*2.2. Context*

During the study period, several infection control programs were active in the ICUs involved, with specific leadership and scope. Surveillance cultures (tracheal aspirate, rectal swab, urinary culture) were performed weekly; universal screening for carbapenemproducing *Enterobacterales* (CPE) and *A. baumannii* using rectal swabs was performed upon admission to the ICU and then once a week. In mechanically ventilated patients, the surveillance of respiratory samples (tracheal aspirates or bronchoalveolar lavage) was also performed at least once a week, with some differences between the different centers. Blood cultures or bronchoalveolar lavage cultures were performed on clinical decision.

#### *2.3. Definitions*

Pneumonia by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) was defined based on real-time polymerase chain reaction (RT-PCR) on at least one low respiratory tract specimen [28].

The occurrence of colonization and/or infection with *A. baumannii* was assessed from the date of ICU admission to ICU discharge. It was considered only once at the time of the first incidence of a positive sample. Colonization was defined as bacterial isolation without clinical signs or symptoms suggestive of infection. Infection was defined according to the Centers for Disease Control and Prevention (CDC) criteria [29]. Carbapenem resistance was defined according to the EUCAST criteria [30].

All episodes of VAP and/or BSI, as well as the development of septic shock with the requirement of vasoactive drugs [31], were registered according to the European Centre for Disease Prevention and Control (ECDC) current definitions [32].

#### *2.4. Microbiology*

*A. baumannii* and CPE strains from blood, respiratory, and rectal samples were collected in accordance with active surveillance screening and following local guidelines. Rectal swabs were collected from hospitalized patients and screened for CPE by combining culture-based detection and the identification of carbapenemase type.

We identified CR-Ab according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) criteria of carbapenem resistance. Cultures were analyzed with the BD BACTECTM FX system (Becton Dickinson) according to EUCAST breakpoint tables. The identification of microorganisms was conducted with mass spectrometry MALDI-TOF (Matrix-Assisted Laser Desorption Ionization Time-of-Flight) and VITEK®, whereas susceptibility to antibiotic molecules was tested using VITEK 2 (VITEK® according to EUCAST breakpoint tables). The whole-genome sequencing of CR-Ab isolates collected from blood cultures and respiratory samples was not available in the pandemic context. The clonal relationship of CR-Ab isolates was currently not investigated.

#### *2.5. Statistical Analysis*

Data were entered and analyzed using SPSS version 27. Statistical significance was defined as less than 0.05. Descriptive analysis was reported as frequencies, percentages, means, and standard deviations. Categorical variables, demographics, and clinical characteristics were compared against mortality using the Chi-squared test. Continuous variables were tested for normality by the Kolmogorov–Smirnov test. Non-normally distributed variables were evaluated using the Mann–Whitney test.

Significant values in the univariate analysis were evaluated with a multivariate model: a logistic regression model for mortality to assess independent predictors. The odds ratio was reported with corresponding 95% confidence intervals.

#### **3. Results**

Sixteen ICUs joined the data collection. Four of them had no cases of CR-Ab in COVID-19 patients. The first data collection was completed in May 2021. The review of the data by independent investigators was completed in the months of June–September 2021.

During the study period, 913 COVID-19 patients were admitted to the participating ICUs. Of them, 19% became positive for CR-Ab, either colonization or infection (*n* = 176). The ICU mortality rate in patients with *A. baumannii* was as high as 64.7% (*n* = 112) (Table 1).



**Table 1.** *Cont.*


List of abbreviations: intensive care unit, ICU; carbapenem-resistant *Acinetobacter baumannii*, CR-Ab; number, *n*; Body Mass Index, BMI; extracorporeal membrane oxygenation, ECMO; Simplified Acute Physiology Score, SAPS; Sequential Organ Failure Assessment, SOFA; Acute Respiratory Distress Syndrome, ARDS; Ventilator-Associated Pneumonia, VAP; Bloodstream infection, BSI; *K.pneumoniae* producing KPC; methicillin-resistant *Staphylococcus aureus*, MRSA; vancomycin-resistant *Enterococcus*, VRE. bold was used for *p* < 0.05.

The majority of patients were males (136; 78.6%), with a median age of 65 ± 10.3 years. The average Simplified Acute Physiology Score (SAPS) II and Sequential Organ Failure Assessment (SOFA) scores were 42 ± 13.37 and 8.3 ± 3.7, respectively. Leading comorbidities were cardiovascular diseases (118 patients, 67%), obesity (52 patients, 29.5%), diabetes (39 patient, 22.1%), and chronic pulmonary disease (22 patients, 12.5%). Around 31% of patients were transferred from one hospital to another; 93.2% of them presented acute respiratory distress syndrome upon ICU admission. The mean length of stay in the ICU was 24 ± 18 days. On average, patients developed colonization or infection within 10 ± 8.4 days from ICU admission.

The scores of SAPS II and SOFA were significantly higher in the deceased patients (43.8 ± 13.5, *p* = 0.006 and 9.5 ± 3.6, *p* < 0.001, respectively). Furthermore, the mortality rate was significantly higher in patients with extracorporeal membrane oxygenation (ECMO; 12; 7%, *p* = 0.03), septic shock (61; 35%, *p* < 0.001), and in elders (66 ± 10, *p* < 0.001) (Table 1).

Among the 176 patients enrolled in the study, 129 (73%) had invasive infection with CR-Ab, distributed as follows: 105 (60.7%) VAP and 46 (26.6%) BSI. In 22 cases (6.5%), VAP was associated with concomitant BSI. Colonization was reported in 165 patients (93.7%). Of note, 118 patients previously colonized by CR-Ab developed invasive infections, while 11 patients developed infection without any known previous colonization. Mortality was significantly higher in patients with VAP (*p* = 0.009). Colonized patients who did not develop invasive infections had a higher survival rate (*p* < 0.001; Table 1). Being colonized by CR-Ab was associated with a higher risk of developing invasive infections (*p* < 0.001).

Co-infections with carbapenem-resistant Klebsiella pneumoniae and enteric pathogens were seen in 29 (17%) and 55 (32%) patients, respectively.

Most of the CR-Ab isolates (159, 90.3%) were sensitive to colistin. Colistin was used to treat the majority (100, 56.8%) of the patients. Most commonly, it was administered with meropenem, ampicillin-sulbactam, and rifampicin (21%, 18.1%, and 17%), respectively. However, no difference in mortality rate was observed between different therapies.

In the multivariate analysis (Table 2), risk factors that were significantly associated with mortality were age (OR = 1.070; 95% CI (1.028–1.115) *p* = 0.001) and CR-Ab colonization (OR = 5.463 ic 96% 1.572–18.988 *p* = 0.008).


**Table 2.** Multivariate analysis for mortality.

List of abbreviations: Simplified Acute Physiology Score, SAPS; Ventilator-Associated Pneumonia, VAP; carbapenem-resistant *Acinetobacter baumannii*, CR-Ab. bold was used for *p* < 0.05.

#### **4. Discussion**

Bacterial and fungal superinfections represent a severity factor with a high impact on the morbidity and mortality of critically ill patients with COVID-19 [33,34]. This aspect is even more essential in countries burdened by a high rate of multidrug-resistant bacteria, such as Italy [35], where an increasing number of CR-Ab infections have been seen in the last years.

In the present multicentric study, conducted on 16 ICUs in the Piedmont region during the COVID-19 pandemic, it was found that 19% of ICU COVID-19 patients became positive for CR-Ab, either colonization or infection, during an ICU stay. Although the whole-genome sequencing of CR-Ab isolates was not available in the pandemic context and the clonal relationship of CR-Ab isolates was currently not investigated, this elevated percentage and some epidemiological factors deserve very high attention. Furthermore, the mortality rate in patients with CR-Ab was as high as 64.7%, significantly higher than the overall mortality in critically ill COVID-19 patients [36].

To the best of our knowledge, this is the first multicenter regional study reporting the impact of CR-Ab colonization and severe infection in ICUs during the COVID-19 pandemic. Interestingly, our analysis refers to the so-called Italian "second-wave" of the pandemic, when the global emergency scenario of the first months of the pandemic had extensively changed. A recent multicenter, cross-sectional study compared the rates of colonization and infection with carbapenemase-producing *Enterobacterales* (CPE) and/or CR-Ab in two study periods, pre and during the COVID-19 pandemic. No significant change in either incidence rate ratios and weekly trends in CPE colonization and infection was observed, while the incidence rate ratios of colonization and infection with CR-Ab increased by 7.5- and 5.5-fold, respectively, during the COVID-19 period. A clonal lineage was demonstrated and appointed for the occurrence of horizontal transmission [26].

Other authors previously highlighted that, during the first wave of the COVID-19 pandemic, several factors could have favored the emergence and spread of antimicrobialresistant bacteria in hospitals [25], such as the overload of hospitalized patients, especially in intensive care, favoring patient-to-patient transmission [37]; the initial overuse of antibiotics for suspected bacterial co/super-infections [38]; the possible delay in providing microbiological culture and sensitivities results due to the COVID-19 overload [39]. During the first months of the pandemic, in several countries, including Italy, a lack of appropriate protective personal equipment and health personnel hired on an emergency basis to respond to the COVID-19 pandemic, sometimes impeding adequate training in infection prevention and control, were common. However, that may not be completely true in the period of our study, when the first pandemic phase with its need for reorganization was already over.

Other factors may have contributed to the described spread of CR-Ab infections.

First of all, the need for the referral of critically ill patients (e.g., requiring ECMO [40]) and the high number of patients transferred from one hospital to another may have facilitated the dissemination of cases at the regional level. Even the structural characteristics of ICUs (new, re-opened, or already functioning before the COVID-19 pandemic) may also have played a role, in terms of spaces dedicated to patients and workstations, devices, and hospital pathways between departments (e.g., emergency department, radiology). In fact, CR-Ab cross-transmission between equipment (ventilators, infusion pumps, hemodialysis machines, ultrasound devices) and COVID-19 patients may also partly explain the onset of this outbreak.

Focusing on the identification and characterization of Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, *Acinetobacter baumannii*, Pseudomonas aeruginosa, and Enterobacter spp. (ESKAPE) bacteria and their possible clonal spread in medical devices, patients, and medical personnel in the ICU, a recent work [41] has shown that 91% of the analyzed sites were colonized by bacteria (pathogenic and commensal), where *S. aureus* and *A. baumannii* MDR showed a high incidence, and *A. baumannii* MDR showed a clonal distribution in surfaces, patients, and health personnel.

It is in fact known that even when there is the scrupulous protection of medical personnel to avoid the transmission of SARS-CoV-2 from patients to health personnel, the transmission of other pathogens such as ESKAPE bacteria is not automatically avoided. In a previous study in ICUs, it was shown that the bacterial recontamination of contact surfaces occurred after 4 h after standard cleaning with detergents with chlorine-releasing agents, isopropyl alcohol, and sodium hypochlorite [42]. Moreover, COVID-19 critically ill patients often require prolonged hospitalizations, and it is known that staying in an intensive care setting for a long time—as well as immunosuppression, the need for prolonged previous antibiotic therapies, and the invasiveness of care—are known risk factors for infections with multidrug-resistant pathogens.

In our analysis, the median ICU length of stay was high (24.27 ± 17.9 days), with a time lag before the development of colonization or the onset of invasive infection of 17.31 ± 13.3 days of hospital stay and 10.69 ± 8.4 days of ICU stay, respectively.

Some other factors must be taken into account in the analyzed population. Certainly, patient severity had an impact on mortality, with statistical significance for the need for ECMO support, higher SAPS and SOFA scores, and the presence of septic shock as infection presentation. Similar to other settings, the use of steroids might be related to a higher risk of developing MDR infection [43]. Concerning the impact of VAP in CR-Ab infected patients, the diagnosis of VAP may have been made difficult by the presence of the radiological and clinical signs of COVID-19 pneumonia, which made it even more difficult to apply the classical criteria and the consequent definition of VAP.

The presence of colonization preceding the infection represented, in our series, a risk factor with respect to mortality. It is well known from the literature that colonization does not require any "pre-emptive" therapy if the patient has no clinical signs of infection, but these data confirmed the finding that colonization remains one of the main risk factors for invasive infections and represent a "wake up call" regarding the frailty of our patients. Therefore, implementing an early pre-emptive therapy in cases of known colonization, at the time of clinical worsening, is one of the main steps to improve survival in this setting.

As previously reported in the literature, the role of combination therapy is widely debated in the absence of definitive evidence [44,45]. The data are insufficient for a more completed analysis, but the unmet need for new and effective therapies is of paramount importance considering the mortality of these critically ill patients.

The presence of multi-bacterial co-infections is a further interesting fact, able to describe not only the fragility of the patients but also the delicate hospital ecology and to reinforce the need for effective and strict control measures. In particular, the combination of various Gram-negative pathogens describes the context of our ICUs and may be the consequence of the high use of empiric broad-spectrum antibiotic therapies used in COVID patients not only at home but also in the early stages of hospitalization.

Our study has several limitations. First, the retrospective nature of the study and therapeutic management on the risk of *A. baumannii* infection. Secondly, the lack of data on the total number of COVID-19 ICU patients did not allow a comparison of risk factors and outcomes. Third, as the clonal relationship was not investigated, it is impossible to define the common origin of the burden of infections or a relationship, at least in the high number of referral patients. Moreover, it was not possible to obtain a cumulative antibiogram for antibiotic classes to show the overall sensibility of different strains. Finally, the local epidemiology and the need to re-organize the capacity, spaces, and staff of our ICUs during the pandemic could limit the generalizability of our results.

#### **5. Conclusions**

The need to not neglect antimicrobial stewardship principles during the COVID-19 pandemic has already been recently underlined [46], as well as the importance of enhancing infection control activities directed against antimicrobial resistance. In continuity with this message, our study remarks on the need to pursue antimicrobial stewardship principles during the COVID-19 pandemic, and infection control activities targeted against the spread of antimicrobial resistance inside and between hospitals.

During a pandemic, not only in the first phases, but especially later in the time course, infection control activities should be revised and eventually re-modulated according to the new organizational structures. Constant infection-control measures are necessary to stop the spread of *A. baumannii* in the hospital environment, prevent outbreaks, and lower mortality rates, especially at this time of the SARS-CoV-2 pandemic. Stricter barrier measures need to be implemented, increasing the effectiveness of screening and surveillance for *A. baumannii*, especially when resistant to carbapenems. The active surveillance culture and efficient performance of a multidisciplinary team will be highly important in detecting and controlling the CR-Ab outbreak in COVID-19 ICUs.

**Author Contributions:** Conceptualization, G.M., S.C., L.B. and F.G.D.R.; Data curation, G.M., T.L., N.S., C.O., M.P., A.P., D.C., A.R., S.B., F.R., V.B., F.D.C., S.G., A.D.S., E.R., N.B., M.C., P.C., G.B. (Giacomo Berta), C.C., B.S., M.G., M.P.G., G.B. (Gabriella Buono), F.F., S.E., P.F.S., G.F., A.C., S.L., D.S., F.A., M.B., M.N., S.V., E.C., M.M.L. and E.M.; Formal analysis, G.M., S.C., T.L., N.S. and F.G.D.R.; Investigation, G.M., S.C., T.L., C.O., M.P., A.P., D.C., A.R., S.B., F.R., V.B., F.D.C., S.G., A.D.S., E.R., N.B., M.C., P.C., G.B. (Giacomo Berta), C.C., B.S., M.G., M.P.G., G.B. (Gabriella Buono), F.F., S.E., P.F.S., G.F., A.C., S.L., D.S., F.A., M.B., M.N., S.V., E.C., M.M.L. and E.M.; Methodology, G.M., S.C., L.B. and F.G.D.R.; Project administration, L.B. and F.G.D.R.; Resources, L.B. and F.G.D.R.; Software, N.S.; Supervision, G.M., P.C., L.B. and F.G.D.R.; Validation, S.C. and P.C.; Writing—original draft, G.M., S.C., M.C. and M.G.; Writing—review & editing, T.L., N.S., C.O., M.P., A.P., D.C., A.R., S.B., F.R., V.B., F.D.C., S.G., A.D.S., E.R., N.B., P.C., G.B. (Giacomo Berta), C.C., B.S., M.P.G., G.B. (Gabriella Buono), F.F., S.E., P.F.S., G.F., A.C., S.L., D.S., F.A., M.B., M.N., S.V., E.C., M.M.L., E.M., L.B. and F.G.D.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Data acquisition and analysis were performed in accordance with the protocols approved by the local Ethics Committee (Ethics Committee: Comitato Etico Interaziendale A.O.U. Città della Salute e della Scienza di Torino—A.O. Ordine Mauriziano—A.S.L. Città di Torino; ethics approval number 0031285). The study was conducted according to the guidelines of the Declaration of Helsinki.

**Informed Consent Statement:** Written informed consent was waived according to Italian regulations due to the retrospective nature of this study.

**Data Availability Statement:** The datasets used and analyzed during the current network metaanalysis are available from the corresponding author upon reasonable request.

**Acknowledgments:** We would thank all the study collaborators in each ICU involved, and in particular: Rosario Urbino, Città della Salute e della Scienza Hospital, Torino; Marinella Zanierato, Città Della Salute e Della Scienza Hospital, Torino; Ilaria De Benedetto, Department of Medical Sciences, Infectious Diseases, University of Turin; Alessandro Bianchi, Ospedale Cardinal Massaia, Asti; Silvio Borrè, AO S.Andrea, Vercelli; Lucio Boglione, University of Eastern Piedmont, Italy; Serena Querio, Maggiore della Carità Hospital, Novara; Arianna Abascià, AO Ordine Mauriziano Torino; Valerio Del Bono, AO Santa Croce e Carle, Cuneo; Guido Chichino, AO SS. Arrigo e Biagio, Alessandria; and Chiara Scaletti, Ospedale di Rivoli, Italy.

**Conflicts of Interest:** The authors declare no conflict of interest regarding this manuscript.
