Respiratory Medicine

Point-of-care lung ultrasound score for predicting escalated care in children with respiratory distress

a b s t r a c t

Purpose: Respiratory distress due to lower respiratory illnesses is a leading cause of death in children. Early rec- ognition of high-risk populations is critical for the allocation of adequate resources. Our goal was to assess whether the lung ultrasound (US) score obtained at admission in children with respiratory distress predicts the need for escalated care.

Methods: This prospective study included 0-18-year-old patients with respiratory distress admitted to three emergency departments in the state of Sao Paulo, Brazil, between July 2019 and September 2021. The enrolled patients underwent lung US performed by a pediatric emergency physician within two hours of arrival. Lung ul- trasound scores ranging from 0 to 36 were computed. The primary outcome was the need for high-flow nasal cannula , noninvasive ventilation (NIV), or mechanical ventilation within 24 h.

Results: A total of 103 patients were included. The diagnoses included wheezing (33%), bronchiolitis (27%), pneu- monia (16%), asthma (9%), and miscellaneous (16%). Thirty-five patients (34%) required escalated care and had a higher lung ultrasound score: median 13 (0-34) vs 2 (0-21), p < 0.0001; area under the curve (AUC): 0.81 (95% confidence interval [CI]: 0.71-0.90). The best cut-off score derived from Youden’s index was seven (sensitivity: 71.4%; specificity: 79.4%; odds ratio (OR): 9.6 [95% CI: 3.8-24.7]). A lung US score above 12 was highly specific and had a positive likelihood ratio of 8.74 (95% CI:3.21-23.86).

Conclusion: An elevated lung US score measured in the first assessment of children with any type of respiratory distress was predictive of severity as defined by the need for escalated care with HFNC, NIV, or mechanical ventilation.

(C) 2023

  1. Introduction

Abbreviations: aOR, adjusted Odds Ratio; AUC, Area Under the Curve; CI, Confidence Interval; CRS, Clinical Respiratory Score; HFNC, High Flow Nasal cannula; LHR, Likelihood Ratio; NIV, Noninvasive Ventilation; OR, Odds Ratio; PED, Pediatric Emergency Department; PEM, Pediatric Emergency Physician; RDS, Respiratory Distress Syndrome; ROC, Receiver operating curves; SD, Standard Deviation; US, Ultrasound.

* Corresponding author at: Instituto da Crianca da Faculdade de Medicina da Universidade de Sao Paulo, Av. Dr. Eneas Carvalho de Aguiar, 647, Cerqueira Cesar, Sao Paulo – SP – Brasil, CEP – 05403.000, Brazil.

E-mail address: [email protected] (E.P.C. Giorno).

Respiratory disorders represent a major cause of emergency depart- ment visits by children and are currently the leading cause of death among children under five years of age worldwide [1]. Although a wide range of acute or chronic lung diseases may progress to respiratory failure, asthma and acute lower respiratory tract infections remain the most common causes of hospitalization for severe illnesses in children and place a significant burden on the healthcare system [1-3].

While most children admitted for respiratory distress only require

minor interventions such as oxygen support [4], some will progress to

https://doi.org/10.1016/j.ajem.2023.02.035

0735-6757/(C) 2023

a more severe illness, and as many as 10% [5] of hospitalized children may require escalated care with high-flow nasal cannula (HFNC), non- invasive ventilation (NIV), or mechanical ventilation. Timely detection of these high-risk patients in primary care or emergency departments is critical for Close monitoring and early transfer to a facility capable of providing escalated care if necessary.

Several clinical features that predict poor outcomes and the need for escalated care in pediatric respiratory distress have been identified. Hypoxia, chest retractions, and tachypnea are the most concerning iso- lated signs that frequently precede clinical deterioration [6]; however, isolated signs are rarely sufficient to properly triage patients. Further- more, the role of imaging modalities in predicting illness severity re- mains unclear. Chest radiography might predict the severity and outcomes of pneumonia [7], but it has limited utility for lower respira- tory tract viral infections [8,9].

Lung ultrasound (US) is an ionizing radiation-free imaging modality that promptly provides bedside diagnosis of many pulmonary condi- tions [10-12] in the emergency department, and enables clinicians to monitor lung aeration and its variations [13-15]. Because different US patterns correspond to varying degrees of aeration loss, many lung US scores have been increasingly used in adult critical care to quantify lung aeration as a whole or in a specific area of interest. Lung US score has been shown to accurately predict post-extubation distress in adult patients undergoing ventilator weaning [16] and is significantly corre- lated with the need for surfactant replacement in preterm neonates [17,18], and intensive care needs in infants with bronchiolitis [19].

The above-mentioned studies have addressed specific conditions, but little is known about lung US score performance in a general popu- lation of pediatric patients with respiratory distress presenting to the Pediatric Emergency Department (PED). As the lung US score can pro- vide a semi-quantitative measure of lung aeration loss caused by vari- ous pathological conditions, we hypothesized that a higher lung US score could predict the need for escalated care in children with respira- tory distress due to any type of lower respiratory illness. Our main ob- jective was to evaluate the ability of the lung US score obtained at admission to predict the need for escalated care within the next 24 h in children with respiratory distress.

  1. Methods
    1. Setting and participants

We conducted a prospective study from July 2019 to September 2021 in the PEDs of three academic hospitals affiliated with the Univer- sity of Sao Paulo and the University of Campinas in the state of Sao Paulo, Brazil. Patients were enrolled as a convenience sample, and re- cruitment was determined by the availability of lung US-trained emer- gency physicians.

Patients aged 0-18 years were eligible if they presented with respi- ratory distress, as defined by the presence of any of the following signs: tachypnea, intercostal and substernal retraction, nasal flaring, or cyano- sis [20]. Respiratory distress caused by both obstructive and restrictive pulmonary diseases were included because they frequently overlap and cover a wide range of lower respiratory illnesses that present to the PED. We included children with any type of lower respiratory illness who presented to the PED, rather than only those with a specific pulmo- nary disease, for two reasons. First, our intention was to investigate the performance of the lung US score as a rapid triage tool that could be used as soon as the patient arrives, when in most real-world scenarios, the etiology of respiratory distress has not been established yet. Second, we hypothesized that some patients may have overlapping conditions, such as atelectasis or infection-related interstitial or alveolar involve- ment, which could be responsible for disease severity, and that these children would not be investigated if a specific diagnosis was chosen.

Patients were excluded if they had respiratory failure due to central

or Neuromuscular dysfunction, if prompt intubation was indicated, if

they had already been transferred from primary care to escalated care, or if the patient was on HFNC, NIV, or mechanical ventilation before lung US was performed. Patients with the following conditions were also excluded: cardiac disease requiring diuretics, daily oxygen supple- mentation, or any chronic lung disease other than asthma.

    1. Data collection

Age, sex, vital signs upon presentation, and final diagnosis as re- corded in the discharge summary were obtained. Heart and respiratory rates upon presentation were assessed as normal or elevated according to the pediatric advanced life support guideline reference values for age [21]. chest radiographic findings were classified as normal or abnor- mal based on the attending emergency physician’s interpretation described in the patient files.

The clinical respiratory score (CRS; Supplemental Table S2) was used to assess clinical severity, which incorporates the following parameters: respiratory rate, pulmonary auscultation, use of accessory muscles, room air oxygenation, mental status (normal, agitated, or lethargic), and color (normal, pale, or cyanotic) [22]. Clinical features upon presentation and CRS were recorded in a standardized abstract form by the attending physician who performed the physical examination.

A nasopharyngeal swab for 17 respiratory viruses was collected depending upon kit availability.

Lung US was performed by one of the seven pediatric emergency medicine (PEM) academic teachers with more than five years’ experi- ence in teaching and supervising lung ultrasound. Images were ob- tained using B-mode, basic preset, with depth and 2D-gain adjusted according to each patient biotype. A high-frequency linear transducer was used (Logic E, GE Healthcare, Chicago, IL, United States, or portable device M2D, Mobissom, Brazil), and the technique was similar to that described by Copetti and Catarossi [23] in which each lung was divided into six regions (upper anterior, lower anterior, upper lateral, lower lat- eral, upper posterior, and lower posterior), resulting in a total of 12 re- gions. For each zone, a 0-3 score was given according to the pattern found. The patterns are described below and illustrated in Fig. 1.A global lung US score ranging from 0 to 36 was obtained by summing the individ- ual scores for all the regions. The worst finding in each region was used to describe the findings and calculate the lung US score. To evaluate inter- observer agreement, images of 75 patients were reviewed and scored by an expert sonologist blinded to all clinical information.

The four validated US patterns and the scores given for each were as

follows [13,15]: 1) normal pattern, presence of Lung sliding and artifac- tual horizontal A-lines (0 points); 2) B-pattern, presence of three or more well-defined vertical B-lines extending from the pleural line, indi- cating the presence of fluid in the interstitium (1 point); 3) severe B- pattern, multiple confluent vertical B-lines extending from the pleural line with or without small (<1 cm) subpleural consolidations, indicat- ing fluid in the alveolar space (2 points); and 4) lung consolidation, presence of a tissue structure with or without hyperechoic punctiform images resembling air bronchograms (3 points).

    1. Study protocol

Eligible patients had undergone lung US within two hours of arrival whenever one of the seven trained PEM physicians was available and not in charge of patients’ care. The attending physician was unaware of the total lung US score. The primary outcome was escalated care within 24 h of arrival in PED, defined by the need for HFNC, NIV, or me- chanical ventilation. Despite the absence of established criteria for esca- lated care [24,25], these interventions are generally indicated for patients with Refractory hypoxemia and/or worsening respiratory dis- tress. The decision regarding whether to provide these interventions was made by the attending physician who was not involved in the study. The lung US score was not used to guide escalated care and was not recorded in the patient files.

Image of Fig. 1

Fig. 1. Lung ultrasound (US) score.

Scores are given as follows: 0 for the presence of only A-lines (stars), 1 for the presence of more than two well-spaced B-lines (arrowheads), 2 for the presence of coalescent B-lines, and 3 for consolidations >1 cm with hyperechoic punctiform images resembling air bronchograms (arrow).

This study was approved by the local ethics board and written in- formed consent was obtained from the patients’ parents or guardians.

    1. Statistical analysis

Patient characteristics are expressed as absolute (n) and relative (%) frequencies for qualitative variables and as the mean, median, mini- mum, maximum, and standard deviation (SD) for quantitative vari- ables. To assess the possible association between two qualitative variables, the chi-square test or Fisher’s exact test was used, as appropri- ate. Independent groups were compared using either the Student’s parametric t-test or the Mann-Whitney U non-parametric test. The data were tested for normality using the Kolmogorov-Smirnov test. The capacity to discriminate between events and non-events was assessed using receiver operating curves (ROC) and the area under the curve (AUC) with 95% confidence intervals (CI). The optimal cutoff point was determined as the point at which the Youden index was max- imized. positive and negative likelihood ratios (LHR), along with the 95% CI, were reported. To quantify the association between the US score and the outcome, a univariable logistic regression model was fitted to the data. Another multivariable logistic regression model was fitted considering the lung US score as a covariate adjusted for sex, age, pulse oximetry, mental status, capillary refil time, chest radiograph, and CRS. Odds ratios (ORs) and 95% CIs were also calculated. The goodness-of-fit of the models was evaluated using the Hosmer- Lemeshow model. The correlation between the lung US score and CRS was computed using Spearman rank correlations. The Cohen weighted kappa coefficient was calculated to estimate inter-observer agreement. All hypotheses were tested at the 5% significance level. Data analysis was performed using the R software (version 4.0 [R Core Team, 2020]). The sample size was calculated based on the data derived from an earlier pilot study of 25 patients, of whom 52% required escalated care and the estimated AUC was 0.86 (95% CI: 0.703-0.999). A sample size of 80 patients using a two-sided z-test at a significance level of 0.05,

achieved 90% power to detect an AUC of at least 0.7.

  1. Results

During the study period, 103 patients were enrolled. The median (min-max) age was 11 (0.53-213) months. The baseline characteristics of the study population are shown in Table 1. The median time elapsed from arrival to lung US was 40 (0-120) min. Lung US was generally well tolerated and lasted 9.8 +- 1.9 min, on average (+-SD). The mean duration of symptoms before admission to the PED was 3.59 +- 2.3 d.

The discharge diagnoses of our patients were wheezing (33%), bron- chiolitis (27%), virus-positive pneumonia (10%), presumed or con- firmed Bacterial pneumonia (6%), asthma (9%), and miscellaneous (16%). Nasopharyngeal swabs for respiratory viruses were collected from 64 patients, and 52 (81%) were positive for at least one virus. Two patients had Staphylococcus aureus-positive blood cultures and ex- tensive consolidation on chest radiography. The chronic conditions were as follows: mild neurological disease (n = 14), asthma (n = 11), prematurity without bronchopulmonary dysplasia (n = 5), mild con- genital heart defect (n = 4), liver disease (n = 3), non-metastatic can- cer (n = 3), endocrine disease (n = 2), hemangioma (n = 1), kidney disease (n = 2), congenital cytomegalovirus infection (n = 2), immu- nodeficiency (n = 1), spherocytosis (n = 1), gastroschysis (n = 1), repaired esophageal atresia (n = 1), and genetic syndrome (n = 1).

Thirty-five patients (34%) required escalated care. Thirteen of them required only HFNC, five needed NIV, and five started with HFNC or NIV but had to be switched to mechanical ventilation (Fig. 2). Twelve patients were placed on mechanical ventilation without any prior at- tempts at HFNC or NIV. The average time between arrival and con- nection of HFNC, NIV, or mechanical ventilation was 6.64 +- 5.94 h. Age, altered mental status (agitated, or lethargic), color change (pale, or cyanotic), and CRS were associated with escalated care (p < 0.05).

Patients who required escalated care had a higher lung US score (Fig. 3A): median 13 (0-34) versus 2 (0-21), p < 0.0001. The OR was

1.22 (95% CI: 1.12-1.32) for each additional point in the score. The AUC, assessing the ability of lung US to predict escalated care, showed good accuracy: 0.81 (95% CI:0.71-0.90) (Fig. 3B). ROC analysis revealed that the best cut-off value was seven (sensitivity,71.4%; specific- ity,79.4%). Odds ratio was particularly high in patients whose lung US score was above seven (OR:9.6 [95% CI 3.8-24.7]).

The analysis was completed by searching cut-offs that prioritized a sensitivity or specificity above 90% (Fig. 3C). According to our findings, the lowest score of zero was not highly sensitive and does not exclude the need for escalated care with near-certainty, with a negative LHR of

0.26 (95% CI:0.10-0.68). The final diagnosis for the four patients with zero score who required escalated care were bronchiolitis (2/4), viral wheezing (1/4), and asthma (1/4). A score above 12 was highly specific for predicting escalated care and had a positive LHR of 8.74 (95% CI:3.21-23.86). The inter-observer weighted kappa coefficient was

0.68 (95% CI 0.64-0.72; p < 0.0001).

On multivariate analysis (Supplemental Table S3), higher lung US score and CRS were statistically significantly associated with the need for escalated care, with an adjusted OR of 1.20 (95% CI: 1.07-1.35) and

1.45 (95% CI: 1.02-2.06), respectively.

Table 1

Summary of basic population details.

Variable

Category

Escalated care

p-value

No (68)

Yes (35)

Sex

Male

30 (44.1%)

22 (62.9%)

0.111a

Chronic Disease

38 (55.9%)

14 (40%)

0.187a

Fever

31 (46.3%)

16 (48.5%)

0.999a

Heart rate

Elevated

30 (44.8%)

11 (31.4%)

0.275a

Respiratory rate

Elevated

61 (89.7%)

33 (94.3%)

0.681a

Pulse oximetry

<94%

42 (62.7%)

29 (82.9%)

0.061a

Pulse oximetry

<90%

28 (41.8%)

20 (57.1%)

0.206a

Color

Altered

15 (22.1%)

16 (45.7%)

0.016b

Mental Status

Altered

24 (35.3%)

23 (65.7%)

0.013b

Virus panel collected

39 (57%)

25 (75%)

Virus positive

29/39 (74.4%)

23/25 (92.0%)

Coinfections

9

6

RSV

8

11

Rinovirus

6

1

Sars Cov2

2

5

Coronavirus 229E

1

Parainfluenza

2

Bocavirus

1

Chest Radiography

Altered

31 (60.8%)

26 (81.3%)

0.087a

Age (months)

Median (min- max)

18.0 (0.6-213.5)

4.1 (0.5-143.8)

<0.0001c

Day of symptoms

Median (min-max)

3 (1-10)

3 (1-10)

0.358c

CRS

Median (min-max)

5 (2-9)

7 (2-11)

<0.0001c

a Chi square with continuity correction.

b Fisher’s exact test.

c Mann-Whitney U test.

CRS was positively correlated with the lung US score, with a Spear- man coefficient of 0.33; p < 0.001.

  1. Discussion

Point-of-care lung US has been gaining an increasing role in many clinical assessments and has been shown to offer prognostic informa- tion in patients with acute respiratory distress syndrome [14,15] and newborns with respiratory distress syndrome [17,18]. Here, we demon- strate that it also provides prognostic information in the first assess- ment of children presenting with respiratory distress. We found that an elevated lung US score detected on the initial evaluation was associ- ated with a higher chance of escalated care within the next 24 h. This finding suggests that the lung US score could be used as an additional tool to stratify patients, especially when decisions surrounding the level of admission or early transfer must be promptly made. In many Low- and middle-income countries, emergency departments are fre- quently overwhelmed, resources are limited, and high-dependency beds are scarce, prompting clinicians to prioritize children who are likely to need further support [26].

Clinical assessment remains the cornerstone of risk stratification; however, in children, it may sometimes be challenging. Several com- monly used clinical parameters, including heart rate, respiratory rate, and color change, can vary owing to fever, crying, anxiety, and

Image of Fig. 2

Fig. 2. Study flow of all included patients.

HFNC: high-flow nasal cannula; NIV: noninvasive ventilation; MV: mechanical ventilation.

coexisting non-pulmonary conditions. Furthermore, most of the com- posite Clinical scores proposed are unsuitable for children, require com- plex auscultative skills, or lack proper validation [27,28].

The clinical score used in our study (CRS) was recently tested in pe- diatric patients with any type of respiratory illness and was shown to be predictive of intensive care needs [22]. The authors found that values above 3 had a sensitivity of 94% (95% CI:79.8-99.3) and specificity of 40% (95% CI:35-45), in predicting outcomes. We also found that CRS was independently associated with escalated care, but most of our study population had CRS values >3, including those who did not re- quire escalated care (median CRS of five). In a population like ours, which is composed of children with already concerning Clinical scores, the lung US score was able to provide useful information that may aid clinicians in their decision-making process. Our findings support the concept that point-of-care lung US should be considered an extension of physical examination, with the potential to help overcome some in- herent difficulties of clinical evaluation by providing additional informa- tion using direct pulmonary imaging.

When comparing all available imaging modalities, the US seems to be a better option than chest radiography or chest tomography when- ever possible for several reasons: it is radiation-free, can be performed quickly at the bedside, is cost-effective, and allows periodic assessments [29]. Although chest radiography is the most commonly used imaging modality in the assessment of children with respiratory distress, previ- ous studies have shown that it has poor inter-observer agreement [30] and lacks sensitivity in many clinical scenarios, including viral lower re- spiratory infections [8,31,32] and Acute chest syndrome [33]. In children with bronchiolitis, lung US can identify abnormalities not revealed by chest radiography, and findings typically increase gradually with increasing illness severity [9,34].

In the adult critical care setting, there are already Evidence-based guidelines addressing the use of lung US scores to guide care for restrictive pulmonary disorders [35]. In the pediatric population, the lung US score is gaining recognition, particularly for bronchiolitis [19,36]. A recent study of infants with bronchiolitis showed that the lung US score had a higher AUC than the clinical score for predicting intensive care unit admission [19]. The best cut-off point was six, which was close to the value found in our study. Accordingly, another study found that children with

Image of Fig. 3

Fig. 3. Summary of the main results.

a. Violin plot depicting the density of each variable level; b: receiving operating characteristic curve with an area under the curve of 0.81. c: inconclusive limits of lung ultrasound score (dashed area) determined by cut-offs of sensibility or specificity above 90%.

bronchiolitis with higher lung US scores had a higher likelihood of requir- ing respiratory support (p = 0.001) [36], but the score used was slightly different from the one used in our study, giving 1 point to confluent lines and 2 points to subpleural consolidations >1 cm.

The current study adds to the validity of previous work on the lung US score by documenting its performance in a broader population of children with respiratory distress of different etiologies. Our patient population was composed of children with various types of pulmonary diseases, including restrictive, mixed, or purely obstructive. It is well known that lung US detects several abnormalities in restrictive disor- ders, but it should be normal in purely obstructive disorders regardless of severity because it cannot detect over-distension or air-trapping [37,38]. Despite the fact that our study included a mixed pediatric pop- ulation with respiratory distress, and the A-line pattern commonly

observed in obstructive pulmonary disease would correlate with lower lung US scores, we could still find a link between higher lung US scores and the need for escalated care. This suggests that disease se- verity might be related to overlapping conditions, such as interstitial in- volvement, consolidations, and atelectasis. In fact, in a study that included children aged 2-17 years with asthma exacerbations, positive lung ultrasound findings were found in 45% of patients and correlated with an increased need for hospital admission and resources [39].

As a triage tool for respiratory distress arising from several etiologies in PED, a lower lung US score indicated reduced chance of requiring es- calated care but did not exclude the outcome with near-certainty. For example, a low US score should not be reassuring for patients whose re- spiratory distress is primarily caused by obstructive disease. Conversely, lung US score > 12 had an elevated specificity and good positive

Image of Fig. 4

Fig. 4. Lung ultrasound (US) images from a patient case.

A previously healthy 49-day-old infant presented with flu-like symptoms and respiratory distress with an initial clinical respiratory score (CRS) of six. Lung US showing consolidation in the right posterior lung field (A), confluent B-lines in the right lateral and posterior lung fields bilaterally (B, C, D), and well-spaced B-lines in the anterior lung fields, with a total score of 14. Twelve hours after admission, the patient was placed on high flow nasal cannula , but on the following day, he needed mechanical ventilation. First-day chest radiography was unremarkable (E), but bilateral infiltrates and upper right lobe atelectasis were observed on the seventh day. He tested positive for the respiratory syncytial virus.

likelihood ratio. Patients with a lung US score above this threshold should be given absolute priority and promptly transferred to an appro- priate facility or hospital unit capable of providing strict monitoring, and with teams skilled in airway support. A cut-off value of seven has an el- evated OR and the best combined sensitivity and specificity, but it falls into the gray zone. This can be useful for clinical purposes if integrated with additional tools to build predictive models.

We performed an exploratory analysis of the correlation between the lung US score and CRS. CRS positively correlated with the lung US score, with a Spearman coefficient of 0.33; p < 0.001. However, this cor- relation could not be considered strong. Many patients had elevated CRS upon arrival but improved after minor interventions, such as oxygen supplementation and/or administration of bronchodilators/corticoste- roids. It is possible that patients with low lung US scores would respond better to bronchodilators/corticosteroids, as they have a more purely obstructive disease and, thus, a more reversible disease. Nevertheless, we did not collect data on the treatment received or CRS changes after treatment to confirm this assumption. In adult patients, there is evi- dence supporting the view that patients with respiratory failure and normal lung ultrasound findings should be treated for asthma or chronic obstructive pulmonary disease [38]. However, young children frequently experience viral wheezing with varying degrees of intersti- tial involvement and varying degrees of responses to bronchodilators/ corticosteroids. To date, no study has addressed whether young chil- dren with viral wheezing and normal ultrasound findings would re- spond better to first-line asthma treatment. On the other hand, nine out of 35 patients who needed escalated care had an ultrasound score equal or above 12, but a mild (0-3) or moderated (4-7) CRS upon ar- rival to PED. In these cases, lung ultrasonography revealed the degree of lung compromise before clinical deterioration. One of such cases is depicted on Fig. 4.

Our study has some limitations that must be acknowledged. We used a convenience sample, which is known to introduce a selection bias in emergency department studies. However, in the present study, the researchers worked both on day and night shifts, and both during week and weekend periods in a variable-week schedule. Therefore, we believe that our sample is representative of the entire cohort. Given the heterogeneous nature of the population, we had a relatively small sample size. Nonetheless, lung US has already been successfully correlated with outcomes in disparate lung diseases; therefore, our findings were consistent with that of previous research and may be re- producible in larger populations. Furthermore, the indication for esca- lated care, which is our main outcome, is not internationally standardized and depends on the clinician’s perception of severity and local resource availability. We attempted to overcome this limitation by collecting data from three different hospitals with varying resources, staffing, and practice standards. Another limitation was that no formal training or inter-observer agreement was made for the CRS. Finally, blinding the clinical conditions of the physicians performing ultrasound is not feasible. Nonetheless, lung US image interpretation was based on objective signs, and a significant [40] agreement was obtained by a Blinded physician.

  1. Conclusion

An elevated lung US score measured in the first assessment of chil- dren with respiratory distress predicts severity as defined by the need for HFNC, VNI, or mechanical ventilation. Our study provides support for further research, ideally combining clinical and imaging data, to identify children with respiratory distress who are at the highest risk of clinical deterioration.

Financial disclosure

The authors have no financial interests to declare relating to this article.

Funding source

This research did not receive any specific grant from funding agen- cies in the public, commercial, or not-for-profit sectors.

CRediT authorship contribution statement

Eliana P.C. Giorno: Conceptualization, Writing – original draft, Methodology, Investigation, Project Administration, Writing – review & editing. Flavia K. Foronda: Writing – review & editing, Writing – original draft, Investigation. Milena De Paulis: Writing – review & editing, Investigation, Data curation. Danielle S.N. Bou Ghosn: Writing

– original draft, Investigation. Thomaz B. Couto: Investigation. Fernanda V.M. Sa: Investigation. Andrea M.A. Fraga: Supervision. Sylvia C.L. Farhat: Supervision, Project administration, Methodology, Conceptualization. Marcela Preto-Zamperlini: Writing – original draft, Supervision, Methodology, Investigation, Conceptualization. Claudio Schvartsman: Writing – original draft, Supervision, Methodol- ogy, Conceptualization.

Declaration of Competing Interest

The authors have no conflicts of interest relevant to this article to disclose.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2023.02.035.

References

  1. Forum of International Respiratory Societies. The global impact of respiratory dis- ease. 2nd ed. Sheffield: European Respiratory Society; 2017.
  2. Zar HJ, Ferkol TW. The global burden of respiratory disease–impact on child health. Pediatr Pulmonol. 2014 May;49(5):430-4. https://doi.org/10.1002/ppul.23030.
  3. McDermott KW, Stocks C, Freeman WJ. Overview of Pediatric Emergency Depart- ment Visits, 2015: Statistical Brief #242. 2018 Aug 7. Healthcare cost and utilization project (HCUP) statistical briefs [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2006 Feb Available from: https://www.ncbi.nlm.nih. gov/books/NBK526418/.
  4. Damore D, Mansbach JM, Clark S, Ramundo M, Camargo Jr CA. Prospective multicen- ter bronchiolitis study: predicting intensive care unit admissions. Acad Emerg Med. 2008 Oct;15(10):887-94.
  5. Freire G, Kuppermann N, Zemek R, Plint AC, Babl FE, Dalziel SR, et al. Predicting es- calated care in infants with bronchiolitis. Pediatrics. 2018 Sep 1;142(3).
  6. Hedstrom AB, Andre-von AO Von Saint, Grassia KL, Nielsen KR. Markers of pediatric respiratory distress predictive of poor outcome in low-and middle-income coun- tries: a systematic review. J Glob Health Rep. 2020 Aug 10;4:e2020072.
  7. McClain L, Hall M, Shah SS, Tieder JS, Myers AL, Auger K, et al. Admission chest radio- graphs predict illness severity for children hospitalized with pneumonia. J Hosp Med. 2014 Sep;9(9):559-64. https://doi.org/10.1002/jhm.2227.
  8. Caiulo VA, Gargani L, Caiulo S, Fisicaro A, Moramarco F, Latini G, et al. Lung ultra- sound in bronchiolitis: comparison with chest X-ray. Eur J Pediatr. 2011 Nov;170 (11):1427-33.
  9. Kharasch S, Duggan NM, Cohen AR, Shokoohi H. Lung ultrasound in children with re- spiratory tract infections: viral, bacterial or COVID-19? A narrative review. Open Ac- cess Emerg Med. 2020;12:275. https://doi.org/10.2147/OAEM.S238702.
  10. Pereda MA, Chavez MA, Hooper-Miele CC, Gilman RH, Steinhoff MC, Ellington LE, et al. Lung ultrasound for the diagnosis of pneumonia in children: a meta-analysis. Pediatrics. 2015 Apr;135(4):714-22. https://doi.org/10.1542/peds.2014-2833.
  11. Lichtenstein DA, Meziere G, Lascols N, Biderman P, Courret JP, Gepner A, et al. Ultra- sound diagnosis of Occult pneumothorax. Crit Care Med. 2005 Jun 1;33(6):1231-8. https://doi.org/10.1097/01.CCM.0000164542.86954.B4.
  12. Yousefifard M, Baikpour M, Ghelichkhani P, Asady H, Nia KS, Jafari AM, et al. Screen- ing performance characteristic of ultrasonography and radiography in detection of pleural effusion; a meta-analysis. Emergency. 2016;4(1):1.
  13. Bouhemad B, Mongodi S, Via G, Rouquette I. Ultrasound for “lung monitoring” of ventilated patients. Anesthesiology. 2015 Feb;122(2):437-47. https://doi.org/10. 1097/ALN.0000000000000558.
  14. Bouhemad B, Liu ZH, Arbelot C, Zhang M, Ferarri F, Le-Guen M, et al. Ultrasound as- sessment of antibiotic-induced pulmonary reaeration in ventilator-associated pneu- monia. Crit Care Med. 2010 Jan 1;38(1):84-92. https://doi.org/10.1097/CCM. 0b013e3181b08cdb.
  15. Bello G, Blanco P. Lung ultrasonography for assessing lung aeration in acute respira- tory distress syndrome: a narrative review. J Ultrasound Med. 2019 Jan;38(1): 27-37. https://doi.org/10.1002/jum.14671.
  16. Soummer A, Perbet S, Brisson H, Arbelot C, Constantin JM, Lu Q, et al. Ultrasound as- sessment of lung aeration loss during a successful weaning trial predicts postextubation distress. Crit Care Med. 2012 Jul 1;40(7):2064-72. https://doi.org/ 10.1097/CCM.0b013e31824e68ae.
  17. De Martino L, Yousef N, Ben-Ammar R, Raimondi F, Shankar-Aguilera S, De Luca D. Lung ultrasound score predicts surfactant need in extremely preterm neonates. Pediatrics. 2018 Sep 1;142(3).
  18. Brat R, Yousef N, Klifa R, Reynaud S, Aguilera SS, De Luca D. Lung ultrasonography score to evaluate oxygenation and surfactant need in neonates treated with contin- uous positive airway pressure. JAMA Pediatr. 2015 Aug 1;169(8):e151797.
  19. Bobillo-Perez S, Sorribes C, Gebelli P, Lledo N, Castilla M, Ramon M, et al. Lung ultra- sound to predict pediatric intensive care admission in infants with bronchiolitis (LUSBRO study). Eur J Pediatr. 2021 Jul;180(7):2065-72.
  20. Nitu ME, Eigan H. Respiratory failure. Pediatr Rev. 2009;30(12):470-8.
  21. Pediatric advanced life support: provider manual. In: Chameides Leon, Ralston Mark, editors. PALS, American Academy of Pediatrics, American Heart Association. Dallas, TX: American Heart Association; 2011 c2011.NLM ID:101589947 [book].
  22. Nayani K, Naeem R, Munir O, Naseer N, Feroze A, Brown N, et al. The clinical respi- ratory score predicts paediatric critical care disposition in children with respiratory distress presenting to the emergency department. BMC Pediatr. 2018 Dec;18(1): 1-8. https://doi.org/10.1186/s12887-018-1317-2.
  23. Copetti R, Cattarossi L. ultrasound diagnosis of pneumonia in children. Radiol Med. 2008 Mar;113(2):190-8. https://doi.org/10.1007/s1154 7-008-0247-8.
  24. Willson DF, Horn SD, Hendley JO, Smout R, Gassaway J. Effect of practice variation on resource utilization in infants hospitalized for viral lower respiratory illness. Pediat- rics. 2001 Oct 1;108(4):851-5.
  25. Fedor KL. Noninvasive respiratory support in infants and children. Respir Care. 2017 Jun 1;62(6):699-717. https://doi.org/10.4187/respcare.05244.
  26. Emukule GO, McMorrow M, Ulloa C, Khagayi S, Njuguna HN, Burton D, et al. Predicting mortality among hospitalized children with respiratory illness in West- ern Kenya, 2009-2012. PloS One. 2014 Mar 25;9(3):e92968. https://doi.org/10. 1371/journal.pone.0092968.
  27. Eggink H, Brand P, Reimink R, Bekhof J. Clinical scores for dyspnoea severity in chil- dren: a prospective validation study. PloS One. 2016 Jul 6;11(7) e0157724.2016.
  28. Bekhof J, Reimink R, Brand PL. Systematic review: insufficient validation of clinical scores for the assessment of acute dyspnoea in wheezing children. Paediatr Respir Rev. 2014 Mar 1;15(1):98-112. https://doi.org/10.1016/j.prrv.2013.08.004.
  29. Gargani L, Volpicelli G. How I do it: lung ultrasound. Cardiovasc Ultrasound. 2014 Dec;12(1). https://doi.org/10.1186/1476-7120-12-25. 1-0.
  30. Voigt GM, Thiele D, Wetzke M, Weidemann J, Parpatt PM, Welte T, et al. Interob- server agreement in interpretation of chest radiographs for pediatric community ac- quired pneumonia: findings of the pedCAPNETZ-cohort. Pediatr Pulmonol. 2021 Aug;56(8):2676-85. https://doi.org/10.1002/ppul.25528.
  31. Testa A, Soldati G, Copetti R, Giannuzzi R, Portale G, Gentiloni-Silveri N. Early recog- nition of the 2009 pandemic influenza a (H1N1) pneumonia by chest ultrasound. Crit Care. 2012 Feb;16(1):1-8. https://doi.org/10.1186/cc11201.
  32. Heussel CP, Kauczor HU, Heussel G, Fischer B, Mildenberger P, Thelen M. Early detec- tion of pneumonia in febrile neutropenic patients: use of thin-section CT. AJR Am J Roentgenol. 1997 Nov;169(5):1347-53.
  33. Preto-Zamperlini M, Giorno EP, Bou Ghosn DS, Sa FV, Suzuki AS, Suzuki L, et al. Point-of-care lung ultrasound is more reliable than chest X-ray for ruling out acute chest syndrome in sickle cell pediatric patients: a prospective study. Pediatr Blood Cancer. 2022 May;69(5):e29283.
  34. Basile V, Di Mauro A, Scalini E, Comes P, Lofu I, Mostert M, et al. Lung ultrasound: a useful tool in diagnosis and management of bronchiolitis. BMC Pediatr. 2015 Dec;15

(1):1-8. https://doi.org/10.1186/s12887-015-0380-1.

  1. Volpicelli G, Elbarbary M, Blaivas M, Lichtenstein DA, Mathis G, Kirkpatrick AW, et al. International evidence-based recommendations for point-of-care lung ultrasound. Intensive Care Med. 2012 Apr;38(4):577-91.
  2. Supino MC, Buonsenso D, Scateni S, Scialanga B, Mesturino MA, Bock C, et al. Point- of-care lung ultrasound in infants with bronchiolitis in the pediatric emergency de- partment: a prospective study. Eur J Pediatr. 2019 May;178(5):623-32. https://doi. org/10.1007/s00431-019-03335-6. [PMID: 30747262].
  3. Bouhemad B, Brisson H, Le-Guen M, Arbelot C, Lu Q, Rouby JJ. Bedside ultrasound as- sessment of positive end-expiratory pressure-induced lung recruitment. Am J Respir Crit Care Med. 2011 Feb 1;183(3):341-7.
  4. Lichtenstein DA, Meziere GA. Relevance of lung ultrasound in the diagnosis of acute respiratory failure*: the BLUE protocol. Chest. 2008 Jul 1;134(1):117-25.
  5. Dankoff S, Li P, Shapiro AJ, Varshney T, Dubrovsky AS. Point of care lung ultrasound of children with acute asthma exacerbations in the pediatric ED. Am J Emerg Med. 2017 Apr 1;35(4):615-22.
  6. Viera AJ, Garrett JM. Understanding Interobserver agreement: the kappa statistic. Fam Med. 2005 May 1;37(5):360-3.