Cholesterol After 90 and the Limits of the Sardinia Study

Sponsored by VERTU PROJECT150

Higher cholesterol after 90 is not a reason to stop a statin. A small Sardinian study found an association between higher LDL cholesterol and longer survival, but it did not test whether raising LDL improves health or whether stopping treatment is safe.

For anyone reading a cholesterol result alongside a longevity headline, the useful question is what the evidence can establish. Blood measurements, survival patterns and treatment benefits are related, but they are not interchangeable.

What did the Sardinia cholesterol study find?

The study found longer observed survival in participants with higher baseline LDL-C. It was an observational analysis of 168 adults aged 90 or older from Sardinia’s longevity Blue Zone, rather than a treatment trial.

The 2025 Nutrients paper reports recruitment waves in 2018 and 2023 in its methods section, with follow-up ending in December 2024. The maximum observation period was approximately six years; this was not six years of follow-up for every participant. People taking statins or loop diuretics were excluded.

The grouping threshold was LDL-C 130 mg/dL, approximately 3.4 mmol/L.

Reported mean survival was about 3.8 years in the higher group and 2.8 years in the lower group.

Table 4 reports an adjusted all-cause mortality hazard ratio of 0.606, with a 95% confidence interval of 0.403–0.912, for the higher group relative to the lower group.

That hazard ratio describes an association after adjustment for measured factors. It does not mean that deliberately increasing someone’s LDL will reduce their chance of dying by about 40%.

Why does the cholesterol paradox need careful interpretation?

The cholesterol paradox describes an unexpected association, not an established protective effect of high LDL. A finding in people who have already reached an exceptional age cannot automatically explain how others should reach that age.

One explanation is reverse causation: illness or declining health can lower cholesterol while also increasing mortality. Another is survivor selection: people who reach 90 may differ from those who died earlier in ways that influence both cholesterol and survival. Statistical adjustment cannot fully account for influences that were not adequately measured.

The Sardinian authors considered reverse causation less likely after additional analyses. That is their interpretation, not proof that the possibility has been eliminated. The small sample, restricted eligibility and single baseline lipid measurement also limit what can be concluded.

Cholesterol has essential biological functions, including roles in cell membranes and hormone synthesis. Those functions do not establish a desirable blood LDL level or demonstrate that appropriately prescribed lipid-lowering treatment deprives the body of essential materials.

Does cholesterol become unimportant after 80?

No. A weaker relative association at older ages does not mean that cardiovascular risk becomes negligible.

A 2007 analysis of 61 prospective studies found that the relationship between total cholesterol and death from ischaemic heart disease weakened with age but remained present. At ages 70–89, each 1 mmol/L lower total cholesterol was associated with about 17% lower relative risk of death from ischaemic heart disease.

Three distinctions matter. This was total cholesterol, not specifically LDL-C; the outcome was death from ischaemic heart disease, not every cardiovascular outcome; and an observational association is not a direct estimate of treatment benefit. It should not be summarised as “cholesterol barely matters after 80”.

What does treatment evidence show for older adults?

Randomised trials support LDL lowering in appropriate patients, including older adults. They answer a different question from the Sardinian observational study: what happens when treatment is assigned?

A 2010 meta-analysis of 26 statin trials, involving approximately 170,000 participants, found about a 22% reduction in the relative risk of major vascular events per 1 mmol/L reduction in LDL-C. This does not mean a reduction of 22 percentage points in everyone’s absolute risk.

A 2019 meta-analysis of 28 randomised trials also found reductions in major vascular events across age groups. Direct evidence was less extensive for primary prevention in people over 75 without established occlusive vascular disease. Primary prevention means trying to prevent a first cardiovascular event.

The 2025 National Lipid Association and American Geriatrics Society consensus emphasises individual assessment and shared decisions in older adults. Existing cardiovascular disease, frailty, life expectancy, other medicines, tolerability and personal priorities can all affect a treatment decision. Some people with life-limiting illness may appropriately have treatment reduced or stopped under clinical supervision; age alone does not decide the matter.

How should an LDL result of 3.4 mmol/L be discussed?

A single LDL result cannot determine whether someone needs medication. Approximately 3.4 mmol/L is equivalent to 130 mg/dL, but the study’s grouping threshold is not a universal treatment target.

For example, someone with established coronary heart disease and someone without known cardiovascular disease may need different approaches despite having the same result. Diabetes, blood pressure, smoking, family history and other health conditions also change the assessment.

A useful appointment starts with clear records and specific questions:

• Bring previous lipid results, including dates, rather than only the latest flagged value.
• List current medicines, doses and any symptoms you want to discuss.
• Ask whether treatment is intended to prevent a first event or a further event.
• Agree on the treatment goal, follow-up timing and what should prompt an earlier review.

These steps organise a clinical conversation. They do not substitute for a clinician’s assessment or authorise a medication change.

Frequently asked questions

Does high cholesterol make people over 90 live longer?

That has not been established. The Sardinian finding was an association in a selected group, not evidence that increasing cholesterol extends life.

Should someone over 90 stop taking a statin?

Not on the basis of this study. Any decision to continue, reduce or stop treatment should be discussed with the prescribing clinician in the context of the person’s health and goals.

Is LDL-C of 130 mg/dL a healthy target for everyone?

No. It was a grouping threshold in this study. An appropriate treatment target depends on individual cardiovascular risk and clinical circumstances.

Can middle-aged adults apply this finding to themselves?

No. Findings in selected people who have already survived to 90 cannot directly establish a treatment strategy for middle-aged adults.

Does this research validate a longevity service?

No. The study did not evaluate a commercial service, supplement or longevity programme, including VERTU PROJECT150.

Conclusion

The cholesterol paradox is a reason to interpret research carefully. This study adds a finding about survival in a particular older population; it does not overturn randomised treatment evidence or establish a reason to raise LDL. Decisions about cholesterol should remain grounded in individual risk, clinical evidence and discussion with a qualified clinician.

This article provides general health information, not individual medical advice. Discuss starting, adjusting or stopping medication with your clinician. Unit conversions are approximate.

VERTU PROJECT150 Personal Physician Service

A cholesterol result is easier to discuss when previous reports, current medicines and follow-up arrangements are available together. For readers seeking ongoing support with this process, VERTU PROJECT150 offers a Personal Physician Service that provides health information management and care coordination. The service itself does not provide medical diagnosis or treatment and does not prescribe.

The service starts with consent to compile relevant reports, medical history, medicines, allergies and family history in a personal health record. Depending on the agreed service package, support includes organising these records, coordinating appointments with appropriately licensed medical institutions and doctors, transmitting information, and arranging follow-up reminders. Specialist appointments can be coordinated when needed, within the agreed scope.

Medical assessment, diagnosis and prescribing are provided independently by appropriately qualified doctors at licensed medical institutions, acting within their authorised scope of practice and retaining responsibility for clinical decisions. VERTU PROJECT150 does not employ those doctors or control their medical decisions. The service does not replace emergency care or guarantee specialist availability, improvements in cholesterol or a longer life. Specific inclusions, additional charges and delivery arrangements should be confirmed before enrolment.