Reproducibility and reference values of diaphragmatic shortening fraction for term and premature infants

Background New ultrasound measurements to diagnose diaphragmatic dysfunction, including diaphragmatic shortening fraction (DSF), have been studied in adults and children, but there are no data on reference values for neonates. Objective To describe DSF reference values for term neonate (TN) and preterm neonate (PTN), and to calculate its reproducibility. Methods We included asymptomatic TN and PTN during their first 24 hours of life. We measured DSF at the zone of apposition in both hemithoraces. Reproducibility of image acquisition, including inter- and intra-rater agreement of the measurements were calculated among an experienced and a novel operator (after completion of a 1-day course on lung ultrasound [LU] and performance of 10 diaphragm ultrasounds [DUs] under supervision), and a more-trained examiner (completion of a 1-day course on LU and performance of 60 DUs under supervision). Results Two groups of 33 TN and 33 PTN were studied. Median DSF values did not differ between the groups, although diaphragm thickness was higher in the TN group. Intra-observer reproducibility: the intraclass correlation coefficient (ICC) was 0.95 (95% confidence interval [95% CI] 0.86-0.98). Interobserver reproducibility with novel operator had an ICC of 0.42 (95% CI -0.74 to 0.81), and with a more experienced operator improved to 0.76 (95% CI 0.27-0.92). Both intra- and interobserver agreement were high. Conclusions Asymptomatic TN and PTN have similar DSF values in the first 24 hours of life. The intra- and interobserver agreement is high. Reproducibility is acceptable, but intensive training is necessary to perform adequate DU.