Research on pneumonia exacerbation in patients infected with SARS-CoV-2 in Wuhan, China

Background SARS-CoV-2 pneumonia occasionally exacerbates to critical condition that is hard to manage. We aim to describe exacerbations of SARS-CoV-2 pneumonia among inpatients. Methods We included confirmed SARS-CoV-2 patients with pneumonia exacerbation admitted to Wuhan Pulmonary Hospital, Hubei Province, China between January 6 and February 17, 2020 and discharged or died before February 25. Their demographic characteristics, clinical symptoms, laboratory tests, CT manifestations, complications and clinical outcomes were collected. Results A total of 158 patients were collected, among them 107 patients were stable and discharged after recovery, 24 patients were already critically severe at hospital admission. 14 patients were excluded for insufficient clinical data. Eventually, 13 confirmed cases were included. The mean age was 65 (± 9.81) years. Ten of the 13 (76.9%) patients were female. Nine (69.2%) had underlying comorbidities. Fever and cough were the most common symptoms (12/13, 92.3%). 10/13(76.9%) patients had their exacerbation in the second week of disease course. All patients had both negative and positive nucleic acid test (NAT) results during the course. Increased range of ground-glass opacity (GGO) on CT imaging are consistent to disease exacerbation. ARDS, MODS, respiratory failure were found in 5/13(38.5%), 3/13(23.1%), 6/13(46.2%) patients respectively. Five (38.5%) patients did not survive. features, clinical characteristics and computed tomography (CT)

SARS-CoV-2 pneumoina 3-7 . The aim of this study is to report the disease exacerbation during hospitalization.

Study design and material
This is a retrospective, single ¬center, observational study. Laboratory-confirmed SARS-CoV-2 patients with pneumonia exacerbation after admitted to Wuhan Pulmonary Hospital, Hubei Province, China (a COVID-19-designated hospital in the event of epidemic outbreak) between January 6 and February 17, 2020 and discharged or died before February 25 are included.
Diagnosis of SARS-CoV-2 pneumonia were confirmed by viral nucleic acid test (NAT) using highthroughput sequencing or ramplification of open reading frame 1ab (ORF1ab) and nucleocapsid protein (NP) genes fragments from sputum, pharyngeal swab or lower respiratory tract samples as descried in previous study 3 . In this study, sampling site also included urine and stool.
NAT-confirmed patients were divided into three types according to the severity grading: 1) moderate: with fever and respiratory symptoms, or with pneumonia performance on imaging. 2) severe: met one of the following criteria: respiratory distress, respiratory rate>=30 times/min; pulse oxygen saturation <93% at rest; oxygenation index (artery partial pressure of oxygen / inspired oxygen fraction (PaO2/FiO2) ≤ 300 mmHg; obvious progression of >50% lesion within 24-48 hours as shown on CT. 3) critically severe: met one of the following criteria: respiratory failure and mechanical ventilation required; shock; multi-organ failure required intensive care unit (ICU) care. Disease exacerbation is defined as the escalation of severity grading during the course of disease.
We counted the detection results of ORF1ab and NP genes, respectively. Double-positive is defined as both ORF1ab and NP positive while double-negative is defined as both ORF1ab and NP negative.

Data collection
We extracted the age, gender, underlying comorbidities, clinical symptoms, vital signs, laboratory findings on admission and CT during hospitalization. All laboratory testing and examinations are performed according to the clinical care needs of the patient.

Definitions
The course of illness of non-survivor cases was defined as the duration from onset to death. The course of illness of survivor cases was defined as the duration from onset to the sampling date of the second negative NAT of two consecutive NATs before discharge. We adopted this new method of definition to reflect the duration of detectable SARS-CoV-2 RNA as accurate as possible.
Early Admission was defined as hospitalized within 7 days from onset. Late admission was defined as hospitalized after 7 days from onset.
Lymphopenia was defined as lymphocyte count<0.8*10^9/L. Thrombocytopenia is defined as platelet count < 100*10^9/L. Anemia is defined as hemoglobin < 110g/L. Hyponatremia is defined as serum sodium < 135mmol/L. Hypokalemia is defined as serum potassium < 3.5mmol/L. Hypoalbuminemia is defined as plasma albumin < 35g/L. Elevation or reduction criteria of each laboratory value were listed on Table1.
The survival outcome was defined as survival to hospital discharge.

Statistical analysis
All continuous variables were described as mean (standard deviation (SD), range) and categorical variables were defined as number and percentage. SPSS software (version 19) was used for the statistical analysis.

Nucleic acid tests
All NAT results and the corresponding sampling sites of 13 patients are collected and shown on Fig. 2.
All cases had both double-negative result and double-positive result in different during disease course. NP gene positive with ORF1Ab negative result are shown in four cases (case 6,8,9,13). It is worth noting that two cases (case 2 and 4) tested positive post hospital discharge whereas their two consecutive NAT tests showed double-negative results before discharge.
One patient with unilateral abnormality at onset progressed to bilateral in 3 days. As the disease progressed, the lesions became consolidated, and central zone and more lung lobes were involved.
Pleural effusion and thickened pleura were detected in 2 different patients.

Exacerbations,complications and outcome
Seven cases (case 1,2,3,4,11,12,13) presented as moderate type at admission, among them only case11 progressed to the critically severe type. The other six cases progressed to the severe type.
Review the age distribution of 13 patients, 11(11/13 85%) had exceeded 60 years old. The majority of patients (9/13 69%) had underlying comorbidities, including but limited to hypertension, diabetes, heart disease, chronic kidney disease, hepatic disease. Two patients under 60 years of age had obesity, hyperlipidemia (case1), and coronary artery disease, hypertension (case2) separately. Aging and underlying diseases may collaborate and contribute to the decline of host immunity that is the prerequisite of virus infection and disease exacerbation.
In this study, 10/13(76.9%) cases were female. This ratio seems to be quite contradictory to previous report on COVID-19 that claimed approximately 70% infections occur in male8. and also inconsistent with the proportion of male patients (67%) in severe SARS-CoV-2 pneumonia reported in another study 8 . Since this study had excluded patients admitted but has not been discharged or died before February 25, the total number and proportion of male patients could be underestimated.
Disease exacerbation mostly occurred in the second week of disease course (10/13, 76.9%). Patients (case 1,4,5,10,11,12,13) admitted early (within 7days) of the disease course had a relatively better prognosis than those (case 2,3,6,7, 8,9) admitted later (after the first 7days). The early group demonstrated lower probability of complications and death. These results indicate that early admission is crucial to restrain deterioration and reduce mortality.
The overall mortality rate was 38.4%(5/13), and in critically severe cases,the mortality rate was 71.4%(5/7), slightly higher than former study 8 . These results may be attributed to the fact that the inclusion criteria of this study are different from those of the above studies. These data do not apply to the entire herd of infected patients.
In our study, fever and cough were the most common symptoms at hospital admission. Initial reports of the COVID-19 virus suggested that symptoms of SAR-CoV-2 pneumonia are similar to that of SARS and MERS, since most patients presenting with fever, cough, fatigue and hypodynamia 9,10 . However, one patient (case12) did not manifest fever during the entire course of disease. This is an 80 years old female patient presented to clinic with a complain of chest distress and palpitation. Since she was in the epidemic area, she underwent a CT scan which demonstrated pneumonia. Then nucleic acid test was administrated and this case is confirmed. Besides, patients presented diarrhea as the initial symptom of disease have been reported 11 . These extra-pulmonary initial symptoms deserve more attention outside of fever clinic.
Elevation of C-reaction protein, D-Dimer and lactate dehydrogenase, reduction of lymphocyte count and albumin are observed in more than half of the cases. Anemia, transaminase abnormality, elevated myocardial index and renal dysfunction are also detected as disease exacerbate. These changes are consistent with previous reports 3, 12,13 . Ground-glass opacity and consolidation are the main changes in CT imaging. CT manifestations are indicative of the exacerbation of pneumonia. In the process of exacerbation, more lobes were involved, range of GGOs were expanded and density of consolidation was increased. Additional signs on CT imaging include vascular enlargement, interlobular septal thickening in crazy-paving pattern, air bronchogram sign and discrete pulmonary nodules, which were previously reported in some studies 14,15 . CT imaging may serve as a standard method in the rapid diagnosis of COVID-19.
Previous studies have reported that the sensitivity of chest CT is significantly greater than that of RT-PCR (98% vs 71%, respectively, p < .001) 16 . And false negative rate is very low (3.9%)16.

Nevertheless, CT is still incapable of distinguish between different viruses.
Nucleic acid RT-PCR testing is a standard test for suspected cases to confirm the SARS-CoV-2 infection. Sampling site plays an important role in virus detection. Case 10 was sampled at 4 sites (pharyngeal swab, urine, venous blood, stool) on the same day, only stool sample demonstrated positive result. The patient's medical records showed severe diarrhea and significant weight loss, which suggests that stool specimen in patients presenting gastrointestinal symptoms is probably more sensitive to the virus than pharyngeal swabs. Controversial NAT results were also observed in case 8 whose sputum and urine sample were taken on the same day. Urine sampling is not of high priority in clinical routine due to lack of sufficient data to support it as a usual shedding route of coronavirus. For different sampling sites, bronchoalveolar lavage fluid exhibited the highest positive rate, followed by sputum, nasal and pharyngeal swabs showing poor positive rate in patients with fever 17 . Another study reported the detection rates of SARS-CoV-2 from sputum specimens are significantly higher than throat swabs 18 . SARS-CoV-2 has the ability to transmit through multiple routes 19,20 and manifest diverse clinical symptoms which should be taken into account while sampling. We infer that while determine where specimens are collected, considering the location of initial symptoms may increase positive rate.
In 4 cases (case 6,8,9,13), NAT showed ORF1ab negative and NP positive result. Low viral load might lead to this result since PCR kits currently used in clinical practice are generally more sensitive to the amplification of the N protein gene than ORF1ab. On the other hand, crossovers of other coronaviruses may cause the same result since ORF1ab sequence is more conservative than N protein gene. Patients presented single-positive result should be re-checked by test kits from a different manufacture or different test method.
NAT results has turned out to be controversial in 2 cases(case2,4). After discharge from hospital (20   days for   The course of illness of all patients.  High-Resolution computed tomographic imaging of case 2 and case 10. The imaging of these two cases showed us that the features of CT imaging were presented as ground glass opacity (GGO), peripheral distribution and bilateral lung involvement. As the disease progressed, the lesions became consolidated, central zone and more lung lobes are involved. Panel A1, A2 and A3 demonstrated an exacerbation process of case 10. Her condition rapidly deteriorated into critically severe type after admission and did not survive eventually. Meanwhile, CT imaging manifested increased range of GGO and consolidation. A typical "white lung" can be seen in Panel A3. CT imaging of case 2 was shown in Panel B1, B2 and B3. In Panel B1, the proportion of lesions were 35%. In the process of pneumonia exacerbation, the proportion climbed up to 90% (shown in Panel B2). The range of lesions decreased to 57.5% (shown in Panel B3) when her condition is relieved. *number of associated lung lobes ※CT imaging was unavailable because CT examination was completed before admission in another hospital. # CT imaging of case7 and case 8 are absent because their critical condition does not permit a CT scan.

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