Long-term sedation in the ICU: enteral versus parenteral feeding.

An open, prospective study was carried out on 45 patients with multiple injuries to compare the mortality and incidence of sepsis between those given early total enteral nutrition (TEN) when sedated with propofol and historical controls who had been given total parenteral nutrition (TPN) and sedated with midazolam. TEN was instituted immediately after surgery via gastrostomy and/or jejunostomy tube inserted during laparotomy or via an endoscope and was continued for the whole stay in the intensive care unit (ICU). Dramatic reductions in both mortality (24.4% vs 35.1% in the controls; p = 0.025) and the incidence of sepsis (8.9% vs 23.8% in the controls; p = 0.025) were found when early enteral feeding was given. The absence of pressure sores and gastro-intestinal bleeding (for example, stress ulcus), which had previously been a common occurrence in this intensive care unit, was remarkable with the introduction of TEN and propofol sedation. Tolerance problems (mainly diarrhoea) arose in only 17% of patients in the study group.