The Hallway Consult — Issue #18: Septic Shock Fluids, ED Buprenorphine, and Late-Window Stroke

ARISE FLUIDS: The Fluid-vs-Vasopressor Debate Finally Has a Trial, and It's a Tie

How to resuscitate early septic shock has been argued for two decades without a definitive randomized trial to settle it. The conventional approach is to lead with fluid and start vasopressors when the fluid isn't working. That approach has never been formally tested against the alternative: restrict fluid from the start and push vasopressors earlier. ARISE FLUIDS, published June 11, 2026 in the New England Journal of Medicine, is the largest randomized trial to take the question on directly.

The investigators enrolled 1,000 adults presenting to EDs with septic shock and randomized them to one of two strategies for the first 6 to 24 hours: restricted fluids with early vasopressors, or larger fluid volumes with vasopressors started later. Over 24 hours, the restrictive arm received approximately 1,100 mL less IV fluid and started vasopressors about an hour sooner. After the intervention window, care followed usual practice at each site.

The results were striking in their similarity. Days alive and out of the hospital at 90 days were a median of 76 in both arms (P=1.00). Ninety-day mortality was 16.4% in the restricted-fluids group and 14.4% in the liberal-fluids group, not significantly different. Days at home and organ-support-free days were also similar.

The clearest difference was pulmonary edema: 5.0% in the liberal-fluids arm versus 0.6% in the restricted-fluids arm (RR 0.12, P less than 0.001). More than eight times the rate. This was an open-label trial, so ascertainment bias in the liberal arm is possible, but a gap that large is hard to dismiss as noise.

What ARISE FLUIDS tells you is that the volume itself may not be the critical variable. Both strategies worked equally well in aggregate. The harder question is which patients will actually respond to the next bolus and which will just accumulate fluid. That is one this trial was not designed to answer. That is where physiology-guided resuscitation, dynamic responsiveness testing, and the growing literature on harm from non-responsive fluid loading will take the conversation next.

Bottom line: In early septic shock, a fluid-restrictive strategy with early vasopressors produced the same 90-day outcomes as a liberal-fluid strategy with later vasopressors. The restricted arm had dramatically less pulmonary edema. Both approaches are defensible, but giving more fluid does not help and may flood lungs that were not going to respond anyway.

ARISE FLUIDS Investigators. Vasopressors or Fluids in Early Septic Shock. N Engl J Med. 2026.

ED INNOVATION: The Shot Is Now a Legitimate First-Line Option for ED Buprenorphine

The case for ED-initiated buprenorphine for opioid use disorder is now established. The question the ED INNOVATION trial asked is whether formulation matters: does a single 7-day extended-release injectable shot do better than the standard discharge prescription for daily sublingual buprenorphine?

The CTN-0099 trial enrolled adults across 29 EDs in the United States with untreated OUD and a Clinical Opiate Withdrawal Scale (COWS) score of 4 or higher. They were randomized to either a single 24 mg injection of 7-day extended-release buprenorphine or a 7-day prescription for daily sublingual buprenorphine at 16 mg/day. Results were published in JAMA in early 2026.

The primary outcome was engagement in OUD treatment at 7 days, and it was nearly identical: 40.5% in the injectable group versus 38.5% in the sublingual group (adjusted difference 1.6%; 95% CI -2.8% to 6.0%). At 30 days, engagement remained comparable: 43.8% vs 44.9%. The trial was not powered to show superiority and did not find it.

Where the injectable arm pulled ahead was in secondary outcomes. Patients who received the injection reported lower craving scores at 7 days, fewer days of illicit opioid use in the past week, and higher treatment satisfaction scores. Precipitated withdrawal, a meaningful concern given that 76% of participants were fentanyl-positive, occurred in less than 1% of patients in either group. That finding alone deserves attention: the 7-day XR formulation can be initiated even with substantial fentanyl exposure and low withdrawal scores, without triggering the precipitated withdrawal that has made providers cautious.

The practical angle for the ED: the shot removes the adherence and diversion variables entirely. A patient who gets a 7-day injection on the way out the door does not need to remember to fill a prescription, fill it correctly, or take it every day, and there is nothing to divert. For the patient who is ambivalent, skeptical of their own follow-through, or living in chaos, the injection may offer an edge that the engagement numbers alone do not fully capture.

Bottom line: In a large multicenter RCT, 7-day injectable buprenorphine and sublingual buprenorphine produced equivalent OUD treatment engagement at 7 and 30 days. Secondary outcomes including craving scores, illicit opioid use, and patient satisfaction favored the injectable. Both formulations are safe even in fentanyl-predominant populations. The shot is now a legitimate first-line option for ED-initiated OUD treatment.

ED INNOVATION Trial Group. ED-Initiated Buprenorphine for OUD. JAMA. 2026.

OPTION: The Thrombolysis Window for Non-LVO Stroke Just Grew to 24 Hours

The standard thrombolysis window for ischemic stroke is 4.5 hours. For large vessel occlusion, mechanical thrombectomy has extended effective treatment to 24 hours in selected patients. But for non-LVO strokes, which represent the majority of ischemic strokes presenting to your ED, the question of whether thrombolysis has any benefit beyond 4.5 hours has been largely unanswered. The OPTION trial, presented at the International Stroke Conference in February 2026 and simultaneously published in JAMA, took that question directly.

Investigators enrolled 566 patients with acute non-LVO ischemic stroke across 48 centers in China. Patients were eligible if they presented between 4.5 and 24 hours from symptom onset and had imaging evidence of salvageable brain tissue on CT perfusion imaging. They were randomized to intravenous tenecteplase or standard medical care.

The primary outcome was excellent functional outcome defined as mRS 0 to 1 at 90 days, and it favored tenecteplase: 43.6% versus 34.2% in the control arm (adjusted RR 1.32; 95% CI 1.08 to 1.61). The number needed to treat for one additional excellent outcome was approximately 11.

The safety signal is real and cannot be glossed over. Symptomatic intracranial hemorrhage within 36 hours occurred in 2.8% of the tenecteplase group and 0% of controls (P=0.004). The number needed to harm was approximately 35. Despite that bleeding risk, there was no statistically significant difference in 90-day mortality: 5.0% in the tenecteplase arm versus 3.2% in controls (P=0.28).

The major external validity caveat is that all 566 patients were enrolled at Chinese centers. Whether these results translate to Western populations with different baseline stroke care, different imaging availability, and different patient demographics remains an open question. The trial also required CT perfusion imaging for eligibility, which limits applicability to centers where advanced imaging is part of the late-presentation stroke protocol. Most US EDs are not there yet.

That said, the signal is credible and the NNT is compelling. For EM clinicians who work in centers with perfusion imaging capability and a stroke neurology team engaged in late-window decision-making, OPTION is the trial that justifies the conversation.

Bottom line: In non-LVO ischemic stroke presenting 4.5 to 24 hours after onset, tenecteplase improved excellent functional outcomes at 90 days (43.6% vs 34.2%, NNT approximately 11) with a meaningful but manageable symptomatic hemorrhage risk (2.8% vs 0%, NNH approximately 35). The trial was conducted entirely in China and required CT perfusion imaging, limiting immediate US applicability, but the window is open and the conversation with stroke neurology is now evidence-based.

OPTION Trial Investigators. Tenecteplase for Non-LVO Ischemic Stroke at 4.5 to 24 Hours. JAMA. 2026.

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