Clinical considerations in determining which modality is best suited for each patient following CXL.

Corneal cross-linking (CXL) was authorized in Canada in 2008. Since that time, a significant number of patients in Canada have undergone the procedure. For this reason, optometrists in Canada have a great deal of experience both with this technology and with co-managing patients with ophthalmologists. This article summarizes our extended clinical exposure in fitting and refitting contact lenses post-CXL.


CXL is an invasive treatment with the goal to make the cornea stiffer and to stabilize it, mostly in the presence of ectasia. Contrary to the general belief, cross-links are formed between amino terminals of the collagen side chains and the proteoglycans of the extracellular matrix and not between and within collagen fibers.1 For this reason, the term collagen cross-linking should no longer be used.

The clinical application of this procedure is to prevent the progression of corneal ectasia, whether genetically acquired or as a side-effect of refractive surgery.2 This technique is also considered beneficial in controlling recalcitrant bacterial infections or in the presence of fungi or amoebae.2

The Dresden protocol (standard) implies instillation of 0.1% isotonic riboflavin solution (vitamin B2) in 20% dextran drops to load the corneal tissue before exposing it to a solid-state ultraviolet-A (UVA) light source with a wavelength of 370nm at an irradiance of 3 mW/cm2 for 30 minutes.2 To facilitate the process, expedite the procedure, and make it more convenient for patients, CXL may be performed using a higher irradiance (9 mW/cm2) for a shorter time (10 min)3; this is known as accelerated CXL.

The standard procedure states that corneal epithelial cells must be debrided within an area of 9mm centrally to favor riboflavin impregnation of the stroma (epi-off technique). But, to reduce the duration of the procedure, to alleviate corneal haze, and to increase patient comfort, a trans-epithelial technique was developed (epi-on). Two sub-methods were also considered. In the first, a hypertonic solution with chemical enhancers (benzalkonium chloride [BAK], trometamol, and/or ethylenediaminetetraacetic acid [EDTA]) is instilled to break the tight junctions between epithelial cells, thereby enhancing riboflavin absorption. In the second option, called iontophoresis, riboflavin is instilled through an electrically charged cup. The solution is then absorbed more easily because of ionic attraction.4

To protect endothelial cells, corneal thickness should be 400µm or greater. In such cases, riboflavin reaches two-thirds of the corneal thickness as visible through optical coherence tomography (OCT) assessment. If the cornea is more than 350µm but less than 400µm, an ultraviolet-barrier-free soft contact lens (0.09mm thickness, 14mm diameter) soaked in riboflavin is placed on the cornea after corneal impregnation. This increases the corneal thickness artificially and protects its deeper layers.5 Otherwise, the CXL technique cannot be performed safely.

Adverse effects associated with CXL include mild diffuse corneal edema and paracentral infiltrates, peripheral corneal vascularization, permanent corneal haze, subepithelial infiltrates, and anterior chamber inflammation.6 This occurs to between 1% (< 35 years old) and 3% (> 35 years old) of patients. Over-correction can also be induced if a topography-guided laser surgery is performed at the same time.7


In general, CXL generates immediate biomechanical and histological effects. Stiffening of the tissue occurs progressively as the cross-links develop. This is true with the standard epi-off procedure and with iontophoresis.8 This means that the efficacy of the treatment is improving over time. Tissue modification leads to stromal compaction and thinning, especially in the anterior stroma. Finally, because of the UV radiation, there is a high level of keratocyte apoptosis that occurs deeper than the stromal compaction.9 Their regeneration occurs after several months, and this process remains unaffected with contact lens use.10 Similarly, corneal fibrosis markers, such as Collagen III and α-smooth muscle actin, are significantly downregulated in treated corneas. To the contrary, fibroblasts seem to be not affected, with no loss observed.11

On the surface, the epithelial layer becomes more fragile.12 Epithelial cells show signs of stress, especially on former lens wearers’ corneas. In fact, increased epithelial cell size and a decrease in basal epithelial cell density occur. Corneal sub-basal nerve plexus density also decreases.10 This loss of nerve fibers is transient; 90% of the fibers were not visible one month post-op, improved at six months,13 but fully restored at one year.14 This temporary loss of corneal sensitivity may explain why some patients can more easily tolerate smaller GP lenses in the weeks following CXL. It also implies that, as in dry eye patients, signs and symptoms don’t always match.


Initially, visual acuity is reduced compared to baseline, but it usually improves during the next 12 months.15 Corneal thinning associated with keratocyte apoptosis explains why visual acuity is reported as reduced and why haloes and glare are more significant at one month postoperatively. At three months, epithelial cells are thicker and stromal edema has dissipated, and this contributes to the improvement in visual performance. This better performance also results from reduced higher-order aberrations; the third, fourth, and fifth orders as well as coma, coma-like, and total high-order aberration parameters are significantly decreased after a standard CXL procedure, even four years postoperatively.16

Increased collagen compaction can be associated with corneal flattening as shown by topographic measurements.17 On average, the cornea becomes 2.0D flatter compared with baseline.18 Results depend on the technique used and the severity of ectasia. In fact, accelerated CXL showed less flattening effect than did the standard protocol3 and is less associated with visual acuity improvement. More advanced cones flatten more compared to less severe ones.19 All of the corneal changes are intrinsically linked with visual acuity modification over time. This applies to all patients, even those under 18 years of age.15

Analysis of the topographic maps indicates that the thinnest corneal pachymetry measurement, keratoconus index (KI), index of highest asymmetry (IHA), and index of highest decentration (IHD) are statistically different if compared one year postoperatively versus baseline. The KI and IHA are the most significant ones in daily clinical practice, as their variation is independent of initial corneal thickness.20 These differences may be explained by a deeper penetration of riboflavin with the standard procedure. This dose/time/response is confirmed when we compare the results of epi-on versus epi-off techniques.

In general, the epi-on technique is considered a safe approach, more comfortable, but associated with a higher rate of ectasia progression. One study showed that 23% of the patients still progressed after undergoing the epi-on technique.21 This is why epi-on is not recommended for progressive keratoconus patients,22 especially those under 30 years of age. This safer approach may evolve, considering new findings. In fact, when the cornea undergoes an extended pre-treatment time of 60 minutes, riboflavin impregnation is enhanced, and then corneas treated with the epi-on approach have similar results compared to epi-off.23


From the findings reported above, it is possible to identify a few elements that will influence what modality of lenses should be fitted or refitted to fully restore the visual acuity after a CXL procedure.

Keep in mind that the patient condition is evolving with time, more in the first six months. Several months are minimally needed to get a clear picture of the potential outcome.

The following characteristics describe a patient’s condition following CXL:

  • Post-CXL corneas may take up to a full year to be restored.
  • Symptoms and signs do not always match.
  • The cornea becomes:
    • stiffer after the treatment, with increased effect over time.
    • less sensitive to the nerve plexus, as 90% of it is destroyed.
    • thinner—any enhancement with topography-guided laser may increase this phenomenon.
  • With the standard procedure (epi-off, iontophoresis), keratometry (K) readings flatten by 2.0D on average.
  • The epithelium remains fragile (epi-off).
  • Visual acuity is worst at the beginning due to temporary hazing and stromal disruption.

Some corneas will show fewer modifications from baseline:

  • Those undergoing accelerated CXL.
  • Those undergoing the epi-on technique.
  • Those not wearing contact lenses before CXL.24

Other factors to also take into account:

  • Patients do not want to wait three to six months to restore their vision and resume lens wear.
  • In moderate and severe cases, glasses do not provide functional VA.
  • Refraction varies over time.
  • In most patients, lenses were worn before CXL.

Practitioners may consider different options to refit these patients: custom soft lenses, corneal GP lenses—with or without piggybacking over a soft lens—scleral lenses, and hybrid lenses.


Custom soft lens designs can be made thicker to compensate for corneal irregularity. Made of materials with low- to mid-Dk values, enhanced tear exchange under their surface is mandatory to alleviate hypoxic stress on the cornea, especially postoperatively. If available, the highest Dk is recommended.

Clinical Indication Consider custom soft lenses for patients who were wearing this type of lens before CXL or if they are intolerant to corneal GP lenses and sclerals.

Precaution Order lenses for three months only, and check for any curvature and refractive changes during the first year post-CXL. Given the thickness of these lenses and the poor oxygen transmission (Dk/t), also watch for early signs of corneal vascularization.


The severe loss of nerve plexus in the initial months following CXL may favor the use of corneal GP lenses by enhancing their tolerance. Corneal GPs are known to provide enhanced visual acuity due to the unique optical characteristics of rigid materials.

Corneal GP lenses are habitually designed to generate a three-point touch on the cornea, allowing for lens stability and optimal tear exchange. Under the same principle, modern keratoconus and irregular cornea designs are fitted with a feather-touch approach. This may increase the mechanical pressure on the cornea and lead to surface alteration.

It is recommended to wait for at least three months post-CXL (epi-off) to fit/refit patients in small GP lenses to prevent any mechanical stress on the fragile epithelium.

A longer delay is certainly not convenient for patients. It is possible to overcome this problem by using a piggyback technique, with a high-Dk silicone hydrogel daily disposable lens as a carrier. In the case of moderate-to-severe ectasia, select a soft lens carrier with a higher sagittal depth (i.e., select a steeper base curve and a larger diameter). Remember that 22% of the soft lens power should be taken into consideration in the optical system. This means that a +1.00D soft lens will modify the final power by +0.25D only.25

If an epi-on technique was used, including iontophoresis, lens wear can resume one month post-CXL without the need to consider a soft carrier, except when discomfort is present or to help with lens centration.

No matter the technique used, the fact that corneas are flattening during the first year post-procedure implies that the refractive power of these eyes will evolve as well. Consequently, any GP lens fitted or refitted soon after CXL treatment will have to be revisited periodically to check for optimal alignment and power accuracy. The average variation is around 1.0D for flatter and mean K and 1.5D to 2.0D for apical K readings,26 which are considered enough to change lens parameters.

Clinical Indication Consider corneal GP lenses for patients who were already fitted with this modality or for patients who cannot be fitted with scleral lenses (handling, low endothelial cell count, etc.).

Precaution Chair time and the number of follow-up visits are expected to be higher with corneal GP lenses post-CXL compared to other modalities because of the need to optimize the lens parameters and powers as the cornea heals and stabilizes over time.


Scleral lenses are increasingly considered a go-to option to fit or refit irregular cornea patients. In the case of post-CXL patients, the primary advantage of scleral lenses is the fact that they vault over the cornea, with no contact on the fragile epithelial tissue. This means that scleral lens wear can be resumed or lenses can be fitted as soon as three weeks post-CXL.

If a patient was already fitted with scleral lenses, the small changes in the corneal profile would be compensated by the tear fluid reservoir, with no need to modify the lens parameters. However, in cases of larger variations, it is expected that 60% of the patients will need modifications of the lens parameters or power.27

For a young, progressing keratoconus patient, this is why we consider discussing CXL before fitting contact lenses, then wait 21 days and proceed to the scleral CL fit.

To alleviate hypoxic stress on the fragile cornea, especially if haze is still visible, keep the lens thickness minimal (250µm) and the clearance optimal (200µm) after lens stabilization, regardless of the lens diameter.28 However, it was suggested to start with fitting the smallest lens possible, depending on the dimension of the horizontal visible iris diameter (HVID) and the limbus width, and to consider larger lenses only when issues occur.29

Clinical Indication Clinical indications for the use of scleral lenses post-CXL are numerous. We believe that it should be the first option to consider, as it helps with resuming lens wear as soon as three weeks post-CXL. Second, scleral lenses do not put mechanical stress on the cornea. Also, the loss of corneal sensitivity may prevent patients from feeling when something is wrong and lead to greater damage with any other type of lenses that have even the smallest bearing on the cornea.

Precaution Once properly designed and fitted, scleral lenses should be monitored to ensure that they do not induce even a small level of hypoxic stress on the fragilized epithelium.


The primary advantages of hybrid lenses are that they provide the full benefit of GP lenses while achieving a level of comfort similar to that with custom soft lenses. In theory, this makes hybrid lenses a good option to fit/refit post-CXL patients.

First, consider the CXL technique used, as modern hybrid lens designs are fitted with some support on the cornea, especially at the rigid inner landing zone area, and with limited clearance under the central GP zone. This means that in the case of the epi-off technique, the fitting/refitting process should be delayed at least three months postoperatively to prevent disturbing the fragile epithelial tissue.

Once fitted, the same warnings as for corneal GP lenses apply: lens parameters may need to be changed as the cornea flattens and the refractive power evolves. Regular follow up of patients is therefore mandatory.

Clinical Indication Clinical indications for hybrid lenses include patients already fitted in this modality.

Precaution Fitting must be delayed until at least three months post-CXL (epi-off).


Preservative and buffer agents can alter the ocular surface and promote inflammatory reactions. For example, this has been shown for borate buffers and polyhexamethylene biguanide (PHMB) biocide.30 GP lens solutions can also alter cells’ viability in the long term, especially if soaking and cleaning solutions are not well rinsed or are used directly on the eye.31

The safer approach to keep the fragile post-CXL cornea healthy is to rely on a nonpreserved solution such as hydrogen peroxide, especially if the lens has a polyethylene glycol (PEG) coating. In such cases, we do not recommend a particular hydrogen peroxide solution that is enhanced with a moisture matrix component. In our clinical experience, this peroxide solution leaves a film on the surface of GP lenses that may compete with/alter the PEG outcome and may lead to reduced or fluctuating visual acuity. Because a PEG solution seemed to increase corneal permeability in the epi-on study, we can assume that any toxicity related to any other chemical present in the reservoir will likely be increased.32

Clinical Indication Thus, standard hydrogen peroxide solution, with no wetting agent, is preferable.


Intrastromal corneal ring segment (ICRS) implantation is an effective procedure for visual and refractive improvement in keratoconic eyes. It may be considered as an adjunct therapy to CXL.33 Numerous questions concerning ICRSs remain, namely on the long-term efficacy of this technology and the ICRS-induced biomechanical effect. The optimal method for combined CXL and ICRS is still to be determined.34

In Canada, the use of ICRS implantation was popular five to 10 years ago, but it has tended to decline in recent years. The development of topography-guided laser surgery, in conjunction with CXL, is gaining more ground. ICRS implantation is now reserved for candidates who are not interested in or who cannot wear contact lenses of any type, relying only on glasses to improve their vision.

ICRS implantation will stretch the cornea and will make epithelial cells bulge over the segments. It may make the epithelium even more fragile, especially in the first months following the procedure. It is obvious that any previous GP or hybrid lenses will need to be modified after the procedure.

In such cases, custom soft lenses, piggyback systems, and scleral lenses are the only options that may be considered safe to protect the epithelial surface from erosion. One issue is that corneas implanted with ICRSs or modified with a topography-guided laser no longer have a keratoconic profile. GP lenses specifically designed for keratoconus correction may no longer be easily fitted, and designs for irregular corneas may be a preferable option. A trial and error approach is required, making the process less convenient for both practitioner and patient. Scleral lenses are certainly a better option to consider given that any irregularity on the surface will be well compensated by the fluid reservoir.

Precaution Neovascularization may occur postoperatively. In such cases, scleral lenses can help, if designed to alleviate hypoxic stress to the cornea.35


CXL is a procedure that is becoming widely used to stabilize keratoconus progression. Most of the patients treated will still need contact lenses post-CXL. While many options exist, it is imperative to consider the pathophysiological changes occurring during the first 12 months post-CXL to select the appropriate type of contact lenses and to manage patients safely. Scleral lenses offer the optimal option. GP, hybrid, and custom soft contact lenses may be considered, but their parameters may have to be modified over time. The potential impact of the mechanical or hypoxic stresses on the weakened epithelium should dictate the lens modality and design. In all cases, a close follow up of the patient condition is mandatory. CLS


  1. Zhang Y, Conrad AH, Conrad GW. Effects of ultraviolet-A and riboflavin on the interaction of collagen and proteoglycans during corneal cross-linking. J Biol Chem. 2011 Apr;286:13011-13022.
  2. Balparda K, Maldonado MJ. Corneal collagen cross-linking. A review of its clinical applications. Arch Soc Esp Oftalmol. 2017 Apr;92:166-174.
  3. Choi M, Kim J, Kim EK, Seo KY, Kim TI. Comparison of the Conventional Dresden Protocol and Accelerated Protocol With Higher Ultraviolet Intensity in Corneal Collagen Cross-Linking for Keratoconus. Cornea. 2017 May;36:523-529.
  4. Bikbova G, Bikbov M. Transepithelial corneal collagen cross-linking by iontophoresis of riboflavin. Acta Ophthalmol. 2014 Feb;92:e30-e34.
  5. Jacob S, Kumar DA, Agarwal A, Basu S, Sinha P, Agarwal A. Contact lens-assisted collagen cross-linking (CACXL): A new technique for cross-linking thin corneas. J Refract Surg. 2014 Jun;30:366-372.
  6. Sykakis E, Karim R, Evans JR, et al. Corneal collagen cross-linking for treating keratoconus. Cochrane Database Syst Rev. 2015 Mar 24:CD010621.
  7. Kanellopoulos AJ, Asimellis G. Keratoconus management: long-term stability of topography-guided normalization combined with high-fluence CXL stabilization (the Athens Protocol). J Refract Surg. 2014 Feb;30:88-93.
  8. Lombardo M, Serrao S, Rosati M, Ducoli P, Lombardo G. Biomechanical changes in the human cornea after transepithelial corneal crosslinking using iontophoresis. J Cataract Refract Surg. 2014 Oct;40:1706-1715.
  9. Del Buey MA, Lanchares E, Cristóbal JÁ, Junquera SR, Gotor CY, Calvo B. Immediate effect of ultraviolet-a collagen cross-linking therapy on the biomechanics and histology of the human cornea. J Refract Surg. 2015 Jan;31:70-71.
  10. Sehra SV, Titiyal JS, Sharma N, Tandon R, Sinha R. Change in corneal microstructure with rigid gas permeable contact lens use following collagen cross-linking: an in vivo confocal microscopy study. Br J Ophthalmol. 2014 Apr;98:442-447.
  11. Sharif R, Hjortdal J, Sejersen H, Frank G, Karamichos D. Human in vitro Model Reveals the Effects of Collagen Cross-linking on Keratoconus Pathogenesis. Sci Rep. 2017 Oct 2;7:12517.
  12. Li Z, Wang Y, Xu Y, Jhanji V, Zhang C, Mu G. The Evaluation of Corneal Fragility After UVA/Riboflavin Crosslinking. Eye Contact Lens. 2017 May;43:100-102.
  13. Ünlü M, Yüksel E, Bilgihan K. Effect of corneal cross-linking on contact lens tolerance in keratoconus. Clin Exp Optom. 2017 Jul;100:369-374.
  14. Ozgurhan EB, Celik U, Bozkurt E, Demirok A. Evaluation of subbasal nerve morphology and corneal sensation after accelerated corneal collagen cross-linking treatment on keratoconus. Curr Eye Res. 2015 May;40:484-489.
  15. Peyman A, Kamali A, Khushabi M, et al. Collagen cross-linking effect on progressive keratoconus in patients younger than 18 years of age: A clinical trial. Adv Biomed Res. 2015 Nov 23;4:245.
  16. Naderan M, Jahanrad A. Higher-order aberration 4 years after corneal collagen cross-linking. Indian J Ophthalmol. 2017 Sep;65:808-812.
  17. Mazzotta C, Caporossi T, Denaro R, et al. Morphological and functional correlations in riboflavin UV A corneal collagen cross-linking for keratoconus. Acta Ophthalmol. 2012 May;90:259-265.
  18. 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 Apr;149:585-593.
  19. Kasai K, Kato N, Konomi K, Shinzawa M, Shimazaki J. Flattening effect of corneal cross-linking depends on the preoperative severity of keratoconus. Medicine (Baltimore). 2017 Oct;96:e8160.
  20. Kránitz K, Kovács I, Miháltz K, et al. Changes of corneal topography indices after CXL in progressive keratoconus assessed by Scheimpflug camera. J Refract Surg. 2014 Jun;30:374-378.
  21. Soeters N, Wisse RP, Godefrooij DA, Imhof SM, Tahzib NG. Transepithelial versus epithelium-off corneal cross-linking for the treatment of progressive keratoconus: a randomized controlled trial. Am J Ophthalmol. 2015 May;159:821-828.e3.
  22. Raiskup F, Veliká V, Veselá M, Spörl E. [Cross-Linking in Keratoconus: “Epi-off” or “Epi-on”?]. [Article in German.] Klin Monbl Augenheilkd. 2015 Dec;232:1392-1396.
  23. Cruzat A, Shukla AN, Arafat SN, et al. Ex Vivo Study of Transepithelial Corneal Cross-linking. J Refract Surg. 2017 Mar 1;33:171-177.
  24. Koppen C, Gobin L, Mathysen D, Wouters K, Tassignon MJ. Influence of contact lens wear on the results of ultraviolet A/riboflavin cross-linking for progressive keratoconus. Br J Ophthalmol. 2011 Oct;95:1402-1405.
  25. Michaud L, Brazeau D, Corbeil ME, Forcier P, Bernard PJ. Contribution of soft lenses of various powers to the optics of a piggy-back system on regular corneas. Cont Lens Anterior Eye. 2013 Dec;36:318-323.
  26. Singh K, Bhattacharyya M, Arora R, Dangda S, Mutreja A. Alterations in contact lens fitting parameters following cross-linking in keratoconus patients of Indian ethnicity. Int Ophthalmol. 2017 Jun 23. [Epub ahead of print]
  27. Visser ES, Soeters N, Tahzib NG. Scleral lens tolerance after corneal cross-linking for keratoconus. Optom Vis Sci. 2015 Mar;92:318-323.
  28. Giasson CJ, Morency J, Melillo M, Michaud L. Oxygen Tension Beneath Scleral Lenses of Different Clearances. Optom Vis Sci. 2017 Apr;94:466-475.
  29. Fadel D. Modern scleral lenses: Mini versus large. Cont Lens Anterior Eye. 2017 Aug;40:200-207.
  30. Oh S, McCanna DJ, Subbaraman LN, Jones LW. Cytotoxic and inflammatory effects of contact lens solutions on human corneal epithelial cells in vitro. Cont Lens Anterior Eye. 2017 Dec 13. [Epub ahead of print]
  31. Choy CK, Cho P, Boost MV. Cytotoxicity of rigid gas-permeable lens care solutions. Clin Exp Optom. 2013 Sep;96:467-471.
  32. Caruso C, Ostacolo C, Epstein RL, Barbaro G, Troisi S, Capobianco D. Transepithelial Corneal Cross-Linking With Vitamin E-Enhanced Riboflavin Solution and Abbreviated, Low-Dose UV-A: 24-Month Clinical Outcomes. Cornea. 2016 Feb;35:145-150.
  33. Elsaftawy HS, Ahmed MH, Saif MY, Mousa R. Sequential Intracorneal Ring Segment Implantation and Corneal Transepithelial Collagen Cross-Linking in Keratoconus. Cornea. 2015 Nov;34:1420-1426.
  34. Park J, Gritz DC. Evolution in the use of intrastromal corneal ring segments for corneal ectasia. Curr Opin Ophthalmol. 2013 Jul;24:296-301.
  35. Kramer EG, Boshnick EL. Scleral lenses in the treatment of post-LASIK ectasia and superficial neovascularization of intrastromal corneal ring segments. Cont Lens Anterior Eye. 2015 Aug;38:298-303.