This Week in Critical Care — Sep 30, 2026

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

Welcome to This Week in Critical Care. This week we're covering 10 notable papers spanning haemodynamic targets and vasoactive drug exposure, the control of breathing and sedation in acute respiratory distress syndrome, and how we staff and organise intensive care. Let's dive in.

Let's start with the question of how hard we should chase blood pressure, because three papers this week circle it from different angles. The biggest is the PRESSURE trial in JAMA, led by Inwald and colleagues, a pragmatic randomised trial across 23 paediatric intensive care units in three countries, enrolling 1,900 hypotensive, invasively ventilated children. Half were assigned a permissive mean arterial pressure target above the fifth percentile for age, and the rest received usual care with the target set by the attending physician. The permissive strategy worked as intended physiologically, cutting the median norepinephrine-equivalent total dose by about 38 percent, and it achieved a median pressure a few millimetres of mercury lower than usual care. But the composite primary outcome of mortality and duration of invasive ventilation at 30 days showed no significant difference between the groups, and six of the seven secondary outcomes were also similar. The one exception was paediatric intensive care unit length of stay in survivors, which was marginally shorter with the permissive target — a difference of a few hours on a median of around a week. Kidney replacement therapy and adverse events did not differ. So this is a neutral trial on its primary endpoint, and the honest reading is that a permissive target appears safe and drug-sparing in ventilated hypotensive children without demonstrating outcome benefit [2].

Two observational studies push the same theme toward the adult bedside. In the Journal of Critical Care, Özakın and colleagues studied 30 mechanically ventilated adults with septic shock, performing thenar near-infrared spectroscopy with vascular occlusion testing after initial resuscitation, once mean arterial pressure was at or above 65 with vasopressor support. Higher norepinephrine-equivalent dose was strongly and inversely associated with dynamic microvascular reactivity — a lower tissue oxygen saturation recovery slope and smaller hyperaemic response — and that association persisted after adjustment for severity and oxygen transport variables. Notably, central venous oxygen saturation did not track these dynamic microvascular indices at all, which is a caution about using it as a surrogate for microcirculatory reserve. With 30 patients and an observational design, this cannot tell us whether the vasopressor dose is causing the microvascular impairment or simply marking sicker vasculature [9]. Alongside that, in Critical Care Medicine, van Steenwijk and colleagues report on 440 adults with severe cardiogenic shock supported with venoarterial extracorporeal membrane oxygenation across six Dutch hospitals. Inotrope doses stayed largely unchanged after cannulation, even though organ perfusion no longer depends entirely on native cardiac output, and higher cumulative inotropic exposure was associated with a modestly increased risk of the composite weaning failure endpoint — death, ongoing extracorporeal support, or need for heart replacement therapy — which occurred in just over half of the cohort. Higher exposure also tracked with 30-day mortality, arrhythmias and slower lactate clearance. The authors are explicit that prospective work is needed before anyone concludes that inotropes can be tapered or stopped on support [6].

Turning to the lungs, two papers address respiratory drive and sedation in acute respiratory distress syndrome, and they point in opposite directions on pharmacology. In Chest, Cavayas and colleagues tested a genuinely novel idea: that serotonin acting on 5-HT3 receptors on pulmonary vagal C-fibres drives excessive respiratory effort, and that blocking those receptors might permit safer spontaneous breathing. This was a single-centre, single-blind, non-randomised crossover experiment in ten patients with acute respiratory distress syndrome ventilated for more than 48 hours on pressure support, with a placebo phase followed three hours later by intravenous ondansetron at 0.15 milligrams per kilogram. Mean inspiratory pressure-time product fell by roughly a fifth, respiratory rate fell by under two breaths per minute, minute ventilation and peak diaphragm electrical activity both declined, tidal volume did not change, carbon dioxide rose by about 3 millimetres of mercury, and the oxygenation ratio improved by about 23 points. It is a mechanistically clean proof of concept in nine analysable patients, unrandomised and unblinded to sequence, so it is hypothesis-generating rather than practice-shaping [7]. Against that, in Intensive Care Medicine, O'Gara and colleagues review volatile anaesthetic sedation in acute respiratory distress syndrome and deliver a sobering account: despite promising early data on inflammatory mediators and oxygenation, a recent multicentre randomised trial found that sevoflurane sedation in moderate-to-severe disease reduced ventilator-free days at day 28 and increased mortality compared with propofol. The authors conclude that routine clinical use of volatile sedation in this population is not recommended outside trials, while arguing that dose, agent, duration and sub-phenotype heterogeneity all deserve reconsideration in future studies [4]. Two narrative reviews round out the airway and antibiotic picture. De Jong and colleagues, in the Journal of Critical Care, trace intubation in the intensive care unit from a technical act to a physiologically complex procedure, noting that current evidence supports individualised respiratory preparation with noninvasive ventilation or high-flow nasal oxygen in selected hypoxaemic patients, head-up positioning, early vasopressor support rather than fluid loading alone, and videolaryngoscopy with a stylet or bougie to improve first-pass success [3]. And in the American Journal of Respiratory and Critical Care Medicine, Timsit and colleagues synthesise the antibiotic duration literature, concluding that randomised trials consistently show short courses of under seven days are noninferior to longer ones in bacteraemia, intra-abdominal infection and ventilator-associated pneumonia after effective source control, while flagging persistent uncertainty for Staphylococcus aureus bacteraemia, non-fermenting Gram-negative organisms, invasive candidiasis, immunocompromised patients and retained prosthetic material [1].

The third theme is organisation of care, where two papers disagree about what staffing buys you. In Chest, Smith and colleagues report a multicentre observational study of nearly 29,000 adults across 64 centres in the United States, Jordan and Saudi Arabia, examining intensive care unit pharmacist staffing. The median pharmacist-to-patient ratio was about one to 17, and each additional patient per pharmacist was associated with a very small increase in the odds of in-hospital death — a one percent relative increase, statistically significant but tiny per patient. More striking, patients who missed pharmacist-delivered comprehensive medication management on even one day of their stay had about 20 percent higher odds of dying in hospital, and ratios above roughly one to 15 carried about 10 percent higher odds than leaner ratios. This is observational and vulnerable to confounding by the kind of hospital that staffs pharmacists well, but it is the largest dataset of its kind [5]. Contrast that with a systematic review and meta-analysis in the Journal of Critical Care by Razvi and colleagues on enhanced nighttime or continuous on-site intensivist coverage. Across four studies of strict on-site night coverage totalling just over 13,000 patients, there was no significant reduction in hospital mortality, and intensive care unit mortality and both length-of-stay outcomes were also no different. Heterogeneity was substantial and the certainty of evidence was rated very low throughout, so the fair conclusion is that the existing studies cannot isolate any added benefit of night coverage beyond established daytime intensivist care — not that specialist expertise doesn't matter [8]. Finally, a single-centre retrospective study, also in the Journal of Critical Care, from Xia and colleagues, examined 97 patients with hypertensive intracerebral haemorrhage under multimodal neuromonitoring and found that when actual cerebral perfusion pressure sat within 5 millimetres of mercury of the individually calculated optimum, pressure reactivity, mean flow and oximetry indices all clustered near zero, indicating preserved autoregulation, with hypoperfusion appearing more damaging than hyperperfusion. It extends optimal perfusion pressure concepts from traumatic brain injury into haemorrhage, but it is retrospective and needs prospective validation [10].

If you only have time for one paper this week, make it the PRESSURE trial in JAMA [2]. It is the first large randomised test of permissive blood pressure targets in critically ill children, and while it does not show that less vasopressor improves outcomes, it reframes the question of whether higher pressure targets were ever buying anything.

Here is what this week's evidence adds up to in Critical Care. First, a permissive mean arterial pressure target in ventilated hypotensive children substantially reduced vasoactive exposure without improving mortality or ventilation duration, and without signals of harm — a neutral primary result that nonetheless narrows the rationale for aggressive pressure targets in this group. Second, two observational datasets, one in septic shock microcirculation and one in extracorporeal support for cardiogenic shock, independently associate higher vasoactive and inotropic exposure with worse physiology and worse outcomes, but neither can establish causation and both call for prospective trials. Third, in acute respiratory distress syndrome, volatile sedation now carries randomised evidence of harm and the reviewers advise against routine use outside trials, while 5-HT3 blockade to modulate respiratory drive is an intriguing but very early single-centre signal in nine patients. Fourth, short-course antibiotics after adequate source control are supported by consistent randomised noninferiority data, with important exceptions still unresolved. And fifth, on staffing, daily pharmacist medication management is associated with lower mortality in a large observational cohort, whereas pooled evidence for strict on-site night intensivist coverage shows no mortality benefit at very low certainty — a contrast worth holding loosely given the study designs involved.

That's your roundup for This Week in Critical Care. 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 if it saves you time, a quick rating in your podcast app helps other physicians find the show.

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