Portable and head-mounted corneal cross-linking systems for keratoconus and infectious keratitis: a narrative review

Introduction

Background

Progressive corneal thinning and protrusion are hallmarks of keratoconus, a condition that can result in visual morbidity if left untreated. The disease frequently begins during adolescence or early adulthood and may progress aggressively in pediatric patients (1,2). Reported incidence rates vary considerably across populations, ranging from 1.5 to 22.3 per 100,000 individuals annually (3).

Pathophysiologically, keratoconus is characterized by progressive stromal thinning and biomechanical weakening of the cornea, producing irregular astigmatism, progressive myopia, and reduced vision; habitual eye rubbing and atopy are recognized associations. Infectious keratitis, in contrast, is a pathogen-driven stromal infection presenting with pain, epithelial defect, stromal infiltrate, and, in severe cases, stromal melting and perforation. Understanding these distinct mechanisms clarifies why corneal collagen cross-linking (CXL) is used to biomechanically stabilize the ectatic cornea in keratoconus and, in photoactivated chromophore for keratitis corneal cross-linking (PACK-CXL), to add an antimicrobial and anti-collagenolytic effect in keratitis (4,5).

Over the last two decades, CXL has become the primary intervention to halt the progression of keratoconus and other corneal ectatic disorders. The treatment combines riboflavin with ultraviolet A (UVA) irradiation to enhance corneal biomechanical strength through the formation of additional collagen cross-links within the stromal matrix (6,7).

The photochemical reaction generated by riboflavin and UVA produces reactive oxygen species (ROS) capable of damaging microbial structures. This process led to the development of PACK-CXL as an adjunctive therapy for infectious keratitis. Evidence suggests that PACK-CXL may contribute to pathogen reduction and reduce stromal degradation (8,9).

Rationale and knowledge gap

High equipment costs and the need for dedicated clinical infrastructure restrict the use of traditional tabletop CXL in resource-limited settings (2,8). Improving access to corneal therapies remains an important priority, particularly in low- and middle-income countries where ophthalmic infrastructure is limited, and shortages of corneal specialists are common (10,11). These circumstances have encouraged the development of alternative ultraviolet delivery platforms that can be used in a wider clinical setting. Slit-lamp-based, portable, and head-mounted CXL systems represent some of the most recent developments in this field (8,12).

Despite the rising interest in these alternative platforms, the available literature remains scattered among laboratory investigations, pilot clinical studies, and technical evaluations. To date, few reviews have specifically examined portable and head-mounted CXL systems within the broader context of keratoconus and infectious keratitis management.

Objective

This review provides an overview of current CXL technologies, with a focus on the broader class of alternative ultraviolet delivery systems (slit-lamp-based, office-based, portable, and head-mounted) for keratoconus and infectious keratitis. It uses the head-mounted Ultraviolet Gadjah Mada (UV-GAMA) device as one representative example rather than the sole focus, and evaluates the potential of these platforms to expand access to corneal therapies in resource-limited settings. We present this article in accordance with the Narrative Review reporting checklist (available at https://aes.amegroups.com/article/view/10.21037/aes-2026-0035/rc).

Methods

The literature included in this review was selected to provide an overview of both established and emerging CXL technologies. Searches were conducted in PubMed, Scopus, and Google Scholar between 5 November and 15 December 2025. Keywords related to CXL, PACK-CXL, pediatric keratoconus, infectious keratitis, and ultraviolet delivery systems were used in various combinations.

Priority was given to studies providing information on device characteristics, clinical outcomes, antimicrobial applications, and technological developments. For our analysis, we selected original studies, clinical trials, observational studies, systematic reviews, and meta-analyses published in English up to December 2025. Studies lacking sufficient methodological detail, conference abstracts, editorials, letters, and publications unrelated to CXL were excluded. Additional relevant articles were identified through manual screening of the reference lists of selected publications (Table 1).

For transparency, approximately 85 records were identified and screened by title and abstract, of which approximately 50 were included after full-text assessment. Screening and selection were performed independently by two reviewers; as this is a narrative review, no formal risk-of-bias assessment was undertaken.

Current evidence on conventional and evolving cross-linking technologies

Principles of CXL

CXL functions by generating a network of new covalent linkages among stromal collagen fibers to strengthen biomechanical stability. Ever since Wollensak and his colleagues published their findings, this technique has served as the mainstay for managing progressive keratoconus and other ectatic disorders, aiming directly at halting tissue deformation before significant visual deterioration occurs. The standard Dresden protocol is still a definitive benchmark for clinical CXL. The procedure combines epithelial debridement, topical riboflavin application, and UVA irradiation at an intensity of 3 mW/cm² for 30 minutes, delivering a total energy dose of 5.4 J/cm² (11). Riboflavin plays a dual role in CXL, functioning as both a photosensitizer and a protective agent by absorbing UVA radiation and limiting exposure of deeper ocular structures to phototoxic effects (12).

The epithelium-off Dresden protocol remains the guideline-supported, first-line standard of care for mild-to-moderate progressive corneal ectasia, and epithelium-off CXL is approved by the United States Food and Drug Administration for patients aged 14 to 65 years with progressive keratoconus or ectasia following keratorefractive surgery (4,13). In the standard protocol the corneal epithelium is removed, riboflavin is instilled every 2 minutes for 30 minutes to saturate the stroma, and the cornea is then irradiated with UVA at 365 nm for a further 30 minutes with continued riboflavin instillation, so that the procedure takes at least 60 minutes (4); the original Dresden description used riboflavin 0.1% in 20% dextran with irradiation at 370 nm and 3 mW/cm² over the central 7 mm of the cornea, delivering a total dose of 5.4 J/cm² (11). In rabbit eyes, the cytotoxic irradiance threshold for the corneal endothelium after riboflavin and UVA treatment was 0.36 mW/cm² (0.65 J/cm²), a level that can be reached at a corneal thickness below 400 µm when 3 mW/cm² is applied at the epithelial surface (11); a stromal thickness below 400 µm at the time of irradiation is therefore listed as a contraindication to conventional CXL (4). To reduce treatment time, accelerated protocols deliver the same 5.4 J/cm² fluence over shorter durations; irradiation at 9 mW/cm² for 10 minutes has demonstrated safety and efficacy in stabilizing progressive ectasia (14,15). Reported outcomes are best when treatment is given early in the disease process, and cross-linking is generally less effective in ectasia occurring after keratorefractive surgery than in keratoconus (4). Portable and head-mounted systems discussed below are intended as alternative delivery methods for these established protocols, not as new treatment paradigms.

Upon UVA irradiation, riboflavin within the corneal stroma undergoes photoactivation, generating ROS through oxygen-dependent photochemical reactions. These ROS facilitate the formation of new intermolecular and intramolecular collagen cross-links, leading to the increase in corneal stiffness observed following CXL treatment (7,12). Published clinical data demonstrate that CXL can slow or stop keratoconus progression and may improve keratometric and visual outcomes in selected patients (14,15).

Recognizing the practical limitations of prolonged treatment times, some studies introduced accelerated protocols that use higher irradiance levels over shorter exposure periods. For example, irradiation at 9 mW/cm² for 10 minutes provides the same total fluence as the conventional Dresden protocol (14-16).

Furthermore, the therapeutic potential of UVA-activated riboflavin extends to antimicrobial applications. The generated ROS induce irreversible oxidative damage to microbial cell walls and nucleic acids, a principle that forms the basis of PACK-CXL (17-19). At the same time, the cross-linked stroma becomes highly resistant to the collagenolytic enzymes released by invading pathogens, thereby reducing the risk of stromal thinning and corneal perforation (18,19).

Conventional CXL systems and their limitations

Stationary tabletop UVA irradiation systems have formed the foundation of clinical CXL since the introduction of the Dresden protocol (14,20). These platforms provide standardized irradiation conditions and have demonstrated favorable safety and efficacy in the treatment of progressive corneal ectasia (11,14,20,21). However, conventional CXL often depends on specialized equipment and appropriate infrastructure such as an operating room. These requirements may restrict its widespread availability (2,8). Operationally, the patient is required to remain in a supine position during the irradiation process (22). This positioning requirement may be difficult for certain patient populations, including children, older adults, individuals with impaired mobility, and those with comorbid conditions such as chronic obstructive pulmonary disease, cor pulmonale, heart failure, or obesity (10).

These practical challenges directly restrict care in low- and middle-income regions, where access to advanced ophthalmic technology is limited. Consequently, there is an increasing clinical push toward the development of portable CXL devices.

Any alternative delivery platform must also reproduce the patient-selection framework and the complication profile established for conventional systems. CXL is relatively contraindicated when corneal thickness is below 400 µm, and in eyes with prior herpes simplex virus keratitis (because ultraviolet irradiation may reactivate the virus), corneal stromal scarring, severe ocular surface disease, or autoimmune disorders associated with corneal thinning (4). Recognized complications include punctate keratitis, corneal striae, photophobia, dry eye, eye pain, infectious keratitis, sterile infiltrates, corneal haze, corneal scarring, non-healing epithelial defects, and corneal edema attributable to endothelial damage; these occur more frequently in patients older than 35 years, when corneal thickness is below 400 µm, and when preoperative corrected visual acuity is poor (4). Treatment failure, defined as progression of keratoconus after cross-linking, is associated with an age of 35 years or older, a preoperative corrected distance visual acuity worse than 20/25, and a preoperative maximum keratometry above 58.00 D (4). These are the reference values against which portable and head-mounted platforms should be judged.

Portable and head-mounted CXL systems

The move toward portable, office-based CXL is not entirely new. Hafezi et al. illustrated this by demonstrating the clinical viability of slit-lamp-based CXL and PACK-CXL. Their work proved that effective treatment could be administered to patients in a seated position, within a clinic setting, eliminating the need for a stationary tabletop system without compromising safety or efficacy (8). Knyazer et al. reported the successful use of slit-lamp-based PACK-CXL in a wheelchair-dependent patient with Duchenne muscular dystrophy, highlighting the potential advantages of upright treatment for patients who are unable to tolerate the conventional supine position (23). This shift is further supported by Salmon et al., whose findings indicated that upright irradiation does not significantly alter the distribution of riboflavin within the cornea. Consequently, the photochemical environment achieved in the seated position appears to be functionally equivalent to the traditional supine approach (24).

One recent development in this field is the UV-GAMA device, a wearable, head-mounted platform designed to facilitate portable CXL delivery (Figure 1). By utilizing lightweight polymers like polyvinyl chloride (PVC), rubber, and polycarbonate, the device weighs less than 240 g. Its adjustable headband frame ensures that the device fits securely on any patient. This design allows CXL procedures to be performed in sitting, standing, or semi-recumbent positions and may improve procedural flexibility in patients who have difficulty tolerating traditional treatment setups. Integrated within the device are a battery management module and an automated timer to regulate power and ensure dosing consistency. For extended clinical sessions, the system also supports operation via an external power supply (10). UV-GAMA is presented here as one representative head-mounted example among the broader class of alternative delivery platforms; comparable slit-lamp and office-based systems are discussed above.

Figure 1 UV-GAMA portable head-mounted corneal cross-linking device. Reproduced from Wibowo et al. (10) under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0), which permits non-commercial reuse with attribution; no separate written permission from the copyright holder is therefore required. UV-GAMA, Ultraviolet Gadjah Mada.

The class of alternative platforms is not limited to slit-lamp and head-mounted designs, and one handheld system has now been assessed clinically. The CXL pen is a 30 mm titanium tube carrying a light-emitting diode with a peak wavelength of 370 nm at its distal end and a removable 8 mm silica meniscus lens at its proximal end, powered by a rechargeable battery pack, and applied directly to the riboflavin-soaked cornea under a suction of approximately 43.5 mmHg (25). In a prospective phase I/II study, 21 eyes of 21 patients aged 14 to 27 years with progressive keratoconus were treated unilaterally with an accelerated epithelium-off protocol at 9 mW/cm² and 370 nm. At 12 months, topographic astigmatism decreased by a mean of 0.56 D [95% confidence interval (CI): 0.10 to 1.03; P=0.02], maximum keratometry decreased by 0.63 D without reaching significance (P=0.33), and best-corrected visual acuity improved from 0.54 to 0.17 logarithm of the minimum angle of resolution (logMAR) (P<0.0001); pachymetry, intraocular pressure, and endothelial cell density were stable, the last changing from 2,563 to 2,698 cells/mm², and the only reported adverse events were early postoperative pain in three eyes (14.3%) that resolved within 1 week (25). The study was single-arm and uncontrolled; the sample was small, the device inventors are among the authors and hold a financial interest in it, and a correction to the article has since been published, so the findings describe feasibility and 12-month safety rather than comparative efficacy. The reported outcomes and technical characteristics of the conventional, slit-lamp-based, head-mounted, and handheld platforms discussed above are summarized in Table 2.

CXL in pediatric keratoconus

Keratoconus in children typically presents with a more aggressive clinical course compared to the adult form, often characterized by an earlier onset, more rapid deterioration, and a heightened risk of significant visual loss (2). Various contributing factors such as habitual eye rubbing, atopic conditions, vernal keratoconjunctivitis, connective tissue disorders, and genetic predisposition have been linked to this accelerated progression (2,28).

Current clinical standards identify CXL as the primary intervention to slow or stop the advancement of pediatric keratoconus. Multiple studies have reported favorable outcomes following treatment, including stabilization of corneal topography and reduced keratometric progression in pediatric populations. Early intervention is often emphasized, as delayed treatment may increase the likelihood of advanced ectasia, corneal scarring, and eventual need for corneal transplantation (14,29). In a prospective interventional case series of 40 eyes of 40 patients aged 9 to 18 years (mean 14.2±1.7 years) with stage II keratoconus, uncorrected visual acuity improved from 0.79±0.21 to 0.58±0.18 logMAR and best spectacle-corrected visual acuity from 0.39±0.10 to 0.20±0.09 logMAR at 24 months, simulated keratometry in the flattest meridian decreased from 46.32 to 45.30 D (P=0.04) and minimum keratometry from 42.95 to 39.47 D (P=0.01), and endothelial cell density was unchanged (3,221±212 to 3,209±178 cells/mm²) (29). That series was non-randomized and had no control arm, so it documents the post-treatment course rather than efficacy relative to the natural history of the disease.

Beyond the procedure itself, a substantial challenge in pediatric keratoconus lies in the diagnostic pathway rather than in the treatment step alone (30). Obtaining reliable corneal topography and tomography can be difficult in young or uncooperative children: in a prospective pediatric study, reliable completion of corneal testing was achieved in only 55% of children with trisomy 21, compared with 87% and 88% in the other pediatric risk groups, with significantly poorer tomographic quality scores (31). Documenting progression is equally demanding, because it requires reproducible serial measurements; the definition of progression in pediatric keratoconus remains inconsistent across studies (32), and current consensus requires the observed change to exceed the measurement noise of the testing system before it is accepted as genuine progression (13). Partly for the same reasons of limited cooperation and fixation, CXL in children is frequently performed under sedation or general anesthesia rather than topical anesthesia alone, particularly in younger children and those with developmental delay (33). Portable, position-flexible devices therefore address only part of this pathway; accurate diagnosis and reliable monitoring remain equally important.

Nevertheless, treatment in children presents unique challenges. Maintaining adequate fixation throughout the procedure may be difficult, and anxiety associated with treatment can affect patient cooperation (34). Interest in portable and wearable CXL technologies has partly arisen from these practical considerations. Devices capable of accommodating different treatment positions may simplify treatment delivery and improve patient tolerance. To date, however, no clinical outcome data for head-mounted UVA delivery in keratoconus have been published in the peer-reviewed literature. The available evidence for the UV-GAMA platform is limited to a single pre-clinical in vitro study of antimicrobial efficacy (10), which reports no keratometric, visual, or endothelial outcomes. Prospective clinical evaluation reporting keratometric, visual, endothelial, and pachymetric outcomes over at least 12 months is therefore required before any comparison with conventional CXL can be made.

PACK-CXL in infectious keratitis

Infectious keratitis frequently causes irreversible corneal blindness, particularly in developing countries due to delayed interventions and limited access to specialist care (35). Although antimicrobial therapy remains the standard of treatment, clinical management may be complicated by poor drug penetration, treatment failure, and the growing challenge of antimicrobial resistance (36).

Interest in PACK-CXL arose from the observation that riboflavin-UVA photochemistry exhibits antimicrobial activity in addition to its biomechanical action on the cornea. During irradiation, ROS can induce oxidative damage to microbial cellular structures, including nucleic acids, proteins, and cell membranes (17,37). PACK-CXL may also help stabilize corneal structure by reinforcing stromal collagen. This reinforcement increases tissue resistance to the collagenolytic enzymes that typically drive stromal melting and perforation during severe infection (38).

Recent preclinical data highlight the potential of head-mounted CXL technology in this field. Wibowo et al. reported that riboflavin activated by the UV-GAMA system reduced bacterial and fungal colony counts in vitro to a degree that did not differ significantly from a conventional CXL device. Pronounced reductions in viability were observed among Gram-positive organisms, including Staphylococcus aureus, Staphylococcus epidermidis, and Streptococcus pneumoniae. Consistent with previous reports, Pseudomonas aeruginosa showed a more limited response, likely reflecting the protective barrier function of the Gram-negative bacterial outer membrane (10,39,40). Antifungal activity was also observed against Aspergillus fumigatus, Candida albicans, and Fusarium species, supporting the feasibility of portable ultraviolet delivery for antimicrobial applications (10). These findings derive from a single pre-clinical study in which bacterial and fungal suspensions in microtiter plates, rather than infected corneas, were irradiated at a working distance of 5 cm with nine replicates per condition; reported mean colony counts fell from 286,933 to 33 colony-forming units (CFU)/mL for S. aureus and from 67,000 to 20,578 CFU/mL for P. aeruginosa after riboflavin plus UV-GAMA (10). The absence of a statistically significant difference between the two devices was not tested as an equivalence or non-inferiority hypothesis and therefore does not establish comparable performance, and a suspension model does not reproduce ultraviolet attenuation, riboflavin gradients, or oxygen dynamics within the corneal stroma.

These in vitro effects should be interpreted cautiously, because the clinical evidence is stratified by infiltrate depth as well as by organism. PACK-CXL is an adjunct to, and not a replacement for, appropriate antimicrobial therapy, and it is used off-label for this indication, without United States Food and Drug Administration approval for microbial keratitis (5,41). For bacterial keratitis, current guidance reports more evidence of benefit when cross-linking is added to standard antibiotic therapy in more anterior infections, particularly in difficult cases, while noting that pooled results across published series have been variable (5); a Cochrane review of three trials similarly concluded that it remains very uncertain whether PACK-CXL added to standard antibiotic therapy improves re-epithelialization and complete healing compared with antibiotics alone (risk ratio 1.53, 95% CI: 0.88 to 2.66) (41). For fungal keratitis, the picture is mixed rather than uniformly negative: pooled randomized evidence indicates that adjuvant PACK-CXL shortens the time to corneal healing (42) without reducing infiltrate size at 1 week, whereas umbrella-review sources report no effect in deep stromal disease and mixed results in superficial infection (41). Infiltrate depth appears to be the critical determinant: PACK-CXL may be effective when the infiltrate is confined to the anterior and middle stroma, but it is ineffective and raises safety concerns in deep infiltration or melting involving the posterior third of the stroma, because ultraviolet energy is absorbed within the anterior cornea (5,41). Evidence in Acanthamoeba keratitis is too limited to support a recommendation, and cross-linking is not currently recommended for this indication because of efficacy concerns (41). PACK-CXL is contraindicated in viral keratitis (41).

Clinical evidence supports the use of PACK-CXL as an adjunct to, and not a replacement for, standard antimicrobial therapy. Reported case series and a systematic review describe reduced infiltrate and improved healing in selected bacterial and treatment-resistant keratitis when PACK-CXL was added to antimicrobial treatment (43,44). More recently, a prospective, multicenter, unmasked, randomized phase 3 trial by Hafezi et al. compared standalone PACK-CXL with standard antimicrobial therapy in early-to-moderate infectious keratitis of presumed bacterial, fungal, or mixed origin. Thirty-nine adults with infiltrates of 4 mm or less in diameter and a maximum depth of 350 µm were randomized 1:1 to PACK-CXL alone (n=18) or antimicrobial therapy alone (n=21), and 35 eyes entered the analysis of the primary endpoint, time to re-epithelialization. Median time to re-epithelialization was 7.0 days in both arms (P=0.828; 95% CI for the between-group difference: −12.66 to 3.38 days), and treatment success was 88.9% (16 of 18) with PACK-CXL versus 90.5% (19 of 21) with medication (45). Two qualifications apply: the trial evaluated PACK-CXL as a standalone first-line treatment rather than as an adjunct, and it was neither designed nor powered as a non-inferiority study, so the absence of a significant difference in 35 eyes does not establish equivalence to antimicrobial therapy. In the meta-analysis by Ting et al., adjuvant PACK-CXL was associated with a shorter mean time to complete corneal healing (mean difference −7.44 days, 95% CI: −10.71 to −4.16) and a smaller infiltrate at 7 days (mean difference −5.49 mm², 95% CI: −7.44 to −3.54), with no significant difference in epithelial defect size, corrected distance visual acuity, or risk of adverse events (41,44). Across the systematic reviews summarized in a recent umbrella review, however, protocols were heterogeneous, particularly with respect to de-epithelialization before irradiation, healing outcomes were inconsistent, and the certainty of evidence for every recommendation was graded no higher than moderate (41).

Randomized evidence published since 2022 has not resolved this uncertainty and, for bacterial keratitis, has moved against the intervention. The Steroids and Cross-Linking for Ulcer Treatment II trial was a double-masked, sham- and placebo-controlled randomized trial of 280 patients with smear- or culture-positive bacterial corneal ulcers and presenting acuity of 20/40 or worse, allocated to topical moxifloxacin with placebo and sham cross-linking, to moxifloxacin with difluprednate and sham cross-linking, or to moxifloxacin with difluprednate and cross-linking delivered by a modified Dresden protocol at 365 nm and 3 mW/cm² for 30 minutes (46). Adding cross-linking to topical corticosteroid did not improve the primary endpoint of best spectacle-corrected visual acuity at 6 months (difference 0.04 logMAR, 95% CI: −0.09 to 0.17; P=0.62), did not increase microbiological cure at day 2 (84% versus 85%), and did not significantly reduce perforation or the need for therapeutic penetrating keratoplasty (hazard ratio 0.48, 95% CI: 0.14 to 1.67; P=0.25), whereas scar size at 6 months was larger in the cross-linking arm (difference 0.56 mm, 95% CI: 0.20 to 0.92; P=0.02) and reported pain at day 3 was higher (difference 0.43, 95% CI: 0.11 to 0.75; P=0.01) (46). The investigators note that standard-fluence Dresden irradiation may be below the level required for a bactericidal effect and that the perforation analysis was underpowered (46).

For fungal keratitis the most recent randomized data are more favorable. In a multicenter, prospective randomized trial with masked outcome assessment, 49 patients with early- to mid-stage fungal keratitis received either accelerated PACK-CXL added to systemic and topical antifungal therapy or antifungal therapy alone (47). Median time to clinical cure was halved, from 60 to 30 days (hazard ratio 2.76, 95% CI: 1.40 to 5.44; P=0.003), median time to clinical improvement fell from 14 to 7 days (hazard ratio 2.61, 95% CI: 1.33 to 5.11; P=0.005), and the corneal transplantation rate was lower with combined treatment (4.17% versus 28.00%; P=0.049), while the change in best-corrected visual acuity at 1 month did not differ between groups (P=0.771) and no adverse events were reported (47). The trial was small, was conducted at centers within a single country, could not mask the treating surgeon, and used five co-primary endpoints, so its estimates remain preliminary.

These divergent results are consistent with the pooled randomized evidence. A meta-analysis restricted to randomized trials identified seven trials with 283 patients and found that adjuvant PACK-CXL shortened the time to corneal healing in fungal keratitis (mean difference −1.13 months, 95% CI: −1.83 to −0.42; I²=0%), while the risk of adverse events did not differ from standard antimicrobial therapy alone in either fungal (risk ratio 0.78, 95% CI: 0.38 to 1.60) or bacterial disease (risk ratio 0.36, 95% CI: 0.08 to 1.71); no benefit was demonstrated for visual acuity, epithelial defect size, or infiltrate size at 1 week, every included trial was judged to be at high risk of bias, and the certainty of the evidence was graded low (42). Taken together, the current randomized evidence supports PACK-CXL as an adjunct with a plausible effect on healing time in early or superficial fungal disease, and does not support it as a means of improving visual outcome or preventing perforation in bacterial disease.

Another notable development has been the introduction of slit-lamp-based PACK-CXL. Early reports suggest that treatment may be delivered safely outside the conventional operating room, potentially increasing accessibility in settings where dedicated CXL infrastructure is unavailable. Olshaker et al. treated 20 eyes with bacterial, fungal, or mixed keratitis at the slit lamp using a high fluence of 10.0 J/cm² delivered at 30 mW/cm² over 5 minutes and 33 seconds, and reported complete re-epithelialization at 8.2±2.8 days, no eye requiring keratoplasty, and no change in endothelial cell density (2,562.1±397.3 versus 2,564.8±404.5 cells/mm²; P=0.96); the series had no control arm and a mean follow-up of only 1.7±1.0 months (26). Knyazer et al. described a single wheelchair user with Duchenne muscular dystrophy in whom two slit-lamp sessions of 10.8 J/cm² within 5 days were followed by complete epithelial closure 6 days after the second treatment, with a demarcation line at 260 µm (23). Both reports illustrate the practical advantages of alternative treatment positioning, particularly for patients who are unable to tolerate the conventional supine position, but neither provides comparative or long-term data (23,26). In published correspondence on the original slit-lamp report, the same group stated that no case of infectious keratitis had occurred in more than 700 procedure-room CXL treatments over more than 5 years, which is an uncontrolled and self-reported observation rather than a comparative safety estimate (27). Collectively, these findings indicate that flexible ultraviolet delivery platforms may help broaden the clinical applicability of PACK-CXL.

Safety considerations for portable and head-mounted UVA delivery

The safety profile of any UVA-delivery platform depends on parameters that are well characterized for standard tabletop systems but require independent validation for each portable device. Irradiance calibration and power stability determine whether the intended fluence is actually delivered; even small deviations alter the biological effect and, with portable or battery-powered units, must be verified over the full treatment. Working distance and beam homogeneity affect the uniformity of stromal exposure, while accurate centration and stable fixation are essential because head-mounted systems typically lack integrated eye tracking. Endothelial safety is governed by the minimum stromal thickness (generally at least 400 µm after epithelial removal) required to keep endothelial irradiance below the cytotoxic threshold; limbal and limbal stem-cell exposure should be minimized. Oxygen availability influences the photochemical reaction, which is relevant for accelerated protocols. Finally, automated timers, dosimetry feedback, and battery/fail-safe mechanisms are needed to prevent over- or under-treatment. Until these parameters are independently validated for each portable system, safety claims should remain provisional (4,11,12). For the UV-GAMA device, the reported parameters are an emission wavelength of 373 nm, measured against 370 nm for the comparator system, an irradiance of 3 mW/cm² applied for 30 minutes, a fixed 5 mm beam diameter produced by a plano-convex lens, and a 5 cm working distance, with output verified using an HR4000CG UV-NIR spectroradiometer certified by the national standardization agency (10). The corresponding fluence of 5.4 J/cm² is calculated from the stated irradiance and exposure time and is not reported as such in the source. These values were obtained on the bench; irradiance stability across a full treatment, beam homogeneity at the corneal plane, and calibration under clinical conditions have not been published.

The phase I/II handheld-device study also illustrates what per-treatment dosimetric verification looks like in practice. Before each treatment, the UVA intensity was measured through the device lens with an ultraviolet meter, because output from the portable battery pack varies, and the exposure time was then adjusted according to the Bunsen-Roscoe reciprocity law so that the intended total fluence was still delivered (25). Verification of this kind before every treatment, rather than a single bench measurement at the time of device development, is what a battery-powered platform requires before its delivered dose can be assumed to match its nominal specification.

The corneal thickness threshold deserves particular emphasis, because it is where dosimetric precision and endothelial safety intersect. Current practice guidance lists a stromal thickness below 400 µm as a contraindication to conventional CXL (4,48). A recent systematic review and meta-analysis of 29 studies and 470 eyes with corneal thickness below 400 µm nevertheless reported flattening of maximum keratometry of 1.4 D (95% CI: 0.9 to 2.0) and a mean improvement in corrected distance visual acuity of 0.09 logMAR (95% CI: 0.06 to 0.13) at 12 months, with no major complications and only minimal reductions in corneal thickness and endothelial cell density (48). These outcomes were obtained with protocol modifications, most commonly hypotonic riboflavin to swell the stroma, and also contact lens-assisted, lenticule-assisted, iontophoretic, and customized epithelial-island techniques (48).

Two implications follow for portable and head-mounted platforms. First, the evidence supporting treatment of thin corneas rests almost entirely on retrospective, single-arm case series of suboptimal methodological quality, with substantial statistical heterogeneity, scarce follow-up beyond 24 months, and no randomized comparisons; the pooled endothelial cell loss of approximately 103 cells/mm² at 12 months was itself highly heterogeneous, and adverse events after CXL have been reported as late as 6 years postoperatively (48). Second, every one of these modified protocols depends on tightly controlled irradiance, exposure time, and intraoperative pachymetry, in some reports with irradiation time individualized to the measured stromal thickness (48). A delivery platform whose irradiance output has not been independently verified across the full treatment duration therefore cannot be assumed to be safe in these scenarios, even if it performs acceptably in corneas of normal thickness.

Socioeconomic implications and global health perspectives

The challenges of corneal disease management are not solely related to treatment effectiveness but also to treatment accessibility. Although CXL represents a definitive benchmark for halting ectasias and an expanding adjunct for keratitis, substantial infrastructure barriers continue to restrict its clinical availability across developing countries (22,38,49). The financial burden associated with conventional tabletop equipment and the need for specialized clinical facilities often hinder implementation in resource-limited settings (2,8,10).

Portable CXL technologies offer a potential solution to these barriers by reducing equipment costs, minimizing infrastructure requirements, and expanding access to treatment (2,8,10). Wearable and head-mounted systems may also help overcome some of the practical limitations associated with conventional tabletop devices, potentially improving access to treatment for patients who have difficulty undergoing standard procedures (8,10). Such clinical utility is particularly relevant to pediatric fields, where treatment cooperation can be a significant challenge. Alternative delivery platforms that accommodate different patient positions may simplify treatment administration and improve feasibility in selected patient populations (2,10,34).

Beyond individual care, portable CXL systems may have broader implications for healthcare delivery. Their mobility creates opportunities for outreach programs, mobile eye-care services, and remote healthcare settings (10,50). Although further evaluation is required, such technologies could contribute to reducing disparities in access to corneal care.

Strengths and limitations

An important strength of this review is its synthesis of evidence related to alternative CXL delivery platforms, including slit-lamp-based and head-mounted systems. By integrating findings from experimental investigations, clinical studies, and technological reports, the review highlights emerging developments that may improve the accessibility of CXL in the future.

However, the current evidence base remains limited and heterogeneous. While conventional CXL has been extensively investigated, portable systems are supported by a comparatively small body of literature. Variability in technical specifications, irradiation protocols, and reported outcomes limits direct comparisons among available devices. Furthermore, long-term studies evaluating safety and effectiveness remain scarce. Finally, as a narrative review, this study does not follow the structured methodology of a systematic review or meta-analysis, which may introduce a degree of selection bias.

More specifically, much of the evidence for portable and head-mounted systems derives from small, often single-center studies, short follow-up, and, in several instances, single case reports or a single randomized trial. For PACK-CXL, the absence of a standardized protocol and organism-dependent efficacy further limit generalizability. These constraints should temper any inference of equivalence to conventional CXL.

Conclusions

CXL has established itself as the primary intervention for halting the progression of keratoconus, with its application in infectious keratitis (PACK-CXL) explored as an adjunctive therapy alongside standard antimicrobial treatment. Access to treatment remains influenced by the availability of specialized equipment and infrastructure that may limit the use of conventional CXL systems in many healthcare settings. Alternative delivery platforms, including slit-lamp-based and head-mounted CXL devices such as the UV-GAMA system, have been developed to address these practical challenges. Early, mostly short-term experience suggests these technologies may allow more flexible treatment delivery; however, comparability to conventional tabletop CXL has not been established and rests on limited evidence, including small single-center studies and case reports. Any potential advantages, including for pediatric patients and for expanding access in resource-limited settings, therefore remain to be confirmed. Continued clinical validation, standardized dosimetry, and longer-term safety and efficacy data are required before these systems can be widely integrated into routine practice.

Acknowledgments

The authors thank the professional language editor who checked the manuscript for grammar, spelling, punctuation, and clarity; this contribution does not meet the criteria for authorship.

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://aes.amegroups.com/article/view/10.21037/aes-2026-0035/rc

Peer Review File: Available at https://aes.amegroups.com/article/view/10.21037/aes-2026-0035/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://aes.amegroups.com/article/view/10.21037/aes-2026-0035/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Santodomingo-Rubido J, Carracedo G, Suzaki A, et al. Keratoconus: An updated review. Cont Lens Anterior Eye 2022;45:101559. [Crossref] [PubMed]
2. Olivo-Payne A, Abdala-Figuerola A, Hernandez-Bogantes E, et al. Optimal management of pediatric keratoconus: challenges and solutions. Clin Ophthalmol 2019;13:1183-91. [Crossref] [PubMed]
3. Chung IK, Kim BS, Han KD, et al. Ten-year incidence of keratoconus in relation to sex, age, and thyroid gland dysfunction: a nationwide population-based cohort study (2009-2018). Ann Transl Med 2024;12:45. [Crossref] [PubMed]
4. Jhanji V, Ahmad S, Amescua G, et al. Corneal ectasia Preferred Practice Pattern. Ophthalmology 2024;131:205-46.
5. Rhee MK, Ahmad S, Amescua G, et al. Bacterial keratitis Preferred Practice Pattern. Ophthalmology 2024;131:87-133.
6. Wu D, Lim DK, Lim BXH, et al. Corneal Cross-Linking: The Evolution of Treatment for Corneal Diseases. Front Pharmacol 2021;12:686630. [Crossref] [PubMed]
7. Rubinfeld RS, Caruso C, Ostacolo C. Corneal Cross-Linking: The Science Beyond the Myths and Misconceptions. Cornea 2019;38:780-90. [Crossref] [PubMed]
8. Hafezi F, Richoz O, Torres-Netto EA, et al. Corneal Cross-linking at the Slit Lamp. J Refract Surg 2021;37:78-82. [Crossref] [PubMed]
9. Papachristoforou N, Ueno A, Ledwos K, et al. A Review of Keratoconus Cross-Linking Treatment Methods. J Clin Med 2025;14:1702. [Crossref] [PubMed]
10. Wibowo E, Respatika D, Mahayana IT, et al. Newly developed head-mounted ultraviolet-activated riboflavin device for corneal cross-linking therapy. Pak J Ophthalmol 2025;41:12-8.
11. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-a-induced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol 2003;135:620-7. [Crossref] [PubMed]
12. Santhiago MR, Randleman JB. The biology of corneal cross-linking derived from ultraviolet light and riboflavin. Exp Eye Res 2021;202:108355. [Crossref] [PubMed]
13. Gomes JAP, Hafezi F, Ambrósio R Jr, et al. Global Consensus on Keratoconus and Ectatic Diseases-Edition 2. Cornea 2026;45:888-908. [Crossref] [PubMed]
14. Choi M, Kim J, Kim EK, et al. Comparison of the Conventional Dresden Protocol and Accelerated Protocol With Higher Ultraviolet Intensity in Corneal Collagen Cross-Linking for Keratoconus. Cornea 2017;36:523-9. [Crossref] [PubMed]
15. Özülken K, Aksoy Aydemir G, Aydemir E, et al. Comparison of Two Different Accelerated Corneal Cross-linking Procedure Outcomes in Patients with Keratoconus. Balkan Med J 2020;37:131-7. [Crossref] [PubMed]
16. Bouheraoua N, Jouve L, Borderie V, et al. Three Different Protocols of Corneal Collagen Crosslinking in Keratoconus: Conventional, Accelerated and Iontophoresis. J Vis Exp 2015;105:53119. [Crossref] [PubMed]
17. Martins SA, Combs JC, Noguera G, et al. Antimicrobial efficacy of riboflavin/UVA combination (365 nm) in vitro for bacterial and fungal isolates: a potential new treatment for infectious keratitis. Invest Ophthalmol Vis Sci 2008;49:3402-8. [Crossref] [PubMed]
18. Khan YA, Kashiwabuchi RT, Martins SA, et al. Riboflavin and ultraviolet light a therapy as an adjuvant treatment for medically refractive Acanthamoeba keratitis: report of 3 cases. Ophthalmology 2011;118:324-31. [Crossref] [PubMed]
19. Papaioannou L, Miligkos M, Papathanassiou M. Corneal Collagen Cross-Linking for Infectious Keratitis: A Systematic Review and Meta-Analysis. Cornea 2016;35:62-71. [Crossref] [PubMed]
20. Meiri Z, Keren S, Rosenblatt A, et al. Efficacy of Corneal Collagen Cross-Linking for the Treatment of Keratoconus: A Systematic Review and Meta-Analysis. Cornea 2016;35:417-28. [Crossref] [PubMed]
21. 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. [Crossref] [PubMed]
22. Burton MJ, Ramke J, Marques AP, et al. The Lancet Global Health Commission on Global Eye Health: vision beyond 2020. Lancet Glob Health 2021;9:e489-551. [Crossref] [PubMed]
23. Knyazer B, Hillen M, Hafezi F. Corneal cross-linking for infectious keratitis at the slit lamp in wheelchair users. J Refract Surg Case Rep 2021;1:e36-9.
24. 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. [Crossref] [PubMed]
25. Saks AA, Saks C, Bert B. Visual rehabilitation in keratoconus by altering corneal shape using the corneal crosslinking pen: initial results. Graefes Arch Clin Exp Ophthalmol 2025;263:2299-305. [Crossref] [PubMed]
26. Olshaker H, Achiron A, Chorny A, et al. Accelerated high fluence photoactivated chromophore for infectious keratitis-corneal cross-linking (PACK-CXL) at the slit lamp: a pilot study. Front Pharmacol 2023;14:1229095. [Crossref] [PubMed]
27. Peyman A, Pourazizi M. Corneal Cross-linking at the Slit Lamp: Concern About Risk of Corneal Ulcer. J Refract Surg 2021;37:572. [Crossref] [PubMed]
28. Léoni-Mesplié S, Mortemousque B, Touboul D, et al. Scalability and severity of keratoconus in children. Am J Ophthalmol 2012;154:56-62.e1. [Crossref] [PubMed]
29. Vinciguerra P, Albé E, Frueh BE, et al. Two-year corneal cross-linking results in patients younger than 18 years with documented progressive keratoconus. Am J Ophthalmol 2012;154:520-6. [Crossref] [PubMed]
30. Price LD, Larkin DFP. Diagnosis and management of keratoconus in the paediatric age group: a review of current evidence. Eye (Lond) 2023;37:3718-24. [Crossref] [PubMed]
31. Neustein RF, Lenhart PD. Detecting Keratoconus: Feasibility and Findings in Three Pediatric Risk Groups. J Pediatr Ophthalmol Strabismus 2022;59:94-101. [Crossref] [PubMed]
32. Antunes-Foschini R, Doná H, de Mello PHS, et al. Natural history and predictors for progression in pediatric keratoconus. Sci Rep 2023;13:4940. [Crossref] [PubMed]
33. Wójcik-Niklewska B, Filipek E, Janik P. Corneal Cross-Linking for Pediatric Keratoconus. Diagnostics (Basel) 2024;14:1950. [Crossref] [PubMed]
34. Chatzis N, Hafezi F. Progression of keratoconus and efficacy of pediatric corrected corneal collagen cross-linking in children and adolescents. J Refract Surg 2012;28:753-8. [Crossref] [PubMed]
35. Termote K, Joe AW, Butler AL, et al. Epidemiology of bacterial corneal ulcers at tertiary centres in Vancouver, B.C. Can J Ophthalmol 2018;53:330-6. [Crossref] [PubMed]
36. Egrilmez S, Yildirim-Theveny Ş. Treatment-Resistant Bacterial Keratitis: Challenges and Solutions. Clin Ophthalmol 2020;14:287-97. [Crossref] [PubMed]
37. Hamblin MR, Hasan T. Photodynamic therapy: a new antimicrobial approach to infectious disease?. Photochem Photobiol Sci 2004;3:436-50. [Crossref] [PubMed]
38. World Health Organization. World report on vision. Geneva: World Health Organization; 2019.
39. Thakuri PS, Joshi R, Basnet S, et al. Antibacterial photodynamic therapy on Staphylococcus aureus and Pseudomonas aeruginosa in-vitro. Nepal Med Coll J 2011;13:281-4. [PubMed]
40. Marrie A, Abdullatif AM, Gamal El Dine S, et al. Corneal cross-linking guards against infectious keratitis: an experimental model. Int Ophthalmol 2023;43:1241-8. [Crossref] [PubMed]
41. Farhadi F, Salehi-Pourmehr H, Arasteh A, et al. Corneal cross-linking for infectious keratitis of various causes: an umbrella review. BMC Ophthalmol 2025;25:238. [Crossref] [PubMed]
42. Liu S, Fang S, Zhang L. Photoactivated chromophore-corneal cross-linking accelerates corneal healing in fungal keratitis: an updated meta-analysis. Syst Rev 2023;12:208. [Crossref] [PubMed]
43. Makdoumi K, Mortensen J, Crafoord S. Infectious keratitis treated with corneal crosslinking. Cornea 2010;29:1353-8. [Crossref] [PubMed]
44. 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;17:624-34. [Crossref] [PubMed]
45. 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. Eye Vis (Lond) 2022;9:2. [Crossref] [PubMed]
46. Prajna NV, Lalitha P, Chandru S, et al. Steroids and Cross-Linking for Ulcer Treatment: The SCUT II Randomized Clinical Trial. JAMA Ophthalmol 2025;143:723-30. [Epub ahead of print]. [Crossref] [PubMed]
47. Zhou X, Liu C, Chen T, et al. Evaluation of the clinical efficacy of accelerated photoactivated chromophore for keratitis-CXL for early to mid-stage fungal keratitis: a prospective, multicenter randomized controlled trial. Int Ophthalmol 2026;46:298. [Crossref] [PubMed]
48. Mohammadi F, McGuinness MB, Mustafa MZ, et al. Effectiveness and Safety of Cross-Linking in Keratoconus Patients With Corneal Thickness <400 µm: A Systematic Review and Meta-analysis. Am J Ophthalmol 2026;281:363-83. [Crossref] [PubMed]
49. Whitcher JP, Srinivasan M, Upadhyay MP. Corneal blindness: a global perspective. Bull World Health Organ 2001;79:214-21. [PubMed]
50. Gairola S, Joshi P, Balasubramaniam A, et al. Keratoconus Classifier for Smartphone-based Corneal Topographer. Annu Int Conf IEEE Eng Med Biol Soc 2022;2022:1875-8. [Crossref] [PubMed]