Excimer laser photorefractive keratectomy (PRK) has become a popular surgical option for the treatment of myopia.1–16 Hundreds of thousands of eyes are now treated each year in the United States.
Concomitant astigmatism can be treated by concurrent photorefractive cylindrical ablation combined with the spherical correction in photorefractive astigmatic keratectomy (PARK); however, the results appear less accurate than for the correction of myopia.17–25 Reasons for the reduction in accuracy and means to improve the accuracy of combined myopic and astigmatic correction are not obvious.
The goals of our study were to provide precise information about the accuracy of PARK for defined levels of myopia and astigmatism and to better understand potential sources of inaccuracy of astigmatic ablation. We report the refractive outcome and visual results of the largest series of eyes treated with PARK published to date with in-depth analysis of the outcomes and compare them with the results in eyes treated with spherical PRK only during the study period.
Patients and Methods
Data were prospectively collected on a computerized patient-record database for consecutive patients at Optimax Laser Eye Clinics from 1994 to 1996. The subjective refraction was recorded in minus cylinder form. A plus-1 blur test was used in each case. The attempted correction was not emmetropia in every eye; for analysis purposes, the magnitude of the attempted correction was used to calculate results. No data for retreatment were included. (Eyes requiring retreatment were not retreated for a minimum of 12 months after the initial treatment.) Autorefraction, keratometry, and corneal topography were performed in all eyes.
Each eye was categorized by the mean spherical equivalent (MSE) and the cylindrical component of the refraction. For the cylindrical component, eyes were grouped in 0.50 diopter (D) intervals (–0.50 D pre-operative cylinder group, eyes with –0.50 D and –0.75 D of cylinder; –1.00 D group, eyes with −1.00 D and −1.25 D of cylinder, etc.) or 1.00 D intervals if the analysis was by preoperative cylinder and MSE to simplify representation (–1.00 D preoperative cylinder group, eyes with –0.50 to –1.50 D of cylinder; –2.00 D group, eyes with −1.75 to –2.50 D of cylinder, etc.). The eyes were similarly split into MSE groups.
In each eye, the refraction, MSE, uncorrected visual acuity (UCVA), and best corrected visual acuity (BCVA) (Snellen fractions) prior to PRK and at the most recent follow-up visit were assessed. The presence of corneal haze was recorded at the latest follow-up. Corneal haze was assessed by slitlamp examination and recorded on a qualitative scale: 0 = clear cornea; 1 = mild haze not interfering with refraction; 2 = moderate haze; 3 = marked haze obscuring iris detail. Photographs of haze were available to aid grading. Due to the large series, refraction and assessment were done by a group of optometrists and surgeons.
Photoastigmatic refractive keratectomy was performed using the Nidek EC-5000 scanning excimer laser. The astigmatic component was sequential to the spherical component; ie, the sphere treated first. The magnitude of treatment for the spherical and astigmatic components were based on manifest refraction. The cylindrical axis was determined on subjective refraction only, and for the purposes of this study, the axis was not marked on the eye as no definitive advantage to marking the eye has been shown. Informed consent was obtained.
The optical zone was 6.5 mm, with an additional 1.0 mm transition zone in all cases. Manual epithelial debridement was carried out under topical anesthesia. Fixation was maintained on a fixation light. Postoperatively, a topical antibiotic agent (chloramphenicol) was used routinely and topical nonsteroidal antiinflammatory drugs (diclofenac or ketorolac) were used by some doctors for 48 hours. Topical steroids (fluorometholone) were reserved for patients in whom myopic regression or excessive haze formation was observed. Bandage contact lenses were not used.
Data analysis was performed with SPSS for Windows 8.0. Vector analysis was performed using Alpins technique.26 In this technique, the target induced astigmatism (TIA) is the planned astigmatic treatment; the surgically induced astigmatism (SIA), the vector of the astigmatic change that results from surgery; the angle of error, a measure of the misalignment of the astigmatism treatment; the correction index (CI), the proportion of astigmatism treatment achieved (SIA/TIA); and the index of success (IS), a relative measure of the success of astigmatism surgery (DV/TIA), in which DV is the magnitude of the vector of the untreated astigmatism. An ideal CI is 1.0, and an ideal IS is 0.
Eyes in patients that had a 12-month (52-week) follow-up were included (n = 6418 for PARK and 3249 for PRK); 321 and 245 eyes, respectively, did not complete a 1-year follow-up. Data from 6097 eyes that had PARK and for myopia with astigmatism and 3004 that had PRK were available for analysis. The mean patient age was 35.2 years ± 9.0 (SD) (range 21 to 76 years). The preoperative cylinder groups are shown in Table 1. The mean follow-up was 84 weeks (range 52 to 160 weeks).
Uncorrected Visual Acuity
A UCVA of 20/40 or better was achieved by 5560 PARK eyes (91.2%) (94.3% of PRK eyes) and of 20/20 or better by 2544 PARK eyes (41.7%) (59.1% of PRK eyes). The UCVA results are presented in Figures 1 to 3. The visual results depended on the amount of preoperative cylinder; they were good in the low-cylinder groups and decreased as the cylinder increased. The visual results also decreased as the preoperative MSE increased.
The preoperative MSE of the 6097 PARK eyes was −4.63 ± 1.96 D, with a mean sphere of –4.06 ± 1.91 D and a mean cylinder of –1.12 ± 0.73 D. The final MSE, measured at the most recent follow-up examination, was –0.02 ± 0.79 D, with a mean sphere of +0.23 ± 0.81 D and a mean cylinder of –0.49 ± 0.47 D. After treatment, 69.8% of eyes had a final SE within ±0.50 D of emmetropia and 87.9%, within ±1.00 D. The post-PARK refractive results are shown in Figures 4 to 6. The spherical results decreased steadily as the preoperative MSE increased. However, the higher cylinder groups had a relatively higher hyperopic result and increased regression (particularly the –4.0 D cylinder group in which the highest spherical result was +2.0 D and the lowest was –1.9 D) (Figure 5). The postoperative cylinders were undercorrected, with higher undercorrection in the higher cylinder groups (Figure 6).
The preoperative MSE in the 3004 PRK eyes was −3.74 ± 1.61 D and the postoperative MSE, −0.07 ± 0.66 D. After treatment, 73.2% of eyes had a final SE within ±0.50 D of emmetropia and 90.7%, within ±1.00 D.
Visual Loss and Haze
Visual loss and haze increased gradually as preoperative cylinder increased. However, the 2 were not directly linked (Figures 7 and 8). The mean haze varied from 0.23 in eyes with no cylinder, to 0.32 in eyes with 2.0 to 2.5 D cylinder, up to 0.44 in eyes with 5.0 D cylinder.
The outcomes for MSE, sphere, cylinder, haze, and BCVA were worse in the higher than the lower cylinders.
Statistical significance (P < .001, analysis of variation) was achieved for MSE, sphere, cylinder, haze, and visual acuity (BCVA and UCVA) based on the preoperative cylinder; ie, the outcomes for the above factors were highly dependent on the preoperative cylinder.
The mean induced vector (SIA) was consistently smaller than the TIA (Table 2). This was probably due to nomogram problems. The mean axis error was also relatively large and contributed considerably to undercorrection of the cylinder. The CI remained between 0.50 and 0.73; ie, a 27% to 50% undercorrection. In keeping with this, the mean IS ranged from 0.05 to 0.37.
This was the largest study to date of PARK. The large population enabled group analysis by refractive error in 0.50 D intervals of cylinder to evaluate the effect of the astigmatic correction on the final outcome. Conceptually, the addition of astigmatic correction could result in inaccuracy not only in the astigmatic component of the final refraction but also in the SE of the final correction. This study was designed to assess these outcomes.
Performing spherical PRK in the presence of preexisting astigmatism induces 0.47 D of astigmatism per diopter of preoperative cylinder and 0.04 D per diopter of spherical treatment.27 Thus, the trend in this series of patients was to treat even low levels of astigmatism such as 0.50 D. The poor accuracy of low cylindrical ablations may be partly because of the induced astigmatism seen in PRK as a result of decentration or uneven ablations.
The accuracy of PARK varies from study to study. Previous studies17–25 have found PARK effective but not as accurate as PRK for the same degree of myopia without astigmatism. We found that at low levels of astigmatism, the addition of astigmatic correction did not significantly affect the accuracy of the spherical correction or the percentage of patients achieving 20/40 and 20/20 visual acuity. With low spherical myopia, we have shown that PRK outcomes28 are similar to PARK outcomes; this was demonstrated here unless the cylinder was high. For example, 93% in the –1.0 D cylinder group achieved a visual acuity of 20/40.
Our results showed a steady decline in the percentage of patients achieving 20/20 and 20/40 UCVA as the preoperative cylinder increased. There was a particularly rapid decline after −3.5 D of preoperative cylinder in eyes achieving 20/40 or better. The MSE remained relatively steady across all preoperative cylinders. However, this hid the tendency for the sphere to be overcorrected and the cylinder to be undercorrected, a tendency that increased as the cylinder increased.
Our results showed a gradual increase in visual loss with increasing preoperative cylinder, as well as a concomitant increase in haze. With a 2-Snellen-line loss, the percentage appeared to rise exponentially. There was a linear rise in mean haze in relation to the preoperative cylinder. The increased levels of haze contributed to the increased loss of BCVA, but this was probably partly due to increased levels of irregular astigmatism.
The absolute magnitude of axis error was used to calculate the vector analysis figures; other articles do not comment on whether positive and negative axes were simply added together. The axis error had a tendency to decrease as the cylinder size increases, implying that this is due to errors in preoperative refractions, postoperative refractions, or both as the axis is easier to determine accurately with higher magnitudes of cylinder. However, factors such as torsion of the eye on lying down, head tilt, and differential healing may also contribute to this. Many laser manufacturers recommend marking the axis on the eye with the patient sitting up so the axis alignment is more accurate; however, this is not the recommended technique for Nidek lasers. The discrepancy between the anterior corneal cylindrical axis and the subjective axis may have some bearing on the outcome of PARK. (The tendency worldwide still appears to be to treat the subjective cylinder even if the anterior corneal cylinder magnitude and axis are quite different from the subjective results.)
With the Nidek EC-5000 scanning excimer laser, we would suggest caution in treating eyes with more than 4.0 D of astigmatism (PARK) based on the poor refractive results, increased haze, and in particular the increased loss of BCVA. This agrees with the current U.S. Food and Drug Administration approved uses of other lasers and of the Nidek EC-5000.
Our results are comparable to those in other studies. They indicate that the outcome of PARK in myopia is predictable and associated with a low risk of BCVA loss. We think it is important to offer patients accurate prognostic information regarding the potential benefits and risks of PARK relative to their preoperative myopia. Further research is required to improve results.
1. Carones F, Brancato R, Morico A, et al. Evaluation of three different approaches to perform excimer laser photorefractive keratectomy for myopia. Ophthalmic Surg Lasers 1996; 27(suppl):S458-S465
2. Aron-Rosa DS, Colin J, Aron B, et al. Clinical results of excimer laser photorefractive keratectomy: a multicenter study of 265 eyes. J Cataract Refract Surg 1995; 21:644-652
3. Kalski RS, Sutton G, Bin Y, et al. Comparison of 5-mm and 6-mm ablation zones in photorefractive keratectomy for myopia. J Refract Surg 1996; 12:61-67
4. Talley AR, Hardten DR, Sher NA, et al. Results one year after using the 193-nm excimer laser for photorefractive keratectomy in mild to moderate myopia. Am J Ophthalmol 1994; 118:304-311
5. El-Maghraby A, Salah T, Polit F, et al. Efficacy and safety of excimer laser photorefractive keratectomy and radial keratotomy for bilateral myopia. J Cataract Refract Surg 1996; 22:51-58
6. Seiler T, Wollensak J. Myopic photorefractive keratectomy with excimer laser; one-year follow-up. Ophthalmology 1991; 98:1156-1163
7. Maguen E, Salz JJ, Nesburn AB, et al. Results of excimer laser photorefractive keratectomy for the correction of myopia. Ophthalmology 1994; 101:1548-1556; discussion by MJ Mannis, 1556
8. Shah SI, Hersh PS. Photorefractive keratectomy for myopia with a 6-mm beam diameter. J Refract Surg 1996; 12:341-346
9. Hersh PS, Schein OD, Steinert R. Characteristics influencing outcomes of excimer laser photorefractive keratectomy; the Summit Photorefractive Keratectomy Phase III Study Group. Ophthalmology 1996; 103:1962-1969
10. Tengroth B, Epstein D, Fagerholm P, et al. Excimer laser photorefractive keratectomy for myopia; clinical results in sighted eyes. Ophthalmology 1993; 100:739-745
11. Epstein D, Fagerholm P, Hamberg-Nyström H, Tengroth B. Twenty-four-month follow-up of excimer laser photorefractive keratectomy for myopia; refractive and visual acuity results. Ophthalmology 1994; 101:1558-1563; discussion by AB Nesburg, 1563-1564
12. Heitzmann J, Binder PS, Kassar BS, Nordan LT. The correction of high myopia using the excimer laser. Arch Ophthalmol 1993; 111:1627-1634
13. Menezo JL, Martinez-Costa R, Navea A, et al. Excimer laser photorefractive keratectomy for high myopia. J Cataract Refract Surg 1995; 21:393-397
14. Sher NA, Hardten DR, Fundingsland B, et al. 193-nm excimer photorefractive keratectomy in high myopia. Ophthalmology 1994; 101:1575-1582
15. Chatterjee A, Shah S, Bessant DA, Doyle SJ. Results of excimer laser retreatment of residual myopia after previous photorefractive keratectomy. Ophthalmology 1997; 104:1321-1326
16. Palliarkis IG, Siganos DS. Excimer laser in situ keratomileusis and photorefractive keratectomy for correction of high myopia. J Refract Corneal Surg 1994; 10:498-510
17. Vidaurri-Leal JS, Helena MC, Talamo JH, et al. Excimer photorefractive keratectomy for low myopia and astigmatism with the Coherent-Schwind Keratom. J Cataract Refract Surg 1996; 22:1052-1061
18. Gallinaro C, Toulemont PJ, Cochener B, Colin J. Excimer laser photorefractive keratectomy to correct astigmatism. J Cataract Refract Surg 1996; 22:557-563
19. Kaskaloglu M. Early results of photorefractive keratectomy for myopic astigmatism. J Refract Surg 1996; 12:S258-S260
20. Snibson GR, Carson CA, Aldred GF, Taylor HR. One-year evaluation of excimer laser photorefractive keratectomy for myopia and myopic astigmatism; the Melbourne Excimer Laser Group. Arch Ophthalmol 1995; 113:994-1000
21. Goggin MJ, Kenna PF, Lavery FL. Photoastigmatic refractive keratectomy for compound myopic astigmatism with a Nidek laser. J Refract Surg 1997; 13:162-166
22. Vinciguerra P, Radice P, Azzolini M. Double clear zone photorefractive keratectomy to correct compound myopic astigmatism. J Refract Surg 1997; 13:S450
23. 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
24. Kremer I, Gabbay U, Blumenthal M. One-year follow-up results of photorefractive keratectomy for low, moderate, and high primary astigmatism. Ophthalmology 1996; 103:741-748
25. Zadok D, Haviv D, Vishnevskia-Dai V, et al. Excimer laser photoastigmatic refractive keratectomy; eighteen-month follow-up. Ophthalmology 1998; 105:620-623
26. Alpins NA. A new method of analyzing vectors for changes in astigmatism. J Cataract Refract Surg 1993; 19:524-533
27. Shah S, Chatterjee A, Doyle SJ, Bessant DAR. Astigmatism induced by spherical photorefractive keratectomy corrections. Ophthalmology 1997; 104:1317-1320
28. Shah S, Chatterjee A, Smith RJ. Predictability of spherical photorefractive keratectomy for myopia. Ophthalmology 1998; 105:2178-2184; discussion by PS Binder, 2184-2185