Sources report side effects of low-dose naltrexone (LDN) differently, and they differ in dose, design and size. A 2025 meta-analysis of randomized trials in chronic pain syndromes (dose not stated) reported that adverse events were increased with LDN compared to placebo.
Trials and pilot studies of LDN
| Study | Dose and design | Safety findings |
|---|---|---|
| Danish trial, women with fibromyalgia | 6 mg once daily. Single-centre, randomised, double-blind, placebo-controlled, 1:1. | Adverse events: 41 (84%) of 49 on LDN, 43 (86%) of 50 on placebo. Discontinuations due to adverse events: 4 (8%) of 49 on LDN and 3 (6%) of 50 on placebo. One serious adverse event, in the placebo group. No deaths. |
| Crossover trial, women with fibromyalgia | 4.5 mg oral naltrexone daily. Small randomized, double-blind, placebo-controlled crossover. | LDN was rated equally tolerable as placebo. No serious side effects were reported. |
| Pilot, women with fibromyalgia | 4.5 mg (low doses). Single-blind crossover pilot. | Side effects, including insomnia and vivid dreams, were rare and described as minor and transient, per the abstract. |
| Multiple sclerosis pilot | 4.5mg nightly for 8 weeks. Single-center, double-masked, placebo-controlled, crossover pilot. | The authors say high dropout and data management errors substantially reduced statistical power, LDN was well tolerated, and serious adverse events did not occur. |
| Children with Crohn's disease | 0.1 mg/kg. Pilot clinical trial. | Oral naltrexone was well tolerated without any serious adverse events. |
Reviews and meta-analyses
A fibromyalgia meta-analysis of randomized trials versus placebo (dose not stated) could not associate LDN use with the incidence of headache (RR 1.68; 95% CI 0.84, 3.36; sensitivity analysis RR 0.98; 95% CI 0.72, 1.34). Vivid dreams (RR 2.41; 95% CI 1.77, 3.28) were significantly higher with LDN than placebo.
A Cochrane review in Crohn's disease covered two studies: 12 weeks of LDN (4.5 mg/day) versus placebo in adults (N = 34), and eight weeks of LDN (0.1 mg/kg, maximum 4.5 mg/day) versus placebo in pediatric patients (N = 12). Pooled data: no statistically significant differences in withdrawals due to adverse events or in specific adverse events (sleep disturbance, unusual dreams, headache, decreased appetite, nausea, fatigue). No serious adverse events were reported in either study. The review found insufficient evidence for firm conclusions on efficacy and safety.
A 2018 review says evidence supports safety and tolerability in multiple sclerosis, fibromyalgia and Crohn's disease, but fewer studies support efficacy, most using subjective measures. A 2023 fibromyalgia review calls the evidence limited. A 2014 review calls LDN use for chronic disorders highly experimental, with low sample sizes and few replications.
Observational studies, which cannot show cause and effect
- Academic pain clinic, retrospective cohort (dose <10 mg for at least one month): 87% (27) discontinued LDN: 52% (16) for lack of benefit, 23% (7) for loss of benefit, 10% (3) for side effects, 3% (1) for other reasons. Seven (23%) reported side effects.
- Veterans with chronic pain, retrospective chart review (dose not stated): 32% (13 of 41) reported adverse effects including vivid dreams, drowsiness, dizziness, and nausea.
- One physician's practice, retrospective cohort (dose not stated): adverse effects occurred in 49.5%, most frequently nausea and fatigue. No serious adverse effects were reported.
- Post COVID-19 Syndrome, single-centre pre-post study (1 mg month one, 2 mg month two): Thirty-eight participants (73.1%) were known to commence LDN; two (5.3%) stopped due to new onset diarrhoea and also described fatigue.
The 50 mg label: a different product
These figures come from the prescribing information for naltrexone hydrochloride tablets (50 mg) and describe people with alcohol dependence or opioid addiction. They are not low-dose naltrexone figures.
- Two randomized, double-blind, placebo-controlled 12-week trials in alcohol dependence: most patients tolerated it well. Of 93 given 50 mg once daily, five discontinued because of nausea. No serious adverse events were reported.
- Open label safety study, approximately 570 individuals with alcoholism, new-onset adverse reactions in 2% or more: nausea (10%), headache (7%), dizziness (4%), nervousness (4%), fatigue (4%), insomnia (3%), vomiting (3%), anxiety (2%) and somnolence (2%).
- In opioid addiction trials, reported at baseline and during the trials at more than 10%: difficulty sleeping, anxiety, nervousness, abdominal pain/cramps, nausea and/or vomiting, low energy, joint and muscle pain, and headache.
- Postmarketing data: most events usually occur early and are transient, and are not always distinguishable from withdrawal syndrome symptoms. Events reported include anorexia, asthenia, chest pain, fatigue, headache, hot flushes, malaise, changes in blood pressure, agitation, dizziness, hyperkinesia, nausea, vomiting, tremor, abdominal pain, diarrhea, palpitations, myalgia, anxiety, confusion, euphoria, hallucinations, insomnia, nervousness, somnolence, abnormal thinking, dyspnea, rash, increased sweating, vision abnormalities, and idiopathic thrombocytopenic purpura.
What this does not tell you
Disclosures: the conflict of interest statement of the Cochrane review says one person named in it received consultancy fees from AbbVie, Ferring, Janssen, Takeda, Lupin, Shire and Allergan and speakers' fees from AbbVie, Takeda, Shire and Allergan, all outside the submitted work. An author of the children's Crohn's pilot has intellectual property rights and is a co-inventor on a US patent for the use of naltrexone in inflammatory bowel disease. The Post COVID-19 Syndrome study reports financial support from UCD Foundation (Gilead, Pfizer, GSK Newman Fellowship).
These sources give no side effect rate for any one person and do not show LDN is free of side effects. "Well tolerated" describes the groups studied. A licensed provider decides what is suitable for a given person.