Transcranial Doppler ultrasound (TCD) monitoring and CT-scanner perfusion are useful but imperfect tools to identify vasospasm and allow intervention to avoid infarction.

Permanent monitoring of cerebral tissue oximetry (rSO2) by NIRS, a noninvasive method could allow better vasospasm detection.

This study will evaluate diagnostic accuracy of cerebral oxymetry (NIRS) -by rSO2 measurement - in order to detect vasospasm in patient with severe subarachnoid hemorrhage compare to standard monitoring tools.

Delayed vasospasm is a serious complication of aneurismal subarachnoid haemorhage (SAH) significantly influencing morbidity and mortality. Mostly observed between days 4 and 10 after aneurismal rupture, its incidence is higher in cases of severe SAH. Vasospasm strongly affects prognisis generating delayed cerebral ischemia.

Clinical deterioration (focal neurological deficit) is the best way to detect severe vasospasm. Unfortunately, most of severe SAH are intubated and sedated prohibiting neurological evaluation. In these frequent situations, Transcranial Doppler (TCD), clinical and biological monitoring, CT-scanner (angio-CT and Perfusion-CT), MRI and cerebral angiography are routinely used to detect vasospasm. Yet, these tools have imperfect sensitivity and specificty delaying diagnosis and treatment.

Near-InfraRed Spectroscopy (NIRS) is a non-invasive method measuring tissue oxygenation by regional saturation of capillary-oxygenated hemoglobin (rSO2). This technology demonstrated ability to measure cerebral oxygenation and has previously been reported to monitor carotid and pediatric surgery. To date a few studies also reported NIRS monitoring feasability in post-SAH vasospasm. Our aim is here to evaluate NIRS monitoring for the diagnosis of vasospasm in severe SAH.

Inclusion Criteria:

• Men and women
• Age ≥ 18 years
• HSA of aneurysmal etiology less than 4 days before inclusion diagnosed by clinical presentation and emergency imaging
• HSA "severe" defined according to the WFNS ≥ III
• Intubated-ventilated patient or any other reason preventing a contributing neurological examination
• Affiliate or beneficiary of a social security scheme
• Free, informed and written consent signed by the representative

Exclusion Criteria:

• Age <18 years
• Significant vasospasm as soon as the detection was detected on the initial imaging
• Possible Neurological Surveillance
• Patient in limitation of active therapeutics or with high probability of soon limitation of active therapeutics
• Intra-parenchymal frontal haematoma limiting the quality of measurement, diagnosed on imaging (Less than 25mm deep from the skin)
• Patient in limitation of active therapeutics or with high probability of soon limitation of active therapeutics
• Adhesive allergy to measuring patches
• Patient under guardianship or safeguard of justice
• Refusal of legal representative