Abstract

OBJECTIVE:

To study the clinical and laboratory characteristics of patients with severe sepsis and baseline hyperglycemia and investigate the impact of hyperglycemia on the final outcome.

PATIENTS:

A total of 265 patients admitted with severe sepsis in 3 major hospitals in South-Western Greece, during a 1-year period, were included in the study. Patients were divided in 3 groups according to their glycemic profile at admission: patients with stress hyperglycemia (group SH, n=47), with diabetes mellitus (group DM, n=65), and with normal glucose level (group NG, n=153). Hyperglycemia was defined as an admission or in-hospital fasting glucose level of >or=126 mg/dL or a random blood glucose level of >or=200 mg/dL on >or=2 determinations.

RESULTS:

A total of 42.2% of patients with severe sepsis had baseline hyperglycemia with 17.7% having sepsis-induced stress hyperglycemia. No family history was noted in the SH group. A higher percentage of septic patients with stress hyperglycemia died compared with patients with normal glucose levels (42.5% versus 13.7%) and diabetics (42.5% versus 24.6%). Group DM had also a poorer prognosis than group NG (24.6% versus 13.7%). A positive correlation was detected between the fasting blood glucose levels of group SH and the severity of sepsis indicated by sepsis-related organ failure assessment score.

CONCLUSION:

Baseline hyperglycemia, including stress-induced hyperglycemia, is common in patients with severe sepsis. Stress-induced hyperglycemia is related to a more severe disease and poorer prognosis.