Intestinal Failure - Feeding Access and Nutrition

Placing a 3-4 French feeding tube in distal bowel and bringing it out as a stent (not maturing as mucous fistula) allows easy distal feeding by syringe injection without catheter access issues.

When creating a jejunal feeding access, dividing the bowel 2-3 cm distal to ligament of Treitz and bringing the distal bowel up as a chimney with side-to-end anastomosis allows placement of a Mickey button without luminal obstruction.

When operating on a 2-month-old with jejunal atresia unable to advance feeds beyond 20 mL/kg, if the anastomosis is patent but dilated bowel is present and the child has 50% estimated bowel length, tapering is preferable to STEP.

STEP procedures done in the first year of life, especially in infants not progressing with enteral feeds, are not beneficial unless specifically avoiding line infections and bacterial overgrowth.

Dilated bowel does not become motile because you did a STEP procedure on it; if bowel hasn't been able to advance feeds, STEP will not make non-functioning bowel work.

The one thing that makes bowel adapt is feeding the bowel; procedures that cause problems in reinitiating feeds cause damage to the patient.

In the STEP registry paper on early neonatal STEP, only 3 patients came off parenteral nutrition, whereas natural data predict 80-90% of such patients would be off TPN at 12 months without surgery.

Since developing a mature multidisciplinary intestinal rehabilitation program, internal transplant candidates have declined by 75%, and intestinal transplants are almost non-existent.

Breast milk provides significant benefit to time to wean off parenteral nutrition in gastroschisis and NEC patients compared to formula-only feeding.

Donor breast milk is less advantageous than maternal breast milk because it is usually from mothers 10-14 months postpartum, has lower caloric density and protein, and freezing may inactivate trophic factors.

About 60% of patients weaned off TPN have deficiencies in one micronutrient or another.

Citrulline is produced by the gut and may predict ability to wean off TPN when measured as a total number, but as a weekly or monthly lab it does not inform clinical care changes.

Prebiotics (oligosaccharides) induce a more healthful flora; probiotics are not routinely recommended due to case reports of central line infections, though these likely occurred through direct line contamination rather than translocation.

Breast milk contains non-nutrient oligosaccharides (2-fucosyllactose, 3-fucosyllactose) that are immunomodulatory and affect microflora, not present in current formulas.

Metronidazole is the wrong choice for bacterial overgrowth because it has limited spectrum, knocks off anaerobes, and facilitates aerobic overgrowth; 2/3 of bloodstream infections were seen in patients on prophylactic Flagyl.

If using antibiotics for bacterial overgrowth, selective decontamination with non-absorbable aminoglycosides (tobramycin) and colistin may be preferable to metronidazole.

Anti-inflammatory agents (5-ASA products, steroid-based enemas) can be helpful in managing hypermotility when there is documented inflammation, particularly in challenging Hirschsprung's patients with dysbiosis.

Cholestyramine in effective doses will bind nutrients, fat-soluble vitamins, and fats, with risk of hyperchloremic acidosis and bezoars; homeopathic doses may improve stool appearance but not reduce volume.

Ursodiol (Actigall) in very short gut can contribute to diarrhea due to osmotic component and has no real benefit in preventing cholestasis.

Pancreatic enzymes in short gut patients go through before they've released, and there is concern they may cause strictures or stoma problems in areas of stenosis and dysmotility.

In infants, only pancreatic proteases are present in adult quantities; amylases appear between 6-12 months and lipase reaches adult levels by end of first year.

Acceptable stoma output is not defined by a hard number; patients with 40-50 cc/kg output can continue feeding if electrolyte profile is acceptable and CO2 is not dropping into the teens.

Knee-jerk reactions to volume of output (unless patient is sick with acidosis and abnormal electrolytes) should be avoided; variable feeding amounts over longer periods are detrimental to weaning from TPN.