Corneal topography of excimer laser photorefractive keratectomy

Corneal topography of excimer laser photorefractive keratectomy

Corneal topography of excimer laser photorefractive keratectomy Stephen D. Klyce, Ph.D., Michael K. Smolek, Ph.D. ABSTRACT The application of the 193...

4MB Sizes 0 Downloads 63 Views

Corneal topography of excimer laser photorefractive keratectomy Stephen D. Klyce, Ph.D., Michael K. Smolek, Ph.D.

ABSTRACT The application of the 193 nm excimer laser for keratorefractive surgery promises to deliver a higher degree of precision and predictability than traditional procedures such as radial keratotomy. The development and evaluation of keratorefractive surgery have benefited from the parallel advances made in the field of corneal topography analysis. We used the Computed Anatomy Topography Modeling System (TMS-1) to analyze a Louisiana State University (LSU) Eye Center series of patients who had photorefractive keratectomy for the treatment of myopia with the VISX TwentyfTwentyTM excimer laser system. The excimer ablations were characterized by a relatively uniform distribution of surface powers within the treated zone. In the few cases that exhibited marked refractive regression, corneal topography analysis showed correlative changes. With topographical analysis, centration of the ablations relative to the center of the pupil could be evaluated. Marked improvement in centration occurred in the patients of LSU Series liB in which the procedure to locate the point on the cornea directly over the pupil's center during surgery was refined. Corneal topographical analysis provides objective measures of keratorefractive surgical results and is able to measure the precise tissue removal effect of excimer laser ablation without the uncertainties caused by measuring visual acuity alone. Our observations forecast the need for improved aids to center the laser ablations and for the development of a course of treatment to prevent post-ablation stromal remodeling. Key Words: clinical diagnosis, cornea, corneal surgery, corneal topography, excimer laser, noninvasive diagnosis, photorefractive keratectomy, refractive surgery

In the past, predictability of keratorefractive surgery was best evaluated by comparing the induced change in ocular refraction and the attempted change. However, it is well accepted that refractive change after surgery does not always accurately reflect the induced change in corneal curvature, particularly when there are likely to be alterations in other ocular dimensions that can obscure or confound the refraction values. Therefore, the evaluation of a keratorefractive surgical procedure is most objectively accomplished by directly

measuring the optical consequences of a physical alteration in corneal shape. While it is obviously necessary to measure refraction, this parameter alone cannot adequately assess the surgical process as it will not distinguish changes in accommodative state, axial length, or other extracorneal changes from the preoperative condition. Therefore, distinct, objective measurements of corneal surface topography should be a primary factor in determining the efficacy of keratorefractive surgeries.

From the Lions Eye Research Laboratories. LSU Eye Center. Louisiana State University Medical Center School of Medicine. New Orleans. Presented in part at the Symposium on Cataract. IOL and Refractive Surgery, San Diego. April 1992. Supported in part by Public Health Service grants EY03311. EY08101. and EY02377 from the National Eye Institute. National Institutes ofHealth. Bethesda. Maryland. and by research grants and equipment donations from Computed Anatomy. Inc.. New York. New York. Dr. Klyce has been a paid consultant to Computed Anatomy. Inc.; he has no financial or proprietary interests in any other devices or treatments mentioned in this chapter. Reprint requests to Stephen D. Klyce. Ph.D .. LSU Eye Center. 2020 Gravier Street. Suite B. New Orleans. Louisiana 70112. 122

J CATARACT REFRACT SURG-VOL 19. SUPPLEMENT 1993

EARLY METHODS OF CORNEAL TOPOGRAPHY ANALYSIS Keratorefractive procedures have undergone substantial development and widespread use in the United States since 1979 and have stimulated a revolution in the principles of corneal topographical analysis and instrumentation. Keratometry, the first practical quantitative method of corneal contour measurement, is still widely used and remains one of the most accurate means of describing corneal curvature. However, the principle behind the keratometer illustrates its inadequacy for assessing many surgical procedures. Keratometers determine the corneal curvature from only four points separated by 3 mm to 4 mm on the paracentral cornea and are specifically designed to measure spheres and any regular or uniform cylinder as would be represented on the surface of an ellipsoid. Since keratorefractive surgery often generates localized topography changes and each type of procedure may produce a different optical zone diameter of modified corneal surface power, it is evident that keratometer readings inadequately characterize corneal curvature in postsurgical corneas. Small treatment zone diameter, unintended decentration of treatment, nonuniform effects oftreatment, and pre-existing irregular astigmatism will produce erroneous keratometer readings, particularly when the measuring mires fall beyond the optical zone. Photokeratoscopes were developed to capture the reflected images of Placido disk mires from the specular surface of the cornea photographically. While the visual keratoscope principle is quite old, the inability to record or measure the magnitude of surface distortion relegated the device to more of a curiosity than a practical clinical instrument. With early photokeratoscopes, such as the Nidek Model PKS 1000 or the Kera Corneascope, clinicians were able to record and compare distortions inherent in individual corneas, albeit in a qualitative fashion. Using this approach, regular corneal cylinder is diagnosed by the elliptical shape of the mires, with the short axis of the ellipse corresponding to the axis of high cylindrical power. Irregular corneal astigmatism is evidenced by nonelliptical distortion of the mires. Similar visual inspection techniques remain in use with hand-held keratoscopes to assess intraoperative astigmatism following corneal transplantation. Unfortunately, even with an experienced surgeon, corneal transplantation produces on average 4 diopters (D) of regular astigmatism. This error cannot be greatly improved upon when intraoperative keratoscopy does not permit the appreciation of corneal cylinder much less than 3 D. More devastating to spectacle corrected visual acuity following corneal transplantation is irregular astigmatism, which can have a serious impact on visual performance. While a qualitative judgment of the severity of irregular astigmatism can be made by inspecting mire distortion, keratoscope mires can look

remarkably distortion free even in the presence of severe corneal topographical alterations. 1

COMPUTER-AUTOMATED CORNEAL TOPOGRAPHY ANALYSIS SYSTEMS To provide quantitative curvature data for clinical diagnosis, a number of advances have been made in the development of computerized keratoscopy;2-9 these permit the analysis of corneal contour over a broad region of the corneal surface. The initial approach to corneal topography analysis relied on comparing photographed mire patterns to standardized patterns and later manual digitization of keratoscope photographs2,3-a costly, labor-intensive effort begging for automation. Soon thereafter, modem digital video image capture and image processing techniques provided a powerful, yet costeffective technology to support automatic digitization of mire patterns. Currently at least three corneal topography analyzers based on videokeratoscopy are available. 10 Each system uses circular mires projected onto the specular corneal surface, electronic edge detection of the reflected image, and a reconstruction algorithm that transforms the twodimensional mire pattern into a representation of the three-dimensional shape of the cornea. Between 256 and 360 data points along each mire ring are acquired and form the raw data set used for surface reconstruction. The proprietary reconstruction algorithms of each system necessarily make approximations 11 as there is no known exact analytical solution. In general, such estimates lead to errors in corneal surface power that are greatest in the corneal periphery and smallest centrally. 12 A major difference among these instruments is in the area of coverage and in the density of points collected and analyzed. The greater the number of points analyzed per unit area without oversampling (i.e., points packed so tightly that the same video pixel is sampled numerous times), the finer will be the detail ultimately provided by the device. Perhaps the most important feature common among the videokeratoscopes is the color-coded contour map of corneal surface power developed at the LSU Eye Center in New Orleans4-a map in which cool colors (violets, blues) represent low corneal power and warm colors (orange, reds) represent high corneal powers. With this approach to the clinical presentation of corneal surface shape distortions, it has become clear that all the clinically relevant information can be obtained when the interval between the contours of these power plots is set to 1.5 D as in the international standard scale. To cover the entire range of powers that could theoretically be encountered, the high and low ends of the scale were assigned 5.0 D increments rather than 1.5 D. A more popular scale uses a range from 28.0 to 65.5 D which maintains the most useful portion of the scale and allows a 1.5 D increment to be used throughout the entire scale. Alternative displays include diopter-at-a-point plots,

J CATARACT REFRACT SURG-VOL 19, SUPPLEMENT 1993

123

astigmatism overlays, and isometric presentation. All the data can be presented in the form of corneal power (diopters) or radius of curvature (millimeters); the selection is usually based on the application. For example, to evaluate corneal topography for keratorefractive surgery, diopters are often selected, whereas for contact lens fitting, radius of curvature in millimeters is often preferred. Despite the utility and significance of color maps in providing quantitative information about corneal power, global topography patterns have remained limited to a descriptive analysis. For corneal topography to become an even more practical tool in clinical trials in which quantitative values are required to permit comparisons of corneal surface topography, it has become necessary to develop numerical descriptors of the corneal surface.'3,14 These have taken various forms, but the most useful appear to be global values that simulate keratometry values (SimK) and the corneal surface regularity index (SRI), both derived from topographic analysis data. The SRI has been particularly useful; it was derived from a clinical correlation between the best spectacle-corrected visual acuity of eyes with 20/20 or better visual potential determined by rigid contact lens acuity measurement and a measure of the optical quality (smoothness) of the cornea overlying the entrance pupil. Full details of its implementation are available; 14 however, while the correlation coefficient for the SRI was strong (0.8) and highly significant (P < .001), work in correlating global corneal shape with visual performance continues. Improvements in contrast sensitivity tests or optical modulation transfer functions will likely playa part in the process of developing stronger numerical correlates between corneal shape and visual performance.

COLOR-CODED CONTOUR MAPPING OF CORNEAL TOPOGRAPHY FEATURES The characteristics of the normal cornea can be determined with the video keratoscope. 15 First, the normal cornea is usually steeper in the center than in the periphery. Second, fellow corneas often appear to be enantiomorphs: nonsuperimposable mirror images of each other. Finally, the contour map of corneal power appears to be unique for each cornea, comparable to the individuality of fingerprint patterns. A commonly encountered form of topographic distortion is regular astigmatism. With-the-rule regular corneal cylinder takes the form of a vertically aligned, bowtie pattern contour map. Against-the-rule corneal astigmatism takes the form of a horizontally aligned bow-tie pattern with the higher power axis running along the 0 to 180 degree meridian. Irregular corneal astigmatism is mapped as irregular contours, and surface distortions of this type are often overcome only with the use of rigid contact lens wear. With the color-coded contour map, one can easily detect patterns associated 124

with the signs of early keratoconus, namely a localized region of steepening. 16 Corneal topography analysis also has been used to refine specific refractive surgical procedures such as epikeratophakia. 17 ,18 Using corneal topography analysis, early myopic epikeratophakia lenticles were found to have a smaller than expected optical zone, and the position of the lenticle over the pupil was not always optimal. Thus, the design of the myopic epikeratophakia lenticle was modified to enlarge the diameter of the optical zone and changes were made in the surgical attachment of the lenticle. Radial keratotomy is the most widely used and studied form of surgical correction for myopia. The PERK Study 19 reports an 85% success rate in the predictability of outcome (i.e., eyes within 2 D of emmetropia) under the carefully controlled protocol of the multicenter clinical trial. In general, the predictability of outcome is influenced by a number of variables-perhaps the greatest variable is the surgery itself. The micron precision of incision depth needed for reproducible results is not a skill that is easily conveyed to all surgeons, and the tissue conditions vary during the act of making mUltiple incisions. A large degree of variation can result even in the two eyes of the same patient, at times requiring optical correction of aniseikonia.

TOPOGRAPHIC ANALYSIS OF EXCIMER LASER PHOTOREFRACfIVE KERATECfOMY FOR MYOPIA Many clinical studies involving the use of the 193 nm excimer laser for the photoablative correction of myopia have been reported. 2o- 26 As with all keratorefractive procedures, success is limited by the biological wound healing response. While spatial imprecision and biological wound healing variations complicate most surgical procedures, surgery with the excimer laser largely overcomes the precision limitations of both lathe and scalpel methods. The excimer laser ablates stromal tissue at a rate of 0.25 Ilm per pulse, a level of precision exceeding other surgical methods. The optical quality of the ablated corneal surface is ultimately the result of the quality and uniformity of the laser beam. Constant attention must be paid to maintain, inspect, and calibrate the excimer laser prior to surgery. The potential for surface contour precision with the excimer laser is best evaluated through color-coded topography mapping. Quantitative descriptors can be used to evaluate three salient features: the optical quality of the treated zone, the stability of the procedure over time, and the accuracy of centration ofthe ablation zone with the pupil. Some information concerning the application of topographic analysis to evaluate the procedure has appeared. 27 ,28 The LSU clinical study results will be described in detail. All eyes were treated by the same physician (Marguerite B. McDonald). In all treatment series, 159 hu-

J CATARACT REFRACT SURG-VOL 19, SUPPLEMENT 1993

man eyes have had the photorefractive keratectomy (PRK) excimer laser procedure. Only Series III was accepting new patients at the time these data were compiled. All eyes received preoperative and postoperative corneal topography exams with the TMS-I. Postoperative examinations were scheduled at the one, three " six and 12 month visits. The laser used was the 193 nm excimer laser system prototype and the production version of the Twenty/Twenty excimer laser system (VISX, Inc., Sunnyvale, CA).

The LSU Blind Eye Series

The LSU blind eye study consisted of nine eyes and was initiated after extensive animal studies. 29 ,3o In this series, corrections as high as - 11 D were attempted to characterize the procedure in humans. 22 Jt was not possible to align these corneas precisely for topographic analysis. Therefore, the topography results were not used to evaluate centration of the ablation zone. The preoperative topography for the first patient was normal, with a relatively spherical central area and a flatter peripheral cornea. One month after treatment to achieve a-II D power change, the actual power of the central cornea had been altered by -11.3 D. The apparent optical zone was close to the intended diameter of 5 mm and exhibited a uniform power distribution. Over the course of one year, there was an initial loss of approximately 5 D of the correction, but the power in the treated area remained unchanged between the six and 12 month examinations. In another patient with a-II D intended correction, the expected decrease in corneal surface power was confirmed at one week postoperatively. At nine weeks there was a pronounced reduction in the diameter of the apparent optical zone, and by 13 weeks essentially no evidence of the ablation remained. This rapid transition suggested that the power decline was due primarily to epithelial hyperplasia. With the patient's permission, the central cornea was de-epithelialized, 1% atropine drops and gentamicin ointment administered, and the eye patched. At 17 weeks postoperatively, a final topographic examination appeared the same as at 13 weeks. During the de-epithelialization process, the anterior central corneal surface was noted to have a loose, gelatinous, and filamentous texture. The conclusion was that the stromal remodeling resulted in a loss in corneal surface power. While the stromal remodeling in this patient appeared to proceed rapidly, stromal keratocytes are migratory and have an immense capacity for synthesis when stimulated. Recognizing this potential drawback renewed attention to topography studies at LSU, since only through such analysis could the temporal nature and magnitude of the regression effect be characterized.

The Initial LSU Sighted Eye Studies: Phase IIA

In the LSU Phase I1A series, 19 eyes were treated with attempted corrections ranging from 2.25 to 8.00 D.25 Seventeen eyes were available for follow-up with corneal topography analysis. 27 In the analysis, certain charac-

teristics of the procedure became evident. Corneal topography underwent a transition during the postoperative period from an early (one to two weeks) reduction in the treatment effect that gradually diminished and ultimately resulted in a stabilization by six months.(Figure 1). These topography changes were assessed by measuring the power at the ablation center. The mean change in corneal surface power at the center closely matched the cohort data based on manifest refraction 25 and provided independent, objective verification of a clinical data set with potentially confounding variables such as progressive increases in axial length.

The Centration Issue

The optimal keratorefractive surgical procedure predictably alters the curvature of the entire cornea. Whereas radial keratotomy can provide a large optical zone of flattened corneal power, it also increases light scatter by scarring the paracentral stroma and thereby elicits the "star-burst" visual complaint. The excimer laser keratectomy procedure itself is not without drawbacks for optical performance for two reasons: (1) The diameter or area of the ablation zone depends on the magnitude of correction needed; thus, increasing the size of the treatment zone requires that the amount of laser energy also be increased. (2) The greater the power change and the larger the desired optical zone, the greater the excision depth of stromal tissue. Among the findings in the LSU excimer laser Phase IIA series was an apparent problem with centration of the procedure over the intended corneal area (Figure 2). Previously, refractive surgical procedures were centered over the line of sight, but it now appears that the best visual performance occurs with a uniform change in corneal curvature centered on the entrance pupil. 31 Maloney32 stressed that keratorefractive surgical procedures should be centered on the pupil while the patient is fixating coaxially with the surgeon. He added that small or decentered optical zones may decrease visual acuity, reduce contrast sensitivity scores, and produce glare. When the average displacement of the visual axis from the pupil center is plotted with respect to the vertical and horizontal directions on the cornea in the PRK IIA series patients, the visual axis tends to appear displaced below the pupil center (Figure 3). Considering a theoretical schematic eye, one would expect the visual axis to be displaced nasally and superiorly at the cornea because of the displacement of the fovea temporally and inferiorly to the optic or pupillary axis. 33 However, PRK patient data showed four of 14 right eyes and two of five left eyes with a temporal displacement of the visual axis at the cornea, and 17 of 19 eyes with an inferiorly displaced visual axis. These data may reflect the individual variation of ocular component locations and axes, yet they also point to a critical need in re-evaluating the use of schematic eye parameters to establish and describe surgical alignment of the living eye. Uozato, Guyton,

] CATARACT REFRACT SURG-VOL 19. SUPPLEMENT 1993

125

52

52

...III

50

50

Q.

48

CII

0

:s

f!

.: 46

;0

Q.

.1 .2 .3 .4 .5 06 67

44

C

.2 42 1ii

:c1\1

-...

40

"i 38 c CII

(J

-10

48

'1:1

;

46

0

44

a.

c ,g

!l D

• ~.,

0

36 34

~

,g

.8 .9

.10 .11 .12 a13

42 40

~;g~~~~:'14

.15 016 017

38 36

0

10

20

30

40

50

60

70

~10

Weeks

30

40

50

60

70

Weeks

Fig. 1.

(KJyce) Change in central corneal power following PRK for myopia from videokeratography: Phase IIA series. Left: Attempted corrections $5.0 D. Right: Attempted corrections >5 .0 D .Note the total regression of one eye in this series. (Reprinted with permission from Wilson SE, KJyce SD, McDonald MB: Ophthalmology 1991 ; 98:1338-1347.)

Fig. 2.

(KJyce) LSU PRK series IIA. Left: Preoperative videokeratograph. Right: Two years postoperative appearance of -8 D attempted correction. Note the decentration of the ablation zone in the 50-degree semimeridian and approximately 1.7 mm toward the periphery.

and Waring34 provide a detailed overview of measurement and error in centering surgical procedures. Predictions about centering the ablation over the entrance pupil seem to be borne out by recent clinical results with the Summit Technology UV200 excimer laser, which has been used to produce ablations with 4 mm diameter optical zones. 35 Under typical photopic conditions, the adult human pupil diameter averages 4 mm to 5 mm and exceeds this value under scotopic conditions. Nonuniform powers within the entrance pupil will degrade optical performance 14 and therefore it is not unexpected that results from the United Kingdom/ Summit study found 78% of the patients reporting a halo effect at night when pupils are relatively dilated. The halo effect is not a diffraction phenomenon as is Sattler's veil, which is caused by epithelial edema,36 but is said to result from the imaging oflight upon the retina 126

after passing through both the treated and portions of the untreated cornea that overlie the dilated pupil. This particular type of halo was thought to occur any time the ablation zone diameter is smaller than the active pupil and/or when marked decentration of the laser procedure exists. The halo complication was said to diminish with time; presumably by a neural adaptation phenomenon, as the optical aberration is essentially permanent. Unfortunately, corneal topographical assessment of patients in this study was not provided; this sort of objective correlate would have been a powerful tool with which to support the concluding hypothetical origin of the diminished visual performance. The centration of a keratorefractive surgical procedure can be measured in a quasi-objective manner using the current capabilities of the TMS-l corneal topography system. 27 In brief, every color-coded corneal topog-

J CATARACf REFRACf SURG-VOL 19, SUPPLEMENT 1993

duce an inherent decentration error along a specific axis. To examine the possibility of centration improvement AVERAGE DISPLACEMENT OF VISUAL AXIS FROM PUPIL CENTER IN SIGHTED LSU PRK as additional experience was gained by the surgeon with E E each surgery, the magnitude of decentration of the ab'-' ....c:: lation zone from the pupil center was plotted for each eye in five laser sessions during the PRK IIA series E I (Figure 5). The data showed no significant trend toward 0 .0 al m I I 'E. improved centration. ;; The visual consequences of decentration were eval~ <0 uated for the Phase IIA series. The lack of a correlation .~ .... between Snellen visual acuity and the amount of de'"' > centration for each eye is shown in Figure 6. This abo 00 ~ -0.5 sence of an effect on acuity is encouraging but may be - 0 .5 0.0 0 .5 misleading as the Snellen acuity test involves high conHorizontal displacement (mm) trast targets. Accurate contrast sensitivity measurements and well-designed psychometric tests should be carried Fig. 3. (Klyce) Displacement ofthe visual axis from the center out before and after surgery to evaluate fully the potenof the entrance pupil in PRK IIA eyes. Minus values for the horizontal axis indicate displacements to the exam- tial relationship between decentration and diminished iner's left, while minus values for the vertical axis indi- visual performance. cate a downward displacement. Most values for the Considerable attention was given to alignment during visual axis are displaced below the pupil center. the LSU series Phase lIB (19 eyes) with an improvement in centration error realized (Figure 7). In this series, the average distance of the ablation centers was 0.47 ± raphy map of every eye in each series at LSU was 0.06 mm from the center of the pupil (Figure 4, right). screened by measuring the magnitude and direction of This improvement occurred solely through careful atthe displacement of the center of the ablation from the tention to centration by both the surgeon and the ascenter ofthe pupil in order to evaluate surgical protocols sistants. Briefly, the patient is physically aligned with the laser delivery system until the fixation light is visible by for centration. In LSU Phase IIA, ablation decentration averaged the patient. As the patient fixates on the light when it is 0.88 ± 0.11 mm from the line of sight and 0.79 ± set to a low power, the surgeon marks the epithelium 0.11 mm from the center of the pupil (Figure 4, left). over the center of the pupil with a probe while viewing However, there were no systematic errors of direction of the cornea through the surgical microscope of the laser offset, suggesting that the optics and alignment of the delivery system. This procedure is repeated at two higher centering mechanisms in the excimer laser did not in- fixation light intensities that cause the pupil to constrict 0 .5

Cl)

Cl) (J

pi

C/)

I~

i~ imI

Cl)

AVERAGE DISPLACEMENT OF ABLATED AREA FROM VISUAL AXIS IN SIGHTED LSU PRK

6

----S

6 PRK lIB. N=19

S

... i5

Limbus

Q)

-'

Q) ()

3

~

3

v

8v ~

O+---~~~~~~~~+-~

P.. en

so ...

;0

..... -3 Q) ()

~

-6+-~--.-~~~~--~~--.

-6

-3

0

3

6

Horizonlal displacemenl (mm) Fig. 4.

~~~~

....ttl .~

o

- 6 +-----""""T---=---r---==:::......-,.------/ 6 o -3 3 -6

Distance from pupil center (mm)

(Klyce) Measured centration of ablations from videokeratography. Left: Phase IIA. Right: Marked improvement of centration in Phase lIB. (Left figure reprinted with permission from Wilson SE, Klyce SD, McDonald MB: Ophthalmology 1991; 98:1338-1347.) J CATARACT REFRACT SURG-VOL 19, SUPPLEMENT 1993

127

,-.,

e e

'-'

2.5

20/50

...,II> ~

r::

II>

-.s..

20/45

o

2.0

~ 20/40

()

::I

Q..

0

~

CJ

-< OJ

1.5

8 0

...,11> II> ()

0

0.5

r::

.~ ...,

«l

;:I

0

>

0 0 0

:0

«

Fig. 5.

0

2

o

~

0

0 0

o

128

.t.

20/25

o

0

.t.

""

o

""

0

0

.t.ce I!II!!O

~

0

Il.l

I!!I

.t.1!!I

CD

20/10 20/5

3

4

5

6

(Klyce) No visible trend in ablation decentration as a function of laser session.

.t.

0.0

Laser session

by relative amounts, thereby producing three epithelial marks. Ideally these marks would be coincident with one another. Two assistants aid the surgeon by viewing the procedure from specific positions outside the surgical field and commenting on the location of the entrance pupil relative to the probe tip according to their view of the procedure. A corneal ring marker with a specific diameter is centered on the probe marks by the surgeon using the cross hairs of the ring marker as a guide. The ring marker is pressed onto the epithelium and deposits an ink stain. The epithelium is removed from the area within the ring mark, and a hand-held vacuum suction ring is placed on the eye. The circular margin of the denuded corneal surface is now aligned with the con-

Fig. 7.

20/30

I'i ~ 20/20 Qj I'i 20/15

o o

0 0.0

20/35

OJ

1.0

~

r::

0

;:I

e

....

PRK II A

0.5

1.0

1.5

2.0

2.5

Pupil Center to Ablation Center Distance (mm)

Fig. 6.

(Klyce) Poor correlation between Snellen visual acuity and decentration in the PRK Phase IIA study. ~ = best visual acuity at 180 days; 0 = best visual acuity at 365 days.

centric rings of the surgical microscope reticle, and the tissue ablated. Unfortunately, although this method has improved the centering of the ablation, no additional insight was gained concerning potential optomechanical alignment devices that might be able to sense the entrance pupil's location automatically and relay this information back to the laser delivery system.

The LSU Sighted Eye Study: Phase III

In the LSU Phase III series to date (5/28/92), 82 patients (102 eyes) have received treatment with maximum follow-up at 12 months. The topographical data from this series are similar to the results presented for LSU PRK Phase lIB over the same postsurgical interval.

(Klyce) Left: Postoperative videokeratograph. Right: One and one-half years postoperative appearance of -4.78 D attempted correction. J CATARACf REFRACT SURG-VOL 19, SUPPLEMENT 1993

PREOPERATIVE TOPOGRAPHICAL SCREENING OF REFRACTIVE SURGICAL PATIENTS A cornea that has progressed to marked asphericity, such as in keratoconus, probably cannot be maintained with a surgically induced spherical shape with a high degree of reliability. Thus, careful preoperative screening for corneal shape anomalies is no longer an option; it is absolutely essential. Color-coded contour mapping offers the most sensitive method available with which to diagnose keratoconus, particularly in its earliest manifestations. Keratoconus appears on contour maps as a localized region of surface power noticeably greater than its surround. Although there is a tendency for cones to occur inferonasally, they can be located anywhere on the corneal surface. 37 Screening potential excimer laser refractive surgical patients is particularly important because early keratoconus patients may consider keratorefractive surgery if their vision with spectacle correction is suboptimal. Topographical analysis from eight of22 eyes screened prior to laser refractive surgery revealed keratoconus; this incidence was much higher than would normally be found in a random sample of the population. Furthermore, both corneas of surgical candidates must be contour mapped, as unilateral subclinical keratoconus may exist in the fellow cornea and could well be an early warning sign for the incipient development of a cone in the prospective surgical eye. In the preoperative topographic examinations for keratorefractive surgical candidates, a contact lens history should be mandatory. If a keratoconus-like pattern is observed and the patient has recently worn or currently wears contact lenses, evaluation should be repeated every two weeks until the topography is stable, whereupon the keratoconus issue can be resolved. 38 A final point on the issue of contact lens wearers and excimer screening procedures is that a fraction of lens wearers will have contact lens warpage that can alter their corneal power by 3 D or more. If contact lens wear is discontinued only two weeks before preoperative evaluation, the final refractive result may be unsuitable because the time for re-establishment of corneal stability after contact lens warpage may be as long as five (soft lenses) to 15 (poly[methyl methacrylate] lenses) weeks. 39

CONCLUSIONS AND PREDICTIONS FOR THE FUTURE Excimer laser area ablation clearly has enormous potential for becoming the standard keratorefractive surgical procedure to correct low to moderate am~unts of myopia. Methods to control stromal remodehng and epithelial hyperplasia are within the realm of modern cell biology and will be discovered, and stable high myopia corrections will eventually be achieved with the excimer laser. The solution to the centration issue will be the application of a real-time excimer beam guidance system

or, perhaps, a simple, yet clever optical alignment aid. Eventually, real-time corneal topographic analysis will replace surgical keratometers40 and guide surgeons intraoperatively in reducing naturally occurring and induced astigmatism. With such a system for general anterior segment use, a separate device for excimer laser guidance may be unnecessary. For the proper evaluation of excimer laser ablations, algorithms will be implemented that rotate and translate preoperative and postoperative corneal topographies so that when they are in perfect registration, a subtractive method can be used to evaluate the quality of the photoablative surgery. Quantitative descriptors of corneal optical performance will be enhanced by inclusion of pupil size considerations. 41 The fundamental aspects of visual performance will be derived from the development of spatially resolved refractometry with comparison to corneal topography.42 Finally, an advanced ray tracing approach28 will be implemented with topography systems that will simulate patient vision, and this will be particularly valuable for diagnosis when unusual visual anomalies such as monocular diplopia are experienced. REFERENCES 1. Maguire U, Klyce SO, Sawelson H, et al. Visual distortion after myopic keratomileusis: computer analysis of keratoscopephotographs. OphthalmicSurg 1987; 18:352-356 2. Ooss JO, Hutson RL, Rowsey JJ, Brown OR. Method for the calculation of corneal profile and power distribution. Arch Ophthalmol1981; 99:1261-1265 3. Klyce SO. Computer-assisted corneal topography: highresolution graphic presentation and analysis of keratoscopy. Invest Ophthalmol Vis Sci 1984; 25:1426-1435 4. Maguire U, Singer OE, Klyce SO. Graphic presentation of computer-analyzed keratoscope photographs. Arch Ophthalmol 1987; 105:223-230 5. Gormley OJ, Gersten M, Koplin RS, Lubkin V. Corneal modeling. Cornea 1988; 7:30-35 6. Arffa RC, Warnicki JW, Rehkopf PG. Corneal topography using rasterstereography. Refract Corneal Surg 1989; 5:414-417 7. Belin MW, LitoffO, Strods SJ, et al. The PAR technology corneal topography system. Refract Corneal Surg 1992; 8:88-96 8. EI Hage SG. A computerized corneal topographer for use in refractive surgery. Refract Corneal Surg 1989; 5:418424 9. Koch 00, FoulksGN, MoranCT, WakilJS. The Corneal EyeSys System: accuracy analysis and reproducibility of first-generation prototype. Refract Corneal Surg 1989; 5: 424-429 10. Wilson SE, Klyce SO. Advances in the analysis of corneal topography. Surv Ophthalmol 1991; 35:269-277 II. Wilson SE, Wang J, Klyce SO. Quantification and mathematical analysis of photokeratoscopic images. In: Schanzlin OJ, Robin JB, eds, Corneal Topography. Measuring and Modifying the Cornea. New York, SpringerVerlag Inc, 1992; 1-9 12. Wang J, Rice OA, Klyce SO. A new reconstruction algo-

J CATARACf REFRACf SURG-VOL 19. SUPPLEMENT 1993

129

13.

14.

15. 16. 17.

18. 19.

20.

21.

22.

23.

24.

25.

26.

27.

130

rithm for improvement of corneal topographical analysis. Refract Corneal Surg 1989; 5:379-387 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 descriptors of corneal topography: a clinical study. Arch Ophthalmol1991; 109: 349-353 Dingeldein SA, Klyce SD. The topography of normal corneas. Arch Ophthalmol1989; 107:512-518 Maguire U, Bourne WM. Corneal topography of early keratoconus. Am J Ophthalmol 1989; 108:107-112 Maguire U, Klyce SD, Singer DE, et al. Corneal topography in myopic patients undergoing epikeratophakia. Am J Ophthalmol 1987; 103:404-416 Reidy JJ, McDonald MB, Klyce SD. Corneal topography of epikeratophakia. Refract Corneal Surg 1990; 6:26-31 Waring GO III, Lynn MJ, Culbertson W, et al. Three-year results of the Prospective Evaluation of Radial Keratotomy (PERK) study. Ophthalmology 1987; 94: 1339-1354 L'Esperance FA Jr, Taylor DM, Warner JW. Human excimer laser keratectomy: short-term histopathology. J Refract Surg 1988; 4:118-124 Taylor DM, L'Esperance FA Jr, Del Pero RA, et al. Human excimer laser lamellar keratectomy: a clinical study. Ophthalmology 1989; 96:654-664 McDonald MB, Frantz JM, Klyce SD, et al. Central photorefractive keratectomy for myopia: the blind eye study. Arch Ophthalmol 1990; 108:799-808 Seiler T, Kahle G, Kriegerowski M. Excimer laser (193 nm) myopic keratomileusis in sighted and blind human eyes. Refract Corneal Surg 1990; 6: 165-173 Gartry DS, Muir MGK, Marshall J. Photorefractive keratectomy with an argon fluoride excimer laser: a clinical study. Refract Corneal Surg 1991; 7:420-435 McDonald MB, Liu JC, Byrd TJ, et al. Central photorefractive keratectomy for myopia: partially sighted and normally sighted eyes. Ophthalmology 1991; 98:13271337 Zabel RW, Sher NA, Ostrov CS, et al. Myopic excimer laser keratectomy: a preliminary report. Refract Corneal Surg 1990; 6:329-334 Wilson SE, Klyce SD, McDonald MB, et al. Changes in corneal topography after excimer laser photorefractive

28.

29. 30.

31. 32. 33. 34.

35. 36. 37. 38. 39. 40. 41. 42.

keratectomy for myopia. Ophthalmology 1991; 98: 13381347 Maguire U, Zabel RW, Parker P, Lindstrom RL. Topography and raytracing analysis of patients with excellent visual acuity 3 months after excimer laser photorefractive keratectomy for myopia. Refract Corneal Surg 1991; 7: 122-128 McDonald MB, Beuerman R, Falzoni W, et al. Refractive surgery with the excimer laser. Am J Ophthalmol 1987; 103:469 McDonald MB, Frantz JM, Klyce SD, et al. One-year refractive results of central photorefractive keratectomy for myopia in the nonhuman primate cornea. Arch Ophthalmol 1990; 108:40-47 Uozato H, Guyton DL. Centering corneal surgical procedures. Am J Ophthalmol1987; 103:264-275 Maloney RK. Corneal topography and optical zone location in photorefractive keratectomy. Refract Corneal Surg 1990; 6:363-371 Duke-Elder S, ed. System of Ophthalmology. Vol 5: Ophthalmic Optics ~nd Refraction. London, Henry Kimpton, 1970; 134-138 Uozato H, Guyton DL, Waring GO III. Centering corneal surgical procedures. In: Waring GO III, ed. Refractive Keratotomy for Myopia and Astigmatism. St Louis, Mosby-Year Book, 1992; 491-505 Gartry DS, Muir MGK, Marshall J. Photorefractive keratectomy with an argon fluoride excimer laser: a clinical study. Refract Corneal Surg 1991; 7:420-435 Lambert SR, Klyce SD. The origins of Sattler's veil. Am J Ophthalmol1981; 91:51-56 Wilson SE, Lin DTC, Klyce SD. Corneal topography of keratoconus. Cornea 1991; 10:2-8 Wilson SE, Lin DTC, Klyce SD, et al. Rigid contact lens decentration: a risk factor for corneal warpage. CLAO J 1990; 16:177-182 Wilson SE, Lin DTC, Klyce SD, et al. T'Opographic changes in contact lens-induced corneal warpage. Ophthalmology 1990; 97:734-744 Frantz JM, Reidy JJ, McDonald MB. A comparison of surgical keratometers. Refract Corneal Surg 1989; 5:409413 Applegate RA, Gansel KA. The importance of pupil size in optical quality measurements following radial keratotomy. Refract Corneal Surg 1990; 6:47-54 Webb RH, Penney CM, Thompson KP. Measurement of ocular local wavefront distortion with a spatially resolved refractometer. Appl Optics 1992; 31(19):3678-3686

] CATARACT REFRACT SURG-VOL 19, SUPPLEMENT 1993