Background: There is a need for reliable diagnostic biomarkers in necrotizing enterocolitis (NEC). Whereas fecal calprotectin (fCP) has been reported to have insufficient sensitivity and specificity, no previous study has stratified for gestational and postnatal age. Objective: We aimed to provide developmental specific fCP data in premature infants and to analyze its value in detecting intestinal distress and episodes of NEC. Methods: Between April 2008 and December 2009, 1,899 fecal samples were obtained from 206 very low birth weight infants. Results: Mean gestational age (GA) was 28.5 weeks and birth weight 1,057 g. 19 (9.2%) patients developed NEC stage II+, of whom 5 had fulminant NEC with unusually low fCP concentrations in meconium and afterwards. fCP levels showed significant gestational and postnatal age dependent dynamics with particularly low levels in extremely premature infants. In infants with a GA <26 + 1 weeks using GA-adapted reference values, the sensitivity for discriminating moderate NEC from healthy infants and infants with intestinal distress was 0.89 for a cut-off of 180 and 210 µg/g, respectively, at onset of symptoms. Specificity was 0.96 and 0.84. Fulminant NEC was characterized by unusually low fCP concentrations with a cut-off of <24 µg/g having a sensitivity of 0.84 and a specificity of 0.72 for identifying those cases. Conclusions: fCP levels depend on gestational and postnatal age and in contrast to adults, there is a lower limit in premature infants. Taking these observations into account when defining reference values and interpreting fCP data in the clinical context, fCP can be a useful marker in identifying premature infants with gastrointestinal distress and NEC in particular.

Our early fed infants achieved better weight gain than those recently published receiving late enteral nutrition, but nevertheless fell below the 10th percentile of intrauterine curves. Which postnatal growth is ideal for extremely low birth weight infants infants is unclear. Our growth curves should not be taken as reference curves of a "normal population" but may help to identify infants with growth failure.

Despite successful resuscitation, infants between 23 and 26 weeks have a very poor prognosis for survival when presenting with bradycardia, cyanosis and no respiratory efforts (1-min Apgar = 1) at birth. According to our data, initiating active treatment for an infant at 23 weeks with bradycardia and apnoea is almost always unsuccessful, whereas by 26 weeks gestation, the chance of survival is higher than the probability of death.

Despite changes in obstetric and neonatal care during the 1990s, mortality and major morbidity rates did not change significantly after the introduction of surfactant in 1991. Comparison of local, regional, national, and international mortality and morbidity rates are becoming more important in allocating resources and in decision-making at the limits of viability.

No abstract