O2147: Case-matched comparison of outcomes following neonatal mucous fistula refeeding: 13 year institutional review

Background: Mucus fistula refeeding (MFR) was first introduced to provide neonates with short gut physiology additional enteral nutrition utilizing the surgically diverted distal bowel. Refeeding proximal ostomy effluent into the residual bowel via the mucus fistula has been associated with decreased parenteral nutrition needs and length to stay. However, in practice MFR can have complications, can be challenging to adopt within an institution and the literature is not yet definitive. Amid increasing adoption of the practice nationally, the limitations of MFR need to be defined. We present a case matched retrospective cohort comparison of neonates treated with and without MFR.

Methods: All neonates who underwent surgical enteral diversion between 2010-2023 were identified. Patients with necrotizing enterocolitis (NEC), spontaneous intestinal perforation (SIP), bowel atresia and midgut volvulus, meconium ileus, and gastroschisis were included. Neonates who underwent primary anastomosis, discharged or died prior to enterostomy closure (EC), or those with anorectal malformations or Hirschsprung’s disease were excluded. Data was obtained from the Childrens Hospital Neonatal Consortium database as well as institutional records. The MFR cohort was defined by neonates who received ≥7 days of MFR. The MFR and non-MFR cohorts were then propensity score matched by birth weight, gestational age, and ostomy level (duodenum, jejunum, ileum). Demographics, clinical details and outcome data were obtained. Outcomes evaluated included cholestasis, time to full feeds and ability to achieve full feeds prior to EC, time interval to EC and MFR complications including perforation, ostomy stricture, leak or prolapse.

Results: In total, there were 178 MFR neonates and 150 non-MFR eligible neonates during the study period. Of these, 262 neonates were able to be case matched (131 MFR and 131 non-MFR). Age, height, weight on admission and weight at initial laparotomy were comparable (Table 1). There were more neonates with NEC in the non-MFR vs MFR cohort (p < 0.001). We found no difference in ability of infants to achieve full feeds prior to EC, weight/height/head circumference at EC, peak bilirubin, time to EC or time to full feeds following EC. MFR neonates did have a lower incidence of cholestasis (p < 0.001), but no difference in central line (CVL) days or rate of CVL infections, number of abdominal reoperations and complications following EC, or length of stay. MFR specific complications occurred in 24 (18%) neonates and number of abdominal RO prior to ostomy closure was higher in MFR neonates (p=0.032).

Conclusion: MFR is an important tool in the postoperative management of diverted neonates. However, when critically analyzed via statistical case matching, our study shows that the benefits may be more nuanced. Randomized control data is needed.