putcutc

(et(tc0fk.e

7

Simultaneous Topography-Guided PhotorefractiveKeratectomy Followed by Corneal Collagen

Cross-linking for Keratoconus

GEORGE D. KYMIONIS, DIMITRA M. PORTALIOU, GEORGE A. KOUNIS, ALIKI N. LIMNOPOULOU,

GEORGIOS A. KONTADAKIS, AND MICHAEL A. GRENTZELOS

t

ttptp

(ssk�lo

● PURPOSE: To present the long-term results after simul-taneous photorefractive keratectomy followed by cornealcollagen cross-linking for keratoconus.● DESIGN: Prospective, interventional, consecutive caseseries.● METHODS: In this study, 26 patients (31 eyes) withrogressive keratoconus were included. All patientsnderwent customized topography-guided photorefrac-ive keratectomy immediately followed by cornealollagen cross-linking with the use of riboflavin andltraviolet A irradiation. Epithelium was removed byransepithelial phototherapeutic keratectomy in allases.

● RESULTS: Mean follow-up was 19.53 � 3.97 monthsrange, 12 to 25 months). Mean preoperative sphericalquivalent was �2.3 � 2.8 diopters (D), whereas athe last follow-up examination, it was significantlyP < .001) reduced to �1.08 � 2.41 D. Logarithm ofhe minimal angle of resolution uncorrected and best-orrected visual acuity were reduced significantly by.46 and 0.084 (P < .001), respectively, at the lastollow-up examination. Finally, mean steep and flateratometry readings were reduced by 2.35 (P <001) and 1.18 (P � .013) at the last follow-upxamination.

● CONCLUSIONS: Simultaneous photorefractive keratec-tomy followed by corneal collagen cross-linking seems tobe a promising treatment alternative in our series ofkeratoconic patients. (Am J Ophthalmol 2011;152:748–755. © 2011 by Elsevier Inc. All rights reserved.)

C ORNEAL ECTATIC DISORDERS, THE MOST COMMON

of which is keratoconus, represent a group ofconditions characterized by a bilateral state of

tectonic corneal weakness that can lead to thinning of thecornea, production of myopia and irregular astigmatism,and visual acuity deterioration.

Accepted for publication Apr 25, 2011.From the Institute of Vision and Optics University of Crete, Herak-

lion, Crete, Greece (G.D.K., D.M.P., G.A.Kou., A.N.L., G.A.Kon.,M.A.G.).

Inquiries to George D. Kymionis, Institute of Vision and Optics,University of Crete, Medical School, 71003 Heraklion, Crete, Greece;

e-mail: kymionis@med.uoc.gr

© 2011 BY ELSEVIER INC. A48

Treatment options comprise 2 general approaches:first, vision restoration by means of spectacles or rigidgas permeable contact lenses, and second, restoration ofthe tectonic integrity of the cornea by means of intra-corneal ring segments1 and corneal collagen cross-linking (CXL).2 In advanced cases of keratoconus,lamellar or penetrating keratoplasty3 are consideredessential treatment possibilities to improve patients’quality of life.

CXL is a minimally invasive technique using riboflavinand ultraviolet A irradiation for the enhancement ofcorneal rigidity. In particular, CXL augments the biome-chanical strength of the cornea by inducing interfibrillarcross-links of the stroma,4 resulting in the stabilization ofhe disorder. Results5–7 are extremely promising, but poor

vision quality in patients remains largely unchanged, withminimal improvement.

Several combined procedures have been proposed tooptimize the CXL result (such as conductive keratoplastyfollowed by CXL8), but it seems that the most effective ishe combined topography-guided photorefractive keratec-omy (PRK) followed by CXL.9 In this case series, weresent the long-term results of simultaneous customizedopography-guided surface ablation followed by CXL inatients with keratoconus.

METHODS

IN THIS STUDY, 26 PATIENTS (31 EYES; CONSECUTIVE

series), 18 men and 8 women, with progressive kerato-conus were included. Mean age was 29.3 � 8.5 yearsrange, 19 to 49 years). Inclusion criteria were progres-ive keratoconus (keratoconus was described as progres-ive when there was an increase in the cone apexeratometry of �0.75 diopters [D] or an alteration of0.75 D in the spherical equivalent refraction in the

ast 6 months), expected corneal thickness at the apexf the cone after PRK more than 400 �m, and no other

corneal pathologic signs.

● CLINICAL EVALUATION: Preoperative evaluation con-sisted of general and ocular health history assessment;

corneal topography (iTrace; Tracey Tech, Houston, Texas,

LL RIGHTS RESERVED. 0002-9394/$36.00doi:10.1016/j.ajo.2011.04.033

icobtzaWwasco

anda

USA); assessment of uncorrected visual acuity (UCVA),best-corrected visual acuity (BCVA), and manifest refrac-tion; scotopic pupillometry; central ultrasound pachymetry(Corneo-Gage Plus; Sonogage, Inc, Cleveland, Ohio,USA) and slit-lamp examination of the anterior andposterior segments of the eyes. Patients’ preoperative andpostoperative data are shown in the Table.

All data were analyzed for normality. Because not allvariables followed normal distribution, the Wilcoxonsigned-rank paired test (PASW Statistics 18; SPSS, Inc,Chicago, Illinois, USA) was used. All diagrams and fre-quency analyses were performed by Microsoft Excel 2007(Microsoft, Inc, Redmond, Washington, USA). Harrisnotation was used to calculate the mean values of refrac-

FIGURE 1. Bar graph showing change in best-corrected visual afollowed by corneal collagen cross-linking for keratoconus confiBCVA at the last follow-up examination. PostOp � after surg

TABLE. Preoperative and Last follow-up Refractive, Visual,Topography-Guided Photorefractive Kera

MRSE (D) UC

Preoperative data

Mean � SD –2.3 � 2.8 0.

Range 1.63 to �12.88 0.5

Mean refraction (range) –2.06/–0.48@77°

Postoperative data (mean

follow-up, 19.53 � 3.97

mos; range, 12 to 25 mos)

Mean � SD –1.08 � 2.41 0.

Range 2.5 to �8.75 0.4

Mean refraction (range) –0.81/–0.54@74°

BCVA � best-corrected visual acuity; D � diopters; K � keratom

months; MRSE � manifest refraction spherical equivalent; SD � st

tions in the Table. u

LONG-TERM RESULTS OF SIMULTANEOUSVOL. 152, NO. 5

● SURGICAL PROCEDURE: All procedures were performedn our institution by the same surgeon (G.D.K.) under sterileonditions. After topical anesthesia with tetracaine 1% andxybuprocaine 0.4% eyedrops, the epithelium was removedy transepithelial phototherapeutic keratectomy (PTK). Theransepithelial PTK ablation was performed in an 8.0-mmone in an intended depth of 50 �m. A solid-state laser with

wavelength of 213 nm (Pulzar Z1; CustomVis, Perth,ashington, USA) was used for the PRK procedure. The

avelength is generated using a major neodymium:yttrium–luminum–garnet laser system of 1064 nm, and throughpecial cultivated crystals, the 213 nm is used finally. Theustomization was performed based on the topographic databtained by the iTrace technology. System software allows

y (BCVA; safety) before and after photorefractive keratectomyg that approximately 50% of patients gained 1 line or more of

Keratometric Data in Keratoconic Patients after Combinedmy and Corneal Collagen Cross-linking

gMAR) BCVA (logMAR) Steep and Flat Keratometry (D)

0.18 0.81 � 0.65 49.8 � 5.3

0.06 2 to 0.1 67.7 to 42.7

3 20/131 44.39 � 4.8 (62.75 to 37.1)

0.15 0.35 � 0.36 47.46 � 4.3

0.04 1.2 to �0.04 53 to 37.48

7 20/46 43.21 � 3.4 (53 to 37.48)

s; logMAR � logarithm of the minimal angle of resolution; mos �

rd deviation; UCVA � uncorrected visual acuity.

cuitrminery.

andtecto

VA (lo

21 �

4 to �

20/3

12 �

6 to �

20/2

etric

sing a percentage of customization from 0% to 100%.

PRK AND CXL FOR KERATOCONUS 749

doCUa

ucFt

Because of continuous flattening after CXL, as shown inprevious studies,10 the relevant attempted correction was upto 60% of sphere and cylinder of patient refractive error,whereas customization was adjusted from 0% to 100%. Using0% would be equivalent to a conventional laser treatment,and 100% would be equivalent to a full customized treatment.Adjusting this percentage could lower the maximum depth oftissue removed (upper limit, 50 �m). Treatment modifica-tions (attempted correction and percentage of customization)were based on preoperative corneal pachymetry, correcteddistance visual acuity, and manifest refraction to arrive to amaximum ablation depth of 50 �m.

Next, riboflavin (0.1% solution of 10 mg riboflavin-5-phosphate in 10 mL dextran-T-500 20% solution) wasapplied every 3 to 5 minutes for approximately 30minutes until the stroma was penetrated completely andaqueous was stained yellow (riboflavin shielding). Acommercially available ultraviolet A system (UV-Xillumination system, version 1000; IROC AG R&D,Zürich, Switzerland) with Koehler optics was used forultraviolet A irradiation. Before treatment, the in-tended 3-mW/cm2 surface irradiance (5.4 J/cm2 surface

ose after 30 minutes) was calibrated using the ultravi-let A meter YK-34UV (Lutron Electronic EnterpriseO, LTD, Taipei, Taiwan), which is supplied with theV-X device. During treatment, riboflavin solution was

pplied every 3 to 5 minutes to ensure saturation.After the treatment, a bandage contact lens was applied

ntil the epithelium healed completely, followed by appli-ation of fluorometholone 0.1% eyedrops (FML Liquifilm;ALCON pharmaceuticals, Ltd, Fort Worth, Texas, USA)

FIGURE 2. Bar graph showing uncorrected visual acuity (UCVby corneal collagen cross-linking for keratoconus representingor better at the last follow-up examination.

wice daily for 2 weeks.

AMERICAN JOURNAL OF750

RESULTS

ALL VALUES ARE EXPRESSED AS MEAN � STANDARD DEVIATION

and range. Visual acuity is expressed as logarithm of minimalangle of resolution (logMAR) visual acuity. All values presentedstatistically significant differences between preoperative and lastpostoperative period. Patient postoperative data at the last

fficacy) before and after photorefractive keratectomy followedcrease in the percentage of eyes that achieved UCVA 20/100

FIGURE 3. Box plot showing flat and steep keratometric readingsafter photorefractive keratectomy followed by corneal collagen cross-linking before surgery, at 1, 3, 6, and 12 months after treatment, andat the last follow-up, showing a significant decrease that remainedstable throughout the follow-up. PreOp � before surgery; PostOp �after surgery.

A; ean in

follow-up examination are shown in the Table.

OPHTHALMOLOGY NOVEMBER 2011

iL0fp

1(eoshflazaess

cdtaTa

s

Mean follow up was 19.53 � 3.97 months (range, 12 to25 months). Mean preoperative spherical equivalent was�2.3 � 2.8 D, whereas at the last follow-up examination,t was reduced significantly (P � .001) to �1.08 � 2.41 D.ogMAR UCVA and BCVA were reduced significantly by.46 and 0.084 logMAR units (P � .001) at the lastollow-up examination. There were no intraoperative or

FIGURE 5. Scatterplot showing the spherical equivalent refrac-tion changes at 1 year and during the last follow-up period aftercombined topography-guided photorefractive keratectomy andcorneal collagen cross-linking, confirming a continuous refractionimprovement after surgery (PostOp), even after the first 12months from the simultaneous treatment. D � diopters; SEQ �pherical equivalent refraction.

FIGURE 4. Graphs showing stability of (Left) uncorrected(BCVA) after combined photorefractive keratectomy and co24 months after treatment, and at the last follow-up. logMARsurgery.

ostoperative complications.

LONG-TERM RESULTS OF SIMULTANEOUSVOL. 152, NO. 5

Regarding the method’s safety, 48% (15/31) of eyes gainedline or more of UCVA at the last follow-up examination

Figure 1). At the last follow-up examination, 10% (3/31) ofyes lost 1 line of BCVA. As shown in Figure 2, 58% (18/31)f eyes were had UCVA of 20/100 (Snellen) or better afterurgery, whereas at the last follow-up, 87% (27/31) of eyesad UCVA of 20/100 (Snellen) or better. Mean steep andat keratometry readings were reduced by 2.35 D (P � .001)nd 1.18 D (P � .013), respectively (Figure 3). The stabili-ation of uncorrected distance and corrected distance visualcuity is demonstrated in Figure 4. Additionally, sphericalquivalent refractive changes in the follow-up period can beeen in Figure 5. Patient topographic improvement can beeen using the iTrace technology (Figure 6).

In 50% (16/31) of the eyes, a posterior linear stromal hazeorresponding to the central treated area of the cornea wasetectable by slit-lamp examination 1 year after surgery. Duringhe follow-up period, this finding gradually became less densend demonstrated an anterior movement (Figure 7, Top left andop right). This finding can be graded as mild haze (grade 1)ccording to the Fantes anterior stromal haze scale.11

In these eyes, using corneal confocal microscopy, we detectedan area with high reflectance at the level of the posterior stromacharacterized by spindle-shaped and hyperreflective structures(Figure 7, Bottom). The reflectance was located adjacent to theendothelial cell layer 1 year after surgery and gradually movedanteriorly throughout the follow-up. The highly reflectivespindle-shaped structures have been linked with migra-tion and activation of keratocytes, which become trans-formed during locomotion, assuming spindle- or needle-shaped morphologic features.12

DISCUSSION

CXL IS A SAFE AND EFFECTIVE TECHNIQUE FOR THE MAN-

al acuity (UCVA) and (Right) best-corrected visual acuityl collagen cross-linking treatment before surgery, at 12 andogarithm of the minimal angle of resolution; PostOp � after

visurnea

� l

agement of ecstatic disorders. It has been proven that CXL

PRK AND CXL FOR KERATOCONUS 751

FIGURE 6. Two examples (Left and Right) of comparative topographic maps before (Bottom right) and after (Upper right)photorefractive keratectomy followed by corneal collagen cross-linking treatment for keratoconus showing significant improvement(difference topographic map left).

AMERICAN JOURNAL OF OPHTHALMOLOGY752 NOVEMBER 2011

is effectual in stopping the progression of keratoconus byquasifreezing of the cornea,13 and in many cases helps toavoid corneal transplantation. Nevertheless, a patient’sbasic problem, which is the deterioration of their vision,remains. CXL combined with topography-guided PRK hasbeen developed to achieve an important goal: to offerpatients stability of their disorder and at the same time toassist them in achieving functional vision.

In this article, we report the long-term results of topography-guided PRK followed by CXL. All of our patients showedtopographic progression of their ectasia over a period of 6 months(although 1 of our patients was 49 years of age). In our case series,all the parameters analyzed—spherical equivalent refraction,

FIGURE 7. (Top left and Top right) Slit-lamp photographs o(PRK) followed by corneal collagen cross-linking (CXL) 1 yeatreated area. (Bottom) Corneal confocal microscopy image obreflective area and keratocyte-activated nuclei.

UCVA, BCVA, and mean keratometrics—showed a significant

LONG-TERM RESULTS OF SIMULTANEOUSVOL. 152, NO. 5

improvement that remained stable throughout the follow-upperiod. The safety index of the combined procedure was1.21 at the last follow-up examination. No eye has lostmore than 2 lines of UCVA at the last follow-up period.None of our patients showed topographic or clinicalsigns of keratoconus progression during the follow-upperiod. All patients showed marked improvement ofcorneal irregularity and visual acuity.

The epithelium was removed in all cases by transepithelialPTK. It is known that transepithelial PTK uses an excimer laserablation to remove the epithelium and to smooth the anteriorirregular cornea.14,15 The aim of transepithelial PTK was epithe-lial removal and anterior cornea smoothening to decrease further

atient treated with simultaneous photorefractive keratectomyer surgery. The linear posterior haze is evident in the centraled after combined PRK and CXL treatment showing a high

f a pr afttain

the irregular astigmatism.

PRK AND CXL FOR KERATOCONUS 753

ahctvp

gcP

otaac

In 50% of our patients, a posterior linear stromal hazedeveloped that was detectable by slit-lamp biomicroscopy.The posterior haze gradually moved anteriorly and becameless dense during the follow-up period. This finding could beattributed to the interaction between CXL and photoabla-tion, which results in keratocyte activation.16–18 This mayexplain the high reflectance revealed in our case series,because our patients underwent photoablation in combina-tion with CXL, and thereby photoablation may have inducedkeratocyte activation, causing posterior stromal hyperreflec-tance formation. Furthermore, the hyperreflective structurescould be associated with increased collagen deposition, col-lagen disorganization, and excessive production of extracel-lular material from the activated keratocytes.

The aim of customized PRK was to remodel the shape ofthe cornea and to decrease irregular astigmatism. Customiza-tion could be a percentage of the fully customized treatmentfor reducing the maximum depth of tissue removed over thecorneal irregularity. A maximum ablation depth of 50 �m hasbeen chosen in all cases to achieve a decrease in astigmatism,but also to avoid removing a significant amount of tissue thatwould jeopardize the biomechanical integrity of the cornea. It

has been demonstrated that even with a relatively small k

Giapitzakis I, Pallikaris IG. Conductive keratoplasty followed

1

1

1

1

1

AMERICAN JOURNAL OF754

correction, because of the thickness limitations, patients maybenefit substantially.

CXL alone is effective in achieving a cessation of thekeratoconus progression. However, studies confirm a visualacuity improvement that is not always considered sufficient tomeet patients’ needs.19 It seems that PRK followed by CXL isn improved version of CXL alone, because it combinesalting of ectasia and significant vision improvement. Thisombined treatment seems to be more appropriate for pa-ients with early keratoconus staging, because in more ad-anced cases, thinning of the cornea could limit theossibility of tissue removal by topography-guided PRK.

One of the limitations of our study is the lack of controlroup. The existence of a control group could facilitate theomparison between CXL alone and topography-guidedRK followed by CXL.In conclusion, long-term results confirm the advantages

f the combined PRK and CXL procedure. This combinedreatment is capable of offering patients a functional visionnd stabilization of the ecstatic disorder. Further follow-upnd additional cases must be reviewed to draw finalonclusions about the benefit of this surgical technique in

eratoconic patients.

THE AUTHORS INDICATE NO FINANCIAL SUPPORT OF FINANCIAL CONFLICT OF INTEREST. INVOLVED IN CONCEPTION DESIGNof study (G.D.K., M.A.G.); Data collection (A.N.L., G.A.Kon.); Management of data (D.M.P.); Analysis and interpretation of data (G.A.Kou.,G.A.Kon.); Statistical expertise (G.A.Kou.); Literature search (A.N.L.); Technical support (G.A.Kon.); Provision of patients (M.A.G.); Preparation ofmanuscript (D.M.P.); Writing the manuscript (A.N.L., M.A.G.); Review of manuscript (G.D.K., D.M.P.); Critical revision of the manuscript(G.A.Kou.); and Approval of the manuscript (G.D.K.). Patients were informed thoroughly about the experimental nature of the intervention, thepossible outcomes, and the current clinical experience and have written consent according to the Declaration of Helsinki and the institutionalguidelines. University of Crete Institutional Review Board/Ethics Committee approval was obtained.

REFERENCES

1. Kymionis GD, Siganos CS, Tsiklis NS et al. Long-termfollow-up of Intacs in keratoconus. Am J Ophthalmol 2007;143(2):236–244.

2. Tan DT, Por YM. Current treatment options for cornealectasia. Curr Opin Ophthalmol 2007;18(4):284–289.

3. Yildiz EH, Cohen EJ, Virdi AS, Hammersmith KM, Laibson PR,Rapuano CJ. Quality of life in keratoconus patients after penetrat-ing keratoplasty. Am J Ophthalmol 2010;149(3):416–422.

4. Wollensak G. Cross inking treatment of progressive kerato-conus: new hope. Curr Opin Ophthalmol 2006;17:356–360.

5. Coskunseven E, Jankov MR 2nd, Hafezi F. Contralateral eyestudy of corneal collagen cross-linking with riboflavin andUVA irradiation in patients with keratoconus. J Refract Surg2009;25(4):371–376.

6. Wittig-Silva C, Whiting M, Lamoureux E, Lindsay RG,Sullivan LJ, Snibson GR. A randomized controlled trial ofcorneal collagen cross-linking in progressive keratoconus:preliminary results. J Refract Surg 2008; 24(7):S720–S725.

7. Caporossi A, Mazzotta C, Baiocchi S, Caporossi T. Long-term results of riboflavin ultraviolet a corneal collagencross-linking for keratoconus in Italy: the Siena eye crossstudy. Am J Ophthalmol 2010; 149(4):585–593.

8. Kymionis GD, Kontadakis GA, Naoumidi TL, Kazakos DC,

by collagen cross-linking with riboflavin-UV-A in patientswith keratoconus. Cornea 2010; 29(2):239–243.

9. Kymionis GD, Kontadakis GA, Kounis GA, et al. Simulta-neous topography-guided PRK followed by corneal collagencross-linking for keratoconus. J Refract Surg 2009;25(9):S807–S811.

0. Vinciguerra P, Albè E, Trazza S, Seiler T, Epstein D.Intraoperative and postoperative effects of corneal collagencross-linking on progressive keratoconus. Arch Ophthalmol2009; 127(10):1258–1265.

1. Fantes FE, Hanna KD, Waring GO 3rd, et al. Wound healingafter excimer laser keratomileusis (photorefractive keratec-tomy) in monkeys. Arch Ophthalmol 1990;108(5):665–675.

2. Wilson SE. Analysis of the keratocyte apoptosis, kerato-cyte proliferation, and myofibroblast transformation re-sponses after photorefractive keratectomy and laser in situkeratomileusis. Trans Am Ophthalmol Soc 2002;100:411–433.

3. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-a-induced collagen crosslinking for the treatment of keratoco-nus. Am J Ophthalmol 2003;135(5):620–627.

4. Muller LT, Candal EM, Epstein RJ, Dennis RF, MajmudarPA. Transepithelial phototherapeutic keratectomy/photore-fractive keratectomy with adjunctive mitomycin-C for com-plicated LASIK flaps. J Cataract Refract Surg 2005;31:291–

296.

OPHTHALMOLOGY NOVEMBER 2011

1

1

15. Holzer MP, Auffarth GU, Specht H, Kruse FE. Combinationof transepithelial phototherapeutic keratectomy and autolo-gous serum eyedrops for treatment of recurrent cornealerosions. J Cataract Refract Surg 2005;31:1603–1606.

16. Kymionis GD, Diakonis VF, Kalyvianaki M, et al. One-yearfollow-up of corneal confocal microscopy after corneal cross-linking in patients with post laser in situ keratomileusisectasia and keratoconus. Am J Ophthalmol 2009;147(5):774–778.

17. Mazzotta C, Balestrazzi A, Traversi C, et al. Treatment of

progressive keratoconus by riboflavin-UVA-induced cross-

LONG-TERM RESULTS OF SIMULTANEOUSVOL. 152, NO. 5

linking of corneal collagen: ultrastructural analysis byHeidelberg Retinal Tomograph II in vivo confocalmicroscopy in humans. Cornea 2007;26(4):390–397.

8. Møller-Pedersen T, Li HF, Petroll WM, Cavanagh HD,Jester JV. Confocal microscopic characterization of woundrepair after photorefractive keratectomy. Invest OphthalmolVis Sci 1998;39(3):487–501.

9. Vinciguerra P, Albè E, Trazza S, et al. Refractive, topo-graphic, tomographic, and aberrometric analysis of kerato-conic eyes undergoing corneal cross-linking. Ophthalmology

2009;116(3):369–378.

PRK AND CXL FOR KERATOCONUS 755