Both drugs eased symptoms by about half, and naltrexone got there faster
STUDY: Savard J et al, BMJ Ment Health 2026
STUDY TYPE: Randomized controlled trial (open label)
FUNDING: Swedish Research Council, regional grants, private foundations
Background
Compulsive sexual behavior disorder (hypersexuality or sex addiciton) is characterized by sexual impulses or urges that cause significant distress, interfere with life (such as risky behaviors or taking up a lot of time), and are difficult to control, lasting at least six months. It is recognized in ICD (6C72) but not in DSM.
Therapy is first line, and guidelines suggest an SSRI or naltrexone when therapy alone falls short. SSRIs lower sex drive, while naltrexone blocks opioid receptors and may dampen craving. Only two randomized, placebo-controlled trials have tested these, with one showing benefits for citalopram (20-60 mg/day). The other produced mixed results for paroxetine and naltrexone, but was probably underpowered to detect a difference.
This first compares the two medication options, without a placebo control.
The Study
- 80 adults (79 men) in Stockholm with about 15 years of symptoms. 66 completed treatment.
- Fluoxetine 20 mg (40 mg if needed, which 20% did) vs naltrexone 50 mg for 8 weeks, then 6 weeks off medication.
- Primary outcome: Hypersexual Disorder scale (HD:CAS, scored 0 to 24; baseline median 14).
Results
There was no significant difference between the two at 8 weeks (reduction on scale of 7 with fluoxetine vs 8.5 with naltrexone), but naltrexone showed benefits by 2 weeks while fluoxetine took 6 weeks.
Symptoms returned after both medications were stopped.
About 40% in each group wanted to restart their drug at the end.
Side Effects
- Fluoxetine: lower libido (72%), delayed ejaculation (46%), fatigue (26%). Two patients stopped for hives or elevated liver enzymes.
- Naltrexone: lower libido (60%), fatigue (58%), nausea (40%).
Limitations
- Small, open label with no placebo group, so all changes could be due to placebo effect, even the worsening off the medication.
- Nearly all men, brief follow-up.
Practice Implications
- This study at least suggests the two meds recommended in guidelines are reasonable. The choice can be personalized: naltrexone for craving-driven behavior or a substance use history, fluoxetine for comorbid depression or intrusive sexual thoughts.
- Start naltrexone at 25 mg, raise to 50 mg after 3 to 5 days, and recheck symptoms at 2 weeks. Main side effect is nausea, but warning: liver enzyme elevation.
—Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report
