BMC Medicine (Jul 2024)

Short- and long-term impact of aspirin cessation in older adults: a target trial emulation

  • Zhen Zhou,
  • Katherine L. Webb,
  • Mark R. Nelson,
  • Robyn L. Woods,
  • Michael E. Ernst,
  • Anne M. Murray,
  • Andrew T. Chan,
  • Andrew Tonkin,
  • Christopher M. Reid,
  • Suzanne G. Orchard,
  • Brenda Kirpach,
  • Raj C. Shah,
  • Nigel Stocks,
  • Jonathan C. Broder,
  • Rory Wolfe

DOI
https://doi.org/10.1186/s12916-024-03507-8
Journal volume & issue
Vol. 22, no. 1
pp. 1 – 11

Abstract

Read online

Abstract Background The net benefit of aspirin cessation in older adults remains uncertain. This study aimed to use observational data to emulate a randomized trial of aspirin cessation versus continuation in older adults without cardiovascular disease (CVD). Methods Post hoc analysis using a target trial emulation framework applied to the immediate post-trial period (2017–2021) of a study of low-dose aspirin initiation in adults aged ≥ 70 years (ASPREE; NCT01038583). Participants from Australia and the USA were included if they were free of CVD at the start of the post-trial intervention period (time zero, T0) and had been taking open-label or randomized aspirin immediately before T0. The two groups in the target trial were as follows: aspirin cessation (participants who were taking randomized aspirin immediately before T0; assumed to have stopped at T0 as instructed) versus aspirin continuation (participants on open-label aspirin at T0 regardless of their randomized treatment; assumed to have continued at T0). The outcomes after T0 were incident CVD, major adverse cardiovascular events (MACE), all-cause mortality, and major bleeding during 3, 6, and 12 months (short-term) and 48 months (long-term) follow-up. Hazard ratios (HRs) comparing aspirin cessation to continuation were estimated from propensity-score (PS) adjusted Cox proportional-hazards regression models. Results We included 6103 CVD-free participants (cessation: 5427, continuation: 676). Over both short- and long-term follow-up, aspirin cessation versus continuation was not associated with elevated risk of CVD, MACE, and all-cause mortality (HRs, at 3 and 48 months respectively, were 1.23 and 0.73 for CVD, 1.11 and 0.84 for MACE, and 0.23 and 0.79 for all-cause mortality, p > 0.05), but cessation had a reduced risk of incident major bleeding events (HRs at 3 and 48 months, 0.16 and 0.63, p < 0.05). Similar findings were seen for all outcomes at 6 and 12 months, except for a lowered risk of all-cause mortality in the cessation group at 12 months. Conclusions Our findings suggest that deprescribing prophylactic aspirin might be safe in healthy older adults with no known CVD.

Keywords