The Hallway Consult — Issue #28: Thrombolysis for Retinal Artery Occlusion, CT Coronary Angiography After a Negative Workup, and IV Magnesium for Headache

TenCRAOS + THEIA: IV Thrombolysis for Central Retinal Artery Occlusion

Acute central retinal artery occlusion is a stroke of the eye. The patient shows up with sudden, painless, profound monocular vision loss, and there's no proven treatment. Open-label studies and meta-analyses suggested early IV thrombolysis might help, and some stroke teams started treating CRAO inside the 4.5-hour window the same way they treat a brain stroke. Two phase 3 RCTs have now compared a lytic directly against aspirin.

TenCRAOS (NEJM, 2026) was a double-blind, double-dummy trial at 16 sites in six countries. Adults with acute nonarteritic CRAO within 4.5 hours were randomized to IV tenecteplase 0.25 mg/kg or oral aspirin 300 mg. Only 78 patients were enrolled (40 vs. 38). Vision recovery at 30 days, defined as best corrected acuity of 20/100 or better, happened in 20% with tenecteplase versus 24% with aspirin (risk difference −3.7 points, 95% CI −22.0 to 14.7, P=0.69). Secondary visual outcomes didn't differ substantially. Adverse events were more common with tenecteplase, including one fatal intracranial hemorrhage, and the authors flagged serious safety concerns.

THEIA (Lancet Neurology, 2025) was a double-dummy French trial at 16 stroke-unit hospitals. It compared IV alteplase 0.9 mg/kg with aspirin 300 mg within 4.5 hours in patients with severe vision loss (worse than 20/400). It enrolled 70 patients, and 56 had primary endpoint data. At 1 month, an acuity improvement of at least 0.3 logMAR was seen in 66% with alteplase (19/29) versus 48% with aspirin (13/27), but this wasn't significant (adjusted OR 1.1, 95% CI 0.07 to 18.39, P=0.95). There was one asymptomatic intracranial hemorrhage and no symptomatic or major bleeding, and the investigators reported no alteplase-related safety concerns. They also said the trial was likely underpowered.

Read together, these two trials don't match perfectly, and they used different primary endpoints: TenCRAOS used an absolute acuity threshold, and THEIA used a degree of improvement. TenCRAOS was negative on its primary endpoint and had a safety signal with tenecteplase. THEIA leaned numerically toward alteplase, found no alteplase safety concerns, and was too small to say anything firm. Its confidence interval can't exclude a large benefit or real harm. Neither trial shows a significant benefit, so current RCT evidence doesn't support routine lytics for CRAO. The investigators themselves are calling for pooled patient-level analyses with upcoming trials.

Bottom line: Two small phase 3 RCTs (78 and 70 patients) found no significant improvement in vision with IV tenecteplase or alteplase compared with aspirin for acute nonarteritic CRAO within 4.5 hours, and TenCRAOS reported serious safety concerns including a fatal ICH. Routine IV thrombolysis for CRAO isn't supported by current RCT evidence. These trials don't change the rest of standard care: early ophthalmology and stroke involvement, screening for giant cell arteritis, and an embolic workup.

Ryan SJ, Jørstad ØK, Skjelland M, et al. A Randomized Trial of Tenecteplase in Acute Central Retinal Artery Occlusion. N Engl J Med. 2026;394(5):442-450.

Préterre C, Gaultier A, Obadia M, et al; THEIA collaborators. Intravenous alteplase versus oral aspirin for acute central retinal artery occlusion within 4·5 h of severe vision loss (THEIA). Lancet Neurol. 2025;24(11):909-919.

TARGET-CTCA: Outpatient CT Coronary Angiography After MI Is Ruled Out

You've ruled out MI with high-sensitivity troponin. The patient is going home, and the question is whether they need anything more. Some of these patients remain at risk for future cardiac events. Sending them for CT coronary angiography seems like a reasonable way to find and treat coronary disease before it causes a heart attack.

TARGET-CTCA tested exactly that at 14 UK hospitals. It didn't enroll everyone who ruled out. It enrolled patients whose MI had been excluded but whose peak hs-troponin was above 5 ng/L (detectable but not diagnostic of MI), a group at intermediate risk of future events. Patients were randomized to outpatient CTCA-guided care or standard care. Between 2019 and 2023, 3,170 patients were enrolled (median age 61, 30% female). The intervention was delivered: 92% of the CTCA group had the scan within 90 days, versus 2% of the standard care group. CTCA-related adverse events were rare (0.4%).

After a median of 3 years, MI or cardiac death occurred in 7.1% of the CTCA group and 7.3% of the standard care group (adjusted HR 0.95, 95% CI 0.73 to 1.23, P=0.71). The scans were done, but that didn't translate into fewer heart attacks or cardiac deaths.

The practical point for the ED is narrow. This doesn't mean CTCA is useless, and the abstract doesn't tell us how it applies to patients with ongoing symptoms or a specific clinical concern. It does mean a reflexive "everyone with a slightly elevated but non-diagnostic troponin gets a CTCA" pathway doesn't improve hard outcomes.

Bottom line: In 3,170 intermediate-risk patients whose MI had been ruled out with hs-troponin, routine outpatient CTCA-guided care did not reduce MI or cardiac death at a median of 3 years compared with standard care (7.1% vs. 7.3%, HR 0.95). Standard care did just as well. Save CTCA for patients with a specific clinical reason, not as a default step after a negative workup.

Lee KK, Wereski R, Lowe D, et al; TARGET-CTCA Investigators. Targeted Use of Computed Tomographic Coronary Angiography in Acute Chest Pain. N Engl J Med. Published online August 29, 2026. doi:10.1056/NEJMoa2608903.

IV Magnesium Added to Acetaminophen for Acute Headache in the ED

IV magnesium shows up on a lot of ED headache cocktails. This Annals trial asked a simple, practical question: if you've already given acetaminophen, does adding magnesium help?

This was a randomized, double-blind, placebo-controlled trial in Tunisian EDs. Adults with acute nontraumatic headache all got 1 g of oral acetaminophen, then were randomized to 2 g IV magnesium sulfate or a saline placebo, infused over more than 30 minutes. A total of 1,028 patients were analyzed (506 magnesium, 522 placebo). The primary outcome was at least a 30% reduction in pain score at 30 minutes.

Success was more common with magnesium: 78.9% versus 65.1% (difference 13.8 points, 95% CI 8 to 19). On secondary outcomes, fewer magnesium patients needed rescue analgesia (7.1% vs. 15.3%, an absolute 8-point drop), and satisfaction was higher (91.7% vs. 85.1%). But the authors are clear that the actual pain-score differences at every time point were below the accepted 1.3-point threshold for a clinically important difference. Adverse events were also more common with magnesium (15.4% vs. 11.1%), and the authors describe them as mild.

So the result is statistically positive but clinically modest. The halved need for rescue medication may be the most useful signal, but it's a secondary outcome. It's a single-country trial, and it compared magnesium plus acetaminophen with acetaminophen alone, not with the dopamine antagonist and NSAID regimens most US EDs use first.

Bottom line: In this 1,028-patient double-blind RCT, adding 2 g IV magnesium to oral acetaminophen for acute nontraumatic headache increased the proportion of patients with a 30% pain reduction at 30 minutes (78.9% vs. 65.1%) and roughly halved rescue analgesia, but pain-score differences fell below the threshold for clinical importance and mild side effects were more common. Magnesium is a reasonable, cheap adjunct. It isn't a reason to change your first-line headache regimen.

Messous S, Toumia M, Dhaoui R, et al. Intravenous Magnesium Sulfate for Acute Nontraumatic Headache in the Emergency Department. Ann Emerg Med. 2026;88(3):321-327.

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