STREAM Trial: Stopping Statins Clears 2-Year Safety Bar but Raises Long-Term Heart Risk

For older adults who never had a heart attack or stroke and are already managing several chronic conditions, stopping a statin is not immediately dangerous — but it is not risk-free either. That is the central finding of STREAM, a randomized controlled trial published November 7, 2026, in the New England Journal of Medicine that is the largest rigorous test of statin discontinuation in older primary-prevention patients to date.

The trial enrolled 1,880 adults aged 70 or older across 24 clinical sites in Switzerland, the Netherlands, and France. All participants had at least two chronic conditions and were taking statins for primary prevention of cardiovascular events — meaning none had a prior heart attack or stroke. By design, STREAM asked whether it was safe to stop: patients were randomly assigned either to discontinue their statin or to continue it, then followed for a median of 103 weeks.

Over that two-year window, the discontinuation group experienced a restricted mean event-free survival difference of −2.0 weeks (95% CI −3.7 to −0.3), which fell within the trial's prespecified non-inferiority margin of −5.2 weeks. The result formally confirmed that stopping statins was non-inferior to continuing them on the primary composite endpoint, allowing principal investigator Nicolas Rodondi of the University of Bern and co-investigators to report that discontinuation cleared the two-year safety bar.

What the STREAM Trial Found

The formal non-inferiority verdict, however, tells only part of the story. When researchers analyzed the full follow-up period using a conventional hazard ratio, the picture shifted. The HR for major cardiovascular events or death across the entire observation window was 1.38 (95% CI 1.05–1.81), indicating that patients who stopped statins faced a 38% higher relative risk of a primary-endpoint event compared with those who continued. That translates to roughly one additional cardiovascular event for every 29 people who discontinued.

Event rates over the median 103-week follow-up illustrate the gap concretely: 12.7 percent (119 of 938) of the discontinuation group experienced a primary-endpoint event, against 9.2 percent (87 of 942) in the continuation group. Secondary endpoints reinforced the pattern. Nonfatal myocardial infarction occurred in 1.4 percent of patients who stopped versus 0.6 percent who continued; ischemic stroke in 1.9 percent versus 0.8 percent; and all-cause death in 8.2 percent versus 6.9 percent.

Serious adverse events were comparable between arms — 43.0 percent in the discontinuation group and 39.8 percent in the continuation group — an important finding for multimorbid elderly patients for whom medication burden and adverse effects are often the primary motivation for considering a deprescribing conversation.

The trial's subgroup analysis by recruitment setting revealed a potentially clinically meaningful difference: patients recruited through general practitioners, the majority of the study population, showed a −3.4-week event-free survival difference when they stopped statins, while those recruited through hospital outpatient settings showed only a −1.1-week difference. That gap suggests the effect of statin discontinuation may be more pronounced in community-based primary care patients than in patients already under specialist monitoring.

What Happens to Cholesterol When Statins Stop

A separate substudy within STREAM tracked the lipid consequences of discontinuation in 235 participants. Among those who stopped statins, LDL cholesterol rebounded to a mean of 176.0 mg/dL at 12 months — compared with 99.4 mg/dL in the continuation group. That 76.6 mg/dL absolute difference represents a substantial lipid rebound and aligns with what pharmacology would predict, but it also quantifies the cardiovascular exposure created by stopping treatment.

The rebound data provide a mechanistic link between discontinuation and the elevated event risk seen in the HR analysis, and they carry direct clinical relevance: a patient who stops statins, even temporarily, is not returning to their pre-statin baseline but rather to a lipid level that reflects the absence of active treatment.

No Quality-of-Life Benefit Found

One argument commonly offered in support of statin deprescribing in older adults is that patients may gain quality-of-life benefits by reducing pill burden and avoiding side effects such as muscle pain. STREAM was designed to test that hypothesis directly, and the results did not support it. Quality-of-life scores were similar between the two groups at all measured time points, with no statistically significant benefit emerging in the discontinuation arm.

That null finding matters for clinical practice. A common framing in deprescribing discussions with patients is that stopping a statin might improve how they feel even if it carries some risk. The STREAM data do not support that trade-off in a population of multimorbid older adults.

The Independent Safety Board's Signal

The clearest signal from STREAM may have come not from the primary endpoint analysis but from the trial's independent data and safety monitoring board. After reviewing the accumulating evidence, the DSMB recommended that participants who had been randomized to the discontinuation arm should be advised to discuss restarting statins with their physicians.

That recommendation from an independent oversight body — a standard clinical trial safeguard empowered to act when accumulating data suggest participant risk — is itself a clinical signal. It does not constitute a formal finding of harm, but it reflects the DSMB's judgment that the direction and magnitude of the data warranted proactive clinical communication rather than waiting for the trial's formal conclusion.

Co-investigator Rita Redberg, MD, MS, FAHA, a professor of medicine at the University of California, San Francisco, summarized the trial's nuanced result: "one cannot make a general recommendation about stopping statins in this patient population," but also acknowledged that "discontinuation is safe for most people who are in this age group and without a previous CVD event."

Three 2026 Trials, Three Angles on the Same Question

STREAM does not stand alone. Three separate major trials published in 2026 have collectively shifted what clinicians know about statin therapy in older adults, and they reach conclusions that point in different directions depending on the question being asked.

STAREE, led by Sophia Zoungas at Monash University in Australia, enrolled 9,971 community-dwelling adults aged 70 and older to test whether starting atorvastatin 40 mg, compared with placebo, reduces mortality and disability. Its results, presented at the European Society of Cardiology meeting and simultaneously published in the New England Journal of Medicine on August 29, 2026, found that atorvastatin significantly reduced a composite cardiovascular endpoint including major cardiovascular events, with an HR of 0.70 (95% CI 0.61–0.82; P<0.001). That is a substantial cardiovascular benefit for starting statins in older primary-prevention patients. However, the trial's primary endpoint — disability-free survival — showed no statistically significant difference (HR 0.94; P=0.25), complicating a simple interpretation.

The SAGA/SITE trial, published in Lancet Healthy Longevity in August 2026, took yet another angle. Conducted in France among 1,160 adults aged 75 and older, it found that statin discontinuation was non-inferior to continuation for 36-month mortality (7.9 percent versus 7.2 percent) and that major adverse cardiovascular events were comparable (4.4 percent versus 5.0 percent), with a composite endpoint of 11.4 percent versus 12.0 percent also showing non-inferiority.

Together, the three trials — STREAM, STAREE, and SAGA/SITE — capture a genuine clinical tension. For patients already on statins for primary prevention in older age, continuation appears to carry meaningful cardiovascular benefit over stopping. For patients not yet on statins, initiating therapy in older age reduces cardiovascular events but may not translate to the broader disability-free survival benefit that often motivates treatment decisions in elderly patients.

Guideline Pressure and the PREVENTABLE Question

STREAM arrives at a moment of significant pressure from the guideline side. The 2026 AHA/ACC dyslipidemia guideline, published March 13, 2026, in Circulation and the Journal of the American College of Cardiology, introduced the PREVENT-ASCVD calculator to replace the older Pooled Cohort Equations. The transition materially expands the number of adults formally recommended for statin therapy: a JAMA study by Anderson and colleagues estimated that 87.5 million US adults aged 30 to 79 — representing 56.6 percent of that population — now meet criteria for primary prevention statins under the updated guideline, including 21.5 million newly eligible compared with the 2018 framework. Among adults aged 70 to 79 specifically, the eligibility figure exceeds 93 percent.

That expansion creates a population-level context in which STREAM's findings carry direct policy relevance. If more older adults are now formally recommended for statins, then understanding the consequences of stopping — whether due to side effects, patient preference, or polypharmacy concerns — becomes a central clinical question.

The most consequential piece of evidence still pending is the PREVENTABLE trial, a National Institutes of Health-funded randomized trial enrolling approximately 20,000 adults aged 75 and older in the United States under principal investigator Karen Alexander, MD, at the Duke Clinical Research Institute. With results expected in December 2026, PREVENTABLE is the largest trial yet designed to address statin initiation in elderly primary-prevention patients, and its findings will likely anchor guideline updates for years.

What Clinicians and Patients Should Take Away

The STREAM results offer a measured answer to a question that clinicians and older patients frequently raise: the answer is that statin discontinuation in multimorbid older primary-prevention patients meets a formal non-inferiority standard over two years, but it comes with a 38% relative increase in cardiovascular risk over the full follow-up period, no quality-of-life benefit, a meaningful LDL rebound, and an independent safety board's recommendation to consider restarting therapy.

For clinicians, STREAM does not endorse deprescribing statins as routine practice in this population. The trial's investigators and the DSMB's signal both point toward caution. For patients, it validates that a shared decision-making conversation with a physician — one that weighs their individual cardiovascular risk, medication burden, functional status, and preferences — remains the appropriate framework. STREAM narrows the uncertainty around what happens when statins stop, but it does not resolve it in favor of stopping.