Article Text

Download PDFPDF

Paroxetine and sertraline delayed ejaculation in heterosexual men with lifelong rapid ejaculation
  1. Jose Catalán, MSc, DPM, FRCPsych
  1. Imperial College School of Medicine London, UK

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

OpenUrlCrossRefPubMedWeb of Science

QUESTION: In heterosexual men with lifelong rapid ejaculation, is paroxetine, sertraline, or nefazodone more effective than placebo in delaying ejaculation?


Randomised (allocation concealed*), blinded (clinicians and patients)*, placebo controlled trial with 6 weeks of follow up.


Outpatient Department of Psychiatry and Neurosexology at Leyenburg Hospital, The Netherlands.


48 heterosexual men (mean age 39 y) who had lifelong rapid ejaculation and a stable relationship with a female partner. Rapid ejaculation was defined as an intravaginal ejaculation latency time (IELT) of ≤1 minute. Exclusion criteria included erectile dysfunction, alcohol or substance abuse, mental disorders, physical illnesses, and concomitant medications. 83% of patients completed the study.


12 patients were allocated to each of the 4 treatment regimens: paroxetine (20 mg/d), sertraline (50 mg/d), nefazodone (400 mg/d), and placebo in the form of 2 identical capsules per day given in a morning and evening dose for 6 weeks.

Main outcome measure

IELT defined as the time between the start of vaginal intromission and the start of intravaginal ejaculation measured by partner with stopwatch. Mean IELTs were determined every week.

Main results

Analysis was by intention to treat. The four treatment groups differed for IELT during the study (p=0.002) and at the study endpoint (wk 6) (p=0.002) (table). The paroxetine (p<0.001) and sertraline (p=0.024) groups differed from placebo for increase in IELT from baseline. The nefazodone and placebo groups did not differ for the increase in IELT from baseline (p=0.85). Compared with baseline values, paroxetine exerted the strongest delay in ejaculation (9 fold increase), followed by sertraline (4 fold increase), nefazodone (2 fold increase), and placebo (2 fold increase).

Nefazodone, paroxetine, and sertraline v placebo for rapid ejaculation in heterosexual men†


In heterosexual men with lifelong rapid ejaculation, paroxetine and to a lesser degree sertraline were more effective than placebo in delaying ejaculation.


Erectile dysfunction and its treatment seem to have been attracting most of the attention in the field of psychosexual dysfunction, and it is a welcome development to see that other problems are also being addressed, even if within the relatively narrow focus of pharmacological interventions. In everyday clinical practice, the management of premature ejaculation [PE] remains essentially a cognitive behavioural one, based on the work of Masters and Johnson, but its relatively limited efficacy for PE makes it important to consider other alternatives. The study by Waldinger et al is an exciting development, but one that raises a number of important practical questions that need answering before advocating the routine clinical use of selective serotonin reuptake inhibitors (SSRIs) for the treatment of PE.

Firstly, less than a third of potential participants with PE (48/142) were included in the study, and while reasons for exclusion were sensible (eg, relationship or personal problems, physical disease, and partner′s sexual dysfunction), they raise the question of how many men with PE would actually be eligible for drug treatment alone. Secondly, the main outcome measure (time from penetration to ejaculation) is seemingly a good one, but one that does not take into account the female partner′s own assessment of the effects of the intervention. Admittedly, it would have been difficult for her to operate the stopwatch as well as allow herself to become aroused and then try to rate her own arousal level, and understandably the focus of the research here was only the ejaculation latency time. While the statistical significance of the results is strong, their clinical significance is less so: would an ejaculation latency time of a little over 2 minutes (paroxetine) or just under 1 minute (sertraline) be of practical value to these men and their partners? The answer is far from obvious at this stage. The third question that arises concerns the duration of treatment. Although the effects of 2 of the experimental drugs are apparent within a week of intervention, their maximum efficacy is reached after 5 weeks, and this would suggest that brief intervention (ie, less than a week) is not appropriate. No information is available on the effects of discontinuing SSRIs on sexual function, but it could be assumed that long term use may be necessary to maintain ejaculation latency time, and this raises other questions about the use of such substances as a long term treatment. It is also interesting to note that a small number of participants dropped out because of drug side effects, while another small group reported reduction in sexual desire.


  • Source of funding: Bristol Myers Squibb Holland.

  • For correspondence: Dr M D Waldinger, Department of Psychiatry and Neurosexology, Leyenburg Hospital, Leyweg 275, 2545 CH, The Hague, The Netherlands.

  • * See glossary.