TrialsgrambetaRCTs are beautiful
AHA Scientific Sessions 2026Presented Sat 7 Nov · 15:15 GMT-6LBS.03 Late-breaking Science: Hot Topics in CVD PreventionAll Late-Breaking Science trials →

STREAM

2026

STREAM: Discontinuing Statins in Multimorbid Older Adults Without Cardiovascular Disease, a Randomized Non-inferiority Clinical Trial

In adults aged 70 or older with at least two chronic conditions taking a statin for primary prevention, statin discontinuation did not significantly increase heart attack, stroke or death at about two years compared with statin continuation.

Result

All-cause death, non-fatal myocardial infarction or non-fatal ischaemic stroke at 24 months
Absolute difference
3.5 points

The absolute difference is derived from the event rates.

Also reported: At about two years, discontinuation of statins was safe, without a significant increase in heart attack, stroke or death compared with continuation. Over two years, for every 40 individuals who discontinued statins, one additional cardiovascular event occurred. There were no differences between the groups in quality of life.

Original publication
Subtopic
Statin discontinuation
Design
Pragmatic randomised non-inferiority trial (Inselspital Bern; 24 sites in Switzerland, France and the Netherlands) of statin discontinuation versus continuation in multimorbid adults aged 70 or older taking a statin for primary prevention
Enrolled
1,880 patients
Source
Rodondi N, Haller ML, Bardoczi JB, et al. N Engl J Med. 2026; published online 7 November.doi 10.1056/nejmoa2606634

Discussion & critique

In older adults with multiple chronic conditions taking a statin for primary prevention, stopping the statin did not significantly increase heart attack, stroke or death at about two years. Over follow-up of up to four years, events were more frequent after discontinuation (12.7% vs 9.2%), and the study author described this as a new finding that argues against a general recommendation to stop. The release gives no primary endpoint definition, non-inferiority margin, hazard ratio, confidence interval or P value, and does not say how participants were allocated or whether the trial was blinded. About 14% of participants crossed over to the other group, which the release says further analyses did not change.

Discussed in podcasts