Diurnal fluctuations in corneal topography 10 years after radial keratotomy in the prospective evaluation of radial keratotomy study

Diurnal fluctuations in corneal topography 10 years after radial keratotomy in the prospective evaluation of radial keratotomy study

Diurnal fluctuations in corneal topography 10 years after radial keratotomy in the Prospective Evaluation of Radial Keratotomy study Jonathan R. Kemp,...

349KB Sizes 4 Downloads 56 Views

Diurnal fluctuations in corneal topography 10 years after radial keratotomy in the Prospective Evaluation of Radial Keratotomy study Jonathan R. Kemp, MD, Carlos E. Martinez, MD, Stephen D. Klyce, PhD, Steven J. Coorpender, MS, Marguerite B. McDonald, MD, Lucciano Lucci, MD, Michael J. Lynn, MS, George O. Waring III, MD ABSTRACT Purpose: To correlate clinically observed fluctuations in manifest refraction, visual acuity, keratometry, and intraocular pressure (IOP) with changes in the anterior corneal surface as measured by videokeratography in patients 10 years after radial keratotomy (RK). Setting: Four clinical centers in the United States that participated in the Prospective Evaluation of Radial Keratotomy (PERK) study. Methods: Thirty-two eyes of 20 PERK patients who noted diurnal fluctuations in vision had clinical examination and videokeratography (TMS-1, Computed Anatomy Inc.) in the morning and evening of the same day a mean of 10.3 years (range 7.8 to 11.7 years) after RK. The videokeratographs were analyzed in terms of various indexes generated by custom-designed software. Morning-to-evening changes in the means of the various clinical and videokeratographic values were assessed using pairwise methods. Results: The mean increase in myopia was 0.36 diopters (D) ⫾ 0.58 (SD) from morning to evening (P ⬍ .01). Analysis of the videokeratographs showed a corresponding increase in average corneal power (ACP), reflecting a steepening of 0.52 ⫾ 0.45 D (P ⬍ .001). The change in ACP was correlated with a change in the manifest spherical equivalent refraction (R ⫽ 0.39, P ⫽ .03) and a change in best spectacle-corrected visual acuity (R ⫽ 0.38, P ⫽ .03) over the same period. Similarly, simulated keratometry (SimK) readings correlated with the change in the manifest spherical equivalent refraction (R ⫽ 0.38, P ⫽ .03 for SimK1; R ⫽ 0.37, P ⫽ .35 for SimK2; R ⫽ 0.4, P ⫽ .02 for average SimK), although the standard clinical keratometric data did not (P ⫽ .26 for K1, P ⫽ .11 for K2, and P ⫽ .09 for the mean K). The elevation depression magnitude, a measure of the low-frequency irregularities of the cornea, showed a decrease of 0.32 ⫾ 1.59, which also correlated with the change in the manifest spherical equivalent refraction (R ⫽ 0.37, P ⫽ .04). Intraocular pressure tended to decrease from morning to evening (mean change of ⫺0.97 ⫾ 3.29 mm Hg), but the difference was not significant. Variations in IOP in individual patients, however, were correlated with changes in the manifest spherical equivalent refraction (R ⫽ 0.37, P ⫽ .04). Conclusions: Diurnal fluctuations in corneal topographic indexes can be used to evaluate the diurnal fluctuations in refraction and visual acuity after RK. The study findings provide statistical support for the idea that IOP contributes to the diurnal fluctuation in visual acuity after RK. J Cataract Refract Surg 1999; 25:904 –910 © 1999 ASCRS and ESCRS

© 1999 ASCRS and ESCRS Published by Elsevier Science Inc.

0886-3350/99/$–see front matter PII S0886-3350(99)00090-5

DIURNAL FLUCTUATIONS IN CT AFTER RK

D

iurnal fluctuations in visual acuity after radial keratotomy (RK) have been well documented. Such changes may occur in up to 60% of patients1,2 and have been noted to persist for as long as 11 years after surgery.3 In particular, diurnal fluctuations in corneal curvature, manifest refraction, uncorrected visual acuity (UCVA), and average central keratometric power have been reported.4 – 6 The diurnal (morning to evening) change in manifest refraction tends to be a myopic shift of approximately 0.5 diopters (D), although some patients have shown a hyperopic shift.3 Previously, intraocular pressure (IOP) was believed to be responsible for the increase in corneal curvature and the corresponding refractive change.1–5 However, little correlation between the diurnal changes in keratometric curvature, manifest refraction, and IOP has been found.3–7 Experimental attempts to show a relationship between IOP and refractive error have yielded conflicting data.8,9 Schanzlin et al.6 showed that the change in UCVA from morning to evening does not correlate with the change in manifest refraction or the change in keratometric power. It may be that these measurements do not show a correlation because keratometry is not an adequate measure of central corneal power after RK. Standard keratometry measures the curvature of 4 points in the central 3 to 4 mm of the cornea but gives no information about the curvature central or peripheral to these points. After RK, the entrance pupil may be

multifocal and the 4 points measured by keratometry may be below or above the central corneal curvature, depending on the size of the optical zone and the power within the transition zone between the treated and untreated cornea. With its detailed analysis of the central cornea, videokeratography might be a better measure of the central corneal power after refractive surgery. Analysis of corneal topography after RK has shown corneal steepening from morning to evening.4 –7 Furthermore, McDonnell and coauthors5 have shown a high degree of correlation between diurnal fluctuation in visual acuity and the postoperative shape of the central optical zone on videokeratography. In this study, we evaluated quantitative as well as qualitative changes in corneal topography using the Tomey TMS-1 videokeratoscope (Computed Anatomy Inc.). Clinically, we studied diurnal changes in manifest refractive error, best spectacle-corrected visual acuity (BSCVA), and IOP. Topographically, we looked at several indexes of corneal power and varifocality of the anterior corneal surface, including average corneal power (ACP) over the entrance pupil, surface regularity index (SRI), elevation depression magnitude (EDM), and simulated keratometry (SimK) readings. Finally, we tried to correlate the clinical data with the topographic data to better understand and predict why some patients experience diurnal fluctuations in vision.

Patients and Methods

Accepted for publication March 10, 1999. From LSU Eye Center, Louisiana State University Medical Center School of Medicine (Kemp, Martinez, Klyce, Coorpender), and the Refractive Surgery Center of the South (McDonald, Lucciano), New Orleans, Louisiana, USA; Department of Biostatistics, Emory University School of Public Health (Lynn), and the Department of Ophthalmology, Emory University School of Medicine (Waring), Atlanta, Georgia, USA. Presented in part at the 1995 Pre-Academy Meeting of the International Society of Refractive Surgery, Atlanta, Georgia, USA, October 1995. Supported in part by U.S. Public Health Service grants EY03311 (Klyce), EY02377 (LSU Core grant), EY03752 (Waring), and EYO3761 (Waring) from the National Eye Institute, National Institutes of Health, Bethesda, Maryland, USA. Dr. Klyce is a paid consultant to Computed Anatomy, Inc., New York, New York. None of the other authors has a financial or proprietary interest in any product mentioned. Reprint requests to Stephen D. Klyce, PhD, LSU Eye Center, 2020 Gravier Street, Suite B, New Orleans, Lousiana 70112-2234, USA.

Study Group All patients were part of the Prospective Evaluation of Radial Keratotomy (PERK) study group. The PERK study took place at 9 centers; the design10 and clinical results at 5 and 10 years11,12 have been published. Our study group was a subset of the PERK study group consisting of 32 eyes of 20 patients who, at 3 months after surgery, admitted to fluctuations in vision when asked by the clinical coordinator at 4 of the centers (Emory University, Atlanta, Georgia; LSU Eye Center, New Orleans, Louisiana; University of Southern California, Los Angeles; University of Minnesota, Minneapolis). At the 10 year examination, the mean patient age was 43.5 years (range 33.0 to 51.4 years). Seventeen eyes were right eyes and 15, left.

J CATARACT REFRACT SURG—VOL 25, JULY 1999

905

DIURNAL FLUCTUATIONS IN CT AFTER RK

Surgical Method The standardized surgical protocol used in the PERK study has been described.10 The basic technique consisted of 8 single-pass, freehand, centrifugal radial incisions made equidistant around the cornea. In the first operated eye of each patient, the diameter of the optical clear zone (3.0, 3.5, or 4.0 mm) was determined by the spherical equivalent of the cycloplegic refraction. In the second operated eye, the outcome in the first eye was considered in determining the diameter of the clear zone.11 The incision depth was 100% of the thinnest of the 4 paracentral ultrasonic pachymetric measurements taken intraoperatively. Radial keratotomy was repeated in some eyes.11 Examination Clinical examinations and videokeratography were performed before 9:00 AM and again after 4:30 PM on the same day at the 10 year follow-up visit. The mean time after surgery was 10.3 years (range 7.8 to 11.7 years). A certified PERK coordinator performed both morning and evening measurements in the same room using the same standardized equipment and forms.10 The examination included measurement of IOP with a Goldmann tonometer, evaluation of manifest spherical equivalent refraction using a fogged technique, and determination of UCVA and BSCVA with the modified Bailey⫺Lovie visual acuity charts as described.3 A calibrated keratometer was used to measure the central keratometric power. Videokeratography was performed with the Tomey TMS-1, which uses a 31-ring cone to generate a color-coded topographic map of the cornea. Only patients who had morning and evening videokeratographs that were well-focused and well-aligned were included in the study. Topographic Analysis To quantitatively evaluate the dioptric data as a supplement to the color-coded maps, a custom-designed software package was developed to calculate the following indexes: ACP over the entrance pupil, SRI, surface asymmetry index (SAI), coefficient of variation of corneal power (CVP), EDM, corneal eccentricity index (CEI), SimK, and cylinder (Cyl).13,14 Briefly, the ACP is an area-compensated average of corneal power from the central cornea extending to the edge of the apparent entrance pupil. The SRI is calcu906

lated from a summation of local high-frequency power fluctuations along 256 equally spaced semimeridians on the central 10 mires.14 The SRI increases with increasing irregular astigmatism and is a sensitive indicator of potential visual acuity. The SAI is a centrally weighted summation of differences in corneal power between corresponding points 180 degrees apart on the mires on 128 equally spaced meridians13; it approaches zero for a radially symmetric surface and increases as the shape becomes more asymmetric within specific meridians. The CVP measures the distribution of powers in a videokeratograph and increases as the range of powers increases. The EDM is another measure of irregular astigmatism, focusing on low-frequency distortions. It was originally designed as an objective measure of central island size. The CEI measures the eccentricity of the central cornea and is calculated by fitting an ellipse to the average curve obtained from the 256 semimeridians out to the 25th mire. The SimK readings are calculated using the greatest power from an average of 6 to 8 rings along every meridian (SimK1) and the power and axis of the meridian orthogonal to the highest power (SimK2). The Cyl is calculated as the difference between SimK1 and SimK2. Regression analysis was used to calculate correlation coefficients, and paired t tests were used to analyze the data. Data are given as means ⫾ standard deviations.

Results The mean increase in myopia from morning to evening in the 32 eyes was 0.36 ⫾ 0.58 D (P ⬍ .01) (Table 1). Individually, the changes in manifest spherical equivalent refraction ranged from a 1.38 D increase in minus power to a 1.38 D decrease in minus power. Overall, 5 eyes (15.6%) experienced no change, 5 (15.6%) experienced a hyperopic shift, and 22 (68.8%) experienced a myopic shift. Of the 27 eyes that experienced a shift in refraction, 21 (77.8%) had a shift smaller than 1.0 D. There was no significant change in regular astigmatism by refraction (P ⫽ .36; Wilcoxon signed rank test). Mean IOP was lower in the evening than in the morning (⫺0.97 ⫾ 3.29 mm Hg), but individual changes ranged from ⫺7.33 to ⫹6.00 mm Hg. Regression analysis revealed a correlation between changes in the spherical equivalent and changes in IOP (r2 ⫽ 0.13;

J CATARACT REFRACT SURG—VOL 25, JULY 1999

DIURNAL FLUCTUATIONS IN CT AFTER RK

Table 1. Diurnal changes in clinical and topographic parameters. Parameter

Morning

Evening

Difference

95% CI

Intraocular pressure (mm Hg)

17.75 ⫾ 2.95

16.78 ⫾ 4.21

⫺0.97 ⫾ 3.29

⫺2.15 /0.22

Manifest spherical equivalent refraction (D)

⫺0.12 ⫾ 1.26

⫺0.48 ⫾ 1.52

⫺0.36 ⫾ 0.58

⫺0.57 /⫺0.15

P Value*

Clinical parameters .11 ⬍.01

Uncorrected visual acuity (letters read)

51.09 ⫾ 12.53

49.09 ⫾ 13.42

⫺2.00 ⫾ 5.81

⫺4.09 /0.09

.06

Best spectacle-corrected visual acuity (letters read)

60.66 ⫾ 4.20

60.25 ⫾ 5.39

⫺0.41 ⫾ 3.45

⫺1.65 /0.84

.51

Average keratometry (D)

39.47 ⫾ 1.48

39.73 ⫾ 1.49

0.26 ⫾ 0.47

0.09 /0.43

⬍.01

Average corneal power (D)

38.53 ⫾ 1.81

39.05 ⫾ 1.73

0.52 ⫾ 0.45

⫺0.36 /0.68

⬍.001

Surface asymmetry index

0.42 ⫾ 0.21

0.42 ⫾ 0.20

0.00 ⫾ 0.15

⫺0.05 /0.06

.92

Surface regularity index

⫺0.53 ⫾ 0.25

0.50 ⫾ 0.18

⫺0.03 ⫾ 0.20

⫺0.10 /0.04

.38

Corneal eccentricity index

⫺0.61 ⫾ 0.51

⫺0.53 ⫾ 0.57

0.08 ⫾ 0.19

0.01 /0.15

.03

Coefficient of variation of corneal power

0.71 ⫾ 0.40

0.70 ⫾ 0.41

⫺0.02 ⫾ 0.12

⫺0.06 /0.03

.40

Cylinder (D)

1.18 ⫾ 0.90

1.24 ⫾ 0.95

0.05 ⫾ 0.26

⫺0.04 /0.15

.26

39.39 ⫾ 1.77

39.94 ⫾ 1.70

0.55 ⫾ 0.37

0.42 /0.68

3.60 ⫾ 4.30

3.27 ⫾ 4.26

⫺0.32 ⫾ 1.59

⫺0.89 /0.25

Topographic indexes

Average simulated keratometry (D) Elevation depression magnitude

⬍.001 .26

All values are mean ⫾ standard deviation CI ⫽ confidence interval *Paired t test

Figure 1. (Kemp) Comparison of diurnal changes in manifest spherical equivalent refraction with diurnal changes in IOP in 32 eyes. Intraocular pressure increased in some patients and decreased in others; spherical equivalent changed according to the magnitude and direction of the IOP change (R ⫽ 0.37; P ⫽ .04). A decrease in IOP was accompanied by an increase in myopia. Changes were calculated as evening values minus morning values. Regression lines indicate 95% confidence intervals.

R ⫽ 0.37; P ⫽ .04) (Figure 1). Correlations of IOP versus SimK and IOP versus ACP were not statistically significant. The calculated ACP showed a mean steepening of 0.52 ⫾ 0.45 D (range 1.89 to ⫺0.29 D; P ⬍ .001) from morning to evening. The ACP change correlated significantly with the change in the manifest spherical equivalent refraction by regression analysis (r2 ⫽ 0.16; R ⫽ 0.39; P ⫽ .03) (Figure 2, left). The SimK readings correlated significantly with the change in manifest spherical equivalent refraction for SimK1 (r2 ⫽ 0.15; R ⫽ 0.38; P ⫽ .03), for SimK2 (r2 ⫽ 0.14; R ⫽ 0.37; P ⫽ .04), and for average SimK (r2 ⫽ 0.16; R ⫽ 0.40; P ⫽ .02), although the correlation of the refractive data with the standard clinical keratometric data was not significant (P ⫽ .26 for K1; P ⫽ .11 for K2; and P ⫽ .09 for mean K). The diurnal change in ACP also correlated significantly with the diurnal change in BSCVA (total number of letters read) (r2 ⫽ 0.14; R ⫽ 0.38; P ⫽ .03) (Figure 2, right). The EDM showed a diurnal variation of 0.32 ⫾ 1.59. This change correlated significantly with the change in the manifest spherical equivalent refraction (r2

J CATARACT REFRACT SURG—VOL 25, JULY 1999

907

DIURNAL FLUCTUATIONS IN CT AFTER RK

Figure 2. (Kemp) The morning-to-evening changes in ACP correlated with the changes in both the manifest spherical equivalent refraction and the BSCVA over the same period of time in these 32 eyes. Left: The diurnal changes in ACP and the manifest spherical equivalent refraction were significantly correlated (R ⫽ 0.39; P ⫽ .03). Right: The ACP also correlated with the fluctuation in BSCVA, as measured by the total number of letters read on the modified Bailey⫺Lovie visual acuity charts (R ⫽ 0.38; P ⫽ .03). Changes were calculated as evening values minus morning values. Regression lines indicate 95% confidence intervals.

⫽ 0.14; R ⫽ 0.37; P ⫽ .04) (Figure 3). The correlation between the change in EDM and the change in BSCVA was not statistically significant (P ⫽ .21). There were no statistically significant fluctuations in the other topographic indexes (CVP, SAI, CEI, and SRI), nor were there any significant correlations between them and the clinical measurements (IOP, manifest refraction, visual acuity, and keratometry).

of the small range of visual acuities and small sample size, further studies are needed. Several authors have noted the lack of correlation between changes in UCVA, spherical equivalent, and

Discussion A morning-to-evening change in ophthalmic measurements after RK has been well documented and can persist for up to 11 years after surgery.3,4,6 Such fluctuations have been measured in BSCVA, UCVA, manifest spherical equivalent refraction, and central keratometric power measured using standard keratometry. In our study, 70% of the eyes experienced a diurnal myopic shift, which is consistent with previous reports.3,4,6 We found no significant difference in UCVA or BSCVA in this self-selected group of patients who complained of diurnal fluctuations in vision. It is possible that the changes perceived by patients are not identified by the use of high-contrast Snellen targets. However, because 908

Figure 3. (Kemp) Changes in the EDM correlated with changes in the manifest spherical equivalent refraction from morning to evening (R ⫽ 0.37; P ⫽ .04). Changes were calculated as evening values minus morning values. Regression lines indicate 95% confidence intervals.

J CATARACT REFRACT SURG—VOL 25, JULY 1999

DIURNAL FLUCTUATIONS IN CT AFTER RK

Figure 4. (Kemp) The corneal topography of eyes that experienced diurnal fluctuations in vision showed characteristic changes from morning to evening. This eye has a centrally flat, dumbbell-shaped optical zone in the morning (left). In the evening (right), the optical zone is divided and the ACP has increased 1.2 D.

keratometric data.5,6 Our findings are similar to those of others in that there was no correlation between keratometric power and manifest refraction. This does not mean, however, that diurnal fluctuations in refractive error are not caused by changes in corneal power. As noted, it is more likely that the lack of correlation is attributable to the fact that standard keratometry measures only a small percentage of the anterior cornea and is, therefore, a poor representation of the curvature of the central cornea after refractive surgery. To test whether videokeratography provides a better overall measure of the central cornea curvature, we created videokeratograph-based measures of ACP and correlated them with the change in manifest refraction and BSCVA. The ACP measures the corneal power over the entire entrance pupil, making it a more accurate representation of the average anterior corneal curvature. Simulated keratometry is an index that attempts to simulate standard keratometry and calculates the steepest meridian from data on mires 6 through 8 and the meridian orthogonal thereto. Both the ACP and SimK correlated well with the diurnal changes in BSCVA and manifest spherical equivalent refraction. Our finding that the ACP does correlate with the change in refraction and BSCVA, whereas keratometric measurements do not, suggests that after RK the central cornea is not of uniform power. This indicates that ACP may be a better measurement of corneal power than standard keratometry. The ACP has also been shown to be a better indicator of central corneal power than simulated keratometry.

In a previous study,15 normal and astigmatic corneas showed no disparity between average SimK and ACP, whereas 7% of eyes that had had RK and 25% of eyes that had had photorefractive keratectomy showed a 0.5 D disparity. In the current study, 21.9% of eyes had a disparity of 0.5 D or more between ACP and average SimK in the morning and evening. We also found that the correlation coefficients for changes in ACP and SimK in relation to changes in manifest spherical equivalent refractions were comparable. This indicates that the central corneal surfaces of these post-RK eyes tend to be fairly uniform radially and that either ACP or SimK can be used as a measure of corneal power. The EDM, a measure of irregularity of the central cornea, correlated significantly with the change in the manifest spherical equivalent refraction, indicating that the surface becomes more irregular as it becomes steeper. This correlation explains why the ACP remained a better correlate with the final BSCVA than the SimK readings. It may also account for some of the diffuse changes in vision experienced by post-RK patients throughout the day. This multifocality can result in glare, ghost images, and decreased contrast sensitivity. The mechanism for diurnal fluctuation in refractive error after RK is thought to be the result of changes in corneal hydration and the instability of the peripheral cornea after the collagen lamellae are incised.16 Other factors may also play a role in changing the shape of the cornea, including tension from extraocular and orbicularis muscles. There have been conflicting data in the literature regarding the correlation between changes in IOP and

J CATARACT REFRACT SURG—VOL 25, JULY 1999

909

DIURNAL FLUCTUATIONS IN CT AFTER RK

changes in refraction after RK.8,9 Previously, a correlation between the diurnal change in IOP and the change in manifest refraction has not been demonstrated in PERK patients. However, our substudy did show such a relationship, and despite the lack of significance in previous studies, we believe this correlation is intuitively plausible because reduced IOP should result in peripheral corneal flattening and central steepening. The cause of the myopic shift is probably multifactorial and likely includes both corneal hydration and IOP. In agreement with previous reports, we found no correlation between the changes in IOP and in standard keratometric power. McDonnell and coauthors 5 demonstrated a correlation between the shape of the central optical zone on postoperative topography and the presence of diurnal fluctuations in visual acuity. They described a dumbbell-shaped, flat, central optical zone in the morning that was divided into 2 optical zones in the evening in eyes that experienced diurnal fluctuations in vision. We also noted these topographic phenomena in several of our patients who experienced a significant fluctuation in ACP (Figure 4). This study shows that topographic indexes may be a better way to measure change in corneal power than standard keratometry and thus may more accurately reflect the changes seen in eyes after RK.

5.

6.

7.

8.

9.

10.

11.

12.

References 1. Hoffer KJ, Darin JJ, Pettit TH, et al. UCLA clinical trial of radial keratotomy; preliminary report. Ophthalmology 1981; 88:729 –736 2. Bores LD, Myers W, Cowden J. Radial keratotomy: an analysis of the American experience. Ann Ophthalmol 1981; 13:941–948 3. McDonnell PJ, Nizam A, Lynn MJ, et al. Morning-toevening change in refraction, corneal curvature, and visual acuity 11 years after radial keratotomy in the Prospective Evaluation of Radial Keratotomy Study. Ophthalmology 1996; 103:233–239 4. Santos VR, Waring GO III, Lynn MJ, et al. Morningto-evening change in refraction, corneal curvature, and visual acuity 2 to 4 years after radial keratotomy in

910

13.

14.

15.

16.

the PERK study. Ophthalmology 1988; 95:1487– 1493 McDonnell PJ, McClusky DJ, Garbus JJ. Corneal topography and fluctuating visual acuity after radial keratotomy. Ophthalmology 1989; 96:665– 670 Schanzlin DJ, Santos VR, Waring GO III, et al. Diurnal change in refraction, corneal curvature, visual acuity, and intraocular pressure after radial keratotomy in the PERK Study. Ophthalmology 1986; 93: 167–175 Kwitko S, Gritz DC, Garbus JJ, et al. Diurnal variation of corneal topography after radial keratotomy. Arch Ophthalmol 1992; 110:351–356 Busin M, Yau C-W, Avni I, et al. The effect of changes in intraocular pressure on corneal curvature after radial keratotomy in the rabbit eye. Ophthalmology 1986; 93:331–334 Busin M, Arffa RC, McDonald MB, Kaufman HE. Change in corneal curvature with elevation of intraocular pressure after radial keratotomy in the primate eye. CLAO J 1988; 14:110 –112 Waring GO III, Moffitt SD, Gelender H, et al. Rationale for the design of the National Eye Institute Prospective Evaluation of Radial Keratotomy (PERK) Study. Ophthalmology 1983; 90:40 –58 Waring GO III, Lynn MJ, Nizam A, et al. Results of the Prospective Evaluation of Radial Keratotomy (PERK) Study five years after surgery. Ophthalmology 1991; 98: 1164 –1176 Waring GO III, Lynn MJ, McDonnell PJ, The PERK Study Group. Results of the Prospective Evaluation of Radial Keratotomy (PERK) Study ten years after surgery. Arch Ophthalmol 1994; 112:1298 –1308 Dingeldein SA, Klyce SD, Wilson SE. Quantitative descriptors of corneal shape derived from computer-assisted analysis of photokeratographs. Refract Corneal Surg 1989; 5:372–378 Wilson SE, Klyce SD. Quantitative description of corneal topography; a clinical study. Arch Ophthalmol 1991; 109:349 –353 Maeda N, Klyce SD, Smolek MK, McDonald MB. Disparity of keratometry-style readings and corneal power within the pupil after refractive surgery for myopia. Cornea 1997; 16:517–524 Ingraham HJ, Guber D, Green WR. Radial keratotomy; clinicopathologic case report. Arch Ophthalmol 1985; 103:683– 688

J CATARACT REFRACT SURG—VOL 25, JULY 1999