Date of Award

January 2014

Document Type

Open Access Thesis

Degree Name

Medical Doctor (MD)



First Advisor

Karen Diefenbach

Second Advisor

Richard Ehrenkranz

Subject Area(s)

Medicine, Nutrition


The purpose of this study was to determine whether feeding difficulties in post-operative neonates correlate with intraoperative findings.

A retrospective study of neonates undergoing gastrointestinal surgery between January 2002 and December 2005 was performed. Operative notes were used to classify infants into four groups based on post-operative anatomy and anticipated intestinal function: class 1: anatomically normal/normal function (n=22); class 2: anatomically normal/dysfunction (n=21); class 3: anatomically short/normal function (n=31); and class 4: anatomically short/dysfunction (n=21). Class 3 was further divided into two subgroups based on ostomy location: proximal ostomy (class 3a, n=11) vs. distal ileostomy (class 3b, n=21). Anatomically short was defined as loss of >50% of small bowel or high ostomy. Dysfunction was defined as decreased motility or absorptive capacity of the small bowel due to dilation, inflammation, or ischemia. Data were collected from the first day of enteral feeding until the infant reached full feeds or was discharged. Outcomes included: time to 50% and to full enteral feeds, days on TPN/lipids, and episodes of feeding intolerance (large aspirates, emesis) or malabsorption (increased volume or watery consistency of stools). Statistical analyses were performed using Kruskal-Wallis test for continuous variables and chi-square test for dichotomous variables.

We enrolled 95 patients. Time to full feeds was longer in anatomically short infants (class 3a and 4) than in anatomically normal infants (class 1 and 2, p<0.05). The same trend was seen in median days of exposure to TPN and lipids. Class 3b infants behaved more like anatomically normal infants despite having an ileostomy. Feeding intolerance occurred in 81% and 71% of infants in classes 2 and 4 respectively, which was significantly higher than in classes 1 (5%), 3a (55%), and 3b (30%), all p<0.05. The median days of feeding interruption due to intolerance were significantly higher in classes 2 and 4 (p<0.05). Malabsorption affected 62% and 64% of patients in classes 3a and 4, respectively, which was significantly higher than in classes 1 (5%), 2 (19%) or 3b (20%), all p<0.05. The median days of feeding interruption due to malabsorption were significantly higher in classes 3a and 4 (p<0.05).

These data demonstrate that surgeon-described post-operative anatomy and anticipated gastrointestinal function correlate with feeding difficulties in the post-operative period. We also found that infants with a distal ileostomy behave similarly to those who are anatomically normal, indicating feedings for these infants can likely be advanced more quickly. Feeding guidelines based on this classification system should be evaluated prospectively.