RESUMEN
OBJECTIVE: To determine correlates of clinical outcomes in patients with short bowel syndrome (SBS). METHODS: Retrospective medical record review of neonates treated between 1986 and 1998 who met our criteria for SBS: dependence on parenteral nutrition (PN) for at least 90 days after surgical therapy for congenital or acquired intestinal diseases. RESULTS: Thirty subjects with complete data were identified; 13 (43%) had necrotizing enterocolitis, and 17 (57%)had intestinal malformations. Mean (SD) residual small bowel length was 83 (67) cm. Enteral feeding with breastmilk (r = -0.821) or an amino acid-based formula (r = -0.793) was associated with a shorter duration of PN, as were longer residual small bowel length (r = -0.475) and percentage of calories received enterally at 6 weeks after surgery(r = -0.527). Shorter time without diverting ileostomy or colostomy (r = 0.400), enteral feeding with a protein hydrolysate formula (r = -0.476), and percentage of calories received enterally at 6 weeks after surgery (r = -0.504) were associated with a lower peak direct bilirubin concentration. Presence of an intact ileocecal valve and frequency of catheter-related infections were not significantly correlated with duration of PN. In multivariate analysis, only residual small bowel length was a significant independent predictor of duration of PN, and only less time with a diverting ostomy was an independent predictor of peak direct bilirubin concentration. CONCLUSIONS: Although residual small bowel length remains an important predictor of duration of PN use in infants with SBS, other factors, such as use of breast milk or amino acid-based formula, may also play a role in intestinal adaptation. In addition, prompt restoration of intestinal continuity is associated with lowered risk of cholestatic liver disease. Early enteral feeding after surgery is associated both with reduced duration of PN and less cholestasis.