The American Psychiatric Association’s Diagnostic and Statistical Manual, 4th edition, text revision (DSM-IV-TR) defines premature ejaculation as ‘… persistent or recurrent ejaculation with minimal stimulation before, on, or shortly after penetration, and before the person wishes it’. DSM-IV-TR goes on: ‘The clinician must take into account factors that affect duration of the excitement phase (such as age, novelty of the sexual partner or situation, and recent frequency of sexual activity); whether the disturbance causes marked distress or interpersonal difficulty; and whether the premature ejaculation is not due exclusively to the direct effects of a substance (e.g., withdrawal from opioids).’

This multivariate definition is manifestly inadequate, with a low predictive value for premature ejaculation, but it does encompass the main dimensions of premature ejaculation – ejaculatory latency, control, sexual satisfaction, and distress.

Epidemiology of premature ejaculation

Premature ejaculation is often reported, perhaps erroneously, as one of the most common male sexual disorders, and has been estimated to occur in 4–39% of men in the general community. However, it remains poorly defined and inadequately characterized. As a result, there is a substantial disparity between the reported incidence of premature ejaculation in many epidemiological studies (which rely heavily on self-reported premature ejaculation) and the incidence suggested by community-based normative stopwatch intravaginal ejaculatory latency time (IELT) studies.

Most community-based epidemiological studies are limited by their reliance on either patient self-report of premature ejaculation or inconsistent and poorly validated definitions of premature ejaculation. A recent multinational, community-based, age-ranging study of an unselected ‘normal’ population of 500 heterosexual couples, which involved stopwatch timing of the intravaginal ejaculatory latency time during sexual intercourse, has provided previously lacking normative data. This study demonstrated that the distribution of the intravaginal ejaculatory latency time was positively skewed, with a median intravaginal ejaculatory latency time of 5.4 minutes (range, 0.55–44.1 minutes). The median intravaginal ejaculatory latency time decreased with age and varied between countries. The authors regarded the 0.5 and 2.5 percentiles as acceptable standards of disease definition in this type of skewed distribution, and proposed that men with an intravaginal ejaculatory latency time of less than 1 minute (belonging to the 0.5 percentile) have ‘definite’ premature ejaculation, while men with intravaginal ejaculatory latency times between 1 and 1.5 minutes (between 0.5 and 2.5 percentile) have ‘probable’ premature ejaculation.

In a study of 1326 consecutive men with premature ejaculation, lifelong premature ejaculation was present in 736 men (74.4%), and acquired premature ejaculation was present in 253 men (25.6%). Men with premature ejaculation appear younger than those without, and after adjusting for concomitant erectile dysfunction (ED) the risk of premature ejaculation significantly decreased with aging. Higher levels of education, divorce, and the presence of social phobia appear to increase the risk of premature ejaculation. A decreased risk of premature ejaculation has been reported in men with treated diabetes, and no association was found with hypertension, cardiac disease, hypercholesterolemia, and peripheral or central neuropathy. Men with self-reported premature ejaculation have a lower frequency of sexual intercourse and higher levels of intercourse-related anxiety, and they note greater impairment in intercourse satisfaction and sexual relationship satisfaction compared with men without premature ejaculation. However, they do not report a reduced quality of life, reduced sexual desire, or a reduced ability to become sexually aroused.

There are few published data on impact of birth country, religion, or culture on the prevalence of premature ejaculation. An increased susceptibility to premature ejaculation in men from the Indian subcontinent has been reported. Kinsey’s observation that Asian men have shorter times to ejaculation than Caucasian men, who in turn have shorter times to ejaculation than Afro-Caribbean men, has been interpreted to suggest that some races are more ‘sexually restrained’ than others. A recent study reported a preponderance of men from Middle Eastern and Asian backgrounds presenting for treatment of premature ejaculation which exceeded the representation of these ethnic groups in the local population.

Classification of premature ejaculation

The premise that premature ejaculation is a psychosomatic disturbance and due to a psychologically overanxious personality was first suggested by Schapiro in 1943. He classified premature ejaculation as either primary (lifelong) or secondary (acquired). The behavioristic view that chronic premature ejaculation was the result of performance anxiety related to a disturbing initial episode of premature ejaculation was first proposed by Masters and Johnson. Most of the behavioral treatments currently used are based on this premise. Waldinger has extended Schapiro’s classification to include lifelong premature ejaculation, acquired premature ejaculation, natural variable premature ejaculation, and premature-like ejaculatory dysfunction (ejaculatory dysfunction).

Lifelong premature ejaculation is a syndrome characterized by a cluster of core symptoms including early ejaculation at nearly every intercourse, within 30–60 seconds in the majority of cases (80%) or between 1–2 minutes (in 20%), with every or nearly every sexual partner and from the first sexual encounters onwards.

Acquired premature ejaculation differs in that sufferers develop early ejaculation at some point in their life having previously had normal ejaculation experiences; it may be due to psychological or relationship problems, ED, prostatitis or thyroid dysfunction.

In natural variable premature ejaculation, the ejaculation time is never consistently rapid but merely coincidentally and situationally rapid. This type of premature ejaculation should not be regarded as a normal variation in sexual performance and is characterized by inconsistent and irregular early ejaculation, often with reduced ejaculatory control.

Men with premature-like ejaculatory dysfunction complain of premature ejaculation but have a normal ejaculatory latency of 3–6 minutes. It is characterized by a preoccupation with a subjective perception of rapid ejaculation, with an intravaginal ejaculatory latency time within the normal range, but often with reduced ejaculatory control.

Defining premature ejaculation

Medical literature contains several univariate and multivariate operational definitions of premature ejaculation. The definitions of premature ejaculation in DSM-IV-TR and the International Classification of Diseases (ICD)-10 of the World Health Organization differ substantially and are authority-based and not evidence-based, having no support from controlled clinical trials or epidemiological studies. This lack of agreement as to what constitutes premature ejaculation has hampered basic and clinical research into the etiology and management of this condition. Quantitative measures of intercourse, such as the IELT, and subjective measures such as voluntary control over ejaculation or self-efficacy, the extent of patient sexual satisfaction and the level of bother or distress, have been employed as patient reported outcomes (PROs) in premature ejaculation clinical trials. Each of the three criteria above has been operationalized, although not always with consistency.

Intravaginal ejaculatory latency time

Operationalization of premature ejaculation using the length of time between penetration and ejaculation, the IELT, forms the basis of most current clinical studies on premature ejaculation. There is considerable variance of the latencies used to identify men with premature ejaculation, with intravaginal ejaculatory latency times ranging from 1 to 7 minutes, and none of the definitions is based on normative data or offers any supportive rationale for their proposed cut-off time. An average duration of intercourse of 4–7 minutes was reported by Gebhard, suggesting that ejaculation before 4 minutes after intromission should be considered premature.

IELT is measured by a stopwatch operated by the partner. Treating physicians must interpret patient self-report of premature ejaculation and self-estimation of intravaginal ejaculatory latency time with some caution, since the estimation of ejaculatory latency by men and women may correlate poorly with stopwatch-recorded IELT. However, several authors have reported that patient self-estimation of intravaginal ejaculatory latency time correlates reasonably well with subsequent stopwatch IELT.

Waldinger et al. reported intravaginal ejaculatory latency times of less than 30 seconds and less than 60 seconds in 77% and 90% of 110 men with premature ejaculation, respectively. McMahon et al. reported similar results in 1346 consecutive men with premature ejaculation and a mean intravaginal ejaculatory latency time of 43.4 seconds. Predominant anteportal ejaculation (during foreplay) occurred in 5.6% of men. Recent normative data parallel these findings in demonstrating a median intravaginal ejaculatory latency time of 5.4 minutes and suggests that men with an intravaginal ejaculatory latency time of less than 1 minute have ‘definite’ premature ejaculation, while men with intravaginal ejaculatory latency times between 1 and 1.5 minutes have ‘probable’ premature ejaculation.

Voluntary control

Kaplan and other authors have suggested that an inability to defer ejaculation voluntarily defines premature ejaculation. Patrick et al. reported ratings of ‘very poor’ or ‘poor’ for control over ejaculation in 72% of men with premature ejaculation compared with 5% in a group of normal controls. However, control is a subjective measure that is difficult to translate into quantifiable terms, and it is the most inconsistent dimension of premature ejaculation. Control has yet to be adequately operationalized to allow comparison across subjects or across studies. Grenier and Byers failed to demonstrate a strong correlation between ejaculatory latency and subjective ejaculatory control. Several authors report that diminished control is not exclusive to men with premature ejaculation and that some men with a brief intravaginal ejaculatory latency time report adequate ejaculatory control and vice versa, suggesting that the dimensions of ejaculatory control and latency are distinct concepts. Furthermore, there is a higher variability in changes in control compared with intravaginal ejaculatory latency time in men treated with SSRIs.

However, Rosen et al. report that control over ejaculation, personal distress, and partner distress was more influential in determining premature ejaculation status than IELT. In addition, the effect of IELT upon satisfaction and distress appears to be mediated via its direct effect upon control. Clearly, the patient’s perception of control over ejaculation is central to understanding how premature ejaculation is associated with satisfaction with sexual intercourse and ejaculation-related distress.

Sexual satisfaction

Men with premature ejaculation report lower levels of sexual satisfaction than men with normal ejaculatory latency. Patrick et al. reported ratings of ‘very poor’ or ‘poor’ for sexual satisfaction in 31% of men with premature ejaculation compared to 1% in a group of normal controls. The inability to control and defer ejaculation until the female partner was sexually satisfied on at least 50% of intercourse attempts was proposed as a definition of premature ejaculation by Masters and Johnson. An inherent problem exists here in defining a man as dysfunctional based on the sexual responsiveness of his partner. This definition implies that any male whose female partner has difficulty in reaching orgasm should be labeled as a premature ejaculator. This definition is at odds with the report that only 30% of women achieve orgasm during sexual intercourse regardless of the extent of their partner’s ejaculatory control and latency. Rowland reported that over 89.4% of men with self-reported premature ejaculation regarded fulfilling their partner’s sexual needs as very or extremely important.

However, caution should be exercised in attributing improved satisfaction solely to the effect of drug treatment, and contributions from other difficult-to-quantify issues such as intimacy, friendship, sexual attraction, and communication should not be ignored. Furthermore, men with premature ejaculation are easily pleased and report improved satisfaction with minimal improvements in IELT. There is an inherent problem in using sexual satisfaction as a measure of treatment efficacy. It is at best a surrogate for treatment efficacy and must be regarded as a ‘soft’ study end-point compared to objective end-points.


Existing definitions of premature ejaculation include distress as an important dimension of premature ejaculation. However, the word ‘distress’ has negative social implications and its existence is denied by most men with premature ejaculation. This dimension of premature ejaculation is better captured by the word ‘bother’. One study reported that 64% of men with premature ejaculation rated their extent of personal distress as ‘quite a bit’ or ‘extreme’ compared to 4% in a group of normal controls. The extent of psychological impact on patients, partners, and the overall relationship are perhaps the most important aspects of treatment-seeking behavior and best define the severity of premature ejaculation.

Although partner distress is a significant contributor to treatment-seeking behavior, there is limited information regarding the effect of premature ejaculation on the partner. Several studies have reported that the effects of premature ejaculation on the female partner are integral to understanding the impact of premature ejaculation on the male and on the sexual relationship as a whole. Patrick et al. reported that 44% of partners of men with premature ejaculation rated their extent of personal distress as ‘quite a bit’ or ‘extreme’ compared to 3% in a group of partners of normal controls. Patrick et al. also reported that partner PRO measures differentiated men with premature ejaculation from men without premature ejaculation and correlated moderately with measures of intravaginal ejaculatory latency time and subject PRO measures. However, partner perceptions of premature ejaculation generally indicated less dysfunction than those of subjects. Although premature ejaculation adversely affects partner sexual satisfaction, it appears to have minimal impact upon relationship satisfaction. Furthermore, partners of men with premature ejaculation report relatively high levels of female sexual dysfunction. The observation that premature ejaculation often pre-dates the time of onset of the women’s sexual symptoms suggests that premature ejaculation may be a risk factor for female sexual dysfunction.

The design of all future studies on any aspect of premature ejaculation should include a uniform operationalized multivariate definition of premature ejaculation in which the dimensions of latency, control, satisfaction, and distress or bother are defined, measured, and analyzed as continuous variables without arbitrary cut-off values.

The etiology of premature ejaculation

Historically, attempts to explain the etiology of premature ejaculation have included a diverse range of biological and psychological theories. Most of these proposed etiologies are not evidence-based and are speculative at best. Psychological theories include the effect of early experience and sexual conditioning, anxiety, sexual technique, the frequency of sexual activity, and psychodynamic explanations. Biological explanations include evolutionary theories, penile hypersensitivity, central neurotransmitter levels and receptor sensitivity, degree of arousability, the speed of the ejaculatory reflex, and the level of sex hormones.

There is little empirical evidence to suggest a causal link between premature ejaculation and any of the factors thought to cause premature ejaculation. There is, however, limited correlational evidence to suggest that lifelong premature ejaculation is a genetically determined biological variable related to the inherited sensitivity of central 5-HT receptors and that acquired premature ejaculation is due to high levels of sexual anxiety, ED, or lower urinary tract infection.

Ejaculatory latency time is probably a biological variable, which is genetically determined and may differ between populations and cultures, ranging from extremely rapid through average to slow ejaculation. Hyposensitivity of the 5-HT-2C or hypersensitivity of the 5-HT-1A receptors (or both) have been suggested as possible explanations of lifelong premature ejaculation. Men with low 5-HT neurotransmission and probable 5-HT-2C receptor hyposensitivity may have their ejaculatory threshold genetically ‘set’ at a lower point and ejaculate quickly and with minimal stimulation, whereas men with a higher set-point can sustain more prolonged and higher levels of sexual stimulation and can exert more control over ejaculation. Men with a very high set-point may experience delayed or absent ejaculation despite achieving a full erection and prolonged sexual stimulation. Treatment with an SSRI activates the 5-HT-2C receptor, elevates the ejaculatory threshold set-point, and delays ejaculation. The extent of ejaculatory delay may vary widely in different men according to the dosage and frequency of administration of the SSRI and the genetically determined ejaculatory threshold set-point. Cessation of treatment results in re-establishment of the previous set-point within 5–7 days in men with lifelong premature ejaculation.

Anxiety has been reported as a cause of premature ejaculation by multiple authors and is entrenched in the folklore of sexual medicine as the most likely cause of premature ejaculation despite scant empirical research evidence to support any causal role. Several authors have suggested that anxiety activates the sympathetic nervous system and reduces the ejaculatory threshold as a result of an earlier emission phase of ejaculation. The possibility that high levels of anxiety and excessive and controlling concerns about sexual performance and potential sexual failure might distract a man from monitoring his level of arousal and recognizing the prodromal sensations that precede ejaculatory inevitability has been suggested as a possible cause of premature ejaculation by several authors. The causal link between anxiety and premature ejaculation is speculative, is not supported by any empirical evidence, and is in fact contrary to empirical evidence, from some researchers.

Recent data demonstrate that almost half of men with erectile dysfunction also experience premature ejaculation. Men with early erectile dysfunction may intentionally ‘rush’ sexual intercourse to prevent premature loss of their erection and ejaculate with a brief latency. This may be compounded by the presence of high levels of performance anxiety related to their ED, which serves only to worsen the prematurity. In the absence of a thorough sexual history, these men may be incorrectly diagnosed as suffering from premature ejaculation and not the underlying ED.

Premature ejaculation drug trial design

The results of premature ejaculation clinical drug trials are only reliable, interpretable, and capable of being generalized to patients with the disorder studied when conducted in well-defined and consistent populations, using a double-blind placebo-controlled study design, and consistent objective physiological measures or sensitive, validated outcome assessment instruments as study end-points. Subjects with lifelong and acquired premature ejaculation should be treated as separate premature ejaculation subgroups, and subjects with erectile dysfunction or other co-morbid sexual disorders should be either excluded or treated as a separate subgroup.

In premature ejaculation studies, the study population should be well characterized, representative of the overall patient population, and defined using a multivariate definition of premature ejaculation. As the population of men with premature ejaculation is not homogeneous, lifelong and acquired premature ejaculation should be treated as demographically and etiologically distinct disorders and analyzed as separate premature ejaculation subgroups. Subjects should be involved in a stable, monogamous, heterosexual relationship, prepared to attempt intercourse on a regular basis and to provide written informed consent. The presence of comorbid erectile dysfunction should be evaluated using a validated instrument such as the International Index of Erectile Function (IIEF) and patients with any degree of erectile dysfunction should be either excluded from the study or treated as a separate subgroup. Patients with hypoactive sexual desire or other sexual disorders, urogenital infection, major psychiatric disorders, a history of drug and alcohol abuse, or contraindications to the study drug should be excluded from the study.

Measurement of the intravaginal ejaculatory latency time by stopwatch is the best method to diagnose premature ejaculation and the response to treatment, and should be used as a primary efficacy end-point and reported as the fold increase of median intravaginal ejaculatory latency time over pre-treatment IELT. It is well recognized that intravaginal ejaculatory latency time in both the general population and in men with premature ejaculation is distributed in a positively skewed pattern. Waldinger’s assertion that reporting arithmetic means overestimates both baseline and end-point intravaginal ejaculatory latency times and his suggestion that either the median or geometric mean intravaginal ejaculatory latency time values are more representative of treatment response are well-founded. The arithmetic mean intravaginal ejaculatory latency time may exceed median values by as much as 45%. Furthermore, because a typical study population has a broad range of baseline intravaginal ejaculatory latency time values (0–90 seconds), reporting mean raw study-end intravaginal ejaculatory latency time is confusing since it incorrectly suggests that all men respond to that extent. The study-end fold increase in median or geometric mean intravaginal ejaculatory latency time is more representative of true treatment outcome and must be regarded as the contemporary universal standard.

Laboratory studies of ejaculatory dysfunction may be simplified by the use of the Sexual Assessment Monitor (SAM), an electronic data collector that comprises a vibrator to induce ejaculation and a sensor to measure time to erection and intravaginal ejaculatory latency time by the detection of ejaculatory pulses, but the role of such devices in large at-home phase 3 clinical trials is limited. Recent normative intravaginal ejaculatory latency time data support earlier suggestions by several authors that intravaginal ejaculatory latency times of less than 1 minute or less than 2 minutes be regarded as cut-off points for inclusion in a clinical trial.

Subjective PROs of ejaculatory control, sexual satisfaction, and bother or distress are important additional efficacy end-points and can be evaluated using validated PRO instruments. Symonds et al. report the development and validation of a brief self-administered five-item questionnaire, the Premature Ejaculation Diagnostic Tool (PEDT), to diagnose premature ejaculation in clinical trials. A pilot development tool of nine questions was derived from qualitative research involving focus-group and individual interviews of men with either physician-diagnosed or self-reported premature ejaculation; this research was psychometrically analyzed and distilled into a five-item, 0–25 score, unidimensional measure that captures the essence of DSM-IV-TR and the dimensions of control, satisfaction, personal distress, and interpersonal distress. The PEDT has good convergent validity and re-test reliability, and sensitivity– specificity analysis suggests that a score ≥11 indicates premature ejaculation. The PEDT is limited in several respects, but represents a significant development towards simplifying the methodology of premature ejaculation drug studies. Future development of this diagnostic tool would be incomplete without further validation of this tool to determine the potential relationship between score, severity of premature ejaculation, and response to treatment.

The reliability of stopwatch intravaginal ejaculatory latency time alone in assigning premature ejaculation status, the use of PROs to replace stopwatch IELT, or the predictive value of single-item PRO measures compared with multiple-item measures are incompletely understood issues. PRO measures, while providing important information, are at best subjective and relate to highly interpretable and imprecise dimensions of ejaculation, and their significance is weighted differently for different patients. On the other hand, intravaginal ejaculatory latency time may not adequately categorize patients since some patients with a brief intravaginal ejaculatory latency time report little or no bother and are therefore asymptomatic and so not ‘suffering’ from premature ejaculation. Clearly, none of the dimensions of premature ejaculation can universally distinguish men with premature ejaculation from non-premature ejaculation men. The current consensus is that a combination of stopwatch intravaginal ejaculatory latency time and a validated PRO of control, satisfaction, personal distress, and interpersonal distress can adequately identify premature ejaculation status in prevalence studies and in the screening phase of drug trials, and can measure response to treatment.

Clinical drug trial design in sexual medicine has largely relied on retrospective reporting of adverse effects after an interval of up to 4 weeks, with the consequence that patient recall of the details, frequency, severity, duration, and temporal relationship of adverse events to dosing may be unreliable. It has been suggested that adverse events be reported prospec-tively in a patient diary within 24 hours using a validated questionnaire, such as the UKU side-effect rating scale, and this would represent a significant advance in basic clinical trial design.


Selections from the book: “Textbook of Erectile Dysfunction”, 2009

Tagged with:  
Share →