In the Netherlands, more than 85% of the preterm infants born <32 weeks gestational age get antibiotics directly after birth because of the risk of infection with a bacteria. However, only 1 in 70 of these preterm babies actually has a bacterial infection. The use of antibiotics after birth can lead to problems on short term (bowel infection, infection with a bacteria later on or death) or long term (asthma, allergy, obesity).

The goal of the PRESAFE trial is to investigate whether addition of a biomarker (presepsin) to the Dutch early-onset neonatal sepsis (EOS) guideline safely reduces unnecessary empirical antibiotic exposure after birth in preterm infants born before 32 weeks gestational age. In this 874-subject multicenter, randomized clinical trial with a concurrent observational cohort, the hypothesis to be tested is that by adding presepsin to the national guideline the amount of unnecessary empirical antibiotic exposure after birth will be reduced with at least 30% without increase in infants with untreated sepsis. The study targets a population of clinical stable very preterm infants with risk factors for eary-onset neonatal sepsis. Antibiotic administration after birth is started to pre-emptively treat EOS.

By adding a presepsin-guided step to the Dutch EOS guideline for those infants qualifying for antibiotic treatment, it is assumed that the rate of antibiotic administration can be reduced. However, it is imperative that this reduction in antibiotics is not outweighed by an increase in (culture proven) EOS. Therefore, the co-primary outcomes of the study are: 1) the incidence of culture-proven EOS (non-inferiority) and 2) unnecessary antibiotics prescription i.e. antibiotic administration for ≤ 3 days when started within the first 72 hours after birth (superiority). Secondary outcomes include sepsis-related severity of illness, total number of antibiotic days when started < 72 hours after birth, and the composite outcome of necrotizing enterocolitis (NEC), late-onset sepsis (LOS), or death until discharge from the initial hospital.

STUDY POPULATION

All infants born at a gestational age of 24+0 to 31 6/7 weeks are eligible for enrollment. As part of standard care all infants will be screened and classified according to the Dutch EOS guideline for the indication of starting empirical antibiotics:

i. Infants at low risk of EOS who do not have an indication for empirical antibiotics according to the Dutch EOS guideline;

ii. Infants at high risk of EOS defined as: 1) suspected or confirmed diagnosis of maternal sepsis; 2) suspected or proven EOS in other infants (in case of multiple births) or infants born to mothers with previous infant with GBS disease/infection; 3) unexplained respiratory insufficiency requiring invasive mechanical ventilation and FiO2>0.40 or non-invasive ventilation with FiO2 >0.60 at time of randomization; 4) ongoing hemodynamic instability requiring inotrope medication or more than one 10 ml/kg fluid bolus at time of randomization; 5) strong clinical concern for sepsis due to physical exam findings (i.e. minimal responsiveness, poor tone).

iii. Infants at moderate risk of EOS who should be treated with empirical antibiotics based on the Dutch EOS guideline. In this group of infants the risks and benefits of receiving empiric antibiotics remain unclear and clinical equipoise is suggested.

Infants in category i. and ii. will be excluded for randomization and included in the observational part of the study; these patients will be further treated according to the Dutch EOS guideline. Infants in category iii. are eligible to be included in the RCT-part of the study.

RANDOMIZATION

Included category iii infants will be evaluated for EOS with blood culture and randomly assigned (1:1) to one of the following study groups:

Intervention-group, presepsin-guided therapy: The decision to start empirical antibiotics in this group will be based on presepsin measurement within four hours after birth. When the presepsin value is above the cut-off value of 645 pg/ml, antibiotics will be ordered and administered within 4 hours of life and discontinued following the criteria of the Dutch EOS guideline. Cessation of antibiotic treatment typically occurs when blood culture results remain sterile within 3 days after birth.

When the presepsin level is below the cut-off value of 645 pg/ml, the treating physician will not start antibiotic treatment. These infants will be closely observed for at least 72 hours. In case of deterioration of the clinical condition within this observation period, the clinician can decide to perform a sepsis evaluation and start with antibiotic treatment. Investigators anticipate that <5% of infant subjects will experience worsening of conditions. When the initial blood culture in this group turns back positive without clinical signs and symptoms of EOS sepsis, a new blood culture and antibiotic treatment will be advised.

Comparator-group, standard care: Standard care according to the Dutch EOS guideline. Presepsin will be determined with the treating physician blinded for the test result. In the Dutch EOS guideline maternal and neonatal risk factors for EOS are categorized as red flags or minor criteria. In the presence of 1 red flag or ≥ 2 minor criteria it is advised to perform a blood culture (= sepsis evaluation) and start empirical antibiotics for EOS suspicion.2 Antibiotic treatment will be advised to discontinue when blood culture turns back negative, reassuring the infants clinical condition with no other indicators of possible infection (e.g. CRP).

Randomization will be performed centrally, using web-based block randomization stratified by study site with random permuted blocks within the strata. Randomization sequence is generated using Castor EDC, which is regulatory compliant software, ensuring allocation concealment. Multiples (i.e. siblings) will be randomized to the same treatment arm.

Inclusion Criteria:

• Infants born at a gestational age of 24+0 to 31 6/7 weeks
• Moderate risk of early-onset neonatal sepsis, i.e. infants who should be treated with empirical antibiotics based on the Dutch EOS guideline.

Exclusion Criteria:

• low risk of early-onset neonatal sepsis who do not have an indication for empirical antibiotics according to the Dutch EOS guideline;

• high risk of early-onset neonatal sepsis defined as:

  1. suspected or confirmed diagnosis of maternal sepsis;
  2. suspected or proven EOS in other infants (in case of multiple births) or infants born to mothers with previous infant with GBS disease/infection;
  3. unexplained respiratory insufficiency requiring invasive mechanical ventilation and FiO2>0.40 or non-invasive ventilation with FiO2 >0.60 at time of randomization;
  4. ongoing hemodynamic instability requiring inotrope medication or more than one 10 ml/kg fluid bolus at time of randomization;
  5. strong clinical concern for sepsis due to physical exam findings (i.e. minimal responsiveness, poor tone).

Data will become available after publication of the main article in peer-reviewed journal.