Material and methods We conducted a prospective study over three years (2014–2016) on 61 newborns, 38 preterm infants with clinical signs of NEC and 23 infants used as control group. We simultaneously measured cerebral and splanchnic saturations through NIRS. We calculated the Receiver Operating Characteristics Curve for all the parameters, in order to assess their value as predictive factors for NEC.

Results The preterm infants who developed NEC had a mean gestational age of 31 weeks (IQR=27–33 weeks) and a mean birth weight of 1311 grams (IQR 900–1850). Splanchnic saturations had values between 15% and 69% in infants with NEC and 54%–88% in infants without NEC, with statistically significant differences between the two groups (p<<<0.01). Cerebral saturations had values between 15% and 85% in infants with NEC and 62%–95% in infants without NEC, also with significant differences (p<<<0.01). SCOR had values of 0.25–1.03 when NEC was diagnosed and 0.61–1.16 when NEC was absent (p<<<0.01). The area under the curve (AUC) is 0.914 for splanchnic saturation, 0.840 for cerebral saturation and 0.747 for SCOR (p<<<0.01).

Conclusions SCOR is significantly different in infants with and without NEC. Both its’ sensitivity and specificity are lower than those of splanchnic saturation, but SCOR can be preferred in the clinical settings due to its’ smaller variations.