(Image by Juan Gaertner on Shutterstock)
In A Nutshell
- Blood amyloid-beta, a protein tied to Alzheimer’s, ran about a third higher in healthy adults after a chest-pain drug and a vibrating mattress.
- Bigger blood pressure drops and older age went along with bigger rises.
- Brain scans and mouse tests offered support, but nothing confirmed the brain was actually clearing itself.
- Early and small: 29 adults, no memory outcomes measured, and nowhere near a treatment.
Blood levels of amyloid-beta ran roughly a third higher in a small group of healthy adults after they took chest-pain tablets and rested on a vibrating mattress, according to a new study. That protein builds up in the brains of people with Alzheimer’s disease, and the jump is consistent with the brain sending more of it into the bloodstream.
Published in Nature Communications, the study paired nitroglycerin, a decades-old chest-pain medication that widens blood vessels, with a vibrating mattress tuned to stimulate blood vessel walls. In 29 healthy volunteers, the same treatment was also associated with higher levels of GFAP, a protein tied to brain support cells. The rises grew with age and with the size of the blood pressure drop.
Nobody involved had dementia, and this was not a trial of an Alzheimer’s treatment. Researchers wanted to know whether two existing, noninvasive tools, one chemical and one mechanical, could push more waste out of the brain’s internal cleanup system. Scientists have long suspected that sluggish cleanup lets proteins like amyloid-beta pile up years before symptoms appear. If that system could be safely jump-started, it might open a new direction for prevention research.
29 Healthy Adults Tried a Vibrating Mattress and Nitroglycerin for Alzheimer’s Research
Participants ranged from 27 to 61 years old, and 15 were women. Each came in for two sessions on separate days, in random order. One session used vibration alone, delivered through a mattress that pulses the whole body at low frequencies. The other paired the vibration with two nitroglycerin tablets dissolved under the tongue. The first came during an initial brain scan and the second about 20 minutes later, during the vibration. Data came from 26 people for the combined session and 27 for vibration alone. A separate group of 12 took a single tablet with no vibration, so researchers could isolate the drug’s own effect. Another 13 volunteers rested for 45 minutes between two brain scans as a comparison, and their scans did not show the same changes.
Nitroglycerin releases nitric oxide, a molecule that relaxes blood vessels and lowers blood pressure. The vibration was tuned to trigger a pressure-sensing protein in vessel walls that earlier animal research linked to fluid draining from the brain.
Blood was drawn before and after the combined session and tested for brain-related proteins. Brain scans that pick up water and protein signals were taken before and after, while sensors tracked blood pressure. The second blood draw came roughly an hour and a half after the first tablet.
Blood Amyloid-Beta Rose by About a Third After the Combined Treatment
Blood pressure fell about 15 percent on average during the combined session. Both forms of amyloid-beta that were measured rose by roughly a third, and GFAP rose by a similar amount. Researchers adjusted the results for the slight dilution of blood that nitroglycerin causes, and each protein was measured in 22 to 25 participants.
Two larger proteins, tied to nerve damage and Alzheimer’s, did not budge. Everything that rose was relatively small, hinting that size may decide what slips out of the brain. The authors add that the effect may apply mainly to loose, soluble proteins before they clump into larger deposits.
Brain scans added supporting evidence, with limits. Signals from brain tissue weakened by about 4.6 percent on average after treatment, a change the authors say may reflect less water and protein inside the tissue. Large blood-draining channels at the back of the head gave off stronger signals, a possible sign of more protein, and that strength loosely tracked blood amyloid-beta levels. Other fluid spaces went in different directions or stayed the same, so the picture was uneven, and the scans cannot identify amyloid-beta itself.
Mouse experiments used a different blood-vessel-relaxing drug plus vibration. A glowing tracer dye showed more material moving from fluid around the brain into the blood, which the authors describe as support for the idea rather than confirmation of it in humans.
Nitroglycerin’s blood pressure drop and stronger rhythmic brain pulses also echo parts of sleep, the state most strongly linked to efficient waste clearance in the brain, according to the authors.
Alzheimer’s Research Shows Mild Side Effects and Many Open Questions
Side effects were mild and short-lived. Four participants reported headaches and three reported nausea after nitroglycerin, and one person stopped the session after vomiting. Every symptom cleared within 30 minutes, nobody found the vibration uncomfortable, and there were no serious adverse events. Sessions took place under controlled experimental conditions.
Still, the study was small and short-term, and it tracked blood proteins and scans rather than memory or health outcomes.
A vibrating mattress and a heart pill are nowhere near ready to prevent Alzheimer’s in anyone. Even so, a drug already approved for heart conditions, paired with a mechanical device, was followed by measurable changes in blood proteins. That is a concrete lead for studies in people at risk, where the open questions are whether the extra amyloid-beta in blood reflects real clearance from the brain and whether that would ever lower dementia risk.
Disclaimer: This article reports on early research and is for informational purposes only. It is not medical advice. Nitroglycerin is a prescription drug that lowers blood pressure, and nothing in this study supports using it, with or without a vibrating mattress, to protect the brain. Questions about medications or Alzheimer’s risk belong with a doctor.
Paper Notes
Limitations
Sample sizes were modest overall. The nitroglycerin-only session included 12 people, the vibration-only brain scans were available for 12, and individual blood protein results came from 22 to 25 participants. The authors say age-related patterns in the separate nitroglycerin and vibration experiments should be interpreted cautiously for this reason. The nitroglycerin-only session and the resting comparison group, which included 13 people ages 26 to 67 and received brain scans only, were exploratory and analyzed separately from the main endpoints. Blood protein results were adjusted using a correction for blood dilution that was estimated from blood counts in 10 participants. Participants were healthy adults recruited through university and hospital staff mailing lists, so results may not apply to older adults or people with memory problems. Brain scans measure water and protein signals indirectly and cannot identify amyloid-beta, and the pathway by which brain proteins reach the blood has not been directly confirmed in humans. The mouse experiments used a different nitric oxide drug and, according to the authors, offer general biological support rather than direct validation of the mechanism suggested by the human data.
Funding and Disclosures
Funding came from the Research Council of Finland, the EU Joint Program for Neurodegenerative Disease Research, the Jane and Aatos Erkko Foundation, Oulu University Hospital research grants, the Finnish Medical Foundation, the Finnish Brain Foundation, and several other Finnish foundations. U.S. and Nordic funders included the National Institutes of Health, the Simons Foundation, the Cure Alzheimer’s Fund, the Ludwig Family Foundation, the Novo Nordisk Foundation, the Lundbeck Foundation, and the Dr. Miriam and Sheldon G. Adelson Medical Research Foundation. The vibration device was a commercially available mattress from Oy Neurosonic Finland Ltd. The authors declare no competing interests.
Publication Details
Titled “Vasomechanic treatment is associated with increased efflux of human brain proteins,” the paper was published online November 3, 2026, in Nature Communications (volume 17, article 11164). It was received February 10, 2026, and accepted October 5, 2026. Authors are Johanna Tuunanen, Katariina Hautamäki, Tommi Väyrynen, Matti Järvelä, Vesa Korhonen, Niko Huotari, Mika Kaakinen, Janne Kananen, Heta Helakari, Lauri Raitamaa, Jari Jukkola, Sanna-Kaisa Herukka, Katariina Laurén, Ulla Salmi, Lauri Eklund, Juha Nikkinen, Maiken Nedergaard, and Vesa Kiviniemi. Tuunanen and Kiviniemi are the corresponding authors. Affiliations include the University of Oulu and Oulu University Hospital in Finland, Lund University, the University of Eastern Finland, Kuopio University Hospital, the University of Rochester, and the University of Copenhagen. The trial is registered as EudraCT 2021-000625-27. DOI: 10.1038/s41467-026-78894-y.
