Comparison of the Silverman-Andersen Score and the Downes Score

The present study was designed as a prospective, multicenter, observational study conducted with neonates undergoing noninvasive ventilation. The study will be conducted in 4 centers (Mugla Training and Research Hospital, Mugla, Turkey; Akdeniz University School of Medicine, Antalya, Turkey; Antalya Training and Research Hospital, Antalya, Turkey; Sanliurfa Training and Research Hospital, Sanliurfa, Turkey). Written informed consent will be obtained from the parents of each neonatal patient prior to enrollment in the study.

Participants:

Preterm infants born at >28 weeks of gestation and term infants born at ≥37 weeks of gestation who developed respiratory distress within the first 6 hours after birth (defined as the presence of at least two of the following: respiratory rate ≥60/min, subcostal/intercostal retractions, and grunting), whose respiratory distress persisted for more than 6 hours, who were within the first 12 hours of life, had a Silverman-Andersen score ≥3, and received NIPPV support will be included in the study (1).

Infants with major congenital malformations, airway or pulmonary anomalies, cardiovascular or respiratory instability due to sepsis, cyanotic congenital heart disease, severe intraventricular hemorrhage, contraindications for NIPPV therapy (such as congenital nasal anomalies, congenital diaphragmatic hernia, or abdominal wall defects), those who required intubation or chest compressions in the delivery room, required intubation within the first hour of life, died within the first 24 hours of life, were born at ≤28 weeks of gestation, or were diagnosed with air leak syndrome (pneumothorax) prior to initiation of NIPPV therapy will be excluded from the study.

Respiratory support protocol Infants ≥34 weeks of gestation with respiratory distress within the first 6 hours of life will be observed for 2 hours; if distress persists and the Silverman-Andersen score is ≥3, NIPPV will be initiated. Infants \<34 weeks with respiratory distress and a Silverman-Andersen score ≥3 will be directly started on NIPPV. NIPPV is a routine, noninvasive respiratory support modality widely used in our unit and globally. Ventilatory support will be delivered via short binasal prongs using a mechanical ventilator.

Initial NIPPV settings will be: PIP 15-20 cmH₂O, PEEP 5-6 cmH₂O, rate 40-50 breaths/min, inspiratory time (Ti) 0.35-0.45 s, and flow 8-10 L/min. PIP may be increased by 1-2 cmH₂O increments up to a maximum of 25 cmH₂O if required. FiO₂ will be titrated to maintain oxygen saturation between 90-94%.

Assessment of respiratory severity Silverman-Andersen and Downes scores will be assessed 20-30 minutes after initiation of NIPPV using standardized scoring systems. Blood gas analysis will be performed within 30 minutes after NIPPV initiation. Additional indices of respiratory severity-including the respiratory severity score (RSS), FiO₂, mean airway pressure (MAP), and oxygen saturation index (OSI)-will be recorded at 20-30 minutes. RSS will be calculated as FiO₂ × MAP; MAP as [(Ti × PIP) + (Te × PEEP)] / (Ti + Te); and OSI as [MAP (cmH₂O) × FiO₂ (%)] / SpO₂ (%)] (18, 19). All scoring and calculations will be performed by a pediatrician, neonatology fellow, or neonatologist, each trained in standardized assessment prior to the study.

Definition of NIPPV failure and success NIPPV failure will be defined as the presence of severe respiratory acidosis (pH \<7.20, PaCO₂ >60 mmHg), oxygen saturation \<90% despite FiO₂ >50%, >3 apnea episodes within 1 hour, >2 apnea episodes requiring positive pressure ventilation with a bag-valve-mask device within 24 hours, or shock requiring inotropic support. Infants with NIPPV failure will be intubated and managed with conventional mechanical ventilation. Infants extubated within 1 hour following ENSURE (ENtubation, SURfactant, Extubation) will not be considered NIPPV failures, whereas those remaining intubated for >1 hour after ENSURE will be classified as failures.

Weaning from NIPPV will be attempted when PIP is reduced to 14-15 cmH₂O, PEEP to 5 cmH₂O, respiratory rate to 20-30/min, and FiO₂ \<30%, provided that the Silverman-Andersen score is \<3 and both clinical findings and blood gas parameters are acceptable. Infants successfully weaned from NIPPV will be considered to have achieved NIPPV success.

Surfactant and caffeine therapy Surfactant therapy is routinely performed in our unit and is not a novel intervention. Surfactant will be administered at a dose of 200 mg/kg via either LISA (Less Invasive Surfactant Administration) or ENSURE when FiO₂ ≥40% is required to maintain target oxygen saturation of 90-94%. In preterm infants with RDS who require intubation for respiratory failure, surfactant will be administered via an endotracheal tube (2). A second dose will be given if FiO₂ requirement persists at ≥40%. All infants born before 32 weeks of gestation or with a birth weight \<1500 g will routinely receive prophylactic caffeine therapy as standard practice.

Study parameters and outcomes The following parameters will be compared between infants with and without NIPPV failure: Silverman-Andersen score, Downes score, respiratory severity score, oxygen saturation index, SpO₂/FiO₂ ratio, duration of NIPPV, surfactant requirement, mortality, blood gas parameters, and demographic characteristics.

The primary outcome is to determine the most appropriate scoring system for predicting noninvasive ventilation failure by comparing the predictive performance of the Silverman-Andersen and Downes scores within the first 24 and 72 hours of life. The secondary outcome is to assess the correlation of both scores with duration of noninvasive ventilation, SpO₂/FiO₂ ratio, oxygen saturation index, respiratory severity score, and blood gas parameters (pH and PaCO₂), and to determine which scoring system demonstrates stronger correlation. Additionally, the predictive value of both scores for mortality and surfactant requirement in infants born before 34 weeks of gestation will be compared.

Sample size calculation Previous studies have shown that initiating respiratory support with NIPPV in infants with respiratory distress reduces the incidence of respiratory failure and the need for intubation compared to NCPAP (3). Permatahati et al. (4) reported an NCPAP failure rate of 37% when applied as primary respiratory support in preterm infants with respiratory distress. In the present study, the investigators hypothesized that the NIPPV failure rate would decrease from 37% to 20% when NIPPV is used as the primary respiratory support in neonates with respiratory distress. Based on this hypothesis, a power analysis was performed using G*Power version 3.1.9.4, with a type I error of 5% and 80% power, resulting in a required sample size of 220 patients.

Statistical analysis:

Data will be analyzed using the SPSS statistical software. Descriptive statistics will be presented as percentages (%), means, or medians, as appropriate. The Kolmogorov-Smirnov, Shapiro-Wilk, skewness, and kurtosis tests will be used to assess the normality of the data. Comparisons between infants with and without NIPPV failure will be performed using the Student's t-test, Mann-Whitney U test, and chi-square test, as appropriate.

Receiver operating characteristic (ROC) curve analysis will be used to calculate the area under the curve (AUC), sensitivity, and specificity of the Silverman-Andersen and Downes scores for predicting NIPPV failure. Comparisons of AUC values will be performed using MedCalc software. Optimal cut-off values for both scores in predicting NIPPV failure will be determined using the maximum Youden index.

Pearson and Spearman correlation analyses will be used to assess the relationship between the Silverman-Andersen and Downes scores and other respiratory distress parameters, including respiratory severity score, oxygen saturation index, blood gas pH and PaCO₂, and SpO₂/FiO₂ ratio. A p-value \<0.05 will be considered statistically significant.