This Week in Anesthesiology — Sep 21, 2026

Welcome to This Week in Anesthesiology. This week we're covering 10 notable papers spanning perioperative aspiration risk and preoperative preparation, regional anaesthesia and the problem of rebound pain, and a cluster of studies on monitoring, technology and the wellbeing of the people doing the work. Let's dive in.

We start with what is probably the most consequential paper of the week, in Anaesthesia, where Potter and colleagues report a national prospective multicentre cohort study of nearly 47,000 adults across 119 United Kingdom sites having procedures under the care of an anaesthetist [1]. Patients were screened for use of a glucagon-like peptide-1 receptor agonist in the preceding three months. About one in 36 patients was taking one of these drugs, and perioperative management was all over the map — variable cessation practices, variable anaesthetic technique, variable airway management. The headline number is the one that will change conversations in your pre-op clinic: pulmonary aspiration or regurgitation occurred in roughly one in 71 patients taking a GLP-1 receptor agonist, compared with about one in 800 in everyone else. That is more than a tenfold difference in odds, and it persisted despite clinicians clearly attempting risk mitigation with a range of airway strategies. Importantly, the events clustered at emergence, not at induction — which is exactly where our attention tends to relax. This is observational data with the usual confounding caveats, and these patients differ systematically from non-users in body habitus and diabetes status. But the signal is large, the denominator is enormous, and the practical message is that a documented GLP-1 receptor agonist should prompt a deliberate plan that extends through extubation and recovery, not just a decision about whether to use a rapid sequence induction.

Sitting alongside that, the British Journal of Anaesthesia publishes an editorial from Camilleri Podesta and Rüggeberg arguing that preoperative clear liquid fasting durations remain unacceptably long worldwide and represent one of the most common and least recognised sources of avoidable perioperative harm [10]. Their argument is about implementation rather than evidence: we already have consensus recommending more liberal clear fluid protocols, and what is missing is routine measurement and audit of actual fasting times against those recommendations. Their framing is worth repeating — failing to implement a strategy that prevents known harm is not a passive omission, it is an active decision. There is a nice tension between these two papers. One tells us a specific population is aspirating far more than we thought; the other tells us that most patients are being starved far longer than they need to be. Both point to the same corrective, which is measuring what actually happens to your patients rather than assuming the protocol is being followed. That thread continues in a consensus statement in Anaesthesia from Bailey and colleagues, produced jointly by the Association of Anaesthetists, the British Association for Day Surgery and the Association of Paediatric Anaesthetists [2]. With day surgery rates in the United Kingdom now above 85 percent but still varying widely between hospitals, a two-round modified Delphi process generated recommendations covering patient selection, pre-assessment and optimisation, the handling of frailty, obesity and obstructive sleep apnoea, urgent and emergency day-case surgery, anaesthetic technique and discharge processes, in both adults and children. It is a pragmatic organisational document, and its real value is as a benchmark against which to audit your own unit.

Our second theme is rebound pain after brachial plexus block — that transient but severe pain as the block wears off — and two journals tackled it from different directions this week. In the European Journal of Anaesthesiology, Ma and colleagues report a Bayesian network meta-analysis of 22 randomised trials and just under 1,700 patients, comparing dexamethasone, esketamine, dexmedetomidine, liposomal bupivacaine, continuous nerve block and combined multi-approach blocks [4]. Compared with placebo, dexamethasone, dexmedetomidine and continuous catheter techniques all significantly reduced the incidence of rebound pain, with dexamethasone and continuous block also lowering pain scores at 12 hours. Continuous nerve block ranked highest across outcomes, though the authors are appropriately restrained — these are ranking probabilities, the certainty of evidence is limited, and they explicitly frame the work as hypothesis-generating rather than definitive. Then in Regional Anesthesia and Pain Medicine, Zhang and colleagues take a very different approach with a single-centre, sham-controlled randomised trial of transcutaneous auricular vagus nerve stimulation in 120 adults having arthroscopic shoulder surgery under single-shot interscalene block [5]. Active stimulation applied perioperatively and continued for two postoperative days roughly halved rebound pain within 24 hours — one in five patients in the stimulation group versus just over four in ten with sham — with a number needed to treat of around four. Resting and movement pain scores were modestly lower, quality of recovery and sleep quality were better, and there were no stimulation-related adverse events. Worth noting that cumulative sufentanil consumption did not differ significantly between groups, so this is an experience-and-comfort benefit rather than a demonstrated opioid-sparing one, and it is a single centre using a non-standard intervention. Taken together, the practical message is that if you place a single-shot brachial plexus block in a patient at high risk of rebound pain, doing nothing further is no longer a neutral choice — a catheter, perioperative dexamethasone, or possibly adjuncts like this deserve active consideration.

Third, technology and monitoring, where this week delivers one promising feasibility study and one useful negative trial. In Anesthesiology, Hoshino and colleagues studied 60 mechanically ventilated adults admitted to intensive care after major surgery at high risk of pulmonary complications, using electrical impedance tomography to classify regional ventilation patterns [6]. Thirty patients with heterogeneous ventilation received individualised positive end-expiratory pressure titrated to the dorsal fraction of ventilation, and phenotype conversion to a homogeneous pattern was achieved immediately in 29 of 30, with 28 of 30 still in target four hours later. In the ventral-dominant group, PEEP went up from 5 to 9 centimetres of water with meaningful gains in compliance and oxygenation; in the dorsal-dominant group, PEEP came down from 7 to 3, improving ventral compliance without changing global compliance. Individualised PEEP also lowered mean pulmonary artery pressure and raised cardiac index. An exploratory comparison suggested fewer severe pulmonary complications against a historical cohort — 25 percent versus 41 percent — but that difference was not statistically significant, and this was a non-randomised single-centre feasibility study. The concept that PEEP should sometimes be lowered, not raised, is the part worth carrying home. Contrast that with a prospective randomised trial in Regional Anesthesia and Pain Medicine from Kim and colleagues, who assigned 80 ultrasound novices — medical students and anaesthesiology residents — to conventional B-mode training or to training with a real-time artificial intelligence nerve tracking system, then tested them with the artificial intelligence switched off [8]. First-attempt independent identification of the median nerve and the supraclavicular brachial plexus did not differ significantly between groups. The artificial intelligence group reported slightly higher confidence identifying the plexus, which is arguably the least reassuring possible result — more confidence without more competence. Brief exposure to automated nerve tracking did not teach novices to see nerves on their own.

Also in Anesthesiology, Nagy and colleagues randomised the hand and muscle monitored in 44 patients receiving rocuronium and sugammadex, recording electromyography simultaneously from the adductor pollicis and the abductor digiti minimi [7]. Overall there were no significant differences in onset time or in recovery to a train-of-four ratio of 0.9 — broadly reassuring for the many clinicians who monitor the abductor digiti minimi because the electrodes sit more reliably. The caveat is handedness: on the dominant right hand, the abductor digiti minimi recovered about a minute and a half earlier than the adductor pollicis, roughly a quarter faster. On the non-dominant arm the two muscles behaved the same. So monitor the non-dominant arm when you can, and be aware you may be declaring adequate reversal slightly early otherwise. Rounding out the section, a narrative review in Anesthesiology from Chabanne and Futier on anaesthesia for mechanical thrombectomy in large vessel occlusion stroke concludes that current evidence does not show superior functional outcomes with procedural sedation over protocol-driven general anaesthesia, that general anaesthesia is associated with higher reperfusion rates, and that the decisive variables are strict haemodynamic control — above all avoiding hypotension — and ventilatory management [3]. Choose your technique on patient factors, procedural complexity and institutional expertise, and then defend the blood pressure.

Finally, and it deserves its own moment, Anesthesiology publishes a contemporary survey from Margolis and colleagues on the impact of perioperative catastrophes, adapted from the landmark 2012 instrument and sent to United States members of the American Society of Anesthesiologists [9]. Among 1,725 respondents — a response rate of only about seven percent, so selection bias is real — over 94 percent had been involved in at least one perioperative catastrophe. After a memorable event, roughly nine in ten reported psychosocial symptoms, and about seven in ten felt their ability to deliver anaesthesia care was impaired in the first four hours. Nearly one in five said they never fully recovered emotionally, essentially unchanged from 2012, and nearly a quarter of respondents had considered leaving anaesthesiology because of such an event. Informal peer support was common; formal debriefing and institutionally supported time off were not. Thirteen years of cultural awareness has not moved the recovery numbers, and the gap is in structural relief, not sentiment.

If you only have time for one paper this week, make it the national GLP-1 receptor agonist cohort study in Anaesthesia [1]. It quantifies, at national scale, a risk many of us have been managing on intuition, and it tells you where in the anaesthetic that risk actually materialises.

Here are the key takeaways from this week in Anesthesiology. Patients on GLP-1 receptor agonists aspirated or regurgitated at more than ten times the rate of other patients, most often at emergence, so extend your risk plan through extubation. Prolonged clear fluid fasting remains widespread avoidable harm, and fixing it requires measuring actual fasting times, not reissuing the guideline. For patients at high risk of rebound pain after a brachial plexus block, continuous catheter techniques ranked best in network meta-analysis, and auricular vagus nerve stimulation roughly halved rebound pain in a single-centre trial. Individualised PEEP guided by electrical impedance tomography reliably homogenised postoperative ventilation, sometimes by lowering PEEP rather than raising it, though the outcome benefit remains unproven. Monitoring the abductor digiti minimi is equivalent to the adductor pollicis on the non-dominant arm but recovers faster on the dominant one. And brief artificial intelligence-assisted ultrasound training raised novices' confidence without improving their independent nerve identification.

That's your roundup for This Week in Anesthesiology. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And think of one colleague — in any specialty — who never has time to keep up with the literature. Tell them about AudioScholar: a free ten-minute weekly for every specialty, to listen to in any podcast app, or to read at audioscholar dot C C.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.