Citicoline (CDP-choline) is sold in the United States as a dietary supplement, not as an FDA-approved drug. It has been tested in fairly large clinical trials, and those trials give us a reasonable picture of its side effect profile. The approved claim is straightforward: citicoline was generally well tolerated at doses of 250 to 2,000 mg per day in trials. The side effects reported most often were mild stomach upset, headache, and insomnia. Data on pregnancy and breastfeeding are lacking, so no safety statement can be made for those groups.
Saying a compound is "well tolerated" is not the same as saying it is safe for everyone. Tolerability in a trial means that most enrolled participants finished the study without serious problems. It does not cover people who were excluded from trials, and it does not cover long-term use outside of a study. This article walks through what the evidence actually shows about citicoline side effects, where the gaps are, and what to raise with your own clinician. You can find more context on our safety hub, the broader evidence base, and our guide to dosage.
Evidence at a glance
| Question | Best evidence | Tier | Bottom line |
|---|---|---|---|
| How well is citicoline tolerated overall? | Trials using 250, 2,000 mg/day, including large stroke and TBI studies | T1 | Generally well tolerated; mild GI upset, headache, insomnia were the main complaints |
| Did the largest trial show harm? | ICTUS (Dávalos, Lancet 2012, about 2,300 people) | T1 | No improvement in 90-day stroke recovery versus placebo; tolerability was acceptable, but the trial was not designed to prove long-term supplement safety |
| Did the TBI trial show harm? | COBRIT (Zafonte, JAMA 2012) | T1 | No improvement in functional or cognitive outcomes versus placebo; again, tolerable but not a safety guarantee |
| Is it safe in pregnancy or breastfeeding? | No adequate human data | , | Data are lacking; cannot be called safe |
| Does it interact with medicines? | Mechanistic and clinical observations with levodopa | T3b/T4 | May enhance levodopa effects; people with Parkinson's should involve their neurologist |
What "generally well tolerated" actually means
When researchers say a compound is well tolerated, they mean that side effects in the trial were mostly mild and did not cause many people to drop out. In citicoline trials, the complaints that came up most often were mild gastrointestinal upset, headache, and insomnia. These are the kinds of effects that are annoying rather than dangerous for most people.
However, trials have limits. Participants are screened before they join. People who are very ill, pregnant, or taking complicated medication lists are often excluded. That means the trial population is usually healthier and simpler than the real-world population of supplement users. A clean tolerability record in trials is reassuring, but it is not a promise that every person will have the same experience.
The large trials: what they showed and what they did not
The two biggest modern trials are worth naming because readers often assume a large trial proves safety. ICTUS, published in The Lancet in 2012 by Dávalos and colleagues, enrolled about 2,300 people with acute stroke. Its main result was negative: citicoline did not improve 90-day recovery compared with placebo. COBRIT, published in JAMA in 2012 by Zafonte and colleagues, tested citicoline after traumatic brain injury and also found no improvement in functional or cognitive outcomes versus placebo.
These trials matter for safety in two ways. First, they show that citicoline could be given to large numbers of patients without a major safety signal stopping the studies. Second, and just as important, neither trial was designed to answer the questions a supplement buyer actually has, such as what happens with years of daily use, or in people with conditions the trials excluded. A negative efficacy result plus acceptable short-term tolerability is not the same as proven long-term safety.
Gaps: pregnancy, breastfeeding, and long-term use
Pregnancy and breastfeeding data are lacking. There is no adequate human evidence to say citicoline is safe in either situation. The honest position is that we do not know, and "we do not know" should be treated as a reason for caution, not reassurance.
Long-term daily use is another gap. Most trials ran for weeks or months. If you plan to take citicoline for years, you are moving beyond what the evidence covers. That does not mean harm is likely; it means the question has not been well studied. This is a good topic to bring to your doctor, and our page on questions for your doctor can help you structure that conversation.
Interactions and special situations
The clearest interaction concern involves levodopa. Citicoline may enhance levodopa effects. For a person with Parkinson's disease, that could change how their medication behaves. Anyone with Parkinson's should involve their neurologist before adding citicoline.
Because citicoline is a supplement in the US, product quality also matters. CDP-choline and citicoline are the same compound, and Cognizin is a branded form of it. The name on the label does not change the safety profile, but manufacturing quality can vary between products, which is a general supplement issue rather than a citicoline-specific one.
Expert insight
I want to be precise about what the register supports. Citicoline was generally well tolerated at 250 to 2,000 mg per day in clinical trials, with mild gastrointestinal upset, headache, and insomnia as the most common complaints. That is a Tier 1 statement grounded in large trials like ICTUS and COBRIT, even though both trials were negative on efficacy. What I cannot tell you is that citicoline is safe for everyone, because the evidence does not say that. Pregnancy and breastfeeding data are lacking, and long-term daily use has not been well studied. If you take levodopa for Parkinson's disease, please involve your neurologist, because citicoline may enhance levodopa's effects. My general advice is to treat any supplement decision as a conversation with your clinician, not a solo experiment.
Questions readers ask
What are the most common citicoline side effects?
In trials, the most commonly reported effects were mild stomach upset, headache, and insomnia. Most were not severe enough to stop people from continuing the study.
Is citicoline safe for everyone?
No source supports that claim. It was generally well tolerated in trials at 250 to 2,000 mg per day, but trial populations do not include everyone. Pregnant and breastfeeding people, and anyone with complex medical conditions, fall outside the evidence.
Can I take citicoline while pregnant or breastfeeding?
There is no adequate human data for pregnancy or breastfeeding. Because data are lacking, it cannot be called safe in these situations. Discuss it with your obstetric clinician.
Does citicoline interact with medications?
The clearest concern is levodopa. Citicoline may enhance levodopa effects, so people with Parkinson's disease should involve their neurologist before taking it. For other medications, the evidence is limited, which is another reason to review your full medication list with your doctor or pharmacist.
Did the big stroke trial prove citicoline is safe?
ICTUS showed citicoline could be given to about 2,300 stroke patients with acceptable tolerability, but the trial found no improvement in 90-day recovery versus placebo. It was not designed to prove long-term supplement safety.
Is Cognizin different from plain citicoline for side effects?
Cognizin is a brand of citicoline, which is the same compound as CDP-choline. The safety evidence does not suggest a different side effect profile for the branded form.
Can citicoline cause insomnia?
Insomnia was among the mild effects reported in trials. If you notice sleep changes after starting it, raise that with your clinician rather than adjusting on your own.
Sources
This article draws on the following studies and reviews: ICTUS (Dávalos and colleagues, The Lancet, 2012, acute stroke, about 2,300 participants); COBRIT (Zafonte and colleagues, JAMA, 2012, traumatic brain injury); the Cochrane review by Fioravanti and Yanagi on chronic cerebrovascular cognitive decline; the IDEALE study (Cotroneo and colleagues) on mild vascular cognitive impairment; small glaucoma adjunct trials from the Parisi and Rossetti groups; and the attention studies by McGlade and colleagues (2012, adult women; and a separate study in adolescent males), both funded by the maker of Cognizin. PMIDs are not listed here; they will be added only after each one is confirmed on PubMed.
