Introduction

The influence of ionizing radiation on the eye has been studied for a long time.1, 2, 3, 4, 5, 6, 7, 8 Cataract is one of the most common complications of radiotherapy,4 and many clinical and experimental studies are available on radiation-induced cataract.5, 6, 7, 8 Optic neuropathy9 and chorioretinopathy10 related to radiation exposure have also been studied clinically. Keratopathy is another form of complication associated with ionizing radiation. Merriam et al11 reported that a dose of 6000 rad within 5–6 weeks caused corneal ulceration in three of 25 patients treated for orbital tumours. In two patients, the cornea perforated and the eye was enucleated 4–12 months after the completion of radiation therapy. Despite these clinical reports, there has been a paucity of basic research on the influence of ionizing radiation on the cornea. Most of the previous studies evaluated the corneal epithelium because of its high sensitivity to ionizing radiation.12, 13, 14, 15 Moreover, histopathological techniques were mainly used in those studies.

The specular microscope is a useful tool in analysing corneal endothelial and epithelial cells. It has been used to evaluate the cornea following refractive surgery16, 17 and ultraviolet irradiation.18 There has been no specular microscopic study, however, on the corneal morphological changes after ionizing radiation, and thus the time course of changes in the corneal epithelium and endothelium after irradiation remains unknown. The current study was aimed at investigating the long-term influence of moderate ionizing radiation on the corneal epithelial and endothelial cells.

Materials and methods

Animals and radiation exposure

Five healthy mature albino rabbits (2.5–3.0 kg) were used. The animals were used in accordance with the guidelines laid down by the ARVO statement for the Use of the Animals in Ophthalmic and Vision Research. Rabbits were anaesthetized with an intravenous injection of 1.5 g/kg urethane. The deep therapy X-ray was used for the source of ionizing radiation. The X-ray unit was operated at 200 kVp, 12 mA. After the eyelid of a rabbit was forced open by a lid retractor, one eye of each rabbit received a single dose of 20 Gy. The X-ray beam was collimated to a diameter of 15 × 20 mm through a lead collimator of 5.0 mm thickness, so as to prevent exposure of the tissue around the eye ball (Figure 1). The entire cornea including the limbus was irradiated. The contralateral eye served as a control. The rabbits were euthanized by pentobarbital sodium overdose after the study.

Figure 1
figure 1

Radiation exposure method. (a) The rabbit is placed under the X-ray unit. (b) Lead collimator is set over the rabbit eye to protect surrounding tissues.

Full size image

Examinations

The cornea was observed with a portable slit-lamp microscope with fluorescein staining. Examinations using a wide-field specular microscope (Keeler-Konan, Tokyo) were conducted before and 1, 4, 8, 12, 16, 20, 24, and 36 weeks afterwards. A drop of sodium hyaluronate was instilled on the cone of the microscope to obtain clear observations of the epithelial and endothelial cells. Specular microscopy was performed on the central cornea. Corneal thickness was also measured with the pachymeter (Keeler-Konan, Tokyo) attached to the specular microscope. Photographs of the epithelium and endothelium taken with the specular microscope were enlarged by 40 times and quantitatively analysed with a computerized cell analysis system. Cell shape was determined by digitizing the apex of each cell. A total of 40 epithelial cells were examined per eye, and the mean area of the superficial corneal epithelial cells was calculated. A total of 100 endothelial cells were examined per eye. We evaluated two parameters: mean cell area and the percentage of hexagonal cells.

Statistical analysis

Data are reported as mean±SD. Values were compared between the irradiated and the control eyes. The significance of changes was assessed with paired t-test. Differences were considered statistically significant at P<0.05.

Results

Slit-lamp biomicroscopy

Slight, transient erythema of the eyelid margin with conjunctival injection was noted in all animals 1 week following irradiation. Two of five animals exhibited cataract within 20 weeks after irradiation, and the remaining three animals developed cataract within 24 weeks. No treatment was performed on these cataracts. On slit-lamp microscopic examination, there were no apparent abnormalities in the cornea and the anterior chamber. Any signs of limbal deficiency such as pannus formation were not observed during the study.

Specular microscopy

The mean area of the superficial corneal epithelial cells before the irradiation was 468±19 μm2 (Figure 2a). The time course of change in the mean cell area after irradiation is shown in Figure 3. Enlargement of the epithelial cells was found at 4 weeks post-treatment (Figure 2b). The mean cell area in the treated group was significantly larger than that in the control group from 4 to 12 weeks after procedure (P<0.01). At 16 weeks and thereafter, there were no abnormal findings in the epithelium (Figure 2c).

Figure 2
figure 2

Specular microscopic image of epithelial cells (black bar=100 μm): (a) before irradiation, (b) large and irregular cells were observed 4 weeks after irradiation, (c) cellular morphology normalized 16 weeks after irradiation.

Full size image
Figure 3
figure 3

Mean cell area of the superficial corneal epithelium. Error bars indicate standard deviation. -•-, irradiated eye; -▴-, control eye; *P<0.01, significantly higher than the control group.

Full size image

The changes in endothelial cells following irradiation were analyzed in terms of the mean cell area and percentage of hexagonal cells. The irradiation significantly increased the mean cell area (Figure 4a) and decreased the percentage of hexagonal cells (Figure 4b). There was no significant intergroup difference 4 weeks after treatment, but values at 8 weeks and thereafter in the treatment group were significantly different from those in the control group (P<0.05). After irradiation, several giant cells were seen in the endothelium (Figure 5).

Figure 4
figure 4

Changes in endothelial cell morphology: (a) mean cell area, (b) percentage of hexagonal cells. Error bars indicate standard deviation. -•-, irradiated eye; -▴-, control eye; *P<0.05, **P<0.01: significant difference was observed between groups.

Full size image
Figure 5
figure 5

Specular microscopic image of endothelial cells (white bar=100 μm); (a) before irradiation, (b) Giant cells were seen (arrow) 36 weeks after irradiation.

Full size image

There were no statistically significant changes in corneal thickness throughout the follow-up period of this study (Figure 6).

Figure 6
figure 6

Corneal thickness after irradiation. Error bars indicate standard deviation. -•-, irradiated eye; -▴-, control eye.

Full size image

Discussion

The single dose of X-ray or gamma-ray that caused slight corneal injury has been reported to be 500–2700 rad (5–27 Gy) in experimental sutudies.12, 13, 14 Serious injury such as corneal ulcer was produced by doses over 2000 rad (20 Gy).13, 14, 15 In the present attempt to elucidate the effect of moderate ionizing radiation on the cornea, we selected a dose of 20 Gy in order to avoid serious corneal damages. In the current study, slit-lamp microscopic examinations revealed no abnormal findings in the rabbit cornea irradiated with 20 Gy of X-rays. Owing to the relatively low dose of X-irradiation administered and the usage of a lead collimator (Figure 1), damage of the lacrimal gland seemed minimal. We thought this was important because alterations in the tear secretion might have modified the physiological status of corneal structure.

Although specular microscopic analysis is an established technique in evaluating the corneal endothelium, it has not been widely used to assess the corneal epithelium. Tsubota et al19 reported that evaluation of the mean cell area is useful for detecting subtle changes in the superficial corneal epithelium. Enlargement of epithelial cells was demonstrated in diabetic aphakic patients, extended soft contact lens wearers, and aphakic patients who used 0.5% indomethacin eye drops.20 Such conditions may influence the metabolism of epithelial cells, resulting in alteration of differentiation and mitosis of epithelial cells. In contrast, small epithelial cells were commonly observed in patients with dry eye.21

The mechanism of the cell injury caused by ionizing radiation has not been fully elucidated. Both lethal and sublethal damages can occur in the cells,22 and sublethal damages are repairable.23 A delay in cell division and a suppression of mitosis have been confirmed in the basal cells of the corneal epithelium after sublethal damages.24 Specular microscopic examination in the present study revealed significant increases in the mean cell area of the superficial corneal epithelium from 4 weeks after irradiation, but changes were transient. At 16 weeks and thereafter, there were no abnormal findings in the epithelium. These results indicate that the corneal epithelial injury observed herein represent sublethal damage of the basal cells in corneal epithelium. In addition, because any signs of limbal deficiency were not observed during the study, we considered that stem cell damage of the cornea was minimal under the condition of X-irradiation used in this experiment. The sensitivity of tissues to irradiation depends on the turnover rate of the cells.25 The corneal epithelial cells possess very high division potential.26, 27 It is not surprising that corneal epithelium showed early response to X-irradiation, and changes were practically reversible. The recovery of corneal epithelium is expected unless stem cells are damaged completely.28

The current study represents the first in vivo, longitudinal investigation on the corneal endothelial injury after X-ray irradiation. The results showed that there were no significant morphological changes in the endothelium immediately after irradiation, presumably due to the low turnover rate of corneal endothelial cells. The turnover rate of vascular endothelial cells is similarly low,29 and it has been reported that these cells also show delayed responses to X-ray irradiation.30 Calvo et al30 reported that the structural abnormality involving damaged nucleus and loss of endothelial cells in the choroid plexus of rat brain developed 12 weeks after a single dose of X-ray irradiation. In the current study, increases in the mean cell area and decreases in the percentage of hexagonal cells were observed at 8 weeks post-treatment. These endothelial changes persisted, or even progressed throughout the 36-week study period. De Gowin et al31 reported that X-irradiation inhibited mitosis of human vascular endothelial cells in culture. Yao demonstrated that X-ray induced destruction of chromosome in the cultured corneal endothelial cell of rat kangaroo.32 It seems that X-ray exposure in the present study inhibited cell division and/or mitosis of rabbit endothelium, leading to progressive and irreversible changes in the endothelial cellular morphology. On the other hand, corneal thickness remained unchanged throughout the experiment, suggesting that the corneal endothelial function in the irradiated eyes was maintained during the current observation period. In comparison to humans, rabbits hold greater healing potential of the corneal endothelium, including migration, elongation, and proliferation of the adjacent cells.33 Therefore, functional capacity after endothelial trauma can be much larger in rabbits, and thus the results obtained herein may not directly apply to human eyes. Moreover, the primary anatomical event after ionizing radiation was reported to be endothelial cell damage, followed by alterations of the surrounding tissues in the retina and optic nerve.34, 35, 36 Further studies will be needed to discuss the long-term structural integrity of the cornea after ionizing radiation.