Search for a command to run...
Fluid overload is common after resuscitation in critically ill patients and is independently associated with unfavorable outcomes. Ultrafiltration (UF) during renal replacement therapy is a common strategy to reverse this condition. Nevertheless, the optimal timing, dose, and safety limits remain unclear. Observational studies suggest a U-shaped relationship between net UF rate and mortality, highlighting a narrow therapeutic window. Insufficient UF prolongs exposure to congestion, whereas excessive UF may precipitate hemodynamic instability and tissue hypoperfusion. This phenomenon, often labeled "UF intolerance", lacks a standardized definition and is frequently attributed to intravascular volume depletion alone, overlooking broader cardiovascular and autonomic mechanisms. This perspective synthesizes the physiological effects and cardiovascular response to ultrafiltration and reframes "ultrafiltration intolerance" as a multidimensional construct determined by four compensatory axes: vascular refilling, cardiac response, venoconstriction/venous capacitance, and systemic arteriolar resistance. We delineate how renal replacement therapy-related factors interact with underlying critical illness to impair these axes and precipitate tissue hypoperfusion. Building on this framework, we (i) argue for proactive, individualized ultrafiltration prescriptions; (ii) propose dynamic predictors that emulate fluid removal to reveal physiologic reserve (iii) outline pragmatic, bedside endotypes of ultrafiltration intolerance: preload dependence, cardiac dysfunction, decreased vascular tone, autonomic dysfunction, and impaired vascular refilling. We propose a holistic framework that emphasizes prevention over rescue and guides tailored ultrafiltration prescription based on endotypes. Ultrafiltration intolerance in critically ill patients is multifactorial and not solely a problem of volume depletion. A physiology-based, preventive, and personalized ultrafiltration strategy (anchored in dynamic testing, ultrasound-based congestion profiling, and perfusion monitoring) may improve hemodynamic tolerance, decongestion efficacy, and patient-centered outcomes. Prospective studies should validate these concepts to establish evidence-based protocols for personalized ultrafiltration in critical care.