To evaluate the 1-year results of keratoconic eyes with thin corneas that were treated by a hypo-osmolar riboflavin solution and ultraviolet A collagen cross-linking (CXL).
Retrospective, nonrandomized study.
setting: Department of Ophthalmology, Carl Gustav Carus University Hospital, Dresden, Germany. study population: Thirty-two eyes of 29 patients with progressive keratoconus and a corneal thickness of less than 400 μm (without the epithelium). intervention: Application of a hypo-osmolar riboflavin solution to the cornea after its de-epithelization followed by ultraviolet A collagen cross-linking. main outcome measures: Thirty-two eyes with a follow-up within 1 year were evaluated before and after the procedure. Examinations comprised an evaluation of visual acuity, corneal topography, slit-lamp microscopy, and corneal thickness measurements.
Before surgery, the mean corneal thickness (with the epithelium) was 382.3 ± 41.9 μm, and after the removal of epithelium, the thickness of the cornea was reduced to 337.0 ± 51.9 μm. After the application of the hypo-osmolar riboflavin solution, this value increased to 451.8 ± 46.7 μm. Before surgery, the mean K-value of the apex of the keratoconus was 65.6 ± 11.2 diopters, and 1 year after treatment, this value remained unchanged at 64.9 ± 11.0 diopters ( P = .839). Mean best-corrected visual acuity at the time of the treatment was 0.63 ± 0.37 logarithm of the minimal angle of resolution, and 1 year after the treatment, this value was not statistically different (0.59 ± 0.42 logarithm of the minimal angle of resolution; P = .662). At the last follow-up examination, which was 1 year after the procedure, all corneas were transparent, without any scarring lesions in the stroma.
The results of our study, using hypo-osmolar riboflavin solution in a cross-linking procedure of thin corneas, show a stability of keratoconus one year after cross-linking. Application of the hypo-osmolar riboflavin solution preserved cross-linked corneas from developing stromal scars.
Corneal collagen cross-linking (CXL) with riboflavin and ultraviolet A (UVA) light has been used for several years as a treatment option for progressing keratoconus to increase the biomechanical and biochemical stability of the corneal tissue. This is accomplished by inducing cross-links within and between collagen fibers. Currently, this procedure is the only semisurgical therapeutic approach in patients with progressing keratoconus. It has been shown to delay or even stop corneal ectasia, thus reducing the necessity of keratoplasty. The confocal microscopic findings of the cornea after the procedure showed modifications of the corneal microstructure. This revealed increased cross-link formations, syntheses of well-structured collagen, and new fibrillar interconnections.
At Carl Gustav Carus University Hospital’s Department of Ophthalmology in Dresden, Germany, corneal CXL in patients with progressive keratoconus has been performed in more than 600 eyes since 1998. The therapeutic potential of the method to treat the progressing primary corneal ectasia was demonstrated in a clinical pilot study, an uncontrolled, retrospective study, and controlled prospective studies. With the increasing number of performed CXL procedures, the number of recorded temporary complications increases as well. These complications fall into 3 categories: complications (persistent epithelial defects, infectious keratitis, and noninfectious keratitis), permanent complications (developing of stromal scars), and disastrous complications (melting of corneas ending with corneal perforations).
In our last article, we reported that in patients with thin corneas, a permanent stromal scar tends to develop after a CXL procedure. All these patients were treated with iso-osmolar riboflavin solution. In the current study, we analyzed the tendency for stromal scar development of the cornea in the patients with thin corneas who were treated with hypo-osmolar riboflavin solution.
In the current study, patients with progressive keratoconus and a corneal thickness of less than 400 μm (without epithelium) were included. An increase in the maximum topographic K-value and a reduction of corneal thickness with or without changes in uncorrected visual acuity and best-corrected visual acuity (BCVA) within the last year were considered as indicators of progression.
Corneal CXL was performed in the outpatient service area. After topical anesthesia using proxymetacaine hydrochloride 0.5% eye drops, 9 mm of epithelial tissue was removed. De-epithelialization was followed by measuring the corneal thickness with an ultrasound device (Tomey SP-3000; Nishi-ku, Nagoya, Japan) to validate that the thickness was less than 400 μm. As a photosensitizer, 0.1% hypo-osmolar riboflavin solution was applied to the cornea every 2 minutes for 30 minutes before the irradiation. The application of riboflavin was followed by another measurement of corneal thickness at the thinnest point to confirm that its value was more than 400 μm. UVA of 370-nm wavelength was used for the irradiation. A slit-lamp examination was performed to ascertain that the riboflavin penetrated the cornea. An 8-mm diameter of the central cornea was irradiated with an irradiance of 3 mW/cm 2 (UV-irradiation device, UV-X; Company Peschke, Nürnberg, Germany). During the 30 minutes of irradiation, drops of hypo-osmolar riboflavin solution were applied to the cornea every 2 minutes to sustain the necessary concentration of the solution and to avoid any desiccation of the cornea.
After CXL, a soft therapeutic contact lens, antibiotic drops, and artificial tears were applied. Analgesics also were prescribed. The contact lens was applied until complete re-epithelialization of the cornea was achieved. To prevent epithelial healing disturbances and corneal infection, topical steroids were prescribed only after complete epithelial closure and were applied for 3 weeks.
Follow-up examinations were performed daily until re-epithelialization was complete and then again at 1, 6, and 12 months. During each examination, refraction, BCVA with glasses or contact lenses, corneal topography, slit-lamp microscopy, and corneal thickness were recorded.
To quantify the corneal CXL effect, the maximum K-value at the apex and the maximum K-values in the 3-mm zone of corneal topography were recorded. The changes were estimated by a subtraction of each parameter at the respective follow-up examination (minus that at the day of corneal CXL). Statistical evaluation of values before corneal CXL and at 12 months after corneal CXL was performed using the Student t test with SPSS software version 17 (SPSS GmbH Software, Munich, Germany).
The analysis included 32 eyes from 29 patients (20 males and 9 females) with a mean age of 27.4 ± 9.4 years and stage 1 to 3 keratoconus, according to the Krumeich classification. All eyes had transparent corneas before the procedure and had preoperative Vogt striae of different intensity and number.
Before surgery, the mean corneal thickness, including the epithelium, was 382.3 ± 41.9 μm, and the thickness was reduced to 337.0 ± 51.9 μm after the removal of the epithelium. Finally, after the application of hypo-osmolar riboflavin solution, the thickness increased to 451.8 ± 46.7 μm ( Figure 1 ).
Before surgery, the mean K-value from the apex of the keratoconus was 65.6 ± 11.2 diopters; 1 year after treatment, this value was maintained at 64.9 ± 11.0 diopters ( P = .839; Figure 2 ).
Mean BCVA at the time of the treatment was 0.63 ± 0.37 logarithm of the minimal angle of resolution; 1 year after treatment, this value was not statistically different (0.59 ± 0.42 logarithm of the minimal angle of resolution; P = .662; Figure 3 ).