Abstract
Wastewater is considered a hotspot niche of multi-drug and pathogenic bacteria such as Enterobacteriaceae-producing extended-spectrum beta-lactamases (ESBL-E). Thus, the aim of this meta-analysis was to evaluate the prevalence of ESBL-E in different wastewater sources. Different databases (Medline, EMBASE, and Cochrane Library) were searched from inception to March 2021. Data were analyzed using random-effects modeling, and subgroup and meta-regression analyses were used to ascertain heterogeneity among the subgroups. Fifty-seven observational studies were selected, and the pooled prevalence of ESBL-E in wastewater was 24.81% (95% CI, 19.28–30.77). Escherichia coli had the highest ESBL prevalence. The blaCTX-M genes were the most prevalent in the selected studies (66.56%). The pooled prevalence of ESBL was significantly higher in reports from America (39.91%, 95% CI, 21.82–59.51) and reports studying hospital and untreated wastewaters (33.98%, 95% CI, 23.82–44.91 and 27.36%, 95% CI, 19.12–36.42). Overall, this meta-analysis showed that the prevalence of ESBL-E in wastewater is increasing over time and that hospital wastewater is the most important repository of ESBL-E. Therefore, there is a need for developing new sewage treatment systems that decrease the introduction of resistant bacteria and antibiotic residues.
HIGHLIGHTS
The global prevalence of Enterobacteriaceae-producing extended-spectrum beta-lactamases (ESBL-E) in wastewater was found to be 24.81%.
The pooled prevalence of ESBL-E was significantly higher in reports studying hospital wastewater.
The highest prevalence of ESBL-E was in America, and the lowest prevalence was in Europe.
Among ESBL genes, blaCTX-M genes had the highest prevalence, followed by blaTEM and blaSHV.
INTRODUCTION
Enterobacteriaceae are responsible for causing infections in humans and animals. They rely on their ability to resist antibiotics and on their virulence arsenal that facilitates their dissemination. The emergence of antimicrobial resistance in Enterobacteriaceae has become a significant concern to public health (Azuma & Hayashi 2020); it is believed that wastewater can contribute to propagating their antibiotic resistance (Meletis 2016). Wastewater is considered a hotspot niche of pathogenic bacteria and genetic exchange of genes encoding antibiotic resistance (Lepuschitz et al. 2019). Antibiotics released into wastewater increase the selective pressure on Enterobacteriaceae, thus causing antibiotic resistance to proliferate (Fouz et al. 2020). Moreover, wastewater treatments do not entirely eliminate the microbial contaminants; therefore, wastewater is mostly discharged to the receiving rivers enabling resistant Enterobacteriaceae and genes encoding antibiotic resistance to reach agricultural soils and water bodies used for domestic purposes (Korzeniewska & Harnisz 2013a). From there, these bacteria can disseminate into human and animal populations (Gatica & Cytryn 2013).
β-Lactams are one of the most commonly used antibiotics that are usually released in wastewater; these antibiotics can induce the production of β-lactamases in Enterobacteriaceae (Bonomo 2017). Among these β-lactamases, extended-spectrum beta-lactamases (ESBLs) are among the most widely spread resistance mechanisms (Teklu et al. 2019). They are a group of enzymes that can break down penicillins, β-lactamase inhibitors, and third- and fourth-generation cephalosporins and monobactams (Rahman et al. 2018). Most of them have been developed through spontaneous mutations of reduced spectrum β-lactamases (Reinthaler et al. 2010). Moreover, ESBLs are encoded by genes belonging mostly to three groups called blaCTX-M, blaTEM, and blaSHV (Pishtiwan & Khadija 2019). These genes are mostly located on conjugative plasmids, facilitating the horizontal gene transfer among bacterial groups (Jesumirhewe et al. 2020). Other less-studied types of ESBLs have also been found, including OXA (oxacillinase), VEB (Vietnamese extended-spectrum β-lactamase), PER (Pseudomonas extended-resistant), and GES (Guyana extended-spectrum β-lactamase) (Amirkamali et al. 2017).
The prevalence of ESBL-producing Enterobacteriaceae in wastewater varies with respect to geographical differences, wastewater source, and antimicrobial prescription patterns, most of the studies on ESBL-E in wastewater are limited to geographical areas such as Asia and Europe, and there has been no meta-analysis to the best of our knowledge that evaluated the prevalence of ESBL-E in different wastewater sources. Therefore, this meta-analysis aimed to determine the prevalence of ESBL-producing Enterobacteriaceae in different wastewater sources and analyze their influencing factors.
MATERIALS AND METHODS
Study and data collection
This meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher et al. 2009). A comprehensive literature search of PubMed/MEDLINE, EMBASE, and the Cochrane Library was performed until March 2021 to find potentially relevant articles. We also reviewed manually the references cited by the studies included to identify additional studies. The terms applied in electronic searches are listed in Supplementary Material, Table S1.
Inclusion and exclusion criteria
Studies were included if they met the following criteria: observational cross-sectional investigating the prevalence of ESBL-producing Enterobacteriaceae in wastewater. Exclusion criteria were included: (1) the absence of the total number of isolates, (2) the ESBL identification method is unclear, and (3) studies with the format of a congress abstract, a review article, or a book chapter.
Data extraction and quality assessment
Two authors independently screened the literature and extracted data, and the following variables were extracted: (1) authors' names, (2) location, (3) year of publication, (4) sample size, (5) the use of treatment or not, (6) prevalence of ESBL isolates, (7) species detected, (8) source of samples, (9) ESBL genes, and (10) ESBL identification methods. Since all the included studies were cross-sectional, each article's quality was assessed using Joanna Briggs Institute's quality assessment checklist (Munn et al. 2015) (Supplementary Material, Table S2).
Statistical analysis and data synthesis
STATA 16.0 (StataCorp LP, College Station, TX, USA) was used for the meta-analysis. The prevalence of ESBL-producing Enterobacteriaceae was estimated. The heterogeneity among studies was examined by the forest plot and the I2 heterogeneity test, in which 0–40, 50–60, 50–90, and 75–100% represented low, moderate, substantial, and considerable heterogeneity, respectively. Due to the high level of heterogeneity, the random-effects model was selected for analysis. Freeman–Tukey double arcsine transformation was used to avoid excluding studies where the prevalence of ESBL-E was close to 0% (Freeman & Tukey 1950; Nyaga et al. 2014). A subgroup analysis was performed, with a breakdown by the continents, use of treatment, year of publication, and the wastewater origin. A value of p ≤ 0.05 indicated statistical significance in the pooled effect. Meta-regression was used to ascertain heterogeneity among the subgroups. The selected independent factors are the wastewater origin (hospital wastewater, municipal wastewater, wastewater from rivers, hospital and municipal wastewaters and their receiving rivers, wastewater from other sources (farms and slaughterhouses)), continents (Africa, Europe, Asia, and America), the use of treatment (treated wastewater, untreated wastewater, and treated and untreated wastewater), and year of publication (2007–2015, 2016, 2017, 2018, 2019, and 2020). The number of studies from 2007 to 2015 is small. Therefore, they were put in the same category, so they will not be omitted by the software. The risk of publication bias across the studies was assessed by the funnel plot and the Egger test; the asymmetry of the funnel plot and statistical significance of Egger's regression test (p-value <0.05) were suggestive of publication bias (Egger et al. 1997). Sensitivity analysis was assessed to evaluate the robustness of the meta-analysis.
RESULTS
Eligible articles
The literature search process is shown in Figure 1. The database search yielded a total of 394 publications. After eliminating duplicates, 300 were selected to review their titles and abstracts, resulting in the exclusion of 214 publications; 86 studies were assessed then for eligibility by full-text reviewing based on the review's inclusion and exclusion criteria, and 29 were removed. Finally, 57 articles were ultimately included in the analysis.
PRISMA flow diagram of the selection process of the included studies.
Characteristics of the eligible studies
The main characteristics of the included studies are summarized in Table 1. The most used method for ESBL-E detection was the double-disk synergy method. In total, 28 countries from four continents studied ESBL-E prevalence in wastewater (Europe: 18, Asia: 18, Africa: 13, and America: 8). Among the 57 included studies, 17 studies examined hospital wastewater, 15 studies examined municipal wastewater, 13 studies examined hospital and municipal wastewaters and their receiving rivers, seven studies examined wastewater effluents in rivers, and five studies examined wastewater from other sources such as farms and slaughterhouses. Moreover, 23 studies examined the occurrence of ESBL-E in untreated wastewater, nine studies in treated wastewater, and 25 studies in untreated and treated wastewater. The selected studies were published between 2007 and 2020, 23.59% were carried out between the years 2007–2015 (n = 13), 21% were carried out in 2019 (n = 12), and 18.54% were carried in 2020 (n = 11).
Characteristics of the included studies
References . | No. of ESBL . | Enterobacteriaceae . | Wastewater origin . | Year . | Treatment . | Methods/guideline . | Country . |
---|---|---|---|---|---|---|---|
Prado et al. (2007) | 20 | 43 | Hospital wastewater | 2007 | Treated and untreated | CDT, CLSI | Brazil |
Sabaté et al. (2008) | 10 | 85 | Wastewater from other sources (farms and slaughterhouses) | 2008 | Treated and untreated | E-test, CLSI | Spain |
Łuczkiewicz et al. (2010) | 1 | 259 | Urban wastewater | 2010 | Treated and untreated | Phoenix Automated Microbiology System, CLSI | Poland |
Chagas et al. (2011) | 96 | 213 | Hospital wastewater | 2011 | Treated | DDST, CLSI | Brazil |
Mokracka et al. (2012) | 12 | 1,832 | Urban wastewater | 2012 | Treated and untreated | DDST, CLSI | Poland |
Diallo et al. (2013) | 9 | 1,248 | Wastewater from other sources (farms and slaughterhouses) | 2013 | Treated | E-test, CLSI | France |
Korzeniewska & Harnisz (2013a) | 150 | 310 | Hospital wastewater | 2013 | Untreated | CDT, CLSI | Poland |
Korzeniewska & Harnisz (2013b) | 34 | 246 | Urban wastewater | 2013 | Treated and untreated | CDT, CLSI | Poland |
Bessa et al. (2014) | 29 | 144 | Hospital and municipal wastewaters and their receiving rivers | 2014 | Treated | Disk approximation test, CLSI | Portugal |
Chandran et al. (2014) | 87 | 190 | Hospital wastewater | 2014 | Untreated | CDT | India |
Čornejová et al. (2015) | 27 | 104 | Urban wastewater | 2015 | Treated and untreated | Microdilution, CLSI | Slovakia |
Kwak et al. (2015) | 73 | 1,894 | Hospital and municipal wastewaters and their receiving rivers | 2015 | Treated and untreated | AREB microplates, CLSI | Sweden |
Kotlarska et al. (2015) | 5 | 774 | Hospital and municipal wastewaters and their receiving rivers | 2015 | Treated and untreated | DDST, EUCAST | Poland |
De Oliveira & Van Der Sand (2016) | 32 | 60 | Hospital wastewater | 2016 | Treated and untreated | DDST, CLSI | Brazil |
Drieux et al. (2016) | 25 | 389 | Hospital wastewater | 2016 | Untreated | DDST, CA-SFM | France |
Dropa et al. (2016) | 7 | 200 | Urban wastewater | 2016 | Untreated | DDST, CLSI | Brazil |
Egbule (2016) | 4 | 96 | Hospital wastewater | 2016 | Untreated | DDST, CLSI | Nigeria |
Lenart-Boroń (2016) | 23 | 196 | Wastewater-receiving rivers | 2016 | Untreated | DDST, EUCAST | Poland |
Maheshwari et al. (2016) | 34 | 103 | Hospital wastewater | 2016 | Untreated | CDT, CLSI | India |
Sultana et al. (2016) | 30 | 166 | Hospital wastewater | 2016 | Untreated | DDST | Bangladesh |
Caltagirone et al. (2017) | 30 | 132 | Hospital and municipal wastewaters and their receiving rivers | 2017 | Treated and untreated | DDST, EUCAST | Italy |
Conte et al. (2017) | 55 | 152 | Hospital and municipal wastewaters and their receiving rivers | 2017 | Treated and untreated | VITEK-2 system and MALDI-TOF, CLSI | Brazil |
Lien et al. (2017) | 115 | 265 | Hospital wastewater | 2017 | Treated and untreated | CDT, CLSI | Vietnam |
Obasi et al. (2017) | 6 | 7 | Urban wastewater | 2017 | Untreated | Vitek 2, E-test strips, CLSI | Nigeria |
Tafoukt et al. (2017) | 3 | 20 | Wastewater-receiving rivers | 2017 | Untreated | DDST, CLSI | Algeria |
Adelowo et al. (2018) | 48 | 98 | Hospital and municipal wastewaters and their receiving rivers | 2018 | Untreated | DDST, CLSI | Nigeria |
Daoud et al. (2018) | 51 | 70 | Hospital wastewater | 2018 | Untreated | DDST, E-test ESBL strips, CLSI | Lebanon |
Debabza et al. (2018) | 143 | 254 | Hospital and municipal wastewaters and their receiving rivers | 2018 | Treated and untreated | DDST, CA-SFM | Algeria |
Falodun et al. (2018) | 30 | 189 | Wastewater-receiving rivers | 2018 | Untreated | DDST, CLSI | Nigeria |
Flach et al. (2018) | 89 | 4,028 | Urban wastewater | 2018 | Treated and untreated | DDST, EUCAST | Sweden |
Park et al. (2018) | 14 | 75 | Wastewater from other sources (farms and slaughterhouses) | 2018 | Treated and untreated | DDST, EUCAST | South Korea |
Siddiqui et al. (2018) | 175 | 506 | Wastewater-receiving rivers | 2018 | Untreated | PDCT, CLSI | India |
Tokajian et al. (2018) | 21 | 34 | Hospital and municipal wastewaters and their receiving rivers | 2018 | Untreated | DDST, CLSI | Lebanon |
Vital et al. (2018) | 27 | 147 | Wastewater-receiving rivers | 2018 | Untreated | DDST, CLSI | Philippines |
Adekanmbi et al. (2019) | 17 | 58 | Hospital wastewater | 2019 | Untreated | DDST, CLSI | Nigeria |
Bartley et al. (2019) | 20 | 40 | Hospital and municipal wastewaters and their receiving rivers | 2019 | Treated and untreated | VITEK-2, CLSI | Brazil |
Chaudhry et al. (2019) | 17 | 112 | Hospital wastewater | 2019 | Untreated | DDST, CLSI | Pakistan |
Falodun & Oladimeji (2019) | 35 | 200 | Hospital wastewater | 2019 | Treated and untreated | DDST, CLSI | Nigeria |
Haberecht et al. (2019) | 47 | 70 | Urban wastewater | 2019 | Treated and untreated | CHROMagar ESBL | USA |
Li et al. (2019) | 50 | 70 | Wastewater from other sources (farms and slaughterhouses) | 2019 | Treated and untreated | DDST, CLSI | China |
Mahato et al. (2019) | 6 | 13 | Hospital wastewater | 2019 | Treated | DDST, CLSI | Nepal |
Miyagi & Hirai (2019) | 141 | 249 | Hospital and municipal wastewaters and their receiving rivers | 2019 | Untreated | DDST, CLSI | Japan |
Paulshus et al. (2019) | 314 | 3,123 | Urban wastewater | 2019 | Untreated | AREB microplates, CLSI | Norway |
Raven et al. (2019) | 192 | 388 | Urban wastewater | 2019 | Treated and untreated | Brilliance ESBL agar, Vitek | UK |
Sghaier et al. (2019) | 58 | 123 | Urban wastewater | 2019 | Treated | DDST, CLSI | Tunisia |
Tesfaye et al. (2019) | 6 | 54 | Wastewater from other sources (farms and slaughterhouses) | 2019 | Untreated | DDST, CLSI | Ethiopia |
Adekanmbi et al. (2020) | 12 | 33 | Hospital wastewater | 2020 | Treated and untreated | DDST, CLSI | Nigeria |
Banjo et al. (2020) | 12 | 23 | Hospital wastewater | 2020 | Untreated | DDST, CLSI | Nigeria |
Khan et al. (2020) | 35 | 61 | Urban wastewater | 2020 | Treated | CHROMagar ESBL, DDST, CLSI | UAE |
King et al. (2020) | 14 | 130 | Hospital and municipal wastewaters and their receiving rivers | 2020 | Treated and untreated | MASTDISCS D68C AmpC/ESBL kits, EUCAST | South Africa |
Lenart Boroń et al. (2020) | 0 | 50 | Wastewater-receiving rivers | 2020 | Treated | DDST, EUCAST | Poland |
Saima et al. (2020) | 1 | 10 | Wastewater-receiving rivers | 2020 | Untreated | DDST, CDT, CLSI | Pakistan |
Smyth et al. (2020) | 89 | 498 | Urban wastewater | 2020 | Treated | DDST, CLSI | Ireland |
Surleac et al. (2020) | 8 | 34 | Hospital and municipal wastewaters and their receiving rivers | 2020 | Treated and untreated | ChromID ESBL agar, CLSI | Romania |
Urano et al. (2020) | 5 | 64 | Urban wastewater | 2020 | Treated and untreated | DDST, CLSI | Japan |
Urase et al. (2020) | 13 | 264 | Urban wastewater | 2020 | Treated | DDST, CLSI | Japan |
Zagui et al. (2020) | 11 | 34 | Hospital and municipal wastewaters and their receiving rivers | 2020 | Treated and untreated | DDST, CLSI | Brazil |
References . | No. of ESBL . | Enterobacteriaceae . | Wastewater origin . | Year . | Treatment . | Methods/guideline . | Country . |
---|---|---|---|---|---|---|---|
Prado et al. (2007) | 20 | 43 | Hospital wastewater | 2007 | Treated and untreated | CDT, CLSI | Brazil |
Sabaté et al. (2008) | 10 | 85 | Wastewater from other sources (farms and slaughterhouses) | 2008 | Treated and untreated | E-test, CLSI | Spain |
Łuczkiewicz et al. (2010) | 1 | 259 | Urban wastewater | 2010 | Treated and untreated | Phoenix Automated Microbiology System, CLSI | Poland |
Chagas et al. (2011) | 96 | 213 | Hospital wastewater | 2011 | Treated | DDST, CLSI | Brazil |
Mokracka et al. (2012) | 12 | 1,832 | Urban wastewater | 2012 | Treated and untreated | DDST, CLSI | Poland |
Diallo et al. (2013) | 9 | 1,248 | Wastewater from other sources (farms and slaughterhouses) | 2013 | Treated | E-test, CLSI | France |
Korzeniewska & Harnisz (2013a) | 150 | 310 | Hospital wastewater | 2013 | Untreated | CDT, CLSI | Poland |
Korzeniewska & Harnisz (2013b) | 34 | 246 | Urban wastewater | 2013 | Treated and untreated | CDT, CLSI | Poland |
Bessa et al. (2014) | 29 | 144 | Hospital and municipal wastewaters and their receiving rivers | 2014 | Treated | Disk approximation test, CLSI | Portugal |
Chandran et al. (2014) | 87 | 190 | Hospital wastewater | 2014 | Untreated | CDT | India |
Čornejová et al. (2015) | 27 | 104 | Urban wastewater | 2015 | Treated and untreated | Microdilution, CLSI | Slovakia |
Kwak et al. (2015) | 73 | 1,894 | Hospital and municipal wastewaters and their receiving rivers | 2015 | Treated and untreated | AREB microplates, CLSI | Sweden |
Kotlarska et al. (2015) | 5 | 774 | Hospital and municipal wastewaters and their receiving rivers | 2015 | Treated and untreated | DDST, EUCAST | Poland |
De Oliveira & Van Der Sand (2016) | 32 | 60 | Hospital wastewater | 2016 | Treated and untreated | DDST, CLSI | Brazil |
Drieux et al. (2016) | 25 | 389 | Hospital wastewater | 2016 | Untreated | DDST, CA-SFM | France |
Dropa et al. (2016) | 7 | 200 | Urban wastewater | 2016 | Untreated | DDST, CLSI | Brazil |
Egbule (2016) | 4 | 96 | Hospital wastewater | 2016 | Untreated | DDST, CLSI | Nigeria |
Lenart-Boroń (2016) | 23 | 196 | Wastewater-receiving rivers | 2016 | Untreated | DDST, EUCAST | Poland |
Maheshwari et al. (2016) | 34 | 103 | Hospital wastewater | 2016 | Untreated | CDT, CLSI | India |
Sultana et al. (2016) | 30 | 166 | Hospital wastewater | 2016 | Untreated | DDST | Bangladesh |
Caltagirone et al. (2017) | 30 | 132 | Hospital and municipal wastewaters and their receiving rivers | 2017 | Treated and untreated | DDST, EUCAST | Italy |
Conte et al. (2017) | 55 | 152 | Hospital and municipal wastewaters and their receiving rivers | 2017 | Treated and untreated | VITEK-2 system and MALDI-TOF, CLSI | Brazil |
Lien et al. (2017) | 115 | 265 | Hospital wastewater | 2017 | Treated and untreated | CDT, CLSI | Vietnam |
Obasi et al. (2017) | 6 | 7 | Urban wastewater | 2017 | Untreated | Vitek 2, E-test strips, CLSI | Nigeria |
Tafoukt et al. (2017) | 3 | 20 | Wastewater-receiving rivers | 2017 | Untreated | DDST, CLSI | Algeria |
Adelowo et al. (2018) | 48 | 98 | Hospital and municipal wastewaters and their receiving rivers | 2018 | Untreated | DDST, CLSI | Nigeria |
Daoud et al. (2018) | 51 | 70 | Hospital wastewater | 2018 | Untreated | DDST, E-test ESBL strips, CLSI | Lebanon |
Debabza et al. (2018) | 143 | 254 | Hospital and municipal wastewaters and their receiving rivers | 2018 | Treated and untreated | DDST, CA-SFM | Algeria |
Falodun et al. (2018) | 30 | 189 | Wastewater-receiving rivers | 2018 | Untreated | DDST, CLSI | Nigeria |
Flach et al. (2018) | 89 | 4,028 | Urban wastewater | 2018 | Treated and untreated | DDST, EUCAST | Sweden |
Park et al. (2018) | 14 | 75 | Wastewater from other sources (farms and slaughterhouses) | 2018 | Treated and untreated | DDST, EUCAST | South Korea |
Siddiqui et al. (2018) | 175 | 506 | Wastewater-receiving rivers | 2018 | Untreated | PDCT, CLSI | India |
Tokajian et al. (2018) | 21 | 34 | Hospital and municipal wastewaters and their receiving rivers | 2018 | Untreated | DDST, CLSI | Lebanon |
Vital et al. (2018) | 27 | 147 | Wastewater-receiving rivers | 2018 | Untreated | DDST, CLSI | Philippines |
Adekanmbi et al. (2019) | 17 | 58 | Hospital wastewater | 2019 | Untreated | DDST, CLSI | Nigeria |
Bartley et al. (2019) | 20 | 40 | Hospital and municipal wastewaters and their receiving rivers | 2019 | Treated and untreated | VITEK-2, CLSI | Brazil |
Chaudhry et al. (2019) | 17 | 112 | Hospital wastewater | 2019 | Untreated | DDST, CLSI | Pakistan |
Falodun & Oladimeji (2019) | 35 | 200 | Hospital wastewater | 2019 | Treated and untreated | DDST, CLSI | Nigeria |
Haberecht et al. (2019) | 47 | 70 | Urban wastewater | 2019 | Treated and untreated | CHROMagar ESBL | USA |
Li et al. (2019) | 50 | 70 | Wastewater from other sources (farms and slaughterhouses) | 2019 | Treated and untreated | DDST, CLSI | China |
Mahato et al. (2019) | 6 | 13 | Hospital wastewater | 2019 | Treated | DDST, CLSI | Nepal |
Miyagi & Hirai (2019) | 141 | 249 | Hospital and municipal wastewaters and their receiving rivers | 2019 | Untreated | DDST, CLSI | Japan |
Paulshus et al. (2019) | 314 | 3,123 | Urban wastewater | 2019 | Untreated | AREB microplates, CLSI | Norway |
Raven et al. (2019) | 192 | 388 | Urban wastewater | 2019 | Treated and untreated | Brilliance ESBL agar, Vitek | UK |
Sghaier et al. (2019) | 58 | 123 | Urban wastewater | 2019 | Treated | DDST, CLSI | Tunisia |
Tesfaye et al. (2019) | 6 | 54 | Wastewater from other sources (farms and slaughterhouses) | 2019 | Untreated | DDST, CLSI | Ethiopia |
Adekanmbi et al. (2020) | 12 | 33 | Hospital wastewater | 2020 | Treated and untreated | DDST, CLSI | Nigeria |
Banjo et al. (2020) | 12 | 23 | Hospital wastewater | 2020 | Untreated | DDST, CLSI | Nigeria |
Khan et al. (2020) | 35 | 61 | Urban wastewater | 2020 | Treated | CHROMagar ESBL, DDST, CLSI | UAE |
King et al. (2020) | 14 | 130 | Hospital and municipal wastewaters and their receiving rivers | 2020 | Treated and untreated | MASTDISCS D68C AmpC/ESBL kits, EUCAST | South Africa |
Lenart Boroń et al. (2020) | 0 | 50 | Wastewater-receiving rivers | 2020 | Treated | DDST, EUCAST | Poland |
Saima et al. (2020) | 1 | 10 | Wastewater-receiving rivers | 2020 | Untreated | DDST, CDT, CLSI | Pakistan |
Smyth et al. (2020) | 89 | 498 | Urban wastewater | 2020 | Treated | DDST, CLSI | Ireland |
Surleac et al. (2020) | 8 | 34 | Hospital and municipal wastewaters and their receiving rivers | 2020 | Treated and untreated | ChromID ESBL agar, CLSI | Romania |
Urano et al. (2020) | 5 | 64 | Urban wastewater | 2020 | Treated and untreated | DDST, CLSI | Japan |
Urase et al. (2020) | 13 | 264 | Urban wastewater | 2020 | Treated | DDST, CLSI | Japan |
Zagui et al. (2020) | 11 | 34 | Hospital and municipal wastewaters and their receiving rivers | 2020 | Treated and untreated | DDST, CLSI | Brazil |
CDT, combined disc diffusion method; CLSI, Clinical & Laboratory Standards Institute; DDST, The Double Disc Synergy Test; PDCT, Phenotypic disc confirmatory test; CA-SFM, Comité de l'Antibiogramme de la société Française de Microbiologie.
The most common producer of ESBL in wastewater was found to be Escherichia coli (51 studies), followed by Klebsiella (29 studies) and Enterobacter (18 studies). Moreover, the blaCTX-M gene was the most studied (30 studies), followed by the blaTEM and blaSHV genes (25 studies).
Pooled prevalence of ESBL-E
A total of 2,618 isolates of ESBL-E from 20,230 Enterobacteriaceae were found in the selected publications (Table 2), and the estimated prevalence varied widely across the studies ranging from 0.00% (95% CI, 0.00–7.13) to 85.71% (95% CI, 48.69–97.43) with substantial heterogeneity (χ2 = 4461.79, I2 = 98.74%, p < 0.001). The random-effect estimated pooled prevalence of ESBL-E was 24.81% (95% CI, 19.28, 30.77). Figure 2 shows the pooled prevalence of ESBL-E.
Stratified pooled prevalence estimates of ESBL-producing Enterobacteriaceae in wastewater
Subgroups . | No. of studies . | No. of ESBL . | No. of Enterobacteriaceae . | Pooled estimate (%) of ESBL . | 95% Confidence interval . | Heterogeneity chi-squared (χ2) . | Heterogeneity test, I2 (%) . | Heterogeneity test, p-value . |
---|---|---|---|---|---|---|---|---|
Continents . | . | |||||||
Europe | 18 | 1,112 | 15,700 | 10.19 | 5.73, 15.72 | 1,645.83 | 98.97 | <0.001 |
Asia | 18 | 830 | 2,433 | 33.95 | 24.05, 44.58 | 449.28 | 96.22 | <0.001 |
Africa | 13 | 388 | 1,285 | 29.31 | 17.86, 42.17 | 250.07 | 95.20 | <0.001 |
America | 8 | 288 | 812 | 39.91 | 21.82, 59.51 | 206.09 | 96.60 | <0.001 |
Overall | 57 | 2,618 | 20,230 | 24.81 | 19.28, 30.77 | 4,461.79 | 98.74 | <0.001 |
Wastewater origin . | . | |||||||
Hospital wastewater | 17 | 743 | 2,344 | 33.98 | 23.82, 44.91 | 442.77 | 96.39 | <0.001 |
Municipal wastewater | 15 | 929 | 11,267 | 18.83 | 11.15, 27.88 | 1,410.51 | 99.01 | <0.001 |
Wastewater from rivers | 7 | 259 | 1,118 | 13.67 | 5.36, 24.64 | 96.29 | 93.77 | <0.001 |
Hospital and municipal wastewaters and their receiving rivers | 13 | 598 | 3,969 | 29.41 | 15.03, 46.19 | 1,123.05 | 98.93 | <0.001 |
Wastewater from other sources (farms and slaughterhouses) | 5 | 89 | 1,532 | 18.43 | 1.04, 48.62 | 273.15 | 98.54 | <0.001 |
Overall | 57 | 2,618 | 20,230 | 24.81 | 19.28, 30.77 | 4,461.79 | 98.74 | <0.001 |
The use of treatment . | . | |||||||
Treated wastewater | 9 | 335 | 2,614 | 21.11 | 7.40, 39.17 | 649.12 | 98.77 | <0.001 |
Untreated wastewater | 23 | 1,229 | 6,350 | 27.36 | 19.12, 36.42 | 956.04 | 97.70 | <0.001 |
Treated and untreated wastewater | 25 | 1,054 | 11,266 | 23.93 | 15.87, 33.03 | 2,286.92 | 98.95 | <0.001 |
Overall | 57 | 2,618 | 20,230 | 24.81 | 19.28, 30.77 | 4,461.79 | 98.74 | <0.001 |
Year of publication . | . | |||||||
2020 | 11 | 200 | 1,201 | 19.32 | 10.31, 30.17 | 144.56 | 93.08 | <0.001 |
2019 | 12 | 903 | 4,500 | 37.62 | 22.18, 54.43 | 766.42 | 98.56 | <0.001 |
2018 | 9 | 598 | 5,401 | 33.93 | 14.01, 57.34 | 1,191.61 | 99.33 | <0.001 |
2017 | 5 | 209 | 576 | 34.73 | 22.53, 47.96 | 28.30 | 85.86 | <0.001 |
2016 | 7 | 155 | 1,210 | 15.51 | 7.02, 26.45 | 126.54 | 95.26 | <0.001 |
2007–2015 | 13 | 553 | 7,342 | 15.08 | 7.17, 25.19 | 1,276.01 | 99.06 | <0.001 |
Overall | 57 | 2,618 | 20,230 | 24.81 | 19.28, 30.77 | 4,461.79 | 98.74 | <0.001 |
Subgroups . | No. of studies . | No. of ESBL . | No. of Enterobacteriaceae . | Pooled estimate (%) of ESBL . | 95% Confidence interval . | Heterogeneity chi-squared (χ2) . | Heterogeneity test, I2 (%) . | Heterogeneity test, p-value . |
---|---|---|---|---|---|---|---|---|
Continents . | . | |||||||
Europe | 18 | 1,112 | 15,700 | 10.19 | 5.73, 15.72 | 1,645.83 | 98.97 | <0.001 |
Asia | 18 | 830 | 2,433 | 33.95 | 24.05, 44.58 | 449.28 | 96.22 | <0.001 |
Africa | 13 | 388 | 1,285 | 29.31 | 17.86, 42.17 | 250.07 | 95.20 | <0.001 |
America | 8 | 288 | 812 | 39.91 | 21.82, 59.51 | 206.09 | 96.60 | <0.001 |
Overall | 57 | 2,618 | 20,230 | 24.81 | 19.28, 30.77 | 4,461.79 | 98.74 | <0.001 |
Wastewater origin . | . | |||||||
Hospital wastewater | 17 | 743 | 2,344 | 33.98 | 23.82, 44.91 | 442.77 | 96.39 | <0.001 |
Municipal wastewater | 15 | 929 | 11,267 | 18.83 | 11.15, 27.88 | 1,410.51 | 99.01 | <0.001 |
Wastewater from rivers | 7 | 259 | 1,118 | 13.67 | 5.36, 24.64 | 96.29 | 93.77 | <0.001 |
Hospital and municipal wastewaters and their receiving rivers | 13 | 598 | 3,969 | 29.41 | 15.03, 46.19 | 1,123.05 | 98.93 | <0.001 |
Wastewater from other sources (farms and slaughterhouses) | 5 | 89 | 1,532 | 18.43 | 1.04, 48.62 | 273.15 | 98.54 | <0.001 |
Overall | 57 | 2,618 | 20,230 | 24.81 | 19.28, 30.77 | 4,461.79 | 98.74 | <0.001 |
The use of treatment . | . | |||||||
Treated wastewater | 9 | 335 | 2,614 | 21.11 | 7.40, 39.17 | 649.12 | 98.77 | <0.001 |
Untreated wastewater | 23 | 1,229 | 6,350 | 27.36 | 19.12, 36.42 | 956.04 | 97.70 | <0.001 |
Treated and untreated wastewater | 25 | 1,054 | 11,266 | 23.93 | 15.87, 33.03 | 2,286.92 | 98.95 | <0.001 |
Overall | 57 | 2,618 | 20,230 | 24.81 | 19.28, 30.77 | 4,461.79 | 98.74 | <0.001 |
Year of publication . | . | |||||||
2020 | 11 | 200 | 1,201 | 19.32 | 10.31, 30.17 | 144.56 | 93.08 | <0.001 |
2019 | 12 | 903 | 4,500 | 37.62 | 22.18, 54.43 | 766.42 | 98.56 | <0.001 |
2018 | 9 | 598 | 5,401 | 33.93 | 14.01, 57.34 | 1,191.61 | 99.33 | <0.001 |
2017 | 5 | 209 | 576 | 34.73 | 22.53, 47.96 | 28.30 | 85.86 | <0.001 |
2016 | 7 | 155 | 1,210 | 15.51 | 7.02, 26.45 | 126.54 | 95.26 | <0.001 |
2007–2015 | 13 | 553 | 7,342 | 15.08 | 7.17, 25.19 | 1,276.01 | 99.06 | <0.001 |
Overall | 57 | 2,618 | 20,230 | 24.81 | 19.28, 30.77 | 4,461.79 | 98.74 | <0.001 |
Pooled prevalence of ESBL-producing Enterobacteriaceae in wastewater. Studies are sorted alphabetically, squares represent effect sizes of individual studies and the diamond indicates the estimated pooled effect size. CI, confidence interval.
Pooled prevalence of ESBL-producing Enterobacteriaceae in wastewater. Studies are sorted alphabetically, squares represent effect sizes of individual studies and the diamond indicates the estimated pooled effect size. CI, confidence interval.
Variables associated with ESBL-E prevalence
Subgroup meta-analysis results are presented in Table 2. The analysis by continent showed that the highest prevalence of ESBL-E was in America at 39.91%, and the lowest prevalence was in Europe at 10.19% (95% CI, 5.73–15.72). The pooled prevalence of ESBL was significantly higher (p < 0.001) in reports studying hospital wastewater (33.98%, 95% CI, 23.82–44.91) and, more specifically, in untreated wastewater (27.36%, 95% CI, 19.12–36.42). The number of reports on ESBL-E prevalence in wastewater increased over time, and the highest number of studies per year was reported in 2019 (n = 12). Forest plots of the ESBL-E prevalence among studies stratified by continents, wastewater sources, the year, and the use of treatment are outlined in Figures 3–6.
Pooled prevalence of ESBL-producing Enterobacteriaceae stratified by continents.
Pooled prevalence of ESBL-producing Enterobacteriaceae stratified by continents.
Pooled prevalence of ESBL-producing Enterobacteriaceae stratified by the year of publication.
Pooled prevalence of ESBL-producing Enterobacteriaceae stratified by the year of publication.
Pooled prevalence of ESBL-producing Enterobacteriaceae stratified by treatments.
Pooled prevalence of ESBL-producing Enterobacteriaceae stratified by treatments.
Pooled prevalence of ESBL-producing Enterobacteriaceae stratified by the wastewater origin.
Pooled prevalence of ESBL-producing Enterobacteriaceae stratified by the wastewater origin.
Among Enterobacteriaceae species, E. coli had the highest ESBL prevalence (15.02%, 95% CI, 11.12–19.36) (Table 3) followed by Klebsiella spp. (9.60%, 95% CI, 19.54 38.29) and Proteus spp. (3.80%, 95% CI, 1.61, 6.76). Moreover, among ESBL genes, the blaCTX-M gene had the highest prevalence (66.56%, 95% CI, 53.98–78.15), followed by blaTEM (49.88%, 95% CI, 35.01–64.76) and blaSHV (21.58%, 95% CI, 13.71–30.49).
Meta-analysis of ESBL-E at the species level and ESBL-E gene prevalence in wastewaters
Subgroups . | No. of studies . | No. of ESBL . | No. of Enterobacteriaceae . | Pooled estimate (%) of ESBL . | 95% Confidence interval . | Heterogeneity χ2 . | Heterogeneity test, I2 (%) . | Heterogeneity test, p-value . |
---|---|---|---|---|---|---|---|---|
Enterobacteriaceae species . | ||||||||
E. coli | 51 | 2,541 | 19,881 | 15.05 | 11.14, 19.41 | 2,902.78 | 98.28 | <0.001 |
Klebsiella | 29 | 1,222 | 5,732 | 9.66 | 5.83, 14.25 | 665.17 | 95.79 | <0.001 |
Enterobacter | 18 | 1,139 | 5,125 | 2.39 | 1.08, 4.09 | 149.25 | 88.61 | <0.001 |
Citrobacter | 8 | 714 | 1,846 | 2.69 | 0.50, 6.23 | 87.10 | 91.96 | <0.001 |
Shigella | 7 | 619 | 1,578 | 2.10 | 0.66, 4.19 | 28.58 | 79.01 | <0.001 |
Proteus | 6 | 323 | 900 | 3.80 | 1.61, 6.76 | 18.02 | 72.26 | <0.001 |
Serratia | 5 | 481 | 1,036 | 3.37 | 0.27, 8.89 | 48.36 | 91.73 | <0.001 |
Other species | 12 | 785 | 2,064 | 3.07 | 1.04, 5.96 | 169.26 | 93.50 | <0.001 |
ESBL genes . | No. of studies . | No. of ESBL . | No. of Enterobacteriaceae . | Pooled estimate (%) of ESBL genes . | 95% Confidence interval . | Heterogeneity χ2 . | Heterogeneity test, I2 (%) . | Heterogeneity test, p-value . |
blaCTX-M | 30 | 1,621 | 7,532 | 66.82% | 52.57, 79.79 | 691.61 | 96.10 | <0.001 |
blaTEM | 25 | 1,292 | 4,888 | 51.01% | 35.52, 66.40 | 661.40 | 96.37 | <0.001 |
blaSHV | 25 | 1,044 | 4,875 | 24.59% | 15.42, 34.91 | 239.93 | 90.41 | <0.001 |
Subgroups . | No. of studies . | No. of ESBL . | No. of Enterobacteriaceae . | Pooled estimate (%) of ESBL . | 95% Confidence interval . | Heterogeneity χ2 . | Heterogeneity test, I2 (%) . | Heterogeneity test, p-value . |
---|---|---|---|---|---|---|---|---|
Enterobacteriaceae species . | ||||||||
E. coli | 51 | 2,541 | 19,881 | 15.05 | 11.14, 19.41 | 2,902.78 | 98.28 | <0.001 |
Klebsiella | 29 | 1,222 | 5,732 | 9.66 | 5.83, 14.25 | 665.17 | 95.79 | <0.001 |
Enterobacter | 18 | 1,139 | 5,125 | 2.39 | 1.08, 4.09 | 149.25 | 88.61 | <0.001 |
Citrobacter | 8 | 714 | 1,846 | 2.69 | 0.50, 6.23 | 87.10 | 91.96 | <0.001 |
Shigella | 7 | 619 | 1,578 | 2.10 | 0.66, 4.19 | 28.58 | 79.01 | <0.001 |
Proteus | 6 | 323 | 900 | 3.80 | 1.61, 6.76 | 18.02 | 72.26 | <0.001 |
Serratia | 5 | 481 | 1,036 | 3.37 | 0.27, 8.89 | 48.36 | 91.73 | <0.001 |
Other species | 12 | 785 | 2,064 | 3.07 | 1.04, 5.96 | 169.26 | 93.50 | <0.001 |
ESBL genes . | No. of studies . | No. of ESBL . | No. of Enterobacteriaceae . | Pooled estimate (%) of ESBL genes . | 95% Confidence interval . | Heterogeneity χ2 . | Heterogeneity test, I2 (%) . | Heterogeneity test, p-value . |
blaCTX-M | 30 | 1,621 | 7,532 | 66.82% | 52.57, 79.79 | 691.61 | 96.10 | <0.001 |
blaTEM | 25 | 1,292 | 4,888 | 51.01% | 35.52, 66.40 | 661.40 | 96.37 | <0.001 |
blaSHV | 25 | 1,044 | 4,875 | 24.59% | 15.42, 34.91 | 239.93 | 90.41 | <0.001 |
Meta-regression evaluating the effect of confounding factors on the prevalence of ESBL-E was performed, and the results are summarized in Table 4. In the univariable meta-regression, an association was found between ESBL prevalence and several factors at p ≤ 0.25. In the final multivariable meta-regression model, a positive association was found between ESBL-E carriage and studies reported in America (Coef. = 0.22, p = 0.01).
Summary results of univariable and multivariable meta-regression of the effects of confounding factors on ESBL-E occurrence in wastewaters
. | Coef.b (95% CIc) . | SE . | p-value . | Coef. (95% CI) . | SE . | p-value . |
---|---|---|---|---|---|---|
Moderators . | Univariable regression . | Multivariable regression . | ||||
Continents . | . | |||||
Europea | – | – | – | – | – | – |
Asia | 0.22 (0.08, 0.35) | 0.06 | 0.002 | 0.14 ( − 0.01, 0.31) | 0.08 | 0.078 |
Africa | 0.16 (0.007, 0.31) | 0.07 | 0.040 | 0.06 ( − 0.11, 0.24) | 0.09 | 0.507 |
America | 0.26 (0.08, 0.44) | 0.09 | 0.003 | 0.22 (0.04, 0.39) | 0.08 | 0.012 |
Wastewater origin . | . | |||||
Hospital wastewatera | – | – | – | – | – | – |
Municipal wastewater | − 0.13 ( − 0.27, 0.016) | 0.07 | 0.081 | −0.11 ( − 0.27, 0.05) | 0.08 | 0.183 |
Wastewater effluents in rivers | − 0.17 ( − 0.37, 0.017) | 0.10 | 0.074 | − 0.28 ( − 0.41, 0.008) | 0.10 | 0.060 |
Hospital and municipal wastewaters and their receiving rivers Wastewater from other sources (farms and slaughterhouses) | − 0.03 (−0.19, 0.11) −0.13 (−0.34, 0.08) | 0.07 0.10 | 0.650 0.222 | − 0.05 (−0.21, 0.11) −0.14 (−0.36, 0.06) | 0.08 0.10 | 0.557 0.163 |
Treatments . | . | |||||
Treated and untreated wastewatera | – | – | – | – | – | – |
Untreated wastewater | 0.02 ( − 0.10, 0.14) | 0.06 | 0.733 | 0.016 ( − 0.13, 0.16) | 0.07 | 0.829 |
Treated wastewater | − 0.01 ( − 0.18, 0.14) | 0.08 | 0.816 | 0.04 ( − 0.11, 0.20) | 0.08 | 0.604 |
Year of publication . | . | |||||
2007–2015 | – | – | – | – | – | – |
2016 | − 0.01 ( − 0.19, 0.16) | 0.09 | 0.851 | − 0.11 ( − 0.31, 0.09) | 0.10 | 0.276 |
2017 | 0.16 ( − 0.06, 0.38) | 0.11 | 0.158 | 0.08 ( − 0.14, 0.31) | 0.11 | 0.483 |
2018 | 0.15 ( − 0.01, 0.32) | 0.08 | 0.073 | 0.14 ( − 0.06, 0.35) | 0.10 | 0.175 |
2019 | 0.19 (0.03, 0.34) | 0.08 | 0.018 | 0.13 ( − 0.03, 0.31) | 0.08 | 0.122 |
2020 | 0.02 ( − 0.14, 0.19) | 0.08 | 0.784 | − 0.01 ( − 0.20, 0.18) | 0.09 | 0.911 |
. | Coef.b (95% CIc) . | SE . | p-value . | Coef. (95% CI) . | SE . | p-value . |
---|---|---|---|---|---|---|
Moderators . | Univariable regression . | Multivariable regression . | ||||
Continents . | . | |||||
Europea | – | – | – | – | – | – |
Asia | 0.22 (0.08, 0.35) | 0.06 | 0.002 | 0.14 ( − 0.01, 0.31) | 0.08 | 0.078 |
Africa | 0.16 (0.007, 0.31) | 0.07 | 0.040 | 0.06 ( − 0.11, 0.24) | 0.09 | 0.507 |
America | 0.26 (0.08, 0.44) | 0.09 | 0.003 | 0.22 (0.04, 0.39) | 0.08 | 0.012 |
Wastewater origin . | . | |||||
Hospital wastewatera | – | – | – | – | – | – |
Municipal wastewater | − 0.13 ( − 0.27, 0.016) | 0.07 | 0.081 | −0.11 ( − 0.27, 0.05) | 0.08 | 0.183 |
Wastewater effluents in rivers | − 0.17 ( − 0.37, 0.017) | 0.10 | 0.074 | − 0.28 ( − 0.41, 0.008) | 0.10 | 0.060 |
Hospital and municipal wastewaters and their receiving rivers Wastewater from other sources (farms and slaughterhouses) | − 0.03 (−0.19, 0.11) −0.13 (−0.34, 0.08) | 0.07 0.10 | 0.650 0.222 | − 0.05 (−0.21, 0.11) −0.14 (−0.36, 0.06) | 0.08 0.10 | 0.557 0.163 |
Treatments . | . | |||||
Treated and untreated wastewatera | – | – | – | – | – | – |
Untreated wastewater | 0.02 ( − 0.10, 0.14) | 0.06 | 0.733 | 0.016 ( − 0.13, 0.16) | 0.07 | 0.829 |
Treated wastewater | − 0.01 ( − 0.18, 0.14) | 0.08 | 0.816 | 0.04 ( − 0.11, 0.20) | 0.08 | 0.604 |
Year of publication . | . | |||||
2007–2015 | – | – | – | – | – | – |
2016 | − 0.01 ( − 0.19, 0.16) | 0.09 | 0.851 | − 0.11 ( − 0.31, 0.09) | 0.10 | 0.276 |
2017 | 0.16 ( − 0.06, 0.38) | 0.11 | 0.158 | 0.08 ( − 0.14, 0.31) | 0.11 | 0.483 |
2018 | 0.15 ( − 0.01, 0.32) | 0.08 | 0.073 | 0.14 ( − 0.06, 0.35) | 0.10 | 0.175 |
2019 | 0.19 (0.03, 0.34) | 0.08 | 0.018 | 0.13 ( − 0.03, 0.31) | 0.08 | 0.122 |
2020 | 0.02 ( − 0.14, 0.19) | 0.08 | 0.784 | − 0.01 ( − 0.20, 0.18) | 0.09 | 0.911 |
aRef., reference category; Coef., regression coefficient; CI, confidence interval.
Analysis of publication bias and sensitivity
The publication bias among the included papers was evaluated with Egger's test, which showed some evidence of publication bias, with a bias co-efficient of 1.56 (p = 0.002). The funnel plot's shape was not symmetrical (Supplementary Material, Figure S1). Thus, there is a potential for publication bias. Moreover, the sensitivity analysis showed no individual study that influenced the overall meta-analysis estimate, which indicates the robustness of the results (Supplementary Material, Table S3).
DISCUSSION
Wastewater is an important environmental supplier of antibacterial-resistant bacteria and antimicrobial-resistance genes (Marti et al. 2013). In this study, the occurrence of ESBL-producing Enterobacteriaceae in different wastewater sources has been evaluated. There are some big differences between hospital, municipal, and animal wastewaters. Thus, we have chosen to do the subgroup analysis and study the prevalence of ESBL-E in each source separately. In total, the pooled prevalence of ESBL-E in wastewater was substantially important (24.81%), with the highest prevalence in untreated wastewater (27.36%). Untreated wastewater is loaded with different pollutants that can facilitate the proliferation of resistant bacteria (Preisner 2020). However, Du et al. (2015) suggested that biological treatments in wastewater treatment plants can also enhance the bacterial proliferation and genetic exchange.
Moreover, hospital wastewater had the highest prevalence of ESBL-E (33.98%). This could be due to the high usage of antibiotics in hospitals compared with public usage. Kümmerer (2009) found that the concentration of antibiotics in hospital effluents is 100 times higher than in municipal effluents. Moreover, the intensive use of antibiotics in hospitals could also contribute in the ESBL emergence (Debabza et al. 2018), which turns the hospital into a highly selective environment for ESBL-producing bacteria (Duong et al. 2008). Wastewater effluents in rivers had the lowest prevalence of ESBL-E. Wastewaters are usually treated before being released to waterways, which could explain the low prevalence of ESBL-E in rivers (Numberger et al. 2019).
Europe had the highest number of studies on ESBL-E and the lowest prevalence of ESBL-E. On the other hand, America (mainly South America because 7/8 of studies are in Brazil) and Asia had the highest prevalence of ESBL-E (39.91 and 33.95%, respectively), and according to the meta-regression results, studies reported in America are significantly associated with the presence of ESBL-E in wastewater. In developing countries, wastewater generated from farming, communities, and hospital effluents does not obtain appropriate treatment, and few functioning treatment facilities are available. This inadequate management can lead to environmental contamination with resistant bacteria such as ESBL-E into the water environment (Behnam et al. 2020; WHO 2018).
The highest prevalence of ESBL-E in wastewater was reported in studies undertaken in 2019. Gelband et al. (2015) analyzed data from 73 countries over the past 10 years and found that antibiotic use is growing steadily worldwide, which can explain, in part, the increase of ESBL-E in wastewater in the last 2 years. On the other hand, the biggest number of studies on ESBL-E in wastewater was also observed between the years 2019–2020; this could possibly mean that public awareness of the challenges related to the presence of resistant bacteria in wastewater is starting to rise lately.
The studies included in this meta-analysis reported that E. coli was the most isolated ESBL-producing Enterobacteriaceae (15.02%). These bacteria are indicators of fecal contamination, agents of several kinds of nosocomial and community-acquired infections, and are often associated with therapy failure when cephalosporins are used (Kola et al. 2007; Shakya et al. 2017). The high prevalence of these enzymes in E. coli is maybe due to the ability of E. coli to survive for long periods and multiply in wastewater without being affected with wastewater treatments, which leads to the selection and emergence of ESBL-producing E. coli (Jang et al. 2017). Moreover, the blaCTX-M genes were the most prevalent in the selected studies, accounting for a significant percentage of the ESBL genes detected (66.56%). This high prevalence of CTX-M-type ESBL producers could be due to the global spreading of clones with epidemic and pandemic potential, such as the extra-intestinal pathogenic E. coli (ExPEC) ST131 known for its ability to produce extended-spectrum β-lactamases, such as CTX-M-15 (Cantón et al. 2012; Tanaka et al. 2019).
These findings enhance our understanding of ESBL-producing Enterobacteriaceae in wastewater. However, some limitations need to be addressed; in some recent studies, ESBL-E isolates were directly isolated using selective media without isolating Enterobacteriaceae first, and these studies were not included in our study because the total number of isolates is absent. Moreover, most of the included studies on ESBL-E in treated wastewater did not report the type of treatment used, so we were obligated to put three categories (treated and untreated, treated, and untreated), and another potential limitation of the present study is that statistical results may have been influenced by publication bias based on the visual evaluation of the funnel plot and the Egger test results; this is maybe due to the inherent bias toward reporting positive results; despite our effort to use search strategies that enhance the reporting of negative results, we have only found one study (the study of Lenart-Boroń et al. (2020)) that reported negative data.
CONCLUDING REMARKS
Based on the 57 studies analyzed in this meta-analysis, we concluded that the prevalence of Enterobacteriaceae-producing ESBL in wastewater is increasing over time and that hospital wastewater is the most important repository of ESBL. This study also revealed a high prevalence of the blaCTX-M genes in Enterobacteriaceae isolated from wastewater, which is an alarming indicator of the global spreading of epidemic resistance plasmids.
Hence, these results highlight the need to develop effective strategies and measures adapted for removing ESBL-producing bacteria in wastewater and preventing the dissemination and transmission of antibiotic resistance from wastewater to different aquatic systems. Moreover, further research must focus on developing new sewage treatment systems that decrease the introduction of resistant bacteria and antibiotic residues.
FUNDING
This work received no external funding.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
AUTHORS’ CONTRIBUTIONS
NZ: conceptualization, writing and editing; SB and NS: investigation and writing.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.