Extubation criteria:
Respiratory management in this center primarily used CPAP to support premature infants with GA >28 weeks, otherwise infants were routinely intubated. Intubated infants were supported with mechanical ventilation for several days. Weaning off mechanical ventilation to extubation followed standardized guidelines. Gradual weaning of FIO2 and peek inspiratory pressure (PIP) or tidal volume (VT) were performed guided by assessment of chest excursion, oxygen saturation and blood gases results. Infants were extubated when ventilatory rate was <25 breaths/min and PIP at 16-18 cmH2o that delivered desired VT. The process of weaning would typically require several days. The weaning practice did not have any additional “readiness test” before extubation. Infants were extubated to nasal CPAP or nasal non-invasive ventilation based on the observed work of breathing and presence of irregular breathing or apnea.
Lung ultrasound protocol 12:
A total of six lung zones were scanned, 3 on the right lung and 3 on the left lung [13]. These zones are summarized in Figure 1. Four patterns of lung findings were described (Figure 2) with graded severity score; the lowest score is zero and the highest is 3 for each zone (Supplementary Figures S1-S4). Therefore, the total lung score is 18, which is the sum of the 3 lung zones on both right and left sides (3x3x2 = 18) 12.
Focused cardiac ultrasound 9:
FCUS was done on postnatal days 3 and 7 by the same radiologist (R.H.) who was experienced with neonates and was blinded to clinical details of the studied infants. LVEI was measured at end-systole from the parasternal short axis 2D image at the mid-papillary muscle level. The formula (LVEI = D2/ Dl) was used where: Dl is the ventricular diameter perpendicular to the interventricular septum bisecting D2, and D2 is the diameter parallel to the interventricular septum. LVEI was considered abnormal when the ratio was >1.0 and a higher LVEI was considered a marker of increased pulmonary vascular resistance13-15. In addition, full echocardiography was done by pediatric cardiologist to screen for congenital heart disease and patent ductus arteriosus (PDA) and to measure PAP assessment on day 3 of life. Both lung and cardiac ultrasounds were performed using the Toshiba Nemio 30 Ultrasound System with the Toshiba PLM-805AT linear array ultrasound transducer probe at a frequency of 5-10 MHz (Toshiba, Tokyo, Japan) gray scale B mode examination.
PDA was visualized using 2D echocardiography and confirmed by color Doppler, PDA was considered hemodynamically significant if the left atrium to aortic root ratio in motion mode echocardiography exceeded 1.4 and if PDA size exceeded 1.5 mm 16. The pulmonary systolic pressure was determined by calculating the systolic pressure gradient peak between the right ventricle and the right atrium assessed through the simplified Bernoulli 7 equation (4 × V2), where V = peak systolic velocity of tricuspid regurgitation (TR) measured using continuous Doppler. The right atrial pressure was measured according to the collapse of the inferior venous cava during inspiration. The right ventricle systolic pressure, in absence of right ventricle outlet obstruction, is similar to the pulmonary systolic pressure. An estimated systolic pulmonary artery pressure exceeding 35 mmHg was considered pulmonary hypertension in preterm neonates17.