Clinical Features and Outcomes in Imported and Non-Imported Patients With COVID-19: A Retrospective, Multi-centre, Descriptive Study

Background: Global outbreak of novel coronavirus 2019 (COVID-19) continues to constitute a public health emergency of international concern. However, few studies have directly compared the differences of clinical features and outcomes between imported and non-imported COVID-19 patients. We aimed to directly compare the differences of clinical features and outcomes between imported and non-imported COVID-19 patients. Methods: we enrolled 275 patients (53 imported and 222 non-imported cases) with laboratory-conrmed COVID-19 from four designated hospitals in Chongqing, China. Epidemiological, demographic, clinical, laboratory, treatment, and outcome data were collected and analyzed. Independent samples t- test, Mann-Whitney U test, Wilcoxon test,χ 2 test, or Fisher’s exact test were adopted to compare the difference between imported and non-imported groups where appropriate Results: The imported group had a higher proportion of asymptomatic patients (18.9% vs. 3.8%, p=0.016), lower lymphocyte count (1.1×10 9 /L vs. 1.5×10 9 /L, p=0.018), higher neutrophil-to-lymphocyte ratio (NLR) (2.6 vs. 2.3, p=0.0237), higher proportion of CD4 + T cells reduction (70% vs. 44.6%, p=0.043), longer duration of viral shedding (20 days vs. 18 days, p = 0.0416) than the non-imported group. The symptomatic non-imported group had a lower lymphocyte count (1.3×10 9 /L vs. 1.6×10 9 /L, p=0.003), lower CD4 + and CD8 + T cells, longer viral shedding (20 days vs.13 days, p< 0.001) and hospital stay (17 days vs.14 days, p=0.015) than the asymptomatic non-imported group. Furthermore, when compared with imported group, asymptomatic non-imported group showed higher lymphocyte count (1.6×10 9 /L vs. 1.1×10 9 /L, p< 0.001) and CD4 + T cells(567 vs.375, p=0.029), while the symptomatic non-imported group showed no difference in lymphocyte count, CD4 + or CD8 + T cells (p ≥ 0.0894). Conclusion: Our results indicated that when compared with imported patients, damage to the immune system by SARS-Cov-2 in non-imported patients was milder, and this might be mainly due to its higher asymptomatic proportion.


Introduction
The ongoing pandemic of novel coronavirus 2019 (COVID-19) has led to more than 6,064,439 cases and 370,248 deaths globally by the end of May, 2020. According to data from China National Health Commission (NHC) daily report[1], the mortality rates of COVID-19 in Wuhan and non-Wuhan areas are quite different. Wuhan, the capital of Hubei Province, China, where the rst Chinese COVID-19 patient has been reported, has a substantially higher mortality rate (2.9%) than that in the rest of China (0.4%) as of February 11, 2020 [2]. Despite that a high mortality rate in Wuhan was attributed in part to the lack of medical resource available at the time of immediate need, it has been reported that patients with COVID-19 in non-Wuhan areas exhibited mild or moderate symptoms, as compared with cases in Wuhan [3][4][5]. . {J, 2020 #3;XY, 2020 #12} Thus, it is speculated that the discrepancy in mortality rate between Wuhan and non-Wuhan areas may also be due to disease differences at these areas. Currently, epidemiological and clinical characteristics of patients with COVID-19 have been reported [5][6][7][8]{N, 2020 #6;C, 2020 #7}{N, 2020 #6;C, 2020 #7}. However, few studies have been conducted to directly compare the differences in COVID- 19 patients between Wuhan and non-Wuhan areas.
Since the available medical resources and treatment options vary greatly in different areas, it is of great signi cance to investigate the clinical characteristics and outcomes between imported and non-imported cases in a designated city. As the largest municipality adjacent to Hubei, Chongqing has been predicted to be the next outbreak city because of its close interaction with Wuhan. According to the mathematical model deduction published on the Lancet, as many as 150,000 new cases would be con rmed every day in Chongqing at the height of the epidemic [9]. It has estimated that about ve million people had already left Wuhan before the city's lock down, and Chongqing ranked at the second top destination city [10]. Therefore, Chongqing is a good model to investigate the clinical characteristics and outcomes in imported and non-imported COVID-19 cases, which could provide important information to further understand the disease and provide important information to improve the prevention and control strategies for this disease.
A total of 275 patients with laboratory-con rmed COVID-19 were enrolled in this study from four designated hospitals. According to epidemiological characteristics, they were classi ed into imported cases and non-imported cases. The clinical, laboratory and radiological characteristics, as well as the treatment and clinical outcomes were compared between the two groups.

Study design and participants
Our multi-center retrospective study included 275 laboratory-con rmed COVID- 19 [11], a con rmed case of COVID-19 was de ned as positive for SARS-Cov-2 nucleic acid by real-time uorescent RT-PCR for respiratory or blood specimens.
We obtained the medical records and compiled data from January 15 to March 1. The data cutoff for the study was March 1, 2020. This study is approved by Ethics Committees of the four hospitals mentioned above and written informed consent was obtained from each enrolled patient.

Grouping
It has been reported that the incubation period of the SARS-CoV-2 could be as long as 27 days [12]. The included patients were divided into the imported group (with a history of travel or residence in Wuhan city within 30 days before the onset of the disease) and the non-imported group (with no history of travel or residence in other cities 30 days before the onset of the disease). A total of 275 patients were included in our study, including 53 imported patients and 222 non-imported patients.

Data collection
The collected information including epidemiological, demographic, clinical, laboratory, treatment, and outcome data extracted from electronic medical records. All the clinical information and data during their hospitalization were followed up till Mar 1, 2020. During the process of data collection, we communicated with the attending physicians and other nursing staff directly if there was requirement of veri cation or loss of data. All the data were examined by two physicians (Zhu Xiaoqian and Yuan Guodan). As described previously, the incubation period was de ned as the time from exposure to the onset of illness, which was estimated among patients who could provide the exact date of close contact with individuals with con rmed or suspected COVID-19 [5]. Viral shedding duration was de ned as the time from illness onset to the rst day of continuous negative tests with intervals of at least 24 hours [13].The illness severity of COVID-19 was de ned according to the Chinese clinical guidance for COVID-19 pneumonia diagnosis and treatment (7th edition) [11].

Statistical analysis
All statistical analyses were performed with SPSS 22.0 software program and p values < 0.05 were considered signi cantly. Continuous variables were presented as mean (SD) if they were normally distributed or median (IQR) if they are not, and categorical variables were presented as count (%).We used the independent samples t-test, Mann-Whitney U test, Wilcoxon test,χ 2 test, or Fisher's exact test to compare the difference between imported and non-imported groups where appropriate.

Epidemiological clinical features
The demographic and clinical characteristics of the patients are shown in Table 1. 275 patients with laboratory-con rmed COVID-19 were enrolled in this study, including 53(19.3%) imported cases and 222(80.7%) non-imported cases. Among them, 133 were male and 142 were female, with age ranging from 10 to 86 years and an average age of 47.4 years. The proportion of males patients in the imported group was higher than that in the non-imported group (62.3% vs. 44.6%, p= 0.015). The age distribution in the imported and non-imported groups were similar (47.2±14.4 vs. 47.5±16.8, p=0.101). In total, 62 of the 275 patients (22.6%) had at least one coexisting illness (e.g., hypertension or coronary heart disease). Except that non-imported group had a higher proportion of coronary heart disease of 14.4% (32/222) than that in the imported group (1.9%, 1/53, p=0.011), there was no signi cant difference between these two groups.
In terms of clinical classi cation, 13 (24.5%) patients were severe type including 6 (11.3%) critically ill cases in the imported group. In the non-imported group, 50 (22.5%) patients were severe type including 22(9.9%) critically ill cases, according to the Chinese clinical guidance for COVID-19 pneumonia diagnosis and treatment (7th edition) [11]. No statistical difference was observed in the proportion of severe or critical illness between the two groups (p=0.523). The median incubation period was eight days in the non-imported group and six days in the imported group, both were longer than four days reported previously [14,15]. Although the incubation period of the non-imported group was slightly longer than that of the imported group, there was no statistical difference between the two groups (p=0.334).The median time from symptoms onset to hospital admission are 4 days in both imported and non-imported groups.
The most common symptoms at illness onset in the imported group were cough and fever (both were 58.5%), fatigue and sore throat (both were 20.8%), diarrhea (13.2%) and headache (11.3%). The most common symptoms in the non-imported group were cough (54.1%), fever (36%), and fatigue (12.7%). When the two groups were compared, the proportion of asymptomatic patients in the non-imported group was signi cantly higher than that in the imported group (18.9% vs. 3.8%, p=0.016), and the proportions of symptoms of fever, sore throat, hemoptysis and diarrhea were all lower than those in the imported group (p≤0.048). Table 2 shows the radiologic and laboratory ndings. Of 218 computer tomography (CT) scans that were performed at the time of admission, 78.9 % revealed abnormal results including ground-glass opacity, local patchy shadowing, bilateral patchy shadowing and interstitial abnormalities. The most common patterns on chest CT in both imported and non-imported cases were bilateral patchy shadowing and local patchy shadowing. The abnormal patterns of two or more types were 33.3% in the imported group and 37.4% in the non-imported group. In general, there was no statistical difference in the proportion of the imported group and the non-imported group for each image performance (p≥ 0.175).

Radiologic features
As to asymptomatic patients, 33.3% had no abnormal chest CT images. The most common patterns on chest CT were ground-glass opacity (20 %) and local patchy shadowing (15.6%). The proportion of bilateral patchy shadowing and interstitial change was 6.7% and 4.4%, respectively. 22.2% of the patients had two or more of the above abnormal images on chest CT. When compared with the symptomatic patients, the ground-glass opacity was more common in asymptomatic patients (20% vs. 5.2%, p=0.004), while the proportions of bilateral patchy shadowing (6.7% vs. 22.5%) and multiple manifestations (22.2% vs. 39.9%) were lower (p≤0.02).

Laboratory ndings
Lymphocytopenia was presented in 59.6% of the enrolled patients, and the proportion of lymphocytopenia in the imported group and the non-imported group was 69.8% and 57.2%, respectively.
The difference was not statistically signi cant (p=0.093). However, the absolute value of lymphocyte count in the imported group was signi cantly lower than that in the non-imported group (1.1×10 9 /L [IQR 0.9-1.6] vs. 1.5×10 9 /L [IQR 1.0-1.8], p=0.018). 35.6% of the patients had increased C-reactive protein (CRP). The proportion of increased C-reactive protein in the imported group was 49.1%, which is higher than that (32.4%) in the non-imported group (p=0.02). The neutrophil-to-lymphocyte ratio (NLR) of the imported group was 2.6 [IQR 2.0-3.6], which is higher than that of the non-imported group (2.3 [IQR 1.5-3.6]) (p=0.0237). In total, 65.5% of the patients had elevated levels of activated partial thromboplastin time (APTT). 34.2% had elevated erythrocyte sedimentation rate (ESR), 24% had elevated Lactate dehydrogenase (LDH), and 23.3% had elevated procalcitonin (PCT). Less common were elevated levels of D-dimer, alanine aminotransferase (ALT), aspartate aminotransferase (AST), prothrombin time (PT) and creatine kinase (CK). There was no statistical difference between the two groups in all of these indicators (p≥0.065).
Furthermore, we investigated the CD4 + and CD8 + T cells between imported and non-imported groups. We found that the proportion of CD4 + T cells reduction in the imported group was higher than that in the nonimported group (70% vs. 44.6%, p=0.043), while the count of CD4 + T cells was not different between the imported group and the non-imported group (375/μL [IQR 274-509.5] vs. 509 /μL[IQR 281-612], p=0.15), although the imported group indeed had a lower median value. The difference in proportion of CD8 + T reduction or the count of CD8 + T cells between the imported and non-imported groups was not statistically signi cantly different (p≥0.275).
As to asymptomatic and symptomatic patients, higher lymphocyte count as well as CD4 + T and CD8 + T lymphocyte was found in asymptomatic patients compared with symptomatic patients (p≤0.029).
As of March 1, 2020, all of the 53 patients in the imported group were discharged from the hospital, while 188 of the 222 patients in the non-imported group were discharged, and no patients had died. The median duration of hospitalization was 17 days (IQR 11.8-24.3) in the imported cases and 15 days (IQR 12-20) in the non-imported cases. There was no statistical difference in the two groups. Median duration of viral shedding was 20 days (IQR 13.8-27) in the imported cases, which was longer than that of 18 days in the non-imported cases (IQR 12-23) (p = 0.0416).
Furthermore, the median duration of viral shedding in symptomatic patients was 20 days, which was longer than that of 13days (p<0.001) in the asymptomatic patients , and the median duration of hospital stay in the symptomatic patients was longer than that in the asymptomatic patients (17 days vs. 14 days p=0.015).

Discussion
It was found in our retrospective study that, compared with the imported cases, the non-imported cases displayed higher lymphocyte count, lower proportion of CD4 + T cells reduction, lower NLR value, shorter duration of viral shedding, but more atypical symptoms and a much higher asymptomatic proportion. As to asymptomatic non-imported patients, they had higher lymphocyte count including both CD4 + T and CD8 + T lymphocyte, shorter duration of viral shedding and length of hospital stay when compared with symptomatic non-imported group. Further subgroup analysis suggested that counts of lymphocyte and CD4 + T cell from non-imported asymptomatic group were higher than those in the imported patients, but there were no difference between non-imported symptomatic and imported patients.
Lymphocytopenia is a dominant feature of patients with COVID-19 and the decrease of lymphocyte count was found to be more apparent in severe group compared with non-severe group [16]. It has been suggested currently that the severity of lymphocytopenia re ects the severity of SARS-CoV-2 infection, serving as a key indicator for disease progression and outcome [17]. However, there was no signi cant difference between the imported and non-imported groups in the proportion of severe type or critically ill type in our study. We speculated that this might be due to the relatively better general condition in imported cases, while the critically ill or frail patients were trapped in Wuhan due to their intolerance of long-distance travel. It is consistent with the fact that the proportion of severe cases in our imported group was much smaller than that previously reported in Wuhan area [7], which is consistent with previous reports [3,18].
Lymphocytes and their subsets play a decisive role in maintaining immune homeostasis and in ammatory response throughout the body [17], which was highly involved in the pathological process of COVID-19 [19]. It has been reported that the immune responses induced by SARS-CoV-2 infection are two phase: immune defense-based protective phase and the second in ammation-driven damaging phase. Once a protective immune response was damaged, virus propagation and extensive tissues destruction occurred and a wide range of symptoms developed. Previous results have shown that SARS-Cov-2 mainly damaged T lymphocytes, especially CD4 + T lymphocytes [19,20]. Our ndings of higher lymphocyte count and lower proportion of CD4 + T cells reduction in the non-imported group suggested that damage to the immune system in non-imported patients was milder. It may explain, at least in part, the observation in our study that the non-imported group has lower NLR and shorter duration of viral shedding than those in the imported group. Furthermore, higher lymphocyte count including both CD4 + T and CD8 + T lymphocyte was found in asymptomatic non-imported COVID-19 patients indicated a stronger immune defense in these patients, which might contribute to the shorter duration of viral shedding and the shorter length of hospital stay.
The neutrophil-to-lymphocyte ratio (NLR), a well-known marker of systemic in ammation and infection, has been studied as a prognostic indicator for patients suffering from various diseases. It has been indicated that high NLR was associated with high levels of in ammation [21]. Systemic in ammation is associated with the development and progression of COVID-19. Higher NLR value has been demonstrated in severe COVID-19 cases when compared with the non-severe cases [19]. NLR has been indicated as an independent prognostic biomarker affecting disease progression in COVID-19 [19]. Thus, the higher NLR value in the imported group indicated a more serious in ammatory response.
Our study found that the median duration of viral shedding was shorter in the imported group. Viral load, virulence, immune response, therapeutic regimen are the main factors that affect the duration of viral shedding [22][23][24][25][26]. However, so far, there is no report of any clinically approved antiviral drugs that are effective against COVID-19[27-29]. In addition, our results displayed no signi cant difference in treatment strategy or the interval from symptoms to admission between the imported and non-imported groups. Previous research has shown that the lower the lymphocyte count is, the higher the viral load is [30]. A recent study has found that the SARS-CoV-2 RNA load in COVID-19 patients of nasopharyngeal was negatively correlated with lymphocyte count [31]. Therefore, we speculated that the duration of viral shedding varied between the imported and the non-imported groups was mainly due to difference in lymphocyte count in these groups.
The asymptomatic proportion was 18.9% in the non-imported group, which was signi cantly higher than that in the imported group (3.8%). Furthermore, only 36% of the patients in the non-imported group had fever, signi cantly lower than the 58.5% in the imported group. Meanwhile, the incidences of sore throat, haemoptysis and diarrhea in the non-imported group were also lower than that in the imported group, which suggested that atypical symptoms were more common in the non-imported group. Thus, symptombased screening alone would fail to detect a high proportion of infectious cases. Also, 33.3% of asymptomatic patients had no abnormal chest CT ndings. Thus, active contact tracing, strict health monitoring coupled with nucleic acid testing as well as serological test when necessary should be highlighted to identify asymptomatic infections.
It is noted that this study has some limitations. First, due to the retrospective study design, not all laboratory tests were done in all patients, including cytokine. Some difference in clinical features might be underestimated. Second, interpretation of our ndings might be limited by the sample size. Despite these limitations, this study is, to our knowledge, the largest case series to date of direct comparison between imported and non-imported patients as well as symptomatic and asymptomatic patients. It is an extended and a more in-depth investigation of the previous report.

Conclusion
When compared with imported patients, damage to the immune system by SARS-Cov-2 in non-imported patients was milder, and this might be mainly due to its higher asymptomatic proportion.

Consent for publication
Not applicable.

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

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