https://orcid.org/0000-0003-1510-8711
Figures
Abstract
Background
Bloodstream infection due to beta-lactamase and carbapenemase-producing gram-negative bacteria poses a substantial challenge to the effectiveness of antimicrobial treatments. Therefore, this study aimed to investigate the magnitude of beta-lactamase, carbapenemase-producing gram-negative bacteria, and associated risk factors of bloodstream infections in patients at a tertiary care hospital, in Addis Ababa, Ethiopia.
Methods
An institutional-based cross-sectional study was conducted with convenience sampling techniques from September 2018 to March 2019. Blood cultures were analyzed from 1486 bloodstream infection suspected patients across all age groups. The blood sample was collected using two BacT/ALERT blood culture bottles for each patient. Gram stain, colony characteristics, and conventional biochemical tests were used to classify the gram-negative bacteria at the species level. Antimicrobial susceptibility testing was carried out to screen beta-lactam and carbapenem drug-resistant bacteria. The E-test was conducted for extended-spectrum-beta-lactamase and AmpC-beta-lactamase-producers. A modified and EDTA-modified carbapenem inactivation method was conducted for carbapenemase and metallo-beta-lactamases producers. Data collected using structured questionnaires and medical records were reviewed, encoded, and cleaned using EpiData V3.1. software. The cleaned data were exported and analyzed using SPSS version 24 software. Descriptive statistics and multivariate logistic registration models were used to describe and assess factors associated with acquiring drug-resistant bacteria infection. A p-value <0.05 was considered statistically significant.
Result
Among 1486 samples, 231 gram-negative bacteria were identified; of these, 195(84.4%) produce drug-hydrolyzing enzymes, and 31(13.4%) produce more than one drug-hydrolyzing enzyme. We found 54.0% and 25.7% of the gram-negative bacteria to be extended-spectrum-beta-lactamase and carbapenemase-producing, respectively. The extended-spectrum-beta-lactamase plus AmpC-beta-lactamase-producing bacteria account for 6.9%. Among the different isolates Klebsiella pneumonia 83(36.7%) was the highest drug-hydrolyzing enzyme-producing bacteria. Acinetobacter spp 25(53.2%) was the most carbapenemase producer. Extended-spectrum-beta-lactamase and carbapenemase-producing bacteria were high in this study. A significant association between age groups and extended-spectrum-beta-lactamase producer bacterial infection was seen, with a high prevalence in neonates (p = <0.001). Carbapenemase showed a significant association with patients admitted to the intensive care unit (p = 0.008), general surgery (p = 0.001), and surgical intensive care unit (p = 0.007) departments. Delivery of neonates by caesarean section, and insertion of medical instruments into the body were exposing factors for carbapenem-resistant bacterial infection. Chronic illnesses were associated with an extended-spectrum-beta-lactamase-producing bacterial infection. Klebsiella pneumonia and Acinetobacter species showed the greatest rates of extensively drug-resistant (37.3%) and pan-drug-resistance (76.5%), respectively. According to the results of this study, the pan-drug-resistance prevalence was found to be alarming.
Conclusion
Gram-negative bacteria were the main pathogens responsible for drug-resistant bloodstream infections. A high percentage of extended-spectrum-beta-lactamase and carbapenemase-producer bacteria were found in this study. Neonates were more susceptible to extended-spectrum-beta-lactamase and AmpC-beta-lactamase-producer bacteria. Patients in general surgery, caesarean section delivery, and intensive care unit were more susceptible to carbapenemase-producer bacteria. The suction machines, intravenous lines, and drainage tubes play an important role in the transmission of carbapenemase and metallo-beta-lactamase-producing bacteria. The hospital management and other stakeholders should work on infection prevention protocol implementation. Moreover, special attention should be given to all types of Klebsiella pneumoniae and pan-drug resistance Acinetobacter spp transmission dynamics, drug resistance genes, and virulence factors.
Citation: Beshah D, Desta AF, Woldemichael GB, Belachew EB, Derese SG, Zelelie TZ, et al. (2023) High burden of ESBL and carbapenemase-producing gram-negative bacteria in bloodstream infection patients at a tertiary care hospital in Addis Ababa, Ethiopia. PLoS ONE 18(6): e0287453. https://doi.org/10.1371/journal.pone.0287453
Editor: Melaku Ashagrie Belete, Wollo University, ETHIOPIA
Received: February 7, 2023; Accepted: June 6, 2023; Published: June 27, 2023
Copyright: © 2023 Beshah et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This work was supported by the Addis Ababa University vice president for research and technology transfer office thematic research funding program with Grant numbers awarded to the author - VPRTTPY-0402018. The funders had no role in study design, data collection, and analysis, the decision to publish, or the preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Gram-negative bacteria (GNB) are one of the major global health concerns due to beta-lactamase and carbapenemase production, which leads to antimicrobial resistance (AMR) [1, 2]. It has become an important cause of morbidity and mortality worldwide [3, 4]. In 2050, multidrug-resistant bacteria are expected to be the cause of 10 million annual deaths [5]. Over 19% of nosocomial infections are caused by extended-spectrum beta-lactamase (ESBL) enzyme-producing bacteria. The ESBL-producing bacteria were also expected to cause a 57% higher mortality rate among bloodstream infections (BSI) than a strain that doesn’t produce ESBL [6]. Production of ESBL and carbapenemase are causes of drug resistance, particularly in GNB, and these pauses serious challenges to the effectiveness of contemporary antibiotic treatment [6, 7]. The ESBL enzyme can hydrolyze penicillin, monobactam, and third-generation cephalosporins. AmpC beta-lactamase (AmpC-BL) can also break down penicillins, third-generation cephalosporins, and cephamycin, while carbapenemase and metallo-beta-lactamase (MBL) can break down carbapenems drugs [6–9]. Recently, it has been observed that the multi-drug resistance (MDR) problem has shifted from gram-positive bacteria to GNB, which in turn worsens the problem by the dearth of new antimicrobial drugs having an effect against them [10]. The GNB, particularly ESBL and carbapenemase-producing Klebsiella pneumoniae, Escherichia coli, Acinetobacter baumannii, Proteus mirabilis, and Pseudomonas aeruginosa, showed increasing drug-resistance mechanism [4]. Research conducted in an Ethiopian context showed that there is a significant association between death and AMR in patients with BSI [11]. Similar research had also reported a prevalence of ESBL and AmpC-BL in GNB to be 38.8% and 2.4% [12], 67% and 2% [13], 51% and 25% [14], and in Enterobacteriaceae specifically, it was reported to be 78.6% and 12.1% [15], respectively. Another research conducted on GNB in an Ethiopian context has also shown a magnitude of 94.5% of the isolates to be MDR of which 8.8% and 0.8% were due to extensively drug-resistance and pan-drug resistant, respectively [13]. However, most of these studies did not address the magnitude of ESBL and carbapenem-producing GNB and the associated factors among all age groups of BSI in tertiary hospitals. Therefore, the objective of this study was to address the identified gaps. Mainly the study aimed to show the magnitude of ESBL, AmpC-BL, MBL, and Carbapenemase-producing GNB and associated risk factors among bloodstream infections patients in Tikur Anbessa Specialized Hospital (TASH), Addis Ababa, Ethiopia.
Materials and methods
Study design, study area, study population, and sample size
An institutional-based cross-sectional study was conducted from September 2018 to Mar 2019 in TASH which is a tertiary hospital located in Addis Ababa having over 700 beds, 201 doctors, 627 nurses, more than 115 other health professionals, and 1300 administrative staff that are dedicated to providing healthcare services [16]. The study population consists of all patients who have visited TASH for medical care during the study period. Convenience sampling techniques were deployed to all age groups of BSI-suspected patients who agreed to offer a blood sample and sign a consent form to participate in the study. Patients that were not willing to participate in the study and had been on antibiotic treatment within ten days were excluded. A total of 1486 patients were enrolled in this study and the sample size calculation was based on a single sample size estimation [17, 18] with a prevalence of 32.8% taken from the study conducted in Addis Ababa, Ethiopia [19]. A sample size (n) calculation with n = Z2p (1 –p)/d2, and n = 1.9620.328(1–0.328)/(0.025)2 = 1354.80 with 10% contingency gives the size to be = 1490. To increase the accuracy of the result a margin of error (d) = 0.02 was used to maximize the sample size. Z stands for Z score with a level of confidence of 95%, which is a conventional value with a Z value of 1.96.
Data collection
Patient demographic data and clinical information were collected using a standardized questionnaire and review of patient medical records by trained nurses. Blood samples were collected by qualified professional nurses and laboratory technologists who had prior experience in collecting research data and blood samples. All data and sample collectors were trained by the principal investigator on the detailed instructions of using prepared standardized questionnaires, blood sample collection procedures, and sample transportation to the laboratory. Follow-up of microbial growth in the blood culture bottle and the downstream steps were accomplished by the recruited professional microbiologists and the principal investigator in the laboratory.
Sample collection
All age groups of hospitalized patients with a bloodstream infection who signed an informed consent form and were willing to participate in the study were included. The conditions of bloodstream infection were defined by the 1991 consensus conference, which was later updated in 2001 [20, 21]. We classified eligible patients as bloodstream infection cases based on the attending physician’s judgment. Blood samples were collected aseptically from bloodstream infection suspected patients who meet the criteria set in this study. The vein puncture site was disinfected with 70% alcohol and 2% tincture of iodine before blood collection. Approximately 1–4 ml of blood from pediatrics (less than 12 years) and 5–10 ml of blood from adults (greater than 12 years) were taken and inoculated into a ready-made BacT/ALERT blood culture bottle [22]. Two aerobic BacT/Alert FA Plus and BacT/Alert PF Plus bottles were used for blood culture, among adults and pediatrics, respectively. Similar growth in both bottles was considered positive. Blood culture bottles were incubated with an automatic BacT/ALERT® 3D machine at 37°C of 5% CO2 gas concentration which was inspected daily for the sign of bacterial growth for up to 5 days [23, 24].
Bacteria identification
Blood samples that were turbid before five days and blood samples that were not turbid at five days were sub-cultured on blood agar (Oxoid, England) (5% sheep blood plates), MacConkey agar (BD, USA), and chocolate agar (Oxoid, England), at 37°C for 24 hours. The McConkey plates were incubated in aerobic conditions, while chocolate and blood agar plates were incubated in microaerophilic conditions (containing 5% CO2) [25]. Bacterial pathogens were identified using CLSI 2017 standard bacteriological procedures [26]. Hence a positive culture from blood samples was characterized by their colony characteristics and gram stain. The GNB bacteria were further classified at the species level by using conventional biochemical tests including indole, urea agar, triple sugar iron agar, citrate utilization, mannitol, malonate, motility, lysine decarboxylase, and oxidase test [27].
Antimicrobial susceptibility testing
Antimicrobial susceptibility tests (AST) were performed by using disk diffusion and the susceptibility range was determined based on CLSI guidelines [26]. With that, each isolate was spread onto a Mueller–Hinton agar plate. The antimicrobials for disc diffusion testing was obtained from BD and Oxoid for GNB with the following concentrations: for GNB Amikacin (AN) (30 μg), Amoxicillin + clavulanic acid (AMC) (20/10μg), Ampicillin–Sulbactam (SAM) (10/10 μg), Ampicillin (AM) (10 μg), Azithromycin (ATM) (30 μg), Cefepime (FEP) (30 μg), Cefotaxime–clavulanate (CTL) (30/20μg), Cefotaxime (CTX) (30 μg), Cefotetan (CTT) (30 μg), Cefoxitin (FOX) (30 μg), Ceftaroline (CPT) (30 μg), Ceftazidime–clavulanate (CAL) (30/10 μg), Ceftazidime (CAZ) (30 μg), Ceftriaxone (CRO) (30 μg), Cefuroxime (CXM) (30 μg), Chloramphenicol (C) (30 μg), Ciprofloxacin (CIP) (5 μg), Dorsapenem (DOR) (10 μg), Doxycycline (DOX) (30 μg), Ertapenem (ETP) (10 μg), Gentamicin (GM) (10 μg), Imipenem (IMP) (10 μg), Levofloxacin (LEV) (5 μg), Meropenem (MEM) (10 μg), Piperacillin–Tazobactam (TZP) (100/10μg), Tetracycline (TET) (30 μg), Tobramycin (NN) (10 μg), and Trimethoprim-Sulfamethoxazole (SXT) 1(.25/23.75μg) [26, 28].
ESBL primary screening
Initial screening for ESBL was conducted by the diameter zone of inhibition produced by Ceftazidime (30μg), Ceftriaxone (30μg), and Cefotaxime (30μg) which is found to be within the clinical and laboratory standard institute (CLSI) screening criteria. These breakpoints indicative of ESBL production were CAZ ≤ 22mm, CRO ≤ 25 mm, and CTX ≤ 27mm. Phenotypic detection of ESBL production was confirmed by a double-disk synergy test and combined disk test using an E-test according to EUCAST and CLSI guidelines respectively [26, 29].
Double disk synergy test (DDST) for ESBL detection.
The organism to be tested were inoculated onto a Mueller–Hinton agar plate. The antibiotic disks that deployed for this purpose were Ceftriaxone (30 μg), Cefotaxime (30 μg), Ceftazidime (30 μg), Aztreonam (30μg), and Amoxicillin/ Clavulanic acid (20/10 μg). The four antibiotics were then placed at distances of 20 mm (edge to edge) from the Amoxicillin/Clavulanic acid disk, which was placed in the middle of the plate. After 24-hrs of incubation, the occurrence of an enhanced zone of inhibition between either of the cephalosporin antibiotics and the Amoxicillin/Clavulanic acid disk was considered ESBL positive [26]. The primary AST screening has integrated the DDST and later on the confirmatory E-test for ESBL and AmpC-BL identification.
E-test for both ESBL and AmpC beta-lactamase detection.
The E-test was conducted for ESBL detection using Ceftazidime (CAZ) 0.5–32 μg/ml alone with Ceftazidime+ Clavulanic acid (CAL) 0.064–4 μg/ml combination, and Cefotaxime (CTX) 0.25–16 μg/ml alone with Cefotaxime+ Clavulanic acid (CTL) 0.016–1.0 μg/ml combination strip were used for phenotypic confirmation. For detection of ESBL, CTX ≥ 0.5 and CTX/CTL ratio ≥ 8 or CAZ ≥ 1 and CAZ/CAL ratio ≥ 8 were considered as positive, whereas CTX < 0.5 or CTX/CTL ratio < 8 and CAZ < 1 or CAZ/CAL ratio < 8 was considered as negative. The AmpC-BL detection, hence, was done by using cefotetan 0.60–32μg/ml alone with cefotetan-cloxacillin 0.5–32μg/ml combination. The ratio of CTT/CXT ≥8 indicates the production of an AmpC BL [30–32].
Carbapenem-resistant detection
The presence of Carbapenemase detection was implemented by using the modified Carbapenemase inactivation method (mCIM) and EDTA-modified Carbapenemase inactivation method (eCIM), except for the case of Acinetobacter spices. This was conducted for all the Carbapenem-resistant or intermediate with either Imipenem, Meropenem, Dorsapenem, or Ertapenem except for none fermenter GNB where Ertapenem was not used. To perform the mCIM, a suspension was made by taking 1μl loopful of bacteria from an overnight blood agar plate and then adding it to a 2 ml trypticase soy broth. Subsequently, Meropenem (10 μg) disc was immersed in the suspension and incubated for 4 hours ± 15 minutes at 37°C. After the incubation, the disc was removed from the suspension using 10 μl inoculation loops and placed on a Mueller-Hinton agar plate inoculated with a susceptible E. coli indicator strain (ATCC 29522). Then, the results reading was recorded after 18–24 hours of incubation at 37°C. Consequently, for the eCIM case taking 1μl loopful of bacteria from an overnight blood agar plate and then adding it to 2 ml having 5 molar EDTA concentration trypticase soy broth. Then continued the mCIT test procedure for the identification of MBL production. For mCIM zone of inhibition of 6 to 15 mm around a disk and a zone of inhibition of 16 to 18 mm around a disk including the presence of colonies in the zone of inhibition was considered as positive. A zone of inhibition with a distance of 19 mm or greater around a disk was considered negative. A negative result suggests that the isolate is resistant to carbapenems by a mechanism other than the production of carbapenemase. If mCIT shows carbapenem production and an increase diameter of eCIT greater or equal to 5mm then mCIT was considered as MBL positive [26, 29, 33, 34].
Determination of MDR, XDR, and PDR isolates
Multidrug resistance (MDR), extensively drug-resistant (XDR), and pan-drug-resistant (PDR isolates were categorized by using Magiorakos et al. standard definitions for drug resistance bacteria [35].
Quality control
Standard operating procedures were strictly followed to verify that quality control parameters were fulfilled. Visual inspections of cracks in media, unequal fill, hemolysis, evidence of freezing, bubbles, and contamination were performed. According to standard operating procedures, blood samples were drawn and then transferred to blood culture broth and immediately transported to the microbiology laboratory [22, 24]. Quality control was performed to check the quality of the media. The resistance and susceptibility were interpreted according to the CLSI laboratory standards. Escherichia coli (ATCC 25922), Staphylococcus aureus (ATCC 25923), and Pseudomonas aeruginosa (ATCC 27853) were used as reference strains for culture and susceptibility tests [26]. Klebsiella pneumoniae ATCC® 700603 and E. coli ATCC® 35218 were used for positive control and negative control for E-test. Klebsiella pneumoniae ATCC® 700603 (ESBL positive) was used as a negative control and Klebsiella pneumoniae ATCC® BAA-1144 was used as a positive control for E-test AmpC detection. Klebsiella pneumoniae ATCC® BAA-1705™ and Klebsiella pneumoniae ATCC® BAA-1706 were used as positive and negative controls for the carbapenemase test respectively [31, 32]. These strains were obtained from the TASH.
Statistical analysis and interpretation
Data collected using a structured questionnaire was encoded and cleaned using EpiData V3.1. software. The cleaned data were exported and analyzed using SPSS version 24 software. The collected quantitative data were analyzed using simple descriptive statistics i.e., percentages, frequency, mean, and standard deviation moreover, the data were analyzed using univariate, and multivariate logistic registration models to assess dependent and independent variables association factors. A p-value of less than 0.05 was considered a statistically significant association. Finally, the results were tabulated hereunder.
Ethics approval and consent to participate
The study was approved by “The College of Natural and Computational Sciences Institutional Research Ethics Review Board (CNS-IRB) on 30/03/2018 with minutes of meeting reference no IRB/032/2018 at Addis Ababa University”. Permission was also obtained from the College of Health Sciences Institutional Review Board (IRB). This study procedure has fulfilled the requirements set by the Declaration of Helsinki [36]. For participants whose age was below 18 years, data were collected with the assent (12–18) and consent of the patient’s legal guardians. However, in the cases of above 18 years of age, we have collected consent from the patients themselves. The purpose and procedures of the study were explained to the participants’ parents, and/or guardians during the study period. Those patients who were given informed consent were selected and enrolled as the study participants and patient results were communicated to the attending physicians.
Results
Culture result
Among 1486 BSI-suspected patient samples, 28.1% were culture positive. Most of the bacteria 233 (54.2%) isolated were GNB. The detail of the bacteria identified and the antimicrobial susceptibility pattern of the isolates was described in our published article [37]. After excluding one Corynebacterium gleum and one Neisseria meningitidis, a total of 231(54.1%) GNB bacteria were further analyzed for determining their drug resistance mechanisms. Among the 231 GNBs, 84.4% of the isolates were positive for one or two types of drug hydrolyzing enzyme. Out of the total 231 identified GNB, Klebsiella pneumoniae (32.5%), Acinetobacter spp (20.4%), and Escherichia coli (16.5%), were the predominant isolates. A total of 71% of bacteria produced one of the drug hydrolyzing enzymes. Of these 13.4% GNB produced a double drug hydrolyzing enzyme, with ESBL + AmpC-BL producers accounting for 6.9%, ESBL + carbapenemase producers for 5.2%, and ESBL + MBL producers for 1.3%. Of the GNB that produced double drug hydrolyzing enzymes Klebsiella pneumoniae and Acinetobacter spp account for 35.5% and 25.8%, respectively as shown in Table 1.
Carbapenemase, MBL, ESBL, and AmpC-BL, producer gram-negative bacteria
Out of the total 231 gram-negative bacteria ESBL, Carbapenemase, AmpC-BL, and MBL producers accounted for 52.8%, 25.1%, 14.3%, and 5.6%, respectively. Most Klebsiella pneumoniae (77.3%) and Klebsiella oxytoca (77.8%) were ESBL producers likewise, most Acinetobacter spp (53.2%) were carbapenemase producers (Table 2).
Factor associated with carbapenemase and MBL-producing gram-negative bacterial infection
From the 417(28.06%) culture-positive patients, 224(53.72%) were infected with GNB. Of these, 189(84.4%) isolated s bacteria showed the production of one or two drug-hydrolyzing enzymes (S1 Table). However, in cases of polymicrobial growth and double drug hydrolyzing enzyme production, the higher-grade drug hydrolyzing enzyme production was considered for the associated factor analysis.
The multivariate regression analysis of the study revealed significant associations between carbapenemase and MBL producers of GNB and the departments where the patient was admitted. Among the departments where the patients were admitted, general surgery (p = 0.001), surgical intensive care unit (SICU) (p = 0.007), and intensive care unit (ICU) (p = 0.008) were sites where patients will acquire almost five times and two times more likely to acquire carbapenemase and MBLs producer bacteria than those who were admitted in other wards. A significant association to acquire infection with carbapenemase and MBLs producing GNB and care that involves the insertion of medical instruments had been observed (p = 0.001). Moreover, patients in whom medical instruments were inserted during medical care were almost four times more likely to acquire infection with carbapenemase and MBLs-producing bacteria than those who don’t require instrumentation. The use of intravenous lines (p = 0.045), drainage tubes (p = 0.032), and the use of suction machines (p = 0.040) were also associated with carbapenemase and MBL-producing GNB infection.
Chronic wound infection on the skin (p = 0.030), and urinary tract infection (UTI) (p = 0.007), had a significant association with carbapenemase and MBL producers GNB infection (Table 3). Patients with wound infection on the skin and ureteral tract infection were almost 4 times more likely to acquire carbapenemase and MBLs producer bacteria than those not have wound and ureteral tract infected patients (Table 2).
Mode of delivery of neonates was associated with carbapenemase and MBL producer bacterial infection (p = 0.036), caesarean section delivery was almost 10 times more likely to result in acquiring these bacteria than normal delivery (Table 4).
Factor associated with ESBL and AmpC-BL producer gram-negative bacterial infection
The multivariate regression analysis of the study demonstrated significant associations between ESBL and AmpC-BL-producing GNB infection and insertion of instruments into the body during medical care (p = <0.001) and chronic illnesses (p = 0.027). Patients having UTI were almost 4 times more likely to acquire ESBL and AmpC BLs producer bacteria than those who don’t have UTI (p = 0.001). Again, patients having respiratory tract infection (RTI) (p = 0.001) and tissue sarcoma (p = 0.018) have a double chance to acquire ESBL and AmpC-BL-producing bacterial infection than those who don’t have them. There was a significant association between the age group and ESBL and AmpC-BL-producing GNB infection (p< 0.0001). Age less than one month was 3.5 times more likely to acquire ESBL and AmpC-BL-producing GNB infection than > 45 years of age (p = <0.001). There was a significant association between ESBL and AmpC-BL-producing GNB infection and patients’ admission departments (p< 0.0001). Among patients admitted in wards of B6 (NICU) (p = <0.001), D7 (Pediatrics Hematology) (p = 0.044), and B7 (pediatrics, medical & surgery) (p = 0.014) were almost 15, 2.5 and 4.5 times more likely to acquire ESBL and AmpC producer bacteria than those admitted in other wards., respectively (Table 5).
Neonatal incubation has an association (p = 0.006) with ESBL and AmpC-BL-producing bacterial infection (Table 6).
Factors associated with MDR gram-negative bacterial infection
A total of 215(51.56%) patients who acquired MDR GNB were analyzed for the associated factors with the infection acquired. The multivariate regression analysis of the study revealed that patient admission department (MICU(p = 0.027), NICU (p = 0.023), General Surgery (p = 0.001)), instrument usage during medical care (p = <0.001), (an intravenous line (p = 0.029), and drainage tube (p = 0.033)) and having a chronic illness (p = 0.041) like RTI (p = 0.027) have significant proportion of MDR GNB bacteria. Neonates with a history of neonatal incubation were identified to have the possibility of acquiring MDR bacteria three times more than neonatal patients who are not treated in neonatal incubation. Chronically ill patients, those with UTIs were five times more likely to acquire MDR bacteria than those without UTIs. Neonates delivered by Caesarean section were more likely to acquire MDR bacteria than those delivered with normal vaginal delivery (S2 Table).
Discussion
Comprehending the so-called "nightmare" of MDR bacteria is key to the successful treatment and control of hospital-acquired infection [38]. Therefore, this study was intended to identify the magnitude of MDR and their phenotypic drug-resistance mechanisms including the production of ESBL, AmpC-BL, MBL, and Carbapenemase along with their associated factors among patients with GNB bloodstream infection in a tertiary care hospital in Ethiopia. A high percentage of ESBL + AmpC-BL bacteria (67.1%) were identified meanwhile carbapenemase + MBL was found to be 30.7%. In this study, the rate of BSI was 28.1% which is comparable to other related studies conducted in Ethiopia with prevalence rates of 28.0% [39], 27.9% [40], and 25.8% [41]. However, the current result was higher than rates reported elsewhere18.2% [42], 15.8% [43], and lower than other studies conducted in Ethiopia which were 32.8% [19], and 39.2% [44]. A lower percentage was also reported from Bangladesh 13.6% [45]. Several confounding factors including differences in infection prevention practices, crowdedness of the hospitals, study settings, study’s design, study area, patient type, clinical situation, etiological factors, and seasonal variations could account for the differences observed.
The higher prevalence of GNB (54.2%) was also reported in the present study, which accords with earlier Ethiopian studies [44] 52.3%, [40] 51.8%; however, it was higher than other reports from the country such as 31% [42], and 22.6% [19]. The GNBs are becoming the major etiology of BSI in hospitalized patients [44, 46, 47]. In this research, Klebsiella pneumoniae, Acinetobacter spp., and Escherichia coli were the three most common pathogens isolated. This was similar to a study from Paris and CDC reports that also identified these three pathogens as the most common isolates [48, 49]. This could be due to the sharing of drug-resistant genetic materials associated with ESBL, AmpC-BL, and Carbapenemase-producing genes among these bacteria [4, 44, 46, 47].
The total percentage (84.4%) of bacteria in the current study that produced carbapenemase, MBL, ESBL, and AmpC-BL was higher than the study done in Addis Ababa Ethiopia (70%) [13]. However, this was lower than the study conducted in Nigeria (98.8%) [50]. Of 195 drug resistance enzyme-producing bacteria, we found 15.9% that produced two or more enzymes which were lower than the Nigerian report showing co-production of ESBL, AmpC-BL, and/or Carbapenemase in 69.6% of isolates [50]. Whereas when it comes to the rate of single enzyme production our study showed ESBL (52.8%), carbapenemase (25.1%), AmpC-BL (14.3%), and MBL (5.6%) producers which was comparable with the study from Nigeria with rates for ESBL (53.9%) and AmpC-BL (20.9%). The carbapenemase and MBL-producing bacteria in this study were 30.7%, which was higher than studies conducted in Nigeria that was 25.2% [50], and Ethiopia which was 2% [13] and 12.12% [15]. These indicate that the prevalence of GNB-producing drug-hydrolyzing enzymes is increasing over time. Carbapenemases, with versatile hydrolytic capacity against β-lactams, are now an important mechanism of drug resistance in GNB [51]. In this study, 30.7% of carbapenemase-producing GNB was higher than that of a study done in Latin America with 20.8% [52]. This could be due to the high percentage of Acinetobacter baumannii which accounts for 76.5% of the total possible PDR and 63.8% of Acinetobacter spp isolated in this study were carbapenem drug-resistant. The genes encoding for the acquired carbapenemase production were associated with a high potential for dissemination [51]. Carbapenem-resistant Acinetobacter spp. that carry mobile genetic elements were associated with nosocomial infection in our study settings and some of these bacteria were pan-drug-resistant [34]. Therefore, special attention must be given in our hospital to all types of Klebsiella pneumoniae that were associated with carbapenem resistance and possible PDR Acinetobacter spp identified, more specifically about their transmission dynamics.
The prevalence of ESBL and AmpC-BL-producing bacteria was 67.1%, comparable to a study conducted in Addis Ababa, which had reported 67% [13] and 78.6% [15]. In our study, the percentage (52.8%) ESBL producing GNB isolated from patients with BSI was similar to the nine-year retrospective study conducted in Ghana which accounts for 50.2% [53]. A lower result was also reported by previous similar studies conducted in Ethiopia (38.6%) [12], Asia (47%) [54], and Thailand (37.8%) [55]. The difference could account for differences in the research methodology, sample type, geographical variation, and hospital infection prevention practice.
In our study, Klebsiella pneumoniae was the highest drug-resistance enzyme-producing bacteria accounting for 36.7% followed by Acinetobacter (21.7%). The Acinetobacter spp. was the high producer of carbapenemase and MBL producers with 53.2% and 10.6% prevalence, respectively. According to a study conducted on urine samples in Nigeria, the percentage of Klebsiella pneumoniae strains resistant to cephalosporins due to ESBL production increased from 18.2% in 2010 to 29.4% in 2011 and further to 34.1% in 2012 [56]. A study conducted in Kenya revealed Klebsiella pneumoniae, E. coli, and Acinetobacter baumannii had resistance to cephalosporins with a range of 57–84%, 16–43%, and 81%, respectively [57]. The systematic review of 20 years conducted in China on BSI hospitalized non-ICU patients with BSI had shown carbapenem-resistant Klebsiella pneumoniae to be below 5% in the years 1998–2012 but this figure had increased to 34.9% during 2013–2017. In contrast, sadly, the detection rate for carbapenem-resistant Klebsiella pneumoniae in ICU patients has increased from 0% in 2013 to 75% in 2016 [58].
Having a double drug resistance enzyme-producing bacteria implies having a high drug resistance capacity for the bacteria and this in turn worsens the success rate of the treatment to the worst level. In this study, about 15.9% of the isolated bacteria produced more than one drug hydrolyzing enzyme that was higher than that of China, which reported 7% of ESBL-producing Klebsiella pneumoniae isolates manifested “cross-resistance to carbapenems [56]. In this research, Klebsiella pneumoniae, Acinetobacter spp., and Klebsiella oxytoca were the three dominant bacteria that produced more than one drug hydrolyzing enzymes. This implies that Klebsiella pneumoniae, Klebsiella oxytoca, Acinetobacter spp, and Escherichia coli were becoming aggressive drug resistance bacteria. In addition to this, having more than one drug-hydrolyzing enzymes strengthens the bacterial pathogens to be more resistant to antibiotics, increased treatment failure rate, longer hospital stays, high economic burden, and unpredicted emotional costs to patients.
The higher percentage of Carbapenemase and MBL producer GNB reported in general surgery and SICU department, were almost five times more likely to be contagious than other admission departments. Neonates delivered by caesarean section were almost 10 times more likely to acquire carbapenemase and MBL producer bacteria than those delivered by normal procedure. In recent years, the rate of caesarean section happening is increasing, and whether that is indicated or not is an issue but the fact that the procedure is a risk factor that exposes the neonates to MDR bacteria calls for serious attention in the national maternal health programs. Patients that pass through surgical procedures including caesarean section had an association with carbapenem infections. These results will be directly associated with a lack of infection prevention protocol implementation in the caesarean section room and sterilization of medical instruments used during surgical procedures. Fumigation of the room, proper sterilization technique for the instruments, and standard surgical procedure should be implemented. These factors were also found in another study conducted in Latin America where ICU admission and general surgery have an association with carbapenem drug-resistant bacterial infection [52].
The other major factors associated with carbapenem resistance bloodstream infection in the current study were the usage of instruments especially inserted devices during medical care were almost four times more likely to be infectious for carbapenemase and MBL producer bacteria. Usage of antibiotic treatment ten days before sample collection was also associated with carbapenemase and MBL producer bacteria. The misuse of antibiotics will be one of the causes of this problem. From patients with chronic illness, patients with wound infection on the skin and UTI were almost 4 times more likely to acquire carbapenemase and MBL producer bacteria than those that do not have a wound and UTI. Wound infection on the skin was one of exposing factors for nosocomial infection due to the lack of mechanical barrier of the skin. The association between wound infection and UTI with Carbapenemase producer GNB BSI was also reported in another study [52]. The instruments used for medical care like a suction machine, an intravenous line, and a drainage tube were associated with Carbapenemase and MBL-producing bacterial infection. Therefore, surgical materials and indwelling medical equipment during medical care should be monitored for proper sterilization.
The NICU wards were almost 15 times more likely to acquire ESBL and AmpC-BL producer GNB in this study. The room temperature which is favorable for microbial growth, high workload, and crowdedness could be a major problem in NICU BSI. In addition, plasmid transfer genes that play a great role in nosocomial infection are common in GNB, especially in E. coli and Klebsiella species found in these wards. As per the research conducted in 11 African countries among neonates, GNB was the predominant cause of early-onset neonatal sepsis, with a high prevalence of ESBL-producing bacteria [59]. This implies that NICU was more affected by third-generation cephalosporin-resistant bacteria meanwhile, the caesarean section delivery room was more affected by Carbapenem-resistant bacteria.
Among chronic illness patients those having UTI were almost 4 times more likely to acquire ESBL and AmpC-BL producer bacteria than those who don’t have UTI. The patients with respiratory tract infection(RTI) and tissue sarcomas were also two times more likely to acquire ESBL and AmpC-BL producers than the others. Immune suppression and prolonged admission in chronic illness patients could be the exposing factors for hospital-acquired infection.
The prevalence of MDR, XDR, and PDR was indicated in our previous article [37]. The ICU department, neonate’s instrument usage during medical care, chronic illness, UTI, RTI, and history of incubation was the predominant factor for having the highest MDR GNB infection. Another study conducted in Australia showed a three-fold increased risk of death in the ICU department [60]. Likewise, another study in the USA, Australia, and India showed that MDR-GNBs were typically isolated in ICUs [61–63]. Our result is also supported by the result of the study conducted in Vellore, South India, Ghana, and Australia where neonatal ICU admission has an association with MDR infection [53, 60, 62]. The study done in Greece showed that PDR in GNB was typically isolated in ICUs [64]. Surgery and surgical admission were identified as independent risk factors in America and Australia for MDR-BSI [60, 65]. A two-fold exposure to MDR infection was also found in patients with a medical instrument used during medical care [62]. A study conducted in Thailand showed RTI and wound infection as significant risk factors for MDR-GNB infection [55]. Another research conducted in the USA also indicated the usage of urinary catheters as a risk factor for the development of MDR in GNB [66]. The neonatal incubation and mode of delivery by caesarean section had three and two-fold exposure to MDR BSI [51, 67]. From this result, we can recommend sterilization of the medical equipment used for the ICU and surgical procedures, and the infection prevention protocol should be strengthened.
Conclusion
In this study, a large number of GNBs were drug-resistant enzyme producers. Two or more drug-resistant enzymes were produced by 15.9% GNB. High carbapenemase and MBL producer bacteria were found in this study. Most patients in neonatal ICUs were susceptible to ESBL and AmpC-BL-producing GNB, whereas, in general surgery, caesarean section delivery room, and SICU patients were more susceptible to carbapenemase-producing GNB. The suction machines, intravenous lines, and drainage tubes were the major cause of acquiring carbapenemase and MBL producer bacteria. Among chronic illnesses, wound infection, and UTI were associated with carbapenemase and MBL producer bacteria. The ESBL and AmpC-BL producers were also associated with UTI, RTI, and tissue sarcomas. The 7.4% of possible PDR bacteria was an alarming result in this study. Klebsiella pneumoniae and Acinetobacter spp. were the most prevalent bacteria with XDR and possible PDR, respectively. The hospital management would extend their effort to address the concern of infection prevention protocols in the NICU department in particular and the hospital in general. Special attention should be given to all types of Klebsiella penumoniae and possible PDR Acinetobacter spp transmission dynamics, drug resistance genes, and virulence factors.
Limitation
The obligate anaerobes were not identified in this study. We used conventional biochemical tests rather than API 20 or MALDI-TOF mass spectrometry because of unavailability and high expense.
Supporting information
S1 Table. Gram-negative bacteria and their enzyme-producing drug resistance mechanism from BSI suspected patients.
https://doi.org/10.1371/journal.pone.0287453.s001
(DOCX)
S2 Table. Factors associated with MDR prevalence crude and adjusted odds ratio of BSI suspected patients.
Bivariate and multivariate logistic regression of demographic and other parameters, taken as predictive variables for BSI patients with MDR acquired 215 patients compared with 1271 BSI patients.
https://doi.org/10.1371/journal.pone.0287453.s002
(DOCX)
Acknowledgments
Tikur Anbessa Specialized Hospital microbiology laboratory staff Mequaninit Mitiku, Tesfaye Seboka, Fuirehiwot Teklehaymanot, and Manie Asres. Addis Ababa University College of Natural and Computational Science Microbial Cellular Molecular Biology Infection Biology stream for coaching and mentorship.
References
- 1. Vasoo S, Barreto JN, Tosh PK. Emerging issues in gram-negative bacterial resistance: an update for the practicing clinician. Mayo Clinic proceedings. 2015;90(3):395–403. Epub 2015/03/07. pmid:25744116.
- 2. Elamreen FA. Neonatal sepsis due to multidrug-resistant Klebsiella terrigena in the neonatal intensive care unit in Gaza, Palestine. Critical Care. 2007;11(4):P12.
- 3. Gaynes R, Band J. Epidemiology, pathogenesis and microbiology of intravascular catheter infections. UpTo Date versión; 2009.
- 4. Timsit JF, Soubirou JF, Voiriot G, Chemam S, Neuville M, Mourvillier B, et al. Treatment of bloodstream infections in ICUs. BMC Infect Dis. 2014;14(1):489. pmid:25431091; PubMed Central PMCID: PMC4289315.
- 5. Dadgostar P. Antimicrobial Resistance: Implications and Costs. Infect Drug Resist. 2019;12:3903–10. Epub 2020/01/08. pmid:31908502; PubMed Central PMCID: PMC6929930.
- 6.
CDC. Antibiotic resistance threats in the United States, 2013: Centres for Disease Control and Prevention, US Department of Health and Human Services; 2013.
- 7. Shaikh S, Fatima J, Shakil S, Rizvi SM, Kamal MA. Antibiotic resistance and extended spectrum beta-lactamases: Types, epidemiology and treatment. Saudi J Biol Sci. 2015;22(1):90–101. Epub 2015/01/07. pmid:25561890; PubMed Central PMCID: PMC4281622.
- 8.
CLSI. Performance Standards for Antimicrobial Susceptibility Testing. The Clinical and Laboratory Standards Institute (CLSI) 2018;31st Edition(Pennsylvania, US).
- 9. Jacoby GA. AmpC beta-lactamases. Clin Microbiol Rev. 2009;22(1):161–82, Table of Contents. Epub 2009/01/13. pmid:19136439; PubMed Central PMCID: PMC2620637.
- 10. Brusselaers N, Vogelaers D, Blot S. The rising problem of antimicrobial resistance in the intensive care unit. Ann Intensive Care. 2011;1:47. Epub 2011/11/25. pmid:22112929; PubMed Central PMCID: PMC3231873.
- 11. Seboxa T, Amogne W, Abebe W, Tsegaye T, Azazh A, Hailu W, et al. High Mortality from Blood Stream Infection in Addis Ababa, Ethiopia, Is Due to Antimicrobial Resistance. PLoS One. 2015;10(12):e0144944. pmid:26670718; PubMed Central PMCID: PMC4682922.
- 12. Tekele SG, Teklu DS, Tullu KD, Birru SK, Legese MH. Extended-spectrum Beta-lactamase and AmpC beta-lactamases producing gram negative bacilli isolated from clinical specimens at International Clinical Laboratories, Addis Ababa, Ethiopia. PLoS One. 2020;15(11):e0241984. Epub 2020/11/13. pmid:33180785; PubMed Central PMCID: PMC7660541 work.
- 13. Beyene D, Bitew A, Fantew S, Mihret A, Evans M. Multidrug-resistant profile and prevalence of extended spectrum β-lactamase and carbapenemase production in fermentative Gram-negative bacilli recovered from patients and specimens referred to National Reference Laboratory, Addis Ababa, Ethiopia. PLoS One. 2019;14(9):e0222911. Epub 2019/09/26. pmid:31553773; PubMed Central PMCID: PMC6760794.
- 14. Gashaw M, Berhane M, Bekele S, Kibru G, Teshager L, Yilma Y, et al. Emergence of high drug resistant bacterial isolates from patients with health care associated infections at Jimma University medical center: a cross sectional study. Antimicrob Resist Infect Control. 2018;7:138. Epub 2018/11/28. pmid:30479751; PubMed Central PMCID: PMC6245755.
- 15. Legese MH, Weldearegay GM, Asrat D. Extended-spectrum beta-lactamase- and carbapenemase-producing Enterobacteriaceae among Ethiopian children. Infect Drug Resist. 2017;10:27–34. Epub 2017/02/10. pmid:28182124; PubMed Central PMCID: PMC5279835.
- 16.
AAU. Background of Tikur Anbessa Hospital | College of Health Sciences (aau.edu.et). College of Health Sciences. 2021.
- 17. Naing L, Winn T, Rusli B. Practical issues in calculating the sample size for prevalence studies. Archives of orofacial Sciences. 2006;1:9–14.
- 18. Daniel WW, Cross CL. Biostatistics: a foundation for analysis in the health sciences. 1995.
- 19. Kitila KT, Tadesse BD, Tinsae K. Assessment of bacterial profile and antimicrobial resistance pattern of bacterial isolates from blood culture in Addis Ababa regional laboratory, Addis Ababa, Ethiopia. Clinical Microbiology. 2018;7:2.
- 20. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest. 1992;101(6):1644–55. Epub 1992/06/01. pmid:1303622.
- 21. Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Intensive Care Med. 2003;29(4):530–8. Epub 2003/03/29. pmid:12664219.
- 22. Dreyer AW. Blood Culture Systems: From Patient to Result. Sepsis-An Ongoing and Significant Challenge: InTech; 2012.
- 23. Flayhart D, Borek AP, Wakefield T, Dick J, Carroll KC. Comparison of BACTEC PLUS blood culture media to BacT/Alert FA blood culture media for detection of bacterial pathogens in samples containing therapeutic levels of antibiotics. J Clin Microbiol. 2007;45(3):816–21. Epub 2006/12/15. pmid:17166960; PubMed Central PMCID: PMC1829095.
- 24. Lee DH, Kim SC, Bae IG, Koh EH, Kim S. Clinical evaluation of BacT/Alert FA plus and FN plus bottles compared with standard bottles. J Clin Microbiol. 2013;51(12):4150–5. Epub 2013/10/11. pmid:24108609; PubMed Central PMCID: PMC3838062.
- 25.
Cheesbrough M. District laboratory practice in tropical countries: Cambridge university press; 2006.
- 26.
CLSI. Performance standards for antimicrobial susceptibility testing; 27th informational supplement. Wayne, 2017. Clinical and Laboratories Standards Institute. 2017;37(1).
- 27. Rods GP. UK Standards for Microbiology Investigations. 2014.
- 28.
CLSI. C.L.S.I. Performance standards for antimicrobial susceptibility testing: 23rd informational supplement, Wayne: CLSI; 2015 CLSI document M100S-S25. 2015;35(3).
- 29. EUCAST. Breakpoint tables for interpretation of MICs and zone diameters. European Committee on Antimicrobial Susceptibility Testing. 2017;7(1).
- 30.
CLSI. C.L.S.I. Performance standards for antimicrobial susceptibility testing. CLSI document M100S-S22. 2012; twenty second informational supplement ed.
- 31.
Liofilchem. MIC Test Strip Technical Sheet ESBL; For in vitro confirmation of ESBL. Quality Control Italy2014.
- 32.
Liofilchem. MIC Test Strip Technical Sheet AmpC; Synergic Inhibitory effect of Cloxacillin and Phenylboronic acid. Quality Control. Italy2014.
- 33.
CDC. Modified Hodge test for carbapenemase detection in Enterobacteriaceae. 2013.
- 34. Kadri SS. Key takeaways from the US CDC’s 2019 antibiotic resistance threats report for frontline providers. Critical care medicine. 2020.
- 35. Magiorakos AP, Srinivasan A, Carey RB, Carmeli Y, Falagas ME, Giske CG, et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clin Microbiol Infect. 2012;18(3):268–81. pmid:21793988.
- 36. Association WM. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. Jama. 2013;310(20):2191–4. Epub 2013/10/22. pmid:24141714.
- 37. Beshah D, Desta A, Belay G, Abebe T, Gebreselasie S, Sisay Tessema T. Antimicrobial Resistance and Associated Risk Factors of Gram-Negative Bacterial Bloodstream Infections in Tikur Anbessa Specialized Hospital, Addis Ababa. Infection and Drug Resistance. 2022:5043–59. pmid:36068835
- 38. Raphaely T, Marinova D, Marinova M. The Future of Antibiotics and Meat. Impact of Meat Consumption on Health and Environmental Sustainability: IGI Global; 2015. p. 178–200.
- 39. Wasihun AG, Wlekidan LN, Gebremariam SA, Dejene TA, Welderufael AL, Haile TD, et al. Bacteriological profile and antimicrobial susceptibility patterns of blood culture isolates among febrile patients in Mekelle Hospital, Northern Ethiopia. Springerplus. 2015;4:314. Epub 2015/07/15. pmid:26155453; PubMed Central PMCID: PMC4489972.
- 40. Negussie A, Mulugeta G, Bedru A, Ali I, Shimeles D, Lema T, et al. Bacteriological Profile and Antimicrobial Susceptibility Pattern of Blood Culture Isolates among Septicemia Suspected Children in Selected Hospitals Addis Ababa, Ethiopia. Int J Biol Med Res. 2015;6(1):4709–17. Epub 2016/03/22. pmid:26997847; PubMed Central PMCID: PMC4793966.
- 41. Alemnew B, Biazin H, Demis A, Abate Reta M. Bacterial Profile among Patients with Suspected Bloodstream Infections in Ethiopia: A Systematic Review and Meta-Analysis. International Journal of Microbiology. 2020;2020:8853053. pmid:32963541
- 42. Dagnew M, Yismaw G, Gizachew M, Gadisa A, Abebe T, Tadesse T, et al. Bacterial profile and antimicrobial susceptibility pattern in septicemia suspected patients attending Gondar University Hospital, Northwest Ethiopia. BMC Res Notes. 2013;6:283. Epub 2013/07/24. pmid:23875886; PubMed Central PMCID: PMC3723433.
- 43. Kumalo A, Kassa T, Daka D, Tadesse AH. Bacterial Profile of Adult Sepsis and their Antimicrobial Susceptibility Pattern at Jimma University Specialized Hospital, South West Ethiopia. Health Science Journal. 2016;10(2).
- 44. Hailu D, Abera B, Yitayew G, Mekonnen D, Derbie A. Bacterial blood stream infections and antibiogram among febrile patients at Bahir Dar Regional Health Research Laboratory Center, Ethiopia. Ethiopian Journal of Science and Technology. 2016;9(2):103–12.
- 45. Ahmed D, Nahid MA, Sami AB, Halim F, Akter N, Sadique T, et al. Bacterial etiology of bloodstream infections and antimicrobial resistance in Dhaka, Bangladesh, 2005–2014. Antimicrob Resist Infect Control. 2017;6:2. Epub 2017/01/11. pmid:28070309; PubMed Central PMCID: PMC5217397.
- 46.
CDC. Gram-negative bacteria infections in healthcare settings. Atlanta, GA: Centers for Disease Control and Prevention. 2011.
- 47.
Oliveira J, Reygaert WC. Gram Negative Bacteria. StatPearls. Treasure Island (FL): StatPearls Publishing LLC; 2023.
- 48. Zarrilli R, Bagattini M, Esposito EP, Triassi M. Acinetobacter Infections in Neonates. Curr Infect Dis Rep. 2018;20(12):48. Epub 2018/10/12. pmid:30306354.
- 49.
CDC. Preventing Healthcare-Associated Infections brocher. 2011.
- 50. Adeyemi F, Akinde S. ESβL, AmpC and carbapenemase co-production in multi-drug resistant Gram-negative bacteria from HIV-infected patients in southwestern Nigeria. African Journal of Clinical and Experimental Microbiology. 2021;22(1):38–51.
- 51. Jean SS, Lee WS, Lam C, Hsu CW, Chen RJ, Hsueh PR. Carbapenemase-producing Gram-negative bacteria: current epidemics, antimicrobial susceptibility and treatment options. Future Microbiol. 2015;10(3):407–25. Epub 2015/03/31. pmid:25812463.
- 52. Villegas MV, Pallares CJ, Escandón-Vargas K, Hernández-Gómez C, Correa A, Álvarez C, et al. Characterization and clinical impact of bloodstream infection caused by carbapenemase-producing Enterobacteriaceae in seven Latin American countries. PloS one. 2016;11(4):e0154092. pmid:27104910
- 53. Deku JG, Dakorah MP, Lokpo SY, Orish VN, Ussher FA, Kpene GE, et al. The Epidemiology of Bloodstream Infections and Antimicrobial Susceptibility Patterns: A Nine-Year Retrospective Study at St. Dominic Hospital, Akwatia, Ghana. J Trop Med. 2019;2019:6750864. Epub 2019/10/24. pmid:31641359; PubMed Central PMCID: PMC6770298.
- 54. Kang CI, Kim SH, Park WB, Lee KD, Kim HB, Oh MD, et al. Bloodstream infections caused by Enterobacter species: predictors of 30-day mortality rate and impact of broad-spectrum cephalosporin resistance on outcome. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America. 2004;39(6):812–8. Epub 2004/10/09. pmid:15472813.
- 55. Chaisathaphol T, Chayakulkeeree M. Epidemiology of infections caused by multidrug-resistant gram-negative bacteria in adult hospitalized patients at Siriraj Hospital. J Med Assoc Thai. 2014;97 Suppl 3:S35–45. Epub 2014/04/30. pmid:24772579.
- 56. Lucero-Prisno DE, Adebayo AY, Ekpenyong A. Antimicrobial Resistance in Africa. Africa Action on Antibiotics Resistance: strategies and innovative solutions. 2017;8.
- 57. Wangai FK, Masika MM, Lule GN, Karari EM, Maritim MC, Jaoko WG, et al. Bridging antimicrobial resistance knowledge gaps: the East African perspective on a global problem. PLoS One. 2019;14(2):e0212131. pmid:30742669
- 58. Tian L, Zhang Z, Sun Z. Antimicrobial resistance trends in bloodstream infections at a large teaching hospital in China: a 20-year surveillance study (1998–2017). Antimicrob Resist Infect Control. 2019;8:86. Epub 2019/06/05. pmid:31161033; PubMed Central PMCID: PMC6540536.
- 59. Williams PCM, Isaacs D, Berkley JA. Antimicrobial resistance among children in sub-Saharan Africa. The Lancet Infectious diseases. 2018;18(2):e33–e44. Epub 2017/10/17. pmid:29033034; PubMed Central PMCID: PMC5805911.
- 60. Prowle JR, Echeverri JE, Ligabo EV, Sherry N, Taori GC, Crozier TM, et al. Acquired bloodstream infection in the intensive care unit: incidence and attributable mortality. Critical care (London, England). 2011;15(2):R100. Epub 2011/03/23. pmid:21418635; PubMed Central PMCID: PMC3219371.
- 61. Jarvis WR, Murphy C, Hall KK, Fogle PJ, Karchmer TB, Harrington G, et al. Health care-associated bloodstream infections associated with negative- or positive-pressure or displacement mechanical valve needleless connectors. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America. 2009;49(12):1821–7. Epub 2009/11/17. pmid:19911973.
- 62. Snyder BM, Montague BT, Anandan S, Madabhushi AG, Pragasam AK, Verghese VP, et al. Risk factors and epidemiologic predictors of blood stream infections with New Delhi Metallo-b-lactamase (NDM-1) producing Enterobacteriaceae. Epidemiol Infect. 2019;147:e137. Epub 2019/03/15. pmid:30869056; PubMed Central PMCID: PMC6518792.
- 63. Di Franco S, Alfieri A, Pace MC, Sansone P, Pota V, Fittipaldi C, et al. Blood Stream Infections from MDR Bacteria. Life (Basel, Switzerland). 2021;11(6). Epub 2021/07/03. pmid:34207043; PubMed Central PMCID: PMC8233890.
- 64. Karakonstantis S, Kritsotakis EI, Gikas A. Pandrug-resistant Gram-negative bacteria: a systematic review of current epidemiology, prognosis and treatment options. The Journal of antimicrobial chemotherapy. 2020;75(2):271–82. Epub 2019/10/06. pmid:31586417.
- 65. de Campos LRP, Côrtes MF, Deo B, Rizek C, Santos S, Perdigão L, et al. Risk factors for bloodstream infection by multidrug-resistant organisms in critically ill patients in a reference trauma hospital. American Journal of Infection Control. 2022;50(6):673–9. pmid:34756966
- 66. Patolia S, Abate G, Patel N, Patolia S, Frey S. Risk factors and outcomes for multidrug-resistant Gram-negative bacilli bacteremia. Ther Adv Infect Dis. 2018;5(1):11–8. Epub 2018/01/19. pmid:29344356; PubMed Central PMCID: PMC5761922.
- 67. Kawakita T, Landy HJ. Surgical site infections after cesarean delivery: epidemiology, prevention and treatment. Matern Health Neonatol Perinatol. 2017;3:12. Epub 2017/07/12. pmid:28690864; PubMed Central PMCID: PMC5497372.