Approach to therapy

Having diagnosed the presence of bacterial infection, specimens for bacterial culture and susceptibility testing should, wherever possible, be collected before administering an antimicrobial drug. Since there will be some delay in obtaining laboratory results it is important to initiate therapy (on an informed empirical basis). The choice of drug can be based on clinical experience and examination of stained material (apply Gram stain to direct smear).

The laboratory results of bacterial culture and MIC determination (when required) give precise data that may be used in selecting the antimicrobial drug, its route of administration and dosing rate for continuation of treatment (specific therapy). The usual dosage regimen for an antimicrobial product is based on providing plasma concentrations of the drug that will be effective against the majority of susceptible microorganisms and will not cause adverse effects in the horse. The assumption is made that the systemic clearance of the drug is not changed by the disease state, even though this may not always be the case.

Quantitative bacterial susceptibility

The need to determine quantitative susceptibility depends on the microorganism isolated, i.e. the causative organism of the infection. The antimicrobial susceptibility of bacteria such as β-hemolytic streptococci, Rhodococcus (Corynebacterium) equi and Corynebacterium pseudotuberculosis is usually predictable, whereas it is generally necessary to determine the susceptibility of other pathogenic bacteria, especially coagulase-positive staphylococci and enteric microorganisms (Escherichia coli, Klebsiella, Proteus and Salmonella spp.).

Such susceptibility determinations must follow standard laboratory procedures. Quantitative (MIC) data are preferable to qualitative (disk diffusion, Kirby Bauer method) information because of the possibility of tailoring drug dosage to susceptibility of the causative bacterial pathogen.

Location of infection

When selecting the antimicrobial drug with which to commence therapy, knowledge of the pathogenic bacteria that most commonly cause infections at various locations is useful but should be supported by examination of a stained (Gram or Romanowsky-type Wright–Giemsa) direct smear. Bacterial pathogens that have been isolated from various sites of infection are listed in Table 4.1. Because the principal bacterial species isolated from any site of infection may vary with geographic region, specimen collection for bacterial culture prior to commencing antimicrobial therapy is of paramount importance.

Drug selection for empirical therapy

The suggested drug of choice and alternative drug for initial treatment of infections caused by various microorganisms are presented in Table 4.2. The information in Table 4.2 should be used in conjunction with direct smear examination for drug selection (empirical therapy) while awaiting the results of bacterial culture and quantitative susceptibility testing (when considered necessary). Selection of the drug to use is influenced by location of the site of infection.

In mixed infections with microbiologically diagnosed anaerobic involvement, the concurrent use of metronidazole and benzylpenicillin or gentamicin, depending on whether the primary aerobic bacterial pathogen is Gram-positive or Gram-negative, may represent the antimicrobial drug treatment of choice.

The penicillins, cephalosporins (apart from cefadroxil), aminoglycosides and oxytetracycline must be administered parenterally to horses, while trimethoprim–sulfonamide combination preparations, rifampicin, metronidazole and enrofloxacin (to mature horses only) may be administered PO as paste formulations or by nasogastric tube as oral solutions or aqueous suspensions.

Interpretation of bacterial susceptibility

There are no agreed guidelines for the interpretation of quantitative susceptibility results for equine bacterial pathogens. Suggested interpretative guidelines for MIC breakpoint values are presented ( Table 4.3). The categories are:

Even though the spectrum of activity of individual drugs within each antimicrobial class or subclass in the case of β-lactam antibiotics is, in general, similar, the quantitative susceptibility of microorganisms to individual drugs somewhat differs. Tetracyclines may constitute an exception in that variation among individual tetracyclines in clinical efficacy is largely due to differences in lipid solubility, which influences tissue concentrations attained.

The MIC (q.v.) is the quantitative value normally used to define the in vitro susceptibility of a microorganism. It is used, in conjunction with pharmacokinetic parameters that describe the bioavailability and disposition of antimicrobial agents, to calculate appropriate dosing rates. This approach allows appropriate dosage adjustment to be made for organisms of different susceptibilities. It is the basis for calculating optimal dosage. Even though quantitative susceptibility (an in vitro measurement of activity) generally correlates with clinical efficacy, it cannot be guaranteed to predict the response to therapy. Furthermore, use of the antimicrobial agent to which a pathogenic microorganism is most susceptible in vitro might not be clinically indicated because of its potential to produce adverse effects.

Usual dosage regimens aim to maintain plasma concentrations that are several times the MIC₉₀ for the majority of usually susceptible microorganisms and could be expected to provide effective therapy. At these plasma concentrations, the penicillins, cephalosporins, aminoglycosides, fluoroquinolones, trimethoprim–sulfonamide combinations, rifampicin and metronidazole produce a bactericidal effect, while tetracyclines, sulfonamides (used alone), erythromycin and chloramphenicol produce a bacteriostatic effect on susceptible microorganisms.

Although the terms bactericidal and bacteriostatic generally apply to the effect produced, they are relative depending on the concentration attained and environmental conditions at the site of infection. For antimicrobial agents that exert a bacteriostatic effect it is especially important that plasma concentrations be maintained well above the MIC₉₀ for the pathogenic microorganism throughout the course of therapy. For antimicrobial agents that produce a bactericidal effect the relationship between plasma concentration and bacterial killing depends on whether the mechanism depends on concentration or time.

For drugs with a concentration-dependent killing mechanism (fluoroquinolones and aminoglycosides), it is usual to recommend achievement of C_max:MIC ratios of 10 or greater, whereas for time-dependent bacterial killing drugs (β-lactam antibiotics) it is desirable to maintain the plasma concentration between 1 and 4 times the MIC for at least half the inter-dose interval.

Impairment of host defense mechanisms is a variable, not generally taken into account in dosage calculations, which contributes to a discrepancy between expected and actual response to therapy, particularly for bacteriostatic drugs. Inadequacy of the type, quality and quantity of the immunoglobulins, alteration of the cellular immune system, or either a qualitative or, more importantly, a quantitative defect in phagocytic cells may result in therapeutic failure despite the use of otherwise appropriate and effective drugs. The effectiveness of therapy may be influenced by local factors at the site of infection. These include pus, the presence of a foreign body, and in the case of abscesses penetrative capacity of the drug and the pH or anaerobic environment within the abscess cavity.

Optimal dosage

Optimal dosage of a drug that produces a bactericidal effect is required in severe infections such as meningitis and endocarditis (q.v.), and in immunocompromised patients. Because of variations in the state of host defenses, the post-antibiotic effect or acquisition of bacterial resistance, and disease-induced changes in the disposition of an antimicrobial agent, optimal dosage cannot be precisely calculated. Disease conditions that may alter pharmacokinetic parameters include fever, dehydration, hypoalbuminemia (e.g. chronic liver disease), uremia and decreased renal excretion (q.v.).

The pathophysiology of the disease condition determines the changes in volume of distribution, clearance and, depending on the changes in these two parameters, the half-life of the drug may be altered. At least some of the pharmacokinetic alterations can be accommodated by adjustment of dosage. The ultimate criterion of optimal dosage is the effectiveness of therapy, as indicated by bacteriologic cure.

Routes of administration

The route of administration of a drug is governed mainly by the formulation of the dosage form (antimicrobial drug preparation). Parenteral preparations of antimicrobial drugs that are formulated as aqueous solutions can be administered either by slow IV injection or IM. Those formulated as aqueous suspensions are designed to provide sustained release (slow absorption), and thereby a prolonged duration of effective plasma concentrations, and should be administered only by IM injection. Prolonged-release products, which cause tissue irritation, and drugs in oily vehicles should never be administered to horses. This excludes the use in horses of some so-called “large animal parenteral preparations”.

At the present time, few antimicrobial drugs are available as oral dosage forms that are convenient to administer to horses. Oral paste preparations that are palatable and of suitable consistency may represent the best type of oral dosage. The addition of an antimicrobial agent to the feed as a powder is an unreliable method of dosing, due to variability in the dose ingested and to the effect of food on the extent of absorption of the drug.

The action of some antimicrobial agents, whether administered orally or parenterally, on the cecal and colonic commensal microorganisms of horses can cause serious digestive disturbances. This adverse effect makes penicillins unsuitable for oral administration to horses other than newborn foals and limits the use of macrolides and, under certain circumstances, of tetracyclines. Foals appear to be less susceptible than adult horses to disturbance in the balance of commensal microbial flora in the large intestine. It is likely that the microbial balance is related to the composition of the usual diet.