Empirical Advanced Orthokeratology Through Corneal Topography: the University of Houston Clinical Study

Empirical Advanced Orthokeratology Through Corneal Topography: The University of Houston Clinical Study

Sami El Hage, O.D., Ph.D., D.Sc., F.A.A.O., Norman E. Leach, O.D., M.S., F.A.A.O., William Miller, O.D., Ph.D., F.A.A.O., Thomas C. Prager, Ph.D., M.P.H., Jason Marsack, M.S., Katrina Parker, O.D., F.A.A.O., Angela Minavi, O.D., and Amber Gaume, O.D., F.A.A.O.

Purpose. Traditionally, orthokeratology has used diagnostic lenses unaided visual acuity. First described in 1962 by Jessen,1 the to determine the best fit. The purpose of this study was to orthokeratology procedure and lens designs have evolved greatly determine the efficacy of fitting empirically from corneal topog- during the last 40 years.2–39 Lenses were originally made of raphy, without the use of diagnostic lenses. Methods. Twenty-nine polymethylmethacrylate and usually fitted flatter than the flat subjects, 18 to 37 years old, with myopia of 1.00 to 4.00 diopters keratometric reading. This concept has limitations with conven- (D) and astigmatism of no more than 1.50 D, were entered into this 6-month study. Corneal topography, scanning slit topography and tional lens designs, especially when the amount of central base corneal thickness (Orbscan), confocal microscopy, ultrasound cor- curve flattening is significant. Typically, the spherical flat-fitting neal thickness, aberrometry, and biomicroscopy were used to lens would rock over the cornea and result in induced corneal assess corneal changes. Unaided logMAR high-contrast visual astigmatism. The solution was to change the base curve gradually acuity, subjective refraction, and questionnaires were used to with time as the cornea changed. However, this procedure required monitor vision and symptoms. Follow-up visits were scheduled several months of treatment to achieve only 1 or 2 diopters (D) of after 1 day, 1 week, 2 weeks, 1 month, 3 months, and 6 months. myopia reduction. Because of only modest and unpredictable Results. For 6-month data, unaided logMAR acuity improved improvements in uncorrected visual acuity, orthokeratology fell Ϯ Ϯ from 0.78 0.26 in the right eye and 0.75 0.22 in the left eye into disfavor among most eye care practitioners and was declared to 0.06 Ϯ 0.18 in the right eye and 0.04 Ϯ 0.16 in the left eye. not to be a viable option for the correction of myopia in the Myopia decreased from –2.55 Ϯ 0.87 D in the right eye and 40–47 Ϫ2.47 Ϯ 0.89 D in the left eye to ϩ0.45 Ϯ 0.74 D in the right eye optometric and ophthalmologic literature. and Ϫ0.17 Ϯ 0.69 D in the left eye. Shape factor, using corneal In 1989, Wlodyga and Bryla introduced accelerated orthokera- topography, increased from 0.85 Ϯ 0.13 in the right eye and 0.85 Ϯ tology, with a report on 15 patients fitted with a new series of 0.15 in the left eye to 1.28 Ϯ 0.32 in the right eye and 1.30 Ϯ 0.29 reverse-geometry lens designs manufactured by Contex, Inc. (Van in the left eye. Both eyes showed a decrease in lower-order Nuys, CA), called OK lenses.48 Keratometry was used to design aberrations (i.e., defocus) and an increase in higher-order aberra- the lens and monitor corneal changes. The difference between tions (i.e., spherical aberrations and coma). Conclusions. Myopia central and peripheral keratometric readings was used to establish reduction after 1 week was clinically insignificant from the the corneal shape factor and to predict the amount of myopia 1-month results, indicating that the full effect is achieved by 1 reduction. It may be argued that because of the large mire sepa- week. Neither total nor epithelial corneal thickness varied significantly from baseline measurements. ration, the keratometer is not the best instrument to use for Key Words: Confocal microscopy—Corneal topography—Opti- monitoring orthokeratology results and that the difference between cal aberrations—Optical pachymetry—Reverse geometry. a central and temporal keratometric reading is not truly represen- tative of the corneal shape or shape factor. Nevertheless, during the next 15 years, new technologies in the manufacturing of the lenses Orthokeratology is a procedure that uses gas-permeable contact and in monitoring their effects on the cornea have revitalized lenses to reshape the cornea to temporarily reduce or modify interest in orthokeratology worldwide.49–65 myopia and astigmatism to achieve a transient improvement in Modern contact lenses for overnight orthokeratology are made of enhanced oxygen-permeable materials and have a more com- plex posterior design. The back surface of newer lens designs may From Eye Care Associates (S.E.); Texas Eye Research and Technology be divided into four or more zones. Each zone influences the lens Center (N.E.L., W.M., J.M., K.P., A.M., A.G.), College of Optometry, positioning over the cornea and the resulting visual outcome. The University of Texas; University of Texas-Houston Medical Center, Her- four zones are the optical zone (central), the reverse curve(s) mann Eye Center (T.C.P.), Houston, Texas. Address correspondence and reprint requests to Dr. S. El Hage, 5320 (paracentral), the alignment curve (intermediate), and the periph- Richmond Ave, Houston, TX 77056; e-mail: [email protected] eral curve (edge lift). Depending on the specific manufacturer, the Accepted March 30, 2007. contact lens has different nomenclature for each zone and different DOI: 10.1097/ICL.0b013e318065b0dd ways of calculating the posterior curvatures of the lens. For

224 ORTHOKERATOLOGY THROUGH CORNEAL TOPOGRAPHY 225 instance, some manufacturers prefer a lens design in which the extended-wear conventional gas-permeable lenses and for daily- central optical zone is calculated first, followed by the alignment wear orthokeratology lens designs. In June 2004, Boston Equalens curve, and finally, the two segments are joined by a return curve of II material was approved additionally for overnight ortho- specific geometric shape. Other manufacturers may choose a keratology. different approach in which the alignment curve (peripheral seg- This clinical trial used a different approach for designing and fitting ment) is calculated first, followed by the optical zone (central orthokeratology contact lenses, which is based on corneal topography segment), and then, the two segments are joined by the reverse and tear layer thickness. The corneal topographer provided three curve patterned after a specific geometric shape. With the afore- coordinates (x, y, and z) of each point on the cornea and allowed mentioned design, the practitioner sends the keratometric readings reconstruction of the shape of the cornea and design of the lens from and refraction to the laboratory for lens design and fabrication. corneal topography information using a built-in software program. Lacking information about the asphericity of the individual cornea, Another consideration is the tear layer thickness between the contact the laboratory assumes an average eccentricity, usually 0.5, when lens and the cornea. In normal gas-permeable lens designs, the tear manufacturing the lens. In another fitting modality, the practitioner thickness is approximately 20 ␮m between the cornea and the back uses a diagnostic set of contact lenses to trial fit the patient. In most surface of the lens. This concept is even more important in fitting cases, the laboratory does not indicate the eccentricity used in their reverse-geometry contact lenses in orthokeratology. A change of a lens design. In yet a third fitting modality, the practitioner uses a few micrometers in the cornea–to–contact lens relationship, for ex- corneal topographer to obtain the apical radius of the cornea, the ample, in the central segment of the lens, may induce a diopter or sagittal height at a chord of 9.35 mm, and the eccentricity or shape more in refractive change. A change of only a few micrometers in the factor. From this information, a diagnostic lens is chosen for an return curve segment or in the alignment segment will affect the overnight trial before ordering the lenses. This last fitting modality centration of the contact lens over the cornea. For instance, to center has an advantage over the previous methods because of the added the contact lens over the patient’s cornea, increasing the return curve information about the shape of the cornea, mainly eccentricity or depth of a few micrometers may be needed to tighten the fit. Alter- shape factor. The authors prefer to use the term shape factor natively, to loosen the lens, the return curve depth may be decreased. because it differentiates between prolate and oblate elliptical The same applies to the alignment curve. By varying the slope of the surfaces. For example, if the shape factor is between 0 and 1.0, the alignment curve, it can be brought closer or farther away from the shape of the cornea lies on the steeper side of the ellipse (i.e., a cornea. Bringing the alignment curve closer to the cornea will tighten prolate shape). If the shape factor is greater than 1.0, the shape of the fit of the contact lens, and moving the alignment curve away from the cornea is on the flat side of the ellipse (i.e., an oblate shape). the cornea will loosen the fit. In this study, the CKR fitting program Because of these innovative designs, lenses may be fitted with base was integrated with the Focal Points software program (Advanced curves significantly flatter than those used in conventional orthok- Medical, SRL, Milan, Italy) to design and manufacture the lenses eratology to result in larger and more rapid changes in corneal (Fig. 1). curvature and thus faster improvements in unaided visual acuity. After designing the orthokeratology lens from the computer Whereas some forms of accelerated orthokeratology typically nomogram, the lens order is sent to the laboratory electronically. rely on traditional keratometry,60,61,66,67 controlled kerato-reforma- The expected benefit of such an empirical system would be to tion (CKR) (Eye Research Associates, Inc., Houston, TX), by simplify the overnight therapy process of achieving temporary definition, is based on computer-assisted videokeratography.68 improvement in unaided visual acuity better than or equal to 20/40 These instruments produce a comprehensive topographic map of while yielding a clinically acceptable physiologic response. Mod- the cornea, which allows the practitioner to more accurately ern orthokeratology requires, at minimum, to fit the lens according monitor the corneal changes that lenses produce with time. The to sagittal height and corneal asphericity, eccentricity or shape CKR procedure consists of fitting reverse-geometry, aspheric, factor, and at best based on corneal topography. The traditional gas-permeable contact lenses designed with the aid of the Optivi- notion of fitting contact lenses on the keratometric reading, flatter sion EH-290 CornealMap (Optivision, Inc. Houston, TX) comput- than the keratometric reading, or steeper than the keratometric erized corneal topographer. The base curve selected depends on reading is not applicable in modern orthokeratology. Corneal the desired amount of myopia reduction, and the corneal shape topography is generally accepted as the standard of care in orthok- factor determines the remaining curves. eratology for follow-up and assessment of the fitting and position- Preliminary retrospective studies conducted by El Hage and ing of the contact lens over the cornea and the health of the cornea. Leach68,69 indicate that perhaps shape factor, as determined by the Along with visual acuity and over-refraction, corneal topography corneal topographer, would be a better indicator of refractive changes. provides the best indication of the efficacy of the procedure. In their retrospective study of CKR on 51 patients, reverse-geometry lens designs were based on videokeratography and manufactured by MATERIALS AND METHODS Metro Optics, Inc. (Dallas, TX). The amount of myopic reduction ranged from 0.37 to 6.25 D and required from as little as 3 days to as The protocol was approved by the University of Houston Col- much as 42 weeks to reach a plateau.68,69 lege of Optometry Institutional Review Board and the University The purpose of this clinical study was to determine the efficacy Committee for the Protection of Human Subjects, in accordance and ease of empirically fitted gas-permeable contact lenses, man- with the Declaration of Helsinki, before the beginning of the study. ufactured in Boston Equalens II material (oprifocon A) (Bausch & Thirty-two subjects were screened, and 29 subjects were entered Lomb, Rochester, NY), for orthokeratology. At the time of this into the study (11 men and 18 women). Nine subjects withdrew or study, the Boston Equalens II material, with a Dk of 85 ϫ10Ϫ11 were discontinued for various reasons. One refused to wear the 3 ⅐ ⅐ ⅐ 3 ⅐ (cm O2 cm)/(sec cm mm Hg) at 35°C, was marketed with lenses because of discomfort and they were not dispensed. Two approval by the Food and Drug Administration for daily- and were lost to follow-up; one was unable to be fitted satisfactorily;

Eye & Contact Lens, Vol. 33, No. 5, 2007 226 EL HAGE ET AL.

FIG. 1. Focal Points Lens Designer display. and one did not achieve acceptable unaided visual acuity. The contact lens (i.e., the overall diameter [OAD], the optical zone average age of the 20 subjects completing the study was 26 Ϯ 3.7 [OZ], the amount of aimed myopia reduction (Pwr), the shape years (range, 20–32 years). Women accounted for 73% of the factor of the lens design (SF), the depth and width of the invagi- cohort, and men accounted for 27%. Each subject was examined at nation or reverse curve (Inv), the anchorage (Anch) or alignment the initial visit to determine eligibility. To be enrolled in the study, zone and its slope, and the edge lift. The computer software allows subjects were required to have normal ocular and systemic health, customization of any of these parameters to achieve an optimally myopia between 1.00 and 4.00 D, astigmatism no greater than 1.50 fitting lens. For example, by changing the slope of the anchorage D, and no history of gas-permeable lens wear. The study was zone, the edge lift will increase (loosen) or decrease (tighten) and explained to the subjects, and they were asked to read and sign a affect the lens position. Likewise, the location of the alignment statement of informed consent. Once enrolled, corneal topography curve and the diameter of the lens may be customized to influence measurements were obtained; study lenses were ordered; and the the fit. The right side of the nomogram shows the values of the subject was scheduled for a dispensing visit. The design of the keratometric readings, corneal astigmatism, shape factor, eccen- lenses was determined directly from corneal topography data and tricity, Q (asphericity), central radius of curvature, surface regu- the CKR computerized fitting nomogram using Focal Points, larity index, visible iris diameter, edge lift, chord, and sagittal without the use of diagnostic lenses. The values of shape factor, height. The asphericity of the cornea is expressed in terms of shape eccentricity, and Q are shown for each cornea and the fitting factor (␳), eccentricity (e), or asphericity (Q), where ␳ ϭ 1 Ϫ e2 nomogram (Fig. 2). The CKR nomogram shows the lens charac- and Q ϭϪe2. After the lenses are designed, the data are electron- teristics and the fluorescein and tear layer thickness under the ically sent to the laboratory through the Internet, and the lenses are contact lens. This program produces an aspheric back surface made. orthokeratology lens design. Figure 2 shows the CKR fitting In addition to the basic eye examination and measurements of nomogram as it appears on the computer monitor. Shown on the corneal topography, confocal microscopy (ConfoScan3; Nidek, Inc., left side is the simulated fluorescein pattern (i.e., bull’s eye) Fremont, CA), Shack–Hartmann aberrometry, ultrasound corneal showing the lens-to-cornea fitting relationship of the designed thickness measurements (CorneaGage Plus; Sonogage, Cleveland, contact lens. On the right side is the tear layer thickness profile OH), and scanning slit topography and corneal thickness (Orbscan II; between the contact lens and the anterior corneal surface. The Bausch & Lomb, Rochester, NY) data were collected. At the initial software shows the amount of tear thickness required at the center dispensing, subjects were asked to wear the lenses overnight, and of the cornea versus the amount required at the junction between follow-up visits were scheduled after 1 day, 1 week, 2 weeks, 1 month, the return curve and the alignment curve. The lower left side 3 months, and 6 months. Corneal topography, subjective refraction, and allows for simulated horizontal and vertical movement of the high-contrast aided and unaided logMAR visual acuity were measured contact lens over the cornea. Finally, the lower right side of the at each office visit. LogMAR visual acuity is a logarithmic transfor- CKR nomogram display shows the parameters of the designed mation of an optotype’s size in minutes of arc, which creates a

Eye & Contact Lens, Vol. 33, No. 5, 2007 ORTHOKERATOLOGY THROUGH CORNEAL TOPOGRAPHY 227

FIG. 2. Optivision CornealMap CKR lens-fitting nomogram.

geometric progression and simplifies statistical analysis. Each subject unnecessary gains in power caused by repeated measures over was also asked to complete a questionnaire regarding lens comfort time. To correct for significance owing to multiple comparisons, and their unaided visual acuity when performing certain tasks. Scan- the Tukey–Kramer correction was applied in all calculations. ning slit topography and corneal thickness, confocal microscopy, Visual Acuity pachymetry, and aberrometry were taken before fitting and then at the Baseline visual acuity values were compared to each subsequent 1-month, 3-month, and 6-month follow-up visits. visit. When compared to baseline, the t value for each visit showed a significant change (PϽ0.0001). Over the seven time points at RESULTS which vision without correction was assessed, there was a signif- All data were analyzed by a mixed-effect repeated-measures icant difference (F ϭ 59.99, PϽ0.0001). An initial improvement in analysis of variance. In the mixed-effect model, all the data are unaided high-contrast logMAR visual acuity was noted after the used to compute the correlation matrix and standard deviations. first night of lens wear, and most subjects reported functional Thus, all eyes were used without averaging. SAS Proc Mixed, the vision for most of the day. The peak of visual acuity improvement computer subroutine, corrects for correlations between eyes and was reached at the end of the first week (Fig. 3).

0.80 0.77 0.70

0.60

0.50

FIG. 3. Uncorrected visual acuity for 20 0.40 subjects wearing CKR lenses for overnight orthokeratology during a 6-month 0.30 period.

0.20 High Contrast logMAR Acuity 0.17

0.10 0.03 0.03 0.02 0.06 0.05 0.00 Baseline 1-Night 1-Week 2-Weeks 1-Month 3-Months 6-Months Wearing Time

Eye & Contact Lens, Vol. 33, No. 5, 2007 228 EL HAGE ET AL.

Wearing Time

Baseline 1-Night 1-Week 2-Weeks 1-Month 3-Months 6-Months

0.00 -0.23 -0.33 -0.31 -0.39 -0.35

-0.50 -0.77

-1.00 FIG. 4. Myopia reduction for 20 subjects wearing CKR lenses for overnight orthokeratology during a 6-month period. -1.50

-2.00

-2.51

Spherical Eq. Refractive Error (D) Error Refractive Eq. Spherical -2.50

-3.00

Refractive Error Corneal Topography Subjects achieved a 66% to 80% reduction in refractive error Corneal topography was shown to change significantly over (Fig. 4). Baseline spherical values were compared to each subse- time (F ϭ 21.21, PϽ0.0001) and, when compared to baseline, was quent visit. When compared to baseline, the t value for each visit significant at each subsequent visit, after correcting for multiple showed a significant change (PϽ0.0001). Dioptric spherical power comparisons. The shape factor showed a significant change (F ϭ changed significantly over time (F ϭ 36.30, PϽ0.0001) from 13.78, Pϭ0.0003), with the average shape factor increasing in Ϫ Ϫ 2.51 D to 0.31 D after 180 days. On the first overnight proportion to the change in refractive error from a prolate to oblate Ϫ Ϫ follow-up visit, the mean was 0.77 D and 0.35 D by day 14. ellipsoidal shape (Fig. 6). When the shape factor increased by 0.1 The change in refractive error as a function of the baseline units, the spherical equivalent decreased by 0.8153 Ϯ 0.1412. refractive error is shown in Figure 5. Also, when the shape factor increased by 0.1 units, the logMAR Ϯ Keratometry results improved by 7.58 2.04 letters, or a gain greater than one Keratometry in the horizontal axis did not change clinically or line of vision on the Bailey–Lovie Chart. statistically over time, and follow-up visits did not differ from baseline values. However, vertical keratometry was statistically Higher-Order Aberrations significant over time (F ϭ 5.33, Pϭ0.002), but comparisons from Higher-order aberrations were determined by using an in house baseline showed small, clinically insignificant changes. Shack–Hartmann aberrometer. Measurements were taken in total

-5.00

-4.50

y = 0.9973x + 0.3151 -4.00 R2 = 0.6719

-3.50

-3.00

-2.50 FIG. 5. Change in refractive error as a function of original refractive error. -2.00 Change in Refraction (D) Refraction in Change -1.50

-1.00

-0.50

0.00 0.00 -0.50 -1.00 -1.50 -2.00 -2.50 -3.00 -3.50 -4.00 -4.50

Original Refraction (D)

Eye & Contact Lens, Vol. 33, No. 5, 2007 ORTHOKERATOLOGY THROUGH CORNEAL TOPOGRAPHY 229

1.40

1.33 1.32 1.34 1.31 1.29 1.20 1.17

1.00

0.85 FIG. 6. Changes in corneal topogra- 0.80 phy shape factor (1 Ϫ e2) for 20 subjects wearing CKR lenses for overnight orthokeratology during a 6-month pe- 0.60 riod. Shape Factor

0.40

0.20

0.00 Baseline 1-Night 1-Week 2-Weeks 1-Month 3-Months 6-Months Wearing Time darkness without pupil dilation, and the Zernike polynomials were throughout the entire corneal thickness at 5-␮m intervals. The calculated for a 4-mm pupil diameter (Fig. 7). As expected, there confocal microscopy images showed that there does not appear was a decrease in lower-order aberrations, primarily defocus, and to be any significant structural changes secondary to orthokera- an increase in higher-order aberrations, primarily coma and spher- tology (Fig. 9). ical aberration. These results were not unexpected and are similar to those found after refractive surgery. As with refractive error, Corneal Thickness defocus and higher-order aberrations did not change after the Corneal thickness measurements were taken with three different 1-month visit (Fig. 8). instruments: the ConfoScan3, the CorneaGage Plus, and the Orb- Confocal Microscopy scan II. The CorneaGage Plus, according to the manufacturer, can The Nidek ConfoScan3 confocal microscope is a noninvasive measure total and epithelial corneal thickness, whereas the Orb- instrument that images the cornea. It is a commonly used instru- scan II is capable of providing a profile of total corneal thickness ment in highly specialized corneal practices for the express pur- across the corneal surface. The ConfoScan3 instrument is not pose of diagnosis, especially of diseases of the cornea. Each generally used for obtaining corneal thickness measurements, but cornea is imaged in nearly 60 seconds with 350 images gathered was said to have that capability.

FIG. 7. Change in wavefront before and after CKR treatment.

Eye & Contact Lens, Vol. 33, No. 5, 2007 230 EL HAGE ET AL.

Baseline Post 2.5 baseline Treatment Lo RMS: 2.574836 Lo RMS: 0.147388 post treatment 2 hi RMS: 0.223408 hi RMS: 0.339782

1.5 FIG. 8. Zernike coefficients before and after 1 CKR treatment.

Coefficient Value 0.5

0 34567891011121314151617181920 -0.5 Zernike Terms

The total central corneal thickness was measured with the Orbscan central corneal thickness data are similar to those of the Orbscan II II, ConfoScan 3, and CorneaGage Plus (Fig. 10). The Orbscan II did and did not show any thickness change over time, nor did the Orbscan not detect any significant thickness change over time, nor did it detect II detect differences from baseline and subsequent visits. Total central differences from baseline and subsequent visits. Total central corneal corneal thickness was also unchanged with this instrument. The thickness was unchanged with this instrument. The confocal total CorneaGage Plus total central corneal thickness did not show thick-

FIG. 9. Confocal microscopy of the corneal epithelium, anterior basal lam- ina, and anterior stroma before and after CKR treatment.

Eye & Contact Lens, Vol. 33, No. 5, 2007 ORTHOKERATOLOGY THROUGH CORNEAL TOPOGRAPHY 231

560

556 550 553 552 549 547 546 546 545 540

530

FIG. 10. Total central corneal thick- 520 ness before and after CKR treatment, 519 as measured with the ConfoScan 3, 510 CorneaGage Plus, and Orbscan II. 510 509 506 Corneal Thickness (um) Thickness Corneal 500

490

480 Baseline 1-Month 3-Months 6-Months Wearing Time

Confocal SonoGage Orbscan II ness change over time, nor did it detect differences from baseline and When measuring central epithelial thickness, the ConfoScan3 subsequent visits. Total central corneal thickness was unchanged with showed a barely significant epithelial thickness change over time this instrument. by 10 ␮m and detected a difference from baseline only at the Total inferior corneal thickness was also measured with each 6-month visit (Fig. 12). There seemed to be a decreasing trend, but instrument (Fig. 11). The Orbscan total inferior corneal thickness it probably was not clinically significant given test–retest variabil- did not show thickness change over time, nor did it detect differ- ity. The central epithelial thickness with the CorneaGage Plus ences from baseline and subsequent visits. Inferior corneal thick- showed a significant epithelial thickness change over time (F ϭ ness was unchanged with this instrument. The confocal total 7.69, PϽ.0001) by only 1 ␮m and detected a difference from inferior corneal thickness showed a barely significant inferior baseline only at the 1-month visit. thickness change over time, but did not detect differences from The inferior epithelial thickness with the ConfoScan3 failed to baseline and subsequent visits. Values changed by as much as 50 show a significant epithelial thickness change over time, with just ␮m, but the change was random over time and did not show a 5 ␮m in variability, and did not detect a difference from baseline consistent trend. The CorneaGage Plus showed a significant infe- and any subsequent visit. The inferior epithelial thickness with the rior thickness change over time (F ϭ 7.92, PϽ0.0001), but CorneaGage Plus also failed to show a significant epithelial detected differences from baseline only at the 3-month visit. thickness change over time, with less than 1 ␮m in variability, and Values changed by as much as 28 ␮m, but the change appeared to did not detect a difference from baseline and any subsequent visit. be random over time and did not show a consistent trend. (Fig. 13).

700 680 677 649 647 638 645 648 646 600 589 562 546 539 500

400 FIG. 11. Peripheral total corneal thickness before and after CKR treatment, as measured with the ConfoScan 3, Cor- 300 neaGage Plus, and Orbscan II approximately 4.0 mm inferior to the geometric

Corneal Thickness (um) Thickness Corneal center of the cornea. 200

100

0 Baseline 1-Month 3-Months 6-Months Wearing Time Confocal SonoGage Orbscan II

Eye & Contact Lens, Vol. 33, No. 5, 2007 232 EL HAGE ET AL.

50 49 48 47 48 48 45 46 43 40 38 35

30 FIG. 12. Central corneal epithelial 25 thickness before and after CKR treatment, as measured with the ConfoScan 20 3 and CorneaGage Plus.

Corneal Epithelium Thickness (um) 10

0 Baseline 1-Month 3-Months 6-Months Wearing Time Confocal SonoGage

Subjective Comfort reduced. Certainly, the effects of contact lenses on the corneal Subjects were asked to rate the overall comfort of the lenses on epithelium have been well documented. Holden et al.70 showed a scale of 0 to 10, with 0 being not tolerable and 10 being unable that a bound silicone elastomer lens resulted in the covered corneal to feel. Comfort is a subjective estimate, and in this study, there epithelium becoming thinner while the adjacent, uncovered epi- was no control lens. Thus, there could be a bias toward patients thelium became thicker. Carkeet et al.63 measured and compared giving information to please the investigators (i.e., the halo effect). central and peripheral corneal thickness in spectacle wearers, However, there was a large change (F ϭ 75.01, PϽ0.0001) in standard gas-permeable lens wearers, and orthokeratology lens perception of comfort over time, from a mean comfort rating of 2.73 wearers. They did not find any significant difference between the at the initial dispensing visit to 7.30 at the 6-month visit (Fig. 14). different treatment groups. Iskeleli et al.71 found a steady decrease in corneal thickness of 17 ␮m in response to orthokeratology during a 6-month period in 29 subjects. Swarbrick et al.,64 using DISCUSSION the Payor–Holden optical micropachometer, measured the differ- The mechanism of orthokeratology is not clear. Is it corneal ence in the total and stromal corneal thickness to obtain epithelial bending? Is it epithelial cell migration? Is it epithelial cell com- corneal thickness at eight locations across the horizontal corneal pression? Is it posterior corneal changes, or is it a combination of meridian on six subjects. For each corneal location, they took three all of the above? Current evidence supports changes in corneal measurements. They were the first to show a decrease in total and epithelial thickness as the means by which a refractive error is epithelial (7.1 Ϯ 7.1 ␮m) central corneal thickness and an increase

50 50

48 48 47 48 47 FIG. 13. Peripheral corneal epithelial thickness before and after CKR treat- 46 ment, as measured with the ConfoScan 46 3 and CorneaGage Plus approximately 4.0 mm inferior to the geometric center 45 45 of the cornea. 45

44 Cornea Epithelium Thickness (um)

42 Baseline 1-Month 3-Months 6-Months Wearing Time Confocal SonoGage

Eye & Contact Lens, Vol. 33, No. 5, 2007 ORTHOKERATOLOGY THROUGH CORNEAL TOPOGRAPHY 233

5 FIG. 14. Subjective comfort rating of the CKR lenses over time. 4

3 Comfort Rating ScaleComfort (0-10)

0 Baseline 1-Night 1-Week 2-Weeks 1-Month 3-Months 6-Months Wearing Time in midperipheral corneal thickness (13.0 Ϯ 11.1 ␮m) after 1 month eratology treatment. Perhaps the variation in the results reflects the of treatment. Fan et al.72 measured the corneal thickness of 54 inherent physical limitations of the instrument used to measure subjects (age range, 11–15 years) with myopia between Ϫ1.25 and corneal thickness. It would be expected that confocal microscopy Ϫ10.75 D and astigmatism up to Ϫ3.00 D. They did not find any is a more appropriate method in measuring the total corneal statistically significant difference in total corneal thickness after 6 thickness, including the epithelial layer. However, the accuracy of months when using A-scan ultrasonography. Nichols et al.73 found measuring corneal thickness may vary among different confocal a reduction in total corneal thickness with the Orbscan II instru- microscopes.77 Perhaps the difference in instrumentation could ment on 10 subjects during a 60-day treatment period. Chow74 explain why no statistically significant reduction in epithelial or measured corneal thickness on 593 subjects by using an ultrasound total corneal thickness was found in the current study. Also, the pachometer and found an average reduction in corneal thickness of Sonogage ultrasound pachymeter did not show any significant 34.8 Ϯ 19.0 ␮m. Mitsui et al.,75 measuring the corneal thickness variation in the epithelium measurements or total corneal thickness on 60 subjects by using the Orbscan II, found an increase in central measurements. The scanning slit optical pachymeter (i.e., Orbscan corneal thickness of 18 ␮m and an increase at the periphery of 22 II) likewise did not show any statistically significant reduction in ␮m. Wang et al.,76 using an optical coherence tomography system, total corneal thickness. In summary, confocal microscopy, ultra- measured the effect of Corneal Refractive Therapy contact lenses sound pachymetry, and optical pachymetry did not show any (Paragon Vision Sciences, Inc., Meza, AZ) on corneal thickness, as statistically significant changes in corneal thickness. compared to a control group wearing regular gas-permeable lenses. Corneal topography showed a shift from a prolate to oblate They found a central epithelial thinning of 5.1% Ϯ 4.9% and ellipsoidal shape and showed a good correlation with changes in midperipheral thickening of 1.9% on the temporal side and 2.4% refractive error. As shown in Figures 4 and 6, as myopia decreased on the nasal side of the cornea. There was no change in the control over time, shape factor increased. Aberrometry showed a decrease group wearing standard gas-permeable lenses. Although the per- in defocus and an increase in higher-order aberrations (i.e., third- centage of standard deviation is almost equal to the percentage of and fourth-order aberrations). This result was not unexpected and the corneal thinning, the change was found to be statistically is similar to results found after myopic refractive surgery. Re- significant (Pϭ0.005). In reviewing the findings of the aforemen- cently, Joslin et al.78 also found an increase in spherical aberration tioned authors (Table 1), five groups of researchers found a and coma after orthokeratology. The increase in spherical aberra- decrease in corneal thickness; two groups did not find any change; tion is attributed to the change of the corneal shape from a prolate and one group found an increase in corneal thickness with orthok- to an oblate shape. The increase in coma could be the result of slight decentration of the treatment zone. Most subjects had func- TABLE 1. Corneal Thickness Measurements in Orthokeratology tional vision after the first night of lens wear. Patients From Eight Investigators

Authors Increase Decrease No change CONCLUSION Carkeet et al.63 ϩ Iskeleli et al.71 ϩ The CKR empirical lens design is an effective and safe method for 64 ϩ Swarbrick et al. temporarily reducing myopia and improving unaided visual acuity. Fan et al.72 ϩ Nichols et al.73 ϩ The amount of myopia reduction found at the 1-week visit was Mitsui et al.75 ϩ clinically insignificant from the 1month results, indicating that the full Chow74 ϩ effect is achieved after 1 week. There is an increase in higher-order Wang et al.76 ϩ aberrations, primarily coma and spherical aberration (i.e., third and

Eye & Contact Lens, Vol. 33, No. 5, 2007 234 EL HAGE ET AL. fourth order of Zernike polynomial), with a decrease in lower-order 28. Kerns R. Research in orthokeratology. Part IV: Results and observa- aberration, primarily defocus (second-order aberration). tions. J Am Optom Assoc 1977;48:227–238. Changes in total and epithelial corneal thickness measurements 29. Kerns R. Research in orthokeratology. Part V: Results and observa- tions: Recovery aspects. J Am Optom Assoc 1977;48:345–359. using different instrumentation are contrary to the results of many 30. Kerns R. Research in orthokeratology. Part VI: Statistical and clinical other researchers.79,80 Epithelial and total corneal thickness mea- analyses. J Am Optom Assoc 1977;48:1134–1147. surements showed variations as a function of the instrumentation 31. Kerns R. Research in orthokeratology. Part VII: Examination of used and the time of measurements. For instance, the Orbscan II techniques, procedures, and control. J Am Optom Assoc 1977;48: system showed central corneal thickness results different from the 1541–1553. 32. Kerns R. Research in orthokeratology. Part VIII: Results, conclusions, CorneaGage Plus and ConfoScan3. The ConfoScan 3 results and discussion of techniques. J Am Optom Assoc 1978;49:308–314. showed a small decrease in central corneal epithelial thickness and 33. Binder PS, May CH, Grant SC. An evaluation of orthokeratology. a slight increase in peripheral epithelial thickness noted at the Ophthalmology 1980;87:729–744. 3-month visit (Figs. 12 and 13). However, the large variance in the 34. Brand RJ, Polse KA, Schwalbe JS. The Berkeley Orthokeratology data prevents any conclusive statements regarding epithelial thick- Study, part I: General conduct of the study. Am J Optom Physiol Opt 1983;60:175–186. ness. The ConfoScan 3 showed a greater level of variability in 35. Polse KA, Brand RJ, Schwalbe JS, et al. The Berkeley Orthokeratol- epithelial thickness measurements, whereas the CorneaGage Plus ogy Study, part II: Efficacy and duration. Am J Optom Physiol Opt did not. Data are similar for each instrument used in assessments 1983;60:187–198. of corneal thickness. In view of these results, no clear differences 36. Polse KA, Brand RJ, Keener RJ, et al. The Berkeley Orthokeratology in total or epithelial corneal thickness were noted over time. Study, part III: Safety. Am J Optom Physiol Opt 1983;60:321–328. 37. Polse KA, Brand RJ, Vastine DW, et al. Corneal change accompany- ing orthokeratology. Plastic or elastic? Results of a randomized controlled clinical trial. Arch Ophthalmol 1983;101:1873–1878. REFERENCES 38. Coon LJ. Orthokeratology: Part I: Historical perspective. J Am Optom Assoc 1982;53:187–195. 1. Jessen GN. Orthofocus techniques. Contacto 1962;6:200–204. 39. Coon J. Orthokeratology: Part II: Evaluating the Tabb method. JAm 2. Neilson RH, Grant SC, May CH. Emmetropization through contact Optom Assoc 1984;55:409–418. lenses. Contacto 1964;8:20–21. 40. Polse KA. Orthokeratology as a clinical procedure [editorial]. Am J 3. Ziff SL. Orthokeratology: Part I. J Am Optom Assoc 1968;3:143–147. Optom Physiol Opt 1977;54:345–346. 4. Ziff SL. Orthokeratology: Part II. J Am Optom Assoc 1968;39:243– 41. Tredici TJ, Shacklett DE. Orthokeratology: Help or hindrance? Trans 254. Am Acad Ophthalmol Otolaryngol 1974;78:425–432. 5. Nolan JA. Approach to orthokeratology. J Am Optom Assoc 1969;40: 42. Eger MJ. Orthokeratology: Fact or fiction [editorial]? J Am Optom 303–305. Assoc 1975;46:682–683. 6. Grant SC, May CH. Orthokeratology: A therapeutic approach to 43. Safir A. Orthokeratology II. A risky and unpredictable ‘treatment’ for contact lens procedures. Contacto 1970;4:3–16. a benign condition. Surv Ophthalmol 1980;24:291–302. 7. Kerns R. Research in orthokeratology. Part I: Introduction and back- 44. Tredici TJ. Symposium: Clinical management of physiologic myopia. J Am Optom Assoc ground. 1976;47:1047–1051. Role of orthokeratology: A perspective. Ophthalmology 1979;86:698– 8. Nolan JA. Orthokeratology. J Am Optom Assoc 1971;42:355–360. 705. 9. Nolan J. Orthokeratology with steep lenses. Contacto 1972;16:31–37. 45. Binder PS. Orthokeratology: Symposium on medical and surgical 10. Gates RC. Orthokeratology and the Air Force. Contacto 1971;15:8–16. diseases of the cornea. In: Transactions of the New Orleans Academy 11. Fontana A. Orthokeratology using the one-piece bifocal. Contacto of Ophthalmology. St. Louis, CV Mosby, 1980, pp 149–166. 1972;16:45–47. 46. Wilson DR, Keeney AH. Corrective measures for myopia. Surv 12. Patterson TC. Orthokeratology: Changes to the corneal curvature and Ophthalmol 1990;34:294–304. the effect on refractive power due to the sagittal length change. JAm Surv Ophthalmol Optom Assoc 1975;46:719–729. 47. Rubin ML, Milder B. Myopia: A treatable ‘disease’? 13. Ziff SL. Orthokeratology J Am Optom Assoc 1976;47:376. 1976;21:65–69. 14. Ziff SL. Orthokeratology procedures. Optometric Weekly 1976;67: 48. Wlodyga RJ, Bryla C. Corneal molding: The easy way. Contact Lens 430–434. Spectrum 1989;4:14–16. 15. Freeman RA. Orthokeratology and the Corneascope computer. Opto- 49. Lebow KA. Orthokeratology. In: Bennett E, Weissman B, eds. Clin- metric Weekly 1976;22:37–39. ical Contact Lens Practice. Philadelphia: JB Lippincott; 1992:2–5. 16. Potts AV. Functional vision through orthokeratological reduction of 50. Horner DG, Wheeler WH, Soni PS, et al. A noninvasive alternative to astigmatism. Optometric Weekly 1976;22:29–33. radial keratotomy. Ophthal Vis Optics Tech Dig 1992;3:42–45. 17. Williams B. Orthokeratology update. Contacto 1976;20:34–38. 51. Horner DG, Richardson LE. Reduction of myopia with contact lenses. 18. Shed FW. Orthokeratology using the corneal contact lens computer. Pract Optom 1992;3:64–68. Contacto 1977;21:23–26. 52. Horner DG, Armitage KS, Wormsley KA. Corneal molding recovery 19. Carter T. Gas-permeable lenses: Orthokeratology. Contacto 1977;21: after contact lens wear. Optom Vis Sci 1992;69(suppl):156–157. 30–33. 53. Riley CM, Horner DG, Soni PS. Polycon II vs. OK-3 lenses in acute 20. Paige N. Orthokeratology for non-orthokeratologists. Contacto 1977; reduction of myopia. Optom Vis Sci 1992;69(suppl):156. 21:31–36. 54. Kame RT, Hayashida JK. A programmed approach to RGP control of 21. Roth HW. Orthokeratology on children: Experience with myopia myopia. Rev Optom 1992;129:94–98. progression. Contact Lens J 1980;9:6–7. 55. Harris D, Stoyan N. A new approach to orthokeratology. Contact Lens 22. Grant SC. Practical orthokeratology. Contacto 1981;25:13–21. Spectrum 1992;7:37–39. 23. Levy B. Permanent corneal change in a patient undergoing orthokera- 56. Soni PS, Horner DJ. Orthokeratology. In: Bennett E, Weissman B, eds. tology. Am J Optom Physiol Opt 1982;59:697–699. Clinical Contact Lens Practice. Philadelphia: JB Lippincott; 1993: 24. Paige N. Full ortho-k with two lens sets. Contact Lens Forum 1982; 1–7. 7:42–43. 57. Joe JJ, Marsden HJ, Edrington TB. The relationship between corneal 25. Indianer LE. A rigid gas permeable lens for ortho-k. Contact Lens eccentricity and improvement in visual acuity with orthokeratology. Forum 1982;7:31–35. Optom Vis Sci 1993;70(suppl):139. 26. Kerns R. Research in orthokeratology. Part II: Experimental design, 58. Carney LG. Orthokeratology. In: Ruben M, Guillon M, eds. Contact protocol, and method. J Am Optom Assoc 1976;47:1275–1285. Lens Practice. London, Chapman & Hall; 1994:878–884. 27. Kerns R. Research in orthokeratology. Part III: Results and observa- 59. Winkler TD, Kame RT. Orthokeratology Handbook. Boston, Butter- tions. J Am Optom Assoc 1976;47:1505–1515. worth Heinemann, 1995.

Eye & Contact Lens, Vol. 33, No. 5, 2007 ORTHOKERATOLOGY THROUGH CORNEAL TOPOGRAPHY 235

60. Winberry JP. The art and science of accelerated orthokeratology. Pract thickness in response to extended wear of rigid gas permeable contact Optom 1995;6:60–72. lenses. CLAO J 1996;22:133–135. 61. Kame RT. Flatten the cornea the nonsurgical way. Rev Optom 1995; 72. Fan L, Jun J, Jia Q, et al. Clinical study of orthokeratology in young 4:61–64. myopic adolescents. Int. Contact Lens Clin 1999;26:113–116. 62. El Hage SG, Leach NE, Colliac JP, et al. Controlled Kerato-Reforma- 73. Nichols JJ, Marisch MM, Nguyen M, et al. Overnight orthokeratology. tion (CKR): An alternative to refractive surgery. Presented at the Optom Vis Sci 2000;77:252–259. American Academy of Optometry, New Orleans, LA, 1995. 74. Chow E. Night therapy orthokeratology and long term corneal cell 63. Carkeet NL, Mountford JA, Carney LG. Predicting success with change. Presented at Global Orthokeratology Symposium, course 7F, orthokeratology lens wear. A retrospective analysis of ocular charac- Toronto, Canada, August 2002. teristics. Optom Vis Sci 1995;72:892–898. 75. Mitsui I, Yamada Y. Changes on the corneal thickness and curvature 64. Swarbrick HA, Wong G, O’Leary DJ. Corneal response to orthokera- after orthokeratology. In: Manns F, Soderburg PG, Ho A, eds. Oph- tology. Optom Vis Sci 1998;75:791–799. thalmic Technologies XIV, volume 5314. Bellingham WA, SPIE, 65. Mountford J. An analysis of the changes in corneal shape and refrac- 2004, pp 24–26. tive error induced by accelerated orthokeratology. Int Conact Lens 76. Wang J, Fonn D, Trefford LS, et al. Topographical thickness of the Clin 1997;24:128–143. epithelium and total cornea after overnight wear of reverse-geometry 66. Wlodyga RJ, Harris D. Accelerated orthokeratology techniques and rigid contact lenses for myopia reduction. Invest Ophthalmol Vis Sci procedures manual. Part 1. Chicago, National Eye Research Founda- 2003;44:4742–4746. tion, 1993, pp 1–7. 77. McLaren JW, Nau CB, Erie JC, et al. Corneal thickness measured by 67. Lakin D, Estse S, Carter W. Reshaping your ideas. Contact Lens confocal microscopy, ultrasound, and scanning slit methods. Am J Spectrum 1995;10:25–30. Ophthalmol 2004;137:1011–1120. 68. El Hage SG, Leach NE, Colliac JP, et al. An Interactive Software for 78. Joslin CE, Wu SM, McMahon TT, et al. Higher order wavefront CKR Nomograms. Contact Lens Spectrum 1997;8:54–56. aberration in corneal refractive therapy. Optom Vis Sci 2003;80:805– 69. El Hage SG, Leach NE, Shahin R. Controlled Kerato-Reformation: An 811. alternative to refractive surgery. Practical Optometry 1999;10:230–36. 79. Alharbi A, Swarbrick HA. The effects of overnight orthokeratology 70. Holden BA, Sweeney DF, Vannas A, et al. Effects of long-term lens wear on corneal thickness. Invest Ophthalmol Vis Sci 2003;44: extended contact lens wear on the human cornea. Invest Ophthalmol 2518–2523. Vis Sci 1985;26:1489–1501. 80. Sridharan R, Swarbrick HA. Corneal response to short-term orthok- 71. Iskeleli G, Oral AY, Celikkol L. Changes in corneal radius and eratology lens wear. Optom Vis Sci 2003;80:200–206.

Eye & Contact Lens, Vol. 33, No. 5, 2007