Six trials of low-dose naltrexone (LDN) in fibromyalgia are described here, with doses from 0.75 mg to 6 mg. The 2 earliest small trials, from 2009 and 2013, reported reductions in symptoms or pain. In 3 later randomized placebo-controlled trials (58 patients, 99 women and 98 women), the authors concluded the results did not show a benefit over placebo. Reviews reach different conclusions. LDN is an off-label use.
The trials
| Trial | Dose, design, size, length and what the abstract reports |
|---|---|
| 2009 pilot, Pain Medicine | 4.5 mg. Single-blind crossover pilot trial with a placebo comparison. 10 women. Length not given. Fibromyalgia symptoms were reduced in the entire cohort, with a greater than 30% reduction of symptoms over placebo. |
| 2013 trial, Arthritis & Rheumatism | 4.5 mg of oral naltrexone daily in the active phase. Randomized, double-blind, placebo-controlled, counterbalanced crossover. 31 women. Length not given. A significantly greater reduction of baseline pain with LDN than with placebo (28.8% reduction versus 18.0% reduction; P = 0.016). |
| 2020 dose-response trial, Pain Medicine | Test doses 0.75-6 mg. Single-blinded. 27 women included, 2 withdrawn, 25 evaluable. Length not given. Dose estimates were 3.88 mg for ED50 and 5.40 mg for ED95. The passage does not say what outcome these refer to. |
| 2023 study, PAIN Reports | 4.5 mg once daily. Randomized, double-blind, placebo-controlled crossover. 58 patients. Length not given. The median difference in FIQR scores between LDN and placebo was -1.65 (IQR 18.55; effect size = 0.15; P = 0.3). The authors concluded that outcome data did not indicate any clinically relevant analgesic efficacy. |
| 2024 Danish trial, The Lancet Rheumatology | 6 mg once daily for 12 weeks. Single-centre, double-blind, placebo-controlled, randomised 1:1 to LDN or an identical-appearing placebo. 99 women. The aim was to investigate whether LDN was superior to placebo for reducing pain. The authors concluded the study did not show that LDN was superior to placebo in relieving pain. |
| 2026 INNOVA trial, European Journal of Pain | 4.5 mg/day for 12 months. Randomised, double-blind, placebo-controlled, single-centre. 98 women. The authors concluded LDN was well tolerated but did not demonstrate a clinically meaningful benefit over placebo for pain-related outcomes. |
What the reviews concluded
- A 2023 systematic review concluded that the strength of evidence from randomized controlled trials to support the use of LDN among patients with fibromyalgia is low.
- Another 2023 systematic review said proving the efficacy and safety of LDN is a future possibility based on current study data, but the level of scientific evidence is limited.
- A 2025 systematic review and meta-analysis said LDN marginally reduced pain and symptom severity from baseline, but these effects were not superior to placebo, and that LDN may not provide significant clinical benefit over placebo.
- A 2025 meta-analysis of 5 randomized controlled trials (the abstract does not state the dose) reported LDN was superior to placebo in alleviating pain in its primary analysis (SMD -0.61; 95% CI -1.14, -0.08) and sensitivity analysis (SMD -0.87; 95% CI -1.28, -0.46) but not in raising mechanical pain threshold (SMD 0.24; 95% CI -0.09, 0.56).
These reviews reached different conclusions. This guide does not decide between them.
Side effects the abstracts reported
- In the 2024 Danish trial (6 mg), discontinuations due to adverse events were 4 (8%) of 49 in the LDN group and 3 (6%) of 50 in the placebo group. Adverse events occurred in 41 (84%) of 49 versus 43 (86%) of 50. One serious adverse event occurred in the placebo group and no deaths occurred.
- The 2013 crossover trial (4.5 mg) reported LDN was rated equally tolerable as placebo and no serious side effects were reported.
- The 2009 pilot (4.5 mg) reported side effects, including insomnia and vivid dreams, were rare, and described as minor and transient.
- The 2025 meta-analysis reported that neither its primary analysis (RR 1.68; 95% CI 0.84, 3.36) nor its sensitivity analysis (RR 0.98; 95% CI 0.72, 1.34) could associate LDN use with headache, and that vivid dreams (RR 2.41; 95% CI 1.77, 3.28) were significantly higher in the treatment group than placebo.
What this does not tell you
These are descriptions from abstracts, not full papers. The trials differ in dose, design and size, and the passages used here give a length for only 2 of the 6, so results are not directly comparable.
The 2009 pilot and the 2020 dose-response trial were small and single-blind, not double-blind, so they cannot show cause and effect. Where the abstracts say what was measured, it was pain, symptom or FIQR scores in the groups studied. Group scores do not tell you how one person would respond. This guide has no disclosure or funding statements for these fibromyalgia studies, so it cannot say whether any author or funder ties exist. A result in fibromyalgia does not apply to other conditions. A licensed provider decides whether LDN is appropriate for a person.