Context : Moderate to severe head trauma with altered state of consciousness is an extremely common pathology (between 60 and 120 cases per 100 000 people per year depending on the country and age group), and is responsible for 30% of deaths by trauma. It is complicated in 30-60% of cases by subarachnoid hemorrhage (SAH), which makes it the leading cause of SAH. SAH and its complications are well described when the origin is aneurysmal, notably cerebral vasospasm (CV) because it promotes delayed cerebral ischemia with a major prognostic impact. This is why the screening and prevention of this vasospasm are well established in the literature and in practice, in the nosological context of aneurysmal SAH. Research problem : However, when it comes to post-traumatic SAH, CV is a more maligned entity, with a much less detailed description. However, when we know the prognostic interest that it could have for patients, it seems legitimate to seek to define its physiopathological and epidemiological contours. On a prospective cohort of 290 subjects, Oertel et al. (2005) demonstrated, in head trauma patients, an incidence of approximately 40% of compatible signs with the recognized criteria of CV. To date, the literature remains sparse on this subject. Proposed study : In view of the incomplete scientific literature, the study team wish to carry out a prospective epidemiological study in moderate to severe head trauma patients complicated by SAH and hospitalized at the Martinique University Hospital, with the aim of better characterizing the incidence of the occurrence, and evolution of CV with clinical impact in these patients. One of the original aspects of the proposed study is the use of CT scan with perfusion sequence, which has shown its superiority to Transcranial Doppler. The other particularity is its prospective aspect and triggered by an alteration in the clinical state of the patient presenting a traumatic SAH, then directly linking the pathophysiology (cerebral ischemia) and the clinical impact. Thus, the diagnosis of traumatic CV will be made on a cerebral CT scan by the association of the 50% reduction in the caliber of one or more cerebral arteries and a perfusion defect in the perfusion sequence in a context of alteration of neurological clinical examination or deterioration of neurological monitoring parameters. Finally, few studies have monitored the evolution of these patients at 1 and 6 months after the initial event. Hypothesis : The research hypothesis is that in the population of moderate to severe head trauma patients hospitalized at the Martinique University Hospital, when a new neurological symptomatology or a deterioration in the state of consciousness occurs, it could be a post-truamatic CV in 15 to 20% of cases. Indeed, the rare studies find frequencies of radiologically confirmed CV in head trauma patients of around 30-45%, with low numbers of subjects, retrospective studies, or not correlated with the clinic and with the clinical and paraclinical data necessary for the positive diagnosis of this entity. The reported frequency of traumatic CV with clinical impact ranges between 15-20%. The study team therefore expect an incidence of 15 to 20% of CV with clinical impact in patients with traumatic SAH in Martinique. CV could be responsible for sudden deterioration of the neurological state in patients suffering from traumatic SAH between the 3rd and 12th day inclusive of treatment (according to retrospective studies already carried out) and responsible for its specific morbidity linked to cerebral ischemia localized in the spasmed area manifested by a worsening of the neurological prognosis on the modified Rankin scale.