2.2.1

Selection of participants

The ED tracking board was screened on a non-linear basis by a physician trained in bedside LUS. Potential subjects were identified by age (12 months and younger) and respiratory complaint. The treating physician was then approached to clarify the patient’s diagnosis. The diagnosis of AB was entirely clinical and based on the treating physician’s judgment. Exclusion criteria were subjects with any of the following: immunodeficiency or other immunosuppression; moderate to severe prematurity (<34 weeks); chronic pulmonary disease; chronic moderately to severely depressed heart function based on most recent echocardiogram; sickle cell disease; chronic neuromuscular disease; or, diagnosis of pneumonia within previous 14 days prior to ED presentation. Written informed consent was obtained from subjects’ caregivers, and video interpreters and translated informed consent documents were used for non-English speaking caregivers.

2.2.2

LUS acquisition

All LUS images were obtained during the subject’s ED visit, and were acquired and interpreted by one of three physician sonographers: two were point-of-care ultrasound (POCUS) fellowship-trained physicians, and the third was a pediatric emergency medicine fellow who demonstrated LUS competency prior to the start of the study [ ]. LUS images were securely stored on an external hard drive and interpretations were recorded in real time (Supplement 1).

The ultrasound machine used was a Sonosite X-porte. The L25x 13-6 Hz linear transducer was used for all scans set to the lung preset [ , ]. To perform the LUS, the subject was placed in positions that allowed access to all lung fields, while also ensuring the subject remained calm. Scans were divided into 12 lung fields as described by Supino, et al., namely, examination of the posterior, lateral, and anterior lung fields, divided in half into superior and inferior segments at the nipple line [ ]. In each segment, the sonographer held the probe in longitudinal position to the patient and started at the superolateral portion of the segment. Then, the sonographer scanned lateral to medial, then moved the probe inferiorly to scan the remaining intercostal spaces of that segment, moving medial to lateral, in a “lawnmower” approach [ ] ( Fig. 2 , Supplement 2).

2.2.3

LUS scoring and documentation

Previous publications report several LUS scoring criteria [ , , ]. After review of the literature, we used the scoring system by Brat, et al. [ ], which is described and scored as follows and seen in Fig. 1 :

• a)
  

  normal lung sliding, mostly A lines, and/or <3 B lines per lung segment; score of 0.

  
  
• b)
  

  ≥3 B lines per lung segment, but not consolidated/“white out”; score of 1.

  
  
• c)
  

  consolidated B lines/“white out,” but no subpleural consolidation or pleural effusion; score of 2.

  
  
• d)
  

  subpleural consolidation with any of the findings of b or c; score of 3.

Four lung ultrasound findings.

Example of the four lung ultrasound findings and associated scores. (a) shows A lines (white arrows) with <3 B lines (no B lines shown in this image), scores 0, (b) shows ≥3 B lines per lung segment (3 B lines shown in this image by white arrows), scores 1, (c) shows consolidated B lines, or “white out,” (white arrows) scores 2, (d) shows subpleural consolidation (shown with white *) with consolidated B lines, scores 3. In this final image, white arrows indicate air bronchograms, a common feature in subpleural consolidations.

Scanning technique.

12 scanning fields were used, divided as shown: anterior, lateral, and posterior, subdivided superiorly and inferiorly. The technique was performed by starting at the superolateral portion of the lung segment with the probe in longitudinal orientation, and moving the probe medially, then once reaching the edge of the segment, moving the probe inferiorly and scanning medial to lateral, and so on until all intercostal spaces were evaluated.

The physician sonographer scored each lung field and summated for a total possible score of 0–36 and recorded the scores on a pre-printed paper form. Additionally, start and end times were also documented (Supplement 1). To avoid introducing bias, the LUS findings were not shared with the treating physician including incidental findings unless, in the judgment of the physician sonographer, the results would place the patient at risk for rapid decompensation such as a subjectively large pleural effusion or pneumothorax. However, no significant incidental findings were identified.

2.2.4

Inter-rater reliability and quality control

All scans, except for one that could not be retrieved, were reviewed for quality assurance and scored by a separate pediatric emergency medicine physician with POCUS fellowship training, blinded to the original bedside sonographer’s scores. Additionally, a random sampling of subjects was scanned and scored by a second bedside sonographer within 30 min after the first bedside sonographer scanned the subject, but blinded to the first’s images and scores. Inter-rater reliability (IRR) was calculated with a linearly weighted Cohen’s kappa and defined as ±0.81 to ±1.00 excellent, ±0.61 to ±0.80 good, and ±0.41 to ±0.60 moderate.

2.2.5

Chart review

The primary author collected clinical data from the subjects’ electronic health record at least 7 days after discharge from the ED or the hospital if admitted. Data were manually harvested, then recorded in a secure spreadsheet. Data included demographics, vitals at triage and disposition, emergency severity index (ESI) at triage, ED findings including respiratory virus testing and CXR findings, admission and discharge diagnoses, and disposition from the ED. For admitted subjects, collected data included RS at 12 h and 24 h, as well as maximum RS during admission, hospital LOS as determined by time from admission order placed in the ED until discharge, and, if admitted to the floor, whether or not the patient was escalated to the intensive care unit (ICU) during their hospitalization. Also noted were the subjects initially discharged from the ED who returned to the ED within 7 days and were admitted for AB on this subsequent visit.