By Mr. Muhammed R Siddiqui , Dr. Atif Shafique
Corresponding Author Dr. Atif Shafique
Liverpool, - United Kingdom
Submitting Author Mr. Muhammed R Siddiqui
Other Authors Mr. Muhammed R Siddiqui
Mayday Hospital, 23 Malvern Road - United Kingdom TN24 8HX


Keratoconus; Collagen cross linking

Siddiqui MR, Shafique A. Corneal Collagen Cross Linking in Keratoconus. WebmedCentral OPHTHALMOLOGY 2012;3(5):WMC003351
doi: 10.9754/journal.wmc.2012.003351

This is an open-access article distributed under the terms of the Creative Commons Attribution License(CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Submitted on: 09 May 2012 10:54:07 AM GMT
Published on: 10 May 2012 06:41:32 PM GMT


Keratoconus is a non-inflammatory degenerative acquired condition affecting the shape of the cornea. The cornea begins to protrude like a cone and this eventually results in distortion, apical thinning and scarring (1).

Keratoconus usually begins during puberty and progresses through adolescence and beyond (3). It may progress for up to 10-20 years before slowing down or stabilising. Most cases affect both eyes with each eye being affected differently but it can be unilateral (2). It is associated with eye rubbing (3).  The incidence of keratoconus is approximately 1 in 20 000 (4).

Although the cause of keratoconus is unknown, there are many theories which indicate that there may be environmental (allergies and eye rubbing), endocrine or genetic elements that play a part (6). Some studies have shown that keratoconus affected corneas lack important anchoring fibrils that structurally stabilise the cornea and that this extra flexibility allows the cornea to protrude forward into a cone like shape (2).

Histological features include thinning of the stroma, iron deposits in the epithelial basement membrane and breaks within the bowman’s layer (3).

Patients present initially with vision deterioration often in one eye due to progressive myopia and astigmatism (4).  This may be described by patients as blurred vision, increased sensitivity to light, difficult night time driving, haloes, headaches or eye strain (2).

Diagnosing keratoconus may be difficult in the early stages but may be done by identifying many of the subtle changes by using a slit lamp microscope (4). Some signs that may be visualised on the slit lamp may include corneal thinning, apical scarring, Fleischer’s rings (yellow-brown iron deposits in the peripheral cornea) and vogt’s striae (vertical stress lines in the stroma and descemet’s membrane) (2,5). Following diagnosis, ophthalmologists would often like to measure the curvature of the cornea which is done by keratometry and computed corneal topography. Keratometry measures the central anterior curvature and measures corneal astigmatism (2). Corneal topography makes three dimensional maps of the cornea and has been the main way to diagnose keratoconus earlier and monitor progression accurately (2).

Acute hydrops may occur in some cases of keratoconus. This occurs when aqueous humour flows through a split descemet’s membrane and into the corneal stroma causing oedema and haziness which may cause severe pain and deteriorate vision further (4,6).

Management of keratoconus is aimed at correcting refractive errors and this is achieved initially by wearing glasses or soft contact lenses. Rigid gas permeable contact lenses may be worn if there is progression of keratoconus (2). A variety of surgical options exist if glasses or contacts prove unsuccessful. These can include insertion of intracorneal rings, however this does not affect progression. Severe cases of keratoconus in which the refractive error is too large to be treated by optical means or when there is substantial scarring may require corneal transplantation (penetrating keratoplasty) in order to improve vision (7). This applies to 10-20% of keratoconus patients who after a long recovery period after the corneal transplantation may still need to wear contact lenses (6). Although the prognosis after transplantation is good, there is still a chance of transplant rejection (6).

Riboflavin ultraviolet light induced corneal collagen crosslinking (CXL) has been a new promising method of treating keratoconus. It was developed in 1998 in Germany and is the only form of treatment that aims to treat the underlying cause of keratoconus by increasing collagen cross-links within the cornea which subsequently strengthens it (6). However this treatment is not curative, it aims to halt any further progression and glasses or contact lens may still be required (2). Early on in this procedure, the epithelium needs to be removed which is done by instilling alcohol onto the corneal for a short period and then scraping it off. Although removal of the epithelium may cause some discomfort and increase the risk for infection, studies have found that cross-linking does not occur in the presence of the epithelium as it acts like a barrier (8).

The aim of this audit is to look the outcomes of patients who have had corneal collagen cross linking (CXL) at the Royal Liverpool University Hospital (RLUH) Ophthalmology Department and compare this to other available literature.


Background reading on keratoconus was initially done using books from the RLUH and University of Liverpool. This was followed by accessing a variety of websites including the National Keratoconus Foundation and UK Keratoconus Group.

Details of patient who had CXL done at the RLUH ophthalmology department were obtained and this was then looked at to find suitable patients. Inclusion criteria included patients with keratoconus who had CXL done between November 2010 and August 2011. These dates were chosen so that there would be a sufficient sample size with a minimum of six months follow up. Furthermore, November 2010 is the earliest date that the department offered the procedure. Patients were excluded if analysis of their records deemed them unsuitable.

Data that was recorded when analysing records included: demographics (Age, gender, left/right eye), history of atopy or eye rubbing, contact lens or spectacle use, intacs implants, intraocular pressure, visual acuity both unaided and best corrected pre and post CXL, Orbscan readings (Sim K, K2 max, K1 min and thinnest) pre and post CXL, pacymetry readings and post CXL, complications and outcomes after CXL.

Online literature was obtained by using a variety of databases including Medline (via Ovid) and Pubmed. To reduce the number of results to relevant articles only limits including availability in English, human species and free full text were applied.  Articles that were deemed irrelevant were disregarded. 


After the limits were applied, there were 30 articles available of which four were suitable for analysis.

From the patient database, 29 patients (45 eyes) had CXL done from February 2011 to March 2012. Details of patients who had their CXL before February 2011 were not available. Of the 29 patients, 8 patients (10) eyes) met the inclusion criteria. Two patients were not included as one patient record was unavailable and another became available after the data collection period. This left 6 patients (7 eyes) to be analysed.

Mean age of the patients was 21 at the time of the procedure (range 15-27, 5 male 1 female, 5 left eyes, 2 right eyes).  Table 1 summarises the data collected from patient records. Table 2 to Table 7 show the data collected from Patient 1 to 6 consecutively.


CXL procedure protocol at RLUH

All eyes have topical anaesthetic and corneal thickness measured before a doctor administers a few drops of 20% alcohol onto the cornea and using a surgical scalpel to remove the epithelium.

The corneal thickness should be measured again and if it is greater than 400?m then isotonic riboflavin with dextran is used as it may maintain or marginally swell the cornea (up to 2%). If corneal thickness is less than 400?m then hypotonic riboflavin without dextran is used as it causes the cornea to swell. One drop of riboflavin is given every 5 minutes for a total period of 30 minutes. Corneal thickness should be checked again and if still less than 400?m then hypotonic riboflavin without dextran is given every 2-3 minutes for 10 minutes until the cornea swells to more than 400?m. Riboflavin diffusion into the cornea should then be checked by the slit lamp if available.

The UV aiming beam should then be adjusted correctly to be in position 45mm above the cornea and at the appropriate spot size as to not irradiate stem cells in the corneal limbus. The machine should be set to deliver 3mw/cm2 for 30 minutes and riboflavin with dextran drops administered every 5 minutes. Proxymetacaine drops can be given as required by the patient for comfort.

Patients are given proxymetacaine, cyclopentolate and chloramphenicol drops after the procedure along with a bandage contact lens, eye pad, take home analgesia, written and oral aftercare advice, emergency contact information and an outpatient appointment in one week.

Current literature

Wittig-silva et al conducted a randomised controlled trial of CXL in 66 eyes with documented progression of keratoconus (9). They found that eyes in the intervention group had flattening of the K-max value by an average of 0.74 D, 0.92 D, and 1.45 D at 3,6 and 12 months consecutively (9). This was also associated with an improvement in BCVA.  Eyes in the control group had steepening of the K-max by 0.60 D, 0.60 D and 1.28 D at 3, 6 and 12 months consecutively (9). This was also associated with a decline in BCVA (9).  They concluded that CXL had temporarily stabilised keratoconus in the treated eyes  (9).

Another retrospective longitudinal study analysed the outcomes of CXL on 46 eyes after two years and found that despite a temporary reduction in vision, the long term outcomes showed an improvement in visual acuity in the majority of the eyes (10).

Shetti et al found that CXL was safe for the surgeons despite very small quantities of radiation reaching them which was dampened further by wearing gloves, a gown and UV protective glasses (11).

One study used a femtosecond laser to create a 100?m deep, 7mm diameter intrastromal pocket which was instilled with riboflavin and irradiated with 7mW/cm2 ultraviolet A (UVA) for fifteen minutes (12). This was done on ten eyes with early keratoconus (12). They found that mean unaided visual acuity improved from 20/40.5 (6/12) to 20/32.5 (6/9.5) (12). Best corrected visual acuity remained unchanged at 20/20 (6/6) (15). Maximum mean keratometry reduced from 48.7 D to 47.90 D (12). Mean corneal thickness was found to decrease initially by 5% but returned to pre operative levels by 18 months (12). There were no adverse effects reported in any of the eyes (12). The author concluded that this novel epithelium sparing method of intrastromal riboflavin instillation with 7 mW/cm2 UVA was effective, safe and reduced post operative pain (12). By removing the need to remove the epithelium, this study avoided the risk of keratitis which could occur in the 3-4 day re-epithelialisation interval. Furthermore, all patients resumed full activities within a day and therefore had a rapid visual rehabilitation period. However, as the study had a small number of patients, further studies with a larger sample size and longer follow up are needed to validate the results.

Comparison to our audit

Patients in our audit had an average pre-CXL corneal thickness of 438.7?m whilst in the femtosecond laser study(15) all patients had a corneal thickness over 500?m indicating that the audit patients had more advanced keratoconus. Furthemore, the patients from our audit had a pre-CXL maximum mean keratometry value of 53.15 D compared to 48.7 D from the study, again indicating that our patients had more advanced keratoconus.

Corneal thickness is important as one study reported that the corneal endothelium, lens or retina will not experience any damage during CXL as long as the corneal thickness is the recommended minimum thickness of 400?m (13). All seven eyes from the audit had a thinner cornea post-CXL compared to pre-CXL. Only two eyes subsequently had a thicker cornea, both after six months. Patient 3’s eye was very thin six months after the procedure at 260?m. All other eyes did not have a second post-CXL orbscan which would have otherwise been useful to compare whether the cornea had become thicker.

The average Sim K value had reduced marginally from -5.6 D pre-CXL to the most recent post-CXL reading of -5.77 D. However two eyes had an improvement in Sim K value whilst all others had a reduction. The mean maximum keratometry value (K2) also increased marginally from 53.15 D pre-CXL to the most recent reading post-CXL reading of 53.47 D.  However two eyes had a reduction in K2. Although the orbscan changes may not have been significant, visual acuity changes were significant. Six of the seven eyes had an improvement in visual acuity when comparing their most recent pre-CXL unaided visual acuity to their latest recorded unaided visual acuity. Two of the six eyes had an improvement with their vision when corrected with spectacles or contact lens.

Four eyes have only had one orbscan reading post-CXL, with all of them done in a relatively short period after the procedure. This does not give us the mid to long term outcomes of CXL or allow us to monitor if there has been any progression of keratoconus. Two eyes had two orbscan readings after CXL and one eye had three orbscan readings post CXL.

After observing patients undergoing CXL and discussing with the corneal nurse specialist, corneal thickness during CXL is measured reliably at the RLUH ophthalmology department. This is done by taking 3 readings using an ultrasound pachymeter and recording the mean of three results. Furthermore, documentation of corneal thickness after debridement of the epithelium and after instillation of riboflavin are now always being recorded, something which was not done every time on the patients included in this audit.

Clinical Recommendations

Ultrasound pachymetry is more accurate than orbscan readings for corneal thickness and is considered the gold standard (14). This method of recording the corneal thickness could be used pre and post CXL for accurate monitoring of corneal thickness. 

Patient two should have had hypotonic riboflavin drops administered instead of isotonic riboflavin drops in order to try and swell the cornea to above 400?m. For this reason, the RLUH ophthalmology protocol for CXL needs to be followed continuously throughout the procedure. This will ensure that the correct solution of riboflavin which is dependent on the corneal thickness is selected and no steps are missed out, such as recording the corneal thickness.

Although patients are followed up regularly in clinic and visual acuity measured, orbscan measurements are not always carried out regularly which would be helpful to monitor the outcome of CXL or any further progression of keratoconus.



Although there are newer more novel methods of CXL that appear to have many benefits such as using a femtosecond laser to create intrastomal pockets, they need further, more expanded studies to validate existing results. Other studies have found that CXL is safe and effective as it has improved visual acuity in the majority of patients.

From the audit point of view, CXL is a safe procedure as there were no complications apart from conjunctivitis and keratitis, both of which resolved quickly. It was also an effective procedure with nearly all the eyes reporting an improvement in visual acuity. As keratoconus is a progressive condition, it was difficult to identify whether any poor outcomes, apart from corneal thickness, were due to CXL failure or keratoconus progression. A more expanded audit with a longer follow up and more regular orbscan measurements will be necessary to make a more valid and reliable conclusion.




1. Derakhshan A, Shandiz JH, Ahadi M, Daneshvar R, Esmaily H. Short-term Outcomes of Collagen Crosslinking for Early Keratoconus. J Ophthalmic Vis Res, 2011;6:155-9.
2. Keratoconus. National Keratoconus Foundation. (accessed 25/03/12)
3. Rabinowitz YS. Keratoconus. Surv Ophthalmol, 1998;42:297-319.
4. Batterbury M, Bowling B, Murphy C. Ophthalmology: An Illustrated Colour Text. Edinburgh: Churchill Livingstone, 2009.
5. Güngör IU, Beden U, Sönmez B. Bilateral horizontal Vogt’s striae in keratoconus. Clin Ophthalmol, 2008;2:653-5.
6. Hydrops and keratoconus. UK Keratoconus Self-Help and Support Association. (accessed 25/03/12).
7. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-a-induced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol, 2003;135:620-7.
8. Bottós KM, Dreyfuss JL, Regatieri CV, Lima-Filho AA, Schor P, Nader HB, et al. Immunofluorescence confocal microscopy of porcine corneas following collagen cross-linking treatment with riboflavin and ultraviolet A. J Refract Surg, 2008;24:715-9.
9. Wittig-Silva C, Whiting M, Lamoureux E, Lindsay RG, Sullivan LJ, Snibson GR. A randomized controlled trial of corneal collagen cross-linking in progressive keratoconus: preliminary results. J Refract Surg, 2008;24:720-5.
10. Kampik D. Koch M. Kampik K. Geerling G. Corneal riboflavin/UV-A collagen cross-linking (CXL) in keratoconus: two-year results. Klin Monatsbl Augenheilkd, 2011;228:525-30.
11. Shetty R, Shetty R, Mahendradas P, Shetty BK. Are the surgeons safe during UV-A radiation exposure in collagen cross-linking procedure? Cornea, 2012 31:167-71.
12. Kanellopoulos AJ. Collagen Cross-linking in Early Keratoconus With Riboflavin in a femtosecond Laser-created Pocket: Initial Clinical results. J Refract Surg, 2009;25:1034-37.
13. Spoerl E, Mrochen M, Sliney D, Trokel S, Seiler T. Safety of UVA-riboflavin cross-linking of the cornea. Cornea, 2007;26:385-9.
14. Cordeiro Barbosa MM, Barbosa JB Jr, Hirai FE, Hofling-Lima AL. Effect of cross-linking on corneal thickness in patients with corneal edema. Cornea, 2010;29:613-7.

Source(s) of Funding


Competing Interests



This article has been downloaded from WebmedCentral. With our unique author driven post publication peer review, contents posted on this web portal do not undergo any prepublication peer or editorial review. It is completely the responsibility of the authors to ensure not only scientific and ethical standards of the manuscript but also its grammatical accuracy. Authors must ensure that they obtain all the necessary permissions before submitting any information that requires obtaining a consent or approval from a third party. Authors should also ensure not to submit any information which they do not have the copyright of or of which they have transferred the copyrights to a third party.
Contents on WebmedCentral are purely for biomedical researchers and scientists. They are not meant to cater to the needs of an individual patient. The web portal or any content(s) therein is neither designed to support, nor replace, the relationship that exists between a patient/site visitor and his/her physician. Your use of the WebmedCentral site and its contents is entirely at your own risk. We do not take any responsibility for any harm that you may suffer or inflict on a third person by following the contents of this website.

0 reviews posted so far

0 comments posted so far

Please use this functionality to flag objectionable, inappropriate, inaccurate, and offensive content to WebmedCentral Team and the authors.


Author Comments
0 comments posted so far


What is article Popularity?

Article popularity is calculated by considering the scores: age of the article
Popularity = (P - 1) / (T + 2)^1.5
P : points is the sum of individual scores, which includes article Views, Downloads, Reviews, Comments and their weightage

Scores   Weightage
Views Points X 1
Download Points X 2
Comment Points X 5
Review Points X 10
Points= sum(Views Points + Download Points + Comment Points + Review Points)
T : time since submission in hours.
P is subtracted by 1 to negate submitter's vote.
Age factor is (time since submission in hours plus two) to the power of 1.5.factor.

How Article Quality Works?

For each article Authors/Readers, Reviewers and WMC Editors can review/rate the articles. These ratings are used to determine Feedback Scores.

In most cases, article receive ratings in the range of 0 to 10. We calculate average of all the ratings and consider it as article quality.

Quality=Average(Authors/Readers Ratings + Reviewers Ratings + WMC Editor Ratings)