SPRINT: What Intensive Blood-Pressure Treatment Proved

SPRINT changed hypertension care by testing a treatment strategy, not merely comparing two numbers. Investigators randomized 9,361 adults at elevated cardiovascular risk to a systolic blood-pressure target below 120 mm Hg or below 140 mm Hg. Participants had standardized measurements, frequent visits, medication adjustment, laboratory surveillance, and structured adverse-event assessment.

The intensive strategy reduced a composite of major cardiovascular events and reduced all-cause mortality during the trial. It also increased hypotension, syncope, electrolyte abnormalities, and acute kidney injury or failure. The finding applies most directly to people who resemble those enrolled, measured and treated in a comparable way. It does not tell you to drive a casual home reading below 120.

The trial question#

SPRINT asked whether treating toward a lower systolic target would improve outcomes in adults with hypertension and increased cardiovascular risk. Eligible participants were at least 50 years old, had screening systolic pressure in a protocol-defined range, and had one or more risk features such as established cardiovascular disease other than stroke, chronic kidney disease, elevated calculated risk, or age 75 or older.

The trial deliberately did not include people with diabetes or a previous stroke. It also excluded several conditions that complicated safety or interpretation, including symptomatic heart failure or markedly reduced ejection fraction and a standing systolic pressure below 110 mm Hg, and participants living in nursing homes were not represented in the same way as ambulatory volunteers attending trial visits. Those criteria matter because benefit and harm depend on baseline cardiovascular risk, competing illness, frailty, orthostatic symptoms, kidney function, medication burden, and the ability to follow the monitoring plan. A large trial can be internally strong and still be conditional in how it applies to you.

Targets are treatment strategies, not achieved averages#

The groups were assigned targets below 120 and below 140 mm Hg. A target is an instruction for medication titration, not a claim that every reading or participant will stay under it. During follow-up, the mean systolic pressure separated substantially, approximately 121 to 122 mm Hg in the intensive group and about 135 to 136 mm Hg in the standard group during the main intervention.

Intensive participants typically needed more medications and more active adjustment. Protocol staff monitored electrolytes, creatinine, symptoms, and standing pressure and could modify treatment for safety. Medication classes and timing were selected within a structured algorithm, with clinical judgment.

This distinction prevents a common error. The trial did not compare a naturally occurring pressure of 119 with 139. It compared packages of care intended to reach different goals. Benefits and adverse events belong to those packages.

The cardiovascular and mortality results#

The original report was released after the intervention stopped early because the monitoring process found clear benefit. The primary composite included myocardial infarction, other acute coronary syndrome, stroke, acute decompensated heart failure, or cardiovascular death.

The 2021 final report incorporated additional adjudicated events through the intervention and described post-trial observation, and during a median 3.33 years of trial follow-up, the primary outcome occurred at 1.77% per year in the intensive group and 2.40% per year in the standard group. The hazard ratio was 0.73, with a 95% confidence interval from 0.63 to 0.86.

All-cause mortality occurred at 1.06% per year versus 1.41% per year, with a hazard ratio of 0.75 and confidence interval from 0.61 to 0.92. Relative effects convey proportional change; absolute rates convey how many events occurred over time. If your baseline risk is lower, the absolute benefit can be smaller even when a similar relative effect applies.

Early stopping can sometimes make a result look larger, especially when few events have accrued. SPRINT had prespecified monitoring, many sites, adjudicated outcomes, and consistent major findings in final analyses. Even so, follow-up was shorter than originally planned, which matters for long-latency outcomes and durability.

Harms were part of the same randomized comparison#

Serious adverse events overall were not simply higher for every category, but specific events occurred more often with intensive treatment. These included hypotension, syncope, electrolyte abnormalities, and acute kidney injury or acute renal failure. The original report did not find a significant increase in injurious falls. A lack of increase in that outcome does not mean your dizziness or fall risk can be ignored.

Kidney findings require nuance. An acute creatinine rise can reflect altered filtration pressure as well as structural injury, and some changes reverse. The trial still recorded more qualifying acute kidney events with intensive treatment. Among people without chronic kidney disease at baseline, intensive treatment was also associated with more incident low estimated filtration by trial criteria. Cardiovascular benefit and renal safety signals should be considered together rather than labeling one side the “real” result.

Monitoring in practice includes symptoms, sitting and standing pressure when relevant, kidney function, sodium, potassium, medication interactions, dehydration risk, and falls. Your tolerance can change during acute illness, heat, weight loss, or the addition of another medicine. SPRINT does not support unsupervised dose escalation.

Measurement protocol changes how the number travels#

SPRINT used an automated Omron 907XL device, an appropriate cuff, positioning, a five-minute rest period, and an average of three readings taken about one minute apart. The protocol minimized conversation and other common sources of error.

The trial has often been described as using “unattended” automated office blood pressure. Site practices varied. Some staff left the room during rest and measurement, while others remained for part or all of the process; analyses and surveys found that staff presence was not uniform, so it is inaccurate to define the whole trial by a single unattended method.

The transferable lesson is standardized technique. A rushed single reading taken after you have walked in, talked, had coffee, needed the bathroom, or been fitted with the wrong cuff is not comparable with the trial average. Your home readings have their own validated protocol too. How blood pressure should be measured explains preparation, cuff choice, repeated readings, and home confirmation.

Measurement difference does not yield one universal conversion factor. Office devices, staff behavior, room conditions, patient state, and home technique vary. Clinicians should make decisions from reliable repeated readings rather than subtracting an assumed number from every clinic value.

What SPRINT did not establish#

Because diabetes and prior stroke were exclusions, SPRINT does not directly establish the same below-120 strategy for those groups. Other trials and guidelines inform their targets. It also does not settle treatment for pregnancy, children, acute stroke, hospitalized instability, or severe symptomatic orthostatic hypotension.

Trial volunteers received frequent follow-up and free protocol medications, and the clinical centers had titration support. The service you actually attend may have fewer visits, fragmented records, cost barriers, and less timely laboratory testing. Implementation should reproduce the safety processes as well as the target discussion.

Subgroup analyses can suggest consistency but are not separate trials. A nonsignificant interaction does not prove identical benefit in every age, sex, kidney, race, or baseline-pressure subgroup. Conversely, a nominal subgroup difference may be chance when many comparisons are tested. Applying it to you starts with the overall result and then considers biological plausibility, precision, and external evidence.

SPRINT MIND and cognition#

SPRINT MIND extended the question to mild cognitive impairment and probable dementia, and the original cognitive analysis did not find a statistically significant reduction in probable dementia alone, partly because fewer cases accrued than planned after early trial stopping. It did find lower rates of mild cognitive impairment and of the composite of mild cognitive impairment or probable dementia with intensive treatment.

A 2025 long-term analysis followed cognitive outcomes for a median around seven years, although re-contact and ascertainment were incomplete. Probable dementia occurred at 8.5 versus 10.2 per 1,000 person-years, with a hazard ratio of 0.86 and confidence interval from 0.72 to 1.02. Because the interval included 1, the result did not establish a reduction in probable dementia alone.

Mild cognitive impairment had a hazard ratio of 0.87 with a confidence interval reaching 1.00, and the composite of mild cognitive impairment or probable dementia had a hazard ratio of 0.89 with a confidence interval from 0.79 to 0.99. Those results support a cognitive-impairment benefit while requiring care about missing follow-up, outcome hierarchy, and the difference between MCI and dementia.

No randomized result justifies promising you dementia prevention. Blood-pressure treatment is one component of vascular risk management, and cognition has many causes.

How the 2025 guideline uses the evidence#

The 2025 ACC and AHA high blood pressure guideline states an overarching treatment goal below 130/80 mm Hg for adults, with additional considerations. It uses overall cardiovascular risk, existing disease, kidney disease, pregnancy status, age, frailty, adverse effects, and other clinical factors to guide when and how treatment begins.

That current target is not a contradiction of SPRINT. Guidelines integrate multiple trials, measurement realities, populations, feasibility, and harm. A randomized comparison below 120 versus below 140 in a selected high-risk group can support moving general practice away from the older 140 target without making below 120 obligatory for everyone. The safest translation is a sequence: confirm the pressure accurately, assess cardiovascular and kidney risk, identify secondary causes or orthostatic symptoms when indicated, choose lifestyle and medicine interventions, set an individualized target within current guidance, and monitor response and harms.

Reading SPRINT without a slogan#

“Lower is better” omits the stopping point, population, measurement, and monitoring. “SPRINT readings were special and cannot apply” discards a strong randomized result. Both slogans fail.

A complete interpretation says that intensive systolic treatment produced substantial cardiovascular and mortality benefit in ambulatory high-risk adults without diabetes or prior stroke, when delivered with standardized measurement and safety follow-up. It also increased several adverse events. Current guidelines translate that evidence into a broader goal below 130/80 with individual considerations.

That formulation is less dramatic but more actionable. It preserves the trial's benefit without turning a protocol target into a home dosing instruction.

References#

1. SPRINT primary report, NEJM 2015
2. SPRINT final report, NEJM 2021
3. SPRINT blood-pressure measurement methods
4. Long-term SPRINT cognitive outcomes, Neurology 2025
5. 2025 ACC and AHA high blood pressure guideline
6. NHLBI BioLINCC SPRINT study resources