* Summit PRK Phase III Study Group:
Atlanta, GA: George O. Waring, MD, R. Doyle Stulting, MD, PhD, Keith Thompson, MD
Birmingham, AL: Marc Michelson, MD, John Owen, MD
Boston, MA: Roger Steinert, MD, Carmen Puliafito, MD, Michael Raizman, MD
Colorado Springs, CO: John R. Wright, DO
Galveston, TX: Daniel H. Gold, MD, Bernard A. Milstein, MD
Kansas City, MO: Daniel Durrie, MD, Timothy Cavanaugh, MD, John Hunkeler, MD
New Orleans, LA: Stephen Brint, MD
San Diego, CA: Michael Gordon, MD
St. Louis, MO: Jay Pepose, MD, PhD
Tulsa, OK: J. Harley Galusha, DO
From the Department of Ophthalmology, UMDNJ-New Jersey Medical School, Newark, New Jersey and the Cornea and Laser Institute, Hackensack University Medical Center, Hackensack, New Jersey1; the Wilmer Ophthalmologic Institute, Johns Hopkins University School of Medicine, Baltimore, Maryland2; and Ophthalmic Consultants of Boston, Boston, Massachusetts3.
Presented in part at the American Academy of Ophthalmology Annual Meeting, November, 1995.
Supported in part by an unrestricted grant to the Department of Ophthalmology from Research to Prevent Blindness, Inc., New York and Summit Technology, Inc., Waltham, MA. Drs. Hersh, Steinert, and Schein are each consultants for Summit Technology, Inc.
Abstract
Purpose: To identify preoperative and intraoperative characteristics associated with outcomes of photorefractive keratectomy (PRK).
Methods: In the Phase III multicenter clinical trials of the Summit Technology excimer laser for corrections of 1.5 to 6.0 D of myopia, 3 principal outcomes of PRK on 612 patients were examined: (1) uncorrected visual acuity (UCVA) of 20/40 or better, (2) predictability of refractive outcome within 1.0 D of attempted correction, (3) stability of refractive result between 12 and 24 months. Multiple logistic regression was used to test for independent associations of multiple pre- and intraoperative characteristics with each of these outcomes.
Results: Older age was independently associated with lesser likelihood of achieving 20/40 or better uncorrected vision (odds ratio = 1.08 per incremental year of age, 95% CI = 1.04-1.12) and with decreased predictability, specifically with overcorrection (odds ratio = 1.09, 95% CI = 1.06-1.12), but age was not associated with stability of refraction. Greater attempted correction was independently associated with a decreased likelihood of 20/40 or better postoperative vision (odds ratio = 2.78 for corrections of 3.5 to 5.5 diopters, 95% CI = 1.18-6.75; odds ratio = 4.19 for corrections of 5.5 diopters or more, 95% CI = 1.66-10.58), with decreased predictability (odds ratio = 1.72 for corrections of 3.5 to 5.5 diopters, 95% CI = 1.05-2.85; odds ratio = 2.95 for corrections of 5.5 diopters or more, 95% CI = 1.65-5.26), and with a reduced likelihood of stability of refraction (odds ratio = 3.46 for corrections of 5.0 diopters or more, 95% CI = 1.32-9.11). No intraoperative characteristics were associated with any of the outcomes assessed.
Conclusions: Using this specific excimer laser system with an optical zone of 4.5 or 5.0 mm, patient age and attempted correction are important preoperative characteristics associated with postoperative uncorrected visual acuity and predictability of PRK; stability of refraction is strongly associated with attempted correction. Such information may help guide patient selection, determine timing of fellow eye treatment, and suggest changes in the laser treatment algorithm for individual patients. Although these findings may be representative of PRK in general, similar analyses should be performed before modifying patient treatments using a 6.0 mm treatment zone or other laser systems.
Introduction
Excimer laser photorefractive keratectomy (PRK) is a promising modality for the correction of myopia and has recently been approved by the U.S. Food and Drug Administration (1-5). To date, procedures have been performed based on theoretical optics and mathematical algorithms designed to remove a specified tissue lens from the corneal surface (6,7). This has been done without regard to individual patient characteristics which might have an influence on the outcomes of the procedure. In the future, it is likely that biologic and wound healing differences amongst patients will be identified and incorporated into individual PRK treatments.
Clinically, the influence of operative and perioperative variables on results of PRK are, as yet, unclear. In this multicenter, Phase III study of PRK using the Summit Technology excimer laser with a 4.5 or 5.0 mm beam diameter, we have analyzed associations of preoperative and operative characteristics with defined outcomes of PRK. Such associations should aid in elucidating factors predictive of either good or poor outcomes following the excimer laser procedure. This knowledge should help to improve patient selection, suggest appropriate variations in laser treatment algorithms, and guide postoperative care strategies for the individual patient.
Patients and Methods
Study Design
As part of the Phase III multicenter clinical study of the Summit Technology, Inc. excimer laser (Waltham, MA) conducted in accordance with U.S. Food and Drug Administration regulations, an observational study of 701 patients was performed to assess the safety and efficacy of photorefractive keratectomy for the treatment of myopia. The treatments were performed at ten clinical centers between June, 1991 and February, 1992. All patients entered in the study had between -1.25 and -7.25 diopters of myopia (spherical equivalent) and less than or equal to 1.00 diopter of regular astigmatism. Attempted corrections ranged from 1.50 to 6.00 diopters with a mean of 4.24 diopters. Data analyzed in this study was from the 2 year followup visit; this data was available on 612 eyes of 612 patients. Only the first eye treated of each patient was included in the analysis.
All study centers conformed to standardized patient entry criteria under an Investigational Device Exemption granted to Summit Technology, Inc. Approvals from appropriate Institutional Review Boards were obtained. Preoperative and followup visits included a detailed ophthalmologic examination with both manifest and cycloplegic refractions by two independent observers, visual acuity under controlled lighting conditions by trained technicians using an ETDRS chart (Lighthouse for the Blind, New York), manual keratometry, glare testing with the Brightness Acuity Tester (Mentor, Norwell, Mass), contrast sensitivity (VectorVision, Dayton, OH), videokeratography (EyeSys Laboratories, Houston, Texas), and anterior segment photography.
Photorefractive Keratectomy Procedure
Photorefractive keratectomy was performed with the Summit ExciMed( UV200LA excimer laser system. Laser parameters included a repetition rate of 10 Hertz, radiant exposure at the corneal plane of 180 millijoules/cm2, and pulse duration of 14 nanoseconds, resulting in an ablation rate of corneal stromal tissue of approximately 0.25 microns per pulse. Treatment zone size was 4.5 mm in 251 (35.8%) and 5.0 mm in 450 (64.2%) cases.
The standardized photorefractive keratectomy procedure used has been published previously (8). In brief, to ensure appropriate laser energy and beam homogeneity, ablation and beam profile characteristics were tested at the beginning of each treatment day by the rate and pattern of ablation of a 100 micron thick gelatin filter (Kodak #1497890, Rochester, NY) and standardized ablations on a polymethylmethacrylate disc. The operative eye received pilocarpine 1% and topical anesthetic drops. The ablation was centered over the entrance pupil as suggested by Uozato and Guyton (9). The centration procedure is described in detail elsewhere (10).
Two training sessions were performed to familiarize the patient with the procedure and to ensure proper fixation subsequently. The first training session involved the application of methylcellulose 1% to the cornea before ablation to block the incoming laser beam. The second session was performed on dry epithelium using 25 pulses of the laser at its maximum aperture. The optical zone was then marked around the entrance pupil with a 6.0 mm optical zone marker, the epithelium within this area was removed with a microsurgical blade, and the laser ablation was performed. In general, the manifest refraction spherical equivalent at the spectacle plane was programmed into the laser, although differences of up to +/- 1.0 diopter from this amount were allowed at the surgeon's discretion. In these cases, the intended correction was entered.
Postoperatively, a combination antibiotic-steroid ointment and a patch were applied. The ointment was continued up to 5 times daily until the cornea had reepithelialized. Fluorometholone 1.0% drops were then administered five times daily for 1 month. The dosage was then reduced to fluorometholone 0.1% drops four times daily for one month and then three times daily for one month. Corticosteroid dosage after the completion of the third month was at the discretion of the individual surgeon based on the patient's refraction and degree of corneal haze.
Data Analysis
Definitions of Principal Outcomes: Outcomes studied included (1) uncorrected visual acuity (UCVA) of 20/40 or better at 2 years; (2) predictability of postoperative refraction within 1 diopter of attempted correction at 2 years; (3) stability of the postoperative refraction, assessed by comparing the refractive spherical equivalent at 12 versus 24 months and considering a refraction to be stable if there was 1 diopter or less change in spherical equivalent between these 2 visits. All visual acuity measurements were reported on the LogMAR scale (10).
Characteristics Assessed: Preoperative patient characteristics and operative variables were analyzed for possible associations with each of the principal outcomes. The patient characteristics studied included age, gender, medical history, biomicroscopic findings of the anterior segment, measures of visual acuity, refraction, astigmatism, average corneal power as assessed by mean keratometry readings, and intraocular pressure. Operative characteristics analyzed included the attempted correction, optical zone size, treatment center, total PRK time, and the time from epithelial removal to laser treatment.
Statistical Analysis: Data were entered from uniform study forms submitted from each investigational site. Bivariate analyses were initially performed to test for individual associations between the preoperative and operative characteristics and the outcomes measured. Contingency tables were constructed for categorical variables. For continuous variables, mean values were compared across groups. A significance level of 0.05 was used for subsequent inclusion in the multivariate models. Multivariate models were constructed using the variables found to be significant in the bivariate analyses (p<0.05) and additional variables thought to be of demographic or clinical significance such as treatment center or optical zone size. Odds ratios, indicating the strength of the independent association were calculated and are presented with their 95% confidence intervals. Statistical analysis was performed using the Statistical Analysis System 6.07 (SAS Institute, Inc., Cary, NC).
RESULTS
A total of 701 patients entered the study cohort. Clinical data were available on 612 (87%) at 2 years following PRK. Comparing those lost to follow-up with the study cohort, there were no differences at baseline in age, gender, or manifest spherical equivalent. The mean age of the 612 patients followed was 38 years, and 55% were male. Treatment zone size was 4.5 mm in 198 (32.3%) and 5.0 mm in 414 (67.6%) of the 612 cases.
Characteristics Associated with Uncorrected Visual Acuity
Overall, for the 612 patients analyzed in this study, 563 (92%) achieved a postoperative uncorrected visual acuity (UCVA) of 20/40 or better at the 2 year followup visit. Bivariate analyses indicated possible associations of uncorrected acuity of 20/40 or better with preoperative age, contact lens use, refractive cylinder, attempted correction, intraocular pressure and total PRK time. Table 1 presents the multiple logistic regression model of preoperative and operative characteristics associated with uncorrected visual acuity worse than 20/40 at the 2 year follow up. This analysis indicates independent associations of age, attempted correction, and intraocular pressure.
The effect of age was large; an individual's likelihood of having uncorrected visual acuity of 20/40 or better decreases by approximately 8% with every additional year of age. For instance, over 95% of patients between the ages of 21 and 40 years achieved 20/40 or better compared with 76% of those between 51 and 60 (Table 2).
Similarly, attempted correction (ie the correction programmed into the laser) was found to be associated with lesser likelihood of achieving 20/40 or better uncorrected visual acuity. Table 3 shows patients stratified to one-half diopter increments of attempted correction. As can be seen, patients with lesser amounts of attempted correction fared the best. Compared to those with attempted correction of 2.5 to 3.5 diopters, those with attempted corrections of 3.5 to 5.5 diopters had approximately a three-fold greater likelihood of not achieving 20/40 or better uncorrected visual acuity and those with corrections of >5.5D had over a four-fold greater risk of having uncorrected vision worse than 20/40.
Conversely, those patients with higher intraocular pressures were less likely to have an uncorrected visual acuity worse than 20/40. No independent effects of other preoperative and operative characteristics including optical zone size, study center, and total PRK time were seen.
Characteristics Associated with Procedure Predictability
Overall, for the 612 patients analyzed in this study, 78% achieved a refractive change within one diopter of the attempted correction at the 2 year followup visit. Bivariate analyses indicated possible associations of a number of preoperative and operative characteristics including patient age, attempted correction, study center, refractive astigmatism, optical zone size, presence of lens opacities, and the time from the start of epithelium removal to laser ablation.
Table 4 shows the multiple logistic regression model of preoperative and operative characteristics associated with predictability outside +/- 1.0 diopter at 2 year followup. As with the analysis of uncorrected visual acuity, significant and comparable independent effects were seen for age and attempted correction. An independent association was also seen for iris color; individuals with light irides were found to have less predictable outcomes. No effect was seen for intraocular pressure, optical zone size, PRK time, or study center.
With each one year increment in age, there was an approximately 5% decreased likelihood of achieving within one diopter of attempted correction. A second regression analysis was performed, separating those individuals who were overcorrected from those undercorrected. This analysis indicated that age was associated specifically with overcorrections (but not undercorrections) of more than one diopter (odds ratio = 1.09 for each yearly increment, 95% CI = 1.06-1.12). Table 5 illustrates predictability stratified by decade of life, demonstrating the association of increasing age with lesser predictability as well as overcorrection. For example, over 82% of patients between the ages of 21 and 40 years achieved within one diopter compared with 59% of those between 51 and 60.
Table 6 shows predictability stratified to one-half diopter increments of attempted correction . As can be seen, proportionately fewer patients with higher attempted corrections achieved this success outcome. Compared to those with attempted correction of 2.5 to 3.5 diopters, those with attempted corrections of 3.5 to 5.5 diopters had approximately a two-fold greater likelihood of not achieving within +/- 1.0 diopter of attempted correction and those with corrections of >5.5D had nearly a three-fold greater risk of not achieving within one diopter of attempted correction (Table 4). When the analysis was repeated separating those under and overcorrected, the association between attempted correction and predictability was found to exist specifically for undercorrections (but not overcorrections) of more than 1.0 diopter. Compared to those with attempted corrections of 2.5 to 3.5 diopters, those with attempted corrections of 3.5 to 5.5 diopters had 3.32 times (95% CI = 1.42 - 7.73) greater odds of being undercorrected by one diopter or more; those with attempted correction of 5.5 or more diopters had a 7.88-fold (95%CI = 3.22-19.24) excess risk of being undercorrected.
Characteristics Associated with Stability of Refraction
Table 7 shows the proportion of patients stable within one diopter between 12 and 24 months, stratified by preoperative spherical equivalent. Over this time interval, 89.9% of patients overall were stable within 1.0 diopter. There appeared to be a greater tendency toward myopic shifts with higher degrees of correction. Table 8 shows the multiple logistic regression model of characteristics associated with changes of more than 1.0 diopter between the 12 and 24 month examinations. There was no significant effect of age on stability. A large independent effect was seen for attempted correction; higher refractive corrections were associated with less likelihood of refractive stability. Those patients with attempted corrections of 5.0 diopters or greater were over 3 times more likely to be less stable (within +/- 1.0D) between 1 and 2 years.
One study center was found to have a lesser likelihood of refractive stability. This center was not associated with any variation in either predictability or the proportion of patients achieving uncorrected visual acuity of 20/40 or better. Otherwise, no independent effects of optical zone size, or other characteristics could be demonstrated.
Discussion
To date, patients undergoing excimer laser photorefractive keratectomy have generally received treatment based solely on refractive error without regard to individual patient characteristics. This "one-size-fits-all" approach has, thus far, led to encouraging results in numerous clinical trials (1-4,12-14). The identification of pre- and intraoperative characteristics influencing outcomes of PRK,however, should improve future results both by clarifying patient selection and suggesting changes in treatment algorithms. In our study of 612 patients in a rigorous multicenter trial, we have identified a number of preoperative and intraoperative characteristics independently associated with specific important clinical outcomes of PRK.
Effect of Patient Age
In this study, we found that older patients had a decreased chance of achieving 20/40 or better uncorrected vision compared with younger patients; each one-year increment in age was associated with a decreased likelihood of achieving >20/40 of approximately 8%. Older patients were also less likely to achieve within 1.0 diopter of attempted correction and were more likely to be overcorrected. Regarding this latter finding, it is possible that accomodation in younger patients may have augmented the age differences found in this study.
Photorefractive keratectomy in general may be associated with an early overcorrection which diminishes over time with expected stromal wound healing (13,15-17). In the older patient, therefore, the relative tendency toward overcorrection could be secondary to a diminished stromal wound healing response, with more patients retaining the initial overcorrection seen in the early postoperative period. This may be analogous to the increased correction seen in older radial keratotomy patients, where diminished healing and contraction of the incisions leads to a greater refractive response to surgery (18). Age, in contrast, was not associated with stability of refractive correction from the 1 year to the 2 year examination. This implies that the interaction between age and wound healing which causes a tendency toward undercorrection occurs in the first year; the refraction thereafter remains equally stable as for the younger patient.
Given this finding of increased overcorrection in older patients, the decreased likelihood of uncorrected vision or 20/40 or better found in older patients could, in part, be a result of the inability of presbyopic patients who are hyperopic postoperatively to focus to good distance acuity. Moreover, the novel optics of the cornea resulting from the change in normal asphericity and topography induced by PRK may interact with other aging changes of the eye, thus potentially resulting in the relationship of age to uncorrected vision found -- optical aberrations following PRK could possibly decrease visual acuity in the aging eye more than in the young eye (20). Actual aging changes such as cataract and age-related macular degeneration, however, were not found on examination of the patients enrolled in this study. Furthermore, no age effect upon topography classification or corneal surface irregularity was found in previous studies of a subset of this patient cohort (8, 21). Such interactions between the optics of the post-PRK cornea and the aging eye thus remain speculative.
Effect of Attempted Correction
Attempted refractive correction also was found to be independently associated with postoperative uncorrected visual acuity; patients with higher attempted corrections had less likelihood of achieving 20/40 or better vision. This finding may be related to the greater likelihood of undercorrection with higher attempted corrections. Moreover, other sequelae causing optical aberrations, such as treatment zone edge effects (21,22), decentration (10), and corneal asphericity (19,23), also could be increased with higher corrections and possibly lead to a diminished likelihood of good uncorrected vision following PRK.
While, in general, there was less likelihood of achieving within 1.0 diopter of attempted correction with higher attempted corrections, patients undergoing higher attempted corrections were specifically more likely to be undercorrected by more than 1.0 diopter. For patients in whom 5.5 or more diopters of correction were attempted, the odds of being undercorrected were nearly 8-fold greater than patients in whom attempted correction was less than 3.5 diopters; over 20% of patients with attempted corrections >5.5 diopters were undercorrected by more than 1.0 diopter.
Since the algorithm governing ablation of a spherical tissue lens by the excimer laser is theoretically correct (6,7), explanations for these findings must rest either in the laser-tissue interaction itself (24-26) or subsequent corneal wound healing (27). The laser-tissue interaction may be affected by corneal hydration during the procedure, possibly with increases in stromal hydration secondary to shockwave phenomena during the procedure leading to a decreasing ablation rate as lasing proceeds. In particular, shockwave phenomenon could drive water into the ablation zone, thus decreasing the ablation rate and removing less stromal tissue than anticipated, with a consequent tendency toward
undercorrection (8,28,29). More likely, corneal epithelial or stromal wound healing may play a role in the decreased predictability of patients undergoing higher refractive ablations. Relatively increased wound healing and stromal remodelling with higher corrections could lead to greater "fill-in" of the ablation zone with resulting decreased achieved correction. Patients in this trial had optical zones of 4.5 or 5.0 mm. Larger optical zones or bevelled transition zones might moderate corneal healing and lead to increased predictability for higher attempted corrections (11,29).
Postoperative refractive stability was inversely associated with the patient's refractive error. Patients with myopia of 5.0 or more diopters were over three-fold more likely to have a change of refraction of more than 1.0 diopter between the one and two year examinations. Whereas 93.5% to 100% of patients with a preoperative myopia of less than 4.0 diopters were stable over this one year period, only 80% to 89.4% of those with >4.0 diopters were found to meet the stability outcome. Again, corneal wound healing is an important factor in stability; those patients with greater ablations may undergo more substantial and lengthy wound remodeling (27). Clinically, this may help to determine the timing of treatment of the fellow eye. To allow the first eye to stabilize, a longer time lag between treatments should be planned for eyes undergoing high attempted corrections.
Other Factors Associated with Outcomes of PRK
Treatment zone diameters in this study were either 4.5 or 5.0 mm. While size of the treatment zone was not associated with postoperative uncorrected visual acuity or overall predictability, patients with the smaller diameter were more likely to be undercorrected. One possible explanation is that the distance from the edge of the ablation to the center of the optical zone modulates overall wound healing, larger zones tending to regress less than larger zones (30).
No center effect was found for the outcomes analyzed in this study although one center did show less refractive stability than the others. This one finding may possibly be associated with measurement error since no variation in any of the other principal outcomes was found. All surgeons received similar training and conformed to standard operative and perioperative protocols. When PRK is in more widespread use at multiple sites with numerous lasers and surgeons, it will be important to reexamine the possibility of an effect of treatment center on outcomes.
The tendency of higher intraocular pressure to be independently associated with lesser odds of not achieving 20/40 or better uncorrected visual acuity is difficult to explain. Similarly, the finding of an effect of iris color on predictability was unanticipated. Finally, the finding that average preoperative keratometry readings (ie average corneal power) did not influence outcomes of best visual acuity or predictability was expected based on theoretical optics of the PRK procedure (7, 23).
Conclusions
PRK algorithms may need to be modified to account for predictive characteristics such as age and attempted correction. For instance, using the excimer laser system in this study, procedure predictability probably could have been improved by intentionally undercorrecting older patients. Moreover, increasing the nominal attempted correction in the case of higher desired corrections, especially for younger patients may improve the results in this subset of patients. It must be emphasized that the specific results presented in this study can be attributed only to the specific laser system, configuration, and optical zone diameter used for these patient treatments. Although our findings may be representative of PRK in general, similar analyses should be performed before modifying patient treatments using a 6.0 mm treatment zone or other laser systems. With this in mind, however, knowledge of the influence of individual patient and surgical variables should be an important addition to patient treatments as the photorefractive keratectomy procedure continues to develop.
References
1. Waring GO, O'Connell MA, Maloney RK, Hagen KB, Brint SF, Durrie DS, Gordon MG, Steinert RF. Photorefractive keratectomy for myopia using a 4.5 mm ablation zone. J Refractive Surgery 1995; 11:170-180.
2. Maguen E, Salz JJ, Nesburn AB, et al. Results of excimer laser photorefractive keratectomy for the correction of myopia. Ophthalmology 1994;101(9):1548-56.
3. 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(3):304-11.
4. Dutt S, Steinert RF, Raizman MB, Puliafito CA. One year results of excimer laser photorefractive keratectomy for low to moderate myopia. Arch Ophthalmol 1994;112:1427-1436.
5. Trokel SL, Srinivasan R, Braren B. Excimer laser surgery of the cornea. Am J Ophthalmol 1983;96:710.
6. Munnerlyn CR, Koons SJ, Marshall J. Photorefractive keratectomy: A technique for laser refractive surgery. J Cataract Refract Surg. 1988;14:46-52.
7. Blaker JW & Hersh PS. Theoretical and clinical effect of corneal curvature on excimer laser photorefractive keratectomy. Refractive Corneal Surg. 1994;10:571-574.
8. Hersh PS, Schwartz-Goldstein BH, Summit PRK Topography Study Group. Corneal topography of Phase III excimer laser photorefractive keratectomy: characterization and clinical effects of treatment zone topography. Ophthalmology 1995;102:963-978.
9. Uozato H, Guyton DL. Centering corneal surgical procedures. Am J Ophthalmol 1987;103:264-75.
10. Schwartz-Goldstein BH, Hersh PS, Summit PRK Topography Study Group. Corneal topography of Phase III excimer laser photorefractive keratectomy: optical zone centration analysis. Ophthalmology 1995; 102:951-962.
11. Holladay JT, Prager TC. Mean visual acuity [letter]. Am J Ophthalomol 1991;111:372-374.
12. Sher NA, Hardten DR, Fundingsland B, et al. 193-nm excimer photorefractive keratectomy in high myopia. Ophthalmology 1994;101(9):1575-82.
13. Seiler T, Wollensak J. Myopic photorefractive keratectomy with the excimer laser. One-year followup. Ophthalmology 1991;98:1156-1163.
14. Epstein D, Fagerholm P, Hamberg NH, Tengroth B. Twenty-four-month follow-up of excimer laser photorefractive keratectomy for myopia. Refractive and visual acuity results. Ophthalmology 1994;101(9):1558-63.
15. Tuft SJ, Zabel RW, Marshall J. Corneal repair following keratectomy: A comparison between conventional surgery and laser photoablation. Invest Ophthalmol Vis Sci 1989;30:1769-1777.
16. Malley DS, Steinert RF, Puliafito CA, Dobi ET. Immunofluorescence study of corneal wound healing after excimer laser anterior keratectomy in the monkey eye. Arch Ophthalmol 1990;108:1316-1322.
17. Gartry D, Kerr Muir M, Lohmann CP, Marshall J. The effect of topical corticosteroids on refractive outcome and corneal haze after photorefractive keratectomy: a prospective, randomized, double-blind trial. Arch Ophthalmol 1992;110:944-952.
18. Waring GO3, Lynn MJ, McDonnell PJ. Results of the prospective evaluation of radial keratotomy (PERK) study 10 years after surgery. Arch Ophthalmol 1994;112(10):1298-308.
19. Hersh PS, Shah S, Holladay JT. Corneal asphericity following excimer laser photorefractive keratectomy. Ophth Surg Lasers 1996;27S:421-428.
20. Maguire LJ. Keratorefractive surgery, success, and the public health. Am J Ophthalmol 1994;117:394-398.
21. Hersh PS, Shah S, Geiger D & Holladay J. Corneal optical irregularity following excimer laser photorefractive keratectomy. J. Cat. Refract. Surg. 1996;22:197-204.
22. Maguire LJ, Bechara S. Epithelial distortions at the ablation zone margin after excimer laser photorefractive keratectomy for myopia [letter]. Am J Ophthalmol 1994;117(6):809-10.
23. Patel S, Marshall J. Corneal asphericity and its implications for photorefractive keratectomy: a mathematical model. J Refract Surg 1996;12:347-351.
24. Krueger RR, Trokel SL. Quantitation of corneal ablation by ultraviolet laser light. Arch Ophthalmol 1985;103:1741-1742.
25. Krueger RR, Trokel SL, Schubert HD. Interaction of ultraviolet laser light with the cornea. Invest Ophthalmol Vis Sci 1985;26:1455-1464.
26. Seiler T, Kriegerowski M, Schnoy N, Bende T. Ablation rate of human corneal epithelium and Bowman's layer with the excimer laser (193nm). J Refract Corneal Surg 1990;6:99-102.
27. Wu WCS, Stark WJ, Green WR. Corneal wound healing after 193 nm excimer laser keratectomy. Arch Ophthalmol 1991;109:1426-32.
28. Maguen E, Nesburn AB, Papaioannou T, Salz JJ, Macy JI, Warren C. Effect of nitrogen flow on recovery of vision after excimer laser photorefractive keratectomy without nitrogen flow. J Refract Corneal Surg 1994;10(3):321-6.
29. Fagerholm P, Hamberg NH, Tengroth B. Wound healing and myopic regression following photorefractive keratectomy. Acta Ophthalmol (Copenh) 1994;72(2):229-34.
30. O'Brart DP, Gartry DS, Lohmann CP, Muir MG, Marshall J. Excimer laser photorefractive keratectomy for myopia: comparison of 4.00- and 5.00-millimeter ablation zones. J Refract Corneal Surg 1994;10(2):87-94.
Table 1: Multiple Logistic Regression Model of Preoperative and Operative Characteristics Associated with Uncorrected Visual Acuity <20/40 Two Years Following PRK (N=611)
| Characteristic | Odds Ratio | 95% CI |
Preprocedure examination | Similar | Similar |
Age (1 year increment)* | 1.08 | 1.04-1.12 |
Attempted Correction* | | |
<2.5D | 0.34 | 0.04-2.87 |
2.5-<3.5D | 1.00 | - |
3.5-<5.5D* |
2.78 | 1.18-6.75 |
>5.5 D* | 4.19 | 1.66-10.58 |
Intraocular Pressure (1 mm increment)* | 0.85 | 0.75-0.96 |
Total PRK time (1-minute increment) | 0.96 | 0.68-1.36 |
Optical Zone <5.0 mm | 1.26 | 0.62-2.56 |
Table 1 (continued): Multiple Logistic Regression Model of Preoperative and Operative Characteristics Associated with Uncorrected Visual Acuity <20/40 Two Years Following PRK
| Characteristic | Odds Ratio | 95% CI |
Study Center | | |
1 | 0.89 | 0.23-3.38 |
2 | 0.79 | 0.23-2.76 |
3 | 0.70 | 0.16-3.03 |
4** | 1.00 | - |
5 | 1.07 | 0.28-4.14 |
6 | 0.26 | 0.05-1.53 |
7 | 0.28 | 0.03-2.81 |
8 | 0.36 | 0.03-3.85 |
9 | 1.16 | 0.29-4.60 |
10 | 0.21 | 0.02-1.97 |
* Statistically significant
** Reference: Center with percent uncorrected visual acuity 20/40 or better closest to the overall mean
Table 2: Patient Age and Post-PRK Uncorrected Visual Acuity
| Age | N | % Patients with UCVA>20/40 |
21-30 | 140 | 95.7% |
31-40 | 241 | 95.4% |
41-50 | 179 | 89.4% |
51-60 | 41 | 75.6% |
>60 | 9 | 88.9% |
Table 3: Attempted Correction and Post-PRK Uncorrected Visual Acuity (N=612)
| Attempted Correction | N | UCVA>20/40 |
1.5-<2.0 D | 16 | 93.8% |
2.0-2.5 D | 40 | 100.0% |
2.5-3.0 D | 55 | 98.2% |
3.0-3.5 D | 73 | 97.3% |
3.5-4.0 D | 81 | 92.6% |
4.0-4.5 D | 86 | 89.5% |
4.5-5.0 D | 69 | 92.8% |
5.0-5.5 D | 65 | 89.2% |
5.5-6.0 D | 60 | 88.3% |
6.0D | 67 | 85.1% |
Table 4: Multiple Logistic Regression Model of Preoperative and Operative Characteristics Associated with Predictability Outside +/- 1.0 Diopter 2 Years Following PRK (N=612)
| Characteristic | Odds Ratio | 95% CI |
Age (1 year increment)* | 1.05 | 1.04-1.07 |
Attempted Correction* | | | |
<2.5D | 0.74 | 0.29-1.85 |
2.5-<3.5D | 1.00 | - |
3.5-<5.5D* | 1.72 | 1.05-2.85 |
>5.5 D* | 2.95 | 1.65-5.26 |
Total PRK time (1-minute increment) | 0.89 | 0.72-1.10 |
Optical Zone <5.0 mm | 1.18 | 0.75-1.87 |
Iris Color* | | |
Dark and medium | 1.00 | - |
Light | 2.03 | 1.33-3.09 |
Table 4 (continued): Multiple Logistic Regression Model of Preoperative and Operative Characteristics Associated with Predictability Outside +/- 1.0 Diopter 2 Years Following PRK
| Characteristic | Odds Ratio | 95% CI |
Study Center | | |
1 | 0.93 | 0.44-1.98 |
2** | 1.00 | |
3 | 1.89 | 0.78-4.62 |
4 | 1.02 | 0.41-2.53 |
5 | 0.78 | 0.37-1.64 |
6 | 0.49 | 0.19-1.26 |
7 | 1.99 | 0.79-4.97 |
8 | 2.05 | 0.76-5.54 |
9 | 1.92 | 0.83-4.45 |
10 | 1.59 | 0.71-3.58 |
*Statistically significant
** Reference: Center with percent predictable closest to the overall mean
Table 5: Stratified Age Groups and Predictability Outcome
| Age | N | +/-1.0D | Overcorrection
>1.0D | Undercorrection
>1.0D |
21-30 | 140 | 83.6% | 4.3% | 12.1% |
31-40 | 241 | 82.4% | 8.7% | 8.7% |
41-50 | 179 | 71.0% | 19.6% | 9.5% |
51-60 | 41 | 58.5% | 29.3% | 12.2% |
>60 | 9 | 77.8% | 11.1% | 11.1% |
Table 6: Attempted Correction and Predictability of PRK
| Attempted Correction | N | +/-1.0D | Overcorrection | Undercorrection |
1.5-<2.0 D | 16 | 75.0% | 25.0% | 0.0% |
2.0-2.5 D | 40 | 92.5% | 5.0% | 2.5% |
2.5-3.0 D | 55 | 89.1% | 9.1% | 1.8% |
3.0-3.5 D | 73 | 79.9% | 15.1% | 5.5% |
3.5-4.0 D | 81 | 81.5% | 13.6% | 4.9% |
4.0-4.5 D | 86 | 77.9% | 11.6% | 10.5% |
4.5-5.0 D | 69 | 71.0% | 18.8% | 10.1% |
5.0-5.5 D | 65 | 76.9% | 12.3% | 10.8% |
5.5-6.0 D | 60 | 73.3% | 5.0% | 21.7% |
6.0D | 67 | 65.7% | 11.9% | 22.4% |
Table 7. Stability of Refraction Between 12 and 24 Months Stratified to Refractive Error
| Preoperative Spherical Equivalent (D) |
| Overall | <2.0 | 2.0-2.9 | 3.0-3.9 | 4.0-4.9 | 5.0-5.9 | >6.0 |
Hyperopic | | | | | | | |
Shift >1.0D | 3.2% | 0.0% | 5.2% | 0.7% | 3.5% | 5.0% | 3.0% |
Stable +/-1.0D | 89.9% | 100.0% | 93.5% | 94.9% | 89.4% | 86.7% | 80.0% |
Myopic | | | | | | |
|
Shift >1.0D | 6.8% | 0.0% | 1.3% | 4.4% | 7.0% | 8.3% | 16.7% |
Table 8: Multiple Logistic Regression Model of Preoperative and Operative Characteristics Associated with Changes of >1.0 Diopter in Manifest Spherical Equivalent Refraction Between 12 and 24 Months Following PRK (N=555)
| Characteristic | Odds Ratio | 95% CI |
Age (1 year increment) | 1.02 | 0.99-1.05 |
Attempted Correction* | | |
<3.0D | 1.00 | - |
3.0-<5.0D | 1.56 | 0.60-4.0 |
>5.0 D* | 3.46 | 1.32-9.11 |
Total PRK time (1-minute increment) | 0.91 | 0.66-1.26 |
Optical Zone <5.0 mm | 0.59 | 0.32-1.10 |
Intraocular Pressure (1 mm increment) | 0.99 | 0.89-1.10 |
Table 8 (continued): Multiple Logistic Regression Model of Preoperative and Operative Characteristics Associated with Changes of >1.0 Diopter in Manifest Spherical Equivalent Refraction Between 12 and 24 Months Following PRK
| Characteristic | Odds Ratio | 95% CI |
Study Center | | |
1 | 0.38 | 0.09-1.86 |
2** | 1.00 | - |
3 | 0.56 | 0.11-2.86 |
4 | 1.86 | 0.60-5.79 |
5 | 0.88 | 0.28-2.79 |
6* | 2.90 | 1.08-7.80 |
7 | 1.04 | 0.21-5.19 |
8 | 0.90 | 0.17-4.70 |
9 | 1.70 | 0.55-5.25 |
10 | 2.60 | 0.92-7.37 |
*Statistically significant
** Reference: Center with percent stability in manifest refraction spherical equivalent closest to the overall mean
