Key takeaways
- Early interest rested on small trials, including a 2013 fibromyalgia study in which pain fell more on LDN than on placebo. 2013 crossover trial.
- A later trial in 99 women found no better pain relief than placebo. Danish randomised trial.
- The 2026 INNOVA trial also found no clinically meaningful benefit for pain, although people generally tolerated it well. INNOVA trial.
- Ideas about LDN calming immune cells and pain signalling are still hypotheses. Nothing shows that it “resets” the immune system. Mechanistic review.
- Naltrexone blocks opioid receptors. It can stop opioid painkillers from working and trigger withdrawal in someone who is dependent on opioids. FDA Revia prescribing information.
Low-dose naltrexone, usually shortened to LDN, has been used off-label for fibromyalgia for more than a decade, on the strength of small crossover trials and a great deal of word of mouth. It is also inexpensive, which matters to anyone paying for a long list of treatments that have not worked.
Two larger placebo-controlled trials have since reported. In the Danish one, pain scores fell by 1.3 points on a 0 to 10 scale with LDN and by 0.9 points with placebo, a difference the trial could not distinguish from chance. Danish randomised trial. In INNOVA, published in 2026, the placebo group improved slightly more than the LDN group at three months. INNOVA trial.
Why did LDN become a chronic-pain talking point?
A small trial with a positive result, plus a plausible theory, made it worth studying.
Naltrexone is an established medicine in addiction treatment. “LDN” means using it at much lower exposure for other purposes, including chronic pain, which is an off-label use.
The 2013 crossover trial, itself a follow-up to an earlier study, followed 31 women with fibromyalgia, each of whom spent time on naltrexone and on placebo. Pain fell by 28.8% from baseline on naltrexone and by 18.0% on placebo, and 32% of participants met the trial’s definition of a response on the medicine against 11% on placebo. Mood and satisfaction with life improved too, while fatigue and sleep did not. The authors themselves called for larger parallel-group studies. Early trial.
The theory involves microglia, immune cells that take part in signalling within the nervous system. A 2014 review of that idea, whose lead author also led the 2013 trial, still called LDN highly experimental and noted how small the published trials were and how few had been repeated. None of it shows that your pain comes from an immune process, or that LDN corrects one. Mechanistic review.
What did the newer trials find?
Neither of the two larger trials found a clear pain benefit over placebo.
| Study | Participants and follow-up | Main finding |
|---|---|---|
| 2013 crossover trial | 31 women with fibromyalgia | Greater pain reduction during LDN treatment |
| Danish parallel-group trial | 99 women; 12 weeks | No statistically significant superiority for pain |
| 2026 INNOVA trial | 98 women; followed for 12 months | No clinically meaningful benefit for pain-related outcomes |
Sources: 2013 trial, Danish trial, INNOVA.
The Danish researchers screened 158 women, randomised 99 and lost none of them to follow-up. Side effects were about as common on placebo as on LDN: 84% of the LDN group and 86% of the placebo group reported at least one. The team did see a possible effect on memory problems and suggested testing it properly, which is a reason for more research rather than evidence that LDN helps thinking.
INNOVA took its main reading at three months, when pain had fallen by 0.33 points on LDN and by 0.64 points on placebo. It then followed its 98 participants to 12 months. Secondary measures, from anxiety and depression to disability and worry, moved little and inconsistently, and responder rates were low in both groups.
Every participant in these trials was a woman with fibromyalgia. Men were not studied, and neither were other conditions LDN is used for, such as Crohn’s disease and multiple sclerosis, both named in the 2014 review. Mechanistic review. The trials do make “proven fibromyalgia relief” an inaccurate description.
Can patient experience still matter when a trial is negative?
Yes, although improving during treatment and improving because of it are separate questions.
The placebo groups show why. In the Danish trial, women taking placebo saw their pain fall by 0.9 points on the 0 to 10 scale, and in INNOVA the placebo group did slightly better than the LDN group. Fibromyalgia pain varies from week to week, and other parts of life and care change at the same time as a new medicine starts. Belief may play a part as well: in exploratory analyses, INNOVA participants who thought they had received LDN showed some improvement, although the effects were inconsistent. INNOVA trial.
The 2013 trial asked its participants to rate their pain every day. Borrowing that idea, with a daily 0 to 10 score started before the first dose, gives you and a prescriber something firmer to judge by than a general impression. Agree in advance what change would justify carrying on, and when you would stop. Our pregabalin review asks the same question of pregabalin: which pain conditions does the evidence actually cover? Fatigue that persists needs its own assessment, and our article on thyroid treatment and ongoing symptoms covers one setting where that happens.
Why does the opioid interaction matter even at low exposure?
Even at low exposure, naltrexone still blocks opioid receptors.
A prescriber needs to know about any codeine, tramadol, morphine, methadone or other opioid you take, including combination painkillers that contain codeine. Naltrexone can stop an opioid painkiller from working. For someone physically dependent on opioids, starting it can bring on abrupt withdrawal. Taking extra opioid to overcome the blockade is dangerous. FDA prescribing information.
No washout timetable copied from a blog is safe for everyone. The prescriber has to account for the specific opioid, any dependence, other illnesses and the possibility of emergency or surgical pain. Commercial presentations and compounded preparations should also be checked individually rather than assumed equivalent.
Do and don’t
Do: Ask which condition-specific evidence supports a trial of LDN, and tell the prescriber about every medicine you take that contains an opioid.
Don’t: Describe LDN as an immune reset, combine it with opioids on your own or treat an anecdote as a guarantee.
