DISCUSSION
To our knowledge, this is the first multicenter national study assessing LUS findings and their prognostic role in a relatively large cohort of children with microbiologically confirmed SARS-CoV-2 infection. Overall, we found that children with higher LUS scores and with subpleural consolidations have a significantly higher risk of being hospitalized or require oxygen support after initial assessment in the PED.
Our study confirms, on a larger populations, what initially suggested by smaller pediatric cohorts from Italy, Turkey and Spain, showing that children with LRTI during SARS-CoV-2 infection can have a cohort of LUS findings such as vertical artifacts and subpleural consolidations.13-22 Authors from four different hospitals in Italy, in particular, found vertical artifacts and subpleural consolidations to be the most common findings, while pleural effusions were more rare and more described in children with MIS-C. However, cohorts were mostly limited to less than 50 patients. In general, these LUS artifacts are in line with what we have learnt from the past decade of LUS practice in pediatrics. Although the initial role of LUS has mostly been detecting pneumonia in children, its role has significantly evolved. 8 The better understanding of different semeiotic LUS patterns and of their physical mechanismsled authors to investigate if specific LUS patterns may better discriminate different lung conditions. 23 For example, pediatricians from Rome found LUS patterns (like large consolidations, complicated effusions, fix or liquid bronchograms) as more predictive of bacterial or more severe pneumonia. 24, 25Similarly, two independent teams investigated if some LUS patterns may be more associated to viral or bacterial pneumonia, both finding that small subpleural consolidations and vertical artifacts are more frequent in viral LRTIs, while large consolidations with bronchograms more in bacterial etiologies. 26, 27 Therefore, our findings that vertical artifacts and small subpleural consolidations are more frequent in Covid-19 pneumonia is in line with what expected from previous LUS literature in adults with Covid-19 or children with other viral conditions. 8
Our multicenter study allowed us to include a larger number of patients and provide information about prediction of severity, but also subanalyses according to age groups. Unsurprisingly, our study found that children with higher LUS scores of subpleural consolidations had a higher risks of needing hospitalization or oxygen support. These findings are in line with a few pediatric studies which included very small number of children with severe disease, but also confirm studies from adults with Covid-19.13-19 In adults, several studies have documented how LUS performed in the ED can predict hospitalization, ventilation support and deaths.10-12Our findings are not unexpected in light of recent understanding of the physical bases of LUS. 23 In particular, there is growing agreement in literature that vertical artifacts represent peripheral lung abnormalities that generate acoustic traps, eventually seen as vertical lines on LUS. As these abnormalities represent areas of dysventilation and possibly altered gas exchange, it is not unsurprising that those children with more abnormalities on LUS may have a higher risk of developing more severe disease. Similar evidence is already available from other pediatric respiratory conditions like acute bronchiolitis, where several studies have documented that children with higher LUS score have a higher risk of hospitalization, respiratory support and intensive care unit admission.28
In our study, we performed subanalyses according to age groups. In general, we found that lung involvement was more significant in children older than 1 year of age, which is in line with a well-established although not yet fully understood gradient of more severe disease according to increasing age.1 Some authors have suggested that better innate immunity in the upper airways might have contributed to lesser degree of LRTIs in children. 29In our cohort, children younger than 1 year of age had, in fact, less frequently vertical artifacts and subpleural consolidations, supporting this hypothesis. However, this age groups may also have been protected by maternal antibodies, since maternal vaccinations have started during the study period, although we did not collect this information.30 Conversely, when we divided our cohorts in children younger than 5 years, 5-11 and older than 12 years of age (which reflects the different age groups that have had access to vaccination), the characteristics of LUS patterns were similar, suggesting that children younger than 5 years of age have a similar rate of LRTI involvement than older one. These data may have clinical implications, as can provide further information to both healthcare workers and parents about the decision of vaccinating or not younger children, a still debated topic.31
Our study has limitations to address. The most important limit is the low number of children with critical Covid-19 that required mechanical ventilation. However, such a severe outcome is very rare and would require significantly larger populations, a limit difficult to overcome, even with multicenter studies. Also, our study did not include populations at higher risk of more severe Covid-19, including children with comorbidities, black and latino communities, therefore our findings may not be generalized to different epidemiological contexts. Also, these data refer to pre-omicro era, and therefore more studies are needed understand the impact of LUS in these cohorts. Last, we did not included a cohort of vaccinated children, therefore we have not been able to evaluate the impact of vaccination on the development of LRTI during Covid-19 in children.
In conclusion, this national study on children with Covid-19 confirmed that LUS is able to detect Covid-19 low respiratory tract involvement, which is characterized by pleural line irregularities, vertical artifacts and subpleural consolidations. Importantly, children with higher LUS score have a higher risk of required hospitalization or oxygen support, further supporting LUS a valid and safe point-of-care first level tool for the assessment of children with Covid-19. Further studies will be needed to understand how vaccinations and new variants may determine a different degree of LRTI in children with Covid-19.