Chatbots as Patient Education Resources for Aesthetic Facial Plastic Surgery: Evaluation of ChatGPT and Google Bard Responses.

Background: ChatGPT and Google Bard™ are popular artificial intelligence chatbots with utility for patients, including those undergoing aesthetic facial plastic surgery. Objective: To compare the accuracy and readability of chatbot-generated responses to patient education questions regarding aesthetic facial plastic surgery using a response accuracy scale and readability testing. Method: ChatGPT and Google Bard™ were asked 28 identical questions using four prompts: none, patient friendly, eighth-grade level, and references. Accuracy was assessed using Global Quality Scale (range: 1-5). Flesch-Kincaid grade level was calculated, and chatbot-provided references were analyzed for veracity. Results: Although 59.8% of responses were good quality (Global Quality Scale ≥4), ChatGPT generated more accurate responses than Google Bard™ on patient-friendly prompting (p < 0.001). Google Bard™ responses were of a significantly lower grade level than ChatGPT for all prompts (p < 0.05). Despite eighth-grade prompting, response grade level for both chatbots was high: ChatGPT (10.5 ± 1.8) and Google Bard™ (9.6 ± 1.3). Prompting for references yielded 108/108 of chatbot-generated references. Forty-one (38.0%) citations were legitimate. Twenty (18.5%) provided accurately reported information from the reference. Conclusion: Although ChatGPT produced more accurate responses and at a higher education level than Google Bard™, both chatbots provided responses above recommended grade levels for patients and failed to provide accurate references.

Device-related outcomes following hypoglossal nerve stimulator implantation.

STUDY OBJECTIVES: Hypoglossal nerve stimulation (HGNS) has been widely used to treat obstructive sleep apnea in selected patients. Here we evaluate rates of revision and explant related to HGNS implantation and assess types of adverse events contributing to revision and explant. METHODS: Post-market surveillance data for HGNS implanted between January 1, 2018 and March 31, 2022, were collected. Event rates and risk were calculated using the post-market surveillance event counts and sales volume over the same period. Indications were categorized for analysis. Descriptive statistics were reported and freedom from explant or revision curves were grouped by year of implantation. RESULTS: Of the 20,881 HGNS implants assessed, rates of explant and revision within the first year were 0.723% and 1.542%, respectively. The most common indication for explant was infection (0.378%) and for revision was surgical correction (0.680%). Of the 5,820 devices with three-year post-implantation data, the rate of explant was 2.680% and of revision was 3.557%. During this same interval, elective removal (1.478%) was the most common indication, and for revisions, surgical correction (1.134%). CONCLUSIONS: The efficacy of HGNS is comparable in the real world setting to published clinical trial data. Rates of explant and revision are low, supporting a satisfactory safety profile for this technology.

Predictive capacity of immune-related adverse events and cytokine profiling in neoadjuvant immune checkpoint inhibitor trials for head and neck squamous cell carcinoma.

OBJECTIVES: Certain low-level immune-related adverse events (irAEs) have been associated with survival benefits in patients with various solid tumors on immune checkpoint inhibitors (ICIs). We aimed to investigate the association between irAEs and response to neoadjuvant ICIs in patients with head and neck squamous cell carcinoma (HNSCC) and to identify differences in circulating cytokine levels based on irAE status. METHODS: This was a retrospective cohort study including three neoadjuvant clinical trials from July 2017 to January 2022: NCT03238365 (nivolumab ± tadalafil), NCT03854032 (nivolumab ± BMS986205), NCT03618654 (durvalumab ± metformin). The presence and type of irAEs, pathologic treatment response, and survival were compared. Canonical linear discriminant analysis (LDA) was performed to identify combinations of circulating cytokines predictive of irAEs using plasma sample multiplex assay. RESULTS: Of 113 participants meeting inclusion criteria, 32 (28.3%) developed irAEs during treatment or follow-up. Positive p16 status was associated with irAEs (odds ratio [OR] 2.489; 95% CI 1.069-6.119; p = 0.043). irAEs were associated with pathologic treatment response (OR 3.73; 95% CI 1.34-10.35; p = 0.011) and with higher OS in the combined cohort (HR 0.319; 95% CI 0.113-0.906; p = 0.032). Patients with irAEs within the nivolumab cohort had significant elevations of select cytokines pre-treatment. Canonical LDA identified key drivers of irAEs among all trials, which were highly predictive of future irAE status. CONCLUSIONS: irAEs are associated with response to neoadjuvant ICI therapy in HNSCC and can serve as clinical indicators for improved clinical outcomes. irAEs can be predicted by concentrations of several circulating cytokines prior to treatment.