Corneal cross-linking versus conventional management for keratoconus: a lifetime economic model

Trends and Sociodemographic Patterns in Keratoconus Management 2015-2020: An American Academy of Ophthalmology IRIS® Registry Analysis

PURPOSE

Investigate trends in keratoconus (KCN) treatment patterns and diagnosis age from 2015 to 2020 and evaluate sociodemographic associations with the treatment approach.

DESIGN

Retrospective cohort study.

PARTICIPANTS

Patients with a new KCN diagnosis from 2015 to 2020 were identified in the Academy IRIS® Registry (Intelligent Research in Sight).

METHODS

Associations between sociodemographic factors and treatment were evaluated using multivariable logistic regression.

MAIN OUTCOME MEASURES

Outcomes included percentages and rates of each treatment (collagen crosslinking [CXL], keratoplasty, or no procedure) from 2015 to 2020, age at diagnosis during this period, and sociodemographic factors associated with treatment type.

RESULTS

A total of 66 199 patients with a new diagnosis of KCN were identified. The percentage of patients undergoing CXL increased from 0.05% in 2015 to 29.5% in 2020 (P = 0.008). The average age (standard deviation) of KCN patients decreased from 44.1 (±16.9) years in 2015 to 39.2 (±16.9) years in 2020 (P < 0.001). In multivariable analyses comparing CXL versus no procedure and keratoplasty versus no procedure, patients undergoing CXL tended to be younger with the odds of having CXL decreasing with increasing age, for example, comparing CXL and no procedure patients, using ages 0-20 years as reference, the odds ratio (OR) (95% confidence interval [CI]) decreased from 0.62 (0.57-0.67; P < 0.0001) for patients aged 21-40 years to 0.03 (0.02-0.04; P < 0.0001) for patients aged > 60 years. Men were more likely than women to have CXL (OR, 1.31; 95% CI, 1.23-1.40; P < 0.0001) and keratoplasty (OR, 1.30; 95% CI, 1.19-1.42; P < 0.0001). Black patients were less likely than White patients to have CXL (OR, 0.70; 95% CI, 0.63-0.77; P < 0.0001) and more likely to have keratoplasty (OR, 2.24; 95% CI, 2.01-2.50; P < 0.0001). Likewise, Hispanic patients had higher odds of CXL (OR, 1.12; 95% CI, 1.00-1.24; P < 0.05) and keratoplasty (OR, 1.29; 95% CI, 1.12-1.50; P < 0.001) compared with non-Hispanic patients. Collagen crosslinking and keratoplasty also varied by region and insurance status.

CONCLUSIONS

A significant increase in use of CXL was noted from 2015 to 2020. Sociodemographic differences in treatment among KCN patients may reflect differences in access, use, or care patterns, and future studies should aim to identify strategies to improve access for all patients.

FINANCIAL DISCLOSURE(S)

The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Assessment of the Predictive Ability of Theranostics for Corneal Cross-linking in Treating Keratoconus: A Randomized Clinical Trial

PURPOSE

To validate the ability of theranostic imaging biomarkers in assessing corneal cross-linking (CXL) efficacy in flattening the maximum keratometry (Kmax) index.

DESIGN

Prospective, randomized, multicenter, masked clinical trial (ClinicalTrails.gov identifier, NCT05457647).

PARTICIPANTS

Fifty patients with progressive keratoconus.

INTERVENTION

Participants were stratified to undergo epithelium-off (25 eyes) and epithelium-on (25 eyes) CXL protocols using an ultraviolet A (UV-A) medical device with theranostic software. The device controlled UV-A light both for performing CXL and assessing the corneal riboflavin concentration (riboflavin score) and treatment effect (theranostic score). A 0.22% riboflavin formulation was applied onto the cornea for 15 minutes and 20 minutes in epithelium-off and epithelium-on protocols, respectively. All eyes underwent 9 minutes of UV-A irradiance at 10 mW/cm2.

MAIN OUTCOME MEASURES

The primary outcome measure was validation of the combined use of theranostic imaging biomarkers through measurement of their accuracy (proportion of correctly classified eyes) and precision (positive predictive value) to classify eyes correctly and predict a Kmax flattening at 1 year after CXL. Other outcome measures included change in Kmax, endothelial cell density, uncorrected and corrected distance visual acuity, manifest spherical equivalent refraction and central corneal thickness 1 year after CXL.

RESULTS

Accuracy and precision of the theranostic imaging biomarkers in predicting eyes that had >0.1 diopter (D) of Kmax flattening at 1 year were 91% and 95%, respectively. The Kmax value significantly flattened by a median of -1.3 D (IQR, -2.11 to -0.49 D; P < 0.001); both the uncorrected and corrected distance visual acuity improved by a median of -0.1 logarithm of the minimum angle of resolution (logMAR; IQR, -0.3 to 0.0 logMAR [P < 0.001] and -0.2 to 0.0 logMAR [P < 0.001], respectively). No significant changes in endothelial cell density (P = 0.33) or central corneal thickness (P = 0.07) were noted 1 year after surgery.

CONCLUSIONS

The study demonstrated the efficacy of integrating theranostics in a UV-A medical device for the precise and predictive treatment of keratoconus with epithelium-off and epithelium-on CXL protocols. Concentration of riboflavin and its UV-A light mediated photoactivation in the cornea are the primary factors determining CXL efficacy.

FINANCIAL DISCLOSURE(S)

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Crosslinking with UV-A and riboflavin in progressive keratoconus: From laboratory to clinical practice - Developments over 25 years

Changes in the biomechanical and biochemical properties of the human cornea play an important role in the pathogenesis of ectatic diseases. A number of conditions in primarily acquired (keratoconus or pellucid marginal degeneration) or secondarily induced (iatrogenic keratectasia after refractive laser surgeries) ectatic disorders lead to decreased biomechanical stability. Corneal collagen cross-linking (CXL) represents a technique to slow or even halt the progression of ectatic pathologies. In this procedure, riboflavin is applied in combination with ultraviolet A radiation. This interaction induces the production of reactive oxygen species, which leads to the formation of additional covalent bonds between collagen molecules and subsequent biomechanical corneal strengthening. This procedure is so far the only method that partially interferes etiopathogenetically in the treatment of ectatic diseases that slows or stops the process of corneal destabilization, otherwise leading to the need for corneal transplantation. Besides, CXL process increases markedly resistance of collagenous matrix against digesting enzymes supporting its use in the treatment of corneal ulcers. Since the discovery of this therapeutic procedure and the first laboratory experiments, which confirmed the validity of this method, and the first clinical studies that proved the effectiveness and safety of the technique, it has been spread and adopted worldwide, even with further modifications. Making use of the Bunsen-Roscoe photochemical law it was possible to shorten the duration of this procedure in accelerated CXL and thus improve the clinical workflow and patient compliance while maintaining the efficacy and safety of the procedure. The indication spectrum of CXL can be further expanded by combining it with other vision-enhancing procedures such as individualized topographically-guided excimer ablation. Complementing both techniques will allow a patient with a biomechanically stable cornea to regularize it and improve visual acuity without the need for tissue transplantation, leading to a long-term improvement in quality of life.

The Pathophysiology of Keratoconus

PURPOSE

Keratoconus is a progressive disease characterized by changes in corneal shape, resulting in loss of visual function. There remains a lack of comprehensive understanding regarding its underlying pathophysiology. This review aims to bridge this gap by exploring structural failures and inflammatory processes involved in the etiology and progression of keratoconus.

METHODS

A literature review was conducted using PubMed and Google Scholar databases, screening for articles published in English using the keyword combinations of "keratoconus" with "pathophysiology," "pathology," "metabolism," "inflammatory," "oxidative stress," "cytokines," "enzymes," "collagen," and "cornea." Articles published between January 1, 1970, and June 1, 2023, were queried and reviewed, with greater emphasis placed on more recent data. Fifty-six relevant studies were examined to develop a thorough review of the pathophysiological mechanisms at play in keratoconus.

RESULTS

Biomechanical structural failures in the cornea seem to be the primary militating factors in keratoconus etiology and progression. These include disruptions in the arrangement in the collagen lamellae, a decrease in collagen levels, a decrease in natural collagen crosslinking, and changes in lysosomal enzyme activity. Immunologic changes have also been identified in keratoconus, challenging the traditional view of the condition as noninflammatory. Elevated levels of proinflammatory cytokines like IL-1b, IL-6, IL-17, and TNF-α have been observed, along with increased apoptosis of keratocytes. Increased oxidative stress leads to the activation of collagenase and gelatinase enzymes.

CONCLUSIONS

Keratoconus is a complex condition influenced by both structural defects and inflammatory processes. Understanding these mechanisms can inform clinical management and potentially lead to more effective treatments.

Prevalence and Economic Burden of Keratoconus in the United States

PURPOSE

To assess the prevalence and economic burden of keratoconus in the United States.

DESIGN

Retrospective cohort study.

METHODS

Patients enrolled in Medicaid and Children's Health Insurance Program (CHIP) who were diagnosed with keratoconus between 2016 and 2019 were included. The data reported to the Centers for Disease Control and Prevention (CDC) Vision and Eye Health Surveillance System (VEHSS) were analyzed. The crude prevalence rates (national and statewise) were obtained from the database and extrapolated to estimate the keratoconus case count in the United States. The keratoconus prevalence was compared between male and female individuals using the Mann-Whitney test, whereas Brown-Forsythe 1-way analysis of variance was used to compare prevalence between age and racial groups. The Dunnett T3 multiple comparison test was used for intergroup comparison. Finally, the economic burden of keratoconus was assessed by inflation-adjusted direct costs to patients and total cases in the country.

RESULTS

In the cohort of 69,502,000 patients enrolled for Medicaid and CHIP, the national prevalence of keratoconus was computed to be 0.04% in 2019 and had increased from 0.03% in 2016. The highest prevalence of keratoconus was observed in patients 18 to 39 years of age, followed by patients 40 to 64 years of age; comparable prevalence rates were observed in these age groups in the Black population. The prevalence was moderately higher in female compared to male individuals; however, significantly higher keratoconus prevalence was observed in Black female individuals compared to male individuals. A significantly high prevalence of keratoconus was observed in the Black population, followed by Hispanic population. In 2019, the average inflation-adjusted lifetime cost of keratoconus treatment was USD 28,766.69, with a cumulative economic burden of USD 3.8 billion.

CONCLUSIONS

In the United States, keratoconus is most prevalent in individuals 18 to 39 years of age. The keratoconus prevalence is higher in the Black population, specifically female individuals, and the diagnosis is often delayed in these patients.

Real-world socioeconomic determinants of corneal crosslinking in a national cohort

PURPOSE

To characterize recent socioeconomic trends in patients with keratoconus/corneal ectasias undergoing corneal crosslinking (CXL).

SETTING

A deidentified administrative medical claims database comprised commercial and Medicare Advantage health claims from across the United States.

DESIGN

Population-based retrospective cohort study.

METHODS

This study identified 552 patients with keratoconus/corneal ectasia who underwent CXL and 2723 matched controls who did not undergo CXL based on Current Procedural Terminology coding from a U.S. national insurance claims database from 2016 to 2020. For each patient, characteristics, including sex, race, age, household net worth, education level, insurance plan type, and geographic region, were extracted. Multivariate logistic regression was conducted to determine the odds of undergoing crosslinking.

RESULTS

Age 30 years or older (odds ratio [OR], 0.34, P < .001) was associated with decreased likelihood of undergoing CXL. Sex, race, education, and patient income were not associated with odds of undergoing CXL. Patients with health maintenance organization insurance had lower odds of undergoing CXL (OR, 0.64, P = .047). Geographically, patients on the east coast (OR, 0.37, P < .001) and Lower Midwest (OR, 0.31, P < .001) had statistically lower odds of undergoing crosslinking.

CONCLUSIONS

This is the first study to identify socioeconomic determinants of CXL, and it highlights that geographic location and insurance type may limit accessibility to patients.

Factors Associated With Receipt of Crosslinking for Keratoconus

PURPOSE

The aim of this study was to identify factors associated with receipt of standard fluence epithelium-off crosslinking (CXL) for keratoconus (KCN).

METHODS

This retrospective, cross-sectional study reviewed electronic health records of treatment-naive patients with KCN seen at the Wilmer Eye Institute between January 2017 and September 2020. Tomographic data were derived from Pentacam (Oculus, Wetzlar, Germany) devices. Multivariable population-average model using generalized estimating equations adjusting for age, sex, race, national area deprivation index, vision correction method, and disease severity was used to identify factors associated with receipt of CXL.

RESULTS

From 583 patients with KCN, 97 (16.6%) underwent CXL for KCN. Patients who received CXL in at least 1 eye were significantly younger (mean 24.0 ± 7.8 years) than patients who had never undergone CXL (33.4 ± 9.3 years) ( P < 0.001). In multivariable analysis, Black patients had 63% lower odds of receiving CXL for KCN (OR: 0.37, 95% CI, 0.18-0.79) versus White patients, and older age was protective against receipt of CXL (OR: 0.89 per 1-year increase, 95% CI, 0.86-0.93). Comparison of characteristics by race demonstrated that Black patients presented with significantly worse vision, higher keratometric indices (K1, K2, and Kmax), and thinner corneal pachymetry at baseline versus White or Asian patients.

CONCLUSIONS

In this clinical cohort of patients with KCN from a tertiary referral center, Black patients were less likely to receive CXL presumably because of more advanced disease at presentation. Earlier active population screening may be indicated to identify and treat these patients before they become ineligible for treatment and develop irreversible vision loss. Such strategies may improve health equity in KCN management.

Definitions for Keratoconus Progression and Their Impact on Clinical Practice

PURPOSE

There is currently no consensus on which keratoconus need cross-linking nor on how to establish progression. This study assessed the performance of diverse progression criteria and compared them with our clinical knowledge of keratoconus evolution.

METHODS

This was a retrospective, longitudinal, observational study. Habitual progression criteria, based on (combinations of) keratometry (K MAX ), front astigmatism (A F ), pachymetry (P MIN ), or ABCD progression display, from 906 keratoconus patients were analyzed. For each criterion and cutoff, we calculated %eyes flagged progressive at some point (R PROG ), individual consistency C IND (%examinations after progression detection still considered progressive), and population consistency C POP (% eyes with CIND >66%). Finally, other monotonic and consistent variables, such as front steep keratometry (K 2F ), mean radius of the back surface (R mB ), and the like, were evaluated for the overall sample and subgroups.

RESULTS

Using a single criterion (e.g., ∆K MAX >1D) led to high values of R PROG . When combining two, (K MAX and A F ) led to worse C POP and higher variability than (K MAX and P MIN ); alternative criteria such as (K 2F and R mB ) obtained the best C POP and the lowest variability ( P <0.0001). ABC, as defined by its authors, obtained R PROG of 74.2%. Using wider 95% confidence intervals (95% CIs) and requiring two parameters over 95CI reduced R PROG to 27.9%.

CONCLUSION

Previous clinical studies suggest that 20% to 30% of keratoconus cases are progressive. This clinical R PROG value should be considered when defining KC progression to avoid overtreatment. Using combinations of variables or wider margins for ABC brings R PROG closer to these clinical observations while obtaining better population consistency than current definitions.

Multimodal diagnostics for keratoconus and ectatic corneal diseases: a paradigm shift

Different diagnostic approaches for ectatic corneal diseases (ECD) include screening, diagnosis confirmation, classification of the ECD type, severity staging, prognostic evaluation, and clinical follow-up. The comprehensive assessment must start with a directed clinical history. However, multimodal imaging tools, including Placido-disk topography, Scheimpflug three-dimensional (3D) tomography, corneal biomechanical evaluations, and layered (or segmental) tomography with epithelial thickness by optical coherence tomography (OCT), or digital very high-frequency ultrasound (dVHF-US) serve as fundamental complementary exams for measuring different characteristics of the cornea. Also, ocular wavefront analysis, axial length measurements, corneal specular or confocal microscopy, and genetic or molecular biology tests are relevant for clinical decisions. Artificial intelligence enhances interpretation and enables combining such a plethora of data, boosting accuracy and facilitating clinical decisions. The applications of diagnostic information for individualized treatments became relevant concerning the therapeutic refractive procedures that emerged as alternatives to keratoplasty. The first paradigm shift concerns the surgical management of patients with ECD with different techniques, such as crosslinking and intrastromal corneal ring segments. A second paradigm shift involved the quest for identifying patients at higher risk of progressive iatrogenic ectasia after elective refractive corrections on the cornea. Beyond augmenting the sensitivity to detect very mild (subclinical or fruste) forms of ECD, ectasia risk assessment evolved to characterize the inherent susceptibility for ectasia development and progression. Furthermore, ectasia risk is also related to environmental factors, including eye rubbing and the relational impact of the surgical procedure on the cornea.

The Lifetime Expenditure in People with Keratoconus in Saudi Arabia

AIM

This study measured and evaluated the socioeconomic burden of people living with keratoconus in Saudi Arabia.

METHODS

This study employed a cross-sectional design, a Keratoconus Economic Burden Questionnaire, and a convenient sample of 89 keratoconus patients (58.4% male) drawn from multiple regions in Saudi Arabia. It was conducted using online surveys, and the data were analyzed using appropriate quantitative techniques.

RESULTS

The mean age and annual income of the participants were 33.24 (SD = 6.82) years and USD 9046.52 (SD = 16,866.48), respectively, with only 37% being employed for wages. Up to 94.4% needed glasses or contact lenses at least once a week, and 73.0% received care from optometrists. The condition forced 45.9% of the respondents to change careers or leisure activities, with a further 51.3% having to take time off work. The mean annual out-of-pocket expenses for buying and maintaining glasses or contact lenses, as well as traveling and accommodation for keratoconus-related treatment were USD 2341.76 (SD = 3053.09), with 48.32 incurring upwards of USD 3240 over the period. The treatment costs increased with disease duration, r(89) = 0.216, p < 0.05. Regression results showed that the existence of comorbid eye disease, changing glasses at least once a year, and wearing either glasses or contact lenses at least once a week individually had statistically significant negative effects on the total annual keratoconus treatment costs, while disease duration, utilization of optometrists, and taking time off had a statistically significant increase on the total cost (p < 0.05).

CONCLUSION

With a prevalence rate of 1 in 375, progressive debilitation, and the lifetime nature of the disease, keratoconus is a critical public health concern in Saudi Arabia. The resulting visual impairment and discomfort, as well as both direct and indirect economic burdens, have considerable impacts on the patient's quality of life.

Corneal Crosslinking With Riboflavin and UVA Light in Progressive Keratoconus: Fifteen-Year Results

PURPOSE

To analyze the 15-year results of corneal crosslinking (CXL) in progressive keratoconus.

DESIGN

Retrospective follow-up analysis of interventional study patients.

METHODS

This study included keratoconic eyes with progressive disease treated from 2001 to 2006 at the Department of Ophthalmology, Carl Gustav Carus University Hospital, TU Dresden, Germany. CXL was performed by applying riboflavin and ultraviolet A (UVA) light according to a standard protocol. The best-corrected distance visual acuity (BCVA), slitlamp examination, and corneal topography as well as corneal thickness values were recorded preoperatively and 15 years after the treatment.

RESULTS

A total of 42 eyes received a complete follow-up of 15 years. The mean age of the patients at baseline was 26.9 (95% CI: 25.0-28.8) years. The maximum keratometry was 61.6 (95% CI: 58.2 - 64.9) diopters (D) preoperatively and 55.1 (95% CI: 51.6-58.4) D postoperatively; the decrease was statistically significant (P < .001). The mean keratometry value changed from 50.3 (95% CI: 48.3-52.4) D to 47.5 (95% CI: 45.3-49.4) D (P < 0.001). Furthermore, the thinnest corneal thickness decreased statistically significantly by 40 (95% CI: 24-56) µm (P < .001). The BCVA improved statistically significantly from 0.4 to 0.2 logMAR after the treatment. Retreatment was needed in 14% of cases. Mild scarring of the superficial stromal corneal layers was observed in 36% of the eyes, and in 67% of them visual acuity was stable or even improved.

CONCLUSIONS

The CXL procedure proved to be an effective method in the treatment of keratoconic eyes in the progressive stage of the disease, and achieved long-term stabilization without the occurrence of serious complications or side effects.

Benefits of Scleral Lens in the Management of Irregular Corneas and Dry Eye Syndrome After Refractive Surgery

OBJECTIVES

To evaluate the utility of the SPOT scleral lens (Oxygen Permeable Scleral Lens of Thonon; LAO, Thonon-les-Bains, France) in the management of the irregular cornea after refractive surgery.

METHOD

We included 19 patients (35 eyes) with irregular corneas after refractive surgery. Patients were fitted with scleral lenses after the failure of conventional contact lenses adaptation. The ophthalmologic examination included measurement of best-corrected visual acuity (BCVA), slitlamp examination, and evaluation of ocular aberrations (Objective Scattering Index [OSI] and higher-order aberration [HOA]).

RESULT

Scleral lens fitting increases significantly the BCVA from 0.33 (±0.25) to 0.08 (±0.13) LogMAR (P<0.001). There was also a significant decrease in Ocular Surface Disease Index from 66.2±22.8 to 42.4±18.9 (P<0.001). Ocular aberrations (OAs) are also significantly reduced by the scleral lenses, the mean OSI goes from 7.2 (±4.2) to 3.0 (±1.8) (P<0.001), OA from 2.58 (±1.34) to 1.98 μm (±2.31) (P=0.035), and HOA from 0.94 (±0.51) to 0.48 (±0.23) (P=0.0018).

CONCLUSION

Fitting with scleral lenses improves patients' optical and ocular surface problems. Scleral lens restores BCVA and the quality of life. Fitting with scleral lenses is an alternative to further surgery on these fragile eyes and is sometimes the only viable treatment option for the patient.

Personalized Model to Predict Keratoconus Progression From Demographic, Topographic, and Genetic Data

PURPOSE

To generate a prognostic model to predict keratoconus progression to corneal crosslinking (CXL).

DESIGN

Retrospective cohort study.

METHODS

We recruited 5025 patients (9341 eyes) with early keratoconus between January 2011 and November 2020. Genetic data from 926 patients were available. We investigated both keratometry or CXL as end points for progression and used the Royston-Parmar method on the proportional hazards scale to generate a prognostic model. We calculated hazard ratios (HRs) for each significant covariate, with explained variation and discrimination, and performed internal-external cross validation by geographic regions.

RESULTS

After exclusions, model fitting comprised 8701 eyes, of which 3232 underwent CXL. For early keratoconus, CXL provided a more robust prognostic model than keratometric progression. The final model explained 33% of the variation in time to event: age HR (95% CI) 0.9 (0.90-0.91), maximum anterior keratometry 1.08 (1.07-1.09), and minimum corneal thickness 0.95 (0.93-0.96) as significant covariates. Single-nucleotide polymorphisms (SNPs) associated with keratoconus (n=28) did not significantly contribute to the model. The predicted time-to-event curves closely followed the observed curves during internal-external validation. Differences in discrimination between geographic regions was low, suggesting the model maintained its predictive ability.

CONCLUSIONS

A prognostic model to predict keratoconus progression could aid patient empowerment, triage, and service provision. Age at presentation is the most significant predictor of progression risk. Candidate SNPs associated with keratoconus do not contribute to progression risk.

The Safety Profile of FDA-Approved Epithelium-Off Corneal Cross-Linking in a US Community-Based Healthcare System

Purpose

To determine the occurrence of early post-operative complications following FDA-approved epithelium-off corneal cross-linking in the United States.

Materials and Methods

This multicenter, retrospective cohort study identified patients who underwent epithelium-off (epi-off) corneal cross-linking (CXL) for keratoconus and post-refractive keratectasia within the Kaiser Permanente Northern California healthcare system between 2016 and 2018. Post-operative complications including delayed epithelial healing, infection, and loss of visual acuity were recorded.

Results

The study included 878 eyes of 654 patients. The mean age was 27±9.4 years (range 7–71). Five hundred ninety-nine patients (91.6%) had keratoconus while 55 had post-refractive corneal ectasia (8.4%). Forty-seven eyes had prolonged follow-up because of the occurrence of complications in the early post-operative period. The respective rates of delayed epithelial healing, and keratitis were 3.9% (95% CI 2.7–5.3%), and 1.5% (95% CI 0.8–2.5%). Four approaches for management of delayed epithelial healing were compared; epithelium healing duration was the longest in the repeat bandage contact lens (BCL) group (23.8 days) and the shortest in the antibiotic ointment group (14.3 days), with statistically significant differences (p < 0.05) in the healing time between these 2 groups.

Conclusion

The concern for early clinical complications after epi-off CXL often leads to delayed CXL intervention and further keratoconus progression with increased economic burdens. A large retrospective review of 878 eyes found that FDA-approved epi-off CXL protocol appears to be safe with low occurrence rates of early post-operative complications. The recommended management for delayed epithelial healing is using antibiotic ophthalmic ointment.

Machine Learning Algorithms to Detect Subclinical Keratoconus: Systematic Review

BACKGROUND

Keratoconus is a disorder characterized by progressive thinning and distortion of the cornea. If detected at an early stage, corneal collagen cross-linking can prevent disease progression and further visual loss. Although advanced forms are easily detected, reliable identification of subclinical disease can be problematic. Several different machine learning algorithms have been used to improve the detection of subclinical keratoconus based on the analysis of multiple types of clinical measures, such as corneal imaging, aberrometry, or biomechanical measurements.

OBJECTIVE

The aim of this study is to survey and critically evaluate the literature on the algorithmic detection of subclinical keratoconus and equivalent definitions.

METHODS

For this systematic review, we performed a structured search of the following databases: MEDLINE, Embase, and Web of Science and Cochrane Library from January 1, 2010, to October 31, 2020. We included all full-text studies that have used algorithms for the detection of subclinical keratoconus and excluded studies that did not perform validation. This systematic review followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) recommendations.

RESULTS

We compared the measured parameters and the design of the machine learning algorithms reported in 26 papers that met the inclusion criteria. All salient information required for detailed comparison, including diagnostic criteria, demographic data, sample size, acquisition system, validation details, parameter inputs, machine learning algorithm, and key results are reported in this study.

CONCLUSIONS

Machine learning has the potential to improve the detection of subclinical keratoconus or early keratoconus in routine ophthalmic practice. Currently, there is no consensus regarding the corneal parameters that should be included for assessment and the optimal design for the machine learning algorithm. We have identified avenues for further research to improve early detection and stratification of patients for early treatment to prevent disease progression.

Forecasting Progressive Trends in Keratoconus by Means of a Time Delay Neural Network

Early and accurate detection of keratoconus progression is particularly important for the prudent, cost-effective use of corneal cross-linking and judicious timing of clinical follow-up visits. The aim of this study was to verify whether a progression could be predicted based on two prior tomography measurements and to verify the accuracy of the system when labelling the eye as stable or suspect progressive. Data from 743 patients measured by Pentacam (Oculus, Wetzlar, Germany) were available, and they were filtered and preprocessed to data quality needs. The time delay neural network received six features as input, measured in two consecutive examinations, predicted the future values, and determined the classification (stable or suspect progressive) based on the significance of the change from the baseline. The system showed a sensitivity of 70.8% and a specificity of 80.6%. On average, the positive and negative predictive values were 71.4% and 80.2%. Including data of less quality (as defined by the software) did not significantly worsen the results. This predictive system constitutes another step towards a personalized management of keratoconus. While the results obtained were modest and perhaps insufficient to decide on a surgical procedure, such as cross-linking, they may be useful to customize the timing for the patient’s next follow-up.