Comparison of the Clinical Characteristics of 2019 Novel Coronavirus Disease and Avian Inuenza A (H1N1) Complicated with Acute Respiratory Distress Syndrome1

Objective: Both 2019 novel coronavirus disease (COVID-19) and avian inuenza A (H1N1) are serious acute respiratory diseases with a predisposition to acute respiratory distress syndrome (ARDS). Our aim was to compare the clinical characteristics of patients with COVID-19 and H1N1 inuenza complicated with ARDS. Methods: We retrospectively studied data of 12 patients with ARDS (7 with COVID-19; 5 with H1N1 inuenza) who were managed at The Second Aliated Hospital of Xiamen Medical College and Xinglin Branch of the First Aliated Hospital of Xiamen University between December 20, 2019 and February 29, 2020. We extracted the clinical information and outcomes from the hospital medical charts. Results: Patients with COVID-19 were older and were more likely to have underlying diseases. Low-to-moderate fever was more frequent and upper respiratory tract symptoms were less common in COVID-19 patients. Chest computed tomography of patients with COVID-19 more frequently revealed bilateral nodular patchy ground-glass opacities in the subpleural and central lobular regions. Heart disorders and pleural effusion were less frequent, and coagulopathy was more common in patients with COVID-19. The average duration of stay in the respiratory intensive care unit was longer in patients with COVID-19. The disease severity and clinical outcomes did not differ signicantly between the two groups. Conclusion: Older age, higher comorbidity frequency, abnormal coagulation responses, longer hypoxemia duration, pulmonary brosis, and poorer clinical outcomes are the main characteristics in patients with COVID-19 who have ARDS. This calls for closer dynamic monitoring and more rigorous follow-up. were more likely to have comorbidities, and had higher incidences of abnormal coagulation and pulmonary brosis, a longer duration of hypoxemia, and poorer clinical outcomes than H1N1 patients with ARDS. Closer dynamic monitoring and rigorous follow-up should be implemented in patients with COVID-19 complicated with ARDS.


Introduction
The novel coronavirus disease (COVID-19) outbreak has rapidly spread worldwide, resulting in more than 6.5 million con rmed cases and more than 383,792 deaths worldwide as of June 5, 2020 (1). The mean case fatality rate of COVID-19 in mainland China was estimated to be 6.1% (2), which was signi cantly higher than that of other severe acute respiratory diseases, such as avian in uenza A (H1N1; 1.3% in 2009), which has also been associated with global pandemics (3).
The clinical manifestations of COVID-19 are heterogeneous, ranging from mild to critical illness. Acute respiratory distress syndrome (ARDS) has been a common complication of both COVID-19 and H1N1 in uenza in critically ill cases and is a crucial driver of respiratory failure or multiple organ dysfunction syndrome the risk factors for death (4). However, it remains unclear whether the clinical characteristics and course of ARDS differ between patients with COVID-19 and those with H1N1 in uenza.
To address this issue, we sought to compare the clinical characteristics of patients with ARDS associated with COVID-19 or with H1N1 in uenza, to aid clinicians in identifying the de ning characteristics of COVID-19.

Methods
This was a retrospective observational study. The study was approved by the ethics committee of both participating centers, and written informed consent was obtained from patients before enrollment. The procedures followed in our study were in accordance with the Declaration of Helsinki.

Participants
Between December 20, 2019 and February 29, 2020, we recruited patients who were admitted to the Respiratory department and respiratory intensive care unit (RICU) of the Second A liated Hospital of Xiamen Medical College and the Xinglin Branch of the First A liated Hospital of Xiamen University (the designated COVID-19 hospital of Xiamen City, China). All patients were diagnosed as having ARDS according to the Berlin criteria (5) and tested positive for either severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) or H1N1. All patients underwent H1N1 in uenza virus and SARS-CoV-2 nucleic acid testing. SARS-CoV-2 was tested using a reverse transcription polymerase chain reaction based test on nasopharyngeal swabs. H1N1 in uenza virus was tested using second-generation gene sequencing technology after hospital admission. We excluded patients younger than 14 years of age, those whose ARDS was caused by community-acquired pneumonia of unknown etiology, and those whose ARDS was due to known causes other than SARS-CoV-2 or H1N1. A total of 12 patients, including 7 patients with ARDS associated with COVID-19 and 5 patients with ARDS associated with H1N1 in uenza virus infection, were admitted to our study sites.

Clinical data collection
We collected data on the sex, age, body mass index, underlying diseases, epidemiological history, clinical manifestations, laboratory examinations (at onset of ARDS), imaging examinations, treatment, and prognosis of the two groups of patients from the electronic medical record case system. The data were reviewed by a trained team of physicians.

Statistical analysis
The data were compared and analyzed descriptively, and the P values were not presented, because the sample size was too small for statistical analysis and subjects were not randomly collected. Mean ± SD or median (interquartile range) was presented for the continuous variables, and count (percentage) was demonstrated for the description of categorical variables.
Epidemiological and clinical characteristics upon development of ARDS As shown in Table 1, patients with COVID-19 were older than those with H1N1 in uenza (median age: 56 vs. 43 years). None of the patients with H1N1 in uenza (residents of Xiamen City) had any underlying diseases, whereas patients with COVID-19 had various underlying diseases, such as atrial brillation (n = 1), hypertension and obstructive sleep apnea syndrome (n = 1), diabetes (n = 1), diabetes and hypertension (n = 1), coronary artery disease and hypertension (n = 1), and malignancy (n = 1).
Five patients with COVID-19 were from Hubei province, and one was from Quanzhou City (Fujian province). Patients with H1N1 in uenza had neither a recent travel history nor any family-based clusters of cases. The seven patients with COVID-19 had a clustered onset within a family.
Patients with COVID-19 presented with a signi cantly longer period from the onset of symptoms to ARDS development (7 vs. 4 days) and longer duration of stay in the RICU (mean, 18 vs. 14 days) than those with H1N1 in uenza.
Clinical manifestations upon development of ARDS Fever was the most common symptom, and low-to-moderate fever was signi cantly more common in patients with COVID-19 than in those with H1N1 in uenza (85.7% vs. 20.0%). Upper respiratory tract symptoms were signi cantly less common in patients with COVID-19 than in those with H1N1 in uenza (14.3% vs. 80.0%); however, lower respiratory tract symptoms (including cough, chest tightness, and dyspnea) were exhibited by all patients in both groups. Hemoptysis was documented in two patients with H1N1 in uenza but in none of the patients with COVID-19. Gastrointestinal tract symptoms were equally rare ( Figure 1).

Laboratory investigations upon development of ARDS
In the ARDS stage, patients with COVID-19 were more likely to have abnormal coagulation than patients with H1N1 in uenza, such as increased D-dimer levels (100.0% vs. 20.0%). However, patients with H1N1 were more likely to have abnormal cardiac function, such as increased myocardial injury markers (troponin I, 80.0% vs. 14.3%) and increased heart failure indicators (brain natriuretic peptide, 80.0% vs. 14.3%) than patients with COVID-19. Leukopenia, decreased albumin levels, and elevated C-reactive protein levels were common in both groups, as were lymphopenia and elevated serum transaminase levels (which occurred in all patients). In addition, hypoxemia was recorded in all patients ( Table 2).
Blood culture results in all patients were negative. Only one case tested positive for mycoplasma antibodies (titer, 1:640). Of the ve patients with H1N1 in uenza, two cases tested positive for Aspergillusfumigatus in galactomannan assays of bronchoalveolar lavage uid.

Imaging characteristics
Computed tomography (CT) imaging showed that all patients with COVID-19 had multiple, bilateral lung lesions. In all cases, the lesions had a multi-lobar, multi-segmental distribution pattern (overlap between subpleural and lobular cores). Two cases showed ground-glass opacities only, while ve cases had ground-glass opacities combined with consolidation. In all cases, there were nodular and patchy-type lesions. We also noted pleural parallel sign (6) in ve cases, the ne-grid sign in six cases, the bronchogram sign in ve cases, the vascular thickening sign in ve cases, bronchiectasis in three cases, and the halo sign in two cases. There was a single case with anti-halo sign, one case with pleural effusion, and one case with pulmonary bulb. Representative images of patients with COVID-19 are shown in Figure 2.
On chest CT, all patients with H1N1 in uenza had multiple lung lesions, all of which were diffusely distributed along the bronchial vascular bundle. All were ground-glass opacities with consolidation. All ve cases showed bronchogram signs, three cases showed central lea et nodules, three cases had a thickened lea et septum and ne-mesh sign, and three cases presented with pleural effusion. Figure 3 shows the representative CT images of a patient with H1N1 in uenza.

Treatment and prognosis
Antiviral drugs were initiated in all patients. Patients with COVID-19 received a-interferon nebulization, lopinavir/ritonavir, arbidol, ribavirin, and chloroquine phosphate according to the 4 th -7 th edition of the Pilot Program for the Diagnosis and Treatment of Novel Coronavirus Pneumonia (7)(8)(9)(10)(11). Patients with H1N1 in uenza received systemic oseltamivir according to the In uenza Diagnosis and Treatment Plan 2019 Edition (12). Gamma immunoglobulin and anticoagulation therapy were initiated in patients with COVID-19, but only for a single patient with H1N1 in uenza.
Nasal high-ow humidi ed oxygen therapy was initiated in four patients with COVID-19 (all had mild ARDS) and three patients with H1N1 (with mild-tomoderate ARDS). Alternating use of nasal high-ow humidifying oxygen therapy and non-invasive ventilator-assisted ventilation was initiated in three patients with COVID-19 due to moderate-to-severe ARDS, but in none of the patients with H1N1 in uenza. Moreover, tracheal intubation with ventilator-assisted ventilation was initiated in two patients with H1N1 in uenza (with severe ARDS), but in none of the patients with COVID-19.
Septic shock, thromboembolism, and renal insu ciency were absent in both groups. All patients had acquired pneumonia and abnormal liver function.
Pneumothorax and mediastinal emphysema were found in a single patient with COVID-19, while gastrointestinal bleeding occurred in a case in each group. Myocardial damage and cardiac dysfunction were more common in the H1N1 group than in the COVID-19 group (80.0% vs 14.3%). Overall, patients with COVID-19 achieved viral assay conversion from clinical diagnosis in a signi cantly longer duration as compared with the patients with H1N1 in uenza.
Furthermore, the duration of hypoxemia as well as the duration of stay in the RICU was longer in patients with COVID-19 ( Table 3). None of the patients died throughout the course of our study due to timely clinical interventions.

Discussion
There have been few reports of COVID-19 and H1N1 in uenza complicated with ARDS to date. This study summarized the clinical characteristics, laboratory test results, and chest CT ndings of the two diseases that might readily predispose to ARDS.
The seven patients with COVID-19 in our study formed a cluster of family cases, whereas all patients with H1N1 in uenza were sporadic, suggesting that COVID-19 would be more contagious, with a stronger potential for human-to-human transmission (4,12,13). Although the general population is susceptible, patients with COVID-19 in our study were mostly older and had underlying diseases, which was consistent with a previous report (14). Our ndings showed that patients with COVID-19 were more likely to have low-to-moderate fever and minor upper respiratory tract symptoms than those with H1N1 in uenza.
Our ndings suggested that lymphopenia and liver dysfunction were common features of both diseases, while other laboratory tests differed between COVID-19 and H1N1 in uenza. For example, in patients with COVID-19, coagulopathy was more common (100% vs. 20%). The more serious the ARDS, the greater the increase in the D-dimer level, which was consistent with the reports by Chen et al (14). This mechanism was consistent with the pathological change of pulmonary microvascular transparent thrombosis after human infection with COVID-19 (15), which emphasized the need to closely monitor coagulation dynamically and provide anti-coagulation treatment as early as possible in COVID-19 cases. In patients with H1N1 in uenza, myocardial injury and cardiac dysfunction were more common (100% vs. 20%). The more serious the ARDS, the higher the level of troponin and brain natriuretic peptide (16), emphasizing the need for early recognition and treatment in patients with H1N1 in uenza.
When comparing the results of blood gas analysis, we found that the duration of hypoxemia in patients with COVID-19 was longer. This may be related to the signi cant coagulation abnormality in the pulmonary microcirculation of COVID-19 patients (17). Our study showed that the period from diagnosis to achieving viral assay conversion in patients with COVID-19 was signi cantly longer than in patients with H1N1 in uenza (average: 19 vs. 7 days). Taken together with a previous report (18), we speculated that the longer viral shedding time may cause more serious damage to the lung tissue and longer lasting hypoxemia, which may help explain why patients with COVID-19 had a longer stay in the RICU. This emphasizes the need for clinicians to manage COVID-19 patients with positive oxygen treatment measures in the early stage of hypoxemia (17).
In COVID-19 patients complicated with ARDS, the subpleural ground-glass opacities developed rapidly into bilateral diffuse in ltration or massive consolidation, often accompanied by the pleural parallel sign (100%) and vascular thickening sign (71.4%). Although the imaging manifestations of ARDS were similar, patients with H1N1 initially showed patchy ground-glass opacities, which were mainly distributed in the bronchovascular bundle (19,20), and were accompanied by bronchiectasis (100%), pleural effusion (60%), and increased heart shadow (60%). Therefore, the combination of the distribution of ground glass opacities and the accompanying pleural parallel sign, enlarged heart shadows, and pleural effUsion sign would be of great value for the differential diagnosis. At the 3 rd month of follow-up, the chest CT showed that the lesions in patients with COVID-19 resolved slowly, and pulmonary brotic changes were more common in these cases (57.1% vs. 13.3%). Therefore, pulmonary brosis after COVID-19 pneumonia cannot be ignored, and patients need close follow-up after discharge (21).
Overall, our patient population had a good prognosis, and no death was documented.
These ndings might be partially explained by the inclusion of patients with mild-to-moderate ARDS in our study.
The interpretation of our ndings was limited by the relatively small sample size and single-center retrospective study design. More multicenter prospective studies are needed in the future.

Conclusions
Patients with COVID-19 who had ARDS were older, were more likely to have comorbidities, and had higher incidences of abnormal coagulation and pulmonary brosis, a longer duration of hypoxemia, and poorer clinical outcomes than H1N1 patients with ARDS. Closer dynamic monitoring and rigorous follow-up should be implemented in patients with COVID-19 complicated with ARDS. The study was approved by the ethics committee of both participating centers, and written informed consent was obtained from patients before enrollment.

Consent for publication
All presentations of case reports in the manuscript all have our institutional consent to publish.

Data availability
All data of our study are presented in this article.

Funding
This research did not receive any speci c grant from funding agencies in the public, commercial, or not-for-pro t sectors.

Con icts of interest
The authors declare that they have no con ict of interest.
Authors' contributions   Rapidly progressing pulmonary lesions with major ground-glass opacities and consolidation, particularly in the middle and lower lung. C1-C3 (February 27, 2020): Signi cantly absorbed ground-glass opacities in both lungs.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.