**1. Introduction**

COVID-19, a pulmonary disease from an infection with the single-stranded RNA virus SARS-CoV-2, has evolved into a global pandemic since March 2020. Clinical manifestation is highly variable. Asymptomatic courses, mild respiratory diseases, severe pneumonia, and severe organ dysfunction that can be accompanied by shock and death have been

**Citation:** de Hesselle, M.L.;

Borgmann, S.; Rieg, S.; Vehreshild, J.J.; Spinner, C.D.; Koll, C.E.M.; Hower, M.; Stecher, M.; Ebert, D.; Hanses, F.; et al. Invasiveness of Ventilation Therapy Is Associated to Prevalence of Secondary Bacterial and Fungal Infections in Critically Ill COVID-19 Patients. *J. Clin. Med.* **2022**, *11*, 5239. https://doi.org/10.3390/ jcm11175239

Academic Editors: Luca Brazzi and Giorgia Montrucchio

Received: 5 August 2022 Accepted: 2 September 2022 Published: 5 September 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

81

described [1–6]. A certain proportion of patients develop an increased respiratory rate (>30/min), a decrease in oxygen saturation with hypoxemia, and respiratory insufficiency, which requires intensive care, usually due to dyspnea [3–7].

During the course of the pandemic, specific therapies for COVID-19 have been developed. However, the corresponding drugs should be applied within the first days after infection [8]. Relative risk reduction with respect to hospitalization or adverse outcome by the administration of antivirals or neutralizing monoclonal antibodies was described with the initiation of therapy at 3 to a maximum of 6 days after symptom onset [8]. Therefore, in an intensive care unit (ICU), only supportive treatment options are available to alleviate symptoms. In severe respiratory failure, intubation and invasive ventilation is the standard therapy in clinical practice [9]. It is a life-saving measure and usually ensures a safe airway and sufficient oxygenation, along with carbon dioxide elimination [9]. Early intubation counteracts the progressive deterioration of lung function due to increased respiratory stress [4,6]. It has also been reported that the critical delay of intubation in the event of failure of non-invasive ventilation options is associated with a poorer prognosis [4,6]. However, invasive ventilation may be the cause of ventilator-associated lung injury [6,7,9]. In addition, the safe airway required for invasive ventilation can promote serious, even lethal, infections [10]. The scientific literature, therefore, also contains reports recommending the avoidance of intubation as long as it is not essential [11].

Patients with viral infections are known to be predisposed to secondary infections [12–16]. In particular, bacteria may benefit from viral infections, and even those that are normally harmless could turn into opportunistic pathogens. The viral facilitation of bacterial pathogenesis is based on complex and multifactorial processes that, ultimately, promote bacterial adherence, disrupt epithelial layers, lead to the displacement of commensal bacteria, and subvert the host immune response [16]. There are multiple reports associating SARS-CoV-2 with co-infections of, primarily, bacterial but also fungal origin. The most common bacterial microorganisms in respiratory cultures from COVID-19 patients are *Pseudomonas aeruginosa*, *Klebsiella* species, *Staphylococcus aureus*, *Escherichia coli*, and *Stenotrophomonas maltophilia* [13,14]. The main fungal pathogens identified are *Aspergillus* and *Candida* species, but there are also reports of secondary infections with *Mucormycetes*, *Histoplasma* spp, *Cryptococcus* spp, and *Pneumocystis jirovecii* [12]. Alarmingly, such secondary infections have been linked to a severe clinical course with possible poor outcome [15,16].

Infections are a common problem in ICUs. A critical condition, an impaired immune response, and invasive treatments (i.e., mechanical ventilation and catheterization) all pose risk factors for nosocomial infections [10,17–21]. It is of concern that secondary infections in viral diseases of the respiratory tract, such as influenza, have been described as causes of morbidity and mortality [22–25]. However, the prevalence and clinical impact of healthcareassociated infections of bacterial or fungal nature in COVID-19 patients treated in ICUs is not well-understood and constitutes a serious knowledge gap. There is also insufficient knowledge on whether bacterial colonialization present on admission impacts disease severity and outcome. More data on community-acquired colonializations, as well as nosocomial infections in ICUs, are needed to optimize the management and treatment of the most severe COVID-19 cases. This could not only help to save lives but also to improve antimicrobial stewardship [26–29].

The aim of the present study is to unravel the prevalence of community-acquired colonializations with multidrug-resistant bacteria, as well as healthcare-associated secondary bacterial and fungal infections, in critically ill COVID-19 patients treated at an ICU. The primary objective was to determine whether (i) there is an association between a patient's infection status and the ventilation therapy used and whether (ii) co-infections are related to mortality. The secondary objectives are to examine the frequency of use and the clinical benefit of antimicrobial therapy in critically ill COVID-19 patients.

#### **2. Materials and Methods**

#### *2.1. Patient Cohort*

This study analyzed patient data from the Lean European Open Survey on SARS-CoV-2-Infected Patients (LEOSS) cohort [30]. The LEOSS project represents a non-interventional, multicenter network that aims at addressing the lack of in-depth knowledge on the epidemiology and clinical course of COVID-19. Established in March 2020, the LEOSS registry encloses data mainly on hospitalized COVID-19 patients. In the LEOSS protocol, patients can be included via PCR confirmed diagnosis or rapid antigen tests as an acceptable alternative. Detailed information on LEOSS can be found on the project's website (https://leoss.net, accessed date: 5 August 2022). The study was registered at the German Clinical Trials Register (DRKS, No S00021145).

Clinical data are reported in an electronic case report form (eCRF) using the online platform ClinicalSurveys.net, which was developed by the University Hospital of Cologne (UHC), Germany, and is hosted by QuestBack, Oslo, Norway, on servers of the UHC [31]. Anonymized patient data are added to the LEOSS registry retrospectively at the end of the acute treatment setting, i.e., when either the treatment is completed or the patient has died. In order to ensure anonymity in all steps of the analysis process, an individual LEOSS Scientific Use File (SUF) was created, which is based on the LEOSS Public Use File (PUF) principles described in Jakob et al. [31]. Re-identification is prevented by vertical (categorical assessment of numerical variables) and horizontal data aggregation (data aggregation within the phases of disease). Categorization is based on four phases, which can be roughly characterized as asymptomatic or mild symptoms (uncomplicated phase), need for oxygen supplementation (complicated phase), need for critical care (critical phase), and the recovery phase. A detailed description of the clinical phases as defined in the LEOSS registry, as well as of the recorded data items, can be found on the project's website (https://leoss.net; accessed on 4 August 2022) and in [32].

### *2.2. Study Design*

This analysis included data of 840 patients who were documented by a LEOSS partner site between 23 March 2020 and 12 October 2020 due to COVID-19 disease diagnosed and treated between February 2020 and October 2020. Only patients who reached the critical phase according to the definitions of the LEOSS database [32] during the course of their COVID-19 disease were included in the analysis. The onset of the critical phase was declared if at least one of the following criteria was present: need for catecholamines, life-threatening cardiac arrhythmia, need for unplanned mechanical ventilation (invasive or non-invasive), prolongation (>24 h) of planned mechanical ventilation, liver failure with Quick <50% or INR >3.5, a qSOFA score of ≥2, or acute renal failure with need of dialysis. Dedicated intensive care data items were developed by a working group of specialized intensive care physicians (LEOSS Intensive Care Group) and implemented in the LEOSS registry. From this set, the following data items were analyzed: (i) the colonialization status of the patients with regard to multidrug-resistant pathogens at baseline, i.e., day of positive SARS-CoV-2 diagnosis (multidrug-resistant, Gram-negative bacteria (3MRGN/4MRGN), methicillin-resistant *Staphylococcus aureus* (MRSA), and vancomycin-resistant enterococci (VRE)), as well as bacterial and fungal superinfections in the critical phase; (ii) the ventilation treatments performed (non-invasive ventilation, invasive ventilation, or extracorporeal membrane oxygenation (ECMO)); (iii) the medications used; and (iv) the outcome (recovery or death). 3MRGN and 4MRGN are enterobacteriaceae, *Pseudomonas aeruginosa*, and *Acinetobacter baumannii* exhibiting resistance to three or four of these antibiotics or antibiotic groups: piperacillin, carbapenems, quinolones, and cephalosporins of the third generation. Two endpoints were defined: (i) the prevalence of community-acquired colonializations and healthcare-associated secondary infections in patients in need of or receiving a specific ventilation therapy (non-invasive ventilation, invasive ventilation, or ECMO) and (ii) the effect of community-acquired colonializations and healthcare-associated secondary infections on patient outcome.

### *2.3. Statistical Analysis*

All data were presented as categorical variables (numbers and percentages). To compare categorical variables, Pearson's chi-squared or Fisher's exact test was used where appropriate. The level of significance was set at *p* < 0.05. The data management, statistical analysis, and computation of figures were conducted using R (R Development Core Team, Vienna, Austria, Version 4.1.1, 2021).

### **3. Results**
