Risk Factors Associated With Acute Respiratory Distress Syndrome and Death in Patients With Coronavirus Disease 2019 Pneumonia in Wuhan, China | Critical Care Medicine | JAMA Internal Medicine | JAMA Network
[Skip to Content]
[Skip to Content Landing]
Figure.  Survival Curve in Patients With Acute Respiratory Distress Syndrome Who Did and Did Not Receive Methylprednisolone Treatment
Survival Curve in Patients With Acute Respiratory Distress Syndrome Who Did and Did Not Receive Methylprednisolone Treatment

Administration of methylprednisolone reduced the risk of death (hazard ratio, 0.38; 95% CI, 0.20-0.72; P = .003).

Table 1.  Demographic Characteristics of Patients With Coronavirus Disease 2019 Pneumonia
Demographic Characteristics of Patients With Coronavirus Disease 2019 Pneumonia
Table 2.  Initial Laboratory Indices of Patients With Coronavirus Disease 2019 Pneumonia
Initial Laboratory Indices of Patients With Coronavirus Disease 2019 Pneumonia
Table 3.  Clinical Characteristics and Initial Laboratory Indices Among Patients With and Without ARDS
Clinical Characteristics and Initial Laboratory Indices Among Patients With and Without ARDS
Table 4.  Bivariate Cox Regression of Factors Associated With ARDS Development or Progression From ARDS to Death
Bivariate Cox Regression of Factors Associated With ARDS Development or Progression From ARDS to Death
1.
World Health Organization. Coronavirus disease 2019 (COVID-19): situation report—37. February 25, 2020. Accessed February 26, 2020. https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200226-sitrep-37-covid-19.pdf?sfvrsn=6126c0a4_2.
2.
Zhu  N, Zhang  D, Wang  W,  et al; China Novel Coronavirus Investigating and Research Team.  A novel coronavirus from patients with pneumonia in China, 2019.   N Engl J Med. 2020;382(8):727-733. doi:10.1056/NEJMoa2001017PubMedGoogle ScholarCrossref
3.
Huang  C, Wang  Y, Li  X,  et al  Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China.   Lancet. 2020;395(10223):497-506. doi:10.1016/S0140-6736(20)30183-5Google ScholarCrossref
4.
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 study.   Lancet. 2020;395(10223):507-513. doi:10.1016/S0140-6736(20)30211-7PubMedGoogle ScholarCrossref
5.
Wang  D, Hu  B, Hu  C,  et al  Clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus-infected pneumonia in Wuhan, China.   JAMA. Published online February 7, 2020. doi:10.1001/jama.2020.1585Google Scholar
6.
World Health Organization. Clinical management of severe acute respiratory infection when novel coronavirus (nCoV) infection is suspected: interim guidance. January 28, 2020. Accessed March 5, 2020. https://www.who.int/publications-detail/clinical-management-of-severe-acute-respiratory-infection-when-novel-coronavirus-(ncov)-infection-is-suspected.
7.
Fine  MJ, Auble  TE, Yealy  DM,  et al.  A prediction rule to identify low-risk patients with community-acquired pneumonia.   N Engl J Med. 1997;336(4):243-250. doi:10.1056/NEJM199701233360402PubMedGoogle ScholarCrossref
8.
Schell-Chaple  HM, Puntillo  KA, Matthay  MA, Liu  KD; National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome Network.  Body temperature and mortality in patients with acute respiratory distress syndrome.   Am J Crit Care. 2015;24(1):15-23. doi:10.4037/ajcc2015320PubMedGoogle ScholarCrossref
9.
Channappanavar  R, Perlman  S.  Pathogenic human coronavirus infections: causes and consequences of cytokine storm and immunopathology.   Semin Immunopathol. 2017;39(5):529-539. doi:10.1007/s00281-017-0629-xPubMedGoogle ScholarCrossref
10.
Wang  YH, Lin  AS, Chao  TY,  et al.  A cluster of patients with severe acute respiratory syndrome in a chest ward in southern Taiwan.   Intensive Care Med. 2004;30(6):1228-1231. doi:10.1007/s00134-004-2311-8PubMedGoogle ScholarCrossref
11.
Nicholls  JM, Poon  LLM, Lee  KC,  et al.  Lung pathology of fatal severe acute respiratory syndrome.   Lancet. 2003;361(9371):1773-1778. doi:10.1016/S0140-6736(03)13413-7PubMedGoogle ScholarCrossref
12.
Ng  DL, Al Hosani  F, Keating  MK,  et al.  Clinicopathologic, immunohistochemical, and ultrastructural findings of a fatal case of Middle East respiratory syndrome coronavirus infection in the United Arab Emirates, April 2014.   Am J Pathol. 2016;186(3):652-658. doi:10.1016/j.ajpath.2015.10.024PubMedGoogle ScholarCrossref
13.
Min  CK, Cheon  S, Ha  NY,  et al.  Comparative and kinetic analysis of viral shedding and immunological responses in MERS patients representing a broad spectrum of disease severity.   Sci Rep. 2016;6:25359. doi:10.1038/srep25359PubMedGoogle ScholarCrossref
14.
Kim  ES, Choe  PG, Park  WB,  et al.  Clinical progression and cytokine profiles of Middle East respiratory syndrome coronavirus infection.   J Korean Med Sci. 2016;31(11):1717-1725. doi:10.3346/jkms.2016.31.11.1717PubMedGoogle ScholarCrossref
15.
Lew  TW, Kwek  TK, Tai  D,  et al.  Acute respiratory distress syndrome in critically ill patients with severe acute respiratory syndrome.   JAMA. 2003;290(3):374-380. doi:10.1001/jama.290.3.374PubMedGoogle ScholarCrossref
16.
Goronzy  JJ, Fang  F, Cavanagh  MM, Qi  Q, Weyand  CM.  Naive T cell maintenance and function in human aging.   J Immunol. 2015;194(9):4073-4080. doi:10.4049/jimmunol.1500046PubMedGoogle ScholarCrossref
17.
Zhou  P, Yang  XL, Wang  XG,  et al.  A pneumonia outbreak associated with a new coronavirus of probable bat origin.   Nature. Published online February 3, 2020. doi:10.1038/s41586-020-2012-7PubMedGoogle Scholar
18.
Hoffmann  M, Kleine-Weber  H, Krüger  N, Müller  M, Drosten  C, Pöhlmann  S.  The novel coronavirus 2019 (2019-nCoV) uses the SARS-coronavirus receptor 2 ACE2 and the cellular protease TMPRSS2 for entry into target cells.  Preprint. Posted online January 31, 2020. bioRxiv. doi:10.1101/2020.01.31.929042
19.
Gu  J, Gong  E, Zhang  B,  et al.  Multiple organ infection and the pathogenesis of SARS.   J Exp Med. 2005;202(3):415-424. doi:10.1084/jem.20050828PubMedGoogle ScholarCrossref
20.
Li  T, Qiu  Z, Zhang  L,  et al.  Significant changes of peripheral T lymphocyte subsets in patients with severe acute respiratory syndrome.   J Infect Dis. 2004;189(4):648-651. doi:10.1086/381535PubMedGoogle ScholarCrossref
21.
Cui  W, Fan  Y, Wu  W, Zhang  F, Wang  JY, Ni  AP.  Expression of lymphocytes and lymphocyte subsets in patients with severe acute respiratory syndrome.   Clin Infect Dis. 2003;37(6):857-859. doi:10.1086/378587PubMedGoogle ScholarCrossref
22.
Kim  KD, Zhao  J, Auh  S,  et al.  Adaptive immune cells temper initial innate responses.   Nat Med. 2007;13(10):1248-1252. doi:10.1038/nm1633PubMedGoogle ScholarCrossref
23.
Zhao  J, Zhao  J, Perlman  S.  T cell responses are required for protection from clinical disease and for virus clearance in severe acute respiratory syndrome coronavirus-infected mice.   J Virol. 2010;84(18):9318-9325. doi:10.1128/JVI.01049-10PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    9 Comments for this article
    EXPAND ALL
    Analysis may be confounded, mis-specified
    Jesse Heitner, ScD, Global Health | Aceso Global
    This study is very important because it  investigates what treatments are effective for COVID-19. It concludes that methylprednisolone may be beneficial; however it also states that patients who died were less likely to be treated with antiviral therapy (P<.001). All analyses are bivariate, which makes treatment effects un-interpretable, because bivariate analysis allows for high levels of confounding by age, comorbidity, and presenting disease severity. Worse, it does not account for which patients received both steroids and antivirals to investigate which, if either, was having a beneficial effect. The question of treatment efficacy is of extreme importance at this stage of the pandemic, and it must be analyzed in a multivariable mode.

    Further, a logit or probit model would be more appropriate for a dichotomous outcome such as died vs. recovered; a cox model (which is a time-to-event) model assumes all patients will reach the same outcome, with the differences being when, not if, a patient will die/reach ARDS.

    In sum, this study is timely and of potentially great use in treatment decisions, but only if a better (multivariable) analysis is done of the association between treatment methods and survival.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Mode of transmission
    Raouf Allim, MB BCh DA DRCOG | General Practitioner, UK
    Do we know anything about the effect of mode of transmission on the severity of illness? For example, presuming that the virus can spread by aerosol (5-10 micrometre droplets) as well as large droplets and surface contamination, is there a difference in disease severity with different modes of transmission?
    CONFLICT OF INTEREST: None Reported
    Risk Factors for COVID-19 and Conditioning Sets
    Michael McAleer, PhD(Econometrics),Queen's | Asia University, Taiwan
    The detailed diagnostic and empirical analysis of the effects of acute respiratory distress syndrome (ARDS) and 84 deaths from confirmed COVID-19 pneumonia, for a sample of 201 older age men and women from a single hospital in Wuhan with compromised immune systems, is highly informative.

    Not surprisingly, older age was found to be a leading risk factor leading to death.

    It would be instructive for healthcare workers if the sample were extended to a larger number of patients, according to differences in ages, stages and severity of pneumonia, types and levels of chronic underlying conditions, comorbidities,
    tobacco and alcohol history, types of coinfections, including influenza, follow-up of recovered patients for reinfection, discharge rates, and alternative model specifications.

    There are at least three possible categorical outcomes, namely full recovery, partial recovery (after follow-up), and death, and numerous deterministic factors, which affect the model specifications that can be used.

    Issues of how to deal with missing data, which reduces the sample size, a larger number of statistical tests, and diagnostic checks regarding the underlying assumptions of the models to provide more robust empirical analysis, would be useful in future research.

    Overall, the list of variables in the conditioning set is more extensive than has been incorporated in the invaluable empirical study.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Additional COVID-19 Information about Italy
    Stefano Olgiati, PhD (Epidemiology) | University of Bergamo, Bergamo 24129, Italy

    Wu et at (2020) report in the Introduction to their study that the coronavirus: "... has subsequently spread to other regions of China and 37 countries, including the United States, Japan, Australia, and France.

    The WHO (2020) Situation Report 37 has reported that  Italy is the country with the highest number of positive confirmed cases and deaths after China. This information is important with regard to the early chain of transmission of the virus.

    As of March 13, 2020, Italy still had the highest number of confirmed cases (24747) and of deaths (1809) after China (Dong et
    al, 2020). This is important because with regard to the chain and magnitude of transmission of the virus.

    Stefano Olgiati
    stefano.olgiati@unibg.it
    s.a.olgiati@gmail.com

    REFERENCES

    Wu C, Chen X, Cai Y, et al. Risk Factors Associated With Acute Respiratory Distress Syndrome and Death in Patients With Coronavirus Disease 2019 Pneumonia in Wuhan, China. JAMA Intern Med. Published online March 13, 2020. doi:10.1001/jamainternmed.2020.0994

    World Health Organization. Coronavirus disease 2019 (COVID-19): situation report—37. February 25, 2020. Accessed February 26, 2020. https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200226-sitrep-37-covid-19.pdf?sfvrsn=6126c0a4_2

    Dong E, Du H, Gardner L. An interactive web-based dashboard to track COVID-19 in real time. Lancet Infect Dis; published online Feb 19. https://doi.org/10.1016/S1473-3099(20)30120-1.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Application of methylprednisolone in COVID-19 pneumonia
    Yin Wang, M.D.,Ph.D. | Department of Cardiovascular Surgery, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology
    We read with interest in the original research article by  Wu et al. describing the clinical characteristics and factors associated with developing ARDS and progression from ARDS to death in patients with COVID-19 pneumonia. This is a very informative and timely study published in the context of COVID-19 pandemic and challenge worldwide. We have several concerns about the application of methylprednisolone in patients with COVID-19 pneumonia. Did the authors have any additional information regard to the dose and duration of methylprednisolone treatment, and were there any corticosteroid-induced complications observed in the patients who received methylprednisolone?

    The authors found administration
    of methylprednisolone appeared to reduce the risk of death in COVID-19 pneumonia patients with ARDS; however, of those who received methylprednisolone treatment, 23 of 50 patients died. This is rather high morality of approximate 50%; therefore, we are concerned that the evidence is not sufficient to declare the benefit of methylprednisolone treatment in the ‘real-world’ practice. Cytokine storm and viral evasion of cellular immune responses are thought to play important roles in the development of COVID-19 pneumonia, and histologic examination has shown diffuse alveolar damage and mucinous exudate similar to acute respiratory distress syndrome. Given corticosteroids as the conventional treatment for ARDS, it should be reasonable to consider the application of methylprednisolone in the treatment of severe COVID-19 pneumonia, but the issues of crucial importance are timing, dosage and duration.

    The study by Wu et al. highlights an important issue regard to the challenges in treatment of COVID-19 pneumonia. We agree with the authors that methylprednisolone should be applied in the treatment of severe cases, however, in terms of the indication, timing, dosage and duration, the application of methylprednisolone warrants further investigation. Therefore, the clinical applicability of the findings needs to be tempered owing to the unanswered questions that remain.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Conclusions on Methylprednisolone
    Hans-Joachim Kremer, Dr. rer. nat. | Medical Writing Service
    The authors’ conclusions on methylprednisolone and risk of deaths are unfounded and misleading. They wrote that methylprednisolone „decreased the risk of death” and based this statement on a Kaplan-Meier analysis with an impressively low p-value of 0.003. However, such an analysis is related to median time of survival rather than risk. In fact the median time in this selected subset differed by 15 days. But, are these 15 days really of any clinical relevance? Are these 15 days of benefit for the patients, considering that they spent this time not in freedom and good health condition but in the ICU and most likely under mechanical ventilation?
    Even these concerns are only lightly touching the crucial problem. In fact, 23 of the 50 (46%) with ARDS who received methylprednisolone died, compared with 21 of the other 34 (62%) with ARDS who did not receive this drug. At least for testing robustness of the conclusions, the authors should have performed an analysis of this difference by the Fisher’s Exact test; the 2-tailed version yields a p-value of 0.1856. This level cannot be considered as significant by any rule, thus, it cannot support the conclusions.
    The basic facts are even topping this concern, however. The data of this paper indicate that 23 of 62 (37%) who received methylprednisolone died, to be compared with 21 of the 139 (15%) who did not receive this drug, yielding a p-value of 0.000815 (Fisher’s Exact as above). Remember that all 201 patients reported here were selected because they were hospitalized and had pneumonia and evidence for COVID-19.
    The estimates of 37% vs. 15% are well in line with other reports from China [1, 2, 3].


    1. Wang D, Hu B, Hu C,et al. Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus–Infected Pneumonia in Wuhan, China. JAMA 2020;323(11):1061-1069. doi:10.1001/jama.2020.1585.
    2. Guan WJ, Ni ZY, Hu Y, Liang WH et al. Clinical Characteristics of Coronavirus Disease 2019 in China. N Engl J Med. 2020 Feb 28. doi: 10.1056/NEJMoa2002032.
    3. Zhou F, Yu T, Du R, Fan G, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. Lancet. 2020 Mar 11. pii: S0140-6736(20)30566-3. doi: 10.1016/S0140-6736(20)30566-3.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Disseminated pulmonary microvascular thromboembolism in COVID-19: a mechanistic link between coagulopathy and respiratory failure?
    anne sofie andreasen, MD, PhD, EDIC | Department of Anesthesiology and Intensive Care, Herlev and Gentofte Hospital, Denmark
    In this article, Wu and colleagues report that the marked COVID-19-associated coagulopathy is associated with a higher risk of developing acute respiratory distress syndrome (ARDS). This finding may suggest a causative link between deranged coagulation and the pulmonary pathophysiology of COVID-19. I posit that pulmonary thromboembolism, particularly disseminated pulmonary microvascular thromboembolism, is a major contributor to respiratory failure in these patients.

    Our preliminary clinical experience in Denmark is consistent with recently published data that COVID-19-associated respiratory distress differs from the usual ARDS presentation. Hence, despite severe hypoxemia, static pulmonary compliance is typically only mildly affected, and relatively low
    peak end-expiratory pressure levels during mechanical ventilation (i.e. often less than 10 cm H2O) are required to achieve sufficient oxygenation (1,2). Furthermore, chest CT findings of ground glass opacities, septal thickening, and linear opacities (3,4) are quite consistent with the lung parenchymal changes of diseases in the pulmonary vasculature, such as chronic embolic disease or chronic sickle cell disease (5). Finally, a retrospective study based on 449 COVID-19 patients found that anticoagulant therapy is associated with a lower mortality in patients with D-dimer levels > 6 fold of upper limit of normal (1). This study furthermore refers to a yet unpublished lung dissection that shows evidence of pulmonary microvascular thrombi and occlusion in a critically ill COVID-19 patient (1). Like the study by Wu and colleagues, the abovementioned findings are all consistent with pulmonary thromboembolism, notably disseminated pulmonary microvascular thromboembolism, as an important contributing factor to respiratory failure in critically ill COVID-19 patients.

    Regrettably, and although many of the patients did exhibit cardiovascular risk factors, Wu and colleagues do not report whether anticoagulant therapy, as it is noted for methylprednisolone treatment, is associated with a reduced risk of ARDS and/or mortality. Furthermore, it would be relevant to know whether any relevant chest imaging or autopsies were performed, and if so, whether any signs of pulmonary thromboembolic disease were noted. If so, I believe that prospective clinical trials on anticoagulant therapy for critically ill COVID-19 patients should be considered in the very near future.

    References
    1. Liu X, Liu X, Xu Y, et al. Ventilatory ratio in hypercapnic mechanically ventilated patients with COVID-19 associated ARDS. American Journal of Respiratory and Critical Care Medicine. 2020;In press. doi:10.1164/rccm.202002-0373LE
    2. Meng L, Qiu H, Wan L, et al. Intubation and ventilation amid the COVID-19 outbreak: Wuhan’s experience. Anesthesiology. 2020;In press. doi:10.1097/ALN.0000000000003296
    3. Liu KC, Xu P, Lv WF, et al. CT manifestations of coronavirus disease-2019: a retrospective analysis of 73 cases by disease severity. European Journal of Radiology. 2020;126. doi:10.1016/j.ejrad.2020.108941
    4. Li Y, Xia L. Coronavirus disease 2019 (COVID-19): role of chest CT in diagnosis and management. American Journal of Roentgenology. 2020;In press:1-7. doi:10.2214/AJR.20.22954
    5. Montani D, Achouh L, Dorfmüller P, et al. Pulmonary veno-occlusive disease: Clinical, functional, radiologic, and hemodynamic char
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Questions about Generalizability of Findings
    Stephen Mchugh, M.D. | University of Pittsburgh School of Medicine
    Dr. Wu and colleagues performed a study to describe demographic, laboratory, and treatment factors associated with the development of ARDS and death in patients with COVID-19 pneumonia.(1) However, specific details from this single-center study raise questions about widespread generalizability of the results.

    Treatment of ARDS frequently requires intubation and invasive mechanical ventilation. In their study, Wu et al report that 41.8% of patients developed ARDS and 21.9% died, but only 3% received invasive mechanical ventilation. Previous studies have suggested that greater than 85% of critically ill patients with ARDS will eventually need intubation and mechanical ventilation.(2,3) The reason for
    the limited use of this therapy was not discussed in the article. Did it represent a systematic difference in patterns of care or perhaps a scarcity of resources during a pandemic? This uncertainty makes it difficult to generalize their results, especially in regions where invasive mechanical ventilation is regularly used to treat ARDS.

    Furthermore, only 32.8% of patients in the study had pre-existing comorbidities. This is surprising in a population with a median age of 51 years. Additional discussion of this seemingly low degree of comorbidities is not provided. Given the ongoing COVID-19 pandemic, rapid publication of information like that in this study is important. However, more details on the patients and especially the treatment decisions for ARDS in the study would be helpful in guiding other clinicians’ decisions.

    Yin Wu, MD, MBA
    Julian H. Broad, MD
    Stephen M. McHugh, MD
    Department of Anesthesiology and Perioperative Medicine
    University of Pittsburgh School of Medicine


    References:
    1. Wu C, Chen X, Cai Y, et al. Risk Factors Associated With Acute Respiratory Distress Syndrome and Death in Patients With Coronavirus Disease 2019 Pneumonia in Wuhan, China [published online ahead of print, 2020 Mar 13]. JAMA Intern Med. 2020;e200994.
    2. Bellani G, Laffey JG, Pham T, et al. Epidemiology, Patterns of Care, and Mortality for Patients with Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries [published correction appears in JAMA. 2016 Jul 19;316(3):350]. JAMA. 2016;315(8):788–800.
    3. Kangelaris KN, Ware LB, Wang CY, et al. Timing of Intubation and Clinical Outcomes in Adults with Acute Respiratory Distress Syndrome. Crit Care Med. 2016;44(1):120–129.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Misleading language used to describe methylprednisolone results
    Kyle Koster, MD, MS | Weill Cornell Medicine, New York, New York
    Wu and colleagues reported that methylprednisolone administration "reduced the risk of death" in patients with acute respiratory distress syndrome (ARDS) associated with coronavirus disease 19 (COVID-19). We caution against this implication of causation, given the observational study design and lack of statistical adjustment for potential confounders. Furthermore, we contend that the data do not in fact demonstrate that methylprednisolone was associated with a reduced risk of death.

    The authors reached their conclusion based on a hazard ratio of 0.38 (95% CI, 0.20-0.72; P = .003), which they derived from the Kaplan-Meier method with log-rank test [1]. The hazard ratio
    estimates the odds that an individual in the group in the numerator reached the endpoint first [2]. Thus, the hazard ratio reported by Wu et al. tells us that a randomly selected patient who received methylprednisolone had an odds of 0.38 of dying sooner than a randomly selected patient who did not receive methylprednisolone. Converting odds to probability, it can be stated that patients receiving methylprednisolone had a 28% probability of dying sooner (or inversely, a 72% probability of living longer). However, the hazard ratio does not tell us whether the risk of death was different at the end of the study period, which may be a more important comparison, particularly when the duration of follow-up is short.

    Wu et al. reported that, among patients with ARDS, 23 of 50 (46%) of those treated with methylprednisolone died versus 21 of 34 (62%) of those not treated with methylprednisolone. Based on these data, the risk ratio of death at the end of the study period was 0.74 (95% CI, 0.50-1.11). The confidence interval includes the possibility of a substantial amount of harm, and, as pointed out by Dr. Kremer in an earlier comment, the result does not reach statistical significance (P = 0.19 using 2-sided Fisher exact test). Therefore, we should not infer that methylprednisolone was associated with a reduced risk of death.

    Understanding the difference between hazard ratio and risk ratio is important, as is the language we use to describe these different methods of analysis. We submit that a more accurately stated conclusion might be as follows: Among COVID-19 patients with ARDS, methylprednisolone administration was associated with a delay in time to death, but by the end of the study period, probability of death was similar in both groups.

    Kyle M. Koster, MD, MS
    Brett G. Fischer, MD
    Weill Department of Medicine
    Weill Cornell Medicine
    New York, New York

    References:

    1. Bland JM, Altman DG. The logrank test. BMJ 2004;328(7447):1073. doi:10.1136/bmj.328.7447.1073

    2. Spruance SL, Reid JE, Grace M, Samore M. Hazard ratio in clinical trials. Antimicrob Agents Chemother. 2004;48(8):2787–2792. doi:10.1128/AAC.48.8.2787-2792.2004
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Original Investigation
    March 13, 2020

    Risk Factors Associated With Acute Respiratory Distress Syndrome and Death in Patients With Coronavirus Disease 2019 Pneumonia in Wuhan, China

    Author Affiliations
    • 1Department of Pulmonary Medicine, QingPu Branch of Zhongshan Hospital Affiliated to Fudan University, Shanghai, China
    • 2Infection Division, Wuhan Jinyintan Hospital, Wuhan, China
    • 3Department of Pulmonary and Critical Care Medicine, Zhongshan Hospital, Fudan University, Shanghai, China
    • 4Tuberculosis and Respiratory Department, Wuhan Jinyintan Hospital, Wuhan, China
    • 5Department of Anesthesiology, International Peace Maternity and Child Health Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    • 6Department of Infectious Diseases, Fengxian Guhua Hospital, Shanghai, China
    • 7Department of Pulmonary Medicine, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    • 8Department of Critical Care Medicine, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    • 9Department of Respiratory Medicine, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    • 10Department of Emergency Medicine, Shanghai Pudong New Area Gongli Hospital, Shanghai, China
    • 11Shanghai Respiratory Research Institute, Shanghai, China
    • 12Department of Urology, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    • 13National Clinical Research Center for Aging and Medicine, Huashan Hospital, Fudan University, Shanghai, China
    JAMA Intern Med. 2020;180(7):934-943. doi:10.1001/jamainternmed.2020.0994
    Key Points

    Question  What clinical characteristics are associated with the development of acute respiratory distress syndrome (ARDS) and progression from ARDS to death among patients with coronavirus disease 2019 (COVID-19) pneumonia?

    Findings  In this cohort study involving 201 patients with confirmed COVID-19 pneumonia, risk factors associated with the development of ARDS and progression from ARDS to death included older age, neutrophilia, and organ and coagulation dysfunction. Treatment with methylprednisolone may be beneficial for patients who develop ARDS.

    Meaning  Risk for developing ARDS included factors consistent with immune activation; older age was associated with both ARDS development and death, likely owing to less robust immune responses.

    Abstract

    Importance  Coronavirus disease 2019 (COVID-19) is an emerging infectious disease that was first reported in Wuhan, China, and has subsequently spread worldwide. Risk factors for the clinical outcomes of COVID-19 pneumonia have not yet been well delineated.

    Objective  To describe the clinical characteristics and outcomes in patients with COVID-19 pneumonia who developed acute respiratory distress syndrome (ARDS) or died.

    Design, Setting, and Participants  Retrospective cohort study of 201 patients with confirmed COVID-19 pneumonia admitted to Wuhan Jinyintan Hospital in China between December 25, 2019, and January 26, 2020. The final date of follow-up was February 13, 2020.

    Exposures  Confirmed COVID-19 pneumonia.

    Main Outcomes and Measures  The development of ARDS and death. Epidemiological, demographic, clinical, laboratory, management, treatment, and outcome data were also collected and analyzed.

    Results  Of 201 patients, the median age was 51 years (interquartile range, 43-60 years), and 128 (63.7%) patients were men. Eighty-four patients (41.8%) developed ARDS, and of those 84 patients, 44 (52.4%) died. In those who developed ARDS, compared with those who did not, more patients presented with dyspnea (50 of 84 [59.5%] patients and 30 of 117 [25.6%] patients, respectively [difference, 33.9%; 95% CI, 19.7%-48.1%]) and had comorbidities such as hypertension (23 of 84 [27.4%] patients and 16 of 117 [13.7%] patients, respectively [difference, 13.7%; 95% CI, 1.3%-26.1%]) and diabetes (16 of 84 [19.0%] patients and 6 of 117 [5.1%] patients, respectively [difference, 13.9%; 95% CI, 3.6%-24.2%]). In bivariate Cox regression analysis, risk factors associated with the development of ARDS and progression from ARDS to death included older age (hazard ratio [HR], 3.26; 95% CI 2.08-5.11; and HR, 6.17; 95% CI, 3.26-11.67, respectively), neutrophilia (HR, 1.14; 95% CI, 1.09-1.19; and HR, 1.08; 95% CI, 1.01-1.17, respectively), and organ and coagulation dysfunction (eg, higher lactate dehydrogenase [HR, 1.61; 95% CI, 1.44-1.79; and HR, 1.30; 95% CI, 1.11-1.52, respectively] and D-dimer [HR, 1.03; 95% CI, 1.01-1.04; and HR, 1.02; 95% CI, 1.01-1.04, respectively]). High fever (≥39 °C) was associated with higher likelihood of ARDS development (HR, 1.77; 95% CI, 1.11-2.84) and lower likelihood of death (HR, 0.41; 95% CI, 0.21-0.82). Among patients with ARDS, treatment with methylprednisolone decreased the risk of death (HR, 0.38; 95% CI, 0.20-0.72).

    Conclusions and Relevance  Older age was associated with greater risk of development of ARDS and death likely owing to less rigorous immune response. Although high fever was associated with the development of ARDS, it was also associated with better outcomes among patients with ARDS. Moreover, treatment with methylprednisolone may be beneficial for patients who develop ARDS.

    Introduction

    Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was first reported in Wuhan, Hubei Province, China and has subsequently spread to other regions of China and 37 countries, including the United States, Japan, Australia, and France.1 SARS-CoV-2, which belongs to a unique clade of the sarbecovirus subgenus of the Orthocoronavirinae subfamily, was identified as the pathogen of coronavirus disease 2019 (COVID-19) in January 2020.2

    As reported by Huang et al,3 patients with COVID-19 present primarily with fever, myalgia or fatigue, and dry cough. Although most patients are thought to have a favorable prognosis, older patients and those with chronic underlying conditions may have worse outcomes. Patients with severe illness may develop dyspnea and hypoxemia within 1 week after onset of the disease, which may quickly progress to acute respiratory distress syndrome (ARDS) or end-organ failure.4 Certain epidemiological features and clinical characteristics of COVID-19 have been previously reported.3-5 However, these studies were based on relatively small sample sizes, and risk factors leading to poor clinical outcomes have not been well delineated. In this study, we report the clinical characteristics and factors associated with developing ARDS after hospital admission and progression from ARDS to death in patients with COVID-19 pneumonia from a single hospital in Wuhan, China.

    Methods
    Study Population

    This is a retrospective cohort study of 201 patients aged 21 to 83 years with confirmed COVID-19 pneumonia hospitalized at Jinyintan Hospital in Wuhan, China. All patients were diagnosed with COVID-19 pneumonia according to World Health Organization interim guidance.6 According to hospital data, patients were admitted from December 25, 2019, to January 26, 2020. Of 201 patients, 10 have been described previously by Chen et al4 and Huang et al.3 The ethics committee of Jinyintan Hospital approved this study and granted a waiver of informed consent from study participants.

    Procedures

    A trained team of physicians and medical students reviewed and collected epidemiological, clinical, and outcomes data from electronic medical records. Patients were followed up to February 13, 2020. The individual components of all definitions of clinical outcomes were recorded separately and checked by 2 authors (C.W. and X.C.). Patient confidentiality was protected by assigning a deidentified patient identification, and the electronic data was stored in a locked, password-protected computer.

    To identify SARS-CoV-2 infection, throat swab samples were obtained from all patients at admission and tested using real-time reverse transcriptase–polymerase chain reaction assays according to the same protocol described previously.3 The pathogenic detection was determined in 4 institutions (Chinese Center for Disease Control and Prevention, Chinese Academy of Medical Sciences, Academy of Military Medical Sciences, and Wuhan Institute of Virology of the Chinese Academy of Sciences) as described previously.4 Other respiratory pathogens, including respiratory syncytial virus, adenovirus, parainfluenza virus, influenza A virus, and influenza B virus, were also detected by real-time reverse transcriptase–polymerase chain reaction assays in 173 patients. Possible bacterial or fungal pathogens were detected by sputum culture. Additionally, patients underwent blood gas, blood routine test, coagulation, and biochemical tests and chest x-rays or computed tomography. The most intense level of oxygen support during hospitalization (nasal cannula, noninvasive mechanical ventilation [NMV], invasive mechanical ventilation [IMV], or IMV with extracorporeal membrane oxygenation [ECMO]) was recorded. The majority of the clinical data used in this study was collected from the first day of hospital admission unless indicated otherwise. To minimize interference of treatment during hospitalization, the highest patient temperature was defined using the self-reported highest temperature prior to hospital admission. Older age was classified as 65 years or older. Fever and high fever were classified as 37.3 °C or higher and 39 °C or higher, respectively.

    Outcomes

    Two outcomes were evaluated: development of ARDS and death among those with ARDS. World Health Organization interim guidance was used to define ARDS.6

    Statistical Analysis

    Descriptive analyses of the variables were expressed as median (interquartile range [IQR]), or number (%). Differences in distributions of patient characteristics by outcome subgroups are reported using differences with 95% CIs. Categorical data were compared using the χ2 test or the Fisher exact test. Nonnormal distributed continuous data were compared using Mann-Whitney-Wilcoxon test.

    Bivariate Cox proportional hazard ratio (HR) models were used to determine HRs and 95% CIs between individual factors on the development of ARDS or progression from ARDS to death. Sample size varied because of missing data (summarized in Tables 1 and 2). Survival curves were developed using the Kaplan-Meier method with log-rank test. Time to events (ARDS or death) were defined as the time from hospital admission to events.

    The analyses regarding different factors were based on nonmissing data, and missing data were not imputed. All tests were 2-sided, and a P value less than .05 was considered statistically significant. All analyses were performed with SPSS, version 23.0 (IBM SPSS), or R software, version 3.6.0 (R Foundation for Statistical Computing).

    Results
    Demographics and Characteristics

    A total of 201 patients were included in this study (Table 1). The median age was 51 years (IQR, 43-60 years), and 128 (63.7%) were male. The most commonly self-reported symptoms at onset of illness were fever (n = 188 [93.5%]), cough (n = 163 [81.1%]), productive cough (n = 83 [41.3%]), dyspnea (n = 80 [39.8%]), and fatigue or myalgia (n = 65 [32.3%]). The majority (n = 154 [76.6%]) of patients had fever with cough; 74 (36.8%) had fever with dyspnea; 66 (32.8%) had fever with fatigue, myalgia, or headache; and only 13 (6.5%) presented with fever alone (eTable 1 in the Supplement). A total of 191 (95.0%) patients had findings of bilateral infiltrates on radiographic imaging, while 10 (5.0%) patients had unilateral infiltrates. Sixty-six (32.8%) patients had comorbidities, including hypertension (n = 39 [19.4%]), diabetes (n = 22 [10.9%]), liver disease (n = 7 [3.5%]), nervous system disease (n = 7 [3.5%]), chronic lung disease (n = 5 [2.5%]), chronic kidney disease (n = 2 [1.0%]), endocrine system diseases not including diabetes (n = 2 [1.0%]), and tumors (n = 1 [0.5%]). Most (n = 173 [86.1%]) patients were tested for 9 additional respiratory pathogens. Bacteria and fungi cultures were collected from 148 (73.6%) patients. Only 1 patient was coinfected with influenza A virus.

    Treatments in Hospital

    Of the 201 patients, 165 (82.1%) required oxygen support in the hospital (Table 1). The most intense level was recorded, including nasal cannula (n = 98 [48.8%]), NMV (n = 61 [30.3%]), IMV (n = 5 [2.5%]), or IMV with ECMO (n = 1 [0.5%]). Among 201 patients, most (n = 196 [97.5%]) received empirical antibiotic treatment and antiviral therapy (n = 170 [84.6%]), including oseltamivir (n = 134 [66.7%]), ganciclovir (n = 81 [40.3%]), lopinavir/ritonavir (n = 30 [14.9%]), and interferon alfa (n = 22 [10.9%]). More than half (n = 106 [52.7%]) of patients received antioxidant therapy, including glutathione and N-acetyl-L-cysteine. Methylprednisolone was given to 62 (30.8%) patients, and immunomodulators, including immunoglobulin, thymosin, and recombinant human granulocyte colony stimulating factor, were given to 70 (34.8%) patients.

    Laboratory Indices

    Laboratory findings on hospital admission are summarized in Table 2. Of 194 patients, 166 (85.6%) demonstrated increased high-sensitivity C-reactive protein. More than half (126 of 197 [64.0%]) of this cohort had lymphocytopenia. About one-third (68 of 197 [34.5%]) of patients had neutrophilia. Approximately one-quarter (46 of 197 [23.4%]) of patients had leukocytosis. Some patients demonstrated liver injury with elevated aspartate aminotransferase (AST; 59 of 198 [29.8%]) and alanine aminotransferase (ALT; 43 of 198 [21.7%]). Most patients presented with an elevated myocardial indices: 135 of 198 (68.2%) had elevated lactate dehydrogenase (LDH), and 9 of 198 (4.5%) had an elevated creatine kinase muscle-brain isoform. Few patients had kidney injury indicated by elevated plasma urea (9 of 198 [4.5%]) and serum creatinine (9 of 198 [4.5%]). Of 195 patients, 4 (2.1%) presented with prolonged prothrombin times (PTs).

    Clinical Outcomes

    As of February 13, 2020, 144 of the total 201 patients (71.6%) were discharged from the hospital. The median hospital stay was 13 days (IQR, 10-16 days), and 13 (6.5%) patients were still hospitalized. Of the entire cohort, 84 (41.8%) patients developed ARDS, 53 (26.4%) were admitted to the intensive care unit, 67 (33.3%) received mechanical ventilation, and 44 (21.9%) died. Among the 67 patients who received mechanical ventilation, 44 (65.7%) died, 14 (20.9%) were discharged from the hospital, and 9 (13.4%) remained hospitalized. The median time from admission to developing ARDS was 2 days (IQR, 1-4 days). All of the patients who died had developed ARDS and received mechanical ventilation.

    Table 3 demonstrates that when compared with patients without ARDS, patients with ARDS were older (difference, 12.0 years; 95% CI, 8.0-16.0 years; P < .001) and had higher temperature prior to admission (difference, 0.30 °C; 95% CI, 0.00-0.50 °C; P = .004). More patients with ARDS presented with initial symptoms of dyspnea compared with those without ARDS (difference, 33.9%; 95% CI, 19.7%-48.1%; P < .001). Compared with patients without ARDS, patients with ARDS had a higher proportion of comorbidities, including hypertension (difference, 13.7%; 95% CI, 1.3%-26.1%; P = .02) and diabetes (difference, 13.9%; 95% CI, 3.6%-24.2%; P = .002). In addition, when compared with patients who did not have ARDS, patients who developed ARDS were less likely to be treated with antiviral therapy (difference, −14.4%; 95% CI, −26.0% to −2.9%; P = .005) and more likely to be treated with methylprednisolone (difference, 49.3%; 95% CI, 36.4%-62.1%; P < .001). Of 84 patients with ARDS, 61 (72.6%) received NMV, 17 (20.2%) received nasal cannula, 5 (6.0%) received IMV, and 1 (1.2%) received IMV with ECMO.

    Compared with patients without ARDS, for patients with ARDS, the value of liver damage indices (total bilirubin [difference, 1.90 μM; 95% CI, 0.60-3.30 μM; P = .004]), renal dysfunction indices (urea [difference, 1.69 mM; 95% CI, 1.10-2.29 mM; P < .001]), inflammation-related indices (interleukin-6 [IL-6] [difference, 0.93 pg/mL; 95% CI, 0.07-1.98 pg/mL; P = .03]), and coagulation function indices (D-dimer [difference, 0.52 μg/mL; 95% CI, 0.21-0.94 μg/mL; P < .001]) were significantly elevated. However, lymphocyte counts (difference, −0.34 ×109/L; 95% CI, −0.47 to −0.22 ×109/L; P < .001) and CD8 T cells (difference, −66.00 cells/μL; 95% CI, −129.00 to −7.00 cells/μL; P = .03) were significantly decreased.

    As summarized in Table 4, older age (≥65 years old), high fever (≥39 °C), comorbidities (eg, hypertension, diabetes), neutrophilia, lymphocytopenia (as well as lower CD3 and CD4 T-cell counts), elevated end-organ related indices (eg, AST, urea, LDH), elevated inflammation-related indices (high-sensitivity C-reactive protein and serum ferritin), and elevated coagulation function–related indicators (PT and D-dimer) were significantly associated with higher risks of the development of ARDS. Patients who received treatment with methylprednisolone appear to have been sicker than patients who did not receive it. Specifically, a higher proportion of patients who received methylprednisolone were classified into a higher grade on the Pneumonia Severity Index7 compared with patients who did not receive methylprednisolone (P = .01; eTable 2 in the Supplement).

    In the subgroup of patients who developed ARDS, patients who ultimately died were older (difference, 18.0 years; 95% CI, 13.0-23.0 years; P < .001) and had lower proportion of high fever (difference, −31.8%; 95% CI, −56.5% to −7.1%; P = .007) than those who survived. They also had higher proportions of hypertension (difference, 18.9%; 95% CI, −2.0% to 39.7%; P = .05). The patients who died were less likely to be treated with antiviral therapy (difference, −40.7%; 95% CI, −58.5% to −22.9%; P < .001). Regarding the most intense level of oxygen support among the 44 ARDS patients who died, 38 (86.4%) received NMV, 5 (11.4%) received IMV, and 1 (2.3%) received IMV with ECMO.

    For patients with ARDS who died, the value of liver damage indices (total bilirubin [difference, 2.60 μM; 95% CI, 0.30-5.20 μM; P = .03]), renal dysfunction indices (urea [difference, 1.50 mM; 95% CI, 0.50-2.70 mM; P = .004]), inflammation-related indices (IL-6 [difference, 3.88 pg/mL; 95% CI, 2.20-6.13 pg/mL; P < .001]), and coagulation function indices (D-dimer [difference, 2.10 μg/mL; 95% CI, 0.89-5.27 μg/mL; P = .001]) were significantly elevated compared with patients with ARDS who survived. However, lymphocyte counts (difference, −0.23 ×109/L; 95% CI, −0.41 to −0.07 ×109/L; P = .004) and CD8 T cells (difference, −134 cells/μL; 95% CI, −221 to −10 cells/μL; P = .05) were significantly decreased (Table 3).

    Bivariate Cox models showed that several factors related to the development of ARDS were not associated with death, which included comorbidities, lymphocyte counts, CD3 and CD4 T-cell counts, AST, prealbumin, creatinine, glucose, low-density lipoprotein, serum ferritin, and PT. However, IL-6 was statistically significantly associated with death (Table 4). Although high fever was associated with higher likelihood of developing ARDS (HR, 1.77; 95% CI, 1.11-2.84), it was negatively associated with death (HR, 0.41; 95% CI, 0.21-0.82).

    Finally, among the patients with ARDS, of those who received methylprednisolone treatment, 23 of 50 (46.0%) patients died, while of those who did not receive methylprednisolone treatment, 21 of 34 (61.8%) died. The administration of methylprednisolone appears to have reduced the risk of death in patients with ARDS (HR, 0.38; 95% CI, 0.20-0.72; P = .003) (Figure).

    Discussion

    In this cohort study, we reported the clinical characteristics and risk factors associated with clinical outcomes in patients with COVID-19 pneumonia who developed ARDS after admission, as well as those who progressed from ARDS to death. Patients who received methylprednisolone treatment were much more likely to develop ARDS likely owing to confounding by indication; specifically, sicker patients were more likely to be given methylprednisolone. However, administration of methylprednisolone appeared to reduce the risk of death in patients with ARDS. These findings suggest that for patients with COVID-19 pneumonia, methylprednisolone treatment may be beneficial for those who have developed ARDS on disease progression. However, these results should be interpreted with caution owing to potential bias and residual confounding in this observational study with a small sample size. Double-blinded randomized clinical trials should be conducted to validate these results.

    The risk factors related to the development of ARDS and progression from ARDS to death included older age, neutrophilia, and organ and coagulation dysfunction (eg, higher LDH and D-dimer). In addition, we observed that several factors associated with the development of ARDS were not associated with death (eg, comorbidities, lymphocyte counts, CD3 and CD4 T-cell counts, AST, prealbumin, creatinine, glucose, low-density lipoprotein, serum ferritin, PT). Moreover, the difference in median D-dimer between the death and survival groups was larger than that between the ARDS and non-ARDS groups, which suggests that disseminated intravascular coagulation was on the pathway to death in some patients. Interestingly, although high fever was positively associated with development of ARDS, it was negatively related to death, which is consistent with results noted in a study by Schell-Chaple et al.8 However, the differences in patient temperature between the groups were very small and self-reported before hospital admission, thus the data regarding high fever should be cautiously interpreted.

    The pathogenesis of highly pathogenic human coronavirus is still not completely understood. Cytokine storm and viral evasion of cellular immune responses are thought to play important roles in disease severity.9 Neutrophilia was found in both the peripheral blood10 and lung11 of patients with SARS-CoV. The severity of lung damage correlated with extensive pulmonary infiltration of neutrophils and macrophages and higher numbers of these cells in the peripheral blood in patients with Middle East respiratory syndrome.12-14 Neutrophils are the main source of chemokines and cytokines. The generation of cytokine storm can lead to ARDS, which is a leading cause of death in patients with severe acute respiratory syndrome 15 and Middle East respiratory syndrome.14 In this study, patients with COVID-19 pneumonia who had developed ARDS had significantly higher neutrophil counts than did those without ARDS, perhaps leading to the activation of neutrophils to execute an immune response against the virus, but also contributing to cytokine storm. This may partly explain the positive association of high fever and ARDS found at the early stages of COVID-19. In addition, considering that older age is associated with declined immune competence,16 the results of the present study showed that older age was associated with both ARDS and death. Therefore, older age related to death may be due to less robust immune responses.

    The results of this study show that higher CD3 and CD4 T-cell counts might protect patients from developing ARDS, but similar results were not observed when examined for death, possibly because of limited sample size. CD8 counts were significantly higher in those who were alive. These results indicate the important roles of CD4 and CD8 T cells in COVID-19 pneumonia. Earlier studies have revealed that SARS-CoV, which was reported to share the same cell entry receptors with SARS-CoV-2,17,18 could infect immune cells, including T lymphocytes, monocytes, and macrophages.19 The CD3, CD4, and CD8 T-cell counts decreased at the onset of illness; this decrease persisted until the recovery period of SARS-CoV pneumonia.19 In addition, CD4 and CD8 T-cell counts decreased in the peripheral blood specimen of patients with fatal SARS-CoV pneumonia10,20,21, which was consistent with these results that patients with COVID-19 pneumonia and ARDS presented with lymphocytopenia (CD3, CD4, and CD8 T cells). Studies demonstrated that T-cell responses can inhibit the overactivation of innate immunity.22 T cells were reported to help clear SARS-CoV, and a suboptimal T-cell response was found to cause pathological changes observed in mice with SARS-CoV.23 We hypothesized that persistent and gradual increases in lymphocyte responses might be required for effective immunity against SARS-CoV-2 infection. Further studies are needed to characterize the role of the neutrophil and lymphocyte response or that of CD4 and CD8 T-cell immune response in SARS-CoV-2 infection.

    Limitations

    This study has several limitations. First, owing to limited medical resources, only patients with relatively severe COVID-19 pneumonia were hospitalized during this period. Second, this study was conducted at a single-center hospital with limited sample size. As such, this study may have included disproportionately more patients with poor outcomes. There may also be a selection bias when identifying factors that influence the clinical outcomes. A larger cohort study of patients with COVID-19 pneumonia from Wuhan, China, other cities in China, and other countries would help to further define the clinical characteristics and risk factors of the disease.

    Conclusions

    Older age was associated with greater risk of developing ARDS and death, likely because of less rigorous immune response. Although fever was associated with the development of ARDS, it was also associated with better outcomes. Several factors related to the development of ARDS were not associated with death, which indicates that different pathophysiological changes from hospital admission to development of ARDS and from development of ARDS to death may exist. Moreover, treatment with methylprednisolone may be beneficial for patients who develop ARDS. Double-blinded randomized clinical trials to determine the most effective treatments for COVID-19 are still needed.

    Back to top
    Article Information

    Published Online: March 13, 2020. doi:10.1001/jamainternmed.2020.0994

    Correction: This article was corrected on May 11, 2020, to fix data errors and correct units of measure used in the Results section and Tables 2, 3, and 4.

    Accepted for Publication: March 3, 2020.

    Corresponding Authors: Yuanlin Song, MD, Department of Pulmonary and Critical Care Medicine, Zhongshan Hospital, Fudan University, 180 Fenglin Rd, Shanghai 200032, China (ylsong70_02@163.com); Junhua Zheng, MD, Department of Urology, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, 85 Wujin Rd, Shanghai 200080, China (zhengjh0471@sina.com).

    Author Contributions: Drs Song and Zheng had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Wu, X. Chen, Cai, and Xia contributed equally and share first authorship. Drs Song and Zheng contributed equally to the study.

    Study concept and design: Wu, X. Chen, Cai, Xia, Zheng, Y. Song.

    Acquisition, analysis, or interpretation of data: X. Chen, Xing Zhou, S. Xu, Huang, L. Zhang, Xia Zhou, Du, Y. Zhang, J. Song, Wang, Chao, Yang, J. Xu, Xin Zhou, D. Chen, Xiong, L. Xu, F. Zhou, Jiang, Bai.

    Drafting of the manuscript: X. Chen, Xia, Xing Zhou, S. Xu, Huang, L. Zhang, Xia Zhou, Du, Y. Zhang, J. Song, Wang, Chao, Yang, J. Xu, Xin Zhou, D. Chen, Xiong, L. Xu, F. Zhou, Jiang, Bai.

    Critical revision of the manuscript for important intellectual content: Wu, X. Chen, Cai, Xia, Du, Zheng, Y. Song.

    Statistical analysis: X. Chen, Y. Song.

    Obtained funding: Wu, Zheng, Y. Song.

    Administrative, technical, or material support: Cai, Xia, Xing Zhou, S. Xu, Huang, L. Zhang, Xia Zhou, Yang, J. Xu, Xin Zhou, D. Chen, Xiong, L. Xu, F. Zhou, Jiang, Bai, Zheng, Y. Song.

    Study supervision: Zheng, Y. Song.

    Conflict of Interest Disclosures: None reported.

    Funding/Support: This study was supported by a grant from Prevention and Treatment of Infection in Novel Coronavirus Pneumonia Patients from the Shanghai Science and Technology Committee (to Dr Yuanlin Song), the Special Fund of Shanghai Jiaotong University for Coronavirus Disease 2019 Control and Prevention (2020RK47 to Dr Junhua Zheng), and Academic Leader of Shanghai Qingpu District Healthcare Commission (WD2019-36 to Dr Chaomin Wu).

    Role of the Funder/Sponsor The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Additional Contributions: We thank Weibing Wang, PhD, from the Department of Epidemiology, School of Public Health, Fudan University; Yan Liu, MD, and Hongcai Shang, PhD, from Key Laboratory of Chinese Internal Medicine of Ministry of Education, Dongzhimen Hospital, Beijing University of Chinese Medicine; Xiaojia Huang, PhD, from the Institute of Biomedical Engineering and Health Sciences, Changzhou University; and Yaohui Li, MD, from Zhongshan Hospital, Fudan University for their statistical analysis discussion. We also thank Dongni Hou, MD, Sucheng Mu, MD, Donghui Zhang, MD, and Ke Lang, BD, from Zhongshan Hospital, Fudan University for their data collection. They were not compensated for their contributions.

    References
    1.
    World Health Organization. Coronavirus disease 2019 (COVID-19): situation report—37. February 25, 2020. Accessed February 26, 2020. https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200226-sitrep-37-covid-19.pdf?sfvrsn=6126c0a4_2.
    2.
    Zhu  N, Zhang  D, Wang  W,  et al; China Novel Coronavirus Investigating and Research Team.  A novel coronavirus from patients with pneumonia in China, 2019.   N Engl J Med. 2020;382(8):727-733. doi:10.1056/NEJMoa2001017PubMedGoogle ScholarCrossref
    3.
    Huang  C, Wang  Y, Li  X,  et al  Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China.   Lancet. 2020;395(10223):497-506. doi:10.1016/S0140-6736(20)30183-5Google ScholarCrossref
    4.
    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 study.   Lancet. 2020;395(10223):507-513. doi:10.1016/S0140-6736(20)30211-7PubMedGoogle ScholarCrossref
    5.
    Wang  D, Hu  B, Hu  C,  et al  Clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus-infected pneumonia in Wuhan, China.   JAMA. Published online February 7, 2020. doi:10.1001/jama.2020.1585Google Scholar
    6.
    World Health Organization. Clinical management of severe acute respiratory infection when novel coronavirus (nCoV) infection is suspected: interim guidance. January 28, 2020. Accessed March 5, 2020. https://www.who.int/publications-detail/clinical-management-of-severe-acute-respiratory-infection-when-novel-coronavirus-(ncov)-infection-is-suspected.
    7.
    Fine  MJ, Auble  TE, Yealy  DM,  et al.  A prediction rule to identify low-risk patients with community-acquired pneumonia.   N Engl J Med. 1997;336(4):243-250. doi:10.1056/NEJM199701233360402PubMedGoogle ScholarCrossref
    8.
    Schell-Chaple  HM, Puntillo  KA, Matthay  MA, Liu  KD; National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome Network.  Body temperature and mortality in patients with acute respiratory distress syndrome.   Am J Crit Care. 2015;24(1):15-23. doi:10.4037/ajcc2015320PubMedGoogle ScholarCrossref
    9.
    Channappanavar  R, Perlman  S.  Pathogenic human coronavirus infections: causes and consequences of cytokine storm and immunopathology.   Semin Immunopathol. 2017;39(5):529-539. doi:10.1007/s00281-017-0629-xPubMedGoogle ScholarCrossref
    10.
    Wang  YH, Lin  AS, Chao  TY,  et al.  A cluster of patients with severe acute respiratory syndrome in a chest ward in southern Taiwan.   Intensive Care Med. 2004;30(6):1228-1231. doi:10.1007/s00134-004-2311-8PubMedGoogle ScholarCrossref
    11.
    Nicholls  JM, Poon  LLM, Lee  KC,  et al.  Lung pathology of fatal severe acute respiratory syndrome.   Lancet. 2003;361(9371):1773-1778. doi:10.1016/S0140-6736(03)13413-7PubMedGoogle ScholarCrossref
    12.
    Ng  DL, Al Hosani  F, Keating  MK,  et al.  Clinicopathologic, immunohistochemical, and ultrastructural findings of a fatal case of Middle East respiratory syndrome coronavirus infection in the United Arab Emirates, April 2014.   Am J Pathol. 2016;186(3):652-658. doi:10.1016/j.ajpath.2015.10.024PubMedGoogle ScholarCrossref
    13.
    Min  CK, Cheon  S, Ha  NY,  et al.  Comparative and kinetic analysis of viral shedding and immunological responses in MERS patients representing a broad spectrum of disease severity.   Sci Rep. 2016;6:25359. doi:10.1038/srep25359PubMedGoogle ScholarCrossref
    14.
    Kim  ES, Choe  PG, Park  WB,  et al.  Clinical progression and cytokine profiles of Middle East respiratory syndrome coronavirus infection.   J Korean Med Sci. 2016;31(11):1717-1725. doi:10.3346/jkms.2016.31.11.1717PubMedGoogle ScholarCrossref
    15.
    Lew  TW, Kwek  TK, Tai  D,  et al.  Acute respiratory distress syndrome in critically ill patients with severe acute respiratory syndrome.   JAMA. 2003;290(3):374-380. doi:10.1001/jama.290.3.374PubMedGoogle ScholarCrossref
    16.
    Goronzy  JJ, Fang  F, Cavanagh  MM, Qi  Q, Weyand  CM.  Naive T cell maintenance and function in human aging.   J Immunol. 2015;194(9):4073-4080. doi:10.4049/jimmunol.1500046PubMedGoogle ScholarCrossref
    17.
    Zhou  P, Yang  XL, Wang  XG,  et al.  A pneumonia outbreak associated with a new coronavirus of probable bat origin.   Nature. Published online February 3, 2020. doi:10.1038/s41586-020-2012-7PubMedGoogle Scholar
    18.
    Hoffmann  M, Kleine-Weber  H, Krüger  N, Müller  M, Drosten  C, Pöhlmann  S.  The novel coronavirus 2019 (2019-nCoV) uses the SARS-coronavirus receptor 2 ACE2 and the cellular protease TMPRSS2 for entry into target cells.  Preprint. Posted online January 31, 2020. bioRxiv. doi:10.1101/2020.01.31.929042
    19.
    Gu  J, Gong  E, Zhang  B,  et al.  Multiple organ infection and the pathogenesis of SARS.   J Exp Med. 2005;202(3):415-424. doi:10.1084/jem.20050828PubMedGoogle ScholarCrossref
    20.
    Li  T, Qiu  Z, Zhang  L,  et al.  Significant changes of peripheral T lymphocyte subsets in patients with severe acute respiratory syndrome.   J Infect Dis. 2004;189(4):648-651. doi:10.1086/381535PubMedGoogle ScholarCrossref
    21.
    Cui  W, Fan  Y, Wu  W, Zhang  F, Wang  JY, Ni  AP.  Expression of lymphocytes and lymphocyte subsets in patients with severe acute respiratory syndrome.   Clin Infect Dis. 2003;37(6):857-859. doi:10.1086/378587PubMedGoogle ScholarCrossref
    22.
    Kim  KD, Zhao  J, Auh  S,  et al.  Adaptive immune cells temper initial innate responses.   Nat Med. 2007;13(10):1248-1252. doi:10.1038/nm1633PubMedGoogle ScholarCrossref
    23.
    Zhao  J, Zhao  J, Perlman  S.  T cell responses are required for protection from clinical disease and for virus clearance in severe acute respiratory syndrome coronavirus-infected mice.   J Virol. 2010;84(18):9318-9325. doi:10.1128/JVI.01049-10PubMedGoogle ScholarCrossref
    ×