Secondary Logo

Journal Logo

Review Article

Computed Tomography Features of Coronavirus Disease 2019 (COVID-19)

A Review for Radiologists

Goyal, Nikhil MD*; Chung, Michael MD; Bernheim, Adam MD; Keir, Graham MD; Mei, Xueyan MS§; Huang, Mingqian MD; Li, Shaolin MD; Kanne, Jeffrey P. MD

Author Information
Journal of Thoracic Imaging: July 2020 - Volume 35 - Issue 4 - p 211-218
doi: 10.1097/RTI.0000000000000527
  • Free


Toward the end of 2019, an outbreak of pneumonia was first reported in central China’s Hubei province, causing simultaneous severe illness of numerous patients that, in some cases, resulted in hospitalization, intensive care unit admission, and death. The responsible pathogen was found to be a novel coronavirus named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).1 Although the majority of initial cases were linked to an animal market in the city of Wuhan,2 direct human-to-human transmission quickly developed among those with contacts of infected people.3 The infection, subsequently named Coronavirus Disease 2019 (COVID-19), has quickly expanded worldwide, and continues to be a serious global health emergency.

SARS-CoV-2 belongs to a family of viruses known as coronaviridae. In 2002, a novel coronavirus (SARS-CoV) led to an epidemic in southeast China known as severe acute respiratory syndrome (SARS), which ultimately resulted in 774 deaths out of 8098 total cases.4 Another coronavirus (MERS-CoV) outbreak known as Middle East Respiratory Syndrome (MERS) began in Saudi Arabia in 2012 and ultimately resulted in 858 deaths out of 2494 infected individuals.5 The current COVID-19 pandemic has provoked global alarm and has triggered an international effort to contain the spread of disease, as the number of infected patients has surpassed both SARS and MERS by at least an order of magnitude,6 albeit with an overall lower mortality rate (Table 1).3,7–10

Mortality Rates of Global Pandemics

The SARS outbreak of 2002 was eventually eradicated, and no cases have been reported since 2003.11 MERS, in contrast, continues to have occasional small outbreaks, usually limited to one locale.12

Current detection methods rely on reverse transcriptase polymerase chain reaction (RT-PCR) to test mucosal swabs of suspected patients. However, despite health agencies’ attempts to increase access to testing, the availability of test kits varies widely geographically, and shortages remain in some locales. The role of chest computed tomography (CT) in the management of COVID-19 patients is yet to be determined. In this review, we aim to familiarize readers with the CT imaging appearances of COVID-19. Ideally, based on screening and triage, suspected patients are isolated before imaging. However, in the rare instance that COVID-19 is not suspected and CT findings are suggestive, prompt recognition and reporting facilitate isolation protocol execution to limit spread of disease.


It is estimated that the first cases of COVID-19 originated in early December 2019 in Wuhan, the capital city of the Hubei Province, China, with subsequent cases increasing at a rapid rate. On December 31, 2019, China alerted the World Health Organization (WHO) about several pneumonia cases in this province. By January 29, 2020, 6000 new cases were reported in China. In the middle of February 2020, the Chinese government broadened the diagnostic criteria to include patients with chest CTs showing typical imaging features of disease, which sharply increased the number of new cases by ~15,000 in just 1 day. In the most recently published version of diagnostic criteria in China, chest CT was removed.13 Despite China’s efforts to contain transmission, in today’s interconnected world of high-volume global travel, the virus quickly spread across continents, with large outbreaks in Iran, South Korea, and northern Italy. As more widespread testing is made available, these figures are anticipated to climb substantially. On March 11, 2020, the WHO officially labeled COVID-19 a worldwide pandemic.9 As of April 2, 2020, there are 956,588 cases reported in 180 countries.6 In the United States, to date, there are 216,722 cases reported, with new cases being reported daily from community spread.6

Currently, the number of deaths attributed to COVID-19 infection worldwide is 28,320,6 with a current estimated mortality rate of 3.6%.14 (Mortality estimates vary by region and are related to the local health care systems’ capacity to treat large numbers of patients who suddenly present.) When adjusted for the 14-day incubation period of the virus, one study has reported the mortality rate to be as high as 5.6%.14 However, given than many infected patients have few or no symptoms, they may not be tested, and true mortality rates may be lower. What is known is that older patients and patients with comorbidities have higher mortality rates.12 In epidemiological terms, the basic reproduction number (designated R0) is an estimate of the average number of new infections that can arise from contact with an infected individual. An R0 value above 1.0 indicates infection spread, and R0 <0 indicates a decline in affected number of individuals. The estimated R0 for COVID-19 was recently reported to be 2.2,3 which exceeds that of many other viral pneumonias such as influenza. Worldwide, the number of patients who have recovered (at the time of submission) is 202,728.6 South Korea, after implementing aggressive testing and containment measures, has been able to decrease spread rapidly.15 In late March of 2020, China reported no new community cases of COVID-19, and all recent cases were related to people traveling into China who were quarantined on arrival.13 With increased testing and aggressive containment measures, other nations aim to similarly report improving statistics.


Given that COVID-19 affects the lower respiratory tract, most patients present with cough, fever, dyspnea, and myalgias. Because most cases are mild and symptoms are nonspecific, it is suspected that many people may not seek medical attention. On the basis of preliminary data, the median incubation period is ~5 days, but may range from 2 to 14 days.16 Viral shedding can occur before the onset of symptoms. In one study, the median duration of viral shedding was 20 days, with the longest survivor shedding for 37 days.17 Acute respiratory distress syndrome develops in 17% to 29% of patients.18,19 Because there is significant overlap between COVID-19, influenza, and other viral illnesses, laboratory findings play a key role in diagnosis. In one study of 1099 COVID-19 patients, lymphocytopenia was present in 83%, 36% had thrombocytopenia, and 34% had leukopenia.3 Ultimately, RT-PCR for the detection of SARS-CoV-2 RNA is the reference standard for diagnosis.16

Children with COVID-19 typically have less severe infection, and all pediatric patients in the early phase of outbreak in Wuhan were discharged home after about 7.5 days in the hospital, with only one child requiring ICU admission.20 The largest review of children with COVID-19 included 2143 children in China. Only 112 (5.6%) of 2143 children had severe disease (defined as hypoxia) and 13 (0.6%) children developed respiratory or multiorgan failure or acute respiratory distress syndrome.21

Chen et al22 reviewed charts of 9 pregnant women with PCR-positive COVID-19 infection and found no evidence of vertical transmission to babies born in this group. At present, there are scant data about vertical transmission overall. A research letter dated March 26, 2020 in JAMA documented a possible vertical transmission case with elevated IgM antibodies in the newborn.23 The current Centers for Disease Control (CDC) data state that no infants born to mothers with COVID-19 have tested positive for the COVID-19 virus. According to the CDC, SARS-CoV-2 has not been detected in breastmilk, but it is unknown whether mothers with COVID-19 can transmit the disease via breastmilk.16

Certain cohorts of adults have more severe clinical courses and a higher mortality rates. Variables associated with increased risk of mortality include older age, D-dimer levels >1 μg/mL, and higher sequential organ failure assessment (SOFA) score.17 Coronaviruses that are pathogenic to humans express a spike protein on their surfaces that binds to angiotensin-converting enzyme 2 receptors (ACE-2 receptors). ACE-2 receptors are found in the lung, kidney, intestines, and on the surfaces of blood vessels.24 Some reports indicate that most prevalent comorbidities in nonsurvivors of COVID-19 are hypertension, diabetes, and coronary heart disease.8 Consequently, others have questioned if conditions in which ACE-2 receptor expression is upregulated (namely diabetics and hypertensives taking ACE-inhibitors) may be more prone to infection with COVID-19.25

The CDC states that people with chronic lung disease or moderate to severe asthma are at increased risk for severe illness by SARS-CoV-216; however, no formal data exist on the role of pre-existing lung pathology and its relation to the severity of COVID-19.


Findings on chest radiographs in patients with COVID-19 range from normal (especially in the early stages) to unilateral (Fig. 1) or bilateral peripheral consolidation to diffuse lung opacities in severely affected patients (Fig. 2). Findings may be subtle and overlap with those encountered in other viral pneumonias such as influenza26 and organizing and eosinophilic pneumonia and other acute lung injuries. This, combined with inconsistent use of terminology in reporting chest radiographs findings, renders radiography of limited value in the assessment of COVID-19 infection. However, portable radiographs can be useful for the initial evaluation of patients presenting with a respiratory complaint without transporting potentially infected patients around the hospital.

A 32-year-old man presented to an outpatient imaging center with cough and known contact with a COVID-19-positive patient. Posteroanterior chest radiograph shows consolidation (arrow) in the periphery of the right lower lung. This is a nonspecific appearance, but may reflect infection, given contact with a known COVID-19 individual.
A 65-year-old man with “flu-like illness” 2 weeks earlier initially treated with antibiotics presented with increasing shortness of breath and hypoxia and was diagnosed with COVID-19 by PCR. Posteroanterior chest radiograph shows bilateral patchy lung consolidation in a peripheral distribution.

CT is more sensitive than radiography for identifying lung abnormalities in patients with COVID-19. The chest CT findings in COVID-19 infection evolve as the illness progresses in a somewhat predictable nature. Specifically, findings evolve similarly to other causes of acute lung injury. In one study of early cases from China,27 imaging findings related to disease progression were divided into phases on the basis of the number of days from symptom onset to initial CT—early (0 to 2 d); intermediate (3 to 5 d); and late (6 to 12 d) (days=days between presentation and initial CT). The investigators found that 56% of early-stage patients had a normal chest CT, but only 4% of late-stage patients had a negative CT. Patients imaged early in the course of disease frequently had unilateral findings, and bilateral lung involvement was more common as time from symptom onset lengthened (28% early → 88% late). Typical findings included focal ground-glass opacities (often (54%) round in shape or sometimes more confluent) with a somewhat basilar and peripheral distribution (Figs. 3–5).27 This has been reproduced in several studies.28–31 At least 2 strains of the virus have been documented, designated L and S subtypes,32 and the question of whether CT features differ in patients infected with L or S is yet to be answered.

A 46-year-old man from Wuhan, China, with COVID-19 presenting with fever and cough. Unenhanced, thin-section CT performed 2 days after the onset of symptoms shows bilateral ground-glass attenuation lung nodules and masses.
A 42-year-old man from China diagnosed with a COVID-19 history presenting with 3 days of fever and productive cough. Axial (A) and sagittal (B) thin-section unenhanced CT images show ground-glass opacities in the right upper lobe and left lower lobe with a striking peripheral distribution.
A 56-year-old woman with COVID-19 and recent travel to Wuhan, China, presenting with 5 days of fevers. Axial and coronal thin-section CT images show multiple bilateral ground-glass opacities (arrows) with rounded morphology.

The evolution of lung findings on chest CT mirrors other forms of acute lung injury. As lung tissue reacts to the initial insult, ground-glass opacities may become more confluent (Figs. 6, 7) or dense, and some areas may eventually exhibit both septal lines and intralobular lines. As time from initial injury increases, findings can evolve into dense consolidation or a crazy-paving pattern (Figs. 8–13). Interestingly, dense lobar consolidation without ground-glass abnormality was rarely seen.28 An alternative diagnosis such as co-infection with bacteria should be considered if consolidation is the lone finding. A subset of patients may develop an “atoll” or a “reverse-halo” sign, as can occur with other causes of organizing pneumonia (Figs. 14, 15). The frequency of this finding in the Wuhan cohort was 4%. As with other forms of acute lung injury, linear opacities, subpleural curvilinear bands, and bronchial dilation can develop with organization and fibrosis. Importantly, multiple studies have consistently shown the lack of pleural effusions, discrete solid pulmonary nodules, lymphadenopathy, and cavities.33 The ability to distinguish COVID-19 from other forms of acute lung injury is a vital topic to consider, especially because there is overlap between the appearances of COVID-19 and other entities such as influenza (Fig. 16). This is especially important as subspecialty thoracic radiology readings may not be available in many communities or may be limited if large numbers of patients are scanned in a relatively short amount of time as the pandemic increases. One study found the accuracy of 4 US radiologists to be 83% to 97%, sensitivity to range from 70% to 93%, and specificity to be 93% to 100%. Three Chinese radiologists also demonstrated accuracy values from 60% to 83%, and sensitivity and specificity ranges from 72% to 94%. Both sets of radiologists were blinded to the PCR results in this study.34 Although this may seem promising, this study had a number of limitations including study design, small number of cases, and exclusion of noninfectious causes of organizing pneumonia.

A 56-year-old male with COVID-19 presenting with fever, dry cough, and shortness of breath. He was initially diagnosed with influenza, but no lab testing was done. He presented to the Emergency Department 8 days after symptom onset with worsening shortness of breath. (At the time of CT, COVID RT-PCR was pending and was reported positive 2 days later.) Contrast-enhanced CT image shows confluent ground-glass opacities in the upper lobes (black arrows) with minimal septal thickening and more focal peripheral ground-glass opacity in the lower lobes.
A 74-year-old woman presented with cough, congestion, and fever ×2 days. Axial CT demonstrates bilateral lower lobe ground-glass opacities, some rounded and some more confluent.
A 59-year-old male presented with hypoxia and progressive shortness of breath for 10 days. Axial CT image at this stage of presentation shows ground-glass opacity (solid arrow) along with more dense areas of peripheral consolidation (dashed arrow) in the lower lobes.
A 49-year-old woman with COVID-19 and no exposure history presenting with 3 days of fever. Thin-section, unenhanced CT images show peripheral ground-glass opacities bilaterally with interlobular septal thickening and discrete peripheral foci of crazy-paving (arrows).
A 43-year-old woman with COVID-19 and history of recent travel to Wuhan, China, presenting with 6 days of fever and dry cough. Thin-section, unenhanced CT image shows peripheral ground-glass opacity and a focus of crazy-paving (arrow) in the left lower lobe.
A 60-year-old man with COVID-19 and recent travel to Wuhan, China, presenting after 7 days of fever and dry cough. A, Thin-section unenhanced CT image shows peripheral (solid arrow) and rounded (dashed arrow) ground-glass opacities in the middle and lower lobes. B, Thin-section, unenhanced CT image from 4 days later shows marked increase in now diffuse ground-glass opacity with scattered septal lines (arrows).
A 61-year-old male with a history of 3 to 4 days of weakness, followed by cough. Patient was hypoxic in the emergency room and intubated. COVID-19 PCR was reported positive a few hours after the CT scan. Note extensive areas of crazy-paving pattern throughout both lungs. (Case courtesy of Jonathan Scheiner, MD).
A 44-year-old man with COVID-19 and with no exposure history presenting with 2 days of fever and dry cough. A and B, Thin-section, unenhanced CT images show peripheral, rounded ground-glass opacities in the peripheral right upper lobe and both lower lobes. C and D, Thin-section unenhanced CT images from 3 days later show increase and confluence of ground-glass opacities with new crazy-paving and right lower lobe nodular consolidation (arrows).
A 22-year-old woman with COVID-19 presenting with 10 days of fever and nonproductive cough. Coronal thin-section unenhanced CT image shows an area of faint ground-glass opacity in the left upper lobe with a surrounding ring of denser consolidation (arrow, “reverse-halo sign”) and a curvilinear focus of consolidation in the superior segment of the lower lobe.
A 52-year-old woman with COVID-19 and worsening cough and fever for 1 month. Thin-section, unenhanced CT image shows an atoll sign in the left lower lobe (arrow), characterized central ground-glass opacity, and surrounding thin rim of consolidation.
A 44-year-old woman with influenza A infection presenting with fever, myalgia, malaise, and cough. Contrast-enhanced, thin-section CT image shows discrete peripheral and central ground-glass opacities (black arrows).


At present, the CDC, American College of Radiology (ACR), Society of Thoracic Radiology (STR), and American Society of Emergency Radiology (ASER) do not recommend chest radiography or CT as a first line to screen for COVID-19 infection.16,26,35,36 RT-PCR remains the only specific method of diagnosis, even if imaging is normal. Some institutions where rapid PCR testing is limited have considered chest CT as a decision-making tool for patient admission or as criteria for testing for COVID or treatment for an alternate diagnosis. The ACR currently strongly cautions against disposition decisions based solely on CT findings, as a normal chest CT does not effectively exclude COVID-19 and an abnormal CT is not specific for COVID-19.26 CT may be used in hospitalized patients with specific clinical indications such as to identify complications such as pulmonary embolism, empyema, or co-infection.

RT-PCR, considered to be the reference standard, does have its own limitations and variability. A number of factors influence the results of the RT-PCR assay including site of specimen (nasal or bronchial), chronicity of illness (early or late) at the time of sampling, and reliability of the testing kit.37 Xie et al38 demonstrated that 3% of (5/167) patients with a negative RT-PCR examination had positive CT findings, whereas another study39 reported up to 30%. A third study at a different medical center in China showed that 29% (15/51) of patients with positive CT findings having negative RT-PCR at presentation (PCR eventually became positive within 1 to 7 d).40 However, the accuracy of these early PCR findings has not been validated. As the pandemic evolves, the recommendations for disposition ultimately will depend on clinical evaluation, availability and turnaround time of RT-PCR, and, in some very select instances, imaging.


COVID-19 has rapidly emerged as a global health emergency. Early detection and containment of suspected cases and social distancing are necessary to prevent further spread of infection. Currently, laboratory testing with RT-PCR remains the primary method of diagnosis. Although the CT findings of COVID-19 have been well described both at presentation and throughout the course of the disease, the current role of CT scanning is limited to identifying complications of infection. Nevertheless, radiologists should be familiar with the imaging appearances of COVID-19 so as to consider the diagnosis when it is unexpected and to recognize the overlap with other causes of acute lung injury.


1. Zhu N, Zhang D, Wang W, et al. A novel coronavirus from patients with pneumonia in China, 2019. N Engl J Med. 2020;382:727–733.
2. Lu R, Zhao X, Li J, et al. Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding. Lancet. 2020;395:565–574.
3. Li Q, Guan X, Wu P, et al. Early transmission dynamics in Wuhan, China, of novel coronavirus-infected pneumonia. N Engl J Med. 2020;382:1199–1207.
4. World Health Organization. Summary of probable SARS cases with onset of illness. 2003. Available at: Accessed March 23, 2020.
5. World Health Organization. Middle east respiratory syndrome coronavirus (MERS-CoV). 2019. Available at: Accessed March 23, 2020.
6. Center for System Science and Engineering. Coronavirus COVID-19 Global Cases by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins. 2020. Available at: Accessed April 2, 2020.
7. de Wit E, van Doremalen N, Falzarano D, et al. SARS and MERS: recent insights into emerging coronaviruses. Nat Rev Microbiol. 2016;14:523–34.
8. Guan WJ, Ni ZY, Hu Y, et al. Clinical characteristics of coronavirus disease 2019 in China. N Engl J Med. 2020. [Epub ahead of print].
9. World Health Organization. 2020. Available at: Accessed March 23, 2020.
10. Taubenberger JK, Morens DM. 1918 Influenza: the mother of all pandemics. Emerg Infect Dis. 2006;12:15–22.
11. Wilder-Smith A, Chiew CJ, Lee VJ. Can we contain the COVID-19 outbreak with the same measures as for SARS? Lancet Infect Dis. 2020. [Epub ahead of print].
12. Nassar MS, Bakhrebah MA, Meo SA, et al. Middle East Respiratory Syndrome Coronavirus (MERS-CoV) infection: epidemiology, pathogenesis and clinical characteristics. Eur Rev Med Pharmacol Sci. 2018;22:4956–4961.
13. National Health Comission of the People’s Republic of China. 2019. Available at: Accessed March 23, 2020.
14. Baud D, Qi X, Nielsen-Saines K, et al. Real estimates of mortality following COVID-19 infection. Lancet Infect Dis. 2020. [Epub ahead of print].
15. Choi SC, Ki M. Estimating the reproductive number and the outbreak size of Novel Coronavirus disease (COVID-19) using mathematical model in Republic of Korea. Epidemiol Health. 2020:e2020011.
16. Centers for Disease Control. CDC Coronavirus Disease 2019 (COVID 2019) Basics. 2020. Available at: Accessed March 23, 2020.
17. Zhou F, Yu T, Du R, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. Lancet. 2020;395:1054–1062.
18. Chen N, Zhou M, Dong X, et al. Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study. Lancet. 2020;395:507–513.
19. Huang C, Wang Y, Li X, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet. 2020;395:497–506.
20. Liu W, Zhang Q, Chen J, et al. Detection of Covid-19 in children in early January 2020 in Wuhan, China. N Engl J Med. 2020;382:1370–1371.
21. Dong Y, Mo X, Hu Y, et al. Epidemiological characteristics of 2143 pediatric patients with 2019 Coronavirus Disease in China. Pediatrics. 2020:e20200702.
22. Chen H, Guo J, Wang C, et al. Clinical characteristics and intrauterine vertical transmission potential of COVID-19 infection in nine pregnant women: a retrospective review of medical records. Lancet. 2020;395:809–815.
23. Dong L, Tian J, He S, et al. Possible vertical transmission of SARS-CoV-2 from an infected mother to her newborn. JAMA. 2020. [Epub ahead of print].
24. Zhang H, Penninger JM, Li Y, et al. Angiotensin-converting enzyme 2 (ACE2) as a SARS-CoV-2 receptor: molecular mechanisms and potential therapeutic target. Intensive Care Med. 2020;46:586–590.
25. Fang L, Karakiulakis G, Roth M. Are patients with hypertension and diabetes mellitus at increased risk for COVID-19 infection. Lancet Respir Med. 2020;8:E21.
26. American College of Radiology (ACR). ACR recommendations for the use of chest radiography and computed tomography (CT) for suspected COVID-19 infection. 2020. Available at: Accessed March 23, 2020.
27. Bernheim A, Mei X, Huang M, et al. Chest CT findings in Coronavirus Disease-19 (COVID-19): relationship to duration of infection. Radiology. 2020:200463.
28. Chung M, Bernheim A, Mei X, et al. CT Imaging features of 2019 Novel Coronavirus (2019-nCoV). Radiology. 2020;295:202–207.
29. Yoon SH, Lee KH, Kim JY, et al. Chest radiographic and CT findings of the 2019 Novel Coronavirus Disease (COVID-19): analysis of nine patients treated in Korea. Korean J Radiol. 2020;21:494–500.
30. Zhou S, Wang Y, Zhu T, et al. CT features of Coronavirus Disease 2019 (COVID-19) pneumonia in 62 patients in Wuhan, China. AJR Am J Roentgenol. 2020. [Epub ahead of print].
31. Zu ZY, Jiang MD, Xu PP, et al. Coronavirus Disease 2019 (COVID-19): a perspective from China. Radiology. 2020:200490.
32. Tang X, Wu C, Li X, et al. On the origin and continuing evolution of SARS-CoV-2. Natl Sci Rev. 2020. [Epub ahead of print].
33. Kanne JP, Little BP, Chung JH, et al. Essentials for radiologists on COVID-19: an update-radiology scientific expert panel. Radiology. 2020:200527.
34. Bai HX, Hsieh B, Xiong Z, et al. Performance of radiologists in differentiating COVID-19 from viral pneumonia on chest CT. Radiology. 2020:200823.
35. American Society of Emergency Radiology. 2020. Available at:
36. Society of Thoracic Radiology. STR/ASER COVID-19 position statment. Available at: Accessed March 23, 2020.
37. Yang Y, Yang M, Shen C, et al., Evaluating the accuracy of different respiratory specimens in the laboratory diagnosis and monitoring the viral shedding of 2019-nCoV infections. 2020. [Epub ahead of print].
38. Xie X, Zhong Z, Zhao W, et al. Chest CT for typical 2019-nCoV pneumonia: relationship to negative RT-PCR testing. Radiology. 2020:200343.
39. Ai T, Yang Z, Hou H, et al. Correlation of chest CT and RT-PCR Testing in Coronavirus Disease 2019 (COVID-19) in China: a report of 1014 cases. Radiology. 2020:200642.
40. Fang Y, Zhang H, Xie J, et al. Sensitivity of chest CT for COVID-19: comparison to RT-PCR. Radiology. 2020:200432.

Coronavirus Disease 2019; coronavirus; imaging

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.