Chapter 24 — ECPR Failure and Withdrawal

101 kB. 13 sections, 35 callouts, 8 tables, 1 mermaid flowchart. COMPLETES PART IV. Scope finding: a tracker query established that no chapter owned brain-death determination on a circuit, organ donation after ECMO, or normothermic regional perfusion; all three placed here on physiological grounds (brain death is an ECPR event, the apnoea test is defeated by the sweep gas, and regional perfusion is an ECPR circuit with the arch vessels ligated). Chapters 74-77 keep the ethics. Organising insight from the one ethnography that observed rather than surveyed (380 hours, two units): treatment is escalated until options are exhausted and family discussion did not occasion decision-making moments - so decision points must be manufactured in advance. Second structural claim: ECPR is begun without consent and in some jurisdictions may be stopped only with it. Three controversies: who may insist on continuing, what substitutes for a failed apnoea test, and whether donation may influence the pathway. Six numerical flags printed uncorrected, including an adjusted odds ratio of -3.1 with interval 2.18-2.8, and the fourth Part IV instance of the same analysis reported twice with different denominators. Safety finding: in two series as many patients had treatment withdrawn with absent brainstem reflexes as were declared dead - complete the determination before discussing withdrawal. Most actionable protocol line: reduce the sweep, never the blood flow; both radial samples simultaneously. All external evidence from structured abstracts; DOIs reproduced from the search record and not independently resolved; every seeded row Verified = No. 12 evidence rows and 2 guideline rows added.