Clinical and CT manifestations of coronavirus disease 2019 (COVID-19): comparison of suspected cases of COVID-19 in isolation and non-COVID-19 pneumonia in a single-center study conducted in Beijing, China

Background: The clinical and CT manifestations of COVID-19 pneumonia and non-COVID-19 pneumonia in the same period have not been compared in detail. The purpose of this study is to analyze the clinical and CT manifestations of COVID-19 pneumonia and perform a comparison of those isolated patients for presumed COVID-19 infection and of non-COVID-19 pneumonia in the same period. Methods: 173 patients with pneumonia from January 1, 2020 to March 20, 2020 were retrospectively enrolled and classified into three groups: patients with COVID-19 pneumonia (Group I, N=4), patients in hospital-isolation for presumed COVID-19 pneumonia (Group Ⅱ, N=5), and patients with non-COVID-19 pneumonia (Group III, N=163). Clinical symptoms, laboratory test results and CT imaging features were compared among three groups. Results: Fever and cough were the most common clinical symptoms in the three groups. 30/163 (18.4%) patients were asymptomatic in Group III. Leukopenia, lymphocytopenia, and elevated C-reactive protein was identified in 1 (25%), 1 (25%), and 1 (25%) patient in Group I; 1 (20%), 1 (20%), and 2 (40%) patients in Group II; 10/157 (6.4%), 33/157(21.0%), and 94/136 (69.1%) patients in Group III. Demarcated GGO/mixed GGO, ill-defined GGO/mixed GGO, consolidation, centrilobular nodule, tree-in bud opacity, bilateral involvement, peripheral distribution, posterior part/lower lobe predilection was observed in 3/4 (75%), 2/4 (50%), 4/4 (100%), 2/4 (50%), 0, 3/4 (75%), 3/4 (75%), and 2/4 (50%) patients, respectively in Group I; 1/5 (20%), 5/5 (100%), 4/5 (80%), 4/5 (80%), 3/5 (60%), 4/5 (80%), 2/5 (40%), and 3/5 (60%) patients in Group Ⅱ; 1/163 (0.6%), 87/163 (54.3%), 115/163 (70.6%), 117/163 (71.8%), 95/163 (58.3%), 52/163 (31.9%), 9/163 (5.5%), and 9/163 (5.5%) patients in Group III, respectively.


Introduction
In December 2019, cases of "unknown origin" pneumonia caused by a novel coronavirus, which was designated coronavirus disease 2019 (COVID-19) by the World Health Organization (WHO), were first reported in Wuhan, China (1)(2)(3). COVID-19 was caused by a novel virus, named as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). On March 11, 2020, the WHO declared that the outbreak of COVID-19 had been characterized as a global pandemic. As of May 4th, 2020, a total of 3,565,899 infected cases and 248,367 death cases had been reported worldwide.
Owing to the lack of specific therapeutic drugs or vaccines for COVID-19, early diagnosis, isolation, and treatment have been proven to play an extremely important role in controlling the spread of the epidemic. According to the program for the diagnosis and treatment of COVID-19 pneumonia published by the National Health Commission, the diagnosis of COVID-19 must be confirmed by reverse transcription polymerase chain reaction (RT-PCR) or gene sequencing for respiratory or blood specimens (4). However, owing to limited capacity for the RT-PCR test, improper sample collection or transportation and inconsistent kit performance, the successful RT-PCR analysis for throat and nasal swab samples was initially limited (5)(6)(7)(8)(9). In contrast with RT-PCR, chest CT has a higher sensitivity for COVID-19 detection than RT-PCR test (8)(9)(10). The CT imaging characteristics of COVID-19 include GGO, consolidation, crazy-paving pattern, interlobular septa thickening with a predilection of peripheral distribution, and a tendency for bilateral involvement (9)(10)(11)(12). However, they were not specific or overlapping (11,13,14). Moreover, as the period of the COVID-19 outbreak coincides with the influenza epidemic period, many influenza virus pneumonia and community-acquired pneumonia cases have emerged during the epidemic of COVID-19 pneumonia (13)(14)(15). Recently, an increasing number of cases of asymptomatic COVID-19 infection have been reported. Therefore, distinguishing COVID-19 pneumonia from non-COVID-19 pneumonia is of particular importance, which allows for early implementation of isolation, helps avoid unnecessary quarantine, public panic, excessive consumption of medical resources, and reduces the risk of cross infection. To the best of our knowledge, few studies have focused on the differences between COVID-19 and non-COVID-19 pneumonia in the same period. Therefore, this study aimed to analyze the clinical and CT manifestations of COVID-19 pneumonia and perform a comparison of those isolated patients for presumed COVID-19 infection and of non-COVID-19 pneumonia in the same period to help early disease identification and avoid unnecessary patient isolation.

Study design
This study was approved by the Medical Ethical Management Committee of our institutional review board. As a retrospective study, the requirement for patients' informed consent was waived in accordance with the CIOMS guidelines. A total of 368 consecutive patients diagnosed with the diagnosis of "pneumonia" by chest CT scan in our institutional PACS from January 1 to March 20, 2020 were reviewed. The diagnosis of pneumonia was in accordance with the 2007 guidelines of the Infectious Diseases Society of America and the American Thoracic Society (16), and the exclusion criteria included pneumonia-like CT appearance caused by mosaic perfusion effect or a hypostasis effect, obstructive atelectasis caused by tumor, compressed atelectasis caused by pleural effusion, and other non-infectious or non-inflammatory diseases. Finally, 172 patients were enrolled in this study ( Figure 1).

Clinical Data Collection
The available clinical data, including age, sex, clinical history, laboratory findings, time course of symptoms, and travel and exposure history, were extracted from electronic medical records (Table 1).
Laboratory assessments consisted of blood leucocyte count, lymphocyte count, and C-reactive protein, and detection of common respiratory tract infection pathogens, including adenovirus, respiratory syncytial virus (RSV), influenza virus, legionella pneumophila, human parainfluenza virus (HPIV), Mycoplasma pneumoniae, and Chlamydia pneumoniae; other pathogenic agents were also recorded if detected.

Chest CT and Imaging Interpretation
All patients underwent chest CT examination in a supine position during end-inspiration without intravenous contrast material and a 2.5 mm-thick slice was imaged using a BrightSpeed scanner (GE Medical Systems, Milwaukee, WI, USA). Two thoracic radiologists (with more than 5 and 8 years of experience) blinded to the clinical data reviewed the CT images independently; in cases of disagreement, consensus was achieved by consulting with a third radiologist with 18 years of experience. All images were reviewed on both lung (width, 1500 HU; level, -400 HU) and mediastinal (width, 400 HU; level, 40 HU) settings.
The following CT features were recorded: ground-glass opacity (GGO)/mixed GGO, consolidation, interlobular septa thickening, crazy-paving pattern, centrilobular nodular, reticulation, bronchial wall thickening, tree-in-bud opacity (TIB), traction bronchiectasis, cavitation, calcification, vascular enlargement, pleural effusion, and lymphadenopathy (defined as lymph node size of ≥10 mm in the short-axis dimension). The detailed descriptions of the features mentioned above were provided by the Fleischner Society (17). GGO/mixed GGO was subdivided as demarcated or ill-defined based on the boundary. In accordance with preliminary studies (5, 18), we introduced a scoring system to semiquantitatively assess the severity of COVID-19 based on the degree of involvement of the five lung lobes: score 0, no involvement of a lung lobe; score 1, 1%-25% involvement; score 2, 26%-50% involvement; score 3, 51%-75% involvement; and score 4, 76%-100% involvement. The overall severity score of the five lung lobes ranged from to 0-20. The distribution of lesions was recorded as peripheral distribution, posterior part or lower lobe predilection, and bilateral involvement. For those patients who received follow-up chest CT, we also analyzed all the CT films of each examination.

Patient demographics
Based on the clinical symptoms, laboratory results, and CT imaging findings, we classified a total of 172 patients (95 men and 77 women; age range, 19-98 years of age; mean age, 63.6 years of age) into three groups: Group I, four patients with COVID-19 pneumonia (4 female patients, 23-78 years of age); Group Ⅱ, five patients in hospital-isolation for presumed COVID-19 pneumonia (4 male and 1 female patients, 25-67 years of age, median age, 38 years of age); and Group III, 163 patients with non-COVID-19 pneumonia (91 male and 72 female patients, 19-98 years of age, mean age, 64.1 years of age). All the five patients in Group II were isolated in our institutional fever observation department for presumed infection of COVID-19 pneumonia. With the exception of one patient who was transferred to a designated institution for COVID-19 for a definite exposure history (husband of a confirmed case of COVID-19 in Group I) who was eventually determined not to be infected, the other four patients were relieved from quarantine later with a series of negative results of RT-PCR test and improvement of lesions on follow-up CT scans. The patients' clinical data and laboratory test results are summarized in Table 1.
Note-RT-PCR: reverse transcription polymerase chain reaction; & : median age (IQR) for Group Ⅰ (N=4) and Group Ⅱ, mean age ± SD for Group Ⅲ N/A: No investigation or test performed; * : Not all patients received the test # : The number in parentheses represented total number of patients received test

Exposure history
One patient in Group II had a history of traveling to Wuhan 16 days before the onset of fever. The other patients in Group II and all four patients in Group I had unknown exposure history. As this was a retrospective study, the investigation of exposure history was not performed for patients in Group III.

CT imaging findings
The CT features of all patients enrolled in this study are listed in Table 2. Note * : The number in Group Ⅰ and Ⅱ was median and mean in Group Ⅲ . GGO: ground-glass opacity  Figure 3). In addition, ill-defined GGOs and centrilobular nodules were found in two patients with leukocytosis ( Figure 2, 3). No patient showed tree-in-bud signs. Pure consolidation with incomplete halo was observed in only one patient who was initially diagnosed with Mycoplasma pneumoniae ( Figure 4). Owing to recurrent fever during hospitalization, a second CT scan 9 days later showed obvious progression with an increased lung involvement score from 1 to 5, and the diagnosis of COVID-19 infection was made due to the positive RT-PCR results for the third time ( Figure 4).
Group III: Consolidation was observed in 117 (71.8%) patients in Group III. Of these 117 patients, air bronchogram, bronchial wall thickening, incomplete halo, and complete halo was were in 46.0%, 46.6%, 1.8%, and 7.4% of patients, respectively. Centrilobular nodules (71.8%) and TIB (58.3%) were more common than those in Group I. Ill-defined GGO was observed in 53.4% of patients and demarcated GGO, which was more common (75%) in Group I, was observed in only one patient (0.6%). Only 5.5% and 31.9% of patients showed peripheral distribution and bilateral involvement, respectively. The mean lung involvement score was 2.8, and the highest involvement score was 14, and the minimum was 1 ( Figure 10). Moreover, interlobular septa thickening, crazy-paving pattern, reticulum, pleural thickening, and pleural effusion were observed in 20.9%, 4.3%, 4.9%, 13.5%, and 14.7% of patients, respectively.

Discussion
The purpose of this study was to identify patients with COVID-19 from those patients who presented with pneumonia in CT imaging and to isolate suspected patients as early as possible and avoid unnecessary isolation. In this retrospective study, the most common clinical symptoms of Group I (COVID-19 pneumonia) were fever and cough, same to Group II (presumed COVID-19 pneumonia) and III (non-COVID-19 pneumonia). 30 (18.4%) patients in Group III were asymptomatic. Myalgia (4, 2.5%), nausea (6, 3.7%), vomiting (6, 3.7%), and diarrhea (1, 0.6%) were only observed in Group III and only one patient presented with diarrhea in Group II.
Normal or slightly decreased white blood cell count and decreased lymphocyte count are the main laboratory manifestations of early stage of COVID-19 (1,7,19,20). The incidence of lymphopenia, normal or decreased WBC count, and elevation of C-reactive protein in this study was 25% (1/4), 50% (2/4), and 75% (3/4), respectively. Three (75%) patients in Group I were initially excluded from the suspicion of COVID-19 as leukocytosis was observed in two and positive mycoplasma result observed in one.
Consolidation was considered as progression of GGO in COVID-19 (20,29). Pure consolidation, focal lesions and lesions involving single lung lobe were less common in COVID-19 (20). Bilateral lung involvement, peripheral distribution, and posterior part or lower lung lobe predilection were reported as typical distribution of COVID-19 (11,12,21,24). Bilateral multiple GGOs and patchy consolidations were observed in three patients in Group I, one patient in Group II, and 52 (31.9%) patients in Group III showed bilateral lung involvement. Only 9 (5.5%) patients showed peripheral distribution and posterior or lower lobe predilection.
Our study has some limitations. Firstly, there were only four cases of COVID-19 pneumonia. As Beijing is not an epidemic area of COVID-19 and our institution is not a designated hospital for COVID -19, there are relatively few cases of COVID-19. Secondly, there may have been some selection bias.
Exposure history and RT-PCR test were not available in Group III, and the diagnosis of non-COVID-19 pneumonia was made by clinical symptoms, elevated inflammatory biomarkers, and abnormal opacities on CT imaging (16). As a result, some patients may have been incorrectly classified. Thirdly, follow-up CT scans were not available for all patients.

Conclusions
In summary, bilateral involvement of multifocal demarcated GGO and well-defined consolidation with a peripheral distribution were typical CT manifestations of COVID-19 pneumonia. Ill-defined GGO consolidation with halo, centrilobular nodule surrounded by GGO, and tree-in-bud opacity without a peripheral distribution or bilateral involvement are preferred for non-COVID-19 pneumonia. These research findings of our study suggest that chest CT has potential in early identification of COVID-19 and implementation of isolation for appropriate case.

Ethics approval and consent to participate
This study was approved by the Medical Ethical Management Committee of our institutional review board. As a retrospective study, the requirement for patients' informed consent was waived in accordance with the CIOMS guidelines.

Consent for publication
Not applicable.

Availability of data and materials
The datasets during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding:
This work was supported by the National Natural Science Foundation of China (NSFC) (Grant No.

Author contributions
Guarantors