Responsibility

An equine practitioner accepting responsibility for the antimicrobial prescription needs an accurate diagnosis, a risk assessment including adverse consequences of antimicrobials, as well as prudent‐use practices that minimize the selection and spread of AMR.

Overreliance on combination antimicrobials represents more a lack of certainty of the diagnosis. Thus, the prescribing veterinarian should accept responsibility for an accurate diagnosis that includes precise standards for a diagnosis of infection. For example, an infection diagnosis based solely on a fever, or reduced feed intake, elevated white blood cell count or other nonspecific blood biomarkers, and lack of culture and susceptibility is insufficient for an antimicrobial prescription.

An unfortunate trend in equine medicine is the widespread use of the surrogate acute‐phase protein biomarker serum amyloid A (SAA) to indicate an infection somewhere in the horse’s body. Acute‐phase proteins, like SAA, are secreted during the acute phase of inflammation, but respond rapidly and nonspecifically to relatively trivial inflammatory stimuli (e.g., vaccination) or noninfectious diseases (e.g., hyperlipemia). However, SAA is promoted as a specific marker of infection (Jacobsen and Andersen, 2007), whereby horses are treated with antimicrobials, based on SAA levels, without culture and susceptibility results, until SAA returns to normal. There are neither robust studies that have shown that SAA can differentiate between inflammatory and infectious conditions in horses, nor any credible scientific basis that SAA can reliably indicate an infectious process (Isgren, 2021).

Treatment based on SAA levels is both extra‐label use as well as a misuse of antimicrobials. For example, Hoeberg et al. (2022) investigated the utility of SAA as a biomarker to detect sepsis and predict outcome in 590 hospitalized foals. They found a poor sensitivity (30.2%) (measure of SAA to identify true positives) for an optimal cut‐off for sepsis, but high specificity (90.7%) (measure of SAA to identify true negatives), an even worse sensitivity (22.1%) for an optimal cut‐off for nonsurvival prediction, but similar specificity (90.8%). Noting the major flaws of this analysis in that sensitivity/specificity were not adjusted for the prevalence of septicemia (both general and hospital‐based populations), sensitivity/specificity assumes a gold standard for diagnosis (not known for antemortem septicemia), sensitivity/specificity are only for population dynamics and not individual test characteristics, then positive/negative likelihood ratios (LR) are preferred for individual patient accuracy and independent of disease prevalence. Recalculation of the results of Hoeberg et al. (2022) reveals that SAA as a marker of foal septicemia has a barely moderate positive LR of 3.2 and a very poor negative LR of 0.8, and is even worse for nonsurvival prediction with a positive LR of 2.4 and negative LR of 0.9. To provide context, a positive LR ≥10 and negative LR ≤0.1 provide strong evidence to rule in or rule out diagnoses, respectively, in most circumstances.

Adverse effects can occur from antimicrobial therapy, ranging from hypersensitivity reactions (e.g., urticaria), immune‐mediated disease (e.g., immune‐mediated hemolytic anemia/thrombocytopenia), arthropathy, nephrotoxicity, ototoxicity, dysrhythmia, neurological factors, anorexia, and antimicrobial‐associated colitis (AAC) (Khusro et al., 2020). AAC develops in temporal association with antimicrobial therapy, leading to dysbiosis or proliferation and toxin production from Clostridioides difficile. Occurrence of AAC is difficult to predict, but there is some risk with any antimicrobial administered via any route. Drugs that have low oral bioavailability or are excreted in bile or through enterocytes pose a higher risk because of the drug concentrations achieved within the gastrointestinal tract. Horses with AAC have presumably greater disturbances of the bacterial communities of the hindgut and subsequent metabolic changes that result in more severe clinical symptoms (Arnold et al., 2021). The likelihood and consequences of AAC should be carefully considered when deciding whether antimicrobials are necessary. Although many horses with AAC are treated with metronidazole, resistance builds quickly, with up to 43% of metronidazole‐resistant C. difficile isolates from horses being reported (Magdesian et al., 2002), and is likely increasing.

Antimicrobial‐resistant bacterial infections are a reality of modern equine practice, with major impacts on treatment failures, acquisition of nosocomial hospital infections and zoonosis, and increased use of critically important antimicrobials. Most horses have a close association with their owners, which makes them part of a One Health perspective (Lönker et al., 2020), potentially spreading zoonotic pathogens as well as transferring multiresistance pathogens. This includes methicillin‐resistant Staphylococcus aureus (Baptiste et al., 2005), Salmonella spp., or Enterobacterales with extended spectrum beta‐lactamases (ESBL), Enterobacter cloacae, Escherichia coli, Klebsiella oxytoca, and Citrobacter freundii. Furthermore, equine bacteria carrying mobile resistance genes to other antimicrobials, with both ESBL and mcr‐9 (colistin resistance), have all been reported in horses (Börjesson et al., 2020).

Several of these pathogens are also “nosocomial” pathogens associated with many equine hospitals. AMR among nosocomial equine hospital pathogens are also “sentinels,” reflecting hospital biosecurity/disinfection practices and antimicrobial policies, and increase the risk of postoperative infections. For example, a reckless hospital management policy of using a fourth‐generation cephalosporin (cefquinome) for equine surgical prophylaxis resulted in ESBL E. coli postoperative infections with increased morbidity and mortality, compared to previous hospital management without cephalosporin prophylaxis (Damborg et al., 2012). These pathogens convey a poor prognosis for infected foals and adult horses, as well as posing risks for veterinarians, nurses, and other employees working at the clinics (Maddox et al., 2012). Some infection control programs exist in referral equine clinics, including cause‐of‐effect investigations of (nosocomial) infections (Rohweder et al., 2021). This is important especially for some bacterial species (Gram‐negatives: Pseudomonas spp., Acinetobacter spp., E. coli; Gram‐positives: Enterococcus spp., Staphylococcus spp.) which, due to their high intrinsic resistance, have limited treatment options for infected horses (Isgren, 2021).

Reduction

Empirical antimicrobial therapy is still routinely employed in horses, although it is generally recognized that submission of culture and antimicrobial susceptibility testing are important parts of good antimicrobial stewardship. Reasons include potential seriousness of the disease (septic arthritis), difficulty in collecting diagnostic samples, and economic constraints. As empirical antimicrobial therapy is common, it represents an issue of reflection for reduction in usage. For example, conditions that are potentially debilitating/life‐threatening infections are targeted for empirical therapy (e.g., severe pneumonia, pleuropneumonia, peritonitis, septic arthritis, neonatal septicemia), while awaiting culture results. There are also disease conditions in which empirical antimicrobial therapy is not recommended, such as upper respiratory tract infection (i.e., likely viral in origin), diarrhea in adult horses, and superficial wounds (not involving a joint or tendon sheath). Supportive care and close monitoring should suffice for these.

Conventional culture and susceptibility has limitations in terms of speed (taking at least three days) and accuracy. For example, false‐negative results can occur due to intermittent shedding of the pathogen, presence of fastidious or unculturable organisms, failure to use specialized microbiological media, improper sample collection, improper sample storage or shipping, or prior antimicrobial therapy.

Recently, more advanced, rapid methods have become available for horses. Single‐plex and multiplex quantitative PCR (qPCR) assays useful for R. equi and rapid diagnostic testing with detection of MLS_B resistance (via specific erm genes detection) have been designed and validated (Narváez et al., 2022). Another example is loop‐mediated isothermal amplification (LAMP) assay and PCR, followed by high‐resolution melt (HRM) curve analysis (Garner et al., 2023) for mares suspected of endometritis from S. zooepidemicus. These methods allow short turn‐around times and a high level of accuracy for detection of S. zooepidemicus (six hours for PCR‐HRM, two hours for LAMP, with slightly lower specificity equal to 89.5%).

Antimicrobial surgical prophylaxis constitutes a major use and could be reduced. There are two components for perioperative surgical prophylaxis: preoperative and postoperative. The evidence base is weak for perioperative prophylaxis due to a lack of large‐scale, randomized clinical trials. Nevertheless, recommendations are to restrict antimicrobial use for surgical procedures to where the incidence of infection exceeds 5% without prophylactic antimicrobial use. For clean surgical procedures, antimicrobial prophylaxis for horses remains controversial (see Chapter 24, Table 24.2). Documented heavy use of perioperative prophylactic antimicrobials occurs despite the fact that the incidence of postoperative infections is very low (0–0.9%) for common elective surgeries (e.g., carpal arthroscopy) (Ridge, 2011; Weese and Cruz, 2009). Penicillin and gentamicin in combination remains the most frequently used prophylactic measure for colic surgeries (Southwood, 2014; Stöckle et al., 2021). Unfortunately, equine surgical antimicrobial prophylaxis has also included critically important antimicrobials such as cefquinome, ceftiofur, and enrofloxacin (Ceriotti et al., 2021).

Redpath et al. (2021) reported an electronic survey on the common use of gentamicin empirically without antimicrobial susceptibility testing (across the continents, as respondents were veterinary surgeons from Europe, North America, and Australia/Asia). Gentamicin use only after susceptibility testing was more frequently reported for generalists than specialists, for a range of indications (respiratory diseases, septic peritonitis, acute febrile diarrhea, cellulitis, and contaminated limb wounds). Gentamicin was more often used prophylactically for high‐risk procedures or contaminated surgeries (86% and 74%, respectively) compared with clean surgery (32%). Gentamicin was often used perioperatively in horses undergoing exploratory celiotomy and more commonly used in horses undergoing an enterotomy (90%) than those which did not undergo an enterotomy (79%).

Stöckle (2019) retrospectively assessed datasets of 652 patients, of which 259 (39.7%) received antimicrobial perioperative prophylaxis, compared to 393 (60.3%) horses which did not. Antimicrobial administration started prior to surgery and continued for 3–5 days. The majority of surgeries were arthroscopy; others included fasciotomy, splint bone extraction, tenotomy, neurectomy, tendovaginoscopy/bursoscopy, or combined surgical interventions. There were no significant differences for postoperative complications between the 97/259 (37.5%) antimicrobial‐receiving patients and the 158/393 (40.2%) controls. This is reasonable evidence that perioperative antimicrobial prophylaxis in uncomplicated, elective orthopedic surgical procedures in equines is not justified. For other elective procedures (e.g., castration), delaying surgery is acceptable to ensure the patient is fully healthy with appropriate updated preventive measures (e.g., tetanus toxoid prophylaxis). It is important to emphasize that antimicrobials should not be used in place of good surgical technique, a proper surgical environment, good management, and optimal infection control practices.

In another larger American survey of 761 hospitalized horses, a total of 511 (67.2%) received an inappropriate amount of antimicrobials preoperatively (Dallap Schaer et al., 2012). The majority of these horses underwent colic surgery. Underdosing was the most common inaccuracy observed. In addition to this, timing of antimicrobial administration was inadequate (e.g., ≥1 hour before surgery), with only 88 (11.6%) of horses receiving the antimicrobial at the appropriate time. Antimicrobial therapy was continued for an average of 3.8 days. Out of the 761 horses followed, 680 received the combination of penicillin and gentamicin (89.3%), 16 received ceftiofur and gentamicin (2.1%), and only 22 horses (2.9%) received a single antimicrobial.

Prophylaxis recommendations for equine surgeries are covered under the general principles discussed in Chapter 24.

Refinement: Right Drug, at the Right Time, at the Right Dose, and Right Duration

The need for refinement in AMS is to reduce, prevent, and slow the development of AMR. Common clinical scenarios that contribute to AMR include underdosing of antimicrobials and prolonged duration of treatment, treatment failures due to the pathology, and antimicrobial agents used for the treatment of nonbacterial conditions. Published studies following equine practitioners have found a proportion of practitioners choosing critically important antimicrobials as first‐line therapy (e.g., third/fourth‐generation cephalosporins), frequently not adhering to dosing recommendations, and failing to utilize antimicrobial susceptibility testing in a large percentage (49.2%) of cases (Ryan et al., 2023). These are clear targets for refinement to improve AMU in daily equine practice.

Due to the practicalities and challenges of handling and medicating horses, there is a strong bias towards oral antimicrobials in horses. Examples include trimethoprim‐sulfonamide as well as extra‐label use of doxycycline and enrofloxacin. Limitations of oral antimicrobials in horses include being linked with AAC as well as oral bioavailability affected by feed intake whereby it is difficult to maintain clinically relevant lung concentrations, for both trimethoprim‐sulfonamide (Winther et al., 2011a), and doxycycline (Winther et al., 2011b). Thus, a major challenge of refinement is to move away from oral antimicrobials as first choice, with more consideration given to correct antimicrobials given via the correct route of administration, that will achieve concentrations above the bacterial MICs for the given infection.

Third/fourth‐generation cephalosporins are commonly used and misused in horses because of their spectrum of activity against many Gram‐negative and Gram‐positive bacteria as well as their favorable systemic safety profile. As a result, AMR to these cephalosporins is common and increasing, especially when used for prophylaxis. In several countries, these cephalosporins are marketed only for equine streptococcal infections. Although occasional reports suggest low‐level penicillin resistance in equine streptococci (Fonseca et al., 2020), none have proven the phenomenon with any molecular studies. Therefore, third/fourth‐generation cephalosporins are not warranted for treatment of equine streptococcal infections, since penicillin susceptibility of equine streptococci is highly prevalent.

Furthermore, ceftiofur is a poorly understood cephalosporin. In vivo, ceftiofur rapidly hydrolyzes to its primary metabolite, desfuroylceftiofur, which remains as the primary molecule with antimicrobial activity. However, antimicrobial susceptibility is based on ceftiofur, since desfuroylceftiofur is unstable in vitro. Little is known about bacterial MICs against desfuroylceftiofur, but they may be very different for staphylococci compared to ceftiofur MICs. Also, desfuroylceftiofur is very highly (>90%) protein bound, which affects clinically relevant concentrations because only free (i.e., unbound) drug exerts antimicrobial effects. This also limits the total body distribution of desfuroylceftiofur, as the protein‐bound concentration cannot penetrate the blood–brain barrier.

Thus, third/fourth‐generation cephalosporins should not be used for first‐line antimicrobial therapy but reserved for mixed infections based on culture and susceptibility results.

Replacement

Unfortunately, there are still examples in equine medicine of antimicrobials endorsed for nonantimicrobial purposes. This practice is an obvious target for replacement with suitable nonantimicrobial treatments. Examples include equine asthma, uncomplicated wounds, and viral respiratory disease. Reasons for this are multifaceted, but it is important for practitioners to step back from regular routines and habits, and reflect more on clinical cases. The correct diagnosis is more important than a quick decision. Pride in taking the time to do a professional assessment, including taking and waiting for diagnostic sample results, is important to communicate to owners.

It has become common practice to inject neonatal foals born with contracted tendons with 1–3 high doses of oxytetracycline (40–70 mg/kg; 3 grams/foal of 50 kg). The use of oxytetracycline for this purpose is due to a unique side‐effect that causes temporary tendon relaxation in foals, possibly related to calcium chelation. With tendon relaxation, the foal becomes more ambulatory which eventually resolves the condition. While it is quicker and easier to treat all cases of foals with contracted tendons with oxytetracycline, compared to splints, oxytetracycline doses of 40–70 mg/kg in foals are well into the known toxic range (hepato‐ and nephrotoxicity) for a nonantibacterial indication. Thus, this treatment option should only be used for severe contractures, utilizing as few doses as possible. The principle of treatment for contracted tendons is to keep the foal’s leg straight enough for it to walk on its toe. The act of walking and then stretching will lead to correction of the issue. For mild to moderate contracted tendons, all efforts should be put into specially padded wrap‐around splints, bandages or casts as valid alternatives which may be applied to hold the fetlock, pastern, and toe straight and forward. Analgesics are also an important component of the treatment plan.

Polymyxins are critically important in human medicine as last‐resort antimicrobials for multiresistant Gram‐negative bacteria. However, polymyxin B is also used in equine medicine for a nonantibacterial purpose at subtherapeutic doses for the treatment of systemic inflammatory response syndrome (SIRS) due to endotoxemia in horses and foals (Morresey and MacKay, 2006). SIRS associated with endotoxemia remains a leading cause of morbidity and mortality in both neonatal and adult equids (Mercer et al., 2023). Polymyxin B binds to bacterial toxins and therefore reduces activation of the proinflammatory cascade. Horses and humans share a pronounced sensitivity to the effects of endotoxemia. Endotoxins (free‐floating) are produced within the equine gastrointestinal tract and absorbed systemically, secondary to a gastrointestinal disease or due to a Gram‐negative bacterial infection.

Polymyxin B dose for SIRS in horses varies between 5000 and 10,000 IU/kg IV every 8–12 hours (Barton et al., 2004; Morresey and MacKay, 2006; Werners, 2017). Doses up to 25,000 IU/kg/day are needed to reach true antimicrobial effects in humans, compared to much lower doses for antiendotoxic effects in horses. There is limited evidence that prophylaxis with polymyxin B improves survival associated with SIRS (Barton et al., 2004). However, polymyxin B is typically given to horses and foals when there is already full‐blown SIRS. Once SIRS is triggered by endotoxemia, with associated clinical signs, polymyxin B is unlikely to be clinically effective since it has no direct effect on SIRS.

Administering polymyxin B intravenously to horses results in serious adverse effects (Schwarz et al., 2013; van Spijk et al., 2022). A study showed neurotoxicosis in healthy horses receiving seven doses of polymyxin B 6000 IU/kg IV q12h. Mild‐to‐moderate ataxia was reversible after cessation of therapy. Number of polymyxin B doses and co‐administration of gentamicin increased the severity of ataxia and nephrotoxicity (van Spijk et al., 2022). Conversely, polymyxin B is not used in human medicine for endotoxemia due to serious adverse effects. Use of polymyxin E (colistin) as a growth promoter in food animals has resulted in global transmissible colistin resistance in animal and human populations. Plasmid‐mediated colistin‐resistant genes have been identified in horses (Börjesson et al., 2020). With widespread global colistin resistance and the critical importance of polymyxins for human medicine, equine clinicians must replace polymyxin use as much as possible for the sake of One Health (Isgren, 2021). Alternatives to low‐dose polymyxin B for SIRS include hydration fluids, NSAIDs, or intravenous lidocaine (Peiró et al., 2010).

The most prominent proinflammatory cytokines in equine SIRS, linked to clinical signs, are TNF‐alpha, IL‐1beta, and IL‐6, all of which are also regulators of innate immunity (Tadros and Frank, 2012). Since prostaglandins affect the formation of proinflammatory cytokines, COX inhibitors (i.e., nonsteroidal antiinflammatory drugs – NSAIDs) remain central for the treatment of SIRS associated with endotoxemia (Mercer et al., 2023). Furthermore, a multicenter, blinded, randomized clinical trial comparing the use of flunixin meglumine or firocoxib in horses with small intestinal strangulating obstruction (Ziegler et al., 2019) found that firocoxib significantly reduced a major biomarker of endotoxaemia more compared with flunixin meglumine, while continuing to provide similar levels of pain control.

A third nonbacterial use is doxycycline as a treatment for osteoarthritis in horses (Maher et al., 2014). The authors report that low‐dose, low‐frequency oral administration of doxycycline attains in vivo synovial fluid concentrations capable of chondroprotective effects through reduction of matrix metalloproteinase (MMP)‐13 activity, while remaining below the MIC₉₀ of most equine pathogens. Given that several novel therapies are available for equine osteoarthritis (e.g., stem cells, polyacrylamide gel), it is not justified to give subtherapeutic doxycycline to horses and it should be replaced with alternative options.

Review

Respiratory Infections

Respiratory tract diseases are common in horses, and one of the most frequent reasons for antimicrobial administration (Table 27.1). Clinical signs are often indistinguishable for infectious versus noninfectious causes. Many upper respiratory diseases in horses are noninfectious or viral and do not require antimicrobial treatment. Furthermore, bacterial pneumonia secondary to viral infection is rare. Broad‐spectrum antimicrobial use for respiratory tract diseases based on clinical signs alone (e.g., fever, cough, nasal discharge) can no longer be justified. There are a number of diagnostic techniques for lower airway diseases (e.g., transtracheal wash, tracheal aspirate, bronchoalveolar lavage, lung biopsy) that are simple and safe to perform whereby representative samples can be obtained for culture and susceptibility testing. Adequate stall rest for three weeks in a well‐ventilated stable and supportive care, including good‐quality hay and water, are important adjunctive components of therapy.

Inhalation therapy for direct delivery of antimicrobials to the lower airways is an enticing method to deliver maximal drug concentrations to the site of infection and achieve a rapid onset of action, while minimizing systemic exposure. Aerosol particle sizes between 1 and 5 μm are thought to be ideal for therapy, using ultrasonic or jet nebulizers. Aminoglycosides are the most commonly reported aerosolized antimicrobial agents because they remain bioactive when aerosolized and are poorly absorbed across epithelial surfaces, thus remaining within the pulmonary tree where they exert concentration‐dependent effects. However, inhalation antimicrobial therapy has remained controversial in human medicine because of the potential risk of pulmonary contamination with environmental bacteria and poor drug delivery to consolidated lung regions. Irritation from the drug may induce bronchoconstriction and aerosol administration on airway surfaces containing large numbers of diverse bacteria (especially biofilm producers) may select for antimicrobial‐resistant bacteria. For most nebulizer systems, it is estimated that only 10% of the drug reaches the lungs during disease due to excessive mucus secretions, bronchospasm, and higher and more turbulent air flow rates with tachypnea.

Only gold members can continue reading. Log In or Register to continue

Share this:

• X
• Facebook

Related

Related posts:

1. Principles of Antimicrobial Drug Selection and Use
2. Antimicrobial Therapy in Dogs and Cats
3. Antimicrobial Resistance and Its Epidemiology
4. Antimicrobial Therapy in Dairy Cattle

Stay updated, free articles. Join our Telegram channel

Join

Tags: Antimicrobial Therapy in Veterinary Medicine

Mar 15, 2026 | Posted by admin in GENERAL | Comments Off on Antimicrobial Drug Use in Horses

Full access? Get Clinical Tree

Get Clinical Tree app for offline access

Get Clinical Tree app for offline access

Site	Diagnosis	Common infecting organism(s)	Comments	Suggested drug(s)	Alternative drug(s)
Upper respiratory tract	Strangles	Streptococcus equi	Classic strangles (fever, depression, swollen mandibular lymph nodes, nasal discharge) does not require antimicrobial treatment. While S. equi is universally susceptible to penicillin, S. equi is protected within an already formed abscess.	None	Penicillin Ga
	Guttural pouch empyema	Streptococcus equi, S. zooepidemicus, rarely Gram‐negatives	Local irrigation with saline is the treatment of choice. Lowering the horse’s head facilitates drainage and reduces the risks of aspiration. Systemic or topical antimicrobials are rarely indicated. Important to culture to identify S. equi carriers, that may spread infection to other horses	None	Penicillin Ga
	Fungal rhinitis	Aspergillus spp. Other opportunistic fungi	Surgical removal of the mycotic plaque and associated necrotic tissue, combined with topical antifungal therapy	Topical miconazole	Topical natamycin;
	Sinusitis, primary	S. zooepidemicus	Antimicrobials are not necessary with adequate removal and flushing of purulent debris. Nonresponsive cases may require sinusotomy	Penicillin Ga	Trimethoprim‐sulfonamideb
	Sinusitis, secondary	Mixed opportunistic aerobic and anaerobicd infection	Usually requires treatment of primary problem, i.e., removal of diseased tooth	Penicillin Ga	Trimethoprim‐sulfonamideb
Lung	Bacterial pneumonia	Opportunistic aerobic pathogensc	Neonatal pneumonia often a part of a generalized infection affecting many different organ systems	Penicillin Ga, if just streptococci. Broad‐spectrum antimicrobialse for polymicrobial infection	First‐ or third‐generation cephalosporins
	Bacterial lung abscesses	S. zooepidemicus	Most common cause of pneumonia /bronchitis in older foals	Penicillin G – while penicillin does not penetrate the abscess wall, it can prevent further spread of bacteria from the abscess