key: cord-0698160-7rv402oc
authors: Pasquini, Zeno; Barocci, Iacopo; Brescini, Lucia; Candelaresi, Bianca; Castelletti, Sefora; Valentina, Iencinella; Mazzanti, Sara; Procaccini, Gaia; Orsetti, Elena; Pallotta, Francesco; Amadio, Giorgio; Giacometti, Andrea; Tavio, Marcello; Barchiesi, Francesco
title: Bloodstream infections in the COVID-19 era: results from an Italian multicenter study
date: 2021-08-18
journal: Int J Infect Dis
DOI: 10.1016/j.ijid.2021.07.065
sha: 6039c589355177a72bd169fc29a739d096d2a28c
doc_id: 698160
cord_uid: 7rv402oc

BACKGROUND: Correlation between COVID-19 and superinfections has been investigated, but remains to be fully assessed. In this multicenter study we report the impact of the pandemic on bloodstream infections. METHODS: This study includes all patients with a BSI admitted to four Italian hospitals between January 1st and June 30th, 2020. Clinical, demographic, and microbiologic data were compared with those of patients hospitalized during the same period of 2019. RESULTS: Among 26,012 patients admitted in the first semester of 2020, 1,182 had COVID-19. In patients with COVID-19, we observed 107 BSIs, with an incidence rate of 8.19 episodes per 1000 patient-days. This incidence was significantly higher than in patients without COVID-19 (2.72/1000 patient-days) and in patients admitted in 2019 (2.76/1000 patient-days). In comparison with patients without COVID-19, BSIs onset in patients with COVID-19 was delayed during the course of hospitalization (16.0 vs 5 days). Thirty-day mortality among patients with COVID-19 was 40.2%, significantly higher than in patients without COVID-19 (23.7%). BSIs in patients with COVID-19 were frequently caused by MDR pathogens, which were often center-dependent. CONCLUSIONS: BSIs are a frequent secondary infection in patients with COVID-19, characterized by an increased risk during hospitalization and potentially burdened with high mortality.

The worldwide spread of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection has caused a massive global health challenge (Lupia et al., 2020) (Pasquini et al., 2020) : the Coronavirus Resource Center of the Johns Hopkins University in Baltimore (Maryland, United States) reported that at the end of February 2021, 114,338,204 cases confirmed and 2,535,737 deaths globally (Home, n.d.) . Patients affected by Coronavirus disease 2019 , the SARS-CoV-2-related syndrome, often require hospital admission, and a large percentage need invasive treatments in intensive care unit (ICU) (Petrilli et al., 2020) .

As shown by several studies on other respiratory viruses, primary infection confers an increased susceptibility to develop co-infections (Lee et al., 2011) (Chertow and Memoli, 2013 ) and secondary infections due to bacteria and fungi (Lee et al., 2011) (Chertow and Memoli, 2013) . Co-infection is defined as acute infection detected on presentation of the primary infection (Langford et al., 2020) , secondary infection is defined as an infection detected ≥48 hours after admission with a positive culture of a new pathogen from a lower respiratory tract specimen or blood taken ≥48 hours after hospital admission (Huang et al., 2020) (Lansbury et al., 2020) . These conditions contribute to increase both the severity of the disease and the mortality (Lee et al., 2011) (Chertow and Memoli, 2013) , and their burden during SARS-CoV-2 pandemic has not been fully assessed yet ) (Bengoechea and Bamford, 2020) . Several studies and two systemic reviews have already attempt to assess the risk of co-infections and secondary infections (Lansbury et al., 2020) (Langford et al., 2020) (Vaughn et al., 2020) (Bartoletti et al., 2020) . Based on these studies, co-infections are generally uncommon in patients with COVID-19, while secondary infections are favored by many factors such as severity of illness, admission in ICU, need for mechanical ventilation, and longer hospital stay.

BSIs has also been evaluated, but only in small studies (Engsbro et al., 2020) , and especially in patients hospitalized in ICUs (Buetti et al., 2021) (Giacobbe et al., 2020) .

Our study aims to evaluate the burden of BSIs in patients with COVID-19 in terms of incidence, ecology and mortality, in four Italian hospitals during the first wave of SARS-CoV-2 pandemic.

This was an observational retrospective multicenter study conducted in four Italian hospitals located in the Marche Region. The hospitals involved were: 1. "Azienda Ospedaliera Ospedali Riuniti Marche Nord" that includes two secondary hospitals (Pesaro hospital and Fano hospital), mentioned as Marche Nord hospitals; 2. "Azienda Ospedaliera Universitaria Ospedali Riuniti di Ancona", a tertiary University hospital mentioned as Ancona hospital; 3. "Ospedale Augusto Murri", a secondary hospital mentioned as Fermo hospital.

We included in this study all patients (older than 18 years) with bacterial or fungal BSI admitted to these hospitals between January 1st and June 30th, 2020. Epidemiological data were compared with those of the same time period of 2019. Patients admitted in 2020 were analyzed as "patients with SARS-CoV-2 infection" or "patients with COVID-19" and "patients without SARS-CoV-2 infection" or "patients without COVID-19". Patients were further divided 7 into two subgroups: those who were hospitalized in ICU at the time of infection and those who were in ordinary wards.

The primary objective of this study was to estimate the incidence of BSIs in patients with COVID-19 admitted to four Italian Hospitals. Secondary outcomes included the evaluation of risk factors, mortality and ecology of these infections.

Patients with bacterial or fungal BSI were collected retrospectively through the hospital's patient management software. The Institutional Review Board of each Center granted retrospective access to the data without need for individual informed consent. The consent was not given since the data were analyzed anonymously. The present research was performed in accordance with the ethical standards of the 1964 Declaration of Helsinki and its later amendments.

For each patient we collected the following data: demographic characteristics, Charlson Comorbidity Index (CCI), arterial hypertension, date of first positive swab for SARS-Cov-2, date of hospital admission, date of ICU admission, date of hospital discharge, date of death, date of first BSI, isolated pathogens, hospital ward at the time of BSI, days between hospitalization and first BSI, days between first BSI and discharge, days between first BSI and death, and death after 30 days from BSI.

All patients with COVID-19 had a SARS-CoV-2 infection confirmed by reverse-transcriptasepolymerase-chain-reaction assay. Blood cultures from peripheral access or central venous catheters (CVC) were performed based on physician's advice in patients presenting clinical deterioration associated with suggestive laboratory findings. The cultured microorganisms were identified by standard techniques. Single blood cultures from contaminants pathogens such as coagulase-negative Staphylococci (CoNS) were not considered.

Normally distributed continuous data are reported as the mean ± standard deviation (SD) and compared using the two-sided Student's t test. Non-normally distributed continuous data are reported as the median and interquartile range (IQR) and compared using the Mann -Whitney test. Categorical variables were analyzed with Chi-squared test or Fisher's exact test, depending on best applicability. BSIs time of onset in patients with and without SARS-CoV-2 infection was evaluated using the Kaplan-Meier method and compared with the log-rank test. IBM SPSS Statistics version 24 (SPSS Inc. Chicago, IL, USA) was employed for statistical analysis. Statistical significance was set at p <0.05.

Between January 1 st and June 30 th 2020, a total of 26,012 patients were hospitalized in the four hospitals and 616 of these were admitted in ICUs (2.37%). Patients with confirmed SARS-CoV-2 infection were 1,182 (4.54%) and 155 of these (13.22%) needed intensive care.

In contrast, during first semester of 2019, 34,712 patients were hospitalized in these hospitals and 628 (1.81%) were admitted in ICUs (Table 1) .

The relative increase of admission in ICUs between 2019 and 2020 (1.81% vs 2.37%, p<0.001), although minimal, led to the increase in number of intensive care beds, which rose from 44 to 91.

There were 665 patients with BSIs hospitalized in 2020, 107 of them were patients with COVId-19 (Table 2) 

Among 1,182 patients with SARS-CoV-2 infection hospitalized in the four hospitals involved, we observed 107 BSIs, a duration of hospitalization of 11.05 days and an incidence of 8.19

BSIs per 1000 patients-days (Table 3 ). This rate was higher than for patients admitted in the same time period without COVID-19 (2.72 BSIs per 1000 patients-days, p<0.001) and for patients admitted in 2019 (2.76 BSIs per 1000 patients-days, p<0.001) ( Table 3 ). Duration of hospitalization in these two groups of patients was 8.25 and 7.83 days, respectively.

Analyzing BSIs incidence according to the department of onset, we observed a significant increase in ordinary wards both for patients without COVID-19 admitted in 2020 (5. Correlation between BSI onset and time after hospital admission was also analyzed through Kaplan-Meier curves (Figure 1 ). Up to the third day of hospitalization the incidence of BSIs was similar between patients with or without COVID-19, while thereafter incidence increased significantly in patients with COVID-19 ( Figure 1 ) The overall mortality in patients with COVID-19 was 40.2%, being significantly higher than in patients admitted during the same period without COVID-19 (23.7%, p<0.001) and in patients admitted in 2019 (25.2%, p < 0.001). For patients hospitalized in ordinary wards, we found a significantly increase in mortality for COVID-19 patients (41.3% vs 23.6% and 24.3%, p < 0.001), while the difference was not significant for patients hospitalized in ICUs. (Table 4 ). Table 5 describes the pathogens causing BSIs in 2019 and 2020. Overall, the incidence rate (n°/1000 hospitalizations) of four pathogens increased significantly from 2019 to 2020:

Enterococcus faecium (from 0.7 to 1.7, p<0.001), carbapenem-resistant Klebsiella pneumoniae (from 0.5 to 1.2, p = 0.009), Acinetobacter baumannii (from 0.5 to 1.0, p = 0.043) and coagulase-negative Staphylococci (from 3.5 to 5.0, p = 0.006). Considering only patients with COVID-19, the incidence rate of all pathogens increased significantly from 2019 to 2020 (p < 0.001).

The highest increase of incidence rate was observed for A. baumannii (29.3 fold, i.e.: from 0.5 to 16.1/1000 hospitalizations) followed by carbapenem-resistant K. pneumoniae (27.7 fold), By considering only patients w/o COVID-19, the only significant difference between 2019 and 2020 was a 2.2 fold increase of E. faecium (from 0.7 to 1.5, p = 0.003). (Table 5 ).

The variation of incidence for specific pathogens was center-dependent ( The burden of the SARS-CoV-2 epidemic during the considered period was different in the three centers. The percentage of patients with SARS-CoV-2 infection in Marche Nord hospitals was significantly higher than in the Ancona hospital (7.0% vs 2.8%; p<0.001) and

the Fermo hospital (7.0% vs 4.6%; p<0.001). The Fermo hospital had more hospitalizations with SARS-CoV-2 infection than Ancona hospital (4.6% vs 2.8%; p<0.001). (Table 6 ).

The pressure on ICUs was also different. Marche Nord hospitals and the Fermo hospital had similar rates of hospitalization in ICUs (20.0% and 20.1% respectively), while the Ancona hospital had only 9.1% of patients with COVID-19 admitted to ICUs. Despite these differences, there was an increase of BSIs in patients with COVID-19 in all three centers, both compared to other patients hospitalized in 2020 and to patients admitted in 2019. In Marche Nord hospitals, the BSI incidence among SARS-CoV-2 patients was 9.27 per 1000 patients-days vs 2.13 and 3.07; in the Ancona hospital it was 9.42 vs 2.88 and 2.39; and in Fermo hospital it was 4.17 vs 3.14 and 3.52 (data not shown).

On the contrary, large differences in mortality were observed between centers. Marche Nord hospitals and the Fermo hospital showed higher mortality rates in COVID-19 patients (respectively 51.8% and 41.7%) than the Ancona hospital which showed a 30-day mortality in line with SARS-CoV-2 negative patients (23.1 vs 19.9; p 0.802) and with the previous year (23.1 vs 24.6; p 0.983).

Our data show that BSIs are a frequent complication in patients with COVID-19. The increase in incidence of BSIs was evident both in ordinary wards and in ICUs, compared with the same time period of 2019. BSIs in patients with COVID-19 were mainly secondary infections, and the increase in incidence was observed from the third day of admission and continued to increase over time, well beyond the viral infection itself. This trend differs greatly from that observed for seasonal influenza, where superinfections usually occur in the first six days after symptoms onset (Chertow and Memoli, 2013) (MacIntyre et al., 2018) . This difference may suggest that, in patients with COVID-19, the altered microbiological clearance of lung alveolar epithelium, due to the viral cytopathic effect, plays a minor role in promoting superinfections than in other respiratory viruses (Morris et al., 2017 ) (Rynda-Apple et al., 2015 ) (Wilder-Smith et al., 2004 . The increased risk of secondary infection in patients with COVID-19 could be explained by four not mutually exclusive factors. First, the immune system dysregulation mostly due to two mechanisms: the "cytokine storm", triggered in response to the virus (McGonagle et al., 2020 (McGonagle et al., ) (E et al., 2020 (Fajgenbaum and June, 2020) , and the marked reduction in IFN-γ production with the consequent reduction of Th1 polarization of CD4+ T cells and cytotoxic activity (Diao et al., 2020) (Qin et al., 2020) (Yao et al., 2021) . Second, the longer hospitalization time with higher rate of admission in ICUs, which increases the risk of contracting nosocomial infections (Ripa et al., 2021) . Third, the large use of immunosuppressive treatments (e.g. corticosteroids, anti-IL6 drugs) (Bengoechea and Bamford, 2020) (Campochiaro et al., 2020 ) (Rojas-Marte et al., 2020 . Fourth, the gut-lung axis dysfunction due to changes in the gut microbiota (Dumas et al., 2018) (Ahlawat et al., 2020) (D and A, 2020) (Zuo et al., 2020) .

Although this study is not strong enough to asses the burden of these four factors, we have tried to limit some potential confounders to better describe this increase in incidence. First, we have analyzed the incidence of BSIs in number of cases per 1000 patient-days in order to adjust the different length of hospitalization. Second, we have calculated the incidence of BSIs in ICUs and ordinary wards separately, in order to limit the effect of the higher rate of admission in ICU for patients COVID-19. Doing this we have highlighted how this increase in incidence was more evident in ordinary wards, while in ICUs it was significant only if compared with 2019 and not with patients without COVID-19 hospitalized in the first half of 2020. Similarly, also mortality showed a significant increase only in ordinary wards, suggesting that particular attention is required to BSIs occurring in non-intensive settings.

With regard to mortality, contrarily to what has been observed for BSIs incidence, we observed large differences between centers. The Marche Nord hospitals and the Fermo hospital, which were the most hit by the pandemic, had a higher mortality in patients with COVID-19 than the Ancona hospital, where the pandemic did not impact the regular functioning of the hospital. This confirms that higher hospitalization rate has an independent harmful impact on the mortality in patients with COVID-19 (Soria et al., 2021) (Ji et al., 2020 ) (Carenzo et al., 2020) (Wu et al., 2020) .

From a microbiological view, we found a wide variability of pathogens causing BSIs in COVID-19 patients with high proportion of MDR organisms. Interestingly, also if carbapenemresistant K. pneumoniae isolates increased significantly in all centers, the prevalence of etiology was center-dependent. This suggests that the use of antibiotics and the local ecology play a fundamental role in the selection of MDR pathogens among COVID-19 patients. These data confirm the importance of limiting the use of antibiotics in patients with COVID-19 as already suggested (Vaughn et al., 2020) .

Despite our efforts to limit some of the potential confounders, this study still has numerous limitations manly related to the retrospective cohort design and the lack of clinical data such as SOFA score, CVC presence, data about source control, laboratory and radiographic data, and records about the treatments performed, in particular regarding immunosuppressive agents (e.g. dexamethasone, tocilizumab and baricitnib).

These limitations do not allow to investigate further with a multivariate analysis risk factors related with the increased incidence of BSIs and mortality. Further studies are needed to identify which risk factors impact on the BSIs development and which measures are best to limit it.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Nothing to declare

The Institutional Review Board of each Center granted retrospective access to the data without need for individual informed consent. The consent was not given since the data were analysed anonymously. The present research was performed in accordance with the ethical standards of the 1964 Declaration of Helsinki and its later amendments.

Immunological co-ordination between gut and lungs in SARS

CoV-2 infection

Epidemiology of invasive pulmonary aspergillosis among COVID-19 intubated patients: a prospective study

Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America

Bacterial and fungal superinfections in critically ill patients with COVID-19

SARS-CoV-2, bacterial co-infections, and AMR: the deadly trio in COVID-19?

COVID-19 increased the risk of ICU-acquired bloodstream infections: a case-cohort study from the multicentric OUTCOMEREA network

Efficacy and safety of tocilizumab in severe COVID-19 patients: a single-centre retrospective cohort study

Hospital surge capacity in a tertiary emergency referral centre during the COVID-19 outbreak in Italy

The microbial coinfection in COVID-19

Bacterial coinfection in influenza: a grand rounds review

Gut microbiota and Covid-19-possible link and implications

Reduction and Functional Exhaustion of T Cells in Patients With Coronavirus Disease 2019 (COVID-19)

The role of the lung microbiota and the gut-lung axis in respiratory infectious diseases

Immune Response and COVID-19: A mirror image of Sepsis

Predominance of hospital-acquired bloodstream infection in patients with Covid-19 pneumonia

Bloodstream infections in critically ill patients with COVID-19

Johns Hopkins Coronavirus Resource Center n

Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China

Potential association between COVID-19 mortality and health-care resource availability

Bacterial co-infection and secondary infection in patients with COVID-19: a living rapid review and meta-analysis

Co-infections in people with COVID-19: a systematic review and meta-analysis

Complications and outcomes of pandemic 2009 Influenza A (H1N1) virus infection in hospitalized adults: how do they differ from those in seasonal influenza?

novel coronavirus (2019-nCoV) outbreak: A new challenge

The role of pneumonia and secondary bacterial infection in fatal and serious outcomes of pandemic influenza a(H1N1)pdm09

The Role of Cytokines including Interleukin-6 in COVID-19 induced Pneumonia and Macrophage Activation Syndrome-Like Disease

Secondary Bacterial Infections Associated with Influenza Pandemics

Effectiveness of remdesivir in patients with COVID-19 under mechanical ventilation in an Italian ICU

Factors associated with hospital admission and critical illness among 5279 people with coronavirus disease

Dysregulation of Immune Response in Patients With Coronavirus 2019 (COVID-19) in Wuhan, China

Secondary infections in patients hospitalized with COVID-19: incidence and predictive factors

Outcomes in patients with severe COVID-19 disease treated with tocilizumab: a case-controlled study

Influenza and Bacterial Superinfection: Illuminating the Immunologic Mechanisms of Disease

The high volume of patients admitted during the SARS-CoV-2 pandemic has an independent harmful impact on in-hospital mortality from COVID-19

Bloodstream infections in critically ill patients: an expert statement

Antibacterial Therapy and Community-onset Bacterial Co-infection in Patients Hospitalized with COVID-19: A Multi-Hospital Cohort Study

Hospitalized patients with bacterial infections: a potential focus of SARS transmission during an outbreak

Risk Factors Associated With Acute Respiratory Distress Syndrome and Death in Patients With Coronavirus Disease

Cell-Type-Specific Immune Dysregulation in Severely Ill COVID-19 Patients

Alterations in Gut Microbiota of Patients With COVID-19 During Time of Hospitalization