Umeå University's logo

Objective: To evaluate the effects of customized corneal collagen cross-linking (CXL) on higher-order aberrations (HOAs) in keratoconus (KC): vertical coma (VC), horizontal coma (HC), spherical aberration (SA), trefoil (TF) and astigmatism, compared with the same effects in healthy eyes undergoing CXL for low-grade myopia.

Methods: This mixed-designed study included 38 eyes of 38 patients with KC, treated and followed prospectively, who received customized epi-on CXL in high oxygen, and a retrospective control group of 23 eyes from 23 patients who underwent central 4-mm CXL treatment for low-grade myopia. VC, HC, SA, TF and keratometry values were obtained from Pentacam HR® measurements at baseline and at 1, 6, 12 and 24 months post-treatment. Statistical analyses included paired T-tests for changes over time and Pearson correlation tests to assess relationships between aberrations, best spectacle-corrected and low-contrast visual acuities (BSCVA and LCVA, respectively) and CXL parameters.

Results: Reduced HOAs and improved visual acuities were observed in KC. A 20% reduction in VC was observed at 24 months (from −1.82 ± 1.15 μm to −1.46 ± 1.01 μm; 95% CI: [0.155, 0.629], p = 0.002), while a 17% reduction in HC was observed at 12 months (from −0.35 ± 0.56 μm to −0.29 ± 0.62 μm; 95% CI: [0.003, 0.096], p = 0.037). A positive correlation was found between baseline VC and the level of improvement in VC at 24 months (R2 = 0.200, p = 0.015). SA increased by 126% at 24 months (from −0.21 ± 0.62 μm to 0.054 ± 0.52 μm; 95% CI: [0.143, 0.347], p ≤ 0.001). TF and astigmatism did not alter from the treatment. In myopia, the natural positive SA increased by 57% post-treatment (from 0.14 ± 0.061 μm to 0.22 ± 0.076 μm at 24 months; 95% CI: [0.067, 0.098], p ≤ 0.001), while changes in VC and HC were minor and BSCVA remained stable.

Conclusion: Customized CXL effectively reduces HOAs in KC. For VC the improvement is larger in cases with higher preoperative VC, indicating that the concept of customization has its intended effect. Accordingly, SA and visual acuities improve in KC whereas CXL for low-grade myopia tends to increase corneal SA unfavourably.

Purpose: To compare clinical outcomes of customized transepithelial (epi-on) corneal crosslinking (CXL) in high oxygen and customized CXL with epithelial removal (epi-off) in room air for keratoconus (KC).

Setting: Umeå University Hospital, Umeå, Sweden.

Design: Prospective, randomized, single-masked, intraindividually comparing study.

Methods: 32 participants with bilateral progressive KC were treated with bilateral customized topography-guided CXL, 30 mW/cm2; 7.2 to 15 J/cm2and were randomized to epi-on in one eye (32 eyes) and epi-off in the fellow eye (32 eyes). Uncorrected distance visual acuity (UDVA) and corrected distance visual acuity (CDVA), maximal keratometry (Kmax), subjective ocular discomfort, low-contrast visual acuities (LCVAs) at 10% and 2.5% contrast, ocular and anterior corneal wavefront aberrations, manifest refractive spherical equivalent, endothelial cell count (ECC), and adverse events were assessed through 24 months.

Results: Both treatments showed improvements at 24 months in UDVA; -0.16 ± 0.24 (P <.001) and -0.13 ± 0.20 logMAR (P =.006), respectively, CDVA; -0.10 ± 0.11 (P <.001) and -0.10 ± 0.12 (P =.001), Kmax; -1.74 ± 1.31 (P <.001) and -1.72 ± 1.36 D (P <.001). LCVA 10% improved for both protocols (P <.001), but LCVA 2.5% improved for epi-on CXL only (P =.001). ECC was unaltered, and no adverse events occurred. The epi-on eyes had significantly less discomfort symptoms during the whole first week posttreatment (P <.05).

Conclusions: High-oxygen customized epi-on CXL is a viable alternative to room air customized epi-off CXL, with faster improvements in CDVA and LCVA and less early ocular discomfort.

Purpose: To analyse the temperature of the corneal surface in keratoconus during corneal customized crosslinking (CXL) with a preserved epithelium (epi-on) under oxygen flow, and epi-off CXL in room air, and to assess the effect of pre-heating the oxygen.

Methods: This masked, intra-individual comparing randomized study included 14 participants with bilateral progressive keratoconus treated with bilateral CXL: one eye with epi-on CXL under a flow of 2.5 L/min oxygen; the fellow eye with epi-off CXL in room air. In a second setting involving 12 healthy participants, room-tempered oxygen was flushed over one eye and oxygen pre-heated to 37°C over the fellow eye. The corneal surface temperature was assessed with infrared photography.

Results: A reduction in corneal surface temperature was seen from the pre-treatment application of topical riboflavin in the epi-off group (−1.1 ± 1.0°C, p < 0.001). The temperature increased during the first half of the CXL treatment in both groups (+0.7 ± 1.2°C, p = 0.041 for epi-on; +0.7 ± 0.9°C, p = 0.023 for epi-off CXL, respectively). In epi-on CXL an overall temperature increase was seen during the treatment (+0.8 ± 1.2°C, p = 0.016). In the second setting, pre-heating the oxygen rendered a surface temperature increase of +1.8 ± 0.2°C (p < 0.001).

Conclusion: In epi-off CXL, the application of topical room-tempered riboflavin decreases the corneal surface temperature, likely due to increased evaporation. A slight temperature increase is seen during CXL with both epi-on and epi-off CXL, albeit far below the corneal safety limit. The corneal temperature can, however, be increased by applying pre-heated oxygen, a possible approach to modify or augment the treatment effect in CXL.

Purpose: To compare two annular epithelium-on (epi-on) high oxygen photorefractive intrastromal cross-linking (PiXL) illuminations protocols for treatment of low-grade myopia.

Methods: In this randomized, single-masked, intra-individually comparative study, healthy individuals with bilateral low-grade myopia (manifest refractive spherical equivalent (MRSE) −0.75 diopters (D) to −2.50 D) were treated with high oxygen epi-on PiXL. One eye was randomized to receive pulsed accelerated 365-nm ultraviolet-A illumination in a central annular zone of 4.0 mm (1 second on, 1 second off; 30 mW/cm2), and the fellow eye in a 3.5 mm annular zone (0.5 second on, 1 second off; 45 mW/cm2). Uncorrected distance visual acuity (UDVA), MRSE, low-contrast visual acuity (LCVA), best spectacle corrected visual acuity (BSCVA), endothelial cell count (ECC) and Scheimpflug light scattering depths were assessed through 24-month follow-up.

Results: Twenty-seven participants (54 eyes) were included. The 3.5 mm protocol rendered less subjective ocular discomfort posttreatment and a larger improvement than the 4.0 mm protocol in UDVA: −0.52 (−0.72, −0.32) logMAR (medians and interquartile ranges, IQR) and −0.38 (−0.50, −0.22), p = 0.003 and MRSE: +1.25 D (0.75, 1.50) and +1.0 (0.75, 1.0), p = 0.037. The transient reduction in LCVA was larger with the 3.5 mm protocol (p < 0.01). No adverse events, and no reductions in ECC or BSCVA were noted.

Conclusion: Epi-on PiXL in high oxygen reduces myopia in healthy eyes. A larger reduction of myopia and less early posttreatment subjective ocular discomfort can be seen with a smaller treatment zone, but likely at the expense of a transient decrease in low-contrast visual acuity.

Purpose: To assess the effect of two high-oxygen epi-on PiXL treatments for low-grade myopia.

Methods: This prospective, randomized, intra-individually comparing, single-masked study included 23 healthy volunteers (46 eyes) aged 18-35 years with mild myopia, -0.75 to -2.50 D manifest refractive spherical equivalent (MRSE). One eye was randomized to a 4.0-mm homogenous treatment zone and the fellow eye to a 4.0-mm annular zone (16:40 min at 30 mW/cm(2), fluence 15 J/cm(2)). Uncorrected distance visual acuity (UDVA), MRSE, best spectacle-corrected visual acuity (BSCVA), Scheimpflug light scattering depths, mean keratometry (K-mean) and endothelial cell count (ECC) were assessed through 24 months.

Results: Similar improvements in UDVA were seen for the homogeneous and annular protocols at 1 month: -0.52 (-0.59, -0.39) and -0.49 (-0.59, -0.39) logMAR, respectively (medians and interquartile ranges, IQR), p = 0.91, and MRSE: +1.0 D (0.94, 1.31) and +1.0 D (0.69, 1.25), p = 0.17. Light scattering depths were 496 (465, 527) and 349 (247, 378) mu m, respectively, and the reduction in mean keratometry was -0.8 D (-1.1, -0.7) and 0 D (-0.1, 0.1), p < 0.001. The treatment effect remained stable throughout 24 months. At 1 week, the participants reported less ocular discomfort with the annular protocol. No reductions were seen in BSCVA or ECC. No adverse events were reported.

Conclusion: PiXL can reduce low-grade myopia and improve uncorrected vision in healthy eyes. The initial ocular discomfort may be reduced with an annular treatment zone. Further studies are needed to optimize PiXL treatment parameters.

Purpose: To evaluate photorefractive intrastromal cross‐linking (PiXL) treatment for low‐grade myopia, comparing three treatment protocols.

Methods: Healthy individuals, 25.6 ± 3.6 years of age, with low‐grade myopia underwent epi‐on PiXL with either: 4‐mm zone treated in high oxygen environment (4 mm‐HIGH; n = 15), 4‐mm/room air (4‐mm LOW; n = 6), or 6‐mm/high oxygen (6‐mm HIGH; n = 6). Efficacy was determined by change in uncorrected visual acuity (UCVA), manifest refractive spherical equivalent (MRSE) and corneal curvature (K_mean) over a 12‐month follow‐up. Safety was determined by best spectacle corrected visual acuity (BSCVA), corneal endothelial cell loss and registration of side‐effects.

Results: Twenty‐seven subjects were included. Due to insufficient effect with the 4‐mm LOW treatment and an unacceptable degree of initial light sensitivity/ocular irritation in the 6‐mm HIGH group, the inclusions to these treatments were stopped after inclusion of 6 patients in each group. The 4‐mm HIGH treatment showed a significantly larger improvement in UCVA (−0.45 ± 0.27 LogMAR) and MRSE (+0.99 ± 0.44 D) at 1, 6 and 12 months compared with the 4‐mm LOW treatment (p < 0.05). At 12 months posttreatment, endothelial cell count and BSCVA were unaltered. More initial side‐effects were noted with the 6‐mm HIGH treatment, compared with the 4‐mm HIGH treatment (p < 0.05).

Conclusion: Epi‐on PiXL may become a safe and effective non‐ablative treatment for low‐grade myopia. The effect is augmented by high oxygen environment and remains stable for 12 months. The initial ocular irritation is acceptable with a 4‐mm treatment zone. The present results justify further clinical studies on PiXL, including refinements of the technique and long‐term results.

Purpose To investigate the central and paracentral astigmatism and the significance of centration and measurement zone diameter compared to a 3-mm pupil-centred measurement zone in keratoconus and in healthy eyes.

Methods Twenty-eight right eyes from 28 KC patients with an inferotemporal cone were selected according to specified criteria based on Oculus Pentacam HR (R) measurements and were matched with healthy control eyes. The flat (K1) and steep (K2) keratometry readings were registered from the Total Corneal Refractive Power' (TCRP) display as well as the anterior and posterior corneal astigmatism displays (ACA and PCA, respectively). Astigmatic power vectors KP0 and KP45 were calculated and analysed for a 6-mm and two 3-mm zones centred on the corneal apex and the pupil, and for 8 paracentral 3-mm zones.

Results The astigmatism was generally higher in KC. Many astigmatic values in KC differed between the 3-mm pupil-centred and the 3- and 6-mm apex-centred zones in KC. In the controls, no corresponding differences between measurement zones were seen, apart from PCA, which differed. The magnitude and direction of KP0 and KP45 varied greatly between the paracentral measurements in KC.

Conclusion Centration and measurement zone diameter have great impacts on the astigmatic values in KC. A small pupil-centred measurement zone should be considered when evaluating the astigmatism in KC.

PURPOSE: To assess the refractive improvements and the corneal endothelial safety of an individualised topography-guided regimen for corneal crosslinking in progressive keratoconus.

METHODS: An open-label prospective randomised clinical trial was performed at the Department of Clinical Sciences, Ophthalmology, Umeå University Hospital, Umeå, Sweden. Thirty-seven patients (50 eyes) with progressive keratoconus planned for corneal crosslinking were included. The patients were randomised to topography-guided crosslinking (photorefractive intrastromal crosslinking (PiXL); n=25) or uniform 9 mm crosslinking (corneal collagen crosslinking (CXL); n=25). Visual acuity, refraction, keratometry (K1, K2 and Kmax) and corneal endothelial morphometry were assessed preoperatively and at 1, 3, 6 and 12 months postoperatively. The PiXL treatment involved an asymmetrical treatment zone centred on the area of maximum corneal steepness with treatment energies ranging from 7.2 to 15.0 J/cm(2); the CXL treatment was a uniform 9 mm 5.4 J/cm(2) pulsed crosslinking. The main outcome measures were changes in refractive errors and corneal endothelial cell density.

RESULTS: The spherical refractive errors decreased (p<0.05) and the visual acuity improved (p<0.01) at 3, 6 and 12 months after PiXL, but not after CXL. The between-groups differences, however, were not significant. K2 and Kmax decreased at 3, 6 and 12 months after PiXL (p<0.01), but not after CXL (p<0.01 when comparing the two treatments). No corneal endothelial cell loss was seen after either treatment.

CONCLUSIONS: Individualised topography-based crosslinking treatment centred on the ectatic cone has the potential to improve the corneal shape in keratoconus with decreased spherical refractive errors and improved visual acuity, without damage to the corneal endothelium.

PURPOSE: To compare objective refraction using small-zone eccentric laser ray tracing (LRT) wavefront aberrometry to standard autorefraction in keratoconus (KC), and whether the visual acuities achieved with these refractions differ from corresponding values in healthy eyes.

METHODS: Twenty-nine eyes of 29 patients with KC and 29 eyes of 29 healthy controls were included in this prospective unmasked case-control study. The uncorrected (UCVA) and spectacle-corrected (SCVA) Early Treatment Diabetic Retinopathy Study (ETDRS) visual acuities based on refractions derived from LRT in central and four eccentric zones were compared to those achieved with standard autorefraction. The spherical equivalent (M) and two astigmatic power vectors (C0 and C45) were calculated for all refractions. Pentacam HR(®) was used to generate keratometry readings of the corresponding zones.

RESULTS: In KC, the refraction from the upper nasal zone rendered a higher SCVA than the standard autorefraction more often than in the controls (p < 0.001). There were no significant variation in M between the different LRT measurement points in the control group, but central data provided the best SCVA. The UCVA:s and SCVA:s were worse in KC, and the KC eyes showed inferior myopia and superior hyperopia. Multiple refractions rendered similar SCVA:s in KC. Pentacam HR(®) showed higher keratometry readings infero-temporally, but also lower readings supero-nasally, compared to controls.

CONCLUSION: In KC, eccentric LRT measurements gave better SCVA than standard autorefraction more often than in healthy eyes. Eccentric LRT may become a valuable tool in the demanding task of subjective refraction in KC.