Randomised controlled trial of fluoxetine versus naltrexone in compulsive sexual behaviour disorder
In brief
Naltrexone did not outperform fluoxetine for compulsive sexual behavior
In an 8-week trial of 80 people, almost all men, both drugs reduced symptom scores, with no significant difference between them at the main endpoint. Improvement patterns and side effects differed, and more participants stopped naltrexone than fluoxetine; the small, short trial leaves open which patients might benefit more from either treatment.
- Journal
- BMJ mental health (Q1)
- Published
- 6 October 2026
- Study design
- Randomized controlled trial
- Evidence level
- Level 1, High (CEBM 1b)
- Authors
- Josephine Savard, Marta Piwowar, Roberth Adebahr, Markus Byström, Jonas Hallberg, Cecilia Dhejne, et al.
- PMID
- 42838712
- DOI
- 10.1136/bmjment-2026-302953
Why clinicians should know about it
- Picked for Psychiatry and Mental Health (paper of the day, 7 October 2026): Open-label RCT comparing fluoxetine and naltrexone for CSBD
Abstract
BACKGROUND: Compulsive sexual behaviour disorder (CSBD), characterised by preoccupation with and loss of control over sexual behaviours, is associated with negative consequences in several areas of life. Evidence for pharmacological treatment remains limited, with only two previous randomised controlled trials and no studies comparing fluoxetine and naltrexone. OBJECTIVE: To investigate whether naltrexone was more efficient and tolerable than fluoxetine in the treatment of CSBD. METHODS: In this open-label superiority trial, 80 participants diagnosed with CSBD according to International Classification of Diseases-11 criteria were randomly assigned (1:1) to receive fluoxetine (starting dose 20 mg/day) or naltrexone (starting dose 25 mg/day) for 8 weeks followed by 6 weeks without treatment. Symptom severity was recurringly assessed using the Hypersexual Disorder: Current Assessment Scale (HD:CAS). Linear mixed models were used for intention-to-treat and per-protocol analyses of treatment effects. FINDINGS: 79 men and 1 woman were randomised to receive fluoxetine (n=40) or naltrexone (n=40). Analysis showed no statistically significant difference in HD:CAS score reduction between treatments at primary endpoint; naltrexone was not superior to fluoxetine. However, participants in the two treatment groups showed different patterns of symptom improvement over time.Two participants (5.1%) receiving fluoxetine were withdrawn due to adverse events (urticaria and elevated liver enzymes) and four participants (10.3%) discontinued treatment; six participants (15%) receiving naltrexone discontinued treatment. CONCLUSIONS: Although no statistically significant between-group difference was observed at week 8, symptom trajectories and adverse effect profiles differed between groups. CLINICAL IMPLICATIONS: Symptom-change patterns and adverse effect profiles may guide individualised treatment decisions.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.