Rogério Aparecido Dedivitis

Decision-making is one of the most difficult and important tasks performed by surgeons. Clinical guidelines and, more recently, Artificial Intelligence (AI) have been implemented to support these decisions. However, there is a paucity of research exploring the impact of AI on surgical decision-making strategies. To evaluate the accuracy of AI in informing clinical decision-making for the treatment of advanced oral cancer. Structured questions based on seven selected clinical recommendations were formulated and processed through multiple large language model (LLM) platforms. Claude (Sonnet 3.5) was used to identify discrepancies between established guidelines and LLM-generated responses. Subsequently, three subject matter experts assessed the identified differences to evaluate their clinical significance. Analysis of the 28 generated responses revealed that 21 (75%) showed congruence between LLM outputs and established guideline recommendations. In three instances, guidelines provided significantly more comprehensive content than LLM responses. One LLM response contained additional relevant information, while three responses were contradictory to the guidelines. The LLMs demonstrated 75% concordance with guideline recommendations and should be considered complementary tools for established clinical guidelines.

Background/Objectives: Laryngeal cancer has a high incidence and mortality rate, especially in advanced stages. Stoma recurrence after total laryngectomy is rare but lethal, and its risk factors remain controversial. We aimed to assess whether the presence of a tracheotomy performed prior to surgical treatment for advanced laryngeal cancer represents a risk factor for recurrence of the disease in the stoma and interferes with patient survival. Methods: A retrospective study was carried out in 138 patients who underwent total laryngectomy. Clinical-pathological variables were related to the presence of a preoperative tracheotomy. Results: Fifty-two patients without and 86 patients with preoperative tracheotomy were studied. There was a significant predominance of transglottic tumors (p = 0.02) and no patients with early-stage disease (T1 and T2, p = 0.002) in the tracheotomy group. There was no statistically significant difference in stoma recurrence (p = 0.53). On multivariate analysis, the risk of stoma recurrence was not significantly associated with the presence of a preoperative tracheotomy. Conclusions: The presence of a tracheotomy prior to total laryngectomy was not a risk factor for disease recurrence in the stoma or for reduced survival in the group.

Introduction: Introduction: The use of portable ventilators (PV) adapted to tracheostomy has been observed in critical patients in clinical practice. Objective: The study aimed to identify the characteristics associated with the indication for PV, patient outcomes, and survival. Methods: This was a prospective cohort study conducted with adult tracheostomized patients. Data collected included demographics, severity scores, reasons for ICU admission and intubation, days of mechanical ventilation (MV) until tracheostomy, cause of tracheostomy, and outcomes (ICU and hospital). Univariate analysis compared the groups with and without PV. Odds ratios (OR) assessed the likelihood of PV use, and age-adjusted logistic regression evaluated associated variables. Hospital survival was studied using the Kaplan-Meier method. Results: A total of 95 patients were included. The mean age was 71 years, with 58% being female. The median SAPS II was 55, and the Charlson Comorbidity Index was 4. Patients with an indication for PV (n=71) were older and had a longer continuous sedation time. Those intubated due to acute respiratory failure had a lower chance of using PV than those intubated for other causes. Hospital survival was higher in the PV group; however, the hospital mortality rate did not differ between the groups. Conclusion: PV use in tracheostomized patients was greater in the elderly and those intubated due to neurological causes or COPD. Despite higher survival, the similar hospital mortality rate suggests the need for further interventions. The literature reinforces that a multidisciplinary approach is essential to manage the complex dysfunctions in these patients with prolonged hospitalization.

Resumo: As grandes lesões da língua representam um desafio significativo, que exige uma reconstrução tridimensional minuciosa para alcançar a ressecção oncológica e as funções ideais de fala e deglutição para o paciente. Neste relato de caso, apresentamos uma reconstrução por meio de um retalho microcirúrgico livre anterolateral da coxa em forma de catedral para restaurar os dois terços anteriores da língua e o assoalho da cavidade oral após a ressecção de câncer de língua avançado. O paciente teve uma rápida recuperação e restauração da função fisiológica.

SUMMARY BACKGROUND: Moral distress is defined as a psycho-emotional experience that occurs when individuals recognize the morally appropriate course of action but are unable to carry it out due to external constraints such as institutional hierarchies, organizational norms, or authority pressures. In medical education, this phenomenon is particularly frequent, given the deeply ethical nature of clinical decisions and patient care. OBJECTIVE: The aim of this study was to translate and adapt the Moral Distress from Professionalism Dilemmas checklist into Brazilian Portuguese. METHODS: The translation and cross-cultural adaptation process followed internationally recognized guidelines. It included independent translations by two professionals (one with and one without medical training), synthesis of the versions, back-translation, expert committee review, pretesting with 200 medical interns, and final evaluation. RESULTS: Of the 200 medical students invited, 160 (80%) completed the checklist, with no significant differences by year or gender. Depending on the item, between 0 and 57.5% of participants reported exposure to the professionalism dilemmas investigated. Among exposed students, the intensity of moral distress varied considerably across situations. CONCLUSION: Moral distress is a prevalent phenomenon in medical education. The adapted checklist demonstrated semantic and conceptual equivalence to the original and may serve as a valuable tool to identify, quantify, and understand moral distress among medical students in clinical training.

This systematic review investigates the cytogenotoxicity of various food preservatives in mammalian cells, including sodium benzoate and potassium sorbate, through a comprehensive analysis of studies retrieved from PubMed, SCOPUS, and Web of Science. An orderly search conducted in March 2025 identified 19 relevant studies (from an initial 594), which employed assays, such as the micronucleus test and the comet assay to assess DNA damage, and MTT assay and polychromatic/normochromatic erythrocytes (PCE/NCE) ratio for evaluating cytotoxicity. Among these, 13 studies (68 %) reported genotoxic effects, with sodium benzoate being the most frequently associated with micronucleus formation and chromosomal abnormalities. Additionally, 12 studies (63 %) described cytotoxic effects, evidenced by decreased cell viability, altered proliferation indices, or nuclear alterations. As for the quality assessment, 18 studies (out of 19) were categorized as strong (n = 15) or moderate (n = 3) and, therefore, we consider our findings to be trustworthy. In summary, the consistent association between exposure to food preservatives and cytogenotoxic outcomes highlights the importance of monitoring such compounds and establishing clearer safety thresholds to protect human health. Certainly, these findings are important for clarifying the role of biomarkers related to cytogenotoxicity due to food preservative consumption in humans.

Objective: To evaluate the accuracy of artificial intelligence (AI) in establishing clinical decision‐making in the treatment of advanced laryngeal cancer. Methods: A structured question was elaborated for each of the seven recommendations chosen. Each Large Language Model (LLM) platform answered the questions. The Claude platform identified the differences between the guidelines and the responses generated and three specialists evaluated the impact of such differences. Results: Of the 28 analyzed responses, 22 (78.6%) demonstrated content similarity with existing guidelines. Two responses showed that guidelines contained significantly more comprehensive content, three responses from LLMs provided additional content not demonstrated in the guidelines, and one response showed direct disagreement with established guidelines. Conclusion: There was a 78.6% overlap in responses between guideline recommendations and LLMs. Therefore, while AI holds promise for transforming guideline creation, its integration into clinical practice must be carefully evaluated to ensure that it complements, rather than replaces, established expert‐driven protocols. Level of Evidence: 4.