Outcome measures
The study outcomes were composite. Composite adverse neonatal outcome included neonatal death, adverse respiratory outcomes (RDS,TTN ), hypoglycemia, newborn sepsis, confirmed seizures, stroke, IVH, cardiopulmonary resuscitation, mechanical ventilation, umbilical-cord-blood arterial pH < 7.0 or base excess < -12.5, a 5-minute Apgar score ≤ 3, and prolonged hospitalization (≥ 5 days). As indicated by previous studies, these outcomes were chosen as they are associated with significant risks of neonatal mortality or long-standing neonatal morbidities, including hypoxic ischemic encephalopathy [16, 12]. Secondary outcomes included neonatal resuscitation (metabolic acidosis and/or resuscitation at birth), metabolic complications (hypoglycemia and/or difficulties feeding), and respiratory support (both invasive and non-invasive).
Non reassuring fetal status was defined as catergory III [17] or persistent category II fetal heart rate pattern with abnormal labor progress [18], non-reactive NST associated with recurrent decelerations among non-laboring women [19], absent or reverse umbilical artery end diastolic flow in the setting of IUGR [20]. Clinical chorioamnionitis included maternal fever in association with uterine fundal tenderness, maternal tachycardia, fetal tachycardia, purulent or foul amniotic fluid [21]. Mild metabolic acidosis was defined as pH= 7.20 – 7.30 and BE = -6 − -12, moderate acidosis consisted in pH= 7 – 7.19 and BE = -12 − -16, while severe acidosis was pH < 7 and BE < -16. CPAP and oxygen administration represented non-invasive respiratory support, as opposed to mechanical ventilation, the invasive respiratory support. Neonatal sepsis was defined as a clinical syndrome prompting antibiotic treatment, with or without positive cultures. Cerebral lesions were suspected clinically, screened by neonatal brain ultrasound, and confirmed on MRI. (For details about specific definitions of outcomes please see Table S1) .