Deresuscitation during continuous renal replacement therapy: A before-after pilot study (The EARLY DRY COHORT)

Background: Active fluid removal has been suggested to improve prognosis following the resolution of acute circulatory failure. To standardize the deresuscitation strategy, we have implemented a routine care protocol to guide fluid removal during continuous renal replacement therapy (CRRT). We designed a before-after pilot study to evaluate the impact of this deresuscitation strategy on the cumulative fluid balance. Methods: : Consecutive ICU patients suffering from fluid overload and undergoing CRRT for acute kidney injury underwent a perfusion-based deresuscitation protocol combining a restrictive intake, continuous net ultrafiltration (UFnet) of 2 mL/kg/h, and both clinical and laboratory monitoring of perfusion (early dry group, N=42) and were compared to an historical group managed according to usual practices (control group, N=45). The primary outcome was the cumulative fluid balance at day 5 or at discharge. Secondary outcomes addressed the protocol safety. Adjustments were done with inverse probability of treatment weighting propensity score analysis. Results: : Adjusted cumulative fluid balance was significantly lower in the early dry group (median [IQR]: -7784 [-11833 to -2933] mL) compared to the control group (-3492 [-9935 to -1736] mL, p=0.04). The difference was mainly driven by a greater daily UFnet (31 [22-46] mL/kg/day vs. 24 [15-32] mL/kg/day, p=0.01). There was no significant difference between both groups regarding maximal arterial lactate level and maximal norepinephrine dose requirement. Conclusion: Our perfusion-based deresuscitation protocol achieved a greater negative cumulative fluid balance compared to standard practices and was hemodynamically well tolerated. Those data suggest the feasibility of an interventional randomized clinical trial using a similar protocol.