Journal of Pediatric Ophthalmology & Strabismus最新文献

Purpose: To evaluate the quality and readability of ChatGPT-4o-generated (ChatGPT) (OpenAI) patient education materials (PEMs) about pediatric ophthalmologic surgical procedures and compare these to PEMs from the American Association for Pediatric Ophthalmology and Strabismus (AAPOS) website.

Methods: The authors prompted ChatGPT-4o to provide PEMs on four procedures-strabismus surgery without adjustable sutures, strabismus surgery with adjustable sutures, pediatric cataract surgery, and nasolacrimal duct probing. The prompt requested responses at a 6th grade level in both Spanish and English. English ChatGPT responses were compared to AAPOS PEMs on quality (using the Quality of Generated Language Outputs for Patients [QGLOP] scale) and readability. English and Spanish ChatGPT responses were also compared on quality and readability.

Results: Based on average scores from the four procedures, AAPOS PEMs were superior to English Chat-GPT responses on the accuracy, currency, and tone subscales of the QGLOP score (4.0 ± 0 vs 2.79 ± 0.79, P = .0021; 3.79 ± 0.26 vs 3.38 ± 0.71, P = .033; 4.0 ± 0 vs 3.42 ± 0.69, P = .042, respectively). There was no significant difference in readability between AAPOS PEMs and English ChatGPT responses. English and Spanish ChatGPT responses did not significantly differ on quality or readability.

Conclusions: ChatGPT-4o-generated PEMs on pediatric ophthalmologic surgical conditions are currently inferior in quality to PEMs on the AAPOS website. However, because ChatGPT is continually being updated and trained, this study should be repeated in the future to determine whether metrics improve over time.

A healthy 17-year-old boy was diagnosed as having blastic plasmacytoid dendritic cell neoplasm (BPDCN), a rare hematologic malignancy. Vision changes prompted ophthalmology consultation, revealing peripheral field changes, retinal findings, and bilateral optic disc edema. Imaging confirmed bilateral optic neuropathy. This case highlights the potential for pediatric ocular involvement in BPDCN.

Purpose: To investigate the impact of strabismus surgery on fundus hemodynamics using swept-source optical coherence tomography angiography (SS-OCTA).

Methods: A total of 124 patients with strabismus undergoing horizontal rectus muscle surgery were included in this non-randomized contemporaneous controlled trial. Patients were divided into unilateral recession (UR) and recession-resection (RR) groups based on the number of rectus muscles operated on. The UR group was further subdivided into lateral rectus (LR) and medial rectus (MR) subgroups. SS-OCTA imaging was performed preoperatively, and at 1 day, 1 week, and 1 month postoperatively, using a 12 × 12 mm macular scan to quantify hemodynamic parameters: vessel density (VD) of superficial capillary plexus (SCP) and deep capillary plexus (DCP), choroidal thickness (CT), choriocapillaris perfusion (CCP), choroidal vascular index (CVI), and choroidal vascular volume (CVV).

Results: Horizontal rectus muscle surgery induced transient fundus hemodynamic changes, with significant differences observed near the fovea and peripheral retina, but not in the operated muscle quadrant. SCP and DCP increased in the peripheral retina postoperatively, while remaining stable near the fovea. CCP increased, with earlier changes near the fovea. CVV and CT increased at 1 week postoperatively, whereas CVI decreased, all returning to baseline by 1 month. The choroid exhibited more pronounced hemodynamic changes than the retina. RR surgery caused greater changes than UR, with LR surgery having a greater impact than MR surgery.

Conclusions: This study highlights transient ocular hemodynamic changes following horizontal rectus muscle surgery, varying by retinal region and number of muscles operated, providing new insights into fundus hemodynamics using SS-OCTA.

Purpose: To evaluate the success of Chat Generative Pre-trained Transformer (ChatGPT; OpenAl), Google Gemini (Alphabet, Inc), and Microsoft Copilot (Microsoft Corporation) artificial intelligence (AI) programs, which are offered free of charge by three different manufacturers, in answering questions related to pediatric ophthalmology correctly and to investigate whether they are superior to each other.

Methods: ChatGPT, Gemini, and Copilot were each asked 100 multiple-choice questions from the Ophtho-Questions online question bank, which is widely used for preparing for the high-stakes Ophthalmic Knowledge Evaluation Program examination. Their answers were compared to the official answer keys and categorized as correct or incorrect. The readability of the responses was assessed using the Flesch-Kincaid Grade Level, Flesch Reading Ease Score, and the Coleman-Liau Index.

Results: ChatGPT, Gemini, and Copilot chatbots answered 61 (61%), 60 (60%), and 74 (74%) questions correctly, respectively. The Copilot AI program had a significantly higher rate of correct answers to questions than ChatGPT and Gemini (P = .049 and .035). Three readability analyses revealed that Copilot had the highest average score, followed by ChatGPT and Gemini, which were more challenging than the recommended level.

Conclusions: Although AI chatbots can serve as useful tools for acquiring information on pediatric ophthalmology, their responses should be interpreted with caution due to potential inaccuracies.

Purpose: To evaluate the effect of recession with Y-splitting of the superior rectus muscle (SR) on correction of dissociated vertical deviation (DVD).

Methods: Twelve consecutive patients with manifest DVD greater than 10 prism diopters (PD) received 7-mm SR recession with Y-splitting, with concurrent treatment of any horizontal deviation. Preoperative and postoperative evaluation of ductions and versions, ocular deviation and pattern was done. A successful outcome was considered as 0 to 6 prism diopters (PD) of vertical and 0 to 10 PD of horizontal deviation.

Results: Surgery was performed on 19 SR muscles of 12 patients. Successful correction of DVD was achieved in all operated eyes (100%) after a mean follow-up of 6.9 ± 4.05 months. DVD became manifest in the fellow eye in 2 of 5 patients following unilateral surgery. Hypotropia (that became occasional after amblyopia therapy) was seen in one eye after unilateral surgery. Mild SR underaction was seen in 2 eyes and marked underaction in 1 eye postoperatively. Y-pattern was documented in 2 cases, V-pattern in 1 case, and X-pattern in 1 case postoperatively.

Conclusions: SR recession with Y-splitting could be a promising option for correction of DVD. Bilateral surgery is advocated even if unilaterally manifest. Further studies on larger series of patients are recommended to consolidate these preliminary results.

Purpose: To describe a rare inherited disorder in five patients presenting with horizontal gaze palsy along with characteristic abnormalities on neuroimaging.

Methods: A retrospective review was conducted from January 2016 to December 2022 of all patients with the diagnosis of horizontal gaze palsy with progressive scoliosis (HGPPS). Neuroimaging was performed in all patients and genetic testing in those who consented.

Results: All five patients had limitation of abduction and adduction in both eyes on attempted horizontal gazes with intact elevation, depression, and convergence. Two patients presented with synergistic convergence. Esotropia requiring strabismus surgery was present in one patient. Magnetic resonance imaging of the brain and orbit in all patients showed brainstem hypoplasia with "split pons" sign and a "butterfly configuration" of the medulla, which are characteristic of HGPPS. Genetic analysis of 3 patients confirmed the presence of ROBO3 gene mutations. Scoliosis was noted in 2 patients.

Conclusions: Early diagnosis and a multidisciplinary approach is important in the management of such cases. Phenotypic variations such as in the cases described may be observed in HGPPS.

Purpose: To evaluate the quality and accuracy of artificial intelligence (AI)-generated images depicting pediatric ophthalmology pathologies compared to human-illustrated images, and assess the readability, quality, and accuracy of accompanying AI-generated textual information.

Methods: This cross-sectional comparative study analyzed outputs from DALL·E 3 (OpenAI) and Gemini Advanced (Google). Nine pediatric ophthalmology pathologies were sourced from the American Association for Pediatric Ophthalmology and Strabismus (AAPOS) "Most Common Searches." Two prompts were used: Prompt A asked large language models (LLMs), "What is [insert pathology]?" Prompt B requested text-to-image generators (TTIs) to create images of the pathologies. Textual responses were evaluated for quality using published criteria (helpfulness, truthfulness, harmlessness; score 1 to 15, ≥ 12: high quality) and readability using Simple Measure of Gobbledygook (SMOG) and Flesch-Kincaid Grade Level (≤ 6th-grade level: readable). Images were assessed for anatomical accuracy, pathological accuracy, artifacts, and color (score 1 to 15, ≥ 12: high quality). Human-illustrated images served as controls.

Results: DALL·E 3 images were of poor quality (median: 7; range: 3 to 15) and significantly worse than human-illustrated controls (median: 15; range: 9 to 15; P < .001). Pathological accuracy was also poor (median: 1). Textual information from ChatGPT-4o and Gemini Advanced was high quality (median: 15) but difficult to read (Chat-GPT-4o: SMOG: 8.2, FKGL: 8.9; Gemini Advanced: SMOG: 8.5, FKGL: 9.3).

Conclusions: Text-to-image generators are poor at generating images of common pediatric ophthalmology pathologies. They can serve as adequate supplemental tools for generating high-quality accurate textual information, but care must be taken to tailor generated text to be readable by users.

Purpose: To present a comparative analysis of the economic and workforce issues impacting access to pediatric ophthalmology care in Italy and the United States in the post-pandemic era.

Methods: Data collection for this cross-sectional study occurred between August 2023 and July 2024, using publicly available workforce and demographic data to calculate supply-demand (provider-patient) ratios. Additionally, qualitative data were obtained through a structured questionnaire distributed to fellowship-trained pediatric ophthalmologists in Italy.

Results: The provider-patient ratio was 1:69,700 in the United States and approximately 1:445,000 in Italy, reflecting significant workforce shortages in both countries, with a greater deficit observed in Italy. Contributing factors in Italy include the absence of a formal fellowship training program, inadequate reimbursement, and a reduction in critical resources and facilities. In the United States, similar challenges have been exacerbated by the coronavirus disease 2019 (COVID-19) pandemic, resulting in reduced practice viability, limited surgical training, and declining interest in the sub-specialty among trainees.

Conclusions: Although health care infrastructures differ between Italy and the United States, the long-standing crisis in Italy may serve as a cautionary model for the potential trajectory of pediatric ophthalmology in the United States should current trends persist. Without strategic interventions to improve workforce retention and training, both countries remain at risk for erosion of access to timely and equitable pediatric ophthalmic care.