Carbon dioxide made fluid move more strongly through the brain

Carbon dioxide is normally treated as a waste gas that the body needs to expel. Yet brief, carefully controlled increases in CO₂ produced a different effect in the brain: blood vessels repeatedly widened and narrowed, accompanied by stronger movement of cerebrospinal fluid.

Several proteins associated with Alzheimer’s and Parkinson’s disease also rose in the bloodstream after the procedure. The researchers suggest that some of these molecules may have been transported out of the brain as part of its waste-clearance process.

The experiment was conducted in people rather than laboratory animals. The main study included 63 participants: 30 people with Parkinson’s disease and 33 healthy adults of a similar age. While undergoing MRI, they alternated between breathing a gas mixture that raised their carbon dioxide level and a mixture that returned it to baseline. Each period of elevated CO₂ lasted 35 seconds, and the full procedure took about five minutes.

The scans showed that the increase in carbon dioxide dilated cerebral blood vessels. When CO₂ fell again, the vessels returned towards their initial state. These fluctuations were followed by an inflow of cerebrospinal fluid, the clear liquid surrounding the brain and spinal cord.

This motion is thought to contribute to the glymphatic system: a network of fluid pathways along blood vessels that helps transport dissolved waste out of brain tissue. Glymphatic activity appears to be particularly strong during deep sleep, when slow changes in neural activity and cerebral blood volume generate rhythmic fluid movement.

People with Parkinson’s disease showed a weaker vascular response to carbon dioxide and a smaller associated cerebrospinal-fluid inflow than the healthy participants. The result is consistent with the possibility that the brain’s waste-removal machinery becomes less effective in neurodegenerative disease.

The most provocative part of the study was also much smaller. Five participants with Parkinson’s disease and five healthy volunteers completed three sessions lasting about ten minutes each. Across the sessions, they underwent 24 short exposures to a mixture containing 5% CO₂.

Blood was collected twice before the intervention and again roughly 45, 90 and 150 minutes after it began. At the first follow-up, plasma concentrations had increased for amyloid-beta peptides Aβ40 and Aβ42, total alpha-synuclein and several other proteins produced largely in the nervous system.

Aβ42 rose in eight of the ten participants, while Aβ40 rose in nine. The individual increases ranged from 0.3% to 8.7% for Aβ42 and from 2.5% to 7.8% for Aβ40.

Phosphorylated tau-217 increased in seven participants, although the overall result for this biomarker did not reach the conventional threshold for statistical significance. One participant with biomarker evidence of coexisting Alzheimer’s pathology showed much larger amyloid changes, but was treated as an outlier and excluded from the main statistical analysis.

More protein appearing in blood does not prove that pathological deposits were removed from the brain. The researchers did not measure amyloid plaques or tau aggregates in the brain before and after the intervention. Some of the proteins could also have come from peripheral tissues or entered the circulation through changes in blood–brain barrier permeability.

The biomarker experiment involved only ten people, had no sham-air control condition and followed participants for just a few hours. It did not test cognition, disease symptoms or long-term changes in the brain.

Intermittent carbon dioxide exposure therefore cannot be described as a treatment for Alzheimer’s or Parkinson’s disease. It would also be unsafe to attempt outside a controlled setting: the gas concentration, exposure pattern and participants’ physiological responses were monitored with medical equipment.

The study demonstrates a narrower but striking possibility. Fluid movement in the brain can be influenced through deliberately timed changes in the calibre of cerebral blood vessels. Whether that motion genuinely accelerates the removal of harmful proteins — and whether repeated sessions could alter disease progression — remains unresolved.