Enteral deeding tolerance in critically ill patients

Abstract

Objective: To assess tolerance of enteral feeding in intensive care unit (ICU) patients. Design: A prospective study conducted over 8 months (1 March - 31 October 1997). Setting: ICU at King Edward VIII Hospital, Durban, a tertiary referral hospital. Patients: Haemodynamically stable patients with a normal or functioning gastro-intestinal tract were enrolled immediately. Those with ileus were commenced on enteral feeding when the nasogastric output was < 500 ml/24 hours and bile was clear. Bowel sounds did not influence commencement of feeding. Patients were fed commercial semi-elemental or polymeric diets. Feeds were commenced at 20 ml/hour and the maximum rate was 80 ml/hour. Residual gastric volumes were measured daily. The endpoints of the study were discharge from the ICU, cessation of feeds because of intolerance, and death. Results: There were 96 patients (male/female ratio 2:1). Mean age was 34 ± 17.2 years. Mean time to commencement of feeds was 4.5 ± 3.7 days. Forty-five patients had no bowel sounds at commencement of feeding. Mean duration of feeding was similar in patients with intestinal anastomosis (18.24 ± 17.40 days) and those without (13.09 ± 9.77 days ) (P = 0.287). Twenty-two patients (23%) suffered feed-related complications, such as abdominal distention (13), vomiting (10) and diarrhoea (3). Feeding had to be stopped in 4 patients. Residual volumes were 35.95 ± 61.62 ml in patients with complications and 20.71 ± 35.25 ml in those without complications (P = 0.094). Twenty-nine patients died (30%). Conclusion: The majority of ICU patients tolerate enteral feeding, which may be commenced in the absence of bowel sounds. There was no significant difference in gastric residual volumes in patients with complications compared with those without complications. The presence of intestinal anastomosis did not influence duration of feeding.