Anemic Conditions Acceptable in Restrictive Transfusion Practice Induce Gut Inflammation and Injury in an Animal Model of Preterm Infants

Very low birth weight (VLBW) infants often require multiple transfusions within the first weeks of life. In an effort to minimize transfusion-associated risks, many neonatologists have opted for more restrictive neonatal transfusion practices. However, whether a restrictive transfusion strategy is effective in limiting transfusions without increasing morbidity and mortality in this population remains unclear. Recently, the increasing use of more restrictive neonatal transfusion practices has been associated with a concomitant increase in the incidence of necrotising enterocolitis (NEC) in VLBW infants. Retrospective studies and a meta-analysis of trials comparing restrictive vs. liberal transfusion strategies suggest a higher risk of NEC among preterm infants managed by restrictive transfusion approaches (pooled relative risk 1.6; 95% CI 0.8-3.1). A recent multicenter, prospective observational study evaluating time-varying exposure to anemia also reported a higher risk of NEC among infants with severe anemia compared to those without severe anemia (adjusted hazard ratio 6.0; 95% CI 2.0-18.0). Importantly, while previous studies suggest that transfusion itself may cause transfusion-associated gut injury (TRAGI), this study failed to find an association between transfusion and NEC, strongly suggesting that anemia itself may represent a key risk factor linking increased incidence of NEC to restrictive transfusion practices.

To determine whether anemia may directly affect the development of NEC we employed a pre-clinical murine model that utilizes gradual phlebotomy-induced anemia (PIA). Pups were bled twice-daily (5mL/g) beginning at postnatal day 3 (P3) to achieve a hematocrit of 23-25% by P7. Anemic and littermate control pups were then examined for intestinal hypoxia, inflammation and gut barrier disruption at P8. Levels of intestinal hypoxia were evaluated by injection of hypoxyprobe (Hypoxyprobe, Inc.) into control and anemic pups, followed by sacrifice at 1 hour post-injection and detection of protein adducts of pimonidazole in hypoxic tissue by confocal analysis. Inflammation was assessed by evaluation of key immune populations, including macrophages, dendritic cells and T cells for number and function by flow cytrometry. Finally, gut barrier function was assessed by quantification of fluorescein isothiocyanate (FITC)-dextran (4kDa) in the serum 4 hours following oral gavage.

While hypoxia within the gut is normally confined to the lumen, neonatal pups exhibited significant increases in gut hypoxia following PIA, which extended beyond the gut lumen to intestinal crypts where key immune populations reside. Comparative analysis of immune populations in the gut between control and anemic neonates indicated no impact of PIA on macrophage, dendritic cell or T cells numbers. However, gut resident macrophages in anemic mice produced significantly more IFNg compared to controls. In contrast, no changes in dendritic cell or T cell cytokine production were observed, strongly suggesting that gut resident macrophages may play a primary role in anemia-induced increases in intestinal inflammation. As inflammation can directly impact gut barrier function, we next investigated intestinal barrier permeability as a marker of gut injury that may precede NEC. When FITC-dextran was administered by oral gavage to anemic and control mice, followed by quantification in the blood 4 hours later, significantly higher blood levels of FITC-dextran could be detected in anemic recipients, demonstrating increased intestinal barrier permeability following PIA.

While physiological hypoxia in healthy individuals is localized to epithelial cells adjacent to the lumen, our data demonstrate hypoxic staining throughout the mucosa in neonates following PIA. Additionally, PIA appears to increase macrophage pro-inflammatory cytokine secretion and also compromises gut barrier function, both key factors that can predispose neonates to NEC. Thus, data from this preclinical model suggest that restrictive transfusion guidelines which allow for hematocrits as low as ~24% prior to transfusion may increase morbidity and mortality associated with NEC in preterm infants.