Delaying Renal Replacement Therapy Could Be Harmful in Patients with Acute Brain Injury.

To the Editor:

We read with interest the article by Gaudry and colleagues on the recent advances regarding the timing of the initiation of renal replacement therapy (RRT) for acute kidney injury (AKI) in critically ill patients (1). The authors conducted the two most recent large-scale studies in this area (2, 3) with concordant results, and propose a potential algorithm for RRT indication and timing. Schematically, their proposal is that unless severe complications related to AKI occur (e.g., hyperkalemia, severe metabolic acidosis, severe fluid overload with pulmonary edema, or neurological symptoms associated with uremic encephalopathy), RRT should be postponed. The use of this strategy did not change the mortality rate of general critically ill patients or those with severe septic shock, and was associated with reduced use of RRT, suggesting a benefit for the “delayed” initiation strategy. The authors should be commended for conducting these studies, which will surely impact the daily practice of ICU physicians. However, we would like to draw attention to a subset of patients who may not benefit from such a delay. Patients with acute brain injury and at risk for cerebral edema and elevated intracranial pressure (e.g., patients with brain trauma, severe stroke, subarachnoid hemorrhage, post–cardiac arrest, meningitis, hepatic encephalopathy, encephalitis, or other brain infections) frequently present with an increased brain volume and reduced brain compliance. The slow increase in serum osmolality related to increased concentrations of metabolites as a result of failing kidney function will have little impact. In addition to variations in cerebral blow flow and arterial pressure, initiation of RRT will induce a rapid osmotic shift due to a drop in serum osmolality, and the extent of this shift is mostly driven by initial urea levels. The osmotic shift will then cause an increase in brain volume secondary to the osmolar gradient, with potential catastrophic consequences such as severe intracranial hypertension and brain death. According to the Monro-Kellie doctrine, the intracranial space is a fixed volume inside the skull and the cerebral pressure–volume correlation is initially linear (compensation), becoming exponential (compliance is reduced after compensatory mechanisms have reached their limits), meaning that a small increase in volume will induce a major increase in intracranial pressure (the so-called Langfitt curve). Therefore, patients with brain injury are at high risk of reaching the right inflection point of the curve and developing severe intracranial hypertension. Several case reports and reviews have described these complications (4, 5). Even the use of recommended “soft” RRT methods, such as sustained or continuous low-efficiency dialysis for patients with AKI and brain injury (6) will hardly moderate this shift, which occurs within the first minutes of RRT. We suggest not using the delayed RRT initiation strategy in patients at risk for elevated intracranial pressure. We believe that the best strategy for RRT modalities and initiation in this subset of patients remains to be determined.