Citicoline (CDP-choline) is not a treatment for dementia. The memory record is limited and should be read that way.
Sudden facial droop, arm weakness, or speech difficulty can be a stroke. Call 911 immediately.
Small trials in people with vascular cognitive impairment reported modest improvements. That sentence is as far as the claim goes. Citicoline is not a treatment for dementia, and it is not described here as reversing memory loss.
The Cochrane review by Fioravanti and Yanagi suggests modest short-term benefit on memory and behaviour in chronic cerebrovascular cognitive decline. The trials were short and of low quality. IDEALE, an observational study in Italy, associated citicoline with stable cognitive scores compared with decline in an untreated group. An untreated comparison in an observational study is not a randomized proof.
If memory is the reason you looked this up, the next step is a conversation with your own doctor, not a stack. Questions to ask is written for that visit.
| Question | Best evidence | Tier | Bottom line |
|---|---|---|---|
| Chronic cerebrovascular cognitive decline | Cochrane review, Fioravanti and Yanagi | T2 | Modest short-term benefit on memory and behaviour; trials short and low quality |
| Mild vascular cognitive impairment | IDEALE, Cotroneo et al. | T3a | Stable scores vs decline in an untreated group; observational |
Also read side effects and doses used in studies.
When people search for citicoline (CDP-choline) and memory, they usually mean one of three very different situations. The first is a person who has noticed their own memory slipping and wants to know whether a supplement will sharpen it. The second is a family member watching someone decline with a vascular or mixed dementia and wondering whether anything can slow it. The third is a healthy adult, often younger, who wants better focus and recall for work or study.
The research does not treat these as one question, and neither should you. The strongest body of citicoline memory research involves older adults with cerebrovascular cognitive decline, thinking problems tied to reduced or damaged blood flow in the brain. That is a specific medical condition, not ordinary forgetfulness. Results in that group cannot be stretched to cover a healthy 30-year-old, and they cannot be stretched to cover Alzheimer's disease either.
This page stays inside what the registered evidence actually says. Where the evidence is weak, we say so. Where a large trial failed to confirm an earlier hope, that failure stays on the page.
The central piece of evidence for citicoline and memory is the Cochrane review by Fioravanti and Yanagi, which pooled trials in people with chronic cerebrovascular cognitive decline. The review suggests modest short-term benefit on memory and behaviour.
Every word in that sentence does work.
"Modest" means the measured effects were small. "Short-term" means the trials did not run long enough to say whether any benefit holds up over months or years, and memory conditions are, by nature, long-term problems. "Suggests" means the reviewers did not treat the finding as settled.
The most important limit is quality. The trials behind the review were short and of low quality. In practical terms, that means small groups, limited follow-up, and methods that leave room for bias. A finding built on low-quality trials is a lead, not a conclusion. It justifies more research. It does not justify telling a family that a supplement will protect a parent's memory.
IDEALE, conducted by Cotroneo and colleagues in Italy, followed people with mild vascular cognitive impairment. The group taking citicoline showed stable cognitive scores over time, while an untreated comparison group showed decline.
That sounds encouraging, and it is the kind of result that gets quoted in marketing. But the design matters more than the headline. IDEALE was observational. People were not randomly assigned to citicoline or to nothing. The two groups may have differed in ways the study could not measure, in other medications, in general health, in how closely they were followed by doctors. Any of those differences could explain part of the gap between stable and declining scores.
An untreated comparison group in an observational study is not the same as a placebo group in a randomized trial. Randomization exists precisely to strip out those hidden differences. Without it, the honest conclusion is that IDEALE found an association worth testing properly, not that citicoline was shown to hold cognition steady.
A fair reading of citicoline and cognition has to include the studies that failed, because they are larger and better designed than most of the positive ones.
The ICTUS trial, published by Dávalos in The Lancet in 2012, enrolled about 2,300 people with acute stroke. It is the largest citicoline trial ever run. Earlier meta-analyses of smaller stroke trials had suggested benefit. ICTUS found no improvement in 90-day recovery compared with placebo. When the biggest and most rigorous test of a treatment contradicts the smaller studies that came before it, the bigger trial carries the weight.
The COBRIT trial, published by Zafonte in JAMA in 2012, tested citicoline in traumatic brain injury. It did not improve functional or cognitive outcomes compared with placebo. Brain injury is not the same as vascular cognitive decline, but COBRIT matters here because it was a serious, well-powered test of citicoline for cognition after brain damage, and it came back negative.
Neither trial was a memory trial in the narrow sense. Both are relevant to anyone being told that citicoline "protects the brain." The best trials of that idea did not confirm it.
Sudden facial droop, arm weakness, or speech difficulty can be a stroke. Call 911 immediately. No supplement is part of stroke care, and memory worries that arrive suddenly are a reason to seek emergency help, not to shop.
Two small trials funded by Cognizin, a brand of citicoline, tested attention rather than memory per se. McGlade and colleagues reported better attention-test scores in women taking 250 or 500 mg, and gains in attention and psychomotor speed in adolescent males. Both trials were small and industry funded. Industry funding does not make a result false, but it is a reason to want independent confirmation before treating the finding as established, and that confirmation does not yet exist.
Attention is also not memory. A faster or more sustained attention score on a lab task does not show that everyday recall improved. If you are a healthy adult hoping for a memory edge, the evidence does not currently answer your question either way. For ADHD specifically, there are no adequate trials, and there is no good evidence citicoline treats ADHD. Our citicoline and ADHD page covers that gap directly.
Citicoline provides raw material involved in membrane phospholipid synthesis, and evidence suggests it supports that pathway. That is a plausible mechanism, not a proven clinical effect. The neuroprotection story, the idea that citicoline shields brain cells from damage, is preclinical only. It comes from laboratory and animal work, and the large human trials designed to test it, ICTUS and COBRIT, did not show the hoped-for benefit. A reasonable mechanism plus failed large trials means the mechanism, at tested doses in those settings, did not translate into measurable recovery.
Across trials, citicoline at 250 to 2,000 mg per day was generally well tolerated. Reported side effects were mild: stomach upset, headache, and insomnia. Data in pregnancy and breastfeeding are lacking, so those groups are outside what the evidence can speak to.
One interaction deserves attention in exactly the population that memory research touches. Citicoline may enhance the effects of levodopa. Parkinson's disease and vascular cognitive problems often overlap in older adults, and a person with Parkinson's who is considering citicoline should involve their neurologist before starting it. Caregivers should not add any supplement to someone's regimen quietly, the medication list for an older adult is usually long, and the prescriber needs to see the whole picture. Our pages on side effects and doses used in studies go into more detail.
The honest summary is short. In chronic cerebrovascular cognitive decline, a Cochrane review suggests modest short-term benefit on memory and behaviour, on trials that were short and low quality. In mild vascular cognitive impairment, one observational study associated citicoline with stable scores against decline in an untreated group. In acute stroke and traumatic brain injury, the two largest and most rigorous trials found no benefit over placebo. In healthy people and in ADHD, the evidence is thin or absent.
None of this makes citicoline a treatment for dementia, and nothing on this page should be read as reversing memory loss. If memory change is the reason you are here, the most useful next step is a medical evaluation, because memory problems have many causes, some reversible, some not, and the supplement question comes after the diagnosis question. Questions to ask your doctor is written for that appointment. For the broader picture across every condition, see the citicoline evidence overview, and for background on the compound itself, what citicoline is.
No. It is not a treatment for dementia.
It suggests modest short-term benefit on memory and behaviour. The trials were short and low quality.
An observational study that associated citicoline with stable scores rather than decline in an untreated group.
No. Observational association is a weaker kind of evidence than a large randomized trial.
No. Talk to the person's doctor, especially if other medicines are in use.
On the side-effects page. Trials describe it as generally well tolerated, with gaps in pregnancy and breastfeeding.
This site reports what studies found. It does not choose a dose for you.