RESUMO
BACKGROUND: We aimed to define the capability of three different publicly available large language models, Chat Generative Pretrained Transformer (ChatGPT-3.5), ChatGPT-4 and Google Gemini in analysing retinal detachment cases and suggesting the best possible surgical planning. METHODS: Analysis of 54 retinal detachments records entered into ChatGPT and Gemini's interfaces. After asking 'Specify what kind of surgical planning you would suggest and the eventual intraocular tamponade.' and collecting the given answers, we assessed the level of agreement with the common opinion of three expert vitreoretinal surgeons. Moreover, ChatGPT and Gemini answers were graded 1-5 (from poor to excellent quality), according to the Global Quality Score (GQS). RESULTS: After excluding 4 controversial cases, 50 cases were included. Overall, ChatGPT-3.5, ChatGPT-4 and Google Gemini surgical choices agreed with those of vitreoretinal surgeons in 40/50 (80%), 42/50 (84%) and 35/50 (70%) of cases. Google Gemini was not able to respond in five cases. Contingency analysis showed significant differences between ChatGPT-4 and Gemini (p=0.03). ChatGPT's GQS were 3.9±0.8 and 4.2±0.7 for versions 3.5 and 4, while Gemini scored 3.5±1.1. There was no statistical difference between the two ChatGPTs (p=0.22), while both outperformed Gemini scores (p=0.03 and p=0.002, respectively). The main source of error was endotamponade choice (14% for ChatGPT-3.5 and 4, and 12% for Google Gemini). Only ChatGPT-4 was able to suggest a combined phacovitrectomy approach. CONCLUSION: In conclusion, Google Gemini and ChatGPT evaluated vitreoretinal patients' records in a coherent manner, showing a good level of agreement with expert surgeons. According to the GQS, ChatGPT's recommendations were much more accurate and precise.
RESUMO
BACKGROUND: We aimed to explore microvascular changes evaluated with optical coherence tomography angiography (OCTA) in patients undergoing epiretinal membrane (ERM) pars-plana vitrectomy (PPV) combined with intravitreal Ozurdex implantation, compared with standard PPV. (2) Methods: Prospective interventional analysis on 25 eyes undergoing PPV + Ozurdex (Group A) and 25 eyes undergoing PPV alone. Best corrected visual acuity (BCVA) and OCTA parameters, such as vessel density (VD) of the superficial and deep capillary plexi (SCP and DCP) in the whole 6.4 mm × 6.4 mm and fovea area, were evaluated preoperatively and 3 months after surgery. (3) Results: Postoperative BCVA significantly improved in both groups. No cases of post-operative cystoid macular edema (CME) were reported in Group A vs. two eyes in Group B. In Group A we found a statistically significant increase of SCP's VD in either the whole image (from 42.1 ± 4.1 to 45.6 ± 4.3%, p = 0.01) and the fovea image (from 38.5 ± 7.5 to 41.7 ± 4.2%, p = 0.03). In Group B, we reported no significant variations in the SCP's VDs. In the DCP, VD significantly increased only in the whole image in Group A. Stage 4 ERMs showed the greatest improvement in VD, especially in Group A. (4) Conclusions: Intraoperative Ozurdex prompted a significant BCVA recovery and limited the occurrence of postoperative CME compared to the standard procedure. Moreover, Ozurdex implant is associated with a better restoration of microvascular structure in SCP and DCP.
