From consensus to practice: How artificial intelligence can advance the core objectives of anaesthesia

We have read with great interest the recent article by Berger Estilita et al., "The definition and aims of Anesthesia: a Delphi based consensus statement" [1]. We commend the authors for establishing a comprehensive, patient-centered framework that defines the core objectives of modern anaesthetic practice. The consensus on 49 distinct aims across preoperative, intraoperative, postoperative, long-term, patient-centeredness, and sustainability domains provides a much-needed roadmap for advancing the quality and safety of perioperative care. The stated aims—ranging from preoperative risk stratification and cognitive protection to ensuring long-term quality of life and environmental sustainability—are ambitious and reflect the growing complexity of anaesthesiology. While the consensus statement provides the "what," we believe the "how" will be significantly shaped by the integration of new technologies such as Artificial Intelligence (AI), wearable devices, and telemedicine. These technologies offer powerful tools to not only meet but exceed the laudable goals established by this Delphi consensus. In the preoperative phase, new technologies can revolutionize risk assessment. The aim to "identify frail and other high-risk patients" (PRE7) can be substantially enhanced by ML models that analyze vast datasets from electronic health records, imaging, and genomics to predict individual patient risk for complications like delirium, organ failure, or prolonged recovery with greater accuracy than traditional scoring systems [2]. Furthermore, AI can help "protect cognitive functions" (PRE5) by identifying patients at high risk for postoperative cognitive dysfunction, enabling clinicians to implement personalized neuroprotective strategies. During the intraoperative phase, the goal of maintaining homeostasis and providing "anaesthesia so that surgery can occur" (INT2, INT4) while ensuring a "sufficient level of sedation/ anaesthesia and to avoid oversedation" (INT7) is a prime area for AI-driven improvement. Closed-loop systems, guided by AI algorithms analyzing real-time physiological data (e.g., EEG, hemodynamics), can titrate anaesthetic and vasoactive drugs with a precision that surpasses human capability [3]. This not only ensures optimal surgical conditions but also minimizes the risks of intraoperative awareness, hemodynamic instability, and excessive anaesthetic depth, which are linked to adverse postoperative outcomes. The postoperative and long-term outcomes outlined in the consensus—such as ensuring prompt recovery (POS1), avoiding complications (DIS2), and preventing the chronification of pain (LTO7)—can also be addressed by AI. Predictive analytics can forecast patients’ recovery trajectories, identify those at high risk for complications, and enable early, targeted interventions. For instance, ML models can predict which patients are most likely to develop chronic postsurgical pain, allowing for the implementa tion of preemptive, multimodal analgesic plans [4]. By continu ously learning from patient-reported outcomes (PROs), AI systems can help refine and personalize care pathways to "maintain the patient’s quality of life" (LTO2). Moreover, AI is pivotal for achieving the aim of a "continuous and systematic review with learning from outcomes" (PC7). AI powered data platforms can automate the collection and analysis of vast amounts of perioperative data, turning every anaesthetic procedure into a learning opportunity. This creates a virtuous cycle of continuous quality improvement that is unachievable through manual audits alone. Finally, even the novel and crucial aim of sustainability (SUS1) can be supported by AI. By optimizing operating room workflows, predicting supply needs, and minimizing waste, AI can contribute to a more efficient and environmentally responsible practice, reducing the ecological footprint of anaesthesia [5]. In conclusion, the consensus statement by Berger-Estilita et al. provides an essential framework for the future of anaesthesia. We propose that the next logical step is to systematically explore, validate, and implement AI-driven solutions to achieve these aims. By embracing AI, the global anaesthesiology community can accelerate the transition towards a future of precision, patient centered, and sustainable perioperative care, turning these aspirational goals into a clinical reality for every patient.