SEARCH

SEARCH BY CITATION

Keywords:

  • alkali injury;
  • cornea;
  • inflammatory cytokines;
  • betamethasone;
  • cyclosporin A

Abstract.

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Purpose:  To evaluate the effects of early systemic administration of betamethasone or cyclosporin A (CsA) on inflammatory cytokine expression and corneal damage after alkali injury.

Methods:  Sixty-five Wistar rat corneas injured with 1N NaOH were divided into three groups: untreated, betamethasone-treated and CsA-treated. Both agents were administered systemically and daily during the first 7 days after injury. Interleukin (IL)-1α, -6 and -8 concentrations in the injured corneas were measured with ELISA at 2, 4 and 7 days after injury. Corneal damage was evaluated by scoring clinical findings.

Results:  In untreated injured corneas, IL-1α, -6 and -8 were markedly elevated during the 7-day period following injury. Both betamethasone and CsA significantly suppressed IL-1α and IL-8 at day 4. Only betamethasone significantly suppressed IL-6 at days 4 and 7. Both agents significantly reduced corneal opacity at day 4.

Conclusion:  Early systemic administration of betamethasone or CsA after alkali injury may be of benefit by suppressing inflammatory cytokine expression in the cornea.


Introduction

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Alkali injury to the eye is a serious ocular disease that can permanently affect visual function. It immediately induces severe inflammatory reaction and destroys ocular tissues, causing persistent epithelial defects and stromal melting. Acute inflammatory reaction often prolongs to the late phase, resulting in an opaque cornea with scarred stromal and conjunctivalized corneal surfaces. It is well recognized among clinicians, especially corneal specialists, that adequate management of acute inflammation after alkali injury is essential to a positive visual prognosis.

A number of therapies have been reported for alkali injury, including ascorbate (Levinson et al. 1976; Pfister & Paterson 1980; Pfister et al. 1981, 1982), citrate (Pfister 1983; Haddox et al. 1989; Pfister et al. 1991), fibronectin (Phan et al. 1991), epidermal growth factor (EGF) (Gonul et al. 1995) and matrix metalloproteinase inhibitors (MMPIs) (Schultz et al. 1992; Wentworth et al. 1992; Paterson et al. 1994; Pfister et al. 1997; Sotozono et al. 1999). However, none of these have proven to be of definite therapeutic value.

Steroids are well known for their anti-inflammatory effects. However, there is a wide divergence of opinion regarding steroid usage in alkali injury. There are clinical reviews that recommend steroid use, but the proposed terms of use either vary (from 1 to 3 weeks) or are not mentioned precisely (Horwitz 1966; Leibowitz 1980; Wagoner 1997; Brodovsky et al. 2000). Some animal studies have evidenced steroid efficacy; however, assessment was based on clinical corneal manifestation and did not refer to the mechanism (Donshik et al. 1978). On the other hand, some reports describe the disadvantages of steroids for alkali injury. Brown et al. (1970) indicated that collagenase activation by steroids accelerates corneal melting. Chung et al. (1998) showed that steroids retard corneal epithelial healing. Steroid therapy for alkali injury thus remains controversial.

Recently, we demonstrated that interleukin (IL)-1α and IL-6 were up-regulated in the early stages (especially the first 7 days) after alkali injury in mice (Sotozono et al. 1997). Interleukin-1 and IL-6 are involved in acute corneal inflammation, and are thought to play essential roles in controlling corneal tissue remodelling after alkali injury. However, overexpression of these inflammatory cytokines can contribute to undesirable damage processes, such as cell infiltration, corneal melting and neovascularization (Sotozono 2000). It is thought that suppressing initial cytokine overexpression may lead to successful management after alkali injury. We therefore hypothesized that consideration of inflammatory cytokine responses may help to determine how to use steroids and that steroid administration in the early stages after alkali injury may suppress overexpression of inflammatory cytokines, leading to a reduction of corneal damage. To study this hypothesis, we investigated corneal cytokine level changes induced by early systemic steroid administration and followed corneal damage using the rat alkali injury model. To our knowledge, there have been no studies investigating steroid effects on inflammatory cytokine expression, or on clinical appearance, after alkali injury in vivo.

In addition, we studied the effects of cyclosporin A (CsA). We frequently encounter patients who cannot receive steroids due to side-effects. In such cases, CsA can occasionally be used instead. We therefore investigated whether CsA offers therapeutic potential for alkali injury, and whether it can serve as an alternative to steroids.

In the present study, we evaluated the effects of systemically administered betamethasone and CsA on inflammatory cytokine expression and corneal damage in the early stages after alkali injury.

Material and Methods

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Animals

Ten-week-old female Wistar rats (body mass 200–220 g), purchased from Shimizu Laboratory Supplies (Kyoto, Japan), were used in this study. The animals were housed in plastic cages in a room with a 12-hour light/12-hour dark cycle. All rats were free from ocular disease, and were handled in accordance with the ARVD Statement for the Use of Animals in Ophthalmic and Vision Research.

Alkali injury and treatment groups

The animals were operated on under general and topical anaesthesia. A filter paper disc 5 mm in diameter, soaked with 1N NaOH, was placed on the centre of each right cornea for 1 min in 65 rats. The corneas were then washed with sterile saline solution for 1 min. The rats were then divided into three groups: group A (n = 22) was not treated with any drugs and served as a control group; group B (n = 21) received one daily subcutaneous injection of 0.1 mg/kg betamethasone; group C (n = 22) was treated with 5 mg/kg CsA once daily by intramuscular injection. Administration of each drug commenced immediately after corneal washing and continued for 7 days. Six uninjured normal corneas of six rats were used to determine the normal levels of each cytokine.

Cytokine quantitation

Interleukin-1α -6 and-8 in the cornea were quantitated at 2, 4 and 7 days after injury. Cytokine expression levels were elucidated using enzyme-linked immunosorbent assay (ELISA) systems for IL-1α (Amersham Pharmacia Biotech UK Ltd, Buckinghamshire, UK), IL-6 (TFB Inc., Tokyo, Japan) and IL-8 (Immuno-Biological Laboratories Co. Ltd, Gunma, Japan), as described previously (Sotozono et al. 1997, 1999; Sano et al. 1998). Briefly, after the eyes were enucleated, the corneas were carefully excised and frozen in liquid nitrogen, then smashed and homogenized in phosphate-buffered saline (PBS) at a ratio of 100  per cornea. Supernatants were collected by centrifugation at 1500 g for 10 min and stored at − 80 ° until use.

Clinical score

The eyes were examined daily with slit-lamp microscopy. At days 2, 4 and 7 after injury, masked observers assigned each cornea a score for corneal opacity, neovascularization (NV) and area of corneal defect. A scoring system was devised to describe in semiquantitative terms the extent of corneal opacity (0–4) and NV (0–4). The corneal opacity scoring system was:

0 = clear cornea;

1 = minimal superficial (non-stromal) opacity;

2 = minimal deep (stromal) opacity;

3 = moderate stromal opacity, and

4 = intense stromal opacity.

The NV scoring system was:

0 = absent;

1 = vessel length less than half the corneal radius in less than two quadrants;

2 = vessel length more than half the corneal radius in less than two quadrants;

3 = vessel length less than half the corneal radius in more than two quadrants, and

4 = vessel length more than half the corneal radius in more than two quadrants.

The area of corneal epithelial defect was counted as the ratio of defect area long axis : corneal diameter.

Statistical analysis

All data are expressed as mean ± SEM. Values below the detection limit were counted as zero pg/ml. The Mann–Whitney U-test was used to statistically compare untreated injured corneas (controls) and betamethasone-treated or CsA-treated corneas regarding cytokine levels and clinical scores at each time-point. The Mann–Whitney U-test was also used to compare cytokine levels between normal corneas and untreated injured corneas. P-values of less than 0.05 were considered significant.

Results

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Cytokine quantitation

Untreated injured corneas

In untreated injured corneas, IL-1α levels were markedly increased at 2 days after injury, compared with normal corneas. Interleukin-1α levels were still high at day 4, then diminished by day 7 (Figs 1A and 2A). Interleukin-6 was not detected in normal corneas. In untreated injured corneas, IL-6 was not detected at day 2, but increased significantly at days 4 and 7 (46.1 ± 10.3 and 55.6 ± 12.1 pg/cornea, p = 0.0163 and p = 0.0101, respectively) (Figs 1B and 2B). Interleukin-8 concentration in normal corneas was 35.4 ± 10.4 pg/cornea. In untreated injured corneas, IL-8 was significantly increased at 2 days after injury (126.4 ± 31.2 pg/cornea, p = 0.0332), reaching a concentration of 357.6 ± 49.0 pg/cornea at day 7 (p < 0.0027) (Figs 1C and 2C).

image

Figure 1. Effects of betamethasone on cytokine expression during the first 7 days after rat cornea alkali injury: (A) IL-1α; (Β) IL-6; (C) IL-8. Two µl of 1N NaOH was applied for 1 min to the corneas of Wistar rats. Untreated injured corneas were used as controls. Uninjured corneas were used to confirm normal cytokine levels. Symbol < indicates below assay sensitivity level (8 pg/cornea). Each column and error bar shows mean value and standard error, respectively. Asterisks show statistical significance of difference as compared with controls (*: p < 0.05) (Mann–Whitney U-test).

Download figure to PowerPoint

image

Figure 2. Effects of CsA on cytokine expression during the first 7 days after rat cornea alkali injury: (A) IL-1α; (Β) IL-6; (C) IL-8. Two µl of 1N NaOH was applied for 1 min to the corneas of Wistar rats. Untreated injured corneas were used as controls. Uninjured corneas were used to confirm normal cytokine levels. Symbol < indicates below assay sensitivity level (8 pg/cornea). Each column and error bar shows mean value and standard error, respectively. Asterisks show statistical significance of difference as compared with controls (*: p < 0.05) (Mann–Whitney U-test).

Download figure to PowerPoint

Treated corneas

In the betamethasone-treated group, IL-1α levels were 108.1 ± 18.5 pg/cornea, significantly lower than in the control group at day 4 (p = 0.0039) (Fig. 1A). Interleukin-6 levels were significantly lower than in controls at days 4 and 7 (< 8 pg/cornea, and 13.0 ± 8.6 pg/cornea, p = 0.0163 and p = 0.0279, respectively) (Fig. 1B). Interleukin-8 levels were 106.9 ±  15.8 pg/cornea, significantly lower than in controls at day 4 (p = 0.0039) (Fig. 1C). In the CsA-treated group, IL-1α levels were 247.5 ± 55.2 pg/cornea, significantly lower than in controls at day 4 (p = 0.0374) (Fig. 2A). Interleukin-8 levels were 175.8 ± 26.0 pg/cornea, significantly lower than in controls at day 4 (p = 0.0374) (Fig. 2C).

Clinical score

Corneal opacity in both betamethasone-treated and CsA-treated groups was significantly less than in the control group at 4 days after injury (p = 0.0163 and p = 0.0082, respectively) (Fig. 3A). There was no significant difference between the treated groups and the control group in scores for NV or corneal defect (Fig. 3B, C).

image

Figure 3. Effects of betamethasone and CsA on clinical scores during the first 7 days after rat cornea alkali injury: (A) corneal opacity; (B) neovascularization (NV); (C) corneal epithelial defect. Untreated injured corneas were used as controls. The same eyes as for cytokine assay were used to assign clinical scores. Symbol < indicates no NV. Each column and error bar shows mean value and standard error, respectively. Asterisks show statistical significance of difference as compared with controls (*: p < 0.05) (Mann–Whitney U-test).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

The present study demonstrated that levels of IL-1α, -6 and -8 in untreated injured corneas increased markedly during the 7-day period following alkali injury. These results were in good accordance with our previous report using the mice alkali injury model (Sotozono et al. 1997). The present study shows that early up-regulation of inflammatory cytokines in the alkali-injured cornea occurs similarly in mice and rats.

Interleukin-1, the initial cytokine induced by inflamed injury, acts as a trigger for expressing other inflammatory cytokines, including IL-6 and IL-8 (Kishimoto 1989; Elner et al. 1991; Cubitt et al. 1993, 1995). It is considered that these cytokines are involved in acute corneal inflammatory reaction, and are responsible for corneal tissue remodelling and destruction (Sotozono et al. 1997). It has been reported previously that secondary damage occurs in the corneal alkali wound after the first week (Chung et al. 1987; Chung & Fagerholm 1989). We have also proposed the concept of secondary injury in the cornea induced by inflammatory cytokine overexpression after initial injury (Sotozono 2000). Once secondary injury is induced, a vicious cycle may occur between cytokine production and corneal damage, and may contribute to undesirable corneal destructive processes. Therefore, the regulation of inflammatory cytokine expression in the initial stages is considered essential for successful management after alkali injury. Recently, Yamada et al. (2003) reported that early administration of topical IL-1 receptor antagonist (IL-1ra) after alkali injury could suppress corneal inflammation and promote corneal transparency, indicating the usefulness of anti-inflammatory treatment with reference to cytokine response after injury.

In this study the steroid administration significantly suppressed the expression of IL-1α, -6 and -8 during a 7-day period following injury, as compared with untreated corneas. The decreased corneal opacity at day 4 in the steroid-treated group may support these results. Steroid therapy in the management of alkali injury remains controversial. However, on the evidence of our investigation, we consider early steroid administration to be useful in suppressing the initial overexpression of inflammatory cytokines. It should be emphasized that steroid administration is effective in the stage of cytokine elevation, that is, in the early and short-term stage after injury.

When steroids are used only in the short-term there is little in the way of adverse effects such as delayed re-epithelialization and stromal melting. Our results in terms of clinical scores confirmed this; the corneal defect area in the steroid-treated group showed no tendency to enlarge at day 7, in comparison with the untreated group. In addition, the corneal defect was healed by day 14 (data not shown) in all three groups. We feel that consideration of cytokine responses may be a reasonable and informative approach to determining the usage of steroids.

In humans, the levels and duration of cytokine up-regulation after alkali injury are not fully understood. Knowledge of cytokine dynamics on the ocular surface may be needed to determine the dose and period of steroid administration. Nakamura et al. (1998) reported that IL-8 was measurable in the basal tears of the normal eye: the concentration was 731.4 ± 116.2 pg/ml. Moreover, in our preliminary investigation, a surprisingly high concentration of IL-8, about 30 times more than normal, was detected in the basal tears of alkali-burned eyes (data not shown). Thus, cytokine measurement in tears may offer potential for cytokine monitoring.

The present study showed that CsA administration also significantly suppressed IL-1α and -8 expression and decreased corneal opacity during the first 7-day period. Intriguingly, recent studies have indicated that CsA has both down-regulatory and up-regulatory effects on inflammatory cytokines. Turner et al. (2000) showed a decrease in IL-6 in the conjunctival epithelium of moderate to severe dry eye patients treated with topical CsA. On the other hand, CsA has been shown to increase inflammatory cytokines in vitro, for example, IL-6 and -8 in conjunctival fibroblast or IL-1β and -8 in human orbital fibroblasts (Burnstine et al. 1998; Leonardi et al. 2001). In our results, CsA appears to suppress inflammatory cytokine expression in vivo. We therefore suggest that CsA may serve as an alternative to steroids in the treatment of alkali injury in patients contraindicated for steroids.

Interestingly, retinal inflammatory response via inflammatory cytokines may be involved in the damage process of alkali injury (Miyamoto et al. 1998). Yoshida et al. (2001) reported that IL-8 in human retinal pigment epithelial cells induced by mechanical injury may be involved in retinal inflammation, and that dexamethasone and CsA inhibited lL-8 expression, although their study was in vitro. To investigate the effect of betamethasone and CsA on retinal inflammation in vivo, we also examined changes in retinal cytokine expression in the present model, clarifying the suppressive effects of betamethasone and CsA on IL-6 and IL-8 expression in the early stages after injury (Den et al. 2001). The results of those studies suggest that steroid or CsA therapy may suppress ocular inflammation not only in part of the anterior segment, but also in the retina, leading to a reduction in retinal dysfunction. Therefore, systemic administration seems to be more beneficial than topical administration.

In conclusion, alkali injury to the eye is characterized by severe inflammatory ocular disease. Ocular inflammation is an important factor influencing the clinical outcome following such injury. Control of acute inflammatory reaction is the key to successful management of alkali injury. We suggest that steroid or CsA administration in the early stages after alkali injury may be beneficial as therapy, through the suppression of inflammatory cytokine expression. It may, therefore, be helpful to consider inflammatory cytokine dynamics on the ocular surface in order to determine the usage of these anti-inflammatory agents and to achieve adequate treatment of alkali injury. We expect that cytokine monitoring will become a reality in the future.

Acknowledgement

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

This work was supported in part by a grant-in-aid for scientific research (10877275) from the Japanese Ministry of Education, Culture, Sports, Science and Technology.

References

  1. Top of page
  2. Abstract.
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References