Purpose: To report delayed anterior stromal melt following corneal allogenic intrastromal ring segment (CAIRS) implantation in a patient without known systemic or ocular risk factors. Methods: A 62-year-old man with advanced bilateral keratoconus underwent femtosecond laser–assisted CAIRS implantation at 260 µm depth (900 µm tunnel width). Three months postoperatively, CDVA improved from 20/150− and 20/400 to 20/80 bilaterally and Kmax improved from 59.67 and 67.56 diopters (D) to 54.39 and 64.72 D in the right and left eyes, respectively. Optical coherence tomography (OCT) measured the segment at 624 µm in the right eye. Results: Ten months postoperatively, the right eye demonstrated focal epithelial thinning with fluorescein staining and melt localized over the inferotemporal segment; overlying stromal thickness decreased from 133 µm (week 1) to 7 µm on OCT. The left eye remained unaffected. Topical loteprednol, lubrication, and systemic doxycycline were initiated. Nineteen months postoperatively, OCT and topography demonstrated stromal stability. Despite astigmatic regression, CDVA remained better than baseline, and the CAIRS was retained. Conclusions: The large segment size placed superficially may have contributed to localized biomechanical stress, particularly in eyes with advanced keratoconus. Safe thresholds for segment/tunnel sizes, depth, and location warrant further studies.
Purpose: To evaluate refractive outcomes of large language model (LLM)–based selection of the optimal intraocular lens power calculator according to individual biometric parameters. Methods: Two hundred eight eyes undergoing routine cataract surgery were included. Ten leading LLMs (Gemini, Chat-GPT, Perplexity, Grok, DeepSeek, Kimi, Qwen, Doubao, ERNIE, and Hunyuan) were prompted to recommend “the single most appropriate formula” from Barrett Universal II (BUII), Cooke K6, EVO 2.0, Hill-RBF 3.0, Kane, and PEARL-DGS, based on axial length (AL), keratometry, anterior chamber depth, and lens thickness. Gemini was excluded because it consistently selected Kane. Precision (SD) and accuracy (absolute spherical equivalent prediction error [SEQ-PE]) were assessed with the Eyetemis Analysis Tool and compared with BUII and SRK/T, both overall and within AL subgroups. Results: Formula preferences varied markedly among LLMs. Overall, all LLMs and BUII demonstrated significantly greater precision than SRK/T, with Hunyuan, ChatGPT, and BUII showing highest precision and significantly improved accuracy compared to SRK/T (all P < .05). The leading percentages of eyes within the ±0.50 diopter (D) threshold were achieved by Perplexity (82.69%), ChatGPT (82.21%), Qwen (82.21%), and BUII (81.73%). Across AL subgroups, at least one LLM numerically outperformed BUII in both precision and accuracy; at the ±0.50 D threshold, certain models increased this percentage over BUII by approximately 3% when AL was ≤ 22 mm or ≥ 26 mm, and by 8% when 24.5 mm ≤ AL < 26 mm. Conclusions: Compared with BUII, current LLMs exhibit limited capacity to improve refractive outcomes by selecting the optimal formulas for routine practice; however, certain LLMs show promise as adjunctive tools for personalized selection tailored to specific ALs.
Purpose: To evaluate the 5-year safety and efficacy of the ELZA-sub400 protocol, an individualized stromal thickness–guided corneal cross-linking (CXL) technique, in eyes with progressive keratoconus and corneal stroma thickness of less than 400 µm. Methods: In this retrospective interventional case series, 39 eyes of 33 patients with progressive keratoconus or primary ectasia and stromal thickness between 214 and 398 µm underwent CXL using the ELZA-sub400 protocol. Following epithelial removal and riboflavin loading, ultraviolet-A irradiation at 3 mW/cm 2 was applied for durations tailored to stromal thickness, targeting a demarcation line (DL) 70 µm anterior to the corneal endothelium. Outcome measures included anterior radius of curvature over the central 3 mm (ARC3mm), maximum keratometry (Kmax), corrected distance visual acuity (CDVA), refractive error, stromal thickness, and corneal densitometry, assessed by Scheimpflug tomography (Pentacam HR; Oculus Optikgeräte GmbH) annually for 5 years. Mean follow-up was 3.6 ± 1.5 years (range: 1 to 5 years), with 18 of 39 eyes (46.2%) completing the 5-year follow-up. Results: At 5 years after CXL, 84.6% of eyes (33/39) remained stable or flattened (Kmax change ≤1.00 diopters [D]). Kmax decreased significantly from baseline (mean: −2.74 ± 2.65 D, P < .001). ARC3mm-derived dioptric power decreased significantly over 5 years (−1.29 D, P = .014). Median CDVA was unchanged (0.40 logMAR, P = .572). No significant change was observed in spherical ( P = .211) or cylindrical ( P = .280) refractive error. DL depth averaged 275 ± 61 µm; no endothelial decompensation occurred. Minimum corneal thickness declined modestly (median −10 µm, P = .049). Anterior densitometry increased slightly (+2.40 grayscale units, P = .007), remaining within mild clinical ranges. Conclusions: The ELZA-sub400 protocol safely stabilized keratoconus in 84.6% of ultrathin corneas over 5 years, with no signs of endothelial damage. Individualized fluence enabled effective treatment of corneas previously ineligible for standard CXL.
