The use of synthetic inhibitors of
metalloproteinases (
SIMP) or
medroxyprogesterone (MP) can prevent or significantly delay the ulceration of
alkali-injured corneas by influencing
collagen degradation. We have examined the remodeling of rabbit corneal stroma following
alkali injury and have assessed the effect of
SIMP and MP treatment. Following a defined
alkali injury to the rabbit cornea, animals were divided into three subgroups, one group treated with topical beta-mercaptomethyl tripeptide (
SIMP), one treated by subconjunctival injection of MP and one treated with a control
solution. The corneal tissue was taken at 3 days, 1, 2, 3, 4, 9 and 26 weeks after
alkali injury and prepared for light microscopy and transmission electron microscopy (TEM). A quantitative measurement of birefringence, in terms of the optical path difference (OPD), was made using a modified polarized microscopy technique based on the analysis of interference colours. The results showed that
SIMP effectively prevented deep corneal ulceration. MP could delay the ulceration and the corneas treated with MP appeared to have better transparency than the other groups. There was a significant difference of the OPD between the anterior (5.8 +/-0.3 nm) and posterior (7.8 +/-0.4 nm) stroma of the normal cornea (P<0.001). The OPD values from the central corneas from
alkali-injured eyes were generally lower than normal during the first 4 weeks and then gradually recovered to the normal level or above, except for the posterior stroma of the MP-treated eyes. We found that the OPD changes were very dependent on the presence of corneal lesions. The stroma near corneal ulceration,
scar tissue, calcified stroma and the retro-corneal
collagen layer showed a significant reduction of birefringence (lower OPD values). These OPD values remained much lower than normal up to the end of the experiment. TEM showed disrupted corneal stroma in all three groups, with thinner
scar tissue in the MP group. The fibril diameters did not change significantly 3 days and 1 week after the
alkali burns (27.1+/-2.3 nm in the control group, 27.3+/-2.2 nm in the
SIMP group and 27.7+/-2.1 nm in the MP group) and there were no differences compared with 29.7+/-1.7 nm of the normal cornea (P>0.05). After 2 weeks of tissue remodeling, the fibril diameters in
alkali-injured corneas showed a large variation (the range was between 11.5 and 80 nm) with a bimodal distribution, especially in the control group. The technique presented here for birefringence evaluation can provide an alternative way to monitor wound healing and tissue remodeling, both visually and quantitatively.