Metallo-beta-lactamase (MBL) activity has emerged as one of the most feared resistance mechanisms because of its ability to hydrolyze virtually all beta-lactams, including carbapenems. However, MBLs are unable to hydrolyze monobactams. Based on the molecular studies, carbapenemases i.e., enzymes hydrolyzing carbapenems are classified into four groups: A, B, C and D. Metallo-beta-lactamases belong to Amber class B type of beta-lactamase and act on a broad spectrum of substrates including penicillins, cephalosporins, and carbapenems [1]. MBL producing Pseudomonas aeruginosa was first reported in Japan in 1991, since then it has been reported in various parts of the world including Asia, Europe, Australia, South America, and North America [2–9]. The detection of MBL-producing Gram-negative bacilli is crucial to control the spread of resistance and for the optimal treatment of patients, particularly the critically ill and hospitalized patients [10].

In context of Nepal data on the burden of MBL producing Gram negative bacteria is limited [15]. Since, the knowledge about the types of enzymes that may be present can serve to guide the infectious disease physician toward choosing appropriate therapy without the need for extensive secondary testing, detection of MBL producing isolates is of paramount importance in clinical setting. Given the background, this epidemiological study was designed to generate updated information on the burden of metallo-beta-lactamase producing Gram negative bacteria from a tertiary care hospital in Nepal so that an effective antimicrobial stewardship policy can be formulated and implemented to circumvent the rising threat of antimicrobial resistance.

A total of 4765 different clinical samples (urine, pus, body fluids and sputum) were included from patients admitted or attending outdoor patient (OPD) during the study period of 6 months from May to November 2014. Out of them 664 (13.9%) samples were growth positive, among which 362 (54.5%) were Gram-negative isolates with E. coli being the most prevalent i.e., 199 (54.9%). Out of the total 362 Gram negative isolates, 196 (54%) samples were from the outdoor patients (OPD) and the remaining 166 (46%) samples were from the indoor (hospital admitted) patients. MBL was detected in 21 (5.8%) of the culture positive Gram-negative bacteria (Table 3).

Metallo-beta-lactamases are a large and diverse group of beta-lactamases that are now disseminating on mobile genetic elements among clinically important Gram-negative pathogens, limiting treatment options for life-threatening infections [1, 17, 18]. Infection with the metallo-beta-lactamase (MBLs) producing organisms are associated with higher rates of mortality, morbidity, and health care costs [19]. In any nosocomial setting, carbapenems are used as the last resort for treatment of MDR Gram-negative bacterial infection. However, since last 15 years, acquired resistance to this life saving antimicrobial has been increasingly reported not only in P. aeruginosa and Acinetobacter spp. but also among members of Enterobacteriaceae which is mainly mediated by Klebsiella pneumoniae carbapenemases (KPC) [20].

In this study, among the 362 Gram-negative isolates recovered, 5.8% were found to be MBL producer which is higher compared to the prevalence rate of 1.3% reported from a similar study conducted in Nepal during 2012 that reported MBL producers from samples of lower respiratory tract (LRT) infection cases [15]. The findings of current study suggest that there is a continuous proliferation of MBL producers in Nepal. During the surveillance period of 6 months, only P. aeruginosa and Acinetobacter spp. were found to be MBL producers out of the 362 Gram-negative isolates recovered: similar to the results of previous study [15], which also reported the MBL production among non-fermentative bacteria only with no case of other members from Enterobacteriaceae showing MBL production. However, two different studies conducted in India reported production of MBL by members of Enterobacteriaceae including E. coli, Klebsiella spp., Enterobacter spp., and Citrobacter spp. [21, 22].

The highest prevalence rate of MBL producers in this study, was detected from urine and pus sample i.e. 7/21 (33.3%) of the total MBL producers isolated. All the MBL producing P. aeruginosa were isolated from urine and maximum MBL producing Acinetobacter spp. 7/21 (33.3%) were isolated from pus. Meanwhile, blood samples were not considered for present surveillance study owing to the lack of BACTEC™ instrumented culture system in the hospital and also because blood culture is recommended mostly for cases related to enteric fever complaints or those presenting with symptoms of sepsis which upon culture yield Salmonella spp. or Gram positive cocci in most of the attempts.

A PCR based method is usually considered to be the best method for detecting MBL-producing isolates. However, the increasing number of types of MBLs is creating difficulties in detection of MBLs, since primers used for PCR are usually designed to detect a single gene type [14]. Furthermore, for a low income county like Nepal use of PCR based technique in surveillance process would be expensive and undesirable from financial aspect to test every single suspected isolate. To circumvent this problem we have used phenotypic detection technique using imipenem-EDTA combined disk method, which in one hand has sensitivity and specificity of 100 and 98%, respectively and on the other hand is cost effective as well [16].

With respect to antimicrobial susceptibilities, 159 isolates i.e., 100 isolates from the member of Enterobacteriaceae family, 24 P. aeruginosa and 35 Acinetobacter spp. were found to be resistant against ceftazidime upon initial screening. Among the ceftazidime resistant isolates belonging to Enterobacteriaceae family, the most sensitive antibiotic was Nitrofurantoin, 91% followed by amikacin i.e. 85%. Polymxin B was found to be the most effective drug with 100% susceptibility to cefatzidime resistant P. aeruginosa followed by imipenem i.e., 91.67% which is in concordance with the findings of Mishra et al. [15]. In this study, 14/24 (58.33%) of P. aeruginosa were found to be resistant to Piperacillin which is in agreement with 69.1% resistance of P. aeruginosa to Piperacilin as reported by Aibinu et al. [23]. CLSI recommends ofloxacin as supplemental when the P. aeruginosa is isolated from urine sample but for the uniformity, in this study it was used to every isolates of P. aeruginosa irrespective of the sample type.

Likewise, among the ceftazidime resistant Acinetobacter spp., tigecycline was found to be the most sensitive drug i.e. 30/35 (85.7%), which is in agreement to the findings of Mishra et al. [15]. Since Tigecycline disk diffusion breakpoint for Acinetobacter spp. is not recommended by CLSI, for the purpose of this study, U.S. Food and drug administration (FDA) tigecycline break point criteria for Enterobacteriaceae i.e. ≥19 mm for susceptibility and ≤14 mm for resistance was used which in recent years have been followed by other researchers as well [24]. In this study, 24/35 (68.5%) of the ceftazidime resistant Acinetobacter spp. were sensitive to Imipenem. Similar reports of higher sensitivity of Imipenem to Acinetobacter spp. were reported in other studies [25, 26].

The highest rate of MDR in our study was seen among E. coli strains followed by Acinetobacter spp. and P. aeruginosa. Among the total P. aeruginosa isolates 44.4% were found to be MDR which is similar to 49.8% MDR reported by Strateva et al. [27]. However, some other studies conducted in Nepal have reported slightly higher MDR cases i.e. 51.3% and 65.3%, respectively [15, 28]. Likewise, among the total Acinetobacter spp. isolated, 82.7% were found to be MDR which is similar to previous study conducted in Nepal, where 95% of Acinetobacter spp. isolated were reported to be MDR [15]. Similarly, in another study conducted at National Institute of Neurological and Allied sciences, Kathmandu, 85.4% isolates of Acinetobacter spp. were reported MDR [28]. The current findings is an alarming sign, since almost half of the MBL producers have been found to be MDR strains leaving the medical practisoners with limited therapeutic options to combat such pathogens.

The armamentarium against MDR Gram-negative microorganisms has almost been exhausted especially after the advent of carbapenem resistance among them. Until last year, parenteral colistin available as colistin methanesulfonate (CMS) showing potent activity in vitro against MDR nosocomial P. aeruginosa, Acinetobacter spp., Stenotrophomonas maltophilia, Enterobacter spp. and Klebsiella spp., including ESBL and carbapenemase-producers [29, 30] was the ultimate treatment option. However, recent reporting of plasmid-mediated colistin resistance in Escherichia coli isolated from animals, food, and patients in China by Liu et al. in November, 2016 [31], has left us with no option. Hence, proper implementation of infection control strategy, active antimicrobial stewardship approach, improved laboratory detection, judicious use of antimicrobial agents, along with regular national level surveillance can be some of the arbitration measures to control as well as aiding formulation of strategy in tackling drug resistance issues like the current one under discussion.

The findings of our study demonstrated a higher prevalence of MDR and MBL positive P. aeruginosa and Acinetobacter spp., which have been globally incriminated with adverse clinical outcome including a higher morbidity and mortality rate. The study results demonstrate the serious therapeutic and epidemiological threat of the spread of metallo-beta-lactamase producers. Since, antimicrobial resistance is a growing threat worldwide with increasing resistance to third generation cephalosporins becoming a cause of concern among Enterobacteriaceae; early detection and infection control practices are the best defense against these organisms.