Determining the reference range of blood presepsin in term and preterm neonates

Introduction Sepsis is still a major cause of morbidity and mortality in neonates, especially in preterm infants. Mortality can reach 60-70% in very low birth weight infants (birthweight 0.20). This reduced model explains 3.8% of the total sum of squares. After adjustment for all the factors in the model, presepsin levels appear to be significantly lower in twins (496 pg/ml  65.5 vs 655 pg/ml  11.8) and in neonates with Apgar at 1 min ≥8 (644 pg/ml  11.8 vs 774 pg/ml  56.2). So none of the above factors seems worth to be taken into account in determining the reference limits for presepsin blood levels in healthy term neonates. Preterm neonates. The largest differences in presepsin level are observed between small for gestational age (SGA) (903 pg/ml  57.1) and adequate for gestational age (AGA) neonates (703 pg/ml  26.7), between neonates with and without mechanical ventilation at blood sampling (1090 pg/ml  86.9 vs 711 pg/ml  24.7) and at delivery (855 pg/ml  87.3 vs 729 pg/ml  25.8), between neonates with and without venous catether (801 pg/ml  47.5 vs 716 pg/ml  28.9), between neonates who underwent blood sampling after the 4th day or before (797 pg/ml  46.2 vs 716 pg/ml  29.2), between males and females (778 pg/ml  35.1 vs 701 pg/ml  34.7). All these factors, when simultaneously introduced into a multivariable linear model, explain only 18.8% of the total sum of squares. A second multivariable linear model was fitted after removing the factors that showed the lowest effect on presepsin level (those associated with a p-value >0.50). This reduced model explains 13.4% of the total sum of squares. A third and more parsimonious multivariable linear model was fitted after removing the factors that showed the lowest effect on presepsin level (those associated with a p-value >0.20). This reduced model explains 12.3% of the total sum of squares. After adjustment for all the factors in the model, presepsin levels result to be significantly lower in AGA neonates (706 pg/ml  25.7 vs 890 pg/ml  55.0) and between neonates with and without mechanical ventilation at blood sampling (1074 pg/ml  85.3 vs 712 pg/ml  24.2). Even in this case, none of the above factors is expected to substantially affect the reference limits for presepsin blood levels in preterm neonates. Conclusion Presepsin blood levels seem to be quite independent of most of maternal and neonatal conditions examined in this study both in preterm and term neonates. The factors exerting significant effects (multiple birth and Apgar at 1 min, in term neonates, weight by gestational age and mechanical ventilation in preterm neonates) are expected to affect presepsin reference limits only to minor extent. References [1] Evaluation of a newly identified soluble CD14 subtype as a marker for sepsis. Yaegashi Y., Shirakawa K., Sato N., Suzuki Y., Kojika M., Imai S., Takahashi G., Miyata M., Furusako S., Endo S. J Infect Chemother. 2005;11:234-8. [2] CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infection in the acute care setting. Horan T.C., Andrus M., Dudeck M.A. Am J Infect Control. 2008;36:309-32.