401 - Lung Ultrasound Heterogeneity Index (LUSHI) Score and Furosemide Response in Neonates with Bronchopulmonary Dysplasia – a Pilot Study

Publication Number: 1384.401

Zohaib Sayyed, Connecticut Children's Medical Center, Manchester, CT, United States; Naveed Hussain, Connecticut Children's Medical Center, Farmington, CT, United States; Jacob Kelner, Connecticut Children's Medical Center, Hartford, CT, United States

Background: The effectiveness of diuretics in bronchopulmonary dysplasia (BPD) remains uncertain, but the use of furosemide is still common practice in some neonatal intensive care units. BPD is characterized by heterogeneous lung aeration, allowing for a "highest" and "lowest" LUS score to be given to each chest region. Using a lung ultrasound heterogeneity index (LUSHI) score, analogous to Cohen's d, the degree of variation in LUS scores across multiple chest regions can be quantified.

Objective: This study investigates whether LUSHI scores, used as a marker of lung heterogeneity, are associated with diuretic response.

Design/Methods: Prospective, single-center, pre-post observational study in a Level IV NICU of infants < 32 weeks gestation on non-invasive respiratory support who received furosemide. LUS of the anterior, anterolateral, and posterolateral chest was performed bilaterally within 12 hours before and 18-24 hours after the furosemide dose. The highest and lowest LUS score was assigned to each area. Infants were eligible for a second scan one week after the last furosemide dose.
Linear regression was performed to determine the association between LUSHI scores and furosemide efficacy. Efficacy was assessed by changes in LUS score or respiratory severity score (RSS = CPAP level × FiO2). Infants were grouped into homogeneous (LUSHI ≤ 0.8) and heterogeneous (LUSHI > 0.8) phenotypes. A Cohen's d cut-off of 0.8 represents a large effect size. Nonparametric testing was then performed to compare the two groups.
LUSHI = (Avg High - Avg Low Score)/(SD of Individual Scores)

Results: Sixteen pre-post LUSs were performed on 12 infants. Demographics are presented in Table 1. LUSHI score was negatively correlated with the change in the lowest LUS (r = -0.70, r2 = 0.49, p = 0.003). Infants with homogeneous lungs had a greater decrease in the lowest LUS score, with an average reduction of 1.44 (SD = 0.88), compared to an average increase of 0.71 (SD = 1.50) in infants with heterogeneous lungs (p = 0.01). There were no significant differences between phenotypes in changes in the highest LUS or RSS.

Conclusion(s): We hypothesize that infants with homogeneous lung patterns had more fixed lung disease mixed with regions of pulmonary edema that improved with furosemide, whereas those with heterogeneous lungs had fixed disease intermixed with non-edematous regions and thus showed minimal response. This novel observation suggests that LUSHI scores may help predict diuretic responsiveness in infants with BPD. However, the lack of significant change in RSS suggests that its clinical utility warrants further evaluation.

Demographics, Respiratory Support, and Lung Ultrasound Scores
Table 1: Demographics, Respiratory Support, and Lung Ultrasound Scores

Linear regression: LUSHI Score VS Change in Lowest LUS Score
Figure 1: Correlation Between LUSHI Score and Change in Lowest LUS Score
Linear regression shows a significant inverse relationship (r = −0.70, r² = 0.49, p = 0.003) between LUSHI and the change in the lowest LUS Score.

Change in Lowest LUS Score by LUSHI group
Figure 2: Comparison of the Change in Lowest LUS Score Between Homogeneous (LUSHI ≤0.8) and Heterogeneous (LUSHI >0.8) Groups
Box plot demonstrating significantly increased changes in the lowest LUS scores in the low LUSHI group compared to the high LUSHI group.