The Evidence: Rationale for Setting the Targets for Glucose Control in the Hospital
|
 |
Van den Berghe and colleagues performed a prospective, randomized, controlled study of 1548 adults in a surgical intensive care unit; 13% of the study patients had a history of diabetes. With intensive insulin therapy, BG levels of 103 ± 19 mg/dL (5.7 ± 1.06 mmol/L) were associated with a mortality rate of 4.6% compared with conventional therapy, in which BG levels of 153 ± 33 mg/dL (8.5 ± 1.8 mmol/L) were associated with a mortality rate of 8.0% (P<0.04).1 Hypoglycemia (BG <40 mg/dL) occurred in 5.2% of patients in the intensive therapy group, with no serious complications.1
In another prospective, randomized, controlled interventional trial, Malmberg and colleagues reported the results of insulin-glucose infusion followed by subcutaneous insulin therapy for 3 months in patients with diabetes and acute myocardial infarction (n=620). An absolute decrease in mortality of 29% overall at 1 year, which was maintained for 3.4 years, was reported. Mean BG at 24 hours in the intensive therapy group was 172.8 ± 59.4 mg/dL (9.6 ± 3.3 mmol/L); in the conventional therapy group, mean BG was 210.6 ± 73.8 mg/dL (11.7 ± 4.1 mmol/L).2 Hypoglycemia was observed in 15% of patients in the intensive-treatment group.
Data from other clinical settings support an association between hyperglycemia and outcomes.
Table 1. Evidence for Association of Blood Glucose Level With Clinical Outcomes summarizes key reports that serve as the basis for our current understanding of the association between hyperglycemia and outcomes in the hospital.
