This article is adapted from a presentation originally delivered at the 2024 AECOS Winter Symposium in Aspen, Colorado.
Corneal ectasia is a rare but serious complication of refractive surgery, first identified in 1998.1 Recent advances in multimodal imaging, powered by AI, are transforming the assessment of ectasia susceptibility. These new methods enable more than just the early detection of mild keratoconus cases; they also hold the potential to prevent progressive ectasia.
For more than 12 years, the Brazilian Artificial Intelligence Networking in Medicine (BrAIN; www.brain.med.br) has pursued an objective, technology-driven approach to ectasia risk assessment. This collaborative initiative combines cutting-edge AI technology with clinical expertise, aiming to elevate ectasia detection and foster ophthalmic innovation. BrAIN welcomes all those who share in this mission to advance the field of refractive surgery.
BACKGROUND: ECTASIA RISK AND DETECTION
An early case report of iatrogenic corneal ectasia following LASIK noted central corneal steepening visible on corneal topography.2 Initial estimates of post-LASIK ectasia incidence have since declined markedly, from 0.66% to a recent low of 0.033%.3-7 This encouraging trend is attributed to improvements in preoperative screening and surgical techniques, along with a better understanding of corneal biomechanics and environmental stressors that influence ectasia risk.
MULTIMODAL IMAGING IN ECTASIA DIAGNOSIS
A practical assessment of a patient’s ectasia risk requires multimodal imaging, which provides a comprehensive profile of corneal health to gauge ectasia susceptibility better. Combining traditional Scheimpflug tomography and corneal biomechanical analysis offers distinct insights into corneal structure. Additional diagnostic tools, such as OCT-based layered or segmental tomography for epithelial thickness mapping, may provide further valuable data. A thorough preoperative assessment should also include ocular aberrometry, axial length measurement, and ocular surface evaluation.
In addition to imaging, a comprehensive ectasia risk assessment integrates expanded diagnostics, encompassing patient history, overall ocular health, and even genetics and molecular biology. This multilayered approach allows clinicians to provide targeted counseling, including guidance on behaviors such as eye rubbing, which can weaken corneal biomechanics.
AN AI-ENHANCED ASSESSMENT OF ECTASIA RISK
AI can help analyze the extensive data produced by multimodal diagnostic devices. The Applied Ancient Intelligence and Applied Artificial Intelligence algorithm combines clinical experience with AI-driven analysis to provide real-time support for ectasia risk detection. Rather than replace human decision-making, this approach aims to augment physicians’ expertise by providing them with tools to interpret complex data more precisely. Tomographic and biomechanical data are leveraged to detect subtle indications of ectasia susceptibility that may not be evident with conventional topography.
The Corneal Biomechanical Index (CBI)8 from the Corvis ST (Oculus Optikgeräte) and the Tomographic and Biomechanical Index (TBI), which includes tomographic data from the Pentacam (Oculus Optikgeräte), represent advances in ectasia detection. The CBI leverages high-speed Scheimpflug imaging to capture intricate details of corneal biomechanical behavior, such as deformation amplitude and corneal velocity, which provide insights into corneal stiffness. This information not only supports accurate IOP estimation but also aids in evaluating biomechanical stability. Key biomechanical metrics, including deformation amplitude ratio and Ambrósio’s relational thickness index, contribute to a robust assessment.
The TBI further refines ectasia risk evaluation by integrating corneal shape into biomechanical data, providing a more individualized risk profile. Enhanced by a random forest model optimized with data from a multicenter database, the TBI demonstrates improved sensitivity and specificity for ectasia detection.9 Together, the CBI and TBI offer a data-driven screening approach that allows clinicians to identify high-risk patients before ectasia manifests clinically.
ADVANCES IN AI FOR LASER VISION CORRECTION RISK ASSESSMENT
The Relational Tissue Altered (RTA) index, developed by Aydano Machado, MD, PhD, represents a substantial advancement in the assessment of ectasia risk. This AI-powered tool integrates essential parameters, including corneal thinnest pachymetry, patient age, LASIK flap characteristics (or laser-assisted lenticule extraction cap parameters), and ablation depth (or lenticule thickness), to provide a comprehensive evaluation of the biomechanical impact of laser vision correction (Figure). Both sensitivity and specificity of the RTA index surpass those of earlier methods, such as residual stromal bed and percentage tissue altered. A robust study of over 3,000 stable cases and more than 100 cases with ectasia validated the RTA’s superior accuracy in identifying high-risk procedures.
Figure. The BrAIN interface displays ectasia risk assessment parameters for a 27-year-old patient with corrected visual acuity of 20/15. For a -4.75 D surface ablation, the RTA index was elevated at 0.456 (above the 0.2 threshold), with a PTA of 26% and a residual stromal bed (RSB) of 364 µm. This is the fellow eye in a case of very asymmetric ectasia, showing relatively normal topography. The BrAIN TBI score was high at 0.96, along with a BAD-Dv3 of 1.86 and an inferior–superior (IS) value of 1.25, indicating susceptibility to ectasia. This comprehensive risk profile demonstrates the utility of combining indices to detect subtle ectatic changes.
The BrAIN Enhanced Ectasia Software combines the complementary strengths of the TBI and the RTA index into a unified, advanced ectasia risk assessment platform. This scoring system incorporates the two-hit hypothesis, which considers both intrinsic corneal susceptibility and the impact of laser vision correction. By integrating these factors, the software enables clinicians to stratify each patient’s ectasia susceptibility with unparalleled precision, improving surgical planning and patient safety.
CONCLUSION
Preventing corneal ectasia after refractive surgery requires advanced imaging, AI-assisted analysis, and a thorough understanding of corneal biomechanics. Tools such as the CBI, TBI, and RTA index can improve physicians’ ability to assess ectasia risk and minimize complications during surgical planning. The goal is to provide an individualized risk assessment based on each patient’s corneal characteristics and the specifics of the laser vision correction procedure to optimize both safety and visual outcomes.
1. Seiler T, Koufala K, Richter G. Iatrogenic keratectasia after laser in situ keratomileusis. J Refract Surg. 1998;14(3):312-317.
2. Seiler T, Quurke AW. Iatrogenic keratectasia after LASIK in a case of forme fruste keratoconus. J Cataract Refract Surg. 1998;24(7):1007-1009.
3. Pallikaris IG, Kymionis GD, Astyrakakis NI. Corneal ectasia induced by laser in situ keratomileusis. J Cataract Refract Surg. 2001;27(11):1796-1802.
4. Spadea L, Cantera E, Cortes M, Conocchia NE, Stewart CW. Corneal ectasia after myopic laser in situ keratomileusis: a long-term study. Clin Ophthalmol. 2012;6:1801-1813.
5. Rad AS, Jabbarvand M, Saifi N. Progressive keratectasia after laser in situ keratomileusis. J Refract Surg. 2004;20(5 suppl):S718-S722.
6. Moshirfar M, Smedley JG, Muthappan V, Jarsted A, Ostler EM. Rate of ectasia and incidence of irregular topography in patients with unidentified preoperative risk factors undergoing femtosecond laser-assisted LASIK. Clin Ophthalmol. 2014;8:35-42.
7. Bohac M, Koncarevic M, Pasalic A, et al. Incidence and clinical characteristics of post LASIK ectasia: a review of over 30,000 LASIK cases. Semin Ophthalmol. 2018;33(7-8):869-877.
8. Vinciguerra R, Ambrosio R, Wang Y, et al. Detection of keratoconus with a new Corvis Biomechanical Index optimized for Chinese populations. Am J Ophthalmol. 2023;252:182-187.
9. Ambrósio R Jr, Lopes BT, Faria-Correia F, et al. Integration of Scheimpflug-based corneal tomography and biomechanical assessments for enhancing ectasia detection. J Refract Surg. 2017;33(7):434-443.
10. Ambrósio R Jr, Machado AP, Leão E, et al. Optimized artificial intelligence for enhanced ectasia detection using Scheimpflug-based corneal tomography and biomechanical data. Am J Ophthalmol. 2023;251:126-142.
