ARTICLE

Safety of high-inten

Q 2014 ASC

Published by

sity corneal collagencrosslinking

George D. Kymionis, MD, PhD, Michael A. Grentzelos, MD, Vardhaman P. Kankariya, MD,Dimitrios A. Liakopoulos, MD, MSc, Dimitra M. Portaliou, MD, Konstantinos I. Tsoulnaras, MD,

Alexandra E. Karavitaki, MD, Aristophanis I. Pallikaris, MSc

RS an

Elsev

PURPOSE: To evaluate the safety of a new high-intensity corneal collagen crosslinking (CXL)treatment protocol for keratoconus.

SETTING: Vardinoyiannion Eye Institute of Crete, Faculty of Medicine, University of Crete, Heraklion,Crete, Greece.

DESIGN: Prospective interventional case series.

METHODS: Patients with progressive keratoconus had CXL using a new treatment protocol with9 mW/cm2 irradiance for duration of 10 minutes. The rate of reepithelialization, endothelial celldensity (ECD), corrected distance visual acuity (CDVA), and steep and flat keratometry (K) valueswere evaluated preoperatively and 3 months postoperatively.

RESULTS: Nine patients (10 eyes) were enrolled. No intraoperative or early postoperative compli-cations were observed in any patient. The ECD did not change significantly 3 months postopera-tively (PZ.169). The CDVA improved from 0.19 logMAR G 0.20 (SD) preoperatively to 0.10 G0.16 logMAR 3 months postoperatively; however, the improvement did not attain significance(PZ.141). No eye lost lines of CDVA. The mean steep K readings decreased significantly from48.04 G 2.57 diopters (D) preoperatively to 46.51 G 2.81 D 3 months postoperatively(PZ.047); the mean flat K readings did not change significantly postoperatively (PZ.285).

CONCLUSIONS: Corneal collagen crosslinking at 9 mW/cm2 irradiance for 10 minutes did not causesignificant changes in ECD or intraoperative or early postoperative complications. None patient losta line of CDVA 3 months after the procedure.

Financial Disclosure: No author has a financial or proprietary interest in any material or methodmentioned.

J Cataract Refract Surg 2014; 40:1337–1340 Q 2014 ASCRS and ESCRS

The introduction of corneal collagen crosslinking(CXL) in routine clinical practice has changed theman-agement of keratoconus. Corneal collagen crosslinkingis a technique that uses riboflavin (photosensitizer,vitamin B2) and ultraviolet-A (UVA) irradiation. Thephotochemical reaction between the 2 in the cornealstroma leads to the development of chemical bondsbetween collagen fibrils. Thereby, CXL strengthensthe cornea and slows or even stops the progressionof keratoconus and corneal ectasia.1–6

The efficacy and safety of the standard CXL proce-dure for stabilizing keratoconus has been reported inseveral studies.4–6 In the Dresden protocol, after epithe-lial debridement and riboflavin stromal saturation for30minutes, UVA light (365 nm) is used for illumination

d ESCRS

ier Inc.

at the intensity of 3 mW/cm2 for 30 minutes.1 Thisdelivers 5.4 J/cm2 total energy to the cornea. Toshorten the treatment span of the CXL procedure forimproving patient cooperation and comfort, an increasein illumination intensity has been used. By increasingintensity, it is hypothesized that a fundamental photo-chemical law, the Bunsen-Roscoe law of reciprocity,can be applied in CXL. According to this law, theeffects of the photochemical reaction are similar ifthe intensity and time is changed while the totalenergy is maintained. Thus, the total energy deliveredand amount of crosslinks induced in standard CXL(3mW/cm2 for 30minutes) should be similar to irradia-tion at 9 mW/cm2 for 10 minutes and 18 mW/cm2 for5 minutes, all delivering 5.4 J/cm2.

0886-3350/$ - see front matter 1337http://dx.doi.org/10.1016/j.jcrs.2013.11.041

1338 SAFETY OF HIGH-INTENSITY CXL

The purpose of this study was to evaluate safety of anew high-intensity CXL (9 mW/cm2 for 10 minutes)treatment protocol for keratoconus.

PATIENTS AND METHODS

This prospective interventional case series comprised eyeswith progressive keratoconus. Patients were thoroughlyinformed about the nature of the procedure, the possiblerisks, and the current clinical experience. Institutional reviewboard approval was obtained, and all patients gave writteninformed consent in accordance with institutional guidelinesof the Declaration of Helsinki.

Inclusion criteria comprised progressive keratoconus(progression of the ectatic disorder indicating necessity oftreatment based on an increase in cone apex keratometry[K] of 0.75 diopter (D) or on alteration of 0.75 D in sphericalequivalent refraction during the past 6 months), cornealthickness more than 400 mm, no central or paracentralsubepithelial or stromal corneal scarring, no pregnancy orlactation, absence of previous anterior segment ocularsurgery, absence of autoimmune or collagen disease, activeocular infection, and severe dry eye.

Preoperative and 3-month postoperative data obtainedfrom the patient records included ocular and general healthhistory assessment, age, sex, corrected distance visual acuity(CDVA), corneal topography (iTrace, Tracey Technologies),slitlamp examination, rate of reepithelialization, and endo-thelial cell density (ECD) using corneal confocal microscopywith a modified confocal scanning laser ophthalmoscope(HRT II, Heidelberg Engineering GmbH).

Surgical Technique

The same surgeon (G.D.K.) performed all proceduresunder sterile conditions. After topical anesthesia withproxymetacaine hydrochloride 0.5% eyedrops (Alcaine) wasapplied, the corneal epithelium was removed using transepi-thelial phototherapeutic keratectomy (Cretan protocol).4 Afterepithelial removal, riboflavin (0.1% solution of 10 mgriboflavin-5-phosphate in 10 mL dextran-T-500 20% solution,Medicross, Medio Haus Medizinprodukte GmbH) wasinstilled on the center of the cornea every 3 minutes for ap-proximately 30 minutes. Ultraviolet-A irradiation was per-formed using a UVA optical system (CCL-365, Peschke

Submitted: October 1, 2013.Final revision submitted: November 13, 2013.Accepted: November 13, 2013.

From Vardinoyiannion Eye Institute of Crete (Kymionis, Grentzelos,Kankariya, Liakopoulos, Portaliou, Tsoulnaras, Karavitaki, Pallika-ris), Faculty of Medicine, University of Crete, Heraklion, Crete,Greece; the Department of Ophthalmology (Kymionis, Kankariya,Portaliou), Bascom Palmer Eye Institute, University of Miami MillerSchool of Medicine, Miami, Florida, USA.

Supported in part by the special research account of the Universityof Crete.

Corresponding author: George D. Kymionis, MD, PhD, University ofCrete, Faculty of Medicine Vardinoyiannion Eye Institute of Crete,71003 Heraklion, Crete, Greece. E-mail: kymionis@med.uoc.gr.

J CATARACT REFRACT SURG -

Meditrade GmbH). Before treatment, an intended irradianceof 9.0 mW/cm2 was calibrated using a UVA light meter(YK-35UV, Lutron Electronics Co., Inc.), which is suppliedwith the UVA optical system. Irradiance was performed for10 minutes in an intended irradiance of 9.0 mW/cm2,corresponding to a total surface dose of 5.4 J/cm2. DuringUVA irradiation, riboflavin solution was applied every 1 to2 minutes to maintain corneal saturation with riboflavin. Atthe end of the procedure, a silicone–hydrogel bandage contactlens (lotrafilcon B, Air Optix, Ciba Vision; 14.0 mm diameter,8.6 base curvature, oxygen permeability 140 barrers) wasapplied until full reepithelialization.

Postoperative medication included ofloxacin (Exocin) 4times daily and chloramphenicol–dexamethasone drops(Dispersadron) 4 times daily until the epithelium healedcompletely. After full reepithelialization, the bandagecontact lens was removed and patients received fluoro-metholone 0.1% drops (FML), which were tapered over thenext 2 weeks. Patients were encouraged to use artificial tearsat least 6 times a day for 3 months postoperatively.

Confocal Microscopy

Confocal microscopy was performed using the modifiedconfocal scanning laser ophthalmoscope in all patients aspreviously described.4

Statistical Analysis

All data were analyzed for normality. Because normalitywas not valid for all data, the Wilcoxon signed-rank test(SPSS software, version 20, SPSS, Inc.) was used. All valuesare expressed as mean values G standard deviation. A Pvalue less than 0.05 was considered significant. Visual acuityis expressed in logMAR notation.

RESULTS

The study evaluated 10 eyes of 9 patients. The meanage of the 5 men and 4 women was 24.90G 5.24 years(range 16 to 33 years). The follow-up was 3 months inall cases. No intraoperative or early postoperativecomplications were observed.

The mean time of reepithelialization was 4.2G 0.79days (range 3 to 5 days). Table 1 shows the parametersevaluated preoperatively and at the 3-month postop-erative visit. There was no statistically significantchange in the ECD between preoperatively and 3months postoperatively (PZ.169). The logMARCDVA improved, although the improvement wasnot statistically significant (PZ.141). No eyes lost linesof CDVA (Figure 1). The decrease in the mean steep Kreadings from preoperatively to 3 months postopera-tively was statistically significant (PZ.047), althoughthe decrease in the mean flat K readings was not statis-tically significant (PZ.285).

DISCUSSION

Several studies have established the safety and efficacyof standard CXL for the stabilization of corneal ectaticdisorders.4–6 Nevertheless, the Dresden protocol

VOL 40, AUGUST 2014

Table 1. Preoperative and postoperative patient data.

Parameter

Mean G SD

P ValuePreoperative Postoperative

LogMAR CDVA 0.19 G 0.20 0.10 G 0.16 .141Mean steep K (D) 48.04 G 2.57 46.51 G 2.81 .047Mean flat K (D) 43.88 G 1.67 43.55 G 1.88 .285ECD (cells/mm2) 2688 G 130 2640 G 127 .169

CDVA Z corrected distance visual acuity; ECD Z endothelial celldensity; K Z keratometry Figure 1. Change in CDVA (safety) 3 months postoperatively

(CDVA Z corrected distance visual acuity).

1339SAFETY OF HIGH-INTENSITY CXL

requires up to 1 hour of surgical time (up to 30minutesfor riboflavin saturation and 30 minutes for UVAirradiation at 3 mW/cm2), which may be inconvenientto the patient as well as the surgeon.1 In an attempt toshorten the duration of treatment, 2 approaches havebeen proposed. One is riboflavin application byiontophoresis technique for a rapid stromal saturationand the other, use of high-intensity UVA irradiation.7,8

In accordance with the photochemical law of reci-procity (Bunsen-Roscoe law), it is possible to achievethe same photochemical effect (induced crosslinking)with a reduced irradiation time and correspondinglyincreased irradiation intensity so the total energydose remains the same. Recently,Wernli et al.8 showedthat the Bunsen-Roscoe reciprocity law is only validfor illumination intensities up to 40 to 50 mW/cm2

and illumination times more than 2 minutes. Howev-er, UVA irradiation may cause damage in keratocytes,corneal endothelial cells, the crystalline lens, and theretina.9 Riboflavin works as a photosensitizer andfunctions as a protective barrier against the harmfulinfluence of UVA irradiation on the underlying tis-sues. The cell-damage threshold of UVA irradiationcombined with riboflavin is 10 times lower thanUVA irradiation alone in standard CXL.10 Althoughthe total energy applied in high-intensity CXL issimilar to that with the standard protocol, the intensityof irradiation is higher andmay have acute side effects(eg, on the corneal endothelium).

Two recent studies11,12 report transient cornealendothelial changes after accelerated CXL for thetreatment of progressive keratoconus. In both studies,there was a significant decrease in ECD on cornealspecular microscopy at the 1-month and 3-monthpostoperative intervals; the ECD returned to pre-operative values by 6 months postoperatively.11,12 Incontrast, using confocal microscopy, Touboul et al.13

found no significant changes in ECD after acceleratedCXL.

In our case series of high-intensity CXL (9 mW/cm2

for 10 minutes), there was no significant change in

J CATARACT REFRACT SURG -

ECD; Touboul et al.13 report similar results but for adifferent irradiation time and irradiation intensity(30 mW/cm2 for 3 minutes). The results in the 2 recentstudies that showed transient decrease in ECD at the1-month and 3-month postoperative intervals11,12

may be attributed to the use of specular microscopybecause in the presence of typical post-CXL cornealhaze, increased scattering and limited visualizationof the corneal endothelium may result. In contrast,confocal microscopy (used in our and the Touboulet al. study13) allows the examination of cornealendothelium in the presence of edema and haze; thecomposite nature of the images obtained ensuresthat confocal biomicroscopy is not affected by thescattering usually present in the third postoperativemonth after CXL.13,14 In addition, in our study therewere no intraoperative or early postoperative compli-cations and no eye had delayed reepithelialization.The CDVA remained stable postoperatively, and nopatient lost a line of CDVA. Themean steep K readingsdecreased significantly postoperatively, while meanflat K readings did not change significantly.

Limitations of our case series are the small numberof patients and the short follow-up; thus, no sufficientconclusions can be made. Another limitation is thelack of a control group of standard CXL, which couldfacilitate the comparison between standard CXL(3 mW/cm2 for 30 minutes) and high-intensity CXL(9 mW/cm2 for 10 minutes).

In conclusion, in our preliminary study of a high-intensity CXL treatment protocol, CXL at 9 mW/cm2

irradiance for 10 minutes did not affect the ECD whilethe CDVA remained stable postoperatively. Eventhough preliminary results are not sufficient for anevaluation of the safety profile, CXL at 9 mW/cm2

irradiance for 10 minutes did not result in significantchanges in ECD and or in intraoperative or earlypostoperative complications. In addition, no patientlost a line of CDVA 3 months after the procedure.Larger case series with a longer follow-up are requiredto confirm these encouraging results.

VOL 40, AUGUST 2014

1340 SAFETY OF HIGH-INTENSITY CXL

WHAT WAS KNOWN

� Standard CXL (30 minutes UVA irradiation at intendedirradiance of 3.0 mW/cm2 with total surface dose of5.4 J/cm2) is a safe and efficient surgical procedure forstabilizing keratoconus. According to the Bunsen-Roscoe law of reciprocity, a 10-minute illumination at9.0 mW/cm2 should provide the same effect with a30-minute illumination at 3.0 mW/cm2.

WHAT THIS PAPER ADDS

� Corneal collagen crosslinking at 9 mW/cm2 irradiance for10 minutes did not result in significant changes in ECDand no intraoperative or early postoperative complicationsoccurred. No patient lost a line of CDVA 3 months after theprocedure.

REFERENCES1. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-A–

induced collagen crosslinking for the treatment of keratoconus.

Am J Ophthalmol 2003; 135:620–627. Available at: http://

grmc.ca/assets/files/collagen_crosslinking_2003_wollensak.pdf.

Accessed March 22, 2014

2. Hafezi F, Kanellopoulos J,Wiltfang R, Seiler T. Corneal collagen

crosslinking with riboflavin and ultraviolet A to treat induced ker-

atectasia after laser in situ keratomileusis. J Cataract Refract

Surg 2007; 33:2035–2040

3. Kymionis GD, Karavitaki AE, Kounis GA, Portaliou DM, Yoo SH,

Pallikaris IG.Management of pellucidmarginal corneal degener-

ation with simultaneous customized photorefractive keratec-

tomy and collagen crosslinking. J Cataract Refract Surg 2009;

35:1298–1301

4. Kymionis GD, Grentzelos MA, Kounis GA, Diakonis VF,

Limnopoulou AN, Panagopoulou SI. Combined transepithelial

phototherapeutic keratectomy and corneal collagen cross-

linking for progressive keratoconus. Ophthalmology 2012;

119:1777–1784

5. Raiskup-Wolf F, Hoyer A, Spoerl E, Pillunat LE. Collagen

crosslinking with riboflavin and ultraviolet-A light in keratoco-

nus: long-term results. J Cataract Refract Surg 2008;

34:796–801

J CATARACT REFRACT SURG -

6. Caporossi A, Mazzotta C, Baiocchi S, Caporossi T. Long-term

results of riboflavin ultraviolet A corneal collagen cross-linking

for keratoconus in Italy: the Siena Eye Cross Study. Am J

Ophthalmol 2010; 149:585–593

7. Vinciguerra P, Rechichi M, Rosetta P, Romano MR,

Mastropasqua L, Scorcia V, Azzolini C, Vinciguerra R. High flu-

ence iontophoretic corneal collagen cross-linking: in vivo OCT

imaging of riboflavin penetration [letter]. J Refract Surg 2013;

29:376–377; reply by H Nagaraj, R Shetty, RS Kumar, H

Veluri, C Malhotra, KB Shetty, 377

8. Wernli J, Schumacher S, Spoerl E, Mrochen M. The efficacy of

corneal cross-linking shows a sudden decrease with very high

intensity UV light and short treatment time. Inv Ophthalmol

Vis Sci 2013; 54:1176–1180. Available at: http://www.iovs.org/

content/54/2/1176.full.pdf. Accessed March 22, 2014

9. Spoerl E, Mrochen M, Sliney D, Trokel S, Seiler T. Safety of

UVA–riboflavin cross-linking of the cornea. Cornea 2007;

26:385–389

10. Wollensak G, Sp€orl E, Reber F, Pillunat L, Funk R. Corneal

endothelial cytotoxicity of riboflavin/UVA treatment in vitro.

Ophthalmic Res 2003; 35:324–328

11. Cing€u AK, Sogutlu-Sari E, C‚ ınar Y, S‚ ahin M, T€urkc‚ €u FM,

Y€uksel H, S‚ ahin A, C‚ ac‚ a _I. Transient corneal endothelial

changes following accelerated collagen cross-linking for the

treatment of progressive keratoconus. Cutan Ocul Toxicol

2013 Jul 17 [Epub ahead of print]

12. C‚ ınar Y, Cing€u AK, T€urkc€u FM, C‚ ınar T, Y€uksel H, €Ozkurt ZG,

C‚ ac‚ a I. Accelerated corneal collagen cross-linking for progres-

sive keratoconus. Cutan Ocul Toxicol 2013 Oct 22 [Epub ahead

of print]

13. Touboul D, EfronN, SmadjaD, PraudD,Malet F,Colin J. Corneal

confocal microscopy following conventional, transepithelial, and

accelerated corneal collagen cross-linking procedures for kerato-

conus. J Refract Surg 2012; 28:769–776

14. Hara M, Morishige N, Chikama T, Nishida T. Comparison of

confocal biomicroscopy and noncontact specular microscopy

for evaluation of the corneal endothelium. Cornea 2003;

22:512–515

VOL

40, AUGUST 2014

First author:George D. Kymionis, MD, PhD

Vardinoyiannion Eye Institute of Crete,Faculty of Medicine, Universityof Crete, Heraklion, Crete, Greece