Two-thirds of patients admitted to the Intensive Care Unit after a cardiac arrest die in the context of treatment withdrawal after a multimodal evaluation that determines an unfavorable neurological outcome.

This study will evaluate the Pupillary Pain Index (PPI) in the neurological prognosis after cardiac arrest. The PPI is determined by recording of pupillary dilatation with a videopupillometer after a calibrated and incremented nociceptive stimulus on a cutaneous metamere.

Cardiorespiratory arrest is associated with high mortality and morbidity rates. The direct consequence of a cardio-circulatory arrest is the absence of blood flow allowing oxygenation of the organs and consequently formation of ischemic lesions. Anoxic cerebral lesions are common in the aftermath of a cardiac arrest and often lead to the death of patients when active therapies are stopped after a multimodal prognostication that indicates that a poor outcome is very likely.

It is of paramount importance to optimize the sensitivity of the prognostication strategy in detecting good neurological outcome. A multimodal approach to the prognostic assessment is essential, and must include at least clinical examination, electrophysiology exploration (electroencephalography and/or evoked potentials) and biomarker analysis.

Although the most reliable predictors did not give false positives in most studies, none of them, considered individually, can establish an unfavorable prognosis with an absolute degree of certitude. For these reasons it is interesting to evaluate new prognostication tools.

The videopupillometry allows precise, reproducible and repeated measurement of changes in pupil diameter in response to a painful or a luminous stimulus. Pupillary pain reflex analysis is usually used to assess the degree of analgesia in a non-communicative patient during general anesthesia and neuromuscular blockade. The PPI score is determined at the bedside by recording pupillary dilatation after a calibrated and incremented nociceptive stimulus (electrical current between 5 and 60mA) applied to a skin metamere with two electrodes.

Automated pupillometry measurement has been recently developed to help support prognostication, with a quantitative pupillary light reflex measurement. The aim of this study is to evaluate the Pupillary Pain Index in the neurological prognosis after a cardiac arrest by correlating the PPI at 48h from the patient's arrival to the CPC score at 3 months.

Inclusion Criteria:

• Admission to the ICU after cardiac arrest
• Age > 18 years
• Presenting a Glasgow score of <8 at admission

Exclusion Criteria:

• Pregnancy
• Ocular pathology making pupillometry impossible.
• Patient under guardianship or curatorship

Patients remaining comatose after cardiac arrest