key: cord-0826102-gq8le026 authors: Das, Karuna M.; Alkoteesh, Jamal A.; Al Kaabi, Jumaa; Al Mansoori, Taleb; Winant, Abbey J.; Singh, Rajvir; Paraswani, Rajesh; Syed, Rizwan; Sharif, Elsadeg M.; Balhaj, Ghazala B.; Lee, Edward Y. title: Comparison of chest radiography and chest CT for evaluation of pediatric COVID‐19 pneumonia: Does CT add diagnostic value? date: 2021-02-25 journal: Pediatr Pulmonol DOI: 10.1002/ppul.25313 sha: d4cf2752eb8ed447b8057361b56f7d7f61dd87f0 doc_id: 826102 cord_uid: gq8le026 PURPOSE: First, to investigate the added diagnostic value of chest computed tomography (CT) for evaluating COVID‐19 in symptomatic children by comparing chest CT findings with chest radiographic findings, and second, to identify the imaging signs and patterns on CT associated with COVID‐19 pneumonia in children. MATERIALS AND METHODS: From March 2020 to December 2020, 56 consecutive children (33 males and 23 girls; mean age ± SD, 14.8 ± 5.0 years; range, 9 months–18 years) with mild to moderate symptom and laboratory confirmed COVID‐19 (based on Centers for Disease Control criteria) underwent both chest radiography and chest CT on the same day within the first 2 days of initial presentation to the hospital. Two experienced radiologists independently evaluated chest radiographs and chest CT studies for thoracic abnormalities. The findings from chest radiography and chest CT were compared to evaluate the added diagnostic value of chest CT for affecting patient management. Interobserver agreement was measured with Cohen's κ statistics. RESULTS: Eleven (19.6%) of 56 patients had abnormal chest radiographic findings, including ground‐glass opacity (GGO) in 5/11 (45.4%) and combined GGO and consolidation in 6/11 (54.5%). On chest CT, 26 (46.4%) of 56 patients had abnormal CT findings, including combined GGO and consolidation in 19/26 (73.1%), GGO in 6/26 (23.1%), and consolidation in 1/26 (3.8%). Chest CT detected all thoracic abnormalities seen on chest radiography in 11/26 (42.3%) cases. In 15/26 (57.7%), chest CT detected lung abnormalities that were not observed on chest radiography, which included GGO and consolidation in 9/15 (60%), GGO in 5/15 (33.3%), and consolidation in 1/15 (6.6%) cases. These additional CT findings did not affect patient management. In addition, chest CT detected radiological signs and patterns, including the halo sign, reversed halo sign, crazy paving pattern, and tree‐in‐bud pattern. There was almost perfect interobserver agreement between the two reviewers for detecting findings on both chest radiographs (κ, 0.89, p = .001) and chest CT (κ, 0.96, p = .001) studies. CONCLUSION: Chest CT detected lung abnormalities, including GGO and/or consolidation, that were not observed on chest radiography in more than half of symptomatic pediatric patients with COVID‐19 pneumonia. However, these additional CT findings did not affect patient management. Therefore, CT is not clinically indicated for the initial evaluation of mild to moderately symptomatic pediatric patients with COVID‐19 pneumonia. Since the first case of coronavirus disease 2019 (COVID-19) was reported in China in November 2019, most COVID-19 cases were epidemiologically linked to exposure to Wuhan's Huanan seafood market, where wild animals are traded. 1, 2 This highly contagious, novel coronavirus infection is rapidly transmitted mainly via droplet inhalation, and has spread extensively among humans across many countries and continents throughout the world. 3 The World Health Organization (WHO) has declared COVID-19 a public health emergency of international concern. As of February 7, 2021, there have been 106,673,989 confirmed cases globally with 2,326,773 deaths. 4 Initially, there was a misconception that COVID-19 spares the pediatric population. However, emerging data have shown that children are indeed affected by this novel viral infection. In fact, there have been several reports describing COVID-19 infection across all pediatric age groups, even affecting neonates. [5] [6] [7] [8] Early and accurate diagnosis of pediatric COVID-19 is paramount for optimal pediatric patient management and limiting the spread of disease, as it has been suggested that asymptomatic or mildly symptomatic children can be a reservoir of unrecognized COVID-19 infection. 9 Chest imaging, mainly chest radiography and CT, are two main imaging modalities currently used for evaluating acute infection in the pediatric population. [10] [11] [12] [13] [14] Although there have been several publications focusing on chest radiographic and CT findings of pediatric COVID-19 pneumonia, [10] [11] [12] [13] [14] [15] [16] to our knowledge, there is no published information regarding the added diagnostic value of chest CT in comparison with chest radiography for detecting thoracic abnormalities related to COVID-19 pneumonia in the pediatric population. In addition, there is limited information regarding the imaging signs and patterns seen on CT in symptomatic pediatric patients with COVID-19 pneumonia. Therefore, the purpose of this study is twofold: (1) to investigate the added diagnostic value of chest CT for evaluating COVID-19 in symptomatic children by comparing chest CT findings with chest radiographic findings and, (2) to identify the imaging signs and patterns on CT associated with COVID-19 pneumonia in children. The institutional review board approved this retrospective study for the review of chest radiographs, chest CT studies, and electronic medical records. The informed consent was waived. Patient confidentiality was maintained in accordance with Health Insurance Portability and Accountability Act (HIPAA) guidelines. Using our radiology department information system in the Al Ain All chest CT studies were obtained with a 64-slice helical CT scanner (Sensation 64; Siemens Healthcare) using a tube kilovoltage (kV), 100-120 kV; tube current (mAs), automatic exposure control; collimation, 2.0 mm; pitch, 1; reconstruction algorithm, iterative-based reconstruction; reconstruction slice thickness, 0.5 mm; interslice gap, 0 mm and reformatted with lung (width, 1500 HU; level, −500 HU) and soft tissue (width, 350 HU; level, 50 HU) window settings. Intravenous contrast was not administered. With the patient in the supine position, CT images were obtained in a single breath-hold at end-inspiration. The scanned area included the entire lungs from the thoracic inlet to the level of the diaphragm. Two radiologists (K. M. D. and J. K.), each with more than 20 years of experience in interpreting pediatric chest radiographs and CT studies, independently evaluated all chest radiographs and CT images. Reviewers were blinded to all clinical information, prior imaging studies, and the original reports of chest radiographs and CT studies. In addition, the order of review was randomized. In the case of discrepant findings by two initial reviewers, a third pediatric radiology fellowship-trained and subspecialty-certified reviewer (E. Y. L., with 20 years of experience in interpreting pediatric chest radiography and CT study) served as an arbitrator, without knowledge of the findings and interpretations of the other two initial reviewers. All chest radiographs and CT studies were reviewed in a picture archiving and communication system (Cedara I-Read 5.2 P11; Cerner Image Devices). The chest CT images were evaluated by using standard lung (width, 1500 HU; level, −500 HU) and soft-tissue (width, 350 HU; level, 50 HU) window settings. For both chest radiographs and CT studies, the reviewers could manually zoom into areas of interest. For chest CT studies, the reviewers routinely used multiplanar reformats (e.g., coronal and sagittal) and manually adjusted window level and width settings for evaluation of thoracic abnormalities. Chest radiography image quality evaluation Chest radiographs were first evaluated for image quality, including anatomic coverage, the presence of motion, and visualization of the chest radiographic findings. Suboptimal image quality chest radiographs, defined as incomplete anatomic coverage of chest, substantial motion, and limited visualization of thoracic anatomy, were excluded. The reviewers then reviewed chest radiographs for abnormalities in the thoracic structures including: (1) lung parenchyma and airway (ground-glass opacity [GGO], consolidation, peribronchial thickening); (2) pleura (pleural effusion), and (3) mediastinum (lymphadenopathy) based on the established criteria from the Fleischner Society's glossary of terms for thoracic imaging in following section. [18] [19] [20] In addition, the reviewers were also instructed to record any other thoracic findings that were not included in the aforementioned diagnostic categories. To decrease potential bias, chest CT evaluation was performed 2 weeks after chest radiographic evaluation, and chest CT studies were also re-randomized before review. Chest CT images were first evaluated for image quality, including anatomic coverage, the presence of motion, and visualization of the chest CT findings. Suboptimal image quality chest CT studies, defined as incomplete anatomic coverage of chest, substantial motion, and limited visualization of thoracic anatomy, were excluded. The reviewers then reviewed chest CT studies for abnormalities in the thoracic structures including: (1) lung parenchyma and airway (GGO, consolidation, peribronchial thickening); (2) pleura (effusion), and (3) mediastinum (lymphadenopathy) based on the established criteria from the Fleischner Society's glossary of terms for thoracic imaging in following section. [18] [19] [20] In addition, the presence of radiological signs or patterns ("halo" sign, "reversed halo" sign, crazy paving pattern, and tree-in-bud pattern) was also evaluated. The "halo" sign was diagnosed when there was a GGO surrounding a pulmonary nodule or consolidation. The "reversed halo" sign was diagnosed when there was a focal rounded area of GGO surrounded by a more or less complete ring of consolidation. 18 Crazy-paving pattern was defined as the presence of thickened interlobular septa superimposed on a background of GGO. 18 Tree-in-bud pattern was diagnosed when multiple areas of centrilobular nodules connected with a linear branching pattern were visualized. 18 In addition, the reviewers were also instructed to record any other thoracic findings that were not included in the aforementioned diagnostic categories. One of the investigators (K. M. D.) in this study reviewed the patient's medical records to assess whether additional findings on CT DAS ET AL. | 3 affected patient management, specifically medication choice, hospital and intensive care unit (ICU) admission decision, and discharge decision. Then, the investigator recorded the findings. Normally distributed continuous variables (e.g., age and the average days of clinical symptoms before image studies) were expressed as the mean ± standard deviation and range. The frequency and percentage of the presence of abnormalities seen in the lung parenchyma and airway, pleura, and mediastinum on the chest radiography and chest CT studies were calculated. The interobserver agreement was determined using Cohen's κ statistics. The The total number of patients admitted and seen at our institution over the study period was 187 consecutive pediatric patients (95 males and 92 females) and all of them underwent chest radiography. Among these 187 patients, 56 patients underwent CT on the same day as chest radiography based on symptoms, as determined by the referring physicians. Therefore, the final patient population consisted of 56 consecutive children (33 males and 23 females; mean age ± SD, 14.8 ± 5.0 years; range, 9 months-18 years). All patients had positive RT-PCR test for COVID-19 and had either mild or moderate symptoms. None of them required supplemental oxygen. All patients were subsequently admitted to the hospital, but none of them required ICU admission. The duration of their hospital admission ranged from 1 to 11 days (mean = 2.6 days). Upon discharge from the hospital, all patients were detained in a quarantine center for a total period of 15 days, including days in the hospital. The details of the initial clinical symptoms are summarized in The chest radiographic findings are summarized in Table 2 Findings of chest CT studies The chest CT findings are summarized in Table 3 On CT, the most frequently seen radiological signs or patterns were: "halo" sign (11/56, 42.3%), followed by "reversed halo" sign (5/56, 19 .2%), crazy paving pattern (2/56, 7.6%), and tree-in-bud pattern (1/56, 3.8%). The comparison of chest radiographic and chest CT findings is summarized in Table 4 . Chest CT studies detected all thoracic abnormalities seen on Among all 224 evaluations (112 evaluations for the chest radiography and 112 evaluations for CT studies, respectively) between two reviewers, discrepancies were noted regarding peribronchial thickening versus normal on two chest radiographs and GGO versus consolidation on one chest CT study. The arbitrator assessed these cases independently and determined that two chest radiographs were normal and chest CT showed consolidation. There was almost perfect interobserver agreement between two reviewers for detecting findings on chest radiographs (κ, 0.89, p = .001) and chest CT studies (κ, 0.96, p = .001). Chest CT detected lung abnormalities, including GGO and/or con- Timely and accurate recognition of radiological signs and patterns typically associated with COVID-19 can lead to optimal pediatric COVID-19 patient management. Our study showed that the "halo" sign, "reversed halo" sign, crazy-paving pattern, and tree-inbud pattern were detected on chest CT studies in symptomatic pediatric patients with COVID-19 infection. Among them, the most frequently seen radiological signs or patterns was "halo" sign (42.3%), followed by "reversed halo" sign (19.2%), crazy paving pattern (7.6%), and tree-in-bud pattern (3.8%). We believe that the "halo" sign, which is currently considered to be present during early phase of COVID-19 infection in the pediatric population, 10, 15, 16 is particularly important because it is unique to pediatric patients with COVID-19 infection in comparison with adults with COVID-19 infection. As pediatric COVID-19 pneumonia progresses, crazy-paving pattern (likely due to underlying alveolar edema coupled with interstitial inflammation) and "reversed halo" sign (likely representing the healing phase of lung infection) are more frequently seen. 10, 11, 15, 27 Our findings of crazy-paving pattern and "reversed halo" sign on chest CT studies of symptomatic pediatric patients with COVID-19 correlate well with findings from previously published studies in the pediatric population. 10, 11, 15 Interestingly, one patient (3.8%) in our study showed tree-in-bud pattern on chest CT study, which has not been previously reported in pediatric patients, however has been rarely described in adult COVID-19 infection. 28 We believe that the tree-in-bud pattern on chest CT study is not typical for COVID-19 infection and is usually indicative of other underlying disease (e.g., aspiration bronchopneumonia) or co-infection (e.g., pulmonary tuberculosis). Future investigation on chest CT findings of tree-in-bud pattern in pediatric patients with COVID-19 infection in the larger patient population will be needed for clarification. Edward Y. Lee https://orcid.org/0000-0001-8453-4920 Early transmission dynamics in Wuhan, China, of novel coronavirus-infected pneumonia Clinical features of patients infected with 2019 novel coronavirus in Wuhan Perceived impact of COVID-19 on pediatric radiology departments around the world: WFPI COVID-19 Task Force survey results from 6 continents COVID-19 CORONAVIRUS PANDEMIC Epidemiology of COVID-19 among children in China Clinical and epidemiological features of 36 children with coronavirus disease 2019 (COVID-19) in Zhejiang, China: an observational cohort study Clinical features of pediatric patients with coronavirus disease (COVID-19) COVID-19 in 7780 pediatric patients: a systemic review Coronavirus infections in children including COVID-19: an overview of the epidemiology, clinical features, diagnosis, treatment and prevention options in children Practical guide for pediatric pulmonologists on imaging management of pediatric patients with COVID-19 CT features of coronavirus disease (COVID-19) in 30 pediatric patients CT features of novel coronavirus pneumonia (COVID-19) in children Private tour guide to pediatric coronavirus disease of 2019 and multisystem inflammatory syndrome in children in 10 minutes: what thoracic radiologists need to know Thoracic imaging findings of multisystem inflammatory syndrome in children (MIS-C) associated with COVID-19: what radiologists need to know now Pediatric SARS, H1N1, MERS, EVLAI, and now coronavirus disease (COVID-19) pneumonia: what radiologists need to know Clinical and CT features in pediatric patients with COVID-19 infection: different points from adults COVID-19). Overview of testing for SARS-CoV-2 Fleischner Society: glossary of terms for thoracic imaging Swine-origin influenza a (H1N1) viral infection in children: initial chest radiographic findings Pediatric pulmonary disease: assessment with high-resolution ultrafast CT Measurement of observer agreement The measurement of observer agreement for categorical data Differences in clinical and imaging presentation of pediatric patients with COVID-19 in comparison with adults High-resolution computed tomography manifestations of COVID-19 infections in patients of different ages Interpretation of chest radiographs in infants with cough and fever Chest CT findings in coronavirus disease-19 (COVID-19): relationship to duration of infection Chest CT features of coronavirus disease 2019 (COVID-19) pneumonia: key points for radiologists Clinical features and chest CT manifestations of coronavirus disease 2019 (COVID-19) in a single ACR Recommendations for the use of chest radiography and computed tomography (CT) for suspected COVID-19 infection International expert consensus statement on chest imaging in pediatric COVID-19 patient management: imaging study reporting and imaging study recommendations What all physicians should know about the potential radiation risk that computed tomography poses for paediatric patients Clinical application of "justification" and "optimization" principle of ALARA in pediatric CT imaging: "how many children can be protected from unnecessary radiation? Managing radiation dose from thoracic multidetector computed tomography in pediatric patients: background, current issues, and recommendations Comparison of chest radiography and chest CT for evaluation of pediatric COVID-19 pneumonia: Does CT add diagnostic value?