ASCRS News: EyeWorld Journal Club
June 2022

by Neal Patel, MD, and Nandini Venkateswaran, MD

Over the last decade, corneal crosslinking (CXL) has rapidly grown in popularity among ophthalmologists for its ability to halt the progression of keratoconus and other corneal ectasias. The Dresden protocol is widely regarded as the standard for CXL, against which other protocols for CXL are compared.¹ In this standard epithelium-off (epi-off) technique, the central 8 mm of corneal epithelium is removed before saturation of the corneal stroma with riboflavin and subsequent treatment with UV-A.² Removing the epithelium has been shown to allow improved absorption of riboflavin and UV-A penetration into the corneal stroma.^3–5 However, this technique is relatively contraindicated in corneas with pachymetry less then 400 microns after epithelium removal due to risk of endothelial damage by UV-A.⁶ Additionally, epithelial debridement is associated with complications including increased postoperative pain, corneal haze, and risk of infective keratitis.⁷

Several alternative techniques have been described to overcome these limitations of epi-off techniques and improve the safety of CXL in thinner corneas. Among these techniques are transepithelial CXL (TECXL) and contact lens-assisted CXL (CACXL). TECXL avoids epithelial removal and uses a riboflavin solution containing substances that enhance epithelial penetration.⁸ This technique has been demonstrated to have improved postoperative comfort and reduced corneal haze, however, the efficacy of this technique as compared to standard epi-off CXL has been questioned.^9–11 CACXL involves the use of a riboflavin-soaked UV barrier-free contact lens placed on a deepithelialized cornea to compensate for reduced corneal thickness and prevent damaging levels of UV-A from reaching the endothelium.¹² In this study, the authors retrospectively compared outcomes of TECXL and CACXL techniques along with standard epi-off CXL in patients with progressive keratoconus.

Design and methods

This retrospective, single-center study included 58 eyes that underwent CXL for progressive keratoconus at the Cornea Services of Advanced Eye Center in Chandigarh, India, between January 2013 and June 2018. Patients were included if they had keratoconus with evidence of progression, as defined by consistent changes in keratometry, refraction, and/or pachymetry on two consecutive visits and had post-procedure follow-up of at least 2 years. Keratometry and corneal thickness measurements were performed with Scheimpflug imaging on Pentacam (Oculus). The standard epi-off technique was used only in eyes with minimal corneal thickness (MCT) of at least 450 microns as measured with an intact epithelium. TECXL and CACXL were used for patients with MCT between 350 microns and 450 microns. There were no predetermined criteria or systematic approach to select among epi-off, TECXL, and CACXL techniques in patients, except that TECXL was initially avoided in patients under the age of 18 years due to concern of possible reduced efficacy in the pediatric population.¹³

All patients were examined on the first and fifth postoperative day, 1 month, 3 months, and 6 months, followed by every 6 months. Efficacy of each CXL technique was judged based on the ability of the procedure to halt or stabilize ectasia. Progression, stabilization, or regression of ectasia was determined by change in maximum keratometry (Kmax) at the 2-year follow-up as compared to baseline, with stability defined as change in Kmax within ±1 D of baseline and progression or regression falling greater than 1 D above or below this range, respectively. Additional outcome measures included corrected distance visual acuity (CDVA), manifest refractive spherical equivalent (MRSE), corneal densitometry, and topographic measurements such as Kmax and simulated keratometry (SimK).

Results

Among the 58 eyes that met inclusion criteria, standard epi-off CXL was performed in 34 eyes, TECXL in 10 eyes, and CACXL in 14 eyes. Patients were mostly male (71%) with a mean age of 19.26 years. Baseline parameters were largely comparable between groups, with no significant difference found among CDVA, MRSE, or Kmax in the groups. Patients who underwent TECXL were significantly older than those who underwent standard epi-off CXL and CACXL. Mean thinnest pachymetry was similar between the CACXL and TECXL groups and, as expected, significantly thinner when compared to those that underwent standard epi-off CXL. Of note, there was a trend for eyes in the CACXL group to have a higher Kmax at baseline as compared to the two other groups (p=0.09).

All three techniques were similarly effective at halting progression of ectasia at 2 years (p=0.61). In the standard epi-off CXL group, regression was seen in 62% of eyes (n=21), stabilization in 32% (n=11), and progression in 6% (n=2). In the CACXL and TECXL groups, regression was seen in 79% (n=11) and 80% (n=8) of eyes, respectively, while the remaining eyes demonstrated stabilization. No progression was seen in any eyes in the CACXL or TECXL groups. At the 2-year follow-up, the largest reduction in Kmax was seen in the CACXL group at –3.18±2.74 D, followed by –2.83±3.35 D in the standard epi-off CXL and –2.02±1.66 D in the TECXL group. Between group comparison of reduction in Kmax as well as reduction in the flat, steep, and mean simulated keratometry at 2 years was comparable among all three groups.

Assessment of CDVA at 2 years demonstrated improvement in all groups when compared to baseline, however, this reached statistical significance only in the standard epi-off CXL group (p=0.006). Between group comparison revealed that improvement in CDVA was similar among all groups (p=0.46). A statistically significant improvement in MRSE was seen within all three groups that also remained similar in all groups throughout the follow-up period.   

One patient in the TECXL group developed herpetic epithelial keratitis (geographic ulcer) at the 3-week follow-up and was managed successfully with topical ganciclovir. No other complications were reported.

Each of these three techniques must be chosen judiciously and can play a useful role in the treatment of patients with progressive keratoconus …

Discussion

Variability in CXL protocols among studies, especially for newer techniques, can make comparison of outcomes difficult. This is the first study that compares standard epi-off CXL, TECXL, and CACXL simultaneously in one clinic setting. The results of this study suggest that both TECXL and CACXL are comparable to standard epi-off CXL in regard to structural outcomes, as demonstrated by halting progression of ectasia and reduction in Kmax, as well as functional outcomes, as evidenced by improvement in CDVA. Although the exact cut-off for a “thin” cornea remains a topic of debate, TECXL and CACXL performed on corneas as thin as 357 microns in this study showed effective results and no evidence of endothelial toxicity. While the reduction in Kmax was significant among all groups, the greatest reduction was seen in the CACXL group, which was possibly related to the relatively higher baseline Kmax in this group, or the thinner baseline pachymetry—two factors that have both been associated with greater corneal flattening after CXL in previous publications.^13,14

Detailed evaluation of depth of treatment demarcation lines as well as endothelial cell structure and function would have been useful to have but were unfortunately not performed in this retrospective study. Complications were minimal and interestingly, only occurred in the TECXL group that did not undergo true epithelial debridement; however, exposure of the corneas to significant amounts of benzalkonium chloride in the riboflavin solution may have rendered the epithelium more irregular and susceptible to keratitis in these patients. The retrospective nature, limited sample size particularly in the TECXL and CACXL groups, and lack of randomization or systematic selection of CXL techniques are notable limitations that may have prevented some observed trends from reaching statistical significance.

Conclusions

The results of this study suggest that both TECXL and CACXL are comparable to standard epi-off CXL in regressing and/or stabilizing progressive keratoconus over a 2-year follow-up period. Each of these three techniques must be chosen judiciously and can play a useful role in the treatment of patients with progressive keratoconus, with TECXL and CACXL being particularly useful in patients with the thinnest pachymetric readings below 400 microns. More research with larger sample sizes and standardized treatment protocols are needed to investigate these techniques further and help clinicians determine when and in which patients each of these techniques should be implemented to treat corneal ectasias. 

---

Comparison of contact lens-assisted and transepithelial corneal crosslinking with standard epithelium-off crosslinking for progressive keratoconus: 24-month clinic results

Malhotra C, et al. J Cataract Refract Surg. 2022;48(2):199–207

• Purpose: To compare outcomes of contact lens-assisted corneal crosslinking (CACXL) and transepithelial corneal crosslinking (TECXL) with standard “epithelium-off” (“epi-off”) crosslinking (CXL) for progressive keratoconus
• Setting: Advanced Eye Centre, Post Graduate Institute of Medical Education and Research, Chandigarh, India 
• Design: Retrospective, comparative study
• Methods: Patients with progressive keratoconus undergoing CXL with a minimum follow-up of 24 months were included. CACXL and TECXL was performed in patients with “epithelium-on” minimal pachymetry between 350 µm and 450 µm. Main outcome measures included change in maximum keratometry (Kmax), corrected distance visual acuity (CDVA), and efficacy in halting progression (increase in Kmax ≥ 1 diopter [D]).
• Results: Standard “epi-off” CXL, CACXL, and TECXL was performed in 34, 14, and 10 eyes, respectively. Baseline Kmax and CDVA were comparable for all groups. Kmax was reduced significantly by –2.83±3.35 D, –3.18±2.74 D, and –2.02±1.66 D in the standard “epi-off” CXL (p<0.01), CACXL (p=0.001), and TECXL (p=0.004) groups, respectively; the reduction was comparable for all groups (p=0.63). CDVA improved by –0.14±0.24, –0.04±0.19, and –0.12± 0.17 logMAR units in the standard “epi-off” CXL (p=0.006), CACXL (p=0.42), and TECXL (p=0.05) groups, respectively; the reduction was comparable for all groups (p=0.46). Progression was documented in 2 eyes (6%) of the standard “epi-off” CXL group and 0% eyes of the CACXL and TECXL groups (p=0.61).
• Conclusion: CACXL and TECXL were comparable to standard “epi-off” CXL for progressive keratoconus.

---

References

1. Subasinghe SK, et al. Current perspectives on corneal collagen crosslinking (CXL). Graefes Arch Clin Exp Ophthalmol. 2018;256:1363–1384. 
2. Wollensak G, et al. Riboflavin/ultraviolet-a-induced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol. 2003;135:620–627. 
3. Podskochy A. Protective role of corneal epithelium against ultraviolet radiation damage. Acta Ophthalmol Scand. 2004;82:714–717. 
4. Malhotra C, et al. In vivo imaging of riboflavin penetration during collagen cross-linking with hand-held spectral domain optical coherence tomography. J Refract Surg. 2012;28:776–780. 
5. Wollensak G, Iomdina E. Biomechanical and histological changes after corneal crosslinking with and without epithelial debridement. J Cataract Refract Surg. 2009;35:540–546. 
6. Spoerl E, et al. Safety of UVA-riboflavin cross-linking of the cornea. Cornea. 2007;26:385–389. 
7. Dhawan S, et al. Complications of corneal collagen cross-linking. J Ophthalmol. 2011;2011:869015. 
8. Leccisotti A, Islam T. Transepithelial corneal collagen cross-linking in keratoconus. J Refract Surg. 2010;26:942–948. 
9. Cifariello F, et al. Epi-off versus epi-on corneal collagen cross-linking in keratoconus patients: A comparative study through 2-year follow-up. J Ophthalmol. 2018;2018:4947983. 
10. Magli A, et al. Epithelium-off corneal collagen cross-linking versus transepithelial cross-linking for pediatric keratoconus. Cornea. 2013;32:597–601. 
11. Soeters N, et al. Transepithelial versus epithelium-off corneal cross-linking for the treatment of progressive keratoconus: a randomized controlled trial. Am J Ophthalmol. 2015;159:821–828.e3
12. Jacob S, et al. Contact lens-assisted collagen cross-linking (CACXL): A new technique for cross-linking thin corneas. J Refract Surg. 2014;30:366–372. 
13. Greenstein SA, Hersh PS. Characteristics influencing outcomes of corneal collagen crosslinking for keratoconus and ectasia: implications for patient selection. J Cataract Refract Surg. 2013;39:1133–1140. 
14. Spadea L, Mencucci R. Transepithelial corneal collagen cross-linking in ultrathin keratoconic corneas. Clin Ophthalmol. 2012;6:1785–1792. 

Contact 

Lorch: Alice_Lorch@meei.harvard.edu
Patel: neal_patel@meei.harvard.edu
Venkateswaran: nandini_venkateswaran@meei.harvard.edu