Evaluation of Prevalence and Changes in Antimicrobial-Resistant Fecal Organisms in Fecal Sludge and Wastewater Treatment Plants, Naivasha, Kenya

4 Poster Track: Applied Research » 4 Health, safety and hygiene

Speaker: Naomi Korir

Summary – for publication in conference brochure:

Antimicrobial resistance (AMR) is a global public health threat; sanitation systems may mitigate-or exacerbate-transmission of AMR pathogens. We quantified extended-spectrum beta-lactamase-producing (ESBL) E. coli, an AMR pathogen of global concern, in influent and effluent of wastewater and fecal sludge treatment plants in Naivasha, Kenya to examine AMR in fecal waste before and after treatment. We observed prevalent ESBL E. coli (≥5 log10/100mL in influent, ≥4 log10/100mL in effluent) with higher proportions of ESBL E. coli/total E. coli than is reported from treatment plants in high-income countries. These data underscore the need to monitor AMR in sanitation systems worldwide.

Introduction, methods, results and discussion:

Although antimicrobial resistance (AMR) is a pervasive, global threat, levels of AMR, including fecal AMR pathogens such as extended-spectrum beta-lactamase-producing (ESBL) E. coli, are not known across low- and middle-income countries (LMICs). Though improving water, sanitation, and hygiene (WASH) systems can prevent AMR- and non-AMR infections, WASH conditions remain poor and may even contribute to AMR spread in LMICs. For example, recent global estimates indicate about 25% of ESBL E. coli in feces is unsafely managed (discharged into unimproved systems or openly defecated).1 Levels of AMR pathogens and effectiveness of treatment processes before discharge should be evaluated in local sanitation systems to enumerate effective AMR mitigation strategies in communities.


In Kenya, 95% of fecal sludge is discharged into the environment without treatment. Only 17% of Kenyans have access to sewerage2 against targets of 40% by 2022. In Naivasha sub-county, a fecal flow study indicated that 75% of the population used non-sewered sanitation.3 Following this, Sanivation partnered with Naivasha Water and Sanitation Company (NAIVAWASCO, designed to serve 50,000 people on sewers, but was serving 143,000 on sewered and non-sewered systems) to process the non-sewered fecal sludge and instead focus on treating wastewater.

From November 2019-March 2020, we sampled the fecal sludge treatment plant (FSTP) 10 times for influent, effluent, and biosolids (final material used to make non-carbonized briquettes used as a firewood substitute) and sampled the NAIVAWASCO wastewater treatment plant (WWTP) 11 times for influent and effluent. All samples were tested for total E. coli, and ESBL E. coli via IDEXX Colilert® (supplemented with cefotaxime for ESBL E. coli quantification).

In the FSTP, influent E. coli (total) and ESBL E. coli levels averaged 6.3 (95% confidence interval (CI): 6.1, 6.5) log10most probable number (MPN) of coliform forming units/100mL and 5.0 (95% CI: 4.7, 5.3) log10MPN/100mL, respectively. Effluent E. coli and ESBL E. coli levels averaged 5.2 (95% CI: 4.7, 5.7) log10MPN/100mL and 4.1 (95% CI: 3.7, 4.4) log10MPN/100mL, respectively. No biosolids samples had detectable total or ESBL E. coli (all < 4 log10MPN/100mL). ESBL E. coli made up 6% (95% CI: 4%, 8%; range: 1-15%) of all E. coli measured in influent samples, and 7% (95% CI: 3%, 11%; range: 2-26%) of effluent samples. We observed a mean decrease of 1.0 log10MPN/100mL in concentrations of each type of E. coli after treatment. .

At the WWTP, influent E. coli and ESBL E. coli levels averaged 7.2 (95% CI: 7.0,.7.3) log10MPN/100mL and 6.4 (95% CI: 6.2, 6.7) log10MPN/100mL, respectively. Effluent E. coli and ESBL E. coli levels averaged 5.6 (95% CI: 5.4, 5.9) log10MPN/100mL and 4.7 (95% CI: 4.4, 5.0) log10MPN/100mL, respectively. ESBL E. coli comprised 23% (95% CI: 14%, 31%; range: 8-41%) of E. coli measured in influent samples and 13% (95% CI: 11%, 16%; range: 3-21%) of effluent samples. We observed a mean decrease of 1.5 log10MPN/100mL in concentrations of total E. coli and 1.7 log10MPN/100mL for ESBL E. coli after treatment.

Conclusions and implications:

ESBL E. coli, an important AMR pathogen, is readily detected at high concentrations in both influent (≥5 log10 MPN/100mL) and effluent (>4log10 MPN/100mL) from treatment plants in Naivasha, Kenya. These results suggest both frequent transmission of ESBL E. coli in local populations as well as relatively poor treatment of these and other organisms within existing sanitation plants, suggesting high levels of environmental discharge into receiving waters. In particular, ESBL E. coli entering WWTP and FSTP made up more of the overall E. coli levels than observed in high-income settings, suggesting a need for optimizing treatment to reduce AMR pathogens in addition to susceptible ones. To-date, AMR pathogens in community settings, including in sanitation systems, in LMICs have not garnered attention; however, the availability of field-practical methods, such as the WHO TriCycle or modified IDEXX methods (used here) make environmental surveillance a feasible and necessary future step in understanding AMR burden and how improving sanitation systems can reduce this burden in the environment and communities.

Relevant references:

1. Berendes D
Kirby A
Brown J
Wester AL. Human faeces-associated extended-spectrum β-lactamase-producing Escherichia coli discharge into sanitation systems in 2015 and 2030: a global and regional analysis. Lancet Planet Heal. 2020;4(6):e246-e255. doi:10.1016/S2542-5196(20)30099-1
2. Water Services Regulatory Board (WASREB). Impact: A Performance Report of Kenya’s Water Services Sector-2018/2019. 2020
3. Bohnert K. SFD Report: Naivasha Sub-county
Kenya. 2017.

Download Attachment