Cardio-thoracic imaging and COVID-19 in the pediatric population: A narrative review

Paolo Ferrero, Isabelle Piazza

Paolo Ferrero, ACHD Unit-Pediatric and Adult Congenital Heart Centre, IRCCS-Policlinico San Donato, San Donato Milanese 20097, Milan, Italy

Isabelle Piazza, Department of Emergency Medicine, ASST Papa Giovanni XXIII, Bergamo 24127, Italy

Abstract Worldwide experience about coronavirus disease 2019 (COVID-19) pandemics suggests that symptomatic disease is significantly less frequent in the pediatric age range. Nevertheless, multi-system inflammatory syndrome has been consistently reported in children and has been associated with severe acute respiratory syndrome coronavirus 2 exposure. In this paper we give an overview of the multimodality chest imaging of pediatric patients with suspected COVID-19,focusing on relevant differences with adults.

Key Words: COVID-19; Radiology; Imaging; Chest; Pediatric; SARS-CoV-2

INTRODUCTION

Coronavirus disease 2019 (COVID-19) in its most frequent clinical manifestation causes a respiratory syndrome, due to a single stranded RNA beta coronavirus infection, that may results in acute respiratory distress syndrome. Due to the clinical association of this new virus with the onset of a respiratory syndrome it has been named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). However, it has been shown that many organs and tissues, including heart, vessels, and brain may be involved in SARS-CoV-2. Furthermore COVID-19 has been repeatedly reported in the pediatric population. Among the 962 million people infected worldwide so far, the pediatric age group constitute less than 2%. Overall atypical and asymptomatic infections are more frequent in children. This might be due to a lower exposure or less susceptible lung barrier due to immature angiotensin enzyme-2 receptors.

Likewise in adults, COVID-19 in infants is characterized by a background of inflammatory activation that may involve different tissues and organs in different phases. In those area particularly hit by the virus during the first pandemic phase, we observed a surge of mucocutaneous inflammatory syndrome resembling Kawasaki disease. This condition has been subsequently named multisystem inflammatory syndrome in children (MIS-C) associated with COVID-19.

The proteus clinical presentation of COVID-19 in children together with the possible occurrence of MIS-C prompt a different diagnostic approach in this age group as compared to adults. (Table 1) Differently from the general population data about image findings in pediatric COVID-19 are scant and sometimes conflictual.

We aim to provide an overview of peculiarity of thoracic diagnostic algorithm in the pediatric age group, focusing relevant differences with adults.

METHODS

For this narrative review the following keywords: COVID-19, SARS-CoV-2, radiology,imaging and pediatric in different combinations. Papers were screened accordingly to the information provided in the abstract. Only manuscripts written in English and focused on the thoracic district were deemed eligible for inclusion. Within the pediatric age group we did not consider any specific limit of range and all papers dealing with patients younger than 18 years were included point of view papers and editorials were also excluded. Additional data were retrieved from the references of individual papers, whenever appropriate.

Pathophysiology

Several pathophysiologic issues have been advocated to explain differences in clinical presentation of pediatric patients. In particular, an association between SARS-CoV-2 infection and a multi systemic inflammatory syndrome has been consistently reported. Diagnostic criteria of MIS-C include fever, hypotension, evidence of cardiac or other end-organ injury together with at least two of the following: maculopapular rash, non purulent conjunctivitis, mucocutaneous inflammation and gastrointestinal symptoms.

This particular presentation, characterized by multi organ inflammation rather than severe respiratory syndrome, may be due to a different immune system reactivity and a more immature and different distribution of angiotensin-converting enzyme II receptor, which is the entry receptor for the virus. The association of syndromes and the degree of lung maturity can further contribute to modulate the clinical presentation. As far as the thoracic involvement is concerned, coronary involvement is a rare peculiar feature of MIS-C sharing the clinical and anatomic presentation with Kawasaki disease.

Characteristics of chest X-ray changes in pediatric COVID-19

Although the concept that radiological manifestations of COVID-19 vary among different age group, information about sensitivity, specificity and characteristic findings of chest X-ray in the pediatric population are limited and conflicting.

Chest X-ray may be negative in more than 1/3 of patients in the pediatric group age.Reported sensitivity of chest X-ray vary from 25% to 69%.

Commonly encountered chest X-ray abnormalities are consolidations and ground glass opacities (GGO), differently from the adult variant, peribronchial thickening is more frequent in this clinical setting (60% to 80%). Peribronchial thickening is however aspecific and can be seen in other variant of viral pneumonia in children,while hyperinflation, which is another recognized hallmark, was not reported. Reason for chest X-ray low sensitivity is the higher prevalence of lower density, smaller size and basal opacities, obscured by the diaphragm and hepatic dome.

Table 1 Differences of coronavirus disease 2019 relevant characteristics between adult and pediatric population

Halo sign, which is deemed to be highly specific of COVID-19, has been observed also in 50% of pediatric cases in a published series, while other finding such as crazy paving pattern or organizing pneumonia pattern, which have been reported as typical in adult with COVID-19, were not consistently reported in pediatric series.

Finally, pleural effusion is a rare manifestation of COVID-19 in adults and is almost uniquely seen in the pediatric age range being associated with involvement of other serous cavities in patients with MIS-C.

Computer tomography

Although computer tomography (CT) has a much higher sensitivity likewise chest Xray, according to some reports, pediatric patients are three time more likely to have normal CT scan as compared to adult. However, pooling together literature data,percentage of negative CT is very variable, ranging between 10% and 30%.

In adults, several series have found that COVID-19 typically presents with peripheral and posterior GGO. Opacities may present in a confluent fashion or as more delimited scattered round opacities (so called ‘crazy paving patter’). Other peculiar findings typical of organizing pneumonia are ‘halo sign’ or ‘reversal halo sign’.

It has been hypothesized that the predominant localization of radiological findings at the periphery of the lung and the presence of enlargement of the vessel feeding the involved lung area (feeding vessel sign) may be explained by the inflammatory involvement of small vessels. Based on these pathogenetic models, standardized reporting methodologies have been devised providing also a ranking of suspicious according to the presence of cluster of signs. However application of diagnostic CT scores based on adult cohorts resulted in low probability of the disease in the pediatric population. According to this background, utilization of low dose CT as screening tool in children with laboratory inconclusive findings is controversial.

As compared to adults, pediatric series show a lower total number of pulmonary lesions and smaller size of them. Bilateral GGO confirmed to be the most common finding accounting for almost 90% of the positive scans. Likewise general population GGO may present different stage of consolidation (halo sign) in 50% of cases. As previously discussed about chest X-ray, peribronchial and bronco-vascular thickening are more common in children.

Role of thoracic ultrasounds

Lung ultrasounds (LUS) is consistently used in the diagnostic process of different lung diseases in both adults and children.

Among patients with suspected or ascertained COVID-19, elementary ultrasonographic findings in the context of lung injury are: Normal lung sliding, B lines, that are perpendicular hyperechogenic streaks due to loss of aerated space, subpleural nodules, pleural effusion. Clusters of these signs have been described in association with different probability and severity of the disease. Evidences about diagnostic accuracy in patient with COVID-19 in the pediatric age group is limited to small series.According to these data LUS, although less sensitive than CT, is more sensitive than chest X-ray. In the largest multi-center cohort of 40 children with suspected COVID-19 undergoing LUS, the diagnosis of pulmonary involvement was done in 10 out of 12 patients with positive CT findings, in seven of whom chest X-ray was normal.

Common findings were: A line in 72%, various pattern of B line in 27%, while parenchymal nodular consolidation were more rare as compared with adults (10%).

As previously mentioned, clusters of inflammatory syndrome associated with SARS-CoV-2 (MIS-C) have been consistently reported. Even though these patients usually present without respiratory symptoms, LUS showed loss of aeration or pleural effusion in all of them in a small case series.

Despite these limited data, use of LUS in this clinical setting might be advocated since it allow early and repeated bedside assessment avoiding X-ray exposure and patient transport in the imaging department, which might potentially increase the risk of virus spread.

Additional specific image modalities

Coagulation disorders have been recognized as a major complication of COVID-19 significantly affecting the prognosis. In this context the ventilation/perfusion single photon emission computed tomography, either as single modality or combined with computed tomography (V/Q SPECT) can be used in selected case to diagnose pulmonary embolism in case of iodinated medium allergy or to integrate CT images.In the pediatric population clinically relevant pulmonary embolism are rarely reported therefore the need to exclude this complication is far less compelling. Furthermore radionuclide exposure and the risk of infection spread across different departments,deeply limits the room for this diagnostic resource in the clinical practice.

Multimodal cardiovascular imaging

Since the first COVID-19 outbreak, different degree of myocardial injury have been reported. Biomarker evidence of myocardial injury has been associated with a higher mortality risk in COVID-19 patients. Although typical clinical presentation of COVID-19 in the pediatric age group is rare, association of MIS-C with SARS-CoV-2 exposure has been reproducibly observed. This syndrome, originally labelled ad Kawasaky-like, is characterized by various degree of myocardial and coronary inflammatory involvement. Even-though clinical manifestation of this syndrome is extremely variable abdominal and gastrointestinal symptoms are key diagnostic features, being present in up to 80%, while cardiovascular involvement my go initially unrecognized. Although abdominal imaging is not the focus of this review, it has to be highlighted that ultrasound of this district is usually the first diagnostic exam performed, furthermore cardiovascular involvement and abdominal imaging are strictly related. Abdominal anechoic space and hepatomegaly are the most frequent findings. In one study screening abdominal echo was able to disclose associated pleuro-pericardial effusion and cardiomegalia in almost 37% and 12% of case,respectively. From the practical point of view, given the high incidence of gastrointestinal symptoms in children with COVID-19 related MIS-C, abdominal screening echo may help in differentiate this condition from other abdominal urgencies.

Trans-thoracic echo has a high diagnostic sensitivity in the acute phase by demonstrating ventricular dysfunction and coronary remodeling. Consistently with echocardiografic diagnostic criteria in typical complete Kawasaky syndrome, coronary size is standardized according to the deviation from the median in the general population with the same body surface area (z score). Cardiac depression, although sometimes severe at presentation, requiring inotropic or even mechanical circulatory support,recovers in around 70% of cases. Nevertheless, given the limited knowledge about the virus pathogenicity and natural history of the disease in the long term, echocardiographic follow up at one month and one year, in patients with documented cardiac or coronary involvement (z score ＞ 25), is recommended.

Cardiac CT scan can provide accurate evaluation of the coronary artery anatomy and may be considered in patients with difficult acoustic windows or with extensive coronary involvement. Cardiac magnetic resonance imaging may be useful during the initial hospitalization or approximately 3 mo post-acute illness to evaluate ventricular function and myocardial characteristics including edema, diffuse fibrosis, and scar by myocardial late gadolinium enhancement.

Differences of COVID-19 crucial characteristics between adult and children

Table 1 summarizes main differences in COVID-19 that are relevant in planning the diagnostic algorithm. We considered the following nosological variables: Epidemiology, pathophysiology, natural history, image sensitivity.

So far about 96 million COVID-19 cases have been reported, of whom only less than 2% occurred in patients less than 18 years old. Furthermore, as compared with adult cohorts, fewer patients have a severe or critical course (6%20%).

Distribution of rate of hospitalization displays a cluster in the age group lower than two years and higher than 10 d. Although infants younger than two years rarely present pneumonia, admission is motivated by poor tolerance of fever. As far as the diagnostic algorithm is concerned, although lung involvement in SARS-CoV-2 infection in the pediatric age range may largely reproduce those reported in the general population, sensitivity of chest X-ray is significantly lower. As a consequence tradi-tional chest imaging is a poor screening test in children with suspected COVID-19. On the other hand, fast bedside ultrasound screening has a high diagnostic yield as may easily disclose multi organ involvement consistent with the infection.

CONCLUSION

COVID-19 pandemics has prompted worldwide rapid reorganization of imaging departments. While the incidence of clinically relevant COVID-19 in pediatric population was previously deemed very low, the observation of a late peak of SARSCoV-2 related disease in this age group prompted the development of specific management algorithms (Figure 1).

Knowledge about imaging diagnostic findings has significant grown, however significant differences in the pathophysiology and clinical presentation in children as compared with adults must be taken into account. Observational data indicate that both chest X-ray and CT have a lower diagnostic yield and can show peculiar findings,such as broncovascular thickening and pleuro-pericardial effusion, in pediatric patients. The thoracic echo may have a relevant role in the diagnostic algorithm in order to screen and monitor lung involvement as well as specific features of MIS-C in this clinical setting. Furthermore, X-ray exposure and risk of virus spread during patient transport should be taken into account when considering repeated traditional imaging.

This review is based mainly on small case series with heterogeneous populations and sometimes contradictory conclusion about findings and appropriateness of the various diagnostic tools. This limitation does not allow to pool together the data and to provide general recommendation with a sufficient grade of evidence.

Figure 1 Proposed multimodel diagnostic algorithm. COVID-19: Coronavirus disease 2019; CT: Computer tomography.