Net ultra filtration (NUF) is one of the most important parameters during renal replacement therapy (RRT) whose role is to control fluid balance by water removal.

To our knowledge, there are no prospective studies or guidelines about the setting of this parameter.

In the NEPTUNE study, we aim to compare the hemodynamic effect of three NUF rates during RRT: 1 ml/kg/h, 2 ml/kg/h and 3 ml/kg/h.

The research hypothesis is that one of the three flow rates evaluated induces the fewest hemodynamic instabilities related to RRT, while guaranteeing the best possible fluid balance.

Renal renal replacement therapy (RRT) for severe acute kidney injury is one of the most widely used life-support techniques in intensive care. One of its main functions is to maintain the water balance in oligo-anuric patients by means of net ultrafiltration (NUF), defined as the volume of water removed from the patient by the RRT per unit time and indexed to the patient's weight.

Surprisingly, the setting of this parameter, which is one of the most important, there are no guidelines. If net ultrafiltration is too low, it may prolong RRT dependency and length of stay in the ICU, with all the associated care-related complications, and increase mortality. If net ultrafiltration is too high, it may transiently induce hypovolemia, leading to hemodynamic instability related to renal replacement therapy (HIRRT).

A recent survey of French practices shows that the average NUF flow rate used in intensive care units in France is 119 ± 77 ml/h, and the median flow rate is 100 ml/h [min-max 20-300]; in this study, NUF was not indexed by weight, but for a patient with an average weight of 75 kg, corresponding to an average NUF of 1.6 ± 1 ml/kg/h.

In the NEPTUNE study, the investigators aim to compare the hemodynamic effect of three net ultrafiltration rates during RRT: 1 ml/kg/h, 2 ml/kg/h and 3 ml/kg/h. These three flow rates are within the range of those usually used in clinical practice.The research hypothesis is that one of the three flow rates evaluated induces the fewest hemodynamic instabilities related to RRT, while guaranteeing the best possible fluid balance.

Inclusion Criteria:

• Patients hospitalised in the intensive care unit of one of the two participating centres

• Patients with Kidney Disease: Improving Global Outcomes 3 (KDIGO3) acute kidney injury (AKI) requiring a continuous RRT during their stay in the intensive care unit, regardless of the aetiology of the AKI

• Need to prescribe water loss by net ultrafiltration (NUF), defined by at least one of the following sub-criteria:

  1. weight gain ≥ 1 kg relative to entry weight
  2. oligo-anuria ≥ 24 hours
  3. clinical impact of fluid overload as judged by the clinician: acute lung oedema clinical or at CT-scan, difficulty of weaning from mechanical ventilation.

• Hemodynamic stability in the 2 hours preceding the start of NUF, defined by all of the following sub-criteria:

  1. absence of vasopressors (noradrenaline) or stability or reduction in their dosage
  2. no need for resuscitative fluids as judged by the clinician

• Patient or his/her trusted support person/legal representative/family member having given free and informed consent, and having signed the consent form or patient included in an emergency situation.

• Patient affiliated to or benefiting from a health insurance scheme.

Exclusion Criteria:

• Patient moribund, with life expectancy too low to benefit from treatment, or with decision to discontinue treatment
• Patient participating in an another interventional study
• Patient in exclusion period determined by another study
• Patient under court protection or guardianship
• Patient/trusted person/legal representative/family member for whom it is impossible to give informed information.
• Pregnant, parturient or breast-feeding patient.