Effect of Posterior Sub-Tenon Administration of Triamcinolone Acetonide on Leukocyte Dynamics in Rat Retinal Microcirculation after Panretinal Photocoagulation

Abstract

purpose. Macular edema is one of the serious side effects associated with panretinal photocoagulation (PRP). The inhibitory effect of triamcinolone acetonide (TA) on leukocyte–endothelial cell interactions in vivo after PRP was evaluated.

methods. Argon laser photocoagulation was performed in one half of the retinas in male Brown Norway rats. Experimental rats were injected with 2 mg TA (50-μL volume) in the posterior sub-Tenon space, and the vehicle-treated rats were injected with the same amount of saline (50 μL) immediately after PRP. Untreated rats were used as the control. Leukocyte dynamics in retinal microcirculation and retinal vessel diameters were evaluated 1 day after laser photocoagulation with the use of acridine orange digital fluorography. Retinal thickness was evaluated with optical coherence tomography.

results. The number of rolling leukocytes and accumulating leukocytes in the retina decreased by 66% in the TA-treated rats (P < 0.01) and by 24% (P < 0.05), respectively, compared with the number in the vehicle-treated rats. Retinal thickness in the vehicle-treated rats was significantly thicker than that in control rats 1 day after laser photocoagulation (P < 0.01). Retinal thickness in the TA-treated rats was significantly suppressed compared with that in the vehicle-treated rats (P < 0.05).

conclusions. Sub-Tenon administration of TA significantly suppressed leukocyte dynamics in rat retinal microcirculation and decreased retinal edema after laser photocoagulation. The results suggest that the suppression of leukocyte–endothelial cell interactions in retinal microcirculation may be one mechanism responsible for the therapeutic effect of sub-Tenon TA on postlaser retinal edema.

Macular edema, which results in transient or persistent visual disturbances, is one of the serious side effects associated with panretinal photocoagulation (PRP), despite the treatment’s significant suppressive effect against neovascularization. ¹ Although it has been suggested that autoregulatory changes in the distribution of retinal blood flow are involved in the pathogenesis of postlaser macular edema, the exact mechanism remains unclear. Nonaka et al. ² reported that PRP increased leukocyte rolling and subsequent accumulation in both the photocoagulated and the untreated portions of retinal microcirculation in a rat model. It also increased the accumulation of leukocytes involved in augmented vascular permeability in the untreated retina, which resulted in retinal edema. Leukocyte accumulation in the untreated half of the retina peaked 24 hours after PRP and returned to normal levels by 48 hours. Leukocyte–endothelial interactions are regulated by multistep processes, ³ with each step mediated by distinct adhesion molecules. ⁴ Leukocyte rolling is the first step in a cascade of events that lead to firm adhesion and transmigration through the endothelium. Leukocyte strong adhesion is induced by the upregulation of CDllb/CD18 and activation of intercellular adhesion molecule (ICAM)-1. In this study, to investigate the induction of ICAM-1 in the retina by PRP, immunohistochemical analysis was performed 24 hours after PRP. 

Recently, triamcinolone acetonide (TA) injection has been reported to be useful in the management of macular edema caused by uveitis, ^{5 6} retinal vein occlusion, ^{7 8} and diabetes. ^{9 10 11} Zacks and Johnson ¹² reported that combination treatment with TA and PRP by intravitreous injection is beneficial in patients with diffuse diabetic macular edema who are in need of urgent PRP for proliferative diabetic retinopathy, to prevent the exacerbation of macular edema. 

We have reported retinal microcirculatory disorder involved in leukocyte–endothelial cell interactions using acridine orange (AO) digital fluorography to visualize leukocytes and to evaluate their behavior in retinal microcirculation in vivo. ^{13 14 15 16} We quantitatively evaluated the inhibitory effect of TA on leukocyte–endothelial cell interactions in retinal microcirculation of rats after PRP by using AO digital fluorography. We also investigated retinal edema by using optical coherence tomography (OCT). 

Materials and Methods

Animal Model

Male Brown Norway rats weighing approximately 200 to 250 g each were used in accordance with the ARVO Statement for the Use of Animals in Vision and Ophthalmic Research. Only one eye of each rat was used. The rats were anesthetized for all procedures with a mixture (1:1) of xylazine hydrochloride (4 mg/kg) and ketamine hydrochloride (10 mg/kg), and the ocular surface was then anesthetized with topical instillation of 0.4% oxybuprocaine hydrochloride. The pupils were dilated with 0.5% tropicamide and 2.5% phenylephrine hydrochloride. Argon laser photocoagulation was delivered through a slit lamp biomicroscope; 400 laser spots were placed in half of the retina, and the other half was left untreated. The laser settings used were as follows: spot size, 100 μm; duration, 0.05 seconds; and power, 50 mW. Immediately after photocoagulation, a single dose of 2 mg TA (Kenakort-A; Bristol-Myers Squibb, Tokyo, Japan) in a volume of 50 μL (40 mg/mL) was slowly injected into the posterior sub-Tenon space with a 30-gauge needle under a dissecting microscope. For the vehicle-treated rats, 50 μL physiologic saline was injected into the posterior sub-Tenon space. Nonsurgical rats that had not undergone PRP were used as control subjects. The number of animals for each phase of the study is shown in Table 1 . 

Immunohistochemistry

For the investigation of ICAM-1 expression induced by PRP in the retina of the control and vehicle-treated rats, the eyes were enucleated 24 hours after PRP. Immunohistochemical analysis was performed as previously reported. ^{17 18} The eyes were bisected behind the limbus and kept in 4% PFA at 4°C overnight and then were cryoprotected with PBS (0.1 M phosphate buffer, [pH 7.4] 0.15 M NaCl) containing 20% sucrose. The cryosections (10 μm) were mounted on slides, incubated with blocking buffer (PBS containing 10% goat serum, 0.5% gelatin, 3% BSA, and 0.2% Tween 20), and incubated with a primary antibody against ICAM-1 (dilution 1:100; Santa Cruz Biotechnology, Santa Cruz, CA). The sections were then incubated with secondary antibody including goat anti-rabbit immunoglobulin G conjugated to Alexa Fluor 546 (Invitrogen-Molecular Probes, Eugene, OR) and diluted 1:200 in blocking buffer for 1 hour. For double staining with isolectin B4 as a marker for vascular endothelium, the sections were washed in PBS and incubated in fluorescein Griffonia simplicifolia lectin, isolectin B4 (Vector Laboratories, Burlingame, CA) for 20 minutes. The slides were then washed in PBS and mounted in antifade medium with DAPI (4′,6′-diamino-2-phenylindole; Vectashield; Vector Laboratories). Photomicrographs were taken with fluorescence microscopy (AX70; Olympus, Tokyo, Japan). In the vehicle-treated rats, the nonphotocoagulated half of the retina underwent immunohistochemical analysis. 

AO Digital Fluorography.

AO digital fluorography was performed as previously described. ^{19 20} With this technique, a scanning laser ophthalmoscope (Rodenstock Instruments, Munich, Germany), coupled with a computer-assisted image analysis system obtains continuous high-resolution images of the fundus stained with metachromatic fluorochrome (AO; 0.1% solution in saline; Wako Pure Chemical, Osaka, Japan), which emits a green fluorescence when interacting with DNA. The spectral properties of AO–DNA complexes are very similar to those of sodium fluorescein, with an excitation maximum at 502 nm and an emission maximum at 522 nm). An argon blue laser, with a regular emission filter for fluorescein angiography, was used as the illumination source. Immediately after the AO solution was infused intravenously, the leukocytes were stained selectively among the circulating blood cells. The nuclei of vascular endothelial cells were also stained. AO easily infiltrates through vessel walls and diffuses into the retina because of its membrane permeability. Accordingly, a few minutes after AO injection, the fluorescence of the circulating leukocytes was diminished by washout. In contrast, the leukocytes that had been trapped in the retinal microcirculation remained fluorescent for approximately 2 hours and were recognized as distinct fluorescent dots 30 minutes after AO injection. The obtained images were recorded on digital videotape at a rate of 30 frames/second for further analysis. 

Immediately before AO digital fluorography, the rats were anesthetized and the pupils were dilated. A contact lens was placed on the cornea, to maintain corneal clarity throughout the experiment. Each rat had a catheter inserted into the tail vein and was placed on a stereotaxic platform. Body temperature was maintained at 38 ± 0.5°C throughout the experiments. AO (0.1% solution in saline) was injected continuously through the catheter for 1 minute at a rate of 1 mL/min. The fundus was observed with a scanning laser ophthalmoscope in a 40° field for 5 minutes. Thirty minutes after the injection of AO, the fundus was observed again to determine leukocyte accumulation in the retinal microcirculation. 

Twenty-four hours after laser photocoagulation, AO digital fluorography was performed in the control, vehicle-treated, and TA-treated rats. Six different rats were used in each group. After the experiment was completed, the rats were killed with an anesthetic overdose, and the eyes were enucleated to determine a calibration factor to convert values measured on a computer monitor (in pixels) into real values (in micrometers). 

Image Analysis.

The digital video recordings were analyzed with an image-analysis system described in detail elsewhere, ^{2 19 20} with a slight modification. In brief, we used a computer equipped with software (DVgate; SONY, Tokyo, Japan) that stores the digital images (640 horizontal and 480 vertical pixels) in real time (30 frames/second) with an intensity resolution of 256 steps into a personal computer. Using this system, we evaluated rolling leukocytes along major retinal veins and the number of leukocytes that accumulated in the retinal microcirculation. In the vehicle-treated rats and the TA-treated rats, the nonphotocoagulated half of the retina underwent image analysis. All image analyses were performed in a masked fashion. 

Rolling leukocytes are defined as leukocytes that move much slower than free-flowing leukocytes. The process of differentiating rolling leukocytes from free-flowing leukocytes has been described in a previous article. ² In brief, leukocytes rolling along major retinal veins were easily recognized on a video monitor, because even the fastest rolling leukocytes moved almost 300 times more slowly than the average for free-flowing leukocytes. Because no leukocytes with an intermediate velocity were observed, it was not difficult to distinguish rolling leukocytes from free-flowing leukocytes. Their number was calculated from the number of rolling cells passing a fixed line in all major veins (4–7 veins) at a distance of 1 disc diameter from the center of the optic disc per minute. The average number of rolling leukocytes in individual major veins was used as the number of rolling leukocytes for each rat. 

The number of leukocytes that accumulated in retinal microcirculation was evaluated 30 minutes after AO injection. The number of fluorescent dots in the retina within four different 100-pixel-square areas at a distance of 1 disc diameter from the edge of the optic disc was counted. The average of the four areas was used as the number of accumulated leukocytes for each rat. 

The number of leukocytes infiltrating the vitreous was determined by counting the number of fluorescent dots in the vitreous within a circle with a radius of 1 disc diameter from the center of the optic disc, 30 minutes after AO injection. 

The diameters of major retinal vessels were measured at 1 disc diameter from the center of the optic disc. Each vessel diameter was calculated as the distance between endothelial cells stained by AO on each side of the vessel. The averages of the individual arterial and venous diameters of the retina were used as the arterial and venous diameters in each rat. 

Optical Coherence Tomography.

OCT is based on low-coherence interferometry and provides high-resolution cross-sectional images of the retina. Retinal thickness was measured with OCT (OCT3; Carl Zeiss Meditec, Inc., Dublin, CA), as described in detail elsewhere, ²¹ with a slight modification. After maximal papillary dilatation, anesthetized rats were mounted in a head holder. The optic disc was placed in the center of the OCT image, and the scanning line was aligned to pass through both the inner and outer canthi. The scan length was 5.0 mm in all cases. Retinal thickness was measured by OCT at 1-disc diameters from the optic disc margin in the peripheral retina with an accessory program of the OCT instrument in the control, the vehicle-treated (24, 72, and 168 hours after PRP), and the TA-treated (24 and 72 hours after PRP) rats. The mean retinal thickness of one eye was defined as the average of the three measurements. Six different rats were used at each time point in each group. In the vehicle-treated rats and the TA-treated rats, retinal thickness was measured in the nonphotocoagulated half of the retina. All image analyses were performed in a masked fashion. 

Statistical Analysis

All values are expressed as the mean ± SD. Data were compared by one-way analysis of variance (ANOVA). Post hoc comparisons were tested by using the Bonferroni procedure. P < 0.05 was considered to be statistically significant in all statistical analyses. 

Results

Fundus Image after Panretinal Photocoagulation

A representative fundus image after panretinal photocoagulation is shown in Figure 1 . Scattered laser burns were placed in half of the retina, and the other half was left untreated. 

Immunohistochemistry

Twenty-four hours after PRP, immunostaining for isolectin B4 and ICAM-1 was performed in retinal specimens from the control and vehicle-treated rats. On sections incubated with isolectin B4 antibody, immunoreactivity was present predominantly in the vascular endothelium of the retinas of all the rats. Intense ICAM-1 immunoreactivity was present in the venous endothelium of both the photocoagulated and nonphotocoagulated sides of the retina in the vehicle-treated rats, whereas only faint immunoreactivity was seen in the retinas of the control rats (Fig. 2) . 

Leukocyte Rolling

Immediately after AO had been infused, many free-flowing leukocytes were visualized. No rolling leukocytes were observed along the major retinal veins in the control rats. In the vehicle-treated rats, some rolling leukocytes were observed along the major retinal veins, making intermittent adhesive contact with the vascular endothelial cells among the many free-flowing leukocytes (Fig. 3A) . The number of rolling leukocytes in the vehicle-treated rats was 14.7 ± 4.3 cells/minute 24 hours after photocoagulation. No rolling leukocytes were observed along any major retinal arteries throughout the experiments. The number of rolling leukocytes in the TA-treated rats was 5.0 ± 2.5 cells/min and decreased by 66% (P < 0.01) when compared with the number in the vehicle-treated rats (Fig. 3) . 

Leukocyte Accumulation in Retinal Microcirculation and Vitreous

Leukocytes that had been trapped in the retinal microcirculation remained fluorescent for approximately 2 hours and were recognized as distinct fluorescent dots 30 minutes after AO injection (Figs. 4A 4B) . Very few leukocytes accumulated in the retinal microcirculation of the control rats. PRP induced substantial leukocyte accumulation in the untreated half of the retina (Fig. 4A) . On the other hand, a significant reduction in leukocyte accumulation was seen in TA-treated rats in the untreated half of the retina (Fig. 4B) . The number of leukocytes accumulating in retinal microcirculation is shown in Figure 4C . In the retinal microcirculation of the control rats, very few leukocytes were recognized. In the eyes of the vehicle-treated rats, the number of accumulated leukocytes in the nonphotocoagulated half of the retina increased 24 hours after photocoagulation and was significantly higher than in the control rats (12.8 ± 4.8 and 40.0 ± 7.0 cells/mm², respectively; P < 0.01). After TA treatment, the number of accumulated leukocytes decreased by 34.8% (P < 0.01, compared with that in the vehicle-treated rats). The number of leukocytes infiltrating the vitreous is shown in Figure 4D . No leukocytes were observed in the vitreous of the control rats. The number of infiltrating leukocytes in the vitreous of the vehicle-treated rats increased 24 hours after photocoagulation and was significantly higher than in the control rats (21.0 ± 3.9 cells/mm²; P < 0.01). After treatment with TA, the number of leukocytes infiltrating the vitreous decreased by 54.7% (P < 0.01 compared with that in the vehicle-treated rats). 

Diameters of Major Retinal Vessels

The diameters of the major retinal vessels in rats 24 hours after photocoagulation are shown in Figure 5 . There was no significant difference in arterial diameter in the three groups (Fig. 5A) . 

In veins, vessel diameters in the vehicle-treated rats increased significantly by 125.4% 24 hours after photocoagulation (P < 0.01 compared with the control value). In the TA-treated rats, venous vasodilatation was significantly suppressed after photocoagulation compared with that in the vehicle-treated rats (P < 0.01; Fig. 5B ). 

Retinal Thickness

Changes in retinal thickness of the nonphotocoagulated half of the retina 24 hours after scatter photocoagulation are shown in Figure 6 . In the vehicle-treated rats, retinal thickness in the nonphotocoagulated half of the retina increased 24 and 72 hours after treatment (208 ± 16.1 and 212.1 ± 21.6 μm, respectively) and was significantly higher than that in the control rats (174.5 ± 10.2 μm; P < 0.01). After treatment with TA, retinal thickness was suppressed by 8.97% (P < 0.05) and 9.48% (P < 0.01), respectively, 24 and 72 hours after laser photocoagulation (Fig. 6D) . There was no significant difference in the retinal thickness in the control and vehicle-treated rats 168 hours after laser photocoagulation. 

Discussion

The exact pathogenesis of macular edema induced by retinal vascular disease has not been identified. Several mechanisms may be responsible, including an altered blood–retinal barrier (BRB) due to hemodynamic changes, alterations in capillary basement membranes, and pericyte loss. Leukocytes adhere to vascular walls and are trapped in the retinal capillaries after PRP, which may cause BRB breakdown by blocking blood flow, releasing various kinds of inflammatory cytokines. ²  

Leukocytes are vital in the defense against infections and in the healing process, but they also contribute to negative conditions such as inflammation. Leukocyte–endothelium interactions are regulated by multistep processes, ³ with each step mediated by distinct adhesion molecules. ⁴ Leukocyte rolling is the first step in a cascade of events that lead to firm adhesion and transmigration through the endothelium. Leukocyte rolling that represents mild adhesion between leukocytes and endothelial cells is induced mostly by the selectin family at an early stage; strong adhesion is then induced because of the upregulation of CDllb/CD18 and activation of ICAM-1. In this study, ICAM-1 immunoreactivity in the endothelium of retinal vein was increased in vehicle-treated rats 24 hours after PRP, even in the nonphotocoagulated retina. This indicated that the increase in leukocyte accumulation in the nonphotocoagulated retina after PRP may be mediated by the upregulation of ICAM-1 in the vascular endothelium. In this study, we used AO digital fluorography to measure leukocytes infiltrating the vitreous. In general, leukocytes infiltration is observed under severe inflammatory conditions, such as in the endotoxin-induced uveitis model, ^{22 23 24 25} indicating that PRP causes severe ocular inflammation. Moreover, an investigation using the fluorophotometric technique developed by Taguchi et al. ²⁶ has demonstrated that hydrogen peroxide diffuses into the vitreous after retinal scatter photocoagulation. Hydrogen peroxide was found to induce the upregulation of adhesion molecules on the vascular endothelium and subsequent leukocyte–endothelial interaction. ²⁷ Accordingly, scatter photocoagulation may cause inflammatory leukocyte behavior such as rolling and accumulation in the nonphotocoagulated retina through oxidative stress-induced upregulation of adhesion molecules on the vascular endothelium. 

In our study, TA injection significantly suppressed both leukocyte rolling and accumulation in the retinal microcirculation. Penfold et al. ^{28 29} have reported that TA modulated permeability and ICAM-1 expression in human choroidal endothelial cells and in a human epithelial cell line that was used as a BRB model. Mizuno et al. ³⁰ have reported that TA injection into the posterior periocular space significantly decreased P-selectin and ICAM-1 gene expression in rat retinal ischemia–reperfusion model, suggesting that TA suppressed the activation of leukocyte–endothelial interactions after ischemia. The downregulation of adhesion molecules by TA was suggested as one of the reasons for the reduction of leukocyte circulatory disturbance by TA. 

In this study, we demonstrated rat retinal thickness quantitatively by OCT. OCT revealed that retinal thickness in the nonphotocoagulated half of the retina increased both 24 and 72 hours after PRP in vehicle-treated rats. Similar to findings in other organs, in the retina, BRB breakdown has been suggested to occur as a consequence of leukocyte infiltration in various pathologic conditions. ^{31 32 33} A histologic study using a rat model of experimental autoimmune uveoretinitis demonstrated that infiltrated leukocytes cause BRB breakdown in ocular inflammation. ³¹ Miyamoto et al. ^{32 33} have reported that prevention of increased leukocyte entrapment by inhibiting ICAM-1 with its antibody reduces the vascular leakage induced by diabetes or VEGF injection. Thus, the cytotoxic properties of leukocytes accumulating in retinal microcirculation after photocoagulation may contribute to increased retinal vascular permeability in the nonphotocoagulated retina. Nonaka et al. ² have reported that accumulating leukocytes may be involved in the augmented vascular permeability in the untreated retina, resulting in retinal edema after photocoagulation. In this study, treatment of TA significantly reduced retinal thickness 24 and 72 hours after laser photocoagulation. A decrease of accumulating leukocytes by TA may have reduced subsequent retinal vascular permeability and retinal edema. 

In the present study, OCT revealed that retinal thickness in the nonphotocoagulated half of the retina returned to a normal level 168 hours after PRP in vehicle-treated rats. On the other hand, macular edema after PRP, especially in diabetic patients, is often prolonged clinically. The number of accumulating leukocytes returns to normal level 48 hours after PRP in normal rat retina. ² We speculate that vascular endothelium in normal rat retina has self-healing ability, while diabetic retinopathy has vascular leakage based on microangiopathy before PRP. 

As shown in Figure 5B , retinal vessels showed remarkable vasodilation after photocoagulation. The inflammatory stimulus upregulates inducible nitric oxide synthetase (iNOS) in leukocytes, which produces large amounts of nitric oxide, in contrast with the relatively small amounts of nitric oxide produced by constitutive nitric oxide synthetase in the vascular endothelium. ³⁴ Leukocytes accumulating after scatter photocoagulation and producing nitric oxide are likely to contribute to significant vasodilation in the major retinal veins. ² In TA-treated rats, venous vasodilatation was significantly suppressed after photocoagulation compared with that in vehicle-treated rats. Glucocorticoids are potent inhibitors of immune responses, inflammation, and endotoxic shock. One of the target enzymes of glucocorticoid inhibition is NOS II. ^{35 36 37} Inhibition of leukocyte accumulation and NO production by TA may prevent venous vasodilation. 

In conclusion, sub-Tenon administration of TA significantly suppressed both leukocyte dynamics in rat retinal microcirculation and retinal edema after PRP. Our results suggest that the suppression of leukocyte–endothelial cell interactions in the retinal microcirculation may be one mechanism responsible for the therapeutic effect of sub-Tenon TA on retinal edema after laser treatments. 

 Submitted for publication October 6, 2007; revised December 18, 2007; accepted March 14, 2008.

 Disclosure: D. Mizuno, None; A . Matsubara, None; Y. Ogura, 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.

 

Table 1.

 

Experimental Design and the Number of Animals for Each Phase of the Study

	PRP	IHC	AO	OCT (Retinal Thickness)
Control	—	3	6	6
Vehicle-treated	+	3	6	18 (6 at 24, 72, and 168 h after PRP)
TA-treated	+	—	6	12 (6 at 24 and 72 h after PRP)

Figure 1.

 

Fundus image after panretinal photocoagulation with clearly visible laser spots.

Figure 2.

 

Immunostaining of frozen sections in the nonphotocoagulated half of the retina for isolectin B4 and intercellular adhesion molecule (ICAM)-1. (A, B, C) Control rats; (D, E, F) vehicle-treated rats. Isolectin B4 is a marker of vascular endothelium. Compared with control retina, ICAM-1 immunoreactivity in retinal vein increased in vehicle-treated rats. Bar, 50 μm.

Figure 3.

 

Fundus image with AO digital fluorography 24 hours after laser photocoagulation. (A) Vehicle-treated rats. (B) TA-treated rats. Circles: rolling leukocytes. (C) The number of rolling leukocytes 24 hours after laser photocoagulation. Data are expressed as the mean ± SD. *P < 0.01 compared with vehicle-treated rats.

Figure 4.

 

Leukocyte accumulation in the untreated half of the retina 24 hours after panretinal photocoagulation in vehicle-treated rats (A) and in TA-treated rats (B). The number of leukocytes accumulating in the retina (C) and vitreous (D). Data are the mean ± SD. *P < 0.01 compared with control rats; †P < 0.01 compared with vehicle-treated rats.

Figure 5.

 

Major retinal arterial (A) and venous (B) diameters after panretinal photocoagulation. Data are expressed as the mean ± SD. *P < 0.01 compared with control rats; †P < 0.01 compared with vehicle-treated rats.

Figure 6.

 

Representative OCT images in the nonphotocoagulated half of retina 24 hours after laser photocoagulation. (A) Control rats. (B) Vehicle-treated rats. (C) TA-treated rats. (D) Time course of retinal thickness after laser photocoagulation. Data are expressed as the mean ± SD. *P < 0.01 compared with control rats; †P < 0.05 compared with vehicle-treated rats; ‡P < 0.01 compared with vehicle-treated rats.