Aiming for lower blood pressure after a stroke cut repeat events

People who had already had an ischaemic stroke or a transient ischaemic attack (TIA) had fewer heart attacks, strokes and related hospital stays when their doctors aimed for a top blood pressure reading below 120 mm Hg, compared with the usual goal of below 140. That is the main finding of OPTIMAL-STROKE, a trial of 4,368 people in Brazil presented at the American Heart Association (AHA) Scientific Sessions in Chicago on 7 November 2026 and published the same day in the New England Journal of Medicine.

Over a median of 3.9 years, 10.9% of people in the lower-target group had one of the trial's main events, against 13.5% with the standard target. The price was more kidney problems, mostly seen as a drop in kidney function on blood tests. It is the first large trial built specifically to test how low blood pressure should go in people who have survived a stroke.

This is a study about long-term prevention. It does not concern emergency treatment in the hours after a stroke, such as the clot-removal procedures we covered in MILD-MT and LATE-MT. The people in OPTIMAL-STROKE were clinically stable, and most joined about a year after their stroke.

Why has the right target been unclear?

High blood pressure is the biggest treatable cause of stroke, and almost everyone leaves hospital after a stroke on blood pressure tablets. What doctors have not known is how far to push it down.

The best-known trial of a very low target, SPRINT, showed in 2015 that aiming below 120 cut heart events in high-risk Americans. It left out people who had had a stroke. "Intuitively, lowering blood pressure prevents stroke and cardiovascular events, but it's never been studied in the poststroke, post-TIA population," Christopher Kramer of the University of Virginia, immediate past president of the American College of Cardiology, told TCTMD.

Some evidence came from ESPRIT, a trial of 11,255 high-risk adults in China published in The Lancet in 2024. It found fewer cardiovascular events with a target below 120 (hazard ratio 0.88), and the effect looked similar in people with and without a past stroke. But stroke survivors were only part of that trial. Doctors have also worried that pushing pressure too low could starve the brain of blood, especially in people with narrowed arteries in the neck or head.

Cheryl Bushnell, a stroke neurologist at Wake Forest University School of Medicine, spoke to that worry at the AHA media briefing. Patients often leave hospital with a narrowed carotid or middle cerebral artery, she said, and "everybody's nervous about lowering their blood pressure too much." In her view the new trial "abolishes some of the concerns" neurologists and family doctors have had about blood flow to the brain. She also pointed out that even small drops matter at a population level, citing data suggesting that a 5 mm Hg fall in blood pressure can cut stroke risk by about 13%.

Who ran OPTIMAL-STROKE and who took part?

The trial (NCT04036409) was led from Hospital Israelita Albert Einstein in São Paulo and funded by the hospital and the Brazilian Ministry of Health, through a public programme that supports research for the national health system, according to Metrópoles. No drug company sponsored it. The principal investigator is Gisele Sampaio Silva, professor of neurology at the Federal University of São Paulo and head of neurology clinical research at Einstein. The senior author, Otavio Berwanger, is at the George Institute for Global Health UK and Imperial College London.

Between August 2019 and August 2022, 30 sites across Brazil enrolled adults with a past ischaemic stroke or TIA and a systolic pressure (the top number) of 130 to 180 mm Hg. They had to have been clinically stable for at least 48 hours. People with severe disability, known secondary causes of high blood pressure, or very poor kidney function were not eligible, according to HCPLive's report of the paper.

Of the 4,368 people enrolled, 2,172 were assigned to the intensive target and 2,196 to the standard one. Their average age was 64, and 47.9% were women. About 92% had had a stroke and 8% a TIA, and their average starting blood pressure was 146.7/81.6 mm Hg. Roughly 30% joined within 120 days of their stroke or TIA.

Patients and doctors knew which target each person had, which makes this an open-label trial. A committee that did not know the group assignments judged every event. Our guide to open-label and blinded trials explains what that design can and cannot rule out.

How much lower did blood pressure go?

After the first three months, the average systolic pressure was 122.0 mm Hg in the intensive group and 135.2 mm Hg in the standard group. The average gap was 13.1 mm Hg (95% CI, 12.5 to 13.8). The median in the intensive group was 119.6.

Getting there took more medicine. People in the intensive group took an average of three classes of blood pressure drug, against about two in the standard group, Berwanger said at an AHA media briefing. By the last visit, half the intensive group were on a calcium channel blocker, against about a third of the standard group.

Even in a carefully run trial, many people did not hit the lower goal. The share of the intensive group below 120 rose from 55% at one year to 62.4% at two years and 68.9% at three, TCTMD reported. Daichi Shimbo of Columbia University, who discussed the trial at the meeting, said it is "really hard to get the systolic blood pressure less than 120," and probably harder still in everyday practice.

What did the trial find?

The main outcome combined death from cardiovascular causes, non-fatal heart attack, non-fatal stroke, and hospital admission for unstable angina or heart failure.

Main outcome and blood pressure figures from the NEJM abstract. Other rows from TCTMD, HCPLive and Portal Afya reports of the full paper.

The main result was statistically significant (p=0.007). In event rates, that is 3.25 against 4.11 events per 100 person-years.

The trial tested its secondary outcomes in a fixed order. The second one, cardiovascular death, heart attack or stroke, was also significant. The third, cardiovascular death on its own, showed no difference, so formal testing stopped there. That means the stroke results, which point to about a quarter fewer strokes, count as supporting evidence. They are not separate proof.

By our own arithmetic, the absolute difference in the main outcome is 2.6 percentage points, or about 26 fewer people with an event for every 1,000 treated for around four years. Shimbo put the number needed to treat at 38. Headlines quoting a 21% reduction are using the hazard ratio, which takes account of when events happened. A simple comparison of the two percentages gives a relative reduction closer to 19%, again by our calculation.

The benefit looked consistent across age groups. People whose event was a TIA made up only 8% of the trial, and their result was too uncertain to read either way. Their hazard ratio was 1.34, with a confidence interval running from 0.68 to 2.64.

What were the downsides?

The clearest harm involved the kidneys. A combined kidney outcome, defined as a sustained fall of at least 50% in estimated kidney function (eGFR) or progression to kidney failure, affected 3.9% of the intensive group and 2.2% of the standard group. Most of these events were falls in eGFR on blood tests. Six people in the intensive group and three in the standard group started long-term dialysis.

Shimbo set the two sides out plainly. "If you treat a thousand patients for a median of almost 4 years, 26 will benefit but 17 will be harmed," he said, while noting that the benefit was fewer major events and the harm was mostly "a creatinine bump." He put the number needed to harm at 59. Sampaio Silva told TCTMD the kidney effect was "mostly a laboratory effect," but said it was a warning to check kidney function more often.

Low blood sodium (below 130 mmol/L) was more common with the lower target, at 3.3% against 2.2%. Falls that caused injury were less common, at 1.0% against 1.7%. Serious adverse events overall were almost identical, at 15.1% and 15.3%.

A companion trial in diabetes found no clear benefit

The same Brazilian team ran a sister trial, OPTIMAL-DIABETES, with the same targets and the same main outcome in 9,476 people with type 2 diabetes and high cardiovascular risk. It was published in NEJM on the same day. There, aiming below 120 did not significantly reduce major events: 6.0% against 6.7%, hazard ratio 0.90 (95% CI, 0.77 to 1.06; p=0.20).

The two results together suggest that the case for a very low target may depend on who the patient is. People who have already had a stroke start at high risk of another event, so there is more to gain.

What are the trial's limits?

The investigators named three. The trial was open-label, so doctors and patients knew the targets, which can affect how carefully people are monitored. All patients were in Brazil, so the results may not carry over exactly to other health systems. And the trial had less statistical power than first planned, after a blinded revision of the sample size.

There is also the gap between target and result. People in the intensive group averaged 122, which led Kramer to suggest that getting stroke survivors "under 130, probably under 125, is probably ideal," with future guidelines deciding the exact number.

What does this mean for stroke survivors in the UK?

The 2023 National Clinical Guideline for Stroke for the UK and Ireland already recommends a clinic systolic target below 130 mm Hg for most people after a stroke or TIA, or below 125 on home readings. It sets a higher target, 140 to 150, for people with severe narrowing of both carotid arteries in the neck. OPTIMAL-STROKE did not change any guideline overnight. It will feed into the next round of updates in the UK, Europe and the US.

The trial does not mean anyone should cut their own doses or add tablets. Lowering blood pressure further means more medicine, more blood tests and closer follow-up, especially of kidney function and sodium levels. Those trade-offs need a conversation with the person who manages your blood pressure.

Questions to ask at your next review

If you have had a stroke or TIA and take blood pressure medicine, you could ask:

• What is my target blood pressure, and am I reaching it at home and in clinic?
• Would a lower target be right for me, given my kidneys and other conditions?
• How often should my kidney function and sodium be checked?
• Do I have any narrowing of the arteries in my neck that changes my target?
• Would a home blood pressure monitor help, and how should I use it?

This article is for general information and is not medical advice. Do not change your blood pressure medicines without talking to your doctor or stroke team.

What remains uncertain?

OPTIMAL-STROKE gives the clearest evidence yet that a lower blood pressure goal can prevent further heart attacks and strokes in people who have already had an ischaemic stroke. About 26 people in every 1,000 avoided a major event over four years, while about 17 in 1,000 had a worsening in kidney blood tests. Whether the same balance holds outside Brazil, in people who struggle to take three or more drugs, and over longer periods, is still open. So is the best target number, which guideline committees will now have to weigh.