Comparative Analysis of Prevalence and Antibiotic Resistance in Vancomycin-Resistant Enterococcus from Clinical Samples – Demographics and Phenotypes

Background: Of all enterococci species, the most renowned clinically as multidrug-resistant pathogens are Enterococcus faecium and Enterococcus faecalis. Vancomycin-resistant Enterococcus (VRE) species are the principal cause of opportunistic hospital-acquired infections, due to numerous resistance mechanisms. Methods: In this study, the prevalence and antibiotic resistance profiles of VRE according to clinical sources from three selected hospitals in Southwest-Nigeria were investigated. Altogether, 431 samples (urine, rectal, and wound swabs caesarian section (CS), automobile accidents, and other skin lesions and abrasions) were collected from three selected hospitals in Osun State, Nigeria. Established techniques were employed for the recovery of enterococci and screening for VRE while antibiotic susceptibility tests were carried out by disc diffusion technique. Results: Altogether, 208 (48.3%) enterococci strains were recovered from which 85 (40.9%) were VRE. E. faecium predominated at 71.8% (61/85) and E. faecalis at 28.2% (24/85) as determined by phenotypic characterization. VRE isolates exhibited 100%, 97.6%, and 92.9% resistance to ampicillin, clindamycin, and quinupristin-dalfopristin (Q/D) respectively. The least resistance in-vitro was to tigecycline (27.1%). None of the antibiotics exhibited 100% activity against all the isolates. vanA resistant phenotype was prevalent at 65.9%. E. faecium from all study locations displayed higher levels of resistance than E. faecalis. Multiple antibiotic resistance (MAR) indices in all VRE isolates were ≥0.2, all being multidrug-resistant. Conclusions: The high prevalence rate along with the high level of multidrug resistance observed in the present study is worrisome and poses a continuous threat in the therapy of illnesses triggered by VRE as vancomycin was perceived as a drug of choice to curb enterococcal infections.

described in detail in the specie E. faecium (9). Several risk factors exist for colonization by and consequent infection with vancomycin-resistant Enterococci. Predominant among these is prior exposure to antimicrobials (10), which possibly leads to bowel flora modification. Besides, patients with advancing age, immunocompromised states, and those with serious underlying illnesses -such as patients in institutions for long-term care, on prolonged antibiotics therapy, extended hospital stays, as well as proximity to other patients colonized with VRE -are at increased risk (11).
The part of enterococci in clinical infections has been insufficiently studied and reported in Nigeria. Previous reports have hinted that the development of resistance to glycopeptides among enterococci did not occur (12); but later, another study on the prevalence of VRE reported the detection of VRE from hospital samples, and hands of health care staff in Southwest Nigeria (13,14). Enterococcal isolates resistant to vancomycin are not habitually screened for in many clinical laboratories in Nigeria perhaps because of the perceived low incidence of resistance to vancomycin among enterococcal isolates, thereby obscuring the detection of vancomycin-resistance in enterococcal strains. Our study was therefore undertaken to evaluate the prevalence and antibiotic sensitivity pattern of vancomycin-resistant enterococci from clinical samples from three selected hospitals in Southwest Nigeria in order to influence treatment choices of infections triggered by VRE in the hospital environment.

Study Locations
This study was carried out at three designated hospitals -State Specialist Hospital, Osogbo (7.

Sample Collection and Processing
This was a cross-sectional study of urine, rectal swabs, and wound samples. The minimum sample size for the study (384 participants) was calculated with the formula n = z 2 pq/d 2 where n is the least sample size mandatory, z is the centile of the standard normal distribution fixed at 1.96, p is the most probable prevalence rate of selected indicators (and in this study 50%), q is (1-p) and d is the degree of accuracy (at 0.05). Altogether, 431 participants were included in the study based on individual or parental consent for addition to the study.
The wound samples were obtained from wounds of caesarian sections (CS), trauma from automobile accidents, and other skin lesions and abrasions obtained from both out-patients and in-patients. Wound and rectal swabs were obtained with sterile swab-sticks dipped in sterile ringer solution, one swab per sample. The sterile swab-sticks were gently wiped on the exterior of the wound/rectum, then carefully rotated to sample the epithelial wall, removed, and inoculated aseptically into 10 mL of sterile Tryptone Soy Broth (TSB) (Oxoid, UK). The tubes were incubated at 37±2 o C for 24 hours and then streaked onto Slanetz and Bartley agar. Urine samples collected mid-stream in sterile universal bottles were streaked out on Slanetz and Bartley agar and placed in the incubator at 37±2 o C overnight. All samples were analyzed immediately upon arrival in the laboratory within 2 hours of collection. Pale pink or maroon colonies on Slanetz and Bartley agar were further identified by microscopic, morphologic, and biochemical characterization. Pure cultures of the recovered isolates were stored in freshly prepared Tryptone Soy Broth supplemented with 15% glycerol and maintained at -20 o C.

Identification on Bile Aesculin and Mannitol Salt Agar
Overnight growth from Slanetz and Bartley agar plates were inoculated onto Bile Aesculin Agar and Mannitol Salt Agar, then incubated at 37±2 o C overnight. Growth on Bile Aesculin Agar indicated the capacity to grow with bile in the media, while a change in color of the agar to dark brown was indicative of aesculin hydrolysis, both being characteristic of enterococci species. E. faecalis grows and ferments mannitol on Mannitol Salt Agar (yellow colonies), while E. faecium lacks this trait (15).

Screening for Vancomycin Resistance
Vancomycin resistance was screened for with Brain Heart Infusion (BHI) Agar supplemented with 6 ug/ mL vancomycin. Distinct colonies from an overnight culture on Nutrient agar were suspended in ringer solution, equivalent to 0.5 McFarland suspension, and spot-inoculated onto the screening agar. The plates were dried, placed in the incubator for 24 hours at 37±2°C in an inverted position. Growth of >1 colony indicated a vancomycin-resistant organism, while no growth denoted susceptibility to vancomycin.

Results
Altogether, 431 patients participated in the study based on individual consent for inclusion (130 males, 30.2%; and 301 females, 69.8%). The age range of participants fell between 15 and 83 years (mean 33.3 years). A high percentage of participants (65.7%) were in the age bracket 21-40 years, eighty-two of who were male and 201 females. Details of the demographic information of the participants are shown in Table 1 (Table 2).

Antibiotic Resistance Profile of Recovered VRE Strains
All the screened isolates exhibited resistance to ampicillin (100%). The resistance was also high to clindamycin and Q/D at 97.6% and 92.9% respectively. None of the antibiotics exhibited 100% activity against all the isolates; and the highest susceptibility rate was observed with tigecycline at 72.9%, and was closely followed by nitrofurantoin at 71.8% ( Figure 1). E. faecium from the three study locations exhibited higher resistance rates to all the antibiotics than E. faecalis, and this was evident for all the antibiotics. All isolates from the three selected hospitals were multidrugresistant. The details of the resistance patterns of VRE isolates by species and study locations are shown in Table 3.
One VRE strain displayed resistance to all the ten classes of antibiotics. This isolate with vanA phenotype (E. faecium) was recovered from the State Specialist Hospital, Osogbo from the urine sample of a 35-year-old female patient. Thirty-four VRE isolates expressed resistance to seven distinct classes of antibiotics (40.0%), alongside an additional 15 isolates resistant to 8 classes and above, making up 57.6% of the isolates resistant to 7 classes or more. Twenty-three isolates were resistant to 6 classes.  Figure 2).    Screening for vanA and vanB resistance phenotype was done by evaluating their susceptibility patterns to vancomycin and teicoplanin. Dual resistance to vancomycin and teicoplanin indicates the possible incidence of the vanA resistance gene, while resistance to only vancomycin is indicative of vanB resistance phenotype. The vanA resistant phenotype was higher at 65.9% (56/85) than the vanB resistant phenotype with a rate of 34.1% (29/85). The breakdown of the occurrence of van phenotypes of E. faecium and E. faecalis is depicted in Figure 3.

Discussion
Enterococci are found in both the human and animal gastrointestinal tract, and as such is a constituent of the natural gastrointestinal microbiome. As they are proficient at surviving under a varied range of severe situations, they often cause opportunistic infections in the health-care setting. Predominant in human infections are E. faecium and E. faecalis.
The number of female participants was higher, probably because a high number of urine samples were collected from females at the antenatal clinic, a gender-specific activity as it involves only females. The State Specialist Hospital, Osogbo had the highest number of participants as it was the largest amongst the hospitals selected. It is a tertiary hospital established by the State Government which caters to the inhabitants of Osogbo and serves as a referral center to nearby towns. This is however at variance with the study of Moosavian et al (17) as most participants (68.0%) in their study were male.
Enterococcal infections are of immense importance in worldwide health challenges, resulting in extensive illness in the populace. The prevalence of enterococci in this study was 48.3%, and E. faecium was the major isolated specie. This rate correlates with that of another study in India (18), wherein E. faecium was also the major recovered specie.  In this study, however, the percentage of enterococci colonization was lower than that of a study from Ethiopia (19) with a prevalence of 63%. This disparity could be due to the fact that samples in their study were recovered from HIV patients. Various infections caused by VRE including urinary tract, wound, and bloodstream infections are on the increase in Nigeria (20). Prevalence rates of VRE obtained from wound swabs, urine and rectal swabs in different studies vary drastically. This study reports a VRE prevalence rate of 40.9% (85/208) of which E. faecium was 71.8% (61/85) and E. faecalis 28.2% (24/85). These rates agree with the reports of other authors from different parts of the world where similar trends have been recorded (17). However, in Africa and in other parts of the world, different prevalence rates have been recorded in clinical samples vacillating between 5.7% to 88.9% (21). In similar studies from other authors, nevertheless, lower values of prevalence were recorded for E. faecium and higher prevalence for E. faecalis (22,23). These differences in prevalence rates are most likely due to variations in geographic locations. The prevalence rate recorded by our study is remarkably high enough to be worrisome. Previous studies have reported VRE transmission from the hands of caregivers in hospital wards (13,14), admission into ward spaces hitherto inhabited by patients colonized with VRE (24), contaminated surfaces of hospital wards such as floors, doorknobs, hand-rails, blood pressure cuffs, and hospital gowns (25), any of which factors is a possible means of transmission in all three hospitals included in the present study.
Reports of studies where higher prevalence rates among females than their male counterparts were recorded abound (19,21,26). This corroborates the present study as the infection rate of VRE in females was higher than that of males with a rate of 46.2% (61/132), while the VRE infection rate of males was 31.6% (24/76). However, this study revealed that infection by enterococci in males from the three selected study hospitals was essentially higher than that of females even though we had a higher number of female participants recorded during sample collection as 58.5% of the males were infected (76/130), while enterococcal infection rate in the female was 43.9% (132/301).
It is worthy of note that the major reasons that enterococci continues to exist in hospital settings are their ability to survive in diverse environments, their intrinsic ability to resist certain antibacterial agents, coupled with the development of diverse resistance patterns to many antibiotics (17) probably due to mutation or horizontal resistant-gene transfer. Within the last ten years, acquired resistance to aminoglycosides and glycopeptides, specifically vancomycin and teicoplanin employed in the therapy infections caused by enterococci, has become rampant (17). VanA and vanB phenotypes are most commonly detected in clinical VRE strains. In our study, the vanA phenotype was found to be higher at 65.9% (56/85) than the vanB phenotype (34.1%). Praharaj et al (27) reported that the majority of the VRE isolates (96.8%) in their study had the vanA phenotype, 3.3% of VRE isolates with vanC, while the vanB phenotype was not identified in any VRE strain. Even though we report only phenotypes in the present study, it has been reported that the vanB gene is clustered and takes up a significant portion of the chromosome, much unlike the vanA gene. As such, it is less likely to be spread among strains. Moreover, the vanB gene is largely associated with epidemics and food contamination, while the vanA gene is linked to hospital isolates (3).
Antibiotic sensitivity test results in our study revealed that none of the antibiotics exhibited 100% activity against all the isolates. The most effective drugs in-vitro are tigecycline (72.9%), nitrofurantoin (71.8%), ciprofloxacin (69.4%), and norfloxacin (63.5%). This finding is at variance with a study (28) where isolates showed a much higher rate of sensitivity to nitrofurantoin at 93.0%. The least resistance in this study was observed with tigecycline; a tetracycline at 27.1%. The reasons for this pattern are not quite clear but increased production of tetracycline resistance factors tet(L)-encoded MFS (Major Facilitator Superfamily) pump and tet(M)-encoded ribosomal protection protein have been reported to be proficient at bestowing tigecycline resistance to clinical isolates of enterococci (29). Studies conducted in Nigeria (20) and Iran (28) revealed 100% susceptibility of isolates to linezolid in contrast with our finding as 40% of our isolates were susceptible to linezolid. This could probably be a consequence of prior exposure to the drug, as 47.8% of our study population had been previously exposed to antibiotics (10). Other postulated reasons for the variances in outcomes could be the result of discrepancies in the study population, sample size, types of specimens, different procedures administered during isolation, level of hygiene of patients, environmental factors, regional differences, as well as contact with hospital settings.
In this study, all the recovered isolates were ampicillin resistant and multiply drug-resistant (100%), corresponding with a previous study (26). A high resistance rate to clindamycin (97.6%) was recorded in our study, a finding correlating with a study (30) that reported that >92% of E. faecium isolates had 96.0% resistance to clindamycin, while 100% resistance was recorded for the E. faecalis isolates. This is most likely due to the intrinsic resistance of enterococci generally to lincosamides (31).
Resistance to Q/D was also high at 92.9%. This is at extreme variance with an earlier report by Wang et al (32) with an incredibly low resistance rate of 1.0% to Q/D, although higher rates of resistance of enterococci to streptogramin (Q/D) was reported in Korea where resistance rates to Q/D by E. faecium was 10.0% and E. faecalis was 81.5% (33). As far as we know, there have been no reports of resistance of enterococci species and more specifically, VRE isolates to Q/D in Southwest Nigeria, and so, our study presents novel data in that regard. These high values of Q/D resistance in Southwestern Nigeria are worrisome as the drug is not frequently prescribed by clinicians. The resistance to Q/D was observed prior to its commercial usage use in the United States, suggesting that the appearance of the phenomenon may be attributed to other factors (33), and not necessarily associated with exposure to the drug (32). The resistance of enterococci species to Q/D has been reported to relate to enzymatic acetylation, efflux of the drug, and di-methylation of the 23S rRNA target site (34).
Higher levels of multidrug resistance were observed in E. faecium than in E. faecalis from all the study locations. This corresponds with various studies by different authors (17,22,23) as higher values of multidrug resistance in E. faecium were also reported. Multidrug-resistant strains of enterococci constitute serious problems in treatment regimens in patients with enterococcal infections (28). A previous study (35) reported that the percentage of VRE rose regarding resistance to a higher number of antibiotic classes, as 16%, 33%, and 100% of VRE isolates were resistant to 3, 4, and 5 antibiotic classes, respectively. While E. faecalis has been reported to be more implicated in human infections as a result of increased virulence, resistance to vancomycin is more associated with E. faecium isolates as it frequently exhibits multi-resistance characteristics (5) which are now progressively more credited to human infections such as bacteremia, endocarditis, urinary tract, and surgical site infections (4). This involvement may well be justified by their intrinsic and acquired resistance to several antibiotics together with great genomic variability, higher flexibility in the clinical environment, along incidence of a variety of virulence factors (4).
The high prevalence rate, as well as the elevated level of multidrug resistance, detected in our study is worrisome and continues to be a threat in the therapy of infection elicited by these strains as vancomycin was identified as a last line of defense against enterococcal infections. Proper regulation guiding the use of antimicrobials in medical practice in addition to resolute control of indiscriminate usage of antibiotics -oftentimes without prescriptions, must be ensured to eliminate selective pressure.

Conclusion
Herein, our study reports that the prevalence of VRE. faecium and VRE. faecalis in three selected hospitals in Southwest Nigeria was higher in females than in males. The antibiotic sensitivity pattern also revealed in-vitro effectiveness of tigecycline and nitrofurantoin against recovered VRE isolates in our study but records an excessive resistance rate to Q/D, a novel report in this region. All isolates were multidrug-resistant; this poses a great risk, as infections resulting from these organisms may complicate therapy and culminate in increased morbidity, as well as mortality.

Author Contributions
FMA designed, and supervised the research, participated in the bacteriological and molecular study, took part in data verification, interpretation, and analyses, and prepared the manuscript; N-AY contributed to study design, collected samples, performed the bacteriological and molecular investigation as well as other laboratory procedures, contributed to data interpretation. RRA and OOO participated in the bacteriological study and other laboratory procedures; KAA-N: analyzed and interpreted the data, contributed to manuscript preparation, and did the final proofreading. All authors read and approved the final version of the manuscript.

Conflict of Interests
The authors have declared that there are no competing interests.

Ethical Approval
Ethical approval for the study covering the three selected hospitals was gotten from the Health Research Ethics Board of the State Specialist Hospital, Osogbo (approval number -HREC/ SSHO/11/478).

Funding
No funding was gotten for this study