SPS3 Blood Pressure
(2013)Objective
To assess whether targeting a lower systolic blood pressure (<130 mm Hg) reduces recurrent stroke in patients with recent lacunar stroke compared to a higher target (130–149 mm Hg).
Study Summary
• A lower systolic blood pressure target (<130 mm Hg) did not significantly reduce recurrent stroke compared to a higher target (130-149 mm Hg) in patients with recent lacunar stroke; • The lower blood pressure target significantly reduced the risk of intracerebral hemorrhage.
Intervention
Antihypertensive treatment to maintain systolic BP <130 mm Hg vs. 130–149 mm Hg. BP managed by local physicians using a standardized protocol and automated monitors.
Inclusion Criteria
Patients ≥30 years old with recent symptomatic MRI-confirmed lacunar stroke within 180 days, without cardioembolic source or large artery disease.
Study Design
Arms: Lower-target BP (<130 mm Hg) vs. Higher-target BP (130–149 mm Hg)
Patients per Arm: Lower BP: 1501, Higher BP: 1519
Outcome
• Primary stroke outcome was non-significantly reduced (HR 0.81, 95% CI 0.64–1.03, p=0.08). <br>• Intracerebral hemorrhage was significantly lower in the lower BP group (HR 0.37, 95% CI 0.15–0.95, p=0.03)..
Bottom Line
Targeting a systolic blood pressure of less than 130 mm Hg in patients with recent lacunar stroke resulted in a non-significant reduction in the rate of recurrent stroke but a significant reduction in intracerebral hemorrhage. The authors conclude that this target is likely to be beneficial.
Major Points
- SPS3-BP is the ONLY randomized trial specifically testing blood pressure targets in lacunar stroke patients — the most common subtype of small vessel disease. It showed a STRONG trend favoring intensive BP lowering (<130 mmHg) with HR 0.81 (p=0.08), just missing significance.
- 3,020 patients with MRI-confirmed lacunar stroke (≤2.0 cm) within 180 days. Open-label BP targets: <130 mmHg vs 130–149 mmHg. Mean follow-up 3.7 years. Part of the SPS3 2×2 factorial design (other arm: DAPT vs aspirin alone).
- Primary outcome (all recurrent stroke): 2.25%/yr intensive vs 2.77%/yr standard (HR 0.81, 95% CI 0.64–1.03, p=0.08). The 19% relative reduction was clinically meaningful but did NOT reach statistical significance — largely because the event rate was lower than expected.
- Intracerebral hemorrhage was SIGNIFICANTLY reduced: HR 0.37 (95% CI 0.15–0.95, p=0.03) — a 63% reduction. This is the strongest signal that aggressive BP control prevents the hemorrhagic complications of small vessel disease.
- Ischemic stroke also trended lower (HR 0.84) — suggesting BP lowering prevents both hemorrhagic AND ischemic manifestations of cerebral small vessel disease, consistent with the shared pathophysiology of lipohyalinosis.
- Mean achieved SBP: 127 mmHg (intensive) vs 138 mmHg (standard) — an 11 mmHg separation. This modest gap (similar to ATACH-2's ~12 mmHg) may have limited the trial's ability to show significance.
- Treatment-related adverse events were REASSURING: hypotension requiring treatment in only 1.5% intensive vs 1.0% standard (p=0.20) — lower than in SPRINT (2.8% vs 1.7%), suggesting lacunar stroke patients tolerate aggressive BP lowering well.
- Together with SPRINT and ACCORD, SPS3-BP supports a target SBP <130 mmHg for secondary stroke prevention in lacunar disease. The 2021 AHA/ASA secondary prevention guidelines recommend <130/80 mmHg (Class IIb, Level B-R for lacunar stroke).
- The 2×2 factorial design with the antiplatelet arm showed NO interaction between BP and antiplatelet randomizations — allowing independent interpretation of each treatment effect.
- Open-label BP target assignment was a necessary limitation (cannot blind BP targets) but outcomes were adjudicated by a blinded central committee, minimizing ascertainment bias.
Study Design
- Study Type
- Randomised, open-label trial with a two-by-two factorial design.
- Randomization
- Yes
- Blinding
- Open-label for blood pressure targets, but outcome adjudication was blinded.
- Sample Size
- 3020
- Follow-up
- Mean of 3.7 years.
- Centers
- 81
- Countries
- North America, Latin America, Spain
Primary Outcome
Definition: Reduction in all stroke (ischemic and hemorrhagic).
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 2.77% per year (152 events) | 2.25% per year (125 events) | 0.81 (0.64-1.03) | 0.08 |
Limitations & Criticisms
- Borderline p-value of 0.08 — failed to reach conventional significance. The HR of 0.81 was clinically meaningful (19% RRR) but the trial was likely underpowered due to lower-than-expected event rates (2.5%/yr vs anticipated 4%/yr).
- Open-label BP target assignment — impossible to blind BP targets, but potential Hawthorne effects could have improved BP control in both groups, reducing the treatment difference.
- Only 11 mmHg mean SBP separation between groups (127 vs 138 mmHg) — modest contrast may have limited statistical power. Some standard-group patients achieved SBPs <130 mmHg, further reducing the between-group difference.
- Lacunar stroke confirmed by MRI but no systematic assessment of white matter disease severity, cerebral microbleeds, or other markers of small vessel disease burden — these could have modified treatment response.
- Enrollment up to 180 days after qualifying event — much later than typical acute stroke trials. Some patients may have already adapted their BP medications, potentially diluting the intervention effect.
- No protocolized antihypertensive regimen — drug choice was left to treating physicians, introducing variability. Specific drug classes (ACE inhibitors, ARBs) may have different effects on cerebral small vessel disease beyond BP lowering.
- The ICH reduction (HR 0.37, p=0.03) was based on a small number of events (6 vs 16) — while statistically significant, the absolute numbers are small and may not be replicated in larger trials.
- Latin American enrollment (31% Hispanic) — genetic variation in renin-angiotensin system and salt sensitivity may affect BP treatment response differently across populations.
- The companion antiplatelet arm was stopped early for harm — potentially affecting recruitment and follow-up for the BP arm, though no statistical interaction was detected.
Citation
Lancet 2013; 382: 507-15