Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, January 2006, p. 262-268, Vol. 50, No. 1
0066-4804/06/$08.00+0 doi:10.1128/AAC.50.1.262-268.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
The University of Texas Health Science Center at San Antonio, San Antonio, Texas,1 Singapore Eye Research Institute, The National University of Singapore, Republic of Singapore;,2 Brooke Army Medical Center, Fort Sam Houston, San Antonio, Texas3
Received 8 June 2005/ Returned for modification 21 July 2005/ Accepted 10 October 2005
|
|
|---|
|
|
|---|
|
|
|---|
In vitro susceptibility. In vitro susceptibility for voriconazole was determined by using the CLSI (formerly NCCLS) method modified for mycelial pathogens (14). The MIC of voriconazole for P. lilacinus isolate no. 00-39 was 0.25 µg/ml at 24 h and 0.5 µg/ml at 48 h.
Animals. New Zealand White rabbits, weighing 2 to 3 kg each, were maintained at The University of Texas Health Science Center at San Antonio Animal Unit. Animals were maintained in compliance with the Association for Research in Vision and Ophthalmology Resolution on the Use of Animals in Research and studied under an Institutional Animal Care Committee approved protocol.
Drug. Voriconazole (UK-109496) powder with 99.9% purity was obtained directly from Pfizer Central Research (Sandwich, United Kingdom). It was suspended in 0.3% Noble agar to concentrations of 5 and 10 µg/ml. Voriconazole powder suspended in Noble agar was applied topically twice a day to each treated eye, from day 3 to day 11 after infection (8 days total).
Infection. Corneas of anesthetized rabbits were infected under an operating microscope. A 1.5-mm triangular lamellar midstromal dissection into the right central cornea was performed using an Alcon 3.0-mm keratome. The internal apex of the wound was further extended intrastromally by use of a 27-gauge cannula on a tuberculin syringe, which dissected a further 4-mm narrow intrastromal pocket. Then, 0.02 ml (3.2 x 107) of the conidial suspension was injected under the flap and into the pocket via the 27-gauge cannula. The inoculum was confirmed by cultures of serial dilutions of the conidia. Inoculum is reported as viable conidia.
Treatment. Narcotic analgesics were administered subcutaneously on a fixed schedule to control pain and minimize discomfort to the animals in all groups throughout all experiments.
Control eyes. Five rabbits were infected with P. lilacinus in the right eye only. This group assessed the development of keratitis in untreated eyes (infected untreated control). Measurements of the corneal infiltrates were made twice daily using a ruler and a x4 magnifier. The uninfected treatment control was voriconazole applied in 0.1 ml of Noble agar to the left eyes of these same rabbits. Control rabbits were sacrificed when the fungal corneal lesions reached 4.0 mm in diameter. This occurred by day 5 in all controls.
Infected/treated eyes. In 10 rabbits, voriconazole was initiated on day 3, when keratitis was clearly established, with typically oval lesions with at least one axis of 2 mm in diameter. One group received 5 µg voriconazole suspended in 0.1 ml Noble agar in the right eye, and a second group received 10 µg voriconazole suspended in 0.1 ml Noble agar, also in the right eye. The uninfected left eyes also received the same dose of voriconazole in Noble agar as a control. Ultimately, four eyes in the 5-µg treatment group and all five eyes in the 10-µg treatment group developed corneal infections with one axis of at least 2 mm by day 2 after intrastromal inoculation. All treated rabbits were sacrificed on day 11.
Dissection. All animals were euthanized using intravenous sodium pentobarbital (100 to 150 mg/kg of body weight). Both eyes were dissected with initial removal of a 1- by 1-cm section of conjunctiva from the lower nasal corner of the eye. Corneas were removed with a 9-mm trephine centered on the stromal lesion and then carefully hemisected through the longitudinal axis of the lesion with a 15-degree super-sharp blade. The vitreous and chorioretina were also hemisected. Tissues were placed into different tubes. One sample was used for quantitative cultures, and the other was used for voriconazole assay. This latter sample was stored at 0°C. Prior to analysis, the samples of conjunctiva, cornea, and retina were weighed on a Mettler electronic microbalance.
Voriconazole assay. A high-performance liquid chromatography-based quantitative assay was developed for the analysis of voriconazole by liquid chromatography-electrospray ionization mass spectrometry (4, 19, 21). The separation was achieved on a reversed-phase C18 column eluted with 70% acetonitrile and 0.01% trifluoroacetic acid against 30% water with 0.01% trifluoroacetic acid. The correlation between the concentration of voriconazole to peak area was linear (R2 = 0.998) between 0.04 to 10 ng. The coefficient of variance was less than 3%. The limit of determination was estimated to be 0.1 ng/ml voriconazole in aqueous humor (500 times more sensitive than the conventional high-performance liquid chromatography-UV detection method). Intraday and interday imprecision were both less than 3% over the whole analytical range.
Microbiology. Half of the dissected tissues (i.e., cornea and conjunctiva) was sent for microbiological counts. The samples were homogenized, and serial dilution counts were made. These were incubated at 37°C for 2 days. The limiting threshold count was 5 CFU/ml.
Statistics. Student's t test was used for comparisons of keratitis lesion size [(vertical + horizontal diameter)/2] in millimeters between treatment groups and to compare concentrations of voriconazole between groups. A P value of <0.05 determined significance. Bonferroni's t test and one-way analysis of variance were used for comparisons of tissue voriconazole concentrations. The Mann-Whitney test was used for comparisons of fungal counts in tissues.
|
|
|---|
![]() View larger version (9K): [in a new window] |
FIG. 1. Average diameter of keratitis [(vertical + horizontal diameter)/2] in millimeters, plotted against number of days from the surgical inoculation of the midstromal central cornea among the untreated/infected eyes.
|
2.5 mm in diameter. Figure 2C reveals the corresponding histopathology within the cornea of an animal sacrificed on day 5 without treatment, with massed fungal hyphae spanning the stroma in multiple sections. Figure 2D shows the marked recovery by day 10 of the same eye shown 8 days earlier in Fig. 2A, after receiving 1 week of twice-daily 5-µg topical voriconazole therapy.
![]() ![]() View larger version (170K): [in a new window] |
FIG. 2. Course of infection with and without treatment. (A) Typical external ocular appearance on day 3 with no treatment. (B) Characteristic changes in an untreated eye on day 5, after 2 further days without therapy. Note the pronounced conjunctival hyperemia, complete iridial obscuration by hypopyon, and focal corneal lesion size of 2.5 mm in diameter. (C) Histopathology within the cornea of an animal sacrificed on day 5 without treatment, with massed fungal hyphae spanning the stroma in multiple sections. (D) Marked recovery by day 10 of the same eye shown 8 days earlier in panel A, after receiving 1 week of twice-daily 5-µg topical voriconazole therapy.
|
2 mm in either dimension within 3 days of inoculation.
![]() View larger version (16K): [in a new window] |
FIG. 3. Average size of keratitis [(vertical + horizontal diameter)/2] in millimeters, plotted against number of days from the surgical inoculation of the midstromal central cornea in the in the 5-µg and 10-µg twice-daily (bid) topical voriconazole-treated eyes. Treatment was started on day 3, when the focal keratitis lesion reached a diameter of at least 2 mm in either the vertical or horizontal axis. One animal failed to produce a corneal lesion of 2 mm in either dimension by day 3.
|
Distribution of voriconazole among different eye tissues. Consistent patterns related to the treatment concentrations of infected and noninfected eyes in the cornea, chorioretina, and vitreous were seen (Table 1). First, 10-µg dosing gave concentrations approximately twofold higher than did 5-µg dosing in all compartments of infected and noninfected eyes. Second, in all three compartments at each concentration, the relative corneal concentrations were markedly higher than the posterior segment concentrations, and the chorioretinal concentration was approximately twice that found in the vitreous. Third, noninfected eyes had twofold-higher concentrations than infected eyes. A single exception was the cornea, where concentrations at 5 µg for infected and noninfected eyes were similar. In any case, all tissues at both doses, infected or uninfected, had voriconazole concentrations much higher than the MIC of P. lilacinus.
|
View this table: [in a new window] |
TABLE 1. Ocular tissue concentrations of voriconazolea
|
![]() View larger version (7K): [in a new window] |
FIG. 4. Quantitative cultures of P. lilacinus in corneas of infected rabbits. Each symbol represents one rabbit, one eye. Horizontal lines are medians.
|
|
|
|---|
The experimental model approved by our Institutional Animal Care and Use Committee allowed us to conservatively assess the initiation of infection in untreated eyes and then, applying humane endpoints, evaluate the potential therapeutic efficacy and quantify the corneal and intraocular distribution of voriconazole in infected eyes and uninfected fellow control eyes for 11 days or until endpoints were reached, whichever occurred first. A nontreated infected fellow eye subgroup was considered but would have necessitated sacrifice of each animal according to our approved protocol criteria once either eye succumbed. Since progressive increase in lesion size was observed with all five infected eyes, with accompanying changes necessitating euthanization of all five animals by day 4, parallel study of infected/untreated fellow eyes was not approved.
Treatment responses were measured by multiple criteria, including reduction in size of infection, fungal tissue colony counts, and histopathologic evaluation. Voriconazole inhibited the progression of corneal infection by P. lilacinus beyond a single-axis 2-mm diameter in 8 of the 10 infected eyes receiving topical therapy. Three rabbits demonstrated complete resolution of their fungal keratitis. Quantitative fungal cultures showed reduction in keratitis in eyes treated with 5 µg and 10 µg voriconazole.
The rabbits responded equally well to both voriconazole treatment regimens, with no apparent adverse effect. Voriconazole is poorly water soluble, and either dose is likely to have saturated the tear film upon application. Clumping of voriconazole crystals is the likely source of the high concentrations of the drug seen in the cornea and may be the reason for our failure to observe a dose-response effect. The demonstration of quantifiable drug penetration into the posterior segment may be of clinical relevance, since many topically applied agents fail to accumulate in the vitreous or chorioretina. Hua and colleagues confirmed that direct intravitreal voriconazole injections of 25 µg/ml cause no electroretinographic or histopathologic abnormality in the retina (10), suggesting that doses far higher than those tested in the present study might be safely tolerated. However, even topical application such as we have done here may be adequate to prevent or treat extension of infection to the deeper eye tissues. Future dose-response studies to address this question are planned. These will explore the use of better solvents to allow higher concentrations of voriconazole to be deposited topically. Such studies will also include comparison of topical, oral (systemic), and combined topical and oral voriconazole to maximize delivery of drug to intraocular tissues and determine the extent of corneal and intraocular drug penetration that can be safely achieved.
Voriconazole has been shown to be highly effective against filamentous organisms (e.g., aspergillosis) and is more potent in invasive aspergillosis than amphotericin B (9, 13). The present study suggests that topical voriconazole may also be useful for the treatment of mycelial infections of the eye. Systemic voriconazole has been used for a few patients with fungal eye infections, and there is a recent clinical report of topically applied voriconazole for Fusarium keratitis (12). Unfortunately, because systemic voriconazole and corneal transplantion were also used, this study was inconclusive for the value of the topical agent. However, it is reasonable to consider that topically applied voriconazole might be superior in delivering drug superficially for the treatment of keratitis and may even have the potential to help counter proliferative fungal endophthalmitis.
This study was sponsored in part by grants from Pfizer (J.G.), Odyssey Ophthalmic, and Research to Prevent Blindness, New York, N.Y. (W.S.).
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»