In ICU, fluid challenge represents one of the cornerstones of hemodynamic care. However, fluid overload due to an excessive and/or inappropriate fluid administration could be associated with morbidity or mortality. Unfortunately, there is currently no continuous non-invasive devices to monitor fluid content at bedside. Bio reactance is a non-invasive, rapid and continuous technology developed in order to measure body fluid compartment. Monitoring devices functioning with such technology are promising to evaluate fluid overload in ICU.

During the resuscitation phase of shock, fluid administration represents one of the cornerstones of care in order to increase cardiac output and improve microvascular blood flow. However, inappropriate fluid administration can increase tissue edema which compromises recovery after the resolution of shock state. Recently, Sakr et al. demonstrated that a higher fluid balance at 72 hours was associated with hospital mortality after septic shock. Furthermore, an administration of more than 5 liters of fluid during the first ICU stay was independently associated with an increase in mortality and hospital costs. Therefore, treating fluid removal appears to be a key component of the de-escalation phase of shock. Thus, all valuable parameters which potentially reflect tissue edema may help clinicians to individualized the necessity of fluid removal during the stabilization and de-escalation phase of shock. Among them, extra-vascular lung water (EVLW) measured with trans-pulmonary thermodilution is able to detect changes in thoracic fluid content but needs to be monitored invasively and only sequential values are recorded (each thermodilution measure). On the other hand, lung ultrasonography may help clinicians to assess fluid overload but its ability to quantify thoracic fluid content is difficult and subjective.

Bio reactance is a non-invasive, rapid and continuous method to measure body fluid compartment. All measures can be performed at bedside. Bio reactance monitoring devices allow measurement of hemodynamic parameters such as cardiac index or stroke volume but also Thoracic Fluid Content (TFC). TFC is measured through the changes in impedance of thoracic tissue to the electrical current. This parameter represents the whole fluid content in the thorax (intravascular, extravascular and intra-pleural). TFC has already been evaluated in several context. During hemodialysis, TFC is correlated to the amount of fluid removal and might help clinician to improve hemodialysis session management in ICU. In cardiac surgery, electrical impedance is correlated with changes in fluid balance. In ICU, TFC is able to predict a mechanical ventilation weaning failure with a moderate accuracy (AUC 0.69 [0.57 - 0.8], bet cut-off value > 50 k.Ω-1) in patients with moderate to severe alteration of left ventricular ejection fraction.

The main objective of the current study will evaluated correlation between thoracic fluid content (TFC) measurement and other valuable indices of fluid overload used at bedside. Secondary objectives will be to evaluate association between TFC and other clinical outcomes (organ dysfunction, mortality and quality of life after hospital living).

Inclusion Criteria:

• Adult (Age >18 years old)

• Septic shock according to Sepsis-3 definition:

  • A suspected or confirmed infection
  • Persisting hypotension, despite adequate fluid resuscitation, requiring vasopressor to maintain a Mean Arterial Pressure (MAP) ≥ 65 mmHg
  • Lactate level > 2 mmol/l

• Predictive ICU length of stay > 3 days

Exclusion Criteria:

• Admission in ICU for more than 3 days
• Refusal to participate
• Moribund patients
• Decision of therapeutic withdrawal
• Curators

Adult patients admitted in ICU for septic shock

All collected data that underlie results in a publication

Data will be available beginning three years and ending fifteen years following the final study report completion

Data access requests will be reviewed by the internal committee of Brest University Hospital. Requestors will be required to sign and complete a data access agreement