RESUMO
OBJECTIVE: In this study, we aimed to characterize the frequency and distribution of ocular surgeries in patients with inherited retinal diseases (IRDs) and evaluate associated patient and disease factors. DESIGN: Retrospective cohort. PARTICIPANTS: Subjects aged ≥ 18 years who were followed at the Johns Hopkins Genetic Eye Disease Center. METHODS: We studied a retrospective cohort of patients with an IRD diagnosis to analyze the occurrence of laser and incisional surgeries. Subjects were categorized into 2 groups: central dysfunction (macular/cone/cone-rod dystrophy, "MCCRD group") and panretinal or peripheral dysfunction (retinitis pigmentosa-like, "RP group"). Genetic testing status was recorded. The association of patient and disease factors on the frequency, distribution, and timing of surgeries was analyzed. MAIN OUTCOME MEASURES: Prevalence, prevalence odds ratio (POR), hazard ratio (HR) of ophthalmic procedures by phenotype. RESULTS: A total of 1472 eyes of 736 subjects were evaluated. Among them, 31.3% (n = 230) had undergone ocular surgery, and 78.3% of those (n = 180/230) had a history of more than 1 surgery. A total of 602 surgical procedures were analyzed. Cataract extraction with intraocular lens implantation (CEIOL) was the most common (51.2%), followed by yttrium aluminum garnet capsulotomy, refractive surgery, retinal surgery, and others. Cataract extraction with intraocular lens implantation occurred more frequently in RP than in MCCRD subjects (POR, 2.59; P = 0.002). Retinitis pigmentosa subjects underwent CEIOL at a younger age than patients with MCCRD (HR, 2.11; P < 0.001). CONCLUSIONS: Approximately one-third of patients with IRD had a history of laser or incisional surgery. Cataract extraction with intraocular lens implantation was the most common surgery; its frequency and timing may be associated with the IRD phenotype. This data may inform the design of prospective research. Such efforts may illuminate routine clinical decision-making and contribute to surgical strategy development for cell and gene therapy delivery. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
RESUMO
PURPOSE: To classify fleck lesions and assess artificial intelligence (AI) in identifying flecks in Stargardt disease (STGD). METHODS: A retrospective study of 170 eyes from 85 consecutive patients with confirmed STGD. Fundus autofluorescence images were extracted, and flecks were manually outlined. A deep learning model was trained, and a hold-out testing subset was used to compare with manually identified flecks and for graders to assess. Flecks were clustered using K-means clustering. RESULTS: Of the 85 subjects, 45 were female, and the median age was 37 years (IQR 25-59). A subset of subjects (n=41) had clearly identifiable fleck lesions, and an AI was successfully trained to identify these lesions (average Dice score of 0.53, n=18). The AI segmentation had smaller (0.018 compared with 0.034 mm2, p<0.001) but more numerous flecks (75.5 per retina compared with 40.0, p<0.001), but the total size of flecks was not different. The AI model had higher sensitivity to detect flecks but resulted in more false positives. There were two clusters of flecks based on morphology: broadly, one cluster of small round flecks and another of large amorphous flecks. The per cent frequency of small round flecks negatively correlated with subject age (r=-0.31, p<0.005). CONCLUSIONS: AI-based detection of flecks shows greater sensitivity than human graders but with a higher false-positive rate. With further optimisation to address current shortcomings, this approach could be used to prescreen subjects for clinical research. The feasibility and utility of quantifying fleck morphology in conjunction with AI-based segmentation as a biomarker of progression require further study.