Aim: To determine the clinical characteristics of severe hypoglycaemia (SH) in a nonselected German population. SH was' defined as an event requiring intravenous glucose or glucagon injection. Methods: The prospective population-based study screened sensitively for SH in a region with 200000 inhabitants between 1997 and 2000. All 30 768 patients who presented to the regional central hospital emergency department, and 6631 (85 %) of 7804 patients attended by the emergency medical service in the region were given an initial blood glucose test to detect atypical hypoglycaemia. Results: Altogether, 264 cases of SH were registered, which occurred either spontaneously (n = 14; 5 %), in subjects with type 1 (n = 92; 35 %) or type 2 diabetes (n = 148; 56 %), or in subjects with a non-classified form of diabetes (n = 10; 4 %). On the basis of the estimated local number of diabetic patients the annual rate of SH was i.5 episodes per 100 patients in insulin-treated type 2 diabetics compared with a rate of 0.4 episodes per 100 patients for the overall group of type 2 diabetic patients. Nocturnal hypoglycaemia accounted for 44% of episodes in patients with type 1 diabetes on intensified therapy but for only 25% in patients with type 2 diabetes. 26% of the hypoglycaemic individuals with type 1 diabetes had an impaired awareness of hypoglycaemia and thus recurrent hypoglycaemic episodes. Irrespective of the treatment, the most frequent contributing factors for SH in type 2 diabetic patients were advanced age (76 12 years), multimorbidity (3.6 +/- 2.6 concomitant diseases) - in particular renal impairment (54% [80/148]) - and polypharmacy (4 +/- 2.7 concomitant drugs). 34% (50/148) of the subjects with type 2 diabetes lived in nursing homes or were cared for by a home nursing service. With standardised treatment zero mortality of SH in diabetic patients was achieved, only one non-diabetic died due to hepatic failure. Conclusion: In elderly, multimorbid patients approaching the insulin-deficient end of the spectrum of type 2 diabetes the risk of developing SH increases considerably, nearing that in patients with type 1 diabetes. In order to avoid SH in geriatric patients, the treatment targets should be defined critically, taking into account individual quality of life and,life expectancy. Hypoglycaemia unawareness is a major risk factor for SH in type 1 diabetes.
