Articles

Clinical course and outcomes of critically ill patients with

SARS-CoV-2 pneumonia in Wuhan, China: a single-centered,
retrospective, observational study
Xiaobo Yang*, Yuan Yu*, Jiqian Xu*, Huaqing Shu*, Jia’an Xia*, Hong Liu*, Yongran Wu, Lu Zhang, Zhui Yu, Minghao Fang, Ting Yu, Yaxin Wang,
Shangwen Pan, Xiaojing Zou, Shiying Yuan, You Shang

Summary
Background An ongoing outbreak of pneumonia associated with the severe acute respiratory coronavirus 2 (SARS-CoV-2) Lancet Respir Med 2020
started in December, 2019, in Wuhan, China. Information about critically ill patients with SARS-CoV-2 infection is Published Online
scarce. We aimed to describe the clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia. February 21, 2020
https://doi.org/10.1016/
S2213-2600(20)30079-5
Methods In this single-centered, retrospective, observational study, we enrolled 52 critically ill adult patients with
*Contributed equally
SARS-CoV-2 pneumonia who were admitted to the intensive care unit (ICU) of Wuhan Jin Yin-tan hospital (Wuhan,
Department of Critical Care
China) between late December, 2019, and Jan 26, 2020. Demographic data, symptoms, laboratory values, comorbidities, Medicine (X Yang MD, Y Yu MD,
treatments, and clinical outcomes were all collected. Data were compared between survivors and non-survivors. The J Xu MD, Prof H Shu MD,
primary outcome was 28-day mortality, as of Feb 9, 2020. Secondary outcomes included incidence of SARS-CoV-2- Prof H Liu MD, Y Wu MD,
related acute respiratory distress syndrome (ARDS) and the proportion of patients requiring mechanical ventilation. Y Wang MD, S Pan MD,
Prof X Zou MD, Prof S Yuan MD,
Prof Y Shang MD), Institute of
Findings Of 710 patients with SARS-CoV-2 pneumonia, 52 critically ill adult patients were included. The mean age of Anesthesiology and Critical
the 52 patients was 59·7 (SD 13·3) years, 35 (67%) were men, 21 (40%) had chronic illness, 51 (98%) had fever. Care Medicine (X Yang, Y Yu,
J Xu, Prof H Shu, Prof H Liu, Y Wu,
32 (61·5%) patients had died at 28 days, and the median duration from admission to the intensive care unit (ICU) to
Y Wang, S Pan, Prof X Zou,
death was 7 (IQR 3–11) days for non-survivors. Compared with survivors, non-survivors were older (64·6 years [11·2] Prof S Yuan, Prof Y Shang),
vs 51·9 years [12·9]), more likely to develop ARDS (26 [81%] patients vs 9 [45%] patients), and more likely to receive Union Hospital, and
mechanical ventilation (30 [94%] patients vs 7 [35%] patients), either invasively or non-invasively. Most patients had Department of Critical Care
Medicine, Tongji Hospital
organ function damage, including 35 (67%) with ARDS, 15 (29%) with acute kidney injury, 12 (23%) with cardiac
(Prof M Fang MD), Tongji
injury, 15 (29%) with liver dysfunction, and one (2%) with pneumothorax. 37 (71%) patients required mechanical Medical College, Huazhong
ventilation. Hospital-acquired infection occurred in seven (13·5%) patients. University of Science and
Technology, Wuhan, China;
Jin Yin-tan Hospital, Wuhan,
Interpretation The mortality of critically ill patients with SARS-CoV-2 pneumonia is considerable. The survival time of
China (X Yang, Prof J Xia MD,
the non-survivors is likely to be within 1–2 weeks after ICU admission. Older patients (>65 years) with comorbidities Prof H Liu, Prof T Yu MD,
and ARDS are at increased risk of death. The severity of SARS-CoV-2 pneumonia poses great strain on critical care Prof Y Shang); Department of
resources in hospitals, especially if they are not adequately staffed or resourced. Critical Care Medicine,
Xiangyang Central Hospital,
Affiliated Hospital of Hubei
Funding None. University of Arts and Science,
Hubei, China (L Zhang MD); and
Copyright © 2020 Elsevier Ltd. All rights reserved. Department of Critical Care
Medicine, Renmin Hospital
of Wuhan University, Wuhan,
Introduction mortality. In this study, we investigated critically ill China (Prof Z Yu MD)
Severe acute respiratory syndrome coronavirus 2 patients with confirmed SARS-CoV-2 pneumonia who Correspondence to:
(SARS-CoV-2) pneumonia is a newly recognised illness were admitted to Wuhan Jin Yin-tan hospital. The Prof You Shang, Department of
that has spread rapidly throughout Wuhan (Hubei baseline SARS-CoV-2-associated morbidity and mortality Critical Care Medicine, Union
Hospital, Tongji Medical College,
province) to other provinces in China and around the data from this study will be of considerable value for the
Huazhong University of Science
world.1–4 As of Feb 19, 2020, the total number of patients early identification of individuals who are at risk of and Technology, Wuhan 430022,
has risen sharply to 74 283 in the mainland of China, becoming critically ill and who are most likely to benefit China
with 2009 (2·7%) deceased. The clinical spectrum of from intensive care treatment. [email protected]

SARS-CoV-2 pneumonia ranges from mild to critically ill

cases. Previous studies have only described the general Methods
epidemiological findings, clinical presentation, and Study design and participants
clinical outcomes of patients of SARS-CoV-2 pneumonia.1,2,5 This single-centre, retrospective, observational study was
However, specific information characterising critically ill done at Wuhan Jin Yin-tan hospital (Wuhan, China),
patients remains unknown. which is a designated hospital to treat patients with
The data on the clinical characteristics and outcomes of SARS-CoV-2 pneumonia. All patients, except infected
critically ill patients with SARS-CoV-2 infection are healh-care workers from Jin Yin-tan hospital, were
scarce, but are of paramount importance to reduce transferred from other hospitals. We retro­ spectively

www.thelancet.com/respiratory Published online February 21, 2020 https://doi.org/10.1016/S2213-2600(20)30079-5 1

 Articles

Research in context
Evidence before this study 32 (61·5%) patients had died at 28 days, and the median
The novel coronavirus disease 2019 is a disease that has duration from intensive care unit (ICU) admission to death was
affected populations around the world. We searched PubMed 7 (IQR 3–11) days in non-survivors. Compared with survivors,
for articles published up to Feb 11, 2020, using the keywords non-survivors were older (64·6 years [11·2] vs 51·9 years [12·9]),
“2019 novel coronavirus”, “2019-nCoV”, “COVID-19”, or “SARS- more likely to develop acute respiratory distress (ARDS; 26 [81%]
CoV-2”. We identified eight articles that describe the patients vs 9 [45%] patients), and more likely to receive
epidemiological and clinical characteristics of patients infected mechanical ventilation (30 [94%] patients vs 7 [35%] patients),
with severe acute respiratory syndrome coronavirus 2 either invasively or non-invasively.
(SARS-CoV-2). However, none of these studies focused on
Implications of all the available evidence
characterising critically ill patients infected with SARS-CoV-2,
The mortality of critically ill patients with SARS-CoV-2
who are at increased risk of death.
pneumonia at 28 days is considerable. The survival time of non-
Added value of this study survivors is likely to be within 1–2 weeks after ICU admission.
We report the clinical courses and clinical outcomes of Older patients (>65 years) with comorbidities and ARDS are at
52 critically ill patients from 710 laboratory-confirmed cases of increased risk of death. The severity of SARS-CoV-2 pneumonia
SARS-CoV-2. 35 (67%) patients had acute respiratory distress poses great strain on critical care resources in hospitals,
syndrome (ARDS) and 37 (71%) required mechanical ventilation. especially if they are not adequately staffed or resourced.

analysed patients from Dec 24, 2019, to Jan 26, 2020, who hospital admission (fever, cough, dyspnoea, myalgia,
had been diagnosed with SARS-CoV-2 pneumonia, malaise, rhinorrhoea, arthralgia, chest pain, headache,
according to WHO interim guidance, and who were and vomiting), vital signs at ICU admission (heart rate,
critically ill.6 Laboratory confirmation of SARS-CoV-2 respiratory rate, blood pressure), laboratory values on
infection was performed by the local health authority as admis­ sion (haemoglobin concentration, lymphocyte
previously described.1,2 Critically ill patients were defined count, platelet count, arterial blood gas analysis, FiO2,
as those admitted to the intensive care unit (ICU) who partial pressure of oxygen (PaO2), and lactate concen­
required mechanical ventilation or had a fraction of tration), coexisted infection, treatment (oxygen therapy,
inspired oxygen (FiO2) of at least 60% or more.7–9 vaso­constrictive agents, antiviral agents, antibacterial
Identification of critically ill patients was achieved by agents, corticosteroids, and immuno­ globulin), as well
reviewing and analysing admission logs and histories as living status. During the outbreak of SARS-CoV-2
from all available electronic medical records and patient infection, the number of critically ill patients exceeded the
care resources. For patients who were alive by Jan 26, 2020, capacity of ICUs. Therefore, two provisional ICUs were
their living status was confirmed on Feb 9, 2020. urgently established in Jin Yin-tan hospital and hence
The Ethics Commission of Jin Yin-tan hospital most mechanical ventilator settings and recordings were
approved this study (KY-2020–06.01). Written informed not recorded, except records of positive end-expiratory
consent was waived due to the rapid emergence of this pressure in some cases. As a routine, electronic medical
infectious disease. data were archived onto a local server, from which we
retrieved these data.
Data collection
We reviewed clinical electronic medical records, nursing Outcomes
records, laboratory findings, and radiological exam­ The primary outcome was 28-day mortality after ICU
inations for all patients with laboratory confirmed admission. Secondary outcomes were incidence of
SARS-CoV-2 infection. The admission data of these SARS-CoV-2-related acute respiratory distress syndrome
patients were collected. Data were evaluated and (ARDS). ARDS and shock were defined according to the
collected, by using a case record form modified from the guidance of WHO for novel coronavirus disease 2019
standardised International Severe Acute Respiratory and (COVID-19).6 Acute kidney injury was identified on the
Emerging Infection Consortium case report forms.10 Any basis of serum creatinine.11 Cardiac injury was diagnosed
missing or uncertain records were collected and clarified if the serum concentration of hypersensitive cardiac
through direct communication with involved health-care troponin I (hsTNI) was above the upper limit of the
providers and their families. reference range (>28 pg/mL), measured in the laboratory
We collected data on age, sex, exposure history, chronic of Jin Yin-tan Hospital.
medical histories (chronic cardiac disease, chronic
pulmonary disease, cerebrovascular disease, chronic Statistical analysis
neurological disorder, diabetes, malignancy, dementia, The aim of this study is to report the clinical courses and
malnutrition, and smoking), symptoms from onset to clinical outcomes of critically ill patients being cared for

2 www.thelancet.com/respiratory Published online February 21, 2020 https://doi.org/10.1016/S2213-2600(20)30079-5

 Articles

710 patients with confirmed SARS-CoV-2 pneumonia as of Jan 26, 2020

Survivors Non-survivors All patients
(n=20) (n=32) (n=52)
Age, years 51·9 (12·9) 64·6 (11·2) 59·7 (13·3)
509 excluded Age range, years
509 admitted after Jan 12, 2020
30–39 6 (30%) 0 6 (11·5%)
40–49 3 (15%) 3 (9%) 6 (11·5%)
201 with confirmed SARS-CoV-2 pneumonia 50–59 4 (20%) 9 (28%) 13 (25%)
60–69 6 (30%) 11 (34%) 17 (33%)
146 excluded 70–79 1 (5%) 7 (22%) 8 (15%)
146 had oxygen therapy with ≥80 0 2 (6%) 2 (4%)
FiO2 <60%
Sex
Female 6 (30%) 11 (34%) 17 (33%)
55 critically ill patients admitted to the ICU Male 14 (70%) 21 (66%) 35 (67%)
Exposure
Exposure to Huanan 9 (45%) 8 (25%) 17 (33%)
3 excluded
seafood market
1 cardiac arrest on transportation
2 cardiac arrest immediately Exposure to patients* 2 (10%) 8 (25%) 10 (19%)
after admission Chronic medical illness 4 (20%) 17 (53%) 21 (40%)
Chronic cardiac disease 2 (10%) 3 (9%) 5 (10%)
52 critically ill patients included in study Chronic pulmonary 2 (10%) 2 (6%) 4 (8%)
disease
Cerebrovascular disease 0 7 (22%) 7 (13·5%)
Figure 1: Study flow diagram
SARS-CoV-2=severe acute respiratory syndrome coronavirus 2. FiO2=fraction of Diabetes 2 (10%) 7 (22%) 9 (17%)
inspired oxygen. Malignancy 0 0 2 (4%)
Dementia 0 0 1 (2%)
Malnutrition 0 0 1 (2%)
in the hospital during the study period. There were,
Smoking 0 0 2 (4%)
therefore, no formal hypotheses being implemented to
drive the sample size calculation and we included the Data are n (%) or mean (SD), unless otherwise specified. SARS-CoV-2=severe
acute respiratory syndrome coronavirus 2. *Patients who have confirmed
maximum number of patients who met the inclusion SARS-CoV-2 infection or are highly suspected of being infected.
criteria.
We expressed descriptive data as mean (SD) or median Table 1: Demographics and baseline characteristics of patients with
(IQR) for continuous variables and number (%) for severe SARS-CoV-2 pneumonia

categorical variables. We assessed differences between

survivors and non-survivors using two-sample t test or
Wilcoxon rank-sum test depending on parametric or non- infection. 21 (40%) patients had chronic diseases,
parametric data for continuous variables and Fisher’s including cerebrovascular diseases in seven (13·5%)
exact test for categorical variables. We used a Kaplan- patients, all of whom died at 28 days. All patients had
Meier plot for survival data. bilateral infiltrates on chest x-ray.
Tests were two-sided with significance set at α less The most common symptoms were fever (98%), cough
than 0·05. The Stata/IC 15.1 software (StataCorp, College (77%), and dyspnoea (63·5%; table 2). Among 52 critically
Station, TX, USA) was applied for all analyses. ill patients, six (11%) did not experienced fever until
2–8 days after the onset of symptoms related to
Results SARS-CoV-2 infection. The median duration from onset
By Jan 26, 2020, 710 patients had been admitted to of symptoms to radiological confirmation of pneumonia
Wuhan Jin Yin-tan hospital with confirmed SARS-CoV-2 was 5 (IQR 3–7) days. The median duration from onset of
pneumonia, of whom 658 (93%) were considered symptoms to ICU admission was 9·5 (7·0–12·5) days.
ineligible, including three patients who had cardiac The median Acute Physiology and Chronic Health
arrest immediately after admission. 52 (7%) critically ill Evaluation II (APACHE II) score of all patients was
patients were included in this study (figure 1). All 17 (IQR 14–19; table 3). Most patients had organ function
patients were residents of Wuhan City and were damage, including 35 (67%) with ARDS, 15 (29%) with
transferred from other hospitals. The mean age was acute kidney injury, 12 (23%) with cardiac injury, 15 (29%)
59·7 years (SD 13·3), and 27 (52%) were older than with liver dysfunction, and one (2%) with pneumothorax
60 years (table 1). 35 (67%) patients were men. 17 (33%) (table 2). The median hsTNI was 161·0 (IQR 41·8–766·1)
patients had a history of exposure to the Huanan pg/mL. Hospital-acquired infection was noted in
seafood market, and 10 (19%) had exposure to patients seven (13·5%) patients, including one (2%) patient who
with confirmed or highly suspected SARS-CoV-2 had pulmonary and blood stream infection of carbapenem-

www.thelancet.com/respiratory Published online February 21, 2020 https://doi.org/10.1016/S2213-2600(20)30079-5 3

 Articles

Survivors Non-survivors All patients Survivors Non-survivors

(n=20) (n=32) (n=52) (n=20) (n=32)
Symptoms Duration from onset of symptoms to 5 (3–9) 5 (3–7)
Fever 20 (100%) 31 (97%) 51 (98%) radiological confirmation of
pneumonia, days
Cough 15 (75%) 25 (78%) 40 (77%)
Duration from onset of symptoms to 9 (6–12) 11 (7–14)
Dyspnoea 12 (60%) 21 (66%) 33 (63·5%) ICU admission, days
Myalgia 2 (10%) 4 (12·5%) 6 (11·5%) Heart rate, beats per min 89 (20) 89 (15)
Malaise 4 (20%) 14 (44%) 18 (35%) Systolic blood pressure, mm Hg 133 (20) 140 (21)
Rhinorrhoea 0 3 (9%) 3 (6%) Ratio of PaO2 to FiO2, mm Hg 103·8 62·3
Arthralgia 1 (5%) 0 1 (2%) (66·6–126·7) (52·0–74·1)
Chest pain 1 (5%) 0 1 (2%) APACHE II score on day 1 14 (12–17) 18 (16–20)
Headache 1 (5%) 2 (6%) 3 (6%) SOFA score on day 1 4 (3–4) 6 (4–8)
Vomiting 1 (5%) 1 (3%) 2 (4%) Haemoglobin concentration, g/L 127 (20) 129 (14)
Comorbidities Lymphocyte count, × 10⁹/L 0·74 (0·84) 0·62 (0·37)
Acute respiratory distress 9 (45%) 26 (81%) 35 (67%) Platelet count, × 10⁹ /L 164 (74) 191 (63)
syndrome Prothrombin time, s 10·9 (2·7) 12·9 (2·9)
Acute kidney injury 3 (15%) 12 (37·5%) 15 (29%) Total bilirubin concentration, μmol/L 13·1 (4·3) 19·5 (11·6)
Cardiac injury 3 (15%) 9 (28%) 12 (23%) Serum creatinine concentration, 76·3 (27·4) 80·7 (32·3)
Liver dysfunction 6 (30%) 9 (28%) 15 (29%) μmol/L
Hyperglycaemia 7 (35%) 11 (34%) 18 (35%) Lactate concentration, mmol/L 1·4 (1·3–1·6) 2·9 (1·4–3·2)
Gastrointestinal 0 2 (6%) 2 (4% Data are median (IQR) or mean (SD). COVID-19=novel coronavirus disease 2019.
haemorrhage APACHE II=Acute Physiology and Chronic Health Evaluation II. FiO2=fraction of
Pneumothorax 1 (5%) 0 1 (2%) inspired oxygen. PaO2=partial pressure of oxygen. SARS-CoV-2=severe acute
Hospital-acquired 40 (20%) 2 (6%) 7 (13·5%) respiratory syndrome coronavirus 2. SOFA=Sequential Organ Failure Assessment.
pneumonia
Table 3: Differences in intensive care measures and vital signs between
Bacteraemia 0 1 (3%) 1 (2%) survivors and non-survivors of severe SARS-CoV-2 pneumonia
Urinary tract infection 0 1 (3%) 1 (2%)
Treatment
six (11·5%) with prone position ventilation, six (11·5%)
High flow nasal cannula 17 (85%) 16 (50%) 33 (63·5%)
with extracorporeal membrane oxygenation (ECMO),
Mechanical ventilation 7 (35%) 30 (94%) 37 (71%)
nine (17%) with renal replacement therapy, and
Non-invasive 6 (30%) 23 (72%) 29 (56%)
18 (35%) with vasoconstrictive agents (table 2). 23 (44%)
Invasive 3 (15%) 19 (59%) 22 (42%)
patients received antiviral agents, 49 (94%) received
Prone position ventilation 2 (10%) 4 (12·5%) 6 (11·5%)
antibacterial agents, and 30 (58%) patients received
Extracorporeal membrane 1 (5%) 5 (16%) 6 (11·5%)
oxygenation
glucocorticoids (table 2). Oseltamivir was given to
Renal replacement 1 (5%) 8 (25%) 9 (17%)
18 (35%) patients, ganciclovir to 14 (27%), and lopinavir
therapy to seven (13·5%).
Vasoconstrictive agents 2 (10%) 16 (50%) 18 (35%) For the primary outcome, among 52 critically ill
Antiviral agents 13 (65%) 10 (31%) 23 (44%) patients with SARS-CoV-2 infection, 32 (61·5%) patients
Antibacterial agents 19 (95%) 30 (94%) 49 (94%) had died at 28 days, and the median duration from ICU
Glucocorticoids 14 (70%) 16 (50%) 30 (58%) admission to death was 7 (IQR 3–11) days in the non-
Immunoglobulin 9 (45%) 19 (59%) 28 (54%) survivors (figure 2). Compared with survivors, non-
survivors were more likely to develop ARDS (26 [81%] vs
Data are n (%). SARS-CoV-2=severe acute respiratory syndrome coronavirus 2.
9 [45%]) and were more likely to receive mechanical
Table 2: Symptoms, comorbidities, and treatments of patients with ventilation (30 [94%] vs 7 [35%]). 30 (81%) of 37 patients
severe SARS-CoV-2 pneumonia requiring mechanical ventilation had died by 28 days.
Of the 20 patients who survived, eight patients were
resistant Klebsiella pneumoniae. Other microorganisms discharged. Three patients were still on invasive
identified from respiratory tract secretions in five (10%) ventilation at 28 days, including one patient who was also
patients included Aspergillus flavus, A fumigatus, extended- on ECMO. One patient was on non-invasive ventilation,
spectrum β-Lactamase (ESBL)-positive K pneumonia, two were using high-flow nasal cannula, and six were
ESBL-positive Pseudomonas aeruginosa, and ESBL-negative using common nasal cannula.
Serratia marcescens, with each microorganism found in Compared with survivors, non-survivors were older
one patient each. Candida albicans was identified in the (64·6 [SD 11·2] vs 51·9 [12·9]) and were more likely to
urine culture of one (2%) patient (table 2). have chronic medical illnesses (17 (53%) patients vs
33 (63·5%) patients were treated with high-flow 4 (20%) patients; table 1). Neither the median duration
nasal cannula, 37 (71%) with mechanical ventilation, from onset of symptoms to radiological confirmation of

4 www.thelancet.com/respiratory Published online February 21, 2020 https://doi.org/10.1016/S2213-2600(20)30079-5

 Articles

pneumonia or from onset of symptoms to ICU admis­ 100

sion were different between survivors and non-survivors
(table 3). The ratio of partial pressure of oxygen (PaO2) to
FiO2 was significantly lower in non-survivors. Based on 75

Survival probability (%)

APACHE II score and SOFA score at ICU admission,
non-survivors were in a more critical condition than 50
survivors. In the cohort, lymphocytopenia occurred in
44 (85%) patients, with no significant difference between
25
the two groups. Compared with survivors, non-survivors
Median duration from ICU admission to death was 7
were more likely to develop ARDS, and to receive (IQR 3–11)
mechanical ventilation, either invasively or non- 0
0 7 14 21 28
invasively.
Time since admission to intensive care unit (days)
Numbers at risk 51 39 28 22 21
Discussion
We report on 52 critically ill patients with con­firmed Figure 2: Survival of critically ill patients with SARS-CoV-2 pneumonia
SARS-CoV-2 infection, characterised by severe hypo­ Dashed lines represent 95% CIs. One patient died within 24 h after admission to
xaemia. 32 (61·5%) of critically ill patients had died at the intensive care unit (ICU).

28 days. Of all included patients, 37 (71%) required

mechanical ventilation and 35 (67%) had ARDS. reasonable that the mortality at 28 days of severe
Since no specialised medication to treat SARS-CoV-2 SARS-CoV-2 pneumonia is similar to the mortality of
infection has been identified at this time, the mainstay of severe ARDS, which is near 50%.17 With a substantial
treatment has been supportive care. Patients are being increase in the number of critically ill patients infected
treated in isolation, and their close contacts are being by SARS-CoV-2, more provisional ICUs are being
quarantined. For non-critically ill patients, close follow- established in Wuhan, China. Qualified specialists are
up is likely to be sufficient to manage the disease.1–3 For coming to Wuhan from other provinces of China and are
critically ill patients, however, aggressive treatments and currently working in these provisional ICUs, fever
intensive care are needed. To our knowledge, this is the clinics, and newly constructed hospitals.18 As the clinical
first study to characterise critically ill patients infected by capacity to treat patients improves, the mortality of
SARS-CoV-2. In three previously published studies of critically ill patients with SARS-CoV-2 pneumonia is
crtically ill patients, the patient numbers were too small expected to decrease.
to summarise the characteristics and mortality of these As mentioned in previous studies, nearly 70% of
patients with SARS-CoV-2 pneumonia.1,2,5 patients infected by SARS-CoV-2 were men.1,2 The
Like SARS-CoV and Middle Eastern respiratory patients are older in our study than in previous studies.1,2
syndrome (MERS)-CoV, SARS-CoV-2 is a coronavirus that We observed that non-survivors were older than
can be transmitted to humans, and these viruses are all survivors. Based on previous studies, evidence suggests
related to high mortality in critically ill patients.12 However, that older, male patients are the most susceptible to
the mortality rate in patients with SARS-CoV-2 infection SARS-CoV-2 infection,2 which is supported by our data.
in our cohort is higher than that previously seen in As previously reported, patients with a history of
critically ill patients with SARS. In a cohort of 38 critically cerebrovascular disease are at increased risk of becoming
ill patients with SARS from 13 hospitals in Canada, critically ill or dying if they have SARS-CoV-2 infection.2,5
29 (76%) patients required mechanical ventilation, In our cohort, fever is the most common symptom in
13 (43%) patients had died at 28 days, and six (16%) patients with SARS-CoV-2 pneumonia, which is in
patients remained on mechanical ventilation.8 17 (38%) of accordance with previous studies, but not all patients had
45 patients and 14 (26%) of 54 patients who were critically fever.1,2,5 We also found that fever was not detected at the
ill with SARS infection were also reported to have died at onset of illness in six (11·5%), and that it was in fact
28 days in a Singapore cohort13 and a Hong Kong cohort,14 detected 2–8 days later. The delay of fever manifestation
respectively. The mortality rate in our cohort is likely to be hinders early identification of patients infected with
higher than that seen in critically ill patients with MERS SARS-CoV-2—if patients are asymptomatic identification
infection. In a cohort of 12 patients with MERS from two of suspected cases is more difficult.19,20 The median
hospitals in Saudi Arabia, seven (58%) patients had died at duration from onset of symptoms to radiological
90 days.15 Since the follow-up time is shorter in our cohort, confirmation of pneumonia was 5 (3–7) days, meaning
we postulate that the mortality rate would be higher after that early or repeated radiological examinations are useful
28 days than that seen in patients with MERS-CoV. in screening patients with SARS-CoV-2 pneumonia.4
The fundamental pathophysiology of severe viral As for laboratory tests, lymphocytopenia occurred in
pneumonia is severe ARDS. Men and people of an older more than 80% of critically ill patients in our cohort.
age (>65 years) are more likely to develop ARDS than Lymphocytopenia is a prominent feature of critically ill
women or those of a younger age.16 Therefore, it is patients with SARS-CoV infection because targeted

www.thelancet.com/respiratory Published online February 21, 2020 https://doi.org/10.1016/S2213-2600(20)30079-5 5

 Articles

invasion by SARS-CoV viral particles damages the cohort study or potentially some randomly controlled
cytoplasmic component of the lymphocyte and causes its trials. Second, some specific information from the ICU
destruction.21 Additionally, lymphocytopenia is also was missing, such as mechanical ventilation settings.
common in the critically ill patients with MERS infection, The data on radiographical examination, supportive
which is the result of apoptosis of lymphocytes.22,23 treatment, living status, and the duration from ICU
Therefore, we postulate that necrosis or apoptosis of admission to death, however, are indisputable. Third,
lymphocytes also induces lymphocytopenia in critically ill this is a retrospective study. The data in this study permit
patients with SARS-CoV-2 infection. In a previous study, a preliminary assessment of the clinical course and
mainly in non-critical patients infected with SARS-CoV-2, outcomes of critically ill patients with SARS-CoV-2
35% of patients had only mild lymphocytopenia,2 suggest­ pneumonia. Further studies are still needed.
ing that the severity of lymphocytopenia reflects the In conclusion, the mortality of critically ill patients with
severity of SARS-CoV-2 infection. SARS-CoV-2 pneumonia is high. The survival term of the
Mechanical ventilation is the main supportive non-survivors is likely to be within 1–2 weeks after ICU
treatment for critically ill patients. The PaO2/FiO2 ratio admission. Older patients (>65 years) with comorbidities
was lower in our patients than in patients admitted to and ARDS are at increased risk of death. The severity of
Zhongnan Hospital.5 The substantial difference in SARS-CoV-2 pneumonia poses great strain to hospital
PaO2/FiO2 ratio between survivors and non-survivors in critical care resources, especially if they are not adequately
our study, indicates this ratio is associated with the staffed or resourced.
severity of illness and thus prognosis. Barotrauma seems Contributors
less severe in patients with SARS-CoV-2 infection who XY, YY, JXu, HS, HL, and YS collected the epidemiological and clinical
are being mechanically ventilated than that seen in data. JXi, YWu, LZ, ZY, MF, and TY summarised all data. XY, YY, HL,
JXi, YWa, SP, and YS drafted the manuscript. XZ and SY revised the
mechanically ventilated patients with SARS-CoV. In our final manuscript.
study, barotrauma occurred in only one (2%) patient,
Declaration of interests
who had been hospitalised for nearly 1 month, and they We declare no competing interests.
are currently on a ventilator and receiving ECMO. In
Data sharing
patients with SARS, barotrauma occurred in about 25% After publication, the data will be made available to others on reasonable
of patients on mechanical ventilation.14 The lower requests to the corresponding author. A proposal with detailed
occurrence of barotrauma in our cohort is probably description of study objectives and statistical analysis plan will be needed
related to the widely accepted strategy of protective for evaluation of the reasonability of requests. Additional materials
might also be required during the process of evaluation. Deidentified
ventilation in mainland China.24 At the same time, prone participant data will be provided after approval from the corresponding
position and ECMO have been used to treat patients with author and Wuhan Jin Yin-tan Hospital.
SARS-CoV-2 pneumonia. Acknowledgments
Without solid evidence, nearly half of the patients were We thank all patients and their families involved in the study.
given antiviral agents, and more than half were given References
intravenous glucocorticoids. Patients treated with lopinavir 1 Huang C, Wang Y, Li X, et al. Clinical features of patients infected
were from an ongoing clinical trial registered on Chinese with 2019 novel coronavirus in Wuhan, China. Lancet 2020;
395: 497–506.
Clinical Trial Registry (ChiCTR2000029308). Remdesivir 2 Chen N, Zhou M, Dong X, et al. Epidemiological and clinical
was given to the first patients with SARS-CoV-2 pneumonia characteristics of 99 cases of 2019 novel coronavirus pneumonia in
in the USA.4 Trials on remdesivir are about to recruit Wuhan, China: a descriptive study. Lancet 2020; 395: 507–13.
3 Zhu N, Zhang D, Wang W, et al. A novel coronavirus from patients
both mild to moderate patients (NCT04252664) and with pneumonia in China, 2019. N Engl J Med 2020; published
severe patients (NCT04257656) infected with SARS-CoV-2. online Jan 24. DOI:10.1056/NEJMoa2001017.
Although, intravenous glucocorticoids were commonly 4 Holshue ML, DeBolt C, Lindquist S, et al. First case of 2019 novel
coronavirus in the United States. N Engl J Med 2020; published
used in patients with severe SARS or MERS pneumonia, online Jan 31. DOI:10.1056/NEJMoa2001191.
their efficacy remains controversial and their use to treat 5 Wang D, Hu B, Hu C, Clinical characteristics of 138 hospitalized
SARS-CoV-2 infection is also controversial.5,25 An ongoing patients with 2019 novel coronavirus-infected pneumonia in
Wuhan, China. JAMA 2020; published online Feb 7. DOI:10.1001/
clinical trial (NCT04244591) might shed some light on the jama.2020.1585.
safety and efficacy of these drugs as treatment. 6 WHO. Clinical management of severe acute respiratory infection
This study has several limitations. First, only when novel coronavirus (nCoV) infection is suspected. Jan 11, 2020.
52 critically ill patients were included. However, the https://www.who.int/publications-detail/clinical-management-of-
severe-acute-respiratory-infection-when-novel-coronavirus-(ncov)-
population from which they were sampled was much infection-is-suspected (accessed Feb 8, 2020).
larger than that of the three studies previously 7 Dominguez-Cherit G, Lapinsky SE, Macias AE, et al. Critically ill
published.1,2,5 We included all the critically ill patients patients with 2009 influenza A(H1N1) in Mexico. JAMA 2009;
302: 1880–87.
being cared for in the ICU of Jin Yin-tan hospital who 8 Fowler RA, Lapinsky SE, Hallett D, et al. Critically ill patients with
met the inclusion criteria. Due to the exploratory nature severe acute respiratory syndrome. JAMA 2003; 290: 367–73.
of the study, which was not driven by formal hypotheses, 9 Kumar A, Zarychanski R, Pinto R, et al. Critically ill patients with
2009 influenza A(H1N1) infection in Canada. JAMA 2009;
the sample size calculation was waived. Instead, we hope 302: 1872–79.
that the findings presented here will encourage a larger

6 www.thelancet.com/respiratory Published online February 21, 2020 https://doi.org/10.1016/S2213-2600(20)30079-5

 Articles

10 The International Severe Acute Respiratory and Emerging Infection 18 USA TODAY. China built a hospital in 10 days to battle coronavirus.
Consortium (ISARIC). https://isaric.tghn.org/ (accessed https://www.usatoday.com/story/news/world/2020/02/03/
Jan 25, 2020). coronavirus-photos-show-wuhan-huoshenshan-hospital-built-10-
11 Kidney disease: improving global outcomes (KDIGO) acute kidney days/4643377002/ (accessed Feb 8, 2020).
injury work group. KDIGO clinical practice guideline for acute 19 Rothe C, Schunk M, Sothmann P, et al. Transmission of 2019-nCoV
kidney injury. March 2012. https://kdigo.org/wp-content/ infection from an asymptomatic contact in Germany. N Engl J Med
uploads/2016/10/KDIGO-2012-AKI-Guideline-English.pdf (accessed 2020; published online Jan 30. DOI:10.1056/NEJMc2001468.
Feb 8, 2020). 20 National Health Commission of the People’s Republic of China.
12 Parry J. Wuhan: Britons to be evacuated as scientists estimate Asymptomatic cases complicate efforts. http://en.nhc.gov.cn/2020–
44 000 cases of 2019-nCOV in the city. BMJ 2020; 368: m351. 02/01/c_76086.htm (accessed Feb 8, 2020).
13 Lew TW, Kwek TK, Tai D, et al. Acute respiratory distress syndrome 21 Gu J, Gong E, Zhang B, et al. Multiple organ infection and the
in critically ill patients with severe acute respiratory syndrome. pathogenesis of SARS. J Exp Med 2005; 202: 415–24.
JAMA 2003; 290: 374–80. 22 Chu H, Zhou J, Wong BH, et al. Middle East respiratory syndrome
14 Gomersall CD, Joynt GM, Lam P, et al. Short-term outcome of coronavirus efficiently infects human primary T lymphocytes and
critically ill patients with severe acute respiratory syndrome. activates the extrinsic and intrinsic apoptosis pathways. J Infect Dis
Intensive Care Med 2004; 30: 381–87. 2016; 213: 904–14.
15 Arabi YM, Arifi AA, Balkhy HH, et al. Clinical course and 23 Liu WJ, Zhao M, Liu K, et al. T-cell immunity of SARS-CoV:
outcomes of critically ill patients with Middle East respiratory implications for vaccine development against MERS-CoV.
syndrome coronavirus infection. Ann Intern Med 2014; Antiviral Res 2017; 137: 82–92.
160: 389–97. 24 Liu L, Yang Y, Gao Z, et al. Practice of diagnosis and management
16 de Wit E, van Doremalen N, Falzarano D, Munster VJ. SARS and of acute respiratory distress syndrome in mainland China: a cross-
MERS: recent insights into emerging coronaviruses. sectional study. J Thorac Dis 2018; 10: 5394–404.
Nat Rev Microbiol 2016; 14: 523–34. 25 Russell CD, Millar JE, Baillie JK. Clinical evidence does not support
17 Bellani G, Laffey JG, Pham T, et al. Epidemiology, patterns of care, corticosteroid treatment for 2019-nCoV lung injury. Lancet 2020;
and mortality for patients with acute respiratory distress syndrome 395: 473–75.
in intensive care units in 50 countries. JAMA 2016; 315: 788–800.

www.thelancet.com/respiratory Published online February 21, 2020 https://doi.org/10.1016/S2213-2600(20)30079-5 7