The Global Prevalence of Daptomycin, Tigecycline, and Linezolid-Resistant Enterococcus faecalis and Enterococcus faecium Strains From Human Clinical Samples: A Systematic Review and Meta-Analysis

Background and Aim: The predominant species of the Enterococcus, Enterococcus faecalis (E. faecalis) and Enterococcus faecium (E. faecium) cause great variety of infections. Therefore, the expansion of antimicrobial resistance in the Enterococcus is one of the most important global concerns. This study was conducted to investigate the prevalence of resistance to linezolid, tigecycline, and daptomycin among enterococcal strains isolated from human clinical specimens worldwide. Methods: Several databases including Web of Science, EMBASE, and Medline (via PubMed), were carefully searched and reviewed for original research articles available in databases and published between 2000 and 2020. A total of 114 studies worldwide that address E. faecalis and E. faecium resistance to linezolid, tigecycline, and daptomycin were analyzed by STATA software. Results: The overall prevalence of antibiotic-resistant E. faecalis and E. faecium was reported to be 0.9 and 0.6%, respectively. E. faecalis and E. faecium were more resistant to the linezolid (2.2%) and daptomycin (9%), respectively. The prevalence of tigecyline-resistant E. facium (1%) strains was higher than E. faecalis strains (0.3%). Accordingly, the prevalence of linezolid-resistant E. faecalis was higher in Asia (2.8%), while linezolid-resistant E. faecium was higher in the America (3.4%). Regarding tigecycline-resistance, a higher prevalence of E. faecalis (0.4%) and E. faecium (3.9%) was reported in Europe. Conclusion: In conclusion, this meta-analysis shows that there is an emerging resistance in Enterococcus strains. Despite the rising resistance of enterococci to antibiotics, our results demonstrate that tigecycline, daptomycin, and linezolid can still be used for the treatment of enterococcal infections worldwide.


INTRODUCTION
Enterococci are known as opportunistic pathogens which are common inhabitants of the intestines of humans and animals (1,2). They are not only a significant component of the commensal microbiota (3), but also are able to cause a wide variety of serious infections such as bacteremia, endocarditis, intra-abdominal and pelvic infection, and urinary tract infection (UTI) (4-6). The predominant species belonging to the genus Enterococcus in clinical specimens are Enterococcus faecalis (E. faecalis) and Enterococcus faecium (E. faecium). These two species are currently considered as the second and third most important nosocomial pathogens in the world (7,8). Currently, expansion of antimicrobial resistance is one of the most important global concerns (9). Acquired resistance to several antimicrobial agents is more frequently observed in E. faecium than in E. faecalis. The World Health Organization (WHO) considered vancomycin-resistant E. faecium as a "high priority pathogen" urgently requiring new antibiotics for targeted treatment (5). E. faecium belongs to the "ESKAPE" pathogens, which includes six bacteria with growing multidrug resistance and virulence: E. faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp. These bacteria are responsible for majority of nosocomial infections and they can escape from the biocidal action of antimicrobial agents (10,11). Resistance is associated with increased morbidity and mortality rates (12,13). Vancomycin was considered as one of the last lines of treatment against multidrug-resistant bacteria. However, the emergence of Vancomycin-Resistant Enterococci (VRE) among hospitalized patients who had received long lasting courses of antibiotic treatment, posed a serious threat for other patients and health care professionals (14)(15)(16)(17). More than 40% of E. faecium bloodstream isolates was resistant to vancomycin, and the prevalence of VRE was different in various countries (9-12.5%) (8,18). Therefore, the emergence of VRE has prompted the use of novel and modified therapeutic agents including linezolid, daptomycin, and tigecycline, although resistance to those agents has already been reported in clinical settings (19,20). Linezolid is the first US FDA approved oxazolidinone antibiotic and it has great therapeutic efficacy for severe infections caused by multidrug-resistant Gram-positive organisms including VRE (21)(22)(23). Mechanisms of linezolid resistance in Enterococcus spp. is often due to mutations in the 23S ribosomal RNA (rRNA) genes and ribosomal protein-coding regulatory genes such as rplC, rplD, and rplV, mutations leading to amino acid substitutions in several ribosomal proteins [L3, L4 and/or L22], and the acquisition of more generic resistance genes such as cfr, cfr(B), poxtA/optrA (24)(25)(26). In 2003, the FDA approved the therapeutic use of the bactericidal lipopeptide antibiotic daptomycin against complicated skin and soft tissue infections arising from a broad spectrum of Gram-positive bacteria, including VRE (27,28). Emergence of daptomycin non-susceptibility in enterococci is associated with mutations in several genes including the stresssensing response system YycFGHIJ and LiaFSR and alterations in phospholipid biosynthesis enzymes such as cardiolipin synthetase cls and glycerophosphoryl diester phosphodiesterase gdpD (29)(30)(31). A third antibiotic that was shown to be beneficial against VRE infection is the bacteriostatic tigecycline that blocks bacterial protein synthesis at the elongation stage (32). Mutations in various efflux pumps is the main mechanism that associated with tigecycline-resistance in the enterocooci. Other resistancerelated mechanisms are deletions in ribosomal protein gene rpsJ and elimination of transcriptional regulation of the ribosomal protection protein (33,34). In conclusion, several new treatment modalities for enterococcal infections have been introduced over the past decades and now is a good moment to define the emergence of resistance to these relatively novel agents used. In the present study, a systematic review and metaanalysis was conducted to define the current prevalence of resistance to linezolid, daptomycin, and tigecycline among E. faecalis and E. faecium strains isolated from human clinical specimens worldwide.

Search Strategy and Selection Criteria
We reviewed original research articles available in databases and published between 2000 and 2020. These databases include Medline (via PubMed), Embase, and Web of Science. We searched the databases on December 2020. Enterococcus faecalis or E. faecalis OR Enterococcus faecium or E. faecium and linezolid and daptomycin and tigecycline were the keywords used in our search strategy. The searches in this study selectively included articles published in the field of epidemiology of E. faecalis and E. faecalis strains isolated from human specimens and targeted to define the prevalence of Antibiotic-Resistant Enterococci (ARE), Linezolid-Resistant E. faecalis (LREF), Tigecycline-Resistant E. faecalis (TREF), Daptomycin-Resistant E. faecalis (DREF), Linezolid-Resistant E. faecium (LREFA), Tigecycline-Resistant E. faecium (TREFA), and Daptomycin-Resistant E. faecium (DREFA). It is worth noting that the latest version of the CLSI guideline (2020) stated that linezolid-resistance in Enterococcus spp. as determined by disk diffusion (DD) must be confirmed by an MIC-generating method. As a result, studies that have used only the DD method to determine susceptibility to linezolid were excluded from the present study. Other excluded studies were review articles, case reports, and publications on basic research of resistance mechanisms for the mentioned antibiotics. Still, the bibliographies of excluded literature were explored to recognize further studies.

Inclusion and Exclusion Criteria
All studies on human clinical samples with complete information about the prevalence of ARE were evaluated. These data included prevalence or frequency of infection by Enterococcus spp. (E. faecalis and E. faecium), the country of origin, and the resistance assessment methods applied. The information presented in each study was evaluated using titles, abstracts, and, ultimately, the full text. Studies that qualified include original articles that provide sufficient information on the prevalence of ARE isolated from human specimens. Studies that used validated molecular techniques to diagnose antibiotic-resistant enterococci and presented data regarding the number of enrolled patients were all included. Excluded studies were those performed on non-human cases, studies that investigated other species of Enterococcus spp., covering other types of antibiotics, meta-analysis, systematic reviews and review articles, congress abstracts, and duplicate publications for the same investigation. Two authors separately reviewed inclusion and exclusion criteria and jointly selected appropriate articles.

Data Extraction and Definitions
The following data were extracted from the studies that met the inclusion criteria: the first author's last name, study period, year of publication, country, numbers of antibiotic-resistant E. faecalis and E. faecium strains, source of samples, and detection techniques (including genotypic identification methods applied). In order to do more accurate data extraction, this was done by two independent individuals and eventually confirmed by another researcher. In order to reach consensus, the reviewers finally had a closing discussion.

Quality Assessment
For all the studies which were entered in our searches based on the desired keywords, a quality evaluation (designed by the Joanna Briggs Institute) has been performed and only the studies with high-quality evaluation were selected for the final analysis (35).

Meta-Analysis
STATA (version 14.0) software was used for data analysis. Two models were used to pool the obtained data: fixed effects model (FEM) (36) and a random effects model (REM) (37). Various statistical methods were used to assess statistical heterogeneity, and then the heterogeneity was evaluated by Qtest and I2 statistical methods (38). P-value < 0.1 was regarded as statistically significant (38).

Characteristics of Included Studies
Among 4,198 articles, which were selected after an initial review of electronic resources and databases, 2,947 duplicate articles were excluded and 1,251 unique articles remained. After the assessment of the title and abstract, 895 articles were excluded and 356 articles remained of which 242 were excluded upon full-text search (see Figure 1 for more detailed explanation). Finally, 114 articles met the inclusion criteria of this study and were selected for final statistical analysis. Out of these 114 studies, 40 articles (35.08%) were from the Asian continent, 28 Europe (24.56%), 45 from America (39.47%), and 1 (0.87%) from Africa. There were no studies from Oceania. Most of the articles reviewed in this study were from USA (40 articles) and China (25 articles), respectively. In Supplementary Tables 1-3, the main features of the 114 selected articles are summarized. Most of the studies included in this study were published between 2012 and 2020 (more than 50%).

DISCUSSION
VRE are considered as an important nosocomial pathogen among the hospitalized patients and they are known to cause life threatening infections in humans. Various factors are predictive of VRE infections. These include long-term hospitalization, extensive use of antibiotics, and increased occupancy rate especially with malignancy in the intensive care unit (ICU) (39,40). Surviving under harsh conditions and outstanding adaptation to environmental conditions render enterococci into a significant reservoir for the transmission and spread of antibiotic-resistance determinants (41). Higher mortality rates, extended length of hospital stay, and higher treatment costs are the consequences of VRE bacteremia (17). Furthermore, to limit mortality, early administration of appropriate antibiotic therapy is essential (42). According to studies, most Enterococcus spp. isolates from human infections belong to E. faecalis (43). In our systematic study and meta-analysis, E. faecalis (47.6%) was identified to be the most dominant species than E. faecium (40.6%). The rate of antibiotic-resistance was higher in E. faecalis. While daptomycin is more active against E. faecalis, linezolid and tigecycline can be used at low concentrations for both E. faecalis and E. faecium (44). Daptomycin has received FDA approval against complicated skin and softtissue infections due to VRE and MRSA in 2003 (45). The results of the present study show that daptomycin had the best remaining inhibitory effect on E. faecalis with a resistance rate of 0.1%, which is in accordance with previous studies showed that most enterococcal isolates (>99.8%) are susceptible to daptomycin on a worldwide scale. As mentioned above the potential reason is its higher activity levels against E. faecalis. It seems that prior but insufficient daptomycin exposure can be related to the occurrence of daptomycin-resistance among enterococci (8). Furthermore, statistical analysis demonstrated that DREFA were mostly isolated from bloodstream infections (BSI) in the United States. Despite of lack of FDA confirmation, daptomycin is used frequently for VRE-BSI. The results of previous studies showed that linezolid and daptomycin have similar results for the treatment of VRE-BSI. Since linezolid is bacteriostatic, and daptomycin is bactericidal, it is assumed that in the immunocompromised patients, daptomycin, achieves superior clinical and microbiologic response rates (42). However, spontaneous resistance to daptomycin seems to arise rarely (28,(46)(47)(48)(49). Previous studies have reported that pulmonary surfactants have inhibitory effect on daptomycin, preventing this antibiotic to be effective on BSI following the lower respiratory tract infection and pneumonia (50). Furthermore, linezolid may provide a better therapeutic effect against BSI (51). Hence, daptomycin MICs should be screened in more detail to prevent treatment failure and the emergence of resistance. Studies showed in vitro synergistic interactions between daptomycin Because the countries that reported the prevalence of TREF strains were not large enough to be sub grouped, the prevalence of these strains in the countries was marked with *.
and various β-lactams in susceptible and non-susceptible enterococcal strains (52)(53)(54). This finding can be because of increased susceptibility of bacteria or enhanced surface binding by daptomycin and β-lactams (52). Linezolid is one of the last lines of defense to treat skin and lower respiratory tract infections. It inhibits bacterial growth by suppressing protein synthesis in bacteria (55 (42,56). The Infectious Diseases Society of America (IDSA) recommended linezolid for VRE intravascular catheter-related bacteremia but significant side effects such as myelotoxicity led to its limited use, especially among immunocompromised patients (57). The multidrug-resistance gene cfr is one of the linezolidresistance mechanisms in enterococci (8). This plasmid-defined mechanism can be occurring even across bacterial species and genera, as it is similar in Staphylococcus spp. and Enterococcus spp. This gene was first isolated from Staphylococcus sciuri of animal origin. Among enterococci it was first reported from an E. faecalis strain of the animal origin that can lead to resistance to at least five classes of antibiotics including linezolid (8). Tigecycline, a bacteriostatic agent which has a broad spectrum therapeutic effect against MDR Gram-positive bacteria including VRE and MRSA in addition to β-lactamase-producing Enterobacterales and anaerobes (58). Tigecycline is a potential treatment choice for complicated soft tissue and intra-abdominal infection, conversely, due to insufficient serum concentration, the use of it is restricted to bloodstream infections (59,60).  (62). Moreover, most of the studies reported <1% tigecycline-resistance and that depended on the various geographical regions and the analyzed species (8). Compared to our previous study (63), the percentage of resistance to linezolid, tigecycline and daptomycin is higher in Enterococcus spp. than in Staphylococcus spp. in the world. It should be noted that a higher level of resistance to the mentioned antibiotics in some parts of the world may not imply a higher resistance to these antibiotics in these areas and may be consistent with regular microbial susceptibility testing programs or the number of studies and the studied isolates which was carried out in these countries. Therefore, by performing regular surveillance programs, the accurate prevalence of antibiotic-resistance can be determined effectively. It should be noted that there are several limitations to our study. First, only published full-text research articles were evaluated in our study. Second, only the studies on clinical enterococci isolates were evaluated and other studies on environmental samples were excluded. Third, since there is insufficient information from many countries, we were not able to provide a truly global representation. Fourth, the failure to differentiate the clinical samples, which eventually did not conclude the prevalence of Enterococcus in various infections. Fifth, as some countries do not systematically monitor resistant E. faecalis and E. faecium, the number of reported antibiotic-resistant enterococci isolated from clinical specimens may not be realistic and it can be less than the actual amount. In some cases, bacteria may colonize in the patient's body without clinical signs. Also, failure to comply with the guidelines on the isolation and identification of bacteria resistant to antimicrobial agents can lead to unrealistic results. Sixth, studies indicate that infections caused by antibiotic-resistance Enterococcus spp. may be associated with mortality. Seventh, we did not distinguish according to the technology used for resistance assessments. Different methods can lead to different results and in this study; we simply merged all data obtained by various methods. Despite the importance of this issue, studies on the impact of antibioticresistance Enterococcus on mortality are rare and the necessity of such studies is clear today.

CONCLUSION
The present study reveals higher rates of resistance to daptomycin and tigecycline among E. faecium strains, whilst resistance to linezolid was higher in E. faecalis. By the way, our results show tigecycline, linezolid, and daptomycin still remain active against enterococcal isolates and can be used for the treatment of enterococcal infections worldwide. Obviously, monitoring of the rising resistance of VRE to these agents, appropriate antibiotic-resistance testing programs, and adequate antibiotic stewardship are extremely important in the successful reduction of resistance to the mentioned antibiotics, especially in VRE isolates.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.