Fasting hyperglycaemia and in-hospital mortality in elderly population.

BACKGROUND: Admission hyperglycaemia has shown to be a marker of poor clinical outcome. The prevalence of admission hyperglycaemia and its relationship with in-hospital mortality in elderly population has not been clearly defined. We assessed the prevalence and prognostic significance of admission fasting hyperglycaemia in aged patients. METHODS: A total of 808 elderly patients were studied. Patients were classified into group I (serum glucose < 126 mg/dl), II (126-180 mg/dl) and III (> 180 mg/dl). Groups II and III were considered newly recognised fasting hyperglycaemia (NRFH) in non-diabetic patients. RESULTS: NRFH was present in 18.6%. After excluding diabetic patients (n = 206, 25.5%), the distribution of patients (n = 602, 74.5%) was as follows: group I (n = 452, 55.9%), group II (n = 122, 15.1%) and group III (n = 28, 3.5%). In the whole cohort, median fasting glucose was lower in patients who survived [105 mg/dl (88-135)] than in those who died [127 mg/dl (93-159), p < 0.001]. This significant difference was maintained only when non-diabetic patients were considered [100 mg/dl (87-122) vs. 118 mg/dl (92-149), p < 0.001]. In-hospital mortality rate in groups I, II and III was 8.5%, 14.1% and 22.9%, respectively (p < 0.001). Mortality rate was 8.4%, 18.0% and 32.1% (p < 0.001) in groups I, II and III, respectively in non-diabetic population. Both low albumin and high glucose serum concentrations were the only independent risk factors for in-hospital all-cause mortality in non-diabetic patients. CONCLUSIONS: In non-diabetic elderly patients admitted for acute disease, serum glucose concentration is an important, simple and independent predictor of hospital mortality.