• c-reactive protein
• diagnosis
• EOS
• preterm infants
• procalcitonin
• sepsis

Background: sepsis still represents today one of the most important causes of morbidity and mortality in the neonatal period, especially in preterm infants. Forms that begin in the first 72 hours of life are defined as early and arise following a vertical mother-infant transmission. Diagnosis is difficult to make by the presence of subtle and nonspecific signs and symptoms and the absence of sufficiently accurate chemical-clinical tests. To date, the diagnostic gold standard is represented by blood culture, but this, in addition to having low sensitivity, can take up to 48-72 hours to become positivized. In recent years, therefore, much effort has been made to identify early, sensitive, and specific diagnostic markers. Among these, particular attention was paid to procalcitonin and to the C reactive protein. Although both present a physiological increase in the first hours of life, they seem to be two potential and useful biomarkers of early neonatal sepsis, but the studies in the literature are still contradictory.
Objectives: the aim of the thesis was to evaluate and compare the average temporal trend of PCT and CRP in the first 72 hours of life in the septic and non-septic newborn, to evaluate the diagnostic accuracy of PCT and CRP in predicting the presence of early neonatal sepsis and its potential use in monitoring the response to empirical antibiotic therapy, to study and to interpret the possible interference of neonatal factors such as critical conditions at birth and respiratory distress and finally create reference ranges for PCT in the first 72 hours of life in uninfected preterm infants.
Methods: in the period between November 2015 and December 2020, newborns with gestational age ≤32 weeks + 6 days or neonatal weight ≤1500 grams were enrolled in the U.O. of Neonatology of Pisa and the blood levels of PCT and CRP were assessed at birth and every 24 hours until the third day of life. All patients, according to the care protocol in place in our department, underwent empiric antibiotic therapy from birth, with suspension after 72 hours in case of exclusion of sepsis. Other neonatal variables were also recorded such as critical conditions at birth and respiratory distress. The eligible patients were classified into septic and non-septic according to the international definitions of sepsis.
Results: 343 patients were enrolled, of whom 28 were septic and 315 non-septics. The two groups were homogeneous in terms of gestational age, weight, and sex. The analysis of the data revealed a statistically significant difference in PCT values at 24 hours between septic and non-septic infants (peak 97.66  103.94 ng/ml vs 15.12  23.54 ng/ml; p <0.001). Comparing the trend of CRP and PCT values in septic patients also showed a greater diagnostic accuracy of PCT from birth, with sensitivity and specificity at 24 hours of 64% and 96% respectively for PCT versus 80% and 82%. for CRP. PCT also proved to be a useful index of therapeutic response showing, in septic infants, an early drop in values already at 48 hours of life, unlike CRP which, in some patients, was still increasing. Furthermore, the other variables taken into consideration, including the severity of the clinical conditions and respiratory distress, seem to have some influence on PCT levels and less on CRP levels.
Conclusions: in the absence of biomarkers capable of identifying 100% of newborns with early sepsis, PCT and CRP are useful parameters for the diagnosis of sepsis. However, PCT is currently the most reliable index and could help in deciding to discontinue empirical antimicrobial therapy early. This would avoid an improperly protracted use of birth-initiated antibiotics in all preterm infants, helping to reduce neonatal mortality and morbidity, as well as drug resistance development and health care costs.