Acute hyperglycaemia following thermal injury: friend or foe?

INTRODUCTION: Hyperglycaemia and insulin resistance are common in severely burned patients, even if they have not previously had diabetes. Conventionally, hyperglycaemia is considered a part of the hypermetabolic stress response and blood glucose levels up to 215 mg/dl are tolerated before insulin therapy is initiated. Recent studies suggest that hyperglycaemia and insulin resistance are harmful and that correcting blood glucose to normal levels with insulin might improve the prognosis significantly. STUDY
OBJECTIVE: The purpose of this clinical study was to evaluate blood glucose levels in severely burned patients with conventional management and to analyse the association between early hyperglycaemia and clinical outcome.
DESIGN: Clinical, prospective, descriptive study. PATIENTS: Thirty seven severely burned adults (>25% total body surface area).
INTERVENTIONS: Hyperglycaemia was treated according to conventional clinical practice. This included the infusion of insulin based on a blood glucose level >215 mg/dl and the maintenance of the glucose level between 180 and 200 mg/dl. MEASUREMENTS AND
RESULTS: Measurements of whole-blood glucose were performed at 8, 16, 24, 36 and 48 h after the thermal injury. Additional measurements were performed if indicated. A total of 185 measurements were obtained and significant elevations of blood glucose levels (>140 mg/dl) were found in 108 (64%) of the measurements. Peak blood glucose values exceeded 140 mg/dl in all but three of the patients; however, only 17 patients received insulin treatment during the shock period. The inadequacy of the insulin treatment is shown by the mean glucose values, which exceeded 200 mg/dl in 27% of the patients. Despite a non-significant difference in the extent of burn (P=0.055), patients who died showed significantly higher maximum glucose values than patients who survived the thermal injury (P<0.05). Even though not statistically significant, blood glucose control was poorer in patients who later developed sepsis or acute renal failure (P>0.05). No correlation was found between burned surface area (TBSA) and mean plasma glucose levels during the first 48 h of resuscitation (r=0.12).
CONCLUSION: Hyperglycaemia is very frequent during the resuscitation period of thermal injury and current guidelines for insulin therapy are inadequate to correct plasma glucose to normal levels. As an association between early hyperglycaemia and subsequent mortality seems to exist, more aggressive manoeuvres to reduce blood glucose may be warranted in this group of patients.