PHOTOREFRACTIVE KERATECTOMY (PRK) IS LESS INvasive

Transcription

1 Epithelial Healing and Clinical Outcomes in Excimer Laser Photorefractive Surgery Following Three Epithelial Removal Techniques: Mechanical, Alcohol, and Excimer Laser HYUNG KEUN LEE, MD, KYUNG SUB LEE, MD, JIN KOOK KIM, MD, HYEON CHANG KIM, MD, KYUNG RYUL SEO, MD, AND EUNG KWEON KIM, MD PURPOSE: To evaluate epithelial healing, postoperative pain, and visual and refractive outcomes after photorefractive keratectomy () using three epithelial removal techniques. DESIGN: Prospective, nonrandomized, comparative trial. METHODS: SETTING: Department of Ophthalmology, Yonsei University College of Medicine and Balgensesang Ophthalmology Clinic, Seoul, Korea. INTERVENTIONS: For the procedure, the corneal epithelium was removed in one of three ways: mechanically (conventional []) in 88 eyes of 44 patients; using excimer laser (transepithelial [t]) in 106 eyes of 53 patients; or using 20% diluted alcohol, laser-assisted subepithelial keratomileusis () in 106 eyes of 53 patients. MAIN OUTCOME MEASURES: Epithelial healing, postoperative pain, uncorrected visual acuity (UCVA), best spectaclecorrected visual acuity (BSCVA), and remaining refractive error. RESULTS: The mean postoperative pain scores were for, for t, and for (P.125). The mean epithelial healing rates were for, for t, and mm 2 /day for (P <.001). Accepted for publication Aug 19, From the Institute of Vision Research (H.K.L., E.K.K., K.R.S.), and Public Health (H.C.K.), and Brain Korea 21 Project for Medical Science (E.K.K.), Yonsei University College of Medicine; Balgensesang Ophthalmology Clinic (K.S.L., J.K.K.), Seoul, Korea. This work was supported by a grant from the Korea Health 21 R&D Project, Ministry of Health and Welfare, Republic of Korea (grant 02-PJ1-PG1-CH02-003). Inquiries to Eung Kweon Kim, MD, Institute of Vision Research, Department of Ophthalmology, Yonsei University College of Medicine, 134 Shinchon-Dong, Seodaemun-Gu, Seoul, South Korea, ; fax: ( 82) ; eungkkim@yumc.yonsei.ac.kr. The postoperative 6-month remaining mean spherical equivalents (diopters) were for, for t, and for (P.01). The group showed less favorable UCVA than other groups. There was no significant difference in BSCVA between the groups. CONCLUSIONS: Postoperative pain, subepithelial opacity and BSCVA were similar regardless of the epithelial removal procedure. A faster epithelial healing rate did not result in better visual or refractive outcomes. Using the same nomogram, t resulted in a slight overcorrection, and resulted in a slight undercorrection. (Am J Ophthalmol 2005;139: by Elsevier Inc. All rights reserved.) PHOTOREFRACTIVE KERATECTOMY () IS LESS INvasive than laser in situ keratomileusis (LASIK) and does not structurally weaken the cornea. 1,2 However, it is associated with significant postoperative pain, slower visual recovery, and corneal hazing, which are deterrents to patient and surgeon acceptance of the technique. 3,4 Several modifications of the conventional procedure have been introduced to reduce pain, epithelial healing time, and corneal haze. Most alternative methods modify the means by which the corneal epithelium is removed. Laser-assisted subepithelial keratomileusis () was introduced by Camellin in It involves the use of a dilute alcohol solution and creates a complete flap of the epithelium. This technique has become popular for correcting refractive errors, with some reports, particularly in cases with moderate degrees of myopia, suggesting that the refractive and visual outcomes in patients were better than in conventional patients. 5,6 In contrast, BY ELSEVIER INC. ALL RIGHTS RESERVED /05/$30.00 doi: /j.ajo

2 others report that and outcomes were similar in relation to postoperative pain and visual outcomes. 7,8 Transepithelial (t) uses an excimer laser to remove the epithelium. In an animal study, Helena and associates 9 reported that corneal epithelial cell removal using transepithelial methods did not cause keratocyte apoptosis. 10 Considering that keratocyte apoptosis is the initial step of subepithelial fibrosis, the suggested advantages of t over are the potential for faster visual and refractive rehabilitation with less corneal hazing. To our knowledge, there are no reports comparing the visual and refractive outcomes of the three epithelial removal techniques. This article reports on a prospective comparison of subjective postoperative pain, epithelial healing, and visual and refractive results after using the three techniques and performed by a single surgeon. PATIENTS AND METHODS THIS STUDY WAS A PROSPECTIVE, DOUBLE-MASKED COMparative clinical trial performed at the Department of Ophthalmology, Yonsei University, and the Balgensesang Ophthalmology Clinic, Seoul, Korea. It included 276 eyes from 138 patients with myopia. Preoperative examinations included visual acuity, manifest refraction, cycloplegic refraction, slit-lamp examination, pachymetry, applanation tonometry, keratometry and videokeratometry readings, and a fundus examination. Tear function was assessed using a Schimer s test without corneal anesthesia, and the tear film breakup time was determined. Before surgery, a study author (H.K.L) explained to patients the merits, demerits, and complications associated with the three possible surgical procedures for removing corneal epithelium;, t, and. Patients were then allowed to select the method they preferred. If the patient was unable to choose, a procedure was randomly assigned using a random number table. Following this process, the three groups comprised conventional mechanical (), 88 eyes of 44 patients; excimer laser (t), 106 eyes of 53 patients; and 20% diluted alcohol (), 82 eyes of 41 patients. All procedures conformed to the tenets of the Declaration of Helsinki, and informed consent was obtained from all patients after gaining approval from the institutional review board. The exclusion criteria were patients with anterior segment pathology, any evidence of lid disease, progressive or unstable myopia and keratoconus, a history of herpetic keratitis, and previous intraocular or corneal surgery. Soft contact lens use was discontinued for a minimum of 3 days, and a rigid gas-permeable contact lens was used for a minimum of 3 weeks before examination and treatment. All surgical procedures were performed by the same surgeon (J.K.K.). Conventional : Conventional was performed using the VISX Star S3 excimer laser using a standardized protocol. Preoperatively, patients were given one drop of 0.5% proparacaine (Alcaine, S.A. Alcon-Couvreur, Puurs, Belgium), 0.1% diclofenac (Optanac, Samil, Korea), and 0.3% ciprofloxacin (Ciloxan, Alcon-Couvreur, Puurs, Belgium). Lashes and lids were treated with a povidone-iodine swab. A closed-loop lid speculum was placed between the lids of the eye to be treated, and the other eye was occluded. A 7.0-mm optical zone marker was applied to the cornea, centering it over the image of the pupil. A crescent knife was used to remove the central 7.0 mm of the corneal epithelium. The loose epithelium was removed using a blunt spatula. This was followed by stromal ablation using the VISX Star S3 laser (VISX, Santa Clara, California, USA). t: The transepithelial technique is based on the difference in the fluorescence spectrum between the corneal epithelium and stroma when the cornea is irradiated with 193-nm wavelength light. After the same preoperative preparation as described for the conventional method, the cornea underwent ablation (7.0-mm diameter and 45- m depth) using a phototherapeutic keratectomy (PTK) mode. All lights in the operation room were then turned off, and the remaining thin layer of epithelium was removed using an excimer laser by observing the disappearance of blue fluorescent light as the epithelium was ablated. Immediately after the blue fluorescence disappeared, the laser system was reprogrammed with the appropriate myopic correction, and stromal ablation was immediately performed. : A speculum was applied to the patient s eye, and 0.5% proparacaine hydrochloride was instilled. An alcohol solution cone (J2905, Janach, Como, Italy) with an 8.5-mm diameter was placed on the eye, and a 20% alcohol solution was instilled inside the cone and left for approximately 30 seconds, after which it was carefully washed off with a balanced salt solution so that the epithelium around the flap would not be disturbed. The epithelial flap was gently lifted with an epithelial microhoe (Janach J 2915A). The flap was then peeled toward the 12 o clock position using a spatula (Janach J2910A), maintaining a one-sheet layer form. An excimer laser was used in the usual manner employing the nomogram for the photorefractive keratectomy () with the same laser system. The flap was washed with balanced salt solution and then repositioned carefully using a spatula. For all treatment groups, ablated stroma was immediately irrigated with a balanced salt solution, and one drop of both 0.3% ciprofloxacin and 0.1% diclofenac were instilled on the surgical site. In addition, a soft contact lens (Focus, Ciba Vision, Duluth, GA) was placed on the cornea at the completion of the procedure. VOL. 139, NO. 1 THREE EPITHELIAL REMOVAL TECHNIQUES 57

3 Postoperative Medication: All patients were monitored daily until the epithelial defect had healed completely and were instructed to apply one drop of both diclofenac and ciprofloxacin every 6 hours. Artificial tears (hyalein, 0.1% hyaluronic acid, Santen, Osaka, Japan) were applied every hour until complete epithelial healing had occurred. Complete epithelialization was determined using daily slit-lamp observation and photography (as described later). Once the epithelium had healed completely, the therapeutic contact lenses were removed from the cornea, and 0.3% ciprofloxacin and 0.1% fluorometholone (Fluorometholone, Santen, Osaka, Japan) were then administered four times daily for 1 week and twice daily for 1 month. Epithelial Healing: Epithelial wound healing was assessed using photographs taken immediately after surgery and then every 24 hours until complete epithelial healing had occurred. All photographs were taken at a fixed focus length using a slit lamp (Haag-Streit, Koeniz, Switzerland) with a digital camera attached (Coolpix 4500, Nikon, Tokyo, Japan). Epithelial defect areas were measured using image analyzing software (Scion image beta 4.02, Scion, Frederick, Maryland). An optometrist blinded to the surgical procedures took the photographs and made the measurements. Postoperative Examinations: For the subjective pain scores, the patients were given a Faces Pain Scale and were asked to rate the pain level in each of their eyes on a scale from 0 (no pain) to 10 (worst pain) until complete epithelial healing had occurred. The Faces Pain Scale is a self-report measure used to assess the intensity of patient s pain. The method uses a piece of paper on which six faces with expressions ranging from happy (Face 0 no pain) to very sad (Face 10 the maximum pain you can imagine) are printed. Patients chose the face that best described how they were feeling. The uncorrected visual acuity (UCVA), best spectacle-corrected visual acuity (BSCVA), manifest refraction, tonometry, and slit-lamp biomicroscopic examinations were performed by one investigator (K.S.L) at 1 week, 1 month, 3 months, and 6 months after surgery. At 1, 3, and 6 months after surgery, two authors (K.E.K., S.K.R.) separately graded subepithelial haze in a masked fashion. Haze levels were determined using a slit lamp according to the method of Helena and associates. 9 Statistical Analysis: The preoperative and postoperative characteristics of participants were analyzed using a repeated-measures analysis of variance with multiple comparison of the Tukey method. UCVA and BSCVA were compared using a chi-square or Fisher exact test for tables. Correlations between preoperative and postoperative independent variables were analyzed using Pearson correlation coefficient. Statistical analysis was performed using a Statistical Analysis System (Ver. 6.12, SAS Institute, Cary, North Carolina, USA). A P value.05 was considered statistically significant. RESULTS TABLE 1 SHOWS THE PREOPERATIVE CHARACTERISTICS OF the patients. There were no significant differences in preoperative variables between the groups. Up to 1 month postoperatively, no patients were lost to follow-up. At 3 months postoperatively we were able to follow up 84 eyes in the, 100 in the t, and 78 in the groups, and at 6 months we were able to follow up 76, 92, and 72 eyes, respectively. Epithelial Healing: The time required for reepithelialization differed between the three groups. The mean period of reepithelialization was days in the group, days in the t group, and days in the group (P.001). Note that in the group analysis, to avoid any risk of bias, we excluded cases where the epithelial flap was not a complete single flap. Because each surgical method removed a different amount of epithelial area ( mm 2 with, mm 2 with t, and mm 2 with patients; mean SD), we compared the epithelial healing rate for each procedure. The data in Figure 1 show the velocity of epithelial healing. We found that the mean epithelial healing rate was mm 2 /day with, mm 2 /day with t, and mm 2 /d with (P.001). Thus, gave the fastest epithelial healing rate. Most eyes in this study had an almost identical healing pattern over time, following a curvilinear slope, with a slower initial phase and a faster intermediate phase. Of the preoperative and intraoperative variables, the preoperative spherical equivalent (r.52, P.001) and intraoperative ablation depth (r.64, P.001) were found to correlate inversely with corneal epithelial healing velocity (Pearson s correlation coefficient). Pain Score: The subjective pain scale score was applied up until postoperative day 2, because by that time some cases were showing complete epithelial healing. The pain scale scores for the three surgical groups on postoperative days 1 (P.125) and 2 (P.283) were not significantly different (Table 2). Visual Acuity: UCVA and BSCVA data were not statistically different between the groups up to 3 months postoperatively (Table 3). At 6 months, 53 eyes (69.7%) from the, 64 eyes (69.6%) from the t, and 48 eyes (66.7%) from the group had a UCVA of 20/20 or better. However, 3 eyes (4.0%) from the, 3 eyes (3.3%) from the t, and 5 eyes (6.8%) from the group had a UCVA of 20/50 or worse (P.032). 58 AMERICAN JOURNAL OF OPHTHALMOLOGY JANUARY 2005

4 TABLE 1. Preoperative Characteristics of Patients (Mean SD) (n 88) t (n 106) (n 82) P Value Age (yrs) Mean Range Sex (M/F) 9/35 10/43 9/ Refractive error Spherical equivalent (D) Mean Range 2.00 to to to 9.50 Magnitude of cylinder (D) Mean Range 0 to to to 4.50 Corneal thickness ( m) Mean Range Schirmer value (mm) Mean Range ( ) ( ) ( ) BUT (sec).092 Mean Range D diopter; F female; laser-assisted subepithelial keratomileusis; M male; photorefractive keratectomy; t transepithelial. Repeated-measures analysis of variance and multiple comparison of Tukey method; chi-square test. FIGURE 1. Corneal epithelial wound closure rates after photorefractive keratectomy using a conventional mechanical procedure (), a transepithelial procedure (t), and laser subepithelial keratomileusis (). laser-assisted subepithelial keratomileusis. photorefractive keratectomy; t transepithelial. * P <.0001, repeated-measures analysis of variance. VOL. 139, NO. 1 THREE EPITHELIAL REMOVAL TECHNIQUES 59

5 TABLE 2. Postoperative Subjective Pain Scores After, t, and Pain Score POD 1 POD 2 t t Mean SD Range P value laser-assisted subepithelial keratomileusis; POD postoperative day; photorefractive keratectomy; t transepithelial. Repeated-measures analysis of variance. With respect to BSCVA data, at 6-month follow-up, no patient showed loss of more than 3 lines. A decrease of more than 1 Snellen line occurred in 6 eyes (7.8%) in the group, 8 eyes (8.7%) in the t group, and 6 eyes (8.3%) in the group, respectively (P.158, chi-square test; Figure 2). Refractive Error, Spherical Equivalent: During the 6-month follow-up, a greater hyperopic shift was observed in the t group compared with the group, and a greater myopic shift was observed in the group (P.001; Table 4). Six months after surgery, the remaining mean spherical equivalents (diopters) were in the group, in the t group, and in the group (P.001). Forty-six patients (60.5%) from the, 66 (71.7%) from the t, and 41 (56.9%) from the group achieved a postoperative manifest correction within 0.50 diopters of the attempted myopic correction 6 months after the operation (P.001). In contrast to the spherical component, the magnitude of the cylinder did not change significantly from 1 month through to 6 months after surgery in any group. Subepithelial Opacity: Up to 6 months postoperatively, subepithelial haze was graded separately by two masked practitioners. We found no difference in subepithelial opacity between the three groups. Almost all patients showed a subepithelial opacity below grade 2. However, two eyes, one each from the and t groups, showed opacity grade 3 (Table 5). DISCUSSION PAIN, CORNEAL HAZE, AND MYOPIC REGRESSION ARE UNavoidable problems associated with the conventional procedure. techniques have been modified in several ways to overcome these postoperative problems. Modified techniques include laser transepithelial debridement, deepithelialization with diluted ethanol, a rotating brush, cooling of the ablation stromal surface, and use of the amniotic membrane flap However, because of varying surgical nomograms, applications, laser systems, and intraoperative and postoperative procedures, as well as the different surgeons performing the procedure, comparing the clinical outcomes among the techniques is difficult, and identifying which method provides the best outcome remains controversial. 7,8,16 Therefore, the aim of this study was to prospectively compare the clinical outcomes after three epithelial removal techniques performed by the same surgeon with the same nomogram and follow-up programs. We found that all three epithelial removal techniques (, t, and ) appeared to be safe and effective for treating a wide range of myopias and myopic astigmatisms. Subjective pain, epithelial healing time, and refractive and visual outcomes were found to be acceptable for each of the methods up to 6 months after surgery. The study found there was no difference in subjective pain scores between the three groups and, in particular, no advantage in using in terms of reducing postoperative pain. This finding contrasts with those of a prospective, nonmasked study reporting that -treated eyes suffered less postoperative pain and corneal haze. 6 Similarly, Kanitkar and associates 17 reported that the procedure was superior to t in terms of postoperative pain. However, consistent with our findings, Litwak 7 and Pirouzian 8 recently described prospective masked studies in which there was no significant difference between and procedures in terms of pain and surgical outcomes. We found that the t group showed the shortest epithelial healing time of the three methods. However, considering the difference in the initial deepithelialized surface, the fastest epithelial healing rate was in the group. In the procedure, diluted alcohol separated the epithelium and stroma between the basement membrane or Bowman s layer level without mechanical trauma on the stromal surface. Previous ultrastructural examination of corneal epithelium in patients who underwent alcohol-assisted epithelial removal showed an intact epithelial cell layer with normal desmosomes and hemidesomsomes. 18 Chen et al reported that corneal epithelial cells from cadaver eyes were viable when exposed to 20% alcohol for up to 45 seconds. 19 The procedure would appear advantageous, with others concluding that the fastest epithelial regeneration rate is likely to occur in eyes that have an environment favorable for epithelial cell regeneration, such as a smooth stromal bed, a protective epithelial flap, and less inflammation. 20 Compared with, t can theoretically provide a smooth deepithelialized stromal surface through use of a high-energy laser with a shorter wavelength. Kanitkar and associates 17 reported that the epithelial healing time was no different after or transepithelial excimer laser debridements, although the ablation area was not mentioned in that 60 AMERICAN JOURNAL OF OPHTHALMOLOGY JANUARY 2005

6 TABLE 3. Postoperative Uncorrected Visual Acuity After, t, and, n (%) 1 Month 3 Months 6 Months t t t (n 88) (n 106) (n 82) (n 84) (n 100) (n 78) (n 76) (n 92) (n 72) 20/20 or better 72 (81.9) 84 (79.2) 64 (78.0) 64 (76.2) 75 (75.0) 58 (74.4) 53 (69.7) 64 (69.6) 48 (66.7) 20/25 to 20/40 16 (18.1) 21 (19.8) 14 (20.8) 18 (21.4) 23 (23.0) 17 (21.8) 20 (26.3) 25 (27.1) 19 (26.4) 20/50 or worse 0 (0) 1 (1.0) 1 (1.2) 2 (2.4) 2 (2.0) 3 (3.8) 3 (4.0) 3 (3.3) 5 (6.9) P value laser-assisted subepithelial keratomileusis; photorefractive keratectomy; t transepithelial. Chi-square test. FIGURE 2. Changes in the Snellen lines of the best spectacle-corrected visual acuity at 6 months postsurgery. laser-assisted subepithelial keratomileusis; photorefractive keratectomy; t transepithelial ; P.158, chi-square test. TABLE 4. Postoperative Mean Refractive Error After, t, and 1 Month 3 Months 6 Months t t t (n 88) (n 106) (n 82) (n 84) (n 100) (n 78) (n 76) (n 92) (n 72) Mean SD P value laser-assisted subepithelial keratomileusis; photorefractive keratectomy; t transepithelial. Repeated-measures analysis of variance. VOL. 139, NO. 1 THREE EPITHELIAL REMOVAL TECHNIQUES 61

7 TABLE 5. Postoperative Subepithelial Opacity 6 Months After, t, and, n (%) Grade (n 76) t (n 92) (n 72) 0 21 (27.6) 33 (35.9) 18 (25.0) (44.7) 39 (42.4) 35 (48.6) 1 17 (22.4) 16 (17.4) 16 (22.2) 2 3 (4.0) 3 (3.2) 3 (4.2) 3 1 (1.3) 1 (1.1) 0 (0) 4 0 (0) 0 (0) 0 (0) laser-assisted subepithelial keratomileusis; photorefractive keratectomy; t transepithelial. P.094 Chi-square test. report. In contrast to alcohol and laser epithelial removal, the mechanical removal technique can cause a rough stromal wound bed that may hamper epithelial healing. Bowman s layer defect, an irregular anterior stromal surface, and retained islands of residual epithelium have been noted in the mechanical removal technique. 21,22 Some studies have concluded the epithelial healing rate has a large influence on refractive and visual outcomes. 23,24 We found, however, that the epithelial healing rate did not have a large influence on clinical outcomes as long as healing was not severely delayed. All three groups showed good refractive results up to 6 months after surgery. After 1 month, t patients showed a greater hyperopic shift and patients showed more myopic results, compared with patients. Although we could not identify the exact reasons for these observations, the refractive results may suggest the extent of stromal hydration during the procedure affected on the postoperative remaining refractive error, compared with mechanical scraping, using either alcohol or an excimer laser to remove corneal epithelium may have some advantages. The time required for mechanical debridement can be longer than that required for laser or alcohol scrape techniques, even for experienced surgeons. 11 This may cause stromal dehydration due to evaporation of stromal surface fluid, which can affect the refractive predictability. For the t procedure, 45 m was initially ablated, and the remaining epithelium was removed until the autofluorescence, which was used to monitor possible ablation of the basal lamina, had disappeared. However, total epithelial removal using a laser is difficult, and some regions of the basal lamina and Bowman s layer may be removed unintentionally with the epithelium in the PTK mode. Therefore, overablation using the t technique might explain the hyperopic shift trend observed in our study. Gimbel and associates 25 reported that t is more precise and has a more consistent time span than conventional, so variable stromal dehydration becomes less of a concern. However, the transepithelial technique can leave variable central islands of epithelium and cause unintended stromal ablation. 26 In contrast, stromal dehydration is not considered a problem with the procedure. Moreover, water from the diluted alcohol may diffuse into the stroma during. If these possibilities were indeed true, then the relatively hydrated corneal stroma during could result in an undercorrection when the stroma was ablated with the same nomogram as for. Despite the differences in the remaining refractive errors, UCVA and BSCVA were similar for each group up to 3 months after surgery. At 6 months, the t group showed a more favorable UCVA than the other groups. These results may correlate with the postoperative remaining spherical equivalent. The progressive myopic shift in the remaining spherical equivalent in all three groups was found up to 6 months postoperatively, and we believe that this myopic shift caused the unfavorable UCVA. Therefore, the initially hyperopic t group had a more favorable visual outcome than the other groups. A limitation of the present study was that we did not randomize the patients into the three groups; the patients chose the procedure they preferred. This may have reduced surgeon or examiner bias, but this cannot be regarded as a randomized study, and such a study is recommended for future research. In addition, our study lacks comparison with currently performed corneal epithelium removal methods such as long-term alcohol drip and then removal with a Weck cell sponge or rotary brush. In conclusion, we found that there was little difference in terms of clinical outcome, postoperative pain and subepithelial opacity among patients undergoing three epithelial removal techniques:, t, and. The t technique resulted in slight overcorrection, which may have been caused by an ablation of the superficial stroma. The technique showed a slight undercorrection, which may be related to stromal hydration. However, these amounts of under- and overcorrection were within the acceptable range. Although, t, and methods of epithelial removal are all safe and effective, continuing investigation using additional prospective and randomized large scale studies may allow for further refinement. REFERENCES 1. Jacobs JM, Taravella MJ. Incidence of intraoperative flap complications in laser in situ keratomileusis. J Cataract Refract Surg 2002;28: Melki SA, Azar DT. LASIK complications: etiology, management, and prevention. Surv Ophthalmol 2001;46: Loewenstein A, Lipshitz I, Varsanno D, et al. Complications of excimer laser photorefractive keratectomy for myopia. J Cataract Refract Surg 1997;23: AMERICAN JOURNAL OF OPHTHALMOLOGY JANUARY 2005

8 4. Alio JL, Artola A, Claramonte PJ, et al. Complications of photorefractive keratectomy for myopia: two year follow-up of 3000 cases. J Cataract Refract Surg 1998;24: Dastjerdi MH, Soong HK. (laser subepithelial keratomileusis). Curr Opin Ophthalmol 2002;13: Lee JB, Seong GJ, Lee JH, et al. Comparison of laser epithelial keratomileusis and photorefractive keratectomy for low to moderate myopia. J Cataract Refract Surg 2001;27: Litwak S, Zadok D, Garcia-deq Uevedo V, Robledo N, Chayet A. Laser-assisted subepithelial keratectomy versus photorefractive keratectomy for the correction of myopia; a prospective comparative study. J Cataract Refract Surg 2002; 28: Pirouzian A, Thornton JA, Ngo S. A randomized prospective clinical trial comparing laser subepithelial keratomileusis and photorefractive keratectomy. Arch Ophthalmol 2004;122: Helena KD, Pouliquen YM, Waring GO III, et al. Corneal wound healing in monkeys after repeated excimer laser photorefractive keratectomy. Arch Ophthalmol 1992;110: Campos M, Hertzog L, Wang XW, et al. Corneal surface after deepithelialization using a sharp and a dull instrument. Ophthalmic Surg 1992;23: Abad J-C, Talamo JH, Vidaurri-Leal J, et al. Dilute ethanol versus mechanical debridement before photorefractive keratectomy. J Cataract Refract Surg 1996;22: Pallikaris IG, Karoutis AD, Lydataki SE, Siganos DS. Rotating brush for fast removal of corneal epithelium. J Refract Corneal Surg 1994;10: Kitazawa Y, Maekawa E, Sasaki S, et al. Cooling effect on excimer laser photorefractive keratectomy. J Cataract Refract Surg 1999;25: Lee YG, Chen WYW, Petroll WM, Cavanagh HD, Jester JV. Corneal haze after photorefractive keratectomy using different epithelial removal techniques: mechanical debridement versus laser scrape. Ophthalmology 2001;108: Lee HK, Kim JK, Kim SS, et al. Effect of amniotic membrane after in laser assisted subepithelial keratomileusis on epithelial healing, clinical and refractive outcomes. J Cataract Refract Surg 2004;30: Claringbold TV II. Laser-assisted subepithelial keratectomy for the correction of myopia. J Cataract Refract Surg 2002; 28: Kanitkar KD, Camp J, Humble H, Shen DJ, Wang MX. Pain after epithelial removal by ethanol-assisted mechanical versus transepithelial excimer laser debridement. J Refract Surg 2000;16: Azar DT, Ang RT, Lee JB, et al. Laser subepithelial keratomileusis: electron microscopy and visual outcomes of flap photorefractive keratectomy. Curr Opinion Ophthalmol 2001;12: Chen CC, Chang JH, Lee JB, Azar DT. Human corneal epithelial cell viability and morphology after diluted alcohol exposure. Invest Ophthalmol Vis Sci 2002;43: Shah S, Sarhan ARS, Doyle SJ, Pillai CT, Dua HS. The epithelial flap for photorefractive keratectomy. Br J Ophthalmol 2001;85: al-abdulla NA, Jabbur NS, O Brien TP. Astigmatism outcomes following spherical photorefractive keratectomy for myopia. J Refract Surg 1998;14: Campos M, Hertzog L, Wang XW, et al. Corneal surface after deepithelialization using a sharp and a dull instrument. Ophthalmic Surg 1992;23: Nakamura K, Korosaka D, Bissen-Miyajima H, Tsubota K. Intact corneal epithelium is essential for the prevention of stromal haze after laser assisted in situ keratomileusis. Br J Ophthalmol 2001;85: Lohmann CP, Patmore A, Reischl U, Marshall J. The importance of the corneal epithelium in excimer-laser photorefractive keratectomy. Ger J Ophthalmol 1996;5: Gimbel HV, DeBroff BM, Beldavs RA, van Westenbrugge JA, Ferensowicz M. Comparison of laser and manual removal of corneal epithelium for photorefractive keratectomy. J Refract Surg 1995;11: Clinch TE, Moshirfar M, Weis JR, Ahn CS, Hutchinson CB, Jeffrey JH. Comparison of mechanical and transepithelial debridement during photorefractive keratectomy. Ophthalmology 1999;106: VOL. 139, NO. 1 THREE EPITHELIAL REMOVAL TECHNIQUES 63