A review of posterior chamber phakic intraocular lenses.

PURPOSE OF REVIEW: Posterior chamber phakic intraocular lenses (pIOLs) are increasing in popularity as a viable alternative to laser refractive surgery. The purpose of this review is to evaluate the recent updates to pIOLs and to assess the advancements and safety of the procedure. RECENT FINDINGS: Accurate lens sizing is the key determinant to suitable vault prediction, advancements to sizing formulae including the use of very high frequency (VHF) digital ultrasound and the application of artificial intelligence and machine learning has led to improved vault prediction and safety. The introduction of the central aquaport has been shown to reduce the formation of cataract and is now adopted in most myopic pIOLs. Recently published studies have demonstrated that pIOLs have an excellent safety profile with no increased risk of retinal detachment or endothelial cell loss. Advancements have led to the introduction of extended depth of focus pIOLs for the correction of presbyopia, further research is required to evaluate the efficacy of new lens designs. SUMMARY: pIOL surgery is experiencing traction with improved lens design and increased lenses choices such as larger optical zone and presbyopic options. Accuracy of implantable collamer lens sizing is paramount to the safety and clinical outcomes, greater predictability is likely to encourage more posterior chamber pIOL users due to fewer sizing related complications.

Epithelial thickness mapping for corneal refractive surgery.

PURPOSE OF REVIEW: As more devices become available that offer corneal epithelial thickness mapping, this is becoming more widely used for numerous applications in corneal refractive surgery. RECENT FINDINGS: The epithelial thickness profile is nonuniform in the normal eye, being thinner superiorly than inferiorly and thinner temporally than nasally. Changes in the epithelial thickness profile are highly predictable, responding to compensate for changes in the stromal curvature gradient, using the eyelid as an outer template. This leads to characteristic changes that can be used for early screening in keratoconus, postoperative monitoring for early signs of corneal ectasia, and for determining whether further steepening can be performed without the risk of apical syndrome following primary hyperopic treatment. Compensatory epithelial thickness changes are also a critical part of diagnosis in irregular astigmatism as these partially mask the stromal surface irregularities. The epithelial thickness map can then be used to plan a trans-epithelial PRK treatment for cases of irregularly irregular astigmatism. Other factors can also affect the epithelial thickness profile, including dry eye, anterior basement membrane dystrophy and eyelid ptosis. SUMMARY: Epithelial thickness mapping is becoming a crucial tool for refractive surgery, in particular for keratoconus screening, ectasia monitoring, hyperopic treatment planning, and therapeutic diagnosis and treatment.

Comparison of Epithelial Thickness Mapping in Normal Corneas with Different Types of Astigmatism.

SIGNIFICANCE: Corneal epithelial thickness mapping has a wide range of applications including screening for keratoconus, screening for anterior basement membrane dystrophy, and assessment of dry eye. It also plays an important role in corneal laser refractive surgery. These all require an understanding of the epithelial thickness profile in a normal cornea as a reference. PURPOSE: This study aimed to compare corneal epithelial thickness along different meridians in normal eyes with different types of corneal astigmatism. METHODS: In this cross-sectional study, corneal topography and epithelial thickness mapping by anterior segment optical coherence tomography was performed for 154 normal eyes. Corneal astigmatism was classified based on the orientation of the steepest corneal meridian (with-the-rule [WTR], against-the-rule [ATR], and oblique) and the amount (low, moderate, and high). On a 9-mm epithelial thickness map, the thickness along the horizontal (180°), vertical (90°), and diagonal (45 and 135°) meridians was calculated. RESULT: With-the-rule, oblique, and ATR astigmatisms were observed in 40, 31, and 29% of the eyes, respectively. In all types of astigmatism, the mean epithelial thickness was greater along the 180° meridian and thinner along the 90° meridian. There was no statistically significant difference in mean thickness along different meridians for WTR (P = .24) and oblique (P = .46) astigmatism, whereas the difference along the 180 and 90° meridians (P = .003) for ATR astigmatism was statistically significant. Also, there was a statistically significant difference in thickness in different types of astigmatism separately for each meridian (P < .05) so that the highest and lowest thicknesses were seen for ATR and oblique astigmatisms, respectively. The epithelial thickness showed no statistical difference in various amounts of astigmatism along different meridians. CONCLUSIONS: Thicker epithelium along the horizontal meridian in ATR may point to the surface regularizing functions of the epithelium to create a regular and smooth surface, but thickness assessment along the different meridians in oblique and WTR showed no curvature-dependent changes.

Improved lenticule shape for hyperopic femtosecond lenticule extraction (ReLEx FLEx): a pilot study.

The aim of this study is to establish and to prove a new lenticule shape for the treatment of hyperopia using a 500 kHz femtosecond laser and the femtosecond lenticule extraction (ReLEx FLEx) technique. Improved lenticule shapes with a large transition zone of at least 2 mm adjusted to the 5.75 mm optical zone were designed. A prospective pilot study on nine eyes of five patients who underwent an uncomplicated FLEx using VisuMax femtosecond laser (Carl Zeiss Meditec AG) for spherical hyperopia was performed. Patients' mean age was 55.5 years, and the preoperative manifest spherical equivalent (SE) was +1.82 D (range +1.25 to +3.00 D). Because of the presbyopic age and in order to compensate for a possible regression, the treatment was aimed at low myopia (mean target SE was -0.88 D with a mean treatment refraction of +2.69 D). At the last follow-up, after 9 months, 33% were within ±0.50 D and 78% within ±1.00 D of intended correction. Thirty-three percent lost one line, and 11% gained one line corrected distance visual acuity (CDVA). On average, the centre of the optical zone was 0.34 ± 0.17 mm from the corneal vertex. No adverse effects were observed. This pilot study confirms that the improved lenticule's design with a large optical and transition zone can achieve good centration and acceptable results for spherical hyperopia using FLEx. The next steps are to extend the study to spherocylindrical hyperopic treatments and to increase the number of eyes for better assessment of refractive outcome.

Comparison of higher-order aberration induction between manual microkeratome and femtosecond laser flap creation.

PURPOSE: To compare the mean change in aberrations produced by a mechanical microkeratome and femtosecond laser. METHODS: This was a retrospective study of 62 consecutive near emmetropic eyes that underwent LASIK and satisfied the following criteria: negligible laser ablation (for spherical equivalent of 0.00 to +0.50 diopters and maximum meridian of +0.50 diopters), and preoperative and at least 3 months postoperative Placido-based corneal aberrometry (ATLAS; Carl Zeiss Meditec, Jena, Germany). Eyes were divided into two groups according to the method used for flap creation: mechanical microkeratome (Hansatome zero-compression microkeratome; Bausch & Lomb, Rochester, NY [mechanical microkeratome group]) or femtosecond laser (VisuMax; Carl Zeiss Meditec, Jena, Germany [femtosecond laser group]). The root mean square total and individual higher-order aberrations were compared between the two groups. RESULTS: Corneas with mechanical flaps, on average, possessed statistically significantly higher trefoil and horizontal coma (P < .001). There was no change in higher-order aberrations, except for spherical aberration in the femtosecond laser group. Average change in coma did not correlate with hinge position. Both groups showed statistically significant changes in spherical aberration (P < .001), although this was most likely due to the small hyperopic ablation performed. CONCLUSIONS: There was greater induction of specific aberrations with the microkeratome than the femtosecond laser. Hinge position did not appear to influence the induction of coma directly, contrary to previously published reports. The difference in aberrations induction between the two groups might be due to the differences in flap thickness profiles.

Comparison of Corneal Epithelial Thickness Measurement Between Fourier-Domain OCT and Very High-Frequency Digital Ultrasound.

PURPOSE: To compare measurements of corneal epithelial thickness using optical coherence tomography (OCT) and very high-frequency digital ultrasound (VHFDU). METHODS: Retrospective analysis of 189 virgin corneas and 175 post-laser refractive surgery (LRS) corneas that had corneal epithelial thickness measurement with RTVue Fourier-domain OCT (Optovue, Inc., Fremont, CA) (tear film included) and Artemis VHFDU (ArcScan Inc., Morrison, CO) (tear film excluded). Averages were calculated for the central 2-mm diameter zone and for two further concentric annuli of 1.5- and 0.5-mm width, each divided into eight sectors. Agreement was analyzed by mean difference (OCT - VHFDU), 95% limits of agreement (LoA) (1.96 standard deviation of the difference), and Bland-Altman analysis. RESULTS: In virgin epithelium, mean central thickness was 53.4 ± 3.20 µm (range: 46 to 62 µm) with OCT and 54.1 ± 2.96 µm (range: 48 to 61 µm) with VHFDU; OCT measured thinnest in 70% with a mean difference of -0.71 µm (95% LoA of ± 3.94 µm, P < .001). In post-LRS epithelium, mean central thickness was 57.9 ± 6.08 µm (range: 42 to 77 µm) with OCT and 60.5 ± 6.47 µm (range: 42 to 79 µm) with VHFDU; OCT measured thinnest in 88%, with a mean difference of -2.48 µm (95% LoA of ± 5.33 µm, P < .001). A larger difference between methods was more common with thicker epithelium. CONCLUSIONS: Corneal epithelial thickness measurements using OCT were found to be slightly thinner than for VHFDU. In contrast to VHFDU, OCT measurement includes the tear film, so the true difference is probably approximately 4 µm more than reported. The difference was greatest inferiorly and higher for post-LRS eyes and in thicker epithelium.

Outcomes for Myopic LASIK With the MEL 90 excimer laser.

PURPOSE: To evaluate the visual outcomes of myopic LASIK performed with the MEL 90 excimer laser (Carl Zeiss Meditec AG, Jena, Germany) using the Triple-A profile with a 500-Hz pulse rate. METHODS: Retrospective analysis of the first 286 myopic LASIK procedures (147 patients) by two experienced surgeons in which the VisuMax femtosecond laser and MEL 90 excimer laser (Carl Zeiss Meditec) were used following a standardized surgical technique. Inclusion criteria were preoperative spherical equivalent refraction (SEQ) up to -10.38 diopters (D), cylinder up to 5.00 D, and corrected distance visual acuity (CDVA) of 20/25 or better. No nomogram adjustments were made. Patients were observed for 3 months. Flap thickness was between 80 and 110 µm and optical zone was between 6 and 7 mm. Standard outcomes analysis was performed. RESULTS: Preoperatively, mean SEQ was -3.83 ± 1.83 D (range: -0.13 to -10.38 D) and mean cylinder was -0.94 ± 0.86 D (range: 0.00 to -5.00 D). Mean age was 36.4 years (range: 18.2 to 74.1 years) with 50% female patients. Of this population, 138 eyes were treated by one surgeon and 148 eyes by another. The mean predictability of SEQ was -0.13 ± 0.34 D (range: -1.00 to +1.00 D). Postoperative SEQ was ± 0.50 D in 88% and ± 1.00 D in 100% of eyes. Preoperative CDVA was 20/20 or better in 97% of eyes. Postoperative uncorrected distance visual acuity was 20/20 or better in 92% and 20/25 or better in 99% of eyes. One line of CDVA was lost in 6% of eyes and no eyes lost two or more lines. There was statistically significant improvement in mesopic contrast sensitivity (CSV-1000) at 3 (P = .021), 6, 12, and 18 (all P ≤.001) cycles per degree. CONCLUSIONS: The MEL 90 excimer laser using the Triple-A ablation profile with a 500-Hz pulse rate was found to achieve a small but real increase in contrast sensitivity and high efficacy for myopia up to -10.00 D and cylinder up to 5.00 D without the need for a nomogram adjustment.

Stromal surface topography-guided custom ablation as a repair tool for corneal irregular astigmatism.

PURPOSE: To illustrate the concept of using a stromal surface topography-guided procedure for therapeutic repair after a complication following primary laser refractive surgery. METHODS: One case example of therapeutic retreatment for short nasal flap after primary LASIK performed in September 2000 is presented. The Artemis very high-frequency digital ultrasound arc-scanner (Arc-Scan, Inc., Morrison, CO) was used to obtain layered corneal thickness including epithelial thickness profile. Corneal front surface elevation was measured with the Orbscan II (Bausch & Lomb, Salt Lake City, UT). Stromal surface height was then calculated by subtracting epithelial thickness data from corneal front surface elevation data and used to calculate the ablation profile applied to the eye. The treatment was performed using the Ultralink system (ArcScan, Inc.), linking the ultrasound corneal thickness data with the Technolas 217c laser (Bausch & Lomb). Postoperative data were available at 30 days and 13 years. RESULTS: One month after treatment, the epithelial thickness map demonstrated that the difference in thickness between the thinnest and thickest points located 2.5-mm nasally was reduced by 26 µm (from 56 to 30 µm). The axial difference map demonstrated an increase in corneal curvature of approximately 4 diopters where the cornea was the flattest nasally, thereby reducing the corneal asymmetry. The anterior elevation map also showed a reduced depression nasally. The patient reported significant improvement of her night vision. CONCLUSIONS: This case example of stromal surface topography-guided treatment demonstrated a significant reduction in the irregularity of the stromal surface and an improvement in the topography, and the visual quality. Stromal surface topography-guided ablation might become the tool of the future for therapeutic repairs because it offers advantages over the current alternative of transepithelial phototherapeutic keratectomy.

Biomechanical Modeling of Femtosecond Laser Keyhole endokeratophakia Surgery.

PURPOSE: To apply a finite element model to endokeratophakia and evaluate anterior and posterior corneal surface changes. METHODS: Spatial elevation data (Pentacam HR; Oculus, Wetzlar, Germany) were obtained for the front and back corneal surfaces of an eye prior to undergoing an endokeratophakia procedure. These were used to warp a spherical template finite element model of the cornea to create a patient-specific finite element mesh and the initial stress distribution was computed with an iterative approach. The finite element model (Optimeyes; Integrated Scientific Services, Biel, Switzerland) included non-linear elastic characteristics of the stroma. The endokeratophakia procedure was recreated in the model: a donor lenticule (-10.50 diopters [D], 5.75-mm zone, 127-µm thick) was inserted into a lamellar pocket (180-µm deep, 6.25-mm diameter) and two 2-mm small incisions were made at 150° and 330°. Anterior and posterior surfaces, computed by the finite element model, were compared to clinical data to assess accuracy and reliability of finite element modeling. RESULTS: The postoperative axial curvature produced by the model closely resembled the patient data; average curvature was 48.01 D clinically and 48.23 D in the simulation, and corneal astigmatism was 3.01 D clinically and 2.88 D in the simulation. The posterior best-fit sphere elevation map also matched the patient data, replicating inward bulging of the posterior surface by approximately 40 µm. Stress distribution modeling predicted a stress increase by 159.94% ± 73% in the cap and a stress decrease by 32.41% ± 21% in the stromal bed. CONCLUSIONS: Finite element modeling of the cornea reproduced the clinically observed anterior and posterior corneal surface changes following an endokeratophakia procedure. This case sets the stage for further study to refine and yield predictive finite element modeling for the evaluation of corneal refractive surgical procedures.

LASIK-induced aberrations: comparing corneal and whole-eye measurements.

PURPOSE: To compare the change in aberrations produced by laser-assisted in situ keratomileusis (LASIK) measured by either front surface corneal aberrometry or Hartmann-Shack whole-eye aberrometry. METHODS: This was a retrospective case series from consecutive patients treated with LASIK at the London Vision Clinic, London, UK. Corneal and ocular wavefront data were collected before and after LASIK for myopia and hyperopia. To control for centration of corneal and whole-eye aberration measurements, only patients with pupil offset (measured as the distance between the corneal vertex and the entrance pupil center) of 0.25 mm or less were recruited. Corneal front surface wavefront was obtained from a Placido-based corneal topography system (Atlas), and whole-eye wavefront was measured by Hartmann-Shack aberrometry (WASCA) before and at least 3 months after surgery. All aberrations were calculated up to the fourth order for the 6-mm pupil diameter. Change in HORMS (higher-order root mean square [RMS]) and all individual Zernike coefficients from the second to the fourth order were compared. RESULTS: One hundred myopic eyes and 50 hyperopic eyes were included for analysis. For myopic LASIK, the change measured by corneal aberrometry or whole-eye aberrometry differed on average by 0.077 µm for HORMS, by 0.024 µm for secondary astigmatism, and by 0.065 µm for spherical aberration (p ≤ 0.003). For hyperopic LASIK, there was no statistically significant difference between the change measured by corneal aberrometry and whole-eye aberrometry for third- and fourth-order individual Zernike coefficients (p ≥ 0.034). However, the change in HORMS measured by whole-eye aberrometry was 0.095 µm greater than that measured by corneal aberrometry (p < 0.001). CONCLUSIONS: This study demonstrated that when measuring the change in higher-order aberrations induced by LASIK, corneal front surface aberrometry measurement was not interchangeable with whole-eye ocular aberrometry.

Lenticule thickness readout for small incision lenticule extraction compared to artemis three-dimensional very high-frequency digital ultrasound stromal measurements.

PURPOSE: To evaluate the accuracy of the lenticule thickness readout for small incision lenticule extraction (SMILE) with the VisuMax femtosecond laser (Carl Zeiss Meditec, Jena, Germany). METHODS: Artemis very high-frequency digital ultrasound (ArcScan, Inc., Morrison, CO) measurements were obtained before and 3 months after SMILE in 70 eyes of 37 patients. The Artemis measured lenticule thickness was calculated as the maximum difference in stromal thickness before and after treatment. Comparative statistics and linear regression analysis were performed between the VisuMax readout lenticule thickness and Artemis measured maximum stromal change. Central epithelial thickness was measured and a similar analysis was performed using corneal thickness. Variability of the data were compared to ablation depths for a matched group of eyes from a previously published LASIK population treated with the MEL80 excimer laser (Carl Zeiss Meditec). RESULTS: The mean maximum myopic meridian treated was -7.81 ± 2.33 diopters (range: -2.25 to -12.50 diopters). On average, the VisuMax readout lenticule depth was 8.2 ± 8.0 µm thicker (range: -8 to +29 µm) than the Artemis measured stromal change (P < .001). On average, central epithelial thickness was 15.0 ± 5.2 µm thicker (range: 5 to 30 µm) after the procedure. The VisuMax readout lenticule thickness was 23.2 ± 10.9 µm thicker (range: +5 to +49 µm) than the Artemis measured corneal thickness change. The R(2) of 0.868 for the SMILE group was higher than 0.738 for the LASIK group (P = .015). CONCLUSIONS: The accuracy of SMILE lenticule thickness was found to be higher than actual measured stromal thickness change; however, predictability for SMILE lenticule thickness appeared higher than predictability for excimer laser ablation depth. The VisuMax readout lenticule depth was found to be 8 µm thicker than the achieved stromal change. This can be partly explained by alignment errors between preoperative and postoperative scans. However, this appears to show some biomechanical changes occur after SMILE.

Outcomes of small incision lenticule extraction (SMILE) in low myopia.

PURPOSE: To report the visual and refractive outcomes of small incision lenticule extraction for low myopia using the VisuMax femtosecond laser (Carl Zeiss Meditec, Jena, Germany). METHODS: A retrospective analysis of 120 consecutive small incision lenticule extraction procedures was performed for low myopia. Inclusion criteria were preoperative spherical equivalent refraction up to -3.50 diopters (D), cylinder up to 1.50 D, and corrected distance visual acuity of 20/20 or better. Outcomes analysis was performed for all eyes with 1-year follow-up according to the Standard Graphs for Reporting Refractive Surgery, and also including mesopic contrast sensitivity. RESULTS: One-year data were available for 110 eyes (92%). Preoperatively, mean spherical equivalent refraction was -2.61 ± 0.54 D (range: -1.03 to -3.50 D) and mean cylinder was 0.55 ± 0.38 D (range: 0.00 to 1.50 D). Postoperatively, mean spherical equivalent refraction was -0.05 ± 0.36 D (range: -0.94 to +1.25 D) and mean cylinder was ± 0.50 D in 84% and ± 1.00 D in 99% of eyes. Uncorrected distance visual acuity was 20/20 or better in 96% of eyes and 20/25 or better in 100% of eyes. One line of corrected distance visual acuity was lost in 9%, but no eyes lost two or more lines. There was an initial overcorrection in mean spherical equivalent refraction on day 1 (+0.37 D) as expected, which regressed to +0.10 D at 1 month and -0.05 D at 3 months, after which stability was reached (mean spherical equivalent refraction was -0.05 D at 1 year). Contrast sensitivity at 1 year was slightly increased at 3, 6, 12, and 18 cycles per degree (P < .05). CONCLUSIONS: Small incision lenticule extraction for low myopia was found to be safe and effective with outcomes similar to those previously reported for LASIK.

Transepithelial phototherapeutic keratectomy protocol for treating irregular astigmatism based on population epithelial thickness measurements by artemis very high-frequency digital ultrasound.

PURPOSE: To report the outcomes of transepithelial phototherapeutic keratectomy (TE-PTK) in the treatment of irregular astigmatism and define a standard treatment protocol based on population epithelial thickness measurements. METHODS: Retrospective analysis of 41 TE-PTK procedures in cases of irregular astigmatism after refractive surgery or with corneal irregularities. The TE-PTK ablations were performed according to preoperative epithelial thickness maps obtained using an Artemis very high-frequency digital ultrasound arc-scanner (ArcScan, Inc., Morrison, CO). Visual and refractive outcomes were analyzed 12 months after the procedure. The efficacy of the stromal surface regularization was evaluated as the change in epithelial thickness range (ie, the difference between minimum and maximum epithelial thickness). A further refractive ablation was performed immediately after TE-PTK in 12 eyes. RESULTS: Corrected distance visual acuity was improved by one or more lines in 58% of eyes, whereas 1 eye lost one line and no eyes lost two lines. Significant stromal surface regularization was achieved with epithelial thickness range reduced on average from 41 to 29 µm. There was an unpredictable refractive shift in the TE-PTK only group with a change of more than 0.50 diopter (D) in 59% of eyes. Refractive accuracy in the TE-PTK with refractive ablation group was reasonably good, although there were two outliers (18%) outside ±1.00 D. A therapeutic window was identified between the highest thinnest epithelium of 51 µm and lowest thickest epithelium of 60 µm. CONCLUSIONS: TE-PTK can be a safe and effective method of reducing stromal surface irregularities by taking advantage of the natural masking effect of the epithelium. There can be a significant refractive shift due to lenticular epithelial masking. A standard protocol of targeting an initial TE-PTK ablation for 55 µm will likely achieve breakthrough of the thinnest epithelium without total epithelial removal, allowing the treatment to be continued in a stepwise fashion.

Enhancing the Automated Detection of Implantable Collamer Lens Vault Using Generative Adversarial Networks and Synthetic Data on Optical Coherence Tomography.

PURPOSE: To investigate the efficacy of incorporating Generative Adversarial Network (GAN) and synthetic images in enhancing the performance of a convolutional neural network (CNN) for automated estimation of Implantable Collamer Lens (ICL) vault using anterior segment optical coherence tomography (AS-OCT). METHODS: This study was a retrospective evaluation using synthetic data and real patient images in a deep learning framework. Synthetic ICL AS-OCT scans were generated using GANs and a secondary image editing algorithm, creating approximately 100,000 synthetic images. These were used alongside real patient scans to train a CNN for estimating ICL vault distance. The model's performance was evaluated using statistical metrics such as mean absolute percentage error (MAPE), mean absolute error (MAE), root mean squared error (RMSE), and coefficient of determination (R2) for the estimation of ICL vault distance. RESULTS: The study analyzed 4,557 AS-OCT B-scans from 138 eyes of 103 patients for training. An independent, retrospectively collected dataset of 2,454 AS-OCT images from 88 eyes of 56 patients, used prospectively for evaluation, served as the test set. When trained solely on real images, the CNN achieved a MAPE of 15.31%, MAE of 44.68 µm, and RMSE of 63.3 µm. However, with the inclusion of GAN-generated and algorithmically edited synthetic images, the performance significantly improved, achieving a MAPE of 8.09%, MAE of 24.83 µm, and RMSE of 32.26 µm. The R2 value was +0.98, indicating a strong positive correlation between actual and predicted ICL vault distances (P < .01). No statistically significant difference was observed between measured and predicted vault values (P = .58). CONCLUSIONS: The integration of GAN-generated and edited synthetic images substantially enhanced ICL vault estimation, demonstrating the efficacy of GANs and synthetic data in enhancing OCT image analysis accuracy. This model not only shows potential for assisting postoperative ICL evaluations, but also for improving OCT automation when data paucity is an issue. [J Refract Surg. 2024;40(4):e199-e207.].

Creating realistic anterior segment optical coherence tomography images using generative adversarial networks.

AIMS: To develop a generative adversarial network (GAN) capable of generating realistic high-resolution anterior segment optical coherence tomography (AS-OCT) images. METHODS: This study included 142 628 AS-OCT B-scans from the American University of Beirut Medical Center. The Style and WAvelet based GAN architecture was trained to generate realistic AS-OCT images and was evaluated through the Fréchet Inception Distance (FID) Score and a blinded assessment by three refractive surgeons who were asked to distinguish between real and generated images. To assess the suitability of the generated images for machine learning tasks, a convolutional neural network (CNN) was trained using a dataset of real and generated images over a classification task. The generated AS-OCT images were then upsampled using an enhanced super-resolution GAN (ESRGAN) to achieve high resolution. RESULTS: The generated images exhibited visual and quantitative similarity to real AS-OCT images. Quantitative similarity assessed using FID scored an average of 6.32. Surgeons scored 51.7% in identifying real versus generated images which was not significantly better than chance (p value >0.3). The CNN accuracy improved from 78% to 100% when synthetic images were added to the dataset. The ESRGAN upsampled images were objectively more realistic and accurate compared with traditional upsampling techniques by scoring a lower Learned Perceptual Image Patch Similarity of 0.0905 compared with 0.4244 of bicubic interpolation. CONCLUSIONS: This study successfully developed and leveraged GANs capable of generating high-definition synthetic AS-OCT images that are realistic and suitable for machine learning and image analysis tasks.

Coaxially sighted corneal light reflex versus entrance pupil center centration of moderate to high hyperopic corneal ablations in eyes with small and large angle kappa.

PURPOSE: To determine whether centering ablations on the coaxially sighted corneal light reflex (CSCLR) in eyes with large angle kappa leads to poor visual outcomes when compared to patients with eyes with negligible angle kappa that by default would be centered on the entrance pupil. In eyes with no angle kappa, the CSCLR coincides with the entrance pupil center, whereas eyes with large angle kappa possess an offset between the CSCLR and the entrance pupil center. METHODS: This study was a retrospective case series of consecutive patients treated by hyperopic LASIK using the MEL80 excimer laser (Carl Zeiss Meditec, Jena, Germany). All ablations were centered on the CSCLR using the standard non-wavefront-guided ablation profile. Angle kappa was classified according to pupil offset defined as the distance in the corneal plane between the entrance pupil center and the corneal vertex. Eyes were divided into two discrete groups according to the pupil offset: small angle kappa for pupil offset of 0.25 mm or less (n = 30) and large angle kappa for pupil offset of 0.55 mm or greater (n = 30). Safety, accuracy, cylinder vector analysis, contrast sensitivity, vertex centered corneal aberrations, entrance pupil centered whole eye aberrometry, and night vision disturbances were compared between the two groups. RESULTS: There were no statistically significant differences in safety, accuracy, induced astigmatism, contrast sensitivity, or night vision disturbances between the two groups. There was also no statistically significant difference between groups for vertex centered corneal aberrations; however, as expected, coma was higher in the large angle kappa group for entrance pupil centered aberrometry because the treatment had been centered on the CSCLR rather than the entrance pupil center. CONCLUSION: Refractive outcomes of high hyperopic LASIK were not found to be worse for eyes where ablation was centered more than 0.55 mm from the entrance pupil as determined by CSCLR in eyes with large angle kappa. The absence of poor quality visual outcomes in cases, which by entrance pupil centration are considered significantly "decentered," supports the notion that centration relative to the CSCLR may be preferable. This provides evidence that refractive corneal ablation should not be systematically aligned with the entrance pupil center.

Improved effectiveness of transepithelial PTK versus topography-guided ablation for stromal irregularities masked by epithelial compensation.

PURPOSE: To demonstrate improved effectiveness of transepithelial phototherapeutic keratectomy (PTK) where topography-guided ablation is degraded by epithelial compensation for localized irregularities. METHODS: Artemis very high-frequency digital ultrasound (ArcScan Inc., Morrison, CO) scanning was performed on five eyes (four patients) after LASIK in which a truncated nasal flap was created but the ablation was still performed, resulting in irregular astigmatism and double vision. The stromal ablation pattern of a transepithelial PTK ablation was modeled and compared to topography-guided ablation simulations. Artemis-assisted transepithelial PTK procedures were performed in three eyes (two patients). RESULTS: In all cases, Artemis very high-frequency digital ultrasound demonstrated a crevice on the stromal surface caused by ablation on the underside of the flap nasally, which had been compensated for by epithelial thickening, thus masking a significant proportion of the stromal surface irregularity from corneal surface topography. The transepithelial PTK ablation pattern appeared logically to reduce the ridge nasal to the crevice, whereas topography-guided ablation patterns demonstrated poor targeting of the stromal ridge. Artemis-assisted transepithelial PTK greatly reduced the stromal irregularity, resulting in smoother epithelium and topography. In one eye, the epithelium had compensated fully for the irregularity and the patient reported 95% subjective improvement. One patient (two eyes) reported little change in visual symptoms despite significant anatomical regularization due to the partial nature of epithelial compensation, demonstrating that multiple procedures may often be required. CONCLUSION: Compensatory epithelial remodeling masks stromal irregularities from the corneal surface, meaning that topography-guided (or wavefront-guided) ablations may not provide an adequate treatment option and may potentially worsen the irregularity. The effectiveness of transepithelial PTK will depend on the proportion of epithelial masking due to remodeling, related to the rate of change of curvature of the stromal surface.

Mathematical model to compare the relative tensile strength of the cornea after PRK, LASIK, and small incision lenticule extraction.

PURPOSE: To develop a mathematical model to estimate the relative differences in postoperative stromal tensile strength following photorefractive keratectomy (PRK), LASIK, and small incision lenticule extraction (SMILE). METHODS: Using previously published data where in vitro corneal stromal tensile strength was determined as a function of depth, a mathematical model was built to calculate the relative remaining tensile strength by fitting the data with a fourth order polynomial function yielding a high correlation coefficient (R(2) = 0.930). Calculating the area under this function provided a measure of total stromal tensile strength (TTS), based only on the residual stromal layer for PRK or LASIK and the residual stromal layers above and below the lenticule interface for SMILE. RESULTS: Postoperative TTS was greatest after SMILE, followed by PRK, then LASIK; for example, in a 550-µm cornea after 100-µm tissue removal, postoperative TTS was 75% for SMILE (130-µm cap), 68% for PRK, and 54% for LASIK (110-µm flap). The postoperative TTS decreased for thinner corneal pachymetry for all treatment types. In LASIK, the postoperative TTS decreased with increasing flap thickness by 0.22%/µm, but increased by 0.08%/µm for greater cap thickness in SMILE. The model predicted that SMILE lenticule thickness could be approximately 100 µm greater than the LASIK ablation depth and still have equivalent corneal strength (equivalent to approximately 7.75 diopters). CONCLUSIONS: This mathematical model predicts that the postoperative TTS is considerably higher after SMILE than both PRK and LASIK, as expected given that the strongest anterior lamellae remain intact. Consequently, SMILE should be able to correct higher levels of myopia.

Accuracy and reproducibility of cap thickness in small incision lenticule extraction.

PURPOSE: To evaluate the accuracy and reproducibility of cap thickness for small incision lenticule extraction (SMILE) with the VisuMax femtosecond laser (Carl Zeiss Meditec, Inc., Jena, Germany). METHODS: Artemis very high-frequency digital ultrasound (ArcScan Inc., Morrison, CO) measurements were obtained before and 3 months after SMILE in 70 eyes of 37 patients with intended cap thickness between 80 and 140 µm. True cap thickness at the time of creation was calculated as the addition of the preoperative epithelial thickness and the postoperative stromal component of the flap and mapped for the central 5-mm diameter zone. Cap thickness accuracy was calculated as the difference between the mean and intended cap thickness. Reproducibility was evaluated as the cap thickness standard deviation between eyes. Accuracy and reproducibility of cap thickness were calculated for all eyes and grouped by intended cap thickness. Within-eye variation in cap thickness was calculated as the standard deviation of all data within the central 5-mm diameter zone. RESULTS: Mean cap thickness accuracy was between -2.3 and 6.5 µm and was -0.7 µm centrally (range: -11 to +14 µm), including all eyes. Cap thickness reproducibility was 6 µm or less for the majority of the central 5-mm diameter zone and was 4.4 µm centrally. Cap thickness accuracy and reproducibility were similar for different intended cap thicknesses. Average within-eye variation in cap thickness was 4.3 µm. CONCLUSIONS: SMILE cap thickness using the VisuMax femtosecond laser was found to be accurate and reproducible across the central 5-mm diameter zone for intended cap thicknesses over the range of 80 to 140 µm.

Comparison of postoperative vault height predictability using white-to-white or sulcus diameter-based sizing for the visian implantable collamer lens.

PURPOSE: To compare vault height predictability of Implantable Collamer Lens (ICL; Staar Surgical) sizing using a sulcus diameter-based formula or the manufacturer-recommended white-to-white-based method. METHODS: In 50 myopic eyes, ICL size was calculated using both a formula including sulcus diameter and the traditional formula based on white-to-white diameter. Sulcus diameter was measured using Artemis 2 very high-frequency (VHF) digital ultrasound (ArcScan Inc). Implantation was based on the sulcus diameter derived size. Actual postoperative vault height achieved was measured by VHF digital ultrasound scanning. Circle segment trigonometry was used to calculate the vault height that would have resulted had lens sizing been based on the white-to-white formula. RESULTS: The same lens size would have been used in 60% of eyes, a smaller lens would have been used in 34% of eyes and a larger lens in 6% of eyes had lens sizing been based on the white-to-white formula. Mean vault for eyes with lenses sized using the sulcus diameter formula was 0.37±0.16 mm (range: 0.08 to 0.92 mm), with 2% <0.09 mm, the recognized low-vault height for risk of cataract. Circle segment trigonometry predicted that the vault height would have been 0.24±0.28 mm (range: -0.31 to 0.92 mm), with 26% <0.09 mm had lens sizing been based on the white-to-white formula. CONCLUSIONS: Significantly better predictability of postoperative vault height was achieved by including sulcus diameter into the ICL sizing formula compared with using the traditional white-to-white-based formula.