Discontinuation of statins for primary prevention of atherosclerotic cardiovascular disease in adults aged 75 years or older (SAGA/SITE): a multicentre, open-label, pragmatic, non-inferiority randomised trial.

Discontinuation of statins in older adults aged 75 and above who have no history of cardiovascular disease does not seem to increase the risk of death compared to continuing statin therapy.

Summary and findings

This study assessed the non-inferiority of statin cessation versus continuation in terms of all-cause mortality among individuals aged 75 years or older who were prescribed statins for primary prevention of atherosclerotic cardiovascular disease. A total of 1160 participants were analyzed over a 36-month follow-up period. The results indicated no significant difference in mortality rates between the two groups.

Abstract

<h4>Background</h4>Statins are among the most widely used drugs; however, their benefits for primary prevention of atherosclerotic cardiovascular disease (ASCVD) in individuals aged 75 years or older without the disease remain uncertain. We aimed to assess the non-inferiority of statin cessation in terms of all-cause mortality under real-life conditions in individuals aged 75 years or older who were prescribed statins for the primary prevention of ASCVD.<h4>Methods</h4>We performed a multicentre, open-label, parallel-group, phase 3 randomised controlled trial in 297 primary care offices. Individuals who were aged 75 years or older, were prescribed any statin for at least 1 year for the primary prevention of ASCVD, had no history of ASCVD, and could provide informed consent were randomly assigned (in an unbalanced 5:4 ratio) to continue or stop their statin treatment and were followed up for 36 months. Participants were excluded if they had a progressive disease with a life expectancy of 3 months or less, were diagnosed with dementia, had known homozygous or double heterozygous familial hypercholesterolaemia, or were unable to provide informed consent. Randomisation was computer-generated and implemented via a central electronic system; masking was not done. The primary endpoint was all-cause mortality at 3 years, analysed in participants according to the primary estimand (participants who met all key eligibility criteria and were randomly assigned to statin discontinuation or continuation less than 90 days after inclusion). Missing data were handled using multiple imputation. The non-inferiority margin was 5% for the absolute between-group difference in mortality. This trial is registered with ClinicalTrials.gov, NCT02547883 (completed).<h4>Findings</h4>Between June 15, 2016, and Jan 7, 2020, 1180 participants were randomly assigned, and 1160 participants were included in the analysis of the primary estimand. 639 participants were randomly assigned to the statin continuation group, and 521 participants were assigned to the statin discontinuation group. The median age was 80 years (IQR: 78-84), and 775 (66·8%) of 1160 participants were women; 342 (29·5%) of 1160 participants had diabetes, and 896 (77·2%) of them had hypertension. At 36 months, 48 (7·9%) of 604 participants in the statin continuation group and 35 (7·2%) of 484 participants in the statin discontinuation group had died. The absolute difference in all-cause mortality between groups was -0·68% (95% CI -3·95 to 2·60). Statin discontinuation was non-inferior to continuation for 3-year all-cause mortality as the upper bound of the between group difference 95% CI did not exceed the prespecified non-inferiority margin. Adverse events occurred in 461 (73·1%) of 631 participants in the continuation group, and in 390 (74·0%) of 527 in the discontinuation group. Non-cardiovascular adverse events occurred in 456 (72·3%) of 631 participants in the continuation group and 383 (72·7%) of 527 in the discontinuation group.<h4>Interpretation</h4>In persons aged 75 years or older receiving statins for primary prevention and with no previous history of ASCVD, discontinuation of statins was non-inferior to continuation in terms of all-cause mortality over 3 years. These findings support individualised decision-making regarding statin discontinuation in older adults.<h4>Funding</h4>French Ministry of Health within the framework of the Medico Economical Research Program (PRME) 2014.

Key findings

• At 36 months, 48 (7.9%) of 604 participants in the statin continuation group and 35 (7.2%) of 484 participants in the statin discontinuation group had died.
• The absolute difference in all-cause mortality between groups was -0.68% (95% CI -3.95 to 2.60).
• Adverse events occurred in 461 (73.1%) of 631 participants in the continuation group and in 390 (74.0%) of 527 in the discontinuation group.
• Non-cardiovascular adverse events occurred in 456 (72.3%) of 631 participants in the continuation group and 383 (72.7%) of 527 in the discontinuation group.

Limitations

• open-label design may introduce bias
• 36-month follow-up may not capture long-term effects
• non-inferiority margin set at 5% may influence interpretation
• unbalanced randomization ratio of 5:4

Elsewhere in the IGF-1 DES (Des(1-3) IGF-1) corpus