Extended-spectrum β-lactamase blaCTX-M-1 group in gram-negative bacteria colonizing patients admitted at Mazimbu hospital and Morogoro Regional hospital in Morogoro, Tanzania

Objective The objective of this study was to determine the proportion of extended spectrum β-lactamase producing gram-negative bacteria (ESBL-GNB) colonizing patients admitted at Mazimbu hospital and Morogoro Regional hospital, in Morogoro, Tanzania. Rectal colonization with ESBL-GNB increases the risks of developing bacterial infections by extra-intestinal pathogenic ESBL-GNB. Results Of the 285 patients investigated, 123 (43.2%) carried ESBL-GNB in their intestines. Five of the 123 ESBL positive patients were colonized with two different bacteria, making a total of 128 ESBL producing isolates. Escherichia coli (n = 95, 74.2%) formed the majority of ESBL isolates. The proportion of CTX-M-1 group genes among ESBL isolates tested was 94.9% (93/98). History of antibiotic use (OR: 1.83, 95% CI: 1.1–3.2, P = 0.03), being on antibiotic treatment (OR: 2.61, 95% CI: 1.5–4.53, P = 0.001), duration of hospital stay (OR: 1.2, 95% CI: 1.1–1.3, P < 0.001) and history of previous admission (OR: 2.24, 95% CI: 1.2–4.1, P = 0.009) independently predicted ESBL-GNB carriage.

Colonization with ESBL producing gram-negative bacteria (ESBL-GNB) increases the risk of developing multidrug resistant (MDR) bacterial infections e.g., bloodstream infection, urinary tract infection or wound infection [12]. Infections with MDR bacteria are associated with increased days of hospitalization, healthcare Open Access BMC Research Notes *Correspondence: vsilago.silago2@gmail.com 2 Department of Microbiology and Immunology, Weill Bugando School of Medicine, Catholic University of Health and Allied Sciences-Bugando, P. O. Box 1464, Mwanza, Tanzania Full list of author information is available at the end of the article costs and mortalities from treatment failure and/or limited therapeutic options [13].
In Tanzania, previous studies from national and zonal referral hospitals have reported magnitudes of rectal/ intestinal carriage of ESBL producing gram-negative bacteria (ESBL-GNB) ranging from 15% to 59.7% among hospitalized patients [14][15][16][17]. ESBL producing E. coli (ESBL-EC) and ESBL producing K. pneumoniae (ESBL-KP) are frequently reported with proportion ranging from 30% to 68.7% and 28.2% to 77.1%, respectively [14,15,17]. The magnitude of ESBL rectal colonization and associated factors among hospitalized patients in other tiers of the healthcare facilities like regional and district hospitals has not been well studied in developing countries including Tanzania. The objectives of this study was to determine the magnitude and factors associated with rectal colonization with ESBL producing gram-negative bacteria (ESBL-GNB) among hospitalized patients at Mazimbu hospital and Morogoro Regional hospital in Morogoro, Tanzania. Therefore, this study's findings provide baseline information to improve measures of infections prevention and control (IPC).

Study design, population, duration and settings
This cross-sectional analytical study was conducted among patients admitted at Mazimbu hospital (~ 30 beds capacity) and Morogoro Regional hospital (~ 450 beds capacity) in Morogoro region, Tanzania between May and July 2017. A minimum sample size of 280 was obtained using Kish and Leslie formula (1965) and a prevalence of 24% [5]. Participants stayed ≥ 24 h in hospital wards were eligible to be enrolled in this study. A standardized data collection tool was used to collect socio-demographic and clinical associated data relevant to study's objectives.

Sample collections and laboratory procedures
Sterile swabs (Mast Diagnostica GmbH, Germany) in Amies transport media were used to collect a single time rectal swab from participants. Then, transported to Microbiology laboratory at Morogoro Regional hospital within 4 h of collection for laboratory analysis. Screening of presumptive ESBL-GNB was done by direct inoculation of rectal swab samples on MacConkey agar (MCA; Oxoid, UK) plates supplemented with 2 µg/ml cefotaxime (MCA-C) incubated in ambient air at 37 °C for 24 h [18,19]. CHROMagar ESBL plates (BD BBL ™ CHROMagar ™ ESBL, Germany) were used for primary identification while physiological and biochemical characteristics (lactose fermentation; production of CO 2 , H 2 S, indole, urease and oxidase; motility; and utilization of citrate) were used for secondary identification of isolates to species level as reported [20]. Discs combination method (ceftazidime 30 µg and cefotaxime 30 µg with and without clavulanic acid 10 µg) was used for phenotypic confirmation of ESBL production in E. coli, K. pneumoniae and K. oxytoca as recommended by Clinical and Laboratory Standards Institutes (CLSI) [21]. All isolates were archived in vials containing 20% glycerol in brain heart infusion (BHI; Oxoid, UK) broth and stored at − 40 °C untill molecular analysis.

Determination of minimum inhibitory concentration (MIC) of cefotaxime
The minimum inhibitory concentrations (MICs) of ESBL-GNB to cefotaxime were determined using agar incorporation method [22,23] on Mueller Hinton agar (MHA; Oxoid, UK) plates supplemented with 4 µg/mL, 8 µg/mL and 16 µg/mL cefotaxime. Inoculated plates were incubated in ambient air at 37 °C for 18-24 h. The MICs were recorded as greater than the highest concentration tested or the lowest concentration when no growth occurs on any of the agar plates.

DNA extraction and molecular detection of bla CTX-M-1 group
Out of 128 isolates, 98 ESBL-GNB (73 E. coli, 18 K. pneumoniae and 7 K. oxytoca) were selected and successful recovered for molecular characterization of the bla CTX-M-1 group. Selection of isolates for molecular characterization was limited by the availability of PCR reagents. The isolates were sub-cultured on plain MCA (Oxoid, UK) plates followed by crude DNA extraction using boiling method as previously described [24]. Out of five major phylogenetic groups of CTX-M genes (CTX-M-1, CTX-M-2, CTX-M-8, CTX-M-9, and CTX-M-25), we chose to test only for CTX-M-1 due to predominance of its members especially the bla CTX-M-15 and from insufficient resources. PCR amplifications of the bla CTX-M-1 group was carried out in thermal cycler machine (PCR Gene AmpR System) with primers CTX-M3G-F (5′-GTT ACA ATG TGT GAG AAG CAG) and CTX-M3G-R (5′-CCG TTT CCG CTA TTA CAA AC) and procedures reported previous [25]. PCR products were electrophoresed (at 110 V for 90 min) by using 2% agarose gel which was stained by SBR-Safe DNA gel stain (Ther-moFisher Scientific, UK) and visualized under UV light.

Quality control
Known ESBL-GNB from [26] and E. coli ATCC 25922 were used as control organisms.

Data analysis
STATA software version 13.0 was used for data analysis as per objectives of this study.

Socio-demographic and clinical characteristics of study participants
A total of 285 patients with median age (IQR: interquartile range) of 18  years were enrolled with the majority (53%, n = 151) being males ( Table 1).

Discussion
This study identified a high carriage of ESBL-GNB in Morogoro regional hospital and Mazimbu hospital. The overall prevalence (43.2%) observed in this study is comparable to a study in Gabon [27]. Although the carriage in our study is relatively lower compared to (50.4%) a study conducted at Tanzanian National Hospital, Muhimbili National hospital (MNH), in Dar es Salaam [28]. Being a national referral hospital, MNH receives patients with multiple antibiotics exposure from other healthcare facilities mainly regional and zonal referral hospitals, increasing the risk of carriage of ESBL-GNB. E. coli followed by K. pneumoniae are predominant ESBL producers colonizing patients. Similar findings were reported previous in Ethiopia, Turkey and other regions of Tanzania [15,16,29,30]. Pathogenic potential of E. coli (e.g., E. coli ST131) and K. pneumoniae (e.g., K. pneumoniae ST14), and frequent acquisition of conjugative plasmids encoding for antimicrobial resistance genes (ARGs i.e., ESBL genes) facilitates rapid exchange and dissemination of ARGs in E. coli and K. pneumoniae [31]. These isolates, ESBL-GNB, colonizing patients are potentially shaded of to contaminate patient's immediate inanimate surroundings as previous reported [32,33]   ESBL-GNB among vulnerable patients (immunocompromised and critically ill) associated with increased mortality from treatment failures and limited antibiotic therapeutic options [34,35]. Therefore, this study's findings alert for the strengthening of infections prevention and control measures and AMR surveillance in line with the Tanzania National Action Plan in order to combat AMR in the country in all tiers of health facilities [36].
This study found high proportion (94.9%) of ESBL-GNB carrying CTX-M-1 group genes colonizing patients. The CTX-M-1 group genes particularly bla CTX-M-15 are predominantly reported in clinical, colonization and environment isolates in Tanzania and elsewhere [6-8, 15, 16, 37]. Horizontal gene transfer (HGT) of mobile genetic elements (MGEs) including plasmids, transposons, and intergrons facilitates rapid dissemination and spreading of CTX-M-1 group genes, mostly in E. coli and Klebsiella spp., [8,[38][39][40] in the hospital environment. These findings hint the possibility of the common genetic elements or resistant strains carrying CTX-M-1 genes in healthcare facilities in Tanzania, necessitating the strengthening of IPC and antimicrobial stewardship in Tanzania.
Hospital admission, previous and current antibiotic use, and longer hospital stay significantly predicted carriage of ESBL-GNB. These findings are in consistency with other studies [5,16,27]. Hospital admission and longer stays increases the odds of being exposed to antibiotics mostly beta-lactams i.e., ampicillin and ceftriaxone as they make first-and second-lines of therapy [41]. Therefore, increasing antimicrobial selection pressure favoring the proliferation of resistant bacterial strains colonizing patients' gastro-intestinal tracts as observed in this study [42,43].
Antibiotic exposure creates essential pressure which select the small fraction of resistant bacteria of the intestinal microbiota therefore giving rise to the emergency and establishment of an entirely resistant population of bacteria [42]. With poor IPC practices especially in lowand middle-income countries, these superbugs may be cross-transmitted between patients resulting to subsequent invasive infections such as BSIs, UTIs and SSTIs. Presence of these bacteria in the gut and environment may also result in exchange of resistance genes to the highly virulent bacteria making the infection difficult to treat hence high morbidity and mortality [42].

Limitations
From limited funds and resources: ESBL isolates were conventionally identified to possible genus and species; agar dilution method was used to determine MICs for cefotaxime only; and other ESBL alleles contributing about 5-10% of ESBL genes in our setting and genetic relatedness of ESBL isolates were not determined.