Facial Plastic Surgery & Aesthetic Medicine最新文献

Objective: To investigate the effectiveness of isotretinoin use in patients with thick nasal skin to improve rhinoplasty outcomes. Methods: The Population Intervention Comparison Outcome framework was utilized for selecting relevant studies for review. The databases of Medline, Embase, Cinahl, Central, Scopus, and Web of Science were searched in June 2024. Original research in adults (≥18 years) who underwent rhinoplasty and administered Isotretinoin pre- and/or postoperatively was included. Case reports, case series (<5 patients), and other letters were excluded. Results: This review identified, 3 (50%) retrospective studies and 3 (50%) randomized controlled trials on the topic. Although four studies demonstrated an improvement in patient's aesthetic satisfaction up to at least 6 months after rhinoplasty, only two studies without control groups, demonstrated a sustained improvement in patient's aesthetic satisfaction up to a year postoperatively. Conclusion: Based on the results of this study, the effectiveness of isotretinoin in rhinoplasty though promising could not be determined due to inconsistent use of treatment protocols, outcome measures, control groups, and long-term follow-up.

Importance: Bell's palsy (BP) is reported as the most common cause of facial paralysis, yet literature lacks a standardized definition of BP. Objective: To identify and categorize how the term "Bell's palsy" is defined and applied in published medical literature. Evidence Review: Randomized controlled trials, clinical trials, systematic reviews, meta-analyses, and reviews containing "Bell's palsy" were identified in MEDLINE, Embase, and CENTRAL databases from inception until April 2, 2024. Articles were indexed by their definition of BP: (1) acute (≤72-h onset) idiopathic facial paralysis with no known etiologies, (2) acute "idiopathic" facial paralysis despite providing etiologies, (3) acute non-idiopathic facial paralysis, (4) non-acute but "idiopathic" facial paralysis with or without providing etiologies, (5) synonymous with facial paralysis, and (6) no definition. Ascribed etiologies were recorded. Original articles' exclusion criteria for patients with BP were also categorized. Findings: Of 4,395 articles obtained, 924 met the criteria. Based on the aforementioned categories, incongruent definitions of BP exist in the literature: (1) 12.88%, (2) 14.72%, (3) 4.00% (4) 15.26%, (5) 8.12%, and (6) 45.02%. Ascribed etiologies and exclusion criteria for BP also varied. Conclusions and Relevance: BP is a common condition seen by providers of many specialties, and as such patient diagnosis, management, and research rely on consistency and a shared language. This systematic review found inconsistent definitions and applications of the term "Bell's palsy" in the literature.

Background: Various large language models (LLMs) can provide human-level medical discussions, but they have not been compared regarding rhinoplasty knowledge. Objective: To compare the leading LLMs in answering complex rhinoplasty consultation questions as evaluated by plastic surgeons. Methods: Ten open-ended rhinoplasty consultation questions were presented to ChatGPT-4o, Google Gemini, Claude, and Meta-AI LLMs. The responses were randomized and ranked by seven rhinoplasty-specializing plastic surgeons (1 = worst, 4 = best) considering their quality. Textual readability was analyzed via Flesch Reading Ease (FRE) and Flesch-Kincaid Grade (FKG). Results: Claude provided the top answers for seven questions while ChatGPT provided the top answers for three questions. In overall collective scoring, Claude provided the best answers with 224 points, followed by ChatGPT's 200, Meta's 138, and Gemini's 138 scores. Claude (mean score/question 3.20 ± 1.00) significantly outperformed all the other models (p < 0.05), while ChatGPT (mean score/question 2.86 ± 0.94) outperformed Meta and Gemini. Meta and Gemini performed similarly. Meta had a significantly lower FKG than Claude and ChatGPT and a significantly lower FRE than ChatGPT. Conclusion: According to ratings by seven rhinoplasty-specializing surgeons, Claude provided the best answers for a set of complex rhinoplasty consultation questions, followed by ChatGPT. Future studies are warranted to continue comparing these models as they evolve.