Grand challenges in anesthetic pharmacology: toward personalized, outcome-focused practice

Introduction

Modern general anesthesia has evolved into balanced anesthesia, combining hypnosis, analgesia, amnesia, immobility, autonomic stability, and recovery through pharmacological and non-pharmacological strategies (1). In contemporary practice, general anesthesia usually involves an intravenous hypnotic for induction, opioid or non-opioid analgesics, neuromuscular blocking agents when required, and maintenance with either volatile anesthetics or intravenous hypnotics (2–6). Although general anesthesia is remarkably safe in the short term, anesthesiology is now facing broader questions that extend beyond intraoperative survival and immediate recovery. These include how pharmacokinetic/pharmacodynamic (PK/PD) models can support personalized dosing; how drug-drug interactions in multimodal regimens can be anticipated rather than treated empirically; and how molecular effects such as mitochondrial dysfunction, endoplasmic reticulum (ER) stress, and neuroinflammation may influence postoperative delirium, persistent pain, cancer-related outcomes, and neurodevelopment (7–11). This editorial review discusses the major challenges that anesthetic pharmacology will face in future clinical practice.

Current standard method of general anesthesia

Volatile anesthetic-based balanced anesthesia

The most common pattern of general anesthesia remains intravenous induction followed by maintenance with a volatile anesthetic. In adult practice, this often includes propofol induction, analgesia with fentanyl, sufentanil, or remifentanil, rocuronium to facilitate airway management, and maintenance with sevoflurane, desflurane, or isoflurane (1, 12). Sevoflurane is widely used because of its favorable airway profile and suitability for both adult and pediatric anesthesia. Desflurane allows rapid emergence because of its low blood-gas solubility, whereas isoflurane remains useful in cost-conscious settings (1, 12). Nitrous oxide (N2O) remains in worldwide use as an adjunct to general anesthesia and as a rapidly titrated analgesic and anxiolytic in dentistry, prehospital and emergency care, and labor analgesia. Its low blood-gas solubility permits rapid onset and offset and can reduce requirements for more potent volatile agents and opioids, although it is not sufficiently potent to provide surgical anesthesia alone. N2O did not increase death or major cardiovascular complications, supporting its cardiovascular safety when appropriately used (13).

Volatile anesthesia offers practical advantages, including measurable end-tidal concentrations, predictable delivery through modern anesthesia workstations, and easy titration. However, volatile anesthetics are associated with postoperative nausea and vomiting, malignant hyperthermia susceptibility, greenhouse gas emissions, and potential effects on neuroinflammation and tumor biology, although clinical evidence for long-term outcomes remains inconsistent (11, 12, 14–16).

Total intravenous anesthesia

Total intravenous anesthesia (TIVA) generally refers to maintenance anesthesia based on intravenous hypnotics such as propofol or remimazolam, usually combined with continuous remifentanil administration (1). TIVA is particularly useful in patients at high risk of postoperative nausea and vomiting, patients susceptible to malignant hyperthermia, and situations where rapid, clear emergence is desired (14, 15, 17). Unlike volatile anesthesia, plasma and effect-site concentrations are usually estimated rather than directly measured. Recent systematic reviews suggest that inhalational anesthesia and TIVA are broadly comparable in overall safety, although propofol-based TIVA tends to reduce postoperative nausea and vomiting and may reduce emergence agitation in selected populations (1, 14). Limitations of TIVA include dependence on infusion pumps and reliable venous access, the possibility of unrecognized interruption of drug delivery, and the need for appropriate monitoring of hypnotic effect.

Target-controlled infusion (TCI) is a model-based method of administering intravenous anesthetics that allows the clinician to select a predicted plasma or effect-site target concentration while a microprocessor adjusts the infusion rate (1). TCI can make titration more systematic than a fixed manual infusion and can support individualized care by combining patient covariates, contemporary general-purpose PK/PD models, and continuous clinical and electroencephalographic feedback (18). However, TCI displays predicted rather than measured concentrations; model misspecification, extreme physiology, obesity, frailty, low cardiac output, organ failure, and propofol-opioid interactions can produce substantial prediction error.

Combined general-regional anesthesia and multimodal analgesia

Regional anesthesia is increasingly integrated with general anesthesia to improve postoperative recovery. Epidural analgesia, peripheral nerve blocks, fascial plane blocks, and wound infiltration, can reduce opioid requirements and improve pain control (6). These techniques may reduce anesthetic requirements, facilitate early mobilization, and improve quality of recovery, even when general anesthesia remains necessary. Modern anesthesia is therefore moving away from opioid-heavy practice toward opioid-sparing, rather than strictly opioid-free, strategies. Acetaminophen, nonsteroidal anti-inflammatory drugs, ketamine, dexmedetomidine, local anesthetics, and regional anesthesia are used to reduce opioid exposure while maintaining adequate analgesia (6). The goal is to reduce respiratory depression, postoperative nausea and vomiting, ileus, urinary retention, opioid-induced hyperalgesia, and persistent opioid use after surgery.

Principal drugs used in contemporary general anesthesia

Intravenous hypnotics

Propofol remains the dominant intravenous induction agent because of its rapid onset, smooth recovery profile, antiemetic effect, and compatibility with TIVA (1). Its limitations include hypotension, respiratory depression, injection pain, and rare but serious propofol infusion syndrome during prolonged high-dose administration. Remimazolam is an ultrashort-acting benzodiazepine metabolized by tissue esterases. It may offer hemodynamic stability and reversibility with flumazenil. Recent systematic reviews suggest that remimazolam reduces hypotension compared with propofol, while delirium-related outcomes in older patients appear broadly comparable (3, 4). At present, remimazolam is best regarded as an alternative for selected patients, particularly older or hemodynamically fragile patients (3, 4). Ketamine provides hypnotic and analgesic effects through N-methyl-D-aspartate receptor antagonism and is useful in trauma, shock, opioid tolerance, and pain sensitization, although emergence reactions and sympathomimetic effects limit routine use. Esketamine is used as an intravenous anesthetic in some regions and as a nasal spray for treatment-resistant depression in the United States and elsewhere (19). Etomidate produces rapid hypnosis with relatively little cardiovascular depression and is therefore useful for induction in patients with severe hemodynamic instability or limited cardiac reserve. It does not provide analgesia, and its adverse effects include myoclonus, injection pain, postoperative nausea and vomiting, and dose-dependent inhibition of adrenal 11beta-hydroxylase. In a recent multicenter randomized trial of 2,365 critically ill adults requiring emergency tracheal intubation, ketamine did not reduce 28-day in-hospital mortality compared with etomidate, whereas cardiovascular collapse during intubation occurred more often with ketamine (22.1% vs. 17.0%) (20).

Volatile anesthetics, opioids, and neuromuscular blocking agents

Sevoflurane, desflurane, and isoflurane remain the major volatile anesthetics. All can trigger malignant hyperthermia in susceptible individuals and should be avoided in such patients (15, 17). Fentanyl remains a standard intraoperative and postoperative analgesic, while remifentanil allows precise intraoperative titration because of rapid esterase metabolism. However, inadequate transitional analgesia after discontinuation of remifentanil may contribute to severe early postoperative pain, acute tolerance, and opioid-induced hyperalgesia (21). Intravenous lidocaine, when appropriate for the procedure and patient, is a systemic non-opioid adjunct that can reduce perioperative opioid requirements and improve early postoperative analgesia as part of a multimodal regimen (6). Rocuronium is widely used because it can be reversed specifically by sugammadex. Cisatracurium is useful in patients with renal or hepatic impairment because of organ-independent Hofmann elimination. Succinylcholine remains important for rapid sequence induction and treatment of laryngospasm, but its use is restricted by the risks of hyperkalemia and malignant hyperthermia (5, 15, 17). Residual neuromuscular blockade is a major preventable cause of postoperative respiratory complications (5).

Perioperative outcomes may be determined by combined exposure to hypnotics, analgesics, neuromuscular blockers, reversal agents, and adjuvant drugs rather than by any single agent. Quantitative monitoring, dose-response modeling, and transition planning from intraoperative to postoperative analgesia are essential components of precision anesthetic pharmacology.

Major problems and controversies

Volatile anesthesia versus TIVA

The comparison between volatile anesthesia and TIVA remains a central debate in anesthesiology (2). TIVA may reduce postoperative nausea and vomiting and emergence agitation in selected patients (1, 14). However, whether TIVA improves major long-term outcomes remains unresolved. Observational studies have sometimes favored propofol-based anesthesia for neurocognitive or oncological outcomes, but confounding by patient selection, surgical factors, perioperative inflammation, and institutional practice remains difficult to exclude. Recent reviews and comparative studies continue to show mixed results (1, 2, 7, 22). Therefore, anesthetic technique should currently be selected according to patient risk, surgical context, institutional expertise, and recovery goals rather than an assumed universal superiority of either approach.

Postoperative delirium and neurocognitive disorders

Postoperative delirium and postoperative neurocognitive impairment are multifactorial syndromes. Age, frailty, pre-existing cognitive impairment, hypotension, sleep disruption, pain, systemic inflammation, and surgical stress may be more important than the anesthetic drug alone (4, 22). Recent meta-analyses suggest that delirium-related outcomes for remimazolam and propofol in older patients are generally similar, while some cardiac surgery studies suggest possible differences between volatile anesthesia and TIVA (4, 22). From a pharmacological standpoint, neurocognitive protection should focus not only on drug selection but also on avoiding excessive anesthetic depth, hypotension, hypoxemia, uncontrolled pain, inflammatory burden and residual neuromuscular blockade (2, 4, 5, 22, 23). Electroencephalography-based monitoring may support individualized hypnotic titration, particularly in older or neurologically vulnerable patients, although its impact on long-term outcomes remains debated.

Pediatric neurotoxicity

Regulatory warnings have raised concern about repeated or prolonged exposure to anesthetics and sedatives in very young children (24). No specific anesthetic has been shown clinically to be safer than another with respect to neurodevelopment. Systematic reviews continue to support caution regarding prolonged or repeated exposure during early neurodevelopment (8, 24). At the same time, necessary surgery and adequate analgesia should not be withheld, and current clinical interpretation suggests that a single short exposure is unlikely to cause major neurodevelopmental harm (8, 24).

Cancer recurrence and immune modulation

Mechanistic studies suggest that volatile anesthetics, opioids, local anesthetics, propofol, surgical stress, inflammation, and regional anesthesia may influence immunosurveillance and tumor biology (7, 16). However, clinical evidence remains inconsistent. A recent randomized multicenter trial did not demonstrate a clear oncological benefit of propofol-based maintenance anesthesia over volatile anesthesia (7). Although regional anesthesia and local anesthetics may have biologically plausible antitumor effects, current evidence is insufficient to select an anesthetic technique based solely on oncological considerations (16). Future work should examine perioperative immune pharmacology, opioid exposure, local anesthetic biology, and surgical inflammatory stress as interacting determinants rather than isolated variables.

Dexamethasone and chronic postsurgical pain

Dexamethasone is routinely used to prevent postoperative nausea and vomiting and can reduce acute postoperative pain and opioid consumption, but its relationship with chronic postsurgical pain is controversial. A prespecified wound-pain analysis similarly found more surgical-wound pain at 6 months after dexamethasone 8 mg (11.5% vs. 9.6%; relative risk 1.20), although pain severity and neuropathic features did not differ (25). In contrast, a propensity score-matched analysis of ENIGMA-II data found no increased risk of chronic postsurgical pain at 12 months (adjusted relative risk 0.88, 95% confidence interval 0.61–1.27) (26). Current evidence therefore raises a signal that deserves dedicated randomized study but does not establish that a single antiemetic dose of dexamethasone causes chronic postsurgical pain.

Environmental sustainability

Environmental impact has become an important determinant of anesthesia strategy. Desflurane has a high global warming potential compared with other commonly used volatile anesthetics, and many institutions are reducing or eliminating its use while emphasizing low-flow anesthesia and minimizing nitrous oxide exposure (12, 27). However, TIVA is not environmentally neutral because of plastics, syringes, infusion pumps, manufacturing processes, and propofol waste (12). The most reasonable current approach is to prioritize patient safety while adopting clinically appropriate methods with the lowest feasible environmental impact.

Malignant hyperthermia and trigger avoidance

Malignant hyperthermia is a rare but life-threatening pharmacogenetic complication associated with volatile anesthetics and succinylcholine. Current guidelines emphasize trigger avoidance, preparation of alternative anesthesia machines, availability of dantrolene, and rapid recognition and treatment (15, 17). In susceptible patients, TIVA remains the standard trigger-free general anesthesia technique.

ER stress, mitochondrial dysfunction, and inflammation

At the cellular level, anesthetics interact with calcium homeostasis, mitochondrial function, oxidative stress, the unfolded protein response (UPR), neuroinflammation, apoptosis, and autophagy (9, 10). Volatile anesthetics may disturb ER calcium handling and protein-folding capacity, thereby activating ER stress-related pathways. The UPR is not a single response but a coordinated network involving PERK-eIF2alpha-ATF4-CHOP signaling, IRE1-XBP1 and stress kinase pathways, and ATF6-mediated transcriptional regulation. Transient activation may be adaptive, whereas excessive or prolonged ER stress can promote mitochondrial dysfunction, inflammatory signaling, synaptic injury, and cell death (28).

Different anesthetic and adjuvant drugs may influence this stress network in distinct ways. Volatile anesthetics have been reported to modulate ER stress and neuroinflammatory cascades (9). Propofol has shown antioxidant, anti-inflammatory, and immunomodulatory effects in experimental settings (29), whereas dexmedetomidine may attenuate perioperative stress and inflammatory responses while preserving immune function (11). Ketamine and local anesthetics may also influence neuroimmune and nociceptive signaling through NMDA receptor-dependent and sodium-channel-dependent mechanisms.

Traditional anesthetic endpoints–loss of consciousness, immobility, hemodynamic stability, and recovery time–do not capture ER stress, mitochondrial dysfunction, or microglial activation. Future pharmacological studies should incorporate biomarkers such as inflammatory cytokines, neuronal injury markers, oxidative stress markers, metabolomics, mitochondrial function assays, and potentially neuroimaging or cerebrospinal fluid biomarkers in selected high-risk populations.

Pharmacogenetics and its impact on anesthesia

Pharmacogenetic information is beginning to convert selected rare or unpredictable anesthetic responses into actionable preoperative decisions. Pathogenic RYR1 (ryanodine receptor isoform 1) or CACNA1S gene variants can support trigger-free anesthesia in patients with malignant hyperthermia susceptibility (30), while BCHE (pseudocholinesterase) gene variants explain prolonged neuromuscular block after succinylcholine or mivacurium (31). The m.11232T > C mitochondrial DNA variant may be at increased risk of severe neurologic deterioration after exposure to anesthetic agents like sevoflurane (32). Associations involving CYP2B66* with propofol disposition, CYP2D6 with codeine, tramadol, and ondansetron response, and OPRM1 or COMT with opioid response and pain sensitivity may eventually refine perioperative drug selection (33).

However, given the variability in the magnitude of effect, reproducibility, turnaround time, cost-effectiveness, and clinical utility, the routine use of broad panel testing for the majority of surgical patients is not yet justified. Priority should instead be given to targeted testing—particularly in cases involving personal or family history, prior unexplained adverse drug reactions, or established, actionable gene-drug combinations associated with significantly higher pre-test probability—and the results should be integrated with PK/PD modeling and real-time monitoring.

Future perspectives

Future general anesthesia will be increasingly personalized, monitored, multimodal, sustainable, and outcomes-oriented. First, anesthetic dose and technique will be adapted to patient vulnerability rather than determined by fixed drug protocols. PK/PD models should be continuously refined for elderly, frail, obese, critically ill, pediatric, and organ-failure populations, which may guide the choice and dose of hypnotics, analgesics, and neuromuscular blocking agents (2, 4, 5, 22). Second, objective monitoring will become more central. Quantitative neuromuscular monitoring should be standard whenever neuromuscular blocking agents are used (5), and brain monitoring may support individualized hypnotic administration, especially in elderly patients and during TIVA. Third, anesthesia will continue to shift from opioid-heavy techniques to rational multimodal analgesia. Prevention of chronic postsurgical pain may become an increasingly important endpoint, linking intraoperative analgesic strategies with long-term recovery. Multimodal analgesia should evolve from empirical drug combination to interaction-aware pharmacological design. Fourth, environmental sustainability will become an ethical, institutional, and regulatory priority. Reduction of desflurane and nitrous oxide use, low-flow anesthesia, careful management of propofol waste, and anesthetic gas capture technologies are likely to expand (12, 27). Finally, research into neuroinflammation, immunomodulation, mitochondrial dysfunction, ER stress, chronic pain, and pharmacogenetics may generate new strategies that evaluate anesthesia not only by intraoperative stability but also by delirium, quality of recovery, persistent pain, cancer-related outcomes, and environmental impact.

Conclusion

Modern general anesthesia is safe, advanced, and highly adaptable. However, important controversies remain. The central grand challenges in Anesthetic Pharmacology are to predict the effects of drugs through PK/PD modeling, manage drug-drug interactions in multimodal regimens, integrate quantitative monitoring as pharmacological feedback, and understand molecular mechanisms such as ER stress, mitochondrial dysfunction, and neuroinflammation. The future of general anesthesia will be shaped by patient-centered, outcomes-oriented, and sustainable practice.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited and Reviewed by: Thomas Schricker, McGill University, Canada