This Week in Critical Care — Aug 28, 2026

Welcome to This Week in Critical Care. This week we're covering 10 notable papers spanning ICU recovery and rehabilitation, the physiology of managing two failing organs at once, and the practical business of diagnosis and monitoring at the bedside. Let's dive in.

We start with the biggest randomized signal of the week, and it's a negative one. In Intensive Care Medicine, Chung and colleagues report the EVER trial, a multicentre open-label randomized study across five tertiary hospitals in South Korea, enrolling 169 adults with sepsis or acute respiratory failure expected to need invasive ventilation for at least 48 hours [1]. A structured six-step early mobilization programme was compared with usual care, and the intervention clearly did what it was designed to do on the process side: patients started moving earlier, at a median of about 30 hours versus 46, they received roughly eleven sessions instead of four, and they accumulated nearly three times as much mobilization time. Despite all that, functional status at ICU discharge, the primary outcome, was essentially identical between groups, and the long-term picture out to twelve months, covering quality of life, physical and mental component scores, post-traumatic stress symptoms and cognition, improved in both arms with no separation. There is one hypothesis-generating observation worth holding lightly: patients who actually reached step four or higher, meaning sit-to-stand or beyond, had meaningfully better function at discharge than matched usual-care patients. That is a comparison of those who could mobilize, not of those who were assigned to try, so it should not be read as proof that pushing harder works. The practical message is that simply protocolizing more sessions and more minutes did not translate into better function, and the field's attention should probably shift to who can achieve genuine upright activity and why.

Staying with recovery, Critical Care publishes a mixed-methods systematic review from Oh and colleagues on how patients and families psychologically cope with ICU delirium [9]. Ten qualitative studies from seven countries were synthesized, and notably not a single quantitative study met eligibility criteria, which is itself a statement about how thin this evidence base is. Patients described active sense-making, deliberate reality orientation, avoiding known delirium triggers, seeking social connection, and later reframing the experience to restore a sense of identity. Families were doing double duty, working to reorient and stimulate the patient while managing their own distress through information-seeking, avoidance, and positive reappraisal. The clinically useful point is that family caregivers function as a primary coping resource for the delirious patient, which argues for structuring family presence and family education as part of delirium care rather than treating visitors as incidental.

The second theme running through this week is the problem of protecting one organ without harming another. Intensive Care Medicine carries two reviews on ARDS in special populations that converge on the same conclusion. Robba and colleagues tackle ARDS in acute brain injury, a combination affecting up to a third of critically ill patients with brain injury and driving worse neurological outcomes [3]. The tension is direct: low tidal volumes and higher positive end-expiratory pressure protect the lung but permit carbon dioxide to rise, which can raise intracranial pressure and compromise cerebral perfusion, while tight carbon dioxide control pulls you away from conventional lung protection. Oxygen targets create the same bind, since both hypoxaemia and hyperoxaemia worsen secondary brain injury. Their recommendation is to move away from protocol and towards multimodal neuromonitoring, using intracranial pressure and brain tissue oxygenation to individualize ventilator settings. Lassola and colleagues cover ARDS in trauma with a similar philosophy, describing a multi-hit model in which the initial injury is compounded by transfusion, infection, or fat embolism [4]. There, management hinges on early risk stratification with clinical scores and multimodal imaging including lung ultrasound and computed tomography, alongside adequate analgesia, haemodynamic optimization, and timely surgery, because inadequate pain control and delayed source control directly worsen respiratory mechanics.

That organ cross-talk theme extends to the kidney. In Critical Care, Manca and colleagues review acute kidney injury as a systemic syndrome, reminding us that it affects around fifteen percent of hospitalized patients and close to half of ICU admissions, and that its consequences reach the lungs, brain, heart, liver and immune system through inflammation and metabolite accumulation [5]. Their most forward-looking argument concerns trial design: traditional endpoints like major adverse kidney events are hard to power in unenriched populations, and endpoints capturing remote organ complications may be more informative. Alongside that, Ostermann and colleagues offer a state-of-the-art review on beta-blockade in critical illness in Intensive Care Medicine [2]. Their reading of the evidence is deliberately unenthusiastic: benefit is established in acute myocardial infarction, tachyarrhythmias, hypertensive emergencies, thyroid storm and variceal bleeding prophylaxis, but evidence in septic shock, traumatic brain injury, acute heart failure and burns is limited or conflicting, with real risks of blunting compensatory responses, dropping cardiac output, and impairing organ perfusion. Whether you are continuing chronic therapy, restarting after withdrawal, or initiating de novo during acute illness matters, and there is currently no reliable marker to identify who benefits.

The third theme is diagnosis and measurement, and here we have two studies that change what you might do tomorrow. In the Journal of Critical Care, Li and colleagues prospectively compared oesophageal, bladder and tympanic thermometry against intravascular temperature in 22 ICU patients with over 800 paired measurements [10]. Oesophageal readings tracked intravascular temperature almost perfectly, with a bias of about a tenth of a degree. Bladder temperature lagged badly during targeted temperature management, agreeing with the direction of change only about half the time, and that lag worsened as cardiac output fell, which is precisely the shocked, cooled patient in whom you most want accuracy. Tympanic measurements were wide and erratic in a way unrelated to haemodynamics, and the authors conclude they are unreliable for continuous monitoring. If you are cooling or rewarming a patient in shock, this argues for oesophageal probes as the default surrogate.

Also in the Journal of Critical Care, Mokart and colleagues analysed 708 patients with haematological malignancies admitted with hypoxaemic respiratory failure, asking whether the pattern of immune dysfunction predicts the cause [7]. Bacterial pneumonia accounted for about forty percent of cases, viral infection just under a fifth, and invasive fungal infection about one in six, while roughly one in eight cases remained undiagnosed. Phagocytic dysfunction roughly doubled the odds of infiltrative causes and raised bacterial risk; humoral and cellular dysfunction both pointed towards viral disease, with cellular dysfunction more than doubling the odds of Pneumocystis pneumonia. Neutropenia was strongly associated with Gram-negative infection, and corticosteroid exposure roughly doubled the likelihood that no diagnosis was ever reached. Unsupervised clustering separated three phenotypes with mortality ranging from just under forty percent in chronic lymphoid disease to above fifty percent in the myeloid infiltration group. The practical implication is that the immune deficit profile should shape your empiric workup and therapy rather than a one-size-fits-all bronchoscopy-and-broad-spectrum approach.

Two further papers round out the week. In the Journal of Critical Care, Cruciger and colleagues analysed over 61,000 severely injured ICU patients in the German trauma registry and found that sepsis developed in about six percent, but carried outsized consequences: median ICU stay of nineteen days versus three, multiple organ failure in around seven in ten septic patients compared with fifteen percent of the rest, and sepsis-associated mortality of about a third [8]. The dominant focus was pulmonary in roughly two thirds of cases, with abdominal sepsis carrying the highest case fatality at around thirty-eight percent, and early ventilation at ICU admission showed the strongest association with subsequent sepsis, nearly tripling the odds. And in Critical Care, Neuberger and Hartl offer a methodological post-mortem on protein dosing, cataloguing the biases, immortal time, confounding by indication, wash-in effects, collider bias, that plausibly manufactured the apparent mortality benefit of high protein intake in older observational work, a benefit that EFFORT Protein, PRECISe and TARGET Protein have all failed to confirm [6].

If you only have time for one paper this week, make it the EVER trial in Intensive Care Medicine [1]. It is the only randomized outcome trial in the batch, and it directly tests a practice that many units have already protocolized on the assumption that it works.

Here are the key takeaways from this week in Critical Care. First, a protocolized early mobilization programme delivered more sessions and more minutes but did not improve function at ICU discharge or at twelve months, so intensity alone is not the lever. Second, in patients with both ARDS and brain injury, or ARDS after trauma, abandon rigid protocols in favour of physiology-guided, monitored, individualized ventilation. Third, beta-blockade in critical illness remains context-dependent, with genuine benefit outside sepsis and traumatic brain injury and genuine hazard within them. Fourth, if you are running targeted temperature management in a shocked patient, use an oesophageal probe and distrust bladder and tympanic readings. Fifth, in haematology patients with respiratory failure, let the specific immune deficit steer your differential. And finally, when observational nutrition data and randomized trials disagree, the trials are usually right.

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 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.