Skip to main content

Impact of glycemic control on survival in critical care patients with sepsis


Several studies have shown an increased mortality and morbidity in various groups of patients with hyperglycemia requiring admission to the ICU. Contrary to these studies, a recent large multicentric international trial in ICU patients has shown an increased mortality in patients where normoglycemia was maintained. However, very few studies specifically evaluated the association between hyperglycemia and mortality in critical care patients with sepsis. Since hyperglycemia is known to be associated with immunosuppression and increased risk of life-threatening infections, we aimed to assess the impact of glycemic control (normoglycemia) in ICU patients with sepsis.


We studied all patients admitted to our ICU with sepsis between July 2004 and May 2009. Patients were excluded if their blood sugar levels (BSL) were not measured at least once every 6 hours during their ICU stay. Patients were divided into two groups based on their glycemic control in the ICU: normoglycemic group (BSL 3.9 to 6.2 mmol/l) and conventional glucose control group (BSL 8 to 10 mmol/l). BSLs were controlled using an infusion of actrapid insulin. Both groups were compared with regard to demographics, co-morbidities, severity of illness at ICU admission (APACHE III and SAPS II scores), ICU and hospital mortality and length of stay.


Over the 5-year study period a total of 314 patients were admitted with sepsis and 297 patients were included. On average every patient included in the study had BSL once every 4.2 hours throughout their ICU stay. Of these patients, 68 had normoglycemia (mean BSL 5.6) and 229 had conventional glucose control (mean BSL 8.5) during their ICU stay. There was no statistically significant difference between the groups in age, sex, APACHE III and SAPS II scores. The ICU mortality was significantly higher in the normoglycemic group (30.8% vs 14.8% P < 0.01). There was no statistically significant difference in hospital mortality between both groups (33.8% vs 25.3% P = 0.16). The ICU (median and IQR days: 1 (1 to 3) vs 4 (2 to 7) P = < 0.01) and hospital duration of stay (6.0 (2.00 to 13.75) 6.0 (2.00 to 13.75) vs 12.0 (7.00 to 27.00); P < 0.01) were significantly higher in the conventional group.


The ICU mortality in patients admitted to the ICU with sepsis who had normoglycemia was higher than in those patients who had conventional glucose control during their ICU stay. Aiming for a normal blood glucose levels is not indicated in patients with sepsis while they are in the ICU.

Author information



Corresponding author

Correspondence to R Tiruvoipati.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Tiruvoipati, R., Lewis, D., Ong, K. et al. Impact of glycemic control on survival in critical care patients with sepsis. Crit Care 14, P572 (2010).

Download citation


  • Hyperglycemia
  • Glycemic Control
  • Hospital Mortality
  • Blood Sugar Level
  • Conventional Group