Background: Premature infants are at risk of intestinal failure, most often due to necrotizing enterocolitis and short bowel syndrome (SBS). Parenteral nutrition (PN) is essential for these infants, however, is complicated by sepsis and liver disease and is costly. Therefore, early enteral autonomy, through intestinal adaptation, is critical. Currently, there are no direct in vivo methods to measure adaptation in babies on PN. Plasma citrulline (CIT) is a proposed biomarker for adaptation, due to its correlation with intestinal length. Our aim was to evaluate the utility of CIT relative to the histological gold standard measure of adaptation, jejunal villus length. We evaluated two SBS anatomical subtypes: a mid-intestinal resection with jejunoileal anastomosis, and a distal resection of the ileum and ileocecal valve, which in our experience exhibits minimal adaptation.

Methods: Neonatal piglets (2-4d) were randomly allocated to either: 75% mid small intestinal resection (jejunoileal/JI, n= 5), 75% distal small intestinal resection (jejunocolic/JC, n=9), or sham control (n=8). A jugular catheter was inserted for PN and a gastric tube for trophic enteral nutrition. At laparotomy day 7, small intestine length and weight were measured, followed by jejunal tissue collection for blinded histopathological assessment of villus height and crypt depth. Baseline and terminal plasma CIT were measured using DC/LC-MS. Statistical analysis included independent t tests, one-way ANOVA, and linear regression.

Results: Amongst all piglets, mean CIT declined from baseline; sham (-42 %), JI (-64%) and JC (-58%). Day 7 CIT correlated with intestinal length (R²=0.34; p=0.003) and mucosal mass (R²=0.34; p=0.003). For SBS piglets alone, CIT did not correlate with intestinal length (p=0.7), mucosal mass (p=0.8) or jejunal villus height (p= 0.9). JI compared to JC piglets demonstrated an increase in small intestinal length (+18.3 vs -15.9cm; p<0.001) and greater mucosal mass (187.2 vs 153.5mg/cm; p=0.03). However, CIT did not differ between the SBS groups (729.0 vs 785.9pM; p=0.6).

Conclusion: In neonatal piglets plasma citrulline levels correlate with large differences in small intestinal length, and hence ‘identify’ SBS. However, citrulline did not discriminate between JI and JC surgical anatomy and failed to identify adaptive changes that were occurring in JI animals. Therefore, in this preclinical model of SBS, citrulline was not a useful biomarker of intestinal adaptation.