Corneal Crosslinking: Present and Future : The Asia-Pacific Journal of Ophthalmology

Secondary Logo

Journal Logo

Review Article

Corneal Crosslinking: Present and Future

Angelo, Lize MBChB, PGDipOphthBS; Gokul Boptom, Akilesh PhD; McGhee, Charles DSc, FRCOphth; Ziaei, Mohammed MD, FRCOphth

Author Information
Asia-Pacific Journal of Ophthalmology: September/October 2022 - Volume 11 - Issue 5 - p 441-452
doi: 10.1097/APO.0000000000000557
  • Open


Keratoconus is a progressive, corneal thinning disorder that leads to myopia, irregular astigmatism, and scarring.1 Keratoconus is a complex clinical entity where late diagnosis, poor surveillance, and delayed treatment can result in preventable vision loss, disproportionately impacting patients’ quality of life and adversely affecting public health resources.2

The reported prevalence of keratoconus varies between ethnicities with high rates in Middle Eastern, Indian, and Maori populations.3 In New Zealand, a pediatric population-based study estimated that Maori ethnicity had prevalence rates 4 times higher than the general population (2250/100,000 vs 520/100,000). A hospital-based study in the UK estimated that people of Asian descent (Indian, Pakistani, Bangladeshi, and other Asians) had a 4-times higher prevalence compared with Europeans (229/100,000 vs 57/100,000).3 The highest prevalence to date has been reported in a pediatric population in Saudi Arabia (4790/100,000).4 Without treatment, approximately 20% of patients require corneal transplantation, with some countries reporting keratoconus as the leading indication, accounting for 40% to 45% of corneal transplants annually.5,6

The pathophysiology of keratoconus is multifactorial, with multiple reported genetic mutations and numerous implicated environmental factors such as climate, geography, and external factors such as eye rubbing and atopy.5 Studies have shown that the thinning and protrusion of the cornea leads to an increase in corneal power caused by changes in the structures of the epithelial basement membrane, keratocytes, stromal collagen fiber, and extracellular matrix leading to reduced corneal stiffness.7 Proteomic evaluation of the corneal epithelium and stroma has revealed alterations of structural extracellular matrix proteins such as decorin, lumican, and keratocan, and collagen types I and V, while cytokine dysregulation, oxidative stress, as well as alterations in the extracellular matrix have also been described.7 The above changes may represent multiple processes coinciding that ultimately result in the phenotypic characteristics of keratoconus.7

Corneal crosslinking (CXL) is a technique used to strengthen corneal tissue. CXL utilizes riboflavin (vitamin B2) as a photosensitizer and ultraviolet-A light (UVA) to ultimately slow or prevent disease progression in patients with keratoconus by creating strong covalent bonds within the corneal stroma, increasing stiffness, which withstands keratoconic progression.8 The technique has been proven to be effective, with some advocating for immediate bilateral sequential treatment of patients to avoid unnecessary delays and potential progression of the ectatic process.9 With almost 20 years since the introduction of CXL, this article aims to review current protocols, emerging techniques and potential future developments in the field.10


Crosslinking encompasses the creation of bonds that connect polymer chains. These bonds between proteins or other large molecules make the material more resistant to degeneration.11 Crosslinking of collagen and elastin molecules was first proposed in the 1970s. Then in 1997, Spoerl et al12 used this principle in the laboratory on porcine eyes with several different protocols using a combination of UVA, blue light, sunlight, and riboflavin. Riboflavin is thermostable and photosensitive and acts as the primary transducer in the crosslinking reaction.13 Its alkyl isoalloxazine structure allows light absorption across a broad spectrum, peaking at 300 and 370 nm within the UV spectra and 430 nm within the blue visible light spectra.13 Spoerl et al12 demonstrated that the combination of riboflavin and UVA successfully increased corneal stiffness, and this technique was subsequently implemented in vivo by Wollensak et al10 on patients with progressive keratoconus in 2003. This method was refined into the conventional CXL procedure commonly known as the Dresden protocol today.

The process of CXL occurs between riboflavin and UVA in a photo-oxidative manner. Although the exact mechanism of crosslinking is unknown, experimental data have shown that the 2 main mechanisms are a mixture of both aerobic and anaerobic photochemical reactions.13 In the early aerobic phase, the excited riboflavin molecule in its triplet state reacts directly with ground-state oxygen, producing reactive oxygen species (ROS), a “type 2” reaction. Once the oxygen supply is consumed, which occurs in the first 15 seconds of UVA initiation, a “type 1” anaerobic reaction occurs whereby the riboflavin triplet state reacts directly with stromal proteins, producing ROS. Rapid replenishment of oxygen occurs again once the UVA irradiation is discontinued.13 ROS are created as byproducts, and they mediate the formation of crosslinks by further reacting with stromal and collagen proteins increasing corneal stiffness.1,14

Postoperative Cellular and Ultrastructural Changes

The typical postoperative findings associated with crosslinking are stromal haze and cellular morphology changes. Anterior corneal stromal haze usually occurs in the first month following crosslinking and resolves within 20 weeks.1,15 CXL also leads to a change of stromal refractive index due to increased collagen fiber diameter and spacing and is represented clinically as a stromal demarcation line representing the depth of crosslinked tissue.1 In vivo confocal microscopy performed immediately after crosslinking shows evidence of stromal edema, a reduction of sub-basal nerve fiber density and keratocyte density.16 These changes slowly resolve within 12 months, with corneal thickness and sub-basal nerve and stromal keratocytes density returning to baseline levels.16


In this conventional protocol, the central 7 mm corneal epithelium is debrided, riboflavin 0.1% in 20% dextran applied for 30 minutes, and the cornea irradiated with UVA (370 nm, 3 mW/cm2) at a 1-cm distance for 30 minutes for a total surface energy exposure of 5.4 J/cm2 (Fig. 1).10 In a study of 23 eyes published in 2003, disease progression was stopped in all eyes with 4 years of follow-up, with 70% of eyes exhibiting a regression of keratometric and refractive parameters.10 Ten-year follow-up of 34 eyes treated using the Dresden protocol has demonstrated good long-term stability and safety with an average maximum keratometry reading (Kmax) reduction of 3.64 diopters (D) and an increase of 0.14 logMAR best-corrected distance visual acuity (CDVA).17 Only 2 eyes did not stabilize and required repeat CXL at 5 and 10 years, respectively.

Intraoperative image of crosslinking demonstrating corneal fluorescence during ultraviolet-A light irradiation.

To date, 9 randomized control trials (RCT) have reported on the clinical outcomes of conventional CXL in the treatment of keratoconus (Table 1). The majority of the RCTs were of a crossover design and included between 15 and 90 eyes. These studies reported a 73.3% to 100% efficacy, with all studies demonstrating an improvement in Kmax ranging from 0.22 to 2.0 D.

TABLE 1 - List of Randomized Control Trials Looking at Conventional CXL Versus No CXL
References No. Treated Eyes Mean Age (y) Mean Follow-up (mo) % of Stabilization (Improvement of Kmax/Kmean) Mean Change in Kmax (D) Mean Change in SE (D) Mean Change in UCVA Mean Change in CVA CCT (μm) Permanent Complications Crosslinking Device
Wittig-Silva et al18 33 26.9 12 100 (>50) 1.45* 0.12 logMAR* >400 None UV-X IROC
Hersh et al19 49 12 90 (51) 2.0* 0.85 0.05 logMAR 0.13 logMAR* >400 (Hypo-osmolar solution if <400) UV-X IROC
O’Brart et al20 22 29.6 18 100 (23) 0.62* 0.82 0.06 Snellen decimal equivalent 0.12 Snellen decimal equivalent* >400 None CMB Vega X-linker & Roithner Lasertechnik
Wittig-Silva et al21 46 25.6 36 98 (—) 1.03* 0.61 0.15 logMAR* 0.09 logMAR* >400 1 eye corneal scar UV-X IROC
Lang et al22 15 28 36 73.3 (11) 0.35* 0.22 logMAR >400 None
Sharma et al23 23 19.7 6 1.2* 0.5 0.11 logMAR* No change >400 None UV-X IROC
Seyedian et al24 26 25.6 12 88.5 (3) 0.22* 0.54 0.13 logMAR* >400 None UV-X IROC
Hersh et al25 90 31 12 94 (84) 1.6* 0.1 4.4 letters 5.7 letters* >400 1 eye corneal scar KXL system
Meyer et al26 38 21.1 60 94.3 (49) 1.45* 0.1 0.13 logMAR* 0.04 logMAR >400 1 eye microbial keratitis and scarring
CCT indicates central corneal thickness; CVA, corrected visual acuity; CXL, crosslinking; SE, spherical equivalent; UCVA, uncorrected visual acuity.
*Significant result.

A significant improvement in mean CDVA within the treatment group was observed in 66% of studies following CXL. While there was a trend toward improving uncorrected visual acuity (UCVA) and spherical equivalent (SE), only 3 studies reported a significant increase in UCVA, and no study reporting statistically significant changes in SE.

Overall, there were few complications reported. Three studies reported persistent corneal haze lasting up to 60 months.22,25,26 Corneal scars were noted in 3 studies,21,25,26 and there were 2 cases of delayed epithelial healing,20,24 the worst case lasting 9 months.20 Two inflammatory responses with infiltrates were treated as microbial keratitis and fully resolved with appropriate treatment,18,21 and 1 study had permanent scarring from microbial keratitis.26

Meta-analyses by Kobashi and Rong27 and Li et al28 also provide high-quality evidence to show that CXL effectively halts the progression of keratoconus, on average reducing Kmax by 0.84 and 2.05 D and improving CDVA between −0.09 and −0.1 logMAR. It is important to note that there was little variation within these RCTs regarding CXL protocols, strengthening the quality of results. However, both reviews combine retrospective studies, thus utilizing data with substantial heterogeneity, including major biases in control group characteristics due to some studies using contralateral eyes for control versus other patients and different follow-up periods.


More recently, CXL techniques have departed from the Dresden protocol. In accelerated CXL, high-intensity UVA is used to shorten treatment duration, reduce patient discomfort, and subsequently minimize postoperative complications such as infection, haze, and persistent epithelial defects. while transepithelial CXL utilizes different formulations and delivery methods of riboflavin to avoid epithelial debridement.1

Accelerated Protocols

Accelerated CXL (ACXL) reduces treatment time by utilizing the Bunsen-Roscoe law of photochemical reciprocity, which states that the same photochemical effect can be achieved with a reduced irradiation interval, provided that total energy is kept constant through a corresponding increase in irradiation intensity.1 In accelerated protocols, high-energy settings of up to 43 mW/cm2 are applied, reducing the irradiation time to as little as 2 minutes. The limit of total dosage is therefore obeyed to minimize the risk of complications which have been theorized to occur with a fluence of more than 5.4 J/cm2.11

A recent meta-analysis performed by Kobashi and Tsubota29 examined 6 RCTs totaling 379 eyes, comparing conventional and accelerated protocols. They concluded that ACXL and conventional CXL provided comparable results and safety profiles in halting keratoconus progression with 12-month follow-up. The conventional CXL group significantly improved in CDVA compared with the ACXL group; however, the change in UCVA was not significantly different between the 2 groups. ACXL provided a greater reduction in cylinder, but there were no differences in changes in Kmax, central corneal thickness (CCT) UCVA, SE, corneal biomechanical properties, or corneal endothelial cell density between the 2 protocols. Both techniques were safe, with no reports of permanent complications. Although conventional CXL showed superiority over ACXL with better CDVA at 12 months, the difference was less than 1 visual acuity line. Therefore, it is debatable whether this statistically significant difference has a clinically meaningful impact. It is important to note that there was substantial heterogeneity between RCTs due to differences in baseline patient characteristics, such as the mean baseline Kmax and corneal thickness, which lessen the quality of the review.

Moreover, while this meta-analysis showed no difference in keratoconus progression and visual outcomes in the first year, in a study by Iqbal et al,30 no child with conventional CXL progressed during the second year of treatment, while 5.4% of subjects who had ACXL showed signs of progression. At 2 years, 70.3% of patients treated with conventional CXL experienced a reduction of Kmax >1 D, and 29.7% showed stability. In contrast, there were insignificant improvements in Kmax in the ACXL group, and stability was achieved in only 30.4% of eyes.

These shortcomings of ACXL could be attributable to multiple factors. First, there is evidence that the demarcation line in ACXL is shallower than in conventional CXL, possibly due to shorter soaking time.30 We know that corneal biomechanical efficacy has a relationship with the depth of the demarcation line, so this could translate into reduced efficacy.30 Moreover, high irradiance may increase oxygen consumption, affecting efficacy.31.

However, the potential benefits of conventional CXL need to be compared with the overall need for CXL in the population and in the context of health economics. Given that the procedure takes 2 to 3 times as long as accelerated protocols and the significant burden of care, it seems logical to assume that accelerated protocols would be a more cost-effective way of delivering care. A health economics study is eagerly awaited.2

Pulsed Crosslinking

Pulsing the UVA light during crosslinking theoretically restarts the photodynamic type 2 reaction by allowing oxygen replenishment of stromal tissue, allowing more singlet oxygen release and enhanced crosslinking of collagen molecules. Pulsing the UVA irradiation can theoretically result in fewer side effects at more intense irradiation levels while facilitating a more effective crosslinking process.32 Ziaei et al32 examined the outcomes of various crosslinking protocols with 24-month follow-up in 80 eyes. Continuous ACXL (c-ACXL) appeared to offer superior refractive and tomographic outcomes when compared with pulsed ACXL (p-ACXL), but this did not translate into better visual outcomes.32 Kang et al33 and Mazzotta et al34 also reported no difference in visual acuity between c-ACXL and p-ACXL at 12-month follow-up, however, suggested that p-ACXL resulted in more flattening of keratometric parameters and a deeper demarcation line.

In recent years, the validity of the Bunsen-Roscoe law of reciprocity in CXL has been challenged with evidence that there is a reduction in crosslinking efficacy at higher UVA intensity treatment protocols (around 45 mW/cm2). Furthermore, the duration of p-ACXL needs to be re-examined, as oxygen depletion occurs after 15 seconds of UVA initiation, but replenishment of oxygen only occurs 3 minutes after UV discontinuation.32 Therefore, the efficacy of short pulsing duration is questionable, and the protocol needs to be further optimized. Finally, the doubling of the time required to complete p-ACXL compared with c-ACXL with no apparent clinical benefit reported should be carefully considered in countries with a high prevalence of keratoconus and a significant burden of care.

Transepithelial Crosslinking

Epithelial debridement is a major contributor to postoperative complications such as infective keratitis and abnormal wound-healing.35 This has perpetuated interest in developing an epithelium-on or transepithelial crosslinking technique. The main challenge associated with this technique is the limited diffusion of the riboflavin molecule through the lipophilic cornea and epithelial tight junctions.1 The diffusion of riboflavin can be achieved by several techniques, including modifying corneal epithelial permeability, changing the physicochemical properties of the riboflavin molecule, and direct delivery of the riboflavin molecule into the corneal stroma by the creation of an epithelial flap or pocket.36,37

To date, 15 RCTs have reported on the clinical outcomes of epithelium-off CXL versus transepithelial CXL in the treatment of keratoconus. Two Cochrane systematic reviews have reported the combined outcomes of these studies.38,39 The most recent systematic review by Ng et al38 examined 13 RCTs totaling 661 eyes, of which 345 eyes underwent transepithelial CXL and 316 underwent epithelium-off CXL. The transepithelial studies had considerable heterogeneity in participant characteristics, topographic/tomographic devices, and surgical techniques. The epithelium-off studies were more uniform, most followed the Dresden protocol, but there was heterogeneity in crosslinking devices and the type of riboflavin used. Postoperative topical steroid drop regimens also varied. This review showed no evidence of a significant difference between interventions concerning mean change in Kmax from baseline or measurements of Kmax at any time point following treatment. There was some evidence suggesting that epithelium-off CXL may have a more stabilizing effect when compared with transepithelial CXL. However, the large amount of heterogeneity between the studies made the certainty of evidence low. Differences in CDVA were only a few letters which were not significant between groups. Adverse effects were reported in only a few studies, which showed moderate evidence for increased corneal haze and scarring in subjects undergoing epithelium-off crosslinking (14 vs 0 patients, respectively).

In contrast, another meta-analysis and systematic review of 12 RCTs (966 eyes) looking at transepithelial versus epithelium-off CXL was performed recently by Nath et al39 and reported inferior efficacy for transepithelial CXL. There was increased disease progression noted with transepithelial CXL at the 12-month follow-up point, with 2% progression in the epithelium-off group and 7% in the transepithelial group. In addition, transepithelial CXL did not appear to “catch up” with conventional CXL either, with the gap in Kmax reduction between the 2 protocols increasing over time. The meta-analysis reported superior corneal flattening and UCVA with conventional CXL, but this did not translate into better CDVA. However, transepithelial CXL was associated with significantly fewer complications than conventional epithelium-off CXL (2% vs 4%, respectively).

Despite these reviews being the most comprehensive and accurate meta-analyses to date, limitations still exist. The analyzed transepithelial techniques described in these studies vary widely, and blinding of subjects and surgeons performing the procedure was not possible due to the visible differences between the 2 techniques and obvious differences in postoperative recovery.

Transepithelial Iontophoresis-Assisted Crosslinking

Iontophoresis utilizes repulsive electromotive forces created by a small electrical current applied to an iontophoretic chamber to deliver riboflavin into the corneal stroma without removing the corneal epithelium.40 The primary mechanism of action is thought to be electromigration. Iontophoresis CXL (I-CXL) requires a specifically formulated riboflavin solution consisting of 0.1% riboflavin, no dextran or sodium chloride, but with possible different enhancers (such as benzalkonium chloride, high concentration sodium chloride or sodium ethylene diamine tetra acetic acid) to facilitate penetration across the corneal epithelium.41 A corneal ring electrode is placed onto the cornea, filled with 0.5 mL of riboflavin solution, and then connected to a constant current generator, which transports the riboflavin through the epithelium over 5 minutes. Finally, the cornea is irradiated with UVA according to standard ACXL protocols. A meta-analysis of 586 eyes, including 3 RCTs and 7 comparative studies, demonstrated equivalent efficacy to conventional CXL, with similar changes in uncorrected/corrected distance VA, Kmean, Kmax, central corneal thickness, higher-order aberration, spherical aberration, coma, stromal keratocyte density, and demarcation line depth.42 However, I-CXL had a threefold reduction in complications such as corneal edema and haze compared with conventional CXL. The long-term efficacy of I-CXL is still debatable as Vinciguerra et al43 demonstrated progression in 5 (26%) of 19 treated eyes over 7 years of follow-up.

A modified technique called enhanced fluence pulsed light iontophoresis (EF I-CXL) has been investigated to improve outcomes above standard I-CXL. EF I-CXL has 2 modifications to the standard procedure: UVA dose is increased to 7 J/cm2 to compensate for the epithelial photoattenuation, and UVA irradiation is pulsed to improve oxygen replenishment.44 A pilot investigation reported promising results at 1 and 3 years with improvement in CDVA, apical flattening, and no complications such as haze or endothelial damage.44

Ultimately, the overall efficacy of I-CXL should be interpreted with caution as there are significant clinical and methodological heterogeneities between the comparative studies.42 Moreover, the RCTs consisted of different riboflavin enhancers, and 1 RCT used a different iontophoresis voltage and application time. These discrepancies, in addition to small sample sizes, necessitate a careful evaluation of the data before widespread adoption of this technique can be considered. RCTs with larger sample sizes, a standardized UVA fluence and riboflavin formulation, and longer follow-up periods are required to determine the long-term effectiveness of I-CXL.


Pediatric patients are often affected by more severe keratoconus due to the differences in corneal microstructure and behavioral differences compared with adults.45 Furthermore, children often have more advanced disease and a higher risk for complications due to noncompliance and increased rates of ocular comorbidity.45

Tuft et al46 retrospectively examined 2723 subjects with keratoconus and found that subjects below 18 years of age progressed to transplantation more quickly than adults. Léoni-Mesplié et al47 found that of 216 subjects diagnosed with keratoconus, 22.7% of subjects were 15 years or under, and 27.8% had advanced disease compared with only 7.8% of adults. Larkin et al48 found that, on average, subjects undergoing CXL had a 90% reduced risk of progression compared with conservative management. However, a systematic review of 1508 pediatric subjects by Achiron et al49 reported that on average, 19.9% of subjects progressed despite undergoing epithelium-off CXL.

Early diagnosis and treatment are imperative given the increased prevalence of severe keratoconus and the more aggressive disease progression pattern in the pediatric population. To date, 2 meta-analyses have reported CXL outcomes in the pediatric population. Fard et al45 examined 28 reports totaling 1300 eyes, and Kobashi et al50 examined 26 studies totaling 1718 eyes. Both reported that CXL is effective in halting the progression of keratoconus in the pediatric population. There were some discrepancies regarding the best protocol, with some studies proposing ACXL and others conventional CXL yielding the best results for CDVA. Kmax improved significantly in the conventional CXL group in both meta-analyses, however, Fard and colleagues reported a significant increase in Kmax in the ACXL group. Both reviews concluded that transepithelial CXL was less efficient, yielded less desirable results, and showed no improvement in keratometric parameters and no statistically significant change in CDVA. These meta-analyses should be interpreted with caution as there is significant heterogeneity of studies, making them difficult to compare, with a distinct lack of RCTs.


The reported rate of keratoconus in Trisomy 21 (Down syndrome) has been as high as 39.8%, which is much higher than the general population.51 Çakmak et al52 assessed 2025 eyes undergoing CXL and found that 2.3% had Trisomy 21. They also found a statistically significant relationship between Trisomy 21 and postoperative complications such as corneal haze and sterile infiltrates.

Hashemi et al53 completed an RCT of conventional CXL versus ACXL in subjects with Trisomy 21. The group studied 27 subjects under 20 years of age and showed that both conventional CXL and ACXL procedures effectively halted disease progression. However, in more advanced cases of keratoconus, the efficacy of ACXL seems to be delayed until the second year. Both groups improved on average by 1 Snellen line of CDVA by the second year. Although these are promising results, given the small number of studies examining CXL in the Trisomy 21 population, more clinical trials need to be performed and the efficacy of a transepithelial technique better examined. This will allow clinicians to consider performing bilateral sequential CXL, which obviates the need for a second general anesthetic, often required in this patient population.


In 2014, the barrier of crosslinking thin corneas was challeneged by introducing contact lens-assisted CXL (CACXL), which artificially “thickens” the cornea by approximately 100 μm through the use of a presoaked UV barrier-free soft contact lens.54 A review by Srivatsa et al54 highlights the safety profile in 4 studies totaling 69 eyes with up to 1-year follow-up, showing no difference in demarcation line depth and endothelial cell density to other CXL protocols. In a study evaluating 2-year follow-up, significant changes were observed in the CACXL groups in Kmax, simulated keratometry and SE, with larger reduction in Kmax compared with standard epi-off CXL, although this could be attributable to higher baseline Kmax in the CACXL group.55 No complications were observed.

Furthermore, the development of SMILE lenticule-assisted CXL has also endeavored to increase corneal thickness by introducing a donor lenticule over the cornea before irradiation. Typically, the thickest part of the lenticule is placed over the thinnest area of cornea. A study of 10 eyes undergoing epi-off-lenticule-on CXL showed that CDVA and keratometry values were stable after 12 months.56 There are limitations to this technique, such as unpredictability of donor thickness and riboflavin absorption capacity, dependency on human donor tissue and the associated need for serological testing and storage facilities.

These are promising techniques which require further assessment with longer-term follow-up and larger patient cohorts but suggest that even thin corneas which have traditionally been seen to be unfit for CXL, may be suitable for treatment using modern protocols.


One of the barriers to CXL is its requirement to be performed in a surgical theater or procedure room. Given that the procedure duration is short, there would be significant benefits to completing this procedure in an aseptic manner within a clinic room at the slitlamp. Goh et al2 showed that 39.6% of keratoconic eyes progressed while awaiting CXL, with a mean wait time of 153 days and procedure room availability cited as a potential barrier to early treatment.

Hafezi et al57 recently described a technique in which a slitlamp-mounted UVA device (C-Eye; EMAGine AG, Switzerland) is used to conduct ACXL at 9 mW/cm². The biggest theoretical concerns of CXL at the slitlamp are whether there are any gravitational effects on riboflavin distribution while performing CXL, whether the subjects will tolerate the upright position for the duration of treatment, and whether there is an increased incidence of infectious keratitis performing CXL outside of the sterile operating theater. A study by Salmon et al58 on porcine corneas reported that gravitational influence on the riboflavin distribution was observed only after 60 minutes of vertical positioning.

Moreover, the CXL procedure creates ROS and disables pathogen replication by oxidizing DNA when undergoing photoactivation.59 These 2 processes ultimately create an aseptic treatment; thus, performing the procedure at the slitlamp is theorized to not lead to significantly higher rates of infectious keratitis. Finally, introducing an office-based CXL protocol could reignite the more widespread completion of the CXL procedure by allied health care professionals, reducing patient wait times and associated procedural costs.


Although CXL has been shown to be effective in halting the progression of keratoconus, the visual outcomes are not as predictable. Whilst many studies have demonstrated a modest improvement in visual acuity and topographic parameters, most subjects require ongoing use of contact lenses or spectacles for optical correction.1 The concept of “CXL Plus” or combined treatments was born to simultaneously address disease progression and poor visual acuity.

Athens Protocol

Krueger and Kanellopoulos60 first reported the Athens protocol. The protocol involves manual epithelial debridement, partial correction of refractive error aiming to correct ~70% of cylinder and sphere with a maximum ablation depth of 50 μm using topography-guided photorefractive keratectomy (PRK) followed by 0.02% of mitomycin C application and subsequent conventional CXL.

The long-term clinical outcomes of the Athens protocol versus CXL treatment alone have been reported with up to 3 years of follow-up.61 This protocol has reported a significant improvement in UCVA and CDVA in the PRK-CXL group, and keratometry in both PRK-CXL and CXL alone, but corneal flattening was more prominent in the PRK-CXL group.61 The effect of simultaneous versus sequential treatment has also been studied by Kanellopoulos,62 with the simultaneous group demonstrating significantly greater improvements in both UDVA and CDVA, and a greater mean reduction in SE and keratometry with less reported corneal haze.

The Athens protocol has yielded promising results; however, studies are small and typically single center in nature, therefore, further prospective multicenter studies are required to corroborate the reported outcomes.

Cretan Protocol

The Cretan protocol was developed by Kymionis et al63 and constituted a technique alternative to the mechanical debridement of the epithelium. Instead, the epithelium is removed by transepithelial phototherapeutic keratectomy (t-PTK) ablation at an intended depth of 50 μm in a 7.0-mm zone. The de-epithelialized area is then enlarged by mechanical debridement until the targeted diameter of 8.0 mm is achieved, followed by conventional CXL.63 In a study examining 38 eyes, t-PTK-treated eyes had significant improvement in visual acuity and keratometric outcomes compared with conventional CXL.63 This group also showed safety and efficacy at a 12-month follow-up, with only 9% of subjects losing 1 line of visual acuity, while 26% of subjects achieved UCVA of 6/24 or better.64

Other studies have since demonstrated that the Cretan protocol is superior in visual and keratometric outcomes to conventional CXL with a follow-up of 4 years.65 However, 1 study could not find significant differences between the 2 techniques after 2 years of follow-up.66 These studies show t-PTK to be superior to CXL alone; however, the results vary regarding follow-up time, and the small sample sizes preclude meaningful conclusions.

Cretan Protocol Plus

Cretan protocol plus is an extension of the Cretan protocol to include adjunctive PRK for supplemental refractive correction with a maximum ablation depth of 50 μm and a maximum optical zone of 5.5 mm followed by conventional CXL.67 In a study of 55 eyes, there was a significant improvement in UDVA, CDVA, keratometry, astigmatism, and SE.67 However, 2% of eyes lost 2 lines of CDVA, and 4 eyes developed clinically, but not visually significant, stromal haze.

Tel-Aviv Protocol

The Tel-Aviv protocol (ePRK-CXL) is a modification of the Cretan protocol and involves excimer laser epithelial removal followed by correction of 50% of the manifest refractive astigmatism (on the same axis) with the potential to add a spherical component for a maximum total ablation depth of 50 µm and subsequent ACXL.68

A retrospective study of 20 eyes reported a significant improvement in UCVA and reduction of Kmax by 2.21 D with less tissue ablation than the Athens protocol (46 vs 70 μm).68 A more extensive retrospective study conducted to compare ePRK-CXL with conventional CXL, totaling 131 eyes with a 12-month follow-up, found no significant difference in corneal thickness. However, the results concurred that ePRK-CXL resulted in a greater improvement in UCVA, CDVA, and SE compared with conventional CXL.69 In addition, ePRK-CXL subjects achieved better visual outcomes compared with the conventional CXL patients.

Combined treatments offer great potential for keratoconic subjects that desire a stabilization of their condition and improvements in UCVA. However, the techniques are limited by the requirement for expensive and sophisticated laser systems. The thin nature of the cornea in many patients precludes them from safe treatment and the potential for developing postoperative complications such as corneal haze. An RCT of various techniques compared with conventional treatment and a cost-effectiveness study is required before combined treatments can become mainstream.


Over the last decade, the clinical indications of CXL have expanded considerably. Currently, limited studies are looking into the application of CXL in treating and managing patients with bacterial keratitis, bullous keratopathy (BK), progressive myopia, and surgical patients undergoing refractive surgery and keratoplasty.

Bacterial Keratitis

Infectious keratitis is the leading cause of corneal blindness in the world.70 The concept of photoactivated chromophore for infectious keratitis (PACK-CXL) was conceived to address this growing need. PACK-CXL has been suggested to provide added resistance to enzymes produced by bacterial and fungal pathogens.

A meta-analysis by Ting et al70 examined 46 articles, of which 4 were RCTs, totaling 435 subjects. They compared standard antimicrobial treatment alone or with PACK-CXL as adjuvant therapy. They concluded that the average healing time was 7 days less for subjects treated with adjuvant PACK-CXL. However, there was no significant difference between the epithelial defect size, rates of adverse events, and CDVA at final follow-up. A Cochrane systematic review of 3 trials totaling 59 eyes reported no real benefit with PACK-CXL compared with standard antibiotic therapy.59 It concluded that there was low-quality evidence on the effectiveness of PACK-CXL for bacterial keratitis due to the clinical heterogeneity of outcomes and small sample sizes.

However, more recently, a multicenter phase III RCT of 18 eyes undergoing PACK-CXL versus 21 eyes with antimicrobial therapy alone showed that success rates were 88.9% in the PACK-CXL group and 90.5% in the antimicrobial therapy group.71 They found no significant difference in time to complete corneal re-epithelialisation or CDVA between the groups. These are encouraging results which suggest PACK-CXL may be a suitable first-line treatment alternative for infectious keratitis or an option for drug-resistant micro-organisms.

Bullous Keratopathy

There is some research examining the use of CXL in BK with limited evidence to show that ACXL may improve pain and corneal edema in subjects with BK. However, the effects are only transient, with recurrence of edema and bullae formation within a few months.72 In a randomized study by Choy and colleagues, CXL reduced corneal thickness in subjects with BK, at least for the first month. However, there were no significant improvements in pain, corneal clarity, and vision.73 The short-term benefit was also unlikely to outweigh the potential risk of recurrent epithelial defect (12%), with two-thirds of the subjects who developed an epithelial defect requiring amniotic membrane grafting.

Myopia Progression

Pathological myopia is a significant public health burden with a prevalence of up to 3.1%.74 Scleral collagen crosslinking (SXL) has the potential to stabilize myopia progression by increasing scleral thickness and preventing aberrant scleral remodeling. Human scleral crosslinking in vitro has been shown to increase Young modulus (a stiffness parameter) by up to 201%, but this process is location specific due to the variability in scleral thickness.75

SXL can be performed utilizing physical and chemical methods. The physical method involves the use of riboflavin with UVA or blue light. Although this method has shown promising increases in Young modulus, adverse damage to the retina is observed in animal studies.74 Chemical SXL creates new covalent bonds through chemical reagents containing reactive groups (amino, thiol) reacting with scleral proteins.76 Animal experiments suggest that the SXL procedure may potentially control the pathologic progression of myopia. However, such treatments’ safety and procedural protocol require further investigation and refinement before they can be utilized in human studies.

Refractive Surgery

CXL has been used to prevent postrefractive surgery ectasia and regression in subjects undergoing laser refractive surgeries such as laser in situ keratomileusis (LASIK), small incision lenticule extraction (SMILE), and PRK. The combined treatments are termed LASIK Xtra, SMILE Xtra, and PRK Xtra, respectively. Kankariya et al77 examined 11 studies totaling 917 eyes undergoing LASIK Xtra, SMILE Xtra, and PRK Xtra. They concluded that combination surgeries did not show any additional refractive advantage. However, the studies suffered from high heterogeneity due to varying protocol and baseline patient characteristics, low sample sizes, and short follow-up periods.

Another review by Lim and Lim78 examining 10 studies totaling 1189 eyes concluded that simultaneous ACXL and refractive surgery are effective for treating myopia but reiterated that it is unclear if the additional CXL step reduces the incidence of iatrogenic ectasia. A recent retrospective study of 48 eyes with 2-year follow-up showed good stability with only 8.3% of eyes changing by 0.5 D or more, and 81% of eyes achieving vision of 6/7.5 or better, with no eyes losing lines of CDVA.79 Despite promising results, longer-term follow-up is required to appropriately assess the effect on postoperative ectasia.

Use in Keratoplasty

Mukherjee proposed a method to reduce postkeratoplasty ectasia by crosslinking the donor corneal tissue before transplantation.80 In an in vitro randomized study, the use of crosslinked porcine donor tissue significantly reduced postoperative mean root-mean-square wavefront aberration and mean keratometric astigmatism by an average of 1.26 μm and 4.76 D, respectively. Huang et al81 then progressed this proposal to human trials with 116 subjects randomly allocated to receiving untreated or CXL donor corneas. At 3-year follow-up, Kmax was significantly less in the CXL group by an average of 2.52 D, and keratometric astigmatism was significantly less by an average of 1.45 D. The CXL group also achieved significantly improved UCVA and CDVA, with 39.7% of eyes reaching vision better than 6/7.5 comapred with 20.7% in the untreated group.

Given the biomechanical changes that CXL provides, Ziaei et al82 hypothesized that CXL of the corneal periphery 3 months before corneal transplantation could also reduce the incidence of recurrent ectasia. In a case report of a 22-year-old female with advanced keratoconus, they were successfully able to show the safety and feasibility of CXL before keratoplasty. The protocol involved ACXL (30 mW/cm2 for 3 min at 5.4 J/cm2) of peripheral corneal tissue (6.5–9.5 mm), sparing the central cornea and limbus. Repopulation of the peripheral cornea with keratocytes with no significant endothelial cell loss and a routine postoperative course over 12 months was reported following deep anterior lamellar keratoplasty.


Noninvasive Crosslinking

One of the biggest concerns with CXL is associated with epithelial removal. Schaeffer et al83 proposed a completely noninvasive technique in which high doses of oral riboflavin, ranging from 800 mg to 2 g daily, are ingested with 15 minutes of UV exposure from natural sunlight per day of the study, totalling 6 months. This 3-case series described promising results with an average reduction of 1.075 D Kmax with subjective and objective improvements in CDVA and no reported adverse effects. Further trials are underway and could lead to a novel and inexpensive medical management option.

Customized CXL Based on Topography

Customized treatment patterns localized on specific corneal zones have been suggested to aid the efficacy of CXL treatment in paracentral and peripheral cones. One such method is customized topography-guided treatment consisting of modifying irradiation to 3 concentric circular zones centered on maximum posterior elevation to mimic a graded treatment pattern.84 This technique has been shown to significantly improve the regularization of the cornea while yielding similar corneal flattening compared with standard CXL.84 However, limitations of this method include the inability to treat cases with no discernible posterior cone or smaller diameter treatments. An “enhanced Athens protocol” has also been presented, in which topography-guided PRK is used in conjunction with a customized, variable-pattern CXL technique, which shows improvements in UCVA and keratometry values that remain stable after 3 years.85

Sub400 Protocol

The Sub400 protocol is a technique developed by Hafezi et al86 that allows for conventional CXL equipment on subjects with corneal thicknesses of less than 400 μm. This protocol involves adjusting the UVA illumination time and irradiance to achieve safer crosslinking depths while using the same materials and equipment as standard CXL. A pilot study of 39 eyes reported significant improvements in Kmax and stability at 12 months with no adverse reactions and stable endothelial cell counts.86

Oxygen Supplementation

Oxygen supplementation of the cornea has been shown to improve CXL results by increasing the strength and depth of CXL.31 This is achieved through a wearable Boost mask (Avedro, MA), which creates a hyperoxic periocular environment. Larger reductions in Kmax with a significantly deeper demarcation line when compared with control subjects undergoing conventional treatment have been reported.87

Antimicrobial Photodynamic Therapy

Antimicrobial photodynamic therapy (aPDT) utilizes the activation of a photosensitizer in the presence of oxygen to kill micro-organisms, the same technique used in PACK-CXL. Several advantages include local treatment without systemic effects, no reports of resistance yet, and selective binding to micro-organisms rather than human cells.88 Photosensitizers that could be used in aPDT for infectious keratitis include toluidine blue O, methylene blue, chlorin, and hematoporphyrin, which show good selective action against multiple organisms in vitro.88 Rose bengal is the only other photosensitizer, other than riboflavin, to have been evaluated in vivo, with a case series evaluating 18 patients (0.1% or 0.2% of rose bengal solution applied over 30 min, then irradiated with 6 mW/cm2 green LED light for 15 min for a total energy exposure of 5.4 J/cm2). This led to clinical resolution of infection in 72% of patients. Unfortunately, the other 28% progressed to corneal perforation and required therapeutic penetrating keratoplasty.89 Rose bengal has also been shown to increase corneal stiffness in rabbit eyes in vitro, with shallower demarcation lines, and may become an alternative for riboflavin CXL especially in thin corneas.90 The main disadvantage of aPDT appears to be the penetration depth of the photosensitizer into the cornea, but certain strategies, including the use of penetrating-enhancing compounds, microemulsions, and the incorporation into nanoparticles, could improve ocular drug delivery. aPDT is certainly an intriguing concept and requires further evaluation in vitro and in vivo.

UVA Emitting Devices

CXL devices are large and expensive and require the patient to be at a specific location and stationary during the procedure. KeraVio is a new portable CXL treatment modality in which UVA emitting spectacles are used along with self-administration of topical transepithelial riboflavin for 3 hours daily over 6 months. A pilot study of 40 eyes showed promising results with a significant reduction in Kmax and a good safety profile.91

A novel CXLens on-eye UVA light-emitting contact lens device (TECLens LLC, Stamford, CT) was also developed. A pilot study of 9 patients with severe keratoconus used this fiber optic-based UVA light-emitting scleral contact lens device with riboflavin application, achieved through a scleral reservoir lens, and reported a significant reduction in Kmax.92

These new devices could allow for the crosslinking procedure to be done outside the clinical setting without epithelial debridement. Moreover, procedures can be performed bilaterally at the same time. Although promising, these devices are not commercially available, and clinical studies comparing their utility to established CXL protocols are awaited.

Artificial Intelligence

Chen et al93 considered using convolutional neural networks technique to create an artificial intelligence (AI) modality of keratoconus screening. This multicenter study using color-coded maps of axial, anterior, and posterior elevation and pachymetry allowed for development of a convolutional neural networks model that could detect keratoconus versus healthy eyes with an accuracy of 0.9785. This model could also stage keratoconus according to the Amsler-Krumeich scale between accuracies of 0.8537 and 0.9032. Shetty et al94 have also considered the development of AI, which detects keratoconus progression by increasing Kmax values. There was high sensitivity in detecting no progression, however, progression was predicted correctly in only 61.2% of cases.

These are both promising studies and with further developments in these AI techniques, sensitivity and specificity can be increased, allowing for improved detection of disease progression. These possible screening modalities will substantially benefit the ongoing management of keratoconus through earlier diagnosis and more timely CXL treatment.


Keratoconus is a progressive corneal ectasia with a significant health burden worldwide. In the last 2 decades, keratoconus management and treatment have rapidly evolved. Timely treatment with CXL can prevent visual loss and is reported to reduce the need for corneal transplantation by more than half.2,5

Many modifications of the original Dresden technique have been developed to enhance postoperative outcomes and minimize potential complications. Accelerated protocols have shown comparative results to standard CXL while providing increased benefits such as reduced procedural time, increased efficiency for health care systems, and reduced associated costs. Despite few studies, immediate sequential bilateral CXL has shown to have no increased risk than delayed sequential CXL, further reducing the burden on the health care system.

Furthermore, combined protocols have allowed for simultaneous addressing of disease progression and visual rehabilitation with the recent development of more novel techniques such as the STARE-X protocol.95 These combined procedures reduce the rehabilitation burden by improving postoperative UCVA and CDVA but unfortunately are limited by equipment cost.

CXL as a treatment entity has also expanded from its initial indication of dealing with ectasia, showing benefit in treating infective keratitis and BK. PACK-CXL has shown promising results in improving outcomes of microbial keratitis when used as an adjunct to antimicrobial therapy, reducing the risk of perforation and the requirement of therapeutic keratoplasty. The possibility of PDT as an independent treatment could also reduce the developing issue of antimicrobial resistance. There is also substantial research looking into alternative uses, such as scleral crosslinking in myopia management and as an adjuvant to refractive surgery, reducing the incidence of postoperative ectasia.

The future of CXL seems to be at our doorstep with advances in customized CXL based on individual topography, specifically targeting the biomechanically compromised areas of the cornea. Ongoing advancements have improved safety and efficacy, with developments reducing procedural time and cost, including the possibility of slitlamp CXL and noninvasive techniques, including oral delivery of riboflavin. The field is ever-evolving, with a rapid accumulation of information over the last 2 decades and various CXL protocols available to clinicians. We postulate that the field of CXL will continue to evolve in the next decade. We eagerly await new developments, which will emphasize further customization of treatments, ongoing safety improvements, simultaneous stabilization of ectasia, reduction of refractive error, and improvement of visual outcomes.


1. Ziaei M, Barsam A, Shamie N, et al. Reshaping procedures for the surgical management of corneal ectasia. J Cataract Refract Surg. 2015;41:842–872.
2. Goh YW, Gokul A, Yadegarfar ME, et al. Prospective clinical study of keratoconus progression in patients awaiting corneal cross-linking. Cornea. 2020;39:1256–1260.
3. Crawford AZ, Zhang J, Gokul A, et al. The enigma of environmental factors in keratoconus. Asia Pac J Ophthalmol (Phila). 2020;9:549–556.
4. Althomali TA, Al-Qurashi IM, Al-Thagafi SM, et al. Prevalence of keratoconus among patients seeking laser vision correction in Taif area of Saudi Arabia. Saudi J Ophthalmol. 2018;32:114–118.
5. McGhee CNJ, Kim BZ, Wilson PJ. Contemporary treatment paradigms in keratoconus. Cornea. 2015;34:16–23.
6. Kim BZ, Meyer JJ, Brookes NH, et al. New Zealand trends in corneal transplantation over the 25 years 1991–2015. Br J Ophthalmol. 2017;101:834–838.
7. Soiberman U, Foster JW, Jun AS, et al. Pathophysiology of keratoconus: what do we know today. Open Ophthalmol J. 2017;11:252.
8. Chan C. Corneal cross-linking for keratoconus: current knowledge and practice and future trends. Asia Pac J Ophthalmol (Phila). 2020;9:557–564.
9. Pagano L, Gadhvi KA, Borroni D, et al. Bilateral keratoconus progression: immediate versus delayed sequential bilateral corneal cross-linking. J Refract Surg. 2020;36:552–556.
10. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-a-induced collagen cross-linking for the treatment of keratoconus. Am J Ophthalmol. 2003;135:620–627.
11. Wollensak G. Corneal collagen cross-linking: new horizons. Expert Rev Ophthalmol. 2010;5:201–215.
12. Spoerl E, Huhle M, Seiler T. Induction of cross-links in corneal tissue. Exp Eye Res. 1998;66:97–103.
13. Santhiago MR, Randleman JB. The biology of corneal cross-linking derived from ultraviolet light and riboflavin. Exp Eye Res. 2021;202:108355.
14. Ziaei M, Yoon JJ, Vellara HR, et al. Prospective one year study of corneal biomechanical changes following high intensity, accelerated cornea cross-linking in patients with keratoconus using a non-contact tonometer. Eur J Ophthalmol. 2022;32:806–814.
15. Ziaei M, Gokul A, Vellara H, et al. Prospective two year study of changes in corneal density following transepithelial pulsed, epithelium-off continuous and epithelium-off pulsed, corneal crosslinking for keratoconus. Cont Lens Anterior Eye. 2020;43:458–464.
16. Jordan C, Patel DV, Abeysekera N, et al. In vivo confocal microscopy analyses of corneal microstructural changes in a prospective study of collagen cross-linking in keratoconus. Ophthalmology. 2014;121:469–474.
17. Raiskup F, Theuring A, Pillunat LE, et al. Corneal collagen cross-linking with riboflavin and ultraviolet-A light in progressive keratoconus: ten-year results. J Cataract Refract Surg. 2015;41:41–46.
18. Wittig-Silva C, Whiting M, Lamoureux E, et al. A randomized controlled trial of corneal collagen cross-linking in progressive keratoconus: preliminary results. J Refract Surg. 2008;24:720.
19. Hersh PS, Greenstein SA, Fry KL. Corneal collagen cross-linking for keratoconus and corneal ectasia: one-year results. J Cataract Refract Surg. 2011;37:149–160.
20. O’Brart DPS, Chan E, Samaras K, et al. A randomised, prospective study to investigate the efficacy of riboflavin/ultraviolet A (370 nm) corneal collagen cross-linkage to halt the progression of keratoconus. Br J Ophthalmol. 2011;95:1519–1524.
21. Wittig-Silva C, Chan E, Islam FMA, et al. A randomized, controlled trial of corneal collagen cross-linking in progressive keratoconus: three-year results. Ophthalmology. 2014;121:812–821.
22. Lang SJ, Messmer EM, Geerling G, et al. Prospective, randomized, double-blind trial to investigate the efficacy and safety of corneal cross-linking to halt the progression of keratoconus. BMC Ophthalmol. 2015;15:1–8.
23. Sharma N, Suri K, Sehra SV, et al. Collagen cross-linking in keratoconus in Asian eyes: visual, refractive and confocal microscopy outcomes in a prospective randomized controlled trial. Int Ophthalmol. 2015;35:827–832.
24. Seyedian MA, Aliakbari S, Miraftab M, et al. Corneal collagen cross-linking in the treatment of progressive keratoconus: a randomized controlled contralateral eye study. Middle East Afr J Ophthalmol. 2015;22:340.
25. Hersh PS, Stulting RD, Muller D, et al. United States multicenter clinical trial of corneal collagen crosslinking for keratoconus treatment. Ophthalmology. 2017;124:1259–1270.
26. Meyer JJ, Jordan CA, Patel DV, et al. Five year results of a prospective, randomised, contralateral eye trial of corneal cross-linking for keratoconus. Clin Exp Ophthalmol. 2021;49:542–549.
27. Kobashi H, Rong SS. Corneal collagen cross-linking for keratoconus: systematic review. Biomed Res Int. 2017;2017:8145651.
28. Li J, Ji P, Lin X. Efficacy of corneal collagen cross-linking for treatment of keratoconus: a meta-analysis of randomized controlled trials. PLoS One. 2015;10:e0127079.
29. Kobashi H, Tsubota K. Accelerated versus standard corneal cross-linking for progressive keratoconus: a meta-analysis of randomized controlled trials. Cornea. 2020;39:172–180.
30. Iqbal M, Elmassry A, Saad H, et al. Standard cross-linking protocol versus accelerated and transepithelial cross-linking protocols for treatment of paediatric keratoconus: a 2-year comparative study. Acta Ophthalmol. 2020;98:e352–e362.
31. Seiler TG, Komninou MA, Nambiar MH, et al. Oxygen kinetics during corneal cross-linking with and without supplementary oxygen. Am J Ophthalmol. 2021;223:368–376.
32. Ziaei M, Gokul A, Vellara H, et al. Prospective two-year study of clinical outcomes following epithelium-off pulsed versus continuous accelerated corneal cross-linking for keratoconus. Clin Exp Ophthalmol. 2019;47:980–986.
33. Mazzotta C, Traversi C, Paradiso AL, et al. Pulsed light accelerated crosslinking versus continuous light accelerated crosslinking: one-year results. J Ophthalmol. 2014;2014:604731.
34. Kang M, Hwang J, Chung S. Comparison of pulsed and continuous accelerated corneal cross-linking for keratoconus: one-year results at a single center. J Cataract Refract Surg. 2021;47:641–648.
35. Khoo P, Cabrera-Aguas M, Watson SL. Microbial keratitis after corneal collagen cross-linking for corneal ectasia. Asia Pac J Ophthalmol (Phila). 2021;10:355–359.
36. Borroni D, Bonzano C, Hristova R, et al. Epithelial flap corneal cross-linking. J Refract Surg. 2021;37:741–745.
37. Borroni D, Bonzano C, Hristova R, et al. A new surgical technique to deliver riboflavin beneath corneal epithelium: the corneal cross-linking epi-pocket. Asia Pac J Ophthalmol (Phila). 2021;10:495–498.
38. Ng SM, Hawkins BS, Kuo IC. Transepithelial versus epithelium-off corneal cross-linking for progressive keratoconus: findings from a Cochrane Systematic Review. Am J Ophthalmol. 2021;229:274–287.
39. Nath S, Shen C, Koziarz A, et al. Transepithelial versus epithelium-off corneal collagen cross-linking for corneal ectasia: a systematic review and meta-analysis. Ophthalmology. 2020;128:1150–1160.
40. Lombardo M, Serrao S, Lombardo G, et al. Two-year outcomes of a randomized controlled trial of transepithelial corneal crosslinking with iontophoresis for keratoconus. J Cataract Refract Surg. 2019;45:992–1000.
41. Vinciguerra P, Montericcio A, Catania F, et al. New perspectives in keratoconus treatment: an update on iontophoresis-assisted corneal collagen crosslinking. Int Ophthalmol. 2021;41:1909–1916.
42. Wan KH, Ip CK, Kua WN, et al. Transepithelial corneal collagen cross‐linking using iontophoresis versus the Dresden protocol in progressive keratoconus: a meta‐analysis. Clin Exp Ophthalmol. 2021;49:228–241.
43. Vinciguerra R, Legrottaglie EF, Tredici C, et al. Transepithelial iontophoresis-assisted cross linking for progressive keratoconus: up to 7 years of follow up. J Clin Med. 2022;11:678.
44. Mazzotta C, Bagaglia SA, Sgheri A, et al. Iontophoresis corneal cross-linking with enhanced fluence and pulsed uv-a light: 3-year clinical results. J Refract Surg. 2020;36:286–292.
45. Fard AM, Reynolds AL, Lillvis JH, et al. Corneal collagen cross-linking in pediatric keratoconus with three protocols: a systematic review and meta-analysis. J AAPOS. 2020;24:331–336.
46. Tuft SJ, Moodaley LC, Gregory WM, et al. Prognostic factors for the progression of keratoconus. Ophthalmology. 1994;101:439–447.
47. Léoni-Mesplié S, Mortemousque B, Touboul D, et al. Scalability and severity of keratoconus in children. Am J Ophthalmol. 2012;154:56–62.
48. Larkin DF, Chowdhury K, Burr JM, et al. Effect of corneal cross-linking versus standard care on keratoconus progression in young patients: the KERALINK randomized controlled trial. Ophthalmology. 2021;128:1516–1526.
49. Achiron A, El-Hadad O, Leadbetter D, et al. Progression of pediatric keratoconus after corneal cross-linking: a systematic review and pooled analysis. Cornea. 2021;41:874–878.
50. Kobashi H, Hieda O, Itoi M, et al. Corneal cross-linking for paediatric keratoconus: a systematic review and meta-analysis. J Clin Med. 2021;10:2626.
51. Mathan JJ, Gokul A, Simkin SK, et al. Topographic screening reveals keratoconus to be extremely common in Down syndrome. Clin Exp Ophthalmol. 2020;48:1160–1167.
52. Çakmak S, Sucu ME, Yildirim Y, et al. Complications of accelerated corneal collagen cross-linking: review of 2025 eyes. Int Ophthalmol. 2020;40:3269–3277.
53. Hashemi H, Amanzadeh K, Seyedian M, et al. Accelerated and standard corneal cross-linking protocols in patients with down syndrome: a non-inferiority contralateral randomized trial. Ophthalmol Ther. 2020;9:1011–1021.
54. Srivatsa S, Jacob S, Agarwal A. Contact lens assisted corneal cross linking in thin ectatic corneas—a review. Indian J Ophthalmol. 2020;68:2773.
55. Malhotra C, Gupta B, Jain AK, et al. Comparison of contact lens-assisted and transepithelial corneal crosslinking with standard epithelium-off crosslinking for progressive keratoconus: 24-month clinical results. J Cataract Refract Surg. 2022;48:199–207.
56. Cagini C, Riccitelli F, Messina M, et al. Epi-off-lenticule-on corneal collagen cross-linking in thin keratoconic corneas. Int Ophthalmol. 2020;40:3403–3412.
57. Hafezi F, Richoz O, Torres-Netto EA, et al. Corneal cross-linking at the slit lamp. J Refract Surg. 2021;37:78–82.
58. Salmon B, Richoz O, Tabibian D, et al. CXL at the slit lamp: no clinically relevant changes in corneal riboflavin distribution during upright uv irradiation. J Refract Surg. 2017;33:281.
59. Davis SA, Bovelle R, Han G, et al. Corneal collagen cross-linking for bacterial infectious keratitis. Cochrane Database Syst Rev. 2020;6:CD013001.
60. Krueger RR, Kanellopoulos AJ. Stability of simultaneous topography-guided photorefractive keratectomy and riboflavin/UVA cross-linking for progressive keratoconus: case reports. J Refract Surg. 2010;26:827–832.
61. Kontadakis GA, Kankariya VP, Tsoulnaras K, et al. Long-term comparison of simultaneous topography-guided photorefractive keratectomy followed by corneal cross-linking versus corneal cross-linking alone. Ophthalmology. 2016;123:974–983.
62. Kanellopoulos AJ. Comparison of sequential vs same-day simultaneous collagen cross-linking and topography-guided PRK for treatment of keratoconus. J Refract Surg. 2009;25:S812–S818.
63. Kymionis GD, Grentzelos MA, Kounis GA, et al. Combined transepithelial phototherapeutic keratectomy and corneal collagen cross-linking for progressive keratoconus. Ophthalmology. 2012;119:1777–1784.
64. Kymionis GD, Grentzelos MA, Kankariya VP, et al. Long-term results of combined transepithelial phototherapeutic keratectomy and corneal collagen cross-linking for keratoconus: Cretan protocol. J Cataract Refract Surg. 2014;40:1439–1445.
65. Grentzelos MA, Liakopoulos DA, Siganos CS, et al. Long-term comparison of combined t-PTK and CXL (Cretan protocol) versus CXL with mechanical epithelial debridement for keratoconus. J Refract Surg. 2019;35:650–655.
66. Gaster RN, Ben Margines J, Gaster DN, et al. Comparison of the effect of epithelial removal by transepithelial phototherapeutic keratectomy or manual debridement on cross-linking procedures for progressive keratoconus. J Refract Surg. 2016;32:699–704.
67. Grentzelos MA, Kounis GA, Diakonis VF, et al. Combined transepithelial phototherapeutic keratectomy and conventional photorefractive keratectomy followed simultaneously by corneal cross-linking for keratoconus: Cretan protocol plus. J Cataract Refract Surg. 2017;43:1257–1262.
68. Kaiserman I, Mimouni M, Rabina G. Corneal cross-linking for keratoconus: the Tel-Aviv protocol. J Refract Surg. 2019;35:377–382.
69. Rabina G, Mimouni M, Kaiserman I. Epithelial photorefractive keratectomy vs mechanical epithelial removal followed by corneal cross-linking for keratoconus: the Tel-Aviv Protocol. J Cataract Refract Surg. 2020;46:749–755.
70. Ting DSJ, Henein C, Said DG, et al. Photoactivated chromophore for infectious keratitis - Corneal cross-linking (PACK-CXL): a systematic review and meta-analysis. Ocul Surf. 2019;17:624–634.
71. Hafezi F, Hosny M, Shetty R, et al. PACK-CXL vs. antimicrobial therapy for bacterial, fungal, and mixed infectious keratitis: a prospective randomized phase 3 trial. J Eye Vis. 2022;9:1–11.
72. Khan MS, Basit I, Ishaq M, et al. Corneal collagen cross linking (cxl) in treatment of pseudophakic bullous keratopathy. Pak J Med Sci. 2016;32:965.
73. Choy BNK, Ng ALK, Zhu MM, et al. Randomized control trial on the effectiveness of collagen cross-linking on bullous keratopathy. Cornea. 2020;39:1341–1347.
74. Wang M, Zhang F, Liu K, et al. Safety evaluation of rabbit eyes on scleral collagen cross-linking by riboflavin and ultraviolet A. Clin Exp Ophthalmol. 2015;43:156–163.
75. Wang M, Zhang F, Qian X, et al. Regional biomechanical properties of human sclera after cross-linking by riboflavin/ultraviolet A. J Refract Surg. 2012;28:723–728.
76. Zhang F, Lai L. Advanced research in scleral cross-linking to prevent from progressive myopia. Asia Pac J Ophthalmol (Phila). 2021;10:161–166.
77. Kankariya VP, Dube AB, Grentzelos MA, et al. Corneal cross-linking (CXL) combined with refractive surgery for the comprehensive management of keratoconus: CXL plus. Indian J Ophthalmol. 2020;68:2757.
78. Lim EWL, Lim L. Review of laser vision correction (LASIK, PRK and SMILE) with simultaneous accelerated corneal crosslinking—long-term results. Curr Eye Res. 2019;44:1171–1180.
79. Sánchez-González J, Rocha-de-Lossada C, Borroni D, et al. Prophylactic corneal crosslinking in myopic small-incision lenticule extraction—long-term visual and refractive outcomes. Indian J Ophthalmol. 2022;1:73–78.
80. Mukherjee A, Hayes S, Aslanides I, et al. Donor cross-linking for keratoplasty: a laboratory evaluation. Graefes Arch Clin Exp Ophthalmol. 2015;253:2223–2228.
81. Huang T, Ye R, Ouyang C, et al. Use of donors predisposed by corneal collagen cross-linking in penetrating keratoplasty for treating patients with keratoconus. Am J Ophthalmol. 2017;184:115–120.
82. Ziaei M, Gokul A, Vellara H, et al. Peripheral cornea crosslinking before deep anterior lamellar keratoplasty. Med Hypothesis Discov Innov Ophthalmol. 2020;9:127.
83. Schaeffer K, Jarstad J, Schaeffer A, et al. Topographic corneal changes induced by oral riboflavin in the treatment of corneal ectasia. Invest Ophthalmol Vis Sci. 2018;59:1413.
84. Sachdev GS, Ramamurthy S, Soundariya B, et al. Comparative analysis of safety and efficacy of topography-guided customized cross-linking and standard cross-linking in the treatment of progressive keratoconus. Cornea. 2021;40:188–193.
85. Kanellopoulos AJ. Management of progressive keratoconus with partial topography-guided PRK combined with refractive, customized CXL—a novel technique: the enhanced Athens protocol. Clin Ophthalmol. 2019;13:581.
86. Hafezi F, Kling S, Gilardoni F, et al. Individualized corneal cross-linking with riboflavin and UV-A in ultrathin corneas: the sub400 protocol. Am J Ophthalmol. 2021;224:133–142.
87. Aydin E, Aslan MG. The efficiency and safety of oxygen-supplemented accelerated transepithelial corneal cross-linking. Int Ophthalmol. 2021;19:1–3.
88. De Paiva ACM, Da Costa FM, Da Fonseca AS. Photodynamic therapy for treatment of bacterial keratitis. Photodiagnosis Photodyn Ther. 2022;10:102717.
89. Naranjo A, Arboleda A, Martinez JD, et al. Rose Bengal photodynamic antimicrobial therapy for patients with progressive infectious keratitis: a pilot clinical study. Am J Ophthalmol. 2019;208:387–396.
90. Cherfan D, Verter EE, Melki S, et al. Collagen cross-linking using rose bengal and green light to increase corneal stiffness. Invest Ophthalmol Vis Sci. 2013;54:3426–3433.
91. Kobashi H, Torii H, Toda I, et al. Clinical outcomes of KeraVio using violet light: emitting glasses and riboflavin drops for corneal ectasia: a pilot study. Br J Ophthalmol. 2021;105:1376–1382.
92. Dackowski EK, Logroño JB, Rivera C, et al. Transepithelial corneal crosslinking using a novel ultraviolet light-emitting contact lens device: a pilot study. Trans Vis Sci Technol. 2021;10:5.
93. Chen X, Zhao J, Iselin KC, et al. Keratoconus detection of changes using deep learning of colour-coded maps. BMJ Open Ophthalmol. 2021;6:e000824.
94. Shetty R, Kundu G, Narasimhan R, et al. Artificial intelligence efficiently identifies regional differences in the progression of tomographic parameters of keratoconic corneas. J Refract Surg. 2021;37:240–248.
95. Rechichi M, Mazzotta C, Oliverio GW, et al. Selective transepithelial ablation with simultaneous accelerated Corneal Cross-linking for corneal regularization of keratoconus: the STARE-X Protocol. J Cataract Refract Surg. 2021;47:1403–1410.

keratoconus; cornea; crosslinking; ectatic disorders

Copyright © 2022 Asia-Pacific Academy of Ophthalmology. Published by Wolters Kluwer Health, Inc. on behalf of the Asia-Pacific Academy of Ophthalmology.