Clinical and CT imaging characteristics of COVID-19 cases in Wenzhou city: A retrospective analysis

Background: In December 2019, Wuhan witnessed the outbreak of an “unexplained pneumonia” caused by a novel coronavirus strain infection and was dubbed the COVID-19 by the WHO. The disease quickly spread to China. This study aimed to investigate the disease’s evolving epidemiological history, as well as analyze the clinical and CT imaging characteristics, treatment regimens, and patients’ prognosis. Methods: This was a retrospective study whose cases were 64 patients with a confirmed diagnosis of COVID-19. The clinical data were obtained from patients who were admitted to the isolation ward from 21 January 2020 to 19 February 2020. Results: 60 out of 64 patients had a definitive history of exposure to people who had traveled from Wuhan City. The median time from onset of symptoms to first hospital admission was 3.9±1.9 days. The initial symptoms included fever (46/64), dry cough (38/64), fatigue or myalgia (23/64), sore throat (10/64), diarrhea (3/64) along with late-onset symptoms like chest pains (2/64) and headaches (2/64). The majority of the patients (43/64) had normal white blood cell counts while 29.7 % (19/64) had leukopenia. Only two patients (3.1 %) presented with leukocytosis. 58 of the 64patients had abnormal radiological findings on chest CTs. The first chest CTs (within 2 days) was more sensitive in detecting COVID-19 infection (85.9 %) compared to the initial RT-PCR (56.3 %; p<0.01). The CTs showed lesions in multiple lung lobes in three-quarters of the patients while 15.6 % had lesions localized to one lobe. Most of the lesions were typically dense with ground-glass opacity co-existing with consolidation or cord-like shadows. Most of these patients (50/64) have recovered and got discharged giving a mean length of hospital stay of 13.5±4.8 days. Our hospital unit has not reported any COVID-19 related death so far. Conclusions: Early intervention in COVID-19 disease improves patients’ prognosis. Our data demonstrate the superiority of early radiological tests ahead

interferon-alpha (inhaled-50 μg, twice daily) along with antiviral treatment lopinavir (400 mg, twice daily) and ritonavir (100 mg twice daily), and arbidol (200 mg, twice daily). The guideline does not recommend taking more than three antiviral drugs. Based on individual clinical needs, some of the patients also received Asmeton (73 mg, three times a day) and Cheno (0.3 g, three times a day). In addition, patients whose resting respiratory rate was more than 30 breaths per minute, oxygen saturation was below 93 %, or multiple pulmonary lobes showed more than 50 % progression of disease within 48 hours on chest CT received low dose corticosteroid treatment (40-80 mg per day) and gamma globulin (15-20 g per day) for 3-5 days [8,10]. Release from medical isolation and discharge standards were as follows: 1. Temperature returns to normal for more than 3 days. 2.
Significant improvement in respiratory symptoms. 3. Improved evidence on chest radiography. 4. The results of two real-time reverse transcription-polymerase chain reaction tests taken 24 hours apart were negative for COVID-19 antigens. In addition to meeting the above four conditions, the patients who get discharged from hospital-based quarantine remain in intensive community based medical observation for 14 days after discharge [10].

Statistical analysis
We used the chi-square test to test the difference between the sensitivity of first chest CTs (early phase, 0-2 days) for COVID-19 infection and the sensitivity of the initial RT-PCR. For continuous variables, we computed means and standard deviation or medians with interquartile ranges. For categorical variables, we computed the percentages of patients in each category. All analyses were performed using the SPSS software, version 22.0.

Epidemiological characteristics
From January 21 to February 19, 2020, clinical
The majority (90.6 %, 58/64) of the patients had abnormal chest CT findings. The sensitivity of first chest CTs (early phase, 0-2 days) for COVID-19 infection was 85.9 % (55/64) compared to the sensitivity of the initial RT-PCR was 56.3 % (36/64). This difference was statistically significant (p<0.01). Interestingly, a 26 years old man came to the hospital because of diarrhea and fever ( 38.1°C), his initial RT-PCR was positive, while on three occasions his chest CT scan results were negative (the times were January 23, January 27 and February 2), and he was discharged after an 11 day stay in the hospital. The remaining 20 patients had positive chest CTs while their initial RT-PCRs were negative. The chest CTs showed that 15.6 % (10/64) of the patients had opacities only in one lobe per lung while 3.1% (2/64) had at least two lobes with opacities in each lung. Nearly three quarters (71.9 %, 46/64) of the patients had bilateral opacities and consolidation in their lungs (Table   4). A typical CT scan of COVID-19 patients showed bilateral or multiple lobular or subsegmental areas of dense, ground-glass opacity that co-existed with consolidation or cord-like shadows ( Figure 1 (Table 5).

Discussion
Coronaviruses are a class of enveloped RNA viruses that live in many places, including humans, other mammals, and birds. To date, six coronaviruses have been found to infect humans [11]. There was a serious outbreak of SARS-CoV in China between 2002 and 2003 [12,13]. MERS-CoV also led to a serious epidemic in the Middle East in 2012, and caused mass mortality [14]. Cross-species infection and occasional spillover events may lead to the recurrent emergence of new coronaviruses [15]. In December 2019, Wuhan city in Hubei province of China became an epicenter of the new coronavirus outbreak. As of 24 February 2020, the confirmed people infected with COVID-19 were 79 331, including 2069 people across 29 other countries [5]. These figures are being updated and increasing daily.
During the work of treating COVID-19 patients, we observed peculiar differences of our patients' clinical features from those reported in early cases in Wuhan city. We had 50 patients who had recovered and got discharged. According to our data, none of the infected patients at our teaching hospital had been exposed to the Wuhan city seafood wet market, instead, 29 patients had a history of familial cluster interactions which strongly suggested human to human transmission. At present, the primary source of infection appears to be from the COVID-19 patients as well as the asymptomatic ones. Respiratory and close contact transmission are the main modes of transmission.
In this vein, aerosol propagation may pose a grave risk of transmission, especially to those exposed to high concentrations of aerosols for a relatively long period of time in a confined environment [10].
Therefore, containment procedures such as quarantine, avoiding close contacts with suspected or infected people as well as mass health education, avoiding large gatherings and taking precautionary measures against exposure to COVID-19 should continue to be implemented to prevent large scale spread.
Compared with the initial patients infected in Wuhan city [6], the symptoms of patients in Wenzhou city are relatively mild. Wenzhou patients' initial symptoms included fever, coughing, fatigue or muscle pain, pharyngalgia, diarrhea and rarely chest pain and headache. Atypical initial symptoms also deserve particular attention. For example, there were 3 patients with COVID-19 who presented with diarrhea as the initial symptom at disease onset in this study. Such observations are important because they point to clinicians being alert to the potential of misdiagnosis and differential diagnosis follow up. Among the Wenzhou patients, the median time from onset of symptoms to first hospital admission was relatively shorter. The news of the epidemic was now widely publicized officially by the Chinese government, as such these patients were very concerned about their health status and hence these early health-seeking behaviors. This could be the reason why the patients in the current study had relatively mild clinical symptoms. None of them was admitted to the hospital for breathing difficulties, a sign of severe disease progression. Early intervention in the treatment of COVID-19 pneumonia increases the likelihood of better prognosis among patients. Viral nucleic acid detection using RT-PCR is the current standard. Non-contrast chest CT scans should also be considered for early diagnosis of this viral disease. Fang et al, reported that 50/51 (98%) patients had evidence of abnormal CT findings compatible with viral pneumonia at baseline while one patient had a normal CT. Such an observation was also confirmed in the current study. Of the 50 patients with abnormal CT results in Fang and colleagues' study, 36 (72%) patients had typical CT scan manifestations (e.g. peripheral, subpleural ground-glass opacities, often in the lower lobes). In addition, difference in the detection rate on the initial CT scan (50/51 [98%, 95% CI 90-100%]) was significantly greater than observed from the first RT-PCR (36/51 [71%, 95%CI 56-83%]) (p<.001) [16].
Chung et al, also reported that the chest CT scans of 3 out of 21 patients were negative for viral COVID-19 pneumonia at initial presentation [17]. On the other hand, Xie et al reported that 5 out of 167 (3%) patients had negative RT-PCR results for COVID-19 at initial presentation despite having chest CT scan findings typical of viral pneumonia [18]. In our data, the sensitivity of first chest CTs (early phase, 0-2 days) was 85.9% (55/64), the sensitivity of the RT-PCR was 56.3% (36/64) demonstrating that early chest CT scanning was more sensitive than the initial RT-PCR (p<.001).
There are some possible reasons for this: 1) the specificity and sensitivity of the detection technique itself are not good enough; 2) variation in the detection rate of kits from different manufacturers; 3) low patient viral load; or 4) improper clinical sampling [16]. We support the view that chest CT scans should be used as part of routine examination of COVID-19 Pneumonia at admission, especially when RT-PCR test results are negative.
The health statuses of patients with COVID-19 sometimes change very rapidly. This may be related to many factors such as treatment plan and patient's premorbid physique as well as preexisting chronic diseases such as hypertension and diabetes. In clinical practice, some patients do have recurrent symptoms and suddenly slide into acute respiratory distress syndrome or multiple organ failure even after a relatively stable period of time. This phenomenon is thought to be related to the "cytokine storm syndromes" in the patient's body [19]. Although there are currently no vaccines or any known special cure for this disease, the treatment regimens provided for in the guideline for treating the novel coronavirus pneumonia [10] were able to increase the number of recovering patients, prevent deaths and shorten hospital stay. A lot of pathological changes could be happening in the lesioned lungs of the patients. The temporal characteristics observed on the initial CT imaging as we all as on follow up scans COVID-19 patients is a rich data source of the novel disease's pathogenesis. In most of these discharged cases, the early stage of admission (0-5 days) chest CT usually shows Ground-Glass Opacities (GGO) and Consolidation [20]. However, with the progression of the disease, the GGO lesions gradually decrease, and consolidation increases at first and then remains relatively stable for some days. If the patient's condition gets controlled and improves, at the last stage of the disease, chest CT scans often show different degrees of pulmonary fibrosis in the lungs. This trend could help clinicians to make correct judgments which are good for the resolution of the disease. The characteristics, including temporal CT changes among COVID-19 patients, can provide a lot of important information that feeds into clinical decision making.
In the absence of effective drugs against COVID-19, the diagnosis and treatment protocol for novel coronavirus pneumonia manual continues to inform practice and would also benefit from data such as generated in this study for its future revisions. Previously, the large-dose glucocorticoids used to treat SARS caused some serious adverse reactions [21] and also failed to effectively decrease the mortality rate associated with the disease [22]. In light of this, we treated the patients with lower dose (30-80 mg/day) and for a shorter time (3-5 days) using methylprednisolone to alleviate the pulmonary exudates and inhibit a possible cytokine storm syndromes. However, the impact of such an adaptation requires further studies.

Limitations Of This Study
Our study has several limitations. Firstly, only 64 patients were included and the study was motivated by the need to timeously communicate outbreak-related information since the disease is rapidly      Figure 1 Large-scale pulmonary interstitial changes and fibrosis evolution in a patient with COVID-19.
Male, 48 years old, fever for 6 days accompanied by chills, cough expectoration, with COVID-19 confirmed patients close contact history. A: CT examination on the day of admission, an axial CT image shows two lungs scattered multiple consolidative opacities, with a striking peripheral distribution in the two lower lobes. B 8 days later, an axial CT image shows the lesions were partially absorbed and the range was reduced. C: 17 days later, an axial CT image shows the range of lesions was further reduced, two lungs scattered multiple rough strips, which suggested pulmonary fibrosis.