This Week in Anesthesiology — Aug 17, 2026

Welcome to This Week in Anesthesiology. This week we're covering 10 notable papers spanning the long-running debate over intravenous versus volatile anaesthesia, intraoperative haemodynamic and respiratory monitoring, and perioperative risk stratification and quality of care. Let's dive in.

The biggest story this week is the apparent closing of the propofol-versus-volatile question. In JAMA, Jhanji and colleagues report the VITAL trial, a pragmatic, open-label randomised trial across 49 United Kingdom National Health Service hospitals that enrolled 2,508 patients aged 50 or older for elective major noncardiac surgery, randomising them to maintenance with a propofol infusion or a volatile agent, with all other care left to clinician discretion [1]. The primary outcome, days alive and at home at 30 days, was essentially identical — about 22 and a half days in each arm, with an incidence rate ratio of 1.00. There was no difference at 90 days, no difference in mortality out to six months, no difference in quality of recovery at day three, and no difference in major complications, which occurred in roughly one in eight patients overall. Delirium was the same in both groups, with about 88 percent of patients showing no delirium at day three. What did differ were minor patient-reported symptoms: less thirst, less hoarseness, and less nausea and vomiting with total intravenous anaesthesia. The two cases of certain or probable unintentional awareness both occurred in the intravenous arm — a small absolute number, but a reminder that depth monitoring matters when you switch off the volatile agent.

Anaesthesia published two papers that bracket the VITAL result nicely. Frempong and colleagues ran the within-trial economic evaluation alongside VITAL, from the National Health Service and personal social services perspective over six months [6]. Costs and quality-adjusted life years were statistically indistinguishable — an incremental cost difference of about 145 pounds in favour of intravenous anaesthesia with confidence intervals spanning well over a thousand pounds either way, and an incremental quality-adjusted life year difference of essentially zero. The probability of cost-effectiveness for total intravenous anaesthesia hovered around 56 to 57 percent at standard willingness-to-pay thresholds, which is another way of saying a coin toss. Their conclusion is that anaesthetic choice should continue to be guided by patient factors, clinician expertise and local context. And in the same journal, Mustafa and colleagues published a systematic review and meta-analysis of eleven randomised trials, nearly 7,900 patients undergoing curative cancer surgery, comparing propofol-based with volatile-based anaesthesia [3]. There was no significant difference in overall survival, recurrence-free survival, mortality, or recurrence, with minimal heterogeneity and what the authors grade as high-certainty evidence for mortality. Trial sequential analysis suggests the mortality finding is robust, while recurrence remains inconclusive. Taken together with VITAL, the honest reading is that for both short-term recovery and long-term oncological outcomes, this choice is not the lever we hoped it was.

Turning to monitoring, two papers in Anesthesiology ask whether sophisticated technology beats simpler alternatives. Wu and colleagues randomised 100 adults having major noncardiac surgery with invasive arterial monitoring either to Hypotension Prediction Index-guided management, triggered at an index of 85 or above, or to a proactive strategy of treating whenever mean arterial pressure fell to 73 millimetres of mercury or below, both following the same haemodynamic protocol [2]. The primary endpoint, time-weighted average hypotension below a mean pressure of 65, was numerically lower in the prediction-index group but the difference was not statistically significant, and neither was the area under the hypotension threshold. Hypertension burden, noradrenaline dose, length of stay and 30-day mortality were all similar. The authors are careful: the algorithm was not superior to simply treating earlier at a higher pressure target, and because the trial was not powered for equivalence or non-inferiority, this is a failure to demonstrate superiority rather than proof that the two are the same. Still, it supports the argument that much of the apparent benefit reported in earlier open-label prediction-index trials may simply reflect intervening sooner at a higher threshold — something you can do today without proprietary software. Also in Anesthesiology, Zhang and colleagues offer a proof-of-concept study in 30 orally intubated adults on pressure support ventilation, simultaneously recording oesophageal pressure and endotracheal tube cuff pressure across a downward pressure support titration [10]. Over 840 breaths, the tidal swing in cuff pressure tracked oesophageal pressure closely, with a mean bias of a quarter of a centimetre of water, though limits of agreement of roughly plus or minus five and a half centimetres mean it cannot replace oesophageal manometry for precise quantification. What it did do well was discriminate extremes: areas under the curve of 0.93 to 0.95 for detecting low and high inspiratory effort, and around 0.88 to 0.89 for identifying high transpulmonary driving pressure and high mechanical power. If this replicates, the cuff you already inflate could become a continuous, equipment-free screen for injurious patient effort.

The third theme is risk stratification and quality of care. In the British Journal of Anaesthesia, Weiss and colleagues retrospectively studied 3,353 patients aged 70 or older having elective noncardiac, noncranial surgery, phenotyping them with two bedside tools — the Mini-Cog and the Clinical Frailty Scale — into four groups [8]. Overall delirium incidence, assessed with the 4A's Test in recovery and on the first two postoperative days, was 7.5 percent, but it rose from just over 4 percent in cognitively intact, non-frail patients to 24 percent in those who were both cognitively impaired and frail, who carried nearly five times the adjusted odds of delirium. Notably, cognitive impairment alone conferred more risk than frailty alone, and median postoperative length of stay lengthened from about two and a half to four and a half days across phenotypes. Two short screening tests in preassessment clinic can therefore identify the 8 percent of older patients who most warrant delirium prevention bundles. On the pharmacological side, Choi and colleagues in Anesthesiology analysed a large electronic health record network, propensity-matching surgical patients with type 2 diabetes taking a glucagon-like peptide-1 receptor agonist against those on metformin, sodium-glucose cotransporter-2 inhibitors, or dipeptidyl peptidase-4 inhibitors, using data predating the 2023 American Society of Anesthesiologists guidance to hold these drugs [4]. Compared with metformin, receptor agonist use was associated with roughly half the 14-day postoperative mortality — about 1 percent versus 2.2 percent — and compared with dipeptidyl peptidase-4 inhibitors, with lower mortality and about half the rate of bacterial pneumonia. Critically, there was no increase in aspiration pneumonitis against any comparator. This is observational and confounding by indication cuts both ways, but it adds to the argument that blanket preoperative withholding may not be risk-free. And Francke and colleagues, also in Anesthesiology, examined nearly 290,000 caesarean deliveries in the Multicenter Perioperative Outcomes Group database and found adherence to guideline-supported best practices was highly uneven [7]. Avoidance of general anaesthesia and maintenance of systolic pressure above 90 after spinal anaesthesia both exceeded 96 percent, and timely antibiotics reached 86 percent, but only about 57 percent of cases included active normothermia measures and only about 55 percent used a post-spinal vasopressor infusion. Adherence was worse in sicker patients, in evening and overnight cases, and at hospitals without an obstetric anaesthesia fellowship — a clear map for local quality improvement.

Two final papers address analgesia and airway safety. Desai and colleagues in Anaesthesia updated the PROSPECT recommendations for major oncological breast surgery, drawing on 176 randomised trials and eight systematic reviews [5]. Paracetamol, a nonsteroidal anti-inflammatory or cyclooxygenase-2 inhibitor, and dexamethasone remain the backbone; gabapentin is no longer recommended because of problematic adverse effects; and a range of single-shot regional techniques — erector spinae plane, interpectoral and pectoserratus, superficial or deep serratus anterior plane, thoracic paravertebral, and local infiltration — are recommended and considered equivalent to one another. Continuous thoracic paravertebral catheters are no longer advised, since single-shot blocks cover the worst of the acute pain trajectory. Perineural adjuncts and complementary therapies are not recommended, while pre- and postoperative physiotherapy is. Finally, Olsen and colleagues in Anesthesiology propose a mechanical approach to aspiration prophylaxis: alginates, the buoyant gel raft agents used for decades in reflux disease, as an ingestible physical barrier to regurgitation at induction [9]. This is a hypothesis-generating argument rather than trial data, and it should be read as a proposal awaiting testing.

If you only have time for one paper this week, make it the VITAL trial in JAMA [1]. It is the largest pragmatic randomised comparison of intravenous versus volatile maintenance in older patients having major surgery, and its neutral result — reinforced by the parallel economic analysis and the oncology meta-analysis — should settle a debate that has consumed a great deal of departmental energy.

Here are the key takeaways from this week in Anesthesiology. First, in older adults having major noncardiac surgery, propofol-based and volatile-based maintenance produce equivalent recovery, mortality, complications and cost, and equivalent long-term cancer outcomes, so choose on patient factors, expertise and context — but keep depth-of-anaesthesia vigilance high when using total intravenous anaesthesia. Second, the Hypotension Prediction Index was not superior to simply treating when mean arterial pressure fell to 73 or below, so a proactive higher pressure target remains a reasonable low-technology strategy. Third, two brief bedside tests, the Mini-Cog and the Clinical Frailty Scale, stratify postoperative delirium risk from 4 percent up to 24 percent and should trigger targeted prevention. Fourth, observational data suggest preoperative glucagon-like peptide-1 receptor agonist use was associated with lower short-term mortality and no excess aspiration pneumonitis, which argues for prospective confirmation rather than reflexive withholding. And fifth, in caesarean delivery, normothermia measures and post-spinal vasopressor infusions are used in only about half of cases and are the most obvious targets for quality improvement, especially out of hours.

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 a quick rating on Apple Podcasts or Spotify helps other physicians discover the show.

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