Annotation

In this retrospective case series, chest CT scans of 21 patients from China with clinical manifestations infected with the new 2019 coronavirus (2019-nCoV) were analyzed; emphasis was placed on identifying and characterizing the most common features. Typical CT signs included bilateral “frosted glass” thickening of the pulmonary parenchyma and consolidation zones, which sometimes had a rounded shape and extended to the peripheral parts of the lungs. Notably, pulmonary cavitations, solitary nodules, pleural effusion and lymphadenopathy were absent. At further CT control in this group of patients during the study period, mild or moderate progression of changes was often revealed, which was manifested in the increase in the degree and density of areas of lung tissue thickening.

Brief overview

Infection caused by the new 2019 coronavirus shows characteristic signs on chest CT scans that aid the radiologist in early detection and diagnosis of this new global public health emergency.

Key findings

• Of the 21 patients with 2019-nCoV infection, 15 (71%) had lesions in more than two lobes on chest CT, 12 (57%) visualized frosted glass-like consolidations, seven (33%) had rounded consolidations, seven (33%) had peripheral spread, six (29%) had foci of both consolidation and frosted glass-like consolidations, and four (19%) had a crazy-paving pattern.
• Pulmonary cavitations, solitary pulmonary nodules, pleural effusion and lymphadenopathy were absent.
• Fourteen percent of patients (three of 21) had no pathologic changes on CT scan.

Introduction

On 31 December 2019, the World Health Organization (WHO) was notified of several cases of a respiratory disease of unknown origin detected in Wuhan City, Hubei Province, China, with clinical manifestations resembling viral pneumonia symptoms and presenting with fever, cough, and shortness of breath. As of January 30, 2020, WHO has designated the outbreak as a global health emergency. Chest X-rays often showed pulmonary darkening in patients. By January 3, 2020, 44 patients with this illness of unclear etiology had been reported to WHO, with preliminary epidemiological investigation indicating that most patients worked or were frequent visitors to the Huangan seafood wholesale market [1]. Analysis of bronchoalveolar lavage fluid samples and electron microscopy showed that the causative agent was coronavirus, with its characteristic ultrastructural morphology in electron microscopy, which is due to the presence of spike-like outgrowths emanating from the viral envelope, resembling the appearance of a corona. The discovered virus has been provisionally named the new 2019 coronavirus (2019-nCoV). Coronaviruses belong to the family Coronaviridae, a genus Nidovarales and family that includes viruses that cause diseases ranging from the common cold to severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS) [2]. The SARS outbreak began in southern China and was recognized as a global public health threat in March 2003, and from November 2002 to July 2003 resulted in 774 deaths out of 8098 infected individuals [3]. MERS was first reported in 2012 in Saudi Arabia, and to date there have been 858 deaths per 2494 infections [4]. As of January 28, 2020, there are 4,593 confirmed cases worldwide. Of these, 4,537 are in China and 56 from 14 other countries. In addition, there are another 6,973 suspected cases in China. As of January 28, 2020, there have been 106 fatal cases – all in China [5]. On January 30, 2020, WHO declared Chinese coronavirus a global health emergency. An initial prospective analysis of clinical data from 41 initially admitted patients with laboratory-confirmed 2019-nCoV in Wuhan showed that 2019-nCoV caused severe illness clinically similar to SARS, sometimes resulting in the need for ICU hospitalization (13 of 41 [32%]) and death (6 of 41 [15%]). All patients with pneumonia in this study had changes on chest computed tomography: preliminary reports showed bilateral thickening of the lung parenchyma in all patients [6]. In this article, we describe and characterize the key CT features found in a cohort of 21 patients infected with 2019-nCoV in China to familiarize radiologists and other clinical team members with the radiologic manifestations of this emerging disease. Early recognition of the disease can expedite treatment and provide early patient isolation, it will also enable early surveillance, containment and public health response to this infectious disease.

Materials and methods

Patients and chest CT

Our institutional review board waived the requirement to obtain written informed consent for retrospective analyses of this case series, in which depersonalized data were evaluated and no potential risk to patients was anticipated. To prevent potential breaches of confidentiality, patient-researcher communication was not available. Between January 18, 2020 and January 27, 2020, 21 patients with confirmed 2019-nCoV infection were admitted to three hospitals in three provinces in China; all underwent chest CT scans. Ten patients from Zhuhai (Guangdong Province) underwent CT scans with 1-mm-thick slices using a CT CT scanner UCT 760 (United Imaging, Shanghai, China). Nine patients were from Nanchang (Jiangxi Province) Tomography with 8-mm-thick slices using an Emotion 16 scanner (Siemens Healthineers). Two patients were from Qingdao (Shandong Province) and underwent a study with a slice thickness of 5 mm, one using a BrightSpeed scanner (GE Medical Systems) and one using an Aquilion ONE scanner (Toshiba Medical Systems). All studies were performed in the patient’s supine position at inspiratory height, without contrast agent administration. All patients had confirmed 2019-nCov infection by laboratory testing of bronchoalveolar lavage fluid, endotracheal aspirate, nasopharyngeal swab, or oropharyngeal swab. Patient selection for this study was continuous at each of the three institutions, and no exclusion criteria were applied (Table 1). In addition to information on age and sex, the clinical information collected included data on the severity and course of symptoms, and epidemiologic history, including information on patient travel. Table 1 | General patient data (n = 21)

Analysis of CT studies

Each CT study was reviewed by two radiologists trained in cardiothoracic radiology with approximately 5 years of experience each (M.C. and A.B.) using a viewing console. Studies were reviewed independently and final results were formed by consensus. If there was disagreement between the two main interpretations, a third qualified cardiothoracic radiologist with 10 years of experience made the final judgment. Negative control and blinded method were not used. For each of the 21 patients, the initial (conducted on admission–note.) CT examinations were evaluated for the following characteristics: (a) the presence of frosted-glass type seals, (b) the presence of consolidation, (c) the number of lobes to which frosted-glass type seals and consolidation areas extended, (d) the extent of lobe involvement in addition to a general “lung lesion severity scale”, (d) presence of nodules, (e) presence of pleural effusion, (j) presence of thoracic lymphadenopathy (defined as lymph node size ≥10 mm in the short axis dimension), and (z) presence of underlying lower respiratory disease such as emphysema or fibrosis. Other CT symptoms, such as cavitations, reticulation, thickening of the interval septa, calcifications, and bronchiectasis, were also noted. Frosted-glass consolidation was defined as an increase in lung tissue density in the form of haze with preserved visualization of bronchial and vascular margins, and consolidation was defined as an increase in lung tissue density with darkening of vascular margins and airway walls [7]. Each of the five lung lobes was graded according to the degree of involvement and categorized as absent (0%), minimal (1-25%), mild (26-50%), moderate (51-75%), or severe (76- 100%). Absence of involvement corresponded to 0 points, minimal involvement corresponded to 1 point, mild involvement corresponded to 2 points, moderate involvement corresponded to 3 points, and severe involvement corresponded to 4 points. The result of the overall “lung lesion severity scale” was generated by summing the scores from each of the five lung lobes (range of possible scores 0-20). Eight patients underwent a control chest CT scan during the study. These scans were also evaluated to determine change or progression over time, which was qualitatively assessed using a consensus approach by two radiologists (M.C. and A.B.).

Results

Thirteen men and 8 women in the age range 29-77 years participated in the study; mean age ± standard deviation, 51 years ±14.

Frosted glass seals and consolidation

Of the 21 initial CT studies, three (14%) patients did not have frosted-glass consolidations; in essence, these three patients had no CT signs of chest pathology on admission (Table 2). Of the 18 patients in whom areas of “frosted glass” type consolidation or consolidation were visualized, only “frosted glass” type consolidations were noted in 12 (no consolidation). No patient had areas of consolidation of lung tissue without “frosted glass” consolidation. Table 2 | CT signs found in 21 patients on initial examination

Prevalence and patterns of increased lung tissue density

With the exception of three patients with a normal initial chest CT scan, the remaining 18 of 21 patients (86%) had changes affecting at least one lobe by definition. Of the 21 patients, one patient (5%) had one lobe affected, two patients (10%) had changes defined in two lobes, three patients (14%) had three lobes affected, four patients (19%) had four lobes affected, and eight patients (38%) had pathologic changes extending to all five lobes. The right upper lobe was involved in 14 of 21 patients on initial CT examination (67%), the right middle lobe in 12 (57%), the right lower lobe in 16 (76%), and changes in the left upper lobe were noted in 14 (67%) patients and the left lower lobe in 14 (67%). Of the 18 patients with increased lung density, 16 had bilateral changes and two had unilateral changes (both had right lung lesions). The total score on the “lung lesion severity scale” ranged from 0 (with three normal CT studies) to a maximum of 19 with a mean score of 9.9. The patient with the highest severity score was admitted to the ICU (Figure 1). Seven (33%) patients had “frosted glass” type seals and/or rounded consolidation areas on initial examination (Fig. 2). Three of the 21 patients (14%) had a predominantly linear pattern and four (19%) had a cobblestone symptom (Fig. 3). Seven patients (21%) had spread of changes to the peripheral lung (Fig. 4). No patient had pulmonary cavitations, isolated pulmonary nodules, pleural effusion(s), lymphadenopathy, pulmonary emphysema or fibrosis.

CT control data in dynamics

During the study period, eight patients underwent repeat chest CT scans, and one patient underwent two control CT scans (Table 3). The mean time between the initial chest CT and the follow-up CT was 2.5 days (range 1-4 days). In one of eight patients (13%), the initial and follow-up chest CTs showed no pathologic features, with no change in follow-up. None of the patients had positive CT dynamics. Five of eight patients (63 %) had slight progression of changes, and two (25 %) had moderate progression. No patient demonstrated significant negative dynamics. One patient with a slight worsening of the CT pattern over time, as determined by the first follow-up CT scan performed 1 day after the initial examination, underwent a second follow-up CT scan 3 days later (4 days after the initial CT examination). On repeat CT follow-up, there was a marked progression of changes in the form of an increase in the number of areas of consolidation and consolidation density. Finally, in one patient who had no changes on the initial chest CT scan, a new single, single, rounded, frosted-glass-type consolidation in the periphery was found on control CT 3 days later (Figure 5). Table 3 | Qualitative assessment of changes on follow-up CT scanning over time in eight patients

Discussion

The 2019 new coronavirus (2019-nCoV) outbreak has potentially far-reaching public health implications. Chest CT is a key component of the diagnostic workup of patients suspected of having this infection, and our study showed some radiologic signs that are commonly seen in those infected. Among the 21 patients examined, “frosted glass” type consolidations were observed in 12 patients (57%) and areas of consolidation in six patients (29%). In this infection, pathologic changes were highly likely to extend to more than two lobes, (15 of 21 patients, 71 %) with bilateral involvement (16 of 21 patients, 76 %). Secondary findings included rounded masses (7 of 21 patients, 33%), masses with a predominantly linear pattern (3 of 21 patients, 14%), and cobblestone symptom (4 of 21 patients, 19%). Peripheral spread of thickening was also quite common (7 of 21 patients, 33 %). Negative findings included the absence of isolated pulmonary nodules, pulmonary cavitations, pleural effusion and lymphadenopathy. Viruses are a frequent cause of respiratory infection. The radiation pattern of different viral pneumonias is different and may overlap with that of other infectious and inflammatory lung diseases. Viruses belonging to the same family lead to similar disease pathogenesis, hence CT may help to identify characteristic features and peculiarities of the radiation pattern in immunocompetent patients [8]. These preliminary data show that CT findings in 2019-nCoV infection have many similarities with the CT pattern in other coronavirus infections: SARS and MERS. The SARS and MERS outbreaks were also caused by coronaviruses, so experience from these epidemics may be useful for the current outbreak. It may also be useful to correlate the radiology findings in patients with SARS and MERS with those of patients with 2019-nCoV infection. Similarities are found in that “frosted glass” type consolidations and areas of consolidation are the first signs on chest CT. Peripheral lung involvement has also been seen in SARS and MERS. Similarly, in previous coronavirus pneumonias, the “cobblestone symptom,” defined as thickening of the interlobular septa and intralobular linear consolidations combined with “frosted glass” type consolidations, was noted in some of our patients. The absence of pulmonary cavitations, pleural effusion and lymphadenopathy noted here was also characteristic of previous SARS studies [9]. Multifocal involvement in 2019-nCoV infection was more common in our study than in earlier studies of coronavirus pneumonias (e.g., unifocal lung involvement was typical in studies for SARS) [10]. Moreover, in a previous study of CT patterns in patients with MERS, a trend toward basilar and subpleural involvement was determined [11]. However, no definite characteristic distribution was observed in patients in our group. After recovery from MERS pneumonia, marked improvement was noted, but with often residual fibrotic changes. Although it is too early to judge the changes in CT pattern in patients with 2019-nCoV infection in subacute, chronic stage or in reconvalescents, it can be assumed that the pattern of changes in dynamics will be similar [12]. What was unique in our study compared with other published data on 2019-nCoV infection was that three of our patients had a normal CT lung picture on admission. One of these patients presented with a solitary rounded thickening of the right lower lobe 3 days later, suggesting that this pattern may be the very first radiologic sign in some patients with 2019-nCoV. In another of these three patients, a control CT scan taken 4 days after the initial CT scan remained completely normal. The evaluation of negative radiologic findings in patients with confirmed 2019-nCoV infection suggests that chest CT does not have full sensitivity and cannot alone reliably and completely rule out this disease especially in the early stage. This may be due to the fact that 2019-nCoV has an incubation period of several days and may have a prodromal period with the presence of clinical but not radiologic signs. The Centers for Disease Control and Prevention notes that symptoms of 2019-nCoV infection can appear from 2 days to 2 weeks after exposure, which is similar to the incubation period of MERS [13]. Our study had several limitations. We had a relatively small number of patients, including only eight of 21 patients who underwent dynamic CT follow-up. We did not consider chest radiographs. Instead, we limited our study to chest CT scans only, as CT is probably more sensitive to early and/or mild changes, as was the case in previous outbreaks of coronavirus infections [14]. Nevertheless, chest radiography may be useful as it could potentially serve as a screening method in health facilities in high-prevalence regions with limited resources. Finally, 11 patients underwent CT with a slice thickness of 5 mm or more, in which case unexpressed changes such as early and/or minimal emphysema and tiny nodules could be missed. However, the thicker slice images provided for this review fairly accurately reflect the realities of the radiologist’s daily practice in some areas of the world affected by the outbreak. Thus, this work represents an early study of chest computed tomography findings in patients with new coronavirus infection (2019-nCoV), revealing frequent manifestations of this disease. The radiologist plays a crucial role in the rapid identification and early diagnosis of new cases, which can benefit not only the patient but also large parts of the public health surveillance and response system. It must be recognized that the CT signs of 2019-nCoV infection are similar to those of other viral pneumonias, especially within the same causative virus family (SARS and MERS). Public health measures worldwide are being updated and evolving daily to control the current outbreak. As new cases are identified, other unique radiation symptoms may be discovered in this patient population. Future studies will be able to determine how the radiation pattern of the disease changes over time after treatment. If you are interested in buying a CT scanner United Imaging Healthcare, then leave your request and our managers will contact you and provide a commercial offer with a full description of the system, as well as with a competitive price in the market of medical diagnostic equipment in Ukraine. Source of information: https://medach.pro/post/2299. Transcript: Zoya Kupfer Edited by Alice Sknar. Original version in English: https://pubs.rsna.org/doi/full/10.1148/radiol.2020200230#fig1a.