Vasoactives in cardiogenic shock, videolaryngoscopy grading, and a negative ICU-VR trial

ESC Congress 2026 opened in Munich today, and the first wave has landed: the 2026 ESC heart failure guidelines and the Fifth Universal Definition of Myocardial Infarction, both in the European Heart Journal. Those are society guidelines and belong to the guidelines watch rather than here — see the guidelines section for today’s entry. Expect Hot Line trial readouts through 31 August.

1. Vasoactive agent selection strategies in cardiogenic shock

Journal · Published: Circulation, 24 August 2026 — genuinely new, 4 days old (clinical primer, not new trial data) Bohula & Morrow

A framework paper rather than a study. It describes the pharmacologic profiles of the vasoactive agents in current use and the available comparative-effectiveness data, reviews the pathophysiology of the common phenotypic forms of cardiogenic shock, and sets out a strategy for agent selection and titration that integrates pharmacology, pathophysiology and the haemodynamic derangements defining each phenotype.

The practice-based sequence it proposes:

1. Emergently protect the mean arterial pressure
2. Define the haemodynamic phenotype of the shock, and identify and address potential aetiologies and contributors in parallel
3. Set additional haemodynamic goals beyond MAP
4. Tailor pharmacotherapy through iterative reassessment of perfusion and ongoing derangements

Interpretation. Bohula and Morrow are the Critical Care Cardiology Trials Network principals, so this is the group with the largest contemporary cardiogenic-shock registry data setting out how they think the problem should be approached. The phenotype-before-drug ordering is the substance of it: the field has struggled to show that any single vasoactive agent beats another across undifferentiated shock, and this argues the comparison was always underspecified. Note it lands the same week as the ICM β-blockade review (26 August), which reached a similar phenotype-driven conclusion from the opposite direction — both are the field declining to give a population-level answer.

Read the paper · PMID 42636314

2. Classification of videolaryngoscopy: a systematic review

Journal · Published: Anaesthesia, 26 August 2026 — genuinely new, 2 days old Grün, Dankert, Wünsch et al.

Systematic review of the tools used to classify and document videolaryngoscopic tracheal intubation, assessing accuracy, performance and inter-rater reliability, with QUADAS-2 quality assessment.

• 13 eligible studies covering seven distinct classification tools in adults and children: Cormack-Lehane, percentage of glottic opening (POGO), intubation difficulty scale, Fremantle score, video classification of intubation score, VIDIAC, and the paediatric PeDiAC score. Eight were clinical studies in patients.
• Most studies reported only inter-rater reliability or concordance, leaving accuracy, performance, generalisability and clinical utility uncertain.
• Accuracy or performance metrics existed for only three tools — Cormack-Lehane, VIDIAC and PeDiAC — and Cormack-Lehane showed only limited performance.
• Empirically derived thresholds were absent for most tools.
• Overall risk of bias was high, particularly for the reference standard and for flow and timing, with substantial applicability concerns.

Interpretation. The finding that matters for daily practice is the one about Cormack-Lehane: the grading everyone writes in the airway record was developed for direct laryngoscopy, has limited measured performance under videolaryngoscopy, and carries no empirically derived threshold. Since that documentation is what the next anaesthetist plans from, a grade that doesn’t predict much is worse than an honest free-text description. The review doesn’t crown a replacement — VIDIAC and PeDiAC are the only alternatives with any performance data at all — so the practical takeaway is to record what was actually seen and done rather than to trust the grade.

Read the paper

3. ICU-specific virtual reality for mental health after critical illness

Journal · Published: Critical Care, 24 July 2026 — ~5 weeks old, not new Drop, Vlake, van Bommel et al.

Date note: criticalcarereviews.com listed this under 27 August. Its actual first publication was 24 July. This is the aggregator’s known date-versus-listing mismatch — worth checking every time before calling something new.

International, three-arm, multicentre RCT across eleven hospitals. Adults with an ICU stay ≥72 h and mechanical ventilation ≥24 h randomised 1:1:1 to standard care, early ICU-VR (within 7–15 days post-ICU), or late ICU-VR (≈3 months, during aftercare). Primary endpoint: PTSD symptom severity at six months on the Impact of Event Scale-Revised (0–88).

• 344 randomised (median age 61, 62% male); six-month follow-up among survivors 80%
• PTSD severity at six months: early ICU-VR vs control β=0.32 (95% CI −3.38 to 4.02, p=0.87); late ICU-VR vs control −0.82 (−4.62 to 2.97, p=0.67)
• No differences in anxiety, depression or health-related quality of life (all adjusted p>0.05)
• Patients nevertheless rated ICU-VR highly for improving their understanding of the ICU experience, and for satisfaction (median 8/10)

The authors are explicit about scope: the findings “should not be generalized to multi-session, therapist-guided, or personalized VR interventions,” and future trials should target high-risk patients with mandatory multi-session, personalised VR.

Interpretation. A well-conducted negative trial of a single-session intervention, and the gap between “patients liked it” and “nothing measurable changed” is the interesting part — satisfaction is not a treatment effect. It joins the run of null PICS-recovery trials in this archive: NEXIS (23 August) and EVER (26 August). Three negatives in a fortnight, all testing single, generic, protocol-driven interventions against a syndrome that is neither single nor generic. The common thread in all three discussion sections is the same: stop asking whether the intervention works, start asking who it works for.

Read the paper · PMID 42638131