Clinical Findings of COVID-19 Patients Admitted to Intensive Care Units in Guangdong Province, China: A Multicenter, Retrospective, Observational Study

Background: Information about critically ill patients with coronavirus disease 2019 (COVID-19) in China but outside of Wuhan is scarce. We aimed to describe the clinical features, treatment, and outcomes of patients with COVID-19 admitted to the intensive care unit (ICU) in Guangdong Province. Methods: In this multicenter, retrospective, observational study, we enrolled consecutive patients with COVID-19 who were admitted to seven ICUs in Guangdong Province. Demographic data, symptoms, laboratory findings, comorbidities, treatment, and outcomes were collected. Data were compared between patients with and without intubation. Results: A total of 45 COVID-19 patients required ICU admission in the study hospitals [mean age 56.7 ± 15.4 years, 29 males (64.4%)]. The most common symptoms at onset were fever and cough. Most patients presented with lymphopenia and elevated lactate dehydrogenase. Treatment with antiviral drugs was initiated in all patients. Thirty-six patients (80%) developed acute respiratory distress syndrome at ICU admission, and 15 (33.3%) septic shock. Twenty patients (44.4%) were intubated, and 10 (22.2%) received extracorporeal membrane oxygenation. The 60-day mortality was 4.4% (2 of 45). Conclusion: COVID-19 patients admitted to ICU were characterized by fever, lymphopenia, acute respiratory failure, and multiple organ dysfunction. The mortality of ICU patients in Guangdong Province was relatively low with a small sample size.


INTRODUCTION
In December 2019, human infection with a novel coronavirus, known as severe acute respiratory syndrome coronavirus disease 2 (SARS-CoV-2), was confirmed in Wuhan, China, and spread rapidly beyond Wuhan and around the world (1,2). By May 1, 2020, a total of 84,388 patients were infected in mainland China, with 4,643 deaths, according to a Chinese Center for Disease Control and Prevention report (3). Previous studies have mainly focused on the general epidemiological findings, clinical presentation, and clinical outcomes of mild and moderately affected patients with coronavirus disease 2019 (COVID-19) (4-7). One recent study reported the characteristics of critically ill patients in a single center in Wuhan with 61.5% mortality (8). However, the clinical characteristics of patients with COVID-19 admitted to intensive care units (ICUs) in China but outside of Wuhan have not been described, including in Guangdong Province, where by May 1, 2020, more than 1,000 people had been confirmed as having COVID-19 (3). Here, we describe the characteristics and treatment of ICU patients with COVID-19 in Guangdong Province.

Study Design
This multicenter, retrospective, observational study was  Table 1). The study was approved by the Ethics Commission of the First Affiliated Hospital of Guangzhou Medical University. The informed consent requirement was waived because the study was retrospective.

Patients
Consecutive patients were enrolled from January 14, 2020, to February 20, 2020, and all had confirmed SARS-CoV-2 infection by real-time polymerase-chain-reaction testing of throat swab specimens. The illness severity of COVID-19 was defined according to the Chinese management guideline for COVID-19 (version 7.0) (9). Severe cases were defined as meeting any of the following criteria: (1) respiratory distress (≥30 breaths/ min); (2) oxygen saturation ≤93% at rest; (3) arterial partial pressure of oxygen (PaO 2 )/ fraction of inspired oxygen (FIO 2 ) ≤300 mm Hg. Critical cases were defined as meeting any of the following criteria: (1) respiratory failure and requiring mechanical ventilation; (2) shock; (3) with other organ failure that required ICU care. In our study, the criteria for ICU admission included both the severe cases and critical cases. For the enrolled patients, their living status, intubation, weaning, and ICU and hospital discharge dates were confirmed on April 20, 2020.

Data Collection
Medical records of patients who were in the ICU from January 14, 2020, to April 20, 2020, were extracted and sent to the data collection center in the First Affiliated Hospital of Guangzhou Medical University. A team of ICU doctors who had been treating patients with COVID-19 collected and reviewed the data. If information was not clear, the central working group contacted the doctor responsible for the treatment of the patient for clarification. Information recorded included demographic data, underlying conditions, symptoms, laboratory and chest radiograph findings, comorbidities, intubation rates, and ventilator settings prior to and during ICU therapy.

Definition
Fever was defined as axillary temperature of at least 37.3 • C. The incidence of COVID-19-related comorbidities was identified, including acute respiratory distress syndrome (ARDS), septic shock, cardiac injury, acute kidney injury, liver dysfunction, and gastrointestinal hemorrhage. ARDS was diagnosed according to the Berlin definition (10), and septic shock was identified by the Sepsis-3 definition (11). Acute kidney injury was identified on the basis of elevated serum creatinine according to the Kidney Disease: Improving Global Outcomes guideline (12). Cardiac injury was recognized by increased cardiac troponin I or electrocardiography abnormalities of non-specific ST-T waves (13,14). Transaminitis was defined by aspartate aminotransferase or alanine aminotransferase levels >80 U/L. Gastrointestinal hemorrhage was identified by a positive fecal or gastric fluid occult blood test (15).

Statistics
Continuous variables are presented as either mean ± standard deviation (SD) or median with interquartile range (IQR), in accordance with either normal or non-normal distributions. For categorical variables, the frequency and percentage of patients in each category were calculated. Differences between intubated and non-intubated patients were assessed with two-sample t-test or Wilcoxon rank-sum test, depending on parametric or non-parametric data for continuous variables and χ 2 test for categorical variables. The Spearman correlation coefficient was used to test the correlations between clinical variables. All analyses were performed using SPSS 23.0 (SPSS Inc., Chicago, IL, USA). A p < 0.05 (two-sided) was considered statistically significant.

Demographic, Epidemiologic, and Baseline Characteristics of the Patients
Of 332 patients with COVID-19, 45 patients requiring ICU admission were identified in the study hospitals. The first case included was confirmed on January 14, 2020, and the last one was confirmed on February 20, 2020. All patients had positive throat swabs for SARS-CoV-2 and bilateral infiltrates on chest radiographs (Figure 1) and were admitted to the ICU.
To identify the potential risks for intubation in patients with COVID-19, a χ 2 test showed that older age (age ≥60 years), higher SOFA (≥4) and APACHE II (≥15) scores, high LDH (≥255 U/L), and lower lymphocyte count (≤0.8 × 10 9 /L) at ICU admission were associated with a higher risk for intubation (Supplementary Table 3 Our study population had many of the clinical characteristics of patients with COVID-19. Lymphocytopenia occurred in more than 90% of COVID-19 patients admitted to ICU, which was similar to the Wuhan cohort (8). We found that intubated patients had significantly lower lymphocyte counts than non-intubated patients. We found that patients with lower lymphocyte counts were at higher risk for intubation and that lymphocyte counts were significantly negatively correlated with SOFA score at ICU admission. Lymphocytopenia is common in viral pneumonia, particularly in SARS and Middle East respiratory syndrome (MERS) (17,18). It was reported that a lymphocyte   count <0.8 × 10 9 /L was an independent risk factor for 90-day mortality in viral pneumonia (19). Therefore, lymphocytopenia may reflect the severity of COVID-19. Dynamic monitoring of the lymphocyte counts might be useful in terms of prognosis during the intensive care phase of the critical illness. Although our patients were treated with antiviral drugs, including oseltamivir and lopinavir-ritonavir, to date no effective antiviral to treat COVID-19 has been identified. The current approach to clinical management is general supportive, supplemented with critical care and organ support when necessary (20,21). It has been suggested that the administration of high-titer anti-influenza immune plasma derived from convalescent or immunized individuals may be clinically beneficial for the treatment of SARS, MERS, and seasonal influenza (22)(23)(24). Convalescent plasma is a potential treatment in coronavirus infection and has been suggested for use in COVID-19 (25). Therefore, convalescent plasma of patients with COVID-19 was administered to six patients (13.3%) in our cohort, and no transfusion reactions occurred. However, our findings cannot provide evidence of the efficacy of convalescent plasma in critically ill patients with COVID-19 because of limited sample sizes, short observation times, and lack of a randomized controlled group. Thus, treatment for the current outbreak of COVID-19 via convalescent plasma, particularly in patients with critical illness, should be carefully considered before well-designed clinical trials are conducted.
The 60-day mortality of ICU patients with COVID-19 was 4.4% (2 of 45) in our cohort, which was lower as compared to reported mortality of ICU patients in Wuhan and around the world (8,(26)(27)(28)(29). There are a couple of factors associated with lower mortality. First, the numbers of infected cases and mortality rates related to COVID-19 vary from country to country. There is evidence that SARS-CoV-2 has undergone genetic changes in the process of spreading to other parts of the world after affecting China (30). A SARS-CoV-2 mutation was identified that was thought to have created a more "aggressive" form of the virus (31). Moreover, SARS-CoV-2 genomic variations were also reported to be associated with a higher mortality rate of COVID-19 (32)(33)(34). Therefore, fewer genetic changes of SARS-CoV-2 in the early beginning of China may account for the relative low mortality in China and in our study. Second, an intensivists-led multidisciplinary team was established and participated in combatting the COVID-19 pandemics at the very beginning of outbreak in the seven hospitals. A well-designed Web-based video consultation system for critically ill patients with COVID-19 was established and applied across different cities in Guangdong Province. This allowed a multidisciplinary team shared their experience and help with the management of critical illness remotely across hospitals. Third, a rapid response for patients with sudden  clinical deterioration was also established to identify patients who developed into severe case and admit them to ICU for further monitor and interventions in a timely manner. The relatively lower APACHE II and SOFA scores (the median APACHE II and SOFA scores were 14 and 4.0, respectively) of all patients at ICU admission indicating a proportion of less critically ill patients were admitted to ICU to receive high-intensity monitor and prompt intensive care interventions. Fourth, the lack of ICU beds and invasive mechanical ventilators, patients management by a non-intensivists-dominated medical team may contribute to a high mortality in the epicenter of the coronavirus outbreak (35). Therefore, the high mortality of the patients (8,(26)(27)(28)(29) may reflect the crisis of critical care medicine rather than the nature of COVID-19. With great difference, the relatively smaller number of patients in Guangdong Province, adequate ICU beds, and nurse-to-patient ratio (Supplementary Appendix) may partly contributed to a low mortality. Treatments, including convalescent plasma, invasive mechanical ventilation, CRRT, and ECMO, were applied for critically ill patients. Taken together, the mortality was relatively lower in ICU patients with COVID-19 in our cohort.

Limitations
Our study has several limitations. First, the study included only 45 COVID-19 patients admitted to ICU from Guangdong Province. The conclusion about the small mortality may be inadequate because of a small sample. More clinical features related to the critical illness may have been identified if a larger sample size had been studied. Second, at the time we wrote this article, a small number of patients had not yet been discharged from ICU, so we were unable to document the exact length of ICU stay, the number of ventilation-free days, the case fatality rate, or the predictors of fatality. A longer follow-up period is needed. Third, the low number of patients prevents performing further statistical analysis such as multivariable logistic regression for independent predictors of intubation.

CONCLUSIONS
The patients with COVID-19 from Guangdong Province in our study had a lower ICU mortality (2 of 45, 4.4%) but with a small sample size (n = 45). Larger cohort was still needed in future studies. Current treatment for critically ill patients with COVID-19 consists of appropriate support in ICU and careful monitoring until effective drugs are developed.

DATA AVAILABILITY STATEMENT
All datasets generated for this study are included in the article/Supplementary Material.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Ethics Commission of the First Affiliated Hospital of Guangzhou Medical University. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.