Time Is Brain: What the Stroke Treatment Evidence Actually Shows

The short answer

The stroke reperfusion trials tell a consistent story: opening a blocked brain artery helps, and the help shrinks with every hour of delay. A pooled analysis of the major clot-dissolving drug trials found that the odds of a good recovery were highest when treatment started within 90 minutes and faded toward statistical uncertainty past roughly four and a half hours. Yet the same body of evidence shows that carefully selected patients can still benefit many hours later, when brain imaging reveals tissue that is starving but not yet dead. The 2026 American Heart Association and American Stroke Association guideline builds on both findings, treating speed as the default and imaging as the tool that widens the door for some patients. This article explains how those trials were designed and what the evidence actually supports. It is educational and not medical advice.

Why time became the organizing idea

The slogan "time is brain" is not a rhetorical flourish. It traces to a 2006 quantification by Jeffrey Saver in Stroke, which estimated that in a typical large-vessel ischemic stroke the brain loses about 1.9 million neurons per minute while blood flow is cut off, along with billions of synapses and miles of nerve fibers. Put another way, an untreated stroke ages the affected brain by years over the course of a few hours. Those numbers are model estimates, not measurements from a single patient, and later work has emphasized wide variability from person to person. Still, the direction is not in dispute, and it framed how every subsequent trial was designed: if tissue dies on a clock, then the interventions that restore flow should be tested against that clock.

What the thrombolysis trials were built to show

Intravenous thrombolysis dissolves the clot with a drug. The foundational trials, including NINDS, the ECASS series, and ATLANTIS, randomized patients to the drug or placebo within defined time windows and measured disability months later, typically using the modified Rankin Scale, a standard zero-to-six ordinal scale of function.

The most informative appraisal comes from the pooled analysis by Lees and colleagues, published in The Lancet in 2010, which combined individual patient data across several of these studies rather than relying on any single trial. It reported adjusted odds of a favorable outcome that declined step by step as treatment was delayed: strongly positive in the first 90 minutes, still positive but weaker from 90 to 180 minutes, weaker again from 180 to 270 minutes, and no longer statistically clear beyond about 270 minutes, or four and a half hours. This is the empirical backbone of the treatment window, and it rewards careful reading, because the pooled data describe a fading gradient of benefit rather than a wall at which the drug abruptly stops working. The 4.5-hour figure is a defensible cutoff drawn from where the confidence intervals cross into uncertainty, and it reflects a balance: thrombolysis also carries a risk of bleeding into the brain, and as expected benefit falls, that fixed risk weighs more heavily.

How imaging reopened the late window

For years the time window was also, in practice, an eligibility window for thrombectomy, the mechanical removal of a large clot with a catheter. Two 2018 trials changed that by selecting patients on physiology rather than the clock alone. DAWN, published in the New England Journal of Medicine, enrolled patients 6 to 24 hours after they were last known well, but only those with a mismatch between a severe clinical deficit and a still-small area of finished infarct on imaging. The idea is that a large clinical deficit paired with a small dead core implies a large volume of threatened but salvageable tissue. In DAWN, functional independence at 90 days reached roughly 49 percent with thrombectomy versus about 13 percent with standard care, a difference large enough that the trial stopped early. DEFUSE-3 applied a related logic in a 6-to-16-hour window, using perfusion imaging to estimate the mismatch between dead and dying tissue.

The lesson is subtle and important for evidence literacy. DAWN and DEFUSE-3 did not repeal "time is brain." They showed that elapsed time is a proxy for the thing that actually matters, which is how much brain remains salvageable, and that imaging can measure that directly in selected patients. Fast collateral circulation lets some people preserve tissue long after the average person would have lost it, and those are precisely the patients these trials identified.

What the 2026 guideline weighs

The 2026 AHA/ASA guideline for early management of acute ischemic stroke, published in Stroke in January 2026, integrates both principles. For eligible patients within 4.5 hours, it supports thrombolysis with either alteplase or tenecteplase and does not require advanced imaging to select them, reflecting the default that speed drives benefit in this window. It describes equipoise between the two drugs, meaning the evidence does not establish one as clearly superior. Beyond that window, the guideline turns to imaging: it addresses extended-window thrombolysis in appropriately selected patients and supports thrombectomy in selected patients out to 24 hours when imaging demonstrates salvageable tissue.

Reading a guideline well means separating what is established from what remains open. The authors themselves flag persisting gaps in knowledge, and equipoise between agents is an honest statement of uncertainty rather than a recommendation to treat the drugs as interchangeable in every case. The strength of a recommendation reflects both the size of the benefit and the quality of the evidence behind it, which is why a 4.5-hour, no-imaging pathway and a 24-hour, imaging-gated pathway can coexist without contradiction.

The through line

The stroke evidence rewards a habit worth generalizing: read the trial design before the headline. The number that made "time is brain" famous is a model estimate. The 4.5-hour window is a threshold pulled from a continuous decline in benefit. The 24-hour thrombectomy window does not loosen the case for urgency; it shifts the measurement from elapsed time to surviving tissue. Each claim is only as strong as the population that was studied and the outcome that was counted, and the 2026 guideline is most useful when read as a weighing of those specifics rather than a set of slogans.