Sepsis Resuscitation Shifts Focus to Tissue Perfusion Over Fluid Rules

When a patient arrives in the intensive care unit with septic shock, one of the first and most consequential decisions clinicians face is how to restore blood flow to failing organs: push fluids early, start a vasopressor such as norepinephrine sooner, or some combination of both. For decades, early fluid administration has been a cornerstone of septic shock management, and the 2026 Surviving Sepsis Campaign still suggests giving 30 mL per kilogram of body weight within the first three hours. Yet that recommendation is deliberately conditional, supporting an individualized approach rather than a fixed-volume target, and it reflects a broader trend in critical care toward more judicious, goal-directed fluid resuscitation. Against that backdrop, a commentary published in Anesthesiology and Perioperative Science by Wissam Maroun argues that the field should stop framing the question as fluids versus vasopressors at all. The real endpoint, the author contends, is tissue perfusion, restored and maintained while minimizing the harm each treatment can itself cause.

The enthusiasm for giving vasopressors earlier did not emerge in a vacuum. It arose largely in response to mounting concern about fluid overload. Excessive fluid administration, which can lead to pulmonary edema and tissue congestion, has repeatedly been associated with worse outcomes, including increased mortality, prolonged mechanical ventilation, acute kidney injury, and longer intensive care unit stays. Smaller randomized and observational studies have suggested that starting norepinephrine earlier in septic shock may achieve more rapid hemodynamic stabilization while reducing the amount of fluid a patient receives, and some reports even indicated improved clinical outcomes. But the commentary is careful to note the caveats: those studies were limited by small sample sizes, single-center designs, and vulnerability to multiple confounders, so their findings should be interpreted cautiously.

The first major test of the early-vasopressor approach came from the CLOVERS trial, one of the earliest large multicenter studies to evaluate early vasopressor use in sepsis-induced hypotension. In that trial, patients were assigned to a restrictive strategy emphasizing earlier vasopressors and less fluid, or to a liberal strategy prioritizing additional intravenous fluid before any vasopressor escalation. Importantly, the protocol allowed clinicians to deviate from the assigned approach when necessary: vasopressors could still be started in the liberal group, and additional fluid could be given in the restrictive group for severe hypotension, elevated lactate, or other signs of inadequate perfusion. Despite meaningful separation in the amounts of fluid and vasopressor actually used after randomization, 90-day mortality was similar between the groups, and so were secondary outcomes including organ support and intensive care utilization.

CLOVERS, however, may not have fully determined what an early vasopressor protocol can do. Substantial resuscitation had already occurred before patients were randomized, and many participants had already received two to three liters of intravenous fluid by the time they entered the trial. That limitation set the stage for the recently published ARISE-FLUIDS trial, which was designed, in part, to address exactly this problem. Unlike its predecessor, ARISE-FLUIDS achieved earlier randomization and clearer separation between treatment groups by excluding patients who had already received more than two liters of fluid before enrollment. As a result, patients assigned to the early vasopressor strategy were genuinely less likely to have received substantial fluid resuscitation before the trial began.

ARISE-FLUIDS also made its intervention more pragmatic in a distinctive way. Clinicians managing patients in the trial were guided by signs of hypoperfusion, including severe hypotension, elevated lactate, delayed capillary refill, tachycardia, or reduced urine output. The trial therefore evaluated, in a real-world style, whether earlier vasopressors combined with less routine fluid administration could improve outcomes. Despite these methodological improvements over CLOVERS, the main findings were strikingly similar. The median number of days alive and out of the hospital at day 90 did not differ significantly between the treatment groups. Major adverse events were generally similar as well, with one notable exception: the restrictive, vasopressor-predominant group experienced a lower incidence of pulmonary edema.

Taken together, the two trials suggest that neither an early vasopressor-predominant strategy nor a liberal fluid strategy is universally superior for early sepsis resuscitation. The optimal approach, the commentary argues, likely depends on the individual patient’s hemodynamics, markers of tissue perfusion, and comorbidities. Both trials were fundamentally designed around restoring tissue perfusion, and although they employed predefined resuscitation protocols, those protocols allowed some flexibility, with treatment adjusted according to serial markers of perfusion and the overall clinical picture. In the author’s view, this flexibility permitted a degree of individualized resuscitation within each treatment arm, and it may partly explain why clinical outcomes did not differ significantly between the groups.

Why should tissue perfusion, rather than any single number on a monitor, be the guiding target? The overarching goal of hemodynamic management in early sepsis is to prevent tissue hypoxia by rapidly restoring adequate perfusion to the capillary beds of vital organs. Because tissue hypoxia is difficult to measure directly at the bedside, clinicians rely on surrogate markers. These include biochemical indicators such as lactate and clinical signs such as capillary refill time. But no single parameter accurately captures tissue perfusion in isolation, and each has critical limitations that can complicate interpretation. Equally important, normalizing macrohemodynamic variables such as mean arterial pressure does not necessarily mean that tissue perfusion has been restored. Macrohemodynamic targets should therefore not be interpreted on their own, but in the context of concurrent perfusion markers and how they evolve over time. Multiple indices should be assessed together and serially, with resuscitation adjusted according to their trajectory, the overall clinical picture, and the patient’s comorbidities.

The lack of a clear mortality difference between the two strategies does not mean the choice is unimportant, and it certainly does not mean either intervention is harmless. Excessive fluid administration may contribute to pulmonary edema, venous congestion, and tissue edema. Vasopressors, in turn, carry risks of excessive vasoconstriction, including peripheral ischemia. These risks do not preclude the use of either therapy during early resuscitation, but they underscore the need to select and titrate each intervention to the individual patient. The consequences of a given strategy can also vary considerably with the patient’s underlying physiology. In someone with reduced left ventricular systolic function, for example, additional fluid may carry a greater risk of pulmonary congestion and edema. In that setting, dynamic assessment of fluid responsiveness using tools such as passive leg raising or point-of-care ultrasonography can help determine whether more fluid is likely to improve stroke volume. If fluid responsiveness is absent, vasopressors can then be titrated as needed to support adequate tissue perfusion.

The practical message for clinicians is a reframing rather than a new protocol. A successful intervention, the commentary argues, should be defined by improvement in tissue perfusion at successive points in time, with the potential benefits of fluids or vasopressors weighed against their possible risks for each specific patient. The absence of a mortality advantage for one resuscitation strategy over another should not be read as implying that strategy choice is trivial. Instead, it reinforces a shift away from fixed resuscitation protocols and the pursuit of a single predetermined target, whether that target is a fixed fluid volume or an early vasopressor start time.

For a field that has long debated which intervention should come first in septic shock, the convergence of CLOVERS and ARISE-FLUIDS on similar outcomes, despite their different designs, is itself informative. Both large trials found no significant difference in their primary endpoints between restrictive, vasopressor-leaning approaches and liberal fluid approaches, and both embedded flexibility that allowed treatment to respond to each patient’s evolving perfusion status. The commentary concludes that the future of early sepsis resuscitation lies not in declaring a winner between fluids and vasopressors, but in restoring tissue perfusion while minimizing treatment-related harm, guided by serial assessment of multiple perfusion markers and a careful balance of both therapies according to the patient’s physiology, comorbidities, and evolving response.

Subject of Research: Individualized sepsis resuscitation guided by tissue perfusion rather than fixed fluid or vasopressor strategies

Article Title: The endpoint is tissue perfusion: moving beyond the fluids-versus-vasopressors debate

Article References: Maroun, W. (2026). The endpoint is tissue perfusion: moving beyond the fluids-versus-vasopressors debate. Anesthesiology and Perioperative Science, 4(1), Article 45. https://doi.org/10.1007/s44254-026-00193-y

Image Credits: AI Generated

DOI: 10.1007/s44254-026-00193-y

Keywords: sepsis, septic shock, tissue perfusion, fluid resuscitation, vasopressors, CLOVERS trial, ARISE-FLUIDS, norepinephrine, lactate, capillary refill time, Surviving Sepsis Campaign, critical care