Corneal Cross-linking and Orthokeratology
This pilot study investigated the use of riboflavin corneal cross-linking and orthokeratology to treat myopia.
By Sami El Hage OD, PhD, DSc, FAAO, & Theo Seiler MD, PhD
Spectacles, contact lenses and surgery are commonly used to correct most refractive errors; however, there is growing interest in the nonsurgical technique of orthokeratology to correct or improve visual acuity by temporarily reshaping the cornea. This effect is achieved when patients wear reverse-geometry lenses for a specified time overnight while sleeping, and the effect is maintained after lens removal during waking hours. In some cases, patients may wear the lenses one or two nights a week, while others must wear them every night to maintain the desired corneal shape and functional vision during waking hours.
Corneal cross-linking (CXL) with riboflavin and ultraviolet A (UVA) has been shown to strengthen corneal structures biomechanically and biochemically (Spoerl et al, 1998; Schnitzler et al, 2000; Wollensack et al, 2003) and is now widely used to treat keratoconic eyes. A comparative study of the potential damage to ocular tissue during CXL by means of riboflavin/UVA (365nm) (Spoerl et al, 2007) found that as long as the cornea has a minimum thickness of 400μm, the corneal endothelium will not be damaged nor will deeper structures, such as the lens and the retina, be damaged. The efficacy and safety of this procedure were confirmed by Mazzotta et al in 2007.
Study Materials and Methods
We conducted a pilot study of five patients adapted to overnight orthokeratology to investigate the utility of cross-linking agents and UV light to extend the orthokeratology effect.
After obtaining ethics committee approval and the patients' consents, we enrolled four women and one man, ranging in age from 18 years to 32 years. Subjects had been wearing orthokeratology lenses overnight for 12 months to 24 months. The average myopia in spherical equivalent refraction was −2.97D, and the average shape factor was 0.736.
After overnight lens wear, patients came to the clinic the next morning wearing their lenses. After removing the lenses, we measured visual acuity, performed slit lamp examination with fluorescein, refraction, and corneal topography displayed in the tangential dioptric plot (Figure 1).
After confirming good, functional visual acuity, we began the cross-linking treatment with the patient in a supine position. We instilled three drops of preservative-free tetracaine 1% solution in the cul-de-sac of the eye to be treated. After placing the lid speculum, we gently debrided an area of the epithelium 8mm in diameter to help diffuse the riboflavin throughout the cornea. We instilled a solution of riboflavin mixed with dextran (0.1% riboflavin, Peschke Meditrade) every three minutes for a total of 30 minutes. Using blue light at the slit lamp, we confirmed the riboflavin had reached the anterior chamber of the eye. UVA treatment was administered using an LED-device radiating light at 365nm with an intensity of 3mW/cm2 (UVX, Peschke Meditrade). The duration of the UVA light exposure was 30 minutes. During the exposure, the patient was instructed to look at the light source while the eye was irrigated with a combination of riboflavin solution and anesthetic (tetracaine 1%) every three minutes.
After the procedure, patients were fitted with soft contact lenses, which were used as bandage lenses to protect the eye. The patients were instructed to instill antibiotic drops (Vigamox, Alcon) four times a day for one week and to use paracetamol 500mg as needed for pain management. Follow-up visits, which included routine eye examinations, were scheduled at one day, three days, one week, one month, three and six months after the procedure. The bandages were worn for three days and the patients stopped ortho-k lens wear for 6 months.
Controlled Kerato-Reformation—CKR (Bausch + Lomb [B+L]) is part of the orthokeratology family of treatment procedures. The design of the CKR lenses is based on corneal topography measurements and shape factor (SF) instead of keratometric readings. This method uses computer-assisted video-keratography and a software program to produce a comprehensive topographical map of the patient's cornea, permitting accurate design, fitting, and monitoring of corneal changes after treatment with CKR orthokeratology contact lenses over time (El Hage et al, 2007; El Hage et al, 1995; El Hage et al, 1997; El Hage et al, 1999).
At the conclusion of this study, we observed an overall reduction in SF in the treated corneas; changes in corneal rigidity and the R factor (a numerical calculation of the rigidity of the cornea) were deduced. We related the changes in SF (or ψ), as a numerical calculation of the asphericity (Q) of the cornea, to the R factor. The SF (ψ) is related to eccentricity (e), while the R factor is a function of rigidity, flexibility, hysteresis and reverse elasticity. The combined treatment also resulted in improved corneal symmetry, improved tilt and first harmonic in Fourier analysis.
We also noted some increases in spherical aberration, as observed in most keratorefractive procedures when the shape of the cornea changes from a prolate to an oblate ellipse. Stillitano et al (2007) found that spherical aberration and coma increased after orthokeratology lens wear and regressed during the day after lens removal. In the pilot study, the correction of corneal shape and the improvement of visual acuity achieved by overnight wear of the orthokeratology lenses was extended to longer-lasting correction by cross-linking of the corrected cornea, as analyzed in patients up to three months post-treatment.
All corneal topography maps were produced using Optivision EH-300 CornealMap. Figure 1 shows Patient A's corneal topography at baseline (left) and six months after orthokeratology treatment (right). The color scale shows the relative dioptric plot, and the differences are shown (bottom). Refractive power at the center of the cornea decreased from 45.06D to 42.27D (difference of −2.80D), the radius of curvature increased from 7.49mm to 7.98mm (difference of 0.49mm) and the SF increased from 0.651 to 1.424 (difference of 0.773), changing the central cornea from the prolate to oblate side of the ellipse.
Figure 1. A difference plot of Patient A, pre- and post-CKR treatment showing vector analysis.
The side-by-side plots in Figure 2 show corneal topography before (left) and after CXL (right). Comparing the data shown in Figures 1 and 2, the changes across the combined treatment are demonstrated for Patient A. The SF increased after reshaping treatment but decreased after the CXL procedure. Figures 3 through 6 show the effect of the CXL procedure to improve corneal symmetry in this patient.
Figure 2. Side-by-side corneal topography maps of Patient A's cornea pre- and post-cross-linking treatment.
A summary plot of an asymmetrical refractive analysis of this cornea before corneal reshaping shows asymmetry, particularly when comparing the wavefront data shown on the bottom right side of Figure 3, which shows a coma (vertical) surface aberration. Fourier analysis of the pretreatment data in Figure 4 indicates that decentration, or tilt, is present. In contrast, 12 days after the CXL procedure, the same plots show more symmetry (Figure 5). While astigmatism and trefoil aberrations are present, the aberration is flipped from superior to inferior, as shown in Figures 3 and 6, and centration and symmetry were improved as shown in the comparison between Figures 4 and 5.
Figure 3. Corneal topography summary plot showing an asymmetrical refractive plot of Patient A's cornea pre-CKR treatment.
Figure 4. Corneal topography map showing Fourier analysis of Patient A's cornea pre-CKR treatment.
Figure 5. Corneal topography map showing Fourier analysis of Patient A's cornea at 12 days post-cross-linking treatment.
Figure 6. Corneal topography map showing a summary symmetrical refractive plot of Patient A's cornea 12 days post cross-linking treatment.
Changes in corneal topography after CXL are shown for Patient B in Figure 7. Side-by-side dioptric plots of the cornea are shown prior to the CXL procedure, for day 12 post-CXL and at three months.
Figure 7. Corneal topography map showing side-by-side dioptric plots of Patient B's cornea at 1 day, 12 days and 3 months post-cross-linking treatment.
These data indicate an improvement in topography with cross-linking; the treatment zone is better centered on the cornea. Changes in central keratometry readings from day one to day 12 measure from 41.27D to 43.46D, respectively, but they return to pre-CXL levels (41.43D) by three months post-CXL. Similar results for the radius of curvature occurred: 8.18mm at day one, 7.75mm at day 12 and 8.08mm at three months.
The difference plot shown in Figure 8 demonstrates an overall improvement in the irregular topography of Patient B's cornea. The central flattening zone (blue area) is more centrally located and more symmetrical.
Figure 8. Corneal topography map showing difference plots of Patient B's cornea at 1 day, 12 days and 3 months post-cross-linking treatment.
We observed an interesting effect when comparing the pre-orthokeratology corneal topography with the topography 12 days post-CXL. As shown in Figure 9, Patient C's asymmetrical astigmatism was reversed after CXL.
Figure 9. Corneal topography map showing reversal of asymmetrical astigmatism of Patient C's cornea from the pre-CKR topography to the topography present after 6 months post-CXL treatment.
In the cross-linking procedure, riboflavin drops are applied to the cornea with or without debriding the epithelium. The riboflavin in the cornea is then activated by ultraviolet light, thus creating radicals, which induce chemical reactions, most probably within or between the collagen molecules. This treatment seems to enhance the amount of collagen cross-linking in the cornea and thereby increases the biomechanical rigidity of the cornea. Our objective in this pilot study was to determine if the cross-linking of a cornea reshaped by means of orthokeratology lenses would reduce the periodicity of lens wear needed to maintain the reshaping and/or vision correction from about once daily to about once per week or once per month or longer.
In a separate prospective study of five keratoconic eyes of four patients (three women and one man) who could not tolerate conventional contact lenses, Calossi et al (2008) measured the corneal shape before and after orthokeratology/CXL. All five eyes showed an improvement in corneal shape after treatment, with a significant reduction in corneal aberration. Moreover, the investigators reported no adverse reactions during the three months of wear of orthokeratology lenses by these patients with keratoconus (unpublished data).
Table 1 summarizes our findings from a literature review of the mechanism of orthokeratology. Five groups of researchers found a decrease in corneal thickness; six groups found no change, and one group found an increase in corneal thickness with orthokeratology. The variations in the results may reflect the inherent physical limitations of the instrument used to measure corneal thickness.
|Change in Corneal Thickness in Orthokeratology|
|Carkeet et al|
|Iskeleli et al|
|Swarbrick et al|
|Fan et al|
|Nichols et al|
|Mitsui et al|
|Wang et al|
|El Hage et al|
|Fukuda et al|
|Jurkus et al|
One would expect confocal microscopy to be a more appropriate method for measuring total corneal thickness, including the epithelial layer (El Hage et al, 2007); however, accuracy may vary depending upon the microscope used. Conceivably, the difference in instrumentation could explain why we did not find a statistically significant reduction in either epithelial or total corneal thickness.
A recent study by Jurkus et al (2009) at the Illinois College of Optometry emphasized the use of confocal microscopy in measuring corneal thickness. Researchers found an increase in corneal thickness the first day after orthokeratology treatment, a significant decrease in corneal thickness one week after the procedure and a return to baseline corneal thickness after one month. In addition, the ultrasound pachometer (SonoGage) showed no significant variation in the epithelium measurements or total corneal thickness measurements. The scanning slit optical pachometer (Orbscan II-B+L) likewise did not show a statistically significant reduction in corneal thickness. The question remains, however, as to where in the cornea the orthokeratology effect takes place.
In summary, confocal microscopy, ultrasound pachometry and optical pachometry did not show statistically significant changes in corneal thickness
The changes in refractive error induced by the combined orthokeratology/CXL procedures performed in this pilot study were multifactorial. We, therefore, conclude that CXL combined with orthokeratology may be a viable treatment option for some patients and warrants further study. Some potential enhancements to this treatment to improve results should be explored, such as overcorrection prior to using riboflavin and UVA light, the use of an ortho-k lens during epithelium healing, injection of riboflavin into the cornea through a small laser incision (without debriding the cornea and wearing the contact lens after CXL), or a transepithelial procedure in which riboflavin is used without epithelial debridement.
We believe the potential is there, and CXL could play a major role in improving orthokeratology treatment results. CLS
For references, please visit www.clspectrum.com/references.asp and click on document #193..
|Dr. El Hage received his OD degree from the Pennsylvania College of Optometry, and his PhD and DSc from the University of Paris. He has taught at the University of Paris and the University of Houston College of Optometry, where he was a tenured full professor and graduate faculty.|
|Dr. Seiler studied medicine, mathematics and physics at the Universities of Heidelberg and Berlin. He is credited with being the first physician to treat a human eye using an excimer laser. He is the chairman of the Institute for Refractive and Ophthalmic Surgery in Zurich.|