Atrial fibrillation (AF) is a cardiac rhythm disorder particularly common in intensive care patients. Some meta-analyzes report a prevalence of new onset AF ranging from 4.5% to 29.5% in polyvalent intensive care. In our department, a recent month-long survey showed that more than 30% of the patients who were unhealthy on admission suffer from an episode of new onset AF during their stay.

The occurrence of AF in intensive care has a pejorative effect on the patient's outcome, and this through two factors. On the one hand, the decreasing of cardiac output by degradation of the ventricular filling in diastole time, on the other hand the FA is responsible for an over-risk of ischemic stroke. In fact, it has been shown that the occurrence of new onset AF in intensive care is associated with a higher level of severity and a higher mortality. It is also important to underline the medico-economic impact of this rhythmic disorder as complication of shock due to the frequent prescription of various anti-arrhythmic or anticoagulant medication.

Various factors have been mentioned to explain the frequent occurrence of AF in shocked patients. The shock state, whatever its origin, is characterized by the occurrence of a systemic inflammatory response syndrome in which is observed a particularly important releasing of stress hormones and endogenous catecholamines involved in the occurrence of a rapid multi-organ failure without treatment. Systemic humoral elements are possibly involved in the occurrence of new onset AF, such as high level of inflammation that characterizes shock states. In addition, physiological factors such as hypoxia, hypovolemia, hyperthermia or ionic disorders are also implicated, but their non-systemic association with intensive care new onset FA suggests that humoral factors may play an important and independent role. Among these humoral factors, the proteins of chromogranin family particularly Vasostatin-I (VS-I) seem possibly involved in the genesis of AF in the aggressed intensive care patients. Several studies have highlighted the beneficial regulatory role of VS-I on the cardiovascular system, particularly in a study on a canine model Stavrakis and al. have shown the VS-I protective role on the FA occurrence.

However, as has been demonstrated in a prospective study in intensive care, the rates of circulating VS-I were significantly higher in the most severe patients and those whose prognosis was the most pejorative, thus not supporting the thesis of the protective effect of VS-I. An explanation exists for this discrepancy: VS-I is present in two distinct forms in the circulating blood. In vitro work carried out within the U1121 INSERM team with has made it possible to highlight the coexistence of two forms of VS-1: an aggregated "inactive" form and an "active" disaggregated form. In our hypothesis, the inactive aggregated form would be predominant during the states of acute pathological aggressions such as the shock and thus would not exert the anti arrhythmic and cardio protector expected functions.

The first aim of our study is therefore to confirm that the onset of new onset FA during the shock state is associated with a significant decrease in the VS-I plasma level in its monomeric form called "active", even when high levels of total VS-1 are detected by ELISA in the plasma of patients.

Our project is a pilot and unpublished translational work. The link between VS-I and new onset AF in intensive care severe patients has never been studied in vivo, and the recent work of the associated INSERM team provides advances in understanding the function of VS-I over time shock conditions. Nevertheless, our experimental hypothesis require confirmation in humans.

A better understanding of the factors influencing the occurrence of cardiac arrhythmias in intensive care patients is a major ambition as it would be a step forward in the development of a preventive strategy or new treatment for the benefit of patients.

Inclusion Criteria:

• Major patients (no upper age limit), male or female.
• Consenting to participate in the study (consent sought from the person of trust in case of inability to consent). Informed consent was given in writing.
• Subject affiliated to a social health insurance regiment
• Presenting a state of shock characterized by a SAP <90mmHg or MAP <65mmHg despite 1000ml of fluid resuscitation and requiring the use of norepinephrine to obtain a SAP >90mmHg or MAP >65mmHg.
• arterial lactatemia is greater than 2mmol / L at baseline
• equipped with radial or femoral arterial catheter (set up as part of the usual care)

Exclusion Criteria:

• Refusal to participate in the study
• Patient subject to legal protection (guardianship, curatorship or legal safeguards)
• Pregnant patient
• History of paroxysmal or permanent FA
• Recent history of cardiac surgery (<15 days)