SPRINT MIND
Effect of Intensive vs Standard Blood Pressure Control on Probable Dementia: A Randomized Clinical Trial
Year of Publication: 2019
Authors: The SPRINT MIND Investigators for the SPRINT Research Group
Journal: JAMA
Citation: JAMA 2019;321(6):553-561
Link: https://doi.org/10.1001/jama.2018.21442
PDF: https://jamanetwork.com/journals/jama/fullarticle/2723256
PDF: Clinical Question
Does intensive blood pressure control (<120 mmHg SBP) reduce the risk of dementia or mild cognitive impairment compared with standard control (<140 mmHg) in hypertensive adults without diabetes or stroke?
Study Overview
Objective
Intensive (<120 mmHg) vs standard (<140 mmHg) systolic blood pressure control — to evaluate effect on incident probable dementia and mild cognitive impairment in hypertensive adults.
Study Summary
- Intensive blood pressure control to SBP <120 mmHg did NOT significantly reduce probable dementia alone (HR 0.83; 95% CI 0.67-1.04; P=.10).
- Intensive control SIGNIFICANTLY reduced mild cognitive impairment by 19% (HR 0.81; 95% CI 0.69-0.95; P=.007).
- The composite of MCI or probable dementia was significantly reduced by 15% (HR 0.85; 95% CI 0.74-0.97; P=.01).
- SPRINT was terminated early for benefit on cardiovascular outcomes, limiting follow-up duration and statistical power for dementia.
- Adverse-event data from the intervention phase were reported previously in the parent SPRINT trial; no additional AE data were collected in the extended cognitive follow-up.
- First large randomized trial to show intensive BP control can reduce early cognitive impairment, supporting BP management as a modifiable dementia risk factor.
Intervention
Intensive BP treatment targeting SBP <120 mmHg vs standard treatment targeting SBP <140 mmHg. Drugs were chosen per guideline; no specific regimen mandated.
Patients per Arm
Intensive 4678; Standard 4683 (N=9361)
Bottom Line
In hypertensive adults without diabetes or prior stroke, intensive blood pressure control (<120 mmHg SBP) did NOT significantly reduce probable dementia (HR 0.83; 95% CI 0.67-1.04; P=.10) but did significantly reduce mild cognitive impairment (HR 0.81; 95% CI 0.69-0.95; P=.007) and the composite of MCI or probable dementia (HR 0.85; 95% CI 0.74-0.97; P=.01) over median 5.1 years. These findings support intensive BP control as a modifiable strategy to prevent early cognitive decline.
Major Points
- Randomized controlled trial of 9361 hypertensive adults ≥50 years
- Enrolled at 102 sites in US and Puerto Rico
- Randomized to intensive BP treatment (target <120 mmHg SBP) or standard treatment (target <140 mmHg)
- SPRINT trial was stopped early in August 2015 for primary cardiovascular benefit and all-cause mortality
- Cognitive outcomes follow-up continued through July 2018
- 8563 of 9361 (91.5%) completed ≥1 cognitive assessment
- Median intervention period: 3.34 years; median total follow-up: 5.11 years
- Primary cognitive outcome: adjudicated probable dementia
- Probable dementia: 149 (intensive) vs 176 (standard); 7.2 vs 8.6 per 1000 person-years
- HR for probable dementia: 0.83 (95% CI 0.67-1.04); P=.10 — NOT statistically significant
- MCI (secondary): 287 vs 353 events; 14.6 vs 18.3 per 1000 PY; HR 0.81 (95% CI 0.69-0.95); P=.007
- Composite MCI or probable dementia (secondary): 402 vs 469 events; 20.2 vs 24.1 per 1000 PY; HR 0.85 (95% CI 0.74-0.97); P=.01
- Early trial termination likely limited power for dementia-alone endpoint
- Baseline mean age 67.9 years; 3332/9361 (35.6%) women overall (Intensive 36.0%; Standard 35.2%)
- Intervention-phase adverse events were reported in prior SPRINT publications; no additional AE data were collected during extended cognitive follow-up
- First large randomized trial to show cognitive impairment prevention from BP control — shifted hypertension guidelines toward lower targets for cognitive protection in addition to CV protection
Design
Study Type: Randomized controlled multicenter trial (ancillary cognitive study of SPRINT)
Randomization: 1
Blinding: Open-label treatment; blinded adjudication of cognitive outcomes
Enrollment Period: Randomization began November 8, 2010; enrollment ended March 2013
Follow-up Duration: Median 5.11 years (SPRINT main trial stopped early Aug 20, 2015; final cognitive follow-up July 22, 2018)
Centers: 102
Countries: USA, Puerto Rico
Sample Size: 9361
Analyzed: 8563
Analysis: Intention-to-treat; Cox proportional hazards stratified by clinical site
Inclusion Criteria
- Age ≥50 years
- Hypertension: SBP 130-180 mmHg at screening visit
- ≥1 cardiovascular risk factor: clinical or subclinical cardiovascular disease, chronic kidney disease (eGFR <60 mL/min/1.73 m²), Framingham 10-year risk ≥15%, or age ≥75 years
Exclusion Criteria
- Diabetes mellitus
- History of stroke
- Nursing home residence
- Diagnosis of dementia (based on medical record review)
- Treatment with medications primarily used for dementia therapy
- Advanced kidney disease and symptomatic heart failure (noted in Limitations as populations not enrolled)
Baseline Characteristics
| Characteristic | Control | Active |
|---|---|---|
| N | 4683 | 4678 |
| Age mean (SD) | 67.9 (9.5) | 67.9 (9.4) |
| Aged ≥75 y | 1319 (28.2%) | 1317 (28.2%) |
| Women | 1648 (35.2%) | 1684 (36.0%) |
| Black race | 1493 (31.9%) | 1454 (31.1%) |
| Baseline SBP mean (SD), mm Hg | 139.7 (15.4) | 139.7 (15.8) |
| Baseline DBP mean (SD), mm Hg | 78.0 (12.0) | 78.2 (11.9) |
| History of CVD | 937 (20.0%) | 940 (20.1%) |
| eGFR <60 mL/min/1.73 m² | 1316 (28.3%) | 1329 (28.5%) |
| MoCA median (IQR) | 23.0 (20.0-26.0) | 23.0 (20.0-26.0) |
Arms
| Field | Control | Intensive BP control |
|---|---|---|
| N | 4683 | 4678 |
| Intervention | Antihypertensive therapy targeting SBP <140 mmHg per guideline | Antihypertensive therapy targeting SBP <120 mmHg per guideline |
| Duration | Median intervention 3.34 years | Median intervention 3.34 years |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| Adjudicated probable dementia during follow-up | Primary | 176 events / 20,378 person-years; 8.6 per 1000 PY | 149 events / 20,569 person-years; 7.2 per 1000 PY | 0.83 | P=.10 |
| Mild cognitive impairment | Secondary | 353 events / 19,281 PY; 18.3 per 1000 PY | 287 events / 19,690 PY; 14.6 per 1000 PY | HR 0.81 (95% CI 0.69-0.95) | P=.007 |
| Composite of MCI or probable dementia | Secondary | 469 events / 19,488 PY; 24.1 per 1000 PY | 402 events / 19,873 PY; 20.2 per 1000 PY | HR 0.85 (95% CI 0.74-0.97) | P=.01 |
| Note | Adverse | This paper reports no adverse event rates. Per the Methods and Duration of Follow-up sections, intervention-phase adverse events (including serious AEs, hypotension, syncope, electrolyte abnormalities, acute kidney injury, and injurious falls) were reported in prior SPRINT publications (Wright 2015 NEJM; Williamson 2016 JAMA; Sink 2018 JAGS). No additional AE data were collected during the extended cognitive follow-up visits. | |||
Subgroup Analysis
For probable dementia, no significant treatment × subgroup interactions across age (<75 vs ≥75), sex, race (black vs nonblack), history of CVD, chronic kidney disease, baseline SBP tertile, or orthostatic hypotension. For MCI, a nominally significant interaction with baseline CKD status was noted (P=.04), which would not remain significant after correction for multiple testing. No significant subgroup interactions for the composite of MCI or probable dementia.
Criticisms
- Primary dementia endpoint did not reach statistical significance — early termination reduced follow-up and power
- Excluded diabetics and prior stroke patients — limits generalizability to these high-risk groups
- Also did not enroll patients with advanced kidney disease or symptomatic heart failure
- Intensive BP control requires multiple drugs and monitoring — real-world implementation may differ
- No specific drug regimen mandated — heterogeneity in actual therapeutic exposure
- MoCA thresholds used to trigger extended testing may have underestimated MCI frequency (though non-differential by treatment)
- Baseline cognitive status was not adjudicated — cannot exclude influence of prevalent MCI at randomization
- Attenuation of between-group SBP difference during extended follow-up may have diluted effect estimates
Funding
NIH — National Heart, Lung, and Blood Institute (NHLBI), National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), National Institute on Aging (NIA), and National Institute of Neurological Disorders and Stroke (NINDS); contracts HHSN268200900040C, HHSN268200900046C, HHSN268200900047C, HHSN268200900048C, HHSN268200900049C, and Inter-Agency Agreement A-HL-13-002-001. Additional support from the US Department of Veterans Affairs (resources and use of facilities) and the NIA Intramural Research Program. Extended follow-up supported by NIH grant R01-AG055606. Study medications (azilsartan, and azilsartan combined with chlorthalidone) contributed by Takeda Pharmaceuticals International Inc. Additional support from the Kulynych Family Foundation and the Oristano Family Foundation. Support also acknowledged from multiple NCATS Clinical and Translational Science Awards (Case Western, Ohio State, Penn, Stanford, Tufts, Illinois, Pittsburgh, UT Southwestern, Utah, Vanderbilt, George Washington, UC Davis, Florida, Michigan, Tulane, and Wake Forest centers).
Based on: SPRINT MIND (JAMA, 2019)
Authors: The SPRINT MIND Investigators for the SPRINT Research Group
Citation: JAMA 2019;321(6):553-561
Content summarized and formatted by NeuroTrials.ai.