Detection of Carbapenemase-producing Klebsiella pneumoniae isolated from Environmental Sources in a Tertiary Health Institution in Nigeria.

Objective The acquisition of carbapenemase-producing organisms in healthcare settings is a major threat and has serious implications for public health. Previous reports regarding carbapenemase-producing Enterobacteriaceae from fomites are limited. This study aimed at analysing the antimicrobial resistance patterns, and prevalence of carbapenemase-producing Klebsiella pneumoniae in the ward environments of a tertiary health institution in Nigeria. Results A total of 142 bacteria were isolated from 534 fomites in the hospital wards, and of the 142 isolates, 15(10.6%) were conrmed to be Klebsiella pneumoniae . The prevalence of Klebsiella pneumoniae in all the 534 samples was 15/534(2.8%), while the prevalence of carbapenemase-producing Klebsiella pneumoniae was 8/534(1.5%). Multi-drug resistance was detected in 15/15(100%) of the Klebsiella pneumoniae isolated. Although no Klebsiella pneumoniae Carbapenemase ( bla KPC ) gene was expressed in any of these isolates, 8/15(53.3%) of these isolates were conrmed positive for carbapenemase production using the Modied Hodge Test. The commonest sites that harboured carbapenem-resistant Klebsiella pneumoniae were the beds 6/15(40%). Maximum resistance (100%) was observed against ampicillin, trimethoprim-sulfamethoxazole, cefuroxime, and tetracycline. In conclusion, the prevalence of carbapenemase-producing Klebsiella pneumoniae fomite colonization in the NAUTH ward environment is low, thus buttressing the need to reinforce strict infection control policies the hospital.


Introduction
Klebsiella pneumoniae are Gram-negative, non-motile, encapsulated bacilli belonging to the family of bacteria called the Enterobacteriaceae. [1] They are considered the second most common cause of healthcare-associated sepsis, remaining for long periods in hospital environments and equipment, and may be spread to patients by contact with these environmental surfaces. [2][3] They develop resistance by various mechanisms, but by far, the most troublesome of these are the carbapenemases which make the organisms resistant to almost all forms of antibiotics, especially carbapenems which have been considered as agents of last resort in the treatment of infections caused by MDR Gram-negative bacilli. [4] [5] The burden of anti-microbial resistance (AMR) in developing countries has increased remarkably in recent years. [6] [7] In a 2017 review of AMR in Africa, only about 60% of the countries had available data on AMR. There was a strikingly high median resistance (MR) rate for the Enterobacteriaceae to ampicillin (MR= 88.1%). [6] Resistance was however uncommon for the carbapenem group of antibiotics. In particular, Klebsiella spp. which has intrinsic resistance to ampicillin, was resistant to 34.2% and 46.7% of ceftriaxone or cefotaxime respectively, suggesting high-level extended-spectrum beta-lactamase (ESBL) production. However, the median resistance rate for K. pneumoniae against Imipenem, a carbapenem was 3.0%. [6] In another survey involving Africa and Asia, resistance to ampicillin and ceftriaxone were reported as 67.2% and 25.9% respectively. [7] The most frequently detected carbapenemases include class A-Klebsiella pneumoniae Carbapenemase (KPC) types), class B-metallo-β-lactamases (MBLs) viz Verona integron-encoded metallo-β-lactamase (VIM) and NewDelhi metallo-β-lactamase (NDM) types, and class D-oxacillinases (OXA-48-like enzymes).
[8] Furthermore, the KPCs have been documented as major causes of nosocomial outbreaks. [9][10] [11] Several studies done previously on carbapenemase detection focused more on isolates from clinical specimens of patients, but limited information is found in the literature on the prevalence of carbapenemase-producing Klebsiella pneumoniae in the hospital environment. One environmental study worthy of note was that in which the prevalence of carbapenemase-producing Klebsiella pneumoniae was determined in environmental sites of Intensive Care Units (ICUs) in Cairo, Egypt. [12] This study, therefore, aimed at determining the occurrence of carbapenemase-producing Klebsiella pneumoniae in the ward environments of a tertiary health institution in Nigeria.

Study Population
One hundred and forty-two human bacterial pathogens were isolated from 534 environmental specimens obtained in the wards of NAUTH, Nnewi, a major referral centre serving individuals from most parts of South-East, Nigeria. The bacteria were collected between January 2018 and June 2018, and the specimens included swabs collected from; patients beds, bedside tables, bedside cupboards, trolleys, sphygmomanometers, water taps, antiseptics, disinfectants, hand wash solutions, hand sanitizers, forceps, wheel chairs, kidney dishes, door handles, drip stands, drug mortars, methylated spirits, suction tubes, nurses desks, doctors desks and pulse oximeters.

Bacterial Isolation
Duplicate swabs were collected by rolling moistened sterile swab sticks over the sites mentioned above for about 5 seconds. These swabs were sent to the laboratory immediately after collection and cultured on chocolate and Mac Conkey agar (Oxoid, UK) and incubated at 35-37°C for 24 hours. [10][12] The isolates were Gram-stained, and the Gram-negative rods were subjected to con rmatory identi cation of Klebsiella pneumoniae using the Microbact TM Gram-negative bacteria identi cation kit (Oxoid, UK). [10] Antimicrobial Susceptibility testing The Modi ed Kirby-Bauer antimicrobial susceptibility testing technique was performed on all isolates con rmed as Klebsiella pneumoniae. [13] [14] A lawn of each bacterial inoculum equivalent to 1.5 X 10 8 CFU/ml, was made on the surface of a Mueller-Hinton agar (Oxoid, UK) plate using a sterile swab stick and left to dry for 3-5 minutes. Antibiotics were then placed on the lawn, and the plates incubated aerobically at 35-37 o C for 16-18 hours. The zones of growth inhibition around each antibiotic disc was measured and reported based on the guidelines of the CLSI. [14] Screening for suspected carbapenemase production Screening involved placing 10μg carbapenem (ertapenem and meropenem) discs (Oxoid, UK) on the surface of Mueller Hinton agar (Oxoid, UK) plates inoculated with each isolate. Following incubation of the plates for 16-18 hours at 35-37 0 C, zones of growth inhibition around each antibiotic were read off. Klebsiella pneumoniae isolates that showed a zone of inhibition ≤ 22mm in diameter for meropenem or ≤ 21mm for ertapenem were considered as suspected carbapenemase producers and were subjected to phenotypic con rmation by the Modi ed Hodges Test (MHT) [11] [14].
Phenotypic con rmation of carbapenemase production (Modi ed Hodges Test) In this method, a suspension of E. coli ATCC 25922 equivalent to 0.5 McFarland turbidity standard was prepared. The E. coli suspension was then diluted 1:10 by adding 0.5 ml of the E. coli suspension to 4.5 ml of saline. A lawn of the 1:10 dilution of E. coli ATCC 25922 was evenly streaked onto Mueller Hinton agar plates using sterile cotton swabs and then allowed to dry for 3-5 minutes. One disc of Meropenem (10µg), was placed on the centre surface of the MHA plate. In a straight line, using a sterilized wire loop, the test organisms were streaked from the edge of each Meropenem disc to the edge of the plate. The plates were incubated at 37 o C for 24 hours. After incubation, they were examined for a clover leaf type indentation at the intersection of the test organism and E. coli ATCC 25922 within the zone of inhibition of the meropenem disc as described by the CLSI. [14] K. pneumoniae ATCC 1705 and K. pneumoniae ATCC 1706 were used as positive and negative controls. [14] Molecular Detection of bla KPC Bacteria DNA from the Klebsiella pneumoniae isolates was extracted using a previously described boiling method for DNA extraction with slight modi cations. [15][16] The extracted DNA was quanti ed and tested for purity using the NanoDrop ® ND-1000 spectrophotometer (Additional le 1: Table S1). The bla KPC gene was detected using a conventional PCR reaction that was based on the protocols and primer sequences previously published by Shanmugam et al., [17] with slight modi cations. (Additional le 2: Table S2).
The PCR conditions for bla KPC detection were as follows: initial denaturation at 94ºC for 3 minutes, followed by 30 cycles of denaturation at 94ºC for 1 minute, annealing at 60ºC for 1 minute, extension at 72ºC for 1 minute, then nal extension at 72º C for 5 minutes. The products were then resolved at 130V for 25 minutes on 1.5% agarose gel stained with 0.5μg/ml ethidium bromide solution (Nippon Genetics, Europe GmbH) in an electrophoresis tank containing 1 mMol Tris-Borate EDTA (TBE) buffer. The gels were observed under UV gel Transilluminator (UV DOC, England) at 280nm and the band pattern observed.

Data Analysis
Statistical analysis was done using STATA version 13 (Stata Corp LP, Texas, USA). Frequency distribution tables were used to determine rates.

Results
The prevalence of bacterial contamination in the total sample population was 142/534(26.6%). Out of these, 15(10.6%) were identi ed as Klebsiella pneumoniae, thus, the prevalence of the Klebsiella pneumoniae in the entire sample population was 15/534(2.8%). (Additional le 3: Table S3).
The Male Surgical Ward had the highest proportion of Klebsiella pneumoniae isolates 5(33.3%), followed by the Male and Female Medical wards which had 3(20%) each. (Additional le 4: Table S4).

Discussion
Klebsiella pneumoniae is a frequent cause of nosocomial infections, accounting for up to 10% of all nosocomial infections. [18] Carbapenems are the drugs of choice for the treatment of infections caused by drug resistant Enterobacteriaceae. [19] Unfortunately, rising bacterial resistance to carbapenems has been well documented. [20] Previous studies have shown that Klebsiella pneumoniae strains of environmental origin are similar to those of clinical origin in terms of biochemical patterns, virulence, and pathogenicity. However, clinical Klebsiella pneumoniae have been observed to be significantly more resistant to antibiotics when compared with environmental Klebsiella pneumoniae. [21] Klebsiella pneumoniae was isolated from 15/534 (2.8%) of the study population. A slightly lower rate was obtained in environmental isolates of Klebsiella pneumoniae in an Egyptian hospital, where 4/100 (0.04%) of the study population was found to harbour Klebsiella pneumoniae. [22] Out of 142 isolated organisms, 15 (10.6%) were con rmed to be Klebsiella pneumoniae with 8(53%) of these observed to be producing carbapenemases. A higher rate was observed in the northern region of Brazil, where 25/25 (100%) of the Klebsiella pneumoniae isolates were con rmed as carbapenemase producers, [23] but much lower values were observed for clinical isolates of Klebsiella pneumoniae in a Chinese study 4/153 (2.6%). [24] In Kano, Nigeria, a low prevalence of carbapenemase-producing Klebsiella pneumoniae was also observed 6/73 (8.2%). [11] The varying prevalence of carbapenemase production could be a result of varying selection pressures from different antibiotic prescribing preferences in different countries. These varying observations were highlighted in a statement by Oduyebo et al., that carbapenemase production among the Enterobacteriaceae has been widely reported with prevalence ranges between 2.8% and 53.6%. [10] The most frequent site of isolation was in beds 6/15 (40%), followed by bedside cupboards 4/15 (26.7%), and then bedside tables 2/15 (13.3%). This nding was similar to that observed in Egypt, where the Klebsiella pneumoniae isolated from several ICUs were found more in beds, bedside tables, suction tubes, and ventilator tubes. [12] However no Klebsiella pneumoniae was isolated from the ICU in this study. This variation in the detection of the organisms from the ICUs of the different hospitals could be attributed to the maintenance of strict infection control measures in the ICU of NAUTH, Nnewi.
The antibiotic susceptibility patterns of the Klebsiella pneumoniae isolates revealed that the organisms had maximum resistance (100%) to Ampicillin, Sulfamethoxazole-Trimethoprim, Cefuroxime, and Tetracycline, but were most susceptible to the Carbapenem class of antibiotics, in which Imipenem showed the most sensitivity (73.3%). Contrasting ndings were observed in an Egyptian study which revealed 100% resistance to Meropenem. [12] The reduced rates of resistance to the carbapenems in this study could be attributed to the limited use of carbapenems due to the high cost of purchase of these antibiotics in the country.
None of the 15 isolates of Klebsiella pneumoniae produced bla KPC . Although this was similar to ndings observed in previous Nigerian studies which dealt with clinical isolates of Klebsiella pneumoniae,[10] [25] contrasting observations were seen in Maiduguri, Nigeria (6.5%). [11] A signi cantly different nding was also observed in a Brazilian study that revealed that 100% of the Klebsiella pneumoniae isolates carried the bla KPC gene. [23] The contrasting rates may be due to long term high use of carbapenems in Brazil, but the still recent introduction of these drugs in Nigeria.
The Klebsiella pneumoniae isolates were phenotypically positive for carbapenemase production on Modi ed Hodge Test but were negative for bla KPC gene on PCR. This could be because these isolates harboured other carbapenemase-producing genes (including bla NDM, bla VIM, bla OXA-48 etc), which were not searched for in this study.

Conclusion
Although the prevalence of carbapenemase production in the Klebsiella pneumoniae isolates was high, the rate of colonization of fomites with these pathogens in the NAUTH ward environment was still quite low, thus buttressing the need to reinforce strict infection control policies the hospital.

Limitations
All the genes responsible for carbapenemase production were not searched for. Although this limitation did not adversely affect the aim of this study, which was to determine carbapenemase production in the organisms, it would have been more accurate to detect all the genes responsible for its production. The phenotypic detection method (MHT) used in this study, helped to curb this limitation. Larger sample size may also have helped to improve the accuracy of the study.

Availability of data and materials
The necessary data generated or analysed during this study are included in this article (and its supplementary information les).
Competing interests