Comparison of Hospitalized Patients with Severe Pneumonia Caused by COVID-19 and Highly Pathogenic Avian Inuenza (H7N9): A Retrospective Study from A Designated Hospital

The clinical data of 23 cases of H7N9 and 23 cases of COVID-19 with severe pneumonia were collected. The comparisons were performed with the t test, Mann-Whitney U test, Fisher exact test or the chi-squared test, and multivariable logistic regression analysis.


Background
The coronavirus disease 2019 (COVID-19), which is caused by 2019 novel coronavirus (2019-nCoV) imposes a grand immediate challenge for global public health. So far, COVID-19 has spread to many other countries with thousands of infected cases. Despite many articles have established the clinical features of COVID-19 patients [1][2][3], comparison of hospitalized patients with severe pneumonia caused by COVID-19 and highly pathogenic avian in uenza (HPAI) has not been reported yet. Because of distinct treatments and prognoses, it is important for clinicians to accurately identify these two respiroviral infections via their differential clinical manifestations [4][5].
In this study, we systematically compared severe pneumonia patients infected with COVID-19 versus H7N9 viruses in terms of clinical presentations, laboratory tests, virologic shedding, image characteristics, complication and clinical outcome in context of critical bundle and intensive management, in order to provide some guidance for their differential consideration.

Patients
All of the severe COVID-19 pneumonia and severe H7N9-induced pneumonia subjects were con rmed by laboratory tests and were hospitalized at The Fifth People's Hospital of Suzhou. The date of disease onset and hospital admission date, as well as the severity of disease were recorded. The severe COVID-19 pneumonia cases were hospitalized from January 2020 to March 2020, who was de ned according to the diagnostic and treatment guideline for COVID-19 pneumonia issued by the National Health and Family Planning Commission of P.R.China (Version 1-7). The severe H7N9 cases were hospitalized from 2014 to 2016, who met the following criteria published by the National Health and Family Planning Commission of P.R.China (the 2nd edition, 2013).

Data collection
The data of all patients were collected from an electronic case report form and included the following: demographic characteristics (age and sex), comorbidities, clinical symptoms, laboratory tests (blood routine test, arterial blood gas analysis (ABG), and blood chemistry), virologic test, microbiological ndings and images of the lung (chest CT). Antimicrobiological therapy, respiratory support, complications and outcomes were also recorded. The Ethics Committee of The Fifth People's Hospital of Suzhou approved this study (2020-005).

Study design
This was a retrospective case-control study. We compared two independent cohorts of patients infected with either COVID-19 or H7N9 in terms of initial onset, further course and outcome, which included clinical presentations at diagnosis, virologic shedding, peak values of laboratory tests, time course of image characteristics, the worst SOFA and PSI score, extrapulmonary complication, secondary bacterial infection, intensive therapy, and clinical outcome.

De nitions
Septic shock was de ned according to the 2016 Third International Consensus De nition for Sepsis and Septic Shock [6]. Secondary bacterial infection was diagnosed when patients showed clinical symptoms or signs of pneumonia or bacteraemia and a positive culture of a new pathogen was obtained from lower respiratory tract specimens (quali ed sputum, endotracheal aspirate, or bronchoalveolar lavage uid) or blood samples after admission [7]. Extrapulmonary complication was de ned as: i acute kidney injury was diagnosed according to the KDIGO clinical practice guidelines; ii acute cardiac injury was diagnosed if serum levels of cardiac biomarkers (eg, high-sensitive cardiac troponin I) were above the 99th percentile upper reference limit; iii acute liver injury was diagnosed if serum levels of ALT or TBIL was above the 2fold of upper reference limit; iv coagulopathy was de ned as a 3-second extension of Prothrombin time (PT) [8][9]. RNA shedding duration was de ned the interval from admission to the date of the rst RNA negative result before discharge.

Statistical analysis
Data were described as the mean ± sd, median (interquartile range, IQR) or number (%). The comparisons of the features between the different subtypes of virus (H7N9 and COVID-19) were performed with the t test to compare the mean ± sd of continuous variables, Mann-Whitney U test to compare the medians of continuous variables, and Fisher exact test or the chi-squared test to compare proportions. To identify risk factors associated with severe COVID-19 infection, we performed a multivariable logistic regression analysis adjusted for baseline covariates. Statistical analyses were performed using SPSS, version 24.0 for Windows, probabilities were 2-tailed, and a 2-tailed p value of < 0.05 was considered signi cant.

Dermatologic feature
As shown in Table 1, the median age of severe COVID-19 patients was 49 years old, which was comparable to that of severe H7N9 patients (54 years old, p>0.05). The proportion of males in severe COVID-19 patients was 69.57%, which was also comparable to that of severe H7N9 patients (65.22%, p>0.05). 26.09% of severe COVID-19 patients has a history of underlying diseases, whereas that of severe H7N9 patients was signi cantly higher, at 56.52% (p=0.04). The majority of severe H7N9 patients suffered to hypertension (p<0.01). There was no signi cant difference in the history of diabetes, chronicairway diseases between the two groups (p>0.05).
Worst Oxygenation Index (OI) during the hospitalization predicts deteriorated respiratory failure. The minimal level of OI in severe COVID-19 patients was 228.30±11.01 mmHg, which was signi cantly higher than the 113.7±11.18 mmHg of severe H7N9 patients (p<0.05).

Imaging ndings
In terms of imaging characteristics (Table 2), ground-glass opacity in initial chest CTs was common in COVID-19 patients (43.48%) and in H7N9 patients (60.87%, p>0.05). In contrast, consolidation and pleural effusion were more common in H7N9 patients than in COVID-19 patients (p<0.05 for each), crazy-paving pattern was less seen in H7N9 patients (p<0.05).

Further course and intensive treatment
Over the course of the viral infections (Table 3), septic shock occurred in 13.04% of patients with severe H7N9. There was no signi cant difference in the duration of onset to ARDS between severe H7N9 and severe COVID-19 patients (8.39±0.59 days, 8.09±0.72 days, p=0.74). The highest sequential organ failure assessment (SOFA) score and pneumonia severity index (PSI) score of severe COVID-19 patients were 3.57±0.31 and 61.22±2.99, respectively, which were lower than the scores of 5.35±0.77 (p=0.04) and 89.96±7.72 (p=0.00) for severe H7N9 patients.
All patients have received medical bundle intervention, including antimicrobial therapy, uid administration, respiratory support, or steroid therapy.
The two groups presented with a variety of accompanying secondary bacterial infection. 39.13% of severe H7N9 patients had positive culture of pathogen isolated from quali ed lower respiratory tract specimens, whereas that of severe COVID-19 patients was signi cantly lower (8.70%, p<0.05).
In terms of respiratory support, during the entire process of treatment, the proportions of severe H7N9 patients who received non-invasive mechanical ventilation (NIV) was signi cantly higher than that of COVID-19 patients (69.57% vs 21.74%, p<0.05). The failure rates of NIV in severe H7N9 patients were comparable to than those in COVID-19 patients (4.35% vs 0%, p>0.05).
In addition to treatments above, 82.61% of COVID-19 patients received glucocorticoids, which was comparable to the proportion of 95.65% in H7N9 patients (p>0.05). Low-molecular-weight Heparin was administered in 43.48% of COVID-19 patients, which also was comparable to that administered to H7N9 patients (34.78%, p<0.001).

Virologic outcomes and prognosis
All of the patients received antiviral therapies. Oseltamivir was administered in all of the H7N9 patients. However, COVID-19 patients had a variety of antiviral treatments, included 43.48% with lopinavir/ritonavir, 8.70% with arbidol, and 47.83% with combination. The duration of severe COVID-19 RNA shedding from the admission was 13.91±1.56 days, which was longer than that of severe H7N9 patients (9.86±0.81 days, p=0.03).
Based on the follow-up of chest CT, the time to beginning absorption on chest CT (TTBAC) was established. Severe H7N9 patients have longer TTBAC than that in severe COVID-19 patients (11.64±0.83 vs 9.13±0.70 days, p=0.02).
In terms of prognoses, although the in-hospital mortality of H7N9 patients with ARDS was 13.04%, it did not reach the statistical signi cance when compared to COVID-19 patients (0, p<0.001).

Discussion
In this study, we compared the clinical features and courses of patients with severe pneumonia caused by COVID-19 and highly pathogenic avian In uenza (H7N9). As su cient medical staff and medical supplies have affected the treatments and prognoses of severe cases, this retrospective study was conducted in a resident designated hospital, where medical resources reached standardized respiratory support in accordance with related guidelines.
Because of different therapies, it is important to differentiate these two diseases via clinical presentations. We found that compared with features in H7N9 patients, COVID-19 patients were less inclined to exhibit dyspnea, myalgia, and nasal congestion. Therefore, we speculate from previous research that severe H7N9 infection may present as more systemic and catarrh symptoms compared with the COVID-19 infection. Furthermore, based on the proportions of underlying diseases in these two groups, it was indicated that the combination of underlying diseases had a signi cant effect on the severities of H7N9 infection [1][2][3][4][5].
The main manifestations on chest CT for both diseases were characterized by ground-glass opacities (GGOs) [10]. Additionally, we found that crazy-paving pattern was more common in severe COVID-19 patients than in H7N9 patients, whereas consolidation and pleural effusion were more frequent in severe H7N9 patients. It was suggested that the combination of radiologic ndings might have a certain value in the differential diagnosis of the two diseases.
Clinically, although severe COVID-19 patients had similar extra-pulmonary complication to those observed in the severe H7N9 patients, they were less inclined to surfer to secondary bacterial infection, severe thrombocytopenia, acute cardiac injury and impairments in coagulation. The majority of patients had increased coagulation activity, marked by increased D-dimer concentrations. High levels of D-dimer have a reported association with 28-day mortality in patients with infection or sepsis [11]. These experimental indexes were considered to be closely related to the severities of the severe infections [9,12].
In term of disease severity, duration from the onset to con rmed diagnosis of ARDS in severe H7N9 and COVID-19 patients was comparable. However, severe COVID-19 patients were more inclined to surfer to relative better disease severity score. Also, according to the PaO 2 /FiO 2 in severe COVID-19 patients, the corresponding OI was signi cantly higher than that of severe H7N9 patients, suggesting the relatively moderate conditions of the severe COVID-19 patients. As supported, serial chest CT showed that the absorption of the lesions among the H7N9 patients was slow. The main cause might have been more severe damage to the lung tissue.
There are still no speci c antiviral medicines or vaccines recommended for COVID-19 infection. Under the circumstances, comparison of duration of virus shedding between severe COVID-19 and H7N9 infection may produce more interesting ndings. Surprisingly, the severe COVID-19 group had longer duration of viral shedding from the admission than the severe H7N9 group. This has important implications for the guidance around the length of antiviral treatment.
We also found that severe COVID-19 patients received a wider variety of treatments that were similar to severe H7N9 patients. The application of glucocorticoids and LMWH were both common in COVID-19 and H7N9 patients in our present study. The treatments for ARDS of the two groups were mainly based on NPPV. More of the patients with severe H7N9 needed NPPV and ultimately cured. However, additional prospective and comparative trials that address the need for intubation and the mortality rate are required [13-14].

Conclusion
Taken together, compared to COVID-19, patients with H7N9 complicated by severe pneumonia had much more severe disease. Given su cient medical staff and medical supplies, the relatively lower mortality rate of these two severe cases is to be expected. However, there were some limitations of our present study. First, this was a retrospective study that included data from one independent single-center cohorts, which may have resulted in unavoidable bias. Second, more observations are needed to clarify further clinical features by a large scale of investigation. Informed consent was waived due to the retrospective nature of the study. Patients data were obtained from an electronic case report form. The data were analyzed anonymously.

Competing interests
The authors declare that they have no competing interests.
Author contributions: T.P.J and C.C. conceived the idea, designed, and supervised the study, drafted the manuscript. G.B.B. and Y.L. collected data and had full access to all of the data and took responsibility for the integrity of the data. Z.X.Y. analyzed data and performed statistical analysis. All of the authors reviewed and approved the nal version of the manuscript.