SUMMARY OF EVIDENCE FOR CXL
Corneal Cross-Linking (CXL) is a treatment to prevent keratoconus getting worse. It is successful in over 90% of cases. After treatment, spectacles or contact lenses will still be needed. Although vision is often hazy at first, most patients can resume contact lens wear and return to work after 1 week. As with all operations there are risks: CXL is safe, but there is a small chance (about 1 in 30) of worse vision afterwards.
Does it work?
Evidence for the standard, epithelium-off corneal cross-linking (CXL) is found in 3 randomised-controlled trials (RCTs) that have followed patients for at least 12-months after treatment.1-3 A study from Australia reported 3 year outcomes in 46 treated eyes. CXL worked in all but one patient, with most treated eyes benefitting from improvements in the shape of the cornea. Vision in spectacles also improved by an average of one line on the reading chart . Only two eyes developed minor complications that did not affect their vision in the long term. A second trial led by David O’Brart at St. Thomas’ Hospital in London reported 18-month outcomes in 24 patients with progressive keratoconus. This study confirmed improvements in corneal shape and vision in all treated eyes . A third trial from the USA allowed untreated eyes in the ‘control’ group to be treated after just 3 months, undermining any useful comparison between treated and untreated patients beyond this time point .
Longer-term outcomes are available up to 6 years after treatment [4-8]. The Siena Eye Cross Study followed 44 patients with progressive keratoconus: one eye was treated with CXL and fellow eye left untreated. The average follow-up was 52 months (range 48-60 months). Untreated eyes, in which disease progression was less rapid than treated eyes prior to CXL, were observed initially then crossed over to CXL after 2 years. Their data confirmed CXL worked in all treated eyes, with impressive improvements in corneal shape improvements and vision (2 lines on the reading chart). Fellow untreated eyes deteriorated in the first 2 years; once these eyes were also treated, they similarly improved . One problem with this type of study is that although keratoconus usually occurs in both eyes, very often one eye is progresses much more rapidly that the other. This means that the effects of a standard ‘dose of CXL may not be the same in both eyes. David O’Brart and colleagues reported follow-up data in 29 treated eyes with between 4 and 6 years follow-up. They authors confirmed more modest, but nonetheless statistically significant, improvements in corneal shape and vision . Five year results were published in the same year (2013) by Hashemi and colleagues in Iran  40 eyes of 30 patients were treated, all of which had stable corneal shape during the 5 years of monitoring. No sight threatening complications were reported.
There is very little data on the use of CXL in children. One study reported results in 40 children with an average age of 14 years. Their results 2 years after the treatment showed impressive improvements in vision (2 lines on the vision chart) with stabilisation of the corneal shape. Further studies are required in children to establish whether the response to treatment is the same as in adults and, crucially, whether the treatment lasts. It may be that patients treated as young teenagers need monitoring well into their twenties and may require repeat treatment if the keratoconus starts progressing again.
Newer treatment variations
Accelerated treatment protocols for CXL based on shorter ultraviolet (UV) light exposure times and higher UV power are emerging. The underlying premise is that delivering a similar total UV energy over a shorter period of time will not compromise safety or effectiveness in comparison with the standard treatment. There is very little published data on accelerated CXL. One study  has compared 13 eyes undergoing conventional CXL with 13 eyes treated with accelerated CXL. With only 6 months’ follow-up, the authors demonstrated similar improvements in corneal shape in each group, with no apparent harm to the corneas treated with higher power settings. Other studies comparing more accelerated UV protocols, including both pulsed and continuous wave UV light exposure, are currently being undertaken.
Is it safe?
Although generally safe, a range of complications may occur following cross-linking. Localised inflammation of the cornea (sterile infiltrates) occur in up to 7% of patients;  this is usually treated with additional steroid drops and does not normally result in vision loss. Corneal infections have also been reported, despite the use of antibiotic drops after the treatment . Mild temporary corneal haze is common after CXL and typically resolves within a year. Anecdotally, haze appears to be much less common after accelerated CXL. About 3% of patients will lose some vision in the treated eye as a result of infection, scarring or corneal surface shape irregularity. In most cases this visual loss is potentially reversible with a corneal transplant. It is worth remembering that without CXL treatment, at least 20% of all patients with keratoconus will eventually require a corneal transplant .
 Wittig-Silva, C. et al. A Randomized, Controlled Trial of Corneal Collagen Cross-Linking in Progressive Keratoconus. Ophthalmology 1–10 (2014).
 O’Brart, D. P. S., Chan, E., Samaras, K., Patel, P. & Shah, S. P. A randomised, prospective study to investigate the efficacy of riboflavin/ultraviolet A (370 nm) corneal collagen cross-linkage to halt the progression of keratoconus. British Journal of Ophthalmology 95, 1519–1524 (2011).
 Hersh, P. S., Greenstein, S. A. & Fry, K. L. Corneal collagen crosslinking for keratoconus and corneal ectasia: One-year results. J Cataract Refract Surg 37, 149–160 (2011).
 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 149, 585–593 (2010).
 Hashemi, H., Seyedian, M. A., Miraftab, M., Fotouhi, A. & Asgari, S. Corneal Collagen Cross-linking with Riboflavin and Ultraviolet A Irradiation for Keratoconus. Ophthalmology 120, 1515–1520 (2013).
 O’Brart, D. P. S., Kwong, T. Q., Patel, P., McDonald, R. J. & O’Brart, N. A. Long-term follow-up of riboflavin/ultraviolet A (370 nm) corneal collagen cross-linking to halt the progression of keratoconus. British Journal of Ophthalmology 97, 433–437 (2013).
 Vinciguerra, R. et al. Corneal Cross-Linking as a Treatment for Keratoconus. Ophthalmology 120, 908–916 (2013).
 Raiskup-Wolf, F., Hoyer, A., Spoerl, E. & Pillunat, L. E. Collagen crosslinking with riboflavin and ultraviolet-A light in keratoconus: Long-term results. J Cataract Refract Surg 34, 796–801 (2008).
 Cınar, Y. et al. Comparison of accelerated and conventional corneal collagen cross-linking for progressive keratoconus. Cutan Ocul Toxicol 1–5 (2013). doi:10.3109/15569527.2013.834497
 Koller, T., Mrochen, M. & Seiler, T. Complication and failure rates after corneal crosslinking. J Cataract Refract Surg 35, 1358–1362 (2009).
 Snibson, G. R. Collagen cross-linking: a new treatment paradigm in corneal disease – a review. Clin Experiment Ophthalmol 38, 141–153 (2010).
 Tuft, S. J., Moodaley, L. C., Gregory, W. M., Davison, C. R. & Buckley, R. J. Prognostic factors for the progression of keratoconus. Ophthalmology 101, 439–447 (1994).