The utility of ChatGPT as a generative medical translator.

PURPOSE: Large language models continue to dramatically change the medical landscape. We aimed to explore the utility of ChatGPT in providing accurate, actionable, and understandable generative medical translations in English, Spanish, and Mandarin pertaining to Otolaryngology. METHODS: Responses of GPT-4 to commonly asked patient questions listed on official otolaryngology clinical practice guidelines (CPG) were evaluated with the Patient Education materials Assessment Tool-printable (PEMAT-P.) Additional critical elements were identified a priori to evaluate ChatGPT's accuracy and thoroughness in its responses. Multiple fluent speakers of English, Mandarin, and Spanish evaluated each response generated by ChatGPT. RESULTS: Total PEMAT-P scores differed between English, Mandarin, and Spanish GPT-4 generated responses depicting a moderate effect size of language, Eta-Square 0.07 with scores ranging from 73 to 77 (P-value = 0.03). Overall understandability scores did not differ between English, Mandarin, and Spanish depicting a small effect size of language, Eta-Square 0.02 scores ranging from 76 to 79 (P-value = 0.17), nor did overall actionability scores Eta-Square 0 score ranging 66-73 (P-value = 0.44). Overall a priori procedure-specific responses similarly did not differ between English, Spanish, and Mandarin Eta-Square 0.02 scores ranging 61-78 (P-value = 0.22). CONCLUSION: GPT-4 produces accurate, understandable, and actionable outputs in English, Spanish, and Mandarin. Responses generated by GPT-4 in Spanish and Mandarin are comparable to English counterparts indicating a novel use for these models within Otolaryngology, and implications for bridging healthcare access and literacy gaps. LEVEL OF EVIDENCE: IV.

Artificial Intelligence in Rhinology.

Rhinology, allergy, and skull base surgery are fields primed for the integration and implementation of artificial intelligence (AI). The heterogeneity of the disease processes within these fields highlights the opportunity for AI to augment clinical care and promote personalized medicine. Numerous research studies have been published demonstrating the development and clinical potential of AI models within the field. Most describe in silico evaluation models without direct clinical implementation. The major themes of existing studies include diagnostic or clinical decisions support, clustering patients into specific phenotypes or endotypes, predicting post-treatment outcomes, and surgical planning.

The ophthalmologic crescent blade in endonasal surgery.

KEY POINTS: The angled tip and small size of the crescent blade provide versatility for its use in a variety of endonasal procedures. The crescent blade enables cutting along 180° from the tip, ensuring a tangential cut through the mucosa, which is important for flap viability. The disposable nature of the blade ensures that it is always sharp, allowing for its use in mucosal and cartilaginous cuts.

Redefining "Value" in Surgery: Development of a Comprehensive Value Score for Outpatient Endocrine Surgery.

OBJECTIVE: The value-based healthcare model aims to improve the quality of care and lower health care costs. The standard value equation (ie, Value = Quality/Cost), while conceptually useful, is grossly oversimplified and lacks clinical relevance. This study introduces a more detailed value equation that generates disease-specific value scores and incorporates real-world clinical and cost data to demonstrate its use. STUDY DESIGN: Prospective observational study. SETTING: Tertiary institution. METHODS: A comprehensive new health care value equation was developed that includes 23 unique inputs. Sixteen inputs represent quality (numerator) and 7 inputs represent cost (denominator). Patients undergoing thyroid or parathyroid surgery were enrolled, and data were entered into the new equation to generate surgery-specific value scores for each patient. A subanalysis was performed for telehealth visits. RESULTS: Ten patients were enrolled (60% female) with an average age of 62 years. The average total monetary cost per patient was $41,884 ($27,885 direct). Across all patients, the average total quality score was 0.99, and the cost score was 6.1, resulting in a final value score of 0.19. A subanalysis showed that changing a postoperative visit from in-person to telehealth would increase the value score by 0.66%. CONCLUSION: This analysis creates a comprehensive value equation for surgical services that incorporates the complexity of modern surgical care. The new equation includes objective and subjective outcomes and health equity, quantitatively compares the value of different surgical interventions and health care services, illustrates how specific interventions can lead to the higher value of care, and can serve as the framework for future value equations.

Mind + Machine: ChatGPT as a Basic Clinical Decisions Support Tool.

Background Generative artificial intelligence (AI) has integrated into various industries as it has demonstrated enormous potential in automating elaborate processes and enhancing complex decision-making. The ability of these chatbots to critically triage, diagnose, and manage complex medical conditions, remains unknown and requires further research. Objective This cross-sectional study sought to quantitatively analyze the appropriateness of ChatGPT (OpenAI, San Francisco, CA, US) in its ability to triage, synthesize differential diagnoses, and generate treatment plans for nine diverse but common clinical scenarios.  Methods Various common clinical scenarios were developed. Each was input into ChatGPT, and the chatbot was asked to develop diagnostic and treatment plans. Five practicing physicians independently scored ChatGPT's responses to the clinical scenarios.  Results  The average overall score for the triage ranking was 4.2 (SD 0.7). The lowest overall score was for the completeness of the differential diagnosis at 4.1 (0.5). The highest overall scores were seen with the accuracy of the differential diagnosis, initial treatment plan, and overall usefulness of the response (all with an average score of 4.4). Variance among physician scores ranged from 0.24 for accuracy of the differential diagnosis to 0.49 for appropriateness of triage ranking.  Discussion ChatGPT has the potential to augment clinical decision-making. More extensive research, however, is needed to ensure accuracy and appropriate recommendations are provided.

Head-to-Head Comparison of ChatGPT Versus Google Search for Medical Knowledge Acquisition.

OBJECTIVE: Chat Generative Pretrained Transformer (ChatGPT) is the newest iteration of OpenAI's generative artificial intelligence (AI) with the potential to influence many facets of life, including health care. This study sought to assess ChatGPT's capabilities as a source of medical knowledge, using Google Search as a comparison. STUDY DESIGN: Cross-sectional analysis. SETTING: Online using ChatGPT, Google Seach, and Clinical Practice Guidelines (CPG). METHODS: CPG Plain Language Summaries for 6 conditions were obtained. Questions relevant to specific conditions were developed and input into ChatGPT and Google Search. All questions were written from the patient perspective and sought (1) general medical knowledge or (2) medical recommendations, with varying levels of acuity (urgent or emergent vs routine clinical scenarios). Two blinded reviewers scored all passages and compared results from ChatGPT and Google Search, using the Patient Education Material Assessment Tool (PEMAT-P) as the primary outcome. Additional customized questions were developed that assessed the medical content of the passages. RESULTS: The overall average PEMAT-P score for medical advice was 68.2% (standard deviation [SD]: 4.4) for ChatGPT and 89.4% (SD: 5.9) for Google Search (p < .001). There was a statistically significant difference in the PEMAT-P score by source (p < .001) but not by urgency of the clinical situation (p = .613). ChatGPT scored significantly higher than Google Search (87% vs 78%, p = .012) for patient education questions. CONCLUSION: ChatGPT fared better than Google Search when offering general medical knowledge, but it scored worse when providing medical recommendations. Health care providers should strive to understand the potential benefits and ramifications of generative AI to guide patients appropriately.

Time-Driven Activity-Based Cost Comparison of Thyroid Lobectomy and Radiofrequency Ablation.

OBJECTIVE: Radiofrequency ablation (RFA) of benign thyroid nodules has gained traction for its therapeutic effectiveness, thyroid function preservation, and minimally invasive nature. While a growing body of evidence reports positive outcomes from thyroid RFA, financial comparisons between both procedures remain limited. This analysis aims to more accurately measure the direct cost of thyroid RFA in comparison to thyroid lobectomy. STUDY DESIGN: Bottom-up financial cost analysis. SETTING: Tertiary endocrine head and neck surgery center. METHODS: Time-driven activity-based costing was utilized to obtain unit-based cost estimates. The care cycles for thyroid lobectomy and RFA were defined, and process maps were developed comprising all personnel and work in the care cycle. Time estimates were calculated for all personnel involved, and public government data were used to obtain capacity cost rates for each component of the care cycle. Consumable supply and overhead costs were obtained for both procedures, and overall costs were compared. RESULTS: For thyroid lobectomy, total personnel costs were $1087.97, consumable supplies were $942.68, and overhead costs $17,199.10. For thyroid nodule RFA performed in an office setting, the total personnel cost calculated was $379.90, consumable supplies $1315.28, and overhead $7031.20. Overall, the total cost for thyroid lobectomy was $19,229.75 compared to $8726.38 for RFA. CONCLUSION: In-office thyroid nodule RFA is associated with lower direct costs than thyroid lobectomy, and overhead is the greatest cost driver for both procedures. If clinical and patient-centered outcomes are comparable, then RFA may provide higher value for appropriately selected patients.

Comparison Between ChatGPT and Google Search as Sources of Postoperative Patient Instructions.

This qualitative study rates the level of understandability, actionability, and procedure-specific content in postoperative instructions generated from ChatGPT, Google Search, and Stanford University.

Assessment of Opioid Use and Analgesic Requirements After Endoscopic Sinus Surgery: A Randomized Clinical Trial.

Importance: The opioid epidemic has generated interest in optimizing opioid prescribing after common surgeries. Recent studies have shown a broad range of analgesic prescription patterns following endoscopic sinus surgery (ESS). Objective: To compare the efficacy of different analgesic regimens after ESS. Design, Setting, and Participants: This multi-institutional, nonblinded randomized clinical trial was conducted at 6 tertiary centers across the US and Canada and included participants who underwent ESS for acute or chronic rhinosinusitis. The study was conducted from March 2019 to March 2020, and the data were analyzed in November to December 2020. Interventions: All participants received acetaminophen, 650 mg, as the first-line analgesic. From there, patients were randomized to either oxycodone rescue (oxycodone, 5 mg, as second-line therapy) or ibuprofen rescue (ibuprofen, 600 mg, as second-line therapy, with oxycodone, 5 mg, reserved for breakthrough pain). Main Outcomes and Measures: Baseline characteristics and disease severity were collected at enrollment. Medication logs, pain scores, and epistaxis measures were collected until postoperative day 7. The primary outcome was the postoperative visual analog scale score for pain. Brief Pain Inventory Pain Severity and Pain Interference Scores were also collected. Results: A total of 118 patients were randomized (62 [52.5%] oxycodone rescue, 56 [47.5%] ibuprofen rescue; mean [SD] age, 46.7 [16.3] years; 44 women [44.0%]; 83 White [83.0%], 7 Black [7.0%], and 7 Asian individuals [7.0%]). After exclusions for loss to follow-up and noncompliance, 51 remained in the oxycodone rescue group and 49 in the ibuprofen rescue group. The groups had similar demographic characteristics and disease severity. Thirty-two (63%) in the oxycodone rescue group had adequate pain management with acetaminophen only, while 19 (37%) consumed at least 1 oxycodone dose. In the ibuprofen rescue group, 18 (16%) required only acetaminophen, 28 (57%) used only acetaminophen and ibuprofen, and the remaining 13 (26%) consumed 1 or more oxycodone doses. The groups had similar average acetaminophen (9.69 vs 7.96 doses; difference, 1.73; 95% CI, -1.37 to 4.83) and oxycodone (1.89 vs 0.77 doses; difference, 1.13; 95% CI, -0.11 to 2.36) use. Both groups had similar postoperative visual analog scale scores. A subanalysis that compared opioids users with nonusers showed clinically significant lower pain scores in nonusers at multiple postoperative points. Conclusions and Relevance: In this randomized clinical trial, most patients who underwent ESS could be treated postoperatively using a nonopioid regimen of either acetaminophen alone or acetaminophen and ibuprofen. Ibuprofen as a second-line therapy did not reduce overall narcotic consumption, but the overall narcotic use was low in both groups. Trial Registration: ClinicalTrials.gov Identifier: NCT03783702.

Oral Corticosteroids Following Endoscopic Sinus Surgery for Chronic Rhinosinusitis Without Nasal Polyposis: A Randomized Clinical Trial.

Importance: Although oral corticosteroids are commonly prescribed following endoscopic sinus surgery (ESS) for chronic rhinosinusitis (CRS) without nasal polyposis, there are little data to suggest that this is a beneficial practice. Objective: To assess the efficacy of oral corticosteroids following ESS in CRS without polyps. Design, Setting, and Participants: This prospective double-blinded, placebo-controlled, randomized noninferiority clinical trial conducted in a single academic tertiary rhinology practice included adults with CRS without polyps undergoing ESS. Of 81 patients recruited, 72 completed the study. Interventions: Patients were randomized into 2 treatment groups: a 12-day postoperative taper of oral prednisone vs matched placebo tablets. All study patients also received a uniform 2-week postoperative regimen of oral antibiotics, fluticasone nasal spray, and saline rinses. Main Outcomes and Measures: The primary outcome measures were Sinonasal Outcome Test-22 (SNOT-22) scores and Lund-Kennedy endoscopy scores, collected preoperatively and postoperatively at 1 week, 1 month, 3 months, and 6 months. Scores were compared between treatment groups at each time point using longitudinal difference between treatment groups and analyzed using 2-way, repeated measures analysis of variance. Secondary outcome measures included treatment-related adverse effects. Results: Overall, 72 patients (mean [SD] age, 49.4 [14.9] years; 36 men, 36 women) completed the study, with 33 in the prednisone arm and 39 in the placebo arm. When comparing longitudinal differences between treatment groups, there was no clinically meaningful difference observed in SNOT-22 total (F[4254] = 1.71, η2 = 0.01 [95% CI, 0.00-0.05]) or Lund-Kennedy scores (F[4247] = 1.23, η2 = 0.02 [95% CI, 0.00-0.50]). In SNOT-22 subdomain analyses, there was no clinically meaningful difference between treatment groups for rhinologic, extranasal rhinologic, ear/facial, or sleep subdomains. However, the prednisone group had worse longitudinal scores for psychological dysfunction compared with the placebo group (F[4254] = 3.18, η2 = 0.05 [95% CI, 0.02-0.09]). Reported adverse effects were similar between the 2 treatment groups. Conclusions and Relevance: In this randomized clinical trial of patients with CRS without polyps, oral prednisone following ESS conferred no additional benefit over placebo in terms of SNOT-22 total scores, SNOT-22 rhinologic subscores, or Lund-Kennedy endoscopy scores up to 6 months after surgery. Patients receiving prednisone, however, did demonstrate worse SNOT-22 psychologic subdomain scores. These results suggest that the risks of oral corticosteroids may outweigh the benefits; thus use of oral corticosteroids after ESS for CRS without polyps should be carefully considered. Trial Registration: ClinicalTrials.gov Identifier: NCT02748070.

Facial Paralysis and Communicative Participation: The Importance of Facial Symmetry at Rest.

OBJECTIVE: There is a paucity of research devoted to understanding the communication restrictions encountered by facial paralysis patients. We aim to explore the relationship between patient-reported restrictions in communicative participation and objective facial paralysis severity using validated scales of facial movement. METHODS: We performed a pilot retrospective study using a consecutive series of adult patients with a diagnosis of unilateral facial paralysis. In addition to baseline demographics, subjects were evaluated using the Communicative Item Participation Bank Short Form (CPIB), Electronic Facial Assessment by Computer Evaluation (eFACE), and Sunnybrook Facial Grading System (SFGS). RESULTS: Twenty patients were included, 10 (50%) of whom were female with a mean age of 61 ± 13 years and mean duration of facial paralysis of 53 ± 82 months. The mean CPIB score was 14.6 ± 10.0 (range 0-29) and was comparable to scores of patients with conditions known to cause significant communicative disability. The mean eFACE scores were 67.4 ± 29.2, 44.2 ± 30.1, and 73.8 ± 30.0 for the static, dynamic, and synkinesis domains, respectively, with a composite smile score of 58.5 ± 16.9. After adjusting for age, gender, and duration of facial paralysis, significant moderate correlations were observed between the CPIB and the static eFACE domain (r = -0.51, P = .03) and smile composite score (r = 0.48, P = 0.0049), in addition to between the CPIB and SFGS synkinesis domain (r = 0.48, P = 0.04). CONCLUSIONS: Patients with unilateral facial paralysis experience significant limitations in communicative participation. These restrictions demonstrate moderate to strong correlations with objective assessments of facial paralysis and quality of life measures. Communicative participation may be a helpful means of tracking response to treatment. LEVEL OF EVIDENCE: IV.

Acute Exacerbations in Recurrent Acute Rhinosinusitis: Differences in Quality of Life and Endoscopy.

OBJECTIVES/HYPOTHESIS: Research surrounding outcome differences for patients with recurrent acute rhinosinusitis (RARS) is scarce. This investigation explored quality of life (QOL) and sinonasal attributes in patients during acute episodes (AEs) and in-between AEs of RARS. STUDY DESIGN: Retrospective outcomes research. METHODS: Data from patients with RARS were collected from two academic institutions between 2009 and 2017 using prospective and retrospective methodology. During clinical presentation, subjects were classified as with or without an AEs using guideline definitions of acute bacterial rhinosinusitis (ABRS). Between-group differences in 22-item Sino-Nasal Outcome Test (SNOT-22) survey and Lund-Kennedy (LK) endoscopy scores were assessed. RESULTS: Four hundred twenty-three clinical visits from 202 patients were included. Visits during an AE (168/423, 40%) were associated with significantly worse SNOT-22 total scores compared to between AEs (255/423, 60%; median = 53.0 [interquartile range (IQR) = 24.0] vs. 34.0 [IQR = 29.5]) and all SNOT-22 subdomain scores (all P < .001). LK scores were available for 167 visits, with 56 (34%) completed during an AE. Compared to visits without an AE, endoscopy findings associated with an AE were less frequently normal (LK score = 0, 45% vs. 62%, P = .031) with worse median LK scores (2.0 [IQR = 4.0] vs. 0.0 [IQR = 2.0], P = .005). CONCLUSIONS: AEs are associated with significantly worse QOL and mildly worse endoscopic findings. Almost half of visits during AEs had negative endoscopy, identifying a disparity between patient symptoms and objective findings and calling into question alternative or concomitant diagnoses. Diagnostic criteria for ABRS or AEs in RARS do not require objective confirmation of inflammation, presenting a conundrum for clinicians. The potential for overdiagnosis of ABRS and AEs should be considered when determining the risk/benefit ratio of treatments for RARS. LEVEL OF EVIDENCE: 2c Laryngoscope, 2019.

Appropriateness criteria predict outcomes for sinus surgery and may aid in future patient selection.

OBJECTIVES: Appropriateness criteria to determine surgical candidacy for chronic rhinosinusitis (CRS) have recently been described. This study stratified patients who underwent endoscopic sinus surgery (ESS) according to these new appropriateness criteria and evaluated postoperative improvements among appropriateness categories. METHODS: Adult patients with uncomplicated CRS electing ESS were prospectively enrolled in a multi-institutional cohort study between March 2011 and June 2015 to assess outcomes. Subsequently, appropriateness criteria that consider preoperative medical therapy, 22-item SinoNasal Outcome Test (SNOT-22) scores, and Lund-Mackay computed tomography scores were retrospectively applied. RESULTS: A total of 92.6% (436 of 471) were categorized as "appropriate" ESS candidates, 3.8% (18 of 471) as "uncertain," and 3.6% (17 of 471) as "inappropriate." Among uncertain patients, two-thirds (12 of 18) had identifiable reasons for undergoing ESS, most commonly oral corticosteroid intolerance (n = 6). Postoperative follow-up was available for 79% (n = 372). Clinically significant SNOT-22 improvements occurred in both appropriate and uncertain groups (all P < 0.050) but not among the inappropriate group. The inappropriate group reported less mean improvement in SNOT-22 total score compared to appropriate (P = 0.008) and uncertain (P = 0.006) groups. CONCLUSION: The vast majority of patients (∼93%) who underwent ESS in a multi-institutional research program were identified as appropriate candidates for surgical intervention, as defined by current appropriateness criteria. Valid considerations frequently exist for offering ESS to patients categorized as uncertain. Appropriate and uncertain candidates report similar, clinically significant SNOT-22 improvements following surgery. Patients classified as inappropriate reported significantly less improvement following ESS. Surgical appropriateness criteria may assist in predicting outcomes of ESS. LEVEL OF EVIDENCE: 2b. Laryngoscope, 2448-2454, 2018.

'Addressed to you not as a smoker but as a doctor': doctor-targeted cigarette advertisements in JAMA.

BACKGROUND AND AIM: During the mid-20th century tobacco companies placed advertisements in medical journals to entice physicians to smoke their brand and, more importantly, to recommend it to their patients. They have been little studied, in part because advertising sections in medical journals are almost universally discarded before binding. This study aimed to define the themes and techniques used in doctor-targeted tobacco advertisements that appeared in American medical journals in the mid-20th century and determine the motivations and tactics of the tobacco industry in engaging the medical profession in this way. METHODS: Doctor-targeted tobacco advertisements from JAMA and the New York State Medical Journal appearing between 1936 and 1953 were studied. These were obtained from the New York Academy of Medicine and the UCSF Truth database of tobacco industry documents. Content analysis of advertising slogans and imagery was conducted. Using internal tobacco industry documents, we examined the relationship between tobacco advertisers and medical journals. RESULTS: Among the 519 doctor-targeted advertisements, 13 brands were represented, with two (Philip Morris and Camel) accounting for 84%. Correspondence between tobacco advertisers and medical journal editors reveals the potent influence of revenue to the sponsoring society and personal compensation derived from consulting arrangements. Content analysis of the advertisements revealed much flattery of doctors and arguments professing the harmlessness of the company's brand. CONCLUSIONS: Analysis of doctor-targeted tobacco advertisements in American medical journals from 1936 to 1953 suggest that tobacco companies targeted physicians as a potential sales force to assuage the public's fear of health risks and to recruit them as allies against negative publicity. Tobacco companies also appeared to try, through the substantial advertising revenue passed by journals to their parent medical societies, to temper any possible opposition by organized medicine.

Epithelial separation theory for post-tonsillectomy secondary hemorrhage: evidence in a mouse model and potential heparin-binding epidermal growth factor-like growth factor therapy.

OBJECTIVE: To provide histological evidence to investigate a theory for post-tonsillectomy secondary hemorrhage (PTH) in a mouse model and to evaluate the potential for heparin-binding epidermal growth factor-like growth factor (HB-EGF) treatment on wound healing in this model. METHODS: A prospective randomized single-blinded cohort study. A uniform tongue wound was created in 84 mice (day 0). Mice were randomized to HB-EGF (treatment, n = 42) or saline (control, n = 42). In treatment mice, HB-EGF 5 µg/ml was administered intramuscularly into the wound daily (days 0-14). In control mice, normal saline was administered daily. Three mice from each group were sacrificed daily through day 14 and the wounds evaluated histologically by blinded reviewers. RESULTS: Key stages of wound healing, including keratinocyte proliferation and migration, wound contraction, epithelial separation, and neoangiogenesis, are defined with implications for post-tonsillectomy wound healing. Epithelial separation (59 vs. 100%, p = 0.003) and wound reopening (8 vs. 48%, p < 0.001) were reduced with HB-EGF. Epithelial thickness (220 vs. 30 µm, p = 0.04) was greater with HB-EGF. Wound closure (days 4-5 vs. day 6, p = 0.01) occurred earlier with HB-EGF. CONCLUSIONS: In healing of oral keratinocytes on muscle epithelial separation secondary to muscle, contraction occurs concurrently with neoangiogenesis in the base of the wound, increasing the risk of hemorrhage. This potentially explains why post-tonsillectomy secondary hemorrhage occurs and its timing. HB-EGF-treated wounds showed greater epithelial thickness, less frequent epithelial separation and wound reopening, and earlier wound closure prior to neovascularization, suggesting that HB-EGF may be a potential preventative therapy for PTH. LEVEL OF EVIDENCE: NA-animal studies or basic research.

Evidence for a 'preinvasive' variant of fungal sinusitis: Tissue invasion without angioinvasion.

Clinical experience has suggested the existence of an intermediate form of fungal sinusitis between the categories of non-invasive fungal sinusitis (non-IFS) and invasive fungal sinusitis (IFS). This fungal sinusitis variant demonstrates unhealthy mucosa by endoscopy with fungal invasion, but lacks angioinvasion microscopically, representing what clinically behaves as a 'pre-invasive' subtype of fungal sinusitis. Unlike non-IFS disease, patients with pre-invasive fungal sinusitis were still felt to require anti-fungal medications due to histologic presence of invasive fungus. While sharing some clinical features of IFS, these 'intermediate' patients were successfully spared extended and repeated surgical debridements given the microscopic findings, and have been successfully treated with shorter courses of antifungal therapy. These select patients have had favorable outcomes when managed in a judicious and semi-aggressive manner, in an undefined zone between the treatments for routine fungal ball and aggressive IFS.