A Non‐invasive Novel Method to Assess Necrotizing Enterocolitis in Rodents

Abstract

Necrotizing enterocolitis (NEC) is the leading cause of mortality due to gastrointestinal (GI) disease in premature neonates. The precise pathogenesis of NEC is unknown, but early stages are thought to be characterized by damage to the intestinal mucosa and changes to the microbiota; while full intestinal necrosis, perforation, and multi‐system failure occur during the later stages. The early diagnosis of NEC would drastically reduce morbidity and mortality, but there are currently no predictive tests available to clinicians.Heart rate variability (HRV) is a non‐invasive way to measure changes in vagal tone, and our laboratory has shown that the high frequency power spectra, an indicator of vagal tone, is a useful factor to predict NEC in neonates before the onset of clinical symptoms (Doheny et al. Neurogastroenterol. Motil., 2014). Technical issues prevent the same power analysis in rodents; however, time‐domain analysis of HRV analyzes inter‐beat‐interval differences (IBID) to discriminate vagal tone. Our unpublished data indicate a correlation between IBID and GI motility, thus validating the usefulness of cardiac vagal tone measures as a non‐invasive indicator of GI motility (Meister and Travagli, unpublished).In the present work we hypothesize that: i) in rodents IBID is a valid marker of NEC; and, ii) the alterations of the neurochemical phenotype of myenteric and vagal preganglionic neurons are proportional to the severity of NEC.Increasing severity stages of NEC are induced in newborn Sprague‐Dawley rats including i) mild NEC (<Stage 2) through twice daily exposure to cold stress and hypoxia, ii) mild NEC and vagotomy (Stage 2–3), and iii) NEC (Stage 4) through maternal separation and formula feeding in addition to cold stress and hypoxia. Rats underwent HRV measures using bilateral electrodes before and after stress exposure on postnatal days (PD) 1 and 5, before being sacrificed on PD9. Gastrointestinal and brain tissues were fixed and analyzed with H&E (to identify NEC stages) and fluorescence immunohistochemistry (to assess the neurochemical phenotype).Control rats displayed typical developmental increases in IBID, while NEC treatments attenuated this effect. Control and all NEC rat groups showed similar numbers of neurons in the myenteric plexus (MP; PGP 9.5 staining) and similar proportion of choline acetyltransferase immunoreactivity (ChAT) in both the MP and DVC. NEC rats showed increases in nitric oxide synthase immunoreactivity (NOS) in both the MP and DVC. Finally, both mild NEC and NEC rats showed decreases in tyrosine hydroxylase immunoreactivity (TH) in the MP, but increases in the DVC. Data are summarized in the Table below.Our data indicate that: i) NEC rodents had an attenuated IBID, consistent with human HRV data; and, ii) myenteric and vagal preganglionic neurons undergo alterations in the NEC model proportional to the severity of the treatments.We suggest that HRV measures are a promising predictive measure of decreased vagal tone and therefore NEC in rodents. This novel NEC model provides a non‐invasive tool to test the efficacy of preventative experimental measures.Support or Funding InformationNIH DK99350 IBIDPD1 IBIDPD5 PGP 9.5 (neurons/field) ChAT NOS TH (MP: % of total; DVC: neurons/slice) MP MP DVC MP DVC MP DVC ControlN=2–7 1.039 ± 0.106 2.361 ± 0.505* 52 ± 2 67 ± 4 157 ± 16 22 ± 4 5 ± 1 26 ± 11 2 ± 1 Mild NECN=2–7 1.004 ± 0.146 1.300 ± 0.011 48 ± 3 60 ± 1 168 ± 20 31 ± 1 ‐‐‐ 18 ± 5* 40 ± 18* Mild NEC + vagotomyN=2–7 1.161 ± 0.218 1.465 ± 0.171 ‐‐‐ ‐‐‐ ‐‐‐ ‐‐‐ ‐‐‐ ‐‐‐ ‐‐‐ NECN=2–7 ‐‐‐ ‐‐‐ 58 ± 10 56 ± 2 194 ± 19 44 ± 3* 11 ± 4* 10 ± 1* 24 ± 6* p<0.05