The Hallway Consult — Issue #24: Capillary Refill–Guided Resuscitation, Tenecteplase Before Thrombectomy, and Steroids for Pneumonia

Capillary Refill–Guided Resuscitation in Septic Shock: ANDROMEDA-SHOCK-2

Quick refresher: capillary refill time (CRT) is how fast color returns to a fingertip after you press and release it — a fast, free bedside check many of us already do. The original ANDROMEDA-SHOCK trial hinted that guiding septic shock resuscitation off CRT might beat using serial lactate levels; the mortality numbers trended better but weren't statistically proven. ANDROMEDA-SHOCK-2 was built to get a real answer, testing a full 6-hour resuscitation protocol that used CRT, pulse pressure, and bedside echo to guide fluids, vasopressors, and dobutamine.

Researchers enrolled 1,501 adults with septic shock and randomized them to this CRT-guided protocol or usual care; 1,467 were analyzed. The CRT group got their capillary refill back to normal faster (86% vs 62% by 6 hours), got less IV fluid, and got more dobutamine. The trial measured three things together as one combined score: death at 28 days, how long patients needed life support (ventilator, dialysis, pressors), and length of hospital stay.

Here's the part worth slowing down on: the CRT group "won" that combined score (49% vs 42%, a real statistical difference) — but pulling it apart, the win came almost entirely from one piece: patients needed about one less day of dialysis, on average. The two numbers that actually matter to a patient — did they survive, and how long were they hospitalized — were identical between groups. Mortality was 26.5% vs 26.6%. Hospital length of stay didn't differ either. And this trial was unblinded, meaning the treatment team always knew which protocol a patient was on — which makes a soft, judgment-based outcome like "one less day of support" easier to nudge in the expected direction without anyone meaning to.

For the ED, this doesn't change anything about checking capillary refill — keep doing it, it's cheap and sensible. What it doesn't support is treating the full aggressive 6-hour CRT protocol as something that saves lives, because the one outcome that really counts — survival — didn't move at all.

Bottom line: In this 1,467-patient open-label trial, a 6-hour resuscitation protocol built around capillary refill time "won" on a combined score — but only because patients needed slightly less dialysis time, not because fewer of them died. 28-day mortality was unchanged (26.5% vs 26.6%), as was hospital length of stay. Checking capillary refill remains reasonable bedside practice; the full aggressive protocol is not proven to save lives.

ANDROMEDA-SHOCK-2 Investigators (Hernandez G, Ospina-Tascón GA, Kattan E, et al.). Personalized Hemodynamic Resuscitation Targeting Capillary Refill Time in Early Septic Shock: The ANDROMEDA-SHOCK-2 Randomized Clinical Trial. JAMA. 2025 Oct 29.

Tenecteplase Before Thrombectomy in the Extended Window: TNK-PLUS

Thrombolysis before thrombectomy already has a track record of improving reperfusion in the standard window. The open question has been whether that benefit extends into the 4.5-to-24-hour window, where more patients are now being treated thanks to perfusion-based imaging selection. TNK-PLUS was built to answer exactly that question.

Investigators at 40 EVT-capable centers in China randomized 391 patients with proximal middle cerebral artery occlusion presenting 4.5 to 24 hours after onset to IV tenecteplase (0.25 mg/kg, max 25 mg) before endovascular treatment, or EVT alone. The primary outcome was functional independence (modified Rankin Scale 0–2) at 90 days.

The result was flat. Functional independence occurred in 44% of the tenecteplase group versus 43% of the EVT-alone group — no meaningful difference. Ninety-day mortality was numerically lower with tenecteplase (12.7% vs 14.2%), but symptomatic intracranial hemorrhage within 36 hours was numerically higher (5.1% vs 2.6%). Neither difference reached a level that changes the risk-benefit calculus in tenecteplase's favor; if anything, the bleeding signal is the one worth watching.

This result lines up with how thrombolysis is understood to work mechanistically: its main value is dissolving clot before mechanical retrieval becomes available, which matters most when EVT is delayed or the clot burden is distal and hard to retrieve. In patients already selected for and proceeding straight to EVT in the extended window, pre-treating with tenecteplase doesn't appear to add benefit and comes with a bleeding cost. This is a negative trial, and it argues against routinely layering tenecteplase onto EVT once a patient is already 4.5-plus hours out and headed to the cath lab.

Bottom line: In this 391-patient RCT of extended-window stroke patients with proximal MCA occlusion, intravenous tenecteplase before EVT did not improve 90-day functional independence compared with EVT alone (44% vs 43%) and was associated with a higher rate of symptomatic intracranial hemorrhage (5.1% vs 2.6%). Routine pretreatment with tenecteplase before EVT in the 4.5-to-24-hour window is not supported by this trial.

Xiong Y, Che F, Wang H, et al. Intravenous Tenecteplase Prior to Endovascular Treatment for Ischemic Stroke at 4.5 to 24 Hours: The TNK-PLUS Randomized Clinical Trial. JAMA. Published online May 8, 2026.

Steroids for Community-Acquired Pneumonia: SONIA

The steroid-in-pneumonia question has mostly been answered for one population: critically ill, ICU-level CAP, where CAPE COD and similar trials showed a mortality benefit from hydrocortisone. What's been unclear is whether that benefit extends to the much larger population of patients admitted to a general medical ward with CAP that isn't yet ICU-level severe — and whether it holds in a low-resource setting where baseline CAP mortality is already higher than in wealthy countries.

SONIA (Steroids for Pneumonia in Adults in Kenya) was a pragmatic, open-label RCT across 18 Kenyan hospitals. Investigators randomized 2,180 adults hospitalized for CAP (1,091 usual care, 1,089 glucocorticoids) to usual care — a beta-lactam plus a macrolide per WHO guidelines — with or without 10 days of a low-dose oral glucocorticoid (dexamethasone, hydrocortisone, methylprednisolone, prednisolone, or prednisone, based on local availability). Patients with a separate clear indication for steroids, such as asthma or confirmed COVID-19, were excluded. Median age was 53, 46% were women, and roughly a third had an oxygen saturation under 90% on admission — meaningfully sick, but general-ward sick, not uniformly ICU-level.

The primary outcome, 30-day all-cause mortality, favored glucocorticoids: 22.6% versus 26.0% with usual care (HR 0.84, 95% CI 0.73–0.97, P = .02). Mortality at earlier timepoints (7, 14, 21 days) was consistent with the primary result. Adverse events were notable but manageable: hyperglycemia was far more common with steroids (16.6%), while acute kidney injury was more common with usual care (8.0%); pulmonary TB reactivation occurred in both arms at similar rates.

The caveat worth sitting with: this trial was conducted in a setting where baseline CAP mortality (26% in the control arm) is dramatically higher than what's typically seen in well-resourced U.S. EDs and wards, where in-hospital CAP mortality is usually in the single digits. That doesn't invalidate the result, but it does mean the absolute benefit seen here may not translate one-for-one to a lower-mortality population — the relative effect may hold, but the number needed to treat would likely be larger in a U.S. ward. All the drugs studied are standard formulary items in the U.S., so availability isn't the limiting factor; interpretation in a different-risk population is.

Bottom line: In this 2,180-patient pragmatic RCT of hospitalized (non-ICU) CAP patients in Kenya, 10 days of low-dose oral glucocorticoids reduced 30-day all-cause mortality compared with usual care (22.6% vs 26.0%, HR 0.84, P = .02), at the cost of more hyperglycemia. This extends the steroid-in-CAP benefit seen in ICU populations to general-ward patients, though the much higher baseline mortality in this cohort means the absolute benefit may not translate directly to lower-mortality U.S. settings.

Lucinde RK, Gathuri H, Mwaniki P, et al. A Pragmatic Trial of Glucocorticoids for Community-Acquired Pneumonia. N Engl J Med. 2025;393(22):2187-2197.

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