Corneal Topographic Alterations in Normal Contact Lens Wearers

Corneal Topographic Alterations in Normal Contact Lens Wearers

Corneal Topographic Alterations in Normal Contact Lens Wearers Jaime Ruiz-Montenegro, MD, 1 Carlos H. Mafra, MD, 1 Steven E. Wilson, MD, 1 J. Michael ...

2MB Sizes 21 Downloads 89 Views

Corneal Topographic Alterations in Normal Contact Lens Wearers Jaime Ruiz-Montenegro, MD, 1 Carlos H. Mafra, MD, 1 Steven E. Wilson, MD, 1 J. Michael Jumper, 1 Stephen D. Klyce, PhD, 2 Edward N. Mendelson, OD1 Purpose: The purpose of this study is to investigate the corneal topography of visually normal asymptomatic eyes that wore rigid and soft contact lenses compared with visually normal eyes that had never worn contact lenses. Methods: Thirty-seven normal corneas and 7 4 corneas in asymptomatic eyes that wore rigid (12 polymethylmethacrylate and 23 gas-permeable) and soft (26 daily-wear and 13 extended-wear) contact lenses for refractive correction underwent slit-lamp examination, keratometry, computer-assisted topographic analysis, refraction, and rigid contact lens over-refraction. Results: Topographic abnormalities tended to be more common and more severe in corneas that wore rigid contact lenses, but significant changes were noted in some eyes that wore daily-wear or extended-wear soft contact lenses. A number of eyes in the rigid polymethylmethacrylate (9 of 12) and rigid gas-permeable (6 of 23) contact lens groups had a correlation between the most frequent resting position of the contact lens and the corneal topography, with relative flattening of the corneal contour beneath a decentered lens. A total of 10 eyes in the rigid contact lens groups had a 1-line decrease in best spectacle-corrected visual acuity attributable to contact lens-induced topographic abnormalities. Conclusions: Corneal topographic alterations are common in asymptomatic contact lens wearers and are frequently detectable only with computer-assisted topographic analysis. It is important that topographic abnormalities be excluded in contact lens wearing eyes before refractive surgical procedures. Ophthalmology 1993;100:128-134

Many studies performed with the keratometer have demonstrated corneal topographic changes induced by the

Originally received: February 24, 1992. Revision accepted: July 13, 1992. 1 Department of Ophthalmology, University of Texas Southwestern Medical Center at Dallas, Dallas. 2 LSU Eye Center, Louisiana State University Medical Center School of Medicine, New Orleans. Supported in part by U.S. Public Health Service grants EY09379 and EY02377 from the National Eye Institute/NIH, Bethesda, Maryland and an unrestricted grant from Research to Prevent Blindness, Inc, New York, New York. Dr. Wilson is a Research to Prevent Blindness William and Mary Greve International Research Scholar. Dr. Klyce is a paid consultant to Computed Anatomy, Inc, New York, NY. None of the other authors or their family members have a proprietary or commercial interest in the instruments used in this study. Reprint requests to Steven E. Wilson, MD, Department of Ophthalmology, University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Blvd, Dallas, TX 75235-9057.

128

wear of contact lenses.'-'' The findings in these studies have varied, with relative steepening of the corneal contour being detected in some and flattening in others. None of these investigations detected a correlation between the contact lens fit and the lens-induced alterations in the corneal topography. 10• 11 The keratometer, however, provides only limited information regarding the paracentral corneal surface and cannot detect changes that occur in the central or peripheral cornea. 12•13 In addition, measurements obtained with the keratometer are only accurate for surfaces that are spherocylindrical. The limitations of the keratometer frequently lead to a lack of appreciation of clinically significant alterations of the corneal contour. 13 In recent studies, 14• 15 computer-assisted analysis was used to monitor topographic alterations that were present in patients with a reduction in visual acuity from contactlens induced corneal warpage. Only corneas that were noted to have distorted keratometer mires or a change in keratometer measurements compared with prefitting values were included in these investigations. The majority of patients wore rigid contact lenses, but alterations also

Ruiz-Montenegro et al · Topography in Normal Contact Lens Wearers were noted in some patients who wore soft contact lenses. Corneas with warpage were noted to have central irregular astigmatism, loss of radial symmetry, and/or frequent reversal of the normal topographic pattern of progressive flattening of curvature from the central to peripheral cornea.14·15 The majority of corneas that wore a rigid contact lens and had warpage were noted to have a correlation between the topography and the resting position of the contact lenses. Commonly, there was a relative flattening of the corneal contour underlying a contact lens that was decentered with respect to the limbus and, in some cases, a relative steepening of the topography outside of the resting position of the contact lensY- 15 For example, a superior riding contact lens frequently induced reversible, inferior steepening of the corneal contour that simulated early keratoconus. In some cases, the correlation only became manifest after removal of the contact lenses for a period of days to weeks. 15 Computer-assisted topographic analysis provides a sensitive method for monitoring corneal topographic changes in normal, asymptomatic wearers of contact lenses. In prior studies, we have examined topographic alterations in patients who were symptomatic from contact lens-induced visual dysfunction; in this study, we examined the corneal topography of asymptomatic, normal contact lens wearing patients compared with the topography in control subjects who had never worn contact lenses. We used computer-assisted corneal topographic analysis to evaluate the corneal contour of corneas that wore rigid polymethylmethacrylate, rigid cellulose acetate butyrate (gas-permeable), daily-wear soft, and extendedwear soft contact lenses compared with normal corneas that had never worn contact lenses.

Patients and Methods All subjects in the contact lens groups were successful, asymptomatic contact lens wearers and upon specific inquiry had no complaints of decreased visual acuity with glasses. All contact lenses were worn only for correction of refractive error. Control subjects had never worn contact lenses. None of the subjects had diseases affecting the cornea, a history of ocular trauma, previous ocular surgery, or corneal edema. Thirty-seven eyes of 19 visually normal subjects, 12 eyes of 6 subjects who wore rigid polymethylmethacrylate contact lenses, 23 eyes of 12 subjects who wore rigid gaspermeable contact lenses, 26 eyes of 13 subjects who wore daily-wear soft contact lenses, and 13 eyes of 7 subjects who wore extended-wear soft contact lenses (9 eyes wore disposable contact lenses and 4 eyes wore traditional extended-wear contact lenses) were included. One eye with a macular hole was excluded from the control group. Informed consent was obtained from each participant before inclusion in the study. At the time of enrollment, an examination was performed with a slit lamp, and subjects with corneal abnormalities such as scars or thinning were excluded. The

most frequent position of the contact lens on the corneal surface with respect to the anatomical center of the cornea was noted at the slit lamp and recorded. The contact lens position in some subjects was photographed with a Zeiss slit-lamp camera (Carl Zeiss, Oberkuchen, Germany). The amount of contact lens movement in millimeters with blinking was noted during slit-lamp examination andrecorded. Contact lens wearing subjects had worn contact lenses for a minimum of 7 hours on the day of the examination. Measurements for each eye included a manifest refraction performed with a 0.25-diopter Jackson cross cylinder, measurements performed with a keratometer (Bausch and Lomb, Rochester, NY) calibrated according to the manufacturer's instructions, and a rigid contact lens over-refraction. Computerized topographic analysis was performed with a Topographic Modeling System (TMS-1, software version 1.1, Computed Anatomy, Inc, New York, NY) that had a 25-ring collimated video keratoscope and digitized 256 points along each mire. The TMS-1 was calibrated by Computed Anatomy before the study. TMS-1 measurements were performed according to the manufacturer's instructions. Scanning laser slits were superimposed on the vertex of the cornea with the subject viewing the central fixation light of the collimated videokeratoscope at the time of capture. All measurements for each subject were performed by a single observer. Three keratoscope images were obtained from each eye. The image that was judged to have the sharpest focus without mire irregularity or interference from the eyelids was selected for processing. A study was only accepted if there was a 100% correspondence between the observed location of the mires and the mire positions identified by the instrument. A topographic scale with 1-diopter intervals was obtained for each cornea. All measurements were performed within 1 hour after contact lens removal in the contact lens wearing subjects. Data from each computerized topographic analysis were evaluated statistically using computerized algorithms that determined the surface asymmetry index and the surface regularity index. 16·17 The surface asymmetry index provides a quantitative measure of the radial symmetry of the four central photokeratoscope mires surrounding the vertex of the cornea. The higher the degree of central corneal symmetry the lower the surface asymmetry index. A high degree of central radial symmetry is characteristic of normal corneas. 18 The surface regularity index is a quantitative measure of central and paracentral corneal irregularity derived from the summation of fluctuations in corneal power that occur along semimeridians of the 10 central photokeratoscope mires. The more regular the anterior surface of the central cornea the lower the surface regularity index. The surface regularity index has been shown to have a high correlation with best spectacle-corrected visual acuity. 16 Statistical comparisons were performed with the Wilcoxon signed-rank test. A P value less than 0.05 was considered statistically significant. Errors were expressed as the standard errors of the mean. Standard deviations also were determined for the visually normal control group. 129

Ophthalmology

Volume 100, Number 1, January 1993

Table 1. Topography in Normal Contact Lens Wearers

Number of eyes SAl SRI

Control

RigidPMMA

Rigid Gas Permeable

Daily-wear Soft

Continuous-wear Soft

37

12 0.86 ± 0.22 (P = 0.005)

23 0.48 ± 0.09 (P = 0.04)

26 0.48 ± 0.11 (P = 0.36)

13 0.46 ± 0.08 (P = 0.02)

1.17 ± 0.34 (P = 0.009)

0.93 ± 0.18 (P = 0.005)

0.52 ± 0.08 (P = 0.27)

0.51 ± 0.06 (P = 0.40)

3

9

13

8

3

NA

9

6

0

0

7

3

0

0

0.35 ± 0.03 Standard deviation ±0.19 0.41 ± 0.04 Standard deviation ±0.22

Abnormal topography Correlation between topography and contact lens fit Decrease in BSCVA

PMMA = polymethylmethacrylate; NA = not applicable; BSCVA = best spectacle-corrected visual acuity. Errors are expressed as the standard error of the mean. Standard deviations also are provided for the control group.

Results There were no differences in the mean wearing times between the soft daily-wear, rigid gas-permeable, and rigid polymethylmethacrylate contact lens groups. All patients in the daily-wear soft and rigid contact lens groups wore their contact lenses a minimum of 10 hours and a maximum of 18 hours each day. All extended-wear soft contact lens patients wore lenses continuously for l to 3 weeks before cleaning and reinsertion. Table l provides data for the mean surface asymmetry index and mean surface regularity index values of each

>< Cl)

..

5

><

Cl)

~4

'tJ

c

..

~3 Cl)

~2

.. ..

~

tn

ca1 Cl)

..

.. ..

....

.. *

~··

.. ..

+ : *

~

.. ..

·a:

..!

.. ..

t ...

:I

*

::::s

0-1~--------~--------------------~ Control

Rigid GPCL

Rigid PMMA

Group

Daily SCL

...

C) Cl)

~ o~~:~•~l~!~*•~:~----·*=!~~~·=*-"~____.._.. _...._..____:_..~•-·*~~~~----·~,rL-·-~Extended SCL

Figure 1. Surface asymmetry index of control corneas and corneas that wore rigid gas-permeable contact lenses, rigid polymethylmethacrylate contact lenses, soft daily-wear contact lenses, and soft extended-wear contact lenses. Thick horizontal bars are the means and the fine vertical bars are ± the standard error of the mean for each group. *Indicates that the mean for the group was significantly different from the mean for the control group.

130

group in the study. In addition, the P value is provided for the difference between each contact lens group and the normal control group. Figures 1 and 2 show the data for surface asymmetry index and surface regularity index, respectively, for each of the contact lens groups and the control group. The surface asymmetry index and surface regularity index values for the rigid polymethylmethacrylate and rigid gas-permeable contact lens groups were significantly higher than the values for the control group. There were no differences for surface asymmetry index or surface regularity index between the daily-wear soft contact lens group and the control group. The surface asymmetry index values for the extended-wear soft contact

Cl)

u ca

't: :I 0

4 3.5 3 2.5 2 1.5 .. 1 .. .. .... 1- .. .5 ::~+ ••• • 0

...

Control

....

.. .... ..

*

·:+ . ·-·+ ·t t""' .... :

....

*

h

Rigid GPCL

Rigid PMMA

Group

.... .. ..t .. .. .... ··:•t+ .. t ...... .. .. ..

"'"41.-+-

Daily SCL

Extended SCL

Figure 2. Surface regularity index of control corneas and corneas that wore rigid gas-permeable contact lenses, rigid polymethylmethacrylate contact lenses, soft daily-wear contact lenses, and soft extended-wear contact lenses. Thick horizontal bars are the means and the fine vertical bars are ± the standard error of the mean for each group. *Indicates that the mean for the group was significantly different from the mean for the control group.

Ruiz-Montenegro et al · Topography in Normal Contact Lens Wearers lens group were significantly higher than the surface asymmetry index values for the control corneas. There was no significant difference for the surface regularity index between the extended-wear soft contact lens group and the control group. Significant differences between the control corneas and the contact lens-wearing corneas can be attributed in part to marked changes in a small number of corneas in each contact lens group. If the common statistical criterion was applied, such that values greater than 2 standard deviations above the mean of the control group were considered abnormal, then 12 contact lens wearing corneas had a surface asymmetry index that was abnormal (surface asymmetry index greater than 0.73; 2 rigid gas-permeable, 4 rigid polymethylmethacrylate, 4 soft daily-wear, and 2 soft extended-wear) (Table l). Two control group patients had a surface asymmetry index greater than 0.73. Similarly, 22 contact lens wearing corneas had a surface regularity index that was abnormal (surface regularity index greater than 0.85; 9 rigid gas-permeable, 5 rigid polymethylmethacrylate, 7 soft daily-wear, and l soft extended-wear) (Table l). One control group patient had a surface regularity index greater than 0.85. Table l shows the number of corneas in each group that were judged to have topography that was not characteristic of normal corneas. Corneas were considered abnormal if they had central irregular astigmatism, a lack of central radial symmetry, or a reversal of the normal topographic pattern of progressive flattening of curvature from the central to peripheral cornea. 18 Three of 37 control corneas had topographies that were considered abnormal based on these criteria. Figure 3 provides examples of normal corneal topography and Figure 4 shows examples of abnormal topography in corneas that wore contact lenses. Table l shows that there was a significant number of eyes in the rigid polymethylmethacrylate and rigid gaspermeable contact lens groups that had a correlation between the most frequent resting position of the contact lens and the corneal topography. In each case, the correlation occurred in an eye in which the contact lens was decentered relative to the anatomical center of the cornea. In each eye with a correlation, there was relative flattening of the cornea beneath the decentered contact lens and in some cases a relative steepening of the area of the cornea outside of the resting position of the lens. Figures 4C-F provide examples of corneas in which there was a correlation between contact lens position and the corneal topography. There were no corneas with a superior riding, rigid contact lens that did not have flattening beneath the resting position of the contact lens and relative steepening of the cornea outside the resting position of the contact lens. No correlations between contact lens position and corneal topography were detected in the soft contact lens groups. Table l also provides the number of patients in each group that had a decrease in best spectacle-corrected visual acuity. One control eye had a degree of central corneal irregularity and had a best spectacle-corrected visual acuity of 20/25, but could see 20/20 with a rigid contact lens. In the contact lens groups, decreases in best spectacle-

corrected visual acuity were only noted in the eyes that wore rigid lenses and each affected eye had a decrease of only l line (20/25). Figure 4E shows an example of the corneal topography in a patient with decreased best spectacle-corrected visual acuity.

Discussion Computer-assisted topographic analysis is a sensitive clinical and research tool that allows subtle, but clinically significant, alterations of corneal contour to be detected that were not appreciated with the keratometer or keratoscope. The results of this study demonstrate that the normal, asymptomatic wearing of contact lenses may frequently be associated with alterations of corneal topography. The topographic changes noted in the normal contact lens wearers were similar to those that had been previously described in patients with contact lens-induced corneal warpage 14•15 and included central irregular astigmatism, a lack of central radial symmetry, and a reversal of the normal topographic pattern of progressive flattening of curvature from the central to peripheral cornea (Figs 4A-F) in a significant proportion of patients. Alterations tended to be more common and more severe in patients who wore rigid contact lenses, but also were detected in a proportion of soft contact lens wearers. Statistically significant increases in corneal asymmetry (surface asymmetry index) and irregularity (surface regularity index) were present in the corneas that wore rigid contact lenses compared with the control corneas that had never worn contact lenses (Table l) (Figs l and 2). Similarly, the surface asymmetry index values in the extended-wear soft contact lens group were significantly different from the control values. These statistically significant increases in surface asymmetry index and surface regularity index can, however, be attributed to high values for a small number of corneas in each of the contact lens groups since the majority of patients in each group had surface asymmetry index and surface regularity index values that fell within the range of the normal control group (Figs l and 2). Topographic alterations also were noted, however, in a proportion of patients who wore daily- or extended-wear soft contact lenses (Figs 4A and B). Decreases in best spectacle-corrected visual acuity were only noted in eyes that wore rigid contact lenses. In each case, however, the decrease was limited to only one line of visual acuity from the level that was obtainable with rigid contact lens over refraction. The previously described 14•15 correlation (Figs 4C-F) between a decentered contact lens and alterations in corneal topography was only noted in corneas that wore rigid contact lenses. In each case, there was relative flattening of the corneal contour beneath the decentered lens. In some corneas, there also was relative steepening of the area of the cornea outside of the most common resting position of the contact lens. As we have previously noted, 14- 16 corneas with superior riding rigid contact lenses may have a topographic pattern that simulates early keratoconus (Figs 4C and E). The finding of a correlation between a decentered rigid contact lens and the altered

131

Ophthalmology Volume 100, Number 1, January 1993

132

Ruiz-Montenegro et al · Topography in Normal Contact Lens Wearers

Figure 5. A and B, photokeratoscope images that were analyzed to generate the color-coded topographic maps in Figures 4C and 4E, respectively. Notice that the marked topographic alterations of each cornea would not have been appreciated by visual inspection of the photokeratoscope mires.

pattern of the corneal topography suggests that decentration of a rigid contact lens is a risk factor for contact lensinduced changes in corneal contour. 15 We believe that, when possible, rigid contact lenses should be fit so that the most common resting position is centered on the cornea. As has been previously reported in the literature, 13•14•19- 21 topographic alterations that are detected by computerassisted analysis are frequently not appreciated by visual inspection of keratometer or photokeratoscope mires. Many of the topographic abnormalities noted in the corneas that wore contact lenses would not have been noted without computerized analysis. Figure 5 provides examples illustrating the difficulty that may be encountered in detecting the alterations by inspection of the videokeratoscope images alone. Clearly, computer-assisted topo-

graphic analysis is an important method for detecting subtle, but clinically significant, changes in corneal topography caused by contact lenses, disease, and surgical manipulations. We do not routinely discontinue contact lens wear in patients who are asymptomatic and have mild alterations in corneal topography, even if the changes are associated with a small decrease in best spectacle-corrected visual acuity. If patients have spectacle blur or discomfort that cannot be attributed to other abnormalities, we consider discontinuation of lens wear until the topographic alterations resolve. We have previously shown that 5 to 8 months, and in rare cases even longer, may be needed for severe alterations to resolve in corneas that wear rigid contact lenses. 14• 15 In a small number of patients, significant abnormalities may persist without change beyond

Figure 3. Top left, A, control cornea with 0. 7 diopters of with-the-rule astigmatism. Top right, B, control cornea with 3.5 diopters of slightly oblique regular astigmatism. Figure 4. Second row left, A, cornea that wore a soft, disposable, extended-wear contact lens that had central irregular astigmatism with a best spectacle-corrected visual acuity of 20/20. Second row right, B, cornea that wore a soft daily-wear contact lens with inferior peripheral steepening. The soft contact lens was well centered on the cornea and there was no correlation between the resting position of the contact lens and the corneal topography. Best spectacle-corrected visual acuity was 20/20. Third row left, C, cornea that wore a superior-riding rigid gas-permeable contact lens that had central irregular astigmatism and the typical pattern of relative flattening beneath the decentered contact lens and relative steepening inferiorly outside of the resting position of the contact lens. The topographic pattern simulates early keratoconus. The best spectacle-corrected visual acuity was 20/20. Third row right, D, superior resting position of the gas-permeable contact lens worn on the cornea in Figure 4C. Bottom left, E, cornea that wore a superior-riding rigid gas-permeable contact lens that had central irregular astigmatism and the typical pattern of relative flattening beneath the decentered contact lens and relative steepening inferiorly outside of the resting position of the contact lens. The topographic pattern simulates early keratoconus. The pattern was similar in the opposite eye. The best spectacle-corrected visual acuity was 20/ 25 in both eyes. Bottom right, F, superior resting position of the gas-permeable contact lens worn on the cornea in Figure 4E.

133

Ophthalmology

Volume 100, Number 1, January 1993

this extended period of time. 14• 15 In these patients, consideration is given to permanent discontinuation of contact lens wear. An important corollary to the current study is that a large proportion of the patients who are candidates for refractive surgical procedures such as excimer laser photorefractive keratectomy and radial keratotomy wear soft or rigid contact lenses. Our results show that a significant proportion of these patients are likely to have lens-induced corneal topographic alterations. The corneal topography of contact lens-wearing patients should be examined by computer-assisted topographic analysis before any refractive surgical procedure. Contact lens wear should be discontinued and the return of a normal and stable topographic pattern should be documented before surgery. This is especially important for procedures such as excimer laser photorefractive keratectomy because the laser superimposes its spherical or spherocylindrical correction on any existing topographic abnormalities. Lens-induced alterations are likely to change after the surgical procedure and may do so over an extended period of time. Failure to detect these preoperative abnormalities could be an important source of poor predictability and less than optimal results after refractive surgical procedures.

References I. Hartstein J. Corneal warping due to wearing of corneal contact lenses: a report of 12 cases. Am J Ophthalmol 1965;60: 1103-4. 2. Mobilia EF, Kenyon KR. Contact lens-induced corneal warpage. Int Ophthalmol Clin 1986;26:43-53. 3. Miller D. Contact lens-induced corneal curvature and thickness changes. Arch Ophthalmol 1968;80:430-2. 4. Rengstorff RH. Corneal curvature and astigmatic changes subsequent to contact lens wear. J Am Optom Assoc 1965;36:996-1000. 5. Rubin ML. The tale of the warped cornea: a real-life melodrama [letter]. Arch Ophthalmol 1967;77:711-2.

134

6. Levenson DS. Changes in corneal curvature with long-term PMMA contact lens wear. CLAO J 1983;9:121-5. 7. Levenson DS, Berry CV. Findings on follow-up of corneal warpage patients. CLAO J 1983;9:126-9. 8. H0vding G. Variations of central corneal curvature during the first year of contact lens wear. Acta Ophthalmol 1983;61: 17-28. 9. Koetting RA, Castellano CF, Keating MJ. PMMA lenses worn for twenty years. J Am Optom Assoc 1986;57:45961. 10. Rengstorff RH. The relationship between contact lens base curve and corneal curvature. J Am Optom Assoc 1973;44: 291-3. II. Hill JF, RengstorffRH. Relationship between steeply fitted contact lens base curve and corneal curvature changes. Am J Optom Physiol Opt 1974;51:340-2. 12. Klyce SD, Wilson SE. Methods of analysis of corneal topography. Refract Corneal Surg 1989;5:368-371. 13. Wilson SE, Klyce SD. Advances in the analysis of corneal topography. Surv Ophthalmol 1991;35:269-77. 14. Wilson SE, Lin DTC, Klyce SD, eta!. Topographic changes in contact lens-induced corneal warpage. Ophthalmology 1990;97:734-44. 15. Wilson SE, Lin DTC, Klyce SD, eta!. Rigid contact lens decentration: a risk factor for corneal warpage. CLAO J 1990; 16: I 77-82. 16. Wilson SE, Klyce SD. Quantitative descriptors of corneal topography. A clinical study. Arch Ophthalmol 1991;109: 349-53. 17. Dingeldein SA, Klyce SD, Wilson SE. Quantitative descriptors of corneal shape derived from computer-assisted analysis ofphotokeratographs. Refract Corneal Surg 1989;5: 372-8. 18. Dingeldein SA, Klyce SD. The topography of normal corneas. Am J Ophthalmol 1989;107:512-8. 19. Maguire LJ, Bourne WM. Corneal topography in early keratoconus. Am J Ophthalmol 1989;108:107-12. 20. Rabinowitz YS, McDonnell PJ. Computer-assisted corneal topography in keratoconus. Refract Corneal Surg 1989;5: 400-8. 21. Wilson SE, Lin DTC, Klyce SD. Corneal topography of keratoconus. Cornea 1991;10:2-8.