O2157: The Impact of Time to Intestinal Resection on Neurodevelopmental Outcomes in Necrotizing Enterocolitis: Findings From a Multicenter Retrospective Study

Background: Necrotizing enterocolitis (NEC) is associated with a substantial risk for neurodevelopmental impairment (NDI), particularly in surgical NEC where the incidence is as high as 57%. Animal studies have demonstrated that NDI in NEC is due to white matter injury that results from inflammation and TLR-4 mediated activation of microglial cells. Despite this and findings of an increased risk for NDI with peritoneal drainage, presumably due to prolonged exposure to affected intestine, there remains variation among surgeons in indications for resection in NEC. We hypothesized that a longer interval from NEC diagnosis to intestinal resection is associated with an increased incidence of NDI.

Methods: A retrospective study in 10 US children’s hospitals was conducted. Patients with NEC who underwent intestinal resection from 2010-2019 were included. Exclusion criteria included incomplete records, major congenital abnormalities, and first operation more than 10 days from diagnosis. The time of NEC diagnosis was determined by sites based on a combination of clinical and imaging factors. The primary outcome was NDI at 16-26 months of age, defined as: any blindness; hearing loss requiring amplification; inability to walk; cerebral palsy; and/or Bayley cognitive, language or motor scores < 70. Secondary outcomes included complications and length of stay. Comparisons were made between patients with and without NDI. After univariate analysis was completed, an iterative multiple regression model was used to assess the relationship between log-transformed time to resection and NDI controlling for factors associated with NDI on univariate analysis.

Results: We identified 347 patients with NEC. Neurological outcomes were documented in 199. Of this population, 65 (52.4%) were male and 74 (37.2%) had NDI (Table 1). Lower birth weight, earlier gestational age, lower weight at diagnosis, intraventricular hemorrhage (IVH), and longer time from diagnosis to resection (OR=1.74) were significantly associated with NDI. In multiple regression models, time from diagnosis to resection was an independent predictor of NDI when controlling for sex. The effect was slightly mitigated after controlling for GA and IVH (Figure).

Conclusion: In a large, multicenter study, time from diagnosis to bowel resection was associated with an odds ratio of 1.74 for NDI in infants with NEC. Multivariable analysis suggested that time to resection was an independent risk factor for NDI, but a more definitive conclusion could have been drawn from a better powered study in which more than 199 (57%) subjects had documented neurodevelopmental assessments. The relative independence of time to resection from IVH as a risk factor confirms the experimental finding of NEC-specific mechanisms of NDI. Our findings highlight the impact of surgical decision making in NEC on neurodevelopmental outcomes and suggest that surgeons should account for the risk of delay in resection on NDI in clinical management.