In our previous article titled, “Imaging and COVID-19: Preparing the radiologist for the pandemic,” we discussed the role of imaging in the management of coronavirus disease 2019 (COVID-19), along with the guidelines to help in precise communication and efficient use of healthcare resources.[¹] However, considering the community spread of the disease and the challenges faced by the radiologists in differentiating COVID-19 from other types of community-acquired and atypical pneumonias on a larger scale, an updated review of imaging-related literature seemed prudent. Therefore, we have reviewed the most recent literature on imaging recommendations for the patients with COVID-19. These recommendations are based on meta-analyses, the Fleischner society guidelines, the international expert consensus statement for chest imaging in pediatric COVID-19, and the COVID-19 reporting and data system (CO-RADS) and COVID-19 imaging reporting and data system (COVID-RADS).

META-ANALYSIS ON COMPUTED TOMOGRAPHY IMAGING FEATURES IN CORONAVIRUS DISEASE 2019

In our previous article, we discussed the meta-analysis by Salehi et al., on 919 patients with COVID-19.[¹²] In a recent meta-analysis that evaluated a total of 13 studies published until February 2020, Bao et al. reported that the pooled positive rate for routine CT imaging was 89.76% and for thin CT section imaging was 90.35%. Peripheral, multilobar, and bilateral involvement were the predominant findings seen in 76.95%, 70.81%, and 78.2% of the cases, respectively.[³] Similarly, Zhu et al., in their meta-analysis on the findings from 4121 patients reported that bilateral, multilobar lesions were the most common presentation in patients with COVID-19.[⁴] In the meta-analysis by Wan et al., that comprised 14 studies and 1115 patients, multilobar involvement with a predominant peripheral distribution was observed in the majority of patients.[⁵] These meta-analyses additionally reported that ground glass opacities (GGOs), consolidation, and air bronchogram were among the most common CT findings, whereas pleural effusions and lymphadenopathy were less common.[³⁴⁵]

CONSENSUS STATEMENTS FOR IMAGING IN CORONAVIRUS DISEASE 2019

The Radiological Society of North America along with the Society of Thoracic Radiology and the American College of Radiology have released a consensus statement.[⁶] In this statement, the COVID-19 computed tomography (CT) appearances have been classified into four categories, namely typical, indeterminate, atypical, and negative for pneumonia, as described in our previous article;[¹] this helps in standardizing the reporting language across the radiology fraternity. The reporting template can be accessed at “ .”

The Fleischner society has provided guidelines for the role of imaging in the management of COVID-19.[⁷] The consensus statement recommends that imaging should not be used as a screening tool for COVID-19 in suspected cases or patients with mild symptoms, unless they show clinical deterioration or have comorbidities that might predispose them to rapid disease progression.[⁷]

Foust et al., in their article titled, “International Expert Consensus Statement on Chest Imaging in Pediatric COVID-19 Patient Management,” concur with the reporting guidelines from Simpson et al. and the Fleischner society.[⁸] The pediatric cases can be divided into two groups, namely mild and moderate to severe. Symptoms in the mild cases include fever, cough, and rhinorrhea, whereas those in the moderate to severe cases include dyspnea, hypoxemia, hypotension, chest pain, and tachycardia. For mild cases, a reverse transcription-polymerase chain reaction (RT-PCR) is recommended as the screening test; imaging is not recommended for majority of the patients, except for children with clinical deterioration or underlying comorbidities, including chronic pulmonary illnesses, such as asthma and cystic fibrosis and congenital, pulmonary or cardiac diseases; those undergoing treatment for malignancies; and those with chronic infections such as human immunodeficiency virus or tuberculosis, with or without immunosuppression. For moderate to severe cases, a chest radiograph is indicated at the time of presentation, and CT thorax is reserved for cases in which the imaging findings can impact the management of the patient (diagnosing potential complications).[⁸]

CORONAVIRUS DISEASE 2019 REPORTING AND DATA SYSTEM AND CORONAVIRUS DISEASE 2019 IMAGING REPORTING AND DATA SYSTEM

The Dutch Association for Radiology has proposed a scoring system for COVID-19 based on CT imaging. The system assigns a score named CO-RADS that ranges from 0 to 6. The score 0 indicates that the findings are not interpretable, whereas the score 6 represents CT findings in an RT-PCR-proven case of COVID-19. The CO-RADS scores 1, 2, 3, 4, and 5 represent very low, low, equivocal/uncertain, high, and very high level of suspicion, respectively, for pulmonary involvement. The authors reported a reasonable interobserver variation, with a kappa of 0.47. The CO-RADS has a high discriminatory power for diagnosing COVID-19 with a mean area under the receiver operating characteristic curve of 0.91.[⁹]

Another scoring system named COVID-RADS was proposed in late April 2020 by a group of radiologists from the University of Southern California. This CT-based scoring system categorizes the findings into 5 groups based on a score ranging from 0 to 3, where 0 and 1 represent a low level of suspicion, 2a and 2b represent moderate levels of suspicion, and 3 represents a high level of suspicion.[¹⁰]

CONCLUSION

Even though CT imaging is not recommended as a tool to diagnose COVID-19, it may have a role in the management of COVID-19 in the following scenarios: (1) triaging patients, which indirectly helps in the prediction of disease worsening and disease improvement and (2) understanding the imaging findings that may pose a challenge when the disease hits the community and the radiologists are faced with the problem of differentiating COVID-19 from other conditions. In these difficult times, the CO-RADS and COVID-RADS may not only improve the communication between the radiologists and clinicians, but will also standardize the reporting of the findings across the globe, thus helping in data mining in future.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.