Topical anaesthetic use for treating premature ejaculation: a double-blind, randomized, placebo-controlled study


W.F.S. Busato Junior, Urology Surgery Service, Marieta Konder Bornahausen Hospital, Itajaí, Brazil. e-mail:



To assess the use of a topical anaesthetic mixture to improve premature ejaculation (PE), for which penile hypersensitivity might be a cause.


The study included 42 men divided in two groups; group A used a lidocaine-prilocaine solution and group B used an inert cream. The tubes of cream were distributed randomly and participants asked to note any unpleasant symptoms, difficulties and the results of each attempt at intercourse, assessed by the intravaginal ejaculatory latency time (IELT).


There was a significant increase in the mean (sd) IELT, from 1.49 (0.9) to 8.45 (0.9) min (P < 0.001) in group A but not in group B, at 1.67 (0.7) to 1.95 (0.12) min (P > 0.05).


We suggest that anaesthetic cream might be effective for treating PE.


premature ejaculation


intravaginal ejaculation latency time


erectile dysfunction


International Index of Erectile Function




Despite the difficulties with published definitions of premature ejaculation (PE) [1], it is the most common male sexual disorder worldwide [2–4]. By some definitions [5,6] PE is present in men who fail to control ejaculation before their partners reach orgasm in at least half of attempts at sexual intercourse. However, Kaplan [7] defined PE as the inability to inhibit the ejaculatory reflex regardless of the time, number of penile thrusts or occurrence of orgasm in the partner. The American Psychiatric Association [8] defines PE as persistent and recurrent ejaculation (not related to the use of any substance) in response to minimal sexual stimulation, immediately after penetration and before the desired moment, which causes interpersonal difficulties.

Men with PE show changes in the autonomic reflex pathways related to ejaculation [9], including a lower vibratory threshold for ejaculation, shorter bulbocavernous reflex latency time, and higher bulbocavernous evoked potentials [10]. These changes are the rationale for dampening the sensory input from the glans and increasing the intravaginal ejaculatory latency time (IELT). It is commonly accepted that a normal IELT should not be < 2 min [11] but very few studies using a topical anaesthetic have been published to date [11].

Therefore, the present double-blind, randomized, placebo-controlled clinical trial was conducted to test a topical anaesthetic cream aimed at extending the duration of sexual intercourse in men with PE.


The study included 42 men, assessed from April 1996 to August 1999; the inclusion criteria were age 18–50 years, a regular sexual life, and those seeking medical help spontaneously. Exclusion criteria were the presence of an organic cause for PE (anatomical abnormality, genital infection, depression and neurological disorder), history of erectile dysfunction (ED), detectable risk factors for ED (alcohol or substance abuse, diabetes mellitus, hypertension, etc.), and use of the study cream in less than half of attempts at sexual intercourse during the study period. All the participants had PE, using the criteria that there is no specific time to ejaculate but rather focusing on whether the couple were satisfied or not with their sexual life [8].

All participants had a medical and sexual evaluation, and a physical examination, and completed the International Index of Erectile Function (IIEF) questionnaire [12]. After the first evaluation, patients were asked to measure the IELT before treatment during a 2-week period, using the stopwatch method; the baseline IELT was determined as the arithmetic mean of five measurements. As this was a controlled study the IELT was defined based on the control group (group A); values were lower than those previously published because the patients had PE. However, question 7 of the IIEF is related to the values used in this study. The ethics committee of our institution approved the study, and all patients provided informed consent.

Patients received tubes of cream (active or placebo) at random, which were not identifiable (the tags were provided later by a member of the study who was unaware of the results); the data were kept in a sealed envelope until data collection was complete. Thus neither the researchers nor the patients knew who was taking the active drug or placebo. After decoding the tubes the two groups were identified (Table 1); group A comprised 24 patients who used lidocaine 25 mg and prilocaine 25 mg cream; and group B, 18 patients who used an inert cream similar in appearance to the first (placebo).

Table 1.  The patients’ characteristics and the results of the assessments
VariableGroup AGroup B
  • *

     P < 0.001.

Initial N2418
N completing1613
Mean age, years33.432.3
(overall 32.9, range 20–50)
Previous treatment 3 3
Married, %8790
Mean (sem):
baseline IELT, min 1.49 (0.9) 1.67 (0.7)
2-month 8.45 (0.9)* 1.95 (0.12)
IIEF score: baseline/2-month
Erectile function26/2828/27
Orgasmic function10/9.710/9.9
Sexual desire10/1010/10
Intercourse satisfaction 4/8.7* 3/4
Overall satisfaction11/1412/11
Sexual satisfaction score: n at baseline/n at 2 months
Bad 1/1 0/0
Regular 5/2 5/4
Good 4/2 5/6
Great 6/6 3/3
Excellent 0/5 0/0

All participants were asked to apply a thin layer of cream on the glans penis, extending the coverage for up to 2 cm on the penile shaft. They were then asked to cover the penis with a condom for 10–20 min, to avoid losing the cream in their clothes and to allow the patient to move freely. After this time they could either have sexual intercourse with the condom on or take it off and wash the penis, to completely remove the cream.

Patients were instructed to use the cream before most of their attempts at sexual intercourse over 30–60 days. During this time patients were asked to assess the IELT, sexual satisfaction and complications. The level of satisfaction level was defined, using a linear scale of 1–10, as bad (1–4), regular (5,6), good (7), great (8,9) or excellent (10); this is similar to the IIEF, where the subjective assessment of satisfaction is defined as excellent when ‘always’, great as ‘almost always’, good as ‘more than half the time’, regular as ‘less than half the time’ and bad ‘almost never’. After 60 days patients completed the IIEF again. The results before and after using the cream were compared within each group using the chi-square test and Student's t-test.


Only 29 participants completed the study (16 in group A and 13 in group B). Of the 13 discontinuing, seven did not return to complete the study, three used the cream for a shorter time than specified, two lost their partners during the study period, and one lost the tube provided. The characteristics of the patients are shown in Table 1. None of the participants had associated diseases and only six had had any treatment for PE (all with antidepressant drugs), with no satisfactory results because of side-effects (Table 1). Fourteen patients (48%) were considered as having primary PE (life-long), with five (17%) having secondary PE (<1 year duration).

The IELTs before and after treatment are shown in Table 1; the IELT was significantly longer in group A (P < 0.001) than in group B. Twelve men from group A had a longer IELT, compared with only two men in group B.

Although erectile function, orgasmic function and sexual desire were normal at baseline (Table 1), intercourse and overall satisfaction were low in both groups. There was an increase in only two variables of the IIEF (intercourse satisfaction, P < 0.001). The number of men in the initial and final sexual satisfaction categories are also shown in Table 1; no patient in group B considered sexual satisfaction to be excellent either before and after treatment. In group A the mean (sd) grade before treatment was 6.68 (1.85) and afterward was 7.81 (2.45); in group B the respective values were 6.77 (0.92) and 6.92 (1.03).

The results for two men in group A were worse than their baseline values; one reported a decrease in penile sensitivity and consequent difficulties in reaching orgasm (>30 min) and the other, whose IELT was > 10 min initially, reported the occurrence of several ejaculations of shorter duration after using the cream. In this case we considered that anxiety was the most important causal factor of PE.

Only three participants reported sexual intercourse with the condom, all in group A. Difficulties in applying the cream were reported by five men; in one because of obesity (group A), and in the others because they had several partners (two each in groups A and B). However, we evaluated sexual satisfaction only in men who completed the study and who had a regular partner.

Adverse reactions were reported by five men (17%); two reported retarded ejaculation (>30 min), with both noting a decrease in penile sensitivity (group A); despite completely removing the cream with water before sexual intercourse there was one case of decreased vaginal sensitivity and two of penile irritation after applying the cream (group A). In group B there were no reports of adverse effects.


PE implies that a man is unable to exert voluntary control over the ejaculatory reflex; as a result, once he is sexually stimulated, he reaches orgasm rapidly [13]. PE is the most prevalent form of male sexual dysfunction [14], and it is known that the IELT is higher in younger than in older men when only patients with normal ejaculation are considered [15]. However, studies assessing the IELT in patients with PE are rare [16].

PE can be classified as primary and secondary; the former, or lifelong PE, is defined as that occurring at the beginning of the patient's sexual life; secondary PE refers to that occurring after a period of normal sexual function [10]. Secondary PE may have several causes, e.g. diabetes mellitus, atherosclerosis, prostatitis, UTI and neurological diseases [17]. Other theories include psychological aspects related to behavioural anxiety or depression [6], but to date no evidence-based psychological study has been published [16]. According to this criterion, half of the present patients were considered as having primary PE and half secondary PE.

More recently, a neurobiological approach was introduced [16]; this was supported by the good results obtained with selective serotonin re-uptake inhibitors, e.g. fluoxetine [18] and paroxetine. Ejaculation is mediated peripherally by α-1 noradrenergic stimulation, probably with cholinergic influences. Selective serotonin re-uptake inhibitors have no sympathicolytic effects on the parasympathetic system and thus the effects of these drugs on delaying ejaculation must be in the CNS [19]. Adler-Graschinsky et al.[20] suggested that serotonin has an inhibitory effect on the noradrenergic mechanism of orgasm, particularly on the presynaptic neurones, which facilitate sympathetic neurotransmission, i.e. serotonin inhibits the sympathetic nervous system, thus retarding ejaculation.

Several studies using animal models give evidence that serotonin receptors are involved in the ejaculatory process, particularly 5-HT2C [21]. For example, Ahlenius et al.[22] consider that stimulating 5-HT2C receptors delayed ejaculation, and reported that the stimulus of 5-HT1A receptors in rats resulted in a shorter ejaculation latency.

The reported assessment of the effectiveness of treatment for PE varies, but prolonging the IELT is an objective measure [16,18,23]. However, no consensus has been reached as to how long the IELT should be to define PE or normality. Waldinger et al.[24] suggested < 1 min, but others report < 2 min [25] or even < 4 min [26]. Schover et al.[27] suggested that < 7 min can be considered normal. In the present study the definition of PE was based on the American Psychiatric Association [8], in which there is no specific time to ejaculate but rather whether the couple is satisfied or not with their sexual life.

Another criterion frequently used is the self- and partner-reported rate of satisfaction [18,23–27]. In addition, a standard tool used to estimate treatment efficacy is the IIEF [12], which gives information on various aspects of sexual function and overall satisfaction. However, the IIEF is not specific for assessing PE because it does not determine the IELT.

Patients with PE might have penile hypersensitivity [9–11], which may represent an important contributing factor to the condition. Therefore, treating the penile hypersensitivity may be an option for treating the PE, because it can retard the ejaculatory process by numbing or desensitizing effects [11]. The present study was based on the possibility that penile hypersensitivity could be an important factor contributing to PE; there was a significant increase in IELT in group A but not in group B. Xin et al.[28] reported shorter latencies and greater amplitudes of somatosensory evoked potentials from the glans penis in men with primary PE than in controls. These results suggest that men with PE have a greater cortical representation of sensory stimuli from the glans penis than do normal controls [16].

Desensitizing creams should be applied directly to the penis; the local topical anaesthetic combination of prilocaine and lidocaine is among the most effective formulations. These off-label agents have been successfully used for treating PE [13].

There is no agreement in other reports about the time of exposure to anaesthetic cream. Choi et al.[23] instructed their patients to apply a herbal extract (SS-cream) 1 h before sexual intercourse and then wash the cream off just before intercourse. They reported an increase of > 2 min in IELT in 80% of the patients, compared with 15% of those treated with placebo. In addition, sexual satisfaction reached a higher grade in 82% of treated patients, compared with 20% of those using a placebo. In the present study the anaesthetic cream was applied for 10–20 min before sexual intercourse and 75% of the men had a greater IELT, vs 15% of those treated with placebo. In a review Morales [11] noted that the anaesthetic cream should be applied for 30 min, but in the study by Atan et al.[29] it was applied for 20 min before sexual intercourse. In an attempt to reduce the application time, Morales [11] tried to produce a ‘eutectic mixture’ in aerosol formulation, with which the application time was reduced to 10–15 min. Atikeler et al.[30] compared different periods (20, 30 and 45 min) elapsing between applying the cream and sexual intercourse, noting that 20 min was the optimum.

Berkovitch et al.[13] postulated that when applying lidocaine-prilocaine cream to the penile skin the drug would be absorbed, resulting in partial anaesthesia of the penis and thus enabling a delay in ejaculation. However, the disadvantage of topical desensitizing creams is the unpleasant effect of penile numbness. In addition, in some cases, female partners complain of vaginal or clitoral anaesthesia, especially when the man does not use a condom. Using a condom would prevent the cream from being applied to the genitalia of the partner and could avoid this effect. However, carefully washing the penis when a condom is not used is very important; in the present study adverse reactions were reported by 17% of men in group A, including retarded ejaculation, decreased vaginal sensitivity in the partner and penile irritation (none of the patients used a condom)

In conclusion, despite the few patients completing the present study, the results suggest that using lidocaine-prilocaine anaesthetic cream might be an effective treatment for PE, possibly by decreasing penile hypersensitivity, a suggested cause of PE. These promising results should motivate the development of a double-blind placebo-controlled trial including more patients and a long-term study, to confirm our findings and determine the acceptability of this method over a longer follow-up.


None declared.