To date, the end-expiratory occlusion test in infants or children has never been evaluated as a marker of preload dependence. It therefore appears clinically relevant to evaluate this new indicator to predict fluid responsiveness in all infants hospitalized in pediatric and neonatal intensive care.

The main objective of this study is to determine whether the hemodynamic effects of a 15-second end-expiratory occlusion were able to predict fluid responsiveness in the mechanically ventilated infant or newborn in pediatric intensive care.

Volume expansion remains one of the first resuscitation treatments for states of shock. Nonetheless, the decision to perform volume expansion should be based on a "functional" hemodynamic assessment, i.e., filling should only be performed if changes in cardiac preload result in significant changes in stroke volume. This is then referred to as preload dependency. Unnecessary volume expansion induces fluid overload and is associated with increased morbidity and mortality for shock conditions in infant and children.

In addition, several studies in adults and children have demonstrated the lack of effectiveness of so-called "static" preload variables as witnesses of preload dependence. Only "dynamic" preload variables, based on the notion of cardiopulmonary interaction during mechanical ventilation, can predict this preload dependence, however with very variable sensitivities and specificities depending on the studies.

The end-expiratory occlusion test is a dynamic preload dependency test that can be performed in patients on invasive ventilation. The test consists in interrupting the ventilator at end-expiration for 15 seconds, delaying the next insufflation thus allowing an increase in venous return to the right heart and then to the left heart. If the heart is preload dependent, an increase in venous return is accompanied by an increase in stroke volume and cardiac output (Franck-Starling's law). Volume expansion is usually considered effective or positive if cardiac output increases by more than 15% from baseline.

To date, the end-expiratory occlusion test in infants or children has never been evaluated as a marker of preload dependence. It therefore appears clinically relevant to evaluate this new indicator to predict fluid responsiveness in all infants hospitalized in pediatric and neonatal intensive care.

The main objective of this study is to determine whether the hemodynamic effects of a 15-second end-expiratory occlusion were able to predict fluid responsiveness in the mechanically ventilated infant or newborn in pediatric intensive care.

Inclusion Criteria:

• Newborn and infant <2 years.
• Born at or near term (> 36 weeks of amenorrhea).
• Hospitalized in pediatric and neonatal intensive care.
• Requiring the use of invasive mechanical ventilation.
• In a state of shock defined by a cardio-circulatory dysfunction associating arterial hypotension, an alteration in organ perfusion, one or more organ dysfunctions or vasopressor's requirement. Clinical signs retained: heart rate ≥ 90th percentile associated with at least one sign of peripheral hypoperfusion represented by a a capillary refill time ≥ 3sec, or the presence of mottling or coldness of the extremities or urine output ≤ 0.5ml / kg / h or disturbance of consciousness.
• The practitioner in charge of the patient has decided to perform volume expansion.
• Non-opposition expressed by the holders of parental authority.

Exclusion Criteria:

• Any serious and urgent clinical hemodynamic situation, the management of which would be delayed by inclusion in the protocol.
• Patient with unoperated congenital heart disease.
• Patient in spontaneous or non-invasive ventilation or with spontaneous ventilatory cycles in invasive ventilation.
• Patient in prolonged cardiac arrest (> 5min) or refractory.
• Patient under venous-arterial extracorporeal membrane oxygenation.
• Refusal of the holders of parental authority to exploit the collected data.