key: cord-0795880-rkupc9w0 authors: Liu, Chengyin; Wen, Ying; Wan, Weiguo; Lei, Jingchao; Jiang, Xuejun title: Clinical characteristics and antibiotics treatment in suspected bacterial infection patients with COVID-19 date: 2020-11-03 journal: Int Immunopharmacol DOI: 10.1016/j.intimp.2020.107157 sha: 6ab8a02cd49c4901bfb6ffe483c4366364ae77ad doc_id: 795880 cord_uid: rkupc9w0 Coronavirus disease 2019 (COVID-19) pandemic has brought challenges to health and social care systems. However, the empirical use of antibiotics is still confusing. Presently, a total of 1123 patients with COVID-19 admitted to Renmin Hospital of Wuhan University was included in this retrospective cohort study. The clinical features, complications and outcomes were compared between the suspected bacterial infection and the no evidence of bacterial infection. The risk factors of mortality and the incidence of acute organ injury were analyzed. As a result, 473 patients were selected to suspected bacterial infection (SI) group based on higher white blood cell count and procalcitonin or bacterial pneumonia on chest radiography. 650 patients were selected to the no evidence of bacterial infection (NI) group. The SI group had more severely ill patients (70.2% vs. 39.8%), more death (20.5% vs. 2.2%), and more acute organ injury (40.2% vs. 11.2%). Antibiotics were found associated with improved mortality and an increased risk for acute organ injury in hospitalized patients with COVID-19. Intravenous moxifloxacin and meropenem increased the death rate in patients with suspected bacterial infection, while oral antibiotics reduced mortality in this group. Moreover, penicillin and meropenem treatments were associated with increased mortality of the patients with no evidence of bacterial infection. In conclusion, patients with suspected bacterial infection were more likely to have negative clinical outcomes than those without bacterial infection. Empirical use of antibiotics may not have the expected benefits. Lower respiratory tract infections had a major impact on public health over the past two decades [1] . Severe acute respiratory syndrome (SARS) [1] , Middle East respiratory syndrome (MERS) [4] , and Coronavirus disease 2019 (COVID-19) [5] have presented an unprecedented challenge for the healthcare community worldwide. In previous influenza pandemics, studies have considered that bacterial infection accompanied with viral infection is an important factor affecting mortality [3] . Bacterial co-infection and secondary bacterial infection are considered critical risk factors for the severity and mortality rates of COVID-19. The prevalence of bacterial infection was range from 14% to 100% in intensive care unit (ICU) patients [7] [8] [9] . A recent meta-analysis suggested that only 7% of COVID-19 patients were expected to have bacterial infection [10] . However, it should be noted that most patients received antibiotics before they were diagnosed with bacterial infection, which may influence the prevalence of bacterial infection. A low rate of laboratory-confirmed bacterial coinfection was observed in patients with COVID-19 in the latest studies [11] [12] . Thus, the widely accepted view was the rate of bacterial infection with COVID-19 is not high. Using antibiotics to treat viral diseases may increase drug resistance and raise the risks of allergic reactions [13] . Herein, determining the presence of bacterial infection is essential to guide the empirical use of antibiotics, thereby reducing antibiotic abuse. On the other hand, the diagnosis of bacterial infection is sometimes challenging because clinical symptoms of infections are similar. Therefore, it may be hard to differentiate viral from bacterial infection. Raised procalcitonin observed in COVID-19 could be due either to bacterial infection [14] and non-elevated procalcitonin, which may be a good predictor of the absence of bacterial infection. Studies have reported that patients infected with other pathogens had higher white blood cell (WBC), neutrophil counts, Ddimer, C-reactive protein (CRP) and procalcitonin (PCT) levels than those of infected with SARScoronavirus 2 (SARS-CoV-2) homogeneously [15] . The empirically use of the antibiotics was used for patients with suspected bacterial infection. However, it was unknown whether this empirically use of the antibiotics had positive outcomes. The diagnosis of COVID-19 was confirmed as a positive result for a nasopharyngeal swab and respiratory pathogen nucleic acid test with high-throughput sequencing or real-time reverse transcriptase-polymerase chain reaction (RT-PCR). The severity of COVID-19 was categorized as moderate and severe. The moderate type represents patients with non-pneumonia and mild to moderate pneumonia. The severe type was characterized by (1) dyspnea (respiratory rate ≥30/min); (2) blood oxygen saturation ≤93%; (3) PaO 2 /FiO 2 ratio <300 or lung infiltrates >50% within 24-48 h [16] . If one of the above items was met, it was classified as severe. For bacterial infection, the COVID-19 patients were grouped as suspected bacterial infection (SI) and no evidence of bacterial infection (NI). The patients were classified as the SI group if the any one of (1)-(4) and (5) were met: (1) Recent cough, expectoration or aggravation of original respiratory disease symptoms, and purulent sputum, with or without chest pain; (2) White blood cell count>10×10 9 /L; (3) PCT>0.1 ng/ml; (4) Fever (body temperature is over 37 ℃), (5) Chest radiography showed bacterial pneumonia. The NI group was characterized by normal white blood cell count and PCT, and the chest radiography showed viral pneumonia. The acute organ injury was described as acute heart injury, acute kidney injury and acute liver injury newly discovered after admission. Acute kidney injury was defined on the basis of highest serum level of creatinine according to the Kidney Disease Improving Global Outcome (KDIGO) clinical practice guidelines [17] . Cardiac injury was defined as a serum level of cardiac troponin I above the 99th percentile upper baseline limit [18] . Acute liver injury was identified as an increase in alanine aminotransferase of 5 times the upper reference limit or alkaline phosphatase increase of twice the upper reference limit. IBM SPSS Statistics (version 23.0) was adopted for statistical analysis. Categorical variables were expressed as proportions and compared between groups using the Chi-square test. Continuous data were expressed as median (interquartile range) and compared between groups using the Mann-Whitney U test. Multivariate logistic regression models were utilized to explore the risk factors associated with in-hospital death. Forest plots were employed to display logistic regression analysis results. The incidence of acute organ injury was examined with the same method of the mortality model. A 2-sided α of less than 0.05 was considered statistically significant. Of 1500 COVID- 19 and glucocorticoid treatment were associated with an increased incidence of acute organ injury in patients with COVID-19 (Figure 1 B) . Since different groups had significantly different outcomes, stratified analysis for the SI group and NI group was made to find the association between death and antibiotic therapy. The specific types of antibiotics were incorporated into the statistics. In the SI group, older age, severer illness, and intravenous moxifloxacin and meropenem were associated with increased risks of death. In contrast, receiving oral antibiotics enhanced mortality (Figure 2 A) . Meanwhile, older age, male gender, intravenous antibiotics and receiving meropenem were associated with the elevated incidence of acute organ injury in the SI group (Figure 2 B) . On the contrary, receiving azithromycin and female gender were related to decreased acute organ injury incidence in the SI group. For the NI group, receiving penicillin, meropenem, and glucocorticoids treatments was associated with an increased death rate (Figure 3 A) . In contrast, older age, male gender, severe illness, receiving antivirals, glucocorticoids and meropenem were linked to the increased incidence of acute organ injury (Figure 3 B ). Our results showed that patients with suspected bacterial infection were more likely to have negative clinical outcomes than those with no evidence of bacterial infection. The suspected bacterial infection group included severely ill patients, and the outcomes showed that the SI group had more death and complications. Similarly, a recent review paper suggested that a mortality rate of 10.9% (53 of 482) was observed in all cases of viral pneumonia due to secondary infections [19] . It is probably because that increased procalcitonin values are associated with a nearly 5-fold higher risk of severe SARS-CoV-2 infection [20] , and bacterial infection is likely to worsen an already poor prognosis [21] . Based on past experience, clinicians often take more drastic treatment for patients with severe illness. We can also see this trend in our research. The proportion of patients receiving antibiotics and glucocorticoids in the SI group was significantly higher than that of the NI group. We identified several risk factors for in-hospital death of the patients with COVID-19. Receiving antibiotics was associated with higher in-hospital death. Previous studies demonstrated that the administration of multiple antibiotics did not change the outcomes of the disease [22-23[23] . In our center, antibiotic therapy seemed to increase the risk of death. It's because clinicians are more likely to use antibiotics for severe patients, who often die of SARS-CoV-2. The rate of bacterial infection is not high, but 70.5% of patients received antibiotics in our research, which means most patients would not benefit from antibiotics. Evidence suggests that very high proportion of COVID-19 patients were receiving unnecessary antibiotic treatment Error! Reference source not found.[25] [26] . This increase in antibiotic administration can cause pressure on bacterial pathogens which lead to antibiotic resistance [27] . Yet, there is a serious problem that the presence of antimicrobial elements in the environment can stimulate antimicrobial resistance [28] and the potential consequence of the COVID-19 pandemic is the long-term propagation of antimicrobial resistance [29] . Both guidelines for COVID-19 management suggest prescribing antibiotics therapy for suspected bacterial infection [30] [31] . In the present study, we found that if laboratory tests indicated suspected bacterial infection, the empirical use of oral antibiotics might reduce mortality of patients with COVID-19. Previous studies demonstrated that procalcitonin and neutrophilic leukocytosis are valuable markers for bacterial infection. Timothy MU [32] believes that procalcitonin is not specific in differentiating viral and bacterial pneumonia. We still chose these ways because when this new virus occurred in China, there is no standard strategy. As a result, some doctors chose routine empiric antibiotic use in patients with elevated WBC or PCT. In terms of the value of PCT, there is a lot of controversies. In Meisner's research [33] , the patients with PCT value >0.25 µg/liter were defined bacterial infection and those who had PCT value ≤0.25 µg/liter were defined nonbacterial [33] . In our center, the PCT value >0.1 µg/liter indicated the presence of bacterial infection. Hence, we chose PCT value >0.1 µg/liter as the indicators of grouping. From our study, the WBC and procalcitonin may not be the biomarkers for distinguishing bacteria from SARS-CoV-2. Facing the increased WBC or procalcitonin, clinicians should not immediately choose antibiotics, especially intravenous antibiotics. Empiric antibiotic use in patients with COVID-19 will not offer considerable benefit. The reliable way is to rely on the microbiological results. Receiving azithromycin may decrease the incidence of acute organ injury in suspected bacterial infection. The results were interesting. Azithromycin has been the focus of intense scholarly debate. The in vitro evidence suggests that azithromycin has antiviral properties [34] , including against SARS-CoV-2. A retrospective study found that treatment with hydroxychloroquine and azithromycin reduced COVID-19-related mortality [35] . However, another retrospective cohort study of 1438 patients in New York [36] showed that treatment with hydroxychloroquine and azithromycin was not associated with lower in-hospital mortality. Kome Gbinigie [37] [37] reported that no evidence supports the use of azithromycin for the treatment of COVID-19, unless it is used to treat bacterial super-infection. Collectively, no link was found between azithromycin and mortality, but azithromycin repressed organ damage from suspected bacterial infection. Nevertheless, the number of patients receiving azithromycin in our research is small and more prospective studies are needed to clarify the association. For the patients with no evidence of bacterial infection, receiving penicillin and meropenem increased mortality, and meropenem elevated the incidence of acute organ injury. It was confirmed that antibiotic therapy should only be used for patients with a presentation suggestive of bacterial infection or supportive, positive microbiology [38] . Hantoushzadeh S and his colleagues [39] reported that the consumption of some antibiotics in the lack of bacterial infection could lead to undesirable outcomes. We found that meropenem was related to increased mortality and organ damage in both SI group and NI group. It means meropenem may not be the first choice for empirical antibiotic use in patients with COVID-19. Our study presents several limitations. Firstly, due to the imperfect clinical records, there was no record of symptom changes, which limited our analysis of the clinical efficacy. Secondly, our study was retrospective, and the impact of the treatments on the patients was inferred. In addition, the number of patients and treatment drugs were limited. In the future, more rigorous case-control studies will be needed to clarify the effect of antibiotic treatment, and more work is needed on biomarkers to help identify bacterial infection. Patients with suspected bacterial infection were more likely to have negative clinical outcomes than those with no evidence of bacterial infection. Antibiotic therapy was associated with increased mortality and most patients would not benefit from antibiotics. Empirical use of antibiotics may not have the positive outcomes as expected. We are in the pandemic of COVID-19, understanding the potential for bacterial infection is important. Detailed antimicrobial policies and guidelines should be applied and promoted in the times of COVID-19. Conceived and designed the experiments: Xuejun Jiang; Performed the experiments: Chengyin Liu, Ying Wen,Weiguo Wan, Jingchao Lei; Statistical analysis: Weiguo Wan, Jingchao Lei; Wrote the paper: Chengyin Liu, Ying Wen. All authors read and approved the final manuscript. The authors declare that they have no competing interests. Our study was approved by the Ethics Review Board of Renmin Hospital of Wuhan University. The data sets used and analyzed during the current study are available from the corresponding author on reasonable request. Challenges to the system of reserve medical supplies for public health emergencies: reflections on the outbreak of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) epidemic in China SARS and other coronaviruses as causes of pneumonia SARS--beginning to understand a new virus Understanding the Latest Human Coronavirus Threat. Viruses COVID-19 pandemic: A global health burden Bacterial and viral infections associated with influenza. Influenza Other Respir Viruses Secondary Bacterial Infections in Critical Ill Patients With Coronavirus Disease Bacterial and viral co-infections in patients with severe SARS-CoV-2 pneumonia admitted to a French ICU Evaluation of bacterial co-infections of the respiratory tract in COVID-19 patients admitted to ICU Co-infections in people with COVID-19: a systematic review and meta-analysis Improving antibiotic stewardship in COVID-19: Bacterial co-infection is less common than with influenza Bacterial and fungal coinfection among hospitalized patients with COVID-19: a retrospective cohort study in a UK secondarycare setting Drivers of Irrational Use of Antibiotics in Europe Procalcitonin for patient stratification and identification of bacterial co-infection in COVID-19 Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention KDIGO clinical practice guidelines for acute kidney injury Secondary Bacterial Infections in Patients With Viral Pneumonia Procalcitonin in patients with severe coronavirus disease 2019 (COVID-19): A meta-analysis Investigation of subsequent and co-infections associated with SARS-CoV-2 (COVID-19) in hospitalized patients Early administered antibiotics do not impact mortality in critically ill patients with COVID-19 Clinical Features of 85 Fatal Cases of COVID-19 from Wuhan. A Retrospective Observational Study Limited Role for Antibiotics in COVID-19: Scarce Evidence of Bacterial Coinfection Antimicrobial resistance in the age of COVID-19 Will the overuse ofantibiotics during the coronavirus pandemic accelerate antimicrobial resistanceof bacteria? Bacterial co-infections with SARS-CoV-2 Environmental side effects of the injudicious use of antimicrobials in the era of COVID-19 Antimicrobial use, drug-resistant infections and COVID-19 COVID-19 rapid guideline: managing suspected or confirmed pneumonia in adults in the community Clinical management of COVID-19 Update on Diagnosis, Treatment, Chemoprophylaxis, and Institutional Outbreak Management of Seasonal Influenzaa Update on procalcitonin measurements Clinical Pharmacology Perspectives on the Antiviral Activity of Azithromycin and Use in COVID-19 Early treatment of COVID-19 patients with hydroxychloroquine and azithromycin: A retrospective analysis of 1061 cases in Marseille, France Association of Treatment With Hydroxychloroquine or Azithromycin With In-Hospital Mortality in Patients With COVID-19 in Should azithromycin be used to treat COVID-19? A rapid review COVID-19 rapid guideline: antibiotics for pneumonia in adults in hospital Possible Cause of Inflammatory Storm and Septic Shock in Patients Diagnosed with (COVID-19) This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.