Poor neuroprognosis after sudden cardiac arrest: end-of-life practices and potential organ donors in a cardiac intensive care unit cohort
M Urpina Matias, C Santos-Jorge, M I Soares, R Montalvao, M Presume, J Machado, A R Bello, M Sousa Paiva, J Presume, J Ferreira, C BrizidoAbstract
Introduction
Hypoxic-ischemic brain injury after sudden cardiac arrest (CA) is central to CA management, as major efforts focus on its prevention during pre- and post-CA care, and it is a key determinant of prognosis after ROSC. European Resuscitation Council (ERC) guidelines define criteria for poor neuroprognosis, but their application does not invariably lead to withdrawal of life-sustaining therapies (WLST). Moreover, these patients may be candidates for solid organ donation after brainstem or circulatory death (DBD or DCD, respectively).
Purpose
To characterize end-of-life practices in post-CA patients classified as having poor neuroprognosis according to ERC guidelines, and to identify potential DBD and DCD donors.
Methods
Single-center retrospective study of consecutive post-CA patients admitted to a CICU between 2015-2025, presenting comatose (GCS <8) after ROSC and undergoing multimodal neuroprognostication. The 2021 and 2025 ERC guidelines algorithms were applied to classify neuroprognosis as good, indeterminate or poor. Patients were evaluated regarding end-of-life decisions (WLST or No Escalation of Treatment [NoET]) and cause of death. For donation purposes, potential donors were categorized as 18-45 years old (ideal), 45-55 (acceptable) and 55-70 (marginal); patients > 70 were considered unlikely donors.
Results
Among 94 patients, 26% (n=24) had good, 37% (n=35) indeterminate and 37% (n=35) poor neuroprognosis, according to ERC criteria. Of the latter, only 6 (17%) progressed to brainstem death; 2 went into circulatory arrest after WLST, and the remaining 4 after NoET. Among the remaining 29 patients with poor neuroprognosis, WLST was instituted in 11 (38%) and NoET in 8 (28%). Ten patients either died from "refractory circulatory shock" (n=5), suggesting treatment escalation, or had no clear end-of-life plan (n=5). Five patients (17%) with poor neuroprognosis underwent tracheostomy, all aged > 55 years. None of the 6 brainstem-dead patients were considered for DBD, although all but one were <70 years. Regarding DCD, 19 (66%) patients with poor neuroprognosis were potential donors (2 ideal, 3 acceptable, 14 marginal). From those, 2 were resuscitated with ECPR and could have been considered Maastricht II donors, while the remaining 17 could have been Maastricht III donors if national legislation allowed; none were evaluated for DCD.
Conclusion
Around 40% of resuscitated CA patients were classified as having poor neuroprognosis, yet only a minority (one fifth) of hypoxic-ischemic brain injuries evolved to brainstem death. Despite a substantial proportion of potential donors, neither DBD nor DCD was implemented. End-of-life decisions, particularly WLST, were underused in this context (< 40%).For image description, please refer to the figure legend and surrounding text.