ABSTRACT
OBJECTIVE: To determine the number, type, precipitating factors and mortality associated with diabetic emergencies admitted to Groote Schuur Hospital. DESIGN: Administered questionnaire survey. SETTING: Groote Schuur Hospital, Cape Town. METHODS: All patients admitted from September 1991 to January 1992 with elevated blood glucose concentrations who required intravenous insulin and fluids were evaluated by an administered questionnaire prior to discharge from hospital. Patients were divided into four groups according to presence of ketosis, standard bicarbonate level and serum osmolality: (i) mild diabetic keto-acidosis (DKA); (ii) severe DKA; (iii) hyperosmolar state; and (iv) hyperglycaemia. RESULTS: There were 131 admissions in 122 patients. Sixty-five occurred in non-insulin-dependent diabetics, 45 in insulin-dependent diabetics and 12 in pancreatic diabetics. There were similar numbers of admissions in the four groups of hyperglycaemic emergencies. The mortality rate was 3% in the mild DKA and hyperglycaemic groups, significantly lower (P < 0.05) than in the hyperosmolar group (19%). The mortality rate was 11% in the severe DKA group (P > 0.05 v. other groups). Sixty-two per cent of participants had achieved Standard 4 to Standard 8 education. Only 39% were employed at the time of interview. Diabetic knowledge and general compliance were assessed as poor. Infections and poor compliance with hypoglycaemic therapy were the dominant precipitating factors. CONCLUSION: Hyperglycaemic emergencies precipitated largely by infections and poor compliance and associated with considerable morbidity and mortality commonly require admission to hospital. Improvement in diabetic education is required to reverse the current situation.
