Tritrichomonas foetus diarrhoea in cats: does ronidazole clear infection and resolve signs vs no specific treatment? — PROBABLE

PROBABLE: Ronidazole clears Tritrichomonas foetus infection in cats, on the strength of one decisive randomised, double-blind, placebo-controlled trial in naturally infected cats, supported by two small placebo-controlled experimental studies — with the important dose caveat that the low dose tested first failed. In Grellet 2017 (PMID: 26662037; 47 cats naturally infected with T. foetus in French catteries, randomly allocated to guar gum-based ronidazole capsules 30 mg/kg orally once daily for 14 days, n = 25, or placebo, n = 22), T. foetus was still detected by conventional PCR at the end of the study in 16% of treated cats (4/25) vs 82% of placebo cats (18/22), P < 0.001 — an absolute difference of 66 percentage points in persistent infection, i.e. clearance in 84% vs 18% — and no clinical signs of adverse drug reactions were observed. In Lim 2012 (PMID: 22711930; 6 experimentally infected kittens, 3 treated and 3 placebo), ronidazole 50 mg/kg twice daily for 14 days made all 3 treated cats negative by direct smear, culture and nested PCR during treatment and 4 weeks of follow-up, while all 3 placebo cats remained positive throughout. In Gookin 2006 (PMID: 16734086; publication type Randomized Controlled Trial; 10 specific-pathogen-free kittens experimentally infected, 5 placebo and 5 ronidazole), the initial dose of 10 mg/kg orally twice daily for 14 days produced initial improvement but infection relapsed in all 5 treated cats 2 to 20 weeks after treatment; when the relapsed and placebo cats were subsequently treated at 30 or 50 mg/kg twice daily for 14 days (an uncontrolled second phase), 10/10 cats were negative for follow-up of 21 to 30 weeks, and in the one naturally infected cat in that study ronidazole first abolished diarrhoea and infection for 85 days before relapse, then eradicated both for over 407 days on retreatment — the only carded diarrhoea-resolution data, and they come from a single cat plus the experimental cats' course. Design class (a): randomisation and blinding are established by Grellet 2017, a controlled, randomised, double-blind trial answering this shelf's own comparison (ronidazole vs placebo) in this shelf's own population (naturally infected cats); Lim 2012 is placebo-controlled and its full text states randomisation ('Cats were randomly divided into 2 equal groups') but does not state masking, and Gookin 2006's high-dose phase had no concurrent control, so neither would reach class (a) alone. The stamp is PROBABLE for clearing infection at the tested effective doses (guar gum-based capsules 30 mg/kg once daily, or 30–50 mg/kg twice daily in the experimental studies), not for the 10 mg/kg twice-daily dose, which failed in every cat. Two qualifications keep it from a stronger reading: the decisive trial tested a guar gum-based (delayed-release) capsule formulation, so its result is a result for that formulation at that dose; and diarrhoea resolution — half of this shelf's question — was not a reported outcome in the decisive trial's abstract, resting instead on Gookin 2006's single naturally infected cat and on the uncontrolled colony series cross-pointed below (PMID: 31796193: 11 of 14 initially positive cats remained negative with no diarrhoea observed six to eight weeks after ronidazole 30 mg/kg once daily, but 9 cats showed mild-to-moderate neurological signs during treatment, six of whom had shown similar signs before treatment). Resistance and natural history also bound the claim: ronidazole resistance is documented in vivo and in vitro (PMID: 20492492), a resistant isolate pair self-resolved without antimicrobial intervention over roughly three years (PMID: 34735847), and placebo cats in Grellet 2017 were still PCR-positive at 82% at study end, so short-term spontaneous clearance is uncommon even though long-term self-resolution can occur. Tier C is not engaged: no mortality or survival claim is made. Drafted Nov 10, 2026 (VetMed build, topic #735); figures from the abstracts, fetched and verified at drafting.