Treating in the main tank: the molecules that shut down your biofilter
Published by Planète Pets · managing editor: Clément Lacaille · how our guides are written
The scenario is a classic and it ends badly: a fish falls ill, you pour the treatment into the main tank, the disease recedes, and then fish start dying a week later. The culprit is neither the disease nor the medication directly, but a rarely explained collateral damage: the treatment sterilised the biofilter, and ammonia climbed. One study measured precisely which molecules do this and which do not. Our maintenance guidance is gathered in the fish care section.
Contents · 4 sections
Fifteen molecules, two lists
Nimenya, Delaunois, La Duong, Bloden, Defour, Nicks and Ansay published an assessment in 1999, in the journal ATLA, of the short-term toxicity of fifteen pharmacological agents on nitrification in a closed aquatic system (see the study). The result reads as two columns.
- Dose-dependent inhibition of both ammonia removal and nitrate production: ampicillin, chloramine T, enrofloxacin, erythromycin, levamisole, methylene blue and polymyxin B.
- No significant effect: chloramphenicol, copper sulphate, kanamycin, malachite green, neomycin, penicillin G potassium, tetracycline, and the trimethoprim-sulfadoxine combination.
One case is worth flagging: enrofloxacin inhibited nitrate production without blocking ammonia oxidation, which is the signature of nitrite accumulation — only the second link in the chain is hit. The authors explicitly conclude that a separate hospital tank is needed for these molecules, to avoid ammonia poisoning.
How long does the filter stay down?
Collins, Gratzek, Dawe and Nemetz had shown as early as 1975, in the Journal of the Fisheries Research Board of Canada, that a single 5 mg/L dose of methylene blue was enough to halt nitrification completely for sixteen days in closed recirculating systems (see the study). Formalin, malachite green, copper sulphate, potassium permanganate and salt, by contrast, had no effect at the therapeutic doses tested.
Formalin deserves a more recent nuance: Fredricks, Cupp, Schleis, Erickson and Gaikowski showed in 2018, in Aquaculture Research, that from 37 mg/L upward total ammonia nitrogen and nitrite rose, with nitrification still not back to baseline eight days after exposure. Harmless at usual doses, damaging in repeated or overdosed courses.
Four habits that follow
- A hospital tank is mandatory for the seven molecules named. This isn't a precaution on principle, it is the authors' explicit recommendation. What goes into a quarantine tank is detailed in our hospital and quarantine tank kit guide.
- If the treatment genuinely has to cover the whole volume, favour the molecules documented as non-inhibitory — bearing in mind that "no effect on the filter" does not mean "harmless": copper remains lethal to shrimp and snails.
- Test ammonia and nitrite daily during treatment and for the following two to three weeks, and plan water changes. A "nitrite rising, ammonia normal" profile points to damage confined to the second link.
- Keep a seeded sponge permanently in the main filter: it starts a hospital tank instantly and reseeds the main filter if a treatment has sterilised it. Activated carbon removes the molecule; it does not bring bacteria back.
Frequently asked questions
Do these figures apply to my aquarium?
With caution. The 1999 study is in vitro over 24 hours, in synthetic water, with no fish and no plants: the ranking of molecules is solid, the threshold doses in a real aquarium are not. The 1975 one covers aquaculture systems, and the 2018 one professional biofiltration media.
Do off-the-shelf products count?
Many are formulations combining several molecules, which have not been tested as sold. Reading the composition and spotting the seven inhibitory active ingredients is more reliable than trusting the brand name.
Why do two aquariums react differently to the same treatment?
A 2016 study by Sipos and Urakawa shows that sensitivity to methylene blue varies a thousandfold depending on which nitrifying micro-organisms are present. One tank's bacterial population is not the neighbour's — hence the value of testing rather than assuming. For the reminder on ammonia toxicity itself, see our article on NH3 ammonia by pH and temperature. Find all our guides on the fishkeeping hub.
An accessory from our shop for this use: Hang-on isolation and breeding box for aquariums (€28.90) · the whole shop
This guide is part of Planète Pets’s Fishkeeping universe. Our advice is general in nature: for any health concern, your veterinarian remains the only reference.
