Performance and Consistency of ChatGPT-4 Versus Otolaryngologists: A Clinical Case Series.

OBJECTIVE: To study the performance of Chatbot Generative Pretrained Transformer-4 (ChatGPT-4) in the management of cases in otolaryngology-head and neck surgery. STUDY DESIGN: Prospective case series. SETTING: Multicenter University Hospitals. METHODS: History, clinical, physical, and additional examinations of adult outpatients consulting in otolaryngology departments of CHU Saint-Pierre and Dour Medical Center were presented to ChatGPT-4, which was interrogated for differential diagnoses, management, and treatment(s). According to specialty, the ChatGPT-4 responses were assessed by 2 distinct, blinded board-certified otolaryngologists with the Artificial Intelligence Performance Instrument. RESULTS: One hundred cases were presented to ChatGPT-4. ChaGPT-4 indicated a mean of 3.34 (95% confidence interval [CI]: 3.09, 3.59) additional examinations per patient versus 2.10 (95% CI: 1.76, 2.34; P = .001) for the practitioners. There was strong consistency (k > 0.600) between otolaryngologists and ChatGPT-4 for the indication of upper aerodigestive tract endoscopy, positron emission tomography and computed tomography, audiometry, tympanometry, and psychophysical evaluations. Primary diagnosis was correctly performed by ChatGPT-4 in 38% to 86% of cases depending on subspecialty. Additional examinations indicated by ChatGPT-4 were pertinent and necessary in 8% to 31% of cases, while the treatment regimen was pertinent in 12% to 44% of cases. The performance of ChatGPT-4 was not influenced by the human-reported level of difficulty of clinical cases. CONCLUSION: ChatGPT-4 may be a promising adjunctive tool in otolaryngology, providing extensive documentation about additional examinations, primary and differential diagnoses, and treatments. The ChatGPT-4 is more effective in providing a primary diagnosis, and less effective in the selection of additional examinations and treatments.

Accuracy of ChatGPT-3.5 and -4 in providing scientific references in otolaryngology-head and neck surgery.

INTRODUCTION: Chatbot generative pre-trained transformer (ChatGPT) is a new artificial intelligence-powered language model of chatbot able to help otolaryngologists in practice and research. We investigated the accuracy of ChatGPT-3.5 and -4 in the referencing of manuscripts published in otolaryngology. METHODS: ChatGPT-3.5 and ChatGPT-4 were interrogated for providing references of the top-30 most cited papers in otolaryngology in the past 40 years including clinical guidelines and key studies that changed the practice. The responses were regenerated three times to assess the accuracy and stability of ChatGPT. ChatGPT-3.5 and ChatGPT-4 were compared for accuracy of reference and potential mistakes. RESULTS: The accuracy of ChatGPT-3.5 and ChatGPT-4.0 ranged from 47% to 60%, and 73% to 87%, respectively (p < 0.005). ChatGPT-3.5 provided 19 inaccurate references and invented 2 references throughout the regenerated questions. ChatGPT-4.0 provided 13 inaccurate references, while it proposed only one invented reference. The stability of responses throughout regenerated answers was mild (k = 0.238) and moderate (k = 0.408) for ChatGPT-3.5 and 4.0, respectively. CONCLUSIONS: ChatGPT-4.0 reported higher accuracy than the free-access version (3.5). False references were detected in both 3.5 and 4.0 versions. Practitioners need to be careful regarding the use of ChatGPT in the reach of some key reference when writing a report.

Exploring the potential of Chat-GPT as a supportive tool for sialendoscopy clinical decision making and patient information support.

INTRODUCTION: Sialendoscopy has emerged in the last decades as a groundbreaking technique, offering a minimally invasive approach for exploring and managing salivary gland disorders. More recently, the advent of chatbots, powered by advanced natural processing language and artificial intelligence algorithms, has revolutionized the way healthcare professionals and patients access and analyze medical information and potentially will support soon the clinical decision-making process. MATERIALS AND METHODS: A prospective, cross-sectional study was designed to assess the level of agreement between Chat-GPT and 10 expert sialendoscopists aiming the capabilities of Chat-GPT to further improve the management of salivary gland disorders. RESULTS: The mean level of agreement was 3.4 (SD: 0.69; Min: 2, Max: 4) for Chat-GPT's answers while it was 4.1 (SD: 0.56; Min: 3, Max: 5) for the group of EESS (p < 0.015). The overall Wilcoxon signed-rank test yielded a significance level of p < 0.026 when comparing the level of agreement between Chat-GPT and EESS. The mean number of therapeutic alternatives suggested by Chat-GPT was 3.33 (SD: 1.2; Min: 2, Max: 5), while it was 2.6 (SD: 0.51; Min: 2, Max: 3) for the group of EESS; p = 0.286 (95% CI - 0.385 to 1.320). CONCLUSION: Chat-GPT represents a promising tool in the clinical decision-making process within the salivary gland clinic, particularly for patients who are candidates for sialendoscopy treatment. Additionally, it serves as a valuable source of information for patients. However, further development is necessary to enhance the reliability of these tools and ensure their safety and optimal use in the clinical setting.

Effectiveness of Platelet-Rich Plasma for COVID-19-Related Olfactory Dysfunction: A Controlled Study.

OBJECTIVE: To investigate the effectiveness of platelet-rich plasma (PRP) injection into the olfactory clefts of coronavirus disease 2019 (COVID-19) patients with persistent olfactory dysfunction (OD). STUDY DESIGN: Controlled study. SETTING: Multicenter study. METHODS: From March 2022 to November 2022, COVID-19 patients with persistent OD were recruited from three European hospitals to undergo PRP injections into the olfactory clefts. Olfactory function was evaluated at baseline and 10 weeks postinjection with the Olfactory Disorder Questionnaire (ODQ) and threshold, discrimination, and identification (TDI) test. Data were compared with a control group of untreated patients. RESULTS: Eighty-one patients who underwent PRP injection and 78 controls were included. Sixty-five PRP patients (80.3%) experienced subjective smell improvement after a mean duration of 3.4 ± 1.9 weeks. The parosmia, life quality statement, and ODQ sub- and total scores significantly decreased from pre- to 10-week postinjection in the PRP group. The TDI sub- and total scores significantly increased 10 weeks postinjection. In controls, the ODQ score did not change over time, while the discrimination, identification, and total TDI scores significantly increase after 10 weeks of follow-up. The 10-week TDI and ODQ scores were significantly better in the PRP group compared with the controls. CONCLUSION: Patients who underwent PRP injection reported better 10-week subjective and objective smell outcomes than controls. Future randomized-controlled studies using saline injection into the olfactory cleft of controls are needed to determine the superiority of PRP over placebo.

Validity and reliability of an instrument evaluating the performance of intelligent chatbot: the Artificial Intelligence Performance Instrument (AIPI).

OBJECTIVES: To evaluate the reliability and validity of the Artificial Intelligence Performance Instrument (AIPI). METHODS: Medical records of patients consulting in otolaryngology were evaluated by physicians and ChatGPT for differential diagnosis, management, and treatment. The ChatGPT performance was rated twice using AIPI within a 7-day period to assess test-retest reliability. Internal consistency was evaluated using Cronbach's α. Internal validity was evaluated by comparing the AIPI scores of the clinical cases rated by ChatGPT and 2 blinded practitioners. Convergent validity was measured by comparing the AIPI score with a modified version of the Ottawa Clinical Assessment Tool (OCAT). Interrater reliability was assessed using Kendall's tau. RESULTS: Forty-five patients completed the evaluations (28 females). The AIPI Cronbach's alpha analysis suggested an adequate internal consistency (α = 0.754). The test-retest reliability was moderate-to-strong for items and the total score of AIPI (rs = 0.486, p = 0.001). The mean AIPI score of the senior otolaryngologist was significantly higher compared to the score of ChatGPT, supporting adequate internal validity (p = 0.001). Convergent validity reported a moderate and significant correlation between AIPI and modified OCAT (rs = 0.319; p = 0.044). The interrater reliability reported significant positive concordance between both otolaryngologists for the patient feature, diagnostic, additional examination, and treatment subscores as well as for the AIPI total score. CONCLUSIONS: AIPI is a valid and reliable instrument in assessing the performance of ChatGPT in ear, nose and throat conditions. Future studies are needed to investigate the usefulness of AIPI in medicine and surgery, and to evaluate the psychometric properties in these fields.

Usefulness of pepsin saliva measurement for the detection of primary burning mouth syndrome related to reflux.

OBJECTIVES: To study the diagnostic value of salivary pepsin tests for detecting laryngopharyngeal reflux (LPR) in patients with primary burning mouth syndrome (BMS). METHODS: Patients with BMS and asymptomatic individuals were consecutively recruited from September 2018 to June 2023. Patients underwent hypopharyngeal-esophageal impedance pH-monitoring (HEMII-pH) and saliva collections to measure pepsin. Stomatology evaluation was carried out to exclude other causes of BMS. Oral, pharyngeal and laryngeal signs and symptoms were evaluated with Reflux Sign Assessment (RSA) and Reflux Symptom Score (RSS). Sensitivity, specificity, positive (PPV) and negative (NPV) predictive values of pepsin test were calculated considering the highest values of pepsin tests at ≥ 16, ≥ 36, and ≥ 100 ng/mL cutoffs. Receiver operating characteristic curve (ROC) was evaluated. RESULTS: Forty-nine patients with both BMS and LPR at the HEMII-pH and 21 asymptomatic individuals were recruited. Pepsin test was 83.7%, 79.6%, and 71.4% sensitive at cutoffs ≥ 16, ≥ 36, and ≥ 100 ng/mL, respectively. The ROC analysis reported that a threshold of ≥ 21.5 ng/mL was associated with sensitivity, specificity, PPV and NPV of 81.6%, 81.0%, 90.1% and 65.4%, respectively. The severity score of burning mouth symptom was significantly associated with the saliva pepsin concentration (rs = 0.263; p = 0.029) and the oral RSA (rs = 0.474; p = 0.007). CONCLUSION: Pepsin test is a valuable diagnostic approach for detecting LPR in patients with BMS. Patients with high level of saliva pepsin reported more severe burning mouth symptoms. Future studies are needed to confirm the role of LPR in the primary BMS.

Laryngeal verrucous cell carcinoma.

PURPOSE OF REVIEW: To summarize the recent literature on epidemiology, clinical findings, treatment, and survival of laryngeal verrucous cell carcinoma (LVC). RECENT FINDINGS: Epidemiological studies report that LVC accounts for 1-3% of all laryngeal cancers. The incidence is decreasing, while most patients are male individuals and smokers. LVC are commonly detected in early stages because they are more frequently located in the glottic region. Tobacco, alcohol overuse, and, possibly, human papilloma virus are the main contributing factors. Recent studies confirm that surgery is the primary therapeutic approach with better prognosis when compared with other treatment modalities. Surgery alone is associated with 86.8% disease-free and 80.3% overall survival rates, while metastases are anecdotal. SUMMARY: LVC presents different clinical, pathological, and survival outcomes when compared with the classic laryngeal squamous cell carcinoma. Biopsies need often to be repeated before getting the most appropriate diagnosis; this supports the need of large-sample biopsy during the tumor diagnosis and staging. The glottic location of most LVC leads to detection of this lesion in its early stages, with ensuing better survival and outcomes after surgery compared with the classic form of squamous cell carcinoma. Future studies are needed to understand the biology of LVC and its related better prognostic outcomes when compared to other laryngeal malignancies.

Indications, Efficacy, Safety, and Clinical Outcomes of 585 nm Pulsed Dye Laser in Non-Malignant Laryngeal Lesions: A Systematic Review.

The objective of this manuscript was to review the indications, efficacy, and safety of a 585 nm pulsed dye laser (PDL) in non-malignant laryngeal lesions. Following the PRISMA statement recommendations, three independent authors searched for articles published in PubMed/MEDLINE, the Cochrane Library, Google Scholar, Scielo, and Web of Science. A bias analysis was performed following NICE guidance tools. From the 506 identified publications, 19 observational studies met the inclusion criteria. The PDL improves vocal quality objectively and subjectively in vascular lesions (p < 0.005) and improves vocal quality in patients with dysplasia/leukoplasia without changing the natural history of the disease compared to other treatments. Reinke's edema and granulomas require an average of 1.5 PDL sessions for resolution. Treatment of recurrent respiratory papillomatosis requires multiple sessions, with complete remission achieved in 50-70% of patients. Regardless of the lesion, the tolerance of the procedure under local anesthesia is exceptional (84-97%), and the results in terms of regression and vocal quality are promising. The complication rate is minimal, and the procedure does not interfere with other treatment alternatives. There is no consensus on laser settings. The lack of consistent use in evaluating vocal outcomes, whether objective or subjective, prevents the comparability between studies. The 585 nm pulsed dye laser appears to be an effective and safe therapeutic option in patients with non-malignant laryngeal pathology. Future controlled studies are needed to compare the 585 nm pulsed dye laser with other lasers or cold instrument procedures.

Is ChatGPT-4 Accurate in Proofread a Manuscript in Otolaryngology-Head and Neck Surgery?

ChatGPT is a new artificial intelligence-powered language model of chatbot able to help otolaryngologists in clinical practice and research. We investigated the ability of ChatGPT-4 in the editing of a manuscript in otolaryngology. Four papers were written by a nonnative English otolaryngologist and edited by a professional editing service. ChatGPT-4 was used to detect and correct errors in manuscripts. From the 171 errors in the manuscripts, ChatGPT-4 detected 86 errors (50.3%) including vocabulary (N = 36), determiner (N = 27), preposition (N = 24), capitalization (N = 20), and number (N = 11). ChatGPT-4 proposed appropriate corrections for 72 (83.7%) errors, while some errors were poorly detected (eg, capitalization [5%] and vocabulary [44.4%] errors. ChatGPT-4 claimed to change something that was already there in 82 cases. ChatGPT demonstrated usefulness in identifying some types of errors but not all. Nonnative English researchers should be aware of the current limits of ChatGPT-4 in the proofreading of manuscripts.

Comparison of prevalence and evolution of COVID-19 olfactory disorders in patients infected by D614 (wild) and B.1.1.7. Alpha variant: a brief report.

OBJECTIVES: To investigate the prevalence and the evolution of olfactory disorders (OD) related to coronavirus disease 2019 (COVID-19) in patients infected during the first and the second European waves. METHODS: From March 2020 to October 2020, COVID-19 patients with OD were recruited and followed over the 12-month post-infection. The following data were collected: demographic, treatments, vaccination status, and olfactory function. Olfaction was assessed with the Olfactory Disorder Questionnaire (ODQ), and threshold, discrimination, and identification (TDI) test. Outcomes were compared between patients of the first wave (group 1: wild/D614G virus) and the second wave (group 2: B.1.1.7. Alpha variant) at 1-, 3- and 12-month post-infection. RESULTS: Sixty patients completed the evaluations accounting for 33 and 27 patients in group 1 and 2, respectively. The 1-month TDI score (23.7 ± 5.3) was significantly lower in group 2 compared to group 1 (29.8 ± 8.7; p = 0.017). Proportion of normosmic patients at 1-month post-infection was significantly higher in group 1 compared to group 2 (p = 0.009). TDI scores only significantly increased from 1- to 3-month post-infection in anosmic and hyposmic patients. Focusing on There was a negative association between the 1-month ODQ and the 1-month TDI (rs = - 0.493; p = 0.012). ODQ was a significant predictor of TDI scores at 3- and 12-month post-infection. The 12-month prevalence of parosmia was 60.6% in group 1 and 42.4% in group 2, respectively. There was no significant influence of oral corticosteroid treatment, adherence to an olfactory training and vaccination status on the olfactory outcomes. CONCLUSIONS: Patients of the second wave (Alpha B.1.1.7. variant) reported significant higher proportion of psychophysical test abnormalities at 1-month post-infection than patients infected during the first wave (D614G virus).

Diamine Oxidase Activity Deficit and Idiopathic Rhinitis: A New Subgroup of Non-Allergic Rhinitis?

Idiopathic rhinitis represents more than 50% of non-allergic rhinitis, a heterogeneous group that involves the symptomatic inflammation of the nasal mucosa. The TRPV1 receptor of unmyelinated C-type neurons appears to be involved in its pathophysiology. Histamine, whose main catabolic enzyme is DAO, is one of the mediators that can activate this receptor. The failure of DAO causes an increase in the level of histamine in the body and, consequently, the activation of TRPV1. The objective was to investigate the existence of a DAO enzyme activity deficit in idiopathic rhinitis and its correlation with symptoms. A cross-sectional study was conducted in 116 idiopathic rhinitis patients, and DAO activity, nasal peak inspiratory flow, and rhinitis severity were recorded. The prevalence of a DAO activity deficit was 41.38% (95%CI 0.33−0.50; p = 0.05). The DAO activity in patients with mild rhinitis was 52.93 ± 8.72 HDU/mL, in those with moderate rhinitis it was 120.33 ± 71.63 HDU/mL, and in those with severe rhinitis it was 92.58 ± 27.75 HDU/mL (p = 0.006). The NPIF in patients with a DAO activity deficit was 107.92 ± 34.05 L/min, compared to 72.35 ± 27.16 L/min in patients with normal enzymatic activity (p < 0.001), demonstrating a linear correlation between activity levels and nasal obstruction (−0.45; p < 0.001). Therefore, patients with a DAO deficiency and idiopathic rhinitis could present a milder disease course, because the repeated and continuous activation of TRPV1 led to a partial or total decrease in their response (desensitization). This new theory represents a different perspective for the study of idiopathic rhinitis and its relationship with TRPV1, with the regulation or modulation of the desensitization of TRPV1 being an important therapeutic target for patients with idiopathic rhinitis in the future.