Enteral nutrition versus glucose-based or lipid-based parenteral nutrition and tight glycaemic control in critically ill patients

OBJECTIVE: Early administration of nutrition and attention to tight glycaemic control are both associated with improved outcomes in critically ill patients. We hypothesized that blood glucose control would be better achieved in patients receiving enteral rather than parenteral nutrition and, in the latter group, would be better achieved using a "balanced" glucose plus lipid solution than a dextrose-based solution as calorie source. METHODS: We conducted a retrospective interrogation of the intensive care database as part of a clinical audit of a 12-bed mixed medical and surgical ICU in a tertiary referral teaching hospital between September 2003 and March 2004. Patients expected to stay in the ICU for longer than 48 hours were treated according to an intensive insulin therapy protocol. They received enteral nutrition (EN) or, if EN was not tolerated, parenteral nutrition (PN) or combined EN and PN. PN comprised a glucose-based solution (GluPN) during the first 3 months of the study and a balanced glucose plus lipid solution (LipPN) during the second 3 months. RESULTS: 96 patients were treated according to the protocol. Patients receiving PN (n = 26) had significantly longer ICU length of stay and greater daily caloric intake than did those receiving EN (n = 70) during both study periods. Mean blood glucose, percentage of blood glucose measurements within the target range, and daily insulin dose did not differ significantly between patients receiving EN and PN or GluPN and LipPN. CONCLUSION: When used in association with a tight glycaemic control regimen, PN is not associated with poorer glycaemic control in critically ill patients than EN.