How Reliable Are Clinical Indicators of Readiness for Enteral Feeding?
Gloria Walters, MD; A. Joseph Layon, MD, FACP;
T. James Gallagher, MD, FCCM; Andrea Gabrielli, MD; Lawrence J. Caruso, MD;
Neil T. Bennett, MD.
Department of Anesthesiology, University of Florida College of Medicine
Background: The ability to absorb enterally administered nutrients is
dependent upon the intact functional status of the gastrointestinal (GI)
tract. Typically, we presume an intact functionality (and, by extension,
absorptive capacity) unless indications to the contrary exist and, when the
patient meets clinical criteria, enteral feeds are initiated as the
preferred route. The purpose of this investigation is to assess whether
these assumptions are, in fact, correct and whether or not the clinical
indicators of readiness to feed correlate with the ability to absorb an
enterally delivered nutritional load. Additionally, we will assess the
consequences of beginning a nutritional program in a patient with clinical
indicators of readiness to feed who demonstrates laboratory evidence to the
contrary.
Methods: 10 critically ill patients admitted in February 2003 to the
surgical intensive care unit of a large teaching hospital and having APACHE
II severity of injury scores greater than 18 will be maintained in an unfed
state for 2-5 days per routine following non-GI surgery. Gut function will
be evaluated using 4 independent assays [absorption of D-xylose,
L-rhamnose1, lactulose, and the lipid compound vitamin A) and nutritional
status will be assessed at postoperative day 1 and postoperative day 10
using prealbumin and retinol binding protein (RBP). As of day 5, it is
anticipated that all patients will meet the clinical criteria for enteral
feeding (no diarrhea, vomiting, or fistulae; audible bowel sounds) and will
have had placed either a Dobbhoff tube or a Gastrostomy-Jejunostomy tube. On
postoperative day 5, GI tract absorption will be measured by evaluating
urinary recovery of D-xylose, L-rhamnose, lactulose, and vitamin A after
enteral administration of an oral test solution containing 5 g of lactulose,
1 g of L-rhamnose, 0.5 g of D-xylose, and 600,000 IU Vitamin A dissolved in
water to a final volume of 100 mL. Urine will be collected via Foley
catheter for 5 hr starting immediately after administration of the test
solution and the saccharide content of the urine estimated and expressed as
a percentage recovery of the oral test solution. Vitamin A absorption
(examining retinol palmitate) will be performed by measuring vitamin A in
the blood three hours after ingestion of the oral solution (normal range: 0
min, 30-90 mg/dL; 180 min, >500 mg/dL). Serum creatinine, lactate2, and
arterial blood gases will be measured at the midpoint of urine collection
and gastric contents aspirated at the end of the collection period to
confirm that the test solution had been passed into the small intestine.
Results: It is anticipated that patients will segregate into those with
normal absorptive function and those with abnormal functional indices.
Prealbumin and RBP will also be assayed at two time points and change in
nutritional parameters will be cross-correlated with the result of the
absorption studies to assess for linkages. Complications will be catalogued
for all patients; anticipated complications include, but are not limited to:
diarrhea, C. difficile colitis, bloating, ileus, obstruction, vomiting, and
increased nasogastric tube output. If the study shows that there is
non-absorption coupled with no improvement of nutritional parameters in a
subgroup of patients (wasted feeding), there will be a discussion regarding
the cost of the study tests in comparison to the cost of waste feeding or of
temporizing institution of intravenous hyperalimentation as a bridge
measure.
1L-rhamnose, which is absorbed by passive unmediated diffusion, is sensitive
to changes in the functional absorptive capacity of the gut, in contrast to
D-xylose, which is absorbed via specific carrier mechanisms.
2In a study by John D. Johnston, et.al. (CRITICAL CARE MEDICINE
1996;24:1144-1149), serum lactate was inversely correlated with the
functional absorptive capacity of the intestines but not to serum creatinine
concentration.
