Macular phototoxicity after corneal crosslinking

Alfayyadh, Mohammed A. MBBS; AlAbdulhadi, Halla A. MBBS, SB-Ophth; Ahad, Muhammad A. MBBS, FRCS, MRCOphth, PhD

Author Information
JCRS Online Case Reports 10(2):p e00078, April 2022. | DOI: 10.1097/j.jcro.0000000000000078
  • Open


Keratoconus is a bilateral, progressive, ectatic corneal disorder characterized by corneal thinning and protrusion. Its clinical features widely vary depending on the disease stage, ranging from no symptoms and normal slitlamp biomicroscopy findings to profound visual loss and readily detectable clinical signs.1

Corneal crosslinking (CXL) is a procedure that includes the combined administration of riboflavin and UV-A to strengthen the cornea and halt disease progression by altering the biomechanical properties of the cornea.2,3 CXL to treat keratoconus in humans was first performed in 2003 by Wollensak et al.4 Subsequently, several studies have shown that CXL is effective for halting disease progression, and the popularity of this procedure has increased in the past decade.5,6 The administration of riboflavin results in the irradiated UV-A being confined to the anterior 300 μm of the cornea. Hence, a minimum corneal thickness of at least 400 μm must be maintained after epithelial removal to avoid phototoxic damage to the corneal endothelium.7

Although CXL is generally considered to be safe, ocular side effects can occur.8,9 Macular phototoxicity is unlikely to occur as most of the UV-A is absorbed by the anterior segment; therefore, this side effect has not been frequently investigated. In this study, we describe the clinical features of a patient who experienced this side effect and the structural and angiographic retinal changes induced by UV-A exposure during CXL.

Patient Consent Statement

The patient provided written informed consent for the publication of this report.


A 37-year-old man with keratoconus sought medical advice to improve his vision. His medical, surgical, and drug histories were unremarkable. On examination, his corrected distance visual acuity (CDVA) with rigid gas-permeable contact lenses was 20/25 in both eyes. Anterior segment examination revealed clear corneas and clear crystalline lenses bilaterally. The intraocular pressure was within the normal range. No abnormal findings were noted on posterior segment examination. Preoperatively, the minimum corneal thickness in the right eye was 464 μm.

The patient underwent topography-guided custom ablation and CXL of the right eye using the Avedro KXL System. First, 9 mm of epithelium was removed using alcohol 20%, following which riboflavin 0.1% (VibeX Rapid, Avedro, Inc.) was applied to the cornea every 2 minutes for 10 minutes. The accelerated pulsed (15 mW/cm2 for 16 minutes) and high-fluence (7.2 J/cm2) protocols were followed, and the procedure was uneventful. Subsequently, a bandage contact lens was applied. The patient received a topical antibiotic for 1 week and a topical steroid for 8 weeks.

At the 1-week postoperative visit, the patient complained of decreased vision in the operated eye. Slitlamp biomicroscopy revealed a deep stromal/pre-Descemet membrane haze. Fundus examination was normal. Although the haze improved with time, the vision continued to deteriorate (Figure 1).

Figure 1.:
Slitlamp biomicroscopy photographs obtained with diffuse (A) and narrow focal slit (B) beams show a deep stromal haze.

After 6 months, there was still subjective deterioration in vision. A thorough examination was performed, which revealed changes in retinal pigment epithelium (RPE) in the foveal region (Figure 2, A); however, the crystalline lens was clear. An examination of the left eye revealed normal findings. Spectral-domain optical coherence tomography (SD-OCT), which was performed to evaluate the retinal structure, revealed disruption of the ellipsoid zone and outer retinal atrophy in the fovea of the right eye (Figure 3). Fluorescein angiography (FA) revealed window defects and staining in the foveal region, which were consistent with the SD-OCT findings, respectively. There was no fluorescein leakage or evidence of choroidal neovascular membrane formation (Figure 2, B–D). The patient was diagnosed with macular phototoxicity. An examination with a potential acuity meter and laser interferometry revealed no improvement in visual acuity. The minimum corneal thickness was still more than 400 μm (416 μm) (Figure 4). The results of ancillary tests of the left eye were unremarkable. The patient's CDVA with rigid gas-permeable contact lenses stabilized at 20/100 in the subsequent visits until the last follow-up (30 months).

Figure 2.:
Optos photograph of the right fundus (A) shows retinal pigment epithelium changes. Fluorescein angiography images of the right eye in the early arteriovenous (B), late arteriovenous (C), and early recirculation (D) phases show window defects and foveal staining.
Figure 3.:
Horizontal (A), vertical (B), and oblique (C and D) spectral-domain optical coherence tomography images of the right eye show disruption of the ellipsoid zone and outer retinal atrophy.
Figure 4.:
Corneal tomography images of the right eye before (A) and after (B) topography-guided custom ablation with corneal crosslinking. The minimum corneal thickness is maintained above 400 μm.


CXL is now considered a therapeutic option for progressive keratoconus.6 Several studies have proven its efficacy in halting disease progression, decreasing maximum keratometry readings, and improving CDVA.2,10 Studies have emphasized the importance of ensuring that the minimum corneal thickness is at least 400 μm before treatment to avoid damage to internal ocular structures, especially the lens and retina.11 Several recent in vivo and in vitro studies have revealed that UV-A exposure can cause damage to the corneal endothelium.8,12

Light can damage the retina through 3 mechanisms, namely, the photothermal, photomechanical, and photochemical effects.13 As in the case of UV light exposure, the photochemical effect is believed to be the most common mechanism by which retinal damage occurs after exposure to free radicals.13,14 Therefore, other ocular structures, such as the retina, might be susceptible to phototoxic damage caused by UV-A exposure during CXL.

Based on our patient's age and sex, a diagnosis of central serous choroidopathy was considered. However, it was ruled out as there was no clinical or angiographic evidence to support this diagnosis (Figure 3).

Certain medications, such as chloroquine, lomefloxacin, hematoporphyrin, hypericin, and phenothiazine, are known to cause ocular phototoxicity, particularly in the retina, when they are administered with UV-A.13 However, our patient had used none of these medications. The lack of improvement seen on examination with a potential acuity meter and laser interferometry indicates a nonrefractive source of visual loss.

Very few studies have examined the structural, angiographic, and electrophysiological changes in the retina after CXL using SD-OCT, FA, and multifocal electroretinography. In a pilot study, it was observed that in 17 patients who experienced a slight reduction in CDVA and an increase in corneal thickness at 7 days after treatment, both CDVA and corneal thickness returned to the baseline values 30 days later.15 Similarly, another study showed that a transient increase in macular thickness and microstructural alterations in the macular structure that were observed on SD-OCT returned to normal after 6 months.7 It was hypothesized by the authors that the changes in the former study were due to the occurrence of a self-limiting inflammation after UV-A exposure, whereas the structural alterations in the latter study were due to changes in the corneal characteristics and amount of astigmatism after CXL.7,15 In this case, we observed significant disruption of the inner segment/outer segment junction layer (ellipsoid zone) on SD-OCT, which is a typical sign of phototoxicity. Moreover, changes in RPE were evident on ophthalmoscopy. These findings are consistent with the fact that the outer retina and RPE are likely to sustain phototoxic damage because they contain chromophores, which are believed to mediate UV-induced damage.13 Although posterior segment imaging is not a part of the preoperative evaluation for CXL, the preoperative excellent vision supported the normal posterior segment findings observed during ophthalmoscopy. Besides, the unilaterality of the postoperative symptoms and signs emphasized the relation between the macular damage and the CXL. The use of an accelerated and a high-fluence CXL protocol (total energy of 7.2 J/cm2) instead of the Dresden protocol (5.4 J/cm2), which was used in the abovementioned studies, might have contributed to the occurrence of this side effect. Although accelerated protocols are used to decrease the duration of UV-A exposure and subsequent side effects, the minimum corneal thickness required for accelerated high-fluence protocols should be re-evaluated. Moreover, the deep stromal corneal haze observed in our patient might have indicated that the internal ocular structures had been exposed to toxic UV light.

In this case, FA revealed window defects and foveal staining. These findings are consistent with those observed on SD-OCT. However, no abnormalities were detected in the pilot study.15

Most patients in previous studies showed multifocal electroretinography abnormalities, some of which persisted for at least 6 months.7,15 Electrophysiological studies were not performed in our patient.

Although there exists reports of phototoxic damage to the macula after light exposure, it has never been observed after UV-A exposure combined with riboflavin administration. Therefore, in patients with a negative drug history and no history of macular disease, the high total energy administered when accelerated and high-fluence CXL protocols are used could be a contributing factor to macular phototoxicity. However, further reports of similar observations are needed to support this claim.


  • Corneal crosslinking (CXL) is commonly used to treat progressive corneal ectasia.
  • UV-A therapy can lead to ocular damage.
  • Patients undergoing CXL using the Dresden protocol should have a minimum corneal thickness of at least 400 μm to ensure safety of the internal ocular structures.


  • To our knowledge, this is the first report of macular phototoxicity and significant nonrefractive visual loss after corneal CXL.
  • When using accelerated or high-fluence protocols, practitioners should carefully select patients to avoid this rare but possibly serious side effect.


1. Rabinowitz YS. Keratoconus. Surv Ophthalmol 1998;42:297–319
2. Suri K, Hammersmith KM, Nagra PK. Corneal collagen cross-linking: ectasia and beyond. Curr Opin Ophthalmol 2012;23:280–287
3. Scott McCall A, Kraft S, Edelhauser HF, Kidder GW, Lundquist RR, Bradshaw HE, Dedeic Z, Dionne MJC, Clement EM, Conrad GW. Mechanisms of corneal tissue cross-linking in response to treatment with topical riboflavin and long-wavelength ultraviolet radiation (UVA). Invest Ophthalmol Vis Sci 2010;51:129–138
4. Wollensak G, Spoerl E, Seiler T. Riboflavin/ultraviolet-A-induced collagen crosslinking for the treatment of keratoconus. Am J Ophthalmol 2003;135:620–627
5. O'Brart DPS, Kwong TQ, Patel P, McDonald RJ, O'Brart NA. Long-term follow-up of riboflavin/ultraviolet A (370 nm) corneal collagen cross-linking to halt the progression of keratoconus. Br J Ophthalmol 2013;97:433–437
6. Caporossi A, Mazzotta C, Baiocchi S, Caporossi T. Long-term results of riboflavin ultraviolet A corneal collagen cross-linking for keratoconus in Italy: the Siena Eye Cross Study. Am J Ophthalmol 2010;149:585–593
7. Mirzaei M, Bagheri M, Taheri A. Influence of standard corneal cross-linking in keratoconus patients on macular profile. J Curr Ophthalmol 2018;30:330–336
8. Sharma A, Nottage JM, Mirchia K, Sharma R, Mohan K, Nirankari VS. Persistent corneal edema after collagen cross-linking for keratoconus. Am J Ophthalmol 2012;154:922–926.e1
9. Thorsrud A, Nicolaissen B, Drolsum L. Corneal collagen crosslinking in vitro: inhibited regeneration of human limbal epithelial cells after riboflavin–ultraviolet-A exposure. J Cataract Refract Surg 2012;38:1072–1076
10. Raiskup-Wolf F, Hoyer A, Spoerl E, Pillunat LE. Collagen crosslinking with riboflavin and ultraviolet-A light in keratoconus: long-term results. J Cataract Refract Surg 2008;34:796–801
11. Spoerl E, Mrochen M, Sliney D, Trokel S, Seiler T. Safety of UVA-riboflavin cross-linking of the cornea. Cornea 2007;26:385–389
12. Wollensak G, Spörl E, Reber F, Pillunat L, Funk R. Corneal endothelial cytotoxicity of riboflavin/UVA treatment in vitro. Ophthalmic Res 2003;35:324–328
13. Glickman RD. Phototoxicity to the retina: mechanisms of damage. Int J Toxicol 2002;21:473–490
14. Youssef PN, Sheibani N, Albert DM. Retinal light toxicity. Eye (Lond) 2011;25:1–14
15. Barbisan PRT, Viturino MGM, Souto FMS, Tian B, Pinto RDP, Quagliato LB, Nascimento MA, de Castro RS, Arieta CEL. Macular phototoxicity after corneal cross-linking. Clin Ophthalmol 2018;12:1801–1807
Copyright © 2022 Published by Wolters Kluwer on behalf of ASCRS and ESCRS
Data is temporarily unavailable. Please try again soon.