Clinical characteristics and risk factors for severe-critically ill COVID-19 adult patients in Jiangsu, China: a multiple-centered, retrospective study

Background A number of reports have documented the clinical characteristics of patients with severe coronavirus disease 2019 (COVID-19) in Wuhan. Clinical features of severe-critically ill COVID-19 patients in Jiangsu, outside Wuhan, remains unknown. Methods This multi-centered retrospective study collected the information of 631 laboratory-confirmed COVID-19 patients hospitalized at 28 authorized hospitals in Jiangsu province between January 23, 2019 and March 13, 2020. Epidemiological and demographic information, clinical and radiological characteristics, laboratory results and treatment of these patients were analyzed. Results A total of 583 adult patients with laboratory-confirmed COVID-19 were enrolled for final analysis, including 84 severe-critically ill patients and 499 mild-moderate patients. Median age of the severe-critically ill patients was 57.0 years [interquartile range (IQR), 49.0-65.8], and 50 (59.5%) were males. Multisystemic laboratory abnormalities were observed on admission in severe-critically ill patients. The severe-critically ill patients showed more noticeable radiologic abnormalities and more coexisting health issues as compared to mild-moderate patients. Most of the severe-critically ill COVID-19 patients become deteriorated in two weeks after diagnosis. Age [odds ratio (OR) 1.08, 95% confidence interval (CI) (1.03-1.14)], D-dimer (OR 3.21, 95% CI 1.39-7.40), and lymphocytes (OR 0.28, 95% CI 0.04-0.88) were independently associated with the progression of severe-critically illness. Conclusions Older age, higher D-dimer levels and less lymphocyte counts on admission are potential risk factors for COVID-19 patients to develop into severe and critically illness. The results would help clinicians to identify high-risk patients in advance. will be asked from the involved health-care providers who subsequently will collect or communicate with patients’ families. We collected data about age, gender, smoking history, exposure history, coexisting disorders (hypertension, diabetes, chronic obstructive pulmonary disease (COPD), coronary heart disease, cerebrovascular disease, chronic renal diseases, hyperlipidemia, hepatitis B infection, connective tissue disease, cancer, pregnancy), symptoms and signs on admission (fever, cough, sputum production, hemoptysis, shortness of breath, sore throat, nasal congestion, rhinorrhea, headache, chest pain, fatigue, nausea, vomiting, diarrhea, mylgia, arthralgia, chill, throat congestion), radiographic imagings, laboratory test results (leukocyte, lymphocyte, neutrophils, platelet, hemoglobin, C-reactive protein (CRP), procalcitonin, lactose dehydrogenase (LDH), aspartate aminotransferase (AST), alanine aminotransferase (ALT), total bilirubin, creatine kinase (CK), urea nitrogen, creatinine, D-dimer, prothrombin time (PT), activated partial thromboplastin time (APTT), sodium, potassium), complications (sepsis, septic shock, respiratory failure, ARDS, acute kidney injury, acute cardiac injury, secondary infections), and treatment (antiviral drugs, antibiotics, antifungal administration, systemic corticosteroids, oxygen therapy, mechanical ventilation (MV), extracorporeal membrane oxygenation (ECMO), renal replacement therapy, intravenous immunoglobin). For severe-critically ill patients, the acute physiology and chronic health evaluation II score (APACHE II) and sequential organ failure assessment (SOFA) were determined to assess disease severity. records of 583 laboratory-confirmed patients hospitalized in 28 The study of COVID-19 patients outside Wuhan is of paramount significance for an in-depth understanding of the clinical characteristics of COVID-19. The purpose of the study aims to further define the risk factors for the progression into severe-critically illness. Our results showed that older age, higher D-dimer levels and less lymphocyte counts are closely related to the development into serious illness.


Background
Coronavirus disease 2019 (COVID-19) is an infectious disease caused by the most recently detected novel coronavirus, named as SARS-CoV-2 [1]. This coronavirus and disease remained unknown until the outbreak in Wuhan, the capital of central China's Hubei Province, in December 2019. The exponential growth of cases and its expanding geographical transmission have raised international concerns. On May 9, 2020, there were 3,855,788 confirmed COVID-19 patients in 215 countries, areas or territories in all five continents, and 265,862 deaths have been reported all over the world [2]. At present, COVID-19 is a pandemic infectious disease recognized by the World Health Organization (WHO) [3].
So far, China have significantly declining epidemics. Of the exceeding 80,000 reported cases in China, more than 90% have recovered and been discharged. Several studies have described the clinical characteristics and mortality of the COVID-19 [4][5][6][7]. The disease can rapidly develop into severe pneumonia, respiratory failure, acute respiratory distress syndrome (ARDS), sepsis, multiple organ failure, and even death. Early identification of the severe ill patients is a top priority mission for clinicians. Although multiple studies have reported the epidemiology, clinical features and the outcomes of COVID-19 patients with severe illness, the results differed from region to region [7][8][9][10][11].
The first case of SARS-CoV-2 infection in Jiangsu province was found on January 22, 2020. By February 19, 631 laboratory-confirmed local cases have been reported in Jiangsu accumulatively. From February 19 to March 30, no more local COVID-19 patient was reported in Jiangsu. All of these 631 COVID-19 patients have been cured and discharged on March 14. At present, no specific information for clinicians to identify severely ill patients and risk factors for these patients in Jiangsu has been reported. The objective of this study was to characterize the clinical characteristics of severely afflicted patients in Jiangsu province and to reveal the potential high-risk factors associated with serious illness.

Study design and participants
Patients included in the study were recruited in Jiangsu (30°45'-35°20' N latitude, 116°18'-121°57' E longitude) with an area of 107,200 km 2 . Jiangsu province consists of 13 municipalities and 96 counties (districts), and there are 80.70 million inhabitants at the end of 2019. Currently, there are 28 hospitals authorized to accept and treat patients with COVID-19 and 541 medical institutions with fever clinics across Jiangsu. This multiple-centered, retrospective study was done at these 28 hospitals (Jiangsu, China). The ethics committee of Zhongda Hospital Affiliated to Southeast University approved this study (No. 2020ZDSYLL013-P01 and 2020ZDSYLL019-P01). A waiver of written informed consent was granted by the ethics commission.
Patients aged 18 years or older were admitted if their diagnostic specimen was positive on reversetranscriptase-polymerase-chain-reaction (RT-PCR) assay for SARS-CoV-2. The clinical spectrums are categorized into mild to critically ill cases according to COVID-19 guidelines (the seventh version) made by National Health Commission of the People's Republic of China [12]. The mild ill patients had mild symptoms and normal radiological images in both lungs. The moderate ill patients had fever, cough and other typical respiratory symptoms and radiological lung images suggesting pneumonia. The severe ill patients had any one of the following conditions: respiratory rates ³ 30 per minute, pulse oxygen saturation ≤ 93% on ambient air, and partial pressure of oxygen (PaO 2 )/fraction of inspired oxygen (FiO 2 ) ≤ 300mmHg. The critically ill patients had any one of the following conditions: respiratory failure in need of invasive ventilation, signs of shock, and failure of any other organ when ICU care is necessary. The patients were divided into two groups, study group that includes severe-critically ill patients, and the control group that includes mild-moderate patients.

Data collection
The epidemiological information, medical history, exposure history, clinical characteristics, laboratory results, comorbidities, radiological features, therapies and outcomes of the COVID-19 patients were obtained from the electronic medical records (EMRs) between January 22 and March 11, 2020. Any data missing or ambiguousness will be asked from the involved health-care providers who subsequently will collect or communicate with patients' families.

Definition
ARDS was diagnosed according to the Berlin definition [13]. Sepsis and septic shock were defined according to the third international consensus definition for sepsis and septic shock (Sepsis-3) criteria [14]. Acute cardiac injury was diagnosed when serum levels of high-sensitive cardiac troponin I were above the 99 th percentile upper reference limit [4]. Acute kidney injury was identified on the basis of the highest serum creatinine level and urine output [15]. Secondary infection was diagnosed when patients showed clinical symptoms or signs of hospital-acquired pneumonia or bacteraemia combined with a positive culture of a new pathogen from lower respiratory tract specimens (sputum, endotracheal aspirate, or bronchoalveolar lavage fluid) or blood samples obtained at least 48 hours after admission [4]. The date of disease onset was defined as the day when the symptom was first noticed. The changes in disease status from onset to hospital admission, severe disease, critically ill disease and discharge were recorded.

Statistical analysis
Categorical variables were sorted according to frequencies and percentages, and then analyzed with Pearson's Chi-square test or Fisher exact test as appropriate. Continuous variables were described as median and 25%, 75% quartiles [median (IQR)], and analyzed using t-tests or Mann-Whitney test as appropriate. Univariable and multivariable logistic regression models were used to assess the risk factors associated with severe-critically ill cases. To assess the discriminatory ability of the model, receiver-operating characteristic (ROC) curves were prepared, and the areas under the ROC curves (AUCs) were determined. Statistical analyses were performed using the SPSS version 25.0 software and GraphPad Prism version 8.0 software. P0.05 was considered as statistically significant.
Of those patients, 84 patients were divided into the study cohort and 499 into the control group, respectively. The median age of the severe-critically ill patients was 57.0 (IQR 49.0-65.7) years ( Table 1), whereas that of the mild-moderate ill patients was median 47.0 (IQR 33.0-56.0) years. The median age of the former was significantly older than the later (P0.001). Patients aged 50 years or older accounted for a higher proportion of the severe-critically ill patients (70.2% vs. 40.3%). Among the severe-critically ill patients, (58.3%) of them were males. No gender biased difference of COVID-19 was found between two groups. The smoking history was also similar in both groups.
In the severe-critically ill patients, the most frequently observed symptoms were fever (69.0%) and cough (57.1%) ( Table 1). Other common symptoms included fatigue (38.1%), shortness of breath (31%) and sputum (23.8%). The severe-critically ill patients had a significantly higher percentage of shortness of breath than the control group (31% vs.7.4%, P0.001). The severe-critically ill patients tended to have coexisting diseases. 57.1% (48/84) of severe-critically ill patients had one or more coexisting diseases ( Table 1). The most common coexisting health issues for the severe-critically ill patients were hypertension (32.1%) and diabetes (29.8%). Compared with the mild-moderate patients, the severe-critically ill patients were more likely to suffer from coexisting diseases, including hypertension, diabetes, COPD, coronary heart disease, cerebrovascular disease and cancer.

Radiological and laboratory examinations
All of the severe-critically ill patients had radiologic abnormalities on chest imaging, which were significantly more prominent than the mild-moderate patients (P0.001) ( Table 2). 79.8% (67/84) of the severe-critically ill patients showed bilateral pneumonia, while only 58.3% (293/499) of the mild-moderate patients showed bilateral involvement (P0.001). Figure 3 shows CT findings of severe type confirmed COVID-19 pneumonia.
As shown in Table 2, there were numerous differences in laboratory findings between the mild-moderate and the severe-critically ill patients. 86.9% (73/84) of the severe-critically ill patients had lymphopenia (lymphocyte counts ≤ 1.5´10 9 /L) on admission. Median lymphocyte counts of the severe-critically ill patients were significantly lower than those of the mild-moderate patients (P = 0.022). Hemoglobin levels, platelet counts and albumin values of severe-critically ill patients at admission were all lower than the mild moderate ill patients. The levels of ALT, AST, LDH, CRP, ESR, D-dimer, PT and fibrinogen were all significantly higher in the severecritically ill patients than the mild-moderate patients.
The median onset-admission interval was 5.8 (IQR 2.2-9.4) days for the severe-critically ill patients and 6.0 (IQR 2.3-10.7) days for the mild-moderate patients. There was no significant difference in the duration from symptom onset to hospital admission between these two groups of patients. The mild-moderate patients had a shorter hospitalization time than the severe-critically ill patients [median (IQR), 22.0 (12.0-32.75) days vs. 16.0 (9.75-25.0) days, P = 0.018]. The median time for COVID-19 to become severe disease was 7.0 (IQR, 4.0-9.5) days, and 10.0 (IQR, 7.5-12.0) days to critically ill disease.

Risk factors
Results of univariable analysis showed that the probability for the patients with shortness of breath, hypertension, diabetes, COPD, coronary heart disease, cerebrovascular disease, and cancer to develop into severe-critically illness increased. Age, bilateral patchy shadowing, lymphocytes, hemoglobin, platelet, ALT, AST, LDH, albumin, CRP, ESR, D-dimer, PT and fibrinogen which were also associated with the progression into severe-critically illness ( Table 4). Results of multivariable logistic regression analysis showed that age (OR 1.08, 95%CI 1.03-1.14), D-dimer (OR 3.21, 95%CI 1.39-7.40) and lymphocytes (OR 0.28, 95%CI 0.04-0.88) were independent risk factors of developing into severe-critically illness ( Table 4).
The ROC curves are shown in Figure 4. The AUC of age for predicting severe-critically illness was 0.68 (95% CI 0.62-0.74), while that of D-dimer was 0.79 (95% CI 0.73-0.85) and lymphocytes was 0.74 (95% CI 0.68-0.79). There was substantially superior performance for the combination of these three factors to predict the severe critically illness, and the AUC was 0.87 (95% CI 0.83-0.92). Therefore, the model performed well in predicting the development into severe-critically illness.

Discussion
In this study, we analyzed the clinical characteristics and risk factors for the adult COVID-19 patients to develop into severe-critically illness using the clinical records of 583 laboratory-confirmed patients hospitalized in 28 authorized hospitals in Jiangsu province, China. The study of COVID-19 patients outside Wuhan is of paramount significance for an in-depth understanding of the clinical characteristics of COVID-19. The purpose of the study aims to further define the risk factors for the progression into severe-critically illness. Our results showed that older age, higher D-dimer levels and less lymphocyte counts are closely related to the development into serious illness.
Severe-critically illness occurred in 14.4% of the COVID-19 patients in Jiangsu Province, but no corresponding death has been reported so far. The percentage of severe-critically ill COVID-19 patients in Jiangsu was lower than that in Wuhan. The regional specific difference was caused by many factors. In the early stage of COVID-19 outbreak in Wuhan, the diagnostic capacity is limited. Some patients were not transferred to hospital until their conditions became deteriorated, and their medical treatments were delayed. At the start of the epidemic, there was a great shortage of medical staff and resources at hospitals in Wuhan, and medical workers were overloaded every day. Due to the shortage of sick beds in the hospitals, a majority of COVID-19 patients cannot receive medical treatment in hospitals and be effectively quarantined. In contrast, Jiangsu provincial government applied active surveillance in the early warning of the novel coronavirus. In particular, people with recent travel history to SARS-CoV-2 affected regions are under observation, and those in close contact with COVID-19 patients are traced. Thus, clinical cases could be found efficiency, which was reflected by a shorter onset-admission interval than Wuhan [16].
None of the severe-critically ill patient had an exposure history to Huanan seafood marker, and nearly 85% of the patients did not travel or live in Wuhan. The majority of the diagnosed cases were the two-generation cases, maybe third-generation cases, and even fourth-generation cases. Consistent with previous literatures [7,8,17], the most frequently observed symptoms on admission were fever and cough. Short of breath was quite common in severe-critically ill patients. There were more evident lesions on chest radiographs in patients with severecritically illness, suggesting a potential correlation between the extent of lung injury and the severity of illness. More than half of the severe-critically ill patients had one or more underlying diseases. Patients with pre-existing diseases, such as hypertension, diabetes, COPD, cardio-cerebrovascular diseases and cancer, are vulnerable to the development of severe illness. A number of laboratory abnormalities were observed on admission in the severe-critically ill patients, including lower levels of lymphocytes, hemoglobin, platelet, albumin, and elevated content of ALT, AST, LDH, CRP, ESR, D-dimer, PT, fibrinogen. These results suggest that the severe-critically ill patients may be associated with more serious immune deficiency, coagulatory dysfunction, nutritional insufficiency, hepatic injury and inflammatory reaction, indicate a multisystem involvement.
Patients with COVID-19 is likely to develop severe illness within two weeks after disease onset. Previously, older age was reported to indicate poor clinical outcomes of COVID-19 [5,6,10,11]. In this study, we also found that the seniors were associated with the development of severe illness.
Declined immunocompetence is relatively common in the older patients who are more prone to severe infection. The incidence of severe-critically illness will be increased by 3.21 times, with increment of one standard deviation in D-dimer. It was reported early that D-dimer can be a significant prognostic factor in patients with infection or sepsis [18]. Activation of coagulation system is an early and common event in patients with infection [19], which could be reflected by D-dimer values. Therefore, D-dimer levels remains important as it can be potential therapeutic targets to resolve the coagulation disorder aiming at reducing the incidence of severe cases. Lymphocyte count was found to be an independent protective factor for severe-critically illness. For every one standard deviation increase in the lymphocyte level, the risk for developing into severe-critically illness will be decreased by about 72%. Earlier studies implied that depressed lymphocyte is a prominent feature of critically ill patients with SARS-CoV and MERS infection [20, 21]. SARS-CoV-2 was reported to use the same cellular entry receptor as SARS-CoV [22]. Thus, we hypothesized that coronavirus particles may invade lymphocytes, damage the cytoplasmic component and cause their destruction. Patients with lymphocytopenia are thought to have a lower rigorous immune response against SARS-CoV-2 and an enhanced susceptibility to severe infection.
All of the severe-critically ill COVID-19 patients in this study were given antiviral agents without any solid evidence. Currently, there are several ongoing clinical trials to investigate the efficacy and safety of those drugs. Nearly half of the patients were treated with intravenous corticosteroids. The use of corticosteroids at low-tomoderate dose in patients with coronavirus infection was supported [11,23]. The treatment with methylprednisolone tended to reduce the death risk in COVID-19 patients with ARDS [11]. Until now, no specific therapy has been recommended for severe and critically ill patients except for the meticulous supportive treatments. Notably, double lung transplants for two COVID-19 patients were successfully performed in Wuxi, East China's Jiangsu Province. The transplanted lungs functioned well in oxygenation, and two patients got improved and discharged.
Our study, however, has some limitations. First, this is a retrospective study. The uncertainty of recall bias and variation of EMRs in different hospitals might have unavoidably affected our evaluation. Second, not all of the laboratory indicators were tested in every patient, such as CD3, CD4, CD8 T cells, IL-6 and IL-8. Especially, previous studies have revealed that CD3, CD4 and CD8 T cells played vital roles in coronavirus pneumonia [11,24,25]. The roles of these laboratory indicators were underestimated in predicting the progression of severecritically ill patients. Third, the data on radiographical examination were not well described in detail. The imagines of some patients are merely left, right or bilateral pneumonia. Some specific information, such as ground-glass opacities, patchy shadow and interstitial abnormalities, was missing.

Conclusions
This study conducted a comprehensive analysis of the clinical characteristics and risk factors of the severecritically ill COVID-19 patients in Jiangsu province, China. The factors closely related to the progression of severecritically illness are older age, elevated D-dimer values and declined lymphocyte cell contents. Our results may facilitate to establish targeted interventions and to reduce the mortality of pandemic.

Ethics approval and consent to participate:
The ethics committee of Zhongda Hospital Affiliated to Southeast University approved this study (No. 2020ZDSYLL013-P01 and 2020ZDSYLL019-P01). A waiver of written informed consent was granted by the ethics commission.

Availability of data and material:
All data generated or analyzed during this study are included in this published article

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

Funding:
National Natural Science Foundation of China (81470206).

Authors' contributions:
Jiangnan Zhao and Yi Shi take responsibility for the study design, accuracy of the data analysis and drafting the manuscript. Meiying Zhu, Xin Su, Mao Huang, Yi Yang, Jianan Huang, Songshi Ni, Quan Cao, Qin Gu, Jun Li, Jiashu Li, Wenjing Zhao and Bin Shi were responsible for the revision of the manuscript. All authors read and approved the final manuscript.   COPD, chronic obstructive pulmonary disease  Data are presented as median (25%, 75% quartiles) or number (%).

Figure 1
Flow chart for patients' enrollment in the study. Note: The designations employed and the presentation of the material on this map do not imply the expression of any opinion whatsoever on the part of Research Square concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. This map has been provided by the authors.  CT findings of severe type confirmed COVID-19 pneumonia. A 66-year-old man with close contact history presenting with fever, cough and dyspnea. Chest CT showed diffusely subpleural distributed ground-glass opacities with consolidation of bilateral lungs.