JAMA otolaryngology-- head & neck surgery最新文献

Importance: A greater proportion of older patients (age ≥65 years) are undergoing thyroidectomy amidst an aging population. Given the elusive results in the literature, clarifying whether older patients face different thyroidectomy safety profiles is crucial for optimizing risk assessment and patient counseling.

Objective: To evaluate the association between older age (age ≥65 years) and thyroidectomy safety by comparing endocrine-related complications, including recurrent laryngeal nerve injury, hypocalcemia, hematoma, and mortality in older compared with younger patients.

Data sources: Ovid MEDLINE, CINAHL, Ovid Embase, and the Cochrane Library were systematically searched from inception to May 29, 2025; only English-language studies were included.

Study selection: Cohort studies comparing thyroidectomy outcomes between older and younger patients and reporting endocrine-related complications were included. Two independent reviewers screened abstracts and full texts. Of 3192 studies identified, 11 met inclusion criteria.

Data extraction and synthesis: Two independent reviewers extracted data following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Risk of bias across studies was evaluated using the ROBINS-I (Risk of Bias in Nonrandomized Studies of Interventions) tool. Random-effects models were used to pool data.

Main outcomes and measures: Incidence of recurrent laryngeal nerve injury, hypocalcemia, and hematoma; odds ratios (ORs) with 95% CIs were calculated.

Results: The analysis included 11 studies with a total of 427 886 patients of whom 104 232 (24.4%) were older adults (≥65 years). Older adults had greater odds of overall recurrent laryngeal nerve injury (2.3% vs 1.0%; OR, 1.58; 95% CI, 1.15-2.16; I2 = 76%), hematoma (2.4% vs 1.0%; OR, 2.32; 95% CI, 1.70-3.16; I2 = 0%), and mortality at last follow-up (0.3% vs 0.01%; OR, 11.09; 95% CI, 1.77-69.52; I2 = 90%) compared with younger patients. However, rates of overall hypocalcemia (OR, 0.80; 95% CI, 0.61-1.05; I2 = 71%) and wound complications (OR, 1.38; 95% CI, 0.92-2.06; I2 = 0%) were similar.

Conclusions and relevance: This systematic review and meta-analysis found that thyroidectomy is safe in older patients, with only small absolute increases in recurrent laryngeal nerve injury, hematoma, and mortality despite statistical significance. These findings support individualized counseling and perioperative planning while acknowledging the overall safety of thyroidectomy in this population.

Importance: Large language models (LLMs), a rapidly advancing domain of artificial intelligence (AI), are poised to transform administrative, clinical, and research paradigms. Although adoption across medicine has accelerated, otolaryngology represents only a fraction of this progress compared with other specialties. This review highlights the advantages of LLMs and innovations from other fields to provide otolaryngologists with a foundation for advancing these technologies. Given the unique reliance of otolaryngology on multimodal data (text, imaging, electrophysiology, and video) and its symptom-driven complexity, LLMs represent a powerful but underused tool for advancing patient care.

Observations: LLMs efficiently combine the strengths of deep learning and natural language processing, enabling revolutionary language abilities. Most otolaryngology research has focused narrowly on question-answering tasks with limited clinical integration. In contrast, other specialties have demonstrated broad methodological applications, including (1) converting unstructured data (eg, notes, reports) into structured variables; (2) automated phenotyping and subphenotyping to advance precision medicine; (3) streamlining administrative tasks; (4) developing domain-specific decision support tools; and (5) using multimodal language models that integrate text with images. Importantly, LLMs are accessible and adaptable, requiring far less labeled data than traditional machine learning models, enabling broader applications to researchers. Yet, in otolaryngology, most studies remain limited to feasibility evaluations with closed-source models, constraining their clinical utility and translational potential.

Conclusions and relevance: LLMs hold significant promise to enhance biomedical research and patient care. Otolaryngology, with its rich clinical, functional, and multimodal data, is well positioned to benefit from these tools. However, progress requires moving beyond feasibility studies toward clinical trial-like validation and implementation research. Future efforts should prioritize open-source model development, domain-specific fine-tuning, secure multi-institutional deployment, and legal oversight to ensure transparency, reproducibility, and generalizability. By aligning with advancements in other specialties, otolaryngology can leverage its unique resources to pioneer responsible LLM applications that improve efficiency, augment decision support, and transform patient outcomes.

Importance: While hypoglossal nerve stimulators (HGNS) have been shown to be associated with reduced apnea-hypopnea index (AHI) scores in many continuous positive airway pressure-intolerant patients with obstructive sleep apnea (OSA), it is not effective for all patients, and it is unclear who would benefit most.

Objective: To explore baseline factors associated with response to HGNS and create a prognostic stratification system.

Design, setting, and participants: This retrospective cohort study included patients with OSA who underwent implant with an HGNS from April 2019 to October 2023 at a single tertiary care center and received a postimplant sleep study. During that time, 194 patients underwent HGNS implant. Data were analyzed from February 2024 to April 2024.

Intervention or exposure: HGNS implant and sleep study.

Main outcomes and measures: Response according to the modified Sher criteria (>50% reduction in AHI score and AHI <15 postimplant).

Results: Among the 119 patients included, the median (range) age was 63 (33-79) years, and 36 (30%) were female. Of the 119 included, 83 (70%) were responders. Neck size, body mass index (BMI; calculated as weight in kilograms divided by height in meters squared), AHI score, and comorbidity burden were associated with response and used to create a 4-category clinical severity staging system. Patients with small necks (≤14 inches for women or ≤16 inches for men), a BMI of less than 30, and an AHI score of 30 or less had the highest response rate (11 [91%]). Patients with large necks (>14 inches for women or >16 inches for men), a BMI of 30 or greater, an AHI score of 30 or greater, and comorbidities had the lowest response rate (16 [38%]). The staging system had moderate discriminative power, with a C statistic of 0.68 (95% CI, 0.57-0.78).

Conclusions and relevance: The results of this cohort study suggest that neck size, BMI, AHI score, and comorbidity burden are 4 clinically relevant and easily obtainable values that are associated with response to HGNS. While this staging system may help physicians identify patients who are most likely to benefit from implant, these results need to be validated in a separate cohort.

Importance: Excessive bleeding during endoscopic sinus surgery (ESS) impairs visualization and increases complications. Controlled hypotension is widely used; however, comparative effectiveness of available pharmacological agents remains uncertain.

Objective: To compare the effectiveness and safety of hypotensive drugs for reduced bleeding in ESS.

Data sources: PubMed, Scopus, Web of Science, and Cochrane databases were searched from inception to December 2024, and updated to include studies up to May 2025.

Study selection: Randomized clinical trials of patients undergoing ESS and receiving dexmedetomidine, clonidine, esmolol, labetalol, bisoprolol, metoprolol, diltiazem, nifedipine, or verapamil vs placebo. A total of 1258 records were screened.

Data extraction and synthesis: Data were extracted and assessed using the Cochrane risk of bias tool. Certainty was rated using the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) framework. A random-effects meta-analysis was conducted in accordance with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement guidelines.

Main outcomes and measures: Surgical bleeding score, intraoperative mean arterial pressure (MAP), and mean intraoperative heart rate. Secondary outcomes included blood loss, surgery duration, and emergence time.

Results: The analysis included 52 randomized clinical trials involving 3526 participants; of these, 43 studies contributed to the network analysis. Several agents were associated with reduced bleeding vs placebo: diltiazem (mean difference [MD], -1.25; 95% CI, -2.13 to -0.37), esmolol (MD, -1.16; 95% CI, -1.80 to -0.51), dexmedetomidine (MD, -1.09; 95% CI, -1.48 to -0.70), labetalol (MD, -1.00; 95% CI, -1.58 to -0.42), and clonidine (MD, -0.69; 95% CI, -1.23 to -0.16). Dexmedetomidine was associated with the most reduced MAP (MD, -30.30 mm Hg; 95% CI, -47.91 to -12.69), followed by clonidine (MD, -28.61; 95% CI, -53.11 to -4.11), esmolol (MD, -27.62; 95% CI, -49.73 to -5.51), and labetalol (MD, -26.54; 95% CI, -47.20 to -5.87). At 60 minutes, bisoprolol (MD, -58.30; 95% CI, -67.50 to -49.10 beats per minute [bpm]), verapamil (MD, -49.90; 95% CI, -58.98 to -40.82 bpm), and labetalol (MD, -43.89; 95% CI, -54.59 to -33.18 bpm) produced the largest heart rate reductions. Esmolol (MD, -3.67; 95% CI, -4.21 to -3.13 minutes) and labetalol (MD, -3.64; 95% CI, -4.79 to -2.49 minutes) shortened emergence time.

Conclusions and relevance: In this systematic review and network meta-analysis, controlled hypotension improved surgical field quality in ESS with comparable safety. Diltiazem, esmolol, dexmedetomidine, labetalol, and clonidine were effective; dexmedetomidine was associated with reduced MAP the most, and esmolol favored recovery.

Importance: Merkel cell carcinoma (MCC) is a rare neuroendocrine malignant neoplasm often affecting the head and neck. Sentinel lymph node biopsy (SLNB) is routinely performed; however, the accuracy and reliability of SLNB for head and neck MCC remains unknown.

Objective: To determine the accuracy of SLNB in head and neck MCC and inform pre- and post-SLNB risk stratification.

Design, setting, and participants: This was a single-center cohort study of patients with clinically node-negative head and neck MCC who underwent SLNB from January 2006 to January 2025; median (IQR) follow-up was 1.9 (0.8-4.6) years. Data analysis was conducted from April to November 2025.

Interventions: SLNB outcome and nodal disease status.

Main outcomes and measures: SLNB accuracy measured by sensitivity, negative predictive value (NPV), and false negative (FN) rate. FN SLNB was defined as regional nodal recurrence at the first site of recurrence after a negative SLNB without concurrent local or in-transit recurrence. Factors associated with failed SLNB (defined as nonvisualization on lymphoscintigraphy, inability to identify the sentinel lymph node intraoperatively, or absence of nodal tissue in the pathology specimen) were evaluated using effect-size measures. Cox proportional hazards model identified factors associated with nodal disease, comparing patients who developed nodal disease (identified by SLNB or subsequent clinical and/or imaging evaluation) with those who did not develop nodal disease during follow-up.

Results: Of 86 participants (mean [SD] age, 75.6 [9.6] years; 36 female [41.9%] and 50 male [58.1%] individuals) with clinically node-negative head and neck MCC, 69 (80.2%) underwent SLNB. Occult nodal disease was identified in 19 participants (27.5%). SLNB sensitivity was 52.6% (10 of 19 patients), NPV was 88.6% (31 of 35 patients), and FN rate was 26.7% (4 of 15 patients). SLNB failed in 29.0% (20 of 69 patients), with tumor location, female sex, and older age having medium to large effect on SLNB failure. Lymphovascular invasion, infiltrative growth pattern, and tumor size had the largest effect size on nodal disease with Cox analysis.

Conclusions and relevance: In this large single-site cohort study, SLNB remained a useful tool in head and neck MCC, identifying more than half of occult nodal disease. Given reduced performance in this region, patients with failed SLNB or negative SLNB plus high-risk features may benefit from tailored management, including intensified therapy and surveillance.