Assessing ChatGPT's theoretical knowledge and prescriptive accuracy in bacterial infections: a comparative study with infectious diseases residents and specialists.

OBJECTIVES: Advancements in Artificial Intelligence(AI) have made platforms like ChatGPT increasingly relevant in medicine. This study assesses ChatGPT's utility in addressing bacterial infection-related questions and antibiogram-based clinical cases. METHODS: This study involved a collaborative effort involving infectious disease (ID) specialists and residents. A group of experts formulated six true/false, six open-ended questions, and six clinical cases with antibiograms for four types of infections (endocarditis, pneumonia, intra-abdominal infections, and bloodstream infection) for a total of 96 questions. The questions were submitted to four senior residents and four specialists in ID and inputted into ChatGPT-4 and a trained version of ChatGPT-4. A total of 720 responses were obtained and reviewed by a blinded panel of experts in antibiotic treatments. They evaluated the responses for accuracy and completeness, the ability to identify correct resistance mechanisms from antibiograms, and the appropriateness of antibiotics prescriptions. RESULTS: No significant difference was noted among the four groups for true/false questions, with approximately 70% correct answers. The trained ChatGPT-4 and ChatGPT-4 offered more accurate and complete answers to the open-ended questions than both the residents and specialists. Regarding the clinical case, we observed a lower accuracy from ChatGPT-4 to recognize the correct resistance mechanism. ChatGPT-4 tended not to prescribe newer antibiotics like cefiderocol or imipenem/cilastatin/relebactam, favoring less recommended options like colistin. Both trained- ChatGPT-4 and ChatGPT-4 recommended longer than necessary treatment periods (p-value = 0.022). CONCLUSIONS: This study highlights ChatGPT's capabilities and limitations in medical decision-making, specifically regarding bacterial infections and antibiogram analysis. While ChatGPT demonstrated proficiency in answering theoretical questions, it did not consistently align with expert decisions in clinical case management. Despite these limitations, the potential of ChatGPT as a supportive tool in ID education and preliminary analysis is evident. However, it should not replace expert consultation, especially in complex clinical decision-making.

Lower urinary tract symptoms in young-middle aged males with a diagnosis of obstructive sleep apnea syndrome.

BACKGROUND: The aim of the current study is to measure the prevalence and the potential role of International Prostate Symptom Score (IPSS) score as a predictor of obstructive sleep apnea syndrome (OSAS) in male experienced lower urinary tract symptoms (LUTS). METHODS: A cross-sectional web-based Italian survey was administered via Google Forms between July 17 and October 31, 2022. The urinary functioning was measured through the IPSS questionnaire. Specifically, we considered symptoms occurring more than "about half the time" (score ≥ 3) as bothering symptoms. Multivariable logistic regression models (LRMs) adjusting for age, body mass index (BMI), International Index of Erectile Function-5, IPSS, and hypertension were fitted to predict OSAS in the cohort of men responding to the survey and experiencing LUTS. RESULTS: Overall, 58 (24.4%) patients had a confirmed diagnosis of OSAS. The overall median IPSS was 5 (inter quartile range [IQR]: 3-8), respectively. According to IPSS items, 24 (10%), 44 (18.4%), 12 (5%), 12 (5%), 12 (5%), 11 (4.6%), 63 (26.4%) patients exhibit incomplete bladder emptying, urinary frequency, intermittency, urgency, weak stream, straining, nocturia with a score ≥ 3, respectively. After multivariable LRMs predicting the developing OSAS, age (odds ratio [OR]: 1.09, p < 0.001), BMI (OR:1.12, p < 0.001) and IPSS total score (OR:1.08, p = 0.02) were independent predicting factors. CONCLUSION: This analysis revealed that the IPSS total score, age, and BMI are independent predictors of OSAS in males. In this context, the use of IPSS in daily practice could be helpful in assessing the LUTS presence and in supporting physicians to identify a hidden sleep apnea condition.

Reliability of large language models in managing odontogenic sinusitis clinical scenarios: a preliminary multidisciplinary evaluation.

PURPOSE: This study aimed to evaluate the utility of large language model (LLM) artificial intelligence tools, Chat Generative Pre-Trained Transformer (ChatGPT) versions 3.5 and 4, in managing complex otolaryngological clinical scenarios, specifically for the multidisciplinary management of odontogenic sinusitis (ODS). METHODS: A prospective, structured multidisciplinary specialist evaluation was conducted using five ad hoc designed ODS-related clinical scenarios. LLM responses to these scenarios were critically reviewed by a multidisciplinary panel of eight specialist evaluators (2 ODS experts, 2 rhinologists, 2 general otolaryngologists, and 2 maxillofacial surgeons). Based on the level of disagreement from panel members, a Total Disagreement Score (TDS) was calculated for each LLM response, and TDS comparisons were made between ChatGPT3.5 and ChatGPT4, as well as between different evaluators. RESULTS: While disagreement to some degree was demonstrated in 73/80 evaluator reviews of LLMs' responses, TDSs were significantly lower for ChatGPT4 compared to ChatGPT3.5. Highest TDSs were found in the case of complicated ODS with orbital abscess, presumably due to increased case complexity with dental, rhinologic, and orbital factors affecting diagnostic and therapeutic options. There were no statistically significant differences in TDSs between evaluators' specialties, though ODS experts and maxillofacial surgeons tended to assign higher TDSs. CONCLUSIONS: LLMs like ChatGPT, especially newer versions, showed potential for complimenting evidence-based clinical decision-making, but substantial disagreement was still demonstrated between LLMs and clinical specialists across most case examples, suggesting they are not yet optimal in aiding clinical management decisions. Future studies will be important to analyze LLMs' performance as they evolve over time.

Reliability of large language models for advanced head and neck malignancies management: a comparison between ChatGPT 4 and Gemini Advanced.

PURPOSE: This study evaluates the efficacy of two advanced Large Language Models (LLMs), OpenAI's ChatGPT 4 and Google's Gemini Advanced, in providing treatment recommendations for head and neck oncology cases. The aim is to assess their utility in supporting multidisciplinary oncological evaluations and decision-making processes. METHODS: This comparative analysis examined the responses of ChatGPT 4 and Gemini Advanced to five hypothetical cases of head and neck cancer, each representing a different anatomical subsite. The responses were evaluated against the latest National Comprehensive Cancer Network (NCCN) guidelines by two blinded panels using the total disagreement score (TDS) and the artificial intelligence performance instrument (AIPI). Statistical assessments were performed using the Wilcoxon signed-rank test and the Friedman test. RESULTS: Both LLMs produced relevant treatment recommendations with ChatGPT 4 generally outperforming Gemini Advanced regarding adherence to guidelines and comprehensive treatment planning. ChatGPT 4 showed higher AIPI scores (median 3 [2-4]) compared to Gemini Advanced (median 2 [2-3]), indicating better overall performance. Notably, inconsistencies were observed in the management of induction chemotherapy and surgical decisions, such as neck dissection. CONCLUSIONS: While both LLMs demonstrated the potential to aid in the multidisciplinary management of head and neck oncology, discrepancies in certain critical areas highlight the need for further refinement. The study supports the growing role of AI in enhancing clinical decision-making but also emphasizes the necessity for continuous updates and validation against current clinical standards to integrate AI into healthcare practices fully.

Validation of the Quality Analysis of Medical Artificial Intelligence (QAMAI) tool: a new tool to assess the quality of health information provided by AI platforms.

BACKGROUND: The widespread diffusion of Artificial Intelligence (AI) platforms is revolutionizing how health-related information is disseminated, thereby highlighting the need for tools to evaluate the quality of such information. This study aimed to propose and validate the Quality Assessment of Medical Artificial Intelligence (QAMAI), a tool specifically designed to assess the quality of health information provided by AI platforms. METHODS: The QAMAI tool has been developed by a panel of experts following guidelines for the development of new questionnaires. A total of 30 responses from ChatGPT4, addressing patient queries, theoretical questions, and clinical head and neck surgery scenarios were assessed by 27 reviewers from 25 academic centers worldwide. Construct validity, internal consistency, inter-rater and test-retest reliability were assessed to validate the tool. RESULTS: The validation was conducted on the basis of 792 assessments for the 30 responses given by ChatGPT4. The results of the exploratory factor analysis revealed a unidimensional structure of the QAMAI with a single factor comprising all the items that explained 51.1% of the variance with factor loadings ranging from 0.449 to 0.856. Overall internal consistency was high (Cronbach's alpha = 0.837). The Interclass Correlation Coefficient was 0.983 (95% CI 0.973-0.991; F (29,542) = 68.3; p < 0.001), indicating excellent reliability. Test-retest reliability analysis revealed a moderate-to-strong correlation with a Pearson's coefficient of 0.876 (95% CI 0.859-0.891; p < 0.001). CONCLUSIONS: The QAMAI tool demonstrated significant reliability and validity in assessing the quality of health information provided by AI platforms. Such a tool might become particularly important/useful for physicians as patients increasingly seek medical information on AI platforms.

Recovery from olfactory and gustatory dysfunction following COVID-19 acquired during Omicron BA.1 wave in Italy.

BACKGROUND: Despite alterations in the sense of smell and taste have dominated the symptoms of SARS-CoV-2 infection, the prevalence and the severity of self-reporting COVID-19 associated olfactory and gustatory dysfunction has dropped significantly with the advent of the Omicron BA.1 subvariant. However, data on the evolution of Omicron-related chemosensory impairment are still lacking. OBJECTIVE: The aim of the present study was to estimate the prevalence and the recovery rate of self-reported chemosensory dysfunction 6-month after SARS-CoV-2 infection acquired during the predominance of the Omicron BA.1 subvariant in Italy. METHODS: Prospective observational study based on the sino-nasal outcome tool 22 (SNOT-22), item "sense of smell or taste" and additional outcomes conducted in University hospitals and tertiary referral centers in Italy. RESULTS: Of 338 patients with mild-to-moderate COVID-19 completing the baseline survey, 294 (87.0 %) responded to the 6-month follow-up interview. Among them, 101 (34.4 %) and 4 (1.4 %) reported an altered sense of smell or taste at baseline and at 6 months, respectively. Among the 101 patients with COVID-19-associated smell or taste dysfunction during the acute phase of the disease, 97 (96.0 %) reported complete resolution at 6 months. The duration of smell or taste impairment was significantly shorter in vaccinated patients (p = 0.007). CONCLUSIONS: Compared with that observed in subjects infected during the first wave of the pandemic, the recovery rate from chemosensory dysfunctions reported in the present series of patients infected during the predominance of the Omicron BA.1 subvariant was more favorable with a shorter duration being positively influenced by vaccination.

Evaluating long-term smell or taste dysfunction in mildly symptomatic COVID-19 patients: a 3-year follow-up study.

INTRODUCTION: No studies have reported data on 3-year prevalence and recovery rates of self-reported COVID-19-related olfactory and gustatory dysfunction. The aim of the present study was to estimate the 3-year prevalence and recovery rate of self-reported COVID-19-related chemosensory dysfunction in a cohort of patients with antecedent mild COVID-19. METHODS: This is a prospective observational study, measuring the prevalence of altered sense of smell or taste at follow-up and their variation from baseline, on adult patients consecutively assessed at Treviso and Trieste University Hospitals, who tested positive for SARS-CoV-2 RNA by polymerase chain reaction during March 2020. RESULTS: Overall, out of 403 respondents, 267 patients (66.3%) reported an altered sense of smell or taste (SNOT-22 > 0) at baseline, while 56 (13.9%), 29 (7.2%), and 21 (5.2%) reported such alterations at 6-24 months, 2 years, and 3 years, respectively. Among the 267 patients with COVID-19-associated smell or taste dysfunction at baseline, 246 (92.1%) reported complete resolution at 3 years. Of the patients who still experienced smell or taste dysfunction 2 years after COVID-19, 27.6% and 37.9% recovered completely and partially, respectively, at the 3-year follow-up. CONCLUSION: Among subjects with antecedent mildly symptomatic SARS-CoV-2 infection, the 3-year prevalence and recovery rate of COVID-19-related alteration in sense of smell or taste was 5% and 92%, respectively. In approximately two-thirds of patients experiencing chemosensory dysfunction still 2 years after COVID-19, it is still possible to observe a delayed complete or partial recovery after a period of 3 years, while the remaining one-third of individuals continues to have unchanged persistent chemosensory alteration.

Assessing the accuracy of ChatGPT references in head and neck and ENT disciplines.

PURPOSE: ChatGPT has gained popularity as a web application since its release in 2022. While artificial intelligence (AI) systems' potential in scientific writing is widely discussed, their reliability in reviewing literature and providing accurate references remains unexplored. This study examines the reliability of references generated by ChatGPT language models in the Head and Neck field. METHODS: Twenty clinical questions were generated across different Head and Neck disciplines, to prompt ChatGPT versions 3.5 and 4.0 to produce texts on the assigned topics. The generated references were categorized as "true," "erroneous," or "inexistent" based on congruence with existing records in scientific databases. RESULTS: ChatGPT 4.0 outperformed version 3.5 in terms of reference reliability. However, both versions displayed a tendency to provide erroneous/non-existent references. CONCLUSIONS: It is crucial to address this challenge to maintain the reliability of scientific literature. Journals and institutions should establish strategies and good-practice principles in the evolving landscape of AI-assisted scientific writing.

Self-reported smell and taste recovery in coronavirus disease 2019 patients: a one-year prospective study.

PURPOSE: The aim of the present study was to estimate the 1 year prevalence and recovery rate of self-reported chemosensory dysfunction in a series of subjects with previous mild-to-moderate symptomatic COVID-19. METHODS: Prospective study based on the SNOT-22, item "sense of smell or taste" and additional outcomes. RESULTS: 268/315 patients (85.1%) completing the survey at baseline also completed the follow-up interview. The 12 months prevalence of self-reported COVID-19 associated chemosensory dysfunction was 21.3% (95% CI 16.5-26.7%). Of the 187 patients who complained of COVID-19 associated chemosensory dysfunction at baseline, 130 (69.5%; 95% CI 62.4-76.0%) reported complete resolution of smell or taste impairment, 41 (21.9%) reported a decrease in the severity, and 16 (8.6%) reported the symptom was unchanged or worse 1 year after onset. The risk of persistence was higher for patients reporting a baseline SNOT-22 score ≥ 4 (OR = 3.32; 95% CI 1.32-8.36) as well as for those requiring ≥ 22 days for a negative swab (OR = 2.18; 95% CI 1.12-4.27). CONCLUSION: A substantial proportion of patients with previous mild-to-moderate symptomatic COVID-19 characterized by new onset of chemosensory dysfunction still complained on altered sense of smell or taste 1 year after the onset.

Correlations between IL-6 serum level and olfactory dysfunction severity in COVID-19 patients: a preliminary study.

BACKGROUND: Interleukin 6 (IL-6) is a proinflammatory cytokine that is secreted by cells infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and it is widely recognized as a negative prognostic factor. The purpose of this study was to analyze the correlations between the olfactory scores determined by psychophysical tests and the serum levels of IL-6 in patients affected by coronavirus disease 2019 (COVID-19) METHODS: Patients underwent psychophysical olfactory assessment with Connecticut Chemosensory Clinical Research Center test and IL-6 plasma level determination within 10 days of the clinical onset of COVID-19. RESULTS: Seventy-four COVID-19 patients were included in this study. COVID-19 staged as mild in 34 patients, moderate in 26 and severe in 14 cases. There were no significant differences in olfactory scores across the different COVID-19 severity groups. In the patient series, the median plasma level of IL-6 was 7.7 pg/mL (IQR 3.7-18.8). The concentration of IL-6 was found to be significantly correlated with the severity of COVID-19 with a directly proportional relationship. The correlation between IL-6 plasma concentrations and olfactory scores was weak (rs = 0.182) and not significant (p = 0.12). CONCLUSIONS: In COVID-19 patients, psychophysical olfactory scores did not show significant correlations with the plasma levels of a well-recognized negative prognostic factor such as IL-6. This observation casts some shadows on the positive prognostic value of olfactory dysfunctions.

Machine Learning Algorithms as a Computer-Assisted Decision Tool for Oral Cancer Prognosis and Management Decisions: A Systematic Review.

INTRODUCTION: Despite multiple prognostic indicators described for oral cavity squamous cell carcinoma (OCSCC), its management still continues to be a matter of debate. Machine learning is a subset of artificial intelligence that enables computers to learn from historical data, gather insights, and make predictions about new data using the model learned. Therefore, it can be a potential tool in the field of head and neck cancer. METHODS: We conducted a systematic review. RESULTS: A total of 81 manuscripts were revised, and 46 studies met the inclusion criteria. Of these, 38 were excluded for the following reasons: use of a classical statistical method (N = 16), nonspecific for OCSCC (N = 15), and not being related to OCSCC survival (N = 7). In total, 8 studies were included in the final analysis. CONCLUSIONS: ML has the potential to significantly advance research in the field of OCSCC. Advantages are related to the use and training of ML models because of their capability to continue training continuously when more data become available. Future ML research will allow us to improve and democratize the application of algorithms to improve the prediction of cancer prognosis and its management worldwide.

Trans-Oral Robotic Surgery: 14 Cases of Pleomorphic Adenoma of the Parapharyngeal Space.

PURPOSE: The aim of this study has been to describe our experience with pleomorphic adenomas of the parapharyngeal space (PPS) treated with trans-oral robotic surgery (TORS). Tumors arising from the PPS comprise less than 0.5% of all head and neck tumors. Salivary gland tumors account for 40% to 50% of PPS lesions with pleomorphic adenomas representing the most common salivary tumors (80%-90%). Parapharyngeal space tumors cause nonspecific symptoms and may be difficult to diagnose. METHODS: In our study a preoperative diagnosis was conducted by fine needle aspiration biopsy and magnetic resonance imaging and the results were used to plan the correct surgical approach. RESULTS: In all cases we were able to employ TORS, a minimally invasive procedure that allows us to operate in narrow and anatomically complex spaces that we can only reach thanks to the use of well-articulated hand pieces. CONCLUSIONS: This report indicates that TORS is a safe surgical procedure for the excision of benign tumors of the PPS in selected cases.

Six-Month Psychophysical Evaluation of Olfactory Dysfunction in Patients with COVID-19.

This study prospectively assessed the 6-month prevalence of self-reported and psychophysically measured olfactory dysfunction in subjects with mild-to-moderate COVID-19. Self-reported smell or taste impairment was prospectively evaluated by SNOT-22 at diagnosis, 4-week, 8-week, and 6-month. At 6 months from the diagnosis, psychophysical evaluation of olfactory function was also performed using the 34-item culturally adapted University of Pennsylvania Smell Identification Test (CA-UPSIT). 145 completed both the 6-month subjective and psychophysical olfactory evaluation. According to CA-UPSIT, 87 subjects (60.0%) exhibited some smell dysfunction, with 10 patients being anosmic (6.9%) and seven being severely microsmic (4.8%). At the time CA-UPSIT was administered, a weak correlation was observed between the self-reported alteration of the sense of smell or taste and olfactory test scores (Spearman's r = -0.26). Among 112 patients who self-reported normal sense of smell at last follow-up, CA-UPSIT revealed normal smell in 46 (41.1%), mild microsmia in 46 (41.1%), moderate microsmia in 11 (9.8%), severe microsmia in 3 (2.3%), and anosmia in 6 (5.4%) patients; however, of those patients self-reporting normal smell but who were found to have hypofunction on testing, 62 out of 66 had a self-reported reduction in sense of smell or taste at an earlier time point. Despite most patients report a subjectively normal sense of smell, we observed a high percentage of persistent smell dysfunction at 6 months from the diagnosis of syndrome coronavirus 2 (SARS-CoV-2) infection, with 11.7% of patients being anosmic or severely microsmic. These data highlight a significant long-term rate of smell alteration in patients with previous SARS-COV-2 infection.

Translation and validation of the short version of the Questionnaire of Olfactory Disorders-Negative Statements to Spanish.

INTRODUCTION: There are more than 400 million of native Spanish speakers around the world, being the second most spoken language in regard to the number of native speakers. For this reason, a valid questionnaire to access the olfaction of our patients is necessary. MATERIAL AND METHODS: Validation and cross-cultural adaptation of the svQOD-NS questionnaire to the Spanish language. Internal consistency of svQOD-NS measured with Cronbach α. RESULTS: 40 patients met the inclusion criteria. 41 ± 153 (range 21-82), and 32 (80%) were female. 20 patients (57,1%) were male and 15 (42,9%) were female. There was a normal distribution among patients included according to the Shapiro-Wilk test (p = 0.175). Internal consistency of svQOD-NS measured with Cronbach α was 0.861. The intraclass correlation coefficient was 0.849 (confidence interval [CI] 95%: 0.766-0.911). CONCLUSION: The Spanish Language is the second most spoken language with regard to the number of native speakers and the svQOD-NS translation represents a valid option for the Spanish-speaking medical community, from which a large number of patients can benefit.

Have There Been any Changes in the Epidemiology and Etiology of Maxillofacial Trauma During the COVID-19 Pandemic? An Italian Multicenter Study.

ABSTRACT: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has had a significant impact on people's behavior. The aim of this study has been to evaluate how the SARS-CoV-2 pandemic has impacted the incidence and the features of maxillofacial fractures presented at 6 Italian tertiary centers. Clinical records of all the patients diagnosed for facial fractures between February 23 and May 23, 2019 and 2020 were retrospectively reviewed. Any differences in patient number and characteristics and fracture etiology and site between the 2 groups were then statistically analyzed.There has been a 69.1% decrease in the number of incoming patients during the pandemic. The number of foreign patients has decreased significantly (23.3% versus 9.6%, P = 0.011) while the average age has increased (38.6 versus 45.6 years old, P = 0.01). Specific statistical significant differences for accidental falls (31.8% versus 50.1%, P = 0.005) and sports injuries (16.9% versus 1.4%, P < 0.001) were found. Concerning fracture sites, significant differences have been found in relation to nasal (22.5% versus 11.4%, P = 0.009) and frontal sinus (0.9% versus 4.4%, P = 0.037) fractures. In conclusion, SARS-CoV-2 pandemic has significantly changed the epidemiology and the etiology of facial traumas.

Reconstruction of full-thickness cheek defect with chimeric facial artery free flap: A case report.

Full-thickness cheek defects represent a difficult challenge for reconstructive surgeons. Multiple techniques have been described, including local, regional, and free flaps. In this report, the reconstruction of a through and through left cheek defect, resulting from a self-inflicted gunshot injury, with a chimeric facial artery free flap is presented. The patient underwent reconstruction with a chimeric full-thickness facial artery free flap, with 9 × 3 cm diameter skin paddle and a 6 × 5 cm myomucosal paddle, was harvested on the contralateral cheek with a 7-cm length of facial pedicle. The flap was transferred via a microsurgical technique and an end-to-end microsurgical anastomosis was performed between the donor and the contralateral facial vessels. The flap was transplanted successfully and there were no donor or recipient site complications with a satisfactory esthetic result 5 years after surgery. This technique may be a good reconstructive option for medium size, full-thickness, cheek defects allowing a "like with like" reconstruction with minimal donor-site morbidity.

The Influence of Socioeconomic Factors on the Epidemiology of Maxillofacial Fractures in Southern Italy.

Maxillofacial fractures represent a serious public health problem. Their epidemiology is extremely variable and its analysis is crucial to establish effective treatment and prevention of these injuries. The aim of this multicentric retrospective study was to analyze causes, demographics, incidence, characteristics of 987 patients diagnosed with maxillofacial trauma between 2011 and 2015 at Complex Operative Unit of Maxillofacial Surgery of Federico II University of Naples and Magna Graecia University of Catanzaro, Italy; 657 male and 310 female patients were admitted in the study. The most frequently observed fracture involved the mandible (399 patients, 35.4%), followed by zygomatic complex (337 patients, 29.9%), orbital walls (160 patients, 14.2%), and nasal bones (129 patients, 11.4%). The most frequent cause of fracture was assaults (30.4%), followed by road traffic injuries (27.2%), falls (23.2%), sport accidents (15.4%), and others causes (2.6%). Significant variations of etiology have been detected between the 2 hospitals in relationship with different migration flow trends and cultural and socioeconomic features. Epidemiological analysis of maxillofacial fractures is crucial to identify the trauma burden and to help in developing a more efficient system to plan resource allocation and to deliver care and preventive measures establishing clinical and research priorities for effective treatment and prevention of these injuries.