Refractive Surgery Candidates Research Articles

Enhancing ectasia screening using advanced AS-OCT: a case series of challenging refractive candidates.

Ectasia screening in candidates for laser refractive surgery is mandatory during preoperative evaluation. Despite the availability of modern imaging techniques, refractive surgeons often face borderline decisions when patients present with suspicious tomographic findings. This case series presents refractive candidates with suspicious tomographic findings and demonstrates how to interpret them using Scheimpflug imaging and additional anterior segment optical coherence tomography (AS-OCT). Department of Ophthalmology, University Hospital, LMU Munich. This case series examines six potential candidates for refractive surgery with a mean age of 29.2 ± 3.9 years, whose corneal assessments using Scheimpflug imaging raised suspicion for ectasia. Each candidate was additionally examined with AS-OCT and reevaluated. The mean manifest subjective spherical equivalent was -3.67 ± 1.8 diopters. The total corneal thickness measured 537 µm ± 30 µm at its thinnest point. None of the candidates had any reported underlying corneal or ophthalmic diseases, and slit lamp examinations revealed no abnormal morphological findings. Both Scheimpflug imaging and AS-OCT are appropriate tools for screening refractive candidates for ectasia. While topographic and elevation analyses yielded comparable results regarding corneal structure, the epithelial mapping provided by AS-OCT played a critical role in decision-making for cases with borderline tomographic findings. Establishing a global consensus on the use of epithelial mapping in ectasia screening is necessary.

AvaGen Genetic Testing versus Ocular Screening Assessments Including the Keratoconus Severity Score (KSS) and Randleman Ectasia Risk Score System (ERSS) in Refractive Surgery Candidates.

To determine whether the AvaGen (AG) Genetic Eye Test provided additional information for screening for the presence of keratoconus (KC) and assessing KC risk in refractive surgery candidates, as compared to the Keratoconus Severity Score (KSS) and Randleman Ectasia Risk Score System (ERSS). This retrospective study analyzed patients seeking refractive surgery at an eye clinic in the United States between January 2022 and July 2023. The inclusion criteria encompassed those with a family history of KC, positive KC indices, or both. Corneal evaluations and demographic information were recorded and analyzed. KSS and ERSS criteria were utilized to evaluate postoperative KC and ectasia risk, respectively. Patients were categorized on how the AG genetic test compared to KSS and ERSS criteria. Clinicians assessed topographic indices, criteria scoring, and AG testing to deliver a definitive surgical recommendation. Among the 19 patients evaluated for ectasia risk, AG testing showed lower KC risk than ocular screening in three patients (15.8%), equal risk in three patients (15.8%), and higher risk in 13 patients (68.4%). The mean AG scores were 45.7 ± 7.0, 49.0 ± 3.46, and 61 ± 13.0 for these respective categories. The most frequently identified KC risk genes were ADAMTS18, COL2A1, and COL4A1. The AG test modified the physician's recommendation for refractive surgery in nine cases (47.4%). Despite the promising application of AG testing for assessing KC risk, further research and development are needed to enhance its applicability for screening refractive surgery candidates, in addition to standard ocular screening approaches.

The Effects of Whole-Corneal and Whole-Eye Higher Order Aberrations on the Discrepancy Between Refractive and Corneal Astigmatism in Otherwise Healthy Candidates for Refractive Surgery.

To evaluate the effect of different whole-corneal and whole-eye higher order aberrations (HOAs) on levels of axis discrepancy. This was a retrospective study including healthy candidates for refractive surgery, with one eye being randomly selected. A total of 360 eyes were included. Whole-corneal and whole-eye HOAs were measured twice with a Pentacam AXL Wave (Oculus Optikgeräte GmbH), and subjective manifest refraction was obtained. Axis discrepancy was defined as the absolute difference between Total Corneal Refractive Power flat keratometry axis and manifest refractive axis. Two multiple linear regression models that sought to explore the effect of HOAs in predicting axis discrepancy while adjusting for corneal and refractive confounders were built. Mean age was 29.1 ± 5.8 years and 63.9% of the patients were women. Mean manifest sphere and cylinder were -3.09 ± 2.36 and -1.45 ± 1.37 diopters (D), respectively. Mean cylinder axis discrepancy was 14.4 ± 14.5°. On multiple linear regression, the only variables significantly associated with axis discrepancy were corneal cylinder and corneal lower order aberrations [F(5,339) = 29.746; P < .001; adjusted R2 = 0.295]. Lower levels of corneal cylinder are by far the main contributor to astigmatism axis mismatch (ß = -1.164). There was not a single HOA, either corneal or ocular, that significantly loaded into any models. Astigmatism axis mismatch decreases rapidly with increasing levels of corneal astigmatism. Corneal and whole-eye HOAs have no role in astigmatism axis mismatch in healthy candidates for refractive surgery. [J Refract Surg. 2024;40(2):e89-e97.].

Assessment of Keratoconus Risk in Very Asymmetric Ectasia Using Corneal Tomographic and Biomechanical Parameters.

To investigate the relationship between corneal tomographic or biomechanical parameters and risk of keratoconus in very asymmetric ectasia (VAE). This retrospective case-control single-centre study included patients with VAE and normal controls. The VAE group had clinical ectasia in one eye and normal topography (VAE-NT) in the fellow eye; VAE-NT eyes were selected for analysis. The control group was selected from corneal refractive surgery candidates; the right eye was enrolled. Scheimpflug-based corneal tomography (Pentacam) and corneal biomechanical assessment (Corvis ST) were performed. Univariate and multivariable logistic regression were performed using Cox proportional hazards models to evaluate keratoconus-associated risk factors. A two-piecewise linear regression model was applied to examine the threshold effect of selected vital paragmeters on the risk of keratoconus according to a smoothing plot. Threshold effect between tomographic integration and risk of keratoconus was observed. Discrepancy between the central corneal thickness and thinnest corneal thickness (discrepancy CCT vs TCT) greater than 5 μm, discrepancy between the apex corneal thickness and thinnest corneal thickness (discrepancy ACT vs TCT) greater than 3 μm, vector distance between CCT and TCT (distance CCT vs TCT) greater than 0.65 mm indicated a significant increased risk of keratoconus. Risk of keratoconus decreased when distance CCT vs TCT was less than 0.65 mm. Discrepancy CCT vs TCT, discrepancy ACT vs TCT, and distance CCT vs TCT can be used as indicators for risk assessment of early keratoconus.

Assessment of the iridocorneal angle pigmentation and structures after the implantation of EVO+ Visian Implantable Collamer Lens.

To evaluate the effect of EVO+ Visian Implantable Collamer Lens (ICL) implantation on the iridocorneal angle pigmentation and structures. Myopic refractive surgery candidates who underwent posterior chamber EVO+ ICL implantation were evaluated preoperatively and 3 and 6months postoperatively. High-resolution images of the iridocorneal angle (nasal, superior, temporal and inferior quadrants) were acquired during gonioscopy. A masked observer, blinded to study visits and patients' information, evaluated the angle width, apparent iris root insertion, iris configuration, and trabecular meshwork pigmentation according to the Spaeth scale in two different occasions. The intra-rater reliability was estimated using the weighted Gwet's Agreement Coefficient (AC2). Differences between visits were analyzed using the Cochran Q test or the Friedman test. Twenty-one patients (13 females and 8 males) aging 31.3 ± 6.3years old were recruited. The intra-rater reliability of gonioscopy assessment was excellent for every parameter assessed (AC2 ≥ 0.97). No significant differences were found among the study visits for any quadrant in the angle width (p ≥ 0.74), apparent iris root insertion (p ≥ 0.22), iris configuration (p ≥ 0.21) and trabecular meshwork pigmentation (p ≥ 0.24). Mean pigmentation of trabecular meshwork for the four quadrants did not vary either among visits (p = 0.25). The EVO+ ICL implantation after uneventful procedures appears not to clinically affect the iridocorneal angle pigmentation and/or structures during a short-medium follow-up using gonioscopy assessment. The intra-rater reliability of glaucoma specialists assessing iridocorneal angle structures using gonioscopy images is very high.

Keratoconus Detection-based on Dynamic Corneal Deformation Videos Using Deep Learning

To assess the performance of convolutional neural networks (CNNs) for automated detection of keratoconus (KC) in standalone Scheimpflug-based dynamic corneal deformation videos. Retrospective cohort study. We retrospectively analyzed datasets with records of 734 nonconsecutive, refractive surgery candidates, and patients with unilateral or bilateral KC. We first developed a video preprocessing pipeline to translate dynamic corneal deformation videos into 3-dimensional pseudoimage representations and then trained a CNN to directly identify KC from pseudoimages. We calculated the model's KC probability score cut-off and evaluated the performance by subjective and objective accuracy metrics using 2 independent datasets. Area under the receiver operating characteristics curve (AUC), accuracy, specificity, sensitivity, and KC probability score. The model accuracy on the test subset was 0.89 with AUC of 0.94. Based on the external validation dataset, the AUC and accuracy of the CNN model for detecting KC were 0.93 and 0.88, respectively. Our deep learning-based approach was highly sensitive and specific in separating normal from keratoconic eyes using dynamic corneal deformation videos at levels that may prove useful in clinical practice. Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Ocular residual astigmatism (ORA) does not seem to correlate with baseline refractive error among refractive surgery candidates.

Ocular residual astigmatism (ORA) is defined as the difference between refractive astigmatism and anterior corneal astigmatism. A high ORA may be correlated with poorer results in patients undergoing corneal-based laser surgery. Is a high baseline refractive error related to a higher degree of ORA? This was a retrospective analytical study including 181 right eyes of an equal number of refractive surgery candidates. Manifest subjective refraction was measured, along with a Pentacam AXL Wave corneal tomography. Via a vector analysis with this methodology, subjective cylinder was translated into the corneal plane and a vectorial subtraction was performed in order to measure ORA. Spearman's rank order test, one-way ANOVA and Chi-square were used to determine whether different levels of baseline refractive error correlate with different levels of ORA. Mean age was 28.33 ± 4.71years with a female preponderance (65.7%). Mean ORA was 0.74 ± 0.39 D, with 33.1% of eyes having an ORA ≥ 0.90 D. There was not a correlation between ORA and level of myopia (rho = - 0.022; p = 0.764), nor between ORA and spherical equivalent (rho = 0.009; p = 0.903). Refractive astigmatism did not demonstrate to be correlated with ORA level either (rho = 0.078; p = 0.329). One-way ANOVA tests failed to demonstrate an association between different classifications of refractive error and level of ORA. In the studied population, ORA is not correlated with baseline refractive error. Every patient presenting for possible corneal-based laser refractive surgery should be evaluated for a possible high level of ORA, irrespective of their baseline ametropia level.

Comparison between wavefront-derived refraction and auto-refraction

BackgroundThis study aimed to compare objective refractive errors and keratometry measurements obtained using the Nidek OPD-Scan II aberrometer/topographer and Topcon KR 8900 autorefractokeratometer. MethodsThe right eye medical records of 176 patients aged 18-35 years who were admitted to our clinic as refractive surgery candidates were tested for refractive status and keratometry measurements with a Nidek OPD-Scan II aberrometer/topographer and a standard table-top autorefractokeratometer (Topcon KR 8900) before and after the induction of cycloplegia. Patients who had undergone any eye surgery and had hereditary, ectatic, or acquired corneal pathology were excluded. Refractive data were compared as spheres, cylinders, spherical equivalents, and power vectors before and after the induction of cycloplegia. Flat and steep keratometry (K1-K2) readings were recorded in diopters (D) and axis degrees, respectively, for each eye. ResultsThe spherical, cylindrical, spherical equivalence, J0-J45 vector values and K1-K2 readings (D, axis) between the two devices were statistically significant before and after the induction of cycloplegia (p<0.05). Bland-Altman analysis identified mean differences (95%CI of limits of agreement) of 0.77 (-0,57 to 2,11) in sphere, 0.74 (-0,54 to 2,01) in spherical equivalent, -0,07 (-0,41 to 0,26) in J0 vector, 0,06 (-0,31 to 0,43) in J45 vector, -0,16 (-0,66 to 0,33) in K1, -0,23 (-0,79 to 0,33) in K2 values before induction of cycloplegia. ConclusionThe refractive and keratometry results of the Nidek OPD Scan II system and Topcon KR 8900 standard table-top autorefractokeratometer are not interchangeable in healthy adult population before and after induction of cycloplegia.

The distribution pattern of corneal volume in Chinese myopic patients from multiple centers

AbstractCorneal volume (CV) is a useful index for detecting forme fruste keratoconus from normal corneas. It can be used to evaluate the whole cornea, since it can measure corneal areas up to 10 mm in diameter. Thus, CV has become the clinicians' interest as a diagnostic tool of corneal ectatic disease and a measure of corneal integrity to determine suitability for refractive surgery. We conducted a cross‐sectional study including 7893 myopic patients from five ophthalmic centers to investigate the distribution pattern of CV. Our study showed that distribution of CV‐3, CV‐5, and CV‐7 mm were slightly positively skewed and the 2.5th to 97.5th percentiles were 3.6–4.4, 10.4–12.8, 22.5–27.5 mm3, respectively. Central corneal thickness (CCT) was significantly correlated with CV in all measurement regions. The correlation between CV and CCT showed an inconsistent trend with the increase of age. The correlation coefficient between CV and CCT did not change significantly with the increase of myopia degree in low to moderate myopia, but fluctuated significantly in high myopia (less than −6.0 diopters). According to our results, corneal volume follows a slightly positively skewed distribution pattern in myopic Chinese patients. The information is useful for screening refractive surgery candidates and assessing the risk of corneal refractive surgery.

The Role of Anamnesis in Modern Ophthalmic Practice: a Clinical Case

Eyes’ dryness continues to be the most frequent complication after LASIK. As a rule, it disappear in the vast majority of patients during the first year after surgery. The conducted studies indicate that complaints of eyes’ dryness in the preoperative period significantly increase the risk of developing a severe form of dry eye syndrome in the postoperative period, especially in middle-aged women (45–59 years) and older. The clinical case presented in this article indicates the development of a severe form of dry eye syndrome after LASIK in a middle-aged patient receiving hormone replacement therapy for menopause, complaining of a feeling of dryness and discomfort in the eyes in the preoperative period, with intolerance to contact lenses and taking antidepressants in the absence of signs of severe dry eye in the preoperative period. Consequently, surgical intervention led to the development of the mentioned above condition. It is confirmed by the literature data indicating that the above factors can cause the development of a severe form of dry eye syndrome after LASIK. Taking into account the fact that tear production tests do not always correlate with the intensity of complaints of dryness, burning sensation, pain in the eyes, the most important component at the stage of deciding on the feasibility of refractive surgery in a particular patient is the collection of anamnesis and clinical picture. When preparing patients with a history of dry eye syndrome for refractive surgery, it is necessary to inform them that they have a significantly increased risk of developing dry eye syndrome. It may persist in the long-term postoperative period, and possibly go into a chronic form. Candidates for refractive surgery in the presence of complaints characteristic of dry eye syndrome can only be those patients who respond well to therapy aimed at relieving these symptoms in the preoperative period. Taking into account the multicomponent nature of pain in dry eye syndrome — nociceptive pain, which is formed at the stage of receptor irritation in the corneal tissue, and neuropathic pain that occurs with the participation of the peripheral and central nervous system, a comprehensive approach is needed in the treatment of dry eye syndrome.

Corneal and Whole-Eye Higher Order Aberrations Do Not Correlate With Ocular Residual Astigmatism in Prepresbyopic Refractive Surgery Candidates.

* Department of Ophthalmology and Visual Sciences, McGill University, Montreal, QC, Canada; and † LASIK MD, Montreal, QC, Canada. Financial disclosures/conflicts of interest: None reported. A. Wallerstein and M. Gauvin equally contributed to this letter.

Determination of the repeatability of corneal densitometry as measured with a Scheimpflug camera device in refractive surgery candidates.

Corneal densitometry analysis provides an objective measurement of corneal transparency in patients undergoing photorefractive keratectomy (PRK). To date, no study that focuses on determining the reliability and smallest real difference (SRD), specifically in PRK candidates, has been published. The study was conducted at Clínica de Oftalmología Sandiego, Medellín, Colombia. This was a prospective and analytical study. Refractive surgery candidates underwent Pentacam measurements twice with five minutes of difference. Repeatability of measurements was evaluated for every area using Wilcoxon signed-rank test, Spearman's rank-order correlation, and the intraclass correlation coefficient, followed by a within-subjects factor (Sw) finishing with a determination of the SRD for all areas. Bland-Altman plots were created and analyzed. A total of 110 eyes were included. The mean total densitometry was 18.67 ± 1.56 grayscale units (GSU) and 18.65 ± 1.49 GSU for the first and second measurements, respectively. Wilcoxon was non-significant (P > 0.05) while the intraclass correlation coefficient demonstrated an excellent agreement. SRD ranged from 0.21 GSU (posterior × 6-10 mm) to 1.79 GSU (total × 10-12 mm). Anterior depths and 10-12 mm annulus had consistently more variability and greater (worse) SRD. The 10-12 mm annulus showed great dispersion on the Bland-Altman plots. Corneal densitometry has an excellent repeatability in refractive surgery candidates for areas below 10 mm. Any densitometry change ≥1.0 GSU in areas under 10 mm should be considered as well above measurement noise and corresponding to a real change in measured parameters. The 10-12 mm area does not seem to be reliable enough in refractive surgery candidates to warrant surveillance after corneal-based surgery.

Evaluation of Ocular Residual Astigmatism in Eyes with Myopia and Myopic Astigmatism and Its Interaction with Other Forms of Astigmatism.

To evaluate the prevalence, magnitude, and direction of ocular residual astigmatism (ORA) in eyes with myopia and myopic astigmatism, and its interaction with refractive, anterior corneal, posterior corneal, and true net power astigmatism. Refractive surgery candidates with myopia and myopic astigmatism were studied. Refractive astigmatism (RA) was measured using the Nidek® AR-310A autorefractometer. Anterior corneal astigmatism (ACA), posterior corneal astigmatism (PCA), and true net power astigmatism (TNP) were measured using the Wavelight® Oculyzer II. Astigmatism was converted from polar to vector notation. ORA was calculated by vector subtraction of ACA from RA vertexed to corneal plane. Compensation factor (CF) was calculated as the ratio of ORA that compensates ACA for both J0 and J45. 154 eyes of 88 patients (mean age 31.7±7.1 years) were included. With-the-rule (WTR) astigmatism was the most common for both RA (55.6%) and ACA (74%), while against-the-rule (ATR) was the most common for PCA (87.7%) and ORA (74.0%). The axes of RA and ACA were within 10° of each other in 46.8% of the eyes, and within 30° of each other in 76.0%. The mean difference in value between the axis of RA and ACA was 25.6°. 71.4% of eyes in the study had an ORA ≥ 0.5D, 44.1% had ORA ≥ 0.75D and 26% had ORA ≥ 1D. There was a statistically significant difference between ACA and each of RA and TNP. Using TNP to calculate ORA instead of ACA reduced its magnitude. RA is positively correlated to ACA and more strongly to TNP. The most common pattern of compensation between ORA and ACA was under-compensation for J0 (49%) and same-axis-augmentation for J45 (35%). ORA, PCA, and the interaction between ORA and ACA can affect results during refractive planning.

Corneal and Whole-Eye Higher Order Aberrations Do Not Correlate With Ocular Residual Astigmatism in Prepresbyopic Refractive Surgery Candidates.

Ocular residual astigmatism (ORA) is the proportion of manifest astigmatism that is not explained by anterior corneal astigmatism. The role of higher order aberrations (HOAs) in the level of ORA has not been profoundly studied. The purpose of this study was to evaluate the effect of different corneal and whole-eye HOAs on levels of ORA using a multivariate modeling approach. This is a retrospective analytical study including a sample of healthy refractive surgery candidates. One eye of every patient was randomly selected. A total of 294 eyes from an equal number of patients were included. Corneal and whole-eye HOAs were measured with a Pentacam AXL Wave, and subjective manifest refraction was taken. Astigmatism values were converted into power vectors and ORA was calculated. The Spearman rank-order correlation was initially used to explore correlation between HOA and ORA. All variables with a P value under 0.10 were included into a multiple linear regression model to explore this correlation adjusting for confounding variables. The mean age was 28.81 ± 5.40 years. Simple bivariate correlation was significant for root mean square (RMS) total, RMS lower order aberrations, RMS HOAs, defocus, and vertical astigmatism. After being included into a multivariate regression model adjusting for confounding variables, the only variable that was significant was RMS total [F (3, 282) = 78.977; P < 0.001; adjusted R 2 = 0.451]. For every μm increase in corneal RMS total, ORA increases by 0.135 diopters. About 45% of variability in ORA can be explained by corneal RMS total, corneal J 0 , and manifest J 0 . No individual HOA correlated with ORA in the multivariate regression model. ORA increases with general corneal irregularity, especially regular astigmatism. Irregular astigmatism expressed by corneal and whole-eye HOAs does not seem to be a significant contributor of ORA in healthy refractive surgery candidates.

Angle κ and its effect on the corneal elevation maps in refractive surgery candidates.

To assess the associations of angle κ and Pentacam decentration indices with elevation maps in normal refractive surgery candidates. Salouti Cornea Research Center, Salouti Eye Clinic, Shiraz, Iran. Retrospective observational study. In this research, the right eyes of 173 refractive surgery candidates were assessed. Data of front and back corneal elevation maps, keratometric data, decentration indices, and corneal astigmatism obtained by Pentacam HR system and angle κ obtained by Orbscan IIz were extracted. Maximum elevation (or depression) for each of the 4 quadrants was recorded. Correlations of elevation values with angle κ, Pentacam decentration indices, keratometry, and astigmatism were examined by Pearson correlation coefficient. 148 age- and sex-matched cases with keratoconus grade 1 were selected as a positive control group, and Pentacam variables were compared between the groups. Overall, data from 173 eyes of 173 normal refractive surgery candidates and 148 eyes of 148 patients with keratoconus were recorded and analyzed. In normal refractive surgery candidates, the mean of angle κ was 5.32 ± 1.36 (SD) degrees. Angle κ had a positive correlation with front and back temporal elevations based on a best-fit sphere (BFS) ( r = 0.339, P = .001; r = 0.300, P < .001, respectively). Front and back keratometric astigmatisms were positively correlated with front and back nasal and temporal elevations ( r ≥ 0.543, P < .001) and negatively correlated with superior and inferior elevations ( r ≤ -0.547, P < .001). These associations no longer existed when using best-fit toric ellipse (BFTE) for calculating elevation data. The thinnest point to vertex decentration was significantly associated with the back temporal elevation (based on the BFTE) in both normal ( r = 0.311, P < .001) and keratoconus ( r = 0.190, P = .021) eyes. This study showed that elevation maps obtained by the Pentacam system using BFS might be affected by both the large angle κ and corneal astigmatism, confounding the preoperative assessment of refractive surgery candidates. Using BFTE as a reference for calculating elevation data should eliminate most diagnostic issues and thus is encouraged in this context.

The role of corneal biomechanical properties assessment in laser vision correction – the introduction

The role of corneal biomechanical properties in patients referred to laser vision correction (LVC) is currently being raised. Understanding of corneal biomechanics may support the proper selection of refractive surgery candidates, improve the refractive outcomes and safety of refractive procedures. The Ocular Response Analyzer (ORA) and Corvis ST are commonly used devices to assess corneal biomechanical parameters in LVC. The vertical corneal incisions have a greater impact on corneal biomechanics weakening than horizontal incisions. Maintaining the high biomechanical strength of the cornea following LVC can decrease the potential risk of postoperative ectasia.

Artificial Intelligence-Based Diagnostic Model for Detecting Keratoconus Using Videos of Corneal Force Deformation.

PurposeTo develop a novel method based on biomechanical parameters calculated from raw corneal dynamic deformation videos to quickly and accurately diagnose keratoconus using machine learning.MethodsThe keratoconus group was included according to Rabinowitz's criteria, and the normal group included corneal refractive surgery candidates. Independent biomechanical parameters were calculated from dynamic corneal deformation videos. A novel neural network model was trained to diagnose keratoconus. Tenfold cross-validation was performed, and the sample set was divided into a training set for training, a validation set for parameter validation, and a testing set for performance evaluation. External validation was performed to evaluate the model's generalizability.ResultsA novel intelligent diagnostic model for keratoconus based on a five-layer feedforward network was constructed by calculating four biomechanical characteristics, including time of the first applanation, deformation amplitude at the highest concavity, central corneal thickness, and radius at the highest concavity. The model was able to diagnose keratoconus with 99.6% accuracy, 99.3% sensitivity, 100% specificity, and 100% precision in the sample set (n = 276), and it achieved an accuracy of 98.7%, sensitivity of 97.4%, specificity of 100%, and precision of 100% in the external validation set (n = 78).ConclusionsIn the absence of corneal topographic examination, rapid and accurate diagnosis of keratoconus is possible with the aid of machine learning. Our study provides a new potential approach and sheds light on the diagnosis of keratoconus from a purely corneal biomechanical perspective.Translational RelevanceOur findings could help improve the diagnosis of keratoconus based on corneal biomechanical properties.

Use of machine learning to achieve keratoconus detection skills of a corneal expert.

To construct an automatic machine-learning derived algorithm discriminating between normal corneas and suspect irregular or keratoconic corneas. A total of 8526 corneal tomography images of 4904 eyes obtained between November 2010 and July 2017 using a combined Scheimpflug/Placido tomographer were retrospectively evaluated. Each image was evaluated for acquisition quality and was labeled as normal, suspect irregular or keratoconic by a cornea specialist. Two algorithms were built. The first was based on 94 instrument-derived output parameters, and the second integrated keratoconus prediction indices of the device with the 94 instrument-derived output parameters. Both models were compared with the tomographer's keratoconus detection algorithms. Out of the 8526 images evaluated, 7104 images of 3787 eyes had sufficient acquisition quality. Of those, 5904 examinations were randomly chosen for construction of the models using the random forest algorithm. The models were then validated using the remaining 1200 examinations. Both RF algorithms had a larger AUC compared with any of the tomographer's KC detection algorithms (p < 10-9). The first constructed model had 90.2% accuracy, sensitivity of 94.2%, and specificity of 89.6% (Youden 0.838). Calculated AUC was 0.964. The second model had 91.5% accuracy, sensitivity of 94.7%, and specificity of 89.8% (Youden 0.846). Calculated AUC was 0.969. Using the RF machine-learning algorithm, accuracy of discrimination between normal, suspect irregular and keratoconic corneas approximates that of an experienced corneal expert. Applying machine learning to corneal tomography can facilitate keratoconus screening in large populations as well as off-site screening of refractive surgery candidates.

Comorbidities in Corneal Refractive Surgery Candidates: A Cross-Sectional Study

Background: Individuals with refractive errors sometimes have other associated ocular and systemic abnormalities. Objectives: To explore ocular and systemic comorbidities in Corneal Refractive Surgery (CRS) candidates and to examine any contraindications for CRS. Methods: This was a cross-sectional study. Medical records of individuals who underwent a refractive surgery screening at Chiang Mai University LASIK Center, Chiang Mai, Thailand, were reviewed. All clinical data of the initial visit were evaluated. Eyes with a history of CRS and phototherapeutic keratectomy were excluded. Results: A total of 1,167 cases (2,334 eyes) were recruited, out of which 643 cases were females (55.09%). Myopia was the most common type of refractive error (2,120 eyes, 90.83%), and 868 (40.94%) eyes had high myopia. Approximately 45% (n=526) of the cases had worn contact lenses. The five most common ocular comorbidities (eyes, proportion) included keratoconus suspect (297, 12.72%), cataract (246, 10.53%), dry eye/probable dry eye (208, 8.91%), glaucoma suspect (184, 7.88%), and amblyopia (149, 6.38%). The most common systemic conditions that affected choices or outcomes of CRS were hypertrophic scar/keloid (169 cases, 14.48%) followed by allergy/asthma (127 cases, 10.88%). A total of 1,028 eyes (44.04%) were excluded from CRS, mainly due to abnormal corneal topography (470/1,028 eyes, 45.72%). Conclusion: Most refractive candidates were myopic and contact lens wearers. Serious ocular sequelae from myopia and contact lens use were not uncommon. Additionally, systemic diseases regularly affected the CRS options. The most common contraindication for CRS was abnormal corneal topography. Our study, thus, emphasizes the need for thorough ocular and systemic screenings of refractive candidates.

The Distribution Pattern of Ocular Residual Astigmatism in Chinese Myopic Patients.

PurposeWe aimed to investigate the distribution of ocular residual astigmatism (ORA) and its associations with age, gender, manifest refraction, and other ocular indicators in Chinese patients with myopia.DesignThis is a multi-center retrospective cross-sectional study.MethodThe study included 7,893 patients with myopia (7,893 eyes) aged 18–40 years from five ophthalmic centers. Anterior segment biometrics of the eyes were collected from the Pentacam. ORA and its summated vector mean were calculated using Alpins vector analysis. Compensation factor (CF) was used to evaluate the relation between ORA and corneal astigmatism. ORA in different age, gender, and refraction groups was compared. The Spearman correlation was adopted to reveal multiple ocular indicators associated with ORA, which were integrated into a multiple linear regression model to predict ORA.ResultsDistribution of ORA was slightly positively skewed (Skewness= 2.111, Kurtosis = 19.660, KS P < 0.0001). Mean ORA was 0.74 ± 0.39 D (95% normal range: 0.14–1.54 D). Among all the subjects, 22.4% of the eyes had an ORA magnitude of 1.0 D or more. Undercompensation or full-compensation of anterior corneal astigmatism (ACA) by ORA prevailed in both J0 (76.99%) and J45 (58.48%). Women had higher ORA power than men (0.77 ± 0.36 D vs. 0.73 ± 0.41 D, P < 0.0001). Participants with less negative spherical equivalent (SE) or higher manifest astigmatism (MA) also had higher ORA (all P < 0.0001). ORA was significantly correlated with ACA (r = 0.405) and posterior corneal astigmatism (PCA, r = 0.356). The multivariate logistic regression analysis showed strong predictability of ORA magnitude >1.0 D using anterior segment parameters (area under the receiver operating characteristic curve: 0.739).ConclusionORA is present in Chinese adults with myopia and is affected by multiple ocular factors. Our findings may provide valuable information about ORA distribution in candidates for refractive surgery, helping optimize the outcome of astigmatism correction.