Gaining Efficiency in Fitting the Post-
Penetrating Keratoplasty Patient

BY DONNA WICKER, O.D., F.A.A.O., PATRICIA BLECKINGER, A.B.O., LYNN KOWALSKI, A.B.O., & KIM WISNIEWSKI, A.B.O.
MAR. 1997

When and how should you incorporate computerized videokeratography into your clinical routine? Here we compare fitting success using a standard keratometer versus a corneal topographer.

Computerized videokeratography provides a detailed analysis of corneal topography, describing both the shape and the refractive quality of most of the corneal surface. Standard keratometry, on the other hand, assumes that the cornea has a spherocylindrical shape and measures the corneal curvature at a three-millimeter optic zone only.

Corneal mapping takes more time than keratometry measurements, and in our department, five doctors share one topographer. We must decide if and when topography may improve our efficiency for some of our most challenging lens fits, eyes with high astigmatism following corneal transplantation.

THE POST-PK PATIENT

Penetrating keratoplasty can result in high irregular astigmatism, preventing a patient from achieving good visual acuity with spectacles and necessitating a rigid contact lens for optimal visual correction. The decision to fit a contact lens after PK is based not only on the amount of astigmatism, but also on anisometropia, surface irregularity, quality of spectacle corrected acuity and type of correction of the contralateral eye. Up to 20 percent of all post-PK eyes and 50 percent of those with PKs performed for keratoconus may need contact lens correction.

METHODS

We conducted a retrospective study of 33 post-PK eyes with at least 2.50D of corneal astigmatism which had been successfully fit with rigid gas permeable contact lenses. In a separate group, we also evaluated six eyes that needed toric RGP contact lenses to achieve adequate lens centration, also with at least 2.50D of corneal astigmatism post-PK. We excluded eyes that required bitoric lenses primarily for residual astigmatism.

The examination included at least visual acuity, slit lamp examination, keratometry using a Bausch & Lomb manual keratometer, verification of contact lens parameters and corneal topography using the TMS-1 Topographic Modeling System by Tomey Technology, Inc.

We took the keratometry reading within 15 minutes of removing the contact lens and within 10 minutes of the topographical measurements. We calculated an initial trial lens using our standard procedure of finding one-third the difference between the flat and steep keratometry readings and adding this to the flat K measurement. For example, if the cornea measured 42.00/46.50, then one-third of the difference (4.50) equals 1.50. The sum of 42.00 and 1.50 equals 43.50, so our initial trial base curve is 7.76mm, corresponding to 43.50D.

For corneal topographical analysis, we took three photographs of each eye with the standard 25-ring cone of the TMS-1 videokeratoscope. The TMS-1 projects concentric light rings onto the cornea, providing corneal reflections at 180-micron intervals from a central point. The photographed rings are then analyzed by a computer program that identifies 256 points along each ring image. We chose the clearest, most complete photograph of each eye. We processed the photo to calculate a simulated keratoscope reading (sim K) using the average of the maximum meridional powers from rings 6, 7 and 8 and the average power for the same rings 90 degrees away (Fig. 1). Using this, we calculated an initial trial lens based on the computerized videokeratoscopy. Again, we added one-third the difference between the flat and steep sim K readings to the flat sim K reading to find the base curve for an initial trial lens.

RESULTS

Of the 33 eyes fit with spherical contact lenses, 16 had a preoperative diagnosis of keratoconus, seven had Fuch's dystrophy, four had traumatic scarring, three had aphakic bullous keratopathy (ABK), two had macular corneal dystrophy and one had herpes simplex scarring (HSV). The total amount of astigmatism (K and sim K) and the calculated initial trial lens base curves are shown in Table 1. Contact lens diameters ranged from 8.8mm to 10.0mm, with an average diameter of 9.45mm. Smaller diameters generally corresponded to steeper corneas and larger diameters to flatter corneas.

Using the paired t-test to analyze the eyes fit with spherical contact lenses (n=33), we found no statistical difference between the total astigmatism measured by standard manual keratometry and that measured by the TMS-1 topographical mapping system (p=0.37). Similarly, we found no statistical difference between the initial trial lens chosen by keratometry and the initial trial lens chosen by the TMS-1 system (p=0.69). Also, we found no statistical difference between the base curves of the successful contact lens and the trial lens calculated with keratometry measurements (p=0.64) or between those of the successful contact lens and the trial lens calculated by the TMS-1 system (p=1.0).

Alternatively, we made a categorical comparison of the trial lens predicted by keratometry with the lens predicted by the TMS-1 using prediction accuracy criterion within ±0.10mm of the lens prescribed. Keratometry predicted a lens with base curve values that were slightly closer to the base curve of the prescribed lens, but there was no statistically significant difference (p=0.78, Chi-square test). Therefore, with respect to the total amount of astigmatism, choosing the base curve of an initial trial lens, and approximating the base curve of the contact lens chosen, there was no statistically significant difference between standard manual keratometry measurements and corneal topographical analysis measurements.

For the eyes fit with bitoric contact lenses (n=6), five had a preoperative diagnosis of keratoconus and one had Fuch's dystrophy. The average total astigmatism using keratometry was 6.80D and using TMS-1 topographical analysis was 7.20D, higher than the 5.40D and 5.70D averages found for eyes fit with spherical lenses. Yet the two eyes with the highest measured astigmatism were fit with spherical RGP contact lenses.

FITTING PHILOSOPHY MATTERS

Computerized videokeratography uses more data to calculate curvature for a central three-millimeter optic zone and provides more data about the corneal contour than standard keratometry. Our purpose was to determine if the additional information provided by the videokeratographer could lead us to an initial RGP trial lens that was different from the lens obtained through standard keratometry, and to determine which method most closely approximated the lenses that were prescribed and successfully worn by our post-PK patients.

In clinical practice, after we evaluate an initial trial lens by biomicroscopy, we modify the fit by trial-and-error. If the accuracy of the initial trial lens could be improved, the fitting process would be more efficient and less time-consuming. However, even with the same measurements, different contact lens fitting philosophies lead to different initial trial lenses for post-PK eyes.

Beekhuis et al. (1991) measured four paracentral keratometry values with fixation lights 30 degrees to the visual axis. The steepest of the two horizontal K values was the base curve of their initial trial lens. Genvert et al. (1985) used the flattest central K reading for the initial trial lens if there was less than five diopters of corneal astigmatism. When fitting corneas that had more than five diopters of astigmatism, their initial trial lens was steeper, although they did not report the calculations they used. Constad (1988) based his initial lens on the average of the two K readings, using a base curve slightly flatter than the average. Lopatynsky et al. (1993) chose the flattest K for the initial trial lens when using standard keratometry. When using a corneal topographer, they made two separate maps of the same eye and found the average of the two flattest curvature readings along the 90-degree axis and three-millimeter optic zone (1.5mm superior location).

Lopatynsky et al. found a significant difference between initial trial lenses calculated by standard keratometry and and those indicated by computerized videokeratography. They found no difference between the trial lenses calculated from topographical analysis and the lenses eventually prescribed. However, the trial lenses they evaluated were those calculated from topographical analysis. Since contact lens fitting begins with an initial lens which is then modified based on clinical performance, fitting philosophy must be considered when reviewing the results of each study.

HYPOTHETICAL LENSES vs. TRIAL LENSES vs. SUCCESSFUL LENSES

The issue of fitting philosophy should be considered when determining the relevance of the results of this study. We chose the base curve for our initial trial lens by adding one-third of the difference between the flat and steep K or sim K measurements to the flat reading. In fitting the lenses, we had the option of altering the lens diameter as well as the base curve to optimize lens centration and movement.

There was no statistically significant difference between the hypothetical trial lens calculated with standard keratometry and the lens chosen by computerized videokeratography. There was also no statistically significant difference between the initial trial lenses and the lenses successfully worn by our patients, and therefore, we will not change our fitting philosophy.

Another point to consider is that our study was retrospective. The contact lenses may have caused corneal molding prior to our measurements. Past studies have shown that contact lens wearers had a greater decrease in corneal astigmatism than spectacle wearers following PK. However, the keratometry and computerized videokeratography measurements recorded here would have been equally affected since the measurements were taken within about 10 minutes of each other.

CONCLUSION

Based on our study, measurements from either a keratometer or the TMS-1 Topographic Modeling System may be used when determining an initial RGP trial lens for eyes with high astigmatism following penetrating keratoplasty. A standard keratometer has the advantage of being quick to use as well as being available in each of our examining rooms. Computerized videokeratography, on the other hand, provides a visual tool for patient education and provides topography information beyond a central three-millimeter optic zone. Here we examined only a small portion of the information that is available from videokeratography.

Future studies and technological advances should enable us to analyze central, mid-peripheral and peripheral contour information to design both spherical and bitoric rigid contact lenses. This could potentially change the current clinical routine, which relies heavily on observation of trial lenses to fit eyes with irregular, highly astigmatic corneas. CLS

References are available upon written request to the editors at Contact Lens Spectrum. To receive references via fax, call (800) 239-4684 and request document #22. (Be sure to have a fax number ready.)

Dr. Wicker is a clinician at W.K. Kellogg Eye Center at the University of Michigan Medical Center in Ann Arbor. Kim Wisniewski is an optician/contact lens technician at W.K. Kellogg Eye Center. Patricia Bleckinger and Lynn Kowalski were optician fellowship students at the time of this study.