Surgical Necrotizing Enterocolitis: Time for a Definition

Abstract

I read with interest the recent article by Hull and colleagues on the mortality and surgical management of necrotizing enterocolitis (NEC) in very low birth weight (VLBW) infants. The authors used the Vermont Oxford Network (VON) database to evaluate birth-weight specific mortality rates for NEC over 5 years and also evaluated the mortality rates for surgical NEC based on whether the infant received a laparotomy, primary peritoneal drainage, or both. These data are certainly important to both surgeons and neonatologists in order to provide accurate and realistic expectations to families whose children have NEC. However, there is a significant limitation to their method of data collection that decreases the value of their findings significantly. Using the VON definition of surgical NEC is problematic. As the authors mention, spontaneous intestinal perforation (SIP) “cannot be incontrovertibly differentiated without a laparotomy” and thereby the peritoneal drainage group “necessarily contains patientswithbothdiagnoses.” The SIP and surgicalNECwere grouped together as 1 entity in their data. Although some may argue that this is inconsequential, we have recently shown that there is significant contamination of surgical NEC cases with SIP in a national database of more than 1,800 cases of surgical NEC. This held true even after excluding patients with dual diagnoses of both perforated bowel and surgical NEC. Spontaneous intestinal perforation is a distinct disease entity that typically occurs in the most immature and smallest infants within the first 2 weeks of life. We have also learned that NEC and SIP have different shortand long-term outcomes. Although both SIP and NEC have higher rates of mortality and neurodevelopmental impairment compared with infants without acquired intestinal disease,Wadhawan and associates found that patients with NEC had higher rates of death before hospital discharge and death or neurodevelopmental impairment compared with infants with SIP; Blakely and coauthors also found that patients with NEC had higher rates of death. In order to truly move our understanding of NEC and SIP forward, we must have a cleaner case-definition of surgical NEC. Including infants with SIP, congenital heart disease, volvulus, and gastroschisis (all subgroups that have NEC-like disease but have different origins of pathogenesis and strategies for prevention) in a dataset of surgical NEC inhibits our ability to truly understand NEC pathogenesis and prognosis. The use of a “clean” definition that is not contaminated with other confounding disease entities should be required for NEC research in the future.