Introduction OpenAI recently introduced the ability to create custom generative pre-trained transformers (cGPTs) using text-based instruction and/or external documents using retrieval-augmented generation (RAG) architecture without coding knowledge. This study aimed to analyze the features of ophthalmology-related cGPTs and explore their potential utilities. Methods Data collection took place on January 20 and 21, 2024, and custom GPTs were found by entering ophthalmology keywords into the “Explore GPTS” section of the website. General and specific features of cGPTs were recorded, such as knowledge other than GPT-4 training data. The instruction and description sections were analyzed for compatibility using the Likert scale. We analyzed two custom GPTs with the highest Likert score in detail. We attempted to create a convincingly presented yet potentially harmful cGPT to test safety features. Results We analyzed 22 ophthalmic cGPTs, of which 55% were for general use and the most common subspecialty was glaucoma (18%). Over half (55%) contained knowledge other than GPT-4 training data. The representation of the instructions through the description was between “Moderately representative” and “Very representative” with a median Likert score of 3.5 (IQR 3.0–4.0). The instruction word count was significantly associated with Likert scores (P = 0.03). Tested cGPTs demonstrated potential for specific conversational tone, information, retrieval and combining knowledge from an uploaded source. With these safety settings, creating a malicious GPT was possible. Conclusions This is the first study to our knowledge to examine the GPT store for a medical field. Our findings suggest that these cGPTs can be immediately implemented in practice and may offer more targeted and effective solutions compared to the standard GPT-4. However, further research is necessary to evaluate their capabilities and limitations comprehensively. The safety features currently appear to be rather limited. It may be helpful for the user to review the instruction section before using a cGPT.

Straatsma syndrome is the triad of myelinated retinal nerve fibers, myopia, and amblyopia and may be associated with strabismus, nystagmus, hypoplastic optic nerve, and heterochromia iridum. The degree of anisometropia, presence of strabismus, extent of myelination, and macular involvement have been reported to be associated with poor visual acuity after occlusion therapy for amblyopia in this syndrome. Here we present two cases of Straatsma syndrome with different responses to occlusion therapy and discuss their treatment responses according to prognostic factors for post-occlusion visual acuity.

This study aimed to report the demographic and clinical characteristics of diabetic macular edema (DME) patients treated with intravitreal injection (IVI) of anti-vascular endothelial growth factors (anti-VEGF) and provide an overview of outcomes during routine clinical practice in Türkiye. This retrospective, real-world study included 1,372 eyes (854 patients) treated with a pro re nata protocol by 21 ophthalmologists from 8 tertiary clinics on the Asian side of the Marmara region of Türkiye (MARMASIA Study Group). Five cohort groups were established by collecting the patients' baseline and 3, 6, 12, 24, and 36-month follow-up data, where each subsequent cohort may include the previous. Changes in best-corrected visual acuity (BCVA, approximate ETDRS letters) and central macular thickness (CMT, μm), number of visits and IVI, and rates of anti-VEGF switch and intravitreal dexamethasone implant (IDI) combination were evaluated. The 3, 6, 12, 24, and 36-month cohorts included 1372 (854), 1352 (838), 1185 (722), 972 (581), and 623 (361) eyes (patients), respectively. The mean baseline BCVA and CMT were 51.4±21.4 letters and 482.6±180.3 μm. The mean changes from baseline in BCVA were +7.6, +9.1, +8.0, +8.6, and +8.4 letters, and in CMT were -115.4, -140.0, -147.9, -167.3, and -215.4 μm at the 3, 6, 12, 24, and 36-month visits (p<0.001 for all). The median cumulative number of anti-VEGF IVI was 3.0, 3.0, 5.0, 7.0, and 9.0, respectively. The overall anti-VEGF switch and IDI combination rates were 18.5% (253/1372 eyes) and 35.0% (480/1372 eyes), respectively. This largest real-life study of DME from Türkiye demonstrated BCVA gains inferior to randomized controlled trials, mainly due to the lower number of IVI. However, with the lower baseline BCVA and higher IDI combination rates in our cohorts, these gains were relatively superior to other real-life study counterparts.