Outcomes for general medical inpatients with diabetes mellitus and new hyperglycaemia

Objectives: To investigate the relationship between admission glycaemic status and inpatient mortality in patients with and without pre-existing diabetes. Design: Prospective observational cohort study. Setting: A general medical ward in an Australian tertiary referral hospital. Participants: 903 patients admitted to the general medical ward between February 2003 and July 2004. Main outcome measure: Inpatient death. Results: The overall inpatient mortality was 5.4% (n = 49). In the total cohort, age > 75 years and admission fasting plasma glucose (FPG) levels >= 5.6 mmol/L were independent predictors of mortality. For patients without a known history of diabetes, each 1 mmol/L rise in admission FPG was associated with a 33% increase in mortality. In these patients, elevated (> 6.0%) and normal glycated haemoglobin (HbA(1c)) levels were associated with mortalities of 11.3% and 4.4%, respectively (odds ratio, 2.47; 95% Cl, 1. 16-5.26). In contrast, in patients with known diabetes, there was no association between admission FPG levels, HbA(1c), and mortality. Length of stay was not independently associated with FPG, HbA(1c), or diabetes status. Conclusions: In patients without known diabetes, the risk of death was increased for admission FPG levels >= 5.6 mmol/L. However, pre-existing abnormal glucose metabolism, reflected by elevated HbA(1c) levels, appeared a more important predictor of inpatient mortality than glucose levels in patients without known diabetes.