PDF: ASPREE
(2018)Objective
Aspirin – To determine whether daily low-dose aspirin increases disability-free survival (survival free from dementia or persistent physical disability) in healthy elderly individuals.
Study Summary
• Associated with increased risk of major bleeding.
• Trial stopped early due to lack of efficacy.
Intervention
Randomized, double-blind, placebo-controlled trial of 100 mg enteric-coated aspirin daily vs. placebo in 19,114 community-dwelling older adults (≥70 years, or ≥65 if Black/Hispanic in the U.S.). Median follow-up 4.7 years. Primary outcome: composite of death, dementia, or persistent physical disability.
Inclusion Criteria
Healthy adults aged ≥70 (or ≥65 for Black and Hispanic participants in U.S.) without cardiovascular disease, dementia, or disability. MMSE ≥78 and independent in basic activities of daily living.
Study Design
Arms: Aspirin vs. Placebo
Patients per Arm: Aspirin: 9525; Placebo: 9589
Outcome
• Death from any cause: 12.7 vs. 11.1 per 1000 person-years; HR 1.14 (95% CI, 1.01–1.29)
• Dementia: HR 0.98 (95% CI, 0.83–1.15)
• Disability: HR 0.85 (95% CI, 0.70–1.03)
• Major hemorrhage: 3.8% vs. 2.8%; HR 1.38 (95% CI, 1.18–1.62; P<0.001)
Bottom Line
In 19,114 healthy elderly persons (median age 74) followed for a median of 4.7 years, daily 100 mg enteric-coated aspirin did not prolong disability-free survival compared with placebo (HR 1.01; 95% CI 0.92-1.11; P=0.79). Aspirin significantly increased major hemorrhage by 38% (3.8% vs 2.8%; HR 1.38; P<0.001), with an additional 2.4 serious bleeding events per 1000 person-years. All-cause mortality was numerically higher with aspirin (HR 1.14; 95% CI 1.01-1.29), driven largely by cancer deaths reported in companion paper.
Major Points
- No benefit on disability-free survival: aspirin 100 mg daily vs placebo, HR 1.01 (95% CI 0.92-1.11; P=0.79), rates 21.5 vs 21.2 per 1000 person-years.
- Significantly increased major hemorrhage: 3.8% vs 2.8% (HR 1.38; 95% CI 1.18-1.62; P<0.001), +2.4 serious bleeding events per 1000 person-years.
- No reduction in dementia: 6.7 vs 6.9 per 1000 person-years (HR 0.98; 95% CI 0.83-1.15).
- Numerically higher all-cause mortality with aspirin: 12.7 vs 11.1 per 1000 person-years (HR 1.14; 95% CI 1.01-1.29), unadjusted for multiple comparisons.
- Deaths dominated the primary endpoint (50%), followed by dementia (30%) and persistent physical disability (20%).
- Trend toward reduced persistent physical disability with aspirin: 4.9 vs 5.8 per 1000 person-years (HR 0.85; 95% CI 0.70-1.03) — not significant.
- Consistent null result across nearly all subgroups: no effect modification by sex, age, country, race, BMI, smoking, diabetes, hypertension, dyslipidemia, cancer history, or prior aspirin use.
- Trial stopped early for futility in June 2017 (median 4.7 years, IQR 3.6-5.7) after interim analysis showed similar primary endpoint rates.
- 19,114 participants aged ≥70 (≥65 for Black/Hispanic in US), median age 74, 56% female, from 50 centers in Australia and US. 4-week placebo run-in required ≥80% adherence.
- Landmark trial that, along with ARRIVE and ASCEND (published simultaneously in 2018), led to revised ACC/AHA guidelines against routine aspirin for primary prevention in adults ≥70.
Study Design
- Study Type
- Randomized, double-blind, placebo-controlled trial
- Randomization
- Yes
- Blinding
- Double-blind (participants, trial staff, investigators, GP associates all blinded). Block randomization stratified by trial center (US) or GP clinic (Australia) and age (65-79 or ≥80). 4-week placebo run-in with ≥80% adherence required.
- Sample Size
- 19114
- Follow-up
- Median 4.7 years (IQR 3.6-5.7). Stopped early June 2017 for futility.
- Centers
- 50
- Countries
- Australia, United States
Primary Outcome
Definition: Composite of death from any cause, dementia, or persistent physical disability (first occurrence)
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 914 events; 21.2 per 1000 person-years | 921 events; 21.5 per 1000 person-years | 1.01 (0.92-1.11) | 0.79 |
Limitations & Criticisms
- Relatively short duration (median 4.7 years) — may be too short for conditions with long latencies like Alzheimer's disease and cancer.
- Does not address benefit of earlier initiation (younger age) or longer duration of aspirin.
- Predominantly white population (91.3%) — limited applicability to Black, Hispanic, and other nonwhite groups.
- Does not address whether healthy older persons already taking aspirin should discontinue.
- Early termination for futility may have reduced power for secondary endpoints.
- Declining adherence: only 62-64% still taking study medication in final 12 months.
- Low prior aspirin use (11%) limits generalizability to that population.
- Secondary endpoints not adjusted for multiple comparisons — mortality signal (HR 1.14) should be interpreted cautiously.
Citation
N Engl J Med 2018;379:1499-508.