What does low-dose naltrexone mean?
Naltrexone is an opioid-receptor blocker. In the United States, the 50 mg oral tablet is approved for alcohol dependence and for blocking the effects of administered opioids as part of addiction management. Low-dose naltrexone, usually shortened to LDN, refers to smaller amounts used off label for other conditions. It is the same drug, and a lower dose does not make its opioid interaction disappear. US prescribing label.
The studies discussed here used oral doses such as 4.5 mg or 6 mg. Those are descriptions of research, not a standard regimen for every person or condition. Fibromyalgia, Crohn's disease, multiple sclerosis and post-COVID fatigue each need their own evidence.
Why are people interested in it?
People with persistent pain or fatigue often want an option that might improve daily life without adding a large burden of side effects. Small early LDN trials produced encouraging results, and researchers have proposed effects involving opioid and immune signalling. Those proposals do not establish that LDN reduces brain inflammation in patients or identify who will benefit. The controlled clinical results matter more than the mechanism story.
What did the fibromyalgia trials find?
A 2013 crossover trial involving 31 women found a greater reduction in pain during LDN treatment than during placebo: 28.8% versus 18.0% from baseline. Mood and satisfaction with life also improved, but fatigue and sleep did not. This was a useful early signal from a small study, not a settled result. 2013 trial.
Three later placebo-controlled trials did not confirm a clear primary pain benefit:
- The 2023 crossover trial: 58 people were randomized to 4.5 mg LDN and placebo periods. Neither the revised Fibromyalgia Impact Questionnaire nor the summed pain ratings showed a clinically relevant advantage. 2023 trial.
- FINAL, published in 2024: 99 women received 6 mg LDN or placebo for 12 weeks. Pain fell by 1.3 points with LDN and 0.9 with placebo on a 0-to-10 scale. The between-group difference was not statistically significant. A possible memory-symptom benefit was a secondary finding for further study. FINAL trial.
- INNOVA, published in 2026: 98 women were randomized to 4.5 mg LDN or placebo as add-on treatment over 12 months. The primary pain outcome at three months did not favor LDN, and later pain-related outcomes did not show a meaningful advantage. Two participants were excluded after randomization, and about 65% of the remaining group stayed in the study at 12 months. INNOVA trial.
Someone can improve while taking a medicine even when the medicine does not outperform placebo in a trial. That does not dismiss their experience. It means that improvement alone cannot tell us how much the drug caused. The newer trials make routine promises of fibromyalgia pain relief difficult to support.
Does the answer change for other conditions?
Crohn's disease: a 2011 trial randomized 34 adults and found better clinical and endoscopic response with LDN. Response means improvement; it is different from remission. A newer trial, published in September 2026, randomized 41 adults and stopped early for futility. It found no improvement in the primary outcome of endoscopic remission, and clinical remission favored placebo in the smaller group with available data. A secondary fatigue score favored LDN, while work impairment favored placebo. Early stopping and missing measurements limit the estimates. This does not support replacing established Crohn's treatment with LDN. 2011 trial; 2026 trial.
Multiple sclerosis: an 80-person pilot, with 60 completing the trial, found improvements in some mental-health quality-of-life measures. Dropouts and data-management problems reduced its statistical power. A separate 96-person trial did not show meaningful differences in most quality-of-life measures. These studies did not establish prevention of relapses or disease progression. First MS trial; Second MS trial.
Depression: a 2026 trial randomized 37 people already receiving antidepressant treatment. Adding LDN did not improve the main depression rating more than placebo at 12 weeks, and it did not lower the measured inflammatory marker, hsCRP. This small study does not settle every psychiatric use, but it does not support an antidepressant claim. 2026 depression trial.
What is new in post-COVID fatigue?
A randomized study posted on October 26, 2026 is a preprint that has not been peer reviewed. It enrolled 160 adults in British Columbia and Ontario with persistent post-COVID symptoms meeting the study's fatigue-syndrome criteria. Eighty-one received LDN and 79 placebo for 16 weeks. The study used oral doses from 1 mg up to 4.5 mg as tolerated. Post-COVID trial preprint.
Fatigue improved in both groups, but LDN did not significantly outperform placebo on the primary fatigue score. Pain, symptom burden, daily steps and quality of life also showed no significant advantage. There were 137 participants with week-16 questionnaire data; the repeated-measures analysis used available observations over time. Grade 3 or 4 adverse events occurred in 5 of 81 LDN participants and 3 of 79 placebo participants. These events are not automatically proof of drug causation, but the result should not be summarized as risk-free.
The linked ClinicalTrials.gov record is completed with actual enrollment of 160 and still has no posted results. The preprint supplies results separately. This matters because a registry can say “no results posted” even when a report is available elsewhere. Trial registry.
What are the benefits and limits worth discussing?
Favorable findings
- Pain improvement in the small 2013 fibromyalgia trial.
- Clinical and endoscopic response in an early Crohn's trial.
- Selected fatigue or quality-of-life outcomes in some studies.
Findings that limit use
- Later fibromyalgia trials missed their primary pain outcomes.
- The 2026 Crohn's trial did not improve remission.
- The post-COVID preprint found no fatigue advantage.
Questions to bring
- Which condition and symptom are we trying to change?
- Could any current medicine interact with naltrexone?
- How will benefit, side effects and a stopping point be assessed?
What does safety look like at a low dose?
In FINAL, adverse events were reported by 84% of the LDN group and 86% of placebo; 8% versus 6% stopped because of adverse events. In INNOVA, the corresponding overall event rates were 68.8% and 72%, with no treatment-related serious events reported. These are reassuring observations within those selected trial populations, but they do not establish safety for every patient or drug combination. FINAL; INNOVA.
The opioid interaction deserves particular attention. Naltrexone can block opioid pain relief and precipitate severe withdrawal in someone who is opioid dependent. The prescribing label contraindicates use with opioid analgesics, current opioid dependence, or acute withdrawal. A published case describes severe precipitated withdrawal after just 1 mg in a person taking long-term opioids. A case report cannot tell us how often this happens, but it shows that low dose does not mean no risk. Prescribing label; Low-dose withdrawal case.
A clinician needs the full medication and substance history, including tramadol, methadone, buprenorphine and any planned procedure requiring pain relief. Liver disease, kidney function, pregnancy and breastfeeding also need review. The naltrexone label warns about hepatic injury and the dangers of trying to overcome opioid blockade. Medication changes and any transition off opioids need an individual clinical plan.
Is LDN an approved product for pain or fatigue?
The oral naltrexone label does not approve these low-dose uses for fibromyalgia, post-COVID fatigue, Crohn's disease or MS. A clinician may consider an off-label prescription, and a low-dose preparation may be compounded. In the United States, compounded drugs are not FDA approved or reviewed for safety, effectiveness and quality in the same way as approved drugs. The prescriber and pharmacist need to establish the appropriate product and the requirements that apply where it will be used. Naltrexone label; FDA compounding guidance.
Fibromyalgia
An early positive pain signal was not confirmed by three later trials.
Other conditions
Crohn's and MS results vary by the outcome measured.
Post-COVID
A 160-person randomized preprint reports no fatigue advantage.
Ongoing work
The CRPS registry remains recruiting with no posted results.
Research continues. The Stanford complex regional pain syndrome record lists recruiting status and an estimated 120 participants, with no posted results as of November 9, 2026. It is a plan to answer a question, not evidence that the treatment works. CRPS registry. Related records are searchable on our Science page.
Questions people ask
Is low-dose naltrexone FDA approved for fibromyalgia?
No. Oral naltrexone has approved uses for alcohol dependence and opioid blockade as part of addiction management. Its low-dose use for fibromyalgia is off label. US prescribing label.
Does LDN reliably reduce fibromyalgia pain?
The small 2013 crossover study reported a benefit, but the 2023 crossover trial, FINAL and INNOVA did not demonstrate a significant primary pain advantage over placebo. The evidence does not support promising reliable pain relief. 2013 trial; 2023 trial; FINAL; INNOVA.
Can a low dose still interact with opioids?
Yes. Naltrexone remains an opioid antagonist. It can interfere with opioid pain relief and precipitate withdrawal in an opioid-dependent person. Severe withdrawal has been reported after 1 mg, so a lower dose does not remove the need for a clinician to review current medicines and opioid exposure. Prescribing label; Case report.
Has a randomized trial reported results for post-COVID fatigue?
Yes. An October 2026 preprint reports a 160-person placebo-controlled trial with no significant benefit on its primary fatigue outcome at 16 weeks. It has not been peer reviewed. The linked registry still has no posted results because the preprint and registry are separate reporting routes. Preprint; Registry.
The evidence
Selected original studies, trial records and regulatory sources checked November 9, 2026. Study doses describe the research and are not instructions for use.
This article is for educational purposes only and is not medical advice, a diagnosis, or a treatment recommendation. Evidence and approvals apply to the products and populations described. Any treatment decision should be made with a qualified physician. Individual results vary.