Diagnostic and Prognostic Value of Chest Radiographs for COVID-19 at Presentation

Abstract

Introduction Pulmonary opacities in COVID-19 increase throughout the illness and peak after ten days. The radiological literature mainly focuses on CT findings. The purpose of this study was to assess the diagnostic and prognostic value of chest radiographs (CXR) for coronavirus disease 2019 (COVID-19) at presentation. Methods We retrospectively identified consecutive reverse transcription polymerase reaction-confirmed COVID-19 patients (n = 104, 75% men) and patients (n = 75, 51% men) with repeated negative severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) tests. Two radiologists blindly and independently reviewed the CXRs, documented findings, assigned radiographic assessment of lung edema (RALE) scores, and predicted the patients’ COVID-19 status. We calculated interobserver reliability. The score use for diagnosis and prognosis of COVID-19 was evaluated with the area under the receiver operating characteristic curve. Results The overall RALE score failed to identify COVID-19 patients at presentation. However, the score was inversely correlated with a COVID-19 diagnosis within ≤2 days, and a positive correlation was found six days after symptom onset.Interobserver agreement with regard to separating normal from abnormal CXRs was moderate (k = 0.408) with low specificity (25% and 27%). Definite pleural effusion had almost perfect agreement (k = 0.833) and substantially reduced the odds of a COVID-19 diagnosis. Disease distribution and experts’ opinion on COVID-19 status had only fair interobserver agreement. The RALE score interobserver reliability was moderate to good (intraclass correlation coefficient = 0.745). A high RALE score predicted a poor outcome (intensive care unit hospitalization, intubation, or death) in COVID-19 patients; a score of ≥5 substantially increased the odds of having a poor outcome. Conclusion Chest radiography was found not to be a valid diagnostic tool for COVID-19, as normal or near-normal CXRs are more likely early in the disease course. Pleural effusions at presentation suggest a diagnosis other than COVID-19. More extensive lung opacities at presentation are associated with poor outcome in COVID-19 patients. Thus, patients with more than minimal opacities should be monitored closely for clinical deterioration. This clinical application of CXR is its greatest strength in COVID-19 as it impacts patient care.

Automated processing of thermal imaging to detect COVID-19

“…First, in line with a previous report 12 , RALE scores were not different between patients with and without COVID-19, and did not provide diagnostic yields ( Figure 5A ). As expected, patients with lung injury had higher RALE scores than control patients ( Figure 5A ).…”

“…To compare our thermal imaging technique to conventional CXR, we used the radiographic assessment of lung edema (RALE) score 12 . We compared thermal imaging scores with RALE scores for all patients who had a CXR within 6 hours of thermal imaging (n = 30).…”

Discrepancies in the clinical and radiological profiles of COVID-19: A case-based discussion and review of literature

“…However, many of the imaging findings in COVID-19 can be nonspecific, and there can be occasional discrepancies between the imaging and clinical features seen in these patients[ 7 , 8 ]. Moreover, on occasions, extrapulmonary disease may dominate in some patients, and pre-existing major illnesses (such as heart failure, liver diseases, chronic kidney disease and malignancies) with the acquisition of COVID-19 illness may pose additional diagnostic dilemmas[ 9 - 14 ].…”

Highlighting COVID-19: What the imaging exams show about the disease

“…This evaluation is a challenge that must be considered, mainly due to the lack of specialized radiologists[ 39 , 41 ]. Therefore, an artificial intelligence (AI) algorithm could be used and programmed to distinguish COVID-19 pneumonia from non-COVID-19 pneumonia through CXR images[ 44 ]. The main goal of AI, when gathering imaging data and clinical information is to read image studies accurately, preferably as a screening tool[ 34 ].…”