Castillejo Becerra, Clara M

ImportanceSpecialized ophthalmology terminology limits comprehension for nonophthalmology clinicians and professionals, hindering interdisciplinary communication and patient care. The clinical implementation of large language models (LLMs) into practice has to date been relatively unexplored.ObjectiveTo evaluate LLM–generated plain language summaries (PLSs) integrated into standard ophthalmology notes (SONs) in improving diagnostic understanding, satisfaction, and clarity.Design, Setting, and ParticipantsRandomized quality improvement study conducted from February 1, 2024, to May 31, 2024, including data from inpatient and outpatient encounters in a single tertiary academic center. Participants were nonophthalmology clinicians and professionals and ophthalmologists. The single inclusion criterion was any encounter note generated by an ophthalmologist during the study dates. Exclusion criteria were (1) lack of established nonophthalmology clinicians and professionals for outpatient encounters and (2) procedure-only patient encounters.InterventionAddition of LLM-generated plain language summaries to ophthalmology notes.Main Outcomes and MeasuresThe primary outcome was survey responses from nonophthalmology clinicians and professionals assessing understanding, satisfaction, and clarity of ophthalmology notes. Secondary outcomes were survey responses from ophthalmologists evaluating PLS in terms of clinical workflow and accuracy, objective measures of semantic quality, and safety analysis.ResultsA total of 362 (85%) nonophthalmology clinicians and professionals (33.0% response rate) preferred the PLS to SON. Demographic data on age, race and ethnicity, and sex were not collected. Nonophthalmology clinicians and professionals reported enhanced diagnostic understanding (percentage point increase, 9.0; 95% CI, 0.3-18.2; P = .01), increased note detail satisfaction (percentage point increase, 21.5; 95% CI, 11.4-31.5; P < .001), and improved explanation clarity (percentage point increase, 23.0; 95% CI, 12.0-33.1; P < .001) for notes containing a PLS. The addition of a PLS was associated with reduced comprehension gaps between clinicians who were comfortable and uncomfortable with ophthalmology terminology (from 26.1% [95% CI, 13.7%-38.6%; P < .001] to 14.4% [95% CI, 4.3%-24.6%; P > .06]). PLS semantic analysis found high meaning preservation (bidirectional encoder representations from transformers score mean F1 score: 0.85) with greater readability than SONs (Flesch Reading Ease: 51.8 vs 43.6; Flesch-Kincaid Grade Level: 10.7 vs 11.9). Ophthalmologists (n = 489; 84% response rate) reported high PLS accuracy (90% [320 of 355] a great deal) with minimal review time burden (94.9% [464 of 489] ≤1 minute). PLS error rate on ophthalmologist review was 26% (126 of 489). A total of 83.9% (104 of 126) of errors were deemed low risk for harm and none had a risk of severe harm or death.Conclusions and RelevanceIn this study, use of LLM-generated PLSs was associated with enhanced comprehension and satisfaction among nonophthalmology clinicians and professionals, which might aid interdisciplinary communication. Careful implementation and safety monitoring are recommended for clinical integration given the persistence of errors despite physician review.

Purpose: To investigate patterns in oculoplastic surgery telemedicine visits at our institution during the pandemic period. Methods: Retrospective chart review of all patients who had a virtual consultation with the oculoplastic surgery service at Mayo Clinic Rochester between March 1, 2020, and March 1, 2021. Results: There were a total of 148 patients. Mean age was 65 years (95% confidence interval [CI]: 62–68). The mean driving distance to the medical center was 468 kilometers (95% CI: 352–586) and 15 (10%) patients came from states with no practicing American Society of Ophthalmic Plastic and Reconstructive Surgery (ASOPRS) members. Out of the 205 virtual visits, 35 (17%) were new, 45 (22%) were return, and 125 (61%) were postoperative. Conversion rate to surgery was 60%. A comparison between type of virtual visits (new vs. return vs. postoperative) revealed that a diagnosis of eyelid malposition was most frequent in postoperative visits ( p = < 0.001), skin malignancy was most frequent in new visits ( p = 0.009), and orbital tumors ( p = 0.018) and thyroid eye disease ( p = < 0.001) were most frequent in return visits. Most virtual encounters had supportive media attached including external photographs in 127 (62%) visits and other imaging or testing in 18 (9%) visits. The average virtual visit lasted 18 min (95% CI: 14–22) with technical difficulties noted in 9 (4%) visits. Unanticipated conversion to in-person visits was noted in three (1%) cases. Conclusions: A wide range of oculoplastic surgery conditions was managed virtually during the COVID-19 pandemic. Utilization of initial virtual visit versus follow-up care varied by diagnosis. Virtual visits may improve access to oculoplastic subspecialty care for patients from underserved areas.

Purpose: The aim of this study was to describe the indications and outcomes of flap amputation after laser in situ keratomileusis (LASIK) at a referral-based institution. Methods: In this retrospective consecutive case series, medical records of patients who underwent LASIK flap amputation at Mayo Clinic, Rochester, MN, between January 1, 1998, and January 31, 2023, were reviewed. Results: Fifteen eyes (15 patients) underwent flap amputation during the study period. The median age was 45 years (range, 25–71 years), and 8 patients (53%) were men. The median uncorrected visual acuity before flap amputation was 20/200 (range 20/40-hand motions). Indications for flap amputation included epithelial ingrowth (n = 6, 40%), infectious keratitis (n = 6, 40%), diffuse lamellar keratitis (n = 1, 7%), vegetative foreign body (n = 1, 7%), and astigmatism from fixed flap striae (n = 1, 7%). The median duration of follow-up was 8 months (range 1–234 months). Subsequent corneal interventions included chelation of calcific band keratopathy (n = 1, 7%), lamellar keratoplasty (n = 1, 7%), penetrating keratoplasty (n = 2, 18%), keratoprosthesis (n = 1, 7%), and rigid contact lens wear (n = 4, 27%). The final median best visual acuity was 20/25 (range 20/20–20/200). Compared with noninfectious indications for flap amputation, eyes with infectious indications had worse baseline median uncorrected visual acuity (hand motions vs. 20/63, P < 0.001), were more likely to undergo major corneal surgical intervention (50% vs. 11%), and had worse final median best visual acuity (20/50 vs. 20/20, P = 0.018). Conclusions: LASIK flap amputation is sometimes necessary to control threatening corneal diseases. Excellent visual outcomes were achieved in most cases, albeit with additional intervention or rigid contact lens wear.

To determine the frequency, characteristics, and clinical course of ophthalmic side effects associated with systemic BRAF inhibitor therapy. Medical records of patients taking BRAF inhibitors for the treatment of systemic malignances at Mayo Clinic, Rochester from 01/01/2010 to 08/30/2021, were retrospectively reviewed. Of 901 patients, 14 (1.6%) patients experienced an ophthalmic side effect. Mean age at presentation of the side effect was 60 years (median 59, range 50–80) and 11 (79%) were male. The most common side effect was uveitis in 7 (50%) patients, followed by dry eye in 4 (29%) patients, and central serous chorioretinopathy in 2 (14%) patients, with singular cases of cranial nerve VI palsy and conjunctival edema. A comparison between individual BRAF inhibitors (vemurafenib vs. dabrafenib vs. encorafenib) revealed that patients taking encorafenib had a shorter interval to any ophthalmic adverse event (mean 55.6 vs. 9.8 vs. 4.0 months, p = 0.03) and were the only patients to experience documented dry eye syndrome (DES) in this series. Outcomes were known in 13 (93%) patients, and ophthalmic adverse effects resolved or were controlled without discontinuing therapy in 10 (77%). Uveitis was successfully treated with topical corticosteroids in 4 patients, while 3 patients with refractory uveitis (2 with panuveitis and 1 with unspecified uveitis) required discontinuation of BRAF inhibitor therapy. Ophthalmic adverse events related to systemic BRAF inhibitor use are rare, with estimated frequency of 1.6%. Most events can be treated with local ophthalmic therapy. BRAF inhibitors provide life-saving therapy, and their discontinuation should be avoided.

AbstractThe protein phosphatase Phlpp1 is an essential enzyme for proper chondrocyte function. Altered Phlpp1 levels are associated with cancer and degenerative diseases such as osteoarthritis. While much is known about the post‐transcriptional mechanisms controlling Phlpp1 levels, transcriptional regulation of the Phlpp1 gene locus is underexplored. We previously showed that CpG methylation of the PHLPP1 promoter is lower in osteoarthritic cartilage than in normal cartilage, and indirectly correlates with gene expression. Here we further defined the effects of DNA methylation on PHLPP1 promoter activity in chondrocytes. We cloned a 1791 bp fragment of the PHLPP1 promoter (−1589:+202) and found that the first 500 bp were required for maximal promoter activity. General methylation of CpG sites within this fragment significantly blunts transcriptional activity, whereas site‐specific methyltransferases HhaI or HpaII decrease transcriptional activation by approximately 50%. We located putative FoxO consensus sites within the PHLPP1 promoter region. Inhibition of DNA methylation by incorporation of 5‐azacytidine increases Phlpp1 mRNA levels, but FoxO inhibition abolishes this induction. To determine which FoxO transcription factor mediates Phlpp1 expression, we performed overexpression and siRNA‐mediated knock down experiments. Overexpression of FoxO3a, but not FoxO1, increases Phlpp1 levels. Likewise, siRNAs targeting FoxO3a, but not FoxO1, diminished Phlpp1 levels. Last, FoxO inhibition increases glycosaminoglycan staining of cultured chondrocytes and leads to concomitant increases in FGF18 and HAS2 expression. Together, these data demonstrate that CpG methylation and FoxO3a regulate PHLPP1 expression.