Antimicrobial stewardship in small animal practice
Three questions to ask before reaching for the prescription pad, how to rank empirical choices, and the four situations where antibiotics are most often given without need.
Admin 5 min readAugust 21, 2026
Antimicrobial resistance is no longer a distant concern; it sits inside daily practice. Multidrug resistance in Escherichia coli and Staphylococcus pseudintermedius isolated from companion animals has climbed steadily across many countries. Methicillin-resistant S. pseudintermedius (MRSP) was an unusual laboratory finding fifteen years ago. Today it is an ordinary explanation for a pyoderma that keeps coming back.
This article sets out a practical order of reasoning for the moment you decide whether to prescribe. The aim is not to avoid antimicrobials. It is to give the right drug, to the right patient, for the right length of time.
Three questions before prescribing
1. Is this actually a bacterial infection?
Fever, leukocytosis and malaise are evidence of inflammation, not of infection. Immune-mediated disease, neoplasia, pancreatitis and sterile inflammation all produce the same picture.
The most frequently confused example is lower urinary tract signs in young cats. In a three-year-old cat with dysuria, pollakiuria and haematuria, the most likely diagnosis is feline idiopathic cystitis; bacterial infection accounts for only a small fraction of cases in this age group. The proportion rises substantially in cats over ten, and in those with chronic kidney disease or diabetes mellitus. Age and comorbidity alone should change the prescribing decision here.
2. Do I need a sample?
Culture and susceptibility testing are unambiguously indicated when:
The infection is recurrent or has failed to respond
The patient has had antimicrobials recently
The site is deep tissue, bone, joint, or the upper urinary tract
The patient is immunocompromised
The patient is critically ill — start empirically, but take the sample before the first dose
Sampling after treatment has begun usually renders the culture uninterpretable. In the critical patient, getting the order right is the only way to preserve the information without delaying treatment.
3. Is a systemic drug required, or will topical treatment do?
Get an email when a new clinical article is published in the knowledge base. No membership needed.
We will send a confirmation link; nothing is sent until you confirm. Every email carries a one-click unsubscribe link.
In superficial pyoderma, otitis externa and infected superficial wounds, topical therapy alone is sufficient in most cases and confines selection pressure to a far narrower field than systemic treatment. Chlorhexidine-based shampoos and solutions perform comparably to systemic antimicrobials in superficial pyoderma.
Ranking the empirical choice
The European Medicines Agency's Antimicrobial Advice Ad Hoc Expert Group (AMEG) sorts veterinary antimicrobials into four categories according to the risk their use poses to human health. The working rule that follows from it:
Start with narrow-spectrum, older drugs. Amoxicillin, amoxicillin-clavulanate, trimethoprim-sulphonamide, doxycycline, first-generation cephalosporins. This group covers the large majority of infections seen in small animal practice.
Fluoroquinolones and third- or fourth-generation cephalosporins are not first-line agents. These classes are critically important in human medicine, and routine first-line use is both unnecessary and inadvisable. Their place is in cases where culture leaves no alternative, or where no other drug reaches the site of infection.
Last-resort human drugs — carbapenems, vancomycin, linezolid — have no routine role in veterinary practice. Any use should be exceptional, justified, and ideally taken in consultation.
For starting doses, work from the licensed product label in your own country together with a current formulary. Nothing in this article substitutes for the product label.
Dose and duration
Two common errors point in opposite directions, and both select for resistance:
Underdosing. The instinct to "go easy on a small patient" produces subtherapeutic concentrations — precisely the conditions that select resistant organisms. Dose is calculated per kilogram and given in full.
Treating for too long. The habitual ten- or fourteen-day course rests more on tradition than on evidence. The International Society for Companion Animal Infectious Diseases (ISCAID) considers three to five days adequate for sporadic bacterial cystitis, in place of the fourteen days taken as standard for years.
Shorter courses do not generalise to every infection. Deep pyoderma, osteomyelitis, prostatitis and discospondylitis require weeks of treatment. The rule is that duration follows clinical response and the site of infection, not habit.
Four situations where antibiotics are most often unnecessary
Acute uncomplicated diarrhoea in dogs. In an afebrile dog that is systemically well, acute diarrhoea is usually self-limiting. Antimicrobials do not shorten recovery, and they disrupt the intestinal microbiota for weeks. Fluid and electrolyte support with dietary management is the actual treatment. Blood in the stool is not on its own an indication; signs of sepsis change the picture.
Lower urinary tract disease in young cats. As above: idiopathic cystitis does not respond to antimicrobials, it resolves on its own, and that resolution is then credited to the drug. It is a textbook case of learning the wrong lesson.
Prophylaxis after clean surgery. Clean, short procedures such as ovariohysterectomy and castration do not call for a postoperative course. Where prophylaxis is indicated, a single perioperative dose is enough.
Feline upper respiratory tract infection. The aetiology is largely viral — herpesvirus and calicivirus. Antimicrobials become meaningful only when secondary bacterial infection declares itself through mucopurulent discharge, persistent fever or inappetence.
What a practice can do
Individual judgement left to good intentions drifts; a written system reduces the drift.
A practice protocol. For the ten infections you see most, write down the first choice, the alternative and the duration. Keep it to one page.
A culture record. Review the resistance profile of isolates from your own practice once a year. Your own patient population's resistance map is more useful than any international average.
Talking to the owner. "We are not giving antibiotics" sounds like neglect when the reason goes unexplained. Saying why, and naming the signs that should bring them back, protects both the treatment and the relationship.
Reassessment. If there is no response at forty-eight to seventy-two hours, question the diagnosis before escalating the drug. The most common reason for failure is not the wrong antimicrobial but the wrong diagnosis.
Sources
Weese JS et al. ISCAID consensus guidelines on urinary tract infections in dogs and cats, The Veterinary Journal, 2019.
European Medicines Agency, AMEG categorisation of antibiotics (Categories A–D), 2020.
WSAVA and WHO antimicrobial stewardship documents.
This article is for general education and does not replace clinical examination, the current product label, or local regulation in the management of an individual patient.