TBS 2026: Best Intensive Care Medicine Papers

This is the third of our paper reviews from the TBS conference in Zermatt 2026. In truth we did not get to present these papers on the day as we sort of ran out of time having too much fun discussing the PHEM/EM papers (which are arguably the focus of the conference). However, we had prepared a few papers to discuss and it seems a waste to not share them with you here. To recap, this was the opening session of the conference with Brian Burns (Sydney HEMS) and Iain Beardsell. You can find links to the EM and PHEM papers in the references.

Precision immunotherapy for sepsis: the ImmunoSep trial

Sepsis trials have historically been limited by biological heterogeneity, with repeated failures of unselected immunomodulatory therapies. The ImmunoSep trial addressed this by testing a precision approach, targeting treatment according to immune phenotype rather than syndrome alone.

This multicentre, double-blind, randomised controlled trial enrolled adults with sepsis due to pneumonia or bacteraemia. Patients were stratified using ferritin concentration and monocyte HLA-DR expression into macrophage activation–like syndrome or sepsis-induced immunoparalysis, and randomised to targeted immunotherapy (anakinra or interferon gamma) or placebo in addition to standard care.

Main results

A ≥1.4-point reduction in SOFA score by day 9 occurred in 35.1% of the precision immunotherapy group versus 17.9% in the placebo group
There was no statistically significant difference in 28-day mortality
Serious adverse events were common in both groups, with higher rates of anaemia and haemorrhage in the intervention arms

This trial demonstrates that immune phenotyping can identify a subgroup of septic patients in whom organ dysfunction improves with targeted therapy. However, the absence of a mortality signal, the complexity of biomarker-driven enrolment, and limited generalisability mean this approach remains experimental. For emergency physicians, this study signals a future direction rather than an imminent practice change.

Cryopreserved versus liquid-stored platelets: the CLIP-II trial

Limited platelet shelf life presents major logistical challenges, particularly for smaller and remote hospitals. Cryopreserved platelets offer a potential solution, but their clinical effectiveness remains uncertain. The CLIP-II trial rigorously evaluated this trade-off.

This double-blind, multicentre, randomised non-inferiority trial compared cryopreserved platelets with conventional liquid-stored platelets in cardiac surgery patients at high risk of bleeding. The primary outcome was chest drain blood loss in the first 24 hours after ICU admission.

Main results

Non-inferiority for the primary outcome was not demonstrated
Cryopreserved platelets were associated with greater intraoperative and total perioperative blood loss
Patients receiving cryopreserved platelets required more transfusion of red cells, plasma, and cryoprecipitate
Ventilation duration, ICU length of stay, and hospital length of stay were longer in the cryopreserved group

This well-conducted trial indicates that improved availability does not compensate for reduced haemostatic effectiveness. For trauma and emergency systems considering cryopreserved products, these findings caution against premature adoption without clearer evidence of equivalence.

Early blood pressure targets in acute spinal cord injury

Augmented mean arterial pressure targets following acute spinal cord injury are widely recommended despite limited supporting evidence. This randomised trial directly tested whether higher targets improve neurological outcomes.

Patients with acute traumatic spinal cord injury were randomised to augmented MAP targets (85–90 mm Hg) or conventional targets (65–70 mm Hg) for up to seven days. Neurological outcomes were assessed at six months.

Main results

No difference in motor or sensory recovery at six months between groups
The augmented MAP group had higher rates of respiratory complications
Duration of mechanical ventilation and non-cardiovascular organ dysfunction were greater in the augmented group
Mortality did not differ between groups

Although underpowered, and a lot of patients were lost to follow up, this trial challenges a long-standing recommendation and suggests potential harm without clear benefit. For emergency and critical care clinicians, it raises important questions about the routine pursuit of aggressive haemodynamic targets in the early phase of care. However, patients were recruited on the ICU and not in the very early phases of care, so we don’t really know how this applies to PHEM/EM practice where there may be even more autonomic dysregulation. For that reason I’m not changing my PHEM/EM practice just yet.

Targeted tissue perfusion versus pressure-guided care in septic shock

Blood pressure targets dominate sepsis resuscitation protocols, yet may not accurately reflect tissue perfusion. The TARTARE-2S trial evaluated whether a perfusion-focused strategy allowing lower MAP targets could improve outcomes.

This multicentre, randomised, open-label trial compared a targeted tissue perfusion strategy, using capillary refill time, skin temperature, lactate, and lower MAP thresholds, with standard MAP-guided care in ICU patients with septic shock.

Main results

No difference in the primary outcome of days alive with normal lactate and without vasopressors at 30 days
No significant differences in mortality or organ support–free days
Lower MAPs were safely tolerated in the tissue perfusion group without excess adverse events

While not practice-changing, this trial supports the safety of permissive lower MAPs when combined with close clinical assessment. It reinforces the concept that shock management should prioritise perfusion rather than isolated pressure targets. It may also have use in lower resource settings. Also note that there is a recent RCT of use of art lines in ICU patients that showed no benefit, so combined we might perhaps reduce invasive procedures and vasopressors without harm to patients. Scott Weingart has a great review on the everdac trial here.

Blood biomarkers for prognostication after cardiac arrest: the TTM2 biobank study

Neurological prognostication after cardiac arrest requires a multimodal approach, yet current blood biomarkers have important limitations. This large prospective study embedded within the TTM2 trial compared established and emerging biomarkers.

Serial blood samples from over 800 patients were analysed for neuron-specific enolase, S100, glial fibrillary acidic protein, and neurofilament light chain, with functional outcome assessed at six months.

Main results

Neurofilament light chain demonstrated the highest prognostic accuracy at 24, 48, and 72 hours (AUROC up to 0.93)
Neurofilament light chain outperformed neuron-specific enolase and S100 at all time points beyond admission
Performance was consistent across centres and unaffected by haemolysis

This study provides the strongest evidence to date supporting neurofilament light chain as a prognostic biomarker after cardiac arrest. While it should not be used in isolation, it represents a meaningful advance in post-resuscitation care.

Final thoughts

These papers challenge the assumption that more intervention or higher targets necessarily lead to better outcomes. Across sepsis, neurological injury, transfusion, and prognostication, they emphasise the importance of physiological understanding, patient selection, and critical appraisal of long-standing practices. For clinicians working at the emergency/ICU interface, the question will be how to balance individual patient strategies against wider institutional/national/guidelines. No easy answers there, and in fact it’s always been like that. We just have more data to prove it these days.

Also do visit Rob McSweeney’s site if you want to keep abreast of the ICM literature. There is nowhere better than Critical Care Reviews. https://criticalcarereviews.com/