Splanchnic Tissue Oxygenation During Enteral Feedings in Anemic Premature Infants at Risk for Necrotizing Enterocolitis

Necrotizing enterocolitis (NEC) is the most common gastrointestinal emergency encountered in
the newborn intensive care unit and represents a significant cause of morbidity and
mortality in infants born prematurely. This disease complicates the management of
approximately 6 - 10% of very low birthweight (VLBW) infants and can result in significant
feeding intolerance, intestinal perforation and/or death despite aggressive treatment (1).
The sequence of events leading to NEC appears to be multifactorial and complex (2,3). While
epidemiologic studies have identified multiple factors that appear to increase an infant's
risk for the development of NEC, other than prematurity, no single predictive risk factor
has been clearly delineated (4,5).

Among possible risk factors, a strong association between elective RBC transfusions in
premature infants with anemia and the subsequent development of NEC has been consistently
observed (6-11). Possible explanations for transfusion-associated NEC have been proposed:
1) the physiological impact of anemia that can initiate a cascade of events leading to
ischemic-hypoxemic mucosal gut injury predisposing to NEC (10); 2) increased splanchnic
blood flow following RBC transfusion leading to reperfusion injury of gut mucosa.

A significant (and increasing) number of VLBW infants with anemia are managed with
erythropoiesis stimulating agents (such as Epo) and iron and do not receive RBC transfusions
during their hospital stay. The present study proposes to study this particular group of
VLBW infants that remain with low (<28 %) hematocrit while receiving full enteral feedings.

Near Infrared Spectroscopy (NIRS) is a non-invasive, FDA approved, bedside technology that
allows determination of regional oxygen saturations (rSO2) in tissues such as the gut
mesentery. Using NIRS, the oxygenation status of hemoglobin in tissues located 2-4 cm below
the skin can be determined and recorded continuously (12). For instance, Dave et al. used
NIRS to demonstrate that splanchnic rSO2, but not cerebral rSO2, increases after feeds in
the stable prematurely born infant tolerating full bolus orogastric feedings (13). Of
importance, the average hematocrit in this study group was 37% (±7) and therefore these
babies had no significant anemia.

While no normative values exist for mesenteric rSO2 in premature infants, recent studies
have explored NIRS use in determining gut hypoxia and ischemia (14). Abdominal NIRS was used
to detect alterations of intestinal rSO2 and perfusion in premature piglets that developed
NEC (15). In a prospective cohort study of 40 neonates with medical or surgical acute
intraabdominal pathology, a cerebro-splanchnic oxygenation ratio (CSOR) of less than 0.75
predicted gut ischemia with 90% sensitivity (16).

While these studies support a role for NIRS monitoring of mesenteric rSO2, it is not clear
whether 1) VLBW with significant anemia have perturbations in intestinal oxygenation and
perfusion, and 2) alterations in mesenteric rSO2 predict the development of NEC in VLBW
infants.

We hypothesize that significant anemia in VLBW infants will be associated with a baseline
low CSOR (<0.75) as measured by NIRS, and that nasogastric feedings (NGF) in those
particular patients will lead to further decreased splanchnic oxygenation. We further
postulate that CSOR values will be significantly lower among VLBW that develop NEC as
compared to infants that do not.

Inclusion Criteria:

1. Premature infants of ≤ 32 weeks gestational age

2. anemia (hematocrit of ≤ 28 %)

3. full enteral feedings

4. stable clinical condition (no mechanical ventilation, no vasopressors, no sepsis)

5. Age ≤ 12 weeks of life

Exclusion Criteria:

1. Lack of parental consent

2. Multiple congenital anomalies

3. unstable clinical condition (mechanical ventilation, vasopressors, sepsis)

4. Previous medical or surgical NEC (defined as ≥ Bell's Stage II disease)