Veterinary medicines in water: moving from detection to risk

Finding a veterinary medicine in river water, effluent or sediment does not automatically mean that fish will become ill. Chemical analysis establishes possible exposure. Risk then depends on concentration, duration, transformation products, organism sensitivity and ecosystem function. Keeping that distinction clear helps direct action without dismissing releases or treating every analytical detection as a predicted ecological injury.

Antibiotics and antiparasitic medicines can reach water through excretion, uneaten medicated feed, cleaning, runoff or inappropriate disposal. Their environmental profiles differ substantially. A useful assessment has to connect field measurements, chronic-effect data and knowledge of local use.

A concentration is not the whole exposure

A grab sample describes one place and one moment. It may miss an intermittent discharge or capture a brief peak that does not represent long-term conditions. Flow, rainfall, season, temperature, sunlight and organic matter influence dilution and breakdown. Some compounds remain mainly in the water column; others bind to particles or move into sediment.

Sampling should cover plausible pathways: upstream and downstream locations, effluent, the mixing zone, sediment and, where justified, sentinel organisms. Detection limits, blanks and sample preservation belong in the report. Without that context, comparisons between campaigns or sites can be misleading.

Bioaccumulation cannot be inferred from persistence in water alone. It depends on a chemical’s properties and ionisation, species metabolism and route of exposure. A residue found in tissue also demonstrates exposure rather than injury by itself; it needs to be interpreted alongside a biological endpoint.

Acute, chronic and indirect effects answer different questions

Acute tests identify rapid effects, often at relatively high concentrations. They may not represent reproduction, growth, development or microbial interactions under prolonged exposure. Lee and colleagues therefore exposed Daphnia magna for 21 days and juvenile medaka, Oryzias latipes, for 40 days to three veterinary antibiotics. No-observed-effect concentrations differed widely by chemical and organism. For Daphnia reproduction they were 27.2 mg/L for amoxicillin, 3.3 mg/L for enrofloxacin and 0.15 mg/L for neomycin; for juvenile medaka survival they were 21.8, 3.2 and 0.87 mg/L respectively.

Those experimental values are not universal water-quality thresholds. They contribute to an assessment that includes uncertainty factors and field exposure. In the same study, predicted no-effect concentrations were 0.078 µg/L for amoxicillin, 4.9 µg/L for enrofloxacin and 3.0 µg/L for neomycin. Risk quotients based on average environmental concentrations remained below 1, while maxima produced quotients of 21.2 for amoxicillin and 6.1 for enrofloxacin. An acceptable average can therefore coexist with local or episodic hotspots.

Indirect effects add another layer. An antibiotic may alter bacteria, biofilms or decomposers before a direct effect on fish is visible. Changes in decomposition or food resources can propagate through a food web. Ecotoxicology consequently needs endpoints at more than one biological level.

Veterinary medicines cannot be ranked as one class

Kołodziejska and colleagues compared doramectin, metronidazole, florfenicol and oxytetracycline in bacteria, algae, duckweed and daphnids. Their results are useful not as a permanent league table but as evidence that the most sensitive endpoint changes with compound and organism. An antibiotic is designed to act on microorganisms; an antiparasitic intended to control invertebrates may also affect non-target invertebrates.

Bundschuh and colleagues examined veterinary pharmaceuticals in a system combining a microbial decomposer and an aquatic detritivore. Their study highlights ecosystem functions and interspecies relationships that a single-species test cannot capture. A broader test battery reduces the chance of drawing reassurance from the least sensitive taxon.

Co-occurring residues create further uncertainty. Mixtures may act additively, antagonistically or synergistically, while most data concern individual compounds. Metabolites and transformation products also need consideration. A precautionary margin is reasonable where exposure is chronic, repeated or close to sensitive habitat.

Antibiotic risk extends beyond conventional toxicity

For antibiotics, selection for resistant bacteria is distinct from fish mortality or reproduction. Concentrations too low to produce an obvious animal effect may still alter microbial communities. Resistance genes and their hosts can then move among water, sediment, biofilms and digestive systems.

A programme focused only on animal toxicity could miss that dimension. Residue chemistry, microbiology, usage and discharge pathways should be assessed together without assigning every resistance finding to one source. Human, veterinary and environmental inputs overlap, making local investigation essential.

In aquaculture, prudent use begins with an appropriate diagnosis and prescription. Medicated feed delivery should minimise loss, reflect actual appetite and be followed through the treatment period. Leftover products, cleaning water and packaging require controlled handling. Preventive health and biosecurity reduce the need for medication, although they cannot promise a disease-free facility.

Build a proportionate and testable response

Four questions organise the response to a detected residue: is the measurement reliable, is exposure brief or sustained, which organisms or functions are sensitive, and which source can realistically be reduced? A targeted follow-up campaign may be more informative than accumulating incomparable analyses.

Priorities include concentrated releases, low-flow periods, biologically active compounds at low concentrations and sites where dilution is limited. Collecting unused medicines, improving effluent management, training staff and recording treatments are practical controls. Their value should be checked using before-and-after indicators.

Communication should preserve units, distinguish measured concentrations from predicted no-effect values and state uncertainty factors. The accurate message is neither “any trace is toxic” nor “dilution solves everything.” Vetofish can support sampling design, result interpretation and source reduction within a One Health framework.