, to differentiate between pre-existing diabetes (HbA 1c ≥ 48 mmol/mol) and stress hyperglycaemia (HbA 1c < 48 mmol/mol). Results A total of 1,637 glucose readings were performed on 262 in-patients, of which 164 (63 %) were in the hyperglycaemic range. Of the 126 eligible for study inclusion, 92 (73 %) had pre-existing diabetes and 11 (9 %) had previously undiagnosed diabetes. The remaining 20 patients (16 %) had stress hyperglycaemia. Conclusions We report a high prevalence of hyperglycaemia (including stress hyperglycaemia) in an in-patient cohort in whom testing was undertaken at the discretion of the treating physician. Our data illustrate the utility of HbA 1c measurement in this setting to help differentiate between pre-existing diabetes and stress hyperglycaemia. Much work remains to be done on how to best identify and treat in-patient hyperglycaemia.