Are artificial intelligence large language models a reliable tool for difficult differential diagnosis? An a posteriori analysis of a peculiar case of necrotizing otitis externa.

Key Clinical Message: Large language models have made artificial intelligence readily available to the general public and potentially have a role in healthcare; however, their use in difficult differential diagnosis is still limited, as demonstrated by a case of necrotizing otitis externa. Abstract: This case report presents a peculiar case of necrotizing otitis externa (NOE) with skull base involvement which proved diagnostically challenging. The initial patient presentation and the imaging performed on the 78-year-old patient suggested a neoplastic rhinopharyngeal lesion and only after several unsuccessful biopsies the patient was transferred to our unit. Upon re-evaluation of the clinical picture, a clinical hypothesis of NOE with skull base erosion was made and confirmed by identifying Pseudomonas aeruginosa in biopsy specimens of skull base bone and external auditory canal skin. Upon diagnosis confirmation, the patient was treated with culture-oriented long-term antibiotics with complete resolution of the disease. Given the complex clinical presentation, we chose to submit a posteriori this NOE case to two large language models (LLM) to test their ability to handle difficult differential diagnoses. LLMs are easily approachable artificial intelligence tools that enable human-like interaction with the user relying upon large information databases for analyzing queries. The LLMs of choice were ChatGPT-3 and ChatGPT-4 and they were requested to analyze the case being provided with only objective clinical and imaging data.

Effects of rapid maxillary expansion on hearing loss and otitis media in cleft palate children.

PURPOSE: Otitis media with effusion (OME), recurrent acute otitis media and conductive hearing loss (CHL) are significantly prevalent in children with cleft palate (CP) and cleft lip and palate (CLP). Rapid Maxillary Expansion (RME) appears to have a positive effect also on middle ear disorders in these patients. The study aims to offer a prospective evaluation of RME effects in a group of patients with CP/CLP in terms of OME, CHL. METHODS: A prospective observational study was conducted. Thirty-four CP, CLP and submucosa cleft patients who received orthodontic indication to RME treatment for OME or conductive hearing loss in a tertiary institutional Care Unit of San Paolo Hospital, Milan (IT), were included. Twenty-two patients matched for age and with analogous inclusion criteria except for indication to RME treatment were enrolled in the control group. Clinical otolaryngological evaluation, pure tone audiometry and tympanogram were performed at the beginning of treatment (T0), at the end of the expansion (T1) and at 6-month follow-up (T2). MAIN OUTCOME MEASURES: Air-bone gaps and tympanogram results at each time interval. RESULTS: In the main group, RME allowed a statistically significant improvement of air-bone gaps (according to frequency, p < 0.001-0.089 T0 vs. T1 and < 0.001-0.044 T0 vs. T2, Friedman's test) and tympanometry results (p = 0.002 T0 vs. T1 and p < 0.001 T0 vs. T2, Friedman's test). Improvements were stable during follow-up and were significantly better in the main group than in the control group. CONCLUSION: CHL and middle ear effusion improved significantly overtime during RME and after 6 months of follow-up.

Adenoidectomy for middle ear disease in cleft palate children: a systematic review.

PURPOSE: Cleft palate children have a higher incidence of otitis media with effusion, more frequent recurrent acute otitis media episodes, and worse conductive hearing losses than non-cleft children. Nevertheless, data on adenoidectomy for middle ear disease in this patient group are scarce, since many feared worsening of velopharyngeal insufficiency after the procedure. This review aims at collecting the available evidence on this subject, to frame possible further areas of research and interventions. METHODS: A PRISMA-compliant systematic review was performed. Multiple databases were searched with criteria designed to include all studies focusing on the role of adenoidectomy in treating middle ear disease in cleft palate children. After duplicate removal, abstract and full-text selection, and quality assessment, we reviewed eligible articles for clinical indications and outcomes. RESULTS: Among 321 unique citations, 3 studies published between 1964 and 1972 (2 case series and a retrospective cohort study) were deemed eligible, with 136 treated patients. The outcomes were positive in all three articles in terms of conductive hearing loss improvement, recurrent otitis media episodes reduction, and effusive otitis media resolution. CONCLUSION: Despite promising results, research on adenoidectomy in treating middle ear disease in the cleft population has stopped in the mid-Seventies. No data are, therefore, available on the role of modern conservative adenoidectomy techniques (endoscopic and/or partial) in this context. Prospective studies are required to define the role of adenoidectomy in cleft children, most interestingly in specific subgroups such as patients requiring re-tympanostomy, given their known risk of otologic sequelae.

Nasal autoinflation devices for middle ear disease in cleft palate children: are they effective?

OBJECTIVE: Cleft lip palate (CLP) and cleft palate (CP) patients have a higher incidence of otitis media with effusion (OME) and conductive hearing problems. This article aims to evaluate the effectiveness of a 6-month course of self-administered autoinflation therapy in paediatric CP/CLP patients in terms of conductive hearing loss (CHL) and OME prevalence. METHODS: Fifty-one patients with surgically corrected CP/CLP and diagnosis of OME received indication to 6-months autoinflation therapy with an Otovent® device. Clinical evaluation, tympanogram and pure tone audiometry were carried out at the time of prescription (T0), at the end of treatment (T1) and at 6-month follow-up (T2). Patients were divided in 2 groups based on therapeutic compliance (29 compliant children, group A, vs 22 non-compliant children, group B). RESULTS: Case series showed better audiological results and tympanometries at both time points (p < 0.001). Group A showed better outcomes at tympanograms and at each frequency, but were statistically significant only in terms of CHL at 250 and 1000 Hz frequencies at T1 (respectively 0.024 and 0.012). CONCLUSIONS: Nasal autoinflation therapy accelerates improvement of OME and hearing thresholds at short-/mid-term, leading to an earlier improved hearing performance.

Isolated olfactory cleft involvement in SARS-CoV-2 infection: prevalence and clinical correlates.

PURPOSE: Smell alterations are a symptom of COVID-19 and have been associated with olfactory cleft mucosal thickening (OCMT). Although their pathogenesis is unclear, evidences link them to viral neuroinvasive potential. This study aims at estimating the prevalence of OCMT in CT scans of COVID-19 patients and investigating its clinical correlates. METHODS: In a single-institution retrospective cross-sectional study, we included all patients hospitalized for COVID-19 undergoing head CT scan for any reason. Exclusion criteria were history of recent head trauma or chronic rhinosinusitis; opacification > 2 mm in any sinonasal space other than the olfactory cleft; CT performed during/after invasive ventilation or feeding via nasogastric tube. We recorded the prevalence of OCMT and related it to age, sex, need for invasive ventilation during hospital stay, outcome, length of hospital stay, diffusion of lung SARS-CoV-19 lesions and outcome. RESULTS: 63 eligible patients were identified (39 male, 24 female; median age 77.82 ± 17.77 years). OCMT was identified in 16 patients (25.4%; 95% CI 15.3-37.9%). Patients with OCMT had longer hospital stays (median 16 ± 4 vs. 9 ± 14.5 days, p = .009, Mann-Whitney U test) and required invasive ventilation more frequently than patients without mucosal thickening (OR 4.89, 95% CI 0.96-24.89, p = .063, Fisher's test). No other difference was observed. CONCLUSION: OCMT affects nearly one in four patients hospitalized for COVID-19. It is associated with a worse disease course irrespective of age, sex and diffusion of lung lesions, although with no direct effect on survival.