Reacción a "A study suggests that stopping statin treatment in people aged 75 and over with no history of heart disease or stroke does not increase the risk of death"

Does the press release accurately reflect the study?

"The press release is technically correct, but I believe it conveys the same error as the paper: in reality, the authors have not demonstrated that discontinuing statins is inferior (or superior) to continuing them, because 1) the study’s power is much lower than anticipated (only half the planned number of patients have been recruited); 2) mortality was half that expected, so the study’s power must have fallen from 90 per cent to less than 50 per cent, meaning the study’s ability to detect the primary endpoint is on a par with flipping a coin, if not even lower.

I believe the press release should be extremely cautious: I would say that the study has not had sufficient power to assess whether, in older people, discontinuing statins affects mortality: there is not enough robust data to say that it affects mortality, nor that it improves or worsens patients’ quality of life, particularly if they are French octogenarians without cardiovascular disease who regularly visit their GPs (external validity).

The headline is a clear example of the risk posed by this paper. “Discontinuing statin treatment in adults over 75 with no history of heart disease or stroke does not increase the risk of death”.

Is the study of good quality? Are the conclusions supported by solid data?

“The study design is of good quality (in theory), but its execution is hampered by a loss of statistical power (i.e. the ability to prove the working hypothesis) which is roughly halved due to recruiting only half the intended number of patients, and perhaps a further 50 per cent because the incidence of the primary outcome is half that estimated, along with other statistical reasons that are not appropriate for discussion in a press release. I therefore believe the paper’s conclusion should be much more cautious.

How does this study fit in with the existing evidence?

“This study is a commendable effort to discontinue a widely used chronic treatment with limited evidence of benefit in older people, but, apart from the weakness of the results mentioned above, it contrasts with the very robust evidence from intervention studies.

This study involved 1,200 patients followed for three years, and its inconclusive result contrasts with the consistent benefits of the meta-analysis also published in The Lancet, which included 186,000 patients, of whom around 14,000 were over 75 years of age and were followed for 4.9 years.

Furthermore, population-based studies indicate that discontinuing statins is associated with an increase in cardiovascular morbidity and mortality. It is true that these are observational rather than interventional studies, and whilst their methodological quality is lower due to potential biases, they involve tens of thousands of patients.”

Have the authors taken confounding factors into account? Are there any significant limitations that need to be considered?

“I have already mentioned the limitations regarding statistical power. Furthermore, the external validity is questionable: the cohort consists of French octogenarians who regularly visit their GP and who have a mortality rate of less than 10 per cent at three years (with or without statins). This may have nothing to do with other populations of octogenarians.”

What are the implications for real-world practice?

“As the accompanying editorial suggests, what seems clear is that withdrawing statins from elderly people who tolerate them well offers them no real benefit: it does not improve their quality of life, nor do other symptoms disappear. Whether or not this alters the risk of mortality cannot be concluded from this study. It must not be interpreted (which carries an extreme risk) as proof that statins cease to be effective at the age of 80. It remains in a zone of uncertainty, but one should not over-interpret a underpowered negative study that is unable to robustly assess its own primary endpoint.

It certainly leaves doctors with the option of discussing with patients what to do when faced with complex decisions regarding the elderly, as should always be the case.

Therefore, this study carries a high risk of being misinterpreted on social media platforms that are highly active in their opposition to statins, without a solid scientific basis.”