Abstract
Intensive insulin treatment is associated with an increased risk of hypoglycaemia. The purpose of this study was to evaluate two different strategies: tight glucose control (TGC) versus intermediate glucose control (IGC). In this quasi-experimental study, 130 critically ill patients were assigned to receive either the TGC protocol (n=65), according to which blood glucose levels were maintained between 4.4 and 6.1 mmol/l, or the IGC protocol (n=65), according to which blood glucose levels were maintained between 4.4 and 8.0 mmol/l. A total of 52 subjects (40%) were diabetic and 63 (49%) were septic. In the IGC group, glucose levels were stabilised in the target range for a longer period of time when compared to the TGC group (63 vs 41%, P <0.001). The median capillary blood glucose level was 6.7 mmol/l in the TGC group (6.2 to 7.2) and 7.9 mmol/l (7.0 to 8.5) in the IGC group (P <0.001). The incidence of hypoglyacemia less than 2.2 mmol/l was 21.5% in the TGC group and 1.5% in the IGC group (P <0.001), and the incidence of hypoglycaemia less than 3.3 mmol/l was 67.7 and 26.2% (P <0.001) in the two groups, respectively. Diabetes (odds ratio 2.88, CI 1.22 to 6.84) and the TGC protocol (odds ratio 7.39, CI 3.15 to 17.35) were identified as independent risk factors for hypoglycaemia less than 3.3 mmol/l. Mechanical ventilation (odds ratio 4.33, CI 1.16 to 16.13), medical illness (odds ratio 2.88, CI 1.20 to 6.99) and hypoglycaemia (<3.3 mmol/l) (odds ratio 2.99, CI 1.21 to 7.41) were independent factors associated with mortality. TGC is difficult to accomplish in routine intensive care unit settings and is associated with a significant increase in the incidence of hypoglycaemia. Hypoglycaemia <3.3 mmol/l is an independent risk factor for in-hospital mortality.