Materials and Methods

Horses—For the first portion of the study (evaluation of ciprofloxacin), 2 geldings, 1 stallion, and 3 mares with a median age of 9.5 years (range, 3 to 18 years) and median body weight of 393 kg (range, 318 to 454 kg) were used. With the exception of 1 mare that was replaced by another mare, the same horses were used for the second portion (evaluation of moxifloxacin). The median age in that group was 9.2 years (range, 3 to 18 years), and the median body weight was 386 kg (range, 318 to 454 kg). Breeds represented included Quarter Horse (n = 3), Thoroughbred (2), Paint (1), and Dutch Warmblood (1).

Results of physical examinations, CBCs, and serum biochemical analyses performed prior to initiation of the study were within reference limits for all horses. In addition, all horses tolerated administration of ocular medications without resistance. Complete ophthalmic examinations before the study began revealed that all horses were clinically normal. These examinations included a Schirmer tear test,^a fluorescein dye^b staining, intraocular pressure estimation via applanation tonometry,^c,d slit-lamp biomicroscopy,^e and indirect ophthalmoscopy.^21,f,g Prior to the second portion of the study, fluorescein dye staining, intraocular pressure estimation, and slit-lamp biomicroscopy (performed to ensure no overt ocular surface damage or anterior uveitis potentially related to the first portion of the study was present) were performed, revealing no ophthalmologic abnormalities. During the study, horses were kept indoors and fed their typical rations of hay and pelleted feed. All horses had access to water at all times. Use of horses was approved and monitored by the North Carolina State University Institutional Animal Care and Use Committee.

Topical ocular drug administration—In the first portion of the study, 0.2 mL of 0.3% ciprofloxacin solution^h was applied at 4-hour intervals (0, 4, 8, 12, 16, 20, and 24 hours) to the right eye of each horse via a 1-mL syringe and the hub of a 25-gauge needle. Following a 2-week washout period and repeated ocular examinations in which no clinical abnormalities of the anterior segment of both eyes were detected, 0.2 mL of 0.5% moxifloxacin solutionⁱ was administered to the left eye of each horse at the same 4-hour intervals, in the same manner. At the same time points in both portions of the study, the eye contralateral to that receiving the antimicrobial treatment received sterile balanced salt solutionⁱ as a control treatment by use of the same administration protocol. Each treatment was performed by the same investigator (ABC), who was aware of the treatment each eye received. The frequency of treatment was chosen because such a regimen is common in the treatment of infectious keratitis in horses.^l Prior to and for 60 seconds after each administration, horses were observed for signs of ocular irritation (eg, blepharospasm, blepharedema, conjunctival hyperemia, or epiphora) and examined via slit-lamp biomicroscopy if any clinical problems were evident.

Sample collection—One hour after administration of the final dose (ie, at 25 hours), AH samples were collected from the treated and control eyes of all horses. This time point was selected as described elsewhere.²⁰ Each horse was sedated with detomidine^k (0.01 mg/kg [0.005 mg/lb], IV). Auriculopalpebral and frontal nerve blocks were performed by use of 2% lidocaine solution¹ (1.5 mL/eye, SC) for eyelid akinesia and anesthesia.²¹ A retrobulbar nerve block was performed with 10 mL of 2% lidocaine solution/orbit²¹ to achieve extraocular muscle paralysis and facilitate sample collection. The corneal and conjunctival surfaces were irrigated with dilute povidone-iodine solution, and corneal anesthetic^c (1 mL, divided) was topically administered approximately 2 to 3 minutes prior to sample collection.

Samples of AH were obtained at the limbus by inserting a 27-gauge needle attached to a 1-mL syringe and gently aspirating 0.3 to 0.4 mL of fluid. Each eye was treated prophylactically with topical application of triple antibiotic ointment^m immediately after AH collection and an additional 3 times in the subsequent 24 hours. Each horse received a dose of flunixin meglumineⁿ (1.1 mg/kg, [0.5 mg/lb], PO) immediately following paracentesis. At 24 (immediately after the final dose), 24.25, 24.5, and 25 hours (the time of AH collection), blood samples were collected via jugular venipuncture into lithium heparin tubes.^c Within 60 minutes after collection, all AH samples were frozen at −80°C in sterile Eppendorf tubes, and all plasma was separated and frozen at −70°C in sterile polypropylene tubes until HPLC analysis.

Analysis of ciprofloxacin and moxifloxacin drug concentrations—Concentrations of ciprofloxacin and moxifloxacin in AH and plasma were determined via HPLC with fluorescence detection at an excitation wavelength of 280 nm and an emission wavelength of 500 nm.^p Plasma samples were prepared by use of a solid-phase extraction technique validated in horses.^22,23 The AH samples were analyzed directly by use of HPLC without extraction. Separation was achieved with a 3.5-μm-pore size C8 column and guard column^q maintained at 40°C. The mobile phase consisted of de-ionized water with 0.1% trifluoroacetic acid and HPLC-grade acetonitrile at 83:17 (vol/vol) for ciprofloxacin and 70:30 (vol/vol) for moxifloxacin. Flow rate was set at 1 mL/min for ciprofloxacin and 1.2 mL/min for moxifloxacin. The samples were maintained at 5°C, and the injection volume was 50 μL for each sample.

Calibration curves were prepared prior to each day's run in plasma or AH for plasma and AH samples, respectively. Curves were linear (R² > 0.99) over a range of 0.001 to 1 μg/mL for plasma and 0.005 to 1 μg/mL for AH and calibration samples were back-calculated to be within 15% of the true concentration. The LLOQ was set as the lowest concentration that was linear on the calibration curve.

Statistical analysis—Mean AH concentrations of antimicrobials were compared with a paired Student t test. For each analysis, a value of P < 0.05 was considered significant. All means, SDs, and probabilities were calculated by use of statistical computer software.^r Plasma pharmacokinetics of moxifloxacin were analyzed by use of a commercially available software program.^s Noncompartmental analyses were performed by use of an extra-vascular input model.^t Results are reported as mean ± SD.

Results

Ciprofloxacin—All 6 horses completed this portion of the study. Ciprofloxacin was detected in the AH of all 6 treated eyes, with a mean ± SD concentration of 0.009 ± 0.008 μg/mL. Ciprofloxacin concentrations were less than the LLOQ in all samples from control eyes and in all plasma samples at all collection points. No ocular or systemic adverse effects were detected in any horse during the study period.

Moxifloxacin—Five of 6 horses completed this portion of the study. One horse was withdrawn approximately 8 hours after testing began because of gas colic, likely related to a change in feed instituted over the previous 2 weeks at the housing facility (2 other horses not in the study had similar gas colic at the same time). Moxifloxacin was detected in the AH of all 5 treated eyes, with a mean concentration of 0.071 ± 0.029 μg/mL, and was lower than the LLOQ in the AH of all control eyes. Moxifloxacin was detected in the plasma of all horses at all sample collection points after the final drug administration (Figure 1). Maximum plasma concentrations of 0.015 ± 0.014 μg/mL were reached at 24.2 ± 0.11 hours after administration of the first dose of moxifloxacin. Results of pharmacokinetic analysis indicated a plasma half-life of 0.35 ± 0.10 hours and an area under the curve extrapolated to infinity of 0.01 ± 0.01 h•μg/mL. There were no significant differences in blood concentrations at any sample collection point. No ocular adverse effects were detected following administration in any horse during the study period. No systemic adverse effects attributable to moxifloxacin administration were detected in any horse during the study period.

View Full Size

Mean concentrations of moxifloxacin in the plasma of 5 horses at 5 points after topical ocular administration of 7 consecutive doses of 0.5% moxifloxacin solution (0.2 mL) given 4 hours apart, beginning at 0 hours and ending at 24 hours.

Citation: American Journal of Veterinary Research 71, 5; 10.2460/ajvr.71.5.564

• Download Figure
• Download figure as PowerPoint slide

Ciprofloxacin versus moxifloxacin—The mean AH moxifloxacin concentration in the treated eyes was significantly (P < 0.001) greater than the mean AH ciprofloxacin concentration.

Discussion

Results of the study reported here indicated that detectable intraocular concentrations of antimicrobials were achieved with repeated topical ocular application of 0.3% ciprofloxacin solution and 0.5% moxifloxacin solution in horses with ophthalmologically normal corneas. Furthermore, repeated topical application of these medications did not cause overt ocular irritation. Comparatively, moxifloxacin AH concentrations were significantly higher than those of ciprofloxacin, consistent with results of studies^13,20 in other species. In humans, topical ocular administration of these drugs prior to routine cataract surgery resulted in a mean moxifloxacin AH concentration of 1.31 μg/mL, whereas the mean ciprofloxacin concentration was 0.15 μg/mL.¹³ In rabbits with Staphylococcus aureus endophthalmitis, a mean AH concentration of 43.3 μg/mL was achieved for moxifloxacin and that for ciprofloxacin was 3.65 μg/mL.²⁰

Although our finding of greater intraocular penetration of moxifloxacin versus ciprofloxacin in horses is consistent with findings in other species, the actual AH concentrations of both drugs were lower than those measured in other species. This difference is likely attributable to both drug and ocular properties. Movement of drugs across major corneal barriers, specifically the lipophilic corneal epithelium and hydrophilic stroma, is considerably influenced by drug solubility; drugs with intermediate solubility profiles based on molecular structure and formulation pH are better able to penetrate the cornea than are drugs with strictly lipophilic or hydrophilic solubility profiles.⁷ Formulation pH also affects drug penetration because compounds that approximate the tear him pH of 7.5 stimulate less reflex tearing and drug washout, increasing ocular surface residence time and thus potential transcorneal movement.⁷ Molecular size also plays a role because smaller-sized molecules more effectively breach the epithelial barrier and cross the cornea but then are more readily removed from the eye.¹⁴ Of the 2 fluoroquinolone antimicrobials evaluated here, moxifloxacin is more lipophilic, has a pH of 6.8 (vs a pH of 4.5 for ciprofloxacin),^h,i and is a larger molecule.^14,24 Each of these factors could result in the higher AH concentrations of moxifloxicin versus ciprofloxacin measured in our study and others.

Also pertinent to the discussion of drug formulation, whereas moxifloxacin is preservative free,ⁱ ciprofloxacin contains the preservative benzalkonium chloride at a concentration of 0.006%,^h which can induce subclinical epithelial damage and increase transcorneal drug penetration.²⁵ To avoid the potential effect of administration of ciprofloxacin first in our study protocol, moxifloxacin was subsequently administered to the opposite eye of all horses, so AH concentrations would be unaffected by previous application of ciprofloxacin.

The disposition of topically applied ophthalmic medications is also influenced by multiple species-specific ocular-related factors, with differences potentially contributing to the lower absolute AH concentrations obtained in horses versus humans and rabbits. Tear film proteins, which vary among species, affect drug bioavailability by readily binding certain drugs and making them unavailable for absorption, whereas the tear film pH may buffer acidic or alkaline drug formulations, altering the ionization status and relative lipophilicity.⁷ The increased corneal thickness of horses (900 μm)²⁶ versus that of humans (600 μm),²⁷ as well as the varied corneal glycosaminoglycan composition in horses,²⁸ may also impact corneal permeability and drug movement. The volume of the anterior chamber in horse eyes (3.04 ± 1.27 mL)²⁹ is substantially greater than that in humans (0.25 to 0.3 mL)³⁰ and rabbits (0.3 mL),³⁰ which may result in lower absolute concentrations via a dilutional effect in horses. Although ocular and drug properties likely affected the AH concentrations reached in the present study, comparison of study results among species is difficult because of different treatment protocols used (ie, preoperative cataract surgery in people vs a treatment regimen for infectious keratitis in horses) and study subjects (eg, rabbits with endophthalmitis vs horses with healthy corneas).

Evaluation of intraocular drug concentrations provides an indication of transcorneal drug penetration and potential therapeutic usefulness of certain drugs in certain ophthalmic conditions such as bacterial keratitis and endophthalmitis. Further characterization of potential therapeutic benefit is achieved by comparing AH concentrations of antimicrobials with previously determined in vitro MICs of antimicrobials for commonly isolated bacterial agents (ie, P aeruginosa, Staphylococcus spp, and Streptococcus spp). In the absence of published data regarding the MICs of ciprofloxacin or moxifloxacin toward equine ocular isolates, values obtained for human ocular isolates may be used for comparison. Ciprofloxacin MICs range from 0.1 to 1 μg/mL versus gram-negative susceptible organisms^31,32 and 2 to 64 μg/mL versus gram-positive organisms.^17,31,32 In contrast, moxifloxacin MICs range from 0.08 to 5 μg/mL^31,32 for gram-negative organisms and 0.03 to 0.2 μg/mL for gram-positive organisms.^18,31,32 Comparison of these MICs with AH concentrations obtained in our study suggests that, at an AH concentration of 0.009 μg/mL achieved in horses with healthy corneas, ciprofloxacin reaches a concentration effective for treating common bacterial eye infections in horses. In contrast, at an AH concentration of 0.071 μg/mL, moxifloxacin may be effective in treating eye infections with most gram-positive and some gram-negative organisms.

It is important to keep in mind that the AH concentrations achieved in the study reported here may not correspond to concentrations in corneal or intraocular tissues. Such tissues can act as drug reservoirs and potentially increase the antimicrobial effect and efficacy for treatment of infectious keratitis and endophthalmitis.¹¹ In the present study, tissue concentrations were not evaluated because obtaining corneal or intraocular tissue samples was considered too invasive in these clinically normal horses. Additionally, antimicrobial MICs, resistance break points, and susceptibility patterns determined from in vitro data developed to determine potential efficacy of systemic antimicrobial treatment have uncertain validity in generalization to an ocular route of administration, through which more frequent topical administration may allow greater achievable drug concentrations.^7,17,32 Ocular disease can also considerably alter drug disposition. For example, corneal ulceration commonly increases drug penetration, whereas corneal and intraocular inflammation may have more variable effects.^33,34 Another point that should be considered is that bacterial isolates from humans with ocular disease may have different susceptibility patterns and MICs than equine ocular bacterial isolates.

In addition to achieving measurable AH concentrations, moxifloxacin was detected in plasma at all time points within 1 hour after the final topical administration, peaking at 15 minutes thereafter (0.15 μg/mL), possibly coinciding with transit time through the cornea, and then rapidly decreasing to < 0.004 μg/mL within 60 minutes. Systemic absorption of topically administered ocular medications may occur via the epithelium of the nasolacrimal duct system subsequent to ocular surface drainage or via the conjunctival vasculature.⁷ Lipophilic drugs are more readily absorbed, providing a possible explanation for the presence of moxifloxacin, and not ciprofloxacin, in the plasma. An additional consideration is that although no miosis, aqueous flare, or hypotony was noticed at the time of the second AH sample collection in the moxifloxacin portion of the study, the initial procedure performed in the ciprofloxacin portion may have induced a subclinical breakdown of the blood-ocular barrier that resulted in movement of moxifloxacin from the eye to systemic circulation. Regardless of the mechanism, the plasma concentrations in horses when they received topical moxifloxacin administration were minimal and are not expected to be associated with any therapeutic effect.

When drugs are systemically absorbed, the potential for toxic effects exists. Parenteral administration of fluoroquinolones in humans has been associated with phototoxic effects, neurotoxic effects, cardiovascular disturbances, gastrointestinal effects, and tendinopathies; however, molecular characteristics of moxifloxacin decrease its toxic potential.²⁴ In horses, the most common and severe adverse effect of systemic fluoroquinolone administration is arthropathy development in foals.^u Systemic administration of ciprofloxacin and moxifloxacin has been associated with the development of moderate to severe diarrhea in horses^23,35; therefore, the use of these drugs is limited to topical administration. It is not believed that the adverse effects in the horse withdrawn from our study were attributable to moxifloxacin administration because the horse's clinical signs and response to treatment were consistent with the gas colic episodes of other horses at the same location in the same interval, the horse did not develop diarrhea, and signs resolved with medical management.

In the present study, the mean AH concentration of moxifloxacin was significantly greater than that of ciprofloxacin after 24 hours of an every-4-hour dosing regimen. Neither drug appeared to cause acute ocular discomfort or clinically detectable deleterious ocular surface changes. In veterinary practice, animals with infectious keratitis and endophthalmitis may be treated more frequently than every 4 hours, which may increase corneal penetration of topically applied drugs, particularly in the presence of corneal ulceration. Considering the incidence and potential severity of infectious ocular disease in equine patients, evaluation of the ocular pharmacokinetic parameters of topically applied ciprofloxacin and moxifloxacin may assist in determining potential efficacy in clinical situations. To better understand and characterize the potential therapeutic efficacy of topical ocular administration of fluoroquinolones, additional research is warranted to determine corneal and intraocular tissue concentrations as well as intraocular penetration of both drugs in the presence of corneal disease.