The World Health Organization recently proposed the Tricycle Protocol for harmonized surveillance of extended-spectrum beta lactamase-producing E. coli (ESBL-Ec) as a proxy for AMR among humans, animals, and the environment. We implemented the Tricycle Protocol with ceftriaxone (CRO) to screen for ESBL-Ec using membrane filtration (TBX-CRO) and Colilert Quanti-Tray (CQT-CRO) workflows in parallel applied to water samples collected from four lakes. ESBL-Ec and E. coli abundance and prevalence, and ESBL-Ec to E. coli ratios were significantly higher in the two highly impacted lakes (receiving sewage treatment plant effluent, open defecation site adjacent to informal settlements) compared to less impacted lakes (wildlife sanctuary, recreational site). The cumulative burden of antibiotic compounds was also higher in the highly impacted lakes. Even in the “pristine” lake in a bird sanctuary, ESBL-Ec and an antibiotic compound were present. Overall, the CQT-CRO format yielded reasonable qualitative (K = 0.603) and quantitative agreement (intra-class correlation coefficient = 0.849) and linearity (r2 = 0.785) with the TBX-CRO format adapted from the Tricycle Protocol. However, CQT-CRO doubled the prevalence of ESBL-Ec in settings with decreased E. coli abundance. Consistent with observations for other third-generation cephalosporins, CRO performed poorly for screening the ESBL phenotype. Further, confirmed ESBL-Ec isolates were not associated with an increased number of resistance phenotypes compared with non-ESBL isolates. Instead, the number of resistance phenotypes was associated with the lake from which the sample was collected. In high-burden settings, AMR surveillance efficiency could be increased by directly testing generic E. coli isolates for ESBL phenotype and antibiotic susceptibility. Our experience also suggests that sequencing ESBL-Ec will likely be required to determine epidemiologically appropriate risk mitigation strategies where ESBL-Ec are prevalent. Resource limitations could jeopardize the relevance of the Tricycle Protocol for changing the trajectory of AMR in high-burden low-resource settings globally.
Decentralized wastewater treatment systems afford a unique opportunity to implement a One Water paradigm through water reuse and resource recovery. Simultaneously, poorly managed on-site wastewater treatment systems (OWTSs) pose significant threats to human health, although their role in environmental antimicrobial resistance (AMR) is understudied. We deployed culture-based methods adapted from the World Health Organization’s Tricycle Protocol to assess the potential role of OWTSs in discharging third generation cephalosporin-resistant E. coli (3GCR-Ec) and extended-spectrum beta lactamase-producing E. coli (ESBL-Ec). We detected 3GCR-Ec in 74% of ATU effluents (geometric mean of 440 CFU/100 mL) and ESBL-Ec in 19%. E. coli isolates in these effluents were resistant to an average of 4.11 antibiotic compounds. ESBL-Ec were detected in up to 19% of surface water samples from rivers known to be impacted by OWTS discharges. Human sewage marker abundance was significantly higher in samples containing 3GCR-Ec (p = 0.0026) and ESBL-Ec (p = 0.0121) and the proportion of samples positive for 3GCR-Ec was strong correlated with river discharge percentile (r > 0.67, p < 0.0292). Our findings emphasize the unique human-environment interface of decentralized wastewater treatments systems must be carefully considered in both One Health and One Water paradigms to safely and sustainably meet our needs.