Photorefractive keratectomy for the correction of spherical myopic refractive errors, particularly those less than 6.00 D, has been shown to be safe and effective.1-3 With advances in the delivery of excimer laser energy to the cornea, sphero-cylindrical ablations are now possible. Preliminary reports of the outcome of such ablations in animals and humans have been published.4-8 The Nidek EC 5000 (Nidek Co. Ltd, Tokyo, Japan) has been in use at this institution for photorefractive keratectomy (PRK) and photoastigmatic refractive keratectomy since August 1993. We present the refractive and visual acuity outcome of eyes undergoing photoastigmatic refractive keratectomy with a minimum of 12 months follow-up.
PATIENTS AND METHODS
One hundred and eighty-two consecutive eyes underwent photoastigmatic refractive keratectomy for primary myopic astigmatism, using the Nidek EC 5000, between January 1994 and June 1995. A retrospective analysis of this population was performed. In the first 3 months after the introduction of this surgery at this institution, 22 eyes were treated using the ablation algorithms programmed into the device's on-board computer. This was found, on early observation, to undercorrect the cylinder power. In light of this experience, and in discussion with other users and the manufacturers, from that time on, 25% was added to the cylinder power for ablation. These first 22 eyes were excluded from the present study, leaving the subsequent 160 consecutive eyes. There were insufficient data at 1 year after the treatment on a further 67 eyes (36 defaulted on follow-up and in 31, examinations were not carried out sufficiently close to the first anniversary of the photoastigmatic refractive keratectomy procedure to be considered a true assessment of the outcome). During the first year, two eyes had a supplementary "mini-radial keratotomy" for residual spherical error that may have affected the remaining astigmatism, a third eye had arcuate keratotomy for residual astigmatism and a fourth eye had supplementary photoastigmatic refractive keratectomy for residual myopic astigmatism. These four eyes were also excluded, leaving 89 eyes in 81 patients, 43 males and 38 females, with a mean age of 34 years (range 20 to 60 years) for analysis at 1 year.
All eyes were examined fully before laser surgery, including videokeratography, keratometry, and manifest refraction. The refraction used to plan the ablation was the subjective refraction, providing the spectacle-corrected visual acuity followed at least 2 weeks free of contact lens use. The mean spherical equivalent refraction, at the spectacle plane, was -5.68 D (SD 2.67) with a mean cylinder power of -1.40 D (SD 0.75; range -0.50 to -5.00 D). The aim of treatment in all eyes was zero refractive astigmatism. It is the policy of this institution to target a residual spherical error of -0.50 D in the non-dominant eye in pre-presbyopic patients, to delay symptoms of presbyopia. Thirtyeight eyes had a baseline spherical equivalent refraction greater than 6.00 D of myopia and 51 eyes, 6.00 D or less of myopia.
Ablations were done by one operator (FL) according to a standard protocol. For the correction of the sphere, 0.25 D was subtracted from the subjective refraction in any patient over 35 years old, because of an empirical observation of the tendency of those patients to be overcorrected. No correction for hyperopic shift induced by the toric ablation was made. The cornea was anaesthetised using topical proxymetacaine 0.5%. At the slit-lamp, the corneal limbus was marked at either end of the horizontal meridian. The patient was then placed under the laser operating microscope, a speculum inserted, and the two limbus marks aligned with the horizontal orientation mire in the microscope to ensure proper orientation of the eye for the toric ablation. The pupil was constricted before surgery with pilocarpine 2%. The ablation, following removal of the central 7.5mm of epithelium, was centered on the visual axis- determined by observation, preoperati vely, of the position of the corneal light reflex in the entrance pupil using a direct ophthalmoscope- and the position of the laser instrument fixation device reflex. The standard program of ablation on the Nidek EC 5000 is sequential. It ablates the required spherical correction first, then the cylindrical element. The cylindrical ablations are achieved by a scanning laser delivery system through an expanding slit and the spherical ablations through an expanding diaphragm. Ablations are carried out using a central optical zone and a peripheral transition zone. All eyes had a peripheral transition zone of 7 to 7.5mm and a central optical zone of 5.5 to 6.5mm. The laser energy was between 125mJ and 175mJ with a pulse repetition rate of 30Hz.
After excimer ablation, fucidic acid (1%) was instilled, the eye padded for 24 hours and fucidic acid (1%) used twice a day until epithelialization was established (usually within 72 hours). Followup examinations were performed 2 days, 2 weeks, and every month until 6 months, and then at 1 year. The manifest subjective refraction giving the best spectacle-corrected visual was acuity was recorded at each visit. Topical medications were restricted to copious tear film supplementation from the time of cessation of the topical antibiotic in those eyes with less than -6.00 D of preoperative myopia, and fluorometholone 0.1% drops three times a day in those eyes with a preoperative spherical equivalent refraction greater than -6.00 D, for the first month in most eyes. All medications were ceased by 3 months. These 89 eyes were followed for a minimum of 12 months.
One year after photoastigmatic refractive keratectomy on 89 eyes, the mean spherical equivalent refraction was -0.44 D (SD 0.87 D). Seventy-one eyes (79.8%) had a spherical equivalent refraction within 1.00 D of the target refraction and 79 eyes (89%) achieved 6/12 or better uncorrected visual acuity. Of the 51 eyes with a preoperative spherical equivalent refraction of -6.00 D or less, 45 eyes (88.5%) were within 1.00 D of the target refraction and 26 (68.4%) of the 38 eyes with a preoperative spherical equivalent refraction greater than -6.00 D achieved this correction. A correction index gauging the effect of the spherical equivalent correction achieved can be derived by dividing the achieved spherical equivalent change by the targeted spherical equivalent change, ie, by comparing the attempted versus the achieved correction at the corneal plane. The mean value was 1.04 (SD 1.22). The ideal index would be 1, indicating a 4% trend toward overcorrection. Of the low to moderate myopes (preoperative spherical equivalent refraction of -6.00 D or less), 50 of 51 eyes (98%) achieved 6/12 or better uncorrected and of the high myopes (preoperative spherical equivalent refraction greater than -6.00 D), 29 of 38 eyes (76%) achieved 6/12 or better, uncorrected. Four of 89 eyes (4.5%) lost more than two lines of spectaclecorrected visual acuity with nine eyes (10%) gaining Snellen acuity when preoperative spectaclecorrected visual acuity was compared with postoperative uncorrected visual acuity.
The mean postoperative cylinder power was -0.36 D (SD 0.28; range 0 to -1.25 D). However, because of the variation of axis from that present before surgery, this figure needs further elucidation. Vector analysis using the Alpins method9 and the Alpins Statistical System for Ophthalmic Refractive Surgery Techniques (ASSORT) software, was carried out to evaluate the efficacy of the toric ablations. The mean magnitude of error (surgically induced astigmatism vector magnitude minus the targeted induced astigmatism vector magnitude) was -0.10 D (SD 0.27)- a small undercorrection-(the optimal value is 0, where surgically induced astigmatism vector magnitude equals targeted induced astigmatism vector magnitude). The mean angle of error (derived from the angle between the surgically induced astigmatism vector and the targeted induced astigmatism vector) was 0.73° (SD 10.91, range -61 to +24° where negative values indicate that the surgically induced astigmatism vector lies further clockwise than the targeted induced astigmatism vector and positive values if the change is further counter-clockwise). The mean difference vector magnitude, ie, the amount of dioptric correction still to be induced to reach the target, was 0.35 D (SD 0.27). Once again the optimum value is 0. The mean angle of correction (the angular separation between achieved and targeted astigmatism axes-optimum value, 0) was -0.12°(SD 46.71). The ratio of the amount of correction achieved to the amount attempted (surgically induced astigmatism magnitude divided by the targeted induced astigmatism magnitude) is the correction index. The inverse of this index is called the coefficient of adjustment. The mean coefficient of adjustment was 1.11 (SD 1.33), again indicating undercorrection. Finally, the mean index of success (the magnitude of the difference vector over the magnitude of the targeted induced astigmatism vector, ie, a measure of the proportion of the plan that remains undone- optimum value, 0) was 0.23 (SD 0.09).
Anterior stromal haze was grade 1 or less10 in all eyes at 12 months after photoastigmatic refractive keratectomy.
One year after photoastigmatic refractive keratectomy for primary myopic astigmatism, 88% of low to moderate myopes had refractive errors within 1.00 D of the target refraction and 98% had uncorrected visual acuities of 6/12 or better. High myopes achieved a refraction within 1.00 D of target in 68% of eyes with 76% achieving an uncorrected visual acuity of 6/12 or better. Translation of this into an adjustment of the spherical ablation for future surgery is made difficult by the variable effect of the hyperopic shift induced by the toric ablation. Greater astigmatism corrections will have greater effects on the spherical result. Loss of more than two lines of spectacle-corrected visual acuity occurred in 4.5% of eyes overall. Postoperative uncorrected visual acuity exceeded preoperative spectacle-corrected visual acuity in 10% of eyes, perhaps as a consequence of the reduction of corneal astigmatism.11 This refractive and visual outcome compares favourably with the reported outcome of PRK for simple myopia6,12,13 and for photoastigmatic refractive keratectomy carried out using other excimer laser instruments.6,7
We did not divide the study population into groups on the basis of the power of the preoperative cylinder (low, moderate or high errors) because there were only 9 eyes with greater than 2.00 D of astigmatism. Recent publications have relied, for many treatments, on cylinder subtraction methods to analyse the results of toric ablation, despite the limitations of such methods.9,14-16 This limits comparison of results with those presented here. All measures of vector power indicate a constant undercorrection despite the routine addition of 25% to the cylinder power for ablation. The mean coefficient of adjustment was 1.11. This is a measure of the adjustment required to improve future surgery on the basis of those eyes analyzed. It suggests that current algorithms be adjusted upward by 11% for our study eyes to overcome the undercorrection. This would be the equivalent of adding 39% to the algorithms resident in the device at present, since the eyes studied all had 25% added to the resident algorithms in the treatments they received. The demonstration of undercorrection in cylinder power and the accuracy of the axis of the correction is consistent with other publications.6,7,14,15 The mean angle of error was 0.73°. That this angle is so close to 0 implies only that there were equal numbers of clockwise and counter-clockwise vector misalignments since clockwise misalignments are given a negative value and counter-clockwise misalignments, a positive value in the Alpins method.9 This fact is important- implying no consistent error in either direction. How much variation in misalignment occurred is better expressed by the standard deviation of this variable (10.91°), still a relatively small value.
Refractive and Visual Acuity Outcomes 1 Year after Photoastigmatic Refractive Keratectomy
One eye was excluded because the patient underwent an arcuate keratotomy, 9 months after photoastigmatic refractive keratectomy, for 3.00 D of residual astigmatism (refraction; +2.25 -3.00 x 120°). Before surgery, the refractive error was -6.75 -1.5 x 120°. The induction of astigmatism following PRK has been described.12 The exclusion of this eye induces a bias in the results, but because the cylinder outcome at 1 year was altered by the procedure before that time, inclusion would be invalid. The fact that this exclusion was applied to only one eye limits the bias in a study population of 89 eyes. Furthermore, two eyes were excluded because they had mini radial keratotomy for residual spherical myopia within the first year and one had a repeat photoastigmatic refractive keratectomy for myopic astigmatism. The table contains the relevant result data calculated from only those eyes included in the study in one column and the same variables calculated from the relevant data from all eyes, including the eyes excluded from the study, with the assumption that measurements carried out immediately before the supplementary treatment would have been stable to 1 year. Variations are small.
The 1-year uncorrected visual acuity of photoastigmatic refractive keratectomy for primary myopic astigmatism is as good as that reported for PRK in eyes with little or no astigmatism and, from this perspective, seems an effective treatment for these errors.6,12,13 The alignment of the correction is adequate. However, the algorithms for the Nidek EC5000 need to be altered to allow for the consistent undercorrection of cylinder power.
1. Gartry DS, Kerr Muir MG, Marshall J. Excimer laser photorefractive keratectomy, 18 month follow-up. Ophthalmol 1992;99:1209-1219.
2. McDonald MB, Liu JC, Byrd TJ, Abdelmegheed M, Andrade HA, Klyce SD, Varnell R, Munnerlyn CR, Clapham TN, Kaufman HE. Central photorefractive keratectomy for myopia. Partially sighted and normally sighted eyes. Ophthalmol 1991;98:1327-1337.
3. McDonald MB, Franz JM, Klyce SD, Beuerman RW, Varnell R, Munnerlyn CR, Claphman TN, Salmeron B, Kaufman HE. Central photorefractive keratectomy for myopia. The blind eye study. Arch Ophthalmol 1990;108:799-808.
4. Shieh E, Moreira H, D'Arcy J, Clapham TN, McDonnell PJ. Quantitative analysis of wound healing after cylindrical and spherical excimer laser ablations. Ophthalmol 1992;99:1050-1055.
5. McDonnell PJ, Moreira H, Garbus J, Clapham TN, D'Arcy J, Munnerlyn CR. Photorefractive keratectomy to create toric ablations for correction of astigmatism. Arch Ophthalmol 1991;109:710-713.
6. Taylor HR, Guest CS, Kelly P, Alpins NA. Comparison of excimer laser treatment of astigmatism and myopia. Arch Ophthalmol 1993;111:1621-1626.
7. Kim YJ, Sohn J, Tchan H, Lee CO. Photoastigmatic refractive keratectomy in 168 eyes: six-month results. J Cataract Refract Surg 1994; 20:387-391.
8. Campos M, Hertzog L, Garbus J, Lee M, McDonnell PJ. Photorefractive keratectomy for severe postkeratoplasty astigmatism. Am J Ophthalmol 1092;114:429-436.
9. Alpins NA. A new method of analyzing vectors for changes in astigmatism. J Cataract Refract Surg 1993;19:524-533.
10. Fantes F, Hanna K, Waring GO, Pouliquen Y, Thompson K, Salvodelli M. Wound healing after excimer laser keratomileusis (photorefractive keratectomy) in monkeys. Arch Ophthalmol 1991;98:1327-1337.
11. Seiler T, Rechmann W, Maloney RK Effective spherical aberration of the cornea as a quantitative descriptor in corneal topography. J Cataract Refract Surg 1993;19:155-165.
12. Goggin M, Algawi K, O'Keefe M. Astigmatism following photorefractive keratectomy for myopia. J Refract Corneal Surg 1994;10:540-544.
13. Epstein D, Fagerholm P, Hamberg-Nystrom H, Tengroth B. Twenty four-month follow-up of laser photorefractive keratectomy for myopia. Ophthalmol 1994;101:1558-1563.
14. Gallinaro C, Toulemont PJ, Cochener B, Colin J. Excimer laser photorefractive keratectomy to correct astigmatism. J Cataract Refract Surg 1996; 22:557-563.
15. Kremer I, Gabbay U, Blumenthal M. One-year follow-up results of photorefractive keratectomy for low, moderate, and high primary astigmatism. Ophthalmol 1996;103:741-748.
16. Niles C, Culp B, Teal P. Excimer laser photorefractive keratectomy using an erodible mask to treat myopic astigmatism. J Cataract Refract Surg 1996;22:436-440.
Refractive and Visual Acuity Outcomes 1 Year after Photoastigmatic Refractive Keratectomy