This week in Critical Care

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 sepsis resuscitation and how we measure it, analgesia and sedation on the ventilator, and cardiogenic shock and resuscitation science, with a closing look at resistant pneumonia and citrate anticoagulation. Let's dive in.

Sepsis dominates the week, and the common thread is that several plausible interventions and measurement tools look less certain under close scrutiny. In Chest, Chen and colleagues report the ADJUST trial, a multicentre, double-blind, placebo-controlled test of ulinastatin, the urinary trypsin inhibitor that has long carried a reputation for survival benefit in sepsis [1]. Among 335 adults with Sepsis-3 sepsis, ulinastatin given every eight hours for up to ten days did not reduce 28-day mortality, which sat at just under a quarter in both arms. Secondary outcomes were similar, and adverse events were more common with ulinastatin, at roughly 57 percent of treated patients versus 43 percent on placebo. The authors conclude that routine use is not supported, and for a drug used in some health systems on the strength of smaller studies, this is a clear negative result, albeit from a single moderately sized trial. In Critical Care, Melo and colleagues tackle fluid dosing in early septic shock with a different idea: using bedside signs of fluid intolerance as safety brakes [2]. In three Chilean ICUs, ventilated patients who were still hypoperfused and fluid responsive were randomized to standard 500 millilitre boluses or to a traffic-light approach in which lung ultrasound, venous congestion scoring, echocardiographic filling pressure and central venous pressure set each bolus at 500 millilitres, 250 millilitres, or none. Oxygenation improved by about 78 points more in the intervention arm, with no difference in lactate or capillary refill time. But the trial stopped at 36 of a planned 62 patients because enrolment was no longer feasible, and cumulative fluid exposure ended up similar between groups. The authors themselves describe the oxygenation effect as uncertain, so this is a feasibility signal for a larger trial rather than evidence to change resuscitation practice.

Staying with fluids, Zampieri and colleagues, in Annals of Intensive Care, return to the BaSICS trial of saline versus Plasma-Lyte across 75 Brazilian ICUs [3]. In about 3,500 patients with chloride measurements, saline raised serum chloride by roughly 2 millimoles per litre more than Plasma-Lyte by day two, with a larger gap in patients with poorer baseline kidney function, about 3 millimoles at an eGFR of 30 versus under 2 at 90. The striking finding is how little the fluid itself mattered: it explained under two percent of the modeled variation in serum chloride, while baseline chloride and unexplained patient-level differences accounted for most of it. That reframes hyperchloraemia as largely a property of the patient rather than the bag, and it cautions against reading observational chloride and outcome associations as fluid effects. Rounding out the sepsis theme, Leland and colleagues, in the Journal of Critical Care, rescored two randomized sepsis trials, ACTS and TRPSS, using the updated SOFA-2 score [4]. SOFA-2 produced lower total scores, with lower respiratory and cardiovascular and higher coagulation subscores. For ACTS, the primary result remained non-significant under SOFA-2 but crossed into significance in an exploratory analysis that added approximated urine output, while in TRPSS the conclusion held but the treatment effect shrank. The practical message is methodological: which SOFA version a trial uses can shift its interpretation, and the authors call for transparent reporting and prospective validation.

Our second theme is comfort on the ventilator, where two papers test newer agents against established ones. In Intensive Care Medicine, Lu and colleagues report CO-ROAM, a 19-centre Chinese trial comparing oliceridine, a G protein-biased opioid, with remifentanil in about 300 mechanically ventilated adults [5]. Oliceridine met non-inferiority for time in the target analgesic range, which exceeded 90 percent in both arms. It was associated with fewer mild-to-moderate adverse drug reactions, no respiratory depression compared with just under 5 percent on remifentanil, no post-extubation respiratory depression compared with about 8 percent, and lower analgesic and sedation costs. Ventilation duration, ICU stay and 28-day mortality did not differ. This is the first solid randomized evidence for oliceridine in critically ill patients, though it comes from one country and one comparator, and the respiratory safety advantage rests on small event numbers. Meanwhile, in the Journal of Critical Care, Nakahara and colleagues pooled six randomized trials of remimazolam versus propofol in 705 ventilated patients [6]. They found no difference in time within the target sedation range, and no detectable difference in mortality, ventilation duration, hypotension or delirium, although several estimates were imprecise. The conclusion is that remimazolam probably sedates about as well as propofol, with its safety and outcome profile still unsettled pending larger trials.

The third theme moves to cardiogenic shock and resuscitation guidance. In The Lancet, Motovska and colleagues report DAPT-SHOCK-AMI, described as the first randomized trial of P2Y12 inhibitor strategies specifically in myocardial infarction complicated by cardiogenic shock [7]. Across 29 hospitals in five European countries, 605 patients received intravenous cangrelor or crushed ticagrelor. In the subset with platelet testing, every cangrelor patient achieved effective platelet inhibition at the end of primary PCI, compared with about one in five ticagrelor patients. Yet the 30-day composite of death, myocardial infarction or ischaemic stroke, which occurred in roughly two in five patients in each arm, did not establish non-inferiority, and major bleeding was similar at around 5 to 6 percent. So a clear pharmacological advantage did not translate into demonstrable clinical non-inferiority, and the authors frame the results as groundwork for future trials rather than a verdict. Alongside it, Resuscitation publishes the 2026 International Liaison Committee on Resuscitation summary from Bray and colleagues [8]. Six task forces, spanning basic, advanced, paediatric and neonatal life support, education and teams, and first aid, graded the newest evidence and issued consensus treatment recommendations, with explicit evidence-to-decision reasoning and a list of priority research gaps. It is the reference document that will feed national guideline updates over the coming cycle.

Finally, two papers address the infection and renal support problems that fill daily ICU rounds. In Critical Care, Mokrani and colleagues pooled 119 patients with difficult-to-treat Stenotrophomonas maltophilia ICU-acquired pneumonia from 31 centres in eight countries [9]. Thirty-day mortality was close to 38 percent. Tigecycline was associated with roughly double the mortality risk, while aztreonam with ceftazidime-avibactam and colistin showed non-significant trends toward lower mortality and cefiderocol appeared neutral. Neither combination therapy nor courses longer than eight days were linked to outcomes. As a retrospective, underpowered cohort, these are hypothesis-generating findings, though the tigecycline signal adds weight to existing caution about that agent in pneumonia. Also in Critical Care, Teh and colleagues offer a perspective on citrate accumulation during regional citrate anticoagulation for continuous renal replacement therapy [10]. They argue that the total-to-ionized calcium ratio is a sensitive but nonspecific marker, confounded by albumin, acid-base status and calcium supplementation, and propose a stepwise, dynamic assessment combining ratio trends with lactate, systemic ionized calcium and acid-base changes. Their key point is that an elevated ratio alone need not end citrate use, since prescription adjustments can often reduce citrate burden. This is expert opinion awaiting validation against direct plasma citrate measurement.

If you only have time for one paper this week, make it DAPT-SHOCK-AMI in The Lancet [7]. It is the first randomized evidence on antiplatelet strategy in infarct-related cardiogenic shock, and it reopens the question of whether faster, more reliable platelet inhibition actually changes outcomes in this sickest group.

That's your roundup for This Week in Critical Care. The full transcript and references are available on the episode page. 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.