Abstract:
Delayed transition and transient tachypnea of the newborn (TTN) are common
causes of respiratory distress in term and near-term infants caused by delayed
postnatal lung liquid clearance. Risk factors for delayed transition and TTN are
prematurity, elective cesarean section, male sex, and perinatal asphyxia. Risk
factors associated with a prolonged course of TTN and respiratory support are
low umbilical cord pH, Apgar score, decreased SpO2, increased respiratory rates,
and increased CO2. Treatment includes close cardiorespiratory monitoring,
supplemental oxygen, and nasal continuous positive airway pressure (nCPAP).
Usually bearing a benign clinical course, delayed transition and TTN may
nonetheless lead to neonatal intensive care unit (NICU) admission with varying
duration and associated mother-child separation. Identifying bedside applicable
prognostic parameters to estimate the duration of nCPAP therapy as a surrogate
for the clinical course of TTN and delayed transition potentially reduces NICU
admission.
We designed a prospective observational study to estimate the duration of
nCPAP therapy in term and near-term infants ≥ 36 0/7 weeks of gestation with
delayed transition and TTN. The main outcome parameter was nCPAP therapy
duration (< 1 vs. ≥ 1 hour). Additional study aims were clinical and lung ultrasound
findings and evaluating interrater agreement of lung ultrasound scores. Thirty and
60 minutes postnatally, nCPAP duration was estimated based on the following
parameters: lung ultrasound score, Silverman-Andersen score, respiratory rate,
FiO2, SpO2, and respiratory acidosis in blood gas analysis. We also included the
previously described risk factors birth weight, gestational age, pH in cord blood
analysis, 1-minute Apgar score, and sex in our analysis. We used univariate and
multivariate analysis to evaluate the risk factors' influence on nCPAP therapy
duration and the intraclass correlation coefficient to test interrater agreement.
Thirty minutes postnatally, a lung ultrasound score > 5, FiO2 > 0.21, and
respiratory acidosis were associated with nCPAP therapy ≥ 1 hour. We
determined two probability cutoffs aiming at either a high sensitivity or high
specificity predicting nCPAP therapy ≥ 1 hour. We confirmed our model using classification and regression tree analysis. With an area under the curve of 0.87
in receiver operating characteristic analysis, our model proved to be a good
diagnostic test, potentially serving as a basis for developing a prognostic tool.
Sixty minutes postnatally, a Silverman-Andersen-Score ≥ 5 and a respiratory rate
> 60/min were associated with nCPAP therapy ≥ 1 hour. Due to unevenly
distributed data, we refrained from further analyses at the 60-minute time point.
None of the already known risk factors were associated with prolonged nCPAP
therapy in our cohort, but we confirmed several already known lung ultrasound
findings in TTN and, for the first time, in delayed transition. We identified the
double lung point and pleural line abnormalities as potential candidates for further
prognostic studies in TTN and delayed transition. We demonstrated a good
interrater agreement for lung ultrasound scoring between different raters: two
neonatologists and two pediatric radiologists.
With regard to our study’s main limitation of determining the lung ultrasound score
cutoff statistically based on our cohort’s data, our findings on risk factors for
prolonged nCPAP therapy and their potential use as a diagnostic tool must be
validated in an independent sample.
Print-on-Demand