Confocal Microscopy of Corneal Stroma and Endothelium After LASIK and PRK Javad Amoozadeh, MD; Soheil Aliakbari, MD; Amir-Houshang Behesht-Nejad, MD; Mohammad-Amin Seyedian, MD; Bijan Rezvan, DDS; Hassan Hashemi, MD ABSTRACT he submicron accuracy of excimer laser ablation is an important factor in the popularity of LASIK and 1,2 PURPOSE: To compare with confocal microscopy the T photorefractive keratectomy (PRK) worldwide. changes in stromal keratocyte density and endothelial Photorefractive keratectomy is a surface ablation procedure cell count due to photorefractive keratectomy (PRK) and that uses the excimer laser to reshape the superfi cial stromal LASIK. layer after removal of surface epithelium. Laser in situ ker- atomileusis involves the creation of a hinged corneal fl ap of METHODS: In this prospective study, 32 eyes (16 130- to 160-µm thickness and delivery of the excimer laser myopic patients) were examined with the NIDEK Con- foscan 3 confocal microscope before and 6 months af- ablation to the underlying stroma. During LASIK, the ante- ter PRK and LASIK. The preoperative mean myopia was rior stroma and epithelium are preserved, which results in Ϫ2.85Ϯ0.99 diopters (D) (range: Ϫ1.00 to Ϫ4.00 D) differences in the healing process compared to PRK.3 Surface in 24 eyes that underwent PRK and Ϫ2.94Ϯ0.96 D procedures such as PRK have been shown to be safe and pre- Ϫ Ϫ (range: 2.00 to 4.25 D) in 8 eyes that underwent dictable for correcting low and moderate refractive errors and LASIK. Keratocyte density in the anterior and posterior stroma and the endothelial cell count were measured. they circumvent fl ap-related complications and biomechani- 4 Statistically signifi cant changes were assessed using the cal instability seen with LASIK. Complications related to sur- t test. PϽ.05 was considered statistically signifi cant. face ablation procedures such as PRK include delayed healing, increased risk of haze, dry eye, and regression of effect.4-6 The RESULTS: Preoperative hexagonal cell percentage in incorporation of mitomycin C during PRK may mitigate haze Ϯ Ϯ the LASIK group was 52.17 11.43 and 51.33 10.98 formation postoperatively.7 Complications of LASIK include in the PRK group. Postoperatively, the percentages were 52.96Ϯ7.55 and 53.34Ϯ10.2, respectively. Six epithelial ingrowth, fl ap complications, keratectasia, biome- 4 months postoperatively, keratocyte density changed by chanical instability, and dry eye. 367.12Ϯ103.35 cells/mm2 (34.7% reduction) in the Despite improvements in surgical techniques and ex- anterior stroma (PϽ.05) and 9.25Ϯ28.28 cells/mm2 cimer laser technology, similar advances in the understand- Ͼ (1.31% reduction) in the posterior stroma (P .05) for ing of the cellular response following LASIK and PRK have the LASIK group. In the PRK group, these values were 2 319.71Ϯ83.45 cells/mm2 (31.13% reduction) in the not been made. Cellular and structural changes induced by anterior stroma (PϽ.05) and 0.17Ϯ38.97 cells/mm2 refractive surgery may aid in understanding the natural cel- (0.02% reduction) in the posterior stroma (PϾ.05). The lular processes and potential complications that occur after changes in keratocyte densities were not statistically each type of surgery. The complex nature of tissue inter- signifi cant between groups (PϾ.05). The mean number of keratocytes decreased by 37.2% in the retroablation zone of the LASIK group (PϽ.05). No changes were From Farabi Eye Hospital, Tehran University of Medical Sciences (Amoozadeh, noted in endothelial cell counts. Behesht-Nejad, Hashemi); and Noor Ophthalmology Research Center, Noor Eye Hospital (Aliakbari, Seyedian, Rezvan, Hashemi), Tehran, Iran. CONCLUSIONS: A signifi cant decrease occurred in the number of stromal keratocytes in the anterior stroma. The authors have no financial or proprietary interest in the materials pre- sented herein. Despite differences in surgery, the change in keratocyte density and endothelial cell counts were similar be- Rich Bains, consultant to NIDEK Co Ltd, assisted in the preparation of the tween LASIK and PRK groups (PϾ.05). [J Refract Surg. manuscript. 2009;25:S963-S967.] Portions of this article have been published previously in the Iranian Journal doi:10.3928/1081597X-20090915-12 of Ophthalmology (2009;21:23-28). Correspondence: Soheil Aliakbari, MD, Noor Eye Hospital, No. 96, Esfandiar Blvd, Vali’Asr Ave, Tehran 1968655841, Iran. Tel: 98 21 82400; Fax: 98 21 88650501; E-mail: [email protected] Journal of Refractive Surgery Volume 25 October (Suppl) 2009 Commercially Sponsored Section S963 Keratocyte Density After LASIK and PRK/Amoozadeh et al action postoperatively may help determine selection the Confoscan 3 preoperatively and 6 months post- criteria for LASIK and PRK. operatively. Confocal microscopy has been used to investigate Confoscan 3 measurements were performed by the cellular morphology and structure of the various an experienced ophthalmologist (H.H.) who did not corneal layers.1-6,8-11 However, the results from these perform the surgeries on this study cohort. Mea- studies are contradictory. Normal keratocyte density is surements were acquired in automatic mode us- 1079 cells/mm2.12 In the present study, in vivo confo- ing the 20ϫ non-contact lens. Topical tetracaine cal microscopy was used to evaluate keratocyte den- 0.5% drops were applied to anesthetize the eyes. sity and endothelial cell count and to compare changes Methyl cellulose drops (Viscotears Gel; CIBA Vi- seen in eyes that underwent LASIK and PRK. sion, Duluth, Ga) were applied on the objective lens of the machine to provide a regular image, and the PATIENTS AND METHODS lens was aligned with the pupil center. For each eye, 350 frames, 5.0-µm apart were acquired. The STUDY POPULATION cells in each layer were marked by an examiner This study was a prospective, non-randomized study (S.A.) (who was masked to the type of surgery) to of 16 patients who were scheduled to undergo LASIK avoid inadvertently counting the same cell twice. (LASIK group) or PRK (PRK group) from March 1 to The number of cells was calculated by counting May 15, 2007. Patients with low to moderate myopia the number of clear and distinct cells in areas of (Ϫ1.00 to Ϫ4.50 diopters [D]) or low to moderate myo- 379ϫ286 µm2. The keratocyte density was studied pic astigmatism (ϽϪ4.50 D sphere and up to 1.50 D in two layers—25 µm posterior to Bowman’s layer astigmatism) were enrolled after signing consent forms (anterior stroma) and 40 µm anterior to Descemet’s for pre- and postoperative examination. Patient prefer- membrane (posterior stroma). The layer 25 µm poste- ence in consultation with the surgeon was the basis for rior to Bowman’s layer was chosen based on previous undergoing PRK or LASIK. Eight eyes of 4 patients had studies,3-5 as the exact amount of tissue removal for LASIK and 24 eyes of 12 patients had PRK. Exclusion each patient could not be predicted preoperatively. criteria were systemic diseases such as diabetes, trau- Postoperatively, the same 25 µm from Bowman’s matic or infectious complications after refractive sur- layer was used, although this may be the midstro- gery, active eye disease, and previous corneal surgery. ma compared to preoperatively. Additionally, the After applying the above exclusion criteria, all pa- Confoscan operator (S.A.) was not aware of the abla- tients were available for analysis. Ten patients used tion depth of each patient postoperatively. By using contact lenses prior to surgery (8 in the PRK group and 2 Bowman’s layer as the reference point, epithelial hy- in the LASIK group). However, all contact lens wearers perplasia was ruled out as a source of error. For deep were required to discontinue use for at least 3 to 28 days stroma, the endothelium was chosen as the reference (depending on contact lens type) prior to preoperative point. evaluation to stabilize keratometry and corneal topogra- phy. Patients were required to have normal keratometry SURGERIES and topography prior to undergoing refractive surgery. All surgeries were performed by a single surgeon (H.H.). The surgeon did not perform confocal micros- PRE- AND POSTOPERATIVE EXAMINATIONS copy measurements. The Technolas 217 Planoscan ab- Pre- and postoperative examinations included mea- lation (Bausch & Lomb, Rochester, NY) was used for surement of uncorrected visual acuity, best spectacle- all patients. For PRK, the epithelium was manually de- corrected visual acuity, slit-lamp evaluation of the cornea brided with a hockey stick spatula. At the end of the and anterior chamber, tonometry, keratometry, topogra- procedure, an O2 Optix bandage contact lens (CIBA phy, pachymetry, cycloplegic refraction, and confocal Vision) was placed, with removal 3 days after surgery microscopy (Confoscan 3; NIDEK Co Ltd, Gamagori, Ja- or upon complete re-epithelialization. pan). Laser in situ keratomileusis was performed using Confocal microscopy was used to determine the the Hansatome (Bausch & Lomb) microkeratome. Post- number of keratocytes in the anterior and posterior operative drop regimen for the LASIK group was chlor- stromal layers, endothelial cell count, and number amphenicol drops for 3 days, betamethasone drops for of hexagonal cells. In the LASIK group, the kerato- 1 week, and artifi cial tears for 2 months postopera- cytes were counted in the retroablation zone, de- tively. Patients who underwent PRK were requested to fi ned as the space 5 µm behind the fl ap to the en- instill diclofenac sodium drops every 8 hours during dothelium. Keratocyte density was measured using the fi rst 24 hours, chloramphenicol drops for 5 days, S964 Commercially Sponsored Section journalofrefractivesurgery.com Keratocyte Density After LASIK and PRK/Amoozadeh et al betamethasone drops for 2 weeks, artifi cial tears for 1 month, and fl uorometholone drops for 12 weeks after 167 154.8 cessation of betamethasone drops.