Targeting blood glucose levels to below 110 mg/dL with insulin therapy prevented morbidity but did not significantly reduce mortality among patients in a medical intensive care unit, said Dr. Greet Van den Berghe and her associates, of Catholic University of Leuven, Belgium.
A total of 1,200 adult patients who were predicted to require medical intensive care for at least 3 days were randomized to either strict normalization of glucose levels (80–110 mg/dL) with the use of infused insulin, or to conventional therapy in which insulin was given only when the blood glucose level exceeded 215 mg/dL and stopped below 180 mg/dL (N. Engl. J. Med. 2006;354:449–61).
The intensive treatment group experienced significantly fewer newly acquired kidney injuries than did the conventionally treated patients (9% vs. 6%), were weaned earlier from mechanical ventilation (hazard ratio 1.21), and were discharged earlier from both the ICU (1.15) and from the hospital (1.16). There was no significant effect on bacteremia, the researchers reported.
Among the 1,200 patients in the intention-to-treat analysis, ICU and in-hospital mortality were not significantly reduced by intensive insulin therapy. At day 3, mortality was 2.8% of the 605 patients randomized to conventional treatment, compared with 3.9% of the 595 in the intensive treatment group. Total in-hospital deaths occurred in 40% and 37%, respectively.
However, when the 767 patients who stayed in the ICU for more than 3 days were examined separately, the in-hospital mortality was reduced significantly, from 53% of the 381 conventionally treated patients to 43% of the 386 in the intensive treatment group, Dr. Van den Berghe and her associates reported.
In contrast, among the 433 patients who stayed in the ICU less than 3 days, mortality was slightly—but not significantly—higher in the intensive treatment group. After censoring 65 patients for whom intensive care had been limited or withdrawn within 72 hours after ICU admission, the in-hospital mortality was 15% for the conventional treatment group and 17% with intensive treatment.
The most likely explanation for the difference in the effect of insulin therapy in the group as a whole compared with those staying in the ICU at least 3 days is that benefits from intensive insulin therapy take time to be realized. Because the intervention is aimed at preventing complications that occur during—and perhaps as a result of—intensive care, it wouldn't be expected to work if the patient has a high risk of death from the disease that prompted the ICU admission, they said.
The mortality findings from these medical ICU patients differ from what the authors reported previously in a study of 1,548 surgical ICU patients, for whom mortality at 12 months was 8% with conventional treatment versus 4.6% with intensive insulin therapy (N. Engl. J. Med. 2001;345:1359–67).
When complications resulting from intensive care contribute to an adverse outcome, a preventive strategy like intensive glucose control is likely to be effective. This would explain why patients with long stays in the medical ICU benefit more than those with short stays, as was shown in the surgical ICU, they said.
Hypoglycemia was more common among the intensively treated patients and was also identified as an independent risk factor for death. However, among those who had hypoglycemia, the intensively treated patients were less likely to die in the ICU than were the conventional treatment patients (46% vs. 67%).
Contributing writer Giancarlo La Giorgia assisted with this report.