Pharmacokinetics of Intraocular Drug Delivery of Oregon Green 488–Labeled Triamcinolone by Subtenon Injection Using Ocular Fluorophotometry in Rabbit Eyes

Abstract

purpose. To evaluate the transscleral delivery of Oregon Green–labeled triamcinolone acetonide (OGTA) into the eye.

methods. Ex vivo experiments were performed on rabbit sclera in a Lucite block perfusion chamber. Two hundred microliters OGTA (5 mg/mL) was placed on the outer surface of the sclera for 24 hours. The exposed sclera was divided into two pieces; one half for a washout of OGTA and the other for histology. The concentration of OGTA that diffused through the sclera (n = 6) was measured by fluorometry. Two hundred microliters of OGTA (5 mg/mL) was also injected subtenon into live (n = 6) and euthanatized rabbits (n = 3). Intraocular OGTA concentrations were measured by ocular fluorophotometry.

results. The permeability constant for the transscleral diffusion (K _trans) of OGTA was 1.12 × 10⁻⁷ ± 0.08 cm/s (n = 8) during the steady state perfusion. Washout tests showed higher OGTA concentration in the sclera exposed to OGTA for 4 hours than in that exposed for 1 hour. Fluorescent microscopy showed OGTA fluorescence throughout the exposed sclera, as evidence of scleral penetration of OGTA. The maximum OGTA concentration in the retina/choroid after subtenon injection was 25.77 ± 10.26 ng/mL in the live rabbit at 3 hours and 84.68 ± 21.04 ng/mL in the euthanatized rabbits at 8 hours.

conclusions. OGTA is capable of diffusing across isolated rabbit sclera ex vivo and into the retina/choroid via transscleral diffusion from a subtenon depot in vivo. Conjunctival and choroidal circulation decreased the drug delivery of OGTA.

Triamcinolone acetonide (TA) is one of the most commonly used steroids in the treatment of retinal diseases because of its duration of sustained release. ^{1 2} Its anti-inflammatory, antivascular permeability, and antiangiogenic effects have also been used to treat uveitis, macular edema, and various intraocular neovascularizations for many years. ^{3 4 5}  

TA can enter the retina after administration via intravitreal, periocular, and systemic routes. Among these routes, intravitreal injection of 4 mg TA (Kenalog-40; Bristol Myers Squibb, Princeton, NJ) appears to be the most effective method of achieving maximum drug concentration. However, this method must be repeated every 8 to 12 weeks and may cause severe complications, including vitreous hemorrhage, retinal detachment, and endophthalmitis. ^{6 7} These complications can be avoided by a subtenon injection of 40 mg TA. Subtenon injections have shown a lower incidence of serious complications than have intravitreal injections. ^{7 8} However, results in a study comparing the effects of 4 mg of intravitreal versus 40 mg of subtenon TA for diabetic macular edema showed that intravitreal TA injection had a greater clinical effect than the subtenon injection. ^{8 9} Inoue et al. ¹⁰ also confirmed much higher vitreous concentrations after an intravitreal TA injection than subtenon injection, which may be linked to an accelerated therapeutic effect. A better understanding of the pharmacokinetics of a subtenon TA injection could be useful in helping clinicians optimize delivery of TA by subtenon injection, perhaps resulting in effectiveness comparable to that of an intravitreal TA injection. 

Few studies have been conducted to investigate the pharmacokinetics of a subtenon TA injection. Mora et al. ¹¹ performed an ex vivo study using sclera and a Franz-type vertical diffusion cell. They found that TA diffused through human sclera, and its permeability rate was 1.47 ± 0.17 × 10⁻⁵ cm/s. Robinson et al. ¹² performed subtenon injections of 10 mg of TA in rabbits and found vitreous drug levels of 16.56 ± 16.44 μg/mL after 3 hours. Although ex vivo studies show hourly changes of TA diffusion across the sclera over 24 hours, no other tissues can be evaluated. Traditional in vivo methods of evaluating ocular pharmacokinetics are invasive and involve either one-time sampling of the aqueous and vitreous in humans during surgery, ^{13 14 15 16} or euthanatizing animals at various time points, ¹¹ followed by enucleation, dissection, and isolation of the vitreous. In vivo animal studies, to measure TA at various time points have not been reported. 

Ocular fluorophotometry is a noninvasive technique that does not require anesthesia, does not disturb ocular structures, and is capable of determining the concentration of fluorescein-labeled compounds in the aqueous, vitreous, and retina on a real-time basis at different time points. Oregon Green 488 (OGTA; Invitrogen-Molecular Probes, Eugene, OR)–labeled TA was synthesized to perform noninvasive in vivo fluorophotometry. 

Ghate el al. ¹⁷ used fluorophotometry to evaluate subtenon sodium fluorescein (NaF) injections in the rabbit in vivo and found that early changes in the intraocular concentration of NaF were easily measured. They found that a subtenon injection achieved the highest vitreous and retina/choroid (909 ± 1014 ng/mL) NaF concentration 1 hour after injection. 

If TA were labeled with fluorescein, in vivo fluorophotometry could be used to measure TA transscleral diffusion. However, a fluorescein label has some inherent limitations such as being readily photobleached ¹⁸ and pH sensitivity. ¹⁹ It is also not easy to label TA with fluorescein. Therefore, we used Oregon Green 488, which is an analogue of NaF to fluorescently tag TA. Oregon Green shows less photobleaching, it is more pH stable, and it has a similar molecular mass and fluorescent spectrum as NaF (376 Da), which has been used clinically to evaluate the integrity of the blood–retinal and blood–aqueous barriers. 

The purpose of this study was to assess the transscleral diffusion of triamcinolone labeled–Oregon Green 488 across rabbit sclera ex vivo and after a single subtenon injection in vivo. 

Materials and Methods

TA is a synthetic acetal derivative of 9α-fluorurate steroid with a molecular mass of 434.5 Da. It has a solubility of 0.012 mg/mL in water, 1.44 mg/mL in 40% ethanol, and 20 mg/mL in acetone. ²⁰ Oregon Green 488 is the analogue of sodium fluorescein, with a molecular weight of 368.29 and the molecular formula C₂₀H₁₀F₂O₅. For these experimental procedures, we obtained 15 mg of custom-made orange, solid OGTA. The molecular formula of OGTA is C₄₅H₃₉F₃O₁₂, with a molecular mass of 828.79 g/mol (excitation maximum: 502 nm; emission maximum, 643 nm; purity, 96% at 254 nm). 

Ex Vivo Study

Scleral Permeability.

Although OG is an analogue of sodium fluorescein, there is poor experimental evidence of scleral penetration of OG. Therefore, before using OG as a fluorescent tag for TA, OG’s penetration of the sclera had to be verified. Because OGTA was an expensive custom-made drug ($7000/5 mg), we wanted to verify the ex vivo sclera penetration before using it in vivo. 

Eighteen scleral sections were obtained from rabbit eyes (Pel-Freeze, Rogers, AR) that had been stored in moist chambers for no longer than 2 days. Episclera and uveal tissue were carefully removed. The sclera was mounted in a specially designed Lucite block perfusion chamber. ²¹ The perfusion chamber clamped the sclera between two Sylgard rings: one on the donor chamber (the episcleral side) and the other on the receiver chamber (the choroidal side). 

OG was prepared as a 2.5 mg/mL solution and used first before the custom-made OGTA (n = 6). Another group of tissue (n = 6) was used for OGTA (5 mg/mL), and another control set with balanced salt solution (BSS; Alcon Laboratories, Fort Worth, TX; n = 6). Two hundred microliters of OG, OGTA, and control were placed in the donor chamber on the episcleral surface. The uveal surface (44.2 mm²) was perfused with balanced salt solution at a rate of 0.003 mL/min. A temperature of 36.5°C was maintained with a circulating water bath. The perfusate of the uveal side was collected in hourly fractions using a fraction collector. OG and OGTA concentration in the fractions were measured with a fluorometer (TD-700; Turner Biosystems, Sunnyvale, CA). 

The diffusion constant (K _trans) was calculated by using a stable flux period for OG, and OGTA and was calculated as:  

\[K_{\mathrm{trans}}{=}{[}R_{\mathrm{total}}/(A{\times}t){]}{\times}(1/D)\]

where R _total equals the total moles of drug diffused through the sclera in time t (in seconds); A, the surface area of the sclera (in square centimeters); and D, concentration of original solution in the donor chamber (moles per milliliter). 

All values were recorded as the mean ± SEM. A t-test was used to determine the significance between mean values of OGTA and balanced salt solution. P < 0.05 was considered significant. 

Fluorescent Microscopy.

To investigate OGTA absorption in scleral tissue, we removed the sclera from the perfusion chamber and cut a scleral button with a 7.5-mm trephine. The button was bisected, and each half was weighed. One half of the bisected sclera was embedded in low-viscosity epoxy medium and frozen for cryostat sectioning. The unstained frozen sections were viewed with an epifluorescent microscope (Nikon, Melville, NY), and photomicrographs were taken at 400× to determine the spatial distribution of drug absorbed by the sclera. Fluorescence in the OGTA-exposed sections were compared with photomicrographs of control sclera exposed to balanced salt solution. 

Scleral Washout.

For the washout studies, the other half of the sclera (∼15 mg), was placed into a well of a tissue culture plate containing 1 mL of balanced salt solution. The sclera section was transferred to a fresh well of balanced salt solution every 30 minutes over a 6-hour period. The OGTA that diffused from the sclera into each well was measured by a fluorometer. 

In Vivo Ocular Fluorophotometry

These experiments were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Nine Dutch-belted rabbits, weighing 3 to 4 lb, were used in all the experiments. The rabbits were anesthetized with 6 mg/kg of xylazine and 15 mg/kg of ketamine intramuscularly before the subtenon injections. 

OGTA was prepared as a 5 mg/mL solution in balanced salt solution. All rabbits received 200 μL of the OGTA solution as a subtenon injection in the superotemporal quadrant of the right eye with a tuberculin syringe and a 30-guage, 5/8-in. needle. The left eye served as a control. The needle was introduced along the sclera beyond the equator. There was a localized ballooning of the Tenon’s space 5 to 6 mm away from the limbus. The injection site was compressed with a cotton swab after the injection to prevent back diffusion. Copious irrigation of the cornea and conjunctiva with more than 300 mL of balanced salt solution was performed before in situ fluorophotometry (Fluorotron Master; OcuMetrics, Mountain View, CA). An external ocular examination, slit lamp examination, and fundus examination were performed before and immediately after the injection, to evaluate inflammation and/or other drug reactions. 

Baseline fluorescence concentration in the anterior segment, vitreous, and the choroid/retina were measured in both eyes using ocular fluorophotometry with a standard objective lens before the subtenon injection of OGTA. Fluorescence concentrations in various ocular tissues (retina/choroid, vitreous, lens, anterior chamber, and cornea) were obtained as data points 0.25 mm apart along an optical axis measured in the anterior to posterior (AP) direction. The fluorophotometer has an internal standard and gives absolute fluorescence concentrations for each data point. It provides linear concentrations from 0 to 2000 ng/mL. Intraocular fluorescence was measured after a subtenon injection at 10 minute; 30 minutes; 1 hour; and 2, 3, 5, 7, 24, and 48 hours, by which time the fluorescence usually reached baseline levels. 

In one group of rabbits (n = 6), fluorophotometry was performed in vivo. Another group of rabbits (n = 3) were euthanatized immediately after a subtenon injection, and then the measurements were performed. Three scans were taken at each time point. The midvitreous fluorescence levels were defined as the maximum fluorescence between 16 and 20 data points anterior to the retina/choroid peak, and the anterior segment fluorescence levels as the maximum fluorescence between 16 and 20 data points posterior to the corneal peak. 

Results

Ex Vivo Study

Scleral Permeability.

Figure 1shows the diffusion of OG and OGTA across the rabbit sclera as a function of time over 24 hours. The study of transscleral diffusion of OGTA was performed after verifying that OG could penetrate the sclera in the OG diffusion test. OG concentration (mean ± SEM, n = 6) in the perfusate reached its peak at 6 hours with a relative steady state flux observed of 1.0 μg/mL between 8 and 22 hours (Fig. 1A) . The cumulative concentration of OG at 24 hours is 10.5 μg/mL (Fig. 1B) . The OGTA concentration (mean ± SEM, n = 6) in the perfusate showed a difference from its balanced salt solution control group at 2 hours (P < 0.01) and reached a peak of 23.5 ± 4.0 ng/mL at 10 hours, with a relative steady state flux of 22.00 ng/mL observed between 17 and 21 hours (Fig. 1C) . The cumulative concentration of OGTA at 24 hours was 245 ng/mL, approximately one-fortieth that of OG (Fig. 1D) . The average background fluorescence in the control sclera exposed to balanced salt solution corresponded to an OG and OGTA concentration of less than 6.0 ng/mL (Fig. 1) . 

From these data, the permeability constant for the transscleral diffusion (K _trans) of OG and OGTA was calculated as 9.19 ± 0.56 × 10⁻⁶ and 1.12 ± 0.08 × 10⁻⁷cm/s, respectively, during the steady state interval in rabbit sclera. 

Epifluorescent Microscopy.

Sections of the rabbit sclera show maintenance of normal scleral structure after the diffusion studies with OGTA over a 24-hour period. After 24 hours of exposure to OGTA, mild scleral edema was observed, but the collagen fibrils appeared intact and normal. Scleral images using epifluorescent microscopy with 50 ms light exposure show strong fluorescence throughout the sclera with OGTA (Fig. 2A) , but a very weak signal in the control sclera (Fig. 2B) . 

Scleral Washout.

The washout studies show that the sclera exposed to OGTA contained a large amount of OG fluorescence compared with the control sclera, which was exposed to balanced salt solution. With time, the concentration of OGTA that diffused from the sclera decreased from 2.78 ± 1.02 to 0.42 ± 0.28 μg/mL (Fig. 3A) . The t _1/2 was approximately 1.5 hours. The cumulative data show that 13.35 ± 0.28 μg of OGTA washed out of the sclera after 6 hours (Fig. 3B) . 

In Vivo Ocular Fluorophotometry

The compiled results of the in vivo experiments are summarized in Tables 1 and 2 . The maximum fluorescence of the anterior segment, midvitreous, and retina/choroid were analyzed. 

Figure 4is a plot of the mean fluorescence in intraocular tissues after a 0.2-mL subtenon injection of 5 mg/mL of OGTA into six live rabbits (Fig. 4A)and three euthanatized rabbits (Fig. 4B) . In Figure 4A , on the downward slope of the retinal peak, a small hump appears at 3 hours, which is representative of the midvitreous. In live rabbits, the midvitreous hump did not appear until the retinal concentration reached its peak. The midvitreous hump appeared in plots for approximately 3 hours. This hump indicates the increase of OGTA in the vitreous. In the euthanatized rabbits, the midvitreous hump showed a higher concentration than in the retina/choroid, when peak concentration was reached at 3 hours (Fig. 4B) . 

Figure 5is a graph of peak OGTA concentrations in the retina/choroid, vitreous, and anterior segment over time compared with its baseline level (t = 0). In the group of live rabbits, the retina/choroid concentration of OGTA reached its peak at 3 hours after injection. This level was maintained for 3 hours and then decreased to the baseline level after 7 hours. In the group of euthanatized rabbits, the retina/choroid concentration of OGTA reached its peak at 6 hours after injection and quickly declined, not fully decreasing to baseline, even after 7 hours. 

In the group of live rabbits, vitreous and anterior segment concentration of OGTA after subtenon injection increased significantly at 2 hours, reached a peak at 3 hours that was sustained for approximately 3 hours, and returned to baseline at 7 hours. In the group of euthanatized rabbits, the vitreous and anterior segment concentration of OGTA increased 3 hours later and peaked at 6 hours. However, even after 7 hours, the concentration did not return to baseline (Fig. 5) . 

The baseline fluorescence in the retina/choroid before injection corresponded to 5 to 6 ng/mL of OGTA. Corresponding values for vitreous and the anterior segment were 1 to 2 ng/mL. The maximum OGTA concentration in vivo in the retina/choroid was 25.77 ± 8.56 ng/mL (n = 6) at 3 hours after a subtenon injection of OGTA; after euthanatization it was 84.68 ± 21.04 ng/mL (n = 3) at 6 hours after injection of OGTA. The vitreous level peaked at 3 hours after OGTA injection in the live rabbits at 6.98 ± 2.09 ng/mL and peaked at 6 hours in euthanatized rabbits at 83.66 ± 5.68 ng/mL. The anterior segment concentration in live rabbits peaked at 3 hours at 1.74 ± 0.42 ng/mL and in euthanatized rabbits at 6 hours at 20.01 ± 0.84 ng/mL (Fig. 6) . 

Discussion

The results from this study demonstrate that OGTA can diffuse through the sclera and accumulate in the vitreous in rabbit eyes after a subtenon injection. All the ex vivo experiments in this study also indicated that OGTA can diffuse across rabbit sclera with a permeability constant of 1.12 ± 0.08 × 10⁻⁷ cm/s, with the maximum diffusion of OGTA occurring at 10 hours. (Fig. 1) . In the scleral washout, the amount of OGTA that diffused from the tissue increased with exposure time and had a t _{[ 1/2 ]} of 1.5 hours. With fluorescence microscopy, OGTA stained the whole sclera after 24 hours. In vivo experiments using ocular fluorophotometry also showed not only that OGTA can diffuse across the sclera, but it can also diffuse into the vitreous after a subtenon injection. 

The dose of 1 mg OGTA used for in vivo transscleral penetration in our study was smaller than that used in rabbit experiments by Robinson et al., ¹² who used 20 mg TA. In human studies, Inoue et al. ¹⁰ found that 5 mg or less of TA injected subtenon resulted in no detectable vitreous drug levels for a median of 5.5 days after injection in most of the patients. Thomas et al. ²¹ found that 4 weeks after a subtenon injection of 40 mg of TA, intravitreal TA concentration varied considerably, from 0 to 4.93 μg/mL. Robinson et al. ¹² encountered no detectable vitreous drug concentration using high-performance liquid chromatography (HPLC) after a 10-mg subtenon injection of TA. This result suggested that the tissues themselves (i.e., sclera, choroid, and retina) and other factors, such as conjunctival lymph and blood flow may play an inhibitory role in transscleral drug delivery to the vitreous. ¹² However, in this study, even 1 mg of OGTA could diffuse across the sclera and be detected in the vitreous. The current results may be related to the difference in measurement method between HPLC and ocular fluorophotometry, but also a difference in the drug itself. Transscleral drug delivery can also be affected by molecular weight, sclera binding, lipid and water solubility, and molecular radius. 

There is evidence that the transscleral permeability constant (K _trans) is inversely related to the molecular weight of a drug. ²² Of the molecules used in previous transscleral delivery studies, dexamethasone, and dexamethasone-fluorescein are comparable to TA and OGTA in nature and molecular weight. The permeability constant (K _trans) of dexamethasone-fluorescein in rabbit sclera is 1.64 ± 0.17 × 10⁻⁶ cm/s with a weight of 841 g/mol. The K _trans is 20% that of dexamethasone in human sclera at 8.94 ± 1.5 × 10⁻⁶ cm/s with a weight of 392 g/mol. ²³ Rabbit sclera is 65% thinner than human sclera, and that could account for why a twofold larger molecule can penetrate the rabbit sclera five times faster than it penetrates the human sclera. Mora et al. ¹¹ found that the permeability constant of TA (weight, 434.5 g/mol) in human sclera is 1.47 ± 0.17 × 10⁻⁵ cm/s, and its penetration of the sclera is approximately 100-fold faster than that (1.12 ± 0.08 × 10⁻⁷ cm/s) of OGTA, which has greater weight (828.8 g/mol). The reason for this difference can be explained by molecular weight. Other factors such as the molecular radius, scleral affinity, the drug’s partition coefficient, and charge interaction with the sclera are also involved. ^{22 23}  

OG is a fluorinated analogue of fluorescein. Two hydrogens were substituted for 2 fluorines at the 2′ and 7′ positions of fluorescein (2′,7′-difluorofluorescein). OG is less pH sensitive than fluorescein and more photostable. Thus, the OG conjugates and the results from its conjugates may also be more stable. The molecular weight of NaF (376.27 g/mol) is slightly higher than but nearly the same as OG (368.29 g/mol), and the absorption and emission spectra of both are very similar. 

We found that OGTA can diffuse into the sclera after subtenon injection. Although the dose was low (1 mg), the vitreous concentration was still measurable by ocular fluorophotometry. Once the OGTA has diffused into the choroid, it has to cross the blood–retinal barrier to the vitreous. The OGTA levels in the midvitreous and anterior segment were observed to peak at 3 to 4 hours after subtenon injection in live rabbits, immediately after the retina/choroid peaks (Figs. 4 5) . This finding suggests that the vitreous and anterior segment concentration of OGTA are closely and quickly affected by the retina/choroid concentration. 

Lymph and blood clearance may also affect transscleral drug delivery. The posterior subtenon space is isolated from the orbital vessels and lymphatics compared with the sub conjunctiva. Slower clearance and enhanced transscleral drug delivery occur from the depot in the posterior subtenon. However, it was clear in the present study that euthanatization influenced the rapid and massive transscleral movement into the vitreous. Transscleral diffusion was also enhanced in a previous study in which hydrophilic contrast agents and magnetic resonance imaging (MRI) were used after euthanatization, which stopped lymph and blood clearance. Studies by Kim et al. ²⁴ using serial MRI and gadolinium (Gd)-DTPA20 or Mn21 have shown that transscleral diffusion is increased after euthanatization. 

Robinson et al. ¹² evaluated the vitreous levels in a rabbit model 3 hours after a subtenon injection of TA. A conjunctival window was used to negate the effect of conjunctival vessels and lymphatics. Cryotherapy was used to negate the effect of choroidal blood flow, and euthanatization was implemented to eliminate both. Their results show that the conjunctival blood vessels/lymphatics are more important barriers to transscleral drug delivery than the choroid. They also found no detectable vitreous TA after subtenon injection of 10 mg of TA; however, in a 10-mg injection with immediate euthanatization, TA was detected in the vitreous at a concentration of 7.13 μg/mL. 

The sclera is considered to be a major barrier to ocular drug delivery. However, conjunctival, lymphatic, and choroidal vessels have also been shown to be barriers to drug delivery. Euthanatization can stop the conjunctival and choroidal circulation and increases drug diffusion through the sclera. ^{11 12 22 23} In the present study, the peak choroid/retinal concentration of OGTA with euthanatization was approximately 3 times greater than that of OGTA without euthanatization, and the vitreous concentration with euthanatization was 12 times greater than that of OGTA without euthanatization (Tables 1 and 2) . These results also indicate more than 3 times the amount of OGTA that entered the retina is cleared by conjunctival and choroidal clearance, and approximately 12 times the amount of OGTA that could enter the vitreous is cleared by conjunctival, choroidal, and retinal clearance. This clearance results in a reduction in vitreous drug concentration as well. In the present study, without immediate euthanatization, the vitreous TA concentration returned to baseline after 24 hours. An explanation for this rapid reduction in vitreous drug concentration is that the subtenon depot becomes smaller over time, and the drug release rate from the depot is not fast enough to overcome the clearance mechanisms. Mason et al. ²⁵ found that after a single intravitreal injection of 4 mg TA in humans, a vitreous concentration of TA was detectable for up to 2.75 months. Chin et al. ²⁶ found that after an intravitreal injection of 0.3 mg of TA in rabbits, vitreous TA concentration decreased more slowly in the nonvitrectomized eye than in the vitrectomized eye, and the half-life of TA was 2.89 days in the nonvitrectomized eye. The rapid reduction in vitreous drug concentration in the present study is related to the drug release rate from a small depot. Based on these results, it is important to develop a unidirectional delivery device to protect the TA from the conjunctival circulation and lymphatics and thus allow unidirectional diffusion into the sclera. 

Mora et al. ¹¹ found that ex vivo sclera penetration equilibrium of TA was achieved on the fourth day and the diffusion rate of the first day was 30% of the fourth day in human sclera. In our in vivo study, the penetration equilibrium of TA was achieved at 7 hours, and the OGTA levels in the choroid/retina increased linearly after half an hour in rabbit eyes. It took approximately 1 hour for the OGTA to diffuse into the choroid across the sclera from the periocular depot site. In live rabbits, the OGTA in the choroid/retina paralleled the concentration in the contralateral control choroid/retina for 30 minutes and then increased above that in the contralateral control eye. However, the contralateral control eye did not show any specific changes after 30 minutes. This suggests that this amount of drug in a subtenon depot does not affect the contralateral vitreous concentration, and thus, does not enter the systemic circulation for hematogenous dissemination, which is another possible route for steroids to enter the vitreous and recirculate back into the eye. ^{15 27 28}  

The rabbit model is commonly used in ocular drug delivery, even though there are some differences between rabbit and human eyes. Compared with that of the human eye, the rabbit sclera is 65% thinner, has 50% the vitreous volume, and higher choroidal flow rates. ^{29 30} However, the scleral permeability, anatomy, and physiology of the lymphatic system is similar to that of humans. ^{31 32}  

Fluorophotometry is a method for tracing a drug in the living eye and was found to be safe and noninvasive to the animal. Without the fluorescent label, the movement of TA in the retina and vitreous could not be traced. TA concentration cannot be measured by extrapolation from fluorescent concentrations; however, pharmacokinetics of a drug diffused into the vitreous after subtenon injection can be determined. 

When subtenon injection is performed, the depot is near the equator, and the highest concentration of drug anywhere in the retina is likely to be adjacent to the depot. The fluorophotometer gives drug levels in the AP direction. Although Figure 4shows a maximum concentration in the retina of 25 ng/mL, this is not necessarily the highest drug concentration that exists in the retina. In what may be a limitation of the technique, the fluorophotometer gives gradients only in 1 dimension and only AP. Nevertheless, fluorophotometry does allow for a continued measurement in the vitreous and retina without the need to kill animals at each time point. Using this in vivo fluorometry, meaningful pharmacokinetic data can be obtained and potentially reduce the number of animals and costs associated with conducting pharmacokinetic studies to determine drug distribution in ocular tissues. 

 Supported by Eye Institute Grants R24-EY017045 and P30-EY06360 and an unrestricted grant from Research to Prevent Blindness.

 Submitted for publication March 7, 2008; revised May 5, 2008; accepted August 22, 2008.

 Disclosure: S.J. Lee, None; E. S. Kim, None; D.H. Geroski, None; B.E. McCarey, None; H.F. Edelhauser, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Concentration of fluorescence (mean ± SEM, n = 8) in the uveal chamber after transscleral penetration of OG, OGTA, and balanced salt solution control over time. The scleral permeability coefficient of OG was 9.19 ± 0.56 × 10⁻⁶ cm/s, and that of OGTA was 1.12 ± 0.08 × 10⁻⁷cm/s.

Figure 2.

 

Scleral absorption of OGTA. Microscopic findings showed strong fluorescence throughout the sclera after 24 hours of OGTA diffusion (A) and weak fluorescence in sclera samples after balanced salt solution control diffusion (B).

Figure 3.

 

The results of ex vivo washout. Concentration of OGTA and balanced salt solution control based on OG fluorescence at (A) different time point through 6 hours, and (B) cumulative data.

Table 1.

 

OGTA Concentration in Ocular Tissues after 0.2 mL Subtenon Injection (5 mg/mL) into Live Rabbits without Immediate Euthanatization

Tissues	Baseline	10 min	30 min	1 h	2 h	3 h	4 h	5 h	6 h	7 h	1 d	2 d
OGTA Exposed (OD)
Retina/choroid	4.38 ± 1.50	4.87 ± 2.05	7.72 ± 3.21	8.01 ± 2.80	15.52 ± 6.15	25.77 ± 8.56	25.72 ± 7.59	23.23 ± 9.99	24.40 ± 10.65	14.63 ± 5.38	5.15 ± 3.17	5.11 ± 1.58
Vitreous	1.81 ± 0.54	2.63 ± 0.51	2.39 ± 1.85	4.67 ± 1.55	6.18 ± 1.07	6.98 ± 2.09	4.83 ± 2.12	5.75 ± 1.62	5.66 ± 1.73	3.23 ± 1.74	1.09 ± 0.71	0.47 ± 0.24
Anterior segment	0.44 ± 0.17	1.07 ± 0.43	1.71 ± 0.65	1.30 ± 0.61	1.47 ± 0.73	1.74 ± 0.42	1.53 ± 0.53	1.67 ± 0.49	1.32 ± 0.58	1.63 ± 0.70	1.06 ± 0.49	1.14 ± 0.47
Balanced Salt Solution Control (OS)
Retina/choroid	4.62 ± 0.96	4.11 ± 1.49	3.61 ± 1.59	5.26 ± 2.48	5.15 ± 1.36	5.59 ± 1.29	6.69 ± 3.45	6.50 ± 1.55	6.30 ± 2.10	5.73 ± 1.61	4.15 ± 1.72	5.11 ± 1.11
Vitreous	0.60 ± 0.29	0.74 ± 0.38	0.69 ± 0.36	0.50 ± 0.31	0.95 ± 0.41	0.86 ± 0.49	0.92 ± 0.62	0.97 ± 0.51	0.92 ± 0.53	0.80 ± 0.39	0.56 ± 0.33	0.50 ± 0.22
Anterior segment	0.69 ± 0.48	0.98 ± 0.43	1.04 ± 0.56	0.99 ± 0.53	1.01 ± 0.51	1.09 ± 0.71	1.25 ± 0.54	1.20 ± 0.45	0.61 ± 0.22	1.05 ± 0.40	0.53 ± 0.23	0.95 ± 0.60

 Data are shown as the mean concentration of fluorescein (ng/mL) ± SD. The mean peak values are shown in bold (n = 6).

Table 2.

 

OGTA Concentration in Ocular Tissues after 0.2 mL Subtenon Injection (5 mg/mL) into Rabbits with Immediate Euthanatization

Tissues	Baseline	10 min	30 min	1 h	2 h	3 h	4 h	5 h	6 h	7 h
OGTA Exposed (OD)
Retina/choroid	2.71 ± 0.90	2.54 ± 0.11	2.73 ± 1.95	8.01 ± 2.80	21.91 ± 6.93	29.54 ± 8.11	27.99 ± 0.92	47.11 ± 6.63	84.68 ± 21.04	48.31 ± 10.57
Vitreous	0.29 ± 0.25	0.26 ± 0.10	2.17 ± 1.38	4.37 ± 2.34	16.94 ± 3.18	23.43 ± 7.78	20.02 ± 1.96	43.92 ± 4.97	83.66 ± 5.68	45.13 ± 9.47
Anterior segment	0.60 ± 0.09	0.54 ± 0.17	2.04 ± 0.49	3.13 ± 0.44	3.95 ± 0.23	4.32 ± 0.49	4.37 ± 0.92	16.13 ± 2.71	20.01 ± 5.84	19.48 ± 1.60
Balanced Salt Solution Control (OS)
Retina/choroid	2.29 ± 0.90	2.40 ± 0.11	2.54 ± 0.19	2.55 ± 0.32	2.14 ± 0.34	1.26 ± 0.45	1.95 ± 0.52	1.82 ± 0.45	1.81 ± 0.40	2.18 ± 0.22
Vitreous	0.49 ± 0.01	0.43 ± 0.10	0.47 ± 0.08	0.48 ± 0.18	0.31 ± 0.11	0.43 ± 0.22	0.41 ± 0.21	0.35 ± 0.30	0.62 ± 0.30	0.36 ± 0.01
Anterior segment	0.93 ± 0.09	1.00 ± 0.62	1.09 ± 0.28	1.85 ± 0.56	1.75 ± 0.97	1.58 ± 0.44	1.98 ± 0.72	1.82 ± 0.70	2.59 ± 1.04	2.46 ± 0.69

 Data are shown as the mean concentration of fluorescein (ng/mL) ± SD. The mean peak values are shown in bold (n = 3).

Figure 4.

 

Fluorophotometric scan of intraocular OGTA. Vitreous concentration of OGTA which diffused through the rabbit sclera reached a peak at 3 hours at 25.77 ± 10.26 ng/mL after 0.2 mL subtenon injection of OGTA (5 mg/mL) in live rabbits (n = 6) (A) and at 84.68 ± 21.04 ng/mL in euthanatized rabbits (n = 3) (B). High midvitreous OGTA concentration was shown in eyes with immediate euthanatization (B). The fluorescence returned to baseline level after 7 hours in the group of live rabbits but not in the group of euthanatized rabbits. Data represent mean concentrations of OGTA.

Figure 5.

 

Fluorophotometry scans of OGTA. The graph shows that the vitreous and anterior segment concentration was affected by the retinal concentration. The vitreous concentration of OGTA after a 0.2-mL subtenon injection of OGTA (5 mg/mL) in live rabbits increased at 1 hour, reached its peak at 3 hours, and returned to baseline at ∼24 hours.

Figure 6.

 

Fluorophotometry scans of OGTA. The graph shows that the (B) vitreous and (C) anterior segment concentrations were affected by the (A) retinal concentration. The vitreous concentration of OGTA after 0.2 mL subtenon injection of OGTA (5 mg/mL) in live rabbits increased at 1 hour, reached its peak at 3 hours, and returned to baseline at ∼ 24 hours. The control (contralateral) eye did not show any specific changes in drug concentration.