Pre-operative nutrition

If the value of nutritional supplementation in the general citizenry of developed countries is unproven, how about for those about to undergo surgery?

A pre-operative cohort of 192 malnourished American veterans were randomized to receive total parenteral nutrition (TPN) and 203 control malnourished patients received conventional nutrition [40]. Malnutrition was defined by serum albumin or prealbumin levels, by a body weight = 95% of ideal or a score = 100 on the Nutrition Risk Index. The Nutrition Risk Index uses body weight and serum albumin to calculate a malnutrition score. The TPN patients received therapy for at least seven preoperative days whereas the conventional patients went to surgery within 3 days. The TPN patients received TPN postoperatively for at least 72 hours. Conventional patients received no parenteral or enteral post operative feedings for 72 hours. After that time, they could receive whatever form of nutrition thought best. Overall infectious risk in a 30-day post operative period was 14.1% in the TPN group and 6.4% in the conventional group (p = 0.01). The bacteremia/fungemia rates for TPN patients were 4.2% vs 2.5% and the rate of pneumonias was almost double that of the conventional group when patients were stratified by degree of malnutrition. The most severely malnourished TPN patients had a 12.9% infectious complication rate as opposed to a 10.5% rate in the severely malnourished conventional patients. The preoperative TPN course did not therefore ameliorate the infectious risk-if anything it exacerbated it.

Early vs. late nutrition

Patients who have undergone trauma or surgery need to be fed eventually. Would early feeding prevent or increase the rate of infections? There are several studies that examine various subsets of these patients.

Nasojejunal tubes were placed in 17 Arizona head trauma patients and feeding started within 36 hours of admission [41]. Feeding of the 15 control patients was by the nasogastric route and never before 72 hours. The patients were studied only for 7 days after admission. In this short time period, there were 14 infections in the late feeding group and only 3 in the early feeding group. The predominant infection was "bronchitis" – a diagnosis whose criteria were not defined. Since this study was not blinded, observer bias would be hard to eliminate. A similar study was done in Tennessee using 30 patients fed nasogastrically before 72 hours or after gastric ileus was resolved [42]. There was no difference in the incidence of pneumonia or infections in general. A small study enrolling cervical trauma patients also failed to show an advantage with early enteral feeding [43].

Sixty-three abdominal trauma patients requiring celiotomy were entered into a study in which half the patients received intravenous glucose for the first five post-operative days and the other half had an immediate percutaneous jejunal feeding tube placed through which was administered an elemental diet [44]. Patients fed immediately had a "sepsis" rate of 9% while those not fed for five days had a rate of 29% in the first seven days (p < 0.05). Unfortunately, "sepsis" was not further defined. A meta-analysis of early feeding in patients undergoing gastrointestinal surgery included 11 studies and 837 patients [45]. The patients were fed either orally or by jejunal feeding. Early was defined as within 24 hours of surgery. The relative risk of all infections was 0.72 in favor of early feeding. The risk of pneumonia was between 0.7 and 0.8 and the risk of intra-abdominal abscess was between 0.8 and 0.9. The rationale behind early enteral feeding in gastrointestinal surgery is based on the putative increase in intestinal epithelial health and decreased translocation of intestinal flora into the systemic circulation. Numerous studies seem to confirm early enteral feedings' value in abdominal surgery, whether or not the rationale is valid. The benefit of early feeding has not been found in acutely burned children who were fed within 24 hours of the burn as opposed to after 48 hours [46], but the mean difference of 33 hours in delaying nutrition cannot be said to be large.

A review of 15 studies incorporating abdominal surgery, trauma, head injury and burns came to the conclusion that early (within 36 hours) enteral feeding yielded a significantly decreased rate of infection (19% vs 41% p = 0.049). There was considerable heterogeneity amongst the studies, with abdominal surgery studies having the most convincing risk reduction [47].

The benefit of early feeding in trauma and surgery patients does not appear to generalize to early enteral feeding in those with severe medical illnesses.

The FOOD study looked at the benefit of early tube feeding in dysphagic stroke patients from around the world [48]. Four hundred and thirty patients were randomized to avoid any enteral tube feedings for 7 days as opposed to 429 allocated to nasogastric or percutaneous endoscopic gastrostomy (PEG) feeding within 3 days of enrollment. Early feeding had no effect on the rates of pneumonia or urinary tract infections. Nor was there an effect on survival by delaying feeding for seven days.

All patients ventilated in a St. Louis ICU received an orogastric tube. Early-fed patients (n = 75) received their nutritional requirements from day 1, while the late fed (n = 75) patients received 20% of their requirements for 4 days and then their feedings were increased to their calculated optimal level [49]. The early feeding group had no fewer bacteremias or UTIs. The early feeding group had 65% more antibiotic days than the late feeding group (p < 0.001) and their length of stay in the ICU was almost twice as long. Ventilator-associated pneumonias occurred in 37% of the early feeding group as opposed to 23% of the late feeding group (p = 0.02). The size of this study and its' quality argues against an infectious benefit by early feeding of medical patients requiring ventilation.