Secondary Logo

Journal Logo


Patient satisfaction with epithelium-off corneal crosslinking

Price, Marianne O. PhD*; Feng, Matthew T. MD; Price, Francis W. Jr. MD

Author Information
Journal of Cataract & Refractive Surgery: March 2018 - Volume 44 - Issue 3 - p 323-328
doi: 10.1016/j.jcrs.2017.12.023
  • Free


Corneal crosslinking (CXL) involves stiffening the cornea through a photochemical reaction. Crosslinking is recommended as a first-line treatment to reduce the risk for disease progression in eyes with keratoconus or corneal ectasia after keratorefractive surgery.1

Since the first report in 2003,2 numerous studies have documented CXL clinical outcomes using the standard Dresden protocol (epithelial removal, soaking with riboflavin 0.1% ophthalmic solution for 30 minutes, and irradiation with ultraviolet-A [UVA] light at 3 mW/cm2 for 30 minutes).3–5 Clinical outcomes have also been documented with accelerated protocols that use a higher light intensity for a shorter duration and with various methods designed to achieve riboflavin penetration and photoactivation without removal of the corneal epithelium.4,5

Despite the wealth of information on clinical outcomes, information about patient perceptions and satisfaction with CXL is lacking.6 The purpose of this study was to address this knowledge gap.


This was a prospective survey study. Review Board approval was obtained from IRBCo, Inc., Buena Park, California, USA. The inclusion criteria were patients treated with CXL at Price Vision Group between February 2008 and March 2016.

Potential participants were invited by email, text message, phone, and regular mail to complete a 1-time electronic survey to assess satisfaction with CXL during March and April of 2017. Those who agreed to participate received an electronic gift card as compensation for their time.

As part of the survey, participants were asked whether they had received a cornea transplant in either eye, the name of their current eyecare provider, and for permission to request medical records from the eyecare provider. Any responses indicating that a participant had keratoplasty in either eye were verified with a review of the medical records at Price Vision Group and by obtaining and reviewing medical records from the participant's current eyecare provider.

The survey study participants were treated with CXL in a series of 5 investigational studies listed on the website (NCT00647699,A NCT00674661,B NCT01143389,C NCT01344187,D and NCT01643226E) and initiated before the U.S. Food and Drug Administration approved CXL in 2016. Three trialsA–C used the standard epithelium-off Dresden protocol parameters (epithelial removal, soaking with riboflavin 0.1% ophthalmic solution for 30 minutes, and UVA irradiation at 3 mW/cm2 for 30 minutes to achieve a total radiant exposure of 5.4 J/cm2).7–9 Two trialsD,E used an accelerated treatment (epithelial removal, soaking with riboflavin 0.12% ophthalmic solution for 20 minutes, and UVA irradiation at 30 mW/cm2 for 4 minutes or pulsed for 8 minutes to achieve a total radiant exposure of 7.2 J/cm2).

Participants were classified as having post-refractive surgical ectasia or keratoconus. Those with keratoconus were further classified as mild, moderate, or severe as described by the Collaborative Longitudinal Evaluation of Keratoconus Study Group.10 The classification was based on the flat keratometry and astigmatism measured by corneal tomography (Pentacam, Oculus Optikgeräte GmbH) before CXL.10 Additional participant characteristics were tabulated, including sex, age at the time of treatment, baseline manifest refraction spherical equivalent (MRSE), baseline maximum keratometry (K) value by corneal tomography, and the type of CXL procedure performed (Dresden protocol or accelerated). The keratoconus classification and baseline data were taken from the first treated eye in bilateral cases.

Statistical Analysis

The potential association between demographic factors and responses regarding vision were assessed with general estimating equations. Factors associated with perceived efficacy were assessed with univariate and multivariable logistic regression. All analyses were 2-sided and performed with Statistical Analysis Software (version 9.4, SAS Institute, Inc.); P values less than 0.05 were considered statistically significant.



Of the 552 participants who met the study inclusion criteria, 448 (80% response rate) completed the survey. Overall, 349 participants (78%) had keratoconus whereas 99 (22%) had post-refractive surgical ectasia, 196 (44%) had bilateral CXL, and 252 (56%) had unilateral CXL. Table 1 shows the participant demographics, including sex, age at the time of CXL, and the MRSE and maximum K values before CXL. The median time from CXL to survey completion was 3.5 years (range 1 to 9 years).

Table 1
Table 1:
Demographics of the CXL survey participants by corneal ectasia classification.


Table 2 shows the perceived visual changes and vision correction methods after CXL. When asked, “how has crosslinking affected your vision,” 183 (41%) said vision had improved, 204 (46%) reported no change, and 61 (14%) said vision had continued to worsen. The proportion who reported visual improvement was highest among those treated for mild keratoconus and lower among those treated for moderate to severe keratoconus or post-refractive surgical ectasia (P = .0003) (Figure 1). Likewise, the proportion who reported continued visual decline was lowest among those with mild keratoconus and higher among those treated for moderate to severe keratoconus or postsurgical ectasia (Figure 1).

Table 2
Table 2:
Perceived visual changes and vision correction methods after CXL.
Figure 1.
Figure 1.:
Perceived effect of CXL on vision as a function of baseline ectasia classification (CXL = corneal crosslinking).

The perceived effect of CXL on vision was not significantly associated with the vision correction method currently being used (P = .89). Sixty-seven participants (15%) reported no use of vision correction in the CXL-treated eye(s), 136 (30%) used only spectacles, 157 (35%) used contact lenses, and 88 (20%) used both spectacles and contact lenses. The proportion using contact lenses was highest among those with previous moderate to severe keratoconus (67 [67%]) and lower among those with previous mild keratoconus (127 [51%]) or postsurgical ectasia (51 [51%]), (P = .019) (Table 2). Ninety-one study participants (20%) used scleral rigid gas-permeable (RGP) contact lenses, 85 (19%) used corneal RGP contact lenses, 58 (13%) used soft spherical or toric lenses, and 23 (5%) were unsure of the lens type used. A higher proportion of those with previous moderate to severe keratoconus relied on RGP contact lenses (54 [54%]) for vision correction as compared with the proportions in the mild keratoconus group (82 [33%]) or postsurgical ectasia group (40 [40%]); however, the difference between groups did not reach statistical significance (P = .10) (Table 2).

Perceived Efficacy

Table 3 shows the associations between demographic factors and the perceived efficacy after CXL. Most of the participants reported that CXL stopped the progression of the keratoconus or ectasia, and the perceived efficacy was not associated with the length of time from CXL to survey completion (P = .90).

Table 3
Table 3:
Associations between demographic factors and perceived CXL efficacy.

Perceived efficacy was significantly associated with 3 factors in univariate analysis as follows: age at the time of CXL, ectasia classification, and sex. The majority of participants who were under 20 years old at the time of treatment thought CXL stopped the disease progression (P = .0003) (Table 3 and Figure 2). Corneal crosslinking was considered effective by a significantly higher proportion of participants classified as having mild keratoconus than those treated for moderate to severe keratoconus or post-refractive surgical ectasia (P = .0004). Men were more likely than women to say that CXL was efficacious (P = .004). The participant age, ectasia classification, and sex were interrelated. In multivariable analysis, both age and ectasia classification were significantly associated with perceived CXL efficacy (P = .028 and P = .029, respectively).

Figure 2.
Figure 2.:
Proportion who considered CXL to be effective in halting disease progression as a function of age at the time of treatment (CXL = corneal crosslinking).

When asked if they still rubbed their eyes, 242 respondents (55%) answered affirmatively, even though all had been instructed not to do so. Self-reported eye rubbing was not significantly associated with perceived CXL efficacy (P = .13) (Table 3).

Opinions about efficacy were comparable between those who had conventional CXL and those who had an accelerated procedure (P = .78). Also, perceived efficacy was not significantly associated with the baseline MRSE (P = .39) or the baseline maximum K value measured before CXL (P = .14) (Table 3).


When asked whether they would recommend CXL to a friend with a similar eye condition, the mean response was 8.6 ± 1.9 (SD) on a scale ranging from 1 (would not recommend) to 10 (definitely would recommend). Those treated for keratoconus were more likely to recommend CXL (mean score 8.8 ± 1.7, range 1 to 10) than those treated for postsurgical ectasia (mean score 7.6 ± 2.5, range 1 to 10) (P < .0001).

A score of 9 or 10 was considered indicative of strong satisfaction with CXL. The strongly satisfied proportion was 169 participants (68%) among those with mild keratoconus, 61 participants (61%) among those with moderate to severe keratoconus, and 42 participants (42%) among those with post-refractive surgical ectasia.

Subsequent Keratoplasty

Including fellow eye treatments, CXL was performed in 644 eyes of 448 study participants. Eleven (1.7%) of 644 treated eyes had subsequent keratoplasty at the time of survey completion with a median follow-up of 3.5 years. In these cases, the participants were contact-lens intolerant and unable to obtain satisfactory vision with spectacles. Twenty-three study participants had a cornea transplant in the non-CXL fellow eye.


This study found that 89% of treated participants thought CXL halted disease progression at a median of 3.5 years after treatment. Treatment at a younger age and at a mild stage of keratoconus was associated with higher satisfaction and perceived efficacy. This was consistent with studies showing that although CXL usually halts disease progression, it does not reliably reverse previous corneal changes.1 The high perceived efficacy among younger participants was a particularly important finding, because younger patients are significantly more likely to experience keratoconus progression if not treated,11 whereas the likelihood of keratoconus progression declines with aging as the extent of CXL in the cornea naturally increases.12

The keratoconus participants were more likely to perceive CXL as being effective and had a higher median satisfaction score than those treated for corneal ectasia after keratorefractive surgery. The latter participants tended to be older (mean age 44 years) and thus were more likely to also experience the challenges of presbyopia. In addition, those who previously paid to have corneal refractive surgery might have had higher expectations of excellent uncorrected vision.

The proportions who would strongly recommend CXL ranged from 42% to 68% depending on the baseline ectasia classification. These are substantially lower than the proportions who would strongly recommend laser in situ keratomileusis (LASIK) (77% to 88%, depending on the baseline method of vision correction), but of course LASIK provides immediate visual improvement whereas CXL alone does not.13

Over one half of the participants (56%) had unilateral CXL. Keratoconus is a bilateral condition but often asymmetric.14 In 23 (11%) of 201 keratoconus participants who had unilateral treatment, the condition had progressed in the untreated fellow eye to the point of requiring keratoplasty. In other participants, the fellow eye might not have qualified for any treatment if the keratoconus were subclinical or not progressing, or if the vision were still correctable to 20/20 on the Snellen chart. Some participants who qualified for bilateral CXL might have chosen not to have the fellow eye treated because of the expense and/or the pain experienced with epithelial removal.

Eleven (1.7%) of 644 CXL-treated eyes had a subsequent keratoplasty by the time of survey completion. It has long been estimated that 10% to 20% of all keratoconus patients will need keratoplasty; however, recent studies suggest that use of CXL has reduced that need.12 Godefrooij et al.15 found that 25% fewer corneal transplants were performed in the 3-year period after the nationwide introduction of CXL in the Netherlands (2012 to 2014) compared with a 3-year period before its introduction (2005 to 2007). Likewise, Sandvik et al.16 found the frequency of keratoplasty at their institution in Norway was more than halved from 2013 to 2014 relative to 2005 to 2006; they attributed this decline primarily to the introduction of CXL. Corneal crosslinking was not approved in the United States until 2016, although Sarezky et al.17 found that American patients with keratoconus were significantly less likely to have penetrating keratoplasty between 2009 and 2012 compared with 2001 to 2008 (odds ratio, 0.57). The decline was attributed to improvements in contact lens technology.18 Both CXL and improved contact lenses might help patients with corneal ectasia avoid the need for keratoplasty.

Among this study's limitations were that self-reported outcomes are intrinsically subjective. Also, the responses might have been influenced by the Hawthorne effect19 (tendency to feel more positive about a subject because of extra attention) and/or preventing cognitive dissonance (discomfort from inconsistent/negative feelings about one's choices, such as having surgery).20 Finally, the study was not designed to validate the subjective responses with objective measures of vision. The study's strengths included the high response rate and the large sample size.

In conclusion, 9 out of 10 participants treated with CXL thought it halted progression of corneal ectasia. Satisfaction and perceived efficacy were higher among those treated at a younger age or earlier stage of disease progression. The time from CXL to survey completion was not significantly associated with perceived efficacy, suggesting no perceptible fading of the effect over the follow-up of 1 to 9 years.


  • Clinical measurements show that epithelium-off CXL can usually halt progression of keratoconus and corneal ectasia after refractive surgery.


  • Patients treated at a younger age and at a milder stage of keratoconus progression were significantly more likely to perceive visual improvement and to consider CXL to be effective at halting disease progression.
  • Compared with patients treated for keratoconus, those treated for post-refractive surgical ectasia were significantly less likely to consider the treatment to be effective and less likely to perceive visual improvement.


1.Gomes JAP, Tan D, Rapuano CJ, Belin MW, Ambrósio R Jr, Guell JL, Malecaze F, Nishida K, Sangwan VS., the Group of Panelists for the Global Delphi Panel of Keratoconus and Ectatic Diseases. (2015). Global consensus on keratoconus and ectatic diseases. Cornea, 34, 359-369, Available at:
2.Wollensak G, Spoerl E, Seiler T. (2003). Riboflavin/ultraviolet-A–induced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol, 135, 620-627, Available at:
3.Wittig-Silva C, Chan E, Islam FMA, Wu T, Whiting M, Snibson GR. (2014). A randomized, controlled trial of corneal collagen cross-linking in progressive keratoconus; three-year results. Ophthalmology, 121, 812-821, Available at:
4.O'Brart DPS. Corneal collagen crosslinking for corneal ectasias: a review. Eur J Ophthalmol. 2017;11:253-269.
5.Meiri Z, Keren S, Rosenblatt A, Sarig T, Shenhav L, Varssano D. Efficacy of corneal cross-linking for the treatment of keratoconus: a systematic review and meta-analysis. Cornea. 2016;35:417-428.
6.Pron G, Ieraci L, Kaulback K. Medical Advisory Secretariat, Health Quality Ontario. Collagen cross-linking using riboflavin and ultraviolet-A for corneal thinning disorders: An evidence-based analysis. Ont Health Technol Assess Ser. 11. 5. 2011. 1-89. Available at:
7.Hersh PS, Stulting RD, Muller D, Durrie DS, Rajpal RK., for the U.S. Crosslinking Study Group. (2017). U.S. multicenter clinical trial of corneal collagen crosslinking for treatment of corneal ectasia after refractive surgery. Ophthalmology, 124, 1475-1484, Available at:
8.Hersh PS, Stulting RD, Muller D, Durrie DS, Rajpal RK., on behalf of the United States Crosslinking Study Group. (2017). United States multicenter clinical trial of corneal collagen crosslinking for keratoconus treatment. Ophthalmology, 124, 1259-1270, Available at:
9.Price MO, Fairchild K, Feng MT, Price FW Jr. Prospective randomized trial of corneal cross-linking riboflavin dosing frequencies for treatment of keratoconus and corneal ectasia. Ophthalmology 2017 Dec 1. [Epub ahead of print].
10.McMahon TT, Szczotka-Flynn L, Barr JT, Anderson RJ, Slaughter ME, Lass JH, Iyengar SK., and the CLEK Study Group. A new method for grading the severity of keratoconus: the Keratoconus Severity Score (KSS). Cornea. 2006;25:794-800.
11.McMahon TT, Edrington TB, Szczotka-Flynn L, Olafsson HE, Davis LJ, Schechtman KB., and the CLEK Study Group. (2006). Longitudinal changes in corneal curvature in keratoconus. Cornea, 25, 296-305, Available at:
12.Jeng BH, Farid M, Patel SV, Schwab I. (2016). Corneal cross-linking for keratoconus: a look at the data, the Food and Drug Administration, and the future [editorial]. Ophthalmology, 123, 2270-2272, Available at:
13.Price MO, Price DA, Bucci FA Jr, Durrie DS, Bond WI, Price FW Jr. Three-year longitudinal survey comparing patient satisfaction with LASIK and contact lenses. Ophthalmology. 2016;123:1659-1666.
14.Rabinowitz YS. (1998). Keratoconus. Surv Ophthalmol, 42, 297-319, Available at:
15.Godefrooij DA, Gans R, Imhof SM, Wisse RPL. (2016). Nationwide reduction in the number of corneal transplantations for keratoconus following the implementation of cross-linking. Acta Ophthalmol, 94, 675-678, Available at:
16.Sandvik GF, Thorsrud A, Råen M, ϕstern AE, Sæthre M, Drolsum L. Does corneal collagen cross-linking reduce the need for keratoplasties in patients with keratoconus? Cornea. 2015;34:991-995.
17.Sarezky D, Orlin SE, Pan W, VanderBeek BL. Trends in corneal transplantation in keratoconus. Cornea. 2017;36:131-137.
18.Downie LE, Lindsay RG. (2015). Contact lens management of keratoconus. Clin Exp Optom, 98, 299-311, Available at:
19.McCarney R, Warner J, Iliffe S, van Haselen R, Griffin M, Fisher P. The Hawthorne Effect: a randomized, controlled trial. BMC Med Res Methodol. 7. 2007. 30. Available at:
20.Festinger L. Cognitive dissonance. Sci Am. 1962;207(4):93-106.


A.U.S. National Institutes of Health Clinical Trials. Corneal Collagen Cross-linking for Progressive Keratoconus (CXL) NCT00647699. Available at: Accessed December 14, 2017
B.U.S. National Institutes of Health Clinical Trials. Corneal Collagen Cross-linking for Ectasia (CXL) NCT00674661. Available at: Accessed December 14, 2017
C.U.S. National Institutes of Health Clinical Trials. Corneal Crosslinking in Patients With Keratoconus and Post-Refractive Ectasia. NCT01143389. Available at: Accessed December 14, 2017
D.U.S. National Institutes of Health Clinical Trials. Safety and Efficacy Study of Corneal Collagen Cross-Linking in Eyes With Keratoconus. NCT01344187. Available at: Accessed December 14, 2017
E.U.S. National Institutes of Health Clinical Trials. Safety and Efficacy Study of Corneal Collagen Cross-Linking in Eyes With Keratoconus. NCT01643226. Available at: Accessed December 14, 2017

Disclosures:Dr. F.W. Price has received lecture fees from Avedro, Inc. None of the other authors has a financial or proprietary interest in any material or method mentioned.

© 2018 by Lippincott Williams & Wilkins, Inc.