Premature ejaculation (PE) is thought to be the most common male sexual dysfunction; however, the prevalence of lifelong (LL)-PE is relatively low. The aim of this study was to investigate the effects of on-demand vardenafil (10 mg) to modify the intravaginal ejaculatory latency time (IELT) in men with LL-PE without erectile dysfunction. Forty-two men (18–35 years) were enrolled in a 16-week, double-blind, placebo-controlled, cross-over study. Primary end point was the modification from baseline of IELT assessed by stopwatch technique; secondary end points were post-ejaculatory refractory time (PERT) and variations of scores at the Index of Premature Ejaculation questionnaire. The changes in geometric mean IELT were superior after taking vardenafil (0.6±0.3 vs 4.5±1.1 min, P<0.01), compared with placebo (0.7±0.3 vs 0.9±1.0 min, ns). PERT dropped significantly after vardenafil (16.7±2.0 vs 4.3±0.9 min, P<0.001), compared with placebo (15.3±2.2 vs 15.8±2.3 min). Patients who took vardenafil (vs placebo) reported significantly (P<0.01) increased ejaculatory control (6±2 vs 16±2), improved overall sexual satisfaction (7±2 vs 15±1) and distress (4±1 vs 8±1) scores, respectively. Multiple regression analysis (r2=0.86) for IELT by the number of attempts at sexual intercourse showed significant differences between the slopes of lines for placebo and vardenafil (P<0.0001). The most common adverse events for vardenafil (vs placebo) were headache (10 vs 3%), flushing (12 vs 0%) and dyspepsia (10 vs 0%), which tended to disappear over the time. In conclusion, in our study, vardenafil increased IELT and reduced PERT in men with LL-PE. Besides, improvements in confidence, perception of ejaculatory control and overall sexual satisfaction were reported.
Although premature ejaculation (PE) is thought to be the most common male sexual dysfunction, with a self-reported prevalence of 21–33%.1 There are indications from controlled studies that the prevalence of lifelong (LL)-PE is much lower (1–8%).2 LL-PE is defined as a ‘male sexual dysfunction characterized by ejaculation, which always or nearly always occurs before or within about 1 minute after vaginal penetration, and the inability to delay ejaculation on all or nearly all vaginal penetrations, and negative personal consequences, such as distress, bother, frustration and/or the avoidance of sexual intimacy.3 Waldinger and Schweitzer4 recently proposed a new ‘classification’ of PE by distinguishing four PE syndromes on the basis of the duration of the ejaculation time, frequency of complaints and course in life. In this new classification, meant for the Diagnostic and Statistical Manual of Mental Disorders (DSM-V), the DSM and International Classification of Diseases (ICD) requirements for a ‘short’ ejaculation time for the diagnosis of PE are maintained, but added with two new PE categories defined in terms of longer ejaculation times and differences in frequency of occurrence of PE complaints. Moreover, the new classification underlines the multidimensional aspects of PE by reintroducing the criterion of control of ability to delay ejaculation yet restricted by the criterion of time, as is also required by the ICD-10. The multifactorial nature of PE suggests strongly that outcome measurements in PE should be more than the intravaginal ejaculatory latency time (IELT) alone and fully validated multidimensional measures to assess different aspects are required. In most studies, IELT is believed to be the most sensitive parameter for measuring the efficacy of treatment of PE.5, 6 We are aware that although it is an accepted tool for assessing PE, currently available IELT measurement methods are not ideal because inherent subjective difficulties in measurement can cause bias in the evaluation of the results. New convenient and efficient technology to measure it, which will be less dependent on the attention of the sexual partners, would be very useful.
At present, there is only one ‘uncontrolled’ study reporting on the use of vardenafil in men with primary (lifelong) or secondary (acquired) PE by Mathers et al.7; in this study, the authors reported that PE grade improved 2.7 points in the vardenafil group vs 1.9 in the sertraline group, and IELT increased by 5.01 min (arithmetic mean) in the vardenafil group and 3.20 min in the sertraline group. However, this was an open trial comparing a phosphodiesterase type-5 inhibitor (PDE5i) with a Serotonin-Selective Reuptake Inhibitor (SSRI) with no placebo arm, and included most individuals suffering from secondary PE, presumably as a result of untreated erectile dysfunction (ED).
This was a randomized, double-blind, placebo-controlled, parallel arms, cross-over study to investigate the effects of on-demand vardenafil on IELT by the stopwatch technique in men with LL-PE and without ED. The secondary end point was the post-ejaculatory refractory time (PERT), which has been shown earlier to drop significantly in normal healthy volunteers after assumption of the PDE5i, sildenafil,8, 9 and may represent an advantage for treating men with LL-PE, in that it shortens the time to re-obtain a new erection for attempting another sexual intercourse. We also used a validated 10-item self-administered tool, the Index of Premature Ejaculation (IPE) questionnaire,10 which accurately assesses the subjective aspects of PE which the patients have identified as being important and relevant to their sexual life.
Patients and methods
Vardenafil näive men aged 18–35 years were included if they met the International Society for Sexual Medicine (ISSM) criteria for LL-PE,3 and had at least 6 months sexual contact with their current partner. This was an investigator-initiated study that received the approval of the Internal Review Board of our Institution. A score 11 of the Premature Ejaculation Diagnostic Tool was also used to confirm the presence of PE.11 Also, patients were included if they had a score of 22 on the erectile function (EF) domain of the International Index of Erectile Function (IIEF).12 They were entered into a 4-week run-in period, during which a diary of all sexual activity was kept. Patients who reported at least one intercourse episode per week and IELT 1 min in 90% of the intercourse attempts during the run-in period were enrolled and randomized to receive vardenafil 10 mg or matching placebo for 8 weeks in the double-blind, placebo-controlled trial. At the end, each patient entered a 4-week cross-over period in which they had to assume at least four tablets of the alternative treatment. Patients had to remain in a stable, single-partner relationship and have at least one sexual intercourse episode per week throughout the treatment period.
Patients were excluded if they had IELT 1 min in 90% of intercourse attempts. Patients were further excluded if they had a history of ED (score of <22 on EF domain of the IIEF)12 or other ejaculatory dysfunctions, or any concomitant treatment with SSRIs. A score of <22 on the EF domain was selected as an exclusion criterion to limit the possibility of a response to vardenafil as a result of treating comorbid moderate to severe ED. Patients were also excluded if they used condoms or masturbated before sexual intercourse for purposes of decreasing penile sensitivity, or used any other treatment for PE. Patients were excluded if they had a history of vascular disease, including stroke, myocardial infarction, unstable angina or life-threatening arrhythmias within the last 6 months, or were using organic nitrates or cytochrome P450 inhibitors. Moreover, screening for hyperthyroidism (thyroid stimulating hormone, FT3, FT4), hypogonadism (total and free testosterone) and prostate smears for detecting acute or sub-acute prostatitis were carried out before the study entry.
Couples were instructed to attempt sexual intercourse four or more times during the 4-week run-in period and four or more times per month during the 12-week treatment period (minimum of 24 h between doses of medication), and to record IELT for the first event after each dose in the event log by the stopwatch technique. IELT was measured on the first coital attempt per 24-h period. Patients were randomly assigned within each stratum 1:3 by a computerized interactive system to receive placebo or fixed-dose vardenafil (10 mg), and were given 15 doses of the study medication (one dose was one tablet; tablets in all groups were identical in appearance as the active principle was encapsulated) each 4 weeks; one dose was to be taken 15–30 min before anticipated sexual intercourse with empty stomach, and only one dose was allowed per 24-h period. Ejaculation-delaying techniques and behavioral therapy were to be avoided. Also, couples were instructed not to use condoms or topical anesthetic cream, not to pause during intercourse or to have interrupted intromission. Furthermore, they were requested to measure the first intercourse IELT along with the time to re-obtain second erection for attempting another sexual intercourse. Patients agreed not to change the type of treatment during the study period. Treatment efficacy was assessed at 8 and 12 weeks.
The primary end point was IELT at week 8, defined as the time elapsed between penetration and ejaculation; ejaculation occurring before penetration was assigned an IELT of 0 min. Secondary end point was PERT, defined as the time to re-obtain erection after ejaculation and was measured by asking each participant to attempt a second sexual intercourse after the first one. Also, measurements of sexual bother related to the presence of PE was investigated by using the IPE questionnaire scores.11 This is a reliable and valid questionnaire for the assessment of control over ejaculation, satisfaction with sex life and distress in men with PE, and has the potential to add value to interpretations of improvements in ejaculation latency resulting from any treatments of PE. The highest are the scores, the better is the reported control over ejaculation. During the screening visit, participants and partners received instructions on the PERT and IELT measurement techniques, in which partners were to activate the supplied stopwatch on vaginal penetration during sexual intercourse and to stop the stopwatch on either intravaginal ejaculation or withdrawal without ejaculation. The time noted on the stopwatch at this point was recorded as the duration of sexual intercourse until ejaculation or withdrawal. Once detumescence occurred, both partners were instructed to re-start stopwatch until a second erection was obtained. This technique has been validated elsewhere.8
We tested for the differences between treatment groups by using analysis of variance. As the IELT distribution of any individual man behaves according to a curve that is more or less positively skewed depending on the investigated drug, we used the logarithmic transformation of individual IELT data to ‘geometric’ mean IELT values both at baseline and on drug treatment.13 Multiple regression analysis was carried out for geometric mean IELT vs sexual intercourse attempt number. A P-value <0.05±s.d. was considered statistically significant. Statistical analysis was carried out using the computer statistical package SPSS/4.0 (SPSS, Chicago, IL, USA) and SAS/6.4 (SAS Institute Cary, NC, USA).
All the observed and volunteered adverse events were recorded for all patients who took at least one dose of study medication.
All patients were seen with their partners and interviewed about their sexual activity and patient's ejaculation function. In total, 42 patients were recruited, but only 40 (95%) completed the whole randomized trial study (30 of 31 in the vardenafil group and 10 of 11 in the placebo group) (Figure 1). Two patients were non-completers. Reasons for non-completion included one adverse effect (headache/pyrosis in the vardenafil group) and one lack of effect on ejaculation (placebo group). No patient was lost at follow-up (Figure 1). The difference in dropout rates was not significant between the groups. There were no statistical differences in the patients' characteristics at baseline at the time of randomization (Table 1) and all patients assumed the medication in a mean time interval before sexual intercourse of 23±7 min. Mean patient age was 24 years (range 18–35). Mean PEDT scores served to recheck the severity of PE at baseline and resulted above the upper normal reference range for both groups of patients (18±0.8 s.d. vs 18±0.6; normal value <11, respectively).
Baseline (geometric mean±s.d.) IELT for patients randomized to vardenafil or placebo was 0.6±0.3 min and 0.7±0.3 min, respectively (Figure 2). At the end of treatment (8 weeks), IELT increased from 0.6±0.3 to 4.5±1.1 (P<0.01) with vardenafil and from 0.7±0.3 to 0.9±1.0 (ns) with placebo. At the time of follow-up (12 weeks), patients who crossed over from placebo to vardenafil reported significant improvements in IELT (from 0.9±1.0 to 2.0±0.9 min to, P<0.05), whereas in the other arm a significant reduction in IELT was found compared with the end of study (from 4.5±1.1 to 3.2±1.2, P<0.05); this latter IELT was still superior to baseline (P<0.01). This represents a mean change per patient of 3.8±1.3 min for patients taking vardenafil and 0.2±0.3 min for patients taking placebo. Thus, the magnitude of the increase in IELT compared with baseline was statistically significant (P<0.01; Figure 2). Compared with placebo, vardenafil significantly (P<0.001) decreased the PERT (16.7±2.0 vs 4.3±0.9 min) (Figure 3). It is interesting that at the time of follow-up, patients who crossed over from placebo to vardenafil reported significant improvements in PERT (from 15.8±2.5 min to 5.1±1.0, P<0.001), whereas in the other arm a significant increase in PERT was found compared with the end of study (from 4.3±0.9 to 14±1.2, P<0.001). Multiple regression analysis (r2=0.86) for IELT by the number of attempts of sexual intercourse showed that a significant difference existed between the slopes of the lines for placebo and vardenafil (P<0.0001, Figure 4).
Scores of each domain of the IPE questionnaire are shown in Table 2. Patients who took vardenafil (vs placebo) reported significantly (P<0.01) increased ejaculatory control (6±2 vs 16±2), improved overall sexual satisfaction (7±2 vs 15±1) and amelioration of distress (4±1 vs 8±1) scores on the IPE questionnaire. The most common adverse events were significantly higher (P<0.01) with vardenafil (vs placebo) after 4 weeks of treatment, and were headache (10 vs 1%), flushing (12 vs 0%), dyspepsia (10 vs 1%), which tended to disappear at the end of the study (Table 2).
As far as we are aware, this is the first controlled study using vardenafil for the treatment of LL-PE in men without ED. In this study, we demonstrate that fixed-dose vardenafil (10 mg), when given 15–30 min before intercourse, determines a robust increase in IELT when compared with placebo. This improvement is much higher in patients with more frequent sexual activity per week, as demonstrated by strong linear relationships between the number of consumed tablets and IELT. Vardenafil also improved PERT and patients' perception of control over ejaculation, satisfaction with sexual intercourse and overall impression of change in condition with modest and transient side effects. These benefits were sustained over the time, as shown by the persistent delay of IELT during the 1-month follow-up period with placebo.
At present, six studies concerning the use of PDE5i in treating PE have been published.14 Two were placebo-controlled studies. None fulfilled the criteria of a level 1 study (double-blind, placebo-controlled, randomized study). Five articles were comparative, using one or two treatment modalities. Four reports assessed therapy with sildenafil as monotherapy and two as combination therapy with an SSRI.15 Only three studies yielded encouraging results to support the use of PDE5i in the treatment of patients with LL-PE with or without additional SSRIs. In a randomized, 8-week, double-blind, placebo-controlled multicenter study, McMahon et al.16 assessed the efficacy of 50–100 mg as needed sildenafil in 157 potent men with LL-PE, defined in 75% of intercourse attempts. IELT was prolonged in patients, who were taking sildenafil, but the difference was not statistically significant; this was probably because of the incorrect definition of PE used (IELT 2 min in 75% of intercourse attempts during the run-in period) or because of the fact that no treatment effect was found regardless of the criteria used to define PE. Salonia et al.17 showed improvements in stopwatch-measured IELT, sexual satisfaction and frequency of intercourse with sildenafil plus paroxetine over paroxetine alone in men with IELT <1 min. In a prospective randomized clinical trial, Wang et al.18 showed that sildenafil was superior to paroxetine and squeeze technique in improving IELT in men with PE and without ED. Given that PE may be psychogenic, at least in part, and possibly related to anxiety, although part of its definition are feelings of lack of ejaculatory control, bother and distress, there is considerable room for a placebo effect in all studies on the management of PE; for this reason we designed a study providing 1-month follow-up during which patients were crossed over to receive the alternative treatment. Following crossover, patients initially treated with placebo for 8 weeks reported a significant increase in IELT after treatment with vardenafil. Furthermore, after the crossover to the placebo arm, patients initially treated with vardenafil for 8 weeks reported a reduction in IELT, which remained significantly higher when compared with baseline. Other SSRIs crossover studies demonstrate duplication of the initial active drug treatment effect in the initial placebo arm following crossover. Furthermore, following cessation of an SSRI in men with LL-PE and crossover to the placebo arm, the IELT usually approximates the baseline pre-treatment IELT within 1–2 weeks. At present, we have no scientific explanation as to why the beneficial effects of vardenafil are not lost beyond the half-life of the drug in most patients; by contrast, the explanation as to why the initial study drug response was not achieved following crossover in previously treated placebo patients may be ascribed to the small number of active drug tablets assumed by the patients, as shown by regression analysis of the number of assumed tablets vs IELT. Another possible explanation comes from the ‘conditioning’ hypothesis of Rowland and Motofei that considers ejaculatory response as a system of integrated and inseparable hardwired and softwired central and peripheral networks, some being readily modifiable, others not.19 Such a view argues that treatment of PE aimed at multiple levels of functioning will be self-enhancing and ultimately more effective in producing positive therapeutic outcomes than strategies relying solely on either psychological or biological approaches.
The mechanisms of action of PDE5i in the treatment of PE still need to be elucidated. PDE5i may exert their influence both centrally and peripherally. The NO/cGMP pathway seems to play a role in controlling sexual behavior through a central effect. Specifically, a decrease of sympathetic tone by NO activity in the Medial Preoptic Area (MPOA) is related to inhibition of ejaculation.20 Moreover, incubation of rat hippocampal slices with vardenafil results in an increase of cGMP.21 Some authors have shown that NO decreases the central sympathetic output to the periphery22 through a cGMP-dependent mechanism or through interactions with other neurotransmitters.23 Many lines of evidence support the presence and activity of the NO/cyclic guanosine monophosphate (cGMP) and NO/cyclic adenosine monophosphate (cAMP) signaling pathways in the seminal vesicle (SV),24 smooth muscles, prostate and urethra. Thus, the peripheral effects of PDE5i have been ascribed to the capability to retard the ejaculatory response due to modulation of the contraction of SV smooth muscle. It has been shown that PDE5i can reverse the tension of isolated human SV tissue and enhance the production of cAMP and cGMP.25 In particular, one study demonstrated that PDE5i can reverse the tension induced by adrenergic agonist in the isolated prostatic tissue and elevate cAMP and cGMP, thus providing a rationale for their use in the treatment of Lower Urinary Tract Symptoms/Benign Prostatic Hyperplasia (LUTS/BPH).26 Kriegsfeld et al.27 reported that mice homozygous for endothelial NOS (eNOS) gene deletion have striking ejaculatory anomalies. A significantly higher percentage of mice with eNOS gene deletion than of normal controls ejaculated during the testing period, requiring less stimulation, and fewer mounts and intromissions. From animal models, we learn that the vasorelaxant properties of different PDE5i are not similar in isolated blood vessels. Rat aorta rings pretreated with adrenergic agonists suggest that vardenafil, but not sildenafil or tadalafil, affects Ca(2+) handling, in addition to increasing cGMP levels through inhibition of PDE5 to cause relaxation.28 Further studies confirmed that vardenafil, in contrast to sildenafil or tadalafil, is able to block Ca2+ fluxes, thus enhancing vasorelaxation of the rabbit pulmonary artery.29
In our study vardenafil increased IELT and reduced PERT in men with LL-PE without ED. Besides, net improvements in the confidence, perception of ejaculatory control and overall sexual satisfaction were reported. We are aware that daily treatment with SSRIs still remains the preferred treatment option by patients with LL-PE30 but cannot be pursued in all patients; especially in those under current treatment with psychoactive drugs, or those young men who are social consumers of alcohol and club drugs, which may determine severe interactions and adverse events. In our patients, improvements in IELTs and PERTs were paralleled by increase in all IPE domain scores, thus indicating better patients' confidence, perception of ejaculatory control and overall sexual satisfaction. Furthermore, there is a real possibility for the motivated couple that combination pharmacotherapy with a PDE5i and behavioral techniques may yield greater improvements in IELT. Further basic science studies to elucidate the exact mechanism of action of vardenafil on ejaculatory reflex (central and/or peripheral) and clinical controlled studies to confirm the beneficial effects in the real sexual life performance are encouraged.
Conflict of interest
The authors declare no conflict of interest.
About this article
Current Sexual Health Reports (2014)