Una nuova variante di cheratoplastica perforante: la PK decagonale con laser a femtosecondi


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Una nuova variante di cheratoplastica perforante: la PK decagonale con laser a femtosecondi

  1. 1. Femtosecond Laser-Assisted Decagonal Penetrating Keratoplasty HÉLÈNE PROUST, CHRISTOPHE BAETEMAN, FRÉDÉRIC MATONTI, JOHN CONRATH, BERNARD RIDINGS, AND LOUIS HOFFART● PURPOSE: To assess the use of a new polygonal treph- gical improvements lead to a gradual change in proceduresination pattern for penetrating keratoplasty (PK) assisted over the next several decades. High astigmatism is one ofby femtosecond laser. the major issues that can compromise a patient’s visual● DESIGN: Prospective, nonrandomized clinical study. rehabilitation after PK.2 Numerous procedures have been● METHODS: Sixteen eyes underwent decagonal PK. tested to correct astigmatic errors, including relaxingNine had Fuchs dystrophy, 4 had pseudophakic bullous procedures, wedge resections, and photorefractive proce-keratopathy, 1 had experienced trauma, 1 had corneal dures.3–9 Femtosecond laser is a step forward in theamyloidosis, and 1 had keratoconus. A Femtec (Tecnolas refinement of corneal surgery. By performing the samePerfectVision) laser was used to create decagonal pene- trephination profile on both donor and recipient tissue,trating cuts on both donor and recipient corneas. All laser-assisted surgery may improve wound adaptation andpatients were evaluated for uncorrected visual acuity, visual outcomes.10best spectacle-corrected visual acuity, pachymetry, to- The femtosecond laser is a focusable infrared laser thatpography, and endothelial cell density. Scanning electron delivers ultrashort pulses in the femtosecond durationmicroscopy was performed on corneal tissue after sur- range. Contiguous pulses are placed at a definite depthgery. within the cornea, thus resecting only targeted tissue. This● RESULTS: All eyes were treated successfully without surgical device allows cutting of corneal tissue in a numberintraoperative complications. The mean follow-up of transplant designs and allows the use of sagittal planestandard deviation was 9.75 3.5 months. Mean post- trephination profiles, such as zigzag,11,12 top-hat,13 Christ-operative best spectacle-corrected visual acuity was 20/ mas tree,14 and mushroom15 shapes to improve wound53, and there was a significant improvement in both stability14 and to reduce postoperative astigmatism.11 Theuncorrected visual acuity (P .0019) and best specta- Technolas laser (Technolas PerfectVision, Heidelberg,cle-corrected visual acuity (P .001). At 6 months, Germany) can be used to create additional geometricmean standard deviation manifest astigmatism was configurations to increase the postoperative results of1.90 1.20 diopters. Mean endothelial cell density was keratoplasty. Herein, we report outcomes of a decagonal-1502 458 cells/mm2. Scanning electron microscopy shaped PK cut profile using a femtosecond laser.displayed straight decagonal cut margins and minor re-maining tissue bridges.● CONCLUSIONS: Use of the decagonal trephination pro-file was effective and safe to perform PK. Short-term METHODSvisual results and refractive results are encouraging ALL SURGERIES WERE PERFORMED IN OUR DEPARTMENTcompared with those of conventional PK studies. Longer- from November 2007 through April 2008. All proceduresterm follow-up and comparative studies are necessary to were performed by one surgeon (H.P.). Corneal trephina-determine precisely advantages the and optimal surgical tion was performed using a Technolas laser. This apparatussettings of this technique. (Am J Ophthalmol 2011; consists of a pulsed solid body (neodymium:glass) laser151:29 –34. © 2011 by Elsevier Inc. All rights reserved.) with a repetition rate of 40 kHz that emits light with a wavelength of 1055 nm and a pulse duration of 800S INCE THE FIRST HUMAN FULL-THICKNESS CORNEAL femtoseconds. The laser was set to obtain a decagonal transplantation, performed by Zirm in 1906, pene- corneal button of 8.0-mm decagonal trephination on trating keratoplasty (PK) has grown to be the most donor tissue and an 8.0-mm decagonal trephination onfrequently performed tissue transplant in the world.1 Sur- recipient tissue. The cut rim angle was set at 90 degrees to the corneal surface tangent. Laser beam energy of 3.2 toAccepted for publication July 20, 2010. From the Department of Ophthalmology, Aix-Marseille University, 3.4 J and a spot separation of 3 m were chosen. AHopital de la Timone Adultes, Marseille, France (H.P., C.B., F.M., J.C., ˇ double-pass procedure was used to reduce the amount ofB.R., L.H.). residual tissular bridges and to ease the opening of the Inquiries to Louis Hoffart, Service d’Ophthalmologie, Hopital de la ˇTimone Adultes, 264 rue Saint-Pierre, 13385 Marseille cedex 05, France; wound (the first laser cut started from 1300 to 200 me-mail: louis.hoffart@ap-hm.fr under the corneal surface and a second laser cut followed0002-9394/$36.00 © 2011 BY ELSEVIER INC. ALL RIGHTS RESERVED. 29doi:10.1016/j.ajo.2010.07.016
  2. 2. TABLE 1. Decagonal Keratoplasty with the Technolas Femtosecond Laser: Patient Demographic Characteristics Patient No. Gender Age (yrs) Eye Primary Diagnosis Secondary Diagnosis 1 F 82 Left Failed DSAEK ARMD 2 M 28 Left Corneal amyloidosis None 3 F 77 Right Pseudophakic bullous keratopathy ARMD 4 M 72 Left Fuchs dystrophy Allograft rejection 5 F 80 Left Failed DSAEK None 6 F 82 Left Fuchs dystrophy None 7 M 78 Right Pseudophakic bullous keratopathy None 8 M 55 Left Pseudophakic bullous keratopathy None 9 F 80 Left Fuchs dystrophy None 10 F 92 Right Fuchs dystrophy None 11 M 90 Right Fuchs dystrophy Macular edema 12 M 72 Left Trauma Epiretinal membrane 13 M 72 Right Fuchs dystrophy Diabetic macular edema 14 F 80 Left Fuchs dystrophy None 15 F 21 Left Keratoconus None 16 F 82 Left Pseudophakic bullous keratopathy None ARMD age-related macular degeneration; DSAEK Descemet automated stripping endothelial keratoplasty; F female; M male; yrs years.the same pattern up to the surface). Donor corneoscleraldiscs were mounted on an artificial anterior chamber(Moria, Anthony, France) filled with BSS plus (Alcon,Fort Worth, Texas, USA) up to a pressure of 20 mm Hg.Corneal thickness subsequently was determined by ultra-sound pachymetry (UP-1000; Nidek, Tokyo, Japan). Cor-neoscleral discs were placed under the laser through aconcave patient interface and were centered using thereflex image of the diode lights of the laser. After trephi-nation, the corneal button was separated by blunt dissec-tion with a Sinskey manipulator (Moria, Anthony,France). On the recipient, the procedure required place-ment of a suction ring. After a proper vacuum seal wasobtained, the patient interface was applied and the laserprocedure was performed as previously described. After FIGURE 1. Slit-lamp photographs showing postoperative fea- tures of decagonal keratoplasty with the Technolas femtosecondcompletion of the cut, the eye was disconnected from the laser. (Left) Slit-lamp photograph of decagonal-shaped cut pene-laser and the corneal button was removed by dissection trating keratoplasty at 1 month. (Right) Perfect apposition ofwith a blunt hook. The corneal graft then was positioned donor and recipient wounds (white arrow). Running sutureon the recipient eye after filling of the anterior chamber penetrates in the middle of each side of the polygon (black arrow).with a cohesive viscoelastic (Healon GV; AMO, Ettlin-gen, Germany) and was secured by a combination of 10interrupted sutures on each angle of the decagon, then a to logarithm of minimal angle of resolution units. Statis-10-bite running suture was added on the middle of each tical analysis was performed using SPSS software versiondecagon side. After surgery, patients received tobramycin 13.0 (SPSS, Inc, Chicago, Illinois, USA). The Wilcoxonplus dexamethasone eyedrops (TobraDex; Alcon) 3 times rank-sum test was used to assess the difference betweendaily for 1 month and then slowly tapered. Selective suture examinations. A P value .05 was considered significant.removal was performed starting at the first-month visit if Endothelial cell density and central corneal pachymetryastigmatism was more than 3 diopters (D). were obtained by noncontact specular microscopy (Top- Postoperative examinations were planned at 7 days and con SP-2000 and ImageNet Computerized Analysis Sys-at 1, 3, 6, 9, and 12 months. The following data were tem; Topcon, Tokyo, Japan).evaluated: uncorrected visual acuity, best spectacle-cor- Objective astigmatism, specular corneal topography,rected visual acuity, slit-lamp biomicroscopy, and dilated surface regularity index, surface asymmetry index, and thefundus examination. All visual acuity data were converted root mean square at 3 mm were evaluated using the OPD30 AMERICAN JOURNAL OF OPHTHALMOLOGY JANUARY 2011
  3. 3. TABLE 2. Decagonal Keratoplasty with the Technolas Femtosecond Laser: Visual Acuity at 6 Months after Surgery Preoperative Preoperative Postoperative Postoperative Patient No. UCVA BSCVA UCVA BSCVA 1 20000 20000 2000 2000 2 2000 200 63 32 3 2000 200 400 400 4 200 200 32 32 5 2000 2000 100 40 6 2000 160 50 40 7 2000 2000 63 25 8 60 50 50 30 9 200 200 50 50 10 2000 2000 63 25 11 2000 2000 100 100 12 400 133 40 40 13 200 133 400 63 14 2000 80 50 40 15 400 100 50 25 16 2000 2000 50 30 Mean standard deviation (lines) 20/783 6 20/356 7.1 20/82 4.6 20/53 5 BSCVA best spectacle-corrected visual acuity; UCVA uncorrected visual acuity.scan station (ARK 10000; NIDEK). All patients under-went anterior segment optical coherence tomography to TABLE 3. Decagonal Keratoplasty with the Technolasevaluate wound alignment and suture configurations (Stra- Femtosecond Laser: Visual Acuity Results at 6 Monthstus OCT 3; Carl-Zeiss Meditec, Inc, Jena, Switzerland). after SurgeryAll donor corneoscleral rims and recipients corneal but- Mean UCVA Mean BSCVAtons were fixed with 2.5% glutaraldehyde for scanning Variable LogMAR Snellen LogMAR Snellenelectron microscopy (Philips XL 30S, Amsterdam, TheNetherlands). Initial examination 1.59 0.60 20/783 1.25 0.71 20/356 Final examination 0.61 0.46 20/82 0.42 0.50 20/53 P value .0019 .001 RESULTS BSCVA best spectacle-corrected visual acuity; logMAR logarithm of minimal angle of resolution; UCVA uncorrectedSIXTEEN EYES OF 16 PATIENTS (7 MEN AND 9 WOMEN) WITH visual acuity.a mean age standard deviation (SD) of 71.4 20.1 years(range, 21 to 92 years) underwent decagonal penetrating resolution was observed after 7 days. Epithelialization waskeratoplasty. Mean follow-up SD was of 9.75 3.5 completed after 3 days in all cases.months. Surgical indication was Fuchs dystrophy in 9 Visual outcomes are summarized in Tables 2 and 3.patients (56.25%), pseudophakic bullous keratopathy in 4 Significant differences between preoperative and postop-patients (25%), keratoconus in 1 patient (6.25%), trauma erative uncorrected visual acuity and best spectacle-cor-in 1 patient (6.25%), and corneal amyloidosis in 1 patient rected visual acuity were observed (P .0019 and P(6.25%). Preoperative examination showed 2 cases of .0010, respectively). If the 5 cases of preoperative retinalage-related macular degeneration, 1 case of diabetic mac- impairment were excluded from the analysis, postoperativeular edema, and 1 case of epiretinal macular membrane. uncorrected visual acuity and best spectacle-corrected All procedures were uneventful. After surgery, 1 case of visual acuity would have been 20/54 0.12 (range, 20/100macular edema was diagnosed and 1 patient showed a to 20/33) and 20/33 0.10 (range, 20/50 to 20/25),retinal detachment at the third postoperative month and respectively.required subsequent surgery. Demographic and diagnostic Refractive data for each patient are summarized in Tabledata are summarized in Table 1. A perfect wound apposi- 4. Mean SD postoperative spherical equivalent wastion was observed in all cases, as illustrated in Figure 1. An 0.10 1.83 D (range, 0.88 to 5.00 D). Mean SDearly and transient postoperative corneal edema was noted manifest and topographic postoperative astigmatism werearound the surgical wound after laser exposure; complete 1.90 1.20 D (range, 0.00 to 5.00 D) and 3.10 1.20 DVOL. 151, NO. 1 DECAGONAL PENETRATING KERATOPLASTY 31
  4. 4. TABLE 4. Decagonal Keratoplasty with the Technolas Femtosecond Laser: Outcomes at 6 Months after Surgery Postoperative Manifest Refraction (D) Postoperative ECD Patient No. Sphere Cylinder K (D) (cells/mm2) 1 1.00 2.00 5.73 2000 2 0.25 1.50 3.54 1200 3 1.50 1.50 1.80 946 4 5.00 2.00 2.25 1216 5 0.00 0.00 2.65 950 6 0.75 2.00 2.79 1400 7 0.25 3.00 4.04 1751 8 1.25 0.75 0.69 1600 9 2.50 3.00 2.45 2500 10 0.25 1.50 2.93 1830 11 2.00 2.00 1.45 993 12 0.00 0.00 4.44 1400 13 4.00 2.00 2.49 800 14 0.50 1.50 2.33 2000 15 3.00 5.00 5.38 1840 16 2.25 2.50 3.43 1600 Mean standard deviation 0.8 1.9 1.9 1.2 3.1 1.2 1502 458 D diopters; ECD endothelial cell density; K topographic cylinder obtained from the diffe- rence of topographic sim K measurements.(range, 0.89 to 5.73 D), respectively. Mean SD postop- plasty. To our knowledge, this study is the first report oferative root mean square was 2.00 1.29 D (range, 0.69 to femtosecond-assisted decagonal keratoplasty outcomes in4.90 D), surface regularity index was 1.39 0.36 (from the literature.0.56 to 2.13), and the surface asymmetry index was 1.33 Femtosecond laser is a focusable infrared laser that0.80 (from 0.52 to 3.91). At 6 months, mean SD central delivers ultrashort pulses in the femtosecond durationcorneal pachymetry and endothelial cell density were range. Contiguous pulses are placed at a precise depth508 32 m and 1502 458 cells/mm2, respectively. within the cornea. The laser spot may be fired in a vertical pattern for trephination cuts or parallel to the corneal surface to achieve lamellar cuts. At the moment, the DISCUSSION United States Food and Drug Administration approve 4 lasers for corneal surgery: IntraLase (IntraLase Corp,RECENT CORNEAL GRAFT TECHNIQUES, INCLUDING LAMEL- Irvine, California, USA), Technolas (Tecnolas Perfectvi-lar keratoplasty procedures, offer significant advantages sion), Femto LDV (Ziemer Ophthalmic Systems AG, Port,over conventional PK, but they cannot address the full Switzerland), and the VisuMax laser (Carl Zeiss Meditecrange of corneal diseases and show limitations.16 –18 This AG, Jena, Germany). The salient feature of the Technolassuggests that improvements in PK will result in significant laser is a curved patient interface that minimizes thebenefits to patients. PK involves a skilled trephination intraocular pressure rise during surgery and induces lesstechnique to avoid irregular or decentered cuts that could posterior stromal distortion during aplanation.23lead to astigmatism and high refractive errors.19 Optimal Compared with mechanical PK, femtosecond laser-postoperative outcome is dependent on a centered perpen- assisted PK creates a more stable wound configura-dicular cut and a well-matched donor button to recipient tion12,14,22,24,25 by producing accurate cuts that allow abed.20 Limitations of manual trephination include inaccu- better fit between donor and recipient. A better appo-rate cut of donor and recipient corneal edges, leading to sition between tissues was observed, and consequently,potential graft– host wound mismatch.21 Advantages of- faster healing was associated with an earlier functionalfered by laser-assisted keratoplasty consist of stronger recovery.13 Several transplant forms have been reportedwound healing by better wound adaptation22 and reduc- in previous studies. Most of them concern the top-hattion of corneal astigmatism by performing same cut profile profile that was found to be biomechanically more stableon both donor and recipient.12,13 Decagonal keratoplasty is than the traditional circular, mushroom, zigzag, ora new trephination pattern that may allow optimizing Christmas tree designs.14 Price and Price reported 1-yearpostoperative refractive results of full-thickness kerato- postoperative outcomes of top-hat PK and concluded32 AMERICAN JOURNAL OF OPHTHALMOLOGY JANUARY 2011
  5. 5. ond laser-assisted corneal graft procedures may gain stability and improved centration from noncircular trans- plant shapes. The advantages of the decagonal cut com- pared with circular transplants consist of the absence of rotation or decentration during the procedure, the decagon corners being distributed evenly over the entire transplant periphery, which is then optimally protected against rota- tions. The decagonal shape decreases the torque effect associated with a circular graft during suture placement and allows a stable fitting of the graft into the recipient, because it provides the strength of an circular shape with the stability given by the angles of the polygonal design. The postoperative refractive results of this study (mean &plusmn SD refractive astigmatism of 1.9 1.2 D at 6 months) compares favorably with that reported in larger series using conventional PK laser-assisted22,26 –28 and mechanical trephination29,30 techniques. The accurate fit of donor and recipient was confirmed in our study by scanning electron microscopy (Figure 2). Also, the safetyFIGURE 2. Scanning electron microscopy image showing do- of the procedure, evaluated by postoperative endothelialnor corneoscleral rims after decagonal keratoplasty with the cell density, was similar to that of previous studies usingTechnolas femtosecond laser. Note the angle of the decagonal- conventional PK.31,32shaped cut (white arrow) and the straight, orthogonal, and crisp We found that decagonal PK performed with femtosec-trephination edge. The stromal surface of the recipient showed ond laser to be an easy, safe, and effective means tominor remaining tissue bridges. improve postkeratoplasty outcomes. Polygonal trephina- tion for PK provides an accurate positioning and an easierthat this configuration allowed for an increased area of suturing of the graft in the recipient bed by reducingwound healing.13 Farid and associates reported similarly rotational slippage of the graft. The potential advantagesgood results using a zigzag profile.12 Moreover, it seems of this new pattern compared with circular keratoplastylogical that the suture pattern and the shape of the may include an enhancement of visual results and recoverytransplant influence rotation and that the rotational times. However, longer follow-up and comparative studiesability of the graft is greater with the zigzag form and are needed to provide more confident estimates of astig-traditional round form than with a decagonal or non- matism reduction proportions for each surgical method.circular form. Subsequently, these studies could seek other means for The use of a polygonal trephination pattern may im- improving visual outcomes, such as an optimal number ofprove the outcomes of keratoplasty. Moreover, femtosec- sides of the decagonal PK.PUBLICATION OF THIS ARTICLE WAS SUPPORTED IN PART BY THE DEPARTMENTAL COUNCIL OF “BOUCHES-DU RHONE” (CG13,Marseille, France). The authors indicate no financial conflict of interest. Involved in Design of study (C.B., H.P., L.H.); Conduct of study (C.B., L.H.,H.P.); Data collection (C.B.); Analysis and interpretation of data (L.H., F.M., C.B.); Statistical expertise (J.C.); Writing the article (L.H., C.B.); andCritical revision of the article (H.P., B.R., J.C.). Approval from the Institutional Review Board, Marseille, France, was obtained. The study wasconducted with institutional approval and in accordance with the Declaration of Helsinki. All patients signed informed consent statements. The authorsthank Joël Courageot, Department of Electronic Microscopy, Faculty of Medicine, Aix-Marseille University, for his technical support in scanningelectron microscopy specimen preparation and imaging. REFERENCES ing keratoplasty: a vector analysis and topographic study. Klin Monbl Augenheilkd 1997;211(3):151–158. 1. Zirm E. Eine erfolgreiche totale keratoplastik (A successful 4. Tuunanen TH, Ruusuvaara PJ, Uusitalo RJ, Tervo TM. total keratoplasty). 1906. Refract Corneal Surg 1989;5(4): Photoastigmatic keratectomy for correction of astigmatism in 258 –261. corneal grafts. Cornea 1997;16(1):48 –53. 2. Claesson M, Armitage W. Ten-year follow-up of graft sur- 5. Fronterre A, Portesani GP. Relaxing incisions for postkera- vival and visual outcome after pentrating keratoplasty in toplasty astigmatism. Cornea 1991;10(4):305–311. Sweden. Cornea 2009;28(10):1124 –1129. 6. Hovding G. Transverse keratotomy in postkeratoplasty astig- 3. Vilchis E, Seitz B, Langenbucher A, Kuchle M, Kus MM, matism. Acta Ophthalmol (Copenh) 1994;72(4):464 – 468. Naumann GO. Limbus-parallel keratotomies with compres- 7. Kirkness CM, Ficker LA, Steele AD, Rice NS. Refrac- sion sutures in treatment of high astigmatism after perforat- tive surgery for graft induced astigmatism after penetratingVOL. 151, NO. 1 DECAGONAL PENETRATING KERATOPLASTY 33
  6. 6. keratoplasty for keratoconus. Ophthalmology 1991;98(12): 20. Perlman EM. An analysis and interpretation of refractive 1786 –1792. errors after penetrating keratoplasty. Ophthalmology 1981; 8. Ghanem R, Azar D. Femtosecond-laser arcuate wedge- 88(1):39 – 45. shaped resection to correct high residual astigmatism after 21. Busin M. A new lamellar wound configuration for penetrat- penetrating keratoplasty. J Cataract Refract Surg 2006;32(9): ing keratoplasty surgery. Arch Ophthalmol 2003;121(2): 1415–1419. 260 –265. 9. Kiraly L, Herrmann C, Amm M, Duncker G. [Reduction of 22. Birnbaum F, Maier P, Reinhard T. Femtosecond laser)as- astigmatism by arcuate incisions using the femtosecond laser sisted penetrating keratoplasty. Ophthalmologe 2010;107(2): after corneal transplantation]. Klin Monatsbl Augenheilkd 186 –188. 2008;225(1):70 –74. 23. Lubatschowski H. Overview of commercially available fem-10. Soong H, Malta J. Femtosecond lasers in ophthalmology. tosecond lasers in refractive surgery. J Refract Surg 2008; Am J Ophthalmol 2009;147(2):189 –197. 24(1):S102–S107.11. Farid M, Steinert R. Deep anterior lamellar keratoplasty 24. Chan C, Ritenour R, Kumar N, Sansanayudh W, Rootman performed with the femtosecond laser zigzag incision for the DS. Femtosecond laser-assisted mushroom configuration treatment of stromal corneal pathology and ectatic disease. J deep anterior lamellar keratoplasty. Cornea 2010;29(3):290 – Cataract Refract Surg 2009;35(5):809 – 813. 295.12. Farid M, Kim M, Steinert R. Results of penetrating kerato- 25. Price FW Jr, Price MO, Grandin JC, Kwon R. Deep anterior lamellar keratoplasty with femtosecond-laser zigzag incisions. plasty performed with a femtosecond laser zigzag incision J Cataract Refract Surg 2009;35(5):804 – 808. initial report. Ophthalmology 2007;114(12):2208 –2212. 26. Hoffart L, Proust H, Matonti F, Ridings B, Conrath J.13. Price FW Jr, Price MO. Femtosecond laser shaped penetrat- Short-term results of penetrating keratoplasty performed ing keratoplasty: one-year results utilizing a top-hat config- with the Technolas femtosecond laser. Am J Ophthalmol uration. Am J Ophthalmol 2008;145(2):210 –214. 2008;146(1):50 –55.14. Bahar I, Kaiserman I, McAllum P, Rootman DS. Femtosec- 27. Farid M, Steinert R, Gaster R, Chamberlain W, Lin A. ond laser-assisted penetrating keratoplasty: stability evalua- Comparison of penetrating keratoplasty performed with a tion of different wound configurations. Cornea 2008;27(2): femtosecond laser zig-zag incision versus conventional blade 209 –211. trephination. Ophthalmology 2009;116(9):1638 –1643.15. Buratto L, Bohm E. The use of the femtosecond laser in 28. Por Y, Cheng JY, Parthasarathy A, Mehta JS, Tan DT. penetrating keratoplasty. Am J Ophthalmol 2007;143(5): Outcomes of femtosecond laser-assisted penetrating kerato- 737–742. plasty. Am J Ophthalmol 2008;145(5):772–774.16. Krumeich JH, Knülle A, Krumeich BM. Deep anterior 29. Bahar I, Kaiserman I, Lange AP, et al. Femtosecond versus lamellar (DALK) vs. penetrating keratoplasty (PKP): a manual dissection for top hat penetrating keratoplasty. Br J clinical and statistical analysis. Klin Monatsbl Augenheilkd Ophthalmol 2009;93(1):73–78. 2008;225(7):637– 648. 30. Bahar I, Kaiserman I, Srinivasan S, Ya-Ping J, Slomovic A,17. Shimazaki J, Shimmura S, Ishioka M, Tsubota K. Random- Rootman DS. Comparison of three different techniques of ized clinical trial of deep lamellar keratoplasty vs penetrating corneal transplantation for keratoconus. Am J Ophthalmol keratoplasty. Am J Ophthalmol 2002;134(2):159 –165. 2008;146(6):905–912.18. Maier P, Reinhard T. Keratoplasty: laminate or penetrate? 31. Nishimura JK, Hodge DO, Bourne WM. Initial endothelial Part 2: lamellar keratoplasty. Ophthalmologe 2009;106(7): cell density and chronic endothelial cell loss rate in corneal 649 – 662. transplants with late endothelial failure. Ophthalmology19. Cheng YY, Pels E, Nuijts RM. Femtosecond-laser-assisted 1999;106(10):1962–1965. Descemet’s stripping endothelial keratoplasty. J Cataract 32. Claerhout I, Beele H, Kestelyn P. Graft failure: I. Endothelial Refract Surg 2007;33(1):152–155. cell loss. Int Ophthalmol 2008;28(3):165–173.34 AMERICAN JOURNAL OF OPHTHALMOLOGY JANUARY 2011