Prevalence of Antibiotic Resistance Among Coliform Bacteria Isolated from Surface Water Samples in Kisumu County, Kenya
Year: 2026
Type: preprint
Abstract: Background Environmental waters and animal-associated matrices can serve as reservoirs and transmission pathways for antimicrobial-resistant bacteria. Screening for antimicrobial resistance (AMR) across interconnected environmental and animal sources is important for One Health surveillance, particularly in settings where humans, livestock and surface-water systems interact closely. Methods We conducted a cross-sectional environmental study across 13 sampling sites representing urban streams and rivers, rural rivers and Lake Victoria shoreline sites. Water samples were assessed for total coliform contamination, while coliform isolates recovered from water, animal feed and animal waste were identified and subjected to antimicrobial susceptibility testing. Resistance was assessed against seven antibiotics representing major antimicrobial classes. Multiple antibiotic resistance (MAR) indices were calculated. Differences in multi drug resistance (MDR) by sample source were evaluated using logistic regression, with water isolates as the reference category. Site-level coliform burden was correlated with mean MAR index using Spearman's rank correlation. Results A total of 79 coliform isolates were recovered, comprising 46 (58.2%) from water, 11 (13.9%) from animal feed and 22 (27.8%) from animal waste. Klebsiella pneumoniae (30/79, 38.0%) and Escherichia coli (27/79, 34.2%) were the predominant species. Overall, 20/79 (25.3%) isolates were MDR. Resistance was highest to amoxicillin (68.4%), followed by amoxicillin-clavulanic acid (40.5%) and gentamicin (29.1%), while no meropenem resistance was detected. Prevalence rates of MDR differed substantially by source: 54.5% in animal feed, 36.4% in animal waste, and 13.0% in water. Compared with water isolates, isolates from animal feed had eight-fold greater odds of MDR (OR 8.00, 95% CI 1.89–37.04; p = 0.005), while isolates from animal waste had approximately four-fold greater odds (OR 3.81, 95% CI 1.14–13.51; p = 0.032). Site-level coliform burden was not associated with mean MAR index (Spearman r = 0.106, p = 0.731; n = 13 sites). Conclusion Substantial antimicrobial resistance and MDR were detected across environmental and animal-associated coliform isolates, with particularly elevated MDR among isolates from animal feed and waste. The absence of an association between coliform abundance and MAR index suggests that bacterial contamination burden alone may not reliably indicate AMR risk. These findings support integrated One Health surveillance incorporating environmental waters, livestock-associated sources and antimicrobial resistance phenotyping.
Source: Research Square
Institution: Uzima University College
Language: English
Subfield: Pollution
Field: Environmental Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7225794609