The impact of artificial intelligence on personalized selection of general anesthetic agents to optimize difficult airway management

This study aimed to evaluate the impact of artificial intelligence (AI)–based decision support systems on the personalized selection of general anesthetic agents in patients with predicted difficult airway, focusing on optimization of airway management and peri-induction safety.

A narrative review of the literature was conducted using PubMed and Scopus databases, including English-language studies addressing AI applications in anesthesiology, airway assessment, anesthetic agent selection, and perioperative risk prediction. Articles describing machine learning models, predictive algorithms, and AI-assisted clinical decision systems were qualitatively analyzed.

The reviewed evidence suggests that AI-based decision support models can integrate patient-specific airway predictors, comorbidities, and physiological parameters to guide individualized anesthetic strategies. In patients with predicted difficult airway, AI-assisted systems may support tailored use of agents such as propofol by identifying individuals at increased risk of apnea or hemodynamic instability, enabling dose adjustment or alternative induction approaches. Furthermore, machine learning models may help optimize the administration of neuromuscular blocking agents, such as rocuronium, balancing optimal intubation conditions with airway safety.

The availability of sugammadex is highlighted as an important component of anesthetic planning, as AI-driven recommendations can incorporate the option of rapid and reliable reversal of neuromuscular blockade. This integration may reduce the likelihood of critical scenarios such as “cannot intubate, cannot ventilate” situations by supporting more strategic and safety-oriented decision-making in high-risk cases.

Overall, AI-based tools demonstrate significant potential to enhance anesthetic decision-making, improve patient safety, and optimize difficult airway management through individualized pharmacological strategies. Although current evidence supports their promising clinical utility, further prospective studies and real-world validation are required before widespread implementation in anesthetic practice.