Metronidazole in Dogs and Cats: A Clinical Evidence Review

Bottom line

Metronidazole has two genuinely evidence-supported roles in small-animal practice: Giardia duodenalis infection in dogs, which carries an FDA approval [1], and infection caused by obligate anaerobes, which its spectrum actually covers [2]. It is not a first-line drug for uncomplicated acute diarrhoea: in a randomised controlled trial in canine acute colitis, dogs given metronidazole alongside an easily digestible diet reached remission in a median of 8.5 days versus 5 days on diet alone [3], and as of August 2026 the ENOVAT guidelines recommend strongly against antimicrobials in dogs with mild or moderate acute diarrhoea [4]. Set against that thin benefit are a measurable faecal microbiome disturbance that outlasts the course [5] and a well-described neurotoxic syndrome [6]. Reserve it for a defined target, use the shortest effective course, and stop reaching for it reflexively.

Drug facts

Class and mechanism. Metronidazole is a 5-nitroimidazole, a group of drugs "that have both antiprotozoal and antibacterial activity." [2] It is effectively a prodrug requiring intracellular reduction: metronidazole "undergoes reduction in bacteria where it is metabolized to cytotoxic derivatives that bind to DNA, causing loss of the helical structure, strand breakage, and impairment of DNA function." [2] That reduction step defines the spectrum — "it is active against obligate anaerobic bacteria; however, aerobic bacteria lack the reductive pathway necessary for its mechanism of activity." [2]

Spectrum. At concentrations readily attained in serum after oral or parenteral administration, metronidazole is active against Bacteroides fragilis, B. melaninogenicus, Fusobacterium spp, and Clostridium perfringens and other Clostridium spp, and is "generally less active against non–spore-forming, gram-positive bacilli such as Actinomyces, Propionibacterium, Bifidobacterium, and Eubacterium spp." [2] Among the antiprotozoals, nitroimidazoles with activity against trichomonads and amebae include metronidazole, tinidazole, nimorazole, flunidazole and ronidazole. [2] Because aerobes are untouched, metronidazole is a combination partner in mixed infection, never a solo agent.

Distribution and metabolism. Metronidazole "penetrates the blood-brain barrier and also attains therapeutic concentrations in abscesses and in empyema fluid," and primary metabolism occurs in the liver. [2] That penetration is what underwrites its use in anaerobic CNS, oral, pleural and hepatic infection.

Formulations and label status. AYRADIA (metronidazole oral suspension, 125 mg/mL, Virbac AH) was approved under NADA 141-572 on 11 October 2023, is prescription-only, and is "indicated for the treatment of Giardia duodenalis infection in dogs" at a dosage regimen of "25 mg/kg (11.3 mg/lb) of body weight, using the supplied syringe, twice daily for five consecutive days." [1] The FDA's Center for Veterinary Medicine describes the product as "the first FDA-approved animal drug for treating Giardia duodenalis infections" and states that it "is only approved for use in dogs." [7] Because that is the label's single stated indication [1] and the product is approved in dogs only [7], every other use — all use in cats, anaerobic infection, hepatic encephalopathy, chronic enteropathy — is extralabel, usually with human-label tablets, capsules, suspension or injectable metronidazole.

Contraindications and cautions. Carcinogenicity and teratogenicity have been reported in laboratory mice after prolonged exposure to metronidazole, "and therefore it has been banned in food-producing animals and should be avoided in pregnant animals." [2] In the US, nitroimidazoles including metronidazole are prohibited from extralabel drug use in all food-producing animal species. [2] Dose reduction is required with hepatic dysfunction, and cumulative exposure drives neurotoxic risk; both are covered below.

Where metronidazole earns its place: Giardia and anaerobic infection

Giardia in dogs is the strongest indication, and the only labelled one. In the FDA dose-confirmation laboratory study, metronidazole oral suspension "was over 99% effective at reducing Giardia trophozoites in treated dogs," and in the field study it "was over 99% effective at reducing Giardia cysts in treated dogs." [1] The pivotal double-masked, vehicle-controlled, randomised multicentre field trial enrolled 180 dogs with 129 included in the efficacy analysis, and reported "a 99.92% reduction in G. duodenalis cyst counts, significantly higher than the reduction in the control group," with adverse events similar between both groups (10%), mainly diarrhoea and vomiting, and treatment "readily accepted by 99% of dogs" at 0.2 ml/kg twice daily for 5 days. [8]

Against fenbendazole, it performs comparably rather than better. A field trial in 24 naturally infected dogs allocated two groups of 12: fenbendazole at 50 mg/kg orally once a day for 5 consecutive days, and metronidazole at 50 mg/kg orally once a day for 5 consecutive days. [9] After the first therapy, on day 7, 4/12 (33.3%) dogs tested positive for Giardia cysts in the fenbendazole group and 5/12 (41.7%) in the metronidazole group; efficacies — calculated as reduction in mean Giardia cyst counts — at days 7, 14, 21 and 50 were 80.9, 94, 100 and 97% for fenbendazole and 70.8, 99, 100 and 97.1% for metronidazole, with no statistically significant difference between them (P = 0.686). [9] Read that table carefully: dogs still shedding after the first course were retreated at day 7, 14 or 21 until negative, so the 100% figure at day 21 reflects repeat courses rather than a single five-day course. [9] The Merck Veterinary Manual lists the giardiasis dose for dogs and cats as "25 mg/kg, PO, every 12 hours for 5 days." [2]

Anaerobic infection is the other defensible target. Where an obligate anaerobe is plausible or confirmed — anaerobic sepsis, deep oral and periodontal infection, pyothorax, hepatic abscessation, some CNS infection — the spectrum and the tissue penetration line up, and metronidazole is a rational component of a combination protocol. [2] This is a very different clinical situation from an otherwise-well dog with three days of soft stool.

Acute diarrhoea: what the controlled trials actually show

The honest answer is that the trials disagree, and the weight has moved against routine use. Four randomised trials bear directly on this, and they should be read individually rather than pooled by impression.

The one positive trial. In a randomised, double-blinded, placebo-controlled trial, dogs with acute diarrhoea in which causation was not determined by routine faecal diagnostics received metronidazole at 10–15 mg/kg PO q12h for 7 days or placebo; 31 dogs completed the study, 14 in the metronidazole group and 17 controls. [10] "The mean ± SD time to resolution of diarrhea for test population dogs (2.1 ± 1.6 days) was less than that for controls (3.6 ± 2.1 days, P = .04)." [10] On carriage, "only 3 of 13 (23.1%) test population dogs had persistent C. perfringens carriage at day 7, which was less than the 11 of 14 (78.6%) controls with persistent growth (P = .007)." [10] The authors concluded that metronidazole "can shorten duration of diarrhea and decrease fecal culture detection of C. perfringens in some dogs with acute nonspecific diarrhea," while adding that "additional studies are needed to assess the benefits and risks of routine use of metronidazole for this purpose because most dogs achieve resolution of diarrhea within several days regardless of treatment." [10]

The null trial. A randomised, double-blinded, placebo-controlled trial of 60 dogs without concurrent comorbidities, randomised into three treatment groups, found dogs "achieved acceptable fecal consistency after 3.5 ± 2.2 days when receiving probiotic, 4.6 ± 2.4 days with oral metronidazole, and 4.8 ± 2.9 days with placebo; statistically significant differences were not identified between treatment groups (p = 0.17)." [11] The authors stated these findings "failed to provide evidence for the common use of metronidazole in this cohort of dogs with acute canine diarrhea." [11]

The trial where metronidazole did worse. In 59 client-owned dogs with noninfectious acute colitis randomised into three placebo-controlled groups, "median remission time was significantly different among the 3 groups (P < .01) with median times of 5 days (range, 4 to 10) for group 1 [easily digestible diet + placebo], 8.5 days (range, 7 to 12) for group 2 [easily digestible diet + metronidazole], and 5 days (range, 3 to 6) for group 3 [psyllium-enhanced easily digestible diet + placebo]." [3] "Metronidazole addition affected the fecal dysbiosis index negatively at days 7 to 10," and the authors concluded that dietary management "proved a superior management strategy compared to metronidazole." [3]

The add-on trial in haemorrhagic disease. In a prospective randomised blinded trial in primary care, dogs with haemorrhagic diarrhoea of less than 3 days' duration requiring IV fluids received either metronidazole or saline in addition to amoxicillin-clavulanic acid, fluids, buprenorphine and omeprazole; 34 cases completed. [12] "There was no significant difference in hospitalisation time between treatment groups (mean for dogs receiving metronidazole was 29.6 hours and for controls was 26.3 hours) nor in daily clinical scores," and the authors concluded that metronidazole "is not an essential addition to amoxicillin-clavulanic acid therapy for treatment of severe cases of haemorrhagic diarrhoea in dogs." [12]

The pooled view. The systematic review and meta-analysis underpinning the ENOVAT guidelines identified six randomised controlled trials of antimicrobial treatment meeting eligibility criteria and found that "high certainty evidence showed that antimicrobial treatment did not have a clinically relevant effect on any outcome in dogs with mild or moderate disease," with certainty of evidence low for dogs with severe disease. [13]

Synthesis. Only one of these trials found metronidazole superior to placebo, in 31 dogs, with a difference of roughly a day and a half in a condition that resolves on its own; the others found no difference, no added value on top of an aminopenicillin, or a longer time to remission than diet alone. A defensible clinical reading is that for uncomplicated acute diarrhoea the expected benefit is small at best, absent in most cohorts, and plausibly negative in colitis — while the microbiome and neurologic costs are real and reproducible. That asymmetry, not any single trial, is the argument against reflex prescribing.

Chronic enteropathy: a shrinking and contested role

The "antibiotic-responsive" category is being actively dismantled. Dogs "that show a marked and long-lasting improvement or resolution of their clinical signs after administering an antibiotic (metronidazole and/or tylosin) for 2 weeks were previously classified as ARE." [14] The problem is durability and collateral damage: "antimicrobial treatment can have a long-lasting effect on the intestinal microbiota, increasing antimicrobial resistance is of major global concern, and most dogs that initially respond to an antibiotic trial will relapse after discontinuing treatment." [14] Consequently, "the true existence of the ARE subgroup (proposed to be renamed 'idiopathic intestinal dysbiosis') of canine CIE and the appropriateness of empirical antimicrobial treatment trials in the diagnostic work-up of canine CIE is currently debated." [14] In protein-losing enteropathy the position is firmer: "absent an infectious etiology, antimicrobial treatment plays no role in canine PLE." [14]

Practically, this means a sequenced diet trial rather than an empirical antibiotic trial is the better first move in chronic signs, and that a response to metronidazole should not be treated as a diagnosis. Where the Merck Veterinary Manual still lists a dose for "inflammatory GI conditions or inflammatory bowel disease (IBD)" of "10–15 mg/kg, PO, every 12 hours" [2], that figure should be read as a formulary entry, not as an endorsement of routine use.

What metronidazole does to the faecal microbiome

The disturbance is large, measurable, and outlasts the course — but the recovery curve is more nuanced than "the microbiome takes a month to recover." In 24 healthy pet dogs divided into three groups, with metronidazole given at 15 mg/kg PO q12h for 2 weeks, metronidazole "significantly changed microbiome composition … including decreases in richness (P < .001) and in key bacteria such as Fusobacteria (q < 0.001) that did not fully resolve 4 weeks after metronidazole discontinuation." [5] The faecal dysbiosis index "was significantly increased (P < .001)," accompanied by "increased fecal total lactate (P < .001), and decreased secondary BAs deoxycholic acid and lithocholic acid (P < .001)." [5]

The time course is worth stating precisely, because the two halves are often conflated. The dysbiosis index itself normalised relatively quickly: "after 2 weeks from the end of metronidazole administration, the DI was already no longer significantly altered compared to baseline (day 28: P = .74, day 42: P > .99)." [5] Specific taxa did not: "four weeks after the cessation of metronidazole (day 42), Fusobacteria abundance, comprising almost exclusively genus Fusobacterium, remained significantly reduced (day 0: 14.49% vs day 42: 1.8%, P = .03)." [5] The authors' conclusion is "a minimum 4-week effect of metronidazole on fecal microbiome and metabolome, supporting a cautious approach to prescription of metronidazole in dogs." [5] The same signal appears in clinical disease, where metronidazole addition "affected the fecal dysbiosis index negatively at days 7 to 10." [3]

The clinically useful framing: a composite index can look recovered while the community has not returned to baseline, and the loss of secondary bile acid producers is a mechanistically plausible route to prolonged loose stool after a course — the outcome the drug was given to prevent.

Neurotoxicity: recognising it, and the dose-duration relationship

Metronidazole neurotoxicosis is a central vestibular and cerebellar syndrome that resolves on withdrawal, and it occurs at lower doses than the classical teaching suggests. The Merck Veterinary Manual notes that "high doses may induce clinical signs of neurotoxicity in dogs, such as tremors, muscle spasms, ataxia, and even convulsions." [2]

The larger and more recent of the two published case series is the most useful for calibration. Across 26 dogs from four referral hospitals, median age at presentation was 7.2 years (range, 0.1–12 years); "median duration of treatment was 35 days (range, 5–180 days); median treatment dosage was 21 mg/kg BID (range, 13–56 mg/kg every 12 h); median resolution of the clinical signs upon discontinuation of metronidazole was 3 days (range, 1–26 days)." [6] MRI of the brain was performed in 19 cases "and only one dog had brain lesions affecting the dentate nuclei," so a normal MRI does not exclude the diagnosis. [6] Critically, those authors "found evidence of neurotoxicity in dogs at much lower doses than previously reported" and "suggest caution when administering metronidazole at doses > 40 mg/kg every 24 h, regardless of the duration of the treatment." [6]

That last point deserves emphasis because it is routinely mis-taught. In the earlier retrospective series of 21 dogs with metronidazole toxicosis, "the average dosage and duration of metronidazole administration for the diazepam-treated and untreated groups were 60.3 mg/kg/d for 44.9 days and 65.1 mg/kg/d for 37.25 days." [15] Those are descriptions of the dogs that became toxic, not a safety threshold — treating roughly 60 mg/kg/day as a ceiling below which the drug is safe is not supported, and the 26-dog series documents toxicity well below it. [6]

Management is withdrawal plus diazepam. Drug discontinuation with supportive therapy alone gives reported recovery times of 1–2 weeks. [15] In the retrospective comparison, 13 dogs treated with diazepam — "an initial i.v. bolus and then diazepam PO q8h for 3 days," with "the average dosage of both the i.v. and PO diazepam" being 0.43 mg/kg — were compared with 8 dogs receiving only supportive care: "the average response time for the diazepam-treated dogs was 13.4 hours compared to 4.25 days for the untreated group," and "recovery time also was markedly shorter for the diazepam-treated dogs (38.8 hours) compared to the untreated group (11 days)." [15] This is a small retrospective study rather than a randomised trial, but the effect size is large and the intervention is low-risk.

Regulatory safety data give a complementary picture of the dose-response. In target animal safety work, dogs were given 1X, 2X or 3X the labelled dose for 15 consecutive days, or 5X the labelled dose for 5 consecutive days; "dogs in the 5X group had increased episodes of ear erythema and diarrhea compared to the other groups," both of which "resolved without treatment," and treatment at five times the intended dosage "was associated with self-limiting episodes of diarrhea and of erythema of the ears." [1] In a separate tolerance study, two dogs given an investigational metronidazole oral suspension at 500 mg/kg/day (10X the intended daily dose) for 7 days "exhibited severe neurologic signs by Days 7 and 8," with "ataxia and lack of ocular reflexes … observed in both dogs" and one dog additionally showing "lateral movements of head and eyes, recumbency, tremors, and mydriasis"; after stopping the drug and giving repeated doses of diazepam and furosemide, "both dogs recovered fully within approximately 24 hours." [1] No adverse neurologic signs were reported in the field study at the labelled dose. [1]

Other adverse effects, hepatic considerations, and cautions in cats

Gastrointestinal upset is the common adverse effect. In the FDA field study, "the most common adverse events seen during the study were vomiting and diarrhea, which were generally isolated and resolved without treatment," and the low incidence made direct attribution to the drug impossible. [1] In the peer-reviewed field trial, adverse events were similar between treated and control groups at 10%, mainly diarrhoea and vomiting. [8] Beyond the trial data, formulary sources such as Plumb's list anorexia and hypersalivation as the other frequently reported complaints, the latter particularly in cats.

Hepatic disease requires dose reduction, not avoidance. Primary metabolism occurs in the liver [2], and "small animals with hepatic dysfunction require lower dose rates of 15–20 mg/kg, PO, every 24 hours." [2] This matters twice over: hepatic encephalopathy is itself an indication, listed at "7.5 mg/kg, PO, every 8–12 hours" [2], and those patients are by definition the ones least able to clear the drug.

Cats: bitter, variably absorbed, and genotoxic in vitro. Metronidazole base is intensely bitter and reliably provokes hypersalivation, foaming and food aversion in cats; formulary sources such as Plumb's describe the benzoate salt as the more palatable option, and it is what feline pharmacokinetic work has used. In a single-dose study, cats received "either 5 mg/kg metronidazole intravenously, or 20 mg/kg metronidazole benzoate (12.4 mg/kg metronidazole base) orally"; systemic absorption was variable (mean = 65 +/- 28%) with a peak of 8.84 +/- 5.4 microg/ml at 3.6 +/- 2.9 h, and the terminal half-life was 5.34 h from the intravenous dose and 5.16 h from the oral dose. [16] The same study found that "genotoxicity was detected at all concentrations of metronidazole in feline PBMC and the T-cell lymphoma line in vitro," and "genotoxicity was also observed in PBMC collected from cats after 7 days of oral metronidazole but resolved within 6 days of discontinuing metronidazole." [16] The reversibility is reassuring; the finding still argues for short, targeted courses in cats rather than open-ended ones, and it is a further reason not to use the drug as an empirical GI sedative.

Drug interactions. These are formulary-level rather than trial-derived: Plumb's and comparable references list metronidazole as reducing the metabolism of warfarin with a resulting increase in anticoagulant effect, note that phenobarbital may accelerate metronidazole metabolism while cimetidine may reduce its clearance, and describe a disulfiram-like reaction with alcohol reported in people. Check the anticoagulant interaction specifically before co-prescribing.

Dosing considerations and monitoring

Doses below are as published; all use in cats and all canine use other than giardiasis is extralabel.

• Giardia, dogs (labelled): "25 mg/kg (11.3 mg/lb) of body weight, using the supplied syringe, twice daily for five consecutive days." [1]
• Giardiasis, dogs and cats (formulary): "25 mg/kg, PO, every 12 hours for 5 days." [2]
• Inflammatory GI conditions or IBD: "10–15 mg/kg, PO, every 12 hours." [2]
• Hepatic encephalopathy: "7.5 mg/kg, PO, every 8–12 hours." [2]
• Hepatic dysfunction: "15–20 mg/kg, PO, every 24 hours." [2]

Two practical checks follow directly from those numbers. First, compare any intended regimen against the neurotoxicity caution of "> 40 mg/kg every 24 h, regardless of the duration of the treatment" [6]: a 10–15 mg/kg q12h regimen [2] sits at 20–30 mg/kg/day, below that caution threshold — though below the threshold is not the same as safe, because the reported dosage range in that same series runs down to 13 mg/kg every 12 h, which is 26 mg/kg/day and falls inside that very band [6]. A 25 mg/kg q12h regimen [1][2] sits at 50 mg/kg/day, above the threshold — acceptable for a defined five-day giardiasis course, but not a regimen to extend open-endedly. Second, duration is the variable most under your control: the ENOVAT panel advises that "antimicrobial therapy should not extend beyond clinical resolution" and that "for the majority of dogs, treatment of 3-7 days is likely adequate to obtain clinical resolution." [4]

Monitoring is clinical rather than laboratory. No therapeutic drug monitoring for metronidazole is in routine clinical use in dogs or cats, and none of the sources cited here describes one, so — as an authorial judgement rather than a guideline directive — the meaningful surveillance is a deliberate neurologic check at each recheck for any course beyond a few days — gait, menace, nystagmus, head tilt — plus owner instructions to stop the drug and call if ataxia, circling, tremors or abnormal eye movements appear. Given that the median duration of treatment in that 26-dog series was 35 days, with a range starting at 5 days [6], "it has only been a week" is not reassurance.

Antimicrobial stewardship: when not to reach for it, and what to use instead

For mild and moderate acute diarrhoea, the guidance is unambiguous. ENOVAT recommends against antimicrobials in dogs with acute non-haemorrhagic diarrhoea and mild disease, and separately in acute haemorrhagic diarrhoea with mild disease — both "strong recommendation, high-certainty evidence," with 100% panel agreement. [4] The same strong, high-certainty recommendation against antimicrobials applies to moderate disease, defined as "dogs with impaired general condition and varying degrees of dehydration/hypovolemia," where systemic signs "will resolve with adequate fluid therapy," in both haemorrhagic and non-haemorrhagic presentations. [4] The stated exception is haematological: "dogs with laboratory values indicative of severe or overwhelming inflammation, such as severe neutrophilia (> 25x10^9/L), neutropenia and/or degenerative left-shift, represent an exception." [4]

For severe disease, antimicrobials are indicated — but metronidazole is not the suggested first-line drug. In dogs with signs of systemic disease persisting despite adequate fluid therapy, the panel suggests systemic antimicrobials (conditional recommendation, very low-certainty evidence), given parenterally, with "ampicillin (alternatively amoxicillin-clavulanate) or trimethoprim/sulfonamides as first line drugs" in non-critical illness, and a four-quadrant protocol in critical illness or where resistance is more likely. [4] Metronidazole's role here is as part of a four-quadrant combination for anaerobic coverage, not as a stand-alone response to bloody stool.

What to do instead in uncomplicated cases. Diet is the intervention with the best trial support: an easily digestible diet, with or without psyllium enhancement, outperformed the same diet plus metronidazole in acute colitis. [3] Fluid and electrolyte correction addresses what actually makes these patients ill. [4] Probiotics are genuinely equivocal — ENOVAT "do[es] not recommend either for, or, against use of probiotics," noting "the trade-offs are closely balanced" on moderate-certainty evidence [4], and in the randomised trial the probiotic arm reached acceptable faecal consistency in 3.5 ± 2.2 days versus 4.8 ± 2.9 days on placebo, a difference that did not reach significance. [11] Symptomatic control of vomiting and nausea with an antiemetic such as maropitant, or a prokinetic such as metoclopramide where hypomotility is the problem, targets the signs that distress the owner without the collateral cost.

The wider pattern. Metronidazole for acute diarrhoea belongs to the same family of reflex GI prescribing as gastroprotectant overuse: a drug with real indications, given far outside them, on the strength of habit and a self-limiting disease that would have resolved anyway. The discipline is the same in both cases — name the target before reaching for the bottle, and set a stop date when you start.

Frequently Asked Questions

Does metronidazole actually work for acute diarrhoea in dogs? The controlled trials disagree, and most do not support it. One randomised placebo-controlled trial in 31 dogs found a shorter time to resolution with metronidazole (2.1 +/- 1.6 days versus 3.6 +/- 2.1 days, P = .04). But a randomised trial in 60 dogs found no significant difference between probiotic (3.5 +/- 2.2 days), metronidazole (4.6 +/- 2.4 days) and placebo (4.8 +/- 2.9 days, p = 0.17); a randomised trial in 59 dogs with acute colitis found metronidazole plus diet took longer to remission (median 8.5 days, range 7 to 12) than diet plus placebo (median 5 days, range 4 to 10, P < .01); and adding metronidazole to amoxicillin-clavulanate in haemorrhagic diarrhoea did not change hospitalisation time (29.6 versus 26.3 hours). The ENOVAT meta-analysis concluded on high-certainty evidence that antimicrobials had no clinically relevant effect on any outcome in mild or moderate disease.

What is the metronidazole dose for Giardia in dogs? The FDA-approved product (AYRADIA, metronidazole oral suspension 125 mg/mL) is labelled at 25 mg/kg (11.3 mg/lb) twice daily for five consecutive days in dogs. The FDA describes it as the first FDA-approved animal drug for treating Giardia duodenalis infections, and it is approved for use in dogs only. The Merck Veterinary Manual lists the same regimen for giardiasis in dogs and cats: 25 mg/kg, PO, every 12 hours for 5 days. Note that 25 mg/kg q12h equals 50 mg/kg/day, above the > 40 mg/kg every 24 h threshold at which one case series advises caution, so it is appropriate as a defined five-day course but should not be extended open-endedly.

At what dose does metronidazole cause neurotoxicity in dogs? Lower than commonly taught, and there is no validated safe ceiling. In a series of 26 affected dogs, the median treatment dosage was 21 mg/kg BID (range, 13 to 56 mg/kg every 12 h) and the median duration of treatment was 35 days (range, 5 to 180 days); those authors found neurotoxicity at much lower doses than previously reported and suggested caution above 40 mg/kg every 24 h regardless of treatment duration. An older series of 21 toxicosis cases recorded average dosages of 60.3 mg/kg/d for 44.9 days and 65.1 mg/kg/d for 37.25 days, but those figures describe the affected dogs rather than defining a threshold, and should not be used as a safety cut-off.

How do you treat metronidazole neurotoxicosis? Stop the drug and give diazepam. With discontinuation and supportive care alone, reported recovery times are 1 to 2 weeks. In a retrospective comparison, 13 dogs given an initial IV bolus then diazepam PO q8h for 3 days (average dose of both the IV and oral diazepam 0.43 mg/kg) had an average response time of 13.4 hours and recovery time of 38.8 hours, versus 4.25 days and 11 days in 8 dogs receiving supportive care only. Expect improvement rather than instant resolution: in a 26-dog series, median time to resolution after stopping the drug was 3 days (range, 1 to 26 days). MRI is often unremarkable, with only 1 of 19 imaged dogs showing dentate nuclei lesions.

How long does the dog microbiome take to recover after metronidazole? Longer than the course, and the answer depends on what you measure. In 24 healthy dogs given 15 mg/kg PO q12h for 2 weeks, the faecal dysbiosis index rose significantly during treatment but was no longer significantly different from baseline two weeks after the course ended (day 28: P = .74, day 42: P > .99). Individual taxa lagged well behind: four weeks after cessation, Fusobacteria abundance remained significantly reduced (day 0: 14.49% versus day 42: 1.8%, P = .03), and secondary bile acids deoxycholic and lithocholic acid were decreased. The authors described a minimum 4-week effect on the faecal microbiome and metabolome.

Is metronidazole safe in cats? It is usable but should be short and targeted, and all feline use is extralabel. Cats find the base intensely bitter, so hypersalivation, foaming and food aversion are common, and the benzoate salt is the more palatable option. In a single-dose study, cats given 5 mg/kg IV or 20 mg/kg metronidazole benzoate (12.4 mg/kg metronidazole base) orally showed variable systemic absorption (mean 65 +/- 28%), a peak of 8.84 +/- 5.4 microg/ml at 3.6 +/- 2.9 h, and a terminal half-life of 5.34 h intravenously and 5.16 h orally. The same study detected genotoxicity at all metronidazole concentrations in feline cells in vitro and in blood mononuclear cells after 7 days of oral dosing, though this resolved within 6 days of discontinuing the drug.

Is metronidazole still indicated for chronic enteropathy in dogs? Its role is shrinking and actively debated. Dogs improving on metronidazole or tylosin for 2 weeks were previously classified as antibiotic-responsive enteropathy, but antimicrobial treatment has a long-lasting effect on the intestinal microbiota, antimicrobial resistance is a major concern, and most dogs that initially respond to an antibiotic trial relapse after discontinuing treatment. The existence of the ARE subgroup, proposed to be renamed idiopathic intestinal dysbiosis, and the appropriateness of empirical antimicrobial trials in the work-up of chronic inflammatory enteropathy are currently debated. In protein-losing enteropathy the position is firmer: absent an infectious aetiology, antimicrobial treatment plays no role. A sequenced diet trial is the better first move.

What should I use instead of metronidazole for uncomplicated acute diarrhoea? Diet and supportive care. In dogs with noninfectious acute colitis, an easily digestible diet plus placebo (median remission 5 days, range 4 to 10) and a psyllium-enhanced version (median 5 days, range 3 to 6) both beat the same diet plus metronidazole (median 8.5 days, range 7 to 12). ENOVAT recommends against antimicrobials in mild and moderate disease on strong, high-certainty evidence, reserving them for dogs with systemic signs persisting despite adequate fluid therapy, or with severe neutrophilia (> 25x10^9/L), neutropenia or a degenerative left shift. Probiotics are equivocal: ENOVAT recommends neither for nor against them. Address hydration and electrolytes, and use antiemetics for symptomatic control.