In initially healthy older adults, low-dose aspirin showed no long-term cardiovascular benefit and increased major bleeding; post-trial, MACE was higher in those originally assigned aspirin.

Background

Guidelines advise against routine initiation of low-dose aspirin for primary prevention in older adults due to small cardiovascular benefits and increased bleeding. The ASPREE randomized trial previously showed no reduction in cardiovascular events and increased bleeding over a median 4.7 years. This extended follow-up (ASPREE-XT) evaluated long-term and post-trial effects on major adverse cardiovascular events (MACE) and major haemorrhage.

Patients

• N = 19,114 community-dwelling adults
• Age: ≥70 years (≥65 years for US minorities); median enrollment age 74
• Key exclusions: prior cardiovascular events, dementia, or independence-limiting physical disability
• Geography: Australia and USA

Intervention

Daily low-dose (100 mg) enteric-coated aspirin for the duration of the in-trial period (median 4.7 years); participants were instructed to stop study drug at trial end.

Control

Matching placebo.

Outcome

• Primary outcome for this analysis: Major adverse cardiovascular events (MACE) = coronary heart disease death, non-fatal myocardial infarction, or fatal/non-fatal ischaemic stroke.
• Key secondary outcomes: Major haemorrhage (intracranial or clinically significant extracranial bleeding), all-cause death. Prespecified bleeding subtypes included intracranial and gastrointestinal bleeding.

Study Design

• Multicenter, randomized, double-blind, placebo-controlled trial (ASPREE), followed by extended observational follow-up (ASPREE-XT) with blinded adjudication of events.
• In-trial adherence assessed by pill counts and medication review; post-trial aspirin use captured by questionnaires and concomitant medication lists.
• Analyses included intention-to-treat for long-term and post-trial effects; sensitivity analyses used inverse probability weighting and as-treated approaches.

Level of Evidence

• Level I: Randomized controlled trial evidence for in-trial effects.
• Observational (post-trial): Non-randomized follow-up assessing potential legacy or rebound effects.

Follow up period

• In-trial: median 4.7 years
• Post-trial: median 4.3 years
• Total (long-term): median 8.3 years

Results

Primary outcome: MACE

• Long-term (in-trial + post-trial): No benefit with aspirin vs placebo (HR 1.04; 95% CI 0.94–1.15).
• In-trial: Trend toward lower MACE with aspirin (HR 0.89; 95% CI 0.77–1.03).
• Post-trial: Higher MACE in those originally randomized to aspirin (HR 1.18; 95% CI 1.02–1.37). The increase persisted beyond the first post-trial year.
• NNT/NNH: Not estimable from hazard ratios without absolute risk differences.

Secondary outcomes

• Major haemorrhage (composite):
  • Long-term: Higher with aspirin (HR 1.24; 95% CI 1.10–1.39).
  • In-trial: Higher with aspirin (HR 1.38; 95% CI 1.18–1.62).
  • Post-trial: No clear difference (HR 1.08; 95% CI 0.91–1.29).
  • Approximate NNH (long-term): ~71 over 8.3 years (derived from reported rates per 1000 person-years; approximation).
• Myocardial infarction:
  • Long-term: HR 1.09; 95% CI 0.94–1.26.
  • Post-trial: HR 1.25; 95% CI 1.01–1.54.
• Ischaemic stroke:
  • Long-term: HR 1.06; 95% CI 0.91–1.23.
  • Post-trial: HR 1.20; 95% CI 0.97–1.48.
• Bleeding subtypes (long-term): Upper GI bleeding HR 1.43 (95% CI 1.13–1.81); intracranial bleeding HR 1.13 (95% CI 0.92–1.39); other sites HR 1.25 (95% CI 1.01–1.54).
• All-cause death (long-term): HR 1.06; 95% CI 0.99–1.14.

Limitations

• Post-trial comparisons are observational and susceptible to selection bias and confounding; inverse probability weighting used but residual bias is possible.
• Aspirin exposure post-trial was not randomized and was captured at annual intervals (timing of changes between visits unknown).
• Multiple outcomes tested without multiplicity adjustment.
• Findings apply to initially healthy older adults; generalizability to younger or higher-risk groups may differ.
• NNT/NNH largely not directly calculable from time-to-event (HR) data without absolute risk differences.

Funding

• National Institute on Aging and National Cancer Institute (USA): U01AG029824, U19AG062682.
• National Health and Medical Research Council of Australia: 334047, 1127060.
• Monash University and Victorian Cancer Agency.
• Bayer AG supplied study drug (aspirin and placebo) with no role in study conduct or reporting.

Citation

Wolfe R, Broder JC, Zhou Z, Murray AM, Ryan J, Chan AT, et al. Aspirin, cardiovascular events, and major bleeding in older adults: extended follow-up of the ASPREE trial. European Heart Journal. 2025;00:1–13. doi:10.1093/eurheartj/ehaf514