Documenting graft outcome to substantiate the organ-donation rationale after out-of-hospital cardiac arrest due to aneurysmal subarachnoid haemorrhage

To the Editor,

We read with interest the multicentre cohort study by Legros and colleagues [1]. We share their view that out-of-hospital cardiac arrest due to aneurysmal subarachnoid haemorrhage (OHCA-aSAH) should not be dismissed as uniformly futile, and that early intensive resuscitation serves two purposes: an individual benefit for the few patients who recover, and a collective benefit, through organ donation, for the many who progress to brain death. The study clearly shows that some patients survive. The open questions are whether the observed outcomes show that aggressive treatment produced recovery, and whether the donation benefit invoked has actually been delivered. We outline the data that would answer each (Fig. 1).

Fig. 1.

Individual and collective rationales for resuscitation in OHCA-aSAH, and the data needed to substantiate each. Reconstructed from the aggregate data reported by Legros et al. The individual-benefit pathway cannot identify which patients recover. The collective-benefit pathway reports the donation rate but not the transplant yield or graft outcome, on which the value of donation depends. Percentages are those published in the source article; figures marked with a tilde (~) are derived. OHCA-aSAH, out-of-hospital cardiac arrest due to aneurysmal subarachnoid haemorrhage; mRS, modified Rankin Scale; DBD, donation after brain death

The authors rightly attribute the more invasive management of patients with favourable outcome to confounding by indication and possible self-fulfilling prophecy. A further, time-related mechanism deserves mention. Aneurysm treatment could only be given to patients who survived long enough to receive it, and the delay from admission to aneurysm securing was not reported, as the authors acknowledge. The comparison between treated and untreated patients is therefore exposed to immortal-time (guarantee-time) bias [2]. All eight patients with favourable outcome underwent coiling, compared with 18.6% of those with poor outcome, and 124 of 164 patients (75.6%) received no aneurysm treatment. Consequently, this association cannot establish a treatment effect, because treatment allocation was strongly conditioned by early survival and clinical selection; notably, no untreated patient had a favourable outcome to serve as a comparator. Nor do conventional severity grades discriminate between groups: after cardiac arrest, patients are comatose, with a Glasgow Coma Scale (GCS) score of 3 and hence World Federation of Neurosurgical Societies (WFNS) grade 5, and modified Fisher grade 4 was near-universal in both groups (100% vs 92.9%). As the authors note, GCS in this setting reflects the absence of cerebral perfusion rather than intrinsic brain injury. Apart from pupillary abnormalities (12.5% vs 74.4%), which may themselves have shaped treatment decisions, these admission variables do not explain who recovered, leaving early clinical selection and subsequent trajectory as plausible contributors. Consistently, the penalized model identified no independent predictor of good outcome; with eight candidate predictors for seven events and no internal validation, its apparent discrimination (c-index 0.95) is likely optimistic. Moreover, meaningful recovery was exceptional: of the eight patients (4.9%) with favourable outcome, seven had a modified Rankin Scale (mRS) score of 2–3 and only one (0.6%) reached mRS 0–1.

This two-pathway message implies an early trajectory that the study does not describe. The very early occurrence of brain death in this cohort (a median of 1 [1–2] day) suggests that the early clinical trajectory, rather than admission severity, may strongly influence treatment selection. This does not argue for deliberately postponing aneurysm treatment: contemporary aSAH guidelines recommend securing the ruptured aneurysm as early as possible, preferably within 24 h, to reduce the risk of rebleeding [3], which predominates within the first 72 h and in poor-grade patients [4]. The more clinically relevant question is whether early aneurysm securing can preserve the individual chance of recovery while neurological prognostication is deferred until the effects of cardiac arrest, sedation, and early brain injury can be assessed more reliably [5]. Magnetic resonance imaging (MRI) at around 48 h, when diffusion abnormalities become apparent and are less likely to be missed than within the first 24 h, could help gauge the extent of hypoxic–ischaemic injury during this window [6]. Reporting the timing of return of spontaneous circulation, aneurysm securing, rebleeding, brain death, and withdrawal of life-sustaining therapy would make this trajectory explicit and help distinguish treatment effect from early selection and survivorship.

When the individual prospect is this narrow, the rationale for maintaining resuscitative management, which the authors explicitly endorse in recommending intensive care unit (ICU) admission for all patients and the avoidance of very early withdrawal, rests substantially on the collective benefit of organ donation. That benefit is genuine and well characterised: transplantation from brain-dead donors, including marginal grafts, is economically dominant over dialysis and gains life-years [7] (Table 1). Precisely because it is real, it warrants measurement here rather than assertion. The study reports that 45.4% of brain-dead patients became donors, consistent with donation rates after resuscitated cardiac arrest [10], but not the number of organs retrieved or transplanted, graft function, or recipient outcome. Whole-body ischaemia-reperfusion injury from the arrest is a legitimate concern for graft quality; reassuringly, organs from brain-dead donors resuscitated from cardiac arrest perform comparably to those from non-resuscitated donors [11–13], yet none of these outcomes is reported for the present cohort. The 45.4% figure is similar to the 48.3% of recorded brain-dead potential donors actually procured in France in 2023 (1,512/3,132; refusal, ~36%, being the leading cause of non-procurement) [14], although the populations differ, and the study does not report why the remaining 54.6% did not donate.

Table 1.

Graft-outcome and cost-effectiveness metrics that would substantiate the organ-donation rationale, with indicative reference values from deceased brain-dead (DBD) donors

Metric (kidney)	Reference value, DBD donors [7–9]	This cohort
Organs transplanted per donor	~2.6 (≈2.9 retrieved)	To be documented
1-year graft survival	~92–93%	To be documented
5-year graft survival	~72–78%	To be documented
Delayed graft function	~15–28%	To be documented
Primary non-function	~2–5%	To be documented
Cost-effectiveness vs dialysis	Economically dominant; cost-effective even for marginal (high-KDPI) grafts	To be documented

Kidney is shown as the organ with a chronic comparator (dialysis). Reference values are indicative figures from the cited transplant literature. Graft survival from donors brain-dead after resuscitated cardiac arrest is comparable to standard DBD; the pathway is promising, and quantifying it in this population would substantiate the donation rationale. DBD, donation after brain death; KDPI, Kidney Donor Profile Index

Two constructive implications follow. First, the development of controlled donation after circulatory death (cDCD; Maastricht category III) may help dissociate the initial therapeutic strategy from the subsequent donation pathway. Initial management can remain directed toward the patient’s individual benefit (including aggressive neurocritical care and early aneurysm securing) followed by appropriately delayed neurological prognostication. If prognosis is subsequently considered catastrophic and a decision to withdraw life-sustaining therapy is made independently of any consideration of organ donation, donation may still be considered after either brain death or controlled circulatory death, according to eligibility and local protocols [15]. Second, a prospective extension of this cohort should capture the timing of aneurysm securing, rebleeding, neurological prognostication, brain death and withdrawal of life-sustaining therapy, as well as the donation pathway, organs retrieved and transplanted, graft outcomes, recipient outcomes, and reasons for non-donation. Such time-resolved data would better quantify both the individual and collective consequences of initial resuscitative management.

In conclusion, this valuable cohort establishes the severity of OHCA-aSAH and shows that survival remains possible; by design, it cannot establish that aggressive treatment produced the few recoveries observed, nor fully quantify the collective benefit of organ donation. Describing the early trajectory, documenting both brain-death and controlled circulatory-death donation pathways, and reporting transplant yield and graft outcomes would allow the author’s clinical recommendations to rest on firmer ground.

Abbreviations

aSAH

Aneurysmal subarachnoid haemorrhage

DBD

Donation after brain death

GCS

Glasgow Coma Scale

ICU

Intensive care unit

KDPI

Kidney Donor Profile Index

MRI

Magnetic resonance imaging

mRS

Modified Rankin Scale

OHCA-aSAH

Out-of-hospital cardiac arrest due to aneurysmal subarachnoid haemorrhage

WFNS

World Federation of Neurosurgical Societies

Author contributions

Conceptualization: NC, BB. Writing: The manuscript was written collaboratively by NC and BB. Review and Editing: All authors reviewed and edited the manuscript for intellectual content and clarity. Final Approval: All authors approved the final version of the manuscript.

Funding

None.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Use of large language models

A large language model (Claude, Anthropic) was used to assist with language editing of this manuscript. All content was reviewed, verified, and approved by all co-authors, who take full responsibility for the final text.

Competing interests

The authors declare no competing interests.