A comparison of keratoconus progression following collagen cross-linkage using standard or personalised keratometry thresholds.

OBJECTIVE: To define how estimates of keratoconus progression following collagen cross-linking (CXL) vary according to the parameter selected to measure corneal shape. MATERIALS AND METHODS: We estimated progression following CXL in 1677 eyes. We compared standard definitions of keratoconus progression based on published thresholds for Kmax, front K2, or back K2, or progression of any two of these three parameters, with the option of an increased threshold for Kmax values ≥ 55D. As corneal thickness reduces unpredictably after CXL, it was excluded from the principal analysis. We then repeated the analysis using novel adaptive estimates of progression for Kmax, front K2, or back K2, developed separately using 6463 paired readings from keratoconus eyes, with a variation of the Bland-Altman method to determine the 95% regression-based limits of agreement (LoA). We created Kaplan-Meier survival plots for both standard and adaptive thresholds. The primary outcome was progression five years after a baseline visit 9-15 months following CXL. RESULTS: Progression rates were 8% with a standard (≥ 1.5D) threshold for K2 or 6% with the static multi-parameter definition. With a ≥ 1D threshold for Kmax, the progression was significantly higher at 29%. With adaptive Kmax or K2, the progression rates were similar (20%) but less than with the adaptive multi-parameter method (22%). CONCLUSIONS: Estimates of keratoconus progression following CXL vary widely according to the reference criteria. Using adaptive thresholds (LoA) to define the repeatability of keratometry gives estimates for progression that are markedly higher than with the standard multi-parameter method.

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.

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.

Accelerated Pulsed High-Fluence Corneal Cross-Linking for Progressive Keratoconus.

PURPOSE: To report on 2-year results of accelerated corneal collagen cross-linking (CXL) in progressive ectasia using the Avedro KXL system. DESIGN: Prospective interventional case series. METHODS: A total of 870 patients (1,192 eyes) attending Moorfields Eye Hospital after CXL were included. All patients undergoing CXL had progressive keratoconus. Corneas with a minimum stromal thickness <375 µm were excluded. Riboflavin 0.1% soak duration was 10 minutes. High-fluence pulsed UVA was delivered at 30 mW/cm2 for 4 minutes, with a 1.5-second on/off cycle (total energy 7.2 J/cm2). Subjective refractive, corneal tomography, and specular microscopy were performed at baseline, 6, 12, and 24 months postoperatively. The primary outcome measure was a change in maximum keratometry (Kmax) at 24 months. RESULTS: Twelve- and 24-month follow-up data were available on 543 and 213 patients, respectively (mean age 25.4 ± 6.6 years). In mild cones (Kmax < 55 diopter [D]), mean keratometry remained unchanged at 24 months. In more advanced disease, we observed modest corneal flattening compared to baseline (Kmax 63.2 ± 6.5 D vs 61.9 ± 8.1 D, P = .02), but no significant changes in central keratometry (K1 or K2). Keratometric stabilization was confirmed in 98.3% of eyes. Mean CDVA, manifest refraction and endothelial cell density did not change. Overall, 2.7% of eyes lost more than 2 lines of CDVA. CONCLUSION: Accelerated pulsed CXL is a safe, effective, and refractively neutral intervention (at 2 years) to halt disease progression in keratoconus.

Combined wavefront-guided transepithelial photorefractive keratectomy and corneal crosslinking for visual rehabilitation in moderate keratoconus.

PURPOSE: To present 24-month results from the transepithelial photorefractive keratectomy (PRK)-corneal crosslinking (CXL) trial using simultaneous accelerated CXL and a new tissue-saving ocular wavefront-guided transepithelial PRK algorithm aiming to reverse visual loss in early-stage keratoconus without compromise to stabilization of disease progression. SETTING: Moorfields Eye Hospital, London, United Kingdom. DESIGN: Prospective case series. METHOD: Patients with progressive grades I to III keratoconus and logarithm of the minimum angle of resolution (logMAR) corrected distance vision acuity (CDVA) worse than 0.00 (20/20) were included. Consecutive matched historical controls treated only with accelerated CXL were the control group. The main outcome measure was change in logMAR CDVA. RESULTS: The study group comprised 47 eyes of 47 patients (mean age 24.6 years ± 3.8 [SD]). The CDVA improved from 0.28 ± 0.21 logMAR (20/60) preoperatively to 0.15 ± 0.14 logMAR (20/30) 24 months after transepithelial PRK-CXL (P < .001). Twelve eyes gained and 1 eye lost 2 lines or more of CDVA. The mean stromal ablation depth at the cone apex was 35 ± 15 µm. Significant reductions in the maximum keratometry (K) reading and coma were evident in topographic comparison maps. The controls (n = 47) had no significant changes in CDVA, higher-order aberrations, or K values. The mean K values in both groups were stable from 6 months after treatment. CONCLUSION: Ocular wavefront-guided transepithelial PRK-CXL resulted in significant gains in CDVA without compromising CXL efficacy over a 24-month follow-up.