The Hallway Consult — Issue #23: Albumin in Septic Shock, TXA for Intracerebral Hemorrhage, and Peripheral Vasopressors

Albumin in Septic Shock: The ARISS Trial

Albumin has a logical appeal in sepsis. Hypoalbuminemia is nearly universal in septic shock, albumin is an endogenous oncotically active protein with anti-inflammatory properties, and the theoretical case for replacing it has been made for years. The SAFE and ALBIOS trials tested albumin in broader ICU and severe sepsis populations respectively, with mixed results. A 2026 multicenter RCT published in JAMA Network Open now tests the question in septic shock specifically — with a replacement strategy designed to actually maintain serum albumin levels, not just administer a fixed dose.

The ARISS trial enrolled 440 adult patients with septic shock across 23 German ICUs between October 2019 and May 2022. This was an open-label trial — blinding was not feasible given the nature of the intervention. Patients were randomized to receive a 60 g loading dose of 20% human albumin, followed by dosing titrated to maintain serum albumin ≥3.0 g/dL for up to 28 days, versus standard fluid management with crystalloids. The primary outcome was 90-day all-cause mortality.

The result: no difference. In the modified intention-to-treat population of 419 patients (210 albumin, 209 control), 90-day mortality was 43.3% versus 45.9% — a relative risk of 0.94 (95% CI, 0.76–1.17; P = .71). No secondary outcomes differed between groups either. Albumin administration appeared safe, with no excess adverse events in the treatment arm.

The important caveat: the trial was prematurely terminated due to low enrollment, ending with 440 of the planned 1,662 patients. That underpowering means the study cannot rule out a benefit — it was insufficiently powered to do so. What it can say is that the signal is not large and the effect is not obvious. For bedside practice, ARISS adds to a consistent pattern: albumin in sepsis sounds right but doesn't prove out in RCTs. The answer is probably not "albumin never helps anyone in septic shock" but rather "we have not yet found the subgroup or timing that reliably benefits." Until we do, routine albumin replacement in septic shock remains an unproven practice.

Bottom line: In this open-label RCT of 419 patients with septic shock, albumin replacement targeting serum albumin ≥3.0 g/dL for up to 28 days did not reduce 90-day mortality compared with crystalloid-only management (43.3% vs 45.9%; P = .71). The trial was underpowered due to early termination, so a modest benefit remains possible, but the data do not support routine albumin replacement in septic shock.

Welte T, Hoffmann M, et al. Albumin Replacement Therapy in Patients with Septic Shock (ARISS): A Multicenter Randomized Clinical Trial. JAMA Netw Open. 2026;9(2).

TXA for Spontaneous ICH: STOP-MSU

Tranexamic acid has an impressive track record in hemorrhage. In traumatic bleeding, CRASH-2 changed management worldwide. In surgical and postpartum hemorrhage, TXA works. The reasoning for applying it to spontaneous intracerebral hemorrhage follows directly: early hematoma expansion is a major driver of poor outcomes in ICH, and TXA inhibits fibrinolysis. If TXA can arrest clot breakdown and limit expansion within the first hours, it might improve neurologic outcomes. The STOP-MSU trial — Stopping Intracerebral Haemorrhage with Tranexamic Acid for Hyperacute presentation including Mobile Stroke Units — was designed to test this in an ultra-early window.

This was a phase 2, international, double-blind, placebo-controlled RCT enrolling 201 randomized participants across 24 hospitals and one mobile stroke unit in Australia, Finland, New Zealand, Taiwan, and Vietnam. Patients with spontaneous ICH presenting within 2 hours of symptom onset received either intravenous TXA (1 g over 10 minutes followed by 1 g over 8 hours) or matched placebo. The primary outcome was hematoma growth — defined as ≥33% relative growth or ≥6 mL absolute growth on CT at 24 hours.

TXA did not reduce hematoma growth. There was no significant difference in the primary outcome between groups — and if anything, the point estimate trended in the wrong direction: hematoma growth occurred in 43% of the TXA group versus 38% of the placebo group (aOR 1.31; 95% CI, 0.72–2.40; P = .37). Secondary endpoints — functional outcomes at 90 days, mortality at 7 and 90 days, and major thromboembolic events — also did not differ. TXA appeared safe, with no increase in thromboembolism, but it did not demonstrate the anticipated biological effect on hematoma expansion even when given within 2 hours of onset.

This matters for ED practice because mobile stroke units and rapid ICH recognition are expanding the window for intervention, and TXA is an attractive, available drug. The STOP-MSU result says: even in the most optimistic circumstances — ultra-early treatment, high-quality phase 2 design, mobile stroke unit delivery — TXA does not arrest ICH growth. It is more likely that the clot dynamics in spontaneous ICH differ fundamentally from traumatic or surgical hemorrhage. There is no role for routine TXA administration in spontaneous ICH based on current evidence.

Bottom line: In this phase 2 RCT of 201 patients with spontaneous ICH treated within 2 hours of symptom onset, intravenous TXA did not reduce hematoma growth at 24 hours compared with placebo. Functional outcomes, 90-day mortality, and thromboembolic events did not differ. There is no current evidence-based indication for TXA in spontaneous intracerebral hemorrhage.

Meretoja A, Churilov L, Strbian D, et al. Tranexamic acid versus placebo in individuals with intracerebral haemorrhage treated within 2 h of symptom onset (STOP-MSU): an international, double-blind, randomised, phase 2 trial. Lancet Neurol. 2024;23(6):577–587.

Peripheral Vasopressors in Septic Shock: The CLOVERS Secondary Analysis

Every clinician who has managed early septic shock has faced this situation: the patient needs norepinephrine, there is a large-bore peripheral IV in place, and the central line is 30 minutes away. Institutional protocol may say hold the vasopressor until the CVC is in. But what does the evidence actually say about the safety of peripheral vasopressor administration?

A secondary analysis of the CLOVERS trial — Crystalloid Liberal or Vasopressors Early Resuscitation in Sepsis, conducted across 60 US hospitals — directly addresses this. The main CLOVERS trial compared fluid-liberal versus vasopressor-first strategies in sepsis-induced hypotension; this secondary analysis compared patients who received vasopressors peripherally versus centrally within that trial. The analysis was published in August 2025.

The findings are clinically important. Peripheral vasopressor administration was associated with a 90-day mortality rate similar to central administration. Complications in the peripheral group were rare — occurring in only 0.6% of patients — compared with 3.7% in the central venous catheter group. There were no cases of skin ulceration or tissue injury from peripheral extravasation. Critically, peripheral administration was initiated a median of 2.3 hours earlier than central administration, a difference that likely reflects both the speed of peripheral access and the time-consuming nature of CVC insertion in a critically ill patient. Patients who received peripheral vasopressors also had lower rates of intubation (17.8% vs 40.5%) and received less fluid in the first 24 hours.

The limitation to contextualize: this is a secondary analysis of an RCT, not a standalone randomized comparison of routes. The groups were not equivalent — central line patients were sicker. That said, the complication data is direct evidence on safety, and the 2.3-hour delay to vasopressor initiation in the central group is a meaningful clinical observation.

For ED practice, the evidence supports what many clinicians are already doing in high-volume centers: starting norepinephrine peripherally while preparing for central access. There is no basis for routinely waiting 30 to 60 minutes to initiate vasopressors while a CVC is placed when a functioning peripheral IV is available. Peripheral vasopressors — using a proximal, large-bore IV, at an antecubital or more central site — is safe for the hours-long window of ED management.

Bottom line: A secondary analysis of the CLOVERS trial found that peripheral vasopressor administration in septic shock was associated with similar 90-day mortality and significantly fewer complications compared with central venous catheter administration (0.6% vs 3.7%), with vasopressors initiated a median 2.3 hours earlier via the peripheral route. There is no evidence-based reason to delay vasopressors until a central line is placed when a functional peripheral IV is available.

Munroe ES, et al. Peripheral Vasopressor Use in Early Sepsis-Induced Hypotension: A Secondary Analysis of the CLOVERS Trial. JAMA Netw Open. 2025;8(8):e2529148.

The Hallway Consult is built for EM clinicians who want the useful version of the literature. Forward it to a colleague if it helped.

www.hallwaymedicine.com

— The Hallway Consult team