Clinical Characteristics and Outcomes of Patients with Coronavirus Disease 2019-associated ARDS and the Early Intubation: A Two-Hospital Retrospective Cohort Study in Hubei, China

Background: More evidence in understanding the heterogeneity of COVID-19-associated acute respiratory distress syndrome (ARDS) and in improving strategy to increase the survival from the critical patients intubated is always needed. The study aimed to comprehensively explore the features of COVID-19-associated ARDS and the features and outcomes between the early and late intubation groups. Methods: This retrospective cohort included 65 adult COVID-19 inpatients with ARDS at two hospitals in Hubei, China. The ARDS in these patients was diagnosed according to the Berlin criteria. We dened intubation within 7 days of ARDS diagnosis as ‘early’ intubation and that performed from the eighth day as ‘late’ intubation based on literatures. The outcomes were invasive mechanical ventilation and in-hospital death. The log-binomial regression models were used to explore the risk factors and the Kaplan-Meier statistic was used to estimate the risk of mortality. Results: The median number of days from symptom onset to ARDS diagnosis was 11.0 (IQR, 8.0–13.0). Up to 84.1% COVID-19-related ARDS patients demonstrated multiple organ injuries. The mortality rates were 41.9% and 85.7% in moderate and severe ARDS. The early intubation and the late intubation had the differences in days from symptom onset/hospital admission/ARDS diagnosis to intubation (P = 0.023, P = 0.011, P < 0.001). Compared with the early-intubation group, the late-intubation group showed less severity at admission (median oxygenation index 159.0 95% CI 134.0-203.0 vs. 133.9 95% CI 98.3-183.2), but required more aggressive therapies (ICU 80% vs. 70%, CRRT 50% vs. 10%, prone-position 50% vs. 30%, and ECMO 50% vs. 10%) and had higher risk to die at hospital (RR, 3.18; 95% CI 1.98-5.12). Conclusion: The ARDS caused by COVID-19 was not typical ARDS due to prolonged onset time, multiple


Introduction
As of 27th February 2021, the coronavirus disease 2019 (COVID-19) pandemic has affected approximately 114 million people and caused over 2 million deaths worldwide [1]. The accumulated data suggest that 5-42% of COVID-19 patients develop severe illness that is characterized by acute respiratory distress syndrome (ARDS) [2,3]. Our understanding about ARDS caused by COVID-19 remains remarkably different [4][5][6]. Some clinicians and researchers thought that the respiratory system mechanics of patients with ARDS, with or without COVID-19, were broadly similar [4], but others said that COVID-19-associated ARDS has distinctive features, including an primary vascular insult, that set it apart from previous ARDS and need to adjust the management strategies [5][6][7]. More evidence is always needed to clarify the features of ARDS to improve the treatment for the ongoing COVID-19 pandemic.
The mortality was even greater than 70% amongst patients aged above 60 years [13]. There are clinicians and researchers who think the timing of intubation is associated with the high mortality for COVID-19 patients [14,15]. The invasive mechanical ventilation thus should be used as proactive but not salvage therapy, to reduce the deaths [7,16]. There is a report that proposed the existence of two primary phenotypes of type L (low lung compliance) and type H (high lung compliance) of COVID-19-associated ARDS [7,17]. Authors therefore advocated to perform proactive or early intubation in patients before excessive inspiratory efforts without worrying ventilator-induced lung injury as the high compliance results in tolerable strain [17]. Direct evidence on the consequences of delayed intubation in COVID-19 patients is not yet available. Early intubated patients seem to have better outcomes through personal experience. Analyses on retrospective data on this matter are urgently needed. In the studies published, there were a number of patients receiving intubation at one week or later after hospital admission compared with those receiving the therapy within one week [15,18]. We therefore wonder if such comparably late intubation is associated with the high mortality.
We performed a two-centre retrospective cohort to extensively investigate the characteristics of COVID-19associated ARDS patients, to explore the features and outcomes between the early and late intubation groups according to 1-week cutting point since ARDS diagnosis, and to study the predictors for the early intubation.

Ethical approval
This manuscript adheres to the applicable CONSORT guidelines. The study was approved by the Chinese

Inclusion and exclusion criteria
We identi ed patients with COVID-19-related ARDS admitted to the two hospitals from 26 January 2020 to 1 April 2020. The ARDS was diagnosed according to the Berlin criteria: development of acute, bilateral pulmonary in ltrates (determined by chest imaging) and hypoxemia (PaO 2 /FIO 2 ≤ 300 mmHg or SpO 2 /FiO 2 ≤ 315 as a surrogate if no PaO 2 data are available in the study) not primarily due to heart failure or volume overload [20]. Patients with COVID-19-related ARDS were included in the cohort independent of their positive end-expiratory pressure (PEEP) requirements. Therefore, patients receiving oxygen via a nasal cannula and facemask were included if they met the criteria of ARDS, in order to recognize more potential inpatients who later presented with hypoxemia respiratory failure warranting IMV. Patients were screened and included by reviewing their electronic medical records and laboratory results. We excluded COVID-19 patients who did not meet the ARDS diagnosis criteria, who did not have laboratory con rmation of ARDS, or had ARDS caused by other pathogens, such as bacteria, fungi, mycoplasma or chlamydia.

Data collection
We reviewed the electronic medical records, laboratory results and radiographic reports of all admitted COVID-19 patients. Using a modi ed version of the standardized case report form developed by the International Severe Acute Respiratory and Emerging Infection Consortium and WHO [21], we extracted data on the demographic characteristics, medical history, comorbidities, illness onset date, initial symptoms, vital signs on admission and other data such as the severity of pneumonia graded by physicians. We also extracted the results of routine hematological, biochemical and in ammation-related indices obtained within 72 hours of hospital admission. Further, we collected information on ARDS diagnosis, including chest imaging; FiO 2 , PaO 2 or SpO 2 values to calculate oxygenation index; oxygen supplement modalities; intubation date; IMV duration; treatment (antiviral drugs, antibiotics, glucocorticoids and vasopressor agents) or advanced therapies (intensive care unit (ICU) care, continuous renal-replacement therapy (CRRT), prone position ventilation and extracorporeal membrane oxygenation (ECMO); complications with death and other clinical outcomes. Reports of chest radiographs were obtained, and any factors associated with acute respiratory failure warranting intubation in the cohort and with death were recorded.

Outcomes
The outcomes were intubation and mechanical ventilation and in-hospital death. Based on 1-week cut-off point, we de ned intubation within 7 days of ARDS diagnosis as 'early' intubation and that performed from the eighth day as 'late' intubation. The patients were then classi ed into three categories: 1) 'early' intubation: IMV performed within 7 days of ARDS diagnosis, 2) 'late' intubation: IMV performed from the eighth day onwards since ARDS diagnosis, and 3) no intubation (non-invasive respiratory support): the use of nasal cannula and facemask to inhale oxygen with no requirement of intubation during the hospital stay.

Statistical analysis
Continuous variables are expressed as medians and interquartile ranges (IQRs). Categorical variables are summarized as counts (n) and percentages (%). Missing data were not imputed. We used Kruskal-Wallis rank test and Fisher's exact test to compare the corresponding results among non-invasive respiratory support and early and late intubation groups. Given the incidences of all outcomes investigated exceeding 15% in which odds ratio would overstate the risk ratio, perhaps dramatically, the log-binomial regression models were used to yield risk ratios of the risk factors for intubation, late intubation and inhospital death [22]. The risk ratios (RRs) were reported if appliable. The Kaplan-Meier statistic and the survival curves were used to estimate the risk of mortality among comparisons. A two-stage procedure was performed for comparing hazard ratios among three groups. Due to the explorative nature of the study, no formal hypotheses were implemented to drive the sample size calculation, and we included the maximum number of patients who met the inclusion criteria. P values less than 0.05 were considered signi cant except for 0.0125 for two-stage procedure statistics. All analyses were performed in SPSS (version 25, IBM Corp, Armonk, NY, USA) and Stata (version 16; Stata Corp, College Station, TX, USA).

Results
Characteristics and clinical outcomes of patients with COVID-19-related ARDS From 26 January to 1 April 2020, 338 patients with COVID-19 con rmed by laboratory ndings or otherwise clinical diagnosis were treated at the two hospitals in Hubei province of China. Of these, 70 patients developed ARDS. We included 65 of these cases with laboratory con rmation and excluded ve cases. The details of enrolment are provided in Fig Laboratory indices tested at admission (Table 2) suggested that most patients had systemic hyperin ammation, indicated by the increased erythrocyte sedimentation rate (ESR), C-reactive protein level and procalcitonin level. Quite a few patients demonstrated reduced albumin/globulin ratio and increased total bilirubin and aspartate transaminase levels, suggesting liver injury, and elevated lactate dehydrogenase and cardiac troponin I levels, indicating heart injury. Approximately four-fths of the patients had lymphocytopenia. Some patients were suspected of having pancreatic injury indicated by the high blood glucose level, and kidney injury based on the elevated serum creatinine level and reduced glomerular ltration rate (GFR). Next, a high international normalized ratio, prolonged prothrombin time and elevated D-dimer level were reported in several patients. Some patients demonstrated leukocytosis, neutrophilia, thrombocytopenia and reduced hemoglobin concentrations.
Comparisons among patients with non-invasive respiratory supports, early, and late intubation Compared with the late-intubation group, patients with early intubation had signi cantly shorter time from symptom onset to intubation (P = 0.023), from hospital admission to intubation (P = 0.011), and from ARDS diagnosis to intubation (P < 0.001) ( Fig. 2A). We observed the differences on OI values (P = 0.015) at ARDS diagnosis and proportions of ARDS classi cations (P = 0.052) among no-intubation, early-intubation, and late-intubation groups. Patients with early intubation presented the lowest OI ratios (median, 133.9; IQR 98.3-183.2) and the highest proportion of moderate and severe ARDS (90%). But patients with late intubation had a slightly higher OI ratios (median, 159.0; IQR 134.0-203.0) compared to early-intubation group. Consequently, these late intubation patients received the most aggressive therapies including glucocorticoids (100.0%), vasopressors (50.0%), CRRT (50.0%), prone position ventilation (50.0%), ECMO (50.0%), and ICU-level care (80.0%) in the context of the associated differences occurring among the three groups (P = 0.016, P = 0.007, P < 0.001, P = 0.001, P < 0.001, P < 0.001). The mortality rate at discharge in the late-intubation group was 90.0% compared with 50.0% in the earlyintubation group and 23.3% in the no-intubation group (P < 0.001, Table 3, Fig. 2B). The hospital survival probability of the late-intubation group was the lowest among the three groups (Fig. 2C).

Discussion
In this study, we found that the ARDS caused by COVID-19 was not the typical ARDS de ned by the Berlin criteria [20] due to longer onset time, multiple organs injuries and higher mortality in moderate and severe form. The late intubation group, but not the early intubation group, with less severity at admission, needed more aggressive therapies and cares, and was highly associated with in-hospital death. The predictors for intubation were not same as those for death.
The study complements the existing studies on COVID-19-related ARDS and enriches the understanding of its clinical features. The median onset time for COVID-19-related ARDS was 11.0 days (IQR, 8.0-13.0), which is inconsistent with the 1-week onset limit reported for ARDS caused by other factors [20]. The ndings should alert physicians to pay more attention to the development of ARDS in COVID-19 patients with a disease course of more than a week for timely treatment. The mortality rates of moderate and severe COIVID-19-related ARDS were 41.9% and 85.7% in our study, which are much higher than 32% and 45% as reported in typical ARDS [23]. One possible explanation is silent hypoxemia as proposed by several experts, that clinical symptoms are inconsistent with the severities of hypoxemia which would result in delay treatment [6,24]. From the perspective of therapy, a more suitable classi cation of ARDS severity that can accurately identify patients requiring a timing therapy for COVID-19 has been proposed [5].
The majority of patients with ARDS in this study were men over 65 years old with hypertension and/or diabetes but not overweight. A large proportion of the patients had palpitations and chest distress at admission and later demonstrated injuries in the heart, liver, kidney, and pancreas during hospitalization. This nding is consistent with the previous ndings that the new coronavirus can attack the tissues of other organs in addition to the respiratory tract [25][26][27][28]. Fundamental experiments revealed the mechanism that SARS-CoV-2 can enter the circulation by infecting the endothelium of lung tissue; the virus then directly attacks the other organs or triggers a systematic hyperin ammation response, resulting in secondary attacks to the organs [29].
We noticed that apart from antiviral therapy, many other antibiotics were administered to combat bacterial co-infections, and some patients received glucocorticoids to alleviate hyperin ammation indicated by the increasing levels of ESR, C-reactive protein and procalcitonin. We observed that patients with COVID-19-related ARDS presented with severe lymphocytopenia on admission, required aggressive therapies such as CRRT and ECMO, needed long hospital stays (median, 21.0 (IRQ, 12.0-34.0) days), but died more in hospital. All in all, updating the management strategy is highly suggested, including monitor longer to the development of COVID-19-related ARDS and re-categorize the atypical ARDS to give timing therapies.
Another important nding was that the groups of early and late intubations and the early intubation was not associated with in-hospital death. By contrast, the late-intubation group required more advanced therapies, presented a higher RR for death and trended the lowest hospital survival probabilities among the three groups. The conclusion is consistent with one study that each additional day between hospital admission and intubation was signi cantly associated with higher in-hospital death [15]. Because most of patients died of acute respiratory failure, the explanation coming forth that cannot be excluded was the delayed intubation. In our cohort, we noticed patients intubated late were not sicker at admission than those intubated early and their median OI ratios at ARDS diagnosis were slightly above the threshold of 150, where the intubation census usually suggests perform invasive mechanical ventilation if OI below 150 [16]. Given that the COVID-19-caused ARDS could present silent hypoxemia [6], where the degrees of dyspnea are inconsistent with the severities of hypoxemia, that would result in delayed intubation once sudden decompensation occurs. Many years ago, Shoemaker et al. [30] reported there is a close association between the oxygen debt accumulated over 48 hour and the chance of survival in critical patients. We noticed that even early intubation group had 50% in-hospital mortality in the cohort, as similar as the overall global data calculated recently by the systematic review [13]. Our current ventilatory approach or intubation criteria is to delay intubation if it clinically appears safe and feasible [18]. However, the ndings of the cohort may suggest proactive intubation instead of salvage means for critical illness of COVID-19. This is supported by some accurate observations that many of COVID-19 patients deteriorated precipitously and that they may be more safely intubated at an earlier stage [6,7]. Additionally, a high work of breathing generating large swings in intrapleural pressure may result in selfin icted lung injury and worsen the disease [7], as certainly a rationale for early intubation. However, the exact de nition of the early intubation for COVID-19 critical illness is challenging. To intubate the patients early before sudden decompensation occurs while avoiding the risk of iatrogenic injury should be considered simultaneously in the face of so much high mortality.
The predictors for intubation and death were not same. The cohort found that the COVID-19 ARDS patients who had hypertension or GFR < 80 ml/min/1.73m 2 were more likely to have the intubation but not death. This may suggest different development mechanism between death and intubation for COVID-19 ARDS patients.
We found lower OI ratio and OI less than 150 were associated with intubation and death. OI ratio less than 200 or 100 was associated with just death. Lower OI ratios mean acute respiratory failure and OI less than 150 already indicates extreme respiratory failure [31]. The OI predictors for intubation or death re ecting the severity of acute lung injury as well were not reported in any study [31][32][33][34]. Our ndings on OI ratio predictors shall support the strategy that "patients with a OI ratio less than 300 mmHg should be prepared for intubation if there are indications, such as poor oxygenation after 2 hours of high-ow oxygen therapy or non-invasive ventilation, and OI less than 150" [16]. That requires early recognition and timing treatment of patients with worsening respiratory function during COVID-19 treatment. Additionally, we also found older ages could predict the death in COVID-19-related ARDS patients. The older age was therefore veri ed as the risk factor for death of critical patients as it for general COVID-19 patients [33].
The limited hospital resources and treatment strategy therefore shall focus on the elder age to reduce the death numbers from critical illness of COVID-19.
The strengths of the study are as follows: the study cohort included patients from two hospitals in two cities of Hubei province, which increases the generalisability of the ndings; phenotyping of clinical characteristics was performed based on respiratory support strategies and IMV-associated groups, de ned as early and late intubation according to 1-week cut-off point; reported adjustment RRs instead of crude RRs. However, the study has some limitations. First, it is a retrospective cohort conducted at two hospitals. As is known, it was hard to perform a perspective cohort when epidemic happened and developed everywhere. As we have sparse new COVID-19 cases now, we hope our new ndings can encourage a larger cohort study or even randomly controlled trials in other regions who are experiencing the outbreak and in the urgent need to adjust the strategy. Second, the information such as respiratory function variables during the treatment course at the early and late intubations should be collected and analysed. The current cochort's ndings propose the need to further explore the better timing of intubation for reducing mortality.

Conclusions
We found that the SARS-CoV-2-caused ARDS was not typical ARDS for which the management strategies should be adjusted accordingly. We studied the early-and late-intubation groups, based on 1-week cutting point, which had signi cant differences in the days from symptoms onset/hospital admission/ARDS diagnosis to intubation. Compared with the early intubation, the late intubation presenting less severity at admission but required more advanced treatment and had the higher risk of death. Predictors of the intubation were not same as those of in-hospital death. The ndings had implications in modulating the management strategy for ARDS of COVID-19, and in supporting the early intubation strategy. This manuscript adheres to the applicable CONSORT guidelines. The study was approved by the Chinese Ethics Committee of Registering Clinical Trials (ChiECRCT20200113), and the requirement for informed consent was waived because of the retrospective study design and given the context of ongoing infectious diseases.    Figure 1 Enrolment of COVID-19-related ARDS patients at Wuhan and Honghu hospitals