Antibiotics are not the most-reported drug class in the FDA's veterinary adverse-event database — that distinction belongs to the parasiticides, led by spinosad and milbemycin. But antibiotics and anti-infectives are the most diverse contributor to veterinary pharmacovigilance, spanning nine distinct mechanistic classes and showing class-specific safety signals that a generic "is this drug safe?" framing will always miss.

An analysis of the FDA Center for Veterinary Medicine (CVM) animal drug adverse-event records, accessed through the openFDA database (analysis run date: 2026-06-13), finds 103,656 reports involving an antibiotic or anti-infective out of 1,340,077 total reports covering 1987 through early 2026 — 7.7% of all veterinary drug adverse-event reports on file. This article breaks those reports down by drug class, by individual molecule, by reaction type, and by species, and connects the patterns to what is known about each class from labels and the peer-reviewed literature.

Every number below is computed directly from the FDA CVM openFDA animal adverse-event records. None of it is estimated.

Three caveats before the numbers

Report volume is a pharmacovigilance signal, not a risk score. Three things must be kept in mind when reading any of the figures that follow:

1. Sick-animal bias is large for antibiotics. Unlike a monthly parasiticide given to a healthy dog, an antibiotic is usually given to an animal that is already ill — sometimes critically. Many of the deaths and serious outcomes in these reports reflect the underlying infection, not the drug. This is the single biggest confounder in any antibiotic adverse-event analysis.
2. Reports are suspicions, not confirmed causality. The FDA states explicitly that reports in this database have not been verified for causality. A report listing enrofloxacin and "vomiting" does not mean enrofloxacin caused the vomiting.
3. Under-reporting is substantial. Only a fraction of adverse events are ever reported, and the fraction varies by product, reporter, and public awareness. Counts tell you where monitoring attention concentrates, not the true event rate.

With those caveats, the class-level patterns are real and useful — and several of them align with known class toxicities that prescribers already weigh.

How often are antibiotic reports flagged serious?

Across the entire openFDA animal database, 25.0% of reports carry a serious-adverse-event flag (death, life-threatening event, or an outcome requiring intervention). Among reports involving an antibiotic or anti-infective, that figure rises to 28.0% — modestly higher, and almost certainly driven by the sick-animal bias above rather than by antibiotics being inherently more dangerous than other veterinary drugs.

The more interesting finding is how much the serious-flag rate varies within the antibiotic class:

The nitroimidazole row is the standout. Metronidazole is one of the most commonly prescribed veterinary antibiotics, and its neurotoxicity at high cumulative doses or long courses is well described — ataxia, nystagmus, head tilt, and seizures that typically resolve after withdrawal. A 50% serious-flag rate among metronidazole reports is consistent with that toxicity showing up as the reason the report was filed in the first place. The relatively low serious rate for sulfonamides (16.0%) at the other end partly reflects how sulfadimethoxine/ormetoprim combinations are used — often as a first-line, outpatient option in less critical patients.

These rates are descriptive, not comparative risk. A class with a higher serious rate is not necessarily more dangerous per prescription; it may simply be used in sicker patients, or have a toxicity dramatic enough that someone files a report.

The drugs behind the reports

Combining salt and synonym variants of the same molecule, the most-reported individual antibiotics are:

Amoxicillin/clavulanate topping the list is a market-penetration story: it is the most widely dispensed companion-animal oral antibiotic, so it accumulates reports simply by volume. The presence of two long-acting injectables high on the list — cefovecin (Convenia) and florfenicol — reflects their use in both companion and food-animal practice, and the fact that a single injection generates a durable, reportable exposure window. Tilmicosin (Micotil) and tulathromycin (Draxxin) are cattle respiratory macrolides; tilmicosin in particular carries a well-known cardiovascular warning (potentially fatal in humans after accidental injection), which contributes to its report profile.

The reaction profile: gastrointestinal, then efficacy failure

Among antibiotic-involved reports, the most common reactions are:

Two patterns deserve attention beyond the expected gastrointestinal cluster (vomiting, anorexia, diarrhea).

Efficacy failure is the second-largest signal after GI signs. "Lack of efficacy" reactions — generic and bacterial-specific — together exceed 11,800 reports. Some of this reflects animals that were severely ill at the start of treatment, but a portion is the pharmacovigilance fingerprint of antimicrobial resistance, undertreatment, or wrong-drug-for-the-bug selection. This is the same signal the FDA is trying to shrink on the food-animal side: in February 2026, CVM finalized Guidance for Industry #273, which asks sponsors to define and justify maximum durations of use for medically important in-feed antimicrobials, an explicit push against prolonged, resistance-selecting exposure.

Neurologic and hepatic reactions track the known class toxicities. Ataxia (3,718 reports) is consistent with the neurologic effects documented for fluoroquinolones and for metronidazole. Elevated ALT (2,637 reports) reflects the hepatic monitoring recommended during fluoroquinolone and certain macrolide courses.

The fluoroquinolone signal in cats, in context

Fluoroquinolones are the class where the openFDA data and the published literature align most tightly. Enrofloxacin (Baytril) can cause acute, sometimes permanent retinal degeneration and blindness in cats. The mechanism is understood: cats carry a variant of the ABCG2 drug transporter at the blood-retinal barrier that allows fluoroquinolones to accumulate in the retina. The original enrofloxacin label allowed up to 20 mg/kg/day; after a cluster of blindness reports in the late 1990s and early 2000s, the label was revised down to a 5 mg/kg/day ceiling. Most affected cats in the published case series received doses above that ceiling, but blindness has been documented at 4.6 mg/kg, and older cats and those with renal impairment are at higher risk. Marbofloxacin (Zeniquin) is generally considered lower-risk for retinal effects in cats.

The fluoroquinolone row in the class table — 13,632 reports, 32.8% flagged serious — is the kind of number that should reinforce, not replace, the stewardship principle: reserve fluoroquinolones for infections where culture and susceptibility show that first-line options will not work, stay at or below label dose, and warn owners to watch for mydriasis as an early ocular sign.

Species — and the people who handle these drugs

The species breakdown among antibiotic reports is:

Dogs dominate, as they do across the entire companion-animal pharmacovigilance database. The cattle and pig figures reflect food-animal macrolide and phenicol use. The row that practitioners and animal-health staff most often overlook is the fourth one: 4,638 reports list a human as the affected species, and "accidental exposure" appears as a reaction in 4,505 antibiotic reports. These are needlesticks, splashes, oral exposures in children, and handler exposures — the occupational-health dimension of veterinary antibiotics. Tilmicosin's cardiovascular warning to humans after accidental injection is the most-cited example, but the data shows the exposure problem is broader than any single product. This is the One Health layer of antimicrobial safety that lives in the same database as the animal events.

What this means for prescribing and monitoring

The practical takeaways from the data are stewardship-shaped:

• Match the drug to the culture, not the habit. The efficacy-failure signal is large enough that empiric, un-titrated courses are a measurable contributor to adverse-event reporting. Culture and susceptibility before escalating to a fluoroquinolone or a higher-tier macrolide.
• Respect the class-specific ceilings. Stay at or below 5 mg/kg/day of enrofloxacin in cats and watch for mydriasis. Be cautious with metronidazole cumulative dose and duration, and recognize ataxia or nystagmus as a withdrawal trigger. Keep aminoglycosides away from patients with renal compromise and stay hydrated.
• Counsel owners on the exposure risk in the home. A child or another pet accessing an oral antibiotic, or a handler receiving a needlestick, is a real event category — store and dispose accordingly, and for injectable food-animal products, follow the handler-safety directions to the letter.
• Report. The single biggest limit on this database is under-reporting. Filing a report when a patient has a suspected antibiotic reaction — through the FDA CVM adverse-event reporting channel — is what makes future patterns detectable.

What this data cannot do

These 103,656 reports are the visible portion of a much larger, unrecorded denominator. They do not rank antibiotics by safety, they do not prove causality, and they do not tell you the per-prescription risk of any individual drug. What they do is show where veterinary pharmacovigilance attention concentrates within the anti-infective class — and confirm that the class-level patterns in the data line up with the toxicities that good prescribing already accounts for.

Sources

• FDA Center for Veterinary Medicine, openFDA Animal & Veterinary Adverse Event API (database of 1,340,077 reports, 1987–2026), https://open.fda.gov/apis/animalandveterinary/
• FDA CVM, Recent Animal Drug Approvals and CVM Updates, https://www.fda.gov/animal-veterinary/news-events/cvm-updates
• FDA, Guidance for Industry #273 — Defining Durations of Use for Approved Medically Important Antimicrobial Drugs Fed to Food-Producing Animals (finalized February 2026), https://www.fda.gov/regulatory-information/search-fda-guidance-documents
• Baytril (enrofloxacin) Freedom of Information Summary, NADA 140-441 (label revision for feline retinal toxicity), https://animaldrugsatfda.fda.gov/
• Gelatt KN et al., "Enrofloxacin-associated retinal degeneration in cats," Veterinary Ophthalmology (2001)
• Wiebe V, Hamilton P, "Fluoroquinolone-induced retinal degeneration in cats," JAVMA (2002), https://www.ncbi.nlm.nih.gov/
• Enrofloxacin-associated ocular disease in cats: a scoping review, PMC, https://pmc.ncbi.nlm.nih.gov/articles/PMC12825762/
• Clinician's Brief, "Top 5 Severe Adverse Effects of Antimicrobials," https://www.cliniciansbrief.com/article/antimicrobials-severe-side-effects-cats-dogs