Alice Lagebrant - Institutionen för kliniska vetenskaper, Lund

Title: Neurological prognostication after cardiac arrest

Main supervisor: 

Marion Moseby Knappe, MD, PhD, Rehabilitation Medicine, IKVL, Lund University.

Reviewers: 

Maria Compagno Strandberg, Associate Professor Neurology, IKVL, Lund University.

Peter Bentzer, Professor Anaesthesiology and Intensive Care Medicine, IKVL, Lund University.

Abstract

Background

Prediction of neurological outcome in unconscious patients after cardiac arrest relies on a multimodal approach, including clinical examinations, neurophysiological tests, neuroimaging, and blood-based biomarkers. In patients with a likely poor neurological outcome, withdrawal of life-sustaining treatment (WLST) is common. However, many patients remain unconscious with an indeterminate prognosis after structured neurological prognostication, emphasising the need for more sensitive prognostic criteria.

Research questions

1. Can blood-based biomarker levels accurately predict hypoxic-ischaemic encephalopathy on brain computed tomography (CT)?
2. Does WLST, based on presumed poor neurological prognosis, result in worse long-term outcomes compared to matched patients in whom life-support is continued?
3. Does incorporating predictors of good outcome into prognostic guidelines improve the overall predictive performance?
4. How does hypothermia treatment affect neurological prognostication and WLST decisions?
5. What clinical cut-off levels of NFL best predicts poor neurological outcomes after cardiac arrest?

Preliminary results

1. The routinely available brain injury marker Neuron-Specific Enolase (NSE) predicts hypoxic-ischaemic encephalopathy on brain CT with high accuracy at 24 and 48 hours.
2. When WLST is performed in accordance with guidelines for predicting poor outcomes, it is not associated with worse outcome compared to matched controls. However, up to 18% of patients who remained unconscious with an indeterminate prognosis – and still underwent WLST due to a presumed poor neurological outcome – could have survived with good functional outcome if WLST had not been performed.
3. Incorporating predictors of good outcome into prognostic guidelines reduces the number of patients who remain with indeterminate prognosis.

Significance

The clinically available biomarker NSE can be used to assess the additive prognostic information from a repeated CT examination. This enables more effective use of resources during neurological prognostication. WLST can safely be performed in patients who meet the guidelines for a likely poor outcome. However, performing WLST in patients with an indeterminate prognosis may reduce their chance of a good recovery, highlighting the importance of adherence to guidelines. Incorporating good outcome predictors into the guidelines reduce the number of patients who remain with indeterminate prognosis, who may otherwise be at risk of inappropriate WLST. 

Published studies:

Lagebrant A, Lee BK, Youn CS, Sandroni C, Bělohlávek J, Cariou A, et al. Effects of withdrawal of life-sustaining therapy on long-term neurological outcome after cardiac arrest - a multicentre matched cohort study. Resuscitation. 2025:110747.

Lagebrant A, Sandroni C, Nolan JP, Bělohlávek J, Cariou A, Carrai R, et al. Prediction of good functional outcome decreases diagnostic uncertainty in unconscious survivors after out-of-hospital cardiac arrest. Resuscitation. 2025;214:110686.

Lagebrant A, Lang M, Nielsen N, Blennow K, Dankiewicz J, Friberg H, et al. Brain injury markers in blood predict signs of hypoxic ischaemic encephalopathy on head computed tomography after cardiac arrest. Resuscitation. 2023;184:109668.