The most frequent cause of decreased visual acuity after extracapsular cataract extraction (ECCE) is posterior capsule opacification (PCO). Posterior capsule opacification is caused by proliferation, migration, and metaplasia to the fibroblasts of residual lens epithelial cells (LECs).1–4 Pharmacological inhibition of PCO using antiproliferative drugs is a promising concept because the LECs that may lead to the complication cannot be sufficiently removed surgically.
In theory, antiproliferative drugs are exclusively targeted toward actively proliferating LECs. Because of the potential ocular toxicity of antiproliferative drugs, we must choose a safe and effective agent and apply it properly. Mitomycin-C is a well-known anticancer drug, and its potent antiproliferative effect has been proved in cases of pterygium and glaucoma surgery.5,6 We performed this study to evaluate whether mitomycin-C can inhibit PCO without causing ocular toxicity.
Materials and Methods
Thirty six eyes of 18 New Zealand albino rabbits weighing 2.0 to 2.5 kg were divided into 3 groups of 12 eyes each based on the substances applied during surgery. The control group received sodium hyaluronate (Healon®) only. The second (experimental) group received mitomycin-C dissolved in sodium hyaluronate (0.2 mg/mL). In the third group, mitomycin-C was dissolved in a balanced salt solution (0.2 mg/mL).
Extracapsular cataract extraction and posterior chamber intraocular lens (IOL) implantation were performed in all cases using general anesthesia (pentobarbital sodium 30 mg/kg intravenous injection and ketamine hydrochloride 25 mg/kg intramuscular injection). One surgeon (H.S.C.) performed all procedures.
The drug application method was as follows: After the lens nucleus and cortex were removed by endocapsular phacoemulsification with a small linear anterior capsulotomy (4.0 mm in length), 1 of the 3 substances (0.2 cc) was injected into the empty capsular bag. Healon was used to fill the anterior chamber to prevent leaking of the drugs from the capsular bag. Three minutes after the injection, the drug was removed completely using an irrigation/aspiration device. A posterior chamber single-piece poly(methyl methacrylate) IOL was implanted in the capsular bag. The biconvex IOL had a 7.0 mm diameter optic and 10 degree posterior haptic angulation (UNI-IOL, Universal Optics). A 20.0 diopter IOL was used in all cases.
Postoperatively, atropine 1%, tobramycin, and fluorometholone eyedrops were instilled twice daily. The animals were killed 3 months postoperatively. All procedures conformed to the Association for Research in Vision and Ophthalmology Statement for Use of Animals in Ophthalmic and Vision Research.
Two methods were used to determine the amount of PCO. One was a grading scale based on the visibility of retina by direct ophthalmoscopy in which 0 = clear; 1 = mild opacity with no disturbance on fundoscopy; 2 = moderate opacity with slight disturbance on fundoscopy; 3 = severe opacity with marked disturbance on fundoscopy. A masked observer used the scale to grade PCO 1 week and 1 and 3 months postoperatively. The second method was observation of the obstruction rate of visible light resulting from the PCO measured by an optical power meter (model 835, Newport Corp.) (Figure 1, left).
Three months postoperatively, the eyes were enucleated and fixed in formalin. All eyes were coronally dissected at the equator of the globe, and both cornea and iris were removed. The specimen, including the IOL in the capsular bag, was placed in a fixation frame. A 250 W tungsten halogen lamp with an infrared filter (IR-filter® model L5 to 371R, Sony) was used as a light source. To project the light through the center (8.0 mm in diameter) of the posterior capsule, the iris and plus spherical lens were placed between the light source and specimen (Figure 1, right). The visible light transmitted through the posterior capsule and IOL optic was measured by the optical power meter with an attenuation filter (model 883-SL, Newport). The detector (model 818-SL, Newport) can measure the power of light with wavelengths from 400 to 1100 nm.
Because infrared light was blocked with a filter, the power of the visible light (wavelengths 400 to 700 nm) was measured. The mean of 2 measurements was used for analysis. The percentage of the obstruction rate of visible light was calculated by measuring the power of the visible light before and after the transmission through the posterior capsule and IOL optic.
Anterior segment inflammation and IOL surface precipitates were evaluated 1 day, 1 week, and 1 and 3 months postoperatively using slitlamp biomicroscopy. Posterior capsule opacification and ocular tissues, including the cornea, iris, ciliary bodies, choroid, and retina, were observed by light microscopy with hematoxylin and eosin staining.
Statistical analysis was by the Kruskal–Wallis and Mann–Whitney U tests.
Three months postoperatively, the mean PCO score was 3.00 ± 0 (SD) in the control group, 1.00 ± 0 in the group receiving mitomycin-C dissolved in sodium hyaluronate, and 2.08 ± 0.52 in the group receiving mitomycin-C dissolved in a balanced salt solution.
Obstruction rate of visible light.
The power of visible light before transmission through the posterior capsule and the IOL optic was 4.6 mW. After transmission, it decreased to a mean of 0.9 ± 0.4 mW in the control group, 3.2 ± 0.5 mW in the group receiving mitomycin-C dissolved in sodium hyaluronate, and 1.3 ± 0.3 mW in the group receiving mitomycin-C dissolved in a balanced salt solution. Mean obstruction rate of visible light was 81.0% ± 8.3%, 30.5% ± 10.1%, and 71.9% ± 6.8%, respectively. Statistically significant differences were found among all 3 groups (P < .05) (Table 1).
There was no significant difference among the 3 groups in anterior segment inflammation or IOL surface precipitates.
Marked LEC proliferation and regenerated cortical lens material were observed in the capsular bags in the control group (Figure 2). These findings were also observed in the group receiving mitomycin-C dissolved in a balanced salt solution; however, the amount of material was significantly less than in the control group. There was no evidence of such material in the group receiving mitomycin-C dissolved in sodium hyaluronate (Figure 3). Light microscopy of ocular tissue showed no abnormal findings in any group.
Pharmacological inhibition of PCO by antimitotics is a promising concept. Several in vitro and experimental studies found that LEC proliferation can be inhibited by antimetabolites such as 5-fluorouracil, daunomycin, methotrexate, and colchicine.7–10 Hartmann et al.11 performed clinical studies using daunomycin in which antimitotics dissolved in solution (eg, balanced salt) were irrigated in the anterior chamber or capsular bag.
In theory, antimetabolites target actively proliferating LECs and do not have toxic effects on nonmitotic cells, such as those of the corneal endothelium, iris pigment, and retina. However, contact of these agents with ocular tissue can be toxic, especially to corneal endothelial cells.
We chose rabbits for this study because of the similarity between the rabbit and human lens (single lens epithelial cell layer in the anterior lens capsule). However, for evaluating toxicity, the rabbit eye has several limitations. We examined toxicity of mitomycin-C in the rabbit eye tissue by slitlamp and light microscopy. These examinations did not provide conclusive evidence that mitomycin-C is not toxic because (1) rabbit corneal endothelial cells can regenerate after damage, and light microscopy 3 months postoperatively may not detect early toxic damage, and (2) the rabbit eye has massive fibrin formation in the anterior chamber in the early postoperative period, making it hard to differentiate the extent of cells and flare in the anterior chamber. We do not claim that the apparent lack of toxicity observed in this experiment can be applied to the human eye. More detailed toxicity studies should precede clinical application of this study.
Shin et al.12 prospectively studied the effect of adjunctive subconjunctival mitomycin-C on PCO after primary trabeculectomy combined with cataract surgery. They found that the probability of PCO requiring neodymium:YAG capsulotomy was significantly lower in the mitomycin group than in a control group.
One study13 used an application method similar to the one we used. It found that the injection of dispase mixed with sodium hyaluronate into the empty capsular bag after endocapsular phacoemulsification can aid in the removal of LECs with negligible damage to the zonules or corneal endothelium. Dispase concentration assay in aqueous humor confirmed that the leakage of dispase from the capsular bag was slight.
In addition to applying mitomycin-C dissolved in sodium hyaluronate, we used 2 other procedures to prevent mitomycin-C leakage from the capsular bag. First, we performed endocapsular phacoemulsification and left the anterior capsule whole by creating a small linear anterior capsulotomy. Second, immediately after the injection of mitomycin-C into the empty capsular bag, we filled the anterior chamber with sodium hyaluronate.
In addition to preventing toxicity, these methods also stopped leaking, increasing the concentration of mitomycin-C in the capsular bag as well as its contact time with LECs. Therefore, the group receiving mitomycin-C dissolved in sodium hyaluronate had significantly less PCO than the group receiving mitomycin-C dissolved in a balanced salt solution. Application of mitomycin-C dissolved in sodium hyaluronate instead of a balanced salt solution boosts the efficiency of the drug by increasing its concentration in the capsular bag and contact time with LECs and makes the drug safer by preventing its contact with other ocular tissues.
In previous studies, several methods, including grading by direct ophthalmoscopy, slitlamp photography, and direct, specular, light, scanning electron, and transmission electron microscopy, were used to evaluate PCO.2,9,13,14 Because of their nonnumeric results, however, these studies lacked accuracy and objectivity. Moreover, the results could not be statistically analyzed.
The optical power meter has been widely used to measure light power in optical research and industry. We used the meter to evaluate PCO by calculating the obstruction rate of visible light. This method provides an objective and accurate quantification of PCO by yielding a percentage. In contrast, direct ophthalmoscopy provides only 4 levels of grading (0 to 3). Three months postoperatively, the PCO grade measured by direct ophthalmoscopy was 3 in all groups. In contrast, the optical power meter found values ranging from 65.2% to 93.5%. Therefore, the meter can detect subtle differences in the amount of PCO that cannot be detected using a grading scale with direct ophthalmoscopy.
Thus, the meter was able to more accurately portray the actual benefit of mitomycin-C dissolved in sodium hyaluronate by providing the obstruction rate of visible light in that group compared with that in the other 2 groups. Because we measured the visible light transmitted through the center (8.0 mm in diameter) of the PCO, we obtained clinical information that may disclose the influence of PCO on the visual prognosis.
We conclude that the application of mitomycin-C dissolved in sodium hyaluronate into an empty capsular bag in ECCE can effectively reduce PCO. More detailed studies of the minimal effective concentration of mitomycin-C dissolved in sodium hyaluronate on human LECs and the potential toxicity of the drug to human ocular tissue are needed before it can be used clinically in extracapsular surgery.
1. Fagerholm PP, Philipson BT. Formation of after-cataract by regeneration of human and rabbit lens epithelium in tissue culture. Acta Ophthalmol 1977; 55:362-371
2. McDonnell PJ, Stark WJ, Green WR. Posterior capsule opacification: a specular microscopic study. Ophthalmology 1984; 91:853-856
3. Apple DJ, Kincaid MC, Mamalis N, Olson RJ. Intraocular Lenses; Evolution, Designs, Complications, and Pathology. Baltimore, MD, Williams & Wilkins, 1989; 313-361
4. Frezzotti R, Caporossi A, Mastrangelo D, et al. Pathogenesis of posterior capsular opacification.: Part II: histopathological and in vitro culture findings. J Cataract Refract Surg 1990; 16:353-360
5. Hayasaka S, Noda S, Yamamoto Y, Setogawa T. Postoperative instillation of low-dose mitomycin C in the treatment of primary pterygium. Am J Ophthalmol 1988; 106:715-718
6. Palmer SS. Mitomycin as adjunct chemotherapy with trabeculectomy. Ophthalmology 1991; 98:317-321
7. McDonnell PJ, Krause W, Glaser BM. In vitro inhibition of lens epithelial cell proliferation and migration. Ophthalmic Surg 1988; 19:25-30
8. Chan RY, Emery JM, Kretzer F. Mitotic inhibitors in preventing posterior lens capsule opacification. In: Emery JM, Jacobson AC, eds, Current Concepts in Cataract Surgery; Selected Proceedings of the Seventh Biennial Cataract Surgical Congress. New York, NY, Appleton-Century-Crofts, 1982; 217-224
9. Ruı'z JM, Medrano M, Alió JL. Inhibition of posterior capsule opacification by 5-fluorouracil in rabbits. Ophthalmic Res 1990; 22:201-208
10. Hansen TJ, Tyndall R, Soll DB. Methotrexate-anticollagen conjugate inhibits in vitro lens cell outgrowth. Invest Ophthalmol Vis Sci 1987; 28:1206-1209
11. Hartmann C, Wiedemann P, Gothe K, et al. Prévention de la cataracte secondáire par application endocapsulaire del ántimétotique daunomycine. Ophtalmologie 1990; 4:102-106
12. Shin DH, Kim YY, Ren J, et al. Decrease of capsular opacification with adjunctive mitomycin C in combined glaucoma and cataract surgery. Ophthalmology 1998; 105:1222-1226
13. Nishi O, Nishi K, Hikida M. Removal of lens epithelial cells by dispersion with enzymatic treatment followed by aspiration. Ophthalmic Surg 1991; 22:444-450
14. McDonnell PJ, Zarbin MA, Green WR. Posterior capsule opacification in pseudophakic eyes. Ophthalmology 1983; 90:1548-1553