key: cord-1055333-pkupfxyr
authors: Roy Choudhury, Shuvro H.; Shahi, Praveen Kumar; Sharma, Shubham; Dhar, Raja
title: Utility of chest radiography on admission for initial triaging of COVID-19 in symptomatic patients
date: 2020-08-31
journal: ERJ Open Res
DOI: 10.1183/23120541.00357-2020
sha: b32e8ed0ca415106654ec01ee4d3fbcb929ad455
doc_id: 1055333
cord_uid: pkupfxyr

A Likert score based on chest radiography of symptomatic patients of #COVID19 can be used as a diagnostic and triaging tool in the emergency room to help physicians identify patients with likelihood of COVID-19 and triage them appropriately https://bit.ly/3gDxzDK

Over a period of 6 weeks, 97 patients presented to the respiratory unit of a private healthcare facility with symptoms causing suspicion of COVID-19. A full clinical, biochemical and haematological assessment was performed on admission along with a chest radiograph. An RT-PCR test was performed on admission in 54 out of 97 patients (those with symptoms exceeding 5 days), and at day 5 of symptom onset (mean of 2.5 days after chest radiograph) in 43 patients. An external radiologist, not involved in the care of patients but aware of the date of symptom onset and the date of the chest radiograph, and blinded to all other parameters, including RT PCR, scored the images. A previously unvalidated Likert score based on radiographic features thought to be related to COVID-19 was developed to objectify the findings. This was based on the CT reporting format suggested by SIMPSON et al. [5] with the addition of one further group: Likert score 3 (table 1) .

Independent matching of the score with the RT-PCR result was then performed. Repeat RT-PCR tests were performed, haematological and biochemical tests continued, and bronchoscopic lavage (n=2) was implemented to exclude COVID-19 in patients with continuing suspicion and an initial negative result, in order to arrive at a definitive diagnosis. 29 out of 97 patients were diagnosed with COVID-19 based on the RT-PCR result. Chest radiography was performed at the time of presentation, which was a mean of 5.56 days (range 1-10 days) after symptom onset. There were 32 posterior-anterior, 58 anterior-posterior erect and seven anterior-posterior supine acquisitions. The sensitivity and specificity using a score of 5 plus 4 for positive diagnosis on chest @ERSpublications A Likert score based on chest radiography of symptomatic patients of #COVID19 can be used as a diagnostic and triaging tool in the emergency room to help physicians identify patients with likelihood of COVID-19 and triage them appropriately https://bit.ly/3gDxzDK radiography for COVID-19 was 75.86% (95% CI 56.5-90%) and 79.41%, respectively; and for a score of 5 plus 4 plus 3 it was 93.1% (95% CI 77-99%) and 63.2%, respectively.

Most radiology societies and the Centers for Disease Control and Prevention do not recommend the use of imaging for diagnosis or screening for COVID-19 [3] . However, in the clinical setting, HRCT has been extensively used worldwide as a diagnostic tool, given the delay of the RT-PCR result as well as its potential false-negative results. Thus far, most reports on CXR and scoring systems have been on monitoring and prognosticating patients with COVID-19 [6] . Based on this preliminary study, a Likert scale based on chest radiography of symptomatic patients can be used as a diagnostic tool in the emergency room, in conjunction with other clinical and laboratory parameters, to help physicians identify patients with a likelihood of COVID-19 and to triage them appropriately regarding isolation and containment pending the RT-PCR result. The tool can be versatile and depending on the clinical situation, the sensitivity can be increased at the cost of specificity by including the indeterminate group as positive for COVID-19 and vice versa. For example, in conjunction with local guidelines, patients with scores of 1 and 2 and fewer symptoms could be considered for supervised home isolation rather than hospital admission. Similarly, highly symptomatic patients with a score of 4 or 5 could have commencement of COVID-19-specific treatment pending RT-PCR result availability (specificity 79.41%). These results may, however, be influenced by the rate of community prevalence of COVID-19, as well as geographical and seasonal variations in the prevalence of other respiratory viral illnesses, and may not be generalisable to all circumstances and geography. The interval between the onset of symptoms and the performance of chest radiography also needs to be taken into consideration when analysing the chest radiograph. This scoring system needs to be further prospectively validated in larger studies. For example, the presence of linear, asymmetrical "interstitial"-looking opacities may occasionally be a radiographic feature of COVID-19 and could be considered in Likert 3 group as opposed to Likert 2. Such a scoring system could be further refined with the addition of clinical and laboratory parameters to improve diagnosis. Based on the high-resolution computed tomography findings described by SIMPSON et al. [5] . CT: computed tomograph; RT-PCR: reverse transcriptase PCR; SARS-CoV-2: severe acute respiratory syndrome coronavirus 2; GGO: ground-glass opacities.

https://doi.org/10.1183/23120541.00357-2020

Correlation of chest CT and RT-PCR testing for coronavirus disease 2019 (COVID-19) in China: a report of 1014 cases

COVID-19 pneumonia: what has CT taught us?

Recommendationsfor-Chest-Radiography-and-CT-for-Suspected-COVID19-Infection. Date last updated: 11

Portable chest X-ray in coronavirus disease-19 (COVID-19): a pictorial review

Radiological Society of North America Expert consensus statement on reporting chest CT findings related to COVID-19. Endorsed by the Society of Thoracic Radiology, the American College of Radiology, and RSNA

COVID-19 outbreak in Italy: experimental chest X-ray scoring system for quantifying and monitoring disease progression

Road, Anandapur, Kolkata, West Bengal 700107, India. E-mail: docaardee@yahoo.com Received: 6 June 2020 | Accepted after revision: 2 July 2020 Data availability: Individual participant data collected are available after deidentification immediately after publication without any end date. The data are available for researchers with a sound research proposal or by the journal editorial team. This can be obtained indefinitely by contacting the corresponding author by e-mail.Conflict of interest: None declared.

Correspondence: Raja Dhar, Respiratory Medicine and Critical Care, Fortis Hospital, #730, Em Bypass