Endocrine Abstracts

Severe inpatient hyperglycaemia is associated with poor patient outcomes. However, its causes, and effects in the general inpatient population remain understudied. We aimed to investigate the causes and consequences of severe hyperglycaemia (blood glucose concentration [BGC] ≥ 20 mmol/l/l) in adult hospitalised patients. We conducted a retrospective observational study comparing 771 admissions with BGC ≥ 20 mmol/l/lto 2,334 admissions with BGC < 20 mmol/l/lbetween July 2024 and May 2025 at Tallaght University Hospital. We analysed patients’ diabetes history, primary causes, and HbA1c and ketone levels. The outcomes included length of stay (LOS), mortality, and intensive care unit (ICU) admission. LOS was significantly longer in severely hyperglycaemic patients (mean +3.4 days, P = 0.022), with findings persisting after adjustment using ordinary least squares regression (β = 0.26, P < 0.001). Mortality did not differ significantly after adjustment. Severe hyperglycaemia was not significantly associated with ICU admission (P = 0.073). Ketonaemia was associated with ICU admission (P = 0.0002). The distribution of causes differed significantly by diabetes type (P < 0.001): diabetic ketoacidosis predominated in type 1 diabetes mellitus (T1DM), infection and steroid use were more common in T2DM and non-diabetic patients, and pancreatic malignancy predominated in T3cDM. HbA1c differed by diabetes status (P = 0.039) but not mortality. Severe hyperglycaemia is associated with markedly prolonged hospitalisation. While high HbA1c and ketone levels can help predict the relative risk of requiring a higher level of care, they are not good predictors of mortality.