Since January 2020 Elsevier has created a COVID-19 resource centre with

free information in English and Mandarin on the novel coronavirus COVID-

19. The COVID-19 resource centre is hosted on Elsevier Connect, the
company's public news and information website.

Elsevier hereby grants permission to make all its COVID-19-related

research that is available on the COVID-19 resource centre - including this
research content - immediately available in PubMed Central and other
publicly funded repositories, such as the WHO COVID database with rights
for unrestricted research re-use and analyses in any form or by any means
with acknowledgement of the original source. These permissions are
granted for free by Elsevier for as long as the COVID-19 resource centre
remains active.
 Journal of Infection 80 (2020) 656–665

Contents lists available at ScienceDirect

Journal of Infection
journal homepage: www.elsevier.com/locate/jinf

Clinical characteristics of coronavirus disease 2019 (COVID-19) in

China: A systematic review and meta-analysis
Leiwen Fu a,1, Bingyi Wang a,b,c,1, Tanwei Yuan a,1, Xiaoting Chen f,1, Yunlong Ao f,1,
Thomas Fitzpatrick d, Peiyang Li a, Yiguo Zhou a, Yi-fan Lin a,g, Qibin Duan h,i, Ganfeng Luo a,
Song Fan e, Yong Lu e, Anping Feng a, Yuewei Zhan a, Bowen Liang a, Weiping Cai f,
Lin Zhang l,m, Xiangjun Du a, Linghua Li f,∗∗, Yuelong Shu a,∗∗, Huachun Zou a,i,j,k,∗
a
School of Public Health (Shenzhen), Sun Yat-sen University, Shenzhen 510080, China
b
State Key Laboratory of Food Nutrition and Safety, Tianjin University of Science & Technology, Tianjin, China
c
College of Food Science and Engineering, Tianjin University of Science & Technology, Tianjin, China
d
Department of Internal Medicine, University of Washington, Seattle, Washington, USA
e
School of Public Health, Sun Yat-sen University, Guangzhou, China
f
Guangzhou Eighth People’s Hospital, Guangzhou Medical University, Guangzhou, China
g
School of Mathematical and Physical Sciences/Statistics, The University of Newcastle, Callaghan, Australia
h
School of Mathematical Sciences, Queensland University of Technology, Brisbane, Australia
i
Kirby Institute, University of New South Wales, Sydney, Australia
j
Shenzhen Center for Disease Control and Prevention, Shenzhen, China
k
School of Public Health, Shanghai Jiao Tong University,Shanghai, China
l
Department of Anesthesia and Intensive Care and Peter Hung Pain Research Institute, The Chinese University of Hong Kong, Hong Kong, China
m
Department of Medicine and Therapeutics, The Chinese University of Hong Kong, Hong Kong, China

a r t i c l e i n f o s u m m a r y

Article history: Objective: To better inform efforts to treat and control the current outbreak with a comprehensive char-
Accepted 15 March 2020 acterization of COVID-19.
Available online 10 April 2020
Methods: We searched PubMed, EMBASE, Web of Science, and CNKI (Chinese Database) for studies pub-
Keywords: lished as of March 2, 2020, and we searched references of identified articles. Studies were reviewed for
COVID-19 methodological quality. A random-effects model was used to pool results. Heterogeneity was assessed
Clinical characteristics using I2 . Publication bias was assessed using Egger’s test.
Meta-analysis Results: 43 studies involving 3600 patients were included. Among COVID-19 patients, fever (83.3% [95%
Systematic review CI 78.4–87.7]), cough (60.3% [54.2–66.3]), and fatigue (38.0% [29.8–46.5]) were the most common clinical
symptoms. The most common laboratory abnormalities were elevated C-reactive protein (68.6% [58.2–
78.2]), decreased lymphocyte count (57.4% [44.8–69.5]) and increased lactate dehydrogenase (51.6% [31.4–
71.6]). Ground-glass opacities (80.0% [67.3–90.4]) and bilateral pneumonia (73.2% [63.4–82.1]) were the
most frequently reported findings on computed tomography. The overall estimated proportion of severe
cases and case-fatality rate (CFR) was 25.6% (17.4–34.9) and 3.6% (1.1–7.2), respectively. CFR and labora-
tory abnormalities were higher in severe cases, patients from Wuhan, and older patients, but CFR did not
differ by gender.
Conclusions: The majority of COVID-19 cases are symptomatic with a moderate CFR. Patients living in
Wuhan, older patients, and those with medical comorbidities tend to have more severe clinical symptoms
and higher CFR.
© 2020 The British Infection Association. Published by Elsevier Ltd. All rights reserved.

Introduction

In December 2019, a cluster of pneumonia cases of unknown

∗
Corresponding author at: School of Public Health (Shenzhen), Sun Yat-sen Uni- cause appeared in Wuhan, China.1 The National Health Commis-
versity, Shenzhen 510080, China.
∗∗
sion (NHC) of the People’s Republic of China later announced that
Corresponding author.
E-mail addresses: [email protected] (L. Li), [email protected] (Y. Shu),
a novel coronavirus, now named COVID-19 by the World Health
[email protected] (H. Zou). Organization (WHO),2 was responsible for the outbreak.3 High-
1
These corresponding authors contributed equally to the manuscript throughput sequencing identified COVID-19 as a betacoronavirus.

https://doi.org/10.1016/j.jinf.2020.03.041
0163-4453/© 2020 The British Infection Association. Published by Elsevier Ltd. All rights reserved.
 L. Fu, B. Wang and T. Yuan et al. / Journal of Infection 80 (2020) 656–665 657

This novel virus is genetically similar to bat coronaviruses, and Data analysis
shares about 79% and 50% of its genetic sequence with the coron-
aviruses responsible for severe acute respiratory syndrome (SARS) Four authors (TY, XC, BW, and LF) independently extracted rele-
and Middle East respiratory syndrome (MERS), respectively.4 Al- vant information, including first author, publication time, study de-
though epidemiological evidence suggests most of the initial pa- signs, city, number of COVID-19 patients, mean or median age of
tients were exposed to the Huanan Seafood Market in Wuhan, the patients, maximum follow-up duration (days), history of exposure
animal source of COVID-19 has not yet been identified.1 Human-to- in Wuhan, smoking history, diagnostic criteria of COVID-19, pres-
human transmission is now responsible for most new infections, ence of medical comorbidities, clinical symptoms, radiologic find-
including those among family members and health care work- ings, laboratory findings, complications, supportive treatment, and
ers.5–7 clinical outcome of COVID-19 patients. We also extracted the origi-
Pneumonia caused by 2019-nCOV, known as COVID-19, is of nal author’s guidelines for defining severe case and screened them
huge global concern, with confirmed cases in 34 Chinese provinces according to Guidelines of Diagnosis and Treatment Of COVID-19
and nearly 30 countries across five continents. The WHO’s Interna- (Sixth Edition) from the NHC.8 We classified patients admitted to
tional Health Regulations Emergency Committee declared this out- intensive care units (ICU) as severe cases when authors did not re-
break constitutes a Public Health Emergency of International Con- port diagnostic criteria for disease severity. Studies that only re-
cern (PHEIC) on 30 January 2020.2 As of 2 March 2020 the cumu- ported data for critically ill patients were excluded in the overall
lative number of confirmed cases and deaths of COVID-19 in China meta-analysis but were included in the meta-analysis restricted to
has reached 80,302 and 2947, respectively. Outside of China, a total severe cases.
of 10,449 cases have been confirmed, including 170 deaths.8 We used the quality assessment tool for case series studies
Only one published systematic review and meta-analysis sum- published by the National Institutes of Health (NIH) to assess the
marized clinical characteristics of COVID-19.9 It reported a case- methodological quality of included studies.53 We scored 0 or 1
fatality rate (CFR) of 4.3% and that fever, sore throat, and mus- point for each item according to the criteria and added scores for
cle soreness or fatigue were the most common symptoms. In all items to generate an overall quality score that ranged from 0
that review the incidence of abnormal chest computer tomogra- to 9. Based on the overall score, we classified studies as low (≥7),
phy (CT) was 96.6%. However, this article analysed results from moderate (5–6), or high risk of bias (≤4). Any disagreement was
only ten studies, including one Chinese Center for Disease Control resolved through discussion by all investigators.
and Prevention (CDC) report that provides epidemiological data We performed data analyses using meta packages in R (ver-
only, and four preprint articles (one was already withdrawn) that sion 3.6.0). Random-effects meta-analysis was used to calculate
are not peer reviewed.10 This article failed to report any clini- pooled estimated prevalence with 95% confidence intervals of clini-
cal laboratory findings, treatments and geographical distribution of cal symptoms, laboratory findings, chest CT findings, complications,
COVID-19 which are essential to a thorough understanding of clin- treatment, and fatality of COVID-19 patients.54 To minimize the
ical characteristics. Many cases have emerged inside and outside impact of studies with extremely small or extremely large preva-
Wuhan over the past month.1 , 5 , 6 , 11–50 Recent publications suggest lence estimates on overall estimates, Freeman-Tukey double arc-
there may be significant differences between clinical outcomes for sine transformation was used to stabilize the variance of specific
COVID-19 between patients inside and outside Wuhan. Xu, et al. prevalence rates before using random-effects meta-analysis mod-
found that patients outside of Wuhan experienced milder illness els to pool data.54
and less pronounced laboratory abnormalities compared to coun- We assessed heterogeneity between studies using I2 , with val-
terparts inside Wuhan.24 ues of 25%, 50%, and 75% representing low, moderate, and high
Although the number of COVID-19 cases continues to grow heterogeneity, respectively.55 If substantial heterogeneity (I2 >75%)
worldwide, little attention has been paid to summarizing the clin- was detected, we further explored the possible source of hetero-
ical signs, risk factors, laboratory and chest CT findings, compli- geneity through subgroup analysis and used the following group-
cations, and treatments of COVID-19. We performed a systematic ing variables: age, sex, region, and underlying medical comorbidi-
review and meta-analysis to provide a comprehensive characteri- ties. We also performed subgroup analyses to explore whether the
zation of COVID-19 to better inform efforts to treat and control the prevalence of outcomes differed by these subgroups. If a meta-
current outbreak. analysis included more than three studies, publication bias was as-
sessed by Egger’s test.56
Methods
Results
Search strategy and selection criteria
Our search produced 2247 publications. Of these, 1648 were
Our systematic review and meta-analysis was undertaken ac- unique records, from which 1434 records were excluded after
cording to PRISMA and MOOSE guidelines.51 , 52 We searched four screening their titles and abstracts (Fig. 1). We assessed the eli-
databases, PubMed, EMBASE, Web of Science and CNKI (Chinese gibility of 214 full-text papers, of which 99 did not report origi-
Database), to identify studies reporting COVID-19. Articles pub- nal data, 47 did not report clinical features of COVID-19 (e.g., epi-
lished on or before March 2, 2020 were eligible for inclusion. demiological characteristics, mathematical models, virus structure),
We used the following search terms: “coronavirus” or “nCoV” or six did not include clear diagnostic criteria, 17 had a sample size
“SARS-CoV-2 or “COVID-19 . References of all retrieved studies smaller than four, two were conducted outside mainland China,
were screened for additional eligible publications. Primary stud- and one focused on patients aged less than one year. After ex-
ies were eligible if they reported any information on COVID-19 pa- cluding these studies, 43 eligible studies with 3600 patients were
tients in China without restriction on study type or study design. included. Among included studies, one study only reported data
We excluded studies that focused on infection in infants, did not on critically ill patients and was excluded from the overall meta-
report original data or clear diagnostic criteria, and no reliable clin- analysis but was included in the meta-analysis restricted to pa-
ical data as well as research outside mainland China. tients with severe illness.1 , 5 , 6 , 11–50
Two independent reviewers (LF and BW) screened the litera- Table 1 summarizes characteristics of included studies. Included
ture search and assessed each study for inclusion. Any disagree- studies were published between 24 January 2020 and 28 Febru-
ment was solved by consulting a senior investigator (HZ). ary 2020, among which 25 (58.1%) were in Chinese and the
 658
Table 1
Characteristics of studies reporting clinical characteristics of COVID-19.
Study Publication Enrolment Maximum Duration Study City No. of Diagnosis Age (median/ Males (%) Traveled to No. Family Current Health Underlying diseases Severe Diagnosis
date duration follow-up between design cases method mean [range/ or resident -cluster Smokers workers (%) Cases (%) of severity
duration onset of (RCS/SD/PS) IQR], years) of Hubei (family) (%)
(days) symptoms Province
and (%)
hospitalizati–
on (median
[range],
days)
Hyper Diabetes Cancer (%) Chronic Having any
tension (%) (%) respiratory coexisting
/lung medical

L. Fu, B. Wang and T. Yuan et al. / Journal of Infection 80 (2020) 656–665

diseases condition
(%) (%)
Guan et al Feb-06 NA NA NA PS Multi-city∗ 1099 L 47† (35–58) 640 (58.2) 676 (61.5) NA 137 (12.4) 32 (2.9) 164 (15.0) 81 (7.4) 10 (0.9) 12 (1.1) 255 (23.2) 173 (15.7) ATS
Chang et al Feb-07 NA NA NA RCS Beijing 13 NA 34† (34–48) 10 (77.0) NA NA NA NA NA NA NA NA NA NA NA
Zhang et al Feb- Jan 18 -Feb NA NA RCS Beijing 9 L 36 (15–48) 5 (55.0) 7 (78.0) 2 NA 1 (11.0) NA 1 (11.0) 0 NA NA NA NA
3
Yu et al Feb-17 Jan 21 NA NA RCS Beijing 40 NA 40 (21–57) 26 (65.0) NA NA NA NA NA NA NA NA NA NA NA
Zhuang et al Feb-19 Jan 1 -Feb 49 NA RCS Beijing 26 L 39.77† (3–79) 18 (77.0) 14 (54.0) NA NA NA 4 (15.0) 3 (12.0) NA NA 9 (35.0) NA NA
18
Li et al Feb-10 Jan 22 -Feb 20 NA RCS Dazhou 17 L 45 (22–65) 9 (53.0) 11 (65.0) NA 3 (18.0) NA 1 (6.0) 0 0 0 3 (18.0) NA NA
10
Chung et al Feb-06 Jan 18 -Jan NA NA RCS Guangzhou 21 L 51† (29–77) 13 (62.0) 18 (86.0) NA NA NA NA NA NA NA NA NA NA
27
Zhang et al Feb-19 Jan 19 -Feb 17 NA RCS Nanjing 42 L 43.02† 23 (55.0) 23 (55.0) 5 NA NA NA NA NA NA 5 (12.0) 0 NA
5 (19–96)
Wang et al Jan-30 Jan 21 -Jan 14 4 (1–11) RCS Shanghai 4 L 47.5 (19–63) 3 (75.0) 3 (75.0) NA NA NA NA 0 0 0 1 (25.0) 2 (50.0) NA
24
Song et al Feb-02 NA NA NA RCS Shanghai 51 NA 49 (16–76) 25 (49.0) 50 (98.0) NA NA NA 1 (2.0) 3 (6.0) NA 1 (2.0) NA NA NA
Lu et al Feb-3 NA NA NA RCS Shanghai 50 L 50 (NA) 28 (56.0) 37 (74.0) NA NA NA 8 (16.0) 3 (6.0) NA 4 (8.0) 18 (36.0) NA NA
Chan et al Jan-24 Jan 10 -Jan 14 7 (6–10) RCS Shenzhen 6 L 50 (10–66) 3 (50.0) 5 (83.3) 1 NA NA 2 (33.0) 1 (17.0) 1 (17.0) 1 (17.0) 4 (67.0) NA NA
15
Liu et al Feb-09 Jan 11 -Jan 10 8.5 (5–16) RCS Shenzhen 12 L 63 (10–66) 8 (67.0) 11 (91.7) 2 NA NA 3 (25.0) 2 (16.7) 0 1 (8.0) 7 (58.0) 5 (42.0) Guidelines
20
Wang et al Feb-07 Jan 1 -Jan 34 7 RCS Wuhan 138 L 56 (22–92) 75 (54.3) 138 (100.0) NA NA 40 (29.0) 43 (31.2) 14 (10.1) 10 (7.2) 4 (2.9) 61 (44.2) 36 (26.1) ICU
28
Huang et al Jan-24 Dec 16 -Jan 37 7 (4–8) PS Wuhan 41 L 49 (41–58)‡ 30 (73.0) 41 (100.0) 1 3 (7.3) NA 6 (14.6) 8 (19.5) 1 (2.4) 1 (2.4) 13 (31.7) 13 (31.7) ICU
2
Liu et al Jan-24 Jan 10 -Jan 15 7 (1–20) RCS Wuhan 137 L 57 (20–83) 61 (44.0) 137 (100.0) NA NA NA 13 (10.0) 14 (10.0) 2 (2.0) 2 (2.0) NA NA NA
15
Li et al Feb-09 NA NA NA SD Wuhan 425 L 59 (15–89) 240 (56.0) 21 (50.0) NA NA 15 (4.0) NA NA NA NA NA NA NA
Chen et al Jan-29 Jan 1-Jan 25 NA RCS Wuhan 99 L 55.5 (21–82) 67 (68.0) 49 (49.0) 1 NA NA 0 13 (13.0) 1 (1.0) 1 (1.0) 50 (51.0) 23 (23.0) ICU
20
Pan et al Feb-6 Dec 30 -Jan31 NA RCS Wuhan 63 L 44.9† (NA) 33 (52.0) 63 (100.0) NA NA NA NA NA NA NA NA NA NA
31
Pan et al Feb-13 Jan 12-Feb 26 NA RCS Wuhan 21 L 40 (25–63) 6 (29.0) 21 (100.0) NA NA NA NA NA NA NA NA 0 NA
6
Chen et al Feb-4 Jan 14 -Jan NA NA RCS Wuhan 29 NA 56 (26–79) 21 (72.0) 29 (100.0) NA 2 (7.0) NA 8 (28.0) 5 (17.0) 1 (3.0) NA 16 (55.0) 14 (48.0) Guidelines
29
Gong et al Feb-18 Dec 20 -JanNA NA RCS Wuhan 33 L 51 (23–79) 13 (39.0) 33 (100.0) NA NA NA NA NA NA NA NA NA NA
22

(continued on next page)

Table 1 (continued)
Study Publication Enrolment Maximum Duration Study City No. of Diagnosis Age (median/ Males (%) Traveled to No. Family Current Health Underlying diseases Severe Diagnosis
date duration follow-up between design cases method mean [range/ or resident -cluster Smokers workers (%) Cases (%) of severity
duration onset of (RCS/SD/PS) IQR], years) of Hubei (family) (%)
(days) symptoms Province
and (%)
hospitalizati–
on (median
[range],
days)
Hyper Diabetes Cancer (%) Chronic Having any
tension (%) (%) respiratory coexisting
/lung medical
diseases condition
(%) (%)

Zhong et al Feb-13 NA NA NA RCS Wuhan 30 L 50 (22–81) 18 (60.0) 30 (100.0) NA NA NA NA NA NA NA 10 (30.0) 8 (26.7) Guidelines

L. Fu, B. Wang and T. Yuan et al. / Journal of Infection 80 (2020) 656–665

Xia et al Feb-18 Jan 15 -Feb NA (7.44±2.99) RCS Wuhan 52 L 54 (23–82) 24 (46.0) 52 (100.0) NA NA NA 25 (48.0) 26 (50.0) NA NA NA 12 (23.0) Guidelines
8
Yang et al Feb-21 Dec 24 -JanNA NA RCS Wuhan 52 L 59 (13.3) 35 (67.0) 52 (100.0) NA 2 (4.0) NA NA 9 (17.0) 2 (4.0) 2 (4.0) 21 (40.0) 52 (100.0) ICU
26
Du et al Feb-9 Jan 27 -Feb NA NA RCS Xian 7 NA 40 (24–55) 4 (57.0) 2 (28.5) 3 0 0 NA NA NA NA NA NA NA
1
Gao et al Feb-6 NA NA NA RCS Xian 10 L 41.8† (22–70) 6 (60.0) 9 (90.0) NA NA NA NA NA NA NA NA 0 NA
Liu et al Feb-18 NA NA NA RCS Xiaogan 41 L 48 (19–64) 32 (78.0) 28 (68.0) NA NA NA 5 (12.0) 2 (5.0) NA NA NA 5 (12.0) NA
Xu et al Feb-20 Jan 10 -Jan NA 2 (1–4) RCS Zhejiang 62 L 41† (32–52) 32 (58.0) 62 (100.0) NA NA NA 5 (8.0) 1 (2.0) NA 1 (2.0) 20 (32.0) 1 (2.0) Guidelines
26
Yu et al Feb-03 Jan 21 -Feb NA 5.5 (3–13) RCS Beijing 25 L 37.9† (3–79) 16 (64.0) 23 (92.0) 3 NA NA 1 (4.0) 3 (12.0) NA NA NA NA NA
2
Huang et al Feb-16 Jan 23 -Feb NA NA RCS Guangzhou 35 L 44 (12–74) 19 (54.0) 20 (57.0) NA 5 (14.0) NA 1 (3.0) 2 (6.0) NA 1 (3.0) NA NA NA
24
Wang et al Feb-15 Jan 19 -Feb NA NA RCS Zhejiang 52 L 44 (13–73) 29 (56.0) 16 (30.0) NA NA NA NA NA NA NA NA NA NA
3
Fang et al Feb-25 Jan 22 -Feb NA NA RCS Hefei 79 L 45.1† (5–91) 18 (75.0) NA NA NA NA 11 (46.0) NA NA NA NA 24 (30.0) Guidelines
18
Chen et al Feb-19 Jan 24 -Feb NA 7 (4–9.5) RCS Wuhan 54 L 58.5 (43–69) 27 (50.0) NA NA NA NA 13 (24.0) NA NA NA NA 31 (57.0) Guidelines
8
Xian et al Feb-17 Jan 21 -Jan NA NA RCS Nanchang 49 L 42.0† (18–78) 33 (67.0) 46 (94.0) NA 3 (6.0) NA 6 (12.0) 2 (4.0) NA NA NA 9 (18.0) Guidelines
27
Cao et al Feb-28 Jan 1 -Feb NA NA RCS Wuhan 36 L 72.5† (61–82) 19 (55.5) NA NA NA NA 17 (47.2) 8 (22.2) NA 0.583 NA NA NA
15
Li et al Feb-24 Jan 26 -Feb NA NA RCS Anhui 12 L 37 (21–71) 8 (66.7) 12 (100.0) NA 0.333 NA 2 (16.7) NA NA NA NA 0 NA
6
Sun et al Feb-24 Jan 21 -Feb NA NA RCS Tianjin 88 L 48.5† (9–91) 49 (55.7) 26 (29.5) NA NA NA 22 (25.0) 10 (11.4) NA NA NA 32 (36.4) Guidelines
8
Ji et al Feb-24 Jan 19 -Feb NA NA RCS Jingzhou 45 L 45.4† (21–67) 27 (60.0) 37 (82.2) NA NA NA NA NA NA NA NA NA NA
1
Wang et al Feb-24 Jan 1 -Feb NA NA RCS Wuhan 159 L 45.5† (20–84) 66 (41.5) NA NA NA NA NA NA NA NA NA NA NA
14
Yu et al Feb-26 Jan 17 -Jan NA NA RCS Wenzhou 40 L 45.9† (23–67) 22 (55.0) NA NA NA NA NA NA NA NA NA NA NA
28
XIAO et al Feb-27 Jan 23 -Feb NA NA RCS Chongqing 143 L 45.1† 73 (51.0) 76 (53.0) NA NA NA 17 (12.0) 10 (7.0) NA 4 (3.0) NA 36 (25.0) Guidelines
8
Wu et al Feb-28 Jan 22 -Feb NA NA RCS Jiangsu 80 L 46.1† 39 (49.0) 80 (100.0) 5 NA NA 25 (31.0) 5 (6.0) 1 (1.0) 1 (1.0) NA 3 (4.0) Guidelines
14
Xu et al Feb-19 Jan 23 -Feb NA NA RCS Guangzhou 90 L 50 (18–86) 39 (43.0) 86 (96.0) NA NA NA 17 (19.0) 5 (6.0) 2 (2.0) 1 (1.0) 45 (50.0) NA Guidelines
4
NA = Not available. RCS = Retrospective case series. SD = Surveillance data. PS = Prospective study. L = Laboratory-confirmed. Guideline = Guidelines of 2019-nCoV infection from the National Health Commission of the People’s
Republic of China. ICU = Being admitted to ICU. ATS = American Thoracic Society guideline on admission. All studies were published in 2020. December belongs to 2019. If there is no mark, the median and range were used
to represent age. ∗ All cases originated from 31 provinces, municipalities and autonomous regions other than Hubei province. †These values are average values. ‡These data are interquartile range.

659
660 L. Fu, B. Wang and T. Yuan et al. / Journal of Infection 80 (2020) 656–665

included studies were retrospective case series (40 [90.3%]), 27

(62.8%) were from cities outside Wuhan, and 34 (79.0%) only in-
cluded patients with laboratory confirmed COVID-19. The number
of patients enrolled in each study ranged from 4 to 1099. Mean
or median age of patients varied from 39 to 72 years (median 41
years; 43 studies). The proportion of male patients ranged from
29.0% to 77.0% (median 56.5%; 42 studies). The proportion of pa-
tients who had ever traveled to or were resident of Hubei Province
varied from 28.5% to 100.0% (median 91.0%; 36 studies). The num-
ber of family-clusters ranged from 1 to 5 (10 studies). The pro-
portion of patients who were current smokers ranged from 0.0%
to 18.0% (median 7.2%; 9 studies), and health workers ranged from
0.0% to 29.0% (median 4.0%; 5 studies). The proportion of patients
with hypertension ranged from 0.0% to 48.0% (median 16.0%;27
studies), diabetes ranged from 0.0% to 50.0% (median 10.1%; 26
studies), cancer ranged from 0.0% to 17.0% (median 1.0%; 15 stud-
ies), chronic respiratory/lung diseases ranged from 0.0% to 17.0%
(median 2.0%; 16 studies), having any coexisting medical comor-
bidity ranged from 12.0% to 67.0%. The proportion of patients diag-
nosed with severe COVID-19 varied from 0.0% to 100.0% (median
26.5%; 21 studies), and the most commonly used diagnostic cri-
teria was The Guidelines on 2019-nCoV Treatment and Prevention
issued by the NHC (70.6) (17 studies). 9 (20.9%) of 43 studies were
rated as low risk of bias, 30 studies (69.8%) as moderate, and all
remaining studies rated as high risk of bias (supplementary Table
1).
We meta-analysed the prevalence of 16 clinical symptoms
among COVID-19 patients (Fig. 2). Fever (83.3% [95% CI 78.4–87.7]),
Fig. 1. Flow diagram of publication selection
∗ cough (60.3% [54.2–66.3]), and fatigue (38.0% [29.8–46.5]) were the
Figure legend: COVID-19: Corona Virus Disease 2019.
most common, followed by increased sputum production, short-
ness of breath, and myalgia, with estimated prevalence just un-
der 30% for each, respectively. Eleven studies reported the pro-
remaining was in English. The earliest enrollment time was 16 De- portion of COVID-19 patients who did not exhibit obvious symp-
cember 2019 and the latest was 27 January 2020. One publica- toms, and the pooled estimated prevalence was 5.6% (1.4–11.6).
tion was a letter, and the remainder were journal articles. Most Among 16 commonly reported laboratory findings (Fig. 3), the

Fig. 2. Meta-analysis of the prevalence of clinical symptoms among COVID-19 patients.

 L. Fu, B. Wang and T. Yuan et al. / Journal of Infection 80 (2020) 656–665 661

Fig. 3. Meta-analysis of the prevalence of laboratory findings among COVID-19 patients.

most common laboratory abnormalities were elevated C-reactive nificantly higher in studies in which the proportion of male pa-
protein (68.6% [58.2–78.2]) and decreased lymphocyte count (57.4% tients was smaller, whereas the reverse was true for the prevalence
[44.8–69.5]), as well as increased lactate dehydrogenase (51.6% of elevated aspartate aminotransferase and lactate dehydrogenase
[31.4–71.6]). Ground-glass opacities (80.0% [67.3–90.4]) and bilat- (all p<0.05), though fatality did not differ by gender.
eral pneumonia (73.2% [63.4–82.1]) and were the most frequent A total of eight studies reported separate results for severe
chest CT findings (Fig. 3). The vast majority of patients received an- cases and non- severe cases. Overall, the existence of clinical symp-
tiviral therapy (90.0% 74.1–99.0]), antibiotic treatment (71.5% [50.0– toms, abnormalities in laboratory and chest CT findings, and com-
89.7]), and oxygen therapy (71•5% [28•0-99•7]). Acute respiratory plications were higher among patients with severe illness com-
distress syndrome (ARDS) was the most common complication pared to patients without severe illness (Table 2), however these
(15.7% [5.0–30.4]). The overall estimated prevalence of severe case differences were not statistically significant due to limited sample
and death was 25.6% (17.4–34.9) and 3.6% (1.1–7.2), respectively size and statistical power (data not shown).
(Fig. 4). Publication bias was found in the following subgroup out-
In subgroup analysis (supplementary Tables 2–5), studies from comes: fever, myalgia, diarrhea, rhinorrhea, hemoptysis, decreased
Wuhan had significantly higher prevalence of death, fever, fatigue, leucocytes, lymphopenia, increased creatine, creatine kinase, and
headache, elevated leukocyte count, and elevated lactate dehydro- procalcitonin, bilateral pneumonia, solid nodules, antiviral ther-
genase, and elevated aspartate aminotransferase compared to pa- apy, and immunoglobulin therapy (Figs. 2–4, all p<0.005 by Egger
tients from other cities (all p<0.05). Similarly, the prevalence of test). Substantial heterogeneity was present within most subgroups
death, ARDS, headache, increased leukocyte count, and increased (Table 2 and Figs. 2–4).
lactate dehydrogenase were significantly higher in studies in which
the proportion of older patients was larger (all p<0.05), and the Discussion
prevalence of diarrhea, and elevated lactate dehydrogenase were
significantly higher in studies in which the proportion of patients Our systematic review and meta-analysis of 43 studies in-
with any coexisting medical condition was larger (all p<0.05). The volving 3600 patients provides the most comprehensive overview
prevalence of fatigue, myalgia, decreased leucocyte count were sig- of clinical features, laboratory findings, chest imaging findings,
662 L. Fu, B. Wang and T. Yuan et al. / Journal of Infection 80 (2020) 656–665

Table 2
Outcomes comparing severe cases and non-severe cases of COVID-19.

Outcomes Critical illness Non-critical illness

No. reports No. patients Prevalence% (95%CI) I2 (%) No. reports No. patients Prevalence% (95%CI) I2 (%)

Clinical symptoms
Fever 6 364 80.8 (41.1–100.0) 97 6 1299 71.2 (23.8–99.9) 98
Cough 6 364 65.6 (51.7–78.2) 67 6 1299 56.7 (39.5–73.2) 88
Sore throat 3 245 16.7 (0.0–53.2) 77 3 1135 11.2 (3.5–22.4) 63
Increased sputum production 3 222 32.1 (15.6–51.0) 19 3 1065 31.4 (23.1–40.5) 14
Shortness of breath 6 364 49.2 (21.5–77.2) 90 5 1216 13.3 (2.2–30.9) 85
Myalgia 5 351 17.6 (8.2–29.5) 57 5 1201 20.8 (10.0–33.9) 85
Fatigue 4 299 41.2 (5.2–84.0) 92 5 1201 34.5 (13.2–59.6) 93
Diarrhea 4 234 7.6 (0.0–24.0) 55 3 1053 4.3 (0.1–12.5) 54
Headache 4 274 11.3 (0.1–33.9) 74 5 1172 11.9 (5.8–19.7) 53
Laboratory findings
Leucocytes (↑) 2 186 27.7 (0.0–100.0) 91 3 838 9.3 (0.0–1.0) 67
Leucocytes (↓) 3 216 33.7 (0.00–95.7) 92 3 957 27.2 (24.3–30.1) 0
Lymphocytes (↓) 3 203 81.5 (18.9–100.0) 94 4 883 59.6 (32.2–84.2) 99
Platelets (↓) 2 169 32.3 (0.0–100.0) 93 3 740 16.4 (0.0–1.0) 88
Aspartate aminotransferase (↑) 2 155 46.1 (0.0–100.0) 56 3 653 15.5 (0.0–50.8) 55
Creatinine (↑) 2 151 6.4 (0.0, 100.0) 57 2 642 2.3 (0.0, 97.1) 76
Creatine kinase (↑) 2 134 28.6 (0.0–100.0) 76 3 563 16.7 (0.0–1.0) 96
Lactate dehydrogenase (↑) 3 173 62.7 (55.7–100.0) 83 3 818 28.1 (0.0, 100.0) 99
C-reactive protein (↑) 2 171 40.3 (0.0–100.0) 99 5 1026 51.2 (38.6–63.8) 71
D-dimer (↑) 2 109 59.6 (50.2–68.7) 0 1 451 43.2 (38.7–47.8) 0
Procalcitonin (↑) 3 165 35.7 (0.0–100.0) 95 4 660 55.2 (0.0–33.8) 95
Chest CT findings
Bilateral pneumonia 2 186 91.0 (0.0–100) 83 1 926 39.7 (36.6–42.9) 0
Complications
ARDS 4 315 38.2 (3.2–83.0) 96 2 130 4.3 (2.8, 6.0) 0
Cardiac failure 4 155 17.1 (1.5–42.2) 78 2 130 1.9 (0.0, 26.0) 0
Shock 3 222 17.4 (0.0, 61.5) 87 . . . .
Renal insufficiency 5 328 9.8 (0.1–28.7) 87 . . . .

ARDS=Acute Respiratory Distress Syndrome.

disease severity, and CFR of COVID-19 patients. Compared with the COVID-19 patients in mainland China (updated through Febru-
only previous published systematic review on the subject, we in- ary11, 2020).10 CFR may have been higher in earlier reports be-
cluded 31 additional studies performed detailed subgroup analyses. cause of belated treatment during the earlier stages of the out-
Particularly our results suggest CFR and proportion of severe cases break or a decline in fatality after sustained human-to-human
are both declining as 2019-nCOV spreads away from Wuhan. transmission.1 , 14 , 19 Of note, roughly half of the studies included in
The dominant clinical features of COVID-19 were fever, cough, our analysis were from outside Wuhan, the epicenter of the cur-
and fatigue, while congestion, rhinorrhea, sore throat and diarrhea rent outbreak, and our subgroup analysis found significantly lower
are rare.13 , 16 , 19 , 24 The most frequently reported laboratory abnor- prevalence of death among patients treated outside Wuhan. This
malities were reduced lymphocyte count, elevated C-reactive pro- may indicate fatality from COVID-19 is declining.
tein, and elevated lactate dehydrogenase, all of which are generally In our analysis, the proportion of severe cases (25.6%) was close
consistent with previous reports of patients with COVID-19.11 , 19 , 24 to the estimate in the China CDC report (18.5%).10 This is consistent
However, all these laboratory markers are very non-specific, with previous studies that patients from Wuhan had significantly
making their clinical utility limited. When evaluating suspected higher prevalence of death, fever, elevated leucocyte count, and
cases, physicians cannot rely on these laboratory abnormalities to elevated aspartate aminotransferase compared with patients from
exclude or confirm the diagnosis of COVID-19. These abnormalities other cities in China (all p<0.05).1 , 14 , 19 Additionally, the China CDC
are similar to those previously observed in patients with SARS and report supports our finding that the overall CFR in Hubei (2.9%) is
MERS.57–59 Previous research suggests these abnormalities may be higher than that outside Hubei (0.4%).10 This interpretation could
related to the cytokine storm brought on by infection.22 Recently, be supported by a study that showed lower fatality in patients who
a study suggested that COVID-19 may primarily affect T lympho- did not have direct contact with the site of the original disease.62
cytes, especially CD4+ T cells, resulting in significant lymphope- Similarly, the CFR, proportion of severe cases, ARDS, headache, in-
nia as well as decreased IFN- γ production.60 Additionally, by us- creased leukocyte count, and increased lactate dehydrogenase were
ing a multiple linear regression model, a study showed that CD4+ significantly higher in studies in which the proportion of older
T lymphocyte count may help predict the duration of viral RNA patients was larger (all p<0.05), which is consistent with previ-
detection in patients’ stools (p = 0.010).61 However, the number of ous publications.62 This finding suggests COVID-19 may dispropor-
cases currently reported is too small to draw firm conclusions, and tionately impact the elderly or people living with medical comor-
further studies are required. The most frequently reported find- bidities. This is consistent with a single-center retrospective study
ing on CT imaging was ground-glass opacities, particularly bilat- found that older patients (>65 years) with comorbidities and ARDS
eral opacities impacting three or more lobes. These results are also were at increased risk of death.45 A multivariate Cox regression
consistent with previous studies,21 and are also frequently identi- analysis results showed age and severe cases were identified as in-
fied in MERS and SARS.57–59 dependent prognostic factors for virus clearance.62 Furthermore, a
In this systematic review and meta-analysis, we found a CFR of study showed that children might be less likely to become infected
3.6%, which is closer to the estimate (2.3%) in a report by the Chi- or, if infected, may show milder symptoms.16 Another study also
nese Center for Disease Control and Prevention (China CDC) that confirmed that the elderly and those with comorbidities including
includes the epidemiological characteristics of 44,672 confirmed diabetes, hypertension, cardiovascular disease, liver diseases, ma-
 L. Fu, B. Wang and T. Yuan et al. / Journal of Infection 80 (2020) 656–665 663

Fig. 4. Meta-analysis of the prevalence of chest CT findings, complications, severe cases, and mortality among COVID-19 patients
∗
Figure legend: ARDS=Acute Respiratory Distress Syndrome.

lignancy were more likely to develop critical illness (62.1%:25.0%, nation of fallacies.9 Authors misuse fundamental terms. They mis-
p<0.001).62 take incidence for prevalence and odds ratio for proportion. They
Our study did not find significant differences between men and demonstrate the proportion of severe cases is 88% and case fa-
women in terms of CFR and proportion of severe cases. This find- tality rate is 42% in figures, which are misleading. PRISMA guide-
ing is similar to a previous study in which there was no difference lines and test for heterogeneity were not mentioned. Authors state
in the proportion of men and women admitted to the intensive in Methods that “Only available data from published articles were
care unit (ICU) for treatment of COVID-19.6 However, this differs collected. Data from unpublished papers were not included.” How-
from another study which found that men are more susceptible to ever 4 out of 10 references were from Medrxiv, a platform that
COVID-19 than women,63 as well as a recent publication reporting publishes non-peer reviewed reports. These reports, as it clearly
that seven of nine infant patients were female.64 There is no clear states on Medrxiv’s website, should not be relied on to guide clin-
explanation as to why men and women would be at different risk ical practice or health-related behavior and should not be reported
of infection, however some have proposed genetic mechanisms or as established information. One reference providing 4021 cases was
sex-specific effects.65 Whether there are differences in risk of in- already withdrawn from publication.66 It is inappropriate to in-
fection between men and women requires further research. clude the China CDC report providing epidemiological character-
We found the prognosis was worse among severe cases com- istics of 44,672 cases of COVID-19 (as of February 11, 2020) in a
pared to non-severe cases, however these differences were not sta- meta-analysis of its clinical characteristics.10 This report, based on
tistically significant, which is likely due to insufficient sample size. national surveillance data, provides epidemiological data only, in-
In our research, there was no significant difference in the degree cluding spatiotemporal distribution. Albeit this report includes a
of lymphocyte decline between severe cases and non-severe cases. large sample, data on clinical symptoms that are not systemati-
This conclusion can be supported by this research that the ex- cally reported, may not be reliable. For example, 53% did not re-
pression level of lymphocyte counts has no significant correlation port if they have co-morbidity or not. 9 out 10 studies included
with the severity of the disease.22 However, some studies showed in the meta-analysis were published/submitted before February 11,
that lymphocytopenia is a prominent feature of severe cases.45 2020 so cases in these 9 studies must have already been included
At present, it is unclear whether lymphocyte count is related to in the China CDC report. It is inappropriate to count an individual
severity of disease. Further investigation is needed to establish twice. After excluding the China CDC report and the four preprint
whether lymphocytosis or lymphopenia can help predict mortality articles, only 369 patients would be reportable in that review. Au-
in COVID-19 patients.62 thors did not list specific imaging performance in abnormal imag-
We found many patients were treated with antiviral and antibi- ing, nor did they list pulmonary fibrosis and its incidence. How-
otic therapy. Currently there is no treatment that can cure COVID- ever in Discussion they use two lengthy paragraphs to explain the
19. Supportive measures may reduce complications and fatality.14 content of pulmonary fibrosis, which may cause readers to mis-
The impact of antivirals and antibiotics on patients’ prognosis re- takenly believe that the imaging abnormality is pulmonary fibro-
mains unknown and requires further clinical evaluation. Currently, sis. Author failed to report any clinical laboratory findings and
clinical trials of lopinavir / ritonavir (LPV/r) and remdesivir regis- treatments of COVID-19 which are essential to a thorough un-
tered in the Chinese clinical trial registry are ongoing. derstanding of clinical characteristics. They also failed to report
The recently published systematic review and meta-analysis on the diagnostic criteria for abnormal chest CT detection and severe
the clinical characteristics of 50,466 patients may reflect a combi- cases.
664 L. Fu, B. Wang and T. Yuan et al. / Journal of Infection 80 (2020) 656–665

Our systematic review and meta-analysis has limitations. First, 6. Wang D, Hu B, Hu C, et al. Clinical characteristics of 138 hospitalized patients
we found substantial heterogeneity between studies and signifi- with 2019 novel coronavirus-infected pneumonia in Wuhan, China. JAMA 2020.
7. Guan W.-j., Ni Z.-y., Hu Y., et al. Clinical characteristics of 2019 novel coron-
cant publication bias among several subgroups. Second, this study avirus infection in China. 2020: https://doi.org/10.1101/2020.02.06.20020974.
performs an analysis during an ongoing outbreak. Many regions 8. National Health Commission of the People’s Republic of China. http://www.nhc.
affected by COVID-19 haven not yet published clinical datasets, gov.cn (Assessed on March 2, 2020).
9. Sun P, Qie S, Liu Z, et al. Clinical characteristics of hospitalized patients with
which may skew the results of this analysis. All these datasets are SARS-CoV-2 infection: a single arm meta-analysis. J Med Virol 2020. doi: 10.
retrospective, which prevents us from exploring risk factors. Ad- 1002/jmv.25735.
ditionally, our meta-analysis focused on Chinese people, not those 10. The Novel Coronavirus Pneumonia Emergency Response Epidemiology Team
The epidemiological characteristics of an outbreak of 2019 novel coronavirus
infected in other countries, so geographical and ethnic differences
diseases (COVID-19) in China. Chinese J Epidemiol 2020;41(2):145–51.
were not excluded. Finally, the meta-analysis was performed by 11. Liu Y, Yang Y, Zhang C, et al. Clinical and biochemical indexes from 2019-nCoV
comparing entire datasets against one another, therefore there was infected patients linked to viral loads and lung injury. Sci China Life Sci 2020.
12. Wang Z, Chen X, Lu Y, et al. Clinical characteristics and therapeutic procedure
no way to analyze data on the level of individual patients.
for four cases with 2019 novel coronavirus pneumonia receiving combined chi-
nese and western medicine treatment. Biosci Trends 2020.
Conclusion 13. Guan WJ, Ni ZY, Hu Y, et al. Clinical characteristics of coronavirus disease 2019
in China. N Engl J Med 2020.
14. Kui L, Fang YY, Deng Y, et al. Clinical characteristics of novel coronavirus cases
This review provides a comprehensive characterization of clin- in tertiary hospitals in Hubei province. Chin Med J (Engl) 2020.
ical features among COVID-19 patients. Patients living in Wuhan, 15. Chung M, Bernheim A. CT Imaging Features of 2019 Novel Coronavirus (2019-
older patients, and those with medical comorbidities tend to have nCoV). Radiology 2020 Apr;295(1):202-207.
16. Li Q, Guan X, Wu P, et al. Early transmission dynamics in Wuhan, China,
more severe clinical symptoms and higher fatality. Better therapeu- of novel coronavirus-infected pneumonia. N Engl J Med 2020. doi: 10.1056/
tics are crucial for the treatment of severe cases. Our comprehen- NEJMoa2001316.
sive characterization of COVID-19 will inform healthcare providers 17. Song F, Shi N, Shan F, et al. Emerging 2019 Novel Coronavirus (2019-nCoV)
Pneumonia. Radiology 2020:200274.
and public health policy makers in their efforts to treat and control 18. Chang, Lin M, Wei L, et al. Epidemiologic and clinical characteristics of novel
the current outbreak. coronavirus infections involving 13 patients outside Wuhan, China. JAMA 2020.
19. 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
Contributors
study. Lancet 2020;395:507–13.
20. Pan Y, Guan H, Zhou S, et al. Initial CT findings and temporal changes in pa-
HZ, YS and LL conceived the study and designed the protocol tients with the novel coronavirus pneumonia (2019-nCoV): a study of 63 pa-
with LF and BW. LF, BW, TY and XC conducted study selection and tients in Wuhan, China. Eur Radiol 2020. doi: 10.10 07/s0 0330- 020- 06731- x.
21. Pan F, Ye T. Time course of lung changes on chest CT during recovery from 2019
data extraction. LF, WB, TY, XC, YA contributed to statistical anal- novel coronavirus (COVID-19) pneumonia. Radiology 2020:200370. doi: 10.1148/
ysis and interpretation of data. LF, BW, TY, XC and HZ drafted the radiol.2020200370.
manuscript with all authors critically revising the manuscript. 22. Chen L, Liu HG, Liu W, et al. [Analysis of clinical features of 29 patients
with 2019 novel coronavirus pneumonia]. Zhonghua Jie He He Hu Xi Za Zhi
2020;43:E005.
Declaration of Competing Interest 23. Zhang MQ, Wang XH, Chen YL, et al. [Clinical features of 2019 novel coronavirus
pneumonia in the early stage from a fever clinic in Beijing]. Zhonghua Jie He He
Hu Xi Za Zhi 2020;43:E013.
The authors declare having no conflict of interest related to this 24. Xu XW, Wu XX, Jiang XG, et al. Clinical findings in a group of patients infected
work. with the 2019 novel coronavirus (SARS-Cov-2) outside of Wuhan, China: retro-
spective case series. BMJ 2020;368:m606.
25. Wu J, Liu J, Zhao X, et al. Clinical characteristics of imported cases of COVID-19
Funding
in Jiangsu province: a multicenter descriptive study. Clin Infect Dis 2020.
26. Xu X, Yu C, Qu J, et al. Imaging and clinical features of patients with 2019 novel
This study was supported by the Natural Science Foundation coronavirus SARS-CoV-2. Eur J Nucl Med Mol Imaging 2020.
27. Chen M, An W, Xia F, et al. Retrospective analysis of COVID-19 patients with
of China Young Scientist Fund [81703278], the Australian National
different clinical subtypes. Herald of Med 2020:1–12.
Health and Medical Research Commission (NHMRC) Early Career 28. H X, N L, L F, et al. Analysis of Chinese medical characteristics of 35 patients
Fellowship (grant number APP1092621), the Precision Targeted In- with novel coronavirus pneumonia. J Emergen Tradit Chinese Med 2020:1–4.
tervention Studies among High Risk Groups for HIV Prevention 29. D YH, J CW, Y J, et al. Clinical features and ct signs of early family clustering
novel coronavirus pneumonia. J Xi’an Jiaotong Uni 2020:1–7.
in China, National Science and Technology Major Project of China 30. Y X, Y H, Y S, et al. Chest CT features of COVID 19. J Pract Med 2020:1–3.
[2018ZX10721102], the Sanming Project of Medicine in Shenzhen 31. G L, Z JP, D YH, et al. CT features of patients with imported 2019-nCov-pneu-
[SZSM201811071], the Australian Research Council centre of Ex- monia. J Xi’an Jiaotong Uni 2020:1–9.
32. G XM, L H, S L. Preliminary explore on ct characteristics of corona virus disease
cellence for Mathematical and Statistical Frontiers [CE14010 0 049], 2019 (COVID-19). Radiol Pract 2020:1–5.
Infectious Disease Specialty of Guangzhou High-level Clinical Key 33. X K, S L, P X, et al. The clinical features of the 143 patients with COVID-19 in
Specialty (2019–2021). north-east of Chongqing. J Third Military Med Uni 2020:1–5.
34. X F, Q M, T Y, et al. Clinical features and treatment analysis of 79 cases of
COVID-19. Chinese pharmacological bulletin 2020:1–7.
Supplementary materials 35. L FM, D HL, G XM. Chest CT performance and clinical characteristics of corona
virus disease 2019 (COVID-19). Radiol Pract 2020:1–3.
36. L YF, Y ZG, W M, et al. Analysis on chinese medical clinical characteristics of
Supplementary material associated with this article can be
50 patients with 2019-nCoV-infected pneumonia. Academic Journal of Shanghai
found, in the online version, at doi:10.1016/j.jinf.2020.03.041. University of Traditional Chinese Medicine 2020:1–5.
37. T. X, J. L, F. X, et al. Analysis of clinical characteristics of 49 patients with novel
References coronavirus pneumonia in Jiangxi province. Chinese J Resp Critical Care Med:1–
7.
1. Huang C, Wang Y, Li X, et al. Clinical features of patients infected with 2019 38. J G, H M, Z Q, et al. CT manifestations and dynamic changes of corona virus
novel coronavirus in Wuhan, China. Lancet 2020;395:497–506. disease 2019. Chinese J Med Imaging Technol 2020:1–6.
2. WHO main website. https://www.who.int (accessed March 2, 2020). 39. C J, Z J, L X, et al. Clinical characteristics and ct signs of corona virus disease
3. Zhu N, Zhang D, Wang W, et al. A novel coronavirus from patients with pneu- 2019 (COVID-19) in the elderly. Med J Wuhan Uni 2020:1–4.
monia in China, 2019. N Engl J Med 2020;382:727–33. 40. Y S, W Z, Q E, et al. Analysis of clinical characteristics of 25 patients with coro-
4. Lu R, Zhao X, Li J, et al. Genomic characterisation and epidemiology of 2019 navirus disease in 2019. Chinese J Integrated Trad Western Med 2020:1–2.
novel coronavirus: implications for virus origins and receptor binding. Lancet 41. W J, L J, W Y, et al. Dynamic changes of chest ct imaging in patients with
2020. corona virus disease-19 (COVID-19). J Zhejiang Uni (Med Sci) 2020:1–13.
5. Chan JF, Yuan S, Kok KH, et al. A familial cluster of pneumonia associated with 42. Sun H, Bi Y, Zhu Z, et al. Analysis of Chinese medical characteristics of 88
the 2019 novel coronavirus indicating person-to-person transmission: a study patients with novel coronavirus pneumonia in Tianjin. J Tradit Chinese Med
of a family cluster. Lancet 2020;395:514–23. 2020:1–4.
 L. Fu, B. Wang and T. Yuan et al. / Journal of Infection 80 (2020) 656–665 665

43. X G, A C, Z C, et al. Clinical study on treatment of 34 cases of COVID-19 with 55. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-anal-
combination of traditional Chinese and western medicine. Journal of Traditional yses. BMJ 2003;327:557–60.
Chinese Medicine 2020:1–7. 56. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis detected by
44. W Y, C J, W X. CT image features of asymptomatic patients with novel coron- a simple, graphical test. BMJ 1997;315:629–34.
avirus. Med J Wuhan Uni 2020:1–5. 57. Leung WK, To KF, Chan PK, et al. Enteric involvement of severe acute respiratory
45. Xiaobo Yang YY, Xu Jiqian, Shu Huaqing, Xia∗ Jia’an, Liu∗ Hong, Wu Yongran, syndrome-associated coronavirus infection. Gastroenterology 2003;125:1011–17.
Zhang Lu, Yu Zhui, Fang Minghao, Yu Ting, Wang Yaxin, Shangwen Pan XZ, 58. Peiris JS, Chu CM, Cheng VC, et al. Clinical progression and viral load in a
Yuan Shiying, Shang You. Clinical course and outcomes of critically ill patients community outbreak of coronavirus-associated Sars pneumonia: a prospective
with SARS-CoV-2 pneumonia in wuhan, china: a single-centered, retrospective, study. Lancet 2003;361:1767–72.
observational study. Lancet Respir Med 2020. 59. Assiri A, McGeer A, Perl TM, et al. Hospital outbreak of middle east respiratory
46. Li X, Pan H, Shu J, et al. Clinical presentations and ct features of imported syndrome coronavirus. N Engl J Med 2013;369:407–16.
corona virus disease 2019. Chinese J Med Imaging Technol 2020:1–4. 60. Chen G., Wu D., Guo W., et al. Clinical and immunologic fea-
47. Y SM, C YF, ZX W, et al. Analysis of the relationship between clinical features tures in severe and moderate forms of coronavirus disease 2019.
and tongue manifestations of 40 cases with novel coronavirus pneumonia. Bei- 2020:medRxiv2020.02.16.20023903.
jing J Trad Chinese Med 2020:1–8. 61. Ling Y, Xu SB, Lin YX, et al. Persistence and clearance of viral RNA in 2019 novel
48. Z X, L L, D GC, et al. A preliminary study on the clinical characteristics and coronavirus disease rehabilitation patients. Chin Med J (Engl) 2020.
Chinese medical syndrome of 42 cases of COVID-19 in Nanjing. J Nanjing Uni 62. Cai Q, Huang D, Ou P, et al. COVID-19 in a designated infectious diseases hos-
Trad Chinese Med 2020:1–5. pitaloutside Hubei province, China. medRxiv 2020. doi: https://doi.org/10.1101/
49. Z FY, Z HF, W BC, et al. CT findings in 2019 novel coronavirus disease 2020.02.17.20024018.
(COVID-19) patients. Med J Wuhan Uni 2020:1–5. 63. Qian K, Deng Y, Tai Y, et al. Clinical characteristics of 2019 novel infected coro-
50. Z YJ, C Z, L J, et al. Clinical and epidemiological characteristics of 26 patients navirus pneumonia:a systemic review and meta-analysis. medRxiv 2020. doi:
diagnosed with novel coronavirus pneumonia. Chinese J Nosocomiol 2020:1–4. https://doi.org/10.1101/2020.02.14.20021535.
51. Liberati A, Altman DG, Tetzlaff J, et al. The prisma statement for reporting sys- 64. Wei M, Yuan J, Liu Y, et al. Novel coronavirus infection in hospitalized infants
tematic reviews and meta-analyses of studies that evaluate healthcare interven- under 1 year of age in China. JAMA 2020.
tions: explanation and elaboration. BMJ 20 09;339:b270 0. 65. Schurz H, Salie M, Tromp G, et al. The x chromosome and sex-specific effects
52. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in infectious disease susceptibility. Hum Genomics 2019;13:2.
in epidemiology: a proposal for reporting. Meta-anal Observat Stud Epidemiol 66. Yang Y, Lu Q, Liu M, et al. Epidemiological and clinical features of the 2019
(MOOSE) Group. Jama 20 0 0;283:20 08–12. novel coronavirus outbreak in China. medRxiv 2020. doi: https://doi.org/10.
53. American National Institute of Health. https://www.nhlbi.nih.gov/health-topics/ 1101/2020.02.10.20021675.
study-quality-assessment-tools.
54. Barendregt JJ, Doi SA, Lee YY, et al. Meta-analysis of prevalence. J Epidemiol
Commun Health 2013;67:974–8.