ADQI XXXVI and ELSO on acute kidney injury and CRRT during ECMO — the first of the series

One new document. All three sweeps ran clean with no query failures: the topic-wide guideline search across all journals for 10–11 September (24 hits, one in scope), the society-acronym sweep, and the French-society sweep — a confirmed zero for the eleventh consecutive day. SFAR, SPILF, SRLF and Infectious Diseases Now have published nothing in guideline form this month.

ADQI XXXVI with ELSO — acute kidney injury and CRRT during extracorporeal support

Societies · Published: The 36th Acute Disease Quality Initiative consensus conference, jointly with the Extracorporeal Life Support Organization. Blood Purification, 10 September 2026 — published yesterday Kashani, Selewski, Gist, Ostermann, Mehta, Ronco, Bagshaw, MacLaren, Askenazi, Rintoul, Singh, Kaddourah

Not on the tracked society list, but ELSO is the authoritative body for extracorporeal life support and ADQI is where the acute kidney injury definitions come from. Squarely primary scope — this sits where intensive care meets cardiothoracic surgery, and the author list is the relevant one: Ronco, Bagshaw, Ostermann and Mehta from the AKI side, MacLaren and Rintoul from ECLS.

The epidemiology, stated plainly:

• AKI occurs in 40–80% of patients on ECMO — among the most frequent complications
• Approximately 40–50% receive continuous renal replacement therapy
• Both AKI and CRRT are independently associated with increased mortality, prolonged duration of support, and long-term risk of chronic kidney disease

The mechanisms it names, which are worth reading as a list because several are specific to the circuit rather than to the patient:

AKI in this setting is multifactorial, reflecting pre-ECMO shock or hemodynamic instability, systemic inflammation, hemolysis, venous congestion, non-pulsatile flow, nephrotoxic exposure, and fluid overload.

What the conference did and what it prioritised:

• International experts in nephrology, critical care, cardiology, perfusion and extracorporeal support convened to evaluate evidence and generate consensus statements on epidemiology, prevention and management
• Emphasis on harmonisation of definitions, identification of modifiable risk factors, optimisation of CRRT strategies, and research priorities
• Stated needs: standardised definitions, prospective multicentre trials, biomarker validation, and integration of nephrology expertise into ECMO programmes

Interpretation. Two things to take from this, and one caution about what it is.

The caution first: this is the epidemiology and risk-factor workgroup, not the management recommendations. ADQI conferences publish as a series of workgroup papers, and a search of the last six weeks finds only this one so far from ADQI XXXVI. The abstract describes itself as “this review… developed in pursuit of” the conference, highlighting patterns and research priorities. So there is no actionable CRRT protocol here yet. The companion papers on prevention, fluid management and CRRT strategy should follow in the coming weeks, and this watch will pick them up.

The substance that is here: the mechanism list is the useful part. Four of the seven named drivers are things the perfusion and ICU team control or modulate — venous congestion, non-pulsatile flow, nephrotoxic exposure, and fluid overload — and only the first two entries (pre-ECMO shock, systemic inflammation) are givens by the time the patient is cannulated. Haemolysis sits in between: circuit-dependent, measurable, and modifiable by flow and cannula configuration.

Of those, venous congestion and fluid overload are the pair this archive has been circling for a month. The ESICM fluid therapy guideline Part 3, on fluid removal at de-escalation, is the relevant companion document, and the renal perfusion phenotype work sent on 8 September — where a semi-quantitative perfusion score discriminated persistent AKI while renal resistive index trajectories did not — points at the same physiology from the other direction. Congestion, not just pressure, and not just creatinine. An ECMO patient with a high central venous pressure and a swollen kidney is the clearest version of that problem, and naming venous congestion explicitly in a consensus mechanism list is a step the field has been slow to take.

Non-pulsatile flow is the one worth flagging for cardiac anaesthesia specifically. It is the same question as pulsatile versus non-pulsatile cardiopulmonary bypass — and the Annals of Thoracic Surgery published a critical perspective on perfusion modalities, “Comparison of Two Forms of Pulsatility,” on 9 September, which surfaced in that day’s sweep and was not written up. Whether pulsatility protects the kidney has been contested for forty years without resolution; ADQI listing it as a mechanistic driver does not settle it, but it does mean the consensus process took it seriously.

“Integration of nephrology expertise into ECMO programmes” is the organisational recommendation, and it is the same shape as the endocarditis team and the heart team — a formal multidisciplinary structure for a problem that falls between specialties. With AKI in 40–80% of ECMO patients and half of them on CRRT, the case for it is hard to argue with: the kidney is not a complication of ECMO, it is part of the routine course.

Access note: the abstract is deposited and was read in full. The consensus statements themselves — the numbered, graded output of the workgroup — are not in the abstract and the full text was not reachable. What is reported above is the conference’s framing and epidemiology, not its statements. Recorded on the outstanding list.

Read the paper · PMID 42721066

---

Also seen and not in scope: a UK consensus framework on familial pulmonary fibrosis (Thorax, 10 September — respiratory medicine and genetics rather than perioperative or critical care, though ILD patients reach theatre and the framework covers relatives’ investigation), a seven-organisation position paper on fracture and fall prevention (European Geriatric Medicine), international consensus guidance on population screening for islet autoantibodies (Diabetologia), 2026 Taiwan migraine guidelines, a Latin American Thyroid Society consensus on anaplastic thyroid cancer, expert recommendations on portal hypertension in cirrhosis, an ACMG position statement on genetic testing in athletes, a Myopia Society nomenclature consensus, and an SFBC update on internal quality controls in medical laboratories — the only French-society document to surface this month, and it is laboratory accreditation, not clinical practice. The Annals of Thoracic Surgery scoping review on adherence to guideline-directed aortic imaging surveillance appeared in this sweep because of “guideline-directed” in its title; it is a review of adherence rather than a guideline, and it was read and reported in today’s literature brief instead.