High-level Gentamicin Resistance among Clinical Isolates of Enterococci in Iran: a Systematic Review and Meta-analysis

Enterococci have been considered as one of the most common causes of nosocomial infections. The spread of antibiotic resistance has posed a serious challenge to treating the enterococcal infections. High-level aminoglycosides resistance leads to failure in the synergistic combination therapy. This study aimed to estimate the prevalence of high-level gentamicin resistance (HLGR) among clinical isolates of enterococci in Iran. Systematic literature search was conducted in the Web of Science, PubMed, Scopus and Google Scholar electronic databases from articles which were published from April 2000 to September 2018. Literature search yielded 918 studies. Eligible studies were selected according to the defined inclusion and exclusion criteria. Statistical heterogeneity was estimated by Q statistic and the I2 index. The Begg’s rank correlation test and Egger’s weighted regression tests were used to evaluate possible publication bias. Nineteen studies were included in this review. According to the meta-analysis results, the prevalence of HLGR among Enterococcus spp. was 49.4% (95% CI: 42.2%-56.6%). It was estimated 44.3% (95% CI: 38.1%-50.8%) and 66.3% (95% CI: 51.4%-78.6%) for E. faecalis and E. faecium, respectively. Since notable rate of HLGR in enterococci was seen in this analysis, improving the implementation of all aspects of the infection control programmes is required. Accurate and regular monitoring of infection control procedures are necessary for reducing the dissemination of such infections.


INTRODUCTION
Enterococci are common inhabitants of the human gastrointestinal tract and they are known to be potent pathogens responsible for several infections in humans. 1 Enterococci have been considered as one of the most common causes of nosocomial infections worldwide. 2 The spread of antibiotic resistance has posed a serious challenge to the treatment of the enterococcal infections. 3 Simultaneous prescription of aminoglycosides and cell wall active agents (such as β-lactams or vancomycin) will be synergistically effective against severe infections. 4 In 1979, high-level aminoglycosides resistance in enterococci was initially described in the US. 5 High-level resistance (HLR) to the aminoglycosides leads to failure in the synergistic combination therapy. 4,6 Moreover, horizontal HLR associated genes transfer is common between various strains and it can increase the risk of the development and dissemination of such resistance in enterococci. 7 According to available data, the prevalence rates of high-level gentamicin resistance (HLGR) in enterococci have been reported from several parts of Iran. 1,[8][9][10][11] However, there is no comprehensive analysis regarding HLGR amongst Enterococcus isolates, obtained from several infections in the country. Therefore, the aim of the present study was to assess the prevalence of HLGR among clinical isolates of enterococci in Iran using a systematic review and meta-analysis conforms to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Search strategies
A systematic literature search was conducted in the Web of Science, PubMed, Scopus, and Google Scholar electronic databases from papers that were published from April 2000 to September 2018 following the PRISMA (the Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. The following keywords, "Enterococcus" or "E. faecalis" or E. faecium and "high level aminoglycoside resistant" or "high level gentamicin resistant" or "HLGR" in combination with "Iran" were searched as scientific terms in the present survey. We also searched bibliographies of retrieved articles for additional references.

Eligibility criteria and study selection
Cross-sectional or cohort studies that reported the prevalence of HLGR in Iran were considered. To determine the articles which had inclusion criteria, the titles, abstracts and full texts were screened independently by two reviewers and any discrepancies were resolved by consensus. The articles published in English or Persian language which were indexed in PubMed or Scopus with the follo-wing characteristics were included: standard used methods for HLGR detection in Enterococcus strains and reported data on number of HLGR among Enterococcus strains. According to the Clinical and Laboratory Standards Institute (CLSI) guidelines 12 , standard methods for detecting HLGR among Enterococcus strains are broth dilution, agar dilution, and disk diffusion methods by high-content gentamicin (120 µg). Additionally, research that has been conducted by non-Iranian authors on the Iranian population or samples were also assessed. Studies with nonstandard methods and without report of HLGR prevalence were excluded. We also excluded studies which their sample size was less than 10 isolates, nonhuman studies, review articles, meta-analyses or systematic reviews, congress abstracts and duplicate publication of the same studies.

Data extraction and definitions
Data collection was performed in parallel by two authors and discrepancies were resolved by a third author. Data collected included: first author's name, the study performing time, publication date, the study setting, sample size (number of Enterococcus, E. faecalis and E. faecium isolates) and prevalence of HLGR.

Statistical analysis
Analysis of data was performed by Comprehensive Meta-Analysis Software v. 2.2 (Bio stat Company). Meta-analysis was performed using random effects model to estimate pooled prevalence and corresponding 95% confidence interval (CI). Statistical heterogeneity between and within groups was estimated by the Q statistic and the I 2 index. The Begg's rank correlation test and Egger's weighted regression tests were used to evaluate possible publication bias (p<0.05 was considered indicative of a statistically significant publication bias).

RESULTS
The literature search yielded 918 studies, 856 were excluded based on their index and review of title and abstract, leaving 62 articles for full text review. Of 62 reviewed studies, HLGR was not detected in 22 studies according to the standard methods, 12 studies did not report the prevalence of Enterococcus or Enterococcus spp., 4 studies had sample size less than 10 isolates, and the results of 5 studies were unclear. Eventually, 19 eligible studies were chosen for final analysis. There were no overlapping study populations in the final review. Fig. 1 shows the study selection process and reasons for exclusion. None of the included studies received direct funding from pharmaceutical companies. The full results of included articles, the prevalence of E. faecalis and E. faecium and the frequency of HLGR amongst them are presented in Table 1  • Studies that had sample size less than 10 isolates (n = 4) • Studies that did not report the prevalence of Enterococcus spp. (n = 12)
Prevalence of resistant enterococcal strains amongst patients may be due to several factors; recently, various studies described the effects of diet on gut microbiota composition. Special diet such as fiber (as a non-digestible carbohydrate) consumption change Enterococcus species population in gastrointestinal tract. 40,41 Also, drug-resistant Enterococcus isolates of animal origin can easily contaminate the food chain and infect humans and affect the intestinal colonization. 42,43 Hospitalization of colonized patients by such strains may lead to dissemination of drug-resistant strains in hospital environment and among patients. Therefore, diverse ranges of high-level gentamicin resistant enterococci isolated from hospitalized patients in various countries may be related to their diverse dietary habits.
Enterococci are able to survive under environmental stresses such as dryness and selective antimicrobial pressure. Therefore, hospital setting has an important role in the spread of resistant strains among patients and high-quality disinfection and cleaning of all surfaces and medical equipment are necessary. [44][45][46] Inadequate efficacy of utilized disinfectant due to reduced biocide sensitivity in enterococci may be one of the major issues in the infection control implementation in hospitals.
In addition, unfavourable condition may induce viable-but-nonculturable (VBNC) strains. 44 This phenomenon can lead to transmission of resistant strains among patients, despite the effective appearance of the infection control policies.
Emergence of drug-resistant bacteria has accelerated by overuse and misuse of antibiotics. 47,48 Ineffective ami- noglycosides prescription may contribute to selection of high-level gentamicin resistant strains. Colonization and dissemination of selected strains in hospital enhance the prevalence of HLGR among hospitalized patients.
HLGR in enterococci often occurs through acquisition of aph(2'')-Ia-aac(6')-Ie gene. 7 This bifunctional gene is most commonly located on a transposon. Therefore, horizontal gene exchange and acquired HLR traits will occur quickly among colonized Enterococcus spp. Other HLGR encoding genes are most frequently in E. faecium. 7 It can explain the greater prevalence of HLGR among E. faecium strains.
Considering the mentioned features of enterococci, it seems that the most effective strategy among the many possible strategies for reducing the transmission of healthcare-associated infections, is hand hygiene compliance amongst healthcare workers. As mentioned in a previous systematic review in Iran, training of healthcare workers seems to be necessary for prevention and control of nosocomial infections. 49 There are some limitations to the present review. First, only the published articles which were indexed in PubMed or Scopus were included in this meta-analysis. Secondly, few studies had been performed in different region of the country and most of the evaluated articles were related to Tehran (the capital of Iran).

CONCLUSIONS
The present systematic review indicates notable prevalence of high-level resistance to gentamicin amongst Enterococcus species in Iranian patients. According to the papers published from April 2000 to September 2018, diverse ranges were reported from Iran. This comprehensive analysis showed that, the prevalence of high-level gentamicin resistant Enterococcus spp. was 49.4%. Also, the prevalence of HLGR in E. faecium strains was greater than that in E. faecalis. Therefore, improving the implementation of all aspects of the infection control programs according to the international standards is necessary.