This paper uses Species Sensitivity Distributions (SSDs) to assess risks of veterinary antibiotics in European waters, but there is a big problem: it mixes toxicity data from different antibiotic classes without considering their different effects. This could lead to wrong conclusions about real environmental risks.
Main Issues:
1. SSDs Combine Antibiotics with Different Mechanisms: Antibiotics like tetracyclines (block protein synthesis) and fluoroquinolones (damage DNA) work in completely different ways.
But the study puts them all into the same SSD model. Is this correct? If one antibiotic kills bacteria and another harms algae, can we really compare them like this?
2.Most Data Comes from Just a Few Species: Over 30% of the toxicity data is from algae and Daphnia, but some antibiotics (like macrolides) mainly affect bacteria.
Are these SSDs really showing the full risk? If cyanobacteria (key for ecosystems) are more sensitive but not included enough, the risk might be underestimated.
3. Risk Quotients (RQs) Ignore Real Exposure Conditions: The study calculates RQs by comparing antibiotic levels in different waters (WWTPs, rivers, hospitals).
But antibiotics behave differently in each place, some break down fast, others last longer. The RQ method treats them all the same. Is this realistic?
4. Missing the Biggest Problem: Antibiotic Resistance: The study looks at direct toxicity (death, growth effects) but ignores resistance genes.
Even low antibiotic levels can make bacteria resistant. Shouldn’t this be part of the risk assessment?
In My Opinion, the Analysis Should Be Performed Differently:
1. Group Antibiotics by How They Work, Not Just Chemical Class: Separate cell wall inhibitors (e.g., β-lactams) from protein synthesis blockers (e.g., macrolides) because they harm different organisms.
2. Test More Bacteria and Cyanobacteria, Not Just Algae and Daphnia: For antibiotics that kill bacteria, the SSD should focus on microbes, not just bigger species.
3. Check Mixture Effects: Antibiotics are never alone in nature. Do they become more toxic together? Models like Concentration Addition should be used.
4. Use Probabilistic Risk Assessment: Instead of simple averages, run Monte Carlo simulations to account for changing antibiotic levels over time and space.
Final Question:
The study says macrolides and fluoroquinolones are the riskiest, but if the SSDs don’t properly separate different antibiotic types or include resistance effects, can we really trust these rankings? A better approach would focus on mechanism-specific toxicity and long-term resistance risks, not just short-term damage.