ABSTRACT
INTRODUCTION: The lung ultrasound score (LUS) has been suggested to predict moderate-severe bronchopulmonary dysplasia (msBPD) in preterm infants. We aimed to assess LUS evolution after birth in preterm infants and the effect of gestational age. METHODS: This multicentre prospective observational study was performed with newborns born before 33 weeks of gestation. We created two groups: group 1 (23-27 weeks) and group 2 (28-32 weeks). We compared LUSs between the groups from birth until 36 weeks of postmenstrual age, and we estimated the LUS evolution in each group with a linear multilevel mixed-effects regression model. The effects of the need for surfactant or an msBPD diagnosis were also studied. RESULTS: We included 339 patients: 122 (36%) in group 1 and 217 (64%) in group 2. The infants in group 1 showed a steady progression in the LUS from birth until 4 weeks of age and a subsequent decrease; the infants in group 2 showed a progressive decrease in the LUS throughout the study. This progression varied significantly in the first weeks of life in infants who required surfactant at birth and after the first week of life in the patients diagnosed with msBPD. DISCUSSION/CONCLUSIONS: Extremely preterm infants showed persistently high LUSs during the first weeks of life, regardless of the progression to msBPD. In this group, the infants who did not require surfactant at birth exhibited an increase in their LUSs after the first week until their values were equal to the remaining infants in their group.
Subject(s)
Bronchopulmonary Dysplasia , Pulmonary Surfactants , Bronchopulmonary Dysplasia/diagnostic imaging , Female , Fetal Growth Retardation , Humans , Infant , Infant, Extremely Premature , Infant, Newborn , Infant, Very Low Birth Weight , Lung/diagnostic imaging , Prospective Studies , Surface-Active AgentsABSTRACT
BACKGROUND: Different lung ultrasound (LUS) scanning protocols have been used, and the results in terms of diagnostic accuracy are heterogeneous. RESEARCH QUESTIONS: What is the diagnostic accuracy of the LUS score to predict moderate to severe bronchopulmonary dysplasia (msBPD)? Does scanning of posterior lung fields improve the diagnostic accuracy? STUDY DESIGN AND METHODS: This was a multicenter prospective, observational study in six centers. Two LUS aeration scores, one involving only anterolateral lung fields and the other adding the posterior fields were obtained at birth, on the third day of life (DOL), on the seventh DOL, on the 14th DOL, and on the 21st DOL. The diagnostic accuracy of both scores to predict msBPD was assessed at each time point. RESULTS: Eight hundred thirty-two LUS examinations in 298 infants were included. Both LUS score using anterolateral and posterior fields and LUS score using only anterolateral fields showed a similar moderate diagnostic accuracy to predict msBPD on the third DOL (area under the receiver operating characteristic curve [AUC] 95% CI, 0.68-0.85 vs 0.68-0.85; P = .97), seventh DOL (AUC 95% CI, 0.74-0.85 vs 0.74-0.84; P = .26), and 21st DOL (AUC 95% CI, 0.72-0.86 vs 0.74-0.88; P = .17). The LUS score using anterolateral and posterior fields was slightly more accurate at 14th DOL (AUC 95% CI, 0.69-0.83 vs 0.66-0.80; P = .01). A cutoff of 8 points in the LUS score using only anterolateral fields on the seventh DOL provided a sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio of 70%, 79%, 3.3, and 0.38, respectively, to predict msBPD. Adding gestational age (GA) and sex improved the discriminative value without significant differences compared with a predictive model based on multiple clinical variables: AUC 95% CI, 0.77-0.88 vs 0.80-0.91 (P = .52). INTERPRETATION: The LUS score is able to predict msBPD from the third DOL with a moderate diagnostic accuracy. Scanning posterior lung fields slightly improved diagnostic accuracy only at the 14th DOL. Adding GA and sex improves the diagnostic accuracy of the LUS scores. The LUS score is useful to stratify BPD risk early after birth.
