Multi-antibiotic resistant extended-spectrum beta-lactamase producing bacteria pose a challenge to the effective treatment of wound and skin infections

Introduction The increasing incidence of antibiotic resistant bacteria is a concern both to the clinicians and the patients due to obvious consequences such as treatment failures, prolonged patients’ stay in hospital and nosocomial infections. The choice of the first antibiotic therapy in emergency wards in hospitals is usually not based on patient-specific microbial culture and susceptibility test result.This study is aimed at profiling extended-spectrum beta-lactamase (ESBL) producing bacteria associated with wound injuries and highlighting their multi-antibiotic resistance character. Methods Sixty-three wound swab samples were collected and cultured on nutrient agar and on selective media. Evaluation for ESBL production was by phenotypic method while the antibiogram screening was by disc-diffusion. Results The wounds evaluated were diabetic sore (14), cancer wounds (12), surgical wounds (17), wounds due to road traffic accidents (10) and wounds from fire burn (10). The result showed that 61 wounds were infected and the prevalence of the infecting pathogens was Escherichia coli 17.46%, Klebsiella Pneumonia 14.28%, Salmonella typhi 12.79%, Pseudomonas Aeruginosa 34.92% and Staphylococcus aureus 17.46%. Thirty four (55.74 %) isolates were ESBL producers, greater than 50% of which being Pseudomonas Aeruginosa. The antibiogram study of the ESBL producers showed multi-drug resistance with resistance highest against ampicillin (100%), followed by cephalosporins: cefuroxime (94.12%) and ceftriaxone (61.76%). No resistance was recorded against the β-lactamase inhibitors: amoxicillin/clavulanate and ceftriaxone/sulbactam. There was a high incidence (55.74 %) of ESBL-producing microbes in the wounds. The isolates were mostly multi-antibiotic resistant. Conclusion Multi-drug resistant ESBL-producing bacteria are common in wound infections in the community. However, amoxicillin/clavulanate or ceftriaxone/sulbactam may be used to treat most patients with such infections in the hospital. This may guide antibiotic selection and use in trauma, most especially in resource limited countries where laboratory test is unaffordable for a majority of patients.


Introduction
The increasing incidence of antibiotic resistant bacteria is a concern both to the clinicians and the patients due to obvious consequences such as treatment failures, prolonged patients' stay in hospital and nosocomial infections. The choice of the first antibiotic therapy in emergency wards in hospitals is usually not based on patientspecific microbial culture and susceptibility test result. Extended spectrum β-lactamases (ESBLs) have been reported to be common cause of hospital-acquired infections and can have severe clinical implications with a corresponding multiple antibiotic resistances [1,2]. The enzymes, β-lactamases, deactivate the molecular antibacterial properties of β-lactam antibiotics by breaking and opening the β-lactam ring. Bacteria harboring these enzymes are usually resistant to β-lactam antibiotics as well as antibiotics from other classes thereby posing a therapeutic challenge to clinicians [3]. ESBLs enzymes are carried in and transferred by bacteria plasmid [4,5] and are responsible for bacteria resistance to βlactam antibiotics but are inhibited by β-lactamase inhibitors [6].
Increase in ESBL-producing enteric Gram-negative bacteria has led to the choice of inappropriate therapy and consequently, the rate of resistance has increased. Antibiotic selection for infections due to ESBL-producing pathogens is still a clinical challenge. In most cases, carbapenems and fluoroquinolones have been used [7].
Horizontal gene transfer by plasmid exchange between E. coli strains is a recognized source of rapid spread of antimicrobial resistance phenotypes [4]. The understanding that plasmids are linked with mobile genetic elements highly suggests that they are responsible for transferring resistant genes to susceptible bacteria recipients [8]. Resistance to different classes of antibiotics had been detected among ESBL producers [9,10]. Selective pressure of the antimicrobials selects those strains that are resistant to the applied antimicrobials causing the resistant strains to multiply and spread [11,12].
Indiscriminate use of antibiotics, poor hygienic practices in hospitals and lack of monitoring for microbial drug resistance create suitable conditions for the emergency and spread of the ESBLs [10]. Other factors reported as fueling the spread of ESBLs in developing economies include extensive self-medication/prescribing and nonprescription use of antimicrobials, poor hygienic conditions even in the hospital environment and very low infection control practice [13,14]. Knowledge on local antimicrobial resistance trends among wound isolates is important not only in guiding clinicians to prescribe appropriate antibiotics but also for evidence-based recommendations in empirical antibiotic treatment of wound and other infections [15,16]. This evidence based prescribing is highly needed everywhere antibiotics are prescribed.
The aim of this study is to identify the microbial contaminants of wounds in patients admitted in a tertiary hospital in Nigeria, determine the incidence and antibiogram of ESBLs producing bacteria and point out the incidence of multi-drug resistant ESBLs producing isolates. This will help to provide evidence-based recommendations for empirical antibiotic treatment of infections caused by ESBL producing bacteria.

Study area
This is a prospective experimental study designed and carried out in Chukwuemeka Odumegwu Ojukwu University Teaching Hospital Amaku-Awka, a 150 bed space hospital. The hospital serves an urban community-Awka, the State capital and several rural communities with a total population of over 2 million persons. It is a state-owned tertiary health institution providing medicare to patient as well as medical and nursing training.

Patients' identification and sample collection
The patients were drawn from medical, surgical, obstetrics, pediatric and emergency (both children and adults) wards as well as the Intensive care unit of the hospital. Every patient in the wards with wound of at least one week duration (except in emergency wards which were two days duration) was recruited and wound swab samples were collected between March and June 2015 by a medical personnel using sterile swab-stick. Samples collected were analyzed within 30 minutes of collection. Verbal informed consent was gotten from the patients or their legal relatives after due explanation of the study protocols and purpose.

Isolation and identification of organisms
The swab sticks used for the collection of the samples were streaked directly on Mac-conkey, Mannitol salt, Cetrimide and Page number not for citation purposes 3 Chocolate blood agars and incubated at 37 o C for 24 hrs. After incubation, cultures were examined for significant growth.
Subcultures were plated onto nutrient agar plates and incubated for another 24 hrs. Grams staining and biochemical tests (Indole production, Oxidase, catalase and Coagulase) were carried out in accordance with standard methods [17].

Preliminary screening of isolates
Gram staining reaction was carried out to distinguish the Gram positive and Gram negative isolates. Briefly, a minute quantity of the isolates colonies was aseptically transferred from the Petri dish on to a sterile slide previously wetted with a few loopful of water. The culture was spread evenly and gently with an inoculation loop to make a small circular thin film and then air-dried and fixed over a gentle flame. Crystal violet stain was added over the fixed culture and allowed to stand for 10-60 seconds. Enough mordant (iodine solution) was added to cover the fixed culture and allowed to stand for another 10-60 seconds. The slide was rinsed with running water, thereafter, carefully cleansed to wipe off the excess water from its surface. A few drops of alcohol solution (a decolorizer) were added such that the solution trickled down the slide. After optimal decolonization, the slide was again rinsed with water and basic fuchsin solution (a counterstain) was applied for 40-60 seconds.
After washing with water, the slide was air-dried and then viewed under the microscope.

Sugar fermentation test
The Mannitol Fermentation Test is a general test that confirms all the Enterobacteriaceae isolates. They produce gas on metabolizing mannitol. Briefly, an inoculum from the pure culture was aseptically transferred to a sterile tube containing phenol red mannitol broth (nutrient broth to which 0.5-1.0% mannitol was added) and bacteria. Two drops of oxidase reagent was placed on a filter paper on a slide, a colony of the test organism was collected using a previously flamed and cooled wire loop, and smeared across the emulsified filter paper.

Indole test:
The indole reagents was prepared as follows; A 3ml of sterile tryptone water was placed in a bijou bottle, after which the isolates were inoculated in these bottles and incubated at 35-37°C for up to 48 hrs. Indole ring formation was tested by adding 0.5ml of Kovac's reagent with a gentle shake.

Screening for ESBL bacteria
This was done using Double Disc Synergy Test (DDST) [18]. Briefly, 20 ml of Mueller Hinton agar was prepared and dispensed aseptically into a Petri dish and was allowed to solidify. The plates were seeded with test organisms pre-adjusted to 0. Method of data analysis: Data was analyzed using simple percentage.

Ethical approval and compliance with ethical standards
The ethical approval was obtained from the Ethical Committee of the Hospital Management (Approval Number: HD92/A2/37). The study was conducted following the international, national, and institutional ethical guidelines and according to the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The participants gave verbal consent, to collect the wound swabs, after the purpose of the study and intentions to publish the findings were explained. Patients were assured that their personal identities will not be linked to any data.

Discussion
This result (Table 1) showed that the rate of isolation of Gramnegative organisms was more than Gram-positive and is consistent with several other reports [10,[20][21][22]. The reason why P.
aeruginosa was the most isolated pathogen in this study could be because it is a common cause of nosocomial infection [21,22] as well an opportunistic pathogen that thrives more wherever there is suppressed immunity as seen in wound infections. The incidence rate of the wounds infections was high (96.83%) probably due to factors associated with the acquisition of nosocomial pathogens in patients with recurrent or long-term hospitalization, complicating illnesses, prior administration of antimicrobial agents, or the immunosuppressive effects of wound trauma. According to Iroha et al., [11], β-lactam antibiotics are the most frequently prescribed antibiotics against aerobic Gram negative bacilli infections in Nigeria, and selective pressure exerted by the extensive use of these βlactam drugs, most likely resulted in strains developing ESBL enzymes. Occurrence and distribution of ESBLs differ from country to country and from hospital to hospital.
The present study ( Table 2) also showed that the ESBL producers were coming mainly from the non-Enterobacteriaceae family-Pseudomonas genera (36.07%). Again, this takes the advantage of suppressed immunity caused by wound trauma. This result differs from the reports by Afunwa et al., [12] and Ankur et al., [23]. They reported that the Enterobacteriaceae family is the major ESBL producer. The cause of the discrepancy could be because most of their samples came from immunocompetent patients. They also identified the multi-drug resistance (MDR) bacteria strains to belong principally to the Enterobacteriaceae family and Pseudomonas genera with a greater percentage skewing towards the Enterobacteriaceae category. Mathur et al., [24] documented the prevalence of ESBL producers to be as high as 68.0%, whereas Kumar et al., [25] reported 19.2% of E. coli isolates as ESBL producers. This is contrary to 20.27 % ESBL producing isolates of P. aeruginosa reported by Aggarwal et al., [26] proving the increasing incidence of the organism in hospitalized patients [27,28].
Salmonella enteric Typhi serovar is a Gram-negative flagellated short rod of the Enterobacteriaceae family. The contamination of the wounds by this organism in our study revealed possible spread from environment, contaminated water and/or undercooked food.  (Table 3, Table 4) is a cause for concern as many clinicians fall back on them for the treatment of Gram-negative pathogens in the face of multi-drug resistance [31]. Many of the ESBL producing isolates (61.76% and 94.12%) were resistant to the cephalosporins used (ceftriaxone and cefuroxime respectively) in the study (Table 3). This is contrary to the study done by Sasirekha et al., [33] in India where they found resistance rates to be as few as 24% and 29% respectively. May be they did not encounter ESBL producers in their study. It was also noted that Pseudomonas and Klebsiella species were more resistant than Escherichia coli. This may be due to their inherent virulence factors like hyper-viscosity, polysaccharide capsules and production of endotoxin [35].
Multi-Antibiotics Resistance results when microbes are resistance to tested antibiotics in at least two of the classes: quinolones, aminoglycosides and β-lactams [36,37]. It is identified by profiling the pattern of resistance of microbial isolates to the antibiotics belonging to the above classes.

What is known about this topic
• Antibiotic resistance is a concern to both the clinicians and their patients; • Extended spectrum β-lactamase (ESBLs) producing organisms are implicated in hospital-acquired infections; • Antibiotic selection for infections due to ESBL-producing pathogens is a clinical challenge.

What this study adds
• High incidence (55.74 %) of ESBL-producing microbes in the wound and skin infections, non-enterobacteriaceae being the dominating isolates; Page number not for citation purposes 7 • The multi-antibiotic resistance index calculation showed that the incidence of multi-antibiotic resistance in the study was 100%; • The emergence of ESBL-producing Salmonella spp.

Competing interests
Authors wish to declare that the author,

Acknowledgments
The authors wish to thank the hospital management where the study was carried out for granting us permission to do the study.